UNIVERSITY OF CALIFORNIA LIBRARY 
 
 THIS BOOK IS NOT TO BE TAKEN FROM 
 THE READING ROOM 
 
 WHEN DONE WITH, RETURN AT ONCE TO ITS 
 PLACE ON THE SHELVES 
 
LIBRARY 
 
 OF THB 
 
 UNIVERSITY OF CALIFORNIA. 
 
 Class 
 
METEOROLOGY 
 
 PRACTICAL AND APPLIED 
 
METEOROLOGY 
 
 PRACTICAL AND APPLIED 
 
 BY 
 
 SIK JOHN MOOKE 
 
 J.C,, M.A., M.D., D.P.H. DUEL., D.Sc. OXON. (Honoris Causa), F.E.C.P.L, 
 
 FELLOW OP THE ROYAL METEOROLOGICAL SOCIETY, 
 EX-SCHOLAR OF TRINITY COLLEGE, DUBLIN 
 
 SECOND REVISED AND ENLARGED EDITION 
 
 NEW YOKE 
 
 EEBMAN COMPANY 
 
 1123 BROADWAY 
 
All Rights Reserved 
 
 LIBn/iiJAfTS FOLD 
 
A FEW EXTRACTS FROM REVIEWS 
 OF THE FIRST EDITION 
 
 THE TIMES. 
 
 " The author gives a lucid and interesting account 
 of modern meteorological methods." 
 
 DAILY EXPRESS, DUBLIN. 
 
 " If there were any real gratitude in the world Dr 
 J. W. Moore's extremely valuable contribution to the 
 science of meteorology ought to be received with 
 universal acclaim, and should quickly become the most 
 widely read of books." 
 
 JOURNAL OF STATE MEDICINE. 
 
 " The book is particularly handy in size and form. 
 The printing is clear, the illustrations numerous and 
 well executed, and the binding distinctive. It is 
 characteristic of the whole book that the information 
 is precise. Dr. Moore's work will serve as a most useful 
 textbook to a far wider circle than that for which it 
 is primarily intended." 
 
 NATURE. 
 " Well written and well illustrated." 
 
 KNOWLEDGE. 
 
 " The whole is well put together. We think Dr. 
 Moore has given his professional brethren and his 
 fellow-observers in meteorology a very useful work." 
 
 BRITISH MEDICAL JOURNAL. 
 
 " This is a good and a pretty book, and one which 
 appeals specially to the medical profession, being the 
 work of an accomplished physician. Among the merits 
 of the book may be mentioned the minute and gener- 
 ally clear descriptions of the numerous meteorological 
 instruments, which are very copiously illustrated,' 
 
 iii 
 
 216634 
 
iv EXTRACTS FROM REVIEWS OF FIRST EDITION 
 
 SCOTTISH GEOGRAPHICAL MAGAZINE. 
 " Gives much valuable information." 
 
 DAILY TELEGRAPH. 
 
 " It may be truly said that after reading his work 
 any person of ordinary education will be acquainted 
 with all the leading principles of meteorological science, 
 and will have reaped pleasure from the clearness with 
 which technical points are explained." 
 
 THE LANCET. 
 
 " May be taken as a full, lucid, and satisfactory ex- 
 position of the present state of knowledge regarding 
 the subject of which it treats. We may say that the 
 work is well printed and handsomely illustrated." 
 
 THE PRACTITIONER. 
 
 " Dr. Moore has given us an attractive and useful 
 book, which will help to popularise the study of 
 meteorology and climatology, and we congratulate 
 both author and publisher." 
 
 EDINBURGH MEDICAL JOURNAL. 
 
 " We would recommend the book as an interesting 
 and thoroughly able treatise upon an interesting 
 science. The volume itself, well printed and illus- 
 trated, and very tastefully bound, is indeed creditable 
 to all concerned in its publication." 
 
 PROVINCIAL MEDICAL JOURNAL. 
 
 " Is brimful of information, besides being expressed 
 in perfect, easily-understood English." 
 
 THE IRISH TIMES. 
 
 " As a textbook, as well as a volume of universal 
 reference, it will take high place in contemporary 
 scientific literature of its order." 
 
PLATE I. 
 
 CIRRUS 
 
 (MARES TAIL) 
 
 CIRRO-STRATUS 
 
 CIRRO-CUMULUS 
 
 [MACKEREL SKY 1 
 
 ALTO-CUMULUS 
 
 10.OOOto2S.OOOA 
 
 ALTO-STRATUS 
 
 IO.OOOtot3.OOOA. 
 
 STRATO CUMULUS 
 
 About. 6f 00 H 
 
 CUMULUS 
 
 CUMULO-NIMBUS 
 
 (STORM CLOUD) 
 WO to 
 
 NIMBUS 
 
 (RAIN CLOUD) 
 3,000 U6.440& 
 
 STRATUS 
 
 V 
 
 4 
 
 
 AND'ES 
 
 (tUMCAGUA) 
 
 . 
 
 HCMT BLANC 
 
 
 /IP 
 
 
 MATTER HORN^ 
 
 n 
 
 mrf BOSTON 
 
 A" 5 
 
 i 
 
 BEN NEVIS 
 I/^IIOWDOI1 
 
 
 H.IMR 
 
 1" i 
 
 CLOUD FORMS. 
 (By permission of R. Inwards, F.R.Met.Soc.) 
 
 Frontispiece. 
 
PREFACE TO THE SECOND EDITION 
 
 WELLNIGH sixteen years have passed since the first edition of this 
 work was published. For several years it has been out of print, 
 but circumstances prevented me from undertaking the prepara- 
 tion of a second edition until some twelve months ago ; and even 
 then the task, congenial though it was, became interrupted from 
 time to time by the exigencies of my practice as a physician and 
 of many collateral interests in a busy life. 
 
 The rapid progress of meteorological science at home and 
 abroad of recent years rendered a thorough revision of the book 
 indispensable. This it has undergone from page to page, so that 
 it will, I trust, be found to reflect to an adequate extent the 
 advances which have been achieved. 
 
 Some notable additions have been made to the text, especially 
 the account, in Chapter VI., of the Meteorological Service of the 
 Dominion of Canada, and the new chapter (XXII.) on the In- 
 vestigation of the Upper Atmosphere. In order not to make 
 the book too long or too bulky, a good deal of the historical 
 account of the United States Weather Bureau has been omitted 
 not, however, in such a way as to make the description of that 
 the largest and most lavishly equipped meteorological service in 
 the world incomplete in any particular. Many new instruments 
 are described in the several chapters, and the method of using 
 them is explained. In many instances they are illustrated 
 through the courtesy of their inventors or makers. 
 
 It only remains to express my grateful acknowledgments to 
 
viii PREFACE TO THE SECOND EDITION 
 
 the many scientific friends who have helped me in my endeavour 
 to make this book a worthy exponent of Modern Meteorology. 
 My special thanks are due to Dr. W. Napier Shaw, LL.D., Sc.D., 
 F.R.S., Director of the Meteorological Office, London, and the 
 members of the staff of that Office in particular, to Mr. R. G. K. 
 Lempfert, M.A., Superintendent of the Statistics and Library 
 Branch ; also to Dr. Hugh Robert Mill, D.Sc., LL.D., Chief of the 
 British Rainfall Organisation, who was good enough to read and 
 criticise the chapters on the Atmosphere of Aqueous Vapour ; to 
 Mr. Ernest Gold, Reader in Meteorology in the University of 
 Cambridge ; to Mr. R. F. Stupart, the Director of the 
 Meteorological Service of Canada ; and to Mr. William Marriott, 
 F.R.Met.Soc., Assistant Secretary of the Royal Meteorological 
 Society, and author of an admirable handbook entitled Hints to 
 Meteorological Observers, prepared by him under the direction 
 of the Council of that Society a work which I have laid under 
 heavy contribution to my pages. 
 
 The very full Index of Subjects and Places and the Index of 
 Proper Names have been compiled with much care by my son, 
 William E. A. Moore, M.A.Univ.Dubl., and will, I think, be 
 found of great use for reference. He has also helped me by read- 
 ing the proof-sheets as they went through the press. 
 
 It will be observed that the volume is copiously illustrated. 
 For this I am indebted to the liberality of the publishers, Messrs. 
 Rebman, Ltd. ; to the Controller of His Majesty's Stationery 
 Office ; to the Meteorological Office and the Council of the Royal 
 Meteorological Society ; to various leading firms engaged in the 
 manufacture of meteorological instruments ; and to Mr. F. 
 Holmes, of Mere, Wiltshire, for the photograph of a flash of 
 lightning taken by its own light in May, 1906. Lastly, my thanks 
 are due, and are hereby accorded, to Dr. W. N. Shaw for the plate 
 illustrating his model of the block of the atmospheric area under 
 observation by ballons-sondes on July 27 and 28, 1908. This 
 ingenious model is described at p. 336. The illustration of 
 Cloud Forms is taken from Weather Lore, by kind permission 
 
PREFACE TO THE SECOND EDITION ix 
 
 of the author, R. Inwards, Esq., and the original photographs 
 were taken by the late Colonel Saunders. 
 
 In conclusion, I crave the indulgence of the readers of this 
 book, and ask them to pardon its shortcomings, of which no one 
 is more conscious than I am myself. 
 
 JOHN WILLIAM MOORE. 
 
 40, FITZWILLIAM SQUARE, WEST, 
 
 DUBLIN, 
 Empire Day, May 24, 1910. 
 
EXTRACT FROM THE PREFACE TO 
 THE FIRST EDITION, 1894 
 
 THE writing of this book has been to me a labour of love. Should 
 the reader derive some pleasure as well as information from the 
 perusal of its pages, the task set before me will not have been 
 undertaken and completed in vain. 
 
 It may be objected that a work on Meteorology should more 
 fitly have been written by an author distinguished for scientific 
 attainments, and whose life-work lay amid the precise sciences. 
 A physician, it may be said and said with truth is daily and 
 hourly exposed to distractions of all kinds in the practice of his 
 profession. He is, in consequence, placed at a disadvantage 
 when he discusses a purely scientific topic. 
 
 In answer to such an objection it may be urged with some 
 force that the physician of all men has the fullest opportunities 
 of observing the far-reaching influence of weather and climate 
 upon human health, happiness, and longevity. If he utilises 
 these opportunities with intelligence and zeal, he is bound to 
 make of such topics a peculiar study. That this often happens, 
 a reference to the Roll of Fellows of the Royal Meteorological 
 Society or of the Scottish Meteorological Society will abundantly 
 prove. In my own case, more than thirty years ago I was 
 already a systematic observer of the weather, and such I continue 
 to be. My " Second Order Station " and the lessons it has 
 taught through all these years afford the needed foil to my more 
 serious professional studies and pursuits. Hence it happens that 
 
 xi 
 
xii PREFACE TO THE FIRST EDITION 
 
 I have been able to bring no small practical experience of meteor- 
 ology to bear in the writing of the pages which follow. 
 
 The work is divided into four parts. A brief introduction is 
 succeeded by a full account of the methods which are employed 
 in practical meteorology. The third part treats of climate and 
 weather necessarily in a somewhat condensed and concise 
 manner. Finally, I have endeavoured to point out, in the fourth 
 and concluding portion of the book, a few of the practical bearings 
 of the subject. In those closing chapters the grave question of 
 the influence of weather and season upon disease is in some 
 measure discussed. 
 
 The time seems opportune for the publication of a popular yet 
 scientific Textbook of Meteorology. The marvellous advances of 
 Preventive Medicine within recent years, the institution of a 
 registrable qualification in Public Health or State Medicine in the 
 United Kingdom of Great Britain and Ireland, the establishment 
 of a new order of public servants drawn from the ranks of the 
 medical profession I allude to Medical Officers of Health the 
 vast development of international telegraphy in modern times, 
 the hearty co-operation of the various national Weather Bureaux 
 all these things have done much within the last quarter of a 
 century to raise Meteorology to the rank of a science, and have 
 given a wonderful impetus to the continuous or the periodic study 
 of the weather. The literature of the subject is therefore daily 
 increasing, and volume after volume is being added to the list of 
 standard works on Meteorology and Climatology. 
 
 Primarily intended for the use of my professional brethren at 
 home and abroad, this book has been so written as to claim the 
 attention of a much wider circle of readers. Its chief object is 
 to convey a clear idea of the Science of Meteorology to anyone of 
 ordinary mental capacity and fair education who had been pre- 
 viously quite unacquainted with so attractive a study. Technical 
 and scientific terms have been as far as possible explained. No 
 pains have been spared to make the description of the different 
 instruments used by meteorological observers as clear as can be. 
 
PREFACE TO THE FIRST EDITION xiii 
 
 In most instances drawings of these instruments have been inter- 
 polated in the text. . . . 
 
 Having said so much, I lay down my pen in the full confidence 
 that the indulgent reader will, in the fascination of the subject, 
 overlook the faults and imperfections of my work, and accept it 
 as the outcome of many years' close observation and attentive 
 study. 
 
 JOHN WILLIAM MOORE. 
 40, FITZWILLIAM SQUARE WEST, 
 DUBLIN, 
 
 September 20, 1894. 
 
CONTENTS 
 
 PART L INTRODUCTORY 
 
 CHAPTER I 
 METEOROLOGY 
 
 PAGES 
 
 Meaning and History of the Term Relations of Meteorology to Medi- 
 cineBuys Ballot's Law Theory of the Winds Cyclonic and 
 Anticyclonic Systems Direction and Force, or Velocity, of the 
 Wind Barometrical Gradients Mr. Ernest Gold's Research on 
 Barometrical Gradient and Wind Force Seasonal Variations of 
 Temperature and Air Movements The Claim of Meteorology to 
 be regarded as a Science - - 1-10 
 
 CHAPTER II 
 THE PHYSICAL PROPERTIES OF THE ATMOSPHERE 
 
 The Atmosphere : what it is, and its Effects The Inclination of the 
 Earth on its Axis The Revolution of the Earth round the Sun 
 Atmospheric Air is Transparent, but has both Substance and 
 Colour Weight of the Atmosphere Its Gaseous Condition It is 
 capable of Compression and of Expansion (Elasticity) Professor 
 Dewar's Experiments on the Liquefaction of Gases by Cold 
 Liquid Air Liquid Oxygen Frozen Air Boyle's and Mario tte's 
 Law The Law of Charles, or Regnault's Law Bearing of these 
 Laws of Compression and Expansion of Gases on Practical 
 Meteorology Elasticity of the Air Rarefaction of the Atmo- 
 sphere at great Altitudes Atmospheric Pressure is exerted 
 equally in all Directions Otto von Guericke's Experiment with 
 the Magdeburg Hemispheres Transparency of the Atmosphere 
 Its Diathermancy and Athermancy - - 11-18 
 
 xv 62 
 
xvi CONTENTS 
 
 CHAPTER III 
 THE COMPOSITION OF THE ATMOSPHERE 
 
 PAGES 
 
 Components of the Atmosphere A Mechanical Mixture of Oxygen 
 and Nitrogen Argon and Helium Nature's Cleansing Opera- 
 tions Volumetric Analysis of Air Proofs that Air is a Mechanical 
 Mixture and not a Chemical Combination Sources of the Carbon 
 Dioxide of the Atmosphere Its Distribution Action of Chloro- 
 phyll upon it Ozone Other Gaseous Constituents of the Atmo- 
 sphere Graham's Law of the Diffusion of Gases Mineral Con- 
 stituents of the Atmosphere Micro-organisms, or Microbes 
 Permanganate of Potassium Test for Organic Matter in the Air 
 Aqueous Vapour - 19-25 
 
 PAKT IL PRACTICAL METEOROLOGY 
 
 CHAPTER IV 
 BRITISH METEOROLOGICAL OBSERVATIONS 
 
 Merle's Journal of the Weather, 1337-1344 Admiral FitzRoy's Charts 
 and Forecasts Synoptic Weather Charts Meteorology of the 
 British Isles Stations of the First Order, or Self-recording Obser- 
 vatories Stations of the Second Order (Normal Climatological 
 Stations) Stations of the Third Order (Auxiliary Climatological 
 Stations) Telegraphic Reporting Stations Anemograph Stations 
 " Isobars " and " Isotherms " Extra Stations Royal 
 Meteorological Society Scottish Meteorological Society 
 Fernley Observatory, Southport Daily Weather Report 
 Beaufort Scale of Weather Weekly Weather Report Monthly 
 Weather Report Accumulated Temperature Numerical Scales 
 for Telegraphic Weather Reports Code System Meteorological 
 Conditions which influence Disease and the Death-rate Equip- 
 ment of a Second Order Station - 26-43 
 
 CHAPTER V 
 
 HISTORY, ORGANISATION, AND WORK OF THE UNITED 
 STATES WEATHER BUREAU 
 
 Historical Sketch Organisation Forecasts Distribution of Fore- 
 castsWeather Signals Scientific Work - - 44-68 
 
 CHAPTER VI 
 
 THE METEOROLOGICAL SERVICE OF THE DOMINION 
 OF CANADA 
 
 History Publications Weather Forecasts - - 63-74 
 
CONTENTS xvii 
 
 CHAPTER VII 
 AIR TEMPERATURE AND ITS MEASUREMENT 
 
 PAOES 
 
 Temperature the most important Meteorological Factor Meaning of 
 the word " Thermometer " History of the Instrument Philo's 
 Thermoscope Fahrenheit's Scale Why Mercury was selected 
 as the Medium for measuring Temperature Celsius' Scale 
 The Centigrade Thermometer Reaumur's Scale Relations of 
 the three Scales to one another Rules for reducing Headings of 
 one Scale to those of another Melting-point of Ice Freezing- 
 point of Water Boiling-point of Water varies with Atmospheric 
 Pressure and with Altitude Definition of a Thermometer Steps 
 in the Construction of this Instrument .... 75-81 
 
 CHAPTER VIII 
 THERMOMETERS 
 
 Standard Thermometers Ordinary Thermometers " Displacement 
 of Zero " Registering Thermometers Phillips'a Maximum 
 Thermometer Negretti and Zambra's Maximum Thermometer 
 Defects in Maximum Thermometers Rutherford's Minimum 
 Thermometer Objections to Spirit Thermometers Six's com- 
 bined Maximum and Minimum Thermometer Casella's mercurial 
 Minimum Thermometer Exposure of Instruments Stevenson 
 Thermometer Stand and Screen The " Wall Screen " Method 
 of Reading the Instruments The Sling Thermometer (Thermo- 
 mltre fronde) Self-recording Thermometers, or " Thermo- 
 graphs " Electrical and Photographic Thermographs Radiation 
 Thermometers Mean Temperature Average Mean Temperature 82-95 
 
 CHAPTER IX 
 RADIATION 
 
 Heat : how transmitted Conduction Convection Radiation Solar 
 Radiation Terrestrial Radiation Effect of Altitude on Tem- 
 perature : how brought about Radiation Thermometers Black- 
 bulb Thermometer in vacuo Bright- bulb Thermometer in vacua 
 Southall's Helio-pyrometer Herschel's Actinometer Angstrom's 
 Electric Compensation Pyrheliometer Pouillet's Pyrheliometer 
 Richards' Actinometer The Wilson Radio-integrator Grass 
 Minimum Thermometers Earth Temperatures Critical Tem- 
 perature at depth of 4 feet Earth Temperature and Diarrhosal 
 Diseases Duration of Bright Sunshine Sunshine Recorders 96-113 
 
xviii CONTENTS 
 
 CHAPTER X 
 ATMOSPHERIC PRESSURE 
 
 PACES 
 
 The Barometer and its Uses Galileo's Observation Torricelli's Dis- 
 covery The " Torricellian Vacuum " Principle involved in 
 Nature's Abhorrence of a Vacuum Fluids used in constructing a 
 Barometer : Mercury, Water, Glycerine Jordan's Glycerine 
 Barometer Estimation of the Height of the Atmosphere 
 Pascal's Experiments Estimation of Mountain Altitudes by 
 means of the Barometer Scaling and Lettering of the Barometer 
 The " Weather Glass," or Wheel Barometer, of Robert Hooke 
 History of the Barometer - 114-119 
 
 CHAPTER XI 
 THE BAROMETER 
 
 The Mercurial Barometer Extreme Limits of Atmospheric Pressure 
 at Sea-level " Torricellian Vacuum " Attached Thermometer 
 Mounting of the Mercurial Barometer Two Difficulties in the 
 Construction of this Instrument How they are Surmounted 
 " Error of Capacity " Capacity Correction The Fortin Baro- 
 meter The Kew Barometer (Adie) The Siphon Barometer (Gay- 
 Lussac) The " Gun Barometer " (FitzRoy) The Wheel Baro- 
 meter, or " Weather Glass " (Hooke) Self- registering Barometers, 
 or "Barographs" King's Mechanical Barograph Ronald's 
 Photographic Barograph Redier's Mercurial (Registering Baro- 
 meter Wheatstone's Electrical Barograph The Aneroid Baro- 
 graph Transmission of Barometric Indications by Electricity 
 (J. Joly) Substitutes for Mercurial Barometers : the Aneroid 
 Barometer Its Altitude Scale Bourdon's Metallic Barometer 
 The Piesmic Barometer Measurement of Heights The " Engi- 
 neering Aneroid " Sympiesometer Hypsometer Barrett's 
 Open-scale Barometer - 120-138 
 
 CHAPTER XII 
 BAROMETRICAL READINGS 
 
 Attached Thermometer Mounting of Barometer Method of taking 
 a Barometrical Observation "Capillary Action" Capillarity 
 The Vernier Graduation of British Barometers Corrections 
 to be applied to Barometrical Readings : Index Error, Capacity, 
 Capillarity, Temperature, Altitude, Gravity Verification of a 
 Barometer The Cathetometer Kew Certificate Schumacher's 
 
CONTENTS xix 
 
 Formula for Reduction of Barometer Readings to 32 F. Ord- 
 nance Datum for Great Britain Ordnance Datum for Ireland 
 Table of Corrections for Altitude controlled by existing Air 
 Temperature and Pressure Laplace's Formula for finding the 
 Difference in Height between Two Places Determination of 
 Mountain Heights by the Barometer Correction for Gravity 139-150 
 
 CHAPTER XIII 
 BAROMETRICAL FLUCTUATIONS 
 
 Periodic and Non-periodic Variations in Atmospheric Pressure 
 Regular and Irregular Variations Regular : Diurnal and Annual 
 Irregular : Cyclonic and Anticyclonic Distribution of Diurnal 
 Variations Explanation Dove's Theory Mr. Strachan's Ob- 
 jections Diurnal Range of Pressure Bibliography Annual 
 Variations of Pressure Periodical Anticyclonic and Cyclonic 
 Systems Explanation of their Formation Trade Winds 
 Irregular Variations in Pressure Isobars : Primary and Secondary 
 Shapes Cyclonic and Anticyclonic Secondary or Subsidiary 
 Depression V-shaped Depression Straight Isobars " Wedge " 
 of High Pressure " Col " of High Pressure Cyclonic and Anti- 
 cyclonic Systems contrasted " Radiation Weather " " In- 
 tensity " Path of Cyclonic Systems Weather Changes accom- 
 panying their Passage " Veering," " Hauling," " Backing " of 
 the Wind Anticyclonic Weather : (1) in Winter ; (2) in Summer 
 
 151-104 
 
 CHAPTER XIV 
 THE ATMOSPHERE OF AQUEOUS VAPOUR 
 
 Aqueous Vapour, or Water in a Gaseous or Aeriform State Its Elastic 
 Force or Tension Influence of Aqueous Vapour on the Baro- 
 meter, on the Weather Evaporation Fogs Capacity of the 
 Atmosphere for Aqueous Vapour Atmometry, or Atmidometry 
 Hygrometry Hyetometry Determination of the Amount of 
 Evaporation Evaporimeters Atmometers, or Atmidometers 
 Saturation Heat made latent in Evaporation Uses of Coolness 
 produced by Evaporation Amount of Evaporation Babing- 
 ton's Atmidometer Von Lament's Atmometer De la Rue's 
 Evaporimeter Richards' Self-recording Evaporation Gauge 
 Pickering's Standard Evaporimeter Mr. Symons's Evaporimeter 
 Observations on Evaporation in Ireland by Mr. James Price, 
 M. Eng., Univ. Dubl., C.E. Mr. Baldwin Latham's observations 
 at Croydon Barton Moss Evaporation Station - - 165-174 
 
xx CONTENTS 
 
 CHAPTEE XV 
 THE ATMOSPHERE OF AQUEOUS VAPOUR (continued) 
 
 PAGES 
 
 Direct and Indirect Hygrometers Organic and Inorganic Hygro- 
 meters Dew-Point Direct Hygrometers : Daniell's, Regnault's, 
 Dines's Trouton's Electrical Indirect Hygrometers : Saussure's 
 Hair Hygrometer Trouton's Gravimetric Recording Hygro- 
 meter The Aquameter Chemical Hygrometer Mason's Dry- 
 and Wet-Bulb Hygrometer The Psychrometer Apjohn's For- 
 mula for calculating the Maximal Vapour Tension for the Dew- 
 Point Glaisher's Hygrometrical Tables Greenwich Factors 
 Examples of their Use Relative Humidity Management of the 
 Wet-Bulb Thermometer Assmann's Ventilated Psychrometer 
 Distribution of Aqueous Vapour in the Atmosphere " Absolute 
 Humidity " History of Hygrometers Crova's Hygrometer 175-189 
 
 CHAPTER XVI 
 THE ATMOSPHERE OF AQUEOUS VAPOUR (continued) 
 
 Hyetometry includes Seven Modes of Condensation of the Watery 
 Vapour of the Atmosphere Dew Two forms of Dew : " Serein " 
 and " Rosee " Dr. McPherson's Theory of Dew Hoar Frost 
 Silver Thaw (Glatteis, Verglas) Interference with Formation of 
 Dew Measurement of Dew Mist and Fog " Scotch Mist " 
 Formation of Fogs, Clouds, and Rain Mr. John Aitken's Re- 
 searches The Influence of Atmospheric Dust Mr. Aitken's Koni- 
 scope The Number of Dust Particles in the Air How Fog or 
 Mist forms Dry Town Fogs Haze Measurement of Fog 
 Densities - 190-204 
 
 CHAPTER XVII 
 THE ATMOSPHERE OF AQUEOUS VAPOUR (continued) 
 
 Definition of a " Cloud "Height of Clouds The " Cloud Line " 
 Luke Howard's Classification of Cloud Forms Upper Clouds : 
 Cirrus, Cirro-cumulus, Cirro-stratus Pallium (Poey) The 
 Mackerel Sky Lower Clouds : Stratus, Cumulus, Cumulo-stratus, 
 Nimbus Roll-cumulus " Pocky " or " Festooned " Cloud 
 Cloud-slopes in Cumulus Various meanings of the word " Nim- 
 bus " Scud Cloud Observations Scale for the amount of 
 Cloud Characters of Thunder- clouds Their Rapid Changes in 
 Formation, Shape, and Density Classification of Cloud System 
 (Mannucci-Abercromby-Hildebrandsson) Direction and Velocity 
 of Clouds The Nephoscope Fineman's Nephoscope Besson's 
 Comb Nephoscope Methods of Stating the Results of Nepho- 
 scope Observations Professor E. G. Hill's Observations in India 
 Besson's Zenith Nephoscope Besson's Spherical Mirror 
 Nephometer - ... 205-219 
 
CONTENTS xxi 
 
 CHAPTER XVIII 
 
 THE ATMOSPHERE OF AQUEOUS VAPOUR (continued 
 and concluded) 
 
 PAOB8 
 
 Relative Amount of Precipitation as Dew and as Rainfall Excessive 
 Precipitation in the Khasi Hills, Assam Mean Annual Rainfall 
 of the World Rainfall Observations in the Seventeenth Century 
 Weighing the Rainfall Hooke's Rain-Gauge (1695) Exhi- 
 bition of Rain-Gauges by the Royal Meteorological Society in 
 1891 Chief Gauges in use: Meteorological Office Gauge, Symons's 
 Snowdon Gauge, the Mountain Gauge, Symons's Storm Gauge, 
 Crosley's Registering Gauge, Yeates's Registering Gauge, 
 Richards' Self-recording Rain-Gauge The best form of Rain- 
 Gauge Measurement of Rain Time and Method of Observing 
 Variation of Rainfall with Elevation : how explained Physical 
 Causes of Rain Influence of Electricity Ionised Air Coloured 
 Rain Snow^ Sleet Hail Measurement of Snow Theory of 
 the Formation of Hail Examples of Hailstorms Soft Hail 
 Relative Size and Weight of Hailstones Distribution of Rain : 
 (1) Geographical, (2) Seasonal, (3) Diurnal Weight and Bulk of 
 Rain 220-203 
 
 CHAPTER XIX 
 ANEMOMETRY AND ANEMOMETERS 
 
 Wind : what it is and how produced .Force or Velocity of Wind de- 
 pends on Barometric Gradients Estimation of Wind Force 
 Beaufort Scale Wind Direction Bearings should be true 
 Variation of the Compass Position of the Pole Star Equation 
 of Time A Windrose : how determined Calms Anemometers : 
 Pendulum, Bridled, Pressure Plate, Pressure on a Fluid, Velocity 
 Robinson's Cup Child's " Step " Evaporation or Tempera- 
 ture, Suction, Direction, Inclination, Musical, Dines's Helicoid, 
 MM. Richards' Anemo-Cinemographe Standard of Wind Velocity 
 (Metres per Second) Lambert's Formula for determining Mean 
 Direction of the Wind Mr. Dines's Comparative Experiments 
 with Anemometers Casella's Self-recording Anemometer 
 Storm of February 26-27, 1903 Life-History of Surface Air-Cur- 
 rents Surface Trajectories of Moving Air (W. N. Shaw and 
 R. G. K. Lempfert) 264 292 
 
 CHAPTER XX 
 ATMOSPHERIC ELECTRICITY 
 
 Identity of Atmospheric Electricity and that obtained from an 
 Electric Machine The Electroscope The Nature of Electricity 
 Atmospheric Electrical Phenomena The Aurora Electrical 
 Density, Force, and Potential Use of the Electroscope The 
 
xxii CONTENTS 
 
 PACKS 
 
 Collector The Electrometer Coulomb's Torsion Balance 
 The Electrophorus The Replenisher The Diurnal and Annual 
 March of Atmospheric Electricity Its Distribution Atmo- 
 spheric Electric Potential or Electric Pressure Thunderstorms : 
 Professor Mohn's Classification Cyclonic and Heat Thunder- 
 storms Geographical Distribution of Thunderstorms Electrical 
 State of the Upper Atmosphere - 293-307 
 
 CHAPTER XXI 
 ATMOSPHERIC ELECTRICITY (continued and concluded) 
 
 Lightning Thunder Varieties of Lightning : zigzag, or forked ; 
 diffused, or sheet ; globular, or ball lightning Fulgurites Prefer- 
 ence of Lightning for certain Trees Rapidity of Lightning 
 St. Elmo's Fire Colour of Lightning Hail The Aurora : how 
 caused ; its Height ; its Colour Ozone Lightning-conductors 
 The Brontometer (Symons) 308-32 i 
 
 CHAPTER XXII 
 THE UPPER ATMOSPHERE 
 
 Kites first used in Meteorology in 1749 by Dr. Alexander Wilson 
 Benjamin Franklin's Historic Experiment (1752) Glaisher's 
 Balloon Ascents (September 5, 1862) Blue Hill Observatory 
 (1894) Milan Commission, 1906 (M. Teisserenc de Bort) 
 Ballons-sondes Dines's Light Meteorographs International 
 Balloon Ascents, 1907-1908 Observations at Bird Hill, Co. 
 Limerick, 1908 Captain Ley's Observations at Sellack, Here- 
 fordshire, 1907 British " Upper Air " Stations Pyrton Hill, 
 Oxfordshire (Observations by Mr. Dines) Dr. W. N. Shaw's 
 Model of a Block of Atmosphere, showing Isothermal Layer 
 Stratosphere Troposphere Advective and Convective Regions 
 Adiabatic Stale Hergessell's Exploration of the Upper 
 Atmosphere in Arctic Regions Vertical Temperature Gradient 
 of the Atmosphere (W. J. Humphreys) Victoria Nyanza Expedi- 
 tion Investigations of Leon Teisserenc de Bort Composition 
 of the '' Free Air "Absolute or Metric Units - - 325-349 
 
 PART III. CLIMATE AND WEATHER 
 CHAPTER XXIII 
 
 CLIMATE 
 
 Meaning of the term" Climate " Definition of Climate Accumulated 
 Temperature Effect of Temperature on the Animal and Vege- 
 table Kingdoms Principal Factors of Climate : Latitude, Alti- 
 tude, Relative Distribution of Land and Water Presence of 
 Ocean Currents ------ 350-3R1 
 
CONTENTS xxiii 
 
 CHAPTER XXIY 
 CLIMATE (continued and concluded) 
 
 PArjES 
 
 Proximity of Mountain Ranges, Soil, Vegetation, Forests, Rainfall, 
 Prevailing Winds Antarctic Climate in 1902-1904 Cold Winds : 
 East Wind of Spring in the British Isles, Mistral, Tramontana, 
 Nortes of Gulf of Mexico, Pamperas, Tormentos, Etesian Winds 
 Hot Winds : Scirocco, Solano, Leveche, Harmattan, Kham- 
 seen, Simoom, Hot Wind of Australia, Fuhn, Leste - - 362-377 
 
 / 
 
 CHAPTER XXV 
 THE CLIMATE OF THE BRITISH ISLANDS 
 
 Division of Climates into (1) Insular, or Moderate ; (2) Continental, 
 or Excessive Distribution of Atmospheric Pressure according 
 to Season Climate of the British Isles : (1) in Summer ; (2) in 
 Winter Isothermals in July and January Features in the 
 Climatology of Great Britain and Ireland Sea Temperatures : 
 in Winter and in Summer Air Temperatures : arrangement of 
 the Isotherms in January, April, July, and October Atmospheric 
 Pressure : its Monthly Distribution Equinoctial Gales - 378-394 
 
 CHAPTER XXVI 
 
 THE CLIMATE OF THE BRITISH ISLANDS (continued and 
 concluded) 
 
 Distribution of Rainfall in the British Islands Regions of heaviest 
 Rainfall How determined : Prevalent Winds, Exposure to these 
 Winds, Mountains Regions of Least Rainfall Geological For- 
 mation Its Local Influence on Temperature Permanent Eleva- 
 tion of Surface Pebble Beds, Sands, and Sandstones Clays and 
 Shales Limestones Crystalline Rocks, whether Slates or Schists 
 Climatological Tables for Dublin Trinity College Meteoro- 
 logical Observatory Is our Climate changing ? Dr. Thomas 
 Rutty's Chronological History of the Weather and Seasons 
 (Dublin) ----.-.. 395-403 
 
 PAET IV. THE INFLUENCE OF SEASON AND 
 OF WEATHER ON DISEASE 
 
 CHAPTER XXVII 
 ACUTE INFECTIVE DISEASES 
 
 Introductory Remarks Dr. William Heberden's Observations on 
 Weather and Disease in 1797 Meteorological Factors which 
 influence the Prevalence and Fatality of Disease : Mean Tempera- 
 
xxiv CONTENTS 
 
 ture, Rainfall, Humidity Acute Infective Diseases : Influenza, 
 Cholera, Diarrhoeal Diseases Seasonal Mortality from Diarrhoaal 
 Disease in Dublin for Thirty Years, 1872-1901 Dr. Edward 
 Ballard's Researches Influence of the Temperature of the Soil 
 The Critical Subsoil Temperature of 56 F. at a depth of Four 
 Feet Dr. E. Meinert's Observations on Cholera Infantum 
 Dr. Ballard's Working Hypothesis The Bacteriology of Diarrhoeal 
 Diseases Influence of Flies in Spreading these Diseases- 409-430 
 
 CHAPTER XXVIII 
 
 ACUTE INFECTIVE DISEASES (continued and concluded) 
 
 Influence on the Prevalence of Enteric Fever of (1) Season ; (2) Tem- 
 perature and Moisture ; (3) Soil and Underground Water Out- 
 breaks at Terling and in Trinity College, Dublin Seasonal 
 Mortality from Enteric Fever in Dublin for Thirty Years, 1872- 
 1901 Typhus Fever a Disease of Winter and Spring Influence 
 of Season, Overcrowding, Defective Ventilation, Temperature 
 and Atmospheric Moisture Seasonal Mortality from Typhus in 
 Dublin for Thirty Years, 1872-1901 Seasonal Prevalence of 
 (1) Smallpox ; (2) Measles ; (3) Scarlatina Seasonal Mortality 
 from Measles and Scarlatina in Dublin for Thirty Years, 1872- 
 1901 431-445 
 
 CHAPTER XXIX 
 
 THE SEASONAL PREVALENCE OF PNEUMONIC OR LUNG 
 FEVER 
 
 Pythogenic Pneumonia Seasonal Prevalence of Bronchitis and 
 
 Pneumonia Compared Pneumonia a Specific Fever - 446-458 
 
 APPENDIX I: 
 
 The Green Flash at Sunrise and Sunset - 459-460 
 
 APPENDIX II. : 
 
 Table of Corrections for reducing Barometric Readings to 
 Standard Gravity, Latitude 45 - 461 
 
 APPENDIX III. : 
 
 Hygrometrical Tables - 462-463 
 
 APPENDIX IV. : 
 
 Conversion Tables for Observations in the Upper Air - 464-465 
 
 APPENDIX V. : 
 
 Bibliography of Snow-Rollers - - 466 
 
 INDEX OF SUBJECTS AND PLACES - 467 486 
 
 INDEX OF PROPER NAMES ..... 487 492 
 
LIST OF ILLUSTRATIONS 
 
 I. Cloud Forms - Frontispiece 
 
 II. The Matterhorn : from the Rifflehaus facing 203 
 III. Temperatures and Pressures in a Block of Atmosphere, Fifteen 
 Miles thick, over a Portion of the British Isles, July 27 and 29, 
 
 1908 . facing 336 
 
 *! FACE 
 
 1. Merle's Journal of the Weather, A.D. 1337-1344 - - 27 
 
 2. Philo's Thermoscope . 70 
 
 3. Kew Observatory Thermometer - - - 83 
 
 4. A, Phillips's, and B, Negretti and Zambra's Maximum Ther- 
 
 mometers . 84 
 
 5. Rutherford's Minimum Thermometer, filled with Pure Alcohol 
 
 for Ordinary Registration, Engine -Divided on the Stem - 85 
 
 6. Six's Thermometer . 87 
 
 7. Casella's Mercurial Minimum Thermometer - 88 
 
 8. Stevenson's Thermometer Stand - . 90 
 
 9. Underground Thermometer - - . 97 
 
 10. Underground Thermometer . . - 97 
 
 11. Solar Radiation Thermometer Stand - - - 101 
 
 12. Richard's Actinometer - ... 103 
 
 13. Wilson's Patent Radio-Integrator - . 104 
 
 14. Casella's Bifurcated Grass Minimum - 105 
 
 15. Symons's Earth Thermometer - ... 195 
 
 16. Diagram illustrating the Relation between Underground Tempera- 
 
 ture and the Death-Rate from Diarrhceal Diseases - - 108 
 
 17. The Campbell-Stokes Sunshine Recorder - - - - 110 
 
 18. The Whipple-Casella Universal Sunshine Recorder - - 111 
 
 19. Jordan Photographic Sunshine Recorder - - - 112 
 
 20. Improved Jordan Photographic Sunshine Recorder - 112 
 
 21. Torricelli's Experiment - - v . . - 115 
 
xxvi LIST OF ILLUSTRATIONS 
 
 FIG. PAGE 
 
 22. Readings of the Jordan Barometer - 117 
 
 23. Fortin Barometer - 123 
 
 24. Wallis's Arrangement for adjusting the Ivory Point - 124 
 
 25. Kew Barometer ... - 124 
 
 26. Gay-Lussac Air-Trap - - 125 
 
 27. Siphon Barometer - 125 
 
 28. Alfred King's Barograph - - 127 
 
 29. The Aneroid Barograph 129 
 
 30. Bartrum's Open-Scale Barometer - - 131 
 
 31. Extra-Sensitive Aneroid Barometer - 132 
 
 32. The Piesmic Barometer - 134 
 
 33. Dines's Self-Eecording Mercurial Barometer - 135 
 
 34. Field's Engineering Aneroid Barometer - - 136 
 
 35. Adie's Sympiesometer - - 136 
 
 36. Casella's Hypsometer - 137 
 
 37. Portable Leather Case for holding Casella's Hypsometer - - 137 
 38 and 39. Method of Reading the Vernier - - 142 
 
 40. Cathetometer constructed for the Indian Government - - 145 
 
 41. Cathetometer, as used at Kew Observatory - 146 
 
 42. Cathetometer, 6| Feet in Height - - 146 
 
 43. Diurnal Oscillation of the Barometer in Various Latitudes - - 155 
 
 44. Cyclonic and An ticyclonic Isobars - - 160 
 
 45. Pickering's Standard Evaporimeter - 170 
 
 46. Daniell's Hygrometer - 176 
 
 47. Regnault's Hygrometer - - 176 
 
 48. Dines's Hygrometer - 177 
 
 49. Vertical View of Dines's Hygrometer - - 177 
 
 50. Mason's Hygrometer - 181 
 
 51. Crova's Hygrometer - 188 
 
 52. Sectional View of Crova's Hygrometer - - 188 
 
 53. Lens in Crova's Hygrometer - 188 
 
 54. Fineman's Nephoscope - - 213 
 
 55. Besson's Zenith Nephoscope - 217 
 
 56. Besson's Spherical Mirror Nephometer - - 218 
 
 57. Hooke's Rain-Gauge - 224 
 
 58. Meteorological Office Rain-Gauge - - 225 
 
 59. Snowdon Rain-Gauge - - 226 
 
 60. Mountain Rain-Gauge - - 227 
 
 61. Bradford Rain-Gauge - 228 
 
 62. Crosley's Self -Registering Rain-Gauge - - 229 
 
 63. Yea tea' s Electrical Self -Registering Rain-Gauge - - 230 
 
 64. Casella's Recording Rain-Gauge - - 231 
 
LIST OF ILLUSTRATIONS xxvii 
 
 FIG. PAGE 
 
 65. Richards' Self-Recording Rain-Gauge (Float Pattern) - - 232 
 
 66. Richards' Self-Recording Rain-Gauge (Balance Pattern) - - 233 
 
 67. Negretti and Zambra's Hyetograph (Upper Part) 234 
 68 and 69. Negretti and Zambra's Hyetograph 235 
 
 70. Forms of Raindrops 241 
 
 71. Casella's Self-Recording Anemometer or Anemograph (see p. 289) - 269 
 
 72. Recording Cylinder of Casella's Anemograph (see p. 289) - - 270 
 
 73. Pressure Plate Anemometer - 272 
 
 74. Dines's Patent Pressure Portable Anemometer - 273 
 
 75. Lind's Anemometer 274 
 
 76. Robinson's Anemometer - - 275 
 
 77. Negretti and Zambra's Improved Robinson's Anemometer - 276 
 
 78. Air-Meter - 277 
 
 79. Vane of Casella's Altazimuth Anemometer - 280 
 
 80. Casella's Altazimuth Anemometer - 281 
 
 81. Vane of Richards' Anemo-Cinemographe - - - 284 
 
 82. Richards' Anemo-Cinemographe - - - 285 
 
 83. Richards' Anemo-Cinemographe (Second Form) - 286 
 
 84. Anemo-Cinemographe in Position - - 288 
 
 85. Diagram of Looped Trajectories for an " Ideal " Storm of Circular 
 
 Isobars and Uniform Wind Tangential to the Isobars, travel- 
 ling with the same Speed as the Wind - 292 
 
 86. Gold-Leaf Electroscope ... - - 297 
 
 87. Thomson's Portable Electrometer - - 301 
 
 88. Thunderstorm at Mere, May 13, 1906 (Photograph taken at 
 
 9.40 p.m.) - . . 309 
 
 89. Kite consisting of Two Strips of Material stretched into Quad- 
 
 rangular Sails on a Lozenge-shaped Plan upon Four Transverse 
 
 Bars kept apart by Two Pairs of Cross-Struts - - 328 
 
 90. Dines's Balloon Meteorograph - ... 329 
 
 91. Recording Plate of Dines's Meteorograph - 330 
 
 92. Map showing the Positions at which Sounding Balloons sent up 
 
 from Certain Observing Stations in 1907-1908 were found - 335 
 
 93. Showing the Relation between Temperature and Height obtained 
 
 by the Ascents of Ballons-Sondes .... 337 
 
 94. Diagram illustrating the Effect of the Perpendicular and the 
 
 Oblique Falling of the Sun's Rays - - - 352 
 
 95. Mortality Curve of Enteric Fever (22 Years, 1869-1890) - - 432 
 
 96. The Mortality Curve for Smallpox (50 Years, 1841-1890) - - 438 
 
 97. The Mortality Curve for Measles (50 Years, 1841-1890) - - 441 
 
 98. The Mortality Curve for Scarlet Fever (30 Years, 1861-1890) - 443 
 
METEOROLOGY 
 
 PART L INTRODUCTORY 
 
 CHAPTER I 
 METEOROLOGY 
 
 THE term Meteorology is more than two thousand years old. 
 It was first used by the philosopher Plato four hundred years 
 before Christ, when he first described Socrates as " a sage, both a 
 thinker on supra-terrestrial things, and an investigator of all 
 things upon the earth beneath." (" SWK/XITT/S, o-o</>6s 
 re fj.T(t)pa (frpovTKTTrjS, /cat TO, i>7To yrjs aTTai/To, 
 Apologia Socratis, cap. ii.) In his Phcedrus the same author 
 employed the very word f) /xcrew/aoAoyta in the sense of a dis- 
 cussion of TO, fj,Tiopa that is, things in the air, natural pheno- 
 mena, the heavenly bodies Cicero's " supera atque caelestia." 
 Fifty years later the philosopher Aristotle wrote a treatise which 
 he styled TO. /xTwpoA.oyi/x, in which he discussed the subjects 
 of air, water, and earthquakes, in this way approaching the 
 modern signification of the word. 
 
 In a lecture on " The Dawn of Meteorology," delivered before 
 the Royal Meteorological Society on March 11, 1908, Dr. G. 
 Hellmann, Professor of Meteorology in the University of Berlin, 
 and Director of the Royal Prussian Meteorological Institute, 
 tells us something of the astro-meteorological system of Mesopo- 
 tamia in the period from 3000 to 1000 B.C. Our knowledge is 
 derived from the astrological cuneiform library of Assurbanipal, 
 discovered by Sir Henry Rawlinson, now preserved in the British 
 
 1 
 
2 INTRODUCTORY 
 
 Museum, and of which some parts have been recently deciphered 
 by Mr. R. Campbell Thompson and the Rev. Franz Xavier 
 Kugler, S.J. The meteorological observations of the Chaldeans 
 were apparently of a quite selective nature, referring particularly 
 to optical phenomena, and especially to the halos. They dis- 
 tinguished clearly the small halo of 22 diameter, called " tar- 
 basu," from the greater one of 45, called " supuru." Besides, 
 they paid much attention to clouds, winds, storms, and thunder. 
 The Babylonians had a windrose of eight rhumbs. They counted 
 the four cardinal points in the order South, North, East, West 
 (sutu, iltanii, sadu, amurra), and by combining them with the 
 word " and " (u), they formed the names for the intermediate 
 directions for example, sutu u sacfa/=S.E., iltanu u amurra = 
 N.W., etc. 
 
 The Greeks were the first to make regular meteorological 
 observations, as we learn from Theophrastus of Lesbos (322 B.C.). 
 From the time of Meton, the astronomer and mathematician of 
 Athens (432 B.C.), the general data of the weather, resulting from 
 observations, were exhibited in the so-called parapegmata 
 (Tra/xxTTTJy/xaTa), a kind of peg almanac fixed on public columns. 
 Fragments of such a parapegma were found recently at Miletus, 
 and are now preserved in the Berlin Museum. 
 
 Originally applied to appearances in the sky, whether atmo- 
 spheric or astronomical in their character, the term Meteorology 
 is at present used in a much stricter and more scientific sense 
 to denote that branch of Natural Philosophy which deals with 
 weather and climate. It includes the study of the physical 
 properties of the atmosphere, a description of the instruments 
 of precision employed in that study, and the application of the 
 knowledge so obtained to the elucidation of the problems of 
 Physical Geography, the advancement of Agriculture, and the 
 promotion of Health, as well as the prevention of Disease. To 
 the physician the securing of these last-named ends is of course 
 of paramount importance. 
 
 From the earliest times the relations existing between Medicine 
 and Meteorology have been most intimate. The far-seeing 
 Fathers of Medicine were not slow to perceive how sensitive to 
 the changes of the weather the delicate human organism is, and 
 
INTRODUCTORY 3 
 
 what an important bearing the study of weather-phenomena 
 should exercise on the practice of Medicine. Two thousand two 
 hundred years ago Hippocrates wrote : " Whoever wishes to 
 study the healing art properly must do this first, he must 
 attentively consider the seasons of the year/' etc. 1 In his other 
 writings, also, he often recurs to this subject. 
 
 Celsus, again, in the second book of his treatise, De Medicind. 
 says : " Saluberrimum Ver est : proxime deinde ab hoc Hiems, 
 periculosior ^Estas, Autumnus longe periculosissimus " a state- 
 ment which to this day is true to the letter as regards Southern 
 Italy, where the words were penned. What can be more graphic 
 than his description of the effects of a north wind, or as we may 
 say of a " nor'-easter " : " Aquilo tussim movet, fauces exasperat, 
 ventrem adstringit, urinam supprimit, horrores excitat, item 
 dolores lateris et pectoris, sanum tamen corpus spissat, et 
 mobilius atque expeditius reddit !" 
 
 There is reason to believe that the suggestions thrown out by 
 these illustrious Greek and Latin physicians were allowed to 
 remain almost a dead-letter. Certain it is that their doctrines 
 as to the close relation of Meteorology and of Climatology to 
 Medicine became dimmed by the rust of time, and were neglected 
 or forgotten. 
 
 Within comparatively recent years, however, a keen interest 
 has been awakened in the subject. Every registered medical 
 practitioner who enters His Majesty's service is required to study 
 the broad facts relating to meteorological observation, and 
 afterwards is called upon to aid in building up a science of climat- 
 ology. Men like Edmund A. Parkes and Ballard in England, 
 Stark and Sir Arthur Mitchell in Scotland, Quetelet in Belgium, 
 Pettenkofer and Buhl in Germany, have studied the weather, and 
 published researches on it and upon its bearing upon health. 
 And now Meteorology, than which no science is more closely 
 analogous to that of Medicine, takes its proper place in the 
 curriculum and in the examination for the Diploma in State 
 Medicine, Public Health, or Sanitary Science, which has been 
 made a registrable qualification under Section 21 of the Medical 
 Act, 1886 (49 and 50 Viet., cap. xlviii.). 
 
 1 Hepl 'A^pojv, 'TSariav, T6TTUV. " 'Irjrpi.K'fji' 6Vm /SouXercu opdws frj 
 r) Troieiv irpurov p.tv e'vdvuteaOou rets wpas rod Ireos," K.T.\. 
 
 12 
 
4 METEOKOLOGY 
 
 Professor Hellmann reminds us that in the epoch of Homer 
 winds were still conceived as absolute beings, like gods ; whereas 
 Anaximander of Ionia, who flourished in the sixth century B.C., 
 is the first to give a scientific definition of the wind which is still 
 valid. He says : "Ave//,oi> efycu pva-iv ae/aos, " The wind is a 
 flowing of air." 
 
 In 1854 the Kev. Humphrey Lloyd, D.D., Provost of Trinity 
 College, Dublin, demonstrated the cyclonic character of most of 
 the gales experienced in Ireland, 1 and so foreshadowed what is 
 now universally known as Buys Ballot's Law a law on which the 
 whole of modern meteorology turns. As applicable to the 
 northern hemisphere, except close to the equator, this law may 
 be concisely stated in the following terms : " Stand with your 
 back to the wind, and the barometer will be lower on your left 
 hand than on your right hand." Similarly, for the southern 
 hemisphere, except close to the equator, the rule holds good : 
 " Stand with your back to the wind, and the barometer will be 
 lower on your right hand than on your left hand." 
 
 So long as the atmosphere is in a state of equilibrium the air 
 is, of course, motionless or " calm "; but the moment the equili- 
 brium is disturbed, an aerial current, which we call " wind," is 
 generated the air moving, or the wind blowing, from the district 
 of greater towards that of less pressure, with the object of re- 
 storing absolute equilibrium. This, however, is in nature hardly 
 ever attained. " The prime cause of atmospheric disturbance," 
 says the Rev. W. Clement Ley, M.A., " is found in the unequal 
 distribution of solar heat over the earth's surface ; in the changes, 
 diurnal and seasonal, in that distribution ; and in the unequal 
 effects thus produced on the tension of the air itself and of the 
 vapour suspended in it." 2 
 
 Experience and reflection have alike proved that air currents 
 flowing in towards an area of low atmospheric pressure do so, 
 not along straight lines, but in curves, so that a gyratory move- 
 ment is developed round the low-pressure area. The determining 
 cause of this phenomenon is the rotation of the earth upon its 
 
 1 " Notes on the Meteorology of Ireland," Eoyal Irish Academy Trans- 
 actions, vol. xxii., " Science," 1854. 
 
 2 Aids to the Study and Forecast of Weather, p. %. London : J. D. Potter. 
 1880. 
 
INTRODUCTORY 5 
 
 axis. A given point on the equator travels round at an immensely 
 greater speed than a similar point near either of the poles, because 
 the equatorial point has to perform a journey of some 25,000 miles 
 in the same space of time (namely, twenty-four hours) that a 
 circumpolar point takes wherein to leisurely traverse a distance of 
 perhaps only 100 miles. The actual speed at which a point is 
 carried round with the earth as it spins on its axis is at the 
 equator, 1,040 miles an hour (namely, 24,900 miles 4- twenty- 
 four hours) ; in latitude 30, 900 miles an hour ; and in latitude 
 60, only 520 miles an hour, or but one-half the equatorial velocity. 
 
 The result of this is that air flowing northwards from the 
 equator outstrips the earth's surface over which it is blowing 
 because of its greater initial velocity, and accordingly trends 
 towards the north-eastward. In this way a south wind is 
 deflected into a south-west wind. Conversely, air flowing south- 
 wards from the north pole lags behind the earth's surface, which 
 is travelling from west to east with increasing speed according as 
 the latitude diminishes ; and so a north wind is deflected into a 
 north-east wind. Supposing, then, an area of low pressure to 
 exist between such south-west and north-east winds, it is evident 
 that these winds must make for that centre so as to fill up its 
 vacuum by curving : the south-west wind through south to south- 
 east and east to the right-hand side, or the eastward, of the low- 
 pressure area ; the north-east wind through north to north-west 
 and west to the left-hand side, or the westward of that area. In 
 this way a circulation in a direction against the hands of a watch is 
 developed round a low-pressure area, the point of lowest pressure 
 in the cyclonic system so formed always lying (in the northern 
 hemisphere) on the left-hand side. 
 
 In the case of the southern hemisphere the reverse of all this 
 holds good. Air flowing southwards from the equator that is, 
 a north wind travels faster than the surface over which it is 
 blowing, and so it trends towards the south-eastward, becoming 
 a north-west wind. Conversely, air flowing northwards from 
 the south pole lags behind the earth's surface, which, as before 
 stated, is travelling from west to east with ever-increasing speed 
 as the latitude diminishes, and so a south wind is deflected into 
 a south-east wind. These north-west and south-east winds, 
 
6 METEOROLOGY 
 
 thus formed, will curve into a vacuum or low-pressure area, in a 
 direction with the hands of a watch : the north-west wind through 
 north to north-east and east to the right-hand side, or the east- 
 ward, of the low-pressure area ; the south-east wind through south 
 to south-west and west to the left-hand side, or the westward of 
 that area. Thus, the point of lowest pressure in the cyclonic 
 system so formed always lies (in the southern hemisphere) on 
 the right-hand side. 
 
 Similar considerations will show that, when air flows out in 
 all directions from an area of high atmospheric pressure, a gyratory 
 movement, or circulation, will be developed, which will be in 
 opposite directions to those just described in the case of each 
 hemisphere north and south of the equator. To these high- 
 pressure systems and circulations the term " anti-cyclonic " is 
 applied, because they are the opposites, or antitheses, of the 
 cyclonic systems already described. " From these considera- 
 tions," writes R. H. Scott, 1 " we gather that round an area of 
 low pressure in the northern hemisphere the wind will circulate, 
 having the lowest pressure on its left, or in a direction against 
 the hands of a watch. Round an area of high pressure in the 
 same hemisphere it will circulate in the opposite direction, or 
 with watch hands. 
 
 " In the southern hemisphere these conditions will be exactly 
 reversed : the wind will move round an area of low barometer 
 readings with watch hands, and round an area of high readings 
 against watch hands." 
 
 It is now nearly sixty years since Professor Adolf Erman first 
 drew attention, in Poggendorff's Annalen (vol. Ixxxviii., 1853, 
 p. 260), to these relations between wind and atmospheric pres- 
 sure. But to the late Professor H. Buys Ballot, Director of the 
 Royal Meteorological Institute of the Netherlands, Utrecht, 
 belongs the credit of having first insisted on their constancy and 
 importance hence the law which expresses them is called " Buys 
 Ballot's Law." 
 
 The application of this law teaches us that, in the northern 
 hemisphere, the wind will be more or less easterly at a given 
 
 1 Elementary Meteorology, p. 254. London : Kegan Paul, Trench and 
 Co. 1883. 
 
INTRODUCTORY 7 
 
 station when the barometer is higher to the north than to the 
 south of it ; more or less southerly when the barometer is higher 
 to the east than to the west ; more or less westerly when pressure 
 is higher to the south than to the north ; and more or less northerly 
 when pressure is higher to the west than to the east. The qualify- 
 ing words " more or less " are used, because the wind seldom 
 blows directly along, or parallel to, the " isobars " (Greek ros, 
 equal ; and ^?a/oos, weight), as the lines of equal barometrical 
 pressure are called. To these lines the wind is often inclined at 
 an angle of some 30 or even 40. 
 
 It is here to be clearly understood that these statements are 
 in general terms, and do not apply to the actual path traversed 
 by a given mass of air in a vast system of atmospheric movement, 
 whether cyclonic or anticyclonic. Such a path has been called 
 by Dr. W. N. Shaw, F.R.S., Director of the Meteorological 
 Office, London, a " trajectory of moving air." His views are 
 stated and explained at p. 291. 
 
 From the foregoing considerations it follows that the direction 
 of the wind, or the point from which the wind is blowing, is 
 determined by differences in atmospheric pressure, which are 
 recorded and gauged by differences in the height of the barometer. 
 But, further, the velocity or force of the wind measured by 
 the Beaufort scale, which will be afterwards explained is found 
 to depend mainly on the amount of those differences, or on what 
 are called the " barometrical gradients." The term " gradient " 
 i s borrowed from the language of engineering. Engineers measure 
 the steepness of a slope or " incline " by the relation which its 
 vertical height bears to its horizontal length. If the ground rises 
 or falls 1 foot in a distance of 60 feet, they speak of the gradient 
 as 1 in 60. 
 
 It is to Mr. Thomas Stevenson, C.E., of Edinburgh, that we 
 owe the application of the term " gradient " to differences of 
 atmospheric pressure as measured by barometrical observations 
 at neighbouring or even distant stations. But barometrical 
 gradients differ from engineering gradients in a very important 
 particular namely, that their vertical and horizontal units of 
 scale are not of the same kind. Their " vertical scale/' says the 
 Hon. Ralph Abercromby, F.R.Met.Soc., " is expressed in units 
 
8 METEOROLOGY 
 
 of barometrical readings, and the horizontal scale in units of 
 geographical measurement/' 1 In the Meteorological Office, 
 London, barometrical gradients are now expressed in decimal 
 parts of an inch of mercury per 15 nautical miles, or about 
 17 statute miles, the line joining the points of barometrical 
 observation necessarily running at right angles to the isobars. 
 This unit of distance 15 nautical miles was adopted by the 
 Permanent Committee of the International Meteorological Con- 
 gress, held at Vienna in 1873, in order to secure uniformity 
 between the gradients of the British scale and those expressed 
 in terms of the metric system. As a hundredth of an inch is 
 nearly equal to a quarter of a millimetre, the English gradient 
 given above corresponds closely with a French gradient expressed 
 in millimetres per 60 nautical miles, or 1 of latitude. 
 
 Barometrical gradients are regarded as slight or moderate 
 when they are below '01 inch, but steep when they exceed 
 02 inch. They seldom exceed *04 inch or '05 inch in the British 
 Islands. An example of the use of these barometrical gradients 
 may be given : when it is said that on a given day there is between 
 Dublin and Holyhead a gradient of -025 for northerly winds, it 
 is implied that the isobars run north and south between those 
 stations, and that the barometer stands as nearly as possible a 
 tenth of an inch higher in Dublin than at Holyhead there being 
 a difference in pressure in favour of Dublin amounting to '025 inch 
 for each unit of 15 nautical miles between the two stations 
 (15 x 4 =60 nautical miles) ('025 inch x 4 ='100 inch, or one-tenth 
 of an inch). 
 
 In general, the steeper the gradient, or the closer the isobars 
 are on a weather chart, the greater the velocity or the force of the 
 wind. But the direct relation between these two factors the 
 gradient and wind force is often interfered with by inequalities 
 of the earth's surface, variations of temperature and of humidity, 
 the existence of cross-currents in the higher strata of the atmo- 
 sphere (" the free air "), and probably also the actual height of 
 the barometer. 
 
 In 1908 Mr. Ernest Gold, M.A., Fellow of St. John's College, 
 
 1 Principles of Forecasting by Means of Weather Charts, p. 4. London : 
 Edward Stanford. 1885. 
 
INTRODUCTORY 9 
 
 Cambridge, Superintendent of Instruments at the Meteorological 
 Office, London, made a highly scientific and technical report to 
 the Director of the Office on the calculation of wind velocity from 
 pressure distribution, and on the variation of meteorological 
 elements with altitude. The latter portion of Mr. Gold's Report 
 contains an account of the results obtained from kite and balloon 
 ascents in Germany and England during 1905 and 1906, and 
 a comparison of the values obtained for the wind velocity and 
 direction at 1,000 metres altitude, by experiment and according 
 to calculation. Immediately a pressure difference arises between 
 two places, the air, if at rest before, will begin to move in the 
 direction of the gradient. As soon, however, as it begins to move, 
 the acceleration due to the rotation of the earth will be called 
 into play, and the air will be deflected in a direction perpendicular 
 to its motion. If we assume the pressure to continue steady over 
 a considerable distance, there will come a time when the force 
 due to the pressure gradient and that due to the earth's rotation 
 balance one another, and keep the air moving along the isobars. 
 For air moving under a steady pressure system along the 
 isobars, the relation between the pressure gradient and the velocity 
 
 v, is given by the equation ---= 2wv sin </>, where o- is the 
 
 & an 
 
 density of the air, p the pressure, n the normal to the isobars, 
 to the angular velocity of the earth about the polar axis, < the 
 
 latitude of the place (u>= ^'1.- = -00007292). 
 
 OD1O4: 1 
 
 The general result of the investigation by Mr. Gold is, in 
 Dr. W. N. Shaw's opinion, to confirm the suggestion that the 
 adjustment of wind velocity to the gradient is an automatic 
 process, which may be looked upon as a primary meteorological 
 law, the results of which are more and more apparent as the 
 conditions are more and more free from disturbing causes, 
 mechanical or meteorological. 2 
 
 Buys Ballot's Law, as originally formulated, was supposed 
 to apply only to those ephemeral and varying systems of atmo- 
 spheric pressure which are called " cyclones " and " anti- 
 
 1 I.e., the number of seconds in a sidereal day. 
 
 2 Barometric Gradient and Wind Force (Meteorological Office, No. 190). 
 London : Wymans and Sons. 1908. Folio. 
 
10 METEOROLOGY 
 
 cyclones "; but it is found to be equally applicable to the far 
 vaster and more permanent seasonal variations of pressure and 
 wind which depend on the alternate heating and cooling of large 
 continents, and the periodical disturbance of the balance of 
 temperature over their surface and that of neighbouring oceans, 
 such as the Atlantic and the Pacific. This topic will be more 
 fitly considered at length in connection with the subject of 
 Barometrical Fluctuations (see Chapter XIII. , p. 160). 
 
 The foregoing reflections will, I trust, vindicate the claim of 
 Meteorology to be regarded as a science not, indeed, an exact 
 science in the sense that mathematics and physics are exact 
 sciences. The phenomena with which it deals are too many and 
 too complex for that ; our knowledge of those phenomena is so 
 imperfect that we cannot systematise it so as to predict or " fore- 
 cast " with certainty. But for this very reason, perhaps, the 
 study of the weather and the elucidation of the laws which govern 
 it possess an interest amounting to fascination, which is quite 
 unfelt by the student of the exact sciences. 
 
CHAPTER II 
 THE PHYSICAL PROPERTIES OF THE ATMOSPHERE 
 
 THE gaseous,_or aerial envelope which surrounds the earth is 
 called the Atmosphere (Greek, (XT/ZOS, vapour ; <r<cu/>a, a globe or 
 sphere). It profoundly influences animal and vegetable life, 
 modifies and retains the heat derived from the sun, facilitates the 
 transmission of sound, causes twilight or the gradual shading of 
 day into night, and is intimately concerned in the production of 
 weather phenomena and geological changes of all kinds. 
 
 Before we proceed to pass in review the properties of the 
 atmosphere, it is necessary to remember that the plane of the 
 earth's equator is inclined at an angle of 2327 / 44 // to its orbit 
 of revolution round the sun, or, as it is technically called, the 
 plane of the ecliptic that is, the apparent annual path of the 
 sun round the heavens, or the real path of the earth as seen from 
 the sun. As the earth revolves round the sun year after year, 
 the plane of the ecliptic cuts the plane of the equator (or the 
 great circle which is equidistant from the poles and perpendicular 
 to the earth's axis of rotation) at two points which are dia- 
 metrically opposite to each other. This happens on March 21, 
 when the sun is on the equator and going northwards, and on 
 September 23, when the sun is again on the equator, but going 
 southwards. The points where the ecliptic and the equator 
 intersect are called the equinoctial points, and the times when this 
 occurs are called the equinoxes, because day and night are then 
 of exactly equal length, the sun being twelve hours above and 
 twelve hours below the horizon. 
 
 From March 21 to June 21 the sun is getting farther and farther 
 north of the equator, and remains longer and longer than twelve 
 hours above the horizon at all places in the northern hemisphere. 
 
 11 
 
12 METEOROLOGY 
 
 On June 21 the sun appears to traverse the heavens at an angular 
 distance north of the equator, amounting at present to 2327'44". 
 He " stands still," as it were, at the Tropic of Cancer and at the 
 summer solstice, before beginning a retrograde journey to the 
 equator and ultimately to the southward of it. Hence the terms 
 " solstice " (that is, solis statio) and " tropic " (from the Greek 
 TpoTrri, a turning round). On December 21 the sun in like manner 
 reaches his greatest southern declination in latitude, at the 
 Tropic of Capricorn ; and in time, at the winter solstice. The 
 expression rpoiral ^eAtoto occurs both in Homer and in Hesiod, 
 the latter first using the phrase as a note of time midsummer or 
 midwinter. Later, the two solstices summer and winter - 
 were distinguished by Greek writers such as Herodotus, Thucy- 
 dides, Plato, and Aristotle, as rpoiral Oepivat and \^i^piva.L 
 respectively. 
 
 The reason why it has been necessary to enter into these 
 particulars about the inclination of the earth on its axis, and the 
 revolution of the earth round the sun, is that the change of 
 seasons on all parts of the earth's surface depends chiefly upon 
 the relations of these two factors to each other. Long days and 
 more or less vertical suns produce summer ; short days and more 
 or less horizontal suns, on the other hand, produce winter. 
 Summer merges into winter through autumn ; winter yields to 
 summer through spring. 
 
 Although invisible to the eye, owing to its transparency, 
 atmospheric air has both substance and colour. That it has 
 substance is evident from the mechanical effects which it pro- 
 duces when in motion. The windmill, the sailing vessel, and the 
 anemometer alike illustrate this. The pressure anemometer has 
 been called upon, in the gusts of great storms, to bear pressures 
 up to 36 or even 40 pounds on the square foot. For example, a 
 pressure of 42 pounds was recorded at Glasgow on January 24, 
 1868, and one of 53 pounds at Greenwich on October 14, 1881. 
 The extraordinary pressure of over 70 pounds per square foot was 
 registered at Bidston Observatory, near Liverpool, on February 1 , 
 1871. This, however, must have been quite a local phenomenon, 
 as a pressure of 49 pounds equals a velocity of 110J miles per 
 hour, and means a " hurricane that tears up trees and throws 
 
PHYSICAL PROPERTIES OF THE ATMOSPHERE 13 
 
 down buildings " (Rouse). In the violent tempest of February 
 26-27, 1903, which wrought immense havoc to trees and buildings 
 in the neighbourhood of Dublin and in Lancashire, a velocity of 
 66 miles an hour was recorded at Kingstown between 4 a.m. and 
 5 a.m. of the 27th ; while 87 miles an hour was registered in squalls 
 from W.S.W. at Southport at 5.55 a.m. of the 27th ; and 88 miles 
 an hour in squalls from S. by W. at Falmouth at 11.50 p.m. of 
 the 26th. 
 
 The principle upon which the parachute is constructed, or the 
 boomerang of the aborigines of Australia, has reference to the 
 substantial nature of air, which is capable of resisting these bodies 
 when passing through it. 
 
 Air, again, has weight, and can be weighed. At a temperature 
 of 32 F., the barometer standing at 29'92 inches, 100 cubic inches 
 of air weigh 32*6 grains nearly. The weight of a cubic foot of 
 air is 573-5 grains. At a temperature of 60 F., and with the 
 barometer at 30'00 inches, the corresponding weights are 30*93 
 grains and 534 '47 grains respectively. According to Dr. Robert J. 
 Mann, 1 13 cubic feet, or a quadrangular block measuring 24 inches 
 in two directions and 39 inches in the third, weighs exactly 
 1 pound ; a room 10 feet square contains 77 pounds of air ; while 
 Westminster Hall holds 75 tons. He adds that air is about 
 760 times lighter, bulk for bulk, than water. 
 
 Under ordinary circumstances, the atmosphere exists in a 
 gaseous state. A gas may be defined as a body whose molecules 
 are in a constant state of repulsion. It is to the late Lord 
 Kelvin (Sir William Thomson), who was President of the Royal 
 Society, and Professor of Natural Philosophy in the University 
 of Glasgow, that we are particularly indebted for a molecular 
 analysis of air. He believed that the atoms of air were so minute 
 that 500,000,000 of them would fit into an inch if arranged 
 in a line. They float at some distance apart from one another, 
 repelling each other very energetically whenever an attempt is 
 made to drive them mechanically together. Like other gases, 
 and, indeed, in consequence of this loose arrangement of the 
 atoms of which it is composed, atmospheric air can be readily 
 compressed to one-half its original volume by doubling the 
 1 Modern Meteorology, p. 3. London : Edward Stanford. 1879. 
 
14 METEOROLOGY 
 
 pressure to which it is subjected. This fixed law of compression 
 of gases was discovered in the seventeenth century (1662) by 
 the Hon. Robert Boyle, F.R.S., and afterwards, independently, 
 in 1679, by Edme Mariotte, a priest who lived at Dijon, in 
 Burgundy. Hence it is known as " Boyle's and Mariotte's 
 Law/' It may be concisely stated thus : The temperature re- 
 maining the same, the volume of a given quantity of gas is inversely 
 as the pressure which it bears. 1 It will be seen, from the statement 
 of this law, that the elastic resistance of air becomes greater and 
 greater the more it is attempted to be compressed, whether 
 mechanically or by means of cold. Notwithstanding this, 
 science has triumphed, and in January, 1893, Professor Dewar, 
 in a discourse on the " Liquefaction of Gases by Cold/' delivered 
 at the Royal Institution of Great Britain, demonstrated the 
 liquefaction of atmospheric air. For the purpose of this experi- 
 ment a temperature of not less than 182 C. (327'6 F.) below 
 the melting-point of ice (32 F.) is required ; in other words, 
 the temperature must be reduced to -295-6 F. If a vessel 
 containing air is chilled to this extent, " the air will condense, 
 trickle down the sides, and accumulate as a liquid at the bottom " 
 (Sir Robert Ball, LL.D., F.R.S.). At this lecture Professor 
 Dewar actually succeeded in pouring half a pint of liquid air from 
 one vessel to another. The professor's subsequent experi- 
 ments have been still more marvellous. He has frozen air into 
 a viscid, jelly-like mass, something between a liquid and a solid. 
 It is supposed that frozen air assumes this form, because oxygen 
 an element which appears to resist independent solidification 
 successfully is probably entangled in a liquid state in the 
 frozen nitrogen of the air. 
 
 It is worth noting that liquid oxygen, also shown by Professor 
 Dewar at his lecture, displays a beautiful blue tint, suggesting, 
 according to Sir Robert Ball, a possible explanation of the colour 
 of the sky that is, of the atmosphere. To this coloration of the 
 air allusion has already been made above. 
 
 Not only is atmospheric air capable of compression, but, in 
 common with all gases, it is capable of expansion also. As com- 
 
 1 Essentials of Physics, p. 61. By Fred. J. Brockway, M.D. Phila- 
 delphia : W. B. Saunders. 1892. 
 
PHYSICAL PROPERTIES OF THE ATMOSPHERE 15 
 
 pression of air takes place in accordance with a fixed law, so also 
 does its expansion. Like other gases, air increases its volume, 
 or expands, by the %} ;T rd of its bulk for every degree of the Centi- 
 grade thermometer, or the fjii-T th of its bulk for every degree of 
 Fahrenheit's scale (Regnault). When air is heated from the 
 melting-point of ice to the boiling-point of water, it is found that 
 1,000 cubic inches become 1,366-5 cubic inches. The fraction 
 i$o* or ISi '^- A (nearly A^) therefore represents the amount of the 
 expansion of a volume of air when raised from 32 to 212 F. 
 that is, through 180. But, as the expansion is equal for each 
 degree, the amount of the expansion for 1 is ...y ^-^; = 1 5^/ '> 
 which, when reduced, becomes ^Y^, as above ; or in decimals, 
 002036 for each degree. It is to be noted that the increase of 
 the unit of volume of a gas for 1 is called its coefficient of 
 expansion. Gases are not only the most expansible of all bodies, 
 but they all have the same coefficient of expansion namely, 
 oi g for 1 C., or 49\ :i for 1 F. Except at very high temperatures 
 their expansion is uniform, no matter what the temperature or 
 the pressure may be. 
 
 The fixed formula or law just stated, by which the expanding 
 effect of heat on a gas is expressed, was first laid down in 1787 
 by M. Charles, then Professor of Physics in the Conservatoire 
 des Arts et Metiers, Paris. It was subsequently arrived at 
 independently by John Dalton, a distinguished meteorologist, 
 physicist, and chemist (1766-1844), whose name is especially 
 identified with the Atomic Theory, which elevated chemistry into 
 a science. In 1802 Louis Joseph Gay-Lussac published a memoir 
 on the subject, and later still Regnault improved upon Gay- 
 Lussac's experiments, so that the law is sometimes spoken of as 
 Regnault's Law as well as the Law of Charles. 
 
 The property which air possesses of contracting in bulk when 
 exposed to pressure, and of expanding again on the removal of 
 that pressure, constitutes what is called elasticity a property 
 which air possesses in no ordinary degree. ~~ 
 
 These laws of the compression and expansion of gases have 
 a most important and direct bearing upon meteorology. Accord- 
 ing to the law of Charles, a volume of air at a constant pressure 
 is proportional to its absolute temperature. According to the 
 
16 METEOROLOGY 
 
 law of Boyle and Mariotte, the density of air varies inversely as 
 its volume. From these two facts it follows that the density of 
 the atmosphere is inversely as its absolute temperature in other 
 words, hot air is specifically lighter than cold air. This aphorism 
 gives a clue to the origin of those great movements of the atmo- 
 sphere which we call " winds/' Wherever the air becomes heated 
 on the earth's surface it expands, and the barometer falls. Where- 
 ever the air is chilled it contracts, and the barometer rises. By 
 these changes in atmospheric pressure are brought about the great 
 wind circulations which have been mentioned in Chapter I. 
 
 Another important consequence of Charles's or Regnault's 
 Law is rarefaction of the atmosphere at great altitudes. Speaking 
 in general terms, the atmosphere exerts at the sea-level a pressure 
 of about 15 pounds (strictly speaking, 14*73 pounds) to the square 
 inch. That is to say, a column of air one inch square, if built 
 up from the earth's surface to the extreme limit of the atmosphere, 
 or to a height of some 200 miles, would weigh about 15 pounds 
 (14*73 pounds). This weight is equal to 2,160 pounds, or nearly 
 1 ton, on a square foot, or 1 kilogramme on a square centimetre, or 
 263,000,000 tons on a square mile. A pressure of 15 pounds upon 
 the square inch is technically spoken of as a pressure of one 
 atmosphere. A column of mercury 30 inches in height and 
 1 square inch in section is found to weigh 14*73 pounds, and so is 
 equivalent to the weight of a column of atmospheric air of the 
 same section. The principle of the construction of the barometer 
 is based upon this fact, as we shall see in a subsequent chapter. 
 Now, if we ascend 2*7 miles, or to a little below the summit of 
 Mont Blanc (15,781 feet), half the weight of the atmosphere will 
 have been left below, and the barometer will read not 30, but 
 only 15 inches. In accordance with Charles's or Regnault's 
 Law, a given bulk of air will, at the height mentioned, expand to 
 twice the volume it would have at sea-level. At a height of 
 5*4 miles its volume would be again doubled, and the barometer 
 would read only 7*5 inches, and so on until at 60 miles above 
 sea-level " the air is probably as rare as the best vacuum that can 
 be produced by the air-pump " (R. J. Mann). The outward 
 limit of the atmosphere must be determined by a counterpoise 
 between gravity on the one hand and centrifugal force and the 
 
PHYSICAL PROPERTIES OF THE ATMOSPHERE 17 
 
 repulsive action of the aerial molecules on the other. Where 
 that limit lies cannot be stated with certainty, but investigations 
 upon the duration of twilight assign to the atmosphere a height 
 of 45 miles at the lowest estimate, and of 190 or possibly 212 miles 
 at the highest. The latter great elevation is inferred by M. Liais, 
 from observations upon the influence of the rarer regions of the 
 atmosphere upon twilight at Rio de Janeiro. 1 
 
 Experiments prove that atmospheric pressure is exerted 
 equally in all directions downwards, as already shown ; but also 
 upwards and horizontally or laterally. Otto von Guericke's 
 classical experiment with the Magdeburg Hemispheres, first per- 
 formed in 1650 before the Imperial Diet at Ratisbon, is con- 
 clusive on this point. Hence it is that objects near the earth's 
 surface are not crushed by the pressure of the atmosphere a 
 pressure so tremendous that an average-sized man sustains a 
 weight of some 15 tons. In this pressure of air in all directions 
 we have a further and effective cause of air movement or wind. 
 A column of cold air being heavier than an equal volume of warm 
 air, its lower strata are pushed towards the area where atmospheric 
 pressure is less, or towards the area of warm and therefore lighter 
 air. The wind, in other words, blows from the area of high 
 barometer towards that of low barometer, not, indeed, in a straight 
 line, but anticyclonically, as explained in Chapter I. 
 
 The atmosphere, when pure and dry, possesses two further 
 remarkable properties transparency and diathermancy. Trans- 
 parency means that pure air is permeable to the vibrations of 
 light. In consequence, we are able to scan the heavens in one 
 direction and to study the effects of light as broken up into colour 
 on the earth in the other direction. When aqueous vapour 
 intervenes*, we gaze with admiration on the glories of sunrise 
 and of sunset, which are due to diffraction of light, absorption 
 of the blue rays of the spectrum taking place because their wave- 
 lengths are small, while the yellow and red vibrations of greater 
 length are allowed to pass through the aqueous vapour and so 
 are reflected to earth. 
 
 Diathermancy (Greek, Sia#e/>/>ios, thoroughly warm] is the 
 property whereby radiant heat, such as that of the sun's rays, 
 
 1 Compies Eendus, tome xlviii., p. 109. 
 
 2 
 
18 METEOROLOGY 
 
 may be transmitted through a medium without raising its tem- 
 perature to any great extent, and this property dry air possesses 
 in a remarkable degree ; it is so freely permeable to radiant heat 
 that both at great altitudes, as on the snow-covered Alps, and 
 in high latitudes, as within the Arctic and Antarctic Circles, the 
 sun's rays may be of extraordinary power, provided only that the 
 atmosphere is extremely dry. Dr. R. H. Scott says i 1 " The 
 observation is as old as the time of Scoresby, that on board a 
 whaler you may see the pitch bubbling out of the seams of the 
 ship where the sun shines on them, while ice is forming on the 
 side of the ship which is in shade." It has been computed that 
 the sun's rays lose, under ordinary circumstances, 20 per cent, 
 of their heat by absorption while passing vertically through the 
 earth's atmosphere. The percentage of loss increases as the path 
 of the heat rays becomes more and more horizontal, until soon 
 after sunrise, or shortly after sunset, a condition of complete, or 
 almost complete, athermancy is reached that is, the power of 
 stopping radiant heat (corresponding to opacity as regards light), 
 is greatest, the heat rays being entirely intercepted by the dense, 
 damp strata of the atmosphere, at sunrise and sunset. 
 
 The heat waves from the sun are long and short. The long 
 waves are absorbed as they pass through the atmosphere towards 
 the earth. The short waves reach the surface, whence they are 
 reflected or radiated back again in lengthened waves, to meet 
 their fate at last in absorption by the aqueous vapour of the 
 atmosphere. In this way radiation into space is checked, and 
 life is preserved upon the face of the globe. 
 
 In the British Islands diathermancy is most decided during 
 the prevalence of clear skies and dry easterly winds in spring 
 and early summer ; it is least marked during the prevalence of 
 damp fogs and mists in late autumn and the winter season of 
 the year. 
 
 1 Elementary Meteorology, p. 57. 1883. 
 
CHAPTER III 
 THE COMPOSITION OF THE ATMOSPHERE 
 
 CAREFUL volumetric analysis shows that atmospheric air consists 
 almost entirely of a mechanical mixture of oxygen and nitrogen 
 (including argon), together with a small and variable quantity 
 of carbon dioxide_or carbonic acid (C0 2 ). There is also present 
 in the air moisture or a quj?o_u8_ vapour, the amount of which 
 varies, especially with the temperature. Peroxide of hydrogen 
 and nitrous and nitric acids are occasional components ; so is 
 sulphurous acid in the vicinity of large towns. Besides the 
 foregoing, very minute traces of ammonia, as well as of sulphide 
 of hydrogen or its ammonia compound, and of helium, besides a 
 variable quantity of organic matter derived from the animal, 
 vegetable and mineral kingdoms, are commonly present in those 
 strata of the atmosphere which are nearest the earth's surface 
 at sea-level. 
 
 The air is purest on the summits of lofty mountains, on open 
 prairies or moorlands, in Arctic regions, and in mid-ocean. 
 It is temporarily purified by gales and thunderstorms, downpours 
 of rain, copious dews and heavy falls of snow or hail all of 
 them great cleansing operations of Nature which the Germans 
 expressively call " Niederschlage " (precipitations). (Cornelius B. 
 Fox). 
 
 The most elaborate volumetric analyses of air have been made 
 by Dr. Angus Smith, Bunsen, and Regnault. In a series of fifteen 
 analyses, Bunsen found the oxygen by volume to vary from 
 2O970 to 20-840 per cent. Regnault' s examinations of air from 
 different parts of the world gave very similar results 20-940 to 
 20*850 per cent. In country air he occasionally found the per- 
 centage volume of oxygen to rise to 21-000. On one occasion 
 
 19 22 
 
20 METEOKOLOGY 
 
 the air of Paris yielded 20*999 of oxygen by volume per cent. 
 Angus Smith, in twenty-two examinations, found 20-938 per 
 cent, of oxygen in the most crowded parts of Perth ; while the 
 air of the heath and of the seashore gave 20'999. For all prac- 
 tical purposes the percentage volume of nitrogen (including 
 argon) in the air may be found by subtracting the foregoing 
 figures from 100, for carbon dioxide is present only in quantities 
 ranging from -025 to '045 per cent., or 25 to 45 per 100,000 parts 
 by volume. 
 
 In order easily to remember the composition of the atmosphere 
 by volume and by weight we may say that in 100 parts there 
 are of 
 
 Volumes. Ojgn. 
 
 Oxygen .. .. .. .. 20-96 23-10 
 
 Nitrogen . . . . . . . . 77'70 l 76-84 
 
 Argon . . . . . . . . 0-80 
 
 Carbon dioxide . . . . . . 0-04 0-06 
 
 Aqueous vapour . . . . . . 0-50 2 
 
 100-00 lOCMX) 
 
 It is right also to mention that, in analyses by weight, the per- 
 centage weights of oxygen and nitrogen may be translated into 
 percentage volumes by dividing the respective specific gravities 
 of these gases into their respective percentage weights. The 
 specific gravity of oxygen is 1*10561 ; that of nitrogen is 0*97135. 
 
 Atmospheric air is not a chemical combination of oxygen and 
 nitrogen. It is simply a mechanical mixture, in which the mole- 
 cules of oxygen are separate and distinct from those of nitrogen, 
 through which they vibrate at inconceivable speed without let 
 or hindrance of any importance. Only when the nitrogen 
 molecules are so compressed by cold as to form a liquid or a solid 
 is the free play of the oxygen molecules so far interfered with as 
 to lead to the formation of the viscid jelly-mass which represents 
 atmospheric air when frozen solid. 
 
 That air is a mechanical mixture and not a chemical com- 
 bination is proved by the following considerations : 
 
 1. There is no chemical formula for air, for the relative propor- 
 tions of oxygen and of nitrogen present in it are not those of their 
 combining weights, or of any simple multiple of those weights. 
 1 May fell to 77 '16. 2 May rise to 1'04. 
 
THE COMPOSITION OF THE ATMOSPHERE 21 
 
 2. When air is artificially made by mixing oxygen and nitrogen 
 together in proper proportions, no change of volume takes place, 
 nor is heat or electricity disengaged as in the production of 
 ordinary chemical combinations. 
 
 3. Air is slightly soluble in water, but oxygen dissolves more 
 readily than nitrogen. If water, in which air has been dissolved, 
 is boiled, the air which is expelled is found to contain nearly 
 35 per cent, of oxygen, instead of only 21 per cent. The air 
 has been oxygenated to the amount of 14 per cent. This could 
 not happen if air was a stable chemical compound. 
 
 4. The refraction of air is the mean of the refraction of oxygen 
 and of that of nitrogen. If air was a chemical compound, it 
 would have a refraction of its own, not the mean refraction of 
 its constituent gases. 
 
 Carbon dioxide is a normal constituent of the atmosphere. 
 The table given above shows that it forms 4 out of every 10,000 
 volumes of air, and weighs 6 grains out of every 10,000 grains of 
 air. If it exceeded this amount to any great extent, it would 
 poison animal life ; if it fell short of this amount, the vegetable 
 kingdom would starve. 
 
 The carbon dioxide of the atmosphere is derived from 
 
 1. The soil and subterranean sources generally. 
 
 2. The respiration of animals. 
 
 3. Combustion. 
 
 4. Fermentation and decomposition. 
 
 5. The burning of limestones in lime-kilns. 
 
 6. Carbonated natural mineral waters. 
 
 Experiments prove that on land the quantity of carbon dioxide 
 in the air is greater by night than by day, because so much of 
 the gas isjjxhalfiiLby plants at night. It increases after rain 
 and towards midday. At sea, it is greater by day (5 volumes per 
 10,000) than by night (3 volumes per 10,000). M. Mene 1 found 
 that the highest percentage of the gas in the air was in October, 
 and that its amount falls to a minimum in December, January, 
 and August. Risler 2 arrived at somewhat analogous results 
 
 1 Comptes Rendus, Ivii., p. 155. 
 
 2 Ibid., xciv., pp. 1390, 1391. 
 
22 METEOROLOGY 
 
 from investigations at Nyon, Switzerland. Frankland, 1 Angus 
 Smith, and M. G. Tissandier 2 all found larger quantities of the 
 gas at considerable elevations than lower down at medium 
 heights. Frankland's experiments were made at the summit of 
 Mont Blanc ; Tissandier's in a balloon. At moderate elevations, 
 however, the quantity of carbon dioxide is not so great as on 
 the ground or at sea-level. 
 
 The composition of the atmosphere, as regards oxygen and 
 carbon dioxide, is maintained by the action of chlorophyll the 
 green granular matter formed in the cells of the leaves of plants 
 which, under the influence of sunlight, has the extraordinary 
 power of splitting carbon dioxide up into its two constituents 
 carbon, which it retains, and oxygen, which it exhales (Wynter 
 Blyth). 
 
 Ozone (Greek ofw, I have a smell) is a colourless, gaseous sub- 
 stance, with a peculiar smell like weak chlorine, which is de- 
 veloped as the immediate result of electrical disturbances. 
 Houzeau has experimentally demonstrated its amount in country 
 air to be 1 volume in 700,000 volumes of air. It is absent in 
 cities, in crowded dwelling-rooms, and over marshes. Unfortu- 
 nately, the tests for it react to other substances in the atmosphere, 
 such as hydrogen dioxide (peroxide of hydrogen) and nitric acid. 
 The whole subject, however, of ozone and of ozone-testing will 
 more fittingly be considered in Chapter XXI. on ^Atmospheric 
 Electricity (see p. 321). 
 
 The element argon (atomic weight = 40) was discovered in 1894 
 by Lord Rayleigh, F.R.S., O.M., and Sir William Ramsay, 
 K.C.B., F.R.S., who described it as a probably inert constituent 
 of the atmosphere ; hence its name from the Greek a, privative, 
 and epyov, work. It shows little affinity for other elements, and 
 is a constituent of atmospheric air to the amount of 0'8 per cent, 
 in volumetric proportion. Neon, krypton, and xenon are three 
 newly discovered atmospheric gases, to which attention has been 
 drawn by Sir William Ramsay. 
 
 Traces of helium (atomic weight = 4) are also present in atmo- 
 spheric air This element receives its name from the Greek 
 
 Journal of the Chemical Society, 18G1. 
 Comptes Rendus, April 12, 1875. 
 
THE COMPOSITION OF THE ATMOSPHERE 23 
 
 word ?]Aio?, the sun, because at the time when it first attracted 
 attention it was supposed not to exist upon our earth. In 1895, 
 however, it was discovered to be also a terrestrial element by 
 Sir William Ramsay, and since then it has been recognised as one 
 of the products given off by radium and as a constituent of 
 certain minerals. 
 
 The other gases which are more or less constantly present in 
 the atmosphere have already been named. Nitric acid is 
 generally present in minute quantities. Sulphurous acid is 
 derived from the combustion of coal in large towns. From 
 experiments undertaken at Lille, A. Ladureau 1 found that it 
 increased in amount during calm weather, while it equally 
 decreased on stormy days. 
 
 Mr. Horace T. Brown estimates the normal amount of ammonia 
 present in the air to be about 6 parts per 1,000,000. Heavy rain 
 lessens the amount for a time. 
 
 The presence of hydrogen sulphide in the atmosphere of large 
 towns is proved by the tarnishing of silver plate and coins. It 
 may be present as ammonium sulphide. 
 
 Besides carbon dioxide, marsh-gas is always present in the air, 
 
 ^r%lthough in minute quantities. Its chemical formula is CH 4 . 
 
 It is a product of decomposition of organic matter in stagnant 
 
 pools. Hence its name. It is also called methane, and constitutes 
 
 the fire-damp of coal-mines. 
 
 These gases diffuse freely through the atmosphere in obedience 
 to a fixed law, which is that the diffusibility of two gases varies 
 in the inverse ratio of the square roots of their densities. This 
 law of the diffusion of gases, commonly called " Graham's Law/' 
 is based upon a consideration of the size and velocity of repulsion 
 of the molecules of each gas. If one molecule, say of oxygen, 
 weighs sixteen times as much as another, say one of hydrogen, 
 then the latter has to move four times as fast as the former in 
 order to strike as effective a blow. Hydrogen, a light gas, 
 diffuses four times as fast as oxygen, a heavy one. The rate of 
 diffusion is in this instance inversely as the square roots of one 
 and sixteen. 
 
 " The process of diffusion," says Professor Miller in his Chemical 
 1 Ann, Chem. Phys. 5, xxix., pp. 427-432, 
 
24 METEOROLOGY 
 
 Physics, " is one which is continually performing an important 
 part in the atmosphere around us. Accumulations of gases 
 which are unfit for the support of animal and vegetable life are 
 by its means silently and speedily dispersed, and this process 
 thereby contributes largely to maintain that uniformity in the 
 composition of the aerial ocean which is so essential to the comfort 
 and health of the animal creation. Respiration itself, but for 
 the process of diffusion, would fail of its appointed end, in rapidly 
 renewing in the lungs a fresh supply of air in place of that which 
 has been rendered unfit for the support of life by the chemical 
 changes which it has undergone/' 
 
 Among mineral constituents of the atmosphere common salt 
 (sodium chloride) is the most frequently met with, especially in 
 the lower strata of the air. Spectroscopic analysis of the Bunsen 
 flame invariably gives the sodium line in consequence of the 
 presence of the salt. Metallic dust of various kinds abounds in 
 the vicinity of manufactories and in the air of mines. 
 
 Vast numbers of micro-organisms, or microbes, infest the air. 
 These belong both to the pathogenic and to the non-pathogenic 
 groups. They are infrequent in the air of mid-ocean and on high 
 mountains, but abound in the air of towns, swarming in that of 
 ill-ventilated dwelling-rooms. Mr. J. B. Dancer, 1 F.R.A.S., 
 examined the solid particles of the air of Manchester microscopi- 
 cally, and came to the conclusion " that 37 J millions of these 
 bodies [particles of both organic and inorganic origin], exclusive 
 of other substances, were collected from 2,495 litres =88 cubic feet, 
 of the ' air of Manchester/ a quantity which would be respired 
 in about ten hours by a^man^of ordinary size when actively 
 employed/* 
 
 According to Mr. A. Wynter Blyth 2 , the best chemical method 
 of estimating organic matter in the air is its approximate estima- 
 tion by means of permanganate of potassium. A known bulk of 
 air is drawn through a little distilled water, and the amount of 
 oxygen consumed is determined by the Forchammer process. 
 Ten cubic centimetres of a standard solution of permanganate 
 
 1 Proceedings of the Literary and Philosophical Society of Manchester, 
 vol. iv. Series 3. 1867-1868. 
 
 2 A Manual of Public Health, p. 96. 1890. 
 
THE COMPOSITION OF THE ATMOSPHERE 25 
 
 of potassium, 1 and ten cubic centimetres of sulphuric acid, diluted 
 to one-third, are added to a known bulk of water say a litre. 
 The whole is then heated for four hours to 80 F. (26-6 C.). At 
 the end of that time the water is titrated that is, has its strength 
 determined with a hyposulphite solution made by dissolving 
 one part of crystallised sodic hyposulphite in a litre of water, 
 using iodide of potassium and starch as an indicator. The value 
 is obtained by running a control with distilled water. 
 
 One of the most important constituents of the atmosphere 
 is aqueous vapour, or water in a gaseous or aeriform state. A 
 vapour, like a gas, is subject to the laws of expansion and of com- 
 pression, which have been already discussed in these pages but 
 only within certain limits. " If," says Dr. R. H. Scott, 2 " these 
 limits be overpast i.e., if the pressure becomes too great or 
 the temperature falls too low a portion of the vapour will pass 
 into the state of liquid. Under any circumstances of pressure 
 and temperature, a given space can contain only a given quantity 
 of vapour. This is as true of vapour mixed with air as of vapour 
 by itself." 
 
 The overwhelming influence of aqueous vapour in practical 
 meteorology arises in part from its sensitiveness to the action 
 of heat even moderate changes of temperature causing it to 
 expand or contract, to evaporate or condense, with great facility ; 
 but more especially from its liability to pass from the gaseous or 
 vaporous, to the liquid or even solid form, at temperatures of 
 everyday occurrence in Nature. It is, however, to the marvellous 
 heat-absorbing powers of aqueous vapour that the attention of 
 the practical meteorologist must in particular be directed, when 
 he seeks an explanation of the phenomena of what we call 
 " Weather." 
 
 1 Made by dissolving 0-395 gramme of potassic permanganate in a litre of 
 water. Each c.c. contains 0-0001 gramme of available oxygen. 
 
 2 Elementary Meteorology, p. 95. 1883. 
 
PAKT IK PRACTICAL METEOROLOGY 
 
 CHAPTER IV 
 BRITISH METEOROLOGICAL OBSERVATIONS 
 
 THE earliest known Journal of the Weather was that kept at 
 Oxford by the Rev. William Merle, Fellow of Merton College, and 
 afterwards Rector of Driby, Lincolnshire, during the seven 
 years 1337-1344. His " Consideraciones Temperiei pro 7 Annis " 
 were discovered in a MS. 1 in the Bodleian Library, Oxford, in 
 1891, and were immediately afterwards reproduced and trans- 
 lated under the supervision of the late Mr. G. J. Symons, F.R.S., 
 to whom British meteorology owes so much (Fig. 1 ) . The observa- 
 tions are climatological, and are written in Latin in Old English 
 characters. They represent the primeval stage of weather study, 
 in which popular weather prognostics came to be drawn from 
 daily scanning of the heavens, untiring observation of the move- 
 ments of animals, including the arrival and departure of migratory 
 birds, of the leafing and flowering of trees and shrubs, of the 
 ripening of harvests and fruits, and of the fall of the leaf in 
 autumn. These phenological observations in bygone days con- 
 ferred a marvellous power of forecasting weather, and so they do 
 at the present day also, so far as local districts are concerned. 
 
 Since the discovery of the barometer in the seventeenth century, 
 isolated observations on atmospheric pressure crude, no doubt^ 
 and unreduced to any standard altitude or temperature 
 afforded an increased power of weather forecasting. 
 
 It was not, however, until 1861 that the systematic application 
 
 1 Digby MS. 176, fol. 4. 
 26 
 
BRITISH METEOROLOGICAL OBSERVATIONS 27 
 
 of telegraphy to the synchronous study of the weather at distant 
 stations revolutionised meteorology and raised it to the dignity 
 of a science. The service of Daily Weather Charts and Forecasts, 
 which was inaugurated in France by M. Le Verrier and in England 
 by the late Admiral FitzRoy in the year named, has been amplified 
 and improved since then, but to them belongs the credit of 
 organising a system of weather study which now extends over 
 the whole civilised world. Every country in Europe ; Egypt ; 
 Canada, the United States, and the Argentine Republic ; India, 
 
 FiG. 1 !. MERLE'S JOURNAL OF THE WEATHER, A.D. 1337-1344. 
 (Reproduced by permission of the Royal Meteorological Society.) 
 
 Japan, China, Australia, and New Zealand all have their 
 Meteorological Offices or Weather Bureaux, at which synoptic 
 weather charts and forecasts are prepared at least once a day. 
 
 The word " synoptic " (Greek O-VVOITTIKO?, from o-wo^is, a 
 seeing all together, a general view) signifies that the weather chart 
 has been prepared from observations taken at the same moment 
 of time over a large tract of country, and that it illustrates the 
 type of weather prevailing throughout the district embraced in 
 the chart at the hour of observation. 
 
 The following account of the method in which the land meteor- 
 ology of the British Islands is studied through the medium of 
 
28 METEOROLOGY 
 
 the Meteorological Office, London, will apply mutatis mutandis 
 to the Weather Bureaux of the British Colonies and of Foreign 
 States. 
 
 The observatories in the United Kingdom in connection with 
 the Central Office may be arranged in five classes : 
 
 1. Stations of the First Order, or Self-recording Observatories, 
 which are furnished with self-registering instruments by which 
 all the principal meteorological phenomena are recorded con- 
 tinuously. Thus, there are continuous records or hourly readings 
 of pressure, temperature, wind, sunshine, and rain, with eye 
 observations of the amount, form, and motion of the clouds, and 
 notes on the weather. These alone afford the materials necessary 
 for the study of the periodic variations of the meteorological 
 elements. The autographic records are checked by frequent eye 
 observations. There are six such stations at present : three in 
 England Falmouth, Kew, and Stonyhurst ; two in Scotland 
 Aberdeen and Glasgow ; and only one in Ireland Valentia, in 
 Kerry. The observatory at Armagh was relegated to Class II. 
 some years ago. Observatories are also maintained at Green- 
 wich (the Royal Observatory), Oxford (Radcliffe Observatory), 
 Bidston (Mersey Docks and Harbour Board), Southport (the 
 Corporation), and Berkhamsted (E. Mawley, Esq., F.R.Met.Soc.). 
 
 2. Stations of the Second Order, or Normal Climatological 
 Stations. At the end of March, 1909, the total number of these 
 stations was 80, including 10 belonging to the Royal Meteorological 
 Society and 13 belonging to the Scottish Meteorological Society. 
 The stations are distributed as follows : 50 in England, 2 in Wales, 
 19 in Scotland, and 9 in Ireland. Reports from the Irish stations 
 are regularly supplied to the Registrar-General for Ireland for 
 his Weekly and Quarterly Returns. At all of these climato- 
 logical stations regular eye observations are taken twice daily 
 at 9 a.m. and 9 p.m. local time (or other accepted combinations 
 of hours), of atmospheric pressure, temperature (dry bulb and 
 wet bulb), wind, amount of cloud, and weather, with the daily 
 maximum and minimum of temperature, the daily rainfall, 
 together with general remarks on the weather. The observers 
 at these stations are all volunteers. All the stations are regularly 
 inspected by inspectors from the Meteorological Office. 
 
BRITISH METEOROLOGICAL OBSERVATIONS 29 
 
 3. Stations of the Third Order, or Auxiliary Climatological 
 Stations, recording observations similar in kind to those at the 
 Normal Climatological Stations, but either (a) less full, or (6) taken 
 only once daily, or (c) taken at hours other than 9 a.m. and 
 9 p.m. On March 31, 1909, 108 stations of this order were at 
 work for, or in communication with, the Meteorological Office 
 namely, 78 in England (8 reporting through the Royal Meteoro- 
 logical Society), 9 in Scotland (2 reporting through the Scottish 
 Meteorological Society), 9 in Wales, and 12 in Ireland. 
 
 4. Telegraphic Reporting -Stations. Twenty-nine are in the 
 British Isles, at which the observations are taken by eye at 
 certain hours determined by the requirements of the telegraphic 
 system. In some cases the eye observations are supplemented 
 by self-recording aneroid barometers, etc. Of the 29 home 
 telegraphic stations, 13 observe at 7 a.m., 1 p.m., and 9 p.m., 
 and thus come under the international definition of Second 
 Order Stations. They are not, however, included in the Second 
 Order Stations mentioned above. 
 
 5. Anemograph Stations, furnished with instruments registering 
 wind only. At Armagh, rainfall and sunshine are in addition 
 recorded. The anemograph stations furnish continuous records 
 of wind velocity (force), and in most cases also of wind direction. 
 The observations from these stations are important in connection 
 with storms, and afford evidence available in courts of law 
 relative to collisions at sea and damage done by wind either on 
 land or at sea. 
 
 The foreign reporting-stations, 44 in number, extend along 
 the entire western coast of the continent of Europe, from Bodo 
 in lat. 67 N. to Lisbon in lat. 38 N. They include 4 stations 
 on the shores of the Baltic, 5 in Iceland, 1 in the Faeroe, 4 in 
 Germany, 1 in Central Sweden, 4 in Norway, 2 in the Azores, 
 and 3 in the Mediterranean. The observations received from 
 Iceland give only the readings of the barometer and of the 
 dry-bulb thermometer, the direction and force of the wind, and 
 the state of the weather. The telegraphic reports from the 
 Azores are furnished through the courtesy of the Portuguese 
 Government and of the Eastern Telegraph Company and the 
 Commercial Cable Company. 
 
30 
 
 METEOROLOGY 
 
 Through the courtesy of the Lords Commissioners of the 
 Admiralty occasional reports of observations at sea off our 
 southern and western coasts are transmitted by wireless telegraphy 
 from the ships of H.M. Navy. 
 
 The remaining 29 telegraphic reporting-stations are scattered 
 throughout Great Britain and Ireland and the adjacent islands. 
 At these stations observations are taken at 7 a.m. and 6 p.m. 
 West European time, and are telegraphed to London according 
 to a special cipher code, which will be hereafter explained (see 
 p. 38). Lisbon observes at 8 a.m. local time, and the Azores 
 at 6 a.m. local time. In addition, observations are taken at 
 1 p.m. daily at certain home and foreign stations, and are at 
 once telegraphed in cipher to London. There is a home-station 
 in St. James's Park, London. 
 
 The following table shows the stations which were working in 
 co-operation with the Meteorological Office, London, in 1909. 
 The facts are taken from the annual report of the year ending 
 March 31, 1909. Particulars of the numbers of Third Order 
 Stations are added, as these now form an important part of the 
 Office work : 
 
 
 SECOND ORDER 
 STATIONS. 
 
 THIRD ORDER 
 STATIONS. 
 
 TELEGRAPHIC 
 STATIONS. 
 
 
 Total 
 Number. 
 
 It! 
 
 M! 
 
 Reporting 
 through 
 R. Met. Soc. 
 
 Total 
 Number. 
 
 Reporting 
 through 
 S. Met. Soc. 
 
 Reporting 
 through 
 R. Met. Soc. 
 
 Telegraphic 
 in 
 British Isles. 
 
 Foreign 
 Telegraphic. 
 
 Scotland . . 
 
 19 
 
 13 
 
 
 g 
 
 2 
 
 
 7 
 
 __ 
 
 England 
 
 50 
 
 
 
 10 
 
 78 
 
 
 
 8 
 
 14 
 
 
 
 Wales 
 
 2 
 
 
 
 
 
 9 
 
 
 
 
 
 2 
 
 
 
 Ireland 
 
 9 
 
 
 
 
 
 12 
 
 
 
 (i 
 
 
 
 Total . . 
 
 80 
 
 
 
 
 
 108 
 
 
 
 
 
 29 l 1 44 
 
 
 
 
 
 
 
 
 
 1 Of these, 13 observe at 7 a.m., 1 p.m., and 9 p.m., and thus come 
 under the international definition of Second Order Stations. These have 
 not been included in the 80 Second Order Stations. The self-recording 
 observatories are Aberdeen, Glasgow, Stonyhurst, Kew, Valentia, Falmouth. 
 As has already been stated, observatories are maintained independently 
 of the Meteorological Office at Greenwich, Oxford, Bidston, Southport, and 
 Berkhamsted. 
 
BRITISH METEOROLOGICAL OBSERVATIONS 31 
 
 As the reports come in, the information is entered on a chart, 
 which is preserved in the Office, and from which the Daily 
 Weather Report is prepared. This Report fills four large quarto 
 pages, and is arranged as follows : 
 
 Page 1 contains the whole of the sixty-six reports from which 
 the maps for the day (given on page 2) are prepared, and the 
 
 6 p.m. reports of the previous day, together with the maximum 
 and minimum temperatures of the air, the rainfall for the 
 previous twenty-four hours, and for the home stations the 
 previous da.y's sunshine, when available. 
 
 Page 2 contains (1) a chart of North-Western Europe, showing, 
 for 7 a.m. on the date of publication, the distribution of atmo- 
 spheric pressure, the prevalent winds, and the sea disturbance, 
 with necessary explanations, together with a table showing the 
 mean temperature for the month at 8 a.m., derived chiefly from 
 observations extending over the thirty-five years, 1871-1905 ; 
 (2) supplementary charts showing the state of the barometer and 
 the direction of the wind at 7 a.m. and also at 6 p.m. of the 
 previous day ; and (3) notes on the general situation at 7 a.m. 
 
 Page 3 contains (1) a similar chart of North-Western Europe, 
 giving the distribution of temperature at 7 a.m., and the weather 
 prevailing at each station, expressed by letters of the Beaufort 
 Scale, explained below ; (2) notes on the weather of the previous 
 day, together with the general inference to be drawn from the 
 
 7 a.m. observations ; (3) a small map of the Forecast Districts of 
 the British Isles, with an explanation of the storm-signals which 
 are exhibited on our coasts as required ; and (4) forecasts for the 
 twenty-four hours commencing at noon of the day of observa- 
 tion. These forecasts are drawn up for each of the eleven Fore- 
 cast Districts of the British Isles, and indicate the weather likely 
 to be experienced within the coming twenty-four hours. 
 
 On the Weather Charts, on pages 2 and 3, lines are drawn 
 through the places where atmospheric pressure and temperature 
 are respectively equal. The lines of equal pressure are called 
 " isobars " (Greek ros, equal ; /3a/oos, weight) ; those of equal 
 temperature are called " isotherms " (Greek i'o-os, equal ; #e/9/^/, 
 warmth). Isobars are by far the most important element in 
 forecasting, while isotherms play a very subordinate part. The 
 
32 METEOROLOGY 
 
 direction of the wind is marked by arrows, which, fly with the 
 wind. They carry a number of " fleches " proportional to the 
 force of the wind estimated by the Beaufort Scale (see p. 39). 
 The readings of the barometer are corrected for the difference in 
 Gravity between the position of the station and latitude 45. 
 
 Page 4 contains information as to (1) the temperature, weather, 
 sunshine, and rainfall recorded at a large number of British and 
 Irish stations on the afternoon or in the evening of the previous 
 day ; (2) the temperature, weather, and rainfall at a number of 
 additional foreign stations, mainly for the twenty-four hours 
 ended at 7 a.m. of the previous day ; and (3) wireless telegrams 
 received from H.M. ships, or from British and foreign trans- 
 atlantic liners, during the twenty-four hours ended at 9.30 a.m. 
 of the day of observation. 
 
 The information relating to weather is indicated by the follow- 
 ing letters, which constitute the " Beaufort Scale " of Weather : 
 6, blue sky ; be, sky half clouded ; c, sky three parts clouded ; 
 d, drizzling rain ; e, wet air, without rain falling ; /, fog ; g, gloomy ; 
 h, hail ; I, lightning ; m, misty (hazy) ; o, overcast ; p, passing 
 showers ; q, squally ; r, rain ; s, snow ; t, thunder ; u, ugly, 
 threatening ; v, visibility, unusual transparency ; w, dew ; 
 x, hoar-frost ; z, dust-haze, or smoke. 
 
 Records are also received from various additional barograph 
 stations, additional thermograph stations, additional autographic 
 rain-gauge stations, some sixty-one sea-temperature stations, kite 
 or balloon stations furnishing observations of temperature, 
 humidity, and wind in the upper air ; while a hygrograph station 
 at Newnham College, Cambridge, sends in a continuous record of 
 the relative humidity of the air. 
 
 The headquarters for the investigation of the upper air are 
 situated at Pyrton Hill, Oxfordshire, where Mr. W. H. Dines, 
 F.R.S., has equipped a station for pursuing this branch of meteoro- 
 logical research by means of kites and balloons. The subject will 
 be discussed fully in a subsequent chapter (see Chapter XXI.). 
 
 A continuous record of the amount of bright sunshine is re- 
 ceived from 146 stations in the British Isles. Of these some are 
 First or Second Order Stations, whilst from others the sunshine 
 record is alone received. 
 
BRITISH METEOROLOGICAL OBSERVATIONS 33 
 
 Such is the official machinery for the scientific study of the 
 weather in the British Isles. It should be mentioned that the 
 Meteorological Office, situated at 63, Victoria Street, Westminster, 
 London, S.W., is under the management of the Meteorological 
 Committee. The Committee, which was constituted by Minute 
 of the Lords Commissioners of His Majesty's Treasury, dated 
 May 20, 1905, consists of seven members, Mr. W. Napier Shaw, 
 LL.D., Sc.D., F.R.S., the Director of the Meteorological Office, 
 London, being the chairman. The Committee administers an 
 annual parliamentary grant for meteorological purposes. It was 
 15,500 in each of the financial years 1907, 1908, and 1909. The 
 Marine Superintendent of the Meteorological Office is Mr. M. W. 
 Campbell Hepworth, C.B., Commander, R.N.R. ; the Super- 
 intendent of the Statistics and Library Branch is Mr. R. G. K. 
 Lempfert, M.A. Cantab. ; the Superintendent of Instruments is 
 Mr. R. H. Curtis ; and the Chief Clerk and Cashier is Mr. John A. 
 Curtis. The telegraphic address of the Office is "Weather, 
 London/' 
 
 Besides the Meteorological Office, the Royal Meteorological 
 Society and the Scottish Meteorological Society have covered the 
 United Kingdom with a network of climatological and pheno- 
 logical stations. 1 Each of these societies also publishes a journal 
 containing many valuable papers on meteorological subjects. 
 Mention should also be made in this connection of the wonderful 
 system of rainfall observation which the late George J. Symons, 
 F.R.S., organised through years of patient and untiring labour, 
 and which is still carried on with ever-increasing success by Hugh 
 Robert Mill, LL.D., D.Sc., Ex-President of the Royal Meteoro- 
 logical Society. In 1908 the number of perfect rainfall returns 
 published in British Rainfall amounted to 4,538 3,326 in Eng- 
 land, 364 in Wales, 604 in Scotland, and 244 in Ireland. England 
 had about one observer for each 17 square miles, Scotland about 
 one for each 50 square miles, and Ireland only about one for each 
 140 square miles. 
 
 As a type of what a fully equipped meteorological observatory 
 should be, the Fernley Observatory, Southport, may be adduced. 
 
 1 Phenological stations are those at which a registry is kept of natural 
 periodic phenomena connected with the animal and vegetable kingdoms. 
 
 3 
 
34 METEOROLOGY 
 
 It includes three stations, of which the principal is situated in 
 Hesketh Park, Southport (lat. 53 39' 24" N. ; long. 2 59' 3" W.). 
 The Marshside Anemograph Station is situate on the coast, over 
 a mile N.N.E. of the Hesketh Park Observatory (lat. 53 40' 18" N.; 
 long. 2 58' 23" W.). The Barton Moss Evaporation Station lies 
 3 miles inland, about 5J miles S.S.W. of the Hesketh Park Obser- 
 vatory (lat. 53 34' 37" N. ; long. 3 I' 12" W.). Besides these 
 stations there is the Hesketh Park Astronomical Educational 
 Observatory (lat. 53 39' 25" N. ; long. 2 59' 8" W.). All these 
 observatories are under the superintendence of Mr. Joseph 
 Baxendell, F.R.Met.Soc., Meteorologist to the Southport Corpora- 
 tion, who publishes an annual report replete with information. 
 
 In relation to the study of the meteorology of the British 
 Isles allusion should be made to two periodical publications of 
 the Meteorological Office apart from the Daily Weather Report. 
 These are (1) the Weekly Weather Report; (2) the Monthly 
 Weather Report, issued as a supplement to the same. 
 
 The Weekly Weather Report has appeared since the beginning 
 of February, 1878. It is published regularly on Thursdays, and 
 gives a summary of the weather of the week ending with the 
 previous Saturday, intended principally for agricultural and 
 sanitary purposes. A division of the British Islands into twelve 
 districts, which are identical with the forecast districts of the 
 Daily Weather Report, is adopted. The districts are further 
 grouped into extreme north, eastern, and western districts, and 
 extreme south (islands in the English Channel). In its present 
 form the Report contains : 
 
 1. General remarks on the meteorological conditions of the 
 week, with a table describing in words the divergence of the 
 warmth, rainfall, and sunshine experienced in each district from 
 the average for the district for the time of the year ; the dates of 
 occurrence of the highest and lowest temperature of the week, 
 for each station. 
 
 2. A table summarising in numerical form the conditions of 
 temperature, rainfall, and sunshine for each district for the week, 
 the current season, and the calendar year. 
 
 3. A table containing the data from stations from which the 
 values for districts are calculated. 
 
BRITISH METEOROLOGICAL OBSERVATIONS 35 
 
 4. A table containing information for selected stations con- 
 cerning the minimum temperature on the grass and the tempera- 
 ture in the ground. 
 
 5. A table giving information of the temperature of the sea- 
 water at a selection of stations on the coast of the British Isles. 
 
 6. A series of maps or Synoptic Charts, showing the distribu- 
 tion of pressure and wind over Europe and Iceland at 7 a.m. and 
 6 p.m. on each day, and the temperature, weather, and sea 
 disturbances at 7 a.m. each day. The maps for each day are 
 accompanied by a brief account of the distribution of weather 
 for that day and the changes which have taken place. 
 
 7. A table giving the results of observations of the upper air 
 taken by means of kites and balloons. These results include 
 particulars as to temperature, humidity, and wind (direction 
 and force) at various levels. 
 
 For the maps and descriptive account the daily telegraphic 
 reports are used, and are supplemented by the information con- 
 tained in the Bulletin International, published in Paris, repro- 
 ducing meteorological telegrams from the whole of Europe, so 
 that the area represented is much larger than that covered by 
 the Daily Weather Report. 
 
 For the statistical summaries the information from the tele- 
 graphic reporting stations in the British Isles is supplemented by 
 returns of daily observations supplied by volunteer observers 
 from about 110 other stations. Of these twenty-seven supply only 
 the daily amounts of bright sunshine. The Reports contain also 
 tables of " Accumulated Temperature.'* These are designed to 
 give persons engaged in agriculture a better means of estimating 
 the manner in which vegetation is affected by temperature than 
 those afforded by the more usual methods of treating the readings 
 of the thermometer. The tables show for each week, and for 
 the whole period from the beginning of the year, the weekly and 
 progressive values respectively of the combined amount and 
 duration of the excess or defect of the air temperature, above 
 or below a suitably fixed standard, or base temperature. The 
 base temperature adopted is 42 F., as being nearly equivalent 
 to 6 C., which has been considered by Continental writers on 
 these subjects to be the critical value, the temperature above 
 
 32 
 
36 METEOKOLOGY 
 
 which is mainly effectual in starting and maintaining the growth, 
 and in completing the ripening of agricultural crops in a European 
 climate. This base is also convenient as being precisely 10 F. 
 above the freezing-point of water, or the melting-point of ice. 
 
 Accumulated Temperature is expressed in Day Degrees a 
 day degree signifying 1F. of excess or defect of temperature above 
 or below 42 F. continued for twenty-four hours, or any other 
 number of degrees for an inversely proportional number of hours. 
 It has been ascertained, by calculation from a considerable series 
 of hourly observations at various places, that the accumulated 
 temperature may be computed, with a very close approximation 
 to the truth, from the observed difference of the mean of the daily 
 maximum and minimum temperatures from the base alone. 
 
 The Monthly Weather Report for 1908 gives a complete resume 
 of the observations dealt with each month at the Meteorological 
 Office. Each monthly issue contained : 
 
 1 . General remarks on the weather over the British Islands. 
 
 2. Summaries, in international form, of observations made at 
 normal climatological stations at 9 a.m. and 9 p.m., and at 
 telegraphic reporting stations at 7 a.m., 1 p.m., and 9 p.m., or 
 7 a.m. and 6 p.m., as the case may be. The international form 
 has been extended to include information regarding the duration 
 of bright sunshine, the earth temperatures at 1 foot and 4 feet 
 (from 1906), the number of observations of fresh or strong winds 
 (forces 4 to 7 of the Beaufort Scale, from 1906), the number of days 
 of fog (from 1906), and of ground frost (minimum temperatures 
 on the grass, 30 and below, from 1908). Summaries for districts, 
 based on observations at the stations of the Weekly Weather 
 Report, have been given for the elements dealt with in that Report. 
 
 3. Abridged summaries of extremes of temperature, rainfall, 
 sunshine, earth temperatures, and grass minimum temperatures 
 for auxiliary climatological stations. I 
 
 4. A plate of four maps, showing 
 
 (1) The monthly distribution of pressure and winds based 
 
 on the morning observations at telegraphic report- 
 ing stations ; also the average distribution of 
 pressure for the month for the period 1871-1905. 
 
 (2) The movements of depressions. 
 
BRITISH METEOROLOGICAL OBSERVATIONS 37 
 
 (3) The distribution of mean temperature over the land 
 
 and in the coastal waters. 
 
 (4) The distribution of bright sunshine. This map was 
 
 added for the first time in the issue for 1908. 
 
 5. A full-page map, showing, by means of isohyetal lines, the 
 distribution of the month's precipitation. 
 
 These maps have been prepared by Dr. H. R. Mill, the Director 
 of the British Rainfall Organisation, and are based on data from 
 nearly 1,000 stations. 
 
 Monthly summaries for a number of additional stations will 
 be found in the Meteorological Record, issued by the Royal 
 Meteorological Society, or in the Journal of the Scottish Meteoro- 
 logical Society. 
 
 The summaries given in the latter are also printed in the 
 Quarterly Reports of the Registrar-General for Scotland. The 
 meteorological data in the Quarterly Reports of the Registrars- 
 General for England and Wales and for Ireland are abstracted 
 from the Monthly Weather Report. Additional information as 
 to rainfall may be found in the annual volumes of British 
 Rainfall. 
 
 The various serial publications of the Meteorological Office are 
 now grouped together under the general title The British Meteoro- 
 logical Year-Book. For the year 1908 the parts of the Year-Book 
 are as follow : 
 
 Part 1. The Weekly Weather Report, with three appendices 
 and a special supplement. It is issued on Thursday of each 
 week. 
 
 Part 2. The Monthly Weather Report,wiih an annual summary. 
 It is issued on the 27th of each month as a supplement to the 
 Weekly Weather Report. 
 
 Part 3. Daily Observations at Stations of the Second Order 
 and at Anemograph Stations. These observations, which are 
 exclusive of those included in the Daily Weather Report described 
 above, are issued in monthly parts, within about six weeks of the 
 close of each month. 
 
 Part 4. Hourly readings at the four Observatories in con- 
 nection with the Meteorological Office namely, Kew, Falmouth, 
 Aberdeen, and Valentia. These readings are issued in monthly 
 
38 METEOROLOGY 
 
 sections for each Observatory within about six weeks of the close 
 of each month. 
 
 To return to the Daily Weather Report, it may interest the 
 reader to learn something about the composition of the weather 
 telegrams which are sent at 7 a.m. by West European time daily 
 to the Meteorological Office, London. The telegraphic reports, 
 which are addressed simply to " WEATHER, LONDON/' consist of 
 two parts. 
 
 The first part is composed entirely of figures, arranged in groups 
 of five each, in accordance with the Code approved by the Inter- 
 national Meteorological Congress, held at Utrecht in September, 
 1874. The second part consists mainly of words, occasionally 
 mingled with figures in groups, and is designed to throw additional 
 light on the information given in the first part of the message. 
 
 The following numerical scales are used in drawing up the 
 telegrams : 
 
 Wind Direction. The thirty-two different points of the com- 
 pass are supposed to be numbered, beginning with 01 =N. by E. 
 and 02 =N.N.E. (true bearings), to 08 corresponding to E., 16 to 
 S., 24 to W., and 32 to N. According to this scale, S.S.E. would 
 be telegraphed " 14," and W. by N. " 25." At the Vienna meet- 
 ing of the International Meteorological Committee (1873) it was 
 agreed that only sixteen directions should be given in the wind- 
 rose, or table showing the direction of the wind according to the 
 points of the compass. In the case of intermediate directions 
 being observed, it was proposed to count them alternately to the 
 one side or the other. 
 
 Wind Force. This is estimated in accordance with the annexed 
 scale, commonly known as the " Beaufort Scale/' because it was 
 originally drawn up by Admiral Sir F. Beaufort, in command 
 of H.M.S. Woolwich in 1805. The wording has been altered so 
 as to suit the- present use of double topsails. The values which 
 were in use for official purposes at the Meteorological Office in 
 1874 have been superseded by those given in the column of the 
 table opposite, headed " Miles per Hour (1909) " : 
 
 It has recently been decided that for statistical purposes in 
 publications of the Meteorological Office winds of force less 
 than 8 shall not be counted as gales, and to avoid ambiguity 
 
BRITISH METEOROLOGICAL OBSERVATIONS 39 
 
 '(6001) 
 juoq aad sajt^ 
 
 CO I> 
 
 oo - 
 
 3 
 
 00 CO O 
 
 put? 'anoq .tad 8o T im o ^ ^ O *G~< 
 
 -ssaadxa aoj ^po 
 
 A'lisaap 
 
 00 O 0^ 
 
 co ib i> o 
 
 S I 
 
 fl 
 
 f ^ 
 
 it 
 
 ^ -^ 
 
 ill I s 
 
 CO * O CO I> QO O O i-H 
 
40 METEOKOLOGY 
 
 implied by the use of the term " moderate gale " for force 7, the 
 Beaufort description has been modified for use in connection 
 with the Daily Weather Service by the description in italics 
 given in the table for forces 7 and 8 namely, "High Wind " and 
 " Gale " respectively. 
 
 As storms in the British Islands are rarely, if ever, so violent 
 as those in tropical latitudes, great caution should be used in the 
 insertion of extreme figures in the telegraphic reports, such 
 as 12 for the wind and 9 for the sea. 
 
 A careful comparison of the Beaufort estimates with the wind 
 velocities recorded simultaneously by anemometers belonging to 
 the Meteorological Office has shown that the most probable 
 equivalent hourly velocity for expressing individual estimates in 
 miles per hour, or vice versa, agrees very closely with the results 
 calculated by the formula 
 
 = 1-87/53, 
 
 where V is the wind velocity expressed in miles per hour and 
 B the Beaufort number. 
 
 The relation between the wind-pressure and the Beaufort 
 numbers is given by the corresponding formula 
 
 P=0-0105B\ 
 where P is the pressure in pounds per square foot. 
 
 The velocity and pressure equivalents calculated from these 
 two formulae have been included in Table I. 
 
 A less detailed statement of the relation between the Beaufort 
 numbers and the corresponding hourly velocity of the wind is 
 given in the following table : 
 
 TABLE II. RELATION BETWEEN BEAUFORT SCALE AND HOURLY 
 WIND VELOCITY. 
 
 Beaufort Scale Number. 
 
 Corresponding Wind. 
 
 Limits of Hourly Velocity 
 in Miles per Hour. 
 
 
 
 Calm 
 
 Under 1 
 
 1 to 3 
 
 4 5 
 
 Light breeze 
 Moderate wind 
 
 1 to 12 
 13 24 
 
 6 7 
 
 Strong Wind 
 
 25 38 
 
 8 9 Gale 
 
 39 54 
 
 10 11 
 
 Storm 
 
 55 75 
 
 12 
 
 Hurricane Above 75 
 
 1 
 
 
BRITISH METEOROLOGICAL OBSERVATIONS 41 
 TABLE III. SCALE OF SEA DISTURBANCE AND WEATHER. 
 
 0=Dead calm. 
 l=Very smooth. 
 2 = Smooth. 
 3= Slight. 
 
 Sea Disturbance. 
 
 4= Moderate. 
 5= Rather rough. 
 6= Rough. 
 
 i 7 = High. 
 | 8=Veryh 
 I 9=Tremer 
 
 igh. 
 idous. 
 
 Weather. 
 
 0=Sky quite clear. 5 = Rain falling. 
 
 1 = Sky a quarter clouded. 6 = Snow falling. 
 
 2 = Sky half clouded. 7 = Haze. 
 
 3 = Sky three-quarters clouded. 8 = Fog. 
 
 4 = Sky entirely overcast. 9 = Thunderstorm. 
 
 Any other phenomena must be reported in words after the 
 groups of figures, such as "lightning last evening/' "heavy 
 dew," " aurora." In this scale the values to 4 refer to the 
 amount of cloud, not to its density. 
 
 Time. 00 or 24 stands for midnight ; 01 for 1 a.m., and so on 
 every hour to 11 p.m., which is represented by 23. 
 
 Armed with the foregoing scales, the observer, having taken 
 the readings at 7 a.m. (Greenwich time), transmits a telegraphic 
 message to " Weather, London," consisting of six groups of five 
 figures each. Here is an example : 
 
 97622 09549 96228 06253 50046 64485 
 
 The first group contains the reading of the barometer (omitting 
 the first figure of the value), reduced to 32 F. and the mean sea- 
 level, for 6 p.m. on the previous day, and the direction of the 
 wind (true, not magnetic) at the same hour. 97622 is thus 
 resolved into : Barometer, 29'76 inches ; wind, W.S.W. 
 
 The second group gives the force of the wind at 6 p.m. on the 
 previous day, the weather and air temperature at the same hour. 
 09549 thus becomes : Wind force, 9, or a strong gale ; weather, 
 rainy ; air temperature, 49. 
 
 The third group supplies the reading of the barometer at 7 a.m., 
 reduced to 32 F. at mean sea-level, and also the direction of the 
 wind. Thus 96228 becomes : Barometer, 29*62 inches ; wind, 
 N.W. 
 
 The fourth group gives the wind force, weather, and air tem- 
 perature at 7 a.m., for 06253 =wind force, 6, or a strong breeze ; 
 2, half-clouded sky ; dry-bulb thermometer, 53. 
 
42 METEOROLOGY 
 
 The fifth group contains the reading of the wet-bulb ther- 
 mometer at 7 a.m., and the amount of precipitation or rainfall, 
 including melted snow and hail, during the last twenty-four hours, 
 in inches, tenths, and hundredths, omitting the decimal point. 
 For example : 50046 = wet-bulb temperature, 50 ; rainfall = 
 0-46 inch. 
 
 The sixth group gives the maximum and minimum temperatures 
 in the last twenty-four hours, together with the amount of sea 
 disturbance at 7 a.m. At inland stations the last figure is of 
 course always 0. Thus, 64485 means that the maximum tempera- 
 ture in the twenty-four hours ending 7 a.m. has been 64, the 
 minimum temperature has been 48, and the sea is " rather rough " 
 at 7 a.m. (i.e., sea disturbance, 5). 
 
 By the adoption of this code system, a very full report can be 
 condensed into what is equivalent to only six words for, under 
 the Post-Office Regulations, five figures, or a letter preceding or 
 following a group of figures, are counted as only one word, and 
 charged for accordingly. The foregoing information has been 
 culled from the official instructions for meteorological telegraphy, 
 prepared for the use of observers exclusively, in accordance with 
 the International Code adopted at Utrecht in September, 1874. 
 
 The meteorological conditions which possess the greatest 
 interest and value for Medical Officers of Health, from their 
 influence on the prevalence of disease and on the death-rate, are, 
 undoubtedly, temperature, humidity, and rainfall. But as these 
 depend to a large extent on the state of the barometer, the direc- 
 tion and force of the wind, and the condition of the sky as regards 
 cloud, fog, and mist or haze, it is necessary to study the whole 
 group of meteorological phenomena. In subsequent chapters, 
 then, a detailed description of the various instruments required 
 by the observer will be given, and the practical application of 
 the information afforded by them will be explained. 
 
 The instruments required for a Second Order Station or Normal 
 Climatological Station are a standard mercurial barometer 
 reading to '002 inch, maximum and minimum thermometers, 
 dry and wet bulb thermometers, and a rain-gauge. The baro- 
 meter and thermometers must have " Kew " certificates from 
 the National Physical Laboratory. All the four thermometers 
 
BRITISH METEOROLOGICAL OBSERVATIONS 43 
 
 named should be suspended in a properly placed Stevenson 
 thermometer screen (see p. 89). These instruments barometer, 
 four thermometers, and rain-gauge are indispensable, but 
 besides them it is desirable to have also a black-bulb maximum 
 thermometer in vacuo, a bright-bulb maximum thermometer 
 in vacuo, and a minimum thermometer (graduated on the stem, 
 without attached scale) for terrestrial radiation, one or more 
 earth thermometers, an anemometer (if the exposure is sufficient), 
 and a sunshine recorder. 
 
 All the thermometers should be graduated on the stem, and 
 only such instruments should be used as have been verified at 
 Kew Observatory, so that the " index-error " may be known. 
 The height of the cistern of the barometer above mean sea-level 
 must be accurately known, since a difference of level of 1 foot 
 gives rise to a difference in the reading of the barometer of very 
 nearly '001 inch. 
 
 The distance of outdoor instruments from any object such as 
 buildings or trees should be twice the height of the object. A 
 suitable site in an open space 300 feet square would afford a 
 quite satisfactory urban exposure. Roofs are not appropriate 
 sites for meteorological observations. The rain-gauge should 
 stand in a grass plot with its rim 1 foot above the grass-level. 
 The sunshine recorders in use at official stations are of the 
 Campbell-Stokes pattern. 
 
CHAPTER V 
 
 HISTORY, ORGANISATION, AND WORK OF THE UNITED 
 STATES WEATHER BUREAU 
 
 THE development of interest in meteorology in the United States 
 dates from the earliest times of its history. It was apparently 
 Benjamin Franklin who first called attention to the progression 
 of weather from west to east. He noted that a north-easterly 
 storm appeared earlier at Philadelphia than at Boston. 
 
 Thomas Jefferson, afterwards President of the United States, 
 was the first to undertake in that country simultaneous meteoro- 
 logical observations. From 1772 to 1777 he carried on such 
 observations at Monticello with Mr. (afterwards Bishop) Madison, 
 who lived at Williamsburg, both places being in Virginia, and 
 about 120 miles apart. 
 
 With the invention of the telegraph by Morse, in 1837, came 
 the idea of collecting at one place simultaneous observations from 
 different parts of the States ; and on this followed, in about ten 
 years, the idea of charting these instantaneous observations, and 
 deducing from this chart some conclusions as to the future weather. 
 Commodore Maury strongly advocated this plan, and in the early 
 " fifties " it was put into operation by Professor Henry, and 
 continued until the breaking out of the Civil War, when it was 
 discontinued. In 1869 Professor Cleveland Abbe, then Director 
 of the Cincinnati Observatory, undertook the collection of data 
 and the forecasting of the weather for the Cincinnati Board of 
 Trade. The data were collected free of cost by the Western 
 Union Telegraph Company, and the map employed by Professor 
 Abbe was made up in the local office of that company by the 
 manager at Cincinnati. 
 
