;:itTO ■ii'^\ S^-Mr^K>.. fVir.i, , , i'ln'.r.'!,!!.?,;;:: ifflst lri',;^'^'" mmi .: wmi^. THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA LOS ANGELES >S5>-/ m^ M.O. 215 (1917). METEOROLOGICAL OFFICE. THE SEx\MAN'S HANDBOOK OP METEOROLOGY. A Companion to the Barometer Manual for the Use of Seamen. THIRD EDITION. ^Juiltsijfti i)p tt)f .Hult)orit),> of tljc ittftforologtcal tCommittef. LONDON : PUBLISHED BY HIS MA.IESTY'S STATIONERY OFFICE. To be piiichasod tlirougli ;niv Bookseller or directly from H.M. STATIOrs'ERY OFFICE ;it the following addresses : Imperial House, Kingsway, London. W.C.2, and 28, Abingdon Stueet, London S.AV.l; 'M, Petkk Stkeet, Manoiiestei! : 1, St. Andrew's CiiESfKNT, Cardiff ; 23, Forth Street, Edinburgh ; or Iroin K. PONSONBY, Ltd.. 116, Grafton Street, Dublin. ib 191H. Price 3.S. Or/. Nrt. 8 n 7 ;j (i S 3 4 Ill TABLE OF CONTENTS. j Page Preface vii The Measckement op Pressure 1 Approximate Correction and Reduction to Sea Level op Barometer Readings in C.G.S. Units for Use on Ships 11 Introduction 18 Introduction to the Third Edition 19 Chapter I. — Tin: Atmosphere and its (TDuuihitlon. — Composition of the Atmos- phere. Heii^ht of the Atmosphere. Density of the Atmosphere. Cause of VVind. Effect of Earth's rotation upon wiud direction. Cyclonic and Anti-cyclonic circulations. Origin of the term Cyc/o«e ... ... ... 20 Chapter II. — Temperature and Humility. — Temperature defined. How interchanges of Temperature take place. Terrestrial Radiation. Decrease of Temperature with height Annual ransje of Temperature. ^ Diurnal variation of Temperature. Humidity. Evaporation. Latent •J Heat. Formation of Cloud. Relative Humidity. Elastic Force of "^ Vapour ... ... ... ... ... ... ... 24 . "^ Chapter III. — Atmospherle Pressure and Wind. — Isobars. Pressure vV*^ Gradients. Gradient Wind. Rril.tion of Gradient to Wind Velocity. Adjustment of Wind Velocity to Gradient. Movement of Wind in Revolving Storms. Analogy to Planetary motion ... ... ... ... 32 Chapter IV. — Clouds. — Luke Howard's Nomenclature. Intermediate modifi- cations. Compound modifica;ious. Cla-'sification of International Comaiittee. Height and motion, speed ami direction, of Clouds. Cloud • distribution in a Barometric Depression. Motion of Upper Clouds ... 38 ^ Chapter V. — .Vist. Fog. Precipitition. — Distinction between Mist and Fog. '^ Sea Fog. Land Fog. Fog associated with High Barometric Pressure. Fog associated with Low Barumecr c Pressure. Sea Fog on Coast. Mist and Fog Frequency round British Isles. Relation of Thick Weather to Wind Direction. Scale of Kog Intensity. Rain. Hail. Snow. Coron-i. Halo. Rainfall Statistics : Rainfal. Frequency ; Snow Frequency ... 50 Chapter VI. — Atmospheric Pressure dLstribution and Weathtr Conditions. — Preparation of Weather Charts. Beaufort notation of Weather. Con- C? figurations assumed by Isobars that are characteristic of definite Weather ^ Conditions. The Cyclone. The Anti-cyclone. Secondary VVind System. ^^ "V "-shaped Depressions. Gusts and Squalls. Line Squalls. The Col. 68 Chapter VII. — Anticipation of Weither hy observations at a single Station. — Approaching Dejjressions : First indications, Bearing of Centre, Path of Centre, Weather Sequence during Passage of Depression over Station. Temperature in Anti-cyclones. Motion of clouds of cirrus type and Weather changes. Weather signs. Recurrence of Warm and Cold Periods ... ... ... ... ... ... ... .- ... ••• 82 Chapter VIII. — Types of Weather Conditions. — South-Westerly. Southerly. Westerly. North- Westerlj'. Northerly. North -Easterly. Easterly. South-Easterly. Typical Thundrtrstorms : Coming from South, Forming locally, Connected with Northern Depression... ... ... ... ... 98 Chapter IX. — Gales on the Coasts of the British Islands. — Prevalence of Gales. Definition of the term (Jale. Classification of Gales in Quadrants. Distribution of Gales. Average Frequency of Gales. Mean Monthly Frequency. Wind direction in Wales. Relative Frequency of Gales from various points of the compass. Relative Frequency of Gales _ on various Coasts of the British I.sles, from different points of the compass. Tracks of Storm Centres. Speed of Storm Centres ... ... 108 (11979— la.) Wt. 159-10— 44. 10,000. 1/18. D A: S. Q. 3. -^2 IV Pagk Chapter X. — Icelicnjs and other forms of Briftiwj Ice. — Ice formation. Tyndall's experiment. Faraday's discovery and explanation of rvcfdution. lUuf'tration by Dewar. The formation of Glaciers. Analogy between river flow, and glacial motion. The colouring of bergs and the causes. Glacial regions. Formation of Icebergs of Glacial origin. Bergs of Ice Barrier origin. Ice Barriers aad their formation. Field Ice, its forma- tion and colour. Floating Ice in other forms. The detection of drifting Ice. The Recording Micro-thermometer ; of Professor Barnes upon its re?ults. The Ice blink. First appearance of Icebergs in thick weather. The Whistle or Siren. Ice in the Northern Hemisphere. Dr. Rink on Glaciers and Icebergs. Ice-friths. The proportion of Icebergs above water. The di>tribution of atmospheric pressure and the disruption and movements of Ice. The dissolution of Icebergs. Drift of Ice from Greenland Sea. Ice off South Coast of Greenland. Ice Frequency in North- Western Atlantic. Ice limits. Phenomenal drifts of Icebergs, and Bergs cf exceptional height. Effect of Weather Conditions upon Ice drift into Atlantic. The loss of s.s. " Titanic ". Velocity of Labrador Current. Ice in the Southern Hemisphere. Phenomenal lengths of seme Siiithern Ocean Iceberjjs. Phenomenal heights of some Southern Ocean Icebergs. The positions in which phenomenally long and phenomenally high bergs were ob-erved. The discoveries of Rcss. Victoria Land. The Great Ice Barrier. Ross on the disruption of Ice Barriers. United States Expedition under Wilkes. Wilkes on the formation of Ice islands. Borchgrevink's Expedition in Smithern Cross. The birth of an Iceberg. Swedish Expedition under Nordenskiold. Scott on the origin of Icebergs. Calving of Icebergs from high cliff. Ice frequency in Southern Ocean. Yearly variation in Ice frequency ... 11!) Chapter XI. — Meteorological Instruments. — Barometers: Torricelli's Experi- ment, Action of Mercury Barometers analogous to that of the Pump. Fishery Barometers. Scale Graduations and use of Vernier. Kew Pattern Barometers and their Scale Graduations. Aneroid Barometer and Baro- graph. Principle, Construction, and Management. Thermometers : Principle, Construction, and Graduations. Thermograph. Wet and Dry Bulb Hygrometer. Maximum and Minimum Thermometers. Sea Water ' Thermometer. Stevenson's Screen. Anemometers : Osier's Pressure Plate. Robinson's Cup. Dines Pressure Tube. The Principle and Construction of the respective instruments. Raingauges : The Snowdon Pattern; The Meteorological Office Pattern ... 143 Ai'PENDix I. — Displacement of the Horizon. — Mirage. lilevated Horizon. Depressed Horizon. Explanation of these Phenomena ... ... ... 175 Appendix II. — Telegraphic Information supplied to the Public hi/ the Meteorological Office. — Daily Weather Reports. Forecasts and Storm Warnings. Wireless Reports from Liner.i. Forecast Districts. The Distribution of Forecasts and Telegraphic Intelligence. Storm Warnings. Storm Sififnals. List of Storm Signal Stations ... ... ... ... 176 Appendix III. — Ofncial provision for the supply of Fishery Barometei's ... 181 LIST OF ILLUSTRATIONS. FIGURES. FipT. Description. Page. A The Exploration of the Atmosphere .. xxxix B Graduation of Kew pattern barometer in millibars and inches . 17 1 Barometer Gradient 35 2 Cloud Distribution in a depression . 47 3 Forecast Districts of United Kingdom 64 4 Line of Squall, Sth February. 1906, 8 a.m. ... 78 5 Line of Squall, 8th February, 1906, 2 p.m. ... 79 6 Line of Squall, Sth February, 1906, 6 p.m. ... 80 7 Bearing of cyclone centre • 84 8 Mean Monthly Frequency of Gales ... 110 9 Relative Frequency of Gales. Seasonal 113 10 Relative Frequency of Gales on Coasts of British '. sles .. 114 11 Tracks of Storm Centres • 117 12 Fishery Barometer . 148 13 Fisher J' Barometer 149 14 Kew Pattern Barometer 149 15 Barometer tube ... 150 16 Errors of parallax 152 17 Method of reading Barometer. Millibar scale 1.52 18 Method of reading Barometer. Millibar scale 1.52 19 Method of reading Barometer. Inches 1.54 20 Method of reading Barometer. Inches 154 21 " Sea Barometer " 156 22 Barograph... 157 23 Barogram 158 24 Wet Bulb 162 25 Stevenson Screen... 16.S 26 Maximum L'hermometer 164 27 Minimum Thermometer... 165 28 Robinson's Cup Anemometer ... 168 29 Dines' Pressure Tube Anemometer 170 30 Dines' Pressure Tube Anemometer, Head of 171 31 Rain gauge. Snowdon pattern ... 173 32 Rain irauge, M.O. pattern 174 33 Refraction 175 34 Refraction... 175 35 Forecast Districts, United Kingdom 177 1197.1 VI List of Illustrations — cont. PLATES. No. Description. To face Page Cloud forms xxxii I Fo^ associated with high and low barometrical pressure 52 II Fog associated with warm air relatively cold sea 54 III Fog and mist round British Isles. January-March 56 IV Fog and mist round British Isles. April-June 56 V Fog and mist round British Isles. July-September 56 VI Fog and mist round British Isles. October-December 56 VIA Solar Halos observed at Aberdeen 64 VII Passage of Cyclone over British Isles 23rd to 25th March, 1909. 70 VIII Passage of Cyclone over British Isles 26th to 28th March, 1909. 70 IX The Anticyclone 72 X " V "-shaped Depression 74 XI The col 82 XII Weather Sequence ... 86 XIII Weather Sequence. Barograms and Thermograms 88 XIV Maximum Wind Force in rear of depression 94 XV Types of weather conditions : South-Westerly, Southerly, Westerly. 100 XVI Types of weather conditions : North-Westerly, Northerly, North-Easterly. Blizzard. Easterly. 102 XVII Types of weather conditions : Easterly type 106 XVIII Types of weather conditions : South-Easterly, Easterly type. Thunderstorm types. 106 XIX Distribution of Ice in North Polar Regions. May 1911 130 XX Distribution of Ice in North Polar Regions. August 1911 ... 130 XXI Extreme limits of Icebergs and Field Ice in the North Atlantic. 132 XXII Phenomenal drifts of Icebergs and Bergs of exceptional height in North Atlantic. 134 XXIII Phenomenally long Bergs of Southern Ocean 134 XXIV Phenomenally high Bergs of Southern Ocean 134 XXV South Polar Regions 140 XXVI South Polar Ice Limits 142 Vll THE SEAMAN'S HANDBOOK OF METEOROLOGY. A Companion to the Barometer Manual for the Use of Seamen. PREFACE. The writing' of this book has been entrusted to Com- mander M. W. Campbell Hepworth, C.B., R.D., R.N.R., Marine Superintendent of the Meteorological Office, who has the advantage of long experience of the practical needs of seamen in connexion with the study of weather. It is now nearly sixty years since the Meteorological Office was established as a Department of the Board of Trade under Admiral R. FitzRoy for organising the collec- tion of Meteorological observations from ships traversing the various oceans of the g^lobe. The purpose was threefold : — First, that trustworthy in- formation about the weather in every part of the navigable seas should be supplied to the seafaring community, and thereby the seaman should be enabled to know what to expect in the way of weather, as a rule, on any voyage which he might have to make. Secondly, by the use of the knowledge so collected, to enable him to make the most of conditions of weather generally favourable, by the choice of his route, or the time of his voyage, or the setting of his course ; and thirdly, by the study of the laws of weather, to enable him to know what sort of weather to expect from day to day, and to make the best of bad weather, if he had to 11979 A 4 viii Preface. face it. These purposes have been steadily pursued at the Office, and the scope of its operations has meanwhile been greatly enlarged in consequence of the public interest in the development of the new method of anticipating coming weather. The Old Method of Using the Barometer. The early years of the Office were of great importance in the history of the study of weather. The barometer, that is an instrument for finding^ out the heio'ht of a column of mercury which would balance the pressure of the atmosphere, or briefly, for measuring the atmospheric pressure, had long been known as a weather-glass ; and the use of the barometer for anticipating changes in the weather was well established. But at that time each individual had only his own instru- ments for the purpose. Guidance in the use of the barometer was for the most part limited to the traditional inscription on a dial, which, unfortunately, may still be seen on many instruments. A height of 31'0 inches of mercury was marked Very dry. ,, 30*5 ,, ,, ,, Set fair. „ 30-0 „ „ „ Fair. „ 29-5 ,, „ „ Change. ,, 29*0 ,, ., ,, Kain. ,, 28'5 „ ,, „ Muchrain. 28-0 ,, „ „ Stormy. It was ah-eady recognised that these stereotyped descrip- tions of the weather belonging to certain barometric heights, although in a way " founded on fact," were often disappoint- ing and perplexing, and they were already regarded as mis- leading. Anyone who watches a barometer can confirm this view for himseK There was an example only a few weeks ago, when an evening paper described the climatic conditions in London on 10th December, 1912, as "utterly wretched," and illustrated them by a picture of a barometer pointing to " Fair," because the mercury read just over 30 inches. One of the early services of the Office to the public was to intro- duce a barometer for the use of fishing communities {see p. 181), which carried an inscription free from objection, and one of its most recent efforts {see p. 156) is to supply a new barometer dial showing pressure in absolute units ranging from about 95 per cent, to 105 per cent, of the middle position, and utilising information which has been gradually compiled about the distribution of pressure over the sea. Cyclonic Depressions. ix Cyclonic Depressions or Sicirls of Air. By the time the Office was established, it had been made out that the tropical hurricanes, the worst type of weather that a ship had then to fear, were swirls of air travelling, not very fast, through a comparatively quiet atmosphere with low pressure in the centre and gradually increasing pressure all round it. Many features of the behaviour of these hurricanes present an unmistakable likeness to those of a whirlpool or vortex such as may be formed when water runs out of a basin through a hole in the bottom. Befor.^ many years had elapsed it was found that the winds and gales of our own coasts, and indeed all over the world, had certain points of likeness to the tropical hurricane, and might indeed be regarded as parts of great swirls with less violent winds than those of the hurricane, but spread over vastly greater areas and generally making their way, sometimes faster, sometimes slower, over sea and land. It would hardly be correct to regard these swirls as making their way through a quiet atmosphere like a tropical storm ; they are rather regarded as forming part of a great procession, so that the whole region is filled with swirls and counter-swirls. All the air takes part in the motion and there is nothing left that one can regard as the undisturbed atmosphere for the swirls to move in. As soon as these swirls, which we call '' cyclonic depressions," were identified, every change in the atmospheric pressure show^n on the barometer became an indication of the approach or departure of one of these swirls, and it was soon seen that the reading of an individual barometer depended upon its position with reference to the centre of the swirl that was passing. The legend on the dial lost its si^ a a a o •M -* tC 03 o o o — cc m O W , ■ ^ 02 o:S1 ^& =1 c2 pi © ip of -185( h wa CQ cc O o >L O o 1— 1 OS QJ — C u a> O (B ^ •I-l bt H *^ he s < cr. > ell-co fort's ould from 'So *g^ = . o cs a; _^j- —T +^ O ^- s»-i -1-i a — cS _ S j=i >-^ a '5 dml 1 sa clea: ^ .t5I ^ ^^ -c 0/ X2 cS ^ 5 iH O O 3J «C t^ CO g & fe o 5 :^' ci :x, o a X Ol tc cS X O g a. a. o 'C -c 'TS ■u si Oi 'C « (B 03 =4-1 a. (!. 03 'D to a (1) L4 o he a o tH « m «H o 1^ ^ a QJ 03 K 03 g. 03 03 03 00 (U — • <53 i- 03 -2 ? I- j2 z: ce he ^ . , t8 he ^„ he fc. to ,a a 03^ a CO o 'C ® o 03 S o £ S in tc t^ CO 05 ^ a ^: a: W s -s S 5 5 - 2 •:- tn CO O Oi o •i-H O •r-l O © ft © •I-H a © © .— '-' O) 05 O »^ ^ §;§ 03 .« o cc -s 2 ^ o d 5 =s a iT o «5 Jag 09 12 Wind Force Scale. XXI Special consideration is re- quired for the specification of the scale for use on board steamships. For this purpose it is recommended that as opportunity occurs use be made of the equiva- lents given in Col. 2. Thus, when the ship is running in a calm at 15 knots, the wind felt in an exposed position on board will be a moderate breeze, which, according to the table, is force 4 on the Beaufort scale, and, if a fcimilar breeze is felt when the ship is running at 15 knots, i\(fht hrfore the wind., the actual speeci oi tne wind will be 30 knots, 7 on the Beaufort scale, accord- ing to the table of equiva- lents. other opportunities occur from time to time for com- paring the speed of the wind with the speed of the .ship. A hand anemometer may be employed if used judiciously and if proper allowance be made for the motion uf the ship. Sufficient wind for working ship. Forces most advan- tageous for sail- ing with leading wind and all sail drawing. Reduction of sail becomes necessary wven with a lead- ing wind. Considerable reduc- tion 01 sail neces- sary even with wind quartering. Close reefed sail when running ; or hove-to under storm sail. No sail can with- stand even when running. , ; y s / ; ^ _, Light breeze \ Moderate breeze ^ 1 1 1 1 Strong wind Gale force Storm force s / ^ ' ^ 1 •*> o o o •■a m o --0 t^ !>-. 00 00 CO 00 C<1 o o to CI 3J o « 3 GO c. CO o 00 'Jj CO 00 CO -*< n CO -*< CO N >-. o lO M lO CO ■J 3> uO OO lO 9 o < - ci CO - »o «o t- CC J> ^ V. xxii Preface. Structure of Wind. The relation between the Beaufort numbers and the cor- responding wind velocities has been based upon the hourly velocity of the wind. The observer has estimated the force upon the Beaufort Scale at one of the fixed hours for observ- ing, say 7h. and the estimate has been set alongside the mean velocity of the wind for the hour between 6h. oOm. and 7h. oCm. as taken from the record of the anemometer. At the time that was the only available mode of procedure. But in reality winds are not steady currents of air, flowing hour by hour without variation. The average hourly wind current of the anemometer is made up of rapidly alternating gusts and lulls and occasionally varied by squalls lasting for some minutes. In light airs there are often " puffs " of wind and even calms are sometimes disturbed by " catspaws." Efforts have been made to give observers more precise directions for estimating wind force and attention is now being paid to the study of wind structure. Some interesting results are given in the Technical Report of the Advisory Committee for Aeronautics, and for a ship at sea Mr. G. I. Taylor made a beginning in 1913 by comparing the wind velocity at 45 ft. above sea level with that at 70 ft. on board the "Scotia," in 1913, when she was on a cruise for the investigation of ice. He found a difference of 7 per cent, in favour of the higher level. What is a Gale ? It will be seen that in the Beaufort specification the winds indicated by the numbers 7, 8, 9, 10, are called Moderate Gale. Fresh Gale, Strong Gale, and Whole Gale respectively, and the new specification suggests that, in the interests of meteorology, a distinction should be drawn between a breeze, a wind, and a gale. It would be easy to find evidence from English literature that making a hard-and-fast distinction between these three words is an innovation. Still, so far as distinguishing between a gale and other winds or breezes, the innovation is a necessary one, because we endeavour to warn the coasts for gales which are not expected to reach what Admiral Beaufort called a storm (Force 11), although the technical name of " storm -warning " or " storm signal" might convey the misleading inference that the issue of a warning implies the probability of a storm in the sense used by Beaufort. It is well, therefore, to make it clear that in deciding whether to issue a " storm-warning " a meteorologist has in Wind of Whole Gale and Storm Force. xxiii mind the probability of a " gale," the warning would be more appropriately termed a " gale-warning " ; and in counting- gales we do not include what Beaufort called " a moderate gale" (Force 7). A good deal of time might be spent in arguing whether or not a " moderate gale " is really a gale in a seaman's judgment without coming to any satisfactory con- clusion. In fact, when it is necessary to draw a distinction for statistical purposes, the implied contradiction of a moderate gale is not satisfactory. In specifying a wind scale for land in the " Observer's Handbook " we have sfiven the description " high wind " to Force 7. At sea winds of that force may perhaps be called " half a gale," but for our purposes a " gale " means a wind of Force 8 or more, that is, the average velocity during an hour should reach at least 38 miles an hour. This definition of a gale is in conformity with that adopted by international agreement for regulating the use of the symbol jf which is employed to denote the occurrence of a gale at meteorological stations on land. Winds of Whole Gale and Storm Force. Over the British Isles it is only during gales of excep- tional severity that the wind reaches what is technically known as storm force, i.e., exceeds 9 of the Beaufort Scale, or a run of 54 miles in an hour. Occasions during the past fourteen years on which a higher velocity than 54 miles in the hour has been recorded on anemometers at stations in various parts of the British Isles in connexion with the Meteorological Office are shown in the accompanying table, (Table A) with the dates, velocities, and places of occurrence. On the eight (jccasions when the figure for the velocity is marked #, Force 11, a "storm," according to Beaufort, has been reached. It must, however, be remembered that the absence of a record on an anemometer does not necessarily mean that the higher forces were not reached. The exposure of anemometers is sometimes not free enough to give the full force of the wind in the locality, because the vane or cups are not carried high enough above the surrounding- buildings to give a measure of the current which may produce its full effect upon a ship at sea, not far away, or upon a projecting obstacle on an exposed shore. XXIV Frejace. Table A, Wind Velocities of Force 10 or upwards of tlie Beaufort Scale, recorded on Anemometers at Stations in connexion ivith the Meteorological Offi.ce, 1899-1913. Station. Date. Average Velocity during an ! Beau- fort Num- Maximum Velocity hour or more. 1 bers. in gusts. Miles Miles per hour. per hour Holyhead 1899, January 2nd — .' 94 Fleetwood „ January 12th *75 11 — Southport ., January 12th — ? 90 Scilly 1900, December 28th HI 10 90 Shields 1903, February 26th-27th... 5.') 10 — Valencia „ February 26th-27th . . . 63 10 — Kingstown .. February 26th-27th... *66 11 — Liverpool February *65 11 88, 86, 84. 82 and 81. Blackpool ,, February — .' 87 Fleetwood „ July 5th-6th .59 10 — Scilly ., September lOth-1 1th *61 11 — Scilly 1904, January 13th-15th ... 62 10 75 Scilly „ February 12th-13th... *65 11 77 Pendennis Castle, 1905, March Uth — .' flOS, 93 Falmouth. Holyhead .. March 15th — 7 84 Scilly Iv^Oe, January Sth 59 10 81 Pendennis ... „ January fith ... »65 11 85 Scilly January 18th 55 10 76 Pendennis ... January 18th 56 10 70 Scilly ,. December 5th-6th ... 62 10 85 Holyhead ... ., December 5th-6th .. 55 10 79 Roche's Point December 5th-6th ... 59 10 75 Deerness .. December 2Bth-28th 62 10 — Deerness 1907, January 28th 59 10 — Fleetwood „ February I9th-21st.. 59 10 — Pendennis .. March 16th-17th ... 55 10 69 Southport ... .. March lHth-17th ... 60 10 81 Fleetwood ., March l'ith-17th ... 56 10 — Pendennis .. October 18th-19th ... 55 10 71 Fleetwood November 12th-13th 59 10 — Fleetwood .. December 13th- Uth 61 10 — Pendennis ... .. December 26th-28th 59 10 71 Scilly 1908, January 7th-8th ... 55 10 — Fleetwood ... February 22nd 55 10 — Deerness .. February 22nd-23rd 59 10 — Pendennis ... .. March 5th-6th 55 10 73 Pendennis ., August 2fith 55 10 69 Pendennis August 31st 58 10 78 Scilly September 1st 56 10 69 Pendennis September 1st 56 10 75 Fleetwood .November 22nd-23ri 56 10 — Deerness .. December 28th-29th 56 10 — Scilly 1909, January 16th 55 10 78 Edinburgh ., January 18th 56 10 — Pendennis .. October 7th 56 10 67 Scilly .. October 23rd *70 11 90 Pendennis ... .. October 23rd 55 10 73 Fleetwood .. November 12th-13th 55 10 — Pendennis November 18th 56 10 75 Scilly ., December 1st 56 10 66 t On the wind record for this date there is an isolated mark which might be taken to indicate a velocity of 106*5 miles per hour, but careful examination at the time left the matter in doubt. Wind Velocities. XXV Wind Velocities of Force 10 or upwards of the Beatcfort Scale, recorded on Anemometers at Stations in connexion with the Meteorological Office, 1899-1913 — continued. Station. Date. Average Velocity during an hourormore, Bean- Num. I ^«i««i*y bers. Maximnm in gnsts. Pendennip ... Southport ... Fleetwood ... Scilly Pendennis ... Kingstown ... Southport ... Pendennis ... Plymouth ... Pendennis ... Kingstown ... Southport ... Scilly Scilly Pendennis ... Deerness Pendennis ... Pendennis ... Pendennis ... Pendennis Castle Eskdaleuiuir Southport ... Fleetwood ... Quilty Pendennis Castle Pendennis Castle Pendennis Castle Scilly Pendennis Castle Pendennis Castle Scilly Pendennis Castle Pendennis Castle Fleetwood ... Pendennis Castle Dwyran Deerness Quilty Pendennis Castle Scilly Roches Point Roches Point Dwyran Pendennis Castle Kingstown ... Quilty Valencia Fleetwood ... Southport ... Pendennis Castle Pendennis Castle Pendennis Castle Pendennis Castle Fleetwood ... Southport ... Pendennis Castle Fleetwood ... 1909, December 2nd ,. December 3rd ., December 3rd 1910, January 24th ,, February 1 ith February 17th February 17th .; February 18th-19th '.. February 20t]i ,, February •20th-21st.. .. February 2Ist ,, February 21st August lst-2nd ,. October 2nd ., October 14th ,, November 7th ., December 7th De.;emher 9th „ December 16th 1911, October 30th ., November r»th .. November 5th ,. November .")th ,, December 4th-.5th . ,. December Bth-7th . ,. December 10th ,. December 13th ., December 13th ., December ISth 1912, March 4th ... ,. March 4th-5th ., March .^>th ... ., March 21st ... ,. November 10th ,. November 13th „ November 2< g o o S o h u s 2, eS-S -a S &^ S-S-^ 0) ^ 12 "^ r/^ 1=2 i: i S 5* M c ^ rt o o 02 O K o ■o rt ^rt C " ^ T3 •"^ a O S 0) ? Mrt i ^ a a ?^ O • r2 0) «> (D rg J3 S ^ "" ^ :=0 +J OJ o ^ ^ ja t« ^-^s r7:K ^ « :^ > ^. O 4) cS ? bo o <^ o; 33 XXXV. •A - o H 5 -^ g I o ^ ;z; I O 5 t -s- H 5 -? r H ? « o O c« Ph to o P O ►J o o m s "" -§ -2 o - 'S 5 t55 c^ « c ■2 S G-2 tr -r< ,^j ^ bn 01 a> 5.a -^ *^ tt s H,S C Ci-g rt K O O >- M ~ — 2 Jos S 5 o8 ft &.;= o o a o H=3^ XXXVl. xxxni. XXXVlll. NOTE ON THE ILLUSTRATIONS OF CLOUD-FORMS. The definitions of the typical Cloud-Forms are given in pp. 40 to 42. These definitions are taken from the International Atlas of Clouds which was approved at the International Conference of Directors of Meteorological Institutes and Observatories at Innsbruck in 1905 and published by Gauthier Villars in 1910. It includes a number of carefully selected illustrations of the typical forms reproduced by chromo-lithography, which are intended as a guide to observers as regards the nomenclature of clouds. Copies of the Atlas can be obtained from the Meteorological Office, price 10s. It was originally intended to make a selection of the illus- trations reproduced in the International Cloud Atlas and include copies of them in this volume, but when the Atlas was published it was felt that it would be unjust to the international selection to pick out some and leave others ; the international selection must be taken in its entirety as illustrating what the Commission and the Conference meant to be included. Meanwhile it is a matter of common experience that the difficulty of the meteorological observer is not so much in recognising a cloud-form when a typical example occurs as in describing what may be called the every-day sky which is often very composite. The Meteorological Office has become possessed of a rich collection of beautiful cloud photographs by Mr. G. A. Clarke, of Aberdeen Observatory, showing all kinds of skies, typical and other, for the naming of which the principles of the international classification ought to be an adequate guide. A selection has therefore been made from the photographs included in Mr. Clarke's album and to these names have been given in accordance with the principles of classification laid down in the International Atlas as understood in the Meteorological Office. It is not suggested that the selection includes all the types which an expert meteorologist will recognise. THE MEASUREMENT OF PRESSURE. The demand for a reprint of the Seamaii's Handhooh of Meteorology must not be allowed to pass without a note about the innovation of the past year, a change in the graduation of the barometers supplied by the Meteorological Office to the stations which report daily by telegraph and to the observers of the mercantile marine. Since the 1st May, 1914, the readings of the barometers published in the Daily Weather Report have been given in "millibars" instead of "inches," and it is intended to ask the marine observers to use the same scale in the meteorological log. What a Millibar means. A millibar is approximately the thousandth part of the ordinary atmospheric pressure at sea level. Face a wind of force 6 in an exposed situation and you will be withstanding a pressure of about a millibar. Multiply that by a thousand and you will get some idea of what you would have to withstand if somebody surreptitiously or otherwise removed the air from behind your back on a calm day. It amounts to about 14J, lbs. on the square inch, and taken over the front of the average human body would mean v/ithstanding a force of seven tons, so that there would be little chance of holding it for long. Fortunately there is not much chance of anyone unawares taking away the air from Ijehind one's back, but, nevertheless when one has to think about the pressure of the atmosphere, it is forces of this kind that one ought to have in mind, forces which are estimated in pounds i)er square inch or, better still, their equivalent in millibars and not the inches of a mercury column which are only the conventional expression of atmospheric pressure. Barometers might have been graduated in pounds per square inch instead of inches ; and the range at sea level would then have been from 13i to log instead of from 27 to 31, and we should then have kept better in mind the 2 Preface. forces which are operative in the problems of the atmosphere. It is still not too late to take such a step for there are many problems of the atmosphere that are not yet solved. Corrections to Barometric Readings. But the natural philosophers to whom we look for enlightenment on such points have tafcen another course. They have found out for us that the pressure-value of a mercury column is different at different temperatures arid in different latitudes, and have explained that when we want to use the readings of a mercury barometer we must make corrections on these accounts as well as on account of the expansion or contraction of the brass scale which is used for the readinof. We must also allow for the index error of the barometer which is determined by means of a comparison with a recognised standard instrument. Even then we are not at an end, for the reading must be reduced to sea level before it can be charted on a map. Now it has been explained that in these days a meteorologist thinks in maps, so before barometer readings can be used in meteorology they have to be corrected for index error, for the temperature of the mercury and the brass scale and for latitude, and also they must be reduced to sea level. The Practice of the last Sixty Years. This string of corrections and reduction has been hitherto regarded as an unnecessary refinement for the seaman who only wants to know roughly what the barometer is doing. All that an observer has been asked to do is to set down in the log the reading of the barometer and attached thermo- meter, and of course the latitude, and leave the rest to the Office. Somewhere in the log will have been entered the height of the barometer cistern above sea level and the number of the barometer of which we keep a record and know the index error. So we have had all the materials necessary for making these elaborate corrections, and they have all been duly made during the last sixty years with the exception of" the correction for latitude, which has only been generally regarded by meteorologists as necessary within the last ten years. Pressure in Pressure-units. It is the introduction of this new correction which has made a difference to our outlook. We have been asking- ourselves what is the real meaning of the philosopher's Pressure in Pressure-units. 3 requirements, that we should reduce barometer readings to a true length of mercury at the freezing point of water in latitude 45° — the reduction to sea level is our own affair — and the answer is because you want to compare pressures. To that our rejoinder is then : Why do you lead us up to the brink and stop short of the final plunge which is the mere multiplication by a single factor ? Why not give pressures ? Why give " inches " of mercury ? And we find that millibars are a very convenient expression of pressure. If there were no other reason, the fact that ordinary sea level pressures are about 1000 millibars is a very good one. There are others, about which there is something to be said later. For the present let us consider why we should aim at giving pressures in pressure luiits. The Consequence of " Thinking in Maps.^^ We have said that a meteorologist thinks in maps, and the sailor at sea if he wishes to take part in the use and practice of modern meteorology must do so likewise. To do so he must use corrected and reduced barometer readings and not the crude uncorrected readings ; they are worse than useless on a meteorological chart ; if you are not sure about them they are distracting to the last degree. Ships within hail by "wireless." on the Atlantic exchange barometer readings by way of courtesy, and quite rightly, for they -might be very useful ; but the barometer is read in inches, corrected for temperature — still inches, reduced to sea level — still inches. None of them are actually inches in real life, but they are called so, and how can the recipient make any use of an observation unless he knows which is meant ? If the sender has omitted to correct for temperature the reading- is very likely out by the tenth of an inch, or for height it may be nearly as nuicli out, and the index error may be almost anything. These differences may seem trivial to the onlooker, but they completely upset the weather on a map. It is fortunate that the observer is out of hearing when the map-maker is brought U]) by a barometer reading that will not fit. The necessity for simplicitu in dealimj with Barometric Readings. It is therefore very desirable that a barometric reading at sea should be reduced, there and then, to its meteorological form BO that it may be immediately put side by side for com- parison with any other reading that comes from another shi[> lll»79 J^ 4 Preface. or from shore, or with a weather chart or a synoptic chart of mean pressures. But the conventional process of correcting and reducing barometer readings is so tedious that only a central Meteor- ological Office could be expected to face the drudgery m the form in which the natural philosophers have presented it. Now we have found that when we confine our attention to the accuracy required in practical work the drudgery is very largely artificial and unnecessary, rooted in the history of the barometer. With the new graduation of the barometer we propose to simplify the process and to arrange that in ordinary circumstances the corrections will be small and tlie trouble infinitesimal. An explanation of the vrhole process is set out at length on pp. 11 to 14, and a specimen in practical form is given on pp. 15, 16. With the old graduation the reading- would have needed correction for temperature and latitude as well as for index error jmd reduction to sea level unless the ship was in latitude 45° and the barometer exposed to a tempera- ture of 28° F., a combination of circumstances which must be very rare ; we propose, on the other hand, to have a baro- meter which will need no correction at all in the most usual conditions and, generally speaking, only a small correction that can be carried out in one's head from a reading of the attached thermometer, instead of requiring an elaborate series of tables of corrections that, so far as we are aware, no seaman ever uses. So our advice to seamen w^ho would help one another with their barometer readings is : — First, use a mercury barometer because aneroid readings will not plot on a map. The index error is always uncertain. Secondly, correct your readings for index error, tempera- ture, latitude and height above sea level, and show that you have done so by quoting the result in millibars. Ahsolate Temperature for the Attached Thermometer. In graduating tlie attached thermometer we have used a scale of temperatures which is named after Lord Kelvin, because it starts from a certain zero which he called absolute, and which is of great theoretical importance. There are reasons for using this scale for all meteorological purposes, Avhich appeal to those who have to deal with the physical problems of the atmos2)here and Avhich will ultimately make themselves felt in ordinary meteorology, but we are not concerned wdth them just now. What we have in mind for the moment is that the temperature of the barometer is one Ahsolute Temperature. 5 thing, it enables one to g-et the pressure of the atmosphere from the barometer reading ; and tlie temperature of the air for meteorological purposes is quite a different thing. The barometer can properly be kept indoors comfortably warmed, but the thermometer that is intended to give the meteoro- logical temperature of the air must be out of doors in a screen which lets the air go through it, and at this stage of our work it is an advantage, rather than otherwise, that the thermometer, which is to give the temperature of the mercury column merely as an auxiliary for getting an accurate value of the pressure should be graduated differently from the thermometer which is intended to take the temperature of the open air. We use the temperature given by the attached thermo- meter to enable us to get a correct measure of pressure from the barometer and for nothing else, so for the time being we use a special thermometer. And when we have employed its readings in the way described on p. 12 we get the pressure in pressiu'e-units which we call millibars. A Word 0/ (■ aid ion. It must be remarked here that since millibars are pressure- units a pressure in millibars represents the final result, and it would be wrong to use the word for anything but the final result. Consecpiently we ought to avoid using it until the final result with the correction for temperature has been obtained. The reduction to sea level is not necessarily included, but it is best to include it because the reduction to sea level is never a large matter at sea, but just large enough to spoil the use of the reading to your neighbour if it is omitted. So the first reading of the barometer is merely a meaningless number until the tem])erature cor- rection is applied and the sea level reduction can be applied at the same operation. Standard Temperature. To show whether any correction is necessary we have the '■ standard temperature " clearly marked on th(? instrument. That is the tem})erature of the attached thermometer at which no correction would be required in latitude 45°. A different temperature would be apj^ropriate in another latitude which we call the " fiducial tenqierature " for that latitude. So that all we require to know for correcting a reading is the fiducial temperature and the actual temperature of the attached thermometer and the appr<»]iriate correction for the difference 111*79 c -' 6 Preface. of the two. If there is no difference there is no correction. When the correction has been applied the process is at an end and the result is the real pressure in millibars and the citation of a pressure in millibars is prima facie evidence that the correction has been applied. There is another way of deal- ing with the matter by which the correction is given by a sliding scale alongside the attached thermometer. It is quite easy to work in practice but lengthy to describe. An account of it is given in a paper by Mr. E. Gold in the quarterly Journal of the Royal Meteorological Society. The final result is, of course, the same. If the original reading of the barometer is cited, please do not forget that the reading of the attached thermometer, the latitude, the index error, and the height above sea level, have all to be given as well before anybody can make use of the observation by plotting on a chart and that is the only way in which barometer readings can be of any real use in these days. Some Facts and Figures about Pressure in Millibars. The reader may be interested in the results obtained by observation of pressure expressed in millibars. Here are some ap])roximate figures for values between which the pressure usually lies at different latitudes along the meridian of 30° W. The values are based upon infor- mation given in terms of inches in the " Barometer Manual for the Use of Seamen," which has been modified by the application of the correction for latitude : — North Latitude. Winter. Summer. 60° 50° 40° 30' mb. mb. 963 to 1024 975 „ 1026 992 „ 1034 1006 „ 1028 mb. mb. 990 to 1024 1000 „ 1027 1012 „ 1032 1013 „ 1027 Tropic of Cancer Equator Tropic of (Jnprieorn 1008 „ 1022 1004 „ 1011 1007 „ 1019 1013 „ 1023 1006 „ 1013 1011 „ 1026 30° 40° 50^ 55° 1003 \, 1022 996 „ 1023 971 „ 1012 962 „ JOOi") 1009 „ 1029 994 ,, 102S 970 „ 1024 959 „ 1018 South Latitude. Summer. Winter. Pressure in Millibars. , . 7 It will be noticed that on the whole a deficieucy of pres- sure is shown in the Southern Hemisphere which, unless it be an accident of figures due to the limitation of the area mapped, is a very remarkable fact. The average frequency of occurrence of pressure in the British Isles is set out in the following table. The numbers of days given are those on which the lowest barometer read- ing of the day has been found to be within certain limits of '• low " barometer, and on the other side when the highest reading is within certain limits of " high " barometer. DiO-950 ... 1 day in 5 years 950-960 ... ldaviii2 years 1 day in 5 years — ' 960-970 ... ?} (lavs a year 2 days a 3'ear 1 day in 10 years. 970-980 ... 10 days a year 6 days a year 3 days a year. 980-990 ... 26 days a year 21 days a year 14 days a year. 990-1000... 57 days a year 42 days a year 39 days a year. 1020-ioao... 109 days a year 123 days a year 133 days a year. 1030-1040... 29 days- a year 37 days a year 34 days a year. i040-]or)0... 1 day a year... ;') days a year 3 days a year. Above 1050. 