* CLIMATES AND WEATHER OF INDIA A PEACTIOAL GUIDE TO THE CLIMATES AND WEATHER OF INDIA, CEYLON AND BUEMAH AND THE STOKMS OF INDIAN SEAS BASED CHIEFLY ON THE PUBLICATIONS OF THE INDIAN METEOROLOGICAL DEPARTMENT BY HENEY F. BLAKFOKD, F.K.S., F.K.MET.S. ** LATE METEOROLOGICAL REPORTER TO THE GOVERNMENT OF INDIA HONORARY MEMBER OF THE IMPERIAL AND ROYAL METEOROLOGICAL SOCIETY OF AUSTRIA AND OF THE METEOROLOGICAL SOCIETY OF THE MAURITIUS CORRESPONDING MEMBER OF THE GERMAN METEOROLOGICAL SOCIETY pontoon MACMILLAJST AND CO. AND NEW YORK 1889 Printed by R. & R. CLARK, Edinburgh, TO LIEUT.-GENERAL RICHARD STRACHEY R.E., C.S.I., F.R.S. PRESIDENT OF THE ROYAL GEOGRAPHICAL SOCIETY THE ORIGINAL PROJECTOR AND PROMOTER OF THE METEOROLOGICAL DEPARTMENT OF THE GOVERNMENT OF INDIA TO WHOSE ADVOCACY ITS ESTABLISHMENT IS MAINLY DUE THIS LITTLE BOOK WHICH IS THE OUTCOME OF THE WORK OF THAT DEPARTMENT IS CORDIALLY INSCRIBED BY THE AUTHOR 130019 PBEFACE OWING mainly to the systematic work of the Meteorological Department, established by the Government of India in 18*75, we now possess a far better knowledge of the weather and climate of India than of those of any other tropical country, and, in some respects, better than of those of many parts of Europe ; thanks to the greater simplicity of the processes concerned, and to the prominence and regular recurrence of the more striking phases of the seasons. But to the great majority of the lay public, of those to whom information of this kind is a matter of daily interest and importance, this knowledge is as a sealed book. It may be found, indeed, embodied in the volumes periodically issued by the meteorological office, or scattered through the journals and transactions of various scientific societies : but all publications of this class are addressed to students and co-workers in the field ; and besides being, in most cases, too technical in form to approve itself to the tastes and requirements of the general public, the literature is too voluminous, and the information it contains too diffuse, to be readily accessible to those who have neither the time nor the opportunity to search through so wide a field. The want of some work that shall afford, in a compen- dious and apprehensible form, such information as is con- stantly in demand by those engaged in administration, in VI 11 PREFACE agriculture, sanitation, engineering works, and the like, and especially in the navigation of Indian seas, or by those who wish to follow, intelligently, the current reports of the weather issued daily at Simla and Calcutta, has long been obvious, and has been definitely expressed in the anniversary address of the President of the Asiatic Society of Bengal, delivered at the annual meeting of the Society in February 1 8 8 7, in the following terms. After noticing the errors of the older storm literature, as represented by the works of Eeid and Piddington, he remarks : "It is much to be desired that steps should be taken by some recognised authority, to prepare a simple and popular manual on the subject [of storms], based on the sounder knowledge which modern science has acquired. And indeed such a work is also needed for the meteorology of the land, to enable the lay public to understand and intelligently appreciate the mass of meteorological literature which the Government Gazettes and the entire press so lavishly place before us. We want something more popular than the very valuable ' Vade Mecum ' of Mr. Blanford, not a scientific investigation of principles, but a brief, clear, and comprehensive explanation of the observations ordinarily made, their objects, uses, and mode of application." It is to meet this want that the present work has been taken in hand. It is not addressed to meteorologists and physicists, although some of the data and facts adduced may be not without interest to those who are principally concerned with meteorology in its scientific aspect ; but more particularly to agriculturists, medical officers, engineers, pilots, and other seafaring men, and to those others of the general public to whom the weather and the climates of India and of its seas are practical and not scientific objects of interest. I have therefore selected from the abundant materials now available, and from the general conclusions from these, PREFACE ix worked out by Messrs. Eliot, Hill, and other Indian meteor- ologists, such portions as are of practical importance to inquirers of this class, and have endeavoured to set them forth in clear and concise language, avoiding, as a rule, all discussion of a technical character, and all technical forms of expression that have not become more or less familiarised to the public through the medium of the weather reports and other literature of a popular character. The explanation of these latter is one of the objects of the work. The work consists of two parts. The first gives a con- cise general description of the kinds of observations made at Government observatories, and the information they directly afford. This constitutes what may be termed the elements of the climate and weather. The second deals with the practical applications of weather knowledge. In the first part, the intensity of the sun's heat in India, the air temperature, humidity, cloud, winds, and rainfall, together with the principal variations they present in differ- ent parts of the Empire, and the changes they undergo in the course of the year, the behaviour of the barometer and the methods of interpreting and utilising its teachings, and the constitution and nature of storms, are briefly described in separate sections. The second and more extensive part is devoted to a detailed notice of the local climates first, those of the chief hill stations, and then those of the different provinces of the low country. Then follow the weather characteristics of the three Indian seasons, illus- trated by the information furnished by the daily weather charts published at Simla ; the tracks and seasons of storms on the seas around India, with some practical deductions for the guidance of seamen; and finally, the statistics of rainfall, evaporation, and wind pressure, which are chiefly required by engineers in dealing with questions of water x PREFACE supply, drainage, and the stability of structures. In this section an attempt has been made for the first time to estimate the average rainfall of some of the chief river basins, a datum of great importance to physical geographers, not less than to engineers. In the Appendices are given, in a condensed form, some portions of the statistical data on which the above de- scriptions are based, much of which have been computed specially for this work. They include the more import- ant climatic elements of 92 places in India, Burmah, and Ceylon, a list of well-recorded storms in the Bay of Bengal, with their approximate places of origin, tracks, and rates of movement, and tables of the average, maximum, and minimum monthly rainfall at 114 places. Among the friends to whom I am indebted for informa- tion and assistance, I must especially mention Mr. James Forrest, Secretary of the Institution of Civil Engineers, to whose kindness I owe much of the information respecting engineering works and professional observations in India, Africa, and Australia, which will be found quoted in the final section of Part II. Also Captain Henry Toynbee, late of the London Meteorological Office, and his successor, Lieutenant C. W. Baillie, K.K, to whom I am indebted for valuable assistance in preparing the section on Storms in Indian Seas. FOLKESTONE, April 1889. CONTENTS PART I. ELEMENTS OF CLIMATE AND WEATHER 1. THE HEAT OF THE SUN IN INDIA PAGE Use of the Sun Thermometer Insolation Temperatures at Hill Stations and on the Plains Duration of Sunshine . . 1 2. THE TEMPERATURE OF THE AIR Observation of Temperature Mean Temperature Diurnal Varia- tion and Range Irregular Changes Annual Range Hottest and Coolest Months ; Annual Variation .... 6 3. THE TEACHINGS OF THE BAROMETER Effect of Elevation Annual Fluctuation Diurnal Fluctuation Magnitude of Variations with Weather Temperature Correction Barometric Changes with Weather Distribution of Pressure Barometric Depressions Reduction to Sea -level Barometric Charts and Isobars Barometric Gradients Persistent and Tem- porary Differences Summary . . . . .19 4. THE WINDS Lightness of the Winds in India Wind Measurement Beaufort's Numbers Wind Pressure Diurnal Variation Annual Variation ; the Monsoons Relations of Winds to Barometric Pressure ; Cyclonic and Anticyclonic Systems . . . .30 5. DAMPNESS AND DRYNESS Relative and Absolute Humidity Variations of Humidity in India Diurnal Variations Relation to the Winds Geographical Distribution of Humidity Annual Variation . . .46 xii CONTENTS 6. HAZE, FOG, AND CLOUD PAGE Haziness of the Sky Dust Haze Fog Forms and Nomenclature of Clouds The Teachings of Clouds Cloud Proportion Distri- bution . . . . 53 7. THE RAINFALL - The Contrasts of Indian Rainfall Geographical Distribution- Seasonal Distribution Statistics of Monthly Rainfall ; Quantity and Frequency Average Heaviness Excessive Rainfalls An- nual Variability The Question of Cyclical Variation Droughts . 63 8. STORMS Thunder-storms and Dust-storms Cyclones and Cyclonic Storms Tornadoes . . .... 81 PART II. CLIMATES AND WEATHER OF INDIA IN RELATION TO HEALTH AND INDUSTRY 1. THE CLIMATES OF INDIA Their Heterogeneity Climates of the Hills : Quetta ; Leh ; Murree ; Simla ; Chakrata; Ranikhet ; Darjiling ; Shillong; Pach- marhi ; Chikalda ; Mount Abu ; Mahableshwar ; Ootacamund ; Wellington ; Newara Eliya ; General Characters of Hill Climates Climates of the Plains : The Punjab ; Sind ; Rajputana ; the North-west Provinces and Oudh ; the Central Indian Plateau ; Behar and Chutia Nagpur ; Bengal and Orissa ; Assam and Cachar ; Central Provinces (Nagpur), and Berar ; the West Coast of India (Konkan and Malabar) ; Khandesh, the Deccan and Mysore ; the Carnatic ; Ceylon ; Burmah ; Andaman and Nicobar Settlements . . . ... . .95 2. WEATHER AND WEATHER REPORTS Climate and Weather Anticyclones and Cyclonic Depressions Barometric Surges Weather of the Cold Season Weather of the Hot Season Weather of the Summer Monsoon Weather Reports and Charts . . . . ... 197 CONTENTS 3. STORMS OF INDIAN SEAS PAGE Piddington's Work Storms of the Bay of Bengal ; Storm Seasons ; Place of Formation, Tracks, and Rate of Travelling Incurvature of Winds and Bearing of Storm Centre The Barometer in Storms Weather around the Storm Area Practical Conclusions and Rules for Guidance Storms of the Arabian Sea ; Storm Seasons on the West Coast of India ; Storm Tracks of the Arabian Sea Storm Warnings and Signals ; Bay of Bengal ; Bombay Coast . . 223 4. RAINFALL IN RELATION TO WATER SUPPLY AND DRAIN- AGE EVAPORATION AND WIND PRESSURE Mean, Maximum, and Minimum Monthly Rainfall Mean and Maximum Daily Rainfall Special Occurrences Special Hourly Rainfalls Proportion of Surface Drainage to Rainfall Evapora- tion from Water Surfaces Rainfall on Indian River Basins ; Rivers of Central India Tableland ; Rivers of Western Bengal ; Rivers of Orissa ; Rivers of the Deccan ; Rivers of the Carnatic ; The Tapti and Nerbudda ; The Looni Wind Pressure . 256 APPENDICES I. Climatic Statistics of 92 Stations . . . .290 II. Tabular Lists of Storms in the Bay of Bengal . . 337 III. Mean, Highest, and Lowest Recorded Monthly Rainfall at 114 Stations in India, Burmah, and Ceylon . . . 342 IV. Directions for the use of Storm Signals at Ports of the Bay of Bengal .358 V. Notification of the Government of Bengal relative to Storm Signals for Calcutta and the Hooghly . . , 360 INDEX .... 365 PAKT I ELEMENTS OF CLIMATE AND WEATHEK THE HEAT OF THE SUN IN INDIA Use of the Sun Thermometer. The question is often asked of meteorologists, "What is the temperature in the sun ? " The reply must be, " That depends entirely on the instrument with which you measure it, and on its surround- ings ; " in other words, the question, so put, does not admit of a definite answer. Still the fact remains that the sun is- hotter in India than in England for example, hotter appar- ently in May than in January, and there must be some way of expressing this difference by a reference to the thermo- meter. And that, in point of fact, there is such a way stands in evidence in the reports of the chief observatories, published daily or weekly in the newspapers, which give, together with the temperature of the air in the shade, the reading of a thermometer exposed to the sun. This ther- mometer is an instrument specially constructed for the purpose. It has the bulb coated with lamp-black, and is- enclosed in a large glass tube hermetically sealed after exhaustion, and it is exposed to the sun on a support four feet above the ground, as far as possible from buildings. Any other kind of thermometer would probably show a different temperature ; and, indeed, thermometers of appar- ently similar construction are often found to differ many degrees in their readings in the sun, notwithstanding that,. B ELEMENTS OF CLIMATE, ETC. in the shade, they read alike. This latter discrepancy can be approximately ascertained and allowed for ; but it is not possible to ascertain the exact effect due to variations in the surrounding objects and in the character of the ground beneath the instrument ; and thus it is found that, under equally clear skies, the excess of the sun thermometer read- ing above that which shows the temperature of the air in the shade, in other words, the heating effect of the sun, shown by the former instrument, is greater at some stations than others. Insolation Temperatures in India. In India this difference is, as a rule, from 50 to 70 Fahr. It does not vary much in the course of the year so long as the sky is unclouded ; but it is higher at hill stations than on the plains, in the cold season considerably higher, showing that, in opposition to the prevalent belief, the power of the sun is really greater on the hills, and that protection of the head and back is at least as necessary a precaution at Simla or Mussoorie (and still more so at greater elevations), as on the plains of India. At Leh, 11,500 feet above the sea, Dr. Cayley succeeded in making water boil by simply exposing it to the sun in a small bottle, blackened on the outside and placed inside an empty quinine phial to protect it from cooling by the wind. As examples of the temperatures recorded with thermo- meters of the kind above described, on the hills and plains of India, the following table shows the highest readings registered in each month of 1885, by no means a hot year, at Simla, Murree, Lahore, Calcutta, and Madras ; and also the greatest differences of these and the highest temperatures in the shade on the same day : [TABLE HEAT OF THE SUN MURREE. SIMLA. LAHORE. CALCUTTA. MADRAS. Temp. co Temp. i Temp. i Temp. i Temp. I in o in o in o in JM in o Sun. w Sun. X H Sun. X H Sun. 1 Sun. X H January 122 79 134 80 128 62 137 55 146 61 February 132 92[?] 136 84 152 76 145 64 146 62 March 144 74 140 75 168 79 155 63 161 69 April 138 82 146 75 172 79 159 62 150 59 May 145 83 144 72 171 73 165 66 150 55 June 151 72 152 70 175 68 162 71 151 53 July 155 71 146 70 173 69 160 71 153 54 August 156 77 145 69 172 75 155 67 154 59 September 147 76 144 73 163 67 155 66 160 65 October 139 68 143 72 157 66 155 67 149 60 November 133 69 131 70 147 60 146 62 148 62 December . 129 78 126 75 146 67 140 65 149 65 In England, the Eev. F. Stow, who has paid much attention to this subject, found that thermometers of the same kind as those used in India, and exposed in the same manner, seldom read above 140, and states his belief that 154 is the highest temperature registered in five years. This would seem to show that the highest temperatures actually recorded in England do not differ very much from those of the hill station Murree, but their excess above the temperatures of the air would seem to be much less, and rarely, if ever, to amount to 70. This latter remark may apply equally to the case of Madras, where, indeed, the excess of the sun temperatures over those of the air is not greater than is occasionally recorded in the summer in England ; but then the air tem- perature itself is 20 or 30 higher; and it would be a grave mistake to suppose that exposure to the sun in Madras can be incurred with the same impunity as exposure to the summer sun in England. On the other hand, experience shows that the more fervid intensity of the sun's rays at the hill stations, in January and February, may be borne not only without danger but even without serious discomfort in a cool atmo- sphere, provided the head and spinal column are adequately ELEMENTS OF CLIMATE, ETC. protected. And it may be concluded that the temperature readings of the sun thermometer, which show the tempera- ture of the air, plus the direct heating effect of the sun, notwithstanding the artificial and purely conventional char- acter of the arrangements for determining them, probably afford a better criterion of the stress imposed on the animal system, than do the figures expressive of the sun's heating effect alone. The glare and intense light reflected from all surrounding objects, and especially from the ground bathed in Indian sunshine, are perhaps not less trying to the un- acclimatised visitor to India, than the heat which they accompany. Duration of Sunshine. The greater intensity of the sun is, however, only one element of its distinctive action in India. Another of not less importance is the greater frequency and duration of sunshine. At St. Aubin's, in Jersey, where, according to a table of sunshine records published by Mr. E. H. Scott, the number of sunshiny hours exceeds that of any station in the British Isles, the average total of sunshine, on the mean of five years, was 1853 hours in the year, or rather over 5 hours per day. At Calcutta, on the mean of three years, it was 2732 hours, or an average of 7^ hours per day; and at Allahabad, also on the mean of three years, 3053 hours or an average of 8^ hours per day. Notwithstanding the greater length of the summer days in the British Isles, in no single month did the number of sunshiny hours in Jersey equal the May average of either Calcutta or Allahabad, and only from June to September, the rainy season of Northern India, did the total duration of sunshine in Jersey surpass that at the Indian stations. The following table, extracted from the Eeport on the Meteorology of India for 1885, gives the data on which these statements are based, and also, for comparison, the corresponding data on St. Aubin's for the average of five years : HEAT OF THE SUN d ctf HS 1 5 cd s ! S o> a p 1-9 3 1-2 to s ^ 1 .J s >' K W ' Q 1 rt - fHours of bright sun- \ | shine. . . f 275 255 294 297 293 154 128 113 158 258 254 253 2732 g X Average per day . "3 I Percentage of possible \_ ^> V. sunshine . . J 8-9 82 9-0 80 9-5 79 9-9 78 9-4 72 5-1 38 4-1 31 3'6 28 5'3 43 8-3 72 8'5 77 8-1 76 7'5 62 3 /Hours of bright sun- 1 J | shine J 263 269 309 307 317 213 162 148 233 286 296 250 3053 jH -^ Average per day . 8-5 9'5 10-3 9'9 10-2 7-1 5'2 4'8 7-8 9-2 9'9 8-1 8'3 5 j Percentage of possible )_ 5j V sunshine . . j" 80 86 84 81 77 52 39 37 64 81 91 77 70 .^ /"Hours of bright sun- 1 5 j shine ...... | 61 93 160 187 267 225 231 227 164 125 74 39 1853 9 -4 Average per day . 2-0 3'3 5-0 6'2 8'6 7'5 7-4 7'3 5'5 4-0 2'5 1-3 5-1 ^ Percentage of possible \ * V sunshine . . ) 23 33 43 45 56 47 48 52 44 38 27 15 39 We have no records of sunshine from Southern India to compare with the above, but it appears from the estimates of the quantity of cloud which have been made for many years past in all parts of India, that, on the general average of the year, the skies are appreciably more cloudy in Southern than in most parts of Northern India, and es- pecially Central and North-western India, and it may be inferred, therefore, that the quantity of sunshine is less. If we take 1000 as representing the total expanse of the sky, the average proportion of cloud in the Punjab, the North- west Provinces, Lower Bengal, and the plains of Madras are respectively as follow : Province. No. of Stations. Cloud Proportion. Punjab 9 283 N.W. Provinces 9 303 Lower Bengal 7 453 Madras 10 507 These figures, being based upon mere eye-estimates, cannot lay claim to the same precision as those of the duration of sunshine, which are derived from self -registering instru- ments, but the data of different stations are very accordant, and as the figures are the averages of, in most cases, from ELEMENTS OF CLIMATE, ETC. 14 to 17 years' observations, I believe they maybe accepted with confidence. THE TEMPERATURE OF THE AIR Temperature Observations. Mean Temperature. In what I have to say on this subject it must be understood that I speak of the temperature of the air outside the house, but in a place well shaded from the sun, where the air is not stagnant, but has free movement from all sides. In practice, at all Indian observatories, the thermometer is suspended under a roof, generally of thick thatch, which is supported either on posts or pillars, so that, while the ground beneath it is well shaded, the wind has free access to the instrument. The temperature shown by a thermometer in such a position rarely agrees with that of a house or even of a verandah, in either of which it is generally lower during the day and higher at night. The observations, usually recorded under these circum- stances, are the highest and lowest temperatures reached in the 24 hours, at whatever time they may occur, and, in addition, the readings of the actual temperature at certain fixed hours ; most commonly 1 A.M. and 4 P.M. From these four observations it is usual to compute what is called the mean temperature of the day. This is assumed to represent the average of all the fluctuations during the 24 hours, not that of the four recorded readings only. And this result is arrived at by various devices founded on a more complete system of observations. As the consideration of these belongs to the technicalities of meteorological observation, they need not be discussed here. Suffice it to say that, although for any single day the result may differ by a degree or two from the real value, on the average of several days these errors should neutralise each other ; so that, if the method adopted be valid, the mean temperature of a whole month should be well within a degree of the truth. AIR TEMPERATURE Diurnal Variation and Range. In fine, cloudless weather, the changes of temperature during the day are remarkably regular in India. The air is coolest a few minutes before sunrise. No sooner does the sun appear above the horizon than the temperature mounts rapidly, the rate of rise gradually increasing up to about half-past 8 o'clock (somewhat earlier or later, according to the hour of sunrise). After this it continues rising, but at a de- creasing rate, until it reaches its highest point about 2 P.M., or somewhat later. It then begins to fall, at first slowly, and afterwards more rapidly, till about sunset, when the fall is as rapid as was the rise in the forenoon. Later on the fall slackens, and after midnight continues slowly and steadily till a little before sunrise. The course of this variation is illustrated by Fig. 1, which represents the average diurnal curves for Calcutta in March and July. M 16 18 M 48 + 6 4-4 \ Fig. 1. DIURNAL VARIATION OF TEMPERATURE AT CALCUTTA. A. March. B. July. The range of temperature during the day depends chiefly on the dryness of the air and the clearness of the skies. It is much greater in the interior of the country, and especially in North-western India, than on the coasts and in the ELEMENTS OF CLIMATE, ETC. neighbourhood of the sea ; and, as a general rule, is greatest in the driest spring months and least in the rainy season. In Sind and Baluchistan, and throughout the dry tract to the west of the Jumna and the Aravalis, the daily range of the thermometer is greatest in October or November, when the difference between sunrise and the afternoon averages not much less than 35 and sometimes exceeds 40. In the northern districts of Bombay, in Kathiawar, Guzerat, Khandesh, and the Northern Mahratta country, the driest time of year is in the earlier months, when land winds from between west and north-west blow most steadily ; and accordingly, in this part of India, the greatest range of the thermometer is in February or March, when it averages nearly 35 at such places as Poona and Malegaon, and in some years, on the mean of a whole month, is as much as 38. In Lower Bengal, March is, on the whole, the driest month ; but at Calcutta and Dacca the average diurnal range does not exceed 21 or 22, and very rarely reaches 30 on any single day. In Behar and the North-west Provinces it averages from 28 to 32 both in March and April. In the interior of Southern India, in February, March, and April, the mean daily range is from 24 to 30, being slightly greater in February than in the succeeding months ; but on the coast of Madras it is much less, not more than 19 at Madras, and 13 or 14 at Negapatam; and at this latter station it is greatest in June, when it amounts to 19 on an average. These variations of temperature, which take place daily with much regularity, much exceed those experienced in most parts of Europe, and constitute an important element of the climate. In different parts of England the range varies from 10 to 20 in the summer, when it is greatest; and in Europe, only in the climate of the Spanish tableland does it amount to 27, and has been observed in a single month as high as 3 1 . 1 . At hill stations, the range varies according to the position of the place. If, like Simla, Mussoorie, and most of Hima- 1 Buchan, Introductory Text-book of Meteorology. AIR TEMPERATURE layan sanitaria, it is situated on a ridge, the difference of the early morning and afternoon temperatures is less than on the plains ; but the same rule by no means holds good if it is on an elevated tableland like Quetta, still less when it is in a deep valley like Leh and Kailang. In some of these cases, in certain months, elevation brings about an exaggeration of the diurnal range. As these changes of temperature during the day are a matter of much importance to invalids, and indeed in their influence on health and comfort generally, it may be useful to give the mean diurnal range of each month at all those elevated stations at which observatories exist or have existed, and which afford, therefore, trust- worthy data. I give for comparison also, in each case, the corresponding ranges at some neighbouring station on the plains. MEAN DAILY RANGE OF THE THERMOMETER SlND FRONTIER. N.W. HIMALAYA AND PUNJAB. N.W. HIMALAYA AND N.W. PROVINCES. |f ! | o| *~ -I wj? g^s it **"3 ~ "22 II "$* 1 = ~' c K ^ 1 3 BTS "^ S || *t 1 to 6 s e o 1 M |- 6 8| 2^ M^- Jan. 24 31 25 14 14 16 17 25 16 13 25 Feb. 21 29 26 14 17 17 22 25 16 15 26 Mar, 24 30 26 15 19 18 19 26 18 16 29 April 26 31 > 26 17 19 20 17 29 19 17 30 May 30 32 27 17 18 20 22 29 18 17 27 June 31 28 [ 30 17 23 18 25 26 15 14 21 July 27 24 i 29 15 16 12 24 18 10 11 14 Aug. 28 22 30 14 13 11 23 18 10 11 13 Sept. 34 25 30 15 21 14 26 22 12 12 15 Oct. 35 33 29 16 25 17 28 31 15 15 23 Nov. 34 34 27 15 25 18 23 32 16 15 29 Dec. 29 32 26 14 22 17 21 28 16 14 27 10 ELEMENTS OF CLIMATE, ETC. SlKKIM AND ASSAM. CL. PRO- VINCES. BERAR. GUZERAT. i ! NlLGIRIS. "STs |j 11 Pachmarlii (tableland) Hoshungabad (plain) Chikalda (hill ridge) Amraoti (plain) Abu (hill summit) 11 Ootacamund (tableland) Wellington (tableland) Coimbatore (plain) January 9 20 20 25 28 17 28 16 32 31 21 23 February 12 18 22 24 30 18 28 16 30 29 24 27 March 13 17 22 25 32 19 31 17 31 27 21 27 April . 15 16 18 23 31 20 31 18 30 21 20 24 May . 13 13 13 20 29 21 29 18 27 18 17 21 June . 10 11 9 14 19 15 21 15 20 12 14 18 July . 10 10 9 9 11 8 14 9 13 10 13 18 August 10 11 9 8 12 8 14 8 13 12 14 17 September 10 10 8 11 14 9 16 11 17 13 15 19 October . 12 14 11 17 22 13 23 15 27 14 13 17 November 13 17 16 23 28 15 26 17 32 16 13 17 December 12 19 18 25 27 17 26 17 32 22 17 19 ! Even at one and the same station, houses on the crest of a ridge have a smaller diurnal range of temperature than those on the lateral slopes. Thus, at Darjiling, the present observatory, which is on the crest of the ridge, shows a differ- ence of night and day temperatures about 2 less than the old observatory, which for many years existed lower down on the side of the same hill (except from June to September, when the difference vanishes) ; and this, notwithstanding that the new observatory is more exposed than the abandoned site. The figures given above are those of the present obser- vatory. At most of the Himalayan sanitaria, and such places as Shillong, Chikalda, and Abu, the range of temperature approximates to that of midland and southern England in summer ; but, as the table shows, it is greatest in the winter, whereas in Europe it is least at that season. Irregular Changes with Weather. Casual and irre- gular changes of temperature, such as accompany changes of wind and weather, are less frequent and for the most part less marked in India than in Europe and most extra-tropical countries. From northerly winds bringing the extreme heat AIR TEMPERATURE 11 or cold of Central Asia, India is shielded by the Himalaya ; and only in rare cases, and such as are temporary and of short duration, are changes of wind accompanied with such striking changes in the warmth of the air as are commonly experienced in Europe, and still more conspicuously in Siberia and North America. The chief exceptions occur in the hot season, when a more or less prolonged spell of dry weather is followed by a burst of rain ; as, for instance, at the onset of one of those afternoon squalls characteristic of this season, which, in Bengal, are known as Nor'- Westers. The cold wind which blows out in advance of one of these little storms sometimes causes the temperature to drop 15 or even 20 in less than as many minutes, but the full effect of the change lasts little longer than the storm, and even the cool- ing effect of the rain, which usually follows the wind, is in a great measure dissipated by another day's sunshine. On the east coast of the peninsula, the setting in of the sea-wind, after a hot land-wind has been blowing for some hours, is also accompanied with a fall of 10 or 15, but the effect is transitory, and is a relief rather than a trial to the bodily system. More important than these are the rapid changes between consecutive days which accompany alternations of rain and fine weather in Upper India ; changes especially frequent in the summer months, but also occurring at other times of year. This subject, which is one of great importance- in relation to health, has hitherto received but little attention in India, although it is well known that chills are a fruitful source of fever and bowel complaints, maladies peculiarly prevalent in this as in most tropical countries. Perhaps the best mode of representing the important facts is that adopted by Dr. Hann, viz. tabulating the number of times in each month that a rise or fall of definite amount (e.g. 5 or 10) takes place in the mean temperatures of successive days at different places. The daily and weekly weather reports fur- nish data which any one interested in the question may readily turn to account, and he may thus compare the vicissitudes of temperature in different seasons or at different 12 ELEMENTS OF CLIMATED ETC. stations with the corresponding health statistics. The fol- lowing table, which shows the number of times that the change between any two consecutive days at. Calcutta and Lahore amounted to 5, 10, and 15 in the years 1884 and 1885, may be taken as an example of this way of represent- ing an important climatic feature, and it also illustrates the. much greater variability of the Punjab climate as compared with that of Bengal. The cases of rise and fall of tem- perature are distinguished by the letters E and F. (^ C 5 to 10" 28 \ 10 to 15 a c *-s | 1 >^ ci s | 1-9 >> "3 i-a to p < 43 f *J ' 25 1 E. F. .. 3 E. F. 1 1 E. F. .. 2 R.F. 2 2 E F. 3 3 .. 1 E.F. 2 2 E. F. 3 3 1 E. F. 2 E.F. B E. F. 1 E. F. E. F. .. 1 * Ir'ttw g \ 10 to 15 - 1 15+ . Calcutta -f 1884 ~ 5 to 10 Calcutta. 1 18S5 _ 5 c t() 1Q . 1 1 .. 1 1 .. .. 2 4 6 1 1 3 2 2 4 .. 2 .. 2 .. 1 .. 1 2 1 1 1 .. .. 1 .. 1 - In Calcutta there was no change amounting to 7 between any two consecutive days during the two years, and only seven amounting to 5. At Lahore, during the same period, there was one fall of 15, six between 10 and 15, and twenty changes of rise and forty-four of fall, amounting to between 5 and 10. Rapid falls of temperature are, therefore, between two and three times as frequent as rises, and, on the whole, greater in amount. In estimating the influence of such changes as these on health it must be borne in mind that, in most cases, they are superadded to a large part of the regular daily fluctuation, which may render the change to which the body is exposed more than double as great ; and while it is unquestionable that pro- longed residence in India renders the system very sensitive to changes of temperature, the thin clothing habitually worn affords but little protection against their injurious effects. Annual Range. The change of temperature in the AIR TEMPERATURE 13 course of the year is very different in different parts of India, not only in magnitude but also in the duration and rapidity of the rise and fall. As regards magnitude, it is greater in dry parts of the country than in those which are damp and rainy ; and the greater also, the higher the lati- tude, or, in other words, the greater the difference of the longest and shortest day. The greatest range is, therefore, at Leh, which is both the most northerly station and situated in one of the driest climates. The difference of the highest and lowest readings, recorded in the year at the Leh observatory, averages 94, and has been as much as 103 in a single year. On the plains of the Punjab it varies, on a general average, between 80 (at Mooltan) and 86 (at Peshawar) at different stations, and in some years exceeds 92. At the hill stations it is much less, viz. 69 at Murree, and 63 at Simla; and has not much exceeded 75 at the former and 69 at the latter within the last ten years. At Darjiling it is only 47. At Quetta the average range in the course of the year is 84, and the greatest in any one of the last eight years 92; while at Jacobabad, on the border of the Cutchee desert, the average is 86 and the greatest 89. From the Punjab, which is the most northerly and, ex- cepting Sind and Western Eajputana, the driest province of India, the annual range of temperature decreases eastwards to Bengal and Assam, and southwards to Travancore and Ceylon, where, at Point de Galle, it reaches its minimum. At Lucknow and Allahabad the average range is 76, at Patna 68, and at Calcutta it has diminished to 54. Still farther east, in the damp province of Assam, it is 54 at Silchar and only 57 at the more northerly station Sibsagar. In the interior of the peninsula, and particularly in the dry tract of the Deccan, it is greater than on either coast, and greater on the east than on the west coast. Thus it amounts to 62 at Poona, 61 at Secunderabad, 54 at Bellary, and 47 at Bangalore, while on the west coast it is only 34 at Bombay, 31 at Mangalore, 28 at Cochin; and on the east cdas% 52 at False Point, 50 at Masulipa- 14 ELEMENTS OF CLIMATE, ETC. tarn, 48 at Madras, and 40 at Negapatam. Finally, in Ceylon the average annual range is only 25 at Colombo, 31 at Trincomallee, and less than 19 at Galle, where in some years it has scarcely exceeded 17. Such is the uniformity of climate on the south-west coast of Ceylon, that the extreme range of temperature in the course of the year is only about as great as that experienced within 24 hours at Simla, and little more than half that of a single day at Jacobabad. These great variations in the annual range of the ther- mometer depend more on the degree of the winter cold than on the varying intensity of the summer heat. Thus, the lowest temperatures hitherto recorded at Leh have fluctuated between 3 above and 17 below zero (29 and 49 below the freezing point), while the highest in different years have varied between 84 and 93. On the Punjab plains the lowest winter temperatures have been between 24*7 (Pesha- war, 1879) and 40 (Mooltan, 1884); in general, a little below the freezing point at the more northerly, and a degree or two above it at the more southerly stations ; while the highest summer readings have varied between 112 and 122. At Allahabad the extreme summer temperatures are as high as in the Punjab, but the lowest winter reading re- corded is 36, and in one year did not fall below 42. At Nagpur also, the summer readings are nearly as high, 117 in the shade having been reached in four of the last ten years ; but in the coolest years the thermometer has not fallen below 43, and sometimes not to 50. And, even at Madras, 113 has been registered in the hottest months, although in the coolest 5 7 '6 is the lowest temperature reached in the last ten years, and in many years it does not fall to 60. Madras, however, although situated on the coast, has a comparatively dry climate at the hottest time of year. In the damper provinces of India, in part of Bengal, in Assam, and on the west coast of the peninsula, the highest summer temperatures rarely or never reach 100. At Calcutta, indeed, the thermometer sometimes rises to 105 or 106 in AIR TEMPERATURE 15 April and May, while it falls to between 45 and 52 at the Alipore observatory in January and February. But in Cachar a reading exceeding 100 has been but once re- corded in eleven years, and at Sibsagar this has occurred but twice in the same period. At Bombay in the same eleven years it has never reached 97, and the highest recorded reading in thirty-seven years is 100'2; at Cochin, notwithstanding its more southerly latitude, it has never marked 100; and at Point de Galle, since the thermometers have had a proper exposure, it is doubtful whether it has ever risen above 90, while the minimum has never fallen to 68. In comparing extreme temperatures, whether high or low, it must be borne in mind that such comparisons are valid only when the same method of exposing the instru- ments has been rigorously observed, and when also the errors inherent to almost every thermometer have been ascertained and corrected. It is easy to obtain a reading a few degrees higher in the daytime or lower at night by slightly diminishing the protection of the instrument. The object sought, since such is understood to be the meaning of a temperature reading, is to ascertain as nearly as pos- sible the temperature of the ambient air, and to avoid the disturbing effect of radiation. In a site distant from build- ings and not too much shaded by trees, the method of exposure adopted at Indian observatories very well fulfils the required conditions. And to ensure the accuracy of the figures quoted, I have restricted the discussion to such obser- vations as have been recorded with all due precautions, and have rejected such as were registered prior to 1876 in the Punjab and some other parts of India, which, on account of improper exposure, I do not consider entitled to confidence. Hottest and Coolest Months. Annual Variation. The popular division of the year into a cold season, a hot season, and " the rains," holds good in the greater part of India, and fairly represents the more obvious phases of the climate. Such is the case in the North-west Provinces and 1 6 ELEMENTS OF CLIMATE, ETC. Oudh, Central India and eastern Kajputana, the Central Pro- vinces, Chutia JSTagpur, and Bengal, and, substituting the word " cool " for " cold," it may be held to apply to the northern and much of the eastern part of the peninsula, although the duration of each season varies not a little in different parts of this region. Wherever this is the case, the temperature continues rising from January or, in some years, February, only up to the setting in of the rains. The first drenching fall of rain sends the temperature down many degrees ; and although it rises again whenever the rains are interrupted, it never attains the same intensity as in the long interval preceding the rainy season, when the whole land surface lay parched and bare of verdure under a torrid sun. In North- western India, where the rains come to an end early in September, they are followed by a few weeks of warmer weather ; but in the latter part of October the temperature falls rapidly, and in Bengal, if, as is often the case, the rain lasts well into October, a fall of temperature sets in soon after their conclusion and gradually brings on the cool season. Thus, throughout Central India, and indeed everywhere east of the Aravalis and south of the Ganges, also in the north of the Peninsula and its eastern half as far south as the Godavery, and on the coasts of Bengal and Arakan, May is the warmest month in the year. In Sind, the Punjab, the region west of the Aravalis, and also on the plain north of the Ganges, where the rains seldom set in much before the latter part or quite at the end of June, or where they are too light and scanty to mitigate the heat, June is warmer than May; and, at the more remote stations of Quetta and Leh, the temperature goes on rising till July, as is generally the case in the temperate zone. For a different reason July is also the warmest month in Sikkim and Assam. In the Sikkim Himalaya, and in the valleys of Assam and Cachar, frequent showers keep down the tempera- ture in April and May and neutralise the hot season. On the other hand, in the Western Deccan (at Poona and Male- AIR TEMPERATURE 17 gaon), and in the south of the peninsula, except on the east coast, April is the warmest month in the year ; and this is also the case in Ceylon and the Burman peninsula, save in the coast province of Arakan. In Burmah, as in Assam, this is due to the greater abundance of the May showers and to the earlier setting in of the monsoon rains ; but in the southern provinces the changes of temperature are smaller, and in the Carnatic it is not until the October rain- fall has cooled the plains and filled the irrigation tanks that there is any considerable mitigation of the heat. From what has been said above, it will be apparent that from January to May the rise of temperature is greatest in Upper Sind, the Punjab, and the North-west Provinces, and diminishes steadily to the south and east. But the fall from May to July is greatest in the heart of the country, in the Central Provinces, where the dryness and heat of May are as great as in the Punjab, while the rains set in earlier and are more abundant. These variations are shown in the following table for all parts of India, together with the annual mean temperatures, those of the hottest and coolest months, the mean highest and lowest temperatures reached in the year, and the mean annual range. Some further particulars are given in the Tables of Appendix I. [TABLE iS ELEMENTS OF CLIMATE, ETC. MEAN AND EXTREME TEMPERATURES AND RANGES AT FIFTY-ONE STATIONS, CHIEFLY HILL STATIONS AND MILITARY CANTONMENTS. (Deg. Fahr.) Stations. Elevation in feet. MEAN TEMPERA- TUBE OF Mean Highest Reading. It S | Rise January to May. Change May to July. Year. Hottest Month. Coolest Month. Kurrachee . . 