 In the meantime very great interest was being taken in the 
 
 44 
 
THE UNITED STATES WEATHER BUREAU 45 
 
 same direction by Professor I. A. Lapham, of Milwaukee, Wisconsin, 
 and it was perhaps Professor Lapham who personally interested 
 a prominent member of Congress from Wisconsin, the Hon. H. E. 
 Paine, in the matter, and he finally introduced into Congress the 
 Bill which, on becoming law, created the Weather Service of the 
 United States. In the session of February 9 to April 20, 1870, a 
 joint resolution was passed by Congress, which required the 
 Secretary of War to take meteorological observations at the 
 military stations in the interior of the continent, and at other 
 points in the States and territories of the United States, and to 
 give notice on the northern lakes and on the sea-coast, by mag- 
 netic telegraph and marine signals, of the approach and force of 
 storms. At the same session an appropriation of $15,000 was 
 made to carry into effect the foregoing resolution. This work was 
 placed by the Secretary of War in the hands of the Chief Signal- 
 Officer, as it involved questions of signalling, and had been recom- 
 mended by him. At that time General Myer was Chief Signal - 
 Officer, and it was very fortunate for the meteorological service 
 of the States that it was first placed in such energetic hands. 
 
 Annual appropriations were made thereafter for the service, 
 and in terms enlarging its scope, until it was transferred, in 1891, 
 to the Department of Agriculture as the Weather Bureau. 
 
 The first weather bulletin of the new service was issued Novem- 
 ber 1, 1870, and the first storm warning a week later. The first 
 weather map appeared on January 1, 1871. This was not the 
 earliest weather map published, for others had been issued 
 previously under the direction of Professor Abbe, and even 
 before that weather maps had been started elsewhere. In 1861 
 Admiral FitzRoy inaugurated the British system of weather 
 charts and forecasts. On September 16, 1863, Leverrier, in 
 Paris, began the publication of a French series of weather maps, 
 which have continued without interruption from that day to 
 this. The American series was the third of those which are now 
 issued daily in various parts of the world. 
 
 On October 1, 1890, a Bill was finally passed which provided 
 that " the civilian duties now performed by the signal corps of 
 the army shall hereafter devolve upon a bureau, to be known as 
 the Weather Bureau, which, on and after July 1, 1891, shall be 
 
46 METEOKOLOGY 
 
 established in and attached to the Department of Agriculture." 
 In accordance with law the service was so transferred, and Mr. 
 Mark W. Harrington, then Professor of Astronomy and Director 
 of the Observatory at the University of Michigan, and founder 
 and editor of the American Meteorological Journal, was placed in 
 charge. 
 
 By the terms of the transfer the Chief of the Weather Bureau, 
 under the direction of the Secretary of Agriculture, had charge of 
 the forecasting of weather, the issue of storm warnings, the 
 display of weather and flood signals for the benefit of agriculture, 
 commerce, and navigation, the gauging and reporting of rivers, 
 the maintenance and working of sea-coast telegraph lines, and 
 the collection and transmission of marine intelligence for the 
 benefit of commerce and navigation, the reporting of temperature 
 and rainfall conditions for the cotton interests, the display of 
 frost and cold-wave signals, the distribution of meteorological 
 information in the interests of agriculture and commerce, and 
 the taking of such meteorological observations as may be neces- 
 sary to establish and record the climatic conditions of the United 
 States, or as are essential for the proper execution of the foregoing 
 duties. 
 
 The first Appropriation Bill set aside a considerable sum for 
 the distribution of forecasts to farmers. For this reason, and 
 because of the transfer of the Bureau to the Department of 
 Agriculture, especial attention has been given since the transfer 
 to the more extensive and complete distribution of the forecasts 
 to country communities, and especially to farmers. 
 
 Among the most important dates in the history of the meteoro- 
 logical service of the United States are the following : 
 
 1871, November 13. First exchange of observations with 
 Canada. This continues to the present, and is very helpful in 
 the forecast work of the Bureau. 
 
 1872, January. As authorised by the Appropriation Bill for 
 that fiscal year, the service arranged for reports of the stages of 
 rivers, and in the following spring these were utilised in forecasts 
 of floods. 
 
 1872, September 3. First balloon ascent of the Signal Service 
 for meteorological purposes. The ascent was made by Samuel A. 
 
THE UNITED STATES WEATHER BUREAU 47 
 
 King, aeronaut, and George C. Schaeffer, jun., as meteorological 
 observer. 
 
 1873. In the autumn of this year the report of observations 
 from the West Indies began. 
 
 1875, July 1. On this date began the publication of the 
 bulletins and charts of international meteorological observations. 
 The first was for the date of January 1, 1875. The series was 
 discontinued at the end of 1887, but the monthly and annual 
 summaries were continued to 1889. 
 
 1876. Stations were established at St. Michael's and St. Paul's 
 in Alaska. 
 
 1881-1884. The international polar explorations were begun 
 by the United States in 1881. The Lady Franklin Bay party 
 returned in 1884. 
 
 1881, April 11. The movement for the assistance of State 
 weather services was initiated by a letter of this date. 
 
 1887. In May the first weather crop bulletin was published. 
 
 1887. The marine meteorological service was surrendered to 
 the Hydrographic Office of the Navy. 
 
 1888. A beginning was made in the installation of auto- 
 matic barographs and thermographs at stations, thus enabling 
 hourly readings to be made of the principal meteorological 
 elements. 
 
 1889. The publication of the Bibliography of Meteorology was 
 begun by the appearance of the first part in this year. 
 
 1891, July. The system of Local Forecast Officials was first 
 put in operation. 
 
 1892, Spring. The special investigation of the Great Lakes 
 was begun. 
 
 1892, Summer. The first systematic study of thunderstorms 
 by the meteorological service. 
 
 1892, August. The first meeting of the Association of State 
 Weather Services. 
 
 1893. Continuous practice work by all forecasters was intro- 
 duced ; the competitive idea for filling professorships with 
 accomplished forecasters was adopted ; the Flood Section was re- 
 organised, and local predictions were placed in the hands of local 
 forecast officials ; the first current chart of the Great Lakes was 
 
48 METEOROLOGY 
 
 issued ; the first annual volume in the form fulfilling the inter- 
 national requirements was published. 
 
 The various bureaux in Washington City are, with one or 
 two exceptions, directly under a member of the Cabinet. The 
 Weather Bureau is under the Secretary of Agriculture. As in 
 the other bureaux, under their proper Secretaries, he can dictate 
 the policy of the Bureau, and can appoint or dismiss any or all 
 employees, with the exception of the Chief of the Bureau, who 
 is appointed by the President and confirmed by the Senate ; and 
 even in this case the wishes of the Secretary would always receive 
 favourable consideration. While the officers of the Bureau may 
 be changed at the will of the Secretary, as a matter of fact such 
 change is infrequent, except in the force of messengers and 
 labourers. The chief officers of the Bureau are continued with 
 little reference to changes of administration, and in the technical 
 observing force such change is practically unknown. 
 
 The appropriations for the Bureau are made annually by 
 Congress, and are a part of the appropriations for the Department 
 of Agriculture. An estimate is carefully made by the officers of 
 the Bureau some months before the session of Congress in which 
 the appropriation must be made, and about a year before the 
 appropriation can become available. This estimate is submitted 
 to the Secretary of Agriculture, and after receiving his approval, 
 passes to the Committee on Appropriations of the House of Repre- 
 sentatives. On receiving their approval it is submitted to 
 Congress. 
 
 Among the various divisions of the central office at Washington 
 stands foremost the Forecast Division. It has charge of fore- 
 casts, of floods, of the telegraphic section, of storm signals, and 
 of the practice which is continuously performed by the forecasters. 
 The forecasts are made twice a day, immediately on receipt of 
 the telegraphic reports of observations from the regular telegraphic 
 stations. As soon as the forecasts are made, the maps are printed 
 in the Bureau office, and the forecasts are given to the Associated 
 and United Press Companies, by which they are distributed over 
 the United States. Each forecast contains statements concerning 
 the weather for divisions of the United States, each division being 
 usually a State, or a large part of the State. The forecast officials 
 
THE UNITED STATES WEATHER BUREAU 49 
 
 on duty at Washington are kept in constant practice. They 
 generally have the rank of professor. Besides this practice, 
 which occupies a fractional part of the day, each professor is 
 entrusted with other and important duties. One has charge of 
 the instrument-room and of all duties relating to instruments. 
 Another has charge of the Monthly Weather Review, which he 
 edits, and also of a great variety of duties relating to theoretical 
 and scientific meteorology. Another has charge of the collection 
 of statistics concerning tornadoes and other destructive storms. 
 And the fourth is entrusted with the special investigation of the 
 relation of meteorology to magnetism. The telegraph section 
 under this division has a force of operators who receive and send 
 telegrams, and have charge of the various coast telegraph lines 
 belonging to the Weather Bure r u. 
 
 The State Weather Service Division is in charge of the weather- 
 crop work, the thunderstorm work, the distribution of tempera- 
 ture and weather signals, and the snow charts of the Bureau. In 
 general its work is essentially climatological. It is the centre 
 of the State Weather Services scattered over the United States. 
 It receives weekly during the crop season the weather-crop 
 reports from the State centres, and digests them into the Weather- 
 Crop Bulletin, which is immediately sent to press, and appears 
 ready for distribution the day the reports are received. It pre- 
 pares and sends to the State centres, for distribution from those 
 points, the signals for temperature and weather, which are 
 intended in general for the agricultural communities ; and in a 
 manner similar to the weather-crop bulletins, during the winter 
 season it collects the data for snow-charts, and has the charts 
 printed and distributed the day the data are collected. It is 
 also in charge of the special thunderstorm work, this work being 
 done through the machinery of the State Weather Services. 
 
 The Records Division is entrusted with the care of the records, 
 with the compilation of data of all sorts required for the work of 
 the Bureau or by the general public, and with the publication of 
 reports, more especially the annual reports made for general 
 information. The accumulation of records in charge of this 
 division has, after very many years of work, become extremely 
 great, and includes not only the records of the meteorological 
 
 4 
 
50 METEOROLOGY 
 
 service which finally ended in the Weather Bureau, but also of 
 that meteorological service which was carried on previously by 
 the Smithsonian Institution, and that also by the Surgeon-General, 
 which to some extent preceded that of the Smithsonian. These 
 records are kept in a fireproof vault in such form as to be readily 
 accessible. Other private records have been added to these, 
 either by purchase or gift, until the collection forms by far the 
 most complete record of climatological interest to be found in 
 the United States so complete, in fact, that its use is entirely 
 indispensable to anyone who wishes to make a competent study 
 of any feature of the weather or climate of the States. In the 
 compilation of data for the general public a great deal of time 
 is spent by the Records Division. All sorts of questions relating 
 to all sorts of features of the weather and climate come con- 
 stantly to the Bureau. The replies must be made with very great 
 care, so as to be thoroughly authentic. No less important is its 
 duty in checking observations and in detection of errors. The 
 system is so complete that the errors are charged up against the 
 individual observers, and at regular times a statement is issued 
 giving the names of the observers who have been the most free 
 from errors in the preceding interval. 
 
 To the Instrument Section is entrusted the question of pur- 
 chase and shipment of instruments, their testing when received, 
 their condition at stations, their repairs when needed, and the 
 examination of the automatic records as they are received from 
 the various stations. It is also occupied with devising new forms 
 of instruments and new methods of taking observations, and 
 performs a large amount of work in physics, more or less directly 
 connected with this purpose. 
 
 The Publications Division is in charge of the publishing and 
 mailing of the material of the Weather Bureau. Those matters 
 which are urgent are published in the Bureau office. In addition 
 to that there are a few other publications made by the Bureau 
 office which are not urgent, but are used to fill up the intervals 
 of time on the part of the compositors and pressmen. Most of 
 the other matter issued by the Bureau is published by the Govern- 
 ment Printing Office at a fixed price, the same being deducted 
 from the appropriation for the Bureau. In a' few special cases 
 
THE UNITED STATES WEATHEK BUREAU 51 
 
 the publications of the Bureau are made by joint-resolution of 
 Congress, in which case there is no charge against the Appropria- 
 tions of the Bureau. The publications prepared in the Bureau office 
 are the maps of all sorts, the reports requiring immediate distribu- 
 tion, and special publications of the same sort. Among the other 
 publications not matters of so much urgency, but actually pub- 
 lished in the Bureau are many of the innumerable " forms " used 
 in the collection and distribution of data, the Monthly Weather 
 Review, which is prepared and printed entirely within the chief 
 office at Washington ; and some of the series of printed bulletins 
 in octavo form issued by the Bureau. Also, lithographed maps 
 and charts, though not urgent, are usually printed by the Bureau. 
 This division has in its charge a draughting-room, in which maps 
 are prepared for printing and other necessary drawings are made ; 
 and the composing-room, in which a considerable number of 
 printers are employed. In this division labour is saved in all 
 possible ways, the most notable one, perhaps, being that of 
 the use of a long series of logographs. In the publication of 
 weather tables and forms the same combination of letters and 
 figures frequently recurs, and in this case a single type has been 
 cast to include these letters and figures. In this division is also 
 included the press-room. A large part of the work of the Bureau 
 is lithographed, because the lithograph makes the cheapest, 
 easiest, and readiest means of publishing urgent data in charto- 
 graphic form. As a result a force of lithographers is kept in 
 connection with the press-room. Also connected with the divi- 
 sion are the folding and stitching room and the mailing-room. 
 The mailing lists are kept with care, in order to economise the 
 publications of the Bureau. 
 
 Connected with the general office at Washington is also a library, 
 containing a number of books and pamphlets, so arranged as to 
 show its annual growth. The library also contains many meteoro- 
 logical and some geographical charts. 
 
 These books are obtained in considerable part by exchange 
 with other Meteorological Services and with various Governments, 
 in part also by gift, but in large part by actual purchase. The 
 result is a technical and special library of unusual size and value. 
 So large and complete is it that with the aid of correspondents 
 
 42 
 
52 METEOROLOGY 
 
 from all over the world the librarian has undertaken to publish 
 a bibliography of meteorology. 
 
 In 1884 the Signal Office began the compilation from the 
 printed literature comprising the books, pamphlets, memoirs, 
 and papers in serial publications of all kinds, relating to meteor- 
 ology and its applications. In 1887 the number of titles collected 
 and classified was about 50,000. 
 
 In addition to these various sections of the Bureau, there is 
 a large mass of correspondence to be cared for. This comes under 
 the direct cognisance of the Chief Clerk, who has a small force of 
 clerks under him to perform this work. The letters received are 
 assigned to the various divisions or individuals most competent 
 to answer them. The replies are drawn up by these divisions or 
 individuals, with the aid of stenographers, and sent to the Chief 
 Clerk for supervision or signature before being mailed. Corre- 
 spondence, manuscripts, and other papers of importance are 
 passed over to a special official called the File Clerk, who has 
 entire charge of the files of the Bureau, and whose duty it is to 
 speedily find any paper of any date required by any officer, and 
 to return it to its place when it comes back to him. 
 
 Especial attention is also paid to comments and criticisms of 
 the Bureau from whatever source they come, and to this duty part 
 of the time of one clerk is devoted. The observers at stations are 
 instructed to send to the central office all comments on the 
 Bureau, and especially criticisms of it. This serves the double 
 purpose of keeping the Bureau in close touch with the popular 
 wants, and of informing the central office of the way in which these 
 wants are filled at the stations. These matters are kept filed in 
 such a way as to be ready for reference. 
 
 What precedes relates only to the central office in Washington 
 City. There is also a large number of employes scattered at 
 numerous stations over the entire United States. These stations 
 are : Regular telegraphic stations, stations in the West Indies 
 (excluding stations in Canada, with which only an exchange is 
 carried on, these stations being controlled by the Canadian 
 Service), river and flood reporting stations, voluntary stations, 
 mountain stations, telegraph-line repair stations, storm-signal 
 stations, temperature and weather stations, and stations in the 
 
THE UNITED STATES WEATHER BUREAU 53 
 
 cotton, rice, and sugar regions ; also stations for special reports 
 of thunderstorms and others. 
 
 The River and Flood Section, under the Forecast Division, 
 has especial charge of the river gauges, which are scattered at 
 frequent intervals up and down the principal rivers, from which 
 reports are received at regular intervals. The duty of the River 
 and Flood Section is the forecasting of the conditions of the rivers, 
 more especially during the period of flood. This work has hereto- 
 fore been entirely done in the central office at Washington, and 
 was in the hands of one officer, who was alone entitled to make 
 forecasts concerning the state of the rivers. It has been found, 
 however, that it is impracticable for a single officer to obtain the 
 intimate familiarity with, and keep in mind the current knowledge 
 of, all the details that are necessary for safe forecasting the entire 
 length of important rivers. There are so many local conditions 
 the width of the river, opportunity for set-back so many condi- 
 tions that may happen accidentally, as when a levee breaks, or 
 when at a certain height the river may pour over the banks into 
 an empty space at one side that it has been found necessary to 
 divide up the work among a considerable number of men, who are 
 stationed in the vicinity of the places where the forecasts are to 
 be made. These men become familiar with the details mentioned 
 above, and hence can perform the work more satisfactorily. This 
 policy has been introduced since the season of the floods of 1893. 
 It is confidently expected that its success will in time be much 
 more considerable than has been experienced heretofore. 
 
 The cotton, rice, sugar, and other special services are intended 
 for the protection of specific crops grown in limited areas. Special 
 reporters are scattered through these regions, who report with 
 special reference to the climatological needs of these individual 
 crops. For the cotton interest this service has been found to 
 be quite successful. There has not been so good an opportunity 
 to test it for the rice and sugar interests. 
 
 Forecasts. 
 
 The main duty of the Weather Bureau is the forecast of the 
 weather, and special attention is paid, therefore, to this part of 
 
54 METEOKOLOGY 
 
 the work. The other features of the work of the Weather Bureau 
 are incidental to this. 
 
 At the hour of eight o'clock, 75th Meridian Time, the observers 
 of the Weather Bureau all over the United States, at each tele- 
 graphic station, proceed to take their observations. These 
 observations are taken, so far as possible, in exactly the same way 
 at each station, with similar instruments, and with exactly the 
 same precautions against error and corresponding provisions for 
 their correction. As soon as these observations are taken a 
 proceeding which usually occupies the observer but a few moments 
 they are at once reduced to the form of a telegram and expressed 
 in the words of the telegraphic " code " employed by the Bureau. 
 They are then promptly taken to the telegraph office, and at once 
 sent on to the Bureau. For a few moments after eight o'clock, 
 morning and evening, all over the United States, all other tele- 
 graphic business gives way to the business of the Weather Bureau. 
 The telegrams are at once forwarded, in order to reach the central 
 office in Washington at the earliest possible moment. They are 
 forwarded, however, in such a way that they can be dropped on 
 their passage at other stations where they are needed. That is 
 to say, they are collected in " circuits," and the telegram for each 
 station in the circuit is dropped at each of the stations where it 
 may be of use. The result is, that the central office at Washington 
 is furnished with the observations from all over the United States > 
 and the individual stations wishing to have them are furnished 
 with the observations which they need. They come into the 
 central office at Washington in circuits one after the other the 
 Southern circuit, the New England circuit, and so on. They are 
 received in the Bureau office, about an hour after they are taken, 
 by operators employed for the purpose, and are at once taken off 
 on the typewriter, and sent by messenger to the forecast-room. 
 On reaching the forecast-room, they are passed to an official, 
 called a translator, whose duty it is to read in ordinary language 
 the telegram expressed in the code, so that the clerks who surround 
 him, and the forecast official, can obtain the information as rapidly 
 as possible. Of these translators there are several in the office, 
 and at each of the outside stations, where a considerable number of 
 the reports are used, one man at least must be expert in the code- 
 
THE UNITED STATES WEATHER BUREAU 55 
 
 This takes us to the forecast-room, which is a very busy 
 place for about two hours after nine o'clock in the morning 
 and nine o'clock in the evening, 75th Meridian Time. This 
 room has a body of clerks to take down the data on indi- 
 vidual maps. At the same time a small force of printers 
 proceed to set up the tables used on the maps, doing this directly 
 from the reading of the translator. One of these is occupied with 
 setting up the symbols for wind direction and weather, as they 
 will appear eventually on the finished map. As rapidly as the 
 translator reads the telegrams the data are placed on maps, of 
 which there are four, the principal map (which is the proper 
 weather map used in the regular forecasts) and three auxiliary 
 maps made by the clerks as the observations come in, and used 
 by the forecast official to aid him in making the official forecast. 
 The map proper, or weather map, which is afterwards published, 
 is made by the forecast official himself, or under his immediate 
 supervision. As soon as these maps are finished and that is 
 almost immediately after the reading of the last telegram received 
 the forecaster proceeds at once to dictate his forecasts to a 
 stenographer beside him. These are made for separate States, 
 or for halves or quarters of larger States, and must be made in a 
 certain fixed order, in which order they are always printed. As 
 they are taken by the stenographer, the compositors set them up, 
 and almost as soon as they are finished by the forecast official 
 they are in print, and the proof copy is taken off. This is read 
 by the forecaster before he leaves the room. After making the 
 forecasts he also decides to what points special signals shall be 
 sent indicating high winds or storms, and gives orders to this 
 effect. 
 
 The principal map, being finished in this way as a manuscript, 
 is taken at once to the lithographer, transferred to stone, placed 
 on the press by pressmen who are waiting to receive it, and run 
 off with all possible celerity. Messengers stand in waiting, so 
 that the first few copies are taken in hand, carried to the trains, 
 or to the points where they are to be left, by means of bicycles 
 or otherwise, so that with all possible despatch the maps are dis- 
 tributed as soon as they are printed. In the meantime the 
 agents of the great news-collecting agencies are at hand to receive 
 
56 METEOEOLOGY 
 
 the forecasts, which they distribute by telegraph to various parts 
 of the United States. The time which elapses from the taking of 
 the observations until the map is finally ready for the messengers 
 varies from two and a half to three hours, depending upon circum- 
 stances. It is rare for it to reach three hours. The usual time 
 is two hours and thirty-five minutes to two hours and forty-five 
 minutes. 
 
 The forecasts were formerly made for the day or the night on 
 which the observations were received. This being the case, not- 
 withstanding the speed that was used in getting them before the 
 public, the period for which they were intended was partly passed 
 before the forecasts could reach their readers. This has now been 
 changed, so that the forecasts that are made from the morning 
 observations, as well as those from the evening observations, are 
 intended for the next day. This amounts to making forecasts 
 for thirty-six hours ahead, instead of twenty-four, and even 
 twelve, as was formerly the case. This has been found to be much 
 more satisfactory to the public, and as a result forecasts can 
 appear in the evening papers which are intended for the next day, 
 so that those interested in the weather of that day can have 
 abundant time to make their preparations. 
 
 After the forecasts have been made comes the question of veri- 
 fication. This is done systematically, both for the official fore- 
 caster in Washington and for those who are on " practice-forecast/' 
 It is also done, from time to time, for the local forecast officials 
 and other forecasters at stations. It is usually done by a series of 
 rules of highly elaborate and technical character, which rules are 
 printed in a code of " Special Instructions to Forecasters/' and 
 with which the forecaster is expected to be entirely familiar. 
 Under these rules precise definitions are given to matters which 
 can be forecasted. Limits are given to the rise and fall of tem- 
 perature, definition is given to rain, to " cold wave/' and other 
 matters which are subject to forecasting. From these somewhat 
 complicated rules a series of averages is drawn up, and these make 
 the rating which the official receives. It has been found by con- 
 siderable experience that high ratings and public satisfaction are 
 not necessarily concurrent. The rules for forecasting are so 
 technical as to confine the forecaster to a limited range of 
 
THE UNITED STATES WEATHER BUREAU 57 
 
 precise expressions, and require in each case a definite forecast, 
 though the forecaster may be unwilling to hazard it on account of 
 uncertain conditions. In cases of this kind it is really better for 
 the public use to give the degree of possibility, probability, or 
 uncertainty under which the forecaster labours from the informa- 
 tion which he has in hand. By so doing, however, he loses his 
 high grading in the verification, so that he stands between the 
 Scylla, on the one side, of precise verification, and the Charybdis, 
 on the other, of complete comprehension by the reading public, 
 and he must make his way between them the best he can. It is 
 said to sometimes result in what is called " hedging " in fore- 
 casting, where expressions are made with special reference to their 
 values in verification. There is no set of rules that can be drawn 
 up that will absolutely prevent this. Thus one may be given a 
 high verification, but fail in usefulness to the public. The custom 
 has therefore grown up of not paying such close attention to the 
 official rating of verification obtained by the forecaster as to the 
 satisfaction shown by the public in the newspapers and elsewhere. 
 Forecasters have been encouraged to state, without technical 
 limitations of language, exactly what they expect to occur from 
 the data in hand, and to give the public all the information which 
 they have in language which, while condensed, may freely express 
 the amount of confidence which they have in their predictions. 
 
 There is also made up and printed with the forecasts a summary 
 of changes in the weather over the United States in the last 
 twenty-four hours, and this is of considerable public interest, 
 and is consulted perhaps as much as the forecasts themselves. 
 By the use of the weather maps and constant consultations of the 
 forecasts, many readers have become so skilful that by means of 
 the summary they can make their own forecasts for their definite 
 purposes, and with more satisfaction to themselves than that 
 afforded by the official forecasts. 
 
 In the forecasts a series of different things is predicted. There 
 is a general statement as to the probable changes of the meteor- 
 ological elements for each district for which forecasts are made. 
 There is also a prediction for local storms, when the forecaster 
 finds indications of them on the maps. When the statement is 
 made that severe local storms may be expected in any particular 
 
58 METEOROLOGY 
 
 district of the United States, it is intended to convey the possi- 
 bility of the occurrence of tornadoes or cloud-bursts. It is a 
 warning to the public to be on the look-out and to prepare for 
 them. It is thought that in this way the public can be warned of 
 the possibility of the tornado without being terrified by the actual 
 prediction for a quarter of the State, when the tornado will occur 
 in any case in only a very small part of that area. Predictions 
 are also made for high or dangerous winds, and special storm 
 signals are ordered up to convey this information to the public. 
 They are made also for cold waves, for frosts, and for stages of 
 rivers where of interest, and for floods. From the general fore- 
 casts specific ones for weather and temperature are taken, which 
 are distributed to inland and country districts. In some cases the 
 local observer can order up cautionary signals at certain ports, 
 also information signals. The latter are intended to notify 
 masters of vessels that additional information can be obtained by 
 calling at the Weather Bureau station, and that this information 
 is of such a nature as to be of importance to them if they are about 
 to leave port. Special bulletins are occasionally sent out when 
 any dangerous storm is in progress ; and specific information can 
 be given in the interval between the two daily maps. 
 
 Somewhat similar forecasts are made at local or district 
 stations. They are, however, not so elaborate as those made at 
 the central office in Washington; although generally more in 
 detail, they are confined to more limited areas. In general, it 
 is intended that the evening forecasts shall be distributed from 
 the central office from the afternoon observations, and forecasts 
 from the local stations sent out from the morning observations. 
 
 The forecasts are, of course, sometimes criticised, as are also 
 the summaries. It is, however, admitted by the general public 
 that all human institutions sometimes go astray, and serious errors 
 occur so seldom as to be excused and overlooked. It is considered 
 more harmful to the public interest to alarm a large area over a 
 doubtful storm than to refrain from mentioning it. The forecast 
 field of the United States is, on the whole, a more favourable one 
 than that of any other country on the globe. It extends from 
 ocean to ocean in the middle latitudes, where storms are more 
 frequent. The storms that come in from Canada can also be 
 
THE UNITED STATES WEATHER BUREAU 59 
 
 noted before their arrival by means of the telegrams sent to the 
 Bureau from the Canadian stations. Practically the Weather 
 Bureau has an outlook over the entire field of the North American 
 continent, north of Mexico and south of latitude 50. Over this 
 field the observations are taken at one simultaneous instant. The 
 maps can be made for this large area with more accuracy than can 
 be done in Europe, where a number of weather services occupy 
 a relatively small territory, while observations are taken in 
 different countries at different hours. The only field for meteoro- 
 logical work which approximates that of the United States is that 
 of Australia, but in this case the dry centre of the island or conti- 
 nent disturbs the progress of cyclones over it much more than 
 they are disturbed in the United States by the Rocky Mountains 
 and the dry plains. 
 
 " Practice forecasts " are carried on by official forecasters not 
 on duty every day in the year. They make forecasts exactly as 
 if they were making them for the general public. They are veri- 
 fied in exactly the same way. To these gentlemen are also en- 
 trusted a series of special problems depending directly upon fore- 
 casting. These they work out in detail from the maps already on 
 file, and report. Their reports are taken into account in future 
 forecasts. 
 
 The items telegraphed from the stations are (in their proper 
 order) : name of station, the corrected readings of pressure and 
 temperature, the direction of wind, state of weather and pre- 
 cipitation, current wind velocity, and minimum or maximum 
 temperature, report of observations of frost, dew-point, upper 
 clouds, lower clouds except forms of nimbus moving with the 
 surface winds maximal wind velocity and direction, .and special 
 monthly reports when required. This will all be included in eleven 
 words, by the telegraphic code. The code employed by the Bureau 
 is of very ingenious construction, and a long trial has proved it to 
 be very complete and satisfactory. It is dissimilar from any other 
 code in use, and has been invented by employes for the purposes 
 for which it is used. This code is intended only for the use of the 
 Weather Bureau, and is not understood by operators generally. 
 It requires on the part of the observers much study to become 
 familiar with it. 
 
60 METEOROLOGY 
 
 The river and flood work is also under the Forecast Division. 
 In the case of this work the difficulties are very great. Each 
 river has a regimen of its own its own peculiarities, idiosyn- 
 crasies, and characteristics. These must be learned for each 
 river, and they depend largely on the size and character of the 
 basin which the river drains. There is also a difference in the 
 effect of precipitation on the height of the river, with the season 
 at which the precipitation occurs, and with the state and character 
 of the surface of the river basin. Should heavy rains fall in 
 August, after dry weather, they will have a very different effect 
 on the river from that of a heavy rainfall in the spring, when the 
 ground is frozen and the rain carries off with it also the melted 
 snow. There is also a series of difficulties of peculiar character 
 in trying to ascertain and forecast the result of the meeting of 
 flood-crests of rivers. For instance, when the Ohio has a flood- 
 crest at Cincinnati, its tributary, the Cumberland River, has 
 its own flood-crest, and the Tennessee River has still another. 
 These all come into the Lower Ohio, between Cincinnati and the 
 mouth of the river at Cairo. It is a matter of extreme difficulty 
 to know what the result will be on the crest already existing in 
 the Ohio. Will it be accelerated in its progress, or retarded ? 
 Will it be heightened, or lengthened ? These are among the 
 questions which it is necessary to decide under such circumstances, 
 and the decision of which presents very considerable difficulties. 
 Other disturbances of the flood-crest may occur, as, for instance, 
 when it reaches a certain height the surface water may pour out 
 into some pocket from which the river is separated at lower levels. 
 This is the case with the great St. Francis marshes, which lie 
 in South-Eastern Missouri and North-Eastern Arkansas. Fairly 
 cut off from the main river when the water is low or moderate, 
 they are easily accessible to the water when the Mississippi is 
 high, and amount to an enormous reservoir, which receives the 
 water when high and gives it out slowly, and at a later date. 
 
 To aid the observers in their work elaborate river gauges have 
 been placed on the rivers, distributed as experience has found to 
 be most necessary. This is the case with most of the important 
 rivers at the present time, but the service has not reached as yet 
 the streams of secondary importance. It is being gradually 
 
THE UNITED STATES WEATHER BUREAU 61 
 
 extended, and as time passes these gauges will be found also on 
 the latter streams, and the service will be extended to all river 
 basins of importance in the area covered by it. 
 
 Even with this complete apparatus floods occur for which the 
 Bureau can hardly hope to make successful forecasts. As an 
 illustration of these floods, the results of a cloud-burst may be 
 mentioned. J?or instance, some years ago one occurred on the 
 side of Pike's Peak and the neighbouring mountains. The water 
 poured down a stream of such volume that miles of railway were 
 carried away, and occasionally the steel rails were bent and 
 twisted by the force of the water, and this through a river bed 
 usually dry. Another illustration of this class is to be found 
 in the Johnstown disaster in Pennsylvania. A large reservoir 
 was sustained near the head of a comparatively insignificant 
 stream. This reservoir had a high retaining wall, and had 
 existed for years. Rains in the mountains of rather unusual 
 character, but not altogether exceptional, carried away the re- 
 taining wall, and the result was a fearful disaster to the town on 
 the river below. To forecast such a flood as this it would be 
 necessary to have a constant watch kept of the dam enclosing 
 the reservoir. A watchman was employed, but the yielding of 
 the dam was so sudden that it was foreseen by him but a short 
 time before it actually occurred. 
 
 Distribution of Forecasts. 
 
 The problems of the distribution of forecasts, after they are 
 made, are quite as important as those involved in making them, 
 and are in some respects more novel. The means actually em- 
 ployed in distributing the forecasts in the meteorological service 
 of the United States are, first, the use of the news-collecting 
 agencies of the newspapers. 
 
 The second way of communicating the forecasts to the public 
 is by means of the storm and cautionary signals. These are 
 under the direct control of the forecaster, and he decides after 
 each forecast to which station the telegrams ordering the observer 
 to display these signals shall be sent. There are a number of 
 minor stations for the purpose of display only, and the signal 
 ordered to be displayed at the centre station is also to be displayed 
 
62 METEOROLOGY 
 
 CHART A. 
 U.S. DEPARTMENT OF AGRICULTURE, WEATHER BUREAU. 
 
 CAUTIONARY SIGNALS. EXPLANATION or CAU- 
 
 (Displayed only on the Grea t Lake., 
 
 ^ . 
 
 i E. fr I 
 
 |0H|| 1^^^. a wllite centre, displayed at 
 
 ! iHHH stations on the Great Lakes, 
 
 'MM H"l indicates that the winds 
 
 expected will not be so 
 severe but well-found, sea- 
 worthy vessels can meet 
 N.W. winds. S.W. winds. N.E. winds. S.E. winds. them without danger. 
 
 STORM SIGNALS. A . red . fla g . with a black 
 
 centre at stations either on 
 
 (Displayed both on the Lakes and seaboard.) the Lake or seaboard indi- 
 cates that the storm is ex- 
 pected to be severe. 
 
 The pennants displayed 
 with the flags indicate the 
 direction of the wind ; red, 
 easterly (from north-east 
 
 N.W. winds. s"w. winds. N.V winds. S.E. winds. to south) ; white, westerly 
 
 (from south-west to north). 
 
 The pennant above the flag indicates that the wind is expected to blow 
 
 from the northerly quadrants ; below, from the southerly quadrants. 
 
 INFORMATION SIGNAL. 
 
 The Information Signal consists of a red pennant, and indicates that the 
 local observer has received information from the central office of a storm 
 covering a limited area, dangerous only for vessels about to sail to certain 
 points. The signal is intended to be a notification to 
 shipmasters that valuable information will be given 
 them upon application to the local observer. 
 
 By night a red light will indicate easterly winds, and 
 a white light below a red light will indicate westerly 
 winds. 
 
 i Signal. The system of weat;her> temperature, and rain sig- 
 nals displayed throughout the country is distinct from the cautionary and 
 storm signals, the latter being principally for the information of maritime 
 interests. They are displayed at the principal ports of the Great Lakes, and 
 on the Atlantic, Pacific, and Gulf coasts. 
 
 HURRICANE WARNINGS. 
 
 Two red flags, with black centres, displayed one above the other, indi- 
 cate the expected approach of tropical hurricanes, and also of those 
 extremely severe and dangerous storms which occasionally 
 move across the Lakes and the Northern Atlantic coast. These 
 warnings are displayed at all Weather Bureau Stations on the 
 Atlantic and Gulf coasts of the United States, and on the fol- 
 lowing islands in the Atlantic : Jamaica, Santo Domingo, Turks 
 Island, Bermuda, Haiti, Curagao, Porto Rico, St. Kitts, 
 Dominica, Barbados, Trinidad, Cuba. Hurricane warnings are 
 not displayed at night. 
 
 The flags employed are 8 feet square. The pennants are 5-feet hoist, 
 12-feet fly. 
 
THE UNITED STATES WEATHER BUREAU 63 
 
 CHART B. 
 U.S. DEPARTMENT OF AGRICULTURE, WEATHER BUREAU. 
 
 EXPLANATION OF FLAG SIGNALS. 
 No. 1. No. 2. No. 3. No. 4. No. 5. 
 
 Fair Weather. Rain or Snow. Local Rains. Temperature. Cold Wave. 
 
 INTERPRETATION OF DISPLAYS. 
 
 No. 1, alone, indicates fair weather, stationary temperature. 
 
 No. 2, alone, indicates rain or snow, stationary temperature. 
 
 No. 3, alone, indicates local rain, stationary temperature. 
 
 No. 1, with No. 4 above it, indicates fair weather, warmer. 
 
 No. 1, with No. 4 below it, indicates fair weather, colder. 
 
 No. 2, with No. 4 above it, indicates warmer weather, rain or snow. 
 
 No. 2, with No. 4 below it, indicates colder weather, rain or snow. 
 
 No. 3, with No. 4 above it, indicates warmer weather with local rains. 
 
 No. 3, with No. 4 below it, indicates colder weather with local rains. 
 
 No. 1, with No. 5 above it, indicates fair weather, cold wave. 
 
 No. 2, with No. 5 above it, indicates wet weather, cold wave. 
 
 EXPLANATION OF WHISTLE SIGNALS. 
 
 The warning signal, to attract attention, will be a long blast of from 
 fifteen to twenty seconds' duration. After this warning signal has been 
 sounded, long blasts (of from four to six seconds' duration) refer to weather, 
 and short blasts (of from one to three seconds' duration) refer to temperature ; 
 those for weather to be sounded first. 
 
 Blasts. 
 One long 
 Two long 
 Three long 
 One short 
 Two short 
 Three short 
 
 Indicate. 
 Fair weather. 
 Rain or snow. 
 Local rains. 
 Lower temperature. 
 Higher temperature. 
 Cold wave. 
 
 INTERPRETATION OF COMBINATION BLASTS. 
 
 One long, alone 
 
 Two long, alone 
 
 One long and one short . . 
 
 Two long and two short . . 
 
 One long and three short . . 
 
 Three long and two short . . 
 
 Fair weather, stationary temperature. 
 Bain or snow, stationary temperature. 
 Fair weather, lower temperature. 
 Bain or snow, higher temperature. 
 Fair weather, cold wave. 
 Local rains, higher temperature. 
 
 By repeating each combination a few times, with an interval of ten seconds 
 between, possibilities of error in reading the forecasts will be avoided, such 
 as may arise from variable winds, or failure to hear the warning signal. 
 
64 METEOROLOGY 
 
 at these minor stations, unless otherwise ordered. The caution- 
 ary, storm, and information signals are as given in Chart A. 
 These signals are intended for inland and country districts. 
 
 The State Weather Service Division has charge of making or 
 discontinuing these display stations sending signals and receiving 
 reports. The series of flags used is given in Chart B. They are 
 of a very simple and suggestive character, are easily learned by 
 anybody in five minutes, and form, on the whole, the most 
 satisfactory series of signals distributed to the general public. 
 
 The Service has made use of a series of railway-train signals. 
 These are signals which are hung on the sides of the baggage or 
 postal car, and can be recognised at a considerable distance by 
 anyone who is in a position to see the passing train. They would 
 be very satisfactory to all within sight of the passing train, were 
 the signals themselves cared for properly. Their care, however, 
 must be left to the employes of the railway, and this is an addi- 
 tional burden placed on their shoulders. The result is, that in 
 some cases they are neglected, and one can occasionally see a rail- 
 way train bearing fair-weather signals when travelling through a 
 storm. 
 
 Another sort of signals which has been used extensively, and 
 which has been found to be fairly satisfactory, is the whistle 
 signal. This is employed only in the case of stationary engines. 
 It produces so little disturbance in the way of whistling as to 
 be generally unobjectionable. But protests are occasionally re- 
 ceived. In one case the whistle was located in the vicinity of a 
 Retreat intended for nervous invalids, and caused so much annoy- 
 ance that it was finally dispensed with. The code consists of a 
 series of long and short whistles, very simple in character and very 
 easy to understand. 
 
 The firing of a cannon has also been made use of for the purpose 
 of conveying information concerning the coming weather. It has 
 been employed chiefly to give information concerning frosts. 
 In one notable case it has been given the credit of having saved 
 an entire crop within the range of its hearing. 
 
 None of these methods can successfully reach all country 
 communities and individual farms. How to do this is the 
 problem yet to be solved. Among the suggested solutions is 
 
THE UNITED STATES WEATHER BUREAU 65 
 
 the extension of free delivery of mail, but this would require an 
 enormous expenditure on the part of the general Government 
 an expense that has been estimated not only in millions, but in 
 hundreds of millions of dollars. Another method proposed is 
 that of the extension of the telephone to farmers' houses. This 
 is entirely practicable, and will perhaps in time be accomplished. 
 Another method proposed is that of small captive balloons, which 
 can be made of different shapes and placed in variable order, 
 and allowed to rise to such a height as to be visible for many 
 miles. A much more promising method for future use is that of 
 the searchlight, rendered possible by the brilliancy of the electric 
 light. As is well known, the shadows of the electric light are 
 easily distinguishable on a cloudy sky, and may be made dis- 
 tinguishable on a clear sky when there is dust in the air. A 
 simple code of signals could be invented which, when projected 
 on low clouds, could be seen through a radius of forty or fifty 
 miles. If they were projected on a clear sky in such a way as to 
 be distinguishable, they could be seen through a radius of much 
 greater distance. This method has been tested to some extent 
 by private enterprise on the top of Mount Washington, and the 
 reports received from it are fairly satisfactory. 
 
 The results of the forecasts are to be found in the very great 
 benefits which this foreknowledge of the weather affords to marine 
 and inland commerce and agriculture, and to business of all sorts. 
 The evidences that these benefits are real and numerous exist in 
 the Bureau in very great numbers, and it would be a hopeless 
 task to endeavour to summarise them. To illustrate the character 
 of these advantages, for instance, to commerce, it may be said 
 that in the case of the two very severe hurricanes in the autumn 
 of 1893, which struck inland on the South Atlantic coast and 
 passed northward, warnings were given in abundant time before 
 the hurricanes struck the coast. Numerous illustrations can be 
 found as to the usefulness of the Bureau to trades and professions, 
 where the general public would not suspect that such usefulness 
 could exist. As one instance, it may be mentioned that a firm 
 carrying on the business of artificially curing wood has written to 
 the Bureau that they depend in their business on the forecasts 
 made by the Bureau, and that they regulate their furnaces by them. 
 
 5 
 
66 METEOROLOGY 
 
 The records are also of great use in all sorts of business. It 
 rarely occurs that a railway has a case for damaged fruit but the 
 Bureau is called on to testify in the case. In a great variety of 
 Admiralty cases the decision frequently depends more or less 
 directly on the testimony given by the Bureau as to the direction 
 of the wind and the character of the weather at the time and 
 place the loss occurred. 
 
 Scientific Work. 
 
 The methods under which the scientific work of the Bureau 
 has been carried on are various. In the early years of the Bureau 
 a study-room under Professor Abbe was organised, and to this 
 room came all the multitudinous questions, more or less directly 
 scientific, which were sent to the Central Office. There they were 
 carefully studied, and thence the results were issued, either in 
 correspondence or in publications. The scientific works produced 
 by members of the meteorological force have sometimes been 
 printed by the Government. Illustrations of these are found in 
 the cases of Professor FerreFs work, of the works of Professor 
 Abbe, and of many others. Sometimes the work of the Bureau 
 has been printed by private enterprise, or otherwise than by the 
 Government. The observations taken on Pike's Peak were 
 printed by the Harvard College Observatory. The results of 
 the electrical work, done under the direction of Dr. Mendenhall, 
 were printed by the National Academy of Sciences. The Service 
 has also encouraged the work of those outside its own ranks and 
 given them such aid as it could, and as a result a series of works 
 by eminent men has been published by the Bureau. For instance, 
 those of Professor Loomis and Professor S. P. Langley. 
 
 Of the work done in general meteorology we find illustration 
 in the publications of Professor Ferrel, and also in the Preparatory 
 Studies made by Professor Abbe. Both of these were printed 
 by the Signal Service as appendices to the annual reports. 
 
 In the matter of instruments the work has been more abundant 
 and more detailed. Professor Abbe's Treatise on Instruments is an 
 appendix to one of the annual volumes. The late Professor Marvin 
 devoted himself to many sides of the study of instruments. 
 Professor Waldo, while connected with the service, occupied him- 
 self with the standardising of the instruments of the service. 
 
THE UNITED STATES WEATHER BUREAU 67 
 
 The study of storms on the part of the meteorological service 
 has continued from its first establishment. The publication of 
 the Daily Weather Map is really a contribution to this subject, 
 and in addition to this, special maps and bulletins have been 
 issued on many individual storms. 
 
 The questions of solar physics, of electricity, and of magnetism, 
 have occupied the service not only directly, but independent 
 studies also have been encouraged by it, and the results have 
 been published by the Service. Dr. T. C. Mendenhall was 
 authorised by General Hazen to organise a special service for the 
 study of atmospheric electricity, and the results of these studies 
 were collected and published for Dr. Mendenhall by the National 
 Academy of Sciences. Professor Bigelow has made a study of 
 the relations of magnetism to meteorology. Clouds have also 
 received the attention of the Bureau officials, and many photo- 
 graphs have been made, and a collection of them has been 
 deposited in the library. 
 
 The work of the service in ballooning has been continued from 
 early days in its history. 
 
 There have been some studies of spectroscopy and its relations 
 to practical meteorology. Professor Upton, under General 
 Hazen, made a special study of the spectroscope, which was 
 printed as a bulletin. Later, Dr. Jewell, under the direction of 
 Professor Rowland, of Johns Hopkins University, made an 
 elaborate study of some new aspects of this problem which give 
 promise of success in its practical application to forecasting. The 
 subject of the radiation from gases is another of the problems in 
 which the Bureau has interested competent students outside of 
 its ranks. 
 
 In climatology the work of the Bureau has been very extensive. 
 A series of climatological monographs has been published on the 
 Western States, on the arid regions, and on the States occupying 
 the great plains. The Monthly Weather Review, which has 
 appeared continuously for so many years, has a number of 
 contributions on the subject of climatology. Rainfall has also 
 occupied the attention of the Bureau to a very great degree. 
 The rainfall maps are published regularly in the Monthly Weather 
 Review, and many special studies have been printed in atlas form, 
 
 52 
 
68 METEOROLOGY 
 
 or otherwise. The work of the meteorological service has been 
 in some sense brought together and condensed in General Greely's 
 American Weather. 
 
 As to the meteorological side of agricultural science, the studies 
 of the service have been numerous, but have been for the most 
 part made under the Weather Bureau rather than under the 
 Signal Service. 
 
 In 1908 the first number of the Bulletin of the Mount Weather 
 Observatory, Virginia, was published, marking an important 
 advance in the practical study of meteorological science. The 
 Bulletin has since appeared quarterly, and each number contains 
 more or less detailed accounts of the researches conducted by 
 the staff of the Observatory, of which William J. Humphreys, 
 Ph.D., is the Director, and William R. Blair, Ph.D., is the 
 Assistant Director. 
 
 Mount Weather Observatory, the name of a group of labora- 
 tories and observatories where the Weather Bureau of the United 
 States has been doing original research work since 1903, is situated 
 in Virginia, on the top of the Blue Ridge Mountains, at a height 
 of 1,725 feet above sea-level, some twenty miles south of Harper's 
 Ferry, and forty-seven miles in a direct line from Washington. 
 The administration building and its contents were totally de- 
 stroyed by fire on the morning of October 23, 1907, but it soon 
 rose, Phoenix-like, from its ashes, and the third part of the second 
 volume of the Bulletin was issued on December 11, 1909. Each 
 part of this important publication is prepared under the direction 
 of Willis L. Moore, D.Sc., LL.D., Chief of the United States 
 Weather Bureau. The Bulletin is profusely illustrated, and con- 
 tains some of the most valuable contributions to the literature 
 of modern meteorology which have been made within the past 
 six years. 
 
CHAPTER VI 
 
 THE METEOROLOGICAL SERVICE OF THE DOMINION OF 
 CANADA 
 
 DURING a recent visit to Canada and the United States (August- 
 September, 1909) I was privileged to visit the new Central Office 
 of the Canadian Meteorological Service, Toronto, Ontario, as well 
 as several of the observatories connected therewith scattered 
 throughout the vast Dominion of Canada. At my request, 
 Mr. K. F. Stupart, the very able Director of the Meteorological 
 Service of Canada, has favoured me with the following account of 
 the establishment and development of that Service. 
 
 In response to representations made by the Royal Society of 
 England and the British Association for the Advancement of 
 Science, the Imperial Government in 1840 established an Observa- 
 tory in Toronto for magnetical and meteorological observations. 
 The land on which the Observatory was located was a block of 
 2J acres, granted by the University for such time as it should be 
 required for scientific purposes. 
 
 The operations of the Observatory as an Imperial establishment 
 were brought to a close early in 1853, but were resumed under the 
 authority of the Provincial Government in July of the same year. 
 
 During the period that the Observatory was maintained by the 
 Imperial Government it was under the direction of an officer 
 of the Royal Artillery, who was assisted by several non-com- 
 missioned officers. When the Provincial Government assumed 
 control of the Observatory, the three non-commissioned officers, 
 who had served during the military regime, resigned from the 
 Army, and continued members of the Observatory staff. 
 
 For two years the duties of the Observatory were carried on 
 under the general supervision of the Professor of Natural Philo- 
 sophy of the University College, Toronto. 
 
 69 
 
70 METEOEOLOGY 
 
 In August, 1855, Professor G. T. Kingston was appointed 
 Director. 
 
 In 1869 Professor Kingston organised a voluntary meteoro- 
 logical system in Canada. 
 
 In the spring of 1871 a grant of $5,000 was made by the Do- 
 minion Government for the promotion of meteorological research, 
 and with a special view of establishing a system of storm-signals. 
 By this time there were forty observers in Canada forwarding 
 weather reports by mail to the Central Office. 
 
 In that year, as at the present time, the meteorologist believed 
 that the only possible way of forecasting storms was by means 
 of a map of a large area of the Earth's surface, on which are 
 written symbols, indicating the reading of the barometer, the 
 temperature, direction and velocity of the wind, etc., such in- 
 formation being supplied by telegraph. Maps showing a large 
 portion of North America were then printed in 1871, and, as a 
 commencement, six stations telegraphed weather reports to 
 Toronto three times a day ; these were forwarded to the United 
 States Weather Bureau, which Bureau in return furnished reports 
 from fifteen American stations. 
 
 In 1872 the grant was increased to $10,000, the number of 
 stations reporting was increased to eight, and storm-signal masts 
 were erected at various ports on the great lakes and in the mari- 
 time provinces. The staff of the Observatory was, at the com- 
 mencement of 1873, composed of the Director, the Assistant 
 Director, and six others, including a messenger. 
 
 Up to the autumn of 1876 the Canadian Service depended wholly 
 on the judgment of the United States Bureau for the issue of storm 
 warnings, which, on advice from Washington, were distributed 
 from Toronto. In September, 1876, warnings were independently 
 issued from Toronto, and in October daily forecasts were issued 
 to some points in the older provinces. At the close of 1876 there 
 were about 120 observers in correspondence with the Toronto 
 Office, and there were 37 storm-signal display stations. The 
 Central Office staff included 12 persons. 
 
 Early in 1880 Professor Kingston resigned office, and Mr. 
 Charles Carpmael, M.A., late Fellow of St. John's College, Cam- 
 bridge, was appointed as his successor. At the close of that year 
 
THE SERVICE OF THE DOMINION OF CANADA 71 
 
 there were 140 observers in Canada, 18 telegraph reporting 
 stations, and 44 storm-signal stations. The staff of the Central 
 Office numbered 17 persons. The Annual Climatological Report 
 in that year was an octavo volume of 365 pages. 
 
 In the year 1894 there were 268 climatological stations, 29 
 stations reporting by telegraph to the Central Office, and 65 
 storm-signal stations. The publications were an Annual Climato- 
 logical Report an octavo volume of about 350 pages ; a Monthly 
 Weather Review, somewhat similar to that published at the 
 present time ; and a Toronto Meteorological Register. Forecasts were 
 issued once a day in the evening to about 1,500 places in Canada. 
 
 The present Director, Mr. R. F. Stupart, was appointed by Order 
 in Council dated December 28, 1894. 
 
 On February 6, 1895, a map showing the climatic conditions 
 of the month just closed was published for the first time. 
 
 The Annual Climatological Report for 1895 is a quarto volume 
 of 274 pages, and contains charts showing the average climatic 
 conditions of the Canadian summer months. 
 
 In August of the year 1896 a daily map was first published, 
 and forecasts published in the forenoon, covering the current 
 and following days, were first issued to a large number of stations, 
 more particularly on the Great Lakes and in the Maritime 
 Provinces, the object being to give better information to mariners. 
 
 In 1900 arrangements were made with the Telegraph Company 
 to manifold a daily meteorological bulletin, and post it at 
 numerous places in some of the larger cities of Ontario, Quebec, 
 and the Maritime Provinces. It was about the year named that 
 the publication of the morning issue of forecasts covering the 
 current and following day became almost general in the after- 
 noon newspapers. 
 
 In 1904 a comprehensive Daily Weather Bulletin, containing 
 weather reports from a large number of stations in the wheat belt, 
 was during the summer for the first time published in Winnipeg 
 and many other important agricultural centres of the North-West. 
 This Bulletin has since been increased in size, and is distributed 
 very freely in the West. 
 
 On July 2, 1907, the Daily Weather Map, hitherto manifolded 
 by the mimeograph, was first printed. 
 
72 METEOEOLOGY 
 
 The Meteorological Service is a branch of the Governmental 
 Department of Marine and Fisheries. Its Central Office, newly 
 erected and admirably equipped, is in Bloor Street, Toronto, 
 Ontario. 
 
 A general statement of the equipment and work of the Meteoro- 
 logical Service in 1907 shows the following facts : 
 
 Number of persons, all told, employed in the Central Office, 
 
 Toronto . . . . . . . . . . 24 
 
 Number of persons receiving pay from the Meteorological 
 
 Service .. .. .. .. .. ..238 
 
 Number of stations reporting by telegraph to the Central 
 
 Office .. .. .. .. .. ..39 
 
 Number of storm-signal stations in operation . . . . 89 
 
 Number of climatological stations . . . . . . 445 
 
 The Government " Appropriation " for the year 1909 
 amounted to $129,300, together with a supplemental grant of 
 $3,200 for a magnetic observatory in connection with the Central 
 Office $132,500 in all. 
 
 PUBLICATIONS COMPILED AT THE CENTEAL OFFICE. 
 
 The Annual Climatological Report : Last issue a quarto volume 
 of 440 pages. 
 
 The Monthly Weather Review : A brochure of 12 pages, contain- 
 ing climatological data from about 300 stations, a general synopsis 
 of weather conditions, etc. 
 
 The Monthly Map : Published three days after the close of each 
 month, showing the temperature conditions which have obtained 
 departures from average, together with rainfall and snowfall, 
 information regarding crops, and the general outlook. 
 
 The Daily Map is compiled and printed in the Central Office. 
 
 The Toronto Meteorological Register, which has been published 
 for over sixty years, and is still continued. 
 
 WEATHER FORECASTS. 
 
 Weather forecasts, covering thirty-six hours in advance, and 
 sometimes a longer interval, are issued twice daily throughout 
 the year. The weather charts on which the forecasts are based 
 have entered on them information obtained by telegraph from 
 thirty-seven stations in Canada and sixty-four stations in the 
 
THE SERVICE OF THE DOMINION OF CANADA 73 
 
 United States ; also reports from St. John's and Bermuda. The 
 forenoon chart is ready for inspection ordinarily about 9.45 a.m., 
 and the forecast official, having drawn the isobars, first issues a 
 bulletin for the Maritime Provinces, including forecasts for the 
 current day and following day for Nova Scotia, New Brunswick, 
 and Prince Edward Island, and also for vessels leaving for the Great 
 Banks and for American ports. Then follows a forecast for the 
 Western Provinces, which is telegraphed without delay to Winni- 
 peg, where a local agent, who has meanwhile received weather 
 telegrams from some twenty-three points additional to those 
 received in Toronto, prepares a bulletin giving a general synopsis 
 of existing weather conditions, and also includes all weather 
 reports received, together with the forecasts from Toronto. This 
 bulletin is then distributed in Winnipeg, and telegraphed to the 
 more important centres in the Prairie Provinces. The Central Office 
 forecast official lastly prepares a bulletin for Ontario and Quebec, 
 which is usually despatched about 10.10 a.m., and is published 
 very widely by the afternoon press, as well as being posted at 
 telegraph-offices, post-offices, and other frequented places. At 
 all the larger towns in these Provinces a special effort has been 
 made to have these bulletins exposed on wharves and docks 
 within easy reach of shipping people and fishermen. 
 
 The evening weather chart, like that of the morning, is usually 
 ready for inspection about 9.45 p.m., and with as little delay as 
 possible a bulletin is prepared for the press, and forecasts are 
 issued for all parts of the Dominion, exclusive of British Columbia. 
 These forecasts are distributed by the telegraph companies to 
 most of the telegraph-offices in the Dominion, and, by arrange- 
 ments, are posted up in a frame hung in a conspicuous place, 
 and nearly every morning newspaper publishes them, generally 
 on the front page. 
 
 During the winter months a very large number of special fore- 
 casts are made for shippers of perishable goods, inquiries being 
 made both by telephone and telegraph. Indeed, there is little 
 reason to doubt that nearly all shippers of such goods in the 
 Dominion now consult the Weather Service before sending forward 
 
 consignments. 
 
 During the winter special warnings of snow and drift are issued to 
 
74 METEOKOLOGY 
 
 all Canadian railways, whenever it is decided that this is necessary. 
 Various electric railways also have made a practice of consulting 
 the Central Office as to the weather of the coming night, the 
 information supplied enabling them either to reduce the working 
 staff on duty to a minimum, or, on the other hand, to take unusual 
 measures to prevent snow blockades. 
 
 During the late autumn many telegrams are received from 
 vessel-masters wishing to cross the Lakes, requesting special fore- 
 casts as to probable winds and weather, and, indeed, in some cases 
 asking a direct opinion as to the advisableness of starting. 
 
 Forecasts and storm-warnings for British Columbia are issued 
 from Victoria, to which place are telegraphed reports from all 
 Canadian stations west of White Kiver, together with some 
 twenty-five reports from the Pacific States. 
 
 The Meteorological Service supplies the meteorological and 
 climatic data to the agricultural departments of the various Pro- 
 vincial Governments, and this entails much work in the Central 
 Office. It is a significant fact that the Government of the North- 
 West Territories for some years has reprinted the crop report from 
 the Monthly Map published by the Meteorological Service. 
 
 The old bench-mark on the south side of the former Central 
 Observatory at Toronto was 350 feet above mean sea-level. The 
 International Deep Water Ways Commission between the United 
 States and Canada have adopted 249-5 feet as the mean level of 
 Lake Ontario above the sea. The new Central Observatory 
 stands at least 100 feet above the level of the lake, from the 
 northern shore of which the city of Toronto rises by a tolerably 
 steep gradient. 
 
CHAPTER VII 
 AIR TEMPERATURE AND ITS MEASUREMENT 
 
 l&o other meteorological factor exercises a more potent influence 
 over the animal and vegetable kingdoms than does temperature. 
 Hence we may fitly commence our study of the weather with an 
 account of the instrument which is in daily use for the purpose 
 of measuring the warmth of the air namely, the thermometer 
 (Greek, Oepprj, heat ; pcrpov, a measure). The principle of the 
 instrument is that it measures temperature by the expansion of 
 bodies. 
 
 In the writings of two Greek physicists, Philo of Byzantium, 
 who lived in the third century B.C., and Hero of Alexandria, of 
 a later though undetermined date, we find descriptions of an 
 apparatus which represents the primitive idea of the thermoscope. 
 
 Philo's description in his work, De Ingeniis Spiritualibus 
 (" On Pressure Engines ") is given as follows by Professor Hell- 
 mann r 1 
 
 " One takes a leaden globe of moderate size, the inside of which 
 is empty and roomy (Fig. 2). It must neither be too thin, that it 
 cannot easily burst, nor too heavy, but quite dry, so that the ex- 
 periment may succeed. Through an aperture in the top is passed 
 a bent siphon, reaching nearly to the bottom. The other end of 
 this siphon is passed into a vessel filled with water, also reaching 
 nearly to the bottom, so that water may the more easily flow 
 out a is the globe, 6 the siphon, and g the vessel. I assert, when 
 the globe is placed in the sun and becomes warm, some of the air 
 inclosed in the tube will pass out. This will be seen, since the 
 air flows out of the tube into the water, setting it in motion and 
 
 1 "The Dawn of Meteorology," Quarterly Journal of the Royal Meteoro- 
 logical Society, October, 1908, vol. xxxiv., No. 148, p. 227. 
 
 75 
 
76 METEOKOLOGY 
 
 producing air-bubbles one after the other. If the globe is placed 
 in the shadow, or any other place where the sun does not penetrate, 
 then the water will rise through the tube flowing into the globe. 
 If the globe is again placed in the sun, the water will return to 
 the vessel, and vice versa. . . . The same effect is produced if one 
 heats the globe with fire or pours hot water over it. ..." 
 
 Somewhat more complicated is the similar apparatus of Hero, 
 to which he gives the name Ai/3as, or drip. 
 
 The thermometer is supposed to have been invented by Sanc- 
 torio, of Padua, in 1590 ; but the history of the instrument is 
 involved in obscurity until 1714, when Fahrenheit, of Dantzig, 
 constructed the thermometer which bears his name. He used 
 
 FIG. 2. PHILO'S THEKMOSCOPE. 
 
 mercury instead of spirit of wine, and introduced a scale, graduated 
 from the melting-point of ice to the boiling-point of water at 
 sea-level, and under ordinary conditions of temperature and 
 pressure. Fahrenheit arrived at his zero point in a curious way. 
 He is believed to have held that for all practical purposes the 
 lowest temperature likely to call for registration was tha reached 
 in an Icelandic winter. It is, however, more likely that his zero 
 was fixed by experiment with a freezing mixture of sn ow and 
 chloride of sodium (common salt) or of snow and chl< ride of 
 ammonium (sal ammoniac). 
 
 Counting, then, from zero, Fahrenheit made the meltir g-point 
 of ice 32 and the boiling-point of water 212, thus dividing the 
 
AIR TEMPERATURE AND ITS MEASUREMENT 77 
 
 distance between these two crucial points into 180, or half the 
 number of degrees in a circle (360). This scale is commonly used 
 throughout the British Dominions and in the United States of 
 America. Its advantages are that, under ordinary circumstances, 
 the minus sign ( ) is not required, while the smallness of the 
 degrees permits very accurate measurements of temperature. 
 
 Mercury was selected as the medium for indicating temperature, 
 because (1) of its equal expansion at different temperatures 
 equal increments of bulk corresponding to equal increments of 
 temperature; (2) of its low freezing-point (-37'9 F.), its high 
 boiling-point (675 ! 1 F.), its high conductivity of heat, its purity, 
 its property of not wetting glass, its low vapour pressure, and 
 its low specific heat, or " the amount of heat required to raise 
 1 pound of mercury one degree, in terms of that necessary to 
 raise 1 pound of water one degree " (R. H. Scott). Dr. Thomas 
 Young 1 defined the degree Fahrenheit as corresponding to an 
 expansion of mercury equal to r<y&oi> P ar ^ f i^ s volume or bulk. 
 This is absolutely true at a temperature of 142 F. For the 
 recording of temperatures below its freezing-point, mercury is 
 replaced by spirit, which medium is also used because of its 
 transparency in the construction of the minimum thermometer in 
 common use. In this instrument a small and light float of glass 
 or enamel, called the " index/' is immersed in the spirit. Hence 
 it is necessary that the spirit should be transparent in order that 
 the exact position of the index may be observed. 
 
 In 1742 Celsius, Professor of Astronomy in the University 
 of Upsala, in Sweden, divided the scale of the mercurial ther- 
 mometer between the melting-point of ice and the boiling-point of 
 water into 100. According to this scale, therefore, the melting- 
 point of ice is zero, and the boiling-point of water is 100 at an 
 atmospherical pressure of 760 millimetres (29'922 inches) in the 
 latitude of Paris. Hence the -name Centigrade, by which this 
 thermometer is usually known. It is extensively used on the 
 Continent of Europe, and, indeed, by scientific men of all nations. 
 
 A third thermometer scale is that introduced about 1731 by 
 Reaumur, a French physicist, who was born at La Rochelle in 
 1683. According to this scale there are only 80 between zero, 
 1 Lectures on Natural Philosophy, p. 485. 
 
78 METEOROLOGY 
 
 the melting-point of ice, and the boiling-point of water. The 
 Reaumur scale is still used in Russia and parts of Germany, but 
 the Celsius or Centigrade thermometer scale is rapidly superseding 
 it in those countries. 
 
 The following figures, copied from Chambers's Encyclopedia 
 (art. " Thermometer "), will render the relations of these three 
 scales easily understood : 
 
 Fahrenheit . . . . 32 77 122 212 
 
 Reaumur .. .. 20 40 80 
 
 Centigrade .. .. 25 50 100 
 
 To reduce a reading taken by one scale to the corresponding 
 reading taken by another becomes quite easy, if we remember 
 the primary relations of the scales to one another namely, 
 80 R. =100 C. =180 F. ; or in the most simple form : 4 R. = 
 5 C. =9 F. The only element of difficulty is the management 
 of 32 F., which corresponds to the zero of the other scales. It 
 is quite clear that when we are reducing Fahrenheit to Centigrade 
 or Reaumur, we must take away, or subtract, 32 from the 
 given Fahrenheit reading ; whereas in reducing either of the 
 other scales to Fahrenheit, we must add 32 to the result. 
 
 The following proportional statements may be of use : 
 1. For Fahrenheit's thermometer : 
 
 QT 
 
 F. : R. : : 180 : 80, i.e., 9 : 4. Therefore F. - -"*'+ 32. 
 
 QC 
 
 F. : C. : : 180 : 100, *.?., 9 : 5. Therefore F. = -~-+32. 
 
 5 
 
 2. For Celsius' thermometer : 
 
 C. : F. : : 100 : 180, i.e., 5 : 9. Therefore C. - (F> - ^ 2) ' 
 
 9 
 
 C. : R. : : 100 : 80, i.e., 5 : 4. Therefore C. =^ 
 
 3. For Reaumur's thermometer : 
 
 R. : F. : : 80 : 180, i.e., 4 : 9. Therefore R. - (F - ~- 32) > 
 
 R. : C. : : 80 : 100, i.e., 4 : 5. Therefore R. =^- 
 
 5 
 
 In speaking of the fixed points on the thermometer scale it 
 will be observed that the expressions used have been "the melting- 
 point of ice " and the " boiling-point of water." 
 
AIR TEMPERATURE AND ITS MEASUREMENT 79 
 
 The melting-point of ice is exactly 32 F., and of both the 
 Centigrade and Reaumur scales. The phrase is used in preference 
 to the " freezing-point of water," because water, if perfectly still, 
 may be chilled several degrees below 32 F. without freezing. 
 Under such circumstances, if it is suddenly agitated, it will congeal 
 instantly. Again, water which holds a salt in solution has a 
 freezing-point considerably below 32 F. Only distilled water 
 is admissible in these delicate experiments. 
 
 The boiling-point of water is a still more variable quantity 
 than the freezing-point, and hence the term must be qualified 
 by the addition of the words " at mean sea-level, the barometer 
 standing at 29-905 inches in the latitude of London." The 
 French standard of pressure already referred to is 760 millimetres 
 in the latitude of Paris. Ebullition takes place the moment the 
 tension of the vapour of water equals the atmospheric pressure. 
 It is evident that the lower the pressure, the more easily will 
 vapour escape from heated water, giving rise to the phenomenon 
 known as " ebullition " or boiling. The converse is equally true, 
 and under a pressure of fifty atmospheres the boiling-point of 
 water is raised to 510 F. In fact, water can be heated to almost 
 any degree without boiling, provided it is subjected to a sufficient 
 pressure. 
 
 As a matter of fact, the boiling-point falls one degree of 
 Fahrenheit's scale for every 0-589 inch of barometric fall at 
 moderate heights. Accordingly, should the barometer fall to 
 27-549 inches at sea-level and this actually occurred on 
 January 24, 1884, and, more recently, on December 8, 1886 the 
 boiling-point would be 208 instead of 212. A still more striking 
 variation in the boiling-point occurs at great elevations. For 
 instance, at Quito, in Mexico, 9,000 feet above the sea, water 
 boils at 194 F. ; on the summit of Mont Blanc (15,781 feet) at 
 183 F. ; and on Gaurisankar, or Mount Everest, the highest 
 peak of the Himalayas (29,002 feet), it would boil at 158 F., 
 were it possible to make the experiment. Advantage has actually 
 been taken of this dependence of the boiling-point on elevation 
 to roughly measure the heights of mountains. Mr. R. Strachan, 
 F.R.Met.Soc., suggests a simple rule for ascertaining the relative 
 elevation of two stations. It is to multiply by 9 the difference 
 
80 METEOEOLOGY 
 
 in barometrical readings between them, taken in hundredths of 
 an inch. The result gives the difference in feet between the 
 stations. This rule depends on the principle that the difference 
 of height corresponding to a difference in barometrical readings 
 of 0-1 inch is approximately 90 feet. For instance, for the change 
 of level from 200 to 290 feet, the difference in barometrical 
 readings (the sea-level reading being 30 inches) is O'lOl inch at 
 40 and 0-098 inch at 50 (R. H. Scott). 
 
 In the case of the Centigrade and Reaumur scales, all tempera- 
 tures below the melting-point of ice have a minus sign ( ) pre- 
 fixed. In the case of the Fahrenheit scale the zero point is 32 
 below the melting-point of ice. The minus sign is, therefore, very 
 seldom required for temperatures occurring in the British Isles. 
 The value -40 represents the same temperature on the 
 Fahrenheit and Centigrade scales. 
 
 A thermometer consists of a capillary glass tube, of uniform 
 bore and blown at one end into a bulb, which is then filled with 
 mercury or spirit, and finally the tube is hermetically sealed at 
 the other end. Any given temperature is measured by the 
 amount of expansion undergone by the liquid in the bulb when 
 exposed to that temperature. The liquid moves up the capillary 
 tube from the bulb as temperature rises, and retreats towards the 
 bulb as temperature falls. In all cases, the scale of degrees by 
 which the readings are made should be engraved on the glass tube 
 itself that is, on the stem. 
 
 The steps in the construction of a thermometer are four in 
 number : (1) calibrating the tube ; (2) filling ; (3) " curing "; 
 (4) graduation. 
 
 (1) Uniformity of calibre is attained in glass-blowing by intro- 
 ducing a bead of mercury, say an inch in length, and noting by 
 measurement if this thread of the liquid metal occupies the same 
 extent of the tube at different points. 
 
 (2) The thermometer is filled, after blowing a bulb at the 
 bottom, by filling a small funnel at the top with mercury and 
 then expelling the air in part from the bulb by heat. As the bulb 
 cools, a vacuum is formed, to fill up which some of the mercury 
 slips back from the funnel into the bulb. The process is repeated 
 until the bulb is quite filled with mercury. 
 
AIR TEMPERATURE AND ITS MEASUREMENT 81 
 
 (3) " Curing " is referred to afterwards. The process consists 
 in laying the instrument aside for a year or so after filling, so that 
 the glass may assume a permanent shape, or " season," and so 
 obviate the error known as " displacement of zero/' Denton has 
 introduced a method by which this can now be done in as many 
 days as it used to take months. 
 
 (4) Graduation is the marking of the scale on the thermometer 
 stem, the fixed points of temperature of melting ice (32 F.), and 
 of the vapour (steam) of water boiling at a pressure of 29*905 inches 
 that is, the pressure of one standard atmosphere in the latitude 
 of London (212 F.) being duly ascertained by direct experiment 
 in each case. 
 
CHAPTEK VIII 
 THERMOMETERS 
 
 THE thermometers used in meteorological observatories are 
 standard thermometers, ordinary thermometers, registering ther- 
 mometers, self-recording thermometers, and radiation thermo- 
 meters. 
 
 1. A standard thermometer is made with every precaution to 
 secure accuracy. It is not intended for daily use, but only for 
 testing from time to time the correctness of the ordinary ther- 
 mometers with which observations are made. The scale should 
 not be cut on the stem for several years after the carefully selected 
 tube and bulb have been filled. This delay is necessary in order 
 to guard against the defect called the " displacement of zero," by 
 which a thermometer is made to read too high. This defect 
 arises from the gradual contraction of the bulb which results 
 from the slowness with which fused glass returns to its original 
 density. Of course, as the bulb contracts, it holds less mercury, 
 which is forced into the tube to a higher level than the tempera- 
 ture warrants. Except for use in extremely cold climates, a 
 standard thermometer should be made with mercury, because of 
 the uniform rate of expansion of this metal. Its scale should 
 range from far below zero to the boiling-point of water. 
 
 2. Ordinary thermometers should be constructed of mercury. 
 They should be scaled from -40 to 110 or 120. In the 
 British Isles a range from - 15 or - 10 to 100 is ample. In 
 every case an ordinary thermometer should carry with it a certifi- 
 cate of verification at Kew Observatory, or other recognised 
 scientific institution. At least once a year each instrument 
 should be tested for the " displacement of zero/' by being plunged 
 into a mass of melting snow or ice. 
 
 82 
 
THERMOMETERS 
 
 83 
 
 The Kew Observatory Thermometer (Fig. 3) is an excellent 
 instrument, particularly adapted for taking reliable observations 
 at sea, as it is proof against the corrosive action of salt water 
 and of damp, and is protected by a copper case. This ther- 
 mometer is 12 inches long, with the degrees etched on the stem, 
 and the figures indelibly burned on the porcelain scale, which 
 ranges from to 120 F. It is made after the Meteorological 
 Office and Admiralty pattern, and carries with it a 
 certificate of verification at Kew Observatory. 
 
 3. Registering thermometers are so constructed as 
 to enable us to read off from them the highest 
 or lowest temperature to which they have been 
 exposed in a given length of time usually twenty- 
 four hours. As it is too inconvenient to read these 
 instruments at the close of a civil day that is, at 
 midnight they are by arrangement read at the 
 latest observing hour of the day, or 9 p.m. in the 
 British Isles. The thermometer which is used for 
 registering the highest or maximal temperature of 
 the day of twenty-four hours is called a " maximum 
 thermometer." Similarly, that which registers the 
 lowest or minimal temperature is called a "mini- 
 mum thermometer." In both maximum and 
 minimum thermometers the contrivance by means 
 of which we are able to read the extremes of 
 temperature is called the " index." 
 
 Maximum thermometers are of two kinds, called 
 after their designers, Phillips' s and Negretti and 
 Zambra's. 
 
 In the instrument invented by Professor Phillips, F.R.S., of 
 Oxford (Fig. 4, A), the index is really a small fragment of the 
 column of mercury, separated from the main body by a tiny 
 bubble of air. When the column expands, this fragment is 
 pushed before it ; but when the column contracts, it remains 
 behind in the capillary tube, so marking the point of highest 
 temperature. This delicate arrangement is apt to get out of 
 order, the air bubble being displaced, and so permitting 
 the index to coalesce with the main thread or column of 
 
 62 
 
 Fi o. 3. KEW 
 OBSERVATORY 
 THERMOMETER. 
 
84 
 
 METEOROLOGY 
 
 mercury, thus converting the instrument into an ordinary 
 thermometer. 
 
 Negretti and Zambra devised a plan which is as ingenious as it 
 is simple (Fig. 4, B). The mercurial thread in the thermometer 
 tube forms itself the index in this way : the bore of the tube is 
 bent at a sharp angle and so much reduced in calibre close to the 
 bulb that, while the expansion of the mercury in the bulb is quite 
 capable of forcing the liquid past the constriction into the tube ; 
 on cooling, the portion of mercury which has so passed into the 
 tube breaks off from the main body, and remains in the tube to 
 register the highest point to which it had reached. In fact, when 
 contracting, cohesion fails, and the mercury in the bulb is unable 
 
 fL 
 
 MAXIMUM I^CASELVA LONDON 
 
 ^ff4Jnli.JnJuMlj.J,nl.J.ML.lnl..n!.J.J.4alJJJ^^^^^ 
 
 ! I I I I I | ! I II I II I I I I I II I ] I II l| ! I I I I I 
 
 1 20 ! 10 I I 10 I 20 I 30 I 40 I iO I 60 170 I 80 I 90 I 100 I 110 1 120 ! 130 ! I4C 
 
 FIG. 4. A, PHILLIPS'S, AND B, NEGRECTI AND ZAMBRA'S MAXIMUM THERMOMETERS. 
 
 to draw back the portion in the tube, a separation taking place 
 where the tube is both bent and constricted. 
 
 The maximum thermometer should be suspended in the ther- 
 mometer stand or Stevenson screen horizontally, or almost hori- 
 zontally. The bulb end should be gently and slightly depressed 
 before reading, so as to secure that the index is in apposition with 
 the constriction in the tube at the near or bulb end. Great care 
 should be taken to handle the instruments as little as possible 
 while reading them ; otherwise erroneous records may result. 
 
 These thermometers are both set in the same way. They are 
 taken in the hand and swung gently bulb downwards or else 
 they may be lightly tapped, bulb downwards, on the wooden 
 
THERMOMETERS 85 
 
 ledge of the stand. In this way the instrument becomes for the 
 time being an ordinary thermometer, showing the existing 
 temperature of the surrounding air. 
 
 Maximum thermometers are liable to two defects : 
 
 1. The mercury may recede from its maximal position to a 
 greater or less extent when the temperature subsequently falls. 
 The observer should accordingly test the instrument occa- 
 sionally by heating it and noting whether the mercury column 
 retains its position in the tube. 
 
 2. The mercury may slip forward when the instrument is 
 brought into a horizontal position after setting. 
 
 These defects may be remedied in most cases by altering the 
 inclination at which the instrument hangs. Should they per- 
 sist, the thermometer must be rejected (The Observer s Hand- 
 book, Meteorological Office, London, 1908). 
 
 FIG. 5. RUTHERFORD'S MINIMUM THERMOMETER, FILLED WITH PURE ALCOHOL FOR 
 ORDINARY REGISTRATION, ENGINE DIVIDED ON THE STEM. 
 
 Minimum thermometers are also of two kinds Casella's, a 
 mercurial thermometer ; Rutherford's, a spirit thermometer. 
 The former is a beautiful instrument, and especially adapted 
 for use in tropical climates, where the intense heat by day causes 
 spirit to volatilise quickly. The latter is the minimum ther- 
 mometer which is in almost universal use at our home and 
 colonial stations (Fig. 5). In its construction a small metallic 
 index is immersed in the spirit. Before using, this index is 
 allowed to run down to the end of the column of liquid by sloping 
 the thermometer with the bulb uppermost. In this way the 
 thermometer is " set/' It should then be placed in a nearly 
 horizontal position in the screen, any slight inclination being 
 towards the bulb, so as to facilitate the backward movement of 
 the index when the spirit contracts with a falling temperature. 
 When this happens, the index is drawn back with the spirit by 
 
86 METEOKOLOGY 
 
 the force of capillary attraction. On a rise of temperature taking 
 place, the spirit expands and flows past the index, which remains 
 behind to mark the lowest or minimum temperature. 
 
 A spirit thermometer such as that just described should be 
 carefully watched and periodically compared with a reliable 
 mercurial thermometer, for some of the spirit is apt to volatilise 
 and afterwards to condense in the distal or further end of the tube, 
 causing the instrument to read too low by two, three, or even 
 more degrees. 
 
 Such an accident is easily remedied by swinging the thermo- 
 meter backwards and forwards, bulb downwards. It may, 
 however, be necessary to cautiously heat the distal end of the 
 tube in the flame, or near the flame, of a spirit-lamp. This causes 
 the condensed spirit again to volatilise. If the instrument is 
 then cooled in the position bulb downwards, the freshly con- 
 densed spirit will gradually trickle downwards and join the main 
 body of the liquid. 
 
 Spirit thermometers are by no means as sensitive as mercurial 
 ones, mercury having a much lower specific heat and a much 
 higher conductivity than alcohol. Hence it is becoming usual 
 to make the bulb of the spirit thermometer of such a shape 
 forked or cylindrical that as large a surface of the spirit as 
 possible shall be exposed to the action of the air. 
 
 Mention should be made of the registering thermometers which 
 were devised in the eighteenth century, but which are now dis- 
 carded as useless for scientific purposes. In 1757 Lord Charles 
 Cavendish, a Vice-President of the Royal Society, read a paper 
 before the Society on a maximum thermometer and a minimum 
 thermometer which he had designed. This paper will be found 
 in the fiftieth volume of the Philosophical Transactions (p. 300). 
 His instruments suggested to Mr. James Six, a quarter of a cen- 
 tury later (in 1782), the idea of an improved registering ther- 
 mometer, which has ever since been known as " Six's Ther- 
 mometer " (Fig. 6). It combines in one instrument a maximum 
 and a minimum thermometer. It consists of a long tube bent 
 parallel to itself in the centre like a siphon, and terminating at 
 each extremity in a bulb, one of which bulbs is larger than the 
 other. The bend of the tube is filled with a plug of mercury. 
 
THERMOMETERS 
 
 87 
 
 The remainder of the tube and both bulbs are filled with spirit, 
 but in the smaller bulb there is also a bubble of highly compressed 
 A small needle index of steel, with a capillary filament 
 
 air. 
 
 
 I 
 
 attached to it, floats on each end of the plug of mercury. 
 
 As temperature rises, the spirit in the larger bulb expands, 
 and pushes the mercurial plug with one of the indexes in front of 
 it into the distal tube. When temperature falls, the spirit, of 
 course, contracts, and the elasticity of the com- 
 pressed-air bubble in the distal bulb causes the 
 mercury to pass back after the retreating spirit, 
 the distal index remaining behind to mark the 
 point of highest temperature. But as the 
 mercury passes back into the proximal tube, 
 it pushes the other index before it towards the 
 larger bulb until temperature ceases to fall. 
 When this happens, the proximal index re- 
 mains to mark the lowest temperature. 
 
 The indexes, being of steel, can be " set " for 
 a new observation by means of a magnet in 
 this way by attraction they are drawn back 
 to the surface of each end of the plug of 
 mercury. 
 
 Since a mercurial minimum thermometer is 
 much better suited for use in hot climates than 
 a spirit thermometer, it may be well to describe 
 Casella's ingenious instrument at some length. 
 Mercury is the only fluid employed in its 
 make. The bulb and column are of the same 
 size as in the standard maximum thermometer, 
 and cold is thus registered under precisely 
 the same conditions as heat. No steel or 
 other index is employed. In this thermometer advantage is 
 taken of a curious property of mercury which tends to adhere 
 to glass in vacuo. 
 
 Various experiments were undertaken by Mr. Louis P. Casella, 
 F.R.Met.Soc., for the purpose of discovering some means by 
 which in a mercurial thermometer the mercury itself might be 
 detained at the point of lowest temperature, and so serve as a 
 
 FIG. (i. 
 Six's THERMOMETER. 
 
88 
 
 METEOROLOGY 
 
 self-registering minimum thermometer. It occurred to him that 
 the adhesive property of mercury for glass in vacuo together 
 with the fact that, where two tubes are united to one bulb, this 
 fluid will rise by expansion in the larger, and recede by con- 
 traction in the smaller, tube might enable him to attain his 
 object. The result was the invention of probably the first 
 instrument known to register past indications without having 
 or forming any separate index. 
 
 The general form and arrangement of this instrument are 
 shown in the accompanying illustrations (Fig. 7), of which the 
 upper drawing represents a full-sized section. A tube with large 
 bore is made to come off from the upper side of the thermometer 
 stem a short distance in front of the bulb. At the distal end of 
 
 FIG. 7. CASELLA'S MERCURIAL MINIMUM THERMOMETER. 
 
 this large-bored tube a flat glass diaphragm is formed by the 
 abrupt junction of a small pear-shaped chamber (ab), the inlet 
 to which at b is larger than the bore of the indicating tube or 
 stem of the thermometer. 
 
 The thermometer being set, as temperature falls, the mercury 
 in the bulb contracts and withdraws the fluid in the stem only. 
 When the minimum temperature is reached, the mercurial 
 column in the stem marks it, not only at the moment, but after- 
 wards also ; for as soon as the mercury in the bulb begins to 
 expand with a rise of temperature, it finds an easier passage into 
 the tube of larger bore, and through it into the pear-shaped 
 chamber beyond. In the tube of smaller bore that is, the stem 
 of the thermometer adhesion or capillary attraction holds the 
 mercury, and prevents its recession from the point indicating the 
 
THERMOMETERS 89 
 
 lowest temperature that had been reached since the instrument 
 was set. 
 
 To set the thermometer, it should be placed in a horizontal 
 position, with the back plate suspended on a nail, and the lower 
 part supported on a hook. The bulb end may now be gently 
 raised or lowered, causing the mercury to flow slowly until the 
 bent part is full and the chamber (ab) is quite empty. At this 
 point the flow of mercury in the long stem of the thermometer 
 is arrested, and indicates the exact temperature of the bulb that 
 is, of the air at the time. When out of use, or after transit, 
 it may happen that, on raising the bulb, the mercury will not at 
 first flow out from the small chamber (ab). In such a case, how- 
 ever, a slight tap with the hand on the opposite end of the instru- 
 ment, with the bulb uppermost, will readily cause it to do so. 
 
 The readings of maximum and minimum thermometers should 
 be compared regularly with those of an ordinary thermometer 
 placed beside them to check their action and to determine the 
 corrections which should be applied to them. 
 
 The maximum and minimum thermometers, the ordinary or 
 dry-bulb thermometer, and the wet-bulb thermometer, by which 
 the temperature of evaporation is shown, as will be afterwards 
 described, 1 should all be suspended in a suitable screen or ther- 
 mometer stand facing north. 
 
 Thermometers should be protected from the direct or reflected 
 rays of the sun, but at the same time should be freely exposed to 
 the open air. These ends are best attained by placing them in a 
 thermometer stand, such as the louvre-boarded box designed by 
 Mr. Thomas Stevenson, C.E., of Edinburgh hence called the 
 Stevenson stand or screen (Fig. 8). The pattern of this screen, 
 which has been approved by the Royal Meteorological Society, 
 is described in the Quarterly Journal of the Society (vol. x., p. 92, 
 1884). The screen is a double-louvred box, its interior being 
 18 inches long, 11 inches wide, and 15 inches high. It has a 
 double roof, the upper one projecting 2 inches beyond the body 
 of the screen on all sides, and sloping from front to back. The 
 front is hinged as a door, and opens downwards. The thermo- 
 meters are suspended on uprights near the middle of the screen, 
 
 1 See p. 181. 
 
90 
 
 METEOROLOGY 
 
 which should be painted white within and without, the finishing 
 coat consisting of white paint and copal varnish. It is desirable 
 that the woodwork should be repainted in the spring of each year. 
 The screen should be mounted on four stout posts over short 
 grass and freely exposed. The posts should penetrate the ground 
 to a depth of fully 2 feet, and the soil surrounding them should 
 
 FIG. 8. STEVENSON'S THERMOMETER STAND. 
 
 be well rammed down. The screen should be arranged so that 
 the bulbs of the dry and wet thermometers shall be 4 feet above 
 the ground. It should not stand in the shade or within 10 feet 
 of any wall, particularly of one with a southern aspect. The 
 door of the screen should open towards the north. It will be 
 a convenience to have a wooden rack placed on the grass imme- 
 
THERMOMETERS 91 
 
 diately in front of the screen for the observer to stand on when 
 reading the thermometers. 
 
 In towns, at many telegraphic stations, and on board ship, 
 a " wall screen " must take the place of the freely exposed 
 Stevenson stand just described. It consists of a covered case, 
 with louvred wings, fixed on a wall facing north, at the 
 height of 4 feet from the ground, by means of large hold- 
 fasts. 
 
 The Stevenson thermometer screen is hardly suitable for use 
 in tropical countries. The following arrangements have been 
 recommended by the Committee of the British Association on 
 the Climate of Tropical Africa : 
 
 The thermometers should be placed within an iron cage, which 
 should at all times be kept locked, so as to prevent interference 
 with the instruments. This cage should be suspended under a 
 thatched shelter, which should be situated in an open spot at some 
 distance from buildings, must be well ventilated, and should guard 
 the instruments from exposure to sunshine or rain, or to radiation 
 from the ground. A simple hut, made of materials available on 
 the spot, would answer this purpose. A gabled roof with broad 
 eaves, the ridge of which runs from north to south, is fixed upon 
 four posts, standing 4 feet apart. Two additional posts may be 
 introduced to support the ends of the ridge beam. The roof at 
 each end projects about 18 inches. In it are two ventilating 
 holes. The tops of the posts are connected by bars or rails, and 
 from a crossbar is suspended the iron cage with the thermometers. 
 These will then be at a height of 6 feet above the ground. The 
 gable ends may be permanently covered in with mats or louvre- 
 work, not interfering with the free circulation of the air, or the 
 hut may be circular. The roof may be covered with palm-fronds, 
 grass, or any other material locally used by the natives as 
 building material. The floor should not be bare, but covered with 
 grass or low shrubs. Care must be taken to fix the cage firmly, 
 so that the maximum and minimum thermometers may not be 
 disturbed by vibration. 
 
 The International Meteorological Committee, at Vienna in 
 1873, and again at Rome in 1879, expressed the opinion that 
 exposure of thermometers in a space which is open and accessible 
 
92 METEOROLOGY 
 
 to all winds, and at a height of 1J to 2 metres, is, as a rule, the 
 most suitable, though not always practicable. 
 
 The Royal Meteorological Society recommends that the 
 observations with the thermometers just described should be 
 made as follows : Having opened the screen, the dry and wet bulb 
 thermometers are to be read first, so that they may not be 
 affected by the nearness of the observer. The maximum ther- 
 mometer is to be read next, by noting the point at which the end 
 of the column of mercury is lying. The minimum thermometer 
 is read last, by noting the position of the end of the index furthest 
 from the bulb. The surface of the column of spirit shows the 
 temperature at the time of observation. A second reading of 
 all the thermometers should be taken to guard against any mis- 
 take in the first entry. The maximum and minimum ther- 
 mometers should then be set. When set, the end of the mercury 
 in the maximum and the end of the index furthest from the bulb 
 in the minimum should indicate the same temperature as the dry 
 bulb. The door of the screen should finally be closed, after fresh 
 (preferably soft) water has been poured over the wet-bulb ther- 
 mometer. 
 
 Sling Thermometer. Under the name of thermometre fronde 
 (sling thermometer) the French meteorologist M. Arago, in 1830, 
 devised a method of measuring air temperature by means of a 
 thermometer attached to a string, and allowed to swing rapidly 
 round for the space of half a minute or so. By this method the 
 use of a screen is dispensed with, and even in full sunshine a close 
 approximation to the true air temperature in the shade may be 
 obtained. This method is, of course, applicable only for isolated 
 observations. 
 
 4. Self-recording thermometers, or thermographs, are so arranged 
 as to record their own readings, independently of the observer, 
 either at frequent intervals in the case of the electrical 
 thermograph, or continuously, as in the photographic thermo- 
 graph. 
 
 In most thermographs the thermometer consists of a slightly 
 curved metal tube filled with spirit (Bourbon tube). One end of 
 this is fixed rigidly to the instrument, while the other is attached 
 to the system of levers which actuates the recording pen. 
 
THERMOMETERS 93 
 
 In the electrical thermographs designed by Dr. Theorell, of 
 Upsala, and Professor F. van Rysselberghe, of Ostend, the ther- 
 mometer tube is open at the upper end, and a wire is introduced 
 into it, which, by a clockwork mechanism long before devised by 
 Sir Charles Wheatstone, is caused to descend at regular intervals 
 until it touches the surface of the mercury. At the moment of 
 contact an electric current is generated, which causes a needle to 
 prick a paper, on which the thermometer scale is marked, at the 
 point corresponding to the height of the mercurial column at the 
 time. The wire is then raised mechanically, and contact is 
 broken (R. H. Scott). In Sir Charles Wheatstone's thermo- 
 graph the mercury became oxidised by the electric spark 
 produced at the moment that the dip separated from the 
 mercury (etincelle de rupture) ; but this inconvenience has been 
 obviated. 
 
 A photographic thermograph is in use in the stations of the 
 First Order managed by the Meteorological Committee (see p. 28, 
 above). In this instrument a bubble of air is introduced into 
 the column of mercury, and this moves up and down with the 
 temperature, the bore of the tube being larger than in Phillips's 
 maximum thermometer. A lamp is placed before the instrument, 
 and a photograph of the space occupied by the air bubble 
 is continuously taken on prepared paper stretched on a drum, 
 which is caused to revolve on its own axis once in forty-eight 
 hours. 
 
 At Greenwich Observatory a thermograph of a rather different 
 construction is employed. It is made somewhat on the principle 
 of the Kew Ba ograph. The light is allowed to pass through the 
 thermometer tube above the level of the mercury on to the 
 sensitised paper. We in this way get a continuous photographic 
 tracing which corresponds along its lower edge with the tempera- 
 ture range, the level of the mercury abruptly cutting off the 
 photographic tracing below. 
 
 5. Radiation Thermometers. The subject of radiation is so 
 important that we shall consider it in the next chapter, and there 
 explain the instruments by which it is measured. 
 
 Having taken observations with the thermometers already 
 described, we are in a position to ascertain the Mean Tempera- 
 
94 METEOROLOGY 
 
 ture, or that temperature which has an intermediate value 
 (1) between the several successive hourly temperatures recorded 
 by a thermograph or read by an observer every hour throughout 
 an entire solar day of twenty-four hours ; or (2) between the 
 extreme readings recorded in that time ; or (3) between the dry- 
 bulb readings taken twice or thrice daily, at 9 a.m. and 9 p.m., 
 as at all British Stations of the Second Order (Normal Climato- 
 logical Stations) ; or at 7 a.m., 1 p.m., and 9 p.m., as in Russia ; 
 or at 6 a.m., 2 p.m., and 10 p.m., as in Austria. 
 
 These are the three methods adopted for ascertaining what is 
 known as mean temperature of a day. The mean temperature 
 of a month is obtained by dividing the figure 31, or 30, or 29, 
 or 28, as the case may be, into the sum of that number of daily 
 values. The mean temperature of a year is similarly obtained 
 by dividing the figure 12 into the sum of that number of monthly 
 values. 
 
 The term average mean temperature is properly applied to that 
 temperature which represents the mean t)f a number of means. 
 For example, in Dublin the mean temperature of March, 1909, 
 was 40'8, but the average mean temperature for March in that 
 city in a long series of years (1866-1905 that is, forty years) 
 was 43*4. We say, then, that the mean temperature of March, 
 1909, was 2-6 below the average. 
 
 When dealing with diurnal extremes of temperature, a sufficient 
 approximation to the mean temperature is obtained by taking 
 the arithmetical mean of the maximum and minimum ther- 
 mometer readings, according to the formula 
 
 Min.+ {Max.-Min.} x 0-5=M.T. 
 
 A careful comparison, however, with the results yielded by 
 thermograms, as the tracings taken by the thermograph are 
 called, has suggested the following empirical formula, in which 
 the coefficient C, a variable quantity from month to month, 
 takes the place of the constant coefficient 0*5. 
 
 Min. + {Max.-Min.} x C.=M.T. 
 
 The annexed table gives the coefficients for the different 
 months. 
 
THERMOMETERS 
 TABLE IV. 
 
 95 
 
 Months. 
 
 Coefficient. 
 
 January 
 December 
 
 } 
 
 0-520 
 
 February 
 November 
 
 / 
 
 0-500 
 
 March 
 October 
 
 
 0-485 
 
 April 
 September 
 
 } 
 
 0-476 
 
 May 
 
 August- 
 
 } 
 
 0-470 
 
 June 
 July 
 
 } 
 
 0-465 
 
 In accordance with this table, the mean temperature of May, 
 1909, in Dublin, was 
 not 
 
 Min. + {Max. - Min.} x 0-5, 
 but 
 
 Min. + {Max. - Min.} x 0-470. 
 
 Interpolating the actual values we have 
 not 
 
 45-1+ {60-1- 45-lx 0-50} =52-6, 
 but^-- 
 
 45-1+ {60-1 -45-lx 0-470} =52-2. 
 
CHAPTEK IX 
 RADIATION 
 
 HEAT is communicated or transmitted from body to body or 
 from place to place, in at least three different ways by con- 
 duction, by convection, and by radiation. 
 
 Conduction is the transmission of heat through a conductor, 
 or a substance or body capable of being a medium for its trans- 
 mission for example, an iron poker, as contrasted with a stick, 
 which latter is a non-conductor. To quote the American Cyclo- 
 pcedia : " The communication of heat from one body to another 
 when they are in contact, or through a homogeneous body from 
 particle to particle, constitutes conduction/' As a rule, con- 
 ductors of heat are also conductors of electricity. 
 
 In practical meteorology we have illustrations of conduction 
 of heat in the propagation of changes of temperature from the 
 surface of the earth to the successive strata of the subsoil ; and 
 in the alternate heating and chilling of the lowest strata of the 
 air through contact with the ground. The subsoil temperature 
 is recorded by means of underground thermometers, such as are 
 figured in the accompanying illustrations (Figs. 9 and 10). 
 
 Convection is the transference or transmission of heat by means 
 of currents generated in liquids and gases by changes of tempera- 
 ture and other causes. When a spirit lamp is applied to the 
 bottom of a vessel of water, the heated water at the bottom ex- 
 pands, becomes specifically lighter, and so rises to the surface, 
 carrying with it or conveying the heat it has received. Thus, 
 by convection, heat is diffused at last through the whole mass of 
 the water in the vessel. A similar experiment on a stupendous 
 scale in Nature causes hot and cold winds in the atmosphere, as 
 well as vast ocean currents in the Atlantic and Pacific, which 
 
 96 
 
RADIATION 
 
 97 
 
 convey heat or warmth to high latitudes along the 
 north-western coasts of Europe and of America. 
 Convection, like conduction, applies to heat and 
 electricity alike. 
 
 Radiation is the transmission from a point or 
 surface of rays of heat along divergent lines (Latin, 
 radius, a semi-diameter of a circle ; hence a beam 
 or ray of light proceeding from a bright object 
 along divergent right lines or radii), not from 
 particle to particle of the 
 same body (as in conduc- 
 tion), but from one body 
 to another, through air, or 
 vacuum, or space. Radiant 
 heat is, in fact, identical 
 with light, only the wave- 
 lengths of the rays of which 
 it consists are at moderate 
 temperatures longer than 
 those corresponding to red 
 light, and so they do not 
 present the phenomena of 
 light. If, however, the 
 temperature of a body is 
 increased, it begins to glow 
 with a dull red light, which 
 passes through shades of yel- 
 low, violet, and blue, until 
 an intensely heated body is 
 said to be incandescent, 
 which means that it gives off 
 a light as white as that of 
 
 the sun, and which contains 
 
 i . . FIG. 10. 
 
 in their proper proportions UNDERGROUND 
 
 all the colours of sunlight. 
 Radiant heat, then, spreads along straight lines, diverging in 
 all directions from the source of heat. " Its intensity," says 
 
 7 
 
 FIG. 9. UNDERGROUND 
 THKRMOMETKR. 
 
98 METEOROLOGY 
 
 Dr. Alex. Buchan, 1 " is proportioned to the temperature of the 
 source, is inversely as the square of the distance from the source, 
 and is greater according to the degree of inclination of the surface 
 on which the rays fall." 
 
 Heat is radiated towards the earth from the fixed stars, the 
 planets, the moon, but, above all, the sun. Indeed, for all prac- 
 tical meteorological purposes we may assume that the sun's rays 
 are the only source whence heat reaches the earth's surface. We 
 speak, therefore, of solar radiation alone in connection with the 
 receipt of heat by the earth. 
 
 In a paper on the " Conservation of Solar Energy," read before 
 the Royal Society on March 2, 1882, Dr. C. William Siemens, 
 D.C.L., F.R.S., observed : " The amount of heat radiated from 
 the sun has been approximately computed by the aid of the 
 pyrheliometer of Pouillet, and by the actinometers of Herschel 
 and others, at 18,000,000 of heat units from every square foot of 
 its surface per hour, or, put popularly, as equal to the heat that 
 would be produced by the perfect combustion every thirty-six 
 hours of a mass of coal, of specific gravity =1*5, as great as that 
 of our earth. 
 
 " If the sun were surrounded by a solid sphere of a radius equal 
 to the mean distance of the sun from the earth (95,000,000 of 
 miles), the whole of this prodigious amount of heat would be 
 intercepted ; but considering that the earth's apparent diameter, 
 as seen from the sun, is only seventeen seconds, the earth can 
 intercept only the 2,250 millionth part." 
 
 In accordance with physical laws, no sooner does the earth 
 receive heat from the sun than it begins to radiate it back again 
 into space in all directions. Hence we speak of terrestrial radiation. 
 
 From what has been stated above in a quotation from Dr. 
 Buchan, it is clear that solar radiation is much less in winter 
 than in summer, owing to increased inclination ; but it is also less 
 in July than in December (taking the earth as a whole), because 
 in the latter month the sun and the earth are some 3,000,000 of 
 miles nearer to each other than in the former. According to 
 Dr. R. H. Scott, F.R.S., with the existing value of the eccen- 
 
 1 Introductory Text-Book of Meteorology, p. 48. William Blackwood and 
 Sons : Edinburgh and London. 1871. 
 
KADIATION 99 
 
 tricity of the earth's orbit, the amount of heat received in peri- 
 helion (the southern summer) is to that received in aphelion (the 
 northern summer) as 1*034 is to 0*967. 
 
 Solar radiation is also interfered with by clouds, but is not 
 materially affected by the air through which it passes, nor is it 
 diverted from a straight course by the wind (Buchan). 
 
 Terrestrial radiation tends to dissipate into space the heat 
 which the earth has received from the sun, and as a consequence 
 temperature falls in winter, when the slanting rays of the sun 
 pour down less upon the earth's surface. Again, not only the 
 seasonal, but also the diurnal, range of temperature depends on 
 radiation. By day, solar radiation predominates and tempera- 
 ture rises ; by night, solar radiation ceases while terrestrial radia- 
 tion continues, and so temperature falls. Just as solar radiation 
 is interfered with by clouds, so an overcast sky interrupts terres- 
 trial radiation. Hence dew is not deposited on a cloudy night, 
 because the thermometer does not fall below the temperature of 
 saturation, or the dew-point. But even an excess of moisture 
 in the atmosphere interferes with terrestrial radiation, so that 
 very low temperatures are never felt in damp weather, while 
 severe frosts occur in spring nights, when the air is very dry and 
 the sky is often clear. 
 
 A covering of snow at one and the same time prevents and 
 facilitates radiation. The explanation of this paradox is that 
 the snow acts like a cloud canopy, and interferes with radiation 
 from the surface of the ground, which is in this way kept warm. 
 But, further, snow is a bad conductor of heat, and so the warmth 
 is imprisoned beneath it through non-conduction. Hence the 
 surface of the snow becomes intensely cold, for no heat reaches it 
 from below, while it radiates freely into space what heat it does 
 already possess. 
 
 Dr. Hann 1 calculated that in a vertical column of absolutely 
 dry air the thermometer should fall 1 F. for every 182 feet of 
 altitude, or 1 C. for every 100 metres. In nature, however, 
 there is practically no such thing as absolutely dry air. The 
 moisture in the atmosphere, then, is liable to be condensed as the 
 temperature falls with increasing altitude. But in the process 
 1 Attgemeine Erdkunde. Third edition. Tempsky : Prague. 1881. 
 
 72 
 
100 METEOROLOGY 
 
 of condensation latent heat is set free in large quantities with the 
 effect of lessening the rate of cooling in the vertical column of 
 air. Sir John Herschel long ago calculated that the slower rate 
 of cooling amounted to about 1 F. for every 300 feet of vertical 
 height, and this value is generally accepted. It is, however, 
 necessary to explain that sometimes in winter conditions of 
 temperature are actually reversed, and an "upbank thaw/' as 
 it is called, may occur on mountains with the arrival on their 
 slopes of a warm air current, while the cold, dense air in the 
 valleys and plains below may cause unbroken frosts at lower levels. 
 Recent observations on the upper air by balloons and kites go to 
 show that inversion of temperature is a very common phenome- 
 non, particularly in winter. By the term is meant that a stratum 
 of warmer air may be superimposed upon one of colder air 
 nearer the earth's surface. 
 
 Further, the descent of temperature with increasing altitude 
 does not go on indefinitely. Within the first two miles from the 
 ground the temperature variations are very complex there is 
 often " inversion/' Above the two-mile limit a very nearly 
 uniform rate of fall of temperature is observed until what is 
 called the " isothermal layer " of the atmosphere is reached, at 
 from six to eight miles above the earth's surface (12 kilometres, 
 or nearly 1\ miles). This question will be discussed later on 
 (see p. 325). 
 
 In estimating the influence of radiation upon climate, it is to be 
 borne in mind that the specific heat of water is much higher than 
 that of land in the proportion of about four to one. Hence solar 
 radiation heats water much more slowly than it heats dry land, 
 and, again, water cools in turn much more slowly than dry land 
 does. In these facts we recognise one explanation of the modify- 
 ing and mollifying effects of the ocean upon climate. Its pres- 
 ence controls temperature, forbidding it to rise quickly in summer 
 or to fall quickly in winter. Of course, by convection also, cur- 
 rents of cool water flow towards warm regions, and currents of 
 warm water towards cold regions. 
 
 We are now in a position to resume the description of various 
 thermometers employed in meteorological observations, which 
 was begun in the preceding chapter. 
 
RADIATION 
 
 101 
 
 5. Radiation Thermometers. Solar radiation is measured by 
 the black-bulb thermometer in vacuo, an instrument which was 
 first suggested by Sir John Herschel. The late Rev. Fenwick W. 
 Stow, M.A., of Aysgarth Vicarage, Bedale, Yorkshire, described 
 this instrument as follows : The insulated solar maximum ther- 
 mometer, usually called the black bulb in vacuo, is a sensitive 
 maximum thermometer, having the bulb and a given portion of 
 the stem covered with lamp-black, the whole being enclosed in 
 a glass tube from which all air and moisture have been removed, 
 so that the heat of the sun's rays is 
 thus obtained, apart from the influence 
 of vapour or passing currents of air. 
 The stem near the bulb must be 
 blackened to prevent reduction of 
 temperature in the bulb through con- 
 duction, the bright stem chilled by 
 radiation in this way affecting the 
 bulb. This delicate instrument should 
 be placed on a stand 4 feet above the 
 ground, in an open space, with its 
 bulb directed towards the south-east, 
 and free from contact with any sub- 
 stance whatever. 
 
 The Royal Meteorological Society 
 recommends the use, in addition to 
 the black bulb, of a bright-bulb ther- 
 mometer in vacuo. The readings of 
 this latter instrument will, of course, 
 be lower than those of the black bulb, 
 because the bright bulb will radiate freely the heat which it 
 receives from the sun's rays. Fig. 11 represents these ther- 
 mometers in situ. 
 
 The black-bulb and bright-bulb thermometers in vacuo should 
 be tested in sunshine at Kew Observatory after enclosure in their 
 vacuum jackets. The corrections usually given on the Kew 
 certificate apply merely to the instruments before they are 
 enclosed in the outer jacket. 
 
 The helio-pyrometer, arranged by Mr. T. Southall, of Birming- 
 
 
 Fio. 11. SOLAR RADIATION- 
 THERMOMETER STAND. 
 
102 METEOEOLOGY 
 
 ham, gives extraordinary readings at times (216, 217, and 
 even 231 '5 in July, 1859), and these readings are confirmed by 
 water being caused to boil violently in a small vessel attached to 
 the apparatus. One of Casella's solar radiation maximum ther- 
 mometers, made on Professor Phillips's principle, is fixed on a 
 cushion at the bottom of a box, the sides of which are also 
 cushioned, and a thick piece of plate glass is laid upon the top to 
 prevent currents of air carrying off the heat as well as with the 
 view of preventing the cooling effects of terrestrial radiation. 
 The box is placed in such a position that the sun's rays may fall 
 as nearly as possible perpendicularly on the glass. A change of 
 position to secure this end may be required twice or three times 
 a day. No doubt a portion of the sun's heat is lost by reflection 
 from the surface of the plate-glass cover, but the amount of the 
 loss can be calculated. 
 
 Other instruments for measuring the intensity of solar radiation 
 which deserve mention are : Sir John Herschel's actinometer 
 (Greek, UKTIS, a ray ; perpov, a measure), Padre Secchi's solar 
 intensity apparatus, and Pouillet's pyrheliometer (Greek, -rrvp, fire 
 or heat ; yjkios, the sun ; ptrpov, a measure}. By means of this 
 last instrument the effect of the sun's heat upon a given area 
 is ascertained by the number of degrees of heat imparted to a 
 given quantity of mercury in five minutes. 
 
 At the International Meteorological Conference held at Inns- 
 bruck in September, 1905, Angstrom's electric compensation 
 pyrheliometer and actinometer were recognised as satisfactory 
 instruments for absolute actinometric measurements. In con- 
 nection with a report on actinometry by M. Violle, the Con- 
 ference resolved that measurements of the total solar radiation 
 be made at central observatories, and at other stations which 
 possess the facilities to do so, regularly each day at 11 a.m., or 
 from 11 a.m. to 1 p.m. Angstrom's compensation pyrhelio- 
 meter should be used exclusively for these measurements, as well 
 as for measurements of terrestrial radiation to be made each 
 evening at 10 p.m., or from 10 p.m. to midnight. 
 
 The Richard system for recording solar heat (actinometer) 
 is partly based upon researches made by Professor Violle, and 
 is represented in Fig. 12. Two thermometers, the bulb of one 
 
RADIATION 
 
 103 
 
 of which is bright, while that of the other is a dull black, are pro- 
 tected by glass spheres and record on a single sheet, so that the 
 difference of their readings, and also the times of their respective 
 maxima, can be easily seen. 
 
 FIG. 12. RICHARD'S ACTINOMETER. 
 
 When using the black bulb in vacuo, observations should also 
 be made with the ordinary maximum thermometer in the shade. 
 The greatest amount of radiation during the day will then be 
 
104 
 
 METEOKOLOGY 
 
 approximately indicated by subtracting the maximal tempera- 
 ture in the shade from the maximal reading recorded by the solar 
 radiation thermometer. The difference may usually be regarded 
 as an index of the intensity of solar radiation. 
 
 The Wilson Radio-Integrator. The late Dr. W. E. Wilson, 
 F.R.S., of Daramona, Streete, Co. Westmeath, shortly before his 
 death designed an ingenious instrument for recording the total 
 amount of solar radiation daily received by the ground. The 
 general form of the instrument is shown in the accompanying 
 illustration (Fig. 13). The radio-integrator, as it is called, con- 
 sists of a sort of sealed retort for the distillation of a volatile 
 liquid, presumably in vacuo, by the heat of the sun. When set 
 for an observation, the whole of the liquid is in the upper bulb, 
 
 which is exposed to the 
 sun, while the lower bulb 
 and tube are sheltered 
 in a white-painted per- 
 forated box. The liquid 
 as it evaporates is con- 
 densed in the lower bulb, 
 and trickles down into 
 the tube, which is gradu- 
 
 Ijljrajl ated according to an arbi- 
 trary scale. The amount 
 of liquid evaporated in 
 ^% : -J '^SjJSIB l&i the previous twenty-four 
 
 FIG. IS.-WILSON'S PATENT RADIO-INTEGRATOR. hours is read daily at 
 
 9 a.m., and recorded in 
 
 terms of the divisions engraved on the glass tube. This instru- 
 ment has been in use by Dr. H. R. Mill at Camden Square, London, 
 since July, 1907. 
 
 Terrestrial Radiation. The thermometer used for registering 
 this meteorological factor is a delicate self-registering spirit 
 minimum thermometer, of Rutherford's construction, which is 
 enclosed in a glass cylinder for protection. To increase the 
 sensitiveness of the spirit " grass minimum," Mr. Casella designed 
 a thermometer in which the bulb, being extended in a forked form 
 exposes a greatly increased surface to the air (Fig. 14). In this 
 
RADIATION 
 
 105 
 
 way the instrument is rendered little, if at all, less sensitive than 
 Mr. Casella's mercurial minimum already described. 
 
 A thermometer intended to measure terrestrial radiation 
 should be suspended over a piece of smooth lawn grass on wooden 
 
 Fia. 14. CASELLA'S BIFURCATED GRASS MINIMUM. 
 
 props shaped like Y's, at a height of only 3 inches or so, in order 
 to escape the disturbing influence of the wind. The amount of 
 terrestrial radiation is determined by subtracting from the 
 minimal temperature recorded in the thermometer screen the 
 minimum registered on the grass. Should 
 the ground be covered with snow, the 
 radiation thermometer should be laid 
 upon the surface of the snow. Where 
 a grass plot is not available, the ther- 
 mometer should be placed on a large 
 black board laid upon the ground. 
 
 Earth Temperatures. In connection 
 with radiation it is desirable to ascer- 
 tain the temperature of the soil at fixed 
 depths. This may best be done by using 
 Symons's earth thermometer (Fig. 15). l 
 It is a sluggish thermometer mounted 
 on a short weighted stick attached to a 
 strong chain. It is lowered by means of 
 this chain into a long, stout, iron pipe, 
 pointed at the lower end, and driven 
 into the earth to any required depth 
 1 foot, 2 feet, 3 feet, or 4 feet below the 
 surface. The top of the iron pipe or 
 tube is closed by a tight-fitting iron cap. 
 driven into the soil below short grass, and in a well-exposed 
 situation. 
 
 Mr. Casella also has designed a self-registering thermometer 
 
 1 " Improved Form of Thermometer for observing Earth Temperature," 
 By G. J. Symons, F.M.S. Quarterly Journal of the Meteorological Society, 
 vol. iii., p. 421, 1877. 
 
 Fio. 15. SYMONS'S EARTH 
 THERMOMETER. 
 
 The tube should be 
 
106 METEOROLOGY 
 
 for immersion to any depth in the earth or wells, where it will 
 record the maximum and minimum temperatures for a required 
 interval of time. 
 
 From observations at the Calton Hill, Edinburgh, Principal 
 Forbes concludes that the seasonal variations of temperature are 
 reversed at a depth of 24 feet the greatest warmth occurring 
 at that depth about January 4, and the greatest cold about 
 July 13. Below 40 feet there is no annual variation in the 
 temperature of the soil. 
 
 The temperature of the soil, as shown by the earth ther- 
 mometer, has a vital bearing on public health. Systematic 
 observations at the City Meteorological Observatory, 299, Oldham 
 Road, Manchester, convinced Dr. John Tatham, formerly the 
 able Medical Officer of Health for that great city, and more 
 recently Medical Superintendent of Statistics at Somerset House, 
 that, in accordance with the views of Dr. Edward Ballard, F.R.S., 
 when the earth-temperature at the depth of 4 feet from the sur- 
 face rises to 56 F. infantile diarrhoea may be expected to become 
 epidemic in the city. Dr. Ballard's proposition was that the 
 temperature of the soil is a far more effective element in raising 
 the death-rate from diarrhoeal diseases than any other meteoro- 
 logical factor. He constructed for London and many other 
 towns in the Kingdom a large number of charts showing week by 
 week for many years the earth-temperature at a depth of 1 foot 
 from the surface and at a depth of 4 feet also, each chart showing 
 in addition the diarrhceal mortality of the corresponding weeks. 1 
 
 The general result shown by these charts is as follows : 
 
 1. The summer rise of diarrhoeal mortality does not commence 
 until the mean temperature recorded by the 4-foot earth ther- 
 mometer has attained somewhere about 56 F., no matter what 
 may have been the temperature previously attained by the 
 atmosphere or recorded by the 1-foot earth thermometer. 
 
 2. The maximal diarrhosal mortality of the year is usually 
 observed in the week in which the temperature recorded by the 
 4-foot earth thermometer attains its mean weekly maximum. 
 
 1 Supplement in Continuation of the Report of the Medical Officer for 1887. 
 Annual Report of the Local Government Board, 1887-88, p. 1, et s:q. 
 London : Eyre and Spottiswoode. 1889. Quarto. 
 
RADIATION 107 
 
 3. The decline of the diarrhoeal mortality coincides with the 
 decline of the temperature recorded by the 4-foot earth ther- 
 mometer, which temperature declines very much more slowly 
 than the atmospheric temperature, or than that recorded by the 
 1-foot earth thermometer. The epidemic mortality may in 
 consequence continue (although declining) long after the last- 
 mentioned temperatures have fallen greatly, and may extend 
 some way into the fourth quarter of the year. 
 
 4. The atmospheric temperature and that of the more super- 
 ficial layers of the soil exert little, if any, influence on the preva- 
 lence of diarrhoea until the temperature recorded by the 4-foot 
 earth thermometer has risen to 56 F. Then their influence is 
 apparent, but it is a subsidiary one, notwithstanding the state- 
 ment made by Dr. August Hirsch that the summer diarrhoea of 
 children makes its appearance as an epidemic only in those dis- 
 tricts whose average temperature for the day in the warm season 
 is rather more than 15 C. (59 F.). 
 
 On January 1, 1904, through the liberality of the Provost and 
 Senior Fellows of Trinity College, a Normal Climatological 
 Station was established within the precincts of the University 
 of Dublin. The station, which is under the supervision of Pro- 
 fessor W. E. Thrift, M.A., F.T.C.D., occupies an open space in 
 the Fellows' Garden, Trinity College, and is fully equipped. At 
 the suggestion of Dr. William Napier Shaw, F.R.S., Director of 
 the Meteorological Office, London, the equipment included two 
 earth thermometers. One of these has its bulb at a depth of 
 12 inches (1-foot earth thermometer) below the surface of the 
 ground. The bulb of the other is sunk in a metal tube to a depth 
 of 4 feet. 
 
 In a paper read before the State Medicine Section of the Royal 
 Academy of Medicine in Ireland on February 10, 1905, I discussed 
 the question of Earth Temperature and Diarrhceal Diseases in 
 Dublin during 1904. The observer, my son, Arthur Robert 
 Moore, M.A., threw the figures into two diagrams, of which the 
 second is here reproduced. It contains two weekly curves for 
 the whole year 1904. The upper of these represents the weekly 
 march of underground temperature at a depth of 4 feet. 
 The lower curve gives the number of deaths from diarrhoeal 
 
108 
 
 METEOROLOGY 
 
 diseases registered week by week in 1904 in the Dublin Registra- 
 tion Area. 
 
 Reference to the curves in the diagram shows that diarrhoeal 
 mortality in the Dublin district in 1904 was trifling till the week 
 ended August 6 that is, the third week after the subsoil tempera- 
 ture at 4 feet had passed above 56 F. The mortality rapidly 
 increased till the week ended August 27, in which thirty-five 
 deaths from diarrhceal diseases were registered, or about 10 per 
 cent, of all the deaths from those diseases in the whole year 1904. 
 
 FIG. 16. ILLUSTRATING THE RKLATION BETWEKN UNDETOROUND TEMPERATURE AND THE 
 DEATH-RATE FROM DIARRHCEAL DISEASES. 
 
 This maximum of mortality followed the maximum of warmth 
 of the soil at 4 feet (58'5) by an interval of just a fortnight. Such 
 a coincidence is remarkable. Diarrhoea kills very young children 
 quickly usually within a week. Then, allowing a few days for 
 delay in registration, we come to the close of the second week. 
 
 The diagram also shows that the 4-foot thermometer stood at 
 56 or upwards from July 10 to September 24 a period of 
 eleven weeks. Starting a similar period of eleven weeks a fort- 
 night later (to allow time for the malady to attack and kill, and 
 for registration of the resulting deaths), we find that in the eleven 
 
RADIATION 109 
 
 weeks beginning July 24 and ending October 8, the diarrkceal 
 deaths were 249, or 71*7 per cent, of the total deaths from 
 diarrhceal diseases registered in 1904, 339 in number. Of these 
 only 18 were registered in the first quarter of the year, only 10 in 
 the second, 243 in the third, and 68 in the fourth quarter. 
 
 There can be no doubt that the prevalence of cholerine or 
 epidemic summer diarrhoea, the prevalence of enteric fever, and 
 that of cholera are all equally determined by this critical subsoil 
 temperature of 56, probably not directly, in the way of cause 
 and effect, but indirectly, by promoting the decomposition of 
 human food-stuffs, especially milk, through the agency of flies 
 or other carriers of saprophytic or pathogenic organisms. 
 
 Duration of Bright Sunshine. Within the past twenty-five 
 years striking observations have been made on this point under 
 the auspices of the Meteorological Office, London, and of the 
 Royal Meteorological Society and the Scottish Meteorological 
 Society. It is manifest that the amount of solar radiation will 
 depend on the amount of bright sunshine. A remarkable instance 
 of this occurred in the spring of 1893, when the registered amount 
 of bright sunshine was much in excess of the average, and when 
 solar radiation was so powerful as to cause a marvellous blossom- 
 ing not only of the ordinary spring flowers and shrubs, but also 
 of shrubs which rarely flower in ordinary years in the climate of 
 the British Isles. 
 
 The instruments which are used for recording the duration of 
 sunshine are (1) the Campbell-Stokes Burning Recorder ; (2) the 
 Whipple-Casella Universal Sunshine Recorder ; (3) the Jordan 
 Photographic Recorder. The principle of the first two of these 
 instruments is the same. It consists in burning a tracing in a 
 piece of mill-board placed in the focus of rays from a glass sphere, 
 which acts as a lens when exposed to bright sunshine. The 
 burning recorder was originally devised in 1854 by Mr. John F. 
 Campbell, F.G.S., of Islay, and was improved by Professor Sir 
 George G. Stokes, Bart., F.R.S., of Cambridge. 
 
 1. The Campbell-Stokes Sunshine Recorder consists of a 
 sphere of crown glass 4 inches in diameter and 3 pounds in 
 weight, supported on a pedestal in a metal zodiacal frame 
 (Fig. 17). It should be fixed in an open position, so that the 
 
110 METEOKOLOGY 
 
 sun's rays may fall upon it at any time between sunrise and 
 sunset. In must face due south, so that the sun's image shall 
 fall upon the meridian line marked on the inside of the ring 
 supporting the recording cards when the sun is itself upon the 
 meridian. The axis of the ring in question must be inclined to 
 the horizon at an angle equal to the latitude of the place, and the 
 instrument must be level as regards east and west. 1 There are 
 three grooves in the ring which supports the card : one holds 
 rectangular cards with hour figures printed upon them suitable 
 for the equinoxes (from March 1 to April 12, and again from 
 September 1 to October 12) ; one, long curved cards similarly 
 
 FIG. 17. THE CAMPBELL-STOKES SUNSHINE .RECORDER. 
 
 time-marked for summer (from April 13 to August 31) ; and one, 
 short curved cards for winter (from October 13 to February 28 
 or 29). A card being fixed in the proper groove according to the 
 season of the year, the sun when shining burns away or chars 
 the surface at the points on which its image falls from moment to 
 moment, and thus a tracing of bright sunshine is given. A card 
 should be removed daily after sunset, and a new one inserted 
 ready for next day. This apparatus costs 9 9s. 
 
 2. The Whipple-Casella modification of this instrument has 
 divided latitude and diurnal circles, so that it can be set for any 
 
 1 Quarterly Journal of the Meteorological Society, 1880, vol. vi., p. 83. 
 " Description of the Card Supporter for Sunshine Recorders adopted at the 
 Meteorological Office." By Professor George Gabriel Stokes, M.A., F.R.S. 
 
KADIATION 
 
 111 
 
 locality and for any day in the year, thus earning its name of 
 " Universal Sunshine Recorder." It is an expensive instrument, 
 costing 15 ; but, owing to its powers of adjustment to time and 
 place, it requires merely a strip of cardboard duly hour-marked 
 instead of Sir George G. Stokes's equinoctial and summer and 
 winter cards (see Fig. 18). 
 
 FIG. 18. THE WHIPPLE-CASELLA UNIVERSAL SUNSHINE RECORDER. 
 
 3. In 1838 an automatic Daylight or Sunlight Recorder was 
 invented and constructed by Mr. T. B. Jordan, who wrote and 
 published an account of his invention in the Sixth Annual Report 
 of the Royal Cornwall Polytechnic Society (p. 185). The instrument, 
 however, which now goes by the name of the " Jordan Photo- 
 graphic Sunshine Recorder," was designed in 1885 by Mr. James B. 
 
112 
 
 METEOROLOGY 
 
 FIG. 10. JORDAN PHOTOGRAPHIC SUNSHINE RECORDER. 
 
 FIG. 20. IMPROVED JORDAN PHOTOGRAPHIC SUNSHINE RECORDER. 
 
 Jordan, of the Home Office. 1 Two forms of the Jordan Photo- 
 graphic Sunshine Recorder are in use. The first pattern, repre- 
 sented in Fig. 19, brought out in 1885, consists of a cylindrical 
 box, on the inside of which a sheet of sensitive cyanotype 
 paper is carefully placed day by day. The sunlight is admitted 
 
 1 Quarterly Journal of the Royal Meteorological Society, 1886, vol. xii., p. 23. 
 
KADIATION 113 
 
 into the box by two small apertures, and acts on the paper 
 chemically, leaving a tracing as the ray travels across it owing 
 to the earth's rotation. The improved pattern Jordan's twin- 
 cylinder recorder (Fig. 20) has two semi-cylindrical boxes, one to 
 hold the forenoon, the other the afternoon, prepared charts of 
 sensitised paper. A slit, through which the beam of sunlight finds 
 entrance, is placed in the centre of the rectangular side of each box, 
 so that the length of the beam within the chamber is the radius of 
 the cylindrical surface on which it is projected. The path of the 
 sunbeam, therefore, follows a straight line on the sensitive paper 
 at all seasons. The instrument must be carefully adjusted to the 
 meridian and to the latitude of the place, and must be firmly 
 fixed (William Marriott, F.R.Met.Soc.). 
 
 The sunshine recorded by any of these instruments should be 
 measured in hours and tenths of an hour, and not in minutes. 
 A " sunless day " is that on which the record of bright sunshine 
 is less than three minutes. 
 
CHAPTEK X 
 
 ATMOSPHERIC PRESSURE 
 
 WE have already seen in Chapter II. that the atmosphere has 
 weight, and can be weighed. The instrument by which this is 
 effected is called the barometer (Greek, /2a/oos, weight ; /xer/aov, a 
 measure). But it would be most misleading to suppose that the 
 only use of the barometer is the mere weighing of the atmosphere. 
 By a careful study of the properties of this marvellous instrument 
 we are enabled to measure the heights of mountains, to ascertain 
 the distribution of atmospheric pressure over the earth's surface 
 by sea as well as by land, and at the different seasons of the year ; 
 to understand in consequence the prevalent winds at all times and 
 in all places, to trace the ever-shifting distribution of atmospheric 
 pressure over vast districts, and finally, to " forecast " the 
 weather. This may be done either by a consideration of baro- 
 metrical observations taken at a single station, or by means of 
 telegraphic information as to a number of such observations taken 
 synchronously (or at the same moment of time) at many stations 
 scattered over a large area, like the west, north-west, and centre of 
 Europe, or the United States of America and Canada. 
 
 Surely such far-reaching potentialities as those now indicated 
 bespeak for so wonderful an instrument our liveliest interest and 
 most attentive study. 
 
 An observation of Galileo Galilei, of Pisa, the father of experi- 
 mental science, that water would not rise in a pump more than 
 " eighteen cubits " (diciotto braccia) above the level of a well, led 
 to the discovery of the pressure of the atmosphere by Evangelista 
 Torricelli, his pupil and successor in the Chair of Philosophy and 
 Mathematics at Florence, who also devised the means of measuring 
 that pressure. Torricelli's famous experiment was made in 1643 
 
 114 
 
ATMOSPHERIC PRESSURE 
 
 115 
 
 He was testing Galileo's dictum that " Nature abhors a vacuum " 
 (up to 32 feet, in the case of water), and for convenience employed 
 mercury. By doing so, he found that Nature's abhorrence of 
 a vacuum varied for different fluids. Torricelli filled a tube 
 (Fig. 21, C D) 3 feet long with mercury, and then inverted it 
 and plunged its lower end into a basin filled half with mercury 
 and half with water. So long as the lower end of the tube 
 remained below the level of 
 the mercury in the basin, the 
 height of the column of mercury 
 in the tube proved to be about 
 30 inches, and a vacant space 
 of 6 inches was left at the top 
 of the tube a space which 
 afterwards came to be, and is 
 still, known as the Torricellian 
 vacuum (Fig. 21, A B). The 
 moment, however, that the 
 lower end of the tube was 
 raised above the surface of the 
 mercury in the basin into the 
 overlying stratum of water, all 
 the mercury in the tube rushed 
 out, its place being taken by the 
 water, which equally readily 
 rushed in and filed the tube 
 completely. Reasoning out the 
 matter, the philosopher con- 
 cluded that some one force ex- 
 isted which was able to support 
 a column of mercury to a height of 30 inches in the tube, but a 
 column of water to a much greater height. This force could be 
 none other than the pressure of the atmosphere on the open sur- 
 face of the fluids mercury and water in the basin. Thus was 
 the barometer discovered. 
 
 Torricelli further proved that Nature's abhorrence of a vacuum 
 was represented by a column of fluid inversely proportional to 
 its specific gravity. Take the very fluids under consideration 
 
 82 
 
 FIG. 21. TORRICELLI'S EXPERIMENT. 
 
116 METEOROLOGY 
 
 water and mercury ; the specific gravity of water being 1, that of 
 mercury is 13*594 and we get this inverse proportion : 
 
 1 : 13*594 : : 30 inches : x = the required height of a column of water 
 
 supported by the atmosphere. 
 13-594 x 30 inches=407-82 inches=33'99 feet. 9 
 
 This principle has been taken advantage of in selecting fluids 
 for the construction of a barometer. Thus mercury is one of 
 the handiest, because, in addition to other recommendations, it 
 requires a tube only 32 inches long in consequence of its high 
 specific gravity. On the other hand, if we could use water in an 
 immense tube 35 or 36 feet in length, the smallest variations in 
 atmospheric pressure could easily be observed. Water, however, is 
 not available because of its high freezing-point. Hence we select 
 glycerine, the specific gravity of which is 1-26, and which, while 
 undiluted, does not freeze at any known terrestrial temperature (a 
 50 per cent, solution freezes at - 31 C., or - 23*8 F.). In practice 
 we find that a column of glycerine 27 or 28 feet high will yield most 
 valuable barometrical indications. The proportional statement is : 
 
 1-26 : 1 :: 33 '99 feet: x=the required height of the column of 
 
 glycerine in feet or inches 
 1-26)33-99(26-976 feet, or 323-7 inches. 
 252 
 
 ~879 
 756 
 1230 
 1134 
 
 960 
 882_ 
 780 
 756 
 
 24 
 
 The proportional statement between glycerine and mercury 
 works out as follows : 
 
 1'26 : 13-594 : : 30 inches : x=the required height of the column of 
 
 glycerine in inches or feet 
 
 l-26)13-594x 30, i.e., 407-82(323-66 inches=323-7 inches, quam proxime 
 378 =26-975 feet. 
 
 "298 
 252 
 "462 
 378 
 840 
 
 756 V" 
 
 '84 
 
ATMOSPHERIC PRESSURE 
 
 117 
 
 Jordan's glycerine barometer, used at the Times office, London, 
 consists of a gas tube, f inch in diameter and 28 feet in height. 
 As glycerine has a singular affinity for water, the glycerine in the 
 cistern of this gigantic barometer is covered with a layer of 
 paraffin-oil. 
 
 The advantage of the glycerine barometer, then, is that it 
 magnifies tenfold, as it were, the readings of the mercurial baro- 
 meter 323 '1 inches on the scale of the glycerine barometer 
 corresponding to 30 inches on that of the mercurial barometer. 
 
 THE TIMES OFFICE, 2 A.M. 
 
 READINGS OF THE JOEDAN BAROMETER (CORRECTED) 
 
 DURING THE PAST TWENTY-FOUR HOURS. 
 
 FEBRUARY 2627. 
 
 1 
 
 
 
 A.W 
 
 L 
 
 
 
 
 
 P.M 
 
 
 
 
 fc 
 
 j? 
 
 Inches. 
 
 i 
 
 \ ( 
 
 \ 
 
 > 8 
 
 | 
 
 l 
 
 * 
 
 r. j 
 
 < 
 
 1 
 
 5 ( 
 
 1 1 
 
 ft 
 
 I. 2 
 | 
 
 * 1 
 Inchea 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 320 _ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 * 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 .29-7 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 319. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -29-6 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 FTG. 
 
 . 
 
 
 
 
 
 
 
 Another interesting application of the principle that the 
 heights of columns of liquids or gases are inversely proportional 
 to their specific gravities is the attempt to determine the height of 
 the atmosphere. As air is about 10,000 times lighter than mer- 
 cury, the height of the atmosphere should on this principle be 
 10,000 times 30 inches, or 300,000 inches that is, 4*7 miles. In 
 Chapter II., however, it has been shown that the density of the 
 aerial column lessens according to its height, and so, as a matter 
 
118 METEOROLOGY 
 
 of fact, the height of the atmosphere is vastly greater than 
 4-7 miles an altitude which falls short of the highest peak of the 
 Himalayas by 4,000 feet. 
 