2 days in 5 years 1 day in 3 years — I'hc highest recorded sea-level pressure in the IJritish Isles is 1U53*5 mb. at Aberdeen on 31st January, 1902, and the lowest 925 mb. at Ochtertyre on 26th January, 1884 ; the same figure was reached at sea in the Atlantic on ,5 til February, 1870. The daily range ot" pressure in tropical seas, the most characteristic feature of continuous reccn-ds of pressure in those regions, is about 2 J mb. In British seas the average daily range is 0*7 mb. and is quite noticeable on a barogram whenever the pressure is steady. The fall of pressure with height is 116 millibars for the the first kilometre or 1 millibar tor 9 metres or 30 feet of height close to sea level. The change for greater heights is very closely represented by 3*4 multiplied by (pressure in millibars divided })y temperature in degrees absolute). B ;{ ii'.try Preface. Mercurii versus Aneroid. Wliy not use an aneroid ? It is very convenient on board yhi}). so easily read with its open scale, requiring no cor- rection for temperature if it is properly compensated and no correction for latitude. That is all delightfully true, but it ha^ an index error which nobody has yet managed to deal AvitJi. The reading depends not merely on the pressure, of which it records the variations with admirable facility, but also upon its past experience which it does not disclose. So that when you have got the reading so easily and minutely, there may still be, and frequently is, an error of which you are vourself quite unconscious but which is big enough to outweigh all the corrections of the mercury barometer and make the readings useless for plotting on a map. The index errors that are reported to us from time to time in connexion with the work of the Office run to tenths of an inch, some- times to half an inch, and when the readings are plotted on a map the result, in the majority of cases, is simple confusion. Education and Metric inifs. Allusion has been made to other advanta^'es of the expression of pressure in millibars which have at present little to do Avith the daily use of the barometer at sea. They have to do with the relation of meteorology to dynamics and physics and to the place of the stud}^ of the weather in public education. It will easily be understood that in reality meteorology is the study of the physics of the atmosphere. jSTowadays in many of our schools they teach the rudiments of physics and Ave hope that presently they will teach the rudiments of meteorology. But if any real good is to come of it they must teach meteorology in relation to physics. Now, the teachers of physics the world over use metric units and so do all those Avho are concerned with electricity and magnetism which are s])ecial depart- ments of physics of world-wide application. The use of our customary units is therefore an obstacle in the way of a young student's learning meteorology effectively, and if we recognise the im])ortance of the future as well as the present it is worth while to remove that obstacle so that meteorology may be taught and learned in combination with physics. Scale of Eqitivalents, 9 Scale oj Equivalents of inches, millimetres and millibars. Schools and colleges have not hitherto used millibars for the expression of pressure. The unit that they have em- ployed is the millimetre of mercury at the freezing point of water in latitude 45°, which has little to recommend it and some disadvantages compared with the inch of mercury in the same locality and conditions. The Conference on the Saving of Life at Sea has recently adopted a code which, following the Continental model, is based on millimetres, so that millimetres have come within the seaman's practice ; therefore it may be convenient to give here a short table of equivalents of the three scales, inches of mercury at 32^ F. (0° C.) and latitude 45"^, millimetres of mercury at 0° C. and latitude 45°, and millibars. Inches. Millimetres. Millibars. Inches. Millimetres. Millibars. 28-0 711-2 948-2 29-6 751-8 . 1002-4 28-1 713-7 951-6 29-7 754-4 1005-7 • 28-2 716-3 954-9 29-8 756-9 1009-1 28-3 718-.8 958 -S 29-9 759-5 1012-5 28-4 721-4 961-7 30-0 762-0 1015-9 28-5 723-9 965-1 30-1 764-5 1019-3 28-6 726-4 968-5 30-2 767-1 1022-7 28-7 729-0 971-9 30-3 769-6 10J6-1 28-8 731-5 975-3 30-4 772-2 1029-4 28-9 734 1 • 978-6 30-5 774-7 1032-8 29 736-6 982-0 30-6 777-2 1036-2 29 1 739-1 985-4 30-7 779-8 1039-6 29 2 741-7 988-8 30-8 782-3 1043 -U 29 3 744-2 992-2 30-9 784-9 1046-4 29-4 746-8 995-6 31-0 787-4 1049-8 29-5 749-3 999-0 31-1 1 789-9 1053 1 Rainfall. With the change of pressure units is associated the change from inches, or in practice hundredths of an inch, to milli- metres in the measurement of rainfall. It is as well to mention that here, because rainfall is such an important item in atmospheric changes over sea or land, though for seamen in practice it does not generally count for more than two or three letters of the Beaufort notation. We wish, however, to know more about rainfall at sea and are taking steps in that direction. Here let me say merely that the millimetre, besides being a metric unit, is a better unit for raini'all than the hundredth of an inch. It is almost exactly four hundredths of an inch, so that usino: millimetres instead ot hundredths of an incli is exactly like counting one's ir.'-9 B 4 1 Preface. money iu pence instead of farthings ; and just for the same reason that the penny is, in practice, better than a farthing for counting money, so iu counting rainfall the millimetre is better than the hundredth of an inch. It economises figures without any sacrifice of practical utility. Practical Utilitij. It is therefore not from any lack of appreciation of the claims of practical utility that we have taken up the change of graduation of barometers and other meteorological instru- ments. On the contrary, immediate utility leads to closer attention and brings with it the greater accuracy which is now wanted for practical purposes. Here, for example, is an interesting point which might perhaps be solved by closer observation by anyone who has both a mercury barometer and a good aneroid. We find an instruction in the Marine Observer's Handbook that when the barometer is "pumping" the lowest point of the excursions of the mercury is to be taken as the reading. This is on the ground that the con- striction of the tube prevents any appreciable flow of mercury along the tube during the up and down motion of the ship and that an oscillation which is visibly present must there- fore take place from a break of the thread at the constriction. This basis of procedure is supported by the instinct and insight of practical observers, but to the theoretical mind it seems unsound ; the mean position should give the true reading of pressure in a pumping barometer. Between these two views there is a difference of about a millibar, so that it is well within the scope of observation. Up to now, while we have been comjjiling mean values, we have com- forted ourselves with the assurance that in the long run such differences will not affect the final result, but now that we think in maps that assurance fails. We are actually in want of a definite answer that will carry conviction both to tJie practical and theoretical mind. The re^dsed method of expressing the measurements of pressure has been adopted for the Meteorological service of the French Government from the beginning of this year. W. ]S^. SHAW. Meteorological Office, June 19, 1917 Note. — Further details for the manipulation of the instruments on board observing ships, and instructions for keeping the " Meteorological Log," are given in The Mmiae Olsei-ver''s Handbooh (M.O. 21(S), which is intended for use with the various forms of meteorological register s^^pplied 1>y the Office. Correction of Barometer Readinfis. 11 CORRECTIOX AND REDUCTION TO SeA-LeVEL OF MiLLIBAR Barometers. Xote. — Barometers graduated to read in millibars are provided with -an attached thermometer graduated according to the absolute scale of centigrade degrees and the references to temperature in the follow- ing instructions are to the readings on that scale. In quoting the temperatiire the degree mark is omitted and instead of it a small "a" follows the number. Thus 273a on this scale corresponds with the freezing point of water, that is 0° C or 32° F, and 283a corresponds with 10° C or ")0° F. A step of 10a in temperature is the same as a step of 18° F. Standard temperature as shown on the Certi- ficate. — The barometer will have been certified as correct in latitude 45° at a certain temperature which we call the standard temperature. The certificate means that when the temperature has the specified value, the barometer reading will give the true value of the pressure in iniJlibars at the level of the barometer cistern in the specified latitude. Example. — Barometer M.O. A. 2074. The standard temperature is 286*5a, tliat is in latitude 45° tl e barometer reads correctly at 2S6"5a which is the same as 56"o° F. With this information it is easy to make allowance for (iifterence of latitude and difference of temperature ; it is also easy to allow for height above sea-level in a similar manner and so put the observer in the position to compare his rea"-60°. Fiducial Temper- Lat. Lati- tude. Fiducial Temper- Lat. 1.-. 30 . Lat. 0°-15^. Lati- tude. Fiducial Temper- Lati- tude. Fiducial Temper- Lati- tude. Fiducial Temper- ature. ature. ature. ature. ature. o N.orS. a. o N.orS. a. N'.orS, a. o N.orS. a. o N.orS. a. 75 310 ()0 304 45 296 30 288 15 282 74 :no 59 304 44 295 29 288 14 282 ?:] :50n 58 303 43 295 28 287 13 281 72 HO!) 57 303 42 294 27 287 12 281 71 :k).s 56 •',02 41 293 26 286 11 281 70 308 55 302 40 293 25 286 10 281 t;u 308 54 301 39 292 24 286 9 281 t; 2S0 •;5 3()(» .')() 290 35 290 20 284 5 280 Cl 306 41) 299 34 290 19 284 4 28(> (;:i 305 48 298 33 289 18 283 3 280 ()2 ;U)5 47 297 32 289 17 283 9 280 C.l 304 46 297 31 288 16 282 1 280 (]() 304 45 296 30 288 15 282 280 16 I'ref(ue. II. Correct for tJie difference between the reading of the attached Thermometer and the Fiducial Temperature by means of tlie following' : — Tables of Corrections for Tp^mperature. (A.) Actual Temperature above the Fiducial Temperature. Actual "^ r Fiducial ^ I Tempera- Tempera- I- 1 l'^ lure. Subtract J I tnre. /I G° ro 1-2 8° I a"" I 10' i i i 1-3! 1-5 I 1-7 mb. (B.) Actual Temperature below the Fiducial Temperature. Fiducial ^ r Actual '- — ' Tempera- Tempera- 11 ture. J I Add ture. ]° •2 2° •3 3° •5 4° 5° •8 6° 1-0 7° 1-2 8° 1-3 9° ; 10^ 1"5 . 17 mb. Relation of millibars to inches. {See Figure ^, opposite.) - — The units on the absolute scale -are related to one another Hfi follow : — 10 millibars = 1 centibar 10 centibars = 1 decibar 10 decibars = 1 bar. The millibar is adopted as the working- unit in the Daily Weather Service (see p. 1). The scale of millibars is related to the ct>nventional scale of mercury inches as follow : — Normal pressure for British Isles, 29'92 mercury inches = 101 o*2 millibars. Highest recorded pressure for the British Isles, ol'll mercury inches = lOoo'o millibars. I>owest recorded presstn-e for the British Isles, 27'o3 mercury inches = 925\5 millibars. 1 millibar = '029 mercury inch. Thus one-tenth of a millibar corresponds with "003 mercury inch, which may be taken as the limit of accuracy to which it is possible to read a barometer under favourable Avorkino" conditions. Correction of Barometer lieadliKis. V GUADUATIOX OF KeW PaTTEKX L')AllO-METF.li IX MlLLIBAliS AND Il^CIIKS, AND OF ITS ATTACHED TlIEintOMETEU IN Dbgkees Absolute and Degrees FAiiKExriEir. K>&- os- 103- --■ 31 5Z Centibars, with luilli- bav divisions. Indies, witii .iV-iiicli divisions. ^%Q■ 310 «— ::= 300- -uo 590- ?-30- -So -50 260 — -20 : ;o Teniperimiiv Temperature in Centigrade Falareulieit. Degrees from Absolute Zeni FivJ. B. 18 THE SEAMAN'S HANDBOOK OF lETEOEOLOOY. INTRODUCTION. The purpose of this Handbook is to supply some iaforma- tion with reference to the conditions of the atmosphere as regards Pressure, Temperature, Humidity, Wind, and Cloud, which collectively constitute what is called iveather, and with reference to the change which is constantly taking place in one or more of these conditions implied in the expression a change of tceather. Also to explain a method of foretelling weather, especially changes in weather conditions which may be attended by an increase of wind to gale force on the coasts of our islands. In this second edition of the Handbook, pressure values in millibars have been added to the reproductions of the Synoptic Charts of the Daily Weather Report which are included among the illustrations. A chapter on Icebergs and other forms of drifting ice is included, because of the close connexion which obviously exists between ice frequency in the North Atlantic and the Southern Ocean, and the meteorological conditions in those oceans and the respective polar regions which adjoin them. In an appendix, mirage and its effects in causing displace- ments of the apparent sea horizon are referred to and these phenomena explained. M. W. CAMPBELL HEPWORTH. \4:th Januani, 1915. 19 INTRODUCTION TO THE THIRD EDITION. To the ocean -g'oino- .sdilor, the Barometer Manual for the Use of Seamen affords information of immediate import, but before much that it contains is intelligible, it is necessary that he become acquainted with the rudiments of meteorology. This Handbook was written, therefore, as mucli for the ofuidance of the ocean-o-oingf sailor as with a view to assist- ing the custodian of the Fishery Barometer to interpret the readings of the instruments entrusted to his charge for the benefit of the fishing community in his district. Although Chapters VII and VIII refer more directly to weather conditions on the Coasts of our Islands than to weather at sea, yet the references to anticipation of changes of wind and weather; the sequence of weather during the passage of a depression over the observer's locality ; the motion of clouds of cirrus type in presaging weather changes ; weather signs ; and types of weather conditions apply to •ocean meteorology as well as to that of our seaboards. When making a passage across the North Atlantic, in- formation included in these pages may assist the sailor to anticipate changes in weather conditions, if he takes into account those changes that occur in his geographical position in relation to that of the changing distribution of barometrical pressure. In the Southern Ocean also, Avhile running down the easting, the same obtains, if for the word " northerly,'' "polar" be substituted and for the word ''southerly," " equatorial." After the observer has become accustomed to reg-ard the utmosphere as an ocean of air of varying density, the distri- bution of which is constantly undergoing changes through the intervention of rapid air-currents ; the oscillations of the mercury coluinii ; the changes of wind, and the motion of upper clouds will become more and more intelligible to him and his anticipation of weather changes increasingly accurate. M. W. CAMPBELL HEPWORTH. April, 1917. 20 CHAPTER I. The Atmosphere and its Circulation. Completely surrounding our globe there exists a layer ol air, the atmosphere, which partakes of its rotatory motion. The mutual rotation of this layer of air, or gaseous envelope, with that of the globe, would remain constant at all points were it not for the occurrence of local chaiiges in the temperature, pressure, humidity, and, therefore, in the density of the air, which have the effect of disturbing its equilibrium and thus producing winds. At anv point, therefore, on the earth's surface where the atmosphere happens to be in a state of equilibrium and a complete calm prevails the air is in truth moving with great rapidity Avith the earth. Composition of the Atmosphere. The atmosphere may be likened to an aerial ocean of a certain depth, consisting of dry air and aqueous vapour intimately mixed ; and, for the most part, existing as invisible gases. It is essentially a mixture of oxygen o,nd nitrogen gases, its average composition by volume being : — Nitrogen ... 77 Tl per cent. Oxygen ... ... 20-65 Aqueous vapour ... ... 1-40 „ Argon ... 0-79 Carbonic acid ... 0-Oi Height of the Atmosphere. The actual height of the atmosphere is not yet known, but the deduction drawn from careful observations, taken at different points, of meteors which become luminous when entering the atmosphere is that it must be at least one hun- dred miles high ; and that in an extremely attenuated form it even reaches to an altitude of fiA'e hundred miles. Density of the Atmosphere. Following the same laws as ordinary gases, dry air is densest near the surface of the earth, and its density diminishes upwards. Seven miles above the surface of the Cause oj Wind. '2\ ■earth it has one quarter the density it has at the surtaoe, at fourteen miles one sixteenth of that density, and at twenty- pens that when the land in some '2''2 The Atmosphere and its Circulation. locality becomes heated the temperature of the air in contact with it increases, the air expands, is displaced by cooler and therefore heavier air, and rises towards the higher regions of the atmosphere, while the upper layers flow from it towards cooler regions. The equilibrium at the surface is thus destroyed, for the baron letrical pressure is greater in the relativel}^ cold regions than it is in the warm, and in order to restore the equipoise the heavier cold air flows in below. Two distinct air currents are thus produced, an upper current setting outward from the heated region and a lower current setting inwards towards it. The currents of wind will have a velocity dependent on the difference between the high and low pressures. At the earth's surface differences in temperature between adjacent localities are constantly occurring, diurnally as well as with the change of season. These differences in tempera- ture arise from a number of causes, among the most potent of which may be mentioned geographical position, with regard principally to latitude ; distribution of land and water ; greater or less abundance of cloud, or rain, or quantity of water-vapour in the air. This explanation of the inequalities of atmospheric pressure at the earth's surface cannot, however, be regarded as by any means the complete solution of the problem, because instances of the association of warm air with high pressure and of cold air with low pressure are common. The question of pressure-distribution is too intricate to be disposed of thus. The interchange of air which takes place between two adjacent areas of unequal barometrical pressure is not direct from ouQ pressure to the other. Although there is an incli- nation of the air current outwards from the high pressure towards a lower pressure, and an inclination of the air current inwards towards the low pressure from the higher pressure, the flow of air is generally round areas of high or low pressure, for the reasons which are given below. Effect of the Earth\s rotation upon the Wind's direction. The velocity of the earth's motion is greatest at the equator and gradually diminishes towards the poles ; so that a current of air flowing from a lower to a higher latitude is deflected to the eastward, since the velocity of the earth's motion is greater in the lower than it is in the higher latitude ; conversely, a current of air flowing from a higher to a lower latitude is deflected to the westward, since the velocity of the earth's motion is not so great in the higher C>/c/onic and Antkyclonic Civeulatmhs. 2% as it is in the lower latitude. For instance, in the Northern Hemisphere a current of air settino; northward — in other Avords, a southerly wind — instead oF retaining its direction is deflected to the right and becomes a south-westerly wind ; and a current of air settino- southward — that is to sav, a northerly wind — is deflected to the rifjht and becomes a north- easterly wind. In the Southern Hemisphere the converse obtains ; the northerly wind is deflected to the left and becomes a north-westerly wind, and the southerly wind is deflected to the left and becomes a south-easterly wind. In this manner, when air currents set towards an area of low barometrical pressure from the relatively high pressure areas by which it is surrounded, they are deflected to the right in the Northern Hemisphere and to the left in the Southern Hemisphere by the earth's rotation, and, instead of flowing direct towards the centre of depression, will acquire a motion round it, but inclined inwards towards the centre. For this reason the wind circulating about an area of low pressure, or depression, in the Northern Hemisphere will have a movement against that of watch hands, while in the Southern Hemisphere the movement will be in the same direction as watch hands. Similarly, when the air from an area of high pressure sets outwards toward the relatively low pressure surrounding it, it is deflected either to the right or to the left according to the hemisphere in which the wind system is situated, and will acquire a motion round the area of high pressure, but inclined outwards from it. Thus the wind circulating about an area of high barometrical j)ressure will in the Northern Hemisphere have a movement in the same direction as watch hands, and in the Southern Hemi- sphere will have a movement in a contrary direction to watch hands. Cyclonic and Anticyclonic Circulations. Winds that have a circulation round an area of low pressure, or depression, are termed cyclonic, and such a distribution of barometrical pressure and wind is known as a cyclonic system, or cyclone ; winds that circulate about an area of high barometrical pressure are called anticyclonic, and such a distribution of pressure and wind is known as an anticyclonic system, or anticyclone. There is a general statement of the facts bearing on the relation of wind direction to ])ressure thus explained, which is known as Buys Ballofs law, because the principle it demonstrates was first publicly announced in Europe ))v 24 Temperature and Hiimidity. Professor Bays Ballot, of Utrecht. It may be enuuciuted thus : — In the Northern Hemisphere, stand with your face to the wind and the barometer will be lower on your right hand than on your left. In the Southern Hemisphere, stand with your face to the wind and the barometer will be lower on your left hand than <-)n vour rio'ht. Origin of the term " Cyclone^ In recent years the term cyclone has largely becDine associated with atmospheric disturbances in which the wind blows with ^'iolence round a central area of low barome- trical pressure, and is frec|uently used to express the force of the wind rather than a characteristic distribution of pressure and wind. Tlie term cyclone was. however, originally adopted by Piddington, in a publication entitled " Sailor's Horn Book" (1848), in connexion with the classification of winds. In introducing it, he says :— I suggest that we might for this last class of circular, or highly- curved, winds adopt the term " c^'clone," from the Greek KucXor (which signifies, amongst other things, the coil of a snake), as neither affirming the circle to be a true one, though the circuit may be complete, yet expressing sufficiently the tendency to circular motion in these meteors. Thus in a cyclonic depression the wind has a teyidency to ■circulate round an area of relatively low pressure ; it may be of moderate force, and in some parts of the system even light, or it can be strong to a gale, and, especially in the tropics, may attain to the force of a hurricane. Cyclonic depressions when formed travel, generally speaking, in temperate latitudes towards some easterly point in either hemisphere. An account of the instruments used in determining the physical condition of the atmosphere and in detecting the ■occurrence of important changes is given in Chapter XI. CHAPTER II. Temperature and Humidity. Temperature. Temperature has been defined as the thermal con- dition of a body which determines the interchange of heat Temperature. 25 between it and other bodies. Tliis interchange of heat occurs in one or more of three ways : by conduction, bv convexion, or by radiation. Heat imparted by conduction involves contact with, or a near approach to, a warmer body ; and is transferred from particle to particle. If one end of an iron rod be placed in a fire, the heat communicated to that part of the rod will be transmitted to other parts by conduction. A similar result may be obtained by substituting a wooden rod for the iron one ; but in this case, although the degree of heat generated may be the same, the heat transmitted to other parts of the rod will be less, because wood is not so good a conductor of heat as iron. Solids are better conductors of" heat than liquids, gases are the worst conductors, metals the best. By conduction changes in the temperature of the earth's surface, which becomes heated durino- the day and cooled during the night, are communicated to the layers of air in contact with, and immediately above, the earth. Interchange of heat by convexion is effected ^vhen air, heated at the earth's surface, ascends and flows away, while the colder air by which it is surrounded flows in to take its place. Over the whole globe heat is constantly being trans- ferred from one locality to another through the agency of winds and of ocean currents. All interchange of heat is taking place, moreover, at all timetj between bodies that are freely exposed to each other ; and the process by which heat is thus communicated is called radiation. The feeling of warmth derived from a fire, or other source of heat at some distance, is produced by radiation. The communication of heat by radiation proceeds, nor from one particle to another, but through air or vacuum, and even throuo'h solids. Radiant heat is not hciit in the ordinary sense of the word, but a form of energy which ]iroceeds in straight lines and diverges in all directions. Tlie* atmosphere derives its heat, directly or indirectlv, entirely from the sun. Deprived of this source of heat the temperature of the atmosphere would fall far below the lowest temperature that has ever been recorded in any geographical position on the earth's surface. The actual temperature of the atmosphere depends not so much upon the direct rays of the sun as upon the conduction and radiation from the surface of the earth heated by the sun's rays. The air is not heated to any perceptible degree by sunshine, tlie sun's heat is traiismitted through air as sunlight is transferred through glass ; the surface of the earth is heated by the sunshine and the air is warmed by contact wiih the earth : bv radiation and bv convexion. 26 Temperature and Humuiifji. Diiferenees in the character of the soil in different localities have a considerable effect on the distribution of atmospheric temperature over large areas ; heavy soils that have their particles closely packed together conduct heat better than soils that are loose and porous, because the air becomes entangled in the latter, and the conducting power of such soils is diminished. Land that has recently been ploughed absorbs and radiates heat more quickly than pasture land. A sandy ^oil heats the atmosphere above it more than a soil covered with vegetation. An expanse of sand heated by the sun's rays is soon cooled after sunset by terrestrial radiation : that is, by radiation from the earth ; but this is not the case with an expanse of water, for when the sun's rays fall on water, the heat produced, instead of being arrested at the surface, as is the case when they fall upon land, penetrates far below the surface, and by vertical and horizontal ciu'rents is diffused to a considerable depth over a large expanse. Water absorbs heat, stores it, and conveys heat whither it flows. The capacity of air to carry heat is inconsiderable as compared with that of water of the same volume. Islands, peninsulas, and all localities situated in the neigh- bourhood of the sea, have for this reason more equable and frequently milder climates than is the case with localities far removed from the seaboard in the same latitude, and in other respects similarly situated. The sun's rays have their greatest effect where they fall perpendicularly on the earth's surface ; their effect diminishes as their obliquity increases, so that where the rays fall vertically, within the tropical belt, the effect is at its maximum, and it decreases poleward. Were it not for the modifying inflaences due to proximity to the sea, to winds, and to ocean currents, differences in the temperature of the air over land areas in different geographical positions on the earth's surface at the same time would depend almost entirely upon latitude. It has been found that hillsides surrounding lakes or other laro'e sheets of water derive heat from the reflection of the o water surfaces ; also that the lemperature of the air in valleys is raised by refleciion and radiation from mountain «ides. In (Contrast with the warming effects due to solar radiation and reflection is the cooling of the earth's surface, and consequent lowering of the air temperature above it, which takes place through terrestrial radiation. During the day- time, as well as during the night, the earth parts with heat received from the sun and pours it into space. The heat Temperature. 27 received from the sun during the day is, as a rule, greater than that which the earth parts with, but as the day declines and the rays of the sun fall more obliquely upon the earth's surface, these conditions are reversed, the temperature com- mences to fall and continues to do so until sunrise. Tc r rest rial Ra dia tio n . The loss of heat by terrestrial radiation is not so great when the sky is overcast or cloudy as when it is clear, because clouds intercept the heat thrown off £i*om the earth and radiate it back ; for this reason nocturnal cooling is greatest when the sky is cloudless. Wind churns the surface layers and mixes them with the upper layers, and therefore any disturbance of stagnant air cooled by radiation is necessarily accompanied by a rise in temperature, the cooling of the earth's surface and of the air resting upon it, by terrestrial radiation, is consequently not so great when the air is in motion as is the case when it is calm ; as the force of the wind increases, other influences remaining the same, the effect of terrestrial radiation is less apparent. Decrease of Temperature with Height. The temperature of the air decreases with height to a great altitude,* for as the height increases there is neces- sarily a loss of warmth derivable, from the heated surface of the earth. Moreover, as most of the sun's heat reaches the earth's surface, very little of it being absorbed in passing through the atmosphere, those layers, or strata, oi" air which lie nearest to the surface acquire by conduction more heat than those at a greater altitude, and tlie layer in contact witli the earth receives most of all. Decrease of temperature with height is caused by yet another process : air in the higher regions of the atmosphere is subjected to less pressure than obtains at lower altitudes, * Investigations of the upper air, conducted by means of hallons- sondeti (sounding balloons) carrying self-recording instruments, have shown that the fall in temperature with height ceases suddenly at au elevation which varies from day to day, but is roughly estimated at about r, miles. Above this height temperature changes but little in a vertical direction, but shows a slight tendency to increase. The thickness of this layer of ecjual temperature, or stratosphere as it is called, has not yet been ascertained ; its lower surface, although irregular in height, may be considered as roughly horizontal. In eciualorial regions, where observations have been made in this con- mjxion. the fall in temperature with altitude has been found to continue to a height of about '.♦ miles. 28 Temperature and Humidity. and, while expanding under pressure, the heat required to cause the expansion becomes latent, or concealed (Latin : latere, to lie hid), thus a fall in temperature ensues. At the same time it is not unusual in still, clear weather during summer for the temperature at a considerable eleva- tion, even on the top of lofty mountains, to rise under the influence of solar radiation to the same height as it attains in the neighbouring valleys or adjacent places at sea level. Seasonal Range of Temperature. Within the United Kingdom the range of temperature, shown by a shaded thermometer in the open air, is ordinarily about o9a (70° Fabr.), that is to say, the readhigs of the instrument during the year may be expected to range from 261a (10° Fahr.) to 300a (80° Fahr.) ; but in the'hardest winter-frosts the temperature occasionally falls below 261a, and on exceptionally hot days in summer not infrequently rises above 300a. In our islands the range of temperature is less than in many other countries, owing to the influence of the ocean. Li the interior of the great northern continents, where no relatively warm sea winds mitigate the cold of winter, and no relatively cool sea breezes allay the heat of summer, the extremes of cold and heat are often great. On the other hand, in islands of smaller extent than our own, situated in warm latitudes, the range o£ temperature is considerably less than obtains in the British Islands. The seasons, as determined by tlie march of temperature, group themselves in the temperate zones into four periods, which in the Northern Hemivsphere are as follow : — Spring — March, April, May ; Summer — June, July, August ; Autumn — September, October, November ; AVinter — December, January, February. On an average July is the warmest month, January the coldest ; April has about the same average temperature as October. The date of the highest mean daily temperature at Greenwich during the 65 years 1841-1905 (29la, 64° Fahr.) occurs on the 15th July, while that of the lowest mean daily temperature (27(ki, 37-5° Fahr.) falls on the 12th January. Diurnal Variation of Temperature. There is also a diurnal variation of temperature which is well marked. Normally, temperature is lowest at about sunrise, highest at about 2 p.m., but during the three summer months later. At Greenwich, however, taking the average of 20 years' observations, 1849-1868, the minimum Humidity. 2^ temperature occurred in suininer at eight Jiiinute.s after sunrise, in winter at 47 minutes before sunrise ; and the maxinnim temperature was recorded at '1 p.m., both in summer and in winter. At Kew the average time of minimum temperature, based on 35 years' observations, 1871-1905, is in summer eight minutes after sunrise, in winter 1 hour 17 minutes before sunrise ; and the time of average maximum temperature in summer is 3 p.m., in winter 2.20 p.m. The distribution of temperature at any given hour over a given area may be shown graphically on a chart by lines uniting localities at which the temperature is equal at the same instant of time. Such lines are called isotherms. Humidity. It has already been shown that air consists principally of nitrogen, oxygen, and aqueous vapour. Now, between the molecules of the two gases, nitrogen and oxygen, which are the chief components of dry air, minute particles of water are floating which have risen into the atmosphere in the form of vapour ; for by evaporation vapour is continually rising from water and other moist surfaces. Evaporation takes place from ice and snow also. The particles of water in the atmosphere are usually invisible because of their trans- parency. \)vy air does not change its gaseous state, and the quantity of dry air in the atmosphere remains constant, but such is not the case with aqueous vapour ; the quantity of vapour in tlie atmosphere is, by the process of evaporation and by condensation, constantly changing, and although where invisible, water-vapour is in a state analogous to that of a gas, nevertheless it can be turned into water and back again from the licpiid to the invisible gaseous state. The general term vaporisation is used to indicate the process of transition of a liquid into a gaseous state, the term traporation refers especially to the slow generation of vapour at the free surface of a liquid or any otlier moist surface. The pressure of all gases in an enclosed space is increased by a rise and diminished by a fall of temperature. As temperature rises the capacity of space for holding moisture increases, and as it falls diminishes and therefore, the warrner the air the larger quantity of moisture can it sustain in an invisible state. As only a limited quantity o{ water-vapour can be held in suspension in an invisible state in a definite volume of air. when this limit is exceeded particles of 30 Temperature and 1 1 umielity . water are formed which become visible as cloud, fog, or mivSt ; and when the air having received its full load of moisture has its temperature reduced, it becomes charged with moisture beyond the point of saturation, and the excess is precipitated in the form of rain, hail, or snow. Latent heat. When a liquid is in process of conversion into a gaseous form a large measure of its heat becomes absorbed, and therefore latent ; but, although the heat thus absorbed cannot be detected by the aid of a thermometer, and is im- perceptible to the senses, it is not lost, but hidden, and will reappear when the vapour is restored to its liquid state. As the conversion of water into vapour by the process of evaporation is attended with a decrease of temperature, due to the absorption of heat, so the conversion of vapour into the liquid state is attended with an increase of temperature, due to the release of latent heat. Formation oj Cloud. When the air, charged with aqueous vapour, is brought below the temperature of saturation cloud is formed, or, should the conversion take place immediately above the earth's surface, fog or mist. The cooling of the air may result from various causes : from intermingling with air of a lower temperature, from radiation, from contact with cold surfaces, and from the absorption of heat resulting from an ascent to higher altitudes. The aqueous vapour in the atmosphere rising from seas, lakes, rivers, and other moist surfaces, is caused by evaporation. The rapidity of evaporation varies with temperature, for a rise of temperature, by increasing the elaptic force of a vapour, will accelerate evaporation. Again, the raj^idity of evaporation varies with the quantity *)i vapour already in the surrounding atmosphere, the drier the air the more vapour is it capable of receiving, no evaporation being possible fr'om a liquid when the air surrounding it is saturated with the vapour of the same liquid. Evaporation is accelerated by renewal of the atmos- phere, for, when the air over a moist surface is stagnant it speedily becomes saturated when vapour is rising. Rapidity of evaporation is influenced also by the extent of the surface of evaporation. Wind, by dispersing vapour as it rises, promotes evapora- tion ; and, because any vertical movement of the atmosphere Relative Humidity. 31- is necessarily attended by a corresponding horizontal move- ment, the process of evaporation is to some extent favourable TO the production of wind. Relative Humidity. The ;»ir at all times contains some water-vapour, held in suspension, and the quantity of vapour it is capable of sus- taining depends ujjon the temperature. The higher the temperature the greater the capacity of the air for moisture When the air contains the largest possible quantity of vapour it is said to be saturated. Referred to a scale of to 100, the latter being assumed to represent saturation, and the former perfect!}' dry air, the quantity of moisture in the air can be expressed in figures, as a percentage of the amount necessary to produce saturation, the temperature remaining the same. This pro])ortion of moisture to saturation is known as relative humidity. The relative humidity is found by dividing the elastic force of aqueous vapour at the temperature of the dew-point by that corresponding to the temperature of the air as shown by the dry bulb thermo- meter ; tlie calculation may be greatly facilitated by the use of hygrometrical tal:>les published originally by the late Mr. James Glaisher, F.R.S., in the year 1845. The temper- ature of the dew-point is the temperature below which the air must be cooled in order to cause condensation of vapour. When, by radiation, the air is cooled below that temperature dew is deposited and latent heat released. The release of tliis heat has the effect of delaying the fall of temperature of the air, but only for a time, because, as radiation proceeds the air is chilled below, the original })oint of vapour condensation or dew -pointy with the result that heat is again released from its latent state ; and thus the process continues to be repeated. The lowest night temperature is, in con- sequence of this fluctuation of temperature, alternately a little above and below the point of vapour condensation, generally that of the dew-point. Elastic force of Vapour. The elastic force of aqueous vapour, or vapour pressure, i^ measured by that part of the barometric column which corresponds with the pressure of water- vapour in tlie atmosphere. For this j)urp()se we may regard the atmos- phere as composed of two separate gaseous ])arts : tirst the dry air, second the water-vapour. Each part contributes to the {treasure as measured by the barometer. The part which 32 Atmospheric Pressure and Wind. the water-vapour can contribute is limited by its saturation pressure, which depends UDon the temperature and nothing else. So long as that limit is not reached the water vapour, the weight of which, volume for volume, is five-eighths that of dry air, behaves like a gas. The course of procedure upon the gradual compression of a mixture of dry air and vapour is as follows: — At first the pressure increases inversely as the volume diminishes, both air and vapour behaving as if gases. Then, supposing the experiment began by vapour taking' 17 millibars (half an inch) and dry air 999 millibars (29^ inches), they would balance the increased pressure in the same pro]iortion until the saturation was reached, say at 25 millibars (three-quarters of an inch), which w^ould be at 295a (71° Fahr.). No more pressure can be put fipon the water vapour, it condenses while maintaining the fixed pressure. Thereafter when the air is comjDressed beyond saturation the dry air has to take any pressure beyond that fixed by the temperature of water vapour. CHAPTER III. Atmospheric Pressure and Wind. In order to illustrate graphically the distribution of pressure over a given area at the earth's surface, at a givea time, simultaneous readino-s of the barometer taken at a number of places are plotted in their respectiv^e positions on a chart, after bavins' been corrected for index error and temperature, and reduced to sea level. Isobars. Through the places at which the corrected barometer readings are equal, lines are drawn connecting them ; these lines of equal barometrical pressure are called isobars. Isobars are usually drawn on a chart for each tenth of an inch c_ for each five millibars. The relation between pressure distribution and wind distribution has been alluded to in Chapter II., where it was shown that over any area where atmospheric pressure is below that of the surrounding region a cyclonic circulation is (leveloped. Air currents, under the action of gravity, set towards an area of relatively low pressure from the relatively high pressure by which it is surrounded ; but, by the earth's- Pressure Gradients. 