49 77 87 65 107 45 62 20 - 1 Quetta . . . 5,501 58 77 40 99 15 84 26 + 11 Jacobabad 186 78 96 57 118 32 86 34 + 3 ^Mooltan . . - 420 76 94 54 114 34 80 35 + 3 D. I. Khan . 573 74 93 52 117 31 86 35 + 4 Leh . . . . 11,503 40 62 18 90 - 4 94 29 + 15 Peshawar . . 1,110 70 89 50 115 29 86 30 + 9 Rawalpindi . 1,652 69 89 49 114 29 85 32 + 6 Murree . . . 6,344 56 71 39 93 24 69 26 + 3 Sialkot. . . 829 73 91 52 117 34 83 33 + 2 v Lahore . . . . 732 75 93 54 117 34 83 34 + 1 Simla . . . 7,012 55 67 41 88 25 63 23 Delhi . . . 718 77 93 59 116 40 76 30 - 2 Meerut . . . 737 76 92 57 112 37 75 32 - 3 Agra . . . 555 79 95 60 116 40 76 34 - 7 Deesa . . . 465 80 92 67 112 40 72 25 - 9 Abu . . . 3,945 68 79 58 96 39 57 21 - 9 Neemuch . . 1,639 75 88 62 111 39 72 26 - 9 Ajmere . . . 1,611 74 89 58 112 34 78 31 - 7 Lucknow . . 369 78 92 61 114 38 76 31 - 6 Allahabad . . 307 78 92 61 116 40 76 31 - 7 Patna . . .. 183 78 89 61 110 42 68 28 - 4 Darjiling . . 7,421 51 61 40 72 25 47 14 + 7 Hazaribagh . 2,007 74 85 61 106 43 63 24 - 7 Berhampore . 66 78 85 65 109 46 63 20 - 2 Calcutta . . 21 78 85 65 102 48 54 20 - 2 Dacca . . . 22 78 83 66 100 48 52 17 Chittagong 87 77 82 67 94 48 46 15 - 1 Sibsagar . . 333 73 84 59 99 42 57 19 + 6 Cuttack . . 80 81 89 70 110 51 59 17 - 6 Saugor . . . 1,769 76 89 63 110 42 68 26 -11 Jubbulpore 1,341 75 90 61 111 36 75 28 -11 Pachmarhi 3,511 69 83 56 100 35 65 25 -12 Nagpur . . 1,025 79 93 67 115 46 69 24 -14 Poona . . 1,849 78 86 72 106 44 62 13 -10 Bombay . . 37 80 85 74 95 61 34 11 - 4 Belgaum . . 2,550 74 81 71 102 51 51 8 - 9 Sholapur . . 1,590 79 89 70 110 47 63 17 -10 Secunderabad . 1,787 78 89 69 109 48 61 19 -12 Bellary . . 1,455 80 89 73 108 54 54 15 - 7 Bangalore . . 2,981 73 80 67 98 51 47 12 - 7 Wellington . 6,200 61 66 55 80 37 43 11 - 3 Madras . . 22 82 88 76 108 60 48 11 - 1 Trichinopoly . 275 82 88 76 106 60 46 12 - 3 Cochin . . . 11 80 84 79 95 67 28 3 - 5 Madura . . 448 82 86 77 105 62 43 9 - 2 Galle . . . 48 80 82 78 89 70 19 4 - 2 ,- Rangoon . . 41 79 84 75 104 58 46 8 - 5 Moulmein 94 78 83 75 99 58 41 7 - 5 Thyet Myo . 134 79 87 68 108 45 63 17 - 5 Tounghoo . . 181 78 85 70 104 51 53 13 - 5 ' BAROMETER 19 THE TEACHINGS OF THE BAROMETER Effect of Elevation. The weather indications formerly inscribed on barometers, such as " set fair," " change," " much rain," and " stormy," have been so generally discarded from modern instruments that it is scarcely necessary to warn the reader that they are quite inapplicable in India. Nowhere in .the world would they be otherwise than misleading in any situation very much above the sea-level. The average reading of a barometer, at any place, depends on its eleva- tion above the sea ; it is the lower the higher the elevation of the place, the fall, roughly estimated, being about one inch for every thousand feet (or one-thousandth of an inch for every foot), provided the elevation does not exceed a few thousands of feet. 1 Until the average value for the place is approximately known, no inference concerning the weather can be drawn from a simple barometer reading. Annual Fluctuation. In India, a further cause of variation must be taken into account, viz. the annual fluc- tuation of the barometer, which is quite independent of weather vicissitudes. At most places on the plains of India the barometer stands highest in December or January, and from this time falls continuously till June or July ; in August it begins to rise again, and continues to do so, amid all its fluctuations, till the end of the year. The whole rise and fall during the year, apart from such changes as accompany changes of weather, is less than a tenth of an inch in Ceylon, about three-tenths in Madras, about half an inch at Calcutta, and not quite six-tenths of an inch in the Punjab. At the hill stations it is less than on the plains. The average monthly readings at a number of stations are given in Appendix I. 1 Thus, at the Alipore observatory, Calcutta, which is only 21 feet above sea-level, the average reading of the barometer is 29*79 inches. At Allahabad, 307 feet above the sea, 29'48 inches ; at Lahore, 732 feet above it, 29'08 inches j and at Simla 7048 feet above it, only 23'24 inches, on the mean of the year. 20 ELEMENTS OF CLIMATE, ETC. Diurnal Fluctuation. There is yet one further oscilla- tion of the barometer which, in India, is remarkably regular, and must be duly taken into account before a change in the barometric reading can be appealed to as indicating a change of weather. Every day, with the utmost regularity, it rises from about 4 in the morning till half-past 9 ; then falls to about 4 or 5 P.M. ; then rises again till 1 P.M. ; and again falls till 3 or 4 A.M. These changes, which amount on an average to about a tenth of an inch, have nothing to do with the weather; but they must not be neglected, since they bear a very appreciable proportion to those that are produced by weather vicissi- tudes. Thus a fall of a tenth of an inch between 10 in the morning and 5 in the afternoon does not betoken any change, merely as regards weather, and so of the other diurnal changes. The course of the diurnal fluctuations at Calcutta in March and July are represented in Fig. 2. M 6 8 10 N 14 16 18 20 22 M *o8 +06 + 04 402 O -02 + 02 O2 -0 4 rtf. z "\ A \ / \ ^ /~ \ A yx* > ^v s \ x^ ^ LX * . V "2_ > ^^ - ^^ _x ^^ ^_ ^ . ^ "^x^ / B \ x^^ \^_ z \ / \ /^ ' w " ^"^ FIG. 2. DIURNAL VARIATION OF BAROMETER AT CALCUTTA. A. March. B. July. Magnitude of Variations with Weather. Lastly, it must be borne in mind by those who have used to consult the barometer as a weather-glass in England or other country in higher latitudes, that the oscillations which, in India, do betoken changes of weather, are much smaller than those to which such experience may have accustomed them; and it is for this reason that the several regular changes above noticed cannot be overlooked or ignored. In BAROMETER 21 this country, a fall of three-tenths of an inch below the average of the time and place occurs only as a prelude to very disturbed weather, and it is only in violent cyclones and in the vicinity of the storm's centre that this amount is very greatly exceeded. In the centre of such a storm it has been known to fall more than two inches below the average of the place and time of year. Temperature Correction. One further remark on the action of the barometer is necessary before any comparison can be made between its readings and those published in the daily weather reports, or with such average values as are also periodically published, and some of which are given in the Appendix to this work. A mercurial barometer is affected by temperature as well as by the pressure of the atmosphere, and acts to some extent like a thermometer, rising with heat and falling with cold. The published readiugs of the barometer are, therefore, not such as are actually recorded, but are reduced to the height which the instrument would have marked if it had been at the tem- perature of melting ice, a condition which, it need hardly be remarked, it rarely indeed has anywhere on the plains of India. This correction is effected by tables computed for the purpose and given in most meteorological handbooks. These remarks apply to mercurial barometers only, not to aneroids, which, if properly compensated by the makers, should not require such correction. But every instrument, whether aneroid or mercurial barometer, generally has some error, which can be ascertained only by comparing its readings with those of a standard instrument, and this should never be neglected. Barometric registers that have been kept for years at the cost of much trouble and self- denial, are of little value to science unless their correction is known, and also unless the temperature of the instrument has been recorded whenever a reading is taken. The baro- meter many years in use at the Calcutta observatory is used as the standard for all Indian stations. Barometric Changes with Weather. With these pre- liminary remarks we may pass on to consider the manner 22 ELEMENTS OF CLIMATE, ETC. in which the readings of the barometer may be interpreted as prognosticating the character of the weather. In addi- tion to the fluctuations just described, which are regularly recurrent and can, therefore, be predicted with confidence, the pressure of the air, which immediately affects the baro- meter and the variations of which are all that it really indicates, oscillates more or less irregularly at all times of year, increasing for two or three days, and then falling in like manner. The rise and fall of the barometer in these movements ordinarily amount to between one and two- tenths of an inch, and are greater in Northern than Southern India. They do not necessarily betoken any important change of weather, though generally there is more cloud about the sky when the pressure is low than when it is higher ; and it is at its low phase, in the hot season, that thunder-storms and dust-storms most frequently occur. In the cold season, in Northern India, a fall is frequently accompanied with an easterly or southerly wind, the wind that precedes and brings the winter rains of Northern India. In the rainy season a fall of greater amount precedes one of those bursts of heavy rain, that throughout this season alternate with intervals of fine weather. When a cyclonic storm is generated far down in the Bay of Bengal, which occurs chiefly in the months of May, October, and November, the barometer on the north of the bay and the surrounding land is often but little affected, and sometimes even rises while the storm is in course of formation ; and it is not until it begins to move northward that any fall sets in. Even then the fall is slow, until perhaps a few hours before the storm reaches the neighbourhood of the place, when it becomes more rapid. Examples of this will be given in the second part of this work. In judging of the probability of a cyclone, other indications must be taken into account, more particularly the direction of the wind and the aspect of the sky. These also will be noticed in another part of this work. Distribution of Pressure. Barometric Depressions. So far I have noticed only such indications as are afforded BAROMETER 23 by the behaviour of the barometer at a single station, and where no information can be obtained of what is taking place in other parts of the country. Before the days of telegraphic weather reports, this, together with the local signs of the weather, was all the practical meteorologist had to rely on for framing his conclusions, and under these cir- cumstances the barometer was little more than an auxiliary guide, far from having the importance which it possesses in these days of telegraphic reports and weather charts. For, while very little of value can be inferred from the simple reading of a barometer, even when compared with the average of all its readings at the same place and time of year and day, a knowledge of the distribution of atmospheric pressure over an extensive country, such as India, affords a firm basis for meteorological reasoning. Primarily, it is some difference in the pressures of different parts of the country and the surrounding seas that determines the courses of the winds, and it is the character of the winds as damp or dry that determines that of the weather. The place where the pressure is least, in other words, where the barometer stands lowest, is usually termed a 'barometric depression or minimum, and the immediate neighbourhood of this depression is the place where rain is most likely, and when it falls most abundant. This last statement is, indeed, not to be taken without some qualification, but the excep- tions are readily recognisable, and will presently be more particularly specified : as a general rule, the maxim holds good. Reduction to Sea -level. It has been stated on a previous page that the reading of the barometer depends on the elevation, falling about one-thousandth of an inch for every foot above sea-level. Differences of readings, due to those of elevation must therefore be allowed for and corrected before the distribution of atmospheric pressure over the surface of a country can be known for the purpose of meteorological reasoning. In the published weather reports this allowance has been made ; all barometric readings of stations on the plains of India are given, not as recorded, 24 ELEMENTS OF CLIMATE, ETC. but reduced to the values they would have at the common level of the sea surface. These pressures are, of course, purely imaginary, and in the case of stations on the hills of India would be probably false and misleading ; but when restricted to such places as are not more than a couple of thousand feet above the sea, the readings, thus reduced, afford a fair idea of the differences of atmospheric pressure that are effective in determining the winds, to which they bear a well-ascertained relation. Barometric Charts. Isobars, To render the facts of pressure distribution evident at a glance, the method now universally practised is to lay them down on a chart. The preparation of such a chart is very simple. The several barometric readings, duly corrected for temperature and reduced to sea-level, are pencilled against the stations to which they severally refer. A series of lines is then drawn between or over these, passing through all those points at which the pressure is the same, the points selected being those where it is the exact tenth, or, in Indian charts, the twentieth of an inch. The interspaces between any two consecutive lines show where the pressure is of intermediate value, and the distance between them is the horizontal dis- tance within which the pressure increases or decreases by one-tenth (or one-twentieth) of a barometric inch. The lines therefore represent differences of pressure in the same way that the contour lines of a map show differences of elevation and the slope of the ground, enabling one to gather correct notions of the form of its surface. They are techni- cally termed isdbaric lines, or simply isobars. A barometric depression is the space enclosed within the isobar of lowest value, and a barometric elevation, or as it is usually termed, a barometric maximum) is that surrounded by the isobar of highest value. Other terms, which will be noticed when we have to speak of the winds, are also used to denote these two important regions. Barometric Gradients. Another term often employed in weather reports and requiring explanation is the baro- metric gradient. This term, which has been adopted from BAROMETER engineers, means the rate at which the atmospheric pressure decreases between any two places. It may be expressed either by the number of miles in which the pressure falls one-tenth of an inch, or by the amount of decrease per hundred miles. On any barometric chart its value, expressed in the former manner, may be found at once by simply measuring on the scale of the chart the distance between two Fia. 3. AVERAGE BAROMETRIC AND WIND CHART FOR JANUARY. isobars differing one-tenth of an inch in value. It is an important datum, because, just as the velocity of a stream of water is the greater, the greater the slope down which it flows, so is the strength of the wind the greater the higher the barometric gradient where it blows. t Wherever, there- fore, the isobars of different values approach each other most nearly, there do the strongest winds prevail. Other relations 26 ELEMENTS OF CLIMATE, ETC. of the winds to the isobars and the distribution of pressure generally will be described in the next section. Persistent and Temporary Barometric Differences. It would be beside the purpose of this work to discuss at any length the circumstances that give rise to these inequali- ties of pressure, which the barometer proves to exist, and FIG. 4. AVERAGE BAROMETRIC AND WIND CHART FOR JULY. which are represented on the weather charts issued from day to day. And indeed, in some of the most important cases, we are not yet able fully to explain them. But it is neces- sary to point out the distinction between such as last for months together and are intimately related to the monsoons, and such as are temporary only and regulate the variable BAROMETER 27 phases of the weather, and more especially those that accom- pany storms. As illustrations of the former class I give two Figures (3 and 4) showing the average distribution of atmospheric pressure in January and July. 1 In the first of these, the seat of the highest pressure is in the north-western corner of India, and it decreases steadily southwards to two regions, one to the south of Ceylon, the other near Sumatra, where are two incomplete isobars, indicating the places of lowest pressure on the chart. In the July chart, the seats of highest and lowest barometer have exchanged places. The former now occupies the equatorial sea, the latter Upper Sind and portions of the adjacent tract, and throughout India and its seas the barometric gradient is reversed. In the interval between January and July this change is brought about entirely by the changes of temperature ; the seat of lowest pressure in both cases being also the seat of the greatest warmth, and that of highest pressure the region where the atmosphere is coolest. Amid all the changes that take place from day to day in the course and position of the isobars, with very rare exceptions, these leading features remain undisturbed ; and, as will be seen in the next section, it is these that determine the general course of the monsoons. Now barometric depressions of this class sometimes form an exception to the rule given above, viz. that rain is most likely, and when it falls, heaviest in the neighbourhood of the depression. Those represented in the January chart conform to the rule, since they are situated over the sea where the supply of vapour is abundant, and the air there- fore damp and capable of yielding rain. But in July, and 1 It is necessary to point out that in these charts the barometric values of the isobars are all somewhat lower than would be assigned to them by the actual readings, corrected only for temperature and elevation as described in the text. A further correction has been applied for gravity, but as this is important only to the physicist, it need not be specially noticed here. The charts show the relative distribution of pressure accurately, and, except in the values of the isobars, the chart differs but little from that which would result from the unconnected data. 28 ELEMENTS OF CLIMATE, ETC. indeed all through the summer monsoon, when the seat of the lowest pressure is the driest land tract in India, the most copious rainfall takes place at a considerable distance from it, viz. in Bengal and Assam, along the face of the Himalaya, and over a broad zone extending across India from the Gulf of Cambay to the Bay of Bengal (see Fig. 6) ; and also, owing to local circumstances, on the west coasts of India and Burmah. Of the irregularities of pressure which are only tem- porary, it is necessary to notice only the depressions, and of these only such as accompany storms. Slight local falls of the barometer occur at all seasons, and have no important significance. Owing to incessant changes of temperature, the atmosphere is never at rest through its whole depth, and every movement is accompanied by some small variation of pressure. Of most of these it is unnecessary to speak. But when, for some days together, the pressure falls continuously over some tract of land or sea, a movement of the air is set up around, which may become strong enough to generate a storm, the constitution and nature of which will be described further on. Such a storm may either exhaust itself in the place where it originated, or may move forward and travel thousands of miles, its centre being marked by the barometric depression which originated it, and which indeed forms one of its leading features. Illustrations of such storms are given in Tigs. 16 to 23 in the latter part of this work. When such a storm approaches any place, as a matter of course the barometer falls continuously until the storm either reaches or passes it ; but in order to know whether any observed fall is due to the approach of such a storm or to some other cause of less importance and significance, it is necessary to be provided with charts of the pressure on successive days, such as are published by the meteorological offices of Simla and Calcutta ; unless, indeed, other weather signs, of which more anon, be sufficiently distinct and charac- teristic to indicate the real character of the disturbance. Summary. These remarks may suffice, for the present, BAROMETER to enable the reader to intelligently interpret the teachings of the barometer. But, before leaving the subject, I will briefly recapitulate the different points that demand atten- tion. First, the readings of the barometer vary with the elevation of the place, and to such an extent that very little of value can be inferred from them until that elevation is known to the nearest foot or two. Secondly, in India, the barometer falls from January to July, and rises during the remainder of the year. This fall and rise are related to the changes of the monsoons. Thirdly, it undergoes a double oscillation daily of about one-tenth of an inch. This has no reference to changes of weather, but is too considerable to be left out of account in using the barometer as a weather- glass. Fourthly, the mercurial barometer is affected by changes of temperature like a thermometer, and in India, where these daily changes sometimes amount to 40, the rise and fall of the barometer owing to this cause alone is very appreciable. Tables are published for correcting it, and in order to render the readings made at different temperatures comparable with each other, it is usual to reduce all to their corresponding values at the standard temperature of the freezing point. Fifthly, every barometer has some error, sometimes a large error, inherent to the instrument. This is to be ascertained by comparing the instrument with some acknowledged standard. And further to render its readings comparable with those of other barometers at different eleva- tions, it is necessary to reduce all to the same level, generally the level of the sea. The oscillations which depend on changes of weather are comparatively small in India, not exceeding a few tenths of an inch, except near the centre of violent cyclones. But oscillations of one or two-tenths of an inch, extending over a few days, follow each other incessantly at all times of year. and do not necessarily betoken any change of importance. As a rule, the fall takes place with a wind from the sea, which is damp, and rises with one from the land, which is dry. The full meaning of any barometric change can only be 30 ELEMENTS OF CLIMATE, ETC. known and understood when it is taken in connection with those in other parts of the country. These are shown by barometric charts, the method of preparing and interpreting which has been described above, and will be further noticed in a subsequent part of the work. The interest of barometric readings centres entirely in their relations to the weather and the physics of the atmo- sphere. As an element of climate, the variations of atmo- spheric pressure are far too small to be sensible to the human system, except indeed at great elevations, where, owing to the decreased pressure, the tenuity of the air is such as seriously to tax the heart to effect the necessary aeration of the blood ; but to persons of good constitution, and with no special feebleness of that organ, this is not felt at the eleva- tion of 8000 feet, which is above that of the hill stations of India, most of them being situated between 6000 and 7500 feet, and some lower. THE WINDS Lightness of the Winds in India. Next to its sunny skies and its notorious and somewhat oppressive heat, perhaps no feature of the Indian climate is more characteristic than the prevailing lightness of the wind. Many a one who has watched the laborious processes by which the Indian ryot raises water to irrigate his patch of sugar-cane or paddy must have asked himself the question, " How is it that the same mechanical skill that has enabled the village carpenter to construct a Persian wheel, or the simple but effective turbine and its gearing, wherewith the hillman drives his primitive mill, has never been turned to economise the work of irrigation, by substituting for the labour of the ryot or his cattle some simple motor driven by the wind ?" The explanation is probably to be found less in his want of mechanical skill than in the insufficiency and uncertainty of the motive power at his command. It is true that on the coasts of India, and especially the west coast, the wind has, WINDS as a rule, sufficient strength to be so utilised, and such is also the case in the Deccan during the summer monsoon ; but in Northern India it seldom has sufficient force, except perhaps in the spring months for some hours of the afternoon, when the hot winds blow, and these are by no means regular, and sometimes fail for many days together. At best, there are but few parts of India where windmills could be usefully employed, and hence probably the fact that they are practically unknown. As illustrating the average rates of the wind's movement in different parts of the country, the stations in the following table have been selected from those for which anemo metric data are given in the Meteorological Eeport for 1885. The height of the anemometer above the ground is given in each case, and those stations only are quoted at which the instrument is comparatively unob- structed. The table shows the number of miles per day registered by the anemometer, the average being in all cases derived from many years' registers. AVERAGE DAILY MOVEMENT OF THE WIND i ii MILES PER DAY. Stations. i" P 5 * 1 Jan. Feb. Mar. April May Tune July Aug. Sept. Oct. Nov. Dec. Peshawar 21 57 78 85 90 109 103 102 94 75 63 51 60 utahore . 28 44 55 72 80 77 83 77 69 48 38 34 35 Jacobabad 44 47 74 91 97 117 113 131 96 75 57 40 36 Kurrachee 22 232 247 283 337 424 451 428 412 351 231 191 223 Deesa . 62 207 215 214 229 315 420 347 292 206 160 184 194 Ajmere . 43 48 63 85 131 174 204 169 129 114 66 37 36 Bareilly 48 68 98 105 111 125 120 100 86 73 47 43 51 Benares 76 68 90 111 120 121 122 113 98 92 60 51 57 Sutna . 26 94 113 132 167 190 229 221 184 153 102 77 85 Patna . 35 49 68 82 106 111 100 85 77 71 45 36 40 Calcutta 54 71 93 132 180 172 147 124 126 110 78 70 73 Silchar . 25 49 67 78 84 83 74 74 72 65 52 50 45 Nagpur . 36 69 87 101 122 156 176 177 138 112 83 65 63 Buldana 24 128 148 178 226 297 278 241 217 206 122 121 124 Bombay 53 243 257 276 273 246 397 459 383 285 232 238 239 Poona . 73 132 157 194 238 338 401 404 342 283 137 137 138 Belgaum 47 272 228 269 320 436 601 686 512 418 240 278 288 Secunderabac 31 90 87 116 120 164 291 318 239 168 82 78 72 Bangalore 32 91 82 84 84 128 203 208 183 162 92 89 92 Madras . 50 150 122 152 191 224 218 202 179 159 123 166 184 Negapatam 27 132 83 82 114 164 180 175 142 126 88 126 163 Thyet Myo Rangoon 52 49 57 104 100 98 124 121 174 137 132 125 191 131 158 136 144 121 113 98 72 77 88 103 113 123 Port Blair 31 174 123 110 126 183 258 286 248 239 160 175 173 32 ELEMENTS OF CLIMATE, ETC. Wind Measurement. At land observatories the wind is measured by means of the anemometer an instrument which consists of four hemispherical cups, mounted on light horizontal cross arms, which, being set in rotation by the wind, communicate the movement through a train of wheel work to an index, which shows on a graduated dial, either the number of revolutions or the distance travelled by the wind. The instrument is set up usually on the highest point of a building, if possible above and free from all obstructing objects, such as trees, higher buildings, etc., around. This condition cannot, however, be always secured, and at some observatories the register of the wind from certain quarters is interfered with to an extent which can only be vaguely estimated. 1 In some places it is mounted on the top of a pole, but as these are of various heights, and every additional foot of elevation reduces the retarding effect of ground surface friction on the wind, under the most favourable circumstances, the anemometric registers of different observatories are only roughly comparable with each other. The direction of the wind, as shown by the wind- vane, is also very much influenced by purely local conditions. In the above table no register is included that has been obtained at a less elevation than 20 feet above the ground, or where obstructions are known to exist in the immediate neighbourhood of the recording instrument. At Calcutta, Nagpur, Bombay, Deesa, Kurrachee, Poona, Belgaum, Madras, and Eangoon, and at some other stations not included in the above list, both the movement and direction of the wind are recorded continuously on a sheet or strip of paper set in movement by clockwork, and at these stations all changes of rate and direction can be determined from hour to hour or any shorter interval, but at the majority of the observa- tories the anemometer dial is read off once or twice only in the 24 hours, and the direction at the time of observation noted by the wind- vane. 1 A list of the elevations of all anemometers at the Indian observatories, with notes of their position and that of obstructions, if any, was given in the Report on the Meteorology of India in 1885, p. 113. WINDS 33 Beaufort's Numbers. At sea, it has long been the practice to estimate the force of the wind instead of its rate of movement, and to express it by the series of numbers 1 to 12, according to a scale proposed by Admiral Beaufort and given below. These are known as Beaufort's numbers. By comparing the forces thus estimated at Holyhead pier lighthouse with the registers of the autographic anemometer on the dome of the lighthouse, an approximate evaluation of Beaufort's numbers, in terms of the rates of movement, has been obtained by Mr. E. H. Scott. These are given in the right hand column of the following table : Approximate Velocity, Force. Beaufort Scale. Miles per hour. 0. Calm to 5 1. Light air. . {orjust w suffieienttogivesteer-| _ _ ^ I or that in which a well-con- \ 2. Light breeze . I ditioned man-of-war, with I 1 to 2 knots 11 ,, 15 3. Gentle breeze . I all sail set, and clean full, >3 ,,4 ,, 16 20 4. Moderate breeze j would go in smooth water I 5 ,, 6 ,, 21 ,, 25 \ from j or that to which she could Eoyals, etc. . . . 26 ,, 30 Single-reefed topsails and ) gi og top-gallant sails ) " Double-reefed topsails, ) Q 7 AA ^OJ.1^ 1/UU.i.ll. IlLOU t ! j ( O/ . 44 o,*_,s, Wl.o 1 Jib, etc. _ f etc. ' ( 45 " 52 Close-reefed topsails and j K0 ,, A > 53 ,, 60 5. Fresh breeze 6. Strong breeze 7. Moderate gale 8. Fresh gale *" ^rfin 'chase, 9. Strong gale J I courses or that with which she could ^ 10. Whole gale . scarce1 ^ be ^[ close -reefed I 61 69 main topsail and reefed j foresail J 11. Storm . / or , tha * which would reduce | 7Q g() 1 her to storm staysails . / 10 TT.