 Torricelli's surmises received their full confirmation at the 
 hands of the French philosopher, Blaise Pascal, of Clermont, in 
 Auvergne. In 1647 he was in Paris, when the thought struck 
 him that if the Torricellian theory of atmospheric pressure was 
 correct, the height of the column of mercury supported by the 
 air should be less on the top than at the foot of a mountain. He 
 accordingly wrote to his brother-in-law, Perrier, who lived at 
 Clermont, to request him to ascend the neighbouring Puy de 
 Dome, with the Torricellian tube in his hands. It was not until 
 September 19, 1648, that Perrier was able to carry out the long- 
 projected experiment. In the presence of a distinguished com- 
 pany of savants in Clermont he on that day repeatedly performed 
 the Torricellian experiment. The party then ascended the 
 mountain, which at a distance of eight miles rises some 3,510 feet 
 above Clermont, and, to Perrier' s surprise, and ultimately to 
 Pascal's delight, the mercury was found to stand 3*33 inches 
 lower on the summit than at Clermont. On the way down, at 
 Font de 1'Arbre, the column was proved to have an intermediate 
 height. Perrier' s observations on this memorable day gave 
 3,458 feet for the height of the Puy de Dome above Clermont, and 
 the actual height is now stated to be 3,511 feet. The account of 
 this experiment was given by Blaise Pascal himself in a pamphlet 
 published in Paris in 1648, and entitled "Recit de la grande Ex- 
 perience de 1'Equilibre des Liqueurs." 1 
 
 During the years 1649-50 readings of the " Torricellian 
 column/' as it was called, were taken daily, and, at the same time, 
 by Pascal at Paris, Perrier at Clermont, and Chanut and Descartes 
 at Stockholm, " in order to see if anything could be discovered 
 by confronting them with one another." Mr. Richard Strachan, 
 F.R.Met.Soc., who gives much of the foregoing information in 
 a lecture delivered under the auspices of the Meteorological 
 Society in 1878, 2 observes : " Pascal was thus the pioneer of the 
 
 1 Neudrucke von Schriften und Karten uber Meteorologie und Erdmagnet- 
 ismus. Herausgegeben von Professor Dr. G. Hellmann. 4to. Berlin : 
 A. Ascher and Co. 1893. 
 
 2 Modern Meteorology, p. 70 et seq. London : Edward Stanford. 1879. 
 
ATMOSPHERIC PRESSURE 119 
 
 synchronous observations upon which modern storm-warnings 
 depend." 
 
 In 1665 the Hon. Robert Boyle observed the Torricellian 
 column in relation to weather, and gave it a scale and lettering. 
 In the same year Robert Hooke invented the " weather glass," 
 or wheel barometer. 
 
 In the Philosophical Transactions for 1666, p. 153, we read 
 that " Modern Philosophers, to avoid Circumlocutions, call that 
 Instrument, wherein a Cylinder of Quicksilver, of between 
 28 and 31 inches in Altitude, is kept suspended after the manner 
 of the Torricellian Experiment, a Barometer or Baroscope 1 . . . 
 to detect all the minute variations in the Pressure and weight of 
 the air." 
 
 A very full historical account of the barometer was communi- 
 cated to the Royal Meteorological Society on March 17, 1886, by 
 the President, Mr. William Ellis, F.R.A.S., of the Royal Obser- 
 vatory, Greenwich. His Presidential Address will be found in 
 the twelfth volume of the Quarterly Journal of the Royal Meteoro- 
 logical Society (No. 59, July 1886, p. 131). 
 
 1 Greek, /3a/>os, weight ; (TKoirtw, I inspect. 
 
CHAPTER XI 
 THE BAROMETER 
 
 THE most usual form of barometer is a glass tube, about 34 inches 
 in length, closed at one end and carefully filled with pure mercury 
 of the specific gravity of 13-594. If necessary, the mercury in the 
 tube should be boiled to expel ail air. The tube is then placed 
 vertically, with its open end dipping into a cup of mercury called 
 the " cistern." When so placed the mercury falls in the tube at 
 sea-level to 30 inches, or some point not very much above or 
 below that level, according to the pressure of the atmosphere at 
 the time. 
 
 In Dublin the monthly mean atmospheric pressure rises to 
 29'994 inches in June, and falls to 29'863 inches in December. 
 The absolute extreme readings of the barometer at any time taken 
 by me were : maximum, 31-020 inches, at 10 a.m. of January 9, 
 1896 ; minimum, 27-758 inches, at 2.30 p.m. of December 8, 1886. 
 These readings assuredly represent the extreme range of atmo- 
 spheric pressure, reduced to 32 F., and to mean sea-level, in 
 Dublin namely, 3*262 inches, rather more than 3J inches. 
 
 Incidentally, I may mention that, in the depression of De- 
 cember 8, 1886, Armagh Observatory recorded a minimum of 
 27*446 inches at 1 p.m., while at 6 p.m. the barometer read only 
 27'41 inches at Barrow-in-Furness. 
 
 On the other hand, at 6 p.m. of January 22, 1907, the baro- 
 meter read 31-58 inches at Riga, and at 8 a.m. of the following day 
 the isobar of 31 inches stretched westward to Ulster, the baro- 
 meter reading 31*01 inches at Donaghadee. Also, on January 31, 
 1902, the barometer rose to 31-118 inches at Aberdeen ; and on 
 January 28, 1905, atmospheric pressure reached, or slightly 
 exceeded, 31 inches all over the southern half of Ireland. Roche's 
 
 120 
 
THE BAROMETER 121 
 
 Point, Cork Harbour, reported 31 '03 inches both morning and 
 evening of the 28th, and at 6 p.m. the reading was 31-06 inches 
 at St. Mary's, Scilly Isles. On that same day at 9 p.m. the 
 reading of 31-007 inches was recorded in Trinity College, and 
 that of 30*999 inches at Fitzwilliam Square, Dublin. But these 
 values by no means represent the extreme range of the barometer. 
 On January 26, 1884, the barometer fell to 27-332 inches at Ochter- 
 tyre, near Crieff, in Perthshire, and on February 5, 1870, the 
 reading of 27*33 inches was recorded on board the Cunard steamer 
 Tarifa in the North Atlantic, in lat. 51 N. and long. 24 W. 
 But even these extremes, all reduced to 32 and mean sea-level, 
 have been exceeded. In a communication to Nature, dated 
 January 6, 1887 (vol. xxxv., p. 344), Mr. Blanford states that 
 " the cyclone which on the morning of September 22, 1885, swept 
 over False Point, on th'e coast of Orissa, gave the lower readings 
 27-135 inches at the beginning of the central calm, and 27*154 
 inches half an hour later (both readings reduced to 32 and sea- 
 level)." These readings were made by a verified standard 
 barometer, and are thoroughly authentic. For comparison with 
 English standards a further subtractive correction of 'Oil inch 
 has to be applied, which would make the lowest reading 27'124 
 inches. 
 
 In an interesting paper 1 on " The Storm and Low Barometer 
 of December 8 and 9, 1886," Mr. Charles Harding, F.R.Met.Soc., 
 quotes from Professor Loomis's Contributions to Meteorology, 
 chap, ii., a reading of 31*72 inches, reduced to sea-level, ob- 
 served at Semipalatinsk, in Western Siberia (lat. 50 24' N., 
 long. 80 13' E.) on December 16, 1877, the reading at Barnaul 
 being 31*63 inches at the same time. This gives a difference from 
 Mr. Blanford's reading, 27*12 inches, of +4*6 inches, which is 
 probably the maximal range of the barometer ever observed at 
 the earth's surface (reduced to sea-level). 
 
 These extreme readings, although rarely observed, should 
 be provided for in the scaling of a barometer, which should 
 range from 32 inches to 26 inches, or less to allow for 
 altitude. 
 
 1 Quarterly Journal of the Royal Metio ological Society, 1887, vol. xiii., 
 p. 201. 
 
122 METEOROLOGY 
 
 The space above the mercury in the barometer tube is still 
 called the " Torricellian vacuum," and, provided the tube has 
 been properly filled, this space should contain nothing except a 
 little of the vapour of mercury. 
 
 As mercury expands with heat, it is essential that a ther- 
 mometer should be attached to every barometer, in order to 
 show the temperature of the mercurial column. Once this value 
 is ascertained, a suitable correction must be applied to the reading 
 of the barometer, so as to reduce it to the fixed or standard 
 temperature of 32 F. 
 
 The mercurial barometer is best mounted in a brass case, 
 because the coefficient of expansion of brass by heat is well known 
 a matter of great importance, as the tables for correcting 
 readings for temperature are based upon the coefficients of ex- 
 pansion of mercury, glass, and brass. Barometers mounted in 
 wood are of inferior value for scientific purposes. 
 
 The attentive reader will at this point suggest to his own mind 
 two difficulties in the construction of the barometer. One is, 
 how to cover the cistern so as to prevent the escape of the mer- 
 cury, and so render the instrument portable without interfering 
 with atmospheric pressure on the surface of the mercury contained 
 in the cistern. This is effected by constructing the bottom of the 
 cistern of leather, as in the Fortin barometer ; or by covering a 
 small cavity in the roof of the cistern with leather, as in the Kew 
 barometer. 
 
 Again, it is evident that the level of the mercury in the cistern 
 will change according as the barometer rises or falls. If it rises, 
 there will be more mercury in the tube and less in the cistern, and 
 the level of the mercury in the latter will fall. On the other hand, 
 if the barometer falls, there will be less mercury in the tube and 
 more in the cistern, in which the level of the mercury will in 
 consequence rise. In a word, " the zero of the scale does not 
 always correspond with the level of the mercury in the cistern " 
 (Fred. J. Brockway). As in all cases the height of the barometer 
 is calculated from the level of the mercury in the cistern, we must 
 apply a correction for the error arising from the change of level 
 in the cistern the " error of capacity/' as it is called. Formerly 
 tables for applying a "capacity correction" were employed, 
 
THE BAROMETER 
 
 123 
 
 but they are not now required, 
 owing to the adoption of barometers 
 of the Fortin, Kew, or Siphon 
 patterns. 
 
 (1) The Fortin barometer has a 
 
 pliable or flexible base to 
 its cistern. 
 
 (2) The Kew barometer has a 
 
 contracted scale. 
 
 (3) The Siphon barometer dis- 
 
 penses with the use of a 
 cistern altogether. 
 
 1. In the Standard Barometer 
 (German, Hauptbarometer ; French, 
 Barometre ctalori), commonly called 
 Fortin's barometer (Fig. 23), the 
 starting-point, or zero, of the scale 
 is formed by an ivory pin, which 
 must be brought into exact contact 
 with the surface of the mercury in 
 the cistern whenever an observa- 
 tion is made. This is effected by 
 fixing a screw below and in contact 
 with the flexible leather bottom of 
 the cistern. The adjustment is 
 made by means of a thumb-screw, 
 which raises or depresses the 
 flexible leathern base of the cistern 
 until the tip of the ivory pin 
 technically called the fiducial point 
 and its image reflected in the 
 mercury in the cistern appear ex- 
 actly to touch each other, when 
 viewed through a glazed aperture 
 in the wall of the cistern. 
 
 In Fig. 24 an ingenious arrange- 
 ment, devised by Mr. Wallis, for 
 facilitating the adjustment of the 
 
 FIG. 
 
 -PoRTiN BAROMETER. 
 
124 
 
 METEOKOLOGY 
 
 barometer scale, is represented. It can be clamped to any of 
 the barometers constructed on the Fortin principle. 
 
 2. In 1853 Mr. P. Adie, of Edin- 
 burgh, invented a barometer for use 
 at sea, which is commonly 
 known as the Kew baro- 
 meter (Fig. 25). It is so 
 called because it was re- 
 commended by the Kew 
 Committee of the British 
 Association for adoption by 
 the Government as best 
 suited for marine observa- 
 tions then about to be com- 
 menced by the Admiralty 
 and the Board of Trade. 
 
 FIG. 24. WALLIS'S ARRANGEMENT T , ,. .. .. , 
 
 FOR ADJUSTING THE IVORY Its distinctive features are 
 a brass frame, a contracted 
 
 tube, having a pipette, a closed cistern, and a scale 
 of contracted inches. In this, the " Marine Baro- 
 meter," the tube is of small calibre throughout the 
 greater part of its length in order to lessen the oscil- 
 lations of the mercury caused by the ship's motion, 
 which are technically known as " pumping." This 
 renders the instrument rather sluggish, but not 
 materially so. The cistern is entirely composed of 
 iron (because brass, being an alloy, is liable to be 
 acted on by mercury), and only a small aperture in 
 its roof is left through which atmospheric pressure is 
 able to exert itself on the contained mercury. This 
 aperture is, as has been said above, covered with 
 leather to prevent the escape of the mercury. KE^'BARO- 
 
 In this instrument the " error of capacity " is com- METER. 
 pensated by contracting the divisions on the scale from above 
 downwards, in proportion to the relative sizes of the tube and 
 the cistern. In ordinary Kew barometers the diameter of the 
 tube is about '25 inch, and that of the cistern about T25 inch. 
 Accordingly, starting from 32 inches correctly marked off from a 
 
THE BAROMETER 
 
 125 
 
 definite point below, the " inches " of the scale are shortened in 
 the proportion of '04 inch for every true inch. 
 
 Every tube is fitted with an " air-trap," which is a small 
 funnel or pipette inserted somewhere between the range of the 
 column and the neck of the cistern. The pipette 
 was first proposed by Gay-Lussac in order to stop 
 the ascent into the Torricellian vacuum of any air 
 or moisture which may work its way from the 
 cistern into the tube between the glass and the 
 mercury (Fig. 26). 
 
 The so-called " Gun Barometer " was designed 
 by Admiral Robert FitzRoy, in 1861, for the naval 
 service. It is a modification of the 
 marine barometer, and is intended to 
 withstand the concussion of heavy 
 ordnance. The glass tube is surrounded 
 wherever possible with vulcanised india- 
 rubber tubing as packing. This checks 
 the vibration from firing, but does not 
 hold the tube too rigidly. 
 
 3. To Gay-Lussac we owe the Siphon 
 barometer, which consists of a bent 
 glass tube of uniform calibre, but with 
 one branch or leg much longer than the 
 other. The longer limb is closed at the 
 
 cury, the shorter limb is quite open, and serves as 
 
 a cistern. As the mercury falls in the long limb, 
 
 leaving the Torricellian vacuum above it, it must 
 
 rise to an equivalent height in the short limb. 
 
 The motion in each limb is exactly one-half of 
 
 what takes place in a Fortin barometer. The 
 
 atmospheric pressure, or " height of the barometer/' 
 
 is the difference between the two levels, so that two BAK01 
 
 readings must on every occasion be taken one, of the level of the 
 
 mercury in one limb ; the other, of the level in the other limb. 
 
 This instrument is the only mercurial barometer suitable for 
 
 mountain climbing, owing to its lightness and portability (Fig. 27). 
 
 FIG. 27. 
 SIPHON 
 
126 METEOROLOGY 
 
 The ordinary wheel barometer, or " weather glass," was in- 
 vented in 1665-66 by Robert Hooke, Secretary of the Royal 
 Society. It is a siphon barometer. Resting on the mercury 
 in the shorter limb is a float connected by a silken cord with a 
 light counterpoise at the other side of a fixed pulley, round which 
 the cord is coiled two or three times. A needle indicator attached 
 to the axis of this pulley rotates with the rise and fall of the 
 mercury round a graduated circular dial, on which are also the 
 words : " set fair," " fair," " change," " rain," " much rain," 
 and " stormy." These words are intended to indicate what 
 weather may be expected when the needle points to each part of 
 the dial. 
 
 Although a popular instrument, the wheel barometer is of 
 no scientific value. Its principle, however, has been applied in 
 the construction of one form of self-registering barometer, or 
 barograph (Greek, f$<ipos, iveight ; y/>a</>w, I write). A pencil is 
 attached to the cord connected with the float, and this pencil 
 is so arranged that it draws a continuous tracing on ruled paper, 
 which is moved by clockwork. 
 
 In a modified barograph of this kind ruled metallic paper spread 
 on a revolving vertical drum of about 4 inches diameter is 
 pierced at given intervals (usually every hour) by a needle shot 
 out by clockwork, and ingeniously connected by means of a 
 pulley with the mercurial surface in the short limb of a siphon 
 barometer. The drum or cylinder in this barograph is made to 
 revolve once a week by means of clockwork. 
 
 One of the costliest barographs in existence was designed in 
 1853 by the late Mr. Alfred King, C.E., of Liverpool (Fig. 28). 
 About 130 pounds of mercury are employed in the construction 
 of this instrument, and the effects of friction, which are the great 
 drawback in wheel barometers, are entirely overcome by the most 
 sensitive mechanical arrangements. In this ingenious barograph 
 the tube is partly supported by the mercury in the fixed cistern, 
 which, as it rises and falls, raises and depresses the tube. A 
 delicate mechanical contrivance records this change of level con- 
 tinuously on a revolving drum. The barometric column is made 
 to show nearly 6 inches for each inch of the ordinary barometer. 
 This instrument has for many years been in use at the Bidston 
 
FIG. 28. ALFRED KING'S BAROGRAPH. 
 
128 METEOROLOGY 
 
 Observatory, near Liverpool. It is fully described in the 
 late Mr. Hartnup's " Report to the Mersey Docks Board for 
 1865." 
 
 Dr. R. H. Scott, F.R.S., speaking of barographs, 1 observes : 
 " The principle of registration generally adopted in this country 
 for the better class of barographs is photographic, not mechanical." 
 
 The late Sir Francis Ronalds in 1847 designed a photographic 
 barograph which, in a modified form, is employed in the First 
 Order Stations of the Meteorological Office, London. The prin- 
 ciple of the instrument is described at length in the " Report of 
 the Meteorological Committee of the Royal Society for 1867 " 
 (pp. 40-42). " The barometer is of the ordinary pattern, and the 
 light is admitted through the Torricellian vacuum, so that the 
 actual height of the mercury itself is photographed without the 
 intervention of any mechanical contrivance " (R. H. Scott). 
 
 Those who are interested in this subject of barographs will find 
 at p. 412 of the second volume of the Quarterly Journal of the 
 Meteorological Society a description of a new barograph invented 
 by M. Louis Redier, which was communicated to the Society on 
 March 17, 1875, by the late Mr. G. J. Symons, F.R.S. The 
 apparatus is so arranged that all the work is done by a powerful 
 clock-movement, and the barometer, of the siphon type, has only 
 to direct the action of the clockwork. 
 
 In 1886 M. Redier, in a pamphlet, described a later form 
 of his mechanical barograph under the title " Nouveau baro- 
 metre enregistreur a mercure." In it the barometer is at rest. 
 A differential clock train keeps a light horizontal arm in con- 
 tinuous slight vertical oscillation close to the point of a stalk 
 rising from the mercury in the lower branch of a siphon tube. As 
 the arm follows the stalk in all its variations of position, the 
 barometric variations, through a pencil, become continuously 
 recorded. 
 
 In addition to mechanical contrivances and photography, elec- 
 tricity has been employed in the construction of the barograph. 
 Sir Charles Wheatstone, in the British Association Report for 
 1842, suggested the adaptation of electricity for the purpose. He 
 proposed that a platinum wire, controlled by a clock, should make 
 1 Elementary Meteorology, p. 77. 
 
THE BAROMETER 
 
 129 
 
 contact at given intervals with the mercury in the tube of a 
 barometer or other instrument for example, the dry and wet 
 bulb thermometers so creating an electric current which should 
 determine both the record and the value of the element (W.Ellis). 
 This principle has been since applied in the barographs included 
 in the combined meteorographs of Salleron (1860), Theorell 
 (1869), and Van Rysselberghe (1873), the records being all inter- 
 mittent. 
 
 The Aneroid Barograph. The motion in most barographs is 
 supplied by a set of aneroid boxes (see p. 132 for a description of 
 
 FIG. 29. THE ANEROID BAROGRAPH. 
 
 the " aneroid barometer ") (Fig. 29). The instruments used at 
 official stations of the Meteorological Office are of two sizes. The 
 parts of the apparatus which affect the size and ruling of the charts 
 are (1) the length of the arm carrying the pen, measured from 
 pivot to pen-point ; (2) the height of the pivot of the pen-arm 
 above the flange on the drum, against which the chart rests ; 
 (3) the magnification of the pen motion i.e., the vertical distance 
 on the chart corresponding with a change of pressure of 1 inch 
 of mercury. The last depends on the number and size of the 
 aneroid boxes and on the arrangement of the levers. The hori- 
 zontal distance on the Meteorological Office charts corresponding 
 
 9 
 
130 METEOKOLOGY 
 
 with an iDtcrval of twelve hours is also given. These dimensions 
 are as follow : 
 
 
 Large Instrument. 
 
 Small Instrument. 
 
 
 Inches. 
 
 Inches. 
 
 Length of pen -arm 
 
 10-24 
 
 7-29 
 
 Height of pivot above flange . . 
 Magnification of scale 
 
 3-14 
 2-04 
 
 1-57 
 1-00 
 
 Twelve hours on time scale . . 
 
 1-05 
 
 0-78 
 
 The diameters of the drums are respectively 4'97 inches and 
 3*65 inches. The scale usually reads from 28 to 31 inches. Both 
 these limits have, however, been exceeded. 
 
 Barographs in which the motion of the pen is furnished by a 
 set of aneroid boxes are subject to changes of zero. When abso- 
 lute pressure values are required, their use must accordingly be 
 confined to interpolating between the readings of standard mer- 
 curial barometers. A barograph needs no special exposure, but 
 it should be protected from shaking and from sudden changes of 
 temperature. 
 
 Transmission of Barometric Indications by Electricity. In 1882, 
 Mr. John Joly (now Professor Joly, F.R.S.), of Dublin, described 
 in Nature (vol. xxv., p. 559) a plan for ascertaining the reading of 
 a distant mercurial barometer, connected with the recording 
 station telegraphically. He carries two wires through the head 
 of the barometer tube. One of these of a given diameter is con- 
 tinued downward into the mercury to a point below which the 
 mercury never falls. The continuation of the other is a fine 
 carbon thread, also of a given diameter, carried to the same point, 
 and there joined to the wire. The outer ends of the wires pass 
 to the recording station, an electric current sent from which 
 traverses both wire and carbon in its passage. The carbon being 
 a substance of high resistance, a very small change in its effective 
 length due to the rise or fall of the mercury in the barometer tube, 
 exposing less or more of the fine carbon thread, will tell on the 
 potential of the returning current. This variation of potential 
 would, in Mr. Joly's opinion, be sufficiently marked to enable an 
 observer at the recording station to measure the barometric 
 
THE BAROMETER 
 
 131 
 
 LON 
 
 W- 
 
 D c o!S s 
 
 '? 
 ET 
 
 
 H 
 
 SL 
 
 30 
 
 
 
 
 
 
 
 - 
 
 
 
 
 
 
 
 
 IfiF 
 
 !'<* 
 
 
 
 8- 
 
 
 7- 
 
 
 - 
 
 
 5- 
 
 
 4- 
 
 
 f- 
 
 
 variations at a station four miles distant, involving eight miles 
 of wire. 
 
 Bartrum's Open-Scale Barometer. This instrument exhibits 
 changes of barometer pressure with great facility 
 and accuracy, and is exceedingly rapid in 
 responding to such changes. 
 
 The lower part of this instrument, of which 
 the sole maker is Mr. James J. Hicks, of Hatton 
 Garden, London, E.G., is formed in the same way 
 as an ordinary mercury barometer. The tube 
 in the neighbourhood of the upper surface of the 
 mercury column is enlarged, and above the surface 
 is again reduced in calibre, and continued up- 
 wards for 27 inches or more. The space above 
 the mercury and for some distance up the narrow 
 tube contains a light red fluid, the position of the 
 upper surface of which, along an attached scale, 
 gives the barometer reading. 
 
 The principle is as follows : On account of the 
 change in calibre of the tube, a rise of the mercury 
 in the enlarged part, which we will call the bulb, 
 will cause a very much greater rise of the light 
 fluid in the upper tube. The change of atmo- 
 spheric pressure causing the rise is represented by 
 the fall of mercury in the cistern, added to the rise 
 of mercury in the bulb, again added to the in- 
 crease in length of the column of light liquid (the 
 last reduced to its equivalent length of mercury). 
 
 As an example, suppose the cistern and bulb 
 to be of the same calibre and each to have a 
 sectional area fifty times as great as that of the 
 upper tube, and suppose the mercury in the 
 cistern to fall O'l inch. The mercury in the bulb 
 will then rise O'l inch and the liquid in the 
 upper tube 5 inches, a change in length of 
 the upper column of 4 '9 inches. The liquid used 
 
 has a density about one-twelfth that of mercury, so that a column 
 of 4'9 inches would correspond to (Ml inches of mercury. Five 
 
 92 
 
132 METEOROLOGY 
 
 inches on the scale will therefore correspond to 0'1 + 0'1 + 0'41 = 
 0*61 inches of mercury, or 8*2 inches of scale to each inch of 
 reading. The range of the scale can be made as open in this 
 barometer, which is 5 feet long, as in a glycerine barometer, which 
 is over six times as long viz., 32 feet. 
 
 The instrument is mounted in a very neat and strong mahogany 
 frame, with glass door. By a momentary adjustment it becomes 
 portable, and can be carried with ease and safety. The effect 
 of change of temperature is inappreciable. Owing to the ex- 
 tremely open range of this barometer (over 8 inches to an inch 
 of mercury), no vernier is required, and a reading can easily be 
 taken to T<yV(>th of an inch. 
 
 Before explaining the method of reading the barometer, it 
 may be well to describe some substitutes for mercurial barometers 
 which have been devised. 
 
 The Aneroid Barometer is the chief of 
 these. It was invented by M. Vidi, of 
 Paris, in 1843, and patented in the 
 following year. The French patent is dated 
 April 19, 1844. As the name implies, it 
 contains no fluid (Greek, a, priv. ; vrjpos, wet 
 or damp, hence liquid or fluid ; and etSos, form 
 FIG. si. EXTRA-SENSITIVE or shape). For this reason the aneroid is also 
 
 ANEROID BAROMETER. 
 
 known as the holostenc barometer, the 
 word " holosteric " meaning " entirely solid " (Greek, 6'Aos, whole ; 
 o-re/oeos, solid). In this ingenious instrument (Fig. 31) the pressure 
 of the atmosphere is measured by its effect in altering the shape 
 of a small, hermetically sealed, partially exhausted metallic box 
 called the " vacuum chamber." This vacuum chamber is com- 
 posed of two discs of corrugated German silver soldered together. 
 Its sides are made in concentric rings, so as to increase their elas- 
 ticity, and one of them is fastened to the back of the brass case 
 which contains the whole mechanism. A strong spring also 
 fixed to the case is so arranged as to act in opposition to the motion 
 of the vacuum chamber, preventing its sides collapsing when the 
 effect of reduced atmospheric pressure is added to that of ex- 
 treme exhaustion of the chamber. A lever, composed of iron and 
 brass, so as to compensate for changes in temperature, connects 
 
THE BAROMETER 133 
 
 the spring, by means of a bent lever at its further end, with a 
 watch-chain, which is wound about an " arbor " (axle or spindle). 
 An index-hand or pointer, fixed to the arbor in front, is by its 
 revolution caused to rotate backwards and forwards over an 
 experimentally graduated dial, and so is made to mark the 
 variations in atmospheric pressure from time to time. A very 
 slight alteration of the size of the vacuum chamber produces a 
 large deviation of the index-hand, ^4<j^h ^ an mcn > causing it to 
 move through 3 inches as marked on the dial. 
 
 When pressure increases, the falling in of the corrugated sides 
 of the vacuum chamber will pull upon the lever, which in turn, 
 acting through the second or bent lever, will pull upon the 
 chain, drawing it off the arbor, and so causing the pointer to 
 move across the dial towards the right, marking high pressures. 
 
 When pressure decreases, the expansion of the vacuum 
 chamber will allow the compensating spring to push away the 
 lever, which will relax the chain, allowing it to be wound round 
 the arbor by a spiral hair-spring, which will move the arbor and 
 pointer towards the left, marking low pressure. 
 
 In 1851 Rusk added an altitude scale to the aneroid barometer. 
 A Metallic Barometer, designed by M. Bourdon in 1851, is a 
 modification of the principle of the aneroid. 
 
 This instrument is described in Besant's Elementary Hydro- 
 statics (chap v., "Notes,"), and in the Report of the Jury on 
 Philosophical Instruments in the Great Exhibition of 1851, as 
 well as in a lecture by Mr. James Glaisher, F.R.S., on these instru- 
 ments. It consists of a thin elastic metal tube of elliptic section, 
 in shape a portion of a circle, closed at its ends and exhausted of 
 air. Alterations in the pressure of the atmosphere are indicated 
 by the ends of the tube approaching towards each other when 
 pressure increases, and receding from each other when pressure 
 diminishes. These motions are communicated by gearing work 
 to an index-hand traversing a dial plate. No definite explana- 
 tion of the principle of action of this instrument was offered until 
 the Rev. E. Hill, M.A., Fellow of St. John's College, Cambridge, 
 communicated a paper on the subject to the Meteorological 
 Society on February 21, 1872. The above particulars are taken 
 partly from Mr. Ellis's address, but chiefly from Mr. Hill's com. 
 
134 
 
 METEOKOLOGY 
 
 munication. His explanation, however, is of too recondite a 
 nature to be reproduced here. 
 
 Aneroids, while very sensitive, are apt to get out of order 
 owing to defects in construction, rusting, or loss of elasticity in 
 the springs. They are, therefore, not used at Second Order 
 Stations or for concerted observations, for which accurate mer- 
 curial barometers are indispensable. 
 
 If an aneroid is employed, its readings should be frequently 
 compared with those of a reliable mercurial barometer reduced 
 to 32 F. It is a popular instrument because of its convenient 
 size and portability. Besides, it requires no correction for its 
 own temperature, for it must be remembered that the " aneroid 
 readings correspond to readings of the mercurial barometer 
 reduced to 32 " (W. Marriott). 
 
 FIG. 32. THE PIESMIC BAROMETER. 
 
 The Piesmic Barometer. This ingenious instrument, invented 
 by Mr. A. S. Davis, M.A., is based on the principle that air is 
 
THE BAROMETER 
 
 135 
 
 more compressible when 
 the barometer is low than 
 when it is high. In the 
 Piesmic Barometer (Greek, 
 Triefw, I squeeze or press) 
 a tubeful of air is taken 
 at atmospheric pressure, 
 and its compressibility is 
 tested by allowing mer- 
 cury to run down the 
 tube and compress the air 
 inside. The depth to which 
 the mercury descends 
 varies with the compressi- 
 bility of the enclosed air, 
 and therefore also with 
 the barometric pressure 
 at the time. The reading 
 of the scale gives the 
 atmospheric pressure in 
 inches of mercury. The 
 instrument is illustrated in 
 the accompanying figure 
 (Fig. 32). By employing 
 an air-tight case, and 
 drying the air before 
 finally closing the cistern, 
 any error arising from the 
 humidity of the air is 
 entirely avoided. 
 
 A self - recording mer- 
 curial barometer (Fig. 33) 
 has been designed by 
 W. H. Dines, Esq., F.R.S., 
 to give a trace from which 
 the height at any time 
 may be determined J to 
 005 inch. This end js 
 
 FIG. 33. DINES'S SELF-RECORDING MERCURIAL 
 BAROMETER. 
 
136 
 
 METEOROLOGY 
 
 FIG. 34. FIELD'S ENGINEERING 
 ANEROID BAROMETER. 
 
 attained by arranging the details of construction so that the 
 friction of all the moving parts, and more particularly that 
 between the pen and the paper, may be very small, and also by 
 
 an automatic temperature correction. 
 The pen is actuated by a float in the 
 lower cistern, the motion being multi- 
 plied by a lever so that a length of 
 1J inches on the paper may corre- 
 spond to a change of 1 inch in the 
 height of the barometer. The float 
 is in the form of a hollow cylinder 
 sealed at the top, and floating mouth 
 downwards in the mercury. A rise 
 of temperature lowers the level of 
 the mercury in the lower cistern, but at the same 
 
 time it expands the air in the float, and makes it 
 swim higher in the mercury. The volume of air is 
 so adjusted that there may be a complete compen- 
 sation. There is an additional pen fixed to the 
 frame, which draws a line of reference on each 
 sheet of paper while it is on the clock drum, and 
 for accurate measurement this line is taken as the 
 zero line, since by this means the error that might 
 be caused by placing the chart unequally on the 
 drum, or by an incorrect printing of the charts, 
 is avoided. The price, complete in glass case, with 
 lock and key, including supply of charts and ink, 
 is 30. 
 
 Measurement of Altitudes. An aneroid in good 
 order will show with precision the difference in 
 height between the various stories in a lofty house, 
 the varying gradients in travelling on a railway, 
 and mountain or balloon elevations it may be up Flo 35 _ ADIE . S 
 to 24,000 feet. One of the chief uses of the aneroid, SYMPIESOMETER. 
 indeed, is the measurement of altitudes. Owing to the elasticity of 
 the atmosphere, the reduction of pressure does not proceed evenly 
 with altitude, and accordingly special altitude scales have been 
 computed, which are engraved on the dial of the instrument. 
 
THE BAROMETER 
 
 137 
 
 A correction for the temperature of the air (not for that of the 
 instrument, for the reason given above) must always be made, 
 and so in the " Engineering Aneroid," invented by Mr. Rogers 
 Field, B.A., M.Inst.C.E., F.R.Met.Soc., and manufactured ex- 
 clusively by Mr. L. Casella, this correction is taken into account 
 by making the scale adjustable for temperature (Fig. 34). 
 
 While on this subject, it will be well briefly to describe two 
 other instruments which are used for ascertaining 
 the altitude one a barometer, the other a 
 thermometer the sympiesometer and the hypso- 
 meter respectively. 
 
 The sympiesometer, or " compression measure " 
 (Greek, (n>//7nrts, compression ; from o-iyzTmfw, to 
 press or squeeze together ; //eiy>ov, a measure), was 
 invented by Mr. P. Adie, of Edinburgh (Fig. 35). 
 It is a sensitive but unreliable kind of barometer 
 (using this term in its strict etymological sense), 
 consisting of a glass tube 18 inches in length 
 and | of an inch in diameter which terminates in 
 a closed bulb above, and, after a sharp bend, in 
 an open cistern below. The pressure of the air , 
 
 FIG. 36. CASELLA'S 
 HYPSOMETKR. 
 
 FIG. 37. PORTABLE LEATHER CASE FOR HOLDING CASELLA s 
 HYPSOMETKR. 
 
 acting through the latter on the surface of a fluid, such as oil or 
 glycerine, in the lower part of the cistern and of the tube, forces it 
 upwards so as to compress an elastic gas, such as hydrogen or 
 air, in the upper part of the tube and in the bulb. The amount 
 of compression is read off on an adjustable scale, the index of 
 which must be set to the division on the scale corresponding to 
 the temperature indicated by an attached thermometer. 
 
138 METEOROLOGY 
 
 The principle of the hypsometer (Greek, ityo?, height ; 
 a measure) is based on the fact, already referred to in Chapter VII., 
 p. 79, that the boiling-point of water falls according as atmo- 
 spheric pressure is reduced. The instrument consists of a vessel 
 for water, with a spirit-lamp for heating it, and an enclosed ther- 
 mometer for showing the temperature of ebullition. In Casella's 
 hypsometer (Figs. 36 and 37), a strong, small-bulbed thermometer, 
 divided and figured on the stem, is sheltered from cold when in 
 use by a double telescopic chamber, into which it is introduced 
 to any required depth through a loose piece of indiarubber at 
 the top. When the water boils, the vapour fills the inner chamber 
 and envelops the thermometer, bulb and stem alike, finally 
 descending in the outer chamber and escaping by a pipe outlet. 
 Mr. Casella has constructed a smaller instrument on the same 
 principle, which is much used by Alpine travellers. 
 
 A New Form of Open-Scale Barometer. Professor W. F. 
 Barrett, F.R.S., has recently devised a simple form of open- 
 scale barometer, or weather-glass, to which a brief reference 
 may be made. It resembles an air-thermometer, but, the 
 bulb and tube being rendered practically impermeable to 
 changes of temperature, the movement of the liquid index indi- 
 cates changes of atmospheric pressure. This is accomplished by 
 using a Dewar liquid air flask for the bulb, and surrounding the 
 index- tube with a wider sealed glass tube from which the air has 
 been thoroughly exhausted. The Dewar liquid air-flask, as is 
 well known, consists of a double or jacketed glass bulb, the space 
 between the two envelopes being highly exhausted in a reflecting 
 film of silver or mercury deposited on the inner surface of the 
 outer envelope. By this means the transmission of heat from 
 the outside is almost wholly prevented. In practice this form 
 of weather-glass has been found to be both very sensitive and 
 reliable. This ingenious instrument was shown, and its action 
 demonstrated, at a scientific meeting of the Royal Dublin Society 
 on February 22, 1910. 
 
CHAPTER XII 
 BAROMETRICAL READINGS 
 
 SINCE mercury expands by heat and contracts by cold, it is 
 necessary that every barometer should carry a thermometer 
 closely attached to its metal case, or preferably to its glass tube 
 and cistern. By means of this ' : attached thermometer " the 
 observer is placed in a position to apply a proper correction to the 
 reading of the barometer, with the view of reducing that reading 
 to the fixed standard of temperature, or 32 F. 
 
 Before any observation is made, the barometer should be 
 mounted in a room with an equable temperature, not near a 
 fireplace or a stove. Its scale should be on a level with the 
 observer's eye 5 feet or 5 feet 6 inches above the floor. It 
 must hang vertically : " Care should be taken that no readings 
 from a barometer which is not hanging truly vertically should 
 ever be recorded." 1 " To facilitate readings, a piece of white 
 paper or of opal glass should be fixed immediately behind the 
 part of the tube at which the readings are taken ; and if the 
 barometer be of the Fortin pattern another piece should be placed 
 behind the cistern." 2 This arrangement may be seen well 
 represented in Fig. 23 above (p. 123). 
 
 The method of taking a barometrical observation is as follows : 
 
 1. The attached thermometer should first be read, no matter 
 what kind of barometer is employed. The temperature of the 
 external air (dry-bulb reading) should also be taken. 
 
 2. Next, in the case of a Fortin or standard barometer, the 
 mercury in the cistern should be adjusted by turning the screw 
 
 1 Instructions in the Use of Meteorological Instruments, compiled by 
 Robert H. Scott, M.A., F.R.S. Reprinted 1885. 
 
 2 Hints to Meteorological Observers, sixth edition, p. 6. By William 
 Marriott, F.R.Met.Soc. 1906. 
 
 139 
 
140 METEOROLOGY 
 
 at the bottom, so that the tip of the ivory pin, or the fiducial 
 point, should barely touch the surface of the mercury. This 
 manipulation is not required, nor is it possible, in the case of the 
 Kew barometer, in which (as has been explained) the scale of 
 shortened inches compensates for the error due to capacity. 
 
 3. The barometer tube should, after this, be gently tapped 
 to overcome any tendency to adhesion between the mercury 
 and the glass, and to allow capillary action to assert itself. 
 
 It is necessary to mention that, in obedience to what is called 
 " capillary action/' a liquid like water, capable of wetting a clean 
 glass tube open at both ends, will rise in such a tube above the 
 level of its surface in the vessel containing it, and higher and 
 higher according to the fineness of the bore of the tube. Hence 
 the term " capillary," from the Latin " capilla," a fine hair. 
 Further, the liquid will stand above the general level in the tube 
 where it approaches the sides, so that its upper surface in the 
 tube will be curved and concave, owing to capillary attraction. 
 On the other hand, a liquid like mercury, incapable of wetting 
 such a tube, will stand in the tube below the level of its surface 
 in the vessel, and where it approaches the sides of the tube, its 
 level will be below its general level in the tube, so that its upper 
 surface will be curved and convex, owing to capillary repulsion. 
 This causes the mercury in a barometer always to stand a little 
 lower than the height due to atmospheric pressure, and necessi- 
 tates a correction for capillarity. Such a correction is less in a 
 barometer in which the mercury has been boiled in the tube 
 than in an unboiled tube, for by the boiling a film of air, which in 
 unboiled tubes adheres to the glass, is expelled. The error is 
 also reduced by widening the bore of the tube ; for example, 
 the depression in a boiled tube of inch in diameter is *02 inch, 
 whereas in a similar tube of J inch diameter it is only '003 inch. 
 
 4. It follows from the foregoing that, in reading a barometer, 
 the height should be taken from the very apex of the convexity, 
 or of the meniscus, as it is called (Greek, /irpiV/cos, a crescent; 
 from fj-rjvrj, the moon). This is done by means of a small movable 
 scale called the vernier, to which a sliding piece at the back of 
 the instrument is connected so as to move with it. To take a 
 reading, the lower edge of the vernier and the lower edge of the 
 
BAROMETRICAL READINGS 141 
 
 sliding piece behind should be brought, by turning the mill-head 
 pinion which moves the rack up or down, to form a tangent with 
 the convex surface of the mercury. As Mr. Marriott well re- 
 marks : " The front and back edges of the vernier, the top of the 
 mercury, and the eye of the observer must be in the same straight 
 line." 
 
 The object of the sliding piece at the back of the instrument 
 is to insure that the observer's eye is at the same level as the 
 domed top of the mercury column. Whenever the index and the 
 scale on which it is read are not in the same plane, serious errors 
 are made, which are known as errors of parallax. 
 
 But what is the vernier ? It is a short scale named after its 
 inventor, Pierre Vernier, made to slide by means of a rack and 
 pinion along the divisions of a graduated scale, such as that of a 
 barometer, and its divisions are so contrived as to be slightly 
 shorter than those of the barometer scale, which is generally 
 divided into inches, tenths, and half-tenths, or five-hundredths 
 (05) of an inch. The vernier is made equal in length to twenty- 
 four half-tenth divisions of the barometer scale, and is then itself 
 divided into twenty-five equal parts. From this it follows that 
 each space on the vernier scale falls short of a space on the barom- 
 eter scale by the twenty-fifth part of *05 inch, or 
 T$ * X *V = v-fis* = ? U = '002 inch. 
 
 Each division of the vernier, therefore, represents a difference 
 of *002 inch, or -.- J^- inch in pressure, while by interpolating 
 a reading between any two divisions of the vernier, we are enabled 
 to read the pressure to *001 inch, or T ^Vo" i ncn ' 
 
 In using the vernier, the division on the barometer scale at or 
 below which the lower edge of the vernier stands after setting it 
 should first be read off. In the accompanying figures (38 and 39) 
 two cases are illustrated. In one (that to the left), the lowest line 
 on the vernier scale exactly coincides with the division 29 '50 on the 
 barometer scale. The reading is therefore 29-500 inches precisely. 
 In the other (that to the right), the reading on the barometer 
 scale gives us 29 '65, but the height of the column of mercury is 
 in reality that amount in inches plus the vernier indication. 
 
 On looking up the vernier scale in this case, we find that both 
 the second and the third divisions above the figure 3 apparently 
 
142 
 
 METEOROLOGY 
 
 coincide with a division on the barometer scale. It is therefore 
 necessary to take a mean reading, and to add the decimal '035 to 
 the first reading, thus : 
 
 29-65 +-035 = 29-685 inches. 
 
 30 
 
 A 
 
 c_ 
 
 1 
 
 n 
 
 ~ NECRETTI&ZAMBRA. LONDON. ^ ^J) 
 
 Q ! co i : : : '.CM 
 
 20 
 
 15 
 1 O 
 05 
 00 
 
 - 
 
 
 
 ~ 
 
 
 
 
 2 
 
 1 
 
 l 
 
 \ 
 
 I 
 
 D 
 
 
 
 
 f 
 
 
 __ 
 
 - 
 
 
 
 - 
 
 c 
 
 
 Case 1. Case 2. 
 
 FIG. 38. FIG. 39. 
 
 METHOD OF READING THE VERNIER. 
 
 In cases where it is hard to say which division of the barometer 
 scale is that below the lower edge of the vernier, the reading of the 
 latter will itself point out which division on the barometer scale 
 should be taken. For example, in the left-hand drawing the 
 
BAROMETRICAL READINGS 143 
 
 correct reading is manifestly 29*500 inches, not 29*50 -f*050 = 
 29 '550 inches. In fact, the value of the vernier reading in the 
 example is zero : 29 -50 +'000 =29 -500 inches. 
 
 It may be well to repeat, what has been already conveyed 
 in different words, that English barometers are usually graduated 
 in the following way : 
 
 1. Every long line cut on the barometer scale represents 
 
 T V inch (-100 inch). 
 
 2. Every short line cut on the barometer scale represents 
 
 ^V inch (-050). 
 
 3. Every long line cut on the vernier scale represents T J^ inch 
 
 (010 inch). 
 
 4. Every short line cut on the vernier scale represents y^ 
 
 inch (-002) inch. 
 
 CORRECTIONS TO BE APPLIED TO READINGS OF THE 
 BAROMETER. 
 
 Before barometrical readings taken synchronously at different 
 places by various observers can be compared with each other 
 and used for scientific purposes a number of corrections must be 
 applied to the recorded readings. Many of these corrections 
 have been already mentioned, but all of them must now be 
 referred to and classified according as they relate to a given 
 instrument, or are applicable to the readings of any instrument 
 taken under the same conditions. 
 
 The corrections of the former class are three in number 
 I. Index error. 
 II. Capacity. 
 
 III. Capillarity. 
 
 Those of the latter class are also three in number 
 
 IV. Temperature. 
 
 V. Altitude, or height above sea-level. 
 VI. Gravity. 
 
 I. Index Error. This is detected by careful comparison with 
 a recognised standard barometer. It includes all errors in gradua - 
 tion of the scale. The detection of the index error is simple 
 in the case of the Fortin barometer, but complicated in 
 
144 METEOROLOGY 
 
 that of the Kew barometer. The latter instrument must be 
 tested at every J inch of the scale from 27 to 31 inches, because 
 its inches are less than true inches. To pass through this ordeal 
 it is necessary to use artificial means of increasing or reducing 
 pressure, and so the instrument and the standard have to be 
 placed in an air-tight chamber connected with an air-pump. 
 The instruments can thus be made to read higher or lower as the 
 air in the chamber is compressed or exhausted. Glass windows 
 through which the instruments can be seen are placed in the 
 upper part of the iron air-tight chamber, but, of course, the 
 verniers cannot be used, as the observer is outside the chamber. 
 To overcome this difficulty an apparatus called a cathetometer 
 (Greek, Ka#eros, let down ; hence, 17 Ka#eros [sc. y/m/ypj], a 
 perpendicular line, a perpendicular height ; f^rpov, a measure] 
 has been devised. This is a vertical scale, on which a vernier 
 and a telescope are made to slide by means of a rack and pinion. 
 The divisions on the scale correspond exactly with those on the 
 tube of the standard barometer. The cathetometer (Figs. 40, 
 41, and 42) is placed at a distance of 5 or 6 feet from the air-tight 
 chamber. The telescope carries two horizontal wires, one fixed, 
 and the other capable of being moved by a micrometer screw. 
 The difference between the height of the column of mercury and 
 the nearest division on the scale of the standard can be measured 
 either with the vertical scale and vernier, or with the micrometer 
 wire. The errors detected in this way include not only the 
 index error, but also the correction required for capillarity. 
 
 II. Capacity. The meaning of this term has been already 
 explained. It will be remembered that the siphon barometer 
 requires no correction for capacity. In the Fortin barometer 
 it is provided for by adjusting the scale to the level of the mercury 
 in the cistern, and in the Kew barometer, by shortening the 
 " inches " cut on the scale. In barometers with closed cisterns 
 and a scale of true inches engraved on the case, there is a certain 
 height of the mercurial column, which is correctly measured by 
 the scale. This is called the neutral point, and it should be marked 
 on the scale by the maker, who should also state the ratio of the 
 interior area of the tube to that of the cistern thus : Capacity = ^. 
 
 From these data the correction for capacity is found by taking 
 
BAROMETRICAL READINGS 
 
 H5 
 
 FIG. 40. CATHETOMETF.R CONSTRUCTED FOR THE INDIAN GOVERNMENT. 
 
 10 
 
146 
 
 METEOKOLOGY 
 
 a fiftieth part of the difference between the height read off and 
 that of the neutral point, adding the resulting value to the reading 
 
 FIG. 41. CATHETOMETER, AS USED AT 
 KEW OBSERVATORY. 
 
 FIG. 42. CATHETOMETER, 6 FEET 
 IN HEIGHT. 
 
 when the column is higher than the neutral point, subtracting 
 it from the reading when it is lower than that point. 
 
BAROMETRICAL READINGS 
 
 147 
 
 III. Capillarity. It has been shown above that the effect 
 of capillary action is always to depress the mercury in a barom- 
 eter. The amount of depression is nearly inversely proportional 
 to the diameter of the tube, and is always greater in unboiled 
 than in boiled tubes. 
 
 The correction for capillarity is always additive (+), and 
 according to the Report of the Committee of the Royal Society 
 on Physics and Meteorology, 1840, it varies from + *004 inch in 
 the case of an unboiled, and + '002 inch in that of a boiled tube 
 of the diameter of '60 inch, to + *142 inch in the case of an un- 
 boiled, and + '010 inch in that of a boiled tube of the diameter 
 of '10 inch. 
 
 The certificate of verification of a barometer issued from 
 Kew Observatory includes the three corrections we have been 
 considering as applicable to the individual instrument. Here 
 is a copy of the certificate which accompanied the barometer 
 in use at my own station of the Second Order : 
 
 Corrections to the Scale Readings of Barometer, 877, by Adie y 
 London. 
 
 in. 
 
 At 27'5 
 
 in. 
 At 28-0 
 
 in. 
 
 At 28-5 
 
 in. 
 At 29-0 
 
 in. 
 At 29-5 
 
 in. 
 At 30-0 
 
 in. 
 At 30-5 
 
 in. 
 At 31-0 
 
 in. 
 + 0-030 
 
 in. 
 + 0-026 
 
 in. 
 + 0-022 
 
 in. 
 + 0-019 
 
 in. 
 + 0-015 
 
 in. 
 
 + 0-011 
 
 in. 
 + 0-008 
 
 in. 
 + 0-004 
 
 When the sign of the correction is + , the quantity is to be added to the 
 observed reading ; and when - , to be subtracted from it. The corrections 
 given above include those for Index Error, Capacity, and Capillarity. 
 
 Pro B. STEWART, 
 T. W. BAKER. 
 
 KEW OBSERVATORY, Jan. 7, 18(57. 
 
 * 
 
 IV. Temperature. Both the mercurial column and the brass 
 scale of a barometer expand by heat, and so the height of the 
 column varies with temperature. It therefore becomes necessary 
 to reduce all observations to what they would have been at a 
 given temperature, which is taken as a standard. This standard 
 temperature is 32 F. 
 
 An elaborate table for reducing the readings of barometers, 
 
 102 
 
148 METEOROLOGY 
 
 mounted in brass frames to 32 F., has been computed from the 
 following formula given by Schumacher : 
 
 .m(t- 32) -s(*-62) . 
 
 Correction = -h ^ - ^ '. in which 
 l + m(t-32) 
 
 h = reading of the barometer ; 
 
 t = temperature of attached thermometer; 
 m=expansion of mercury for 1 F., taken as -0001001 
 of its length at 32 ; 
 
 s = expansion of the substance of which the scale is made ; 
 for brass s, is taken as '00001041 of its length (h) at 
 the standard temperature for the scale, viz., 62 F. 
 
 In this table the sign of the correction changes from + to - 
 at the temperature of 29, as the formula gives negative results 
 for 3 below 32. 
 
 V. Altitude. Every barometrical observation should be 
 reduced to mean sea-level as a standard, because as the barometer 
 measures the pressure of the atmosphere, the height of its 
 column will vary with that pressure, becoming less as we ascend 
 and leave some of the atmosphere, and therefore some of its 
 pressure, below us ; and, on the other hand, becoming greater 
 as we descend and leave more of the atmosphere and more of its 
 pressure above us. For Great Britain the mean sea-level at 
 Liverpool has been selected by the Ordnance Survey as their 
 datum, and the altitude of a barometer at any station in England, 
 Scotland, or Wales may be easily determined by reference to the 
 nearest Ordnance Bench Mark. The Ordnance Datum plane for 
 Ireland differs from that for England and Wales by - 7*4 feet 
 that for Ireland being low-water spring-tides, while that for 
 England is mean tide-level. In order that observations in the 
 two countries should be exactly comparable, the Meteorological 
 Office, in 1890, issued new tables for use in Ireland in reducing 
 barometrical readings to the Liverpool Ordnance Datum. 
 
 Any table of corrections for altitude or reduction to mean 
 sea-level must take cognisance of two disturbing elements, the 
 temperature of the air and the actual air-pressure at sea-level 
 at the time of observation. The air temperature must be taken 
 from the dry-bulb thermometer, not from the thermometer 
 
BAROMETRICAL READINGS 149 
 
 attached to the barometer. Table II. in Appendix I. to The 
 Observer's Handbook, published by the Meteorological Office, 
 London, in 1908, being a new and revised edition of Instructions 
 in the Use of Meteorological Instruments, compiled by direction 
 of the Meteorological Council by Dr. Hobert H. Scott, M.A., 
 D.Sc., F.R.S., contains data for reducing to sea-level barometrical 
 observations made at every 10 feet from 10 to 1,500 feet above 
 the datum, and at temperatures varying by 10 from - 20 F. 
 to 100 F. that is, a range of 120. The table is given for two 
 pressures at the lower or sea-level station namely, 30 and 27 
 inches. For intermediate pressures the correction may be 
 obtained by interpolating proportional parts. 
 
 For heights exceeding those given in the table, the value, 
 at the sea-level, of a barometer reading at a station the height 
 of which is knowD, may be calculated from the following formula : 
 
 . 00268 cos 
 
 From a table of common logarithms, the natural number 
 corresponding to log ,, is found ; or v, =n, 
 
 And h=nh'. 
 
 In this formula 
 
 h and h' = barometer reduced to 32 F. at the lower and 
 
 upper stations respectively, 
 
 t and t' =the temperature of the air at the respective stations, 
 / = elevation of upper station in feet, 
 I = latitude of the place. 
 
 The above formula is merely an inversion of the well-known 
 formula given by Laplace in his Mecanique Celeste, for finding the 
 difference of elevation between any two places by means of the 
 barometer, which, adapted to Fahrenheit's thermometer and 
 English feet and inches, is 
 
 /=601591<4(l + *rJ*) (l + -00268 cos 
 
 In this formula / is the difference of elevation between the two 
 stations, and x is the height of the lower station above the sea- 
 level. 
 
150 METEOROLOGY 
 
 In the last factor an approximate value must be used for /. 
 
 Not only, then, can we reduce the barometer reading at one 
 level to that at another, the relative heights of the two stations 
 being known, but we can conversely determine the difference in 
 height between two stations if we know the barometrical readings 
 and the temperature at each at the same moment of time. In 
 other words, we can determine the height of a mountain by 
 barometrical readings taken simultaneously on the summit and 
 at sea-level. 
 
 VI. Gravity. Since the Earth is not a perfect sphere, but is 
 slightly flattened at the poles, it follows that the centre, the seat 
 of gravity, is more distant from the surface at the Equator than 
 it is at the poles. Hence the force of gravity is less at the Equator 
 than it is at the poles. The lessened force of gravity at the Equator 
 has another cause also namely, the centrifugal force arising from 
 the rotation of the Earth on its axis. This acts in opposition to 
 gravitation, and is necessarily greatest at the Equator, gradually 
 lessening as we move northwards or southwards, till at the poles 
 it is nothing. 
 
 The force of gravity accordingly varies slightly with the latitude, 
 and hence barometric readings require to be reduced to a standard 
 latitude in order to make them strictly comparable with one 
 another in various parts of the world. The standard value of 
 gravity adopted is that prevailing at latitude 45. 
 
 The table in Appendix I., taken from The Observer's Hand- 
 book, published by the Meteorological Office, London, 1908, gives 
 the correction for gravity required for each degree of latitude. 
 It is to be added or subtracted according as the sign in the table 
 is + or - . As the correction is practically constant for any 
 one place and, indeed, for all places in the same latitude it 
 is not as a rule applied to individual readings. It should, however, 
 always be quoted at the head of tables of barometer readings. 
 
CHAPTER XIII 
 BAROMETRICAL FLUCTUATIONS 
 
 ATMOSPHERIC pressure as measured by the barometer is subject 
 to two classes of variations periodic and non-periodic. The 
 first are regular ; the second, irregular. 
 
 The regular or periodic variations are (1) diurnal, (2) annual. 
 
 The irregular or non-periodic variations are (1) cyclonic, 
 (2) anticyclonic. 
 
 1. Diurnal variations are best marked within the tropics, or 
 in the torrid zone. They are less marked in temperate climates, 
 absolutely because their physical cause is there less potent, 
 relatively because the irregular variations in atmospheric pres- 
 sure so frequent in higher latitudes tend to mask them. They 
 gradually sink to zero towards the Arctic and Antarctic Circles 
 and the Poles. So regular is the daily rise and fall of the barom- 
 eter in the tropics that Humboldt said that the time of day might 
 be inferred from it within seventeen minutes. 
 
 As the earth rotates on its axis day by day, the hemisphere 
 facing the sun becomes overheated, the air over it expands^ 
 becomes specifically lighter, rises, and tends to flow away from 
 the day hemisphere to the night hemisphere. The barometer 
 consequently falls in the hottest part of the day, reaching a 
 minimum about 3 p.m. But a second, though less decided, 
 minimum occurs about 3 a.m. This cannot be caused in the 
 same way as the day minimum, for, as a matter of fact, the air 
 is coldest about 3 or 4 a.m. We must therefore seek elsewhere 
 for an explanation. It is to be found, according to Dove (and 
 his theory received the sanction of Sabine), in the state of tension 
 of aqueous vapour in the early morning. As we shall see when 
 
 151 
 
152 METEOROLOGY 
 
 we come to discuss the subject of the moisture of the atmosphere, 
 barometrical pressure is made up of two elements the pressure 
 of dry air, and the pressure of aqueous vapour suspended in the 
 atmosphere. This latter is technically called the elastic force, 
 or tension, of aqueous vapour. Now, long before 3 a.m. dew has 
 fallen heavily in other words, the aqueous vapour has been 
 condensed by the nightly fall of temperature, and has left the 
 atmosphere in the condition of dried or desiccated air. In this 
 way the tension of aqueous vapour is largely withdrawn, and the 
 barometer falls. 
 
 As there are two minima of pressure, so there are two maxima. 
 Of these, the first occurs about 10 a.m., the second about 10 p.m. 
 Condensation of the air after a cold night partly accounts for the 
 forenoon wave-crest of pressure. But another potent cause is 
 rapid evaporation, and consequently increasing tension of aqueous 
 vapour. The evening maximum is no doubt due to a brisk 
 decrease of temperature, causing'condensation of the atmosphere, 
 coupled with the saturated state of the air after the evaporation 
 of the daytime. The vapour tension, or elastic force, in a word 
 attains its maximum. 
 
 The foregoing theory affords a rational explanation of these 
 interesting diurnal fluctuations of pressure, and it receives support 
 from the fact that at stations far distant from the sea, or with 
 a high mean temperature that is, at places where the diurnal 
 ranges of temperature are least interfered with by large evaporating 
 surfaces like the ocean or by moist winds the maximum at 
 10 p.m. and the minimum at 3 a.m., which are largely due to 
 the condition of the aqueous vapour, are only slightly marked. 
 
 Dove's theory, however, does not receive universal acceptance, 
 because (in the words of Mr. R. Strachan, F.R.Met.Soc. 1 ) " the 
 diurnal range of vapour tension does not always and everywhere 
 conform to the simple oscillation." According to Mr. Strachan, 
 in the Island of Ascension we still have a double period for the 
 diurnal range of the barometer, but the vapour tension and the 
 dry air pressure of which it is composed both exhibit a double 
 period also. Such cases completely demolish the theory. Mr. 
 Strachan adds : " A hypothesis then remains yet to be framed 
 
 1 " The Barometer and its Uses," Modern Meteorology, p. 89. 1879. 
 
BAROMETRICAL FLUCTUATIONS 153 
 
 which shall account for the diurnal range of the barometer in all 
 seasons and places." 
 
 Be that as it may, the fact remains that every day of twenty- 
 four hours sees two vast waves of high pressure and two equally 
 vast troughs of low pressure sweep round the globe at a speed equal 
 to the revolution of the earth on its axis. It is as if two solar 
 tides, stupendous in extent, with their alternate ebb and flow, 
 were generated in the atmosphere by the action of the sun, or, 
 as Inspector-General Robert Lawson suggests, by alternate 
 accelerations and retardations of the motion of the atmosphere 
 revolving with the earth on its axis, caused by the relation which 
 the atmosphere bears to the orbital motion of the earth as 
 distinguished from its axial motion. 
 
 The diurnal range of pressure, as the difference between the ex- 
 treme daily oscillations is called, exceeds ^ inch within the tropics 
 at Calcutta it is as great as '127 inch in January (dry north- 
 east monsoon), but only '093 inch in July (moist south-west 
 monsoon) the average for the whole year being '116 inch. At 
 Plymouth and in Dublin it is about '020 inch, or only one-sixth 
 of the tropical value. In St. Petersburg it is '01 2 inch, and within 
 the Arctic Circle it merges gradually into the annual range, 
 owing to the length of the circumpolar day and night (Fig. 43). 
 
 From observations in Dublin, extending over as many as 
 forty-five years, I am prepared to say that the diurnal range of 
 pressure is quite perceptible in anticyclonic weather, especially 
 in spring-time, when the air is dry and the diurnal range of tem- 
 perature is large. It is doubtless even better marked at an 
 inland station like Parsonstown or Armagh under like circum- 
 stances. Observations, carefully analysed by Mr. Francis 
 Campbell Bayard, have shown that it increases steadily from 
 north-west towards south-east over Western and Central Europe. 
 
 The following references to the recent bibliography of diurnal 
 range of pressure may prove of interest. The papers will be found 
 in the Quarterly Journal of the Royal Meteorological Society. 
 
 1. On the Diurnal Variations of the Barometer. By John 
 Knox Laughton, M.A., F.R.A.S. (vol. ii., p. 155 read 
 April 15, 1874). 
 
154 METEOROLOGY 
 
 2. The Diurnal Inequalities of the Barometer and Ther- 
 
 mometer, as illustrated by the observations made at 
 the summit and base of Mount Washington, N.H., 
 during the month of May, 1872. By W. W. Rundell, 
 F.M.S. (vol. ii., p. 217 read June 17, 1874). 
 
 3. On the Diurnal Variation of the Barometer at Zi-KaWei 
 
 (a suburb of Shanghai, 31 15' north latitude), and 
 Mean Atmospheric Pressure and Temperature at Shanghai. 
 By Rev. Augustus M. Columbel, S.J. (vol. ii., p. 232 
 read June 17, 1874). 
 
 4. Suggestions on certain Variations, Annual and Diurnal, 
 
 in the Relation of the Barometric Gradient to the Force 
 of the Wind. By the Rev. W. Clement Ley, M.A., 
 F.M.S. (vol. iii., p. 232 read June 21, 1876). 
 
 5. On the Diurnal Variation of the Barometer at the Royal 
 
 Observatory, Greenwich. By William Ellis, F.R.A.S., 
 of the Royal Observatory (vol. iii., p. 467 read June 20, 
 1877). 
 
 6. On a Method of sometimes determining the Amount of 
 
 the Diurnal Variation of the Barometer on any particular 
 Day. By the Hon. Ralph Abercromby, F.M.S. (vol. iv., 
 p. 198 read June 19, 1878). 
 
 7. The Daily Inequality of the Barometer. By W. W. 
 
 Rundell, F.M.S. (vol. v., p. 1 read May 15, 1879). 
 
 8. Diurnal Variations of the Barometric Pressure in the 
 
 British Isles. By Frederick Chambers, Meteorological 
 Reporter, Bombay (vol. v., p. 133 read February 19, 
 1879). See a paper by the same author in the Philo- 
 sophical Transactions of the Royal Society for 1873 
 " Convection Current Theory/* 
 
 9. The Diurnal Range of Atmospheric Pressure. By Richard 
 
 Strachan, F.M.S. (vol. vi., p. 42 read December 17, 
 1879). 
 
 10. Results of Hourly Readings derived from a Redier Baro- 
 
 graph at Geldeston, Norfolk, for the four years ending 
 February, 1886. By E. T. Dowson, F.R.Met.Soc. 
 (vol. xiii., p. 21 read November 17, 1886). 
 
 11. On the Cause of the Diurnal Oscillation of the Barometer. 
 
Sitka Island, Alaska, 
 56 50'. 
 
 St. Petersburg, 59 5(5'. 
 
 Greenwich, 51 28'. 
 
 Halle, 51 28'. 
 
 Geneva, 4*5 13'. 
 
 Grt. St. Bernard, 45 50'. 
 
 Toronto, 43 38 . 
 
 Philadelphia, 39 50'. 
 
 San Francisco, 37 48'. 
 
 Calcutta, 22 35'. 
 Cumana, 10 27'; 
 
 If 
 
 o g 
 
 ss 
 
 i* 
 
 g -S 
 
 s s 
 
 ^ -2 
 
 11 
 
 8 I 
 5 i 
 
 I * 
 
156 METEOROLOGY 
 
 By Robert Lawson, LL.D., Inspector-General of Hospitals 
 (vol. xiv., p. 1 read November 16, 1887). 
 
 12. The Diurnal Range of the Barometer in Great Britain, 
 
 and Ireland, derived from the Hourly Records of the 
 Nine Principal Observatories in the Kingdom during the 
 years 1876-80. By Francis Campbell Bayard, LL.M., 
 F.R.Met.Soc. (vol. xv., p. 146 read April 17, 1889). 
 
 13. Diurnal Barometric Curves in Valleys. By Dr. A. Buchan 
 
 (vol. xix., p. 60). 
 
 14. On Certain Relationships between the Diurnal Curves of 
 
 Barometric Pressure and Vapour Tension at Kenil- 
 worth (Kimberley), South Africa. By J. R. Sutton, 
 M.A., F.R.Met.Soc. (vol. xxx., p. 41 read December 16, 
 1903). 
 
 2. Annual Variations in atmospheric pressure are on a far 
 vaster scale than the daily ranges we have been considering. 
 As typical examples, we may adduce the high-pressure areas 
 observed in January over the central districts of North America 
 (30-20 to 30-30 inches) and over Central Asia (30-30 to 30-40 
 inches and upwards). These anticyclonic systems on a gigantic 
 scale give place in July to equally well-marked low-pressure, 
 or cyclonic, systems the mean pressure falling over the central 
 parts of North America to 29*80 inches and less, or J inch on the 
 average below the mean pressure of January ; and over Central 
 Asia to 29*60 inches and less, or T 8 ^ inch below the January mean. 
 
 Again, compare the low-pressure areas of January situated 
 over the Pacific Ocean south of Alaska (29*60 inches), and over 
 the Atlantic Ocean south of Greenland and Iceland (29*40 inches), 
 with the comparatively high mean pressures for July in these 
 oceanic regions 29*90 to 30-00 inches over the Pacific, 29-80 to 
 29 '90 inches over the Atlantic. 
 
 Let us seek for an explanation of these phenomena. 
 
 Over the centre of that vast continent of the Eastern Hemi- 
 sphere, or Old World, which is formed by Europe and Asia, the 
 air in summer becomes much warmer than that over the Atlantic 
 Ocean to the west, and over the Pacific Ocean to the east. In 
 consequence, the air is rarefied, and the barometer falls over 
 
BAROMETRICAL FLUCTUATIONS 157 
 
 Russia, Siberia, and other inland countries the isobars, or lines 
 of equal barometrical pressure, curving round the area of lowest 
 pressure while it remains comparatively high over the North 
 Atlantic and North Pacific Oceans. In accordance with Buys 
 Ballot's Law, a circulation of wind will commence round the 
 barometrical depression thus formed : an immense cyclone 
 develops, the winds blowing against the hands of a watch, from 
 south-west in India and China (the south-west monsoon) ; from 
 south, south-east, and east in Japan and North-Eastern Siberia ; 
 from north-east and north in North- Western Siberia and Northern 
 Russia ; from north-west and west over the west and south of 
 Europe and South-Western Asia. 
 
 In winter, on the other hand, the air over the central districts 
 of Europe and Asia, rendered dry by the intense heat of summer 
 and its accompanying excess of evaporation, becomes rapidly 
 chilled to an extreme degree. The autumnal snows cover the 
 ground, cutting off terrestrial radiation and causing a still more 
 decided fall of temperature. By this the air is condensed and 
 the barometer rises at a time when a vacuum is forming over the 
 Atlantic and Pacific Oceans owing to the updraught and lateral 
 dispersion of the light warm air which had been resting upon the 
 surface of those oceans, and which may at the time possess a 
 temperature 60 or even 80 higher than that of the air over the 
 interior of the great Continent. Owing to the advancing season 
 also, and the consequent general decrease of temperature, the air 
 over these oceans becomes saturated with moisture, frequent 
 rains result, and a further reduction of atmospheric pressure 
 results, caused by the latent heat set free in the formation of rain. 
 In this way conditions are brought about which are the reverse 
 of those observed in summer : an immense anticyclone is formed, 
 the winds circulating round and out from the centre of high 
 pressure in a direction with the hands of a watch, blowing from 
 north-west and north in Japan and China ; from north-east in 
 India (the north-east monsoon) ; from east and south-east in Russia 
 and Southern Europe ; from south-west in the British Isles ; and 
 from west in Northern Russia and Siberia. In Central Russia and 
 Siberia a region of calms will exist near the position of the highest 
 atmospheric pressure. In the winter season the predominant 
 
158 METEOROLOGY 
 
 winds over Scandinavia are south-easterly, but this apparent 
 anomaly is in fact a beautiful fulfilment of the very laws it 
 seems to contradict. We have seen that in winter a barometrical 
 depression exists over the North Atlantic Ocean, particularly 
 over that portion of it which is called the Norwegian Sea. It 
 is this which draws the wind from south-east over Sweden and 
 Norway, in strict agreement with Buys Ballot's Law. 
 
 A precisely similar state of things, though on a somewhat 
 reduced scale, holds good in the Western Hemisphere, or New 
 World. In summer a barometrical depression, or cyclonic 
 system, develops over Upper Canada and the Central States of 
 the Great Republic, round which minimum the prevailing 
 winds sweep in a gentle curve against watch-hands. In winter, 
 on the contrary, an area of high atmospheric pressure, or anti- 
 cyclonic system, develops in the same region, and round its central 
 zone of calms the prevailing winds blow with watch-hands. 
 Hence the prevalent north-west and north winds, which bring 
 to Labrador, Lower Canada, and the Eastern States the rigorous 
 winters of the American Atlantic seaboard ; although, of course, 
 the setting of a polar current of iceberg-laden water southwards 
 along that seaboard intensifies the rigours of the climate, just 
 as the warm waters of the North Atlantic north-easterly surface- 
 drift in laving the western shores of Europe temper the climate 
 even further north than the Arctic Circle. 
 
 In the department of Cosmical Physics of Section A (Mathe- 
 matics) at the meeting of the British Association, at Winnipeg, 
 in August, 1909, Mr. R. F. Stupart, Director of the Canadian 
 Meteorological Service, read a paper on the " Distribution of 
 Atmospheric Pressure in Canada." The chief points of the paper 
 were : (1) That the world charts of pressure distribution give an 
 inadequate and even inaccurate representation of the pressure 
 conditions in the Dominion ; (2) that relatively high pressure in 
 the North -West at Dawson City is accompanied by relatively 
 mild winters and low pressure by severe winters, a fact which is 
 directly contrary to the prevailing idea that in winter the higher 
 the pressure the lower the temperature over continental areas. 
 
 In Equatorial regions, where air temperature and moisture 
 are constants throughout the year, the annual variation in 
 
BAROMETRICAL FLUCTUATIONS 159 
 
 atmospheric pressure is trifling in amount. In the southern 
 hemisphere, however, seasonal changes in pressure again become 
 marked, although they are not so pronounced as in the northern 
 hemisphere, where dry land or continent so largely takes the 
 place of ocean. 
 
 Trade, Winds. Leaving out of count the great disturbances 
 of pressure from winter to summer and from summer to winter, 
 caused by the rise and fall of temperature over the continents 
 of the Old and New Worlds, we find that a belt or ridge of com- 
 paratively high pressure, from 30'00 to .30 '20 inches, encircles the 
 earth at the tropics both north and south of the equator, while over 
 the Equator and the immediate vicinity to 10 or 15 north and 
 south, the barometer stands from ^ to ^y mcn lower. In the 
 northern hemisphere this ridge lies approximately along latitude 
 35. In the southern hemisphere it is situated about latitude 30. 
 These areas of high and relatively low pressure oscillate backwards 
 and forwards with the season : in January the northern zone 
 of high pressure approaches the Equator, while the corresponding 
 southern zone recedes from it. Conversely, in July the northern 
 zone retreats northwards, while the southern advances towards 
 the Equator. In obedience to Buys Ballot's Law, permanent 
 winds blow from these respective areas of high pressure towards 
 the Equatorial trough of low pressure, constituting the North- 
 East Trades of the Tropic of Cancer and the South-East Trades 
 of the Tropic of Capricorn. 
 
 Dr. Alexander Buchan, in a masterly analysis of barometrical 
 observations taken at some four hundred stations scattered all 
 over the globe, has ascertained that atmospheric pressure is 
 lowest throughout the year over the Antarctic Ocean, about 
 29 inches. " In the hemisphere where winter reigns, the greatest 
 pressure lies over the land ; the larger the continent, the greater 
 the pressure. In the hemisphere where summer reigns the low 
 pressures are over the land, the high over the oceans." Mr. R. 
 Strachan, whose words we have just quoted, gives the following 
 table of the most remarkable areas of high and low pressures 
 (see p. 160). 
 
 So far periodical variations in atmospheric pressure have been 
 our theme, \Ve have now to consider those irregular variations 
 
160 
 
 METEOROLOGY 
 
 which daily, monthly, and yearly occasion changes in wind and 
 weather over more or less extensive areas of the earth's surface. 
 They are measured or determined by drawing lines of equal 
 barometrical pressure, or isobars, on a map of the area under 
 discussion, which is then called a synoptic weather chart. These 
 
 Period. 
 
 Position. 
 
 Pressure. 
 
 
 
 Inches 
 
 
 / Iceland 
 
 29-4 
 
 December, January, February . . 
 
 I 50 N. 170 W. 
 | 50 N. 100 E. 
 
 29-6 
 30-4 
 
 
 I 0to40S. 
 
 30-0 
 
 June, July, August 
 
 /40 N. 90 E. 
 \.30 N. 40 W. 
 
 29-5 
 30-2 
 
 isobars are drawn for each ^ inch. They tend to assume two 
 primary and five secondary shapes (Fig. 44). If they enclose 
 an area of low pressure, forming a circle or an oval, they are 
 described as cyclonic in shape, from the Greek Kv/cAos, a circle. 
 If, on the contrary, the isobars encircle an area of high pressure, 
 
 Cyclo 
 
 -O 
 
 lone Wedge / S~Gvc)ont 
 . s S\ Jf f i 
 
 )) I \ i I .//' / 
 
 **'' X* 
 V- depress? /'**" " x , Second* 
 
 one 
 <2 9-9 
 
 \\.~. 
 
 Col. 
 
 jo. i 
 
 FIG. 44. CYCLONIC AND ANTICYCLONIC ISOBARS. 
 
 they are described as anticyclonic, the Greek preposition 
 meaning originally over against, opposite hence, in opposition to. 
 
 Thus we have two primary types of isobars cyclonic and 
 anticyclonic. 
 
 The secondary shapes are five in number. They are for the 
 most part modifications of the primary types or connected with 
 
BAROMETRICAL FLUCTUATIONS 161 
 
 either one or other of them. Thus, in a cyclonic system one or 
 more of the isobars sometimes curves outwards from the centre, 
 forming a loop which embraces a secondary area of low pressure 
 in the periphery of the primary cyclone. Such a system is 
 called in consequence a secondary or subsidiary depression. 
 
 Again, isobars embracing an area of relatively low pressure, 
 instead of curving into a cyclone, run or bend into the shape of 
 the letter V. Such a system is called a V-shaped depression. 
 
 Occasionally, in the third place, the isobars run parallel to each 
 other, or nearly so, it may be for hundreds of miles, assuming the 
 form neither of cyclonic nor of anticyclonic isobars. They are 
 then called straight isobars. 
 
 In the fourth place, when cyclonic systems are following each 
 other in rapid succession, and when an anticyclone is in the 
 neighbourhood, a tongue of high pressure inserts itself like a 
 wedge between two areas of low pressure. Such a system 
 resembles an inverted V, but is the converse of a V-shaped 
 depression, because its isobars enclose an area, not of low, but of 
 high pressure. It is called a " wedge." 
 
 Lastly, two anticyclones may be connected with each other 
 by means of a furrow or neck of relatively low or less high pressure, 
 and this system is called a col, because it is analogous to the col 
 which forms a pass between two adjacent mountain-peaks (Ralph 
 Abercromby). 
 
 Speaking in general terms, we may contrast cyclonic with 
 anticyclonic systems as follows : 
 
 1. Cyclonic areas in the northern hemisphere as a rule travel, 
 it may be, at the rate of twenty miles an hour or upwards in 
 Equatorial regions from east to west, in extra-tropical latitudes 
 usually from west to east. The " westing " of cyclonic systems 
 generated between the Equator and the Tropic of Cancer is due 
 to the general westward drift of the atmosphere throughout those 
 low northern latitudes (N.E. trades). The area throughout 
 which this occurs in the North Atlantic Ocean covers the Carib- 
 bean Sea and the region to the eastward of the Windward Islands. 
 As the centres of low pressure draw away from the Equator until 
 the northern limit of the N.E. trades is reached, the cyclonic 
 path usually, but not always, recurves, and turns eastward and 
 
 11 
 
162 METEOROLOGY 
 
 towards the Pole. When this happens, the resulting storm loses 
 the characteristics peculiar to tropical cyclonic storms, which 
 are called " West India hurricanes " in the North Atlantic, and 
 " typhoons " in the Pacific, off the coast of Asia. 
 
 Anticyclonic systems, on the contrary, are often stationary 
 for days or weeks, or their motion is slow and irregular. They 
 frequently move away from the track of cyclonic systems almost 
 at right angles. Occasionally a shallow depression forms within 
 the confines of an anticyclone, and dull, rainy weather ensues, 
 with much haze or fog. 
 
 2. In cyclonic systems the isobars generally approach each 
 other much more closely than do those of anticyclones in other 
 words, the gradients are steeper, and therefore the winds are 
 stronger in cyclones than in anticyclones. 
 
 3. Unsettled, windy, rainy, or showery weather is commonly 
 associated with cyclonic systems. In anticyclones, on the 
 contrary, conditions are as a rule fine, quiet, and dry. In winter, 
 however, dense fogs sometimes accompany the calms of an 
 anticyclone, and in parts of its periphery the sky may be densely 
 clouded. If rain should fall, it is usually drizzling not heavy. 
 In summer, hot sunshine by day and cool nights accompany an 
 anticyclone, and sea-fogs are prevalent when the calm-centre 
 overlies the sea. In any case much haze obscures the horizon. 
 
 4. Thunderstorms are very apt to develop in connection with 
 V-shaped and secondary depressions. In the former, violent 
 shifts of wind and sudden changes of temperature usually occur, 
 these phenomena being accompanied by heavy squalls and showers 
 of rain and hail, or, in winter, snow. 
 
 5. The weather accompanying anticyclones is well described 
 by the Hon. Ralph Abercromby as " radiation weather " hot 
 suns by day and cool nights in summer ; intense frost in winter, 
 so long as the sky is tolerably clear. 
 
 6. The term " intensity " applied to an anticyclone means that 
 the barometer has reached an unusual height in its centre ; 
 further, that the system is of vast extent and also of long dura- 
 tion. The same term applied to a cyclonic system means that 
 the isobars are close together, and that the system is deep and 
 moving quickly. " There is no difference," says the Hon. Ralph 
 
BAROMETRICAL FLUCTUATIONS 163 
 
 Abercromby, 1 " between the cyclones which cause storms and 
 those which cause ordinary weather except intensity." 
 
 The great majority of the atmospheric depressions which pass 
 over the British Islands come in from the Atlantic, and travel most 
 usually in a north-easterly, less frequently in an easterly or south- 
 easterly, direction. Their advent, passage, and departure are 
 attended by definite changes in the weather. First, in front of 
 the disturbance, the sky becomes streaked with cirrus cloud, 
 which spreads out into a thin veil of cirriform cloud or cirro- 
 stratus, in which solar or lunar halos develop, and through which 
 is seen a " watery sun " or a " watery moon," as the case may be. 
 The wind freshens from south-east, the cloud canopy thickens, 
 and a stratum of lower clouds of the scud-cumulus type develops, 
 coming up from south-east or south under the cirriform sheet, 
 which is probably travelling from west or south-west. Drizzling 
 rain next sets in, and finally heavier driving rain and squalls 
 from the southward. As the lowest pressure is reached, the wind 
 may fall calm if the centre of the system is near at hand. Then 
 more or less suddenly and completely the wind shifts to west or 
 north-west, with dense clouds, heavy rain or showers of rain 
 and hail, and a brisk fall of temperature. The clouds now break, 
 and massive cumuli drive past across a deep blue sky, sharp 
 showers probably falling at intervals in the rear of the disturbance. 
 
 In the case we have been supposing, the depression or cyclonic 
 system is travelling north-eastwards. It is to be noted that 
 the wind is said to " veer," or " haul," when it changes with the 
 sun ; for example, when it changes from east to south-east, 
 south, or south-west, or from south to west or north-west, or 
 from north-west to north-east, and so on. Should it change 
 against the sun, it is said to " back," and when it changes com- 
 pletely in direction, as from east to west or south to north, it 
 is said to " shift." 
 
 On rare occasions depressions move slowly westwards or north- 
 westwards from the Continent of Europe towards the British 
 Islands. The western or north-western quadrant of the depres- 
 sion is then its front, and the weather varies accordingly, being 
 " muggy," cloudy, and damp or wet. 
 
 1 Weather, p. 29. London: Kegan Paul, Trench and Co. 1887. 
 
 112 
 
164 METEOROLOGY 
 
 In anticyclones barometrical gradients are comparatively 
 slight, and the force of the wind is correspondingly moderate 
 as a rule. During winter intense cold prevails in the centre 
 and in the south-east and south-west quadrants of the anti- 
 cyclone ; in its north-west and north-east quadrants at least, 
 in Western Europe conditions are milder. A typical instance 
 of such a distribution of temperature occurred in the memorable 
 frost of 1890-91. An anticyclone hung for weeks over Central 
 Europe and the southern half of the British Islands. Intense 
 cold prevailed in these districts, while the west of Ireland, the 
 greater part of Scotland, and Scandinavia, came under the 
 influence of warm south-west and west winds skirting the north- 
 west and north-east quadrants of the anticyclone. The result 
 was that in the extreme north of Scotland, as well as in the west 
 of Ireland, the mean temperature for fifty-nine days (November 25, 
 1890, to January 22, 1891) was 10 higher than in the south-east 
 of England. The mean temperature for the period was 10 or 
 more below the average over the southern Midlands and south of 
 England. In the north of England the deficiency, however, did 
 not amount to 5, and in the extreme north of Scotland it was less 
 than 1. At Sumburgh Head, in the Shetlands, frost occurred 
 on only nine of the fifty-nine days, whereas at Biarritz it occurred 
 on thirty-one days, and at Kome on nine days. At Brussels it 
 froze daily throughout the period. On January 19, 1891, the 
 harbour at Toulon was reported to be frozen over for the first 
 time on record, while the ice floating on the Thames between 
 London Bridge and the Tower was so packed that all movements 
 of vessels had entirely ceased. On January 20, the River Tagus 
 at Lisbon was frozen over, and the Ebro was covered with 19 inches 
 of ice, the first since 1829. In Regent's Park skating lasted 
 forty-three days consecutively (December 13, 1890, to January 24, 
 1891), according to Colonel Wheatley, R.E., of the Office of 
 Works. 
 
 Anticyclonic weather in summer is characterised by dry, quiet, 
 bright weather, hot suns by day being followed by cool nights, 
 except in the north-west quadrant of the system, where the nights 
 are warm and often cloudy. 
 
CHAPTER XIV 
 THE ATMOSPHERE OF AQUEOUS VAPOUR 
 
 AT the close of Chapter III., on the Composition of the Atmo- 
 sphere, it was pointed out that one of its most important con- 
 stituents was aqueous vapour, or water in a gaseous or aeriform 
 state. Moisture is universally present in the atmosphere, bub 
 in very variable proportions as regards both time and place. 
 To its elastic force or tension the height of the barometer is to 
 some extent due, and we might with propriety speak of two 
 atmospheres instead of one atmosphere the atmosphere of 
 aqueous vapour as well as the atmosphere of dry air. No other 
 factor singly exercises so profound, so far-reaching an influence 
 on weather as the aqueous vapour of the atmosphere. Its liability 
 to alter its form from the gaseous to the liquid or solid state and 
 back again, the caloric phenomena which accompany these 
 changes, and the extreme variability in amount of vapour present 
 in the air, these all cause frequent fluctuations in temperature 
 and pressure, in cloud and sunshine, in terrestrial and solar 
 radiation, in wind and weather. 
 
 Watery vapour is constantly distilling into the atmosphere 
 from the surface of oceans, lakes, and rivers, and from the moist 
 soil. In general the tiny molecules, which make up the vapour, 
 are invisible as they rise into the atmosphere to diffuse freely 
 through the air and to float about in the interstices between the 
 atoms of oxygen and nitrogen which compose the atmosphere. 
 If, however, the aerial strata are much colder than the water 
 surface upon which they rest, the evaporating water may appear 
 instantly as steam or fog. This is one cause of winter fogs in 
 the vicinity of large rivers like the Thames. On a frosty day such 
 a river may be seen to literally steam into the atmosphere the 
 
 165 
 
166 METEOROLOGY 
 
 aqueous vapour being condensed as soon as it has separated from 
 the water by evaporation. The fact is, that only a certain 
 quantity of aqueous vapour can diffuse through the air in an 
 invisible form, and that the quantity which can so diffuse varies 
 with the temperature of the air. The warmer the air, the greater 
 the quantity of vapour which it can sustain in an invisible state. 
 The colder the air, the smaller the quantity of vapour it can so 
 sustain. Setting out from 32 P., at which the air can sustain 
 y^ of its weight of transparent vapour, we find that for every 
 increase of temperature of 27 the vapour-sustaining capacity 
 of air is doubled. Thus, at 59, air can sustain the F T a th part 
 of its own weight of vapour, and at 86 the ^ (T th part. Each 
 cubic foot of saturated air at 32 F. contains only 2 '37 grains of 
 aqueous vapour ; at 60 it contains 5'87 grains ; and at 80, 
 10*81 grains. It follows from this that, if the atmosphere is 
 suddenly chilled from 80 to 60, nearly 5 grains of vapour will 
 be condensed out of every cubic foot of air, forming mist or cloud 
 and falling as rain. This is really the explanation of one of the 
 most potent causes of rain. 
 
 It is evident that aqueous vapour, while constantly present 
 in the atmosphere, is equally constantly passing into it 
 by evaporation and passing out of it by condensation. The 
 subject, therefore, naturally falls under three headings, which 
 are well given by Dr. R. H. Scott, F.R.S., in his Elementary 
 Meteorology, as follows : 
 
 " 1. Atmometry, or the determination of the amount of water 
 passing into the air by evaporation. 
 
 " 2. Hygrometry, or the determination of the amount of water 
 present in the air in the vaporous form. 
 
 " 3. Hyetometry, or the determination of the amount of water 
 condensed out of the atmosphere in the form of [dew], [hoar-frost], 
 rain, hail, or snow." 
 
 ATMIDOMETRY. 
 
 Evaporation is the process by which water is changed from the 
 liquid or solid (ice or snow) state into vapour, and is carried off 
 into the atmosphere as such. Atmometry, or, more correctly, 
 atmidometry (Greek, dr/xos or ar/us, steam or vapour ; ptrpov, a 
 measure), is the determination of the amount of evaporation by 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 16? 
 
 means of instruments which are indifferently called evaporim- 
 eters, atmometers, or atmidometers. Evaporation takes place 
 most quickly into dry air at a high or increasing temperature. 
 It is also facilitated by high wind, and to some extent by low 
 barometrical pressure. In Western Europe the process is most 
 active in spring, when the capacity for moisture of the atmosphere 
 is increasing owing to the prevalence of desiccated easterly winds, 
 whose temperature is fast rising. On the other hand, in late 
 autumn (November) evaporation is usually almost at a stand- 
 still, because the temperature of the air is falling fast, and its 
 capacity for moisture is diminishing, so that it is charged with 
 vapour, or saturated. When this last condition is present, 
 evaporation ceases and the slightest additional fall of tem- 
 perature would cause condensation into fog, cloud, or rain. 
 
 In evaporating, every grain of water absorbs heat sufficient 
 to raise 960 grains of water through 1 of Fahrenheit's scale. 
 This heat is extracted from neighbouring objects, and is made 
 latent that is, it lies hid or concealed in the vapour, ready to be 
 used again in the converse process of condensation. Latent heat 
 can no longer excite a sensation of warmth, or be measured by the 
 thermometer. It is, however, existent, and is employed in keeping 
 the vaporous molecules " floating loosely and widely apart " 
 (R. J. Mann). 
 
 The coolness produced by evaporation has been utilised in 
 hot climates in many ways. Porous earthenware j ars are employed 
 to cool drinking-water, and in India railway carriages are cooled 
 by placing damp matting across the windows, while ice is formed 
 by exposing water in shallow pans, laid on straw, to the combined 
 effects of evaporation and radiation at night. 
 
 From data which were collected some years ago by the late 
 Rev. Samuel Haughton, M.D., F.R.S., Senior Fellow of Trinity 
 College, Dublin, 1 it would seem that in nearly all parts of the 
 globe, situated reasonably near the coast, the rainfall is about 
 equal to the evaporation from a free water surface, and that there 
 can be no great transference of vapour from the torrid to the 
 temperate zones (R. H. Scott). 
 
 1 Six Lectures on Physical Geography, p. 165. London: Longmans, Green 
 and Co. 1880. 
 
168 METEOROLOGY 
 
 Even at the present day no entirely satisfactory atmidometer 
 or evaporimeter exists. 1 
 
 On November 24, 1859, Dr. Babington, F.R.S., exhibited to 
 the Royal Society the evaporimeter which now bears the name of 
 " Babington's atmidometer." It consists of an oblong hollow 
 bulb of glass or copper, beneath which, and communicating with 
 it by a contracted neck, is a second globular bulb, duly weighted 
 with mercury or shot. The upper bulb is surmounted by a small 
 glass or metal stem, having a scale graduated to grains and half- 
 grains, on the top of which is fixed a shallow metal pan. The 
 bulbs are immersed in a vessel of water having a circular hole 
 in the cover, through which the stem rises. Distilled water is 
 poured into the pan above until the zero of the stem sinks to a 
 level with the cover of the vessel. As the water in the pan 
 evaporates, the stem ascends, and the amount of the evaporation 
 is indicated in grains. 
 
 In Professor von Lament's atmometer the evaporation pan is 
 a shallow cylinder with a slightly curved bottom, from the 
 middle of which a narrow pipe leads to a vertical cylindrical 
 reservoir of water, containing a closely-fitting piston. The 
 position of this piston in the cylinder is adjustable by means of 
 a screw which moves the piston vertically, and it can be read by 
 a vertical scale attached to the piston, a pointer being carried 
 by the cylinder. The method of observing is as follows : The 
 
 1 The reader who is inter ;sted in the subject will find articles on evapora- 
 tion, in which exhaustive descriptions of the principal instruments in use for 
 measuring evaporation are given, in : (1) Symons's British Rainfall, p. 151. 
 1869. (2) Symons's Monthly Meteorological Magazine, pp. 70-74. 1870. 
 (3) Ibid., pp. 156-159. 1876. (4) Ibid,, p. 2. 1887. (5) Symons's British 
 Rainfatt, pp. 18-43. 1889. (6) Ibid., pp. 17-29. 1890. (7) Quarterly 
 Journal of the Royal Meteorological Society, vol. xvii., pp. 186, 187. No. 79. 
 July, 1891. (8) "Records of Evaporation" in the yearly volumes of 
 British Rainfall. (9) " Measurement of Evaporation," by Richard Strachan 
 F.R. Met. Soc., Quarterly Journal of the Royal Meteorological Society, vol. xxxi., 
 No. 136. October, 1905, p. 277. (10) " Description of the Wilson Radio- 
 integrator," British Rainfall, 1907, p. 44. (11) "Methods and Apparatus 
 for the Observation and Study of Evaporation," by C. F. Marvin, Professor 
 of Meteorology. Monthly Weather Review, April and May, 1909. U.S. 
 Department of Agriculture, Weather Bureau. (12) "Studies on the 
 Phenomena of the Evaporation of Water over Lakes and Reservoirs," by 
 Professor Frank H. Bigelow. Ibid., July, 1907, February, 1908. Annual 
 Summary, 1908. (13) " An Annotated Bibliography of Evaporation," by 
 Mrs. Grace J. Livingston (from A.D. 1670 to the present day). Ibid., June, 
 September, November, 1908 ; February, March, April, May, 1909. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 169 
 
 piston is screwed up so as to allow the water in the evaporation 
 pan to run into the reservoir, leaving the connecting tube quite 
 full, so that the water just makes the curved surface of the bottom 
 of the pan continuous. The scale is then read, and the water 
 is driven by the piston up to within a little of the top of the pan, 
 and the evaporation is allowed to take place. The piston is 
 then raised, so that the water sinks again from the pan to the 
 same point as before, and the scale is read again. The difference 
 of readings in scale divisions gives the depths of water evaporated. 
 
 A manifest fault in these instruments is the exposure of the 
 water in the evaporation-dish to gusts of wind at all seasons, 
 causing waste ; and to frost in winter, stopping their mechanical 
 movements. 
 
 In de la Rue's evaporimeter the water evaporates from a 
 surface of moistened parchment paper, stretched over a shallow 
 drum kept full of water, which is supplied from a cylindrical 
 reservoir giving about 6 inches head. Into this vessel dips a 
 narrow metal tube, forming the only opening into a graduated 
 cylinder of glass about 6 inches high and 1J inches in diameter. 
 The glass cylinder is in the first instance filled with water, and 
 the tube leading from it, which dips into the reservoir, is per- 
 forated laterally. The water in the reservoir is therefore main- 
 tained at a constant level by a flow of water from the glass cylinder 
 whenever the lateral opening becomes exposed to the air. The 
 amount of water evaporated is given by the graduations on the 
 glass cylinder, which are so drawn as to express the evaporation 
 in hundredths of an inch. 
 
 A self - recording evaporation gauge was exhibited by 
 MM. Richard Freres, of Paris, at the Twelfth Annual Exhibi- 
 tion of Instruments held by the Royal Meteorological Society 
 in March, 1891. It consists of a pair of scales, one of which 
 bears a basin of water or a plant. Weights are placed in the 
 opposite scale to establish a state of equilibrium. A style is 
 attached to the scale beam, and the pen records its motions on 
 a revolving drum. The sensitiveness of the scale is regulated 
 by a sliding weight, which being raised or lowered, raises or 
 lowers the centre of gravity of the scale beam. 
 
 Mr. Spencer P. Pickering, M.A., F.R.S., invented and patented 
 
170 
 
 METEOROLOGY 
 
 an instrument which affords a simple means of measuring directly 
 the volume of water evaporated from a moist surface of known 
 area. In " Pickering's Patent Standard Evaporimeter," the 
 moist surface consists of a piece of linen (originally of a sheet 
 of blotting-paper) measuring 100 millimetres by 50 millimetres, 
 held vertically, by means of a hinged frame, over a copper water- 
 reservoir fitted with a graduated glass side-tube, as shown in the 
 figure (Fig. 45). The sheet of linen ends in a tongue which dips 
 into soft, distilled, or rain water, and is thus kept damp. The 
 graduations are such that they give the number of units of volume 
 
 evaporated per unit area exposed. Thus 
 a fall of -24 shows that -24 cubic inch, 
 or cubic centimetre, has evaporated 
 from each square inch, or square centi- 
 metre, of the surface exposed. 
 
 In British Rainfall the late Mr. G. J. 
 Symons, F.R.S., each year, from 1885 
 inclusive, published a return of the 
 evaporation from a water surface at 
 his residence, Camden Square, London, 
 N.W. He kept a daily record of the 
 depth of water evaporated from the 
 surface of a tank 6 feet x 6 feet x 
 2 feet, buried 20 inches in the ground, 
 and in which water to a depth of about 
 22 inches was usually kept. The average 
 annual evaporation from this water sur- 
 face in the years 1885-91 was 14-5 inches. Evaporation in London 
 is greatest in June and July, least in December and January. This 
 record has been continued year by year since Mr. Symons's death 
 by his successor, Dr. Hugh Robert Mill. The monthly average 
 evaporation at Camden Square in the twenty-two years, 1885-1906 
 inclusive, and the monthly evaporation in 1907 were as follow : 
 
 FIG. 45. PICKEB ING'S STANDARD 
 
 EVAPORIMETKR. 
 
 
 Jan. Feb. 
 
 Mar. April. 
 
 May.' June. 
 
 July. 
 
 Aug. 
 
 Sept. Oct. 
 
 Nov. 
 
 Dec. Total. 
 
 1 
 
 
 in. i ip. 
 
 in. in. 
 
 in. in. 
 
 in. 
 
 in. 
 
 in. in. 
 
 in. ! in. ] in. 
 
 Avirage ... 
 
 09 -26 
 
 67 1-52 
 
 2"35 2'94 
 
 3-18 
 
 2-39 
 
 11-35 -62 
 
 23 '08 15-68 
 
 1907 
 
 21 '38 
 
 1-02. 1-27 
 
 1-03 2-47 
 
 2-65 
 
 2-24 
 
 172 -49 
 
 27 
 
 19 14-84 
 
 Difference 
 
 + -12J+ -la 
 
 + 35 -'25 
 
 -42 -'47 
 
 -53 
 
 -15 '37i -'13 
 
 + '04 
 
 + 11 -'84 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 171 
 
 For many years the late Mr. James Price, M.Eng., Univ. 
 Dublin, C.E., of Knockeevin, Greystones, County Wicklow, 
 kept evaporation gauges in connection with rain gauges in Dublin, 
 and at Cavan, Sligo, Galway, and Athlone. Several reliable 
 series of observations of great interest were obtained. For 
 instance, during two consecutive years the evaporation in Dublin 
 and at Galway was 26 inches, whereas in the County Cavan only 
 13 inches of water passed off into vapour. This showed that the 
 habitual dampness of the air was exactly twice as great in Cavan 
 as in Galway or Dublin. In Cavan there is a retentive sticky 
 clay subsoil, with endless lakes and a less wind-movement than 
 on the coast at Galway, which also has a dry gravelly subsoil, 
 with but little stagnant water. The dampness is independent 
 of the rainfall, which is heavier at Galway than in Cavan ; nor 
 does it depend so much on the contiguity of the sea as on the 
 nature of the subsoil and the amount of stagnant fresh water 
 in the district. Mr. Price pointed out that there is something 
 particularly bracing and invigorating in the air of those parts 
 of Galway and Clare where a gravelly subsoil occurs. Indeed, 
 he was so strongly of opinion that habitual dryness of the air as 
 indicated by evaporation gauges has more to do with health 
 than the matter of rainfall, that he suggested that each locality 
 should he YC its evaporation gauges as well as its rain gauges. 
 To make the results of observations on evaporation comparable, 
 Mr. Price recommended that the water level in the evaporation 
 gauge should be kept at a given standard, and he stated that he had 
 devised a plan whereby this can be accomplished automatically. 
 
 At a meeting of the Royal Meteorological Society held on 
 Wednesday, April 21, 1909, Mr. Baldwin Latham, M.Inst.C.E., 
 read a paper on " Percolation, Evaporation, and Condensation/' in 
 which he gave the results of the observations which he had carried 
 out at Croydon on the subjects during the last thirty years. 
 Two percolation gauges were used, both of which were exactly 
 a superficial yard in area, and contained a cubic yard of natural 
 soil, one of chalk and the other of gravel. The average annual 
 amount of percolation through the chalk gauge was 10-84 inches, 
 and through the gravel gauge 10*34 inches. The average annual 
 rainfall was 25-456 inches. It appears that the rate of percolation 
 
172 METEOROLOGY 
 
 is governed by the rate of rainfall, for when once the gauges have 
 become sensitive by being thoroughly wetted, the rate at which 
 rain percolates depends entirely on the quantity of rain imme- 
 diately falling. The evaporator used for determining the 
 evaporation was a floating copper vessel 1 foot in diameter, 
 supported by a life-buoy ring, connected by four arms with the 
 evaporating vessel, the whole being floated in a tank 4 feet internal 
 diameter, containing about 3 feet depth of water. The average 
 annual amount of evaporation by this gauge was 18 - 137 inches, 
 and the average annual amount of what has been termed " nega- 
 tive evaporation " or really condensation was '359 inch. 
 
 Mention was made at p. 34 of the Barton Moss Evaporation 
 Station near Southport, Lancashire. At this station the 
 standard evaporation tank is one of Symons's pattern, 6 feet 
 square and 2 feet deep, and its rim is 3 inches above the ground. 
 The height of the water is measured daily, at 9 a.m., by means 
 of a Halliwell Float and Multiplying Index-Finger Gauge. The 
 amount of evaporation is entered to the previous day, as is also 
 the rainfall. A second evaporation tank, only 3 feet square, 
 but in all other respects similar to the standard one just described, 
 is in use for comparative purposes. 
 
 The rain-gauge is of the Snowdon pattern, 5 inches in diameter. 
 Its rim is 9 inches above the ground. 
 
 The table on p. 173 is taken from Mr. Joseph Baxendell's 
 Meteorological Report to the Southport Corporation for 1908. 
 Barton Moss is only 14 feet above mean sea-level. 
 
 This table shows very clearly how deficient is the evaporation 
 throughout the winter months, how rapidly it increases in spring, 
 how it exceeds the rainfall in early summer, and how quickly 
 it lessens from August to November. 
 
 It is to the United States of America that we must look for 
 investigations into the subject of evaporation on a large scale. 
 In May, 1907, the Salton Sea, Southern California, consisted of 
 a sheet of fresh water, 45 miles long, and about 10 or 15 miles 
 wide, containing 440 square miles of surface area, 205 feet below 
 the mean tide-level of the Pacific Ocean. This body of water 
 had recently been formed by an overflow from the Colorado River, 
 and was protected from further inflow. It was estimated that 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 173 
 
 this lake would probably dry out in ten or twelve years, so that 
 an unusually fine example of evaporation on a large scale in the 
 arid climate of Southern California was offered for study. 
 
 Preliminary observations were undertaken in July, August, 
 and September, 1907, at Reno, Nevada, a district lying to the 
 east of the Sierra Nevada Mountains, which is very favourable 
 for excessive evaporation, without the discomfort of abnormally 
 high summer temperatures, such as occur in the Salton Basin. 
 The expedition to Reno was furnished with some improved 
 apparatus designed by Professor C. F. Marvin, U.S. Weather 
 
 TABLE V. EVAPORATION AND RAINFALL AT BARTON Moss, 
 LANCASHIRE. 
 
 
 Evaporation per 6-foot Tank. 
 
 
 1908. 
 
 Total 
 
 Difference from 
 
 Rninfall. 
 
 
 Evaporation. 
 
 the Average. 
 
 
 
 Inches. 
 
 Inches. 
 
 Inches. 
 
 January 
 
 31 
 
 - -10 
 
 2-11 
 
 February 
 
 68 
 
 + -21 2-33 
 
 March 
 
 1-09 
 
 - -25 
 
 2-48 
 
 April 
 May 
 
 2-32 
 3-04 
 
 + -04 
 + '05 
 
 2-50 
 2'59 
 
 June 
 
 3-27 
 
 - -34 
 
 165 
 
 July 
 
 3-27 
 
 - -40 
 
 3-14 
 
 August 
 
 3-09 
 
 - -10 
 
 3-33 
 
 September 
 
 1-69 
 
 - -35 
 
 3-76 
 
 October 
 
 1-05 
 
 - -17 
 
 2-23 
 
 November 
 
 57 
 
 + -01 
 
 2-45 
 
 December 
 
 57 
 
 + -18 
 
 2-65 
 
 Totals 
 
 20-95 
 
 -T22 
 
 31-22 
 
 Bureau ; Mr. Lynam T. Briggs, U.S. Department of Agriculture ; 
 and Mr. Edgar Buckingham, Bureau of Standards. The instru- 
 ments in question included a measuring micrometer for differ- 
 ential changes in the level of a water surface, an electrical device 
 for maintaining a fixed surface-level and measuring the cubic 
 contents of evaporated water, an improved Piche evaporimeter, 
 an anemometer transformed for reading wind velocities in kilo- 
 metres per hour instead of miles per hour. Evaporating pans 
 and auxiliary contrivances, such as towers for the equipotential 
 surfaces, tubes for the Stefan formula, and so on, were constructed 
 at Reno. 
 
174 METEOROLOGY 
 
 For details as to the results of the experiments reference must 
 be made to Professor Frank H. Bigelow's reports in the Monthly 
 Weather Review of the U.S. Weather Bureau for July, 1907, 
 February, 1908, and the " Annual Summary " of the same pub- 
 lication for 1908. In the Monthly Weather Review, April and May, 
 1909, also, Professor C. F. Marvin describes the methods and 
 apparatus which were employed, and which should be employed 
 in investigations on evaporation. His monograph is a model of 
 scientific writing, and is fully illustrated. 
 
CHAPTER XV 
 
 L THE ATMOSPHERE OF AQUEOUS VAPOUR (continue!) 
 
 HYGROMETRY. 
 
 HYGROMETRY is on a more satisfactory basis than atmidometry. 
 Hygrometers (Greek, 17/005, moist; perpov, a measure) are of 
 two kinds direct and indirect, and the latter class is further 
 subdivided into organic and inorganic organic hygrometers being 
 those which depend for their indications on the effects produced 
 on such organic substances as wool, twine, hair, and seaweed, 
 by the varying humidity of the atmosphere. 
 
 All direct hygrometers experimentally illustrate the theory 
 or principle of the deiv-point that critical temperature at which 
 dew begins to be deposited. We have seen that the capacity 
 of the atmosphere for taking up and holding aqueous vapour 
 in suspension varies with the temperature ; in other words, 
 with the elastic force or tension of aqueous vapour. If tempera- 
 ture falls, and with it the tension of vapour, a point is at last 
 reached at which the air is saturated with moisture. Should the 
 chilling process be continued, a deposition of dew takes place 
 the temperature has fallen below the dew-point. Now, in a 
 direct hygrometer the cooling process is continued until a film 
 of condensed moisture, or " dew/' develops on a surface of polished 
 metal or of glass. At this moment an attached thermometer 
 is read off, giving the temperature of the dew-point. 
 
 Three direct hygrometers call for description Daniell's, 
 Regnault's, and Dines's. 
 
 Professor Daniell, F.R.S., in 1820 described the instrument 
 which bears his name (Fig. 46). It consists of a glass tube, bent 
 twice at a right angle, and terminating at each end in a glass ball 
 
 175 
 
176 
 
 METEOKOLOGY 
 
 or bulb. One of these bulbs is blackened, the other is covered 
 with a jacket of fine linen or muslin. The black bulb is partly 
 filled with pure ether, and encloses the bulb of a delicate ther- 
 mometer, which just touches the surface of the ether. The 
 bent tube and the other bulb are filled with ether vapour, all the 
 air having been carefully removed. When 
 it is desired to find the dew-point, a little 
 ether is allowed to drop on the muslin or 
 linen jacket of the other bulb. It volati- 
 lises quickly, and in so doing makes a large 
 quantity of heat latent. In consequence 
 of this, the ether vapour in the covered 
 bulb condenses, and owing to reduced 
 pressure the ether in the black bulb begins 
 to evaporate. This process, in its turn, 
 chills the black bulb so that a ring of dew 
 begins to form upon its exterior. At this 
 instant the contained thermometer is read 
 off, and the dew-point temperature is 
 
 ascertained. 
 Regnault's direct 
 
 hygrometer (Fig. 
 
 47) is a modifica- 
 tion of Daniell's. 
 
 In it there are two 
 
 thermometers : one 
 
 shows the tempera- 
 ture of the air ; the 
 
 other dips through 
 
 a stopper into a 
 
 small vessel or 
 
 thimble of polished 
 
 silver, and is ex- 
 
 posed during an 
 experiment to the influence of a current of air bubbling through 
 ether contained in the silver vessel. The observer creates the cur- 
 rent of air by opening the tap of an aspirator or jar containing 
 rather less than a gallon of water. As the water flows out of this 
 
 FIG. 46. DANIELL'S 
 HYGROMETER. 
 
 FIG. 47. REGNAULT'S 
 HYGROMETER. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 177 
 
 jar, it draws or aspirates air through a flexible tube connected 
 by air-tight fittings with the silver thimble. To supply the place 
 of the air thus drawn off, a fresh supply bubbles through the ether, 
 drawn from the outer air through a small silver tube which is 
 carried almost to the bottom of the silver thimble. As the air 
 bubbles through the ether it causes it to volatilise, and in this 
 way the temperature is so much reduced that dew is at last 
 deposited on the outside of the polished silver vessel. The tempera- 
 ture indicated at this instant by the 
 contained thermometer is that of the 
 dew-point. In a modified form of the 
 instrument, the observers blow gently 
 into the silver thimble, and the waste 
 
 V r '-, 
 
 B, 
 
 FIG. 43. DINES'S HYGROMETER. 
 
 FIG 49. VERTICAL VIEW OF 
 DINES'S HYGROMETER. 
 
 air is carried off from the silver bottle through a hollow bent tube, 
 conducted into a hollow telescopic stand, which supports the whole 
 apparatus. 
 
 The hygrometer (Figs. 48 and 49), designed by the late Mr. 
 George Dines, F.M.S., is of simple construction, " consisting," 
 says Dr. R. H. Scott, 1 " of a vase, A, fitted with a pipe at the 
 bottom, which is conducted close under a plate of black glass, 
 where it also envelops the bulb of a thermometer, C ; a cock, B, 
 is fitted at the base of the vase. Very cold water, or ice and water, 
 is put into the vase, and the cock is opened ; the glass speedily 
 
 1 Elementary Meteorology, p. 104. 
 
 12 
 
178 METEOROLOGY 
 
 becomes dulled, and the thermometer is read. The cock is then 
 closed again, the water in the tube soon rises in temperature, 
 and the cloud disappears, the moment of its disappearance 
 being that when the dew-point is again reached. The operation 
 may be repeated as long as the water in the vase remains at a 
 temperature below the dew-point." 
 
 An Electrical Dew-point Hygrometer. At the Southport Meeting 
 of the British Association for the Advancement of Science in 1903, 
 Professor F. T. Trouton, F.R.S., exhibited a modified Dines's 
 hygrometer under this title. The moment of deposition of 
 moisture on a hygrometer of the Dines's type is observed from 
 the completion of an electric circuit effected by the deposited 
 moisture. Two long parallel wires are affixed to the surface of 
 deposition. These wires form the electrodes of a circuit containing 
 a battery and indicating instrument. While the circuit is dry 
 there is insulation, but on dew forming, the current can pass 
 between the wires. The apparatus can be adapted for use with 
 an automatic recording instrument for giving a record of the dew- 
 point at frequent intervals. It is also of use in positions where 
 the moment of deposition of dew cannot be observed by the eye. 
 
 Indirect Hygrometers. Many fibrous organic bodies tend to 
 alter their molecular arrangement, or their appearance, when 
 exposed to damp. In the hair hygrometer of de Saussure advan- 
 tage is taken of this tendency. A hair elongates when damp, 
 and contracts when dry. In de Saussure's hygrometer a healthy 
 human hair, freed from grease by careful boiling in an alkaline 
 fluid, fixed at one end, is turned round a pulley, and supports a 
 light weight. Connected with the pulley is an index hand or 
 needle, which moves over a graduated scale, thus roughly showing 
 the percentage humidity of the air, or the hygrometric state or 
 degree of saturation of the air. Strictly speaking, de Saussure's 
 hygrometer is merely a hygroscope (Greek, vypos, moist ; o-KOTrew, 
 / look at), or an instrument which shows whether the air is moist 
 or dry, without measuring the amount of moisture. 
 
 Among inorganic indirect hygrometers, mention should be 
 made, in the first place, of two chemical hygrometers : one, 
 a scientific toy ; the other, an exact method of chemical analysis. 
 The former is the toy ballet-dancer, a French invention, in which 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 179 
 
 a change of colour of the dress from pink to red occurs when the 
 weather becomes damp. The dress is stained with a solution 
 of the nitrate or chloride of cobalt, which salts are hygroscopic 
 in the way described. In the chemical hygrometer a known 
 volume of air is made to pass by aspiration through weighed tubes 
 packed with chloride of calcium, which has a singular affinity 
 for moisture, and so desiccates the aspirated air. After the 
 experiment the tubes are again carefully weighed, and the 
 increased weight represents the amount of watery vapour present 
 in the given volume of air. 
 
 Professor F. T. Trouton showed at Southport, in 1903, a 
 gravimetric recording hygrometer. The principle on which the 
 working of this instrument depends is that the weight of moisture 
 condensed by bodies such as flannel is, within the meteorological 
 range of temperature, approximately a function of the hygro- 
 metric state alone. Thus, when the moisture in the air varies, 
 or the temperature changes, the weight absorbed by a piece of 
 flannel also changes ; not, however, in proportion to the amount 
 of moisture present, but in proportion to the hygrometric state. 
 This alteration in weight is shown by the movement of the arm 
 of a balance from which the flannel is suspended, and is recorded 
 by means of an inked stilus, on graduated paper, revolving with 
 a clock-driven drum. 
 
 The Aquameter. l NLT. William B. Newton, Ph.D., F.C.S., 
 F.I.C., described to the Royal Meteorological Society, on Novem- 
 ber 15, 1905, a quick method of determining exactly the amount 
 of moisture in the air by means of an instrument to which he 
 has given this barbarous name. 
 
 The instrument consists simply of a mercury reservoir con- 
 nected by a rubber tube to a measured glass tube with two taps. 
 While the top tap is open, by raising the reservoir the mercury 
 is put up to the mark 100. On lowering the reservoir till the 
 mercury in the measured tube drops to mark 0, 100 measures of 
 air are drawn in (in the case of the present instrument 100 cubic 
 centimetres). The top tap is then closed. 
 
 The tap in the side tube at the top of the measuring vessel 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. xxxii., p. 11, 
 1906. 
 
 122 
 
180 METEOROLOGY 
 
 is then opened, connecting a small glass bulb containing porous 
 granular phosphoric anhydride. This absorbs the aqueous 
 vapour of the 100 measures of air. On then bringing the mercury 
 in the reservoir and that in the measured tube to the same 
 level (thus giving the air in the measured tube the atmospheric 
 pressure at which it came in), the rise of the mercury above 
 in the measured tube is the exact percentage by volume of the 
 aqueous vapour previously in the air. With an additional 
 tap and bulb containing solid caustic potash the carbonic acid 
 gas in the air can also be determined in the same manner. The 
 additional reading of the scale to that shown after the aqueous 
 vapour is absorbed gives the carbonic acid in parts per ten 
 thousand. 
 
 Before commencing the experiment the top tap is closed, and 
 the taps to the two bulbs are opened in order to remove the aqueous 
 vapour and carbonic acid in the air of the vessel above the 
 mark 100. 
 
 To cause rapid absorption of the aqueous vapour during the 
 experiment, the reservoir of mercury is raised so as to press as 
 much air as possible into the absorbing flask containing the 
 phosphoric anhydride. 
 
 The absorption then takes place in about five minutes, but 
 it is safer to give the instrument ten minutes so as to be certain 
 of complete absorption before taking the reading. 
 
 As the time taken by the experiment is short, there is no need 
 to make allowance for change of temperature. Practically there 
 is no change of temperature in so short a time, unless the position 
 of the instrument is moved. 
 
 The indicator for showing when the surfaces of the two columns 
 of mercury are level is a horizontal brass bar, which is moved 
 up or down close to the mercury tubes. This horizontal bar 
 slides on two brass uprights. 
 
 When the saturation point is passed, for instance in misty 
 weather, the instrument does not register the total moisture. 
 It does not measure any moisture existing as liquid drops, but it 
 records accurately at all atmospheric temperatures so long as the 
 moisture in the air exists as aqueous vapour. 
 
 In 1792 Hutton observed that a thermometer read lower if 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 181 
 
 its bulb was wet. But it is to Sir John Leslie of Edinburgh, 
 and to Mason of London, that we really owe the psychrometer 
 (Greek, j/^xpos, cold or chill ; ptrpov, a measure], or the dry 
 and wet bulb hygrometer. This apparatus, which is now (often 
 under the name of " Mason's Hygrometer ") everywhere employed 
 in hygrometrical observations, consists of two carefully graduated 
 thermometers, placed side by side at a distance of some 4 inches 
 (Fig. 50). One of these thermome- 
 ters marks the air temperature, 
 and is called the " Dry Bulb." 
 Round the bulb of the other a 
 muslin cap is lightly tied, and this 
 is kept moist with water drawn 
 from a small reservoir by means 
 of capillary attraction through a 
 few strands of loosely twisted lamp- 
 wick. As the moisture evaporates 
 from this muslin cap, heat is 
 made latent, and the temperature 
 of the wet-bulb thermometer is 
 depressed in proportion to the 
 amount and rapidity of evaporation. 
 From the respective readings of 
 the dry- and wet-bulb thermom- 
 eters many valuable deductions may 
 be made : for example, the dew- 
 point, the tension or elastic force 
 of vapour (or the amount of baro- 
 metric pressure due to the vapour 
 in the air), the relative humidity, 
 
 the Weight of Vapour in a Cubic FlG ' *>-- MASON>s HYGROMETER. 
 
 foot of air, the amount of vapour required to saturate the air, 
 the weight of a cubic foot of air in grains at the prevailing 
 atmospheric pressure. 
 
 At Vienna, in 1873, the International Meteorological Con- 
 ference concluded that the psychrometer (wet and dry-bulb 
 thermometer) cannot be replaced by any other instrument? 
 though its defects are not to be denied. 
 
182 METEOROLOGY 
 
 Many years ago, August in Germany, and Professor James 
 Apjohn, M.D., of Trinity College, Dublin, independently investi- 
 gated a method of determining, by calculation, the maximal 
 vapour tension for the dew-point from the temperatures of the 
 dry and wet-bulb thermometers. August's researches will be 
 found in Poggendorff's Annalen for 1825 and 1828. In 1834 and 
 1835 Dr. Apjohn laid his investigations before the Royal Irish 
 Academy. They are published in the Philosophical Magazine 
 for 1835, and in the Trans. R.I.A. for 1837 (vol. xvii., p. 277). 
 
 The physical principle assumed by both investigators is pre- 
 cisely the same. Dr. Apjohn states it as follows : 
 
 " When in the moist-bulb hygrometer the stationary tempera- 
 ture is attained, the caloric which vaporises the water is neces- 
 sarily exactly equal to that which the air imparts in descending 
 from the temperature of the atmosphere to that of the moistened 
 bulb ; and the air which has undergone this reduction becomes 
 saturated with moisture." 
 
 Let / = tension of aqueous vapour at the dew-point tem- 
 
 perature which we desire to know. 
 /' = tension of vapour at the temperature of evaporation, 
 
 as shown by the wet-bulb thermometer. 
 
 The values of / and /' for every degree of temperature from 
 F. to 95 F. are known from experiments carefully performed 
 by M. Regnault. 
 
 Further, let a = the specific heat of air. 
 
 e = the latent heat of aqueous vapour. 
 
 (t - 1') or d = the difference between the reading of the 
 dry-bulb thermometer and that of the wet 
 bulb. 
 
 p = the pressure of the air in inches ; then Dr. 
 Ap John's formula is 
 
 - 
 e ' 30 
 
 or, with the coefficient 
 
 This formula, for wet-bulb temperatures above 32, works 
 t thus :/=/'- * ~y 
 substituted for (p - /'). 
 
 out thus :/=/'- * ~ ' x , h being the height of the barometer, 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 183 
 
 V f 
 
 The fraction ^~ usually does not differ much from unity 
 
 at stations near the sea-level (R. H. Scott). 
 
 Consequently, the formula is abbreviated into 
 
 /=/'-'-/, or /=/'-*. 
 For temperatures of the wet bulb below 32, the value is 
 
 Mr. James Glaisher, F.R.S., by instituting a series of com- 
 parisons between synchronous observations of the dry and wet 
 bulb thermometers and of Daniell's hygrometer, made at Green- 
 wich from 1841 to 1854, and also at high temperatures in India, 
 and at low and medium temperatures at Toronto, constructed 
 special tables, based on a series of numbers called the Greenwich 
 Factors, by means of which the dew-point and other hygro- 
 metrical results may be ascertained by inspection. Glaisher's 
 Hygrometrical Tables, as they are called, have passed through 
 many editions, and are now almost universally employed by 
 practical meteorologists for the purpose of deducing the dew-point, 
 vapour tension, and relative humidity (saturation = 100), from 
 observations of the dry- and wet-bulb thermometers. A copy 
 of these useful Hygrometrical Tables is supplied to each ob- 
 server by the Meteorological Committee, constituted in 1905, 
 and a further simplification of them by Mr. William Marriott, 
 F.R.Met.Soc., will be found in Hints to Meteorological Observers, 
 prepared by him under the direction of the Council of the Royal 
 Meteorological Society. 
 
 Although these " shorts cuts " exist, it may be interesting to 
 explain the use of Glaisher's factors by which the dew-point is 
 found. 
 
 From the dry-bulb reading subtract the wet-bulb reading, 
 multiply the difference by the factor corresponding to the dry- 
 bulb reading, and subtract the product from the dry-bulb reading. 
 The result is the dew-point. 
 
 For example 
 
 Dry bulb =53. 
 Wet bulb =49. 
 Then, 
 
 53 -49 = 4x 2-00 (the factor corresponding to 53) = 8, 
 53 _ 8 =45= the dew-point temperature sought for. 
 
184 METEOROLOGY 
 
 A table of Glaisher's Factors will be found in Appendix II. 
 
 To ascertain the dew-point is of practical importance from 
 a health standpoint as well as in agriculture and horticulture. 
 As Dr. Buchan observes : l "It indicates the point near which 
 the descent of the temperature of the air during the night will 
 be arrested/' " Thus, then," he adds, " the dew-point deter- 
 mines the minimum temperature of the night/' The moment 
 the dew-point is reached, dew is deposited and latent heat is 
 given out, causing temperature to rise. After a time, the air 
 is by radiation again cooled down to the dew-point, when the 
 same process is repeated through the night the air temperature 
 gently oscillating round the dew-point, so long as the sky is clear 
 and the air tolerably calm. 
 
 Having once found the dew-point, we can determine the 
 percentage of saturation or the relative humidity, provided we 
 have before us a table of the tension or elastic force of aqueous 
 vapour at ordinary temperatures. 
 
 Assuming the dry-bulb temperature as above to be 53, and 
 that of the dew-point to be 45, we enter Table II. (Appendix II.) 
 and extract the tension at 53 namely, -403 inch, as well as that 
 at 45, namely, '299 inch. If the air were saturated with 
 moisture, the tension would be '403 inch, but it is only -299 inch. 
 From these facts the percentage of saturation that is, the relative 
 humidity is easily calculated by simple proportion 100 being 
 taken to represent saturation and absolutely dry air 
 
 403 : -299 : : 100 : x =74-2 per cent. 
 
 If from the tension of aqueous vapour at the dry-bulb tempera- 
 ture we subtract that at the dew-point, we obtain the force of 
 evaporation. Thus, in the example we have chosen, '403 - -299 
 = -104 inch. 
 
 In order to avoid erroneous deductions from observations 
 made with the dry- and wet-bulb thermometers, it is necessary 
 to keep the wet bulb in working order by frequent douching 
 of the muslin covering and the capillary threads connecting 
 it with the cistern or reservoir. Soft river water, or rain water, 
 
 1 Introductory Text-Book of Meteorology, p. 96. William Blackwood and 
 Sons. 1871. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 185 
 
 or distilled water should be used for this purpose. If spring 
 water or hard lime-waters are used, their calcareous salts are 
 deposited in the meshes of the muslin and the strands of lamp- 
 wick or floss-silk leading to the cistern, capillary attraction is 
 interrupted, the muslin dries, and evaporation ceases. 
 
 In frost, it is a matter of great difficulty to keep the wet bulb 
 acting. In the first place, so long as the water surrounding it 
 is actually freezing, its temperature will remain steadily at 32, 
 although the dry bulb may be some degrees lower. This is 
 brought about by the disengagement of latent heat during the 
 process of freezing. Again, when the wet bulb and its connections 
 are thoroughly frozen, capillary action between the cistern and 
 the bulb will cease, the ice about the bulb will soon evaporate, 
 and the wet bulb will no longer indicate the temperature of 
 evaporation. In such a contingency, the muslin covering of 
 the bulb must be damped with ice-cold water by means of a small 
 wet camel's hair brush about half an hour before the time of 
 observation. In Russia the use of the hair hygrometer has been 
 enjoined in winter, at the instance especially of M. Pernter. 
 
 The " depression " of the wet-bulb thermometer below the 
 dry-bulb reading depends in some measure on the ventilation 
 to which the instruments are exposed. In calm weather the 
 observer may reduce the temperature of the wet-bulb ther- 
 mometer by a degree or upwards by fanning the instrument. 
 In balloon ascents a relative calm is produced by the balloon 
 travelling with the wind. In order to get trustworthy wet-bulb 
 readings during such ascents, Professor Assmann has devised the 
 " ventilated psychrometer." This instrument consists of dry- 
 and wet-bulb thermometers mounted in parallel metal tubes 
 which communicate at their upper ends. A small ventilating 
 fan, driven by clock-work, is placed in the upper end of the tube. 
 By this means an air current of definite velocity can be aspirated 
 past the thermometers whenever readings are required, and in 
 this way comparable results may be obtained. 1 
 
 From experiments carried out in Spitzbergen, M. Ekholm, 
 of Upsala University, deduced the law that -45 C. must be sub- 
 tracted from the reading of the wet-bulb thermometer when it 
 
 1 The Observer's Handbook, Meteorological Office, p. 33. 1908. 
 
186 METEOKOLOGY 
 
 is covered with ice before the figures given in Jelinek's Psychro- 
 meter Tables can be applied. The physical cause of the unduly 
 high reading of the wet bulb is due to the difference between the 
 saturation pressure of water vapour over water and over ice. 
 
 Aqueous vapour is most abundant in the atmosphere near 
 extensive water surfaces. It is very deficient in the centres of 
 continents, and rapidly diminishes in amount as we ascend 
 through the atmosphere. Dr. K. H. Scott says that it has been 
 calculated that one-half the quantity of vapour in the air is 
 contained in the lowest 6,000 feet of the atmosphere, and that 
 the amount contained in the air above 20,000 feet is only one-tenth 
 part of that at the surface of the ground. Hence the burning 
 power of the sun in the arid atmosphere of lofty mountain peaks 
 and slopes. Tyndall 1 well observes that a sheet of vapour acts 
 as a screen to the earth, being in a great measure impervious to 
 heat. When the air is laden with moisture, visible or invisible, 
 the intensity of the sun's rays is controlled by day and terrestrial 
 radiation is checked by night. It is at the same time, perhaps, 
 necessary to explain that moderate heat with a damp atmosphere is 
 singularly oppressive, but this arises from another cause than actual 
 elevation of temperature, and that is interference with evapora- 
 tion. The air being well-nigh saturated, evaporation is checked, 
 and its cooling and beneficial influence is in consequence unfelt. 
 
 The tension or elastic force of aqueous vapour represents the 
 pressure of all the vapour in the air above the place of observa- 
 tion. It is expressed in terms of inches of the barometrical 
 column, and represents the absolute humidity of the atmosphere. 
 It is greatest near the Equator, least near the poles ; greater 
 over the ocean than over dry land, in summer than in winter, 
 by day than by night, at sea-level than in the upper strata of 
 the atmosphere. 
 
 " A Contribution to the History of Hygrometers " was the 
 title of a Presidential Address delivered before the British Meteoro- 
 logical Society on March 16, 1881, by the late Mr. G. J. Symons, 
 F.R.S. It will repay perusal, and is to be found in the Quarterly 
 Journal of the Meteorological Society, July, 1881 (vol. vii., p. 161). 
 
 1 Heat, a Mode of Motion, p. 385. London : Longmans, Green and Co. , 
 1880. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 187 
 
 The dew-point instruments described in the preceding pages 
 are not suited for use in the open air except in calm weather. 
 When the air is still, the layer immediately in contact with the 
 cooled metallic surface may no doubt be in thermal equilibrium 
 with the latter, even though it is surrounded by layers of warmer 
 air, since air is a bad conductor ; but in a fresh breeze the con- 
 stant renewal of the air prevents its attaining the dew-point, 
 unless the instrument is cooled to a considerably lower tempera- 
 ture. On this account these hygrometers, when used in the open 
 air, give results which do not agree with those of the chemical 
 hygrometer, and are even very discordant amongst themselves. 
 The hygrometer invented by M. Crova avoids this defect, and 
 affords very consistent indications. 
 
 Cr ova's Hygrometer. In 1883 M. A. Crova, Professor in the 
 University of Montpellier, invented an instrument, which is 
 described in the late Dr. Thomas Preston's Theory of Heat, 
 the second edition of which was published by Macmillan and Co., 
 London, in 1904, and edited by J. Rogerson Cotter, M.A., Univ. 
 DubL The original description of the instrument will be found 
 in Memoires de I' Academic des Sciences et Lettres de Montpellier 
 (tomex., p. 411, 1883). 1 
 
 The following description is taken from the second edition of 
 Thomas Preston's Theory of Heat (Section 231), and is reproduced 
 here, together with the three accompanying figures, by permission 
 of Messrs. Macmillan and Co., Ltd. : 
 
 " Fig. 51 (see p. 188) gives a general view of the instrument, 
 a section of which is shown in Fig. 52 (see p. 188) ; efgh is a 
 tube of thin brass, nickel-plated inside, and carefully polished. 
 The end ef is closed by a disc of ground glass, which is illumi- 
 nated by daylight or by a lamp, and which is viewed through 
 a lens gh, which closes the other end of the tube. The image 
 of the window ef, seen by reflection in the polished sides of the 
 tube, appears as an annular ring of light ee'ff of three times the 
 diameter of ef (Fig. 53, p. 188). 
 
 " Air can be slowly drawn through the brass tube by com- 
 
 1 Sur VHygrcm'trie. Par M. A. Crova, Professeur a la Faculte des 
 Sciences de Montpellier. Extrait des Memoires de I'Academie des Sciences 
 et Lettres de Montpellier. (Section des Sciences. Tome x , 1883.) Mont- 
 pellier : Boehm et Fils. 1883. 
 
188 
 
 METEOROLOGY 
 
 pressing and slowly releasing the indiarubber ball (Fig. 51), and 
 if the tube is cooled to the dew-point, the deposition of dew 
 is immediately indicated by the darkening of the reflected 
 image of ef. 
 
 " In order to regulate the temperature of the tube, the latter 
 
 FIG. 51. CROVA'S HYGROMETER. Fie. 52. SECTIONAL VIEW OF CROVA'S HYGROMETER 
 
 _ y 
 
 FIG. 53. LENS IN CROVA'S HYGROMETER. 
 
 is surrounded by a brass box abed containing bisulphide of carbon, 
 through which air can be blown from the mouth by means of a 
 rubber tube fitted to the tubulare T. M. Crova prefers carbon 
 bisulphide to ether, because it is more readily obtained pure, and 
 also does not boil in hot weather. Ordinary commercial ether 
 contains water and alcohol, which are left behind when the ether 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 189 
 
 evaporates. But it is possible to attain a lower temperature 
 with ether than with carbon bisulphide. A thermometer gradu- 
 ated in fifths of a degree dips into the liquid, and is in contact 
 with the brass tube. A blackened screen EE' protects the eye 
 from external light ; ii' is a rubber disc insulating the brass box 
 from its stand, through which heat might otherwise be con- 
 ducted. 
 
 " The advantage of this hygrometer is that the whole of an 
 enclosed volume of air is cooled to the temperature of the dew- 
 point, and that it is unaffected by draughts. By attaching a 
 long tube to the opening t, the air experimented on can be drawn 
 from a point out of reach of the influence of the observer or of 
 contamination by the vapour of carbon bisulphide. It can easily 
 be regulated so that the appearance and disappearance of dew 
 are within 01 C. of each other." 
 
CHAPTEE XVI 
 
 THE ATMOSPHERE OF AQUEOUS VAPOUR (continued) 
 HYETOMETRY. 
 
 TAKEN in a practical rather than a strictly etymological sense, 
 the word hyetometry (Greek, iWdc, rain ; /xer/oov, a measure] may 
 be extended in meaning so as to embrace all the different ways 
 in which aqueous vapour is condensed out of the atmosphere 
 and again restored to the earth from which it was originally 
 taken up by evaporation. 
 
 Condensation of the watery vapour of the atmosphere takes 
 place under seven different forms (1) dew ; (2) hoar frost ; 
 (3) mist and fog ; (4) cloud ; (5) rain ; (6) hail ; (7) snow. We 
 shall consider each of these in more or less detail. 
 
 Dew. The origin of dew attracted the attention of physicists 
 towards the close of the eighteenth century, and the names 
 of Pictet of Geneva, Le Koy of Montpellier, Six of Canterbury, 
 and Patrick Wilson of Glasgow, are especially connected with 
 the subject. But it is to Dr. Charles William Wells, a London 
 physician, that we are indebted for a clear explanation of the 
 nature of dew. He published in 1814 his celebrated Essay on Dew, 
 and his theory of dew has ever since been generally accepted. 
 