33 rotation, they are deflected to the right in the Xorthern Hemisphere, and to the left in the Southern ; so that, instead of flowing direct towards the centre of depression, they acquire a motion round it, l)ut inchned inwards towards the centre. Therefore the wind circulating about an area ot low pressure in the Northern Hemisphere will have a movement against that of watch hands, while in the Southern Hemis- phere the movement will be in the same direction as watch hands. Such a distribution of pressure, called by Piddington a cyclone^ is now known as a ci/e/onic depression. Also over an area where ])ressnre is high, and decreases in ail directions trom the maximum, an anticyclonic circulation of wind is developed ; for air setting outward from the centre of a high- pressure area towards the relatively low pressure surrounding it is deflected to the 'right or to the left according to the hemisphere in which the system is situated, and thus acquires a motion round the high pressure, but inclined outward from it. The circulation of the wind in an anticyclonic system in the Northern Hemisphere will therefore be in the same direction as watch hands, but in the Southern Hemisphere it will be in a contrary direction to watch hands, K to a chart on which isobaric lines have been drawn there be added symbols to represent correlative observations of wind direction and force, it will be found that the wind follows the course of the isobars to which they are related, but inclines somewhat tOAvards an area of relati\ely low pressure and somewhat away from an area of relatively high pressure. The angle at which tlie wind inclines towards the isobar showing the lower pressure is considered to be roughly al.out 30 degrees, but no rule can be given. In addition to the relation existing between the distribu- tion of pressure and wind direction, there exists a relation between pressure distriuution and wind force. Pressure Gradients. Tt lias been found that the wind force associated with a difference of barometric pressure at two places at the same time is greater as that difference is greater ; therefore, a relation will be observable between the distances separating isobars drawn on a chart and the force of the wind over the intervening area. Moreover, if one takes a chart on which the i ) > ■— ♦ • (•»lrn, 1-8, 4-6, 7 A 8, 9 iiiy. BAWOMETER, WIKD AND WEATHER *t 7«.m. 2rth MARCH, tOJO. r * I I ^*\ E.XPI.A NATION. 1UK> -»• Korct b to 10 *-— > Force 4 to 7 *■ Force 1 to .-i r.iiiu G Pr-Ki'lPlTiTloN.— Kain falling • Weathfr shown by letlors of the Beaufort Scale. ^> »li3 <5»4a/,<« «/aa» >/<« WrMm * Son Ijfo . UU»o MO Pr««, U'nAon. 8.W Sea Fog on Coast. 53 until the Htli it may be said. to have covered the whole of Europe. During this time the prevailing light air had an easterly or south-easterly direction. Mist and fog were experienced generally. The weather conditions shown in Fig. 1, Plate I., relating to the 5th, are characteristic of those prevailing during the first six days of the month. The barometer was highest, 1088 mb. (30'5 inches), and upwards, in the central area of an anti- cyclone covering- the greater portion of Western and Central Europe. It w^as lowest, 1009 mb. (29*8 inches), and less, over Northern Russia. The weather was foggy over nearly the whole of England ; over parts of Ireland, France, and Germany ; also off the east, west, and south coasts of England ; off the east coast of Ireland, in the Strait of Dover ; and, probably, over the North Sea, English Channel, St. George's Channel and Irish Sea. It was misty off the coasts of Scotland and the south and west coasts of Ireland. Vv^ith a chans^e of wind to the westward on the 6th and an .ncrease in its force, caused by the surge southward of a low pressure system, the fringe of which had on the previous day been noticed over Lapland, the weather became clearer on the 7th, except over the Southern Districts and the Channel, where it remained thick for some days longer. Fog associated with Low Pressure. As affording an instance of the association of land fog with uniform low pressure, the Weather Chart for 8 a.m. on the 20th November, 1887, has been selected (Fig. 2, Plate I.). A comparison of the conditions there shown with those of the oth November, 1901, rev^eals a striking similarity in the trend of the isobars, although the distribution of pressure is inverted, and the barometrical values differ by about 37 mb. : being as low as 996 mb. (29*4 inches) in the " low, " as against 1033 mb. (30*5 inches) in the " high " ; yet, on both, extensive areas of fog and mist are indicated. Sea Fog on Coast. y^n instance of the spread to the coast from seaward of fog occasioned by the passage of a cold air current over relatively warm water is shown in Fig. 8, Plate I., a reproduction of the Weather Ciiarts for 7 a.m., 27th March, 1910, which appeared in the Daily Wciither Report issued by the Meteorological OlHce on that dav. 54 Mist^ Fog, Precipitation. It will be observed that pressure at that time was high : 1023 mb. (30*02 inches) and above over a great part of North Western Europe, highest, 1026 mb. (30*3 inches) and over the southern portion of the North Sea. Light easterly breezes prevailed over the English Channel and France, but in all the more northern, and western parts of our islands the wind had shifted to the southward or south-westward, and soon afterwards the British Islands generally came under the influence of the equatorial wind. The sea off the East Coast of Britain was at least l*7a {'6^ Fahr.) cooler than the air, and had been so for some days past. There is reason, therefore, for believing that the thick weather experienced on the morning of the 27th over the southern portion of the North Sea, and at Fano, Cax- haven, the Helder, Cape Gris Nez, Dungeness, Dover, Clacton-on-Sea, Yarmouth, Spurn Head, and North Shields, was caused by a change in the direction of the air current : the flow of warm, humid air over a cold sea surface. Sea Fog. As affording a more definite illustration of the association of fog with the passage of a warm air current over relatively cold water the weather charts of the North Atlantic for noon of the 1st August, 1882, are shown in Plate 11a and h. They are copies of charts which are included in a set of Synchronous Weather Charts of the North Atlantic cover- ing a period of l^ months, and were prepared in the Meteorological OflSce and published by the authority of the Meteorological Council in 1886. By referring to the chart of Barometer, Wind, and Weather, (Fig. 1) it will be seen that an extensive anticyclone covers the North Eastern Atlantic, the highest barometric pressure within the system being situated immediately to the west of the Bay of l^iscay. The air over the North Eastern Atlantic will be seen to circulate about the baric maximum ; and to the west and north of this area of highest pressure the wind comes from an equatorial quarter, and is therefore warm and humid. The Chart of Air and Sna Temperature (Fig. 2) shows, by the trend of the respec- tive isotherms of air and sea, that the surface temperature of the latter is considerably below the temperature of the former, the difference being as much as 5° over a large portion of the air referred to. For instance : the air iso- therm of 65° Fahr., between the 45th and 20th meridians, ruiiB, for the most part, side by side with the sea isotherm. To fact p. S4, BAROMETER, WIND AND WEATHER. PLATE II. Ca'm,1-3,4-e,TA8.9AM,tlAtS,5fiKUt«. , ^^'/'TT/t^ ,\Jiiat:^.F'JS^Sh,rtMrt/'/// KainWM Fl6. I. AIR AND SEA TEMPERATURE AND WEATHER. Wr, It 8.. M.O. Preii, S.W, Sea Fog. SS- of 60° Fahr. and the air isotherm of 60° Fahr., is found by the side of the sea isotherm of 55° Fahr. In fact, wherever the warm equatorial wind is shown to be passing over water of a lower temperature than that of this wind, there fog or mist for the most part is indicated on these charts. The latest information relating to the formation of fog and mist, however, was contributed by Major G. I. Taylor, formerly Schuster Reader in Meteorology, now Professor of Meteo- rology in the Royal Flying Corps, in a lecture delivered at a meeting of the Royal Meteorological Society ; a concise summary of which has been published in Symons' Meteoro- logical Magazine, for April, 1917, and is as follows : " Fogs are due either to precipitation of water in the air or to a condition of the atmosphere which prevents smoke from being dispersed from the air close over the roofs of a town. The two necessary conditions for the formation of a smoke fog are that the wind velocity must be very small, and the air near the ground must be relatively cold compared with the air higher up for a period sufficiently long to collect enougli smoke to form a fog. " The formation of fog at sea can usually be traced to the cooling of the surface air when it flows from a place where the sea is warm to a place \^^here it is cold, but sometimes a fog is caused by air flowing from a cold to a warm* part of the sea. In the former case the fogs are usually low-lying and thick, while in the latter they are more frequently light fogs which stretch up to a considerable height. " Fogs consisting of small drops of water are formed on land, too, by the cooling of surface air, but in this case the air usually stays still while the lowering of the temperature of the ground by radiation to the sky at night cools the air near the surface. " Fogs of this type are not formed till the temperature has fallen considerably below the dew-point of the air during the day. This is beciuise the tormation of dew drif3s the air near the aground. Theoretical considerations show that the amount by which the temperature must foil below the dew- point before fog is produced de])ends on a complicated series of causes, but an empirical method has been devised for estimating whether, on any given night, there is enough water vapour in the air to form a fog if other conditions are suitable. This method can be used for local forecasting." 5G • Mist, Fog, Precipitation. Fog and Mist round British Isles. The average distribution and frequency of mist and fog over the seas and channels surrounding the British Islands is shown graphically on Plates III. to VI. Mist is represented by open shading ; fog by closer cross shading. In places tbe shading which defines mist is masked by that defining fog, but as the areas of mist are enclosed by a dotted line and the areas of fog by a plain line this does not signify. The results thus illustrated are based on four-hourly observations of weather recorded on board ships of the Royal Navy and the Mercantile Marine during the 30 years ended in December, 190(S. The total number of observations utilised for obtaining these results is no less than '55,126 for all months ; and the number of observations per month ranges from 3,560 for February to 5,577 for July. Nevertheless, there are in some months localities for which the data available for the purpose are insufficient, and a few other localities for which there are no data. Where such occur the words insufficient data or no data, as the case may be, are noted in the unshaded areas. The shading which denotes fog refers to a frequency of 10 or under 10 per cent, of all observations. In no localities to which these charts relate does the frequency of fog exceed 10 per cent, throughout the 12 months. Relation oj Thick Weather to Wind Direction. With respect to the correlation of mist and fog frequency, which may be briefly referred to as thick weather frequency, with wind direction, from a careful examination of a large number of simultaneous observations of wind direction and weather it has been found that, broadly stated, thick weather occurs most frequently in the North Sea with light winds and airs, which may be referred to as air movement, from between south-east and south-west. In the southern half of the North Sea thick weather occurs almost exclusively with air movement from south-westward, but off the north- east coast of Scotland it is generally experienced with air movement from southward and south-eastward. In summer, however, thick weather is often associated with calm ; and in the Strait of Dover and its neighbourhood, in the months of October and November, it may be expected to occur with equal frequency with light breezes from almost any direction when the relation between air and sea surface temperatures is favourable to its formation. Scale of Fog Intensity. 57 Off the south and west coasts of Great Britain and off the coasts of Ireland experience has proved that the air movement most frequently associated with thick weather is from an equatorial quarter. Over the English and Bristol Channels, and the coastal waters of the United Kingdom bordering the North Atlantic, where there is any air move- ment in thick weather, it is generally from a southerly or south-westerly point ; but in St. George's Channel and in the Irish Sea its direction is more frequently south-easterly. In order to assist an observer in estimating and recording grades of atmospheric obscurity due to mist or fog the accompanying scale and specification of fog intensity has been adopted by the Meteorological Office, the Ad- miralty, and the Trinity House. Scale of Fog Intensity. Scale. Name. On Sea. On River. f 1 f 5 No fog or mist Light fog or mist Moderate fog Thick fog Horizon clear. Horizon invisible, but lights and landmarks generally visible at working distances. Lights, passing vessels, and landmarksgenerall}' indistinct under a mile. Fog signals are sounded. Ships' lights and vessels invisible at \ mile or less. Objects indistinct but navigation unimpeded. Navigation im- peded , addi- tional caution required. Navigation sus- pended. Rain, Hail, Snow. Rain is caused by a diminution in the temperature of the air below the point of saturation. When the minute spherical drops of water, of which cloud is com]X)sed, become larger and heavier by the condensation of aqueous vapour and unite they fall as rain. The temperature of the air may be lowered in various ways, but the cooling is brought about chiefly, either by its expansion in ascending or by movements of the atmosphere which brir g relatively warm air into mixture with air of a lower temperature. For instance, cold, dry, and therefore heavy air currents, by forcing warm, moist air into higher ■dH Mist, Fog, Precij)itati07i. regions of the atmosphere lower the temperature of the rising air dynamically and occasion precipitation. The rate at which dry air loses its temperature in ascending is 1° Fahr. for every 180 feet, but if the air be not dry this rate is Solway Fi irth to Aberdovey. To face page 64, Platk VIa. Solar Halos observed at Abekdekx, Reproduced from sketches hi/ G. A. Clarke, Aherdeeti Ohserratorij. Solar Halo of 22" radius. May 27, 1"J12. Complete circular balo, with arc of contact. Semi-major axis of the ellipse of which the arc of contact forms a part was about 29°. 11^79 ^.>l:ii ll;ilM.,t .'_' iM.liii>. M:ircli .'», I'.iO."^, wuh arc of .oiii ar i . in..1 (^I .Si— 1 1—1 1—1 r-lrH.— Ii— 1 M << U lO ^^ GO t>. >c -^ iri O O O >— 1 j D fe +^ £ 55 t- -X> lO O O <:0 ^S> 'JDtr-'-^t^ O" K o a- =M S'c S «■ a;i«ocoo.t.— lOi t^ o-nom '-0 iO >o o >o lO o ^s 'Xi ti:: ^o 03 -U) bo fl r S ^ XI 'i 'i '-b w cr^ Oi di t^ <|-a5 O' Lj C3 CD X lO -^ CO lO 't! r^ "-H rH '-H -o I— 1 ^^|£§ • •-0 lO '* -nP -^ lO '-3 "C lO a O O" » fC =M CV .— 1 ^ ^- r: -H vD CO c ',:: t^ to nn • .... 03 2 O 5 53 00 -x; Ki ic CO to to t^ (-- p: g cci .rH a * '^J (M CO to r-- (M to 10 iiTt T+1 ■!* ^ ITS -* -+ in lO -* Q o & ^-l -z; =<-i O • S ° 00 S CO to c\i X t^ -+ M i bo fl r 03 2 o'5 • CO T'l CO r— 1 r— 1 Cp CO (7>I -— 1 C^. G -^ t^ lb 'jji to M CO Ci cr. t^ > a p^ .1^ 1—1 T—l .—1 '3 «1 CO cr.^ CO r- c^ i^ o co i t^ lO lO -n to to to to -< o cr -, 'M H i^.«a t— .— 1 f X lO t-- iri (M r-H ic r- 03 ^ o >. 3 to -^ 'Th -^ lO uO to lO lO — 1 c^t CO to t^ 'X) c.. .— i —1 1— 1 :::::: '^ : 'ts ::: : d • c • • • • C3 C3 ^ ^ - ^ II . • 32 ^ • OS d 5 r^iSiS^iHiS '^ ^ ^ ^ '^ g § c id d g C^ d^ £ S fl cccoWWHgoH H H^i^H Rainfall and Snow Frequencri. 67 o Q o ::j5 i ^ O^Z 02 .2 -3 '^- ;- nj _: S « s ^ i^-^ -S ^: « _ . o ■■^ S i ^ 02 g:- ^ g g -H u CO 1 02 1 ^ r;! o i^ 02 ^ .-H o ^ a O o "i" Oi T' "— 1 oj 1:5 ^ ^ ^.Si^r? 2-5 S cS o ^% J^ ^ ? .2 ^ ;:: -^ ;5j 3 O > ^oo oowoo o o o c > ooo ooooo o o o -a — 1- 3 COt-3 -^ a S^ 1 o H cCCOfM (M-^OrH.— i -H .— i ^ 1 ^•^ ,_, — ^^ "S 'C — ■ a o. ^'lT- 1 i-H I— i (7i ^H w fl ..^ c3 Its rH . . , ...... ~, bo % I '-^ '■ '-'^-^^ B ^^ 'i; .1 .^ a J ilC^ COu-iQ; re fT<~-J3r- 2^ Tr. i? S g X) « o ; o ,, c p, -■ W ^ ^ ^ ^ O ;• <; ^ « CD •' .- ^ ^ ^-^ CO ~r, ^ •' r |B> £ ^ ^^ If-H ^«i fl^ ^;^Sg-s 5 ^ ^ e:25 5zi OQ^qa-^l a. S > .^ .— I (M CO "O ^c t^ OO cr- o 1— I : : : : TS x; : : • • • a c . c5 . rt ►^ ^ Cfl [^ 7- ''Z : -^t; '^'^'^ -^ A "^ S - J : csccdct^ v^'^ '^ cc 5 ^«^ S^ecS rtr:_ri=Pc2 fl 1 _<^_i^^ ^^O T^C — ^ 1 '/ J 03 w HHco P^ ■/: 1-3 H^ 1 1197S) 68 Pressure Distribution and Weather Conditions. CHAPTER VI. Atmospheric Pressure Distribution and Weather Conditions. The relation between the distribution of atmospheric pressure and wind has been referred to in Chapter III, There exists also a relation between the general distribution of pressure over a given area, as exemplified by the con- figuration of isobars representing it on a synoptic weatner chart and the general conditions which characterise the weather prevailing simultaneously over that area. Preparation of Weather Charts. The methods adopted at the Meteorological Oihce in the preparation of the Daily Weather Charts are as follow : — A blank outline map of Europe is taken, on which the positions of the stations from which meteorological reports are received are represented by dots. The telegrams from these reporting stations arrive at the office in varying order, but as soon as each message is received the observations are plotted in correct position upon the chart ; the details charted comprising : — (a) the direction and force of the wind, shown by small arrows ; (6) the reading of the barometer, corrected ; {(•■) the reading of the dry-bulb thermometer ; and {d) the state of the weather, given in letters of the Beaufort notation as follow : — Letters to indicate the State of the Weather. b Blue sky. q Squall3\ C Clouds (detached). r Rain. d Drizzling rain. S Snow. e Wet without rain. t Thunder. f Foggy. U Ugly (threatening appearance g" Gloomy. of weather). h Hail. V Visibility. Objects at a dis- 1 Lightning. tance unusually visible. m Mihty. W Dew. O Overcast. Z Haze. p Passing showers. It is well to bear in mind that w = dew, but d= drizzle, and e == wet without rain ; p = passing showers of rain, and q = squalls, but s == snow. In the case of a morning chart, the changes which have occurred at each station in the readings of the barometei" and thermometer in the past three hours are then indicated in red and blue figures ; red representing a rise, Weather Charts. 69 and blue ;i fall. When most of the telegraphic reports have been charted, isobaric lines are drawn connecting places or ])ositions in which the barometer stands at the same height. These lines are drawn for each 5 millibars ('15 inch) ; thus, 1005 mb., lOlU mb., 1015 mb., and so on ; but as the barometer readinofs at the various stations are oriven to tenths of a millibar, it is usually necessary to run the lines along positions lying between two stations. Thus, if the reading at one station be o-iven as 1004 mb., and the readinu" at the nearest adjacent station as lOOG mb. the isobaric line repre- senting a barometric pressure of lOOo mb. will be drawn midway between the two stations. But if the reading at one station be given as 1 002*7 mb. and the reading at the next adjacent station as lOOG'-t mb. the line represent- ing 1005 mb. will, of course, be drawn much nearer the latter than the former station. When completed the isobars show^ at a glance the regions in which the barometer is high and those in which it is low ; and by the distance separating the lines it can be gathered whether the differences in pressure over any portion of the regions represented on the chart are oTcat, moderate, or slio-ht. Great differences in pressure, which are technically known as steep //radientSj are recognised by the isobars lying closely together, and these are almost always associated with gales or strong winds. Slight differences in pressure, or slii/ht gradients as they are termed, are, on the other hand, recognisable by the isobars lying widely apart, and these are always associated with winds of little strenofth. By comparing the weather chart for any particular hour, say, 7 a.m. to-day, with those for previous times, say 7 a.m. and 6 p.m. of yesterday, it is possible to follow the changes that have taken place in the interim, more particularly as regards the relative positions of the areas of high and low pressure ; and thus to form an estimate in regard to ])robable changes in the near future. A series of synoptic charts affords, in fact, the bases upon which the official forecasts of the weather are drawn up. To revert to the relation of pressure to weather ; the di^tl'ibution of the former is undergoing changes continually, but the general character of its disposition may be maintained for a considerable time, although the whole layer of air near the earth's surface and for some distiince above may be in motion. There are certain well - delined shapes which isobars assume on a synopti'.- weather chart that are found to be closely allied to certain definite conditions of weather. The shapes representing dispositions of atmospheric pressure liy79 D2 70 Pressure Distribution and Weather Conditions. which appear to exercise the most dominating effect on weather conditions are those which are circular or approxi- mately so ; pressure either increasing or decreasing towards the central area of highest or lowest pressure about which the air circulates. The characteristic distributions of pressure in which the circulation of the air is round a central area of low barometric ]»ressure termed cyclonic^ and that in which the circulation is round a central area of high barometric pressure termed aittici/clonic, have been discussed in a former chapter. Other definite shapes which occur on synoptic weather charts, and are allied to specific conditions of weather, are for the most part modifications of these types, and will be referred to after typical instances of the cyclone and anticyclone have been introduced. The Cyclone. Fig. 3, Plate YII, is mainly a reproduction of the synoptic chart given in the Daily Weather Report issued l)y the Meteorological Office for 7 a.m. of the 25th March, 1909, but the data shown on the original chart have been supple- mented by observations received, for the most part, subse- quent to its publication, by wireless telegraphy from Transatlantic liners, so that the area under observation has been extended westward by these means. In this chart the centre of a cyclonic depression, which passed directly over the British Islands, is indicated near the coast of Yorkshire by the closed isobar of 979 mb. (28*9 inches). The isobars of 982, 985, 989, 992 and 996 mb. successively (29*0, "1, "2, "o and '4 inches) are shown encompassing this central area of lowest barometer reading or baric minimum as it is called. The wind, represented by arrows, is seen to follow the course of the isobars, but in- clines in\N ards towards the centre of the system showing complete circulation about the system. It is easterly to north-easterly, fresh or strong, over the south of Xorway, and over Scotland; north-westerly and westerly, fresh to light, over Ireland, England, and the North of France ; south-westerly and southerly over Germany, Holland, and the Xorth Sea. This cyclonic depression was traced from mid x\tlantic b}" means of reports transmitted by wireless telegraphy. On the 2;^rd March, at 7 a.m., the centre of the system was situated in about 50° N. latitude ; 35° AV. longitude (Fig. 1, Plate VII). Its path from 7 a.m. of the day following (Fig. 2, Plate VII), when it began to infiuence the weather over our islands, to 7 a.m. on the 26th, wlien it was centred To face p. 10. PLATE VII BAROMETER AND WIND 7 am :""": r-n FlB 1 LOW looa — __, 1006- ' (009'' ' ^, 1013 -'^ id* '^ lO'S ^ t 'lao o 3(y6 \M I G »^ -f y — -^ / ' ^^X ■ ,<^ ., O^ Q. ',<.-'' G H ;y '^ «3bo march 1 9f)9 L^ G M 1 90S laoliars drawn to every tenth >>f an im-h : with e«. Wj'inaii >« .Simn, littl., (•itho. U.*) t Wjmaii t .S.>ni., l.t.!.. r.ivho. It" Pre«». I/mdou, S W To face p.lO. BAROMETER AND WIND 7 am. K"^'^^^'^^^ LOW PLATE VII i. 1 ■Awireless M^ssngan Fig. 1. I -v \ ^^ ^ -^^'''^aSTH MARCH. 1909 Ixoliars , S-10 >— »■ , t Jiluis ® Knin •. i??^ '.s5«c/-.* ooof yi4 \\.\miin A Soii^, l.l.l.. Lull". M. O. Pr^sf. I.'.ndoii, S W ~ To face p. II. PLATE- X. BAROMETER, WIND AND SEA 8 a.m., 5th MARCH, 1905. .;^^f|3c-^*H 1 G H ^ / f^o-c^P ( C \ SSW4: I so S 30 2 36 / 3(1 IS g ^ Barometer.— Isobars .ipl' drawn for eaih tfntli of an inch ; with equivalents in millibars. WiNii.— Arrows fly'Rg witii the winri- show Uirant ions and ¥aree. Force above 10 »— Force 8 to 10 > — Force ^ to 7 Force 1 to 1 Hifthf !>??£. -TS^d/V-t /OOtA' Vr«Ai> * SoM liTB.. Litiw.lLO. Smbb. Lon->! ! .F«>^{! : /x ■*-;>-> // . — ^yig.-np-i-.-'--.:.-v^->.z:i:::2:! ^SXPIiATfATKW.— RATtOKKTiCK-. — Isobars-sre ^ranrn for eacli tenth nf an im-li. Wind.— Arn.ws nyiug with the wind show lJire<;tnjn and Korce, thus: — Force 4 t^> 7.—^ Force 1 to 3. — ^ Calm. ® Kaiii lalliiiK.* r Fig; 27 - ->--- — iBSPLATTATIWf. — TKirpmrATirRB.-=^-I»n*herms ar»- drawn for each 10' W e at H BR.— Shown in places by letters of the beaufort scale. Rainf.ai,I-.— A dotted liue separates areas in which raiu has fallen from those without rain. 6 P.M., 29th DECEMBER, 1908. 7 am', 29th DECEMBER, 1908 L 0^4- ^I'l 4N\TT«)N -BAROMKTKK. -Isobars are drawn for tenths of an inch: with equivalent* in millihars Mm, Irrow. ny n" w.th the wind show direction and Korctf. thus .-force above W^* , fcorc, s i,,i(V"- -► Force 4 to 7--->, Force I to 8 -, Calm® . Rain falling •, «"«'* f*"'"!?- * SEA liisTUKHANtK.— Kough^^^ . Uigh;;^:*^ . Wyman A Sons, Ud., I.ithn. M.p. Frew. L-ndou, S W ' The Anticyclone. 78 closed isobar of 1019 mb. (30"1 Ids.) there is complete circula- tion of liMit breezes which are from southward and south- westward to the west and north-west of our islands, from the westward and north-westward over the north of Scotland and northern portion of the North Sea, from north-westward and northward over the southern portion of the North Sea, and from north-eastward over the English Channel and over France. Thus the wind circulation was in the same direction as that of clock hands. Over Eno'land the temperature (Fig. 2, Plate IX) had fallen since the previous day, and was belo^v the normal ; the weather was fine over North-western Europe generally, excepting at Christiansund and in the south-west of Iceland, where rain was fallino;, and on the East Coast of Eno-land where fog was reported, but the atmosphere was rather misty in some localities. Under the influence of a depression spreading from the south-west over Iceland, the barometer was falling in our islands but was rising" over France and the S])anish Peninsula. Next day the anticyclone had moved south-westward over the ocean towards the Azores. Secondary Wind Systems. Accompanying the cyclonic depressions there frequently occurs an interruption to the regular sequence of changes in press are and wind by the formation within the area of disturbance of a subsidiary depression called a secondary, and sometimes termed a satellite depression, because it appears to be thrown off the larger cyclone ov primary. Over these ishinds and their neighbourhood, when a secondary accom- panies a cyclonic depression, it is almost always located to the south of the latter, and is commonly found in the rear ot the lowest barometer reading ; but not infrequently it develops in a ])osition, which, shown on a synoj^tic chart, lies in line with the trough of the primar}'. Occasionally, however, the secondar}^ forms in front of this trough ard moves, usually in an easterly or north-easterly direction, faster than the primary depression. The winds associated with a secondary during its passage over a locality are generally, but not always, moderate to light, and are accompanied V)y rain or snow. Precipitation commences on the near approach of the secondary, when the barometer is falling (piickly ; and ceases or becomes inter- mitteiit after the lowest barometer reading has passed, and the system is retreating, while j)ressure rapidly recovers ; but the amount of j)re(;ij)itation is not large as a rule. llHTit D 4 74 Pressure Distribution and Wectther Conditions. Ill illuHtration of a secondary depression Fig. 4, Plate IX, iti here reproduced from the Supplementary Chart of Baro- meter and Wind, relating to 6 p.m. on the 29th December, 190(S, which appeared in the Daily ^Yeather Report of the following day. The Daily Weather Chart for 7 a.m. on the 29th showed a distribution of pressure, immediately to the west of Great Britain, that is known as a " V^'' -shaped depression, and is so called from the contour of the isobars illustrating it, which run to a point, in the form of a " V " (Fig. 3, Plate IX). The trough of this depression, lying nearly north and south, was situated over, and to the north and south of Ireland. Strong winds from between south and south-east were reported from nearly all parts of England, and gales from some parts of Scotland. At Valencia, Ireland, the wind had veered to north-west and blew freshly. The wind was squally over the United Kingdom, and snow was falling over a considerable portion of England and Scotland, and rain in Ireland. At 6 p.m. of that day (Fig. 4, Plate IX) the centre of a secondary, thrown off the primary cyclonic depression, central to the north-west of Iceland, is indicated over a part of the English Channel and the north of France by the closed isobar of 999 mb. (29*5 inches). The wind, wdiich, as shown, follows the course of the isobars, was gentle to strong round the baric minimum ; from northward and north-westward, over the mouth of the English Channel, and west and south of France ; from southward over the east of France ; and from eastward over the south-east of England. Snow was falling at Liverpool, Portland Bill, La Heve, Paris, Bath, London, Dungeness, Dover, Cape Gris Nez, Clacton-on-Sea, Yarmouth, Spurn Head, and Aberdeen ; rain at Perpignan and Biarritz. Next day this secondary appeared to have parted from the primary depression, and travelled to the south-eastward, while the latter had moved to a position north-west of Scotland, and the barometer at localities between the two systems was rising. This secondary depression was exceptionally well developed ; but frequently a secondary disturbance of the atmosphere, when depicted on a synoptic chart, ajDpears as a loop or kink in an isobar ; nevertheless the conflicting air currents, represented by this irregularity in the course of the isobar, usually give rise to precipitation. Sometimes, on the other hand, the secondary dej^ression assumes large proportions, and occasionally, though less frequently, it ii^ " V^ -shaped Depressions. 7o difficult on this account to determine which of the two wind systems is the primary. " V ^^ -shaped Depressions. The " V "-shaped depression belongs to the type of secondary depression : and is formed by a prolongation of the southern segment of a cyclonic system, and develops, not infrequently, in the lane of low barometrical pressure, separating two adjacent areas of high pressure, termed by meteorologists a col. The extension of the cyclonic system southward, in the manner described, appears to be attributable to abnormal activity of the polar air current in rear of the trouyh of the system. '• V "-shaped depressions, as a rule, move eastward with tiieir primary, and occasionally develop into separate dis- turbances, Plate" X. which is reproduced, irom the Chart of Piarometer, Wind, and Sea, in the Daily Weather Report, issued by the Meteorological Office, for 8 a.m. on the oth March, 1905, affords another illustration of a " \' "- shaped depression, which had advanced from the Atlantic over our islands since the previous day. It extended from beyond the north of Scotland to the extreme south-west of Eno-land, but in the west of Ireland the barometer had begun to rise briskly. The barometer readings ranged from lOi'G mb. (30'3 inches) and upwards over the Spanish Penin- sula, and above 1023 mb. (30*2 inches) over Germany to 1010 mb. (29-82 inches) at Scilly, and to about 99(rmb. (29*4 inches) to the northward of the Hebrides. In front of the trough of the depression or eastward of a line drawn between the Minch and Scill}', the barometer was falling generally ; temperature, under the influence of the wind from warmer regions, southerly to south-westerly, had risen ; and at Oxford, Bath, Portland Bill, Scilly, Pembroke, and Holyhead, rain was falling ; while at other stations the sky was overcast or cloudy. In rear of the trough the wind had veered to north-westward ; the baro- meter was rising ; temperature, under the influence of winds from colder regions, mainly north-westerly, had fallen ; and the weather had become squally and shower}'. Gusts and Squalls. A gust is a marked increase in the force of the wind, sudden and transient ; a squall is a gust of greater intensity and of longer duration. Either may occur whatever be the 76 Prei, and 6. These charts with the whole of the information in reference to the squall which follows are summarised with the author's permission from his published report. The squall, hrst noticed at Stornoway at l!^.30 a.m. on the 8th, travelled to the south-eastward at an average speed of 3S miles per hour ; and at, at least, two stations the velocity of the wind recorded by anemometer was 80 miles per hour. The last station to feel its effects was Hastings, over which place it passed at alxnit 4 p.m. * "Quarterlv Journal of the Royal Meteoi'(»loi,'ical Society-," "^'«>l, XXXI 1. No. 140. October, lUOti. 7.S Pressure Distribution and Weather Conditions. At 6 p.m. on the 7th the barometer commenced to fall rapidly, the pressure gradient soon afterwards becoming steep. The positions of the line of squall are marked on all of the charts by a dotted line, on those for 2 p.m. and 3 p.m. 'tke p<»»tit>»jfcof41je line of aquaJl -Trtje force otthe vrmA has, be*" Indic^tpd Uiiia Forces I Sth- 19oe Bt earn ' The dotted hne« *>iow Fig. 4. by the lower of the two dotted lines, the upper marking the passage of a secondary squall which followed. In front of the line of squall the pressure distribution is for west-south- westerly wands, in rear, it is for north-westerly winds ; the line thus separating the two air currents. Line Squall. 79 The velocity of the wind in the squalls varied consider- ably: at Bidston and Shoeburyness it reached SO miles per hour, about double the velocity at which the line of squall travelled. In the east of England and Scotland the velocity of the Fi G. O. north-west wind, in rear ot the line, was not so i^reat as that of the wind in front of it ; in the south-eastern districts the wind drop])ed to a light breeze after the squall was over, but in tiie west it was strong both in front and in rear of the line, and at Holyhead the wind blew strongest after the passage of the line. At Oxshott, in Surrey, no increase of wind was 80 Pressure Distribution and Weather Conditions. shown by the pressure-tube anemometer : the current which was brisk from south-westward in front of the line fell light after veering suddenly to north-west ; and at Kew the velocity of the wind in the squall appears to have been Fig. 6. about the same as the rate of translation of the line of squall. The barograms lent by observers for the Meteorological Office and others, for use in connexion with the enquiry, all showed a sudden cessation in the diminution of pressure, and, at the English stations, east of the 3rd meridian of The Col. 81 west longitude a sudden increase. At the Scottish stations- the sudden increase of pressure was absent, but the crisis was none the less well marked by the steep angle shown on the barogram caused by the sudden cessation in the rapid diminution of pressure. The time of passage of the line of squall over the respective stations from which thermograms were received was well marked on the traces by the contrast shown between the steady conditions of temperature, associated with the warm south-westerly current of air, and the unsteady conditions obtaining in the cold north-westerly current. . The rainfall in connexion with the squall varied between O'l inch and 0*2 inch, except at stations in the south of England where it exceeded 0*2 inch. The secondary squalls experienced were, in some localities, of considerable intensity. It is a feature worthy of note in connexion with similar dis- turbances that the primary squall is frequently followed by secondary squalls, more or less intense. The air became considerably drier afto" the squall than it was before it ; despite the heavy rain and hail, and the diminution of temperature which accompanied and followed the passage of the line of squall. The Col The weather is quiet but unsettled within a c*o/, or neck, of relatively low pressure, which occasionally forms between two areas of high pressure ; and the winds generally are light and variable. In the warmer months of the year the temperature within a eol is usually above the normal, the weather dull* and, on the coast, mist or fog is common in the autumn. The distribution of pressure is favourable to the occurrence of thunderstorms. In the colder months the temperature within a eol is belo-v the normal, the atmosphere often misty to foggy inland, but, as a rule, clearer near the sea, and frequently the thick weather does not extend to the coast. To instance the formation of a 6'(>/ in early autumn, the Chart of Barometer and Wind at 7 a.m., 12th September, 1909, is reproduced from the Daily Weather Report issued on that day (Fig. 1, Plate XL). At this tiuie the barometer was high over the north-west of Ireland, and over the adjacent portions of the North Atlantic ; also in a band extending from the south of Sweden to the White Sea. It was below 10U9 mb. (29*8 ins.) in a dej)ression over Iceland, and in another depression over France and Spain ; a neck of relatively low pressure or eol 82 Anticipation of Weather. connecting these two low-pressure systems. Calms and light airs or light breezes, from between north-east and north-west, were reported at the majority of stations in the Tnited Kingdom, gentle to moderate breezes at a few of them. The sky, for the most part, was cloudy or overcast, there avms a good deal of mist on the coasts and fog at Spurn Head. The weather was showery in the Hebrides, rain ATas falling at Alalin Head, in Ireland, and was experienced later in most parts of our islands, with thunder at some southern stations ; while severe thunderstorms occurred in different parts of Fi-ance. During the day air temperature rose to 294'7a (71° Fahr.) at Brighton, and 294-la (70° Fahr.) at Eastbourne, Weymouth, and Jersey ; but, while above the normal generally, it remained below 288* 6a (60° Fahr.) in many parts of the United Kingdom. In illustration of a winter col^ the general conditions prevailing over our islands and North Western Europe at 8 a.m. 'on 18th February, 1902, are shown in Fig. 2, Plate XI, which is from the Daily Weather Report of that date. The barometer was then highest, 1029 mb. (30'4 ins.) and upwards, in an anticyclone which had formed over Scandinavia ; another area of relatively high pressure, in which the barometer readings were 1019 mb. (30*1 ins.) and upwards, had appeared over the Bay of Biscay and south- west of France. Two areas of low pressure, one in which 1016 mb. (30*0 ins.) was the highest barometer reading, situated over south Germany, and another in which the readings were 1013 mb. (29*9 ins.) and less, were connected by a neck of relatively low pressure. Thus Great Britain was under the influence of a col, in which temperature was below the normal. The sky was clear in some parts of Scotland, but cloudy or dull generally, with fog, over England and over some parts of the Continent ; but clearer or clear weather prevailed on the coasts. CHAPTER VII. Anticipation of Weather by Observations at a Single Station. In the previous chapter the method of ])reparing the Daily Weather Charts at the Meteorological Office was described; and it was explained how by comparing a chart for any hour with others for previous hours, noting at the same time the To face p. 82 PLATE XI BABOMETEFf, WIND AND WEATHER I'2th SEPTEMBER, 1909. 7am BAROMETER. WIND AHO WEATHER 18th FEBRUARY, 1902, 8 » m. Fig. 1. Pig. 2. KXl'LAHAnoN. BaRiimktkk.— IsdiArs »rc drawn for .iwh tentli of an inch ; with equivalpntB in inillibarR. Win 1,.— Arrows (lyins^ with the win<1 uhow Klrection and Knrce, thus :— I'orte 4 to 7 -*■ Force 1 to S ^ <';ilni Wkathbr ii indicai.cl by lett- rs ot tfw Bejiiifort icaJe ; Ko^ an'l Mist excepted. »» waa <(»»o/^; /uyofl ///t> "WrtL^t k S«in Lrs LIUic 4 Piwis. ijondon 8 W 7 Approach! IK I Depressions. 80 changes that had taken place in the interval, an estimate coald be formed regarding probable changes that wouhl occar in the weather conditions of the near future. The method described calls for, and depends upon, accurate synoptic observations, instrumental and otherwise, trans- mitted by telegraph from a large number of selected statiojis, located in suitable positions. In this chapter it will be shown how by noting from time to time at regular intervals observations of barometer, wind, and clouds, supplemented, if possible, by readings of the wet and dry-bulb thermometers, the trained observer may anticipate, with some degree of accuracy, the more im])ortant changes in wind and weather that are likely to follow. Approaching Depressions. The most marked changes that occur are usually due to the approach of cyclonic depressions or tlieir opposites, anticyclones, and the passage of these or of secondary disturbances in connexion with the former across the country. First hidications. As a rule the approach of a depression is first indicated by the appearance of cirrus clouds ; and, as explained in the chapter on clouds, the observer's position in relation to the path on which the depression is travelling can be roughly determined by watching the motion of the upper clouds, and noting the changes in the direction of the wind at the surface. As the depression draws nearer the observer's locality the wind veers or backs, as the case may be, and soon afterwards the Ixirometer conunences to fall. Bearing of Centre. For the purpose of forecasting weather it is necessary, as soon as the barometer shows a decided fall, to ascertain the direction in which the central area of low barometric pressure of the depression is situated, and the ])ath on which it is travelling. Subsequently it can be determined whether the centre is nearing or receding from the locality. Standing facing the wind the bearing of the central area of a cyclonic de{)ression in our hemisphere will be al)out two points in rear of the observer's right arm outstretched at right angles to the wind's direction. For instance, su])posing an observer, who lias reason to believe that a cyclonic de})ression is approaching his locality, stands facing the wind 84 Anticipation of Weather. from south-south-east at A, Fig. 7, his right arm, out- stretched at right angles to the wind's direction, will point west-south-west, and the bearing of the central area of low pressure will be about west. Fig. Again : suppose the wind he faces to be from east-south- east at B, his outstretched right arm will point south- south- west, and the bearing of the central area of low pressure will be about south-west. Path of Centre. Having determined the bearing of the central area of lowest pressure, changes in the wind's direction must be carefully observed and recorded. If at A the wind veers — that is to say, shifts with watch hands to southward and south- westward — the centre must be taking a path to the north of the observer, the path, for instance, G C D ; and, as the depression moves onward and passes away in a north-easterly direction, the wind will veer to south-west and west. The barometer will then commence to rise and the weather to improve. If, on the other hand, the wind backs — that is to say, shifts againr:t watch hands to south-eastward and eastward — the centre must be taking a path to the south of the observer, the path, for instance, F C E ; and as the depression moves onward, and passes away in a south- south-easterly direction, the wind will continue to back to east, east-north-east, and north-east. Soon after the wind has backed to north-eastward the barometer may commence to rise, but the improvement in the weather will be gradujil. Should the wind remain steady at Weather Sequence. mktrr.