- f or that which no canvas could ) 80 and ' | withstand } - ' ' upwards. These numbers will be sometimes referred to in the section on " Storms in Indian Seas " in the second part of this- work. Wind Pressure. A third mode of expressing wind force is by the pressure it exerts per square foot of surface ; and this is the form in which it is of most use to architects and engineers. The pressure is sometimes computed from the rate of movement, but the result so obtained is of little D 34 ELEMENTS OF CLIMATE, ETC. practical value, since it gives only the mean and not the maximum pressure, and stormy winds always blow in gusts which may vary from 5 or 10 to 50 Ibs. in the course of a few seconds. It is better shown by a pressure gauge, of which several forms have been devised. That which registers the pressure autographically at the Calcutta ob- servatory, known as Osier's anemometer, exposes the surface of an iron plate one foot square, which is maintained facing the wind and transmits its pressure to springs of known resistance. The pressure thus indicated is, however, not the same that would be exerted by the same wind on each square foot of an extended surface, such as the face of a building, and differs with the form of the plate as well as with its dimensions. In the opinion of Professor Stokes and others who have studied the subject, such a plate in- dicates a higher pressure than would be exerted by the same wind on each square foot of a masonry structure. But it may represent more nearly that on a lattice girder, which, like the pressure plate, allows the air to pass and form eddies on the rear face. Diurnal Variation of Wind. Everywhere in India, as far as observation enables us to verify the fact, except per- haps in certain positions on the mountains, the movement of the wind is more rapid in the daytime than at night, and most rapid at the hottest time of day. Indeed it varies very nearly with the temperature of the air. The same thing has been observed in other countries, but, like most of the regular periodical changes of the atmosphere, it is so marked in India, that it must be recognised by any one whose attention is once directed to it. The air is most still just before sunrise, indeed at certain times of year absolutely calm ; and in and about large cities, where, in this stagnant condition, it becomes charged with mephitic gases and vapours, it is extremely questionable whether the Indian habit of taking exercise before sunrise is not rather prejudicial than otherwise to health. But the rising sun at once stimulates its movement, which increases rapidly in the earlier hours of the forenoon, and about 2 P.M., as a WINDS 35 general rule, the wind blows with its greatest force, after- wards declining at almost exactly the same rate as the temperature falls. The hot winds of Northern India in the spring months are essentially diurnal winds, springing up in the forenoon, blowing with some strength during the heat of the day, and sometimes not falling to a lull till long after sunset. They are land winds, and derive their heat from the ground, which is much hotter than the air, while their dryness is such that paper becomes brittle, and furni- ture and woodwork of all kinds shrinks and tends to fall to pieces by the loss of the moisture which, in more ordinary conditions of the air, all such organic substances hold ab- sorbed in an insensible form. On the coasts, as for instance at Bombay, the sea winds * follow a very similar course of variation during the drier months of the year, but, at the height of the monsoon, any diurnal variation in the force of the wind is less appreciable and hardly to be recognised by the casual observer. In some parts of the country, at certain seasons, over and above this variation in the rate of its movement, there is a well-marked change of the wind's direction in the course of the day. On the coasts, as is well known, in the long lull between the monsoons, and generally in calm states of the atmosphere, a land wind at night and in the early morning alternates with the stronger sea breeze of the warmer hours. At Calcutta, sixty-eight miles from the sea, the variation is of a different character. In the spring months, during the hottest hours of the afternoon, the wind frequently blows from the west, being in fact the land wind of the Upper Provinces, and is characterised by the heat and dryness of that wind in a somewhat mitigated degree ; but, in the evening, it yields place to a cool southerly wind, in fact, a late sea breeze, blowing up from the wide estuary of the Hooghly. This latter does not reach the more inland stations, such as Kishnaghur and Berhampore, and such stations, therefore, do not enjoy the refreshing change which ushers in the Calcutta dinner hour, and affords some hours of comparative coolness and rest. Easterly winds are 36 ELEMENTS OF CLIMATE, ETC. most frequent in the early hours of the morning, chiefly in October and November, and are too often tainted with the unsavoury exhalations from the Calcutta sewage and refuse, which are discharged in that direction on the borders of the Salt Lake. In the valleys of the Himalaya, as in all mountainous countries, there is a tendency to an up-valley wind in the daytime and a down-valley wind at night ; but, as a rule, this tendency is shown, not so much by a distinct reversal of the directions in the course of the day, as by the ex- clusive prevalence of one or the other wind at its appro- priate hours, alternating with a comparative calm. The up-valley wind is felt most strongly on the passes, where, as General Strachey has described it, it sometimes blows in the afternoon with such force as to present a serious ob- stacle to travellers. On the other hand, the down-valley wind is strongest where the great valleys debouch on the plains. At Hurdwar, at the mouth of the Ganges valley, where it is known as the Dadu, it sometimes blows with such force, in the early morning hours of the cold weather, as to blow down the tents of persons camping in the vicinity. In point of fact, there is perhaps no station in any part of India which has not a characteristic diurnal variation in the direction of the wind. Such has been shown to exist by Mr. Hill at Allahabad, by Mr. F. Chambers at Kurrachee, and by Mr. C. Chambers at Bombay; and it is equally evident in the records of the anemographs at Lucknow and Nagpur, published monthly by the Meteorological Office. But, in most cases, it is not manifested by a distinct change of the wind, but only in a more or less slight modification of the direction prevalent at the season, and has a scientific rather than a general interest. Annual Variations. The Monsoons. The reversal of the prevailing winds, which takes place twice regularly in the course of the year, has long been familiar under the name of the monsoons to all who have any knowledge of the east. It holds good in all parts of the country, but the WINDS 37 directions are by no means everywhere those by which the monsoons are commonly known. The terms south-west and north-east are, in fact, truly descriptive, only in the Bay of Bengal and the western half of the Arabian Sea. And for this reason, in dealing with the land, it is better to substitute for them the equally distinctive but less mislead- ing designations, summer and winter monsoons, as will be clone throughout this work. Figs. 3 and 4 (pp. 25 and 26) exhibit the average wind directions, both on India and the seas around, in the months of January and July, when the two monsoons are respect- ively at their height. The relative steadiness of the wind is indicated by the lengths of the arrows, some of the longest representing 90 per cent of the observations and upwards. The winds are lightest in the cold season of Northern India, that of the winter monsoon. In the Punjab, even in the daytime, on nearly half the days at this season, there is not wind enough to stir the cups of the anemometer; and all down the Gangetic plain, while a calm state of the atmosphere is very common, light zephyrs are the chief movement of the air, until the sun of March and April has sufficiently heated the ground to set up active convection and give rise to the hot west winds. On the higher ground south of the Ganges and Jumna, the movement is somewhat greater and more northerly, and the same is the case in Lower Bengal ; but, as the table on a preceding page shows, up to the end of February, at nearly all stations it is less than 100 miles a day. In the peninsula at this season the wind is easterly, between north-east and east in November and December, and more easterly and southerly in January and February, while, on the west coast, it is light from north and north-west. At times indeed, and in some seasons more than in others, the light northerly and westerly winds of Northern and Central India are suspended for many days together, and replaced by light winds from the opposite quarters. These are warmer, damp and oppressive, and the vapour 38 ELEMENTS OF CLIMATE, ETC. brought up by them is condensed, first clouding the Hima- laya and then the plains, and finally turning to rain, the cold weather rainfall of Northern India. In the north of the Bay of Bengal the north-east monsoon is very light, and often an absolute calm prevails for days together. In the heart of the bay, and from the mouths of the Godavery southwards, it blows more steadily, and is often strong to the south of Madras, producing a rough sea and, up to January, stormy weather off Ceylon and in the south of the bay. On the Arabian Sea it is very light near the Indian coast, but becomes stronger and steadier farther west, blowing from between north and north-east. In the north-west of the bay, on the coasts of Bengal and Orissa, the winter monsoon may be said to have ceased by the end of January, but light northerly winds continue some weeks longer down the Arakan coast, and in the middle and south of the bay it lasts well into March, becoming less steady and by degrees shifting towards the east. In the Arabian Sea, on the other hand, it backs through north- west to west, as the barometer falls over the land in conse- quence of the increasing heat ; but the weather remains fine, and the Arabian Sea is seldom disturbed by stormy weather, such as, after March, sometimes occurs in the Bay of Bengal in the interval between the winter and summer monsoons. On the coast of Bengal southerly winds begin to set in as early as the end of January, at first merely as afternoon sea breezes. Little by little they penetrate farther inland, blowing night and day, and also extend their range down the coast and farther out to sea. In the east and south of the peninsula the east and north-east winds of the cool season also veer to south-east ; but in Northern India, on the plains of the Ganges and the higher plains and hilly country to the south of that river, the westerly winds strengthen and become hot and drier. In the meeting of these and the sea winds are generated the thunder-storms and dust-storms characteristic of the spring season. Up the Gangetic plain along the foot of the Himalaya, and over the crests of the lower hills, easterly winds advance WINDS 39 earlier than on the plains to the south, feeding the diurnal storms, often accompanied with hail, which are common in the spring, and adding copiously to the snows of the higher ranges ; and, recurving from the hills, they are often felt as north- east winds on the plains of the Ganges, as at Purneah, Patna, Benares, Allahabad, and Lucknow. Meanwhile, in Western India, north-west winds, sometimes changing to west, blow very steadily in Eajputana, Guzerat, Khandesh, Berar, and the north-west of the peninsula ; and these being very dry, rain seldom falls in this part of India during the spring ; nor indeed, except in the rare case of a winter disturbance in some part of this region, or of a cyclone crossing from the Bay of Bengal, during the whole interval between the cessation of one summer monsoon and the beginning of the next. But in the south and east of the peninsula, where the winds of the earlier months are east and south-east, little thunder-storms occur, as in Bengal. Fig. 15 shows the average wind directions in the month of May. In May, while light west and north-west winds still sweep the Arabian Sea, in the Bay of Bengal south-west winds have sprung up, but somewhat unsteadily and often interrupted by calms ; although this condition is less com- mon than in March and April. The summer monsoon, accompanied with heavy rainfall, now sets in on the west coast of Ceylon and Travancore, and, in the eastern peninsula, on the coast of Tenasserim, and in the course of two or three weeks has taken possession of the bay and the Arabian Sea up to their northern coasts. It invades India from these seas in two main currents. That from the Arabian Sea blows on the west coast from directions between west and south-west, sweeps over the western Ghats, on which it dis- charges a very heavy rainfall, and continuing its course steadily across the Deccan, holds undisputed possession of the peninsula up to the end of August. In the interior its direction is most commonly west, especially in the Deccan, where it is strongest and steadiest ; but in the south it often changes to south-west, and, at slack intervals in all parts of the peninsula, sometimes to north-west. 40 ELEMENTS OF CLIMATE, ETC. The other branch sweeps from south-west up the greater part of the bay; a part passing directly over Burmah and Assam towards the eastern Himalaya. The remainder curves to south in Bengal, and farther recurving along the face of the Himalaya, becomes an easterly wind up the Gangetic plain. To some extent it penetrates the north- eastern part of the peninsula, but this is exceptional until near the end of the season, and it sweeps past the Carnatic, withholding from that province more than a scanty share of the rainfall which it carries abundantly to Bengal, Assam, and, in a minor degree, other parts of Northern India. In the Punjab, during the summer monsoon, the winds are somewhat variable. Easterly winds predominate, and these are partly drawn from the current blowing up the Gangetic plain. But the westerly branch of the monsoon from the coasts of Sind, Cutch, and Guzerat blows across Eajputana as a south-west wind, and in part coalesces with the former in the northern Punjab. And, at intervals, the easterly wind yields place to a dry wind from the opposite quarter. This takes place during breaks in the rains ; and, in seasons of drought, west and north-west winds hold possession of North-western India for weeks together, pene- trating eastwards across Central India and down the Gangetic plain, even to Bengal, and southwards to the Deccan and Hyderabad. Such was the case in the dis- astrous years of 1876 and 1877, and during the prolonged suspension of the rainfall in 1880 and 1883. In Lower Sind the summer winds are chiefly south- west ; in Upper Sind south-east or south ; but the atmo- sphere is often calm, and rarely indeed does rain visit this arid tract. On the Baluchistan tableland, which rises to 5000 feet and upwards immediately to the west of Sind, the prevailing wind throughout the season is from north-west. The same wind would doubtless prevail in Upper Sind, were it not for the mountain barrier and the fact that winds do not readily descend from a higher to a lower level. But there is little doubt that, descending in a very flat air cata- ract and brought down by convective intermingling with the WINDS 41 lower atmospheric layer, it is this current that feeds the dry west winds of North-western India. We have seen that while a west wind holds almost un- disputed sway in the peninsula up to the Satpura range, an easterly wind is only less dominant along the northern border of the Gangetic plain and on the Himalaya. The tract between these, which comprises the greater part of the loop of the isobar 2 9 - 5 0, on the July chart on Fig. 4, p. 2 6, is debatable ground in which sometimes one, sometimes the other, predo- minates; more often the stronger westerly current. The distribution of the rainfall in this region will be described later on ; but it may be here mentioned that, throughout the monsoon, the cyclonic storms which are generated in frequent succession in the north of the Bay of Bengal (some- times also over the Gangetic delta, or in Chutia Nagpur), as a rule, travel along this debatable zone. Such, for in- stance, was the case with the storms illustrated on Figs. 16 to 23. Further details on this subject may be reserved until we come to describe more fully the natural history of these storms. On reference to the table on page 31, comparing the rates of movement in the summer with those of the winter months, it will be apparent that the summer monsoon is a much stronger current than its winter correlative; and in India this fact is recognised in popular language, since it is often spoken of distinctively as "the monsoon," the claim of the winter monsoon to the same designation being, for the moment, tacitly ignored. On land, in fact, the winter monsoon blows so short a time, and is so unsteady and fre- quently interrupted, that it is far from impressing the lands- man in the same vivid manner as the seaman, who has perhaps to beat up the bay against a steady head wind ; and his attention is rather turned to that feature of the season which is far more striking and important than the direction of its gentle wind. In his consciousness of awakened energy, the expatriated European feels that it is also the cold season, refreshing and invigorating, and affording him a climate, than which, as experienced in Northern India, Italy itself 42 ELEMENTS OF CLIMATE, ETC. can offer nothing more delightful. The thinly clad native, inured to heat, and living in a draughty hut, with perhaps a single meal a day of less stimulating food, is less enrap- tured with the delights of the cold weather, but he is not less conscious than his European brother of this its most characteristic feature, which he feels in the early morning in benumbed limbs and torpid faculties, and which he endea- vours to meet by swathing his head and mouth in a fold of his body cloth, and cowering over the embers of his little fire, till the warmth of the ascending sun restores him for some hours to his state of normal activity. But in spite of this tacit recognition of its superior force and steadiness, it would be a mistake to suppose that the summer, any more than the winter monsoon is a per- sistent and unvarying movement of the atmosphere. In Northern India, at least, the wind changes frequently both in direction and strength; and even a calm state of the atmosphere, though far less frequent than in the winter months, is by no means so rare as to be regarded as phe- nomenal, even in Bengal and the eastern part of the Gangetic plain, while farther to the north-west, and more particularly on the hills and in their neighbourhood, the record of calms constitutes a very appreciable and, at some stations, a large proportion of the wind registers, throughout the summer monsoon. It has already been mentioned that, in some seasons, it is excluded, for weeks together, from a great part of its ordinary domain, which is then invaded by dry winds from the opposite quarter; and these prolonged periods of its suspension are but a magnified form of those interruptions, intervals of sunny weather known as " breaks," which alternate with bursts of rain as an ordinary feature of the season. In fact, the goal of the monsoon, the place of low barometer to which its course is directed, is constantly changing. Generally it lies beyond the limit of the tropic on the western side of India, or not far within it on the eastern side. The seat of lowest pressure in Sind is pretty stable, but very little of the sea wind ever reaches it ; it is rather to the zone indicated by the loop of the isobar 2 9 '50, WINDS 43 on the July chart Fig. 4, or still farther to the south-east, that the main volume of both branches of the monsoon is directed, and within this tract the seat of lowest pressure is constantly shifting. At times, indeed, both branches of the monsoon relax their strength, and fair weather with a moderate wind pre- vails for days or even weeks over a great part of the adjacent seas. On the Arabian Sea this occurs during those invasions of the continent by north-west winds above spoken of ; on such occasions the mail steamers cross from Bombay to Aden with fine weather and light winds, similar to those usually experienced before the monsoon sets in. Towards the close of the season, in September, the rain- bearing wind ceases to penetrate to North-western India, and is directed rather towards the north-east of the peninsula and the adjacent portion of the Bay of Bengal ; the winds of Bengal become more easterly, and those of the northern provinces of the peninsula more northerly. By degrees, the seat of low barometer moves farther south, till, in the latter part of October and November, it lies over the bay, opposite the coast of Madras. This is the season of the Carnatic rains. In November, the winds are very variable in the south of the bay, between Sumatra and Ceylon, shift- ing to all quarters from south-west round by north and east to south-east, and often calm. In December calms are very common in the south-east of the bay, off Sumatra, and at no time does the north-east monsoon blow so steadily in lati- tudes south of Ceylon as in the Bay of Bengal at this season. Relations of Winds to Regions of High and Low Pres- sure. Forthose whose only object,in consulting these pages,is to gain some general idea of the character of the winds at the different seasons of the year, the above description may suffice. But for those who desire to go farther, to understand the meaning of weather reports and charts, and something of the principles on which weather prediction is based, and for sea- men, to whom it is a matter of paramount importance to gather as much knowledge as possible of the weather round about them, and especially of their position in a storm, from 44 ELEMENTS OF CLIMATE, ETC. the indications afforded them by their barometer and the wind direction, it will be necessary to go somewhat farther into the subject, and to show at greater length than has yet been done, how the winds are related to regions of high and low barometer. The preceding description of the two monsoons, taken in connection with the distribution of pressure and the pre- valent wind directions exhibited on the barometric charts on Figs. 3, 4, and 15, has already afforded a striking illustration of the general fact that winds blow from a place where the barometer is high to that at which it is relatively low. But it has also shown that the course they follow is not direct, but, on the contrary, very circuitous. And since this course is no local peculiarity, but that which winds always FIG. 5. CYCLONE AND ANTICYCLONE [Diagram]. follow under the like circumstances in the northern hemi- sphere, it will be desirable to describe it more particularly in relation to the seats of high and low barometer generally. Except that in a cyclone the action is more concentrated and localised than in the case of the monsoons, the descrip- tion will equally apply to both. To make the facts as clear as possible I will, in the first place, exhibit them in the form of a diagram, which may afterwards be compared with the results of actual observa- tion, represented on the storm charts, Figs. 16 to 23, or with the general description of the monsoons, in connection with the two charts of average pressure and wind distribu- tion on Figs. 3 and 4. Let the right hand circle in Fig. 5 be supposed to represent a region where the barometer is highest, and the left hand circle that where it is lowest, the intermediate lines indicating a number of isobars of WINDS 45 intermediate gradations of pressure. Then the winds that blow outwards from the region of high barometer, if in the northern hemisphere, instead of blowing directly outwards like the spokes of a wheel, all take an increasingly oblique course to the right of the radial direction, as shown by the arrows, and those that blow in towards the region of low barometer all blow obliquely to the right of that course which leads directly to the centre. The result is that the former describe a series of spirals curving round the seat of high barometer in the same direction as the movement of the clock hands ; the latter, a series of spirals circulating round the seat of low barometer against the direction of the clock. The latter represents the movement of the winds in a cyclone. They do not blow, as was formerly supposed, in a circle, but as is represented on the diagram. For proof of this the reader has only to refer to the charts on Figs. 16 to 23, where the arrows represent the winds observed and the small circular isobar the eye of the storm. When, as in the case of the monsoons, neither the seat of highest barometer nor that of lowest barometer is a circle or any other regular figure, the course of the winds is not in such regular spirals as those shown in the diagram. It is determined by the directions of the isobars, and, unless diverted by some local condition, is always obliquely to the right of a perpendicular to the isobar towards which it is directed. The amount of obliquity depends on a number of circumstances, the enumeration of which would be foreign to the purpose of this work. As a general rule, the stronger the wind, the more acute is the angle at which it meets the isobar, and the nearer do they approach parallelism. Near the centre of a cyclonic storm, in the North Atlantic, the spirals become so flat that they approximate to the circles which formerly they were supposed to be, but they are never truly circular. In the Bay of Bengal this is less marked, and I have not been able to detect more than a very slight decrease of the radial component in the published storm charts. Around a region of high barometer the pressure de- 46 ELEMENTS OF CLIMATE, ETC. creases slowly, and, when represented on a chart, the isobars are far apart. In other words, the barometric gradient is low, and, as has been stated in the previous section, under these circumstances the winds are light, and radiate more directly outwards than where the pressure decreases rapidly. This is why, when the barometer stands high, indicating that the seat of maximum barometer is in the neighbour- hood, the winds are generally light, as is the case in North- western India in the winter months. On the other hand, when it shows a very low atmospheric pressure, it frequently happens that a cyclonic storm is in the neighbourhood, and around such a storm the barometric gradients are high and the winds strong. The system of winds around a region of low barometer, even when less strong than in the storms, originally termed cyclones, are frequently spoken of as a cyclonic system and those around a seat of high barometer as an anti- cyclonic system. These remarks, which are of general application, will suffice for the present. The whole subject of storms, and the indications which are of most importance to the mariner, together with a further subject equally essential to his guidance, viz. the direction in which storms travel, will be treated of more conveniently and at length in the subse- quent section on storms and in the second part of this work. DAMPNESS AND DRYNESS Relative and Absolute Humidity. The dampness and dryness of the air, as these terms are generally understood in ordinary language, have reference, not so much to the quantity of moisture which it holds suspended, as to the readiness with which it parts with this moisture to objects capable of absorbing it, or, on the other hand, deprives them of that which they hold absorbed. Now this depends quite as much on the temperature as on the quantity of vapour present in the air. The atmosphere of Great DAMPNESS AND DRYNESS 47 Britain in the winter, when the thermometer stands, let us say, at 40, is usually very damp ; but if this same air were at the May temperature of Calcutta it would be an exceedingly dry atmosphere. Dampness and dryness, in this sense, are the conditions most important to us, both in respect of our own bodies, and also as affecting vegetation of all kinds ; and only, therefore, in this sense, will it be necessary here to discuss the subject. In technical language, it is spoken of as the relative humidity of the air, in con- tradistinction to the absolute humidity, which has reference only to the amount of water vapour present, independently of the temperature. It is convenient to have some way of expressing numeri- cally the degree of dampness (of which dryness is of course the correlative), and the plan universally adopted is to regard completely damp air, usually termed " saturated " air, as a standard, and to express the degree of dampness as a percentage of the saturating quantity. Thus, when it is stated that the relative humidity of the air is 50, the meaning is that it holds just half the quantity of water vapour that it is capable of taking up at that temperature. Methods of Observation. The humidity of the air is an object of observation at all observatories, and not the least important. As an element of climate it is probably one of the most important. The method adopted is an indirect one, and requires attention to some points of detail, which need not be insisted on here. But it is very simple in practice, and when properly attended to is found to give a very near approximation to the truth. It consists in taking the readings of two thermometers suspended at a little distance from each other, under the same conditions as other thermometers (see page 6), one of which shows the temperature of the air, the other, having its bulb covered with a single fold of wet muslin, the temperature of a freely evaporating wet surface. From these two readings the humidity is found by the use of a special table. Variations of Humidity in India. One characteristic 48 ELEMENTS OF CLIMATE, ETC. feature of the climate of most parts of India is its great variability with respect to humidity. In the rains, during a spell of very wet weather, the air is sometimes saturated, or has a relative humidity of 100. But when a hot wind is blowing, such as is common in April and May in the Upper Provinces, it is frequently less than 10, and has been known to fall to 3 ; as dry a condition as is probably to be met with anywhere in the world. At Agra, in May 1884, on the mean of the whole month, it was only 20, and at Eoorkee but little higher. In Scotland, Mr. Buchan states that 73 is about the lowest humidity that occurs in the driest month, which, there also, is May. Fortunately for us, when the heat is very great, the air is always dry ; fortunately, I say, because it is the rapid evaporation from the skin, evaporation which can take place only in a dry atmosphere, that enables us to withstand the heat. As some of those who have made the passage down the Red Sea in August or September may have experienced, an atmosphere at 90 which, though not saturated, is, at all events, very damp, is insupportably oppressive, but the hot winds of the Upper Provinces, with a temperature between 112 and 118, if not exactly agreeable, are borne without serious inconvenience. Diurnal Variation. The humidity of the air varies during the day almost exactly in the opposite manner to the temperature. The atmosphere is most damp in the early morning before sunrise ; in fact, the fogs that are so common in the early morning in the cold weather testify that the air is then saturated, for fog cannot exist in air that is not in that condition ; and the dissipation of the fog soon after sunrise affords evidence of the increasing dry- ness. 1 The driest time of day is the same as that of the highest temperature, and the humidity increases in the later hours of the afternoon and evening, as the temperature falls. 1 A remarkable exception to this ordinary sequence occurs in Assam, which I once had the opportunity of verifying personally. Its probable ex- planation will be given farther on (p. 55). DRYNESS AND DAMPNESS 49 Dryness and Dampness dependent on the Winds. The dryness of the warmer hours of the day is not due solely or, to speak more correctly, not directly to the wanning of the air. It is partly an effect of the movements set up in the lower atmosphere by the greater heating of that layer which rests immediately on the ground, in consequence of which the whole atmosphere is stirred up eventually to a height of some thousands of feet in dry weather; and thus, while the damper and more heated air ascends, drier and originally cooler air descends from a higher level to replace it. This, action, which takes place in all countries more or less in dry warm weather, is developed in India in a very high degree, and most so in the dry atmosphere of the Upper Provinces. This, in fact, is the real explanation of the extreme dryness of the hot winds above referred to. In substance, they are part of an air current coming originally from the western mountains and blowing far above the earth's surface, but brought down to the lower level by the action just described. Originally it is indeed much cooler than the air which it displaces, but it is speedily heated, partly as a necessary consequence of its descent, 1 and partly by contact with the ground glowing in the sunshine. It is not improbable that the well-known healthiness of the hot season is in some measure due to this stirring up of the atmosphere, the removal of all mephitic exhalations, and the invasion of the lower atmosphere by the purer air drawn from the high level of the mountain zone. The part played by land winds in India is only be- ginning to be fully recognised. As has been set forth in the preceding section, they occur at all seasons of the year, while in none are they exclusively prevalent. They are characterised partly by their direction, which, in Northern India, is always westerly and, to some extent, northerly, but chiefly by their dryness, of which some instances have already been given. When they penetrate to Southern India they are 1 Dry air, descending through the free atmosphere, is heated 1 for every 183 feet of its descent, in virtue of the compression which it undergoes as it be- comes subject to the increasing weight of the superincumbent atmosphere. E ELEMENTS OF CLIMATE, ETC. more northerly, but retain somewhat of their westing, except in the cold season, when they are between east and north-east. Winds that come originally from the sea, however great the expanse of land they may have traversed, unless, indeed, like the westerly summer monsoon of Southern India, they have been driven to a great height over a mountain range, are relatively damp winds. As an almost invariable rule in India, except temporarily and locally, they are more or less southerly ; but while in Western India, in the greater part of the peninsula, and in Assam, Arakan, and Burmah, they combine this direction with more or less westing, in the Gangetic plain and the neighbourhood of the eastern coast they are from south-east or even more easterly. Geographical Distribution of Humidity. Different parts of India are characteristically dry or damp, accordingly as one or the other of these winds is most prevalent. North- western India, including all the northern portion of the Bombay Presidency, is more particularly the domain of the former; North-eastern India, Assam, and the western part of the Burman peninsula, together with the western margin of the Indian peninsula, that of the latter. The following table shows the mean humidity of the several provinces of India. It must, however, be remarked that different parts of one and the same province sometimes afford a consider- able range of variation : MEAN ANNUAL HUMIDITY OF THE SEVERAL PROVINCES OF INDIA Provinces. Rel. Hum. Provinces. Rel. Hum. Ladak .... 49 Orissa , . . 76 N.W. Himalaya (lower ranges) 62 Central Provinces (southern) 59 Nepal Valley . . 72 Berar and Khandesh 53 Sikkim Himalaya 84 Konkan and Surat 72 Baluchistan 50 Deccan 55 Sind 1 and Cutch 50 Hyderabad and Bellary 54 Punjab Plains . 55 Northern Circars and Goda^ eiy 71 Raj pu tana 48 Carnatic 67 Ouzerat 1 . 47 Mysore 66 Central India Plateau 50 Malabar an iCoo Kg 79 Nerbudda Valley 55 Ceylon 80 N.W. Provinces and Oudh 59 Arakan > 80 Behar 65 Pegu. 77 Bengal 80 Tenasserim , t 80 Assam 80 Bay Islands 81 Chutia Nagpur . 