UW NKLr METEOROLOGICAL GLOSSARY FFICE. (Fourth Issue) In continuation of The Weather Map, (M.O. 225 i). fcg tf)c authority of tfje iJHetforologiral Committrr. LONDON: N-i ra ' " ins MA.)! - ; i . LIMITED, J. AND .'JIOM | M OEOLOGHCAL 1018, ' -. m*i. M.O. 225 ii. METEOROLOGICAL OFFICE. METEOROLOGICAL GLOSSARY (Fourth Issue) In continuation of The Weather Map, (M.O. 225!,) tg ifje &tttf)orttg of tije jj&eteoro logical Committee. LONDON : PBINTED UNDER THE AUTHORITY OF HIS MAJESTY'S STATIONERY OFFICE Bi DARLING AND SON, LIMITED, ^ACON STREET. E.2. AND TO BE PURCHASED FROM TH-E METEOROLOGICAL OFFICE, EXHIBITION ROAD, LONDON, S.WJ, Price Is. net. For units or systems of units.* F.P.S. is equivalent to foot-pound-second system of units. C.G.S ,, centimetre-gramme-second system of units. B.T.U. ., British Thermal units. .The following are also used : For the expression of Abbrevia- tion. Meaning. Angle O 1 II Degrees, minutes, seconds of arc. Density g/m 3 g/cc Ib/cu. ft. Grammes per cubic metre. Grammes per cubic centimetre. Pounds per cubic foot. Length mm cm m k Millimetre. Centimetre. Metre. Kilometre. Mass ... & Gramme. Kilogramme. Pressure mb cb Millibar. Centibar. Temperature a C F Absolute scale of temperature.* Centigrade scale of temperature. Fahrenheit scale of temperature. Velocity . rn/s mi/hr Metres per second. Miles per hour. Volume cc m 3 Cubic centimetre. Cubic metre. * The abbreviation "a" is al^o us'-vi '-o represent a unit or degree of temperature en the absolute scale. Fov aa iacerval of temperature la is the same as 1 0. METEOROLOGICAL GLOSSARY. TABLE OF CONTENTS. The entries marked (I) are also referred to in The Weather Map. (M.U. 225 i.) Page. Page. Absolute Extremes ... 12 Aneroid barometer .. 30 Absolute Humidity ... 290 Aneroidograph 30 Absolute Temperature 12 Anthelion 30 Accumulated Temper- 293 Anticyclone (I) 30 ature. Aqueous-vapour 31 Actinometer ... 14 Atmosphere (I) 33 Adiabatic 15 Atmospheric Electri- 294 Aerology ... 16 city. Aeroplane Weather ... 16 Audibility 33 Air(i) 19 Aureole 36 Air-Meter 20 Aurora... 298 Airship- Weather 20 Autumn 36 Altimeter 27 Average 37 Altitude 28 Azimuth 37 Alto-cumulus ... 28 Alto-stratus 28 Backing 37 Anabatic 29 Ballon sonde ... 39 Anemobiagraph 29 Balloon Kite ... 42 Anemogram .,. 29 Bar 43 Anemograph ... 29 Barogram 43 Anemometer ... 29 Barograph 43 Anemoscope ... 29 Barometer 43 (13204r 12.) Wt. 26779464. 7000. 3/18. D & S. (S.) Gr. 3. 423582 Glossary. I 'age. i Page. Barometric Tendency 44 Condensation ... ... 70 Beaufort Notation (J ) 44 Conduction ... 71 Beaufort Scale (1) 45 Contingency ... ... 302 Bishop's Ring ... 46 Convection ... 71 Blizzard 46 Corona ... ... ... 71 Blue of the Sky 46 Correction , ... 12 Boiling Points 300 Correlation ... 74 Bora 47 Correlation Ratio ... 302 Breeze ... 48 Cosecant ... 76 Brontometer ... 49 Cosine ... ... 76 Buoyancy 49 Cotangent ... 76 Buys Ballot's Law (I) 57 Counter Sun ... ... 76 Cumulo-stratus ... 77 O.G.S 57 Cumulus ... 77 Calm 57 Cyclone (I) ... 77 Calorie... 301 Cyclostrophic ... ... 77 Celsius ... 57 Centibar 58 Damp Air ... 77 Centigrade 58 Day Breeze ... 77 Centimetre 58 Debacle - ... 77 Cirro-cumulus 58 Dekad ... 78 Cirro-stratus ... 58 Density... ... 78 Cirrus ..- 58 Depression ... 82 Climate 58 Dew ... 82 Climatic Chart 59 Dew-point ... 82 Climatology 60 Diathermancy ... 82 Clouds 60 Diffraction ... 83 Cloud Burst 67 Diffusion ... 84 Clouds, Weight of .. 67 Diurnal ... *5 Col 68 Doldrums ... 88 Compass 68 Drought ... 88 Component 69 Dry Air ... 88 Table of Contents. Page, i Page. Dry Bulb ... 88 Forecast ... 117 Duration of Rainfall ... 303 Freezing ... 118 Dust ... 304 Frequency ... 118 Dust-counter ... ... 305 Friction ... 122 Dynamics ... 89 Frost ... 123 Dynamic* Cooling ... 89 Gale ... ... 124 Earth Thermometer ... 89 Gale Warning... ... 128 Eddy ... 89 Gas ... 129 Electrification of Water- 337 Geostrophic ... 129 drops. Glacier ... ... 307 Electrometer ... ... 91 Glazed Frost ... ... 129 Energy ... 9* Glory ... ... 130 Entropy ... 94 Gradient ... 130 Equation of Time ... 100 Gradient Wind ... 134 Equator ... 100 Gramme ... 138 Equatorial ... 101 Grass Temperature ... 139 Equilibrium ... ... 102 Gravity ... 308 Equinox ... 103 Great Circle ... ... 139 Error ... ... 104 Gulf Stream ... ... 139 Evaporation ... ... 106 Gust ... 140 Expansion ... 109 Gustiness ... 142 Exposure ... 109 Extremes ... 110 Hail ... 142 Halo ... 143 Fahrenheit ... 110 Harmattaii ... 145 Fall ... Ill Harmonic Analysis ... 311 False Cirrus ... ... 306 Haze ... 145 Fluid ... Ill Heat ... 145 Fog ... 112 High (I) ... 152 Fog Bow ... 116 Hoar Frost ... 152 Fohn ... 115 Horizontal ... 152 (jrlossary. Page. | Page. Horse Latitudes ... 153 I Katabatic 182 Humidity (I) ... 154 i Khamsin 182 Hurricane ... 155 Kilometre 183 Hydrometer . . . ... 158 Lake 183 Hydrosphere ... ... 158 Land Breeze 1*3 Hyetograph ... 158 Lapse ... 183 Hygrograph ... ... 158 Lenticular 185 Hygrometer ... ... 159 Level ... 185 Hygroscope ... 159 Lightning 186 Hypsometer ... ... 160 Protection against 324 Line Squall 188 Ice ... . 321 Liquid 189 Iceberg ... 161 Low (I) 189 Incandescence ... 161 Lunar ... 169 Index ... ... 162 Index Error ... ... 162 Mackerel Sky ... 190 Insolation ... 162 Magnetic Needle 190 Inversion ... 163 Magnetism 327 Ion ... 164 Mammato-Cumulus ... 190 lonisation ... 322 Mures 1 Tails ... 191 Iridescence ... 166 Maximum 191 Irisation ... 166 Mean ... 191 Isabnormals ... ... J66 Meniscus 191 Isanomalies ... 166 Mercury 192 Isentropic ... 167 Meteor ... 192 Iso ... 168 Meteorograph ... 193 Isobars ... ... 168 Meteorology ... 193 Isohels 1 Metre ... 193 Isohyets 1 g Tcsn Microbarograph 193 Isopleths [ k JLbU. Millibar 194 Isotherm J Millimetre 195 Isothermal . 181 Minimum 195 Table of Contents. Page. Page. Mirage ... ... 195 Pocky cloud 212 Mist ... 196 Polar 212 Mistral ... .., 197 Pole 212 Mock San ...'197 Potential 213 Mock Sun Ring ... 197 Potential Temperature 213 Monsoon ... 197 Precipitation 329 Moon ... 198 Pressure ... ... 213 Prevailing winds ... 213 Nadir . 198 Probability 214 Nephoscope ... 198 Prognostics ... ... 215 Nimbus ... 198 Psychrometer 216 Normal ... 198 Pumpimr 216 Purple Light ... ... 217 Observer ... 203 Pyrheliometer ... 217 Ombrometer ... ... 203 Orientation ... 203 Radiation 330 Orographic Rain ... 203 Rain ... 217 Ozone ... 204 Hainband . ... 218 Rainbow .. ' ... 218 Rainday 219 Pampero . . . 204 Raindrops, size of, &c. 334 Paranthelion . . ... 204 Rainfall ., ... 219 Paraselenae . . ... 204 Rainfall, duration of . . . 303 Parhelia , . ... 204 Raingauge 219 Pentad ... 205 Rain-spell ... ... 219 Periodical ... 205 Reaumur 219 Persistence . . ... 206 Reduction 220 Persistent Rain ... 206 Reduction to Sea Level 220 Phases of the Moon ... 208 Refraction 222 Phenology ... 210 Registering balloon ... 223 Pilot-balloon ... ... 210 Regression-equation ... 339 Pluviograph ... ... 211 Relative humidity ... 223 Pluviometer ... . 212 Reversal 224 Glossary. Page. Page. Ridge 225 Solar Radiation Ther- 238 Rime ... 225 mometer. River 225 Solstice ... 238 Roaring Forties 226 Sounding ... 238 Spells of Weather ... 238 Spring ... ... 239 St. Elmo's Fire 226 Squall .,. 239 Saturation 226 Stability ... 239 Scotch Mist 340 Standard Time ... 239 Screen 226 State of the Sky ... 239 Scud 226 Statics .. 240 Sea-breeze 227 Station ... 240 Sea-level 227 Statoscope ... 240 Seasons 227 Storm ... ... 241 Secant ... 231 Storm Cone ... 241 Secondary 231 Strato-cumulus ... 241 Seismograph ... 231 Stratosphere ... ... 241 Serein ... 231 Stratus ... 241 Shamal 231 Summer ... 241 Shepherd of Banbury 232 Sun ... 241 Silver Thaw 235 Sun-dial ... 343 Simoon 235 Sun-dogs ... 242 Sine 235 Sun Pillar ... 242 Sine curve 236 Sunset Colours ... 242 Sirocco 237 Sunshine ... 242 Sleet 237 Sunshine-recorder ... 243 Snow ... 237 Sun spot Numbers ... 243 Snow crystals 237 Surge ... 244 Soft hail 343 Synoptic ... 244 Solar Constant 237 Solar Day 238 Tangent ... 244 Solarisation 238 Temperature (I) ... 244 Table of Contents. Page. Page. Temperature Gradient 246 Vapour Tension ... 266 Tension of Vapour .... 247 Vector ... ... 266 Terrestrial 247 Veering ... 267 Terrestrial Magnetism 327 Velocity ... 267 Thaw 247 Vernier ... 268 Thermodynamics ... 247 Vertical ... 268 Therm ogram ... ... 247 Viscosity .,, 269 Thermograph 247 Visibility ... 269 Thermometer 248 Vortex ... 347 Thunder 248 Thunderstorm ... 249 Time ... 252 Water ... 271 Tornado 252 Water-Atmosphere ... 273 Torricelli 253 Waterspout ... 274 Trade Winds 253 Water-Vapour ... 274 Trajectory ... ... 262 Waves ... ... 274 Tramontana 263 Waves of Explosion ... 276 Transparency ... ... 263 Weather (I) ... ... 276 Tropic ... ... ... 263 Weather maxim ... 276 Tropical 263 Wedge ... 279 Tropopause ... ... 263 Weight ... 279 Troposphere 263 Wet Bulb ... 279 Trough 263 Whirlwind ... 280 Twilight 344 Wind ... 280 Twilight Arch ... 264 Wind Rose ... 351 Type 264 Winter ... ... 288 Typhoon 265 Wireless Telegraphy ... 288 Upbank Thaw 265 Zenith ... ...288 V-Shaped depression... 266 Zodiac ... ... 289 Vapour Pressure ... 266 j Zodiacal Light ... 289 10 TABLE OF CONTENTS OF THE WEATHER MAP. AN INTRODUCTION TO THE GLOSSARY. (Now issued as a separate volume.) PAGE. Meteorology and Military Operations 3 Weather Records nd Climate 5 The Necf ssity for Forecasts of Weather ... 7 Modern Meteorology the Work of an Organisation, not of an Individual ... 8 The Meteorologist at Headquarters ... ... ... ... 8 A Map of the Weather ... 9 The Beaufort Notation ... .., 10 A Map of the Winds 12 The Beaufort Scale 13-14 The Atmosphere . , 15 Water Vapour : Evaporation and Condensation .,. ... 17 Temperature and Humidity , 19 A Map of Temperature Pressure and Its Measurement 21 The Barometer 23 A Map of the Distribution of Pressure Isobars ... ... ... ... ... ... ... ... 25 A Map of all the Elements together 26 Lessons from Weather Maps 26 Buys-Ballot's Law and the General Relation of Pressure to Wind .., 26 Weather and Temperature 30 The Sequence of Weather 31 The Travel of the Centres of Cyclonic Depressions ... 35 Barometric Tendency ... ... ... ... ... ... 41 Veering and Backing of Wind 42 Types of Pressure Distribution . 43 The Upper Air. The Dynamics and Physics of the Atmosphere ... ... ... ... ... ... ... 44 Distribution of Rain and Cloud in Cyclonic Depressions... 50 Climatic Supplement, Charts and Diagrams ... 55 11 METEOROLOGICAL GLOSSARY. CONTAINING INFORMATION IN EXPLANATION OF TECHNICAL METEOROLOGICAL TERMS. Details as to the use of meteorological instruments are given in' the Observers Handbook, and as to the numerical computations in the Computer's Handbook, to both of which reference is made in the Glossary when required. In accordance with the practice of the Oxford Diction- ary the initial word of each article is in black type, and words in the body of the text are printed in small capitals when they are the subjects of articles in another part of the glossary. For the articles in this glossary I am principally indebted to the Staff of the Observatory at Benson, W. H. Dines, F.R.S., and E. V. Newnham, B.Sc. ; and of the Branch Office at South Farnborough, Captain C. J. P. Cave, R.E., and R. A. Watson Watt, B.Sc., with Major Taylor, Professor of Meteorology, R.F.C. ; and to the staff of the Forecast Division, especially F. J. Brodie, E. L. Hawke, B.A., and Second Lieutenant T. Harris, R.E., who have passed the MS. through the press, and W. Hayes who prepared many of the illustrations. The revision of the work for the present fourth issue has been carried out by Dr. C. Chree, F.R.S., Superintendent of the Observatory, Richmond. Some new articles have been added at the end of the volume, pp. 290-354. NAPIER SHAW. Meteorological Office, 26 October, 1917. 12 METEOROLOGICAL GLOSSARY. Absolute Extremes. The word extreme is often used with reference to temperature to denote the highest and lowest temperatures recorded at an observing station in the course of time. As the observations are generally summarised for a year, extreme temperatures come to mean the highest and lowest temperatures of a year. When the survey is taken over a longer period, 10 years, 20 years, 35 years or 200 years, according to the duration of the observations, the highest and lowest temperatures observed during the whole period are called the absolute extremes. The journalistic expression would be u the record," high or low as the case may be. The absolute extremes for the British Isles are, highest 310'8a., (100 F.) ; lowest 245'8a., (-17 F.). Those for Belgium, highest 311'2a., (]QO-8 F.) ; lowest 243-2a., (-21-6 F.). The Surface of the Globe, highest 329-7a., (134 F.) ; lowest 203'2a., (-93-6 F.). In the Upper Air, lowest 182'Ja., (- 131'6 F.) at a height of 16^ km. over Java. Absolute Temperature. The temperature of the centigrade thermometer, increased by 273, more properly called the temperature on the absolute, or thermodynamic scale. The absolute scale is formulated by reasoning about the production of mechanical work at the expense of heat (which is the special province of the science of thermodynamics, see ENTROPY), but for practical purposes the scale may be taken as identical with that based on the change of volume and pressure of one of the per- manent gases with l^eat. For thermometric purposes aiming at the highest degree of refinement the hydrogen scale is used, but for the purposes of meteorological Absolute Temperature. Ui reckoning the differences of behaviour of the permanent gases, Hydrogen, Oxygen, Nitrogen are unimportant. In physical* calculations for meteorological purposes the absolute is ihe natural scale ; the densities of air at any two temperatures on the absolute scale are inversely proportional to the temperatures. .Thus the common formula for a gas, P _ ' Po ^ P (273 + " PO C^ 73 + *o) where p is the pressure, p the density and t the tempera- ture Centigrade of the gas at one time, p Q , p , to the corre- sponding values at another, becomes JL ffo where T and Jo are the temperatures on the absolute centi- grade scale. Its most important feature for practical meteor- ology is that from its definition there can be no negative temperatures. The zero of the absolute scale is the tempera- ture afc which all that we call heat would have been spent. In the centigrade scale all temperatures below the freezing point of water have to be prefixed by the negative sign . This is very inconvenient, especially for recording obser- vations in the upper air, which never gives tsmperatures above the freezing point in our latitudes at much above 4 kilometres (13,000 feet), and often gives temperatures below the freezing point nearer the surface. The absolute temperature comes into meteorology in other wavs ; for example, the rate at which heat goes out into space from the earth depends, according to Stefan's Law, upon the fourth power of the absolute temperature of the radiant substance. See RADIATION. 14 Glossary, Temperatures can also be m expressed in an absolute scale of Fahrenheit degrees of 'which the zero is approxi- mately 45',' below the Fahrenheit zero. Some common temperatures on the absolute scales and their equivalents in Centigrade and Fahrenheit are : Centigrade. Fahrenheit. af. The boiling point of 4 269 -452-2 7-2 helium. The boiling point of 77 -196 -320-8 138-6 nitrogen. The freezing point of 234-2 -38-8 -37-8 421-6 mercury. The freezing point of 273 32 491-4 water. The mean temperature 282-7 9'7 49'5 508-9 of London. " Temperate " as shown 285-8 12-8 55 5H'4 on an ordinary ther- mometer. The best temperature 290 17 62-6 522-0 for a living 1 room. A hot summer day 300 27 80-6 540-0 The temperature of the 3io 37 98-6 558'o human body. The temperature of the 6,000 10.000 Sun. Actinometer. An instrument for measuring the intensity of RADIATION received from the sun. In Michelson's Actinometer, for example, the essential element consists of two strips of different metals fixed together. These are heated by the solar radiation which they absorb, and the amount of bending which results Actinometer. 15 from their unequal expansion is a measure of the rate at which they are receiving radiant energy. AdiabatiC The word which is applied in the science of thermodynamics to the corresponding changes which may take place in the pressure and density of a substance when no heat can be communicated to it or withdrawn from it. In ordinary life we are accustomed to consider that when the temperature of a body rises it is because it takes in heat from a fire, from the sun or from some other source, but in the science of thermodynamics it is found to be best to consider the changes which occur when a sub- stance is compressed or expanded without any possibility of heat getting to it or away from it. In the atmosphere such a state of things is practically realised in the interior of a mass of air which is rising to a position of lower pres- sure, or sinking to one of higher pressure. There is, in consequence, a change of temperature which is called mechanical or dynamical, and which must be regarded as one of the most vital of meteorological phenomena because it accounts largely for the formation and disappearance of cloud, and probably for the whole of our rainfall. Tyndall' illustrated the change of temperature due to sudden compression by pushing in the piston of a closed glass syringe and thus igniting a piece of tinder in the syringe. The heafcing of a bicycle pump is a common experience due to the same cause.* On the other hand the refrigeration of air is, often obtained simply by ex- pansion, particularly in the free atmosphere. * Dangerous heating- may result on firing a gun from the sudden compression of gas within the bursting charge of the shell if there are cavities in the explosive therein. 16 Glossary. To plan out the changes of temperature of a substance under compression and rarefaction alone, we have to suppose the substance enclosed in a case impermeable to heat the word adiabatic has been coined to denote impermeable to heat in that sense. The changes of temperature thereby produced are very great, for example : For adiabatic change of pressure decreasing 1 from 1000 nib. by The fall of temperature from 29oa, 62-6 F.. is nib. in. <J. F. 10 c r 0*30 0-9 c r 1*6 100 , 2'95 8-7 I5'7 200 S'9 1 18-2 32-8 300 8-86 28-4 51*1 400 11-81 39*9 71-8 500 14-77 52-8 95*o 600 17-72 67-6 121*7 700 20*67 85-5 153-9 800 23-62 108-1 , 194-6 900 26-58 I4 1 "3 > 254-3 Aerology. The study of the free air; a word that has come into use recently to indicate that part of meteorology which is concerned with the study of the upper air. Some of the results are given under BALLON-SO.NDE and PILOT-BALLOON. Aeroplane weather. The weather most suitable for aeroplanes is calm clear weather with little or no wind. The only conditions which make it impossible for a good pilot to fly a modern aeroplane are a strong To face p. 16. Diagram showing the pressure in the upper air corresponding with the standard pressure (1013-2 mb.) at the surface and adiabatic lines for saturated air referred to height and temperature. (From Neuhoff Smithsonian Miscellaneous Collections, Vol. 51, No. 4, 1910.) The pressure is shown by full lines crossing the diagram, and the adia- batic lines for saturated air by dotted lines. Temperatures are given in the absolute scale. The short full lines between the ground and the level of 1,000 metres show the direction of the adiabatic lines for dry air. Aeroplane Weather. 17 GALE or a FOG. On the other hand many weather con- ditions may prevent useful work from being done by an aeroplane when it is in the air. The weather affects civilian and military flying in quite different ways. When testing aeroplanes, with a view to finding their rate of climb, top speed when flying level, landing speed, or other aerodynamical quantities it is usually necessary to choose a calm day, when eddies, or large ascending or descending currents, or other conditions prejudicial to accurate testing, are unHkely to occur. In flying across country the chief danger is that the engine will stop when the aeroplane is over ground on which it is impossible to land. When the engine has stopped the aeroplane must come down somewhere in- side a circle whose radius is about equal to five times the height of the aeroplane above the ground. An aeroplane flying at a height of one mile will have an area of about 75 square miles in which it may choose its landing ground, while at a height of 2,000 feet on a calm day the machine has less than 12 square miles to choose from. In England it is almost always possible to pick out a possible landing ground in a circle containing 75 square miles ; but it is frequently impossible to do so in an area of 12 square miles. For this reason clouds below 6,000 feet are one of the chief dangers of cross-country flying, and the lower they are the more dangerous they become. In flying under war conditions eddies and vertical currents are almost immaterial provided they are not so violent as to impede observations. On the other hand low clouds make observations over an enemy's lines almost impossible, owing to the accuracy of modern anti-aircraft gnus. Detached clouds iiupede, but do not put a stop to, reconnaissance. On days when the clouds are too low to 20 Glossary : Lithoplate IX. the upper regions, collections of water globules which are called clouds ; in meteorology the dust is regarded as an impurity, and the clouds as an addition to the air, not part of it. The dust, though an impurity, is important, as it makes the formation of cloud and rain possible when- ever the temperature of a mass of air gets below the DEW-POINT. Air-meter. The name given to an apparatus designed to measure the flow of air. It consists of a light wheel with inclined vanes carried by the spokes, and a set -of counting dials to show the number of revolutions of the wheel. Its accuracy can be tested on a whirling table. As generally sold it is the most portable form of ANEMO- METER, its box not being more than four inches each way. But it cannot be used with success by a careless observer. Airship-weather. Favourable weather.- The most favourable conditions for airships are calms or light airs, with good seeing from above, extending over the whole area to be traversed, and persistent for the whole period, say 24 hours. Detached low clouds may be an advantage, but precipitation, whether in the form of rain, snow or hail, would spoil the occasion. The favourable conditions thus defined are characteristic of the central part of an area of high barometric pressure which, in technical language, is called an ANTICYCLONE. Thus, for operating across the North Sea, the primary meteorological conditions will be favourable when the barometer readings at Helder, Yarmouth and Grisnez are higher than those at surround- ing places, because the three places named will then be in the central region of an anticyclone, cf. Fig. 3, p. 25. / During an anticyclone the pressure is, as a rule, above the normal for the place ; thus, pressures above 1,020 AIRSHIP WEATHER. FIG. 1. DISTRIBUTION OF TEMPERATURE, WEATHER, WIND, AND PRESSURE, 6 P.M. 6th SEPTEMBER, 1915. Normal Ttroperature ^^^*^~* ~~ of the Se*tS*pt.*il ISOBARS are drawn for interrals of fire milH- WEATHER. Shown by lh following symbols - bar*. f\ olear sky. f) iky i clouded. bl " rowa the scale 0-12, is iodi- o%Ud by the number of father, l overcatt sky. A rain falling ^ow. A bail* S f OR. Tthwodtr. "K thunderstorm. TEMPERATUR.-OiTen in dfrta Fahrenheit Airship - Weather. 21 millibars (30*1 inches) are generally to be found in anti- cyclones, and this has given rise to the statement that a high barometer in itself indicates favourable conditions for airships and vice versa. This is often, but not always, the case. If one plots the pressure on a map the favour- able area extends outward from the central region where the highest pressures are found until the pressure begins to fall away rapidly, arid then one finds strong winds and, possibly, also rain or snow. An anticyclone is indicated on a map by drawing lines of equal pressure, ISOBARS, which naturally enclose the area of highest pressure. The lines of an anticyclone, generally speaking, run in roughly parallel curves and are easily recognisable on a map, such as the one reproduced here, Fig. 1. The shape is sometimes that of a regular curve, more or less like a circle or an oval, as in this instance, but often it is quite irregular and straggles over a large region. Anywhere near what may be called the top of it, ^.e., the region of highest pressure^ there are calms or light airs. Further away the winds begin to range themselves in circulation round the central region easterly winds on the uouth side, westerly on the north. Further out, as one gets towards the regions of low pres- sure, the winds become brisker, and on the margins of an anticyclone they may be very strong, but on the eastern side they are generally steady, not changeable or squally. It is characteristic of an anticyclone that when it is once set up and well marked it generally lasts two or three days ; sometimes it is persistent for a week or 10 days, occasionally even more. An anticyclone is, in fact, typical of settled weather, and consequently the setting in of a large anticyclone over the area of 22 Glossary : Lithoplate X. operations may be regarded as providing an ample period of favourable weather. An anticyclone in our neighbourhood generally drifts eastward or north-eastward. It has northerly winds on its eastern or front side, so that the setting in of a northerly wind, veering to N.E., generally means that an anticyclone is coming over, and as it will take two or three days at least to pass, the navigator is practically sure of a few days of fair conditions, and while the central region is going over and the wind is slacking down from north-east to calm, and then changing to south or south-west, there is practically certain to be a perfect day, possibly two or three, for operating an airship. All that an airship navigator has to do, therefore, to hit upon a favourable time fo/ a raid is to choose the occasion when there is an anticyclone with its central area over southern England or the Channel, advancing slowly east- ward, as most of them do. He may then reckon on two or three days' favourable weather, and if he watches the map may extend his forecast day by day as he notes the behaviour of the anticyclone. An anticyclone is such a well-recognised creature in meteorological maps that observations from one half of it are sufficient to go upon in forming a judgment as to its existence. Its end comes with the southerly wind that marks its western side, so that a southerly, wind, even a light one, is a warning which no hostile navigator is likely to disregard. In winter it is often foggy in anticyclonic weather, particularly when the anticyclone is going away, and sometimes it is cloudy and gloomy, but there is never heavy rain in the central region, and seldom any rain at all. Unfavourable weather. The most unfavourable weather for hostile operations with airships is cyclonic Plate X. AIRSHIP WEATHER. FlG. 2. DISTRIBUTION OF TEMPERATURE, WEATHER, WIND, AND PRESSURE, 7 A.M. 1st NOVEMBER, 1915. ISOBARS are drawn for intervals of five milli- WEATHER. Shown by the following gym bols : ^. bftrs - C\ clear ky. O k T i clouded. WIND.- Direction Is shown by arrcrws fljing X ;s: . with the wind. dD 8R y ^ clouded. (Jj) sky j clouded. Force, on the scale 0-12, is indi- /Tix overcast sky. A rain falling Umber f I8atheri - Y ow. A hail s fog =uiist, T thunder. 15 thunderstorm. TEMPERATURE. t/iven in degrees Fahrenheit. x f J A irsh ip- Wea ther. 23 weather represented on the map by a region of relatively low barometer round which strong winds circulate. (See figure 2.) It is the opposite of anticyclonic weather. The cyclonic depression passes rapidly across the map and the weather goes through a well-known cycle of phases in the course of twelve to twenty-four hours. A well-developed cyclonic depression makes successful air raiding impossible, partly because of the strength of the wind, which may reach 50, 60 or 70 miles an hour in the upper regions, and still more because of its variability (the wind is gusty and squally and is also liable to regular changes), which may give the ship as much lee-way as traverse. This, in darkness, means losing the course and probably losing the bearings. Besides, there is often heavy rain or snow with cyclonic weather. In the South of England cyclonic weather generally begins with a southerly or south-westerly wind, and as there is frequently a succession of depressions passing along the same track there are successive fallings and risings of the barometer and successive phases of southerly wind, veering to N.W. with the rising barometer, and backing again to S.W. with a falling barometer. Between two successive depressions there is often a day of perfect weather, light, transparent airs and clear skies. An airship commander who started at the right time might use this brilliant interval to make a raid, but with- out extremely expert forecasting, which would require ample telegraphic information, it is too dangerous. He is more likely to wait until the setting in of a northerly or north-easterly wind marks the beginning of an anticyclone. Risky weather. Between these two extremes of easily identifiable weather, favourable or unfavourable, there are a number of conditions which may be called risky, 24 Glossary. periods of slow transition between anticyclonic and cy- clonic, or periods of vague type without marked features. , These require an expert knowledge of meteorology if they are to be dealt with successfully. -So far as we know, hostile aircraft have used special meteorological observa- tions (with pilot balloons) to identify a case in which a strong north-easterly wind, too strong for easy navigation, fell off and became much lighter in the upper air. That is characteristic of easterly and north-easterly winds, but there are exceptions. To make use of the proper occasion in this particular is certainly clever, but it is risky ; because we can only take advantage of such cases when we happen to find them, and we do not know any law of their distribution. In this connexion it may be re- marked that airships are not likely to take the air at night in a strong wind. Not knowing precisely the dis- tribution of air currents, it is impossible to lay a course for an objective across wind, so the objective must be approached ultimately up wind. That means the slowest speed at the most critical point. The most risky weather for an airship is when a cyclonic depression with southerly wind in its front advances rapidly eastward and replaces the light airs in front of it. Light airs, it has been remarked, are characteristic of the central region of an anticyclone, but they are also characteristic of the region in front of an advancing depression. Depressions sometimes advance at a rapid rate, say 25 miles an hour 600 miles a day always in that case from the W 7 est or south-west. In winter the risk which an airship runs depends a good deal upon the position of the centre of the depression. On the northern side of it, or in its rear, there is often snow, in the latter case with strong northerly winds. Airship- Weather. FIGURE 3. 25 Chart of Barometer Reading's Variations of the Barometric pressure at five stations during the passage of an anticyclone in September, 1915. 26 Glossary. FIGURE 4. Chart of Barometer Readings Oct.- Nov., 1915 1020mb Helde-r Yarmouth IQIQmk Holqhead C.GrisNer woo* Skudesnaes Variations of Barometric pressure during the passage of two consecutive cyclonic depressions in October-November, 1915. Airship- Weather. 27 This description is quite typical, and is exactly applic- able to the case of February 17th, 1915, when two airships were lost. Prognostication. It will be understood from what has been written above that forecasting favourable weather for air raids for a few days ahead is an easy matter when an anticyclone establishes itself on the map. It requires only the most elementary knowledge of weather-study. Similarly it is quite easy to recognise a day or two in advance the periods of unfavourable weather. The best way of doing this is to have a daily map upon which the positions of the anticyclone or cyclonic depressions are easily recognised. But if a chart of consecutive barometer readings is preferred, the charting of the readings at Helder, Yarmouth, Holyhead, Grisnez, Skudesnaes may be recommended. Figure 3 shows the chart for the passage of the anticyclone of Figure 1 , and Figure 4 the chart for the passage of the cyclone of Figure 2. Forecasting for the more risky weather is a matter for experts, and entails a careful study of weather maps. What was curious about the late summer and autumn of 1915 was the frequency of anticyclonic periods, and their coincidence with times of new moon. A succession of depressions is generally characteristic of the weather of North-Western Europe, and, consequently, a competent meteorological establishment would naturally lay itself out to catch the occasional opportunity. But this was not at all necessary in the season of 1915. No particular skill in forecasting was required. Altimeter. An aneroid barometer graduated to show height instead of pressure. The most that an aneroid barometer can do is to give a satisfactory measure of the 28 Glossary. pressure of the air. The pressure is very largely affected by the height, and, therefore, whatever indicates the pressure gives a rough indication also of the height. The accurate determination of the height of a position requires a knowledge of the temperature of the air at successive steps, so that a mean temperature may be obtained for the column between the position and the earth. There are various short ways of making estimates of the temperature of the column, but in any case the temperature at the top and bottom should be noted.* Altitude. The angle in a vertical plane subtended at the eye of the observer by the line drawn from the top of an object to the horizon. The word is also used commonly as synonymous with height. AltO-CUmulus. A form of cloud of middle height (10,000 feet to ^5,000 feet). It consists of fleecy groups of cloudlets called by the French " gros-inoutons." The separate cloudlets are thick enough to show a darkening of the white ; the similar groups of smaller and higher clouds, cirro-cumulus, show no shadow. See CLOUDS. AltO-StratUS. A sheet of continuous cloud of middle height, of considerable size and moderate thickness, some- times covering the whole sky. It must be distinguished from cirro-stratus which is higher and thinner, and stratus without any prefix, which is the lowest form of cloud- sheet. * The pressure at the foot of the column must also be known, and the lack of this knowledge is a source of error with a machine travelling over great distances. A special note on the subject prepared for the Handbook of Meteorology can be obtained from the Meteorological Office. Anabatic. 29 Anabatic. Referring to the upward motion of air due to convection. A local wind is called anabatic if it is caused by the convection of heated air ; as, for example, the breeze that blows up valleys when the sun warms the ground. See BREEZE. Anemobiagraph. See ANEMOGRAPH. . Anemogram. The record of an anemograph. Anemograph.. An instrument for recording the velocity or force, and sometimes also the direction of the wind. The best known forms of anemograph are the Robinson Cup anemograph similar to that designed by Beckley for Kew Observatory, the Tube anemograph with direction recorder similar to that designed by Dines for Benson Observatory (which might be called the Harpagraph or gust-recorder), the anemobiagraph designed by Halliweli for Negretti and Zambra, the Dines Tube recorder, with direction recorder designed by Rooker for R. W. Munro. The Royal Observatory at Greenwich and the Observa- tory of the Mersey Docks and Harbour Board, near Liverpool, have pressure-plate anemographs by Osier. Anemometer. An instrument for measuring the velocity or force of the wind. Anemometers register in various ways ; by counting the number of revolutions of cups in a measured time, by the difference of water level in a tube, and in other ways. Information as to the construction and use of various anemometers is given in the Observer's Handbook, Anemoscope. An instrument for indicating the existence of wind and snowing its direction. The one 30 Glossary. best known to the Meteorological Office is that designed by Mr. J. Baxendell which is provided with recording mechanism. It is an observatory -instrument, not a portable one. Aneroid Barometer. An instrument for deter- mining the pressure of the atmosphere. It consists of a shallow air-tight metal box, usually nearly exhausted of air. The distance between opposite faces of the box alters with change in the surrounding atmospheric pressure, the alteration being shown on a dial by a hand actuated by a suitable train of levers. An aneroid is light, portable and convenient, but should be compared occasion- ally with a mercury barometer, as an appreciable change of zero sometimes occurs. It is also subject to "creep," e.g., after a recent large fall of pressure such as may occur when it is used as an ALTIMETER it will, though under a really constant pressure, show a small spurious further fall, which in the course of an hour may amount to 1 or 2 per cent, of the previous fall. u Creep " in the same direction may be perceptible for several hours, but its rate continually diminishes. Aneroidograph.. A self-recording aneroid. An aneroid-barometer provided with mechanism for record- ing the variations of pressure of the atmosphere. See BAROGRAPH. Antlielion. A colourless MOCK SUN (see HALO) appearing at the point of the sky opposite to and at the same ALTITUDE as the sun. Anticyclone. An anticyclone is a region in which the barometric pressure is high, relatively to its surround- ings, and is generally shown on the weather charts by a series of closed isobars, the region of highest pressure Anticyclone. 31 being the central region of the anticyclone. In a well- marked anticyclone the isobars are roughly circular or oval curves, the wind blows spirally outwards in accord- ance with Buys Ballot's Law, and the pressure in the central parts is very seldom under 1,015 millibars or 30-00 inches. See Plate XIII. Certain parts of the earth, notably large parts of the latitude belts of about 30 N. and 30 S., also continental areas in the winter in temperate latitudes, are anti- cyclonic regions. In the Azores-anticyclone in summer the pressure is usually about 1,025 millibars, or 30*25 inches, and in winter about 1,020 millibars, or 30*10 inches, and in the Siberian anticyclone of winter the pressure is often as high as 1,050 millibars, or 31 '00 inches. Anticyclones are characterised by calms and light winds and an absence of rain; the desert regions of the earth are anticyclonic regions. But in the temperate zones, short of gales and strong winds, almost any weather may occasionally occur in an anticyclone. In England they are generally accompanied in winter by dull, cheerless weather and fogs, and in summer by bright, hot weather. The causes of anticyclones are still unknown. We have learnt in recent years that the temperature of the air in them between the heights of 2 and 10 kilometres (1-6 miles) is higher, but at still greater heights lower than its environment. For the anticyclone in relation to weather refer to the Weather Map, and see also AIRSHIP-WEATHER and ISOBARS. Aqueous Vapour.* Aqueous vapour isalways present in the atmosphere, and, although it never represents more . * See also HUMIDITY, p. 154, and ABSOLUTE HUMIDITY, p. 290. 32 Glossary. than a small fraction of the whole, it has physical properties that give it great importance in meteorology. In a closed space wherever there is a free surface of ice or water, evaporation takes place until the water- vapour exerts a definite pressure of saturation, depending only upon the temperature, and not upon the pressure of the surrounding air. This pressure of saturation is very much greater at high than at low temperatures as is shovm in the following table : Temperature. Pressure of Saturation in millibars. Temperature. Pressure of Saturation in millibars. F a. 10 260-8 20 266'3 30 271-9 40 277-5 50 283-0 2-4 3*7 5'8 8-5 I2'2 l? a. 60 288-6 70 294-1 80 299-7 90 305-2 100 310-8 17-6 24-7 34-6 47*8 65-0 A cubic metre of dry air at 1,000 mb. and 289a. weighs l,206g. The mass of water contained in saturated air at different temperatures is given in the following table : Mass in grammes Mass in grammes Temperature. of water vapour per cubic metre Temperature. of water vapour per cubic metre of saturated air. of saturated air. F. a. F. a. 32 273-0 5 70 294-1 18 40 277-5 7 80 2997 25 50 283-0 9 90 305-2 34 60 288-6 13 100 310-8 45 Vapour. 33 It is easily seen from these figures that saturated air must at once yield rain or snow if cooled, and even air that does not contain all the aqueous-vapour possible will ultimately deposit moisture if sufficiently cooled. In the passage from the liquid to the gaseous state great quantities of heat are absorbed, 536 calories for every gramme ^evaporated at the boiling point, and even more if the water is initially cold. Conversely much heat is yielded up when condensation occurs. See p. 70. Tyndall has shown that the heat radiated from a black body at the boiling point of water is readily absorbed by aqueous vapour, which must, therefore, have a correspond- ing power of radiation. Spectrum analysis shows also that some of the visible radiation of the sun is strongly absorbed by the earth's atmosphere. Atmosphere. See Weather Map. M.O. 225 i., p. 15. Atmospheric Electricity. See p. 294. Audibility. The audibility of a sound in the atmos- phere is measured by the distance from its source at which it becomes inaudible. On a perfectly clear, calm day the teound of a man's voice may be heard for several miles, provided there are no obstructions between the source of sound and the listener ; but- quite a small amount of wind will cut down the range of audibility enormously. The sound is not cut down equally in all directions; to leeward, for instance, a sound can usually be heard at a greater distance than it can to windward of the source. This is accounted for by the bending which the sound-rays undergo, owing to the increase in wind-velocity with height above the ground, the rays to leeward of the source being bent downwards while those to windward are bent upwards so that they pass over the head of an 34 Glossary. observer stationed on the ground. The decrease in all directions in the range of audibility of a sound when there is a wind appears to be due chiefly to the dissipation in the energy of sound as it passes through eddying air. A plane wave-front becomes bent in an irregular manner when it passes through air in irregular or eddying movement. It, therefore, ceases to travel uniformly for- ward. Part of its energy is carried forward, while the rest is dissipated laterally. If there were no dissipation of energy in a sound-wave the intensity of the sound would decrease inversely as the square of the distance from the source. Experiments show that, under normal conditions when there is a light wind blowing, the rate of decrease in intensity of sound at a distance of half a mile or more is considerably greater owing to the dissipation of energy than would be expected from the inverse square law. If, for instance, a whistle can be heard at a distance of half a mile, four whistles blown simultaneously should be audible at a distance of a mile ; but the range is actually only increased to about | of a mile. 4 Sounds are usually heard at greater distances during the night than during the day. On calm nights the range of audibility of a sound may be as much as 10 or 20 times as great as it is during the day. This effect is clue partly to the increased sensitiveness of the ear at night owing to the decrease in the amount of accidental disturbing waves, partly to the inversion of temperature which com- monly occurs on calm, clear nights, and has the effect of bending the sound- waves downwards, but chiefly to the diminution of the amount of disturbance in the atmos- phere at night. Between the source of sound and the extreme range of Audibility. 35 audibility areas of silence sometimes appear, in which the sound cannot be heard. This effect has in some cases been attributed to a reversal in the direction of the wind -in the upper layers of the atmosphere. The lower wind would bend the sound rays upwards to windward of the source. On entering the reversed upper wind current these, rays would be bent down to the earth again, and would reach it at a point separated from the source of sound by an area of silence. This explanation is quite adequate in many cases in which the places, where the sound is heard again, are to the windward of the source. There are, however, many cases in which areas of silence appear to leeward of the source, and many others in which an area of silence occurs in the form of a ring enclosing the source and surrounded by an area of distinct audibility. In most of the cases where a ring-shaped area of silence has been observed the outer region of distinct audibility begins at a distance of about 100 miles from the source, and may extend to 150 miles or more. The well-known Silvertown explosion is a good example of a case in which a detached area of audibility was separated from the source of sound by an area of silence. In the accompanying map, which is reproduced by permission from the Quarterly Review, the two areas of audibility are shown. It will be seen that the outer area, which includes Lincoln, Nottingham and Norwich, lies about 100 miles from the source of sound. The inner area sur- rounding the source is not symmetrical, being spread out towards the north-west and south-east. Definite evidence was obtained that no sound was heard at various towns within the area of silence. No very satisfactory explanation of these cases has so far been offered. The wind-distribution necessary to 13204 B 2 36 Glossary. explain them on the wind-refraction theory would be very complicated and would, moreover, in some cases, be of a type which no meteorologist has yet observed. The effect of FOG on the audibility of sound has been the subject of a considerable amount of discussion. The idea that sound is muffled by a fog seems to be commonly accepted ; but on the other hand the experiments of Henry and Tyndall have failed to give any indication of such an effect. They seem rather to show an increase in audibility in a fog. The effect of the waterdrops them- selves has been shown to be too mall to affect the propagation of sound waves to an appreciable extent, while the weather conditions usually associated with the production of fog, the homogeneous state of the atmos- phere and the INVERSION of temperature, are such as to give rise to increased audibility. In calm weather the direction of a hidden source of sound may be estimated to a few degrees by turning the head till the sound appears to come from the point towards which the observer is facing. The .observer, however, is seldom confident that he has attained such accuracy. In windy weather it is more difficult to esti- mate the direction of sound. Aureole. The luminous area surrounding a light seen through a misty atmosphere. Aurora. See p. 298. Autumn. Autumn, in meteorology, comprises the three months of September, October and November, the first three months of the farmer's year. In astronomical text-books it is defined as the period commencing with the autumnal equinox and ending with the winter solstice, i.e., from September 23rd to December 22nd, but Audibility To face p. 36. ^Stfrtv Canter &pry c (Reproduced "by permission from the Quarterly Review. July, 1917.") 38 Glossary. CURVES SHOWING CHANGE OF TEMPERATURE WITH HEIGHT ABOVE SEA- LEVEL OBTAINED FROM BALLON- SONDE ASCENTS 1907-8. V PYRTON HILL 4 PYRTON HILL iPITCHAMWftK pwnw HILL pTDMMMK [MANCHESTER r; toe. MANCHESTER 8 DEC 3, (MANCHESTER' ' SELLACK MANCHESTER PYRTON HiUL WTCHAMfMK MANCHESTER CRiNAN'HARfiS MANCHESTER CRINAN HARB< (&CNNANHUW OITCHAM F*RX JPYRTON HILL pITCHAM PAW (MANCHESTER Jwa. PYRTON HILL . PYRTQN HILL flfae. MANCHESTER IWRTON HILL [MANCHESTER PYRTON HILL MANCHESTER icpTifSELLACK (OITCHAM f 5. OITCHAM R^RH ISELLACK icpre. 01 SOTit PYRTON HIU Z. PYRTON HILL 5. [Oil ISELLACK OITCHAM WRK Z7Ae. PYRTON HILL -40 -EO 2.Q TEMPERATURE IN DEGREES FAHRENHEIT , 3^0 40 250 2^0 -270 fEMPCRATURE ABSOLUTE 300 r The separate curves represent the relation between temperature and height in miles or kilometres in the atmosphere. The numbers marking the separate curves indicate the date of ascent at the various stations as shown in the tabular columns. The difference of height at which the isothermal layer is reached, and the difference of its temperature for different days or for different localities, is also shown on the diagram by the courses of the lines. Brdlon sonde. 39 Ballon SOnde. A small balloon usually made of india- rubber, inflated with hydrogen, and used for carrying self -registering instruments into the free atmosphere and thus obtaining records of the pressure, temperature and humidity aloft. The balloons used in England usually have a diameter of about one metre at starting (40 inches nearly), those on the continent nearly two metres. The balldons generally rise until they burst, on account of the diminished external pressure of the air, which may not happen until they have reached a height of 20 kilometres (12 miles) or more. After the balloon has burst, the material acts as a kind of parachute and breaks the fall of the instruments, so that they reach the ground without injury. A label is attached, offering a reward to the finder for the return of the instrument, and in that way valuable records are secured. Sometimes the balloon fails to burst, but develops pin- holes through which the gas leaks. A long TRAJECTORY is the result. One of our records was returned from a Bavarian forest. In other countries, where much heavier recording instruments are used, two balloons in tandem are em- ployed, one of which bursts, and the other regulates the fall. This mode of arranging the apparatus, with a simple modification, is available for use at sea and many soundings of the air over the sea have been obtained. We have not taken part in that side of the inquiry. Very remarkable facts about the temperature of the free air have been disclosed by soundings with ballons- sondes. Their general characteristics are shown in the diagram on page H8 which exhibits graphs of tempera- ture and height for 45 soundings obtained for the Meteorological Office in 1907-8. The reader should notice that, with one or two exceptions, the balloons reached a 40 Glossary. O O C^ O co t-H CO O co i-i O C M coo <^ co VQ O ^ O v -d- vo CTiOO H-I CT> co n O t- **"> O "3-OO co t^ CO -rf-O O\ coo MOi-it^^ coOO>O-' v-O VOOOOCOO COCOO OO O I *j"> t^ O ir > O i^)O ^ O "st" t^ ) ~ | O O i ' t^- -j- N O O^ O i PL, i_i (_| n i_ i co co cocO'^-'^- vo O t> l~>- O^ O CO co -rj- ^ *~t~> O O !> OO O M 00 co C PH O II s C7\ rh O O co M CO N OO^C^iOO -QOco i-i^O^O^" ' ) rj- rj- vD VO O t> OO O JBallon sonde. H SI H 02 W P to feq O O fe o PQ EH 3 ri ^ g o 02 ^ 2 OQ 3 O -g w S k a g ^? > M > * M "S S P ^ 03 I I I- .CD 3 S r^ - _ f ' ^ to 43' I Qs LO M I>*GO M LOMGO M M 01 ^ GC O COOO VOM * M M OO rt-MGO lOMOO ^ n ^ ro co rj- 10 uox M MSQ COM n M w co rf -Mt^ COCO M \O oo M w \o co w oo voc^co -Nha^coaN C< 01 W CO -t- rt- LO^O vO !>. r^OC GC ^M i>,ir)i-i i>. coco co O co TJ- TJ- LO\O so r^ r^oo GO M iriMGO LOCI OMOMVO C^OO 01 01 CO CO rf IT) LO^O l>s ^00 GO O Tt- OvO H OvO 01 GO CO ON LO ^1 01 01 CO rf rf- LOvO ^O t^ X^GC O 01 \O M OvO M CNJ^ O ^O 01 01 M 01 01 co co *-j- ir) LO\O t^ r^oo OMT^O --t-o r^^ M M 01 01 co co *t- 01 x r^ o M M 01 01 Ol -f-o r^coo rx ro ix H \o M M rj 01 ro co -f- i^i i n M o 42 Glossary* position (somewhere near ten kilometres), after which the temperature ceased to fall, yet the range of temperature at that level for the whole series is larger than the range at the surface. Additional soundings have enabled us to put forward a table of the average pressure and temperature of the free air at different levels in the several months of the year, which is given on pp. 40 and 41. Particulars as to the variation of the meteorological elements, temperature, wind and humidity derived from observations with ballons-sondes, kites and pilot balloons are given in Geophysical Memoirs, No. 5, by Major E. Gold, D.S.O. (published by the Meteorological Office). Balloon Kite. For dealing with the general features of the relation of temperature to pressure or height in the upper air of all parts of the globe the ballon-sonde is most effective. In this general inquiry the details due to the smaller irregularities of diurnal or seasonal variation may be disregarded. Such irregularities are specially noteworthy in the lowest kilometre of the atmosphere and are of importance in aviation and gunnery, because changes in the distribution of temperature are necessarily accompanied by changes in the distribution of pressure, and consequently of wind. The first kilometre or 3,000 feet, therefore, requires special attention. Observations of temperature, humidity and wind can be got by means of kites, when there is wind enough ; by captive balloons, when there is little or no wind ; and by the observation balloon or balloon kite in all weathers, except a gale. Special instruments are required for these observations. A. special form of meteorograph has been designed by Mr. W. H. Dines for use with kites, but suitable provision has still to be made for kite-balloons and captive balloons. Bar. 43 Bar. The unit of atmospheric pressure, being equal to the pressure of one million dynes (one megadyne) per square centimetre. The BAR is equal to the pressure of 29-5306 inches, or 750-076 mm. of mercury at 273a (32F) and in latitude 45. The name was introduced into prac- tical meteorology by V. Bjerknes, and objection has been raised by McAdie of Harvard College on the ground that the name had been previously appropriated by chemists to the C. G. S. unit of pressure, the dyne per square centimetre. The meteorological bar is thus one million chemical bars, and what chemists call a bar we should call a microbar. One bar is 100 CENTIBARS or 1,000 MILLIBARS. See p. 194. Barogram. The continuous record of atmospheric pressure yielded by a self-recording barometer. See p. 156. Barograph. A self-recording barometer, an instru- ment which records automatically the changes of. atmospheric pressure. In one form of mercury barograph the movements of the mercury in a barometer are com- municated by a float to a pen in contact with a moving sheet of paper carried by a revolving drum which is driven by clockwork. The portable barographs which are in common use are arranged to record the variations of pressure shown by an aneroid barometer, and on that account they are some- times referred to as ANEROIDOGRAPHS. Particulars as to the method of using these instruments are given in the Observer's Handbook. (M.O. Publication 191.) Barometer. An instrument for measuring the pres- sure of the atmosphere. The mercury barometer has been found to be the most satisfactory form for general use, The principle underlying this type of instrument 44 Glossary. is quite simple. If a glass tube 3 feet long, closed at one end, is filled with mercury, and the open end is tempor- arily stopped up and immersed in an open vessel of the same liquid, then if the tube is held in a yertical position and the immersed end is re-opened, the mercury will fall until the level inside the tube stands at a height of about 30 inches above the mercury in the trough. The pressure of the atmosphere on the lower mercury-surface balances the tendency of the enclosed column to fall, and the height supported in this way represents the atmospheric pressure at the time. In order to cojnpute the pressure, the length or height of the column of mercury has to be measured. Different mercury barometers vary as regards the method of reading this height, and in all the tem- perature of the mercury and the latitude of the place mast be taken into account. In the aneroid barometer the pressure of the atmosphere causes deformations in a spring inside a closed metallic box, which has been exhausted of air, and these are communi- cated to a pointer moving over a suitably engraved scale. For the purposes of meteorology the pressure of the atmosphere has to be determined to the ten-thousandth part, which is a much higher degree of accuracy than is required in other meteorological measurements. Special contrivances and precautions are therefore required, whicfr are duly set out in the Observer's Handbook. Barometric tendency. The change in the baro- metric pressure within the three hours preceding an observation. See Weather-Map, p. 35. Beaufort notation. A table of letters for weather. See Weather-Map, p. 10. Beaufort Scale. 45 Beaufort Scale. The scale of wind force devised by Admiral Beaufort in 1805. An explanation has been given in the Weather-Ma/), p. 13. Table of Equivalents in Force and Velocity. Pressure of Wind Equiva- | Limits of Velocities. on a Plate lent a velocity & ^ in Ibs. in Milli- in miles f-\ EH Q Statute Miles Nautical ^ , Miles i Metres Feet per square ft. bars ( i o 3 dynes per hour. s 1 per Hour. per H P , S d ' per Second. per cm. 2 ). CQ O Less Less Less Less 1 than i than i than 0*3 than 2 01 'or 2 1 i-3 i-3 o'3-i'S 2-5 08 '04 5 2 4-7 4-6 i -6-3-3 6-1 1 28 13 10 3 -12 7-10 S^-SH 12-18 67 32 15 4 13-18 11-16 5'5-S-o 19-27 I-JI 62 21 * 5 19-24 17-21 8-1-107 28-36 2'3 i-i 27 6 25-31 22-27 10-8-13-8 37-46 3-6 17 35 7 3^-38 28-33 I 3'9- T 7' 1 47-56 5'4 2 ' 6 42 8 39-46 ' 34-4 17-2-207 57-68 77 37 So 9 47-54 41-47 20*8-24-4 69-80 10*5 5-0 59 10 55-63 48-55 24-5-28-4 8i-93 14/0 67 68 11 64-75 56-65 28-5-33-5 94-110 Above : Above Above 12 Above Above 3 3 -6 or Above 17-0 8-1 75 75 65 above. no 46 Glossary. Bishop's ring, so named after its first observer, is a dull reddish-brown ring of about 20 outer radius seen round the sun in a clear sky even at mid-day. That it is a CORONA, and not a HALO, is proved by the fact that at times it has been seen to have a red outer margin (see CORONA). It appeared after the great eruption of Krakatoa in 1883, and remained visible till the spring of 1886, and was no doubt due to. minute particles shot out by the eruption ; these remained suspended in the atmos- phere for a considerable time. The great radius of this corona is explained by the smallness of the particles, and its intensity by their great number. The non-appearance of the other colours of the corona is explained by the presence of particles of many different sizes. Bishop's Ring was seen again after the eruptions of the Souffriere in St. Vincent and Mont Pelee in Martinique in 1902. Blizzard. A gale of wind with the temperature below freezing, the air being filled with fine dry snow. The snow may not be actually descending from the clouds but be merely raised from the snow-covered ground. The fine powdery snow peculiar to these storms is formed only at very low temperatures, so that the phenomenon is practically confined to the polar regions and the large land areas of the temperate zone. During a blizzard the temperature often rises, possibly because the violent wind causes a mixing of the lowest layers of the atmosphere and brings down air that is often warmer than the excessively cold air lyinof near the ground. Blue Of the Sky. Light rays striking particles which are smaller than the wave length of the light are scattered, that is turned aside in all directions. But the short wave's Blue of the Sky. 4? composing the blue and violet end of the spectrum are more completely scattered than the long red and yellow waves. Hence light passing through a medium contain- ing a great number of such particles is left with an excess of red, while light emerging laterally has an excess of blue. It is for this reason that soapy water looks yellow- ish when one looks through it at a source of white light, and bluish when one looks across the direction of illumination. The greater part of the sky appears blue because the light from it consists mainly of light scattered laterally from minute particles in the atmosphere. The smaller the particles the less intense is the light but the greater the proportion of it that is blue. When the particles are larger the proportion of blue is less, as in the whiter sky of a haze. Near the horizon the sky is whiter than at the zenith because the rays of light from that region have passed through a greater thickness of the lower air where large particles are relatively more numerous, Sunset colours are reddish because the rays reaching ue directly have lost much of their blue light by lateral scattering. The sky as seen from high moun- tains and from aeroplanes at a great height is of a deeper but purer blue because there are fewer large particles than at lower altitudes. Boiling* Points. See p. 300. Bora. A cold wind occurring in the northern Adriatic, very violent, which blows from the high plateaus which lie to the northward. These plateaus may become extremely cold in clear winter weather, and passing cyclonic systems allow the air to flow down to lower levels. The actual violence of the wind in a bora is largely due to the weight of the cold air of the plateau causing it to run down the slope like a torrent or cataract 48 Glossary. of water. The wind experienced may therefore be in- dependent of Buys Ballot's Law. The adiabatic warming due to the increased pressure below is riot sufficient to prevent the resulting wind from being cold. In the Meteorological Office it is proposed to call local winds of this character " katabatic " in order to distinguish them from the winds which show the normal relation to the distribution of atmospheric pressure and are called " geostrophic " winds. Breeze. A wind of moderate strength. Glacial-breeze. A cold breeze blowing down the course of a GLACIER, and owing its origin to the cooling of the air in contact with the ice. The movement of the air is due to the gravitation of the air made denser by the cold surfaces on a slope, and a glacial breeze may be classed as a typical example of a katabatic wind. Lake-breeze. A breeze blowing on to the coast of a lake in sunny weather during the middle of the day, part of the convectional circulation induced by the greater heating of the land than of the water. Land-breeze. An off-shore wind occurring at the margin of a sea or lake during a clear night, due to the more rapid cooling of the air over the land than over the water. During the day the conditions are reversed and the wind blows from the sea to the land, constituting a SEA-BREEZE. These phenomena are most marked in the tropics, where the wind arising from other causes is usually not strong enough to mask the convectional effect. See also p. 183. Mountain-breeze. A night breeze blowing down the llrew. 49 valleys, due to the flowing downward of the air chilled by the cold ground. These also, as being due to convection in which the colder air takes the leading part, would be classed as katabatic winds, whereas the one next following in which warmed air plays the leading part would be classed as an u anabatic " wind. Valley -breeze. A day-breeze that blows up valleys when the sun warms the ground. Brontometer, from bronte, a thunderstorm, a com- bination of apparatus for following and noting all the details of the phenomena of weather during a thunder- storm. Buoyancy. Used generally with regard to ships or balloons to indicate the load which a ship could carry without being completely submerged, or the weight which a balloon or airship can carry without sinking. In the case of the balloon or airship the buoyancy is due to the displacement of air by hydrogen which is lighter than air. A cubic metre of perfectly dry air at 1000 mb. and 273 a. weighs 1*28 kilogramme, whereas a cubic metre of hydrogen under the same conditions weighs only *09 kg. It follows that a cubic metre of hydrogen in air at 1000 mb. and 273a. will have a buoyancy represented by the difference, i.e., 1*19 kg. Part of the buoyancy will have to be devoted to sup- porting the envelope which contains the hydrogen and which adds so little to the volume of air displaced by the hydrogen that, for purposes of calculation, we may regard the displacement of the air by hydrogen as the source of the buoyancy, and count the envelope and oth<T accessories of the same kind as " dead weight." 50 Glossary. The gas that is used under the name of hydrogen for filling balloons always contains some impurity that re- duces its buoyancy. Water-vapour in the hydrogen and the surrounding air will reduce the buoyancy by an amount varying from 0*2 per cent, to 2 per cent, in the common range of circumstances and the impurities in- cidental to the manufacture of the gas, or to leakage. may easily reduce the buoyancy by 5 or 10 per cent. Instead, therefore, of taking the buoyancy of a cubic metre of working hydrogen at 1*19 kg. the theoretical figure for pure dry hydrogen in dry air, we may take it at 1*10 kg. per cubic metre at 1000 mb. and 273 a. The volume of a large airship may be 25,000 m 3 , the dimensions being 140 m. in length and 15 in. in diameter. The gross buoyancy at 1000 mb. and 273 a. is, in that case, 25,000x1-10 kg., or 27,500 kg. In those conditions the relation of pressure to temperature is 3*66:1, see p. 53. In other conditions of pressure and tem- perature the relation, and therefore the buoyancy, will be different. The relation between the displacement and the weight supported is given by the equation where Q is the volume of air displaced, p the density of the air, a the specific gravity of the "hydrogen " referred to air at the same temperature and pressure, W the dead weight, L the portable load, and B the ballast. BUOYANCY IN DIFFERENT ATMOSPHERIC CONDITIONS AND THE LIMIT OF HEIGHT THAT AN AIRSHIP CAN REACH. The density of air p becomes less and less as one ascends in the atmosphere because the pressure di- minishes. The temperature diminishes also, and, on that Buoyancy. 51 account, there is some compensation for the fall of pres- sure, but not enough co preserve the buoyancy. We may suppose that <r, the specific gravity of the hydrogen, remains the same throughout, because, in an airship, the pressure of the hydrogen changes with that of the air in which it floats. Its temperature changes likewise, and, if we leave out of account the heat received or lost in the form of radiation, the fall of temperature of the hydrogen of an ascending balloon is generally more rapid than that of the surrounding air. It would take time for the temperature to become equalised, whereas the adjustment of pressure is practically immediate. Therefore, the hydrogen in a rising airship is rather denser than the result of calculation would give. Thus, to suppose <r to remain constant, is a little more favour- able to the navigator than actuality, unless he takes ad- vantage of sunshine. The buoyancy at any level is determined by the density p. With the assumption that <r remains constant the buoyancy can be computed from the density of the air at the level of standard pressure and temperature by the ordinary gas-equation Where p, p, T are the pressure, density and absolute temperature of the air at the selected level p n9 p , T are the standard pressure, density and temperature. The equation of buoyancy becomes 0xx 'poO-T)^ W+ + B. 52 Glossary. For the figures which we have quoted, jp u = 1000 mb, T,, = 273a, Po (1 - *) = 1-1 kg/m 3 and the equation becomes 273 x 1-1 p _ TF+ + 1000 T Q W and L, the dead weight and the portable load, cannot be altered during a voyage without sacrificing something, but the ballast B is carried for the purpose of adjusting the level. The maximum height will be reached when the ballast is exhausted, that is when B is zero. In this equation the value of the ratio p\T which determines the density depends upon the pressure and temperature of the air at the time of the flight, but for aeronauts the most important cause of the variation in these elements, and therefore in their ratio, is the change of pressure and temperature of the atmosphere with height. These are so considerable that they overshadow altogether the changes at any chosen level from day to day or from month to month. We can therefore use a table of average monthly values with advantage. The following tables give the average values ofpjTfor different heights computed from observations of ballons-sondes. From the values pjT the density in grammes per cubic metre can be found by multiplication by 348. (The first 'table is computed from the pressures in the accompanying table and the temperatures in Table IV. 2, Computer's Handbook, Part II., 2, p. 56, : the second has been pre- pared by Mr. W. H. Dines, F.R.S., for the Handbook of Meteorology). Buoyancy. 