— Uuliura are drawn for nacli two tenths of an inch ; wiiti eituivaleiit* In tnilliban. \VIM>. -Arrowi Ilyinn *lth the wirjcl show I)lr»clion and lorce. thui : — Korc.' aiM.\e lu*— ♦■ , Ijorce 8 lu l<)-» — > , iuixi 4 i.. 7 - + , kmce I li> :•! - •■ r»irii '■ *;'.'•§. ^^^(j./** ■.■»ion was followed Ijy a ridge of relatively high pressure, which moved to the east- ward over the Hritisli islands during the 18th and 19th. and was associated, while the wind remained northerlv, with colder and drier weather. The copies of barograms and thermograms on Plate XlII show the fluctuations in atmos[)heric pressure and air tenr- perature at Aberdeen and Crathes respectively, on the north sirle of the cyclone's path ; and at the Meteorological Office 88 Aiiticijjation of Weather. and Warlingham respectively, on its south side, from midniglit on the loth to noon on the 19th December, 1910, during the approach of the depression, the passage of its centre between those stations, and its retreat eastward which has been described in the foregoing. Symbols denoting the corresponding changes in wind direction and force, and in weather have been added to the barograms. J-*ine'tVood Crathes is situated 12 miles to the south- westward of Aberdeen ; Warlingham 13 miles soLith-south- east of the Meteorological Office. The stronofest winds at Aberdeen, also at the Meteoro- logical Office, did not exceed seven of the Beaufort scale, the pressure gradient having become less steep before the centre of the depression arrived over the meridians of those places. Gale force was recorded at most of the reporting stations on the coasts of Ifeland and on the west ccmst of England. The rainfall appears to have been general over our islands, but confined mostly to times when the front of the depression was passing ; in many localities the fall was heavy — more than half an inch. After the passage of the line of lowest pressure over a locality the rain either ceased or became intermittent, even on the left side of the central area usually proliric of rain, and continuing after the passage of the trough, The ridge that moved over the British Islands as the depression passed away was formed by the protrusion north- ward, between the retreating and oncoming depressions, o1 an area of high barometrical pressure or anticyclone, situated on the 18th over the Bay of Biscay, and it passed eastward over the North Sea and Scandinavia on the 20th of the month. As it moved from our islands to the ISTorth Sea the wind backed to the southward on the northern side of the path of the oncoming depression, and to south-westward on the southern side, freshening at the same time ; the temperature quickly rising, and the air becoming: humid. When our islands came directly under the influence of the new depression the barometer commenced to fall, and by 7 a.m. on the 20th a south-west gale was blowing on the west coast of Ireland. Such a sequence of pressure wind, and weather changes is characteristic of certain types of weather conditions over the British Islands and their neiiihbourhood, and mav be repeated during many days, even weeks ; one depresson, with its accompanving ridge of relatively high pressure, following another in regular succession. Oh z s I - ;aj ^.- HI o OF] 00 I To face p. 88. PLATE XIII. OECCMBER. ton. t 1 1 M 1 1 2to M DECEMBEHL WO. Syochronoui BanacruB* and TbetmogiuoM (boBUw ehuifei n Baromi-ti-Le prMsnrr uid Aii tetnpmtun at pbcn litnaUd U«* BorUi and on ibt 3o.nb nc:-t rc»petUvely of a cycloni'. dqrnMon dunnf the paiugc avti tbow sutirmt of the centnl tn* of I pirMUR To the tormn bavc b«ea •ddeo &yiiitrak deooliog chuifei u> wind force to' ' " >nri iD »«atber ■.A 3,,M.O. Pres»,8 WeatJier Sequence. 8i^ The depressions usually follow about the same path ; in this instance, however, the anticyclone over the Bay, which has been referred to, spread slowly northward, with the result that the path of the advancing- depression was deflected more towards the east. Thus a westerly type of weather conditions replaced the south-westerly ty])e which had pre\'iously obtained. xls a rule, while our islands are still under the influence of the ridg-e of relatively high pressure, the weather being fine, cirrus clouds make their appearance in what may be an otherwise cloudless or almost cloudless sky ; and the relative humidity of the air becomes abnormally high soon afterwards. The excess of moisture in the air may at once be detected by readings of the wet and dry-bulb thermometers, and should be recorded in the '* weather '" column of a climato- logical register by the letter " e," which indicates wet without rain. This wetness of the air frequently gives warning of the approach of a low pressure area before the wind has backed or the barometer has commenced to fall. The fact is of considerable importance in foretelling weather by means of observations obtainable at an observer's station oniy. Had the cyclonic depression been slowly followed by a more extensive area of high pressure, or anticyclone, instead of a ridge of high pressure, the centre of which had taken the same path as that t iken by the centre of the cyclone, the direction of the wind would have changed mere slowly, and these changes would have been contrary to the movement of watch hands on the northern side of this path, i.e.; from north-east towards north, north-west, west, and south-west as the anticyclone slowly moved across the United Kindom ; but with watch hands on the southern side of the path, i.e., from north-west towards north, north-east, east, and south-east. During the passage of the central area of highest barometer readings, however, there may have been no wind, and only gentle to light winds or light airs would be experienced in the neighbourhood of the central area. The air in an anti- cyclone is characterised by its dryness. In summer the weather is generally quiet and warm in the daytime, the sky clear, inland, but is sometimes misty in the early morning. On the coast mist or fog commonly occurs and may continue during a considerable part of the day. Temperature in Atificyclones. In the colder months of tlie year thick weather is fre- quently associated on land with anticyclonic conditions, and the atmosphere is thickest where pressure is highest and 90 AiiHcijmtion of Weather. most uniform, and where there is consequently little or no breeze to dispel the foo- or mist. It diminishes in intensity as the distance from the area of highest pressure increases, where the pressure gradient becomes steeper, and the air current stronger. The occurrence, on some occasions, of spells of cold , weather in these islands, when barometrical pressure is high, has led to the assumption that winter anticyclones are generally attended by abaormally low temperatures. This supposition has been shown by an eminent meteorologist, Mr. W. H. Dines, F.R.S., to be erroneous, so far as it may be considered to apply to the anticyclonic conditions that are germane to the British Islands. Mr. Dines found, by examining the Greenwich records, that during the 50 years 1841-1890, 74 periods of frost were shown*. Daring 20 of these periods, which between them represented 216 days of frost, the mean barometric pressure was below 1009 mb. (29*80 inches) ; and during 13 of the periods, representing 93 days of frost, barometric pressure was above 1023 mb. (80"20 inches). Mr. Dines also found that nearly every frost in the fifty years' period, of marked severity or length, occurred during the low-pressure series, and that the mean temperature on all those days on which the barometer was above 1023 mb. (30*20 inches) is close to the mean temperature of winter (276*5a, 38*3° Fahr.). He noticed, moreover, that the per- centiige of frosty days, that is to sa}", days on which the mean temperature for the day was below 273a (32° Fahr.), which was 15, is also the percentage of frosty days for the whole 150 winter months included in the period. Mr. Dines sums up by saying :— The figures at Greenwich show that severe cokl in the south-east o£ England is most likely to occur with the barometer below its average value, and that on days when the barometer is very high the temperature is close to its mean value. He considers the direction of the wind to be the dominant factor, because in Europe daring winter an east wind brings cold and a west wind warmth ; so that the northern segment of an anticyclone and the southern segment of a cyclone are warmed by west winds, while the southern segment of an anticyclone and the northern segment of a cyclone are chilled b}^ east winds. Nevertheless, the clear skies and light breezes or c;dms, which frequently accompany summer anticyclones, and * Symon's Meteorological Magfassine, March and May, IDll. Cirriform Cloud AJotion. 91 condace to excessive solar radiation, occasioning high day temperatures, are not confined to that season. When such conditions obtain during the colder months they favour the causation of excessive terrestrial radiation at night, and occasion corresponding low temperatures, the effect of which remains, to some extent, throughout the day. The foreo'oing- information relatino; to the sequence of wind and weather changes pertaining to different parts of an area over which a cyclonic depression passes, although it refers to one type of weather conditions only, may enable the reader to realise the changes that would take ])lace during the passage of disturbances associated with other characteristic tvpes of weather conditions in the various localities that come under their influence. An acquaintance, therefore, with the more marked types, which will be described in the next chapter, should aid an observer, dependent upon observations taken solely in his own locality, in anticipating from time to time changes of w^ind and weather for many hours in advance, not only in his own locality, but also in other parts of the area over which he anticipates a disturbance will travel, Cirriform Cloud Motion and Weather Changes. In order to assist him in foreseeing these changes an obserxer should be on the watch for the appearance of cirrus clouds, noting their bearing and the direction of their movement. He may thus gain early information in regard to the approach of a depression and its movements, in the mann'cr described in Chapter I\ , which subs^equent observa- tions of wind should, as a rule, confirm. It must, however, be borne in mind that depressions do not always follow the path they appear to be taking ; that instead of passing, for instance, north of the observer a depression may pass south of his locality, and tlie sequence of wind and weather will thus differ from that which he. antici])ated. His antiri])ations may j)rove incorrect from other causes als(i : the progress of a depression may be arrested ; the depression may tnivel faster or slower than it was expected to move ; it may become shallower, that is to say, baro- metrical pressure within the area of disturbance may become more evenly distributed ; and it may become so evenly distributed as to lose its energv, iu other words, it mav " fill up." After the central " Low " of a depression has j)assed over the locality of an observer, situated to the south of its path, 92 Anticipation of Weather. v/hile the barometer is rising, the wind veering and moderating, the weather at the same time improving, it frequently happens that the wind rather suddenly backs, the rise of the barometer is checked, and the mercury commences to fall. The sky then becomes overcast and soon afterwards there is a temporary increase of wind and fall of rain or snow. Such changes occur wheu the primary depression is followed by a secondary disturbance, the characteristics of which have already been described. In certain types of weather conditions depressions follow one another in somewhat quick succession, and when the normal sequence of events is not interrupted by the inter- ference of secondaries between the retreating and advancing primary low-pressure systems, the continued rise of the barometer, while the wind veers and moderates, is before long accompanied, as a rule, by a brief spell of quiet weather. Then, jijg the new depression advances, cirrus clouds appear, and when our western shores come under its influence the incursion is heralded b}^ backing wind and diminishing pressure ; while the cirrus deepens into cirro-stratus, heavy clouds gather, and soon aftervN'^ards precipitation sets in and the breeze freshens. At all seasons of the year, when a cyclonic depression passes over these islands it brings cloudiness, followed by either rain or snow in places where the wind comes from directions between south-east and north-east round by west ; and when precipitation sets in with an easterly or north- easterly wind the fall is generally prolonged. In winter the passage of a depression over our islands is accompanied by a rise in temperature, caused by the intro- duction of warmth by the equatorial wind blowing in front of its centre ; but in summer, by reason of the cloudiness and consequent obstruction of sunshine with which it is attended, it frequently produces a fall in temperature. Cyclonic depressions usually travel at a comjxu'atively rapid rate over our islands, and come, as a rule, from south- westward or westward, rarely from eastward. The force of the wind associated with any area of relatively low pressure depends, not upon the degree to which atmo- spheric pressure is reduced therein, but upon the pressure gradient or difference in pressure between various points within an area of disturban€e and its neighbourhood ; in other words, it depends, not upon the height of the barometer in any place, but upon the difference between that height and the height subsisting over a neighbouring place, the force of the wind being greater as that difference is greater. Cirriform Cloud ^fotion. 93 Cyclonic depressions usually move over our islands com- paratively rapidly ; anticyclones move much more slowly, and not infrequently are stationary, or nearly so. for days. When changes in barometrical pressure are slow quiet weather may be expected. Even when the mercury stands low in any locality winds may be moderate or light and the weather dry in the locality. A low barometer, however, is an indication of atmospheric instabiUty, and violent winds may at anv time arise caused by increase in barometrical pressure beyond the area of low barometer. A slow fall in barometrical pressure is usually followed or accompanied by increased humidity and with rain, not necessarily with much increase of wind. Sudden changes in pressure, whether shown by a sudden rise of the barometer when it stands low, or by a sudden fall when it stands high, indicate a disturbed state of the atmosphere, and are invariably followed by increase of wind. It has been stated in the earlier part of this chapter that when a dee]> depression is followed by an area of consider- ably higher pressure than that which preceded it, the velocity of the wind is greater in the rear of the trough of the depres- sion than it was in front of it, because of the steeper pressure gradients thus produced in the rear. The sequence of weather conditions accompanying the passage eastward of a cyclonic depression that visited our islands on the 29th and oOth September, 1911, may be cited MS an appropriate case in point. At 6 p.m. on the 29th a high-pressure system occupied the eastern portion of the North Atlantic ; pressure being highest, lOoo mb. (30*5 ins.), immediately to the west ot the Bay of Biscay ; lowest, 982 mb. (29 ins.), over Scandinavia. A shallow depression which during the day had advanced from the Atlantic was situated to north-westward of our north-western shores. In the course of the night this depreission deepened, and moved rapidly eastAvard across Ireland and England, and at 7 a.m. on the 30th was centred near Spurn Head, with the barometer at about 999 mb. (29^ in>^-)? ^^^^ i)ressure increasing rapidly in its rear. The winds, showing a complete cyclonic circulation about this centre, blew strongly to a gale in places, and at 6 pm. ( l*late XI V), while the centre lay over the Netherlands, a stee)> gi'adient prevailed over the North Sea, the east of England, and all neighbouring parts of the Continent. Winds from northward, on the eastern and south-ejistern coasts of England and the adjoining sea, as well as in Holland and in the north of ^ Above 10. >< >. Wti»a» * 8«in 1/rr. Litht, M.O Pt»«>- UmhI-wi 8 V 7 Weather Signs. 95 Dew is an indication of coming fine weather. Its formation never begins under an overcast sky, or when there is much wind. Remarkable clearness of the atmosphere, especially near the horizon ; distant objects, such as hills, unusually visible or well defined or raised by refraction, and what is called a good hearing day, may be mentioned among signs of wet, if not wind, to be expected in a short time. When smoke from chimneys does not ascend readily, straight upwards, during calm, unfavourable change is probable. Near land, in sheltered harbours, in valleys or over low ground, there is usually a marked diminution of wind and a dispersion of clouds during the early part of the night. During bright weather and westerlj' (north-west to south-west) airs or light winds, the appearance of very high clouds of the Mare's Tail type (cirrus) moving from north-westAvard is usually an indication of the backing of the wind to the southward, and its increase in force, probably to a fresh or strong gale.* This movement ^f the very high clouds under such conditions is a very decided indication of bad weather, if at the same time a batch of such clouds l)e rising in the west, and the barometer, after rising, is inclined to fall. Again, when the wind is westerly or north-westerly of moderate strength, if high, hair-like or thready (cirrus) clouds appear moving from north or north-north-east they very commonly portend a great increase of wind from north-westward, with snow, sleet, or soft hail in winter." If the wind be easterly, and high clouds appear, similar to those just mentioned, but moving steadily from south-south-west, they point to an increase in the force of the easterly wind,t and during sultry summer weather, to the early approach of thunderstorms, followelace until near Christmas. As regards the cold period in Februar}^ it may perhaps be attributable as nuich to deficiency in solar heat as to the easterly winds that are I'refjuent in that month ; and, although the direction of the wind during the cold and warm periods of 'Inly, August, and December doubtless accentuates the interruption in tlie regular rise or fall in tem| erature in those months, it cannot be considered wholly to account for these recurrences ; for it is noticeable that when the cold j^eriods recur the air current is not always from a polar quarter ; noi is it invariably from an equatorial quarter when the warm j)eriods recur. 98 Types of Weather Conditions. It has been suggested that all periodic interruptions in the seasonal march of temperature are partially due to periodic variations in solar heat, the cause of which research in solar physics has yet to discover. As an aid to weather forecasting generally, and more especially to the presaging of changes by means of observa- tions restricted to a single station, the knowledo:e that annually, on more or less definable dates, departures from the normal sequence of weather conditions that are germane to the several seasons may be expected to recur will be found of no small assistance to the forecaster on the watch for any signs of coming changes. CHAPTER VIII. Types of Weather Conditions. The weather conditions of these islands, and largely of North- Western Europe, frequently maintain, during periods of varynig duration, a similarity as regards their general characteristics. Sometimes for weeks too-ether certain well- defined types of weather prevail, the general distribution of atmospheric pressure remaining, for the most part the same ; the wind backing and veering some points, under the influence of depressions which come chiefly from the Atlantic Ocean. These depressions, which usually approach our shores from the westward or south-westward, cause an increase of wind force, and not infrequenily a gale in some part or parts of our coasts. The sequence of wind and weather changes in these islands, associated with the passage of depressions, depends upon the paths they follow in concordance with the distribu- tions of pressure over North Western Europe and the adjacent se^is. An acquaintance with the salient features of the most characteristic of these types, added to the knowledge of their more than occasional persistence when established, will materially aid an observer in anticipating weather changes. Of the se^'eral types of weather conditions to which reference will be made, the south-westerly type is the commonest. So nth- Westerly Tyi ) e . When this type obtains barometrical pressure is relatively high over the Spanish Peninsula, and the Bay of Biscay, as South-W esterly Type. 99 well as to the south-east of England ; and relatively low over the north-eastern arm of the North Atlantic, including the Icelandic region. The centre of cyclonic depressions advancing from the westward or south-westward usually skirt the high-pressure area and pass away in a north-easterly direction. On the approach of a depression the wind from a south- westerly direction on our western shores backs somewhat and increases in force ; pressure diminishes, temperature rises, and a drizzling rain sets in, which is followed by a heavier fall. When the line of lowest barometer readings, or trough of the depression, has passed the locality of the observer, pressure increases ; the wind veers to south-westward, and perhaps more westerly later ; the temperature falls, the rain becomes intermittent, and soon afterwards ceases, the sky clears, and the wind moderates. If the recovery of pressure, or, in other words,' the rise of the barometer, in rear of the trough be rapid and consider- able, the wind will increase, probably to gale force on some parts of the coast, after the shift of wind to a more westerly point ; and this crisis may be attended by a heavy shower ; the wind will not moderate in that case until the rise of ]^ressure is checked. After a short interval of fair weather, the wind may be expected again to back in consequence of the approach of a new disturbance, and a similar sequence of weather to follow. Fig. 1, Plate XV, copied from the Daily Weather Report of 25th January, 1903, aflbrds a good instance of the severe weather that may attend the passage of a cyclonic depression while the south-westerly type- prevails. iietweeu I2th and 24th January, 1903, an area of high barometrical pressure situated over Europe barred the passage eastward, in the immediate vicinity of our islands, of disturbances coming from the Atlantic. A depression had on the 16th encroached sufficiently on our western shores to occasion a shift of wind from east to south-east, and the break up of a week's frost ; but this " low," repelled by the anticyclone, returned to the Atlantic or filled up. It was not until the approach of successive depressions had caused a considerable diminution in atmospheric pressure over our islands, and the f^rea of high ])ressure had spread southward and westward, that a complete change took place in the type of weather, which lasted well on into February. iii»7y B: 100 Types of Weather Conditions. On the morning of the 2Ath January, a cyclonic system was centred off the Shetlands, causing strong winds to \s gale on the northern and western coasts. At 8 a.m. next day, (see Fig.) this system had passed north-eastward, and a deeper disturbance had advanced from south-westward, to a position off the Hebrides. Over the north and west of Scandinavia, the northern part of the North Sea, and the major portion of tlie firitisli Isles, gradients were very steep, and southerly or south-westerly gales or strong winds prevailed, attaining to storm force on the M'est coast of Ireland. Along the North Atlantic seaboard a rough to very high sea was experienced. Southerly Type. This type closely resembles the south - westerly, irom which it often changes, and into which, it frequently merges. When the southerly type is established, barometrical pressure is relatively high over that portion of Europe which lies to the east and south-east of the United Kingdom. and relatively low over the North-Eastern Atlantic. Cyclonic depressions advancing from the Atlantic are usually held up by the high pressure, and. pass away in a northerly or rorth-north-easterly direction. On the a])proach of a depression, the wind, from a south- westerly or southerly quarter, l)acks and freshens ; pressure diminishes, temperature rises, and rain commences to fall. After the passage of the trough over the observer's locality, pressure increases, temperature falls, the wind moderates, and the weather clears. The Weather Chart for 8 a.m., 16th January, 1903 (Fig. 2, Plate XV), copied from the Daily AVeather Report, for that day, serves as a good illustration of disturbed weather occasioned by the incursion of a cyclonic depression during conditions characteristic of the Southerly type. At 8 a.m. on the loth a larp-e anticvclone was shown over North Western Europe, its centre situated over the southern parts of Norw ay and Sweden, where the barometer stood at 1043 mb. (30*8 inches). Pressure had commenced to gixe way over the United Kingdom, and over France where the wind was from southward and south-eastward, in harmony with anticyclonic circulation. During the night the centre of the anticj^clone moved slightly northward, and at the same time a deep C}'clonic depression, from the Atlantic, spread over our more northern and western districts, greatly increasing the pressure To face p. WO. PLATE XV. Fio. S Diractian wm) (ort« of Wiod Cabn, 1-8, 4-0, 7 & 8, & 10, n & 12, Squ&lk. Pic. « ill! .ij«7/.»< 10000 ifm Wtvu * Son i/r». Litho. M.O. fnm. Londoo, 8.W.7. Southerly and Westerly Types. 101 gradient over, and in the neighbourhood of, our Islands, where the wind, in consequence, steadily increased in force. At 8 a.m. on the 16th {see Fig.) the wind, from l)etween south and south-east over Ireland, Scotland, and the West of England, had increased greatly in force, and blew a fresh to whole gale in many places. Over our eastern and south- eastern districts it did not rjse above a fresh breeze, and was even lioht in places, its direction mainly south-easterly, but in France the wind came from eastward. The sky had become overcast in the north-east of Scotland, and in most parts of Ireland ; rain was falling at Valencia, but in most other parts of our area the sky was clear. The sea, rough or rather rough off the greater portion of our coasts, ran high to very high to the west of Scotland, and to the north and north-west of Ireland. Barred by the extensive area of high pressure referred to, the depression appeared, on the day following, to have broken into two parts, and to have moved subsequently, one in a south-easterly direction, the other north-eastward ; both, at the same time, becoming less deep. No material change appears to have taken place in the general distribution of atmospheric pressure over North Western Europe during the next five days, and the weather conditions of our western and northern districts continued to be affected by the incursions of depressions moving north- ward or north-eastward. On the 2ord January pressure had increased over the Bay of Biscay and the Spanish Peninsula, and diminished over Scandinavia and the northern portion of the North Sea. Thus the southerly type of weather conditions merged into the south-westerly type. Westerly Type. IJuring the prevalence of the Westerly type an area of high pressure, an extension northward of that which is situated at all times between the 20th and 42nd parallels of north latitude and is known as the North Atlantic anticyclone, is constantly to the southward of our islands, while pressure to the east, the west, and most markedly to the north, remains comparatively low. Depres- sions traversing the North Atlantic from west to east pass north of this high pressure, and thus, by producing a steeper })ressure gradient, cause an increase of wind from a westerly direction attended by precipitation. This type sometimes alternates with the south-westerly, into which it develops when pressure diminishes over France and the Bay of Biscay, and increases to the east and 1197'.! • E 2 102 Types of Weather Conditions. Bouth-east of the United Kingdom. If, on the contrary, pressure diminishes to the east and south-east, and the " high " over the Bay of Biscay expands northward, this type merges into the North-westerly and Northerly types. On the approach of a disturbance while the Westerly type obtains, the wind backs, and soon afterwards the barometer -commences to fall ; temperature rises, and subsequently rain «ets in ; but after the trough of the depression has passed the observer's locality the rain becomes intermittent, and the weather squally, with bright intervals. The wind will by that time have veered to the north of west ; temperature will have fallen, rather decidedly ; and the barometer will be rising. Occasionally the low pressure area to the north of our islands spreads southward towards the North Atlantic anti- cyclone, thus producing steep pressure gradients, with the result that strong westerly winds and gales are experienced, with very little change in the direction of the wind. The distriT3utions of barometrical pressure, and wind characteristic of this type of weather conditions, are well illustrated in Fig. 4, Plate ^\, which is copied from the Daily Weather Keport for 8 a.m. on the •23rd January, 1900. Fig. 3 exhibits for the same area the distributions of barometrical pressure and wind obtaining during the South- westerly type which preceded the Westerly. North-Westeiiy Type. The North-westerly type, alternating with the Westerly, is frequently experienced during the month of December, Dut may prevail in any other ' month. During its continu- ance, barometrical pressure is high over the Bay of Biscay and South AVestern Europe, while disturbances coming from the North Atlantic appear to the northward of our islands, and, crossing Scandinavia, take a south-easterly course across the Continent. The Chart for 8 a.m., 7th December, 1895 (Fig. 1, Plate XVI.), reproduced from the Daily Weather Report for that date, is selected to illustrate this type. It will be seen that the barometer was highest over the western and central parts of the Spanish Peninsula, while a cyclone covered Scandinavia, its centre of lowest barometer readings lying over the south of Sweden. Pressure gradients were steep over the North Sea, Denmark, Germany, and the Baltic, and gales from north-westward were experienced ever \^''estern Europe generally. To face p. 10 PLATE XVI. Fio. S e p.m 8th MAflO^\ North Eatteri 8 a.m. tnX FEBRUARY, tB85. Easterly Typ«. '■: 31li. 1^9*0/**. ucei W.it.11 * $k*t b»n« Uib*. M.O. i-Ttm ioodoB, H W.7. Tolactf 102. Northerly Type. 103 Subsequently the antic3^c]one having moved eastward, the paths of depressions crossing Europe became more northerly, and the type merged into the Westerly. These types ruled from the' oOth November to the 16th December, 1895. Northerly Type. When barometrical pressure becomes relatively high to the westward of the British Islands and re•ati^'ely low to the eastward, the wind over our area and its neighbourhood is drawn from northward, and a Northerly type of weather conditions results. Fig. 2, Plate XVf, from the Daily Weather Report for 8 a.m., 16th May, 1891, affords an excellent illustration of this type, which during the first or second week in May, or even later, is generally experienced, and is the cause of decidedly low temperature while it lasts. On the day to which the chart refers an extensive anticyclone over the North Eastern Atlantic, united with jinother area of high barometrical pressure covering Green- land. To the westward of the British [slands, and stretching across the face of the Bay of Biscay, the isobar of 1('19 mb. (30'1 ins.) is shown to lie almost- north and south, and the Kup|)ly of air to Europe consequently was drawn from Green- land and the Greenland Sea, northerly winds prevailing. An area of low barometrical pressure, having two centres of minima bnrumeter readings, lying north-east and south-west of each other, situated respectively over Lapland and over the southern part of Scandinavia, occasioned gales and very inclement weather over, and in the vicinity of, our islands and the North Sea. The conditions commenced on the morning of the loth Ma}" and lasted until the morning of the 19th. The maximum temperature in London on Whit x»Ionday, 18th May, was as low as 278'6a (42° Fahr.), whereas on the I2th and 18th it stood at 298-6a (78° Fahr.). No rth - Ea sterly Type. The characteristics of the North-easterly type, whicli frecpiently is the ])revailing ty])e in the month of March, are. demonstrated in the Weather Chart for 6 p.m., 8th March, 1898, from which Fig. 3, Plate XVI, is drafted. Areas of high barometer readings ai'c shown over the United Kingdom and Scandinavia, while the barometer is relatively low over Portugal, Spain, and the western lialf of the Mediterranean. liy7'J - E3 104 Types of Weather Conditions. The distribution of pressure associated with this type may continue for many days ; in the instance here cited it lasted from the morning of the 7th Marcli to the e^-ening- of the 12th, during which time depressions entering the Mediter- ranean from the Atlantic by crossing over the Spanish Peninsula, produced steep gradients in the vicinity of our shores, with the result that strong winds and disturbed weather were experienced. In rare cases, when the high pressure over Great Britain is less pronounced, these depressions take a more northerly course, and pass up the English Channel, causing in winter heavy falls of snow in our islands. Similar conditions caused a fierce blizzard from the 9th to the 11th March, 1891. (Fig. 4, Plate X-VI.) Reverting: to the foreo:oino; remarks relative to the con- ditions which dominated the weather from the 7th to the 12th March, 1898, inclusive, it may be mentioned in supple- ment that for some time prior to the 7th the area of high barometer was situated over, and to the westward of, the United Kingdom, and extended southward across the Bay of Biscay, but pressure over Scandinavia and Central Europe was relatively low, and the Northerly type, there- fore, prevailed during tha^ period. After the 12th March, barometrical pressure over Scan- dinavia was again relatively low, but to the south of the British Islands it had increased, the area of high pressure extending over the Peninsula, and to the Mediterranean ; thus the type merged into the Westerly, which predominated for many days. Easterly Type. The commonest form of the Easterly type, and one that often prevails in the month of February, is illustrated in Fig. 5, Plate XVI, from the Daily Weather Report for 8 a.m., 3rd February, 1895. The salient features in the distribution of atmospheric pressure associated with this type are as follow : — -The Xorth Atlantic anticyclone is situated farther south, and is less extensive than usual. Its influence is therefore lessened 'as regards our weather, which is dominated by an area of high barometer prevailing over Scandinavia, and a relatively low barometer over Central and Southern Europe, the latter having at times a westerly movement. As disturbances advance from the North Atlantic, approach and partially displace the Scandinavian anticyclone, they either fill up or pass away to the south-eastward ; thus producing easterly Easterly Type. 105 winds, which fluctuate between south-east, with a rise of temperature and foul, rainy weather, and north-east with fairer but colder weather. This alternation goes on while disturbances continue to arrive near our shores ; provided the general relative distri- bution of barometrical pressure remains the same ; and it may retain its characteristics for weeks consecutively. This type is usually preceded by the northerly, or north-easterly, and, with these, especially with the latter, it not infrequently alternates. The accompaniment of rain with east wind, and rising barometer, is a feature of this type. The easterly type of conditions described in the foregoing is only one out of several modifications of this type. Any decided decline in barometrical pressure that may take place south of the English Channel, wliiie an anticyclone lies over or to the north of the British Islands, will occasion easterly to north-easterly winds of more or less intensity and persistency. For instance, a low pressure system or depression situated over Spain, France, the Bay of Biscay, or the Gulf of Lyons may surge northward and produce steep gradients over the southern portion of England, causing stronor winds and gales on our south and south-east coasts, while the centre of the low pressure shows for some time little or no progressive motion ; or such a depression, centred over Europe, to the south-east of England, may expand to the westward, giving rise to similar results. The distribution of pressure and wind shown on the chart of barometer, wind, and weather, at 6 p.m. on the 23rd November, 1911 (Fig. 1, Plate XVII), which is copied from the chart for that time given in the Daily A\''eather Eeport of the following day, is typical of the first of these two cases. On the morning of the 23rd a rather large anticyclone lay immediately to the north, north-east, and north-west of England and Ireland, while a somewhat deep depression was situated over the Bay of Bisca) , the north of Sjxiin, and the south-west of France. This low-pressure system deepened, and spread laterally northward, the barometrical gradient between the two systems thus becoming steep. At 6 p.m. a gale from north-eastward prevailed in the English Channel, and by midnight a ircsh to strong gale was blowing on our southern and south-eastern coasts. The lov/-pressure system which had moved over the Bay from the North Atlantic on the 21st remained almost stationary, and on the 24th com- menced to fill up. Between the 21st and 24th, strong easterly and north-easterly winds and giales were experienced in the Channel and in various parts of the United Kingdom. 11079 E 4 106 Jypes of Weather Conditions. A good instance of the north-easterly to easterly type of conditions resulting from the expansion north-westward of a depression situated over central Europe, when pressure is high to the north-west and north of our islands, is shown in Fig. 2, Plate X^'II, which is reproduced from the synoptic chart for 7 a.m., on the 26th March, 1911, in the Daily Weather Report issued for that day. It will be seen that the central area of a large anticyclone, lying over the north-western arm of the North Atlantic, was situated immediately to the north-west of the British Islands ; and that a depression, which covered central Europe, was centred over A^enetia. On the previous morning the centre of the high-pressure system had been located between Iceland and the Hebrides, Moving south-eastward, and intensifying during the next twenty-four hours', while the depression referred to spread north-westward, the pressure gradient between the two systems became steeii, with the result that the wind, from north-eastward on the south-east coast uf England, the adjacent coasts of the Continent, as well as over the extreme southern portion of the North Sea, and in the Strait of Dover, blew strongly, gale and with force at times. Subsequently the anticyclone spread eastward, and the depression moved in a south-westerly direction ; the air current, therefore, veered to the eastward, the pressure gradient became less steep, and the wind moderated. This alternating easterly to north-easterly type of weather commenced on the 16th of March, and continued to the end of the month, when the conditions -became northerly. South- Easterly Type. In illustration of the distribution of pressure pertaining to the south-easterly type, the chart of pressure and wind published in the Daily Weather Keport, for 8 a.m. 6th January, 1897, is reproduced in Kig. 1, Plate XVIII ; the state of the weather, in letters of the Beaufort notation, taken from the accompanying chart of temperature and veather, being added. Pressure was then highest, 1036 mb. (30*6 ins.) and upwards over the Gulf of Bothnia and the northern part of the Baltic ; lowest, 982 mb. (29*0 ins.\, and less, in a deep depression having its centre off the south-west of Ireland. Gradients were rather steep over all the western parts of our islands; the wind was from south-eastward over western Europe generally, and on our extreme western and northern coasts blew sti ongly to a gale. This type of weather conditions, which had set in o\\ the 4th January, lasted until the 10th of the month. ^acep. 106. PLATE XV TI. BAROMETER, WIND AND WEATHER 6 p.m. 23rd NOVEMBER, 1911. tio-n. — Oo this chart a (lotted line •ep&rates areas over which rain haa fallcD within the ti houn fcrom 7 am. from those ivithout raia. Ueavjr lalls are given in fif uies. To face f. 106. PLATE XVIII. 8 a.m. eth JANUARY, 18S7 South -Eatterly Type. ';^/T3^ I0Z3 ' / I »96\ 8«.m. L J^^ aoih JULY, 1900. j,^*.; -v7- ^ "99 "7 " 1 ^ Li. , 9\ cl -W I ®6 -J- X ' '-A / -' 'J Ac/ I 6o ■~,.^' Pra. 8 DlrectioQ and force of Wind Cahn. 1-S, *-6, 7 & 8. B & 10. 11 & IZ, SqutOk. Fio. 4 'r »?2? lst4C/*4 10000 i/lt. WntAR * Sow Ud.. UUto. MO. Pr«M. Uculo^ ft.W. 7. Thunderstorms. 