57 1 Exclusive of the coast. DRYNESS AND DAMPNESS Hence the driest parts of India have an average humidity a little below 50, the dampest about 80. Mountains, but not high plains, have a damper atmosphere than the plains of the same part of the country. Annual Variation. In those parts of the country that are intermediate between the permanently arid and the permanently damp provinces, the variation in the course of the year is as great as between these geographical extremes, and indeed greater. The rainy season which, in most parts of India, lasts from June to September, is also the dampest in nearly all parts of the country. But this is not the case on the north-western frontier, nor in the Himalaya north of the snowy range. At Quetta, Peshawar, Leh, and even in Lahoul, the winter is the dampest time of year, as is also the case in Western Europe, and indeed as a rule in extra- tropical countries. On the plains of Northern India August is the dampest month, the mean humidity being a little below 90 on the plains of Bengal and Assam, and as much as 9 5 at Darjiling, which is nearly the dampest station at which an observatory exists. It is above 80 everywhere to the east of Allahabad, but not farther westward, except on the hills. On the plains of the Punjab, south of Lahore and Umballa, and throughout Western Bajputana and Sind, the mean humidity of August remains below 70. At Quetta it is only 43, and at Leh, 48. In the north of the peninsula, July and August are about equally damp, their average humidity being about 85 in the provinces east of Nagpur, and 87 at Bombay and in the Konkan generally. But in the Deccan, to the east of the Ghats, it is everywhere below 80, and in the drier parts below 70. Farther south, in Malabar and Travancore, it is about the same as in the Konkan or a little higher, between 87 and 89 ; and at Mercara in Coorg, on the crest of the Ghats, as high as 97 in July and 95 in August. This station, in the rains, is damper therefore than even Darjiling. On the high plateau of the peninsula it is 77 at Bangalore, and only 62 or 63 at Bellary. Throughout the interior and eastern half of the peninsula south of Orissa, September 52 ELEMENTS OF CLIMATE, ETC. is damper than either of the preceding months ; and as we proceed farther south, the dampest season falls later and later, being in October and November at Masulipatam, in November at Madras, and in November and December from Negapatam southwards as far as Hambantotte in the south coast of Ceylon; thus following the rainfall. In Burmah it is generally in July and August, as in the north of the Indian peninsula. The driest month in the year is November or December on the coasts of Bengal and Orissa, February or March in the interior of Bengal and Assam. Everywhere to the west of Bengal April is the driest month ; except in the Punjab and Upper Sind, where it is deferred till May, and at Quetta, Peshawar, and Leh till June. With local exceptions, the average humidity of April is below 30, from Agra south- wards, all over Rajputana and Central India, the Central Provinces as far south as Nagpur, in Berar, Khandesh, the Deccan, and Guzerat, except on the coast; and at some stations in the heart of this tract, such as Akola, Sutna, and Indore, below 25. In Upper Sind and Western Rajputana, and in the Punjab and North-west Provinces, except in the neighbourhood of the hills, also in Chutia Nagpur and the country south of the Lower Ganges, it is between 30 and 40, but eastwards it increases rapidly, and in the drier parts of Bengal averages little below 60, though on certain days it may fall to 3 or even lower. At the Alipore obser- vatory, in a suburb of Calcutta, the average of February and March is 69 ; but in the more closely-built suburb of Chowringhee only 66, showing that small differences of posi- tion may make an appreciable difference in the dryness of the air. In the valleys of Assam and Cachar it is 70 or higher; at Sibsagar, in Upper Assam, as high as 79. On the west coast of the peninsula the driest time of year is between December and February, when the humidity falls to 7 or a little lower ; at Rutnagiri appar- ently to 60 on the average of the driest month, and at Mangalore to 62. At these two stations the observatory is situated on high ground, which is not the case at the other HAZE, FOG, AND CLOUD 53 stations on this coast. They afford another instance of the variation of the dampness of air due to small changes of site. In the Deccan and Hyderabad March and April are the driest months, and this is also very generally the case in the south-east of the peninsula, except on and near the coast, where the driest period is as late as June and even July ; but throughout the Carnatic the air is comparatively dry in all these months, the humidity varying between 61 and 64 on the coast to a little above 50 in the interior, in the driest months. At Trincomalee, on the east coast of Ceylon, July is the driest month in the year. In Burmah it varies between January at Mergui, in the south of Tenasserim, and March or April in the north of Pegu, where, at Thy at Myo, the humidity averages only 56. Even at Eangoon it falls to 62 in February. Observatories have only lately been established in Upper Burmah, and the only exact information that we have of its climate is furnished by a register kept at Mandalay for the first nine months of 1879 and the last two months of 1886. In the former year, March was the driest month, when the mean humidity was as low as 43, and that at 4 P.M. only 22. September was the dampest, when it averaged 87. In this respect, therefore, the climate of Mandalay would seem to be comparable with that of the drier districts of northern Bengal. It is certainly drier than any part of Pegu. From this brief review it results that different parts of India exhibit as great diversity in the seasons of dampness and dryness as they do in their degrees, and it is driest on the east coast of Ceylon and the Carnatic when it is dampest in the north of the peninsula and on the west coast. HAZE, FOG, AND- CLOUD Haziness of the Sky. Dust Haze. However sunny the sky of India, it is hardly to be described as an azure sky. Of cloud indeed, save sometimes in the rainy season, and at such a station as Darjiling, the amount is not such 54 ELEMENTS OF CLIMATE, ETC. as to cause depression of spirits ; but he who wends his way to India in the hope of living under a deep blue heaven, like to that of Italy or Greece, will surely reap disappoint- ment. The artist who seeks to depict a cloudless Indian sky will deal sparingly with his cobalt and ultramarine if he would work truthfully, and not in the spirit of the en- graver in a well-known story who introduced palm trees in a sketch of Punjab scenery. Except now and then, in a fine interval of the rains, or at the hill stations of the Himalaya in September and October, the pallor of the celestial vault is its ordinary and persistent characteristic ; whether it be due to thin diffused cloud matter at a high level, or, as seems equally probable, to the haze of im- palpable dust, which is carried up by the daily commotion of the heated air, and darkens and sometimes even extin- guishes the setting sun before it has dipped to the horizon. So dense is this haze in the hot season, and to such heights does it rise, that even at Simla, the opposite mountain ridge, only four or five miles distant, is ofttimes invisible through the murky atmosphere. That the pallor of the sky is, however, partly due to very attenuated cloud, seems probable from some observations of Mr. Hennessey at Mussoorie. In the course of a long series of actinometric measurements of the sun's heat, it was necessary to watch closely the appearance of the aerial vault, and he frequently descried a thin gauzy veil, which he designated as " steam- cloud," creeping over the otherwise cloudless expanse, reduc- ing the depth of the sky tint, and appreciably also the action of the solar rays on the instrument. On the plains the haze is almost constantly present, and it is the exception rather than the rule, in the dry season, that the slopes of the Himalaya are visible from a distance of twenty or thirty miles. As far as my own experience enables me to form an opinion, the atmosphere of Southern India is less hazy than that of Northern India, but even there a very clear atmosphere is exceptional. Fog. Fog is restricted to the damper provinces and the cold season, and in them as a rule to the early morning. HAZE, FOG, AND CLOUD 55 It is sometimes thick over the rivers, and also in the ex- treme north of the bay off the Sunderbans, impeding navi- gation ; but in Bengal it is dissipated an hour or two after sunrise. In Assam, however, it often lasts till mid-day, and sometimes presents a peculiarity of behaviour to which my attention was first drawn by Sir Charles Elliott and Mr. C. J. Lyell. This is that the atmosphere remains perfectly clear till some time after sunrise ; it then rapidly fills with dense fog, and so remains for several hours, when at length the fog breaks up and becomes dissipated under the in- creased warmth of the sun. During a short visit to Assam in December 1884, I once witnessed this at Gauhati on the bank of the Brahmaputra. The night was clear and starlit, but the heavy drip from the eaves of the bungalow told that the air was completely saturated, depositing dew copiously. The ground on which the station stands is high, but im- mediately in front flows the broad stream of the Brahma- putra, and there are swamps and tanks in other directions around. The air was perfectly still, unstirred by the slight- est breath, and remained clear until the sun had risen about half an hour, when, in the course of a few minutes, every- thing around was obscured by a dense fog. As I was leaving the station the same morning I had no opportunity of investigating the matter, but the probable explanation appeared to me to lie in the facts of the stillness of the atmosphere during the night, of the alternations of land and water surfaces, and of the saturation of the air probably to a considerable height. It is well known that a land surface cools more by radiation under a clear sky than one of water, and the air over each such surface would approximate to the temperature of that on which it rests. Owing to the entire absence of wind, these air masses might remain un- mingled, though at different temperatures, until set in motion by the sun. But it is a well-known physical fact that two masses of air at different temperatures, and both completely charged with invisible vapour, when intermingled, can no longer hold the whole quantity in suspension. The excess must therefore be deposited as fog. 56 ELEMENTS OF CLIMATE, ETC. These suggestions require verification, which might be supplied by any one who would suspend two minimum thermometers half an hour before sunrise at the same level, one midway over a tank or other water surface, the other over the land at a sufficient distance from the shore, both being sheltered to prevent radiation from the instrument. It would thus be ascertained whether the difference of tem- perature is such as is assumed in the hypothesis. For this purpose the minimum temperatures would suffice. Forms and Nomenclature of Clouds. To an ex- perienced eye the forms and movements of the clouds and the general aspect of the sky are eloquent of impending weather ; but save seamen, and among intelligent and ob- servant nations, farmers and herdsmen, there are but few who have learned their language and can rightly interpret it, while such knowledge as these possess is for the most part empirical and little capable of being harmonised with the facts and translated into the language of physical science. Among meteorological observers of the ordinary class it is certainly rare ; it cannot be learned from books, and unlike the art of managing and reading a thermometer or a baro- meter, cannot be imparted in a few easy lessons. The first difficulty in the way of such teaching is the want of a sufficiently exact language to designate the protean forms in which cloud masses present themselves to the eye; forms which, for one and the same cloud, are different when seen overhead and when seen laterally at a distance ; and second, the difficulty of identifying the particular kind of cloud in- tended from a mere verbal description. Owing mainly to this second difficulty, the art of cloud observation, as treated in text-books, and still more in practice, is involved in end- less confusion, and confusion is inimical to all advance in knowledge. At the beginning of this century, Luke Howard distin- guished some of the principal and most characteristic simple forms of cloud, giving them names which have obtained universal acceptance. Two of these, applied respectively to the highest and to the most common and familiar low cloud HAZE, FOG, AND CLOUD 57 form, are easily referred to their proper objects, and are not often misapplied. These are first, cirrus, feather cloud or mare's tails, the wisps of feathery or fibrous cloud often seen against the blue sky far above any other kind of cloud ; and, second, cumulus or heap cloud, the fine weather cloud of the daytime, with a flat base and a rounded upper surface. But a third name of a simple form, viz, stratus or layer cloud, applied by its author simply to a fog bank, whether resting on the ground or floating at some moderate height above it, has been indiscriminately applied to clouds of very different character at great heights in the atmosphere. It is really characteristic of the night and of the cold weather, always on or near the earth, and the use of the term should properly be restricted to such. So far, however, it is comparatively plain sailing. But when we come to deal with the innumerable forms which occur at all intermediate levels, and which are either modi- fications of these or transitional forms, or compounded of them and other simple forms, the difficulty of distinguishing and identifying the several varieties becomes very great ; and very, much yet remains to be ascertained respecting their mode of formation and their movements, all of which are essential to enable us to interpret their meaning and to draw any trustworthy inference as to what they portend. That this state of ignorance, which is a reproach to meteorological science, will be allowed to endure much longer is, however, unlikely. For some years past, a few excellent observers have devoted themselves to the special study of clouds. Not to name others, Mr. Clement Ley in England, Professor Hildebrandsson at Upsala, and the Hon. Ealph Abercromby in the course of two journeys round the world, have not only studied and recorded their diverse forms with the aid of photography, but what is still more important, have established their connection with those other weather conditions in which they occur, and have shown how certain important inferences may be drawn from their appearance. Of the results at which these observers have arrived, 58 ELEMENTS OF CLIMATE, ETC. only a few preliminary sketches have as yet been published ; but the following outline of a scheme proposed by Messrs Hildebrandsson and Abercromby represents what will prob- ably be the basis of the future classification of clouds, and recognises the very important distinction between such as are characteristic of fine and such as betoken or accompany rainy weather. It may be remarked in connection with this scheme, that Mr. Abercromby, as the result of his widely extended experience, finds that the forms of clouds are similar in all parts of the world, the chief distinction being that certain of them are more common in some latitudes and places than others. MESSRS. HILDEBRANDSSON AND ABERCROMBY'S PROPOSED SCHEME OF CLOUD CLASSIFICATION a. Discrete tending to rounded forms fi. Extended and sheet-like forms (principally in dry weather). (rainy weather). A. Highest clouds, mean height 30,000 feet. 1. Fibre cloud [Cirrus or mare's 2. Thin cloud veil [Cirro-stratus]. tails]. B. Medium elevation 13,000 to 20,000 feet. 3. Small globular cloudlets, shining white like silk, 20,000 feet 5. Thicker ash-coloured or blu- [Cirro-cumulus, mackerel sky]. ish-gray sheet, 17,000 feet 4. Larger globular, like white wool, [Straio-cirrus]. 13,000 feet [Cumulo-cirrus]. C. Lower clouds, 5000 to 7000 feet. 6. Great rounded masses or rolls of 7. Ragged sheets of gray cloud gray cloud [Strato-cumulus]. from which rain commonly falls [Nimbus], D. Clouds in ascending air currents. 8. Heap cloud [Cumulus]. Summits at 6000 feet ; bases at 4500 feet. 9. Storm (thunder) clouds [Cumulo-nimbus]. Summits 10,000 to 16,000 feet ; bases 4500 feet. E. 10. Elevated fogs. Below 3500 feet [Stratus]. The elevations given * are furnished by measurements made 1 The originals are given in meters, which I have converted into feet in round numbers, as great exactitude is not claimed for them, nor indeed possible. HAZE, FOG, AND CLOUD 59 in Northern Europe. It is probable, nay, almost certain, that they would have to be extended in India to greater heights. Not only does this seem to follow from the con- sideration that at the ground surface, and at equal heights above it, the temperature is higher in India than in Europe, and that temperature must be one principal condition in determining the character of the cloud ; but it is also con- firmed by direct observation. It is common at Simla to see clouds of the D class shrouding the summits of the snowy range, and rising some thousands of feet above its highest peaks, which range to 20,000 feet; and the plane of the cloud bases evidently slopes regularly upwards from over the plains towards the snowy range. But this does not affect the general validity of the scheme, which shows how cloud forms become modified as they sink lower in the atmosphere, and also how certain forms are characteristic of dry or wet weather respectively. Scud, or torn masses of rain cloud, is regarded as only a slight modification of Nimbus borne along by a gale. It often descends very low ; in cyclones only a few hundred feet above the earth's surface. The tops of thunder clouds are often seen tailing off laterally in a fibrous tuft like cirrus. This, Pro- fessor Hildebrandsson calls false cirrus, and believes that like true cirrus it is composed of ice spicules, not of water corpuscules. The Teachings of Clouds. According to the above scheme the inferences to be drawn from the behaviour of the clouds are obvious. If they show a tendency to form sheets, at first high and thin but gradually becoming darker and more opaque, showing that they are increasing in thick- ness, sinking at the same time lower in the atmosphere, it is a sign of probable rain. But if, on the other hand, such a sheet is breaking up into separate masses, and sunlight begins to be visible through the thinner portions, it is a sign of the weather's clearing. These signs are in full accord w T ith universal experience. Besides these, the clouds teach other lessons. The movements of the higher clouds tell us what winds are 6o ELEMENTS OF CLIMATE, ETC. blowing in those elevated regions of the atmosphere which are otherwise inaccessible to observation. To a casual ob- server this movement may seem to be very slow ; but this is deceptive, and owing to the great height at which it takes place. As a fact, it appears from actual measure- ments made by Dr. Vettin at Berlin, the high feathery cirrus travels at an average rate of forty miles an hour, or about three times that of the surface winds ; and some such result might perhaps have been expected when we bear in mind how small is the retarding friction at great heights in the atmosphere. 1 It is only of late years, and only at two stations in Northern India, viz. Calcutta and Allahabad, that the higher cloud movements have been systematically observed in India. Opportunities for observing such do not often present themselves, and the following tables enumerate the total results of the observations of six years at Calcutta and eight years at Allahabad : MOVEMENTS OF CIRRUS CLOUDS OBSERVED AT ALIPORE (CALCUTTA) DURING THE SlX YEARS 1881 TO 1886 DAYS OF Moi CEMENT FROM Months. Total N. N.E. E. S.E. s. S.W. w. N.W. January 13 27 2 42 February 1 14 16 3 34 March 2 5 18 8 33 April , . n 4 8 3 15 May . ; 1 2 1 2 4 3 1 14 June 2 1 6 5 4 18 July . 3 7 6 16 August 3 7 3 7 20 September 3 3 9 2 7 3 3 30 October 2 1 1 3 5 4 8 24 November 3 1 2 18 27 1 52 December l 3 2 7 27 5 44 1 The recently-published report on the eruption of Krakatoa shows that the ash cloud emitted by the volcano was carried westward round the world in an elevated stratum of the atmosphere at the rate of from 70 to 76 miles an hour. 2 For December, five years only. HAZE, FOG, AND CLOUD 61 MOVEMENTS OF CIRRUS CLOUDS OBSERVED AT ALLAHABAD DURING THE EIGHT YEARS 1878 TO 1885 DAYS OF MOVEMENT FROM Months. N. N.E. E. S.E. s. s.w. w. N.W. lotal. January 1 27 44 20 92 February 2 1 3 18 55 . 9 88 March 1 2 3 3 28 58 8 103 April 2 1 1 10 31 15 60 May . 8 22 7 37 June . 6 6 4 2 15 24 7 64 July . 4 10 3 3 7 8 3 38 August 2 6 10 2 1 7 9 4 41 September October 2 4 1 11 1 5 1 3 6 7 9 5 3 1 43 16 November 1 1 4 6 2 14 December . 1 21 17 3 42 Hence, as far as the occasional high cloud affords any basis for judgment, the winds in the upper part of the atmosphere appear to be chiefly west and south-west over both Calcutta and Allahabad, except in the summer monsoon, when thev are variable, from all quarters at Allahabad, and at Cal- cutta chiefly southerly. When a cyclone is forming in the bay, cirrus cloud from the south-west is often seen passing over Bengal before there is any other distinct indication of its existence. Cloud Proportion. The average cloudiness of different parts of India, at different seasons of the year, varies very much with the dampness of the atmosphere, as already described in the previous section. The usual mode of re- cording it is to estimate the proportion of the sky that is covered by cloud, irrespective of the kind of cloud. The whole expanse of sky overhead, that is, so much as is in- cluded in an angle of 60 on either side of the zenith, is designated by the number 1 0, and the amount of the cloud is estimated as so many tenths of that expanse. This is the meaning of the figure expressing cloud proportion that stands against each station in the daily reports. A com- pletely clear sky is thus indicated by 0, an entirely over- cast sky by 10. The average cloudiness of a day, a month, 62 ELEMENTS OF CLIMATE, ETC. or a place is found in the usual way, by taking the arith- metical mean of all the individual observations included in the period or relating to the place in question, care being taken that the observations are so timed as to represent equally or in due proportion all the phases of variation. Distribution. The comparative serenity of the skies of most parts of India is shown by the fact that, except in the Carnatic, Ceylon, the southern parts of Pegu and Tenasserim, Assam, and Sikkim, the mean cloud proportion is every- where below 5. In Scotland, according to Mr. Buchan, it is between 6 and 7. In Sind and the southern Punjab it is below 2, and in the remainder of the latter province, except on the hills, scarcely exceeds 3, on the average of the whole year. In Baj put ana it varies between 2 and 3 '5, and in the remainder of India between 3 and 5. The cloudiest stations at which observatories exist would appear to be Hambantotte, on the south coast of Ceylon, and Sib- sagar in Upper Assam, at both of which the mean proportion amounts to 7 or more. In the case of Sibsagar this is partly due to the fogs of the cold weather, which are pro- perly reckoned as cloud. In the summer months alone, Mercara, on the crest of the Western Ghats, is the cloudiest station at which an observatory exists, the proportion, from June to August, being between 9*5 and 10. The cloudiest month in the year in by far the greater part of India is July, the proportion at most stations being between 7*5 and 9 ; and August is nearly as cloudy. In the western Punjab and Upper Sind, which are scarcely ever reached by the summer rains, it is February or March, but here the proportion does not much exceed 3 on the average of any month. In Baluchistan, where March is also the cloudiest month, it amounts to nearly 5. In the extreme south of the peninsula, and on the east coast of Ceylon, the last two months of the year are more cloudy than any of the summer months, and also the dampest and rainiest. But in the western half of the island, and also in Malabar, June is the cloudiest. RAINFALL 63 The season of greatest serenity is more variable, ranging in different parts of India from October to March. In the Punjab and Sind October is the clearest month, the average proportion of cloud being below 0*5 at the drier stations, and little exceeding 1 in the north-east of the tract. In Guzerat, Eajputana, Central India, and the Gangetic plain, the skies of November are slightly clearer; and in Bengal and Assam both months are surpassed by December and January. In these provinces and months the average cloud is between 1 and 2. In the peninsula, again, the first three months of the year are the clearest, in general, Feb- ruary ; and as far down as Bellary and Dharwar, the skies are then as clear as in Bengal ; but farther south the pro- portion does not fall much below 2 '5 in the clearest month, and in Ceylon it exceeds 4. In Burmah also the first three months of the year are the clearest, the proportion in the least cloudy month averaging about 0*5. at Tounghoo and as much as 3 at Mergui. THE EAINFALL The Contrasts of Indian Rainfall. What has already been said of the variations of the climate, in respect of dampness and cloudiness, will have prepared the reader to expect a great diversity in the rainfall of different parts of India ; and, in point of fact, no country in the world fur- nishes such contrasts as Northern India. In the east we have the station of Cherrapunji in the Khasi hills, with an annual fall of from 500 to 600 inches; and, in the west, Jacobabad, with an annual average of 4^ inches, and in some years less than 1 inch. 1 Between these extremes we 1 The rain-gauge is so well known an instrument that any detailed de- scription is scarcely needed. The essential parts are a funnel into which the rain falls, and the area of the aperture of which regulates the quantity received ; a receiver, in which the water is collected ; and a graduated glass to measure the quantity. The glass must be adapted to the area of the funnel. It is of such capacity that when filled to the topmost mark it holds half an inch (or one inch) of rainfall, and the graduation shows the tenths 64 ELEMENTS OF CLIMATE, ETC. have ever} 7 possible variation, and the contrasts afforded by large provinces are only less striking than that of these prominent examples. In Southern India the diversity is also very great, but here the relative positions of the dry and wet tracts are reversed. It is the west coast of the peninsula that receives the heavy rainfall of the summer monsoon, while it is withheld from the plains of the south- east coast, and still more from certain portions of the high plains of the interior; and the transition from the zone of heaviest rainfall to that where it is least and lightest is extraordinarily rapid. Thus, on the crest of the Western Ghats are the two stations, Baura Fort and Mahableshwar, with an average fall of 251 and 261 inches respectively. Only sixty-five miles distant from the former is Gokak, the annual average of which is no more than 22 inches, or less than one-tenth of the former ; and this is representative of a broad zone extending, with but slight interruption, all down the peninsular to the east of the Ghats, in which the rain- fall averages from 18 to 25 inches only in the year. See Fig. 6. Scarcely less remarkable is the occasional heaviness of the falls, even at places where the average rainfall is not by any means excessive. That upwards of 40 inches in the 24 hours should have been once recorded at Cherrapunji will perhaps hardly seem surprising, but falls nearly as great (from 30 to 35 inches), in the same interval, have occurred on more than one occasion on the plains of the Ganges valley, at places where the average of the whole year is not more than from 40 to 65 inches; and even in the extremely arid province of Sind as much as 20 inches fell on one day in 1866 at Doorbaji, where the annual average is probably less than 6 inches. Another noteworthy peculiarity in the rainfall of India is the variability of different parts of the country in respect of the regularity of the rainfall. As will presently be shown more in detail, as a general rule, those provinces and hundredths of an inch. An inch of rainfall means that if collected on an impervious level surface it would form a sheet of water one inch deep. RAINFALL 65 which have the lowest rainfall are also those in which it is most precarious, and the most rainy parts of India are also those in which the rainfall is most regular. But there are some striking partial exceptions to the rule ; or rather, among the provinces with a moderate average rainfall, some are subject to great vicissitudes, while in others it is re- markably regular. Thus, Nagpur and other parts of the Central Provinces south of the Satpura range have a very regular rainfall, and never suffer from droughts of any magnitude, although the annual average fall does not exceed 50 inches, whereas the North-west Provinces, with a general average of 36 inches, have repeatedly been subject to disastrous droughts ; not less than seven times in the course of the present century. It is not my purpose, in this work, to discuss the physical circumstances that determine these and other peculiarities in the rainfall of the country. To the physicist and geographer they present problems of the greatest in- terest, and afford a most instructive field of study. But they have lately been discussed at length in a somewhat bulky volume, 1 issued by the Meteorological Department of the Government of India, and the reader, whose interest in the meteorology of the country attaches rather to its physical than its statistical aspects, may be referred thereto for such information as we possess on the subject. The bearing of these rainfall variations on the agriculture and general welfare of the population are more germane to the objects of the present treatise, and only in this connection need they be here discussed. I shall first describe the general distribution of the rainfall over India, then its relations to the seasons, and lastly, the more important vicissitudes to which it is subject. Geographical Distribution. The chart of the average rainfall of India (Fig. 6) shows that the west coasts of both peninsulas are fringed with a belt of very heavy rainfall, where, as may be seen in Fig. 4, the westerly summer monsoon blows full upon the land. In both, at a distance 1 Indian Meteorological Memoirs, Vol. iii. "The Rainfall of India." F 66 ELEMENTS OF CLIMATE, ETC. of 50 miles or less from the sea, a hill range, such as in many countries would be called mountains, runs parallel with the coast, and over the whole intervening space the rainfall is very high, varying from 100 to 250 inches on the coast of India, and probably of larger amount on the coast of Arakan. In the western peninsula this zone of high rainfall contracts rapidly to the north of Bombay, 76 80 84 88 92 96 DIAGRAM SHOWING AVERAGE ANNUAL RAINFALL 3 INDIA U Below 30 Inches &Z% From 30 to 50 Inches " o From 50 to 70 Inches I From 70 to 100 Inches I Above 100 Inches 84 I rattier &Boutau FIG. 6. and dies out at the end of the Ghats, south of the Tapti. But in the eastern peninsula it is continued northward beyond the shores of the Bay of Bengal, and beyond the tropic, through eastern Bengal and Cachar, and along the face of the Khasi hills to the gates of Assam, where it links on to the similar zone, which runs east and west along the face of the Himalaya. In a westerly direction, this latter zone, as denned by a rainfall RAINFALL 67 of over 100 inches, has not been actually traced beyond Sikkim, owing to the absence of rainfall registers from this part of the range ; but it probably extends uninterruptedly to the farther extremity of the Himalayan frontier of Nepal, and is taken up again in the far north-west by the southern slopes of the Dhaoladhar range, overlooking the valley of Kangra. Eastward, there is little doubt that it extends to the upper end of the valley of Assam, where the rainfall in the valley itself, at Dibrugarh, amounts to more than 100 inches. The zone of heavy rainfall on the face of the Himalaya does not extend far into the mountains. It is broadest in Sikkim, where it includes a strip of the plains of northern Bengal; and, in a northerly direction, up to the valley of the Great Eungit, and perhaps somewhat beyond, the annual fall exceeds 100 inches. At Katmandu, in the Nepal valley, the rainfall is only 5 7 inches. In the north- west Himalaya, beyond the Nepal frontier, it is barely 100 inches on the outermost slopes of the range (except, as already mentioned, on the Dhaoladhar), and it decreases rapidly as we penetrate farther into the hills, until, beyond the first snowy range, we enter on the dry region of Tibet, represented by Leh, with an annual fall of only 2^ inches. From the Gulf of Cambay and the Sabarmati river, a band of moderately high rainfall (between 30 and 70 inches) stretches eastwards across India, rapidly broadening out and increasing in average amount in the direction of Bengal. It occupies all North-eastern India, extends to the valley of the Ganges on the north, and on the south covers the drainage basin of the Godavery and its tributaries, down to the delta of that river and the Kistna. From Allahabad an offshoot runs up to the north-west, including the Gan- getic plain north of the Ganges, and continues through the northern Punjab and along the face of the Himalaya as far as the Indus valley. In the south of the peninsula are two outlying tracts with a rainfall above 3 inches ; one on the Eastern Ghats, east of Kurnool ; the other, including all the central districts of the Carnatic and linked on to the 68 ELEMENTS OF CLIMATE, ETC. zone of high rainfall of the Western Ghats, akmg the face of the Eastern Ghats ; and again, farther south, across the hill groups between Trichinopoly and Madura. The remainder of India has an annual rainfall below 30 inches. It consists of (1) the dry north-western region, which includes Eajputana, Sind, Cutch, part of Guzerat, the Punjab (with the exception of the strip of the plains border- ing the Himalaya), and nearly the whole of the Gangetic doab (the plain between the Ganges and Jurnna) ; and (2) the peninsular dry region, which comprises Khandesh and the valley of Berar, the Western Deccan, the southern dis- tricts of Hyderabad, the ceded districts of the Madras Presi- dency, the northern Carnatic, Mysore, Coimbatore, and the plain in the south-east of the peninsular, between Tanjore and Cape Comorin. In the former the rainfall decreases steadily in a westerly direction, till it reaches its minimum in the Cutchee desert, between Jacobabad and the hills of Baluchistan. In the latter are several isolated tracts, in which the rainfall is somewhat below 20 inches, the prin- cipal of which are a large area around Bellary, and a smaller one in Coimbatore, between the Mlgiris and the Shevaroy hills. Also, on the narrow strip of coast plain bordering the Gulf of Manaar, in the extreme south-east of the peninsula, facing Ceylon, the rainfall is equally low. In Burmah, from the crest of the Arakan range, and from the mouths of the Irawadi, the rainfall rapidly decreases towards the interior of Pegu, and falls still lower as one ascends the river towards Mandalay. But it is compara- tively high on the Pegu Yoma, east of the Irawadi delta, on the hills around the source of the Sitang, and on the range west of the Salwin. From Upper Burmah our present in- formation is insufficient to show the features of its distribu- tion. But rain-gauges were established at a number of stations of the newly-acquired province, even before it was fairly pacified, and in the course of a few years it will be possible to extend the rainfall chart of the Burman penin- sula at least to Manipur and Bhamo. The average rainfall of the whole of India up to the foot RAINFALL 69 of the mountain barrier, including Assam and Cachar, but not the Burman peninsula, has been computed at 42 inches in the year ; that is to say, such would be the depth of the sheet of water, if the whole quantity that falls on this area were equally distributed over its whole surface, instead of being concentrated in certain provinces to the great deprivation of others. Of the general character of its distribution some idea may be gathered from the table given below. In this table the country is partitioned out in a number of rainfall provinces, partly conforming to administrative divisions, but with such modifications as to avoid bringing together parts of the country which differ very greatly in their average amounts. AVERAGE ANNUAL RAINFALL OF THE PROVINCES OF INDIA AND BURMAH Eainfall Provinces. Area in Square Miles. Number of Stations. Average Rain- fall. Inches. Local Variation. Inches. Punjab Plains . . 120,000 29 22 6 to 36 X. W. Provinces and Oudh . 83,500 45 36 25 50 Rajputana (Eastern only) 67,000 19 28 14 63 2 Central India States l . 91,000 21 42 32 55 Behar 30,000 14 43 39 48 Western Bengal, Chutia Nagpur, etc. 38,000 10 49 43 61 Lower Bengal .... 54,000 29 66 54 112 Assam and Cachar .... 61,000 17 94 69 475 2 Orissa and Northern Circars . 27,000 16 47 31 70 Central Provinces, South 61,000 19 51 43 79 2 Berar and Khandesh . . . 43,000 11 35 21 69 2 Guzerat . 54,500 14 33 18 47 Sind and Cutch . . . 68,000 10 9 4 19 North Deccan 48,000 14 29 18 49 Konkan and Ghats 16,000 13 141 74 261 2 Malabar and Ghats . . . '". 18,000 8 114 74 132 Hyderabad . . . . 71 000 19 32 23 43 Mysore and Bellary 58,000 17 29 18 36 Carnatic ..... 7^ 000 40 36 20 62 2 Arakan ^ 11,000 6 156 105 214 Pegu 32,500 7 73 46 123 Tenasserim . . . . . 10,500 4 171 142 196 1 Including Jhansi, Saugor, and Damoh, and the Nerbudda valley. 2 In these cases one or more hill stations have a much higher rainfall than any at the lower levels. 70 ELEMENTS OF CLIMATE, ETC. Seasonal Distribution. By far the greater part of the rainfall is brought by the summer monsoon; and over a great part of Western India, including nearly the whole of the Bombay Presidency, it is only during this season, viz. from June to October, that there is any rain of importance. This is the chief season of rainfall in Burmah, Assam, Bengal, the North-west Provinces, the greater part of the Punjab, Central India, Eajputana, the Central Provinces, Hyderabad, the Deccan, Berar, Guzerat, and the whole of the west coast of the peninsula. But the south-east of the peninsula, including the Carnatic and the Eastern Ghats, and the eastern half of Ceylon, receive only an occasional shower so long as the main body of the monsoon is directed towards Northern and Central India, and their season of heaviest rainfall comes later, viz. from October to December, when north-east winds are beginning to blow in the north-west of the bay, and both the incipient north-east monsoon and the residue of the southerly current are drawn towards the Carnatic and the southern half of the bay. In these months the seat of the chief rainfall moves southwards, till at length, by the end of December, the north-east monsoon has established itself in the bay, and, in the south, rain is restricted to a zone a few degrees north of the equator, in which it falls, more or less, all through the year. In the Punjab and Eajputana the summer rainfall is much interrupted. It seldom penetrates to the Indus or to the Himalaya beyond the first snowy range. In the in- terior of this range, and also on the mountains and high plains beyond the north-western frontier, in Afghanistan and Baluchistan, the latter part of the winter and the earlier spring months are the season in which the chief additions are made to the snows ; and on the outer hills and plains of the Punjab, rain, sometimes heavy, occurs at intervals throughout this season. It extends also, though less frequently and heavily, to the Gangetic plain, and some- times, but still more rarely, to the whole of Northern India, including Bengal. Occasionally it occurs even in the Northern Provinces of the peninsula, but this is exceptional. RAINFALL 71 In the spring, when sea winds blow on the south and east of the peninsula, and very steadily and sometimes strongly in Lower Bengal and Cachar, thunder-storms are not infrequent, and sometimes the showers are heavy and prolonged. Chiefly is this the case in the damp province of Assam and Eastern Bengal ; and it is in consequence of these frequent falls of rain that Assam and Cachar are so favourably distinguished as tea -growing provinces. The showers become more frequent as the season advances, and the May rainfall of Assam is not very far short of that of June, the first month of the monsoon rains. In Southern India, too, where the spring rains are known as the " mango showers," they amount to a total of three or four inches, falling chiefly in the months of April and May. But from the Bombay Presidency, north of Dharwar, the spring rainfall is almost entirely withheld. This is the season of north-westerly winds, as has been explained on a previous page, and these, being dry land winds, bring no rain. Sometimes the rain of the spring thunder-storms takes the form of hail, and, though not very common on the plains, it is very destructive, the hailstones being generally large, from the size of marbles up to ice concretions two or three inches across. On the Himalaya hailstones occur more often than on the plains, but as far as I can speak from personal experience the hailstones are generally of moderate size. There are, however, recorded instances of very large hailstones on the hills. Thus, there is no time of year in which rain does not fall in some part or other of India. No sooner have the Car- natic rains come to an end in December, than the winter rains are setting in in the most northern province; and while these are prolonged into the spring months, the spring rainfall of the south and east of the Indian peninsula, of Bengal, Assam, and the Southern Provinces of Burmah, is already preparing the fields for the spring sowings, and bringing out flushes of young leaves on the tea bushes of Assam, Chittagong, and Sikkim. In the interior of Southern India these showers become less frequent in June ; but on 72 ELEMENTS OF CLIMATE, ETC. the south-west coast the monsoon rains set in at the end of May, and reach Bombay in the first or second week of June ; while, in Assam, the spring rainfall graduates into the steadier downpour of the summer monsoon. In September this monsoon ceases to blow up to North-western India, and after concentrating a diminished rainfall in Bengal and the north-east of the peninsula, gradually transfers it south- wards to Madras, and finally to Ceylon and the equatorial sea, when the southerly current has completely yielded place to the winter monsoon of the bay. Statistics of Monthly Rainfall, Quantity and Fre- quency. The following two tables, one giving the average monthly rainfall of a number of important stations in all parts of India and Burmah (to the nearest tenth of an inch), the other the average number of rainy days in each month, will serve to illustrate the above description, and show in somewhat greater detail the relative proportions in which the rainfall is distributed over India, and also through the year. The second table is based on the registers of the last 1 1 years ; the first, in most cases, on those of much longer periods. A more comprehensive table of the rainfall is given in Appendix III. AVERAGE MONTHLY RAINFALL OF EIGHTY STATIONS IN INDIA, CEYLON, AND BURMAH _c RAINFALL IN INCHES. Stations. I* 1 I i t g 1 I 1 p & 5 I 8 Kurrachee 49 0-6 0-3 0-2 0-2 o-i 0-2 3-1 1-8 0-9 O'l o-i 0-2 7'8 Hyderabad 66 0-2 o-i o-i 0-2 o-i 0-4 2-8 3'2 0-8 O'l 8-0 Quetta . 