53 . ir> -f- ro Ol I-H O ONOO r-0 n ^04 hH^ 03 01 1 ir< -}- moo O4 OO vno t-^oo O *- rj-04 O O C4 \n 04 M -sf- O coo ON O4 O a l i CD Tf- CO "^-O l-l O vno r^oo O -H c^vnRo,^ O co rj- t~>. vn cOO ON O4 O - g ii i 1 1 hH M hH I-H l-l 04 O4 04 Ol CO co ~ST in <& vno r^oo O "i coxnr^oo HH cOO ON 04 O ^ hH I-H ,_ _ W tH 04 O4 O4 O4 CO CO 2 ^pts 1 O vno ON cooo -* KH O 00 O ri- O 00 O *>. cOO OO CN vn 'I jg 1 '1 j vno !> ON O i-i rh 04 O 00 O co vn t^oo I-H co ,ONOO O vn co vnoo 04 vn 43 rt H BU I-H I-H I-H M O4 O4 04 O4 CO CO c3 "S bo bi) p kj co vno oo O 04 OO t^ t^ 04 co vnoo M vn <1 fhH hH _ HH hH I-H 04 O4 O4 O4 CO CO S, |-S * t-*.O r- O ^h O co vno oo O O4 OO t^.OO CO co vnoo HH vn 4-3 rH vr> 5 - M 1-4 M I-H I-H O4 O4 O4 04 CO CO 43 p | 3 vr*vn p vno t^oo O >-< Svn^as co O t> ONO coo oo hH vn -2& -a M l-l |_| |_| |_4 hH 04 C4 N O4 CO co ~^ vn vn vnoo co ON vn 04 ON i-i M co vno ON i-i rh 04 hH COOO coO ON 04 vn | IS fl HH M M hH M I-H 04 04 O4 O4 CO CO O! d, vno J>-OO O -i a as as O l-l O4 vn 04 coO ON cX O S S3ZSS5 rj- I-H CT> HH co coo ON c>4 O 02 bo S HH ^H |-| hH I-H I-H 04 04 04 O4 CO CO "- so ^ '-i s -1- co coO O t~>. vno t^oo O '-i co vn j> ON M 00 ^f- vo O 00 cOO ON coO b l-l I-H |_, HH M hH 04 O4 O4 O4 CO CO fl Id *-i * p SKg8S 8>S8 co O ON co O ^11 M I-H M M hH M 04 O4 O4 O4 CO CO <j3 ^ K^g ^ m ^f- co 01 i-i o 0.00 t-o vn ^-COOlhH^ Q^ ^ Glossary* > O 1-^00 O^ O c-i O woo W to eo co < co fO vnoo co -H oo co -H ON C^O t-i M co t-i OO to W O\ vr> - O 1/- > O O O^> O O ^ *$ co TJ- j> cxi o c^ ^ vn oo vo O^ oo co vn co oo co oo co t^ Q d M co *$ -rf- vnvo 1^.00 O *- co xnTj-vnoo w OO -^-Mintxivncooo Ooo -rh^rl-^'^- ^r^vnvO f>.co O M -rh<o O>W>O O^^Ot-^fOi-i criCTiO>-< t- vnC\rf-OvO xn T}- vnoo c^oocoO O>C^>-<vni-" w -^O ^ vn rj- vnoo WOO'^hCNir^-i-icow t> ONOO vn O*> vn O 'rh )_, _< I I h- 1 - C-l Cl CO CO -rf -rf- 1/~>VQ VO t^ s ! O ^1 jjh-ii-H>-it-(t-H(-ii II-HI iciMc<ifO"^- f- vnvD vO t^ J>.oo be I Buoyancy. 55 With these tables the limiting height of flotation can be approximately determined when the dimensions and par- ticulars of the dead weight, portable load and ballast are known. For example, in the case of an airship which displaces 25,000m 3 of air with a dead weight 11,500kg. 9,000kg. ballast, including 2,000kg. of fuel which can be spared, 7,000kg. portable load, including crew, clothing and food, fuel for the return journey> oil armament and working tackle, the right hand side of the equation is reduced to its minimum when the ballast is disposed of, that is when B = o, in that case p\T works out to be 2*46. This will be found in the table between 3k and 4k, rather nearer 3k in summer than in winter. THE EFFECT OF WEATHER UPON BUOYANCY. The changes of weather affect the buoyancy because they alter the value of p\T at the starting point and at every stage of the course. The value at the starting point determines the amount of ballast that can be carried to begin with, and the value at other stages of the journey determines the height to which the airship can rise with all its ballast expended. Assuming that the limit of ascent is determined by the value 2'4 for the ratio pj T 7 , we find that the variation in the course of the year 1913 was from 3'5k to 4'lk. The effect of low pressure or high temperature can easily be seen but it is not very pronounced. The differ- ence between the worst occasion and the best from January to May 1913 was about 1,600 ft. It is now recognised that above one kilometre, when pressure is low, tempera- ture is also low, and, consequently, at any level the value 54 Glossary. of p\ J^is steadier than the variations at the surface would lead one to expect. In these calculations no allowance has been made for leakage, nor, on the other hand, for compressed hydrogen carried to replace losses. DEPRESSION PRODUCED BY THE WEIGHT OF A DEPOSIT OP RAIN OR SNOW. The catchment area may be about 1000 m 2 , and the figures quoted from Josselin by Modebeck for the weight on that area are Maximum depression. k. Dew (light) Dew (heavy) Rain ... Heavy rain Storm rain Weight. Snow kg. 15 80 200 250 4OO 800 to to to to to to kg. 50 240 290 360 480 1000 03 15 18 21 3 60 According to this table the worst effect, even of snow, would be a depression of 2,000 feet. DEPRESSION PRODUCED BY DESCENDING WIND. The calculation is very hazardous ; supposing the vertical component of the wind in ordinary circum- stances to be from 0'5 to 1*5 m/s, and assuming the ordinary law of resistance, which is true for small areas, the downward force would bo '01 nib. = 10 dynes per cm 2 , say 500 x 111 1 x 10 dynes on the envelope (iakmt- ihe area to be f)00 m 2 ), or 5 x 10 7 dynes. 'This is equivalent to a weight of about 50 kg. and is, therefore, unim- portant, but it would become important if the downward Buoyancy. 57 velocity reached 5 m/s, which it probably may do in a disturbed atmosphere, and in the downrush of a line squall this value might be largely exceeded. Buys Ballot's Law (see The Weather Map). The law is that if you stand with your back to the wind the atmospheric pressure in the northern hemisphere de- creases towards your left and increases towards your right. In the southern hemisphere the reverse is true. The law is a necessary consequence of the earth's rotation. C.G.S. Abbreviation for Centimetre Gramme Second, used to denote the organised system of units for the measurement of physical quantities by units which are based upon the centimetre, the gramme, and the second as fundamental units. Calm. Absence of appreciable wind ; on the Beau- fort Scale 0, i.e., less than 1 mile an hour, or three-tenths of a metre per second. Calorie. See p. 301. Celsius, Anders : an astronomer and physicist, the inventor of the Centigrade thermometer, born at Upsala in 1701. The name Centigrade arises from the division of the interval between the freezing point and boiling point of water into one hundred parts. In continental countries the scale is generally named after the inventor Celsius. It is also not infrequently referred to as the centesimal scale. Note. Celsius divided the thermometric interval between freezing and boiling points of water into 100 parts, but he made the former 100 and the latter 0. The Centigrade scale now in use was introduced by Christin of Lyons in 1743 (possibly earlier). It has also been claimed for Liiine, but the priority is extremely doubtful. 58 Glossary. Centibar. A hundredth part of a "bar" or C.G.S. " atmosphere." See BAR, also see Table of equivalents at the end of this volume. Centigrade. A thermometric scale introduced by ANDERS CELSIUS (q.v .), which has zero at the melting point of ice, while 100 represents the boiling point of water at a pressure of 760 mm. of mercury. A centigrade degree is 9/5 Fahrenheit degrees. On the centigrade " absolute " scale the freezing point is at 273a, the unit being the same as a centigrade degree. Centimetre the hundredth part of a metre. The unit of length on the Centimetre-Gramme-Second system which is universally employed for electrical and magnetic measurements. A metre was originally defined as one ten-millionth part of the earth's quadrant, that is, the distance from the equator to the pole. 1 centimetre is equal to *394 inch, and 2*54 centimetres to 1 inch, to a high degree of accuracy. Cirro-cumulus. A group of fleecy cloudlets or small " flocks " of cloud formed at great heights and showing no shadows. They are called " moutons " in French. See CLOUD. CirrO-StratUS. A layer of thin, transparent cloud at a high level. See CLOUD. Cirrus. Clouds at great height, generally formed into " wisps " of thread-like structure. See CLOUD. Climate. A general summary of the weather for any particular locality. When the weather has been observed for a sufficiently long time in any locality we are able to make a useful statement as to the weather which may be To face p. 58. Cirrus clouds, Thread or Feather clouds at a height of from five to six miles, and generally of a white colour. They are composed of ice-crystals. The picture gives an idea of rather more massive struc- ture than is usual with cirrus clouds, but the sweeps and ^vvisps are very characteristic. Climate. 59 experienced at any particular time of the year in that locality. Technically, the climate of a place is represented by the average values of the different meteorological elements, which should include means for each month, as well as means for the whole year. Average extreme values for different periods and ABSOLUTE EXTREMES are of in- terest, and also the number of rain-days, days of snowfall, frost, hail, thunderstorm, gale, &c., and the frequency of occurrence of winds from different directions. Included also under this heading would be the earliest and latest dates of frost and snow, the average depth of snow lying at different times, and particulars of the temperature of the soil at various depths. In places where the type of weather at a given time of year varies greatly, a long series of observations is needed in order to obtain a fair idea of the different climatic elements. Various kinds of climate are characterised, chiefly with regard to moisture and temperature, as continental which is dry, with great extremes of temperature ; insular or oceanic which is moist and very equable in temperature ; tropical in which the seasons depend chiefly upon the time of occurrence of rainfall ; temperate in which the seasons are chiefly dependent upon the variation of the daily course of the sun in the sky ; arctic in which the year is mainly two long periods of sun and no sun. Local climates, such as those of the Mediterranean in general and of the Riviera in particular, are dependent upon the geographical conditions of land and water. Every climate is to some extent affected by the geographical nature of the sur- roundings of the locality. Climatic Chart (see Weather Map, p. 88). A map showing the geographical distribution of some element of 60 Glossary. climate ; temperature, rainfall and sunshine are the most frequently charted. Charts of normal values of these elements for the British Isles for long periods of years ended 1910 are issued by the Meteorological Office. The word also covers any diagram representing the periodic or secular variation of a climatic factor. Climatology. The study of CLIMATE (q.v.). Clouds. Clouds have been classified into certain typical forms, but there are so many intermediate types that it is sometimes difficult to decide to what class any cloud belongs. Certain types of cloud and the direction from which they are moving indicate certain states of weather ; cer- tain types of cloud being usually within certain heights, a knowledge of cloud forms enables the observer to make a rough estimate of the height of clouds. For practical purposes clouds may be divided into cloud sheets and cloud heaps. CLOUD SHEETS. Many forms of cloud are obviously in more or less extended sheets, sometimes covering the whole sky, sometimes only covering a small patch. They vary much in thickness, sometimes no blue sky is visible through the sheet, sometimes small patches of blue are visible, sometimes half the sky is blue, at other times the sheet of cloud is represented by a few detached clouds in the blue. The formation of a sheet of clouds is not well under- stood. But the following may be on some occasions the method of formation ; it is known that the atmosphere is to a certain extent stratified, and that there may be damp Cloud*. 61 and dry layers ; if the pressure over any area is diminished the air in the region will be cooled by expansion ; if sufficiently cooled the dew point will be reached, and any damp layer will become a cloudy layer. Cloud sheets are often seen to be broken up into waves ; the waves may not be formed in the cloud itself ; if waves are set up in the atmosphere they will be propa- gated upwards and downwards ; in a thin cloud layer the air that rises on the wave crest is cooled by expansion, more condensation occurs and the cloud is thicker ; the air that descends in the hollow is warmed by compression and some of the cloud is evaporated, leaving a clear or nearly clear space. Cloud sheets may sometimes be seen forming at several levels at the same time ; over a cyclonic depression there are probably sheets of cloud at several levels. Cloud sheets may be divided up into, three classes which differ in appearance and height. The upper layer, the cirrus clouds, are at heights of from 25,000 to 30,000 feet ; the middle layer, the alto clouds, are from 10,000 to 25,000 feet ; the lower layer clouds are below 10,000 feet. THE UPPER LAYER. The clouds of the upper layer, the CIRRUS CLOUDS, are composed of ice particles, not of water drops as all other clouds. They are of a pure white colour with no shadows except when seen when the sun is low down. They are usually streaky and have a brushed out appearance (mares' tails); they are frequently spoken of afe windy looking clouds, but in spite of this appearance they do not necessarily indicate wind. They are often seen in long streaks stretching across the sky, sometimes in parallel bands, serming by 62 Glossary. the effect of perspective to converge, on the horizon to a V-point. These clouds often move away from the centres of depressions ; when coming from North- West, West or South-West they indicate a depression in a westerly direction, which will probably spread over the observer ; when seen coming from the North they indi- cate a depression to the Northward which will probably move away to the North-East and so not influence the weather. To observe the motion of clouds, especially of the high clouds, get a patch of cloud " on " with the point of a branch or some point of a building; if one moves so as to keep the cloud on the point, one's direction of motion will be towards the direction from which the cloud is coming. It is often difficult to determine the motion of the high clouds without a careful observation of this kind. Bands of cloud do not necessarily move in the direction of their length, nor do parallel bands of clouds necessarily move from their V-point. HALOS and MOCK SUNS (sun dogs) are seen at times in cirrus clouds but in no other forms ; a halo is a ring, sometimes coloured, seen at some distance (22 or more rarely 46) from the sun or moon. Besides the wisps and streaks of cloud of common cirrus there are other forms : CIRRO-CUMULUS : lines or groups of cloud, some- times detached globules, sometimes waves ; mackerel sky (French : moutons) ; these clouds are pure white, with 110 shadows. ClRRO-STRATUS : a bank of tangled web sometimes overspreading the whole sky ; no shadows ; sun pale and " watery." Cloud*. G;> ClRRO-NEBULA : similar to last but no visible structure ; a veil of pale white cloud. Precedes depressions. THE MIDDLE LAYER, ALTO CLOUDS. The alto clouds are composed of water droplets, not ice particles, therefore halos are never seen ; CORON^E may be seen ; these are coloured bands quite close round the sun or moon. The alto clouds are not such a pure white as the cirrus clouds, and shadows are visible on them. They may move away from centres of low pressure, like the cirrus clouds. They are, on the average, only half the height of the cirrus clouds, being from 10,000 to 25,000 feet high. The following are the most important varieties : ALTO-CUMULUS. Very like cirro-cumulus, but the cloud masses are larger, and shadows are visible ; some- times arranged in globular masses (French : Gros moutons), sometimes in waves. ALTO-CUMULUS CASTELLATUS, TURRET CLOUD. The globular masses of Alto-cumulus are developed upwards into hard-edged clouds like miniature cumulus ; some- times a large number of clouds almost of exactly the 'same shape are seen. When coming from a southerly or westerly point, after fine weather, Turret cloud is a sign of approaching thundery conditions. ALTO-STRATUS. Very like cirro-stratus, but a thicker and greyer cloud ; halos never seen ; precedes depres- sions, and does not usually extend so far from the centre as cirro-stratus. , THE LOWER LAYER. The clouds of the lower layer are below 10,000 feet ; thicker clouds with not such a fine structure as the higher clouds. (54 Glossary. STRATO-CUMULUS. Masses of cloud with some vertical structure appearing in rolls or waves, sometimes covering the whole sky ; sometimes the whole sky is covered with cloud, but the hollows of the waves are lighter and obviously thinner ; blue sky is occasionally seen more or less plainly through the thinner parts. This cloud is common in quiet weather in winter, and sometimes per- sists for many days together. In summer it is more broken up, and tends to turn into cumulus clouds. STRATUS. A uniform layer of cloud which resembles a fog, but does not rest on the ground. Small masses of stratus, more or less in the shape of a lens (lenticular cloud), are often seen near thunder clouds ; they appear dark when seen against the white sides of the thunder clouds. NIMBUS. A dark shapeless cloud without structure, from which continuous rain or snow falls ; through open- ings in this cloud, should such occur, layers of cirro-stratus or alto-stratus may almost always be seen. SCUD. Small shapeless clouds with ragged edges ; sometimes seen without other cloud, but usually asso- ciated with nimbus and cumulus. . HEAP CLOUDS. Clouds with considerable vertical structure, not forming horizontal sheets of cloud. Their formation is due to rising currents of air ; as the air rises it is cooled by expansion till it reaches the dew point when cloud begins to form ; as soon as condensation takes place heat is liberated, and though the air cools as it rises still further, it does not cool so rapidly as it would were no heat liberated by the condensing water vapour. The rising current is due to air heated above its surroundings, and the condensation enables the current to rise considerably Between pp. 64 and 65. CLOUDS SEEN FROM BELOW : FIGURE 1. :Strato-Cumulus from below. Cumulo-Nimbus (Thunder-cloud) with "Anvil" extension of. false cirrns. 30th October. 1915. CLOUDS SEEN FROM ABOVE : FIGURE 2. Strato-cuinulus from an aeroplane (4,000 feet). filled with fog, 300 feet deep, in early morning after still night. Clouds. 65 higher than it would have done had no extra heat been available. SIMPLE CUMULUS. This may consist of small clouds with flat base and rounded top or large cauliflower shaped clouds (Woolpack cloud). These clouds commonly form on a summer day, beginning as small clouds, and growing larger by the early afternoon, when the rising currents due to the sun's heat are at a maximum ; they usually disappear before sunset. CUMULO-NIMBUS ; SHOWER CLOUD ; THUNDER CLOUD. Sometimes when cumulus grows to large propor- tions the upper edge is seen to become soft and brushed out into forms somewhat like cirrus (false cirrus) ; at the same time from the under edge of the cloud, which has been growing very dark, rain begins to fall. Sometimes only a slight shower may result ; but if the cloud is large there may be heavy rains and violent thunderstorms. Sometimes the false cirrus is brushed out round the top of the cloud giving it the shape of an anvil ; the anvil cloud is usually associated with very violent thunder- storms and falls of hail. Thunder clouds due to the rising currents of a hot day usually disappear about the time of sunset, but often leave the false cirrus. Cumulus often forms in long lines of cloud presenting the appearance of a succession of clouds or of a wall of cloud extending along the horizon. Any cause that brings masses of air of different temperatures near together brings about the formation of cumulus cloud. It should be noted that rain, hail, and snow only fall from nimbus or cumulo-nimbus clouds, except the slightest and most transient showers, which may some- times fall from alto-cumulus or strato-cumulus. Nimbus causes persistent rain or snow, cumulo-nimbus more or less severe showers of rain, hail or snow. 13204 C 66 Glossary. TABULAR STATEMENT OF THE SEVERAL TYPES OF CLOUDS. I. CLOUD SHEETS. CIRRUS. Mares' tails ; wisps or lines of pure white clouds with no shadows. Upper cloud layer about 30,000 feet. Clouds composed of ice crystals. With these are sometimes seen halos, or rings, at some distance from the sun and moon. Middle cloud layer ; 10,000 feet to 25,000 feet. Clouds com- posed of minute drops of water. Coloured rings sometimes seen quite close to sun and moon, but never halos. Lower cloud layer. Below 7,000 feet.* CIRRO-CUMULUS. Small speckles and flocks of white clouds ; fine ripple clouds : mackerel sky. CIRRO-STRATUS. A thin sheet of tangled web structure, sometimes covering the whole sky ; watery sun or moon. CIRRO-NEBULA. Similar to last, but a veil of cloud with no visible structure. ALTO-CUMULUS. Somewhat similar to cirro- cumulus, but the cloud masses are larger, and show some shadow. ALTO-CUMULUS CASTELLATUS. Turret cloud; alto-cumulus with upper margins of the cloud masses developed upwards into miniature cumulus, with hard upper edges. (Sign of thunder.) ALTO-STRATUS. Very like cirro-stratus and cirro-nebula, but a thicker and darker cloud. STRATO-CUMULUS. Cloud masses with some vertical structure ; rolls or waves some- times covering the whole sky. STRATUS. A uniform layer of cloud resembling a fog but not resting on the ground. NIMBUS. Shapeless cloud without structure, from which falls continuous rain or snow. SCUD. Small shapeless clouds with ragged edges ; sometimes seen without other cloud, especially in hilly country ; but more commonly seen below other clouds, such as cumulus and nimbus. *xThe heights given are only approximate. Thus lower clouds of the strato-cumulus type may rise above 7,000 feet and attain at times the height of the middle cloud layer. Clouds. 67 II. HEAP CLOUDS. CUMULUS. (Woolpack clouds) ; clouds with flat base and con- siderable vertical height. Cauliflower-shaped top. FKACTO-CUMULUS. Small cumulus with ragged tops. CUMULO-NIMBUS. (Anvil-, thunder- or shower-cloud). Towering cumulus with the top brushed out in soft wisps or larger masses (false cirrus) and rain cloud at base. The height of the heap clouds is very variable, Mean height of base, about 4.500 feet ; the height of the top varies from about 6,000 to 25,000 feet. Reproductions of photographs of Strato-cumulus seen from below and from above, also of Cumulo-Nimbus and of a Valley filled with fog are inserted between pages 64 and 65. A photograph representing Cirrus faces page 58, and one representing Mammato-cunmlus, page 190. Cloud-burst. A term commonly used for very heavy thunder-rain. Extremely heavy downpours are sometimes recorded, which in the course of a very short time tear up the ground and fill up gulleys and watercourses ; this happens in hilly and mountainous districts, and is probably due to the sudden cessation of conyectional movement, caused possibly by the supply of warm air from the lower part of the atmosphere being cut off as the storrn moves over a mountain range. With the cessation of the upward current, the raindrops and hail- stones which it had been supporting must fall in a much shorter time than they would have done had the ascensional movement continued. Clouds, Weight of. Measurements on the Austrian Alps of the quantity of water suspended in clouds have given 0*35 g/m 3 to 4- 8 g/m 3 . The water suspended as mist, fog, or cloud may be taken as ranging from 0*1 to 5 g/m 3 . (See Wegener's Thermodynamics of the Atmo- sphere, p. 262.) 13204 C 2 68 Glossary. Col. The neck of relatively low pressure separating two anticyclones (see ISOBARS). One of the most treacherous types of barometric distribution, as it some- times marks a locality of brilliantly fine weather and sometimes is broken up by thunderstorms. See Plate XIV. Compass. A circumference, or dial, graduated into thirty-two equal parts by the points N., N. by E., N.N.E., N.E. by E., N.E. and so on. The cardinal points of the compass are North, South, East and West. The points of the compass are often called ORIENTATION points. A MAGNETIC COMPASS, or MARINER'S COMPASS, is a compass or orientation-card having attached to it one or a series of parallel magnets and supported so that it may turn freely in a horizontal plane. Any magnet sets itself parallel to what is termed the magnetic meridian, but it is only some few countries that have a magnetic meridian the same as the geographical meridian ; they include narrow strips in Arabia, European Russia, Finland, and North Lapland. Even at places within temperate latitudes the angle between the two may amount to 50 or more, and in the Arctic or Antarctic regions the direction of the compass needle may be the opposite of the geographical north and south line. In the neighbourhood of London the needle points about 15 to the west of north, and the amount is slowly decreasing. In the west of Ireland the declination, or variation, as it is called, is more than 20 W. Public wind-vanes are often incorrectly set according to magnetic north instead of true north ; caution must be exercised accordingly. All maps are set out according to true north ; often an orientation mark is given to show the variation of the compass or magnetic declination. Compass. 69 A compass-needle is influenced by magnetic material in its neighbourhood, and therefore in practice should not be used near to articles made of iron or steel. The variation of the compass and correction of its error? are matters of primary importance for aircraft pilots and are provided for by a special manual. Component. A word used to indicate the steps, in their various directions, which must be compounded or combined geometrically in order to produce a given displacement. For example a man going ten steps upstairs arrives in the same position as if he took ten treads forward on the level and then ten rises straight upwards, or equally if he first went up ten rises and then took ten treads forward. The actual distance travelled is the combination of the horizontal distance and the vertical rise. We call the horizontal distance and the vertical rise the components arid the actual distance the geometrical sum or the resultant of the components. It is evident that it is not necessary that the components should be at right angles to each other as the horizontal and vertical are. Any displacement AC is the geometrical sum of the two components AB and BC,, wherever B may be. This analysis of displacement into components, or combination of component displacements to form a resultant displacement, finds an effective illustration in the effect of leeway on the performance of aircraft. The aircraft must necessarily be carried along by the air in which it travels. If we suppose AB to be the travel of the aircraft through the air in an hour, and BC to represent the travel of the wind in the hour, AC, the resultant, will represent the travel of the aircraft with reference to the fixed earth, or its performance. BC is the leeway, AB is the headway made through the air. 70 Glossary. Hence the performance is the geometrical sum of the headway and the leeway. See VECTOR. The law of composition of displacements which is thus set out is equally applicable to the composition of velocities, accelerations, and forces. The resultant is always the geometrical sum of the components, and to obtain the geometrical sura, set out a step or line representing the first component, then another step representing the second component : the resultant is the step from the beginning of the first component to the end of the second. The simplest way of dealing with the magnitudes and angles which come into questions involving geometrical composition is to make a scale-drawing and measure them with a rule and protractor. That is generally accurate enough for most purposes, but the methods of the solution of triangles can be employed if high numerical accuracy is wanted. Condensation. The process of formation of a liquid from its vapour. See AQUEOUS VAPOUR. Conduction. 71 Conduction. The process by which heat is trans- ferred by and through matter, from places of high to places of low temperature, without transfer of the matter itself, the process being one of '* handing on " of the heat- energy between adjacent portions of matter. It is the process by which heat passes through solids ; in fluids, although, it occurs, its effects are usually negligible in comparison with those of CONVECTION. See also p. 146. Convection. In convection heat is carried from one place to another by the bodily transfer of the matter containing it. In general, if a part of a fluid, whether liquid or gaseous, is warmed, its volume is increased, and the weight per unit of volume is less than before. The warmed part therefore rises and its place is taken by fresh fluid which is warmed in turn. Conversely, if it is cooled it sinks. Consequently, if heat is supplied to the lower part of a mass of fluid, the heat is disseminated throughout the whole mass by convection, or if the upper part is cooled the temperature of the whole mass is lowered by a similar process. There are two apparent and important exceptions in meteorology. Fresh water, when below the temperature of 39*1 F., 277a., expands instead of contracting on being further cooled. Hence a pond or lake is cooled bodily down to 39*1 but no further, as winter chills the surface before it freezes. Secondly, heat applied to the bottom of the atmosphere may stay there without being dis- seminated upwards when the atmosphere is exceptionally stable in the circumstances explained under ENTROPY. Corona. A coloured ring, or a series of coloured rings, usually of about 5 radius, surrounding the sun or moon. The space immediately adjacent to the luminary is 72 Glossary. bluish-white, while this region is bounded on the outside by a brownish red ring, these two together forming the " aureole." In some cases the aureole alone appears, but a complete corona has a set of coloured rings surrounding the aureole, violet inside followed by blue, green, yellow to red on the outside. This series may be repeated several times outwards. The corona is produced by diffraction, that is by the bending of rays of light round tl^e edge of small particles, in this case minute water drops, but sometimes dust (see BISHOP'S RING). If the diffracting particles are of uniform size the colours are pure ; a mix- ture of many sizes may give the aureole only. The more numerous the particles involved, the greater is the intensity of the colours, while the radius of the corona is inversely proportional to the size of the particles. Thus a corona whose size is increasing indicates that the water particles are diminishing in size, and vice versa. The corona is distinguished from the halo, which is due to REFRACTION, by the fact that the colour sequence is opposite in the two, the red of the halo being inside, that of the corona outside. Correction. The alteration of the reading of an instrument in order to allow for unavoidable errors in measurement. The measurement of nearly all quantities is an indirect process, and generally takes the form of reading the position of a pointer or index on a scale. When we wish to know the pressure of the atmosphere we read an index on the scale of a barometer ; when we wish to know the temperature of the air we read the position of the end of a thread of mercury in a ther- mometer ; to determine a height we use the pressure, though the scale may be graduated in feet or metres. Correction. 73 Almost all measurements are, in fact, ultimately re- duced to reading a position or length on a graduated scale and, generally speaking, the reading depends mainly, it is true, on the quantity which the instrument is in- tended to measure, but also partly upon other quantities. Thus the readings of barometers are generally affected by temperature as well as pressure, those of thermometers by alterations in the glass containing the mercury or spirit. It is the object of the designer and maker of instruments to get rid of these disturbances of the reading as far as possible, either by the selection of special materials or by introducing some device whereby the disturbing effect is automatically corrected. In that case the error, and often the instrument, is said to be compensated. But in most cases the amount of the error has to be determined and allowed for by a suitable correction. An ANEROID BAROMETER is often compensated for tempera- ture, and for a mercury barometer the effect of temperature is made out and tabulated and a correction introduced, which is derived from a table, when the temperature of the " attached thermometer " has been noted. In a similar manner the correction of a barometer reading for the variation of GRAVITY at different parts of the earth's sur- face is worked by means of tables, the variation of gravity with latitude having been previously reduced to a formula, by means of observations from which the figui e of the earth has been determined and the change of gravity has been ascertained. In some measurements, such as the determination of height by the use of an aneroid barometer, corrections are numerous and complicated : the uncorrected reading may even be only a rough approximation not sufficiently accurate for practical purposes. 74 Glossary. Correlation. Two varying quantities are said to be correlated when their variations from their respective mean values are in some way or other mutually connected with each other. A kind of measure of this connection is given by the " correlation coefficient." This is a decimal lying between + 1 and 1, which is easily calculated. Values of + 1 and of 1 show that the two quantities -are directly or inversely proportional ; that is to say, for any departure from the normal of the one quantity there is a corre- sponding departure from the normal of the other, always in the same sense or always in the opposite sense. On the other hand values that are nearly nothing show that there is little if any connexion. It may be of interest to state that the coefficient of correlation between the phases of the moon and the barometric pressure at Greenwich is insignificant. Modern statistical methods have been frequently applied to meteorological problems. Mr. W. H. Dines, F.R.S., has used the method of correlation with great success in his work on the upper air. His results are published by the Meteorological Office in Geophysical Memoirs No. 2. As examples of correlation coefficients we may give those found by Mr. Dines between the pressure at a height of 9 kilometres and the mean temperature of the air column from 1 to 9 kilometres. For different sets of observations the values found were '88, '96, 1 90, -90, 94. The inference to be drawn from such correlation coefficients is that the variations of temperature of the air column from 1 to 9 kilometres are directly dependent upon the pressure at the top of the air column. Mr. Dines gives correlation coefficients from work on the upper air in various other papers. Correlation. 75 Sir" Gilbert Walker, of the Indian Meteorological Service, has used the method of correlation to predict the amount of the Indian Monsoon Rainfall. He has also investigated the connection between sunspots and tem- perature, sunspots and rainfall, sunspots and pressure by statistical methods. From, various papers the following examples of correla- tion coefficients have been selected : Correla- tion Coefficient. Number of Observations. Variables Correlated. Reference. *49 42 (1867-1908). Height of Nile flood and South Ameri- can mean pressure in March, April, May. G. T. Walker, Cor- relation in Seasonal Variations of Weather. (Memoirs of the Indian Met. Dept., Vol. XXI., Part II.) -'47 30 (1880-1909). Rainfall at Java, October to March, with rainfall at Trinidad the fol- lowing six months. R. C. Mossman. South- ern Hemisphere Seasonal Correla- tions. (Symons Met. Mag., Vol. 48, 1913.) -'43 34 (1877-1910). Annual mean tem- perature at Cairo and annual mean temperature in Eng- land, S.W., and South Wales. J. I. Craig, a See-Saw of Temperature. (Quar. Journal, Roy. Met. Soc., April 1915.) -62 36 (1875-1910). Mean pressure for March at Azores and in Iceland. J. P. Van der Stok. 76 Glossary. Correla- tion Coefficient. 78 -81 80 Number of Observations. 20 (1891-1910). 47 (1869-1915). 20 (1896-1915). Variables Correlated. March barometric gradient at Zikawei (China') and July- August air tem- perature at Miyako in N.E. Japan. Mean pressure for March at Kew and rainfall total for same month at Kew. Sunspot numbers and level of Lake Victoria. Reference. T. Okada (Monthly Weather Review, U.S.A., Jan,, 1916). E. H. Chapman, The Relation between Atmospheric Pres- sure and Rainfall. (Quar. Journal, Roy. Met. Soc., 1916.) M.O. MS. Cosecant. In a right angled triangle the ratio of the hypotenuse to one side is the cosecant of the angle opposite to that side ; the ratio is the reciprocal of the sine. See SINE. Cosine. In a right angled triangle the ratio of one side to the hypotenuse is the cosine of the angle between them. See SINE. Cotangent. In a right angled triangle the ratio of the two sides that form the right-angle is the cotangent of the angle opposite the side taken as the divisor. See SINE. Counter sun. See ANTHELION. Cumulo-stratus. 77 CumulO-StratUS. The name given to a certain com- bination of cloud forms which is no longer used in the international classification. See CLOUDS. Cumulus. The technical name of the woolpack cloud. See CLOUDS. . A name given to a region of low barometric pressure ; now usually spoken of as a DEPRESSION or a LOW. See ISOBARS. Cyclostrophic. See GRADIENT WIND. Damp Air. As distinguished from dry air in me- teorology, damp air implies a high degree of RELATIVE HUMIDITY (q.v.). When its relative humidity equals or exceeds 85 per cent, of saturation air may fairly be called damp. It will deposit some of its moisture in dry woollen fabrics, cordage or other fibrous material, though its water will not condense upon an exposed surface until 100 per cent, is reached. Even the driest air of the atmosphere contains some water vapour, and its relative dampness or dryness can be changed by altering the temperature. Thus the same air may be very dry at 2 o'clock in the afternoon, and very damp, even cloudy, at 8 o'clock in the evening, simply because its tempera- ture has been lowered. At any time of the year, but especially in summer, the dampness of the air is subject to great changes. Day Breeze. See SEA BREEZE or BREEZE. Debacle. Breaking up of the ice in the spring in rivers and seas ; it lasts from two to six weeks, and takes place between the end of January and the beginning of 78 Glossary. May, varying according to locality. The waters are usually free from ice by the end of April-May. At " debacle " the water in rivers rises to the extent of inundating the country for miles around, sometimes stopping all carriage traffic ferry-boats taking their place. This condition may last for as long as three weeks. In Russia there are some 110 stations at which obser- vations of the "debacle" are taken. The event takes place earliest in the Caspian, Black and Azov Seas : it commences at the end of February, and the sea along the coast is free of ice by the end of March the open sea being clear by the end of February. In the Pacific the phenomenon takes place in April, and the sea is clear by May ; in the Baltic usually in March or the beginning of April, the sea being cleared by the beginning of May, but at Reval and Libau this may occur even by the end of March. At Uleaborg the "debacle" is later, there being ice in the sea till the end of May ; and in the Arctic Sea still later, about the end of April, and the sea is not clear till the beginning of June. In Canada, in Ontario, the occurrence takes place in March, freeing the waters by April, and the same is true in the case of the Maritime provinces. In the St. Lawrence it ia a little later, the river being free of ice in May. Dekad, in Meteorology, a period of ten days, but decade is often, used for ten years. Density. The density of air is one of the things you have to know when you want to calculate the lifting power of a balloon of given size. As applied to air, density is a difficult word to explain because the numeri- cal value depends partly upon the composition of the air, partly upon its pressure, and again partly upon its tern- Density. 79 perature. Thus it is often said that moist air or damp air is lighter than dry air, warm air is lighter than cold air, and rarefied air is lighter than compressed air, and all these statements are true provided that in each case we remember to introduce the condition "other things being equal." By the density of a sample of air is to be understood the weight, or better, the mass of a measured volume, a cubic foot, or a cubic metre ; and moist air is lighter than dry air in this sense that a cubic metre of perfectly dry air weighs 1,206 grammes when its barometric pressure is 1,000 millibars, 29*53 inches, and its temperature is 289a. (60*8 C F.), whereas if it were saturated air instead of dry air the cubic metre would weigh 1,197 grammes. But if the barometric pressure rose while the change from dry air to saturation was being effected the density would rise in like proportion ; and if the temperature changed the density would change in reversed proportion to the absolute temperature. The formula for the density of air is, therefore, a complicated one A- where A is the density to be computed, A is the density of perfectly dry air at pressure^ and temperature T . If p = 1,000 mb., 29-53 in., and T = 290a. then A = 1,201 g/m 3 . p is the barometric pressure in mb. of the sample. e is the pressure of aqueous vapour in the sample. The following is a complete example of the deter- mination of the density of a sample of air from a reading of the barometer and wet and dry bulb thermometers. 80 Glossary. Barometer corrected for temperature 1 ,010*1 mb. (29*83 in.). Dry bulb 285*8a. (55*1 P.). Wet bulb 281-3a. (47*0 P.). Vapour pressure from humidity tables = 8*2 mb. (0*239 in.). A = 1,201 g/m3, 1,010-1 - 8*2 x 3/8 290 A = 1,201 x - i )() 00 x 285^8 g / m ' = 1,227 g/m, or 0*0766 Ibs. per cubic foot. Note, One grain per cubic foot is equivalent to 2-29 g/m 3 . Since the reading of a barometer gives the pressure at the level of the mercury in the barometer cistern it must be understood that the density refers to the sample of air at that level. For the value at any other level a correction for level must be made. The variation of the density of air with pressure and temperature is of great importance in meteorology. Pressure is of course higher in an anticyclone than in a cyclonic depression, and it has recently been made cer- tain that above one kilometre (3,281 feet) the air in the high pressure is warmer than in the low pressure, so the effect of difference of pressure may nearly counter- balance the effect of temperature, and in consequence the density of a column of air in a cyclone may be very little different from that in an anticyclone. The influence of moisture upon the density of air is not regarded as having the importance in meteorology which used to be attributed to it when it was held to explain the difference between high and low pressure with the accompanying weather. Density. 81 The following are the average pressures, temperatures and densities of air at different levels above high pressure and low pressure respectively in the British Isles accord- ing to the results obtained with registering balloons by Mr. W. H. Dines, F.R.S. TABLE o^ AVERAGE VALUES OF THE PRESSURE, TEM- PERATURE AND DENSITY OF AIR IN A REGION OF HIGH AND OF LOW PRESSURE. High Pressure. Low Pressure. Height. Pres- sure. Temp. Density. Pres- sure. Temp. Density. lOOO-ft. k. mb. a. g/m 3 mb. a. g/m 3 32*809 IO 273 226 421 247 225 382 29-528. 9 317 2 33 474 288 226 444 26 * 247 8 366 240 53i 335 227 SH 22 * 966 7 422 247 595 388 232 583 19-685 6 483 254 662 449 240 652 16-406 5 552 261 736 5i6 248 724 13-124 4 628 267 818 59i 255 807 9^43 3 713 272 911 675 263 893 6-562 2 807 277 IOI2 767 269 992 3-28l I 913 279 H37 870 275 IIOO o o 1031 282 I27O 984 279 1226 The densities quoted above are calculated on the assumption that the relative humidity is 75 per cent. 82 Glossary. The following are the densities of a few other substances. 'Hydrogen (dry) 88'74 g/m 3 *Carbonic acid gas (dry)... 1,953 g/m 3 Water 1-000 g/cc (at 277a.) Sea water I'Ol to T05 g/cc. Mercury 13'596 g/cc (at 273a.) Petrol ... ... 0-68 to O72 g/cc. See also under BUOYANCY. Depression. A region of low barometric pressure surrounded on all sides by higher pressures. See ISOBARS, and Plate XL Dew. The name . given to the deposit of drops of water which forms upon grass, leaves, &c., when they become cooled, by radiating heat to the sky on a clear night, to such an extent that their temperature is below the saturation or DEW POINT of the air which surrounds them. The last part of the process of the formation of dew is in no way different from that which operates when a glass of ice-cooled water covers itself with water drops indoors, or when a deposit of moisture is formed by breathing on a window-pane. Dew-point. The temperature of saturation of air, that is to say, the temperature which marks the limit to which air can be cooled without causing condensation, either in the form of cloud if the cooling is taking place in the free air, or on the sides of the vessel if it is enclosed. See AQUEOUS VAPOUR. Diathermancy. Diathermanous. The power of allowing heat in the form of radiation to pass in the same way that light passes through glass. Rock salt is * At a pressure of 1,000 mb. and temperature of 273a. Diathermancy. 83 peculiarly diathermanous, water, on the contrary, and glass are not. All forms of energy which are " radiated " and in connection with which the word " ray " is used, such as rays of light, heat rays, X-rays, radio-telegraphic rays, travel by wave motion and have some properties in common. One of the most characteristic is that described as transparency and opacity with regard to light. But the same substances are not similarly transparent for all kinds of rays. X-rays and electric rays make no difficulty about going through walls which stop light ; and sound- waves often find a way where light cannot follow. The question of transparency and opacity for different kinds of rays is one of bewildering complexity. The study of DIATHERMANCY deals with that part of the subject which is concerned with the transmission of heat in the form of wave motion. Diffraction. The process by which rays of different colour are separated one from another when a beam of light passes an obstacle of any shape. In reality the shape of the obstacle must be carefully chosen in relation to the shape of the beam to make the phenomenon easily apparent. Perhaps the simplest experiment is to draw a greasy finger across a plate of glass and to look through the glass at the bright line of an incandescent electric lamp, taking care that the plate is turned so that the lines left by the finger on the glass are parallel to the bright line. Those who are unfamiliar with the experiment will be surprised at the brilliancy of the colours which are produced by the simple process. In more scientific form when the lines are ruled regularly on the glass by a suitable dividing engine we get a diffraction-grating, one of the most delicate of all optical instruments. 84 Glossary. For scientific experiments in diffraction either a bright line of light formed by a slit in front of a lamp with a linear obstacle, or a bright point of light with a circular obstacle may be used, and many remarkable results can be shown with these simple means. For example, when the distances are properly adjusted a bright spot will be found at the central point of the shadow of a circular disc, thrown by a bright point of light, and again on looking past a needle at a line of light parallel to the needle a great play of colours will be seen. The phenomena of diffraction are explained by the hypothesis that the actual transmission of light is not a direct projection of the light along straight lines from any luminous point, but the spreading out of waves with a spherical wave-front central at the luminous point. They are exhibited in the atmosphere principally by the formation of CORONAE round the sun and moon, and sometimes also by the IRIDESCENCE of CLOUDS. Diffusion. The slow molecular process by which supernatant fluids mix in spite of the differences in their density. The molecular forces or motions which come into play in diffusion are perhaps most effectively illustrated by the tenacity with which the mixture main- tains its composition when once the mixing has taken place. For example, whisky is lighter than water and a separate layer of the spirit can be floated on the top of water by judicious manipulation. The spirit and water will then slowly mix by diffusion, even if there be 110 stirring, due to thermal convection or mechanical opera- tion. But when the spirit and water have become mixed no amount of allowing to stand will cause the water to settle to the bottom and leave the whisky in a separate Diffusion. 85 layer at the top. Once mixed they are mixed for ever, owing to the power of diffusion. The process of diffusion follows certain definite rules, which are similar in type to those for the diffusion of heat by thermal conduction, and the diffusion of velocity through a viscous mass, but it takes so long a time for any appreciable effect to be produced, that in practice diffu- sion only completes the process of mixing which has been begun by stirring or convection. Major G. I. Taylor has recently shown that in the atmosphere mixing by tur- bulent motion (see EDDY) follows a similar law, but with a different characteristic constant that brings diffusion by turbulent motion among the operative forces in Meteorology. Diurnal. The word, which means " recurring day by day," is used to indicate the changes in the meteorological elements which take place within the twenty-four hours of the day. Thus, by the diurnal change of pressure is meant a slight rise of the barometric pressure between about 4 a.m. and 10 a.m., and between 16 h. (4 p.m.) and 22 h. (10 p.m.) with corresponding falls between. In this change it is the " semidiurnal " variation which is the most striking because it occurs (with different intensity) all round a whole meridian simultaneously, and sweeps round the globe from meridian to meridian about three and a half hours in front of the sun. Other elements also show noteworthy diurnal variations. We give here diagrams showing the diurnal variation of pressure, temperature, humidity and wind velocity at Kew for January and July as representing winter and summer respectively. 86 DIURNAL VARIATION IN SUMMER. HOURS c so-G CD O c 70-ti 60-: 3 6 9 12 15 18 21 HLMIDITY i\ i-60 HOURS so .310145-1 -2996u -/- WIN _ o 6 - --4 <U 294 <28St 286- TEMPERATURi: :-55 From the Normals for July. Averages of not less than twenty-five years of Hourly Readings at Kew Observatory. For the Diurnal Variation of wind-speed through the different months at the summit of 4 the Eiffel Tower, Paris, see p. 287, 87 DIURNAL VARIATION IN WINTER. HOURS 12 15 18 21 2t HOUR5 HUMIC (D " 90-E 80-: 70-: JIOI65.E fioi&al PRESS ITY E-so : -70 -30005 /VIN O ^ j cu O 4~" C Q ~ 1-3+ a;280^ ^275 PERATURE <274- : 40 ^ 35 From the Normals for January. Averages of not less than twenty-five years of Hourly Readings at Kew Observatory. 88 Glossary. Doldrums.- The equatorial belt of calms and light variable airs, accompanied by heavy rains, thunderstorms and squalls. The belt follows the sun in his movement North and South, but its movement is not so great as that of the sun, and lags behind it by one to two months. Drought. Dryness due to lack of rain. According to the classification of the British Rainfall Organization an absolute drought is a period of more than 14 consecutive days without one-hundredth of an inch of rain on any one day, and a partial drought is a period of more than 28 consecutive days the mean rainfall of which does not exceed * 01 inch per day, or the total fall for the 28 days at most barely exceeds a quarter of an inch. Dry Air. The words are used in two senses. In a book on physics or chemistry dry air means air that carries no water- vapour at all, but in ordinary practice it is used for the atmosphere when it contains a smaller proportion of water-vapour than usual. Water evaporates from wet surfaces exposed to the air unless the atmos- phere is completely saturated. If we call air containing 85 per cent, or more of the possible amount of water- vapour damp, we may call air with less than 60 per cent, dry, and understand thereby that EVAPORATION is rapid and roads, grass, &c., dry quickly. The letter " y " has recently been added to the Beaufort Notation (see Weather Map, p. 10) to signify air with less than 60 per cent, of the possible amount of water-vapour. The following table shows for different dry-bulb tem- peratures the smallest depression of the wet-bulb which would justify the use of the letter "y" in reporting the " present weather." Dry Bulb. 89 Dry Bulb. Depression of Wet-bulb. a a 271 or below 1 273 (F.P.) or la above or below 2 275281 3 282292 4 293305 5 above 305 5^ It is found from the hourly readings of the Meteorological Office Observatories that out of a thousand hours Aberdeen has 37 hours of " dry air " thus defined, Valencia 6, Falmouth 19 and Kew 90. Had the limit been set at 50 per cent, instead of 60, the figures would have been only 4, 1, 1 and 19 respectively (see p. 224). Dry bulb. A curious name given to an ordinary thermometer used to determine the temperature of the air, in order to distinguish it from the wet bulb. No special precautions are taken to keep a " dry bulb " dry beyond protecting it from falling rain in a screen. But it is true enough that if the dry bulb gets a film of water upon it by condensation it will not be in a proper con- dition to record the temperature of the air until it has got quite dry again. Dynamics. The study of the motion of bodies in relation to the forces which control the motion. The fundamental principle of dynamics is that if a moving body is let alone it will go on moving. It is a vulgar error to suppose that it will stop. Dynamic Cooling*. The fall of temperature which occurs automatically throughout a mass of air when it expands on the release of pressure. (See ADIABATIC.) Examples of the expansion under reduced pressure and consequent cooling are to be found in the flow of air up a mountain slope with the formation of cloud at the top. 90 Glossary. Earth - Thermometer. A mercury - thermometer suspended in a tube sunk into the earth, or an electrical resistance thermometer buried in a trench usually at depths of one foot and four feet for measuring the temperature of the ground. See Observer's Handbook. Eddy. " The water that by some interruption in its course runs contrary to the direction Of the tide or current (Adm. Smyth) ; a circular motion in water, a small whirlpool," according to the New English Dictionary. Eddies are formed in water whenever the water flows rapidly past an obstacle. Numbers of them can be seen as little whirling dimples or depressions on the surface close to the side of a ship which is moving through the water. In the atmosphere similar eddies on a larger scale are shown by the little whirls of dust and leaves some- times formed at screet corners and other places which present suitable obstacles. The peculiarity of these wind eddies is that they seem to last for a little while with an independent existence of their own. They sometimes attain considerable dimensions and, in fact, they seem to pass by insensible degrees from the corner eddy to the whirlwind, the dust-storm, the waterspout, the tornado, the hurricane, and finally the cyclonic depression. It is not easy to draw the line and say where the mechanical effect of an obstacle has been lost, and the creation of a set of parallel circular isobars has begun, but it serves no useful purpose to class as identical phenomena the street corner eddy twenty feet high and six feet wide and the cyclonic depression a thousand miles across and three or four miles high. The special characteristic of every eddy is that it must have an axis to which the circular motion can be referred. The axis need not be straight nor need it be fixed in shape or position. The best example of an eddy is the vortex- Eddy. 91 ring or smoke ring which can be produced by suddenly projecting a puff of air, laden with smoke to make the motion visible, through a circular opening. In that case the axis of the eddy is ring shaped ; the circular motion is through the ring in the direction in which it is travelling and back again round the outside. The ring-eddy is very durable, but the condition of its durability is that the axis should form a ring. If the continuity of the ring is broken by some obstacle the eddy rapidly disappears in irregular motion. It is on that account that the eddy motion of the atmosphere is so difficult to deal with. When air flows past an obstacle a succession of incomplete eddies are periodically formed, detached, disintegrated and reformed. There is a pulsating formation of ill-defined eddies. The same kind of thing must occur when the wind blows on the face of a cliff, forming a cliff -eddy with an axis, roughly speaking, along the line of the cliff and the circular motion in a vertical plane. Whenever wind passes over the ground, even smooth ground, the air near the ground is full of partially formed, rapidly disintegrating eddies, and the motion is known as turbulent, to distinguish it from what is known as stream- line motion, in which there is no circular motion. The existence of these eddies is doubtless shown on an anemogram as gusts, but the axes of these eddies are so irregular that they have hitherto evaded classification. Irregular eddy motion is of great importance in meteoro- logy, because it represents the process by which the slow mixing of layers of air takes place, which is an essential part in the production of thick layers of fog. Moreover, all movements due to convection must give rise to current and return current which at least simulates eddy motion. Electrometer. An instrument for measuring elec- 92 Glossary. tromotive force, or potential difference. An ordinary battery has an electromotive force of at least a volt and shows a corresponding potential-difference on an electrometer when the poles of the battery are connected to the electrodes (connecting clamps) of the electrometer. A fully charged secondary battery shows a potential- difference of about two volts. In the atmosphere near the ground there is, on the average at most stations, a po- tential-difference exceeding 100 volts for a difference of level of one metre, due to the electrification of the air. It can be measured by an electrometer using a burning match or a water-dropper or a radio active substance as " collector." The potential-difference in the atmosphere measured in this way is very variable, especially during rain. The potential-difference necessary to cause a spark between two metal balls through one centimetre of air at ordinary pressure is about 30,000 volts, which suggests that the potential-differences necessary to produce a dis- charge of lightning are enormous. Energy. Used frequently in meteorology in the general sense of vigour or activity. Thus, a cyclone is said to develop greater energy when its character, as exhibited by a low barometer, steep gradients and strong winds, becomes more pronounced. But there is a technical dynamical sense of the word, the use of which is some- times required in meteorology, and which must become more general when the physical explanation of the phenomena of weather is studied, because all the pheno- mena of weather are examples of the " transformations of energy " in the physical sense. The most important conception with regard to energy is its division into two kinds, kinetic energy and potential energy, which are mutually convertible. A clock-weight gives a good idea of potential energy. When the clock Energy. 93 is wound up the weight has potential energy in virtue of its position ; it will utilise that energy in driving the clock until it is " run down " and can go no further. Potential energy must be restored to it by winding up before it can do any more driving. The potential energy in this case is measured by the amount of the weight and the vertical distance through which it is wound up. In dynamical measure the potential energy of the raised weight is mgh, where m is the mass of the clock- weight, h the vertical distance through which it is wound up, g the acceleration of gravity. It is to the mysterious action of gravity that the energy is due : hence the necessity for taking gravity into account in measuring the energy. Using the simple product mgh as a measure of the potential energy of gravitation, by a simple formula for bodies falling freely under the action of gravity, we have mgh = fynv 2 where v is the velocity acquired by a body falling through a height 7i, or, speaking in terms of energy, by losing the potential energy of the height h. It thus obtains a certain amount of motion which represents kinetic energy, in exchange for its potential energy. The kinetic energy is expressed by the apparently artificial formula rnv 2 . In virtue of its motion it has the power of doing "work" : if it were not for unavoidable friction the mass could 'get itself up-hill again through the height h by the use of its motion, and thus sacrifice its kinetic energy in favour of an equivalent amount of potential energy. The exchange of potential and kinetic energy can be seen going on in a high degree of perfection in a swinging pendulum. At the top of the swing the energy is all 94 Glossary. potential, at the bottom all kinetic. The swings get gradually smaller because in every swing a little of the energy is wasted in bending the cord or in overcoming the resistance of the air. What we get in return for the loss of energy in friction is a little HEAT, and one of the great conclusions of physical science in the middle of the nineteenth century was to show that heat is also a form of energy but a very special form, that is to say, its transformation is subject to peculiar laws. Heat is often measured by rise of temperature of water (or its equivalent in some other substance). Calling this form of energy thermal energy and measuring it by the product of the " water equivalent," M, and the rise of temperature AA^ produced therein, we have three forms of energy all convertible under certain laws, viz. : Potential energy ... mgli Kinetic energy ... ... \ mv" Thermal energy M (A A Q ) We have mentioned only a lifted clock-weight as an example of potential energy, but there are many others, a coiled spring that will fly back when it is let go, com- pressed gas in a cylinder that will drive an engine when it is turned on, every combination, in fact, that is dormant until it is set agoing and then becomes active. From the dynamical point of view, the study of nature is simply the study of transformations of energy. In meteorology kinetic energy is represented by the winds ; potential energy by the distribution of pressure at any level, by the electrical potential of the air and by the varying distribution of density in the atmosphere, causing convection ; thermal energy by the changes of temperature due to the effect of the sun or other causes. It is the Entropy. 95 study of the interchange of these forms of energy which constitutes the science of dynamical meteorology. Entropy. A term introduced by R. Clausius to be used with temperature to identify the thermal condition of a substance with regard to a transformation of its heat into some other form of energy. It involves one of the most difficult conceptions in the theory of heat, about which some confusion has arisen. The transformation of heat into other forms of energy, in other words, the use of heat to do work, is necessarily connected with the expansion of the working substance under its own pressure, as in the cylinder of a gas engine, and the condition of a given quantity of the substance at any stage of its operations is completely specified by its volume and its pressure. Generally speaking (for example, in the atmosphere) changes of volume and pressure go on simultaneously, but for simplifying ideas and leading on to calculation it is useful to suppose the stages to be kept separate, so that when the substance is expanding the pressure is maintained constant by supplying, in fact, the necessary quantity of heat to keep it so, and, on the other hand, when the pressure is being varied the volume is kept constant ; this again by the addition or subtraction of a suitable quantity of heat. While the change of pressure is in progress, and generally, also, while the change of volume is going on, the temperature is changing, and heat is passing into or out of the substance. The question arises whether the condition of the substance cannot be specified by the amount of heat that it has in store and the temperature that has been acquired just as completely as by the pressure and volume. To realise that idea it is necessary to regard the processes of supplying or removing heat and changing the temperature as separate and independent, and it is this 96 Glossary. step that makes the conception useful and at the same time difficult. . For we are accustomed to associate the warming of a substance, i.e. 9 the raising of its temperature, with supply- ing it with heat. If we wish to warm anything we put it near a fire and let it get warmer by taking in heat, but in thermodynamics we separate the change of temperature from the supply of heat altogether by supposing the substance is "working." Thus, when heat is supplied the temperature must not rise ; the substance must do a suitable amount of work instead ; and if heat is to be removed the temperature must be kept up by working upon the substance. The temperature can thus be kept constant while heat is supplied or removed. And, on the other hand, if the temperature is to be changed it must be changed dynamically not thermally, that is to say, by work done or received, not by heat communicated or removed. So we get two aspects of the process of the transforma- tion of heat into another form of energy by working, first, alterations of pressure and volume, each independ- ently, the adjustments being made by adding or removing heat as may be required, and secondly, alterations of heat and temperature independently, the adjustments being made by work done or received. Both represent the process of using heat to perform mechanical work or vice versa. In the mechanical aspect of the process, when we are considering an alteration of volume at constant pressure, p (vv ) is the work done, and in the thermal aspect of the process HH is the amount of heat disposed of. There is equality between the two. But if we consider more closely what happens in this 97 case we shall see that quantities of heat ought also to be regarded as a product, so that HH should be expressed as 7 7 (0~(/) ) where T is the absolute temperature and the entropy. The reason for this will be clear if we consider what happens if a substance works under adiabatic conditions, as we may suppose an isolated mass of air to do if it rises automatically in the atmosphere into regions of lower pressure, or conversely if it sinks. In that case it neither loses nor gains any heat by simple transference' across its boundary ; but as it is working it is drawing upon its store of heat, and its temperature. falls. If the process is arrested at any stage, part of the store of heat will have been lost through working, so in spite of the adiabatic isolation part of the heat has gone all the same. From the general thermodynamic properties of all substances, it is shown that it is not H, the store of heat, that remains the same in adiabatic changes, but EfT, the ratio of the stcre of heat to the temperature at which it entered. We call this ratio the entropy, and an adiabatic line which con- ditions thermal isolation and therefore equality of entropy is called an isentropic. If a new quantity of heat h is added at a temperature T the entropy is increased by hj T. If it is taken away again at a lower temperature T' the entropy is reduced by hfT'. In the technical language of thermodynamics the mechanical work for an elementary cycle of changes is dp. %v, and the element of heat c T. fy. The conversion of heat into some other form of energy by working is expressed by the equation 32\ 30 = Sp. Sv when heat is measured in dynamical units. 