 Dew is moisture deposited from the atmosphere under a clear 
 sky in the colder hours, and therefore particularly at night. As 
 the sun sinks in the western sky and finally sets, solar radiation, 
 of course, becomes less and finally ceases. But simultaneously, 
 provided the sky is clear and the atmosphere is tolerably dry 
 and calm, terrestrial radiation increases, causing a rapid fall of 
 temperature, so that at last the dew-point is reached. The 
 moment temperature falls below this point, dew is deposited 
 first and chiefly on those substances which are the best and there- 
 fore the most powerful radiators of heat. Such are hair, wool, 
 
 190 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 191 
 
 straw, grass, and herbage in general. There are really two kinds 
 of dew, and in French they are distinguished by separate words 
 serein and rosee. Serein is the falling evening dew, which results 
 from the general chilling of the stratum of air nearest the earth's 
 surface after sundown. Rosee is the dew seen in the morning 
 gathered in drops on the surface of leaves and other cool surfaces, 
 and at the extremities of blades of grass ; it is the morning dew, 
 and its deposition depends on the more rapid radiation from the 
 substances on which it gathers. This causes the moisture of the 
 air which comes into contact with the leaves or grass to be con- 
 densed and to form drops of dew upon their surface. 
 
 This theory of dew has been called in question by at least one 
 eminent physicist. In a lecture on the " Formation of Dew," 
 delivered in 1897, Dr. J. G. McPherson, F.R.S.E., Lecturer on 
 Meteorology in the University of St. Andrews, strongly sup- 
 ports the view originally put forward by Mr. John Aitken 
 that what is really dew mostly rises from the ground. It is 
 watery vapour rising from the soil and condensed by ascending 
 into a chilled stratum of air resting upon the surface of the 
 ground on a clear, calm night. According to this view, dew does 
 not fall from the air, but rises from the soil. The glistening 
 " dew-drops " on blades of grass and the leaves of plants are not 
 really dew at all, but the watery juices of the grasses and plants 
 themselves carried to the edges of the blade or leaf by its veins 
 in order to keep up plant circulation. " The large drops seen 
 on plants at night are falsely called dew ; they are produced 
 from the plants themselves as tokens of their active and healthy 
 growth." l Dr. McPherson finds the most practically convincing 
 proof of the rising of dew from the ground in the formation of 
 hoar-frost or frozen dew. 
 
 Hoar frost. In winter, should the dew-point be below 32, hoar 
 frost instead of dew is deposited, even forest trees assuming a thick 
 coating of rime. This is to be distinguished from a phenomenon 
 which occurs at the beginning of a sudden thaw after a severe frost, 
 and which is called " silver thaw " (German, Rauhfrost ; French, 
 givre), a deposition of rough ice-crystals on a surface still below 
 the freezing-point. A smooth ice-coating formed under like 
 1 Symons's Meteorological Magazine, vol. xxxii., p. 62. 1897. 
 
192 METEOROLOGY 
 
 circumstances is called " glazed frost " (German, Glatteis ; French, 
 verglas). When a warm, damp air comes into contact with frozen 
 surfaces, its moisture is condensed and deposited in a solid 
 form like hoar frost or light snow. Should rain fall on such 
 surfaces, it is at once converted into a sheet of ice, which renders 
 city pavements and country roads equally impassable for the 
 time being. Glazed frost sometimes inflicts great injury on trees 
 and plants. Dr. R. H. Scott quotes a remarkable example of 
 the phenomenon which occurred in France in January, 1879, 
 and was described by M. Godefroy in the eighty-ninth volume of 
 the Comptes Rendus of the French Academy (p. 999). During 
 a severe frost it rained heavily, but the rain instantly was con- 
 gealed on contact with still frozen surfaces, so that branches of 
 trees snapped off, and even the trees themselves were felled by the 
 weight of superincumbent ice. A twig of a lime-tree, 4 inches 
 long, weighed 930 grains, but when freed from ice only 7-5 grains. 
 A laurel leaf carried a coating of ice which weighed 1,120 grains. 
 These figures sufficiently explain the destructive action of a 
 glazed frost. 
 
 The formation of dew is interfered with by (1) a high wind ; 
 (2) a very damp atmosphere even with a cloudless sky, because 
 aqueous vapour checks terrestrial radiation ; (3) a cloudy sky, 
 which radiates back the heat cast off by terrestrial radiation ; 
 (4) the proximity of buildings or of lofty trees. 
 
 The most systematic attempt to measure the amount of dew 
 which has been yet -made was by the late Mr. George Dines, 
 F.M.S. 1 His observations numbered 198, and were carried 
 out in 1877 and 1878 near his residence at Walton-on-Thames, 
 on open grass land, at a height of 52 feet above Ordnance Datum. 
 On only three occasions was an amount of dew exceeding 
 010 inch in depth deposited upon his measuring glasses. Taking 
 the average of all his observations, and multiplying the result by 
 365, the annual depth of dew would appear to be 1-397 inches. 
 If the observations on the grass only are taken, the amount is 
 1-022 inches. Mr. Dines considered that it might fairly be assumed 
 that the average annual deposit of dew upon the surface of the 
 earth falls short of 1-5 inches. 
 
 1 Quarterly Journal of the Meteorological Society, vol. v., p. 157. 1879. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 193 
 
 Mist and- Fog. Possibly the simplest way to describe these 
 phenomena is to say that they are really cloud formations in 
 contact with, or suspended just above, the surface of the ground 
 or ocean. Mist is, in the strict sense of the term, visible watery 
 particles suspended in the atmosphere at or near the surface of 
 the earth. In an applied sense it is coarser watery particles 
 assuming the form of tiny rain-drops, and so floating in the air 
 or falling to the ground. Such is the proverbial Scotch Mist. 
 
 Fog differs from cloud only in being near or in contact with 
 the ground, and from mist only in regard to the fineness of the 
 watery particles of which it is made up. It is to be remembered, 
 however, that, in winter especially, dry smoke-fog often forms 
 over large cities, notably over London and Manchester. Nothing 
 is more remarkable than the property of fog to conduct sounds 
 of all kinds to unwonted distances. Fog and mist are, strictly 
 speaking, not aqueous vapour at all, but water itself in minutest 
 particles or droplets. In the formation of fog and mist, aqueous 
 vapour is condensed and latent heat is set free. These two facts 
 establish the nature of fog and mist they are water, not watery 
 vapour. When objects exposed to their influence are moistened, 
 or when .an appreciable amount of water is collected in the rain- 
 gauge during fog and mist, we speak of a " wet fog." 
 
 In a Presidential Address to the Royal Meteorological Society 
 on January 16, 1889, Dr. William Marcet, F.R.S., classifies fogs 
 into sea fogs, lake fogs, river fogs, waterfall or spray clouds, and 
 town fogs. This interesting address is published in the Quarterly 
 Journal of the Society for April, 1889 (vol. xv., p. 59). 
 
 It is to the scientific researches of Mr. John Aitken, F.R.S.E., 
 of Falkirk, N.B., that we are especially indebted for our know- 
 ledge of the formation of fogs, clouds, and rain. It was in the 
 autumn of 1875, when studying the action of " free surfaces " 
 in water when changing from one state to another, that Mr. Aitken 
 first observed the conditions necessary for cloudy condensation. 1 
 By a " free surface " is meant a surface at which water is free 
 to change its condition. For instance, the surface of a piece of ice 
 in water is a " free surface " at which the ice may change to water, 
 
 1 " On Dust, Fogs, and Clouds." By John Aitken, F.R.S.E. Trans. 
 Royal Soc. Edin., vol. xxx., part i., p. 337, 1883. 
 
 13 
 
194 METEOROLOGY 
 
 or the water change to ice. Again, a surface of water bounded 
 by its own vapour is a " free surface/' at which the water may 
 vaporise, or vapour condense. What are called the " freezing- 
 point " and the " boiling-point " of water are the temperatures, 
 C. and 100 C. respectively, at which these changes take place 
 at such " free surfaces." When there is no " free surface " in 
 the water, we have at present no knowledge whatever as to the 
 temperature at which these changes will take place. It is well 
 known that water may be cooled in the absence of " free surfaces " 
 far below the " freezing-point " without becoming solid. Several 
 years ago Mr. Aitken showed reason for believing that ice, in 
 the absence of " free surfaces," could be heated to a temperature 
 above the " freezing-point " without melting. 1 Professor Thomas 
 Carnelley, of Firth College, Sheffield, has shown this to be pos- 
 sible, and states that he has succeeded in raising the temperature 
 of ice to 180 C. 2 Further, Mr. Aitken has shown in the paper 
 above referred to, that if water be deprived of all " free surfaces " 
 it may be heated in metal vessels while under atmospheric pressure 
 to a temperature far above the " boiling-point," when it passes 
 into vapour with explosive violence. 
 
 From the foregoing considerations it is evident that a necessary 
 condition for water changing its state is the presence of a " free 
 surface," or " free surfaces," at which the change can take place. 
 
 Let us now look, with Mr. Aitken, at the process, as it goes 
 on in nature, of water changing from its gaseous or vaporous 
 to its liquid state in other words, to the cloudy condensation 
 of our atmosphere. 
 
 " As the heat of the sun increases," writes Mr. Aitken, 3 " and 
 the temperature of the earth rises, more and more water becomes 
 evaporated from its surface, and passes from its liquid form to 
 its invisible gaseous condition ; and so long as the temperature 
 continues to increase, more and more vapour is added to the 
 air. This increased amount of vapour in hot air compared to 
 cold air is generally explained by saying that hot air dissolves 
 more water than cold air. This, however, is not the case. Air 
 has no solvent action whatever on water vapour. Water vapour 
 
 1 Trans. Royal Scottish Society of Arts, 1874-75. 
 
 2 Nature, vol. xxii., p. 435. 1880. 
 
 3 " Oa Dust, Fogs, and Clouds," Trans. Royal Soc. Edin., p. 337. 
 
THE ATMOSPHEKE OF AQUEOUS VAPOUK 195 
 
 rises into air to the same amount that it would do into a vacuum 
 at the same temperature, only it rises into air more slowly than 
 into a vacuum, and the amount of vapour which can remain in 
 the air is independent of the amount of air present that is, in- 
 dependent of the pressure of the air and depends only on the 
 temperature. 
 
 " After air has become what is called ' saturated ' with vapour, 
 that is, when the vapour tension is that due to the temperature 
 a momentary condition of stability is attained. Suppose the 
 temperature to fall, a change must now take place. All the water 
 cannot remain as invisible vapour ; some of it must condense 
 out into its visible form. It is this condensed water held in 
 mechanical suspension in the air to which we give the names of 
 fog, cloud, mist, and rain phenomena having some resemblance 
 to each other, yet possessing marked differences. The particles 
 composing a fog, for instance, are so fine they scarcely fall through 
 the air, a cloud is a little coarser in the grain, while a mist is 
 coarser still in texture, and rain is any of these while falling, 
 whether it be a wetting mist or a drenching rain. And the 
 question now comes, Why this difference ? Why should the 
 water vapour condense out of the air in one case in particles so 
 minute they seem to have no weight, and remain suspended 
 in the air, while in another case they are large-grained and fall 
 rapidly ?" 
 
 The key to the answer to this question is given by a very 
 simple and beautiful experiment, which I had the good fortune 
 to see Mr. Aitken repeat during the Congress of the British 
 Institute of Public Health in Edinburgh in July, 1893. Two 
 large glass receivers, A and B, are connected with a small boiler 
 by means of pipes. The receiver A is filled with ordinary air 
 the air of the room. The receiver B is also filled with the air of 
 the room, but before entering the receiver the air is passed 
 through a filter of cotton-wool, and all dust is removed from it. 
 The receivers being so prepared, steam is allowed to pass from 
 the boiler into both receivers. As it enters A it is seen to rise 
 in the globe, forming a beautiful white foggy cloud of condensed 
 vapour a cloud so dense that the observer cannot see through it. 
 On the other hand, when the steam is allowed to enter B, not 
 
 13-2 
 
196 METEOROLOGY 
 
 the slightest appearance of cloudiness is observed in this receiver, 
 although it is as full of water as the receiver A, which remains 
 for some time densely packed with fog. The air is " super- 
 saturated " in both receivers, but only in A does the water 
 condense out and form a cloud ; in B it remains in its in- 
 visible but supersaturated vaporous form. The only possible 
 explanation is that the great difference between the appearance 
 of the two receivers is due to the dust in the air. Dusty air that 
 is, ordinary air gives a dense white cloud of condensed vapour. 
 Dustless air gives no fogging whatever. 
 
 The truth is, that molecules of vapour do not unite with 
 each other and form a particle of fog or mist ; but a " free surface " 
 must be present for them to condense upon. The vapour con- 
 denses on the dust suspended in the air, because the dust particles 
 form " free surfaces," at which the condensation can take place 
 at a higher temperature than where they are not present. Where 
 there is abundance of dust there is abundance of " free surfaces/' 
 and the visible condensed vapour forms a dense cloud ; but where 
 no dust particles are present there are no " free surfaces," and 
 the vapour is not condensed into its visible form, but remains 
 in a supersaturated vaporous state until the circulation of the 
 air in the receiver brings it into contact with the " free surfaces " 
 of the sides of the receiver, where it condenses into droplets of 
 water. If the fog in receiver A is allowed to settle, and more 
 steam is blown in, without allowing any dusty air to enter, a 
 fresh fog is formed, and so on many times in succession. It 
 will, however, be noticed that after each condensation the fog 
 becomes less and less dense, but at the same time more coarse- 
 grained and heavier, until at last no visible fog forms, but the 
 condensed vapour will be seen falling as fine rain. Exactly the 
 same thing may be observed if the experiment is varied by cooling 
 " saturated " air by expansion in a large globular glass flask 
 connected with an air-pump. 
 
 These experiments show clearly 
 
 1. That when water vapour condenses in the atmosphere, 
 it always does so on some solid nucleus. 
 
 2. That the dust particles in the air form the nuclei on which 
 it condenses. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 197 
 
 3. 11: it if there were no dust in the air, there would be no 
 fogs, no clouds, no mists, and probably no rain. 
 
 That the air, when no dust is present, is really supersaturated 
 in these experiments is evident from the fact that when the dust 
 particles become few, the fog particles are not only few, but are 
 much heavier than when they are numerous ; and also from the 
 fact that they increase in size as they fall through the air. Each 
 falling particle becomes a " free surface," at which the super- 
 saturated vapour can condense and increase the size of the 
 drop. 
 
 Mr. Aitken draws a graphic picture of what would occur in 
 Nature if there were no dust in the atmosphere. " When the 
 air got into the condition in which rain falls that is, burdened 
 with supersaturated vapour it would convert everything on the 
 surface of the earth into a condenser, on which it would deposit 
 itself. Every blade of grass and every branch of tree would 
 drip with moisture deposited by the passing air ; our dresses 
 would become wet and dripping, and umbrellas useless ; but our 
 miseries would not end here. The insides of our houses would 
 become wet ; the walls and every object in the room would run 
 with moisture. We have in this fine dust a most beautiful 
 illustration of how the little things in this world work great 
 effects in virtue of their numbers. The importance of the office 
 and the magnitude of the effects wrought by these less than 
 microscopic dust particles strike one with as great wonder as 
 the great depths and vast areas of rock which, the palaeontologist 
 tells us, is composed of the remains of microscopic animals." 
 
 Atmospheric dust, capable of fog and cloud production, is 
 probably composed of fine salt-dust from the spray of the ocean, 
 meteoric dust, volcanic dust, condensed gases, and combustion 
 dust. Mr. Aitken admits the accuracy of Professor Tyndall's 
 observation that extreme heat causes dust motes to become 
 invisible in the sunbeam, but he disputes the accuracy of the 
 conclusion that the heat has destroyed the motes. According 
 to him, the heat would seem to destroy the light-reflecting power 
 of the dust by breaking up the larger motes into smaller ones 
 and by carbonising, or in some way changing their colour, and 
 so making them less light-reflecting. But that the motes are 
 
198 METEOROLOGY 
 
 not destroyed is evident, because the fog-producing power of 
 the air so superheated is actually increased a fact proved by 
 experiment, and explained on the assumption that the number 
 of the particles is increased by being broken up by the heat. 
 
 Mr. Aitken bursts into poetry in prose when explaining one 
 source of the immense quantities of fine sodic-chloride dust 
 ever floating in the air, and its usefulness in the economy of 
 Nature. He says : " The ocean, which under a tropical sun 
 quietly yields up its waters to be carried away by the passing air, 
 almost looks as if he repented the gift, when tossed and angry 
 under tempestuous winds, as he sends forth his spray, which, dried 
 and disguised as fine dust, becomes his messenger to cause the 
 waters to cease from their vaporous wanderings, descend in 
 fertilising showers, and again return to their liquid home/' 
 
 For testing the amount of dust particles in the air what 
 Milton calls " The gay motes that people the sunbeams " 
 Mr. Aitken has designed several ingenious instruments, such 
 as his dust-counter, his pocket dust-counter, and his koniscopc. 
 It would be foreign to the subject-matter of these pages to 
 describe these instruments, but it may be useful to explain what 
 is meant by a koniscope (Greek, /coVis, dust ; O-KOTT^O, I inspect). 
 In the course of his experiments, Mr. Aitken observed that cer- 
 tain colour phenomena took place in cloudy condensation pro- 
 duced by expansion, and it occurred to him that as the colours 
 so produced varied according to the number of dust particles 
 present in the air experimented on, an instrument might be con- 
 structed by means of which, in a rough-and-ready way, no doubt, 
 the amount of dust in the air might be tested by observing the 
 tints produced in it. The instrument consists of an air-pump 
 and a metal tube with glass ends, called the " test-tube." The 
 capacity of the pump should be from half to three-quarters that 
 of the test-tube. Near one end of the test-tube is a passage by 
 which it communicates with the air-pump, and near the other 
 end is attached a stop-cock for admitting the air to be tested. 
 Pointing the test-tube towards some suitable source of light 
 (preferably daylight), so as to illuminate it from end to end, the 
 stop-cock is closed, and one full stroke of the pump is made, 
 when the resultant colour in the test-tube is at once noted. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 199 
 
 This colour would indicate the number of particles. For instance, 
 if there are few particles, one stroke will make the light in the 
 test-tube first blue, then green, then yellow ; and then a second 
 stroke, blue and green, finishing with yellow. But if there are a 
 great many particles present, one stroke will not give the whole 
 of the first series of colours, but may stop at the blue. 
 
 The number of dust particles in the atmosphere is immense. 
 To take a single instance : Mr. Aitken states that he has found 
 that a cigarette smoker sends 4,000,000,000 particles, more or 
 less, into the air with every puff he makes ! He has numbered 
 the dust particles in the atmosphere at many places in Great 
 Britain and on the Continent, and has come to the following 
 conclusions as to the relation between the amount of dust and 
 meteorological phenomena : l 
 
 1st. The earth's atmosphere is greatly polluted with dust 
 produced by human agency. 
 
 2nd. This dust is carried to considerable elevations by the hot 
 air rising over cities, by the hot and moist air rising from sun- 
 heated areas of the earth's surface, and by winds driving the 
 dusty air up the slopes of hills. 
 
 3rd. The transparency of the air depends on the number of 
 dust particles in it, and also on its humidity. The less the dust 
 the more transparent is the air, and the drier the air the more 
 transparent it is. There is no evidence that humidity alone 
 that is, water in its gaseous condition and apart from dust has 
 any effect on the transparency. 
 
 4th. The dust particles in the atmosphere have vapour con- 
 densed on them, though the air itself may not be saturated. 
 
 5th. The amount of vapour condensed on the dust in un- 
 saturated air depends on the " relative humidity," and also on 
 the " absolute humidity " of the air. The higher the humidity, 
 and the higher the vapour tension, the greater is the amount 
 of moisture held by the dust particles when the air is not satu- 
 rated. 
 
 6th. Haze is generally produced by dust, and if the air be 
 dry, the vapour has but little effect, and the density of the 
 haze depends chiefly on the number of particles present. 
 1 Proc. Royal Soc. Edin., vol. xvii., p. 246. 1890. 
 
200 METEOROLOGY 
 
 7th. None of the tests made of the Mediterranean sea air show 
 it to be very free from dust. 
 
 8th. The amount of dust in the atmosphere of pure country 
 districts varies with the velocity and the direction of the wind 
 fall of wind being accompanied by an increase in dust. Winds 
 blowing from populous districts generally bring dusty air. 
 
 9th. The observations are still too few to afford satisfactory 
 evidence of the relation between the amount of dust in the 
 atmosphere and climate. 
 
 Fog or mist forms in different ways : 
 
 1. On a clear, calm night terrestrial radiation so chills the 
 air near the ground that over a level surface like a plain the 
 aqueous vapour of the atmosphere is, through a height of a few 
 inches or feet, condensed into visible water particles, or mist, 
 which is hence called radiation fog. It is best seen on autumn 
 nights over low-lying, flat fields, very locally distributed, and 
 very evanescent should a breeze spring up. 
 
 2. In winter, and even in summer or autumn, provided the 
 night is clear, and calm, and cool enough, white fogs form rapidly 
 over rivers and lakes, the temperature of which is several degrees 
 above that of the contiguous air. Under these circumstances, 
 the water surface may be seen to steam into the atmosphere. 
 While travelling from Ardrossan to Glasgow between 4 and 
 4-50 a.m. on July 27, 1893, I had an opportunity of observing 
 very perfect examples of radiation fog, and of the fog formed 
 over running water. The morning was calm and clear, and at 
 Ardrossan, on the sea-coast, the thermometer had fallen to 49 
 in the screen after a rather warm summer's day. 
 
 "The damp of the river fog 
 That rises after the sun goes down." 
 
 Or, as Shakespeare has it in King Lear 
 
 " You fen-suck'd fogs, drawn by the powerful sun." 
 
 Cities built on the banks of large rivers are liable to suffer 
 from fogs of this kind, the visitation being intensified by the 
 presence of undue quantities of carbon in the atmosphere, so 
 that the fog is no longer white and pure, as in the open country, 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 201 
 
 but assumes the colour of pea-soup, or becomes so dense and 
 murky as to rival the darkness of an overcast sky at midnight 
 so extraordinary is the light-absorbing power of a city-born, 
 smoke-begrimed fog. 
 
 Mr. Aitken asks the pertinent question, Why should the smoke 
 which usually rises and is carried away by the winds fall to the 
 ground when we have fogs ? He thinks that the conditions 
 which account for the fog also account for the smoke falling. 
 He says : l " When we have fogs the atmosphere is nearly saturated 
 with vapour, and the smoke particles, being good radiators, are 
 soon cooled, and form nuclei on which the vapour condenses. 
 The smoke particles thus become loaded with moisture, which 
 prevents them rising, and by sinking into our streets add their 
 murky thickness to the foggy air. This seems to explain the 
 well-known sign of falling smoke being an indication of coming 
 rain. That the colour or blackness of what is called a pea-soup 
 fog is due to smoke is, I think, evident from the fact that a town 
 fog enters our houses and carries its murky thickness into our 
 rooms, and will not be induced to make itself invisible, however 
 warmly we treat it. It will on no account dissolve into thin air, 
 however warm our rooms, for the simple reason that heat only 
 dissolves the moisture and leaves the smoke, which constitutes 
 a room fog, to settle slowly, and soil and destroy the furniture. 
 If the fog was pure that is to say, was a true fog, and nothing 
 but a fog such as one sees in the country, it would dissolve 
 when heated, as every well-conditioned country fog does at 
 least, I never remember meeting a fog in a country house." 
 
 Somewhat in the character of an optimist, Mr. Aitken puts in 
 a good word for a smoke fog as a deodoriser (carbon), and a 
 disinfector and antiseptic (sulphurous acid). To say the least, 
 it is a nauseous remedy, if a remedy it can be regarded. 
 
 Town fogs, just like the smoke fogs which penetrate our 
 dwelling-houses, are frequently dry. Professor E. Frankland, 
 D.C.L., F.R.S., in some experiments on the influence of coal- 
 smoke on foggy air, found that water, when its surface is covered 
 with a film of coal-tar, evaporates much less readily on that 
 
 1 " Dust, Fogs, and Clouds," Trans. Royal Soc. Edin., vol. xxx., part i., 
 p. 353. 
 
202 METEOROLOGY 
 
 account. He suggests that this physical fact affords an explana- 
 tion of the formation of dry town fogs. 1 
 
 3. In winter, when an anticyclone, with its accompanying 
 cold and frost, disperses and gives place to cyclonic conditions 
 and a warm, moist, Equatorial air-current, a fog forms. This 
 is due to the sudden chill of the warm moist air by its impact 
 against the cold surfaces of the ground, trees, and buildings in 
 the locality. 
 
 4. In spring, when in quiet, bright, anticyclonic weather the 
 temperature of the air rises quickly over large islands like Great 
 Britain and Ireland, the adjoining seas, still cold from the pre- 
 ceding winter, condense the vapour of the air into dense fogs. 
 These fogs often envelop our east coasts, where they depress the 
 day temperature perhaps 20, or even 25, below that of inland 
 stations. Thus, on April 5, 1893, the thermometer rose to 70 
 at Cambridge, but did not exceed 46 at Yarmouth, where fog 
 prevailed all day. The same phenomenon on a more extensive 
 scale is observed off the banks of Newfoundland, where a polar 
 current of cold water meets the warm air of the mainland of 
 North America, and dense fogs are the result. 
 
 5. Large icebergs are nearly constantly surrounded by fog, 
 which they have generated by chilling the surrounding warm moist 
 atmosphere. 
 
 6. Promontories, jutting into the sea, and mountains are very 
 apt to generate fog and mist, when they are said to be " cloud- 
 capped." So the poet Longfellow sings in Evangeline : 
 
 "... Aloft on the mountains 
 Sea fogs pitched their tents, and mists from the mighty Atlantic." 
 
 Warm, moist air is forced upwards along their sides to an 
 elevation where saturation is at last reached owing to the lower 
 temperature, and in this way the mist or cloud is produced. 
 Mountain peaks sometimes seem to " smoke " or " steam/' owing 
 to the continuous formation of a column of mist or cloud, which 
 spreads out laterally or horizontally to the leeward of the peak. 
 At times also a cloud or mist seems to be motionless at the top 
 of a mountain, or suspended just above it. In reality, the mass 
 of vapour forming the cloud or mist is moving with the wind, 
 1 " On Dry Fogs," Proc. Royal Soc. Ed-tn., J878. 
 

THE ATMOSPHERE OF AQUEOUS VAPOUR 203 
 
 but the watery particles of which it is composed do not condense 
 until they reach a certain point where the cold mountain-top 
 reduces temperature below the dew-point, while at the other 
 extremity they evaporate and become invisible. The accom- 
 panying plate, which is a photograph of the Matterhorn, taken 
 at Riffel, at an elevation of 8,430 feet above sea-level, by Mr. 
 Greenwood Pirn, M.A., shows this smoke or steam cloud very well. 
 
 In this connection mention may be made of haze, which is 
 often observed in anticyclonic, fine, dry weather, particularly 
 during the prevalence of easterly wind in spring. Haze more or 
 less impedes vision, shutting out from view distant mountain 
 landscapes on shore and at sea, causing the horizon to disappear, 
 and the sea and sky to merge into one grey plane. The atmo- 
 sphere is dry but dense during the prevalence of haze. It is, 
 no doubt, partly caused by the presence of dust and smoke in 
 excess in the air, and probably also by partial condensation of 
 the aqueous vapour by the cold polar air-current which it so 
 often accompanies. At the same time, it must be acknowledged 
 that haze and intense heat are often observed together. A 
 peculiar obscuration of the atmosphere which sometimes appears 
 in summer is called " dust haze " (in German Hohenrauch). It 
 is more common on the Continent than in the British Islands. 
 Its origin is not quite understood, but at times it has been traced 
 to extensive fires on the moors or in the forests of Northern 
 Europe. To its formation dust certainly contributes. 
 
 The optical phenomenon known as the " Spectre of the Brocken " 
 (das BrocJcengespenst) is nothing more than the magnified shadow 
 of the observer, or of some other object, thrown by the setting 
 sun upon the mist or thin cloud stratum which so often veils 
 the summit or slopes of the Brocken (Mons Bructerus, Melibocus 
 of the Romans), the culminating point of the Harz Mountains, 
 in Saxony. The mountain has an elevation of 3,740 feet above 
 sea-level. But the phenomenon may be seen from a much smaller 
 height, as will appear from the following account of the " Spectre 
 of the Brocken," as seen from the Hill of Howth, County Dublin, 
 on September 20, 1907, by Mr. Henry A. Cosgrave, M.A., J.P. 
 The bold foreland of Howth rises to a height of 563 feet above 
 the sea. Mr. Cosgrave wrote to me from Broomfield, Howth, 
 
204 METEOKOLOGY 
 
 under date September 21, 1907 : " As I know what an interest 
 you take in things meteoric, I proceed to tell you of a phenomenon 
 I saw for the first time yesterday. A friend and I were walking 
 on the east cliff about four in the afternoon, when there was a 
 very dense fog seawards. The lower part of the fog was in 
 shadow of the hill, but on the upper part there was brilliant 
 sunshine. We saw our shadows projected on the fog apparently 
 walking upon the part in shadow, and a crown or halo 
 encircled us. We went on some distance, and again saw the 
 same phenomenon. To-day I mentioned the matter to an old 
 fisherman, but he had never heard of it being seen. He said 
 they called the rainbows in the fog 'fog scoffers/ because 
 they intimated that the fog would pass away. I suppose the 
 appearance was the same as the well-known ' Spectre of the 
 Brocken/ ' 
 
 On rare occasions a stratum of haze forms at a great height 
 above the earth, while the lower strata of the atmosphere remain 
 perhaps unusually clear. This happened on Sunday, May 22, 
 1870, over a great part of Ireland. So clear was the lower air 
 that Snowdon was indistinctly seen from the Hill of Howth on 
 the north of Dublin Bay, while a vapour fog or haze, suspended 
 in mid-air, absorbed the blue rays of the solar spectrum, causing 
 the sun to assume a pinkish or carmine tint, and a strange lurid 
 light to spread over the landscape. I described the phenomenon 
 in Symons's Monthly Meteorological Magazine for June, 1870 
 (vol. v., p. 65). 
 
 Measurement of Fog Densities. At a meeting of the Eoyal 
 Meteorological Society held on May 15, 1907, Mr. Joseph W. 
 Lovibond, F.E.Met.Soc., exhibited and described an apparatus 
 which he had designed for measuring fog densities. The method 
 is based on the power of selective absorption resident in suitably 
 coloured glass. When this has been graded into mechanical 
 scales of equivalent colour-value, a beam of white light can be 
 progressively absorbed to extinction, and the luminous value 
 of each successive absorption can be stated in quantitative 
 terms. This analytical power also applies to the colour con- 
 stituents of the beam. 1 
 
 1 Quarterly Journal of the Eoyal Meteorological Society, vol. xxxiii.. 
 p. 275. No. 144. October, 1907. 
 
CHAPTER XVII 
 THE ATMOSPHERE OF AQUEOUS VAPOUR (continued) 
 
 CLOUDS. 
 
 WE pass from fogs to clouds by an easy and rational transition. 
 A cloud is a collection of particles of aqueous vapour condensed 
 into watery particles and floating in the atmosphere at some 
 height above the ground. This height varies from a few hundred 
 feet to several miles, feathery cirri having been observed far 
 above him by Gay-Lussac, in Septembei, 1804, when in a balloon 
 at a height of 23,000 feet, or considerably more than four miles. 
 Tyndall has aptly applied the term " water dust " to the minute 
 particles of water, condensed from aqueous vapour, which go to 
 make up a cloud. 
 
 The " cloud line," or that level below which cloud formations 
 seldom or never take place, varies in different parts of the world. 
 In South America, Dr. E. H. Scott says it is about 9,000 feet ; 
 in the Tyrol it sinks to about 5,000 feet ; and in the British Islands, 
 out of a great number of observed cloud levels, one-third were 
 below 2,500 feet. 
 
 Lamarck, in 1801, first classified clouds, but it is to the dis- 
 tinguished climatologist, Mr. Luke Howard, F.R.S., that we are 
 indebted for a classification of cloud forms which is still in use. 
 In his Essay on the Modifications of Clouds, first published in 
 1803, and re-issued as a third edition by Mr. John Churchill, of 
 London, in 1865, Howard recognised three primary types and 
 four compound types. The primary types are : (1) Cirrus, or 
 " mares' tails " ; (2) stratus, or " ground .fog " ; (3) cumulus, 
 or " wool-pack." The secondary or compound types are : 
 (1) Cirro-stratus, or " sheet-cloud " ; (2) cirro-cumulus, or 
 
 205 
 
206 METEOROLOGY 
 
 " mackerel sky " ; (3) cumulo-stratus, or " shower cloud " ; 
 (4) nimbus, or " rain-cloud/' These cloud-forms arrange them- 
 selves into two groups, when their height is considered namely, 
 upper and lower clouds. 
 
 I. Upper Clouds. There is good reason to believe that the 
 clouds belonging to this class cirrus, cirro-cumulus, and cirro- 
 stratus are usually composed, not of watery particles, but of 
 ice-crystals. The vast height at which these clouds float would 
 suggest this, but they are also the halo-producing clouds, and the 
 phenomena of halos can be explained only by the refraction of 
 light through ice-crystals. Coronse are white or coloured circles 
 round the sun or moon, with a radius of from 6 to 15. Halos 
 are great circles, with a radius of 22 to 46 (22 30' and 45 0'), 
 in which all the colours of the rainbow may be seen. 
 
 1. Cirrus (cir.) (Latin, cirrus, a hair] is the loftiest of all clouds. 
 It consists of delicate wavy sprays, like a wisp of hair, thread 
 fibres, or feathers, often arranged in parallel lines across the 
 sky, these lines apparently converging towards the horizon and 
 diverging near the zenith owing to perspective. Observations 
 of the cirrus cloud are of the first importance in weather fore- 
 casting. There is no doubt that its formation is connected with 
 the overflow, in the upper regions of the atmosphere, of air which 
 has been carried aloft in the front of a cyclonic system which is 
 developing. It generally appears in that quarter of the sky 
 from which the coming disturbance is advancing, but its motion 
 is often quite different from that of the wind or of the lower 
 clouds. Thus the cirrus may be travelling from west, while the 
 lower clouds are coming from south-west or south, and the wind 
 from south or south-east. Most usually detached sprays precede 
 the main body of the cirri, as scouts precede a body of troops in 
 the field or on the march. In consequence of the intimate 
 relation which cirrus bears to atmospheric depressions or cyclonic 
 systems, its appearance in the sky often betokens wind as well 
 as broken, rainy weather. In winter its arrival is one of the 
 earliest signs of a thaw. 
 
 2. Cirro-stratus (cir.-s.) is commonly formed from cirrus by 
 increased condensation, which extends to a lower level in the 
 sky. When bad weather is approaching, a uniform sheet of 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 207 
 
 cirro-stratus overspreads the sky. This " sheet-cloud " has been 
 called Pallium (Latin, a cloak] by Poey. In it solar and lunar 
 halos are apt to form, and we speak of a watery sun or moon 
 when it has begun to interfere with the solar or lunar rays. As 
 it sinks to a lower level, it becomes denser, entirely intercepting 
 sunshine or moonlight, and soon rain begins to fall from it, while 
 scud drifts rapidly before the wind far beneath it. The word 
 scud is applied to fragments of cloud of the stratiform or cumulus 
 type in rapid motion. As regards prognosis of rain, we may say 
 that rain is sure to fall when cir.-s. supervenes on cir. in other 
 words, when the cloud level is descending through the air. On 
 the other hand, when cir.-s. forms while cumuli are in the sky- 
 in fact, by the ascent of the lower clouds the weather will 
 probably take up, for this cloud will then interfere with evapora- 
 tion and sunshine, and so check the formation of cumuli clouds 
 which are very likely to condense into showers. 
 
 3. Cirro-cumulus (cir.-c.) consists of small, well-defined, round, 
 oval, or globular dense, or soft and fleecy masses of cloud at a 
 lower level than cirrus. These white woolly masses resemble 
 a flock of sheep lying down. They have also been compared to 
 the markings on a mackerel, hence the expression " a mackerel 
 sky." Cir.-c. is essentially a fine-weather cloud, and is suspended 
 at a great height in the atmosphere, though not so high as 
 cirrus. 
 
 II. Lower Clouds. These are stratus, cumulus, -cumulo- 
 stratus, and nimbus, and they are usually composed of watery 
 particles that is, condensed vapour except in winter or when 
 they freeze into ice-crystals as they rise into the atmosphere, 
 as those of the cumulus type nearly always do. 
 
 1. Under the name of " Stratus " (str.), Howard described the 
 cloudy formation which spreads over low-lying ground at nightfall, 
 and vanishes as temperature rises in the morning. He also 
 called it "ground fog/' and defined it as "a widely extended, 
 continuous horizontal sheet increasing from below upwards." 
 Dr. Scott 1 says that " stratus is generally a fine-weather cloud, 
 appearing during the evenings and mornings of the brightest 
 days. At times it overspreads the whole sky in the form of a 
 1 Elementary Meteorology, p. 127. 1883. 
 
208 METEOKOLOGY 
 
 low, gloomy, foggy canopy, the atmosphere being more or less 
 foggy under it. All low detached clouds which look like a piece 
 of lifted fog, and are not in any way consolidated into a definite 
 form, are stratus." 
 
 2. " Cumulus " (cum.) is essentially an evaporation cloud, 
 appearing in its prime when evaporation is taking place rapidly, 
 and when strong upborne or ascensional currents are carrying 
 the aqueous vapour rapidly above the line of saturation or of 
 the dew-point. The cloud in consequence appears with a sharply- 
 defined horizontal base, while its upper portions form magnificent 
 globular masses, snow-white in bright sunshine, but elsewhere 
 
 " Rolled in masses dark and swelling 
 As proud to be the thunder's dwelling." 
 
 MOORE : Lalla Rookh. 
 
 Cumulus is seen as either a land cloud or a sea cloud under 
 absolutely contrasted conditions. As a land cloud, it may best 
 be studied in summer or autumn. A rain-bearing depression, 
 we will suppose, has passed away, and the sky has cleared com- 
 pletely at or after nightfall. Terrestrial radiation has full play 
 during the ensuing night, and towards morning the air is damp 
 and very cold for the time of year. The morning breaks without 
 a cloud, but when the sun's power begins to increase, a few soft 
 scud-cumuli begin to fleck the deep blue sky. These clouds 
 rapidly develop in size and density, and rise to a higher and 
 higher plane, as the line of saturation ascends with the rising 
 temperature near the earth's surface. At last the cumuli become 
 piled up into threatening masses, with snow-white, sharply- 
 defined crests. Their summits now begin to spread out in front 
 into a fan-like, cirriform crest, and simultaneously a heavy shower 
 of rain or hail may be observed falling from the base of the cloud 
 mass. Probably a peal of thunder will be heard re-echoing now 
 and again through these nimbi, as they are called. Such a cloud 
 formation as that just described is almost confined to the land. 
 Over the sea the sky will either remain clear or the cumuli will be 
 seen to waste quickly. The reason for this is clear. Over the land 
 there is a strong ascending air-current due to the increasing heat ; 
 over the sea, on the contrary, the air is descending from above 
 to supply the place of the air which has passed in over the land 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 209 
 
 as a sea breeze. These summer cumuli are especially apt to 
 gather round chill mountain-tops. In tropical climates their 
 formation at certain seasons leads to a daily afternoon thunder- 
 storm and torrents of rain (see p. 304). 
 
 Cumulus is not infrequently a sea cloud in high latitudes in 
 winter and during the prevalence of a polar air-current, when there 
 is no cirriform cloud about. Under such conditions sea cumuli 
 form by night. A cold land breeze blows off the coast to supply 
 the place of the warmer air over the sea, which has risen to such 
 a height as to lead to condensation of its aqueous vapour into 
 vast masses of frozen cumulus. On the east coast of Ireland 
 I have often witnessed the development of such night cumuli 
 in winter during the prevalence of a northerly or north-easterly 
 wind. A curious result of their formation is the precipitation at 
 sea and along the coast of heavy showers of cold rain and hail, 
 or sleet and snow, while a few miles inland a clear sky and keen 
 frost may prevail. 
 
 Sometimes cumulus assumes a modified shape. Instead of 
 towering aloft into great snowy masses, it almost covers the sky, 
 spreading out into long cylindrical rolls, between which gleams 
 of sunlight are seen here and there. This modification has been 
 called by the authorities of the Meteorological Office, London, 
 Roll-cumulus. 
 
 Again, but more rarely, the cumulus cloud appears inverted, 
 topsy-turvy, or upside down, its globular or mammillated surface 
 being underneath, while its horizontal layer is uppermost. To 
 this rare appearance (which, however, I have often seen) the 
 name " pocky-cloud " is given in the Orkneys, where it is recog- 
 nised as a sure sign of storm. A more euphonious name for it 
 is " the festooned cloud." It is caused by a warmer and damper 
 stratum of air above, and a colder and drier stratum below 
 inversion temperature. A very faithful illustration of this cloud 
 will be found in Symons's Meteorological Magazine for June, 
 1874 (vol. ix., p. 65). It was seen at Elterwater, near Ambleside, 
 by Mr. Edward Tucker, junior. 
 
 A very interesting feature about cumuli is the peculiar slopes 
 which they commonly assume. As a rule their upper portions 
 travel more quickly than their bases, of which the motion is 
 
 14 
 
210 METEOROLOGY 
 
 retarded by proximity to the earth. Accordingly, the rounded 
 summit of the cloud is seen in advance of the base the cloud 
 appears to be rolling over upon itself, and this really does occur. 
 But the globular head or crown of the cloud is not directly in 
 front it is usually inclined to the right-hand side of the line of 
 advance at an angle of some 45 or so, while the base of the cloud 
 similarly trails behind to the left hand of the line of advance. 
 In fact, the whole cloud seems to slope away from the centre of 
 lowest atmospheric pressure, the situation of which is roughly 
 indicated by that of the " neutral point," or the point of the 
 compass whence the cloud-slope springs. In anticyclonic systems 
 just the reverse sometimes occurs : the base of the cumulus out- 
 strips its apex and the cloud-slope is towards the left hand of 
 the line of advance in front, and towards the right hand behind. 
 In this case the undercurrent travels more rapidly than the upper 
 current, and the set of the air is outward from the area of high 
 pressure, the centre of which lies to the right of the observer, 
 who in every case is supposed to stand with his back to the 
 wind (see p. 4). 
 
 3. " Cumulo-stratus " (cum.-s.) was described by Howard as 
 " the ' cirro-stratus ' blended with the ' cumulus,' and either 
 appearing intermixed with the heaps of the latter, or superadding 
 a widespread structure to its base." R. H. Scott explains that this 
 is the cumulus, as it were, changing into a nimbus, or rain cloud. 
 He adds that it is dark and flat at its base, and is traversed by 
 horizontal lines of dark cloud. 
 
 4. " Nimbus " (nim.). This is the " rain cloud/' according 
 to Howard, who defines it as "a cloud, or system of clouds, from 
 which rain is falling. It is a horizontal sheet above which the 
 ' cirrus ' spreads while the ' cumulus ' enters it laterally and 
 from beneath." As so defined, it is really the " shower cloud," 
 or a mass of cumulus which is being rapidly condensed into 
 rain, hail, or snow. It is a composite cloud, towering from the 
 realms of cumulus into those of cirriform cloud above, and 
 streaked with stratiform cloud beneath, so that Howard called 
 this form of nimbus by the composite name of " cumulo-cirro- 
 stratus." 
 
 But with equal propriety the term nimbus is applied to a diffuse 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 211 
 
 sheet of cirro-stratus from which rain has begun to fall in front 
 of the centre of an area of low pressure (cyclonic system), or, 
 indeed, to any cloud or system of clouds from which rain is actually 
 falling. 
 
 Besides the foregoing classical cloud types, mention should 
 be made of scud, a term which is used to indicate loosely-formed, 
 vapoury, detached clouds driving rapidly before the wind, as 
 the poet Longfellow has it 
 
 " Borne on the scud of the sea." 
 
 In recording observations on clouds, the contractions dr., 
 Cir.-c., Cir.-s., Str., Cum., Cum.-s., Nim., and Scud, should alone 
 be used. The scale for the amount of cloud varies from " blue 
 sky/' or "cloudless/' to 10 "entirely overcast/' The direction 
 from which all clouds are coming should be recorded. Very 
 often cloud direction is far from corresponding with wind direc- 
 tion, and this is especially true of upper clouds. The apparent 
 rate at which clouds move should also be noticed, as well as the 
 radiant points, in the cases of cirrus in particular. 
 
 When thunder threatens, cloud undergoes rapid changes of 
 formation, shape, and density, and nearly always a peculiarly 
 dense cirrus or cirro-stratus is superimposed on massive, lurid 
 cumuli. Whenever it is possible, photographs of thunderclouds 
 and of rare cloud formations in general should be taken. 
 
 In a report 1 issued early in 1894 by the Vatican Observatory 
 (Pubblicazioni della Specola Vaticana, Fasciculus III.), Signor 
 Mannucci, of Rome, gives a brief account of systems of cloud 
 classification. He practically accepts the classification proposed 
 by Abercromby and Hildebrandsson at the International Con- 
 ference held in Munich in 1891, and set forth in the Cloud- Atlas 
 of Hildebrandsson, Koppen, and Neumayer. This classification 
 recognises ten different species, arranged in five principal groups. 
 The first group (A) comprises the highest clouds in our atmo- 
 sphere ; the second group (B) includes clouds at a medium 
 height ; and the third group (C) low clouds. In the fourth 
 group (D) we have clouds in ascending currents ; and, finally, the 
 fifth (E) contains the masses of vapour changing in form. In 
 
 1 See Nature,rFebrua,Ty 8, 1894, p. 341 et seq. 
 
 142 
 
212 METEOROLOGY 
 
 the first four groups the letter (a) is used to distinguish the forms 
 of cloud usually accompanied by fine weather, and (6) for those 
 characteristic of bad weather. The following is the grouping 
 as given by Signer Mannucci : 
 
 GROUP A. 
 
 Clouds from medium altitudes up to an average of 9,000 
 metres (29,528 feet). 
 
 1. Cirrus (a). 
 
 2. Cirro-stratus (6). 
 
 3. Cirro-cumulus. 
 
 GROUP B. 
 
 Clouds having altitudes from 3,000 to 6,000 metres (9,843 to 
 19,686 feet). 
 
 4. Alto-cumulus (a). 
 
 5. Alto-stratus (6). 
 
 GROUP C. 
 
 Clouds the bases of which have altitudes from 1,000 to 2,000 
 metres (3,281 to 6,562 feet). 
 
 6. Strato-cumulus (a). 
 
 7. Nimbus (6). 
 
 GROUP D. 
 
 Clouds on ascending columns of air, with bases about 1,400 
 metres high, and summits from 3,000 to 5,000 metres (9,843 to 
 16,405 feet). 
 
 8. Cumulus (a). 
 
 9. Cumulo-nimbus (6). 
 
 GROUP E. 
 
 Fog-banks up to about 1,500 metres (4,921 feet). 
 10. Stratus. 
 
 The cloud nomenclature adopted by the International Meteoro- 
 logical Committee at Upsala in August, 1894, agrees with the 
 above in all essential particulars. 1 
 
 1 See Quarterly Journal of the Royal Meteorological Society, vol. xxi., 
 p. 16 et seg. No. 93. January, 1895. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 213 
 
 These various cloud forms are exquisitely and artistically 
 portrayed in the Atlas International des Nuages, published by 
 Gauthier-Villars et Fils in Paris in 1896, in accordance with the 
 resolutions of the International Meteorological Committee. The 
 Atlas was prepared under the supervision of MM. H. Hildebrands- 
 son, A. Riggenbach, and L. Teisserenc de Bort, members of the 
 Cloud Sub-Committee of the International Committee. 
 
 Direction and Velocity of Clouds. The direction of motion of 
 clouds is always expressed in terms of the point of the compass 
 from which the clouds are coming. It is best observed by sighting 
 a given cloud against a fixed 
 point, as near the zenith as 
 possible to avoid errors due 
 to perspective. By day the 
 top of a flagstaff, the gable of 
 a house, a tall chimney, or a 
 church-spire may be used as 
 a fixed point. At night, should 
 the cloud canopy be broken, 
 stars near the zenith or the 
 moon, when high in the sky, 
 are suitable fixed points. Ex- 
 perience will enable an ob- 
 server to record a qualitative 
 estimate of the apparent velo- 
 city of clouds by using the 
 adjectives " slow," " fast/' 
 " moderate," etc. 
 
 The Nephoscope. This term is applied to an instrument 
 for observing the direction and rate of motion of clouds. There 
 are two main types of such an instrument : (1) the reflecting 
 nephoscope, (2) the direct vision nephoscope. 
 
 Fineman's nephoscope (Fig. 54) consists of a disc of black glass 
 mounted on a tripod stand, which allows of accurate levelling. A 
 vertical pointer, which can be raised or lowered by a rack-and- 
 pinion movement, is attached to the circumference of the disc in 
 such a way that it can be rotated about the disc. A scale engraved 
 on the edge of the pointer enables us to read off the height of its 
 
 Fia. 54. FINEMAN'S NEPHOSCOPE. 
 
214 METEOKOLOGY 
 
 tip above the glass surface. On this surface three concentric 
 circles are marked, the radii' of the two outer circles being re- 
 spectively twice and three times as great as that of the innermost 
 circle. 
 
 The method of observing is as follows : The observer stations 
 himself in such a position that the image of the cloud in the glass 
 and the central point of the mirror are seen in the same straight 
 line. He then rotates the pointer and adjusts its length until 
 its tip also is brought into this straight line. This done, he 
 moves his head so as to keep the cloud image and the tip of the 
 pointer in coincidence, and notes the radius along which the 
 image appears to travel. This radius marks the direction of 
 cloud drift. A compass needle mounted below the disc enables 
 the observer to identify this direction, but he must bear in mind 
 that in this country the compass needle points about 18 west 
 of true north. 
 
 The angular or, more strictly speaking, the tangential velocity 
 of the cloud may be determined by noting the number of seconds 
 required for the image to travel from the centre of the mirror 
 to the first circle or from one circle to the next. If a be the 
 radius of the inside circle, b be the height of the tip of the pointer 
 above the reflecting surface, and t be the time required for the 
 cloud image to traverse the distance a (both a and b being 
 measured in the same units e.g., millimetres), the value of the 
 tangential velocity as it would appear to an observer at a point on 
 the surface vertically below the cloud is given by the expression 
 Tangential velocity = ". 
 
 Fig. 54 illustrates an adaptation of Fineman's nephoscope, or 
 " cloud-mirror," which Mr. Casella has introduced. As the 
 mirror must be truly horizontal, the instrument is fitted with a 
 circular spirit-level. The earliest form of this cloud-mirror was 
 arranged by J. T. Goddard, and shown in the Great Exhibition 
 of 1851. 
 
 Besson's Comb Nephoscope will serve as an example of a direct 
 vision nephoscope. It consists of a brass rod about 9 feet long, 
 bearing at its upper end a cross-piece 3J feet long, to which a 
 number of equidistant vertical spikes are attached. The rod is 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 215 
 
 mounted in a vertical position by means of a number of rings and 
 clamps screwed into a tall post in such a manner that it can rotate 
 freely. Its height should be adjusted so that a fixed mark on 
 the rod is at the level of the observer's eye. 
 
 When using the apparatus, the observer stations himself in 
 such a position that the cloud selected for observation is seen in 
 the same straight line as the central spike. He then turns the 
 cross-piece until the cloud appears to travel along the line of 
 spikes, while he himself remains motionless. The cross-piece 
 will then be parallel to the line of motion of the cloud, and the 
 direction in which it points can be read off on a graduated circle 
 which is provided for the purpose. The rod may be turned, while 
 the observer stands at some distance away from it, by means of 
 two cords tied to a second shorter cross-piece attached to its 
 lower extremity. 
 
 The tangential velocity may be determined by noting the 
 time taken for the cloud to pass from spike to spike. If a be 
 the distance between the spikes, and b the distance from the 
 upper cross-piece to the marked point on the rod which has been 
 adjusted to the level of the observer's eye, and t the observed 
 time, we have as before 
 
 Tangential velocity = . 
 
 Both a and 6 must be measured in the same units. The differ- 
 ence in level between the cross-piece and the observer's eye should 
 be the same in all experiments, and hence the instrument must 
 be set up on a level site. If slow -moving clouds are being watched, 
 the observer will require a fixed support to steady his head if 
 satisfactory results are to be obtained. He will also need smoked- 
 glass spectacles to protect his eyes. 
 
 METHODS OF STATING THE RESULTS OP NEPHOSCOPE 
 OBSERVATIONS. 
 
 In stating the results of nephoscope observations one cr other 
 of the following methods is generally followed : 
 
 1. All clouds are, somewhat arbitrarily, assumed to be at a 
 
216 METEOROLOGY 
 
 level of 1,000 metres, and the linear velocity, V, is then calculated 
 from the formula 
 
 V= |, x 1000. 
 
 2. The height of the cloud is calculated on the assumption that 
 the linear velocity is 1 metre per second. 
 If H be this height, we have 
 
 H=- metres. 
 a 
 
 The International Meteorological Conference at Munich in 
 1891 recommended that at several stations in each country 
 comparative observations should be instituted of the amount 
 of cloud for the whole sky with an unobstructed horizon, and 
 of zenithal zones of 45 and 60. 
 
 Professor E. G. Hill's " Eeport on Cloud Observation and 
 Measurement in the Plains of the North- Western Provinces of 
 India during the Period December, 1898, to March, 1900," 1 
 is a work of great scientific merit. It was undertaken by Professor 
 Hill at the invitation of the Meteorological Reporter to the 
 Government of the North- Western Provinces of Oudh. 
 
 The measurements were made with a pair of photogrammeters 
 of French construction (ecJiassoux). the standard pattern recom- 
 mended for this work by the International Meteorological Com- 
 mission. The photogrammeters are in reality photographic 
 theodolites, and are fully described in Professor Hill's Report. 
 To measure the cloud velocities in the observations, Professor 
 Hill used the apparatus known as Monsieur C. G. Fineman's 
 nephoscope, which is described above at pp. 213 and 214. 
 
 In the fifteen months from December, 1898, to March, 1900, about 
 900 pairs of plates were exposed in the photogrammeters, and from 
 these nearly 1,000 calculations of heights of clouds have been made. 
 
 The results are of great interest, and form a valuable contribu- 
 tion to the literature of meteorology. 
 
 One instance may suffice. On March 1, 1900, cirrus clouds 
 were observed at a height of 95,577 feet above Allahabad (309 
 feet above sea-level), and were ascertained to be travelling at the 
 stupendous rate of 282*2 miles per hour. 
 
 1 By E. G. Hill, Esq., B.A., Professor of Natural Science, Muir Central 
 College Allahabad. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 217 
 
 On another occasion, November 23, 1899, cirrus was seen to 
 travel at a height of 108,050 feet above Allahabad. 1 
 
 Zenith Nephoscope. In the number of the Annuaire de la Societe 
 Meteor ologique de France for February, 1903, M. Louis Besson, of 
 the Montsouris Meteorological Observatory, gives the following 
 description of a new nephoscope designed by him, and intended 
 
 Fia. 55. BESSON'S ZENITH NEPHOSCOPE. 
 
 to take the place of his comb nephoscope (see p. 214), which is 
 not conveniently adapted for zenith observations : 
 
 " The apparatus consists essentially in a horizontal frame, on 
 which are stretched two right-angled systems of parallel and equi- 
 distant threads forming a square pavement. When the observer 
 places himself underneath such a frame, and looks straight at the 
 
 1 The photogrammeters used in this investigation are similar to those 
 which have been described byM.Hildebrandssonin his Etudes Internationales 
 dcs Nuajcs (Observations et Mesures de la Suede), p. 3 et seq., 1896-97 ; and 
 by P. Jos6 Algue, S.J., in his work, Las Nubes en el Archipi logo Filipino, 
 pp. 40, 41. 
 
218 METEOKOLOGY 
 
 clouds, he can determine their direction by setting the frame so 
 that one of the systems of threads becomes parallel to it ; the 
 other system of threads then appears, as it were, perpendicular 
 to the movement of the clouds, and allows their relative speed 
 to be determined. In reality the observation is not made 
 directly, but by means of a plane mirror placed at an angle un- 
 derneath the frame. This arrangement has a twofold advantage : 
 first, it frees the observer from an inconvenient posture ; in the 
 second place, for a like elevation of the frame, it increases the 
 
 FIG. 56. BESSON'S SPHERICAL MIRROR NEPHOMETER. 
 
 effective length of the instrument for the distance which separates 
 the eye from the mirror. The position of the eye is fixed by 
 means of an eyepiece, which may be protected by smoked glass, 
 if this is deemed necessary " (Fig. 55). 
 
 [Spherical Mirror Nephometer. This instrument (Fig. 56) 
 permits the cloud-percentage (nebulosity) to be measured with- 
 out any fear of an error in the number of the tenths. 
 
 The description of, and the method of using, this new nepho- 
 meter have been given by the inventor, M. L. Besson, in the 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 219 
 
 number of the Annuaire de la Societe Meteorologique de France for 
 September, 1906 (p. 241). The following is an abstract : " The 
 convex mirror is a hemisphere of 30 centimetres in diameter. 
 The celestial hemisphere so formed is seen divided into six parts 
 as follows : Two horizontal circles map out a zone of four-tenths 
 on the horizon, another of four-tenths above the first, and a cap 
 of two-tenths round the zenith. Two vertical great circles, 
 perpendicular to each other, divide each of the two annular 
 zones into four parts ; finally, the zenith-cap is divided into two 
 equal parts by an arc of a vertical great circle, making with the 
 foregoing an angle of 45 degrees. 
 
 " The observer looks along an eyepiece fixed to the pedestal 
 of the mirror. The observer's shadow obstructs only the three 
 squares numbered 8, 9, 10. In order to make an observation, 
 the amount of cloud in each of the seven squares 1 to 7 is noted. 
 Then the instrument is turned through 180, and in this new 
 position the amount of cloud in the squares, 2, 5, and 7 is noted, 
 these now representing the regions of the sky which correspond to 
 the squares 8, 9, and 10 in the first position." 
 
 Both the zenith nephoscope and the mirror nephometer are 
 manufactured by MM. Richard Freres, of Paris. The cost of 
 each is 150 francs (6). 
 
CHAPTER XVIII 
 
 THE ATMOSPHERE OF AQUEOUS VAPOUR (continued) 
 
 HYETOMETRY : PRECIPITATION. 
 
 IN Chapter XVI. it was shown that precipitation in the form 
 of dew or hoar-frost fell short of an average of 1'5 inches 
 per annum at the Earth's surface (G. Dines). 1 It is manifest 
 that this figure represents a very small proportion of the total 
 precipitation, which takes place in the form of rain, hail, or snow. 
 We shall probably be near the mark when we say that the pre- 
 cipitation in the form of dew or hoar-frost is only about one- 
 twentieth of that in the form of rain, hail, and snow. There are 
 districts which are practically rainless ; there are other districts 
 where the rainfall is measured in hundreds of inches. On the 
 Khasi Hills, Assam, some two hundred miles to the north-east- 
 ward of Calcutta, the average downpour is said to be more than 
 500 inches, or about 42 feet. Five-sixths of this astonishing 
 rainfall occurs during the south-west monsoon, when the vapour- 
 laden south and south-west winds are forced up by the hills 
 to an altitude far above the saturation line. At Cherrapunji, 
 situated on the southern verge of the Khasi Hills, just outside 
 the Tropic of Cancer (latitude 25 IT N.), the rainfall in June, 
 1851, amounted to 148 '53 inches, or more than falls in most parts 
 of the British Islands in four years. In 1861 the rainfall is 
 stated to have reached 905*12 inches, of which 336*14 inches 
 are returned for the month of July alone. It is true that these 
 last figures were accepted with reserve by the late Sir John 
 Eliot, 2 and rejected as untrustworthy by Mr. Henry F. Blanford, 
 
 1 See p. 192. 
 
 2 " The Rainfall of Cherrapunji." Quarterly Journal of the British 
 Meteorological Society, vol. viii., p. 41. 1882. 
 
 220 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 221 
 
 F.R.S., F.R.Met.Soc., in a paper on the " Variations of the 
 Rainfall at Cherra Poonjee, in the Khasi Hills, Assam," which 
 he read before the Royal Meteorological Society on April 15, 
 1891. 1 In discussing this paper, however, Mr. Tripp, F.R.Met.Soc., 
 showed that from a comparison of variations in the rainfall 
 at other stations there was nothing improbable in a maximum 
 of over 990 inches, with a mean of 500, at Cherrapunji. 
 
 The Mean Annual Rainfall of the World. The first rainfall 
 map appeared in Dr. Heinrich Berghaus's Physikalischer Atlas, 
 published at Gotha in 1845. Professor Loomis, of Yale University, 
 drew the first isohyets for the world in 1882. 2 Isohyets (Greek, 
 10-05, equal ; and vero?, rain) are lines denoting equal depths 
 of rainfall or precipitation, for the term " rainfall " includes 
 all the forms in which water is deposited on the earth's surface 
 rain, snow, sleet, hail, fog, and dew. Professor Loomis's map 
 was revised in 1887 by Dr. Alexander Buchan, for Sir John 
 Murray's measurements of the world's rainfall, 3 and by Professor 
 Loomis himself in 1889. Two new maps, based on the most 
 recent data, appeared in 1898. One, dealing with the distribution 
 of rainfall over the land, was compiled by Dr. A. J. Herbertson, 
 Ph D., 4 and was reproduced in Plate XVIII. in the magnificent 
 Atlas of Meteorology, which forms the third volume of Dr. J. G. 
 Bartholomew's Physical Atlas, published by Archibald Constable 
 and Co., Westminster, in 1899. The other map was drawn by Dr. A. 
 Supan, and published at Gotha in 1898. 5 The dotted lines show- 
 ing the precipitation over the ocean in Plate XVIII. of the Atlas 
 of Meteorology are taken from Dr. Supan's map, and are based on 
 W. S. Black's discussion on rainfall at sea, 6 supplemented by 
 observations taken on the Novara, Gazelle, and Elisabeth. The 
 most striking feature is the great area of excessive rain (over 2,000 
 millimetres, or 80 inches, per annum) over the Atlantic between 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. xvii., No. 79, 
 p. 146. July, 1891. 
 
 2 American Journal of Science, xxiii. and xxv. New Haven, Conn. 
 1882-83. 
 
 3 Scottish Geographical Magazine, iii. Edinburgh, 1887. 
 
 4 Royal Geographical Society's Extra Publications. London, 1899. 
 
 5 " Die Verteilung des Niederschlags auf der festen Erdoberflache," 
 Petermanns Mitteilungen, Ergdnzungsheft, No. 124. Gotha, 1898. 
 
 6 Journal of the Manchester Geographical Society, vol. xiv., pp. 36-56. 
 Manchester, 1898. 
 
222 METEOROLOGY 
 
 Newfoundland and Ireland, and the western extension of the 
 Sahara conditions (rainfall under 250 millimetres, or 10 inches) 
 to 65 W. This parched area practically does not occur in the 
 South Indian Ocean, and is found in the South Atlantic only as 
 a very narrow tongue running north-west from the Kalahari 
 desert, in the region where the south-east trade winds blow most 
 regularly. In the Indian Ocean, the region between 15 and 
 20 S. has more than 2,000 millimetres (80 inches) of rain ! 
 
 Dr. Herbertson summarises the facts as to land rainfall as 
 follows : 
 
 " The Equatorial regions are wet, and in most places more than 
 1,000 millimetres (40 inches) fall every year. The eastern 
 coasts of both the Old and New World receive a relatively heavy 
 rainfall, especially where they are mountainous ; and this is also 
 true for Africa south of the Equator, when we take Madagascar 
 into account. This rain is heaviest in the belt where the trade or 
 monsoon winds strike the land. Between 20 and 35 from the 
 Equator the rainfall rapidly diminishes, and dry deserts are found 
 west of the region influenced by the trade and monsoon winds. 
 
 " On the polar side of 35 the stormy westerly winds bring 
 rain to the western coasts, which are wetter than the eastern 
 ones, although the latter are also affected by the moving low- 
 pressure areas, which have winds blowing from over the sea in 
 front of their centres. The configuration of the land determines 
 the extent of the influence of these stormy winds. In North and 
 South America, in Scandinavia and Scotland, and in New Zealand 
 the moist west winds strike against great mountains, which deflect 
 the air upwards, cool and condense the water vapour, and yield 
 heavy rains. Beyond the crest of the mountains comparatively 
 little rain falls, except in the summer months. These summer 
 rains reach far inland in the regions on the polar sides of 45. 
 Mountains, wherever they exist, are regions of greater rainfall, 
 as they locally cause ascending currents, and deflect horizontal 
 ones upwards. 
 
 " The coldness of lands within the polar circles prevents heavy 
 rainfall, as comparatively little water can exist in the air as 
 vapour. 
 
 " There is, on the whole, a steady diminution of rainfall, from 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 223 
 
 Equator to Pole, corresponding to the diminution of temperature 
 and of vapour-carrying capacity of the air. Three exceptions 
 should be noted. (1) The coastal lands, where sudden changes 
 of temperature are frequent, and the air is nearly saturated with 
 moisture, are rainy, except where cold currents well up and 
 make a cool area near the coast, as happens near the tropics on 
 the west of the continents. (2) Great temperature changes 
 also occur in mountain lands, and where the air is sufficiently 
 damp, rain is common. (3) The hearts of the continents, far 
 from the source of water vapour in the oceans, and the regions 
 reached by winds blowing out from them over dry land, are very 
 dry." 
 
 Dr. Herbertson says that, "comparing the continental and 
 ocean rainfall, it is seen that the latter is greater than the former 
 in high latitudes, but lower in the trade-wind regions outside 
 the Equatorial rain-belt." He further shows, when discussing 
 the mean monthly distribution of rainfall over the land of the 
 globe, that the rainfall zones move north and south with the sun, 
 and attain their maximal northern and southern positions re- 
 spectively about a month later than it does. 
 
 Rainfall Observations. In a letter to Galileo, dated June 10, 
 1639, B. Castelli, of Perugia, records the earliest authentic 
 measurement of rainfall ; but it was an isolated observation, 
 suggested by an exceptionally heavy downpour of rain, and 
 led to no practical advance. 1 In a " Contribution to the History 
 of Rain Gauges," read before the Royal Meteorological Society 
 on March 18, 1891, 2 Mr. G. J. Symons, F.R.S., tells us that, most 
 curiously, the first rain-gauge designed was not an ordinary one, 
 but a recording gauge. On January 22, 1662, Dr. (afterwards 
 Sir Christopher) Wren showed before the Royal Society his ex- 
 periment of filling a vessel with water, which emptied itself when 
 filled to a certain height. Ten years later a tipping-bucket rain- 
 gauge, on Sir Christopher Wren's plan, was ordered for construc- 
 tion by the Royal Society. 
 
 The earliest published returns of rainfall were made in Paris, 
 
 1 Dr. G. Hellmann, " Die Anfiinge der meteorologischen Beobachtungen 
 und Instruments. " Himmd und Erde, II. Jahrgang, 3 und 4 Hefte. 
 
 2 Quarterly Journal of the Royal Meteorological Society, vol. xvii., 
 No. 79. 
 
224 
 
 METEOROLOGY 
 
 in 1668, by M. Pierre Perrault, who wrote an anonymous work, 
 De I'Origine des Fontaines, and in England by Mr. R. Townley, of 
 Townley, near Burnley, Lancashire, whose observations were 
 begun on January 1, 1677. Three years earlier, in 1674, an 
 unknown observer at Dijon was recording the rainfall, and he after- 
 wards supplied Mariotte, of Paris, with records from that city. The 
 gauge used at Dijon is thus described by Mariotte : " Un vaisseau 
 quarre qui avoit environ deux pieds de diametre, au fond duquel 
 il y avoit un tuyau qui portoit 1'eau de la pluie qui y tomboit 
 dans un vaisseau cylindrique." 
 
 In 1695 Mr. Robert Hooke weighed the rainfall at Gresham 
 College, London. The rain-gauge (Fig. 57) used by him consisted 
 of a large bottle called a " bolt head/' 
 capable of holding more than 2 gallons, 
 and with a neck 20 inches long. Into it 
 was conducted the pipe of a funnel (appa- 
 rently of glass like the bottle) 11*4 inches 
 in diameter. The funnel was steadied by 
 two stays or pack-threads strained by two 
 pins. The glasses that is, the funnel and 
 the bottle were supported in a wooden 
 frame. The collected water was weighed 
 every Monday morning by troy weight. 
 Thus, from August 12, 1695, to the 
 same date in 1696, 131 pounds 7 ounces 
 113 grains of rain fell that is, 29'H 
 inches. 
 
 The Twelfth Annual Exhibition of Instruments held by the 
 Royal Meteorological Society, March 3 to 19, 1891, was devoted 
 to rain-gauges, evaporation-gauges, and like instruments. An 
 official description of the various instruments exhibited will be 
 found in the number of the Quarterly Journal of the Royal Meteoro- 
 logical Society for July, 1891 (vol xvi., No. 79, p. 180). A few 
 of the many instruments only will need description. The largest 
 gauge ever made has been in use at Rothamsted, Harpenden, 
 Herts, since the beginning of 1853. Its receiving area equals 
 ToVu^h f an acre - A coloured drawing of this monster gauge 
 by Lady Lawes, of Rothamsted, was exhibited ; so also was a 
 
 FIG. 57. HOOKE'S RAIN- 
 GAUGE. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 225 
 
 specimen of Colonel Ward's 2-inch gauge, one of the smallest in 
 use. One was more than two thousand times as large as the 
 other, and yet their indications did not differ by anything like 
 5 per cent. The smallest gauge ever used is only 1 inch in 
 diameter, and its readings have been proved by experiments 
 undertaken by Colonel Ward, and continued by the Rev. 
 C. H. Griffith, and the Rev. Fenwick W. Stow to differ from 
 those of a gauge five hundred times its size by less than 2 per 
 cent. 
 
 Theoretically, square gauges are simpler than circular gauges, 
 but in practice the latter are mostly used, because they are not 
 so apt to get out of shape as the former, and the least 
 denting of the rim of a rain- 
 gauge would interfere with 
 its measurement. 
 
 The International 
 Meteorological Congress at 
 Vienna in 1873 suggested 
 for all rain-gauges a circular 
 receiver $ square metre in 
 area (about 14 inches in 
 diameter), and considered 
 that the rim should be 
 formed of a strong turned 
 ring of brass, with bevelled 
 edge. But at the Rome 
 Congress, in 1879, it was 
 agreed that, for stations 
 of the second and third 
 order, rain-gauges 8 or even 4 inches diameter are sufficient. 
 
 1. Meteorological Office Gauge. This gauge, 8 inches in diameter, 
 is in very general use (Fig. 58). It is made of copper, weighs 
 between 8J and 9 pounds, or, with a splayed base, from 10 J to 
 lOf pounds, and has a circular collecting funnel surmounted 
 by a vertical rim, 5J inches in depth, in order to catch snow. 
 On top of this rim or cylinder is a stout brass ring about 1 inch in 
 width, ground to a knife-edge above, so that its circular shape 
 is preserved, and the in-splashing of raindrops is entirely pre- 
 15 
 
 FIG. 58. METEOROLOGICAL OFFICE RAIN-GAUGE. 
 
226 
 
 METEOROLOGY 
 
 vented. The rain, caught in the funnel, flows through a pipe 
 into a large copper can capable of holding 4J inches of rain. 
 From this it is poured into a graduated measure-glass, and its 
 quantity is read off in thousandths, hundredths, and tenths of 
 an inch. A fall of an inch of rain means that a square tray 
 12 inches long and 12 inches wide would be filled up to a height 
 of 1 inch by such a rainfall. 
 
 2. The Snmvdon Gauge (Fig. 59) resembles the foregoing, but 
 has a diameter of only 5 inches. In it as in the Meteorological 
 Office gauge a cylinder rises 4 inches 
 vertically from the edge of the cone of 
 the funnel, constituting what is called a 
 " Snowdon rim." A gauge of this kind 
 in copper is nearly indestructible and 
 independent of frost. The galvanised 
 iron Snowdon gauge is much cheaper, and 
 will last for fifteen or twenty years. 
 
 3. The Mountain Gauge (Fig. 60) is 
 intended for rough mountain work, and 
 for waterworks purposes in wet districts. 
 It is a float gauge, capable of holding 48 
 inches of rain. It is superseded by 
 
 4. The Bradford Gauge (Y\%. 61). This 
 is a modified Snowdon gauge. The funnel 
 is identical in dimensions with that of 
 the Snowdon gauge, the tube being long 
 enough to reach within an inch of the 
 bottom of the inner can. According to 
 the nature of the site for which this 
 
 gauge is intended, the inner can may be made deep enough to 
 contain from 12 to 30 inches of rain, but a total depth of 2 
 feet is usually sufficient. The bottle is dispensed with. 
 
 To check evaporation, a slightly concave sheet of metal should 
 be soldered 1J inches below the mouth of the inner can, with a 
 hole f inches in diameter in the centre, and a crescent-shaped 
 opening at one side to facilitate pouring out the contents. To 
 check the measurement, a graduated dip-rod of cedar, J inch 
 wide and J inch thick, tipped with brass, is dropped vertically 
 
 FIG. 59. SNOWDON RAIN- 
 GAUGE. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 227 
 
 into the inner can after the funnel has been removed. The rod, 
 when withdrawn, shows the approximate amount of rain by the 
 dark colour of the portion which was immersed, and the rain is 
 
 FIG. 60. MOUNTAIN RAIN-GAUGE. 
 
 then measured accurately by means of the measuring-glass, both 
 readings being recorded. 
 
 5. Symons's Storm Rain - Gauge is not intended for general 
 use, or for yielding continuous records, but for enabling observers 
 to record in detail the rate at which heavy rains fall during 
 thunderstorms. With one of these instruments, in London, on 
 
 152 
 
228 
 
 METEOROLOGY 
 
 June 23, 1878, rain was ascertained to be falling for thirty 
 seconds at the rate of 12 inches an hour, or 288 inches a day. 
 The first pattern of this gauge was apt to be broken by frost, and 
 
 Diaphragm 
 Bradford Gauge 
 
 GROUND 
 
 t- 
 
 
 1 
 
 
 
 | 
 
 
 
 | 
 
 
 \ 
 
 
 
 
 \ 
 
 1 
 
 
 \ 
 
 // 
 
 D 
 
 : 
 
 -. -_4m- 
 
 
 "BRADFORD:* 
 
 FIG. 61. BRADFORD RAIN-GAUGE. 
 
 therefore could be put out only in summer time. In a second 
 stronger and more elaborate instrument (Fig. 61) the rain passes 
 into a copper cylinder in which is a float, which rises as the 
 rain falls. The float has a string passing round a pulley, and 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 229 
 
 kept taut by a counterpoise. Therefore, when the float rises, the 
 pulley turns. To the extremity of the axle of the pulley a hand 
 or index is attached, which completes a revolution on a graduated 
 dial when an inch of rain has fallen. Inside the case there is 
 a simple wheel-work whereby another short hand, like the hour- 
 hand of a clock, completes a revolution for 5 inches of rain. 
 With this gauge it is therefore quite easy to read from a window 
 the fall of rain to hundredths of an inch, and by doing this say, 
 every thirty seconds the minutest detail of the fall of rain 
 can be ascertained. This instru- 
 ment is constructed by JSTegretti and 
 Zambra. 
 
 There are several self-registering 
 and recording rain gauges. For ex- 
 ample : 
 
 6. Crosley's registering rain-gauge, 
 of which the area is 100 inches 
 (Fig. 62). Beneath the tube leading 
 from the funnel there is a vibrating 
 divided bucket. When one com- 
 partment of this bucket has received 
 a cubic inch of water that is, when 
 01 inch of rain has fallen the 
 bucket tips, the index advances on 
 the first dial, and the second bucket 
 begins to fill, and so on indefinitely. 
 
 Crosley was a gas-meter maker, and brought out his gauge first 
 in the year 1829. 
 
 7. Messrs. Yeates and Son, of Dublin, have designed a very in- 
 genious modification of Crosley's instrument, by which it is made 
 to record the rainfall electrically. In their electrical self-registering 
 rain-gauge (Fig. 63, p. 230) the funnel is 100 square inches in area, 
 and the measuring bucket (the working parts of which are made of 
 platinum alloy, with agate bearings) is adjusted to turn with 
 1 cubic inch of water. At each turn of the bucket electrical contact 
 is made, which is recorded on a chart. The recording apparatus 
 can be placed in any convenient position indoors. Each 
 
 Fio. 02. CROSLEY'S ELF-REGIS- 
 
 TERINO RAIN-Gf\UGE. 
 
230 
 
 METEOROLOGY 
 
 chart records for one week in a manner similar to the ordinary 
 barograph, and so possesses the advantage of reference at any 
 time. As this apparatus is entirely self-recording, each cubic inch 
 of water, as it is weighed and recorded, is emptied out, so that 
 no error can arise from evaporation. 
 
 8. Casellas Recording Rain-Gauge (Fig. 64). In this pattern the 
 rainfall is measured by means of a new form of tilting bucket, 
 in which provision is made that the water dropping in during 
 the actual tilting of the bucket shall be conveyed to the right 
 compartment of the bucket i.e., to the compartment just coming 
 into position to receive the water. 
 
 FIG. 63. YEATES'S ELECTRICAL SELF-REGISTERING RAIN-GAUGE. 
 
 The recording mechanism consists of an electro-magnet, the 
 armature of which is connected to a cam, by means of which the 
 pen is raised on the revolving drum. This cam is counter- 
 poised so that the force required to lift the pen is the same in all 
 po sitions. 
 
 The chief advantages of this pattern are as follows : 
 
 (1) Rubbing platinum contacts are made use of, which are much 
 more reliable than mercury contacts. 
 
 (2) The instrument is not liable to stick when the pen is near 
 the top of the drum. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 231 
 
 F K , 6-i CASELLA'S RECORDING RAIN-GAUGE. 
 
232 
 
 METEOROLOGY 
 
 (3) Every O01 of an inch is recorded. 
 
 The collecting arrangements are enclosed in a copper or 
 
 FIG. 65. RICHARD'S SELF-RECORDING RAIN-GAUGE (FLOAT PATTERN). 
 
 galvanised iron case, with an 8-inch funnel, and the recording 
 mechanism, which is kept indoors, is usually placed in a polished 
 mahogany case, with glass windows. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 233 
 
 9. MM. Richard Freres, of Paris, have invented a float pattern 
 and a balance pattern self-recording rain-gauge. In the float 
 pattern (Fig. 65) a funnel collects the rain, which is carried by 
 a pipe into a reservoir in which there is a float. A style, carrying 
 a writing pen, follows the motion of the float, rising 4 inches 
 
 FIG. 66. RICHARD'S SELF-RECORDING RAIN-GAUGE (BALANCE PATTERN 
 
 for a rainfall of '4 inch. When the pen reaches the top of a 
 revolving drum, the reservoir empties itself automatically by 
 means of a siphon, the float falls to the bottom, and the pen 
 returns to zero. The siphon is started by an electro-magnet, 
 which, on the circuit of a battery being completed, pulls the float 
 down and causes a sudden rise of the water-level, thereby filling 
 the siphon. In the balance or bucket pattern (Fig. 66) the rain 
 
234 METEOROLOGY 
 
 is led into a tipping bucket divided into two compartments 
 and placed on a balance. One of these compartments being under 
 the funnel, the rain falls into it and causes the balance to descend. 
 A writing pen records this motion on a revolving drum. When 
 the pen has reached the top of the drum ('4 inch of rain), the 
 tipping-bucket reservoir oscillates, and the water filling the 
 first compartment is emptied into a controlling reservoir. This 
 motion causes the second or empty compartment of the bucket 
 
 FIG. 07. NKGRETTI AND Z AM BRA'S HYETOGRAPH (UPPER PART). 
 
 to place itself under the funnel. The filling and emptying of 
 each compartment are alternately and automatically produced, 
 and to each of these double operations a rise and a fall of the 
 writing pen corresponds. 
 
 Negretti and Zambras Hyetograph. For obtaining satisfactory 
 records of the duration and intensity of rainfall it is imperative 
 that the rain-scale on the chart should be an open one. This 
 desideratum may be achieved in various ways, the most obvious 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 235 
 
 of which would appear to be the use of a very high drum (which 
 is a costly matter), or by automatically emptying the collected 
 water at convenient intervals. This latter is usually achieved 
 by the water in the float cylinder being made to siphon away as 
 in the well-known Halliwell's Patent Rain-Gauge, of which 
 
 K 
 
 n 
 
 
 Fios. 68 AND (59. NEORETTI AND Z\MBRA'S HYETOORAPH. 
 
 Messrs. Negretti and Zambra are the makers. An automatic 
 mechanical siphon which can be trusted not to get out of order 
 involves an expense in construction which could not be incurred 
 in a gauge of low price, and for this reason a mechanical device 
 affecting the pen only is substituted in the new instrument styled 
 
236 METEOROLOGY 
 
 the hyetograph, patented in 1908, and marking a new departure 
 in self-recording rain-gauges (Fig. 67). This device enables 
 an open scale of | inch of rainfall to be recorded on a chart 
 measuring about 3 inches in height. 
 
 The hyetograph, as will be seen from the plan (Figs. 68 and 69), 
 is built on a cast-iron plate well protected from rust by a special 
 galvanising process. On the plate and underneath is bolted 
 the copper float chamber C, with its accompanying siphon-tube K. 
 
 As long as rain continues to fall, the copper float D rises, 
 moving in a guide, up to the maximum capacity of 4J inches. On 
 a spindle E, rising from the float D, are a number of projecting 
 pins FF which engage successively with a projection on the lever G, 
 which lever is so pivoted that when the pen reaches the top of the 
 chart, the lever disengages with the pin, and falls by its own weight 
 on to the next lower pin, which is so placed to allow the pen to 
 fall to zero on the chart. The float therefore continually ascends 
 during rainfall, but at each successive J inch of rain the pen 
 descends to zero, and recommences its upward movement. 
 
 As no automatic siphon is used, it is obvious that the rain will 
 collect in the float chamber until it is removed, and the float 
 cylinder is constructed of sufficient size to allow an accumulation 
 of over 4 inches of rainfall, which is the maximum likely to occur 
 in one day in any locality in Great Britain and Ireland 
 (except the wettest parts of the Lake District and on high 
 mountains). 
 
 In order to remove the water the hyetograph is constructed 
 with a specially designed hand-started siphon K, which is 
 actuated when desired, and empties the float chamber of any water 
 which may have accumulated. 
 
 The chart is wound round a clockwork cylinder H, making one 
 revolution in twenty-four hours ; the effective length being 
 10*8 inches, giving 0*45 inch for an hour. 
 
 The whole of the working parts are protected by a stout 
 galvanised iron cover A, hinged at the side, having an observation 
 window, and surmounted by a stout brass rim of 6-inch diameter. 
 
 From the above description it will be seen that the hyetograph 
 has very few parts, is extremely simple to erect, and offers practi- 
 cally no opportunities of going out of order. The moving parts 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 237 
 
 are three only in number, viz., the clock-drum, the float, and the 
 pen lever. 
 
 The hyetograph is constructed especially to conform to the 
 standard instructions for fixing rain-gauges viz., that the funnel 
 should be between 12 and 18 inches above the ground. 
 
 It can be fixed with the base plate on the ground and the float 
 chamber underneath ; in any way, provided a space is allowed 
 for the water to pass rapidly away after siphoning. 
 
 The clock makes one revolution in 24 hours, but need only be 
 wound up once a week. The small dash pot M under the pen 
 lever is filled with the oil supplied, until the piston is just covered 
 when in its highest working position. 
 
 If it is desired to empty the hyetograph when less than J inch 
 of rain has accumulated, about a pint of water may be poured 
 into the float-vessel through the aperture in the cast-iron plate, the 
 siphon being then discharged. 
 
 It is necessary (say once a week, when winding the clock) to 
 lift the float spindle E to its highest limit in order to clear away 
 any soot or grit deposited from the rain water in the cylinder C. 
 Also at least once a month the whole length of the spindle E 
 and pins FF should be carefully wiped with a clean rag just 
 moistened with good sewing-machine oil. 
 
 The case is designed to allow of the circulation of a current of 
 warm air to assist in melting snow as it falls into the funnel. 
 
 A night-light or small spirit-lamp will give the necessary heat. 
 
 The hyetograph, complete with 100 special charts, costs 6 15s. 
 Extra charts, per 100, cost 7s. 6d. 
 
 On May 15, 1907, Dr. Hugh Robert Mill, President of the 
 Royal Meteorological Society, read a paper on " The Best Form 
 of Rain-Gauge." 1 In his opinion the three best patterns are : 
 (1) The Snowdon Rain-Gauge, which he has adopted as the 
 standard for the British Rainfall Organisation, of which he is 
 the very able Director ; (2) the Bradford Rain-Gauge, designed 
 by Sir Alexander Binnie, which is simply a Snowdon gauge of 
 great capacity, made of proportionately stout material very 
 strongly put together, and suitable for monthly readings in wet 
 
 1 Quarterly Journal of the Royal Meteoroloqical Society, vol. xxxiii., 
 No. 144, p. 265. October, 1907. ' 
 
238 METEOROLOGY 
 
 localities, such as mountains or moorlands ; and (3) the Meteoro- 
 logical Office 8-inch Rain-Gauge. 
 
 No matter what rain-gauge is employed, the instrument 
 should be firmly set in a well-exposed position, at least as far 
 in feet from any building, tree, or high wall as the height of 
 that obstacle. " The angle subtended in each azimuth by the 
 nearest obstacle, such as a building or tree, should not exceed 
 30, and the true bearing of the obstacle from the gauge should 
 be carefully measured and noted in the register " (R. H. Scott). 
 The Meteorological Office recommends, in plain language, that 
 the distance between the gauge and the nearest object should 
 be at least twice the height of that object. The gauge should be 
 placed on the ground rather than on a roof, unless in the case 
 of a small town garden, in which it is impossible to obtain a 
 sufficient exposure. The height of the rim of the funnel should 
 be 1 foot above the ground. This should be given in all returns 
 of rainfall, as well as the height of the gauge above mean sea- 
 level. It is essential that the top of the cylinder above the funnel 
 should be absolutely horizontal. The gauges used by the 
 Meteorological Office are now made with a splayed base, which 
 can be firmly embedded in the ground. Secured in this way, the 
 gauge cannot be blown over in a gale or displaced when the funnel 
 is removed for measuring the rainfall. 
 
 Measurement of Rain. At Normal Climatological Stations 
 the rainfall should be measured at 9 a.m. daily, and the amount 
 should be entered to the previous day, for of the twenty-four 
 hours which elapsed since the last measurement, fifteen belonged 
 to the previous day, and only nine to the day on which the 
 measurement is made. The gauge must be examined daily, 
 whether rain has fallen or not, for dew, hoar-frost, or fog may 
 yield an appreciable precipitation. Daily examination also 
 safeguards against errors arising from the accidental or mis- 
 chievous addition of water. If there is no water in the 
 gauge, a line or dash should be inserted in the register. The 
 water is poured from the can in the interior of the gauge 
 into the graduated measure-glass, which is scaled to represent 
 hundredths and tenths of an inch up to half an inch of rainfall 
 (50 inch). If the fall_exceeds half an inch, the measurement 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 239 
 
 must be made in instalments not necessarily of half an inch 
 precisely. For example, we may measure a fall of 1'581 inches 
 thus: -490 +*485 +'496 +'110 inch. In such a case the water 
 first measured should be carefully preserved until the whole 
 rainfall has been registered and the amount written down. It 
 should then be remeasured as a check on the first reading. The 
 measure-glass must be placed on a perfectly level surface before 
 reading, the observer's eye being brought to bear on the surface 
 of the water at a right angle, so as to avoid errors of parallax. 
 Allowance should be made for capillary attraction, by which 
 the water is drawn a little way up the sides of the glass above the 
 general level in the measure. The reading then should be taken 
 at the bottom of the concave meniscus, or curved surface of the 
 water. The reading should be to the nearest hundredth of an 
 inch in all cases, but it may be to three places of decimals, if 
 extreme accuracy is desired. In the former case, falls of less than 
 01 inch, but more than '005, inch may be entered as '01 inch. 
 If the amount collected in the gauge is less than one-tenth of an 
 inch, the decimal point and the first should always be entered 
 in the register. Thus, seven-hundredths should be written 
 07 inch, and, to three places of decimals, seventy-one thousandths, 
 071 inch. Even the minutest quantity ('001 inch) should be 
 recorded, but a day is not to be counted a " rain day " unless 
 the measurement amounts to *005 inch (five thousandths of an 
 inch). Very heavy falls of rain should, if possible, be measured 
 immediately after they occur, and noticed in the " Remarks " 
 column of the Meteorological Register. The amount should, of 
 course, be included in the next regular entry. 
 
 In the modern and improved measuring-glass, the lower part 
 is made to taper to a point internally, so as to enable the critical 
 quantity of '005 inch to be very clearly defined. The divisions 
 for each hundredth of an inch should be distinctly etched on the 
 glass, and the capacity up to each mark should be as follows in 
 grains of water at 60 F. (see table on p. 240). 
 
 More rain is collected on the ground than on the top of a 
 building or of a stand at a height above the ground. In British 
 Rainfall, 1880, will be found papers on this subject by the late 
 Mr. George Dines, F.R.Met.Soc., and the late Mr. G. J. Symons, 
 
240 
 
 METEOROLOGY 
 
 F.R.S. The latter writer believed that the deficiency in the 
 amount of rain collected in gauges on high buildings is wholly 
 due to the position of the gauge, to the configuration of that 
 portion of the building which is close to the gauge, and to the 
 strength and direction of the wind during times of rain. In the 
 many experiments which Mr. Symons quoted there was no evidence 
 of any difference between the fall of rain, at various heights from 
 60 feet to 260 feet above the ground. Observations made at 
 the Wolverhampton Waterworks in the years 1849-52 showed 
 that the rainfall on the top of the water-tower, 180 feet high, 
 was on the average only 76 per cent, of that recorded by a low 
 gauge at the foot of the tower on a post about 7 feet high. Season, 
 
 TABLE VX CAPACITY OF MEASURING -GLASSES FOB, RAIN-GAUGES. 
 
 
 
 
 
 For 5-inch Gauge. 
 
 For 8-inch Gauge. 
 
 
 
 
 
 Grain?. 
 
 Grains. 
 
 Up 
 
 to -005 inc 
 
 h 
 
 
 25 
 
 63 
 
 
 01 
 
 
 
 50 
 
 127 
 
 
 02 
 
 
 
 99 
 
 254 
 
 
 03 
 
 
 
 149 
 
 381 
 
 
 04 
 
 
 
 198 
 
 508 
 
 
 05 
 
 
 
 248 
 
 635 
 
 
 10 
 
 
 
 496 
 
 1,269 
 
 
 20 
 
 
 
 992 
 
 2,539 
 
 
 30 
 
 t 
 
 
 1,488 
 
 3,808 
 
 
 40 
 
 , 
 
 
 1,984 
 
 5,078 
 
 
 50 
 
 
 
 
 2,480 
 
 6,347 
 
 however, had a marked influence, for while the ratio for the 
 summer five months, May to September, averaged 81 per cent., 
 that for the winter and windy five months averaged only 68 per cent. 
 Symons's explanation is now generally accepted, and it has 
 finally disposed of the old theory that the growth of the raindrop 
 by the condensation of particles of aqueous vapour floating in 
 its path through the air was adequate to explain the difference 
 of rainfall with elevation. Sir John Herschel, it is true, had 
 already demolished this theory, by showing that the latent 
 heat of steam being about 1,000 F., drops of rain, if they acquire 
 an increase in weight amounting to 1 per cent, by condensed 
 vapour, must in so doing have their temperature raised 10 F. 
 If they acquire an increase of 5 per cent., they must have their 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 241 
 
 temperature raised about 50 F. In the paper from which we 
 have culled these remarks, Mr. Symons observed : l " Experi- 
 mental evidence proves that Mr. Jevons was quite right in his 
 theoretical view that the fall of rain is practically identical at 
 all elevations, and that the observed differences are due to 
 imperfect collection by the gauges." 
 
 Size of Raindrops. When we consider the enormous mass 
 of material which has been accumulated regarding the quantity 
 of rain which falls, it is remarkable how little attention appears 
 
 FIG. TO. FORMS OF RAINDROPS. 
 
 Complete set of samples from the great general storm of August 20, 11)04. Duration of 
 storm, fifteen hours. One raindrop sample per hour was taken throughout the storm. 
 
 (Reproduced by permission of the United States Weather Bureau.) 
 
 to have been given to the number and size of the drops. A very 
 simple and ingenious method of studying rain-drops is described 
 in a paper in the U.S. Monthly Weather Review for October, 1904, 
 by Mr. Wilson A. Bentley. The raindrops are allowed to fall into 
 a layer of dry flour 1 inch deep, which is exposed to the rain for 
 a few seconds. The flour is allowed to stand for some time, 
 and the pellets of dough, each representing a raindrop, are then 
 picked out and may be preserved. The method was tested by 
 
 1 British Rainfall. 1881. P. 45. 
 
 16 
 
242 METEOROLOGY 
 
 allowing measured drops of water to fall from a height into the 
 flour ; it was found that the dough pellet differed but little in 
 size from the drop which produced it. In the paper a series of 
 interesting photographs of such dough-pellets is given, illustrating 
 the variation in the size of the raindrops during the course of 
 showers of different types. The largest drops met with some- 
 what exceeded i inch in diameter. This is in agreement with 
 the observations of Wiesner (quoted by Hann in his Lelirbucli), 
 which gave 7 millimetres as an upper limit. Mr. Bentley gives 
 tables showing the relative frequency of occurrence of drops 
 of various sizes in rain from various kinds of clouds. 1 
 
 In reference to the remarkable statements of Sir John Herschel 
 just quoted, the effect of rainfall in warming the air may be 
 imagined from a calculation by the late Rev. Dr. Haughton, 
 F.R.S., Senior Fellow of Trinity College, Dublin, 2 that " 1 gallon 
 of rainfall gives out latent heat sufficient to melt 75 pounds of ice, 
 or to melt 45 pounds of cast iron. From this datum it is easy to 
 see that every inch of rainfall is capable of melting a layer of ice 
 upwards of 8 inches in thickness (exactly 8*1698 inches) spread 
 over the ground." From such considerations Dr. Haughton 
 concluded that on the west coast of Ireland the heat derived 
 from the rainfall is equivalent to one-half of that derived from 
 the sun (R. H. Scott). 
 
 The chief physical cause of rain is the sudden chilling of com- 
 paratively warm air, more or less laden with moisture, either 
 by its ascent into the upper and colder regions of the atmosphere, 
 or by its impact against cold mountain slopes or (in winter) 
 the colder surface of the ground, as on the western coasts of 
 Europe. The former cause is more potent in summer, the latter 
 in winter. It was formerly supposed that rain was largely 
 caused by the mixture of masses of air of different temperatures. 
 But, even supposing that any such admixture did take place 
 (which is problematical), Dr. J. Hann, of Vienna, has shown, from 
 a comparison between the units of heat set free by condensation 
 and the weight of aqueous vapour per cubic foot of air at any 
 two given temperatures one high, the other low that the 
 mixture of volumes of air cannot be very effective in causing 
 1 Nature, February 23, 1905. 2 Physical Geography, p. 126. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 243 
 
 precipitation ; in fact, the setting free of latent heat in the process 
 of condensation largely prevents that fall of temperature which 
 is assumed to take place and to cause a rainfall. 
 
 The electricity of the atmosphere also plays an important 
 part in the production of rain. It may not be generally known 
 that, apart from the sublime and often awe-inspiring electrical 
 phenomena which accompany a thunderstorm, electricity is 
 always at work in the earth's atmosphere. And not the least 
 useful product of that work is the raindrop. 
 
 Lucien Poincare, Inspecteur - General de FInstruction 
 Publique, Paris, writes : l " If the pressure of a vapour that 
 of water, for instance in the atmosphere reaches the value of 
 the maximum pressure corresponding to the temperature 
 of the experiment, the elementary theory teaches us that the 
 slightest decrease in temperature will induce a condensation, 
 that small drops will form, and the mist will turn into rain. 
 
 " In reality matters do not occur in so simple a manner. A 
 more or less considerable delay may take place, and the vapour 
 will remain supersaturated. We easily discover that this pheno- 
 menon is due to the intervention of capillary action. On a drop 
 of liquid a surface tension takes effect, which gives rise to a 
 pressure which becomes greater the smaller the diameter of the 
 drop. 
 
 " Pressure facilitates evaporation, and on more closely examin- 
 ing this reaction we arrive at the conclusion that vapour can never 
 spontaneously condense itself when liquid drops already formed 
 are not present, unless forces of another nature intervene to 
 diminish the effect of the capillary forces. In the most frequent 
 cases these forces come from the dust which is always in suspen- 
 sion in the air, or which exists in any recipient. Grains 2 of 
 dust act by reason of their hygrometrical power, and form germs, 
 round which drops presently form. It is possible to make use, 
 as did M. Coulier as early as 1875, of this phenomenon to carry 
 off the germs of condensation, by producing by expansion in a 
 bottle containing a little water a preliminary mist, which purifies 
 
 1 La Physique Moderne, son Evolution (<?/. " The New Physics and its 
 Evolution." The International Scientific Series. Edited by F. Legge. 
 London : Kegan Paul, Trench, Triibner and Co. 1907. Pp. 242-244.) 
 
 2 See p. 195, supra. 
 
 162 
 
244 METEOROLOGY 
 
 the air. In subsequent experiments it will be found almost 
 impossible to produce further condensation of vapour. 
 
 " But these forces may also be of electrical origin. Von Helm- 
 holtz long since showed that electricity exercises an influence 
 on the condensation of the vapour of water, and Mr. C. T. R. 
 Wilson, with this view, has made truly quantitative experiments. 
 It was rapidly discovered after the apparition (i.e., discovery) 
 of the X-rays that gases that have become conductors that is, 
 ionised gases also facilitate the condensation of supersaturated 
 water vapour." 
 
 Rontgen, in 1895, showed that when a current of electricity 
 was made to pass through a vacuum tube, the tube emitted rays 
 which were capable of passing through bodies opaque to ordinary 
 light. These rays could, for example, pass through the flesh of 
 the body, and throw a shadow of the bones on a suitable screen. 
 
 The Rontgen rays, or X rays as they are also called, cause 
 gases, and even liquids and solids, through which they pass to 
 become conductors of electricity. Again, gases exposed to 
 Rontgen rays are found to include particles charged with 
 electricity some with positive, others with negative electricity. 
 " We know from these investigations/' says Sir Joseph Thomson, 1 
 " that electricity, like matter, is molecular in structure that 
 just as a quantity of hydrogen is a collection of an immense 
 number of small particles called molecules, so a charge of elec- 
 tricity is made up of a great number of small charges, each of a 
 perfectly definite and known amount." 
 
 One of the most wonderful and interesting advances ever made 
 in physics was the discovery and investigation of radio-activity 
 the power possessed by certain metallic substances and their 
 compounds of giving out rays which, like Rontgen rays, affect 
 a photographic plate, make certain minerals phosphoresce, and 
 make gases through which they pass conductors of electricity 
 (Thomson). Examples of such radio-active substances are 
 uranium (Becquerel), thorium (Schmidt), radium and polonium 
 (Monsieur and Madame Curie), actinium (Debierue), and potas- 
 sium (Campbell). Professors Rutherford, Soddy, and other 
 
 1 " Presidential Address to the British Association for the Advancement 
 of Science," Winnipeg, Manitoba, August, 1908. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 245 
 
 physicists, through their researches, have recently added many 
 new radio-active substances to this classical list. 
 
 " The radiation emitted by these substances," says Sir Joseph 
 Thomson, "is of three types, known as a, f$, and y rays. The 
 a rays have been shown by Rutherford to be positively electrified 
 atoms of helium, moving with speeds which reach up to about 
 one-tenth of the velocity of light. The /3 rays are negatively 
 electrified corpuscles, moving in some cases with very nearly the 
 velocity of light itself, while the y rays are unelectrified, and are 
 analogous to the Rontgen rays." 
 
 In his Presidential Address Sir Joseph Thomson also tells us 
 that " a knowledge of the mass and size of the two units of 
 electricity, the positive and the negative, would give us the 
 material for constructing what may be called a molecular theory 
 of electricity, and would be a starting-point for a theory of the 
 structure of matter ; for the most natural view to take, as a 
 provisional hypothesis, is that matter is just a collection of 
 positive and negative units of electricity, and that the forces 
 which hold atoms and molecules together, the properties which 
 differentiate one kind of matter from another, all have their 
 origin in the electrical forces exerted by positive and negative 
 units of electricity, grouped together in different ways in the 
 atoms of the different elements." 
 
 On February 1, 1901, Mr. C. T. R. Wilson read a paper before 
 the Royal Society on the " lonisation of Atmospheric Air." The 
 principal results of his investigation were (1) that ions are con- 
 tinually being produced in atmospheric air (as is proved also by 
 Geitel's experiments), and (2) that the number of ions of each 
 kind (positively and negatively charged) produced per second 
 in each cubic centimetre of air amounts to about twenty. 1 
 
 To return to M. Poincare, 2 " We are thus lad by a new road 
 to the belief that electrified centres exist in gases, and that each 
 centre draws to itself the neighbouring molecules of water, as 
 an electrified rod of resin does the light bodies around it. There 
 is produced in this manner round each ion 3 an assemblage of 
 
 1 Proc. Roy. Soc., voL Ixviii., pp. 151-161. May 4, 1901. 2 Loc. cit. 
 
 3 An "ion" IB " any minute material particle which carries an electrical 
 charge." (Hints to Meteorological Observers. By W. Marriott. F.R.Met.Soo. 
 Sixth edition. 1906.) 
 
246 METEOROLOGY 
 
 molecules of water, which constitute a germ capable of causing 
 the formation of a drop of water out of the condensation of 
 excess vapour in the ambient air. As might be expected, the drops 
 are electrified, and take to themselves the charge of the centres 
 round which they are formed ; moreover, as many drops are created 
 as there are ions. Thereafter we have only to count these drops 
 to ascertain the number of ions which existed in the gaseous 
 mass. 
 
 " To effect this counting, several methods have been used, 
 differing in principle, but leading to similar results. It is possible, 
 as Mr. C. T. R. Wilson and Professor J. J. Thomson have done, 
 to estimate, on the one hand, the weight of the mist which is 
 produced in determined conditions, and, on the other, the average 
 weight of the drops, according to the formula formerly given by 
 Sir G. Stokes, by deducting their diameter from the speed with 
 which this mist falls ; or we can, with Professor Lemme, determine 
 the average radius of the drops by an optical process viz., by 
 measuring the diameter of the first diffraction ring produced 
 when looking through the mist at a point of light/' 
 
 Mr. C. T. R. Wilson has observed that the positive and negative 
 ions do not produce condensation with the same facility. Con- 
 densation by negative ions is easier than by the positive. Hence 
 the negative electricity of rain. 
 
 In the second edition of Preston's Theory of Heat (p. 408), the 
 Editor, Mr. J. Rogerson Cotter, M.A. (Dub.), writes : 
 
 " Dust-free air may contain water vapour of a density several 
 times as great as that necessary for saturation. For if a very 
 small drop were to form, it would evaporate unless the vapour 
 pressure were great enough to be in equilibrium with the curved 
 surface. If drops of various sizes were present, the small ones 
 would tend to evaporate and condense on the larger ones. 
 
 " C. T. R. Wilson has shown that if air containing water 
 vapour be freed from dust and supersaturated by a sudden ex- 
 pansion, a cloud or fog will form if the air is ionised by the passage 
 of Rontgen or similar rays, and this will take place with a much 
 smaller expansion than is necessary to produce condensation if the 
 rays are absent. In this case the charged ions appear to act as 
 nuclei for the condensation of vapour." 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 247 
 
 "An ion (atom + electron) can play the part of a condensation 
 nucleus " (L. C. A. Bonacina). 1 
 
 Coloured Rain. Showers of grey, red, yellow, or black rain 
 have been recorded from time to time. Luke Howard, in his 
 Climate of London (1833), mentions a fall of chalky rain, which 
 occurred " in the third region of Mount Mtna " on the morning of 
 April 24, 1781. The ground was wet with a coloured or eta- 
 ceous grey water. The grey matter " was an oxide, produced 
 from a metallic base, or sublimate, ejected, along with a 
 prodigious quantity of steam, from the bowels of the mountain, 
 and condensed along with that into this singular mixed 
 rain/' 
 
 In the chronicles of the Middle Ages " showers of blood " are 
 often alluded to. The red colour was probably due to the 
 Protococcus nivalis, a minute vegetable organism which causes the 
 red snow of Arctic and Alpine regions. Occasionally, however, 
 the colour was caused by the presence of an inorganic substance, 
 such as ferric oxide. 
 
 Towards the end of 1896 a red dust fell heavily in Melbourne 
 and over a considerable area of Victoria. A very clean sample 
 of the dust was collected by Mr. W. E. Appleby, a resident of 
 Moonee Pounds, and analysed by Mr. Thomas Steel, F.L.S., 
 F.G.S., who found that it consisted largely of sand (66 per cent.), 
 but ferric oxide was present to the extent of 4*68 per cent., and 
 ferrous oxide to the amount of *5 per cent. 
 
 L. F. Kamtz, in his Meteorology (1845), wrote : " Formerly, and 
 even at the present time, flowers of sulphur have been said 
 frequently to fall with rain. After heavy showers quiet waters 
 were found covered with a yellowish dust, and, as it was easily 
 inflamed, it was concluded to be sulphur. Every year the news- 
 papers contained notices of it. More accurate researches have 
 proved that this dust is nothing else than the pollen of certain 
 flowers, and of pines in particular, which was swept off by the 
 wind and precipitated with the rain. Elsholtz had said this as 
 long ago as 1676." 2 
 
 1 " Some of the Causes and Effects of Atmospheric Electricity " (Symons's 
 Meteorological Magazine, vol. xli., p. 169, 1906). 
 
 2 See a letter by Mr. H. Sowerby Wallis in Symons's Monthly Meteorological 
 Magazine, November, 1886 (vol. xxi., p. 144). 
 
24:8 METEOROLOGY 
 
 In June, 1879, " showers of sulphur " were reported from 
 various localities in the United Kingdom. Such a shower 
 of yellow rain fell in Dublin, and on examining the resulting 
 water with a low-power microscope, I found the cause of 
 the colour to be merely pollen grains from some species of 
 Coniferce. 
 
 The late Dr. Alexander Buchan, F.R.S., in his Handbook of 
 Meteorology, published in 1868, stated that " the Hack showers 
 which occasionally fall in Scotland are in all likelihood the dust 
 or scoriso from the volcanoes of Iceland transported southwards/' 
 A more probable and reasonable explanation is that the black 
 coloration of the falling rain was due to the presence of soot. 
 In Symons's Meteorological Magazine for February, 1908 (vol. xliii., 
 p. 2), Dr. Otto Boeddicker, observer at the Earl of Rosse's observa- 
 tory at Birr Castle, King's Co., describes a " black rain " which 
 fell over the central counties of Ireland, from Tipperary to 
 Armagh, on October 8 and 9, 1907. Large quantities of soot 
 were deposited in some places. At Lynbury, near Mullingar, 
 Co. Westmeath, the deposit of soot in a recently-cleaned tank 
 was sufficient to choke a J-inch pipe. A soot-laden cloud 
 approached Birr from east-south-east about 2 p.m. of October 9. 
 The last tidings of this cloud reached Dr. Boeddicker from 
 Faelduff, twelve miles to the west of Westport, Co. Mayo, where 
 it again discharged black rain " much darker than bog water." 
 He considers that we have here evidence of a soot-laden cloud, 
 originating probably in South Wales, crossing St. George's 
 Channel and the whole of Ireland, and finally disgorging its soot 
 into the Atlantic. 
 
 Mr. H. Sowerby Wallis, in a letter on " Remarkable Showers/' 
 published in Symons's Monthly Meteorological Magazine, Novem- 
 ber, 1886, quotes Sir J. F. W. Herschell, Bart. (Meteorology, 1862) 
 as follows : " Showers of fish, frogs, flannel (matted Confervce), 
 bread (edible fungus), Infusoria, and other unaccountable sub- 
 stances, are among the more palpable evidences on record of the 
 elevating and transporting power of whirlwinds." 
 
 Under the same heading, " Remarkable Showers," we find 
 references in Symons's Meteorological Magazine from time to time 
 to showers of " frogs " (July, 1842 or 1843, in Suffolk), " hazel- 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 249 
 
 nuts " (May 9, 1867, in Dublin), " sulphur " i.e., pollen (October, 
 1867, at Thames Ditton), " larvse" (April 22, 1871, at Bath, and 
 April 29 of the same year at Aldershot), " hay " (July 28, 1875, 
 at Monkstown, Co. Dublin), " snails " (July, 1886, at Redruth, 
 Cornwall), and " fish " (June 15, 1895, in Co. Clare). 
 
 " Red rain " fell in Sicily and many parts of Southern Italy on 
 Sunday, March 10, 1901, alarming the peasants on account of its 
 resemblance to blood. The red colour was proved, on examination 
 by Professor J. W. Judd, to be due to dust or fine sand raised 
 from the Sahara, and carried across the Mediterranean by a 
 sirocco. 1 At the beginning of the same month the blood-rain 
 plant a motile alga named Sphcerella pluvialis, closely allied 
 to the better-known S. nivalis of red snow was found by 
 Dr. H. R. Mill in the large evaporation tank (6 feet square and 
 2 feet deep) at Camden Square, London, N.W. (See note on 
 p. 264). 
 
 Before making a few remarks as to the Distribution of Rain, 
 it may be well to say something about the other forms in which 
 precipitation takes place snow, sleet, and hail. 
 
 1. Snow consists of watery particles, frozen or congealed into 
 crystalline forms of infinite variety and exquisite beauty. It is 
 white or transparent, and entangles in its loose texture relatively 
 large quantities of atmospheric air (about ten times its own bulk). 
 To this last peculiarity snow owes its property of being a very 
 bad conductor of heat, so that it protects the earth from the 
 effects of terrestrial radiation in winter, the soil underneath it 
 being at times 40 F. warmer than the superincumbent air. The 
 white colour of snow is due to the blending of prismatic colours 
 flashed from the countless surfaces of minute snow-crystals, as 
 well as to the air entangled by these crystals ; it is analogous to 
 the whiteness of pounded glass or of foam. Red snow and green 
 snow have been observed in the Arctic Regions and elsewhere. 
 The coloration is due to the presence of minute micro-organisms, 
 Ttnnr mcn m diameter, called Protococcus nivalis or Sphcerella 
 nivalis. 
 
 In a beautiful word-picture Professor Tyndall describes a fall 
 of snow which he witnessed on the summit of Monte Rosa as 
 1 Nature, March 28, 1901. 
 
250 METEOROLOGY 
 
 " a shower of frozen flowers. All of them were six-leaved ; some 
 of the leaves threw out lateral rib-like ferns ; some were rounded, 
 others arrowy and serrated ; some were close, others reticulated, 
 but there was no deviation from the six-leaved type. Nature 
 seemed determined to make us some compensation for the loss 
 of all prospect, and thus showered down upon us those lovely 
 blossoms of the frost, and had a Spirit of the Mountain inquired 
 my choice the view or the frozen flowers I should have hesi- 
 tated before giving up that exquisite vegetation. It was wonder- 
 ful to think of, as well as beautiful to behold. Let us imagine the 
 eye gifted with a microscopic power sufficient to enable it to see 
 the molecules which composed those starry crystals ; to observe 
 the solid nucleus formed and floating in the air ; to see it drawing 
 towards it its allied atoms, and those arranging themselves as if 
 they moved to music, and ending by rendering that music con- 
 crete. Surely such an exhibition of power, such an apparent 
 demonstration of a resident intelligence in what we are 
 accustomed to call ' brute matter/ would appear perfectly 
 miraculous, and yet the reality would, if we could see it, tran- 
 scend the fancy. If the Houses of Parliament were built up 
 by the forces resident in their own bricks and lithologic blocks, 
 and without the aid of hodman or mason, there would be 
 nothing intrinsically more wonderful in the process than in the 
 molecular architecture which delighted us upon the summit of 
 Monte Rosa." 
 
 A very elaborate series of 151 different forms of snow-crystals, 
 drawn by Mr. James Glaisher, F.R.S., will be found in the Fifth 
 Report of the Council of the British Meteorological Society (now 
 the Royal Meteorological Society), published in 1855. The 
 crystals are either hexagonal plates or six-pointed stars, with 
 angles which are always multiples of 15 or 30, so bearing a 
 close relation to those of a regular hexagon (R. H. Scott). 
 
 Beautiful engravings of snow-crystals will be found in the 
 Philosophical Transactions for 1742 (vol. xlii.), by Dr. Leonard 
 Stocke of Middelburg, in Zealand ; and in the same publication 
 for 1755 (vol. xlix.), by Dr. John Nettie, also of Middelburg. The 
 latter series includes ninety-one different forms of crystals, which 
 were observed in the intensely cold winter of 1740-41. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 251 
 
 For modern work on snow-crystals we are indebted to Dr. G. 
 Hellmann, of Berlin ; x Dr. G. Nordenskiold, of Stockholm ; 2 but 
 especially to Mr. Wilson A. Bentley, of Jericho, Vermont, U.S.A. 
 In the Monthly Weather Review of the United States Weather 
 Bureau for May, 1901, the last-named observer gives a sketch of 
 his twenty years of study among snow-crystals, illustrating it by 
 about twenty-five exquisite photomicrographs of snow-forms. 
 In the Annual Summary of the same publication for 1902 (vol. 
 xxx., No. 13, p. 607) Mr. Bentley published his " Studies among 
 the Snow-Crystals during the Winter of 1901-02, with additional 
 Data collected during Previous Winters." That winter proved 
 to be extremely favourable for observing, and over 200 beautiful 
 photomicrographs were obtained, showing snow-forms of infinite 
 variety, which also greatly exceeded in beauty and interest the 
 contributions of any other single winter. 
 
 The application of photography to this charming research, as 
 was to be expected, has shown that many of the elaborate draw- 
 ings of snow-crystals of former times were very misleading. 
 
 Mr. Bentley's monograph is illustrated by 22 plates, containing 
 255 different forms of snow-crystals. His original paper (May, 
 1901) has 3 plates, including 26 microphotographs selected out 
 of 800. 
 
 Like snow, hoar-frost crystals are divisible into two funda- 
 mental classes or types columnar and tabular. 
 
 Snow-rollers. Dr. W. N. Shaw, Director of the Meteorological 
 Office, communicated to the Royal Meteorological Society on 
 February 19, 1908, a paper by Mr. Charles Browett on this 
 subject. 3 At Ryton-on-Dunsmore, near Coventry, on the after- 
 noon and evening of January 29, 1907, there were several storms 
 of fine snow, which fell very evenly, and without drifting, to a 
 depth of about 1J inches. At 8 a.m. next day Mr. Browett 
 noticed that the snow on the lawn to the east of his house was 
 heaped up, as though someone had run with a spade in front of 
 
 1 Schnzekrystalle, Berlin, 1893. 
 
 2 " Preliminart meddelande rorande en undersokning snokristaller " 
 (Foren. i Stockholm Forhandlingar, Bd. xv., Heft 3, 1893 ; Geological 
 Society of Sweden, 146-158, Plates 5-26). 
 
 3 Quarterly Journal of the Royal Meteorological Society, vol. xxxiv., 
 No. 146, p. 87. April, 1908. 
 
252 METEOROLOGY 
 
 him. On going out he observed that the snow was cleared away 
 to the bare grass (except for slight bars of snow across) in tadpole- 
 like markings, the tails all pointing to N.N.W., the direction 
 whence the wind had been blowing all night, and at the heads the 
 heaped-up snow neatly turned over in a roll. In the discussion 
 which followed the reading of the paper, Mr. H. Mellish said that 
 he had frequently observed the phenomenon. He had looked 
 at his thermograph in 1907 on three occasions when falls of light 
 fluffy snow had been followed by a temperature just above 
 freezing-point. On each of these occasions the minimum had 
 occurred in the early evening, and had been succeeded by a 
 considerable rise before midnight. It therefore appeared that 
 a necessary condition for the formation of these snow-rollers was 
 a temperature slightly above freezing following a rather low 
 reading of the thermometer. (See Appendix V., p. 466). 
 
 Snowflakes vary in size to a remarkable extent. The largest 
 are fully an inch in diameter, and are observed at a comparatively 
 high temperature (32, or slightly above freezing-point), and 
 when the air is very damp. The flakes are seldom less than 
 ^ inch in diameter, but in an extremely cold, dry atmosphere 
 they may not exceed y^ inch ; they then form " snow-dust," 
 the penetrating power of which nothing can resist when it is 
 driven before a strong wind, as in the " blizzard " of North 
 America. 
 
 The following definitions were agreed on at the meeting of the 
 International Meteorological Committee held at Paris in Sep- 
 tember, 1885 : 
 
 Drifting snow (Germ., Schneetreiben ; Fr., Chasse-neige). 
 
 Snowstorm (Germ., Schneegestober ; Fr., Tourmente de neige). 
 
 If more than half of the country surrounding a station is under 
 snow the following symbol is to be employed : [H (a square sur- 
 rounding a star). (Munich Conference, 1891.) 
 
 When snow falls, its measurement demands constant attention 
 on the part of the observer. Should the wind be high and 
 temperature very low, drifting of " snow-dust " will be apt to 
 vitiate the measurement. Snow will be blown out of the gauge 
 on the one hand or drifted into it on the other. The depth of 
 snow in a sheltered place, free from drifting, should be carefully 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 253 
 
 measured by a 2-foot rule. On a very rough estimate, 1 foot of 
 dry snow may be taken to represent an inch of rain. The rain- 
 gauge should be visited frequently during a snowstorm, and the 
 snow carefully removed and thawed in a covered vessel protected 
 from evaporation. If snow is not falling at the hour of observa- 
 tion, the gauge (funnel and receiver) may be brought indoors, 
 when its contents will melt and may then be measured as rain. 
 The funnel should be covered with a large plate to prevent loss by 
 evaporation. It is also recommended to melt the snow in the 
 funnel and cylinder of the rain-gauge suddenly by adding a 
 known quantity of hot water, the amount of which should be 
 deducted from the final measurement. In practice this plan 
 does not work well, for the shrinkage of the water in volume 
 caused by its rapid reduction of temperature introduces a con- 
 siderable element of error into the calculation. In some gauges 
 hot water is poured into an outer casing, and the snow is thawed 
 in the funnel and cylinder without admixture with the water at 
 all. This is an excellent plan, but requires a slightly more costly 
 instrument, such as the snow-melting rain-gauge invented by Mr. 
 James Sidebottom, F.R.Met.Soc. In this instrument the case 
 is double, and warm water is poured into an angular tube, thus 
 melting the snow. When the snow (with which the warm water 
 is never in contact) in the funnel is melted, the water is run off by 
 a tap, and, if needed, a fresh supply is added. By this arrange- 
 ment any mistake from adding a wrong quantity of water is 
 rendered impossible. 
 
 Should the snow have been lifted out of the funnel by the wind 
 a good plan is to take the outside cylinder of the gauge, which has 
 the same diameter as the funnel, and to insert it in the snow, 
 where it lies level and of a uniform depth. The solid cylinder or 
 section of snow thus cut out should then be melted, and the 
 resulting water measured. 
 
 2. Sleet is half-thawed snow, or mingled snow and rain. It is 
 of rare occurrence in rigorous climates, but is frequently observed 
 in a British winter, and in the vicinity of large lakes or the open 
 sea. Sleet is generally formed by falling through a stratum of 
 air much warmer than that in which condensation has taken 
 place and whence it has come. It is, therefore, just the opposite 
 
254 METEOROLOGY 
 
 of " frozen rain " or " silver thaw " a phenomenon which 
 occasionally happens in winter, particularly in connection with 
 " glazed frost " (Germ., Glatteis ; Fr., verglas). In this case 
 raindrops, sometimes of large size, fall from clouds in a warm 
 upper current into a stratum of air near the ground, the tempera- 
 ture of which may still be many degrees below freezing-point. 
 The result is that the raindrops freeze before or when they reach 
 the ground. They fall as particles, or pellets, or spicula of clear, 
 transparent ice, and presently the surface of the ground becomes 
 coated with ice, rendering locomotion on roads and footways well- 
 nigh impossible. 
 
 On January 22, 1867, a " silver thaw " occurred in the south- 
 east of England, which is thus described by Mr. G. J. Symons in 
 the Meteorological Magazine for February, 1867 : " At 7.10 p.m., 
 the ground being then frozen and the temperature of the air below 
 freezing-point, some rather sleety hail began to fall on the pave- 
 ment ; it crackled underfoot, and flattened out into diminutive 
 lozenges. About 8 p.m. it turned to rain, although the temperature 
 of the air was still several degrees below freezing-point, and the 
 ground-temperature was about 24. The necessary result was the 
 coating of everything with a layer of ice. At 9 p.m. the temperature 
 at 4 feet above the ground was 26*2, and it was still raining, and 
 the rain still freezing on pavement, walls, gravel walks, umbrellas 
 in fact, on everything. We never recollect being (meteoro- 
 logically) more mortified than we were at the failure of all our 
 efforts to reach the thermometers 20 feet above the ground ; but 
 climbing an iced pole was a feat beyond us, and we know not 
 what was the temperature at that small elevation. ... Of course, 
 the streets were in a frightful state. . . . There was for some hours 
 (till 3 a.m. in London) no safe mode of traversing the roads or 
 pavements but the very novel one of skating." 
 
 In 1672 accounts reached the Royal Society of a remarkable 
 " silver thaw " which visited Somersetshire and Oxfordshire 
 early in December of that year, and caused great destruction of 
 trees in plantations and orchards. One observer weighed the 
 sprig of an ash-tree of just f pound. The ice on it weighed 
 16 pounds at least. 1 
 
 1 Philosophical Transactions (abridged), vol. i., pp. 455 and 478. London 
 C. and R. Baldwin. 1809. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 255 
 
 3. Hail (Germ., Hagel ; Fr., grele), while as white as snow, is 
 much denser than it, and often consists of a central nucleus of ice 
 or condensed snow, with alternate deposits of hoar-frost and of ice 
 surrounding it, the latter sometimes taking the form of pyramidal 
 crystals. According to Dr. Marcet, the formation of hail requires 
 (1) a large accession of moisture ; (2) a temperature below freezing- 
 point ; (3) the presence of electrified clouds. Typical hailstorms 
 are nearly always associated with thunder and lightning, and 
 Mr. R. H. Scott says that Volta supposed that the hail pellets 
 are kept in a state of constant oscillation between two oppositely 
 electrified clouds, until by continued condensation the stones 
 grow so heavy that at last gravity prevails, and they break 
 through the lower stratum of the clouds and fall to the earth. 
 Hail, strictly speaking, is a day phenomenon, and occurs when the 
 air is warm near the ground, while the upper air is intensely cold. 
 Under these circumstances, dense and electrical cumuli form, 
 and from these clouds the hail descends. Such hailstones vary 
 in size from a small pea to an orange or a goose egg. Before 
 they fall all observers agree that a loud and continuous roar is 
 heard, caused by the stones being dashed hither and thither in the 
 air, either by electrical agency, as Volta thought, or, as seems 
 more likely, by a cyclonic whirl of air in the vicinity of the 
 storm-cloud. This is really Dove's theory. He held that hail- 
 storms are always whirlwinds, but with their axes almost hori- 
 zontal instead of vertical. I thoroughly endorse this view. It 
 is supported by the behaviour of the barometer in what are called 
 " thunderstorm depressions/' by the cyclonic shifting of the wind 
 in any ordinary summer shower, and by the phenomena attending 
 spring tornadoes in North America and elsewhere. One of my 
 earliest recollections is my having been an eyewitness of a 
 tornado of this kind which devastated a part of Dublin on the 
 afternoon of Thursday, April 18, 1850. The Rev. Dr. Lloyd, 
 then President of the Royal Irish Academy, and afterwards 
 Provost of Trinity College, Dublin, communicated a most 
 interesting and graphic account of the tornado to the Academy 
 four days after it occurred, of which the following is an ab- 
 stract : 
 
 " The first indications of the approach of the storm were 
 
256 METEOEOLOGY 
 
 observed soon after three o'clock. Massive cumuli were seen 
 forming in the south-western portion of the sky. These became 
 denser as they approached, until they formed a mass of an ash- 
 grey colour, projected on a sky of a paler tint, while the rugged 
 outliers from the mass, of the peculiar form (between cirrus and 
 cumulus) which indicates a high degree of electrical tension, 
 showed plainly that a storm was approaching. About half- 
 past three o'clock it burst forth. The flashes of lightning 
 (generally forked) succeeded one another with rapidity, and 
 at length the roar of the thunder seemed continuous. Some 
 persons who observed the phenomenon from a distance were 
 able to distinguish the two strata of oppositely electrical 
 clouds, and to see the electrical discharges passing between 
 them. 
 
 " Hitherto the wind was light, and there was that peculiar 
 closeness in the air which is the result of high temperature and 
 excessive humidity. Shortly before four o'clock the rain com- 
 menced ; this was followed almost immediately by discharges of 
 hail, and at 4 p.m. the terrific tornado, which was the grand and 
 peculiar feature of this storm, reached us. 
 
 " This gale, which appears to have been a true whirlwind, first 
 sprung up from the south-east, driving the hail before it im- 
 petuously. It then suddenly, and apparently in an instant, 
 shifted to the point of the compass diametrically opposite, and 
 blew with increased violence from the north-west. The noise 
 about this time of the shifting of the wind was terrific, and arose 
 (as is conjectured respecting similar tropical phenomena) from 
 the confused conflict of hail in the air. The size of the hailstones 
 as well as the vehemence of the gale appeared to be greater during 
 the second phase of the storm than in the first. These masses, 
 many of which were as large as a pigeon's egg, were formed of a 
 nucleus of snow or sleet, surrounded by transparent ice, and this 
 again was succeeded by an opaque white layer, followed by a 
 second coating of ice ; in some of them I counted five alterna- 
 tions. 
 
 " In less than ten minutes the tornado had passed. The wind 
 returned to a gentle breeze from the south-west, and the weather 
 became beautiful." 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 257 
 
 Dr. Lloyd adds : "In the tornado, the vortex is of much 
 smaller dimensions (than that of a cyclone or great revolving 
 storm), and is produced by rapidly ascending currents of air, 
 caused by the heating of a limited portion of the earth's surface 
 under the action of the sun's rays. In the temperate zones, 
 accordingly, it is never produced in winter. These ascending 
 currents are loaded with vapour, which (owing to the rapid 
 evaporation) is in a highly electrical state, and when they reach 
 the colder regions of the atmosphere the vapour is condensed, and 
 electrical clouds are rapidly formed/' 
 
 In the park and garden of Trinity College nineteen trees were 
 uprooted by the tornado, eleven being trees of large size. Ten 
 fell from the south-east, or under the influence of the first half of 
 the gale, and nine from the north-west. The bearings of the 
 fallen trees were accurately taken, and showed that the main 
 direction of the south-east gale was S. 56 E., and that of the 
 north-west gale N. 53 W. The centre of the vortex, therefore, 
 passed over the College Park. 
 
 The barometer fell from 29*964 inches at 1 p.m. to 29*930 inches 
 at 4 p.m., rising again to 29*944 inches at 7 p.m. 
 
 From an unofficial return made by the Metropolitan Police, it 
 appears that damage to the amount of 26,332 Os. lid. was done 
 in the city by the tornado, including the breaking of 388,635 panes 
 of glass. 
 
 On the morning of August 31, 1891, Venice was devastated by 
 a hailstorm, of which an eyewitness writes : "I never before 
 realised how property can be destroyed, and personal injury 
 inflicted, and even death incurred, by simple exposure for a minute 
 or two to the blows of hailstones. Now I do. The storm came 
 on with a suddenness that took us all by surprise. A few dark 
 clouds coming over the city from the north and west, and a few 
 flashes of lightning and rattling peals of thunder, and in a moment 
 a tempest of hail was upon us. The awful noise made me think 
 that the hurricane of wind that was raging was sweeping the roofs 
 clear of their tiles and levelling chimneys to the ground. In part 
 that was the case, but the noise was the noise of hailstones, solid 
 pieces of ice as big as eggs, and some as big as oranges, that did 
 not fall, but were being driven with terrific force on to the roofs of 
 
 17 
 
258 METEOROLOGY 
 
 the houses, on to the pavement, and into the water. ... As they 
 fell into the canal that goes by my door its usually quiet surface 
 seemed to boil furiously. The storm was over in a few minutes, 
 and I went out. Venice seemed to have suffered a bombard- 
 ment. . . . The officials in the office of the Adriatica newspaper 
 secured some of the hailstones as they fell, and had them weighed. 
 Several weighed 250 grammes that is, more than J pound. In 
 some places the streets seemed to be covered with a bed of white 
 stones." 
 
 In cold weather in spring hail sometimes assumes another form, 
 and falls from cumuli in pyramidal soft masses, like miniature 
 snowballs. This is " soft hail " (Germ., Graupel ; Fr., gresil). 
 The front of the pyramid as it falls is convex ; the remainder of 
 the soft hailstone is either conical, when the whole resembles a 
 tiny pegtop upside down, or pyramidal with a hexagonal or square 
 base, as if it originally formed part of a large sphere of hail. 
 Mr. H. A. Cosgrave, M.A., describes the former shape as having 
 characterised hailstones which fell on July 3, 1877, at Kilsallaghan, 
 Co. Dublin j 1 while W. T. Black figures the latter shape in an 
 account of a heavy fall of hail at Leamington on March 31, 1876. 2 
 Hail of moderate size accompanies snow showers in winter, 
 whether by day or by night, coming with northerly (north-west to 
 north-east) winds. Tiny granules of hail also fall from the grey 
 roll-cumulus of winter anticyclones, but never in large quantity. 
 The international symbols distinguish between true hail, & , and 
 soft hail, A . 
 
 An attempt was made many years ago by Mr. G. J. Symons, in 
 his Meteorological Magazine* to determine the weight of hail- 
 stones in relation to their size. He gives a table which is based 
 on the assumptions : 
 
 1. That the hailstones are truly spherical. 
 
 2. That they consist wholly of clear ice. 
 
 A cubic inch of water weighs 253 grains, but the specific gravity 
 of ice is "93, therefore a cubic inch of ice cannot weigh more than 
 235 grains, or a trifle over half an avoirdupois ounce. Each 
 
 1 Sy :nons 's Monthly Meteorological Magazine, vol. xii., p. 86. 
 
 2 Loc. tit., vol. xi., pp. 55, 56. 3 Vol. xiv., p. 115. 1879. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 259 
 
 01 inch of rain in the measuring- jar of an 8-inch gauge contains 
 127 grains. Each '01 inch in that of a 5-inch gauge weighs 
 50 grains. If, therefore, ten selected hailstones, when gradually 
 melted in the measuring-glass of the 5-inch gauge, yield "13 inch, 
 we have : 
 
 13 x 50 
 
 --= grains = 65 grains of water = the weight of each of the ten hailstones. 
 
 A modern work on " Hail " is that from the pen of the Hon. 
 Rollo Russell, F.R.Met.Soc., which was published in 1893 
 (London : Edward Stanford). The author concludes, as the 
 result of a wide experience, that the clouds in which large hail has 
 its origin are commonly at a great height, between 15,000 and 
 40,000 feet, or higher. These clouds are the result chiefly of 
 expansion and refrigeration of warm humid air, of the sudden 
 mixture of masses of air greatly differing in temperature and 
 vapour tension, and of free radiation. The nucleus of a hailstone 
 consists of a snowflake, pellet, or spicule, which falls from the 
 uppermost cloud. The snowflake, pellet, or spicule, is electrified 
 as a result of condensation, and as it falls attaches particles of ice 
 and globules of water below the freezing-point to itself, the 
 particles arranging themselves commonly in a stellate form, or 
 concentrically round the nucleus. The variety of form of the 
 primitive kernel is great, and consequently hailstones of many 
 different shapes may be met with. The ordinary top-shaped hail- 
 stone is produced by the lower side growing more quickly than 
 the upper, as it comes into contact with more particles ; and since 
 the impact is most forcible on the lower side, the ice of the 
 spheroidal base is the hardest. Mr. Russell's book is illustrated 
 by two photographs of hailstones (actual size) taken after a 
 terrific thunderstorm at Richmond, Yorkshire, on July 8, 1893, 
 by Mr. H. J. Metcalfe, photographer, High Row, Richmond, Yorks. 
 Some of the hailstones figured have a diameter of 2 inches. 
 
 The recent literature on the " building of hail " includes the 
 following : 
 
 1. "Die Bildung des Hagels," by Wilh. Trabert. Meteorolog. 
 Zeitschrift, pp. 433-447. October, 1899. 
 
 2. " Beitrage zur Hageltheorie," by P. Schreiber. Meteorolog 
 Zeitschrift, pp. 58-70. February, 1901. 
 
 172 
 
'260 METEOROLOGY 
 
 3. Lehrbuch der Meteorologie, by J. Harm, pp. 682-699. 1901. 
 
 4. "Hailstones/' by F. W. Verz. Trans. Acad. of Science 
 and Art, Pittsburg. (A lecture delivered January 5, 1904.) 
 
 5. " Studies on the Thermodynamics of the Atmosphere," by 
 Professor Frank H. Bigelow. Monthly Weather Review, U.S. 
 Weather Bureau, vol. xxxiv., No. 11, pp. 514-517. November, 1906. 
 