5,501 1-6 1-9 2'4 1-3 0-5 o-i 0-8 0-6 0-2 O'l 0'4 9'9 Jacobabad 186 0-2 0-2 0-3 0-2 o-i o-i 1'4 1-4 0*3 o-i o-i 4-4 Mooltan 420 0'4 0-3 0-5 0-3 0-5 0-4 2'2 1-3 0*8 O'l o-i 0-3 7'2 D. I. Khan 573 0-4 0'7 0-9 0-8 0-4 0-6 1-8 1-6 0-6 o-i o-i 0-3 8'3 Leh 11,503 0-2 0-2 0-2 o-i o-i 0*2 0-5 0-4 0*2 0-5 o-i 2-7 Peshawar 1,110 1-6 1-2 1-8 2'0 07 0-3 1-7 2*0 0*8 0*2 0-6 0-6 13-5 Rawalpindi 1,652 2'4 2-0 1-9 2-3 1-6 1-7 7*4 7'3 3*2 0-6 0-9 1-1 32-4 Murree 6,344 2-8 3-4 37 4'3 3*8 2-4 11 -0 14-0 6-1 2-2 1-7 1-2 56-6 Sialkot 829 1-4 1-8 1-9 1-6 1-2 3'2 11-6 9-1 3'2 0-6 0*4 0-8 36-8 Lahore 732 07 1-1 1-1 0-6 0-9 1-8 7'4 4-6 2'4 0*6 0-2 0*5 21*9 Umballa 902 1-4 1-5 1-1 0-6 i-o 4-0 11-6 8'6 4*1 0*6 0-2 0-7 35*4 Simla 6,953 2'8 2-7 3-0 2'8 4'7 7-9 19-3 18-1 6-0 1-4 0-3 1-1 70'1 Delhi 718 1-0 0-5 0-7 0-4 0*7 3'4 8-5 6'9 4-5 0-5 o-i 0-4 27-6 Roorkee 887 2-0 1-4 i-o 0-4 1-2 5*1 12*5 12-3 5-1 0-6 0-2 0-4 42'2 RAINFALL 73 AVERAGE MONTHLY RAINFALL OF EIGHTY STATIONS IN INDIA, CEYLON, AND BURMAH Continued Stations. Elevation in Feet. RAINFALL IN INCHES. 1 & 1 1 1 1 i i-s >* hs to J3 <1 +J 1 1 1 1 1 Meerut . 737 1-0 07 0-7 0-4 0-8 3-6 9-2 7-2 4-0 0-5 o-i 0-3 28-5 Agra 555 0-5 0-3 0-2 0-2 0-7 2-9 9-8 6-7 4-3 0-4 0-2 26-2 Deesa . 465 o-i 0'2 o-i o-i 0-2 2-2 9-8 8-5 3-3 0-8 o'-i 25-5 Abu 3,945 0'2 0-4 o-i 1-0 5-1 22-2 22-5 9-1 2-1 0-2 0-2 63*1 Jodhpur 1,274 0-3 0-2 o-i 0-6 1-3 3-6 5-2 2-3 0-2 o-i o-i 14-0 Neemuch 1,(539 o-i 0-2 o-i o-i 0-5 3-9 11-2 10-4 5-5 I'O 0-2 33-2 Indore . 1,822 0-4 0-3 o-i 0-6 6-8 10-4 7-8 8-1 1-2 0*-2 0-2 36-1 Ajmere . 1,611 0'2 0-3 o'-4 o-i 0-7 2-5 6-9 7-3 3-4 0-3 o-i 0-3 22-5 Jhansi . 855 0-5 0-2 0-4 o-i 0-3 4-0 13-6 10-5 5-2' 0-8 0-2 35-8 Luckuow 369 0-8 0-3 0-3 o-i 0-9 5-0 10-8 10-4 7-1 1-4 0-5 37-6 Allahabad . 307 0-8 0-4 0-4 0-2 0-3 4-6 11-9 9-6 6-7 2-3 o'-2 0-2 37-6 Benares 267 07 0-5 0-4 0-2 0-5 5-0 12-8 10-7 6-5 2-1 o-i o-i 39-6 Gorakhpur . 256 0-7 0-5 0-4 0-3 1-5 7-7 13-3 11-8 8-8 3-0 0-2 o-i 48-3 Patna . 179 0-7 0-5 0-3 0-3 1-6 7-1 11-0 10-1 7-9 2-9 0-2 0-2 42-8 Darjiling 6,912 0-7 1-2 2-4 3-7 7-1 24-1 30-5 26-0 17-8 6-4 0-2 0-2 120-3 Jalpaiguri 270 0-4 0-4 1-7 3-7 11-2 28-6 26-3 25-9 24-3 5-5 o-i o-i 128-2 Hazaribagh . 2,010 0-4 0-8 0-7 0-4 1-6 8-3 12-6 12-7 8-0 3-4 0-3 0-2 49-4 Bhagulpore . 159 0-5 07 0-4 0-8 2-5 8-3 11-2 10-7 7-8 4-1 < 0-2 o-i 47-3 Berhampore . 66 0-4 1-0 i-o 1-9 4-8 9-7 10-3 10-8 9-8 5-3 ' 0-3 o-i 55-4 i Calcutta 18 0-4 1-0 1-3 2-3 5-6 11-8 13-0 13-9 10-0 5-4 ; 0-6 0-3 65-5 Dacca . 15 0-3 1-1 2-5 5-8 9-2 13-3 12-8 12-4 10-2 5-2 ' 0-7 0-2 73-7 Chittagong . 87 0-4 1-2 1-9 4-6 9-2 23-8 22-2 20-5 14-1 5-7 : 1-6 0-6 105-8 Silchar . 87 0-6 2'6 7-9 13-0 15-7 19-1 20-6 18-2 14-2 6-4 ? 1-0 0-7 120-0 Cherrapunji . 4,455 0-6 2'6 9-0 29-6 50-0 110-0 120-5 78-9 57-1 13-6 1-8 0-3 474-0 Gauhati 370 0-6 0-9 2-5 5-8 10-1 12-9 12-7 11.2 8-1 3-1 0-6 0-3 68-8 Sibsagar 333 1-1 2'2 4-4 9-9 11-1 14-1 15-6 16-0 11-7 5-2 1-3 0-6 93-1 Cuttack . 80 0-4 0-6 1-1 1-5 3-2 10-7 12-6 11-2 9-8 5-8 1-0 0-5 58-4 Samblapur . 451 0-6 0-6 0-7 0-5 1-6 13-0 17-7 15-2 8-7 2-4 0-3 0-4 61-7 Saugor . 1,769 06 0-5 0-2 0-2 0-6 6-3 16-8 11-2 7-3 1-3 0-4 0-7 46-1 Jubbulpore . 1,351 0-6 0-5 0-5 0-2 0-5 8-5 18'6 13-8 8-2 1-5 0-4 0-3 53-6 Paclimarhi . 3,504 0-5 0-3 0-4 0-3 0-6 10-8 28-8 18-2 15-1 1-9 0-4 0-7 78-0 Nagpur . 1,025 0-6 0-4 0-6 0-5 0-8 8-8 13-3 8-9 7-8 2-3 0-4 0-5 44-9 Amraotf 1,213 0-5 0-2 0-3 0-2 0-6 6-9 8-8 7-0 5-3 1-6 0-2 0-6 32-2 Dhulia . 1,000 0-3 O'l 0-4 4-8 4-8 4-0 4-6 2-0 0-5 0-4 21-9 Poona . 1,819 0'2 6-2 6-6 1-6 5-6 6-6 4-1 4-3 4-1 0-8 0-2 28-3 Bombay . 37 o-i 0-5 20-8 24-7 15-1 10-8 1-8 0-5 o-i 74-4 Matheran 2,200 o-i 0-8 35'7 84-4 50-7 31-0 5-3 0-6 o-i 208-7 Mahableshwar 4,540 0-4 d-i 6-4 6-9 1-4 47-3 102-1 68-6 32-9 5-8 1-1 0-4 261-4 Karwar . 44 0-2 o-i 0-3 2'9 34-5 37-7 21-7 ^O 5-2 1-7 0-2 116-5 Belgaum 2,550 o-i 0-5 2-0 2-8 9-3 15-2 9-0 ;3-7 4-7 1-2 0-3 48-8 Jalna o-i 01 o-i 0-8 7-2 6-8 5-3 7-5 3-4 0-8 1-4 33-5 Sholapur 1,819 d-i 0-3 0-7 1-2 4-6 4-3 6-0 r-5 3-7 0-7 0-4 29-5 Secunderabad 1,787 0-3 0-2 0-7 0-7 1-4 3-7 6-0 5-7 5-2 3-3 0-8 0-3 28-3 Bellary . 1,455 o-i 0-6 0-8 1-8 1-8 1-3 2-3 3-7 3-9 1-0 0'3 17-6 -Bangalore 2,981 0-2 6-i 0-6 1-3 5-0 3-2 4-0 5-9 6-3 6-4 1-9 0-7 35-6 Wellington . 6,200 0-8 0-3 2-0 2-9 4-1 3-6 3-2 4-0 4-7 9-8 8-5 4-1 48-0 Bimlipatam . 30 0-3 0-5 0-2 0-2 2-0 3-2 3-6 4-4 6-3 8-1 2-5 1-3 32-6 Masulipatam . 10 0-3 o-i 0-3 o-i 1-7 4-4 5-6 6-0 6-5 8-8 4-0 0'7 38-5 Rajamundry . 68 0-2 0-3 0-3 0-9 3-3 4-5 7'2 6-6 7'1 6-5 1-7 0-2 38-8 Cuddapah 477 o-i 0-2 0-3 1-6 2-6 3-5 5-1 5-8 5-3 3-1 0-7 28-3 Madras . 22 i-o 6-3 0-4 0-6 2-2 2-1 3-8 4-4 4'7 10-8 13-7 5-1 49-1 Trichinopoly. 275 1-0 0-5 07 1-8 3-8 1-3 2-2 4-4 5-3 7-8 5-2 3-1 37-1 Cuddalore 20 1-0 0-3 0-4 0-9 1-5 1-4 2-3 5-1 4-7 8'1 14-1 5-7 45-5 Madura . 448 07 0-4 0-6 2-0 2-8 1-6 1-7 4-7 4-5 8-7 5-1 2-2 35-0 Mangalore 52 0-2 o-i o-i 2-0 8-1 37-8 37-9 23-1 11-3 8-0 1-9 0-5 131-0 Cochin . 11 0-9 0-7 2-1 4-4 12-7 30-7 22-7 12-4 9-4 12-1 5-1 1-9 115-1 Trincomalee . 175 6'2 2-4 1-3 1-6 2-2 1-9 2-2 4-2 4-6 8-9 13-1 13-2 61-8 Colombo . 40 3-0 1-7 5-5 8-8 13-2 8-2 5-5 4-5 4-9 12-9 12-7 6-4 87-3 Galle . 48 4-4 3-3 4-7 8-7 11-6 8-2 5-7 5-3 7-6 13-0 11-5 6-7 90-7 Rangoon Moulmein 41 94 0-2 o-i o-i o-i o-i 1-8 3-0 10-9 19-7 18-4 38-4 21-3 43-9 18-6 43-0 16-0 30-3 8-1 84 3-4 1-5 o-i o-i 99-0 188-5 Thyet Myo . 134 o-i 07 5-3 7-9 8-0 8-5 7-8 4-9 2-3 45-5 Tounghoo 169 0-2 1-5 6-6 13-4 17-5 18-1 11-8 7-4 1-4 6-2 78-1 Akyab . 15 o-i 0-2 0-5 1-6 12-2 51-6 51-0 38-6 23-0 12-4 3-9 0-6 195-7 74 ELEMENTS OF CLIMATE, ETC. MONTHLY AVERAGE OF RAINY DAYS AT EIGHTY STATIONS IN INDIA, CEYLON, AND BURMAH Stations. RAINY DAYS. I & .' 1 I >> qj a 3 1-5 >> 13 1-3 tc 3 "4 -^ - A 1 1 Kurrachee 1 2 1 1 1 6 4 o 1 19 Hyderabad . 1 1 "i 1 5 5 2 ... 16 Quetta . 7 7 "s 8 3 1 2 2 1 i' 40 Jacobabad 1 3 1 1 1 3 4 14 Mooltan 1 1 1 1 1 2 3 2 1 ... 'i' 14 D. I. Khan . 2 3 3 3 2 2 4 3 2 1 i' i 27 Leh . 4 5 2 1 2 1 3 3 1 ... 2 24 Peshawar 3 4 4 5 3 1 3 3 2 1 2' 2 33 Rawalpindi . 5 5 5 6 5 4 9 11 6 3 2 3 64 Murree . 4 5 8 10 9 7 14 18 9 4 3 1 92 Sialkot . 2 4 3 3 3 4 9 9 4 1 1 1 44 Lahore . 2 3 2 3 3 3 7 6 4 1 1 2 37 Umballa 2 2 1 1 2 3 9 7 4 1 1 33 Simla . 3 5 6 6 9 10 21 22 12 2 i 2 99 Delhi . 2 2 2 1 3 4 10 9 5 1 1 40 Roorkee 3 4 3 2 3 6 15 14 7 1 i 2 61 Meerut . 2 2 2 1 3 5 11 11 6 1 1 45 Agra 1 1 2 1 3 4 13 10 7 1 ... 1 44 Deesa . 1 1 1 4 13 12 8 1 42 Abu . 1 1 "i 1 3 8 22 22 15 3 i 1 79 Jodhpur 1 1 1 3 3 7 8 4 ... 28 Neemuch 1 1 "i 3 8 15 14 11 2 1 57 Indore . 1 1 3 11 20 19 15 4 i 1 76 A j mere . 1 2 "i "i 4 4 13 11 17 1 i 1 47 Jhansi . 1 1 1 i 2 7 15 14 9 1 ... 1 53 Lucknow 2 2 2 i 2 6 16 14 9 2 1 57 Allahabad 2 2 1 i 2 7 16 17 11 3 ... 1 63 Benares 2 2 1 i 2 7 18 18 12 4 i 1 69 Gorakhpur . 1 2 1 i 3 7 13 14 9 3 1 55 Patna . 2 2 1 i 4 10 18 18 12 4 1 73 Darjiling 3 4 5 9 19 22 27 26 21 6 1 143 Jalpaiguri 1 1 3 6 15 21 23 21 17 6 1 115 Hazaribagh . 3 3 3 1 8 15 24 23 17 7 2 1 107 Bhagulpore . 2 2 1 2 5 12 19 18 12 4 1 78 Berhampore . 1 2 2 3 9 13 19 19 15 6 1 90 Calcutta 1 3 3 5 12 18 25 26 19 8 i' 2 123 Dacca . 1 3 5 9 15 19 22 22 16 6 2 1 121 Chittagong . 1 2 5 6 13 21 25 23 17 9 4 1 127 Silchar . 2 6 13 17 20 22 24 24 18 8 2 1 157 Cherrapunji . 2 4 10 20 25 26 29 28 24 12 3 1 184 Gauhati 2 4 9 14 19 20 20 18 13 6 1 1 127 Sibsagar 7 9 14 19 20 21 23 21 18 11 3 3 169 Cuttack 1 2 2 3 9 16 23 23 19 9 3 1 111 Sambalpur . 1 2 2 2 5 14 24 24 17 4 1 1 97 Saugor . 2 1 ... 1 2 11 19 17 13 3 1 70 Jubbulpore . 2 2 1 2 3 12 19 18 11 3 1 1 75 RAINFALL 75 MONTHLY AVERAGE OF RAINY DAYS AT EIGHTY STATIONS IN INDIA, CEYLON, AND BURMAH Continued RAINY DAYS. Stations. 1 " fS March. i sA o5 -5 "3 1-5 s>b o Jz; I 1 Pachmarhi . 2 2 1 2 4 14 25 23 17 5 1 2 98 Nagpur 2 1 2 2 4 15 21 18 15 4 1 1 86 Amraoti 1 1 1 1 3 14 18 15 13 5 1 2 74 Dhulia . ... 1 8 14 11 13 5 1 1 54 Poona . ... 1 "2 2 13 21 20 15 8 2 1 85 Bombay * . 1 1 1 ... 2 20 29 26 21 7 1 ... 109 Matheran 2 25 31 30 27 10 2 128 Mahableshwar i 1 "2 4 27 31 30 28 12 4 1 141 Karwar . ... 1 5 26 30 26 20 13 4 125 Belgaum 1 3 6 7 21 28 25 22 13 5 3 134 Jalna 1 1 2 13 16 13 14 5 3 2 70 Sholapur i' 'i' 1 3 4 14 16 17 14 8 3 2 84 Secunderabad i ... 3 4 5 12 17 18 15 8 4 2 88 Bellary 1 3 5 6 8 8 9 9 5 1 55 Bangalore T 2 3 10 13 14 17 13 13 7 3 96 Wellington . 3 2 6 7 14 15 14 16 13 18 16 10 134 Bimlipatam . Masulipatam . 1 1 1 1 1 1 1 1 4 4 5 8 6 15 9 16 11 16 7 12 4 6 1 2 51 83 Rajaraundry . 1 1 1 5 7 11 11 9 7 3 56 Cuddapah ... 1 1 3 6 4 10 8 7 5 i' 46 Madras . 3 1 1 2 10 14 15 12 14 14 9 85 Trichinopoly . 1 1 1 5 3 3 7 8 11 9 6 55 Cuddalore 3 ... 1 1 2 6 9 11 9 11 14 10 77 Madura 1 ... 1 2 5 3 3 7 8 14 9 5 58 Mangalore Cochin . 1 2 "5 3 8 8 17 27 27 30 27 28 22 22 19 15 19 7 12 o 5 142 164 Trincomalee . 10 3 4 4 6 4 2 9 7 16 19 20 104 Colombo 6 4 10 13 20 17 12 13 14 21 18 13 161 Galle . 12 8 11 14 21 22 20 21 20 22 18 16 205 Rangoon ... ... 3 19 26 29 25 27 12 10 151 Moulmein 7 18 28 29 28 27 15 5 1 158 Thyet Myo . 2 11 22 27 23 20 11 3 1 120 Tounghoo 1 4 12 24 26 27 22 14 4 1 135 Akyab . 1 "i 2 14 28 30 29 22 12 4 2 145 Average Heaviness of the Rainfall. If the quantities of rainfall at the several stations in the first of the foregoing tables be divided by the number of rainy days in the second, it will be found that the average fall of each rainy day is between six and seven tenths of an inch on the plains of Bengal and the North-west Provinces, and in parts of Central India and the Central Provinces. In the wetter 7 6 ELEMENTS OF CLIMATE, ETC. parts of Assam, on the coasts of Arakaii and Tenasserim, at Pachmarhi, and on the west coast and Western Ghats of India, it is higher; at Cherrapunji as much as 2 '6 inches, and at Akyab and Moulmein, Mahableshwar and Matheran, from 1 to 2 inches on each rainy day. But in the Deccan and Mysore, and also in the Indus valley, it is only between three and four tenths, and at Leh but one-tenth of an inch. This last is about the same as the general average of the British Isles. It results, therefore, that, day for day, rain is from three to seven times as heavy on the plains of India as in Western Europe, and that the common pro- verbial expression, " it never rains but it pours," is a literal description of the ordinary rainfall of India. This is as true of such places as Jacobabad and Mooltan, where there are no more than about fourteen rainy days in the year, as of Point de Galle, where it rains on 205 days on an average. In consequence of this character of Indian, in common with tropical, rainfall generally, it is less penetrating in pro- portion to its quantity than in countries where much of it falls in a state of fine division, allowing time for its absorp- tion by the ground. Instead of feeding perennial springs, and nourishing an absorbent cushion of green herbage, the greater part flows off the surface and fills the dry beds of drains and watercourses with temporary torrents. In uncul- tivated tracts, where jungle fires have destroyed the withered grass and bushy undergrowth, and have laid bare the soil and hardened its surface, this action is greatly enhanced ; and while all perennial water supplies which depend on the absorbed rain are either greatly reduced or altogether sup- pressed, a rainfall which, if husbanded by nature and art, would suffice for the agriculture and domestic requirements of the population, is thrown into the nullahs and rivers, and not only is wasted and lost for any useful purpose, but by producing floods, becomes an agent of destruction. Under any circumstances, the character of the rainfall is hardly compatible with its economical storage and expenditure in any high degree ; and much more, therefore, than in tern- RAINFALL 77 perate regions is it incumbent on us to safeguard such pro- vident arrangements as nature has furnished for the purpose. Excessive Rainfalls. Seeing, then, that on the plains of Northern and Central India the average fall of each rainy day is not less than six or seven tenths of an inch, such enormous falls as those referred to on page 64 become less surprising as an exceptional occurrence. The following are a few instances of remarkable quantities recorded within 24 hours, in the years 1875-1885, but by no means all of those exceeding 12 inches, which would be much more numerous. A more comprehensive list will be given in the second part of this work. Cherrapunji, Khasi Hills . . 40 '8 inches. 14th June 1876. Purneah, Northern Bengal . . 35 '0 13th September 1879. Nagina, Rohilkhand . . . 32 '4 18th September 1880. Danipur, ,, . . . 30 '4 Najibabad, ,, . . . 28 '5 Hard war, Dehra Doon . . . 19 '5 Rewah, Central India . . . 30 '4 Sutna ,, . . . 22-1 Chandbally, Orissa . . . 20 '3 Delhi, Punjab . . . . 19 "5 Gurgaon ,, .... 10'5 ,, Allahabad, N.W. Provinces . . 15.5 ,, 29th July 1875. Pilibhit, Rohilkhand . . . 13 '6 ,, 2d August 1879. Mozufferpore, Bengal . . . 12'5 ,, 17th June 1883. 16th June 1882. 16th September 1879. 9th September 1875. Although, in their extreme intensity, such falls are gene- rally local, they are only part of a less copious but still heavy fall, often extending over thousands of square miles, and lasting over two or three days. Such was the case on the 18th September 1880 when, in addition to the torrential rainfall in the Bijnore district (in which are three of the four stations enumerated above), a quantity, exceeding 10 inches in two days, was registered over a large part of the adjoining districts of Saharunpore, MozufTernagar, Meerut, Moradabad, Pilibhit, and Kumaon ; and also on the 13th September 1879, when 5, 6, 9 '5, and 12 inches, respectively, were registered in adjacent districts around Purneah. There would seem also to be a tendency to a repetition of heavy falls in about the same part of the country in the same season. Thus, only 78 ELEMENTS OF CLIMATE, ETC. nine days before the above enormous fall of 3 5 inches took place at Purneah, one of 11*5 inches had been registered at the same station, and only five days previously one of 5 inches. In 1875 the excessive fall of September in the eastern districts of the Punjab had been preceded in August by very heavy falls at Sialkot, Jullunder, and Lahore, of 11 inches in two days, 18*8 inches in three days, and 11 '7 inches in three days respectively. Some other similar cases have been recorded in the Annual Reports on the Meteorology of India. These excessive falls are always the result of cyclonic storms, not indeed such as are of destructive violence as regards the wind, but the long-lived cyclonic storms which occur chiefly in the rains, storms in which the barometer is not greatly depressed, and only recognisable in their true character when the barometric readings and the winds are laid down on charts in the manner already described. Another noteworthy point is that they have frequently occurred in years of partial drought, as if, in such seasons, the whole energy of rain formation were concentrated in the storms, to the deprivation of parts of the country not reached by them. Thus in 1875, although in the Punjab it was one of the wettest years on record, the rainfall was very deficient in Bengal and all over the southern half of the peninsula. 1876 was the year that produced the last great peninsular famine, and in both 1880 and 1883 there was a prolonged suspension of the rains over all North- Western and Central India and a part of the peninsula. Annual Variability. It has already been stated that the variability of the rainfall, that is to say, the degree in which that of any given year is liable to deviate from the local average, either in excess or defect, is much greater in some provinces than in others. This liability is propor- tionally greatest in the driest part of the country, viz. in Sind, and is least in Assam, which is one of the wettest. In the former, the mean annual deviation is as much as 37 per cent of the general average, in the latter only 5 per cent, although in certain years these amounts are greatly RAINFALL 79 exceeded. Next to Sind, the North-west Provinces and Oudh are subject to the greatest vicissitudes, the mean annual deviation being 23 per cent; and within the last 22 years it has been as much as 47 per cent above and 39 per cent below the average. Berar and Khandesh, Hyderabad, Behar, and in a somewhat less degree, Guzerat, the Deccan, Mysore, and the Carnatic, also have a comparatively pre- carious rainfall. But all the older provinces of Burmah, viz. Arakan, Pegu, and Tenasserim, also Lower Bengal, Chutia Nagpur, arid the Central Provinces, rank next to Assam in point of regularity. Although certain portions may sometimes suffer from drought, there has not been any extensive and disastrous famine in any of them during the present century. In no single year has the rainfall been either deficient or excessive over the whole of India ; and a recent critical examination of its vicissitudes during the last 22 years has brought to light some curious relations in the way in which certain provinces are prone to vary alike, while others vary in the opposite direction. Thus the Northern Provinces of the peninsula, Khandesh, Berar, the Central Provinces, and Orissa, usually all share either an abundant or a deficient rainfall ; and, what was hardly to be expected, Assam and Eastern Bengal vary in the like manner with Orissa, while Lower Bengal, which lies between them, varies in the opposite direction. Again, when the rainfall is excessive in Bengal, it is usually light and deficient on the Bombay side and in Hyderabad. Many of these peculiarities admit of, at all events, a probable explanation ; they seem to depend partly on the unseasonable prevalence of dry land winds in Western India, partly on the course taken by the cyclonic storms of the rainy season. But others are at present quite inexplicable, such, for instance, as the fact that, in five years out of six, the rainfall of Arakan is deficient or exces- sive when that of Sind varies in the opposite direction. On a general average, about two-thirds of India varies in one way, and one-third in its opposite. But in 1864, 1868, and 1873 the rainfall was deficient, and in 1878 excessive, over a very much larger proportional area. The average rain- 8o ELEMENTS OF CLIMATE, ETC. fall of India being taken, as stated on page 69, at 42 inches, the greatest deficiency during the last 22 years was about 6*6 inches, viz. in 1868, and the greatest 6 - 3 inches in 1878. The Question of Cyclical Variation. Much has been written of late years of the supposed variation of rainfall in a cycle of 1 1 years, coincidentally with the now well-known variation of sun-spots. Without venturing to pronounce any opinion on the question whether any such variation exists in the rainfall of the globe generally, it is certain that that of India, as a whole, has not displayed it during the last 2 2 years. But while this is true of the country as a whole, the rainfall of the Carnatic, which chiefly occurs at a later season than the heavy rains of other parts of India, certainly showed a somewhat striking fluctuation in the 11 years, 1864-1874, and again, though less regular, in the next 11 years, 1875- 1885. It is still a matter of opinion whether this was fortuitous or otherwise, a question which can be decided only by further experience. According to such evidence as is before us, there would appear to be a mean difference of not less than 14 inches between the years of lightest and heaviest rainfall in this cycle, the general average of the province being 36 inches. Some evidence has been adduced by Mr Hill of a similar fluctuation in the winter (but not the heavier summer) rain- fall of Northern India. Droughts. The interest of this last observation lies in the fact that a heavy winter and spring rainfall takes the form of heavy snow on the Himalaya, and, as far as our limited experience goes, seasons in which the accumulation of snow has been excessive, have always been followed by a more or less prolonged suspension of the summer rainfall, as a rule in Northern India, but sometimes in some portion of the peninsula. It must, however, be added that there have also been serious droughts which, as far as our information goes, have not been preceded by any unusual snowfall on the mountains in the earlier months of the year, so that, even if this be a cause of drought, which many considerations STORMS 81 render probable, it is not the sole determining agency. The immediate cause of drought is the unseasonable persistence of dry land winds to the exclusion of the rain-bearing wind of the summer monsoon. Finally, it appears from the list of droughts furnished in the Eeport of Indian Famine Commissioners, that on no less than five occasions in the course of the last century, a serious drought in the peninsula has been followed by one in Northern India in the next year. The following is a list of the cases : /1782. Drought in Bombay and Madras. I 1783. Drought in Upper India. [ 1802. Drought in South Hyderabad and Deccan. \ 1803. Drought in North-western Provinces and Central India. f 1823. Drought in Madras. 1 1824. Drought in Bombay, chiefly Guzerat and North Deccan. f 1832. Drought in northern districts of Madras, etc. \ 1833. Drought in north part of Bombay, Kajputana, Punjab, and North- west Provinces. ( 1876. Drought in all parts of Madras, Deccan, etc. ( 1877. Drought in North-west Provinces, Central Provinces, and Punjab. STORMS Thunder - Storms and Dust -Storms. Any violent wind, accompanied with rain, hail, or snow, may be termed a storm, and in India the use of the term is somewhat further extended, since it is often applied to dust- storms- with little or no rain. In these storms, however, which are common in the hot weather in the Punjab and North- western India, the coolness of the wind, and that of the atmosphere after the storm is over, is hardly to be accounted for otherwise than by supposing that rain is always formed by the cloud overhead, but being re-Evapo- rated before it reaches the earth, the storm is often appar- ently rainless. A complete transition may be traced between dust-storms of this character and the nor'- westers of Bengal r which are accompanied by heavy rain. Even in the dry climate of the Eastern Punjab and North-west Provinces, a G 82 ELEMENTS OF CLIMATE, ETC. slight sprinkling of rain often follows the dust-storm, and in the damper atmosphere of Bengal the dust-storm is, as a general rule, only the first stage of a nor'-wester, and is followed by heavy rain and sometimes hail. Atmospheric disturbances of this class are individually local, though they generally occur simultaneously at different places over a large area of country. On their approach, the barometer is much disturbed, but the action is spasmodic, consisting of a rapid rise, followed by much irregular oscillation, and is very different from that which portends the approach of a cyclone, which is a steady continuous fall. For a day or so before a dust-storm or nor'-wester the barometer is, as a rule, somewhat lower than the average of the time of year ; and indeed they are most common in the north and east of a region of barometric depression. But the depression of the barometer is not very great, and although the wind squalls are sometimes violent at the beginning of the storm, the whole character of the weather is so unlike that which betokens the approach of a cyclone, that a very little ex- perience suffices to enable any one to distinguish between them. Both dust-storms and nor'-westers almost always occur in the afternoon or evening, towards the end of a close warm day, either with little wind or with a damp wind from the sea. But, in the latter case, this lulls some time before the storm approaches, and for an hour or more the air is calm, until violent gusts of cool air blow out in advance of the storm, raising clouds of dust. Sometimes these gusts blow with great force, blowing down trees or breaking their branches ; and on many occasions pressures of from 30 to 50 Ibs. to the square foot have been registered by the Calcutta pressure-gauge. These storms are of precisely the same nature as the thunder-storms of an European summer, except that they are more violent. The accompanying diagram (Fig. 7) of a summer squall, taken from a treatise by Dr. Moller, 1 exactly represents the leading features of an Indian nor'- 1 Meteorologische Zeitschrift, 1884, p. 238. STORMS wester. The strong gusts that blow out in front of the storm, a, a, and which constitute the dust-storm, are air dragged down by the falling rain, and their coolness is due to the partial evaporation of the water drops; therefore to the same cause as that of the air blown through a tattie. The conditions for the production of a storm of this character are that the lower layer of the atmosphere is warm and damp ; that immediately above it, perhaps at a height of 4000 or 5000 feet, dry and cooler; and the wi* FIG. 7. SECTIONAL DIAGRAM OF A SQUALL. AFTER MOLLER. cloud is formed by the ascent of the former, penetrating the latter and becoming cooled by its ascent, the reverse of the process noticed on page 49. Sometimes, when the storm is seen from a distance, this process may be watched, the motion of the air that streams in under the cloud bank in front of the storm (indicated by the short arrows, ~b, in the figure) being rendered visible by the little cloud* tufts that appear in it before it reaches the main body of the storm cloud, and their horizontal motion being gradually arrested, are then carried up and absorbed in the cloud bank above. The storm almost always advances from the west or north- west ; whence its name. This movement is due to the upper air current, which is the same as the west land wind of the Upper Provinces, blowing above the damp air that comes originally from the sea. Of the existence of this upper current ocular demonstration is afforded by the fact that the upper part of the storm cloud (and, in fine weather, 84 ELEMENTS OF CLIMATE, ETC. of the cumulus cloud, of which the storm cloud is only a more developed form) is almost always inclined eastward, and also by the false cirrus (see page 59 and c, c, Fig. 7) which advances at a great height as a sheet of cirro-stratus, with a hard, well-defined margin in advance of the body of the storm cloud. Looking out westward across the open expanse of the Calcutta maidan, I have often watched the advance of these storms, and long before the publication of Dr. Holler's description and figure, had arrived at the con- clusion that the process of their formation is exactly such as is represented in his diagram. Storms of this class are short-lived, and are not an in- dication of any general or important disturbance of the atmosphere, however great their local violence. They are local eddies between opposite air currents, the one above, the other below, and do not affect the general movements of the atmosphere to any great distance from the storm region. Cyclones and Cyclonic Storms. Very different is the other more lasting and sometimes more destructive class of storms, now known as cyclones and cyclonic storms. I use these two terms alternatively, as recognising the fact that, while the disturbances respectively denoted by them are essentially identical in constitution and mode of origin, there is a marked difference in the degree of their violence and destructiveness. The term cyclone is so familiarly as- sociated in men's minds with winds of destructive violence, and frequently with the still more disastrous accompani- ment of the storm- wave, that it will be convenient in a work addressed to practical men, to whom such differences are all important, to restrict its application to storms of this kind, and to call the milder though not less extensive dis- turbances which are experienced in the summer monsoon, and those also which bring about the winter rains of Nor- thern India, by its modified derivative, cyclonic storms. The essential features common to both have already been described in the section on the winds (page 44). Over a limited region, more or less circular or oval in form, the STORMS 85 atmospheric pressure, as indicated by the low barometer, is lower than anywhere for a great distance around. Into this barometric depression the air pours from all sides, following courses which, as we have seen, are neither directly towards the centre, nor, as was formerly believed, in concentric circles around it, but intermediate between the two, in converging spirals. In the northern hemisphere these turn around the centre in a left-handed direction as in Fig. 8, 1 against the hands of the clock, or in the direction in which one turns an ordinary screw in order to extract it. In the southern hemisphere they are right-handed, or move with the clock hands, as in Fig. 9. The strength of the winds, as has been already explained, varies with the steep- ness of the barometric gradient, or the rapidity with which FIG. 8. TROPICAL CYCLONE. FIG. 9. TROPICAL CYCLONE. NORTHERN HEMISPHERE. SOUTHERN HEMISPHERE. the pressure falls from the outside towards the eye and centre of the storm. Hence, when such a storm is ap- proaching, the winds become fiercer, the more rapidly the barometer is falling. In the centre of the storm the air is calm or the winds are light and variable, notwithstanding that the barometer is then at its lowest. But immediately around they blow with their greatest violence, and become by degrees more moderate as the distance from the centre increases. In the centre of a cyclone the barometer often marks an inch, and sometimes as much as two inches, below its readings at a distance of a few hundreds of miles. In a cyclonic storm the depression does not exceed a few 1 The spirals in Fig. 8 represent the course of the winds in a cyclone in the north of the Bay of Bengal. Those in Fig. 9 the winds of a cyclone in the same latitude in the southern hemisphere. 86 ELEMENTS OF CLIMATE, ETC. tenths of an inch, and although at sea the winds are locally strong and squally, sometimes indeed attaining to hurricane force to east and south of the centre, yet on land they seldom blow with much violence : the central barometric depres- sion is larger, and the calm area less denned : but the rain accompanying them is often very heavy, and it is in such storms that those torrential falls take place, some examples of which have been quoted on a previous page. Another distinction of much practical importance be- tween cyclones and cyclonic storms is that cyclones, not- withstanding their fierceness, are, as a rule, less lasting than the milder form of storm, when they have reached the land ; and are often broken up and dissipated when they encounter even a low range of hills interposed in their path. Such was the case, for instance, with the cyclone which, on the 1st November 1876, passed over Backergunge in Eastern Bengal, submerging the large islands at the mouth of the Megna, and sweeping some 100,000 human beings to de- struction ; but was dispersed and annihilated on meeting the Tipperah hills, which are not more than about 2000 feet high. And cyclones which cross the Coromandel coast generally break up and disappear on reaching the Eastern Ghats, which are also little, if at all, over 2000 feet in height. Even if not extinguished, they are much reduced in force when crossing the plateau of the interior, and become reinvigorated only on reaching the Arabian Sea, which they may then traverse for a distance of many hun- dreds of miles beyond. On the other hand, cyclonic storms which are formed at the head of the bay, or over Bengal or Chutia Nagpur during the rains, traverse the hilly country of Chutia Nagpur and the Central Provinces without any great loss of energy, and sometimes pass right across Central India, Guzerat and Cutch, and out to the Arabian Sea, off the coast of Baluchistan. Others again travel to the north- west, generally taking four or five days in the transit, and sometimes lasting much longer. In September 1884, a storm of this kind, formed in the west of the bay about the STORMS 87 16th of the month, crossed the Ganjam coast on the 20th, then travelling slowly on a north-west course, by the 24th had passed the Satpuras, and the centre lay between Jub- bulpore and Saugor ; in two more days it had reached a position between Jhansi and Agra, after which its farther progress was stayed, and it remained virtually stationary till the end of the month, and indeed till the 1st or 2d October, giving very heavy rain to Eohilkund and the Meerut division of the North-west Provinces. The life of this storm therefore exceeded a fortnight. One further point of distinction between cyclones and cyclonic storms is that the former are restricted to the intervals between the monsoons or to their beginning and ending. The latter occur in frequent succession all through the summer monsoon, and, as above mentioned, also in the winter months in Northern India, bringing the winter rainfall. Their movements are, however, quite different at the two seasons. The storms of the late spring and summer almost invariably move on a course between north and west. Some few autumnal storms have taken a north-north-east course up the Bay of Bengal, and those which reach the shores of Bengal often curve to north-east in the final part of their course before breaking up. But the cyclonic storms of the winter and early spring months, that make their first appearance in North-western India, if not stationary, always move eastward, and sometimes, though rarely, a little to the south of east. Cyclones of great violence are comparatively rare. In the Bay of Bengal they are certainly not more numerous than about two in the year on an average, perhaps less ; and in the Arabian Sea are even less frequent, as far as our records go. Whereas, during the summer monsoon, cyclonic storms sometimes succeed each other at the rate of four or five in the month. Cyclones, as experienced on the coasts of Madras, Bengal, and Burmah, always come from the sea ; probably because it is only at sea, where there is a wide expanse of open water, that circumstances admit of their develop- ing a high degree of severity. Otherwise there is no reason, ELEMENTS OF CLIMATE, ETC. why a cyclonic storm originally formed over the land and passing out to sea should not develop into a cyclone. And indeed, more than one instance is on record, in which a cyclone formed in the bay, has reached the Corornandel coast, and in its passage across the peninsula has been so far mitigated that it could only be described as a cyclonic storm ; but after crossing to the Arabian Sea it has again developed its former intensity, and as an unquestionable cyclone has been encountered by vessels making the passage between Aden and Bombay. Such was the case with the cyclone that passed over Madras at the end of November 1885. There is, then, no fundamental distinction between cyclones and cyclonic storms. Mr. Eliot, who has paid much attention to this subject, chiefly in connection with the requirements of the Bengal storm-warning system, has distinguished them simply as large and small storms, the former being characteristic of the periods of transition between the monsoons, the latter of the monsoon season. And from his point of view this terminology is unobjectionable. But inasmuch as the latter class of storms, when laid down on the Indian weather charts, are found to cover quite as large an area as the more severe cyclones, these terms might be open to misconception, and those here proposed seem therefore preferable. These general remarks on cyclones and cyclonic storms may suffice for this part of the work. In the second part they will be described more in detail, with respect to the regions affected by them, to their relative frequency in different months, the circumstances and place of their origin, and the courses they generally follow at each season, and finally the antecedent conditions which are premonitory of their occurrence. Tornadoes. It remains to notice a class of storms which resemble cyclones in the vortical movement of their winds, and which, within the narrow limits of the areas affected by them, surpass these and all other storms in violence and consequent destructiveness ; but in the circumstances of STORMS 89 their origin, and in their brief duration, are more nearly akin to ordinary thunder-storms. Tornadoes or whirlwinds, as these are termed, are comparatively rare in India, there being only four or five recorded instances known to me, and all these were in Bengal. Possibly some of the dust-storms of North-western India, described by Baddeley, 1 may be of the same character ; they have apparently a similar vortical movement, but not the same intensity, and at present it is at least uncertain how far the conditions of their formation are identical with those of tornadoes. The few recorded tornadoes have all occurred in the spring months, between March and May. They have varied from about 200 yards to half a mile in diameter, and while the track of their passage has been marked by com- plete devastation, by trees torn up by the roots, or stripped of their largest branches, by houses unroofed and blown into ruins, and by human beings, cattle, and even more ponderous objects being carried into the air and transported to con- siderable distances, immediately beyond its limits everything has remained undisturbed and uninjured. And the track of destruction has not been continuous. It appears that whirlwinds, at least such as are of small diameter, only reach the ground intermittently, and in the intervals pass overhead without much affecting objects on and near the ground. There is some evidence that such was the case in a tornado that originated over the Jamuna river in Eastern Bengal on the 26th March 1875, and also in the more recent tornado that passed over Dacca on the 7th April 1888, of which an excellent description has been published by Dr. Crombie in the Journal of the Asiatic Society of Bengal? It is hardly possible to draw further general conclusions from these few instances, and it may therefore be useful to consider such as are afforded by storms of the same kind in the United States of America, where they are of frequent occurrence, and have been made the subject of careful study. 