13204 D 93 Glossary. It is useful in meteorology to consider these aspects of the science of heat although they may seem to be far away from ordinary experience because, from certain aspects, the problem of dynamical meteorology seems to be more closely associated with these strange ideas than those which we regard as common. For example, it may seem natural to suppose that if we could succeed in com- pletely churning the atmosphere up to, say, 10 kilo- metres (6 miles) we should have got it uniform in temperature or isothermal throughout. That seems reasonable, because if we want to get a bath of liquid uniform in temperature throughout we stir it up ; but it is not true. In the case of the atmosphere there is the difference in pressure to deal with, and, in consequence of that, complete mixing up would result, not in equality, but in a differrnce of temperature of about 100 C. between top and bottom, supposing the whole atmosphere dry. The resulting state would not, in fact, be isothermal ; the temperature at any point would depend upon its level and there would be a temperature difference of 1 0. for every hundred metres. Bat it would be perfectly isentropic. The entropy would be the same everywhere throughout the whole mass. And its state would be very peculiar, for if you increased the entropy of any part of it by warming it slightly the warmed portion would go right to the top of the isentropic mass. It would find itself a little warmer, and therefore a little lighter specifically than its environment, all the way up. In this respect we may contrast the properties of an isentropic and an isothermal atmosphere. In an isen- tropic atmosphere each unit mass has the same entropy at nil levels, but the temperatures are lower in the upper levels. In an isothermal atmosphere the temperature Entropy. W the same at all levels, but the entropy is greater at the higher levels. An isothermal atmosphere represents great stability as regards vertical movements, any portion which is carried upward mechanically becomes colder than its surroundings und must sink again to its own place, but an ISEN TROPIC atmosphere is in the curious state of neutral equilibrium which is culled " labile." So long as it is not warmed or cooled it is immaterial to a particular specimen where it finds itself, but if it is warmed, ever so little, it must go to the top, or cooled, ever so little, to the bottom. In the actual atmosphere above the level of ten kilo- metres (more or less) the state is isothermal ; below that level, in consequence of convection, it tends towards the isentropic state, but stops short of reaching it by a variable amount in different levels. The condition is completely defined at any level by the statement of its entropy and its temperature, together with its composition which depends on the amount of water-vapour contained in it. Speaking in general terms the entropy increases, but only slightly, as we go upward from the surface through the TROPOSPHERE until the STRATOSPHERE is reached, and from the boundary upwards the entropy increases rapidly. If the atmosphere were free from the complications arising from the condensation of water- vapour the defi- nition of the state of a sample of air at any time by its temperature and entropy would be comparatively simple. High entropy and high level go together ; stability depends upon the air with the largest stock of entropy having found its level. In so far as the atmosphere approaches the isentropic state, results due to convection may be expected, but in so far as it approaches the isothermal state, and stability supervenes, convection becomes unlikely. 13204 D 2 100 Glossary. Equation of Time. The interval between two successive transits of the sun over the meridian is the true solar day, and the time based on this length is called apparent time. The length of the true solar day varies at different times of the year, and to avoid the inconveni- ence of want of uniformity in the length of the day, an imagiuary body called the mean sun may be supposed to revolve uniformly round the Earth and complete each revolution in a time equal to the average length of the true solar day ; the time referred to this standard is called mean time. To convert mean time into apparent time, and vice versa, the correction known as the equation of time must be applied. The equation of time varies at different times of the year ; its value may be obtained from the Nautical, or other Almanac. See also Observer's Handbook. Equator. " The line " of sailors. An imaginary line on the earth's surface separating the northern hemisphere from the southern hemisphere. The use of the word " hemisphere " suggests that the earth is regarded as a sphere, and on a spherical globe representing the earth the equator is the line formed by the intersection with the surface of a plane drawn at right angles to the polar axis and bisecting it. The position of the equatoiys identified by the vertical or plumb line being at right angles to the polar axis, and any complications introduced by the irregularity of the figure of the earth have to work from that datum. Latitude is measured from the equator northward through 90 to the North pole, southward through 90 to the South pole. By geodesic calculation it has been found that the Equator..-. 101 diameter of the globe from a point of the equator to its antipodes is 12,756,776 metres, whereas the polar axis is 12,713,818 metres.* Equatorial. Originally only an adjective derived from equator ; thus, the equatorial regions are the regions in the neighbourhood of " the line," but a meteorologist thinks of them as regions of rather low atmospheric pressure lying between the two belts of high pressure which are found in either hemisphere just north or south of the tropics. In this region the rotation of the earth has little or 110 influence in adjusting the wind to balance the distribution of pressure. The adjustment necessary for persistence can only be reached by the curvature of the air's path. It is perhaps for that reason that the regions about ten degrees north and south of the Equator are the regions in which tropical revolving storms originate. The adjective has also come to be used with regard to wind to mean a wind that is composed of air which has come from lower latitudes, whatever may be its direction at the time, as distinguished from a polar wind which is composed of air that has travelled from higher " latitudes. Typical equatorial winds are generally from South- West or between South and West, and Polar winds from North-East or between North and East. It is a question of considerable meteorological interest to con- sider in special cases whe-ther a South-East wind or a North-West wind is actually equatorial or polar. Equa- torial winds are generally warm, polar winds cold, but north-easterly winds are sometimes very warm and some- times very cold ; a north-wester almost always cold. * U.S. Coast and Geodetic Survey. 102 Glossary. Equilibrium.- Properly speaking, the balancing of two or more forces in such a way that the combined effect is the same as if there were no forces acting at all, so that the body upon which they act, if at rest, remains at rest and if in motion, it goes on moving without any alteration of its velocity. In the case of a hammock slung by a cord at each end, the forces acting along the cords and the weight balance each other and the hammock with its load is supported at rest. It is, therefore, not unusual to say that the load is in equilibrium, but it is an unfortunate use of the word because the state of rest is only a special case. In meteorology we are very frequently concerned with equilibrium of forces associated not with rest but with the uniform motion of the body upon which they act. For example, raindrops are all impelled downwardn by their weight, and their motion is resisted by the air through which they move, and after a very brief interval from the start the resistance of the air balances the weight and the drops fall with a uniform* speed (which depends upon their size) as if there were 110 longer any gravity or any air. The same is equally true of a falling bomb, but the time required to reach the limiting velocity is much greater. From the moment of its release it acquires velocity from its \\ eight, but if the height is sufficiently great it reaches a limiting velocity when the weight is balanced by the friction of the air and no further increase of velocity occurs. So, on the other hand, a pilot balloon rises with a uniform velocity as soon as the balloon is moving upwards fast enough for the buoyancy of the balloon, the weight * Mr. W. H. Dines points out that the speed is not strictly uniform but diminishes wifih the increasing density of the air in the lower layers. Equilibrium. 103 of the balloon and the friction of the air to get into equilibrium. So, also, in the case of a train running with uniform speed along a level ; the rails push the train forward, the resistance of the air holds it back ; there is equilibrium which we recognise by the fact that the speed is uniform. There are some other cases in which the word equili- brium is 'used that are more difficult. It is sometimes said, for example, that there is equilibrium between the wind velocity and the barometric gradient, but it is a peculiar kind of equilibrium : the wind is kept moving without change of speed in a great circle of the earth but not in a straight line through space. The balance of forces in this case is the same as that which obtains when we consider the motion of the moon round the earth, the force of gravitation on the moon is " balanced " by the moon's motion, a convenient form of expression but one which requires some explanation before its meaning is quite clear. Equinox. The time of the year when the astronom- ical day and night are equal, each lasting twelve hours, At the equinox the sun is u on the Equator " or is " cross- ing the line." It is on the horizon in the morning exactly in the east, and exactly in the west in the evening all over the world. Sunrise occurs at the same time all along a meridian. The sun is visible by refraction for a little longer than the duration of the astronomical day. There are two equinoxes. The spring or vernal equinox about the 21st March, and the autumnal equinox about the 22nd September. The currently accepted phrase " equinoctial gales " indicates that the equinoxes are re- garded in some quarters as the times of the year when gales 104 Glossary. are specially frequent ; but on our coasts the equinoxes mark the beginning and the end of the season of gales rather than its culmination. Winter is really the season for gales. Erg. See under HEAT. Error. In all the sciences dependent upon observation of the size of things the word error has a numerical sense which must not be confused with the ordinary trivial sense of a mistake or fault. Errors in the latter sense have to be avoided by skill and care, so they never occur, or hardly ever ; but when everything possible in that respect has been done there are always errors in the technical sense, due to imperfections of the instrument, or its adjustment, or to difficulty arising from changes in the element while it is being measured. In this sense error is difference between the reading of the assumed measure and the true measure of the element. The size of the residual error is a good indication of the degree of nicety to which the measurement can be carried. Pressure is the only meteorological element which is read to a very high degree of accuracy, such as one hundredth per cent. ; the temperature of the air can be read to the tenth of a degree, or within less than one per cent., reckoning on the absolute scale, as one must do for all calculations of density, but the temperature of the air is not " known " to that degree of accuracy, because it is subject to local and temporary fluctuations. An accuracy within one per cent, or even five per cent, is often acceptable. What is aimed at is to improve the instruments and the methods of reading, so that there is no systematic error, that is to say, no error that is known always to be present and to affect the measurement in the same wav, Error. 105 When that stage has been reached, and we have no good reason for thinking that the figure given by the reading is in any way biassed, we have what is called the residual error which has been made the subject of prolonged study, and has led to the application of the laws of mathematical probability. These have a real practical ^application in all cases where we deal with very large numbers of observations. In that case the frequency of occurrence of errors of given magnitude follows a well-known law called the law of error, from which we are able to compute what is called the "probable error" of an observation, a term which frequently occurs. It only means that in any particular case the actual error is no more likely to be greater than the " probable error " than it is to be less, so that the chances of the error being as great as the probable errors are one in two. The chances of an error being twice the probable error on either side of the true value^are 10 in 57, for three times 10 in 238, while for four times the probable error the chances are 1 in 147 and for five times 1 in 1,388. If the chances of an error on one side only are required the second of each pair of figures given above must be doubled. The study of laws of error is of great practical im- portance in all actuarial questions, and now forms part of the science of statistics. There are various forms of numerical error that call for notice. In dealing with accurate timing, for example, by means of a clock or chronometer, there is the index error, or clock error, and the rate error, the amount by which the clock is gaining or losing. Every instrument is liable to index error, on account of the index being inaccurately set, and every instrument is also liable to a scale error, on account of the scale being imperfectly graduated. Before trusting 106 Glossary implicitly to the readings of any instrument it is desirable that the user of it should become acquainted with its ways and habits in respect of error. This is particularly the case with instruments used by aviators, namely, altimeters, anemometers, aneroids, compasses, and so on. Evaporation. The process of conversion of water from the liquid to the gaseous form at the free surface of water in the presence of air, or from the solid to the gaseous form from the surface of ice. It expresses itself by the gradual disappearance of drops of dew after sun- rise, the drying of roofs and roads after rain, the loss of water from cisterns and reservoirs in drought. There is also copious evaporation from the stomata of the leaves of plants. The atmosphere is very rarely completely saturated, so that evaporation is always going on when water surfaces are freely exposed. The rate of evaporation depends upon a number of conditions. One of the chief is the " drying power" of the air and is represented by the difference between the amount of water-vapour which would saturate it and the amount which it holds at the time : that again depends on the relative humidity and the temperature. The drying power of air below the freezing point is very small but, in spite of that, the disappearance of snow by evaporation is surprisingly rapid. The other important condition is the nature of the surface of the water from which the evaporation takes place. There are all stages of cleanliness of the surface from the chemically pure water surface which is very seldom realised in practice, to the complete superficial film of oil which arrests evaporation altogether. Besides Evaporation. 107 I s I 8 ^ 1 1 rfjj ? i^r>.O-tt^v-icovj-ioocMTj- >f O t> "~> co i-i CM co M- f O CM co HH HH M 00 11 C$ d 00 Tj- >-icor^Mt-^coOt^.t^cMOO OO CM O ONMVOOO cOt^O OOO H "*" M xn rj-coM-tOO'-'"-'.cOTh'<d- co * * vn xn vO xn O co vO ON CM O t~ oo vO co w oo ON CM co vn co T$- xn vO ON rt-tt-cMMwuhHCMcOrt-xn OO CO OQ ~ 2 eg o^ ^ g $ co oo CM oo T!- xn vo co ON O t^ ' ^* t-ir^CMTJ-OOcOcMMcOOcO <1o C1 M N t-^^^^-vnvOONcooOfM xo ^^H 00 *-ivOxnoocMvnOMO"^f^ OO c^vnvnrt-"5t-OoooOt-<OOt^ co &|,r<r ^ 00 ^-vOxnCMOO CM Tj-CMOO CM xn OO CM ^ H CM t-*. i-t xn xn co CM CM O xn O CM WMcMfMCMMeMcMcMt-lM O xglaf? E OO CM C>1 CM O O CM t^ OO 1>- xn O H-icMcMfMfMCMcOcOcOcMcM w CO HloC p 5 M O "^ O w ON r^ rj- co vO ON Th (N ^J- l> ON O ON O oo l>. ON t^ OO 00 hH ) i h- 1 1 I M -2"- fl vn iO rj-oo^vnOO*nvowTJ-vo :f!'!'o g S n -iONTt--d-MOOOMMOvo vO i-i H- i vn O oo 1^.0 vn CM i-i ^ ^1 || p O xnOO cOrJ-cor^OO coj^i-ivO CM -t-t^>-ivnrMOcoOoo^-CM 9 ' CM Ol CO CO t^ CM i-i 05 rT fl 00 vooOTj-xnt^OONOOoo>-ivn g M i i co ^o t~>- t^. vn co *-* co o|||| PI" M H i M CM vO xn coONQcocovOcMi-i i-i CM CM CO CM i-t i-( eS S ,|.l| S ' i-gj ! , x ^.i -i 1 1 i H glil-o-g-^llgl SnS-'iS^Hi^iKOfcfi 108 Glossary. these conditions, temperature and wind have to be con- sidered, so that the measure of evaporation is the end of a very intricate story. Still, in dry countries like Egypt, South Africa and Australia it is a matter of serioun economic importance. , It is generally given on the analogy of rainfall as the depth of water evaporated. Some results are given on page 107. The differences shown in this table illustrate quite forcibly the influence which evaporation may exercise upon climate, but the figures must not be taken as strictly comparable as the actual amount of evaporation depends upon so many conditions. The measurements given for Egypt are taken from a Wild's gauge which holds only a small body of water and gives a high figure for evaporation. See Keeling : Evaporation in Egypt and the Sudan (1909). Craig : Cairo Scientific Journal, May (1912). Expansion. The increase in the size of a sample of material, which may be due to heat or to the release of mechanical strain, or the absorption of moisture or some other physical or chemical change. The size may be taken as the length or volume, some- times as the area. In the science of heat the fractional increase of length or volume for one degree of tem- perature is called the coefficient of thermal expansion. Thus, the co-efficient of " linear " expansion with heat of the brass used for barometer scales is 0*0000102 per degree Fahrenheit, which means that for 1F. the length of the scale increases by 10,2 ten-millionth parts of its length at the standard temperature (62 F.). The co-efficient of ''cubical" expansion of mercury is '0001010, which means that the volume of a quantity of mercury increases by 1*01 ten-thousandth of its bulk at the standard tem- perature (32 F.) for 1 F. The corresponding expansions Expansion. J09 and coefficients for 1 C. or la. are larger in the ratio of 18 to 10. Expansion of volume alters the density of a substance, and changes of density are therefore numerically related to expansion. The expansion of a gas may be caused either by reduc- tion of pressure or by increase of temperature. So in order to see the effect of temperature alone we muse keep the pressure constant. In these circumstances the coefficient of expansion is '00366 for la. referred to 273a. as standard, or *002 for 1 F. referred to 41 F. as standard. Exposure. In meteorology, the method of presen- tation of an instrument to that element which it is destined to measure or record, or the situation of the station with regard to the phenomenon or phenomena there to be observed. If meteorological observations are to be of much value attention must be paid to the manner of the exposure of the instruments. Details are to be found in the Observer's Handbook. Uniformity of exposure is of the greatest importance and for that reason the pattern of the thermometer-screen has been standardized in most countries, while in these Islands, a standard height above ground for the rain- gauge has likewise been fixed. A SUNSHINE EECORDER demands an entirely unobstructed horizon near sunrise and sunset at all times of the year. The question of the exposure of ANEMOMETERS is one of great difficulty. The extent of the GUSTINESS of the wind as exhibited on the trace of the tube-anemometer is a fair index of the excellence of the exposure. At Aberdeen Observatory, two anemometers are ex- posed, one at an elevation of 30 feet above the other. 110 Glossary. The results are noticeably different ; and the same state- ment applies in an even more marked degree to the two anemometers at Falmouth, one exposed upon the Obser- vatory roof and one upon the tower of Pendennis Castle. Extremes. Generally used with reference to temper- ature or wind ; in the first case to mean the highest and lowest temperatures recorded at an observing station in a day, a month or a year. The maximum and minimum temperatures are the extremes for the day ; in l'.U4 the extremes for Greenwich for January were 55 F. and 20 F., and for July 92 F. and 45 F. When the observations for a series of years are available we may find the normal or average extremes and ABSOLUTE EXTREMES. With regard to wind the highest wind recorded in a gust shown on a tube - anemogram is the extreme wind, and the highest wind-force on the Beaufort Scale noted by an observer in the course of a gale is logged as the extreme for that gale. The strongest gust for the British Isles is given on p. 142, the highest hourly wind velocity is 34*9 m/s (78 mi/hr) recorded at Fleetwood in 1894. ' Falirenlieit, Gabriel Daniel. The improver of the thermometer and barometer, born 1686 at Dantzig. He used mercury instead of spirit of wine for thermometers and avoided negative temperatures by marking the freezing point of water 32 ; the boiling point of water Avas subsequently marked 212. The Fahrenheit scale is still in common use in English- speaking countries, and it has advantages because the size of 'the degree is convenient and temperatures below F. are of rare occurrence at the Earth's surface, except in the polar (Arctic and Antarctic) regions and the con- tinental countries bordering thereupon. In fact, the range Fahrenheit. Ill F. to 100 F. is a very serviceable range for the climates of the temperate zone. But the investigation of the upper air has necessitated the frequent use of temperatures on the negative side of the Fahrenheit zero, temperatures as much as 100 below zero occur, and to have the zero in the middle of the working scale is very inconvenient. In the physical laboratory too, temperatures approaching 5()0F.- below the freezing point are realised in experiments on the condensation of hydrogen and helium. Probably the best scale for all purposes would be in Fahrenheit degrees measured from 459 below the Fahrenheit zero which is computed to be within half a degree of the zero of absolute temperature. In that case 500 would correspond with 41 F. or 5 C. But the grow- ing prevalence of the Centigrade or Celsius scale in countries which do not speak English has led to the use of temperatures measured in the centesimal degrees from the zero of absolute temperature computed as 273 below the freezing point of water. For a table of conversion of the various scales in use see page 355. Fall. "The fall of the leaf" in common use with American writers for Autumn. Fluid. A substance which flows, to be distinguished from a solid which will not flow. Some fluids are very viscous, like pitch or treacle, and take a long time to flow, others are mobile, like water or petrol, and take very little time to flow until the surface becomes level. Gases are included in the general term fluid, because they also will flow through a pipe. Their peculiarity is that they can be not only compressed by pressure but also expanded indefi- nitely on the release of pressure. The density, i.e., the amount 112 Glossary . that can be got into a limited space, is, in fact, almost exactly proportional to the pressure. So when we find a, large mass of gaseous fluid like the atmosphere lying upon the earth's surface, it is dense in the surface layer which has to carry the weight of all there is above it ; and as the pressure gets less and less, upward, the density gets less and less until space is reached. We do not know what happens where the atmosphere merges into space, but we are sure that the earth, with the aid of gravity, carries its atmosphere along without losing any appreciable amount into the void. Fog*. Obscurity of the atmosphere which impedes navigation or locomotion. It may be due to a cloud of water particles at the surface, as sea-fogs and valley-fogs generally are, but an effective fog can be produced by clouds of dust ; that is often the case off the West coast of Africa during the season of the HARMATTAN. In towns true water-fogs are generally rendered more opaque by loading with smoke, and in some cases in towns obscurity of the atmosphere that hardly amounts to fog may be due to the condensation produced by the gaseous products of combustion under the action of sunlight. Sea-fog is apparently due most frequently to the passage of air over sea water colder than itself ; there is first the cooling of the air by the contact with the cold water, and then the mixing up of the air near the surface by the eddy motion resulting in the cooling of a considerable thickness below the dewpoint. (See G. I. Taylor, Scientific Results of the Voyage of the " Scotia" 1913.) Sea-fog is most prevalent in spring and summer when the air is warming rapidly. It does not often occur in the winter. See p. 121. Fog. 113 Land-fog, on the other hand, is an autumn or winter fog ; it is generally due to cold air passing over relatively warm, moist ground. The process of cooling may be either by radiation or by a change of wind, but again eddy motion is necessary to mix the warm, moist air close to the ground with the cold flood. Fogs of this kind are not infrequent in the early mornings of summer, but they persist sometimes through the day in autumn and winter. Autumn is their special season. Anticyclonic weather with light airs is very favourable for land-fog, and the ending of a period of anticyclonic weather is nearly always fog. Fog on our coasts is generally included in the forecasts for the British Isles when the wind changes from a Northerly or Easterly to a Southerly point. For the monthly percentage-frequency of fog and mist in the English Channel see p. 121. The conditions for fog in London are set out in a Report 3f the Meteorological Office on fogs. (M.O. publication No. 160.) The frequency of fog at the observing hours, according fco the returns for the past 20 years from British Stations for the Daily Weather Report, is shewn in the following table. It should be noted that up to the end of June, 1908, the morning and mid-day hours of observation were <Sh. and 14h. instead of 7h. and 13h., and that at Oxford the observing hours have been, and are still, 8h. and 20h., not 7h. and 18h. The following are the yearly frequencies of observations of fog at Ih. (or 3h.) in the past two years: Wick ... 10 Donaghadee 18 Spurn Head 10 Stornoway Holyhead 13 Yarmouth 8 Malin Head 10 Pembroke 18 Eskdalemuir 3 Blacksod Point 9 Portland Bill 5 Benson ... 2 Valencia... 3 Dimgeness 15 London ... 3 Scilly ... 20 Tynemouth i 114 Glossary. AVERAGE NUMBEK OF OBSERVATIONS OF FOG IN A iNur ofY of O vatic 7h. aber ears bser- ns at 13h. Station. 7h. 13h. Sum- mer. Win- ter. Year. Sum- mer. Win- ter. Year. 20 20 8 20 20 20 2O 20 20 2O 12 20 20 20 20 2O 20 2O 2O 20 20 20 1 20 17 2Q 12 2O 20 20 8 5 20 20 20 20 20 7 20 20 20 20 13 2O 17 20 20 20 20 20 13 4 43 3 i & Sum burgh Head... Stornoway Castle bay , Wick Nairn Aberdeen Leith IO 2 9 II 3 4 2 2 I 2 2 I 7 12 3 ii 12 5 5 9 o 3 8 2 I I I 2 4 IO 2 5 o 4 9 4 i In "S H 6 N. Shields Spurn Head Great Yarmouth... Clacton-on-Sea ... 8 7 5 i 8 12 25 6 16 19 30 7 2 3 I 6 8 ii 2 s 03 0) Malin Head f Belmullet \ \ Blacksod Point J Valencia ... Roche's Point Donaghadee Liverpool (Bidston Observatory). Holyhead Pembroke 8 3 2 8 7 i 16 ii 3 2 2 5 3 6 9 7 ii 5 4 13 10 7 25 18 4 i o 3 2 9 6 I I 1 { 5 2 5 4 13 9 6 3 09 Scilly (6t. Mary's) Jersey (St. Aubin's) j Hurst Castle \ | Portland Bill J Dungeness Dover '! 7 3 6 5 5 10 7 20 7 10 , I 10 4 17 2 10 s 2 3 ii 3 6 7 6 2 1 3 London ... ...13 19 Oxford ... ... 4 24 / Loughborough \ \ Nottingham J Bath i 5 Birr Castle ... 4 i 6 22 28 25 6 10 o o 7 6 2 NOTE. Summer (April to September) contains 183 days, and Fog. YEAR AT VARIOUS STATIONS IN THE BRITISH ISLES. Sum- mer. 18h. Sum- mer. 21h. Win- ter. Station. Number of Years of Obser- ' vations at ?:(* Year. 18h. 2-lh. 6 I 7 7 ^ o i o . i 2 7 9 i 10 1000 9 30 3 8628 Sumburgh Head... Stornoway Castlebay Wick Nairn Aberdeen Leith s I 20 20 8 20 20 20 20 20 20 ? 20 20 20 20 20 20 20 20 20 20 20 20 8 ' 20 17 20 12 20 7 7 7 5 ~ 2 7 7 7 7 7 7 7 *8 7 7 5 7 J i 4 7 9 2 7 10 10 2 4 I 6 6 8 i 10 10 ? N. Shields Spurn Head Great Yarmouth... Clacton-on-Sea ... ll pq jo 6 2 I 4 2 O 9 6 3 2 o 3 2 I 7 5 9 4 i 7 4 i 16 ii i 4 3 8 7 ; o 4 5 6 7 12 12 Malin Head / Belmullet \ ( Blacksod Point / Valencia Roche's Point ... Donaghadee Liverpool (Bidston Observatory). Holy head Pembroke i 1 9 6 ; 4 4 ^ 3 2 15 7 6 5 2 10 2 j 5 i '2 15 3 5 Scilly (St. Mary's) Jersey (St. Aubin's f Hurst Castle ) \ Portland Bill J Dungeness Dover ,p 1 o cc O O o 6 4 6 2 I 6 4 6 2 I I O 8 5 9 5 London ... Oxford / Loughborough \ \ Nottingham / Bath Birr Castle Inland. Winter (October to March) 1*2 days, in leap year 183 davs. 116 Glossary. Fog Bow. A white rainbow of about 40 radius seen opposite the sun in fog. Its outer margin has a reddish, and its inner a bluish tinge, but the middle of the band is quite white. The bow is produced in the same way as the ordinary rainbow but owing to the smallness of the drops, under O025 mm., the colours are mixed and the bow is nearly white. Fohn. The name given to certain dry, warm, relaxing winds of the valleys on the Northern side of the Alps. The general direction of the Fohn winds is from the South. The peculiar character of the air is accounted for by supposing that it comes from over the plains on the Southern side of the ridge. In its elevation it becomes dynamically cooled, and if condensation occurs and rain or snow is formed in it, the fall of temperature is so much restricted on account of the latent heat of the vapour which is condensed and left behind, that the air which forces its way down into the valleys on the north side, being dynamically warmed and dried, appears as a warm, dry wind. Some of the details of the process are still obscure because warm air does not naturally flow downhill, but th'e main outline of the process is certainly established and the subject has been studied in detail by Austrian meteorologists. The Chinook wind of the Western prairies of America which comes down from the Rocky Mountains as a warm, dry wind evaporating a good deal of the prairie snow in winter is of similar character, and various other examples of what is known in meteorology as the Fohn effect occur from time to time on many hill sides. In regions like Norway, Greenland and the Antarctic Continent it com- plicates the temperature measurements very seriously. Forecast. 117 Forecast. The name given by Admiral R. FitzRoy to a statement of the weather to be anticipated in the near future from a study of a synoptic chart or " weather map." In the Meteorological Office the period of antici- pation of a forecast does not exceed twenty-four hours, but when conditions shown on the map are favourable a more general statement of the probable weather for two or three days is given in a form which is called " the further outlook." In practice a forecast includes (1) A statement of the direction and force of the surface-wind and the changes therein which are expected within the period of the forecast. (2) A statement of the state of the sky (as regards clouds), precipitation (rain, hail, snow or sleet) and temperature, whether it will be high or low for the time of year, or higher or lower than at the time of making the forecast. (3) A note as to the probability of such occurrences as night-frost, fog, or thunder. For these statements the forecaster depends upon the changes in the distribution of pressure which are indicated on the map, although these changes are not described in the forecast. They are. however, set out in a preliminary statement called the " general inference," and for the information of airmen the anticipated changes in the pressure gradient over the several districts are formulated and are expressed as an addition to the forecast giving the " wind at 1,500 feet." That height is chosen because the wind at that level is generally in close agreement with that computed from the distribution of pressure at the surface, 118 Glossary. The direction and velocity quoted for 1,500 feet are sufficiently applicable for heights up to 3,000 or 4,000 feet, as changes in the wind above the level of 1,500 feet are generally gradual. A Southerly, South-westerly, Westerly or North-westerly wind generally gets gradually stronger at higher levels, but an Easterly wind often falls lighter and is replaced in the highest levels by a wind from a Westerly quarter, though that is not always the case. The motion of clouds, or the measurement of air-currents by observations of a pilot balloon, are the only means available at present for guidance as to the changes in the higher level. Freezing 1 . With reference to weather this word is used when the temperature of the air is below the freez- ing point of water 32 F., 273a., C. American writers use the term " a freeze " where we are accustomed to use " frost " to indicate freezing conditions persistent for a sufficient time to characterise the weather. Frequency : The number of times that a particular phenomenon of weather has happened in the course of a given period of time, generally a number of years. Here, for example, is a summary of the spells of wind from the Easterly quarter, according to the direction of the isobars, over S.E. England and Northern France in nine years. Taking January, for instance, the nine years supply a total of 279 days, of which 58 were days of East wind. These consisted of one sequence of eight consecutive days of East wind, one sequence of six days, three sequences of four, two sequences of three and seven sequences of two days, with finally twelve isolated days of East wind. frequency. 119 J i ti :s~ ^5 o 1 "7 I ^ > r* c^ ^ E.-S !^SE ^s J " : r^J 'rtf. >< ^ ^ r O W M 00 COO M O CO IO o s -is. ro M O IT) O CO M rf M K H 6 co ro Th co O l>x CO LO M O ^ |>> M CO M H s O M C1 CO g & R 01 pj ^2 -P o O O C 120 Glossary. In this case the number of years over which the obser- vations extend is given, quite an arbitrary number, and thus the numbers for frequency of occurrence have to be considered with reference to the number of years selected. It is, however, usual to reduce frequency figures to a yearly average. Here, for example, is the average frequency of GEOSTROPHIC winds from different quarters over the South- East of England and Northern France obtained from observations for the nine years, 1904-1912. Frequency of winds (geostrophic) from different quarters. Average Number of Days in the several months of the year in which the Wind is from a specified quarter. South-East of England and Northern France. N.E Fi SE. s s.w. w N.W. N Calms. Total. January ... 3 2 2 2 9 6 2 I 4 31 February .. 3 I I 4 7 6 3 2 I 28 March 4 2 I 3 7 5 2 2 5 31 April 5 2 I 2 6 5 2 4 3 30 May 4 2 2 2 5 4 2 2 6 29 June 5 I I 3 6 4 3 4 3 30 July 4 I I i 6 5 3 4 6 3 1 August ... 2 I I 2 10 6 3 2 4 3 1 September 4 3 2 2 5 4 3 4 o J 3 October ... 3 3 2 5 7 3 3 2 4 ^2 November 3 2 I 3 7 7 2 1 4 30 December.. i 2 2 7 10 4 2 I 2 31 The treatment of fractional parts of a day accounts for the two discordant totals. 121 In view of the awkwardness of having to bear in mind the possible number of occurrences, while considering the actual or average number, it is convenient to use the percentage frequency instead of the actual frequency. This plan is often adopted for giving the results of observations at sea, which are made six times a day, or every four.hours. For example," the percentage frequency of fog in the English Channel is given by the figures in the following, table : Percentage Frequency of Fog and Mist in the English Channel. [Based on four-hourly observations from ships during the 15 years 1891-1905.] Percentage of Percentage of Month* Number of observations. whole number of observations. whole number of observations. Fog. Mist. January 1,187 2'5 15-8 February . . . 1,185 2'8 22*9 March 1,241 3'8 2 4*5 April M 2 4 4-8 24-4 May 1,501 4*4 26*9 June 1,363 5'6 30-2 July 1,260 3*8 26*3 August 1,206 3' 2 I7-3 September ... 1,264 r,3 I7'0 October i,454 i-6 16-5 November ... ' 1,284 I *I 15-0 December ... 1,294 i*5 18-2 122 Qlossanj. Friction. A word used somewhat vaguely in meteor- ological writings in dealing with the effect of the surface of the sea or of the land, with its obstacles in the form of irregularity of surface, hills, buildings, or trees upon the flow of air in the lower layers of the atmosphere. The effect of the irregularities of surface is to produce turbu- lent motion in the lowest layer which gradually spreads upwards, if the wind goes on blowing, and consists of irregular eddies approaching to regularity in the case of a cliff eddy which can be noticed when a strong wind blows directly on to a cliff and produces an eddy with a horizontal axis. An account of the eddy caused by the Eastern face of the rock of Gibraltar is gi^en in the Journal of the Aeronautical Society, Vol. 18, 1914, p. 184. The general effect of this so-called friction is to reduce the flow of air past an anemometer so that the recorded wind velocity is below that which would be experienced if the anemometer were high enough to be out of the reach of the surface effect. Numerical values for this effect are of great practical importance, because they are concerned with the change of velocity in the immediate neighbourhood of the ground. But it is not easy to obtain them, because every exposure near land or sea is more or less affected, and, therefore, no proper standard of reference can be obtained by direct observation. Recourse is, therefore, had to the computation of the wind from the distribution of pressure, the so-called " geostrophic " or GRADIENT WIND. From the comparison of a long series of geostrophic and observed winds we conclude that over the open sea, or on an exposed spit of flat sand like Spurn Head, the wind loses one-third of its velocity from "friction," and at other well-exposed stations the loss is, on the average, as Friction. 123 much as 60 per cent., but for any particular anemometer it is different for winds from different quarters because the exposure seaward or landward is different. Infor- mation on this point for a number of Meteorological Office stations is given in a memoir by Mr. J. Fairgrieve (Geophysical Memoirs, Vol. 1, p. 189), and information for other stations is in process of compilation at the Meteorological Office. The consequence of this 4 effect can sometimes be seen in weather maps. On one occasion when the whole of the British Isles was covered with parallel isobars running nearly West and East, all the stations on the Western side gave the wind as force 8 (42 mi/hr) while those on the Eastern side gave force 5 (21 mi/hr), so that the velocity was reduced by one-half in consequence of the "friction" of the land. If the velocity at the exposed Western stations be taken at two-thirds the velocity of the wind free from friction, we get the following interest- ing result which is probably correct enough for practical use : One-third of the velocity is lost by the sea friction on the Western side, and one-third more by. the land friction of the country between West and East. Frost. According to British meteorological practice frost occurs when the temperature of the air is below the freezing point of water (see FREEZING) ; it may be either local, as a ground-frost, a spring-frost or a night- frost often is, or general, such as a frost which gives bearing ice in the course of three or four days. But the word would hardly be used unless there were water or plants or something else to be frozen, so that its use is generally restricted to the lowest levels of the atmosphere. We should hardly speak of a frost in relation to the cold of the upper air, or even of a mountain top. 124 Glossary. Meteorologically, the difference between the conditions for a general frost and those for a local frost is so great that different words are needed. The American meteor- ologists have some reason, therefore, in speaking of a general meteorological frost as " a freeze." The British Isles are accessible for a freeze or general frost in two ways, first by Northerly winds bringing cold air from the Arctic regions over the North Atlantic, round a high pressure lying over the Greenland-Iceland region, secondly by Easterly or South-Easterly winds coming round a high pressure over Scandinavia and Northern Europe, which in the winter is persistently cold. The Northerly wind has to cross a considerable stretch of the north-eastern extension of the Gulf -stream water, so that it has to travel quickly to avoid being warmed. Consequently, Northerly " freezes " are generally short and sharp. The more prolonged frosts are generally caused by the Easterly winds which have only a short stretch of sea to cross. A long freeze may begin with a Northerly wind, and snow, followed by a persistent Easterly wind. The short frosts, or night frosts, may occur with very light winds from any quarter except between South and West ; they are characteristic of clear nights, with great loss of heat from the ground by radiation to the clear sky. The conditions are set out in detail in a pamphlet prepared in the Meteorological Office and reprinted in " Forecasting Weather" Chapter XII. Low temperatures are often quoted as degrees of frost, meaning thereby the nu&iber of degrees below the freezing point of water. Gale. Wind with an hourly velocity of more than 17m/s, or 39mi/hr. The figure is selected as the lower Gale. 125 limit of force 8 on the Beaufort Scale. A wind estimated as force 8 or more is counted a gale. The relation of the estimation to the measured hourly velocity is subject to some uncertainty on account of the incessant fluctuations of velocity in a strong wind which are known as GUSTINESS. They are not shown in the records of the cup anemometer which were used for com- puting the equivalents. The number of gales recorded for any locality depends largely on the exposure of the anemometer, as the table on the next page shows. Judging by this table, anyone who is unacquainted with the practical difficulties of anemometry would be tempted to draw the conclusion that the localities represented by Kew, Falmouth, Aberdeen, Valencia and Yarmouth are immune from gales, or nearly so. For any purpose of aerial navigation such a conclusion would be egregiously untrue. It is true of the anemo- meters, but not of the free air above them. The records, which in these particular cases go back to 1868, are good enough when we are concerned only with comparing the wind of to-day with that of yesterday, or any other day in the last forty-seven years, and of determining normals for reference, the diurnal and seasonal variation, and so on ; but when we want to compare one locality with another we must face the problem of making allowance for the exposure of the anemometer. To meet this requirement we propose to give the basic characteristics of the localities under our observation as regards wind in terms of GRADIENT WIND, or more strictly geostrophic wind. It is a very voluminous inquiry, but is now nearly completed, and some of the results will be included in this volume. 126 Glossary. ,3 '3-^ Q i ^o M i' O O CO H O i-OOO M M M 0} s M M o To o ^ 0> *9UUJ[ o M H O O O O O CO Is J^" 1 4H te i "jCraj^T o M -t- M CO O 01 <*J .3 -** ^ r-H 0> CJ C8 t> }<5{ jiidY . o cn -f M LO O 00 O 4s ZQ a 0-3 ^* -+- o CD .43 P -d fl * rt ^ - !>, r^ ^t-i>. o O^ H CO - *> ^> ^ -^ |1s A'j^n.iqy^ in -i- -h O O^ O O^ W M o 1 j f'S cr 1 -^ ^"8 -^uu,, 01 -t- n oc LO o w 01 01 O 00 N Is *<S> s si |>-: J9qraao9([ IJ^ ON M co o vO M M M rf o o 01 M So * eq 'V Pa *$l P Wqra8A o N ^ 90 H !>. 00 O w "S ^ |ig a o ^ _ci a $ ^qov.0 ,0 r^ co o o M 01 O 'Tf 3 1 'S S Ig^s "I9qui9^d9g M X M -f- O 01 O o 1 g c S) *s s '^snSny 01 M CO O 000 5 $ Ch B-SI .s ^!^nj[' o r^ o M o o o o ^ ^ % Number of recorded b d o 1 VI Valencia G P c -M .s" Fleetwood Holyhead Scilly Falmouth (Observator Deerness Aberdeen Yarmouth CD M Gale. 127 The seasonal variation may be expressed as follows : O fl OD 0nMoooc*0Mp>oovoo>- 1 42 real C^ rH -o^o^^^o^cr^ h rH .5* 'fe *? ! j9qni9AO^[ xi "^~ QO - 'jaqo^OQ oo *-t o OVOOOOOOOOOQ a\ o r* O " 43 a? f 3 o ^ g QD c j8qniaid8Qj t^O rl-^Oxort-C^cow - C^w W QQ QQ S N N N W $<* .9 2 ^ o ^nSny SSSSffSffaSffsaaS 1^ C8 03 I*V> ^ r^J 43 pq J A I n f HI HI HI f rS s '8unp ^^^^^OSN ^ N ^cso^*- SH *"* HI 1- 43 ^f ^ S O '^Bj^r s-sass's R'g.sa- :? *3 .2 .3 " ludy re? a 2 a-:? gulag's- pd HH <> CC qow M ^8*3* ""a:rs'' rf 2 -s 2 > QQ ^Cisnjqa^ ^g^^^-.OK.-^^JJ-OO i 2 ^ S 1 Xnuimr ^^^^^^^^^s:- 2 |f rg 2 2 -2 r* o d 1 frS CO r^ fl 2 r ^ ?H 0) 1 *0 1 COASTS. ^ > ^ ^' ^ ^ ' 1 s = s | ^j*li & JS 2sl (1 The " favourit 21 to 19 against. 128 For special localities tables of local statistics must be consulted. Some guidance may be obtained from the diagrams given under the heading wind. Gale- Warning*. Notice of threatening atmospherical disturbances on or near the coasts of the British Islands are issued by telegraph from the Meteorological Office to a number of ports and fishery -stations. The issue of a warning indicates that an atmospheric disturbance is in existence which will probably cause a GALE (Force 8 by BEAUFORT SCALE) within a distance of (say) 50 miles of the place to which the warning is sent. The place itself may be comparatively sheltered, and the wind may not attain the force of a gale there. The meaning of the warning is simply " Look out. Bad weather of such and 'mch a character is probably approaching you." The fact that such a notice has been received is made known* by hoisting in a conspicuous position a black canvas cone (gale-cone) 3 feet high and 3 feet wide at the base, which has the appearance of a triangle when hoisted. The " South cone " (point downwards) is hoisted in anticipation of gales and strong winds From S.E. veering to S.W., W., or N.W. S.W. W. or N.W. W N W vv . ,, J.-N . vv . The " North cone " (point upwards) is hoisted in antici- pation of gales and strong winds From S.E., E. or N.E., backing to N. N.W. veering to N., N.E., or E. N. N.E. or E. . N.E. E. * The display of cones and issue of notices to the general public has been suspended during the war. Gas. 129 The warning is intended to continue from the time the telegram leaves the Meteorological Office until 8 p.m. the following day. The gale-warning service of the British Isles was established under the direction of the late Admiral FitzRoy in 1861, and has been maintained in operation ever since, with a slight interruption in 1867. Gas. The name used for any kind of fluid which has unlimited capacity for expansion under diminishing pressure. It is to be distinguished from a liquid which has only a limited capacity for expansion under reduced pressure. A liquid may occupy only the lower part of a vessel like a bottle ; it will flow to the bottom of the vessel and leave a "free" surface. But a gas cannot be located in that way ; its volume is determined not by the amount of material but by the size of the vessel which contains it and by the pressure upon its boundaries. There are many different kinds of gas, such as nitrogen, hydrogen, carbonic acid, coal-gas, marsh-gas, and so on ; but the word is often used when coal-gas is meant, and recently it has been used for heavy poisonous gas of unspecified composition. In scientific practice gas means any substance which obeys approximately the gaseous laws ; these laws are two, viz. : 1. When the temperature is kept constant the pressure of a given mass of gas is inversely proportional to the volume which it occupies, or the density is directly proportional to the pressure. 2. When the volume is kept constant the pressure is proportional to the absolute temperature, or when the pressure is kept constant the. volume is proportional to the absolute temperature. GeostropMc.- See GRADIENT WIND. 13204 E 130 Glossary. Glazed Frost. When rain falls with the air-tempera- ture below the freezing point a layer of smooth ice, which may attain considerable thickness, is formed upon all objects exposed to it. This is known as glazed frost. The accumulation of ice is frequently sufficient to bring down telegraph wires. In these islands the phenomenon is one of comparative rarity. It must be distinguished from SILVER THAW, which occurs when a warm, damp wind supervenes upon severe cold, the moisture condensing on still-freezing surfaces and thus producing a coat of ice, similar in appearance to glazed frost. Super-cooled water-drops are said to be the cause of glazed frost. Glory. The system of coloured rings surrounding the shadow of the observer's head on a bank of cloud or fog or even of dew. It is a diffraction-effect due to the bending of rays of light round small obstacles, water- drops in this case. As in all diffraction effects the violet ring is nearest the centre, followed outwards by blue, green, orange, and red on the outside; the blue and violet are seldom seen. A Glory may be seen surrounding the shadow of an aeroplane on a cloud. Gradient. A convenient word rather overworked in modern meteorology. We use it in pressure gradient, temperature gradient, potential gradient, to denote different ideas. In pressure gradient for any locality we imagine the distribution of sea-level pressure to be mapped out by isobars ; take a line through the locality at right angles to the isobars nearest to it on either side and measure the step of barometric pressure which corresponds with a measured distance along the line from high pressure to low. This use of gradient was Gradient. 131 introduced by Thomas Stevenson, C.E., of the Board of Northern Lights. It corresponds with an engineer's use of the word gradient in specifying a slope from a map of contours, but to get the pressure-gradient we have first to determine the line along which the slope is steepest, so that pressure-gradient has a definite direction. There is a convention that the distance to be taken is 15 nautical miles and the step of pressure is to be given in hundredths of an inch. The gradient will work out at practically the same figure if the distance is a geographical degree and the step of pressure is given in millimetres. To get the same figure for the gradient with the step of pressure in millibars the distance would have to be taken as 45 nautical miles. But numerical values of the gradient are very seldom quoted. Temperature gradient may be based on the same idea and give the rate of change of temperature, along the horizontal through a locality, at right angles to the isotherms, as obtained from a chart of isotherms properly corrected for height. But it is much more frequently used to indicate the step of temperature for a kilometre step of vertical height. Used in this sense temperature gradient may be positive or negative, and by international agreement the temperature gradient is positive when the step is towards lower temperature for increasing height, because temperature generally decreases aloft ; but it does not always do so. The change from positive to negative temperature gradient is called an INVERSION of temperature gradient or simply an u inversion ", and so an " inversion " comes to mean a region where tempera- ture increases with height. Potential gradient is used for the change of atmospheric 13204 E 2 132 Glossary. electrical potential in the vertical, and for that alone. It is generally given in volts per metre. That also may be positive or negative and is taken to be positive when the potential increases with height. The word LAPSE (q.v.) has been adopted as a better name than gradient for these rates of change in the vertical. The pressure gradient is the one which comes most frequently into practical consideration, as it is closely related to the direction and force of the wind, so that the idea of pressure gradient should always be present in the mind of a student of weather maps, though the gradient may seldom be evaluated in figures. On looking over a map the localities where the gradient is steep will always be noticeable by the closeness of the isobars. The deter- mination of the pressure gradient is comparatively easy when the isobars in the locality are free from local irregularity and nearly parallel. There is then no difficulty in identifying the direction of the gradient, because the line drawn at right angles to successive isobars is approximately straight for a sufficient distance on the map. Experience is required to make a workable estimate of the gradient when the isobars are irregular. In practice tho gradient is not taken by setting out a length of 15 or 60 or 45 nautical miles, but by scaling the distance apart of consecutive isobars. It is most con- venient to express this distance in nautical miles, because 60 nautical miles make up a degree of latitude, and every map made for meteorological purposes is scaled according to latitude. If, for example, isobars are drawn for steps of JSQ' millibars, and the shortest line drawn to bridge two Gradient. 133 isobars across a station scales out at 75 nautical miles ; the gradient is 1 millibar (0*03 m.) for 15 nautical miles, or 3 on the conventional scale of pressure gradients. For calculation in C.G.S. units it is convenient to have the gradient expressed in terms of millibars for 100 kilo- metres. It is best to use a large scale map for obtaining pressure gradients so that intermediate isobars can be inserted by estimation when those drawn for the ordinary steps are not regular, but with the best maps the estimation of the gradient is sometimes uncertain on account of local irregularities of pressure which may be indicated on la barogram but cannot be allowed for in a map based on telegraphic reports from stations 100 miles (160 kilometres) apart. Some of the steepest authentic gradients that have been noted on British weather maps are : Date and Place. Gradient. Inter- Millibars national meaeure. per 100 kilometre. 1912, August 26, East Anglia (Norwich floods). 1907, February 20, between Ice- land and Faroe. II'O I0'0 13*2 12 'O 1912, November 26 land. , West of Scot- 9*7 II'7 134 Glossary. Gradient Wind. The flow of air which is necessary to balance the pressure-gradient. The direction of the gradient wind is along the isobars, and the velocity is so adjusted that there is equilibrium between the force pressing the air inwards, towards the low pressure, and the centrifugal action to which the moving air is subject in consequence of its motion. In the case of the atmosphere the centrifugal action may be due to two separate causes ; the first is the tendency of moving air to deviate from a GREAT CIRCLE in consequence of the rotation of the earth ; the deviation is towards the right of the air as it moves in the Northern hemisphere, and towards the left in the Southern. The second is the centrifugal force of rotation in a circle round a central point according to the well known formula for any spinning body. In this case we regard the air as spinning round an axis through the centre of its path. This part of the centrifugal action is due to the curvature of the path on the earth's surface. Both components of the centrifugal action are in the line of the pressure gradient : the part due to the rotation of the earth is always tending to the right in the Northern hemisphere, the part due to the curvature of the path goes against the gradient from low to high when the curvature is cyclonic, and with the gradient when it is ancicyclonic, so that in the one case we have the gradient balancing the sum of the components due to the earth's rotation and the spin, and in the other case the gradient and the spin-component balance the action due to the earth's rotation. The formal reasoning which leads up to this result is gi /en at the end of this article. The method used therein Gradient Wind. 135 for calculating the effect of the rotation of the earth was suggested to the writer in 1904 by Sir John Eliot, F.R.S., Director of the Indian Meteorological Service. For the sake of brevity in reference to these two components it is very convenient to have separate names for them. . Let us call the one due to the rotation of the earth the geostrophic component,* and the one due to the curvature of the path the cyclostrophic component. Consider the relative magnitude of these components under different conditions. It will be noticed that the geostrophic component depends upon latitude, the cyclo- strophic component does not, so, other things being equal, their relative importance will depend upon the latitude ; so we will take three cases, one near the equator at latitude 10 within the equatorial belt of low pressure, one near the pole latitude 80 of undetermined mete- orological character, and one, half - way, between, in latitude 45, a region of highs and lows travelling Eastward. Using V to denote the wind-velocity, when the radius of the path is 120 nautical miles the cyclostrophic com- ponent is equal to the geostrophic in latitude 10 when V is 5*6 metres per second ; in latitude 45 when Y is 22*9 metres per second ; in latitude 80 when V is 31-9 metres per second. It will be seen that in the equatorial region the cyclo- strophic component is dominant as soon as the wind reaches a very moderate velocity. * A table to find the geostrophic component is given on pp. 172, 173. 136 Glossary. EQUATION FOR GTEOSTEOPHIC WIND. 2 he Relation between the Earttts Rotation and the Pressure Distribution for Great- Circle-Motion of Air. The rotation co of the earth about the polar axis can be resolved into co sin about the vertical at the place where latitude is and co cos <j> about a line through the earth's centre parallel to the tan- gent line. The latter produces no effect in deviating an air current any more than the polar rotation does on a current at the equator. The former corresponds with the rotation of the earth's surface counterclockwise in the Northern Hemisphere and clockwise in the Southern Hemisphere under the moving air with an angular velocity co sin 0. We therefore regard the surface over which the wind is moving as a flat disc rotating with an angular velocity co sin 0. By the end of an interval t the air will have travelled Vt y where V is the " wind-velocity," and the earth underneath its new position will be at a distance Vt X co t sin 0, measured along a small circle, from its position at the beginning of the time t. Taking it to be at right angles to the path, in the limit when t is small, the distance the air will appear to have become displaced to the right over the earth is Fco -sin 0. This displacement on the "igt 2 " law (since initially there was no transverse velocity) is what would be produced by a transverse acceleration 2 co Fsin 0. . . the effect of the earth's rotation is equivalent to an acceleration 2 co Fsin 9, at right angles to the path directed to the right in the Northern Hemisphere, and to the left in the Southern Hemisphere. In order to keep the air on the great circle, a force corresponding with an equal but oppositely directed acceleration is necessary. This force is supplied by the pressure distribution. Gradient Wind. 137 EQUATION FOB CYCLOSTROPHIC WIND. Force necessary to balance the acceleration of air moving uniformly in a small circle, assuming the earth is not rotating. Let A be the pole of circle PRQ. Join PQ, cutting the radius OA in N. Acceleration of particle moving uniformly along the small circle with V' 2 V 2 velocity Fis -^ along PN = =-. - where R = radius of earth ; Jri\ jj, sin p and o is the angular radius of the small circle representing the path. The horizontal component of this acceleration, that is, the component F 2 cosp F 2 along the tangent at P, is p . = -- cot p. GENERAL EQUATION CONNECTING PRESSURE GRADIENT, EARTH'S ROTATION, CURVATURE OF PATH OF AIR AND WIND VELOCITY. I. Cyclonic motion. The force required to keep the air moving on a great circle in spite of the rotation of the earth must be such as to give an acceleration 2w V sin directed over the path to the left in the Northern Hemisphere. It must also compensate an acceleration due to the curvature of the path, F- cot p JR, by a force directed towards the low pressure side of the isobar. For steady motion these two combined are equivalent to the acceleration due to the gradient of pressure, i.e., *- where I) is the 138 Glossary. density of the air, and y the pressure gradient, directed towards the low pressure side. y F 2 .*. = 2 (j Fsin + cot p. 1 II. Anticyclomc motion. In this case 2 w F sin and ^ are directed outwards from the region of high pressure, and the equation becomes y F 2 -/ = 2 u> Fsin cot p. D R Gramme. The unit of mass on the C.G.S. system. It is one- thousandth part of the standard kilogramme, Gramme. 139 which was originally constructed to represent the weight of a litre (cubic decimetre) of water. A gramme is equivalent to 15*4 grains, or rather more than one-thirtieth of an ounce. A pound is equivalent to 454 grammes. For GRAMME- CALORIE see p. 301. Grass-temperature. For estimating the effect of RADIATION from the Earth's surface at night a minimum thermometer is exposed just above the surface of short grass, so that the bulb does not actually touch the grass. Abroad the thermometer is sometimes laid on the grass itself. Gravity. See p. 308. Great Circle. A line on the earth's surface which lies in a plane through the centre of* the earth's figure. All meridian lines are great circles so is also the equator, but all lines of latitude, with the exception of the equator, are small circles since their planes do not pass through the earth's centre. The visible horizon is a small circle. The great circle which passes through two points on the earth's surface is made up of the shortest and the longest track between the two points. The shortest track is less than a semicircle, the longest greater than a semicircle. Gulf Stream. A warm ocean current that flows out of the Gulf of Mexico along the coast of Florida. It is ascribed to ttie action of the TRADE WINDS which cause a mass of water to flow into the Gulf from the East. The current near Florida is strengthened by water which branches from the main trade wind current and flows outside the Antilles. The Gulf Stream flows Northward into the region of prevailing Westerly winds, which 140 Glossary. cause a current to flow slowly Eastward across the Atlantic ; this current, also called the Gulf Stream, carries water from the Gulf Stream proper to the coasts of Europe. The air no doubt has its temperature slightly raised by the warm current, but our temperate climate is due to prevailing Westerly and South Westerly winds, which are also the cause of the Eastward extension of the Gulf Stream. Gust. A " coup de vent." The word was used originally for any transient blast of wind, but is now limited to the comparatively rapid fluctuations in the strength of the wind which are specially characteristic of winds near the surface of the earth, and are probably due to the turbulent or eddy motion arising from the FRICTION offered^ by the ground to the flow of the current of air. The subject of gusts, as indicated by a tube-anemo- graph, has been investigated for the Advisory Com- mittee for Aeronautics by the Meteorological Office, and the results are contained in four reports on Wind Structure published in the annual reports of the Com- mittee. The number and extent of the fluctuations are very irregular ; they have been counted as seventeen in the minute, but another count would probably give a different figure. If the wind be regarded as fluctuating between a gust and a lull, the range between gusts and lulls is dependent on the one hand on the mean velocity of the wind, and on the other hand upon the nature of the exposure of the anemometer. Expressing the fluc- tuations as a percentage of the mean velocity we get the following results for various anemometers. (Report of the Advisory Committee, 1910.) Gust. 141 Anemometer. Range of Fluctuation as a Percentage of the Mean Velocity. Southport (Marshside) Scilly (St. Mary's) Shoeburyness, E.N.E. wind W. wind Holy head (Salt Island) Falmouth (Pendennis), S. wind ... W. wind... Aberdeen 30 per cent. 50 30 80 50 25 50 IOO Alnwick 80 Kew IOO In this table a fluctuation of 100 por cent, means that a wind with a mean velocity of 30 miles per hour fluctuates over a range of 30 miles per hour, between 15 miles an hour and 45 miles an hour, in consequence of the gustiness. The most gusty exposure within the experience of the Meteorological Office is at Dyce, in Aberdeenshire, where, for the purpose of inquiry, an anemometer was installed by Dr. J. E. Crombie, with its head projecting 15 feet above the tree-tops of a small wood. Gusts are to be distinguished from squalls. A squall is a blast of wind occurring suddenly, lasting for some minutes at least, and dying away as suddenly. A squall is attributable to meteorological causes, whereas gusts are the result of mechanical interference with the steady flow of air. m/s. 1905. Pendennis, 46-0 1906. Scilly, 38-4 1907. Southport, 36-2 1908. Scilly, 37'6 1909. Scilly, 40*2 1910. Pendennis, 38-9 1911. Eskdalemuir, 40*2 1912. Pendennis, 43'8 IQI3- Southport, 38-4 1914- Quilty, 4I*O ^S- Pendennis, 4O'O 1916. r\ Pendennis, 40-8 142 Glossary. The strongest gusts recorded on anemometers of the Meteorological Office in recent years are : mi/hr. 103 86 81 84 90 87 90 98 86 92 89 91 Gustiness. The name given to the factor which is used to define the range of the gusts shown on the record of an anemometer. The gustdness of an interval is the factor, (maximum velocity minimum velocity) -f- mean velocity. The figures given for the fluctuations of wind in the records of various anemometers given a^ove may be called the " percentage gustiness " of the winds. To obtain an estimate of the relative gustiness of the winds in the upper air, Mr J. S. Dines used the pull of a kite wire defining the gustiness as (maximum pull minimum pull) -f- mean pull. Using this method it appears that gustiness falls off rapidly in the first 500 feet of ascent, and thereafter it is irregular. (Second Report on Wind Structure, p. 10.) Hail. Usually described as frozen raindrops, though hailstones are often very much larger than any raindrop an possibly be. Hail is formed in the columns of rapidly Hail. 143 ascending air that are part of the mechanical process of a rain-storm or thunderstorm. They are associated \vi h the cumulo-nimbus type of cl>ud. The convection currents which begin with insta' ility in the atmosphere result, first in heavy cloud, and then in raindrops still carried upward in air which is automatically becoming colder in consequence of the diminished pressure. So the drops may freeze, and then any further upward journey may result in condensation in the form of ice on the already formed hailstone. To maintain a mass of water or ice in the air a very vigorous ascending current is required. If a raindrop reaches a certain size it is broken up into smaller drops by the current which is necessary to keep it from falling, but when the hailstone is once formed there is no limita- tion of that kind upon its growth. From their structure, which is often very composite, it is clear that hailstones have a long history, and from their size, which may be large enough to give measurements, it is said, of three or four inches in diameter, a pound or more in weight, they must have required ascending currents of great velocity to support them. There is, however, evidence to show that some of the strongest winds of the earth are katabatic winds, that is, they are due to falling air, so it requires only a special adjustment of the temperature of the environment to give rise to currents of rising air, anabatic winds, of the most violent character. (SOFT HAIL, see p. 343.) Halo. The term halo is an inclusive one applied to all the optical phenomena produced by regular REFRAC- TION, with or without accompanying reflection, of the 144 Glossary. rays of the sun or moon in clouds consisting of ice- crystals (see CLOTD, Cirro-Nebula). The most common halo is a luminous ring of 22 radius surrounding the sun or moon, the space within it appearing less bright than the rest of the sky. The ring, if faint, is white if more strongly developed its inner edge is a pure red, while yellow and green follow, more faintly. Next in order of frequency of occurrence is a similar but larger ring of 46 radius. MOCK SUNS are simply more brilliant patches occurring at certain definite points in a halo system. There is a great variety of minor and rarer halo phenomena. (For some of these see Observer's Handbook, p. 57.) In polar regions, where ice crystals extend much lowe^ in the atmosphere, halo systems attain great brilliance and complexity. Halos are very varied in form, they are produced by the REFRACTION of the sun's rays, or the moon's rays, through a cloud of ice crystals forming what is called cirro-nebula or cirrus-haze, one of the highest forms of cloud. They are of great interest from the point of view of the physics of the atmosphere, but they have no^ meteorological significance. In weather lore they are' often spoken of as presaging storms and it is possible that the ice cloud is one of the earliest results of the fall of pressure with which the storm is associated ; but the formation of a halo is not by any means a necessary step in the preparation of a storm ; many storms arrive without announcing their coming in that way. Moreover, the appearance of a halo at the end of a spell of dirty weather is said to be a sign of clearing. We may perhaps conclude that cirro-nebula, with no other clouds in the sky to interfere (the condition for seeing a halo), may be found at the beginning or the end of a depression. Harmattan. 