107 Typical Thunderstorm. Thunderstorms that occur over our islands are frequently, but not necessarily, incidental to the transition between two t^'pes of vveather conditions. For the purpose of examining the characteristics of their dev^elopment they can be grouped into three distinct classes : (1) Those coming to us from southward; (2) those forming locally ; (3) those appearing as secondaries to depressions in the north. Most thunderstorms are associated with secondary formations, or with the bend or " kink " in an isobar that may be regarded as a modification of such a formation. Thunderstorm coming from South. The class of thunderstorm first mentioned is the commonest. To cite an instance : on the 18th July, 1900, the weather over the British Islands and Western Europe may be said to have been under the influence of an extensive area of high barometrical pressure. On the morning of the 19th a shallow depression appeared over the Bay of Biscay, and at 6 p.m. of that day had travelled north, its centre, with the barometer at 1009 mb. (29*8 inches), beinof situated off the mouth of the Eno-lish Channel. A " kink " in the isobar of 1013 mb. (29*9 inches) which had spread eastward appeared in the vicinity of Lorient, indica- ting the probable formation of a secondary. Between this time, 6 p.m. of the 19th and 8 a.m. on the following day, thunderstorms occurred at Roche's Point, Pembroke, Scill}', and in London. Fig. 2, Plate XVIII, adapted from the weather charts in the Daily Weather Report for 8 a.m. on the 20th July, shows the distribution of pressure and wind, with the state of the weather at that time ; and the develop- ment over the north of France and eastern half of the English Channel of a secondary will be seen to have taken place as anticipated. The })rimary was then centred south-west of the British Islands. The isobar of 101 o mb. (29-9 inches) exhibits a bend across the southern entrance of St. George's Channel, and this locality was the scene of heavy thunderstorms on the 20tli. It has been found that- when lightning strokes occur most frequently numerous oscillations are shown in the barograpli trace, and that heavy rain falls when the barometer rises sud- denly "3, '1 or even TO mb. (2, 3 and 4 hundredths of an inch respectively). The conflicting currents in the atmosphere during thunderstorms, so noteworthy a feature, is exhibited by the motion of the clouds at varying heights. 108 Gales on the Coasts. Thunderstorms forming Locally. In illustration of the less common class of thunderstorms which form locally the distribution of barometrical pressure and wind and the state of the weather at 6 p.m., 31st August, 1896, are shown in Fig. 3, Plate XVIIT. On the morning of that day a shallow depression appeared over the south of England and the eastern part of the Channel. The 6 p.m. reports, when charted, showed that the disturb- ance had moved south-southeast, and was then centred near Cape Gris Nez. To the westward of this " low," round which there was shown to be complete wind circulation, as well as over Europe .generally, pressure was relatively high, but unevenly distributed. Between 6 p.m. and midnight, thunderstorms, accompanied in most cases with heavy rain, were experienced over the east of England and the centre of Ireland. Thunderstorms connected with Northern Depressions. The synoptic chart of barometer and wind relating to 6 p.m., 29th September, 1897, is reproduced in Fig. 4, with weather added, in letters of the Beaufort notation. It illustrates a disposition of isobaric lines that has been found to be essentially characteristic of the third class of thunder- storms ; those which appear as secondaries to depressions in the north. At 6 p.m. on the 28th September a large anticyclone covered Central and Northern Europe, and an apparently shallow depression was found to be approaching our north- west coasts, which, at 8 a.m., on the following day, had advanced and deepened. At 6 p.m. (Fig. 4, Plate XVIII) this disturbance developed over England and France a " V "-shaped secondary, shown by the isobar of 1013 mb. (29* 9" ins.), and during the night severe thunderstorms occurred over the south-eastern and south-midland parts of England, where the rainfall measured in most places exceeded an inch. The system passed away to the eastward, and it is probable that thunderstorms were experienced over the southern portion of the North Sea. CHAPTER IX. Gales on the Coasts of the British Islands. The statistics and illustrations given in this chapter relating to the prevalence of gales on the coasts of the Prevalence of Gales. 109 British Islands are derived with few exceptions, from Mr. F. J. Brodie's contributions to the Royal Meteorological Society upon that subject,* The results obtained by Mr. Brodie are based upon infor- mation collected in the Meteorologicil Office for the purpose of testing the accuracy of the Storm Warnings issued to stations on our coasts. Prevalence of Gales. During the oO }e-irs to which these results refer about 48 gales occurred annually on the coasts of the British Islands, and of these more than 10 were severe or partially severe : that Is to say, they attained to a force of at least 10 of the Beaufort Scale ; in other words, to a velocity of at least 59 miles an hour. A larger number of gales were, of course, experienced in some years than in others ; but it is noteworthy that those years in which gale frequency was either greater or less than the average, as the case may be, not infrequently followed one another for several years in succession. Dejinitiun of the term Uale. At the Meteorological Office, in accordance with inter- national agreement, the term (ja/e is applied to winds that attain to a force of 8 and upwards of the Beaufort Scale. Classification of Gales in Quadrants. To return to Mr. Brodie's statistics : For the purpose of classification, gales recorded on the coasts of the United Kingdom were grouped into four districts — the Northern, Western, Southern, and Eastern. The Northern District comprised the coasts of Scotland, the extreme northern coasts of England and Ireland, and occasionally a portion of the northern half of the Irish Sea. The Western District included Ireland and the west coasts of England to the mouth of the English Channel. Tlie Southern District comprehended the English Channel from its mouth to the North Foreland. The Eastern District commenced at the mouth of the Thames and followed the eastern coast- line to the Tweed. Distribution of Gales. It is found that on an average 48 gales occur aniuially, and out of this 11 are general ; that is to say, tliey are * "The Prevalence of Gales on the Coasts of the British Islaucls (luring tlie iJO j^ears l(S71-l'.'0l)." — Quarterlv .lournal. Meteorological Society, July, 1902, Vol. XXVIII. ; July, VM\, Vol. XXIX. no Gales on the Coasts. experienced in three out of the four districts ; eight are confined to the Northern and Western Districts ; four to the Western and Southern ; one to the Northern and Eastern Districts ; five are felt in the Western District only ; seven in the Northern only ; three in the Southern only ; and one in the Eastern only. Three gales experienced annually are classed by Mr. Brodie as sporadic, by which it is meant that they are felt at a considerable number of isolated places, scattered over a wide area, and five are classed as local, the term local meaning in this connexion that they are felt at several stations, but in a portion only of one district. Average Frequency of Gales. With regard to the average prevalence of gales in different seasons, 10 occur in spring, three in summer, 15 in autumn, and 20 in winter. The fewest gales, as might be expected, are experienced in the months of June and July ; the largest number in January. In February a falling off in the number of gales is shown, but there is a slight increase in March ; after about the middle of March, gale frequency steadily declines, rapidly until about the middle of April, less rapidly, but none the less steadily, until the middle of June. The mean monthly prevalence of gales is shown in the accompanying diagram (Fig. 8). July Aug Sept Oct. Nov. Dec. Jan. Feb. NAar. April ^'Iay Junl r^\' \x r^-- ''^M. w 4:..-il 1 !"- — ^ g^.xi u I i ^ ^ f I , r^ Ga/es of a// kinds shown /-hus ■ " ■ ''^^^^^ Seirere ^ Parh'al/y Severe Ga/cs s/io*vn f^hus. J»i;^S888S Fig. 8. — Mean Monthly Frequency of Gales. Adapted from the figure given by Mr. Brodie. The average number of gales in each of the 12 months is shown by the height of the columns ; the scale of numbers is shown on eack side of the diagram. Gales on the Coasts. Ill Wind Direction in Gales. Respecting the direction of tiie wind in the gales, out of the 48 referred to as about the average number that visited the coasts of the British Islands, about 10 per cent, blew from northward and north-eastward, 14 per cent, from eastward and south-eastward, o8 per cent, from southward and south-westward, and 80 per cent, from westward and north-westward. Eight per cent, of the gales were associated with cyclonic depressions, the centres of which passed over the more central parts of the United Kingdom, so that, the wind having complete circulation about a central area of minimum pressure, moving eastward, there was no quarter from v,'hich it did not blow in some part or parts of our islands, and attain to the force of a gale, during the progress of the depressions from the Korth Atlantic to the North Sea To such gales the term vortical was applied by the author. The highest proportion of equatorial gales, i.e., gales from between south-east and west round by south, occurred in the months of December, 68'6 percent.,and January, 68*9 percent.; the proportion was, however, nearly as high in February, H7'7 per cent. : in each of these three wiater months more than two-thirds of the gales blew from some equatorial direction. The proportion was lowest in March, 55 per cent. ; but was low also in October, 56*9 per cent, and some- what low in November, 58 per cent. The proportion of gales from polar quarters was greatest in July, 40'9 per cent., which appears singular, but, as Mr. Brodie points out, the total number in that month was so small that the result was not considered satisfactory, and a more extended series of observations would, it was thought, be necessary before such a conclusion could be accepted as representing a general rule. Next to July, the greatest proportion of polar gales in any month, 40'7 f)er cent,, occurred in March, and April iiad the third highest percentage, ;-)7"9 per cent. The smallest proportion by far of polar gales, 19*6 per cent., occurred in December, and this leads to the somewhat paradoxical con- clusion that cold-wind gales were most common in the warmest months of the year, and least common in the coldest months. The })roportion of vortical gales was greatest in -J une, \i'?) per cent. ; but here again the total number of gales recorded was too small to yield a conclusive result. Next to 'Fune the greatest proportion of such gales was found to have occurred in December, when 11'8 per cent, of our gales are due to cyclonic systems, the centres of which advance directly over the United Kingdom. In May, and in each of 112 Gales on the Coasts. the autumn months, September, October, and November, about 10 per cent, of the gales are of this character, but in April the percentage is only 2*4, and in July such gales did not occur. A more minute subdivision of the various directions from which gales blow in different months shows that the pro- portion of gales from north and north-east is greatest in N^ovember and least in June ; the proportion of those from east and south-east is greatest in April, and they do not appear to occur in June ; the proportion of gales from south and south-west is greatest in January and least in April ; and from west and north-west is greatest in July and least in November. \\ ith regard to the direction of. the wind in gales w^hich visited our islands, rather mor.e than 10 per cent, blew from northward, 14 per cent, from eastward and south-eastward, 38 per cent, from southward and south- westward, and 30 per cent, from westward and north-west- ward. Nearly 8 per cent, are vortical. It is apparent from the foregoing that on the coasts of the British Islands generally gales from between south and south-west are the most frequent, and that the largest number blow from south-westward ; after that a'ales from between north and east are the least frequent. fe' Relative Fraquency of Gales from various Points of the Compass. With reference to the relative frequency of gales from various points of the compass {see Fig. 9) during spring, ofales from between south-west and north-west were most frequent, those between north and east least frequent, but more frequent than at any other season of the year. In sunnuer gales blew almost entirely from between south-west and north-west, the largest number being from west. In autumn south-west gales largely predominated, but a number of gales were experienced from south, from west, and from north-w^est. In winter gales from south and south - west were most frequent, those from south - west predominating. Gales from north-west and south-east w^ere not infrequent during winter, but from between north and east they were of comparatively rare occurrence. The wind-roses in diagram. Fig, 9, show the prevalence of gales from various directions in each of the four seasons of the year. The figures enclosed Vv4thin each of the inner circles give the percentage of " vortical " gales, while the thick lines pointing towards these circles show the propor- tion of gales from each of the eight principal points of the Relatice Fre(iaeneij of Gales. 118 compass. The length of these thick lines is determined by the outer circles, each of which, counting from the inner one, represents a portion of 5 per cent. For purposes of com- parison with the general results, the central gale-rose gives the proportion for the entire year. S PR I NG SUM M ER the: year AUTUMN W I NTER Fig. 9. — Relative Frequency of Gales from various Points of the Compass. Relative Frequency of Gales on the respective Coasts of the British Isles, from different Points of the Compass. So far the direction of the wind in gales has been considered only with regard to our coasts generally ; the prevailing directions, however, vary considerably according to the district. {See Fig. 10.) On the north coasts about 1 J per cent, of gales were from northward and north-eastward, about 15^ per cent, from 114 Gales on the Coasts. eastward and south-eastward, about o7 per cent, from south- ward and south-westward, 32J per cent, from westward and north-westward, and about 4 were vortical. GALES SN NORTH CXLtS IN wesT GALES m EAST GALE.S iM SOUT H l^'IG. 10. — Relative Frequency of Gales on the respective coasts of the British Isles, from diflEerent points of the compass. In Fiof. 10 the direction from which o-ales blow in the .... . . mi various divisions is shown graphically by wind roses. The figures enclosed within each of the inner circles give the Relative Frequency of Gales. 115 percentage of " vortical " gales, while the thick lines pointing towards these circles show the propartion of gales from each of the eight principal points of the compass. The length of these thick lines is determined by the outer circles, each of which, counting from the inner one, represents a proportion of 5 per cent. On the south coasts the percentages w^ere respectively about 8 J from northward and north-eastward, nearly 10 from eastward and south-eastward, 4(3 from southward and south- westward, 335 from westward and north - westward, and about 2 were vortical. On the east coasts the percentages of gales were about 21 5 from northward and north-eastward, nearly 25 from eastward and south-eastward, nearly 28 from southward and south-westward, 224 from westward and north-westward, and nearly 3j vortical. On the svest coasts nearly 7^ per cent, of gales were from northward and north-eastward, nearly 13 per cent, from eastward and south-eastward, nearly 45 per cent, from south- ward and south-westward, and rather more than 29 per cent. from westward aud north-westward ; nearly o per cent, were vortical. With reference to the gales associated with cyclonic depressions, the centres of which pass over the more central parts of the United Kingdom, and to which the term vortical has been applied, it may be well at this point to state that almost all gales which are experienced in these islands are associated with wind systems of a cyclonic character, the characteristics of which will be explained in the next chapter, but that the central areas of low pressure in these systems, round which the wind circulates, follow most frequently a path that lies to the northward and westward of our northern and western seaboard. When this is the case our islands come under the influence of winds, related to the seaward semicircle of the cyclonic system : from between south-east and north-west. The large number of southerly and south-westerly gales which visit our northern and western shores are caused by these cyclonic de})ressions, which travel in a north-e.isterly direction across the north-eastern arm of the North Atlantic, and })ass between the north of Scotland and Iceland, or over Iceland. They rarely cause an increase of wind to gale force on our south-east and east coasts, because the centres of the depressions pass far to the north-westward of Ireland and Scotland. 116 Gales on the Coasts. Tracks of Storm Centres. The tracks ordinarily pursued by storm centres are shown in Fig\ IL With reference to this illustration the author mikes the following statement : — The undue preponderance of gales from south-west, and west on our southern coasts is at once accounted for by the fact (doubtless a very familiar one) that the vast bulk of the storm systems move along either in an easterly or a north-easterly direction over the more northern parts of our area, and that wo in the south get the winds blowing round their southern and south-eastern sides. In the west many of the gales from south and south-west are due to cyclonic areas moving along in the directions shown hy the storm tracks A and B*, these disturbances being in very many cases too remote to cause any serious increase of wind in the south and east. In the north a gale is also caused very frequently by these two sets of storm systems, while in the rear of Class A it is not unusual for a gale to spring up from west, or even north-west. The relative frequency of gales from these latter directions in the north is also partially accounted for in disturbances moving in the tracks marked G and D ; the latter must, however, be very large and deep to affect our own coasts at all seriously. In the north, and to a limited extent in the east also, storm systems moving along in the tracks E and J will, if of sufficient depth, cause gales from south-east, and, as these two types are not uncommon, 'we see at once that the percentage of gales from that quarter on the coasts mentioned i3 somewhat large. The high percentage of gales on our east coasts from some easterly direction (south-east to north-east) is accounted for, but only partially, by cyclonic systems travelling in the somewhat low tracks marked G, H, and J. Many of the easterly gales both on this and on our southern coasts are, however, caused by storm systems, the centres of which fail to come anywhei-e near our islands. It is not unusual for a cyclonic area which appears over Spain, or even over the Mediterranean, to remain almost stationary for some time but to spread out laterally in a northerly direction, the surging process resulting in a gradual increase in the easterly gradient over our own southern and eastern districts. Another but rarer type of gale which blows from eastward to north-eastward is due to cyclonic systems surging westward and north-westward from Central Europe ; in some cases the centres actually move out in this direction to the neighbourhood of our eastern and southern coasts, but these instances are so infrequent that I have omitted them altogether from the map of storm tracks. The high proportion of vortical gales on the west coasts is due very largely to the fact that the district in question covers a wide range of latitude, some 350 miles from north to south, and the possibilities of a storm centre moving eastward or north-eastward over the more central parts of the district and causing a complete cyclonic circulation are therefore great. * Since the year 1906, when telegraphic communication with Iceland and Faeroe was established, observations have been received at the Meteorological OfBce daily, and have been utilized in the preparation of the Daily Weather Report. Had such observations been available when Mr. Brodie drafted Fig. 11, he would, doubtless, have placed tiacks A and B much nearer Iceland than he placed them on this map. Tracks of Storm Centres. 17 Next to the west coasts, vortical gales are most common in the north and east, these being due respectively to storms moving in the tracks E and F. In the southern district the range of latitude embraced is very small, and as very few storm centres move across it in a due easterly, and scarcely anj^ in a due northerly or southerly direction, the gales experienced there are seldom of a vortical character. Fig. 11. — Track-i ordinarily pursued by cyclone Centres appearing over various parts of North Western P]urope. (The relative frequency of each track is sliown by the width of the shaded line.) Since 1007 tlie ai-ea under daily observation has included Faeroe and Iceland, ami the track 13 has been found to have been drawn too near the European side. Speed of Storm Centres. As regards the speed at which storm centres travel, Viv. Hrodie found that many of the centres of the larirer 118 Gales on the Coasts. cyclonic disturbances follow paths lying outside our area of observation ; so that although the direction of movement can usually be determined with a reasonable degree of accuracy, it is often impossible to locate the centres of the systems from time to time, and thus measure their rate of travel. Out of oil important storm systems the speed of travel of not more than 267 could be determined. The average rate at which these moved was found to be 24'1 geographical miles per hour. The variations in the speed of travel were so great, how- ever, that the average appears to be of little value ; it included, for instance, numerous storms that showed a speed of not more than 8 or 10 miles per hour ; while in the case of several others it amounted to 40, 50 and even 60 miles per hour. The rapidity of motion was found to vary in the different seasons of the year, and according to the data relating to summer was greatest in that season ; but as in the 30 years' records only seven summer gales could be classed as severe ; the evidence in this connexion, as regards that period of the year, is obviously insufficient. The average speed of travel in the autumn was found to be 21*7 miles per hoar, in the winter 2o'4 mdes, and in the spring 2o"4 miles. Thus, disregarding the results relating to summer, the storms of winter appear to have the fastest rate of travel, the autumn storms the slowest. Average results referred to all months of the year indicate that the storms which attain the maximum rate of travel, which is 25'1 miles per hour, move in an easterly direction, i.e., east-north-eastward, eastward, and east-south-eastward. Storms moving northward and north-eastward, southward and south-eastward, hiive about an average rate of travel 24 miles per hour. Storm systems that moved in irregular paths were found to have irregular rates of travel, and the a\Trage speed of the comparatively few that were examined for this purpose was found to be as low as 19*1 miles per hour. Systems moving in north-westerly and south-westerly directions, instances of which are rare, showed a comparatively slow rate of travel : the average speed being 13*4 and 10*7 miles per hour respectively. There were 60 cases, during the 30 years' period under examination, in which storm systems travelled at a higher speed than 30 geographical miles per hour. Of these 25 travelled at speeds of 31 to 35 miles, 14 at speeds of 36 to 40 miles, 11 at speeds of 41 to 45 miles, and 4 at speeds of Speed of Storm Centres. 119 46 to 50 miles. The remaining 6 cases included 3 that travelled at 51 to 55 miles, 2 at 56 to 60, and 1 ac 62 miles per hour. The exceptional speed of 62 miles was attained by a small secondary cyclone which travelled over Ireland, England and the North Sea in a north-easterly direction on the 24th March, 1895. CHAPTER X. Icebergs and other forms of Drifting Ice. Ice Formation. Ice is a mass of stellate crystals, which form in a freezing fluid while in process of passing from a liquid to a solid state. The crystals usually take the form of six-pointed stars similar to those of which particles of snow are composed. Ice is formed also from snow by pressure, which, by ex- pelling the air intermingled with its particles, renders it hard and transparent ; for it is the close contact of the crystals, blending as it does the individual particles into a compact, continuous mass, that makes ice more or less transparent. TyndaWs Experiment. The beautiful structure of ice may be seen by repeating an experiment that was made originally by the late Professor Tyndall, which may be described as follows : — A thin slab of ice is cut parallel to its plane of freezing. This plane can be iiscertained by examining the direction in which the bubbles in the ice are distributed either thickly, ])arallel to the surface of the water when freezing or sparsely in stria> at right angles to it. If the disintegration of the ice be watched throuo^h a lens ... . ^ while a ray of light is allowed to pass through it, numerous small crystals will be seen gradually assuming star-like forms of rare beauty. In ])olar regicms, and on mountain heights above the anoin line, or limit above which snow perpetually lies, some melting takes place at the surface of the snow under the rays of the summer sun. The water thus melted at the surface percolates to the lower layers of snow, raising their temperature to that of 273a (32° Fahr.) ; and while this is taking place the weight of the snow above presses the lower layers into a coherent mass, 120 Icebergs. from which the air is expelled by further compression, and ice is formed by regelation. Reg-elation is a property of ice which was discovered by Faraday, who found that when two pieces of melting ice are placed together they adhere one to the other by freezing at their places of contact. Faradai/s Discovery and Explanation of " Reg elation. ''"' Faraday's explanation of this phenomenon is, in effect, as follows : — The particles on the surface of a block of ice are held together by cohesion on one side only, the side tl at is iiot exposed to the air ; whereas those particles that are in the interior of the block are pressed together on all sides. With the temperature at 273a (32° Fahr,), the particles exposed melt first, and form a film of Avater over the surface of tiie ice. Now, if the block of ice be cleft in twain, melting takes place of the fractured surfaces also. By placing the two halves together in their original position the melting surfaces, where fractured, are again pressed on all sides by frozen particles, and become one by regelation. The accepted Explanation. The accepted explanation, however, is that given by Prof. J. Clerk Maxwell, in his "' Theory of Heat." It is as follows : — When two pieces of ice at the melting point are pressed together, the pressure causes melting to take place at the portions of the surface in contact. The water so formed escapes out of the way and the temperature is lowered. Hence, as soon as the pressure diminishes the two parts are frozen together with ice at a temperature below 273a (32° Fahr). Illustration by Dewar. In one of the Christmas lectures to children at the Royal Institution, in December and January, 1912-19i3,» Sir James Dewar, in order to illustrate by a well-known experi- ment the fact that ice cannot be cut, furnished also an example of adhesion by the above process. A fine wire, with weights attached to each end, was suspended on a block of ice. The wire gradually cut its way through the ice, but when it emerged at the other side the block was found to be intact. The ice, or water in a solid state, as the lecturer preferred to describe it, had liquefied in front of the wire by pressure, and the cleft parts behind the wire had reunited by regelation. Glacial Regions. 121 The regelating property of ice was applied by the late Professor Tyndall to explain the formation of glaciers. The Formation of Glaciers. Glaciers are rivers of snow, which have their origin in elevated regions, and are compacted into ice in the manner described. This compacted mass of ice is urged dowaiward by gravity, and by the increasing pressure of the masses in the rear, caused by the continuous accumulation of ice and snow. A glacier behaves in all respects like a river : its rate of motion accelerating in its steep descent through narrow passes and becoming slower where its channel widens. Analogy btticeen River Flow and Glacial Motion. Moreover, as a river flows with a greater velocity in the middle of the stream than is the case at its sides, because of the greater friction it has to overcome near its banks, so the motion of a glacier is faster at its centre than at its margins : and the analogy may be carried still farther, for both river and glacier move taster at their surface than near iheir beds. The Colouring of Icebergs and the Causes. The whiteness of icebergs is due to the tine lineal pures, equidistant, parallel, and of about the same size, which are uniformly distributed throughout the greater ])ortion of the mass. In the broad stripes, clear and of a dark-blue colour, which intersect a berg in places; air bubbles are either absent or else they differ in size, and are distributed sparsely and irregularly. Glacial Regions. — Formation of Icebergs of Glacial Origin. [n arctic and antarctic regions glaciers are found at all levels, and they frequently occupy extensive tracts of country ; but in temperate and tropical zones they exist only on mountains at great altitudes ; and near the equator they do not f(jrm below a height of 16,000 feet above sea level. In these zones a glacier melts when it has descended to a level below the snow line, where it becomes the source of a river ; but in polar regicms a glacier continues its descent to the lowest levels and carves its way to the coast. Being still under the influence of pressure from behind, the glacier gradually protrudes seawanl, but its base rests on the ground, until it reaches a sufficient depth to become water 122 Icebergs. borne. This protrusion continues until, as a result of resist- ance from continuous pressure in rear, but also through the agency of sea disturbance, the so-called calving of an iceberg (German : Eis : ice ; Berg : mountain) takes place ; a portion of the floating mass of ice is severed from the parent glacier, and, when the wind is Pavourable to its release, the berg- drifts out to sea. Bergs of Ice-Barrier Origin. All icebergs, however, are not of glacial formation ; a large percentage of these masses of lioating ice originally formed part of ice barriers. Ice Barriers and their Formation. The origin of ice-barriers is yet a matter of uncertainty, but it seems probable that they are formed by a gradual and steady extension seaward of the ice frozen to the shore, which is called the ice-foot, and to the growth Aertically ot the extended ice -foot by the formation of new^ ice on its surface from snow through the process of regelation. Icebergs that have lormed part of an ice-barrier, or of an ice-clif that extends from and is attached to the face of a precipitous foreshore, are generally tabular : a natural result of their formation. Most of the bergs of the southern hemisphere, as wdll be shown later, are in this form. Field Ice : Its Formation and Colour. Field ice, w^hich is flat ice, often occupying a large area and rendering it unnavigable, is formed near the shore, and, w^hen in motion, being more under the influence of wind than of current, it frequently ^^//es, thus becoming uneven. Fragments of bergs, or growlers, as they are termed, which at times become trapped and embedded in field ice, are an additional menace to shipping, although they can, as a rule, ]je detected at some distance in consequence of their dark- blue colour. Floating Ice in Other Forms. Floating ice in other forms is described in the '" JSTew- foundland and Labrador Pilot,"* as follows : — FJoe ice consists of several pieces of field ice frozen or pressed together. * " Newfoundland and Labrador Pilot," Fourth Edition, 1907, pp. 27 and 28. Published by order of the Lords Commissioners of the Admiralty. Detection of Drifting Ice. 123 Land we is field or floe ico attached to the shore since the winter. Hummocky ice is formed by the edges of ice floes meeting in strong breezes, when they are pushed up and formed into pj-ramids, which are then named Jiumr/iocJcs ; and it is of these that the high mounds of ice met with in the Gulf of St. Lawrence are generally composed. Pack ice is a large collection of pieces of ice from broken-up floes or icebergs, which have, to a certain extent, closed together again. The pack is said to be open when it presents leads or lanes of w.tter between the pieces of ice, forming more or less navigable channels ; and close when it is not possible to navigate through the pack. Drift ice it unattached pieces of floating ice, easily navigable. Brash or sludge ice is a collection of very small pieces of broken- up ice through which a ship can easily force her way. Pancake ice is newly-frozen ice of sufficient thickness to prevent navigation, and is sometimes separated into pieces of a form sugges- tive of the name. Bay ice is newly - frozen ice sufficiently thick to prevent navigation. A Floeherg is a thick piece of salt-water ice presenting the appear- ance of a small iceberg. Slob ice, which is the first ice to form, should also be mentioned It is crushed by wind and sea, and piled to a height of from 3 to 10 feet. Sometimes the harbours and bays of Newfoundland are filled by it. The Detection of Drifting Ice. When navigating in regions where ice may be encountered, no possible means for detecting its presence should be' neglected. For many years past the use of the thermometer in this connexion has been discredited ; yet instances are not wanting to prove that timely warning of ice, in the immediate neio'hbourhood of a vessel, which cannot be seen owing to thick weather, may be given by means of observa- tions of sea-surfiice temperature, particularly on occasions when the current is setting from the direction of the ice towards the vessel. The Recording Micro-Thermometer. For registering minute changes in sea temperature a recording micro-thermometer has been devised by Professor H. T. Barnes, F.R.S., Director of the Physics Department in the ^IcGill University, ^Montreal. This instrument, which is designed on the principle of the electrical-resistance thermometer, was tested on board one of the hydrographic survey vessels employed by the Canadian Government during a voyage from the St. Lawrence to Hudson Ixiy, passing through the Strait of Belle Isle, in the year 1910. It was found that a rise in the temperature of the water at five feet below the surface was registered as the vessel drew 124 Icebevifs. near an iceberg ; but that this rise was followed by a fall which continued until the berg was in a position nearest to the vessel. The sea then gradually became warmer as the distance from the berg increased, until the latter ceased to have any effect upon the temperature of the former. The explanation offered by Professor Barnes in regard to the rise in sea tem])erature when neariug the iceberg is in effect as follows :— The fresh water melting from an iceberg- is carried downward by the convective action of the inflowing salt water ; and the surface water, which, he says, has no tendency to sink, because of its horizontal motion, and is not dispersed by sea movement, absorbs and accumulates all the heat radiated from the sun and sk}' ; and in this way the water around the berg becomes warmer than that of the surrounding- sea. This explanation is given by Professor Barnes in his Report to the Deputy Minister of Marine and Fisheries, Ottawa.* It is possible, however, that by dilution with fresh water from an iceberg that has been carried below the surface by convective action, a warm undercurrent of relatively high salinity may be brought to the surface and cause the rise of temperature referred to. The inventor subsequently tried the apparatus during a passage from Halifax to Bristol ; and, in addition to testing- its utility in the detection of drifting ice, he claims to have obtained interestins; results in connexion with the disturbine- influence of land on the temperature of the sea. The Ice Blink. The ice blink : an effulgence, reflected from ice, which is seen in the sky near the horizon in its direction, generally indicates the presence of this danger, and renders a berg that is snow-covered distino^uishable at some distance. At short distances this brightness may assume the appearance of a white cloud. Fi7^st appearance of Icebery in lliick Weather. When approaching an iceberg in thick weather the first appearance it presents is that of a relatively dark object. * " Report on the Influence of Iceberg^s and Land on the Tempera- ture of the Sea, as shown by the use of the micro-thermometer on a trip of the C.G.S. ' Montcalm,' in the Gulf of St. Lawrence and the coast of Labrador, &c." By H. T. Barnes, D.Sc, F.R.S., Director of the Physical Laboratories and Professor of Physics, McGill Uuiver.-ity, Montreal. Sessioual Paper No. ^Ic. Sup])lement to the forty-fifth A.nnual Report of the Department of Marine and Fisheries for the nscal year 1911-12. Ottawa, 19.1 3. he in the \<>rthern [Temlsphtre. 125 The Whistle or Siren. Tlie whistle or siren of a steamship, sounded at frequent intervals, has been tried by some navigators for the detection of icebergs in thick weather, with the object of obtaining an echo from the mass which would reveal its presence in the immediate neio'hbourhood of the vessel in sufficient time to avert a collision with it. If this method has on any occasion been found to succeed, the tact, it is believed, has not been published. Ice in the Northern Hemisphere. The glaciers of Greenland to which the majority of the laro-e icebero-s that drift southward into the Xorth Atlantic doubtless owe their origin, occupy niimense tracts ot country extending far inland. The area of Clreenland, including its outlyijig islands, is estimated at about 827,275 square miles. Its extreme length, assuming its northernmost point to be situated on the north coast of one of the two large islands to the north, in latitude 83° 40' N., longitude 31° 15' W., and Cape Farewell, which is also on an island, its southernmost point, is about 1,<){)0 statute miles. Its extreme breadth, which may be accepted as the measurement between Cape Alexander in 78° 11' N.. 73° 13' \V., and Cape Bismarck in 78° 47' N., 18° 30' W., is about 820 statute miles. The rugged, precipitous coasts of Continental Greenland are, for the most part, characterized by deep glaciated indentations ur fiords ; that have been, and continue to be, the birth-places of icebergs, which break away from pro- truding glaciers ; also by numerous islands, many of which li^* at the entrance of these tiords. This is specially the case between the ()4th anur 2,000 feet to an elevation of 9,000 feet or more. This inland ice sheet is estimated, by General Greely, rhi- famous American Arctic explorer, to have a thicknes-^ ot 1,000 to 5,000 feet.* '^ Handbook of Polar Discoveries. By A. W. (i;of ly, Major General, United States Army. London, T, Fisher Unwin, lOlO, 126 Icebergs. Owing to the barrier, of land and sea ice. the exploration of the East Coast of Greenland is not yet complete ; and the greater part of the coast fro:n Capa Farewell, northward, has been surveyed from time to time under great difficulties arising from the prevalence of fog. The land ice may extend, it is stated,* to ten miles from the shore ; and includes icebsrgs that are calved from inland glaciers ; while outside of this inland ice, a stream of drift ice, from the Polar basin, consisting, presumably, of both sea and land ice, is carried north-westward by a constant cujTent The breadth of this ice-drift varies during the year ; in spring it stretches from about Cape Dan to Cape North, in Iceland ; thence north-eastward to the island of Jan Mayen, and north-westward to Spitzbergen. South of Cape Dan the ice-drift, which there is a comparatively narrow belt, skirts the coast to Cape Farewell. Diverse ophiions are held with regard to the origin of the inland ice : there are those that support the theory, formed by Dr. H. Rink, of Copenhagen, of an ice deluge, that originated in the valleys by the freezing up of the river systems and spread upwards over the water-sheds until the whole land was covered. Others believe that the formation of the ice mantle com- menced on the lieio'hts : and streamino; down from immense masses of perpetual snow, welded together in its descent, the various ramificaticns into an uniform mass, ultimately spreading over the land, covering valleys and heights alike. The first theory presupposes excess of cold, and of pre- cipitation : because the temperature now prevailing in Greenland is not sufficiently low to maintain throughout the summer the ice formed during the winter, even in the smaller rivers; and to allow moreover an increase of ice in the winter followino-. The second theory refers the formation of the ice mantle to excessive precipitation ; contending that that alone is sufficient explanation. As regards the extent and the outward form and elevation of the inland Ice ; the most reliable information is that which is imparted by the intrepid Morwegian explorer, Fridtjof Nansen. in the narrative of his iamous journey '* Arctic Pilot, Vol. II, Second Edition, 1911. Published by order of the Lords Commissioners of the Admiralty. Dr. Rinl: on Glaciers and Icebergs. 127 across the Continent, entitled Ihe First Crossing of Greenland.* Nansen states that the inland ice stretches in an unbroken sheet over that part of the Continent along which he and his party travelled ; and that he feels justified in concluding that from the 75th parallel southward the whole country is similarly covered ; there being no reason to suppose that the atmos- pheric conditions are not approximate! v the same over the whole interior. He holds in fact that his investigations actually supplied him with evidence in that direction. Nansen concluded therefore, with a high degree of con- fidence, that in the southern part of Greenland, across which he made his journey, there were no oases ; although, as exceptions to the general rule, there may be peaks in the interior which project above the ice sheet. He added, that as yet nothing had been observed m Greenland which would lend support to a supposition of their existence. L'he last • nunataks ' f that they passed, he says, on the eastern side of the Continent, were little more than thirty miles from the sea board, and were plainly visible from the mountains on the coast. Dr. Rink on Glaciers and Icebergs. Kink resided for several years in Greenland. He imparted much interesting information upon this point in a paper he contributed to the Koyal (Tcograpliical Society in the year 18oa,J in which he makes the following statement : — For the formation of icebergs a tract of land of a certain extent is necessary, in which the sea forms so fevv and small creeks or inlets, that rivers or water courses of some magnitude must necessarily be present. Where the above-mentioned condition exists, in conjunction with the necessary temperature oi" the climate, the formation of ice does not proceed from certain mountain heights, but the ivhole country is c'jVe red iiith ice to a artain elevation ; mountains and valleys are levelled to a uniform plain ; the river beds are concealed, as well as every vestige of the original form of the cnintry. He goes on to say that the outer edge of this mass of ice is thrust forward towards the sea by a movement which connnences far inland : and that when the ice reaches a frith * The First Crossing of Greenland. By Fridtjof Nansen. Trans- lated from the Norwegian by Hubert Majendie Gepp, B.A. Lecturer at the Univereitv of Upsala. Vol. ii. London, Longmans, Green & Co., 1890. t A nunatak is a peak, hill, nr mountain top, which rises above the surface of the inland ice. X "On the large Continental Ice of Greenland, and the Origin of Icebergs in the Arctic Seas." By Dr. H. Rink, of Copenhagen. Journal of the Royal Geographical Society, Vol. xxiii. 