1 Phil, Magazine, 1850, and Journal As. Soe. Bengal, 1852. 2 Vol. IviL, 1888, Part ii. No. 2. 90 ELEMENTS OF CLIMATE, ETC. In the single year 1884, 171 tornadoes were recorded and reported in the States of the Union, and have been discussed in a Memoir by Lieut. J. P. Finlay. The most important conclusions of a general character, apart from questions of geographical and seasonal distribution, etc., which have chiefly a local interest, may be sum- marised as follows : Tornadoes occur when a barometric depression is traversing the States, usually one of more or less oval or elongated form, having its major axis north and south, and always in a definite part of the depression. From the charts which accompany the Memoir, this appears to be the south or south-east of the depression. They occur to the south and east of a region where there are high contrasts of temperature and also high contrasts of atmospheric humidity, which respectively accompany cool northerly and warm southerly winds. Now, substituting " dry and north-west " for " northerly " winds, most of these conditions are probably fulfilled in the case of the Bengal tornadoes. Such was certainly the case in the only instance for which the actual data are before me, viz. that of the recent Dacca storm above quoted. This storm occurred in the evening, and the Simla weather chart shows that, on the morning of the day, there was a barometric depression in Northern Behar, to the north-west of Dacca. On the morning of the previous day a cyclonic storm had appeared in the Punjab, and it may be inferred from the subsequent changes of the barometer and winds, that this had passed to the eastward over the Himalaya, and that the depression in Northern Behar was a portion of its southern limb. In any case, on the forenoon of the 7th, winds were northerly over North-western India, north-west to west down the Gangetic plain, and between south and south-west in Bengal. The north-west winds down the Gangetic plain were very dry, the relative humidity at 8 A.M. varying between 20 and 36 per cent of saturation at different stations, while the south-west winds in Bengal had a humidity in most cases of over 80 per cent. The difference of temperatures of the two winds near the ground surface STORMS 91 was not indeed very great, and in the afternoon evanescent, but it is well known that in the dry atmosphere of the spring months in Upper India the vertical decrease of temperature is very rapid in the daytime, much more so than in the damp atmosphere of Bengal ; and it is probable that at the elevation of the clouds, in which tornadoes originate, there was a large difference of temperature. It seems very probable, then, that the circumstances under which tornadoes occur in Bengal are similar, in all essential respects, to those which hold good in the United States of America as above denned. No tornado has been traced in Bengal for a greater distance than a few miles. In America, according to Lieut. Finlay, the length of their path has varied from 2 miles to 130 miles, and their rate of progression from 15 to 80 miles an hour. Their average duration at any given spot was only 45 seconds. In the great majority of instances, the direction of the wind's revolution is the same as in cyclones. The exceptional reported instances in which it was said to be " probably " anticyclonic or clockwise, are only 6 per cent of the whole. PAET II CLIMATES AND WEATHEE OF INDIA IN EELATION TO HEALTH AND INDUSTEY PAKT II CLIMATES AND WEATHEE OF INDIA THE CLIMATES OF INDIA Heterogeneity of Indian Climates. In the heading of this section and in the title of the work, I use the word climate in the plural, in order to emphasise the fact that in this respect, and consequently in all that depends on climate, the different parts of India exhibit very great diversity. Northern or extra-tropical India alone, in its most easterly and most westerly provinces, in Assam on the one hand and in Sind on the other, presents us with the greatest possible contrast of dampness and dryness, a con- trast greater than that of the British Isles and Egypt ; and when, further, we compare the most northerly province, the Punjab, with the most southerly, such as Travancore or Tenasserim, we have in the former a continental climate of the most pronounced character, extreme summer heat alter- nating with winter cold that sometimes sinks to the freezing- point, and in the latter that almost unvarying warmth in conjunction with a uniformly moist atmosphere, that is especially characteristic of the shores of a tropical sea. To speak, then, of the climate of India, as we might speak of the climate of Ireland, as if such expression denoted certain definite conditions of heat and moisture, varying only within moderate limits, and nearly uniform in the several provinces, would be as misleading as if we were to speak of its in- habitants in terms implying that they are a homogeneous race, alike in ethnic and social characters, culture and belief. 96 CLIMATES AND WEATHER OF INDIA That they are Indian natives certainly implies that they are human beings, denizens of some part or other of Her Majesty's Indian Empire, and little indeed beyond. And in like manner the term Indian climate means little more than that, on the general average of the year, the sun is higher in the heavens, and the temperature some degrees greater, than in Europe or other lands of the temperate zone. In all else we must be prepared to find the greatest variation, and even the greatest contrasts. In respect of this one feature, the mean temperature, notwithstanding that one-half of the area of India lies without the tropical zone, the country as a whole is remark- able as being one of the warmest regions of the earth's sur- face. It has long been known that what is termed the thermal equator (the line of greatest mean heat of the globe), does not follow the course of the equinoctial line, where, on the average of the year, the sun's action is most direct and intense, but, in the longitude of India, bends northward, running up through Ceylon and the peninsula to beyond the tropic and, passing through Sind, crosses thence to the Arabian peninsula ; furthermore, that where it traverses the low plain of Southern India is one of the hottest portions of this line. But it has been shown in the discussion of the temperature in the earlier portion of this work, and it will presently be explained more in detail, that it is not in Southern India that the most intense heat is experienced in the summer season. Erom May to August the heat of North-western India exceeds that of any part of the peninsula at any time of the year; and probably nowhere on the earth's surface has the -thermometer, when properly shaded in the manner explained on page 6, registered higher temperatures than stand recorded in the dry climate of Upper Sind and western Eajputana. HILL STATIONS Before proceeding to describe the varieties of climate to be met with on the plains and plateaux of India, I will 'CLIMATES OF HILL STATIONS 97 notice those of the hill stations, many of which lie around its borders, and which, as the health resorts of Europeans in the military and civil services of India, and as the permanent abodes of many of those who have become domiciled in the country, lay claim to our especial interest from a sanitary point of view. The climates of the Himalayan stations have indeed been described in a special work on the subject, written for the guidance of medical officers by Dr. F. N. Macnamara, and as that work, in addition to the statistics of the climate of each station, gives much other information relative to the prevalent diseases and their treatment, to all such officers it will remain a valued and familiar guide-book, which the present in no way pretends to supersede. But the latest meteorological data available to its author were those of 1876, only 2 years after the Meteorological De- partment had been established on its present footing ; and the records of 10 additional years, which have now been published, enable me to give much fuller and more trust- worthy data than were then extant, for illustrating the climatic features of those stations that are provided with observatories. To these and to the similar stations in the peninsula and Ceylon, my own remarks will be restricted. They comprise the following : Quetta. Ranikhet. Mount Abu. Leh Darjiling. Wellington. Murree. Shillong. Ootacamund. Simla. Pachmarhi. Newara Eliya. Chakrata. Chikalda. Quetta. The climate of this important military station, within a few miles of the Afghan frontier, has become known only within the last few years, thanks to the excellent observatory established and superintended by Dr. Fullerton, the residency surgeon. In many respects it is unlike that of any other high-level station, but may be taken as repre- sentative of that of the greater part of the western frontier, and especially of Northern Baluchistan. The part of Balu- chistan in which Quetta is situated is a tableland, extend- ing from the summit of the Bolan range, and declining H 98 CLIMATES AND WEATHER OF INDIA gently towards the west. At the Kotal or summit of the Bolan pass, it is 6500 feet above the sea, and consists of broad loamy plains, almost bare of vegetation, intersected and sur- rounded by numerous ridges of bare rocky hills, some peaks of which rise to 11,000 feet above the sea. Quetta is built at an elevation of 5500 feet on one of these valley plains, forming part of the drainage of the Lora river, running north and south for about 30 miles, and varying from 3-^ to 5 miles in breadth. To the north of Quetta it bends westwards. The station is thus surrounded by hills, except to the south and north-west, those on the east being about 2 miles distant, and those on the west little more than 1^- miles. To the north-west the vista is open, nearly from north to west, and 30 miles off is seen the Khojak range, at the near foot of which is the frontier military post Pishin. The whole character of the scenery around Quetta testi- fies to the extreme dryness of its climate. The naked hill scarps, on which every rocky outcrop stands conspicuous as in a geological diagram, the absence of trees and for the most part of bushy scrub on the pebble-strown slopes at their foot, and the grassless flats of loam that occupy the intervals between the ranges, dotted here and there with mounds of earth excavated in the construction of karezes or underground water channels, bear witness, even in the absence of any meteorological record, to the scantiness of the rainfall and the aridity of the prevailing westerly winds. The summer monsoon rains, if indeed they reach Quetta at all, do so only to the extent of a rare shower ; the total rainfall of July and August amounting, on an average, to less than an inch and a half, distributed over four rainy days, and in some years failing completely. Indeed the only rain of importance is that of the first four months of the year, when, on an average, it rains or snows on one day in four ; and the total average rainfall of the year is barely 10 inches. Even this small amount is very uncertain and precarious. In 1879 it was little more than 4 inches, and in the following year barely 6 inches, whereas in 1885 it exceeded 2 1 inches ; and the number of rainy days in CLIMATES OF HILL STATIONS 99 the year has varied from 18 only to 73. January and February are the dampest months, but even then the mean humidity of the air does not exceed two-thirds of saturation, and the driest is September, when it does not much exceed one-third. All through the summer monsoon the average humidity is below 50 per cent. As is invariably the case in dry regions, the ordinary range of temperature is very great, both in the course of the 24 hours and also between winter and summer. The diurnal range is least in February, when it amounts to 21, and is greatest in October, when it averages not less than 35 between the early morning and the afternoon. Although the station is not less than 5500 feet above the sea, the mean daily maximum of July and August is 91 or 92 in the shade, and a reading of 103 has been registered within the last 6 years. The mean temperature of these months, however, is only 77 and 75, and as a rule in the early morning, the thermometer, protected from radiation, sinks to between 63 and 65, and often lower. In January the plateau is swept by storms of snow and sleet with bitterly cold winds, alternating with frosts, which, however, speedily disappear under an unclouded sun, and from November to February the shaded thermometer, on most days before sun- rise, sinks to a temperature below the freezing-point. The nights are always cool throughout the year. As above remarked, the winds are almost always more or less westerly, most commonly north-west, and this is the case throughout the season of the Indian summer monsoon, when in the greater part of Northern India easterly winds prevail. Leh. Situated 4 farther north than Quetta, and 6000 feet higher; this station, in the leading character- istics of its climate, represents that of Western Tibet. Its observatory, 11,500 feet above the sea, is the most elevated in Asia, and has now furnished a register for 12 years. Leh, the chief town of the province of Ladak, is situated in the upper valley of the Indus, which is here from 6 to 8 miles across for a distance of 40 or 50 miles, and shut in on both sides by .portions of the Tibetan plateau, averaging ioo CLIMATES AND WEATHER OF INDIA 16,000 feet above sea-level, and bearing some of the loftiest ranges in the Himalaya. The flats and terraces along the banks of the river are all cultivated, and dotted with numerous villages ; but for a mile on each side, between the fields and the foot of the hills, is a waste of sand, gravel and large boulders, similar to that which fringes the much lower hill ranges of Baluchistan, but on a larger scale. The town of Leh nestles under the hills north of the valley, at a dis- tance of some 4 miles from the river, up a long, gentle, gravelly slope. The atmosphere of the valley is remarkably clear and transparent, and the heat of the sun very great. There is generally a difference of more than 60 between the reading of the exposed sun thermometer in vacuo and the air tem- perature in the shade, and this difference has occasionally exceeded 90. It has been mentioned on a former page that Dr. Cayley succeeded in making water boil by simply exposing it to the sun in a small bottle blackened on the outside, and shielded from the air by inserting it in a larger phial of transparent glass. Owing to the diminished pres- sure of the atmosphere at the elevation of Leh, this would, however, take place at 191 or 192, or about 20 below the normal boiling-point at the sea-level. The mean annual temperature of Leh is 40, that of the coldest months (January and February) only 18 and 19; but it rises rapidly from February to July, in which month it reaches 62, with a mean diurnal maximum of 80, both in that month and August, and an average difference of 29 or 30 between the early morning and afternoon. The mean highest temperature of the year is 90, varying between 84 and 93 in the last 12 years. On the other hand, in the winter, the minimum thermometer falls occasionally below Fahrenheit, and in 1878 reached as low as 17 below zero. The extreme range of recorded temperatures is, therefore, not less than 110. The air is as dry as at Quetta, and rather more uniformly so. In the driest month (June) the mean humidity is but 3 7 per cent of saturation, and in the dampest months (January CLIMATES OF HILL STATIONS 101 and February) only 61 per cent. But the skies are more cloudy than these circumstances might lead one to expect, and in no month does the mean cloud proportion fall much below four-tenths of the expanse. The amounts of rain and snow are, however, insignificant. The average rain (and snow) fall is only 2 '7 inches in the year, and twice that amount is the greatest yet recorded in any one year (1879), while in one year (1876) it was less than half an inch. It snows most frequently in January and February, but the falls are very light in the valley and soon disappear. Eain is most frequent in July and August, but even then it occurs, on an average, only on one day in ten ; between one and two-tenths of an inch being the average fall of each rainy day. Agriculture is, therefore, almost entirely de- pendent on irrigation. The winds are generally light, and depend on the local direction of the valleys. At Leh, which stands at the en- trance of the valley leading up to the Khardong pass, the most common directions are between south and west in the daytime and summer, and from north-east in the night, especially in the later months of the year. In January and February the air is generally calm, and April and May are the most windy months of the year. Murree. This is the chief sanitarium of the Western as Simla is of the Eastern Punjab. It is situated on the summit of the ridge that divides the Jhelum valley from the Patwar (the tableland above the Salt range), and com- mands an extensive view of mountain and plain. The hills around are well wooded, except in the direction of the plains. None of those in the vicinity of Murree are of much greater elevation than the station itself, which stands nearly 7500 feet above the sea. The observatory which was established on its present footing in 1875 is at the Lawrence asylum, the military school, and fully 1000 feet lower than the station, on the crest of a spur that runs down towards the plains, and has a somewhat higher mean temperature than the station itself, while it is screened from northerly winds by the main ridge. 102 CLIMATES AND WEATHER OF INDIA Being thus situated on the crest of one of the outermost spurs of the Himalaya, its climate, like that of other Himalayan sanitaria, is of a very different type from that of Leh and Quetta. Although drier than that of stations such as Mussoorie and Nairn Tal, similarly situated but farther to the south-east, its atmosphere is much damper than that of the plains immediately below, and subject to smaller variations of temperature both annual and diurnal. The mean temperature of the observatory is 56, in January and February 39, and in June 71. From this time it falls gradually to 65 in September, and then rapidly to the end of the year. The lowest temperature is generally reached in February, when the mean minimum reading is 34. The lowest yet recorded, in 1886, is 16*7. Owing to the comparative lightness of the summer rainfall, the temperature, from June to September, is higher than at the more easterly hill stations. Notwithstanding the elevation, the shaded thermometer not infrequently rises above 90 in June, and in 1880, a very dry year, registered as high as 9 8 '7. The diurnal range varies but little in the course of the year. It is rather smaller in the winter and in August, when it amounts to 14, than at other times of the year, and is greatest in the driest months, viz. from April to June, when, however, it does not much exceed 17. These figures, it must be remarked, represent the conditions of the observa- tory only, and it is probable that the amounts would be found to differ considerably, with differences of position and aspect, in different parts of the station, but those general characters which distinguish the climate from those of other hill stations are doubtless shared by all parts of the site. From June to November the air is much drier than that of Simla, or any other hill station of the outer Himalaya, but in February it is apparently damper, and in the remain- ing months of the winter and spring of about the same character. The humidity is highest in August, when it amounts to 80 per cent of saturation. In March the skies are as cloudy as in the rains. In the spring months, from March to May, it rains on CLIMATES OF HILL STATIONS 103 an average about one day in three, most frequently in April, the rainfall on each rainy day averaging about four- tenths of an inch. Eain is less frequent in June, but increases again in July and August, when about half the days are more or less rainy, and the falls heavier than in the spring, averaging three-fourths of an inch on each day of rain. The number of rainy days in the year has varied from 74 only, in the dry year 1880, to 123 in 1885, and the total quantity from 39*1 inches in the former to 71 '8 inches in the latter year. October and November are the clearest months, but even these, and every month up to May, are more cloudy than at Simla or on the plains. On the other hand, in July and August there is less cloud than at Simla or other more easterly stations, as might be expected from the smaller frequency of rain. Simla. The well-known summer capital of the Govern- ment of India, distant 5 8 miles by road from the plains and 2 3 by direct measurement, is built on a ridge, much of which is rather over 7000 feet, separating the valley of the Sutlej from the drainage of the Tonse and Jumna, and therefore forming part of the main watershed of India. The greater part of the station crowns the crest of an amphitheatre of hills facing to the south-west, but shut off from the plains by the peak of Tara Devi opposite the station, and by a more distant ridge, on which are the military sanitaria Dugshai and Kasaoli. This amphitheatre is formed by the main ridge and a spur which runs southwards from its highest point, the forest-clad hill Jako, which rises above its north-east corner to a height of 8048 feet. A small part of the station occupies the northern face of Jako and the crest of a spur which runs off to the north in the direction of the Sutlej. Trees are rigorously pro- tected within the limits of the station, and many of the houses are embosomed in a thick growth of deodars, blue pines, and oaks ; but the lower spurs and hillsides are for the most part bare of timber, presenting steep grassy or rocky slopes, and every available acre of moderately-inclined ground is terraced and cultivated. The descents on all 104 CLIMATES AND WEATHER OF INDIA sides are very steep to the stream beds, about 2000 feet below the crest of the ridge. The observatory is attached to the Government offices, on the southern aspect, a hun- dred feet below the crest of the ridge. With a higher rainfall and, on the whole, a somewhat damper atmosphere than Murree, the climate of Simla is subject to greater variations of humidity at different seasons, though not of temperature. The mean temperature of Simla is 55, or nearly the same as that of Murree, and the same as that of Milan; that of January and February 41, and that of June, the warmest month, 67. The former of these temperatures is the same as that of Florence in January, and the latter that of Frankfort and Vienna in July. The lowest readings reached in the course of the year vary between the freezing-point and 13 below it, 19 '7 being the lowest yet recorded. The highest are generally below 90, but in 1877 rose to 94'4. The course of variation during the year is much the same as at Murree. The extent of the diurnal range is, however, more variable. In April and May, when it is greatest, there is an average difference of 20 between the early morning and afternoon the same as in the south of England in July, while in August this difference is only 11. During the six winter months it is pretty steady, averaging between 16 and 18. The driest month is December, when the mean humidity of the air is only 47. But in May, the mean humidity of which is 49 per cent of saturation, in some years, for short periods; it has been known to fall much below these averages, even as low as 20 per cent. The early part of June is sometimes equally dry, but the rains which, as a rule, set in towards the end of the month, are often preceded by afternoon storms, which mitigate the dryriess of the air and reduce the temperature. In July and August the rains are heavy, and the mean humidity as high as 88 and 91. At this time the station is frequently enveloped in cloud and mist, but less continuously than Darjiling and Nairn Tal, or even than Mussoorie ; and even at the height of the rains the evenings are frequently fine. The rains cease about the CLIMATES OF HILL STATIONS 105 middle of September, and their cessation is followed by the clear mild weather that renders Simla at this season one of the most charming places of residence to be found in the Indian empire or elsewhere. During the latter part of September, and all through October and November, the skies are generally cloudless, and only as an exceptional occurrence is the calm serenity of this season interrupted by a day or two of rainy weather. About Christmas the clouds begin to gather, and the first three months of the year are chequered by occasional storms of snow, and, later on, of rain and hail. Thunder and hail storms become more frequent in May and June, though interrupted by long spells of hot dry weather, and it is during these inter- vals that the exceptionally high temperatures above noticed are experienced. The mean annual rainfall of Simla is 70*1 inches ; higher, therefore, than that of Murree, the excess being entirely due to the greater heaviness of the summer monsoon rains. The average number of rainy days in the year is, however, not much greater, viz. 99. It rains less frequently in the spring months, but more frequently in the rains. In the wettest years there have been 136 rainy days. Snow usually falls several times in January and February, eight being the average number of snowy and rainy days in these two months, on most of which the precipitation takes the form of snow. About once in 8 or 10 years the accumula- tion may be to a depth of 4 or 5 feet in a single fall, but as a rule it is much less. Hail-storms are common in March, April, and May, and occur occasionally in June, the precipitation sometimes, though rarely, taking the form of soft hail, which is unknown on the plains. Though more frequent than on the plains, the hailstones are generally smaller on the hills and less destructive, the commonest size being that of small marbles. The winds at Simla are generally light; those from north and north-west are dry ; and the houses, on the Elysium spur and the north face of Jako, which are most exposed to these winds, have a more bracing atmosphere io6 CLIMATES AND WEATHER OF INDIA than those of Chota Simla and Boileaugunj, which have a southern aspect. The latter also receive more sunshine, and in the rains more cloud and mist. There is thus a very appreciable difference between the northern and south- ern portions of the station, the latter being much the more extensive. Chakrata. This is a small military sanita,rium on the crest of a ridge which rises to the west of the Jumna valley, a few miles above Kalsi, where it makes its exit from the Himalaya. It is 50 miles in a direct line south-east of Simla, and 22 to the north-west of Mussoorie. Except the low range of the Sivaliks and the spurs which run down from Deobund to the Doon, no hills intervene between the Chakrata ridge and the plains. The station lies well within the zone of heavy rainfall that borders the outer Himalaya, and at a height of 7000 feet above the sea. The Deobund range, immediately north of Chakrata, is a few hundred feet higher, but to the north-east falls towards the Jumna, and the station commands a magnificent panorama of Junmotri and the snowy peaks at the head waters of the Jumna and Ganges. The hills around the station, like this part of the Himalaya generally, present bare precipitous slopes to the south, and are more or less wooded on their northern aspect. Being at almost exactly the same elevation as Simla, but more southerly and nearer to the plains, its mean tempera- ture is about 1 higher, viz. 56, and as the course of the annual variation at the two stations is almost identical, there is a similar difference in most months of the year. The diurnal range varies in like manner with the seasons, and is 1 or 2 less than at Simla at nearly all times of the year. The mean annual range is also 2 less, and somewhat less variable from year to year. The mean highest temperature of the year is 85, the lowest 24; the highest yet recorded 8 9 '6, and the lowest 18-7. These differences are not indeed great, but they illustrate the modi- fication that ensues in the climate of the outer Himalaya in the course of only 50 miles, as the chain is followed to the south-east. CLIMATES OF HILL STATIONS 107 Except in March, April, and August, the air is rather damper than at Simla, and most so from October to Febru- ary. But except in July, August, and September, Chakrata is by no means a damp station, and its situation on the crest of a ridge exposes it freely to the usually light winds of the mountain zone. On the average of 17 years the rainfall is only 61*5 inches, and more than half of this falls in the two months July and August. The winter rainfall is about the same as at Simla, that of April and May little more than half as great. The number of rainy days in the year averages 110, and in the wettest year has not exceeded 127, while in the very dry year, 1877, there were but 95. Ranikhet. This, too, is exclusively a military sanita- rium, situated in the Kumaon Himalaya, 120 miles south- east of Chakrata, rather more than 6000 feet above the sea, on a little plateau, 25 miles distant from the plains, from which it is shut out by the higher range, in which lie em- bosomed the station and lake of Naini Tal. This range shields it from the heavier rainfall of the southern face of the hills, and it enjoys a more equable climate than any station yet noticed. The plateau is well wooded with pine groves, amid which clearings have been made for the build- ings. To the south and south-west extends a broad valley clear of forest, leading down to the Eamgunga river, and to the north an extensive depression in the hills of similar character affords a wide view over lower ranges and valleys, terminating some 20 miles off against a higher range, above which towers a gigantic group of snow-clad peaks, including Nanda Devi and the Trisul. To the north-west the view is equally unobstructed, and on clear days the great independent groups of snowy peaks, Badrinath, the Banda Poonch, etc., may be seen rising one beyond the other in diminishing perspective to a distance of more than 100 miles, towering above the lower mountains, and separated by great valleys from the watershed of the Upper Sutlej. The mean temperature of the station is 60, and the average annual range from 30 to 86; the extremes, hitherto recorded, being 26 as the minimum, and 8 7 '8 as io8 CLIMATES AND WEATHER OF INDIA the maximum. The absolute range of temperature is therefore 9 less than that of Chakrata, and 13 less than that of Simla. The mean diurnal range is from 1 to 3 less in different months, except from July to October ; and in April and May, when greatest, does not exceed 17. This comparative equability of temperature is not, however, due to greater dampness. On the contrary, the mean humidity of the air, in most months, is appreciably lower than at Chakrata, and even in July and August is only 85 and 86 per cent of saturation. There would appear also to be somewhat less cloud, especially in the winter months. The rainfall, 49 inches, is less by 12 inches, and little more than half that of Naini Tal ; and there are on an average but 9 5 rainy days in the year ; 1 7 is the greatest number yet recorded. The winds are light and calms frequent at all times of year. Those from the south-west greatly preponderate, owing probably to the situation of the station at the head of the broad valley above mentioned. On the whole, Eanikhet enjoys a more equable, moderate, and at the same time rather drier climate than any other station in the north-west Himalaya from which we have comparable data. Darjiling. An interval of 570 miles separates this station from Eanikhet, the greater part of which is occupied by the kingdom of Nepal. For this part of the Himalaya, save Pithoragarh, an unimportant station 50 miles from Eanikhet, and Katmandu the capital of Nepal, in a valley 4000 feet above the sea, any accurate climatic data are wanting ; and Katmandu is rarely accessible to Europeans, and not available as a sanitarium. Darjiling is situated 16 miles from the plains, on a branch of the Singalelah spur, running southwards from Kanchinjinga, the gigantic peak which rears itself at a distance of only 40 miles to the north of the station. The station is built, partly on the crest of the ridge, partly on its western and south-western slopes, at elevations between 6500 and 7800 feet ; but it is shut off from a direct view of the plains by the Goom Pahar ridge, running east and west between Sinchal (8608 feet), CLIMATES OF HILL STATIONS 109 and Tonglo (10,081 feet). On the north the Darjiling ridge descends rapidly to the valley of the Great Kungit, across which and the ridges beyond, the station commands a magnificent view of the snow-clad monarch which domi- nates it, and with its attendant peaks affords one of the grandest spectacles to be met with in the Himalaya and probably in the world. The highest part of the Darjiling spur is where it runs up to join the Goom Pahar, and it is on the crest of this spur, at 7421 feet above the sea, that is situated the present observatory, established in 1881. This house, now occupied by St. Paul's school, is the same as that at which Dr. (now Sir Joseph) Hooker recorded his observations in 1848. For this site we now possess a meteorological register for a little more than 4 years. For 14 years previously the observatory was at the tele- graph office, 500 feet lower down, on the west face of the ridge, and more on a level with the greater part of the station. An abstract of the results of both is given in the Appendix at the end of this work. The existing data for the new observatory are hardly sufficient to show its climatic difference with great precision, but it would appear that its mean temperature is not less than 3 lower than that of the old observatory, and its atmosphere about 3 per cent of saturation damper. The diurnal range of temperature is moreover 2 or 3 less. Taking the registers of the old observatory as repre- senting the climate of the greater part of the station, it appears that, notwithstanding its more southerly position and (excepting Eanikhet) its somewhat lower elevation, the mean temperature of Darjiling is slightly lower than that of any of the north-western sanitaria, viz. 54. But the most striking point of difference is its much greater dampness, a difference which characterises it in almost every month of the year, and more especially in the spring season. Even in the driest month, March, the mean relative humidity of Darjiling exceeds 70 per cent, and from June to September inclusive is above 90 per cent of saturation. Its cloudiness is not less striking. The clearest and most i io CLIMATES AND WEATHER OF INDIA serene months are November and December, but even in these, on an average, nearly half the sky is clouded. During the rains, from June to October, for many days in succession, the whole station is shrouded in cloud and mist, with almost incessant rain. Even in May more than half the days are rainy, and from June to September the number of rainy days varies from 22 to 28 on an average in each month. As the rains begin early, so also they last late, and it is not till the middle of October that they can be considered fairly at an end, the average rainfall of that month being 6 '4 inches. The mean rainfall of the year is 120 inches, distributed over 149 days, and in 1870 there were 171 rainy days. It is to this dampness and cloudiness, and the frequency of rain throughout the spring and summer, that must be attributed the comparative coolness of the station. The highest temperature reached in the year is, on an average, only 78, which is 8 lower than the average maximum of Eanikhet and 10 lower than that of Simla. The average minimum of the year is 30, which is the same as that of Eanikhet ; and 2 6 is the lowest recorded at the old observatory in 14 years. The mean difference of extreme night and day temperatures varies from 11 in June, July, and August to 17 in March, April, November, and Decem- ber ; a range of variation which is the same in amount as at Eanikhet. The winter months are generally fine, and notwithstanding the frequency of cloud, rain and snow are less frequent at Darjiling at this season than at the north- western stations. The military sanitarium on the Jellapahar is on the same ridge as the new observatory, but 400 feet higher up, and its climate, therefore, differs in the same way from that of the old observatory, and to a greater extent. It is colder, and the site is more frequently enveloped in mist. The gardens which produce the fine teas of Darjiling are all at lower levels than the station, and in a warmer climate. Some gardens, established in the early days of tea cultivation between 6000 feet and 7000 feet, were found CLIMATES OF HILL STATIONS in to yield indeed a tea of fine flavour, but the produce was insufficient to be highly remunerative, and tea cultivation is now carried on at all elevations from 5000 feet down to the plains. Another important product of the Darjiling valleys is cinchona bark, the cultivation of which is carried on at the Government plantation, on a larger scale than anywhere else in India. The gardens are situated below Darjiling, to the east of the station, on the slopes that run down to the Tista valley. A register of temperature and rainfall has been kept here during the last 8 years, under the same conditions as at other Government observatories, at the superintendent's residence at Mongpoo, about 3000 feet above the sea, and affords the following summary of the temperature and rainfall, which will be of much interest to those who are engaged in the cultivation of the cinchona. The rainfall averages are obtained from 13 years' registers. TEMPERATURE AND RAINFALL OF MONGPOO CINCHONA GARDEN TEMPERATURE. RAINFALL. Mean. Mean Max. Mean Min. Mean Range. Ins. Days. Daily. Month. January 54 59 49 10 17 0-8 2 February 55 60 50 10 23 0-8 3 March 63 69 57 12 25 2-1 6 April . 