145 Harmattan. A very dry wind which is prevalent in Western Africa during the dry season (November to March). During these months, (the winter of the Northern Hemisphere) the air over the desert of Sahara cools rapidly, owing to its clearness and lack of moisture, so that it tends to flow outwards to the coast, especially south-westwards to the Gulf of Guinea, and replace the lighter air there. Being here both dry and relatively cool, it forms a welcome relief from the steady damp heat of the tropics, and from its health-giving powers it is known locally as " The Doctor," in spite of the fact that it carries with it from the desert great quantities of impalpable dust, which penetrates into houses by every crack. This dust is often carried in sufficient quantity to form a thick haze, which impedes navigation on the rivers. Harmonic Analysis. See p. 311. Haze. Obscurity of the atmosphere which may occur in dry weather and may be due to dust or smoke, or merely to irregularities of density and consequent irregular refraction of the light by which distant objects are seen. During HARMATTAN winds off the West Coast of Africa, dust haze is thick enough to be classed as fog. At sea the weather is often classed as hazy when there is no distant horizon, and yet no visible mist or fog. The obscurity may, however, be due to water particles. It would therefore be desirable to limit the use of the word haze to occasions when the air is not very damp, that is when there is a noticeable difference between readings of the wet and the dry buib thermometers. Heat. The name used for the immediate cause of the sensation of warmth, a primary sensation which is easily recognised and needs no explanation. As used in relation 146 Glossary. to the weather, heat and cold are familiar words for opposite extremes of temperature of the air. What the American writers call a heat-wave is a spell of hot wenther in which the maximum temperatures reach 90 or 100 F. (above 305a.), and a cold-wave is a spell of the opposite character during which temperatures in the neighbour- hood of the Fahrenheit zero, or 32 degrees of frost, may be experienced. In continental climates, during the passage of severe cyclonic depressions, the transitions from heat to cold are sometimes extremely abrupt and far-reaching ; a difference of temperature of 50 F. in a few hours is not unknown. We have visitations of similar character in this country, but they are less intense. A few days in succession with a temperature over 80 F. would suffice for a heat-wave, and a few days with 10 of frost would certainly be called a cold-wave. One of the most noticeable features of our climate is the succession of cold spells which interrupt the genial weather of late spring and early summer. They are not very intense, but a drop in the mean temperature of the day from 55 F. to 45 F., which roughly defines them, produces a very distinct impression, As used in connexion with the study of the atmosphere heat has another sense which must not be overlooked. It denotes the physical quantity, the reception of which makes things warmer, and its departure makes them colder. If you wish to make water hot, you supply heat to it from a fire or a gas-burner or, in modern days, by an electric heater, a very convenient contrivance for getting heat exactly where you want it. On the other hand, if you want water to become cooler, you leave it where its heat can escape, by CONDUCTION, aided by CONVECTION or by RADIATION. You can also warm water by adding Heat. 147 some hot water to it, or cool it by adding cold water to it. Either process suggests the idea of having the same quantity of heat to deal with altogether, but distributing it, or diluting it, by mixing. The idea of having a definite quantity of heat to deal with, and passing it from one body to another is so easily appreciated and so generally applicable, that the older philosophers used to talk confidently of heat as a substance which they called Caloric, and which might be transferred from one body to another without losing its identity. They measured heat, as we do still, by noting by how much it would raise the temperature of a measured quantity of water. For students of physics the unit of heat is still a gramme-calorie, the heat which will raise a gramme of water through one degree centigrade. To raise m grammes from ^C to t 2 C, m( 2 -i) gramme calories are required. The amount can be recovered, if none has been lost meanwhile, by cooling the water. If we wish to be very precise, a small correction is required on account of the variation in what is called the capacity for heat of water at different temperatures, but that need not detain us. For students of engineering the unit, called the British Thermal Unit, is a pound-Fahrenheit unit instead of the gramme-centigrade unit, and the heat required to raise m pounds of water from ^F to 2 F is m (t 2 - tj B.T.U. It is in many ways a misfortune that students of Physics and Engineering do not use the same unit. It is no doubt a good mental exercise to learn to use either indiscriminately without confusion, but it takes time. From measurements of heat we get the idea that with different substances the same change of temperature requires different quantities of heat ; the substances have 148 Glossary. different capacities for heat. We define capacity for heat as the heat required to raise a unit of the substance (1 gramme or 1 Ib.) through 1 degree. It is a remarkable fact that of all common substances water has the greatest capacity for heat. It take* one unit to raise the temperature of a unit mass of water one degree, it takes less than a unit, sometimes only a small fraction of a unit, to raise the temperature of the same amount of another substance through one degree. We give the name specific heat to the ratio of the capacity for heat of any substance to the capacity for heat of water. Numerically, specific heat is the same as the capacity for heat in thermal units. The specific heat of water is 1, the specific heat of any other common substance is less than 1. The specific heat of copper is only 1/11. So the heat which will raise the temperature of a pound of copper 1 will only raise the temperature of a pound of water 1/11, or the heat which will raise the temperature of a mass of water 1 will raise the temperature of the same mass of copper 11. This peculiar property of water makes it very useful for storing heat and carrying it about. From that point of view it is the best of all substances for cooling the condenser of an engine, for distributing heat at a moderate temperature in a circulating system, and for many other economic purposes. In meteorology its influence is very wide. Large masses of water, of which the ocean is a magnificent example, are huge store houses which take up immense quantities of heat from the air when it is warm and give it out again when the air is cold, with very little change in its own temperature, so that a large lake, and still more the ocean, has a great influence in reducing the extremes HeaL 149 of temperature of summer and winter, and of day and night, in the countries which border it. There is another remarkable storage of heat in which water takes a predominant share that is dealt with in physical science under the name of latent heat. Water at 288a. (59 F.) cannot be evaporated into water-vapour unless every gramme of it is supplied with 589 calories of heat, which produce no effect at all upon the temperature. -.The water is at 288a. to start with, and the water- vapour is at exactly the same temperature and yet 589 calories of heat have gone. They are latent in the water- vapour but produce no effect on the thermo- meter. You can get them back again easily enough if you condense the vapour back again into water, but you must manage somehow to take away the heat while the condensation is taking place. The separation of the u waters that are above the firmament from the waters that are below the firmament," or in modern language, the evaporation of water from the sea or a lake or the wet earth and its condensation in the form of clouds and rain, implies the transference of enormous quantities of heat from the surface to the upper air, the dynamical effect of which belongs to another chapter of the romantic story of heat which deserves more than the few words which we can afford for it. Readers can find an interesting account in Tyndall's Heat a Mode of Motion. The idea of heat as an indestructible substance, caloric, which could be transferred from one body to another without loss, became untenable when it was found that when air was allowed to expand iri a cylinder it cooled spontaneously to an extent that corresponded exactly, so far as could be ascertained, with the means then available, with the amount of mechanical work that the 150 Glossary. cylinder was allowed to do. It was the last step in the process of reasoning by which men had come to the con- clusion that, when mechanical work was devoted to churning water or some other frictional process, heat was actually produced, not brought from some other substance but created by the frictional process. It took many years for men to reconcile themselves to so novel an idea, and a good deal of ingenuity was devoted to trying to evade it, but it has now become the foundation stone of physical science. Heat is not an unalterable indestructible substance but a form of energy. It can do mechanical work in a steam-engine or a gas engine or an oil-engine, but for every foot-pound* of work that is done a corresponding amount of heat must disappear, and in place of it a corresponding amount of some other form of energy is produced. A good deal of heat, besides, may be wasted in the process so far as practical purposes are concerned. In a steam engine, of the whole amount of heat used, only one tenth may be transformed, the rest wasted, as we have said ; but it is still there raising the temperature of the water of the condenser or performing some other unproductive but necessary duty. There is, therefore, a numerical equivalent between heat and other forms of energy. We give the relation : 1 B.T.U. is equivalent to 777 foot-pounds of energy. 1 gramme-calorie = 42,640 gramme-centimetres. = 41,830,000 ergs. * Afoot-pound of work is the work done in lifting one pound through. a distance of one foot. A gramme-centimetre is the work done lifting one gramme through one centimetre. An erg is the absolute unit of work on the C.G.S. system ; 1 gramme centimetre = 981 ergs. Heat. 151 We have led up to this statement in order to point out how extraordinarily powerful heat can be in producing mechanical energy. If, in the operations of nature, one single cubic metre of air gets its temperature reduced by l c 0. in such a way that the heat is converted into work by being made to move air, the equivalent of energy would be a cubic metre of air moving with a velocity of nearly 45 m/s. (101 mi/hr.). So familiar have we become with heat as a form of energy that we measure the heat of sunlight in joules* and the intensity of sunshine in watts per square centi- metre, i.e.. the number of joules falling on one square centimetre per second. THE SPECIFIC HEAT OF AIR. The foregoing statement is necessary to lead up to a matter of fundamental importance in the physics of the atmosphere, namely the heat that is required or used to alter the temperature of air in the processes of weather ; in technical language this is the capacity for heat of air or the specific heat of air. We have explained that when air is allowed to do work on its environment, in expanding, heat disappears, or more strictly is transformed. So the amount of heat required to warm air through a certain number of degrees depends upon how much expansion is allowed during the process. The most economical way of warming air from the * A joule is a more convenient unit than the small unit, the erg ; one joule = ten million ergs (10 7 ergs) and one calorie = 4*18 joules. The icatt is a unit of power, that is, rate of doing work ; a power of one watt does one joule of work per second. 152 Glossary, thermal point of view is to prevent its expanding altogether ; it then has " constant volume " and its specific heat is 0*1715 calories per gramme per degree at 273 a. It is remarkable, but true, that if you have a bottle full of air, it will take more heat to raise the temperature of each gramme of it by a degree if you take the stopper out while the warming is going on, than if you keep it tight. The difference between warming a bottle of air with the stopper in and with it out, simple as it may seem, has got in it the whole principle of heat as a form of energy. The effect of leaving the stopper out is that the pressure of the air inside the bottle is the atmospheric pressure for the time being and is therefore practically constant throughout the brief operation. So we get the specific heat of air at constant pressure 0*2417 gramme-calories per gramme per degree, or 1*010 joules. The specific heat of air at constant volume is 0*72 joules. The difference of the two represents the heat equivalent of the work used in expanding unit mass of the gas against atmospheric pressure. High. Sometimes used as a contraction for high barometric pressure. The technical term anticyclone was coined by Sir F. Galton for the purpose, but, whether for the sake of brevity or for some 'other reason, a " high " is often spoken of. Hoar Frost. A feathery deposit of ice formed upon leaves and twigs in the same way as DEW (q.v.) by the .cooling of exposed objects through the radiation of their heat to the clear sky. Horizontal in the plane of the horizon. The surface of still water is horizontal. In dynamics and physics a Horizontal. 153 horizontal line is a line at right-angles to the direction of the force of gravity which is vertical and identified by the plumb line. The Visible Horizon, or Distance of Visibilitl| for objects of given height 1,000 900 800 700 600 500 400 500 200 100 10.000 9.000 8000 7.000 6,000 5.000 4000 3.000 2.000 1.000 \ / \ / / / =f I / / o | c ft 7T/ y 5 1 I / <^ ^ / S I/ ^ / / </ / ^ Curve I rf/ti r. to L *// Hf nil Ht ight- * (Lf^le. X / . -r TL ght > MMes. 10 20 30 40 50 60 70 80 90 100 110 120 130 Miles Diagram showing the relation between the height of an observation point in feet and the distance of the Visible Horizon in miles {neglecting refraction), or the height in feet of a cloud or other distant object and the distance in miles at which it is visible on the horizon. The " sensible or visible horizon " which is visible from a ship at sea, the line where sea and sky apparently join, is a circle surrounding the observer a little below the plane of the horizon in consequence of the level of the earth's surface being curved and not flat. The depth of the " sensible horizon " below the " rational horizon " or horizontal plane is approximately the same as the elevation of the point from which the " sensible 154 Glossary. horizon " is viewed. Apart from any influence of the atmosphere the distance of the visible horizon for an elevation of 100 feet (30 metres) is about 12 miles. The actual distance is about 2 miles greater on account of refraction. It varies as the square root of the height, so that it would require a height of 400 feet to give a horizon 24 miles off. A level canopy of clouds 10,000 feet high is visible from a point on the earth's surface for a distance of about 125 miles, or the visible canopy has a width of 250 miles. Horse Latitudes. The belts of calms, light winds and fine, clear weather between the TRADE WIND belts and the prevailing Westerly winds of higher latitudes. The belts move North and South after the Sun in a similar way to the DOLDRUMS q.v. Humidity, in a general sense means dampness, but in meteorology it is used for RELATIVE HUMIDITY and means the ratio of the actual amount of aqueous vapour in a measured volume of air to the amount which the volume would contain if the air were saturated. (See AQUEOUS VAPOUR.) In practice, at climatological stations, the humidity of air is determined from the readings of the dry and wet bulbs with the aid of tables prepared for the purpose and called humidity tables or psychrometric tables. But humidity is the most variable of the ordinary meteorological elements, as it depends not only on the sample of air under observation but also on its temperature. Hence the record of a self-recording hair- hygrometer which can be obtained in a form not much different from an ordinary barograph gives a most instruc- tive record. In the spring and summer it sometimes Humidity. 155 shows very high humidity in the night and early morning, approaching or actually reaching saturation, and very great dryness, perhaps only from 15 to 20 per cent, humidity, in the sunny part of the day, with very rapid changes soon after sunrise and towards sunset.* These are the changes which correspond with the characteristic changes in the feeling of the air at the beginning and end of the day. Hurricane in French, ouragan, in German, Danish and Swedish, orkan. " A name [of Spanish or Portuguese origin] given primarily to the violent wind-storms of the West Indies which are cyclones of diameter of from 50 to 1,000 miles, wherein the air moves with a velocity of from 80 to 130 miles an hour round a central calm space which, with the whole system, advances in a straight or curved track ; hence any storm or tempest in which the wind blows with terrific violence " (New English Dictionary), The hurricanes of the Western Pacific Ocean are called typhoons in China, and .baguios in the Philippine Islands. Those of the Indian Ocean, which are experienced in India, are called by the Indian meteoro- logists cyclones of the Arabian Sea or of the Bay of Bengal ; while the hurricanes of the South Indian Ocean which visit Mauritius are also called cyclones. Shakespeare uses the word hurricano for a water-spout. Overleaf is a reproduction of a barogram showing the variation of pressure during a cyclone which passed over Cocos Island, Sumatra, in 1909, November 27th. It is interesting to notice that in spite of the rapid fall of pressure with the onset of the cyclone the diurnal varia- tion of the barometer is still apparent and it reappears before the normal level is recovered. 156 Glossary. Coco s Island ttarogram November, 1909 The occurrence of hurricanes shows a marked seasonal variation. The following table is taken from the Barometer Manual for the Use of Seamen. Variation of the Barometer during a Hurricane. 157 lO 10 00 CS CN vO to Tt" M \O tO ' r^oj, CO CO M CS i i . co * VD ^5 aaqraaooQ; H 5 ' ^ 10 ON OO vO 1>> !>. rh a X W . i i i i 11 S -wqowo S' to cs M IT) <N 00 M CO CO , o \ 00 vO CO \O CS oo \ o J ^Sn ? ^ " + C, | w - ^ M ^*" CO M ** 1 "S \U v ^" <N 1 s "d '9Utl P O H CO CS O O < 3 * | 11 'A'B pff 10 ON ON M l-t IT) OO *-> ft [udy ^ a - ON 10 H H M 00 qo.rej\[ JjJJ ON M to C< to i ^^ 1 CO j <^ tonuqaj t>. ** M 1 M 1 M 1 rs 1 -.fcumn.. to K. " <N 10 1 ^ 1 CO ^N oo : . CO 1 O 2 CO 2 : : o M CO ON ON t I 7 & O o 1^ Goo | ON LI t> s co i3 r- oc oo ^ >j OO hi "5> CO O r1 I 2 00 > S "^r 10 bo fl ^ " O S fl S 1 P I West Indi P! ^ 1 ^s J 1- ! 02 cq cs 02 o> <y 09 a w p3 q_| CS 4-( s S ^ 5 c3 ^3 *- eS Q 158 Glossary. The force of the wind which is experienced in hurri- canes is equalled, if not surpassed, in the tornadoes which occur on the American Continent, but the area affected by a tornado is generally a narrow strip a few miles at most in width. In the Beaufort Scale of wind force the name hurricane is given to a wind of force 12, and its velocity equivalent is set at an hourly velocity exceeding 34 m/s, or 75 mi/hr, but from what has been said under GUST it must be under- stood that at all ordinary exposures a wind with an hourly velocity of 75 miles an hour will include gusts of con- siderably higher velocity, reaching a hundred miles an hour or more. The strongest recorded gust in the British Isles marked 103 mi/hr on the anemometer at Pendennis Castle on March 14, 1905. Hydrometer. An instrument for measuring the density or specific gravity of sea- water. (See Marine Observer's Handbook, M.O. Publication 218.) Hydrosphere. The name given to the layer of water of irregular shape and depth lying on the earth's surface, between the geosphere, or the solid earth below, and the atmosphere, the gaseous envelope above. HyetOgraph.. A self-recording RAIN-GAUGE, an instrument for recording automatically and graphically the fall of rain. (See Observer's Handbook.) Hygrograph.* A self-recording HYGROMETER, an instrument for recording automatically the humidity of the atmosphere. Some form of hair-hygrometer is generally employed for the purpose. Hygrometer. 159 Hygrometer. An instrument for determining the humidity of the atmosphere. Almost all materials exposed to the weather are affected by the humidity of the air, so that it is easy to form a rough estimate of whether the air is damp or dry. Many different materials such as hair catgut, the awm or beard of the wild oat, flannel, have been used in instruments to give an indication of the state of the atmosphere in this respect. But for the purposes of meteorology there are three well-known forms of hygrometer : the hair-hygrometer, the indications of which depend upon the length of a hair or a bundle of hairs exposed to air of different states of moisture ; the dew-point hygrometer, in which a polished surface is artificially cooled until a deposit of dew is produced and the DEW- POINT determined ; and the PSYCHROMETER, or wet and dry bulb hygrometer, in which the temperatures of a bulb covered with moistened muslin and of a dry bulb close to it are read and the humidity determined by tables. The psychrometer is in almost universal use at meteoro- logical stations, as it is the least dependent upon the skill of the observer ; but a few hair-hygrometers are also em- ployed for eye observations, and for automatic records either at the surface or in soundings with kites or BALLONS- SONDES the hair-hygrometer is generally used. HygTOSCOpe. An instrument for showing whether the air is dry or damp. If its indications are sufficiently regular to permit of graduations, it can be made into a HYGROMETER. Any substance which is hygroscopic, that is to say, which is affected in shape, size, or appearance by the variations of moisture in the air can be used as a hygroscope. A bundle of seaweed is sometimes used, (the hygroscopic substance in that case is the salt) ; the 160 Glossary. ordinary " Jacky and Jenny " in a toy house with a catgut support is another example. Hypsometer. The word is derived from hupsos, and means an instrument for measuring height, but it is employed exclusively for apparatus for determining very precisely the temperature of the boiling point of water. That amounts to the same thing as measuring the pressure at which the water is boiled, because the boiling point depends upon the pressure of the atmosphere, and a table of the relation between the two makes the reading of the temperature equivalent to a reading of the barometer. Table of the boiling point of water under pressures occurring in the atmosphere up to about 8000 feet. Pressure. Boiling Point. Millimetres of mercury at 0C. sea level, lat. 45. Millibars. a. mm. mb. 374 787-67 1050* 12 373 760 oo 1013-23 372 733*i6 977'45 3/1 707-13 942-74 37o 681-88 909 * 08 3 6 9 657-40 876-44 368 633-66 844-79 3 6 7 610-64 814- 10 366 588-33 784-36 365 566-71 755-54 364 545-77 727-62 Hypsometer. 161 From the pressure (with a corresponding reading lower down, which may, should, or must be assumed) the height can be computed. The hypsometer has advantages for measuring heights as a substitute for a mercury barometer, which is a troublesome instrument to carry on a journey of ex- ploration. With a pair of thermometers that have all modern improvements, and with careful manipulation, the temperature can be measured to one-thousandth of a degree, corresponding approximately with. '0^1 inch, or 03 millibar, that is to say, the pressure can be determine 1 to the equivalent of one foot of height. That, however, is not to be attained by the inexpert traveller in a hurry. Ice. See p. 321. Iceberg. A large mass of ice that breaks away from the tongue of a glacier running into the sea and floats away. An accoun t of the subject is given in the Keamaris Handbook. Icebergs drift with favourable winds and currents into latitudes of forty or fifty de^r^es. The final period of their life history is not very well understood ; there seems to be a sudden ending that is not accounted for. Nobody has apparently "stood by" an iceberg on the track of Atlantic steamers until the end came. Nor do we know through how "many seasons, for example, a North Atlantic iceberg floats, or lies aground, between its " calving " and its dissolution. It probably weathers one season but collapses in the second. Incandescence. The spontaneous glow of a substance in consequence of its temperature. The word is now quite familiar in consequence of the incandescent or glow-lamp which is luminous on account of the temperature to which the carbon or metallic filament is raised by the electric 13204 F 162 (Glossary. current. Every substance becomes incandescent when heated to a sufficiently high temperature : thus lightning is presumably incandescent air, the sun incandescent vapour. The temperature at which incandescence begins is different with different substances, so the following figures are only roughly approximate : Red hot covers a wide range beginning with 800a. Dull red is about lOOOa. Cherry red 1200a. Orange 1400a. White hot 1500a. The sun GUOOa. Carbon melts at 4300a, platinum at 2000a. Index, the pointer which moves on the scale of an instrument and by which the reading is taken. Two indexes, the two hands or fingers, are required to tell the time by a watch, but only one index is required in read- ing the barometer. The index in this case is the top of the mercury column, so also in the ordinary thermometer the end of the mercury is the index. In a maximum thermometer the outer end of the detached thread of mercury is the index. In a minimum thermometer a special index is introduced into the spirit, which is trans- parent. In both these cases the index has to be set after one reading to make it ready for another. In like manner every measuring instrument has its index. Index-error. See ERROR. Insolation. Originally exposure to sunshine ; solar- isation is also used in the same sense.. It is now applied to the solar radiation received by terrestrial or planetary objects (Willis Moore). The amount of solar-radiation which reaches any particular part of the earth's surface in any one day depends upon (1) the constant of solar-radiation, (2) the area of the intercepting surface and its inclination to the Insolation. 163 sun's rays, (3) the transparency of the atmosphere, (4) the position of the earth in its orbit. The following table quoted from Angot is taken from Willis Moore's Descrip- tive Meteorology. Calculated Insolation Reaching Earth, assuming the mean coefficient of transparency of the atmosphere to be 0'6. (Angot). 1 *> >; b 1 1 1 ^3 f_* ,4 43 a 9 o a a 49 p o & 9 be 5 CS 0> ft p o> H 1-3 ^ s "-3 -5 <1 r jj fc p 5* N. 90 O'O O'O O'O 1-4 6-7 9'9 7*9 2-4 O'l O'O O'O O'O 28-4 80 O'O O'O 0'2 2-7 7*5 10-3 8'5 3*8 0-5 O'O o-o O'O 33*5 60 O'l I'O 3*9 8-2 I2'oi3'8 I2'6 9-2 4*9 i'5 O'2 O'O 67*4 40 3*3 5*7 9*4 12-9 15*316-2 15-6 13*5 I0'2 6-6 3*8 2-7 115-2 20 9-0 II-2 13*6 15-2 15-815-9 15-8 15*3 H'c 11-7 9*4 8'2 I55*i Equator s. 14-0 14-9 15*3 14-6 13*5 I2'8 13*1 14-2 15-0 14-2 13-6 170-2 20 16-815-9 13*9 II'2 8-8 7*7 8'3 10-5 13*1 is;3 i6'6 17-0 155-1 4 16-6 13-9 9*9 6-0 3-4 2'4J 3*0 5*2 8-8 15-9 17*3 115*2 60 I3'4 9*2 4*4 1-3! o-i o-o, o-i 0-8 3*4 7*8 12-3 14-6 67*4 8o c 90 8'8 3*5 2' I 0-4 o-o O'O O'O O'O O'O O'O O'O O'OJ O'O O'l O'O O'O O'O 2-3 I'O TA II'O 10-5 33-5 28-4 The unit is the amount of energy that would be received on unit area at the equator in one day, at the equinox, with the sun at mean distance if the atmosphere were completely transparent. It is 458'4 times the solar constant, or in gramme calories per minute 885, taking the solar constant to be 1-93. 13204 Glossary. Inversion. An abbreviation for u inversion of temperature-gradient" (see GRADIENT). The tempera- ture of the air generally gets lower with increasing height but occasionally the reverse is the case, and when the temperature increases with height there is said to be an " inversion ". There is an inversion at the top of a fog-layer, and generally at the top of other clouds of the stratus type. Inversions are shown in the diagram of variation of temperature with height in the upper air, p. 38, by the slope of the lines upwards towards the right instead of towards the left, which is the usual slope. In the troposphere inversions do not generally extend over any great range of height ; the fall of temperature recovers its march until the lower boundary of the stratosphere is reached. At that layer there is generally a slight inversion beyond which the region is isothermal, so far as height is concerned. For that reason the lower boundary of the stratosphere is often called the fc< upper inversion ". In some soundings with ballons-sondes from Batavia the inversion has been found to extend upwards for several kilometres from the commencement of the strato- sphere. It is important also to note that frequently in anti- cyclonic weather, and especially cold anticyclonic weather, there is often an inversion at the surface ; the temperature increases upwards instead of decreasing. Ion. The name selected by Faraday for the com- ponent parts into which a chemical molecule is resolved in a solution by the electrolytic action of an electric current. Of the two component ions one is always elec- tro-positive, and the other electro-negative. The electro- Ion. 165 negative ions consist of atoms of oxygen, chlorine or some other corresponding element or radicle, and the electro- positive ions consist of atoms of hydrogen, potassium, or some other metallic element or radicle. Each electro- positive ion is called a cation. It is charged with a definite quantity of positive electricity and travels with the electric current to the cathode, the conductor by which the current leaves the solution, while the electro- negative ion, called the anion, is charged with an equal quantity of negative electricity and travels against the electric current to the anode, the conductor by which the current enters the solution. It is supposed that a solution which will conduct an electric current is ionised by the spontaneous dissociation of the components of its molecules and the consequent formation of free ions carrying their appropriate electric charges. In a solution, recombination and dissociation are constantly going on and the electric current causes the free ions charged with positive electricity to move slowly with the current and those charged with negative electricity to move against the current. Similarly, a gas may conduct electricity to a less extent, but in the same way, as a solution when it contains free ions, which may be produced by the action of radio- active agents, ultra-violet light, very hot bodies, the combustion of flame and in other ways. The conduction of electricity through the atmosphere is now, therefore, attributed to the free ions which exist in it, and its capacity for conducting electricity is attributed to its ionisation. The ions in the air may be atoms of hydrogen or oxygen, or they may be aggregates of those atoms witfe some other material. ] 66 Glossary. A certain number of the ions in atmospheric air doubt- less arise from the radio-active materials in the soil. These materials give rise to an emanation, as it is called, which must gradually reach the surface through the pores of the soil, The supply will naturally depend on- the state of the soil, whether damp or dry, frozen or covered with snow, and presumably also on variations of the baro- metric pressure which promote or check the escape of emanation. Other ions must be produced by light from the sun. These will naturally chiefly arise at considerable heights above the ground, where sunlight is stronger and relatively richer in ultra-violet light than near the ground. In addition there seems to be some other powerful source at high altitudes, possibly some form of electrical radiation from the sun. lonisation. See p. 322. Iridescence or Irisation, words formed from Iris, the rainbow, to indicate the rainbow- like colours which are sometimes seen on the edges of clouds ; tinted patches, generally of a delicate red and green, sometimes blue and yellow, occasionally seen on cirrus and cirro-cumuius clouds up to about 25 from the sun. They may be also seen at times on the edges of fracto-cumuius or strato- cumulus clouds. The boundary between the two tints is not a circle with the sun as centre, as in a CORONA, but rather tends to follow the outline of the cloud. They are probably not due to the refraction of light by water drops, which produces the colours of the rainbow, but to the diffraction of light scattered by the very small water drops, and are to be classed like the corona with the iridescence of the opal and the mother-of-pearl. Diffraction-colours formed in artificial clouds in the Isanomalies. 167 Bame way as in the corona, become more brilliant as the cloud gets older and the drops more uniform in size. Hence it seems probable that an iridescent cloud is an old cloud that has been drifting for some time. Isabnormals. See ISANOMALIES. Isanomalies. This word is a combination of the prefix iSo- and the word ANOMALY, which, like the more common adjective anomalous, signifies departure or deviation from normal. The normals used for reference are obtained on various plans, Normals of temperature have been obtained by taking the general mean of the observed temperatures of successive parallels of latitude and thus assigning a normal temperature to each latitude ; isanomalies of temperature are then the departures of mean temperature for any place from the normal for its latitude ; places that are relatively warm for their latitude have a positive anomaly, and places that are relatively cold for their latitude a negative one. Isanomalies are then lines on a map showing equality of departure of the average temperature of any place from, the normal for its latitude. There are, however, not many meteorological elements which can be said to have a normal value for latitude, and it is usual to employ as normals for any place the average or mean value for a long period of years. In that case departures, or differences from the normal for the corresponding period, of the value for any one period, say a month or a year, are called ABNORMALS or abnormalities, and a chart showing equality of departure from the normal a chart of ISABNORMALS. Isabnorinal is an objectionable compound because it is made up of a Greek prefix and a Latin body ; if the departures are to be called abnormals the lines of equal departure ought to 168 Glossary. be called equi-abnorrnals, so that the tendency is to use ISANOMALIJSS for lines of equal departure of a value from its long period average. Isentropic. Without change of ENTROPY (q.v.) gen- erally equivalent in meaning to ADIABATIC. ISO. The prefix ISO- is the Greek equivalent of the Latin EQUI- and implies the setting out of lines on a chart or diagram to show the distribution of set values of some meteorological element. The words with this prefix can generally be interpreted 'by the reader on this basis : some examples are set out under the separate headings below. Thus : ISOBARS, from baros, are lines on a chart showing equal barometric pressure. ISOHELS, from helios, are lines showing equal dura- tion of sunshine. ISOHYETS, from huetos, are lines showing equal amounts of rainfall. ISOPLETHS, from plethos, lines showing equal amounts of a meteorological element. The word isogram was recommended for this pur- pose by Sir Francis Galton. ISOTHERMS, from therme, are lines showing equal temperatures. Isobars. If some of the air were removed from a room the pressure inside would be reduced and the pressure outside would force air through the windows and doors till the room was again filled with the normal amount of air. If over any area some of the air were by any means removed the pressure over that area would be reduced, and the pressure of the air in the surrounding Isobars. 169 districts would tend to force air into the region of deficient pressure. Such areas of deficient pressure are found to exist, but for the following reason the air does not actually {low into them. Anything moving above the surface of the earth will continue to move in a straight line if no force acts on it ; but the earth in its rotation turns under the. moving body ; the moving body is therefore apparently deflected to the right in the Northern Hemi- sphere. This is true of a cannon-ball, and it is true of a moving current of air. Hence the wind does not actually blow into the area of low pressure ; air from the North is deflected to the West side of the area, air from the West to the South side, and so on ; the wind therefore instead of blowing straight in from all sides blows round an area of low pressure counterclockwise in the Northern Hemisphere. We thus have the apparent paradox of a force tending to push the air into the centre of the low pressure area while the air is actually moving round the centre at right-angles to the force that is acting on it. There are many examples of similar things in nature ; the Earth's motion round the sun for instance ; or the water in a basin when there is a hole in the centre from which the plug has been taken out ; the slightest circular motion sets up a swirl, and the water moves round the basin at right-angles to the force of gravity which is tending to force it towards the hole. In the case of an area of high pressure the air blows out from the high pressure, but it is deflected to the right as in the former case, with the result that the wind blows round the area of high pressure in a clockwise direction. In either case if you stand with your back to the wind the low pressure will be on your left hand, this is BUYS BALLOT'S LAW. In the Southern Hemisphere the reverse is the case. 170 Glossary. If the observed heights of the BAROMETER (reduced to sea level) from a number of places are put on to a map and arrows are put in to represent the direction of the wind, we have a weather map which at first sight looks like a disordered collection of figures ; we may make it clearer, however, by drawing lines through places where the barometer stands at the same height ; thus we may draw one line through all places where the barometer stands at 1,015 mb., another through all places where it stands at 1,010, and so on. Such lines are called ISOBARS. It may happen that we cannot find any station where the barometer stands at say 1,015 mb., but if we find one where it stands at 1,016 and another where it stands at 1,014 we take it that the 1,015 line passes midway between the two stations. When the isobars are drawn in we can readily see the shapes of the areas of . low and high pressure, and we see also that the wind blows in accord- ance with Buys Ballot's law. The areas of low pressure are called CYCLONES, DEPRESSIONS, or simply LOWS ; the areas of high pressure ANTICYCLONE?, or HiGHS. The isobars are analogous to contour lines on an ordinary map, the high pressures" corresponding to the hills, the low pressures to the valleys. The moving air does not go straight from the highs to the lows, but it blows round the highs in a clockwise, and round the lows in a counter- clockwise direction. On the contours of the earth we may descend from a height of say 1000 feet to 500 feet by a gentle slope many miles in length, or in another place we may descend by a precipitous scarp ; in the former case a stream will run down sluggishly, in the second it will be a swift torrent full of rapids and waterfalls. So with the pressure ; we may travel a Jong way between a place where the baro- meter reads, say, 10^0 millibars to another where it reads Isobars. 171 1015 millibars, or we may have to go only a comparatively short distance. The steepness of the gradient on a map is measured by the distance between the contour lines ; the steepness of the barometric gradient is measured by the nearness of the isobars. The strength of the wind depends on the steepness of the barometric gradient, just as the velocity of the stream depends on the steepness of the slope, but the analogy is not quite perfect for the stream runs down the slope across the contour lines, whereas the wind blows nearly along the isobars with a slight inward curvature towards the low pressure. In the case of the wind close to the surface if the five- millibar isobars are 400 miles apart the barometric gradient is sJight, and the wind will be about 10 miles per hour ; if the distance apart is 60 miles the gradient is steep, and the wind will be about 70 miles per hour. The wind calculated from the barometric gradient is called the GRADIENT WIND or, if no allowance is made for the cur- vature of the path of the air, the GEOSTROPH1C WIND ; it is in most cases the wind met with at or about 1500. feet ; nearer the surface, owing to the friction, the actual wind is less than the gradient wind. The gradient wind as stated depends principally on the distance apart of the isobars ; it is modified, however, to a small extent by the variations of density of the moving air and therefore by the height of the barometer and by the temperature ; a table is given on pp. 172-173 showing the geostrophic wind for various pressures and temperatures according to the formula of p. 136. The values are dependent upon the latitude and are given in the table for two latitudes 52 and 40. Further tables are given in the Computer's Handbook MX). 223. Section II. 172 Glossary. fl.ls-.9i gainst ?s*1gS . |H*m g. O 43 Agsli &!i bc-d* g rs j;*lfi' ^ a ."S ^ t> >i ^:iw^ g-a^|5a l^oJ^SJg 1^23^. o o * O ^ p,^^ o o F> M M CO ft 00^ ON M S 5 a 1C ^ M CO <N 00 CO M 1 1 1 O M oo & f CO o ^ ^ CO P M ^o ri M CO M PH a 10 M CO o co 10 OO M 0? ON M ^ g> ON K r> H-T co <N d 00 M 1 1 vO -o R 1 H f & "ft j O M CO 1 1 -Si 1 s OQ 1 o S O a o h *OQ c 1 m M 8 M P Itobart. 173 01 CO 0^ vo M CO m to co CO n 10 c^ M 01 00 NO M 3 M NO 01 M 01 H M M M > *4-3 ^ i CO r^ o T^- cs CO ^t- 10 ON 01 NO 0^ co 01 01 00 r^ ON 2 o O Th rf- vJ OQ g H r^ ON ON CO R CO M 10 co OO OO r^ 10 <* co t^ M 3- S a p -1-3 rf o of JJ 3 1 1 CO 10 rf s CO vO M n M oo M M 10 M ON Ol M co M H M M M 0^ 1| g.2 M VO ro T 0) vO co Tf ON n oo M 01 R M 3 M 10 01 M $ & r^ oo "5 ^ 13 & O i do 10 % N CO 8 01 vO M 10 CO M NO M M M H 8, 00 dS -*^r o <M iO 10 M ON CO CO OO r^ c^ ON CO M vO C< fc M t^ 10 1 1 M CO M 01 M M % r^ oo 00 t^ as OQ a> o 05 73 d -u oo !>. ro vO LO t^ . CO r^ 10 cs ON oo M lO M NO 01 M OO M lO ON Tj- OO NO r^ d 5 02 +3 3 LO ^q to ON r^ t^ 1O xO co CO Tf n oo M ^ M 01 01 M ? M M ON M 00 CO r^ O M ir> ro 01 g, M lO CO $ <N vO r^ M O * M r^ M M 8 M OO 00 oo r^ o t^ o 3 ^ g^.-s 2 J5_ a '^ n H M .^J- f^ vO CO CO r^ ^ ON io 10 M K NO CO GO r^ 10 ON OO r^ 8 H- 1 op M M M 1*1 3 * 1 -i i a M I/") M lO M cs 10 Ol co IO co 3- lO * O IO 174 Glossary : Lithoplate XI. If we put down on the weather map besides the baro- meter read ings and the wind arrows, the state of the weather we shall see that in anticyclones it is usually fine, while in the depressions it is rainy and cloudy on th^ East side, and finer on the West side of each depression. The weather in fact is intimately connected with the shape of the isobars. As this is a most important fact in meteoro- logy it will be well to consider a little more closely the areas of high and low pressure. A DEPRESSION is a part of the atmosphere where the baro- meter is lower than in the surrounding parts. (See PI. XI). The isobars round such an area are more or less circular or oval, though there are often irregularities. The size of a depression may vary enormously ; one may cover only a part of an English county, another may fill the whole space between our islands and the Arctic circle. Some are much deeper than others ; a deep depression is one where the barometer is very low near the centre ; a shallow de- pression is one where the barometer, though low near the centre, is not very much lower than in the surrounding districts. North of the Equator the wind blows round the depression in a counter-clockwise direction, and the steeper the barometric gradient, that is the deeper the depression, the stronger is the wind. The sky is dull and overcast on the east side of the depression, with rain near the centre, especially heavy on the north-east side. Near the centre in the region where there is an abrupt change of wind direction from south on the east side, to north on the west side of the depression, there is heavy rain and often squalls. On the west side, in the region of northerly or north-westerly winds, the cloud sheet is broken up into detached clouds which get further apart, fewer, and less rainy the further one goes from the centre. f* ANTICYCLONE. DISTRIBUTION OF WEATHER, WIND, *AND PRESSURE, 7 A,M, 17th NOVEMBER, 1915. oV 2CV ISOBARS are drawn for intervals of five i bars. . WIND. D^ection ia shown bj arrowi flying with the wind. Force, on the oala (M2, ii Incji- oated bj the number of fther. Calm x*"\ ATHIR, Shown by tho following symbol* ) clear sky. Q iky Clouded. ) iky * clouded. ) overcaft ky snow A halt ) iky | clouded, i raiii falling Scfotf. s oait. T thunder. "K thundarstorm. COL. DISTRIBUTION OF WEATHER, WIND, AND PRESSURE, 7 A.M. 1ft MAY, 1015. ISOBAR! are drawn for interval! of ten milH- WEATHER. -Shown by the following symbols - _ . bH f * Q clear iky. f) ky J clouded. WIND. Direction if shown by arrow* flying S< . \Z , with the wind. (ft) tk J * ded. ^J) aky j clouded. Force, on the acale 0-12, it indi- jtv overeat iky A rain falling number of feathers W 8now j^ ^^^ s 1^ s*mist. T thunder IT thunderstorm Ps. 1130, 26779. 464, 6000 1/18- o Wy. & S., M.O. Press, S. W. 7. Isobars: Lithoplate XII. 175 If we note a depression on a weather map for one day we shall usually find that on the map for the follow- ing day the depression has moved in an easterly direction ; a depression seldom remains long in one place, and its drift is usually towards some point between north-east and south-east,, though there are exceptions. Since the air in front of a depression is coming from the south, and in the rear from the north, there will often be a great difference of temperature between the two sides. This is particularly noticeable in winter when the approach of a depression is heralded by warm, and its passing away by culd weather. As the depression moves it carries its weather and wind system with it, so that an observer situated on its track would have the following sequence of weather : The barometer begins to fall, the wind becomes southerly, the sky becomes overcast and the weather muggy, and in winter the temperature is well above the normal ; the clouds get thicker, the wind stronger, rain begins to fall ; as the centre approaches the barometer gets lower, the wind gets stronger and the rain heavier ; as the centre passes over there are often gusts of wind or squalls, with heavy rain, " clearing showers"; the barometer ceases to fall and commences to rise ; the clouds show signs of breaking, and the wind changes round to the north or north-west and of tea blows more strongly than it did before the centre passed ; as the centre moves away the wind lessens, the rain ceases, or only occurs in showers, the sky clears. The rate at which these changes take place depends on the size of the depression and its rate of travel ; 24 hours is an ordinary time for such changes to be gone through, but it may be longer and it may be shorter. The above sequence of weather is a typical one, but there are many 176 Glossary: Lithoplate XIII. differences in individual depressions ; some are rainy without much wind, others bring much wind, but not much rain. The above typical sequence of weather will only bo experienced by an observer who is on the track of the centre of the depression. One further south will have south-westerly winds at first, with dull weather, becoming rainy ; the wind will gradually veer to the west when the centre is passing to the north, the barometer will begin to rise and the wind will veer further to the north-west as the centre passes away. An observer a long way from the track of the centre will perhaps only experience a slight fall of the barometer, with cloudy weather. An observer north of the track will have south-easterly or easterly winds at first, with probably much rain ; the wind will back to the north-east or north as the centre passes to the south. The winds on the north side of a depression are usually less strong than those on the south side, the isobars on the polar side being less crowded together than those on the equatorial side. Thus a depression passing on the south gives less wind, but probably more persistent rain, than one passing on the north. Gloomy days with an east or north-east wind, and rain all day, are usually due to depressions passing to the south of the observer. The easterly current on the north side of a depression frequently brings snow in winter. If high clouds are visible they will frequently be seen to be moving away from the centre of low pressure; thus a south wind on the surface, with high clouds moving from the west, is a sure sign of the existence of a depression to the west. X/ fl^Y''\ DEPRESSION. DISTRIBUTION OF WEATHER, WIND, AND PRESSURE, 6 P.M. 13th FEBRUARY, 1915. ISOBARS are drawn for jinterrals of ten mtlH- WEATHER. Shown by the following symbols . bars. r\ clear sky. O**^ i clouded. WIND.- Direction is shown by arrows flying >c , , , ;: i. with the wiftd. 8k * i doa ded - Iky * clouded - Force, on- the scale 0-12, is indi- |flv overcast sky. A rain falling cated by number of feathers ^ gnow ^ hai ]^ s fog Calm /^S =mist, T thunder. 1C thunderstorm Plate XIV. V-SHAPED DEPRESSION, DISTRIBUTION OF WEATHER, WIND, AND PRESSURE, 7 A.M. 8th OCTOBER, 1916. ISOBARS are drawn for intervals of ten mill!- WEATHER. Shown by thefol!owtogyrnboji bars. WIND. Direction is shown by arrows flying with the wind. Force, on the scale 0-12, is indi- cated by the number of feathers. clear iky. Q ky i clouded. sky \ clouded. ^ sky f clouded. overcwt sky. f rain falling snow. A hail. = tog. Tthnnder. TCthunderiVor Isobars : Lithoplate XIV. Ill At times, especially in summer, very small depressions are apt to form ; several such depressions are sometimes seen on the weather map for the same day ; they follow the same laws as large depressions, but being shallow they do not usually occasion much wind; they bring heavy rains and thunderstorms in the summer. An ANTJ CYCLONE,^ or high-pressure system, is the contrary to a depression ; here the barometer is high in the centre, the isobars are usually more or less circular or oval, they are also usually further apart than is the case in a depression, therefore the winds in an anticyclone are usually lighter. An anticyclone is not so small as the smaller depressions, though the large ones may equal it in size. It often coyers a large area. An anticyclone moves but slowly and irregularly ; it may remain in the same position for many days, or even weeks, at a time. The weather in an anticyclone is usually fine and bright, though extensive cloud sheets may form, and fogs are prevalent in winter ; rain seldom falls, and persistent rain never. The depression and the anticyclone are the main arrangements of isobars, but there are five other shapes each of which has its characteristic weather. The SECONDARY DEPRESSION. Sometimes in the neighbourhood of a depression, usually on its southern side, the isobars take a slight bend outwards, marking the position of a small centre of low pressure ; such a depression usually travels forward in the same direction as the main depression, and may even outstrip it in rate of travel. It usually produces much rain, and sometimes much wind. In the summer secondary depressions are * Itttrstraifttms ~oT~ the various types of isobars are given under the respective heading's ANTICYCLONE, SECONDARY. &c. 17.8 Glossary : Lithoplate XV. often very shallow, and resemble the shallow depressions already noticed ; like them they occasion heavy rain and thunderstorms. When they are secondary to deep depressions they often cause a crowding together of the isobars on their southern side, and thus occasion strong winds. The V-SHAPED DEPRESSION. This is a further extension of the Secondarj T ; the isobars, instead of bulging out slightly, extend out a long way in the form of the letter V. The wind on the east side is from a southerly point, that on the west side from a northerly point, in accordance with Buys Ballot's law ; the east side is a region of cloud and rain, often heavy driving rain ; over the central line there is an abrupt change, very fine weather being experienced on the west side. As the Y-shaped depression moves over any place the observer experiences southerly winds and driving rain, both wind and rain becoming stronger as the central line approaches ; as it passes over there is a sudden change of wind to a northerly point ; the rain stops, and the sky rapidly clears ; the central line is often a region of heavy squalls ; there is a marked fall of temperature with the passage of the central line. See Plate XV. WEDGE OF HIGH PRESSURE. Between two depressions there is oiten a region of high pressure where the isobars are shaped like an inverted V ; the high-pressure wedge usually extends from an anticyclone poleward between two depressions that are skirting the northern edge of the high pressure. The wedge moves forward in an easterly direction between the two depressions ; it is in fact merely the relatively high pressure between the depressions. The front of the wedge is often a region of extremely fine weather with northerly winds, rapidly becoming light as Plate Xl/ SECONDARY DEPRESSION. DISTRIBUTION OF WEATHER, WIND, AND PRESSURE, 7 A.M. 22nd FEBRUARY, 1015. ISOBARS are drawn for internals of ten tan, WIND. Direction ii shown by arrows flyiaf with the wind. Force, on the scale 0-12. is indi- cated by the number of feathers /^N W1AT1IW. -Shown by thetollowlBf symbols - clear iky. Q iky $ clouded. jn sky | clouded. <JJ) sky | clouded. T^ overcast slry. rain falling jT snow. A hail. = f off. =mi8t, T thunder. ~K thunderstorm Plate XVI. WEDGE. DISTRIBUTION OF WEATHER, WIND, AND PRESSURE, 7 A.M. 9th DECEMBER, 1915. ISOBARS are drawn for intervals of five milli- WEATHER. Shown by the following symbol a . 8 ah ft O clear iky. (Qtky $ clouded. *) tky J clouded. ^Jj) sky f clouded D overcArt sky rain failing r< A half. = fo$r. T thunder. "K thunderstorm with the wind. Force, on the scale 0-12, is indi- cated by the number of feathers, Calra/^S Isobars : Lithoplate XVI. 179 the central line approaches ; on the central line there is a calm ; after its passage the wind backs to the south-west, clonds begin to come up, and rain usually follows rapidly as the new depression approaches. After the passage of a depression, if the weather clears up very rapidly and the wind falls quickly, it is usually the sign of the approach of a wedge ; in such a case very fine weather may be forecasted for a few hours, followed again by bad weather. " It has cleared up too quickly to last." The COL. A col is a region between two anticyclones, and may be likened to a mountain pass between two higher peaks. Since the wind is blowing round the two anticyclones in a clockwise direction the col is a region where light airs from very different directions are brought into close proximity. This gives conditions for fog in cold weather, and for thunderstorms in hot weather. STRAIGHT ISOBARS. Occasionally the isobars run straight over a very considerable area. In these latitudes straight isobars usually have the lowest pressure to the North, and thus in accordance with Buys Ballot's law the winds are westerly. There may be a great diversity of weather in a region of straight isobars, for it must be remembered that the northern side extends to a low-pres- sure region and the southern side to an anticyclone ; there- fore in the Northern region we get much cloud and some rain, in the Southern clear skies and fine weather. Major Gold, D.S.O., Commandant of the Meteorological Section, G.H.Q., makes the following comments, which fairly illustrate the difficulty of making positive state- ments about the relation of weather to isobars : Straight Isobars. If the anticyclone is a warm one, the Southern side, also, gets cloudy skies and rainy weather, see January 6th, 1916, 180 Glossary. January 4th, 5th, 1914. January 2nd, 10th, llth, 12th, 17th, 18th, 1910, Perhaps there is a seasonal variation in the weather in straight isobars. In January they nearly always seem to gret rain right to the edge of the anticyclone. Straight W. to E. isobars usually mran cool or moderate temperature in Summer and rather mild in Winter. One gets also straight isobars running from S. to N. (See January, 1913 : not much rain but some clond and mist). Thunderstorm weather when it is very warm to the South. Also N. to S. straight isobars (January 2nd, 12th. 1911 ; December, 1913: squally, sno * , hail and sleet weather in winter. September 29th. 30th, October 2nd. 3rd, 4th, 1915 ; showery ; thunder in Flanders). Wedge : when the dominant anticyclone is to the North, it is more stable than the wedge proper (which is very unstable as a rule) and gives X. winds on the E. side and E. winds on the W. side. It is the business of the forecaster who has a weather map before him to note the arrangements of the isobars, and the positions of high and low pressure ; he must note whether the low-pressure systems are main depressions, secondaries, or V-shaped depressions ; he has to judge from the map the directions in which the disturbances are likely to travel, and, knowing the weather which each kind brings, to warn different districts what wind and what kind of weather they are to expect. In judging the direction of travel the meteorologist is guided by certain rules and by past experience ; a depression usually travels from, some point between south-west and north-west to some point between north-east and south-east ; they frequently skirt the Western seaboard of Europe ; they do not pass through anticyclones ; when an anticyclone is situated to the West of these islands depressions do not come in from the Atlantic, but there is then a tendency for depressions to pass from North to South down our Eastern coasts. The meteorologist must also forecast what temperatures are likely ; in the region of Easterly winds on the North, side of a depression cold weather is Isobars. 181 likely, with snow in winter; the approach of a depression from the west during a frost in winter is sure to bring about a thaw on the southern side of its path. The endless varieties of weather must, as far as possible, be foreseen by the meteorologist with the map before him. The solitary observer who has no means of making a map may however recognise some of the signs of the approach of certain types of weather. Remembering BUYS BALLOT'S LAW, and watching his barometer, he may recognise the approach of a depression, and may even on many occasions roughly plot out its track ; he may often tell whether the fine weather is of an anticyclonic type, or whether it is the result of a wedge and therefore only transitory. In short, if he has a knowledge of the principles disclosed by weather maps and has a barometer he will be in a much better position than his neighbours to forecast the weather from local manifestations. Isothermal of equal temperature. An isothermal line is a line of equal temperature, and, therefore, is the same as isotherm. Isothermal is frequently used in meteorological writings on the upper air for the so-called " isothermal layer " by which is meant the layer indicated in the records of all ballons-sondes, of sufficient altitude, by the sudden cessation of fall of temperature with height and generally by a slight INVERSION (see also GRADIENT) followed by practical uniformity of temperature. The layer is not really isothermal. Its temperature on the occasions when simultaneous soundings have been secured at different places shows a temperature-gradient in the stricter sense of difference of temperature at the same level, and an in- spection of the diagram reproduced under BALLON-SONDE, 182 Glossary. representing the results of a large number of soundings in the British Isles, shows that the range of temperature at the highest layer is greater than the range at the surface. But it is also clear from the diagram that in each single sounding the balloon reaches a region where the thermo- meter ceases to fall. To avoid the misconception which the use of the word isothermal for this region would imply, M. Teisserenc de Bort, who was largely instrumental in its discovery, coined the word stratosphere, while he gave the name of troposphere to the region below. These names have -now been generally adopted. The strato- sphere has also been called the advective region, in contradistinction with the convective region below it. We are still without any effective explanation of the origin of differences of temperature which are found in the stratosphere ; they must probably be classed among the most fundamental characteristics of the general circulation of the atmosphere and among the primary causes of the changes of weather, but hardly any light has been thrown on the mechanism of the process. Katabatic. Referring to the downward motion of air due to convection. A local cold wind is called katabatic if it is caused by the gravitation of cold air off high ground ; such a wind may have no relation to the distribution of atmospheric pressure. See BORA and BREEZE. Khamsin. A hot, dry wind which passes over the Egyptian plain from the southward, forming the front of depressions passing eastward along the Eastern Medi- terranean, while there is an area of high pressure to the of the Kile in Middle Egypt. Kilometre. 183 Kilometre. A length of one thousand metres, approximately five-eighths of a mile. Lake. The water that collects in a hollow or depres- sion in the land's surface. In meteorology a lake serves the purpose of a huge rain-gauge, and, subject to some allowance -for lag and evaporation, indicates the variations in the collective rainfall of the area which it drains. For example, the Victoria Nyanza under the equator, apart from gradual fluctuations of level which in the last twenty years have followed closely the variations in the SUNSPOT-NUMBERS (q<v.\ has a seasonal variation which is connected with the seasonal rainfall of the spring and early summer and of the late autumn in equatorial Africa. (see CLIMATIC TABLES). Land-Breeze. A light wind passing across a coast line from the land, sea\\ard. It generally begins with the setting in of coolness in the evening and disappears with the advance of temperature over the land in the day time, or is replaced by a sea-breeze, and it is therefore regarded as a katabatic wind due to convection between the colder layer over the earth and a warmer layer over the sea. In sunny weather there is a large diurnal change of temperature in the land and hardly any in the sea, and the direction of the wind is regarded as alternating with the direction of the temperature-gradient along the level. Lapse, from the Latin lapsus, a slip, a word suggested for use instead of gradient (which is from gradus a step) to denote the loss of temperature or pressure of the atmosphere with height. So that lapse-rate, or lapse-ratio, for temperature will be the fall of temperature per kilo- metre of height. A lapse-line will be a line representing 184 Glossary. the change of temperature with height. The word is connected with the word labile which means liable to slip, and applies technically to the peculiar state of equili- brium of an isentropic, or thoroughly churned atmos- phere (see ENTROPY). The equilibrium is neither stable nor unstable, that is to say, if it is disturbed by slow mechanical process it will, when left to itself, neither go back to its original state nor go forward, but remain indifferent, in its displaced condition. In these circumstances the air will have the greatest possible lapse-rate of temperature short of instability. The lapse rate of temperature may change its sign and indicate an increase of temperature with height. This apparently always happens in and above a layer of fog, and not infrequently in and above other forms of cloud. In that case the lapse-line showing the change of temper- ature with height will have a slope to higher temperature upward, opposite from and therefore easily distinguished from the ordinary slope to lower temperature upward. We refer to that state as a recovery, instead of using the term " inversion of temperature-gradient" which is used at present and is often shortened to " inversion." Generally speaking, the recovery is only temporary in the journey upward and is followed by a relapse with perhaps a different lapse-rate, unless the point has been reached at which the fall of temperature with height ceases. That is the boundary between the stratosphere and the tropo- sphere. We may call that point on the lapse-line the lapse-limit or TROPOPAUSE. The shape of a lapse-line is a very important index of the condition of the upper air ; it has often been deter- mined by the results obtained with a ballon-sonde. The normal average shape has a lapse rather less than that of Lapse. 