1853. 128 Icebergs. it may be seen to sink and to diverge, and even to extend subsequently for several miles. Then, through the agency of the oUiterated rivers, the glacier is carried forward to the ocean ; and reaching the shore, while still preserving its continuity, protrudes seaward, borne up by the sea ; until, by sea disturbance, by its own weight, or some other agency, equilibrium is destroyed, and a portion of the floating mass parts from tlie parent glacier and an iceberg is born. Commenting on the fact, regarded as remarkable by Scoresby, that icebergs are rarely met with in the neighbour- hood of Spitzbergen, Dr. Rink pointed out that, neither that island nor the narrower parts of Greenland, nor lands adjacent to that continent, are adequate as regards extent to produce the yearly excess of ice which is necessary for their formation. Such conditions, he considered, were found only in Green- land, and were characteristic more especially of that portion of Greenland which lies to the north of the Arctic Circle, where sufficient space is afforded to serve as the cradle of large icebergs. . The opportunities Dr. Rink enjoyed, during a somewhat prolonged residence in Greenland, of becoming acquainted with that part of the west coast which is ^situated between 68° N. and 74° N. latitude, were unique, and lend to his evidence regarding the formation of icebergs in the Arctic a special weight. Ice -friths. He expressed the opinion that the largest icebergs which drift southward in Davis Strait issue from ice-friths, as the friths or fiords which transmit the bergs are called, r.ituated in the above - mentioned region. His knowledge in this connexion was based, partly on his own observations, and partly on information derived from intercourse with the inhabitants. There exist. Rink concluded, five principal ice-friths along the coast, from G7|° IST. to 73° N. latitude, every one of which receives annually many thousand cubic feet of ice. He named and located them as follows : — Jakobshavn, in 69° 10' N. lat. ; Torsukatak, behind the island of Arvemina, in 69° 50' N. lat. ; Karaiak, in 70° 25' N. lat., a large ice-frith ; Kano'erdluo'suak, in 71° 25' N. lat., a still laro-er ice-frith, which with Karaiak is a ramification of the Bay of Umanak, Upernivik, in 73° N. lat., behind a large group of islands. Ice- friths 12t) According to another authority upon the subject, the region that is richest in glaciers, and presumably therefore the most prolific of icebergs, on the West Coast of Greenland, is in the neighbourhood of the Umanak Fiord ; especially on the Nuo'snak Peninsula, which is situated on the Southern side of the Fiord. A large expanse of this peninsula rises above the snow line. The Northern side of the Fiord is formed of a series of small peninsulas and islands that yield, in relation to their sizes, abundant formations of high ice. The greatest number of glaciers on Nugsuak are located on the Eastern end of the peninsula, a low rocky mass between a line of lakes and the Fiord ; on a strip of the Coast, about 62 miles in length. On the North side of the rocky mass there are no less than 28 glaciers debouching to the sea ; the easternmost of these have no connection with the ice-cap of the plateau and belong to the type of pendant glaciers; so-called because they do not reach the surface of the sea or land, as the case may be. There are three other glaciers of the same type farther West ; and to the West of these are situated five of the greatest glaciers of the ])en!nsula, namely, the Sarkak ; the Little and the Great Umiartorfik ; the Asakak, and the Semiarsut. To the left of the Great Umiartorfik on the rise from the valley there is, moreover, a pendant glacier, which ends iij two sharp tongues. West of Semiarsut, another succession of mighty glaciers begins : Kome, Sarfarfik and nine others, these however are all pendant glaciers, and end at a height above the sea, that increases in each case, westward. The few glaciers on the South side of the peninsula all end inland, descending towards the lakes ; but on the West side of the mainland, in the neighbourhood of the peninsula, there are at least se\en important glaciers, most if not all of which descend to the sea. Other glaciers on the west coast of Greenland from which iceberj^s doubtless are derived are Humboldt {^'lacier, in Peabody Bay, between Smith Sound and Kennedy Channel ; a glacier unnamed and Petowik glacier, on the northern coast of P)affin Bay, between 76° and 76i° N. There are, moreover, on the western shores of Baffin Bay, tmd Davis Strait several iceberg - bearing glaciers such as llice's, in 77° N , on the east coast of ]<]llesmere Land ; two glaciers unnamed on the north side of Glacier Strait, at the entrance to Jones Sats with from one-seventh to one-ninth of its bulk above water ; so that when such a mass is drifting under the influence of wind and current, it is, as a rule, acted upon more by the latter than by the former ; at the same time, wind doubtless plays a more or less important part in steering a drifting berg. The Distribution of Atmospheric Pressure and the Disruption am] Movements of Ice. The centres of many cyclonic depressions that influence the weather conditions in Davis Strait and Baffin 13ay during * Jsforholdene i de arktiske Have, 1911. (The state of the ice in the Arctic Seas, 191].) THE STATE OF THE ICE IN THE ARCTIC SEAS 1911. published by the Danish Meteorological Institute. THE STATE OF THE ICE IN THE ARCTIC SEAS 1911 published by the Danish Meteorological Institute. • The Dissolution oj Icebergs. 131 their progress eastward pass to the south of those regions, and the strong winds and gales associated with these disturbances, blowing as they do from eastward in their northern segment, drive the icebergs that have not left the shelter of the friths on the west coast of Greenland into open water. During the spring the average distribution of atmospheric pressure to the eastward and westward of Davis Strait and Baffin Bay is conducive to the prevalence of northerly or north-easterly winds ; pressure is relatively low over southern Greenland, and a large area of low pressure is situated over the ocean immediately to the south-east or south of Green- land. It is relatively high over the Dominion of Canada between the 80th and 120th meridians of west longitude. These winds act with the prevailing current in carrying bergs, with other ice, to the southward, steering them at the same time towards the western shores of the strait, where, it may be assumed, some of them winter, while others are again driven into the open by westerly gales that are experienced when the centres of cyclonic depressions pass to the north of the strait. Ice that has broken away from the western side would also be carried to the eastward by these westerly gales, and to the southward by the current ; after which, it seems evident, that the assemblage would eiiher clear the land and move into the Atlantic or drift towards the south- west coast of Greenland, and there become hampered by land and other fast ice and imprisoned for the winter. Probably a few icebergs, with some field and pack ice that has collected near the southern coasts of Greenland and Labrador, are released by the early winter gales and the accompanying sea disturbance, so that they resume their drift southward into the Atlantic, and, with other free ice from the north, reach the Transatlantic trade routes as early as January or February, and soon after come under the influence of the warm current from the south. The Dissolution of Icebergs. The time which elapses between the entry of an iceberg into the warm water south of about the 42nd parallel of latitude, and its dissolution doubtless varies considerably, depending on its mass, structure, condition, and the tempera- ture of its environment ; possibly upon other circumstances also ; and it may here be suggested that the rock and other earthy matter which adheres to the parent glacier while carving its way to the sea may so materially affect the 11979 p 132 Icebergs. buoyancy of its offspring that the iceberg founders after the greater portion of its mass has dissolved. The number of bergs that survive their encounter with the Gulf Stream and drift to the southward and eastward are few ; but, as will be shown later, small icebero;s and lars^e masses of ice that must at one time have been huge bergs have been seen as far south as 31° N. latitude, and as far east as 5° 50' W. longitude. Drift of Ice from Greenland Sea. Following its disruption in summer most of the ice in Denmark Strait drifts to the south-westward along the east coast of Greenland in the polar current that follows the coastline round Cape Farewell. This ice has its origin in, and in the neighbourhood either of Spitzbergen or Greenland. That which comes from the former region is chiefly of the nature of great level floes or of field ice ; the drift from the latter region consists for the most part of icebergs, which ground in 60 to 70 fathoms of water, and field ice. Arctic explorers, it is stated*, have found the ice to the north of Spitzbergen, and between that island and Green- land, to be movinar in a body to the south-west. Great quantities of field and hummocky ice pass each year between Spitzbergen and Greenland and between Greenland and Iceland, these waters being almost covered with ice, which greatly impedes navigation or renders it impossible. Drift ice appears around Iceland sometimes as early as January, and lasts until the autumn ; usually there is little in the last four months of the year. Ice off South Coast of Greenland. Polar ice arrives off the south coast of Greenland in May, June, and July ; during the four months November to February none is seen in that locality. According to Dr. Hink, who mentioned that he had good authority for the statement, sealing vessels, which in June visit the ice south- ward and eastward of Cape Farewell, from which headland it extends 50 to 80 miles, find it to be moving westward at the rate of eight to ten miles a day. After rounding Cape Farewell this assemblage of ice continues, with the current, to follow the coast to the north- westward until it reaches about the 59th parallel, when, its * Arctic Pilot, Vol. II., Second Edition. Published by order of the Lords Commissioners of the Admiralty, 1911. To face p. 13Z. )£ PLATE XXI. MOD 1901-1912. Intic and Mediterranean. r» sesa iSMO/44. *m* i/m fofteep tSi- EXTREME LIMITS OF ICEBERGS AND FIELD ICE on and in the neighbourhood of THE GRAND BANKS OF NEWFOUNDLAND DURING THE PERIOD 1901-1912. Prepared from the data given on the Monthly Meteorological Charts of the North Atlantic and Mediterranean. r Ice Limits. 133 course being deflected to the westward, it joins the ice which is being carried southward by the current from Baffin Bay. Ice Frequency in North Western Atlantic, Icebergs and field ice reach the trade routes earlier in some, years than in others. Reports of ice increase, as a rule, from January or Febi'uary to May or June, but in some years the maximum quantity reaches the area fre- quented by shipping as early as April or as late as August, During the ten years period 1903-1912 the maximum number of ice reports received per month at the Meteorological Office through various sources, from steamers eng-aored in North Atlantic trades, related in seveu of these years to the month of May, in two of the years to the month of x\pril, and in the remaining year to the month of July. Field and slob ice usually melt away rapidly after April ; it may beset a vessel, for instance, in the evening, yet have disappeared on the following morning. Ice Limits. The southern and eastern limits of ice in the North- Western Atlantic vary from month to month, and from year to year. The extreme limits of icebergs and of field ice on and in the neighbourhood of the Grand Banks of Newfoundland during the twelve years ended December, 1912, is shown for each month in the maps on Plate XXI, which are supplemented by a statement in Table 8.* The earliest date on which ice of any kind was observed in the North Atlantic each year, from 1903 to 1912 inclu- sive, and was subsequently reported from a reliable source to the Meteorological Office, was 6th March, 9th February, 18th January, 2nd January, 2nd February, 1st January, 24th January, 2nd March, 28th January, and 7th January respectively. Drifting ice may, however, be observed in almost any part of the North Atlantic north of 30° N. latitude, about as far east as the 10th meridian of west longitude on the eastern side of the ocean ; and about as far west as the 75th meridian on the western side, north of 35° N. •"Monthly Meteorological Charts of the North Atlantic and Mediterranean for March, 1913." Issued by the Authority of the Meteorological Committee. 11979 F2 134 Icebergs. Table 3. Extreme Limits, 1901-1912. Year. Fabthest South. Icebergs. Field Ice. Lat. X. I Long. W. Month. Lat. N. Long. W. Month 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1901 to 1912 41 42 30 40 50 40 10 40 10 39 40 39 40 40 41 41 42 50 40 10 35 15 35 15 47 25 43 32 45 Oto 55 ■49 45 47 10 48 30 to 49 30 46 30 to 48 48 46 49 48 Oto July July \ Ma) ch j June ( April ( June I May \ May J June March - June. 49 30 57 44 50 June April July October 44 50 October 41 30 41 50 1 42 10 -I 41 45] 42 o| 43 40 41 50 41 40 42 42 10 40 40 48 38 50 Oto 51 48 Oto 49 50 Oto 51 50 Oto 51 58 49 51 49 52 47 10 47 10 April > April I May [ April I May A pril July May March A pril May May Extreme Limits, 1901-: L912. 1 Farthesi ^ East 1 Year' | Icebergs. . Field Ice. Lat. N. Long. W. Month. Lat N. Long . w. Month, 1901 47 11 42 58 July o ' o ' 1902 46 42 47 July 41 30 48 38 April 1903 43 30 38 May 46 20 43 30 April 19(14 46 41 April ■46 50 47 April 1905 43 30 41 Mar 43 30 44 A]>ril 1906 41 38 June 46 45 30 March 1907 41 50 39 50 May 47 50 45 April 1^08 46 15 42 May 43 50 46 30 July 1909 44 40 20 May 46 30 40 March 1910 44 20 39 20 August 46 45 50 A}iril 1911 43 50 41 July 47 30 45 March 1912 42 3 15 26 May 45 36 42 32 April 1901 42 to 3 15 26 May 46 30 40 March 1912 i -i.S 30. <3> ^ ^ « *S a: ts "a 3S Q. ?1 W ! I ; 1- : z ! UJ : _j 1 UJ I H! 13 ! 2 O -^-'^ to M^s I I ! I ■^•S rt be ^"Z-^ J cyo u a til 1-) S It, ^ ► ■^ 5. 3 O C-iJ ?• - 00 'OOOOC^Ott o& ^ "j; f .Z"5 '■(• a a o -f -=i ^ c r^ - -^ M-=i V -^ jz ^ i^ r «■! «•) C-l M frl W ?0 CO CO « CO w E SE o. _^-= ^1 T3 ^ > -^ »n :c r^ X OS o r- 01 CO T ^4 ^ «9 ^ fM M^mAeoC4r-( t; « 3 -3=4 = fl -,1' 2 o = c e <; «5 c/) •< c, a; ?. „ u. J < E u -<»ieo-^>raet~ooo»0' s * . 8 .. n M ! N ? 2 8 « i « 1 / ' S^ 4:-iii ■^ir>c3t^«C50 ---MW' ooQ oa oo oo oo OONOOCOCOOO o^uooooaooaoooc^o "OS OS *5 S 5 = 2- I =°JS,-3~ -J to v.^ joo-yzcj'j-.uNO «■«*^/a5=>2S2 2S29 -vS 1 aooooossaoooooo* |. =111 =Ia =^i c-ico-^tfttot^oooic -^n O O O O'O'O c o o o o , t^ * o c c: o S x 1.-. o n ! -» I- o •- »» TO X C ~ J3 -• T* 3 nl 5 5^ ^ *-• f* UiDfi a 3 _).(/) O J J O H -♦e«c«7««»tor~«)-o»o -• Icebergs. 135 Phenomenal Drifts of Icebergs and Bergs of Exceptional Height, Phenomenal drifts of ice in the Xorth Atlantic are charted on Plate XXII, which shows also the positions in which a number of icebergs ot exceptional height have been seen. The names of the ships from which the respective bergs or fragments of bergs were sighted are given with a reference number at the head of the chart. E-^ect of Weather Conditions upon Ice Drift into the North Atlantic. The quantity of ice which drifts into the ]^orth Atlantic each year depends for the most part probably upon the meteorological conditions that have obtained in that ocean and in the adjoining arctic region during the previous summer ; also upon the velocity of the Labrador current during the intervening period. When the prevailing distri- bution of atmospheric pressure during the summer is such as to direct the majority of east-moving areas of low pressure into paths that invade high latitudes — that is to say, when pressure is relatively high over the temperate zone of the North Atlantic and relatively low over the adjoining Arctic region — the paths of depressions coming from westward are deflected to the north-eastward, with the result that the weather conditions in the latter region are of a stormy character, and therefore favourable to the disruption of ice. Such a distribution of pressure prevailed during the summer of 1910, when the captains of whaling shi])s experienced strong winds and gales in Davis Strait and Baffin Bay during the greater part of the whaling season. Moreover, in the four months January to April of the following year the Labrador current appears to have been unusually active, and the quantity of ice in the North Atlantic increased from January to May, rapidly to April, when it was exceptionally large, and icebergs had drifted unusually far south and east. The Loss ofs.s. " Titanic." On the 14th of April, 1912, the new White Star liner Titanic struck an iceberg in lat. 41° 16' N., long. 50° 14' W., and foundered. Several other steamers sustained damage through collision with ice. 11979 F 8 ,13,6 Icebergs. Velocity of Labrador Current. The rate at which the cold ice-bearing current from Arctic Seas moves southward has been found to vary near the coast of Labrador from 10 to 36 miles a day, and occasionally, when southerly gales prevail, to cease.* The influence of the wind upon its rate of motion is con- siderable, especially near the coast, and its volume is greatly increased during the spring and summer by the melting of ice in the Arctic Seas. The current attains its maximum velocity after northerly winds ; its average rate near the land is 11 miles a day. Although the general trend of the current is to southward, its course is not infrequently deflected to the eastward or westward by the wind, and it has been stated that icebergs, within the average limits of the stream, have been known to move northward, " without any apparent cause."* Ice in the Southern Hemisphere. Phenomenal Lengths of some Southern Ocean Icebergs. The majority of icebergs in the Southern Hemisphere, as previously mentioned, are tabular in form, and originally formed part of an ice barrier. Many of them are remarkable for their great length : south of the 40th parallel bergs from 5 to 20 miles long are frequently sighted, and the length of a few has been estimated at 50 miles and upwards. Phenomenal Heights of some Southern Ocean Icebergs. j^umbers of southern ocean bergs are equally remarkable in consequence of their great height, for an altitude of 800 feet is by no means uncommon. Some that have been reported since the year 1884 are stated to have reached a height of from 800 to 1,700 feet from water line to summit. 2 he Positions in which phenomenally Long and phenomenally High Bergs were Observed. The positions in which many of these icebergs of phe- nomenal length and of phenomenal height were observed are shown in the maps given on Plates XXIIl and XXIV respectively. The name of the ship from which each berg was observed, the year and month on which it was sighted, the length or the height of the berg regarded as phenomenal, * "Newfoundland and Labrador," No. 73, 1899. Published by the United States Hydrographic Office. Discoveries of Ross. 137 as the case may be, and the number by which it can be identified, is indicated in the table at the head of each map. In addition, the extreme northern limit of ice is defined by a fine line. The barrier of ice capable of producing icebergs of such remarkable dimensions as those referred to must extend for a lonor distance seaward from the land, and border uninter- ruptedly a far reaching coastline. The Discoveries of Ross. Such an ice barrier was first discovered by the eminent explorer, Captain (afterwards Admiral) Sir James Clark Ross, R.N., F.K..S., who, with H.M. ships Erebus and Terror under his command, visited the Antarctic in the years 1841 and 1842. Victoria Land. Ross discovered a continent south of 70° 30' S. latitude, west of 71° E. longitude, to which he gave the name Victoria Land ; and subsequently sailed more than 400 miles along an ice barrier, the height of which he estimated at about 150 feet to 200 feet, to the eastward of an island which bears his name, the eastern extremity of which is situated in 77° 29' S. ; 169° 32' E. The Great Ice Barrier. In his narrative of this Expedition of Scientific Research to the Antarctic Reo:ions * Sir James Ross thus records his first impressions of the Great Icy Barrier^ as he termed it :— As we approached the land (Ross Island) under all studding sails, we perceived a low white line extending from its eastern extreme point as far as the eye could discern to the eastward It presented an extraordinary appearance, gradually increasing in height, as we got nearer to it, and proving at length to be a perpendicular clifiC of ice, between one hundred and fifty and two hundred feet above the level of the sea, perfectly flat and level at the top, and without any fissures or promontories on its even seaward face. What was beyond it we could not imagine ; for being much higher than our mast-head we could not see anything except the summit of a lofty range of mountains extending to the southward as far as the seventy- ninth degree of latitude. ****** If there be land to the southward, it must be very remote or of much less elevation than any other part of the coast we have seen, or it would have appeared above the barrier. • " A Voyage of Discovery and Research in the Southern and Antarctic Regions, during the years 18;59-4:3." By Captain Sir James Clark Rose, R.N., Knt., D.C.L., Oxon., F.R.S., &c. 11979 F 4 138 Icebergs. Ross on the Disruption of Ice Barriers. Boss, referring to the disruption of ice-barriers, expresses the opinion that in the winter, when the temperature of the air is probably forty or fifty degrees below zero, and the sea from twenty-eight to thirty degrees above, the unequal expansion of the mass exposed to so great a difference of temperature cannot fail to produce the separation of large portions of the barrier, which are driven to the north by the prevailing wind. In arctic regions, he adds, he had often witnessed the astonishinoc effects of a sudden chans^e of temperature during the winter season, causing great rents and fissures of many miles extent. United States Expedition under Wilkes. AVhile engaged in an exploring and surveying expedition during the years 1838 to 1842 inclusive. Captain Charles Wilkes, U.S.N., with the United States ships Vinceiuies, Porpoise, and Peacock under his command, visited the Antarctic in 1840 prior to the aiTival of Ross in south polar regions. Wilkes discovered land and encountered ice barriers in several localities between the 67th and 64th parallels of south latitude, and the 160th and 76th meridians of east longitude. In some localities the barrier extended as far as the eye could reach, and along one stretch of it where the land appeared behind his ship sailed for more than 50 miles, the obstruction presenting the appearance of a straight and perpendicular wall of ice, the altitude of which he estimated at from 150 to 250 feet from sea level to summit. The icebergs afloat near the barrier were from a quarter of a mile to five miles in length. Wilkes on the Formation of Ice Islands. Captain Wilkes mentions in his account of the voyage* that the formation of ice islands, a name given to the larger icebergs by early navigators of far Southern Seas, claimed much of his attention ; and he arrived at the conclusion that these masses of ice increase in bulk after they have separated D:om the parent barrier by means of the accumulation of rain, snow, and moisture deposited in fogs. It might, he * " Narrative of the United States Exploring Expedition during the years 1838, 1839, 1840, 1841, and 1842." Vol. II. By Charles Wilkes, U.S.N,, Commander of the Expedition, 1845. Birth of an Iceberg. 139 thought, be safely asserted that icebergs are at all times increasing in size while in antarctic seas, because the days are few, according to his experience, in which precipitation in some form, does not take place in those high latitudes. Observations on the temperature of the sea, taken by the expedition, led him to believe that these ice islands undergo little change by the process of melting before they reach the latitude of 60°. Borchgrevink's Expedition in " Southern Cross.^^ Mr. C, E. Borchgrevink, Kt. St. Olaf, the leader of a British Antarctic Expedition undertaken in the ship Southern Cross durino; the years 1898-1900, was the next explorer to visit South Victoria Land and examine the great ice barrier in the Ross Sea. In the book he published* soon after the return of the expedition, Borchgrevink alludes to the formation of antarctic icebergs, and expresses the opinion that they are either of glacial or barrier origin. Tlie Birth of an Iceberg. While collecting specimens of rocks and vegetation at the foot of Mount Terror, Ross Island, he witnessed the calving of an iceberg from a glacier, situated immediately to the west of the little beach on which lie and the captain of the Southern Cross had been landed, and thus he describes the incident : — With a deafening roar a huge body of ice plunged into the sea, and a white cloud of water and snow hid everything before our eyes. The plunge of this mass of ice, weighing, as he affirms, milHons of tons, raised a wave which he judged to have been from lo to 20 feet in height ; and had it not been for a pro- jecting ice slope, which broke the force of the wave as it advanced up the beach, he and his companion would, he believed, have been smashed by it against the rock to which they were both clinging. Swedish Expedition under Nordenskiiild. Dr. Otto Nordenskiold, the leader of the Swedish Ant- arctic Expedition, in the ship Antarctic^ during the years • " First on the Antarctic Continent : being an Account of the British Antarctic Expedition, 1S98-11)00." By C. E. Borchgrevink, F.R.G.S., Commander of the Expedition. 140 Icebergs. 1902-1903, discovered, while cruising in the neighbourhood of Terre Louis Philippe in about 65° S. latitude, 57° W. longitude, a perpendicular barrier of ice, which appeared to rise 130 feet above the water, stretching as far as the eye could reach. Scott on the Origin of Icebergs. In the year 1902, the ship Discovery^ of the National Antarctic Expedition, under the command of the heroic seaman the late Captain Robert Falcon Scott, C.Y.O., R.N., cruised along the full extent of the great ice barrier in the Ross Sea, which was closely inspected, prior to and after the discovery of land to the eastward of the 160th meridian of west longitude, to which he gave the name King Edward VII. Land. In his graphic narrative of this expedition,* Scott says that although the barrier was not more than 70 feet high when they started to cruise along it on the morning of the 23rd January, 1902, by the evening of the 23rd it had risen to a height of 240 feet, and subsequently the Discovery passed close to a sheer wall of ice 280 feet high. With reference to the origin of the icebergs that were observed in the Ross Sea, Captain Scott states that the main supply is derived from this barrier and from King Edward VII. Land, the latter presumably from a glacier ; for, he says, the glaciers in Victoria Land are in a condition of stagnation, and nearly all the bergs that were met with along the coast came from the east. He mentions that from Cape Adair (the northernmost point of South Victoria Land) to Cape Crozier (Ross Island), only two icefloes were observed that were capable of giving off a clean tabular berg of any dimensions. The positions of the localities referred to in the foregoing are shown in the map of the Antarctic on Plate XXV, which is also intended to indicate, in a general manner, the area over which, according to our present imperfect knowledge, the polar ice-cap extends. In his remarks upon the size and form of icebergs in southern seas, he alludes to the early southern voyagers, who (lo7:ibtle'«s had a knowledge of the bergs of the Northern Hemisphere, but in the Southern Ocean met with masses of ice incomparably larger than anything they had seen in the north ; and that to these they gave the name Ice Islands ; a * " The Voyage of the Discovery.'' By Captain Robert F, Scott, C.V.O., R.N. London : Smith, Elder & Co., 1905. To face p. J40 PLATE XXV To /«« P 'f"- ^^^^ POLARREG/0/Vs PLATE XXV. V Yearly Variation. 141 name which even Cook preserves in describing the long tabular iceberg which is typical of the southern regions. Except, Scott remarks, in cases where they have suffered denudation or have lost stability and capsized, the shape of Antarctic icebergs is uniform : they are flat-topped and wall- sided, and appear to have broken away from some huge sheet of ice. Calving of Icebergs from High Clifs. The Discovery passed innumerable bergs aground on the shoals off King Edward YII. Land, some very large ; one or two small bero;s were seen in the act of calvino- from the high cliffs in the neighbourhood, but none were observed separating from the ice barrier. Icebergs are rarely seen in the southern seas between the meridians of 130° E. and 170° W. within the parallels of lati- tude frequented by shipping, and during the seven months April to October that zone is practically ice-free. Drifting ice in temperate latitudes of the Southern Hemi- sphere is exclusively in the form of bergs. Ice Frequency in Southern Ocean. The number ot icebergs observed yearly in those latitudes varies consideraV)ly ; the number reported each month during each of the twenty-eight years 1885 to 1912 is set forth in Table 4 ; the total number of bergs reported in each of these years is also stated, as well as the total number during the whole period. It will be noiiced that in each of the seven years 1885 to 1S91, and again in the years 1898, 1899, and 1900 th number of icebergs reported was small ; whereas in the years 1892, 1908 the number was large, and in the years 1893, 1906 excessively large. Yearly Variation in Ice Frequency . This marked ^•ariation in ice frequency in southern seas may perhaps be due to causes similar to those suggested to explain the yearly xariation in the quantity of ice which drifts into the North VV^estern Atlantic ; it may depend chiefly upon the meteorological conditions over the Southern Ocean and the adjoining polar regions which have obtained during the previous summer, conjointly with the strength subsequently of the polar currents. The map of the South Polar Regions on Plate XXV is supplemented by a map on Plate XXVI, on which the distribution of ice is shown in detail. 142 Icebergs. SS w <41 « ^ 02 g « o ^ ^ o «i o rJ5: Cm '^ O "t* s^ >< o P4 ^ p^ (4-1 1»S u « >i a 1 5^ s- t<< 1885 to 1912. ^ lO '■X)CC ^ 1— 1 1—1 .— 1 coc^ I— 1 Oi 1— 1 r-^ I-l CTl CO rH CO lO I— 1 1—1 1—1 CO '^ CO as Oi 1-1 I-l rH CO iO rH I— 1 1— < I— ( 1— 1 (M(MO OCOCO CO !MCO (M i-l (MO rji t-CO CO COCO t^ t^ -^ 1—1 t^ I— 1 ■^ »o t^ 1— ( COCO T-l t^O X) I— 1 (M (M'* I-l 00 OS OSiO-^ I— 1 lO (M i-( (M (M CO OOQO 1-H CO c6 CC CO CO COiH '^ t-O I— ( CO CO ^ CO 1—1 1—1 I-l 1—1 Oi coo 0(M0 .— 1 I-l -i^ 1-1 Oico CO CO ^ c» .-1 PLATE XXVT The northern liimts ot ice throughout the year lie between the two dotted lines that are drawn through ail meridians. These limits are referred to observations that date back to the year 1772. The symbols employed to distinguish between the ice reports of the respective months are as follow : — WINTER. Bergs. Pack July A VWW^ August a -ru-LTL S^tonber o-o-o-o- atmos])here. At the Meteorological Office the practice of Tofacop- 142 .^.^ POLAR 'OJ U^^.^ TLATE XXVI The northern hmits ot ice throughout the year lie between the two dotted lines that are drawn through all meridians. These limits are referred to observations that date back to the year 1772, The symbols employed to distinguish between the ice reports of the respective months are as follow : — The general distribu- tion of ice, which can be at a glance, relates to the fifteen years 1902 — 1916. The Baromtter. 143 CHAPTER XL Meteorological Instruments. The Barometer. This is the technical name for the weather-glass, and is often called simply the glass. It is an instrument used for ascertaining the pressure of the atmosphere, or, to state the facts more correctly, for determining the pressure of the atmosphere by measuring the height of a column of mercury supported by it ; and it is so named from two Greek words — bar OS, weight, and metron, a measure. Torricellis Experiment. If one end of a glass tube, about 33 inches long, be dipped into a bowl of mercury and an air pump applied to the other end a column of mercury will gradually rise in the tube as the pump is worked. It is pushed up by the air pressing on the surface of the mercury, and when the ope- ration is complete, that is to say, when the mercury ceases to rise in the tube, it will stand at about 30 inches. We thus have a column of mercury supported by the pressure of the air. The space left between the top of the column of mercury and the top of the tube will be a vacuum. In the barometer this is as nearly a perfect vacuum as can be produced. It is known as the Torricellian vacuum, because the original form of the apparatus was made in the year 1643 by an Italian named Torricelli. As the mercury in the tube is held up by the pressure of the atmosphere on the mercury in the bowl — in other words, as the weight of the column of mercury exactly balances the pressure of the atmosphere — any variations in the pressure of the latter will be shown by corresponding variations in the height of the mercury column. The apparatus used in making the experiment referred to is in truth a simple form of the Cistern Barometer, and would require onl}' the addition of a fixed scale by which the height of the mercury column could be measured to make it serviceable for ascertaining the pressure of the atmosphere. In this book the scale of ihe barometer is generally assumed to be a scale of inches of mercury which, when duly corrected for the difference of temperature from the freezing point of water and the difference of latitude from latitude 45°, gives a measure of the pressure of the atmosphere. At the Meteorological Office the practice of 144 Meteorological Instruments. expressing the pressure of the atmosphere in so-called absolute units (centibars or millibars) instead of inches of mercury is now being introduced. The dial of the sea barometer shown in Fig. 21, p. 156, gives the relation of centibars to inches. Action of Mercury Barometer analogous to that of Pump. Mercury being thirteen and a half times as heavy as water, the latter could be raised from a well by means of an ordinary pump to a height of from 32 to 34 feet, provided a perfect vacuum could be obtained,* because a column of water 34 feet long has the same weight as a column of the atmosphere of similar transverse area. The atmosphere, pressing on the surface of the water in the well, forces the latter up the pipe when the piston is raised and the pressure from above thus removed ; but this action ceases as soon as the weight of the column of water in the pipe is equal to the pressure of the atmosphere. Water can be raised by a pump to a somewhat higher level when the pressure of the atmosphere is abnormally great than when it is about normal and vice versa ; in like manner the column of mercury in a barometer tube will rise as the pressure of the atmosphere increases and fall when it diminishes ; and while pressure remains unchanged the mercury column will remain at rest. Thus, any variations in atmospheric pressure will be shown by corresponding variations in the height of the mercury. The instrument known as the Fishery Barometer^ a weather-glass lent by the Meteorological Office for the use of fishing communities, consists of a tube of large bore con- nected with a cistern, both containing pure mercury, the whole being mounted in a solid oak frame to which is attached a fixed scale, marked in inches and tenths of an inch from 26 to 31 inches ; and a sliding scale, or vernier, so called after its inventor, Pierre Vernier (a.d. 1630). In some instruments two verniers are fitted, one on each side of the scale. To the frame of the instrument there is also attached a thermometer, graduated from 0° to 120° Fahr. The vernier is moved by a rack and pinion, the latter fitted with a milled head for convenience of adjustment. The thermometer connected with the instrument, called the attached thermometer^ is required for the purpose of ascertain- ing what allowance should be made for temperature when cor- recting the barometer readings. The column of mercury in * In practice the height is not more than 27 feet. Action of Mercury Barometer. 145 the tube expands when heated, and a longer column of the lighter liquid is required ; it contracts, and a shorter column is required when cooled ; therefore, in order to render com- parable readings of the barometer with those of instruments in other localities taken at the same time, it is necessary to Know the temperature of the mercury in order that the readings may be reduced to a standard temperature. A temperature of S^*-* Fahr. has been adopted universally as the standard to which all barometer reading^s should be reduced. A correction is also required to compensate for the varia- tions of temperature of the brass scale when the scale is made of that metal. Where great accuracy is required it is necessary to apply other corrections to barometer readings than those already mentioned ; one of these is a correction for index error, an error in the scale introduced in the process of graduation. Cistern barometers are subject also to two other sources of error, one of these arising from what is called capillarity and the other from what is known as the error of capacity. The error arising from capillarity is due to the fact that mercury has no such attraction for glass as water has ; it does not wet it, and has therefore to overcome the resistance of the glass, and is in consequence somewhat depressed in the tube. The error of capacity results from changes of level in the cistern ; for wheu the mercury rises in che tube it leaves less in the cistern, thus lowering the level of the mercury in the cistern ; similarly, when it falls in the tube it raises the level of the mercury in this cistern. The capillarity correction, for a barometer, which is always additive, is the same for all readings of the instrument to which it applies. Not infrequently the correction for index error, combined with that for capillarity, is expressed in one number, which is engraved on the scale of the barometer. In a type of instrument known as Fortiii's Barometer^ in order to get rid of the error of capacity, the surface of the mercury in the cistern is temporarily adjusted to a level which corresponds with the zero of the scale. This is effected by raising, or lowering the base of the cistern, which is flexible, by means of a screw. In the Kew pattern of barometer the error due to capillarity, and the error of capacity are both annulled in graduating the scale ; allowance being made, when the scale is laid otf, by comparison with a standard instrument, for the depression of the mercury column in the tube through friction, and for 146 Meteorological Instruments. the differences of level in the cistern created by the rise and fall of the mercury in the tube ; the allowance made for the latter depending upon the relative diameter of the tube and the cistern. When barometer readings from different parts of the world have to be compared by plotting on a chart, and an accuracy of 100th of an inch is required, a correction for gravity also is now applied, because the earth being a spheroid, the force of gravity varies with latitude, and places at the equator are at a greater distance from the earth's centre than places at the poles. Barometer readings, therefore, are reduced to standard latitude, for which the parallels of 45° N. and 45° S. have been adopted. The corrections required in this connexion are as follow : — Corrections for reducing Barometric Readings to Standard Gravity in Latitude 45°. Correction. Corre ctiou. Lat. Lat. N. or S. N. or S. At 27 inches. At 30 inches. At 27 inches. At 30 inches. In. In. In. In. -•070 -•078 28 -•039 - •043 1 •070 •078 29 •037 •041 2 •070 •078 30 •035 •039 3 •070 •077 31 •033 •036 4 •069 •077 32 •031 •034 5 •069 •077 33 •028 •o;')2 6 •068 •076 34 •026 •029 7 •068 •075 35 •024 •027 8 •067 •075 36 •022 •024 9 •067 •074 37 •019 •021 10 •066 •073 38 •017 •019 11 •065 •072 39 •015 •016 12 •064 •071 40 •012 •013 13 •063 •070 41 •010 -Oil 14 •062 •069 42 •007 •008 15 •061 •067 43 •005 •005 16 •059 •066 44 -•002 -•003 17 •058 •064 45 ± + • 18 •057 •063 46 + ^002 + -003 19 •055 •061 47 •005 •005 20 •054 •060 48 •007 •008 21 •052 •058 49 •OiO •Oil 22 •050 •056 50 •012 •013 23 •049 •054 51 •015 •016 24 •047 •052 52 •017 •019 25 •045 •050 53 •019 •021 26 •043 •048 54 •022 •024 27 •041 •046 55 •024 •027 Correction of Barometric Readings. 147 Corrections for reducing Barometric Readings to Standard Gravity in Latitude 45° — cont. Correction. Correction. Lat. Lat. N. or S. N. or S. At 27 inches. At 30 inches. At 27 inches. At 30 inches. In. In. In. In. 56 + -026 + ^029 74 + ^059 + -066 57 •028 •032 75 •061 •067 58 •031 •034 76 •062 •069 59 •033 •036 77 •063 •070 60 •035 •039 78 •064 •071 61 •037 •041 79 •065 •072 62 •039 •043 80 •066 •073 63 •041 •046 81 •067 •074 64 •043 •048 82 •067 •075 65 •045 •050 83 •068 •075 66 •047 •052 84 •068 •076 67 •049 •054 85 •069 •077 68 •050 •056 86 •069 •077 69 •052 •058 87 •070 •077 70 •054 •060 88 •070 •078 71 •055 •061 89 •070 •078 72 •057 •063 90 + ^070 + •078 73 •058 •064 For the correction of barometer readings in C.G.S. units, see pages 11-14. The divisions oi the fixed scale on a Fishery Barometer are each 0*1 of an inch. Eleven of these are taken as the length of the vernier, which is, therefore, 1"1 inch long. The vernier is divided into 10 equal parts, as seen in Fig. 12 from 1 to 10 ; thus each division is 0*11 inch in length, so that the difference in length between a division of the vernier and a division of the scale is O^ll inch — '10 inch = ^01 inch, and this arrangement enables the observer to ascertain the height of the barometer to the nearest 100th of an inch. This is the form of vernier originally proposed by its inventor, and it is one that is commonly attached to the scale of a Fishery Barometer {see Fig. 12). It has an advantage over other forms with respect to the clearness of its divisions, but it has to be numbered backwards, or in a contrary direction to the numbering of the fixed scale of the instrument. To obtain a reading of this class of mercury barometer, the upper edge of the vernier should be brought into coincidence with the to]) of the mercury. Should the upper edge of the vernier happen to be in the same straight line as a division of the scale, then that division points out directly the reading to be recorded. If, however, 148 Meteorological Instruments. the upper end of the vernier fall between two adjacent divisions of the scale, then the reading has to be obtained by use of the vernier. The observer should, therefore, look down the vernier for a division which coincides with one on the scale, and when found this division will indicate the reading to be recorded. For instance, by referring to the right-hand vernier in the accompanying illustration (Fig. 12) Fig. 12. the upper end is seen to lie between 29'7 inches and 29"8 inches hence the barometer reading so indicated is above 29"7 inches, but below 29*8 inches. As it is necessary to determine the reading to the nearest 100th of an inch, the observer must look down the vernier, when h will be seen that the line marked 6 on the vernier coincides with a division of the scale and gives the reading in detail as follows : — Reading on scale Reading on vernier ... Actual barometer reading 29-70 •06 29-76 inchfts. Using the left-hand vernier a similar method is adopted. By referring to the illustration it will be seen that the upper end of the vernier lies between 29-6 and 29*7 on the scale, so thai the barometer reading for that position of the vernier is slightly above 2.