68 73 62 11 21 4'0 9 May . 69 73 64 9 21 9-2 17 June . 72 75 68 7 17 29-9 22 July . 73 78 69 9 14 31-2 25 August 72 76 68 8 15 26-0 24 September October 72 68 76 72 67 63 9 9 17 19 16-5 5-3 16 7 November 61 67 55 12 19 o-i 1 December 55 60 49 11 18 0-2 1 Year 65 125-9 133 Mean highest temperature of year . . . .83 Mean lowest temperature of year . . . .43 Mean range of temperature of year . . . .40 Highest recorded reading . 83 Lowest recorded reading . 40 '5 Absolute range . . . 43 '1 Rainfall of wettest year 155 - 8 ins. Rainfall of driest year . 92 ins. Rainy days in wettest year . 156 Rainy days in driest year . 116 ii2 CLIMATES AND WEATHER OF INDIA The most striking characteristic of the Mongpoo climate disclosed by these registers is its great equability. With an annual mean temperature of 65, nearly the same as that of Athens, Algiers, and Madeira, the extreme variation during the year is rather less than that of Moulmein less than that of any place enumerated in the table on page 1 8, save only Bombay, Cochin and Galle ; and throughout the year the daily range is hardly greater than that of the dampest pro- vinces of India in the rainy season. Moreover, this equability is displayed not only in the average conditions of many years, but perhaps with even greater emphasis in the regularity of their recurrence year after year. Thus, in the course of 8 years, the highest temperature reached in each year has varied only between 81 '5 and 83'6, and the lowest be- tween 40 '5 and 46*5. It has never descended to within 8 of the freezing-point. Shillong. This station, which is the seat of Govern- ment in the province of Assam, is situated on a gently undulating tableland 4800 feet above the sea -level, im- mediately north of the culminating ridge of the Khasi hills, and about midway between the valley of Assam and the plains of Silhet. It is only 30 miles north of Cherrapunji, (long notorious as the wettest known place in the world), but the higher plateau that intervenes effectually shields it from the heavy rainfall of the southern escarpment of the hills, and the rainfall of the station averages only 8 5 inches, or little more than two-thirds of that of Darjiling. The Shillong ridge, immediately to the south of the station, dominates it by more than 1000 feet, but to the north the general level of the plateau falls away gradually towards the Brahmaputra. The surface of the plateau around Shillong, as indeed of the Khasi hills generally, is a rolling green sward, with occasional clumps of pines ; but trees are for the most part restricted to the valleys which, in the course of ages, have been eroded by the hill drainage, the higher parts of the surface being kept clear of forest by the fires annually kindled by the natives for the destruction of the coarse grass and the stimulation of a young growth fitted for pasture. CLIMATES OF HILL STATIONS 113 Partly owing to the difficulty of access from other pro- vinces, and partly to the want of spare house accommoda- tion, Shillong has hitherto been but little visited by others than the Government officials and a few residents of the Assam valley. The former obstacle is now in a great measure removed by the completion of the railway from Calcutta to Dhubri, and the establishment of a line of swift steamers daily from that place to Gauhati, from which Shillong may be reached in 10 hours ; and when new houses are provided to meet the increasing demand, it may be expected that in the course of a few years the place will become better known ; and a station which, in its mild climate and open downs, holds forth to the residents of Bengal many of those attractions which make the Madras Mlgiris so charming a resort, will attract from the over- crowded boarding-houses of Darjiling some of those who seek a temporary refuge from the "branding summers of Bengal." To such persons the following sketch of its climatic features will be welcome. On the average of the 4 years, 1869-1872, the mean temperature of Shillong was 62, which is about the same as that of Constantinople, Barcelona, and Oran ; in fact of an average Mediterranean climate, and the July temperature of the south of Ireland and the more northern midland counties of England. Shillong itself has this temperature in March and October. In the warmest months, June to August, it is below 70, and in April and May intermediate between the two, since in the Khasi hills, as in the pro- vince of Assam generally, rain is so frequent in the spring months that the temperature does not rise to a maximum in May, as on the plains of Northern India, and suffers no abatement when the monsoon rains set in in June. The highest temperature reached during the 4 years was 84, and the maxima of the other years were a degree or two lower. The lowest reading recorded was just above the freezing-point ; the average minimum of the 4 years 34, and the mean temperature of December and January 51. The winter temperature is therefore nearly the same as that of I 114 CLIMATES AND WEATHER OF INDIA Lisbon and Palermo, and the July temperature lower than that of either. In December and January, the most serene months of the year, the mean difference of the early morning and afternoon temperatures is 19 or 20, which is as great as that of Simla in April and May, and is that of the south of England in July. In July it is only half as great. While, in respect of temperature, the climate of Shillong much resembles that of places in the south of Europe, in the dampness of its atmosphere and its rainfall it is emi- nently tropical. In the driest month, March, the humidity is indeed only 59 per cent of saturation, but from July to October inclusive it ranges between 86 and 89, and from June to September, on an average, eight-tenths or more of the sky is clouded. Except the winter months, or from November to March, there is no dry season. Even in April it rains on one day in three, in May on two days out of three, and in the four succeeding months even more fre- quently. On the average of 18 years there have been 150 rainy days in the year. But although it rains more fre- quently than at Darjiling, the rain is much less heavy, since, as above remarked, the average rainfall of the year hardly exceeds 85 inches, of which 70 inches falls in the five months, May to September. In the remainder of the year the climate equals that of the most favoured countries in Europe. Pachmarhi. This is a small station on the Mahadeva or Pachmarhi hills, in the Central Provinces, about 1 south of the Tropic of Cancer. It is the hot weather residence of the Chief Commissioner and some of the higher officials of the Province, and is also used as a sanitarium for a small body of British troops. The Pachmarhi hills form a prominent group in the belt of hilly country that stretches across India to the south of the Sone and Ner- budda valleys, now generally known as the Satpura range. On the south, the Pachmarhi group terminate in a pre- cipitous escarpment of thick-bedded sandstones, and on the north, slope rapidly but less abruptly towards the Nerbudda valley. Their summit forms a little undulating plateau, CLIMATES OF HILL STATIONS 115 grassy and dotted with clumps of trees, which give it the general aspect of an English park, diversified with bluffs and higher peaks of sandstone which, in Dhupgarh, rise to 1000 feet above the general level of the tableland. The scenery is peculiar and very beautiful. The elevation of the station being only 3500 feet above sea-level, and 2500 above that of the Nerbudda valley, its climate is of the same general character as that of the Central Provinces around, with the lower temperature and higher rainfall due to its elevation and the form of the ground. It is sufficiently high to afford cool nights even at the hottest season, but in the daytime the sun is very powerful, and in some years the shaded thermometer rises to above 1 0. As a sanitarium it cannot be considered as a rival to the Himalayan stations and the Mlgiris of Southern India, but it affords a grateful place of refuge from the intense heat of the lower plains from March to May, and at other times of year, except perhaps sometimes in the rainy season, the climate is charming. The mean annual temperature of Pachmarhi is 69; that of December, the coolest month, 56; and that of May, the warmest, 83, rising to 94 in the daytime and sinking to 74 at night. The rains, which usually set in about the middle of June, bring down the day temperature nearly 20, and the night temperature by 5 or 6. The former rises again slightly at the end of the rainy season, in September and October, but this slight rise is compensated by the gradual fall of the night temperature, and after October the decline is rapid. Even in the coldest season the ther- mometer rarely sinks to the freezing-point, 35 being the mean minimum of the year. Except in the four months of the rains, the atmosphere is very dry, especially in the spring or hot season. In April the average humidity is only 26 per cent, and from November to May below 60, the cloud from two to three- tenths of the expanse, and the rainfall from three to seven- tenths of an inch in each month. In July rain is heavy and frequent, amounting to nearly 29 inches on 25 rainy days. In August it is lighter, but almost equally frequent. ii6 CLIMATES AND WEATHER OF INDIA la the cyclonic storms, described on a previous page, which frequently traverse this part of India in the rains, falls of 3 or 4 inches in the 24 hours are by no means un- usual, and occasionally as much as this is registered on two or more days in succession. The rains come to an end in the latter part of September, but there are usually a few heavy showers in October before the weather finally clears. The natural drainage of the plateau is complete and rapid. A little lake has been formed by damming a small water- course that intersects the station, and this is considered to have somewhat impaired the salubrity of the place, which is hardly above fever range. Chikalda. About 100 miles to the south-west of Pach- marhi, this small sanitarium has been established at 3656 feet above the sea, on the highest part of the Gawilgarh hills, which, like the Pachmarhi group, are a part of the Satpura range. It is little known beyond the province of Berar, in which it is situated, and in elevation, and the general character of its climate, much resembles Pachmarhi. The geological formation of the Gawilgarh hills is, however, different, being a portion of the great Deccan basaltic trap. Chikalda stands on the southern margin of the hills, over- looking the plain of Berar, which has here an elevation of over 1000 feet. A Government observatory, established here in 1875, affords the data summarised in the Appendix, from which are gathered the following particulars of its climate. The mean temperature of the year is 70 or 1 warmer than Pachmarhi, and the annual course of its varia- tion is of the same general character, except that its range is much smaller, the winter months being 6 warmer, while those of the monsoon (June to September) are 2 or 3 cooler. The highest temperature of the year is, however, about the same as at Pachmarhi, the mean of May being 94, and the mean highest reading in the year 99, with an absolute maximum of 103 in the dry year 1876. The lowest yet recorded is 3 9 '5, which is 9 '5 higher than that of Pachmarhi ; and therefore the absolute range of recorded CLIMATES OF HILL STATIONS 117 temperature is only 63*5, or 9 less than that of the more northerly station, the elevations of the two being the same within 30 feet. The rainfall is 1 1 inches less than at Pachmarhi ; that of July alone being 1 1 inches less, and that of both August and September 2 or 3 inches less, but there is more rain in October, viz. over 4^ inches on an average, whereas Pach- marhi has but 2. In other respects, the two climates are very similar. Mount Abu. Mount Abu is a lofty isolated hill or cluster of hills, 7 miles to the west of the Aravali range ; the name given to the chain of rocky hills that runs south-west from Delhi through Eajputana to the northern limits of the Nerbudda valley, forming the western margin of the Malwa tableland. Mount Abu is a rocky eminence rising to 5653 feet above the sea-level, and crowned by a Jain temple of great sanctity, a renowned place of pilgrimage to the adherents of that sect, from all parts of India. The station of the same name, which is the hot-weather residence of the chief officials of Eajputana, the seat of a Lawrence asylum for soldiers' children, and is also resorted to by the residents of Guzerat, is built on this hill at an elevation a little below 4000 feet, that of the observatory being 3945 feet above sea-level. Being on the border of the arid region of western Eajputana, it shares the general character- istics of its climate, save that, owing to its elevation, it is much cooler, and has a rainfall between double and treble as great as that of the plains around. Although it lies more than 2 north of Pachmarhi, and is 400 feet higher, the mean temperature of the year is only 1 lower, viz. 68, and the January temperature the same; that of the spring months, March to May, being 3 or 4 lower, and that of the three final months of the year higher by the same amount. In January, the coolest month of the year, the, temperature is 58. The lowest reading varies much in different years, not having fallen below 45'3 in 1882, while in 1878 it was as low as 32'8. On the average of 9 years it was 39. In May, which is the ii8 CLIMATES AND WEATHER OF INDIA warmest month, the mean temperature is 79, but the afternoon average is 89, and the highest readings have varied from 93 to 101 in different years: the nights, however, are cool, the temperature falling, on an average, to 71 be- fore sunrise. Owing probably to the position of the station, on a hill which descends on all sides to plains 3000 feet or more below the station, the daily range of the ther- mometer is smaller than at Pachmarhi, not exceeding 18 on the average of any month in the year, and being only 16 or 1*7 in the winter months, and 8 or 9 in the rains. As already remarked, the air is very dry during the greater part of the year. The mean humidity of March or April is only 29 or 30 per cent of saturation, and from November to May inclusive, in only one month does it ex- ceed 40. In August it rises to 87. From October to May the skies are very clear, the mean proportion of cloud varying from one to three-tenths of the sky expanse, and the rainfall is equally slight, indeed insignificant, not ex- ceeding two-tenths of an inch in any month, on an average, from November to April. Only in July and August is it at all heavy, the average of each of these two months being rather over 22 inches, and this falls in quantities averaging about 1 inch on each rainy day. The mean rainfall of the year is a little over 63 inches, distributed over 78 rainy days. The greatest in any one year was 123 inches, and the least 19*2 inches. From May to August or September the wind is steady from the south-west or west, and of moderate strength, averaging 1 or 12 miles an hour ; in November and Decem- ber northerly winds prevail, and in the remaining months they are more or less variable, but with a preponderance of west winds, the dry winds of all North-western India. Mahableshwar. Except Matheran, within a few hours' railway journey from Bombay, the only hill station in the Bombay Presidency is Mahableshwar, on an eminence of the Western Ghats, 4540 feet above the sea. For this station I have no recent and trustworthy registers of climate, save only of rainfall. This has long been known to be ex- CLIMATES OF HILL STATIONS 119 ceptionally heavy at Mahableshwar, exceptional even in the rainy zone in which it is situated, and is surpassed only by that of Cherrapunji and a few other places in Assam. The residents of Bombay resort to Mahableshwar during the hot season that precedes the rains, but from June to October the favourite place of residence is Poona, which is more accessible from Bombay and, together with a light rainfall, not exceeding 25 inches from May to October, nevertheless enjoys a cool atmosphere throughout the rainy season. The climate of this place will be noticed further on, in connec- tion with that of the Deccan generally. Ootacamund. Beside one or two smaller settlements and numerous plantations of coffee and cinchona (chiefly on the western slopes), two sanitaria, one civil and one for military purposes, have been established on the Nilgiri hills, one of the loftiest of the hill groups dotted over the southern part of the peninsula. The first, Ootacamund, is the summer residence of the Madras Government ; the second, Wellington, formerly known by its native name, Jakatalla, is exclusively occupied by British troops ; but there is, in its vicinity, the civil settlement of Coonoor, the climate of which is very similar. The Mlgiris are a lofty mass of hills at the southern extremity of the Mysore tableland, where the Eastern and Western Ghats converge. Their summit is a grassy un- dulating and hilly plateau about 20 miles across, and aver- aging between 6000 and 7500 feet above the sea. On its western margin it is crowned by a somewhat higher ridge, the western face of which falls away abruptly towards the plains of Malabar, as a continuation of the Western Ghats of the Wynaad and Mysore ; and about midway in its extension it is crossed from north to south by another and higher range of hills, the culminating point of which, Dodabetta, is 8640 feet above the sea. Ootacamund and Wellington are on the opposite sides of this latter range, the former to the west, the latter on its eastern flank, and thus, while Oota- camund is more exposed to the westerly monsoon which blows in the summer months, Wellington receives more rain, 120 CLIMATES AND WEATHER OF INDIA when, in October, the wind has changed to east and north- east ; a change which ushers in the north-east monsoon of the Bay of Bengal. One could hardly desire, and it would certainly be hard to find within the limits of the Indian empire, or perhaps elsewhere, a more charming climate or one more fitted to the European constitution than that of the Nilgiri hills. The Laureate has sung of the " sweet, half-English Nilgiri air;" but it is the air of the English spring and summer, without Atlantic storms or the bitter east winds of March ; a climate where one may inhale the fresh breeze that blows over rolling downs, and brave with impunity the ardour of a tropical sun, and even enjoy the cheerful companionship of an even- ing fire through the greater part of the year. The flowers of Southern Europe and well-flavoured English vegetables flourish with a luxuriance unknown in the torrid zone, save on such as this and its sister hill groups, the Shevaroy and Pulni hills. Shielded by the Kundah and Makurti ranges from the heavy rainfall of the Western Ghats, fine and cool cloudy weather is the rule rather than the excep- tion, even in the summer monsoon ; and in mid-winter the clear frosty atmosphere of the early morning is warmed up daily by a genial unclouded sun, and day follows day un- troubled by snow-storms, and free from their unwelcome sequels, thaws, slush and catarrh. Without the grandeur of the Himalayan gorges, or the majesty of eternal snows, the Mlgiris have a soft beauty of their own, recalling to the Englishman the undulating contours of his own western hills ; and the pedestrian can strike a bee-line over hill and dale, whithersoever his fancy leads him, unimpeded by any obstacle more serious than an occasional peat- bog or a brawling hill stream. Among all the pleasant memories of more than 3 years of Indian life, and an experience of all parts of the empire from Peshawar to Sibsagar and Point de Galle, I can recall no more charming scene and climate than those of the Nilgiri hills. Many an old Anglo-Indian, whom choice or necessity has led to fix his home in India, has found in these hills scenery as beautiful and a climate CLIMATES OF HILL STATIONS 121 as enjoyable as any in the most favoured lands of the Medi- terranean shores. One striking characteristic of the Ootacamund climate, due to its position low down in the tropics, is the compara- tive uniformity of its temperature throughout the year. While the mean temperature of the year is the same as that of Simla (viz. 55), that of May, the warmest month, is only 4 above this mean, and that of January, the coldest, only 7 below it. Even this difference is due much more to the variation of the night temperature than to that of the day, and the latter is affected more by the amount of cloud and rain at different seasons than by the sun's declination and the length of the day. The afternoon temperatures are lowest in July, but vary very little from that month to December, after which they rise steadily till April. From April to July they fall 10. The night temperatures are lowest in January, and rise 18 between that month and May ; from which time they vary only 1 or 2 till October. The highest thermometer reading recorded in 1880, the only year for which we have a trustworthy register, was 7 7 '3 and the lowest 25'3. The daily range is very great in clear weather, as is usually the case on tablelands. In January, February, and March it varied, on an average, between 27 and 31 in the 24 hours, but only from 10 to 14 in the cloudy months, June to October. The atmosphere is not very dry, except in March, the only month of 1880 in which the mean humidity ranged below 50 per cent of saturation. From June to November inclusive, the average of any month fell little below 80 per cent, and in November was 90. From June to October the skies were very cloudy; but as the rainfall register of Wellington shows that 1880 was a remarkably wet year on the Nilgiris, there is reason to believe that Ootacamund is in general a less damp and cloudy station than might be inferred from the table in the Appendix. In 1857, the geological survey of the Mlgiri hills was carried out during the summer monsoon with no more interruption from bad weather than would have been experienced in an English 122 CLIMATES AND WEATHER OF INDIA climate. Notwithstanding the prevailing cloudiness at this season, the rainfall is by no means heavy. From May to August the weather is showery, but the total amount of rain is only 4 or 5 inches in the month, and out-of-door life is as little checked by weather as in an ordinary English summer. The heaviest rain is in October. Wellington. At a level 1000 feet lower, and with an additional screen interposed between it and the westerly monsoon, in the Dodabetta range, which dominates it by 2500 feet, Wellington is 6 warmer than Ootacamund, and has less cloud and rain ; while in some respects it enjoys even greater equability of temperature. During the summer monsoon it is frequently fine and clear to the east of Doda- betta, while completely overcast and raining at Ootacamund ; a contrast which we meet with in a much higher degree on the opposite sides of the culminating range of the Ceylon hills. But, after October, the relations of the two stations to the rainy winds are reversed ; and in the final months of the year Wellington is the rainier station of the two. The mean temperature of Wellington, as deduced from between 12 and 14 years' registers, is 61. That of May, the warmest month, is only 5 higher ; and that of January, the coldest, only 6 lower. The average afternoon tempera- ture of May is 76, and the highest readings recorded in each of 6 years have varied only between 7 9 '5 and 80*7, or little more than 1. The early morning tempera- ture in January averages 45, and the lowest has varied between 34*2 and 37'5, or little more than 3 in different years. In the winter months the range during the 24 hours is considerably less than at Ootacamund. It amounts to 24 in February, and 21 only in January and March; and from June to November does not exceed between 13 and 15 on an average. The air is drier than at Ootacamund. The humidity is above 80 only in October and November, but it is below 70 only from February to May. There is also appreciably less cloud than at Ootacamund, but clear skies predominate CLIMATES OF HILL STATIONS 123 only from January to April. The rainfall is moderate about 48 inches only in the year. From April to August there falls from 3 to 4 inches in each month on an average, and it rains about every second day. In October and Nov- ember it is about three times as heavy, but even then there average 13 or 14 fine days in each month. In January and February rain is rare, and in March and April it rains on an average one day in five, chiefly in afternoon thunder- storms. Newara Eliya. The mountain region of Ceylon is an isolated mass in the south of the island, surrounded by a broad undulating plain on the south-east and east, and a more extensive plain on the north, while, to the south-west, a generally hilly country of greatly inferior elevation and intersected with alluvial flats extends between it and the sea. Most of the highest elevations are on the south, where Adam's Peak and some other eminences rise to over 7000 feet ; and from this range a spur of equal height is given off to the north, running north-north-west across the middle of the mountain tract and dividing it into two nearly equal por- tions. Midway in this spur is Pedro Talla-galla, the highest peak in the island, rising to 8296 feet above sea-level ; and on its south-western flank, 2000 feet below the summit, lies the little sanitarium of Newara Eliya. The station is built on and around a small alluvial flat, formed by a small stream that drains the western slopes of the mountain, and then precipitates itself through the Eangbodde Pass, by which the road descends in the direction of Kandy. The ridge, of which Pedro Talla-galla is the culminating peak, acts as a barrier to the rainfall of the summer monsoon, in the same way as the Dodabetta range on the Nilgiris, but in a much more emphatic manner, producing a very striking contrast of weather on its eastern and western slopes. The following description was written many years ago, after a visit to JSTewara Eliya in the month of June, and its accuracy has been confirmed by persons of much longer experience of the station. In the summer monsoon, day after day and week after week, Newara Eliya lies under 124 CLIMATES AND WEATHER OF INDIA a dense canopy of cloud that shrouds all the higher peaks and pours down almost incessant rain. But let the traveller leave the station by the Badulla road, and crossing the col of the main range at a distance of 5 miles from Newara Eliya begin the descent to Wilson's bungalow, and he emerges on a panorama of the grassy downs of the lower hills bathed in dazzling sunshine, while on the ridge above, he sees the cloud masses ever rolling across from the west, and apparently dissolving away in the drier air to leeward of the hills. The disappearance of the cloud is doubtless due to the descent of the cloud-laden current that has swept over the crest of the range, and to the warming which it neces- sarily undergoes in such descent ; which, according, to theory, must be 1 in about every 450 feet, until the cloud is dissipated. A similar phenomenon has been described by Sir John Herschell on the Table mountain at the Cape of Good Hope, and on a smaller scale, and less continuously, may be frequently witnessed on the ridges of the Himalaya. Newara Eliya is a much rainier though not cloudier station than Ootacamund, being unprotected (unlike the latter) by a higher range to windward. The table in the Appendix shows that it rains on an average 195 days in the year, and the annual rainfall is nearly double that of Ootaca- mund. Indeed, the only season that can be called fine is restricted to the first four or four and a half months of the year, and even in these it rains on one day in three or four. October, which is the first month of the winter mon- soon rainfall, is as rainy as June, July, and August, and even in November it rains on an average on two days out of three. The mean temperature of the station is 59, or 4 higher than that of Ootacamund, a difference due in a great measure to its lower elevation. In no month does the mean vary more than 2 above or below this ; and the chief variation at different seasons is that the extreme temperatures of day and night differ more widely in February, March, and April than in other months of the year. January and February are the coolest months, and it is warmest in May, CLIMATES OF HILL STATIONS 125 just before the setting in of the monsoon. But the highest temperature recorded has not exceeded 79 during the last 11 years, and it has varied only between that and 74 in different years. The lowest trustworthy reading on record is 32, and the average minimum of the year 35. Much lower readings were registered in 1875 and 1876, but there can be little doubt that for two and a half years an untrustworthy instrument was in use, and these accord- ingly I have rejected. The diurnal range of the thermometer is not so great as at Ootacamund, but rather approaches that of Wellington. It varies from 12 in June and July to 23 or 24 in February, March, and April. It is a very damp station, the humidity of the air being between 80 and 90 per cent of saturation from May to the following January, and not falling to 70 on the mean of any month. A damp and cool but not cold atmosphere, varying but little throughout the year, is therefore the leading characteristic of the climate of Newara Eliya. General Characters of Hill Climates. Herewith I conclude my notice of the climates of the hill stations of India and Ceylon, of such of them at least as are repre- sented in the Indian Meteorological Eeports. We have seen that they present almost as great variety as do the plains below them. In all cases their atmosphere is cooler and damper than that of the neighbouring plains ; but while those in the North-west Himalaya are subject to great vicissitudes of heat and cold, dryness and dampness in the course of the year, those of Southern India and Ceylon are comparatively uniform in these respects, and their fine clear season is shorter than at the northern stations, and by no means so dry. Of all the stations enumerated, Ootacamund is that which most frequently reminds one of England ; but the resemblance is not very close, and consists chiefly in its coolness, its cloudy skies and moderate rainfall. The climate of Simla and the Kurnaon hill stations is rather South European in type ; and to Darjiling hardly any parallel can be found, except perhaps, and that imperfectly 126 CLIMATES AND WEATHER OF INDIA and in respect of dampness, in Cornwall and the south-west of Ireland. As to the stations at lower elevations, Shillong, Pachmarhi, Chikalda, and Mount Abu, they retain too much of a tropical character to be put in comparison with any European analogue. They are pleasant places of refuge from the scorching or suffocating heat of the plains, and seem cool by comparison ; but, except in the cold season, when they are charming, they are scarcely such as an European, free to select his place of residence, would prob- ably choose as an abode. CLIMATES OF THE PLAINS I now pass to the description of the climates of the plains of India, beginning with the Northern and Western provinces, then tracing out the changes that gradually ensue as we proceed eastwards to Assam and southwards to Ceylon, and concluding with what is at present known of the climate of the Burman peninsula. The tabular sum- maries of temperature, humidity, etc., at a number of the more important stations given in the Appendix, will serve to illustrate these descriptions, and to afford exact data for many of the military cantonments and larger civil stations, in which the European population chiefly centres. Punjab. Of all the provinces of India, the Punjab, which is the most northerly, is also that in which the vicissitudes of climate are greatest. It has the coldest winters, and, together with Sind, the hottest summers ; and it has been shown at page 12, in the illustrative instance of its capital, Lahore, that the irregular changes of temperature from day to day, which accompany changes of weather, are greater and much more frequent than in the damper province of Bengal, or, it might be added, in any other part of India. Next to Sind and Western Eajputana, which border it on the south, it is in great part also the driest province ; but in this respect there are great differ- ences in different parts of its area. Much of its southern CLIMATES OF THE PLAINS 127 and western districts is virtually desert, not owing to any want of natural fertility in the soil, for, wherever it can be brought under irrigation, no soil is more productive, but solely to the scantiness and uncertainty of the rainfall. But the tract, from 50 to 100 miles broad, that extends along the foot of the Himalaya, receives in most years a rainfall sufficient for successful agriculture ; and as this occurs at two different seasons of the year, viz. from January to April and from July to September, two series of crops are raised annually, the spring crops or rabi consist- ing chiefly of wheat, barley, and gram (cicer arietinum), the autumn crops of millet, maize, various pulses, oil seeds, and sugar cane. Cotton, indigo, and some rice are also among the agricultural produce of the Punjab. How the climate impresses the European is well shown in the following extracts translated from a lecture by the Eev. J. M. Merk, for 16 years a resident in the province. With some slight modifications, to be noticed in due place, the description will apply to the greater part of North- western India. " Like the rest of India, the Punjab has really but three seasons : the summer or hot season, the rains, and the winter, which, in India, we speak of simply as the cold season. The hot season begins in April, but in March it is already so warm that barley and wheat ripen and are harvested. From April to June, as a rule, there is no rain. The west wind holds sway, and, blowing from the sandy wastes of the Indus region, is a veritable hot wind (Glutwind). A denizen of the temperate zone can hardly realise to himself the desiccating, truly scorching heat of this wind. When exposed to it, one may imagine one is facing an open furnace. The thermometer rises in the shade to over 50 C. (122 Fahr.) In order to enjoy fresh air at this season one must take exercise in the early dawn, between 4 and 5 in the morning ; for no sooner has the sun risen than the heat sets in again. After 7 A.M., save of necessity, no European leaves his house, and should business oblige him to do so, he must protect himself from the sun with an umbrella and a thick head-covering. ... At sunrise, or soon after 5 A.M., houses must be closed, only a small door being left open for communication with the outside. Thus the house of a European is more like a gloomy prison than an ordinary dwelling- house. So long as the hot winds blow strongly and steadily, rooms may still be kept in some measure cool by means of tatties or grass 128 CLIMATES AND WEATHER OF INDIA screens set up in front of the doorway, and continually sprinkled with water, or by the fan vanes of the so-called ' thermantidote,' which a servant keeps revolving and sprinkles with water ; and at night the punkah is worked. Whoever cannot provide himself with these arti- ficial cooling appliances must suffer the daily torment of insupportable exhausting heat. Man and beast languish and gasp for air, while, even in the house, the thermometer stands day and night between 35 and 45 C. (95 and 113 Fahr.) Little by little the European loses appetite and sleep ; all power and energy forsake him. Vegetation suffers equally ; almost all green things wither ; the grass seems burnt up to the roots ; bushes and trees seem moribund ; the earth is as hard as a paved highway ; the ground is seamed with cracks ; and the whole landscape wears an aspect of barrenness and sadness. At length, in June, the hot winds cease to blow, and are followed by a calm ; and now indeed the heat is truly fearful : tatties and therman- tidotes avail nought ; all things pine for the rains ; but no rain, not even a shower, can one hope for, till the south and east winds shall have set in. And even then, the rains do not extend to the whole of the Punjab ; Lahore has but little rain, Mooltan scarcely any ; and the peasant of the Western Punjab is dependent entirely on artificial irrigation for the watering of his crops. " The southerly and easterly winds bring first clouds and violent storms with heavy rain showers, which are repeated daily, or, at all events, every 2 or 3 days ; and, finally, the rains which, in the Himalaya, set in at the beginning of July and cease at the end of August or in the middle of September. In July the trees begin a second time to burst into leaf ; grass springs up once more, and soon a vegetation is developed that, fostered by warmth and moisture, is scarce to be kept within due bounds. The peasant now works hard at ploughing, sowing, and weeding his fields. Eice is sown in June, during the great heat ; in September it is reaped ; and within 2 months, maize is sown and harvested. . . . "After from 4 to 6 weeks of heavy rain, often falling uninter- ruptedly for 2 or 3 days in succession, it clears up, and sometimes some weeks pass without further rain ; after which, a week or two more of rainy weather bring the season to a close. Grateful as is the coolness brought by these showers, the more oppressively hot and sultry is it, when the rain ceases and holds off, if only for half a day. The atmosphere weighs on one like a heavy coverlet ; and then comes the daily and nightly plague of mosquitoes. Insect and reptilian life is now active ; of evenings it hums and buzzes and croaks all around ; frogs make their way into the house, and with them more serious and unwelcome visitors, scorpions and snakes ; for which reason it is unwise at this time of year to go about in the dark. " One can hardly picture to oneself in our European climate how serious and disagreeable are the effects of excessive moisture, as ex- perienced towards the end of the rains. Woodwork swells, and doors CLIMATES OF THE PLAINS 129 and windows can be fastened only with much difficulty. Shoes and all articles of leather become thickly coated with fungus, books become mouldy and worm-eaten, paper perishes, linen