185 the isentropic atmosphere of saturated air, or almost one half of the " adiabatic gradient " for dry air. The investigation of the air near the sea surface during fog off the Banks, carried out on the u Scotia," has shown that the effect of the mixing of the air of the surface over the cold water is to replace the normal lapse-line at the lower end by' a line which shows a gradual recovery of temperature from the cold surface to the undisturbed condition at a kilometre more or less in height. In the lower, or colder, part the water- vapour is condensed in fog. It would appear that in a region where convection is going on over an extended area there must be an isen- tropic lapse-rate, that the process of the gradual ascent of warmed air is a gradual formation of a thicker isentropic layer. An isentropic lapse-rate seems also to be indicated when mixing takes place in the surface layers owing to turbulence over water which is not less warm than the air in contact with it. Lenticular In shape like a lens or lentil. The word is used to identify a cloud of characteristic shape formed by a large mass of clustered cloudlets which is apparently disposed horizontally, has well-defined edges, a pointed end and broad middle or base. Sometimes the cloud becomes thin in the broad part and gives one the impression of a horizontal bow or horse-shoe of cloud, foreshortened by being seen from a distant point under- neath it. Level. A surface is level if it is everywhere at right* angles to the force of gravity which is indicated by the plumb-line. When a table is not level the force of gravity makes things roll towards the lower edge, but when we are considering areas so large that the curvature of the 186 Glossary. earth has to be allowed for, the words higher or lower have no meaning unless we can refer the heights to some " level " surface accepted as a datum. The accurate com- parison of levels in different regions of the earth is a problem of the greatest refinement and delicacy. Part of the problem is to determine whether the level of the sea, apart from any disturbance due to waves, is the same all over the world or not. For example, it used to be a debatable question if a cut were made across the Isthmus of Panama from the Atlantic to the Pacific, whether the water would flow from the Atlantic to the Pacific or vice versa. That question has doubtless been solved by the accurate levelling of the engineers of the Panama Canal. On land, levels can be set out with great accuracy by means of a spirit-level, but allowance has to be made for the curvature of the earth. Land-levels in Great Britain are referred to the ordnance datum which is the assumed mean level of the'sea at Liverpool, and is O650 ft. below the mean level of the sea of the British Isles, and in Ireland to low water of spring tides in Dublin Bay, which is 21 ft. below a mark on the base of Poolbeg light- house. The datum in mariners' charts is usually " low water ordinary spring tides." Tidal and river levels in Great Britain are usually referred to Trinity High Water (T.H.W.) 12'47 ft. above Ordnance Datum. Lightning*. The flash of a discharge of electricity between two clouds or between a cloud and the earth. A distinction is drawn between " forked " lightning, in which the path of the actual discharge is visible, and " sheet " lightning, in which all that is seen is the flash of illuminated clouds and which is attributed to the light of a discharge of which the actual path is not visible. Lightning. 187 Since the introduction of photography many photo- graphs of lightning have been obtained, and in general character they cannot be distinguished from photographs of electric discharges of six inches or more in length which are obtained in a laboratory, but the varieties of form of lightning discharges are very numerous. Frequently a flash shows many branches, especially the upper part of a flash 'bet ween the clouds and the earth. Among a collection of photographs thrown upon a lantern screen, Dr. W. J. S. Lockyer once interpolated a photograph of the River Amazon and its tributaries, taken from a map, and the photograph was accepted without comment as a picture of lightning. No satisfactory evidence has yet been produced as to what limits or defines the portion of the atmosphere which is freed from electric stress by a discharge of lightning, nor how the path of the discharge is selected. Lightning-conductors, which are metal rods leading from the salient points of buildings to conductors buried in moist earth, have been used since the time of Benjamin Franklin to protect buildings from damage by lightning. A good deal of attention was devoted to the method of operation of lightning-conductors, especially by Sir Oliver Lodge, whose lectures before the Society of Arts are the best source of information on the subject. The chief use of conductors is supposed to be the relief of stress in the immediate neighbourhood of a building by the so-called silent or brush-discharges from its exposed points. These brush-discharges are often visible in snow storms as discharges from the yards and points of ships or from an ice-axe and other projecting points in high mountains. The phenomenon is known as Corposants, or St. Elmo's fire. For PROTECTION AGAINST LIGHTNING see p. 325. 188 Glossary. Line-Squall. A squall of wind, accompanied by rain or hail, associated \vith a sudden drop of temperature and the passing of a long line or arch of dark cloud. The sequence of events as represented on recording meteoro- logical instruments is one of the most clearly defined and easily recognised of all types of weather. There is a sudden rise of the mercury in the barometer by about 2 mb. (less than *1 in.), a veer of wind through about 8 points, a simultaneous fall of temperature as much as 5 to 10C., or 10 to 20F., a sudden squall of wind, some- times of great violence, lasting for a few minutes. The sequence of phenomena is represented by the sketches made by Mr. G. A. Clarke, of Aberdeen Observatory, with the accompanying records, which are reproduced in figure 1. The four sketches of the line of dark cloud are made at intervals of 2 minutes, and the set represent 6 minutes in the life-history of the cloud. The line of the squall, which is marked locally by the line or arch of the cloud, often extends across the country for hundreds of miles and represents the sudden transition from a southerly wind to a westerly wind, or from a southerly type of weather to a westerly type. The cloud appears to be due to convection between the cold westerly current and the warmer southerly current ; the squall is therefore probably katabatic in its origin, and its violence on the actual passing of the cloud accounted foi in that way ; it represents the dash forward of a breaking wave, or more strictly speaking, of the water of a broken wave. The phenomenon has been studied in the Meteorological Office during the past 10 years. A short account of the results of the investigations is given in Shaw's Fore- casting Weather. To face p. 188. LINE SQUALL : FIGURE 1. LINE-SQUALL AT ABERDEEN! OCT H. i M* CLARKE. Line Squall. 189 Line squalls frequently occur at the time of the passage of the TROUGH of a deep depression when the transition from southerly wind to westerly wind takes place suddenly. They also occur as a preliminary to a thunder- storm, and in such cases the wind of the squall is sometimes very destructive. They form the most serious danger to aircraft ; at the same time, their characteristics lend themselves to forecasting with unusual precision provided their existence is once identified, because the line travels across the country with a very definite velocity. Special arrangements are therefore made to obtain notifications of the passage of a line squall over the stations at the outside edge of our area. Liquid. The name given to a class of fluids. The peculiar property of a liquid is that a limited quantity poured into a sufficiently large vessel forms a definite and permanent layer with a free surface. Gases can also be poured from one vessel to another, but unlike liquids the boundary between the heavy gas at the bottom and the lighter gas above it is obliterated in time and a complete mixture of gases results ; with a liquid the well defined surface of separation remains. Liquids are of all degrees of mobility from pitch which moves only inches in a month, through the stages of treacle, and glycerine, which visibly move, but take time, to water or ether which move at once on tilting and can be " shaken up." Low, used to denote a region of low pressure, in the same way as HIGH is used for the region of high pressure : a depression. See also ISOBARS. Lunar dependent upon lima, the moon ; thus a 190 Glossary. lunar rainbow is a rainbow formed by the rays of the moon, a lunar cycle a cycle dependent upon the moon's motion. A month is really, from its name, a lunar cycle, but the introduction of a calendar month makes it necessary to draw a distinction between it and the lunar month, which is the period from new moon to new moon. In astronomy it is called the synodic month, and is equal to 29*5306 days. The endeavour to bring the month or the revolution of the moon round the earth into relation with the year or the revolution of the earth round the sun, has given rise to the differences of calendar which have been or are in use. Mackerel Sky. A sky covered with cirro-cumulus clouds arranged in a somewhat regular pattern, and show- ing blue sky in the gaps. See CIRRO-CUMULUS and CLOUDS. Magnetic Needle. A strip of steel permanently magnetised and provided with an agate cup for balancing on a point, like the needle of a compass. See COMPASS. MammatO - cumulus. When low clouds have rounded projections, or pap-like protuberances, from their under surface the term mamma to-cumulus is applied to them. They are appropriate to the disturbed atmospheric conditions which accompany the close of a thunderstorm. An example is given in the accompanying illustration, obtained by Captain Cave at Ditcham Park. They do not occur often in England and never persist for long. Their relation to cumulus clouds in the ordinary sense is not GLOSSARY. To face p. 190. Mammato Cumulus (Festoon Cloud) after thunderstorm in August, 1915. This photograph is reproduced in illustration of the commotion which occurs in the atmosphere in the various stages of a thunderstorm. It may be taken as the sequel to the illustration of the Cumulo-Nimbus cloud shown in figure 1 reproduced under CLOUD. The cumulo-nimbus is a thundercloud approaching, the mammato cumulus a thunderstorm receding. Mamma to -cumulus. 191 apparent. Both are bulging clouds, but in these the bulging is downward, while in ordinary cumulus it is upward. Mares' tails. A popular term used to describe cirrus cloud, in which the thread-like filaments are arranged in the form vof fans or plumes. See CIRRUS. Maximum. The highest reading of an instrument during a given period. The context generally shows the period to which reference is made. An instrument, like a maximum thermometer, is often designed for the purpose of giving the highest reading that has occurred since it was last read. The term " absolute maximum " is also used, the meaning of which is generally clear from the context, but see ABSOLUTE EXTREMES. Mean. The mean value of a set of values is the number formed by adding all the individual values together and dividing the sum by the number of values. In some cases there is an ambiguity unless the context makes it clear how the values are classified. For example, the mean temperature of the atmosphere lying over a certain place might indicate the arithmetical mean of the temperatures taken at equal intervals of height, or at equal intervals of pressure, going upwards. See also AVERAGE. Meniscus. The curved upper surface of liquid in a tube. If the tube is of narrow bore the curvature is pronounced, and in estimating the height of the liquid column, allowance must be made for it. In the case of 192 Glossary. water, the meniscus, when viewed horizontally against the light, appears as a dark belt. The upper edge represents the highest point to which the water is drawn up against the glass, and the lower edge the lower part of the surface out in the middle of the tube. When the tube is broad like the measuring glass of a raingauge, the bottom edge should be used in reading ; but in narrow tubes the mid-point would be more suitable. Mercury has a convex upper surface and in the case of the baro- meter the index is adjusted to the top of the meniscus. Mercury. Mercury is a metallic element of great value in the construction of meteorological instruments. In the mercurial barometer its great density enables the length of the instrument to be made moderate, while the low pressure of its vapour at ordinary temperatures makes possible a nearly perfect vacuum in the space above the top of the barometric column. In the mercury thermo- meter there is no risk of condensation in the upper end of the stem, as in the case of the spirit thermometer. Specific gravity = 13 '5955 at 273a. Specific heat = 0'0335 at 273a, Freezing point = 234 *2a. Meteor. A meteor, or shooting star, is a fragment of solid material entering the upper regions of the atmo- sphere from outer space and visible by its own luminosity. The luminosity is attributed to incandescence due to the compression of the air in front of the meteor. (See ADIABATIC.) A larse meteor may leave a luminous trail that persists for half-an-hour or longer. Meteor. 193 Accurate determinations of the track of the meteor by reference to the constellations and of the different posi- tions of the trail, by observers in different parts of the country, may enable the height of the streak and the velocity of the lofty air currents containing it to be determined. From the results it has been conjectured that the height to which the atmosphere extends with sufficient 'density to retard the speed of meteors is 300 k. (188 miles). Meteorograph. A self-recording instrument which gives an automatic record of two or more of the ordinary meteorological elements. Of late the term has been more generally applied to the instruments that are attached to kites or small balloons and sent up to ascertain the pres- sure, temperature and humidity of the upper atmosphere. Meteorology. The science of the atmosphere. The word "meteor" from which the name is derived has now acquired a restricted meaning. It can be, and sometimes is, used for any atmospheric phenomenon. Metre. The unit of length in the metric system. 1 metre = 39 '37 inches = 3 '281 feet. MicrobarogTaph.. An instrument designed for recording small and rapid variations of atmospheric pressure. It consists of an airtight reservoir of ample size containing air, and the difference of the external atmospheric pressure and the internal pressure in the reservoir is made to leave a record on a drum driven by clockwork. The reservoir is well protected from 13204 G 194 Glossary. changes of temperature by a thick covering of felt or other non conducting material, and it is also provided with a small leak, the magnitude of which can be ad- justed. If the external pressure changes slowly the leak allows the internal pressure to follow it closely, but as the leak is small, the internal pressure cannot adjust itself rapidly to any sudden changes in the external pressure, and consequently a record of such changes is obtained. Millibar. The thousandth part of a BAR, which is the meteorological unit of atmospheric pressure on the C.G.8. system. Since the " bar " is equal to a pressure of one megadyne per square centimetre, i.e., to 1,000,000 dynes per square centimetre, a millibar is equivalent to 1,000 dynes per square centimetre.* The millibar has been in general use in the Meteorological Office since May 1st, 1914. The principal advantage of using a unit of this type is that a statement of atmospheric pressure as a certain number of millibars is perfectly definite. Accord- ing to the older practice that a separate unit had to be used for length in reading the height of the mercury in the barometer, generally the inch or the millimetre, but this * It should be explained here that a megadyne is a measure of force. 'Ihe dyne is the unit force of the C.G.S. system of units, and stands for the force which produces unit acceleration in one gramme. As the force of gravity is the mo*t familiarly known of all forces, we may say tha the force of one dyne differs but 'ittle from the weight of a milligramme, and a megadyne stands in the same relation to the we : ght of a kilogramme. The precise numerical relation is dependent upon locality, because the weight of a body, that is, the force which gravity exerts on it, depends upon latitude and the distance from the Earth's centre. At sea-level, in latitude 45, the gramme weighs '6 dynes, the kilogramme 0-9806 megadynes, Millibar. 195 length is not a measure of the atmospheric pressure until the density of the mercury, the temperature of the scale and the value of gravity at the place are allowed for. The " millibar " on the other hand can only be used for pressure. If a barometer graduated in the C.G.S. system is set up at any place, there is a definite temperature called the fiducial temperature at which the scale reading of the mercury column gives the pressure of the air in millibars ; a correction must be applied to the reading when the temperature of the instrument is not the fiducial tempera- ture. (See Observer's Handbook.) 1,000 millibars are equivalent to the pressure of a column of mercury 750*1 millimetres (29*531 inches) high at C. (273a.) in latitude 45. Millimetre. The thousandth part of a metre, 25-4 mm. = 1 inch. Minimum. The opposite of maximum. See MAXI- MUM. Mirage. The image of an object which is seen dis- placed, upwards or downwards, usually vertically, by the REFRACTION of the rays of light in their passage through layers of air of different densities near the ground. Where the density of the layers of air decreases, from the ground upwards, more rapidly than the normal rate, as it does when the ground is covered with a layer of very cold air, the rays of light are bent towards the earth and the image is therefore seen raised above the object, which may even be below the horizon at the time. L3204 196 Glossary. If the density increases rapidly upwards at the ground, as it does over highly heated deserts, the rays are bent upwards and the image is formed below the object. In its commonest form Mirage has the appearance of a sheet of water, often surrounded by banks, reeds and other objects. In this case what appears to be a sheet of water is the image of the sky behind the object at which the observer is looking, the rays of light being totally reflected from the layer of heated air which is in contact with the ground. The banks, reeds, &c., are the images of various objects, repeated with more or less distortion by being viewed through layers of air of different and varying density, so that a dark stone appears as though it were an upright stake, or plant, and so on. Hills situated at a short distance away may appear as detached masses floating on this lake-like surface, their lower portions being invisible under the conditions pre- vailing. In the same way dark stones or gravel capping a gentle roll of the ground in the desert may present the appearance of a distant vertical cliff of considerable height. Mirages may often be seen over smooth road surfaces on calm hot days in England, especially over tarred roads. They simulate pools of water on the road- way in which surrounding objects are reflected. Mist. Cloud at the level of the ground, consisting of minute drops of water suspended in the air. Mist occurs most frequently in the British Isles in the autiynn or winter, especially in still weather. A calm autumn or winter night that commences by being clear will usually become misty towards morning if the air is damp, because the nocturnal cooling lowers the temperature of the air below its dew point. At such times hill tops may have Mist. 197 clear weather, while valleys only a few hundred feet below are covered with a dense blanket of mist.* In wet weather, on the other hand, the clouds may be so low as to cover the hills and produce mist on them while the plains below experience clear weather. (SCOTCH MIST see p. 340.) Mistral. A strong dry, cold wind that is experienced on the Mediterranean coast of France. It blows from the north-west. Mock Sun. An image of the sun, sometimes very brilliant, that occurs most frequently at a distance from the sun equal to the radius of the ordinary HALO, i.e., 22. Mock Sun Ring. A colourless HALO passing through the sun parallel to the horizon, hence it is also called the Horizontal Circle. On it are situated most of the MOCK SUNS. Monsoon. The term is applied to certain winds which blow with great persistence and regularity in opposite directions at different seasons of the year. The monsoon winds are confined to tropical regions, and are most marked on the shores of India and China, where there is a south-west monsoon in the summer months and a north-east monsoon in the winter months. The term is also applied to the rainy season of India which sets in with, and is governed by the monsoon wind that blows from the south-west or west in the summer on the south and west coasts of India. Moon. The only satellite revolving round the earth. The possibility of its influencing the weather has often * Se reproduction of a photograph of Valley-fog under CLOUDS. 198 Glossary. been advanced, but never demonstrated by means of statistical evidence to the satisfaction of meteorologists. The brilliance of the moon is due solely to the sunlight falling upon it. Telescopes show a rugged and clear cut landscape very different from what would be visible if the surface were hot enough to radiate a considerable quantity of heat across a quarter of a million miles to the earth. One of the many fallacies connected with the moon is that its rays are injurious to plants, but no doubt this arose simply because nights of ground frost, harmful to vegetation, are almost always clear ; and so it happens that with the moon suitably placed, the damage is done on the occasions when it is visible and not when it is hidden by cloud. An explanation of the moon's apparent influence in scattering clouds is -given in the Quarterly Journal of the Royal Meteorological Society, vol. 28, under the title of " La lune mange les nuages," and in Shaw's Forecasting Weather, p. 175. For a note on the supposed connexion of the weather with the moon, see PHASES OF THE MOON. Nadir. See ZENITH. NepllOSCOpe. An instrument for measuring the motion of clouds. A description of the different form^ and methods in use is given in the Observers Handbook. A Camera Obscura is a very useful form of nephoscope. Nimbus. Ragged clouds of indefinite shape from which rain or snow is falling. See CLOUD. Normal. The name given to the averages of any meteorological element such as pressure, mean tempera- ture, maximum temperature, minimum temperature, duration of sunshine, velocity of wind, taken for a Normal. 199 sufficient number of cases to form a satisfactory basis of reference, and thus obtain the difference from normal which is the excess or defect of a particular example above or below the normal. Thirty-five years form a very good period for satis- factory normals, but shorter periods have to be used if the figures for 35 years are not available. The formation of a set of normals for all the stations in its region is the first duty of a National Meteorological Institute in respect of climate, and in this respect the Russian and Indian Governments have set a laudable example in the Climatological Atlases which they pub- lished almost simultaneously. In this country we have published successive editions of normals of instru- mental observations for 30 Telegraphic Reporting Stations and for about 150 Climatological Stations (Temperature and Rainfall), many of which date back for 40 years, and about 80 Sunshine Stations with records for 30 years. Monthly maps showing those Climatological normals have also been published as an appendix to the Weekly Weather Report, M.O. 214A, Appendix 4. A selection of these maps is given in The Weather Map. The non-instrumental observations, as wind, fog, snow, &c., can also be usefully summarised in the form of FREQUENCY normals but this is less often done. Observatories with self-recording instruments furnish material for an elaborate series of normals which are most effectively represented by isopleths, of which a number are given in The Weather Map, pp. 72 to 87. As an example of normals expressed in figures, we give the hourly normals for wind velocity at Kew Observatory, and the monthly normals for a number of stations in England and France. 200 Glossary. Table of Normal Hourly Velocities of the Wind in month at Kew Observatory Before noon. Hour. i 2 3 4 5 6 7 8 9 10 ii 12 Metres per second. Jan. ... 3'3 3'3 3'3 3'3 3*4 3'4 3*3 3-4 3'5 3'8 !4'2 4'3 Feb. ... 3*3 3*3 3*3 3'3 3*3 3'3 3*3 3'4 3'8 4*i 47 4'9 March .. 3'i 3*i 3-0 3*i 3'i 3'i, 3'3 3'6 4*3 47 5'i 5' 2 April ... 2"? j 2"] 2'5 2'6 \2'6 2'8 3*3 3'8 4*3 4*7 5'o SV2 May ... 2'3 2'3 2'2 2'2 2'2 2'6 3-2 3-6 4-0 4'3 47 47 June ... 2'1 2'0 . 2'0 I'9 2'I 2-5 3-0 3'3 3-6 3'8 4-2 4-2 July ... 1-9 r8 r8 1-8 r8 2'2 2-6 3-0 j 3-4 37 ; 3'9 4-0 August. 2'0 1-9 i'9 1-9 I'9 2'I 2-5 3'i 3'5 3*8 ,4-1 4-2 Sept. ... rS r8 1-9 1-9 r8 I'9 2'I 2-6 3'i 3-5 3'9 3'9 Oct. ... 2-4. 2'4 2- 4 2'4 24 2-5 2-6 27 3-2 3-6 4-2 4'3 Nov. ... 3*o 3'0 3*0 3-0 ! 3*O 2*9 2-9 3-0 3*3 3*4 4-0 4-2 Dec. ... 3-4 s 3-4 ; 3*3 3-4 ; 3-4 3-4 3-4 3*5 3-6 37 4' i 4'3 1 This table was prepared to furnish a reply to a question as to the best time of year for learning to fly on a machine of small power. The reply given Dy the figures is that, on the average, the wind is strongest from 11 a.m. to 4 p.m. in March and April, lightest from midnight to dawn in June, July, August and September. September, Normal. 201 metres per second for each hour of the day, for each (averages for 30 years 1881-1910). After noon. 24 Day 13 ! 14 15 J 16 17 . 18 19 20 21 22 23 Metres per second. 4'3 4'3 4'9 4'i | 3*8 47 l4'4 4-0 37 3'8 37 37 3-6 3-6 3 ; 4 3*4 3'4 3'4 3'4 3'3 37 Jan. Feb. 5'2 5'2 5'i 4'9 4' 5 3'9 37 3'5 3*5 3*3 3-2 3'i 3'9 March 5-2 5'2 5'2 5-r 4*8 4*3 3*4 3'3 3-0 2-9 27 3'8 April 4*8 47 47 47 4*5 4'i 3'6 3'i 2-9 2-6 2*5 2'4 3'5 May. 4'2 4'3 4'5 4'5 4*2 3'9 3'4 2-9 27 2-5 2-3 2'I 3'i June 4'* 4*2 4-1 4-1 3'9 3-6 3*2 27 2'4 2'2 2T 2'0 2-9 July. 4'3 4*3 4*3 4'2 4-0 3*6 3-0 2'6 2-5 2-3 2'2 2'I 3-0 August 4-0 4-i 3'9 37 3'4 2'8 2-5 2'4 2-3 2'2 2'0 i'9 27 Sept. 4'3 4'2, 3'9 3'5 3-i 2-9 27 2'6 2'6 2-6 2'5 2-4 3-o Oct. 4'3 4-2 3'9 3'5 3'4 3'3 3*3 3'2 3-2 3*i 3-0 3 -0 3*3 Nov. 4'3 4' 2 3'9 37 3-6 3-6 3-6 3'5 3-6 3*5 3'4 3'5 3-6 Dec. on the whole, is the best month, July the next best. March the worst. But in view of the seasonal frequency of strong winds shown in the diagrams on pp. 281 to 285, the answer does not seem complete. For such questions FREQUENCIES give better answers than NORMALS. A similar table giving the values for the top of the Eiffel Tower, Paris (300 metres high), will be found on p. 287. 202 Glossary. aq^ joj A^IO "1 9 A UIB8 I\[ C^ M O CO ^ co O M o oo vn xn co xn O oo ^ m co co CO "J8qin809(j xn rt- *^- 5 Cs C^ vn O CO C^ 'Tj- xr> co CO "rh uaqraaAO \^ Tj- O HH co vb xn y- p r^ vo ON 00 o 0> rf- xn w <N CO SKKXPO W vb : ^t- p oo co "O O W ^ a> ^ "^~ xn co M w jaqina^dag r~ r*" P in oo oo rh in ci xn wnSny r- c<j O O oo xn g^ oo co in w s* g co m TJ- vO w Ar IJj&lJS; ^ P it oo CO CO O vr; O c <O O "^ N t^ ^d" xn co CO CO vp CO co xn rf vo w AH r 1 " r* r co vb ^j~ vb S co t-> "O vo OO TJ- xn co CO CO OJO CO jTidy r^ -^- oo co vb rh vb w 00 Tj- in rj- CO c 'qoj'Bj^ OO P M co ^o \n -^- xn S 2> vb 5- xn CO t xwJuqe* i^ m -ri- co b */"> 1C I? op HH oo xn CO "5- ^CjBnur?j w n p CM O xn b co Cl 5 co vb xn i-* f-1 Averages for o o o 000 Q\ o^ Q> t-- OO a a O^i t^ CO O O O O O oo 00 8 1 1 ! OO 00 00 OO 00 00 i i xn o^i O 00 00 00 i c^ J, O^ O^ O^> oo 00 00 OO OO 1 oo 1 oo oo : s v , g : ~ -i ^S^ fe fn *! x o t>o ** a FJ 1 1 pL| f_^ <5 J 1 M ^.3 49 i 4 : fl g ' ^ | 1 1 M Q F^ c3 3 ^ 2- PQ *= o ^ ^fl^g CJ 8D 33 v^J I | 1-5 PM Observer. 203 Observer, in meteorology, is a person who undertakes, in co-operation with others at a reseau of stations, to make regular simultaneous records of the weather upon an organised plan. Good observing requires punctuality and accuracy, and therefore skill, in reading and setting in- struments, and intelligence in noting occurrences which are worth recording, though they are not in the prescribed routine. The best observer is one who is personally interested in scientific work on a co-operative basis. The work of the observers for the Weather Map is partly represented by the four maps of the gale of December 27-28, 1915, which face Weather Map, p. 88, and which show also the work of the compiler and map-maker at the Central Office. The work of the telegraphist is also necessary, but if that is perfect it makes no show at all on the maps. If it is not perfect the map at once bears evidence of the imperfection. Ombrometer. Another name for RAINGAUGB. Orientation from Or lens (Lat.), the rising of the sun the East. The direction of an object referred to the points of the compass. The meaning is much the same as the meaning of ' azimuth ' except that the azimuth is more particularly referred to the meridian line. The exact orientation of a position is generally best made out by the aid of an ordnance-map, identifying the position and the bearings of some prominent landmarks visible from it. Churches in England are usually oriented East and West, but the orientation is not always very exact. Orographic Rain is produced by the forced ascend- ing currents due to mountains. A horizontal air current striking a mountain slope is deflected upwards, and the 204 Glossary. consequent DYNAMICAL COOLING produces rain if the air contains much aqueous vapour. The dynamics of the process is not altogether free from difficulty, as the lifting of the air requires a certain amount of energy. If the mountains extended to the boundary of the troposphere the air would presumably go round, and not over the moun- tain. On the analogy of flowing water we might expect the air to go round any mountainous obstacle instead of over it. It does sometimes, but not always. Ozone is an allotropic form of oxygen for which the chemical symbol is 3 . It is produced by passing elec- trical sparks through oxygen, or by the action of cathode or ultra-violet rays. There is generally at least a trace of it in pure atmospheric air. The quantity is usually esti- mated by the depth of the colour of so-called ozone papers exposed for a given time. Some observers have described a powerful influence exerted by ozone in increasing the transfer of electricity between the atmosphere and the ground. Pampero. A name given in the Argentine and Uruguay to a severe storm of wind, with rain, thunder and lightning. It is a LINE SQUALL, with the typical arched cloud along its front. It heralds a cool South-Westerly wind in the rear of a DEPRESSION ; there is a great drop of temperature as the storm passes. Parantlielion. A mock sun (see HALO) appearing on the MOCK SUN RING at about 60 from the ANTHELION. Paraselenae Mock moons, i.e., images of the moon, occurring most often at certain points on the ordinary halo of 22 radius. Like parhelia or mock suns they are probably formed by the reflection of light from the surfaces of the snow-crystals in cirro-nebula. 20f> Parhelia or mock suns. Images of the sun occurring in connexion with solar halos. See MOCK SUN. Pentad a period of five days. Five-day means are used in meteorological work, as five days form an exact sub-division ( T V rc O of the ordinary year, an advantage not possessed by the week. Periodical. Recurring at regular intervals. Periodi- cal variations of meteorological elements generally have a period of one day, or one year, corresponding with the rotation of the Earth and its annual progress round the sun. Many attempts have been made to identify the variations of rainfall and other meteorological elements with a period .of years. Thus the period of the frequency of sunspots, about 11 years, has been regarded with some favour, as a meteorological period. A period of 19 years has been suggested with regard to the climatic elements of Australia and the wandering of the anticyclones of the southern tropical belt. Thirty-five years make up the period which seems to fit in best with the variations of climate suggested by Briickner in his examination of the records of rainfall, lake levels, floods, droughts, and other experiences going back in time as far as possible. A period of three or four years is indicated for a recog- nisable oscillation in barometric pressure. The fluctuations of the yield of the wheat harvest between 1885 and 1905 wero shown in the Meteorological Office to be represented with curious fidelity by a combi- nation of oscillations with a common point of mean value, so that the crops appeared to repeat themselves numeri- cally after 11 years. Professor Turner has suggested that these periods may possibly be fractional periods of the period of revolution of the swarm of leonid meteors, which is about 33^ years. 206 Glossary. Up to now there has not been sufficient material for a proper critical examination of these various suggestions. Moreover it is probable that the question cannot be dealt with separately for a small part of the earth's surface when the primary causes are external to the earth's atmosphere. Some method must be found for determining the change in the atmosphere as a whole, and then perhaps some particular feature like the trade winds may be found to form a sort of " pulse " of the circulation and thus act as an indicator of the atmospheric changes. If so, the process of examining for periodic changes will be so much simplified as to offer a very promising field of meteoro- logical inquiry. Persistence, a term used by Hon. R. Abercromby, the repetition of meteorological conditions or what may be called moods of weather. For several months the distri- bution of pressure may be of the same general type, with temporary interruptions. In N.W. Europe cyclones generally arrive from the Atlantic and pass eastwards, alternating with relatively high pressure in connexion with a persistent anticyclone to the south. If the path along which the centres of the lows travel remains the same, the alternation of cyclonic and anticyclonic weather persists. If, on the other hand, the Scandinavian high pressure develops in area and .intensity, the path of depressions may be to the southward of our islands, and easterly weather becomes persistent. On the other hand, the development of high pressure over Greenland, extending southward over the Atlantic, gives a northerly type of weather which is sometimes persistent for a season. Persistent rain. For some reason which cannot be explicitly stated rain generally lasts for only a few hours ; Persistent rain. 207 looking over the published results for the observatories of the Meteorological Office for the year 1912 we find the longest sequences of hours with rain are as follows : Longest Periods of consecutive hours of rainfall 1912. Valencia. Kew. Eskdalemuir. Number of hours. Number of hours. Number of hours. January February March April ..; May . 22 . 19 15 16 T-2 9 6 5 6 24 ii 16' 6 Q June ... IQ 9 II July 12 y IO August September October November December 14 28 9 8 14 9 15 8 6 13 19 29 II 19 At Kew Observatory in the past ten years rain extending over 25 consecutive hours was recorded three times, viz., 1906, November 7th-8th ; 1914, March 8th-9th, and 1915, May 13th-14th. The table does not fully represent all that the inspec- tion of the published figures suggests, for example, in August, at Valencia, only one fine hour inttrvened between two spells of eleven and twelve hours respec- tively, and at Valencia, too, though April had a longer spell of rain than March, there were 228 rain-hours in 208 Glossary. March as compared with April's 47. At Eskdalemuir, a moorland station in Dumfriesshire, in the last three days of March only 28 hours out of the 72 were free from rain. A twelve-hour rain is exceptional at an inland station like Kew, on the eastern side of Britain, but it does occur sometimes. In June, 1903, there was a run of thirty hours, separated by only one rainless hour from a preceding run of twenty hours, and that again by two rainless hours from another run of seven hours, so that there was nearly continuous rain for sixty hours at a stretch. In the same month Valencia had 110 spell of more than eleven con- secutive hours of rainfall, though two separate fair hours broke up a spell of twenty-five hours' rain. The persist- ence of rain in particular localities on special occasions in a subject of great interest which is not at all understood. In it lies the explanation of local floods which are some- times of the most extensive character, such as those of East Anglia in August, 1912, and in Eastern Ireland in August, 1905. Snowstorms are often similarly localised. In a similar way there is considerable difference in the succession of days of rain. Sometimes it rains with very little meteorological provocation, and, on the other hand, the meteorological conditions which are recognised as favourable for rain are sometimes productive of very little. The longest spell of consecutive days of rain at Kew in the 45 years 1871 1915 was one of 36 days which occurred in 1892 from September 16th to October 21st. Two spells in 1891 were nearly as long, viz., one of 31 days from September 26th to October 26th, and one of 34 days from November 7th to December 10th. See DURATION OP RAINFALL p. 303. Phases of the Moon : The appearances of the moon by custom restricted to the particular phases of new Phases of the Moon. 209 moon, when nothing is visible, first quarter, when a semicircle is visible, with the bow on the West, full moon, when a full circle is visible, and last quarter, when a semicircle is visible with the bow on the East. These changes of phase are due to the fact that the moon in its monthly course round the earth, is at one time between us and the sun, while at another time we are between it and the sun. In the first case the half of the moon illuminated by the sun is turned directly away from us, and it is the period of new moon, in the second it is directed towards us, and the moon is full. At first quarter half the side of the moon facing us is lit up, and the remaining part will gradually become so ; at last quarter the appearance is similar, but the bright portion is diminishing. It is a common practice of immemorial antiquity to associate changes in the weather with the phases of the moon ; but it must be remembered that, before the days of the cheap press and the daily newspaper, the phases of the moon were the shepherd's calendar, and the only means of marking intervals of time greater than a day and less than a year. There are about twelve and a half lunar months in a year, and the adjustment of the time-keeping of the moon to the daily and annual periods of the earth is a scientific question of great complexity and long history. For conditions of weather that are of too long duration to be associated with a day, and too short to fill a year, the phases of the moon afford the only natural method of time-keeping, so the primary classification of such events as spells of weather must necessarily count in phases, or weeks. Although we have now progressed beyond that stage and can use decimals in our reckoning, we are not 216 Glossary. altogether free from ancestral habits. If the sun happens to be shining we are accustomed to refer to the occurrence as '*a fine day." It would be regarded as pedantic to speak of a fine hour or a fine minute, but the fine day is sometimes over in an hour, and is replaced by a wet day. A well-known meteorological authority asserts, from his experience as a sailor, that weather changes set in with the "turn of the tide." It is possible that there may be also in this case some association with a single epoch of events which are really distributed over some hours, but no statistical inquiry has been made into the matter. Phenology. The study of the sequence of seasonal changes in nature. All natural phenomena are included, seed-times, harvests, flowering, ripening, migration, and so on, but often in practice the observations are limited to the times on which certain trees and flowering plants come into leaf and flower each year, and to the dates of the first and last appearance of birds and insects. A phenological report is published each year by the Royal Meteorological Society. Pilot balloon. A small free balloon, the motion of which gives information concerning the wind currents aloft. Toy balloons, having a diameter of about J8 inches when inflated, are often used, but a rather larger size is preferable ; they are filled with hydrogen and released, and their progress measured by means of a specially designed theodolite. If two theodolites are used at some distance apart the trajectory of the balloon can be com- pletely determined, but one theodolite may be used if the rate of ascent of the balloon is known, assuming this to be uniform. Pilot balloons have shown that the wind, Pilot-balloon. 211 at a height of 1,500 feet, is usually in close agreement with the GRADIENT WIND. An East wind is often shallow, and there is a REVERSAL of wind direction on many occasions, the upper wind being Westerly ; on some occasions, however, the East wind is maintained up to 'great heights, 'though it seldom increases in velocity above 3,000 feet, at which height an East wind is usually at its maximum. Winds from other than Easterly direc- tions may increase up to 30,000 feet or so, but still higher, when the STRATOSPHERE is reached, there is a decrease in velocity. In the first 3,000 feet there is usually a VEERING of the wind with height, whatever the direc- tion of the wind. Sometimes great changes in direction occur at various heights ; these are usually veerings. When this is the case the velocity falls off near the level of the change, and when the direction is reversed there is generally a region of calm between the opposite cur- rents. During the approach of depressions from the West a Southerly surface-wind changes to a Westerly wind in the upper air. In anticyclonic weather there is sometimes very little wind up to the greatest heights reached, and what little there is varies in direction from one level to another. Ordinary pilot balloons are some- times followed to heights of four or five miles ; to reach greater heights larger balloons must be used, such as those used for sending up recording instruments (BALLONS- SONDES). Instructions for observations with pilot balloons and for calculating the results are given in the Computer's Handbook. Pluviograpll. A self-recording rain-gauge ; the rise of the water in the gauge is recorded by means of a pen 212 Glossary. attached to a float. Some form of device by which the gauge automatically empties itself when the water reaches a certain height is often employed. Pluviometer. -A rain-gauge (q.v.). Pocky Cloud. Cumulus cloud, .with a festooned appearance on the under side. See MAMMATO-CuMULtrs. Polar. Occurring in the regions of the north and south pole. The meteorology of these regions is of a special type caused by the continuous presence of the sun above the horizon during a long period of the year, and its absence during an equal period. This makes the diurnal range of temperature small. The seasonal change on the other hand is large, extremely low temperatures during the long night being followed by less severe cold in the time of continuous daylight, when the thermo- meter often rises above the freezing point. The snowfall has always been found to be moderate ; the great quantity on the ground represents the accumulated fall of many years because there is no loss except by evaporation and that is very slight at the very low temperature. Aurorae, as well as optical effects due to the presence of ice crystals in the air, are of common occurrence. In the south polar- regions a spot has been found where the normal velocity of the wind is beyond the limit of gale. Both poles are separated from the equatorial regions by a great circum- polar whirl of prevailing westerly winds, and regions of relatively low pressure. These features are especially marked in the southern hemisphere. Pole. The geographical poles lie at the extremities of the axis of rotation of the earth. The magnetic poles are at some considerable distance from the geographical poles. Potential. 2i;> Potential as applied to energy indicates the energy which is due to the position of a body. In considering the total amount of energy available, in any case we must consider not only the position but the quantity of work- ing substance that is collected there. If we wish to consider the influence of the position alone we must limit our ideas to a particular amount of the working substance. We naturally choose the unit measure as the amount for this purpose, and the potential energy of unit quantity is called the potential at the point. Thus, the electrical potential at any point in the atmosphere is the amount of energy which one unit of electricity possesses in virtue of its position at the point. Similarly, the gravita- tional potential or geo-potential at any point above the earth's surface is the potential energy of a unit quantity of material, a gramme or a pound, placed there. Potential temperature. The temperature which a specimen of air would acquire if it were brought down from the position to mean sea-level under ADIABATIC conditions. Precipitation. See p. 329. Pressure. Force per unit of area exerted against a surface by the liquid or gas in contact with it. The pressure of the atmosphere, which is measured by means of the barometer, is produced by the weight of the over- lying air. The pressure exerted by the wind is -generally very small in comparison. That due to a wind of force 6 is approximately one-thousandth part of the pressure of the atmosphere. Prevailing: Winds. When a station experiences wind more often from a certain direction than from others, that wind is termed the prevailing wind, The ,'214 Glossary. best example is the trade wind, which blows from the N.E. at many places between the equator and 30N. Lat. with great regularity, and from the S.E. in the correspond- ing belt south of the equator. In latitudes 40 to 60, north or south of the equator, westerly winds are very common and form circumpolar whirls. These are best developed in the southern hemisphere, where there is less land. For England and the neighbouring portion of Europe the best guide to the prevailing winds in different parts is the average distribution of pressure shown in the maps of Mean Pressure in the Monthly Weather Report and its Annual Summary. The prevailing direction is between S.W. and W. : there is more southing in the western districts than in the eastern. In monsoonal regions the prevailing winds are in opposite directions at different seasons, and in others there may not be a favoured direction. In all cases a long series of observa- tions is required to make sure of the normal conditions. Probability. When the occurrence of an event is apparently doubtful, but under similar circumstances has happened before, more often than not, we speak of it as " probable." Mathematically, the probability is repre- sented by the fraction obtained by dividing the number of times that the event has happened by the total number of times that the circumstances have arisen. It is a quantity that must lie between and 1, and is usually denoted by the letter p. It should be clear from the definition that the probability of the event failing to happen, when added to the probability that it will happen, must make 1. For example, we may consider the prob- ability that to-morrow will be fine if to-day is wet. Out of a number N of occasions of wet days we count the Probability. 215 number n when the following day was fine, and the probability of a fine day to-morrow is njN. It may be called the random probability because the occasions are simply chosen at random without any guiding principle. It is, of course, necessary to deal with a large number of observations before a reliable value for the probability is obtained. It will -be noticed that in this sense probability is directly determined by the frequency of occurrence, so that the facts represented in tables of frequency can be equally well represented by tables of probability. When we have no guide as to the expectation of an event except the number of times that a similar event has occurred previously, and we express the probability as a fraction 1/n with 1 as numerator, we may say that the random chance of the occurrence is one in n or that the odds against the occurrence are n-\ to one. Prognostics. Signs of coming weather. Some of them are dealt with under SHEPHERD OF BANBURY and WEATHER MAXIMS. There is a widespread belief that certain animals are in some way aware of the approach of wet weather, and behave in some special manner in consequence, but this seems unlikely. There is no reason for supposing that they can feel anything beyond the changes of temperature, moisture, wind, &c., in the air surrounding them, and changes in these are followed by a great variety of weather. Perhaps the most valuable instrument for pi ognostication is the barometer. Regions of high and low pressure have fairly definite weather- associated with them, and mostly move across our islands from the west or south-wes"t. A southerly wind, there- fore, with a pronounced fall of the barometer, as an 216 Glossary. example, is a fairly reliable indication of an advancing depression, and therefore of rain. A gradual dimming of the clear blue of the sky, and the formation of a thickening sheet of high cloud, which often forms halos round the sun and moon, shows that the stormy area is not far off. By the application of BUYS BALLOT'S LAW (q.v.), it can be seen that the wind will generally veer towards west if the low pressure is going to pass on the north side of the obseryer, but will back if it is going to cross on the south side. In fine weather the wind often drops at night on the ground, while continuing to blow a few hundred feet up. It follows, therefore, that brisk motion of the lower clouds on a still sunny morning indicates a wind which may be expected down below during the middle of the day. PsychrOHieter, the name given to the dry and wet bulbs as forming an instrument for measuring coolness ; the combination of a thermometer having its bulb coated with wet muslin and an ordinary thermometer used for estimating the dampness of the air, by observing the difference between the readings of the tw 7 o thermometers. In dry air evaporation takes place freely and cools the wet bulb. The cooling for a given temperature of the dry bulb depends principally upon the dryness of the air, or its absorbing power for moisture. In the aspiration psychrometer a fan is used so as to produce a draught of definite speed past the instrument. Pumping*. Unsteadiness of the mercury in the barometer caused by fluctuations of the air pressure produced by a gusty wind, or due to the oscillation of a ship. Purple Light. 217 Purple Light. A parabolic glow of colour varying from pink to violet appearing vertex upwards in the western sky at a considerable elevation above the point of sunset after the sun has passed below the horizon. It is a DIFFRACTION glow similar to the white glow round the sun during the day, which is caused by the interference of light scattered by particles of many sizes ; as the sun sets, its light reaches only the more uniform particles of the upper atmosphere, and the coloration becomes purer, culminating in the coloration of the margin. In very clear weather a second, fainter purple light may follow the first. (See also BLUE OF THE SKY and TWILIGHT). PyrheliO meter. An instrument for measuring the radiant heat received from the sun. In the form of instrument devised by Angstrom there are two metal strips, one of which receives the solar heat, while the other is warmed by means of an electric current. The current required to give equal heating of the two strips depends upon the intensity of the sun's rays, and when measured gives the amount of heat received. Radiation. See p. 330. Rain is produced by the condensation of the aqueous vapour in the atmosphere. Each cubic foot, or cubic metre, of air is capable of holding a certain definite amount of water in the form of vapour ; the amount depends greatly . upon the temperature, being large when the temperature is high, and small when it is low. The ^ater vapour is mixed with the air in varying proportions, and when the temperature of the mixture falls sufficiently a point is reached where the vapour is condensed into fine particles 218 Glossary. of water, and a cloud is formed. As the cooling continues more water is condensed to form larger drops which fall as rain. The cooling which produces rain is probably dynamical. See ADIABATIC and PERSISTENT RAIN : also DURATION OF RAINFALL, p. 303, and RAINDROPS, p. 334. Rainband. A dark band in the solar spectrum on the red side of the Sodium D lines, due to absorption by water vapour in the Earth's atmosphere. It may be best seen when the spectroscope is pointed at the sky rather than directly at the sun. The band is strengthened with increase of water-vapour, and also when the altitude of the sun is low, and his light has to shine through a greater thickness of air. It is of doubtful value as a PROGNOSTIC of rain. Rainbow. A rainbow is seen when the sun shines upon raindrops, or indeed upon spherical drops of water produced by a waterfall or by any other means. The drops may be at any distance from the observer, but the centre will always be exactly opposite the sun, and the angular diameter of the circle for each colour is invariable. When sunlight falls upon a drop of water it is reflected and refracted in all directions, but there are certain directions in which the light is much more intense thail in others. An observer therefore looking at the drops in general will see some of them much better than others, and those drops which show up will lie in that particular direction in which they reflect the greatest amount of light. But the particular direction is different for each colour, and hence the rainbow consists of a series of rings of different colours. A similar explanation applies to the secondary bows. In the primary bow the red is outside. A rainbow is usually circular, the head of the observer Rainbow. 219 being the centre of the circular arch, but a horizontal rain- bow can be seen when the sun's rays from behind the observer fall on drops of dew on the grass, or gossamer threads of a meadow. In that case the bow is not circular. Rain-day. A day on which more than a certain specified amount of rain has fallen. It has been usual in the past to measure rainfall in hundredths of an inch, and 01 inch has been the specified quantity, a day on which 005 inch, or more, fell, counting as a rain-day. Rainfall water which falls from the atmosphere. The term is very commonly taken to include snow and hail. Precipitation is the proper inclusive term. The measurement of a definite amount of rain, say, fifteen millimetres, 15 mm., means that if all the water had remained where it fell and not soaked in or run off, the depth of water on the ground would be 15 mm. An inch of rain is equivalent to 101 tons per acre ; a millimetre to a kilogramme per square metre, Or one thousand metric tons per square kilometre. Raingauge. An instrument for measuring the rain- fall. All the rain which falls on a definite area, generally a circle of either five or eight inches in diameter, is collected into a glass vessel, which is graduated to give the amount of rain. Rain-spell* According to the definition of the British Rainfall Organization, a rain-spell is a period of more than fourteen consecutive days, every one of which is a rain-day. On a general average, one or two such periods fall to the lot of most stations in the British Isles within the year. Reaumur Rene Antoine Ferchault de, d. 1757, 220 Glossary. whose name is given to a scale of temperature now almost obsolete. On it the freezing point of water is zero, and the boiling point 80. Reduction, as applied to meteorological observations, generally means the substitution for the values directly observed of others which are computed therefrom and which place the results upon a comparable basis. Thus reduction to sea-level in the case of barometer readings, means estimation according to certain rules of the value which the pressure would have at a fixed level lower than that of the place of observation, and the reduction of a set of mean values extending over a regular series of years to a uniform or normal period indicates a similar procedure based upon comparison with neighbouring stations. Reduction to Sea Level. Both temperature and pressure are " reduced to sea level " before they are plotted on charts. To reduce mean temperature to sea level 1 F. is added for each 300 feet in the elevation of the station ; la for 165 metres ; other rates are used for maximum temperature and minimum temperature (see Computer } s Handbook, Introduction, p. 1 L). This reduc- tion is regarded as necessary in forming maps of ISOTHERMS of regions with a considerable range of level, otherwise the isotherms simply reproduce the contours ; but it reduces the practical utility of the maps because the addition of ten or twelve degrees to the temperature actually observed gives an entirely false idea of the actual state of things in the locality represented. The same objection does not apply to the reduction of pressure to sea level because the human organism has To face p. 221. DIAGRAM for obtaining HEIGHT DIFFERENCFS Iron pRtssuwr- PifTEKtNCES lor different TEMPERATURE ol tt,* an -column from tbe formula H H. =. 0674 7 log* ? where H is the height m kilometres r is the temperature on the absolute scale, and p is the pressure measured in am units, but pre- ferably in millibars H. and f>, aie corresponding values of H and /> at ny definite height, usually ground level The process ol um the diagram u z> follow* - C 1 Tbere are two pcotncton lot temperature ttkmebe o lie^ht u reprewntod by cm (X o 1Z-S cat (ft of tbe luge d For urotractoi A I kilomrtrr is represented by 2 large division: <ENCEb for protractor B I kilometre is represented by 6 large divisions Reduction to Sea Level. 221 no such separate perception of pressure as it has of temperature. The reduction of pressure to sea level is carried out in accordance with the general rule for the relation of difference of pressure to difference of height. This goes according to the equation h - h, = KT (Iog 10 Po - Iog 10 p) where //, }>, h , p Q are corresponding values of height and pressure, T is the absolute temperature, arid k, a constant which is numerically equal to 67*4 when the height is to be given in metres, or to 221*1 when the height is to be given in feet. This equation is derived from the direct expression of the relation of pressure and height g p dh = dp. The best way of working the equation is to use what is called semi-logarithmic paper, that is squared paper which is ruled in one direction in equidistant lines representing equal steps of height, and in the other direction according to the logarithm of the numbers indicated on it, like the graduation of a slide-rule. The relation between height and pressure for any one temperature is represented by a straight line that has a slope that can be calculated when k is known. The regular course is to find the proper point for the height and pressure of the starting point, and travel along the line of proper slope for the tempera- ture so long as that temperature can be accepted, say for half a kilometre. When the half kilometre is reached adjust the slope for the mean temperature of the next half kilometre, and so on until the observed difference of pressure has been traversed. If the line reaches the boundary of the ruled paper, the vertical line or the 222 Glossary. horizontal line as the case may be, begin again on the other side of the ruling. It will be seen that to make an accurate determination of the height the temperature of each stage must be known. If there are no actual measurements for the occasion an approximation may be made by using mean values either for the initial temperature or the lapse of temperature with height. To simplify the process of obtaining the height differ- ences from pressure differences semi-logarithmic paper is provided at the Meteorological Office ruled with the slope lines for given values of the temperature, so that with a parallel ruler the composite line for any particular determination can be easily drawn. A reduced copy of the form is shown in the figure. The humidity of the air makes very little difference to the computation of height in our latitudes where temperature and moisture do not reach tropical figures. The best way for allowing approximately for humidity, which diminishes the density under standard conditions, is to regard the temperature as increased by one tenth of a degree for each millibar of water- vapour-pressure in the atmospheie. Refraction. The name applied to the bending to which rays of light are subjected in passing from one medium to another of different optical density. It plays an important part in many optical phenomena in the atmosphere ; MIRAGE, HALOS, and RAINBOWS are refraction pheno- mena, the colours of the two latter being due to the fact that rays of different colours suffer a different amount of bending. Another refraction effect is that the apparent ALTITUDE of a heavenly body is greater than its real Refraction. 223 altitude because the rays of light entering the atmosphere are passing from a less dense to a more dense medium, and their final direction is nearer the vertical than their original direction. Registering balloon. A small free balloon carrying with it a light meteorograph and sent up to ascertain ihe temperature, humidity, &c., of the air. See BALLON SONDB. Regression Equation. See p. 339. Relative Humidity. All natural air, unless it is artificially dried, contains more or less water in the form of vapour. For each temperature there is a fixed and definite limit to the amount of water in a definite volume of air, such as a cubic foot or a cubic metre. Air which contains this full amount is called saturated air. The actual amount that can be present in a given volume depends on the tem- perature, and increases rapidly as the temperature rises. The relative humidity is the ratio of the amount that is present to the maximum amount that could possibly be present. This ratio is expressed as a percentage, so that saturated air, at whatsoever temperature it may be, always has a relative humidity of 100. Thus a relative humidity of, say, 75 means that a certain volume of air is holding in the form of vapour 75 grammes or ounces of water, whereas it is capable of holding 100 grammes or ounces. When saturated air is cooled by any means it ceases to be able to hold all the water in an invisible form, and fine water drops appear forming a fog or cloud. From an analysis of upwards of 100,000 hourly readings at the observatories of the Meteorological Office during the three years 1907, 1908, 1909, it appears that on the average out of one thousand observations at every hour of Glossary. the day or night throughout the year the frequencies of specified values of relative humidity are as follows : TABLE OP FREQUENCIES OP OCCURRENCE OF SPECIFIED VALUES OF RELATIVE HUMIDITY REFERRED TO A TOTAL OF ONE THOUSAND HOURLY OBSERVATIONS. Relative Humidity. Aberdeen. Valencia. Falmoutk. Kew. Frequency. IOO ... i 14 0'2 I 95 to 99 45 186 104 76 90 to 94 ... 150 191 213 171 80 to 89 378 335 3^8 3*1 70 to 79 270 217 ^34 206 60 to 69 ... 119 5i I O2 135 50*059 33 5 18 71 40 to 49 4 i I 17 30*039 O'2 o O 2J Total IOOO IOOO IOOO IOOO See also ABSOLUTE HUMIDITY, p. 290. Reversal. A large change (more than 90) in direction between the surface current and the wind in the upper air. Reversals are most common with Easterly surface winds, and least common with Westerly. A reversal may take place quite close to the ground, or anywhere up to 15,000 feet or more. The most permanent case of a reversal is over the TRADE WIND, where the North-Easterly surface current is replaced by a South- Westerly current in the upper air. Reversal. 225 When an eruption of the Sou Mere in St. Vincent takes place, the dust, carried by the upper current, falls in Barbados, though it lies 100 miles to windward of St. Vincent (see PILOT BALLOON). Ridge. An extension of a " high " area shown on a weather chart, corresponding to the ridge running out- wards from a mountain system. It is the opposite of a trough of low pressure. Rime. Ice crystals, like small needles, which form on trees and buildings in foggy, frosty weather. The needles point to the direction of the wind, and, in favourable situations, the summit of Ben Nevis for example, may accumulate and form large and heavy masses of ice. River. Geographically a river is simply the flow of water from the higher levels of the land to lower levels and is thus, in meteorology, only part of the great circu- lation of water through evaporation and condensation ; but, from the point of view of climate, the variations of river-level are interesting and important as they represent the result of meteorological causes operating over a large region. The seasonal variation is often different from what might be expected, for example, the River Thames is at its highest in February, four months after the normal period of greatest rainfall. The great historic example of seasonal variation of river-flow is that of the Nile, upon which the fertility of lower Egypt depends; its annual rise begins at Assuan in June and reaches its maximum in the beginning of October. The Tigris and Euphrates, like- most Continental rivers, show their rise in the spring with the melting of the snow in the regions of the head waters. 13204 H See CLIMATIC SUMMARIES appended to The Weather Map. Roaring Forties. The belt between latitudes 40 and 50 South latitude,' characterised by prevailing boisterous Westerly winds. St. Elmo's Fire. Brush-like discharges of electricity sometimes seen on the masts and yards of ships at sea during stormy weather ; it is also seen on mountains on projecting objects. It may be imitated by bringing a sharp pointed object, such as a needle, near a charged Leyden jar. Saturation. When applied to the air this term indicates that all the moisture possible as water vapour at that temperature is present. A reduction of temperature would lead to the condensation of some of it to liquid drops, while a rise of temperature would make the air " dry " and enable it to take up more. Screen. In order to allow thermometers to indicate as nearly as possible the temperature of the air, and therefore to avoid the disturbing effect of the sun's rays and of neighbouring objects, louvred screens are used which allow of a free circulation of the air. The pattern used at the Meteorological Office observatories is a modi- fication of that designed by Thomas Stevenson, C.E., one of the founders of the Scottish Meteorological Society. It is known in this country as the Stevenson screen, on the continent as the English screen. Scud. A word used by sailors to describe small fragments of cloud that drift along underneath nimbus clouds. The meteorological term is fracto-nimbus (Fr. Nb.). See CLOUDS. Scua. 227 In mountainous districts, after the passage of a depres- sion, nimbus often breaks up into scud, which may persist with sunny weather for many hours. Sea-breeze. A breeze that blows from the sea during the day in fine weather and drops at night (see BREEZE). In bright weather the warmed air over the land-surface rises, and there is an inrush of the cooler air from the sea to take its place ; at night, when the temperatures are more or less equalised, the wind dies away. Sea-tevel. The level surface which the sea would have if the waves were smoothed out. Mean sea level (M.S.L.) is the mean position occupied by this surface during the whole year. In England M.S.L. is an arbitrary level at Liverpool. See LEVEL. Seasons. In meteorology the seasons are taken to be as follows : 1. Spring ... March, April, May. 2. Summer ... June, July, August. 