 Professor Bigelow considers that hail is formed at the rear of 
 the rising column of warm air in the front of a storm, at a place 
 of marked changes in the isotherms, when the barometer is 
 beginning to rise rapidly, and the wind shifts from south to 
 north-west. This is the site (locus) of the contact of two counter- 
 currents of air having very different temperatures, and hail 
 formation is one of the results of the rapid progress of the warm 
 and cold layers towards thermal equilibrium. He discusses five 
 distinct theories of the formation of hail, in each of which there is 
 probably an element of truth : (1) The oscillation theory, (2) the 
 orbital theory of Professor William Ferrel, 1 (3) the upward current 
 theory, (4) the electrical attraction theory, and (5) what he calls 
 the stratification theory. According to this (Bigelow's) theory, 
 a hailstorm cloud consists of two component portions, separated 
 from each other by isothermal surfaces inclined forward 
 from the vertical. On the front side the air is much warmer 
 than on the back side, and along the line of separation the 
 contour is strongly stratified by the mutual interpenetration from 
 opposite directions of layers of air having different temperature. 
 
 Distribution of Rain. This topic may be considered under the 
 headings geographical and seasonal. 
 
 The hyetal (Greek, t>erds, rain) equator is the line separating areas 
 whose rainfall follows the seasons of the Northern Hemisphere 
 from those whose rainfall follows the seasons of the Southern 
 Hemisphere: This line is south of the geographical equator in 
 the east, and north of it to the west of the continents (A. Supan). 
 
 I. Geographical 1. British Islands. Thanks in great measure 
 to the energy and organising power of Mr. G. J. Symons, F.R.S., 
 and of his very able colleague and successor, Dr. Hugh Robert 
 Mill, the United Kingdom is now covered with a network of rain- 
 gauge stations, upwards of 4,500 in number in 1908, and the 
 
 1 " Recent Advances in Meteorology," Appendix 71, Annual Report of the 
 Chief Signal Officer for 1885, part ii., pp. 302-315. 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 261 
 
 observations are digested and published in British Rainfall each 
 year under the personal supervision and editorship of Dr. Mill. 
 Mr. Symons's coloured rainfall map gives the leading facts in a 
 striking and intelligible form. The average annual rainfall as a 
 rule decreases from west to east, both in Ireland and in Great 
 Britain. It exceeds 75 inches in the West of Scotland (151 inches 
 on an average of fifteen to eighteen years at the dismantled Ben 
 Nevis Observatory), the English Lake District, in the mountains 
 of North Wales, and on the top of Dartmoor in Devonshire. 
 Seathwaite in Borrowdale, at the south end of Derwentwater, has 
 an annual fall of 137 inches on the average of fifty years ; near 
 Stye Head above it the mean fall is 177 inches, or some 15 feet, 
 rising in wet years above 200 inches, or 1 7 feet. In Ireland the rain- 
 fall is heaviest on Mangerton, in the neighbourhood of Killarney 
 (about 86 inches, based, however, on only eight years' average), at 
 Kylemore, in Connemara (mean for fifteen years being 77*6 inches), 
 smallest in and about Dublin (about 28 inches) and in the Co. Down. 
 On the East Coast of England it falls well below 25 inches. Spurn 
 Head, Yorkshire, on an average of ten years, had only 19*1 inches ; 
 and Shoeburyness, in Essex (in twenty-five years), 20'6 inches. 
 
 The reason for this distribution is that westerly winds are those 
 which prevail most in the British Islands. They reach the 
 mountainous western coasts off the Atlantic, and laden with 
 moisture in consequence, are forced upwards above the saturation 
 line. When these prevalent winds reach the eastern seaboard 
 they have already lost a great deal of their moisture through con- 
 densation, and they are descending a state of things which raises 
 their temperature and increases their capacity for vapour. They 
 are then like the dry, warm, south wind of the northern slopes of 
 the Alps, which is called the Fohn in Switzerland. The Chinook, 
 or warm westerly wind of the Canadian prairies east of the Rocky 
 Mountains, has a similar controlling effect on the rainfall of the 
 Province of Alberta. 
 
 2. In Foreign Parts, the wettest regions are the Equatorial zone 
 of calms over the Atlantic and Pacific Oceans, and " localities 
 where damp winds meet mountain ranges and are forced up- 
 wards " (R. H. Scott). Examples of the latter are the Khasi Hills 
 in Assam (Cherra Poonjee, or Cherrapunji, 464 inches as the mean 
 of thirty-three years), the Western Ghats in India (Mahabaleshwar, 
 
262 METEOROLOGY 
 
 269 inches, the mean of forty years), the western coast of Norway 
 (Bergen, 73 inches on an average of twenty-five years), Sitka in 
 North-West America, Valdivia in Southern Chili, and Hokitika 
 in New Zealand. The driest regions are the Sahara, Egypt, 
 Arabia, and Persia, the southern steppes of Russia, the North- 
 West of India ( Jacobabad, on the Upper Sind Frontier, has only 
 4*1 inches per annum), the Great Salt Lake region in North 
 America, the Kalahari Desert in South Africa, the interior of 
 Australia, and Peru and Northern Chili between the Andes and 
 the sea in South America. The desert of Gobi, in Central Asia, is 
 rendered almost rainless by intercepting chains of lofty mountains. 
 
 II. Seasonal. On the western shores of Europe the winter 
 rainfall exceeds that of summer in persistence and amount. This 
 is due to the prevalence of westerly winds, and to the coldness of 
 the highlands near the Atlantic seaboard. In winter the rains 
 mainly accompany cyclonic storms. Rain falls most heavily 
 along the western coast-line, as the isotherms run almost parallel 
 to it and perpendicular to the prevailing winds (Drs. A. Angot 
 and A. J. Herbertson). On the continent of Europe the summer 
 rainfall exceeds that of winter, at all events in amount. This is 
 doubtless in consequence of the torrential rains which accompany 
 summer thunderstorms. In fact, as Dr. R. H. Scott puts it, 
 the rains of low latitudes are essentially summer rains, as they occur 
 principally when the sun is highest. Another good name for 
 them would be " evaporation rains/' because they fall from cumuli 
 which have been formed by local evaporation and the ascent 
 of the vapour above the saturation or condensation line. The 
 great Indian rains accompany the south-west monsoon that is, 
 they are caused by the condensation of the vapour-laden winds 
 blowing from the Indian Ocean. 
 
 In Dublin the monthly rainfall, on an average of forty years, 
 1866-1905, is least in April (1-913 inches) and June (1'912 inches), 
 greatest in August (3'130 inches) and October (2'803 inches). 
 
 The wettest month in most parts of England and the East of 
 Scotland is October ; in the West and North- West of Scotland, 
 however, it is December or January ; and in the East of England 
 July or August. 
 
 III. The diurnal fall of rain is determined by the seasons. As 
 a rule, in winter more rain falls by night than by day ; in summer, 
 
THE ATMOSPHERE OF AQUEOUS VAPOUR 263 
 
 more rain falls by day than by night. At ordinary British 
 stations falls more than 2 inches in the twenty-four hours are 
 not common. In Dublin, since 1865, such a fall has occurred 
 only on nine occasions : August 13, 1874 (2-482 inches) ; Octo- 
 ber 27, 1880 (2-736 inches) ; May 28, 1892 (2-056 inches) ; July 24, 
 1896 (2-020 inches) ; August 5, 1899 (2-227 inches) ; August 2, 
 1900 (2-135 inches) ; November 11, 1901 (2-037 inches) ; Septem- 
 ber 2, 1902 (2-075 inches) ; and August 25, 1905 (3-436 inches). 
 This last excessive rainfall is especially noteworthy. On no 
 previous or subsequent occasion within the past forty-five years 
 have 3 inches or upwards of rain been measured as the product of 
 twenty-four hours in the city of Dublin. But these downpours 
 pale into insignificance before the record rainfalls of the world, of 
 which, perhaps, the most notable is that which wrought such 
 terrible ruin in Brisbane at the beginning of February, 1893. 
 In the Blackall Ranges, near the city of Brisbane, 77*305 inches 
 of rain fell in the four days ending February 3, 35-714 'inches 
 being the measurement on February 2. But this fall, tremendous 
 as it is, does not stand out as a world's record, for on June 14, 
 1876, 40*80 inches of rain fell within twenty-four hours at Cherra- 
 punji, Khasi Hills, Assam, being at the rate of 1*7 inches per hour. 1 
 Weight and Bulk of Rain. When we speak of an inch of rain, 
 we mean that sufficient has fallen to fill to overflowing a vessel 
 which is 1 inch in length, 1 inch in breadth, and 1 inch in depth 
 that is, a volume of 1 cubic inch. Now, an acre contains 6,272,640 
 square inches, each of which would receive an inch depth of rain 
 if the rainfall was 1 inch. One inch of rain over an acre is there- 
 fore 6,272,640 cubic inches. But, according to recent determina- 
 tions, 2 1 imperial gallon contains 277-123 cubic inches. So that 
 
 6 ' 72 ^t=22,635 gallons ; or 101 tons cwt. 3 qrs. 26 Ibs. 
 
 In round numbers, therefore, a rainfall of 1 inch means a down- 
 pour of 101 tons of water on every acre. As there are 640 acres 
 in a square mile, a rainfall of 1 inch means a precipitation of 
 64,640 tons of water on every square mile. 
 
 1 Professor John Eliot, " The Rainfall of Cherrapunji." Quarterly Journal 
 of the Meteorological Society, vol. viii., pp. 47, 51. 1882. 
 
 2 T able-Book, p. 35. By Rev. Isaac Warren, M.A. Longmans, Green 
 and Co. 1888. 
 
264 METEOKOLOGY 
 
 " Blood-rain " due to a Dust-fall On February 21 and 22, 
 1903, a remarkable dus'-fall occurred over the whole of the 
 South of England, the greater part of Wales, the Low Countries, 
 the German Empire, and Switzerland. The dust fell over 
 nearly all parts of England and Wales south of a line drawn 
 from Anglesey through Wrexham and Northampton to Ipswich. 
 The area over which dust fell comparatively thickly in England 
 and Wales was certainly not less than 20,000 square miles, 
 and the total quantity of deposit in England alone has been 
 roughly estimated at not less than 10,000,000 tons. The dust 
 usually attracted attention either as a dense yellow haze, or 
 as a reddish-yellow powder, lying thickly on trees or roofs. In 
 some instances it fell as a dry powder ; in others it was noticed 
 in the form of drops of muddy rain. The fall was often accom- 
 panied by temperatures considerably above the average, and by 
 remarkably low relative humidities ; at Uccle, near Brussels, for 
 example, with a dry-bulb temperature of 58, the relative 
 humidity was only 32 per cent, of saturation. At a meeting of the 
 Royal Meteorological Society on November 18, 1903, Dr. H. R. 
 Mill, D.Sc., and Mr. R. G. K. Lempfert, M.A., read a paper on this 
 great dust-fall. In this paper the authors attempted to trace the 
 origin of the dust which was brought to our islands by a south- 
 west wind. With the help of a series of daily weather charts, 
 trajectories were drawn for the air which reached the North- West 
 of Europe on the morning of February 22. Four trajectories 
 were obtained. They seem to show that the air which reached 
 the British Isles on the morning of the 22nd was derived from 
 three sources : The North of Scotland, where temperature was 
 low, was supplied with air from the Northern Atlantic ; Ireland, 
 and the North of England were deriving their air from the Central 
 Atlantic ; while over Wales and the South of England, apparently, 
 air was being derived from the North- West Coast of Africa. 
 
 Appended to the communication of Dr. Mill and Mr. Lempfert 
 is a note on the microscopic characters of this " blood-rain/' by 
 Mr. John S. Flett, M.A., D.Sc. He concluded that the only part 
 of the dust which could reasonably be supposed to have come 
 from beyond the British Isles was an exceedingly fine reddish 
 clay, none of the particles of which could greatly exceed '01 milli- 
 metre (-0004 inch) in diameter. 
 
CHAPTER XIX 
 ANEMOMETER AND ANEMOMETERS 
 
 WIND is air naturally in motion with any degree of velocity ; 
 it is a current of air. Wind is produced by differences of 
 atmospheric pressure, and these differences are in turn referable 
 to indeed, largely dependent on variations in temperature. 
 It has already been shown that the force of the wind is governed 
 by the steepness of barometric gradients in other words, by the 
 closeness to each other of the isobars, or lines of equal atmo- 
 spheric pressure. All mechanical obstacles, however, interfere 
 with the velocity or force of the wind for these two terms 
 come to mean the same thing, owing to the substantial nature 
 of the air and, accordingly, we find that in general the force 
 of the wind is greater at sea than on dry land, at a distance 
 above the earth than on the surface of the ground that is, in 
 what is now called " the free air " on the sea-coast than at an 
 inland station, on the slope or summit of a mountain than on a 
 plain, on a bare plain than in a wooded or hilly district. 
 
 In the Narrative of a Voyage to the Southern Atlantic Ocean 
 in the Years 1828-30, performed in H.M. " Chanticleer," Captain 
 the late Henry Foster, F.R.S., 1 Mr. W. H. B. Webster, surgeon, 
 R.N., stated the general principles of the relation of wind to 
 atmospheric pressure, and foreshadowed Buys Ballot's Law. 
 His remarks were based on personal observation of the singular 
 difference between the mean height of the barometer at the 
 Cape of Good Hope and at Valparaiso, on the coast of Chili 
 (about 30 inches), and that at Cape Horn, Staten Island, and 
 New South Shetland (29'3 to 29'4 inches). He says: 2 "If 
 
 1 From the Private Journals of W. H. B. Webster, 2 vols. London, 1834. 
 
 2 Op. cit., vol. i., p. 316. 
 
 265 
 
266 METEOROLOGY 
 
 we suppose that at any time the barometer is high at one place 
 and low at the other, we shall have at Cape Horn the barometer 
 at 28'3 inches, while at Valparaiso or the Cape it will be at 
 30' 6 inches, being an occasional (nay, frequent) difference of 
 more than 2 inches. Now, if we consider these changes to take 
 place principally in the lower strata of the atmosphere, which 
 in fact must be the case, and that they range within the limits of 
 five or six miles' altitude, how great must be the difference of 
 the weights and pressures of the reciprocal columns. It is not 
 surprising, then, that there should be continual gales endeavour- 
 ing to restore the equilibrium. From the foregoing statements 
 it may be safely inferred that * the mean height of the barometer 
 at the level of the sea being the same in every part of the globe ' 
 is by no means correct ; but, on the contrary, that every place 
 has its own peculiar height of the barometer ; and to this per- 
 manent variation, a circumstance not heretofore recognised, 
 may be attributed the perpetual interchange and motions of the 
 atmosphere." 
 
 In this connection it may be incidentally mentioned that it 
 is the steep gradient over the immense Southern Ocean which 
 gives rise to the strong and gusty anti-trades or westerly winds 
 which blow in the " Roaring Forties " of that ocean, and which 
 prevail as far south as lat. 50. 
 
 The oldest method of observing wind is by sensation or by 
 estimation. It is a rough but ready method, and in the hands 
 of a skilled observer yields fairly satisfactory results. 
 
 The earliest attempts at estimating wind force were doubtless 
 made by sailors, from whom we have learned the expressions : a 
 " gale," a " whole gale," a " squall," a " strong breeze," a 
 " capful of wind," a " light breeze," a " dead calm." It has 
 been already stated that such expressions were reduced to a scale 
 by Admiral Sir F. Beaufort in 1805, whose Table of Wind Force 
 (revised) is printed at p. 39 of this book. The nautical part of 
 this table is of little use to a landsman, and, accordingly, the 
 equivalent velocities and pressures have been calculated at the 
 Meteorological Office, London, and included in the table. The 
 mean velocity in English miles per hour being known, the 
 equivalent velocities according to the metric scale i.e., in metres 
 
ANEMOMETRY AND ANEMOMETERS 267 
 
 per second are obtained by multiplying the figures in the fifth 
 column that is, the number of English miles by the factor 
 447. The scale of velocities included in the table must, how- 
 ever, be regarded as merely provisional, and as applicable chiefly 
 to coast stations. At inland stations the velocity corresponding 
 to a given force is much smaller, because the general motion of 
 the air is retarded by inequalities in the surface of the ground, 
 while wind force is naturally estimated from that of the gusts. 
 
 Wind Direction. By this term is meant the point of the com- 
 pass from which the wind is blowing. It may be ascertained 
 by observing for a few moments the movements of a properly 
 set and freely movable vane or weathercock. Chaucer has it 
 " As a wedercok that turneth his face with every wind." 
 
 When a weathercock is not available, the drift of the smoke 
 from exposed chimneys should be carefully noted. Under all 
 circumstances, the bearings should be true Tind not magnetic 
 (by compass). In the British Islands the variation of the compass 
 at the present time ranges from 15 in the extreme east of 
 England to 22 in the extreme North- West of Ireland the mag- 
 netic north lying so many degrees to west of the true north, 
 or the true north so many-degrees to east of the magnetic north. 
 Roughly speaking, a true north and south line lies along the line 
 north-north-east to south-south-west by compass. Accordingly, 
 we get the following table for the conversion of directions 
 observed by mariner's compass in the United Kingdom to 
 approximate true bearings : 
 
 Compass Bearings. True Bearings. 
 
 North 
 
 North-north- east 
 
 North-east 
 
 East-north-east 
 
 East 
 
 East-south-east 
 
 South-east 
 
 S outh- south- east 
 
 South 
 
 S outh- south- west 
 
 South-west 
 
 West-south-west 
 
 West 
 
 West- n orth - west 
 
 North-west 
 
 North-north-west 
 
 North-north-west 
 
 North 
 
 North-north-east 
 
 North-east 
 
 East-north-east 
 
 East 
 
 East-south-east 
 
 South-east 
 
 S outh- south- east 
 
 South 
 
 South- south- west 
 
 South-west 
 
 West- south- west 
 
 West 
 
 West-north-west 
 
 North-west 
 
268 METEOEOLOGY 
 
 In the absence of a mariner's compass, we can ascertain the 
 north point by means of the pole star, or the south point by 
 means of the sun. The pole star, Polaris, is practically due 
 north in January and July at 6 a.m. and 6 p.m., in February 
 and August at 4 a.m. and 4 p.m., in March and September at 
 2 a.m. and 2 p.m., in April and October at noon and midnight, 
 in May and November at 10 a.m. and 10 p.m., in December and 
 June at 8 a.m. and 8 p.m. In order to ascertain the position of 
 due south, we must know the longitude of a given place, and also 
 the equation of time, or the difference between mean and apparent 
 time that is, the difference between the time of day indicated 
 by the sun's position on the meridian and that indicated by a 
 perfect clock going uniformly all the year round. As a matter 
 of fact, the sun is not always on the meridian at 12 o'clock noon. 
 It was so in 1909 on April 16, June 15, September 1, and Decem- 
 ber 25 ; but on November 2, 3, and 4 it reached the meridian 
 16 minutes 21 seconds before noon, and on February 10, 11, 
 and 12 it was 14 minutes 25 seconds late, not arriving at the 
 meridian until hour 14 minutes 25 seconds p.m. Greenwich 
 time is converted into local time by subtracting four minutes for 
 every degree of west longitude, by adding four minutes for every 
 degree of east longitude. Thus, noon at Greenwich becomes 
 11 hours 35 minutes a.m. in Dublin, the longitude of the Irish 
 capital being 6 15' west (6 x 4 m. =24 m.+ 1 m. =25 m.). In 
 Bradshaw's British Railway Guide a map gives longitude from 
 Greenwich in time direct, without calculation. 
 
 A " windrose " may be constructed " by calculating the per- 
 centage proportion of the number of wind observations from 
 each point of the compass, and printing the results either in a 
 tabular form or representing them by a diagram. Windroses 
 may be made to show the force, as well as the direction, of the 
 wind from different points " (R. H. Scott). 
 
 The Meteorological Congress at Vienna, in 1873, decided that, 
 in the construction of windroses, winds of velocity less than 
 \ metre per second (one mile per hour) were to be disregarded, 
 and counted as calms. A year previously, at Leipzig, it was 
 arranged by the Meteorological Conference that calms should be 
 enumerated separately, and designated by a special abbreviation. 
 
ANEMOMETRY AND ANEMOMETERS 
 
 269 
 
 such as the letter " C." The accepted abbreviation at present 
 is the letter " Z," standing for force 0, or zero. 
 
 FIG. 71. CASKLLA'S SELF-RECORDING ANEMOMETER OR ANEMOGRAPH (see p. 
 
 Anemometers. The history of mechanical anemometry (Greek, 
 afe//.os, wind ; fj-trpov, a measure) may be held to date from 
 the year 1667, when the Royal Society published a revised 
 
270 METEOROLOGY 
 
 edition of " Master Rooke's " Directions for Seamen. The editors 
 were Dr. Hooke and Sir Robert Moray, who say, inter alia, 
 " The strength of the wind is measured by an instrument such 
 as is represented." The representation shows a plate suspended 
 by a bar from a pivot, and thus able to swing upwards when 
 pressed by the wind along a graduated quadrant, the quadrant 
 itself, with the plate, turning freely as a vane on a vertical 
 shaft. 
 
 On March 15, 1882, Mr. J. K. Laughton, M.A., F.R.G.S., 
 read his presidential address to the British Meteorological Society, 
 entitling it a " Historical Sketch of Anemometry and Anemo- 
 
 FIG. 72. RECORDING CYLINDER OF CASELLA'S ANEMOGRAPH (see p. 289). 
 
 meters." 1 To this address I am indebted for the foregoing 
 information, as well as for the following classification of anemo- 
 meters : 
 
 A. Pendulum. Such as Hooke's (1667), Pickering's anemo- 
 scope (1744), Dalberg's (1780), Schmidt's, of Giessen (1828), 
 Wild's (1861), Hewlett's (1868). The Vienna Congress (1873) 
 recommended the introduction of Professor Wild's gauge, which 
 is in use in Russia and Switzerland. It consists of a rectangular 
 plate hung on hinges on a horizontal axis. The angle which this 
 makes with the vertical indicates the force of the wind. This 
 
 Quarterly Journal of the Meteorological Sxiety, vol. viii., No. 43, p. 161. 
 
 1882. 
 
ANEMOMETRY AND ANEMOMETERS 271 
 
 instrument measures the force of light winds accurately, but 
 fails in the case of strong winds, because the plate will be kept 
 almost horizontal by even a moderate breeze. 
 
 B. Bridled. Wolf (1708), Leupold (1724), Leutmann (1725), 
 Beaufoy (1821), Francis Galton's " torsion anemometer " (1879), 
 in which a set of Robinson's cups are bridled by a spring on a 
 vertical shaft ; Ronalds (1844), Stokes (1881). In Sir F. Ronalds's 
 instrument the force of the wind is determined by means of a 
 simple balance. 
 
 C. Pressure Plate. Bouguer's (Traite du Navire, 1746), in 
 which a piece of cardboard, 6 inches square, was fixed per- 
 pendicularly on to a light rod, which pressed into a tube against 
 a spring ; Abbe Nollet (L'Art des Experiences, 1770) ; Osier's 
 (1836), in which the plate was separated from the wedge-shaped 
 vane, and acted on a wire passing down the hollow spindle of 
 the vane. In this way Mr. Osier obtained two distinct registers 
 one of direction, the other of pressure. Mr. Osier more recently 
 adapted a windmill vane to give direction instead of the original 
 wedge-shaped vane. 
 
 Jelinek (1850), in order to avoid the vacuum behind the 
 pressure plate, cased it in by a cylinder, closed behind, against 
 which it bore by three spiral springs. Cator (1864) made the 
 back of the plate the base of a cone, and received the pressure 
 on a system of levers instead of a spring. Professor Wilke, of 
 Stockholm (1785), described another form of pressure anemometer, 
 which he called an Anemobarometer. Pujoulx (1830 ?) adopted 
 the same plan in causing the pressure to act on a bladder con- 
 taining air, which by means of a double siphon-shaped tube 
 forced a column of coloured liquid to rise. 
 
 The principle of the pressure plate anemometers is shown in 
 the drawing on the following page (Fig. 73) used to illustrate 
 Mr. Dines's anemometer comparisons at Oxshott (Quarterly 
 Journal of the Royal Meteorological Society, vol. xviii., No. 83, 
 July, 1892, p. 165 et seq.). 
 
 Dines's Patent Pressure Portable Anemometer. This instru- 
 ment meets a want long felt in the shipping interest ; it is 
 very compact, and with moderate care is not likely to be damaged 
 
272 
 
 METEOROLOGY 
 
 or get out of order. It shows accurately the force of the wind, 
 the scale having been calibrated by direct experiment. 
 
 To use the anemometer, hold the case and pull up the projecting 
 nozzle as far as it will go. The case then forms a convenient 
 
 FIG. 73. PRESSURE PLATE ANEMOMETER, 
 
 handle. Unscrew the milled head at the top a few turns, and 
 hold the instrument in a vertical position, with the nozzle facing 
 The velocity is then shown on the scale by the height 
 
 the wind. 
 
ANEMOMETRY AND ANEMOMETERS 
 
 273 
 
 of the coloured liquid in the glass tube. Before 
 replacing in its case and putting away, screw 
 down the milled head gently until the rubber 
 washer inside seals the end of the glass tube, 
 taking care not to screw too hard for fear of 
 breaking the glass. 
 
 When using the instrument, be careful to choose 
 a fully exposed situation, and stand facing the 
 wind, holding it at least 1 foot in front of the 
 body. The nozzle should face the wind as nearly 
 as possible, but the registration is not affected so 
 long as it points within 15 to 20 of the right 
 direction. 
 
 If bubbles get accidentally formed in the glass 
 tube, they may be dislodged by gently sucking the 
 nozzle. 
 
 When the milled head is unscrewed and the 
 instrument is held vertically in still air, the liquid 
 should stand at zero. If it does not, a little must 
 be added or subtracted to make it do so ; but 
 the anemometers are sent out with the right 
 amount of liquid in them, and there is no reason 
 why this adjustment should be required. 
 
 It is desirable, though not necessary, to keep 
 these instruments with the upper end of the scale 
 highest ; hence a loop has been provided, so that 
 they may be kept hung on a nail. The milled 
 head should be screwed down before the instru- 
 ment is removed from the vertical position. This 
 instrument is made by L. Casella. 
 
 D. Pressure on a Fluid. Lind's (1775) anemo- 
 meter consisted of a Pitot's or U-tube, swinging 
 freely on a vertical spindle, so as to form a direc- 
 tion vane (Fig. 75). The tube nearer the spindle 
 was bent back at right angles, so as to present its 
 mouth to the wind, which, acting on water in the 
 bend of the tube, forced it up the other leg of 
 the U, the difference of level giving a measure of 
 
 18 
 
274 
 
 METEOROLOGY 
 
 the wind force. Wollaston (1829) modified this principle in the 
 construction of his " differential barometer." Adie (1836) caused 
 the air to blow down a bell-mouthed tube, led into the inside 
 of a cylinder, air-tight above, but open below, which floats in a 
 vessel containing water. This small " gasometer " rises as the 
 pressure of the air inside is increased. 
 
 E. Velocity. 
 
 1. Wheel with axis, horizontal or vertical, perpendicular to 
 the direction of the wind. Lomonosow (1751). 
 
 2. Windmill sails, or fan, with axis in the direction of the 
 wind. Woltman (1790); Whewell (1837); Rev. W. Foster 
 
 (1844). 
 
 3. Hemispherical cups (Fig. 76). The 
 Rev. W. Romney Robinson, D.D., in 
 1846 applied a fact, " which," he said, 
 " he had learned from the late Richard 
 Lovell Edgeworth, that if hemispherical 
 cups be carried by horizontal arms 
 attached to a vertical axis, with their 
 diametral planes vertical, they con- 
 stitute an effective windmill, which he 
 (Dr. Robinson) had found revolves 
 with one-third of the wind's velocity." 
 " To the bottom of the axis is attached 
 wheel - work actuating a revolving 
 disc, which rotates through a degree 
 for every mile traversed by the wind." 
 The principle of this anemometer is based entirely on the 
 difference of the wind pressure on the concave and convex 
 sides of the cups. Dr. Robinson adopted 3 as a general and 
 constant co-efficient to express this ratio. It is now known that 
 the original " constant " of 3, as settled by the inventor, is far 
 too .high. The co-efficient 2'5 was proposed by Professor Sir 
 George Stokes, Bart., in a paper in the Proceedings of the Royal 
 Society (vol. xxxii., p. 170), but Mr. Laughton considers that this 
 is only an approximation, and ought in strictness to be changed 
 for each individual instrument and every different wind. The 
 Kew authorities have finally decided to use the figure 2, and 
 
 FIG. 75. LIND'S ANEMOMETER. 
 
ANEMOMETRY AND ANEMOMETERS 
 
 275 
 
 Mr. Dines's experiments, referred to later on, indicate that at 
 times 1'85 is nearer the truth (see p. 287 et seq.). 
 
 Dr. Robinson's anemometer is described in the Transactions 
 of the Royal Irish Academy for 1850. The readings on the dials 
 of this anemometer are as follow : One complete revolution of 
 the first stamped index- wheel equals T Vth of a mile ; the second, 
 1 mile ; the third, 10 miles ; the fourth, 100 miles ; the fifth, 1,000 
 miles. Necessarily, in noting such reading, it must be done back- 
 wards, according to the indications on the instrument. The 
 cups travel at a rate equal to one-third that of the wind ; but 
 allowance having been made for this in graduating the circles, a 
 
 Fio. 70. ROBINSON'S ANEMOMETER. 
 
 true reading is at once obtained. Negretti and Zambra's Im- 
 proved Robinson's Anemometer (Fig. 77), as described by Colonel 
 Sir H. James, R.E., F.R.S., has two graduated circles. The 
 outer circle is graduated into five miles, each divided into tenths, 
 and the inner circle from 5 to 505 miles. The velocity of the 
 wind at any particular moment is found by observing this index 
 before and after a certain interval of time as one or five minutes 
 and then multiply the rate by sixty or twelve to find the 
 velocity in miles per hour. 
 
 In recommending a modification of Robinson's cup anemometer, 
 which he calls the " Step " anemometer, Mr. Walter Child points 
 
 182 
 
276 
 
 METEOROLOGY 
 
 out that the shortcomings of " the Robinson " are two its 
 results apply only to the motion of one thin, horizontal stratum 
 of air, and they are, as is nowadays well-known, vitiated by the 
 " sheltering error." This is especially the case with high winds. 
 The cups in their revolution set the contiguous air in rotation 
 in the same direction. In this way the resistance to the motion 
 of the cups is lessened an effect which culminates in a gale. 
 The instrument, on this account, registers what actual wind 
 there is at too high a velocity. Also, when the wind drops, the 
 instrument runs on by the inertia of its part, and so registers 
 
 FIG. 77. NEGRETTI AND ZAMBRA'S IMPROVED ROBINSON'S ANEMOMETER. 
 
 wind when there is none. This may be called the error of " over- 
 running." 
 
 In order to obviate these drawbacks, Mr. Child arranges the 
 four cups of the Robinson anemometer no longer at the same 
 level, but in tiers or " steps " one above another, though at right 
 angles as before. 
 
 The " Step " anemometer was tried at Kew Observatory, and 
 yielded results which were some 25 per cent, less than the 
 standard. This discrepancy, so Mr. Child submits, is the 
 " apparent " measure of the " sheltering error " of the standard 
 at the wind velocities under notice. 1 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. xxxiii., 
 No. 144, p. 295. October, 1907. 
 
ANEMOMETRY AND ANEMOMETERS 
 
 277 
 
 4. Current meter. In a portable magnetic anemometer and 
 current meter for maritime use, designed by Mr. R. M. Lowne 
 in 1874, 1 the measurement of a current of air or water is effected 
 by the revolution of a wheel carrying a number of plates of very 
 thin aluminium, so arranged that their flat surfaces lie at an 
 angle of 45 to the plane of the wheel's motion. When a wheel 
 so formed is placed in a current, it revolves in a given time a 
 number of turns that exactly express the velocity of the current 
 which passes the wheel. The number of the revolutions of the 
 wheel is indicated by pointers turning on a dial, and traversing 
 circles on which the lineal feet 
 of the current are expressed by 
 graduations and figures. 
 
 At the suggestion of the late 
 Sir Edmund Parkes, F.R.S., of 
 the Royal Victoria Hospital, 
 Netley, Mr. Casella, of London, 
 constructed an air-meter for 
 measuring the velocity of cur- 
 rents of air passing through 
 mines, hospitals, and other public 
 buildings (Fig. 78). 
 
 F. Evaporation or Tempera- 
 ture. These anemometers are 
 based on the principle enunciated 
 by Leslie in his Experimental 
 Inquiry (1804), that " the refri- 
 gerant power of a stream of air is exactly proportional 
 to its velocity. Hence we may determine the rate of cooling 
 that corresponds to any given velocity." Leslie's anemo- 
 meter is a thermometer with a bulb larger than usual. Sir 
 David Brewster (1829) adopted the principle that " when water 
 is exposed to wind, the quantity evaporated in a given time 
 is proportional to the velocity of the wind, the capacity of 
 the air for moisture remaining the same. His anemometer con- 
 sisted of a light frame, on which was stretched a surface of sponge 
 
 1 Quarterly Journal of the R-)yal Meteorological Society, vol. ii., p. 285- 
 1874. 
 
 FIG. 
 
 .AIR-METER. 
 
278 METEOROLOGY 
 
 or coarse flannel to be wetted. This frame was fixed perpen- 
 dicularly on a light horizontal rod, pivoted on an upright spindle, 
 round which it turned so as to face the wind. On the other 
 arm of the rod was a sliding weight, the rod being graduated as 
 a steelyard. The observed loss of weight gave a measure of 
 evaporation, and so of the wind's velocity. Phillips (1848-49) 
 re-invented the methods of calculating the velocity of the wind 
 from its power of cooling or evaporating. 
 
 G. Suction. Anemometers coming under this heading are 
 based on a principle, first illustrated by Bernoulli about 1738, 
 that the friction of masses of fluid in motion induces a power 
 of suction as a result of the production of a partial vacuum. 
 Professor Overduyn, of Delft (1854), and M. Bourdon (1882) 
 applied this principle to the measurement of wind. G. A. Hage- 
 mann (1876) designed two anemometers, one for stationary 
 observations, the other for observations on board ship or when 
 travelling. The latter combines the pressure with the suction 
 principle ; the former adopts the suction principle only. 
 
 In his anemometer for stationary use, Hagemann 1 uses only 
 the rarefaction produced by the wind on an open perpendicular 
 tube. This tube is usually a piece of ordinary gas-pipe, J to 
 \ inch in diameter, and is fastened either to a mast or on a 
 prominent place, such as a high chimney or a church tower. 
 The pipe has at the top a gilt brass mouthpiece, with an opening 
 of not less than 3 millimetres. It is carried down to the anemo- 
 meter, properly so called, which consists of a vessel about half 
 filled with pure water. From the bottom of this vessel a pipe 
 enters and opens above the water surface into the cavity of a 
 small gasometer, made of very light tinned sheet brass, and having 
 an upper surface of exactly 100 square centimetres. The gaso- 
 meter dips into the water, and is hung by a strong silk, which, 
 after first having passed over a pulley connected by a wheel 
 with an index-hand, has its other end fastened to a spring 
 properly tempered. The open perpendicular tube is connected 
 by means of a caoutchouc tube with the pipe which enters the 
 anemometer. When the connections are thus made, it is evident 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. v., p. 208. 
 1879. 
 
ANEMOMETRY AND ANEMOMETERS 279 
 
 that any change in rarefaction will turn the index-hand. A 
 rarefaction of 1 millimetre water pressure will act as if a layer 
 of 1 millimetre of water were laid on the whole surface of the 
 gasometer ; this on 100 square centimetres is a weight of 
 10 grammes. Hence, by loading the gasometer with 10, 20, 30, 
 40, up to 100 grammes, the position of the index will correspond 
 to 1, 2, up to 10 millimetres suction, and the scale is divided 
 accordingly. A scale for velocity of wind, in metres per second, 
 is calculated by the formula V=3'9x Jh (h being the water 
 pressure in millimetres), and is also marked. The velocity of 
 the wind is, so to speak, weighed. Neither temperature, pressure, 
 moisture, rain, snow, nor hail has any influence upon this anemo- 
 meter, the action of which is regarded by the inventor as perfectly 
 satisfactory. The Hagemann anemometers are manufactured 
 by Nyrop of Copenhagen. 
 
 H. Direction. Lomonosow (1751) attached to the vane- 
 spindle of his anemometer a vertical wheel, with a tube con- 
 taining mercury running round the greater part of its circum- 
 ference perhaps 300. As this wheel, bridled by a spring, 
 turned on its axis, a small quantity of the mercury was poured 
 out into a tray beneath, divided into thirty-two radiating com- 
 partments ; the compartment in which the mercury was after- 
 wards found indicated the direction of the wind, the quantity of 
 mercury its force. Beaudoux (1777) registered the direction 
 only by fine sand falling into a similarly divided tray. God- 
 dard (1844) used water in the same way. Craveri (1866 ?) 
 adopted a similar method of registry, corn grains being the 
 weight employed. 
 
 I. Inclination. Various instruments have been designed with 
 the object of showing whether any given current of wind has 
 an upward or downward tendency. Benzenberg (1801) caused 
 the windward end of the direction vane to carry a vertical fork 
 open to the wind ; across this was fixed the axis of a horizontal 
 vane, which showed the inclination of the wind. Cacciatore 
 (1840), Director of the Observatory at Palermo, caused the 
 velocity of the wind to be given by a horizontal fan of four 
 curved sails, which, being segments apparently quadrants of 
 
280 
 
 METEOROLOGY 
 
 a cylinder, necessarily revolved in one direction. A similar fan 
 on a horizontal axis was fixed in a rectangular frame fastened to 
 an upright spindle, so as always to swing away from the wind 
 and rotate in a plane at right angles to the wind's direction. 
 It could be acted on only by the vertical component of the wind. 
 More modern instruments for observing the inclination are those 
 J nvented by Professor Hennessy (1856) and Father Dechevrens 
 
 FIG. 79. VANE OF CASELLA'S ALTAZIMUTH ANEMOMETER. 
 
 (1881), of the Observatory of Zi-Ka-Wei, near Shanghai (Sur 
 I'lnclinaison des Vents). 
 
 In 1886, Mr. Louis Marino Casella, F.R.Met.Soc., described 
 to the Royal Meteorological Society an altazimuth anemometer 
 which he had designed and patented. The object of the instru- 
 ment is to record continuously the vertical angle, as well as the 
 horizontal direction and force, of the wind. A full description 
 of his instrument will be found in the Quarterly Journal of the 
 Royal Meteorological Society, vol. xii., No. 60, October, 1886, 
 
ANEMOMETRY AND ANEMOMETERS 281 
 
 p. 246. The existence of air currents moving in a direction more 
 or less inclined to the plane of the horizon, or " inclined currents," 
 as Mr. Casella calls them in his paper, being conceded, the 
 necessity for the study of their inclination and velocity is at once 
 apparent. Mr. Casella's anemometer, which includes in its con- 
 struction the principle of the engineering instrument known as 
 
 FIG. 80. CASELLA'S ALTAZIMUTH ANEMOMETER. 
 
 the altazimuth, records continuously on one sheet of paper the 
 pressure, direction, and inclination of the wind, with all its 
 changes, the pressure plate being always maintained truly at 
 a right angle to the wind. The apparatus for indicating the 
 direction of the wind consists of a vane (Fig. 79), constructed 
 of a pair of diverging blades fixed to a cap, mounted so as to 
 
282 METEOROLOGY 
 
 rotate about a vertical axis, the motion of this vane being trans- 
 mitted by a vertical tubular shaft passing downwards through 
 the usual fixed column to the registering mechanism. This 
 tubular shaft, called the " direction tube/' is made to operate 
 the styles which record its movements through the medium of 
 pinions and wheels, conveying motion to two discs, which are 
 made to carry pencils in a vertical position and equidistant ; 
 three styles are used, so that one is always ready to enter on 
 the scale at one side when another leaves it at the other side 
 (Fig. 80). 
 
 The apparatus for indicating the inclination of the wind 
 that is, its divergence from a horizontal plane consists of a 
 similar vane (composed of a pair of diverging blades), mounted 
 on a horizontal axis within the direction vane, so balanced as 
 to assume normally, when no wind is blowing, a position in 
 which its longitudinal axis is horizontal. To insure this, the 
 vane is brought to a condition of stable equilibrium as closely 
 approximating to that of instability as it is possible to 
 bring it. 
 
 The oscillating motion of this inclination vane is transmitted 
 to its registering mechanism by a tubular connecting rod, jointed 
 to the vane by a pair of links, which move up and down inside 
 the direction tube. The inclination tube is so connected with 
 the carriage of a style. Thus its longitudinal motion affects only 
 the latter, which thus records upon the scale the oscillations of 
 the vane due to the varying inclination of the wind. 
 
 The pressure plate is a disc, having an area of 1| square feet, 
 fixed to a guide rod, fitted to slide between pairs of guide rollers 
 in the frame of the inclination vane, and moving with it. This 
 plate should have had a cone at the back, which was omitted in 
 its construction. In order to prevent the varying positions of 
 the pressure plate affecting the balance of the vane, its motion 
 is constantly and exactly compensated by a movable weight 
 running on rollers, so arranged that the weight moves to a pro- 
 portionate extent in the opposite direction to the pressure plate, 
 so as to maintain the balance of the vane in all positions of th3 
 pressure plate. The motion of the pressure plate is transmitted 
 to the apparatus for measuring the force by means of a chain 
 
ANEMOMETRY 'AND ANEMOMETERS 283 
 
 attached to the guide rod of the plate, and passing down over a 
 pulley through the tubular shaft of the inclination vane. To 
 prevent the weight of this chain affecting the accuracy of the 
 records, it is exactly balanced by a counterpoise hanging in a 
 casing carried by the cap. In this way the pressure plate is 
 kept perpendicular to the direction of the air current, not only 
 in azimuth, but also in altitude. 
 
 The apparatus for measuring the pressure consists of a cistern 
 containing mercury, and a displacement plunger immersed 
 therein, connected to a frame of guide rods joined together 
 above and below the mercury cistern, which works up and down 
 against guide wheels mounted around the cistern, the chain 
 being attached to the bottom of the frame. The plunger has a 
 varying ratio of displacement for successive depths of immersion, 
 so that the scale may be open for the smaller and compressed 
 for the greater (and less frequent) pressures. In order to check 
 the motion of the plunger and avoid inaccuracy in the indica- 
 tions, due to the momentum of the parts, the lower end of the 
 plunger is provided with a disc, fitting more or less closely to 
 the sides of the mercury cistern, so as to prevent the too 
 rapid passage of the mercury from one side of the disc to the 
 other. 
 
 The frame is connected to the carriage of the marker for 
 registering the motions of the plunger by means of a bell-crank 
 lever. The carriage carrying the recording pencil is mounted 
 to travel upon rollers running upon the oppositely bevelled 
 edges of a horizontal bar and rollers running on the front and 
 back surfaces, by which it is truly guided with the least possible 
 friction. The markers are all metallic, sliding in sockets and 
 pressing by their own weight or by springs on the paper. The 
 scales for the different records are marked upon a single sheet 
 of paper wrapped round a cylinder, rotated at a uniform speed 
 by a clock movement in the usual manner. 
 
 As a mere anemoscope, Mr. Laughton considers that none of 
 these elaborate contrivances excels the simple little feather vane 
 in daily use on board our men-of-war. It is a tapering tail, 
 8 inches long by 1J inches wide where broadest, made of the 
 softest down or feathers, and tied to the top of a staff by a 
 
284 METEOROLOGY 
 
 short thread. Let the wind blow how it will, this must stream 
 with it. 
 
 J. Musical Anemometers have been suggested and designed 
 by Hooke (1667), Athanasius Kircher, Leupold (1724), and 
 Delamanon (1782). These instruments were so constructed as 
 to emit musical sounds when the wind blew upon them. They 
 were scientific toys at the best. 
 
 K. The Helicoid Anemometer. In the Quarterly Journal of 
 the Royal Meteorological Society (vol. xiii., p. 218, 1887), Mr. 
 
 Fio. 81. VANE OF RICHARD'S ANEMO-CINEMOGRAPHK. 
 
 W. H. Dines, B.A., F.E.Met.Soc., has a paper on a " New Form 
 of Velocity Anemometer," which he read before the Society on 
 April 20, 1887. In this instrument an attempt was made to 
 measure the velocity of the wind by the rotation of a small 
 pair of windmill sails, the pitch of the sails being altered auto- 
 matically, so that their rate may always bear the same ratio to 
 that of the wind. These sails present what is called a " helicoid " 
 surface (Greek, e'Ai, a spiral; and efSos, resemblance), or one 
 which may be rotated about its axis in a current of air (the 
 
ANEMOMETRY AND ANEMOMETERS 
 
 285 
 
 axis, of course, pointing in the direction of the current) without 
 causing any deflection or whirl in the air passing over it. 
 
 L. The Anzmo-Cinemographe.On May 19, 1892, the late 
 Mr. G. M. Whipple, B.Sc., Superintendent of the Kew Observa- 
 tory, laid before the Royal Meteorological Society the results 
 
 FIG. 82. RICHARD'S ANEMO-CINI^MOGRAPHE. 
 
 of a comparison of Richard's Anemo-Cinemographe with the 
 standard Beckley Anemograph at the Kew Observatory. 1 This 
 ingenious instrument (Figs. 81, 82, 83, and 84) is a modification 
 of the old Whewell fan, or windmill vane, the change being in 
 the shape of each blade of the vane, which is made oval and 
 
 1 Quarterly Journal of the Roijal Meteorological Society, vol. xviii., p. 257. 
 1892 
 
286 
 
 METEOROLOGY 
 
 fitted at an angle of 45 to the axis. The vanes are said by 
 MM. Richard to have been carefully calibrated. The fan is 
 formed by six little wings or vanes of sheet aluminium, 4 inches 
 in diameter, inclined at 45, riveted on very light steel arms, 
 the diameter of which is so calculated that the vane should make 
 exactly one turn for the passage of a metre of wind (Fig. 81). 
 Its running is always verified by means of a whirling frame 
 fitted up in an experimental room where the air is absolutely 
 calm, and, if necessary, a table of corrections is supplied. The 
 recording part of the apparatus is called the Anemo-Cinemographe, 
 
 FIG. 83. KICHARD'S ANEMO-CINEMOGRAPHE (Second Foim). 
 
 and in principle is as follows : The pen, recording on a movable 
 sheet of paper, is lowered at a constant rate by means of a conical 
 pendulum acting through a train of wheel-work, whilst a second 
 train, driven by the fan, is always tending to force it up from 
 the lower edge of the paper. Its position is, therefore, governed 
 by the relative difference in the velocity of the two trains of 
 wheel-work, being at zero of the scale when the air is calm, but 
 at other times it records the rate of the fan in metres per second 
 (Fig. 82). Fig. 83 represents another pattern of the same instru- 
 ment, which, however, does not record the directior of the wind, 
 
ANEMOMETRY AND ANEMOMETERS 287 
 
 MM. Richard having constructed another apparatus for that 
 purpose. 
 
 Another pattern of the anemo-cinemographe (Fig. 84), with 
 endless paper running 1J inches a minute, has been made by 
 MM. Richard for the Bureau Central Meteorologique of France 
 for studying the storms on the top of the Eiffel Tower in 
 Paris. 
 
 Notwithstanding the recommendation of the Vienna Meteoro- 
 logical Congress (1873), Robinson's and Osier's anemometers still 
 hold the field in British observatories, at all events. A Robinson 
 anemometer should be well exposed, its machinery should be kept 
 well oiled, and the following particulars should always be furnished 
 with a register of its indications : (1) Length of arm (axis to centre 
 of cup) ; (2) diameter of cups ; (3) how the registration is effected 
 (mechanically, electrically, or otherwise) ; (4) name of maker ; 
 (5) height above the general surface of the ground (William 
 Marriott). 
 
 The Munich International Meteorological Conference (1891) was 
 of opinion that it is desirable to publish wind velocities in metres 
 per second ; and the International Meteorological Committee, at 
 the Southport meeting in 1903, held that the height of the 
 anemometer above the ground ought always to be given at the 
 head of all published tables of wind velocities. 
 
 The mean direction of the wind may be calculated by the 
 formula proposed by Lambert towards the close of the last 
 century. It is given by Dr. Scott as follows : 
 
 T E 1 -W^(N.E.+ S.E. - S. W. - N. W.) cos 45. 
 
 ' N.-S. + (N.E. + N.W. -S.E. -S.W.) cos 45. 
 
 In this equation <f> is the deviation of the mean direction from 
 north round by east. 
 
 In 1889-90 Mr. W. H. Dines, B.A., F.R.Met.Soc., carried out a 
 series of experiments on the resistance of plates of various forms 
 at oblique incidences to the wind, and communicated the results 
 to the Royal Society in a valuable paper. 1 Mr. Dines subse- 
 quently carried out a series of comparisons between specified 
 anemometers at the request of the Council of the Royal Meteoro- 
 
 1 " On Wind Pressure on an Inclined Surface," Proceedings of the Royal 
 Society, vol. xlviii., pp. 233-257. 
 
288 
 
 METEOROLOGY 
 
 FlG. 84. AN^MO-ClN^MOGRAPHE IN POSITION. 
 
ANEMOMETRY AND ANEMOMETERS 289 
 
 logical Society, the cost being defrayed by the Meteorological 
 Council. The instruments compared were : 
 
 C\. Kew Pattern Robinson Anemometer. 
 2. Self-adjusting Helicoid Anemometer (Dines). This in- 
 Velocity f strument is described, as stated above, in the Quarterly 
 Instruments \ Journal of the Royal Meteorological Society, vol. xiii., 
 
 p. 218. 1887. 
 1 3. Small Air-Meter. 
 
 Pressure ( 4. Foot Circular Pressure Rate. 
 Instruments \5. A special modification of the Tube Anemometer. 
 
 The conclusions arrived at are given in a note appended to the 
 Report of the Meteorological Council to the Royal Society for the 
 year ending March 31, 1892 (pp. 23, 24). They are to the effect 
 that, with proper precautions, the tube anemometer a combina- 
 tion of Lind's and Hagemann's instruments will form a most 
 useful and convenient instrument, that the relation between 
 pressure and velocity for a foot circular plate, and at ordinary 
 barometrical pressure, is P='003 V 2 , and that the factor of the 
 Kew Pattern Robinson Anemometer is practically constant for all 
 velocities except very low ones, that its value must lie between 
 2-20 and 2'00, and that there is a very great probability that it 
 is within 2J per cent, of 2*10. 
 
 Mr. Dines read his paper on " Anemometer Comparisons " before 
 the Royal Meteorological Society on April 20, 1892. It is fully 
 illustrated, and was published in the Quarterly Journal of the 
 Society (vol. xviii., No. 83, July, 1892, p. 165). 
 
 At pages 269 and 270, illustrations are given of an enlarged 
 anemometer or anemograph, constructed by Mr. L. Casella for 
 harbours and public observatories. In this arrangement (Fig. 71) 
 windmill fans are added to the wind vane, causing the mean 
 direction of the wind to be accurately indicated by means of a 
 revolving cylinder (Fig. 72) to which paper is attached. The 
 direction as well as the velocity is continuously shown for every 
 minute of time by means of a clock, which forms part of the instru- 
 ment. The exposed part of this anemometer may be placed at 
 any height, whilst the registering part is kept in a room or other 
 covered place for observation. 
 
 The climate of the British Isles is essentially windy, or even 
 stormy. Hence the following will prove of interest to the student 
 of that climate. 
 
 19 
 
290 METEOROLOGY 
 
 In a paper read before the Royal Meteorological Society on 
 June 20, 1894, Mr. R. H. Curtis stated that the greatest force of 
 an individual gust which he had met with was registered in 
 December, 1891, and amounted to a rate of 111 miles per hour, 
 which, with the old factor, would be equivalent to a rate of about 
 160 miles per hour. Gusts at a rate of from 90 to 100 miles per 
 hour have many times been recorded, but the usual limit for gusts 
 may be taken to equal about 80 miles per hour, which on the old 
 scale would be equivalent to about 120 miles per hour. Gales 
 and strong winds differ much in character. There are gales which 
 are essentially squally. In these the gusts constitute the main 
 feature. In an average gale the ordinary gusts occur at intervals 
 of about ten to twenty seconds ; the extreme gusts at intervals of 
 about a minute. Another class of gales show a tolerably steady 
 wind velocity. In the third class are gales which appear to be 
 made up of two series of rapidly succeeding squalls the one 
 series at a comparatively low rate of velocity, the other at a much 
 higher one, the wind-force shifting rapidly and very frequently 
 from one series to the other. Mr. Curtis has not infrequently 
 found very distinctly marked in the anemometer tracings a 
 prolonged pulsation in the wind-force, which recurs again and 
 again with more or less regularity, sometimes every twenty 
 minutes or half an hour, sometimes at longer intervals of about an 
 hour or so. 
 
 Between sunset of February 26 and noon of February 27, 1903, 
 the British Isles were swept by a storm of most unusual violence. 
 A vast amount of damage was done to trees and buildings by 
 gales from the south or south-west, particularly in the neigh- 
 bourhood of Dublin, where very large numbers of trees were 
 uprooted, and in Lancashire. This storm of almost hurricane force 
 was the subject of an able paper by Dr. W. N. Shaw, D.Sc., 
 F.R.S.. Director of the Meteorological Office, with the assistance 
 of Mr. R. G. K. Lempfert, M.A., and F. J. Brodie, F.R.Met.Soc. 
 This paper was read before the Royal Meteorological Society on 
 June 17, 1903. 1 The appended Table gives the wind velocities 
 from anemograph records 
 
 1 Quarterly Journal of the. Royal Meteorological Society, vol. xxix., No. 128, 
 p. 233 : 1903 ; and Monthly Weather Review, vol. xxxi., p. 218 : 1903. 
 
ANEMOMETRY AND ANEMOMETERS 
 
 291 
 
 STORM OF FEBRUARY 26-27, 1903 (WIND VELOCITIES FROM 
 ANEMOGRAPH RECORDS). 
 
 Observatories. 
 
 Wind 
 Direction. 
 
 Time of Occurrence. 
 
 Maximum Velocity Recorded. 
 
 Valentia 
 
 W.S.W. 
 
 Midnight, 26th, to 
 
 63 miles in the hour. 
 
 
 1 a.m., 27th 
 
 
 
 ( 
 
 Midnight, 26th, to 
 
 37 
 
 Falmouth . . S. by W. -1 
 
 1 a.m., 27th 
 
 
 Armagh 
 
 S.S.E. 
 
 11.50 p.m., 26th 
 2 to 3 a.m., 27th 
 
 88 miles per hour, in squalls. 
 33 miles in the hour. 
 
 Kingstown . . 
 
 W.S.W. 
 
 4 to 5 a.m., 27th 
 
 66 
 
 Holyhead 
 
 S.W. 
 
 5 to 6 a.m., 27th 
 
 52 
 
 Southport 
 Stony hurst . . 
 
 w.s.w.{ 
 
 W.S.W. 
 
 6 to 7 a.m., 27th 
 5.55 a.m., 27th 
 6 to 7 a.m., 27th 
 
 65 
 87 miles per hour,in squalls. 
 43 miles in the hour. 
 
 Glasgow 
 
 W. 
 
 Noon to 1 p.m., 
 
 26 
 
 
 
 27th 
 
 
 Deerness 
 
 E. 
 
 7 to 8 a.m., 27th 
 
 36 
 
 North Shields 
 
 S.W. 
 
 7 to 8 a. m., 27th 
 
 70 
 
 Aberdeen 
 
 S.E. 
 
 3 to 4 a.m., 27th 
 
 33 
 
 Kew 
 
 S.S.W. | 
 
 3 to 4 a.m., 27th 
 4.7 a.m., 27th 
 
 32 
 59 miles per hour, in squalls. 
 
 Oxford 
 
 S.S.W. 
 
 3 to 4 a.m., 27th 
 
 34 miles in the hour. 
 
 Berkhamsted S. 
 
 2 to 3 a.m., 27th 
 
 23 
 
 But the main interest in this highly scientific study of a 
 historic tempest centres in the attempt made and not unsuc- 
 cessfully by Dr. Shaw to trace the actual paths of definite 
 masses of air in a travelling storm. For ease of distinction he 
 calls the actual path of the air a "trajectory," reserving the 
 use of the word " path " for the motion of the storm centre. 
 Dr. Shaw arrives at the following conclusions : 
 
 1. The more central area of the storm circulation is fed from 
 outside by winds passing to a region in front of the centre, with 
 directions in the quadrant between the direction from which the 
 storm comes and the direction at right angles thereto on the left 
 facing the coming storm. 
 
 2. When the storm centre has passed, corresponding winds 
 blow out from behind the centre with directions from the adjacent 
 quadrant on the other side of the path. 
 
 3. The winds from the remaining quarters will be compara- 
 tively transient in any locality, as the changes in them take place 
 with great rapidity. These winds are represented by the parts 
 of the loops of the curves above the path of the centre. 
 
 192 
 
292 METEOROLOGY 
 
 4. No air is taken into the storm area from the " northern " 
 side of the path. 
 
 5. There is a great convergence of winds behind the centre to 
 points in the line of the " trough/' This convergence is associated 
 with corresponding divergence in front of the trough, and apparent 
 crossing of the trajectories at the trough itself (Fig. 85). 
 
 Dr. Shaw adds that the convergence and divergence must have 
 reference to upward and downward convection. 
 
 The whole subject of the surface trajectories of moving air is 
 
 I IKITIM. POSITION OF CEHTKE MO PAITKLB -^ Z fUM. POSIT IM OF CENTRE MD PATICUS 
 
 r fwmon or runcus nxuwts^* ^JISTIEOUSLY im CIKIC 
 
 FIG. 85. DIAGRAM OF LOOPED TRAJECTORIES FOR AN "IDEAL" STORM OF CIRCULAR 
 ISOBARS AND UNIFORM WIND TANGENTIAL TO THE ISOBARS, TRAVKLLING WITH THE 
 SAME SPEED AS THE WIND. 
 
 discussed in detail by Dr. Shaw and Mr. R. G. K. Lempfert, M.A., 
 in the Life History of Surface Air Currents, published by the 
 authority ^of the Meteorological Committee in 1906 (M.O. 174). 
 From that publication is taken this diagram of looped trajectories 
 for an " ideal " storm of circular isobars and uniform wind tan- 
 gential to the isobars, travelling with the same speed as the wind. 
 The curves are represented by the equation : 
 
CHAPTER XX 
 ATMOSPHERIC ELECTRICITY 
 
 IN his Presidential Address to the British Association for the 
 Advancement of Science, delivered at Winnipeg, Manitoba, in 
 August, 1909, Professor Sir Joseph J. Thomson, M.A., LL.D., 
 D.Sc., F.R.S., speaks of " the great ocean of the ether, the 
 substance with which the whole universe is filled/' He goes on 
 to say : " The ether is not a fantastic creation of the speculative 
 philosopher ; it is as essential to us as the air we breathe. For 
 we must remember that we on this earth are not living on our 
 own resources ; we are dependent from minute to minute upon 
 what we are getting from the sun, and the gifts of the sun are 
 conveyed to us by the ether. It is to the sun that we owe, not 
 merely night and day, springtime and harvest, but it is the 
 energy of the sun, stored up in coal, in waterfalls, in food, that 
 practically does all the work of the world. 
 
 " How great is the supply the sun lavishes upon us becomes 
 clear when we consider that the heat received by the earth under 
 a high sun and a clear sky is equivalent, according to the measure- 
 ments of Langley, to about 7,000 horse-power per acre. Though 
 our engineers have not yet discovered how to utilise this enormous 
 supply of power, they will, I have not the slightest doubt, 
 ultimately succeed in doing so ; and when coal is exhausted and 
 our water-power inadequate, it may be that this is the source 
 from which we shall derive the energy necessary for the world's 
 work. When that comes about, our centres of industrial activity 
 may perhaps be transferred to the burning deserts of the Sahara, 
 and the value of land determined by its suitability for the 
 reception of traps to catch sunbeams. 
 
294 METEOROLOGY 
 
 " This energy, in the interval between its departure from the 
 sun and its arrival at the earth, must be in the space between 
 them. Thus this space must contain something which, like 
 ordinary matter, can store up energy, which can carry at an 
 enormous pace the energy associated with light and heat, and 
 which can, in addition, exert the enormous stresses necessary to 
 keep the earth circling round the sun and the moon round the 
 earth. 
 
 " The study of this all-pervading substance is perhaps the 
 most fascinating and important duty of the physicist. 
 
 " On the electro -magnetic theory of light, now universally 
 accepted, the energy streaming to the earth travels through the 
 ether in electric waves ; thus practically the whole of the energy 
 at our disposal has at one time or another been electrical energy. 
 The ether must, then, be the seat of electrical and magnetic 
 forces." 
 
 It is to the operation of these forces that the phenomena of 
 atmospheric electricity and of the aurora are due. 
 
 The identity of atmospheric electricity with that obtained from 
 an electrical machine, foreshadowed by Dr. Wall in 1708, was 
 finally proved by Benjamin Franklin in June, 1752. His classical 
 experiment of obtaining electricity from the clouds by flying a 
 kite need not be referred to in detail. In a letter to Mr. Peter 
 Collinson, F.K.S., dated Philadelphia, October, 1752, Franklin 
 describes his electric kite. 1 Suffice it to say that the experiment 
 is a dangerous one. In June, 1753, M. de Romas in France 
 repeated it, using a fine wire 550 feet long instead of a string, with 
 the result that he obtained flashes 9 or 10 feet in length, which 
 were accompanied by a loud report. On one occasion De Romas 
 was struck down, but not killed, by such a charge. In August of 
 the same year, Professor Richmann, of St. Petersburg, lost his life 
 during a thunderstorm. He approached the end of the conducting 
 wire, when a ball of fire apparently leaped to his head, killing him 
 on the spot. 
 
 Lemonnier proved, by means of insulated metal rods, that the 
 atmosphere is charged with electricity even in fine weather. 
 Volta and de Saussure, subsequently, each constructed an instru- 
 1 Philosophical Transactions, vol. xcv., p. 565. 1752. 
 
ATMOSPHERIC ELECTRICITY 295 
 
 merit, called an electroscope, for collecting atmospheric electricity 
 and demonstrating its effects. In de Saussure's electroscope the 
 electricity conducted from a rod to two little pith balls suspended 
 by fine wires in a glass case caused them to diverge from one 
 another. Alessandro Volta of Pavia substituted two blades of 
 straw an inch long for the pith balls. The most delicate of all 
 electroscopes is that which is called the " gold-leaf electroscope." 
 
 Before, however, I attempt to describe the chief electrical 
 phenomena in nature, it will be desirable to draw attention to 
 certain rudimentary facts relating to the nature of electricity. 
 These are admirably summarised in the following sentences taken 
 from an address on " Atmospheric Electricity," delivered before 
 the Royal Meteorological Society on March 21, 1888, by the late 
 Dr. W. Marcet, F.R.S., then President of the Society. 1 
 
 " Like heat," says Dr. Marcet, " electricity is the manifestation 
 of a peculiar condition of a body, and bodies are said to be 
 electrified when, after having been rubbed, or placed in com- 
 munication with an electrified object, they exercise an attraction 
 or repulsion, more or less great, upon other light bodies. On 
 inquiring into this phenomenon, it is found that the electricity 
 developed by friction from different substances such as glass on the 
 one hand, and sealing-wax on the other, is not identical, and that 
 there are consequently two electricities, varying from each other 
 in some of their characters. It may be said in a general way 
 
 " 1. That there is an attraction between an electrified body 
 and another not electrified. 
 
 " 2. That there is an attraction between two bodies electrified, 
 the one by a rubbed glass rod, the other by a rubbed stick of 
 sealing-wax, or of some resinous substance. 
 
 " 3. That there is a repulsion between two bodies, both of 
 which are electrified either by glass or by sealing-wax. 
 
 " To put these laws of Nature more clearly, electricities of 
 different kinds attract each other, and electricities of the same 
 kind repel each other. 
 
 " In order to distinguish between the two kinds of electricity, 
 the one is called vitreous electricity, from its being generated from 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. xiv., p. 197, 
 1888. 
 
296 METEOROLOGY 
 
 glass, and is known as positive, while the other is called resinous 
 electricity, and is known as negative." 
 
 Atmospheric electricity may, from time to time, reveal its 
 presence by very unequivocal phenomena, of which the chief 
 are (1) thunder and lightning, (2) hailstones, (3) aurora the 
 aurora borealis being peculiar to the northern hemisphere, the 
 aurora australis to the southern. The aurora is an electrical 
 phenomenon of cosmical origin. It is usually associated with 
 magnetic storms, and usually appears as a bright arch beneath 
 which the sky looks darker than in the surrounding regions. 
 Frequently streamers of white or rose coloured light shoot out in 
 long rays from the arch towards the zenith. Sometimes the 
 arch resembles a swaying sheet or curtain of light, or, again, 
 several arches may be seen simultaneously. 
 
 But apart from these manifestations, observations should be 
 made upon the electricity existing in the air under ordinary 
 circumstances, so as to determine, firstly, whether it is positive 
 or negative ; secondly, what is its intensity or tension. 
 
 In a Report on Atmospheric Electricity, drawn up at the 
 request of the Permanent Committee of the First International 
 Meteorological Congress at Vienna, and published by the 
 Meteorological Council in 1878, Professor J. D. Everett, M.A., 
 Queen's College, Belfast, observes that in discussions relating to 
 the electrical condition of the air at a specified point three things 
 must be carefully distinguished : electrical density, electrical 
 force, and electrical potential. 
 
 (a) The electrical density at a point in the air is the quantity 
 of electricity, per unit volume, with which the air at the point is 
 charged. 
 
 (b) The electrical force at a point is the force with which a unit 
 of positive electricity would be acted on, if brought to the point 
 without altering, by its inductive action, the previously existing 
 distribution. 
 
 (c) The electrical potential at a point is the work which would 
 be done by electrical force upon a unit of positive electricity 
 passing from the point to the earth, the movement of this unit 
 being supposed not to disturb the pre-existing distribution. 
 
 In Dr. R. H. Scott's Instructions in the Use of Meteorological 
 
ATMOSPHERIC ELECTRICITY 
 
 297 
 
 Instruments 1 concise information as to the apparatus used in 
 researches on atmospheric electricity will be found. From that 
 source chiefly the following is culled : 
 
 1. The electroscope is intended to show the nature or kind of 
 electricity present in the air. By far the most sensitive 
 instrument for this purpose 
 is the gold - leaf electroscope 
 (Fig. 86), in which electricity 
 collected from the neighbour- 
 ing atmosphere is made to act 
 through a metal rod, called a con- 
 ductor, upon two delicate gold 
 leaves suspended at the end of 
 the rod, and applied closely to 
 each other. The leaves, when 
 brought under the influence of 
 the same kind of electricity, 
 will diverge or repel each other. 
 As very little electricity can be 
 observed near the ground, the 
 conductor should be placed in 
 contact with the air at some 
 height above the earth's surface, 
 by means of a 
 
 2. Collector. This may be a 
 metallic arrow tied to one end 
 of a conducting string, and then 
 shot upwards into the air. The 
 electroscope will be found elec- 
 trified as the arrow mounts. A gilded fishing-rod may be sub- 
 stituted as a conductor, its lower end being insulated that is, 
 surrounded by a non-conductor such as caoutchouc. 
 
 Volta's collector is a flame burning at a height, either in a 
 lantern hung to a mast and connected with the electroscope by a 
 wire, or in the form of a slow-burning match attached to the top 
 of a long metal rod. The electricity of the air, in the neighbour- 
 hood of the flame, by its inductive action on the conductor, causes 
 1 Reprinted, p. 60 et seq., 1885. London : E. Stanford. 
 
 FIG. 86. GOLD-LEAF ELECTROSCOPE. 
 
298 METEOROLOGY 
 
 electricity of the opposite kind to accumulate at the upper 
 extremity, whence it is constantly carried off by the convection 
 currents in the flame, leaving the conductor charged with elec- 
 tricity of the same kind and potential as the air. 
 
 It is necessary again to explain that the term potential, as applied 
 to electricity, means the energy of an electrical charge measured 
 by its power to do work ; it is electro-motive force. " When one 
 body is charged with electricity to a higher potential than 
 another, electricity tends to pass off, so as to equalise the potential 
 on the two bodies " (R. H. Scott). 
 
 Dr. Scott says that " when we speak of the motion of the 
 electricity from one body to another, we say that this is effected 
 owing to difference of potential.'* He adds : " Difference of 
 electric potentials may very well be termed ' difference of electric 
 heights/ " 
 
 The water-dropping collector, invented by Sir William Thomson, 
 afterwards Lord Kelvin, Professor of Natural Philosophy in the 
 University of Glasgow, who died in 1907, is on the same principle 
 as Volta's method. A copper can is placed on an insulating 
 support, either of ebonite, with its surface thinly coated with 
 paraffin, or of glass surrounded with pumice-stone impregnated 
 with sulphuric acid. From the can a small pipe projects far into 
 the air, and terminates in a fine jet. The can being filled with 
 water, and the tap which opens into the jet being turned on, a 
 small stream of water is allowed to flow out guttatim, in drops. 
 In half a minute the can is found to be electrified to the same 
 extent and in the same way as the air at the point of the 
 tube. 
 
 This collector cannot be employed in frost, for the water 
 freezes in the jet. At such times the use of a slow-burning match, 
 made of blotting-paper, steeped in a solution of nitrate of lead, 
 dried and rolled, was recommended by Lord Kelvin. 
 
 Since electrical density is greater on projecting surfaces and 
 less on hollow surfaces than on planes, the collector should not be 
 near trees, or houses, or within a closed space. 
 
 3. Electrometer. This is an instrument which is intended to 
 measure the amount of electricity, or the electric intensity, 
 tension, or potential. The earliest electrometer was Coulomb's 
 
ATMOSPHERIC ELECTRICITY 299 
 
 Torsion Balance, by means of which one of the principal laws of 
 electricity was discovered that two electrified bodies, whose size 
 is very small in comparison with their distance apart, attract or 
 repel each other with a force proportional to the inverse square 
 of the distance which separates them. Lord Kelvin designed 
 two kinds of electrometer (1) the Quadrant Electrometer for 
 observatory use ; (2) the Portable Electrometer. 
 
 In the Quadrant, or modified Divided-Ring, Electrometer, a 
 needle of thin sheet aluminium, cut so as to resemble in form a 
 figure " 8 " with the hollows filled in, and carrying above it a 
 small light mirror weighing only a fraction of a grain, is suspended 
 from its centre by two fine silk threads, the distance between 
 which can be varied at will. The needle swings horizontally 
 inside a shallow cylindrical brass box, which is cut into four equal 
 segments or quadrants, each insulated separately by glass sup- 
 ports, but connected alternately by thin wires. Each pair of 
 quadrants is also connected to a stiff wire passing through the 
 case of the instruments, to form the two electrodes, or terminals, 
 for the attachment of the collecting and earth wires. 
 
 The base of the electrometer contains a Leyden jar, partially 
 filled with strong sulphuric acid, and a platinum wire, hung from 
 the lower surface of the needle, is made to dip into the acid. 
 
 A lamp and a divided scale are placed about a yard in front of 
 the instrument, and the light shining through an aperture in the 
 frame of the scale is reflected by the mirror on the scale, where the 
 position of-the image of a wire stretched across the hole can be 
 accurately observed. 
 
 In order to make use of this electrometer the needle must be 
 charged with electricity from a small electrophorus or electricity- 
 bearer, brought into contact with a wire (charging electrode) 
 dipping into the sulphuric acid at the bottom of the Leyden jar. 
 One of the electrodes connected with the segments is then joined 
 by a wire to the water-dropping collector ; the other is placed in 
 communication with the earth through a wire attached to a gas- 
 pipe, or similar conductor. The needle will then be deflected 
 towards either one side or the other, according as the electricity 
 of the atmosphere is of the nature to repel or attract it, and the 
 extent of repulsion, as measured on the scale, is proportional to 
 
300 METEOKOLOGY 
 
 the amount of difference of potential between the atmospheric and 
 terrestrial electricities. 
 
 To secure that the needle shall remain fully charged, an auxiliary 
 apparatus for the generation of electricity, termed a replenisher, 
 is fixed inside the case, by turning which the charge can be 
 restored to its original potential. This is indicated by a small 
 gauge consisting of a light lever, made of thin aluminium, and 
 fixed to the top of the instrument. One end of this gauge carries 
 an index which moves in front of a small scale. The other end is 
 flattened into a plate of about a square centimetre in area, which 
 is repelled by another plate, similarly electrified, fixed to the top 
 of the instrument, and in metallic connection with the sulphuric 
 acid of the Leyden jar, so as to be charged to the same potential 
 as the indicating needle. The position of the index being there- 
 fore once determined, it is easy, by giving a few turns to the 
 replenisher, at least once daily, to bring the potential of the 
 charge of the instrument up to its original value. 
 
 The scale value of each electrometer must be experimentally 
 determined by means of a galvanic battery of constant intensity 
 such as Daniell's. Knowing the electro -motive force of the cell 
 'in ployed in the battery, the indications of the electrometer scale 
 may be converted into terms of the absolute unit of electro- 
 motive force or " volts/' 
 
 If the electrometer is used as a self-recording instrument, a 
 drum carrying photographic paper, and maintained in rotation 
 at a uniform rate by a chain of clockwork, is substituted for the 
 divided scale, and the aperture is reduced so as to form a mere dot 
 of light on the cylinder. 
 
 Thomson's quadrant electrometer was used, in conjunction 
 with a water-dropping collector, for photographically recording 
 atmospheric electricity at the Kew Observatory from 1874 to 
 1885. It has been in use at the Royal Observatory, Greenwich, 
 since 1877. The instrument was described, with illustrations, in 
 the British Association Report for 1867, p. 489. 
 
 In Lord Kelvin's Portable Electrometer (Fig. 87) the electricity 
 is collected by means of a burning fuse at the extremity of a 
 vertical wire. An illustrated description of it is given in the 
 British Association Report for 1867, p. 501. 
 
ATMOSPHEKIC ELECTRICITY 
 
 301 
 
 Another electrometer which is highly recommended is Peltier's. 
 It was used for more than thirty years by the late M. Quetelet 
 at Brussels, and for upwards of twenty years at Utrecht. This 
 electrometer is described in the Annuaire M eteorologique de France, 
 1850, p. 181, and in the British Association Report, 1849 (" Trans- 
 actions of Sections/' p. 11). 
 
 Quetelet, according to Dr. K. H. Scott, drew the following con- 
 clusions from five years' observations with the electrometer at 
 Brussels : 
 
 1. The diurnal march of electricity, at a constant height above 
 the ground, exhibits two maxima and two minima. 
 
 2. The maxima and minima of electrical tension precede 
 by about an hour those of barometric pressure (see Chapter XIII., 
 p. 151 above). The maxima occur when temperature is either rising 
 or falling most rapidly 8 a.m. and 9 p.m. in summer, 10 a.m. 
 and 6 p.m. in winter. The 
 
 day minimum corresponds 
 with the period of maximal 
 temperature and minimal 
 humidity. The epoch of the 
 night minimum has not been 
 satisfactorily determined, but 
 was referred by de Saussure 
 and Schiibler to shortly before 
 daybreak. 
 
 3. The annual march of 
 electricity presents one maxi- 
 mum in summer (June) and 
 one minimum in winter 
 (January). 
 
 In nature the atmosphere, 
 whether clear or cloudy, 
 always shows an electric reaction it is in a state of electric tension, 
 which increases remarkably with altitude, as has been observed 
 by de Saussure, Erman, Quetelet, Lord Kelvin, and the Hon. 
 Ralph Abercromby (by the last named on the Peak of Teneriffe in 
 July and August, 1878). Lord Kelvin found in the island of 
 Arran, at a height of 9 feet above the ground, a difference of 
 
 FJO. 87. THOMSON'S PORTABLE ELECTROMETER. 
 
302 METEOKOLOGY 
 
 potential equal to 200 to 400 Daniell cells, or from 216 to 
 432 " volts " a volt being the standard of electro -motive force, 
 or " potential/' on the C.G.S. system of fundamental units of 
 mass, length, and time (namely, centimetre, gramme, and 
 second). This difference of potential, or electric pressure, repre- 
 sents a rise of potential of from 24 to 48 volts for each foot of 
 ascent. This is subject to great variations. With north and 
 north-east winds the potential was often six to ten times as much 
 as the higher of the amounts just given. The change of potential 
 is most rapid in cold, dry weather, when the quantity of moisture 
 in the air is at its lowest. 1 
 
 Under a clear sky atmospheric electricity is nearly always 
 positive. According to Peltier, land is always negative in its 
 electrical character, while Becquerel observed that sea-water is 
 always positive. M. de la Bive holds that the positive electricity 
 of the air is derived mainly from the sea. Clouds are in general 
 electrified, positively as a rule, but sometimes negatively. Nega- 
 tive clouds are supposed to result from ground fogs charged with 
 negative electricity, which they retain as they rise into the 
 atmosphere. Or they may become negatively electrified by 
 induction from superimposed positive clouds. In these facts we 
 gain a clue to the origin of thunderstorms and other atmospheric 
 phenomena of an electric character. 
 
 Professor Henry Mohn, Director of the Norske Meteorologiske 
 Institut, Christiania, has classified thunderstorms into two groups 
 Heat Thunderstorms and Cyclonic Thunderstorms. The former 
 type belongs to summer and to hot climates ; the latter to winter 
 and to insular climates. 
 
 Cyclonic Thunderstorms are so called because they accompany 
 deep atmospheric depressions such as traverse the North Atlantic 
 Ocean and the north-western seaboard of Europe, especially in 
 winter. Scarcely a gale of wind of any extreme intensity occurs 
 without attendant electrical phenomena. Occasional flashes of 
 sheet-lightning light up the sky over wide areas during the passage 
 of a winter storm, and here and there sharp hail and sleet squalls, 
 with a few vivid flashes and loud peals of thunder, are experienced. 
 
 1 GanoCs Physics, p. 1075. By Atkinson and Reinold. Sixteenth 
 edition. 1902. 
 
ATMOSPHEBIC ELECTRICITY 303 
 
 While these cyclonic thunderstorms are not so violent, they are 
 quite as dangerous as summer thunderstorms, if not more so, 
 because in them the clouds drift at a lower level, so that the 
 lightning is more likely to strike the ground. 
 
 Heat Thunderstorms are especially associated with sudden and 
 extreme alterations in atmospheric temperature. Perhaps a cool 
 night has been followed by a blazing sun and a light south or 
 south-east wind. Vast quantities of aqueous vapour rise in the 
 atmosphere as the result of rapid evaporation. Massive cumuli 
 form, the upper edges of which become more and more dense, and 
 appear snowy white as the sun shines upon them. These clouds 
 are probably surcharged with positive electricity. Then a light 
 surface current of air arises and blows towards the approaching 
 cumuli, while an angry -looking, lurid cloud stratum, negatively 
 electrified, forms in the lower strata of the air, and is seen con- 
 stantly to change its shape and density. Presently the top of the 
 piled-up cumuli spreads out into a dense cirriform sheet, and at 
 once thunder is heard, and rain or hail begins to fall in great 
 quantities. With flash after flash, peal upon peal, the storm 
 momentarily gathers strength and increases in violence. The hail 
 and rain fall intermittently in drenching showers, and the whole 
 sky becomes overcast, while the wind either falls light, dies down 
 to a calm, or shifts perhaps to the opposite point of the compass 
 in a fierce squall. The rain then lightens, the thunder and light- 
 ning become less frequent and more distant, and gradually the 
 sky clears and the air feels cool and fresh temperature perhaps 
 being 10, 15, or even 20 lower than before the storm. 
 
 From a careful analysis of the thunderstorms of 1888 and 1889 
 over the south and east of England, undertaken at the instance 
 of the Council of the Royal Meteorological Society, Mr. William 
 Marriott, F.R.Met.Soc., arrives at the conclusion that thunder- 
 storm formations are small atmospheric whirls in all respects 
 like ordinary cyclones. The whirl, which is most probably con- 
 fined to a stratum of air at only a short distance from the earth's 
 surface, not more than 4,000 to 6,000 feet, may vary from one 
 mile to ten miles or more in diameter. Thunderstorms usually 
 occur when the isobars show large areas of ill-defined low pressure 
 containing several shallow minima, called " thunderstorm de- 
 
304 METEOROLOGY 
 
 pressions," or when there is a " lane " or " trough " of low pressure 
 between adjoining areas of relatively high pressure. Mist and 
 fog often precede thunderstorms, which may accompany high 
 barometer readings as well as low. On May 21, 1888, there was a 
 thunderstorm with the barometer at 30-40 inches. On March 26, 
 1888, there was one with pressure below 29-00 inches. When 
 isobars are drawn for hundredths, instead of tenths, of an inch, a 
 number of small but distinct areas of low pressure, or cyclones, 
 with regular wind circulation, may generally be recognised in 
 thundery weather. In nearly all cases a sudden upward move- 
 ment of the barometer is noticed when a thunderstorm breaks out. 
 This increase of atmospheric pressure is usually well shown on 
 barographic tracings. 
 
 These " thunderstorm depressions " often circulate round a 
 large but shallow area of low barometer, forming secondary or 
 subsidiary depressions to it as a primary. At other times they 
 travel perhaps for hundreds of miles along a direct line, the rate 
 of progression being sometimes as much as fifty miles an hour. 
 On May 18-19, 1888, a storm passed across England from Christ- 
 church, Hants (8.15 p.m.), to Edinburgh (4 a.m.) and Cupar-Fife 
 (4.5 a.m.). Similarly, on June 2, 1889, a storm travelled north- 
 wards from Wiltshire (3 a.m.) to Edinburgh (10.44 a.m.), and 
 probably to Kirkwall, in the Orkneys (3.37 p.m.), a distance of 
 550 miles, at a uniform rate of 50 miles an hour. 
 
 Thunderstorms often break out over the same line of country 
 on consecutive days, and nothing is more curious than to watch 
 thunder-clouds springing into existence in the sky time after time 
 above a particular place or district, as if there was a direct 
 electrical attraction for the time being between the earth just 
 there and the superincumbent atmosphere, and this no doubt is 
 in reality the case. 
 
 Heat thunderstorms show a diurnal and an annual periodicity. 
 In tropical climates their periodicity is best marked. According 
 to Arago, Jamaica is peculiarly liable to thunderstorms. From 
 November to April the day breaks cloudless, but between 11 a.m. 
 and 1 p.m. the mountains of Port Royal become covered with 
 towering thunder-clouds. At the last-named hour rain falls in 
 torrents, lightning flashes in all directions, and the crash of 
 
ATMOSPHERIC ELECTRICITY 305 
 
 thunder is incessant and deafening. But the storm is quickly 
 spent, and a brilliant evening follows. It is an old observation 
 of Caldcleugh 1 that at Rio Janeiro it was customary to state in 
 invitations whether the guests were to assemble before or after 
 the thunderstorm, which was practically a daily episode. 
 
 Europe presents examples of both types of thunderstorms. 
 On the Continent and in England heat thunderstorms are preva- 
 lent in the summer months, the result being that the rainfall of 
 July and August is particularly heavy, if not excessive. In 
 Ireland, Scotland, and Norway heat thunderstorms are less 
 frequent, while cyclonic thunderstorms are apt to occur, especially 
 in the south-east quadrant of deep winter depressions. In Iceland 
 thunderstorms are almost unknown in summer, whereas they 
 frequently occur in winter. Dr. Alexander Buchan, in papers 
 contributed to the Scottish Meteorological Society on the 
 " Meteorology of Iceland " 2 and on the " Rainfall of Scotland/' 3 
 stated that during the twenty-five years 1846-1870 thirty-one 
 thunderstorms occurred at Stykkisholm, Iceland, in January, 
 seventeen in February, eight in March, six in April, two in May, 
 none in June, two in July, none in August, five in September, five 
 in October, fourteen in November, and twenty-five in December. 
 In the second of the two papers mentioned, Dr. Buchan shows that 
 the thunderstorms of the north-west of Scotland belong to the 
 cyclonic type, while those of the south-east are heat thunderstorms 
 for the most part. 
 
 England is celebrated for its thunderstorms in summer. They 
 are far more severe and far more frequent than those felt in 
 Ireland. This is due to the fact that the southerly winds in front 
 of the depressions, which the thunderstorms accompany, have 
 crossed the sea in the case of Ireland, but land that is, France 
 in the case of England. The contrast of temperature between 
 the south wind in front and the north wind behind the centre or 
 trough of low pressure is, therefore, much greater for England 
 than it is for Ireland. Again, the heated air rising over France 
 is negatively electrified, while the warm sea air in Ireland is 
 
 1 Quoted by Daniell in his Meteorological Essays and Observations. First 
 Edition, p. 335. 1823. 
 
 2 Journal of the Scottish Meteorological Society. New Series, vol. ii., p. 289. 
 1863. 3 Loc . ciL) vo i. ni^ p. 25L 1873. 
 
 20 
 
306 METEOROLOGY 
 
 positively electrified. There is, then, attraction between the 
 negative electricity of the ascending current and the positive 
 electricity of the atmosphere over England, whereas over Ireland 
 both the atmosphere and the ascending current of warm moist air 
 are positively electrified. Hence there is no attraction, and no 
 electrical energy is evoked. At the same time, it is true that 
 negative electricity forecasts rain, of which there is no lack in 
 Ireland. On the other hand, as Dr. Scott points out, a sudden 
 development of positive electricity in wet weather is a certain sign 
 of the sky clearing. 
 
 Electrical State of the Upper Atmosphere. An investigation into 
 this subject has been recently made at the Howard Estate 
 Observatory, Glossop, Derbyshire, by the observers, W. Makower, 
 Margaret White, and E. Marsden. The observatory stands at a 
 height of 335 metres (1,099 feet) above mean sea-level. The 
 results of a large series of experiments were communicated to the 
 Royal Meteorological Society by Mr. J. E. Petavel, F.R.S., 
 F.R.Met.Soc., on November 18, 1908. 1 As is well known, there 
 exists under normal atmospheric conditions a potential gradient 
 in the atmosphere surrounding the earth. The earth being nega- 
 tively charged with respect to the air, a continuous electric 
 current flows from the upper atmosphere to the earth's surface. 
 The magnitude of this current has been estimated at 2-2x 
 10- 16 amperes per cubic centimetre of the ground by Mr. C. T. R. 
 Wilson, F.R.S., 2 and at 24 x 10- 16 amperes per cubic centimetre 
 of the ground by H. Gerdien. 3 An ampere, or the unit of cur- 
 rent, is the current due to an electro-motive force of 1 volt 
 working through a resistance of 1 ohm the unit of resistance. 
 A kite attached to an earth-connected wire will tend to assume 
 the potential of the air surrounding it, and an electric current will 
 flow continuously down the wire to earth through the winding 
 machine with which the wire is connected. The experiments at 
 Glossop were undertaken with the view of determining the mag- 
 nitude of this current when the kite was at different heights 
 above the ground. 
 
 1 Quarterly Journal- of the Royal Meteorological Society, vol. xxxv., No. 149, 
 p. 7. January, 1909. 
 
 2 Proceedings of the Royal Society, vol. Ixxx., p. 537. 
 
 3 Physikalische Zeitschri/t, vol. vi. 1905. 
 
ATMOSPHERIC ELECTRICITY 307 
 
 The authors illustrate their communication with a series of 
 figures showing the strength of the electric current at various 
 specified heights expressed in amperes x 10- 5 . At heights 
 exceeding 3,000 feet the currents were so large that it was found 
 necessary to shunt the galvanometer with low resistances in order 
 to get readable deflections. In general a high value of the current 
 corresponded with a high velocity of the wind, a low value 
 of the current with a low wind-velocity. The wind therefore is 
 an important (though not the only) factor in causing large 
 fluctuations from day to day in the strength of the electric current 
 flowing to earth. 
 
 202 
 
CHAPTER XXI 
 ATMOSPHERIC ELECTRICITY (continued) 
 
 LIGHTNING, according to Professor Balfour Stewart, owes its 
 brilliancy to the generation of heat along the path of the electric 
 discharge so intense as to render the various constituents of the 
 air momentarily incandescent. This generation of heat is due to 
 the resistance of non-conductors in the air to the discharge which 
 takes place when clouds charged with different electricities 
 approach each other. 
 
 Thunder is the noise, or atmospheric vibrations, produced in 
 the first place by the tremendous expansion due to the heat of the 
 lightning flash, and then by an inrush of air to fill up the vacuum 
 so caused. Its prolonged reverberations are merely an acoustic 
 phenomenon an echo on a stupendous scale. When thunder is 
 heard close at hand, it sounds first like a volley of musketry, 
 because a separate report accompanies each zigzag movement on 
 the part of the flash which is pursuing its uneven, if rapid, paths 
 through masses of air of different conducting powers moist air 
 being a better conductor than dry air. As sound travels infinitely 
 less quickly than light, a flash which is a mile away will be seen 
 about five seconds before the thunder is heard. 
 
 Lightning may be described as of three kinds : (1) Zigzag or 
 forked lightning ; (2) diffused, summer, or sheet lightning ; 
 (3) globular or ball lightning. 
 
 Forked lightning does not occur in nature as drawn by artists. 
 We know this, thanks to Mr. James Nasmyth's observations, 
 communicated to the British Association in 1856. His state- 
 ments have been amply confirmed by sixty photographic repro- 
 ductions of lightning flashes, received by the Thunderstorm 
 Committee of the Royal Meteorological Society in 1888, in reply 
 
 308 
 
ATMOSPHERIC ELECTRICITY 
 
 309 
 
 to a circular sent out in June, 1887. The first report of that 
 Committee, drawn up by the Hon. Ralph Abercromby, 
 F.R.Met.Soc., goes to show that lightning assumes certain 
 
 typical forms (1) Stream lightning ; (2) luminous lightning ; 
 (3) ramified lightning ; (4) meandering lightning ; (5) beaded or 
 chapletted lightning ; (6) ribbon lightning. 
 
310 METEOKOLOGY 
 
 Flashes of forked lightning take place horizontally or vertically 
 between oppositely electrified clouds ; or vertically between the 
 negative earth and a positive cloud. The latter is very dan- 
 gerous. And even when a living object is not in the direct path 
 of the discharge, and so killed by the effect of the electricity on 
 the nervous centres or muscular system, death may ensue by 
 nduction from what is called the " return shock/' Suppose two 
 clouds of opposite electricities are hovering over the earth at no 
 great elevation. They will induce opposite electricities to their 
 own in objects on the ground beneath them. A discharge now 
 takes place between the clouds, establishing electrical equili- 
 brium so far as they are concerned. When this takes place, the 
 induced electricity in objects on the ground disappears, causing 
 such a nervous shock to living beings as to deprive them of life. 
 When a telephone gong sounds during a thunderstorm, it indi- 
 cates that a return shock is taking place. Fig. 88 shows a 
 remarkable flash of forked lightning with secondary streams, 
 which was cleverly photographed by Mr. F. Holmes, Castle Hill 
 Studio, Mere, Wiltshire, at 9.40 p.m. of Sunday, May 13, 1906. 
 
 Summer or sheet lightning (German, Wetterleuchten) is the 
 diffused flash of light which illuminates the horizon or the distant 
 clouds when a thunderstorm is raging at a great distance from 
 the observer, perhaps 100 or even 150 miles away far beyond the 
 limits (15, or at the most 20, miles) at which thunder is audible. 
 
 Globular or ball lightning " fire-balls " is more persistent 
 than forked lightning, remaining visible for several seconds, or 
 even as long as three minutes, as happened at Milan in 1841 
 (Arago). It shows itself as a luminous sphere or ball of fire, in 
 diameter varying from a few inches to 2 or 3 feet, which moves 
 slowly, and at last bursts with a loud report like a bomb-shell. 
 Dr. Scott, in his Elementary Meteorology, adduces several instances 
 of this rare form of lightning. 
 
 The destructive effects of lightning are twofold, mechanical 
 and combustible. If a flash of lightning strikes a sandy soil, it 
 fuses or vitrifies the silicious particles into a fulminary tube or 
 fulgurite (Latin, fulgur, flashing lightning). The German term 
 is very expressive Blitzrohren, lightning tubes. As a matter of 
 fact, there is no such thing as a thunderbolt. In a paper on 
 
ATMOSPHEKIC ELECTKICITY 311 
 
 " The Non-existence of Thunderbolts/' contributed by Mr. 
 G. J. Symons, F.K.S., to the Royal Meteorological Society, on 
 March 21, 1888, 1 the author effectually disposes of this myth. 
 
 Investigations made by Dr. Carl Miiller, and reported in Himmel 
 und Erde, show that lightning prefers to strike certain kinds of 
 trees. Under the direction of the Lippe-Detmold Department 
 of Forestry, statistics were gathered showing that in eleven years 
 lightning struck fifty-six oaks, three or four pines, twenty firs, but 
 not a single beech-tree, although seven-tenths of the trees were 
 beech. It would seem, then, that in a thunderstorm one is safer 
 under a beech-tree than under any other kind of tree. 2 The 
 Electrical Review, August 10, 1906, however, reports that, while 
 six men were sheltering under a beech-tree in the English Mid- 
 lands during a severe storm a few days previously, two were 
 killed and the others were struck down insensible. At the 
 inquest, the Coroner said he had specially examined the tree, as 
 for years he had read and understood that there was no record 
 of a beech-tree being struck by lightning. In this case the 
 lightning had not injured the tree to the extent of damaging a 
 leaf. The accident was probably due to the " return shock/' 
 
 The lightning flash moves with inconceivable velocity. Sir 
 Charles Wheatstone, by means of a rapidly revolving mirror, 
 showed that the duration of a spark, '1 inch in length, in air 
 at ordinary atmospheric pressure was about ^l^ second. 
 He also ascertained that its velocity along the insulated wire 
 with which he experimented was nearly 290,000 miles in a second 
 that is, half as great again as the velocity of light, 186,000 miles 
 in a second. 
 
 Dr. R. H. Scott acknowledges, in his Elementary Meteorology? 
 his indebtedness to M. de la Rue for the following calculation 
 of the potential, or electric pressure, necessary to produce a flash 
 of lightning a mile in length. By his and Dr. Miiller's experi- 
 ments (Philosophical Transactions, vol. clxix., p. 118) with his 
 magnificent battery, the striking distance, between points, when 
 11,000 cells were used, the potential of each being T06 " volts," 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. xiv., p. 208. 
 1888. 
 
 2 Symons's Meteorological Magazine, vol. xxxi.. p. 74. 1896. 
 
 3 P. 180. 
 
312 METEOROLOGY 
 
 was -62 inch. This striking distance varies with the square of 
 the number of cells employed. Then, as 1 mile =63,360 inches, 
 
 we have V-^^x 11,000 -3,516,480 cells, as the amount 
 
 requisite to produce such a flash. 
 
 St. Elmo's Fire a luminous electrical display is the Castor 
 and Pollux of the ancients. It is an induction phenomenon, and 
 occurs when an electrified cloud approaches a prominent or 
 pointed obstacle like the mast of a vessel, a flagstaff, a tree-top, 
 or a lightning-conductor. The electricity of the cloud and of the 
 earth combine, not in a flash of lightning, but more slowly and 
 continuously, so that a flame seems to rise from the projecting 
 point. Csesar noticed it after a hailstorm, and described it in 
 the words : " Eadem nocte legionis quintse cacumina sua sponte 
 arserunt." The phenomenon, according to Dr. Scott, is of the 
 nature of the " brush " discharge of the electrical machine. 
 It has received many names, such as " St. Elmo's Fire " and 
 u Comozants " a corruption of " corposants " from the Latin 
 corpus sanctum (Italian, corpo santo). Displays of St. Elmo's 
 Fire, accompanied by hissing or loud crackling sounds, are not 
 uncommon at great elevations : on the top of Ben Nevis, in the 
 Alps, and at the high-level observatory on the summit of Pike's 
 Peak in Colorado, United States of America. Pike's Peak, 
 14,151 feet above the sea, was until lately the highest observatory 
 in the world. 
 
 Colour of Lightning. On May 20, 1908, Mr. Spencer C. Russell, 
 F.R.Met.Soc., communicated to the Royal Meteorological Society 
 the results of observations he had made on the colour of lightning 
 at Epsom during the years 1903-1 907. 1 Forked lightning was 
 observed in fifty-seven storms, sheet lightning on seventy-eight 
 occasions. In the fifty-seven storms red alone occurred nine 
 times, red and blue eight times, and blue alone seven times. 
 Red and blue in combination with other colours occurred four- 
 teen times, red combined with colours other than blue six times, 
 and blue in combination with colours other than red once. White 
 was twice observed alone, and so was yellow. The greatest 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. xxxiv., 
 No. 148, p. 271. October, 1908. 
 
ATMOSPHERIC ELECTRICITY 313 
 
 number of colours seen during a single storm amounted to seven 
 namely, red, blue, violet, orange, yellow, white, and green. 
 This combination happened on three occasions. Cyclonic 
 thunderstorms in the winter months are accompanied by light- 
 ning either red or blue in colour, or in combination. As to sheet 
 lightning, red, yellow, and white alone were each seen on nine 
 occasions. Violet was seen on seven occasions, golden on six, 
 blue and orange on five. In sheet lightning green was the only 
 colour which was not observed alone. 
 
 Mr. Russell thinks that the various colours which lightning 
 assumes appear to depend on (1) The height of the storm-- 
 clouds ; (2) the electrical energy ; (3) the density of the air ; 
 (4) the moisture of the air ; (5) the presence of suspended matter 
 in the air ; and (6) the distance of the flash from the place of 
 observation. In the discussion which followed the reading of 
 Mr. Russell's paper, Mr. W. W. Bryant said that one would expect, 
 for physiological reasons, that, when two flashes occurred at 
 a very short interval, the second would be of the colour comple- 
 mentary to that of the first flash. Storms with frequent flashes 
 would in this way tend to come under Mr. Russell's red and blue 
 class of lightning. 
 
 Hail is intimately related to atmospheric electricity, as was 
 stated in Chapter XVIII. (pp. 255 and 259). Professor Colladon, of 
 Geneva, considers that the heavy rainfall preceding a hailstorm 
 causes a strong downward current of air, which induces by suction 
 a partial vacuum in the upper clouds. At once a fall of tempera- 
 ture takes place, accompanied by a sudden reduction of vapour 
 tension, which causes a condensation of moisture into frozen 
 particles. These are alternately attracted and repelled by in- 
 tensely electrified masses of cloud, thus increasing rapidly in size 
 until they become so heavy that they fall to the ground as hail- 
 stones. Mr. Colladon supposes that the cloud masses become 
 strongly electrified through the agency of lateral up-draughts 
 of air caused by the central down-draught from the shower of 
 rain preceding the hailstorm. It will be observed that this 
 theory closely agrees with the views of Volta as to the formation 
 of hail referred to in Chapter XVIII. (p. 255). 
 
 Mr. Russell, in discussing the colour of lightning, states that 
 
314 METEOROLOGY 
 
 the presence of hail in association with a thunderstorm seems to 
 be intimately connected with blue lightning. 
 
 Aurora. This electrical phenomenon is rarely seen in low 
 altitudes or at the Equator. It is a luminous appearance in 
 the northern sky (Aurora borealis), or in the southern sky (Aurora 
 australis), and assumes most frequently the aspect of an arch 
 of light above a " dark segment " of sky at right angles to the 
 magnetic meridian, from which arch bright rays or luminous 
 pillars, called streamers, shoot up with a wavy, quivering motion 
 towards the magnetic zenith. According to Professor Loomis, 
 in America the aurora borealis appears most frequently between 
 latitude 50 and latitude 62 north. In Europe and Asia the 
 auroral region is situated farther north than in America the 
 region of maximal frequency lying between the parallels of 66 
 and 75. In fact, the aurora is seen oftenest within an oval zone 
 surrounding the North Pole, the central line of which zone crosses 
 the meridian of Washington in latitude 56 north, and that of 
 St. Petersburg in latitude 71 north. Auroral displays are therefore 
 more common in America than in Europe. The shape of this 
 auroral zone bears some resemblance to the line of equal magnetic 
 dip, as well as to a " magnetic parallel " that is, a " line every- 
 where perpendicular to a magnetic meridian." Professor Loomis 
 thinks it probable that an auroral display round the North 
 Magnetic Pole of the earth is uniformly attended by a simul- 
 taneous display round the South Magnetic Pole. That a con- 
 nection exists between the aurora and terrestrial magnetism is 
 proved by the extreme agitation of the magnetic needle during 
 an auroral display. In a note, further, on the relation between 
 sun-spots and weather, 1 Dr. K. H. Scott points out that modern 
 observations show that the appearance of an unusually large 
 spot on the sun's surface is almost invariably accompanied by 
 a " magnetic storm " felt simultaneously in all parts of the 
 globe. When such magnetic storms occur, brilliant displays of 
 the aurora usually take place the aurora thus also exhibiting 
 a periodicity allied to that of sun-spots, which show epochs of 
 greatest frequency every ten or eleven years. The last epoch 
 of this kind fell in 1902-03. 
 
 1 Elementary Meteorology, Appendix V., p. 392. 1883. 
 
ATMOSPHERIC ELECTRICITY 315 
 
 In his presidential address on " Atmospheric Electricity/' 
 delivered at the meeting of the Royal Meteorological Society, 
 March 21, 1888, 1 Dr. W. Marcet, F.R.S., says that the aurora 
 borealis is now generally considered to be due to positive elec- 
 tricity from the sea between the tropics being carried into the 
 upper atmospheric regions, and thence wafted to the poles by 
 the higher aerial currents. In the vicinity of the poles it descends 
 towards the earth and meets the terrestrial negative electricity 
 in a rarefied atmosphere. Luminous discharges then take place, 
 their brightness being increased by the presence of masses of 
 ice-particles in the atmosphere. These phenomena occur in the 
 neighbourhood of what are called the North and South Magnetic 
 Poles, and from this circumstance assume the form of bands so 
 peculiar to these aurorae. 
 
 Researches carried out by Messrs, de la Rue and Miiller 2 
 go to prove that the aurora may appear at any height between 
 a few thousand feet and 80 to 100 miles. Experiments made 
 with M. de la Rue's battery of 11,000 cells established the interest- 
 ing fact that the colour of the discharge with the same potential 
 varied with, and was apparently determined by, the tenuity of 
 the gas or air. The authors give the following Table of the 
 pressures at which they actually obtained discharges, represented 
 by the corresponding calculated heights, and also of the tints 
 at each height : 
 
 He'ghtin Mile.- 
 
 Ti.t. 
 
 Height in Miks. 
 
 Tint. 
 
 81-47 Pale and faint 
 
 27-42 
 
 Carmine 
 
 37-67 Maximal brilliancy 17 '80 
 
 33-96 Pale salmon 12'42 
 
 32-87 Salmon-coloured 11'58 Full red 
 
 30-86 
 
 
 The roseate and salmon-coloured tints are always near the 
 positive source of the electric or magnetic current. The dis- 
 charge at the negative terminal, in air, is always of a violet hue, 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. xiv., p. 207. 
 1888. 
 
 2 Proceedings of the Royal Society, vol. xxx., p. 332. 
 
316 METEOROLOGY 
 
 and, accordingly, this tint in the aurora indicates the proximity 
 of the negative source. 
 
 On Saturday, September 25, 1909, a magnetic storm of un- 
 usual intensity swept over the whole earth. Dr. Charles Chree, 
 the Superintendent of the Observatory Department of the 
 National Physical Laboratory at Richmond, Surrey, states that 
 the records at the Observatory show that the disturbance began 
 at 11.22 a.m. on Saturday, September 25, that the magnetic 
 conditions of the earth remained highly disturbed until 8.30 p.m., 
 and that the disturbance had almost disappeared at 1 a.m. on 
 Sunday, September 26. The most remarkable feature of the 
 disturbance was its extremely oscillatory character a direct 
 evidence that there was aurora well outside the Arctic regions. 
 Clouds interfered with observations over the greater part of the 
 British Isles, but a brilliant display of aurora borealis was reported 
 from Russia, Switzerland, and Northern Italy, while an equally 
 brilliant aurora australis was seen throughout Australia and in 
 South Africa. 
 
 In reply to inquiries as to the magnetic storm from the Times 
 Correspondent in Birmingham, Sir Oliver Lodge, F.R.S., made 
 the following statement : 
 
 " A cosmic electromagnetic disturbance, such as the earth 
 experienced on Saturday, is now believed to be due to solar 
 radioactivity. For in addition to its ordinary radiation on which 
 the earth entirely depends, the sun is at times technically radio- 
 active, and the eruption not only produces sun spots, but also 
 expels crowds of electrons, which fly with prodigious speed in 
 straight lines after the manner of the Beta rays in radium. 
 Whenever a torrent of these minute electrified projectiles rush 
 past the earth, as they do at the rate of some thousand miles a 
 second, they constitute a powerful electric current, and are 
 liable to deflect magnetic needles. 
 
 " Some of them, however, as in the recent case, actually 
 encounter the earth's atmosphere, and though they are mostly 
 deflected to the poles, some of them, especially at the times of 
 the equinox, may come down near the Equator. Those which 
 journey to the poles are accompanied by an opposite current in 
 the crust of the earth from the Equator to the poles, and this it is 
 
ATMOSPHEKIC ELECTKICITY 317 
 
 which"*disturbs the telegraphs, being picked up or tapped by 
 them en route. They also produce auroras in the neighbourhood 
 of the poles. 
 
 " Those which enter the atmosphere elsewhere act as nuclei 
 f or condensation of moisture, and by screening the sun's rays are 
 probably responsible for some of the dull and overcast weather. 
 Local thunderstorms are also a not unlikely result. 
 
 " These atmospheric conditions might be mitigated, at least 
 so far as the dull weather is concerned, by artificial supplies of 
 positive electricity to the upper atmosphere in large quantities ; 
 but no one has as yet thought the experiment worth trying on a 
 sufficient scale. 
 
 " There is no remedy for the magnetic storms due to cosmic 
 causes, nor for the corresponding earth-currents, but telegraphic 
 disturbance can be eliminated by the use of double lines or 
 return wires." 
 
 Lightning-Conductors. B. Franklin devised the lightning-rod, 
 or lightning-conductor, which is now universally adopted. The 
 principle of the lightning-conductor (French, paratonnerre ; 
 German, Blitzableiter) is that electricity selects the better of 
 two conducting passages, and that when it has got a sufficient 
 conducting passage, it is disarmed of all destructive energy. 
 A lightning-conductor is a metallic rod, usually of galvanised 
 iron or of copper, which terminates above in one or more sharp 
 points, and below in moist earth or in a sufficient expanse of 
 water. This metallic rod, when placed on a building or on the 
 mast of a vessel, protects it by affording a ready passage to the 
 electricity of the earth into the atmosphere, so establishing electric 
 equilibrium gradually and silently. As Dr. R. H. Scott well 
 observes, 1 " The action depends on what is called the ' power of 
 points/ The electricity on a sphere is uniformly distributed 
 over the surface ; on an oval figure it tends to accumulate at the 
 ends. On a cylinder this tendency is more strongly developed, 
 and, when the cylinder becomes a fine wire, the tension is so great 
 at the end that the electricity soon forces its way into the sur- 
 rounding air and escapes." 
 
 A lightning-conductor consists of three parts the pointed 
 1 Elementary Meteorology, p. 183. 
 
318 METEOROLOGY 
 
 rod, overtopping the building ; the conductor, or part connecting 
 the top with the ground ; and the part in the ground. In a very 
 able paper, entitled " Remarks on some Practical Points con- 
 nected with the Construction of Lightning-Co nductors," 1 the 
 late Dr. Robert J. Mann, F.R.A.S., at the time (1875) President 
 of the British Meteorological Society, laid down the indispensable 
 conditions for an efficient lightning-conductor as follows : (1) The 
 lightning-conductor must be made of good conducting material, 
 metallically continuous from summit to base, and of a dimension 
 which is sufficient for the ready and free conveyance of the largest 
 discharge that can possibly have to pass through it. (2) It must 
 have ample earth-contacts, and these contacts must be examined 
 frequently, to prove that they are not getting gradually impaired 
 through the operation of chemical and electrical erosion. (3) It 
 must terminate above in well-formed and well-arranged points, 
 which are fixed and distributed with some definite regard to the 
 size, form, and plan of the building. (4) There must be no part 
 of the building, whether it be of metal or of less readily conduct- 
 ing material, which comes near to the limiting surface of a 
 conical space, having the highest point of the conductor for its 
 apex, and having a base twice as wide as the lightning-conductor 
 is high, without having a point projecting out some little distance 
 beyond, and made part of the general conducting line of the 
 lightning-rod by a communication with it beneath. (5) There 
 must be no mass of conducting metal, and, above all things, 
 no gas-pipe, connected with the main, within striking distance of 
 the lightning-rod, lest at any time either the points or the earth- 
 contacts shall have been so far deranged or impaired as to leave 
 it possible for discharges of high tension, instead of continuous 
 streams of low tension, to pass through the rod, and to be diverted 
 from it into such undesigned routes of escape. 
 
 The Royal Meteorological Society many years ago organised 
 a Conference of delegates from various scientific and professional 
 societies to examine into the whole question of lightning-con- 
 ductors, and the Conference drew up a code of rules 2 very much 
 on the lines indicated by Dr. Mann in his paper published in 
 1875. The chief matters to be attended to are these : (1) The 
 
 1 Quarterly Journal of theRoyal Meteorological Society, vol. ii., p. 417. 1875. 
 
 2 Report of the Lightning-rod Conference, p. 16. London : Spon. 1882. 
 
ATMOSPHERIC ELECTRICITY 319 
 
 point of the upper terminal should not be sharp, not sharper 
 than a cone of which the height is equal to the radius of its base. 
 All points should be platinised, gilded, or nickel-plated, so as to 
 resist oxidation. (2) Rods should be taken down the side of 
 the building most exposed to rain, and held firmly, but not too 
 tightly, by holdfasts, so as to allow of contraction and expansion 
 by changes of temperature. (3) The rod should consist of copper, 
 weighing not less than 6 ounces per foot-run, and the conduc- 
 tivity of which is not less than 90 per cent, of that of pure copper, 
 either in the form of tape or rope of stout wires, no individual 
 wire being less than No. 12 B.W.G. Iron may be used, but 
 should not weigh less than 2J pounds per foot-run. (4) Although 
 electricity of high tension will jump across bad joints, they 
 diminish the efficacy of the conductor ; therefore, every joint, 
 besides being well cleaned, screwed, scarfed, or riveted, should 
 be thoroughly soldered. (5) Iron rods should be painted, whether 
 galvanised or not. (6) The rod should not be bent abruptly 
 round sharp corners. (7) As far as practicable, it is desirable 
 that the conductor should be connected to extensive masses of 
 metal, such as hot-water pipes, both inside and outside the 
 building ; but it should be kept away from all soft metal pipes, 
 and from internal gas-pipes of every kind. (8) It is essential 
 that the lower extremity of the conductor should be buried in 
 permanently damp soil, hence proximity to rain-water pipes 
 and to drains is desirable. 
 
 It is a very good plan to make the conductor bifurcate close 
 below the surface of the ground, and adopt two of the following 
 methods for securing the escape of the lightning into the earth : 
 A strip of copper tape may be led from the bottom of the rod to 
 the nearest water main, not merely to a lead pipe, and be 
 soldered to it ; or a tape may be soldered to a sheet of copper 
 3 feet x 3 feet, and -g- inch - thick, buried in permanently wet 
 earth, and surrounded by cinders or coke ; or many yards of 
 the tape may be laid in a trench rilled with coke, taking care 
 that the surfaces of copper are, as in the previous cases, not less 
 than 18 square feet. Where iron is used for the rod, a galvanised 
 iron plate of similar dimensions should be employed. 
 
 A lightning-conductor should be kept away from a gas-pipe, 
 because if there was any defect in the connections, an electric 
 
320 METEOROLOGY 
 
 discharge of high potential might cause a spark and so ignite the 
 gas. 
 
 The Brontometer. In 1890 MM. Richard Freres, of Paris, acting 
 under instructions given by the late Mr. G. J. Symons, F.R.S., 
 made for him an elaborate apparatus, which he called a " Bronto- 
 meter/' or thunderstorm measurer (Greek, fipovrr), thunder). 
 A full description of the instrument by Mr. Symons will be found 
 in the Proceedings of the Royal Society (vol. xlviii., 1890, p. 65). 
 The traces are made in ink, by a series of Richard pens, on endless 
 paper, 12 inches wide, travelling under the various recording 
 pens at the rate of 1 '2 inches per minute, or 6 feet per hour. The 
 first pen records the time, minute by minute. It produces a 
 straight line for fifty-five seconds ; then begins to go, at an angle 
 of about 45, ~ inch to the left, and at the sixtieth second it 
 flies back to its original position. This movement enables the 
 time of any phenomenon to be read off with certainty to a 
 single second of time. The second pen is driven by one of 
 Richard's anemo-cinemographs (see p. 285), and gives a tracing 
 of the wind's force. The third pen, actuated by a handle, indi- 
 cates the intensity of the rainfall. This movement should be 
 made automatic if possible. The fourth pen is actuated some- 
 what like a piano. On the occurrence of a flash of lightning, the 
 observer presses a key, the pen travels slightly to the right, 
 and flies back to zero. The fifth pen works in a similar way ; 
 but, as it is intended to record the thunder, the observer will 
 continue to hold down the key until the roll is inaudible. Re- 
 ferred to the automatic time-scale, the instant at which these 
 keys are depressed is given to the second. The sixth pen, similar 
 to the third, is intended to record the time, duration, and in- 
 tensity of hail. The seventh and last pen is devoted to an 
 automatic record of atmospheric pressure. 
 
 Owing to the rapid motion of the paper, which is indispensable 
 for studying the details of a thunderstorm, it was imperative 
 that the barometer scale should itself be greatly enlarged. To 
 meet this difficulty, a modification of an ingenious instrument 
 of Richard, called a " statoscope," was adopted. This measurer 
 of air-pressure "is so sensitive that it will indicate the opening 
 or shutting of a door in any part of a house, gives a scale for 
 30 inches for each mercurial inch (i.e., about three times that of 
 
ATMOSPHERIC ELECTRICITY 321 
 
 a glycerine barometer), and yet requires only 4 inches breadth 
 of the brontometer paper/' This wonderful apparatus records 
 accurately '001 inch of mercurial barometric pressure. 
 
 Mr. William Marriott, F.R.Met.Soc., has used Symons's bronto- 
 meter, and the records taken at his residence, West Norwood, 
 London, S.E., during a very severe thunderstorm, on June 4, 
 1908, are reproduced in a diagram which appears in the number 
 of the Quarterly Journal of the Royal Meteorological Society for 
 July, 1908 (vol. xxxiv., No. 147, p. 211). The International 
 Meteorological Conference which took place at Innsbruck in 
 September, 1905, expressed the opinion that autographic thunder- 
 storm recorders that is, brontometers are still in the experi- 
 mental stage, and it consequently could not recommend the 
 general adoption of these instruments at observatories. 
 
 Ozone. In 1785 the Dutch physicist Van Marum, while 
 passing a succession of electric sparks from a powerful electric 
 machine through a tube containing oxygen, was attracted by a 
 peculiar odour which developed in the oxygen, and which he 
 attributed to the " electric matter," calling it accordingly the 
 " smell of electricity." In 1840 M. Schonbein, of Basle, named 
 the substance which gave rise to this odour ozone, from the 
 Greek 6'w, I have a smell. His views as to the exact nature of 
 ozone passed through several phases. At first he thought it 
 was an element analogous to the halogen group chlorine, 
 bromine, iodine, and fluorine. Then he considered the possi- 
 bility of its being a constituent of nitrogen, or a higher oxide of 
 hydrogen, because he found that it could also be produced by 
 the action of phosphorus on moist air (1845). Lastly, in 1852, 
 he came to the conclusion, with other observers, de la Rive and 
 Marignac, that ozone is really oxygen in an allotropic state, 
 just as diamond is an allotropic form of carbon, coke or charcoal 
 being the usual forms of this latter element. 
 
 This view was fully confirmed by the experimental researches 
 of Dr. Andrews in 1856. He proved that ozone was really an 
 allotropic form of oxygen, and that it was identical in its nature 
 by whatever process it was prepared. Andrews also demon- 
 strated that ozone can be turned back into oxygen by exposing 
 it to high temperatures (300 C.). 1 
 
 1 Philosophical Transactions of the Royal Society, 1856. 
 
 21 
 
322 METEOROLOGY 
 
 In 1858 Schonbein started a new and plausible hypothesis. 
 He announced that ordinary oxygen was a neutral combination 
 of two oppositely electrical and therefore very active bodies. 
 One of these was ozone, or negative oxygen., which was formed 
 during the electrisation of oxygen or air, the electrolysis of water, 
 the slow oxidation of phosphorus, and the decomposition of most 
 metallic peroxides. He called those substances which evolved 
 this kind of oxygen ozonides. The other, the positive oxygen, or 
 antozone, he failed to isolate, but he assumed its existence in a 
 certain class of peroxides, which he called antozonides, examples 
 being the peroxides of hydrogen and barium in particular, also 
 the so-called ozonised turpentine, cod-liver oil, and ether. This 
 ingenious theory was demolished in 1863 by Sir Benjamin C. 
 Brodie. 
 
 In 1860 Andrews and Tait presented a very important com- 
 munication on the subject of ozone to the Royal Society. These 
 observers, in the first place, confirmed a previously known fact, 
 that only a small proportion of oxygen (one-twelfth, at most) 
 can be converted into ozone by the electric discharge. But they 
 also found that a constant and considerable diminution of volume 
 accompanied the change one hundred volumes of oxygen, 
 subjected to the silent discharge, contracting to ninety- two 
 volumes. Hence ozone must be denser than oxygen. Nor was 
 this all, for when mercury, or some other oxidisable substance, 
 was introduced into the ozonised oxygen, and the ozone entirely 
 absorbed, the residual oxygen had precisely the same volume 
 as it had before the removal of the ozone, so that the density of 
 ozone appeared to be absolutely infinite. 
 
 Dr. Odling suggested that the formation of ozone might really 
 consist in the condensation of another atom of oxygen into 
 each diatomic or dyad molecule of ordinary oxygen. The 
 chemical formula for free oxygen being 2 , that for ozone would 
 therefore be 3 ; and the density of ozone would be one-half 
 greater than that of oxygen. When one hundred volumes of 
 oxygen were reduced by ozonisation to ninety-two, it might be 
 supposed that eight volumes of oxygen combined with sixteen 
 volumes of oxygen to produce sixteen volumes of ozone. The 
 reaction might be represented in this way : 
 80^+160, =160^, 
 
ATMOSPHERIC ELECTRICITY 323 
 
 The absorption of the ozone by the mercury, or iodine, etc., 
 might really be only the removal of the third atom of oxygen, 
 which would, of course, leave the volume unaltered 
 80,+ 8Hg=8(X+8HgO. 
 
 The same view would account for the mutual reduction which 
 ozone and hydrogen dioxide exercise upon one another, and, in 
 fact, for all known reactions of ozone 
 
 ;! +H,0 2 ( = hydrogen dioxide) = 2(X+ H,0 ( = water). 
 
 This beautiful hypothesis received a remarkable experimental 
 verification at the hands of M. Soret, 1 who succeeded in finding 
 a body, oil of turpentine, which, instead of removing only one 
 atom of oxygen from each molecule of ozone, as most substances 
 do, absorbs the entire molecule, the whole three atoms of oxygen. 
 To take our previous illustration, if the ninety-two volumes of 
 ozonised oxygen were treated with oil of turpentine, a dense 
 white cloud would appear, the ozone would vanish, but instead 
 of the volume remaining the same, it would contract to seventy- 
 six volumes, the 160 3 having been removed bodily, instead of 
 being merely reduced to 160 2 . 2 
 
 A great number of experiments have given the atomic weight 
 of ozone as twenty-four, and consequently its molecular weight 
 as forty-eight. This is just the weight of three atoms of oxygen 
 namely, 16 x 3 =48, of which, accordingly, it is now universally 
 believed to be composed. 
 
 Mr. Francis E. Twemlow, F.M.S., in a paper on ozone, read 
 before the British Meteorological Society on March 17, 1875, 
 indicates the chief points of difference between ozone and 
 ordinary oxygen : 
 
 1. It liberates iodine from iodide of potassium. 
 
 2. It oxidises rapidly the precious metals. 
 
 3. It destroys vegetable colours. 
 
 4. It possesses a remarkable smell, like weak chlorine, whilst 
 
 oxygen is odourless. 
 
 Ozone bleaches most vegetable colours, and is a strong oxidiser 
 and so it may be called " nascent," or " active," oxygen. The 
 
 1 Comj,tes Rendus, November 27, 1865. 
 
 2 Medical Times and Gazette, pp. 383, 384. October 5, 1867. 
 
 21 2 
 
324 METEOROLOGY 
 
 latter fact probably affords the real clue to the supposed con- 
 nection between an absence of ozone in the atmosphere and out- 
 breaks of cholera, dysentery, and other like diseases. Their 
 relation attracted the attention of M. Quetelet in Belgium, and 
 of M. Andrand, of Paris, in the cholera outbreak of 1849. Accord- 
 ing to Glaisher, Moffat, Hunt, and others, the occurrence of 
 cholera and choleraic diarrhoea is coincident with an absence 
 or diminution of ozone, and their departure with a return of 
 ozone (C. B. Fox, M.D. 1 ). On the other hand, Dr. Moffat remarks 
 that " the prevalence of influenza, and the spread of catarrhal 
 affections, are invariably connected with an excess of ozone in 
 the atmosphere." The fact is that ozone irritates the mucous 
 membranes, and so has been credited with producing epidemic 
 catarrh. 
 
 Although colourless in its gaseous form, ozone appears as a 
 blue fluid when liquefied by cold and pressure. It decomposes 
 a solution of iodide of potassium and sets free iodine, which 
 gives a blue coloration when brought into contact with a solution 
 of starch. Ozone test-papers are strips of blotting-paper or 
 filter-paper steeped in a solution of potassic iodide and starch. 
 When moistened and exposed to an ozone-laden atmosphere, 
 such test-papers turn blue. 
 
 Testing for ozone is even still in an unsatisfactory state. Any 
 other oxidising agent in the air, such as hydrogen dioxide or 
 nitric acid, will reduce the potassic iodide, so that the test given 
 above is open to serious fallacy. Dr. Cornelius B. Fox, who has 
 paid particular attention to this question in his work on Ozone 
 and Antozone, 2 says that, if we wish to ascertain the amount of 
 ozone present in the air to the exclusion of the other air purifiers, 
 we employ a paper which is alone acted upon by ozone, such as 
 the iodised litmus paper. With this test, we do not take any 
 notice of the amount of iodine set free, but we observe the amount 
 of potash formed by the union of the ozone with the potassium. 
 Potash, being an alkali, has the property of turning red litmus 
 blue. The greater or less conversion of the red litmus into blue 
 shows a greater or less quantity of ozone in the air. 
 
 J Ozone and Antozone. London : J. and A. Churchill. 1873. 
 2 P. 168. London : J. and A. Churchill. 1873. 
 
CHAPTER XXII 
 INVESTIGATION OF THE UPPER ATMOSPHERE 
 
 AT the meeting of the British Association for the Advancement of 
 Science held at Winnipeg, Manitoba, Canada, in August, 1909, 
 an important Report was presented by a Committee consisting of 
 Mr. E. Gold and Mr. W. A. Harwood, on the present state of our 
 knowledge of the Upper Atmosphere as obtained by the use of 
 kites, balloons, and pilot balloons. This Report is the most 
 recent comprehensive scientific contribution to the literature of 
 the subject. 
 
 In a letter to the editor of Symons's Meteorological Magazine, 
 dated July 4, 1896, Mr. A. Lawrence Rotch, F.R.Met.Soc., 
 Director of the Blue Hill Meteorological Observatory, U.S.A., 
 points out that, as far as he knows, kites were first used for 
 meteorological purposes in the year 1749. The credit for intro- 
 ducing this method of investigation belongs to Dr. Alexander 
 Wilson, Professor of Practical Astronomy at Glasgow, who in 
 July of that year, with a student named Thomas Melville, explored 
 the temperature of the atmosphere in the higher regions at 
 Camlachie, near Glasgow, by raising a number of paper kites, 
 4 to 7 feet in height, one above another upon the same line, with 
 thermometers attached to those which were to be most elevated. 1 
 
 In 1752 Benjamin Franklin, in his historic experiment, obtained 
 electrical discharges from a thunder-cloud by means of a cord 
 carried up to the cloud by a kite. 
 
 During the winter of 1822-23 the Rev. George Fisher and 
 Captain Sir Edward Parry, at the island of Igloolik, in the 
 Arctic regions (lat. 69 21' N., long. 81 42' W.), obtained tern- 
 
 1 Cf. " The Biography of Alexander Wilson, M.D.," in the Trans. Roy 
 Soc. Edin., vol. x., pt. ii., pp. 284-286. 
 
 325 
 
326 METEOROLOGY 
 
 peratures in the upper air by means of self -registering thermo- 
 meters attached to kites. 1 
 
 In 1883 Mr. E. -Douglas Archibald, at Tunbridge Wells, raised 
 a Biram's anemometer by means of tandem kites. Four of these 
 anemometers were suspended at different levels. They registered 
 on dials the total wind movement during the time they were 
 suspended. 2 Mr. Archibald appears to have been the first person 
 to fly kites with steel wire, although copper wire was so used by 
 Robert Stevenson when a boy. 
 
 At the instigation of the British Association for the Advance- 
 ment of Science, Mr. James Glaisher, F.R.S., shortly after the 
 middle of the nineteenth century, explored the upper regions of 
 the atmosphere by means of balloons. This work was spread 
 over a period of five years, and embraced about thirty ascents, 
 the most remarkable of which was also one of the earliest. His 
 historic balloon ascent was made from Wolverhampton on Sep- 
 tember 5, 1862, when a height estimated at about 7 miles was 
 reached, and both Glaisher and his aeronaut, Coxwell, narrowly 
 escaped the loss of their lives. Subsequently these " free " ascents 
 were supplemented by others, made with a large " captive " 
 balloon at Chelsea. Glaisher afterwards wrote the article on 
 Aeronautics for the ninth edition of the Encyclopaedia Britannica, 
 and he edited English translations of works on the subject by 
 th3 French writers, MM. Tissandier and Flammarion. An 
 article from his pen, on " The Variation of Temperature with 
 Altitude in the Neighbourhood of the Ground/' was published in 
 the Comptes Rendus of the Paris Academy of Sciences, and in 
 Nature in 1877. In an obituary notice in Symons's Meteorological 
 Magazine for April, 1903, from which the above particulars are 
 taken, the author, Mr. R. H. Curtis, reminds us that Mr. Glaisher 
 was born so long ago as 1809, and at the time of his death, on 
 February 7, 1903, he was within a couple of months of completing 
 his ninety-fourth year. 
 
 In 1898 M. Teisserenc de Bort equipped a kite-station at the 
 
 1 Symons's Meteorological Magazine, vol. xxxii., April, 1897, and article, 
 " Meteorology," by G. Harvey, F.R.S., in the Encyclopedia Metropolitana, 
 1834, p. 73. 
 
 2 Quarterly Journal of the Royal Meteorological Society, 1883, vol. ix., 
 p. 62 ; and Nature, vol. xxxi., p. C6, and vol. xxxiii., p. 593. Also cf. British 
 Association Report, 1884, p. 639 ; ibid., 1885. 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 327 
 
 Observatory of Trappes, near Paris. Four years later (1902) 
 kite experiments were made by Mr. W. H. Dines on land, and also 
 over the sea, from a small steam-vessel, on the west coast of 
 Scotland. In 1907 kite-stations were established in Egypt and 
 at Glossop, Derbyshire. 
 
 The International Commission of Scientific Aeronautics, at a 
 meeting at Milan in 1906, resolved, on the recommendation of 
 M. Leon Teisserenc de Bort, to carry on, during the years 1907 
 and 1908, the investigation of the upper atmosphere in the 
 Northern Hemisphere on a much more extended scale than had 
 hitherto been attempted. The Royal Meteorological Society was 
 invited, and agreed, to take part in the scheme, and the matter was 
 placed in the hands of the Kite Committee, consisting of the 
 President and Secretaries of the Society, Colonel J. E. Capper, 
 R.E., Mr. Richard H. Curtis, and Captain Hepworth. This Com- 
 mittee co-operated with a Committee of the British Association, 
 and with the Meteorological Office. 
 
 In connection with the investigation of the upper air by the 
 Kite Committee, balloons (ballons-sondes) which carried self- 
 recording instruments, and also smaller balloons, were used, the 
 heights and drifts of which were determined by two theodolites 
 placed at the ends of a fixed base. 
 
 A thermograph of Richard's pattern, constructed of aluminium 
 by Mr. Fergusson, of the Blue Hill Observatory, was for the first 
 time raised by kites from Blue Hill in 1894 ; and during the follow- 
 ing year meteorographs, recording several elements, were there 
 lifted by the same method. The Blue Hill Observatory was 
 founded and maintained by Professor Rotch, and much of the 
 experimental work was carried on by H. H. Clayton. 
 
 When describing the apparatus and instruments employed in 
 ascents of balloons and kites, Messrs. Gold and Harwood observe 1 
 that the increasing use of captive balloons, which were subject 
 to sudden shocks and jars, of ballons-sondes (small free balloons), 
 and of kites, gave a strong impetus to the work of designing 
 really satisfactory self-recording instruments. The light self- 
 recording aneroid barometers, Bourdon tube 2 thermometers, and 
 
 1 Report to the British Association, Winnipeg, August, 1909. 
 
 2 See p. 133. 
 
328 
 
 METEOROLOGY 
 
 hair hygrometers of Richard Freres, of Paris, came to be consider- 
 ably used with kites and ballons-sondes. They recorded through 
 levers and metal styles on smoked paper, wrapped round a revolv- 
 ing clockwork drum. They were used with kites at the Blue Hill 
 Observatory, Massachusetts, U.S.A., alongside a meteorograph 
 designed byFergusson, which included also an anemometer; and by 
 Hermite and Besan9on with ballons-sondes in the years 1893 to 1 898. 
 The type of instrument finally evolved for use with ballons- 
 sondes had (1) a completely exhausted Bourdon tube 1 barometer, 
 
 FIG. 89. KITE CONSISTING OF Two STRIPS OF MATERIAL STRETCHED INTO QUADRANGULAR 
 SAILS ON A LOZENGE-SHAPED PLAN UPON FOUR TRANSVERSE BARS KEPT APART BY 
 Two PAIRS OF CROSS-STRUTS. 
 
 which was found to show less fatigue effect than the aneroid 
 barometer ; (2) Teisserenc de Bort's bimetallic thermometer 
 consisting of a blade of German silver fixed in a frame of Guil- 
 laume steel, which had small thermal inertia (requiring only 
 fifteen seconds to indicate a sudden change of temperature of 
 9 C.), and which was not affected by shocks and HergeselFs 
 German-silver tube thermometer ; (3) a hair hygrometer. The 
 working parts of the instrument were enclosed in an aspiration- 
 tube. Similar instruments were designed for use with kites. 
 
 1 See p. 133. 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 329 
 
 The principal self-recording instruments which have at various 
 times been used have been designed by Richard Freres, C. F. 
 Marvin, Fergusson, L. Teisserenc de Bort, R. Assmann, H. Her- 
 gesell, and W. H. Dines. Richard, Marvin, Hergesell, and Dines 
 designed instruments for use 
 with kites ; and Richard, 
 Teisserenc de Bort, Assmann, 
 Hergesell, and Dines, ballons- 
 sondes instruments. 
 
 Special balloon ascents were 
 made in 1907, on July 22 
 to 27, September 4 to 6, and 
 November 6 to 8. Reports 
 on the International Balloon 
 Ascents of July 22 to 27, 
 1907, by W. H. Dines, F.R.S., 
 J. E. Petavel, F.R.S., W. A. 
 Harwood, and Professor 
 W. E. Thrift, M.A., Fellow of 
 Trinity College, Dublin, will 
 be found in the number of 
 the Quarterly Journal of the 
 Royal Meteorological Society 
 for January, 1908. 1 
 
 The instruments Dines's 
 light meteorographs which 
 were used are described in 
 Symons's Meteorological 
 Magazine for July, 1906, 2 
 while a detailed account of 
 Dines's instruments and 
 methods is contained in The 
 Free Atmosphere in the Region 
 of the British Isles (Meteoro- 
 logical Office Publications, No. 202). Briefly, it may be said 
 that Dines's light meteorograph is a baro-thermograph, no 
 measurements of humidity being attempted. The barometer 
 1 Vol. xxxiv., No. 145, p. 1. 2 Vol. xli., p. 101. 
 
 FIG. 90. DINES'S BALLOON METEOROGRAPH. 
 
330 
 
 METEOROLOGY 
 
 is, in general, a partially exhausted German-silver aneroid, 
 and the thermometer is bimetallic, consisting of a strip of 
 aluminium or German-silver and a rod of invar. The partially 
 exhausted aneroid is used because it gives a larger scale than the 
 totally exhausted box, and leaves a record scratched by two 
 hard steel points on a small piece of sheet metal electro-plated 
 with copper (Fig. 91). The scale is very small, but the traces are 
 read under a low-power microscope. The marks consist of two lines 
 roughly parallel ; at least, they would be parallel were there no 
 change of temperature. In the usual form of trace, these lines 
 are about \ inch (12 millimetres) long, and start about -^ inch 
 
 apart, diverging to perhaps 
 YO inch at the top. The 
 distance between the lines 
 gives the temperature. Under 
 a microscope it is easy to 
 read this distance to about 
 001 inch (or -025 millimetre), 
 and this distance corresponds 
 to about 1 C. The record 
 of height is more uncertain. 
 The atmospheric pressure 
 recorded can be ascertained 
 with fair accuracy, but 
 heights where the pressure 
 amounts to only a few inches 
 of mercury (under 150 milli- 
 metres) are frequently reached, and at such heights 40,000 feet 
 and upwards a small error in the pressure means a large error 
 in the height. 
 