^-6 inches. Looking down the vernier it is found that the line marked 3 on the Kew Pattern Barometer. 149 vernier coincides with a division on the scale, and this signifies tha^ (he upper end of the vernier is four hundredths of an inch above the 29*6 inch division of the scale. The reading set out in detail is : — Reading on scale 29-60 Reading on vernier ... ... '04 Actual barometer reading 29*64 inches. In a Fishery Barometer of another pattern (M.O. 187) the divisions of the fixed scale are each O'O inch. Twenty-six of these divisions are taken as the length of the vernier, which is therefore 26 x '05 = 1*3 inches long. This length is divided into 25 equal parts, as seen in Fig. 13, and thus each division of the vernier is 1'3 -=r 25 = 052 inch long ; so that the difference in length between a division of the vernier and one of the scale is '052 inch — "050 inch = "002 inch. Looking down the vernier it is seen that the second line below the figure 1 lies evenly with a line on the scale. The figure 4 indicates '04, and the second subdivision below it "004, thus we have : — Reading on scale ... ... 30"200 f 040 - \ -004 Reading on vernier Actual barometer reading 30-244 inches. Fig. 13. Fig. 14. Kew Pattern Barometer. 150 Meteorological Instruments. Kew Pattern Barometer. At Climatological and Tele- graph Reporting Stations in the British Isles, the Kew Pattern Station Barometer is in general This mercury barometer use (Fig. 14) consists of a glass tube closed at one end, which is filled with pure mercury before being connected with the cistern, all air being carefully excluded ; the tube is then inverted and its open end immersed in the small cistern, also containing mercury, in such a manner as to preclude the pos- sibility of air entering the tube. As air, even a minute quantity, is detrimental to the satisfactory working of the instrument, the utmost care is taken to exclude it. A small hole in the upper part of the cistern H admits access to the air, a washer of leather preventing the escape of mercury from the cistern, while permitting the atmosphere to exercise pres- sure upon it. About eight or nine inches from the top of the tube the bore is small downwards, and at the termination of this con- tracted portion of the tube, or in some other part of it, there is an air trap. In Fig. lo a section of the greater part of a tube is shown, in which C is the con- tracted portion of the tube, A the air trap. The barometer is suspended in gimbals, and this method secures that the scale is vertical when the instrument is hanging quite freely. When in position, the top of the barometer should be at such a height that the observer can o read the scale comfortably while standing upright, and, if possible, it should be so placed that the light comes slant- wise from somewhere behind the observer. ! Fig. 15. Barometer Tube, not drawn to scale. Keic Pattern Barometer. 151 In handling a barometer, it should be borne in mind that it is a delicate instrument. Should it be necessary to move the barometer, it should first be inclined slowly, so as to allow the mercury to How gently to the top of the tube. Then, the tube being full, the instrument may be transported with safety in a horizontal or inverted position (cistern uppermost), but care must be taken not to subject it to concussion. To mount the instrument, lift it carefully out of its case and slip the hinged part of the suspension arm into the socket. Take care that the screws which secure the instrument in its gimbals are screwed home, otherwise it may slip through its supports. No observation should be taken until at least two hours have elapsed after mounting the barometer, in order that the mercury may have time to acquire the temperature of the air. The tube of some station barometers is so much contracted to avoid "pumping" that a considerable time is required for the mercury to reach its level after being first set up. In standard barometers of the Kew pattern which are graduated in inches, the fixed scale is divided into inches, tenths and half-tenths, each of the last named therefore represent "050 of an inch. Twenty-five divisions of the vernier are made to coincide with twenty-four of the smallest divisions of the fixed scale ; a space on the scale, therefore, is larger than a space on the vernier by the twenty-fifth part of '050, that is to say, by '002 of an inch. In standard barometers which are graduated in millibars, the fixed scale is divided into centibars and millibars. The vernier lor this scale covers thirty-nine divisions of the fixed scale, and therefore is less by one millibar-division than the length of forty millibars on the fixed scale. The vernier of this instrument, like that of the Fishery Barometer, is moved by a rack and pinion, the latter carrying a milled head. The vernier is set for reading by turning the milled head until the front edge of the vernier is brought into alignment with the uppermost part of the convex surface ot the mercury column, and the edge of the sliding piece at the back of the instrument. Great care should be taken to acquire the habit of setting the vernier when the eye is exactly on a level with the top of the mercury, thaj; is, with the line of sight at right angles to the tube, which, while the observation is being made, should hang freely in a vertical position, because any inclination will cause the column to rise in the tube. The sliding piece at the back of the instrument is provided in ortler to aid the observer by showing him when his eye is at the same level as the top of the mercury column, for should the observer's eye be too high or too low when setting the vernier, an error in the reading will result ; such errors 152 Meteorological Instruments. are known as errors of ixirallax.^ graphic representations of which are given in the accompanying diagram (Fig. 16). 30 ^ccuR^"^^ p.£ ^o\ng .--Z950-' ACCURATE READING 29 26- '^^CCo^^ '/\7-£: '^Cao '^'G-, '13 Zt EYE TOO HIGH EYE AT CORRECT LEVEU EYE TOO LOW Fig. 16. The mode of reading; off the heioht can be understood best by the aid of the diagrams, Figs. 16 and 17, in which A B repre- sents part of the scale, and C D the vernier, the lower edge of which, D, has been brought to coincide with the top of the mercury column. B Fig. 17. Fig. 18. '^First. — Xote the value of the scale division next below the zero division on the vernier marked D in Figs. !7 and 18. The scale is graduated in millibars^ and numerical values in centibars are figured along it (10 millibars = 1 centil)ar). Kew Pattern Barometer. 153 In order to assist the eye when determining the value of a division, the millibar graduations are of unequal length. In Fig. 17, D is supposed to be in the same straight line with the fifth (the long) division above the scale division num- bered 98, in other words with the graduation 985 millibars. In Fig, 18 the graduation next below D is the second above the graduation numbered 101 ; its value is therefore 1012 millibars. Second. — Look along the vernier for a division which is in one and the same straisfht line with a scale division. The yalue of this division on the vernier gives the decimal place. In Fig, 17 the vernier division is exactly coincident with a scale division ; the reading of the barometer is therefore 985*0. In Fig. 18 the vernier division 7 is exactly opposite a scale division ; the barometer reading is therefore 1012*7. If the vernier has not been shifted between two observations^ it is advisable to check the previous reading before proceeding to a fresh setting. The mode of reading: off the heio^ht. when the vernier has been set and the instrument is graduated in inches, may be learned from a study of the diagrams, Figs. 19 and 20 (p. 154), in which A B represents part of the scale, and C D the rernier, the lower edge of which, D, has been brought to coincide with the top of the mercury column. The scale is readily understood : the bottom line at B represents 29*000 inches ; the first line or division above is 29*050 ; the second line or division 29*100, and so on. The scale division just below D must first be noted ; it is then necessary to ascertain which division on the ▼ernier is on a line exactly with a division on the scale. In Fig. 19 the lower edge of the vernier, D, is represented in exact coincidence with the scale division 29*5 ; the barometer therefore reads, 29*500. It will be seen that the top of the vernier C again coincides with a line on the scale, and that all of the other divisions of the vernier are more or less separated from the divisions of the scale. As one division of the vernier is *002 inch smaller than one division of the scale, consequently with the vernier in the position shown the division a is '002 inch below the nearest line, 2", of the scale. If, therefore, the vernier be moved upward, so as to place a in line with z^ the edge D would be raised *00- inch, and the reading would be 29*502, for this would be the height of D on the scale. In like manner it can be seen that b on the vernier is •004 inch below the line next above it on the scale ; c, "OOG inch below that next above it ; d, "008 inch from that next above it ; and 1, on the vernier, 154 Meteorological Instruments. is '010 inch below y on the scale. For this reason, if the lines ft, c, west temperature of a mixture of ice and salt. The examination of the expansion ot gases shows that if air or any other of the permanent gases were to go on con- tiacting at the same rate with cooling as they do at ordinary 160 Meteorological Instruments. temperatures tliey would have no volume and cease to exert any pressure at a temperature of about 459° below zero, Fahrenheit. This temperature is found to be nearly identical with the temperature computed by Lord Kelvin as that to which the temperature of a working substance must be reduced in order to get the full equivalent in work of heat supplied to it. This latter temperature, which we may take as — 45 9 "4° Fahr., is known as the absolute zero. On Fahrenheit's scale it will be understood then that the freezing point is 32°, and the boiling point, being 180° above the freezing point, is 212°. The sub- division of the scale between the two iixed points is, of course, entirely conventional ; and there are two other methods of graduation employed. Farly in the eighteenth century an eminent French physicist, called Reaumur, invented the thermometer scale which bears his name. The space on the scale of Reaumur's thermometer between the freezing and boiling points is divided into 80 equal parts ; the freezing point is zero, the boiling point 80°. This system of graduation has in recent times fallen largely into disuse. The thermometer in general use throughout Continental Europe was originated in 1 742 by Anders Celsius, a professor of astronomy in [Jpsala. This is the Centigrade thermometer, so called because of the division of its scale into 100 equal parts ; the freezing point being zero, and the boiling point 100°. For some years after a thermometer has been made the glass of the bulb contracts, and in course of time the con- traction may give rise to an error of as much as a degree, or even more. It is therefore necessary to compare the readings of a thermometer with those of a standard instru- ment from time to time, in order to ascertain the correction, if any, to be applied to readings of the former. Th e Thermograph . A self-recording thermometer, or thermograph, is now largely employed for obtaining a continuous record of temperature, and if studied in connexion with the record of a barograph for the same period, will demonstrate the close relation existing between the fluctuations in temperature and pressure respectively. The instrument will be found, after the observer has had some experience with it, a valuable aid in foretelling changes in weather conditions. For instance : a marked rise in temperature, detected by a glance at the thermograph, if Thermograph. 161 associated with a shift of wind to an equatorial (in this country, southerly) quarter, will frequently give warning of the approach of an atmospheric disturbance before the baro- meter has commenced to fall. In most thermographs the thermometer consists of a slightly carved metal tube filled with spirit. One end of this tube is fixed rigidly to a bracket on the frame of the instrument, while the other is attached to the system of levers which actuates the recording pen. Thermographs for meteorological use must be well pro- tected from rain and sun, at the same time, of course, they must be exposed out of doors, hence it is necessary to clean and oil their bearings from time to time. The instrument may be set by comparing its indications with the reading of a standard mercury thermometer, placed beside it, and adjusting the pen accordingly. The setting should be effected only when the temperature is constant or changing slowly, and when the pen is near the middle of its range. As the metal tube containing spirit, which con- stitutes the thermometer, is in thermal contact vvith other parts or the instrument as well as its frame, all of which take an appreciable time to alter in temperature, the instru- ment is apt to respond somewhat sluggishly to rapid changes of temperature. The readings of the thermograph require frequent checking by comparing them with those of a standard thermometer. The Hygrometer. For measuring the amount of water vapour in the atmos- phere an instrument is used which is called a hygrometer (from two Greek words : hygros, wet ; and 7netro?i, a measure). Among a large variety of hygrometers, differing in form and construction, the one best adapted for general use consists of a pair of delicate thermometers placed side by side. Over the bulb of one of these instruments a single thickness of cambric or fine muslin is secured, and this covering is kept damp by means of a few strands of cotton wick, the bight of which is hitched round it, while the ends are nnmersed in water, contained in a small receptacle placed near the bulb, but not under it (Fig. 24). The water is conveyed to the muslin by capillary attraction, and, even though the temperature of the air should be below the freezing point of fresh water (273a, 32° Fahr.), the readings of the wet bulb will be trust- worthy so long as there is a coating of ice on the muslin, for 162 Meteorological Instruments. evaporation takes place from ice as well as from water, A thermometer fitted in the manner described is called a wet-bulb thermometer. The other in- strument is in all respects an ordinary thermometer, which without the muslin covering is called the dry-bulb thermo- meter. This apparatus, known generally as a wet and dry-bulb hygrometer, is also called a psychrometer. It is essential that the two instru- ments, while freely exposed to the air, should be protected from sun and rain, and from the effects of radiation. It is usual, therefore, to suspend them in a specially constructed screen, called a Stevenson Screen, the sides of which are louvred (Fig. 25). Fig. 24. Arrangement of the Thermometers in the Screen.* In arranging the thermometers in the screen the following points must be borne in mind : — (1.) There should be a space of at least three inches between the bulbs of the thermometers and the top, bottom or sides of the screen. (2.) The thermometers should be so arranged that all parts of their scales can be read without the necessity for moving any one of them. (3.) The maximum and minimum thermometers should be clamped down so that strong winds cannot shake them, as jolting often leads to displacement of the indexes. The instruments require to be moved once a day for setting, and hence cannot be screwed in position. A suitable arrangement is shown in Fig. 25. The action of the wet bulb is as follows : — The temperature of the bulb is lowered by evaporation from the moistened muslin in the same manner as the temperature of the water in a canvas or earthenware water cooler is lowered by the evaporation of moisture percolating through the canvas or earthenware, as the case may be. When the air is dry evaporation proceeds rapidly, and the difference between the wet and dry bulb thermometer is relatively large ; when the air is damp evaporation takes place slowly, and the difference referred to is small. When the air is completely saturated with vapour the two thermometers, corrected for instru- mental error, will indicate exactly the same temperature. * From the " Observers' Handbook." lication No. 191. Meteorological Office Pub- Uygrometer 16^ ;r/ -'■;. Fig. 2r». Stevenson Screen with Thermometers. The difference between the readings of tlie wet and dry bulb Iherniometers is occasionally considerable, and the depression ot the wet bulb, as this difference due to evapora- tion is termed, may amount to as much as 20° Fahr. in our islands, and to more than 1^0° in some parts of the world. The difference is usuidly less near the sea than it is inland. 119711 G 164 Meteorological Instruments. The muslin and wick for the wet-bulb thermometer should be well washed before being applied, and in order to secure trustworthy results from readings of the instrument these should be changed or cleansed frequently — say, every two weeks — for accuracy depends much on the care taken to ensure cleanliness. Rain or distilled water only should be used for moistening the muslin, because river and spring water leave a deposit of lime on the bulb, and this impairs its sensitiveness. Even rain water is not entirely free from impurities, containing as it does alkalis, salts, &c., which in time form a thin coating on the bulb. Should any encrustation be found on the bulb when the muslin is removed it should be carefully scraped off with a sharp pen- knife. By means of observations of the dry and wet-bulb thermo- meter the following can be deduced : — The dew point ; the elastic force of aqueous vapour, or vapour pressure ; and the relative humidity. In connexion with any investigation relating to climate, it is of great importance to know the highest temperature which occurs during the day and the lowest temperature during the night. For this purpose specially constructed self -registering thermometers have to be used, for it is only by continuous observations that such records can be obtained by means of an ordinary thermometer, and this will be found to be impracticable. The instruments used for registering the highest and lowest temperatures are called maximum and minimum thermometers (from the Latin onaxim-um, greatest ; and minimum, least). The Maximum Tliermometer. The form of instrument commonly used is a mercury thermometer, the bore of which is much constricted close to the bulb (Fig. 2Q). When the mercury Fig. 2Q. Maximum Thermometer. expands with a rise of temperature it is forced past this constriction u]j the stem, but when it contracts with a Maximum and Minimum Thermometers. 165 fall of temperature that portion of it which is within the stem is prevented, by the constriction, from returning into the bulb and remains in the position to which it was. forced, thus registering the highest temperature attained {see Fig. 26). To set the instrument, it is held bulb down- wards and shaken ; this causes the mercury within the stem to unite with the mercury in the bulb. The thermometer is then hung horizontally. Minimum 1 hermometer. The most serviceable form of instrument for registering the lowest temperature is a spirit thermometer having a small metallic index, which does not move freely within the stem (Fig. 27). When the index is at the end of the column of liquid and the temperature falls, the colunm contracts and draws back the index with it in consequence of adhesion until the temperature ceases to fall. \Yhen the temperature rises the spirit expands, and, flowing- past the index, does not displace it. The index thus registers the lowest temperature that has been reached. (nl C^ ^ ,l.,J^,l „ .l . m*i . , ! .iiil. i i|li. MJ i M!.i . i! ..i JuijJiJmijM Fig. 27. Minimum Thermometer. The thermometer is set by sloping it, bulb uppermost, until the index is made to run down to the end of the coluuni of spirit. The instrument is then hung up horizontally. Like the dry and wet-bulb thermometers, the maxiuiuiir and minimum thermometers should be exposed in the Stevenson's screen, and should be clamped to the battens in the screen from which they are suspended. The arrangement of the thermometers in the screen recom- mended is described on page 162 and is shown in Fig. 25. There should be a space of at least three inches between the bulb of the thermometers and the top, bottom or sides of the screen. The thermometers should be so arranged that all parts of their scales can be read without the necessity for moving any one of them. 11979 G 2 16d Meteorological Instruments. Stevenson's Screen. The screen which has been found most suitable for the exposure of thermometers in this climate, and has therefore been generally adopted in the British Islands, was originally constructed by Thomas Stevenson, Engineer of the Northern Lighthouse Board ; it has already been referred to (Fig. 25). The two sides as well as the front and back of this screen are double louvre-boarded, the louvres sloping in opposite directions. It has a double roof, the main one being perforated to give additional ventilation, and the battens, which form the base of the screen, are separated for the same purpose. Thus, while the thermometers in the screen are protected from radiant heat and rain, they are nevertheless exposed to a free circulation of air. The screen should be well supported on four legs which are buried sufficiently deep in the ground to ensure steadi- ness in strono- winds. It should stand over short grass, its base raised o feet 6 inches above the level of the ground. The face of the screen should open towards the true north, so that at any time when observations are being taken the instruments may not be exposed to the sun's rays. Sea Water Thermometer. Observations of sea surface temperature are taken for the Meteorological Office, twice daily : at about sunrise and at 4 p.m., at a large number of places on our coasts. The thermometer used for this purpose is protected by a metal case, having at its base a receptacle for water, capable of holding sufficient to cover the bulb of the instrument. To observe sea temperature at a shore station a convenient place is selected where the water is not less than six feet deep because shallower water responds too quickly to changes in air temperatui'e, to increase of heat due to solar radiation, and to diminution of heat through cloudiness. Consequently it does not represent fairly the surface tempera- ture of the coatsal water. The thermometer, secure in its metal case, is plunged under water, kept there three minutes, and then promptly read off. If a canvas bucket and line are available the bucket is thrown into water, not less than six feet deep, and left there for, five minutes ; it is then hauled in, full of water, the thermometer is placed in it, and after three minutes' immersion is read, while the instrument is held upright, with its bulb still immersed. Anemometers. 167 A canvas bucket is recommended because of its portable- ness : it is easily thrown to a considerable distance, and easily carried when full of water. It should be about 14 ins. deep and 7 ins. in diameter. Anemometers. Many devices have from time to time been suggested for measuring the velocity or the pressure of the wind. The first, of which there is any authentic record, is that described by Dr. Hooker in the first volume of the Philosophical Transactions (1667). It consisted of a plate hinged in such a way that it could swing upward under the pressure of the wind. The extent of its upward movement was measured upon a graduated arc, and " the force or strengtii of the wind in proportion to the resistance of the flat side of the instrument "' was thus determined. Osler^a Pressure Plate. The form of pressure anemometer which until recently was generally used, was introduced by the late Mr. Osier of Birmingham, and bears his name. It had a vertical plate, generally of one or two square feet area, which was kept facing the wind, and by the wind was driven back against the resistance of a spring. The deflection of the plate from the perpendicular, thus produced, was registered upon a sheet of paper, indicating the force exerted by the wind upon the plate. The chief defect of this instrument lies in the unchecked momentum of the plate ; and the exaggerated pressures which the instrument not infrequently recorded may probably l)e attributed to this cause. A modiried form of this instrument, in which the plate has been prevented from oscillating after being struck by a gust of wind, has been made and used. A pawl is employed for tliis purpose and the plate is thus retained at the extreme limit to which it is pushed in a gust, and it can move no further back until a stronger gust arrives. It moves, therefore, step by step, up to the highest force attained by the wind ; and it is this maximum wind pressure alone, and not a continuous record of the variations of wind pressure, from moment to moment, which is obtained. The mechanical arrangements required by most forms of recording anemometers necessitate the placing of such an instrum^^nt upon a building of some kind, and the result frecpicntly is that its exj)osure to the wind becomes seriously impaired, the record l)eing marred in consequence of eddies to an extent which cannot be accurately estimated. ll'.»7'J Qt 3 168 Meteorological Instruments. RohinsorCs Cup. In the year 1846 Dr. Robinson introduced an anemometer for giving the velocity of the wind, which is commonly known as the Cuj) Anemometer. Fig. 28. It consists of four hemispherical cups fixed to the ends of a horizontal cross which can revolve upon a vertical spindle. The force exerted by the wind upon the cups causes the cross to rotate, and its revolutions are registered either upon dials or upon a sheet of paper round a drum the rotation of which is effected by clockwork. With this iustrument it is obviously necessary, first of all to determine the relation between the speed at v\"hich the cups move, and the speed of the wind which drives them, and it is found that this relation depends upon the size of the cup, and also upon the length of the arm to which it is attached ; it is consequently different for instruments of different pattern. The K obi n son Anemometer Fig. 28. In the standard instrument, adopted by the Meteorological Office, wliich has cups of 9 inches in diameter, fixed to arms measuring 2 feet from the centre of the cup to the spindle, the mechanism should be such as to make it necessary to multiply the distance travelled by the cups by 2"2 in order to obtain the corresponding travel of the wind. In a smaller instrument, which has ctips 3 inches in diameter, carried on arms 7f inches long, the factor has been found to be 2*733 instead of 2'2. Dines^ Pressure Tube. 169 Bines' Pressure Tube. A more recent form of anemometer, and the one which gives information not only of the pressure and velocity of the wind, but also the way in which the velocity varies from moment to moment, is the pressure-tube anemometer invented by Mr. W. H. Dines, F.R.S. This anemometer depends on measuring the small differences of pressure that are produced in the air when the wind passes over an obstacle. In general, when moving air meets with a fixed obstacle there is an excess of pressure on the windward side, and a decrease of pressure on the lee side, but with light winds the differences are small, and difficult to measure. In the tube anemometer a tube with one end closed and one open is kept by a vane with the open end facing the wind, and in consequence there is an excess of pressure inside the tube ; also, of course, inside any closed vessel with which the tube is put into airtight communication by means of a pipe. It was found by experiments conducted by Mr. Dines, that the wind when it blows over a vertical tube of one or two inches diameter, having a ring of small holes round it, exerts a sucking action, and produces a decrease of pressure inside. This fact is made use of in connexion with the construction of the tube anemometer ; and the difference in pressure causation is measured between a tube with its open end facing the wind, and a tube, with a ring of holes round it, in a position roughly vertical : for the exact angle is not of any consequence. For a wind of 100 miles an hour this difference amounts to 7 '31 inches of water ; the value depends on the square of the velocity, and thus for 10 miles an hour it is jIq of the above value, namely '073 inch ; for 20 miles an hour it is 4 X '073 inch : for 30 miles, 3 squared or 9 x "073 inch, and so on. This form of anemometer is very convenient since the head, or part exposed to the wind, when once fixed will run for years without oiling, and the registering part may be under cover at any reasonable distance from the head. If automatic registration be not required ; also if observa- tions of light winds may be excluded, the simplest form of registration is an ordinary U glass tube. With such a tube partially filled with water, the force of winds equal to or above a fresh breeze, can be read (»ff on a scale suitably adjusted. 11979 G 4 170 Meteorological Instruments. For continuous registration the recording apparatus shown in Fig. 29, and the head of the anemometer in Fig. 30 were invented bv Mr. Dines. In connexion with the former the Fig. 29. Recordhig Apparatus. Dines' Pressure Tube Anemometer. clock drum D is employed, and a fine pen, P. which traces a record on a chart adjusted round the druti}, as is the case with the barograph, and thermograph ab'eadv described. The pen is carried by the rod B, which is attached to a Dines' Fressure Tube. 171 floating cylinder F called the jioat. The rod is prevented from rotating by the guide G, to one end ot which it is connected, while the other end moves in a vertical slot C. This float is shaped in a special manner, and in its position of equilibrium a very small force acting upwards on it will raise it a considerable height out of the water in which it floats. Its open end or rim is downwards like a diving bell, 3.nd the air enclosed between the water and its upper part is in communication with the open tube P on the head, Fig. 30, PRESSURE SUCTlClfl Fig. 30. The Head. which is kept facing the wind by the vane V. The commu- nication is effected by a tube (P') of about g inch hi diameter, which runs down from the head and up through the water ■in the float, its mouth opening inside the enclosed air space. The float, and the water which buoys it up, are contained in a sealed vessel T, the rod carrying the recording pen passing out through a hole which it fits closely enough to be air-tight or nearly so, but not so closely that there is any ap[)reciable friction. It is obvious that the wind blowing into this tube, will raise the float, and with it the pen. The top of the sealed vessel T is connected with the vertical tube S, which has been referred to as liaviuij: a rintr of holes at the head, by a -I 111' pipe so that the sucking action of the wind, produced by its j)assage across the holes, mav take effect. 172 Meteorological Instruments. The action of the instrument depends chiefly on the excess of pressure produced by the wind against the mouth of the tube P ; the second communicating pipe S is introduced, not to increase the difference of pressure caused in the two tubes, but for the following reason. The tube anemometer has to measure very small differences of pressure, and, if one connecting pipe only came from the head to the registering apparatus, the registration would depend on the pressure of the air in the room where the apparatus was placed. In windy weather it would in fact depend largely upon what windows and doors were open or shut, so that the regis- tration of correct values would not be possible. Supjoosing the suction tube to be absent, and the registering apparatus to be placed in a small hut, then the act of opening the door of the hut to windward, if the hut were well exposed, would, during a gale, reduce the recorded values from gale force to that of a fresh breeze or possibly even to that of a light breeze. The suction tube removes this source of error, by confining the wind's action upon the instrument to that which is communicated through the two pipes on the head of the anemometer. In the " Observers' Handbook " the following instructions are given in regard to the management of this instrument : — (1.) It should be kept clean, so as to admit of the free working of its parte. (2.) It must be kept level. To test the level remove the collar through which the pen rod passes, and note whether the latter passes through the centre of the circular hole in the top of the recorder. If not, adjust by means of the levelling screws. Replace the pen carriage on the rod before making the adjustment. The screws should bear on a metal surface and not on wood. (3.) The level of the water in the tank should be kept up to the fixed mark in the gauge which projects from the side of the apparatus. (4.) When the pressure is identical on both sides of the float, the float should be quite free, ^.e., it should not be supported on the base of the tank. If it is, there will be a " dead " interval within which the upthrust on the float produced by wind is not sufficient to raise it. The float is so constructed that it rests in contact with the base of the recorder in its zero position without exerting any pressure on it. If the adjustment is not perfect, it should be altered by adding or withdrawing shot from the cup which will be found on the pen carriage. (5.) The position of the pen should correspond with the zero of the scale when the air pressure is the same on both sides of the float, i.e.y when both taps are turned so that communication with the head is shut off. A screw motion is provided for making this adjustment. Pressure tube recorders are primarily graduated upon an empirical basis by Mr. Dines's experiments,* but the indications of different instruments can be compared with a pressure gauge. Meteorological Instruments. 178 Rain Gauges. There are several different forms of this instrument but the'majority of rain gauges in use at the present time are those of the Snow don pattern or a modified form of that pattern to which the principle of the former is applied. The Snoivdon, Fig. 31, is a 5-inch gauge and consists of cylindrical funnel (/) having a rim 4 inches deep, to the edge of which a stout brass ring (r) is firmly fixed ; a vertical cylinder with closed base (c), and shoulder (s) upon which I I 1 1 I I Ic 1 1 Fig. 31. the lower edge of the funnel-cylinder rests when in position, and a can (/>), which rests on the bottom of the lower cylinder. Precipitation is directed from the funnel to the glass bottle by means of a pipe (p) attached to the former, reaching almost to the bottom of the latter : • The brass rinsf, the inside measurement of which is exactly 5 inches in diameter, is bevelled on the outside so as to form a knife edge u})on which no rain can rest. The rim of the cylindrical funnel is made 4 inches deep in order to prevent loss of precipitation by splashing ; also to facilitate its collection when in the form of snow. With the exception of the ring, the instrument is usually made entirely of copper, on account of the durability of that metal . The rain collected is measured by pouring it into a measuring glass (//) graduated to indicate either millimetres or hundredths r)f an inch. Meteorological Office Pattern. 'I'he gauge adopted by the Meteorological Office is eight inches in diameter ; and the rim of the funnel is six inches deep. The lower half of this gauge has a splayed base 174 Meteorological Instruments. as shown in Fig. 32, where (/) is the cylindrical funnel; (s) the splayed base ; (c) the can, and {g) the measuring glass. y Fig. 32. The measurino- olass will hold half an inch of rainfall, an amount which corresponds with a Aveight of 14'50 oz. avoirdupois when collected in an 8-inch gauge or with 5'67 oz. when collected in a 5-inch gauge. Rain gauges should be sunk into level ground, and their rims should be one foot above the surface. They should be fixed firmly in the ground ; the object of the splayed base is to contribute towards its stability when in position. The Meteorological Office pattern of gauge is made of five inches diameter also ; it differs from the Snowdon only in having a splayed base. 175 APPENDIX I. Displacement of the Horizon. The conditions which occasion displacement of the horizon through refraction are analogous to those under which the pheno- menon of mirage occurs, and result from unequal density of the different layers of air by contraction or expansion through contact with chilled or heated surfaces. The air generally diminishes in density ^s'ith height. Rays of light traver.siug a medium of varying i*efractive iudex proceed obliquely upwards from an object, becoming more and more horizontal, but nsually passing away into space. Assuming the density of the air to diminish with height nnusnally quickly, as it does when it is cooler the nearer it is to the surface ; commonly the case in the Red Sea, for instance, where hot air from the land passes over relatively cool water, then the angle of incidence is increased from one layer of the atmosphere to another, so that the obliquely ascending rays may become very nearly horizontal. When this critical angle is reached, as at A, Fig. 33, reflection succeeds refraction, the rays are bent downwards, and again undergo a series of refractions, but in a contrary direction, until they reach the surface at a distant point. The horizon at this point, as seen, is too high ; and erect images of objects, such as 0, that are below the horizon, may thus be observed. Fig. 83. The faint indications of coast line, sometimes noticeable beyond the limits of direct vision, which sailors speak of as the loom of land^ are produced in this manner. Should the density of the air increase with height, as when cold air passes over the surface of relatively warm water, and such con- ditions frequently obtain in temperate latitude.s of the North Atlantic that come under the inlluence of the Gulf Stream, the lower stratum of air becomes warmer, and therefore less dense than the strata above it. Under these circumstances a ray of light from a distance suffers refraction that diminishes with lessened density and is convex towards the sea surface ; with the result that the apparent horizon is seen too low ; and a distant object, such as 0, Fig. 34r, is seen at P inverted, appeiu-ing to come from tlu' direction P A D. 176 APPENDIX 11. Tiie following notice with reference to the supply of Telegraphic Information to the public is taken verbatim from Circular 001 of the Meteorological Office : "Provisions for the supply of information to the public." During the continuance of hostilities, however, the provisions for the supply of information to the Public have been modified and are subject to further modifications : — Daily Weather Reports, Forecasts and Storm Warnings. Between 7.15 a.m. and 9.30 ajn, telegraphic megsages are received daily, reporting meteorological observations at 29 stations (marked T in list of fctations, Section I) in the Bi'itish Isles, chiefly on the coast, at 39 stations on the Continent of Europe, including Gibraltar, and at the Azores, at five stations in Iceland, and one in the Faeroe Islands. The observations are now made at 7 a.m. at all stations, except Oxford, Lisbon, and the Azores. A certain number of stations report evening observations (6 p.m.), also by telegram, and those that do not report in the evening include the evening observations with the following morning reports, so that a complete schedule of morning and evening observations is drawn u)) daily. The information refers to the readings of the barometer, dry bulb thermometer, maximum and minimum thermometers, rainfall, and in some cases, sunshine, with estimates of the direction and force of the wind, and reports of the weather and state of the sea. The observations received from Iceland give only the readings of the barometer and the dry bulb thermometer, the direction and force of the wind, and the state of the weather. These reports are supplemented by a number of additional obser- vations made at various stations in the United Kingdom, and sent either by telegram or by post, by private persons or local officials. Moreover, the " Bulletin International " published in Paris, repro- ducing meteorological telegrams from the whole of Europe, is received by post on the morning of the day after publication, and supplements the information previously received in the Office by telegram. Through the courtesy of the Lords Commissioners of the Admiralty occasional reports of observations at sea of£ our southern :;ind western coasts are transmitted by wireless telegraphy from the ships of H.M. Navy. Wireless reports are also received almost daily from ocean lineT-e crossing the Atlantic. The telegraphic information is tabulated ard charted by about 9 a.m. for the morning observations, and 7 p.m. for the evening ones. A general report is then drawn up, and forecasts of the weather, for iiie twenty-four liours following the next noon, or midnight, as the case may be, are formulated ; a note as to the further outlook is added if the meteorological coniitions are such as to justify an anticipation for more than twenty-four hours ahead. ^t a selection of stations additional observations are t.iken and telegraphed to the Office at 1 p.m., and occasionally modifications are made in the morning forecasts as a result of these observations. This information is usually available by 2 p.m. 177 Forecast Districts.— For the purposes of forecasts of weather the region of the British Isles is divided into land districts and sea districts, as indicated in the accompanying map. Fig. 35. Forecast Districts, Land.* 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Scotland, J (a) Islands. North. ( (J) Mainland. Scotland, East. England, /(a) North. North-East. \ (i) South. England, East. MIDLAND (['j^.