3. Autumn ... September, October, November. 4. Winter ... December, January, February. If an element is described as having simply a seasonal variation, it implies that it goes through its changes in a period of one year. The selection of months to represent the seasons accord- ing to the farmer's year is guided by the consideration that each season shall comprise three months. The uniformity in length opens the way for some paradoxical cases. The warmest week of summer may be in the spring, late May, or in autumn, early September, and the coldest week of winter may be in the autumn, late November, or spring, early March. From the point of view of weather, we have in this country about five months of moderate winter weather between October and 13204 H 2 228 Glossary. April, and four months of summer weather from the middle of May to the middle of September, a short spring and a short autumn ; but the seasonal variations are not nearly so large here as they are in continental countries, and the change from winter to summer and vice verm is much less abrupt. NORMAL TEMPERATURES AT SEA-LEVEL OP EACH WEEK OF THE SEVERAL SEASONS FOR ENGLAND SOUTH-EAST. Week No. Winter. Spring. Summer. Autumn. a a. a. a. i 278 278 287 288 2 78 79 87 87 3 78 79 88 87 4 77 80 89 86 5 . 77 80 89 85 6 77 81 89 84 7 77 81 89 83 8 77 82 89 82 9 78 83 89 82 10 78 83 89 81 ii 77 84 89 80 12 77 '85 8q - 79 13 77 86 88 79 Mean 277 282 289 283 RAINFALL IN MILLIMETRES. mm. mm. mm. mm. England, S.E. 172 131 1 60 216 Scotland, N. 413 251 268 396 Seasons. 229 The seasonal changes for England South-East, are set out in the accompanying table of normal temperatures for each week of the year, to which the seasonal rainfall of Scotland North, as well as of England South-East, has been added. The temperatures are given only to the nearest whole degree of the absolute scale, so that minute differences are not apparent. If we allow for winter the temperatures 277, 278, 279, i.e., from 39 F. to 43 F., and for summer the temperatures 287, 288; 289, i.e., from 57 F. to 61 F., we see from the table that winter temperatures last from the 12th week of auturdii (middle of November) to the 3rd week of spring (middle of March) ; then come ten weeks of slow tran- sition, a degree each fortnight, to summer temperature in the first week of summer (beginning of June) ; the summer temperatures last until the third week of autumn, about the 21st September, then there are eight weeks of rapid transition of a degree each week until winter temper- ature is reached in the middle of November. Some interesting particulars of the temperature of the several seasons in the British Isles are given in Tempera- ture Tables of the British Isles, M.O. Publication 154 (1902). Diagrams are given for the daily temperature at four observatories : Aberdeen, Valencia, Falmouth and Richmond (Kew Observatory), and they show a lag of temperature behind the Sun very similar to that noticed in the diurnal changes of temperature which are figured in the same volume. For rainfall in the South-East of England autumn is the rainy season with 216 mm., compared with 131 mm. in the spring, while in the North of Scotland winter is the rainiest season though spring is again the driest. In considering the seasonal variations of rainfall it is 230 Glossary. important to distinguish between the day rainfall and the night rainfall. We may contrast the two from the summary of forty years' observations at Kew Observatory. Average Daily Rainfall. A. M. P.M. Midt. to 6 a.m. to Noon to 6 p.m. to 6 a.m. Noon. 6 p.m. Midt. mm. mm. mm. mm. January... 0-37 0-37 0-36 0-36 February 0-38 *35 '34 0-29 March ... 0-31 0*30 0-32 '34 April *33 0-34 - 0-39 0-30 May 0-34 '35 0*42 0-26- June ... ... ... i 0*43 Q'45 '54 0-51 July 0-39 0*42 0-67 0-48 August ... ... ... ; 0*36 September 0-49 0*42 0-37 0*59 o*43 0-42 o 46 October ... November December o-59 0-49 0*46 0*40 0-38 0-60 0-48 0*41 0-50 0-49 0-40 It will be seen that there is a maximum in October for each of the four quarters of the day, but the October maximum for the afternoon and for the evening is a sub- sidiary one. The maximum for the whole year belongs to the afternoon in July, when the figure 0*67 is reached. The reader is recommended to draw " isopleths " on this table, that is, lines of equal rainfall, *60 mm., *55 mm., 50 mm., *45 mm., *40 mm., *35 mm., and '30 mm. He will find the grouping of the rainy and of the dry parts of the day in different months of the year very suggestive. Seasons. 231 The idea of four seasons in agriculture appropriate to these islands, the winter for tilling, the spring for sowing and early growth, the summer for maturing and harvest- ing and the autumn for clearing and preparing, depends upon the peculiarity of our climate. Where the land is' ice-bound in winter or rainless in summer another dis- tribution has to be made. Between the tropics there is nothing that can properly be called summer and winter ; the seasons depend upon the weather and rainfall, and not upon the position of the sun, and the periods of growth are adjusted accordingly. In India, or the north-western part of it, the divisions of the year are the cold weather, the hot weather, and the'- rains. It is curious, for example, that the period for growing wheat in Western Australia is locally the winter period, and coincides in actual time with our own, which is a summer period. Secant. In a right angled triangle the ratio of the hypotenuse to one side is the secant of the angle between the two. See SINE. Secondary. A small area of low pressure accom- panying a larger " primary " depression. The secondary may develop into a large and deep cyclone, while the primary disappears. See ISOBARS and Plate XII. Seismograph.. An earthquake recorder, or instru- ment for automatically recording the tremors of the earth. Serein. Fine rain falling from an apparently clear sky. It happens very rarely. Sliamal. From an Arabic word originally meaning u left-hand " and thence 4k North " used to denote the North-Westerly winds of summer over the Mesopotamian plain. See The Weather Map, p. 60. 232 Glossary. Shepherd Of Banbury. The nominal author of " rules to judge the changes of the weather." The following is taken from u The Complete Weather Guide," by Joseph Taylor, 1814 : " Who the shepherd of Banbury was, we know not ; nor indeed have we any proof that the rules called his were penned by a real shepherd : both these points are, however, immaterial : their truth is their best voucher. Mr. Claridge (who published them in the year 1744) states, that they are grounded on forty years' experience, and thus, very rightly, accounts for the presumption in their favour. ' The shepherd,' he remarks, ' whose sole business it is to observe what has a reference to the flock under his care, who spends all his days, and many of his nights in the open air, under the wide-spread canopy ' of Heaven, is obliged to take particular notice of the alterations of the weather ; and when he comes to take a pleasure in making such observations, it is amazing how great a progress he makes in them, and to how great a certainty he arrives at last, by mere dint of comparing signs and events, and correcting one remark by another. Every thing, in time, becomes to him a sort of weather-gage. The sun, the moon, the stars, the clouds, the winds, the mists, the trees, the flowers, the herbs, and almost every animal with which he is acquainted, all these become, to such a person, instruments of real knowledge.' " The rules enumerated are typical of all rules based on experience of the weather ; what of truth or error there is in them the reader may judge ; they are as follows : I, SiTN. If the sun rise red and fiery Wind and rain. II. CLOUDS. If cloudy, and the clouds soon decrease Certain fair weather. III. Clouds small and round, like a dapple-grey, with a north-wind Fair weather for two or three days. IV. If small Clouds increase Much rain. V. If large Clouds decrease Fair weather. VI. In Summer or Harvest, when the ivind has been South two or three days, and it grows very hot, and you see Clouds rise with great white Tops like Towers, as if one were upon the Shepherd of Baribury. 233 Top of another, and joined together with black on the nether side There will be thunder and rain suddenly.* VII. If two such Clouds arise, one on either hand It is time to make haste to shelter. VIII. If you see a Cloud rise aga'ulst the Wind or side Wind, when that Cloud comes up to you The Wind will blow the same way that the Cloud came. And the same Rule holds of a clear Place, when all the Shy is equally thich, except one Edge. IX. MIST. If Mists rise in low Grounds, and soon vanish Fair Weather. X. If Mists rise to the Hill-tops Rain in a Day or two. XI. A general Mist before the Sun rises, near the full Moon Fair Weather. XII. If Mists in the New Moon Rain in the Old. XIII. If Mists in the Old Rain in the New Moon. XIV. RAIN. Sudden Rains never last long : but when the Air grows thick by degrees and the Sun, Moon and Stars shine dimmer and dimmer^ then it is like to rain six Hours usually. XV. If it begin to rain from the South, with a high Wind for two or three Hours, and the Wind falls, but the Rain continues, it is like to rain twelve HoiJrs or more, and does usually rain till a strong North Wind clears the Air. These long Rains seldom hold above twelve Hours, or happen above once a year. XVI. If it begins to rain an Hour or two before Sunrising, it is likely to be fair before Noon, and to continue so that day ; but if the Rain begins an Hour or two after Sunrising, it is likely to rain all that day, except the Rainbow be seen before it rains. XVII. WINDS. Observe that in eight Yeari Time there is as much South- Wext Wind as North- East, and consequently as many wet Years as dry. XVIII. When the Wind turns to North-East, and it continues two * See photograph of Cumulo-Nimbus under CLOUDS. 234 Glossary. Days without Pain, and does not turn South the third Day, nor Pain the third Day, it is likely to continue North- East for eight or nine Days, all fair ; and then to come to the South again. XIX. After a Northerly Wind for the most part of two Months or more, and then coming South, there are usually three or four fair Days at; first, and then on the fourth or fifth Day conies Rain, or else the Wind turns North again, and continues dry. XX. If it turns again out of the Smith to the North- East with Rain, and continues in the North-East two Days without Rain, and neither turns South nor rains the third Day. it is likely to coutinue North-East two or three months. XXI. If it returns to the South within a Day or two without Rain, and turns Northward with Rain, and returns to the South in one or two Days as before, two or three times together after this sort, then it is like to be in the South or bouth- West two or three Months together, as it was in the North before. The winds will finish these turns in a fortnight. XXII. Fair Weather for a Week with a Southern Wind is like to produce a great Drought, if there has been much Rain out of the South before. The H ind usually turns from the North to South with a quiet Wind without Rain; but returns to the North with a strong Wind and Rain. r lhe strongest Winds are when it turns from South to North by West. When the North Wind first clears the Air, which is usually once a Week, be sure of a fair Day or two. XXI II. SPRING AND SUMMER. If the last eighteen Days of Feb- ruary and ten Days of March be for the most part rainy, then the Spring and Summer Quarters are like to be so too ; and I never knew a great Drought but it entered in that Season. XXIV. WINTER. If the latter End of October and Beginning of November be for the most part warm and rainy, then January and February are like to be frosty and cold, except after a very dry Summer. XXV. If October and November be Snow and Frost, January and February are likely to be open and mild. Shepherd of Banbury. 235 The CORRELATION COEFFICIENTS (q.v.) for one or two- of the above rules have been worked out, but they are dis- appointing : XXIII. For 38 years, S.E. England, between rainfall of last 18 days of February and first 10 days of March, and spring 1 rain- fall, the correlation coefficient is + 0'14 ; between the rainfall for the same period and summer rainfall, + (V07. XXIV. For 64 years at Greenwich, between October-November temperature and that of the following January-February r -1- 05 ; between October-November temperature and that of the following December-January-February-March, + 0-25. None of these values indicates a connection of any significance; in the case of XXIV. the Shepherd's pro- position is negatived. Silver Thaw. An expression of American origin. After a spell of severe frost the sudden setting in of a warm damp wind may lead to the formation of ice on exposed objects, which being still at a low temperature cause the moisture to freeze upon them and give rise to a " silver thaw." Simoon. A strong, hot wind, accompanied by clouds of dust, experienced in the Sahara and the Arabian desert. It is probably due to convection movements similar to those in thunderstorms, but there are no clouds, rain or thunder and lightning ; this is probably owing to the extreme dry ness of the air over the desert. Sine. The ratio of the vertical height of a distant object to the distance of its top from the observer is the same for all objects which have their tops in the same line of sight. It is one method of specifying the angles which the objects " subtend " at the eye of the observer, which could also be specified by the ratio of the vertical 236 , Glossary. height to the horizontal distance of its foot (tangent), or by the ratio of the horizontal distance of the foot to the distance of the top from the observer (cosine). The values of these ratios are of great importance in survey- ing, and are called the trigonometrical ratios of the angles. They are formally defined as follows : Let A be a line drawn from the observer to the top of the distant object, BC the vertical, AC the horizontal. Then ABC is a right angled triangle with C as the right angle. The sine of the angle A (sin A) is the ratio BC to AB. The cosine of the angle A (cos A) is the ratio AC to AB. The tangent (tan A) is the' ratio BC to AC. The secant (sec A) is the ratio of AB to AC. The cosecant (cosec A) is the ratio AB to BC. The cotangent (cot A) is the ratio AC to BC. If the angle is greater than a right angle some of these ratios are negative, and the following convention is adopted. A positive angle is measured from AC in the direction opposite to the motion of a watch hand. The line AB is always counted positive, AC is positive if C falls on the right of A, negative if C falls on the left, and BC is positive if B falls above AC, and negative if B is below AC. Thus if the angle is between 90 and 270 AC is negative, and if between 180 and 360 BC is negative. Sine Curve. This curve is obtained by plotting, against horizontal distances representing angles, vertical ordinates representing their SINES. Its simplest equation is y = a sin a?, the more general equation y = a sin (x a) represents the same curve shifted forwards through a dis- tance corresponding to the angle a. The importance of the Sine Curve. 237 curve in Meteorology is due to the fact that it represents the simplest form of PERIODIC variation ; its shape, in fact, is that of the conventional " wave." The diurnal and annual march of temperature, for example, would, in so far as they depend only on solar altitude, each be represented by a sine curve, and are so represented in theoretical work. Any periodic variation, however complex, can be represented by a number of sine curves superposed. The process of finding a set of sine curves to represent a given variation is called HARMONIC ANALYSIS, p. 145 and p. 311 (#.^.). SirOGCO. A name used on the Northern shores of the Mediterranean indiscriminately for any warm Southerly wind, whether dry or moist. Such winds blow in front of depressions advancing Eastward. The typical Sirocco however is hot and very dry, and is probably in many places a F6HN wind. Sleet. Precipitation of rain and snow together. See p. 341. Snow. Precipitation in the form of feathery ice crystals. See p. 342. Snow Crystals. Thin flat ice-crystals of a hexagonal form. There are many patterns, and when snowflakes of various size are observed on any one occasion they differ only by containing a larger or smaller number of these crystals. Solar " Constant." The amount of radiant energy which would be received in one second from every square centimetre of cross-section of a beam of solar radiation if it had undergone no absorption in the atmosphere. Recent investigations indicate that the solar "constant" is not invariable. The mean value is about 135 milliwatts per square centimetre. 238 Glossary. Solar Day. See EQUATION OF TIME. Solarisation. Exposure to direct sunlight; the same as INSOLATION (q.v.). See also RADIATION, p, 330. Solar Radiation Thermometer. A thermometer whose bulb is blackened with lamp black, placed in a vacuum, and exposed to the direct rays of the sun. It is used for obtaining some indication of the intensity of the Sun's RADIATION. Solstice. The time of maximum or minimum decli- nation of the sun, when the altitude of the sun at noon shows no appreciable change from day to day. The summer solstice for the northern hemisphere, when the sun is farthest north of the equator, is about June 21st, and the winter solstice, when it is farthest south, is about December 2Ist. After the summer solstice the days get shorter until the winter solstice and vice versa. Sounding;. Generally means a trial of the depth of the sea, but in meteorology it is used for a trial of heights in the atmosphere with measures of pressure, temperature, humidity or wind -Telocity. u Soundings of the ocean of air " can be carried out by means of kites, PILOT BALLOONS or BALLONS-SONDES. Spells of Weather. Long spells of the same type of weather are often experienced. Anticyclonic weather may maintain itself for weeks. Depressions often follow one another on nearly the same track for weeks or even months ; if they pass North of us we get a warm Southerly, alternating with a Westerly, type, and accord- ingly rains, gales and fine intervals succeeding one another with some regularity, as in the Autumn and Winter of 1915 to 1916. If the depressions pass to the Spells of Weather. South we get Easterly and North-Easterly winds, with cold weather and rain, or snow in Winter, as in the cold spell in February and March, 1916. Spring. Meteorologically in the northern hemisphere, the three months March, April and May. Astronomically, spring is defined as the period from the vernal EQUINOX, March 21st, to the summer solstice, June 21st. See SEASONS. Squall. A strong wind that rises suddenly, lasts for some minutes, and dies suddenly away. It is frequently associated with a temporary shift of the wind, and heavy showers of rain or snow. The thundersquall is a cool outrushing wind, probably katabatic, that often precedes a thunderstorm. Stability. A state of steadiness not readily upset by small events. For stability of the atmosphere, see ENTROPY. Standard Time. Time referred to the mean time of a specified meridian. The meridian of Greenwich is the standard for Western Europe. The standard meridian of other countries is chosen by international agreement, so that it differs from Greenwich by an exact number of hours or half hours. State of tlie Sky. The fraction of the sky obscured by cloud. It is usually measured on a scale, of (quite clear) to 10 (overcast). A rougher classification suitable for synoptic charts divides the cloudiness into four classes represented by the symbols : b, be, c, and o, which corre- spond to cloudiness of 0-3, 4-6, 7-8, and 9-10 respectively. Statics. A branch of mechanics, dealing with the forces which keep a body at rest. 240 Glossary. Station. A place where regular meteorological observations are made. The classification of British stations is : (1.) First order stations of the International Classifi- cation. Normal Meteorological Observatories : at which continuous records, or hourly readings, of pressure, temperature, wind, sunshine, and rain, with eye obser- vations at fixed hours of the amount, form, and motion of clouds and notes on the weather, are taken. (2.) Second order stations of the International Classi- fication. Normal Climatological Station?: at which are recorded daily, at two fixed hours at least, obser- vations of pressure, temperature (dry and wet bulb), wind, cloud and weather, with the daily maxima and minima of temperature, the daily rainfall and remarks on the weather. At some stations the duration of bright sunshine is also registered. (3) Third order stations of the International Classifi- cation. Auxiliary Climatological Stations : at which the observations are of the same kind as at the Normal Climatological Stations, but (a) less full, or (&) taken once a day only, or (c) taken at other than the recognised hours. StatOSCOpe. A very sensitive form of aneroid barometer, used to show whether a balloon is rising or sinking. The range of its index is very small and it has to be set from time to time by opening a tap leading to the interior of the box. Storm. Is commonly used for any violent atmos- pheric commotion, a violent GALE or a THUNDERSTORM, a rainstorm, duststorm or snowstorm. A gale of wind is classed as a storm when the wind reaches force 10. Storm Cone. See GALE WARNING. Strato-cumulus. 241 StratO- cumulus. The most common form of cloud of moderate altitude, sometimes covering the whole sky. It consists of flattish masses, often arranged in waves or rolls. See CLOUDS. Stratosphere. The external layer of the atmosphere in which there is no convection. The temperature of the air generally diminishes with increasing height until a point is reached where the fall ceases abruptly. Above this poirit lies the stratosphere, which is a region where the temperature changes slowly in a horizontal direction, and is practically uniform in the vertical (see BALLON- SONDE). The height at which the stratosphere commences is often about ten kilometres, but varies. It is higher in regions nearer the equator. Stratus. A sheet of low cloud without definite form ; virtually fog above the level of the ground. See CLOUDS. Summer. Meteorologically in the northern hemi- sphere the months of June, July, and August. Astro- nomically, the period from the summer solstice, June 21st, to the autumnal equinox, September 22nd See SEASONS. Sun. The central body of the solar system, round which the various planets revolve. Almost all meteoro- logical processes depend directly or indirectly upon the radiation received from the sun. Sun-dOgS. Another word for MOCK SUNS or PAR- HELIA ; i.e., images of the sun occurring most often on the halo of 22 radius. Sometimes also used' for portions of a rainbow. Sun Pillar. A column of light extending for about twenty degrees above the sun, most often observed at 242 Glossary. sunrise or sunset. The colour is usually white, but some- times red. It is due to the reflection of light from snow crystals. Sunset Colours. See BLUE OF THE SKY and TWILIGHT, p. 344. Sunshine. An important climatological factor that is determined by a sunshine recorder, an instrument in which the rays of the sun are focussed by means of a glass sphere upon a card graduated into hours. To obtain comparable results the instrument must satisfy a precise spec iiicat ion. The sun will also record its appearance on photographic paper and in many other ways, but in dealing with climatological records it is of the first importance that they should be made on a comparable basis. For the British Isles a set of Monthly Maps of normals for the duration of sunshine, together with those for temperature and rainfall, is given in an Appendix to the WeMy Weather Report for 1913. The only point to which attention will be called here is that when one takes the values for the whole year, so that the possible amount is the same for all stations, there is a gradual falling off in the percentage of possible duration shown on the. records, as one goes northward, from nearly 45 per cent, in the Channel Isles to 25 per .cent, in the Shetland Isles. This difference raises the question whether there are any parts of the globe where the average percentage of the possible duration of sunshine is zero, where in fact the screen of cloud ia perpetual. If there is a region of that character, judging from the meteorological conditions that are known to us, we should expect to find it somewhere near the Arctic Circle in the North Atlantic and the North Pacific, and anywhere Sunshine. M3 along the Antarctic Circle in the Southern Ocean. And, on the other hand, we know that, except in those regions where the belts are interrupted by the trade winds and monsoons, there is hardly any interference with the sun's rays by cloud in the high pressure belts along the Tropics of Cancer and Capricorn. We are not able at present to give the figures which represent these conclusions, but they lead on to speculation as to the causes which account for the distribution of cloud and sunshine and as to why cloud and rain are not confined to special localities or regions. PERCENTAGES OP POSSIBLE DURATION OF SUNSHINE FOR THE WHOLE YEAR FOR DISTRICTS IN THE BRITISH ISLES (AVERAGES FROM RECORDS EX- TENDING OVER THE 30 YEARS, 18811910). Western Side. Per cent. Middle Districts. Per cent. Eastern Side. Per cent. Scotland, W. 30 Scotland, N. 26 Scotland, E. ... 30 Ireland, N. ... 2Q England, N.W. ... 32 England, N.E. 3i Ireland, S. ... 32 Midland Counties 3i England, E. ... 36 England, S.W. ... 38 England, S.E. 38 English Channel 43 Sunshine Recorder : See Observer's Handbook. Sunspot-Numbers : the numbers which are used to represent the variation in the sun's surface from year to year as regards spots. The occurrences of dark spots, sometimes large, sometimes small, which are to be seen from time to lime on the sun's face between its equator and forty degrees of latitude north or south, have long 244 Glossary. been the subject of observation. An irregular periodicity in their number was discovered by Schwabe of Dessau in 1851, using 25 years of observation. Professor R. Wolf of Zurich, by means of records in a variety of places, made out a continuous history of the sun's surface from 1610 to hin own time, which is now continued by his successor, Professor Wolfer. The sunspot-number N is obtained by the formula N = k (lOg 4 /), in which g is the number of groups of spots and single spots, /is the total number of spots which can be counted in these groups and single spots combined, k is a multiplier, representing " personal equation " which depends on the conditions of obser- vation and the telescope employed. For himself when observing with a three-inch telescope and a power of 64 Wolf took k as unity. The method of obtaining the number seems very arbitrary, but from the examination of photographic records by Balfour Stewart and others it is proved that the numbers correspond approximately with the " spotted area " of the sun. One hundred as a sunspot-number corresponds with about 1/500 of the sun's visible disc covered by spots including both umbras and penumbras. (See " The Sun," by C. G. Abbot, 1912.) Spots are now regarded as vortical disturbances of the sun's atmosphere. They have a definite relation to the amplitude of the regular diurnal changes in ter- restrial magnetism : their exact relation to magnetic storms is still unknown. Very many attempts have also been made to connect the phenomena of weather with the sunspot-numbers, Indian famines dependent on Indian rainfall, cyclones in the South Indian Ocean, Scottish rainfall, commercial catastrophes have all been Sunspot- Numbers. 245 the subject of investigation. The mean period of fre- quency of spots is ll'l years and anything with a period approximating to 11 years or a multiple or sub-multiple thereof, may suggest a connexion with sunspots. The most recent and the most effective relation that has come to the knowledge of the Meteorological Office is the direct relation between the sunspot-number and the variation of level of the water in Lake Victoria at Port Florence. The CORRELATION in this case is + '8. The following is the list of sunspot-numbers since 1750 : TABLE OP SUNSPOT-NUMBERS, 1750-1916. i i 2 3 4 5 6 7 8 9 1750 83 48 4 8 3i 12 10 i 10 32 48 54 1760 63 86 61 45 36 21 ii 38 70 106 1770 101 82 66 35 31 7 20 92 , 154 126 1780 85 68 38 23 10 24 83 132 I3i 118 1790 90 67 60 47 41 21 16 6 4 y 1800 H 34 45 43 4 8 42 28 10 8 2 1810 i 5 12 14 35 46 4i 30 24 1820 16 7 4 2 8 17 39 50 62 6 7 1830 7i 48 28 8 13 57 122 138 103 86 1840 63 37 24 ii 15 40 62 98 124 96 1850 66 65 54 39 21 7 4 23 55 94 1860 96 77 59 44 47 30 16 7 37 74 1870 139 in 102 66 45 17 ii 12 3 6 1880 32 54 6c 64 64 52 25 13 7 6 1890 7 36 73 85 78 64 42 26 27 12 1900 10 3 5 24 42 64 54 62 49 44 1910 19 6 4 i 10 46 55 246 Glossary. Surge. First used by Abercromby to denote the general alteration of pressure that seems superposed upon the changes related to a low pressure centre. Synoptic. Giving a general or " bird's eye" view. Synoptic charts show the weather at one point of time, or its mean values for the same interval, over a large area upon a single map. Tangent. A straight line that touches a curve and does not cut it even when produced. Trigonometrically, the ratio perpendicular : base. See SINE. Temperature. The condition which determines the flow of heat from one substance to another. Difference of temperature plays the same part in the transfer of heat as does difference of pressure in the transfer of water. Temperature must be clearly distinguished from HEAT, heat being a form of ENERGY, temperature a factor which affects the availability of the energy. Temperature is measured by a THERMOMETER. Temperature-Gradient. A change of temperature with distance (see GRADIENT) ; but the usual meaning of the term is the lapse rate or rate of decrease of tempera- ture that is found as greater altitudes are reached. In most parts of the earth near the surface a fall of temperature of 1F. for every 300 feet occurs, so that a tableland on the summit of a mountain 3,000 feet high will have a mean temperature 10F. lower than stations near sea level in the same neighbourhood. In the free atmosphere the temperature gradient is usually measured in degrees centigrade per kilometre of height. The decrease is a little slower than that found by mountain observations. Ic amounts to about 6C. per kilometre in England in the lower strata, but increases to Tempera ture- Graaien t . 247 from 7 to 8 in the strata that lie between 5 and 9 kilometre height. Above 11 or 12 k. the fall ceases altogether. In the tropics the temperature gradient of 7 or 8 per kilo- metre is continued up to 15 k. height or more. See Tables under BALLON-SONDE and DENSITY. Tension of Vapour. See VAPOUR. Terrestrial. Having reference to the earth. The term Terrestrial Radiation refers to the heat radiated from the earth. * Thaw. The term used to denote the cessation or break-up of a frost ; usually the result of the substitution of a South-Westerly type for a North-Easterly type in this country, or of the sudden incursion of a CYCLONE from the west. In more northern latitudes the u spring thaw " is a periodic event, denoting the seasonal progres- sion, the unlocking of ice-bound seas and the melting of the snow. In these latitudes, in Western Europe, though not in Canada or Asia, the sun is generally at sufficient ALTITUDE about noon, except near mid-winter, to effect a partial thaw by day, even in the midst of a protracted frost, if the sky is clear. Thermodynamics. That part of the science of heat which deals with the transformation of heat into other forms of energy. See ENERGY and ENTROPY. Thermogram. The continuous record of temperature yielded by a thermograph. Thermograph. -- A self-recording thermometer generally consisting of a " Bourdon " tube or a bimetallic spiral with a suitable index. A large spirit thermometer with a float is also used. A mercury thermometer can also be arranged to give a photographic record, as at the observatories of the Kew type. Thermometer. An instrument for recording tern- 248 Glossary. perature, usually by means of the changes in volume of mercury or spirit contained in a glass tube with a bulb at one end, but not infrequently by the change of an electrical resistance. Generally the temperature of the air is required, and this is not easily obtained, particularly in sunny weather. Readings of thermometers exposed in a Stevenson screen, however, are sufficiently accurate for practical purposes. Thunder. The noise that follows a flash of lightning, attributed to the vibrations set up by the sudden heating and expansion of the air along the path of the lightning. The distance of a lightning flash may be roughly estimated by the interval that elapses between seeing the flash and hearing the thunder, counting a mile for every five seconds. It is somewhat astonishing in common experience at what little distance thunder ceases to be audible : the interval between flash and sound seldom reaches a full minute,* which would set a limit of twelve miles to the distance of audibility. Considering the violence of the commotion in the immediate neighbourhood of a flash it might be expected that the sound would be perceptible at far greater distances. Two considerations affect the question first, it is a common experience with balloonists that sounds from the balloon are less easily audible on the ground than sounds from the ground to the balloon and this observation is confirmed by the experience on mountains that sounds from below are more easily audible upwards than sounds from above downwards ; the second consideration is that in thunder-weather there are great discontinuities in the structure of the atmosphere, so that the distortion of the rays of sound, which partly accounts * Capt. Cave cites an occasion on which not less than two minutes elapsed. Thunder. 249 for the smaller audibility of sounds from below, is much exaggerated on the occasion of a thunderstorm. Thunderstorm.* In the British Isles, except at the stations on the Atlantic Coasts, well developed thunder- storms occur most frequently in the summer, and especially during the afternoon. The barometric disturbances with which they are associated are generally too limited in area to be called cyclones, but like cyclones they frequently move towards the East. They are nearly always accom- panied by heavy rain, which is sometimes preceded by a squall that blows outward from the advancing storm, while the barometer rises suddenly and then remains comparatively steady. The squall brings with it cool air. See LINE SQUALL. Lightning is seen, and thunder heard, before the arrival of the storm itself, but the flashes are generally most brilliant during the heavy rain. The thunder then follows the flashes after a very short interval, showing that the discharge has taken place at no great distance. The thunder-cloud seems to be an extreme develop- ment of the cumulus cloud, in which the ascending currents have reached to a considerable height and spread outwards at the top. In consequence they often have the shape of an anvil. The thunder type of cumulus has a rounded summit, with a clearly defined border. Obser- vations of the time and place of occurrence of thunder- storms show that they are generally long and comparatively narrow, and move broadside across the country. The precise conditions that lead to their formation are not understood. In some parts of the tropics thunderstorms are frequent and very violent. * See photographs under CLOUDS and MAMMATO-CUMULUS. 250 Glossary. IMMUNITY PROM THUNDERSTORMS IN VARIOUS PARTS EXTENDING MAINLY OVER THE 25 YEARS Table of " odds against one", expressing the random chance STATION. Spring. Summer. Autumn. Winter. SCOTLAND. ( Sumburgh Hd. (Island) A I Deerness (Island) "S -{ St or no way (Island) ... Wick (E. Coast) I Fore Augustus (Inland) 767 135 192 IIO 108 164 30 77 30 56 569 57 134 103 606 45o 68 102 225 225 ( Nairn (N.E. Coast) ^ | Aberdeen (E. Coast) ... \ Braemar (Inland) K | Dundee (E. Coast) ILeith (E. Coast) 192 135 74 35 32 20 25 15 23 284 142 162 84 228 1,120 1,120 281 750 375 fLaudale (W. Coa*t) ... 43 1 Rothesay (Island) \ Glasgow (W. Coast) ... ^ | Pinmore \ L Douglas (Isle of Man) 34 44 68 92 53 24 33 26 45 24 162 97 ' 49 33 80 204 118 250 IRELAND. f Malin Hd. (N. Coast)... | Blacksod Pt. (W. Coast) { Markree Castle (Inland) | Armagh (Inland) LDonaghadee (E. Coast) 92 109 72 97 36 58 32 25 46 120 152 175 505 175 134 73 78 391 1,120 f Dublin (E. Coast) c/2 J Valencia (S.W. Coast)... t Roche's Pt. (S. Coast)... 61 92 82 18 60 43 99 78 108 250 750 150 g g /Scilly (St. Mary's) ... jfj 5 I Jersey (St. Aubin's) 82 36 42 16 65 28 68 58 Thunderstorm. 251 OF THE UNITED KINGDOM AS SHOWN BY OBSERVATIONS 1881-1905 (COMPILED BY F. J. BRODIE). of a thunderstorm on any day in the several seasons of the year. STATION. Spring. Summer. Autumn. Winter. r N. Shields (E. Coast) ... pq J Durham (Inland) a | York (Inland) * I Spurn Hd. (E. Coast) ... 74 27 27 32 20 10 13 II 95 11 62 375 562 2,250 375 rHillingdon (Inland) ... W . I Yarmouth (E. Coast) ... ^ 1 Norwich (Inland) I. Cambridge (Inland) ... 21 45 19 27 8 13 9 10 37 87 40 47 205 562 209 750 . fWorksop g S 1 Cheadle 3 P J Churchstoke q ,g "J Loughborough 5 o Cheltenham 1 Oxford 27 17 36 3i 29 40 IO Q 15 II 14 H 61 25 76 65 87 63 225 125 220 1,000 f 161 " 750 f London (Brixton) . j Margate (E. Coast) ... K . { Dungeness (S. Coast) ... 00 i Southampton (Inland) I Hurst Castle (S. Coast) 25 55 53 45 74 ii 20 17 17 29 53 103 5i 58 76 225 1,120 562 225 322 fc f Ay sgarth (Inland) J | Stonyhurst (Inland) ... g ^ ^i Liverpool (W. Coast) .., W ^ | Llandudno(W. Coast)... L Holy head (W. Coast) ... 27 23 74 68 72 13 IO 27 29 30 39 27 114 60 69 237 94 321 141 374 ( Pembroke (W. Coast)... 1 Falmouth (S.W. Coast) 00 / Cullompton (Inland) ... 209 177 72 9 2 56 30 95 73 108 562 150 150 252 Glossary. Time.* For all common purposes Greenwich mean civil tinie is now used in all places in Great Britain, except at Canterbury, and clocks are set by telegraphic signal from Greenwich. Previously, each town or village clock kept its own local mean time and had to be set by the local time keeper, usually the parson, with the aid of a sundial or some other means of ascertaining the time from the sun. In Ireland clocks are set according to Dublin time, which is 25 minutes after Greenwich time. In meteorology the hours of the civil day are numbered from 1 to 24, the counting beginning from midnight. Thus the hours of observation for telegraphic reporting are, lh., 7h., 13h., 18h., with a subsidiary observation at 21h. Meteorologists are closely interested in good time- keeping, because punctuality is of importance, both with climatological observations and with those that are made for the maps used in forecasting. For the former local time, and for the latter Greenwich time is taken as the standard. The records of self-recording instruments, when the sheet is changed once a . week, are for many purposes useless unless marks are made on the trace at definite times, so as to allow for irregularities in the running from day to day. See STANDARD TIME. Tornado. A short lived, but very violent wind. In West Africa the tornado is the squall which accompanies a thunderstorm ; it blows outward from the front of the storm at about the time the rain commences, and in all parts of the world similar squalls occur, associated with- thunder. It is also the name applied to small but very * In 1916 " Summer Time," one hour in advance of Greenwich Time, was used in the United Kingdom from May 21st to September 30fch ; in 1917 from April 8th to September 17th ; between the limiting dates (September 30th, 1916, and April 8th, 1917), GLM.T. was used in Ireland. Tornado. 253 violent whirlwinds of one or two hundred yards diameter. These whirlwinds often do immense damage in the United States, where they are. known as cyclones, completely destroying every tree and building in their track. They are not unknown in England, but are less frequent and less violent than in North America. Torricelli, Evangelista.- The inventor of the barometer, born at Piancaldoli in 1608. At the age of 20 he went to Rome to study mathematics. In 1641 he met Galileo, and remained with him as his amanuensis till the death of Galileo three months later. Torricelli then became professor to the Florentine Academy ; he lived in Florence till his death in 1647. Torricelli explained the fact, already known, that water will only rise about 32 feet in the pipe of a suction pump ; he argued that if this was due to the pressure of the atmosphere the column of mercury that would be supported would be a little under 2| feet, since mercury is 13^ times as heavy as water. He performed the experiment that confirmed his theory. He also enunciated various fundamental principles in hydro- dynamics. Trade Winds. The word " trade " in this expression is said to mean " track " and trade winds are winds which keep to a fixed track. We naturally turn to tropical or subtropical regions for track winds. The easterly wind on the margin of the ice in the Antarctic is very persistent but not very steady. The best known examples of track winds are the North East Trade and the South East Trade of the Atlantic Ocean. In a publication of the Meteorological Office, M.O. 203, on the Trade winds of the Atlantic Ocean the areas selected for the observations are for the North East Trade from 10 N. to 30 N. between 30 W. long, and the West Coast of Africa, and for the 254 Glossary. ' South East Trade the two pairs of ten-degree squares to 20 S., to 10 W. and 10 to 30 S., to 10 E. The Canary Islands and Cape Verde Islands come in the region selected for the North East Trade, and St. Helena in that selected for the South East Trade. At St. Helena there is a self-recording anemometer at a point 1,960 feet above sea level, which is maintained for the Meteorological Otfice. The coast of Africa disturbs the regularity of the North East Trade in the Eastern part of the area selected, but the monthly results for the whole area give a wind with a mean direction for the whole year of N. 30 E., varying between N. 18 E. in May and N. 48 E. in January, and a mean velocity for the whole year of 1O6 miles per hour (4* 7 ro/s), varying from 7*4 miles per hour (3*3 m/s) in October to 13*5 miles per hour (6 rn/s) in April. The South East Trade shows a mean direction for the year of S. 38 E., varying from S. 35 E^in February and October to S. 41 E. in August and November, and a mean velocitv of 14'2 miles per hour (6*4 m/s) varying from 13*1 miles per hour (5'9 m/s) in January to 15 miles per hour (6*7 m/s) in April, June and August. Thus, the North East Trade shows more variation in direction than the South East, and its velocity exhibits a marked seasonal variation, with a maximum in April and a minimum in October, which has no counterpart in the South East. These are taken from observations made by ships at sea, the velocities being determined by a scale of equivalents of ** Beaufort estimates." See BEAUFORT SCALE. When the measures of the direction and velocity at St. Helena are taken they show the monthly values oscillating about S. 40 E., from S. 35 E. in October to S. 42 E. in April, and a very marked seasonal change of velocity from Trade Winds. 255 13 miles per hour (5*8 m/s) in May to 20 miles per hour (8*9 m/s) in September. This is very nearly the counter- part of the seasonal variation of the North East Trade. The mean of the velocities of the two " trades" works out at about 11*6 miles per hour (5'2 ni/s) throughout the year. 'The flow which is represented by these winds comprises two streams about 1,000 miles wide, the courses of which are kept steadily from N.E. or from S.E. for about 2,000 miles. These steady currents carry an enor- mous amount of air. Taking the run at 300 miles per day over a thousand mile front the flow for a thickness of 1 mile would be 300,000 cubic miles a day ; it would take rather less than 10 years for the whole of the atmosphere to pass through ; if it be two miles thick the circulation would be complete in five years. And on the same assumption, the two trades acting together yet they use only one-twentieth part of the belt of the earth's surface available for approaching the equator from North and South would deliver the equivalent of the whole of the atmosphere in the course of about two and a half years. So far, we have considered only the trade winds of the Atlantic Ocean between the African and American coasts. Similar winds under similar conditions exist to the West- ward of the American coast where, it may be remarked, the coast line bends away, for a 1,000 miles or more, to the Westward after crossing the equator from the Southward, very much in the same way as the African coast line does, so that the North- East Trade wind of the Eastern Pacific lies to the West of, and not opposite to, its South-East partner, just as in the Atlantic. These are the only regions where the recognised characteristics of the Trade winds are well marked. In the Indian Ocean 256 Glossary. they are replaced by the Monsoon winds, which are con- tinued across the equator from the North East in the winter and from the South East in the summer, when the air current from the south is carried forward as a South-West monsoon over India. A suggestion of " Trade " conditions appears in the Western Pacific to the North East of Australia, but it is less well marked than in the Atlantic and Eastern Pacific. It should be noted that in the West Indian region in June the wind varies from North. East to South East, and the same is true off the coast of tropical South America in December. Locally it is still known as the Trade wind, although it may be blowing from the South East, away from the equator. The explanation of the origin of the Trade winds which is given in all books on Physical Geography is due originally to Edmund Halley, a personal friend of Newton's, secretary of the Royal Society, and subse- quently Astronomer Royal at the beginning of the eighteenth century. It attributes the flow of air south- ward and northward on either side of the line to the replacement of air which has been heated by the warmth of the equatorial belt, and has, in consequence, ascended to the upper air and passed away. * Jete Hadley, also a personal friend of Newton's, associated with him in the invention of the sextant, explained the easterly com- ponent by bringing the rotation of the earth into account. Whatever real ground there may be for a flow of air towards a belt of high temperature along the equator on the ground of local heating, it appears clear from the maps that the great arterial currents which we have described, and which are commonly understood as Trade winds, are really parts of the general circulation of the Trade Winds. 2f>7 Atmosphere, governed by the distribution of pressure. A map of the distribution of pressure and winds over the globe, such as that for January in the Barometer Manual for the Use of Seamen (M.O. publication No. 61) shows that there are two belts of high pressure on either side of the equator, about latitude 30 N. and 30 S. respectively. These belts are not continuous but form a succes-ion of anticyclonic areas, each with its appropriate circulation. The southern hemisphere pr-sents the simpler arrange- ment, because the land areas which cause disturbance of the order are less extensive. Along the parallel of 30 S. latitude we have anticyclonic areas with centres, (1) at 100 W., 30 west of South AmerL-a, (2) at 10 W., nearly midway between South Africa and South America, (3) a system with double centre between the Cape and Australia, which extends along the Southern shore of Australia and develops a secondary centre there. The regularity of the distribution is thus much distorted by the Australian continent. The channels of low pressure between the anticycionic areas are breaches in the belt of high pressure through which the great arterial currents of the trade winds flow over the Eastern Atlantic and Eastern Pacific. And these great currents are in reality features of a circulation round the isolated regions of high pressure. There seems little possibility of any alternative for the distribution thus described. Halley's explanation of the trade winds supposes a low pressure area over the equatorial belt, continually maintained by the convection of the rising air, and continually fed by a flow of air from a belt of higher pressure north or south. So that the flow of air is thought of as from high pressure to low pressure. Whatever may be the actual state of 13204 I 258 Glossary. things close to the equatorial belt, the arterial currents of the trade winds are clearly shown by the map to be great rtreams of air with 2,000 miles of run, and with high pressure on the one side and low pressure on the otht r, such as we may find in all cases of well established air currents over the earth's surface, whether they last only lor a few hours, a few days, as in the intermediate and polar regions, or the whole year, as in the regions of the trade winds. If we cross the great currents from west to east we are travelling towards lower pressure ; from east to west towards higher pressure. Obviously, we cannot continue this process all round the globe, and going westward we see that the pressure soon gets to a maximum and then falls off again, and the falling off is associated with a change in the direction of the current from south-east to east or north-east, or from north-east to east and south-east. The WIND -ROSES show that the western boundary of the high pressure is a fluctuating boundary, not a fixed one. Going Eastward we are brought up by the great land areas of Africa, where our knowledge of the distribution of pressure is little more than guessing from a few isolated stations which are affected by the uncertainties of REDUCTION TO SEA-LEVEL. But further investigation must lead to a distribution which corre- sponds with a low pressure at sea-level. That these great currents are really part of a great circulation which is governed by the distribution of pressure may be illustrated by comparing with the steadiness of the winds at St. Helena* (lat. 16 S., long. 5*42 W.), the following table of the winds at Suva, Fiji, which is in latitude 18 S., long. 17826 h. : * See Trade Winds of the Atlantic Ocean. M.O. publication No. 20H. Trade Winds. 259 TABLE OF WIND DIRECTION AT I) A.M. AT SUVA, FIJI, IN 1911. i N. N.B. E S.E. S. s.w. w. X.W. Calm. January .. 6 3 : 5 4 12 February .. 4 , 5 5 2 2 9 March .. 4 G 2 2 3 13 April .. 7 12 I I 4 4 May .. 2 15 2 f 3 i June .. 3 6 I I I 5 10 July .. i 5 4 I I I 4 ii A.ugust .-4 6 4 5 I 4 7 September .. 2 10 8 3 3 2 2 October .. L 10 7 4 3 2 I 3 Xovember .. 16 6 3 .1 4 December .. 9 10 <) 2 I Year ... 43 104 ! 50 37 8 19 ; 4 i 26 74 From this it appears that the wind conditions at the two places are quite different, though their relations to the equatorial belt of high temperature are altogether similar. We must, therefore, regard the trade winds as the main streams of .air in the general circulation by which the intertropical region is supplied. The greater part of the supply turns eastward and gets away from the equatorial region again by passing round the western boundaries of the anti-cyclonic regions. Some part of it may go to feed the rain storms of the doldrums, but what fraction of the whole supply is so used is not known. The extension of Halley's theory of the trade winds 13204 I 2 2t>6 Glossary. provides that the air after ascending in the equatorial region should flow back again away from th e equator,and on account of the rotation of the earth the northward flow should be diverted towards the east, and thus become a south-west wind. Accordingly, south-west and north-west winds are to be expected above the north-east and south-east trades. Two thousand miles is a long way for the air to travel with no more diversion than 45 from the path of its desire when the rotation is taking place at the rate of lr> sin \ degrees per hour unless the distribution of pres- sure interferes, but the theory seems to be confirmed by an observation made in 1856, by Piazzi-Smyth (then Astronomer Royal for Scotland) of a south-west wind at the top of the peak of Tenerife, 12,500 feet high, over the north-east trade flowing below. The transition is at about 10,000 feet, and the existence of a south-westerly current over the north-east trade over the ocean was verified by balloon observations by Teisserenc de Bort, although the question was the subject of some discussion at the time. If, however, we regard the surface winds of the trades as part of the general circulation of the atmosphere con- trolled by pressure, we cannot ^do otherwise in the case of the upper currents, consequently we ought to find our explanation of the south-westerly current over the north- east trade as evidence of low pressure over the central region of the Atlantic north and south of the equator, and high pressure over the African land adjoining, giving rise to a gradient for equatorial winds up above, or for polar winds below. We should then have to conclude that the high pressure belts of the tropics are reversed in the upper regions, a corclusioii that carries with it some consequences which at first sight are not easily disposed of. Trade Winds. 261. The trade winds have an interest for meteorologists quite independent of their geographical interest and of their place in the general circulation. They are a sort of laboratory in which one can study the properties of a great current of air of known temperature and humidity flowing steadily over the surface of the sea and affected by the turbulent motion caused by the surface. Professor Piazzi-Smyth, spent part of July, August and part of September, 1856, in the main stream of the north- east trade at Tenerife at a height of 8,900 feet or more. He found the air remarkably dry, while below him at a height of about 5,000 feet he could see the long strings of cumulus clouds that are characteristic of the trade wind forming a level horizontal layer upon which it seemed that one might walk to the neighbouring Canary Islands if it were not for a gap between the cloud sheet and the cloud actually in contact with the mountain, which was some 1,000 feet below the trade wind cloud. The wind at 8,900 feet was generally light. Cirrocumulus clouds, moving from south-west, were sometimes visible in the sky above at a height estimated at 15,000 feet. Occasionally the north-easterly wind got nearly to gale force at the high level, and on other occasions the wind blew strongly from the south-west there. The interme- diate region between the north-east trade and the south- west counter trade was found to be generally a region of light winds, while the trade wind clouds were formed at the middle height of the trade. It would appear, therefore,- that we have in the body of the trade wind, an analogy, but on a smaller scale, of the separation of the troposphere from the stratosphere which is universal. The boundary between the two is the limit 262 GUssary. of convection 'from the surface. In the trade winds the boundary of convection is marked by the layer of clouds, above that level the moist air does not penetrate. The uniformity of level suggests that the limit is depen- dent upon the turbulent motion due to the eddies caused by the friction at the surface, the effect of which extends upwards to a height which depends upon the length of the "fetch." The convection is therefore in this case partly dynamical, and it is curious that the clouds>formed are often spoken of as rollers. Piazzi-Smyth speaks of summer and winter conditions as though he were dealing with a climate of temperate latitudes instead of a region to the south of the line of tropical high pressure. The observations of St. Helena, which are made at about 2,000 feet above sea level in the south-east trade, show persistent south-easterly winds, with a mean humidity for the yearj|f of 89 per cent., ranging from 88 in January to 91 in March, and the normal amount of cloud works out at 8'5 (on the scale 0-10, or 85 per cent) for the year. The mean temperature reaches a maximum of 291'9a, 66-lF., in March, and a minimum of 287-la, 57'3F., in September, and the rainfall has a normal maximum of 131 mm. in March and a minimum of 40 mm. in November. Hence, there is a definite seasonal variation, but it lags behind the corresponding changes in higher latitudes of the same hemisphere. Trajectory. The path traced out by a definite particle of air in a travelling storm, or the horizontal projection of the course followed by a pilot balloon : the trajectory, as worked out from theodolite readings, may be plotted on squared paper, and the direction and velocity of the wind at any given height deduced therefrom. Tramontanes. 263 Tramontana. An Italian word for the northerly winds of Italy which blow from the mountains. Transparency. The capacity for allowing rays of light or some other form of radiation to pass. Thus glass is transparent for the visible radiation of light. Rock- salt is specially transparent for the rays of radiant heat. See VISIBILITY. Tropic. One or other of the circles of 23^ of latitude north and * south of the equator, which represent the furthest position reached by the sun in summer and winter in consequence of the tilting of the earth's axis with reference to the plane of the ecliptic. The term applies also to the zone of the earth lying between them. The northern circle is called the* tropic of Cancer, the southern the tropic of Capricorn. Tropical. Belonging to the regions of the tropics, or Hinilar to what is experienced there. The word tropical is often used for the region between the tropics, which is more strictly called intertropical. Tropopause. The lower limit of the STRATOSPHERE. Troposphere. A term suggested by Teisserenc de Bort for the lower layers of the atmosphere. The temperature falls with increasing altitude up to a certain height (see TEMPERATURE GRADIENT), and the part of the atmosphere in which this fall occurs is called the troposphere. In these latitudes (50) it extends from the surface for a thickness of some 7 miles, or 11 kilometres ; in the tropics the thickness may reach 10 miles, or 16 kilometres. Trough. The period of lowest barometer during the passage of a depression. Taking the fluctuations of the barometer to be analogous to the fluctuations of level 264 Glossary. caused by waves, which is, however, not a very good analogy, the trough of the wave suggests as its counter- part the lowest reading of the barometer. With the barometer the passage of the trough is generally marked by phenomena of the type of a LINE SQUALL, a sudden rise of pressure, veer of wind, drop of temperature, and one or more squalls ; nothing of that kind takes place in the trough of a wave. Twilight. See p. 344. Twilight Arch. On a clear evening after sunset a dark arch with a pink edge may be seen to rise from the eastern horizon ; the distinction between the darkness below the arch and the brighter sky above it becomes rapidly less as the arch rises in the sky. The dark space is really the shadow of the earth. In mountainous countries shadows cast by mountains between the sun and the observer may be seen to rise from the twilight arch. The pink edge of the arch is due to reflection from particles in the atmosphere which are illuminated by rays of the sun that have lost nearly all their blue light from lateral scattering (see BLUE OP THE SKY). Type. Different distributions of atmospheric pressure are characterised by more or less definite kinds of weather, and accordingly when a certain form of distribution is seen on a chart the weather is described as belonging to such and such a type. The types are defined as cyclonic, anti- cyclonic and indefinite, and by terms denoting the direction of the isobars. Thus, a " southerly type" denotes a weather chart on which the isobars are shown as more or less parallel lines running north and south. A northerly type will also have isobars running north and south ; the distinction will be that in the southerly type barometric pressure will be high in the east, whereas in the northerly Type. 265 type the higher pressure will be in the west. In each season each type has more or less definite kinds of weather. Thus, the anti-cyclonic type will have dry weather, the cyclonic wet ; the southerly type will in general be warm, the northerly cold. Typhoon. A word of Chinese origin applied to the tropical cyclones occurring in the western Pacific near Japan and the Philippine Islands. They are extremely violent circular storms of 50 to 100 miles diameter, and travel slowly. Exactly similar storms are known as hurricanes in the West Indies, and as cyclones in the Bay of Bengal. The hurricanes of Mauritius are also similar to typhoons, See HURRICANE. Upbank thaw. A state of affairs in which the usual fall of temperature with height is reversed, a thaw, or an increase of temperature occurring on mountains some- times many hours before a similar change is manifested in the valleys. It is due to the superimposition of a warm wind blowing from a direction differing from that of the surface wind, and occurs most usually at the break-up of a frost, on the approach of a cyclonic system, but sometimes during the prevalence of anti-cyclonic con- ditions, when a down-current of air is dynamically heated in its descent from a great height. Under these conditions, at 9 a.m. on February 19th, 1895, at the end of a great frost, the temperature at the summit of 3- Ben Nevis was 9'8a., 17'G F. higher than at Fort William, 4,400 feet below. It is probable that this INVERSION of the normal temperature gradient is the cause of the phenomenon known ;is GLAZED FROST (q.v.}. 266 Glossary. V-shaped depression. Used to describe isobars having the shape of the letter V, which enclose an area of low pressure. The point of the V is always towards the south or east. Vapour-pressure. The pressure exerted by a vapour when it is in a confined space. In meteorology vapour-pressure refers exclusively to the pressure of water- vapour. When several gases or vapours are mixed together in the same space each one exerts the same pressure as it would if the others were not present, and the vapour-pressure is that part of the whole atmospheric pressure which is due to water-vapour. See AQUEOUS VAPOUR, RELATIVE HUMIDITY. Vapour-tension. A now obsolete term for vapour- pressure. There seems now to be no reason why this should be called tension. Vector. A straight line drawn from a definite point iii a definite direction. Thus a radius vector of the earth in its orbit is the line drawn from the sun to the earth. A vector quantity is a quantity which has a direction, as well as magnitude, and of which the full details are not known unless the direction is known. In meteorology the wind and the motion of the clouds are examples of vector quantities ; the directions, as well as the magnitudes, are required, whereas in the case of the barometer or the temperature the figures expressing magnitude tell us everything. They are called scalar quantities. All vectors obey the parallelogram law. That is to say, that any vector A may be exactly replaced by any two vectors B and C, provided that B and C are adjacent sides of any parallelogram, arid A the diagonal through the point where B and C meet. Also the converse holds. The position of an airship a given time after starting is Vector. 267 an example. The two vector quantities that bring about the final result are the velocity and the direction of the airship through the air, and the velocity and direction of the air, i.e., the wind. Suppose an airship pointing S.W. and with speed of 40 miles an hour. After two hours its position on a calm day is 80 miles S.W. of the starting point. Now suppose the airship has to move in a S. wind of 30 miles an hour ; after two hours this wind alone would place the airship 60 miles N. of the starting point. The real position will be given by drawing two lines representing these velocities and finding the opposite corner of the parallelogram of which they form adjacent sides. See COMPONENT. Veering". The changing of the wind in the direction of the motion of the hands of a watch. The opposite to BACKING. Velocity. Velocity is the ratio of the space passed over by the moving body to the time that is taken. It is expressed by the number of units of length passed over in unit time, but in no other sense is it equal to this space. It can be expressed in a variety of units. For winds metres per second, kilometres per hour and miles per hour are most common. When a velocity is variable a very short time is chosen in which to measure it. Thus the statement " at 11 a.m. the wind was blowing at the rate of 60 miles per hour " means that for one second or so at just 11 a.m. the wind had such a rate, that had that rate continued for an hour the air would have travelled ()() miles in that hour. From the time of its establishment in 1854 until the final evaluation of the Beaufort Scale in 1905 it was the custom of the Meteorological Office to measure wind velocity by the cup anemograph which gave the " run " 268 Glossary. of an hour in miles. Miles per hour were accordingly the accepted unit of velocity, but when it became certain that for anemometers of the standard size (9-in. cups and 2-ft. arms) the " factor " was not 3 but 2-2, so that the miles were not really miles after all, some change of nomenclature was necessary to mark so great a change of habit. The pressure-tube anemograph gave the oppor- tunity of measuring the velocity at any instant instead of the run during an hour : a gust that lasts only part of a minute is more appropriately measured by the distance which the wind travels in a second than by the distance which it might travel in an hour if it remained unchanged throughout the hour. Moreover in all questions of dynamical calculation the second is the unit of time. Gunners use the foot per second as their unit. The metre per second is more suitable for units on the C.G.S. system, and is now used in all the publications of the Meteorological Office. Vernier. A contrivance for estimating fractions of a scale division when the reading to the nearest whole division is not sufficiently accurate. The vernier is a uniformly divided scale which is arranged to slide alongside the main scale of an instrument. Informa- tion as to the graduation of a vernier and the method of reading is given in The Observer's Handbook. Vertical. The direction of the force of gravity or of the plumb line, so called because it refers to the vertex, /.#., to the zenith, A vertical line is perpendicular to the surface of still water, which is horizontal. When produced it passes through the zenith and close to the centre of the earth. Two vertical lines can therefore never be parallel, although if they are near together they are very nearly .po. A vertical plane at any point likewise passes Vertical. 