 The heavier of the meteorographs weighs 3J ounces (100 
 grammes). In the second and lighter form of instrument, which 
 weighs 1 ounce (28 grammes), the thermometric arrangement 
 consists of a round steel (invar) rod and a strip of very thin alu- 
 minium, both 6 inches long ; and their difference in length, which 
 is multiplied about twenty times by a lever, produces the varying 
 distance between the scratches. 
 
 As regards the balloon ascents of July, 1907, balloons carrying 
 
 FIG. 91. RECORDING PLATE OF DINES'S 
 METEOROGRAPH. 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 331 
 
 recording instruments the 1-ounce Dines's meteorographs were 
 sent up at 11 a.m. daily between July 22 and 28 from Manchester. 
 The balloons were observed from three trigonometrical stations, 
 unfortunately in unfavourable hazy weather. Indiarubber 
 balloons of the nominal size, 2 feet (60 centimetres) diameter, 
 were used. These were filled to about 39 inches (1 metre) dia- 
 meter. The instruments, together with instructions to the 
 finder, were attached to a small silk parachute (11 inches square), 
 suspended some 6 feet below the balloon. On July 23 the tem- 
 perature of the air at the time of the ascent at Manchester was 
 59 F. ; at a height of 40,000 feet it was -52 F. The ascents 
 confirm the interesting theory put forward by M. Teisserenc de 
 Bort with regard to the existence of a nearly isothermal layer 
 above some 6 miles (10,000 metres). 1 His results were con- 
 firmed immediately afterwards by Dr. Richard Assmann, Director 
 of the Royal Prussian Aeronautical Observatory, Lindenberg, 
 Germany. 2 Its existence over North America was demon- 
 strated by Mr. A. L. Rotch. Teisserenc de Bort found the 
 average height at which the change occurred to be about 11 kilo- 
 metres. He discovered also that the height was greater near 
 centres of high atmospheric pressure than near centres of low 
 pressure, the average heights for the two cases being 12'5 and 10 
 kilometres respectively. Later observations agree on the whole 
 with these results. Between the ground-level and a height of 
 about 5 miles the temperature falls continuously. The average 
 gradients over this range varied during the week of the ascents 
 between 0-28 and 0-33 C. per 100 feet (0-5 and 0-6 C. per 
 100 metres). Above 7 miles all the Manchester curves show a 
 more or less marked temperature inversion. In his report, read 
 before the Royal Meteorological Society on November 20, 1907, 
 Mr. W. H. Dines remarks that " the thirty -four successful ascents 
 that have been made in England since June last (1907) have, in 
 my opinion, proved conclusively the existence of the isothermal 
 conditions above some 7J miles (12,000 metres)/' The heights 
 attained by the balloons ranged to over 12 J miles (20 kilometres), 
 the average being about 7J miles (12 kilometres). 
 
 1 Complex Eendus, April, 1902, vol. cxxxiv., pp. 987-1000 ; January, 1904, 
 vol. cxxxviii., pp. 42-45 ; July, 1907, vol. cxlv., pp. 149-152, etc. 
 
 2 Ergebnisse aeronautischen Obs., Berlin, May, 1902. 
 
332 METEOROLOGY 
 
 In the international balloon ascents of July, 1907, England 
 headed the list for height with two ascents at Manchester of 
 69,000 feet (21,000 metres) ; and out of nine ascents of over 
 65,000 feet (20,000 metres), four took place in England. 
 
 During July and August, 1908, balloon observations were made 
 at Birdhill, Co. Limerick, Ireland, by Captain C. H. Ley, 
 F.R.Met.Soc. Twenty-five balloons were despatched, including 
 seven registering balloons, which latter were sent up on the inter- 
 national days in July namely, the 2nd, 27th, 28th, 29th, 30th, 
 and 31st. One of the balloons was observed, by means of an 
 8-inch special theodolite, to a height of nearly 33,000 feet ; four 
 were observed to over 20,000 feet ; seven to over 15,000 feet ; and 
 four to over 10,000 feet. As regards horizontal distance, several 
 balloons were observed to 126,000 feet (24 miles), and one to an 
 actual range of 130,000 feet. This may be considered the practical 
 limit of vision on an opaque balloon 2J feet in diameter with a 
 35-power telescope in a clear atmosphere. A feature developed 
 during the course of the experiments was the observation of 
 balloons at night by means of naked acetylene lights. After 
 some trouble, these proved quite successful, gave long runs with 
 less risk of being lost in small clouds, and afforded points of light 
 which could be observed with great accuracy. 1 
 
 Investigations of the variations in velocity and direction of 
 wind in the great middle strata of the atmosphere by means of 
 balloons followed by special theodolites were also made by Captain 
 Ley in the summer of 1907 at Sellack, about 3 miles north-west 
 of Ross, Herefordshire. The formulae used for theodolite calcu- 
 lations were the following : 
 
 206085d sin A 
 l.h = a B . 
 
 2. h. d. =h cot A, 
 where h = vertical height in feet. 
 
 h. d. = horizontal distance in feet. 
 d= diameter, in feet, of balloon at start. 
 A = angle of altitude (to minutes). 
 B= observed diameter in seconds. 
 a = expansion percentage. 
 
 1 Quarterly Journal of the Eoyal Meteorological Society, January, 1909, 
 vol. xxxv., No. 149, p. 15. 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 333 
 
 A direct estimation of range of the balloon from its apparent 
 diameter, as measured by cross-threads in a telescope, of which 
 the aperture was 1-9 inches and the focal length was 13 J inches, 
 was also made. 
 
 Based on this series of balloon observations, Captain Ley read 
 a paper on the possibility of a topography of the air before the 
 Royal Meteorological Society on December 18, 1907. * 
 
 Systematic investigations of the upper air are now carried out 
 at four British stations by means of kites and balloons. These 
 stations are : 
 
 1. Pyrton Hill, near Watlington, Oxfordshire, 500 feet above 
 sea-level. 
 
 2. Ditcham Park, near Petersfield, Hampshire, 400 feet. 
 
 3. Brighton, Sussex, 380 feet. 
 
 4. Glossop Moor, Howard Estate, Peak District, Derbyshire, 
 1,100 feet. 
 
 The work is under the general direction of Mr. W. H. Dines, 
 F.R.S., who has designed most of the special apparatus in use. 
 The work carried on may be classified under five heads : 
 
 1. Observations by means of kites. 
 
 2. Observations by means of registering balloons. 
 
 3. Observations by means of pilot balloons followed by one or 
 more theodolites. 
 
 4. The making of apparatus and instruments for use at Pyrton 
 Hill and other stations. 
 
 5. Calibration of the instruments and working up the records 
 obtained. 
 
 The kites carry meteorographs, which give a continuous record 
 of temperature, wind velocity, humidity, and height. 
 
 At the meeting of the Royal Meteorological Society held on 
 November 20, 1907, a paper was read, written jointly by Miss 
 Margaret White, Mr. T. V. Pring, and Mr. J. E. Petavel, F.R.S., 
 in which the authors discussed the observations (some 5,000 in 
 number) made at the British kite-stations during the session of 
 1906-1907. 2 
 
 1 Quarterly Journal of the Royal Meteorological Society, January, 1908, 
 vol. xxxiv., No. 145, p. 27. 
 
 2 Quarterly Journal of the Royal Meteorological Society, January, 1908, 
 vol. xxxiv., No. 145, p. 15. 
 
334 METEOROLOGY 
 
 To the Third Annual Report of the Meteorological Committee 
 for the year ended March 31, 1908, Mr. Dines contributed a report 
 on the work done during that year at the Pyrton Hill station, 
 a little over 14 miles south-east by east of Oxford, and on the 
 north-western slope of the Chiltern Hills. During the year 
 April 1, 1907, to March 31, 1908, 113 kite ascents were accom- 
 plished, and the average height was 3,500 feet. The first register- 
 ing balloon was sent up on June 5, 1907. Up to March 31, 1908, 
 twenty such balloons had been despatched, of which fifteen were 
 found, though only fourteen records were obtained, because in 
 one case the finder abstracted the meteorograph, and returned 
 only the empty case. 
 
 In general, rubber balloons weighing 8 ounces (227 grammes) 
 are used singly : they are filled with hydrogen, and at starting 
 have a radius of about 19 to 20 inches (J metre). They carry a 
 meteorograph, which, with a bright metal cylindrical case, weighs 
 just 2 ounces. The parachute is made out of very thin red silk, 
 10 inches square, with a IJ-inch hole, and threads running from 
 the four corners to a thin wire cross. Apart from the balloon, 
 the total weight to be carried is 2J ounces, and a free lift of about 
 10 ounces is generally given. This affords an ascensional velocity 
 of from 600 to 700 feet per minute. The average height at- 
 tained has been 48,500 feet (14'8 kilometres), and in most cases the 
 isothermal layer has been reached. The balloons are sent up a 
 short time before sunset. 
 
 Up to April 2, 1908, forty-five records had been obtained from 
 these meteorographs by balloons sent up either from Pyrton 
 Hill, Manchester, Ditcham Park, Sellack in Herefordshire, or 
 Crinan in Argyllshire. The map (Fig. 92) shows the geographical 
 positions from which the instruments were returned by the finders. 
 Fig. 93 shows the relation between temperature and height ob- 
 tained from the results. The diagram on the small scale of the 
 reproduction is too confused for the individual curves to be 
 traced, but three points may be made out : (1) A notable compli- 
 cation within the first 2 miles above the surface ; (2) remarkable 
 parallelism in the slope of the curves, showing nearly identical 
 temperature gradients up to 10 or 12 kilometres ; (3) the iso- 
 thermal layer above 12 kilometres. The figures are reproduced 
 
Investigation of the. Upper ft \r IS01 -8 
 
 sent up from the following 
 be&n found 
 
 Cr'.nan 
 
 Fio. 92. MAP SHOWING THE POSITIONS AT WHICH SOUNDING BALLOONS SENT UP FROM 
 CERTAIN OBSERVING STATIONS IN 1907-1908 WERE FOUND. 
 
330 METEOROLOGY 
 
 by the kind permission of the Controller of His Majesty's Sta- 
 tionery Office, London. 
 
 " Perhaps the most remarkable phenomenon revealed by the 
 investigation of the upper air with balloons carrying self-recording 
 instruments," write Messrs. Gold and Harwood, " is the com- 
 paratively sudden cessation of the fall of temperature at a height 
 varying with the time and the latitude. Above this height, 
 which may be regarded as the height of an irregular but roughly 
 horizontal surface dividing the atmosphere into two regions, the 
 temperature at any time varies very little in a vertical direction, 
 showing on the average a slight tendency to increase. This com- 
 parative absence of regular vertical variation of temperature in 
 the upper region led to the name ' isothermal layer or region/ to 
 distinguish it from the lower atmosphere, in which the vertical 
 variation of temperature is about 6 C. per 1,000 metres." 
 
 The upper region has been usually described as the " isothermal 
 layer," but Teisserenc de Bort has recently introduced the terms 
 " stratosphere " and " troposphere " to denote the upper and 
 lower regions respectively. 
 
 It is difficult to form a mental picture of the condition of the 
 atmosphere to be inferred from observations at different places 
 on the same day, and in order to help towards that object 
 Dr. Shaw has constructed a model of the block of the atmosphere 
 over the area of observation in the British Isles for each of the 
 two days, July 27 and July 29, 1908, on which observations were 
 taken on an extended scale. Simultaneous soundings by bal- 
 loons (ballons-sondes) gave data for temperature up to great 
 heights at the corners of a triangle, about 300 miles in the side. 
 The observations used in the representation are from records of 
 balloons sent up at Petersfield (Hants), Watlington (Oxon), 
 Crinan (Argyll), and Limerick, on the 27th ; and at Watlington, 
 Manchester, Crinan, and Limerick on the 29th, together with 
 additional observations of a pilot balloon at Petersfield on the 
 29th. 
 
 A frame of vertical plates of glass represents a block of atmo- 
 sphere, 15 miles thick, standing on the triangle. The horizontal 
 scale of the original models is 25 miles to an inch, the vertical 
 scale 5 miles to 4 inches. The observed temperatures are set 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 337 
 
 CURVES SHOWING CHANGE OF TEMPERATURE WITH HEIGHT ABOVE SEA- LEVEL 
 OBTAINED FROM BALLON- 50NDE ASCENTS 1907-8. 
 
 IUNCS. PYRTON HILL 
 IULY 4. PYRTON HILL 
 
 5&.LACK 
 MANCHESTER 
 PYRTON HILL 
 DITCHAM ftRK 
 MANCHESTER 
 ^INANHARB' 
 
 IULY2& PYRTON HILL 
 MANCHESTER 
 CRINAN HARB* 
 6UU.Y36. CRINAN HARB' 
 DITCHAM 
 
 hAY2a MANCHESTER 
 
 iEPT3.[SELLACK 
 
 ISELLACK 
 )CfT6. 
 iDTie. PYRTON HILL 
 
 PYRTON HILL 
 
 [DITCHAM PARK 
 
 SELLACK 
 
 DITCHAM PMH 
 MANCHESTER 
 I7fta/a MANCHESTER 
 BOecilMANCHESTER 
 PYRTON HIU.' 
 
 DITCHAM 
 PYRTON HILL 
 DITCHAM PARK 
 MANCHESTER 
 
 DITCHAM FARK ?2JANtt PYKTON HILL 
 
 DITCHAM PARK Z3FEB5 
 
 . PYRTON HILL 
 24PEB6. MANCHESTER 
 [PYRTON HILL 
 [MANCHESTER 
 
 Ami. PYRTON HILL 
 Z7A/H2. PYRTON HILL 
 
 DITCHAM f*RK 25MAR5. 
 
 SECTION ACROSS THE BRITISH 
 
 POINT TO V (MOUTH JHCWING 
 
 -40 -20 
 
 TEMPERATURE IN DEGREES FAHRENHEIT 
 240 2^0 ?60 
 
 TFMPFRATURE ABSOLUTE 
 
 FIG. 93. SHOWING THE RELATION BETWEEN TEMPERATURE AND HEIGHT OBTAINED BY 
 THE ASCENTS OF " BALLONS-SONDES." 
 
 22 
 
338 ;. METEOROLOGY 
 
 out at the angles. The distortion due to the horizontal drift 
 of the balloons is neglected. Points of equal temperature are 
 joined by lines on the glass sides ; the thickness of the lines is 
 adjusted to cover half a degree of temperature. These isothermal 
 lines are drawn for every 5 C., and the space between the two 
 lines on either side of the freezing-point, to be found at a height 
 of about 3 kilometres, is filled in so that it appears as a thick 
 band in the photographs. 
 
 The lines as thus drawn suggest the distribution of tempera- 
 ture " in the solid." They give a vivid representation of the 
 reversal at the top of the troposphere. The temperature of the 
 closed ring shown at the south-eastern angle (Petersfield) on the 
 27th (Plate III., Figs. 3 and 4) is - 58 C., and that of the smaller 
 closed ring on the 29th (Plate III., Figs. 1 and 2) at an angle 
 (Watlington), which is nearly identical, is - 68 C. On the 27th 
 the isothermal lines indicate a wedge of cold air ( - 58 C.) in- 
 vading the block at the reversal layer near the south-east corner. 
 On the 29th the cold wedge is shown nearly covering the triangle. 
 Lines of equal pressure are drawn in red on the original models 
 to show the levels of one-fifth and one-tenth of an atmosphere ; 
 they are shown as beaded lines on the photographs. 
 
 Winds, as far as they are known from observations of pilot 
 balloons, are shown by means of arrows representing wind vanes, 
 mounted on long vertical pins. 
 
 Messrs. Gold and Harwood point out that the absence of vertical 
 temperature fall in the upper atmosphere implies that general 
 direct convection in that region is also absent, and that, in 
 general, interchange of air in the stratosphere would be mainly 
 by advection. They accordingly suggest that the two regions 
 of the atmosphere might be approximately named advective and 
 convective regions, so expressing the characteristic difference 
 between them. The terms " convective " and " advective " need 
 somewhat more detailed explanation, which I can best give in 
 Mr. Gold's own words in answer to a query addressed to him by me. 
 
 If a quantity of matter undergoes changes such that no trans- 
 ference of heat takes place between it and external matter, the 
 changes are said to be adidbatic (Greek, d8ia/?aro?, not to be 
 passed). If the transference of air from one level in the atmo- 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 339 
 
 sphere to another takes place adiabatically, the difference of tem- 
 perature between the two places is proportional to their difference 
 of level. It is equal to 9 C. for a height of 1 kilometre. If the 
 air is saturated with water vapour, the difference of temperature 
 is no longer strictly proportional to the difference of level. Near 
 the earth's surface the difference is nearly 5 C. per kilometre, 
 but it increases with increasing height. Of course, it is always 
 less than 9 C. 
 
 If the motion of the atmosphere consists of vertical currents or 
 of horizontal currents with considerable vertical interchange, it 
 is said to be in the convective state. It must then be also approxi- 
 mately adiabatic. Adiabatic and convective are frequently used 
 as interchangeable terms, although they are not strictly so. The 
 idea implied by " convective " is that vertical interchange is the 
 dominating factor in determining the vertical distribution of 
 temperature. 
 
 If vertical interchange is practically absent and the atmospheric 
 currents are nearly altogether horizontal, the state is said to be 
 advective. Of course, there will be some vertical interchange as 
 long as there are horizontal currents, but the vertical motion will 
 exercise only a secondary influence on the temperature distribution. 
 
 In a paper recently published, 1 on " Vertical Temperature 
 Gradients of the Atmosphere, especially in the Region of the 
 Upper Inversion," Professor W. J. Humphreys arrives at the 
 following (among other) conclusions : 
 
 1. Considered from the standpoint of temperature gradients, 
 the explored portion of the atmosphere is divisible into three 
 parts : (1) The region of terrestrial disturbance, extending from 
 the ground to an elevation of about 3,000 metres above its sur- 
 face ; (2) the region of uniform changes, from the top of (1) to, 
 roughly, 10,000 metres above the sea ; (3) the region of permanent 
 inversion, or all that explored portion, at least, that lies above 
 the plane of the upper inversion. 
 
 2. Spectroscopically, the known atmosphere is divisible into 
 three parts : (1) The black body portion, coincident with the 
 region of, and due to, the relatively dense water vapour ; (2) the 
 
 1 Bulletin of the Mount Weather Observatory, March 24, 1909, vol. ii., pt. i. 
 Washington, U.S., Weather Bureau. 
 
 222 
 
340 METEOKOLOGY 
 
 diathermanous, or the dry air next above the water vapour ; 
 (3) the selectively absorptive, or the air of the isothermal layer, 
 presumably rich in ozone. 
 
 On December 10, 1908, Dr. W. N. Shaw communicated to the 
 Royal Society a paper by E. Gold, M.A., Fellow of St. John's 
 College, Cambridge, and Reader in Meteorology, on " The Iso- 
 thermal Layer of the Atmosphere and Atmospheric Radiation." * 
 Having shown that there can be no question of the isothermal 
 layer of the atmosphere being merely a local or temporary 
 phenomenon, Mr. Gold states that it is clear that there cannot be 
 convection currents to any marked extent in this region. He 
 proceeds to show that in an atmosphere which is not transparent, 
 but absorbs and emits radiation, the process of radiation would 
 prevent the establishment of the temperature gradient necessary 
 for convective equilibrium in the upper layers of the atmosphere, 
 and that in the lower layers of our atmosphere it can be main- 
 tained only by transference of energy from the earth to the 
 atmosphere by direct convection or by the process of evaporation 
 of water at the earth's surface, and subsequent condensation in 
 the atmosphere. The heat necessary for the evaporation of 
 water vapour at the earth's surface is supplied mainly by absorp- 
 tion of solar radiation, and is not taken from the atmosphere, 
 but the heat given up on condensation is added almost entirely 
 to the heat of the atmosphere, and in this way we get a supply 
 of heat to the atmosphere at a rate that may be estimated approxi- 
 mately from the annual rainfall. 
 
 The attempts to furnish a reasonable explanation of the 
 phenomenon on theoretical grounds led to various suggestions. 
 Trabert 2 showed that if there were a decrease of temperature 
 in a horizontal direction in passing eastwards over Europe, and 
 if the air moving eastwards also had a small ascending motion, 
 then the adiabatic fall of temperature would not exist in a vertical 
 direction. It appears probable, however, that the causes which 
 produced a horizontal decrease of temperature in one layer would 
 also produce a similar decrease in the layer above it, and in that 
 case Trabert's effect would vanish. 
 
 1 Proceedings of the Royal Society, A. vol. Ixxxii., p. 43. 
 
 2 Met. Zeit., 1907. 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 341 
 
 Fenyi 1 considered the question of the absorption of solar 
 radiation in the upper atmosphere. He concluded that, if the 
 phenomenon were due to this, there must be absorption of dark 
 radiation, since the ultra-violet radiation would be insufficient 
 even if it were all absorbed. Humphreys 2 pointed out that, if 
 the effective radiating power of the earth and atmosphere were 
 the same as that of a black body at temperature T lf the effect 
 on any radiating and absorbing matter near enough to the earth 
 for the radiating surface to be regarded as an infinite plane would 
 be to keep the matter at a constant temperature, such that the 
 radiation from it would be half the radiation from it at tempera- 
 ture T r If the radiating matter were such as to admit of the 
 application of Stefan's law, its temperature would be T, where 
 T^JTj 4 . 
 
 From a careful examination of the results of MM. Dulong and 
 Petit's classical experiments as to the laws of cooling (Annales 
 de Chimie et de Physique, 2 e , tome vii., pp. 225 et 337, 1817), 
 M. J. Stefan was led to the conclusion that the total radiation 
 emitted by any body is proportional to the fourth power of the 
 absolute temperature of the body. 3 
 
 This was shown by Boltzmann to be a necessary relation from 
 thermodynamic considerations. " Thus, if T is the absolute 
 temperature of a ' black ' body," writes Mr. E. Gold, " and R 
 the number of units of energy radiated from unit surface of it in 
 unit time, then 
 
 This is Stefan's Law. " The law strictly applies only to a black 
 body, which means a body radiating for every wave-length, and 
 capable of absorbing all the energy of any wave-length which 
 falls upon it from an external radiation. Such a ' body ' radiates 
 as much energy at a given temperature as it is possible for any 
 ' body ' to radiate. Of course, it would be possible for the law 
 to apply also to other bodies which radiated, say, a definite frac- 
 tion for all wave-lengths of the radiation of a black body. It has 
 
 1 Met. Zeit., 1907. 2 Astrophysical Journ., 1909. 
 
 3 Sitzungsberichte d. k. Akadeniie der Wissenscha/ten in Wien, vol. Ixxix., 
 1879 ; and Journal de Physique, tome x., p. 317, 1881. Cf. Preston's Theory 
 of Heat, 1894, p. 458. 
 
342 METEOKOLOGY 
 
 been suggested that such bodies might be called ' gray' bodies." 
 The whole subject is treated clearly and simply in Poynting and 
 Thomson's Heat. 
 
 The observed value of T agrees with the value deduced from 
 this equation by giving Ti the value estimated by Abbott and 
 Fowle 1 from the value of the solar constant, regard being paid 
 to the proportion of the incident solar radiation which is re- 
 flected, and does not affect the temperature of the earth. 
 
 Gold 2 developed a theory based on the experimental results 
 for atmospheric absorption obtained by Paschen and others. 
 His argument rests on the principle that a necessary condition 
 for convection is that in the upper part of the convective system 
 the radiation from any horizontal layer must exceed the absorp- 
 tion by it. He takes the temperature in the convective region to 
 be given with sufficient approximation by the equation T ] = kp 
 where n = 4 and p is pressure, and represents the radiating power 
 of the atmosphere by a = 1 (q - p), where a and q are constants, 
 in order to allow for the diminution with height arising from the 
 decrease in the amount of water vapour present. 3 He finds that 
 
 1 Annals of Observatory of Smithsonian Institution, vol. ii. 
 
 2 Proceedings of the Royal Society, A. vol. Ixxxii. 
 
 3 Messrs. Gold and Harwood state that it has been suggested that the 
 upper limit of the convective region may be also the upper limit of the 
 water vapour atmosphere. But it appears certain that at this upper limit 
 tho atmosphere must always be saturated with water (ice) vapour, and that 
 in the advective region the water vapour atmosphere will be such that the 
 difference of vapour pressure between two points will be equal to the weight 
 of the vapour in the intervening column. For the processes of diffusion 
 and of convection of water vapour alone would tend to produce a water 
 vapour atmosphere, in which the amount of vapour present at any height 
 in the convective region would be more than sufficient to produce satura- 
 tion at that height for the temperature in the actual atmosphere. The 
 only process which prevents the atmosphere being saturated at all heights 
 is the descent of air carrying with it the water vapour it contained at the 
 beginning of the descent, an amount insufficient to saturate it at lower 
 levels. But at the upper limit of the convective region there can be no 
 considerable descent of air from above, and the air arriving there from below 
 will necessarily be saturated, since it must contain sufficient water vapour 
 to saturate it at the lowest temperature to which it has been exposed i.e., 
 T c . Of course, the actual amount of water \apour present is small com- 
 pared with the amount present near the earth's surface ; but a small amount 
 of water vapour is sufficient, at ordinary temperatures at least, to produce 
 considerable absorption of terrestrial radiation, and the absorption extends 
 through a large part of the spectrum of radiation at terrestrial tempera- 
 tures. In fact, it is probably chiefly due to the presence of this water 
 vapour that it is possible to obtain theoretical results agreeing with the 
 observed facts by using the assumption that the absorption, and therefore 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 343 
 
 for an atmosphere of uniform constitution the adiabatic state 
 cannot exist to a height greater than that for which p = Jp , 
 where p is the surface pressure, because if it extends at any time 
 to a greater height, the absorption in the upper part will exceed 
 the radiation. He shows that for the actual atmosphere the 
 adiabatic state can exist to a limited height only, and that if 
 the atmosphere consist of an adiabatic and an isothermal region 
 the adiabatic state must extend to a height greater than 5*5 kilo- 
 metres, and cannot in general extend to a height greater than 
 10' 5 kilometres. He shows also that the radiation from the 
 lower half of the convective region exceeds the absorption by it, 
 and deduces that its temperature must be maintained by con- 
 vection from the earth's surface and by condensation of water 
 vapour. It follows also from the theory that, if in the upper 
 region the temperature increases with the height, the conditions 
 for thermal equilibrium are satisfied if the convective atmosphere 
 extends to a height greater than that for the case of an isothermal 
 upper region i.e., the limits for Ho are greater than 5*5 and 
 10'5 kilometres. 
 
 Shaw 1 has recently considered the connection between a 
 depression of the lower surface of the advective region and the 
 temperature distribution in tha*t region. He finds that if such 
 a depression is produced artificially or through a disturbance 
 in the convective region, the first effect will be to produce a hori- 
 zontal difference of temperature in the advective region. If 
 the advective region is initially isothermal, it will still be vertically 
 isothermal, but the temperature of the vertical columns will not 
 be the same for all. Over the depression the temperature will 
 be raised. He finds that the value of He is diminished by 3'5 
 times the difference of height of the " homogeneous " stratosphere 
 at the normal and increased temperatures. If the increase of 
 temperature is 20 C., the decrease in He is about 2 kilometres. 
 
 At the Winnipeg meeting of the British Association, on 
 
 also the radiation, is sufficiently extensive to warrant the application of 
 Stefan's law. It follows, also, from this reasoning that the mean amount 
 of vapour present at any height above the lower cloud level will be at least 
 half the sum of the amount for saturation at that height, and the amount 
 necessary for saturation at the height H c namely, the height of the 
 dividing surface between the advective and convective regions. 
 
 1 Perturbations of the Stratosphere, Publications of the Meteorological 
 Office, No. 202. 1909. 
 
344 METEOROLOGY 
 
 August 31, 1909, to the department of Section A. (Mathematics) 
 devoted to Cosmical Physics, Professor Humphreys, of the United 
 States Weather Bureau, communicated a paper on " Seasonal 
 and Storm Vertical Temperature Gradients." The paper dealt 
 mainly with results obtained by ballons-sonies in Europe, and 
 showed that in regions of high pressure (pressure above 770 milli- 
 metres, or 30'3 inches) the mean temperature was higher than in 
 regions of low pressure (pressure < 750 millimetres, or 29 '5 inches) 
 both in summer and winter. This result is in agreement with that 
 found from the manned-balloon observations by von Bezold, 
 and is corroborated by the results given by Gold and Harwood in 
 their Report on the Present State of Our Knowledge of the Upper 
 Atmosphere. In this Report, as has been explained above, 
 the names " Advective " and " Convective " Regions are used 
 to denote the upper and lower parts of the atmosphere, and 
 H c is used to denote the height at which the advective region 
 begins. 
 
 The three sets of terms applied to the same phenomena are 
 therefore : 
 
 Isothermal Layer. Adiabatic Atmosphere. 
 
 Stratosphere. Troposphere. 
 
 Advective Region. Convective Region. 
 
 The European observations showed remarkable minima in the 
 value of He in March and September, and an attempt was made 
 in the Report to connect these with the general circulation of the 
 atmosphere. The interesting law discovered by Egnell, V/> = const, 
 where V is wind velocity and /> air-density, was shown to be only 
 approximately true, and was proved to be a consequence of the 
 difference in temperature between regions of high and low pressure. 
 
 H. Hergesell's exploration of the upper atmosphere over the 
 Northern Atlantic showed that the temperature distribution in 
 the vertical was very irregular, and not like that in the lower 
 latitudes. The higher strata of the air were, on the whole, rather 
 warmer, probably owing to the continued heating by the Arctic 
 sun. Immediately above the sea the temperature frequently fell 
 rapidly, while the humidity rose a bank of clouds would limit 
 this stratum. On the Island of Spitzbergen strong land winds 
 were very persistent, blowing in the same direction even at night, 
 especially with a bright sky. 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 345 
 
 In the Wijde Fjord, which penetrates 100 kilometres southward 
 into the land, the wind blew almost constantly at 7 metres per 
 second from the south. But these cold winds from the glaciers 
 were purely local, and died away over the open sea : they were 
 confined to the lower strata of a few hundred metres, above 
 which kites would not rise. In the higher strata the winds were 
 much stronger, blowing at the rate of 20 or 30 metres per second 
 at an altitude of 10,000 metres, as shown by pilot balloons and 
 telescopes. But the winds were very irregular ; the westerly com- 
 ponent was the most powerful, and winds from the south were as 
 frequent as winds from the north. It will be seen from this most 
 important exploration of the upper atmosphere within the Arctic 
 circle the results of which were communicated to the French 
 Academy of Sciences that the winds from the south are purely 
 local, and that, if any strong winds predominate at all, it is those 
 from westerly points. 
 
 During the months of July to September, 1908, an aerological 
 expedition to Tropical East Africa was carried out by the Royal 
 Prussian Aeronautical Observatory, Lindenberg, Germany, of 
 which Dr. Richard Assmann, honorary member of the Royal 
 Meteorological Society, is Director. 
 
 The object of the expedition was the exploration of the upper 
 air in the heart of a tropical continent, in the very middle of the 
 Equatorial belt. It was intended also to contribute to the elucida- 
 tion of a well-known important problem the origin and interior 
 structure of the monsoon winds of the Indian Ocean the influ- 
 ence of which on precipitation in India and the subsequent crops 
 is of the highest and most practical interest. 
 
 The expedition left Europe in the middle of June, and arrived, 
 via the Uganda Railway and the Victoria Nyanza, on July 24, at 
 Shirati, in German East Africa, a town situated on the east coast of 
 that vast lake in 1 7' S. latitude. The members of the expedition 
 were Professor Berson, director ; Dr. Elias, formerly assistant 
 at the Royal Aeronautical Observatory ; and Mr. Mund, balloon 
 superintendent of that Observatory. 
 
 In the interval between the end of July and the middle of Sep- 
 tember twenty-three ascents of self -registering balloons were made 
 on the lake ; fifteen of the balloons were recovered with their 
 apparatus. In the two highest ascents of these balloons namely 
 
346 METEOROLOGY 
 
 to 65,000 feet (19,800 metres) and 56,000 feet (17,000 metres) 
 the upper isothermal layer was duly found. It is, therefore, 
 now proved to exist in the actual Equatorial belt, as it does over 
 the Arctic Seas. The lowest temperature encountered at 65,000 
 feet was -119 F. (-84 C.), when the thermometer on the 
 ground (3,800 feet- 1,150 metres above sea-level) read 79 F. 
 (26 C.). The variability of temperature at high altitudes was 
 very marked : in two subsequent ascents at 56,000 feet - 105 F. 
 ( - 76 C.) and - 62 F. ( -52 C.) were registered, while the whole 
 annual variation on the ground does not exceed 3 or 4 C. 
 
 Besides these ascents of registering balloons, a large number 
 of smaller pilot balloons, carrying no apparatus, were sent up, 
 soaring in some instances to enormous heights, in order to com- 
 plete a study of the wind. Their flight was observed with theodo- 
 lites from fixed points on the shore. The most surprising result 
 was the discovery of an uppermost current of air blowing nearly 
 from due west, and flowing above the regular easterly current of 
 the Equatorial region. 1 In a note by E. Gold he points out that 
 according to Oberbeck's theoretical solution of the problem of 
 the general circulation, the belt of easterly currents gets rapidly 
 narrower with increase of height, and at very great heights the 
 current will be westerly even close to the Equator. 
 
 The name of Leon Teisserenc de Bort is so intimately associated 
 with the investigation of the upper air that it is allowable to cull 
 the following facts relative to his life from the Minutes of the 
 Council Meeting of the Royal Meteorological Society of October 16, 
 1907, at which meeting it was resolved that the Symons Memorial 
 Gold Medal for 1908 should be awarded to him " for distinguished 
 work done in connection with meteorological science." 
 
 Leon Teisserenc de Bort commenced his scientific work in 1878 
 as a member of the staff of the Bureau Central Meteorologique 
 de France. From 1880 to 1892 he was chief of the department of 
 General Meteorology. He resigned his appointment in 1892 in 
 order to devote himself to experimental research in meteorology. 
 In 1896 he founded an observatory for the study of dynamical 
 meteorology at Trappes, which has become famous for the bold- 
 
 1 Dr. Assmann contributed a full descriptive account of this notable 
 expedition to the Royal Meteorological Society on January 31, 1909 (Quar- 
 terly Journal of the Society, vol. xxxv., No. 149, p. 51). 
 
INVESTIGATION OF THE UPPER ATMOSPHERE 347 
 
 ness of its enterprise and the success of its methods in the study 
 of the upper air. 
 
 M. Teisserenc de Bort has also organised and directed for 
 special purposes in connection with that study schemes for ex- 
 ploration of the upper air in Denmark and Lapland, on the Catte- 
 gat, and the Zuyder Zee. He has transformed a Hull fish-carrier 
 into a floating meteorological observatory, which, under the name 
 of the yacht Otaria, has carried out partly with the assistance of 
 Professor Rotch extensive researches in the Mediterranean and 
 over the intertropical belt of the Atlantic Ocean. 
 
 But M. Teisserenc de Bort is best known for his work upon the 
 upper air by means of ballons-sondes and kites. The equipment 
 used was designed and constructed at Trappes, and the results 
 obtained have added largely to our knowledge of the upper air. 
 The establishment of the existence of the so-called isothermal 
 layer at a height of about 10 kilometres is largely due to the work 
 of the Observatory. The identification of the return anti-trade 
 above the north-east trade-wind ; the structure of the inter- 
 vening layer ; and the comparison of the temperatures in the 
 highest layers in various latitudes, which apparently show a 
 gradient of temperature from the pole to the Equator, instead of 
 one, as might be expected, in the opposite direction, are scientific 
 achievements of M. Teisserenc de Bort, in collaboration, as regards 
 some parts of the work, with colleagues of various countries. 
 
 With Professor Hildebrandsson, M. Teisserenc de Bort has 
 recently completed an historical treatise on dynamical meteorology. 
 The advances made in that subject by his personal exertions and 
 at his own charges must always remain for future historians 
 among the most original achievements of the present generation. 
 
 One of M. Teisserenc de Sort's most recent researches has been 
 on the composition of air at great altitudes, with special reference 
 to argon and its allies. 1 The collecting vessel was a glass tube 
 with a finely drawn-out end, which was sealed after a very perfect 
 vacuum had been made inside. When the tip of the capillary tube 
 was broken at the desired height, the air was expected to enter 
 readily and fill the tube, so that it only remained to fuse the 
 capillary tip in order to secure the sample. The tube is opened by 
 
 1 Quarterly Journal of the Royal Meteorological Society, July, 1908, 
 vol. xxxiv., No. 147, p. 189. 
 
348 METEOROLOGY 
 
 the tip being broken off by the fall of a little hammer, released by 
 an electric contact, and it is sealed by another contact allowing 
 the current from a small accumulator to raise to red-heat a 
 platinum wire wound round the capillary tube, by which means 
 heat is produced sufficient to melt the glass. Both contacts can 
 be made either by the barometer at a pressure previously ar- 
 ranged, or by the clockwork of the meteorograph, and the whole 
 operation is so conducted that no impurity can possibly affect 
 the air in the tube, the apparatus being hung at such a distance 
 below the balloon (ballon-sonde) that there can be no trace of 
 hydrogen in the air. 
 
 The apparatus, as first made on a very small scale, secured 
 several little tubes of air in July, 1907. The quantity of air 
 collected was too small to permit of an ordinary chemical an- 
 alysis, and M. Teisserenc de Bort decided to confine himself to 
 spectrum analysis, paying special attention to argon, neon, and 
 helium. He proceeded by two different methods one, by ab- 
 sorbing all the elements of the air, except helium and neon, by 
 means of carbon ; the other, by first separating the argon. 
 
 The result of the first experiment proved the presence of argon 
 in all the samples of air taken between 8,000 and 14,000 metres 
 (5 to 8 miles), as one would expect. Helium, distinguished by 
 its yellow line in the spectrum, was detected in most of the speci- 
 mens ; but the highest of all, that taken at 14,000 metres, showed 
 no trace of it. Neon was clearly discernible in every case. The 
 attempts to disclose krypton have not as yet given any result, but 
 the experiments on this gas have not been sufficiently numerous to 
 allow an opinion to be formed on the subject (Paris, June 16, 1908). 
 
 From the commencement of 1909 the results of observations in 
 the Upper Air have been stated by the Meteorological Office and 
 other scientific bodies in absolute or metric units. The units 
 adopted for each element are as follow : Altitude Metres (m.) or 
 kilometres (km.). Wind Direction Angular Measures from 
 North (360) through East (90). Wind Velocity Metres per 
 second (m.p.s.). Pressure Megadynes per Square Centimetre 
 (m.g.d.). Temperature The Absolute Scale of Centigrade De- 
 grees (freezing-point of water = 273 ; normal boiling-point of 
 water = 373. Values at or below the freezing-point of water 
 (273) are printed in heavy type. 
 
INVESTIGATION OF THE UPPEE ATMOSPHERE 349 
 
 Tables for converting from British units to absolute or metric 
 units are given in tlie Introduction to the Weekly Weather Report 
 of the Meteorological Office for 1909, pp. 6, 7. 
 
 The explanation of the metric units is as follows : In order to 
 avoid negative values, the temperatures are given in Centigrade 
 degrees from the " absolute zero " or, more strictly speaking, 
 from 273 below the freezing-point of water and in recording 
 the air-pressure for different points of the ascents the unit of 
 one megadyne per square centimetre, or 1,000,000 absolute c.g.s. 
 (centimetre-gramme-second) units of pressure, are employed, 
 because the megadyne represents practically the normal surface 
 pressure at the level of the observing-stations ; and, consequently, 
 the pressure figures in megadynes per square centimetre (m.g.d.) 
 give the fraction of pressure of the atmosphere which is still above 
 the instrument. Thus, an entry in the pressure column of 
 527 m.g.d. means that the instrument was at a point where 
 the atmospheric pressure was reduced approximately to -527 of 
 the surface value. It is noteworthy that the conversion from 
 millimetres of mercury (in latitude 45) to m.g.d. can be carried 
 out with an accuracy of *1 millimetre by adding one-third and 
 moving the decimal point, and that a megadyne per square 
 centimetre exceeds the kilogramme per square centimetre by less 
 than 2 per cent. 
 
 In expressing the results of temperature measurements, the 
 figure in the "hundreds" place is omitted. Thus, the freezing-point 
 of water= 32 F.= C.= 73 A. (i.e., absolute temperature 273). 
 
 The vertical gradient of temperature is expressed in degrees C. 
 per kilometre, and is reckoned positive when temperature dimin- 
 ishes with increasing height. 
 
 The most recent contributions to the investigation of the 
 upper air are comprised in a report by Mr. W. H. Dines, B.A., 
 F.R.S., on apparatus and methods in use at Pyrton Hill, Oxford- 
 shire, with an introduction and a note on the " Perturbations of 
 the Stratosphere," by W. N. Shaw, Sc.D., F.R.S., Director of 
 the Meteorological Office, London. 1 At p. 13 of this monograph, 
 which has only just been published, will be found a very full 
 bibliography of the subject. 
 
 1 The Free Atmosphere in the Region of the British Isles, Meteorological 
 Office Publications, No. 202, 1909. Folio. 
 
PART III. CLIMATE AND WEATHER 
 
 CHAPTEK XXIII 
 CLIMATE 
 
 IN his work on Elementary Meteorology, Dr. K. H. Scott, F.K.S., 
 observes 1 that the old division of the world by Parmenides 2 
 into five zones a central torrid zone, northern and southern 
 temperate and frigid zones has been found to be quite inadequate 
 as a representation of the climatology of the globe. 
 
 In its original and stricter etymological sense, the word climate 
 (Greek, /cAt/xa, a slope or inclination) was applied to one of a 
 series of regions or zones of the earth running parallel to the 
 Equator, from which the earth's surface was supposed to slope 
 to the poles ; hence the Latin rendering of icAt/ua, inclinatio cceli. 
 
 According to this view, put forward by Claudius Ptolemy, 
 the author of the Ptolemaic System of the Universe (A.D. 120-149), 
 climate was determined solely by latitude, and one climate 
 differed from another only as regards the relative length of the 
 midsummer day and the relative altitude of the noontide sun. 
 As a matter of fact, latitude is only one, and that, as we shall 
 see presently, by no means the most important, factor in the 
 determination of climate. 
 
 We may define climate as the condition of a country, district, 
 
 or place, in relation to certain meteorological elements notably 
 
 air temperature, atmospheric pressure and wind, atmospheric 
 
 moisture and electricity viewed more particularly in their 
 
 effects upon animal or vegetable life. It is these effects which 
 
 determine the distribution of a fauna or of a flora in a given 
 
 1 At p. 338. 2 Of Elis. Flourished circa 430 B.C. 
 
 350 
 
CLIMATE 351 
 
 region of the globe. Dr. Scott points out 1 that the distribution 
 of the plants of most importance to mankind, such as the cereals, 
 depends chiefly on the summer temperature, while the distribu- 
 tion of animals is more dependent on the winter temperature. 
 For example, the province of Manitoba in Canada yields mag- 
 nificent crops of wheat, although its winter temperature often 
 falls far below zero of the Fahrenheit scale. Maize, again, succeeds 
 well in extreme climates, the summer season alone being suffi- 
 cient for its whole life. On the other hand, plants which, so to 
 speak, are alive throughout the year, like the fuchsia, the laurel, 
 or even the hawthorn, would perish in the bitter winter of the 
 " Canadian North- West," although they flourish in climates like 
 that of the British Islands, where the fate of the wheat crop trembles 
 yearly in the balance, and where maize entirely fails to grow. 
 
 In connection with this topic, attention has already been 
 drawn to the estimation of accumulated temperature, or warmth 
 available for agricultural purposes, expressed in " day-degrees " 
 (see Chapter IV., p. 36). It may be well to repeat that a " day- 
 degree " signifies 1 F. continued for twenty-four hours, or any 
 other number of degrees for an inversely proportional number 
 of hours, the term " accumulated temperature " indicating the 
 combined amount and duration of an excess or defect of tempera- 
 ture above or below 42 F. for the period named. 
 
 Animals can bear a greater range of temperature than plants, 
 and so their territorial distribution is more extensive than that 
 of individual members of the vegetable kingdom. As Dr. Scott 
 puts it, " The distribution of animals is more dependent on the 
 winter temperature." 
 
 Climate depends chiefly on (1) distance from the Equator, or 
 latitude ; (2) physical configuration of the surface ; (3) elevation ; 
 (4) nearness, or otherwise, of oceans ; (5) prevailing winds. In 
 his Lumleian Lectures on " Aero-therapeutics in Lung Disease/' 
 delivered before the Royal College of Physicians of London in 
 1893, Dr. C. Theodore Williams, M.A., F.R.C.P., gives the 
 principal factors oi^climate as follows : 
 
 1. Latitude Naturally the greatest influence as describing 
 the position of the sun towards the earth in a certain 
 1 Elementary Meteorology, p. 338. 1883. 
 
352 METEOROLOGY 
 
 region, and thus determining the length and intensity 
 of sunshine. 
 
 2. Altitude By which the effects of latitude may be to some 
 
 extent neutralised, for even in the tropics, at a height 
 of 16,000 feet, snow and ice may exist, the temperature 
 falling in ascending mountains 1 F. for every 300 feet. 
 
 3. Relative Distribution of Land and Water, and especially 
 
 the presence of vast tracts of either desert or ocean ; 
 the former accentuating extremes of temperature, and 
 the latter tempering them. 
 
 4 Presence of Ocean Currents, flowing from higher or lower 
 latitudes (as the case may be), and thereby qualifying 
 the climate. 
 
 5. Proximity of Mountain Ranges, and their influence on the 
 
 shelter from wind and on the rainfall. 
 
 6. Soil Its permeability or impermeability to moisture. 
 
 7. Vegetation. 
 
 8. Rainfall Its amount and annual distribution. 
 
 9. Prevailing Winds. 
 
 It will be necessary to consider these factors in more detail. 
 I. Latitude. A writer in Chambers's Encyclopedia (Art. 
 " Climate ") says : " The effect of the sun's rays is greatest where 
 
 FIG. 94. DIAGRAM ILLUSTRATING THE EFFECT OF THE PERPENDICULAR AND THE 
 OBLIQUE FALLING OF THE SUN'S BATS. 
 
 they fall perpendicularly on the surface of the earth, and 
 diminishes as their obliquity increases ; the surface which receives 
 any given amount of the sun's rays increasing with their increased 
 obliquity, as a'b' is greater than ab in the annexed figure ; whilst, 
 at the same time, the oblique rays being subjected to the influence 
 of a greater number of particles of the atmosphere, as c'a' is 
 longer than ca, a greater amount of their heat is absorbed before 
 
CLIMATE 353 
 
 they reach the surface of the earth at all. The greater or smaller 
 extent of surface receiving a certain amount of heat, also makes 
 important differences to arise from exposure by slope towards the 
 Equator or towards the nearer pole." 
 
 II. Altitude. It has been already shown (see Chapter IX., 
 p. 99 et seq.) that temperature decreases with increasing elevation 
 above sea-level, and that this is largely due to diminishing 
 density of the atmosphere, as well as to reduced humidity. Even 
 on the Equator perpetual snow and ice are found above a certain 
 height, and Quito, the capital of Ecuador, at an altitude of 
 9,451 feet, enjoys an eternal spring, the mean temperature of 
 the whole year and of every season being steady at 60 F. 
 
 The fall of temperature with altitude is not, however, uniform 
 at all places having the same latitude. A striking example of 
 this is met with in the climates of the Himalayan mountains. 
 On the southern slopes of this vast and gigantic chain the snow- 
 line, or limit of perpetual snow, is depressed by the precipitation 
 of large quantities of snow and rain from the moisture-laden 
 south-west winds coming from the Indian Ocean. These winds, 
 first chilled and deprived of their moisture, but afterwards 
 warmed by the latent heat set free in the condensation of their 
 vapour into rain or snow, cross the summits of the Himalayan, 
 peaks and descend towards the plains of Central Asia. Being 
 abnormally dry, these winds have an immense capacity for both 
 heat and moisture, and so, as they descend, they become rapidly 
 still warmer, and at the same time lick up both water and snow. 
 The result is that the snow-line on the northern slopes of the 
 Himalayas is at least 4,000 feet higher than it is on the southern 
 side. Strictly analogous phenomena attend the passage of a 
 south wind across the Alps, where it is called the Fohn ; of a south- 
 east wind across the mountainous interior of Greenland, bringing 
 with it comparative warmth to North Greenland and Smith Sound ; 
 of the " Chinook " wind of the Rocky Mountains in Western 
 Canada ; and of a " nor'-wester " across the mountains of New 
 Zealand. This last wind, taking its origin in the South Pacific, 
 deposits its moisture on the western slopes of the New Zealand 
 Alps, and appears in the Province of Canterbury on the east coast 
 as a very dry, and often a hot, wind, unaccompanied with rain. 
 
 23 
 
354 METEOROLOGY 
 
 It is said that, when the Fohn is blowing in the Alps, the snow 
 melts with marvellous rapidity, so that it is popularly called the 
 " snow-devourer " (Schneefresser). 1 
 
 III. Relative Distribution of Land and Water. In our study 
 of climate we may lay down as aphorisms 
 
 First. That hot air is lighter than cold air. 
 
 Secondly. That the rapidity with which the processes of 
 heating and cooling of air goes on is in direct proportion to 
 the amount of aqueous vapour contained in that air dry air 
 becoming heated or cooled more rapidly and more completely 
 than moist air, other conditions being alike. 
 
 Thirdly. That, consequently, the air over large areas of land, 
 being drier, becomes more rapidly heated in summer and more 
 rapidly cooled in winter than air which is in contact with exten- 
 sive water-surfaces ; and 
 
 Fourthly. That the radiation-heating power of dry land is 
 greater than that of water, as also the radiation-cooling power 
 of dry land is greater than that of water. 
 
 This group of facts is of paramount importance in climatology. 
 The effect of them upon the climate of the great continent of 
 Europe and Asia has already been described in Chapter XIII. 
 (see p. 156). 
 
 We can, indeed, form but little idea of the enormous changes 
 of temperature which take place in Central and Northern Asia 
 between the seasons of summer and winter. But that these 
 changes are sufficient to produce the great variation in baro- 
 metrical pressure on which depends the varying wind-system of 
 the continents of Europe and Asia in those seasons may be easily 
 shown by a comparison of the range of temperature between 
 July and January in an insular 2 climate like our own, and at 
 Yakutsk, in Siberia, which is situated close to the centre of 
 lowest and highest barometrical pressures in those months 
 respectively. At Dublin the mean temperature of July is about 
 
 1 For a full account of the Swiss Fohn wind, see a paper by Dr. Wild, 
 the Director of the Imperial Observatory at St. Petersburg, Ueber Fohn und 
 JKiszeit. Bern, 1868. 
 
 2 The terms Insular and Con'inen'al, as applied to climate, usually 
 signify merely that it is characterised by a small or by an extreme range of 
 temperature respectively, without any reference to the geographical position 
 of a place as regards the seaboard. 
 
CLIMATE 355 
 
 60 F., of January about 40 F. a range of only 20. The 
 corresponding mean temperatures at Yakutsk are 66 F. and 
 45 F. respectively a range of 111. For weeks in summer 
 the thermometer ranges between 80 and 90 at this place, while 
 in winter it may descend 90 below the freezing-point of water. 
 Well does Humboldt observe : l 
 
 " The inhabitants of the countries where such continental 
 climates prevail seemed doomed, like the unfortunates in Dante's 
 
 Purgatory 
 
 " ' A soffrir tormenti caldi e geli.' " 
 
 Or, as Milton has so admirably expressed it 
 
 "From beds of raging fire to starve in ice." 
 
 " The capability of man to endure variations and extremes of 
 temperature/' says Dr. Theodore Williams, " has been proved 
 to be very great, for General Greely states that at Fort Conger, 
 U.S.A., in February, 1882, he experienced the low temperature 
 of -66'2 F., and at another time, in the Maricopa Desert, 
 Arizona, he saw noted the air temperature of 114 F., while the 
 metal of his aneroid beside him as he rode assumed a temperature 
 of 144 F." 
 
 The lowest mean monthly temperature ever recorded is 88' 8 F. 
 ( 67 '1 C.) at Werchojansk (or Verkhoyansk), in Siberia, lat. 
 67-5 north, in January, 1886. This cold station lies in the valley 
 of the River Jana, 330 to 460 feet above sea-level. At Poplar 
 River, Montana, North America, the thermometer fell to 63*1 
 in January, 1885. 
 
 The principal laws of distribution of annual range of tempera- 
 ture, given by Professor A. Supan in a paper which appeared 
 in the first volume of Kettler's Zeitschrift fiir wissenschaftliche 
 Geographic, are thus summarised by Dr. R. H. Scott : 
 
 1. The annual range of temperature increases from the Equator 
 towards the poles, and from the coast towards the interior of 
 a contin3nt. It is greatest 100 F. and upwards in Siberia, 
 near Yakutsk. It is least under 20 F. over almost the 
 whole of the sea surface of the globe, in South America and 
 
 South Africa. 
 
 1 Kos.nos, vol. i., p. 352. 
 
 23-2 
 
356 METEOROLOGY 
 
 ^ 2. The regions of extreme range in the northern hemisphere 
 coincide approximately with the districts of lowest temperature 
 in winter. On the whole, the range curves in their course resemble 
 the isotherms of January. 
 
 3. The range is greater in the northern than in the southern 
 hemisphere. 
 
 4. In the middle and higher latitudes of both hemispheres, 
 with the exception of Greenland and Patagonia, the western 
 coasts have a less range than the eastern. 
 
 5. In the interior of the continents the range, in mountainous 
 districts, diminishes with the height above the sea. 
 
 Probably nowhere is the influence of the ocean in restricting 
 the annual range of temperature more marked than off the ex- 
 treme south-west coast of England. In the Scilly Isles the mean 
 temperature of the sea surface ranges between 49 in February 
 and 61 in August that is, through 12. The mean tempera- 
 ture of the air is 46*3 in January and 61 '5 in July a mean 
 annual range of only 15'2. At Yarmouth, also on the sea, but 
 not in the ocean, the temperature of the sea surface ranges from 
 37 in January to 61 in July the mean temperature of the air 
 being 37 '9 and 62*5 in the months named a mean annual 
 range of 24-6. 
 
 Perhaps no single element has a greater influence upon climate 
 than the presence of water. Its specific heat 1 is four times 
 that of dry land, and consequently it absorbs heat more slowly, 
 stores up a larger amount of it, and parts with it less rapidly. 
 Again, owing to condensation by cold, surface water, when 
 exposed to low temperatures, sinks, and its place is taken by the 
 deeper strata of warmer liquid. Equilibrium of temperature is 
 thus maintained in the neighbourhood of extensive areas of 
 water. In warm weather evaporation from water surfaces tends 
 to cool the superincumbent air, to increase the humidity, and to 
 fill the atmosphere with clouds. Hence the equable, cloudy, 
 and moist climates of seaboards. Extensive lakes produce 
 similar effects on a smaller scale. Thus, in the Canadian winter, 
 what may be a storm of rain on the shores of Lake Superior 
 
 1 The specific heat of a substance is the number of units of heat required to 
 raise the temperature of 1 pound of it by 1. 
 
CLIMATE 357 
 
 is often a fall of snow a few miles inland. When districts of 
 country are partly covered with water, the climate is rendered 
 damp and cool, for the evaporation from wet earth is much 
 greater than that from a uniform water surface. Under such 
 circumstances there is reason to believe that a climate would 
 even be improved by changing marshes or morasses into sheets 
 of open water. Like good results often follow the carrying out 
 of effective drainage works, and there can be no doubt that, if 
 the water-shed of the Shannon, for example, were properly 
 drained, the climate of the whole of Ireland would be made drier 
 and warmer. This is a subject of vital importance to all who 
 have agricultural interests in view, as well as to those who have 
 regard to the public health. 
 
 In estimating the effect of water surfaces upon climate it should 
 not be forgotten that, while fresh water attains its maximal 
 density at 39'2F., and freezes at 32-0 F., sea water continues to 
 contract until it is chilled down to 26*2 F., and does not freeze 
 above 28*4 F. The result is that sea water in the open will 
 not begin to freeze on its surface until all its depths have been 
 cooled nearly to freezing-point, whereas fresh water needs to be 
 chilled throughout only to 39*2 F. before ice commences to form 
 on its surface. 
 
 Turning to the question of the influence of dry land on climate, 
 we find that the configuration of the surface of the ground exercises 
 an effect second only to that of the amount of land. Thus, as a 
 rule, when the surface slopes away from the sun, the rigour of 
 the climate is intensified. We have especially striking examples 
 of this in North Germany, Siberia, and those parts of North 
 America which slope towards Hudson Bay. The converse is 
 also true, and is exemplified in the excessive summer heat of 
 Northern Italy, India, and Southern China. Most winter resorts 
 are situated on grounds having a southern, or solar, aspect. 
 But, leaving out of consideration the direct influence of the 
 presence or absence of the sun's rays, the cooling of the air by terres- 
 trial radiation is found to affect localities in very different degrees. 
 Where the surface is uniformly level, as in the case of plains or 
 table-lands, radiation proceeds uniformly, and the whole district 
 is equally chilled. If the air is calm and the sky clear, radiation 
 
358 METEOROLOGY 
 
 goes on rapidly, and the temperature falls. Should the sky be 
 clouded and the air in motion, radiation is uniformly checked. 
 The alternations in either case are similar over the whole area 
 of level ground. In hilly or mountainous districts radiation acts 
 as before, but the air which is cooled becomes specifically heavier, 
 and immediately commences to flow down the mountain sides. 
 As it is replaced by warmer air, the temperature remains com- 
 paratively high and uniform. On high ground, also, the atmo- 
 sphere is seldom calm, so that radiation is usually more or less 
 checked, and warmth is maintained. Valleys, however, experi- 
 ence extreme variations of temperature for two reasons first, 
 because the cooled air flows down into them from the surrounding 
 high grounds ; and, secondly, because, being so much shut in, they 
 are fully exposed in the absence of wind to the influence of radiation. 
 
 Great diurnal range of temperature is a marked feature in 
 inland districts, particularly when the earth's surface is desert 
 or scantily dotted with vegetation. Thus at Mooltan (where the 
 thermometer ranges yearly from 29 to 126 F. in the shade), 
 Rawal Pindi, and other stations in the Punjab, the daily range 
 in April and November may amount to 40 F. The same state 
 of things is observed in Egypt, on the steppes of Southern Russia, 
 and on the prairies of North America. Even in the British Islands 
 in anticyclonic weather, with a dry atmosphere and easterly wind 
 in spring, a large diurnal range of temperature is often noticed. 
 For example, at Nairn, in the north-east of Scotland, the thermo- 
 meter rose to 72 F. in the screen on May 5, 1909, but fell to 32 F. 
 in the course of the following night. On Good Friday, April 17, 
 1908, Mr. Sydney Wilson recorded at Perth a range of 40*2 F. 
 in a few hours. About 6 a.m. of that day a minimum reading 
 of 27*8 F. occurred, whereas about 2 p.m. the thermometer had 
 risen to a maximum of 68 F. 
 
 A third instance of extreme diurnal range of temperature may 
 be given. At Marlborough, Wiltshire, during the great heat- 
 wave of 1906, the thermometer rose on August 30 from a minimum 
 of 37*1 F. in the early morning to a maximum of 83' 6 F. some 
 twelve hours later a swing of 46*5. 
 
 IV. Ocean Currents of either warm or cold water modify 
 climate in a remarkable degree. They are named according to 
 
CLIMATE 359 
 
 the direction towards which they flow. The ameliorating effects 
 of the presence of a vast water-surface are intensified in the case 
 of oceans, such as the Pacific and Atlantic, in winter-time, by 
 the setting towards the Arctic and Antarctic regions of immense 
 surface-currents of warm water. The best known of these 
 currents is the North Atlantic surface drift, commonly, but not 
 quite correctly, called the " Gulf Stream," which flows north- 
 eastward along the western shores of the British Isles and 
 Norway, and to which we are so largely indebted for our 
 wonderfully mild British winters. 
 
 " Its climatic effect," says Mr. J. Knox Laughton, M.A., 1 
 " when stated in measures of heat, is stupendous it is the very 
 poetry and romance of arithmetic." He proceeds to show that 
 the heat brought by the Gulf Stream into the North Atlantic has 
 been fairly estimated as not less than one-fifth of the whole heat 
 possessed by the surface water of that division of the ocean. As- 
 suming, with Sir John Herschel, that the temperature of space is 
 239 F. below zero, and taking the existing temperature of the 
 North Atlantic as 56 F. above zero, we find that the heat which it 
 actually has corresponds to a temperature of 295 F. (namely, 
 239+ 56 F.), the fifth part of which is 59 F. If, then, the fifth 
 part of its heat that is, the heat derived from the Gulf Stream 
 were taken away from it, the surface water of the North Atlantic 
 would have an average temperature of 3 F., or 35 F. below 
 the freezing-point of fresh water. 2 It is roughly estimated that 
 about five billions of cubic feet of water are hourly poured through 
 the Straits of Florida into the North Atlantic. This water has 
 at the time an average temperature of not less than 65 F., but 
 after performing a circuit in the North Atlantic, it returns to 
 the Tropics as an undercurrent with an average temperature not 
 above 40 F. It has imparted to the air over the North Atlantic 
 the heat corresponding to a difference in temperature amounting 
 to 25 F. Now, the British standard measure of heat the 
 thermal unit is the quantity of heat required to raise the tempera- 
 ture of 1 pound of water by 1 F., while a cubic foot of water 
 weighs about 64 pounds. With these data, we find that the heat 
 
 1 In a lecture on " Air Temperature : its Distribution and Range." 
 Modern Meteorology. Edward Stanford. 1879. 
 
 2 Croll's Climate and Time, p. 35 et seq. 
 
360 METEOKOLOGY 
 
 thrown out by the Gulf Stream every hour into the air of the 
 North Atlantic is 25 x 64 x 5,000,000,000,000 thermal units. 
 
 But every thermal unit, according to the "Law of Equivalence/' 
 experimentally established by Dr. Joule, of Manchester, is capable 
 of lifting a weight of 772 pounds through a height of 1 foot. 
 Consequently, the heat hourly dispersed from the water of the 
 Gulf Stream, if stored up and applied as power, would be capable 
 of lifting each hour 772 x 25 x 64 x 5,000,000,000,000 pounds 
 through a height of 1 foot that is, of doing the work of steam- 
 engines having an aggregate horse-power of 3,119,000,000,000, 
 a power equal to that of nearly 400,000,000 ships such as our 
 largest ironclads. 1 
 
 Ocean currents of cold water also exist. Of these the most 
 notable, probably, is that which flows out of Baffin's Bay down 
 the eastern shores of North America, and which is known as the 
 American Arctic Current. Its cooling influence is felt as far 
 south as Cape Cod, in latitude 42. 
 
 On Friday, August 13, 1909, I left Liverpool for Quebec as a 
 passenger on board the Royal Mail Steamer Empress of Ireland, 
 of the Canadian Pacific Railway's Atlantic Service. Her com- 
 mander was Captain J. V. Forster, R.N. Reserve, who courteously 
 placed at the disposal of the passengers full information relating 
 to the ship's " log." This included observations on the state of 
 the barometer, the temperature of the air and of the sea, the 
 direction and force of the wind, made at the various four-hourly 
 " watches," besides the usual noontide record of the distance 
 run in knots and the latitude and longitude of the ship's meridian 
 position. From 8 a.m. on Saturday, August 16, till the afternoon 
 of the following day, the sea-temperature remained steady at 
 about 56 F. From the afternoon of Sunday, the 17th, until 
 4 p.m. on Monday, the sea-water ranged from 54 to 49 F. The 
 position of the ship at the latter hour was approximately 55 
 N. latitude, 40 W. longitude. Four hours later at 8 p.m. of 
 the 16th the temperature of the sea was only 38 F., a fall of 
 11 in four hours. In the course of the next afternoon Tuesday, 
 the 17th several icebergs were sighted, two of unusual size. 
 The mathematicians on board calculated that one berg to the 
 1 Croll's Climate and Time, p. 25. 
 
CLIMATE 361 
 
 northward was distant seven miles on the beam N.N.W., was 
 220 feet high, and 900 feet long. At noon on this day the sea- 
 temperature was 40 F. ; by 8 p.m. it had fallen to 33 F. At 
 10.30 p.m. the lights of the Strait of Belleisle were sighted. It 
 was quite evident that the ship had passed on the afternoon of 
 the 16th from the warm north-easterly surface drift current into 
 the chill waters of the American Arctic current, setting south 
 from the west and east coasts of Greenland past the inhospitable 
 shores of Labrador. The " Pilot Chart " for September, 1909, 
 of the North Atlantic Ocean, issued by the Hydrographic Office, 
 Washington, D.C., shows that icebergs in large numbers were 
 reported during August in and near the Strait of Belleisle, and 
 in smaller numbers on the Grand Banks. From August 10 to 
 13 the Strait was full of bergs, the British s.s. Montfort sighting 
 seventy-nine, and the British s.s. Hesperian eighty-five, of which 
 twenty-two were west of Greenlet Island, Newfoundland. On 
 August 2, Captain Best, of the British s.s. Shimosa, reported a 
 piece of ice, 18 feet long and 5 feet wide, in latitude 37 16' N., 
 longitude 42 06' W. 
 
 An Arctic current of far less magnitude flows into the North 
 Pacific through Behring's Straits. It chills the air over Kam- 
 chatka and Japan in the summer season, throwing the isotherms 
 over the North Pacific into remarkable loops in July, when the 
 current is strongest owing to the melting of the polar ice. In 
 the southern hemisphere the oceanic polar currents form a still 
 more striking feature in the physical geography. According to 
 Sir F. Evans, 1 all the surface water between the Antarctic Circle 
 and the parallel of 45 S. seems to drift northwards and east- 
 wards, causing the isotherms on the western coasts of America, 
 Africa, and Australia to dip down towards the Equator (R. H. 
 Scott). The best-known of these currents is the Peruvian, or 
 Humboldt's current, which washes the west coast of South America. 
 Dr. Scott also points out that the influence of the Antarctic 
 Atlantic current on the west coast of Africa is such that the 
 temperature of the sea near Cape Town is sometimes 20 F. 
 lower than in the corresponding latitude on the eastern side of 
 the continent. 
 
 1 British Association Report, p. 175. 1876. 
 
CHAPTER XXIV 
 CLIMATE (continued) 
 
 V. Proximity of Mountain Ranges. Dr. Alex. Buchan, in his 
 Introductory Textbook of Meteorology, 1 stated that, apart from 
 diverting the winds from their course, the chief effect which 
 mountain ranges have upon the temperature (and so upon 
 climate) is to drain the winds which cross them of their moisture. 
 Colder winters and hotter summers in places to the leeward, as 
 compared with places to the windward, are thus caused ; for the 
 protecting screen of aqueous vapour is partially removed by con- 
 densation, and so the country to the leeward of a mountain 
 chain becomes more fully exposed to both solar and terrestrial 
 radiation. For the same reason, the rainfall is lessened in such 
 sheltered localities, although it is, of course, proportionately 
 increased on the windward side of the mountains. Dr. Theodore 
 Williams cites the extraordinarily dry climate of Colorado, which 
 lies under the lee of the Rocky Mountains, as an instance of the 
 influence of a mountain-range on climate. Nearer home we have 
 similar examples on a much smaller scale in several parts of the 
 British Isles : heavy and continuous rainfalls in the mountainous 
 districts of Kerry, Cumberland, and the west of Scotland, con- 
 trasting with comparatively dry climates in Dublin, the Low- 
 lands of Scotland, and the coasts of the Moray Firth, Nairn- 
 shire, and the Carse of Sutherland. These last-named districts, 
 according to Dr. Scott, owe their good fortune mainly to the fact 
 of their lying on the lee-side of an extensive mountain district. 
 As regards Dublin, a range of mountains lies a few miles south of 
 the city, with summits varying in height from 1,000 to more than 
 2,500 feet. This mountain chain intercepts the vapour-laden 
 1 William Blackwood and Sons : Edinburgh and London. 1871. P. 73. 
 
 362 
 
CLIMATE 363 
 
 winds at all points between south-south-east and south-west. 
 In consequence, the rainfall is diminished and the sky is com- 
 paratively cleared during the continuance of the southerly and 
 south-westerly winds which so frequently prevail. Dublin and 
 its neighbourhood are the only part of Ireland where the annual 
 rainfall falls short of 30 inches it is about 28 inches and this 
 depends on the geographical situation of the city on the east 
 coast and to the leeward of high lands, grouped into mountains 
 to the south-east, south, and south-west, whereby the rain- 
 bearing winds are drained of their superabundant moisture 
 before they reach the valley of the Liffey and the plains lying 
 north of that river. This relatively dry region stretches along 
 the east coast northwards to Dundalk Bay. 
 
 It is everywhere recognised at the present day that the chief 
 cause of the condensation of the aqueous vapour of the atmo- 
 sphere by a mountain chain is the adiabatic cooling of the rising 
 mass of air, for a current of air impinging on high ground must, 
 in order to pass over it, necessarily rise. A thoroughly scientific 
 attempt to investigate the process of adiabatic cooling of ascend- 
 ing air currents quantitatively was made some years ago (in 1901) 
 by Professor F. Pockels, of the School of Technology, Dresden, 
 Germany. His memoir on the subject, entitled " The Theory of 
 the Formation of Precipitation on Mountain Slopes," appeared 
 originally in Annalen der Physik, 1901. * A translation of the 
 article will be found in the Monthly Weather Review, of the United 
 States of America Department of Agriculture, for April, 1901. 2 
 
 VI. Soil. With regard to the absorbing power of heat pos- 
 sessed by soils, Schiibler has arranged them in the following 
 order, 3 100 being assumed as the standard : 
 
 Sand with some lime, 100 ; pure sand, 95*6 ; light clay, 76'9 ; 
 gypsum, 73-2 ; heavy clay, 71*11 ; clayey earth, 68'4 ; pure clay, 
 66-7; fine chalk, 61'8; humus, 4 49. This list shows the high 
 absorbing power of the sands, and the comparative coldness of 
 
 1 Vol. iii., pp. 459-480. 2 Vol. xxix., No. 4, pp. 152-159. 
 
 3 Parkes's Manual of Hygiene, p. 312. Fourth edition. 
 
 4 Humus is the organic matter of the soil, which is made up of the 
 products of the decomposition of vegetable substances. These products 
 may be arranged in three classes: (1) Those soluble in water crenic, apo- 
 crenic, and ulmic acids ; (2) those soluble in alkaline solutions, but not in 
 pure water humic and geic acids ; (3) those insoluble humin and ulmin. 
 
364 METEOKOLOGY 
 
 the clays and humus. The absorbing power of water possessed 
 by soils varies in a similar manner. Sands retain but little 
 water, clays about ten to twenty times as much as sands, and 
 humus double as much again. Clays and humus are compara- 
 tively unsuitable as sites for building, owing to their characters 
 of coldness and dampness. In some diseases they are very 
 injurious, as, for example, in phthisis, rheumatism, and catarrh. 
 If damp soils be exposed to a high temperature, they may 
 cause ill-health, owing to decomposition of the organic matter 
 mixed with them. Marshy soils, alluvial soils, old estuaries, and 
 deltas contain much organic matter, and should be regarded with 
 suspicion. Peaty soils also are largely composed of organic 
 matter, but they are not so injurious to health, owing probably 
 to the preservative properties of peat. Granite, metamorphic 
 and trap rocks, clay-slate, chalk, sandstone, gravels, and the pure 
 sands, are healthy, and suited for building sites. The limestone 
 and magnesian limestone rocks and mixed sands are only 
 moderately healthy. 1 
 
 Three diseases, leaving malaria and mosquitoes out of the 
 question, appear to be intimately connected with the presence of 
 water in the soil. In 1862 Dr. Bowditch, of Boston, U.S., drew 
 attention to the relations between the prevalence of phthisis and 
 the amount of sub-soil water. His researches were amply con- 
 firmed by Dr. (afterward Sir George) Buchanan, 2 who discovered 
 that the death-rate from phthisis in various towns in England 
 was greatly reduced in consequence of efficient drainage and 
 removal of the sub-soil water. Many years ago Pettenkofer of 
 Munich advanced the doctrine of the ultimate dependence of 
 enteric fever and of cholera on the varying level of the sub-soil 
 or " under-ground " water, 3 the most dangerous period, accord- 
 ing to him, being that of the sinking of the water after a previous 
 rise. 
 
 The high absorbent power of loose sandy soils is doubtless 
 due to the presence of large quantities of imprisoned air, which 
 converts the sand into a bad conductor of heat. Hence such 
 soils heat readily in summer and cool readily in winter near the 
 
 1 Cf. Parkes, Manual of Hygiene, p. 314 et seq. 
 
 2 Ninth and Tenth Reports of the Medical Officer of Health to the Privy 
 Counci'. 3 Zeitschrift fur Biologie, 1868. 
 
CLIMATE 365 
 
 surface ; but these extremes of temperature do not penetrate by 
 conduction to any depth. Dr. Williams states 1 that the sandy 
 soil in an Arabian or Egyptian desert may be heated to 120, 
 140, or even 200 F., and when the particles of this hot sand 
 are carried through the air by the terrible simoom, the shade 
 temperature may rise to 125 F. On the other hand, Dr. Buchan 
 mentioned in his Introductory Textbook of Meteorology, 2 that in 
 Scotland, for a period of nine years, the temperature at 3 inches 
 below the surface fell to 26'5 F. in loose sandy soils, but at a 
 depth of 12 inches the freezing-point was only once observed. 
 In clay soils, at 3 inches the lowest temperature recorded was 
 28 F., whilst at 12 inches the temperature often fell to freezing, 
 and even at 22 inches 32 F. was more than once recorded. 
 
 VII. Vegetation. A district covered with a luxuriant growth 
 of plants and forest trees has a comparatively uniform and 
 temperate climate. By day the heat is lessened, because the 
 vegetation intercepts a large proportion of the sun's rays, which 
 would otherwise heat the earth's surface ; also, because the 
 evaporation from leaves and grasses renders heat latent, and so 
 keeps the atmosphere cool. By night radiation from the surface 
 of the ground is checked, and so the fall of temperature is 
 diminished. Forests control evaporation and increase the 
 humidity of the air ; they are also said to increase the rainfall, 
 but this seems not to be satisfactorily established. As moist air 
 prevents excessive heat in summer and excessive cold in winter, 
 forests are thus seen to be of use in mitigating extremes of climate. 
 In winter they afford shelter from storms, and in tropical climates 
 the spread of malaria is said to be prevented by the interposition 
 of a belt of trees between a malarial swamp and a village. Sir 
 Patrick Manson suggests, in explanation, that the trees may 
 filter out the mosquitoes by affording them protection from 
 winds, and so the houses on the leeward of the trees escape 
 infection. 3 
 
 The third number of Petermanns Mittheilungen for 1885 con- 
 tained an article of exceptional interest by Herr A. Wojeikof 
 on the influence of forests on climate. The first step towards a 
 
 1 Lum^eian Lectures, 1893. 2 P. 46. 1871. 
 
 3 Tropical Diseases, p. 104. Fourth edition. By Sir Patrick Manson, 
 K.C.M.G., M.D., LL.D.(Aberd.). London : Cassell and Co. 1907. 
 
366 METEOROLOGY 
 
 scientific investigation of this subject was taken when the 
 Bavarian forest meteorological stations were established, and 
 when Prussia, Alsace-Lorraine, France, Switzerland, and Italy 
 followed the example. As a general rule, it may be laid down 
 that during the warmer season, (1) the temperatures of the earth 
 and air are lower in the forests than in contiguous woodless 
 places ; (2) their variations are less ; (3) the relative humidity is 
 greater. The influence of forests in diminishing evaporation 
 from water and the soil is so great that it cannot be accounted 
 for solely by the lower temperature of the warm months, the 
 greater humidity, or even by the shade the protection from 
 the wind afforded by the trees is regarded by Herr Wojeikof as 
 more important than all these factors together in reducing the 
 amount of evaporation. With respect to the influence of forests 
 on rainfall and snowfall, there is as yet only a single series of 
 observations supplying comparative statistics, and extending 
 over a sufficiently long period namely, six years. These were 
 taken in the neighbourhood of Nancy by the pupils of the School 
 of Forestry of that city, under the direction of M. Mathieu, sub- 
 director of the school. These observations, reported in Poly- 
 biblion, 1882, prove that 
 
 1. Forests increase the quantity of meteoric waters which fall 
 on the ground, and thus favour the growth of springs and of 
 underground waters. 
 
 2. In a forest region the ground receives under cover of the 
 trees as much water as, or more than, the uncovered ground of 
 regions with little or no wood. 
 
 3. The cover of the trees of a forest diminishes to a large 
 degree the evaporation of the water received by the ground, and 
 thus contributes to the maintenance of the moisture of the latter, 
 and to the regular flow of springs. 
 
 4. The temperature in a forest is much less variable than in 
 the open, although, on the whole, it may be a little lower ; but 
 the minima are there constantly higher, and the maxima con- 
 stantly lower, than in regions not covered with wood. 
 
 M. Fautrat, when sub-inspector of forests at Senlis, made 
 observations on forestal meteorology during four years. These, 
 although conducted on a different method, fully corroborate 
 
CLIMATE 367 
 
 those of M. Mathieu in several respects. He adds the following 
 interesting remarks : Rain falls most abundantly over forests 
 with trees in full leaf ; the humidity of the air is much higher 
 over masses of Pinus sylvesiris than over masses of leafed species ; 
 and the leafage and branches of leafed trees intercept one-third, 
 and those of resinous trees one-half, of the rain water, which after- 
 ward returns to the atmosphere by evaporation. 
 
 The influence of forests on the dampness of the soil and on the 
 yield of springs was discussed by Professor H. Gravelius in 
 Petermann's Mittheilungen for March, 1901. The result of 
 the Professor's study of the recent Russian literature on 
 the subject is to show very clearly that forests do not 
 preserve the moisture of the ground or promote the flow of 
 springs. All the experiments showed that the level of ground 
 water was lower under great forests than in open country, even 
 in the Russian steppe. The forest appears to protect the ground 
 altogether from light rains, which are absorbed by the foliage 
 or evaporated from the immense surface formed by the leaves 
 as the drops trickle downwards. Heavy rains reach the ground 
 nearly as freely as in open land, but here the tree roots play their 
 part, and the transpiration of the vegetation keeps the soil dry 
 to a considerable depth. These facts go to prove the immense 
 value of forests on mountain sides for checking floods, and of the 
 planting of woods in swampy country as an aid to drainage in 
 drying the land. 1 
 
 Herr Wojeikof endeavoured to ascertain the influence of 
 forests on the climatic conditions of their neighbourhood in 
 the western parts of the Old World, between the 38th and 52nd 
 degrees N. latitude, the places selected being in all cases in the 
 open. Thus, for the 52nd degree, eight stations were taken 
 between Valentia Island, in Ireland, on the west, and the 
 Kirghiz steppe on the east ; for the 50th, Guernsey on the west, 
 Semipalatinsk on the east, and ten other intervening stations ; 
 and so on for each two degrees of latitude to 38 N. The general 
 result of the observations at fifty stations in six different degrees 
 of latitude is that in Western Europe and Asia large forests have 
 a great effect upon the temperature of places near them. The 
 1 Sjmons's Meteorological Magazine, vol. xxxvi., p. 116. 1901. 
 
368 METEOROLOGY 
 
 normal increase of temperature as we travel eastward from the 
 Atlantic Ocean to the interior of the Continent is not merely 
 interrupted by the influence of forests, but places far removed 
 from the coast through that influence enjoy a cooler summer 
 than those actually on the sea. A striking example of this is 
 Bosnia, where the summer is 4'5 to 8-1 F. cooler than in Herze- 
 govina. Bosnia, separated by lofty mountain ranges from the 
 sea, has extensive forests ; Herzegovina, on the contrary, is 
 almost disafforested. Even on the Island of Lissa, where under 
 the full influence of the Adriatic Sea the summer should be 
 cooler, the temperature is more than 1'8 F. higher than it is in 
 Bosnia. In Portugal, which is poor in forests, the temperature 
 rises very rapidly towards the interior during the almost rain- 
 less summer. The heat is still greater in stony Attica, notwith- 
 standing the proximity of the sea. On the eastern shore of the 
 Caspian, owing to the desert of sand and stone, the summer 
 temperature is extremely high, whereas at Lenkoran, on the 
 western shore of that vast inland sea, a cool though dry summer 
 is enjoyed. In the great Lenkoran forest vegetation is more 
 luxuriant than in any part of Europe, for a tangled mass of climb- 
 ing plants encircles the trees, so that it is always humid in the 
 forest. Yet here the rain curve is a sub -tropical one very little 
 rain falling in the summer, but large quantities in autumn and 
 winter. The water is stored up in the forest, so maintaining 
 evaporation during the summer droughts. 
 
 " To sum up : Forests exercise an influence on climate which 
 does not cease on their borders, but extends over a larger or 
 smaller adjacent region, according to the size, kind, and position 
 of forest. Hence man, by afforestation and disafforestation, can 
 modify the climate around him ; but it is an extreme position to 
 hold that by -afforestation the waste places of the earth can be 
 made fertile. There are places incapable of being afforested, 
 which would not give the necessary nourishment to trees " 
 (Nature, vol. xxxii., 1885, p. 115). 
 
 Some thirteen years ago (February 15, 1897) the following 
 query was addressed to me by a distinguished Medical Journalist : 
 " Is it regarded as insanitary to have ivy and other evergreens 
 clinging to walls of hospitals ?" My reply was as follows : 
 
CLIMATE 369 
 
 An ivy-clad wall is a dry wall. Rain is intercepted when 
 beating against such a wall, and the rootlets (or tendrils) of the 
 ivy absorb all moisture from the mortar, bricks, and even stones, 
 of which the wall is built. 
 
 An ivy-clad wall is a warm wall. Radiation of heat from 
 the wall is interfered with, and temperature does not fall in cold 
 weather under the shelter of the ivy within several degrees of 
 the temperature of the open air. 
 
 Other evergreens produce similar effects, though in a less 
 degree, owing to their inherent greater moisture and less close 
 foliage, which is also less regular. 
 
 An ivy or evergreen covering has the disadvantage of har- 
 bouring insect life near windows and doors. In any case, such 
 a covering should be periodically trimmed, so as not to interfere 
 with the free access of both air and light to windows and doors. 
 
 VIII. Rainfall. The influence of precipitation on climate has 
 already been discussed at some length in Chapter XVIII. (pp. 221, 
 242, 261-263). In estimating it, regard should be had to the 
 monthly, seasonal, and yearly rainfall, both in individual periods 
 and on an average of many such periods. Equally important is it to 
 , know the number of " rain days " (or days upon which '005 inch of 
 rain or upwards is measured in the gauge) which may be expected 
 to occur in each month, season, or year. Nor should the probability 
 of heavy or torrential rainfalls 1 inch or upwards be left out of 
 account. It is evident that a moderate rainfall spread over many 
 days throughout the year may constitute a wet climate, while a 
 still heavier rainfall restricted to a given season may characterise 
 what is essentially a dry climate. Another element worthy of 
 consideration is the frequency of thunderstorms, so often attended 
 by torrential rains, and of hailstorms. Lastly, the seasonal 
 limits of snowfall should be investigated, and the average and 
 extreme duration of an unbroken snow-covering. 
 
 This last topic has been ably handled by Dr. Alexander Wojei- 
 kof in a paper on the " Influence of Accumulations of Snow on 
 Climate," which was read before the Royal Meteorological 
 Society, June 17, 1885. * It has already been shown in these 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. xi., p. 299. 
 1885. 
 
 24 
 
370 METEOROLOGY 
 
 pages that snow is a bad conductor of heat, and accordingly 
 protects the underlying soil from excessive cold. But much 
 depends on the structure of the snow. If it consists of loosely 
 piled small feathery flakes, which entangle air in large quantity, 
 it is a thoroughly bad conductor, and so affords most protection. 
 If, however, by alternate thawing and freezing, it solidifies into ice 
 assuming the form called in Germany Firn, and in France neve 
 it is a far better conductor of heat, and the underlying soil will 
 quickly freeze. 
 
 Again, the air over a snow-covered surface will become ex- 
 tremely cold first, because the snow cuts off from it the warmth 
 of the ground, and, secondly, because dry feathery snow is a 
 good radiator of heat. Then, as there is but little dust in the air 
 over a snow-covered country upon which the solar rays can act, 
 these rays will be unable by themselves to thaw a deep snow 
 covering. How, then, is it that the winter snow does melt in 
 the northern parts of Europe, Asia, and North America ? Dr. 
 Wojeikof has no doubt that the thaw is first caused by winds 
 from warmer quarters, or from open oceans. These warm winds 
 cause the upper layer of snow to melt. After it has been frozen 
 again, it is changed to neve that is, to a condition in which it is 
 somewhat diathermanous to solar heat, and radiates heat much 
 less freely. Once this happens, the melting of snow goes on much 
 more easily. To a small extent the melting of the snow may be 
 helped by dust brought by the winds from continental areas 
 already free from snow. No doubt a great quantity of heat is 
 expended on the melting of the snow, and so the warm winds are 
 chilled, and lose their power of thawing the snow. But near the 
 border of the snow-covered country, when the snow has mostly 
 melted there, the surface of the ground can be heated by the sun. 
 and thus become a source of heat for the country lying still 
 farther to the northward. The melting of the snow is progressive 
 from, say, February to June in the northern hemisphere. It 
 begins close to seas which do not freeze, and continental areas 
 which are not permanently covered with snow even in mid- 
 winter. Thence it proceeds intermittently by leaps and bounds 
 until all the lowlands of our hemisphere are freed from their 
 snow-covering. 
 
CLIMATE 371 
 
 In high southern latitudes no such happy event takes place. 
 Sir James Ross proved that the mean temperature of the shores 
 of the Antarctic Continent, or the islands bound together by 
 glacier ice, and so appearing like a continent, is much below the 
 freezing-point even at midsummer. There is no notable melting 
 of snow, and what is melted is very soon replaced by fresh snow. 
 The geographical position explains this. The shores of the Ant- 
 arctic Continent are washed by an ocean, the surface water of 
 which has a temperature below freezing-point to about 62 S. 
 even in summer, while they lie at a distance of 20 or more from 
 any land area of lower latitude which could supply warmth for 
 the thawing of their eternal snows. We see, then, that the sun's 
 rays are of themselves unable to raise the temperature above 
 freezing-point, notwithstanding the nearness of the sun to the 
 earth in the summer of the southern hemisphere. 
 
 Mr. L. C. Bernacchi, Physicist to the National Antarctic 
 Expedition of 1902-04, states that the mean temperature ob- 
 served by Lieutenant C. W. R. Royds, R.N., at the winter 
 quarters of the Discovery, in latitude 77 50' 55" S., longitude 
 166 55' 45" E., for the two years from February 9, 1902, to 
 January 31, 1904, was - 1*7 F. The lowest mean temperature 
 for any one month was 21*1 F. for July, 1903; and the 
 highest mean temperature 26-1 F. for January, 1903. The 
 absolute maximal temperature observed in 1902 was 39*0 F., 
 in December, and in 1903 42'0 F., also in December. The 
 summers were very cold ; only a few days gave a mean tempera- 
 ture above freezing. Mr. Bernacchi states that the large mass of 
 land ice and the remarkable dryness of the air would seem to 
 be partly responsible for this. The air, he adds, is very trans- 
 parent, fogs are infrequent, and precipitation is slight. An 
 outstanding feature is the abundance of bright sunshine in the 
 summer. The total of 490 hours in December, 1903, is equal to 
 66 per cent, of the possible amount. On one occasion there was 
 continuous sunshine for 87 hours. Solar radiation temperatures 
 were very high, although the sun, when it attained its greatest 
 altitude in December, was more than 60 from the zenith. The 
 mean black bulb temperatures for December and January are 
 only 14 'F. less than the corresponding means (June and July) 
 
 242 
 
372 METEOROLOGY 
 
 at Madras, with an almost vertical sun. The maximal solar 
 radiation reading came within 3 F. of the Madras maximum. 
 
 In our northern hemisphere the melting of the snow keeps 
 the temperature for a long time near freezing-point. This is 
 the reason why April is so much colder than October in Central 
 European Russia, in Canada, and in the more northern of the 
 United States. Hence also the keenness of the easterly winds 
 of spring in Western Europe, even in the British Islands. It 
 .is clear that at the period when the mean temperature begins to 
 rise above the freezing-point, very much depends on the store 
 of cold existing in the vicinity in the form of snow and ice. The 
 larger it is, the slower and more irregular will be the rise of 
 temperature. At the beginning of winter, also, a heavy fall of 
 snow over a large continental area intensifies and gives a per- 
 manency to succeeding cold. 
 
 IX. Prevalent Winds. The division of winds into (1) Per- 
 manent, (2) Periodic, and (3) Variable, must now be considered. 
 Dr. Theodore Williams well observes that, beyond the use of 
 winds for propelling vessels and machinery, they serve a dis- 
 tinctly hygienic object in dispersing noxious exhalations, whether 
 animal or vegetable, in permitting free evaporation, and thus 
 preventing accumulation of moisture, and maintaining the cir- 
 culation of the air, which is necessary for the purification of 
 the atmosphere. Their influence upon climate is indisputable. 
 They raise or lower temperature, increase or diminish humidity, 
 cause or prevent rainfall, interrupt sunshine by bringing up 
 clouds, or clear the sky when descending from the higher strata 
 of the atmosphere. 
 
 Apart from the permanent aerial currents, which are called 
 the " Trade Winds," and which blow within the Tropics as a 
 north-east wind north of the Equator, and as a south-east wind 
 south of the Equator ; and from the periodic aerial currents, of 
 which the most striking examples are the Indian monsoons the 
 south-west monsoon (of summer) and the north-east monsoon 
 (of winter), there are certain local winds, which prevail occasionally 
 and produce very decided effects upon both animal and vege- 
 table life. This class may be subdivided into occasional cold 
 winds, prevalent in winter and spring, and occasional warm winds, 
 
CLIMATE 373 
 
 prevalent in summer and autumn. On this subject Mr. W. 
 Marriott, Secretary of the Koyal Meteorological Society, made 
 a valuable communication at one of the Conferences on " Meteor- 
 ology in Eelation to Health/' held at the International Health 
 Exhibition in London in July, 1884. 
 The cold winds are : 
 
 1. The East Wind of the British spring, which is dry, cold, and 
 keenly penetrating. The late Sir Arthur Mitchell, in a " Note on 
 the Weather of 1867 and on some Effects of East Wind/' 1 shows 
 how much a cold dry wind must chill the surface of the body 
 by conduction, and also by evaporation. He adds : " The 
 quantity of heat which our bodies lose in this way is far from 
 insignificant, and the loss cannot be sustained without involving 
 extensive and important physiological actions, and without 
 influencing the state of health. In feeble and delicate con- 
 stitutions the resources of nature prove insufficient to meet the 
 demand made on them, and a condition of disease then ensues." 
 
 2. The Mistral is a violent north-west wind, dry, cold, and 
 parching, which sweeps the shores of the Gulf of Lions, drying 
 up and withering vegetation, and predisposing to pleurisy and 
 pneumonia in the inhabitants of Provence. Writing of it, Dr. 
 Scott quotes the old couplet : 
 
 " Le Parlement, le Mistral, et la Durance 
 Sont les trois fleaux de la Provence." 
 
 3. The Tramontana is a searching northerly blast, which is 
 felt along the eastern shores of the Adriatic. A similar furious 
 northerly wind is known in Trieste and Dalmatia as the Bora. 
 
 4. The Nortes (Northers) of the Gulf of Mexico have a per- 
 nicious influence upon health and vegetation. Mr. R. Russell, 
 in his North America : its Agriculture and Climate, states that in 
 Southern Texas, in January, 1855, with a Norther, temperature 
 fell from 81 to 18 F. in forty-one hours. 
 
 5. The Pampero is a dry, cold, south-west wind, which prevails 
 on the coast of Brazil, blowing with great force across the pampas, 
 or plains, of the River Plate. In the Argentine Republic similar 
 winds are called Tormentos. 
 
 Dr. C. R. Harper, now of Peckham Rye, London, S.E., in a 
 1 Journal of the Scottish Meteorological Society, vol. ii., p. 80. 
 
374 METEOROLOGY 
 
 letter to me, dated August 8, 1907, describes a pampero which he 
 experienced off the mouth of the Rio de la Plata, South America. 
 He writes : " The evening was overcast and extremely oppressive, 
 considering we were at the mouth of the River Plate. About 
 6 p.m. I noticed grit in my mouth and hair, and the ship's deck 
 was covered with a fine layer of sand. When the storm broke 
 about 8 p.m., it was accompanied by intensely vivid lightning, 
 thunder, and torrents of rain. After about one and a half hours 
 it gradually ceased. We dragged our anchor about half a mile, 
 and at times were broadside on, owing to the violence of the 
 wind. An Italian barque was capsised near us, because she had 
 not made herself snug. We could not have rendered any aid, 
 owing to the great noise caused by the storm. We sustained no 
 damage, our fellows, no doubt, having been warned by the 
 barometer." 
 
 6. The Etesian Winds of South-Eastern Europe blow across 
 the Mediterranean towards North Africa, apparently to supply 
 the place of the heated air which rises from the Sahara and other 
 African deserts. 
 
 The chief hot winds are : 
 
 1. The Scirocco, a hot south-east wind blowing from the 
 immense deserts of Northern Africa. It is a dry wind on the 
 African coast, but blows in Italy and Sicily as a hot, moist wind, 
 from the oppressiveness of which there is no escape. Mr. Marriott 
 states the case well when he says : " Though not fatal to human 
 life, it is deadly to human temper." In Sicily, during its con- 
 tinuance, the thermometer sometimes rises to 110 F. in the 
 shade. 
 
 2. The Solano is the scirocco of Spain. It is a very hot, dry, 
 and dusty south-east wind, most deleterious to health and to 
 temper ; h^nce the Spanish proverb : " Ask no favour during the 
 Solano." So also is the Leveche or hot south-west wind of the 
 Iberian Peninsula. 
 
 3. The Harmattan of the west coast of Africa is a hot easterly 
 wind, laden with dust and sand from the Sahara. It prevails 
 in December, January, and February. 
 
 4. The Khamseen, or Khamsin, is the hot wind from the 
 desert in Egypt. It is so called, not because it lasts for fifty 
 
CLIMATE 375 
 
 days, but because it is liable to occur during the fifty days follow- 
 ing Easter. It blows from S. or S.S.E., the more easterly variety 
 being the most disagreeable. It usually blows for three days, 
 but may last for seven days at a time. The number of Khamseen 
 days in any one year would seem to vary from four to twenty. 
 During its prevalence the air becomes extremely dry, and is 
 filled with fine sand in a highly electrified condition (F. M. Sand- 
 with 1 ). 
 
 5. The dreaded and deadly Simoom of the deserts of Arabia, 
 Kutchee, and Upper Scinde, is really a circular storm, or tornado 
 in fact, a whirlwind which lasts only ten minutes or there- 
 abouts. 
 
 6. The Hot Wind of Australia, locally known as a " Brick- 
 fielder," blows from the north. It is most severe in the months 
 of November, December, and January. In Sydney it may send 
 the thermometer up to 100 F. once it rose to 106'9 F. but 
 in Central Australia the heat is even more intense, Captain 
 Sturt having reported a shade temperature of 131 F. on 
 January 21, 1845. Dr. Hann, in his Handbuch der Klimatologie, 
 p. 639, quotes an observation of Dr. Neumann, formerly Director 
 of the Melbourne Flagstaff Observatory, respecting the hot wind 
 of January 21-22, 1860, that " the apples were literally roasted 
 on the trees, where the north wind had set in." This north 
 wind is displaced by a sudden south wind which is called a 
 " burster," and its effect is to reduce temperature with mar- 
 vellous rapidity. 
 
 7. The Fohn, or warm, dry wind of the valleys in the north- 
 east of Switzerland, has already been described. So also has been 
 described the Chinook wind of the Rocky Mountains, in Western 
 Canada (Chapter XVIII, p. 261, and Chapter XXII., p. 353). 
 
 8. The Leste is a very dry and parching wind, sometimes 
 very hot, which blows over Madeira from E.N.E. or E.S.E., 
 taking its origin in the Sahara. Its dryness is remarkable, for 
 it traverses 300 miles of sea before it reaches the island. 
 
 There is no doubt that, among the elements which make up 
 climate, temperature holds the foremost place. Accordingly, 
 
 1 Egypt as a Health Resort, p. 32. London : Kegan Paul, Trench and Co. 
 .1889. 
 
376 
 
 METEOROLOGY 
 
 CHART C. 
 
 100 
 
 r~ 
 
 l| 130 - 
 
 J: 
 
 3 MURZOUK. 
 
 
 
 120 
 
 1 COOPER'S CREEK. 
 
 
 E 
 
 i 
 
 v| .2 no 
 
 fj BAGDAD. 
 
 i CAIRO. 
 
 
 
 i 
 
 :| SYDNEY. 
 
 
 
 ii Madras. ' 
 i| Calcutta. 90 
 
 Ill GREENWICH. 
 
 
 
 .| Havana. ^ 
 
 
 70 
 
 
 is 
 
 i Cairo. 70 - 
 
 ;| FALKLAND ISLANDS. 
 1 Barbadocs. 
 
 
 
 ! Algiers. ^ 
 60 
 
 | Singapore. 
 
 60 
 
 
 j Jerusalem !u 
 
 i TJrtTV, 50 - 
 
 I! Lima. 
 4 Bombay. 
 
 
 E 
 
 if Constantinople. ^ 
 
 Senegal. 
 
 
 E 
 
 Pekin. 
 
 Adelaide. 
 
 
 E 
 
 Berlin. g 
 
 ;.;: Leipzig. "^^ 
 
 1 Jerusalem. 
 
 
 - 
 
 3 Toronto. <" o> 
 
 H -s 
 
 Ji Sf Ppf Ai-cViiiiry *0^ 10 
 
 V\ Constantinople. 
 
 
 ~ 
 
 ! Hammerfest. 
 
 ^i ; K^ - 
 
 -^ Leh(Ladakh). 
 
 
 = 
 
 tJ 
 
 4 Freezing- Point. _ 1QO 
 
 11 Greenwich 
 
 
 ~ 
 
 fH 
 
 |l Godhaab _ 2 QO 
 
 
 
 E 
 
 | Nertschinsk. 
 
 fi -30 _ 
 
 111 Blackadder (Ber- 
 wick). 
 Toronto. 
 
 20 
 
 ~ 
 
 1 Fort Churchill ^2 
 (Hudson Bay). 40 o 
 
 | Chicago. 
 || Winnipeg. 
 
 
 i 
 
 || Fort Enterprise. 
 
 ; i Yakutsk. '^ -50(_ 
 
 I Montreal. 
 
 f Moscow, 
 
 
 ~ 
 
 1 Winter Island. ^ 
 j Fort Hope. JJ _ 6QO _ 
 
 
 
 
 j 
 
 t Boothia Felix H 
 j Melville Island. | _ 70 o_ 
 
 Sk - 80) t 
 
 Barnaul. 
 1 Fort Reliance. 
 : ^'cmipalatinsk. 
 \\ r erchojansk. 
 
CLIMATE 377 
 
 the preceding Chart (C) will both interest and instruct the 
 reader. It was prepared by Mr. William Marriott, F.R.Met.Soc., 
 some years ago by direction of the Council of the Royal Meteoro- 
 logical Society. The mean annual temperature in the shade 
 in degrees Fahrenheit of certain places in various parts of the 
 world is shown on the thermometer scale to the left-hand side, 
 while the highest and lowest shade temperatures, observed at 
 the specified places in all parts of the world, are drawn on the 
 right-hand scale. When the highest temperatures are indicated 
 the names of the places are printed in SMALL CAPITALS ; when the 
 lowest temperatures, in italics. 
 
CHAPTER XXV 
 THE CLIMATE OF THE BRITISH ISLANDS 
 
 WITHIN the limits of a small book it would be impossible to 
 do full justice to the great theme of Climatology. It must suffice 
 to repeat that, because water by its presence not less than by 
 its motion so profoundly modifies climatic conditions, climates 
 are, by universal consent, divided into insular or moderate, and 
 continental or excessive. Of the former, the climate of the British 
 Islands affords a typical example ; of the latter, the climate of 
 Siberia may be taken as a type. But an insular climate is by 
 no means confined to islands (Latin, insulce). The western 
 shores of all continents enjoy moderate climates which are fully 
 entitled to be described as " insular." On the other hand, the 
 interior of continents and their eastern shores are exposed to 
 extremes of heat and cold, which equally justify the appellation 
 " continental," applied to their climate. 
 
 The great changes which take place in the distribution of 
 atmospheric pressure over the immense continent of Europe 
 and -Asia and the adjoining oceans the North Atlantic and 
 the North Pacific have been already described and explained 
 (Chapter XIII., pp. 156-158). It was there shown that, in summer, 
 a vast cyclonic system develops over Europe and Asia the wind 
 blowing against the hands of a watch in accordance with Buys 
 Ballot's Law, round an area of low barometer formed over the 
 heated inland regions, from S.W. in India and China (the south- 
 west monsoon) ; from S., S.E., and E. in Japan, and North- 
 Eastern Siberia ; from N.E. and N. in North-Western Siberia ; 
 from N.W. and W. over most of Southern Europe and South- 
 western Asia. 
 
 In winter, on the contrary, over the ice-bound, snow-covered, 
 
 378 
 
THE CLIMATE OF THE BRITISH ISLANDS 379 
 
 boundless Eurasian plain, an immense anticyclone is formed, 
 the winds circulating round and out from the centre of high 
 pressure in a direction with the hands of a watch blowing from 
 N.W. and N. in Japan and China ; from N.E. in India (the north- 
 east monsoon) ; from E. and S.E. in Russia and Southern Europe ; 
 from S.W. in the British Isles ; and from W. in Northern Russia 
 and Siberia. 
 
 These considerations facilitate an explanation of the climate 
 of the British Isles (1) in summer and (2) in winter. 
 
 It will easily be seen how the summer continental depression 
 influences the climate of the British Isles. Air is drawn from 
 W. and N.W. over these countries, and as this air blows over 
 the surface of a wide ocean and from high latitudes, it is cool 
 and moist. Do not these two words describe our summer ? 
 These ocean winds prevail chiefly on the W. and N.W. shores of 
 Ireland and Scotland, which have thus the rainiest and the 
 coolest summer, while this season is warmer and drier as we go 
 eastward and southward, to the south-eastern counties of Eng- 
 land. This is well illustrated in Dr. Buchan's Chart 1 of the 
 Isothermals 2 of the British Isles in July. 
 
 It is not necessary to consider at length the influence of the 
 winter system of barometrical pressure on our climate. During 
 the earlier winter months a great stream of warm, very moist 
 air, as a rule, flows north-eastward and northward over these 
 islands round the Atlantic depression, the centre of which lies 
 near Iceland. But this stream does not flow evenly. Along its 
 eastern edge it is in continual conflict with the cold anticydonic 
 air, which is travelling westward from Russia and Siberia, and 
 immense volumes of the latter are frequently rushing in to supply 
 the place of those volumes of the warm air which, owing to their 
 low density, have presumably risen from the earth's surface 
 towards the higher strata of the atmosphere. The close juxta- 
 position of two such opposite currents of air causes our storms, 
 and those violent and rapid alternations of temperature which 
 are so prejudicial to health in the winter months. 
 
 1 " The Mean Temperature of the British Islands." By Alexander 
 Buchan. Journal of the Scottish Meteorological Society, vol. vi. New Series. 
 No. 64, p. 22. 1882. 
 
 2 Greek, fcros = equal, and Btppri = warmth. 
 
380 METEOROLOGY 
 
 The reason for the occurrence of these alternations of tem- 
 perature will be explained when we remember that most of these 
 gales, or bourrasques, as they have been termed, are cyclonic in 
 character, and that they generally cross the British Isles from 
 S.W. to N.E., less frequently from W. to E., and still less fre- 
 quently from N.W. to S.E. The southerly winds which blow 
 over the country in front of the centre of the storms are warm 
 and moist, while the northerly winds, which prevail over those 
 districts already reached and passed by the centre, are cold, 
 and after a time dry. No better examples of this can be given 
 than the remarkable gales of December 8 and 9, 1872, and of 
 February 2, 1873. In front of the former, temperature rose 
 generally to about 50 F. over the south of Ireland, most part 
 of England, and the whole of France ; while it fell almost to 
 the freezing-point over those districts a few hours later when 
 the centre had passed. The second gale referred to was accom- 
 panied by a range of 18 F. over the whole of France. Dr. 
 Scott says 1 that a great contrast of temperature between adjacent 
 stations or, so to speak, a great " thermometric gradient "- 
 being an indication of serious atmospheric disturbance, is the 
 precursor or concomitant of a serious storm. He quotes, as an 
 example, the gale of November 14, 1875, which followed hard 
 upon a difference of 36 F. in temperature at 8 a.m. of the previous 
 day between Scilly (57 F.) and Wick (21 F.). 
 
 The effect of the warm Atlantic air-current on the Isothermals 
 of the British Isles is well represented in Dr. Buchan's Chart 
 for January. 2 
 
 Anticyclonic wind-systems sometimes prevail over Western 
 Europe, but much less frequently than cyclonic systems. They 
 cause dry, often cold, weather, and are much more persistent 
 than cyclones. 
 
 Anticyclones are better marked, as a rule, in winter than in 
 summer, and historical " hard frosts " in the British Isles are 
 almost invariably connected with one of these systems. The 
 
 1 Weather Charts and Storm Warnings, p. 134. London : Henry S. King. 
 1876. 
 
 2 " The Mean Temperature of the British Islands." By Alexander 
 Buchan. Journal of the Scottish Meteorological Society, vol. vi. New Series. 
 No. 64, p. 22. 1882. 
 
THE CLIMATE OF THE BEITISH ISLANDS 381 
 
 great frost of 1890-91, which lasted in the south-east of England, 
 almost without interruption, from November 25, 1890, to 
 January 22, 1891, was connected with the presence of a large 
 area of high barometric pressure which maintained a nearly 
 permanent position over Central Europe. The incoming dis- 
 turbances from the Atlantic could not effect a passage into 
 Europe, but were fended off by the European anticyclone, their 
 centres being kept well out in the Atlantic. Ireland and Scot- 
 land came from time to time under the warming influence of 
 these Atlantic depressions or cyclones, and consequently the 
 frost was neither severe nor continuous in those countries ; 
 but England was not affected by them, and so the cold held in 
 its intensity, particularly in the eastern, south-eastern, and 
 midland parts of the country. Mr. Charles Harding, F.K.Met.Soc. 
 in a paper 1 on this historical frost, states that the very dry 
 character of the weather over England during the frost was also 
 attributable to the fact that the European anticyclone embraced 
 the southern portion of the kingdom, and although on two or 
 three occasions there were some rather heavy falls of snow, the 
 aggregate fall of snow and rain was but trifling in comparison 
 with the average. 
 
 In the chapter on " Weather " in his Elementary Meteorology 
 (p. 360), Dr. K. H. Scott well observes : 
 
 " The w T eather we experience in Western Europe is distinctly 
 related to these areas of depression and anticyclones, to the rate 
 at which they respectively travel over the earth's surface, and 
 to the distance which intervenes between their respective centres. 
 As in a system of either kind we may meet with winds from any 
 point of the compass, which will have different qualities as to 
 temperature, humidity, etc., according as they belong to one or 
 the other, we see the great importance of the consideration, 
 first pointed out by W. Koppen, 2 and subsequently by Captain 
 Toynbee, 3 that the climatic character of a wind depends on its 
 origin i.e., on its belonging to a depression or to an anticyclone." 
 He adds : " Anticyclones are generally more or less stationary, 
 
 1 Read before the Royal Meteorological Society on February 18, 1891. 
 
 2 Repertorium fiir Meteorologie, vol. iv. 1875. 
 
 3 The Meteorology of the North Atlantic during August, 1873, p. 97. 
 London, 1878. 
 
382 METEOROLOGY 
 
 but depressions move over the earth's surface, usually from west 
 to east in these latitudes, their paths as they advance, though 
 chiefly ruled by the distribution of pressure, being liable to 
 modification by the irregularities of the surface over which they 
 pass ; and their effects, as to the amount of cloud and rain to 
 which they give rise, being influenced by the same causes. A 
 south-west wind, for instance, may blow over a flat country 
 with a clear sky, but as soon as the air reaches a hill-side and is 
 forced to ascend, the moisture it contains is condensed, clouds are 
 formed, and rain is frequently the result." 
 
 An important contribution to the climatology of England 
 and Ireland is a paper by Mr. Francis Campbell Bayard, 
 F.K.Met.Soc., which was read before the Royal Meteorological 
 Society on June 15, 1892. * The author carefully analysed the 
 observations taken during the ten years, 1881-1890, at nineteen 
 Second Order Stations (sixteen in England and three in Ireland), 
 and at thirty-three Climatological Stations (thirty-two in England 
 and one in the Channel Islands), and he arrives at the following 
 general conclusions : 
 
 1. With respect to mean temperature, the sea-coast stations 
 are warm in winter and cool in summer, whilst the inland stations 
 are cold in winter and hot in summer. 
 
 2. The mean maximum temperature occurs at all stations in 
 July or August, while the mean minimum temperature takes place 
 mostly in December or January, except at Llandudno and the 
 south-western sea-coast stations, where it is later, taking place 
 in February or March. 
 
 3. Relative humidity is lowest at the sea-coast stations and 
 highest at the inland ones. 
 
 4. The south-western district seems most cloudy in winter, 
 spring, and autumn, and the southern district the least cloudy 
 in the summer months ; and the sea-coast stations are, as a rule, 
 less cloudy than the inland ones. 
 
 5. Rainfall is smallest in April, and, as a rule, greatest in 
 November, and it increases as we travel from east to west. 
 
 Mr. Bayard's paper, it is true, does not include Scotland in 
 
 1 See the Quarterly Journal of the Society, New Series, vol. xviii.,? No. 84, 
 p. 213. 
 
THE CLIMATE OF THE BEITISH ISLANDS 383 
 
 its scope, and three stations in Ireland Londonderry, Dublin, 
 and Killarney are far too few to serve as a basis for climatological 
 conclusions. Neverthe^ss the foregoing sentences epitomise 
 the facts relating to the climate of the British Islands at large. 
 This is shown by referring to a series of elaborate communications 
 on the subject, which Dr. Alexander Buchan from time to time 
 since 1862 laid before the Scottish Meteorological Society. 
 
 I. SEA TEMPERATURES. 
 
 In an article on the " Temperature of the British Islands," * 
 Dr. Buchan observed that a very cursory examination of the 
 British isothermals is enough to show the powerful influence of 
 the sea in modifying their course in the different months of the 
 year. Hence, the temperature of the sea which washes our 
 shores is a question of the first importance in investigating the 
 climate of these islands. Observations on sea temperature 
 have been made at different points round the Scottish coasts 
 since 1855, when the Scottish Meteorological Society was founded, 
 and more recently in Faeroe and Iceland. During the three 
 years, July, 1879, to June, 1882, observations, from which maps 
 of the sea temperature all round the British Isles have been con- 
 structed in the Meteorological Office, London, were taken at 
 certain coastguard stations, lighthouses, and lightships to the 
 number of forty-nine. The results have been embodied in a 
 Meteorological Atlas of the British Isles, published by the 
 authority of the Meteorological Council in 1883. 
 
 In the year ending March 31, 1909, there were in communication 
 with the Meteorological Office fifty -nine sea-temperature stations 
 scattered round the coasts of the British Isles. The Sea-Fore- 
 cast Districts were twelve in number namely, Shetland and the 
 Naze, Great Fishery and Dogger Banks, North Sea north of the 
 Wash, North Sea south of the Wash, Straits of Dover, English 
 Channel east of the Isle of Wight, English Channel west of the 
 Isle of Wight, Bristol Channel, St. George's Channel, Irish Sea, 
 North Channel, the Minch. It will be observed that the whole 
 western seaboard of Ireland is left out of this scheme for the 
 
 1 Journal of the Scottish Meteorological Society, New Series, vol. iii., 1873. 
 p. 102. 
 
384 METEOROLOGY 
 
 protection of shipping by means of Weather Forecasts and Storm 
 Warnings. Of course, such forecasts and warnings may be sent 
 to Malin Head, Co. Donegal, Blacksod Point, Co. Mayo, and 
 Valentia, Co. Kerry, at each of which places there are land 
 stations in telegraphic communication with the central Meteoro- 
 logical Office, London. 
 
 The mean temperatures of the sea surface vary as follows : 
 
 January . . Highest, 49 Cleggan, Co. Galway ; Scilly, Truro, Penzance. 
 
 Lowest, 37 Yarmouth, Berwick. 
 February . . Highest, 49 Scilly, Seven-Stones L.V., 1 Cornwall. 
 
 Lowest, 37 Burntisland, Fifeshire. 
 March . . Highest, 51 Cleggan, Co. Galway. 
 
 Lowest, 40 Dunrobin, Holkham, Leman and Ower L.V. 
 April . . Highest, 51 Cleggan, Valentia Island, Scilly. 
 
 Lowest, 42 Berwick, Leman and Ower L.V., Norfolk. 
 May . . Highest, 56 Valentia Island. 
 
 Lowest, 45 Berwick. 
 
 June . . Highest, 58 West Coast of Ireland, Bristol Channel, Pad- 
 
 stow, Cornwall ; Yarmouth. 
 
 Lowest, 49 North Unst, Shetland ; Berwick. 
 July . . Highest, 62 Bristol Channel. 
 
 Lowest, 52 North Unst, Wick, Berwick. 
 August . . Highest, 64 The Owers L.V., off the Sussex coast. 
 
 Lowest, 52 North Unst, East Yell, Shetland. 
 September . . Highest, 62 Cleggan, Co. Galway ; Dover. 
 
 Lowest, 52 Shetland. 
 October . . Highest, 59 Mouth of the Thames. 
 
 Lowest, 47 Fraserburgh. 
 November . . Highest, 54 Truro. 
 
 Lowest, 44 Fraserburgh, Berwick. 
 December . . Highest, 52 St. Agnes' Head, Cornwall. 
 
 Lowest, 39 Berwick. 
 
 WINTER. According to Dr. Buchan, the high temperature of 
 the northern islands in winter is one of the best illustrations 
 which could be adduced of the powerful influence of the ocean 
 on climate. The conserving influence of the sea on the tem- 
 perature is also seen, though in a less degree, in the openings of 
 the Irish Sea and the English Channel. The isothermals indica- 
 tive of the mildest British climate in winter are seen enveloping 
 Ireland in January. The west and north coasts of Wales share 
 in the genial influence of this offshoot from the warm waters 
 of the Atlantic. The mildest winter climate of Great Britain, 
 however, is found in the peninsula of Devon and Cornwall, which 
 is not only farther south, but is also more completely enveloped 
 1 L.V. = Light- vessel. 
 
THE CLIMATE OF THE BRITISH ISLANDS 385 
 
 by the ocean than any other part of the British Islands. A 
 rapid lowering of temperature takes place from the Land's End 
 eastwards to Kent, because the English Channel is comparatively 
 shallow, is near the colder continent, and is connected with the 
 colder North Sea. 
 
 SUMMER. Owing to the great preponderance of sea over 
 land in the vicinity of the Hebrides, the Orkneys, and the Shet- 
 lands, the temperature of these islands is remarkably reduced in 
 summer. A tendency to northing in the summer winds is also 
 mentioned by Dr. Buchan as another cause for the marked diminu- 
 tion of summer heat in the northern parts of Great Britain. The 
 Irish Sea and the English Channel moderate the heat of summer 
 along their coasts. Conversely, warmth is relatively greatest 
 in those parts of the British Islands which are most removed from 
 the direct and indirect influence of the sea. Hence there is a 
 curving northwards of the isothermals of June, July, and August 
 through the central parts of Great Britain, from the Thames 
 Valley northwards to the Moray Firth. The patch of highest 
 mean temperature corresponds closely with the Thames Valley, 
 and is most marked in the immediate vicinity of London. 
 
 In an account of the climate of Dublin, written in 1907, and 
 printed in the Handbook to the Dublin District, prepared for the 
 meeting of the British Association for the Advancement of Science 
 which took place in September, 1908, 1 showed how beneficial was 
 the influence of the Irish Sea upon that city not only in winter and 
 spring, when it softened and warmed by some 5 F. the keen, dry, 
 searching easterly winds of those seasons but also in summer. In 
 calm, clear weather in summer time, no sooner has the sun mounted 
 high in the heavens than a cool, refreshing sea breeze a typical 
 " inbat/' 1 as the modern Greeks call it sets in towards the land, 
 so that extreme or oppressive heat is rarely experienced. Indeed, 
 an oppressive atmosphere happens only when a damp, warm, 
 south-west wind is blowing, with a more or less clouded sky. 
 On July 15, 1876, the thermometer no doubt did rise in the Irish 
 capital to 87*2 F., but this was altogether a phenomenal occur- 
 rence. Temperatures above 80 F. in the screen in Dublin nearly 
 always coincide with winds off the land, from some point between 
 1 Evidently a derivative from e^aivw. 
 
 25 
 
386 METEOROLOGY 
 
 south and west, and a clear or only slightly clouded sky. On 
 July 14, 1905, the maximum in Dublin was 81'8 F. (practically 
 82), with brilliant and hot sunshine, a freshening south-west 
 wind, and the clouds also coming from the south-west. Since 
 January, 1868, the extreme readings of the thermometer in 
 Stevenson's stand recorded at Fitzwilliam Square, Dublin, have 
 been 87'2 F. on July 15, 1876, and 13'3 F. on December 14, 
 1882 a range of 73*9 F. The average annual range of mean 
 temperature in the forty years 1866-1905 was not quite 19 F. 
 namely, January, 41'6 F. ; July, 60*4 F. that is, 18-8 F. 
 
 Speaking on his paper on the " Mean Temperature of the British 
 Islands," at a meeting of the Scottish Meteorological Society 
 on July 20, 1881, Dr. Buchan laid special stress on the influence 
 of the Atlantic and other surrounding seas upon the temperature. 
 He pointed out the great influence of the Irish Sea in affecting 
 the course of the isothermals, and also that of the Atlantic, 
 particularly off the north-west of Scotland. In winter the 
 temperature of St. Kilda is as high as that of Penzance, and the 
 temperature at Cape Wrath as high as that of the Isle of Wight. 
 Taking the British Islands as a whole, the mean annual tempera- 
 ture in the west, 52'0 F., was represented about the same latitude 
 in the east by a mean annual temperature of 51*0 F. in other 
 words, the west was one degree warmer than the east. 
 
 II. AIR TEMPERATURES. 
 
 i Observations upon this element fully justify Dr. Buchan's 
 dogmatic statement that " the climate of the British Islands 
 is eminently insular that is, it is not subject to great extremes 
 of heat and cold, but is remarkably equable throughout the year 
 being much milder in winter and cooler in summer than in 
 continental regions in the same latitudes/' 
 
 We now possess two series of Temperature Charts, which may 
 be accepted as conclusive evidence of the distribution, annually 
 and monthly, of air temperature throughout the British Isles. 
 The first series was drawn up by Dr. Buchan for the Scottish 
 Meteorological Society, originally in 1871. It was based upon 
 observations extending over thirteen years, beginning with 
 January, 1857, and terminating with December, 1869. These 
 
THE CLIMATE OF THE BRITISH ISLANDS 387 
 
 observations on the daily maxima and minima were taken at 
 seventy-six stations in Scotland, sixty-seven in England, twelve in 
 Ireland, and fifteen in adjoining countries on the Continent. The 
 mean temperatures used by Dr. Buchan are the arithmetical means 
 of the daily maximal and minimal thermometer readings. In 
 order to reduce the means so obtained to their sea-level value, they 
 were increased by an addition at the rate of one degree for every 
 300 feet of elevation above the sea, and were set down, so cor- 
 rected, in their proper position on thirteen charts, from which the 
 isothermals for each of the twelve months and for the year were 
 drawn. Dr. Buchan's paper will be found in vol. iii. (New Series) 
 of the Journal of the Scottish Meteorological Society, 1873, p. 102. 
 
 Of this paper a still more elaborate communication on the 
 " Mean Temperature of the British Islands," based on twenty- 
 four years' observations ending with 1880, and laid before the 
 Scottish Meteorological Society by Dr. Buchan in 1882, 1 may be 
 regarded as a revision. The observations embrace a period 
 nearly double the length of that of the earlier paper, and were 
 taken at 24 places in Ireland, 132 in Scotland, and 138 in Eng- 
 land. Thus, the series of Plates illustrating this second paper, 
 and giving the isothermal lines for each month, may be accepted 
 as faithfully representing the temperature of the British Islands 
 in exceedingly close agreement with the true mean annual 
 temperature of this portion of the globe " a datum/' says 
 Dr. Buchan, "of no small importance in many inquiries which 
 deal with the physics of the atmosphere and underground 
 temperature." 
 
 Plate VI. in the Atlas of Meteorology, which constitutes the 
 third volume of Bartholomew's Physical Atlas, published in 1899, 
 contains a revision of Dr. Buchan's Isothermal Charts of the 
 British Isles, together with two additional Charts specially pre- 
 pared for the Atlas. Of these, one shows the annual range of 
 temperature over these islands. It was prepared by Dr. A. J. 
 Herbertson from Dr. Buchan's figures, and demonstrates very 
 clearly the increasing variability of the seasonal temperature 
 from west to east, and more particularly from north-west to 
 
 1 Journal of the Scottish Meteorological Society. New Series. Vol. vi., 
 p. 22. 1882. 
 
 252 
 
388 METEOROLOGY 
 
 south-east. The second additional Chart, prepared by Dr. A. J. 
 Herbertson, shows the actual annual temperature over the 
 British Isles, which may be obtained by calculation from the 
 ordinary temperature map in which the isotherms are reduced 
 to sea-level, provided the height of the place and the vertical 
 temperature gradient are known. In this chart the mean actual 
 annual temperature is shown not reduced to sea-level. The 
 isotherms in a great measure follow definite contour lines, but 
 they slope downhill from south to north, and also, to a less 
 extent, from west to east. All the fifteen maps are based on the 
 mean temperatures of 400 stations reduced to the forty-year 
 period, 1856-1895. The lowest mean annual temperature in 
 the British Isles is 31'2 F. (-0-4 C.), on the summit of Ben 
 Nevis. The highest mean annual temperature is 52*2 F. (11'2 C.), 
 registered in the Scilly Isles and at Truro. The capitals have 
 the following actual mean temperatures : London, 50*4 F. 
 (10-2 C.) ; Edinburgh, 276 feet (84 metres), 47'1 F. (8'4 C.) ; 
 Dublin, 49-5 F. (9-7 C.). 
 
 The second series of Temperature Charts was drawn up by 
 the authority of the Meteorological Council, and published, in 
 1883, in the Meteorological Atlas of the British Isles. It is based 
 upon observations of the maximum and minimum thermometers 
 made daily during the twenty years 1861 - 1880 inclusive at 
 seventy-five stations thirty-one in Scotland and the adjacent 
 islands, thirty-five in England (including the Channel Islands), 
 and nine in Ireland. The mean temperatures given in the 
 Atlas, which consists of twelve monthly maps and one yearly 
 map, are the arithmetical means of the daily maxima and minima. 
 They have been reduced to their sea-level value by the addition 
 of a correction at the rate of one degree Fahrenheit for each 
 300 feet of vertical elevation. The isotherms in these maps are 
 drawn for each degree, the value for each being inserted in large 
 figures at one of the extremities of each line. 
 
 In 1902 the Meteorological Office published temperature tables 
 for the British Islands, together with a supplement giving 
 " Difference Tables for each Five Years for the Extrapolation 
 of Mean Values." 1 
 
 1 Official Publication, No. 154. 
 
THE CLIMATE OF THE BKITISH ISLANDS 389 
 
 An analysis of the maps in the Meteorological Atlas of the 
 British Isles shows that the isotherm of 46, representing the 
 mean temperature of the whole year, skirts the north coasts 
 of the Hebrides and Scotland. The mean temperature then 
 increases as we travel southwards, with many interesting local 
 irregularities, until the isotherm of 52 is found off the extreme 
 south-west of Ireland, whence it passes east-south-eastward by 
 the Land's End in Cornwall to Jersey. The mean annual tem- 
 perature in the Scilly Islands is 53*1 (nearly a degree higher 
 than Dr. Herbertson's estimate). Scotland lies between the 
 isotherms of 46 (45'8) and 48 (48'3) ; England between 
 those of 47 and 52 ; Ireland between those of 48 (48'4) 
 and 51 (51-3). 
 
 Taking the first month of each quarter of the year, we obtain 
 the following results : 
 
 January. The area of greatest cold is represented in Scotland 
 by the isotherm of 36, which embraces Aberdeenshire. Tem- 
 perature increases from that low value to 40 all along the 
 extreme western coast of Argyllshire and the Hebrides to the 
 Shetlands. In England the area of greatest cold is represented 
 by the isotherm of 37, which covers parts of Norfolk, Lincoln- 
 shire, Huntingdonshire, and Cambridgeshire near the Wash. 
 The isotherm of 42 passes southwards down the west coast 
 of Wales, across the borders of Devon and Cornwall ; that of 
 45 just touches the Land's End, while that of 46 (46-3) 
 passes through the Scilly Islands. In Ireland the isotherm of 
 40 embraces an oval-shaped area on what may be called the 
 " lee-side " of the island, extending from the western, or inland, 
 half of Antrim southwards to the counties Kilkenny and Carlow. 
 The isotherm of 41 passes through Dublin south-westwards 
 to Fermoy, and then in a curve towards north-west and north 
 to the extreme north of the island near Lough Swilly. On the 
 other hand, the isotherm of 45 sweeps southwards down the 
 extreme western coast from Achill Island to Valentia. 
 
 April. This is a transitional month the characteristic winter 
 isotherms are now giving place to those equally characteristic 
 of summer. In Scotland the isotherms run from north-west to 
 south-east, with local interruptions 44 crosses Caithness ; 47, 
 
390 METEOROLOGY 
 
 Wigtonshire and Dumfries. In England the isotherms run 
 in the same direction 46 skirting the north-east coast, and 
 50 showing itself near London north of the Thames, and also 
 over Devonshire and Cornwall. In Ireland temperature is very 
 uniform, ranging from 47 in the extreme north to 49 in Kerry 
 and Cork. 
 
 July. The summer distribution of temperature is now seen 
 to full advantage, inland districts being warmest, and coast 
 districts coolest. In Scotland, the isotherm of 55 sweeps in 
 a convex curve round the north-west and north coasts from the 
 Hebrides to the Orkneys. The almost circular isotherm of 59 
 covers the centre of Scotland, including Perthshire, Lanarkshire, 
 and the Lothians. The English coasts vary from 59 in 
 Northumberland to 62 along the shores of the English Channel 
 and Suffolk. Inland, 63 covers the Midlands, and 64 is 
 found surrounding and especially to the north of London. In 
 Ireland, 58 skirts the north, and 59 the west coast, while 
 61 embraces a large area extending from the southern shores 
 of Lough Neagh to Cork. Tipperary and North Cork, with 
 Kilkenny and Carlow, enjoy a mean temperature of 62. The 
 great central plain extending from Galway to Dublin is somewhat 
 cooler 60*4 to 60*8. This is doubtless due to the immense 
 quantity of water with which the Bog of Allen and other less 
 extensive peat-bogs are charged, as well as to the number of 
 lakes in the centre and west of the country. 
 
 October. In this month the winter distribution of tempera- 
 ture begins to appear in the drawing of the isothermal lines. 
 Scotland varies from below 47 in the north, north-east, and 
 south-east, to 49 in the south - west ; England from 48'4 
 in Durham to 53*6 at Penzance, and 55*1 in the Scilly 
 Islands ; Ireland from 49 in a large oval in the north and 
 centre to 52 in the south-east, and 53 off the promontories 
 of Kerry and South-West Cork. 
 
 With respect to the reduction of mean temperature to sea- 
 level in these charts, I may remind the reader that, while it is 
 expedient from a scientific point of view that such a reduction 
 should be made, it is* quite unnecessary nay, even misleading 
 from either an agricultural or a medical standpoint. We want 
 
THE CLIMATE OF THE BKITISH ISLANDS 391 
 
 to know what are the actual climatic conditions under which 
 both plants and animals live. Further, in a suggestive address 
 on " The Eelations of the Official Weather Services to Sanitary 
 Science/' delivered before the American Public Health Associa- 
 tion some years ago at a Conference at Mexico, Mr. Mark W. 
 Harrington, the very able Superintendent at that time of the 
 Weather Bureau of the United States Government, wisely and 
 properly pointed out that the meteorological data required by 
 sanitarians and physicians may not be furnished in a form 
 suitable for their purpose. For instance, among the tempera- 
 ture data for health resorts especially physicians particularly 
 want to know the extreme range as well as the mean. Two places 
 may have the same mean temperature say, 45 F. but they 
 may be as far apart as the Poles in their relative availability for 
 invalids. One place may have an occasional range of 40 F. 
 within a few hours, the other may not have an absolute annual 
 range of that amount. 
 
 " For hygienic purposes," says the writer of the Address, 
 " the details of temperature are of interest, as on a sunny day 
 the temperature may differ greatly in short distances, depending 
 on the exposure and the character of the surroundings. The 
 meteorologist has defined the air temperature as that of the 
 free air at about the height of a man, the thermometer being 
 protected from all radiation. With such a definition the tem- 
 perature data which could be prepared from observations now 
 taken, and which might be of use to sanitarians, would appear 
 to be as follows : 
 
 " The mean temperatures of the hours, months, seasons, and 
 year. 
 
 " The mean maxima and minima for the months, seasons, and 
 year. 
 
 " The absolute maxima and minima for the same. 
 
 " The mean and absolute amplitudes for the same. 
 
 " Inter diurnal variability (i.e., the mean change in mean daily 
 temperatures). 
 
 " List of sharp changes of temperature of short duration. 
 
 " Frequency of freezing days (mean temperature less than 
 32 F.) ; of frost days (minimum temperature below 32 F.) ; 
 
392 
 
 METEOROLOGY 
 
 of hot days (maximum temperature, 86 F.) ; and of very hot 
 days (maximum temperature, 95 F.). 
 
 " Mean and absolute dates of the last and first frosts. 
 
 " Mean and absolute duration of freezing, of hot, and of very hot 
 weather. 
 
 " Means of temperature of evaporation (i.e., of the wet-bulb 
 thermometer). 
 
 " This is an element which has not been discussed, so far as 
 known to the writer, though Lieutenant Glassford suggested it 
 to him some time ago. Its significance lies in the fact that it 
 would approximately represent the temperature of the person 
 in hot weather. It would help to distinguish between the dis- 
 tressing moist heat of some stations and the more endurable 
 dry heat of others. It should probably be given in means, and 
 associated with the corresponding air temperature (or tempera- 
 ture by the dry-bulb thermometer). It could be given thus : 
 
 I ! 
 
 Air Temperature. Mean. 
 
 Temperature of 
 Evaporation. 
 
 70 to 80 
 80 to 90 
 90 to 100 
 10CP or more 
 
 75 
 85 
 95 
 
 
 " A similar table for low temperatures might be of use, as it 
 is thought that dry, cold weather is less hard to endure than wet, 
 cold weather/' 
 
 III. ATMOSPHERIC PRESSURE. 
 
 The Meteorological Atlas of the British Isles (1883) includes also 
 thirteen maps, showing the distribution of mean barometrical 
 pressure over the United Kingdom for each month and for the 
 whole year, during the twenty years 1861-1880. For purposes 
 of comparison the readings have in all cases been reduced to their 
 mean sea-level value ; but I agree with the Superintendent of 
 the Weather Bureau of the United States Government in think- 
 ing that, from a hygienic or medical point of view, it is the pressure 
 to which an individual is actually exposed, and not that felt 
 
THE CLIMATE OF THE BKITISH ISLANDS 393 
 
 at sea-level, perhaps 1,000 feet below him, which is required in 
 investigations as to the influence of climate on health. 
 
 This view was evidently shared by the compilers of the Atlas, 
 for each map bears the following " Note " : 
 
 " The approximate mean pressure for the month (or year) 
 may be found by subtracting from the pressure indicated by 
 the nearest isobars a correction obtained as follows : From 
 1-21 inches, subtract for each 10 above zero, F., -025 inch ; 
 the residue will be the correction for an elevation of 1,000 feet, 
 and the correction Tor the actual elevation will be proportional 
 to this." 
 
 The barometrical maps, moreover, are full of interest, par- 
 ticularly as the monthly distribution of pressure gives a clue to 
 the direction and force of the predominant winds. 
 
 In this series of maps, isobars have been drawn for the even 
 hundredths of an inch of pressure, so as to correspond as nearly 
 as practicable with the actual observations recorded. The 
 readings are reduced to sea-level, and also to 32 F. 
 
 In the chart for the whole year, and, indeed, in that for every 
 one of the twelve months, the isobars have a cyclonic or concave 
 trend in the north, but an anticyclonic or convex trend over 
 the south of the United Kingdom. This at once explains the more 
 settled weather of the south contrasted with the less settled 
 weather of the north. Next, we observe that throughout the 
 year pressure is on the average lower in the north than in the 
 south. The differences in pressure are not uniform, however, 
 throughout the year. Thus, in January the isobar of 29-66 inches 
 runs across the extreme north of Scotland from south-west to 
 north-east, that of 29'98 inches crosses Kent in a north-easterly 
 direction. Here we have a difference of pressure amounting to 
 32 of an inch, and gradients for south-westerly winds over the 
 whole kingdom. This difference steadily diminishes from 
 January through February (when it is '22 inch 29'74 and 
 29-98 inches), March (-14 inch 29'76 and 29'90 inches), April 
 (10 inch 29-86 and 29-96 inches), to May, when it is only 
 08 inch 29*91 inches and 29*99 inches. It will at once occur 
 to the reader that this equalisation of pressure means the dying 
 out of the strong south-west winds of winter, and the inter- 
 
394 
 
 METEOKOLOGY 
 
 spersing of a large proportion of easterly winds with the pre- 
 dominant westerly winds of our latitudes. Once May has passed 
 a gradual reverting to the winter type of distribution may be 
 noticed, thus : 
 
 June 
 July 
 August 
 Septembe 
 October 
 November 
 December 
 
 N. 29-88 inc 
 N. 29-84 
 N. 29-82 
 N. 29-76 
 N. 29-72 
 N. 29-74 
 N. 29-70 
 
 hes ; S. 30-01 inch 
 S. 30-00 
 S. 29-98 
 S. 29-96 
 S. 29-92 
 S. 29-94 
 S. 29-98 
 
 es diffe 
 
 rence, -13 in 
 16 
 16 
 20 
 20 
 20 
 28 
 
 3h. 
 