^l',, COUNTIES, ^gjj^-f; England, South-East. /(«) Scotland, West. \ (&) Isle of Man. I (a) England, N.W. I (b) North Wales. ( (a) South Wales. ] (b) England, S.W. Ireland, j (a) West. North. I (i*) East. Ireland, | (m Strangford Louoh to IJan'ow Head and North of Lat. 53° N. — Howth, Kingstown, Point of Ayre, Ramsey, Douglas, Silloth, Maryport, Workington, White- haven, Barrow, Walney Island L.H., Morecambe, Fleetwood, Blackpool, Lytham, Preston, Southport, Formby, Liverpool, Run- corn, New Brighton, %Hoylahe, New Ferry, Rhyl, Port Penrhyn, Point Lynas L.H,, Holyhead, South Stack L.H., Carnarvon, Port Dinorwic. VIII. iS7. George's Channel. — Aberystwyth. IX. Bristol Channel, St. David's Head to River Camel. — Smalls L.H., -Milford, St. Ann's Head, Caldy L.H., Tenby, Pembrey (Burry Port), Llanelly, Swansea, Briton Ferry, Porthcawl, Nash L.H., Penarth, Cardiff (Bute Dock and Barry Dock), New- port, Weston-fuper-Mare, Bui-nham, Bridge water, XMinehead, Ilfra- combe. Bull Point L.H., Barnstaple, Appledore, Hartland Point L.H., Lundy Island. X. llnyhind, S. W. — Rirer Camel to River Exe. — Port Isaac, Newquay, \Godrevy L.H., 1 i/a/y/c, St. Ives, St. Sennen, Newlyn West, Padstow, Penzance, Porthleven, Scilly, the Lizard, Looe, Falmoutht (Pendeunis Castle), Fowey, Plymouth, Devonport (Mount Wise and the tDockyard), Prawle- Point, Salcombe, Teign- inoiith, I'iXmouth. XI. Enyland, S. — River Exe to Beachy Head with Channel Isles. — (iuernsey,t Jersey (St. Helier's), Portland Dockyard, Portland L.H., Weymouth, Anvil Pomt L.H., Poole, Hurst Castle L.H., Southampton, Yarmouth (I. of W.), Cowes, Ryde, St. Catherine's Point, Portsmouth (Dockyard and Horse Sand Fort), Little- hamj)t()n, Brighton, *Ncwhaven. * Telegrams only exhibited t Arrangements made for showing signals or illuminating the cone at night. \ Stations of which the names are printed in italics are in abeyance during the war. 180 XII. England, S.E. — Beachy Head to Shoehuryness' — Beachj Head, Eastbourne, fHastings, Rje, XSandgate, Dover, Deal, Rams- gate, Faversbam, Sbeerness, Cbatham, Greenbithe, Sboeburyness.* XIII. England, N.E. — From the Tweed to the Humber. — Blytb, Tjnemoiitb, Soutb Shields, Souter Point L.H., Sunderland, Hartlepool, fMiddlesbrougb, Redcar, Whitby, Filey, Flamborougb, Bridlington. XIV. England, E. — From the Humber to Shoehuryness. — Hull, Goole, Grimsby, Boston, Sutton Bridge, Lynn, Sheringham, Cromer, Great Yarmouth, Gorleston, SouthAvold, Orford Ness L.H,, Ips- wich, Harwich, Gunfleet L.H., West Mersea. C. — Statistical Information. The British Meteorological and Magnetic Year Book. Terms of Subscription. — The Statistical Publications of the Office have been grouped together under the general title " British Meteorological and Magnetic Year Book." For 1913, 1914, 1915 and 1916 the Year Book consists of four parts, as follows : — Part I. — The Weekly Weather Report. Issued on Friday of each week. Price &d. per number. Annual subscription (w^hich includes the Monthly Weather Report, see below) 30^. postage paid. The appendices to the report can be obtained separately, price from 4rf. each. Part II. — The Monthly Weather Report with an annual summary. Issued on the 27th of each month as a supplement to the Weekly Weather Report. Price Qd. each issue. Part III. — (1) Daily Readings at Meteorological Stations of the First and Second Orders. Issued in monthly parts, within about five weeks of the close of each month. Price Qd. each issue. Annual Volume consisting of 12 monthly numbers with Introduction, Annual Supplement, Title Page and Map, price 5s. (2) The Geophysical Journal. Daily observations (Mete- orology, Terrestrial Magnetism, Atmospheric Electricity, Seismology, &c.) at the Meteorological Office Observatories (Valencia, Kew, Eskdalemuir) and principal Anemograph Stations, together Avith records of Temperature, Humiditv and Wind in the free atmosphere, obtained by means of kites, balloons and pilot balloons. Price Is. each issue. Annual Volume consisting of 12 monthly numbers, with Introduction, Annual Supplement and Title Page, price IO5. Part IV. — (1) Hourly Values from Autographic Records. Meteorological Section. Hourly Readings of Pressure, Tem- perature, Wind, Rainfall, Humidity and Sunshine, at the Meteorologi(ial Office Observatories (Valencia, Kew, Eskdalemuir). Issued in monthly sections for each observa- tory. Price 6c?. each section. Annual volume, 2Qs. * Telegrams only exhibited. f Arrangements made for showing signals or illuminating the cone at night. % Stations of which the names are printed in italics are in abeyance during the war. 181 APPENDIX III. The following- notice with reference to the supply of Barometers which the Office lends for the use of fishing communities is taken verbatim from Circular 001 of the Meteorolooical Office, '' J-'rovisions for the Supply of Information to the Public.'' FISHERY BAROMETERS. The Office possesses a number of Barometers which it lends for the use of fishing communities, where it is shown that the instru- ment will be of material service. As a condition of the loan the community is required to provide for the housing of the instrument and to keep and forward to the Office a record of daily readings. Forms on which the record may be plotted are supplied by the Office. Simple instructions concerning the use of the barometer are given on these forms (M.O. 4150). A picture of a frame containing one of them is shown on p. J 86. LlSr OF STATIONS SUPPLIED WITH FISHERY BAROMETERS. District and Port. No. Custodian. District and Port. No. Custodian. EN GLA] VD. ENGLA ND— continued. Lancashire and Cornwall — Western. continued. St. Ives ... 2 Pier Head Light Morecambe 67 Sanitary Inspector. Keeper. Fleetwood UO Marine Supt. L. i: Forth Guarra 26.5 Mr. T. Joyce. N.W. & L. & Y. Penberth Cove ... 241 Mr. A. Jeflfery, Riilway. Carrier. Carnarvon 135 Customs Officer. Mousehole 7 Harbour Master. Aberystwyth 221 R. Kenrick, Esq. Newly n Town 23 Mr. R. Pollard. Trinity Pilot. Milford Haven. Newlyn 137 Mr. E. Morrish. Coverack ... 153 Coast Guard. Milford r>s ]\Ir. H. Lewis. Falmouth 62 Customs Officer. Angle 180 Mr. Davies, Grocer Porthallow 38 Coast Guard. Durgan 81 Mr. E. Downing, Giamorgran. Fisherman. 379 ^ Penryn 63 Customs Officer. Swansea 126 Customs Officer. Portscatho 1-J6 Mr. Wesley Collins. llriton Ferry Dockmaster. Devoran 77 Hon. Sec. of Read- ing Room. Somerset. Gorranhaven 49 Coast Guard. Mevajrissey 41 1) )) lUunhani... .">2 Customs Officer. Devon. Cornv.'all. Bideford 218 Miss A.M. Braund. Fort Isaac 4r, Coast Guard. Exmouth ... 191 Customs Officer. Haylo 11 r Customs Officer. Budleigh Saltcrtou 47 Coast Guard. 182 LIST or STATIONS SUPPLIED WITH FISHERY BA.nOMETEliS-cuHthn; Mr. G. Watt, iMer- chaut. Fisherman. Ardneaskan 252 Mr. D. Mackay, Rosehearty 79 Harbour Master. Fisherman. PituUie (Sand- 85 )i Shieldaig (Apple- 239 Mr. J. Grant. hav< n). cross) Fraserburgh 200 •! •, Badachro... 244 Mr. J. Mackenzie, Inverallochy Itl Mr.' J. Duthie, Fishcurer. Fisherman. Ullapool 151 Mr. W. Urquhart, Poiutlaw 150 Mr. B. Tuck, Ferry- Grocer. man. East Mey 124 Mr. J. Mackaj, Portlethen 61 ]\rr. J. (-laig-, Fisherman, Fisherman. Gills 238 Mr. J. Geddes, Skate raw ... 164 Mr. W. Christie, Crofter. Fisherman. Stroma, Nortn ... 114 Mr. W. Allan, Stonehaven 138 Harbour Master, Fisherman, Arbroaih 210 Mr. T. Suankie, Ship Chandler. „ , Soutli ... 75 Mr. A. S. Robert- sou, Postman. 184 LIST OF STATIONS SUPPLIED WITH FISHERY BAROMETERS— w/t^z«M#^Z. District and Port. No. Custodian. District and Port. IRELAI No. Custodian. SCOTLAND— coniinued. 'JD — continued. Hebrides. South Coast. Ness 69 Mr. J. McLeod, Dunmore East ... 163 Coast Guard. Postmaster. Dungarvan 78 Barbour Master. Carlo way 271 Harbourmaster. Kinsale ... 107 Coast Guard. Marvaig- 262 Mr. W. Kerr, Union Hall 228 >i )) Schoolmaster. Castletownshend... 9 V )) Crossbost ... 266 Postmistress. Baltimore 197 Storuoway 68 Harbour Master. Schull (2) j 3 185 1 Portnaguran 207 Crookhaven Castletown (Here- 154 227 Valtos 209 haven). Lawrence Cove ... 234 )) !) Obb 94 Mr. W. Stewart. Bally donegan 232 i; )i Ballycrovane 229 " !' Beruera ... 255 Mr. F. Paterson^ Merchant. Boreray 208 Mr. D. McDonald, Merchant. West Coast. Lemreway 199 Mr. A. McLeod. Valencia ... 246 Coast Guard. Locli Boisdale ... * Mr. A. MacLennan, Dingle 39 M II Tralee (Fenit) 71 Mr. A. Board. *Large Barogra ph No . 137. Kilronan 25 Coast Guard. Galway .• 236 Customs Officer. Spiddal 196 Coast Guard. Cleggan 220 ;) )' IR] ELAIS rD. Elly Bay Ballyglass 250 226 !I !> East Coast. Ballycastle (Mayo) Mullaghmore 20 15 Belfast 73 Customs Officer. Donegal 166 )) )) Bangor 51 Coast Guard. Tribane 145 )) !) Groomsport 172 Killybegs Teelin 104 144 Portavogie 28 Coast Guard Malinmore 225 Donaghadee Ballyhalbert 189 Station Master. Port Noo 176 I I 187 Harbour Master. Burton Port 65 !> )! Cloghy 251 Coast Guard. Kincashla 201 1! !) Ardglass 72 !) !) Bunbeg 83 :> )• Carlingford 120 )) ;• Inuiscoo Island ... 258 Captain J. Glenarm 130 J! I) O'Donnel Greenore ... 127 ,. ,, Dundalk 36 Town Clerk. North Coast. Dunfanaghy 46 Mr. W. T. Auld. Loughshinney ... 170 Coast Guard. Bathmullen 48 Harbour Constabl Clogher Head ... 242 !) !! Buncrana 132 Coast Guard. Malahide 122 ;) 1! Malin Head 139 )! !: Howth 59 1! !? Moville Greencastle 147 256 !) )) Kingstown (H.M.S. Ranger). 98 Coast Guard. Port Stewart 159 Harbour Master. Portrush ... 128 Coast Guard. Kingstown (Sailors' Home). 123 Harbour Master. Port Ballintrae ... 233 n )j Balliutoy 142 J) )) Bray 179 Coast Guard. Ballycastle (co. 216 !) !) Wicklow 198 !• !I Antrim). Rosslare 113 Supply of Information to the Puhlic. 185 Frames for exhibiting Weather Information at Fishing Ports, &c. In accordance with the tradition initiated 50 years ago, the Meteorological Office has lent barometers to about 230 fishing villages, and in peace time sends storm warnings by telegraph to 250 seaports. These provisions have tor their object the diminu- tion of the loss of life and property at sea. The regulations under which the loans have been made and the telegrams sent provide that the incidental cost of exhibiting the barometers and hoisting the signals have to be borne by the locality. To enable fishermen and others to take advantage of these facilities for making use of our increasing knowledge of the laws of weather, the barometer diagram and the storm warning tele- gram, together with the explanation of the meaning of the warning, should be exhibited where they can be seen by all. From the fact that the several localities have been left to make their own arrangements for exhibiting the Weather Information there has been no organised system for providing suitable frames, and in consequence, in many cases, the information has not been accessible to the people for whom it is intended. The Office has also made a practice of sendiiig post free to a number of seaports cdpic^ of th>' " Daily A\'eather Report." 186 (Jencral ProL'isionti. This Iveport gives detailed observations as received by telegram from tlie [Jnited Kingdom, Atlantic Islands, and European countries, and by " wireless " from ships at sea, together with the charts and the forecasts based on these observations. Both sides of this Report should be visible to the public, but the local pro- vision for exhibiting it is often inadequate and sometimes fails altogether. In order as far as possible to meet this want, a teak frame has been designed of the proper size. It is strong and waterproof. A sliding bolt to catch the sheets of paper which are to be exhibited has been devised, and, on trial, has been found serviceable. Frames of this pattern can now be sent out from the Meteorological Office ready for fixing; they can be secured directly to wood-work or to wall plugs. For the exhibition of the w^hole of the information sent out by the Office in the forms mentioned above, four frames would be required, as shown in the accompanying sketch, but thej^ may be obtained separately to meet the requirements of individual localities. The cost of the frames is £1 lOs. Od. each (carriage to any railway station in the British Isles included). Applications for supply should be addressed to the Director, Meteorological Office, Soilth Kensington, London, S.W. 187 INDEX. A. Absolute Units „ Zero, Lord Kelvin ... Action of Barometer compared with that of pump ,, Wet Bulb Thermometer Admiral Beaufort's Wind Scale ,, „ ., „ , New Alternative Specification Air, Capacity of, for moisture ,, , to carry heat Cooling' of ... ... currents, Motion of ,. , under action of gravity motion ; Dr. H. H. Hildebrandsson Rotation of swirls temperature Water vapour in ... Aitken, Dr. John, F.R.S., Cloud Formation Alternative Specification, Admiral Beaufort's Wind Scale Anemometer ... ... ... , Dines' Pressure Tube... , Earliest ... , Exposure of , Osier's Pressure Plate , Robinson's Cup , Standard Instrument Aneroid Barometers ., „ , Correction of Readings of „ ,, , New Dials for „ ,, , Mercury versus Angle of Indraft Antarctic Bergs, Formation of C. E. Borch«rrevink Captain R. F. Scott, C.V.O., Captain C. Wilkes, U.S.N. Anticyclone, Temperatures in ,The Anticyclonic Wind Circulation Appearance. First, of Iceberg in Thick Weather... Approaching Depres-^ions Atmosphere. Actual Temperature of the „ , Circulation of the „ , Composition of the „ , Cyclonic and Anti-Cyclonic Circulation „ , Density of the ... „ , Heat of, derived from sun „ , Height of the Atmospheric Pressure... „ ,. , Fluctuations in „ „ measured by barometer ... Attached Thermometer „ „ , Absolute Temperature for Average, Definition of „ frequency of gales ... ... „ wind frequeucy, United Kingdom Pa^'e lU 16(> U4 162 xviii, xix, xx, xxi ... XX, xsi, xxii 31 2& 57 22.32,48 ' 32 i5 20 ix, xii 24.25,26 21 39 XX xviii, 167 xxvi. 109, 17i> 167 xviii, xix 167 168 16« 155 155 155 8 33 139 Ll.X 140 138 89,90 23,68 23.33 124 83 25,26 20,22 20 23 20 22 2U 21,32,68 86 143 144 4 xxvii 110 xxvii ISS B. Page. Backing of Wind „. 72 Balloons and Fog xxviii Barnes, Professor, H. T., F.R.S., Recording Micro-Thermometer 123 !!)!,, ,, , Sea Temperature xv Barograph, Cleansing pen of 158 ,, , Portable 156 Barometer, Absolute Units 144 ., , Action of, compared with that of pump 144 Aneroid 155 „ , Corrections required for ... ... 155 Attached Thermometer 144 Cistern 143,145 Correction for Index Error 145 „ Standard Gravity — tables of 146, 147 error, Capacity 145 „ , Capillarity 145 Errors of Parallax ... 151 Fishery, Description of 144,147,149 ,, , Introduction of viii „ , Manual ... xiv, 94 „ (Provision of 181 , Reading of 147, 148, 149 Fortin's 145 Handling of 151 Kew pattern 145, 150 ,, ,, , Graduation of ... 17 Manual for the use of Seamen xiv Method of reading 151,153 New method of using x, xi, xii Old „ , ... ... ... ... viii '•Pumping" ... ... ... 9 Reduction of Readings of, to Standard Latitude 146, 147 „ ., ., „ Temperature 2. 145 Scale of ... 143 Sea ... ... ... , 155 Simultaneous Readings of x, 32 Suspension of 150 Torricelli's Experiment 143 Traditional Inscription on ... ... ... ... ... ... viii Use of Readings of . at Central Office x „ „ , out of reach of Central Office ... ,. Vernier Barometric Gradients ; Rev. Clement W. Ley, M.A., F.M.S. ... ; Dr. W. N. Shaw, F.R.S ,, ,, ; Captain H. Toynbee, F.R.A.S., F.R.G.S., F.R.Met.Soc. Readings, necessity of simplicity Barrier, The Great Ice Barriers and their Formation Bay Ice ... ... ... Bearing of centre of depression Beaufort, Admiral, Wind Scale Bentley, W. A., Twenty Years' Study of Snow Crystals Birth of Icebergs ... Blink, Ice Borchgrevink, C. E., Formation of Antarctic Bergs Brash Ice ... British Isles, Fog round „ „ , Gales on coasts of ,, „ , Mist round ... Brodie, F. J., F.R.Met.Soc, Gales of United Kingdom Bulletin International, Supplementing forecast information .. Buys Ballot, Professor, Law propounded by XI 144,147,151 36 ... 37.38 36 4 137,139 122, 138, 140 123 83 xviii, xix, xx, xxi, xxii 61 127.139 127 139 123 56 113 56 109 176 23,37 189 c. Page. C.G.S. units lo'^ Calculation of Wind Velocity ... ... •• ^'' Calving of Iceberg's ... ... ... ..n ... ... ... ... ...122,141 Capacity of Air to hold Water Vapour ... ... ... ... ... ... 21 Capacity, Error of ... ... ... ... ... 14^5 Capillarity, Error of l-^S Cause of Fog 50,51,52 „ Rain 58 „ Wind 21 Celsius Thermometer Scale ... ... ... ... ... -•• ^^O Centigrade Thermometer Scale ... ... ... ... ... ... ... ^^0 Central Meteorological Office X, xii, xiii, xix, xxiii, xxvi, 4, 63, 68, 88, 168, 174, 18.=1 Centre of Depression, Bearing of ... ... ... ... ... 8S , Path of 84,86 „ Gales, Speed of ... ... ... ... ... ... ... ••■ 1^^ ,, , Tracks of 116 'Jhauge from Sail to Steam ... ... ... ... ... ... ... ... ^^ Charts, Monthly Meteorological xiii, 178 ,, , Preparation of Weather ... ... ... ... ... 68,69 „ , ciynoptic ... ... ... ... ... 69, 178 Circulation of Atmosphere ... ... ... ... ... ... ... ••• 20 „ , Wind, Anticyclonic ... ... ... ... ... ... ... 22,32 „ , „ , Cyclonic ... ... '^s, o^ Cirrifoim Cloud Motion aud Weather Changes ... ... .•• ^1 Cistern Barometer 143,145 „ of Barometer ... ... ... ... ... ... ... ... ••• 1^0 Classification of Clouds by International Committee ... . ... ... ... "^^ ,, Gales in Quadrants ... ... ... ... l^^ Climate 26 Cloud Atlas, International '1^ Cloudiness, Diminution of heat by ... ... ... 27 Clouds XXX, 38 „ and Weather Signs ... 47,86 „ , Classification of, by International Committee **^ ., , Cirrus ... 39,42,46,48 „ , Composition of ... ... ... ... ... ... ... ... 38,39 Cumulus 39,42,44 „ . Cloud forms, illustration of ... ... ... ... ... ••. xxxi „ , Direction of ... ... ... ... ... ... ... ... .•• ^^ ,, , Distribution of, in a depression ... ... ... ... ... ... ^7 „ , Dr. Julius von Hann, on ... ... ... ... ... 41 „ , Formation of... ... ... ... .. ... ... ... ••. 80, 3. > „ , „ ; Dr. John Aitken, F.R.S., on ^^ „ , Height and Speed of ^^' *? „ , Hildebrandsson on ... ... ... ... ... ••• 46, 4< „ , Investigation, International ... ... ... ... ... .-. f^ „ , Luke Howard's nomenclature ... ... ... ... _ ^^ „ , Motion of upper ... ... ... ... ■•. 4.), 49 „ , Seasonal speed of ... ... ... ... ... ... ... ••. ^'* ,, Stratus, Suspension of ... ... ... ... ... ... ••. ^^ Col, The 81 „ , Winter 82 Cold Periods, Recurrence of Warm and ^^ Colouring of Icebergs, Explanation of ... ... ... ... ... ••. 121 Composition of the Atmosphere ... ... ... ... ... 20 Conduction, Heat imparted by 25 Conferences, International Weather xiv Contraction of Thermometer Bulb ... ... ... ... ••• 16^ Convexion, Heat imparted by ... ... ... ... 25 Corona .. ... 62 Correction of Barometer for Standard Gravity ... ... 146 Corrections required for Aneroid Barometers ... ... ... ... ... 155 „ ,, „ Barometric Readings ... ... ... ... ... 2,8 ,, „ „ Temperature 14, 16 1^0 Crystals, Ice ; Sir Napier Shaw, F.R.S ,, , Snow Cup Anemometer, Robinson's Currents, Air, Motion of „ , „ , under action of gravity and Earth's rotation Cyclone ,, Centre, Position of observer with regard to , Exceptional speed of „ . Origin of the term ... „ , Path of ,, , Primary ,, , Secondary ... ,, , Variations in rate of travel of ,, , Vortex of ... ... Cyclonic Backing of Wind ... ,, ' Depressions „ „ associated with Gales ,, Wind Circulation „ Winds ; Sir Napier Shaw. F.R.S Page. 62 ... 60, 62 168 22,23,48 ... 32, 33 ... 23, 70 46 119 24 24,47,48 73 ... 73, 74 118 ix 48, 49, 72 .. ix, 33 115 .. 23,33 37 D. Daily Weather Reports Decrease of Temperature with Height Definition of Average ... „ Gale „ Isobars ... „ Isotherms „ Storm Field Density of Atmosphere Depression ,, , Approaching ,, , Bearing of centre of ... „ , Cyclonic ... ., , First Indications of „ of Wet Bulb ,, , Path of centre of ,, , Satellite ,, , Trough of „ , " V "-shaped ,. , Weather sequence of ... Detection of Ice ... ... ... Dew Point Dewar, Sir J., illustration of Regelation ... Diffraction Diminution of Heat by Cloudiness Dines, W. H., F.R.S. ; Pressure Tube Anemometer Instructions for ment of... ,, „ ; Temperature in Winter Anticyclones ... Direction of Wind in Gales Displacement of Horizon Dissolution of Icebergs Distinction between Fog and Mist Distribution of Gales ... District Forecasts Diurnal variation of Temperature ... Doldrums Dove, H. W., F.R.S., on Hail Storms Drift Ice Drifting Ice, Detection of Drift of Ice, Effect of Weather Conditions on ... „ „ from Greenland Sea ... ., „ , Rate of, in Labrador Current Dry and Wet Bulb Hygrometer ... ,. „ Thermometers ... 176 ... 27,58 xxvii xxii. 109 ... ' 32 29 4S 20 23 83 83 ix, 33 83 163 ... 84,86 ... 46,73 48,71,76,86 ... 75,85 ... 8.5,86 ..123,124 31 120 63 27 xxvi. 169. 170 manage- 172 90 111 175 131 xxvii, 50 109 177 28 76 60 123 123 135 132 136 162 ...162,164 191 151 29 30 E. Page. Earth's rotation. Effect of, on Wind 22 Easterly type of Weather ... ... ... ... ... .•• ^9.. Education and Metric Units xviu Effect of Ice on Sea Temperature ^^ ., , Remarkable, of Hurricanes ... ... ... ... ... ... ••• ^^^ Elastic Force of Vapour ... ... ... ... ... •• ••• ••• ^| Equatorial Gales, Proportion of ^^i Error of Capacity ... Jjr „ Capillarity ]jl Index 145 „ Parallax ... ... Evaporation ,, , Acceleration of... Exceptional Speed of Cyclones 1^^ Expansion of (xases |^^ Explanation of Colourinsr of Icebergs 121 ,, Fog Charts •?o Exposure of Anemometer ... ... ... ... •• xix, xxin Extreme Limits of Ice l^* P. Faraday's discovery of Regelation _^ 120 Fahrenheit, Temperature Scale of , 159,160 Ferrel. VV., on Hail ^9 Fiducial Temperatures, Table of 1^ Field Ice, Formation and Colour of Ip2 First appearance of Iceberg in thick weather ... ... ... ... -•• 12* ., indication of Depression ... ... ... ■•• ••• **^ Fishery Barometer, Description of ... ... ... ... -.• ••• 144,147,149 „ ,, , Introduction of . ^'"^ Manual J^J „ „ , Provision of ... ... ... ... .. ••• ••• ^°'- „ „ , Reading of ^*7 FitzRov. Admiral R •• ••• ^'^ " Float." Dines' Pressure Tube Anemometer .'. •.• 1^1 Floating Ice -^^^i" Floeberg 123 fluctuations in Atmospheric Pressure •• °" Fog and Balloons ^^^"i „,Cau8eof .50,51,52 ,, , Coastal, and Change of Weather .. ^2 ,. , Distinction b tween, and Mist... ... ... ... ••• xxvii, 50 ., . Explanation of Charts ^^ „, Height of -^l „ , Laud ^' „ , Number of ' 'bservations used for Results ... ... ... ■. •■• -^^ ,, , Relation 1,0 Wind Direction ... 56 „ round British Isles ... ... ... •■. ••• ••• ^^ „ , Scale of Intensity ... ... ... ... ... ••• ••• •• " ,Sea... ... ... ... ... ... ... .-• XV, xxvii, xxviii, 50, 53, 54 „ with High Pressure ... ... ... ... ... •• ••• ••• -^2 Low , ... ... ... ... ... ... ... ..• ^*^ Forecast Districts •»•!) 1^7 Formation of Antarctic Bergs ; C. E. Borchgrevink l'^9 : Captain R. F Scott, C.V.O., R..V 140 „ „ ; Captain C. Wilkes, U.S N 138 Clouds 30 ", j', Fog and Mist, Taylor, Major G. L, R.F.C 55 192 Formation of Glaciers ... 121 ,, ., Ice 11» „ ., ., Barriers 122 Forms of Hail Stones 60 Fortin's Barometer ... ... ... HS- Frequency, Average of Gale 110 „ Wind, United Kingdom ... ... ... ... ... xxvii Ice in N.W. Atlantic 133^ „ in Southern Hemisphere ... ... ... ... ... ... 141 ., Yearly Variations in ... ... ... ... 141 Eainfall ... ' 66 Snow ... ... 67 Wind ... .., ... ... ... ... ... ... xxvi, xxvii Friths, Ice 12& G. Gale, Definition of xxii, 109" Gales, Average frequency of ... ... 109 „ , Classification of, in quadrants ... ... ... ... ... ... 109- „ , Distribution of 109 „ , International Symbol of ... ... xxii „ of United Kingdom ; F. J. Brodie, F.R.Met.Soc 109 „ on Coasts of British Isles ... ... ... ... ... ... ... 113 „ , Polar. Proportion of ... ... ... ... ... ... Ill „ , Relative frequency of, from points of compass ... 112 „ , Speed of storm centres in ... ... ... ... ... ... ... 117 „ , Tracks „ ,. ... ... ... 116 „ , Vortical ' 111,112,114,115 „ , Whole and Storm force ... ... ... ... ... xxiii „ , Wind direction in ... ... ... ... .. ... ... ... Ill Gases, Expansion of ... ... ... ... ... ... .■ 159 „ , Pressure of ... ... ... 29 Geographical Positions of Phenomenal Icebergs ... .. ... 136 Glacial Origin of Icebergs ... ... ... ... ... ... ... ... 121 Glaciers, Baffin Bay 129 „ Continental Greenland ... ... ...128.12 „ Formation of < ... ... ... ... ... .. ... ... 121 Glaisher, Jas., F.R.S. ; Hygrometrical Tables ... ... 31 Glass, Weather ... ... ... viii, x, 143 Gold, Ernest, M.A. ; Calculation of Wind Velocity ... 37 Gradient, Barometric ; Rev. Clement W. Ley, M.A., F.M.S 3& „ ; Dr. W. N. Shaw, F.R.S 37,38 „ , „ ; Captain H. Toynbee, F.R.A.S., F.R.G.S., F.R.Met.Soc. 3& „ (Pressure 33,34 „ , ,, and Wind Velocity 36 „ , Wind ■ 3& Graduation of Kew Pattern Barometer ... ... ... ... ... ... 17 Graduation of Thermometer Scale ... ... ... ... ... ... ... 159 (rraujjel ... ... ... ... ... ... ... ... ... ... ... 59 Gravity 121, 14G „ , Standard, Correction of Barometer Readings for ... 146 Ice off South Coast of , 132 „ Sea, Drift of Ice from ... ... ... ... ... .. ... 132 Greenwich Minimum Temperature ... 29 Gresil ... ... ... ... ... ... ... ... ... ... ... 59 Growler... 122 GuBta . xxvi. 75 Greely, General 12» Greenland, Area of ... ... ... ... 125 „ . Continental Glaciers of... ... ... ... ... 128,129 „ , East Coast, incomplete survey of ... ... ... 126 „ , First crossing of, Fridtj of Xansen ... 12f), 129 , Ice drift 12& ., , Ice mantle, formation of ... ... ... ... ... ... 126. „ , Inland Ice, Origin of Dr. H. Rink 12(;, 127 „ , ,, „ Sheet, thickness of the ... ... ... ... ... 125 „ , Nunataks 127 193 H. Hail, Formation of „ , Hon. Rollo Russell on „ -stones, Forms of .. . ... „ -storms ; H. W. Dove, F.R.S „ ,, : W. Ferrel Halos Handling of Barometer Hann, Dr. Julius von ; Clouds Heat, Capacity of Air for „ , „ Water for ,, , Diminution of, by cloudiness... „ imparted by Conduction „ ,, Convexion „ ,, Radiation „ , Latent „ of Atmosphere ... ., , Theory of, Maxwell's ... ,, , Transmission of Sun's ... Height and Speed of Clouds ., of Atmosphere ;, „ Fog Hemisphere, Northern, Ice in „ , Southern, ,, ... . . High Pressure and Fog ... Hildebrandsson, Dr. H. H. ; Air motion ... H.M. Navy, Wirelass from ... Hooker, Dr. ; Anemometer ... Horizon, Displacement of Horn Book, Sailor's ... Humidity ,, , Relative ... Hummocky Ice... Hurricanes, Remarkable effect of ... „ , Velocity of Hygrometer „ , Wet and Dry Bulb Hygrometrical Tables ; Jas. Glaisher, F.R.S. Page. 59 59 60 60 59 63 151 43 26 26 ... 27,166 25 25 ... 25,166 30 25 120 ... 25,26 ... 43, 46 20 51 127 136 52 45 176 167 175 24 29 31 123 . . xvi xvi 161 162 31 z. e , , Analogy between Rivers and Glaciers 119 121 , Barrier, The Great 137 , Barriers, disruption of 138 , „ and their Formation ..122,139 ..Bay 123 , , Blink 124 , , Borchgrevink's exploration of, in '• Southeri a Cross " 139 , , Brash 123 , Crystals ; Dr. W. N. Shaw, F.R.S. ... 62 , , Discoveries of Ross ..137, 1.S8 , , Dr. Rink on 127 , , Drift 123 , Drift, Efifect of Weather Conditions on 135 , , Drift of, from Greenland Sea 132 , , Drift rate of, in Labrador Current 136 , , Drifting, Detection of 123 . , Effect of, on Sea Temperature XV , , Extreme limits of 134 194 Page. Ice, Field, Formation and Colour of 122 Floating xxviii, 122, 123 Floe 122 Floeberg 123 — foot 122 Formation of , by Revelation 123 „ „ ; Tyndall's Experiment 119 Frequency of in N.W. Atlantic 133 ,, „ Southern Ocean 141 ., ,. Yearly Variation in 141 Friths 128 ,, Glaciers. Formation of 121 „ Grovrlers 122 ,, , Hummocky 123 ,, . Islands, Wilkes on Formation of 138 „ , Kinds of ... ... ... ... ... ... 122 , , Land 123 ,, Limits 133, 134 „ Micro-Thermometer and 123 „ Movements and Weather, Connexion between ... xxix. 130 „, Nordenskiold's Expedition's experience of 139,140 „ , Northern Hemisphere 127 „ , Number of Reports of .. 142 „ , Pack 123 „ , Pancake 123 ,,, Sheer Wall of 140 ,, , Slob ... 123 „ , Sludge 123 ,, , South Coast of Greenland off 132 „ , Southern Hemisphere ... ... ... 136 „ Structure ... ... ... ... ... 119 „ , Victoria Land 137 Icebergs 138 , Antarctic Formation of ; C. E. Borchgrevink ... 139 , „ •, ; Captain R. F. Scott, C.V.O., R.N. ... 140 „ „ ; Captain C. Wilkes, U.S.N. 138 , Birth of 127,139 , Calving of 141 , ColoLiriug of ... ... ... . ... ... ... 121 , Detection of ... , 123, 127 , Disruption and movement of ... ... ... ...130, 131 , Dissolution of ... ... 131 , First appearance of in thick weather ... ... ... ... ... 124 , Geographical Positions of Phenomenal... .^. ... 136 , Glacial Origin of 121 . Ice Barrier, Origin of 122 , Phenomenal Drifts and Heights of ... ... ... 135 , Proportion of , above Water 130 , Southern Hemisphere, Phenomenal Heights of ... ... ... 136 Iceland, Observations from ... ... ... ... ... ... ... ... 176 Index Errors ... ... ... ... ... ... ... ... ... ... 145 Indian Ocean Charts, Monthly Meteorological ... ... xiii „ Summer of Eastern Canada ... ... ... 96 Indication, First, of Depression ... ... ... ... 83 Indralt, Angle of ... ... ... ... ... ... ... ... ... 33 Inscription on Barometer, Traditional ... ... ... viii Intensity, Fog, Scale of 57 International Bulletin, Supplementing Forecast Information ... 170 „ Committee's Classification of Clouds ... ... 40 ,, Weather Conferences ... ... ... ... ... ... ... xiv Introduction ... ... ..." ... ... ... 18 Isobars. Angle of Indraft ... ... ... ... ... ... S3 ,, , Definition of ... ... 32 „ , Trend of, and V\'ind 33 ■,, , Use of - 37,38 Isotherms, Definition of ... 29 , Trend of 54 195 K. Payre Kelvin, Lord ; Absolute Zero 4,160 Kew, Minimum Temperature... ... ... ... ... ... ... ... 29 „ Pattern Barometer ... ... ... ... ... ... ... ...145, 150 Kinds of Ice 122. 12:^ Labrador Current. Course and Velocity of ... ... ... ... l^^'J Land Fog- ... ... ... ... ... ... ... ... ... ... 51 ,. Ice 123 Latent Heat 30 Latitude. Reduction of Barometer Readings for 146 LavN', Profcj^sor Buys Ballot's 23,37 Lempfert. i{. G. K., M. A., on Line Squalls 77 Letter? to Indicate State of the Weather ... ... ... ... ... ... 68 Ley, Rev. Clement W.. Af. A., F. M.S. ; Barom«;tric Gradients 36 Light, Retraction of ... ... ... ... .-• ■•■ " 175 Limits, Extreme, of Ice ... ... ... ... ... ... ... ... 134 „ of Ice 133 Line Squalls ... ... ... .. ... ... ... ... ... ... 77 Liners, Wireless Reports from ... ... ... ... ... ... ... 178 Local Thunderstorms ... ... ... ... ... ... ... ... ... 108 Lord Kelvin ; Absolute Zero ... ... ... ... ... ... 160 Low Pressure and Fog ... ... ... ... ... ... ... ... 53 Luke Howard's Nomenclature of Clouds ... .. .. ... 39 Manual, Barometer, for use of Seamen ... ... ... ... ... ... xiv ,, , Fishtry Barometer ... ... ... ... ... .. ... .-■ xiv Maps, Weather ix, s ., , ,, , Subscriptions for xi Marine Barometer ... ... ... ... ... ... ... -•• -.• 1»5 Maritime Congress .. ••• xix Maximum Thermometer ... ... ... ... .. ... 164 Maxwell. Professor J. Clerk, on Regelation 150 „ , ., „ : "Theory of Heat ■• 120 .Menn Rainfall • ••• 58 Measurement of Rainfall ... ... ... ... ... 38 ,, Snow... ... ... ... ... ... ••■ ■• ••. 62 „ Wind Force and Velocity .- xvji Mediterranean, Monthly Meteorological Charts of the North Atlantic .md ... .Kill Mercury, Action of, in Barometer ... ... U'J „ Barometer ver>us Aneroid ... ... ... ■.- ^ Meteorological Charts, Monthly, of the North Atlantic and Mediterranean, and Indian Oceans xiii ,, Observations, Increasing Importance of, to Seamen xiii „ Office X, xii, xiii, xix, xxiii, xxvi, 4, 63, 6C>, SS, 168, 174. 185 „ ., , its purpose ••• vii. xii „ ,, Pattern Raingauges 1'3 Method of reading Barometer 151, 1.t3 „ using Barometer, New ... .. ... ••• ••• ••• x, xiv „ ., " .. .Old viii Metric Units and Education ... ... ... ... ... ... ... ■•• * n07it H 196 Pagre. Micro-Thermometer, Barnes' Recording 123 Millibar, Meaning of ... ... ... ... ... ... ... ... ... l.f> Millimetres in measurement of Rainfnll ... ... ... ... ... .. 9 Minimum Temperature, Greenwich 29 . Kew 29 „ Thermometer ... 16.5 Mist 50 ,, , Distinction between Fog and ... .. ... ... xxvii. 50 .. , Distribution of and Frequency round British Islands .56 ., , Formation of ... ... ... ... ... ... ... ... ... 50 Moisture, Capacity of Air for ... ... ... ... ... ... ... 31 Motion of Air Currents 22,48 Upper Clouds 49 Muslin and Wick for Wet Bulb Thermometef ... ... ... .. ... 164 N. New Method of using Barometer ... ... ... ... ... ... ... x, xiv Nomenclature of Clouds, Luke Howard's ... ... ... ... ... ... 39 Norden?kiold, 0. ; Swedish Expedition ... ... ... .. ... ... 139 Xorth Atlantic and Mediterranean Charts, Monthly Meteorolo.^ical, of the ... xiii ., ,, , Frequency of Ice. in ... ... ... ... 133 Xorth-ea¥terl'. Type of Weather ... ... ... ... ... .. ... 103 Northerly Type of Weather 103 Northern Hemi'^VJhere, Ice in... ... ... ... ... ... ... ... 127 North-westerly Type of Weather 102 " Observer's Handbook " ... ... ... ... ... xv, xxiii, 172 Office. Meteorolo'/ical x, xii. xiii, xix, xxiii, xxvi. 4,(53, 68, S8. 168, 174, 185 Old Method of using Barometer ... .. ... ... ... ... ... viii Origin of Icebergs .... ... ... ... ... ... ... ... .. 122 „ the term Cyclone .. ... .. ... ... ... 24 Osier's Pressure Plate Anemometer .. 167 P. Pack Ice 123 Pancake Ice ... .. ... ... ... ... ,., ... ... ... 123 Parallax. Errors of ... .. ... ... . . . .. ... ... 151 Paraselense .. , ... 42 Parhelia 42 Path of Centre of Depression 84,86 „ „ Cyclone 47,48 „ „ „ Centre. Dependence of Weather on 86 Periods, Recurrence of Cold and Warm 95 Phenomenal Drifts and Heights of Icebergs ... ... ... .. ... 135 Velocities of Wind bv Anemometer ... ... .. .... sxiv. xxv Piddington " ... ; 24,33 Points of (]!om pass. Relative Frequency of Gales from 112 Polar Gal OS. Proportion of ... ... ... ... ... .. .. ... Ill 197 Page. Portable Barograph 156 Precipitation 5-!. «-J, 73, 8C Prefiaration of Weather Charts ... 68 Pressure, Atmospheric ... ... ... ... ... ... ... 21,32,68 ., , Flnctuatious iu ... ... ... ... ... ... 7,86 .- , msasured by Baromet-.T ... ... ... ... ... 113 . Distribution and Wind Velocity 37 Gradient 33,34 ,, ,, and Wind Velocity ... ... ... ... ... .. 36 ,, • , High, associated with Fog ... ... ... ... ... 52 ,. in Pressure Units ... ... ... ... ... ... ... ... 3 ,, in Millibars, the final result 5,6 , Low. associated with Fog ... ... ... ... . 53 of Gases 29 Plate Anemometer, Osier's ... ... ... .. ... . 167 Tube Anemometer, Dines' ... ... ... ... ...167,170 , Wind ... ... ... ... ... ... ... ... ... XX Prevalence oi Gales 109 Primary Cyclone ... ... ... ... ... ... . -.. 73 Proportion of Icebergs above Water ... ... ... ... ... .. 130 Psychroineter ... ... ... ... ... ... ... ... ... ... 162 Pumping of Barometer ... ... ... ... ... ... ... ... 9 Quadrants, Classification of Gales iu 109 a. Radiation, Heat imparted bj'... ... ... ... ... ... ... ... 25,166 . Solar and Terrestrial 26,27 Radio-telegraphy ... ... ... ... ... ... ... ... ... xiv Rain 57 .. , Cause of 58 ., Gauge, M.O. Pattern 174 ., Snowdon 173 ., Gauges . ... 173 ., Measuring GlasH ... ... ... ... ... ... ... 174 Rainfall Frequency 66 , Mean 58 ,. measurement ... ... ... ... ... ... ... ... 58 Rate of Travel of Storms, Variations in ... ... ... ... ... 118 Reaumur Thermometer Scale ... ... ... ... •• ... ■ 160 Recurrences of Cold and Warm periods ... ... ... ... ... ... 95 Reduction of Barometer Readings to Standard Latitude ...146,147 ,. ., ., Temperature ... ... ... 145 Refraction of Light 175 Regelation 60,120 „ ; Sir J. Dewar on ... ... ... ... ... ... ... 120 ., ; Discovery of, by Faraday 120 ., ; Professor J. Clerk Maxwell on 120 ,, ; Professor Tyndall's Application of, in Glaciers 121 Region of Glaciers 121 Relative Frequency of Gales from Points of Compass 112 ., ,. „ ,, „ „ „ ,, on Coasts of British Isles 113, 114, 116 Humidity 31 Reports, Daily Weather ... ... ... ... ... ... ... ... 176 Reports of Icebergs, Numbers of ... ... ... ... 142 1 i)8 Page. Ridge 71,85 Hmk, Dr. ; on Ice 127 Rivers and G-laciers, Analugy bet.ween ... . ... ... ... ... 121 Robinson, Dr. : Anemometer ... ... ... ... ... ... ... 168 Ross, Ice Di.sGoveries of ... ... ... ... ... ... ... ...137, 138 Rotatiou of Air 20 ., ,. Earth, Effect of, on Wind 22 ••Royal Charter" Stnrra xiii Ruesell, Hon. Rollo, on H.iil .. 59 S. Sail, Chang'e from, to Steam ... ... ... ... ... ... ... ... xiii •■ Sailor's Horn Book"... ... ... ... ... ... ... ... ... 24 Satellite Depressions 16,73 oaturaticu of Air 31 Scale of Barometer H3 ,, ,, Fog Intensity... ... ... ... ... ... ... .57 Scott, Captain R. F., C.V.O., R.N., Formation of Antarctic Icebergs 140 Scout Ship xxix Scud 42 Sea Barometer ... ... ... ... ... ... ... ... 155 „ Fog 50, 53, 54 „ Surface Temperature 166 ,, Temperature, Professor H. T. Barnes on xv „ ,. , Effect on by Ice xv ,, Water Thermometer ... ... ... ... ... ... ... ... 166 Seamen. Barometer Manual for the Use of... ... ... ... ... ... xiv ,, . Increasing importance of Meteorological Observations to ... ... xiv Seamen's Weather ... ... ... ... ... ... ... ... ... xv Seasons, The 28 Seasonal Range of Temperature ... ... ... ... ... ... ... 28 Secondary Cyclone 73, 74 „ Wind Systems 73 Shaw, Sir Napier, Sc.D., F.R.S. ; Barometric Gradient 38 ., ; Cyclonic Winds 37 „ ; Ice Crystals ... ... ... ... ... 36 „ ; Winds, Surface Friction of 73 Simjjlest form of Anemometer ... ... ... ... ... 121 Simultaneous Readings of Barometer ... ... ... ... ... ... 32 Sleet 60.62 Slob Ice 123 Sludge Ice 123 Snow xxvii, 57, 60 ,, Crj'stals 60, Gl „ „ , Twenty Years' Study of , Bentley, W^ A. 61 „ frequency 67 ,. , how measured 62 ,', line 119 Suowdon Raingauge 173 Soil, Heat and the ... ... ... ... ... ... ... ... ... 26 Solar and Terrestrial Radiation 26,27 South-easterly type of Weather 106 Southerly tj'pe of Weather 100 Southern Hemisphere, Ice in... ... .. 136 ,, ,, , Phenomenal Heights of Icebergs of the ... ... 136 ,, Ocean, Frequency of Ice in ... ... ... ... 141 South-westerly type of Weather 98 Speed and Height of Clouds ' 43,46 ,, , Exceptional, of Cyclones ... ... ... ... 119 ,, , Seasonal, of Clouds ... 44 Spirit Thermometer ., 165 Squalls 75 „ and Weather 76 „ , Line 77 199 Standard Gravity, Correction of Barometer for ,, Temperature Stevenson's Thermometer Screen Storm Field, Definition of ,, Signal Stations ... StrucLure of Ice ,, Wind Study of Weather Maps Summer, Indian, of Eastern Canada ,, , St. Luke's ... Sun's Heat, Transmission of .. Suspension of Barometer Swedish Expedition ; O. N'orclenskioid Swirls, A.ir ... Synoptic Charts Page. lift .5,11 166 48 179 119 xzii ix, z 96 96 2.5, 26 l.iO 139 ix, xii (J9 T. Taylor, Major >. I., R.F.C.. Formation of Fog and Mist Telegraph, Weather Reports by Temperature, Absolute ,. . Air „ , Ciiuses of difference in ,, , Decrease of, with Evaporation ,, , ,. „ Avith Height „ , Definition of ... „ , Diurnal Variation of ... ., . Effect of. on Humidity „ terrestrial Radiation on ... ,. , Fiducial .. ,. , Greenwich Maximum and Minimum ,, , Highest 3Iean Daily ... ,. , Kew Maximum and Minimum , Latitude and, Connexion between ... „ , Lowest Mean Daily at Greenwich ,, , Lowest I ossible „ . Modified by character of soil ,, , ,, ,. neighbourhood of water ,, , „ ., Winds and Ocean Currents . . „ , Reduction of Barometer Readings to Staudard ,, . Sea, Professor H. T. Barnes on „ , Sea Surface „ , Seasonal Range of ,. , Standard Temiieratures in Anticyclone I'errestrial Radiation ... " Theory of Heat," Maxwell's Thermograph Comparison between, and SCiuidard Thermometer Thermometer Attached Bulb, Gradual Contraction of . Dry and Wet Bulbs , Maximum , Minimum Scale, ,. , Centigrade .. . Fahrenheit ., , Graduations of ., , Reaumur Self-recording... Screen, Stevenson's ... , Sea Water , Spirit 162, 4 24, 31 22 30 27, .58 24 28,29 L».i. 31 27 11,16 29 28 29 26 28 .1.59, 160 26 26 26 14;5 XV 166 28 5,11 89,90 27 120 160 161 169 11,144 160 .162,164 164 16.5 1.59 160 160 159 160 162 16S. 169 166 166 200 Page. Thick Weather and Wind Direction xxvii.xxviii.5fi „ ,. , First Appearance of Iceberg in 124 Thinking in Maps s ,. ,, ,, Consequences of ... ... .. ... ... 3 Thunderstorms cominjj from South 107 „ connected with Northern Depressions ... ... ... ... 108 „ form ng local] y 108 ,, .Typical 107 " Titanic " Disaster xiv, xxviii, 135 Torricellian Va;:;uum 143 Tirricelii's Experiment 143 Toynbee. Captain H. ; Barometric Gradient? 36 Tracks of Storm Centres 116,117 Transmission of Sun's Heat ... ... ... ... ... ... ... ... 25,26 Travel of Cyclone 24 Trend of Isobars and Wind 33 „ Isotherms. Air and Sea. in Fog .. ... ... ... ... ... 54 Trough of Depression ... ... ... ... ... ... ... 48,71 Tyndall, Professor, on Ice Structure 119 Types of VVerUher Conditions... ... ... ... ... ... ... 98 U. United Kingdom. Average Wind Frequency ... ... ... ... ... xsvii „ „ . Gales on Coasts of ... ... ... ... ... ... 108 United States Expedition under Wilkes ... ... ... ... ... ... 138 Units, Absolute ... ■ ... ... ... ... ... ... 144 Upper Clouds, Motion of ... ... ... ... ... ... ... ... 49 Use of Barometer Readings at Central Office ... ... ... x ,, „ ,, ., out of reach of Central Office xi „ „ Isobars 38 Utility, Practical, of change of graduation in Barometers .. 8 " V "-shaped Depressions^ ... ... ... ... ... ... 75,85 Vacuum, Torricellian ... ... ... ... ... ... ... ... ... 143 Vaporisation ... ... ... ... ... .. ... ... ... ... 29 Vapour, Elastic Force of 31 ,, , Water, Behaviour of 32 „ , „ , in the Air 21 Variations in Rate of Travel of Cyclones ... 118 Veering of Wind ... ... ... ... ... ... ... ... ... 72 Velocities of Wind. Phenomenal, by Anemometer ... ... ... xxiv, xxv Velocity of Wind xvi,xvii ,. . Wind Force and. Measurement of ... ... ... ... ... xviii Vernier' 144,147,151 Victoria Land, Ice Barrier ... ... ... ... ... ... 137 Vortex of Cyclone ix Vortical Gales ... ... ... ... ... ... ... ... Ill W. Warm Periods, Recurrence of Cold and ... ... ... ... ... ... 95 201 Page. Warnings, Storm 176 Water, Capacity of , for Heat 26 Water Vapour, Behaviour of 32 „ „ in Air 21 Weather Changes and Cirriforni Cloud Motion 91 Chart, Preparation of x, 68, 69 Charts, daily sj'Qoptic 178 ,; Conditi'^ns and Ice Drift ... ... ... ... ... ... 135 Conferences, International xiv ,, , Easterly Type of 104 „ Forecastiagr ... 82,176 „ Glass viii, x, 143 ., , Ice Movements and. Connexion between ... ... ... xxix, 130 „ . Letters CO Indicate State of 68 ,, Maps, Preparation of x, 68, 69 ,. , Study of ... ... ... ... ... ... ... ... X, xi ,, . Weather Reports and ... ... ... ... ... ... 178 , North-easterlv Type of ... 103 „ . Northerly " ' ., ... 103 ,. . North-%ve3terly „ 102 , Path of Cyclone Centre, and 86 Reports, Daily ... 176 „ , Weekly 178 „ Seamen's xv „ Sequence of Depre?slon ^.5,86 ., Signs 94 ,, , South-easterly Type of 106 , Southerly Type of 100 ,. , South-westerly Type of ... ... ... ... ... ... .. 98 , Squalls and 76 .Thick... ... ... ... ... ... ... ... xxvii, xxviii ., , TypesJ and Conditions of " 98 ,, , Westerly Type of 101 Weekly Weather Report. Gusts Recorded on ... ... ... ... ... xxvi Wedge, The ' 71 Westerly Type of Weather 101 Wet and Dry Bulb Hygrometer 162 „ Bulb, Action of 162 „ „ , Depression of 162 „ ,, , Mus^lin and Wick for .. 164 ,, ,, Thermometers, Dry and 162,164 Wilkes, Captain C, U.SX,, Formation of Antarctic Icebergs 138 „ ,, , United States Expedition under 138 Wind xvi ., , Backing of 72 ., , Cause of 21 ., Circulation, Anticyclonic ... ... ... ... ... ... ... 23,34 „ , Cychmic 23,34 ,. , Cyclonic Backing of ' 49,72 ,. Destrnctiveiiess... ... ... ... ... ... ... .. ... xvi „ Direction in Gales Ill ,, „ , Relation of, to Fog 66,57 ., , Effect of Earth's RotaCion on ... 22 „ Force and Velocity, Measurement of ... ... ... ... ... xviii Frequency xxvi, xxvii . Average, United Kingdom ... .. ... ... ... xxvii Gradient 36 „ Measurements ... ... ... ... ... .. ... ... ... xvii ., , Phenomenal Velocities of, by Anemometer ... ... ... xxiv, xxv Pressure ... ... ... ... ... ... ... xix Scale, Admiral Beaufort's xviii, xix, xx, xxi, xxii . . Structure of ... ... ... ... ... ... ... xxii Systeius. Secondary ... ... ... ... ... ... ... ... 73 ., , Thick We.ither and ... ... ... ... ... ... ... xxvii, xxviii ., , Trend of Isobars and 33 ,. .Veering of ... ... ... ... ... ... ... 72 Velocity xvi, xvii ,. , Calculation of 37 ,, ., , Connexion of. with Pressure Gradient 36 Winds. Cyclonic ; Dr. W. X. Shaw. F.R.S.. on 37 2iY2 Pasre. Winds, Surface Friction of ; Dr. W, X. Shaw, F.R.S., on 38 Wintered 82 Wireless Weather Reports from H.M. Navy 176,178 „ „ „ Mercantile Marine 176,178 Z. Zero. Abwlute. Lord Kelvin ... ... ... ... 160 Printed under the authority of His [Majesty's Stationery Office By DARLING and SOk, Limited, Bacon Street, e'2. 3D55 ' QC Gt.Brit. Meteorological Office - The 863 Seaman's handbook of meteorology. V^*5^ PTO UNIVERSITY OF CALIFORNIA LIBRARY LJ-^ 1 7S j^jjg Angeles '^ This book is DUE on the last date stamped below. MAY 2 71966 l«ftY2 7RErt^ n&v r\t,\] * ", M DEC 519B8 A 000 437 207 4 f l?<