269 through the zenith and very close to the centre of the earth. A vertical circle is a GREAT CIRCLE passing through the zenith and the nadir ; and the vertical circle also passing through the east and west points is called the prime vertical. Viscosity. The property of a liquid or gas whereby it resists the tangential motion of its parts. See DIFFUSION. Visibility. A term used in describing the effect of the atmosphere, and the amount of light in the sky, on the maximum distance at which an object can be seen, and the clearness with which its details can be made out. The visibility of the atmosphere depends chiefly on the amount of solid or liquid particles held in suspension by the air. On a cloudy day it is usually equally good in all directions, but on a sunny day the visibility is usually better, i.e. 9 objects can be seen more clearly, when looking away from the sun than when looking towards it. In England the visibility of the atmosphere is usually bad towards the end of a spell of fine, calm weather ; but in these cases the occurrence of a shower of rain frequently clears the air and gives rise to good visibility. On the other hand during rainy weather the visibility is frequently bad, even when it is not actually raining. The visibility of objects on the ground, when looked at from an aeroplane, is sometimes bad even when the visibility between two points on the ground is good and the sky is cloudless. This condition usually arises in calm, anti-cyclonic weather and is due to a layer of haze at a definite height above the ground. The occurrence of haze during fine dry weather can frequently be connected with the proximity of a large town. A light north-easterly wind sometimes carries haze to a distance of 70 miles south-west of London, QA 270 Glossary. one such occasion, it was found that at Farnborougli, 3~> miles from London, the haze extended to a height of 7,000 feet. Above that height the atmosphere was perfectly clear. It frequently occurs that an aeroplane can be seen from the ground at a time when the ground cannot be seen from the aeroplane. This condition arises when there is a low haze or mist which pre vents a large part of the sun's rays from reaching the ground. The aeroplane itself is brightly lighted by the direct rays of. the sun, while the light reflected upwards from the top of the haze towards the aeroplane overpowers the feeble rays from the lees brightly lighted ground The effect is similar to that of a lace curtain over a window, which enables the occupants of a room to see out, while the interior cannot be seen from the outside. Occasionally it happens that an aeroplane can see the ground while remaining invisible itself. The condition arises only on sunny days, but its cause is not understood. The limit of visibility depends chiefly on the number of dust particles per cubic centimetre of the atmosphere. An apparatus for counting this number has been designed (see DUST-COUNTER), and used by Mr. John Aitken, F.R.8., who has found as few as 10 arid as many as 7,000 dust- particles per cubic centimetre in the open country. The distance of the furthest visible object was found to depend on the number of particles in the atmosphere, and on its humidity. For a given depression of the wet-bulb thermometer, the limit of visibility multiplied by the number of particles per c.c. of air was found to be roughly constant. This constant, however, increases as the air becomes drier. For a given depression of the wet bulb, therefore, the number of particles in a column 1 sq. cm. in cross section Water. 271 and stretching from the observer to the limit of visibility is constant. Mr. Aitken's estimates of its values for different degrees of humidity are shown in the accom- panying table. Depression of wet-bulb. C Number of particles of dust to produce complete haze. 2 to 4 12 x 10 9 4 to 7 17 x 1C 9 1 7 to 10 22 x 10 9 Vortex. See p. 347. Water. The name used for a large variety of sub- stances such as sea-water, rain-water, spring-water, fresh water of which water, in the chemical sense, is the chief ingredient. Chemically pure water is a combination of Hydrogen and Oxygen in the proportion by weight of one part to eight, or by volume, at the same pressure and tem- perature, of two to one, but the capacity of water for dis- solving or absorbing varying quantities of other substances, solid, liquid or gaseous, is so potent that the properties of chemically pure water are known more by inference than by practical experience. They are in many important respects different from those of the water of practical Jife. The most characteristic property of ordinary water is that we find it in all three of the molecular states ; we know it in the solid state as ice, as a liquid, (over a considerable range of temperature so well recognised as to be used for graduating thermometers), and as a gas. Thus, freezing and boiling are the common experience of many specimens of the water of ordinary life, and yet it is difficult to say in what circumstances, if any, perfectly pure water can be made to freeze or to boil. 272 Glossary. Ordinary water is a palatable beverage, and is a medium in which a variety of forms of vegetable and animal life can thrive, but pure water freed from dissolved gases is perfectly sterile and quite unpalatable. Ordinary water has a mass of 1 gramme per cubic centi- metre (62*3 Ibs. per cubic foot) at 277a. Sea- water contains dissolved salts to the extent of as much as 35 parts per 1000 parts of water, and its density varies from 1*01 to 1*05 g/cc Rain-water is the purest form of ordinary water, it contains only slight amounts of impurity in the form of ammonium salts derived from the atmosphere. Spring- water contains varying amounts of salts dissolved from the strata of soil or rock through which it has percolated, The most common of these salts is carbonate of calcium, which is specially soluble in water that is already aerated with carbonic acid gas; sulphates of calcium and other earthy metals are also found, and sometimes a considerable quantity of magnesia. These dissolved salts give the waters of certain springs a medicinal character. In some districts underground water is impregnated with common salt and its allied compounds to such an extent that it is no longer called water, but brine. When impure water evaporates, the gas that passes away consists of water alone, the salts, which are not volatile, are left behind ; similarly when w r ater freezes in ordinary circumstances the ice is formed of pure water, the salts remain behind in the solution ; so that, except for the slight amount of impurity due to mechanical processes, pure water can be got from sea- water or any impure water, either by distilling it, or by freezing it. Besides the solicj constituents which give it a certain degree of what is called " hardness," ordinary water con- tains also small quantities of gases in solution, presumably Water. 273 oxygen and carbonic acid. When the water freezes the ice consists of pure water, and the dissolved gases collect in crowds of small bubbles. The thermal properties of water, in the state of purity represented by rain-water, are very remarkable. Starting from ordinary temperatures, such as 290a. (62*6F.) and going upwards in the scale, the water increases in bulk, and part evaporates from the surface, until the boiling point is reached, a temperature which depends upon the pressure, as indicated on p. 300. Then the water gradually boils away without any increase of temperature, but with the absorption of a great amount of heat. Going down- wards, the bulk of the water contracts slightly until the temperature of 277a. is reached (4C., 39'1F.) : that is known as the temperature of maximum density of water. From that point to the freezing point of water, there is a slight expansion of one eight-thousandth part, and in the act of freezing there is a large expansion amounting to one-eleventh of the volume of water. It is in consequence of this change of density in freezing that ice floats in water with a one-eleventh of its volume pro- jecting, if the ice is clean, solid ice, and the water of the density of fresh water. Salt water would cause a still larger fraction to project, but floating ice carries with it a considerable amount of air cavities and sometimes a load of earth so that the relation of the whole volume of an iceberg to the projecting fraction is not at all definite. Water-atmosphere. A general term used to indi- cate distribution of water- vapour above the earth's surface. The limitation which is imposed upon the quantity of water-vapour in the atmosphere by the dependence of the pressure of saturation upon temperature, places the distribution of water- vapour on a different footing from 274 'Glossary. that of the other gases. The atmosphere is enriched with water- vapour by EVAPORATION at the surface and it is distributed by the process of CONVECTION, but that pro- cess does not extend beyond the TROPOSPHERE, and the water- vapour beyond that limit must be attributed to the action 6f diffusion. Above the surface saturation is pro- duced only by the reduction of temperature on rarefaction caused by convection, so that we cannot expect CLOUDS to be formed beyond the range of convection. Hence for all the ordinary purposes of meteorology which are concerned with the formation of clouds and other forms of precipitation, the water-atmosphere is limited by the boundary of the troposphere. Waterspout. The term used for the funnel-shaped tornado cloud when it occurs at sea. Waterspouts are seen more frequently in the tropics than in higher latitudes. Their formation appears to follow a certain course. From the lower side of heavy Nimbus clouds a point like an inverted cone appears to descend slowly. Beneath this point the surface of the sea appears agitated, and a cloud of vapour or spray forms. The point of cloud descends until it dips into the centre of the cloud of spray ; at the same time the spout assumes the appearance of a column of water. It may attain a thickness (judged by eye) of 20 or 30 feet, and may be 200 to 350 feet in height. It lasts from 10 minutes to half an hour, and its upper part is often observed to be travelling at a different velocity from its base until it assumes an oblique or bent form. Its dissolution begins with attenuation, and it finally parts at about a third of its height from the base and quickly disappears. The wind in its neighbourhood follows a circular path round the vortex and, although very local, is often of consider- Wares. 275 able violence, causing a rough and cimfused, but not high, sea. Water- Vapour .See AQUEOUS VAPOUR. Waves. Any regular periodic oscillations, the most noticeable case being that of waves on the sea. The three magnitudes that should be known about a wave are the amplitude, the wave length, and the period. The ampli- tude is half the distance between the extremes of the oscillations, in a sea wave it is half the vertical distance between the trough and crest, the wave length is the distance between two successive crests, and the period is the time-interval between two crests passing the same point. In meteorological matters the wave is generally an oscillation with regard to time, like the seasonal variation of temperature, and in such cases the wave length and the period become identical. If a quantity varies so as to form a regular series of waves it is usual to express it by a simple mathematical formula of the form y = a sin (/?. + ). Full explanation cannot be given here, it must suffice to say that the method of expressing periodic oscillations by one or more terms of the form a sm(nt + a) is known as 44 putting into a sine curve," u putting into a Fourier series," or as "HARMONIC ANALYSIS." See p. 311. Any periodic oscillations either of the air, water, temperature, or any other variable, recurring more or less regularly, may be referred to as waves. During the passage of sound waves the pressure of the air at any point alternately rises above and falls below its mean value at the time. A pure note is the result of waves of this sort that are all similar, that is to say, that have the same amplitude and wave length. The amplitude is defined in this simple case as the extent of the variation from the 276 Glossary. mean, while the wave length is the distance between successive maximum values. The period is defined as the time taken for the pressure to pass through the whole cycle of its variations and return to its initial value. Another good example of wave form is provided by the variations in the temperature of the air experienced in these latitudes on passing from winter to summer. This is not a simple wave form because of the irregular fluctuations of temperature from day to day, and the amplitude of the annual wave cannot be determined until these have been smoothed out by a mathematical process. Fourier has shown that any irregular wave of this sort is equivalent to the sum of a number of regular waves of the same and shorter wave length. In America u heat waves " and " cold waves " are spoken of. These are spells of hot and cold weather without any definite duration, and do not recur regularly. Waves of Explosions are among the causes which may produce a rapid variation of pressure which begins with an increase, and is followed by a considerable de- crease. The transmission is in the same mode as that of a wave of sound. The damage done by a wave of ex- plosion is often attributed to the low r pressure which follows the initial rise. In the same way the destructive effect of wind is sometimes due to the reduction of pressure behind a structure resulting in the bulging outwards of the structure itself in its weaker parts. Weather. The technical classification of different kinds of weather as given by the letters of the Beaufort notation, set out in detail in the Introduction, p. 10. Weather Maxim. A popular saying or proverb in connexion with the weather, sometimes expressed in rhymes. The best are the sailors' maxims which, at the Wmtlior Ma.rim. 277 Meteorological Office, whether rightly or wrongly, are associated with Admiral FitzRoy. The relation with modern meteorology is often easily apparent. First rise after low Foretells a stronger blow is quite characteristic of the passage of the TROUGH of a depression. If the wind backs against the sun. Trust it not for back ic will run is appropriate for the anticipation of a cyclonic depression in the Northern Hemisphere. Long foretold, long last, Short notice, soon past is also good meteorology in relation to travelling depres- sions. A useful essay might be written on the sailor's maxim (( noted by Sir G. Nares When the rain's before the wind Your topsail halyards you must mind, But when the wind's before the rain You may hoist your topsails up again. Some of the land maxims also represent fair conclusions from experience. If hoar frost come on mornings t.vain The third day surely will have Jain provides a fair indication of the gradual transition from Easterly to Southerly weather. A yellow sunset is regarded as a sign of stormy weather. Admiral FitzRoy's version of the maxim is " A bright yellow sunset presages wind ; a pale yellow wet.'' A voluminous collection of maxims and legends has been compiled by Mr. Richard Inwards, a former President of the Royal Meteorological Society, under the title of Weather Lore," Admiral FitzRoy was perhaps the J;tsr to 278 Glossary. attempt to draw up a scheme of weather prognostics according to the precepts of experience, as he was the first to introduce forecasts based on weather maps. Professor W. J. Humphreys of the United States Weather Bureau has given a physical explanation of many of the best known weather signs in the atmosphere. An examination of Mr. Inwards' collection makes it apparent that the weather wisdom of ancient saws does not lend itself to systematic presentation. Variants of the same maxim sometimes contradict one another. A large number have to do with the saints' days of the calendar, and so with seasonal variations. The St. Swithin's legend has obvdous reference to the transition from spring drought to autumn rainfall (see SEASON), and the fact that the hour of heaviest rainfalLin the year [in London] is the third hour of the afternoon in July. Many maxims are based upon the prevalent notion that every unusual occurrence is a sign of something to come. In modern days we prefer to regard the state of the crops and the behaviour of birds as the natural consequence of the past and present* not as the controlling cause of the future. No doubt, if the course of events in the physical universe is unique, that is to say, if the present is the only possible sequel to the past, then the relation of the future to the present is of the same order as the relation of the present to the past, and while we are looking for the one we may find the other. But what are offered as signs are obviously insufficient as causes. When we read Hark ! I hear the asses bray, We shall have some rain to-day, we are supposed to regard the braying, not as a cause of rain, but as an evidence of superior intelligence in the quadruped, stimulated by sensations which are too delicate Weather Maxim. 279 for our senses ; but as a matter of practice it is doubtful if any serious action was ever based on that intelligent expression of the emotions. Wedge. Short for wedge of high pressure : an exten- sion of a high pressure region, more or less in the shape of a wedge, that separates two neighbouring areas of lower pressure. See Plate XVI. Weight. The force with which the earth attracts bodies near its surface. In dynamics we distinguish between the amount of material substance which a body contains, and the force with which gravity attracts it, but experiments made by Galileo long ago led ultimately to the conclusion that apart from the resistance of the air, all bodies, large or small, are similarly affected by GRAVITY, so that every part of a composite mass is now recognised as separately affected by gravity, the result for the whole being simply proportional to the amount of material substance, irrespective of its particular nature Bodies immersed in a fluid, as a balloon in air, may rise, that is, apparently have less than no weight, because they dis- place fluid, air in this case, which weighs more than the bodies themselves. It should be noted that it is the weight of the heavier surrounding air that furnishes thrt driving force for the ascent of the balloon. Wet Bulb. An ordinary thermometer having the bulb coated with muslin that is kept moist. The evapora- tion cools the bulb, and makes the reading lower than that of a similar plain thermometer. See PSYCHROMETER. Whirlwind. A quite small revolving storm of wind in which the air whirls round a core of low pressure. Whirlwinds sometimes extend upwards to a height of many hundred feet, and cause dust storms when formal over ii dcs<Tt. 280 Glossary. Wind. The motion of the air. It appears certain that the general winds of the earth are maintained by the unequal warming of different parts of the earth by the sun, but the exact manner in which they arise, and the reason of their distribution, is not clearly understood. Some local winds may be explained, as, for instance, the wind accompanying the descent of an avalanche, or the land and sea breezes, but the problem of the general circulation is very much more difficult. In the open sea, away from the disturbing influence of the great continents, the general trend of the winds is as follows : Round the equator are light variable winds, and on either side to 20-30 north or south are to be found the TRADE WINDS, moderate in force from the N.E. in the northern and from the S.E. in the southern hemisphere. Further towards the poles in about latitude 40 and 50 there are winds blowing chiefly from westerly points, but by no means steady. They often reach the force of a gale. Concerning the polar regions comparatively little is known. The calm equatorial belt and the trade winds on either- side follow the movements of the sun, being furthest north in our summer, and furthest south in winter. There is at present no exhaustive analysis of the facts which have been collected concerning the force and direction of the winds of the British Isles. The diagrams on pp. 281-285 represent the monthly frequency of winds of different forces at Deerness (for the northern Area), at Holyhead (for the Irish Sea), at Scilly (for the mouth of the English Channel), and at Yarmouth (for the East Coast). A diagram is also given for the frequency of the wind in January at ten stations. The monthly average duration of winds of gale force at these stations is given under GALE. The real fact which these diagrams illustrate is that no Wind. 281 BEAUFORT NUMBERS 01T[2|3|4|5|6|7|8|9||0 METRES PER SECOND Monthly Average Wind Frequency at Deernes&, 282 Glossary. ' 1 2 ! 5 ' 4 ! 5 ! 6 ! 7 ! 8 ! 9 ! 10 METRES PER SECOND ~ Monthly Average Wind Frequency at Holy head, Wind. Monthly Average Wind Frequency at Scilly. 284 Glossary. Q_ < 30- 20 : 30- 20- : 10- BEAUFORT NUMBERS 01234 56783 10 METRES PER SECOND YARMOUTH BEAUFORT NUMBERS 0|l|2|3|4|5|6h|8|9|K) METRES PER SECOND I i 7 YEARS - Monthly Average Wind Frequency at Yarmouth. Wind. 285 v o c^ aJ i H-S 30,-H 20 10fi 30- 2 Of 30-H 20 10 30-H 20 iof BEAUFORT NUMBERS 0|l|2|3|4| METRES PER SECOND Average Wind Frequency in January for Ten Stations. Note : From an investigation lately completed it appears that on the average the observed wind of force (i is related to the theoretical or sjfeostrophic wind at various stations as follows : Valencia, 62 per cent. Aberdeen, 58 per cent. Scilly, 63 per cent. Yarmouth, 63 per cent. Spurn Head, 71 per cent. 286 Glossary. exact scientific meaning can be attached to the comparison of measurements of wind when the observations are made close to the surface of the earth. It is certain that near the ground or near buildings the velocity of the air is changing rapidly with height. It has for example recently been determined by special observations of wind quite near the surface that the velocity at 4 ft. is from 83 to 90 per cent, of the velocity at 6 ft. above ground according to the nature of the ground. The velocity doubles itself, more or less, within 500 metres ; the actual ligure depends upon the time of day among other things. So a measure of wind is as much dependent upon the exposure of the particular point at which the observation was taken as of the unrestricted flow of air in an unob- structed position. In recent years at the Meteorological Office we have found the wind computed from the isobars, the geostrophic wind, a much more, satisfactory standard of reference than the anemometer readings.* Further information about wind is given under the following headings : BEAUFORT SCALE, EDDY, FRICTION, GALE, GRADIENT WIND, GUST, ISOBARS, LINE SQUALL, SQUALL. See also references in the index to Tables on p. 357. Wind at the earth's surface is subject to considerable diurnal variation, being greatest in the early afternoon and least before dawn (see DIURNAL) : at the tops of mountains and presumably at higher levels generally the reverse is the case, as the following tabular summary of the observations at the top of the Eiffel Tower (300 metres high) clearly shows : * Strictly speaking*, in dealing 1 with winds at a considerable height, we should employ the system of isobars appropriate to that particular level. Wind. 28? <"O N O ** "*t~ ** vrt t*.vC coco t-* O LT O^' -^ ) -i T^ w O^ ^ t^O ' co r^O xr> rh co w xr>woo t^uo i.nvo O C^ "-i rfoo w O CFi O\ OOO 1 oo O O^oo O O t^ O O oo coo roo vrjao >-. co ooo >rh ONOO oo t>. fO-^OOO > >O | -'OOOOrO^ t> t>OO cOOO Cl rJ-VO ifl vT> rf- rf- O^> CT> w> O -^- ~t- -d-vjD vnt^c-i C^l^i-< ir>i-iOOO coi-i i-ioo r^oo >-* i^r^QO r^.vOUD*O xr>wO C^i^Oxrjmi-^H-i O^Tj-^c^criO ^ OO ON M x^oo mvo oo i-tvoci O O^ooo cor>.r^oo TJ-MCO covo Cfr 1-1 >-< O O"*OO t^. t^ t>. COOO O O LO<O t> O> CO C7>00 CO vn t^. iri co co O O G^ ^ O^> O\ O^> O^oo OO t> t~^ t~>. r> t>oo OOG^OOOOOO O^ -+O O covO Tt"vQ t^.M^i-iO-<O>O-i-i coo O >-" vri HH vrj COHH OOO t>.vO cO-n\rit-i ^-vO t>.t^O M rt-O CTi-^-OO C?> ^O . OoOOOl^-OOO rJ-CTiO^MVO covnw W rj-n-vo c^t^OJ O ^- O O"^ x> O O oo i~^ *- w wi x>. o\ -H ci -^- vn r->. ci ^ -i- ^ co co O oo oo co oo oo r>o '-o >-o vr>yp O O CT^O !-> r^- <>> r^oo oo oo oo oo _--,-._ w 1-1 noo oo O r >. t^ 1>.OO OO OO OO i O vnx^.oooo ->J~>^O t^vr>cor>.O vnO voi^.-ioo >-<oo O corj-w r-^o rj-coM i-iOO O O MO r^t^CNOt-i >-oo Thoo O\ O oo O m co cooo O O C^oo oo w>O i-< r>. r^ 1-1 C>o d O co ^ co O >- M -t O CTvC^COO CO >-0 . ON"-" W r}-l^r>.oO i-ir>.O HH TOrt--l- O O O O^<^O ^oooo i>.oooooooooooo ^<^O O O O O o o ooo opoo<^ocoooocoooooc^c^oo ooo o c^r^rj-inr-ric<j t^.-h-^-ThooO ^ooo VT>OSTJ-T>I-VO *-* t^O \C \n i/-j -f- voo O 10 co O^ TT ^\ r^ t^ O^> "'J-oo coo oo oo oo r- r OOOOOOOOOC^C^oooOOOQOO^O^OOOOOOO O ji o I 288 Glossary. Winter.- In meteorology the months of December, January, and February. Astronomically, winter com- mences on December 21st, and ends on March 21st. See SEASONS. Wireless Telegraphy has two types of application in meteorology, according as we deal with electromagnetic waves produced artificially or naturally. The former is exemplified by the wireless weather-reports from vessels at sea, enabling synoptic charts to be extended to ocean regions without the long delays otherwise involved, and by the distribution of meteorological information from high power stations. The other application is in the detection of distant thunderstorms. Very early in the historyipf Wireless Telegraphy it was found that lightning flashes* emitted electromagnetic waves capable of affecting the detecting device then in use the coherer. A coherer may be|made to actuate an electromagnet the armature of which carries a pen for recording on a revolving drum, so that. every lightning flash within 200 miles or more may thu? automatically record itself. In the modern wireless receiver the electromagnetic waves set up by lightning cause clicks in the telephone. It seems probable that a large proportion if not all of the irregular and trouble- some noises called atmospherics, strays, or X's, which are formidable obstacles in long distance wireless telegraphy may be referred to distant lightning. Zenith. The point of the sky immediately " over- head," or in the vertical produced upwards ; the opposite of nadir, which is the point in the sky below one's feet, or in the vertical produced downward beyond the earth's centre and out the other side. Zodiac. 289 Zodiac. The series of constellations in which the sun is apparently placed in succession, on account of the revolution of the earth round the sun, are called the Signs of the Zodiac, and in older writings give their names and symbols to the months, thus : March is associated April May June July August September October November December January February with Aries, the Ram. Taurus, the Bull. Gemini, the Heavenly Twins. Cancer, the Crab. Leo, the Lion. Virgo, the Virgin. Libra, the Scales. Scorpio, the Scorpion. Sagittarius, the Archer. Capricorrius, the Goat. Aquarius, the Watercarrier. Pisces, the Fishes. Owing to precession, the position of the sun relative to the above constellations has altered a good deal since classical times. The sun now enters Aries in April. Zodiacal Light. A cone of faint light in the sky, which is seen stretching along the Zodiac from the Western horizon after the twilight of sunset has faded, and from the 'Eastern horizon before the twilight of sunrise has begun. In our latitudes it is best seen from January to March after sunset, and in the Autumn before sunrise. In the TROPICS it is seen at all seasons in the absence of moonlight. It is supposed to be due to the reflection of sunlight from countless minute particles of matter revolv- ing round the sun inside the Earth's orbit. Its light is usually fainter "than that of the Milky Way. 13204 K 290 Glossary. SUPPLEMENTARY ARTICLES. Absolute Humidity. Aqueous vapour has a very large annual variation but a small diurnal variation, whether one considers the amount of vapour present, or the contribution which it makes to the atmospheric pressure. This is readily seen in consulting the two accompanying tables, which give respectively the quan- tity and the pressure of aqueous vapour at Richmond (Kew Observatory). According to the first table the quantity of aqueous vapour in the atmosphere in the hottest months, July and August, is nearly double that in the coldest months, January and February. The quantity is slightly greater in the afternoon than in the morning hours, but the excess of the afternoon maximum over the morning minimum represents only 6 or 7 per cent, of the mean value. As the second table shows, the mean vapour-pressure in July and August is fully double that in January and February. The diurnal variation is also a little more marked than it was for the quantity of vapour. The maximum pressure in the afternoon occurs decidedly later in the day at midsummer than at midwinter. The morning minimum, on the other hand, occurs a little earlier in summer than in winter. Absolute 291 j -utWK f; : f ; ^ & b b & t 5 t "^ ;f S E S C & b b b\ t~x <b <^> K * CO .**. r "f 5* .* f "* b bs R 2" * -si dc 5 $ S & b b bs <b vi it c; c t 5 c a i rjT~rj 5ni SP<M ; ^ In i* b K ON bs K sb '* ^ j " ^ e: c; z t s ? b b, d i ^ t r < "** < y^ ^3 q-i w M go . ^ ^ <b K ON b O OS OO vO *f\ ^H ~ "&? b "os oo o v> 0^ m O jL, ! - """ E E t5l 2" ?. b\ b rn .** P t^ 00 h O ON r> <3 <^ 'K 22 ~ E E 5 S b b bs oo b !*^ > | 2 ^J M - - - *> - -o 2 b a, K - U K w, ^ ^ b r* bs O Os OO *O v> R ? 2 ^ ^ b K & b o 5 K o U, - g 5 '- - - - K -o b o bs K b ^ - 00 k- OH /y* ^ Vr, Jo <5 K. bs b OS K- b ^ i^ H M f Tj- * M 02 s ^ * "* ^ " ^ * 2 o bs K so *^ o 1 * -, Vo U ^ K 60 o b oo K sb ^ K is 2 5 , - r* .'" r^ p^ . M 9 P 6-S' : C r 5 | M a | : ! N PI rr\ p -> I>N ON n , v \b i- oo bs b bs K sb ^ t Q *" "" a s; s 5 c s b b\ K ff C H H ^*- M N 00 pS a s. ^ c I i t 5^ : ^ ^ ^ o K do o ON r^ so " S ' ^ i ^ & w _o 2 'c t>> s >> s | a 5 a 3 1-5 EL, ^ <! g KJ t- bb "S ^J > o 3 LI * 1 a $ X 13204 292 Glossary. 'A*CI * NO '- OC * 2 JT JT 2 o oo ** ON NO Nb K g bN "M g g s ON K NO ON 5 1 NO NO . o BO NO M 00 00 s K ON 3 00 NO 00 NO r^ CO 00 C4 p p Vj H cc K Os 5 oo NO oo NO K oc "ON Vi "<* "* s b ro do K oo bs 3 ON NO 00 NO N 00 ON >^ O M ON Vl j* Vt- ? b -h CO g, g 2 s ON NO ON NO K b H Vi- V*- vT b cc K ON ON oo ON NO O ** M W^ p P b H "* V 00 b oc r Os ON ON NO fc cc O Vl Vh rr> i >.n b oc K ON ON NO . K. -t- K io M * OS 00 s i r^ CO K e M r p T*- ff M rh y\ oo ^ g IT Tf Os ON S K p 2 b J & & | b g ON JJ K K 00 b J C: t s cc K Os ON M e, e. j Jo o <^ *o t^ NO :-> K OS ON e s K rr> ic 2" ^ j oc r ON ON O 00 NO s K 00 s ^ ? H oc K OO *Os I ON Nb NO ON ^b cc oo N t^ ON ? b do p S oc 1 g NO NO ON NO C50 r^ M r>, oo g 00 "ON do NO ON ^ NO NO NO K s S? 1? 00 NO ON o co NO bs j ' g NO NO K NO & Os NO ON " NO NO NO >0 K SN -M -J & s 5 oc O ON * NO NO NO oc NO K M ^- _, M ON M w rn ^ "CN do OS NO ON Ml NO s NO v^ S "S 2. v^ 00 ON do Os NO ON " NO i ON NO ON S' 'S -J S p oo ON do O b. ~" * NO O K 3 1? ? 5 R s. & bs (H 1 I 1 B 3 O ~ $ 3 K? r^ <J i CO | I 6 Q ' I Accumulated Temperature. 293 Accumulated Temperature. A term used to de- scribe the excess or defect of temperature in relation to a selected base-level, prevailing over a more or less ex- tended period of time, e.g., a week, a month, or even a year. Accumulated temperature is employed mainly in connection with agricultural statistics, and the base-level adopted in this and in most Continental countries is a temperature of 279a, equivalent to 42F., or 10 above the freezing point. This temperature was suggested many years ago by Prof. A. Candolle, an eminent Swiss physicist, as the level above which the growth of vegetation commences and is maintained. Temperatures above 42F. may therefore be regarded as effective, inas- much as they tend to the promotion of active plant - growth. Temperatures below 42F. may be regarded as non-effective at the best, and at certain seasons of the year, when the defect is large, as positively injurious. In the Weekly Weather Report the amount of effective and non- effective heat is expressed in what are described as u day- degrees." A "day-degree " Fahrenheit signifies 1 above or below 42F. continued over a period of 24 hours, or, in inverse proportion, 2 continued over 12 hours, 3 over 8 hours, and so on. At the Meteorological Office the amount of Accumulated Temperature above and below 42F, is computed from the daily readings of the maxi- mum and minimum thermometers, in accordance with formulae proposed more than 30 years ago by Sir Richard Strachey, at that time Chairman of the Meteorological Council. The actual method of computation is described in the preface to the Weekly Weather Report for 1884 and subsequent years. As examples of the results attained, the following statistics may be of interest. In the table the amount of Accumulated Temperature above 294 Glossary. and below 42F. is given respectively for the exceptionally warm summer of 1911, in contrast with the cool summer of 1907 ; and for the cold winter of 1916-17 in contrast with the mild winter of 1912-13. The period embraced is, in each case, the 13 weeks comprised as nearly as may be within the three months June to August and December to February. The winter of 1916-17 was, it need scarcely be said, prolonged far beyond the ordinary winter boundary. The averages with which the actual results are compared are those for the 35 years 1881-1915. ACCUMULATED TEMPERATURE. Above 42F. Below 42F. Day degrees. Difference from average. Day degrees. Difference from average. Summer of 1911 ... 1,940 + 297 ?? > Winter of ?> ?> 1907 ... 1916-17 1912-13 M23 54 278 220 -126 + 98 598 '85 +*37 -176 Atmospheric Electricity treats of the various elec- trical phenomena observed at or near the earth's surface. If we regard the earth as a sphere of radius R, carrying an electrical charge of uniform surface density <r, the elec- trical force on unit charge at any radial distance / not less than R is the same as if the whole charge ^vR 2 were Atmospheric Electricity. 295 collected at the centre, and is thus C x 4:7rrrR*fr 2 , where C is a numerical constant depending on .the units adopted. On the electrostatic system C is unity. Charges of like sign repel one another, and it is usual when talking of electrical force to regard it as the force experienced by a positive unit. Thus the above force is upwards or down- wards according as <r is positive or negative. In fine weather v is normally negative and so the force is down- wards. The force at any surface-point is really deter- mined by the charge in its immediate vicinity, and thus whether * be uniform or not the force is given correctly by 47To-(7, where fa is the surface density at the point. If we take a point at a height k small compared with R and suppose p to be the mean electrical charge per unit volume throughout the height /*, we find in a similar way that the electrical force at the point is 4;r(7 (tr + pJi). If F denote the attraction towards the earth on a unit charge at height h, in order to raise it to a slightly greater height h' an amount F (h' h) of work must be done, which is transformed into a rise of potential (i.e., capacity to do work). If the potential rises during the operation from V to V, then F(h'-h)= V - V. Thus the force downwards is F = (V - F)/(fc'-A). It is usual to consider the change of potential per metre of height, which is known as the potential gradient. If we write P for ( V i - F)/(A' A), we finally get where P,, and P h are the values of the potential gradient at ground level and at height h respectively. A is here a numerical constant, equal to 4?r in the electrostatic system of units. P is positive, i.e. the potential increases as we leave the ground, if (the force on a positive charge is directed down- 296 Glossary. wards, i.e. if <r is negative. This is almost always the case in fine weather. As we go up, P will increase if p is nega- tive like <r, but diminish if p is positive. In fine weather P diminishes as we go up, or p is positive. At Kew Observatory the potential gradient in fine weather averages about 300 volts per metre. At most other stations somewhat lower values have been observed. The potential gradient has a large annual variation, being lowest in summer. There is also a large diurnal variation, with usually two maxima and two minima in the 24 hours. At Kew the lowest value occurs in the early afternoon in summer, but in the early morning in winter. As we go up from the earth, the potential gradient normally diminishes, i.e. each successive metre adds less to the potential, but all contributions being in the same direction the potential goes on mounting, and at the levels attained by aircraft may reach hundreds of thousands of volts. Any body remaining at one level gradually assumes the potential of the surrounding air, the process being accelerated if the body is provided with sharp edges or points, or with an engine emitting fumes, or if it is discharging ballast. But if a body makes a large rapid change of level, it may depart widely in potential from the surrounding air, and in an extreme case this miy lead to discharge by sparking and consequent danger to an airship. Besides its regular changes, potential gradient near the ground shows numerous if not perpetual irregular changes. Clouds sometimes carry large electrical charges, and heavy passing clouds usually cause large fluctuations of potential. During rain the potential gradient at ground-level is often negative. The charge in the air evidenced by the change in Atmospheric Electricity. 297 potential gradient with height may be carried by drops of water, but it is also carried by ions. There are always at lea'st two kinds of these present in the atmosphere, usually known as light (or mobile) ions and heavy (or Langevin) ions. The latter seem the more numerous, especially near smoky towns, but they move very slowly, and as carriers of electricity are relatively unimportant. Also the numbers of heavy ions carrying positive and negative charges seem approximately equal, thus they- neutralise one another so far as influence on the potential gradient is concerned. On the other hand there is usually a marked excess of light positive ions, as suggested by the falling off of the gradient as height increases. The number of light positive and negative ions combined near ground level is usually of the order of 1,000 per c.c. Air is often regarded as a non-conductor of electricity, and it is a very poor conductor compared with copper ; it conducts, however, to a certain extent. The negative charge on the earth and the positive ions in the atmosphere attract one another, a process equivalent to the passing of a current from the air to the earth. This current is extremely small if reckoned in amperes per cm. 2 only in fact about 2 x 10~ 16 on the average but for the earth as a whole this means fully 1,000 amperes, if we can accept the few stations where observations have been made as fairly representative. The process by which this current is maintained is at present a mystery. The difficulty is analogous to that which puzzled philosophers who saw rivers flowing into the sea without the sea becoming fuller. Two explana- tions which seemed promising, though neither suggested a sufficiently regular source of supply, appear to have broken 298 Glossary. down. It was suggested that rain might restore the balance by bringing down negative electricity. But the observations hitherto made too limited as yet perhaps to be wholly conclusive indicate that while rain not infrequently brings down negative electricity, it brings down on the whole more positive. Another suggestion was that the balance might be restored by lightning. Mr. C. T. R. Wilson has, however, recently devised a method of determining the sign and the quantity of the electricity brought to the earth by a lightning flash, and his results suggest that on the whole the charge brought down is at least as often positive as negative. Observations made from balloons indicate that at heights above 1 or 2k a rapid increase takes place in the ionisation of the atmosphere. At heights of 90 to 150k AURORA is a frequent phenomenon in high latitudes, and it is natural to suppose that at such heights the electrical conductivity is much greater than near the ground. Aurora takes a great many forms. In addition to arcs, bands, rays and isolated patches, there are some- times displays resembling curtains or draperies, also so- called " coronae," representing a concentration of rays directed towards a limited space of the heavens. Of the accompanying figures, one shows an arc, the other a curtain. The arc is the most symmetrical and stable form of aurora, sometimes persisting with little visible variation for several minutes. Auroral curtains, when seen in the zenith in high latitudes, seem very thin in the direction perpendicular to their length, und are usually in rapid motion. The lower edge, both of arcs and curtains, is usually much the best defined. Aurora is very rare hi the south of Europe, and is but To face p. 298. P tf .tf 5 ^ 3204 To face p. 291). PQ fe H K <D tf H PQ O .0 >. . H d PH O O CM fi Aurora. 299 seldom seen, in the south of England. But the frequency increases rapidly as we go north, and in Orkney and Shetland Aurora is a comparatively common phenomenon. A /one of maximum frequency passes from the North of Norway to the South of Iceland and Greenland. Aurora is also common in high southern latitudes, but its distri- bution there is still imperfectly known. The visibility of weak aurora is much affected by moonlight, and even the strongest aurora seems to be invisible if th-3 sun is above the. horizon. Thus there is some difficulty in assigning definite laws for its frequency of occurrence. In the British Isles and similar northern latitudes it is most common in the late evenings and near the equinoxes ; but in the north of Norway and Greenland it appears to be most frequent near mid-winter. Of late years Pro- fessor Stormer has devised a method of photographing aurora, includins 1 reference stars in the photograph, and by means of photographs taken simultaneously at the two ends of a measured base he has obtained numerous results for the altitude. Heights exceeding 200 k. are not unusual, but a great majority of the heights lie between 90 k. and 130 k. Aurora is undoubtedly an electrical discharge, and we thus infer that at heights of 100 k. at least in high lati- tudes the atmosphere must often, if not always, be a vastly better conductor of electricity than it is near the ground. Some distinguished travellers have claimed to see aurora come down between them and distant moun- tains. If this be the case, aurora must occasionally come down to much lower levels than any measured by Stormer, and there may be truth in the belief held by some meteorologists that cirrus cloud is sometimes the seat of aurora. 300 Glossary. When visible in England, aurora is nearly always accompanied by a magnetic storm, but this is not the case when it is confined to high latitudes. The spectrum of aurora consists of a number of lines, one of which in the green is particularly characteristic, bat it has not yet been identified with certainty with that of any known gas. It is not at all improbable that eventually we may learn much from the spectrum of aurora a~s to the constitution of the atmosphere at heights far greater than those accessible to balloons. Boiling-points. It is customary to Bay that a thermometer is graduated so that the freezing-point and boiling-point of water com^ at specified figures, e.g., 32 and 212 on the Fahrenheit, or 273 and 373 on the abso- lute scale, and the importance of defining the pressure under which the water is boiling is frequently over- looked. In the process of boiling, bubbles of vapour are formed in the interior of the liquid. If the pressure of the air above the liquid is low the bubbles are formed and grow more readily, i.e., at a lower temperature than when the pressure is high, so that the boiling-point is lowered by decrease of pressure, raised by increase of pressure. Except on rare occasions pressure at sea level in England is between 1,040 mb. and 960 mb. Under the pressure 1,040mb., water would boil at 373*73 a, under 960 mb. at 371-49 a. Under 1,000 mb. the boiling-point is 372*63 a. The standard pressure used in the definition of the temperature 373 a (or 212F.) is that due to 760mm. of mercury at the freezing point ol ! water at sea level in latitude 45, i.e., 1,013*2 mb. When we leave sea-level and ascend to places where the pressure is much lower, the temperature at which Boiling-points. 301 \vater boils is reduced. For example, at Pretoria, 5,200 ft. above sea-level, the average pressure is 871 mb., corre- sponding with a boiling-point 369a. This change in the boiling-point in the course of an ascent is mado use of in the measurement of heights by the HYPSOMETER, a method which is convenient, as the apparatus required is more portable than a mercury- barometer^ and more reliable than an aneroid. At 2 ,600 f t. r the greatest height reached by the Duke of the Abruzzi on the Himalayas, the pressure would be about 420 mb. and the boiling-point 350a. At great altitudes there ,is much difficulty in cooking, owing to the comparatively low tem- perature of the boiling water. It should be mentioned that the figures which have been given above refer to the boiling of pure water. The addition of salts in solution raises the boiling-point. Sea water of average density boils at about 0*5a. above pure water under the same pressure. By adding 40 grammes of salt to 100 grammes of water the boiling-point can be raised by 9a. Calorie or gramme-calorie i.e., the heat required to raise the temperature of 1 g. of water by la. at 288a. is often used in connection with measurements of solar radiation. The amount of solar radiation is accordingly often given as so many gramme-calories per square centimetre per minute. At the Meteorological Office we use instead the milliwatt (per square centimetre), because that is the accepted unit of "rate of working" in Thermodynamics. Even in the total absence of cloud the amount of solar radiation which reaches a given area on the earth's surface at a given time is very variable. It depends on the altitude of the sun and the transparency of the earth's atmosphere. Thus it 302 Glossary. is by no means a simple thing to estimate the ' Solar Constant," i.e., the radiation which would be received in one minute by a square centimetre outside the earth's atmosphere, at the mean distance of the earth from the sun, the incidence of the radiation being normal. From a long series of investigations, Dr. Abbot, of the Smithsonian Institution, Washington, U.S.A., has arrived at 1*93 calories per square centimetre per minute or 135 milli- watts per square centimetre, as the mean value of the solar constant. Dr. Abbot concludes, however, that solar radia- tion, outside the earth's atmosphere, is not really constant, the fluctuations being presumably due to changes in the solar atmosphere. Contingency. In expressing the relationship between two variables, the correlation co-efficient " r " can only be used when both variables are given quantitatively. The correlation ratio rj can be used when one or both variables are expressed quantitatively. If, however, both variables are given qualitatively, it is not possible to use either the correlation co- efficient or the correlation ratio. In such cases it is usual to calculate the co-efficient of mean square contingency, Ci. This co- efficient Ci is of the same nature as the correlation co-efficient r. The method of calculating d can be obtained from any text-book on the Theory of Statistics". An excellent astronomical example is given on page 167, " The Com- bination of Observations," Brunt (Camb. Univ. Press), in which it is shown that there is a close relationship be- tween spectral type and the colour of stars. See Correlation. Correlation Ratio. A measure of the relationship * An Introduction to the Theory of Statistics by G, Udny Yule, Correlation Ratio. 303 between two variables. In calculating a correlation co- efficient it is assumed that the regression is linear, i e., that the observations themselves, or the means of grouped observations when plotted, lie approximately along a straight line. In the calculation of the correlation ratio iy, linear regression is nob assumed. The correlation ratio may be denned as a generalised coefficient which measures the approach towards a curved linear line of regression of any form. If /;, r be respectively the correlation ratio, and correla- tion coefficient, found from a set of observations, // 2 -r 2 is a measure of the deviation from linearity of the curve of regression. See Computer's Handbook, pp. V 29-52. Duration Of Rainfall. A climatic feature of some importance. Many forms of self-recording raingauge have been designed to determine the distribution of rain in time, but it is only in recent years that adequate atten- tion has been paid to the matter, and the pattern of gauge has not yet been standardised in this country. During the 34 years, 1881-1914, at the headquarters of the British Rainfall Organization, in North-west London, the average annual duration of rain (snow and other forms of precipitation included) was 433 hours, about seventeen days, or 5 per cent, of the year, the extreme annual values ranging from 299 hours to 689 hours (12| days to 29 days, 3vr per cent, to 8 per cent.). In the wet regions ot Cumberland, Wales and Western Scotland, the annual duration exceeds 1,000 hours (11*5 per cent, of the year) and in the wettest parts in the wettest years sometimes approaches 2,000 hours, or between a quarter and a fifth of the year. The duration values for the year 1915 (a wet year) are -iven for the Observatories of the Meteorological Office, 304 Glossary. together with the total amount of rainfall and the mean rate of fall per hour : Rainfall. Duration. Rate per hour. mm. hrs. mm. Kew ... 805 461 1*75 Falmouth 1,322 744 i'7 8 Eskdale 1,224 790 i "55 Aberdeen . . . 805 589 1*37 Valencia ... 1,516 761 1-98 Armagh 741 530 1-40 The highest duration value for the year 1915, accord- ing to tables set out in the annual publication British Rainfall was 1,303 hours, or 15 per cent., near Kinloch- leven, Argyll, a few miles S.E. of Ben Nevis, and the lowest value 442 hours at Nottingham. In the London district the mean rate of fall increases with the mean temperature, reaching its maximum in July and minimum in January. The actual duration, on the average, is least in July and September, when it is little more than 25 hours, and greatest in November, when that figure is nearly doubled. The CORRELATION COEFFICIENT between annual amount and annual duration of rainfall for 35 years in N.W. London is +'81. Dust. The atmosphere is permeated by dust up to great heights ; the dust may be derived from deserts, -or from any dry surface, from evaporated sea-spray, from Dust. 3()f> plants in the shape of pollen grains, from the debris of meteorites, from volcanoes. Dust is important mete- orologically in that water vapour condenses on dust particles. Dust -free air may be cooled considerably below the dew-point without condensation occurring. Dust Counter. A.itken has devised three types of instrument for estimating the number of dust-particles in a sample of air ; a large model with artificial illumina- tion, suitable as a permanent part of the equipment of a first-class observatory ; a portable model suitable for testing locally polluted ah, e.g., in sanitary work ; and a pocket instrument for use in country districts where the variations in the dustiness are small. This last instrument will be described briefly, as it embodies the same principle as the others, but with fewer working parts. Into a receiving chamber, lined with moist blotting paper, a measured quantity of the air to be tested is intro- duced. This chamber contains a horizontal glass stage, having fine cross lines 1 mm. apart etched upon it so as to divide the surface into a network of squares. A sudden reduction of the pressure of the saturated air in this receiver, which is effected by means of a pump, causes the aqueous vapour to condense upon the dust-particles, and the small raindrops so formed fall upon the glass stage. The average number that fall upon one of the small squares is then counted, with the aid of a lens let into the roof of the receiver, and so an estimate of the number of dust-particles in 1 cc. of the air can be made. The method of measuring the volume of the sample of air ijs ingeniously simple. When the piston is drawn down in making a stroke of the pump, the air in the receiver expands by a fraction which is read off upon a scale 306 Glossary* engraved upon the barrel of the pump for this purpose. If now the receiver is put into communication with the outer air, a sample of air having this volume enters and restores the pressure inside to that of .the outside air. The complete instrument can be packed in a box 4| x 2^ x 1 inches and weighs barely half-a-pound. The following table shows the number of dust- particles found by Aitken in 1 cc. of air at various localities : Cannes (April) ... 1,500150,000 Simplon Pass (May) 50014,000 Summit of Rigi (May) 200 2,35( ) Eiffel Tower (May) ... ... 226104,000 Paris (Garden of M.O.) (May) ... 134,000210,000 London (Victoria Street) (June) ... 48,000150,000 Dumfries (Oct.-Nov.) 39511,500 Ben Nevis (August) 335 473 False Cirrus." False cirrus may be defined as a type of cloud resembling cirrus but occurring at lower altitudes. It consists of snow, may occur at any height, and may be divided into two main classes : (1) Consisting of isolated tufts of large masses of considerable height. (2) Spread out in extensive sheets. Type (1) is commonly seen on the tops of showers. The rounded top of a cumulus, consisting of minute particles causing diffraction rings, often turns to false cirrus; this may afterwards be carried for considerable distances both vertically and horizontally. As the shower dissolves away, the false cirrus may remain for some time * Contributed by Lieut. C. K. M. Douglas, R.F.C. False Cirrus. #07 afterwards, and may consist of white tufts or dull grey masses, with edges resembling cirrus. It occasionally forms direct, and not from other clouds. For instance, on April 1st, 1917, near Edinburgh large tufts of false cirrus formed at a height of about 6,000 feet, apparently caused by convection currents. Snow showers resulted but only a few flakes reached the ground. Type (2) is distinguished by its more regular formation and covers large areas. It sometimes has a dull uniform grey appearance, being then usually classed as " alto- stratus " ; it has then often very great thickness. Some- times it assumes the well-known cirrus form, resembling cirro-stratus sheets, but never causing a halo. No very definite dividing line can be drawn between "false cirrus " and cirro-stratus. False cirrus sheets may change to thin wavy clouds resembling cirro-cumulus, causing diffraction rings ; this most often happens in the evening, and the clouds may afterwards dissolve away entirely. Dense masses of false cirrus from the south with surface winds from the north-east often precede heavy thunder- storms. Glacier. A river of ice flowing slowly but irresistibly down the valleys of those regions which have a perpetual supply of snow to feed the head of the glacier. The explanation of the gradual flow of ice down valleys under the action of gravity, forms a special section of physics and is another illustration of the peculiarities of the material of which water is composed. Glaciers are not only of climatic importance but in dynamical meteorology, with rivers, they deserve consideration as showing the line along which air flows when the excess of density cine to cooling is the primary reason for the movement. #08 Glossary, From the analogy we may conclude that a gully is no protection against katabatic winds but rather the reverse. Gravity relates to the attraction between material bodies. The law of universal gravitation is that every mass attracts every other mass with a force which varies directly as the product of the attracting masses and in- versely as the square of the distance between them. It is convenient to regard the attracted "body as of unit mass. The law then implies that the force exerted is independent of the temperature or velocity of the attracting body. Both these conclusions have been attacked of late years, but it is not questioned that they are sufficiently exact for meteorological purposes. It is easily shown mathematically that a sphere whose density varies only with the distance from the centre attracts an external body exactly as if the whole ma^s were collected at the centre, and that a similarly consti- tuted spherical shell i.e., a mass bounded by two con- centric spherical surfaces while attracting an external body as if its mass were collected at the centre, exerts no attraction at any internal point. Let us apply this to a point in the atmosphere at height h above the ground, regarding the earth as a perfect sphere of radius 72, and assuming the density, whether of the earth or the atmo- sphere, to vary only with the radial distance. The atmo- sphere outside the spherical surface of radius R + h exerts JU i. no attraction, while the earth's mass M' within the surface * of radius R + h attract as if collected at the centre. Thus the attractive force is G(M + M')l(R + h) 2 , where G is a constant. This becomes g (1 + M'jM) (l + /i/#)~ 2 , where c/ =GMjR- is the corresponding force at the earth's sur- face, i.e., wholly within the atmosphere. Counted 171 Gravity. 309 kilogrammes M' is large, but even if we went to the confines of the atmosphere M\M. would be less than 1/1 million. Thus the attraction of the atmosphere may be neglected, at least for meteorological purposes. The variation with height in the earth's own attraction is much more important. At all heights attained by balloons we may neglect ft 2 /jR 2 , and so replace (l + /*/jR)~- by 1 -2A/R. But at a height of say 10 miles this repre- sents a reduction of one part in 200 in gravity. In reality the earth is not a sphere, but approaches to a spheroid whose equatorial radius is 10*7 k. longer than its polar radius. Also it rotates, and the " centrifugal force " due to the rotation reduces gravity, especially near the equator. The earth's surface is also irregular in outline, and the density variable, at least near the surface. Thus the formulae actually advanced to show the variation of gravity at different parts of the earth's surface are com- plicated. The following formula, due to Helmert, is perhaps the best known. In it g denotes the acceleration of gravity in C.G.S. units, i.e., in cm/s 2 : =978- 000(1+ 0-00531 sin 2 </>) x 2h 37* c: It' (6-0) \ " K 2/2 A 2./2A y ) Here <p is the latitude, R the earth's mean radius, h the height above sea level, h' the thickness of surface strata of low density,. A the earth's mean density (approxi- mately 5'6 times that of water), c mean density of surface strata (usually taken as 2*8), the actual density of surface strata for the region, and y 3.10 Glossary. a so-called orographical correction arising from neigh- bouring mountains and so on. At sea level, supposing 3=0, and tj negligible, this becomes =978-000 (1 + 0-00531 sin 2 f), or more conveniently </= 980 '5966 2*5966 cos 2<l>. Thus g has its mean value where cos 2< vanishes, i.e., where 2^=90, or </>=45. This explains why it is usual to reduce gravity to latitude 45. This means reducing some measure actually made e.g. of the height of the barometer to what it would have been if gravity had possessed its mean value. The formula does not, of course, imply that gravity has the same value at every spot in latitude 45, irrespective of its height above sea level or other local peculiarities. The determination of g absolutely at any spot with the precision which the formulae suggest is extremely difficult, but relative values of /, or, differences between its values at different places, can be determined with very high pre- cision by means of pendulum observations. If t and t' be the times of oscillation of a certain pen- dulum at two stations, the corresponding values g and g' of gravity are connected by the relation g'\g(t\t'}*. This enables gravity at any station to be determined in terms of gravity at a base station. For accurate work corrections have to be applied to the observed times of swing to allow for departures of temperature and pressure from their standard values, also for chronometer- rate and flexure of the pendulum-stand. When all the known corrections are carefully made a very high degree of accuracy is obtainable. For instance, taking 981*200 as the value of g at Kew Observatory, this being the value accepted for the purposes of the Trigonometrical Survey Gravity ,, ,3il of India, the last two comparisons instituted between Greenwich and Kew, the one made by the United States Coast and Geodetic Survey, the other by the Trigono- metrical Survey of India using two different sets of half-second pendulums, gave for g at Greenwich the respective values 981-188 and 981*186. - Pendulum and other geodetic observations have led to a theory of isostasy which has received strong support of late years, especially in the United States. According to this theory if we start at about 100 k. below sea level we find between there and the free surface an approximately uniform quantity of matter irrespective of whether the free surface is mountainous or not. A lesser density under lofty mountains and a higher density under deep seas act as compensations. While the mass of a body is independent of its position, its weight, i.e., the gravitational attraction exerted on it varies with g and so increases as we pass from the equator towards either pole. Denoting by g<p the value of g in latitude </> we obtain from the formula ^0=983-19, 00=978-00. In other words, gravity at the poles exceeds gravity at the equator by 1 part in 189. Harmonic Analysis. There are many meteoro- logical phenomena which recur with some approach to regularity day by day. If the changes of such a variable us temperature are represented by a curve, then the portions corresponding to successive days bear a strong likeness to one another. If for the actual record for each day the record for the average day were substituted, the variation for u long period would be represented by a 31;! Glossary. c^p-KEW OB5ERVATORY, RICHMOND, SURREY -*^> -NORMAL DIURNAL VARIATION IN TEMPERATURE IM MARCH a ZB2 280- figure i jf "rom Hourli^ Mean i a 282 NSET 278 22 282 280 278 \^ 278 276 294 at 280 278 276 *""-* ^.^^^ SUNRISE .s ^^ , , i , , i i i i i | , 1 , , , , 1 i , 'IRST TERM OF HAR FIRST TWO Figure 2 .* ^lOINIC SERIES ^ /^^^ ...^ "^x^ ''Ti^""---^ ,*_-' .-''/^ ^^: , i i , , , i 1 . i i i 1 i i NORMAL DIURNAL VARIATION IM TEMPERATURE IN JULY. 292 290 288 286 / P^~~ \ 200 288 286 ~/_ \ SUNSET TV, / ~\^^ SUNRISE 7 t 1 1 I 1 1 1 1 1 ! .III! 1 1 1 1 1 GMT 2 4 6 8 10 12 14 16 18 20 22 24 Hariri unit' Ann.lt/s-is. 313 curve in which the part corresponding with each day was like its fellows. The simplest curve possessing this property of continuous repetition is a curve of sines. As an example the variation of temperature at Kew Observatory, Richmond, in July, may be cited. The sequence of change throughout the average day is shown in the lower part of the figure. The representative curve is not unlike a curve of sines, but it is not quite symmetrical. The rise which commences at sunrise and lasts until after 15 h. is more. steady than the drop which is rapid in the evening and slow after midnight. A good approximation to the temperature on the average day is given, however, by the expression 289 '9 + 3 '7 sin (15*+ 224^) where t is the time in hours reckoned from midnight. It will be seen that the lowest value is reached at the time given by 15-f2247j:=270 i.e. at 3 h. 6 m. and the maximum comes 12 hours later at 15 h. 6 m. The substitution of a sine- curve for the curve based on the observations would make the minimum too early by an hour, but would not affect the maximum so much. The curve showing the diurnal variation of temperature in March (in the upper part of the figure) is not so near to a sine-curve as that for July, The rise in temperature from minimum to maximum takes little more than six hours. The best sine-curve for representing the variation is given by the formula 0=278 -74 + 2 -47 sin (15^ + 222) and is shown by the broken line in the figure. It will be seen that the agreement is by no means close. To obtain a more accurate expression for the temperature, an additional sine term with a period of 12 hours may be 314 (jlossary. introduced. The best formula containing such a term is 0=278 '74+ 2 -47 sin (15* + 222) + 0-63 sin (30* + 39) The new term 0*63 sin (30* + 39) is positive in the early morning and in the early afternoon, so that it delays the drop to the minimum and makes the maximum earlier. In Figure 2, the continuous curve which corresponds with the proposed formula crosses the simple sine-curve at intervals of six hours. The resemblance to the curve based on the observations is greatly improved. A closer resemblance would be obtained if additional terms 0-08 sin (45* + 330) + 0-12 sin (60* + 190) were included in the formula. The harmonic representation of a diurnal inequality may be expressed in either of the alternative forms r/i cos (15 xf) +a 2 cos (30 x *) +a s cos (45 x *) +a 4 cos(60x*) + . . . + &! sin (15 x *) + 6 2 sin (30 x *) + & 3 sin (45 x *) + 6 4 sin (60 x*) + . . . , Pi sin (15 x * + Ai) + P 3 sin (30 x * + A 2 ) + P 3 sin(45x* + J 3 ) + P 4 sin.(60x*-M 4 ) + . . . , where t denotes the time in hours counting from some fixed hour, usually midnight. The latter is the form which has been adopted in the previous part of this note, as it best exhibits the physical significance of the results, but the first form is that employed for the actual numerical calculation of the harmonic coefficients. We first calculate the a, b coefficients and then derive the P, A coefficients from the relations where n may be 1, 2, 3, 4, &e. Harmonic Analysis. 315 For brevity, let 0, 1, 2 .... 23 represent the algebraical departures from the mean value for the clay of an element such as temperature or pressure at the successive hours midnight (or 0), 1, 2 .... Then using the following closely approximate values 0*966 for cos 15 or sin 75, 0-866 for cos 30 or sin 60, 0-707 for cos 45 or sin 45, 0*259 for cos 75 or sin 15, and noticing that 0*5 is the exact value of cos 60 or sin 30, we have the following mathematical expressions for the a, b coefficients of the first 4 orders : 12ai=(0-12) +0-966{(i + 23)-(ll + 13)} +0-866 {(2 + 22) -(10 + 14)} + 0707{(3 + 21)-(9 + 15)} )-(8 + 16)}+0-259{(5 + 19)-( + (11 + 13)} + 0-5{(2 + 22) -(4+20) - + (10 + 14)}, 12&i=(6-18) + 0-966{(5-19) +(7-17)} +0-866{ (4-20) + (8-16)}+X)'707{(3-21) + (9-15)} + 0-5{(2-22) + (10-14)} +0-259{(l-23) + (ll -13)}, -(10-14)} +0-5 {(1-23) +(5-11)) '-(7-17) -(11-13)}, 316 Glossary. The terms are arranged in pairs for facility of calcula- tion, as will be better understood on consulting the numerical example given presently. It will be noticed that in the case of the b coefficients we have to do invariably with differences, and that the sum of the two numerals (representing the observational hours) which form the pair is invariably 24. Similarly in the case of the a coefficients, with two apparent exceptions, we have the sum of observational values at hours which together amount to 24. The exceptions (0 12) are only apparent ; they really represent {(() + 24) (12 + 12). The hours and 24 alike represent midnight. Take for illustration the case already given of the diurnal variation of temperature at Kew Observatory during March. The 24 hourly differences from the mean of the day on the average of the 45 years 1871 to 1915 were as follows : 0,12 1,23 2,22 3,21 4,20 5,19 -1-35 -1-51 -1-75 -1-91 -2-10 -2'19 + 2-03 -1-04 -0-72 -0-29 +0-11 + 0-70 6,18 7,17 8,16 9,15 10,14 11,13 -2-31 -2-14 -1-62 -O49 + O43 4-1-41 + 1-42 +2-26 +2-69 +2'93 +2-77 +2-58 - The headings denote the hours to which the observa- tional data immediately below refer. For instance, the departures from the mean value of the day at h. and 12 h. were respectively 1*35 and + 2*03. Referring to the formulae it will be seen that we want the algebraical sum and difference of the two entries Harmonic Analysis. 