 I have compared these values with those given in Dr. Buchan's 
 charts of the isobars, showing in inches the mean atmospheric 
 pressure of the British Isles, monthly and yearly, on an average 
 of forty years, ending with 1894, and I find a remarkable agree- 
 ment between the two sets of observations. 
 
 A necessary consequence of the changes in the monthly dis- 
 tribution of atmospheric pressure above indicated is, that January 
 is the stormiest month in the British Isles. A careful analysis 
 of the reports of storms received at the Meteorological Office, 
 London, for the fourteen years, 1870-1883, has led Dr. E. H. Scott 
 to the conclusion that there is no strongly-marked storm-maxi- 
 mum at either equinox in September storm prevalence is 
 increasing from a marked minimum in June and July ; in March 
 it is decreasing from a sharply-defined maximum in January. 
 Equinoctial gales, as such, are non-existent. 1 
 
 1 Quarterly Journal of the Royal Meteorological Society, vol. x., p. 236. 
 1884. 
 
CHAPTER XXYI 
 
 THE CLIMATE OF THE BRITISH ISLANDS (continued) 
 IV. KAINFALL. 
 
 IT is well said by Dr. A. Buchan, in his third paper on "The Climate 
 of the British Islands/' 1 that, as regards these islands, the greatest 
 differences in local climates arise from differences in the rainfall. 
 Thus, on comparing the climate of Skye with that of the southern 
 coasts of the Moray Firth, their mean temperatures in no month 
 differ so much as 2-0, and for several months of the year they are 
 nearly identical. But the annual rainfall of Skye rises to, and 
 in many places exceeds, 100 inches, whereas at Culloden it only 
 amounts to 26*17 inches (24*6 inches in the twenty-five years 
 1866-1890), and at Burghead to 25-23 inches. This difference in 
 the rainfall, with the clear skies and strong sunshine which accom- 
 pany it, renders the south shore of the Moray Firth one of the 
 finest grain-producing districts of Scotland. It is this aspect of 
 the rainfall which gives it so prominent a place in the climatology 
 of a country. 
 
 Dr. Buchan's article on " The Annual Rainfall of the British 
 Islands " is based on observations of the rainfall made at 547 
 stations in Scotland, 1,080 in England and Wales, and 213 in 
 Ireland ; in all, 1,840. The period selected for discussion was 
 the twenty-four years extending from 1860 to 1883, inclusive. 
 Dr. Buchan handsomely acknowledges his obligations to Mr. 
 Symons's British Rainfall, a publication which rendered such an 
 inquiry possible. The observed, but in some instances calculated, 
 twenty-four years' averages were transferred to a map of the 
 British Islands, which was then coloured with six different tints 
 
 1 Journal of the Scottish Meteorological Society. New Series. Vol. vii., 
 p. 131. 1886. 
 
396 METEOROLOGY 
 
 these shadings showing the districts where the mean annual 
 rainfall did not exceed 25 inches (pale pink), was from 25 to 
 30 inches (red), from 30 to 40 inches (dark red), from 40 to 60 inches 
 (pale blue), from 60 to 80 inches (blue), and, lastly, above 80 inches 
 (dark blue). 
 
 In the Meteorological Atlas of the British Isles (1883), the rainfall 
 for the whole year is shown on a map by lines drawn, for each 
 5 or 10 inches of rain, from place to place having the same annual 
 rainfall. This rainfall map was constructed by plotting on a large 
 scale all the mean annual rainfall values for the fifteen years 
 1866-1880, inclusive, which are given in Rainfall Tables of the 
 British Isles, compiled from the records of 366 stations by Mr. G. J. 
 Symons, F.R.S., and published by the authority of the Meteoro- 
 logical Council in 1883. The rainfall indicated by the lines 
 upon a map, drawn on a reduced scale for the Atlas, is the mean 
 quantity actually observed at these selected stations. It is 
 not claimed that this map is perfect, for a much larger amount 
 of rain is known to fall in some places, notably on mountain 
 slopes, of which no such record exists as to admit of its being 
 shown on so small a map. In Ireland and the west of Scotland 
 this defect is aggravated by the fact that for large areas no 
 record whatever is obtainable for the period embraced in the 
 inquiry. 
 
 It should be mentioned that the Rainfall Tables, prepared 
 by Mr. Symons at the request of the Meteorological Council, 
 are illustrated by three coloured maps of England, Scotland, and 
 Ireland respectively, which exhibit, not only the geographical 
 position of the 366 stations furnishing the rainfall records, but 
 also the area in square miles of the river catchment basins in 
 which these stations are severally situated. 
 
 The key to the distribution of rainfall in the British Islands 
 is " the direction of the rain-bearing winds in their relation to 
 the physical configuration of the surface " (Buchan), 
 
 The regions of heaviest rainfall, 60 to 80 inches or upwards, 
 are : Skye and the Western Highlands, the Lake District in 
 Cumberland and Westmoreland, the mountainous district in 
 North Wales, the mountainous district in the south-east of Wales, 
 Dartmoor in Devonshire, the Highlands of West Galway, 
 
THE CLIMATE OF THE BRITISH ISLANDS 397 
 
 and the neighbourhood of Killarney and the Macgillicuddy's 
 Reeks in Kerry. This distribution of heavy rainfall is deter- 
 mined by (1) prevalent south-west winds, blowing vapour-laden 
 from the Atlantic Ocean ; (2) the exposure to these winds of 
 mountains, or high tablelands like Dartmoor, with valleys open- 
 ing to the westward or south-westward. On the mountain slopes 
 the warm, moist air is condensed into mist, cloud, and rain. 
 
 Over the south of Scotland the rainfall is not excessive, because 
 the rain-bearing south-westerly winds have been partially dried 
 in their passage across Ireland before they reach the district in 
 question. 
 
 The absolutely largest annual rainfalls are : In Scotland, at 
 Ben Nevis summit, 4,404 feet above the sea, [fifteen to eighteen 
 years] (151 inches), and at Glencroe, Argyllshire, at an elevation 
 of 520 feet (128*50 inches) ; in England, at the Stye, Cumberland, 
 1,077 feet (185'96 inches so far as yet observed, the heaviest 
 rainfall anywhere in the British Islands), at Seathwaite, Cumber- 
 land, 422 feet (143*21 inches in the twenty-four years 1860-1883 ; 
 139-29 inches in the fifteen years 1866-1880) ; in Wales, at 
 Beddgelert, Carnarvonshire, 264 feet (116-90 inches), Rhiwbrifdir. 
 Merionethshire, 1,100 feet (102'56 inches), Ty-Draw-Treherbert, 
 Glamorgan, 735 feet (96-18 inches), Glyncorrwg, Neath, [twenty 
 years] (87' 5 inches) ; in Ireland, on Mangerton, near Killarney, 
 [eight years] (86 inches), at Kylemore, County Galway, 105 feet, 
 [fifteen years] (77'6 inches), at Foffany, County Down, 920 feet 
 (72-26 inches), at Newcastle, County Down, and at Derreen, 
 Kenmare, County Kerry, 74 feet (69*40 inches). 
 
 A rainfall of 40 inches a year, or upwards, occurs over about a 
 fourth part of the surface of England and Wales, about half of 
 that of Ireland, and considerably more than half of that of 
 Scotland (Buchan). Nowhere along the whole east coast of 
 Great Britain, or for some distance inland, does the average 
 yearly rainfall reach 40 inches. On the east coast of Ireland, 
 however, the rainfall rises to, or exceeds, 40 inches in the moun- 
 tainous districts of Wicklow, Down (the Mourne Mountains), 
 and Antrim. On the other hand, the annual rainfall is well 
 below 30 inches in Dublin and its vicinity, for reasons which have 
 been already explained (see Chapter XXIV., p. 363). 
 
398 METEOROLOGY 
 
 Wherever mountains or " downs " run east and west, a heavy 
 rainfall is propagated eastwards along their southern face, while 
 the precipitation is diminished to the northward of the barrier 
 which they oppose to the rain-bearing south-west winds. Thus 
 the mountains of Sutherland, the Grampians, the Cheviots, the 
 Pennine Range, and the Downs of the south of England, all cause 
 an extension eastward of a heavier rainfall along their southern 
 slopes, but a diminution in the rainfall to the northward and 
 north-eastward. Precisely the same thing on a smaller scale is 
 found in connection with the Pentland Hills, near Edinburgh, 
 the Mourne Mountains in the County Down, and the Dublin and 
 Wicklow mountains. Leith (28'00 inches), Edinburgh (28- 31 
 inches), Donaghadee (31 '08 inches), and Dublin (27' 672 inches in 
 the forty years 1866-1905), all owe their comparatively small pre- 
 cipitation to their geographical position north-east of the moun- 
 tain ranges mentioned. The rainfall at Belfast (Queen's College) 
 is, on the average, 34*73 inches. This is no doubt due to the 
 proximity of Divis and other mountains north-west of the city, 
 heavy rains falling with south-east winds, which impinge upon 
 those hills. 
 
 " The influence of the breakdown of the watershed of Scot- 
 land between the Firths of Forth and Clyde," writes Dr. Buchan, 
 " is strikingly manifested in the overspreading of western parts 
 of Perthshire, Stirlingshire, and Dumbartonshire, with a truly 
 western rainfall as regards amount, and the direction of the winds 
 with which it falls ; and in the extension eastwards, through 
 Kinross-shire, of a rainfall of fully 40 inches, which occurs no- 
 where else over comparatively level plains so far to the east 
 of the watershed separating the western and eastern districts." 
 
 Turning to the regions of least rainfall, we find a large area 
 in England, extending from the Humber to the estuary of the 
 Thames (exclusive of the higher grounds of Lincoln and Norfolk), 
 over which the annual precipitation falls short of 25 inches. 
 In Cambridgeshire it is generally about 23 inches, except at 
 Wisbech Observatory, where it rises to 26J inches. The smallest 
 rainfall of all is at the eastern point of Spurn Head, Yorkshire 
 (ten years), 19-1 inches. At Shoeburyness, in Essex, the average 
 for the twenty-five years 1866-1890 was only 20'6 inches. In 
 
THE CLIMATE OF THE BRITISH ISLANDS 399 
 
 1874 only 14-20 inches fell at this station. On the higher grounds 
 of Lincoln and Norfolk the rainfall exceeds 25 inches, because 
 the precipitation with easterly winds is increased. Similarly, 
 the rainfall of the Yorkshire Wolds is made to exceed that of 
 neighbouring districts. A small patch in the valley of the 
 Thames, from Kew to Marlow, in Bucks, has an annual fall of 
 less than 25 inches Kew Observatory, twenty-four years 1860- 
 1883 = 25-26 inches; fifteen years 1866-1880 = 24-67 inches, 
 Between the valley of the Thames and the Humber the rainfall 
 nowhere reaches 30 inches, except near the Chiltern Hills 
 (Buchan). 
 
 In Scotland the annual rainfall falls short of 30 inches in the 
 north-eastern part of Caithness, round the Moray Firth from 
 Tain to the mouth of the Spey, along the east coast from Peter- 
 head in Aberdeenshire to Burntisland in Fifeshire, the low 
 ground of Midlothian and East Lothian, and lower Tweeddale 
 from Kelso to Berwick. The absolutely smallest rainfalls are 
 observed on the very shores of the Moray Firth from Tarbet- 
 Ness to Burghead (25 to 28 inches), the extreme north-east of 
 East Lothian (25 to 29 inches), and the lower Tweed from Cold- 
 stream to Jedburgh (26 \ inches). Nairn, in the twenty-five years 
 1866-1890, had only 23'3 inches annually on the average. 
 
 The only parts of Ireland where the rainfall falls short of 
 30 inches are Dublin and its vicinity (about 28 inches) and 
 Dundalk (29'9 inches on an average of ten years). The reason 
 for this diminished rainfall has been given above (p. 363). 
 
 V. GEOLOGICAL FORMATION. 
 
 In a lecture on " The Physical Influences which affect the 
 British Climate," delivered early in the year 1893, in the Public 
 Health Department of King's College, London, 1 Mr. H. G. Seeley, 
 F.K.S., Professor of Geography in the College, aptly observes : 
 " The areas of heavier rainfall are the regions of higher land, 
 and a rainfall map in England closely approximates in its broad 
 features to a geological map." Allusion has already been made 
 to the marked influence on rainfall exercised by the high lands 
 and mountain chains in various parts of the United Kingdom. 
 1 See The Journal of State Medicine, vol. i. ; No. 4, p. 165. April, 1893. 
 
400 METEOKOLOGY 
 
 The configuration of the coast-line and the distribution of high 
 and low ground govern the rainfall to a remarkable extent, 
 and in this way control climate generally. As Mr. Seeley 
 remarks, " Next to the situation of our islands upon the earth's 
 surface, the most important element in climate is the geological 
 structure and contour of the surface of the country." 
 
 In his lecture Professor Seeley shows that there are two 
 ways in which the geological structure affects climate : first, it 
 has a local influence on temperature ; secondly, it is a main 
 element in modifying the relative durability of the rock material 
 which determines the elevation of the surface. It would be 
 foreign to the purpose of this book to enter into details as to the 
 geology of the British Islands. Suffice it to say, with Professor 
 Seeley, that the chief geological formations which have a bearing 
 upon climate may be classed as (1) pebble beds, sands, and sand- 
 stones ; (2) clays and shales ; (3) limestones. There are, in 
 addition, certain altered conditions of these simple forms, in 
 which a more or less crystalline texture is developed, which may 
 be the micro-crystalline texture of slate or the micro-crystalline 
 texture of schist. 
 
 1. The pebble beds, sands, and sandstones are commonly 
 warmer and drier than other rocks. Their dryness is due to 
 the existence of porous interspaces between the quartz grains 
 of which these rock forms are so largely composed. This whole- 
 some property may be interfered with by the presence of a 
 cement which will bind the grains of quartz together, or of a bed 
 of clay, which will render the sand impervious. The dryness, 
 or otherwise, of sand strata will also largely depend on the angle 
 at which they are inclined ; horizontal strata are naturally less 
 dry than inclined strata. The warmth of a sandy soil is probably 
 due to its dryness as well as to the low specific heat of the quartz. 
 
 2. The particles of which clays are composed are extremely 
 small, and consist chiefly of silicate of alumina. Some clay 
 soils contain as much as 40 per cent, of alumina, but the usual 
 proportion is much smaller. In Scotland clay soils are found 
 chiefly on the coal-measures, the boulder clay, and as alluvium 
 in the valleys. The last named is the richest form of clay and 
 is known as carse clay. In the North of England, the aluminous 
 
THE CLIMATE OF THE BRITISH ISLANDS 401 
 
 shales of the coal-measures yield soils in their properties very 
 like those in Scotland. England also abounds in clay soils 
 derived from other geological formations, such as London clay, 
 plastic, weald, gault, and blue lias clay. An astonishing quantity 
 of water may be held in a clay soil, which has an almost bound- 
 less affinity for moisture. In a warm, dry summer, wide and 
 deep chasms open in clay, owing to the evaporation of the water 
 it contains. Clay land is looked upon as cold a condition 
 attributed by Professor Seeley, theoretically, to the small size 
 of its constituent particles and the way in which they are divided 
 from each other by films of water. Through evaporation from 
 a clay soil, the superincumbent atmosphere is rendered moist 
 and cool. Precisely the same effect is produced in Ireland by 
 the water-soaked morasses or bogs, which, according to Sir 
 Robert Kane, M.D., cover 2,830,000 acres, or about one-seventh 
 part of the entire surface of the island. The Bog of Allen 
 stretches in a vast plain across the centre of the island, having 
 a summit elevation of 280 feet. Its apparent influence on the 
 mean temperature has been alluded to above (see p. 390). 
 
 Owing to the retentive and non-porous nature of clay soils, 
 there is no deep filtration and underground storage of water. 
 The superficial strata become water-logged, and the imprisoned 
 waters are very liable to contamination with surface impurities 
 or with the products of chemical decomposition in the clay itself. 
 
 3. The third great group of water-formed rocks is that of 
 the limestones (Seeley). The carboniferous limestone covers an 
 immense area in the Pennine Chain and North of England, and 
 forms the base of nearly all our coalfields. It also underlies 
 the peat-moss bogs of Central Ireland. The oolitic limestones, 
 more or less continuous, and chalk, stretch from the Yorkshire 
 coast to the South of England, forming parallel ridges of hills. 
 Being soluble under flowing waters charged with carbonic acid 
 gas, the surface is always deeply scored with valleys (Seeley). 
 The ancient and oolitic limestones are not as absorbent as the 
 newer chalk, which is very pervious to water. As water per- 
 colates through these limestones, it becomes highly charged 
 with lime salts. According to Professor Seeley, limestones 
 always give up a good deal of vapour under sunshine, and have 
 
 26 
 
402 
 
 METEOROLOGY 
 
 a warm steamy atmosphere above them in summer, which is 
 in marked contrast to the bracing air of sandstones with silicious 
 or calcareous cements. When there is even a thin bed of clay 
 on the summit of a limestone ridge, such as forms the insoluble 
 residue left by atmospheric denudation, the climatic conditions 
 are changed. 
 
 4. The crystalline rocks, whether slates or schists, usually 
 occur in elevated country in the West of England, in Scotland, 
 and in parts of Ireland. They are remarkable, chiefly, for their 
 i mpervious and almost insoluble character. " The high and 
 irregular ground which they form, like their western position, 
 causes them to have a great effect in radiating heat, and there- 
 fore in producing winds which descend from the mountainous 
 regions, and in condensing rain." 
 
 Professor Seeley, in the paper from which I have so largely 
 quoted, observes with justice that " the influence of the soil 
 upon climate is complicated by the effects of the climate in 
 transporting and forming the superficial soil/' 
 
 The foregoing sketch of the climate of the British Islands 
 may fitly conclude with Climatological Tables for the City of 
 Dublin, lat. 53 20' N., long. 6 15 /0 W., altitude 18 to 67 feet. 
 
 TABLE VII. SHOWING THE AVERAGE MEAN TEMPERATURE, ATMOSPHERIC 
 PRESSURE, RAINFALL, AND RAIN DAYS IN THE CITY OF DUBLIN DURING 
 THE FORTY YEARS 1866-1905. 
 
 Month. 
 
 Average. 
 
 Mean Temp. 
 
 Pressure. 
 
 Rainfall. 
 
 Rain Days. 
 
 
 F. 
 
 Inches. 
 
 Inches. 
 
 January 
 February 
 March 
 
 41-6 
 42-5 
 43'4 
 
 29-913 
 29-915 
 29-892 
 
 2-287 18-0 
 1-954 15-8 
 2-030 16-9 
 
 April . . 
 
 47-6 
 
 29-896 
 
 1-913 
 
 15-7 
 
 May .. 
 
 51-9 
 
 29-976 
 
 2-027 
 
 14-9 
 
 June . . 
 
 57-7 
 
 29-994 
 
 1-912 
 
 14-3 
 
 July .. 
 
 60-4 
 
 29-948 
 
 2-510 
 
 16-6 
 
 August 
 
 59-5 
 
 29-917 
 
 3-130 
 
 17-0 
 
 September 
 October 
 
 55-8 
 49-5 
 
 29-938 
 29-887 
 
 2-243 
 
 2-803 
 
 15-2 
 17-4 
 
 November 
 
 45-1 
 
 29-896 
 
 2-587 
 
 16-6 
 
 December 
 
 41-9 
 
 29-863 
 
 2-275 
 
 17-5 
 
 Annuals 
 
 49-7 
 
 29-920 
 
 27-672 
 
 195-9 
 
THE CLIMATE OF THE BRITISH ISLANDS 403 
 
 The materials for Table VII. (see p. 402) were culled from the 
 records of observations taken by me in the City of Dublin during 
 the forty years 1866-1905, inclusive. 
 
 The following averages of temperature, rainfall, and duration 
 of bright sunshine are based on observations taken during the 
 thirty-five years, 1871-1905, at three Normal Climatological 
 Stations situated in the Irish metropolis. They are extracted 
 from Appendix III. of the Weekly Weather Report, 1906, published 
 by the Meteorological Office, London. 
 
 Trinity College Meteorological Observatory. In January, 1904, 
 the Provost and Senior Fellows of Trinity College established a 
 Normal Climatological Station within the precincts of the College. 
 The station occupies an open space in the Fellows' Garden, and 
 is fully equipped. In addition to the usual instruments 
 barometer, dry-bulb, wet-bulb, maximum and minimum thermo- 
 meters, and rain-gauge, all of which are read at 9 a.m. and 9 p.m. 
 the equipment includes a Campbell-Stokes sunshine-recorder 
 and two earth - thermometers, of which the bulbs are placed 
 underground at a depth of 1 foot and 4 feet respectively. The 
 Observatory is under the superintendence of Erasmus Smith's 
 Professor of Natural and Experimental Philosophy, W. E. Thrift, 
 M.A., F.T.C.D. 
 
 The tables on pp. 405-406 have been compiled from the records 
 of this Climatological Station, extending over the five years 1904- 
 1908, inclusive. They show the monthly and yearly values of 
 the underground temperatures at 4 feet, and of the .duration of 
 bright sunshine in hours and percentages. 
 
 262 
 
404 
 
 METEOKOLOGY 
 
 TEMPEEATTJRE. 
 
 
 Dublin City 
 (Fitzwilliam Square). 
 
 Royal Botanic Gardens 
 (Glasnevin). 
 
 Ordnance Survey 
 (Phoenix Paik). 
 
 
 Mean 
 Max 
 
 Mean 
 Min. 
 
 Mean . 
 
 Mean 
 Max. 
 
 Mean 
 Min. 
 
 Mean. 
 
 Mean 
 Max. 
 
 Mean 
 Min. 
 
 Mean. 
 
 
 F. 
 
 F. 
 
 o F 
 
 F. 
 
 F. 
 
 F. 
 
 F. 
 
 F. 
 
 op 
 
 January . . 46'0 
 
 37-4 
 
 41-7 
 
 45'8 
 
 35-2 
 
 40-5 
 
 45-5 
 
 347 
 
 40-1 
 
 February .. 46'9 
 
 37-9 
 
 42-4 
 
 47-1 
 
 35-3 
 
 41-2 
 
 46-5 
 
 347 
 
 40-6 
 
 March . . 49'2 
 
 38'2 
 
 437 
 
 49-3 
 
 35-6 
 
 42-4 
 
 487 
 
 35-0 
 
 41-9 
 
 April .. 53-6 
 
 41-5 
 
 47-6 
 
 53-5 
 
 38-4 
 
 46-0 
 
 52-8 
 
 37-8 
 
 45-3 
 
 May . . 58-8 
 
 45-5 
 
 52-2 58-2 
 
 41-9 
 
 50-1 
 
 577 
 
 41-3 
 
 49-5 
 
 June . . 64-5 
 
 51'3 
 
 57'9 ! 64-5 
 
 47'9 
 
 56-2 
 
 63-9 
 
 47-1 
 
 55-5 
 
 July . . 66'8 
 
 54-2 
 
 60-5 67'0 
 
 51-3 
 
 59-2 
 
 66-1 
 
 50-6 
 
 58-4 
 
 August . . 65 '7 
 
 53-6 
 
 59-7 ! 66'2 
 
 50-7 
 
 58-5 
 
 65-4 
 
 50-1 
 
 577 
 
 September 617 
 
 50-2 
 
 55-9 62-4 
 
 47-0 
 
 54-7 
 
 61-6 
 
 46-3 
 
 54-0 
 
 October . . 547 
 
 44-3 
 
 49-5 55-2 
 
 41-4 48-3 
 
 54-5 
 
 41-1 
 
 47-8 
 
 November 49 '9 
 
 40-8 
 
 45-3 ! 50-0 
 
 38-2 
 
 44-1 
 
 49'6 
 
 377 
 
 437 
 
 December . . 46'3 
 
 37-6 
 
 42-0 
 
 46-0 
 
 35-0 
 
 40-5 
 
 45-8 
 
 34-6 
 
 40-2 
 
 Whole Year 55-3 
 
 44-4 
 
 49-9 
 
 55-4 
 
 41-5 
 
 48-5 
 
 54-8 
 
 40-9 
 
 47-9 
 
 
 
 1 
 
 
 
 
 
 
 The respective heights of the three stations above mean sea-level are 
 Fitzwilliam Square, 47 feet ; Glasnevin, 67 feet ; Phoenix Park, 155 feet. 
 
 RAINFALL AND RAIN DAYS. 
 
 
 Dublin City 
 (Fitzwilliam Square). 
 
 Royal Botanic 
 Gardens (Glasnevin) 
 (1875-1905). 
 
 Ordnance Survey 
 (Phoenix Park). 
 
 Month. 
 
 
 
 
 Rainfall in 
 
 Rain 
 
 Rainfall in 
 
 Rain 
 
 Rainfall in 
 
 Rain 
 
 
 Inches. 
 
 Days. 
 
 Inches. 
 
 Days. 
 
 Inches. 
 
 Days. 
 
 January 
 
 2-21 
 
 18 
 
 2-28 
 
 15 
 
 2-24 
 
 20 
 
 February . . 1 2'01 15 
 
 1-99 
 
 14 
 
 1-89 
 
 17 
 
 March . . . . 1'91 
 
 17 
 
 2-05 
 
 14 
 
 2-03 
 
 19 
 
 April 
 
 1-94 
 
 16 
 
 1-94 
 
 14 
 
 1-84 
 
 17 
 
 May 
 
 1-97 
 
 15 
 
 2-01 
 
 13 
 
 1-99 
 
 16 
 
 June 
 
 1-99 
 
 15 
 
 2-08 
 
 13 
 
 2-08 
 
 15 
 
 July 
 
 2-68 
 
 17 
 
 275 
 
 16 
 
 2-84 
 
 19 
 
 August 
 
 3-24 
 
 18 
 
 3-45 
 
 17 
 
 3-32 
 
 19 
 
 September 
 
 2-21 
 
 15 
 
 2-33 
 
 13 
 
 2-22 
 
 16 
 
 October 
 
 2-87 
 
 18 
 
 2-86 
 
 16 
 
 2-85 
 
 19 
 
 November 
 
 272 
 
 17 
 
 278 
 
 15 
 
 2-87 
 
 19 
 
 December 
 
 2-25 
 
 17 
 
 2-29 
 
 16 
 
 2-35 
 
 19 
 
 Totals . . 
 
 28-00 
 
 198 
 
 28-81 
 
 176 
 
 28-52 
 
 215 
 
I 
 
 . 
 THE CLIMATE 
 
 OF THE BEITISH ISLANDS 405 
 
 TABLE VIII. SHOWING THE MONTHLY AND YEARLY DURATION OF BRIGHT SUNSHINE IN HOURS, WITH THE 
 PERCENTAGE OF THE GREATEST POSSIBLE DURATION, TOGETHER WITH THE AVERAGES FOR FIVE YEARS, 
 RECORDED IN TRINITY COLLEGE, DUBLIN. 
 
 f 
 
 Per Cent. 
 
 CO 
 
 88Sii8833 8 
 
 ^eor^t-ooiOT^T^'icO' i ~< 
 
 NCOCOCOCOCOCOCOC<J' 1> 1 CO 
 
 W 
 
 CO 
 
 35r le^xooi iioooot^-o ^f 
 
 1 1 
 
 lO O C^ r~i "^ O5 CO lr* "^ *H CO ' O* 
 
 
 * 
 
 11 ' 
 
 saassssssss- & 
 
 g 
 
 o 
 
 ^ 
 
 OOOOOO5^COI>OCOr-HCO CO 
 
 CO 
 CO 
 
 COrt^OOCOi I^OOi lt-OOi-H O5 
 ^OCOO500l>.-^OOO^<M t^ 
 
 1 1 
 
 < 
 
 II 
 
 O5 
 i-H 
 
 OOOOOOCOCOlO^COt^COO : 00 
 <NCO<NCO<NCOCO<Mr-i' 1> i <N 
 
 Hours. 
 
 9 
 
 !>. 
 
 * 
 
 oooooooo>oop o 
 
 CO OO t^ CO O5 CO ""^ 00 ^O ^ ^ ^O 
 t-COi-Ht^C^t^-lOOSOCOC^ J^ 
 
 i-H 
 
 
 
 11 
 
 t^* ^H ^ T^ t^ C^ IT*- Oi *O ^ tr^* O5 
 
 p-H CO CO Tt* G^ ^ CO CO ^ CO ^^ CO 
 
 Hours. 
 
 
 1 
 
 COOOOlOCOOOC5rt<t^0005 ! "* 
 
 9 
 
 4tl<N<NC<lo4tlcbo5O5COO5 O5 
 00<NOOCOi-HOOt^COO5-*rH > CO 
 
 - 1 
 
 >o 
 
 
 
 *i 
 
 CO 
 
 SSS^S?R^23 . S 
 
 i 
 
 w 
 
 o 
 
 0500t-CO(NC5^^COCO 5 
 
 g 
 
 00 "^ O '"^ f ~* CO C1 C^ Oi ^ CO 00 
 i-H 
 
 i 
 
 ll 
 
 2 
 
 S838SS^83S 8 
 
 Hours. 
 
 CO 
 
 COOOOIOIOOOOOCO<NCO O 
 
 ^ 
 
 COOOCOO5COOOOOOOCOCO ! U5 
 
 s 
 
 
 1" 
 
 1-5 
 
 b J ^^ * 
 
 1 It fr- 7 -f 1 . S 1 #ni 
 
 ^|^|9g^^<SP 
 
406 
 
 METEOROLOGY 
 
 TABLE IX. SHOWING THE MONTHLY AND YEARLY MEAN TEMPERATURE 
 OF THE SUBSOIL AT A DEPTH OF 4 FEET BELOW THE SURFACE, TOGETHER 
 WITH THE AVERAGE SUBSOIL TEMPERATURE FOR THE FIVE YEARS 
 
 1904-1908, RECORDED IN TfilNITY COLLEGE, DjBLIN. 
 
 Month. 
 
 1904. 
 
 1905. 
 
 1906. 
 
 1937. 
 
 1938. 
 
 Avenge. 
 
 
 OF. 
 
 o p 
 
 F. 
 
 op 
 
 OF 
 
 o F> 
 
 January . . 
 
 42-9 
 
 44-7 
 
 447 
 
 44-0 
 
 43-0 
 
 44-6 
 
 February 
 
 42-5 
 
 44'2 
 
 43-4 
 
 42-2 
 
 44-3 
 
 43-3 
 
 March 
 
 43-3 
 
 44-1 
 
 44-0 
 
 44-2 
 
 43-9 
 
 43-9 
 
 April 
 
 46-3 
 
 46-4 
 
 45-9 
 
 46-4 
 
 45-8 
 
 46-2 
 
 May 
 
 50-0 
 
 50-3 
 
 48-6 
 
 49-8 
 
 50-0 
 
 49-7 
 
 June 
 
 54-2 
 
 54-6 
 
 53-9 
 
 52-9 
 
 54-1 
 
 53-9 
 
 July 
 
 567 
 
 58-6 
 
 56-5 
 
 55-7 
 
 57-7 
 
 57-0 
 
 August 
 
 57-9 
 
 58-5 
 
 58-4 
 
 57-3 
 
 58-4 
 
 58-1 
 
 September 
 
 56-4 
 
 56-1 
 
 58-0 
 
 56-3 
 
 56-2 
 
 56-6 
 
 October . . 
 
 53-2 
 
 51-9 
 
 54-0 
 
 53-8 
 
 55-6 
 
 537 
 
 November 
 
 50-0 
 
 467 
 
 48-8 
 
 49-4 
 
 51-2 
 
 49-2 
 
 December 
 
 45-3 
 
 45-6 
 
 46-7 
 
 45-3 
 
 47-1 
 
 46-0 
 
 Yearly means 
 
 49-9 
 
 50-1 
 
 50-2 
 
 49-8 
 
 50-6 
 
 50-1 
 
 Is our Climate changing ? To this question I endeavoured 
 to give an answer in a paper read before the Cosmical Physics 
 Subsection of the Section of Mathematics and Physics, at the 
 Dublin Meeting of the British Association for the Advancement 
 of Science, September, 1908. 
 
 In 1770 Dr. Thomas Eutty published an octavo work of 
 340 pages, entitled A Chronological History of the Weather and 
 Seasons, and of the Prevailing Diseases in Dublin. The results 
 of forty years' observations are recorded in this most valuable 
 volume. More than a century earlier Dr. Gerard Boate, State 
 Physician to the Parliamentary Forces in Ireland, made ob- 
 servations on climate and diseases, and, with the assistance of 
 his brother Arnold, who had been for many years a medical 
 practitioner in Dublin, wrote his work on Ireland's Natural 
 History. It was first published in London in 1652, three years 
 after Gerard Boate's death. A French edition appeared in Paris 
 in 1666. 1 
 
 From 1805 to 1841 Dr. Thomas H. Orpen made observations 
 on the weather in Dublin, which are still preserved in manu- 
 
 1 Census of Ireland, 1851. Part V., "History of Epidemic Pestilences in 
 Ireland, A.M. 2820 to A.D. 1851." 
 
THE CLIMATE OF THE BRITISH ISLANDS 407 
 
 script in the Library of the Royal Irish Academy. Some years 
 of Dr. Orpen's Tables were likewise published in the Dublin 
 Philosophical Journal in 1825, which also contains a meteoro- 
 logical table for 1823-24, by Mr. Semple, of Malahide. From 
 1829 down to the present time a careful series of observations 
 has been made at the Ordnance Survey Office, Mount joy Barracks, 
 in the Phoenix Park, and at the beginning of the nineteenth 
 century the observations taken at the Royal Botanic Gardens, 
 Glasnevin, were published yearly in the Proceedings of the Royal 
 Dublin Society. 
 
 Rutty's remarks on the weather in the early years of the 
 eighteenth century would serve to describe accurately the weather 
 of the twentieth century. Here is one brief quotation : 
 
 1760-61. " The winter very open and warm. December and 
 January windy and stormy." 
 
 And then this shrewd observer remarks : " December (1760) 
 was healthy, as was this whole winter quarter, though uncom- 
 monly wet and warm, an express contradiction to the vulgar 
 tradition that a green Christmas makes a fat churchyard " 
 (p. 254). 
 
 1763. "A remarkably cold and wet summer, but healthy. 
 . . . The wet summer, particularly the month of August, was 
 less productive of diarrhoeas, dysenteries, and the cholera than 
 drier and warmer seasons, which seems to furnish occasion for 
 refuting a vulgar and long-established error and prejudice respect- 
 ing the cause of these diseases, which have been ordinarily 
 attributed to an obstructed perspiration and the use of fruit in 
 the summer and autumn " (pp. 297, 298). 
 
 Having extracted evidence from the records of the past which 
 goes to prove that the climate of the British Isles was much the 
 same long ago as it is at the present day, I submitted, in further 
 proof of this contention, the results of my personal observations 
 in Dublin through a long series of years. As a matter of fact, 
 I have kept a Weather Journal since the year 1861 to the present 
 date. The period dealt with began with the year 1866, and 
 ended with the year 1905 that is to say, an even period of forty 
 years, or of eight lustrums. My son, Maurice Sydney Moore, 
 B.A. Dublin, prepared the tables, which set forth the mean 
 
408 METEOEOLOGY 
 
 temperature, rainfall, rain-days, and atmospheric pressure of 
 the period, and worked out the necessary calculations. 
 
 The average annual mean temperature of the forty years 
 was 49-7 F., the lustrum averages being 50-1, 50-2, 49-3, 
 49-4, 48-6, 49-4, 50-8, and 50-0 F. 
 
 A careful study of the analysis of the behaviour of tempera- 
 ture in the forty years under discussion shows that, no matter 
 what fluctuations take place between individual months in suc- 
 cessive years or between individual years in successive lustrums, 
 the temperature pendulum swings back to its original position 
 at either side of the average. The question of periodicity in 
 such movements awaits solution. 
 
 Even in the matter of rainfall, with the wide swing of the 
 pendulum from 45 per cent, in excess, to 40 per cent, in defect 
 of the average that is, from a maximal yearly fall of 35-566 
 inches to a minimal fall of 16-601 inches we see a tendency to 
 return to that average as the years roll by. 
 
 In the forty years, 1866-1905, the mean annual atmospheric 
 pressure was lowest 29*731 inches in 1872, and highest 
 30-015 inches in 1887. In 1872 the monthly pressure was 
 below the average in every month except in April and August. 
 In 1887 the monthly pressure was above the average except in 
 January, September, November, and December. 
 
 In conclusion, the facts which I put forward in this paper 
 prove that, within the past six centuries at all events, no appre- 
 ciable change has taken place in the climate of the British Isles. 
 
 There is not a scintilla of evidence to show that within 
 historic times any such change has taken place in the past, or 
 is likely to take place in the future. The weather " varium 
 et mutabile semper," as it is resembles the river which, in the 
 words of Horace, " Labitur, et labetur in omne volubilis sevum/ 3 
 
PART IV. THE INFLUENCE OF SEASON 
 AND OF WEATHER ON DISEASE 
 
 CHAPTER XXVII 
 ACUTE INFECTIVE DISEASES 
 
 OBSERVATIONS as to the influence of weather upon health are 
 as old as meteorology itself nay, older, if we admit that Aristotle 
 was the founder of the science. More than four hundred years 
 before Christ, Hippocrates of Cos, the " Father of Medicine," had 
 penned his immortal Aphorisms, and had written " Ile/oi ae/xoi/, 
 I'SaTwi/, TOTTtov " (On Air, Waters, and Places), and " He/at Sicu-njs" 
 (On Regimen). In these works we meet with passages as ap- 
 plicable to-day as they were some twenty-four centuries ago. 
 
 The suggestions thrown out by the Greek physician were 
 allowed to remain almost a dead-letter. His doctrines as to the 
 close relations of Climatology to Medicine became dimmed by 
 the rust of time, and were neglected or forgotten. 
 
 In the Introduction to his splendid Geographical and Historical 
 Pathology, August Hirsch observes that only in a few of the 
 best Greek and Roman medical authors, such as Celsus, Ascle- 
 piades, and Aretseus, do we find here and there indications that 
 they gave some attention to the various effects of " climate " 
 and " diet " upon the human organism in health and disease. 
 Such questions were unfamiliar to the physicians of the Middle 
 Ages, and it was only in the sixteenth century that naturalists 
 and physicians again began to investigate the changing aspects 
 of organic life, including the life of man, in various quarters of 
 the globe. 
 
 That in these countries but little attention was given to the 
 
 409 
 
410 METEOROLOGY 
 
 subject is evident from the antiquity and popularity of the 
 proverb, " A green Christmas makes a fat churchyard/' Even 
 Sydenham stated 1 that a prevailing epidemic ceased on the 
 approach of winter a statement which is no doubt true in the 
 case of Asiatic cholera, but is of by no means universal or even 
 common application. On the whole, however, Sydenham' s 
 observations on the dependence of disease on season are accurate, 
 and well worth perusal. 
 
 The first modern paper on Weather and Disease was a com- 
 munication made to the Koyal Society in 1797 by Dr. William 
 Heberden, jun., F.R.S., on the " Influence of Cold on the Health 
 of the Inhabitants of London/' 2 The author showed that a 
 difference of above 20 F. between the mean temperatures in 
 London in January, 1795, and that in the same month in 1796 
 the former being an excessively cold month, and the latter an 
 equally mild one caused the deaths in January, 1795, to exceed 
 those in January, 1796, by 1,352. 
 
 In my remarks on the influence of season and weather on 
 disease, I shall confine myself almost exclusively to three meteoro- 
 logical factors mean temperature, rainfall, and humidity. Of 
 these the first is the most important, as it is, in truth, the resultant 
 of many other factors. In the following chapters we shall 
 consider the influence of season and weather upon some of the 
 principal acute infective diseases. 
 
 1. Influenza. 
 
 On February 28, 1890, I read a paper before the Eoyal 
 Academy of Medicine in Ireland on the " Influenza Epidemic of 
 1889-90, as observed in Dublin/' The two earliest cases of the 
 disease which came under my notice dated from Thursday and 
 Friday, December 5 and 6, 1889, respectively. The outbreak 
 was at its height in the first half of January, 1890 a month which 
 proved one of the sickliest ever experienced within living memory. 
 The whole " Epidemic Constitution " to use Sydenhara's 
 classical phrase was changed for the worse ; the power of 
 resisting disease was lessened ; and extreme languor and prostra - 
 
 1 Swan's Sydenham, p. 9. 1769. 
 
 2 Philosophical Transactions, vol. Ixxxvi., No. 11. 
 
ACUTE INFECTIVE DISEASES 411 
 
 tion passed over the population like a pandemic. Towards the 
 close of January the epidemic waned, but in the middle of 
 February there was a recrudescence of it. The duration of the 
 outbreak was practically eleven weeks. In Dublin, while the 
 mean temperature of the first two weeks of the epidemic period 
 was about equal to the average, a remarkable excess of tempera- 
 ture afterwards set in, lasting for at least five weeks, and cul- 
 minating in the second and third weeks of the new year, the mean 
 temperatures of which were no less than 7'5 and 7'9 F. respec- 
 tively above the average. Now, if any one fact has been 
 established in relation to the winter death-rate in Dublin, it is 
 that the deaths from all causes, and particulaily from diseases 
 of the respiratory organs, such as bronchitis and pneumonia, 
 vary in number inversely with the temperature. If the ther- 
 mometer is high in winter, the death-rate is moderate or low ; 
 if the thermometer is low, the death-rate is high. 
 
 For example, the mean temperature of the first six weeks 
 of 1881 was only 35-6 F., or 5-4 F. below the average. The 
 mean weekly number of deaths from all causes in that period 
 were 40' 1 above the average in a population of 350,000 ; and 
 the mean weekly number of deaths from diseases of the respira- 
 tory organs were 29 '1 above the average. On the other hand, in 
 1884 the mean temperature of the first six weeks was 45'1 F., 
 or 4 - l F. above the average. The mean weekly number of deaths 
 were 37 '2 below the average, while the mean weekly number of 
 deaths from respiratory diseases were 19 '2 below the average. 
 
 And now we come to the opening six weeks of 1890, when the 
 mean temperature shows an excess comparable with that of 1884 
 it was 44-2 F., or 3'2 F. above the average, and only 0*9 F. 
 below the value for 1884. Under these circumstances, a low rate 
 of mortality from all causes, and especially from respiratory 
 diseases, was to have been looked for. But how different were 
 the facts ! The mean weekly number of deaths were 60'8 above 
 the average 285'0 against 224'2. The mean deaths from 
 respiratory diseases were 40'0 above the average 98' 7, against 
 58 7. The mean weekly deaths from bronchitis were 65 '0, com- 
 pared with the average, 42 - 4 ; while the mean weekly deaths from 
 pneumonia were 23*3, compared with an average of 8'6. 
 
412 METEOROLOGY 
 
 It is of interest to observe that, whereas the deaths referred 
 to bronchitis were only 53 per cent, in excess of the average, those 
 referred to pneumonia were no less than 171 per cent, in excess. 
 
 The prime cause of this heightened death-rate at the beginning 
 of 1890 was manifestly the epidemic of influenza, which proved 
 more pernicious to the population of Dublin than the extreme 
 cold of January, 1881. 
 
 But the incidence of the outbreak in a winter month was 
 only accidental. Recent experiences confirm Hirsch's state- 
 ment that influenza has prevailed in all seasons of the year, in 
 all climates, independently of telluric conditions, and under 
 the most various states of the weather high and low tempera- 
 ture, steady and changeable weather, much or little atmospheric 
 humidity. There is not the slightest ground for assuming a 
 causal relation between the production of influenza and certain 
 states of the barometer. Hirsch gives a table embracing 125 
 epidemics or pandemics, which ran their course independently 
 of one another ; and of these outbreaks, 50 are shown to have 
 begun in winter (December to February), 35 in spring (March 
 to May), 16 in summer (June to August), 24 in autumn (Septem- 
 ber to November). Certainly winter comes out very decidedly 
 as the season of the year most favourable to the setting up of 
 the disease, but we must remember that an epidemic, once 
 developed, runs its course equally through all seasons of the 
 year, of which fact the pandemics of 1580, 1781-82, 1831, 1832-33, 
 1836-37, are striking illustrations. 
 
 We may conclude, then, that the prevalence of this strange 
 disease is absolutely independent of season and weather a 
 fact which distinguishes influenza from epidemic bronchial 
 catarrh. 1 " Et tempore frigidiori et calidiori, et flante tarn 
 Austro quam Borea, et pluvioso et sereno coelo, peragravit 
 hasce omnes Europse regiones, et omnia loca indiscriminatim." 2 
 As Morgagni says, " Tempestate frigida et sicca, coelo die noc- 
 tuque sereno." 
 
 1 Hirsch. Handbook of Geographical and Historical Pathology, vol. i., 
 p. 26. New Sydenham Society. 1883. 
 
 2 Petrus Salius Di versus, cited by Dunning (Med'ca J and Physica 1 Journal, 
 vol. x., p. 43), and quoted by Dr. Thomas Hancock in an excellent article on 
 Influenza in the second volume of the Cyclopaedia of Practical Medicine, 
 published in 1833. 
 
ACUTE INFECTIVE DISEASES 413 
 
 2. Cholera. 
 
 That cholera tends to prevail in the warmer months of the 
 year is sufficiently borne out by the history of the disease. In 
 Table X. on p. 414 are given the deaths from the disease 
 by months in some of the great epidemics of late years, and the 
 figures speak for themselves. In one case that of Limerick, 
 1849 we meet with an early spring epidemic, and in January 
 of the same year a large mortality from cholera prevailed in 
 England. But these are only exceptions. If from the totals 
 we omit the Paris outburst of April, 1832, in which city the 
 epidemic kindled into flame for the second time in July of 
 that year, we have an increasing series of deaths from February 
 to September, and a decreasing series from the last-named 
 month to December. 
 
 " Real epidemics of cholera," writes Professor Faye, 1 " in the 
 more rigorous season of winter have very seldom occurred, while 
 sporadic cases have very frequently shown themselves even in 
 winter. At Breslau a winter epidemic prevailed in 1848-49, 
 continuing from October till March, with the same fatality as 
 had characterised summer epidemics at the same place ; and at 
 St. Petersburg, as in several of the districts of Russia, cholera 
 has prevailed in winter, although to a far less degree than in 
 summer, so that the Russian physicians have often declared 
 that the disease is prevalent in the winter quarter. At Bergen, 
 in Norway, the epidemic of 1848-49 was also a winter epidemic. 
 It is therefore not altogether without reason that cholera has been 
 stated to observe no season ; but if we take into consideration 
 both the relative infrequency of its appearance in winter, and its 
 impaired virulence under intense degrees of cold, this assertion 
 as to the compatibility of the disease with a winter temperature 
 experiences a very important limitation. Perhaps the explana- 
 tion of the matter is not very remote. At Bergen, for example, 
 the winter is often rainy, and the air in proportion mild, so 
 that the freezing of the earth's surface to any depth does not 
 occur ; and the winter of 1848-49 was really of this kind. It 
 
 1 " Om Cholera-Epidemien i Norge i Aaret 1853 " ("On the Cholera 
 Epidemic in Norway in the Year 1853 "). 
 
414 
 
 METEOROLOGY 
 
 | 
 
 !> Oi GO ^ 
 <M O O CO 
 
 CO 
 
 S 
 
 
 
 l-H 
 
 OS 
 
 
 
 OS 
 
 1 
 
 GO 
 
 os 
 
 
 
 H 
 
 l-H l-H CO CO 
 l-H 
 
 CO 
 
 CO 
 l-H 
 
 co 
 
 l-H 
 
 CO 
 
 ^ 
 
 p-H 
 
 i-H 
 
 <N 
 
 
 'S6SI 
 
 o o o o 
 
 
 
 
 
 
 
 I 
 
 i 
 
 r-H 
 
 <M 
 
 co 
 
 Z6SI 
 
 o o o o 
 
 o 
 
 l-H 
 
 OS 
 
 GO 
 
 s 
 
 i 
 
 i 
 
 CO 
 
 CO 
 
 l-H 
 
 '9981 
 
 o o o o 
 
 o 
 
 o 
 
 o 
 
 CO 
 
 s 
 
 o 
 
 O 
 
 O 
 
 '858T 
 
 0000 
 
 
 
 
 
 CO 
 
 CO 
 l-H 
 
 1 
 
 o 
 
 CD 
 
 s 
 
 o 
 
 '0581 
 
 0000 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 O 
 tQ 
 
 t-- 
 
 CO 
 
 
 
 '8881 
 
 o o o o 
 
 o 
 
 
 
 
 
 
 
 
 
 G* 
 
 GO 
 
 CO 
 
 l-H 
 
 ( '9981 
 
 o o o o 
 
 
 
 * 
 
 CO 
 GO 
 
 CO 
 
 ! 
 
 GO 
 
 o 
 
 CO 
 
 11 
 
 '0581 
 
 
 
 
 o 
 
 
 
 CO 
 
 l-H 
 
 O 
 
 GO 
 
 S3 
 
 So 
 
 'nepaMg 
 
 
 
 
 
 <N 
 
 <N 
 
 GO 
 
 CO 
 
 
 fSSI 
 
 .. ,. r ^ 
 
 *. 
 
 I.--. 
 
 8 
 
 OS 
 
 CO" 
 
 1 
 
 GO 
 
 l-H 
 
 CO 
 
 4 '9981 
 
 O O i-i O 
 
 l-H 
 
 o 
 
 <N 
 
 21 
 
 o 
 
 1 
 
 CO 
 
 
 
 "6*8 T 
 
 o co i-i co 
 
 O^ "^f 
 
 ^ 
 
 l-H 
 
 o 
 
 o 
 
 l-H 
 
 O 
 
 O 
 
 o 
 
 'Sa 
 
 <M CO GO C<I 
 
 CO 
 
 g 
 
 5 
 
 r-4 
 
 CO 
 
 CO 
 
 i 
 
 s 
 
 
 
 o 
 
 .6fSt 
 
 s s 
 
 - * 5 S5 
 
 I-H" 
 
 s 
 
 1 
 
 co 
 
 GO 
 
 GO 
 
 Th 
 
 15 
 
 - 
 
 .. 
 
 '2881 
 
 'SLIBJ 
 
 gco 
 CO 
 
 1 1 
 
 <M 
 
 GO 
 
 GO 
 CO 
 GO 
 
 CO 
 
 s 
 
 1 
 
 1 
 
 <N 
 
 CO 
 
 +. 
 
 .. 
 
 '6*81 
 
 i oo ^~ 4 c^l ir** 
 
 CO 
 
 CO 
 
 o^ 
 
 
 
 g 
 
 s 
 
 OS 
 
 CO 
 
 CO 
 
 CO 
 
 CO 
 
 i-H 
 
 'ptrejSujj 
 
 T^ GO OS >"H 
 i 1 O r-H O 
 CO t- >0 * 
 
 | 
 
 CO 
 
 co 
 
 CO 
 
 CO 
 
 r-H 
 CO 
 
 GO 
 
 OS 
 
 o 
 
 i 
 
 
 
 
 p-H p-H 
 
 
 11 
 
 T* 
 
 CO 
 
 10 
 
 T* 
 
 
 
 Months. 
 
 > ^ -> 
 
 111 
 
 * & a < 
 
 1 
 
 
 1 
 
 1 
 
 1 
 
 September . . 
 
 October 
 
 November . . 
 
 1 
 
ACUTE INFECTIVE DISEASES 415 
 
 is well known also that cholera at St. Petersburg in winter-time 
 is almost exclusively confined to the unhealthy houses, situated 
 on the low and swampy banks of the Neva, belonging to an 
 indigent labouring population ; and, indeed, it is not strange 
 that low-lying and overcrowded cellars, beneath which the soil 
 has scarcely stiffened, with a favourable and confined oven- 
 temperature, should foster the contagion and occasion a constant, 
 though tardy, propagation of the disease. Whether conditions 
 of this kind held at Breslau I am unable to say ; but, in any case, 
 it is certain that violent epidemics during severe winter-frost 
 very rarely, if, indeed, ever, occur." 
 
 Professor Faye goes on to say that, while the epidemic (of 
 1853) was at its worst at Christiania, the atmosphere was steadily 
 warm, and the air, in addition, clear and very still. This con- 
 tinued for about three weeks, during which the daily numbers of 
 cases, which were then at the highest, scarcely varied. At this 
 point of time the middle of September the air was set in 
 motion by a strong and stormy north-west wind, and, remark- 
 ably enough, the number of cases fell, next day, to about one-half. 
 Similarly, at Bergen, during the epidemic of 1848-49, a strong and 
 cold north-easterly gale, supervening on a lengthened period of 
 milder temperature, caused a considerable fall in the number 
 of cholera cases. 
 
 In the epidemic of 1866 the acme of mortality from cholera 
 was reached in Dublin about the middle of October, the weather 
 of the preceding week having been continuously calm, cloudy, 
 foggy, damp, with a very high barometer, and a great deficiency 
 of ozone (the latter showing a mean value of only 10 per cent, 
 at the Ordnance Survey Office, Phoenix Park). 
 
 The decrease in the mortality was consequent on a freshening 
 breeze, and a change of wind from north-east to south-west, a 
 diminution of barometrical pressure, a moderate and continued 
 rainfall, a rise in ozone to 70 per cent., and a gradually falling 
 temperature. The coincidence of a high barometer with a great 
 development of cholera has often been remarked, but striking 
 exceptions are also on record. Keeping in view the fact that 
 heavy rain and a strong breeze are most valuable detergents 
 and disinfectants, it seems probable that the calm weather conse- 
 
416 METEOKOLOGY 
 
 quent on slight barometrical gradients, so common in anti- 
 cyclonic, or high-pressure systems, has more influence than the 
 mere height of the barometer itself. In December the epidemic 
 died out rapidly, and no death occurred later than the 29th of 
 that month, on which day, it is most interesting to note, the 
 intense frost of January, 1867, was ushered in by a fall of 
 temperature amounting to 15 F. in a few hours. 
 
 The meteorological conditions just alluded to, and the in- 
 fluence of season, are to be classed among the predisposing 
 causes of cholera. Its exciting cause is, of course, the introduc- 
 tion into the human system, and particularly the intestinal canal, 
 of the specific virus or contagium of the disease, the comma 
 bacillus of Koch. I thoroughly agree with the late Mr. Ernest 
 Hart when he says, with uncompromising dogmatism : " We 
 may lay aside all pedantry and mystery-talk of ' epidemic con- 
 stitution/ ' pandemic waves/ ' telluric influences/ ' cholera 
 blasts/ ' cholera clouds/ ' blue mists/ and the like terms of art 
 with which an amiable class of meteorologists have delighted 
 to cloak ignorance. Asiatic cholera is a filth disease, which is 
 carried by dirty people to dirty places. Cholera/' he adds, 
 " does not travel by air waves or blasts. We can drink cholera 
 and eat cholera, but we cannot ' catch ' cholera in the sense in 
 which we catch measles, scarlatina, or whooping-cough. Cholera 
 is carried by men in their clothing and their secretions along the 
 lines of human intercourse. Earlier epidemics of Asiatic cholera 
 took three years to reach us, by caravan and fitful travel, from 
 its Asian home. It comes now not as a pedestrian or a horse- 
 man, but by locomotive and fast steamboat." I believe that 
 cholera is taken precisely as enteric fever is taken that is, it 
 is most usually swallowed in water, less commonly in milk, or 
 it is eaten in solid food, or, exceptionally, it is inhaled and swal- 
 lowed with the saliva when a liquid medium containing its virus 
 has evaporated, leaving that virus to be air-borne for a short 
 distance. Further, I am satisfied that cholera, like cholera 
 nostras or cholerine, and enteric fever also, becomes much more 
 virulent when the subsoil temperature reaches the critical point 
 of 56 F. at 4 feet below the surface. As this occurs most readily 
 when the level of the subsoil water (German, Grundwasser) is 
 
ACUTE INFECTIVE DISEASES 417 
 
 low, the significance of Pettenkofer's theory is at once evident. 
 That veteran sanitarian says in one of his later papers i 1 The 
 fluctuations in the level of the subsoil water have a meaning 
 for setiology, only because they are traced back to those primary 
 influences by which air and water are made to share, in varying 
 proportion, the possession of the pores of an impregnated soil. 
 Beyond that they have no significance." 
 
 3. Diarrhosal Diseases. 
 
 Under this heading are now included in the third revision of "The 
 Nomenclature of Diseases " of the Koyal College of Physicians, 
 London (1906), and in the Reports of the Registrars-General of 
 the several divisions of the United Kingdom, " Dysentery," 
 " Infective Enteritis," " Epidemic Diarrhoea," and " Diarrhoea 
 due to Food." 
 
 These propositions may be laid down : 
 
 1. In summer and autumn the tendency to sickness and 
 
 death is chiefly connected with the digestive organs 
 diarrhoea, dysentery, and simple cholera or cholerine 
 (cholera nostras), being the affections which are especially 
 prevalent and fatal during these seasons. 
 
 2. In summer and autumn a rise of mean temperature above 
 
 the average increases the number of cases of, and the 
 mortality from, the diseases named. 
 
 3. On the other hand, a cool rainy summer and autumn con- 
 
 trols their prevalence and fatality. 
 
 4. Diarrhoeal diseases are observed to become epidemic when 
 
 the subsoil temperature at a depth of 4 feet below the 
 surface permanently reaches 56 F. This may therefore 
 be called the " critical temperature." Recent investiga- 
 tions would point to this relation as being a coincidence, 
 and not an setiological factor, as was supposed by Dr. 
 Ballard. 
 
 In Table XI., p. 418, facts are given, which support the first of 
 these propositions. Reference to the last two columns of the table 
 will convince the reader that yearly towards the end of July 
 
 * Zeitschrift fur Biologic, Heft vi., p. 527. 1870. Quoted by Hirsch, 
 Lac. cit., vol. i., p. 460. 
 
 27 
 
418 
 
 METEOROLOGY 
 
 TABLE XI., SHOWING THE AVEEAGE AND TOTAL DEATHS FROM DIAREHCEAL 
 DISEASES IN THE DUBLIN REGISTRATION DISTRICT IN EACH OF THIRTEEN 
 FOUR- WEEKLY PEEIODS IN THE THIRTY YEARS, 1872-1901, AND THE 
 PERCENTAGE OF THE SAME IN EACH OF THE SAID PERIODS. 
 
 
 
 
 1 
 
 
 
 Diarrhoeal Diseases. 
 
 
 
 | 
 
 
 
 
 
 
 
 5 -a 
 
 Four- 
 Week 
 Periods. 
 
 Corresponding Periods 
 in Calendar. 
 
 III 
 
 ||i 
 
 III 
 
 III III' 
 
 
 "^ 
 
 !s 
 
 bcH co 
 
 g H CO 
 
 S OrH 
 
 !? 
 
 gB* 
 
 il a 
 
 
 
 
 
 
 o " 
 
 
 
 
 < 
 
 H 
 
 
 H 
 
 IH 
 
 I. 
 
 Jan. 1 to Jan. 28 
 
 12-9 
 
 9-2 
 
 9-1 
 
 312 
 
 3-5 
 
 IT. 
 
 Jan. 29 , Feb. 25 
 
 10-9 
 
 7-7 
 
 8-9 
 
 275 
 
 3-1 
 
 III. 
 
 Feb. 26 , Mar. 25 
 
 10-2 
 
 8-5 
 
 8-2 
 
 269 
 
 3-0 
 
 IV. 
 V. 
 
 Mar. 26 , Apr. 22 
 Apr. 23 , May 20 
 
 12-9 
 9-9 
 
 6-7 
 7-4 
 
 5'8 
 
 277 
 231 
 
 3-1 
 2-6 
 
 VI. 
 
 May 21 , June 17 
 
 9-7 
 
 67 
 
 7-2 
 
 236 
 
 2-6 
 
 VII. 
 
 June 18 , July 15 
 
 10-7 
 
 10-4 
 
 21-8 
 
 429 
 
 4-8 
 
 VIII. 
 
 July 16 , Aug. 12 
 
 24-1 
 
 28-2 
 
 77-1 
 
 1,294 
 
 14-5 
 
 IX. 
 
 Aug. 13 , Sept. 9 
 
 56-1 
 
 67-1 
 
 97-5 
 
 2,207 
 
 24-6 
 
 X. 
 
 Sept. 10 , Oct. 7 
 
 54-9 
 
 66-6 
 
 68-2 
 
 1,897 
 
 21-2 
 
 XI. 
 
 Oct. 8 , Nov. 4 
 
 23-2 
 
 31-7 
 
 27-0 
 
 819 
 
 9-2 
 
 XII. 
 
 Nov. 5 , Dec. 2 
 
 13-6 
 
 14-2 
 
 12-4 
 
 402 
 
 4-5 
 
 XIII. 
 
 Dec. 3 , Dec. 30 
 
 9-6 
 
 10-6 
 
 9-2 
 
 294 
 
 3-3 
 
 Fifty- 
 two 
 Weeks 
 
 | January 1 to "I 
 f December 30 / 
 
 2,587 2,750 
 
 3,605 
 
 8,942 
 
 100-0 
 
 Fifty- 
 third 
 
 I .. ( 
 
 1873=2 1884=3\ 
 1879 = 3 1890= OJ 
 
 1896=4 
 
 12 
 
 
 Week 
 
 J 
 
 
 
 
 
 
 General Totals . 
 
 2,592 2,753 
 
 3,609 
 
 8,954 
 
 
 
 1 
 
 
 ! 
 
 or the beginning of August, on an average, diarrhoeal diseases, 
 and particularly cholerine, assume epidemic proportions in 
 singular obedience to a law of periodicity, and with all the sudden- 
 ness of an explosion. Of every 100 deaths from diarrhoea! 
 diseases taking place annually, only 2 -6 occur in the four weeks 
 ending June 17, only 4'8 in the four weeks ending July 15. Then 
 the percentage runs up to 14'5 in the next four weeks (ending 
 August 12), and to no less than 24 ( 6 in the period ending Septem- 
 ber 9. In the eight weeks ending October 7, 45-8 of every 
 100 deaths from diarrhoea! diseases take place. Similarly, in 
 the case of simple cholera, of 100 deaths occurring in the whole 
 
ACUTE INFECTIVE DISEASES 
 
 419 
 
 year, only 0-6 takes place in the four weeks ending May 20 ; 
 whereas in the period ending September 9, 26'9 take place, and 
 in that ending October 7, no less than -28-8 55*7 per cent, of 
 the annual mortality in eight weeks. 
 
 From the curves of mortality given by Dr. A. Buchan and 
 Sir Arthur Mitchell in their paper on " The Influence of Weather 
 on Mortality from Different Diseases and at Different Ages " 
 (Journal of the Scottish Meteorological Society, vol. iv., p. 187), 
 it would appear that the diarrhoeal and choleraic death-rates rise 
 in London to a yearly maximum about three weeks earlier than 
 in Dublin. 
 
 In support of the second and third propositions, we have only 
 to refer to the Reports of the Registrar-General for Ireland for 
 the years 1868 and 1887 (warm, dry years), and for the cold, wet 
 year 1879. The facts may best be thrown into a short tabular 
 
 statement as follows : 
 
 TABLE XII. 
 
 i 
 1868. 
 
 . 
 
 
 1887. 
 
 
 1879. 
 
 Quarter. fl ^ ji 
 
 
 
 erf, 
 
 1.2 
 
 
 
 = 
 
 e rf, 
 
 1.- S 
 
 ;l| || 
 
 I 
 
 II 
 
 11 
 & 
 
 S| 
 5 
 
 II | 
 
 F. 
 
 
 
 op 
 
 
 F. 
 
 
 
 L 44-7 
 
 39 
 
 
 
 41-9 
 
 31 
 
 2 39-3 
 
 39 
 
 
 
 II. 55'5 
 
 22 
 
 i 
 
 53-1 
 
 27 
 
 49-7 
 
 38 
 
 
 
 III. 60-7 
 
 289 
 
 11 
 
 59-3 
 
 331 
 
 11 56-4 
 
 54 1 
 
 IV. 45-3 
 
 77 
 
 
 
 43-3 
 
 71 
 
 1 
 
 43-8 
 
 54 
 
 1 
 
 51-6 
 
 427 
 
 12 
 
 49-4 
 
 460 
 
 14 
 
 47-3 
 
 185 
 
 2 
 
 In his Weekly Return of Births and Deaths in Dublin for 
 August 22, 1868, the Registrar-General for Ireland wrote : " The 
 number of deaths from diarrhoea registered during the week 
 amounted to forty-nine, showing an increase of twenty-three 
 on the number registered during the week preceding, and being 
 thirty-five more than the average deaths from this disease in the 
 corresponding week of the four previous years." In his return 
 for the corresponding week in 1879, the Registrar-General 
 observed : " Owing chiefly to the low mortality from diarrhoea, 
 the number of deaths from zymotic diseases is considerably 
 
 272 
 
420 METEOEOLOGY 
 
 under the average for the thirty-fourth week of the last ten years/' 
 As a matter of fact, only one death from diarrhoea was registered 
 in the whole Dublin Registration District in that week, and the 
 largest number of deaths from the disease registered in any week 
 during 1879 was eight in the week ending Saturday, September 13. 
 A more striking contrast can hardly be imagined than that be- 
 tween the epidemic prevalence of diarrhoea in the very warm 
 season of 1868 and its absence in the extremely cool summer of 
 1879. 
 
 The year 1868 may be cited as an example of an unusually 
 warm year. There was an almost complete absence of frost, and 
 during ten out of the twelve months the mean temperature 
 was above the average : the excess varying from O5 F. in 
 January to 3 '8 F. in March the warmest March within the 
 twenty years now under discussion namely, 1868-1887. October 
 and November were cold, the deficit of temperature amounting 
 to 2O and 1*1 respectively. Notwithstanding this, the mean 
 temperature of the whole year was 51 '5, compared with an 
 average of 49*7 (excess =1-8). A remarkable drought pre- 
 vailed from the last week in April to August 10, when a tropical 
 rainfall occurred. During this period of nearly three and a 
 half months only 2-797 inches of rain fell in the city. On 
 six occasions during the summer of this year the thermometer 
 rose to 80 in the shade in Dublin the highest readings of all 
 being 86 on July 15, and 85 on July 21. On August 1 the 
 maximum was 82, and even as late as September 6 the high 
 reading of 77 was noted. 
 
 In marked contrast to 1868, and as an instance of a cold year, 
 1879 stands out in bold relief. The annual mean temperature 
 was only 47-3 F. that is, 2-4 below the average (49-7). 
 Every month was colder than usual the deficit of mean tern-, 
 perature ranging from 6'3 in January, 3'1 in April, 3*2 
 in July, and 4-0 in December to T2 in November. Only 
 in October was the mean temperature, 49 '7, slightly in 
 excess of forty years' average, 49'5. Curiously enough, these 
 last-named months were relatively the coldest in the warm year 
 1868. There was a singular absence of summer heat in July 
 and August ; in each of these months the shade temperature 
 
ACUTE INFECTIVE DISEASES 421 
 
 exceeded 70 F. on one day only in Dublin, and on nine days in 
 July it did not reach 60 F. The low temperature was accom- 
 panied with to some extent depended upon a continuous 
 rather than a heavy rainfall. During the six months ending 
 September 30, rain fell on 125 out of 183 days that is to say, 
 on two out of every three days. The amount of cloud during 
 this cold, damp, sunless year was 7'5 per cent, over the average. 
 The cold weather, which persisted almost throughout 1879, set 
 in first on October 21, 1878. This period of low temperature 
 had probably not been paralleled for intensity and duration within 
 the nineteenth century. 
 
 In their classical paper already quoted, Dr. Buchan and Sir 
 Arthur Mitchell speak of " the close and direct relations which 
 the progress of mortality from these (diarrhceal) diseases bears 
 to temperature. This relation is seen in the startling suddenness 
 with which they shoot up during the hottest weeks of the year, 
 and the suddenness, equally startling, with which they fall on 
 the advent of colder weather." The authors point out that the 
 death-rate curves for diarrhoea and cholerine rise and fall about 
 a month earlier than do those for dysentery and epidemic cholera. 
 The annual phases of the former diseases are, in other words, about 
 a month earlier than those of the latter. For all four diseases, 
 the curves are reproduced in all their essential features from year 
 to year. In very hot summers the numbers of deaths are enor- 
 mously increased, and in cold summers, such as 1860, the deaths 
 from bowel complaints are correspondingly few. 
 
 The fourth proposition was first advanced by Dr. Edward 
 Ballard, in his elaborate Report to the Local Government Board 
 for England upon the causation of the annual mortality from 
 " Diarrhoea," which is observed principally in the summer season 
 of the year. 1 That a high atmospheric temperature conduces to 
 a high diarrhceal mortality, and a low atmospheric temperature 
 to a low diarrhceal mortality, Dr. Ballard admits. " It is," he 
 says, " an established fact which no one can dispute." But his 
 inquiry showed that the influence thus exerted is not a direct 
 influence, except in so far as it affects also infant mortality from 
 
 1 Supplement in Continuation of the Report of the Medical Officer for 1887. 
 Seventeenth Annual Report of the Local Government Board, 1887-88. 
 London : Eyre and Spottiswoode, 1889. Quarto. P. 1 et seq. 
 
422 METEOKOLOGY 
 
 all causes. Rainfall, again, exerts an influence on diarrhoea, but 
 apparently not equally in all periods of the diarrhoeal season. 
 The diarrhoeal mortality is greater in dry, less in wet seasons. 
 But here again the influence exerted is not direct (e.g., by a 
 washing of the atmosphere, so to speak), but indirect namely, 
 by its effect mainly in preventing the rise and (probably to a 
 less extent) in hastening the fall of the temperature of the earth. 
 
 Wind and comparative calm affect the diarrhoeal mortality. 
 Other things being equal, calm in the diarrhoeal season promotes 
 it, and high winds tend to lessen it. 
 
 But soil and the temperature of the soil are far more important 
 predisposing causes of diarrhoeal diseases. Their prevalence 
 and fatality is low in dwelling-houses built on a foundation of 
 solid rocL Deep and wide and frequent fissuring of the rock 
 in a town, or superficial alternations of rock with looser material, 
 modify this immunity. On the other hand, a loose soil, more 
 or less freely permeable by water and by air, is a soil on which 
 diarrhoeal mortality is apt to be high. Of all natural soils, sand 
 and surface mould to a considerable depth are " the most diar- 
 rhoeal." Gravel varies in its relation to diarrhoeal mortality 
 according to its texture : fine, sand-like gravel predisposes to 
 diarrhoeal prevalence ; coarse, rock-like gravel is more wholesome. 
 Clay soils do not in themselves favour diarrhoea. A soil which 
 is a mixture of clay, sand, and stones (commonly called a " marl "), 
 is apparently favourable or unfavourable to diarrhoeal mortality 
 in proportion as it is loose and permeable on the one hand, or 
 plastic on the other. The presence of much organic matter in 
 the soil renders it distinctly more conducive to high diarrhoeal 
 mortality than it otherwise would be. Hence, dwellings built 
 upon made ground, the refuse of towns, or the site of market- 
 gardens, are unwholesome. And, of course, a sewage-soaked 
 subsoil is most unwholesome and dangerous. Excessive wetness 
 and complete dryness of the subsoil appear to be alike unfavourable 
 to diarrhoea. Habitual dampness, which is not sufficient to pre- 
 clude the free admission of air to the interstices of the subsoil, 
 favours diarrhoeal prevalence. 
 
 Dr. Ballard, however, considers that the Temperature of the 
 Soil is a far more effective element in the causation than any of 
 
ACUTE INFECTIVE DISEASES 423 
 
 the meteorological factors just mentioned. He constructed for 
 London and many other towns in the kingdom a large number 
 of charts, showing week by week for many years the earth 
 temperature at a depth of 1 foot from the surface and at a depth 
 of 4 feet also, each chart showing in addition the diarrhoeal 
 mortality of the corresponding weeks. The general result shown 
 by these charts is as follows : 
 
 a. The summer rise of diarrhoeal mortality does not commence 
 until the mean temperature recorded by the 4-foot earth ther- 
 mometer has attained somewhere about 56 F., no matter what 
 may have been the temperature- previously attained by the 
 atmosphere or recorded by the 1-foot earth thermometer. 
 
 j3. The maximal diarrhoeal mortality of the year is usually 
 observed in the week in which the temperature recorded by the 
 4-foot earth thermometer attains its mean weekly maximum. 
 
 7. The decline of the diarrhoeal mortality coincides with the 
 decline of the temperature recorded by the 4-foot earth ther- 
 mometer, which temperature declines very much more slowly 
 than the atmospheric temperature, or than that recorded by the 
 1-foot earth thermometer. The epidemic mortality may in 
 consequence continue (although declining) long after the last- 
 mentioned temperatures have fallen greatly, and may extend 
 some way into the fourth quarter of the year. 
 
 8. The atmospheric temperature and that of the more super- 
 ficial layers of the soil exert little, if any, influence on the pre- 
 valence of diarrhoea until the temperature recorded by the 
 4-foot earth thermometer has risen to 56 F. Then their in- 
 fluence is apparent, but it is a subsidiary one, notwithstanding 
 the statement made by Dr. August Hirsch that the summer 
 diarrhoea of children makes its appearance as an epidemic only 
 in those districts whose average temperature for the day in the 
 warm season is rather more than 15 C. (59 F.). 1 
 
 It is interesting to notice that in an excellent article on 
 " Cholera Infantum," which appeared in the Medical Annual 
 for 1893, Dr. E. Meinert, of Dresden, entirely adopts Dr. Ballard's 
 views as to the meteorological aetiology of this disease, while he 
 
 1 Handbook of Geographical anl Historical Pathology, vol. iii., p. 379. 
 New Sydenham Society. 1886. 
 
424 METEOROLOGY 
 
 also expresses his entire concurrence with Dr. Ballard's state- 
 ment that density of buildings, whether dwelling-houses or other, 
 upon area quite apart from density of population upon area 
 promotes diarrhceal mortality to a remarkable degree, particu- 
 larly because crowding together of buildings of whatever sort 
 restricts and offers an impediment to the free circulation of air. 
 Dr. Edward W. Hope, the Medical Officer of Health for Liverpool, 
 has investigated the influence of the mode of feeding of young 
 infants, upon the prevalence and fatality of diarrhoea, and arrives 
 at the following conclusions i 1 
 
 1. Infants fed solely from the breast are remarkably exempt 
 
 from fatal diarrhoea. 
 
 2. Infants fed in whatever way with artificial food, to the 
 
 exclusion of breast milk, are those who suffer most heavily 
 from fatal diarrhoea. 
 
 3. Children fed partially at the breast, and partially with 
 
 other kinds of food, suffer to a considerable extent from 
 fatal diarrhoea, but very much less than those who are 
 brought up altogether by hand. 
 
 4. As regards the use of " the bottle/' it is decidedly 
 
 more dangerous than artificial feeding without the 
 bottle. 
 
 In relation to this part of the subject, Dr. Ballard's observa- 
 tions go to show that the circumstances of food-keeping, of its 
 exposure to telluric emanations (e.g., in underground cellars), 
 or to emanations from accumulations of domestic filth, etc. 
 (e.g., when kept in pantries, etc., to which such emanations have 
 more or less free access), tends to render it liable to produce 
 diarrhoea, especially where the storing place of food is dark, and 
 not exposed to currents of air. 
 
 Dr. Ballard believes that a working hypothesis, or provisional 
 explanation, that would best accord with the whole evidence in 
 his possession bearing on the production of epidemic diarrhoea, 
 may be stated as follows : 
 
 1. The essential cause of diarrhoea resides ordinarily in the 
 superficial layers of the earth, where it is intimately associated 
 
 1 Cf. Dr. Ballard's Report, p. 6. 
 
ACUTE INFECTIVE DISEASES 425 
 
 with the life processes of some micro-organism not yet detected, 
 captured, or isolated. 
 
 2. The vital manifestations of such organism are dependent, 
 among other things, perhaps principally, upon conditions of 
 season and on the presence of dead organic matter which is its 
 pabulum. 
 
 3. On occasion, such micro-organism is capable of getting 
 abroad from its primary habitat, the earth, and having become 
 air-borne obtains opportunity for fastening on non-living organic 
 material, and of using such organic material both as nidus and 
 as pabulum in undergoing various phases of its life-history. 
 
 4. In food, inside of as well as outside of the human body, 
 such micro-organism finds, especially at certain seasons, nidus 
 and pabulum convenient for its development, multiplication, 
 or evolution. 
 
 5. From food, as also from the contained organic matter of 
 particular soils, such micro-organism can manufacture, by the 
 chemical changes wrought therein through certain of its life 
 processes, a substance which is a virulent cJiemical poison. 
 
 6. This chemical substance is, in the human body, the material 
 cause of epidemic diarrhoea. 
 
 To the foregoing we have only to add Dr. Meinert's words : 
 " The poison, or a combination of poisons, appears to work upon 
 the medulla oblongata, for there lies the centre for intestinal 
 secretion, vomitings, convulsions, respiratory and vaso-motor 
 phenomena/* 
 
 Duval and Bassett, working at the Mount Wilson Sanatorium, 
 discovered, in the dejecta of children suffering from summer 
 diarrhoea, a bacillus apparently identical with the organism 
 shown by Shiga (in 1898) to be the cause of epidemic dysentery 
 in Japan. 1 In 1903 the Rockefeller Institute research showed 
 that this organism was present in a large number of cases of 
 so-called " summer diarrhoea." " The laboratory studies," 
 writes Professor Osier, " of Martini and Lentz, Flexner, His, 
 Parke, and others, indicate that there is a group of closely allied 
 forms of bacilli differing slightly from the original Shiga bacillus 
 
 1 Osier, The Principles and Practice of Medicine, p. 243. Seventh edition, 
 1909. 
 
426 METEOROLOGY 
 
 in their action on certain sugars and in agglutinating properties. 
 The type of organisms most frequently associated with the 
 diarrhoeas of children belongs to the so-called ' acid type/ and, 
 unlike the Shiga cultures, ferments mannite with acid produc- 
 tion." * 
 
 In the British Medical Journal for 1906 and 1907, 2 Mr. Harry de 
 Eiemer Morgan, Ernest Hart Memorial Scholar, reported on 
 the " Bacteriology of the Summer Diarrhoea of Infants/' The 
 result of his investigations was the discovery of several different 
 organisms instead of a single specific form. But he did discover 
 a bacillus, not hitherto described so far as he could ascertain, 
 which appeared to him entitled, in the absence of further know- 
 ledge, to be regarded as a factor, perhaps the most important 
 factor, in the causation of the disease. 
 
 Messrs. Orr, Williams, Murray, and Rundle, working upon 
 material from the Liverpool City Hospital at Fazakerley, have 
 found yet another microbe in a case of epidemic diarrhoea, and 
 they denote it as the " Bacillus F." 3 Their summary of con- 
 clusions is as follows : " The Bacillus F. was obtained from a 
 case of epidemic diarrhoea. By its cultural reactions it is readily 
 differentiated from the Bacillus typhosus. The absence of indol 
 formation separates it from Morgan's No. 1 Bacillus. The presence 
 of well-marked motility is sufficient to distinguish it from 
 Morgan's Bacilli Nos. 3 and 4, and from the dysentery bacilli. 
 The agglutination reactions show that there is a relationship 
 between this organism and the Bacillus typhosus and the para- 
 typhoid Bacillus B. The Bacillus F. is able to produce diarrhoea 
 in animals, and can be recovered from their stools. We believe, 
 therefore, that it may be an agent in the production of epidemic 
 diarrhoea. During the summer of 1908 we have further investi- 
 gated this disease, and find that we have obtained a positive 
 agglutination reaction, varying from 1 in 25 to 1 in 100 in nearly 
 half the cases tested against the Bacillus F. The specificity of 
 this reaction we cannot yet affirm, but we hope later to publish 
 a further report on the material obtained during the past 
 summer." 
 
 1 Op. cif., p. 505. 
 
 2 British Medical Journal, p. 16. July 6, 1907. 
 
 3 Abstract in The Journal of Clinical Research, May, 1909. 
 
ACUTE INFECTIVE DISEASES 427 
 
 Of late years much attention has been paid to the part which 
 the common house-fly plays in the transmission of infectious 
 material to articles of food, and thereby in the causation of 
 tuberculosis, epidemic diarrhoea, enteric fever, and cholera. 
 Among the British authorities on this subject are Professor 
 E. Klein j 1 Dr. James T. C. Nash, 2 M.O.H. for Co. Norfolk ; 
 Dr. V. J. Glover, 3 of Liverpool ; Mr. Eobert Newstead, Liverpool 
 School of Tropical Medicine ; and Mr. Ernest E. Austen. 4 The 
 last-named authority, in an article on " Blood-Sucking and Other 
 Flies/' published in Part II. of the second volume of Allbutt's 
 System of Medicine, 5 refers the reader to a memoir by Dr. G. H. F. 
 Nuttall, " On the Role of Insects, Arachnids, and Myriapods 
 as Carriers in the Spread of Bacterial and Parasitic Diseases of 
 Man and Animals/* which was published in the eighth volume 
 of the Johns Hopkins Hospital Reports. He also states that 
 information on this subject will be found in L. 0. Howard's 
 "Contribution to the Study of the Insect Fauna -of Human 
 Excrement (with especial Reference to the Spread of Typhoid 
 Fever by Flies)/' in the second volume for 1900 of the Proceedings 
 of the Washington Academy of Science. 
 
 Mr. Austen holds that certain flies, while incapable of sucking 
 blood, are nevertheless, in some instances, important agents in 
 the dissemination of such diseases as cholera and enteric fever. 
 In these cases the flies act as mechanical carriers of bacilli or 
 other infective matter (" fomites "), and the effect is produced 
 chiefly by the contamination of food. The species concerned 
 are those most closely associated with man, and of these the 
 common house-fly (Musca domestica Linn.) is the most important. 
 
 In December, 1907, Dr. Daniel D. Jackson, S.B., reported to 
 the Committee on Pollution of the Merchants' Association of 
 New York, on the pollution of the harbour of that great city as 
 a menace to health by the dissemination of intestinal diseases 
 through the agency of the common house-fly. 
 
 The investigations upon which this important sanitary report 
 
 1 British Medical Journal, October 17, 1908. 
 
 2 Trans. Epidem. Society, London, 1903. Lancet. 1904. 
 
 3 Lancet, September 5, 1908. 
 
 4 Journal of the Royal Army Medical Corps, June, 1904. 
 
 5 London : Macmillan and Co., 1907. P. 185. 
 
428 METEOROLOGY 
 
 is based, were carried out during the summer months of 1907, 
 under the immediate direction of Dr. Jackson, assisted by a 
 number of observers. They proved that the water front of 
 Greater New York was much contaminated by human excreta. 
 It was found that at many points sewer outfalls had not been 
 carried below the low-water mark, in consequence of which the 
 solid matters from the sewers were exposed on the shores. It 
 was also shown that deposits of this nature may, and did actually, 
 become a source of typhoid fever and certain intestinal diseases 
 through the agency of flies. It is this last point which lends a 
 special value to the New York Harbour Pollution Report. 
 
 The large amount of work which was carried on during the 
 summer of 1907 was divided into two parts first, a thorough 
 inspection of all sources of contamination throughout the entire 
 water front of the city ; second, a study of the prevalence and 
 distribution of flies by fly-cages distributed in all parts of the 
 city in order to demonstrate what proportion of the intestinal 
 diseases in the city were contracted by means of these insects. 
 
 Examinations made of flies at the beginning of the season 
 directly after hibernation showed that many of them carried 
 only a few bacteria and moulds, and little or no faecal matter. 
 Like examinations made later in the year showed the presence 
 of numerous animal and vegetable parasites, faecal matter in 
 abundance, large numbers and many kinds of germs. In some 
 cases an individual fly carried as many as 100,000 faecal bacteria 
 on its legs, in its mouth, and on its body. Over 98 per 
 cent, of the flies found in dwelling-houses belong to one 
 species namely, that known as the common house-fly (Musca 
 domestica). The activity of these flies extends over a very few 
 weeks of the summer, after which most of them perish by cold or 
 are killed by moulds and other parasites. The few which hiber- 
 nate and come out in the spring are observed, in the climate of 
 New York, about the middle of June. 
 
 These flies begin to lay eggs soon after their emergence, pref- 
 erably in horse dung, but also in human excreta and in decaying 
 animal and vegetable matter. The eggs hatch in from six to 
 eight hours. The larvae are white pointed maggots. They grow 
 rapidly, cast their skins twice, and, under favourable conditions, 
 reach full size in four or five days. The outer skin then becomes 
 
ACUTE INFECTIVE DISEASES 429 
 
 hard, swells up, and turns dark brown in colour. Within it the 
 true pupa is found. In about five days the adult fly issues forth 
 from a round hole in the anterior end of the brown pupa-case. 
 The total time required for a single generation is about ten days, 
 and the number of generations during the summer season, stated 
 by some authorities to be as many as twelve, is probably about 
 one-half that number in New York. The number of eggs laid 
 by each female fly during the season is about 1,000. 
 
 A table is given by Dr. Jackson showing the total deaths by 
 weeks from diarrhoeal diseases in New York during the summer 
 of 1907, together with the general prevalence of flies in that city. 
 This table proves that the time of appreciable prevalence of flies 
 in 1907 was the period from July 1 to September 30. By far 
 the greatest number of flies were caught in cages in the weeks 
 ending July 27 and August 3. It will also be seen from the table 
 that deaths from intestinal complaints rose above normal at 
 the same time at which flies become prevalent, culminated at 
 the same high point, and fell off with slight " lag " at the time 
 of the gradual falling off of the prevalence of the insects. A 
 secondary rise of flies in September is reflected in a fresh rise in 
 the number of deaths from intestinal diseases. Dr. Jackson very 
 properly points out that the comparative immunity from diar- 
 rhoea of breast-fed babies and the frequent occurrence of diarrhceal 
 diseases among artificially-fed babies point strongly to the food 
 as a medium of transmission. Much of this actual infection is, 
 in his opinion, undoubtedly due to flies. He adds : " There is 
 crying need for better sanitation on our dairy farms." Of one 
 individual fly, captured on South Street, and found on examina- 
 tion to be carrying in his mouth and on his legs over one hundred 
 thousand (100,000) faecal bacteria, Dr. Jackson says : " He had 
 been walking over human excreta on the water front, and was on 
 his way to the nearest milk-pitcher/' 
 
 This remarkable Report is illustrated by a number of maps, 
 diagrams, photographs, and tables. 
 
 Dr. Glover 1 agrees with those writers who have remarked 
 
 the prevalence of summer diarrhoea with a high ground-air 
 
 temperature, particularly when occasional showers alternate 
 
 with prolonged high atmospheric temperature. These conditions 
 
 1 Lancet, p. 717. September 5, 1908. 
 
430 METEOROLOGY 
 
 favour the growth of micro-organisms. Dr. Glover's theory is 
 that, whilst the female fly walks over the nauseous material 
 in which she lays her eggs and during her act of oviposition, 
 multitudes of micro-organisms, and pre-eminent among them the 
 unknown one of infantile summer diarrhoea, adhere to the moist 
 sucker-like terminations of the hairs on the pulvilli of the fly's 
 legs. The house-fly loves warmth and also food delicacies, and 
 so it haunts our homes, frequenting the sugar-bowl, the jam-pot, 
 and the milk- jug. In this way it carries the contagion. Babies 
 also often sleep with their mouths open and smeared with milky 
 saliva. To such a mouth the fly seems to be attracted, and 
 often will venture slightly inside the lips, so infecting the child, 
 even if breast-fed, with the poison of summer diarrhoea. 
 
 The Right Hon. John Burns, President of the Local Govern- 
 ment Board for England, in 1908 authorised an investigation 
 into the possible carriage of infection by flies, under the general 
 supervision of Dr. S. Monckton Copeman, F.R.S., in co-operation 
 with Dr. G. H. F. Nuttall, F.R.S., Professor of Biology in the 
 University of Cambridge. A Preliminary Report 1 of this in- 
 vestigation contains the following : (1) How to distinguish the 
 more important species of flies found in houses namely, the 
 common house-fly, Musca domestica ; the lesser house-fly, Homa- 
 lomyia canicularis ; the blue-bottle, Calliphora erythrocephala ; 
 and the Muscina stabulans, by Mr. E. E. Austen, of the British 
 Museum. (2) Mr. Austen's notes on flies examined during 
 1908. (3) Mr. Jepson's report on the breeding of the common 
 house-fly during the winter months. Mr. Jepson shows by a 
 series of experiments that flies, provided the temperature is suit- 
 able, may go on breeding in winter. He gives the duration of 
 the various stages at an average temperature of 70 F., as follows : 
 The eggs hatch in twenty-four hours, the larval period occupies 
 eleven days, the pupal stage on an average ten days. He con- 
 cludes that much might be done to reduce the number of, or even 
 to exterminate, flies, if isolated colonies living in certain warm 
 places through the winter months were carefully sought out and 
 destroyed. 
 
 1 Reports to the Local Government Board on Public Health and Medical 
 Subjects. New Series. No. 5. Preliminary Reports on Flies as Carriers 
 of Infection. London : Wyman and Sons, Limited. Edinburgh : Oliver 
 and Boyd, Tweedale Court. Dublin : E. Ponsonby. 1909. 
 
CHAPTER XXVIII 
 
 ACUTE INFECTIVE DISEASES (continued) 
 4. Enteric Fever. 
 
 IT is fitting that this disease should be taken next in order after 
 cholera and the diarrhoeal diseases, to which it presents so many 
 points of analogy. 
 
 Season. Enteric fever is most prevalent in autumn and early 
 winter, hence the names by which it is often described in America, 
 " Autumnal " or " Fall Fever " (Austin Flint). The exciting 
 cause of the disease seems to be called into action only, as 
 Murchison says, " by the protracted heat of summer and autumn, 
 while it required the protracted cold of winter and spring to impair 
 its activity or to destroy it/' An examination of the Returns 
 of the Registrar-General for Ireland shows that enteric fever 
 exhibits, as the summer rolls by, a decided tendency to increase 
 in Dublin at an earlier period than typhus. This is, no doubt, 
 partly due to the fact that the secondary phenomena of enteric 
 fever are generally developed in connection with the digestive 
 system, acute and infective diseases of which system increase 
 towards autumn. 
 
 The diagram on p. 432 (Fig. 95) is reproduced from the Annual 
 Summary of the Registrar-General for England for 1890. 
 
 Temperature and Moisture. Hot, dry, calm summers increase 
 the prevalence of enteric fever, which is less frequent in cold, 
 wet, stormy seasons. Warm, damp weather, however, predisposes 
 to the disease. Floods occurring in badly drained localities may 
 impregnate sources of drinking-water with the germs of enteric 
 fever, and so lead to its outbreak. 
 
 Soil and Underground Water. 1 Professor von Pettenkofer 
 
 1 An excellent resume of various papers on this subject in the Zeitschrift 
 fiir Biologie will be found in the Ugeskrift for Lceger, Copenhagen, January 30, 
 1869. A translation by my father, Dr. W. D. Moore, appeared in the 
 Dublin Journal of Medical Science, vol. xlvii., p. 497, May, 1869. 
 
432 
 
 METEOROLOGY 
 
 I 
 
 and Professor Buhl, of Munich, have shown that when the subsoil 
 water in that city (as measured by the depth of water in the surface 
 wells) is falling, the number of cases of enteric fever increases ; 
 * w when the water level is rising, 
 
 5 | the number of cases diminishes, 
 
 g ^ ^ < ^ Liebermeister and Buchanan 
 
 suppose that these observations 
 simply illustrate the mode in 
 which the disease is communi- 
 cated by means of drinking- 
 water. When the subsoil water 
 ; is low, any noxious matters in 
 it accumulate and acquire a 
 greater virulence. 
 
 In the case of an outbreak of 
 enteric fever at Terling, Essex, 
 in December, 1867, the late Sir 
 Richard Thome Thorne, then 
 an Inspector, and afterwards 
 Medical Officer, of the Local 
 Government Board of England, 
 found that the disease had broken 
 out with great severity precisely 
 when the wells were high. 1 
 
 Two or three years after the 
 introduction of the Vartry Water 
 Supply into Dublin, in 1868, a 
 serious local outbreak of enteric 
 fever took place in Trinity Col- 
 lege, Dublin. It was confined to 
 the resident water-drinkers in 
 the College. An inquiry was 
 instituted into the cause of the 
 outbreak, the Rev. Dr. Haughton, 
 F.R.S., Fellow of Trinity College, Dr. Apjohn, F.R.S., then 
 Professor of Chemistry in the University of Dublin, and Mr. 
 Dowling, then Professor of Engineering in the University, being 
 appointed to act as Commissioners by the Rev. H. Lloyd, D.D., 
 1 Tenth Report of the Medical Officer of the Privy Council, p. 51. 1868. 
 
 4 
 
ACUTE INFECTIVE DISEASES 
 
 433 
 
 at the time Provost of Trinity College. It was found that, owing 
 to high tides in the River Liffey, and the accumulation of water 
 in the subsoil, in consequence partly of the disuse of the pumps 
 after the introduction of the Vartry water, and partly of the leakage 
 of the Vartry water itself from defective house-drains, the foul 
 subsoil water had overflowed into and contaminated the well 
 within the College precincts from which the drinking-water in 
 use in the College was drawn. Ever since that time the level of 
 the subsoil water in Trinity College has been kept low by steam 
 pumping, at a cost of about 300 per annum, with the result that 
 within the past forty years no indigenous outbreak of enteric 
 fever has occurred amongst the residents in the College. 
 
 TABLE XIII. SHOWING THE TOTAL NUMBER OF DEATHS FROM ENTERIC 
 FEVER IN THE DUBLIN REGISTRATION DISTRICT IN EACH OF THIRTEEN 
 FOUR-WEEKLY PERIODS IN THE THIRTY YEARS 1872-1901'; THE 
 AVERAGE YEARLY NUMBER OF DEATHS FROM THIS FEVER IN THE 
 DECENNIAL PERIODS 1872-81, 1882-91, and 1892-1901 RESPECTIVELY; 
 AND THE PERCENTAGE OF THE TOTAL MORTALITY FROM THE SAME 
 FEVER IN EACH OF THE SAID PERIODS. 
 
 
 
 J 
 
 1 
 
 1 
 
 is 
 
 II 
 
 Four- 
 Week 
 Periods. 
 
 Corresponding Periods 
 in Calendar. 
 
 ift 
 
 ||| 
 
 ||1 
 
 US 
 
 "s c . 
 
 f! 
 
 
 
 r 
 
 jhr 
 
 r~ 
 
 I* 8 
 
 5 " 
 
 I. 
 
 Jan. 1 to Jan. 28 
 
 16-1 
 
 12-1 
 
 15-1 
 
 433 
 
 8-9 
 
 II. 
 
 Jan. 29 Feb. 25 
 
 16-0 
 
 10-8 
 
 11-2 
 
 380 
 
 7-9 
 
 III. 
 
 Feb. 26 Mar. 25 
 
 14-6 
 
 13-2 
 
 11-7 
 
 395 
 
 8-2 
 
 IV. 
 
 Mar. 26 Ap. 22 
 
 12-3 
 
 9-8 
 
 8-9 
 
 310 
 
 6-4 
 
 V. 
 
 Ap. 23 May 20 
 
 14-3 
 
 9-6 
 
 8-2 
 
 321 
 
 6'6 
 
 VI. 
 
 May 21 June 17 
 
 8-7 - 
 
 8-1 
 
 7-1 
 
 239 
 
 4-9 
 
 VII. 
 
 June 18 July 15 
 
 10-6 
 
 8-6 
 
 6-9 
 
 261 
 
 5-4 
 
 VIII. 
 
 July 16 Aug. 12 
 
 9-2 
 
 7-9 
 
 6-4 
 
 235 
 
 4-9 
 
 IX. 
 
 Aug. 13 , Sept. 9 
 
 10-7 
 
 10-1 
 
 11-2 
 
 320 
 
 6-6 
 
 X. 
 
 Sept. 10 Oct. 7 
 
 14-3 
 
 10-9 
 
 18-1 
 
 433 
 
 8-9 
 
 XL 
 
 Oct. 8 Nov. 4 
 
 12-7 
 
 19-7 
 
 20-3 
 
 527 
 
 10-9 
 
 XII. 
 
 Nov. 5 Dec. 2 
 
 15-9 
 
 17-1 
 
 17-0 
 
 500 
 
 10-3 
 
 XIIL 1 
 
 Dec. 3 Dec. 30 
 
 13-8 
 
 20-2 
 
 14-9 
 
 489 
 
 10-1 
 
 
 Totals . . 
 
 1,692 
 
 1,581 
 
 1,570 
 
 4,843 
 
 100-0 
 
 1 The thirteenth period included five weeks in 1873 (no deaths), 1879 
 (2 deaths), 1884 (7 deaths), 1890 (7 deaths), and 1896 (3 deaths). These 
 19 deaths raise the periodic averages from 13*6 to 13-8 in 1872-81, from 
 18-8 to 20-2 in 1882-91, and from 14-6 to 14-9 in 1892-1901 ; and the yearly 
 averages from 169-0 to 169-2 in 1872-81, from 156'7 to 158'1 in 1882-91, and 
 
 from 156-7 to 157 -0 in 1892-1901. 
 
 28 
 
434 METEOROLOGY 
 
 The preceding table supplies information as to the seasonal 
 Mortality from enteric fever in Dublin in the thirty years ending 
 1901. The facts are drawn from the Keports of the Registrar- 
 General for Ireland. In this table the year is divided into 
 thirteen periods of four weeks each. In each of the years 1873, 
 1879, 1884, 1890, and 1896, fifty-three weeks are included, in order 
 to bring the Registrar-General's statistics into agreement with the 
 Calendar. In these five additional weeks nineteen deaths from 
 enteric fever were registered, thus raising by so many the number 
 of deaths recorded in the thirteenth four-week period in the 
 table. 
 
 The table shows that, allowance being made for a three weeks' 
 illness before death and registration occur, enteric fever increases 
 in prevalence and fatality towards the end of July that is, with 
 a rise of the subsoil temperature at 4 feet to and above the critical 
 point of 56 F. Its epidemic character becomes pronounced 
 in September, and continues until the close of February, after 
 which the disease becomes less frequent and deadly, reaching its 
 spring minimum at the beginning of May. From this time to 
 the end of June is also the period of its annual minimum, while 
 its annual maximum takes place about the middle of November. 
 These results agree remarkably with the curve for typhoid fever 
 for all ages and both sexes given by Buchan and Mitchell. 1 
 This is a well-marked curve (they, say) resembling the curve for 
 scarlatina in showing the maximal death-rate in October and 
 November, but differing from it in the duration and phases of 
 the minimal period. Scarlatina falls below its average in the 
 beginning of January, typhoid fever not till the last week of 
 February ; scarlet fever has its absolute minimum period from 
 the middle of March to the middle of May, typhoid fever from 
 the middle of May to the end of June ; scarlet fever begins steadily 
 to rise in the second week of May, typhoid not till the beginning 
 of July, when the heat of summer has fairly set in. 
 
 1 " The Influence of Weather on Mortality," Journal of the Scottish 
 Meteorological Society, vol. iv., p. 197. 
 
ACUTE INFECTIVE DISEASES 435 
 
 5. Typhus Fever. 
 
 Typhus is essentially a disease of winter and spring that 
 is, of the colder seasons of the year. Among the predisposing 
 causes of this fever, season and atmospheric temperature are 
 commonly included. 
 
 Season. During twenty-three years January and March were 
 the months in which the number of admissions of typhus patients 
 to the London Fever Hospital reached a maximum the mini- 
 mum falling in September, August, and July. This distribution 
 was from time to time disturbed by an epidemic, outbreaks of 
 typhus commencing and advancing irrespective of season. An 
 examination of the Registrar-General's (Ireland) returns of 
 deaths from typhus in Dublin, undertaken many years ago, led 
 me to the conclusion that the death-rate from typhus attains 
 its maximum in January and its minimum in September. The 
 reason for this is not far to seek. Typhus is often intimately 
 related to overcrowding, and affections of the respiratory organs 
 are among its most frequent complications. Hence we should 
 expect to meet with it especially in the colder seasons of the year. 
 Murchison points out that typhus does not always become more 
 prevalent with the commencement of cold weather, nor does it 
 decline immediately on the advent of summer. He correctly 
 infers from this that the increase of typhus in winter and spring 
 is due not so much to the direct effect of cold as to the continued 
 overcrowding and defective ventilation of the dwellings of the poor 
 in cold weather. 
 
 The accompanying table gives the facts relating to the deaths 
 from typhus in Dublin during the thirty years 1872-1901, in- 
 clusive (see Table XIV., p. 436). 
 
 Apart from our present inquiry, one gratifying circumstance 
 stands prominently out from the figures in the following table, 
 and that is the fact that typhus fever is fast dying out in Dublin. 
 The number of deaths from the disease fell nearly 50 per cent. 
 (49-1) to 507 from 996 in the second decennium discussed in 
 the table, and as much as 85'2 per cent. from 507 to 75 in 
 the third decennium. 
 
 An analysis of the table proves that tne mortality from typhus 
 
 282 
 
436 
 
 METEOKOLOGY 
 
 reaches a minimum in the ninth and tenth periods August 13 
 to October 7 ; while the minimal death-rate from enteric fever 
 has already occurred in the eighth period July 16 to August 12 ; 
 this fever exhibiting, as the summer rolls by, a decided tendency 
 to increase at an earlier period than typhus. The highest per- 
 centage death-rates from typhus are met with in the seasons 
 of winter, spring, and early summer 10'4 per cent, of the fatal 
 cases being registered in the second period (January 29 to 
 February 25), and 1OO per cent, in the fifth period (April 23 to 
 May 20). 
 
 TABLE XIV. SHOWING THE TOTAL NUMBER OF DEATHS FROM TYPHUS 
 FEVER IN THE DUBLIN REGISTRATIO i DISTRICT IN EACH OF THIRTEEN 
 FOUR- WEEKLY PERIODS IN THE THIRTY YEARS 1872-1901 ; THE 
 AVERAGE YEARLY NUMBER OF DEATHS FROM THIS FEVER IN THE 
 DECENNIAL PERIODS 1872-81, 1882-91, AND 1892-1901 RESPECTIVELY ; 
 AND THE PERCENTAGE OF THE TOTAL MORTALITY FROM THE SAME 
 FEVER IN EACH OF THE SAID PERIODS. 
 
 
 
 1 
 
 \ 
 
 1 
 
 J 
 
 0> 
 
 
 
 
 
 a ,H- 
 
 
 & r ' 
 
 I 5 
 
 "S . 
 
 Four- 
 Week 
 Periods. 
 
 Corresponding Periods 
 in Calendar. 
 
 3-Coo 
 
 *iS 
 gig 
 
 5 C: 
 ^ 03 2 
 
 of ! 
 
 s-ao 
 
 fc-sS 
 
 II 5 
 
 A >< 
 
 ^Qo 
 
 a x 
 |J 
 
 -2&S 
 
 fl * Q 
 
 
 
 & <=- 
 > 
 
 < 
 
 \ 
 
 > 
 
 <! 
 
 g" 
 
 2 
 15 
 
 I. 
 
 Jan. 1 to Jan. 28 
 
 9-0 
 
 4-3 
 
 1-0 
 
 143 
 
 9-1 
 
 IL 
 
 Jan. 29 Feb. 25 
 
 9-8 
 
 5'9 
 
 1-0 
 
 167 
 
 10-6 
 
 III. 
 
 Feb. 26 Mar. 25 
 
 8-1 
 
 5-4 
 
 0-6 
 
 141 
 
 8-9 
 
 IV. 
 
 Mar. 26 Ap. 22 
 
 7.4 
 
 5-8 
 
 0-7 
 
 139 
 
 8-8 
 
 V. 
 VI. 
 
 Ap. 23 Mav 20 
 May 21 June 17 
 
 10-2 
 9-6 
 
 4-8 
 3-1 
 
 0-7 
 0-4 
 
 157 
 131 
 
 9-9 
 8-3 
 
 VII. 
 
 June 18 July 15 
 
 7-1 
 
 4-1 
 
 0-3 
 
 115 
 
 7-3 
 
 VIII. 
 
 July 16 Aug. 12 
 
 7-5 
 
 2'4 
 
 0-4 
 
 103 
 
 6-5 
 
 IX. 
 
 Aug. 13 Sept. 9 
 
 4-6 
 
 3'4 
 
 0-6 
 
 86 
 
 5-5 
 
 X. 
 
 Sept. 10 Oct. 7 
 
 5-1 
 
 2-0 
 
 0-5 
 
 76 
 
 4-8 
 
 XI. 
 
 Oct. 8 Nov. 4 
 
 6-0 
 
 3-4 
 
 0-5 
 
 99 
 
 6-3 
 
 XII. 
 
 Nov. 5 Dec. 2 
 
 6-9 
 
 2-9 
 
 0-3 
 
 101 
 
 6-4 
 
 XIII. 1 
 
 Dec. 3 Dec. 30 
 
 8-3 
 
 3-2 
 
 0-5 
 
 120 
 
 7'6 
 
 
 Totals . . 
 
 996 
 
 507 
 
 75 
 
 1,578 
 
 100-0 
 
 According to Buchan and Mitchell, 2 the curve for typhus is 
 above the average from January to the beginning of May, and, 
 
 1 The thirteenth. period included five weeks in 1873 (4 deaths), 1879 (3 
 deaths), 1884 (no deaths), 1890 (no deaths), and 1896 (no deaths). These 
 7 deaths raise the periodic average from 7*6 to 8 '3, and the yearly average 
 from 98-9 to 99'6, in 1872-1881. 
 
 2 Loc. cit., p. 197. 
 
ACUTE INFECTIVE DISEASES 437 
 
 with the exception of the hot season of July and the beginning 
 of August, it is below the average from the middle of May to the 
 end of September. It seems probable that the curve has two 
 maxima the larger in the early months of the year, and the 
 smaller in the height of summer. Buchan and Mitchell's 
 typhus curve is based on only six years' returns of mortality 
 1869-1874. 
 
 Temperature and Moisture in the Atmosphere do not seem to 
 have any marked predisposing influence on typhus, notwith- 
 standing the opinion advanced in 1866 1 by Dr. T. W. Grimshaw, 
 afterwards Registrar-General for Ireland, that a warm moist state 
 of the atmosphere seemed to favour an increase of typhus, 
 whereas dryness with cold had a contrary influence. Murchison 
 was unable to trace any such connection, but points out that 
 exposure to cold and wet, if long continued, depresses the nervous 
 system, and so favours the onset of typhus. 
 
 6. Smallpox. 
 
 Turning now to the principal eruptive fevers, we find that, 
 although the incidence of smallpox is apparently independent 
 of climate, yet the season of the year has a marked influence upon 
 the prevalence of the disease. Nearly all writers are agreed 
 that, while outbreaks of smallpox may occur at all seasons, 
 they mostly begin towards the end of autumn and in the early 
 spring, or in the cold season. In a word, smallpox is essentially 
 a disease of winter and spring. In the British Islands and 
 Western Europe generally, for example, the monthly number of 
 cases is high from November onwards ; but from May a rapid 
 decline in the prevalence of the disease takes place, the least 
 number of cases being observed in September. 
 
 The diagram (Fig. 96, p. 438) is copied from the Annual 
 Summary of Births and Deaths of the Registrar-General for 
 England for 1890. It shows the weekly departure from the 
 average weekly number of deaths from smallpox (17) in London 
 in the fifty years 1841-1890, inclusive. 
 
 In this diagram the thick horizontal line represents the mean 
 
 1 " On Atmospheric Conditions influencing the Prevalence of Typhus," 
 Dublin Quarterly Journal of Medical Science, May, 1866. 
 
438 
 
 METEOROLOGY 
 
 weekly mortality from smallpox in London, on the supposition 
 that the mortality is spread equally over the fifty-two weeks of 
 the year the fifty-third week, when it occurs, being ignored. 
 The curved line represents the amount per cent, by which the 
 
 * 1 average mortality in each week 
 jjj differs from this mean. When 
 
 $ s> S *i & a S | ^ ie Percentage for any week is 
 
 above the mean, the amount of 
 the percentage excess is marked 
 above the horizontal line repre- 
 
 1 senting the mean ; and when the 
 
 2 percentage is below the mean, it 
 ^ is marked below the line. 
 
 It must be remembered that 
 
 * the data on which the curve is 
 - J formed are the deaths registered 
 
 <|:::::::::: :::::j:::: ^ in eacn week , not the deaths 
 
 g which occurred in the week, and 
 
 BJ that the registration is usually a 
 
 p| few days after the death ; and, 
 
 g secondly, that the curve relates 
 
 . to deaths that is, the final ter- 
 
 s urination of the attack of illness, 
 
 and not its commencement. So 
 
 H that, in estimating the effect of 
 
 * season in generating smallpox, 
 allowance must be made for the 
 g average duration of this disease 
 
 when fatal that is, eleven or 
 twelve days. It is, moreover, 
 possible that the curve of mor- 
 1-5-1 tality may, for another reason, 
 
 * $ 1 not accurately represent the 
 * curve of prevalence. For it may 
 
 be that an attack of smallpox is more likely to terminate fatally 
 if it occurs at one season for example, midwinter than if it 
 occurs at another, such as midsummer. 
 
 The diagram shows that at the beginning of February and in 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 n\ 
 
 
 
 
 
 
 
 
 
 
 5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 **v^ 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -h 
 
 
 
 
 
 
 
 
 
 
 -J> 
 
 
 
 
 
 
 
 j_ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 
 
 
 
 
 
 
 
 
 J ' 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 j 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 * 
 
 
 
 
 
 
 
 
 "7 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 < 
 
 
 
 
 
 
 
 
 
 --- 
 
 
 5^- 
 
 
 
 
 
 
 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^y 
 
 
 
 
 
 
 
 
 
 
 > 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 "> 
 
 
 
 
 
 
 
 
 
 < 
 
 
 
 
 
 
 
 
 
 
 
 > 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 ii <> 
 
 
 
 i 
 
 
 
 c 
 
 s 
 
 <: 
 
 i 
 i .. 
 
ACUTE INFECTIVE DISEASES 439 
 
 the second half of May the weekly number of deaths was 35 per 
 cent, in excess of the average weekly number of seventeen deaths 
 represented by the mean line, whereas at the end of September 
 there was a deficit of 43 per cent, in the weekly number of deaths 
 as compared with the same average weekly number over the 
 whole year. 
 
 In Dublin, during the autumn of 1871, the prevalence of, and 
 mortality from, smallpox increased with a fall of mean tempera- 
 ture below 50 F., and the greatest severity of the epidemic was 
 experienced in the first half of the following April, shortly after 
 a period of intense cold for the time of year. With the rise of 
 mean temperature to between 55 and 60 F. in the middle of 
 June, the epidemic declined rapidly. Abundant rainfalls seemed 
 to be followed by remissions in the severity of the epidemic, and 
 the converse was also true. 1 
 
 Buchan and Mitchell say that the curve for smallpox is one 
 of the simplest of the curves, showing that the mortality from 
 the disease is above the average from Christmas till the end of 
 June, the maximum falling in the last week of May, and the 
 minimum in the last week of September. 
 
 From statistics as to the prevalence of the disease in Sweden, 
 by months, in the years 1862-1869 inclusive, 2 it appears that the 
 greatest prevalence of smallpox is observed in May, the cases 
 in that month being 13 ! 7 per cent, of the total cases occurring 
 in the year ; while the least prevalence is observed in September, 
 when only 3'9 per cent, of all the cases in the year occur. From 
 November the monthly number of cases is high, but from May 
 a rapid decline in the prevalence of the disease takes place. 
 
 When due allowance has been made for difference of climate, 
 these results agree very closely with the observations which have 
 been recorded in this country on the relation of smallpox to season. 
 Dr. Edward Ballard, 3 writing of the epidemic of 1871, observed : 
 
 1 Manual of Public Health for Ireland, 1875, p. 298. Dublin : Fannin and 
 Co. See also Buchan and Mitchell's Paper in the Journal of the Scottish 
 Meteorological Society, 1874. 
 
 2 These statistics were compiled from exhaustive annual reports by the 
 late Dr. Wistrand, as to the morbidity of Sweden, and are the direct fruit 
 of an admirable system of disease-registration, which has been in operation 
 for many years in Sweden, and also in the other Scandinavian countries. 
 
 3 Medical Times and Ga-.ette, March 11, 1871. 
 
440 METEOROLOGY 
 
 " There is some reason for believing that the variations of 
 the epidemic (of smallpox) from week to week are influenced 
 to a certain extent by atmospheric conditions, and more especially 
 by variation in temperature." 
 
 He then quoted a series of remarkable coincidences between 
 the fluctuations of mean temperature and those of the small- 
 pox mortality in London during the winter of 1870-71. In the 
 number of the Medical Times and Gazette for May 13, 1871, he 
 wrote : 
 
 " The epidemic has now lasted a good six months. It may be 
 regarded as assuming a distinctly epidemic form in November, 
 shortly after the mean temperature of the air had fallen decidedly 
 below 50 F. In the progress of the seasons we have now arrived 
 at a time when this mean temperature is again reached. The 
 mean temperature of the last three weeks, as recorded at Green- 
 wich, has been 50, 50'7, and 49'7 F. It is customary about 
 the second week in May for some check in the consecutive weekly 
 rises of temperature to take place, but after this, in the ordinary 
 or average progress of events, the steady rise towards the summer 
 temperature may be expected to set in, and with it there is, at 
 least, a hope that the epidemic will begin to fade." 
 
 A week later the same writer said : 
 
 " The sudden fall of deaths in London from smallpox which 
 occurred last week namely, from 288 to 232 occurring about 
 three weeks after the mean temperature of 50 F. was reached, 
 appears to be confirmatory of the favourable hopes we expressed 
 last week, that the epidemic had, for this season, arrived at its 
 climax." 
 
 And so it had, for, although the decline was occasionally 
 interrupted, the virulence of the epidemic was broken in May, 
 in accurate fulfilment of the anticipations which had been 
 grounded on a consideration of the influence of temperature on 
 its progress. 
 
 7. Measles. 
 
 Seasonal Prevalence. Although, like smallpox, apparently 
 independent of climate for it is met with alike amidst Arctic 
 snows, in temperate latitudes, and under the tropical sun 
 measles prevails especially in the spring and autumnal quarters 
 
ACUTE INFECTIVE DISEASES 
 
 441 
 
 H 
 
 I 
 
 S$S S 
 
 * 
 
 V ? 
 
 of the year. An analysis of the weekly returns of deaths from 
 measles in the Dublin Registration District, published by the 
 Registrar-General for Ireland, long since led me to the con- 
 clusion that a mean temperature * 
 above 58*6 F. was not favour- 
 able to the spread of this disease, 
 and that a mean temperature 
 below 42 '0 F. was equally in- 
 imical to its prevalence. 1 These 
 results are in strict accord with 
 those arrived at by Dr. Edward 
 Ballard, who says 2 that the only 
 condition concerned in the arrest 
 of the spread of measles in 
 summer is the rise of the tem- 
 perature of the air above a mean 
 of 60 F., while towards winter 
 a fall below 42 F. also distinctly 
 tends to check the disease. 
 
 The accompanying diagram 
 (Fig. 97), copied from the Annual 
 Summary of Births, Deaths, and 
 Causes of Death in London and 
 other great towns for 1890, by 
 the Registrar-General for Eng- 
 land, is based upon the weekly 
 returns of deaths from measles 
 in London for the fifty years 
 1841-1890, inclusive. In it the 
 mean line represents an average 
 weekly number of thirty-four 
 deaths from the disease under 
 
 
 discussion, and the weekly curve 
 shows a double maximum and a double minimum, the larger maxi- 
 mum falling in November, December, and January, with an extreme 
 excess of 50 per cent, in the fourth week of December, and the 
 
 1 Manual of Public Health for Ireland, pp. 300, 301. 1875. 
 
 2 Eleventh Report of the Medical Officer of the Privy Council, No 3, 
 pp. 54-62. 1868. 
 
442 METEOROLOGY 
 
 smaller in May and June, with an extreme excess of 25 per cent. 
 in the first week of June. The larger minimum falls in August, 
 September, and October, extreme deficit being 45 per cent, below 
 the average in the last week of September, and the smaller 
 minimum in February and March extreme deficit, 30 per cent, 
 below average in the third week of February. 
 
 According to Buchan and Mitchell, who examined the London 
 death-rates for the thirty years beginning with 1845 and ending 
 with 1874, the measles curve is remarkable in showing a double 
 maximum and minimum during the year, the larger maximum 
 occurring in November, December, and January, and the smaller 
 in May and June ; the larger minimum in August, September, 
 and October, and the smaller in February and March. The most 
 rapid fluctuation takes place in the fall observed from Christmas 
 to the middle of February, the weekly deaths falling from 50 per 
 cent, in excess of the average to 30 per cent, below it that is, 
 from fifty-one to twenty-four, the average weekly number of 
 deaths throughout the year being thirty-four. This curve is one 
 of the steadiest from year to year, both the December and the 
 June maxima being well marked in nearly every one of the 
 thirty years analysed by Buchan and Mitchell, and the yearly 
 minima also being well marked. 
 
 In Table XV. is contained an analysis of the deaths from measles 
 registered in the Dublin Registration District during the thirty 
 years ending 1901, with the corresponding figures for scarlet 
 fever. The two annual maxima and minima for measles are 
 shown in the last column, but it will be observed that, as com- 
 pared with London, the Dublin spring minimum is feebly marked, 
 while the incidence of the autumnal minimum is later. Similarly, 
 the summer maximum falls later in Dublin than it does in 
 London. 
 
 It is instructive to compare the figures for measles with those 
 for scarlet fever. It will be seen that these diseases are correla- 
 tive, measles being very much in evidence when scarlet fever is 
 infrequent, and the latter disease attaining its autumnal maxi- 
 mum when the prevalence of measles only is beginning to increase. 
 
ACUTE INFECTIVE DISEASES 
 
 443 
 
 fr 
 
 4 
 
 8. Scarlatina or Scarlet Fever. 
 
 Climatic Influences do not play a prominent part in determining 
 the geographical distribution of this disease, for although the 
 tropical and subtropical ^ 
 regions of Asia and Africa jjj 
 
 have so far almost entirely > 5 
 
 Ji O >5 5 ?> tt *? 
 
 escaped scarlet fever, yet 
 it has often prevailed epi- 
 demically in the tropical 
 countries of South America ; 
 and, on the other hand, in 
 certain cold or temperate 
 climates scarlet fever is 
 among the rarest of 
 diseases. 
 
 There is, however, evi- 
 dence that season does in- 
 fluence its prevalence. 
 " Scarlatina" observes the 
 Registrar-General of Eng- 
 land, 1 "discovers a uniform, 
 well-marked tendency to 
 increase in the last six 
 months, and attain its 
 maximum in the December 
 quarter, the earlier half of 
 the following year witness- 
 ing a decrease." In Dublin, 
 also, the disease is almost 
 invariably most prevalent 
 
 and fatal in the fourth ^ *---.* 
 quarter of the year (see " 
 
 Table XV., p. 444). I < 4 
 
 From an analysis of the * 
 
 weekly death-rate from scarlatina in Dublin, it would seem that this 
 fever shows a tendency to increase when the mean temperature 
 
 1 Twenty -eighth Annual Report of Births, Deaths, and Marriages, p. 38. 
 
444 
 
 METEOROLOGY 
 
 SdS 
 
 win 
 
 111 
 
 III 
 
 PP 
 
 ipS 
 
 J =rS3 
 g g | 
 
 s 
 
 INI 
 
 \ 
 
 <, -2 H 
 
 M < W 
 
 eS! 
 
 G TH 
 N EA 
 
 5 
 
 s 
 
 eJ 
 
 pga 
 
 18P 
 
 
 COOCOrH 
 (NC<l<N(M 
 
 
 
 wj 
 
 &l 
 
 2 $ 
 SyS.S 
 
 rig 
 !?J 
 
 o>* 
 ""I 
 
 1-al 
 
 1 1 
 
ACUTE INFECTIVE DISEASES 445 
 
 rises much above 50 F., while a fall of mean temperature below 
 this point in autumn checks the further rise of the mortality. 1 In 
 this city scarlet fever is most fatal in the forty-sixth week of the 
 year (middle of November) and least fatal in the twenty-fourth week 
 (middle of June). Dr. Edward Ballard long ago drew inferences 
 which confirm these results. 2 The Annual Summary of Births, 
 Deaths, and Causes of Deaths, of the Eegistrar-General of England, 
 for 1890, is illustrated by the diagram on p. 443, showing the 
 weekly mortality curve for scarlet fever in London on an average 
 of thirty years (1861-1890). The curve consists of a single wave, 
 which rises to its crest (60 per cent, above the mean line, which 
 represents an average weekly number of forty-four deaths) in 
 October and November, while the trough extends from February 
 to August. It is a suggestive fact that the corresponding curves 
 for diphtheria and enteric fever are also single-wave curves, 
 closely resembling each other and the curve of scarlatina, in rising 
 to a crest in October and November, and showing a trough from 
 February to August. 
 
 From their analysis of the deaths from scarlet fever registered 
 in London in the thirty years 1845-1874, Buchan and Mitchell 
 concluded that this disease has its maximum from the beginning 
 of September to the end of the year, and its minimum from 
 February to July. The period of the highest death-rate is from 
 the beginning of October to the end of November, being nearly 
 60 per cent, above the average ; and the lowest in March, April, 
 and May, when it is about 33 per cent, below its average. In 
 each of the thirty years the deaths increased at the time of mean 
 maximum, and in all except four of the years the increase was 
 considerable. During ten of the years a high death-rate was 
 continued on into the year immediately following, but in every 
 year the deaths became fewer, and diminished steadily, if not 
 rapidly. 
 
 1 Manual of Public Health for Ireland, pp. 303, 304. 1875. 
 
 2 Eleventh Report of the Medical Officer of the Privy Council, No. 3, 
 pp. 54-62. 1808. 
 
CHAPTER XXIX 
 
 THE SEASONAL PREVALENCE OF PNEUMONIC OR LUNG 
 FEVER ! 
 
 IN April, 1875, the late Dr. T. W. Grimshaw, C.B., afterwards 
 Registrar- General for Ireland, and I, read before the Medical 
 Society of the King and Queen's College of Physicians a paper 
 on what we ventured to call " Pythogenic Pneumonia/' This 
 paper, which was published in the number of the Dublin Journal 
 of Medical Science for May, 1875, 2 was based upon observations 
 of pneumonia in Steevens' and Cork Street Hospitals, Dublin, 
 during the summer of 1874 when an epidemic of the disease 
 prevailed in the Irish capital as well as upon an analysis of the 
 statistics of death from bronchitis and pneumonia registered 
 in Dublin during nine years ending with 1873. In the same 
 communication the meteorological and epidemic conditions of 
 1874 were discussed, and our researches seemed to warrant us 
 in drawing the following conclusions : 
 
 1. That the bibliography of pneumonia indicates the existence 
 of a form of the disease which arises under miasmatic influences, 
 and is contagious. 
 
 2. That this view is supported by the relation which exists 
 between this form of pneumonia and certain zymotic affections 
 notably enteric fever and cholera and by the resemblance 
 between it and epizootic pleuro-pneumonia. 
 
 3. That its aetiology justifies us in regarding the disease as a 
 zymotic affection, and in naming it " pythogenic pneumonia." 
 
 4. That pythogenic pneumonia presents peculiar clinical 
 
 1 Reprinted, by permission, from the Transactions of the Ninth Session of 
 the International Medical Congress, vol. v., p. 45. Washington, B.C., U.S. A . 
 1887. 
 
 2 Vol. lix., No. 41, p. 399. Third Series. 
 
 446 
 
PREVALENCE OF PNEUMONIC OR LUNG FEVER 447 
 
 features which enable us to distinguish it from ordinary pneu- 
 monia. 
 
 5. That much of the pneumonia which prevailed in Dublin 
 during 1874 was of this pythogenic character. 
 
 6. That whereas ordinary pneumonia is specially prevalent 
 during a continuance of cold, dry weather, with high winds and 
 extreme variations in temperature, pythogenic pneumonia 
 reaches its maximum during tolerably warm weather, accom- 
 panied with a dry air, deficient rainfall, hot sun, and rapid 
 evaporation. 
 
 The years which have elapsed since the publication of this 
 paper on " Pythogenic Pneumonia " have been fruitful in the 
 literature of the subject to an unprecedented degree. Among 
 the many monographs on pneumonia which have of late appeared, 
 perhaps the most valuable are that by the late Dr. August Hirsch, 
 Professor of Medicine in the University of Berlin, on the Geo- 
 graphical and Historical Pathology of the Disease, 1 and that by 
 the late Dr. C. Friedlander, of Berlin, on the " Micrococci of 
 Pneumonia." 2 
 
 Hirsch, after pointing out that pneumonia, even in its narrowest 
 acceptation of fibrinous or so-called croupous pneumonia, is an 
 anatomical term that includes several inflammatory processes 
 differing from one another in their aetiology, goes on to observe 
 that the prevalence of the malady depends very decidedly upon 
 certain influences of season and weather. He gives an elaborate 
 table of percentages of pneumonic prevalence in the several 
 months at a large number of places in Europe and America. 
 According to this table, the largest number of cases falls in the 
 months from February to May, the smallest number in the period 
 from July to September. Taking the average for all the places 
 mentioned in the table, it appears that 34*7 per cent, of the 
 patients were attacked in spring (March to May inclusive) ; 
 29 -0 per cent, in winter (December to February) ; 18'3 per cent, in 
 autumn (September to November) ; and 18'0 per cent, in summer 
 
 1 Handbook of Geographical and Historical Pathology, vol. iii. Translated 
 from the second German edition, by Charles Creighton, M.D. London : 
 The New Sydenham Society. 1886. 
 
 2 Fortschritte der Medicin. Band 1, Heft 22. November 22, 1883. 
 Translated for the New Sydenham Society by Edgar Thurston. 1886. 
 
448 METEOROLOGY 
 
 (June to August). The combined percentage for winter and 
 spring is 63'7 ; that for summer and autumn is 36'3. If the 
 number of cases in summer be taken as 1, then autumn has I '02, 
 winter 1-6, and spring 1-9. Nearly all the recorded epidemics 
 of pneumonia have occurred in winter and spring. From the 
 foregoing considerations, Hirsch confidently concludes that the 
 origin of the malady is dependent on weather influences proper 
 to winter and spring, and more particularly on sudden changes 
 of temperature and considerable fluctuations in the proportion of 
 moisture in the air. He holds that any exceptionally large 
 number of cases of " inflammation of the lungs " at the other 
 seasons, more especially in summer, has coincided with the prev- 
 alence of the same meteorological conditions phenomenally at 
 that season. 
 
 " But that conclusion/ 5 he goes on to say, " is still further 
 borne out by the fact that in those northern regions (Russia, 
 Sweden, Denmark, Germany, England, the North of France, 
 and the Northern States of the American Union) where the most 
 sudden and severe changes of temperature fall in spring, the 
 largest number of cases is met with in spring also ; while in the 
 warmer and sub-tropical countries (Italy, islands of the Mediter- 
 ranean, Spain and Portugal, Greece, Algiers, Southern States of 
 the Union, Chili, and Peru), which are subject to those meteoro- 
 logical influences, for the most part, in winter, it is winter that 
 represents the proper season of pneumonia. And that applies 
 not merely to sporadic cases, but, in part at least, to epidemic 
 outbreaks of the malady as well. One other fact deserves to be 
 noticed here namely, that those tracts of country, especially 
 in the tropics, which are highly favoured in their climate or in 
 the steadiness of the temperature from day to day (Egypt, many 
 parts of India, including Bengal and the plain of Burmah, Cali- 
 fornia, etc.), are subject to pneumonia to a comparatively slight 
 extent." 
 
 In the paper on " Pythogenic Pneumonia," by Dr. Grimshaw 
 and myself, will be found a table, compiled from the returns of 
 the Registrar-General for Ireland, which shows the number of 
 deaths from bronchitis and pneumonia registered in the Dublin 
 Registration District in each quarter of the nine years 1865-1873, 
 
PREVALENCE OF PNEUMONIC OR LUNG FEVER 449 
 
 inclusive. According to that table, of every 100 deaths from 
 bronchitis, 44 on the average occurred in the first quarter of the 
 year, 22 in the second, only 10 in the third, and 24 in the fourth 
 quarter. Thus, the mortality from bronchitis was twice as 
 great in the first as it was in the second quarter, and more than 
 four times greater in the first than in the third quarter. 
 
 Very different were the facts as to pneumonia : of every 100 
 deaths from this disease, 32 on the average occurred in the 
 first quarter, 27 in the second, 16 in the third, and 25 in the fourth 
 quarter. The mortality from pneumonia was only one-fifth 
 greater in the first than in the second quarter, and only twice 
 as great in the first as in the third quarter. The extreme winter 
 fatality of bronchitis and its low summer fatality were equally 
 wanting in the case of pneumonia. 
 
 A careful analysis of the weekly returns of the Registrars- 
 General of England and Ireland for ten years ending with 1885, 
 and of the same returns for the year 1886, brings out a similar 
 remarkable contrast between bronchitis and pneumonia, as to 
 the time of year when these diseases are respectively most pre- 
 valent and fatal in London and Dublin. 
 
 Table XVI. contains the figures relating to pneumonia, and 
 Table XVII. those relating to bronchitis. Each table sets forth the 
 weekly average number of deaths in London and in Dublin from 
 pneumonia and bronchitis respectively, in the ten years 1876-1885, 
 as well as the actual weekly number of deaths from these diseases 
 in the year 1886. 
 
 In Tables XVIII. and XIX. these numerical results are thrown 
 into curves. 
 
 It will be observed that the statistics for London and for 
 Dublin agree to a remarkable extent. In both cities bronchitis 
 falls to a very low ebb in the third, or summer, quarter of the year 
 (July to September inclusive), when only 12 per cent, of the deaths 
 annually caused by this disease take place in Dublin, and only 
 11 per cent, in London. In the last, or fourth, quarter (October 
 to December inclusive), the percentage of deaths from bron- 
 chitis rises to 27 in Dublin and to 30 in London. The 
 maximal mortality occurs in the first quarter (January to March 
 inclusive), when it is 38 per cent, in both London and Dublin. 
 
 29 
 
450 
 
 METEOROLOGY 
 
 ll 
 
 co 10 co cs o o > 1 10 i i I-H <N t- 
 
 Ot-GOOGOO:OiC5<N.-i-H<N 
 
 COIOO' iCOTt*(MOCOCO'-HiO' I 
 l>l>GOOOi-H' i<M'-iO' i- i i i 
 
 ^ CD CO (N ^H 10 00 ^ Th 00 
 
 5! 
 
 . . I t- <N CO N OS O5 Oi -* O5 CO ^H CD 00 * 
 
 O" 
 
 IB 
 
 03 Q 
 
 || 
 < 
 
 i-HOOii i 
 
 CD t^ 
 
 1 
 
 I 
 
 
 R 
 
 j-< . 
 O CO 
 
 a do 
 
 
 R- 
 
 il 
 
 '! 
 
PREVALENCE OF PNEUMONIC OR LUNG FEVER 451 
 
 292 
 
452 
 
 METEOROLOGY 
 
 In the second, or spring, quarter (April to June inclusive), the 
 deaths from bronchitis declined to 23 per cent, in Dublin, and to 
 21 per cent in London. 
 
 The mortality from " pneumonic fever " is very differently 
 distributed throughout the year. In the summer quarter more 
 than 14 per cent, of the deaths yearly referable to this disease 
 
 Table- XVIII.<5 'housing f/lf average WffHy ntitntcrofDealhs from 
 Bronchitis, and from, Pnfomoma, in. the Decade 1876-1885 in 
 the DuAUsi Re.gi<stra'ti0n District 
 
 Bron 
 
 Pnum 
 
 are recorded in Dublin, and more than 15 per cent, in London. 
 In the first quarter the figures are : Dublin, 31 per cent. ; London, 
 31 per cent. ; in the second quarter they are : Dublin, 30 per cent. ; 
 London, 26 per cent. ; in the fourth quarter they are : Dublin, 
 24 per cent. ; London, 28 per cent. 
 
 From these numerical results it therefore appears that the 
 
PREVALENCE OF PNEUMONIC OR LUNG FEVER 453 
 
 fatality and (indirectly) the prevalence of pneumonic fever from 
 season to season do not correspond with the seasonal prevalence 
 and fatality of bronchitis. The latter disease be it of primary 
 or secondary origin increases and kills in direct relation to the 
 setting in of cold weather, with excessive relative humidity and 
 increased and frequent precipitation in the form of rain, snow 
 
 Table XI ~ Shelving fAc arercuqc^u'fe/cly numhcrof 
 Deaths from Bronc}u2is. an< from Pneumonia, 
 in. Hie Decade 1876- !86S '-ui London- 
 
 Fit-si Quarter 
 
 Second Quarter 
 
 \ 
 
 'f/urd Quarter 
 
 Fonrtf* Quarter 
 
 or sleet, and hail. It subsides in prevalence and fatality with the 
 advance of spring and the advent of summer. 
 