317 which appear in the same column. These are respect- ively : sums + 0-68 -2-55 -2-47 -2-20 1-99 -1-49 -0-89 + 0-12 + 1-07 + 2-44 + 3-20 + 3-99 differences -3-38 -0-47 -1-03 -1-62 -2-21 -2-89 -3-73 -4-40 -4-31 -3-42 -2-34 -1-17 For calculating the b coefficients we require only the differences ; for the a coefficients in addition to the sums we require only the first difference. We thus have 12i= -3-38 + 0-966 (-2-55-3-99) + 0-866 (-2-47- 3-20) + 0-707 (-2-20-2-44) +0*5 (-1-99-1-07) + 0-259 (- 1-49-0-12). Employing Crelle's Tables, or logarithms, or straight- forward multiplication, we get 12a : = 3*38 6*32 -4-91 -3-28 -1-53 -0-42 = - 19-84, and so a\ = 1*653. Similarly we find 12a 2 = +4'83, and so a 2 = + 0'403 ; 12a 3 = -0-53 03= - 0-044 ; 12a 4 = -0-32 a 4 =- 0-027. Coining next to the &'$, we have UPi =-3-73 + 0-966 (-2-89-4-40) +0-866 (-2-21-4-31) + 0-707 (-1-62-3-42) + 0-5 (- 1-03 -2-34) + 0-259 (-0-47 -1-17), = -3-73 - 7-04 - 5-65 - 3-56 - L-69-0-42 == - 22-09, and so b\ = 1-841. 318 Glossary. Similarly we liiul 12& 3 = + 5'86, and so // 2 =+0'488 ; 126,= + 0-79 6 3 = +0-066; .12& 4 =-l-39 b, = -0-116. The deduction of the P, -4, constants is simple. Take. for example, the case of PI, A\, 'i.e. the 24-hour term. We have tan A l = i = ~= + O8978. bi ' 1'841 The angle whose tangent is +0*8978 may be 4155 / or 415,Y + 180, i.e. 2215f) / . To determine which, we notice that a\ and bi are both minus, so that sin A\ and cos .4i are both minus. Thus AI lies in the third quadrant, between 180 and 270, and consequently is 22155', or to the nearest degree 222. The formula PI = aifemAi gives us PI=~ ^=2'47. O'obo We need not trouble about the sign, as the P's are all positive. The values of A 2 , P 2 , A^ PS, A 4 and P 4 are derivable exactly in the same way. The process of finding the trigonometric series to give the best representation of a periodic function is known as harmonic analysis. The reverse process, determining the value of the function at any time when the components are known, is harmonic synthesis. Both processes can be carried out by suitable machines, and also by arithmetical computation from given data. The latter process is the more .usual except in the case of the prediction of tides. In any term P sin (nt + A) the coefficient P which determines the range is called the amplitude, nt + A is called the phase-angle, A being the phase-angle for mid- night. It may be mentioned that the alternative form Harmonic Analysis. 319 Pcos n(t t ) where t is the time of the maximum, has certain advantages ; it was adopted by General Strachey for the discussion of harmonic analysis of temperature in the British Isles. By comparison of the amplitudes and phase-angles for different places and different seasons, climates may be classified. For example, the amplitude of the whole-day term for temperature in July at Falmouth is 2'la, and the phase-angle for local apparent midnight is 250. In com- parison ' with Kew, the amplitude is small and the maximum occurs early. This difference in phase is typical of the difference in conditions on the coast and inland. It may be stated, however, that as regards tem- perature, harmonic analysis has not yielded information which can not be obtained more readily from the curves showing the daily variation. With pressure more im- portant results have been discovered. For temperature the first or all-day term in the expan- sion in trigonometric series is by far the most important. With pressure the second term is comparable in size with the first, and at most stations there are two maxima and two minima in the course of 24 hours. The first term is found to depend on the situation of the station, whether near the coast or inland, in a valley or on a mountain-side, whereas the second or twelve-hour term depends principally on the latitude. The daily changes represented by the first term are clearly understood, they are the effects of local heating of the air. No adequate explanation of the surgo of pressure which is represented by the second and higher terms has been put forward. 320 Glossary. The daily variation of pressure at Cairo in July is represented graphically by the second Figure. Daily variation of the Barometer at Cairo [Abbassia Observatory] in July. The departure of the pressure from the mean for the day is given in millibars by the expression 92 sin (15* + 17) + : 66 sin (30*-+ 140) + '12 sin (45* + 348) +-05 sin (60* + 250). The first term represents an oscillation with the maximum and minimum at about 5h.and 17h. respectively. It indicates that as the air is warmed in the daytime it expands and overflows from the Nile valley over the sur- rounding high ground and over the neighbouring seas. The second term represents an oscillation with maxima at lOh. 20m. and 22h. 20m. These hours are almost the same all over the globe. The amplitude depends on the latitude and to a certain extent on the time of year. The mean value for the year at Cairo is 0*8 mb. It is about 1*3 mb. at the equator, 0*5 in latitude 45, 0*35 in Lond'on and 0*1 mb. in latitude 60. The third term is interesting as it changes its phase by 180 at the equinoxes. The first maximum occurs at 2h. in summer, the first minimum at the same hour in winter. Harmonic Analysis. 321 It has been mentioned that the all- clay term depends largely on local conditions. An interesting contrast is offered by the British observatories. At Richmond, Surrey, the amplitude of this term in July is about O3mb., and it has about the same value at Cahirciveen, but the phases are opposite : at Richmond the maximum occurs at f)h., whereas at Cahirciveen it is the minimum which occurs in. the morning (at 7h). Harmonic Analysis may be extended to the investigation of changes which are caused by forces having different periods. The classical instance is that of the tides. The , Tides being caused by the attraction of the sun and the moon show as periods the solar and also the lunar day. The process by which the heights and the times of tides are foretold in practice defends on harmonic analysis and synthesis. Ice Owing to the large amount of heat absorbed in melt ing (80 CALORIES for one gramme melted) a mass of ice represents a powerful reservoir of cold. Masses of ice or snow can attain to such dimensions in Nature that the heat absorbed during melting is of climatological im- portance. An excellent example is furnished by the ice- bergs observed by Antarctic explorers. The largest of these appear to be portions of the great Ross Ice-barrier that have broken away during the summer months. They are generally several hundred feet thick and may exceed 20 miles in length. The amount of heat required to meli one 20 miles long, 5 miles broad and 600 feet thick would be sufficient to raise the temperature of the air over the whole British Isles from the ground up to a height of 1 kilometre (3,281 feet) by over 40C. When ice forms upon a pond during frosty weather the cooled water at the surface is continually replaced by 322 Glossary. warmer water from below until the whole "mass has fallen to 277a (39F), which is the temperature at which water has its greatest density. The surface can then cool un- disturbed until the freezing point is reached and ice begins to form, but when ice first begins to form in a flowing river is a complicated question about which Professor H. T. Barnes, F.R.S., has written. When the sea freezes the crystals formed contain no salt but cannot easily be separated from the brine which is mixed up with them, consequently the water obtained by melting genuine sea-ice is salt. When, however, this ice forms hummocks under the action of pressure the brine drains out and leaves pure ice. Newly-formed sea-ice has a surprising degree of flexibility due to the fact that the crystals are separated by layers of brine or salt, and even when it is several inches thick the surface can be moved up and down unbroken by a swell. As the thick- ness increases this can no longer happen, and the sheet is broken up into pieces, which grind together and soon form the beautiful *' pancake-ice " familiar io polar explorers. lonisation in the atmosphere arises from the presence of free + and ions. Charged ions move along the lines of force in an electric field, carrying their charges with them. This is equivalent to an electric current. The current increases with the number of free ions present, and with their mobility (i.e., the velocity which an ion possesses in a field of unit strength). An increase of current for a given strength of field is equivalent to an increase of conductivity in the medium. There are at least two distinct kinds of ions in the atmosphere, usually known as light or mobile ions, and as heavy or Langevin ions, so-called after their discoverer. Itmisation. 323 I'rof. Langevin. The heavy ions ;ire the more numerous, at least, in the polluted air ordinarily. met with near large towns, but their mobility is so small that they are of minor importance so far as the conductivity of the atmosphere is concerned. The light ion has a velocity of the order of 1 cm. per second in a field of 1 volt per cm, Ionic charges in the atmosphere are usually measured with the Ebert apparatus. This consists of a hollow cylindrical tube vertical in the more recent forms con- taining a co-axial cylindrical rod, which is insulated and connected to the fibres of a " string electrometer." The rod is charged to a potential of from 100 to .20 ) volts, the cylindrical tube being earthed. Air is pulled through the tube by a turbine, the amount passing being recorded on an anemometer. Supposing the rod charged negatively, the positive ions in the admitted air are attracted to the rod, and as the air passes give up their charge to it. The length of the rod and the potential to which it is charged are such as to ensure that no mobile ion will escape capture. The readings of the electrometer taken before and after the admission of the air, the capacity of the electrical system being known, inform us how much the charge on the rod has been diminished. A small part of the loss determined by a separate experiment is due to imperfect insulation ; the balance represents the free electrical charge opposite in sign to that on the rod existing in a measured volume of air. Two experiments are made, the rod being charged negatively in the one case, positively in the other. We thus get the free charges present per c.c. in the atmosphere. These charges are highly variable, but are usually of the order of 1000 x 10" 20 in electromagnetic units. The number of ions per c.c. is often published and may be deduced by dividing the 1S204 L 2 324 Glossary. charge, whether + or ,. by the charge on an ion, for which the value accepted at present is 159 x 10~ L ' 2 in electromagnetic units. In all, or nearly all the earlier work done on the subject, the ionic charge was given a value 29 per cent, less than that now accepted, leading to an overestimate of the number of ions. It is also the case that part of the charge is derived from the heavy ions, though the percentage of these caught by the Ebert apparatus is undoubtedly small. The mobility of the ions may also be measured with the Ebert apparatus, but the process is somewhat compli- cated, and the accuracy of individual determinations is not high. Observations at the earth's surface have usually given a decided excess in the number of -f ions. The + and - ions naturally tend to combine, and their presence thus points to the existence of some agency producing ions. Several possible agencies are recognised, including radio- active substances, and solar radiation, especially ultra- violet light. Near the ground the radio-active substances ordinarily present in the ground may be the chief source. Balloon observations have shown that the ionisation diminishes somewhat at first as we leave the ground, but at heights of 1 k to 2 k it begins to increase again, and at heights of 9 k or 10 k it seems to be very much larger than near the ground. The ionising agent there is presumably RADIATION of some kind from the sun. Lightning", Protection from. The region between the earth and a thundercloud is one of great electrical stress. It may appear paradoxical that a lightning-con- ductor, intended to protect a structure within that regicii from damage, acts primarily by increasing this stress. The lightning-conductor consists of a metallic point or set L igh tning, Protection from. 325 of points connected by a metal rod to a conductor of con- siderable area buried in moist earth. This conducting point, projecting some short distance above the highest point to be protected, increases the intensity of the stress between itself and the cloud, till one of two things happens. A silent " brush " discharge may take place, in which the induced electrification of the earth streams comparatively gently from the point (selected because of its well-known property of facilitating brush-discharge), and thus finally reduce the stress to a limit insufficient to produce a violent lightning discharge. If, on the con- trary, a lightning discharge does occur in the region, it will pass to the point because of the increased stress, and the rod will carry the current harmlessly to earth. The old idea of an "area of protection" is no longer tenable, and pointed conductors should therefore be pro- vided on every vertical projection of the structure to be protected, and at intervals along the ridge of a long roof. These should be connected by metal rods or cables and the rods connected at several places to earth. The con- necting rods should run as straight as possible since electrical inertia will make the discharge jump across an air-gap in preference to rounding a sharp bend or loop in the rods. Iron rods are preferable to copper both electrically arid economically, and may be painted for preservation. The rods should be held at a distance of some inches from walls, but should be fixed, not by in- sulators, but by metal holdfasts fastened in or on the walls. Metal roofs, gutters, pipes and other masses of metal should be electrically continuous and connected to earth. Inside the building, pipes, bell-systems and all large metallic masses should also be earth-connected, since vio- 326 Glossary. lent oscillations may be induced in them by a discharge through the external conductor, and they may cause fire by sparking to earth if they are not so connected. The only complete protection is attained by enclosing the structure in a "birdcage " arrangement of conductors, well connected to earth at several points. A metal build- ing, whose parts are electrically connected amongst them- selves and to earth, is probably the most perfect protection possible, but here also internal metal-work should be earth-connected . The personal danger from lightning in the open country is at least twice as great as in towns, owing to the number of buildings in the latter protected intentionally (as by lightning conductors), or unintentionally (as by overhead wiring for light, power, telephones, etc.). It seems to be a definitely established fact that certain trees, particularly the Oak, are more frequently struck than others. The relative danger of lightning stroke, taking the Beech as 1, is Oak 50-60, Scotch Pine 30-40, Spruce 10. but is of course much affected by environment. Isolated and prominent trees are more frequently struck than average forest trees. The safest course in the open though perhaps not the most comfortable is to lie in a ditch or furrow, failing these to lie on the ground. To shelter under isolated trees or on the edges of a wood is dangerous ; well inside the wood the danger is probably not great. Immediately under a line of over-head wires is also a relatively safe area. The protection of aircraft against lightning is a difficult problem, complicated by such appendages as wire cables in the case of balloons and radio-telegraphic aerials on aeroplanes. The trail of hot gases from the engine Magnetism. 327 exhaust acts as a comparatively good conductor, which may work for good or harm according to the position of the craft relative to the thundercloud, but on the whole, by acting as a lightning-conductor and thus facilitating dis- charge along a path passing through the machine, probably adds a good deal more to the danger than it takes from it. * Magnetism. The branch of knowledge relating to magnetic phenomena. Terrestrial magnetism is con- cerned with the earth's magnetism. The earth has been happily .described as a great magnet : the distribu- tion of magnetic force at its surface may as a first approximation be regarded as that due to a uniformly magnetised sphere, whose magnetic axis, however, is inclined at some 10 or 12 to the earth's axis of rotation (polar axis). The magnetic poles are the one to the north of Canada, the other in the Antarctic. At these poles the dip needle is vertical, and the horizontal component of magnetic force vanishes ; the compass needle has no guiding force and points anyhow. At what may be called the magnetic equator, which is nowhere very far from the geographical equator, the vertical force vanishes ; the dip needle is horizontal, and the horizontal force has its largest values. The horizontal force in London is only about half that where the magnetic equator crosses India. The compass needle points in the direction of the local horizontal component of magnetic force, and this direction is exactly north and south only along two lines or narrow belts, one of which at present crosses A.sia Minor and European Russia, while the other crosses the United States and Canada. The position of these " Agonic " lines as they are called, and the direction of the magnetic needle, change gradually with time, having what is known as " secular change." The inclination of the magnetic needle 328 Glossary. to the geographical meridian is called the declination, or sometimes, rather unfortunately, the magnetic variation. As we travel westward from the Asiatic-European Agonic line, the declination becomes increasingly westerly, until we pass the western limits of Europe. Thus, at present, approximate values are at Cairo If W., at Athens 3f W., at Pola 7i W., at Copenhagen 8^ W., at Brussels 12^ W., at London 15 W., and at Valencia Observatory (Co. Kerry) 20 W. In Europe westerly declination is at present diminishing at a rate of nearly 10' a year. Besides the secular change, declination has a regular diurnal variation, and also frequent irregular changes, which when large are known as magnetic storms. In the Northern hemisphere the prominent part of the regular diurnal variation is a westerly movement from about 7h. to 13 h. (1 p.m.), followed by a slower easterly movement. Speaking generally the range of the regular diurnal variation is least near the magnetic equator, where it may average less than 3', and the greatest near the magnetic poles, where it may exceed 1. It varies with the season of the year. At Richmond, for instance, in the average year it varies from about 3^' in December to 11' in August, the mean from all months of the year being about 8'. It also varies considerably from year to year, show- ing a remarkably similar progression to that of sun- spot frequency. In a year of sun spot maximum the range may be 50 per cent, or more larger than a few years earlier or later at sunspot minimum. Few days are wholly free from irregular changes of declination, and these are sometimes much larger than the regular changes. Thus the actually observed daily range at Kew sometimes exceeds 1, and in high latitudes ranges of 5, or even 10 or more, are occasionally observed, In latitudes, however, Magnetism. 329 below 60 in Europe departures of more than 30' from the mean value for the year are rare. On the occasions of very large magnetic changes AURORA is generally observed, even in the South of England, and there is usually some interference with ordinary telegraphy. The direction of the compass-needle is practically independent of the height above the ground, except in localities where some large underground source of dis- turbance exists. In such a case, moreover, the higher one goes up the less is the effect of the local disturbance. All objects of iron are liable to, become magnets, especially when they are long and disposed with their length nearly in the magnetic meridian, and are thus apt to introduce errors in the readings of compasses in their vicinity. This warning applies particularly to large objects like ships, girders and large guns ; but even small articles of iron, such as some buttons or spectacle frames, when close to a compass, may cause quite a serious error. Some forms 01 rock, especially dark-coloured basalt, are strongly magnetic, and large disturbing effects are caused by the strong electric currents used in connection with electric tramways and railways when the system is a direct-current one. Precipitation. A term borrowed from chemistry to denote any one of the results of the conversion of the invisible water-vapour to visible water or ice," thus com- prising not only rain, hail, snow, sleet, dew, hoar-frost and rime, but cloud, mist and fog. In practice, however, the use of the word is limited to appreciable deposit in either the solid or the liquid form at the earth's surface, the definition of "' Day of precipitation " being identical with that of RAIN-DAY (q.v.y. At low levels it is rare for appreciable deposit of water in the rain-gauge to 330 Glossary. result from precipitation of cloud, mist or fog, but in mountainous districts wet-fog and SCOTCH-MIST (q.v.} are responsible for a considerable quantity of " rainfall." See RAIN, HAIL, SNOW, etc. Radiation, Solar and Terrestrial. Radiation is the process by which heat is transferred from one body to another without altering the temperature of the inter- vening medium. All life upon the earth and all meteoro- logical phenomena are dependent upon the radiant heat and light received from the sun. See INSOLATION. The earth itself is always radiating into space, and according to Stefan's law r the rate of radiation is given by o-T 74 , where T is the absolute temperature and <t a constant. During the day normally the earth receives directly from the sun and indirectly from the atmosphere, including the clouds, more radiation than it gives out. At night the reverse is the case. The rate of loss of heat from a plate freely exposed at the earth's surface at night tells us the balance of loss, i.e., the excess of the earth's radiation outwards over what it receives from the atmos- phere. With the aid of Stefan's law we can form an estimation of radiation received from the atmosphere. According to Professor Millikan, . who has recently reviewed the literature of the subject, the most probable value of <r in Stefan's formula in watts per sq. cm. is 5*72xlO~ 12 , or in gramme - calories per sq. cm. per minute 77 x 10~ 12 . For the radiation at 288a this gives 77(288) 4 / klO~ 12 or 0*53 gramme-calories per sq. cm. per minute, and similarly for any other temperature. Combining this with observations of the balance of loss experienced at different heights at different temperatures, we have the following results, the heat-data being in gramme-calories per sq. cm. per minute : , Ho/fir and Terrestrial. 331 Height (metres) 44 950 3100 3100 Temperature absolute 288 267 261 271 Balance of loss 0-13 " J 5 0*20 0*20 Radiation by Stefan's law!.. o'39 0*36 0*42 Radiation received from at- mosphere 0*40 0-24 o- 16 O*22 Recent experiments by Anders Angstrom show that aqueous vapour exercises a potent influence in reducing the balance of loss of heat by nocturnal radiation. In a paper read before the Royal Meteorological Society in February, 19 1 7, and published in the Quarterly Journal for April, Mr. W. H. Dines, F.R.S., gives the following values for the amount of solar radiant energy absorbed and reflected by the earth and its atmosphere, and of radiation of the earth's heat and its absorption and transference. The values are expressed in gramme-calories per sq. centimetre per day. Radiant energy received by the earth per day 720 (An amount capable of warming the atmosphere 3a per day.) Radiant energy absorbed by the earth per day ; for the whole earth ... ... ... ... 300 (The Callendar instrument at the Meteoro- logical Office gives about 200 ; Hann gives 166 for Kiev (50N.) ; ' and 46 tor Tauren- berg Bay (79N.) ; 300 is the value for London at the beginning of May or August.) liadiant energy absorbed by the air ... ... 60 (The value seems low, but there is no observational evidence against it.) 332 Glossary. OutwHi'd radiation from the earth ... ... f)<K.) (Obtained by using the most probable value of <r in Stefan's law [vide supra'].) Heat radiated by earth reflected back by atmosphere ... ... ... ... ... 60 Heat radiated by earth absorbed by atmosphere 360 Heat radiated by earth transmitted beyond the outer limit of atmosphere ... ... ... 80 Transference of heat from earth to air ... ... 200 Radiation from air downwards 340 Radiation from air upwards ... ... ... 280 Effect of the ivhole daily solar radiation when applied to raise the temperature of the air in the first 1, 2 and 8 kms. of the atmosphere. According to Angot (quoted from Hann) the following are the proportionate values of the solar radiation per cm 2 , in each latitude. 10 20 30 40 50 60 70 80 90 350 345 331 308 277 240 199 166 150 14;> If the value of the solar constant be taken as 2 gramme- calories the daily receipt of heat per cm. strip on the equator is 2r x 2 x 60 x 24 where r is the earth's radius, and the receipt per cm 2 , per day is %r x 2 x 60 x 24/27r/'=916. The amounts for each latitude are therefore 10 20 30 40 50 60 70 80 90 916 904 869 807 725 629 521 435 393 380 Taking a mean pressure of 1014 mb. the water-equivalent of the atmosphere is a layer of water 250 cm. deep, hence dividing the above numbers by 250, the number of tiJ.fn- and Tr-rrrxt./'ictl. 333 degrees that the whole radiation is capable of raising the whole atmosphere is given below for each latitude. 10 20 30 40 50 60 70 80 90 3-7 3-6 3-5 3-2 2*9 2'5 2-1 1-7 1-6 1-5 With a mean surface temperature of 288a and a lapse- rate of 6a per k. the percentage of the whole quan- tity of air found under 1 k. height is 11*3, under 2 k. 21*7 and under 3 k. 30'8. The amounts for each latitude are shown in the following table, the figures indicating the rise af temperature in degrees that would occur if the whole solar radiation were concentrated in the layer and no loss of heat occurred. Latitude 10 20 30 40 50 60 70 80 90 Under 1 k. ... 32'5 32'0 30'8 28'6 25'7 22-3 18-4 154 13'9 13'4 Under 2k. ... 16'9 16'6 16'0 14*9 13'4 11-6 9'6 8'0 7'2 7'0 Under 3k. ... 11-9 11-7 11-3 10-5 9*4 8-2 6-7 5-7 5'1 4'9 The above figures represent mean conditions as to density. A fall of pressure will increase the values and so also will a rise of temperature, because with a rise of temperature a smaller proportion of the whole atmosphere will be found in the given layer. The mean conditions hold in about latitude 40 ; in the equatorial regions some 4 per cent, must be added to the values, and in latitude '55 some 4 per cent, deducted. The values given are interesting, but, it must be re- membered that the whole solar heat is not absorbed by the lower strata, probably only a small proportion of the whole, also as the loss per 24. hours is about the same as the gain in the 12 hours that the sun is on the average above the horizon, the rise of temperature, quite apart from convection, would be only half the values given in the table. 334 Glossary. Size and Rate of Fall of Raindrops. Raindrops. The size of raindrops can be measured If, for example, a shallow tray containing dry plaster of Paris is exposed for a few seconds during rain, each drop which falls into the tray will make a plaster cast of itself which can easily be measured. A better method is to collect the drops upon thick blotting paper. If, while still wet, the paper is dusted over \Yith a dye powder a permanent record will be obtained consisting of circular spots whose diameter is a measure of the size of the drops. By comparing the diameters of the discs produced by raindrops with those produced by drops of water of known size, the amount of water contained in the former can be found. The following table contains some results ob- tained by P. Lenard in this way at nine different times : Drops. Diameter. Vol- No. of drops per m.'- per second. ume. mm. in. mm. 3 CD C2) (3) (4) (5) C6) m CS) m '5 0:9 o o6'6 ;IOOO 1600 129 60 100 514 679 7 I'O 039 0-523 200 120 100 280 50 1300 423 524 233 i $ 1-77 140 60 73 1 60 50 500 359 347 2'0 079 4-19 140 200 100 20 150 200 138 295 46 2*5 098 8-19 'O 29 20 156 205 7 3'o 118 14-2 o o 57 200 138 81 o 3'5 138 22-5 o o O O O o 28 32 4' 157 33'5 50 O 20 39 4'5 177 47-8 200 101 c 5*0 1 196 65*5 o o o o 25 Total number ... 1480 1980 486 540 500 2300 1840 2190 500 Rate of rainfall 0-09 O'o6 O'll 0-05 0-32 0-72 0-57 0-34 0-26 (mm./min.) Raindrops^ 335 (i) and (2) refer to a rain " looking very ordinary " which was general over the north of Switzerland. The wind had freshened between (1) and (2). (3) Rain with sunshine-breaks. (4) Beginning of a short fall like a thundershower. Distant thunder. (5) Sudden rain from small cloud. Calm ; sultry before. (6) Violent? rain like a cloudburst, with some hail. (7), (8)' and (9) are for the heaviest period, a less heavy period, and the period of stopping of a continuous fall which at times took the form of a cloudburst. We see then that in a general rain, such as the normal type which accompanies the passage of a depression over Northern Europe, by far the greater number of drops have a diameter of 2 mm. or less. In short showers, especially those occurring during thunderstorms, the frequency of large drops is much greater. In such showers the diameter of the largest drops appears to be about 5 mm. We shall see later that there is a limit to the size of drops determined by the fact that it is im- possible for a drop, whose diameter exceeds 5*5 mm. or rather less than a quarter of an inch, to fall intact. The rate at which a raindrop, or any other object, can fall through still air depends upon its size. When let fall its speed will increase until the air-resistance is ex- actly equal to the weight, when it will continue to move at that steady speed (see EQUILIBRIUM). The manner in which this u terminal velocity," as it is called, varies with the size of the raindrops is shown in the following table, due to Lenard. 336 Glossary. TERMINAL VELOCITIES OF WATER-DROPS FALLING IN AIR. Diameter Terminal Diameter Terminal of drop. Velocity. of drop. Velocity. mm in. m/s. mi/hr. mm. in. m/s. mi/hr. 0*01 0*0004 0-0032 0-007 3-0 0-118 6-9 15-4 O'l 0-0039 0-32 0-71 3*5 0-138 7'4 16-5 *5 0-020 3*5 7'9 4-0 .0-157 7'7 17-2 r-o 0-039 4*4 9*8 4-5 0-177 8-0 17-9 1-5 0-059 5-7 12-6 5'0 ' 200 8-0 17-9 2-0 0-079 5-9 13-2 5*5 0*216 8-0 17-9 2'5 0-098 6-4 14-3 We may look upon this table in another way. The frictional resistance offered by the air to the passage of a drop depends upon the relative motion of the two, and it is of 110 consequence whether the drop is moving and the air still, or the air moving and the drop still, or both air and drop moving if they have different velocities. The velocities given in the tables are those with which the air in a vertical current must rise in order just to keep the drops floating, without rising or falling. The above results were, in fact actually determined by Lenard in this way, by means of experiments with vertical air- currents on drops of known size. We see that beyond a certain point the terminal velocity does not increase with the size of the drops. This is due to the fact that the drops become deformed, spreading out horizontally, with the result that the air-resistance is increased. For drops greater than 5*5 mm. diameter, the deformation i^ Raindrops. 33? sufficient to make the drops break up before the terminal velocity is reached. An important consequence of Lenard's results is that no rain can fall through an ascending current of air whose vertical velocity is greater than 8 m/s. In such a current the drops will be carried upwards, either intact or after breaking up into droplets. There is good reason for believing that vertical currents exceeding this velocity frequently occur in nature. On account of their inability to fall in an air current which -is rising faster than their limiting velocity, rain- drops formed in these currents will have ample oppor- tunity to increase in size, and the electrical conditions will usually be favourable for the formation of large drops. These large drops can reach earth in two ways ; either by being carried along in the outflow of air above the region of most active convection, or by the sudden cessation of or a lull in the vertical current. The violence of the precipi- tation under the latter conditions may be particularly disastrous. (See also Cloudburst and Hail.} Electrification of Waterdrops ly Splashing. If drops of water are allowed to splash upon a metal plate, the water acquires a minute positive charge of electricity, and an equal negative charge is shared by the plate and the air contiguous to the splashing drop. It is possible to show this by means of delicate apparatus. The largest charges are found when distilled water is used, and even small amounts of dissolved substances in the water make a considerable difference to the results obtained. With sea-water, indeed, the effect is reversed, the water becoming negatively charged after splashing. 338 Glossary. The presence of a solid obstacle to cause splashing is not really necessary to produce the separation of electricity. The breaking up of a jet of water into spray and the splitting of large drops of water in a current of air produce similar effects. The last-named case is of par- ticular importance in meteorology because it forms the basis of the theory put forward by Dr. G. C. Simpson to account for the production of the enormous electrical stresses in the atmosphere which precede the discharge of lightning in thunderstorms. The first necessity for a thunderstorm is the formation of a cumulus cloud, and this requires an ascending current of air. In ascending the air expands and gets rapidly cooler, with the result that before long the water- vapour in the air begins to condense and form visible droplets. The cloud is, in fact, the visible result of this condensa- tion. Once formed, the drops rapidly increase in size and would ordinarily fall as rain. But if the ascending current is sufficiently violent the raindrops will not be able to fall through it, but will be carried up with the air. The vertical velocity required to hold up drops of all sizes is 8 m/s., and there is no reason to doubt that such currents can easily be produced. Now in such a current it is impossible for a drop to grow beyond 5*5 mm. in diameter (see Size and Rate of Fall of Raindrops). At that point it becomes unstable and divides into droplets. These in their turn go through the same process of growing and dividing. Each time a division occurs the droplets gain a positive charge, and the air which is carried up with the current gain^ an equal negative charge. In this way the waterdrops in the region of the ascending current rapidly become very highly charged, and as soon as the potential gradient anywhere amounts Raindrops. 339 to 30,000 volts/crn. a lightning-flash will occur. Although the charge produced by a single division is very small, we have only to suppose that the same quantity of water may take part in many hundreds of divisions and there is nothing improbable in this to be able to account for the production of sufficiently high potentials. The negatively charged air will be carried right to the top of the column and there dispersed. Its presence should be shown by a negative charge on the rain which falls some distance from the storm-centre, while that falling near the centre should be positively charged. Such observations as exist tend to support this conclusion of the theory. Regression Equation. A regression equation shows the most probable form of the relationship between two varying quantities insofar as such relationship can be definitely determined from the set of statistical data on which it is based. It is formed from the correlation co-efficient and the matter is best explained by an example. See also under CORRELATION. The strength of the wind and the steepness of the barometric gradient at the same time and place are closely related, and a regression equation may be formed between them. Let W denote the strength of the wind, W m its mean value, and S W the departure from the mean, and let 6r, 6r m and f G be the corresponding values for the steepness of the gradient. Then if the gradient is known the strength of the wind is given by a regression equation in the form W = W m + c G + . In this equation t will in general have some value, posi- tive or negative, differing on each occasion, and the a will 340 , Glossary. be so chosen that the sum of the squares of the e's will be as small as possible. The IF, c Gr and < are variable quantities, the a a constant. It is usual, for the sake of brevity, to write the equation c W = a c Gr, but it must be remembered that the e has been omitted ; e is called the residual error, and since e is often fairly large, it is not permissible to wrUe J-^~=^4^frf. g G* There are two errors involved in a regression equation. The value of a can in general only be found at all correctly when the number of observations is large. The residual error e may be as large as the term a c G unless the correlation co-efficient is nearly 1 or 1. When the correlation is + 1 the term t is nothing ; also, when the correlation is known to be large (it cannot be proved to be large from a few observations) the value of a can be determined with greater certainty than when the correla- tion co-efficient is small. The following are three examples all dependent on fairly high correlation co-efficients. Thickness of TROPOSPHERE = 10,600 + 112 3P 9 metres, where P 9 denotes the pressure of the air at a height of nine kilometres expressed in millibars. (Europe.) Hay crop per acre = 28 + 4 c R cwts., where R denotes the spring rainfall in inches. (East of England.) Number of deaths in England during July, August and September = 150,000 + 7,200 c T, where T denotes the mean temperature in degrees F of June, July and August. (On the assumption that the present population is forty millions.) Scotch-Mist. In mountainous or hilly regions, rain- clouds (nimbus) are often adjacent to the ground, and Scotch-Mist. 341 precipitation takes place in the form of minute water- drops, the apparent effect being a combination of thick mist and heavy drizzle. The upland character of the greater part of Scotland and the consequent frequency of occurrence of the phe- nomenon in that country have secured for it the appella- tion by which it is generally known in the British Isles. The base of a true nimbus or rain-cloud rarely exceeds about 7,000 ft. (2'1 k.) in elevation, and sometimes descends to within a few hundred feet of sea-level, so that Scotch>mist may be experienced in comparatively low- lying regions. In the uplands of the Devon-Cornwall peninsula the same phenomenon, which is there of very frequent inci- dence, is known as " mizzle." Sleet. Precipitation of rain and snow together or of partially melted snow. In America the name " sleet " is used for small dry pellets of snow which might be classed as soft-hail. Sleet is, perhaps, snow that passes through a stratum of comparatively warm air (see INVERSION) and, undergoing partial liquefaction therein to an extent varying with the temperature and thickness of the layer, reaches the ground in a semi-liquid con- dition. If the stratum of warm air is not adjacent to the ground, and if the surface-temperature is below the freezing-point, a phenomenon similar to GLAZED FROST (q.v.) may result, the re-freezing of the half-melted snow occasioning the formation of a layer of ice on all objects exposed to the precipitation. Marked instances of this occurred in London during the winter of 1916-1917, and road traffic, and in some cases even rail-locomotion, was rendered difficult or impossible. )U2 Glossary. Snow. Precipitation in the form of feathery ice- crystals ; other forms of ice-precipitation are the powdery ice-crystals or needles which are commonly experienced in the snow-storms of intensely cold weather on mountain tops and in the arctic or antarctic regions in the snow- storms, in fact, of which the " BLIZZARD " has become the descriptive name; soft hail or graupel. /.., snow in which the needles are agglomerated to form minute snow- balls, sometimes striated in texture, which break with a splash on reaching hard ground ; and true hail, which began as rain frozen and sustained in rapidly-ascending currents of dynamically cooling air. Snow may perhaps be the result of the direct congelation of water- vapour, the omission of the intermediate liquid-state being the essential difference between the hail and snow processes. Snowflakes are formed of one or more ice-crystals arranged in symmetrical hexagonal patterns of which there is an almost infinite variety. Many are figured as photo- micrographs in the Monthly Weather Review of the United States Weather Bureau, Washington, for 1902 (W. A. Bentley). When snow falls with comparatively high temperature, large, wet flakes often result ; with lower temperature the flakes are smaller, and with the thermo- meter reading far below the zero of the Fahrenheit scale, we have the " snow-dust" or " ice-needles " ; fine ice- crystals or needles also characterise the deposits which are formed in foggy, frosty weather, particularly on mountain- tops, where wreaths of such crystals sometimes grow out to windward. The hexagonal formation of a snowflake may be well observed under a low-power microscope ; it will be noticed that each one of the constituent "spiculee" is set at an angle of 60 degrees to its fellows. The ratio which an inch of rain bears to an inch of snow depends Snow. upon the density of the snow ; as a rough approximation for the most common kind of snow a ratio of 12 to 1 is usually taken in this country. In exceptional cases the divergences from this value are very wide indeed ; according to Colonel Ward, the range may be from about 5 to 1 to about 50 to 1 that is to say, a foot of snow on the ground may yield, when melted in the RAIN-GAUGE, the water-equivalent to 2 '4 in. of rain at the one extreme, or to 0-24 in. at the other. One foot of snow to one inch of rain is, however, a convenient generalisation. Soft-Hail. The English term for the form of ice- precipitation known in German as Graupel. It consists in reality of pellets of closely agglomerated ice-needles, sometimes striated in texture, and thus falls under the category of snow, rather than under that of hail. On colliding with any hard substance, soft-hail breaks up with a splash, and may thus be distinguished from true hail, the form of which is not affected by the impact. The French equivalent for " soft-hail " or " graupel " is " gresil." Sun-dial. Little use is made of the sun-dial at the present time, except as an ornament for the garden. There are various forms, the commonest being a horizontal stone slab upon which a rod or style, called the gnomon, is set up in the astronomical meridian, inclined to the horizontal at an angle equal to the latitude of the place, or, in other words, parallel with the earth's axis. The line traced by the shadow of the style at each hour of the day is engraved upon the slab. When the vertical plane through the style lies correctly in the meridian, after applying a correction for the EQUATION OF TIME (q.v.\ such a dial will give 344 Glossary. mean solar time whenever the sun is visible. To obtain Greenwich Mean Time a constant correction must be applied depending upon the longitude of the place. By taking account of the length of the shadow as well as the line on the dial, the time of year can be indicated, and some dials are elaborately graduated as a perpetual calendar as well as time-keeper. Twilight. Twilight is caused by the intervention of the atmosphere between the sun and the earthV surface. With no atmosphere, darkness would set in sharply at the moment of sunset, and would give place suddenly to light at sunrise, as on the moon. But when the sun is some distance below the horizon the upper layers of air are already illuminated, and are reflecting light to us. The amount of reflected light diminishes as the sun's distance below the horizon increases, because higher, and so less strongly reflecting, layers alone are in direct sunlight. So early as the llth century the period of Astronomical Twilight, between sunset and the onset of " complete " darkness, was determined as ending when the sun is 18 below the horizon, and this value has not been modified by later observations. If we assume direct reflection as the sole cause of twilight, this value, 18, would indicate that the atmosphere above a height of some 80 kilometres is incapable of reflecting an appreciable amount of light. Long before the end of Astronomical Twilight, how- ever, the light has become insufficient for ordinary employments, hence another period, Civil Twilight, is recognised, ending when the sun is about 6 below the horizon, and conditioned by the insufficiency of light for outdoor labour after that time. The duration of twilight depends on the season arid the Twilight. 345 latitude. At midsummer the sun is 23^ North of the equator. Hence within the Arctic circle, latitude 90 23i=66|, the sun never 'nets, so thai; there is no twilight! Between the Arctic circle and latitude 90-*3| -18 =48i there is a belt with no true night, twilight extending from sunset to sunrise. At midwinter, in the Arctic circle, the sun does not rise, but up to lati- tude 90 23^ + 18=84i, there is an alternation between twilight and night. North of 84^ there is continuous night. At London (latitude 51 ) Astronomical Twilight has a minimum duration of about 1 hour 50 minutes on March 1st and October 1st, with a secondary maximum in mid- winter of just over 2 hours, and lasts all night at mid- summer. At, the Equator the minimum duration is 1 hour 9 minutes, the solstitial maxima 1 hour 15 minutes. Civil twilight at the Equator varies between 21 and 22 minutes, at London it has minima of 33 minutes in March and October, maxima of 40 minutes in December, and 45 minutes in June. The duration of either twilight at any latitude and season may be found by using the equation sin a sin sin S cos h = - cos cos o where h = sun'e hour angle from the meridian, a = sun's altitude, = latitude, c = sun's declination, a = 50' at beginning of twilight (allowing for sun's semi-diameter and refraction), 6 at end of civil, and 18 at end of astro- nomical twilight ; 346 Glossary* thus to find duration of civil twilight, find cos h for the two cases, a = 50' and a = 6, convert the two values of h to time, and the difference is the required duration. The intensity of twilight depends to some extent on cloud, dust, haze, or other obscurity in the atmosphere. Dust in the higher layers, as in the case of the sunsets of 1883-5 (after the eruption of Krakatoa), may much increase the intensity of illumination during twilight, by increasing the reflected light. In a cloudless sky the intensity falls off rapidly at first, then more slowly from about 35 foot-candles at sunset to 5 foot-candles at the end of civil twilight, and to '0001 foot candle at the end of astronomical twilight; These intensities are to the light of the fuJl moon in the zenith as 1750, 25, and "005 respectively to 1. The optical phenomena of twilight occur in the follow- ing sequence ; for explanations, reference should be made to the corresponding articles in the Glossary. As the sun sinks towards the horizon it is shining through an increasing thickness of haze and dust-laden air, and scattering (see BLUE OP THE SKY) causes less and less of the blue light to reach us, so that the sun appears in- creasingly red. A yellow band now appears on the Western horizon, extending for about 60 to either side of the sun. Gradually the yellow deepens to orange or red as the proportion of blue decreases. As the sun passes below the horizon the pink TWILIGHT ARCH (better called the Anti-Twilight Arch) rises from the Eastern horizon, the space under it being strikingly darker than the rest of the sky. While this arch is rising in the East the PURPLE LIGHT has appeared at an altitude of about 25 in the West, above the point of sunset. This light attains its maximum intensity when the sun is about 4 Q Twilight. 347 below the horizon, and disappears on the Western horizon when the sun is about 6 below, at the end of Civil Twilight. Just before its disappearance the purple light has become a narrow arch over the yellow glow near the horizon and thus forms the " Western Twilight Arch." The purple light is often seen to be intersected by dark blue stripes radiating from the position of the sun. These are the shadows of clouds on or below the horizon, and are frequently called Crepuscular Rays. On very clear nights a second dark segment in the East and a second Purple -Light in the West may be observed. Evening conditions have been assumed above, but obvious inversions make the discussions applicable to the mornings. Vortex. A special form of rotatory motion in fluids. Two forms of vortex have figured much in mathematical literature, the vortex-ring and the long straight vortex, and both are believed to be represented in nature. The mathematical vortex-ring in its simplest form is shaped like a perfect anchor-ring or hoop, whose circular aper- ture is very large compared with the diameter of the circular wire of which it is composed. The cross-section of the material of the ring a liquid or gas is every- where a circle of radius 0, and the centres of all these circles lie on a larger circle, the aperture, of radius n. There is complete symmetry round the axis, i.e., the per- pendicular to the plane of the aperture through its centre. Any plane through the axis cuts the ring at right angles in two circles of radius e, situated at opposite ends of a, diameter of the aperture. If we take any one of these circular sections the liquid within it is everywhere circu- lating round and round within the circle. In the simplest 348 Glossary. case its rotational velocity increases as its distance from the centre, where it vanishes. But in addition to this the ring moves bodily. If it is alone in an infinite liquid, its centre travels in the direction of the axis, with a uniform velocity which is greater the greater ale. 'The straight vortex in its simplest form is a right circular cylinder, or pencil-shaped body, and if the vorticity is uniform over the cross section the simplest case the liquid spins round with a velocity proportional to the distance from the centre of the section, i.e., the liquid forming the vortex turns round exactly as if it were a rigid body. A solitary straight vortex in an infinite liquid has no inherent tendency to translatory movement. The liquid forming the vortex simply goes on spinning round the axis of the cylinder ; the liquid round it also rotates round this axis, but with a velocity which dimin- ishes as the distance from the vortex increases. The assumption ordinarily made that the liquid is infinite means that every part of the vortex is remote from a boundary. But some forms of vortex motion are possible in presence of a plane boundary, and a sphere whose radius is large compared with the largest dimension of a vortex may be treated as a plane. A vortex ring with its aperture parallel to a plane boundary behaves as if face to face in an infinite liquid with an equal " image " vortex, whose distance is double that of the real vortex from the plane. The two vortices repel one another. Again a theoretically possible case is that presented by the half of a complete vortex-ring cut in two, as it were, by the boundary the plane of the aperture being perpen- dicular to the boundary. The motion would be the same as if the ring were really complete and no boundary Vortex. a4l) existed. Similarly there are two possible cases of a straight vortex in presence of a plane boundary. The vortex may be parallel to the boundary. The conditions are the same as if it were in an infinite liquid facing another equal vortex in which the spin is in the opposite direction the distance between the two being double the distance of the real vortex from the boundary. The vortex tends to move parallel to the boundary, in the direction perpendicular to its own length. In the second case the vortex is perpendicular to and abuts on the boundary ; it then behaves as if it extended to infinity on both sides of the boundary, and so has no inherent tendency to translatory movement. Whether the vortex be straight or ring-formed, an essential feature of the mathematical theory is that the liquid, once incorporated in the vortex, remains in it. The beginning and ending of the existence of the vortex are events outside the compass of the mathematical theory. Vortex rings are easily created by human agency. A drop of one liquid falling into another suitable liquid forms a vortex-ring, and smoke-rings are familiar to most people. Whether they occur in nature is a more difficult question. Delicate optical measurements suggest that sunspots are whirls of electrified gases/ It is noticed that sometimes sunspots move in pairs, and that the whirls in them deduced from the optical measurements are in opposite directions. It has been suggested that we have here really to do with the horse-shoe or semi-ring vortex. The two sunspots represent the portions of this which are nearly perpendicular to the sun's surface, and the connecting or crown portion extends into the more rarefied solar atmosphere and is not recognisable from the earth. This is merely a speculation, but the possibility 350 Glossary. of. a similar phenomenon in the earth's atmosphere may be worth considering. What seems to be at least an approach to the long straight vortex is exemplified by water spouts and by the dust-whirls sometimes seen on warm days. But the common belief that it is also exemplified by the ordinary cyclonic storm does not seem well founded. Tho belief is mainly based on the fact that the isobars during a cyclonic storm are often roughly circular. The direction of the wind, it is true, at some height above the ground, approaches that of the isobars. But the centre of the storm, i.e., the centre of the system of isobars, is not stationary but moves with a velocity comparable with that of the wind itself. The actual path of the air is complicated. It is carried from without into the cyclone, but does not remain in it. The mathematical vortex, on the other hand, is composed of the same fluid from start to finish. The mathematical vortex, moreover, is a long thin body like a pencil. An ordinary cyclone, even supposing it extends some distance into the stratosphere, is a disk-like body, the height of which is small compared with its diameter. Again, in the straight mathematical vortex the velocity round the axis is the same at the same axial distance in all cross sections. Ordinarily the wind increases in velocity with the height above the ground. Supposing the core of a cyclonic vortex originally vertical, unless the motion of translation were the same at all heights, the core would depart more and more from the vertical, and, judging by what happens with water-spouts, dissolution would soon ensue. The conditions compatible with real vortex motion in a cyclone are that the velocity should be independent of the height, that the horizontal diameter of the body of Vortex. 351 air possessing the motion should not be large, and that an unchanged body of air should have the translational velocity shown by the isobars. The necessary conditions are certainly not fulfilled near the centre of the ordinary large cyclonic storm. They seem much more likely to be encountered in the small " secondaries " that are some- times met with on the outskirts of large depressions, or in the whirlwinds that occasionally leave a long narrow track of devastation. It is difficult to ascertain the exact meteorological conditions attending these special disturb- ances. A weather map, to show them satisfactorily, would have to be of exceptionally open scale and based on an unusually minute knowledge of local conditions. One or two cases of real vortical storms do seem, however, to have been observed in the British Isles, notably a storrn on March 24, 1895, which caused much damage to trees in the Eastern Counties. Wind Rose. A diagram showing, for a definite locality or district, and usually for a more or less extended period, the proportion of winds blowing from each of the leading points of the compass. As a rule the " rose " in- dicates also the Strength of the wind from each quarter, and the number or proportion of cases in which the air was quite calm. The simplest form of wind rose is represented by the accompanying figure, in which the number or proportion of winds blowing from each of the principal 8 points of the compass is represented by lines converging towards a small circle, the proportion of winds from each direction being indicated by the varying length of the lines. The figures in the circles give the number, or percentage, of cases in which the air was calm. 352 Glossary. A "rose" may be, and occasionally is, devised such a manner as to indicate the relation of other meteoro- logical phenomena, such as cloud, rain, fog, &c., to the direction of the wind. As a result of an investigation recently undertaken in the Meteorological Office a series of roses has been constructed showing that on the western and southern coasts of the British Islands the bulk of the fogs experienced are sea fogs, i.e., they occur with winds blowing (sometimes with considerable strength) from the surface of the ocean. On the north and east coasts summer fogs also come from the sea, but winter fogs more often from the land. The " roses " show further that over the inland parts of England the fogs are radiation fogs, and are accompanied by calm or very light winds blowing from various quarters. The publications of the Meteorological Office have in- cluded from time to time wind roses of various designs. Specimens of these are reproduced on the two succeeding pages. s j he Wind-Pose. 353 Showing- Average Di- rection of Wind by shaded areas converging- towards centre of diagram and Strength of Wind in numbers of Beaufort Scale by dots. Prevalence of Calms indicated by diameter of central circle. (Reproduced from Wind Charts of North Atlantic," published in 1859.) Relative preva- lence of Wind for each point of the compass shown by length of arrows converging to- wards centre. Force of Wind by curve inter- s e c t i n g wind arro\* s. Calms by pro portion of shadec to unshaded por tions of large cen tral area. Additional in formation relate* to Ocean Currents and other matters of interest t< Navigators. (Reproduced from. " Charts of Meteorological Data for Lai. 20 N. to 10 S Long 10 40 W. t " published in 1876.) 13201 M 354 Glossary. Irregularly shaped areas around outside circle indicate relative prevalence of Wind from various directions. Radial lines converg- ing towards central area indicate Wind Force. Shaded portions of outlying areas indicate prevalence of Gales. Small central circle indicates by its diameter the proportion of Calms. Star points around this circle indicate by their length the num- ber of observations. (Re produced from " Charts ofttie Ocean District adjacent to the Cape of Good Hope " published in 1882J) Arrows fly with the Wind towards centre of diagram. Frequency . of Wind from each direction is indicated by length of arrow. Force of Wind is indi- cated thus : Light, Moderate, Strong. Figures in centre of diagram indicate percent- age of Calms. {Similar to Wind Roses published in current issues of Monthly Weather Report and in " Monthly Meteorological Charts of North Atlantic and Mediterranean"^) Conversion Tables. 355 II co vo CM ON vo CM CO Vh V$- vO CO CO CO co co o to CO co CO CO ON O w ^t r* CO CO ^t" >-n vO vb K. t^oo ON p O CM -*J-vO CO i-O b CO SCM -^-vO CO b to O M rj-vO CO b CO O c ThvO oo CO 8, O CM -^-VO CO b CO II vovb ** x>oo o M co O vO to O o>b o ^ c< r^ co O t-. > ^- N CO Vt- ri- vo M t^ Tf HH 00 vb vb t-^oo oo W d OJ W W rj- t-l OO VO C* ON b b M w CM CO CO CO co 8 CM Tj-vO <?O ( O "<*-vO CO b CO O M rj-vo CO b CO O <N ^VO CO CO b CO O CM T}-VO co b CO 3 M O vp to O i ;* O f* T- M CM CO Cp T$- vO r^ ^t- t-t co '^~ M oo m w co ON ONO ^ M vo C<| ON vO CM n to to ^ vo SJ ON ON O O O t-i a; II O N "tf-VO OO VO O C< Tj-vO CO O N -^-vO co O CM T^-VO CO CO O CM -^-vO co ^ CM CM co Vj- rh OO OO CO CO OO ^(- M CO VO w vOSO VO t^-OO OO vo M CO vO CO ON b b '** co CO ON ON ON CM CM to Vh rh vovb o -f^oo co g p w O c -^-vo co O W -^-vO oo O W ^"VO oo CO ON O CM ^VO CO ;3 OD t-i oo yo HI co vo d CO vO tX co ON ON b HH HH f>J CO CO Tj- SSvb K< vo to O t>- "4- oo ON b b >-" t^ i^OO OO CO a H-t O M Tj-vO 00 VO oo O W -^-vo CO vp CO O W ^-VO 00 CO O CM T^-VO 00 00 oo a CM Tj-VO OO m 00 , CM 00 vo ex, ON VO W ONVO to ONVO co O t> CO O t->. CO O t^Th t-Th || oo oo ON O O 1-1 M W CO 4- vO vO vO vO vO VO vO VO vo vO vO vO VOVO VO i-i CM to to Th vO vO VO vO vO 1 CM -^-O CO CO O w -^-vO oo w O M ^VO 00 CO O CM T*-VO OO co CO O CM -^-vO co 7*- 00 w ! 356 Glossary. O VO M j>. CM OO co ON * O vO M J>- CNJ 00 CO ON rhO M t^ CM OO CO c8 oo oo ON o^ o O -! t-H d CO CO Tf Th U^ VO vo vo t^oo oo ON ON O O M CO CO ro CO cO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO O vo M t*- CM OO CO ON rj- O VQ M t^ CM OO co ON -i- O vo M t>. M OO CO O o vr unvo O t^ CO CO CO cO CO r^oo oo ON O CO CO CO co rj- O hH HH CM CO co co rf- vo xr> ^- "* Th Tj- ri- VO O t^ J-^00 TJ- -t- Th Th T^- EH o vnvo r^oo ON M CM CO Tj- O O O ff^^?? O M W CO Tt- vnvO t^OO ON M ^ co oo co ON rh O O M t^ CM oo co ON rh o vo M r^ CM OO co ON T|- cS TJ- rrf- xr> vr><o ON ON ON ON ON CO CO CO CM CO vo t^oo oo ON O"i ON ON ON o^ CM CM CM CM W ON O O O CM co CO CO co CM co co rj- rt- O O O O CO CO CO CO CO w> LOVO VO J^ O O O O CO CO CO CO CO l-l .t^ CO OO CO ON rt-O w J>. C-l OO CO ON T- O vo 1-1 i^ CM OO co ON Tf- 9 W hH CO CM CO CM CO CO CO CM CO r^- U-J iTiVD CM CM CM CN C^ VO t^ t^O OO CM M <N CM C<J ONO O W l-l CM CO CO CO CO CM CM CO CO -<^- CO CO CO CO CO & O M CO CO Tj- vnvo t^oo ON O M CM CO Th vovo r-^oo ON O I-H co co T}- o CO 00 CO ON rh O vo HH t>. CM OO CO ON ^- O VO t-t t-^ C^ OO co ON -rj- O vO 06 O O >H i-" CM OO OO OO OO OO CM CM CO CO CO CO CO ^h Tj- W> 00 00 OO 00 00 CM W W CM CM ir>>O O t^oo OO OO OO OO CO CM CM W CM CM OO ON ON O O OO OO CO ON ON W CM W CM CM M M CM CO CO ON ON ON ON ON CM CM CO CO M <M OO CO C^ Tj- O VO ** t*. CM oo co ON ^- O VO t-t t^ CM 00 co ON rf- O vo O o t>. t^oo oo ON O M M 01 CM CO CO rt-xn vnvo vo t^ r~* oo oo ON O O f=H o w>v5 t^-OO ON O H CM CO TJ- xr> vr \f) \n xr> vnvo t^oo ON xr> \jt vn UTJ \r> O - CM co -ri- w>vo t-^oo ON vc vO vo VO VO CO ON rh O ^D -< t>* CM OO CO ON "* O vo M t>. CO OC CO ON "tf- O vo M t^ 03 O VO t^OO 00 ON ON O O "-< M CJ CO CO rj- ^t- tn to vo vo t>.oo 00 ON ON w 01 o* a o* CM CM CM CM CM CM CM CM W <N CM CM C<J CM CM CO CO CO CO CM t^ M vo O -=t- ON cooo CM JT>. H-4 VO -O SO - ^ CM OO CO ON Th O VO M t-* ? VO ^D >JT> A^> ^J- 1 1 1 1 1 CO CO CM CM HH Mill HH O O M 1 1 + M CM CM CO CO ^t- u-> vr>vo vo ft O M M CO rj- W CM CM CM C< vnvo t^oo ON CM W CM W N O 1-1 CM co rh CO CO CO CO CO xr>vO J-^oc ON CO CO CO C*5 TO O -i CO CO Tj- Index to Tables. 357 INDEX TO TABLES INCLUDED IN THE GLOSSARY. Abbreviations. List of Adiabatic expansion. Change in temperature of air on. Aqueous Vapour Mass of in saturated air ... Pressure of do. do. ... Amount and Pressure of, at Kew. Clouds. Types of Correlation Coefficients. Selected examples ... Density Evaporation of Water Fog and Mist. Average No. of observations of Fog in British Isles. Frequency in English Chan- nel of Gales at some British anemometer stations ... Seasonal variation of Gradients. Steep pressure Gusts. Range of fluctuation of Strongest recorded Hurricanes, cyclones and typhoons recorded in various parts of the World. Insolation. Calculated Insolation reaching Earth. Pressure Units. Conversion table Rain. Consecutive hours of rain in 1912 Rainfall during the four seasons in S.E. England and X. Scotland. Day rainfall and night rainfall Relative Humidity. Frequency of occurrence of various values of Sunshine. Percentages of possible duration of Temperature. Boiling points of water at various pressures in the atmo- sphere up to 8,000 feet. Conversion table Normal weekly temperatures for S.E. England. Some common temperatures ... 13204 See under : p. 2. Adiabatic, Aqueous vapour. Do. Absolute Humidity. Clouds. Correlation. Buoyancy. Evaporation. Fog. Frequency. Gale. Gale. Gradient. Gusts. Gusts. Hurricane> Insolation. p. 355. Persistent rain. Seasons. Seasons. Relative humidity Sunshine. Hypsometer. p. 356. Seasons. Absolute temperature. N 358 Glossary* Thunderstorms, Immunity from Upper atmosphere. Normal pressure at var- ious heights. Average temperature at different levels. Average values of pres- sure, density and tem- perature of air over regions of high and low pressure. Normal factors for the density of air at various heights. Limit of height for the expenditure of ballast. Depression produced on airships by rain or snow. Wind. Monthly normals of wind velocity at some French and British stations. Normal hourly wind velocities at Kew Spells of N.E.-S.E. winds of specified duration over S.E. England and N. France. Frequency of winds from different quarters over S.E. England and N. France. Distance between isobars for various geostrophic winds. Equivalents of wind force Wind direction at Suva, Fiji Hourly velocity at the top of the Eiffel Tower. See under : Thunderstorms. Ballon-sonde. Ballon-sonde. Density. Buoyancy. Buoyancy. Buoyancy. Normal. Normal. Frequency. Frequency. Isobars. Beaufort scale. Trade winds. Wind. Printed .under the authority of His Majesty's Stationery Office BY DARLING AND SON, LIMITED, BACON STREET, E.2. * METEOROLOGICAL GLOSSARY," M.O. 225 ii. (Fourth Issue.) CORRIGENDA. In the Fourth Issue the plates representing various forms of pressure distribution which in the previous issue were placed with the separate articles : ANTICYCLONE, COL, DE- PRESSION, SECONDARY DEPRESSION, V-SHAPED DEPRESSION, WEDGE in alphabetical order are now put together in the article ISOBARS, and should have been re-numbered in order to correspond with the text. But the re-numbering, and in consequence the order, has failed. The following corrections should therefore be made in the numbering and order of the plates : DEPRESSION should be Plate XI. instead of XIII., and face page 174. SECONDARY DEPRESSION should be Plate XII. instead of XIV., and face page 175. ANTICYCLONE should be Plate XIII. instead of XL, and face page 176. COL should be Plate XIV. instead of XII., and face page 177. The foot-note on page 177 should be omitted. Page 75, line 1, "Sir Gilbert Walker" should read -Dr. Gilbert Walker." Page 132, last line, " 65 millibars " should read " 5 millibars." P.ige 256, line 24, " John Hadley," etc., should read " George Had ley, who was a brother of John Hadley," etc. Page 262, line 18, " years" should read " year." Page 271. In the column headed u Depression of Wet Bulb," the temperature scale F. should be indicated. Page 295, line 1, R 1 should read R 2 . line 12, /o<r should read <r. line 26, V 1 should read V'. Page 308, line 27, "Earth's mass M' " should read "earth's mass plus that of atmosphere (M + M')." Page 330, line 28, should read " 77 (288) 4 x lO" 12 ." SW Page 340, line 7, the equation should read 3G = . a C23632 12.) Wt. 7924185. 6000. 6/19. D & S. G. 3 THIS BOOK IS DUE ON THE LAST BATE STAMPED BELOW AN INITIAL FINE OF 25 CENTS WILL BE ASSESSED FOR FAILURE TO RETURN THIS BOOK ON THE DATE DUE. THE PENALTY WILL INCREASE TO 5O CENTS ON THE FOURTH DAY AND TO $1.OO ON THE SEVENTH DAY OVERDUE. 101941 M DEC 5 t944 >X !NTEB-OBfi\ G-r 423582