 Pneumonic fever, on the other hand, increases less quickly 
 in winter, and remains more prevalent and fatal in spring and 
 summer, than bronchitis ; its maximal incidence coincides with 
 the season of dry, harsh winds and hot sunshine in spring, when 
 
454 METEOKOLOGY 
 
 also the relative humidity is low, precipitation is scanty, while 
 the diurnal range of temperature is extreme. 
 
 A closer study of Tables XVIII. and XIX. yields some interesting 
 results. In the first place, we observe that the London curves 
 of deaths both from bronchitis and pneumonia vary less from 
 week to week than the corresponding curves for Dublin, which 
 are much less regular, and, as it were, more serrated. The reason 
 for this evidently is, that in the case of London we have to deal 
 with a population which is about twelve times greater than that 
 of Dublin, hence the law of periodicity fulfils itself with greater 
 exactness in the vast population of London than in the compara- 
 tively small population of Dublin. The death curves of the larger 
 city are, as it were, seen through a magnifying glass of ten 
 diameters, in the corresponding death curves of Dublin, the varia- 
 tions from week to week being magnified or multiplied tenfold. 
 In the second place, it will be noticed that bronchitis is uniformly 
 throughout the year less fatal in proportion to the population 
 in London than it is in Dublin, while the converse is true of 
 pneumonia. According to the Census of 1881, the middle year 
 of the decade with which we are at present concerned, the popula- 
 tion of the London Registration District was 3,893,272 ; that of 
 the Dublin Registration District was 348,293. The average 
 quarterly number of deaths from bronchitis in the ten years, 
 1876-1885, were these : 
 
 First quarter, Dublin, 566*9 ; London, 4358*5. 
 Second quarter, Dublin, 338'2 ; London, 2397-1. 
 Third quarter, Dublin, 172-7 ; London, 1253-8. 
 Fourth quarter, Dublin, 3954 ; London, 3413'2. 
 
 On the other hand, the average quarterly number of deaths 
 from pneumonia in the same ten years were : 
 
 First quarter, Dublin, 112-2 ; London, 1467-2. 
 Second quarter, Dublin, 108'8 ; London, 1222-5. 
 Third quarter, Dublin, 51 "4 ; London, 734'8. 
 Fourth quarter, Dublin, 85'9 ; London, 1350-2. 
 
 The third point of interest in Tables XVIII. and XIX. is the 
 dip in the death curve from bronchitis, both in London and in 
 
PEEVALENCE OF PNEUMONIC OR LUNG FEVER 455 
 
 Dublin, from the seventh to the tenth week of the year. This 
 would seem to depend on several causes firstly, the removal 
 by death at the beginning of the year of those individuals who were 
 most susceptible to bronchitis ; secondly, the acclimatisation 
 of the surviving population to the continued cold of winter ; and 
 thirdly, the prevalence of south-west winds and open weather 
 toward the close of January and early in February. With the 
 setting in of the searching east winds of early spring the death 
 curve again rises at the beginning of March, when also there is 
 a marked rise in the death-toll exacted by pneumonia, more especi- 
 ally in London. 
 
 Another curious point is, that the changes in the contour 
 of the death curves apparently occur a week earlier in London 
 than they do in Dublin. Delay in registration in the latter 
 city seems to be the explanation of this otherwise puzzling 
 circumstance. 
 
 It will be observed that in the foregoing analysis only statistics 
 of deaths are considered, and these, unfortunately, are of minor 
 value compared with statistics of the prevalence of bronchitis 
 and of pneumonia respectively, were such available. Let us 
 hope that the day is not far distant when registration of disease 
 will be compulsory, as registration of the cause of death is at 
 present. Until this much-needed reform is carried into effect, 
 statistical inquiries into the prevalence of disease in localities 
 and in seasons will want much of that precision which alone can 
 give them scientific value. 
 
 How are we to explain the continued frequency of pneumonic 
 fever in summer and autumn ? In my opinion the solution of 
 this paradox is to be sought in the consideration of the pythogenic 
 origin of the disease in many instances, and particularly in the 
 warm season of the year. In a word, I would regard exposure 
 to cold, extremes of temperature, harsh, drying winds, and other 
 personal or climatological conditions as merely so many pre- 
 disposing causes of the disease, while I would reserve for the 
 introduction into the system of a specific virus or contagium 
 the role of an exciting cause perhaps the sole exciting cause of 
 pneumonic fever. As to the exact nature of that virus or con- 
 tagium, we are now comparatively well-informed, thanks to the 
 
456 METEOROLOGY 
 
 researches and discoveries of Klebs, Eberth, Koch, Frankel, and 
 Friedlander. Chief among the bacteria of pneumonia arc 
 the Diplococcus lanceolatus capsulatus pneumonice of Frankel 
 and the Bacillus pneumonice (Pneumoniekokken) of Friedlander. 
 Modern researches in bacteriology are full of promise. We 
 stand on the threshold of a new Science of Medicine, and before 
 long a still greater flood of light will doubtless be shed upon the 
 intimate nature and pathology of pneumonia as well as of other 
 blood diseases. 
 
 In the Medical Report of Cork Street Fever Hospital and House 
 of Recovery, Dublin, for the year 1884, I ventured to assert that 
 the claims of pneumonia to be considered a specific fever rested 
 principally upon 
 
 1. Its not infrequent epidemic prevalence, which the biblio- 
 graphy of the disease places beyond dispute. 
 
 2. Its proved infectiousness in some instances, as, for example, 
 those observed at Dal ton in the spring months of 1883, by 
 Dr. E. Slade King, and Mr. Sloane Michell, M.R.C.S., England. 1 
 
 3. Its occasional pythogenic origin, and the remarkable 
 correlation which appears to exist between it and enteric fever. 
 
 4. Its mode of onset, or " invasion/' which exactly resembles 
 that of the recognised specific fevers. 
 
 5. The appearance of constitutional symptoms before the 
 development of local signs, or even local symptoms in many 
 instances ; in other words, the existence of a " true period of 
 invasion/'' 
 
 6. The critical termination of the febrile movement in all 
 uncomplicated cases. 
 
 7. The presence of local epi-phenomena in connection with 
 the skin, such, for example, as eruptions of herpes, sweat rashes, 
 and the occurrence of desqnamation. 
 
 8. The development of sequelae in some cases, such as an attack 
 of nephritis, followed by renal dropsy, ataxia like that observed 
 after typhus or diphtheria, mania, and so on. 
 
 9. The discovery of a bacterium in pneumonic exudation, 
 to which analogy, at all events, points as pathognomonic. 
 
 In my hospital and private practice I have acquired the 
 1 C/. The Practitioner. April, 1884. 
 
PREVALENCE OF PNEUMONIC OR LUNG FEVER 457 
 
 habit of expressing the relation of the local lesion in pneumonia, 
 or pneumonic fever (lung fever), to the essential disorder, in 
 terms of the intestinal lesions in enteric fever to that disease. 
 Just as physicians and pathologists have long since come to 
 avoid the dangerous error I would even say heresy of Broussais 
 and his school, who held that the pyrexia or feverishness in 
 enteric fever was symptomatic of and secondary to a local in- 
 flammation of the glands of the small intestine, so we shall come 
 in time to avoid the similar and not less dangerous, but more 
 widely disseminated, error of regarding the pyrexia in pneumonia 
 as symptomatic of and secondary to a local inflammation of the 
 lungs. The day is seemingly not far distant when we shall 
 speak of " pneumonic fever " in precisely the same way as we 
 use the term " enteric fever " at present that is, to signify a 
 disease due to a zymotic or specific blood infection, manifesting 
 itself after the lapse of a certain time, by physical phenomena 
 objective and subjective connected, in this instance generally, 
 but by no means of necessity with the lungs. 
 
APPENDIX 
 GREEN FLASH 
 
 The Green Flash on the Horizon at Sunset. In Nature (vol. xxix., 
 p. 7) the Rev. G. H. Hopkins, of Week, St. Mary Rectory, Bude, 
 Cornwall, writes : "In a clear sky, as the disc of the sun sinks 
 down below the horizontal line of the ocean, the parting ray is a 
 brilliant emerald green. . . . The same effect is not produced by 
 the sun setting behind a distant bank of clouds. Probably the 
 first ray from the rising sun would be the same unexpected colour." 
 On p. 76 of the same volume of Nature Mr. William Swan describes 
 a similar appearance which was seen by him at sunrise from the 
 summit of the Rigi on September 13, 1805. He writes : " The 
 very first rays, although necessarily proceeding from the com- 
 paratively obscure limb of the sun, were dazzlingly brilliant, and 
 of a superb emerald green colour." Many references to the 
 " green flash " will be found scattered through the pages of the 
 Journal of the British Astronomical Association, and the pheno- 
 menon attracted great interest among the readers of Symons's 
 Meteorological Magazine some years ago, so that in vol. xli. of 
 that publication we find several interesting communications on 
 the subject. 
 
 Notable among these is an article by Professor Arthur A. 
 Rambaut, D.Sc., F.R.S., Radcliffe Observer in the University 
 of Oxford, who saw the green flash for the first time on the evening 
 of September 27, 1905. He was then returning from the meeting 
 of the British Association in South Africa, on board the s.s. 
 Durham Castle, and, as the sun set behind the grand range of 
 hills terminating in Cape Guardafui, East Africa, he was fortunate 
 enough to see the phenomenon to perfection. As to its cause, the 
 
 459 
 
460 METEOROLOGY 
 
 view had been put forward by Mr. R. C. Cann Lippincott and 
 others that the green flash seen as the last ray of the setting sun 
 disappears is the apparent image or spectrum of the sun which 
 has just sunk below the horizon seen in the complementary 
 green colour. 1 But this physiological theory is disposed of by 
 Professor Rambaut, who states 2 that the large number of 
 observations of the "green flash" at sunrise affords a complete 
 refutation of that hypothesis. According to him, the obvious and 
 satisfactory explanation is that the " green flash " is due to the 
 unequal refraction experienced by rays of different colours in 
 passing through the Earth's atmosphere. It is a well-known 
 fact that the setting sun has actually sunk completely beneath 
 the horizon at the moment when it appears to us to be first in 
 contact with it, and when the upper limb is just about to dis- 
 appear it is actually 36 minutes of arc below the horizon. We 
 see it in virtue of the refraction which bends the rays round the 
 horizon so as to reach our eyes. The red, being the least re- 
 frangible, is the first to be lost, and the different colours of the 
 spectrum vanish one by one in the order red, orange, yellow, 
 green, blue, indigo, and violet. Professor Rambaut remarks 
 that the term " green flash " is unsatisfactory. When best seen, 
 the colour of the remaining segment of the sun at setting turns, 
 at the very last moment, from green to blue. The flash does not 
 appear indigo or violet, partly owing to the relative feebleness 
 with which these rays impress the eye, compared with the yellow, 
 green, and blue rays ; and partly because the selective absorp- 
 tion of the atmosphere cuts down the rays of shorter wave-length 
 in a larger proportion than those which are less refrangible. 
 The violet and indigo are thus most reduced. 
 
 1 Symons's Meteorological Magazine, 1906, vol. xli., p. 11. 
 
 2 Ibid., p. 22. 
 
APPENDIX II 
 
 TABLE OF CORRECTIONS FOR REDUCING BAROMETRIC 
 READINGS TO STANDARD GRAVITY, LATITUDE 45 
 
 I.at. 
 N. 
 
 orS. 
 
 Correction. 
 
 Lat. 
 
 N. 
 orS. 
 
 Correction. 
 
 Lat. 
 
 N. 
 or S. 
 
 Correction. 
 
 Lit. 
 N. 
 
 orS. 
 
 Correction. 
 
 At 27". 
 
 At 30". 
 
 At 27". 
 
 At 30". 
 
 At 27". 
 
 At 30". 
 
 At 27". 
 
 At 30". 
 
 
 
 070 
 
 078 
 
 23 
 
 049 
 
 054 
 
 46 
 
 002 
 
 003 
 
 69 '052 
 
 058 
 
 1 
 
 070 
 
 078 
 
 24 -047 
 
 052 
 
 47 : '005 
 
 005 
 
 70 -054 
 
 060 
 
 2 
 
 070 
 
 078 
 
 25 '045 
 
 050 
 
 48 -007 
 
 008 
 
 71 '055 
 
 061 
 
 3 
 
 070 
 
 077 
 
 26 -043 
 
 048 
 
 49 -010 
 
 Oil 
 
 72 '057 
 
 063 
 
 4 
 
 069 
 
 077 
 
 27 ! -041 
 
 046 
 
 50 -012 
 
 013 
 
 73 '058 '064 
 
 5 
 
 069 
 
 077 
 
 28 j '039 
 
 043 
 
 51 '015 
 
 016 
 
 74 
 
 059 '066 
 
 6 
 
 068 
 
 076 
 
 29 
 
 037 
 
 041 
 
 52 -017 
 
 019 
 
 75 
 
 061 -067 
 
 7 
 
 068 
 
 075 
 
 30 
 
 035 
 
 039 
 
 53 '019 
 
 021 
 
 76 
 
 062 '069 
 
 8 
 
 067 
 
 075 
 
 31 
 
 033 
 
 036 
 
 54 ! -022 
 
 024 
 
 77 
 
 063 -070 
 
 9 
 
 067 
 
 074 
 
 32 
 
 031 
 
 034 
 
 55 
 
 024 
 
 027 
 
 78 -064 -071 
 
 10 
 
 066 
 
 073 
 
 33 
 
 028 
 
 032 
 
 56 
 
 026 
 
 029 
 
 79 
 
 065 '072 
 
 11 
 
 065 
 
 072 
 
 34 
 
 026 
 
 029 
 
 57 
 
 028 
 
 032 
 
 80 
 
 066 
 
 073 
 
 12 
 
 064 
 
 071 
 
 35 
 
 024 
 
 027 
 
 58 
 
 031 
 
 034 
 
 81 
 
 067 
 
 074 
 
 13 
 
 063 
 
 070 
 
 36 -022 
 
 024 
 
 59 
 
 033 
 
 036 
 
 82 
 
 067 -075 
 
 14 
 
 062 
 
 069 
 
 37 '019 
 
 021 
 
 60 
 
 035 
 
 039 
 
 83 -068 -075 
 
 15 
 
 061 
 
 067 
 
 38 -017 
 
 019 
 
 61 
 
 037 
 
 041 
 
 84 -068 
 
 076 
 
 16 
 
 059 
 
 066 
 
 39 i -015 
 
 016 
 
 62 
 
 039 
 
 043 
 
 85 i '069 
 
 077 
 
 17 
 
 058 
 
 064 
 
 40 -012 
 
 013 
 
 63 
 
 041 
 
 046 
 
 86 '069 
 
 077 
 
 18 
 
 057 
 
 063 
 
 41 -010 
 
 on 
 
 64 '043 
 
 048 
 
 87 i '070 
 
 077 
 
 19 
 
 055 
 
 061 
 
 42 -007 
 
 008 
 
 65 i -045 
 
 050 
 
 88 ! -070 
 
 078 
 
 20 
 
 054 
 
 060 
 
 43 '005 
 
 005 
 
 66 i '047 
 
 052 
 
 89 -070 
 
 078 
 
 21 
 
 052 i -058 
 
 44 -002 
 
 003 
 
 67 '049 
 
 054 
 
 90 -070 
 
 078 
 
 22 
 
 050 
 
 056 
 
 45 
 
 
 
 68 '050 
 
 056 
 
 
 
 
 461 
 
462 
 
 METEOROLOGY 
 
 APPENDIX III 
 
 HYGROMETRICAL TABLES (Glaisher) 
 I. TABLE OF FACTORS. 
 
 Reading of 
 ;he Dry -Bulb 
 Thermometer. 
 
 Factor. 
 
 Reading of 
 the Dry Bulb 
 1 hermometer. 
 
 Factor. 
 
 Reading of 
 'he Dry-Bulb 
 'hermometer. 
 
 Factor. 
 
 o 
 
 
 o 
 
 
 o 
 
 
 20 
 
 8-14 
 
 44 
 
 2-18 
 
 68 
 
 1-79 
 
 21 
 
 7-88 
 
 45 
 
 2-16 
 
 69 
 
 1-78 
 
 22 
 
 7'60 
 
 46 
 
 2-14 
 
 70 
 
 1-77 
 
 23 
 
 7-28 
 
 47 
 
 2-12 
 
 71 
 
 1-76 
 
 24 
 
 6-92 
 
 48 
 
 2-10 
 
 72 
 
 1-75 
 
 25 
 
 6-53 
 
 49 
 
 2-08 
 
 73 
 
 1-74 
 
 26 
 
 6-08 
 
 50 
 
 2-06 
 
 74 
 
 1-73 
 
 27 
 
 5-61 
 
 51 
 
 2-04 
 
 75 
 
 1-72 
 
 28 
 
 5-12 
 
 52 
 
 2-02 
 
 76 
 
 1-71 
 
 29 
 
 4-63 
 
 53 
 
 2-00 
 
 77 
 
 1-70 
 
 30 
 
 4-15 
 
 54 
 
 1-98 
 
 78 
 
 1-69 
 
 31 
 
 3-70 
 
 55 
 
 1-96 
 
 79 
 
 1-69 
 
 32 
 
 3-32 
 
 56 
 
 1-94 
 
 80 
 
 1-68 
 
 33 
 
 3-01 
 
 57 
 
 1-92 
 
 81 
 
 1-68 
 
 34 
 
 2-77 
 
 58 
 
 1-90 
 
 82 
 
 1-67 
 
 35 
 
 2-60 
 
 59 
 
 1-89 
 
 83 
 
 1-67 
 
 36 
 
 2-50 
 
 60 
 
 1-88 
 
 84 
 
 1-66 
 
 37 
 
 2-42 
 
 61 
 
 1-87 
 
 85 
 
 1-65 
 
 38 
 
 2-36 
 
 62 
 
 T86 
 
 86 
 
 1-65 
 
 39 
 
 2-32 
 
 63 
 
 1-85 
 
 87 
 
 1-64 
 
 40 
 
 2-29 
 
 64 
 
 1-83 
 
 88 
 
 1-64 
 
 41 
 
 2-26 
 
 65 
 
 1-82 
 
 89 
 
 1-63 
 
 42 
 
 2-23 
 
 66 
 
 1-81 
 
 90 
 
 1-63 
 
 43 
 
 2-20 
 
 67 
 
 1-80 
 
 91 
 
 1-62 
 
APPENDIX III 
 
 463 
 
 HYGROMETRICAL TABLES continued 
 
 TABLE II. TENSION, OR ELASTIC FORCE OF AQUEOUS VAPOUR 
 
 IN INCHES OF MERCURY FOR EVERY DEGREE OF 
 
 TEMPERATURE FROM TO 95. 
 
 
 
 
 
 
 
 
 
 Temp. 
 
 Ten-ion. 
 
 Temp. 
 
 Tension. 
 
 Temp. 
 
 Tension. 
 
 Temp. 
 
 Tension. 
 
 o 
 
 
 o 
 
 
 o 
 
 
 o 
 
 
 
 
 044 
 
 24 
 
 129 
 
 48 
 
 335 
 
 72 
 
 785 
 
 1 
 
 046 
 
 25 
 
 135 
 
 49 
 
 348 
 
 73 
 
 812 
 
 2 
 
 048 
 
 26 
 
 141 
 
 50 
 
 361 
 
 74 
 
 840 
 
 3 
 
 050 
 
 27 
 
 147 
 
 51 
 
 374 
 
 75 
 
 868 
 
 4 
 
 052 
 
 28 
 
 153 
 
 52 
 
 388 
 
 76 
 
 897 
 
 5 
 
 054 
 
 29 
 
 160 
 
 53 
 
 403 
 
 77 
 
 927 
 
 6 
 
 057 
 
 30 
 
 167 
 
 54 
 
 418 
 
 78 
 
 958 
 
 7 
 
 060 
 
 31 
 
 174 
 
 55 
 
 433 
 
 79 
 
 990 
 
 8 
 
 062 
 
 32 
 
 181 
 
 56 
 
 449 
 
 80 
 
 1-023 
 
 9 
 
 065 
 
 33 
 
 188 
 
 57 
 
 465 
 
 81 
 
 1-057 
 
 10 
 
 068 
 
 34 
 
 196 
 
 58 
 
 482 
 
 82 
 
 1-092 
 
 11 
 
 071 
 
 35 
 
 204 
 
 59 
 
 500 
 
 83 
 
 1-128 
 
 12 
 
 074 
 
 36 
 
 212 
 
 60 
 
 518 
 
 84 
 
 1-165 
 
 13 
 
 078 
 
 37 
 
 220 
 
 61 
 
 537 
 
 85 
 
 1-203 
 
 14 
 
 082 
 
 38 
 
 229 
 
 62 
 
 556 
 
 86 
 
 1-242 
 
 15 
 
 086 
 
 39 
 
 238 
 
 63 
 
 576 
 
 87 
 
 1-282 
 
 16 
 
 090 
 
 40 
 
 247 
 
 64 
 
 596 
 
 88 
 
 1-323 
 
 17 
 
 094 
 
 41 
 
 257 
 
 65 
 
 617 
 
 89 
 
 1-366 
 
 18 
 
 098 
 
 42 
 
 267 
 
 66 
 
 639 
 
 90 
 
 1-410 
 
 19 
 
 103 
 
 43 
 
 277 
 
 67 
 
 661 
 
 91 
 
 1-455 
 
 20 
 
 108 
 
 44 
 
 288 
 
 68 
 
 684 
 
 92 
 
 1-501 
 
 21 
 
 113 
 
 45 
 
 299 
 
 69 
 
 708 
 
 93 
 
 1-548 
 
 22 
 
 118 
 
 46 
 
 311 
 
 70 
 
 733 
 
 94 
 
 1-596 
 
 23 
 
 123 
 
 47 
 
 323 
 
 71 
 
 759 
 
 95 
 
 1-646 
 
464 
 
 METEOROLOGY 
 
 M w O 
 
 I I H OD 
 
 g 5 
 
 ^H ^r ^H 
 
 p . pj ^ 
 
 ^ II 
 
 000' I 
 
 000 'I 
 
 000' I 
 
 OOO'I 
 
 1 
 '> 
 
 i ii i(N<M<N<N<N<NCOCO 
 
 ooo'i 
 
 icct-t-ooooosoio 
 
 2 ,2^ 
 
 bo 
 I 2 
 
APPENDIX IV 
 
 465 
 
APPENDIX V 
 
 BIBLIOGRAPHY OF SNOW-ROLLERS 
 
 THE Editors of the Quarterly Journal of the Royal Meteorological 
 Society give the following references to descriptions of pheno- 
 mena which throw further light upon the formation of the snow- 
 rollers which are described at page 251 : 
 
 Handy Book of Meteorology. By Alexander Buchan, second 
 edition, pp. 202, 203. 1868. (Snow-rollers observed by the 
 Rev. Charles Clouston, Sandwick, Orkney, about 9 p.m. on 
 March 5, 1862.) 
 
 Nature, vol. xxvii., p. 483. 1883. (A " Remarkable Pheno- 
 menon Natural Snowballs/' an account furnished to the Courant 
 newspaper of February 22, 1883, by Mr. Samuel Hart, Trinity 
 College, Hartford, Conn., U.S.A.) 
 
 Nature, vol. xxvii., p. 507. 1883. (Letter from Mr. G. J. 
 Symons, referring to a letter from the late Admiral Sir F. W. 
 Grey, which appeared in the Meteorological Magazine for May, 
 1876. The Admiral observed masses of snow, like boulders, at 
 Lynwood, Sunningdale, Staines, after the snowstorm of Thursday 
 night, April 13, 1876.) 
 
 Symons' s Monthly Meteorological Magazine, vol. xx., p. 7. 1885. 
 (Snow-rollers observed by Mr. H. Mellish, F.R.Met.Soc., at 
 Hodsock Priory, Worksop, on January 12 and 13, 1885.) 
 
 U.S. Monthly Weather Review, vol. xxxiv., pp. 325, 326. 19G6. 
 (Snow-rollers observed during the night of January 18, 1906, 
 by Mr. Wilson A. Bentley, at Jericho, Vt. The snow-rolls 
 were formed as the temperature rose slowly from 24 to 34 F., 
 and the lower wind shifted from westerly to southerly points, and 
 blew at times in gusts.) 
 
 U.S. Monthly Weather Review, vol. xxxiv., p. 326. 1906. (Snow- 
 rollers observed at Mount Pleasant, Michigan, by Professor R. D. 
 Calkins, after a light flaky snow on the evening of January 17, 
 1906.) 
 
 U.S. Monthly Weather Review, vol. xxxv., pp. 70, 71. 1907. 
 (Snow-rollers noted at Canton, New York, by M. L. Fuller, on 
 February 19, 1907.) 
 
 466 
 
INDEX OF SUBJECTS AND PLACES 
 
 ABSOLUTE humidity, 186, 199 
 
 Accumulated temperature, 36, 351 
 
 Actinometer, 98 ; Herschel's, 98, 
 102 ; Richard's, 102, 103 ; Ang- 
 strom's electric compensation 
 pyrheliometer and, 102 
 
 Acute infective diseases, 409, 431 
 
 Adiabatic, meaning of term, 338 ; 
 atmosphere, 343 ; cooling of 
 ascending air-currents, 363 
 
 Advective region of atmosphere, 
 338, 343 
 
 Aeronautics, International Commis- 
 sion of Scientific, 327 
 
 Africa, aerological expedition to 
 Tropical East, 344, 345 
 
 Air temperatures of the British 
 Islands, 386 
 
 Air-trap in a barometer, 125 
 
 Air, weight of, 13 ; liquefaction of, 
 14 ; freezing of, 14 ; elasticity 
 of, 15 ; rarefaction of, 16 ; volu- 
 metric analyses of, 19, 20 ; a 
 mechanical mixture, 20, 21 ; 
 composition of, at great alti- 
 tudes, 347, 348 ; temperature 
 and its measurement, 75 ; capa- 
 city for containing aqueous 
 vapour, 166 ; supersaturated, 
 195-197 
 
 Air-meter, 277 
 
 Aitken's dust-counters and koni- 
 scope, 198 
 
 Alcohol, specific heat of, 86 
 
 Allotropic oxygen, 321 
 
 Altazimuth anemometer, Casella's, 
 280 
 
 Altitude, barometric correction for, 
 
 143, 148-150 
 
 effect of, on rainfall, 239, 240, 241 
 on temperature, 99, 100, 352, 
 353 
 
 Altitude, effect of, on electric tension 
 301 ; on climate, 352, 353 
 
 Altitudes, measurement of , 136, 137 ; 
 composition of air at great, 347, 
 348 
 
 Alto-cumulus cloud, 212 
 
 Alto-stratus cloud, 212 
 
 American Arctic current, 360, 361 
 j American Weather, General Greely's, 
 
 68 
 
 I Ammonia in air, 19, 23 
 | Ammonium sulphide in air, 19, 23 
 
 Amount of cloud, 41 
 
 Ampere, 306 
 
 Anemobarometer, Wilke's, 271 
 
 Anemo - cinemographe, Richard's, 
 285-288 
 
 Anemograph, Casella's, 269, 289 ; 
 Beckley, 285 ; stations, 29 
 
 Anemometers, 265 ; history of, 269 ; 
 pendulum, 270; Wild's, 270, 
 271 ; bridled, 271 ; pressure 
 plate, 271 ; torsion, 271 ; 
 Osier's, 271, 287 ; Dines's, 271, 
 284 ; Patent Pressure Port- 
 able, 271, 272, 273 ; pressure on 
 a fluid, 273, 274 ; Lind's, 273, 
 
 274, 289 ; velocity, 274 ; hemi- 
 spherical cup, 274 ; Robinson's, 
 271, 274-276, 287, 289 ; " step," 
 
 275, 276; current-meter, 277; 
 air-meter, 277 ; evaporation or 
 temperature, 277 ; suction, 278 ; 
 Hagemann's, 278, 289 ; direc- 
 tion, 279 ; Lomonosow's, 279 ; 
 inclination, 279 ; Casella's self- 
 recording, 269, 289 ; Casella'3 
 altazimuth, 280 ; musical, 284 ; 
 helicoid, 284 ; Richard's, 285 ; 
 Kew pattern, 285, 287, 289 
 
 Anemometry, 265 ; history of, 269, 
 270 
 
 467 
 
 302 
 
468 
 
 METEOKOLOGY 
 
 Anemoscope, Pickering's, 270 ; 
 
 Laughton's feather-vane, 283 
 Aneroid barometer, 132-134 ; draw- 
 backs to its use, 134 ; engineer- 
 ing, 137 
 barograph, 129 
 Annual barometric variations, 151, 
 
 156 
 Climatological Report, Canada, 71, 
 
 72 
 
 march of electrical tension, 301 
 rainfall of the British Islands, 395 ; 
 
 mean, of the world, 221 
 range of temperature, laws of dis- 
 tribution of, 355, 356 
 Antarctic, Atlantic current, 361 
 Continent, why constantly snow- 
 covered, 371 
 
 Expedition of 1902-04, 371 
 Anticyclones, 9, 156, 157, 158, 160, 
 
 164, 379, 380, 381 
 Anticyclonic, 4, 160, 379, 380, 381 
 
 circulation of air, 6, 17, 157, 
 
 158 ; barometric variations, 
 
 151, 160 
 system contrasted with cyclonic, 
 
 161 ; clouds in, 210 
 weather, in winter, 157, 158 ; in 
 
 summer, 164 
 " Anti-trades," 266 
 Antozone, 322, 324 
 Antozonides, 322 
 Aphelion, 99 
 
 Apjohn's formula, 182, 183 
 Appendices I. to V., 459-466 
 April, mean temperature of, 389, 390 
 Aquameter, 179, 180 
 Aqueous vapour, atmosphere of, 
 
 165, 175, 190, 205, 220 
 vapour in the air, 19, 25, 354 ; 
 
 tension of, 151, 152, 165, 175, 
 181, 186 ; where most abun- 
 dant, 186 
 
 Arctic circle, diurnal range of pres- 
 sure in, 151, 153 
 
 Argon, 19, 22, 347, 348 
 
 Asia, climate of, 156, 354, 355, 367 
 378, 379 
 
 Asiatic cholera, 410, 416 
 
 Athermancy, 18 
 
 Atlas International des Nuages, 213 
 
 Atlas, meteorological, of the British 
 Isles, 383, 388, 389, 392, 393, 
 396 
 
 Atmidometers, 167, 168 
 
 Atmidometry, 166 
 
 Atmometry, 166 
 
 Atmosphere, definition of, 11 ; its 
 physical properties, 11 ; its 
 height, 17, 117-118 ; its trans- 
 parency and diathermancy, 17 ; 
 its composition, 19 ; of aqueous 
 vapour, 165, 175, 190, 205, 220 ; 
 dampness of, 186 ; in a constant 
 state of electric tension, 301 
 inversion of temperature in upper, 
 
 100, 331 
 isothermal layer of, 100, 331, 336, 
 
 343, 345 
 upper, investigation of the, 325- 
 
 349 
 
 Atmospheric air, weight of, 13 ; 
 liquefaction of, 14 ; freezing of, 
 14 ; elasticity of, 15 ; rarefac- 
 tion of, 16 ; where purest, 19 ; 
 its composition, 19 ; a mechani- 
 cal mixture, 20, 21 
 dust, 195 - 198 ; composition of, 
 
 197 
 
 electricity, 293, 308 ; nature of, 
 295 ; its phenomena, 296 ; in 
 the production of rain, 243-246 
 pressure, 17, 114 ; how measured, 
 114 ; distribution of, in Canada, 
 158 ; extremes of, 120, 121 ; 
 variations in, 151 ; influence of 
 aqueous vapour on, 165, 166 ; 
 in British Islands, 392 ; in 
 Dublin, 402 ; periodicity of, 408 
 Atomic theory of Dalton, 15 
 Attached thermometer, 139 
 Attraction, capillary, 140 ; electri- 
 cal, 295 
 
 Aurora, 294, 296, 314 ; borealis, 296, 
 314; australis, 296, 314 ; distri- 
 bution of, 314 ; periodicity of, 
 314 ; explanation of , 315 ; height 
 of, 315 ; tints of, 315 
 Australian " brickfielders," 375 ; 
 
 " bursters," 375 
 | Autumnal distribution of disease, 
 
 417-419 
 fever, 431 
 
 Average mean temperature, 94 
 Axis of rotation, Earth's, 4, 5, 11, 12, 
 151 
 
 BABINGTON'S atmidometer, 168 
 Bacillus pneumonise of Friedlander, 
 
 456 
 
 Backing of wind, 163 
 Bacteriology of summer diarrhoaa, 
 
 425, 426 
 Balance pattern rain-gauges, 233, 234 
 
INDEX OF SUBJECTS AND PLACES 
 
 469 
 
 Ball lightning, 308, 310 
 Balloon ascents for meteorological 
 purposes, 46, 67, 100, 185, 325 ; 
 wet-bulb readings during, 185 
 Ballons-sondes, 327 
 Barograph, 126 ; meaning of the 
 term, 126 ; King's, 126, 127 ; 
 Ronalds's photographic, 128 ; 
 Redier's mechanical, 128 ; elec- 
 tric, 128, 129 ; aneroid, 129, 130 
 Barometer, 26, 1 14 ; history of the, 
 118, 119; meaning of the word, 
 114 ; uses of the instrument, 
 114 ; its discovery, 115 ; its con- 
 struction, 120 ; mercurial, 116 ; 
 120, 121, 122 ; glycerine, 116, 
 117 ; wheel, 119, 126 ; extreme 
 readings of, 120, 121 ; mount- 
 ing, 122 ; Wollaston's differ- 
 ential, 274 
 aneroid, 132, 133, 134 ; holosteric, 
 
 132 ; metallic, 133 
 corrections to be applied to read- 
 ings of, 122, 140, 143-150 
 how to be mounted, 139 
 readings, 139 
 scale, 143 
 
 adjustment of, Wallis's, 123, 124 
 Standard, or Fortin, 123 
 Kew, 123, 124, 125 
 siphon, 123, 125 
 Fitzroy, or " gun," 125 
 self-registering, 126 
 Bartrum's open-scale, 131, 132 
 piesmic, 134, 135 
 Dines's self-recording mercurial, 
 
 135, 136 
 
 sympiesometer, 137 
 Barrett's open-scale, 138 
 Bourdon tube, 133, 328 
 variations in height of, 151 
 Barometrical fluctuations, 151 
 
 gradients, 7, 8, 265 
 Baroscope, 119 
 Barton Moss Evaporation Station, 
 
 34, 172 
 
 Base temperature, 35 
 Beaded lightning, 309 
 Bearings, how determined by com- 
 pass, 267 ; by the pole star, 
 268 ; by the sun, 268 
 Beaufort scale of wind force, 32, 
 36, 38, 39, 40, 266 ; of weather, 
 32 
 
 Beckley anemograph, 285 
 Beech-trees and lightning, 311 
 Behring's Straits current, 361 
 
 Belfast, rainfall in, 398 
 Bench-mark, Ordnance Survey, 148 
 Bergen, cholera at, 413-415 ; mean 
 
 annual rainfall at, 262 
 Besson's nephometer, 218 
 comb nephoscope, 214 
 zenith nephoscope, 216 
 Bimetallic thermometer, 328 
 Black-bulb thermometer in vacua, 
 
 101 
 
 " Black rain," 248 
 Blitzableiter, 317 
 " Blitzrohren," 310 
 Blizzard, 250 
 Blood- rain, 264 
 Blue Hill Observatory, U.S.A., 325, 
 
 327, 328 
 
 " Blue mists," 416 
 Bog of Allen, 390, 401 
 Bogs, influence of, on climate, 390, 
 
 401 
 Boiling-point of water, 76-79, 138, 
 
 194 
 
 Boomerang, principle of the, 13 
 Bora of Trieste and Dalmatia, 373 
 Bourdon tube thermometers, 327 
 
 tube barometer, 133, 328 
 Bourrasques, 380 
 Boyle's and Mariotte's law, 14 
 Bradford rain-gauge, 226 
 Breslau, cholera in, 413, 415 
 " Brickfielders," Australian, 375 
 Bridled anemometers, 271 
 Bright sunshine, duration of, 109 
 Bright-bulb thermometer in vacua, 
 
 101 
 
 Brisbane, torrential rainfall in, 263 
 British Association for the Advance- 
 ment of Science, 69, 91, 124, 
 
 178, 244, 293, 308, 325, 326, 
 
 327, 343, 385, 406 
 Islands, rainfall in, 261, 395 
 
 climate of, 289, 351, 378, 395 ; 
 gales in, 289-291 ; sea-tem- 
 peratures round, 356, 383 ; 
 air-temperatures of, 386 ; at- 
 mospheric pressure in, 392 
 meteorological observations, 27, 
 
 28,29 
 Meteorological Office, 7, 28, 29, 
 
 30, 33, 37, 38, 40, 109, 128, 129, 
 
 148, 149, 150, 209, 225, 238, 
 
 266, 327, 348, 383, 388, 394 
 Meteorological Society. See 
 
 Royal Meteorological Society 
 British Rainfall, 33, 37, 170, 239, 
 
 261, 395 
 
470 
 
 METEOKOLOGY 
 
 Brocken, Spectre of the, 203 
 
 Bronchitis, seasonal prevalence of, 
 448, 449, 451-455 
 
 Brontometer, 320, 321 
 
 Brush discharge and St. Elmo's fire, 
 312 
 
 Bulb, forked or cylindrical thermo- 
 meter, 86, 104 
 
 Bunsen flame, 24 
 
 Bureau, United States Weather, 44- 
 68, 70 ; its cost, 48 ; its depart- 
 ments, 48-51 ; its library, 51, 
 52 ; its regular stations, 52, 54 ; 
 its publications, 50, 51 
 
 Bureaux, weather, 27 
 
 Burning sunshine recorder, 109 
 
 " Bursters," Australian, 375 
 
 Buys Ballot's law, 4, 6, 9, 157, 158, 
 159, 265, 378 
 
 CALIBRATING thermometer tubes, 80 
 Calm, meaning of the term, 4, 268 
 Campbell-Stokes sunshine recorder, 
 
 109-110 
 Canada, climate of, 351, 356 
 
 meteorological observers in, 70, 7 1 
 telegraph-reporting stations in, 
 
 70, 71, 72 
 
 storm-signal stations in, 70, 71, 72 
 climatological stations in (1907), 
 
 72 
 
 atmospheric pressure, distribu- 
 tion of, in, 158 
 Canadian Meteorological Service, 
 
 69-74 
 establishment and development 
 
 of, 69-72 
 
 publications of, 72 
 cost of, 70 
 grant to (1909), 72 
 weather forecasts of, 70, 71, 72-74 
 Director of, 69, 71 
 Central Office and Observatory of, 
 
 Toronto, 69, 72, 74 
 equipment and work of, in 1907, 
 
 72 
 snow and drift warnings by, 73, 
 
 74 
 
 storm -warnings by, 70, 73, 74 
 Cancer, Tropic of, 12 
 Capacity correction in barometers, 
 
 122, 143, 144, 146 
 Capillarity, correction for, 140, 143, 
 
 147 
 Capillary action in barometer, 140 
 
 attraction, 140 ; repulsion, 140 
 Capricorn, Tropic of, 12 
 
 Carbon dioxide in air, 19-22 ; whence 
 
 derived, 21 
 Carse clay, 400 
 
 Casella's mercurial minimum ther- 
 mometer, 85, 86, 87-89 
 electrical rain-gauge, 229 
 self-recording anemometer, 269, 
 289 ; air-meter, 277 ; altazi- 
 muth anemometer, 280-283 
 self - registering earth - thermo- 
 meter, 105, 106 
 
 solar radiation maximum ther- 
 mometer, 102 ; bifurcated grass - 
 minimum thermometer, 104, 
 105 
 
 Castor and Pollux, 312 
 Cathetometer, 144-146 
 Cautionary signals, 62, 63, 64 
 Celsius' thermometer scale, 77-80 
 Centigrade thermometer, 15, 77-80 
 Centrifugal force, 16 
 Chalk, 401 
 
 Chapletted lightning, 309 
 Charles's law, 15, 16 
 Charts, weather, 27, 31, 32, 45, 51, 
 
 55,56,57,67,70, 71,72,73 
 underground temperature, 107- 
 
 109 
 
 Jordan barometer, 117 
 diurnal oscillation of the baro- 
 meter in various latitudes, 155 
 cyclonic and anticyclonic isobars, 
 
 160 
 
 synoptic weather, 27, 35, 160 
 " Chasse-neige," 252 
 Chemical hygrometers, 178, 179 
 Cherrapunji, rainfall at, 220, 262, 
 
 263 
 Chief of the United States Weather 
 
 Bureau, 46, 48 
 Signal Officer, United States, 45, 
 
 46,52 
 Chinook wind of the Canadian 
 
 prairies, 261, 262, 353, 375 
 Chlorophyll, chemical action of, 22 
 Cholera, 413-417 ; and subsoil tem- 
 perature, 109 ; and ozone, 323, 
 324, 415 ; and subsoil water, 364 
 Asiatic, 410, 416 
 " Cholera blasts," 416 
 " Cholera clouds," 416 
 infantum, 423, 424 
 nostras, 416 
 
 seasonal prevalence of, 413-417 
 Cholerine and subsoil temperature, 
 
 106-109, 416 
 Chris tiania, cholera in, 415 
 
INDEX OF SUBJECTS AND PLACES 
 
 471 
 
 Cirro-cumulus cloud, 205, 207, 212 
 Cirro-stratus cloud, 205, 206, 210, 
 
 211,212 
 Cirrus cloud, 205, 206, 212 ; an aid 
 
 in forecasting, 20 5 
 Cistern, barometer, 120, 122 
 City fog, 200-202 
 Classification of clouds, 205, 211, 
 
 212 
 
 gales, 290 
 thunderstorms, 302 
 winds, 372 
 Clays, influence of, on climate, 363, 
 
 400, 401 
 
 Climate, 350, 362 ; meaning of the 
 word, 350 ; influence of radia- 
 tion upon, 100 ; influence of 
 forests upon, 365-368 ; on what 
 it depends, 351, 352 
 and weather, 350 
 definitions of, 350, 351 ; insular 
 and continental, 354, 378 ; 
 moderate and excessive, 354, 
 378 
 
 of the British Islands, 378, 395 ; 
 
 mild winter, of Devon and 
 
 Cornwall, 384, 385 ; influence of 
 
 soils on, 400 
 
 Climatological observatories, 28, 33, 
 
 382 ; in Canada (1907), 72 
 
 Tables for Dublin, 402, 404-406 
 
 Climatology, 3, 53, 67, 350, 354, 378, 
 
 382, 383 
 Cloud, 190, 205 
 
 amount of, 41 ; scale for, 211 
 Cloud-atlas, 211, 213 
 Cloud-line, 205 
 
 Cloud-slopes in cumulus, 209, 210 
 Clouds, classification of, 205, 206, 
 211, 212 ; contractions used in 
 recording observations on, 211 ; 
 electrical state of, 302 
 Clouds, 205 ; photographs of, 67 ; 
 
 height of, 205, 206 
 formation of, by mountains, 202, 
 
 203 
 fine-weather, 212; bad-weather, 
 
 212 
 
 upper, 206 ; lower, 206, 207 
 Cloudy condensation in air, 193, 
 
 194, 195 
 Code, International, 41, 42 ; United 
 
 States Weather, 54, 59 
 Coefficient of expansion of a gas, 
 
 15 
 
 for obtaining mean temperatures, 
 94 
 
 " Col " of high pressure, 161 
 
 " Cold wave," 56 
 winds, 372-374 
 
 Collector, electrical, 297-298 
 
 Colorado, climate of, 362 
 
 Colour of lightning, 312, 313 
 
 Colour phenomena from dust, 198, 
 199 
 
 Coloured rain, 247-249 
 
 Combustion dust, 197 
 
 Comma bacillus of cholera, 416 
 
 Committee, Meteorological, 33, 93, 
 149, 183, 289, 292, 296, 334, 383, 
 388, 396 
 
 Comozants, 312 
 
 Compass bearings compared with 
 true bearings, 267 
 
 Composition of the atmosphere, 19 
 
 Compression, measure, 137 
 of gases, law of, 14 
 
 Condensation, 190 ; its various 
 forms, 190 
 
 Conduction of heat, 96 
 ! Conductors, lightning, 317 
 i Conference, International Meteoro- 
 logical, 8, 102, 181, 211, 216, 
 225, 268, 270, 287, 296, 321 
 
 Configuration of surface, influence of, 
 on climate, 357 
 
 Conservation of solar energy, 98 
 
 Consumption and subsoil drainage, 
 364 
 
 Continental climate, 354, 378 
 
 Contractions used in cloud observa- 
 tions, 211 
 I Convection of heat, 96, 97 
 
 Convective region of atmosphere, 
 
 338, 343 
 j Coronae, 206 
 
 Corposants, 312 
 
 Corrections in barometer readings, 
 122, 140, 143-150 
 
 Cost of United States Weather Ser- 
 vice, 45 
 
 Coulomb's torsion balance electro- 
 meter, 298, 299 
 
 " Critical " temperature of the soil, 
 109, 416, 417, 423, 434 
 
 Crosley's registering rain-gauge, 229 
 ! Crova's hygrometer, 187, 188 
 
 Crystalline rocks, 400, 402 
 
 Crystals, snow, 249, 250 ; ice, 206, 
 207, 254 
 
 Cumulo-cirro-stratus cloud, 210 
 
 Cumulo-nimbus, 212 
 
 Cumulo - stratus cloud, 206, 207, 
 210 
 
472 
 
 METEOROLOGY 
 
 Cumulus cloud, 205, 207, 208, 210, 
 212 ; formation of, 208 ; fes- 
 tooned, 209 ; cloud slopes in, 
 209, 210 
 
 " Curing " thermometer tubes, 81 
 
 Current meter anemometer, 277 
 
 Currents, inclined air, 279 
 
 Cyclones, 9, 157, 158, 160-163, 257, 
 303, 304 
 
 Cyclonic, 4, 160; system, 5, 378, 
 380 ; clouds in, 206, 210, 211 ; 
 contrasted with anticyclonic, 
 161 ; circulation of air, 5, 6, 
 156 - 158 ; barometric varia- 
 tions, 151, 160 
 thunderstorms, 302, 313 
 
 DAILY Weather Bulletin, Canada, 
 
 The, 71 
 
 weather charts, 27, 31, 32, 55, 67 
 Weather Map, Toronto, 71, 72 
 Weather Report, The, 31, 32, 34, 
 
 38 
 weather reports, 31, 32, 34, 35 
 
 Daniell's hygrometer, 175 ; galvanic 
 battery, 300 
 
 Dates in the history of the United 
 States Weather Service, 46, 47 
 
 Datum, Ordnance Survey, 148 
 
 Day degrees, 36, 351 ; meaning of 
 term, 36 
 
 Daylight Recorder, Jordan's, 111 
 
 De la Rue's evaporimeter, 169 
 
 Density, electrical, 296 
 
 of buildings a cause of diarrhoea, 
 424 
 
 Depression, atmospheric, 156, 159 
 secondary or subsidiary, 161, 304 
 thunderstorm, 255, 303, 304 
 V-shaped, 161 
 
 Devon and Cornwall, mild winter 
 climate of, 384, 385 
 
 Dew, 99, 175, 176, 184, 190 - 192, 
 220 ; theories of, 190, 191 ; its 
 measurement, 192, 220 ; its 
 formation interfered with, 192 
 
 Dew-point, 175-178, 181, 183, 184; 
 mode of determining, 183 
 
 Diarrhoea and' subsoil temperature, 
 106-109, 417 
 
 Diarrhceal diseases, 417 ; nature and 
 action of the poison of, 425 ; 
 bacteriology of, 425, 426 ; 
 house-flies and, 427-430; in- 
 vestigations in New York, 427- 
 429 ; investigations in England, 
 430 
 
 Diathermancy, 17, 18, 370 ; mean- 
 ing of term, 17 
 Differential barometer, Wollas- 
 
 ton's, 274 
 
 Diffused lightning, 308, 310 
 Diffusion of gases, 23, 24 
 Dines's helicoid anemometer, 284 ; 
 hygrometer, 175, 177-178 ; 
 patent pressure portable ane- 
 mometer, 271 ; light meteoro- 
 graphs, 329, 330 
 Diplococcus pneumonice of Frankel, 
 
 456 
 
 Diploma in State Medicine, 3 
 Direct hygrometers, 175 
 Direction of wind, 38, 267 ; how de- 
 termined, 7, 267 ; Lambert's 
 formula for, 287 ; true, not 
 magnetic, 41, 267 
 anemometers, 279 
 Disease, influence of season and 
 
 weather on, 409 
 registration, necessity for, 455 
 Diseases, acute infective, 409, 410, 431 
 Displacement of zero, 81, 82 
 Distribution of forecasts in United 
 
 States, 61 
 
 of rain, 249, 260-263 
 Diurnal barometric variations, 151 
 march of electricity, 301 
 rainfall, 263 
 Drainage, subsoil, and consumption, 
 
 364 
 
 Dry-bulb thermometer, 148, 181 
 Dry smoke fog, 193, 201 
 Dublin, extremes of atmospheric 
 pressure in, 120, 121 ; diurnal 
 range of atmospheric pressure 
 in, 153 ; hail-storm in, 255- 
 257 ; rainfall in, 261, 263, 363, 
 397, 398, 399, 402, 404 ; heavy 
 rains in, 263 ; mean tempera- 
 tures in, 354, 355, 388, 402, 404 ; 
 storm of February, 1903, in, 13, 
 290 ; climate of, 362, 363, 385, 
 386 ; climatological tables for, 
 402, 404-406 ; influenza epidemic 
 in, 410 ; cholera in, 415 ; diar- 
 rhceal diseases in, 417-419, 421 ; 
 enteric fever in, 431-434; typhus 
 in, 435, 436 ; smallpox in, 439 ; 
 measles in, 441, 442, 444 ; scar- 
 latina in, 443-445 ; pneumonic 
 fever in, 446, 448, 449 ; bron- 
 chitis in, 449, 451-455 
 Dust as a condenser of aqueous 
 vapour in air, 195, 196, 243 
 
INDEX OF SUBJECTS AND PLACES 
 
 473 
 
 Dust-counter, Aitken's, 198 
 
 Dust-fall, 264 
 
 Dust haze, 203 
 
 Dust particles in air, 195-198 ; com- 
 position of, 197 ; colour-pheno- 
 mena from, 198, 199 ; number 
 in atmosphere, 199 
 water, 205 ; snow, 250, 252 
 
 Dysentery and ozone, 323 
 
 EARTH temperatures, 105-109 
 thermometer, Symons's, 105 ; 
 
 Casella's, 105, 106 
 Earth's axis of rotation, 4, 5, 11, 12, 
 
 151 
 Easterly winds of spring, 373 ; why 
 
 so cold, 372 
 Ebullition, 79 
 Eccentricity of the earth's orbit, 98, 
 
 99 
 
 Ecliptic, 11 
 Edinburgh, mean temperature of, 
 
 388 ; rainfall in, 398 
 Egnell's law, 344 
 Elastic force of aqueous vapour, 151, 
 
 152, 165, 175, 181, 186 
 Elasticity of air, 15 
 Electric kites, 294, 306 
 Electrical barographs, 128, 129 
 collector, 297; Volta's, 297; 
 Thomson's water-dropping, 298 
 density, 296 
 
 dew-point hygrometer, 178 
 force, 296, 298 
 induction, 297 
 intensity, 296, 298 ; diurnal and 
 
 annual march of, 301 
 potential, 296, 298, 311 
 replenishes 300 
 self-registering rain-gauges, 229- 
 
 233 
 state of the upper atmosphere, 
 
 306, 307 
 
 tension, 296, 298 ; diurnal and 
 annual march of, 301 ; effect of 
 altitude on, 301 
 thermographs, 92, 93 
 Electricity, atmospheric, 293, 308 ; 
 nature of, 295 ; phenomena at- 
 tending, 296 
 compared with heat, 295 
 convection of, 97 
 diurnal march of, 301 
 in the production of rain, 243-246 
 negative, 296, 302, 305, 306 
 positive, 296, 302, 306 
 resinous, 295, 296 
 
 Electricity, transmission of baro- 
 metric indications by, 130 
 vitreous, 295 
 
 Electrified bodies, 295 
 
 Electrodes, 299 
 
 Electro-magnetic theory of light, 
 293, 294 
 
 Electrometer, 298 ; Thomson's 
 Quadrant, 299 ; Thomson's 
 Portable, 299, 300; photo- 
 graphic, 300 ; Peltier's, 301 
 
 Electro-motive force, 298 ; expressed 
 in " volts," 300 
 
 Electrons, 316 
 
 Electrophorus, 299 
 
 Electroscope, 295 ; described, 297 
 
 Elevation, effect of, on boiling-point, 
 79, 138; on temperature, 99, 100 
 
 Engineering aneroid, 137 
 
 England, thunderstorms in, 305 ; 
 distribution of rainfall in, 261, 
 396-399 ; mean temperatures of, 
 386-390; soils in, 400, 401 ; 
 scarlet fever in, 443, 445 
 
 Enteric fever and subsoil tempera- 
 ture, 109 ; and subsoil water, 
 364 
 fever and its seasonal prevalence, 
 
 431, 434 
 feverishness in, 457 
 
 " Epidemic constitution," 410, 416 
 influenza, 410-412 
 summer diarrhoea and subsoil 
 temperature, 106-109, 417 
 
 Equation of time, 268 
 
 Equator, 11 ; distance from the, as 
 
 influencing climate, 351 
 snow and ice under the, 353 
 
 Equinoctial gales, 394 
 points, 11 
 
 Equinoxes, 11 
 
 Equivalence, law of, 360 
 
 " Error of capacity," 122 
 
 Errors of parallax, in barometer 
 readings, 141 ; in rainfall read- 
 ings, 239 
 
 Etesian winds of south-eastern 
 Europe, 374 
 
 Etincdle de rupture, 93 
 
 Europe, thunderstorms in, 305 ; 
 climate of, 156, 157, 354, 367, 
 368, 378, 379 
 
 Evaporation, 166 ; coolness pro- 
 duced by, 167 ; force of, 184 
 investigations in England, 172 ; 
 in the United States, 172, 173, 
 174 
 
474 
 
 METEOROLOGY 
 
 Evaporation, anemometer, 277, 278 
 in London, 170 ; in Ireland, 171 
 
 " Evaporation " rains, 262 
 
 Evaporimeters, 167 ; de la Rue's, 
 169 ; Pickering's Patent Stan- 
 dard, 170 
 
 Excessive climates, 378 
 
 Expansion of air, 14 ; coefficient of, 
 15 
 
 Exposure of rain-gauge, 238 
 
 Extremes of atmospheric pressure, 
 120, 121 
 
 FACTORS, Glaisher's, 183, 184, 460 ; 
 Greenwich, 183 
 
 Fahrenheit's thermometer, 15, 76- 
 78, 80, 351 
 
 " Fall Fever," 431 
 
 Fauna, influence of climate on, 350, 
 351 
 
 Feather-vane anemoscope, Laugh- 
 ton's, 283 
 
 Fernley Observatory, Southport, 33, 
 34 
 
 Festooned cloud, 209 
 
 Fever, enteric, and subsoil tempera- 
 ture, 109 ; and subsoil water, 364 
 pneumonic, seasonal prevalence 
 of, 446 ; in Dublin, 446 
 
 Fiducial point in Standard Baro- 
 meter, 123, 140 
 
 Filling thermometer tubes, 80 
 
 Fineman's nephoscope, 213, 214 
 
 Fire-balls, electrical, 310 
 
 Fire-damp in air, 23 
 
 Fire, St. Elmo's, 312 
 
 Firn, 370 
 
 Fitzroy barometer, 125 
 
 Flag-signals in United States, 63, 64 
 
 Flash, the green, 459 
 
 Flies and diarrhceal diseases, 427- 
 430 
 
 Float pattern rain-gauges, 226, 233 
 
 Floods in American rivers, 60, 61 
 
 Flora, influence of climate on, 350, 
 351 
 
 Fog, 165, 190, 193 ; varieties of, 193 ; 
 
 formation of, 193, 194, 200 
 wet, 193 ; dry smoke, 193, 201 
 
 Fog densities, measurement of, 204 
 
 "Fog scoffers," 204 
 
 Fohn wind of north-east Switzer- 
 land, 261, 353, 354, 375; the 
 " snow-devourer," 354 
 
 Fomites, 427 
 
 Food-kesping, defective, and diar- 
 rhoea, 424 
 
 Force of wind, how determined, 7, 
 38, 39, 265, 266; Beaufort 
 Scale, 32, 36, 38-40, 266 
 evaporation, 184 ; electricity, 296 ; 
 
 electro-motive, 298, 300 
 Forchammer process, 24, 25 
 Forecast Division of United States 
 Weather Bureau, 45, 46, 48, 49, 
 53-61 
 Forecasting, cirrus observations an 
 
 aid to, 206 
 
 Forecasts, 10, 27, 31, 45, 46, 48, 49, 
 53-61, 64, 65, 66, 70, 71, 72-74, 
 114 
 
 in Canada, 72-74 
 distribution of, in United States, 
 
 61-65 
 Forests, influence of, on climate, 
 
 365-368 
 Forked lightning, 308 
 
 thermometer bulb, 86, 104 
 Formula for reduction of barometer 
 
 readings, 148, 149, 150 
 Fortin barometer, 122-124, 139, 143, 
 
 144 
 
 " Free air," 8, 265 
 Free water surfaces, 167, 193, 194 
 Freezing, mixtures, 76 
 
 point of water, 76, 79, 194 
 Friction, development of electricity, 
 
 by, 295 
 
 Frost of 1890-91, 164 
 glazed, 191, 192, 254 
 hard, in British Islands, 380, 381 
 Frozen air, 14 ; nitrogen, 14 
 
 rain, 192, 254 
 Fulgurite, 310 
 Fulminary tube, 310 
 
 GALES, classification of, 38, 39, 40, 
 
 290 
 
 equinoctial, 394 
 
 and strong winds compared, 290 
 Galvanic battery, Daniell's, 300 
 Gas, definition of a, 13 ; liquefaction 
 of, 14 ; compression of, 13, 14 ; 
 diffusion of, 23 
 Gases, laws relating to, 14, 23 
 
 condensed, in atmosphere, 197 
 Gauge, evaporation, de la Rue's 
 
 self-recording, 169 
 rain, history of the, 223-225 ; best 
 
 form of, 225 ; Hooke's, 224 
 float, 226, 233 ; storm, 226 ; Ne- 
 gretti and Zambra's, 228 ; self- 
 recording, 228-237 ; Crosley's 
 registering, 229 ; Yeates's elec- 
 
INDEX OF SUBJECTS AND PLACES 
 
 475 
 
 trie self-registering, 229 ; Ca- 
 sella's electrical, 229 ; advan- 
 tages of, 230, 232 ; Richard's 
 self-recording, 233 ; Meteoro- 
 logical Office, 225 ; Snowdon, 
 226 ; Bradford, 226 ; Halliwell's 
 Patent, 235 ; snow-melting, 253 
 position of, 238 
 Geographical distribution of rain, 
 
 260, 261, 395 
 Geological formation as influencing 
 
 British climate, 399 
 Givre, 191 
 Glaisher's hygrometrical tables, 183, 
 
 460 
 
 factors, 183, 184, 460 
 Glasnevin Observatory, 404, 407 
 Glass, weather, 119, 126 
 Glatteis, 192, 254 
 Glazed frost, 191, 192, 254 
 Globular lightning, 308, 310 
 Glossop Observatory, 306 
 
 Upper Air Investigation Station, 
 
 327, 333 
 Glycerine, specific gravity of, 116 
 
 barometer, 116, 117 
 Gold-leaf electroscope, 295 ; de- 
 scribed, 297 
 Gradients, barometrical, 7, 8, 265 
 
 thermometric, 380 
 Graduation of thermometer tubes, 
 
 81 
 
 barometer scale, 141-143 
 Graham's Law, 23 
 Grass-minimum thermometer, Ca- 
 
 sella's bifurcated, 104, 105 
 Graupel, 258 
 Gravity, correction for, in reading 
 
 the barometer, 143, 150 
 specific, 115, 117 
 
 Great Britain, rainfall in, 261 ; 
 thunderstorms in, 305, 306 ; 
 temperature in, 386-390 
 Green Flash, 459 
 Green snow, 249 ; its cause, 249 
 Greenwich Observatory, 28, 93, 119, 
 
 300 
 
 factors, 183 
 
 time, conversion of, to local, 268 
 Grele, 255 
 Gresil, 258 
 Ground fog, 205, 207 
 frost, 191, 192, 254 
 Grundwasser, 416 
 Gulf Stream, 359 ; its climatic effect, 
 
 359, 360 
 Gun Barometer, 125 
 
 Gusts, 290 
 
 Gyratory movement of air, 4, 6 
 
 HAGEL, 255 
 
 Hagemann's anemometer, 278, 289 
 Hail, 190, 249, 255 ; its formation, 
 255, 259, 260 ; soft, 258 ; weight 
 of, 258, 259 ; shape of, 259 ; its 
 relation to electricity, 296, 313, 
 314 
 
 Hailstorms, 255-260, 313, 369 
 Hair hygrometer, 178 ; used in 
 
 winter in Russia, 185 
 Halliwell's patent rain-gauge, 235 
 Halos, solar and lunar, 163, 206, 207 
 Hard frosts in British Islands, 380, 
 
 381 
 Harmattan of the west coast of 
 
 Africa, 374 
 Hauling of wind, 163 
 Haze, 199, 203 ; dust, 203 
 Health and weather, 2 
 
 influence of temperature of the 
 
 soil on, 106-109 
 Medical Officers of, 42 
 Heat, and electricity compared, 295 
 conduction of, 96 
 convection of, 96, 97 
 definition of specific, 356 
 how transmitted, 96-98 
 latent, 167 
 radiant, 17, 18, 97 
 radiation of, 96, 97, 98 
 specific, of mercury, 86 ; alcohol, 
 
 86 ; water, 100, 356 
 unit, 359 
 Heat-waves, 17, 18 ; thunderstorms, 
 
 302, 303 
 Height of the atmosphere, 17, 117, 
 
 118 
 
 of barometer, 125 
 of clouds, 205, 206 
 of columns of liquids or gases in- 
 versely proportional to their 
 specific gravity, 115, 117 
 Helicoid anemometer, 284 
 
 surface, meaning of the term, 284 
 Helio-pyrometer, 101-102 
 Helium in air, 22, 23, 347, 348 
 Hemispheres, Magdeburg, 17 
 Hemispherical cup anemometer, 274 
 High-pressure areas, 156, 160 
 Himalayas, climate of the, 353 
 Hoar frost, 190, 191, 220 
 Hohenrauch, 203 
 Holosteric barometer, 132 
 Hooke's rain-gauge, 224 
 
476 
 
 METEOROLOGY 
 
 Hot wind of Australia, 375 
 winds, 374, 375 
 
 House-flies and diarrhoeal diseases, 
 427-430 
 
 Humboldt's current, 361 
 
 Humidity, 42 
 
 absolute, 186, 199 
 and disease, 410 
 relative, 181, 184, 199 
 
 Humus, definition of, 363 ; influence 
 on climate, 363, 364 
 
 Hurricanes, West India, 162 
 
 Hydrogen sulphide in air, 19, 23 
 
 Hyetal equator, 260 
 
 Hyetograph, Negretti and Zambra's, 
 234 
 
 Hyetometry, 166, 190, 220 ; mean- 
 ing of term, 190 
 
 Hygrometers, 175 ; direct, 175 ; in- 
 direct, 175, 178 ; organic, 175 ; 
 inorganic, 175 ; Daniell's, 175, 
 183 ; Regnault's, 176 ; Dines' s, 
 
 177 ; an electrical dew-point, 
 
 178 ; de Saussure's, 178; chemi- 
 cal, 178, 179 ; a gravimetric re- 
 cording, 179 ; dry and wet bulb, 
 181 ; Mason's, 181 ; Crova's, 
 187, 188 ; history of, 186 
 
 Hygrometrical tables, 182, 183, 460, 
 
 461 
 
 Hygrometry, 166, 175 
 Hygroscope, 178 
 Hypsometer, 137, 138 
 
 ICE, melting-point of, 76-80, 194; 
 specific gravity of, 258, 259 
 
 Icebergs, 360, 361 
 as fog-producers, 202 
 
 Ice-crystals, 206, 207, 254 
 
 Iceland, winter thunderstorms in, 305 
 
 " Inbat," 385 
 
 Inch of rain, meaning of term, 263 
 
 Inclination anemometers, 279 
 
 Inclined currents, 279 
 
 Index error, 43, 143, 144, 147 
 
 Indian monsoons, 262, 372 
 
 Indirect hygrometers, 178 
 
 Induction, 297, 310, 312 ; pheno- 
 mena caused by, 310, 312 
 
 Infective diseases, acute, 409, 410, 
 431 
 
 Influence of season and weather on 
 disease, 409 
 
 Influenza, 410-412 
 
 Inland range of temperature, 358 
 
 Innsbruck International Meteoro- 
 logical Conference, 102, 321 
 
 Instrument Section of tne United 
 States Weather Bureau, 50 
 
 Instruments required for Second 
 Order Station, 42, 43 
 
 Insular climate, 354, 378, 386 
 
 Insulated, meaning of, 297 
 
 Insulation, 297 
 
 Intensity of cyclones, 162 
 electricity, 296 
 
 International - Balloon Ascents of 
 
 July, 1907, 329-332 
 Meteorological Committee, 38, 91, 
 
 212, 213, 252, 287, 296 
 Meteorological Conference, 8, 38, 
 42, 102, 181, 211, 216, 225, 252, 
 268, 270, 287, 296, 321 
 Code, 42 
 
 Inversion of temperature, 100, 331 
 
 Inverted cumulus cloud, 209 
 
 lonisation of atmospheric air, 245 
 
 Ionised gases in production of rain, 
 244 
 
 Ions, 245, 247 ; counting of, in rain- 
 drops, 246 
 
 Ireland, evaporation in, 171 ; rain- 
 fall in, 261, 396-399; thunder- 
 storms in, 305 ; climate of, 357 ; 
 mean temperatures of, 386-390 ; 
 soils in, 401 
 
 Irregular barometric variations, 151, 
 159, 160 
 
 Isobars, 7, 157, 160, 265 ; meaning 
 of term, 31 ; straight, 161 ; in 
 thunderstorms, 303 
 
 Isohyets, 221 
 
 Isothermal layer of the atmosphere, 
 
 100, 331, 336, 343, 345, 346 
 Isothermals, 379, 380 ; British, 383 
 Isotherms, 31, 379, 380; meaning 
 
 of term, 31 
 Ivy on houses, 368, 369 
 
 JAMAICA, thunderstorms in, 304 
 January mean temperature, 389 
 Jordan photographic sunshine re- 
 corder, 109, 111-113 ; improved 
 ditto, 113 
 glycerine barometer, 117 ; chart 
 
 of, 117 
 
 Journal of the Weather, Merle's, 26 
 July mean temperature, 390 
 
 KEW Observatory, 28, 37, 43, 82, 
 83, 101, 147, 276, 285, 300, 399 
 anemograph, 285, 287, 289 
 barograph, 93 
 
INDEX OF SUBJECTS AND PLACES 
 
 477 
 
 Kew Observatory barometer, 122- 
 
 125, 140, 144, 147 
 thermometer, 83 
 
 Khamseen, or Khamsin, of Egypt, 
 374 
 
 Khasi Hills, rainfall in the, 220, 262, 
 263 
 
 Kilda, St. , winter temperature in, 386 
 
 King's barograph, 12G, 127 
 
 Kite, electric, 306 
 
 Franklin's electric, 294, 325 
 
 Kite-stations, 327, 333 
 
 Koniscope, or dust-detector, Ait- 
 ken's, 198 
 
 Krypton, 22, 348 
 
 LAMBERT'S formula for mean direc- 
 tion of wind, 287 
 Lament's atmometer, 168 
 Land and sea breezes, 208, 209 
 and water, influence of, on cli- 
 mate, 352, 354-358 
 Land meteorology of the British 
 
 Islands, 27, 28 
 
 Latent heat, 167 ; its use, 167 
 Latitude, influence of, on climate, 
 
 350-353 
 Law of Buys Ballot, 4, 6, 9, 157, 158, 
 
 159, 265 
 
 Boyle and Mario tte, 14, 16 
 Charles, 15, 16 
 Graham, 23 
 Regnault, 15, 16 
 Stefan, 341 
 
 Laws of electricity, 295, 296 
 Leipzig, International Meteorologi- 
 cal Conference, 268 
 Lenkoran forest, 368 
 Leste of Madeira, 375 
 Leveche of the Iberian Peninsula, 374 
 Leyden jar, 299, 300 
 Library of the United States 
 
 Weather Bureau, 51, 52 
 Light, vibrations of, 17 
 
 electro-magnetic theory of, 293, 
 
 294 
 Lightning, 308, 309 ; what it is, 308 ; 
 
 its varieties, 308 
 
 zigzag or forked, 308, 309, 310 ; 
 diffused, 308, 310 ; summer, 308, 
 310 ; sheet, 308, 310 ; globular, 
 308, 310 ; ball, 308, 310 ; tubes, 
 310 
 
 rapidity of, 311 
 colour of, 312, 313 
 conductors, 317, 318 ; their prin- 
 ciple, 317 ; description of, 317, 
 
 318 ; conditions for an efficient, 
 318-320 
 
 Lightning-rod, 317 ; Conference, 318 
 
 Limestone soils, influence on cli- 
 mate, 400, 401 
 
 carboniferous, 401 ; oolitic, 401 ; 
 ancient, 401 
 
 Lind's anemometer, 273, 274, 289 
 
 Liquefaction of gases by cold, 14 
 
 Local time, conversion of Green- 
 wich to, 268 
 
 London clay, 401 
 
 London, diarrhoeal diseases in, 418, 
 423 ; mean temperature of, 388 ; 
 smallpox in, 437, 438, 440; 
 measles in, 441, 442 ; scarlatina 
 in, 443, 445 ; pneumonic fever 
 in, 449 ; bronchitis in, 449, 451- 
 455 
 evaporation in, 170 ; remarkable 
 
 rainfall in, 226 
 Fever Hospital, 435 
 Meteorological Office, 7, 28, 29, 30, 
 33, 37, 38, 40, 109, 128, 129, 
 148, 149, 150, 209, 225, 238, 
 266, 327, 348, 383, 388, 394 
 
 Lower clouds, 206, 207 
 
 Low-pressure areas, 4, 156, 160, 304 
 
 Luminous lightning, 309 
 
 Lunar halos, 163, 206, 207 
 
 Lungs, inflammation of the, 446, 448 
 
 MACKEREL sky, 206, 207 
 
 Magdeburg Hemispheres, 17 
 
 Magnetic bearings compared with 
 
 true bearings, 267 
 parallel, 314 ; storms, 314 
 storm of September 25, 1909, 316 
 
 Magnetism, terrestrial, and aurora, 
 314, 316 
 
 Malaria, 364, 365 
 
 Maps, distribution of bright sun- 
 shine, 37 
 Monthly Weather, Canada, The, 72, 
 
 74 ' 
 
 rainfall, 67, 221, 395, 396 
 weather, 27, 31, 32, 45, 51, 56, 57, 
 67, 70, 71, 72, 73 
 
 Mares' tails, 205 
 
 Marine barometer, 124 
 
 Marl, 422 
 
 Marsh-gas in air, 23 
 
 Marshy soils, influence of, on health, 
 364 
 
 Mason's hygrometer, 181 
 
 Matterhorn, 203 
 
 Maximum thermometers, 83-85 
 
478 
 
 METEOROLOGY 
 
 Mean, direction of wind, Lambert's 
 
 formula for, 287 
 temperature, 93, 94, 95 ; of British 
 
 Islands, 386 
 reduction of, to sea-level, 387, 388, 
 
 390 
 
 and disease, 410 
 coefficients, 94, 95 
 how obtained, 94-95 
 Meandering lightning, 309 
 Measles, seasonal prevalence of, 440 ; 
 
 in London, 441 
 Measurement of altitudes, 136, 137, 
 
 150 
 
 fog densities, 204 
 rainfall, 238, 239 
 snow, 252, 253 
 Mechanical barograph, Redier's, 
 
 128 
 
 Medical Officers of Health, 42 
 Medicine and meteorology, 2 
 
 State, Diploma in, 3 
 Melting-point of ice, 76-80 
 Meniscus, meaning of the term, 
 
 140 
 Mercurial minimum thermometer, 
 
 85, 86, 87-89 
 barometer, 116 
 
 Mercury, specific heat of, 86 ; 
 specific gravity of, 116 ; as a 
 medium for indicating tempera- 
 ture, 77, 82, 85, 86 ; atmo- 
 spheric pressure, 115 
 Merle's Journal of the Weather, 26, 
 
 27 
 Metallic dust in air, 24 ; barometer, 
 
 133 
 
 Meteoric dust,' 197 
 Meteorographs, 129, 327 ; Dines's 
 
 light, 329, 330 
 
 Meteorological Atlas of the British 
 Isles, 383, 388, 389, 392, 393, 
 396 
 Committee, 33, 93, 149, 183, 289, 
 
 292, 296, 334, 383, 388, 396 
 Committee, International, 38, 91, 
 
 212, 213, 252, 287 
 Conference, International, 8, 38, 
 42, 102, 181, 211, 216, 225, 268, 
 270, 287, 296, 321 
 information bearing on public 
 
 health, 331 
 
 observations, British, 26, 28 
 Office, London, 7, 28, 29, 30, 33, 
 37, 38, 40, 109, 128, 129, 148, 
 149, 150, 209, 225, 238, 266, 
 327, 348, 383, 388, 394 
 
 Meteorological Service, Canadian, 
 
 69-74 
 
 Society, Royal, 28, 29, 33, 37, 89, 
 92, 101, 109, 118, 119, 133, 169, 
 171, 179, 183, 193, 223, 224, 
 250, 251, 270, 280, 284, 285, 
 287, 289, 290, 295, 303, 306, 
 308, 311, 312, 315, 318, 331, 
 333, 369, 373, 377, 382 
 Society, Scottish, 28, 29, 33, 109, 
 
 305, 383, 386, 387 
 Year-Book, The British, 37, 38 
 
 Meteorological Record, the, 37 
 
 Meteorology, meaning of the term, 1, 
 2 ; " Dawn of," a lecture by Dr. 
 G. Hellmann, 1, 75 ; relations 
 existing between medicine and, 
 2 ; a science, 10 ; practical, 26 
 
 Methane in air, 23 
 
 Microbes in air, 24, 425, 426, 430 
 
 Micrococci of pneumonia, 447, 456 
 
 Micro-organisms in air, 24, 425, 426, 
 430 
 
 Mineral constituents of the atmo- 
 sphere, 24 
 
 Minimum thermometers, 83, 85-89, 
 104, 105 
 
 Mist, 190, 193; formation of, 193,207 
 Scotch, 193 
 
 Mistral of the Gulf of Lions, 373 
 
 Mists, cholera blue, 416 
 
 Model illustrating investigation of 
 upper atmosphere, 336, 338 
 
 Moderate climates, 378 
 
 Molecular analysis of air, 13 
 
 Monsoons, Indian, 262, 372 
 
 Monte Rosa, fall of snow on, 250 
 
 Monthly, summary of weather, 34, 36 
 Weather Report, The, 34, 36, 37 
 Weather Review, United States, 
 
 49, 51, 174, 260, 363, 466 
 Weather Review, Canada, 71, 72 
 Weather Map, Canada, 72, 74 
 
 Moon, watery, 163 
 
 Mosquitoes, 364, 365 
 
 Mountain, determination of height 
 
 of, 150 
 
 cloud-capped, 202 
 ranges, effect of, on climate, 352, 
 362, 363 
 
 Munich, International Meteoro- 
 logical Conference, 211, 216, 
 252, 287 
 
 Musical anemometers, 284 
 
 NASCENT oxygen, 323 
 Nature'sabhorrenceof a vacuum, 115 
 
INDEX OF SUBJECTS AND PLACES 
 
 479 
 
 Negative, electricity, 296, 302, 305, 
 
 306 
 
 oxygen, 322 
 Negretti and Zambia's maximum 
 
 thermometer, 84 
 hyetograph, 234 
 
 improved Robinson's anemo- 
 meter, 275, 276 
 storm rain-gauge, 228 
 Neon, 22, 347, 348 
 Nephometer, spherical mirror, Bes- 
 
 son's, 218, 219 
 Nephoscope, 213 
 reflecting, 213, 214 
 direct-vision, 213, 214 
 Fineman's, 213, 214 
 Besson's comb, 214, 215 ; zenith, 
 
 217, 218 
 methods of stating the results of 
 
 observations by, 215 
 Neutral point, in barometer scale, 
 
 144 
 
 in cloud-slopes, 210 
 Neve, 370 
 
 Newfoundland fogs, 202 
 Niederschlage, 19 
 
 Nimbus, 206, 207, 208, 210, 212; 
 different applications of the 
 term, 210, 211 
 Nitric acid in air, 19, 23 
 Nitrogen in air, 19-21 
 Non-periodic barometric fluctua- 
 tions, 151, 159, 160 
 Normal climatological stations, 28, 
 30, 42, 43, 94, 107, 134, 147, 
 238, 382, 403 
 
 instruments required for, 42, 43 
 Nortes (Northers) of the Gulf of 
 
 Mexico, 373 
 North-east trade winds, 159 
 
 OBSERVATIONS, meteorological, 26, 
 27 ; synchronous, 114, 119, 143, 
 183 
 
 barometer, 139 
 
 cloud, 211 
 
 nephoscope, 215 
 
 phonological, 26, 33 
 
 rainfall, 223, 224 
 
 Observatories, self-recording, 28 ; 
 normal climatological, 28, 30, 
 42, 43, 94, 107, 134, 147, 238, 
 382, 403; auxiliary climato- 
 logical, 29 
 
 Observatory, Kew, 28, 37, 43, 82, 
 83, 101, 147, 276, 285, 300, 399 
 
 Toronto, 69, 74 
 
 Observatory, Greenwich, 28, 93, 
 
 300 
 
 Fernley, Southport, 33, 34 
 Glossop, 306 
 
 Occasional cold winds, 372 
 warm winds, 372 
 
 Ocean currents, affecting climate, 
 
 100, 352, 358-361 
 warm, 359 ; cold, 360, 361 
 influence of, on climate, 100, 352, 
 356, 384 
 
 October mean temperature, 390 
 
 Office, Meteorological, London, 7, 
 28, 29, 30, 33, 37, 38, 40, 109, 
 128, 129, 148, 149, 150, 209, 
 225, 238, 266, 327, 348, 383, 
 388, 394 
 
 Officers of Health, Medical, 42 
 
 Ohm, 306 
 
 Opacity, 18 
 
 Open-scale barometer, Bartrum's, 
 131, 132 ; Barrett's, 138 
 
 Ordinary thermometers, 82, 83 
 
 Ordnance Survey, Datum, 148 
 
 Office, Phcenix Park, 404, 407, 
 415 
 
 Organic matter in the air, tests for, 
 24, 25 
 
 Osier's anemometer, 271, 287 
 
 Oxshott, anemometer comparisons 
 at, 271 
 
 Oxygen, liquid, 14 ; in air, 19-21 ; 
 allotropic, 321 ; negative, 322 ; 
 positive, 322 ; nascent, 323 ; 
 active, 323 
 
 Ozone, 22, 321 ; theories as to its 
 nature, 321, 322, 323 ; pro- 
 perties of, 323, 324 ; influence 
 on disease, 323, 324 ; tests for, 
 324 ; and cholera, 415 
 
 Ozonides, 322 
 
 PALLIUM, 207 
 
 Pampero of Brazil, 373, 374 
 
 Pandemic waves, 416 
 
 Parachute, principls of the, 13 
 
 Parallax, errors of, in barometer 
 readings, 141 ; in rainfall read- 
 ings, 239 
 
 Paratonnerre, 317 
 
 Paris, cholera in, 413, 414 ; Inter- 
 national Meteorological Com- 
 mittee, 252 
 
 Peaty soils, influence of, on health, 
 364 ; on climate, 401 
 
 Pebble beds and climate, 400 
 
 Peltier's electrometer, 301 
 
480 
 
 METEOROLOGY 
 
 Pendulum anemometers, 270 
 Percolation and rainfall, 171, 172 
 evaporation and condensation, 
 
 171, 172 
 Perihelion, 99 
 Periodic, barometric fluctuations, 
 
 151 
 thunderstorms, 209, 304 ; winds, 
 
 372, 373 
 Periodicity of temperature, 408 ; of 
 
 rainfall, 408 ; of atmospheric 
 
 pressure, 408 
 Permanent winds, 372 
 Peruvian current, 361 
 Petersburg, cholera in St., 413, 415 
 Phenological observations, 26, 33 
 Phillips' s maximum thermometer, 
 
 83,93 
 Phoenix Park Ordnance Survey 
 
 Office, 404, 407, 415 
 Photogrammeters, 216 
 Photographic thermograph, 92, 93 
 barograph, 128 
 electrometer, 300 
 sunshine recorder, 109, 111 
 Photographs of thunder-clouds, 211 
 Phthisis and subsoil drainage, 364 
 Physical cause of rain, 242 
 Physicians and meteorological in- 
 formation, 391 
 
 Pickering's evaporimeter, 170 
 Piesmic barometer, 134, 135 
 Pitot's tube, 273 
 Pneumonia, 446 ; its claims to rank 
 
 as a specific fever, 456 
 Pneumonic fever, seasonal preva- 
 lence of, 446 
 in Dublin, 446, 447 
 Pocky-cloud, 209 
 Poison of diarrhceal diseases, 425 
 Pole-star, position of the, 268 
 Portable electrometer, Thomson's, 
 
 299, 300 
 Positive electricity, 296, 302, 306 
 
 oxygen, 322 
 Potential, electrical, 296, 298, 311 ; 
 
 difference of, 298 
 Practical meteorology, 26 
 Practice forecasts, 59 
 Precipitation, 220, 369 
 Pressure, atmospheric, 17, 114 
 plate anemometer, 271 
 on a fluid anemometer, 273 
 Prevailing winds in relation to 
 
 climate, 352, 372 
 classification of, 372 
 Protococcus nivalis, 249 
 
 Proximity of mountain ranges, 362 
 Psychrometer, 181 ; ventilated, 185 
 Public Health, Diploma in, 3 
 
 influence of temperature of the 
 
 soil on, 106 
 
 meteorological information bear- 
 ing on, 391 
 
 Publications, Division of the United 
 States Weather Bureau, 50, 51 
 " Pumping " in a barometer, 124 
 Pyrheliometer, of Pouillet, 98, 102 
 Angstrom's electric compensation 
 
 actinometer and, 102 
 Pyrton Hill Upper Air Investigation 
 
 Station, 32, 333, 334, 349 
 Pythogenic pneumonia, 446, 447 
 
 QUADRANT electrometer, Thomson's, 
 
 299-300 
 
 " Quicksilver, Cylinder of," 119 
 Quito, climate of, 353 
 
 RADIANT heat, 17, 18, 97 
 Radiation fog, 200 
 
 of heat, 96-98 ; terrestrial, 98-100, 
 102, 104, 105; solar, 98, 99, 
 101-104, 109 
 
 thermometers, 82, 93, 101-105 
 " Radiation weather," 162 
 Radio-Integrator, the Wilson, 104 
 Rain, 166, 190 ; its chief physical 
 cause, 242 ; atmospheric elec- 
 tricity in production of, 243- 
 246 ; its distribution, 249, 260- 
 263 
 
 cloud, 206, 210 
 coloured, 247-249, 264 
 frozen, 192, 254 
 
 inch of, meaning of term, 263, 264 
 prognosis of, by means of clouds, 
 
 207 
 
 weight and bulk of, 263 
 Rain day, definition of a, 239, 369 
 Rain-drops, size of, 241, 242 
 Rainfall, 42, 67, 221, 223 ; and dis- 
 ease, 410 ; weighing the, 224 ; 
 measurement of, 238, 239 
 British, 33, 37, 170, 239, 261, 395 
 and diarrhoea, 422 
 how affected by altitude, 239, 240, 
 
 241 
 in relation to climate, 261, 262, 
 
 352, 369 
 
 mean annual of the world, 221 
 land, 222 
 
 continental and ocean, 223 
 observations, 33, 223, 224. 395 
 
INDEX OF SUBJECTS AND PLACES 
 
 481 
 
 Rainfall, of the British Islands, 261, 
 
 395 
 
 periodicity of, 408 
 tables prepared by Mr. G. J. 
 
 Symons, 396 
 torrential, 369 
 
 Rain-gauges, history of, 223-225 ; 
 diameters of, 225 ; best form 
 of, 225 ; varieties of, 225-238 ; 
 snow-melting, 253 
 exposure of, 238 
 Ramified lightning, 309 
 Rarefaction of air at great altitudes, 
 
 16 
 
 Rauhfrost, 191 
 
 Reaumur thermometer scale, 77-80 
 Recorders, sunshine, 109-113 
 Records Division of the United 
 States Weather Bureau, 49, 50 
 " Red rain," 249 
 Red snow, 249 ; its cause, 249 
 Registering, rain-gauges, 228-237 
 
 thermometers, 83 
 Registration of disease, 455 
 Regnault's law, 15-16 
 
 direct hygrometer, 176, 177 
 Regular barometric variations, 151 
 Relative humidity, 181, 184, 199 
 
 mode of determining, 184 
 Replenisher, electrical, 300 
 Report, The Daily Weather, 31, 32, 
 
 34,38 
 
 The Weekly Weather, 34, 35 
 The Monthly Weather, 34, 36, 37 
 Report on atmospheric electricity, 
 
 296 
 
 Reporting Stations, British Tele- 
 graphic, 29, 30 ; United States, 
 52 
 Reports, daily weather, 31, 32, 34, 
 
 35 ; weekly weather, 34, 35 
 Repulsion, capillary, 140 ; electrical, 
 
 295 
 Return shock in thunderstorms, 
 
 310, 311 
 
 Review, United States Monthly 
 Weather, 49, 51, 67, 174, 260, 
 363, 466 
 Canadian Monthly Weather, 71, 
 
 72 
 
 Ribbon lightning, 309 
 Richard's anemo - cinemographe, 
 285 ; actinometer, 102 ; self- 
 recording rain-gauge, 233 ; 
 brontometer, 320 ; statoscope, 
 320, 321 
 Rime, 191 
 
 Rio Janeiro, periodic thunderstorms 
 at, 305 
 
 River, and Flood Section of the 
 United States Weather Bureau, 
 53, 60, 61 
 and flood work in United States 
 
 Weather Bureau, 60, 61 
 fog, 200 
 
 " Roaring Forties," 266 
 
 Robinson's cup anemometer, 271, 
 274-276, 287, 289 
 
 Roll-cumulus, 209 
 
 Rome, International Meteorological 
 
 Committee, 91 
 
 International Meteorological Con- 
 ference, 225 
 
 Rontgen Rays, 244 
 
 Rosee, 191 
 
 Rotation of earth on its axis, 4, 11, 
 12, 151 
 
 Royal Meteorological Society, 28, 
 29, 33, 37, 89, 92, 101, 109, 118, 
 U9, 133, 169, 171, 179, 183. 
 193, 223, 224, 250, 251, 270, 
 280, 284, 285, 287, 289, 290, 
 295, 303, 306, 308, 311, 312, 
 315, 318, 331, 333, 369, 373, 
 377 382 
 
 Observatory, Greenwich, 28, 93, 
 119, 300 
 
 Rutherford's spirit minimum ther- 
 mometer, 85, 86, 104 
 
 SALT, common, in atmosphere, 24, 
 
 198 
 
 dust, 197 
 Sands, their influence on climate, 
 
 400 
 
 Sandstones, 400 
 Sanitary Science, Diploma in, 3 
 Saturation, 167 ; percentage degree 
 
 of, 178-181, 184 
 
 de Saussure's hair hygrometer, 178 
 
 Scale, Beaufort, of Wind Force, 32, 
 
 36, 38, 39, 40, 266 ; of Weather, 
 
 32 
 
 barometer, 143 ; adjustment of, 
 
 Wallis's, 123, 124 
 for Amount of Cloud, 211 
 of Sea Disturbance, 41, 42 
 thermometer, 76-80, 82 
 vernier, 140-143 
 weather, 41 
 Scarlatina, seasonal prevalence of, 
 
 434, 442, 443 
 Schists, 400, 402 
 '* Schneefresser," 354 
 
 31 
 
482 
 
 METEOROLOGY 
 
 " Schneegestober," 252 
 
 " Schneetreiben," 252 
 
 Scientific work of the United States 
 
 Weather Bureau, 66 
 Scilly Isles, climate of, 356, 388, 389, 
 
 390 
 Scirocco of North Africa and Sicily, 
 
 374 
 
 Scotch mist, 193 
 Scotland, subsoil temperature in, 
 
 365 ; temperature of, 386-390 ; 
 
 distribution of rainfall in, 261, 
 
 396-399 ; soils in, 400 
 Scottish Meteorological Society, 28, 
 
 29, 33, 109, 305, 383, 386, 387 
 Journal of the, 37, 383, 387 
 Screen, thermometer wall, 91 
 Scud-cloud, 207, 211 
 Sea breezes and land breezes, 208, 209 
 Sea disturbance, Scale of, 41, 42 
 Sea-fogs, 193, 202 
 Sea-Forecast Districts (British 
 
 Isles), 383 
 Sea-level, reduction of barometer 
 
 readings to, 148 ; of tempera- 
 ture to, 390 
 Sea-temperature stations round 
 
 British Islands, 383 
 Sea temperatures round British 
 
 Islands, 356, 383 ; in the Atlan- 
 tic, 360, 361 
 
 Sea -water, electrical state of, 302 
 freezing-point of, 357 
 maximal density of, 357 
 Searchlight signals in meteorology, 65 
 Season, influence of, on disease, 409 ; 
 
 on rainfall, 240 
 Seasonal distribution of rainfall, 
 
 260, 262 ; thunderstorms, 305, 
 
 306 ; disease, 409 
 Secondary barometric depression, 
 
 161, 304 
 
 Self-recording observatories, 28 
 anemometer, Casella's, 269, 289 
 barometers, 126 ; Dines's, 135 
 electrometer, 300 
 evaporation-gauge, Richard's, 169 
 rain-gauges, 228-237 
 thermometers, 82, 92, 93 
 Serein, 191 
 Shales, 400, 401 
 Sheet-cloud, 205, 206, 207 
 
 lightning, 308, 310 
 Shiga bacillus, 425, 426 
 Shock, return, in thunderstorms, 
 
 310, 311 
 Shower-cloud, 206, 210 
 
 " Showers of blood, 247 
 of sulphur," 247, 248 
 remarkable, 248 
 
 Siberia, excessive climate of, 157, 
 354, 355, 367, 378 
 
 Signal-Officer, United States Chief, 
 45, 46, 52 
 
 Signals, storm and cautionary, 62, 64 
 cannon-firing, 64 
 flags, 63, 64 
 railway train, 64 
 searchlight, 65 
 telephone, 65 
 whistle, 64 
 
 Silver thaw, 191, 192, 254 
 
 Simoom of Arabia, Kutchee, and 
 Upper Scinde, 375 
 
 Siphon barometer, 123, 125, 144 
 
 Six's thermometer, 86, 87 
 
 Sky, mackerel, 206, 207 
 
 Slates, 400, 402 
 
 Sleet, 249, 253 ; definition of, 253 
 
 Sling thermometer, 92 
 
 Slopes, cloud, in cumulus, 209, 210 
 
 Smallpox, seasonal prevalence of, 
 437 ; in Dublin, 439 ; in Sweden, 
 439 ; in London, 437, 438, 440 
 
 Smoke-fog, dry, 193, 201 ; wet, 200, 
 201 ; as a deodoriser, disinfec- 
 tor, and antiseptic, 201 
 
 Snow, 190, 249 ; its properties, 249 ; 
 its measurement, 252, 253 ; fall 
 of, on Monte Rosa, 250 
 red, 249 ; green, 249 ; dust, 252 
 
 Snow-covering, 105, 249, 369 
 effects of, on climate, 369, 370 
 
 Snow-crystals, 249, 250, 251 
 
 " Snow-devourer," the, 354 
 
 Snow and drift warnings by Cana- 
 dian Meteorological Service, 73, 
 74 
 
 Snowdon rain-gauge, 172, 226 
 
 Snowfall, limit of, 369 
 
 Snowflakes, 249 
 
 Snow forecasts, 73 
 
 Snow-line, mountain, 353 
 
 Snow-melting, mechanism of, 370, 
 371 
 
 Snow-rollers, 251, 252 
 
 Society, Royal Meteorological, 28, 
 29, 33, 37, 89, 92, 101, 109, 118, 
 119, 133, 169, 171, 179, 183, 
 193, 223, 224, 250, 251, 270, 
 280, 284, 285, 287, 289, 290, 
 295, 303, 306, 308, 311, 312, 
 315, 318, 331, 333, 369, 373, 
 377, 382 
 
INDEX OF SUBJECTS AND PLACES 
 
 483 
 
 Society, Scottish Meteorological, 28, 
 29, 33, 109, 305, 383, 386, 387 
 Sodium chloride in air, 24, 198 
 Soft hail, 258 
 
 Soil, temperature of the, 105, 364, 
 
 365 ; its relation to health, 106- 
 
 109, 416, 417 ; its relation to 
 
 climate, 352, 363-365, 399-402 
 
 a cause of diarrhceal diseases, 422, 
 
 423 
 
 Solano of Spain, 374 
 Solar halos, 163, 206, 207 
 
 intensity apparatus, Padre 
 
 Secchi's, 102 
 
 radiation, 98, 99, 101, 109 ; how 
 measured, 101 ; thermometers, 
 101-104 
 
 Solstice, meaning of term, 12 
 South-east Trade Winds, 159 
 Southport International Meteoro- j 
 
 logical Committee, 287 
 Observatory, 33, 34 
 Specific gravity of mercury, 116 ; of j 
 
 water, 116; of ice, 258, 259 
 gravity, heights of columns of 
 liquids or gases inversely pro- 
 portional to their, 115, 117 
 heat denned, 356 
 heat of mercury, 86 ; alcohol, 86 ; ! 
 
 water, 100 
 
 Spectre of the Brocken, 203 
 Spectroscopy in meteorology, 67 
 Spirit minimum thermometer, 85, 86 
 Spring fog, 202 
 Springs, influence of forests on, 366, 
 
 367 
 
 Stand, thermometer, 43, 84, 89-92 
 Standard, thermometers, 42, 82 
 barometer, 123 
 
 temperature in barometric obser- 
 vations, 147 
 
 State Medicine, Diploma in, 3 
 State Weather Service Division of 
 United States Weather Bureau, 
 49,64 
 
 Stations, of the First Order, 28, 128 
 Anemograph, 29 
 Extra Reporting, 30 
 Foreign Reporting, 29 
 of United States Weather Bureau, 
 
 52, 54 
 
 Phonological, 33 
 Sea-temperature, round British 
 
 Isles, 383 
 
 Second Order, 28, 30, 42, 43, 94, 
 107, 134, 147, 238, 382, 403; 
 instruments required for, 42, 43 
 
 Stations, Telegraphic Reporting, 29, 
 
 30, 52 
 
 Third Order, 29, 30 
 Upper Air Investigation, 32, 327, 
 
 331, 333 
 
 working in connection with the 
 Meteorological Office, London, 
 30 
 
 Statoscope, Richard's, 320, 321 
 Stefan's Law, 341 
 St. Elmo's Fire, 312 
 St. Petersburg, cholera in, 413, 415 
 " Step " anemometer, 275, 276 
 Stevenson, thermometer stand, 43, 
 
 84, 89, 90 
 Storm of February, 1903, 13, 290, 
 
 291 
 
 rain-gauge, 226 
 signals, 62, 64, 67 
 warnings, 62, 64, 65, 67, 73, 74 
 Storms, classification of, 290 
 
 magnetic, 314, 316 
 Straight isobars, 161 
 Strato-cumulus cloud, 212 
 Stratosphere, 336, 343 
 Stratus clouds, 205, 207, 208, 212 
 Stream lightning, 309 
 Subsidiary barometric depression, 
 
 161, 304 
 Subsoil water, influence of, on 
 
 health, 364, 416, 417, 431, 432 
 temperature (in Trinity College, 
 Dublin), 406 ; influence of, on 
 health, 106-109, 416, 417, 422, 
 423, 434 
 
 Suction anemometers, 278 
 Sulphurous acid in air, 19, 23 
 Summer, diarrhoea and subsoil tem- 
 perature, 106-109, 417 
 climate of British Islands, 379 
 distribution of atmospheric pres- 
 sure, 10, 156 ; of thunderstorms, 
 305 ; of disease, 417 
 rams on the Continent of Europe, 
 
 262 
 
 haze, 203 ; lightning, 308, 310 
 sea-temperatures, 385 
 Sun, watery, 163 ; southing of the, 
 
 268 
 
 " Sunless day," 113 
 Sunshine, duration of bright, 109 ; 
 in Trinity College, Dublin, 405 ; 
 measurement of, 109-113 
 recorders, 109-113 
 Sun-spots, and weather, 314 
 and magnetic storms, 314 
 Supersaturated air, 195-197 
 312 
 
484 
 
 METEOROLOGY 
 
 Surface, configuration of, effect on 
 
 climate caused by, 357 
 trajectories of moving air, 291, 
 
 292 
 
 Sweden, smallpox in, 439 
 Symons's brontometer, 320, 321 
 Sympiesometer, 137 
 Synchronous observations, 114, 119, 
 
 143, 183 
 Synoptic Weather Charts, 27, 35, 
 
 160 
 
 TELEGRAPH, section of the Forecast 
 
 Division of the United States 
 
 Weather Bureau, 49 ; lines in 
 
 United States, 49 
 
 Telegraphic Reporting Stations, 29, 
 
 30, 52 ; in Canada, 70, 71, 72 
 transmission of barometric indica- 
 tions, 130 
 Weather Code, United States, 54, 
 
 55 
 
 Telegraphy, application of, to 
 weather study, 26, 27, 49, 70, 
 72-74 
 
 Telephone in meteorology, 65 
 ' ' Telluric influences ' ' in spread of 
 
 disease, 416 
 
 Temperature, accumulated, 36, 351 
 air and its measurement, 42, 75, 
 
 81 
 and diarrhoea, 107, 420, 421 ; and 
 
 enteric fever, 431 
 anemometer, 277 
 base, 35 
 correction for, in reading the 
 
 barometer, 122, 143, 147, 148 
 earth, 105, 365, 416 
 effect of elevation on, 99, 100, 
 352 ; effect of forests on, 367, 
 368 
 
 " inversion " of, 100, 331 
 lowest mean monthly, recorded, 
 
 355 
 mean, 93-95 ; average mean, 94 ; 
 
 and disease, 410 
 of sea round British Islands, 356, 
 
 383 
 
 periodicity of, 408 
 Tempest of February, 1903, 13, 290, 
 
 291 
 Tension of, aqueous vapour, 151, 
 
 152, 165, 175, 181, 186 
 electricity, 290, 298 
 Terling, enteric fever at, 432 
 Terrestrial radiation, 98-100 ; how 
 measured, 102, 104, 105 
 
 Tests for ozone, 323, 324 
 Thaw, silver, 191, 192, 254 
 
 cirrus cloud a sign of, 206 
 Theory of the winds, 4, 5, 6 
 Thermal unit, 359 
 Thermograms, 94 
 Thermographs, 92, 93 
 Thermometer, 75 ; meaning of word, 
 75 ; construction of a, 80-81 ; 
 description of a, 80 
 attached, 139 
 black-bulb, in vacuo, 101 
 bright-bulb, in vacuo, 101 
 dry-bulb, 148, 181 
 earth, 96, 105, 106 
 hypsometer, 137, 138 
 maximum, 83-85 
 minimum, 83, 85-89 
 ordinary, 82, 83 
 radiation, 82, 93, 101-105 
 registering, 83, 86, 87 
 scales, 76-80, 82 
 self-recording, 82, 92, 93 
 solar radiation maximum, 101-104 
 stand, 43, 84, 89-92 
 standard, 82 
 wet-bulb, 181 ; management of, 
 
 184, 185 
 Bourdon tube, 327 ; bimetallic, 
 
 328 
 
 Thermometre fronde, 92 
 Thermometric gradient, 380 
 Thermoscope, Philo's, 75, 76 
 Thomson's water-dropping collector, 
 
 298 ; electrometers, 299 
 Thunder, 208, 211, 255, 308 ; what 
 it is, 308 ; its relation to hail, 
 255 ; to electricity, 296 
 Thunderbolts, non-existent, 310, 
 
 311 
 Thunderclouds, characteristics of, 
 
 211 
 Thunderstorm, depression, 255, 303, 
 
 304 
 at Mere, Wilts, May 13, 1906, 309, 
 
 310 
 
 Committee of the Royal Meteoro- 
 logical Society, 303, 308 
 Thunderstorms, periodic, 209, 304 ; 
 heat, 302, 303 ; cyclonic, 302, 
 313 ; Mohn's classification of, 
 302 ; analysis of, 303 ; occur- 
 rence of, 369 
 Time, Greenwich, conversion of, to 
 
 local, 268 
 
 Tormentos of the Argentine Re- 
 public, 373 
 
INDEX OF SUBJECTS AND PLACES 
 
 485 
 
 Tornadoes, 255-257, 375 
 
 Toronto Meteorological Register, 
 
 71,72 
 
 Torrential rainfalls, 369 
 Torricellian column, 118, 119 
 experiment, 115, 117, 118 
 vacuum, 115, 122, 125 
 Torsion anemometer, 271 ; balance 
 electrometer, Coulomb's, 298, 299 
 " Tourmente de neige," 252 
 Town fog, 200-202 
 Trade Winds, 159, 372 
 Trajectories of moving air, 291, 292 
 Tramontana of the Adriatic, 373 
 Transmission of barometric indica- 
 tions by electricity, 130 
 Transparency of the atmosphere, 17, 
 
 199 
 
 Trappes Observatory, 327, 346 
 Trees and lightning, 311 
 Trinity College, Dublin, enteric fever 
 
 in, 432, 433 
 Normal Climatological Station, 
 
 107, 403 
 tornado in, 257 
 Tropics, meaning of term, 12 ; di- 
 
 u;-nal variation of barometer in 
 
 the, 151 
 
 Troposphere, 336, 343 
 Tube anemometer, 289 
 Twilight, 16, 17 
 Typhoid fever, seasonal prevalence 
 
 of, 431, 434 
 Typhoons, 162 
 Typhus fever, seasonal prevalence of, 
 
 435-437 
 
 UNDERGROUND water and health, 
 
 364, 416, 417, 431, 432 
 thermometer, 96, 97, 105 
 United States Weather Bureau, 44- 
 
 68,70 
 Monthly Weather Review, 49, 51, 
 
 67, 174, 260, 363, 466 
 its scientific work, 66 ; its sta- 
 tions, 52, 54 
 service created, 45 ; its cost, 48 ; 
 
 its publications, 50, 51 
 Units, absolute or metric, in upper- 
 air investigation, 348, 349 
 Universal Sunshine Recorder, 109, 
 
 111 
 
 " Upbank thaw," 100 
 Upper Air, British Stations for In- 
 vestigation of, 32, 333 
 Foreign Stations for Investigation 
 of, 325, 327, 331 
 
 Upper Atmosphere, Investigation of 
 
 the, 325-349 
 history of, 325-328 
 instruments for use in, 185, 328- 
 
 330 
 absolute or metric units in, 348, 
 
 349 
 absorption of solar radiation in, 
 
 341 
 
 composition of air in, 347, 348 
 Upper, clouds, 206 ; atmosphere. 
 
 electrical state of, 306 
 Upsala International Meteorological 
 
 Committee, 212, 213 
 Utrecht International Meteoro- 
 logical Congress, 38, 42 
 
 VACUUM, Torricellian, 115, 122, 125 
 
 Valleys, temperature in, 358 
 
 Vane, 267 
 
 Vapour, aqueous, in the air, 19, 25, 
 165, 166, 354 ; tension of, 151, 
 152, 165, 175, 181, 186 ; atmo- 
 sphere of, 165, 175, 190, 205, 
 220 ; amount of, required for 
 saturation, 181 ; where most 
 abundant, 186 
 
 Variable winds, 372 
 
 Vartry Water Supply, Dublin, 432 
 
 Veer, meaning of, 163 
 
 Vegetation in relation to climate, 
 351, 352, 365-369 
 
 Velocity, of wind, 7, 266 ; in storm of 
 
 February, 1903, 291 
 anemometers, 274 
 
 Venice, hailstorm in, 257, 258 
 
 Ventilated psychrometer, 185 
 
 Verglas, 192, 254 
 
 Verification Certificate, Kew Ob- 
 servatory, 82 
 
 Vernier in the barometer, 140-143 ; 
 method of reading it, 141-143 
 
 V-shaped depression, 161 ; thunder- 
 storms in, 162 
 
 Vibrations of light, 17 
 
 Vienna International Meteorological 
 Conference, 8, 181, 225, 268, 
 270, 287, 296 ; Committee, 38, 
 91, 296 
 
 Volcanic dust, 197 
 
 Volta's collector, 297 
 
 Volts, 300, 311 ; meaning of, 300 
 
 Volumetric analyses of air, 19, 20 
 
 WALES, distribution of rainfall in, 
 
 396, 397 
 Wall screen, thermometer, 91 
 
486 
 
 METEOROLOGY 
 
 Water, boiling-point of, 76-79, 138, 
 
 194 
 
 dust, 205 
 
 free surfaces of, 167, 193, 194 
 freezing-point of, 76, 79, 80, 81, 
 
 194 
 
 influence of, on climate, 352, 356 
 specific heat of, 100 ; specific 
 
 gravity of, 116 
 subsoil, influence of, on health, 
 
 364, 416, 417, 431, 432 
 underground, 364, 416, 417, 431, 
 
 432 
 
 Watery moon, 163, 207 
 sun, 163, 207 
 
 vapour in the air, 19, 25, 165, 166, 
 354; tension of, 152, 165, 175, 
 181, 186 ; where most abun- 
 dant, 186 
 Wave-lengths of light, 17 ; of heat, 
 
 17, 18 
 
 Weather, 26 ; Merle's Journal of the, 
 26 ; telegraphy and, 27 ; 
 bureaux, 27 ; reports, daily and 
 weekly, 31, 32, 34, 35 ; scale of, 
 41 ; bureau of United States, 
 44-68, 70 ; service, United 
 States, 45 ; its cost, 48 ; Service, 
 Canadian, 69-74 ; influence on 
 disease, 409 
 and health, 2 
 anticyclonic, in winter, 157, 158, 
 
 164 ; in summer, 164 
 radiation, 162 
 Weatliercock, 267 
 Weather-crop bulletin, United 
 
 States, 49 
 
 " Weather-glass," 119, 126 
 Weather maps, 27, 31, 32, 45, 51, 
 55, 56, 57, 67, 70, 71, 72, 73, 
 160 
 Weather Report, The Daily, 31, 32, 
 
 34, 38, 40, 41, 42 
 Weather Report, The Monthly, 34, 36, 
 
 37 
 
 Weather Review, Monthly, Unit id 
 Stat2S, 49, 51, 67, 174, 260, 363, 
 466 
 
 Weather telegrams, 38 
 Wedge of high pressure, 161 
 Weekly Weather Report, 34, 35, 37 
 Weight, of air, 13, 181 ; of hailstones, 
 
 258, 259 ; of rain, 263 
 of vapour in a cubic foot of air, 
 
 181 
 Wells' s Essay on Dew, 190 
 
 Werchojansk, low temperature at, 
 355 
 
 West India hurricanes, 162 
 
 Wet-bulb thermometer, 181 ; man- 
 agement of, 184, 185 
 
 Wet fog, 193 
 
 Wetterleuchten, 310 
 
 Wheel-barometer, 119, 126 
 
 Whipple-Casella " Universal " Sun- 
 shine Recorder, 109, 110, 111 
 
 Whirlwind, 375 
 
 Whistle-signals in United States, 63, 
 64 
 
 Wilson Radio-integrator, 104 
 
 Wind, definition of, 4, 265 ; direc- 
 tion of, how determined, 7, 267 ; 
 velocity or force of, 7 ; in storm 
 of February, 1903, 291 ; how 
 generated, 1|>, 16 ; relation be- 
 tween the pressure gradient 
 and velocity of, 9 ; direction of, 
 38, 287 ; force, Beaufort scale 
 of, 32, 36, 38, 39, 40, 266 ; trus 
 direction of, 38, 41 ; veering of, 
 163 ; hauling of, 163 ; backing 
 of, 163 
 
 Windrose, 38, 268 ; how constructed, 
 268 
 
 Winds, Trade, 159, 372 ; prevailing, 
 and climate, 352, 372 ; perma- 
 nent, 372 ; periodic, 372 ; vari- 
 able, 372 ; uses of, 372 ; influ- 
 ence on climate, 372 ; occa- 
 sional cold, 372, 373, 374 ; occa- 
 sional warm, 372, 374, 375 
 
 Winter, distribution of atmospheric 
 pressure, 10, 157, 158 ; of thun- 
 derstorms, 305 ; of disease, 410- 
 412 
 
 climate of British Islands, 379, 380 
 fog, 165, 202 
 sea-temperatures, 384 
 
 Wool-pack cloud, 205 
 
 Xenon, 22 
 X-Rays, 244 
 
 YAKUTSK, extreme temperatures at, 
 
 354, 355 
 
 Year-Book, The Meteorological, 37 
 Yeates's electric self -registering 
 
 rain-gauge, 229 
 
 ZERO, displacement of, 81, 82 
 Zigzag lightning, 308 
 Zones, climatic, 350 
 
INDEX OF PEOPER NAMES 
 
 ABBE, Professor Cleveland, 44, 45, 66 
 
 Abbott, 341 
 
 Abercromby, Hon. Ralph, 7, 154, 
 
 161, 162, 211, 301, 309 
 Adie, P., 124, 137, 147, 274 
 Aitken, John, 191, 193-200, 201 
 Anaximander of Ionia, 4 
 Andrand, 323 
 Andrews, Dr., 321, 322 
 Angot, Dr. A., 262 
 Angstrom, 102 
 
 Apjohn, Professor James, 182, 432 
 Appleby, W. E., 247 
 Arago, Francois, 92, 304, 310 
 Archibald, E. Douglas, 326 
 Aretaeus, 409 
 Aristotle, 1, 409 
 Asclepiades, 409 
 Assmann, Professor R., 185, 329, 
 
 331, 344 
 August, 182 
 Austen, Ernest E., 427, 430 
 
 BABINGTON, 168 
 
 Baker, T. W., 147 
 
 Ball, Sir Roberts., 14 
 
 Ballard, Dr. Edward, 3, 106, 417, 
 
 420-424, 439-441, 445 
 Barrett, W. F., 138 
 Bartholomew, J. G., 221 
 Bartrum, 131 
 Bassett, 425 
 
 Baxendell, Joseph, 34, 172 
 Bayard, Francis Campbell, 153, 156, 
 
 382 
 
 Beaudoux, 279 
 Beaufort, Admiral Sir F., 32, 36, 38, 
 
 39, 40, 266 
 Beaufoy, 271 
 Beckley, 285 
 Becquerel, 244, 302 
 Bentley, W. A., 241, 242, 251, 466 
 
 Benzenberg, 279 
 
 Berghaus, Dr. Heinrich, 221 
 
 Bernacchi, L. C., 371 
 
 Bernoulli, 278 
 
 Berson, Professor, 345 
 
 Besangon, 328 
 
 Besant, 133 
 
 Besson, Louis, 214, 217, 218 
 
 Best, Captain, 361 
 
 Bezold, Von, 343 
 
 Bigelow, Professor Frank H., 67, 
 174, 260 
 
 Binnie, Sir Alexander, 225 
 
 Boate, Dr. Arnold, 406 
 
 Boate, Dr. Gerard, 406 
 
 Boeddicker, Dr. Otto, 248 
 
 Bonacina, L. C. W., 247 
 
 Bort, L. Teisserenc de, 213, 327, 331, 
 336, 346-348 
 
 Bouguer, 271 
 
 Bourdon, 133, 278, 327, 328 
 
 Bowditch, Dr., 364 
 
 Boyle, Hon. Robert, 14, 16, 119 
 
 Black, W. S., 221 
 
 Black, W. T., 258 
 
 Blanford, Henry F., 121, 220 
 
 Blyth, A. Wynter, 22, 24 
 
 Brewster, Sir David, 277 
 
 Briggs, Lynam T., 173 
 
 Brockway, Dr. Fred. J., 122 
 
 Brodie, Sir B. C., 322 
 
 Brodie, F. J., 290 
 
 Broussais, 457 
 
 Browett, Charles, 251 
 
 Brown, Horace T., 23 
 
 Bryant, W. W., 313 
 
 Buchan, Dr. Alexander, 98, 99, 156, 
 159, 184, 221, 248, 305, 362, 
 365, 379, 380, 383-387, 394- 
 399, 418, 420, 434, 430, 437, 
 439, 442, 445, 466 
 
 Buchanan, Sir George, 364, 432 
 
 487 
 
488 
 
 METEOKOLOGY 
 
 Buckingham, Edgar, 173 
 Buhl, Professor, 3, 432 
 Bunsen, 19, 24 
 Burns, Right Hon. John, 430 
 Buys Ballot, 4, 6, 9, 157, 158, 159, 
 265, 378 
 
 CACCIATORB, 279 
 
 Caesar, 312 
 
 Caldcleugh, 305 
 
 Calkins, Professor R. D., 466 
 
 Campbell, 244 
 
 Campbell, John F., 109 
 
 Capper, Colonel J. E., 327 
 
 Carnelley, Professor, 194 
 
 Carpmael, Charles, M.A., 70 
 
 Casella, Louis Marino, 280, 281, 289 
 
 Casella, Louis P., 85, 87, 102, 104, 
 
 105, 109, 110, 137, 138, 229, 
 
 269, 277 
 Castelli, B., 223 
 Cator, 271 
 
 Cavendish, Lord Charles, 86 
 Celsius, 77-80 
 Celsus, 3, 409 
 Chambers, Frederick, 154 
 Chanut, 118 
 Charles, 15, 16 
 Chaucer, 267 
 Child, Walter, 275, 276 
 Chree, Dr. Charles, 316 
 Churchill, John, 205 
 Cicero, 1 
 
 Clayton, H. H., 327 
 Clouston, Rev. Charles, 466 
 Cole, Dr. Frank N., 418 
 Colladon, Professor, 313 
 Collinson, Peter, 294 
 Columbel, Rev. Augustus M., 154 
 Copeman, Dr. S. Monckton, 430 
 Cosgrave, Henry A., 203, 258 
 Cotter, J. R., 187, 246 
 Coulier, M., 243 
 Coulomb, 298 
 Coxwell, 326 
 Craveri, 279 
 
 Creighton, Dr. Charles, 447 
 Croll, 359, 360 
 Crosley, 229 
 Crova, A., 187, 188 
 Curie, M. and Mme., 244 
 Curtis, J. A., 33 
 Curtis, R. H., 33, 290, 326, 327 
 
 DALBERG, 270 
 Dalton, John, 15 
 Dancer, J. B., 24 
 
 Daniell, Professor, 175, 176 
 
 Dante, 355 
 
 Davis, A. S., 134 
 
 Debierue, 244 
 
 de Bort, L. Teisserenc, 213, 327, 
 
 331, 336, 346-348 
 Dechevrens, 280 
 Delamanon, 284 
 Descartes, 118 
 Dewar, Professor, 14 
 Dines, George, 175, 177, 178, 192, 
 
 220, 239, 271, 272, 275 
 Dines, W. H., 32, 135, 284, 287, 289, 
 
 327-331, 333, 334, 349 
 Diversus, Petrus Salius, 412 
 Dove, 151, 152, 255 
 Dowling, Professor, 432 
 Dowson, E. T., 154 
 Dunning, 412 
 Duval, 425 
 
 EBERTH, 456 
 
 Edgeworth, Richard Lovell, 274 
 
 Egnell, 344 
 
 Ekholm, M., 185 
 
 Elias, Dr., 345 
 
 Eliot, Sir John, 220, 263 
 
 Ellis, William, 119, 129, 133, 154 
 
 Elsholtz, 248 
 
 Erman, Professor Adolf, 6, 301 
 
 Evans, Sir F., 361 
 
 Everett, Professor J. D., 296 
 
 FAHRENHEIT, 15, 76-78, 80, 351 
 
 Fautrat, M., 366 
 
 Faye, Professor, 413, 415 
 
 Fenyi, 341 
 
 Fergusson, 327, 328, 329 
 
 Ferrel, Professor William, 66, 260 
 
 Field, Rogers, 137 
 
 Fineman, C. G., 213, 216 
 
 Fisher, Rev. George, 325 
 
 FitzRoy, Admiral Robert, 27, 45, 
 
 125 
 
 Flammarion, 326 
 Flexner, 425 
 Flint, Austin, 431 
 Forbes, Principal, 106 
 Forster, Captain, J. V., 360 
 Fortin, 122-123, 139, 143, 144 
 Foster, Henry, 265 
 Foster, Rev. W., 274 
 Fowle, 341 
 
 Fox, Dr. Cornelius B., 19, 324 
 Frankel, 456 
 Frankland, Professor E., 22, 201 
 
INDEX OF PROPER NAMES 
 
 489 
 
 Franklin, Benjamin, 44, 294, 317, 
 
 325 
 
 Friedlander, Dr. C., 447, 456 
 Fuller, M. L., 466 
 
 GALILEO, Galilei, 114, 115, 223 
 
 Gal ton, Francis, 271 
 
 Gay-Lussac, Louis Joseph, 15, 125, 
 205 
 
 Gerdien, H., 306 
 
 Glaisher, James, 133, 183, 184, 250, 
 324, 326, 460 
 
 Glassford, Lieutenant W. A., 392 
 
 Glover, Dr. V. J., 427, 429, 430 
 
 Goddard, 279 
 
 Godefroy, 192 
 
 Gold, Ernest, 8, 325, 327, 336-343, 
 345 
 
 Graham, 23 
 
 Gravelius, Professor H., 367 
 
 Greely, General A. W., 68, 355 
 
 Grey, 'Admiral Sir F. W., 466 
 
 Griffith, Rev. C. H., 225 
 
 Grimshaw, T. W., late Registrar- 
 General for Ireland, 28, 418, 
 419, 431, 434, 435, 437, 446, 
 448 
 
 Guericke, Otto von, 17 
 
 HAGEMANN, G. A., 278, 279, 289 
 
 Halliwell, 172, 235 
 
 Hancock, Dr. Thomas, 412 
 
 Hann, Dr. J., 99, 242, 260, 375 
 
 Harding, Charles, 121, 381 
 
 Harper, Dr. C. R., 373 
 
 Harrington, Mark W., 46, 391 
 
 Hart, Dr. Ernest, 416 
 
 Hart, Samuel, 466 
 
 Hartnup, 128 
 
 Harwood, W. A., 325, 327, 329, 336- 
 
 339, 341-343 
 Haughton, Rev. Samuel, 167, 242, 
 
 432 
 
 Hazen, General William B., 67 
 Heberden, Dr. William, 410 
 Hellmann, Professor Dr. G., 1,4, 75, 
 
 118, 223,251 
 Helmholtz, Von, 244 
 Hennessy, Professor, 280 
 Henry, Professor, 44 
 Hepworth, Captain, 327 
 Hepworth, M. W. Campbell, C.B., 33 
 Herbertson, Dr. A. J., 221, 222, 223, 
 
 262, 387-389 
 
 Hergesell, H., 328, 329, 344 
 Hermite, 328 
 Hero of Alexandria, 75, 76 
 
 Herschel, Sir John, 98, 100-102, 240, 
 
 242, 248, 359 
 Hesiod, 12 
 Hicks, James J., 131 
 Hildebrandsson, H., 211, 213, 347 
 Hill, Professor E. G., 216 
 Hill, Rev. E., 133 
 Hippocrates, 3, 409 
 Hirsch, Dr. August, 107, 409, 412, 
 
 417, 423, 447, 448 
 His, 425 
 Holmes, F., 310 
 Homer, 12 
 Hooke, Robert, 119, 126, 224, 270, 
 
 284 
 
 Hope, Dr. Edward W., 424 
 Hopkins, Rev. G. H., 459 
 Horace, 408 
 Houzeau, 22 
 Howard, L. O., 427 
 Howard, Luke, 205, 207, 210, 247 
 Howlett, 270 
 Humboldt, 151, 355, 361 
 Humphreys, Professor W. J., 339, 
 Hunt, 324 [341, 343 
 
 Hutton, 180 
 
 JACKSON, Dr. Daniel D., 427-429 
 
 James, Colonel Sir H., 275 
 
 Jefferson, President Thomas, 44 
 
 Jelinek, 186, 271 
 
 Jepson, 430 
 
 Jevons, 241 
 
 Jewell, Dr., 67 
 
 Joly, J., 130 
 
 Jordan, James B., 109, 111-113, 117 
 
 Jordan, T. B., Ill 
 
 Joule, Dr., 360 
 
 Judd, Professor J. W., 249 
 
 KAMTZ, L. F., 247 
 
 Kane, Sir Robert, 401 
 
 Kelvin, Lord, 13, 298, 299, 300, 301 
 
 Kettler, 355 
 
 King, Alfred, 126, 127 
 
 King, Dr. E. Slade, 456 
 
 King, Samuel A., 46, 47 
 
 Kingston, Professor G. T., 70 
 
 Kircher, Athanasius, 284 
 
 Klebs, 456 
 
 Klein, Professor E., 427 
 
 Koch, 416, 456 
 
 Koppen, Dr. W., 211, 381 
 
 Kugler, Rev. Franz Xavier, S. J. , ' 
 
 LADTJREAU, A., 23 
 Lamarck, 205 
 
490 
 
 METEOEOLOGY 
 
 Lambert, 287 
 
 Lament, Professor von, 168 
 
 Langley, Professor S. P., 66, 293 
 
 Lapham, Professor J. A., 45 
 
 Laplace, 149 
 
 Latham, Baldwin, 171 
 
 Laughton, John Knox, 153, 270, 
 
 274, 283, 359 
 Lawes, Lady, 224 
 Lawson, Inspector-General Robert, 
 
 153, 156 
 
 Lemme, Professor, 246 
 Lemonnier, 294 
 
 Lempfert, R. G. K., 33, 290, 292 
 Lentz, 425 
 Le Roy, 190 
 
 Leslie, Sir John, 181, 277 
 Leupold, 271, 284 
 Leutmann, 271 
 Leverrier, 27, 45 
 Ley, Captain C. H., 332, 333 
 Ley, Rev. W. Clement, 4, 154 
 Liais, M., 17 
 Liebermeister, 432 
 Lind, 272, 289 
 Lippincott, R. C. Cann, 460 
 Lloyd, Rev. Humphrey, 4, 255-257, 
 
 432 
 
 Lodge, Sir Oliver, 316 
 Lomonosow, 274, 279 
 Longfellow, H. W., 202, 211 
 Loomis, Professor, 66, 121, 221, 314 
 Lovibond, Joseph W., 204 
 Lowne, R. M., 277 
 
 MADISON, Bishop, 44 
 
 Makower, W., 306 
 
 Mann, Dr. Robert J., 13, 16, 167, 318 
 
 Mannucci, Signor, 211, 212 
 
 Manson, Sir Patrick, 365 
 
 Marcet, Dr. William, 193, 255, 295, 
 
 315 
 
 Marignac, 321 
 Mariotte, Edme, 14, 16, 224 
 Marriott, William, 113, 134, 139, 
 
 141, 183, 287, 303, 321, 373, 
 
 374, 377 
 
 Marsden, E., 306 
 Martini, 425 
 Marum, Van, 321 
 Marvin, Professor C. F., 66, 173, 174, 
 
 329 
 
 Mason, 181 
 Mathieu, 366, 367 
 Maury, Commodore, 44 
 Mawley, E., 28 
 McPherson, Dr. J. G., 191 
 
 Meinert, Dr. E., 423, 425 
 
 Mellish, H., 252, 466 
 
 Melville, Thomas, 325 
 
 Mendenhall, Dr. T. C., 66, 67 
 
 Mene, M., 21 
 
 Merle, Rev. William, 26 
 
 Metcalfe, H. J., 259 
 
 Meton, 2 
 
 Michell, Mr. Sloane, 456 
 
 Mill, Dr. H. R., D.Sc., 33, 37, 104, 
 
 170, 225, 249, 261 
 Miller, Professor, 23, 24 
 Milton, John, 198, 355 
 Mitchell, Sir Arthur, 3, 373, 418, 420, 
 
 434, 436, 437, 439, 442, 445 
 Moffat, Dr., 324 
 Mohn, Professor Henry, 302 
 Moore, Arthur Robert, 107 
 Moore, Dr. William D., 431 
 Moore, Maurice Sydney, 407 
 Moore, Thomas, 208 
 Moray, Sir Robert, 270 
 Morgagni, 412 
 Morgan, H. de Riemer, 426 
 Morse, 44 
 
 Miiller, Dr., 311, 315 
 Mund, 345 
 Murchison, Dr. Charles, 431, 435, 
 
 437 
 
 Murray, Sir John, 221 
 Murray, 426 
 Myer, General, 45 
 
 NASH, Dr. James T. C., 427 
 Nasmyth, James, 308 
 Negretti, 83, 84, 228, 234, 235, 275 
 Nettie, Dr. John, 251 
 Neumann, Dr., 375 
 Neumayer, 211 
 Newstead, Robert, 427 
 Newton, William B., 179 
 Nollet, Abbe, 271 
 Nordenskiold, Dr. G., 251 
 Nuttall, Dr. G. H. F., 427, 430 
 Nyrop, 279 
 
 ODLING, 322 
 
 Orpen, Dr. Thomas H., 406, 407 
 
 Orr, 426 
 
 Osier, Professor, 271, 287, 425 
 
 Overduyn, Professor, 278 
 
 PAINE, Hon. H. E., 45 
 
 Parke, 425 
 
 Parkes, Sir Edmund A., 3, 277, 363, 
 
 364 
 Parmenides, 350 
 
INDEX OF PROPER NAMES 
 
 491 
 
 Parry, Captain Sir Edward, 325 
 
 Pascal, Blaise, 118 
 
 Paschen, 341 
 
 Peltier, 301, 302 
 
 Pernter, M., 185 
 
 Perrault, Pierre, 224 
 
 Perrier, 118 
 
 Petavel, J. E., 306, 329, 323 
 
 Petermann, 365 
 
 Pettenkofer, Professor von, 3, 364, 
 
 417, 431 
 
 Phillips, Professor, 83, 102, 278 
 Philo of Byzantium, 75 
 Piche, 173 
 Pickering, 270 
 
 Pickering, Spencer P., 169, 170 
 Pictet, 190 
 Pirn, Greenwood, 203 
 Pitot, 273 
 Plato, 1 
 
 Pockels, Professor F., 363 
 Poey, 207 
 
 Poincare, Lucien, 243, 245 
 Pouillet, 98, 102 
 Preston, Dr. Thomas, 187, 246 
 Price, James, 171 
 Pring, T. V., 333 
 Ptolemy, Claudius, 350 
 Pujoulx, 271 
 
 QFETELET, 3, 301, 323 
 
 RAMBAUT, Professor, A. A, 459, 460 
 Ramsay, Sir William, 22, 23 
 Rawlinson, Sir Henry, 1 
 Rayleigh, Lord, 22 
 Reaumur, 77-80 
 Redier, Louis, 128 
 Regnault, 15, 16, 19, 175, 176, 182 
 Richard, freres, 102, 169, 219, 233, 
 
 285-287, 320, 327-329 
 Richmann, Professor, 294 
 Riggenbach, A., 213 
 Risler, 21 
 
 Rive, Auguste de la, 302, 321 
 Robinson, Rev. T. Ronmey, 271, 
 
 274-276, 287, 289 
 Romas, M. de, 294 
 Ronalds, Sir Francis, 128, 271 
 Rontgen, 244 
 Rooke, " Master," 270 
 Ross, Sir James, 371 
 Rosse, Earl of, 248 
 Rotch,A. Lawrence, 325, 327, 331, 346 
 Rouse, 13 
 
 Rowland, Professor, 67 
 Royds, Lieutenant C. W. R., 371 
 
 Rue, Dela, 169,311,315 
 
 Rundell, W. W., 154 
 
 Rundle, Dr. Claude, 426 
 
 Rusk, 133 
 
 Russell, Hon. Rollo, 259 
 
 Russell, R., 373 
 
 Russell, Spencer C., 312, 313 
 
 Rutherford, Professor, 85, 104, 244, 
 
 245 
 
 Rutty, Dr. Thomas, 406, 407 
 Rysselberghe, Professor F. van, 93, 
 
 129 
 
 SABINE, 151 
 Salleron, 129 
 Sanctorio, 76 
 Sandwith, F. M., 375 
 Saussure, de, 178, 294, 295, 301 
 Schaeffer, George C., jr., 47 
 Schmidt, 244, 270 
 Schonbein, 321, 322 
 Schreiber, P., 260 
 Schubler, 301, 363 
 Schumacher, 148 
 Scoresby, 18 
 
 Scott, Dr. R. H., 6, 18, 25, 77, 80, 
 93, 98, 128, 139, 149, 166, 167, 
 177, 183, 186, 192, 205, 207, 
 210, 238, 242, 250, 255, 262, 
 268, 287, 296, 298, 301, 306, 
 310-312, 314, 317, 320, 321, 
 350, 351, 355, 361, 362, 373, 
 380, 381, 394 
 Secchi, Padre, 102 
 : Seeley, Professor H. G., 399-402 
 Semple, 407 
 Shakespeare, 200 
 
 Shaw, Dr. W. N., 7, 9, 33, 107, 251, 
 290, 291, 292, 336, 340, 343, 349 
 Shiga, 425, 426 
 I Sidebottom, James, 253 
 Siemens, Dr. C. William, 98 
 Six, James, 86, 190 
 Smith, Angus, 19, 20, 22 
 Socrates, 1 
 
 Soddy, Professor, 244 
 ! Soret, 323 
 I Southall, T., 101 
 Stark, 3 
 
 Steel, Thomas, 247 
 Stefan, 341 
 
 Stevenson, Robert, 326 
 ! Stevenson, Thomas, 7, 43, 84, 89 
 
 Stewart, Professor Balfour, 147, 303 
 i Stocke, Dr. Leonard, 250 
 Stokes, Professor Sir George G., 109, 
 110, 111,246,271,274 
 
492 
 
 METEOROLOGY 
 
 Stow, Rev. Fenwick W., 101, 225 
 Strachan, Richard, 79, 118, 152, 154, 
 
 159 
 
 Stupart, R. F., 69, 71, 158 
 Sturt, Captain, 375 
 Supan, Dr. A., 221, 261, 355 
 Sutton, J. R., 156 
 Swan, W., 459 
 Sydenham, Dr. Thomas, 410 
 Symons, George J., 26, 33, 105, 128, 
 170, 186, 204, 209, 223, 226, 
 228, 239, 240, 241, 247, 248, 
 254, 258, 261, 311, 396, 466 
 
 TAIT, 322 
 
 Tatham, Dr. John, 106 
 Theophrastus of Lesbos, 2 
 Theorell, Dr., 93, 129 
 Thompson, R. Campbell, 2 
 Thomson, Sir Joseph, 244, 245, 246, 
 
 293 
 Thomson, Sir William (see " Lord 
 
 Kelvin "), 13, 298, 299, 300, 301 
 Thome, Sir R. Thorne, 432 
 Thrift, Professor W. E., F.T.C.D., 
 
 107, 329, 403 
 Thurston, Edgar, 447 
 Tissandier, M. G., 22, 326 
 Torricelli, Evangelista, 114, 115, 118 
 Townley, R., 224 
 Toynbee, Captain, 381 
 Trabert, Wilhelm, 259, 340, 341 
 Tripp, 221 
 
 Trouton, Professor F. T., 178, 179 
 Tucker, Edward, jr., 209 
 Twemlow, Francis E., 323 
 Tyndall, Professor John, 186, 197, 
 
 205, 249 
 
 UPTON, Professor Winslow, 67 
 
 VAN MARUM, 321 x 
 
 Van Rysselberghe, Professor F., 93, 
 129 
 
 Vernier, Pierre, 141 
 Verz, F. W., 260 
 Vidi, 132 
 
 Violle, Professor, 102 
 Volta, Alessandro, 255, 294, 295, 
 297, 313 
 
 WALDO, Professor Frank, 66 
 
 Wall, Dr., 294 
 
 Wallis, 123 
 
 Wallis, H. Sowerby, 248 
 
 Ward, Colonel, 225 
 
 Warren, Rev. Isaac, 264 
 
 Webster, W. H. E., 265 
 
 Wells, Charles William, 190 
 
 Wheatley, Colonel, 164 
 
 Wheatstone, Sir Charles, 93, 128, 
 
 311 
 
 Whewell, 274, 285 
 Whipple, G. M., 109, 110, 285 
 White, Margaret, 306, 333 
 Wiesner, 242 
 Wild, Professor, 270, 354 
 Wilke, Professor, 271 
 Williams, 426 
 Williams, Dr. C. Theodore, 351, 355, 
 
 362, 365, 372 
 
 Wilson, C. T. R., 244, 245, 246, 306 
 Wilson, Dr. Alexander, 325 
 Wilson, Dr. W. E., 104 
 Wilson, Patrick, 190 
 Wilson, Sydney, 358 
 Wistrand, Dr. 439 
 Wojeikof, A., 365, 366, 367, 369, 
 
 370 
 
 Wolf, 271 
 Wollaston, 274 
 Woltman, 274 
 Wren, Sir Christopher, 223 
 
 YEATES, George, 229 
 Young, Dr. Thomas, 77 
 
 ZAMBRA, 83, 84, 228, 234, 235, 275 
 
 THE END 
 
 JKebman Ltd., 129 Shaftcsbury Avenue, W.G. 
 

 
 
 HPR231956LU 
 
YC 10847