becomes damp in the presses, and despite the oppressive heat, one must often light a fire on the hearth, only to neutralise in some degree the influence of the damp. " The period which immediately follows the rains up to October is the most unhealthy season in the year. Decaying vegetation under an ardent sun generates miasma, the consequences being fever, dysentery, and not unfrequently cholera. Towards the end of the rains one rejoices indeed to see the heavy dark clouds disappear, but the heat soon becomes once more so great that one longs for the cold season, and more than ever turns an anxious eye to the wind vane, watch- ing for some sign of the cool westerly and northerly winds. With the beginning of October these winds set in steadily, clearing the skies, and now the blue firmament appears in all its splendour, so glorious in the torrid zone. . . . From October to Christmas, as a rule, the weather i& clear and fine, the air is pure and most delicious, and one can hardly imagine a more charming climate ; but it must never be forgotten that an Indian sun shines overhead, and that even in the cold season one must never expose the unprotected head to its rays. The European now once more breathes freely, and it is a delight with the head well covered to move about in the open air. Fruit certainly is over, but the European now bethinks himself of his own garden, for it furnishes him with most of the vegetables of Europe, and our loveliest garden flowers bloom and delight his eye with their well-remembered forms. . . . For 5 or 6 weeks Europeans can work vigorously and with pleasure. " In December and January the fire burns all day long on the hearth, and in the morning and evening is especially grateful. The nights are positively cold ; even on the plains, ice and hoar frost form, and near the ground the thermometer sometimes sinks to 5 C. (23 Fahr.) During the second half of the cold season we have in the Punjab a good deal of rain, without which indeed the barley and wheat harvest is but poor ; the pulses also require the winter rains. In February we have a short spring ; many trees unfold their leaves, and every bush furnishes its quota of flowery adornment. But this spring is of short duration, and in March it is already warm on the plains and the hot summer is at hand ; an occasional dust-storm, however, for a while keeps off the summer heat. A dust-storm is indeed in itself unpleasant, the air being so charged with dust as to bring about an Egyptian darkness, no matter what may be the hour of the day." The above description gives a more vivid picture of the chief features of the climate and will better serve to "bring them home to the mind of the English reader than any mere sum- K 130 CLIMATES AND WEATHER OF INDIA mary of the statistical data. Except, perhaps, that the extreme temperatures are slightly overstated, 1 it represents fairly enough the experience of any resident of Delhi or one of the large military cantonments along the northern border of the province, where, during the two months, July and August, the rainfall is such as the writer describes. But in the west and south rain is much more scanty and uncertain, and although the air is sufficiently damp to be very oppressive to the system and to render nugatory the usual artificial appliances for cooling the atmosphere of houses, even the temporary respite of heavy cooling rain is unfrequent throughout the summer monsoon. At Mooltan it rains on an average on but five days altogether in the two months, at Dera Ishmail Khan on seven only, and at Peshawar there are but six rainy days at this season. In 1877 no rain fell at Peshawar from the end of April till the end of Sep- tember, and at Mooltan there were but four rainy days in the six months from May to November, and thirteen only in the whole year. It now remains to notice more particularly the variations at some of the chief stations in different parts of the province. In its extreme north-western corner, on a plain surrounded by mountains and watered by the Cabul river above its junction with the Indus, is the important frontier station Peshawar. The military cantonment of Nowshera is near the eastern extremity of the same plain. This part of the pro- vince enjoys nearly seven months of fairly cool pleasant weather, for it is not until the latter part of April that it begins to be unpleasantly warm in the daytime, and even in that month the average maximum is only 85 in the shade, and the night minimum 60. The most cloudy and rainy season here is not that of the summer monsoon, but from January to April. The rainfall of the former averages only about 4 inches, and, as in 1877, sometimes fails almost completely, while that of the latter is almost double as great. The winter is cloudy and cold, the temperature not infrequently falling to the freez- ing-point before sunrise ; while there is nothing to mitigate the ardour 1 This is probably due to no inaccuracy on the part of the writer. Before 1876 the instruments at the Government observatories were so exposed as to register higher day and summer temperatures, and lower night and winter temperatures than really represented those of the air. See, on this point, the remark on page 15. In 1876 the system was altered, and the data quoted in the present work are taken from the subsequent registers. CLIMATES OF THE PLAINS 131 of the summer sun, and during the three months, June to August, the average afternoon temperature is 100 and upwards. The highest temperatures reached in June are not less than in the more southerly parts of the Punjab, averaging 115, and having been as high as 119 in the shade. Notwithstanding the proximity of the mountains, the atmosphere of Peshawar is as dry as that of Mooltan, in July and August indeed drier, and, needless to remark, the power of the sun is very great in these months. The average rainfall is only 13*5 inches in the year, and cultivation is therefore dependent on irrigation from hill streams and the Cabul river ; from which latter source the city is well supplied. Of the same general character and only less extreme is the climate of Rawalpindi, 90 miles to the east of Peshawar, on the plain above the Salt Range, 1700 feet above the sea, and within a few miles of the outermost hills of the Himalaya. The atmosphere of Rawalpindi is less dry than that of Peshawar at all times of the year, and the rain- fall between twice and three times as great ; that of July and August alone averaging more than 14 inches. There is more rain also in the winter and spring, and on rare occasions snow has been known to fall and accumulate to a depth of 4 inches in and around the station. This part of the Punjab, lying to the west of the Jhelum and north of the Salt Range, is a small plateau, elevated in places as much as 2000 feet above the sea, and there is a marked difference in its climate and vegetation as well as in the general aspect of the country, as compared with the plains to the east of that river. About 110 miles to the south-east of Rawalpindi lies the important military station Sialkot. It is built on high ground to the east of the Chenab, about 15 miles from the foot of the hills. On the mean of the year this station is about 4 warmer than Rawalpindi, owing to the higher temperature of the winter, spring, and autumn, for in the monsoon months there is but- little difference. The monsoon rainfall is about 6 inches heavier, but that of the winter and spring months about 2tjr inches less, and also less frequent. Notwithstanding the somewhat heavier rainfall, the number of rainy days is little more than two-thirds of that at Rawalpindi. This cantonment is one of the healthiest in the Punjab. Lahore, the capital of the province, lies 64 miles almost due south of Sialkot, at a greater distance from the hills and in a drier zone ; indeed, on the border of the desert tract that occupies the high ground of the Bari Doab between the Ravi and the Sutlej. It is 2 or 3 warmer than Sialkot in almost every month of the year, a difference partly due to its more southern position, partly to its drier atmosphere. On the average of 10 years, the highest temperature registered in the year is 117, and this varies but a few degrees from year to year, 120'3 being the highest yet recorded. In the winter, the shaded thermometer but seldom sinks to the freezing-point, 34 being the mean minimum of the year. The air is dry, the mean humidity of 132 CLIMATES AND WEATHER OF INDIA the year being only 50 per cent of saturation, and that of May, the driest month, 33. Both the summer and autumn rainfall is less than that of either Sialkot or Rawalpindi, the annual total being 22 inches, and on an average it rains on but 37 days in the year. The station is well planted with trees and gardens artificially irrigated, but a few miles to the south begins the desert tract occupied by Capparis bush with an occasional babul tree (Acacia Arabic^ which extends to Mooltan ; save where irrigated from the Bari Doab canal. Such is the case, for instance, in the Changa Manga forest, a tract of 32 square miles in extent, about midway between Lahore and Mooltan, where the annual rainfall is under 16 inches. In the course of 20 years this forest, partly consisting of sissoo (Dalbergia sissooj, has at- tained an average height of 40 feet or more, under the influence of a good annual watering from the canal. 90 miles to the east of Lahore, and somewhat farther south, is Ludhiana, 3 miles from the left bank of the Sutlej and 50 miles from the foot of the Himalaya. Although so much farther from the hills than Sialkot, the climate is very similar, the temperature of the two places scarcely differing more than a degree in any month, and the mean temperature of the year being the same. The winter is rather drier and less cloudy, and in most months of the year the rain is rather lighter ; in other respects there is little difference. Delhi, in the extreme south-east of the Punjab, is not less than 130 miles distant from the Himalaya, but the climate, although dry, is much less so than at stations equally distant from the hills in the more westerly parts of the province. The winter is less cool, and the shaded thermometer, during the last 11 years, has never sunk to within 4 of the freezing-point. But the summer temperatures in May and June are nearly as high as in the drier parts of the province, the mean maximum of the year being 116. In July and August, however, the heat is less intense (though scarcely less oppressive), the average afternoon maximum being 4 lower than at Lahore in these months. The winter rains are less frequent and heavy than at stations nearer the hills, while the summer rainfall is nearly the same as at Ludhiana. Although, therefore, the climate of Delhi is somewhat less extreme than that of places farther to the north-west, it is of the same general type. Returning now westwards to the drier parts of the Punjab remote from the hills, we first come to Sirsa, 150 miles west-north-west from Delhi, and about the same distance south by east from Lahore, almost midway between the Jumna and Sutlej, and on the border of the Bickanir desert. Although the mean temperature of this place is only 1 lower than that of Delhi, yet the winter temperature is 3 lower, and the average minimum of the year 6 lower, while, owing to the scanti- ness and uncertainty of the monsoon rainfall, the day temperatures of July and August are 5 higher than at Delhi. The diurnal range of temperature throughout the year is also from 3 to 9 greater. When at its greatest, in November and December, there is a mean CLIMATES OF THE PLAINS 133 difference of 36 and 34 between sunrise and the afternoon. The humidity of the air is lower than at any station yet noticed, the annual mean being only 47 per cent of saturation, that of April and May 36 and 37, and that of July and August only 60. The winter as well as the summer rains are infrequent and precarious, and the annual average is barely 15 inches, of which about half falls in July and August. The climate of Dera Ishmail Khan, in the extreme west of the province, on the right bank of the Indus, is, in some respects, still drier than that of Sirsa. Owing perhaps to the proximity of the river and the irrigation derived from it, the air of the observatory is slightly damper, but this difference is probably purely local. The rainfall is little more than half as great, viz. little over 8 inches in the year, of which no more than 3^ inches fall in July and August. The mean winter temperature is 3 or 4 lower, the average minimum of the year being 31, and in 1878 it fell to 26. On the other hand, the intense summer heat is more prolonged, the afternoon maximum ranging above 100 from May to September, and having once reached 121 '5, which is nearly as high a reading as any yet recorded by a well-shaded verified thermometer in India. 1 Lastly, Mooltan, 115 miles south of Dera Ishmail Khan, and a little above the junction of the Sutlej and Indus, has a still lower and much less frequent rainfall. Rain falls, on an average, on only 15 days in the year, a difference as compared with Dera Ishmail Khan, principally due to the infrequency of winter and spring rain. Both in 1875 and 1883 it rained on only 9 days in the whole year. The mean temperature of Mooltan is about 2 higher, but in the winter it has fallen to the freezing-point in 4 out of the last 11 years. In the early months of the year the rise is very rapid. Already in April the thermometer often rises to 100, and in May the average maximum is 104. June is the hottest month, as in other parts of the Punjab, but July is nearly as hot, and not till the end of October does the afternoon temperature range below 90. Mooltan and Sirsa may be taken as representative of the climate of the Southern Punjab generally. In dryness and rainlessness they are surpassed, and that but little, only by Upper Sind. Sind. The great plain traversed by the Lower Indus from the Punjab frontier to the sea, is at once the driest and, as a whole, the hottest of all the provinces of India. What the Nile is to Egypt that is the Indus to Sind. Wherever the arid soil can be brought under the influence 1 It was exceeded in 1886 at Pachpadra in Rajputana, where 123 '1 was registered on the 25th May, and at Jacobabad, where 122 '2 was recorded on the same day. 134 CLIMATES AND WEATHER OF INDIA of its vivifying waters, there agriculture flourishes, and even extensive forests of babul and tamarisk afford ample sup- plies of small timber, fuel and fodder. But beyond such limits all is a sandy or stony waste, thinly tenanted by a leafless bush of Capparis apfiylla and the gray -leaved Prosopis spicigera, or, as in the Cutchee desert, a broad plain of hard bare clay, void of vegetable life. The aridity which, in a high degree, characterises the whole province, reaches its highest expression in Upper Sind, represented in our registers by the climate of Jacobabad. The atmo- sphere of this place shows a lower degree of humidity and a lower rainfall than even Sirsa, Dera Ishmail Khan, and Mooltan, and it is probable that even this would be found surpassed by that of the Pat or Cutchee desert to the west, since the country around Jacobabad, extending to about 4 miles beyond it on the west, is under field crops, and the station is shaded by magnificent trees fed by irrigation and subsoil infiltration from canals supplied by the river. As in other dry regions the vicissitudes of temperature are very great. Notwithstanding its more southerly latitude, the extreme winter cold is as great as at Mooltan and Sirsa ; indeed rather greater ; and the extreme summer temperature surpasses that of both. Nevertheless, even Upper Sind is not entirely uninfluenced by the dampness of the summer mon- soon. May is still the driest month of the year, and although June is virtually rainless, the south and south-east winds that prevail from June to September, alternating with oppressive calms, bring a slight increase of moisture, lower- ing the temperature a few degrees in July and August. Rain is, of course, very rare. Indeed, the only occasions on which it falls are when, in the summer monsoon, a cyclonic storm travels from the eastward, and reaches Sind before it is broken up ; or when, in the winter, a barometric depres- sion forms in Sind or passes across it from Baluchistan. On such occasions the rainfall is sometimes very heavy, and owing to the stony impervious character of the soil, and the rapid drainage of the hills on its western frontier, heavy floods sometimes sweep over the low country, breaching the CLIMATES OF THE PLAINS 135 railway and causing serious damage. The climate of Lower Sind is slightly damper and less extreme, and the strong west- south-west winds that blow steadily from April to September, although they bring little or no rain, afford the means of ventilating dwelling-houses, which, as may be seen in the native city of Hyderabad, are surmounted by permanent wind-sails of masonry, to catch the current, and divert it through the apartments below. On the coast and for a long distance inland, this south- west and west wind is remarkably strong and steady, blowing throughout June, July, and August with an average rate of from 20 to 25 miles an hour. The same directions prevail, more or less, through the greater part of the year. On the coast, north-east winds preponderate in December, but in no other month; and even at Hyderabad, where north winds prevail from November to January and north-east winds in February, from the middle of March to the end of October the south- west wind is steadily dominant. Its dryness shows, however, that it is only in part a sea wind ; it is doubtless fed to some extent from the dry atmosphere of the Baluchistan highlands, and indeed it is to this source that the climate of Sind, and more particularly Upper Sind, owes its extreme aridity. As already pointed out, this characteristic feature is displayed in the strongest degree by Jacobabad, which stands on the flat alluvial plain to the west of the Indus, 40 miles from its bank, and 65 miles from the Khirthar range of hills on the west, through which the Bolan Pass leads up to the highlands of Baluchistan. The mean temperature of Jacobabad is 2 higher than that of Mooltan, and that of the winter months from 2 to 4 higher ; but this is due to the greater power of the sun in the daytime, since the night temperatures are quite as low on an average, and in extreme cases somewhat lower than at Mooltan. Consequently, the daily range of the thermometer is greater, and aver- ages between 30 and 34 in 7 months of the year. There is but little cloud at any time of year ; more from February to April than in any other month, but rain is very rare in the winter and spring, and the small quantity that falls in July and August, averaging 3 inches, makes up two-thirds of the annual total. Even in these months it rains on an average on only one day in ten, and this is very uncertain and irregular. The climate of Hyderabad is more moderate. The city stands on 136 CLIMATES AND WEATHER OF INDIA an elevated platform of rock, 3 miles to the east of the Indus, com- manding an extensive view over the alluvial flat of the river. It is thus well exposed to the south-west and west winds that sweep across the river, and the air is somewhat less dry than that of Jacobabad in every month except December. The rainfall of July and August is also double as great, but it amounts to only 6 inches on an average, and is equally precarious. In 1880, it rained on only five days during the whole year at both these places. From September to the follow- ing June, both inclusive, the rainfall is insignificant. The winter temperatures are from 4 to 6 higher than at Jacobabad, and the summer temperatures 5 or 6 lower. The difference of the summer day temperatures alone is even greater, amounting to not less than 8 in July and August. Lastly, Kurrachee, within 3 miles of the sea, is a cooler station than either of the foregoing. Owing to its more southerly position, as well perhaps as its proximity to the sea, the winter mean temperature is 2 or 3 higher than at Hyderabad, and the night temperature, even in January, has never fallen to 41 in the last 11 years. In the summer a maximum reading of 117*6 has been registered, which is nearly as high as the highest at Hyderabad, but this was quite excep- tional, and in most years, 102 to 106 is the highest reached ; tem- peratures which are about the same as the highest of Calcutta. The mean afternoon maximum of the hottest months, May and June, is only 93, those of Hyderabad being 106 and 103, and those of Jacobabad 108 and 111. The strong sea wind which, during the greater part of the year, blows night and day at Kurrachee, much mitigates the heat, and although somewhat relaxing, makes it far superior as a place of residence to other stations in Sind. This wind, however, brings but little rain ; the annual average of the place, 7*8 inches, being slightly less than that of Hyderabad ; and, as in other parts of Sind, trees will not grow unless artificially watered. Rajputana. Both geographically and in point of climate Eajputana is transitional between the very -dry, almost rainless, valley of the Lower Indus and the plateau of Central India, where agriculture can be carried on ex- tensively and regularly, at least in the summer months, without the aid of artificial irrigation. It consists of a con- geries of native (and one small British) states, the more westerly of which, Bickanir, Jeysulmere, and Jodhpur, are included in what is known as the great Indian desert. The rainfall of this region varies from only 5 or 6 inches in the extreme west to a little under 20 inches in the year along the Aravali hills, and, being very precarious, is insufficient to ensure successful CLIMATES OF THE PLAINS 137 cultivation. For water for domestic purposes and the supply of their cattle, the population is in a great measure dependent on deep wells. In describing this region, Mr. W. T. BlanTord remarks 1 that "the name 'Great Desert' usually conferred on it, conveys an imperfect idea, because the tract of country is neither barren nor uninhabited. It is covered with shrubs and bushes in general, and in places small trees are found. Moreover, though the population is thin, villages are scattered throughout, and immense herds of camels, cattle, sheep, and goats are kept and pastured. The desert is, in fact, a great sandy tract entirely destitute of streams of water and with but few hills of rock, and a large portion of its surface consists of sandhills of consider- able height. When rain falls, crops of bajri (Holcus spied) are raised. When rain fails, the population lives princi- pally on the milk of cattle and on imported grain." The boundary between this and the eastern states, Jeypore, Udaipur, and a number of small states, is the Aravali range, a series of rocky ridges rather than a continuous range, which begins in the historic ridge of Delhi and runs to the south-west past Jeypore and Ajmere to Mount Abu. The country to the east of this is the north-western portion of the tableland extending between the Jumna and the Nerbudda, the remainder of which will be noticed under the heading of Central India. Everywhere to the east of the Aravalis the rainfall is generally over 20 inches in the year ; in the extreme south, in the little state of Banswarra, over 40 inches ; and in the hills, wherever protected from the jungle fires, so destructive of all arborial vegetation, exist extensive tracts of forest, described by Sir D. Brandis as consisting of such trees as Anogeissus pendula, Butea fron- dosa, Bomlax mcdabaricum, Acacia catechu, Acacia Senegal, Dendrocalamus strictus and Tecoma undulata, with a number of bushy shrubs, such as Grewia populifolia, two species of Balsamodendron, Balanites Eoxburgliii, Salvadora Persica, Cordia Rotliii, Calotropis procera, and Eeptonia luxifolia. Indeed, much of the preserved forest around 1 Journ. As. Soc. Bengal, Vol. xlv., 1876, Part ii. p. 89. 138 CLIMATES AND WEATHER OF INDIA Ajmere consists of bushes rather than trees. As in the Punjab, two series of crops are sown and reaped yearly ; the summer or rain crops, consisting, as also there, of millets, cotton, and beans ; the winter crops of wheat, barley, gram (Cicer arietinum), vetches, lentils, and peas. In the Indore State the opium poppy is also extensively grown. All through Eajputana the winter rainfall is lighter and less regular than in the Punjab, and in general the crops of that season require some artificial irrigation. The division of the year into a hot season, a rainy season, and a cold season, holds good in Eajputana as in the Punjab ; but the first and last are less strongly contrasted. The extreme summer temperatures are indeed but little below the highest registered in the Punjab, but the winter cold is less ; the latter season is less cloudy and rainy, and even on the high ground of Eastern Eajputana, in but few places does the shaded thermometer ever sink to the freezing- point. In this part of the province, the rains, which set in in the latter part of June, bring about a rapid and consi- derable reduction of the heat, and the westerly wind that blows through the summer months from the coasts of Sind and Cutch affords an atmosphere less intolerable to the human system than the close, oppressive, and almost stagnant air then prevailing in the Punjab. The climate is illustrated in the Appendix by the meteoro- logical statistics of four stations, viz. Bickanir, which repre- sents the northern part of the desert tract ; Jeypore and Ajmere just to the east of the Aravalis, and Deesa, which, though administratively in the Bombay Presidency, is close to the southern border of Western Eajputana and, except that the rainfall is rather higher, may be taken as repre- sentative of that portion of the province. Neemuch also, in Central India, has a climate very similar to that of Southern Eajputana lying east of the Aravalis. The climate of Bickanir is drier, and on the average of the year apparently 4 warmer than that of Sirsa, and the rainfall is 2 inches less ; while the average number of rainy days is rather greater. In the dry season, from October to May, the mean temperature in CLIMATES OF THE PLAINS 139 many months is as much, as 6 higher, and the mean night temperature from 7 to 11 higher. Such at least are the results shown by the registers ; but the station is one of the least accessible in India, and it is possible, though there are no independent grounds for such an assump- tion, that the record may be affected by some undetected cause of error. Ajmere and Jeypore are respectively 6 and 3 cooler than Bickanir, there -being a mean difference of 3 between them, although they are less than 90 miles apart, and their difference of elevation is less than 200 feet. It is not clear to what circumstance this is due, unless indeed to the proximity of the large Anasagar tank and also of considerable hills, from which air cooled by radiation descends at night on the station of Ajmere ; for it is noteworthy that the greatest differ- ences between the two stations is in the night temperatures. The difference of the rainfall is little more than 1 inch. The mean humidity of the air is about the same at both ; but while Jeypore is much the drier station in October and November, and still more so in April and May, it is very much damper in July. During the height of the monsoon, in July and August, it rains on about two days in five, on an average, at Jeypore and Ajmere ; at Bickanir only on about one day in four. In the south-east, in Banswarra, the rainfall of this season is heavier and more frequent. Except in the rains, the climate of Deesa is much drier than either of the above stations, as dry as that of Western Rajputana, the average humidity of the year being only 46 per cent of saturation ; but the monsoon rains are rather heavier than at either Jeypore or Ajmere. The winter months are much less cool, owing partly to its lower elevation, partly to its more southern latitude ; the mean temperature of January being 67, while it is only 58 at Ajmere and 61 at Jeypore. The difference of their summer temperatures is not more than 1 or 2 in any month. The North- West Provinces and Oudh. The Gangetic Plain. These provinces include the whole of the alluvial plain of the Ganges and Jumna, eastwards from the latter river to the boundary of the Bengal Province of Behar. Together with Lower Bengal and Behar, they constitute the most fertile and highly-cultivated and also the most densely- populated provinces of India. Traversed by the great tributaries of the Ganges, two of which, the Gogra and the Gandak, rival in magnitude the parent stream and, like it, bring down the drainage of the Himalaya, depositing from their waters the silt that constitutes the soil of these plains, there would be no portion of them unfitted for agriculture, 140 CLIMATES AND WEATHER OF INDIA were it not that, in certain of the drier tracts, salts left by the constant evaporation of the subsoil waters have accumu- lated to such an extent as to be destructive of all vegeta- tion, save only a coarse grass and a few worthless plants that thrive on saline soil. This class of soils, known by the local names of reh and usar, exists only where the ground is too high to be reached by the flooded rivers, and where, at the same time, the rainfall is too scanty to dissolve the saline efflorescence and carry it away in the surface drainage. With the exception of such tracts and the strip along the foot of the Himalaya, where the more highly sloping marginal zone consists of pebbles and coarse gravel with a filling of sand and earth, the whole of the vast plain is highly productive arable land. The sub -montane zone, known as the Ihabar, is essentially a forest tract, the chief home of the sal tree (Shorea robusta) which furnishes one of the most valuable timbers of Northern India. Although, in the opinion of Indian geologists, there is no reason to question that the whole of the Gangetic plain has been formed by the rivers in past ages with materials washed down from the hills, very much of it is now far above the reach of the highest floods, and in its turn is being eroded and its waste carried down to lower levels. Such portions of the plain are known as Ihcingar land, and except where they can be irrigated from canals or local wells, are suited only for dry crops. The lower grounds, termed Khadar lands, which flank the present rivers, lying in the depressions cut by the latter in the more ancient lhangar, are more or less subject to inundation, and are especially, though by no means exclusively, the rice lands of the pro- vince. As in the Punjab and Eajputana, so also in the North-west Provinces, the summer or Kliarif crops consist of millets, rice, oil seeds, sugar cane, and cotton, the winter or rabi crops of wheat, barley, pulse, gram, and, especially in the eastern districts, of the opium poppy, which is largely cultivated both here and in the adjacent province of Behar. Tobacco, safflower, mustard, and indigo, together with many kinds of vegetables, among which cabbages and cauliflowers CLIMATES OF THE PLAINS 141 rival or surpass in size those ordinarily raised in their European home, are also among the field or garden produce. But in fact there is scarcely any agricultural product of the summer of the temperate zone that cannot be raised on this fertile plain, which, besides being in great part irrigable from wells or from canals led off from the upper waters of the great rivers, has for the most part and in most seasons a rainfall varying from 30 to 50 inches. Only in parts of the doab, the tongue of land that separates the Jumna from the Ganges, and in a portion of Oudh to the north of the latter river, does the average rainfall fall below 30 inches in the year ; and it is in this portion of the province that the reh soil mentioned above is chiefly found. It is, however, a meteorological peculiarity of the North- west Provinces, that, more than any other part of India, with the sole exception of the arid tract farther west, it is peculiarly subject to vicissitudes of the rainfall. Not to speak of the great famines of the last century, in 1804, 1834, 1838, 1861, 1868, and 1877, these provinces, or- dinarily so fertile, have been repeatedly devastated by famine, resulting from the failure of the seasonal rains ; and what- ever may have been the fate of other portions of the plain, the Gangetic doab has invariably been involved in the disaster in its most severe form. It was as a protection to this precarious tract that, after the destructive famine of 1834, the Ganges canal, the greatest work of the kind in India, and indeed in the world, was planned and executed under the direction of Sir Proby Cantley. The trunk canal commenced in 1848 was completed and opened in 1854; and although the districts which it supplies have since suffered from seasons of drought, almost if not quite as severe as that of 1834, in no year has the resulting suffer- ing approached in intensity that of that memorable year. The climate of the North-west Provinces differs from that of the Punjab in many important particulars. The cool season is less cold, less rainy and cloudy, and comes to an end in March, when strong hot winds set in from the west with great persistence, lasting well into May. These 142 CLIMATES AND WEATHER OF INDIA winds are diurnal, beginning 2 or 3 hours before noon, blowing through the heat of the day and lulling towards the evening or after sundown. They are intensely dry, a humidity as low as 6 per cent having sometimes been in- dicated by the hygrometer ; but they are very healthy, and indeed the season when they prevail is, to acclimatised Europeans as well as to natives, the healthiest in the year. The heat is greatest at the end of May or the beginning of June ; when, on certain days, the thermometer reaches as high a point as at most of the Punjab stations. As the hot winds flag, the clouds gather more frequently, and occasion- ally the heat is temporarily relieved by an afternoon dust- storm. Sometimes, and especially in the eastern districts and near the hills, the dust-storm terminates in rain ; and in some seasons, after a light easterly wind for a day or two has held possession of the plain, one or two days of steady rain follow. This is known as the chJwti larsat or " little rains." It has, however, by no means the regularity frequently attributed to it in the popular belief. The rains set in some time in the latter half of June ; rarely earlier. In some years they begin simultaneously all over the province ; in others, first of all in the eastern dis- tricts. While they last, it rains more often than in the Punjab ; but rain is by no means continuous for more than a day or two at a time, and these rainy spells are separated by intervals of finer but steamy and oppressive weather. At Benares there are, on an average, 2 6 rainless days in the two months July and August ; and in the drier parts of the province, as at Agra, rain falls on only about one day in three. In most seasons there occur intervals of a week or more during which rain is almost or quite suspended, such interruptions being termed "breaks" in the rains, and in certain years these breaks last for several w T eeks, and west- erly winds set in, less hot and dry but more oppressive than the hot winds of the spring. In 1877 such was the case almost throughout the summer monsoon. Save, how- ever, in such exceptional seasons of drought, the temperature of July and August is several degrees lower than in the Punjab. CLIMATES OF THE PLAINS 143 The rains cease, as a rule, in September, earlier or later in different years. Generally they last a week or more longer in the eastern than in the western districts. A few weeks of close and warmer weather follow, but under the clear skies of the lengthening nights the temperature gradu- ally falls ; and if, as sometimes happens, a late and final fall of rain comes at the end of the month or in October, its cooling effect is rapid and permanent. Light airs begin to move from the west and gradually strengthen till they become the steady cool wind of the winter months. As a rule, the weather remains fine and clear, with cloudless skies, till near Christmas, when the heavens become covered with a sheet of filmy cloud, and after some days of rather close weather, and either a still atmosphere or light southerly and easterly winds, the first rain of the cold weather falls, followed by a great fall of temperature and a stronger and cooler wind from the west. These falls may be repeated once or twice or oftener in January and February, but they are not lasting, and the rain is lighter than that of the summer months. As already mentioned, the rainfall of this season is less frequent than in the Punjab, except on the hills of Kumaon and Garhwal. The meteorological statistics of seven stations are given in the Appendix, illustrative of the climate of the North-west Provinces and Oudh. The first of these, Dehra, lies close to the foot of the Himalaya, in the north-west corner of the province, or rather between the Himalaya and a range of low hills, the Siwaliks, which run parallel to it at a distance of 8 miles, constituting a sort of outwork to the main chain. The intervening strip of plain, 2000 feet or more above the sea, is known as the Dehra Dun, and enjoys a more temperate climate than the great plain to the south of the Siwaliks. Tea is grown here in tracts cleared from the forest, and some Europeans have chosen it as a per- manent place of residence. The annual rainfall is as much as 76 inches, two-thirds of which falls in the months of July and August. The mean annual temperature is 71, that of June 84, and the average afternoon temperature in May and June 93 and 94, and at the annual maximum 105, which is fully 10 lower than on most parts of the western half of the Gangetic plain. From December to February the mean of the day is from 55 to 57. 144 CLIMATES AND WEATHER OF INDIA Koorkee lies 15 miles south of the Siwaliks, and nearly 900 feet above the sea. It is the headquarters of the engineering staff of the Ganges canal. The mean annual temperature is 75, and the average rainfall 42 inches, which is higher than on the plains farther south. Meerut is a large military and civil station on the Gangetic doab, 40 miles north-east of Delhi, and 75 miles from the hills ; with a rainfall of only 28 inches in the year, and a mean annual temperature of 76 ; and Agra, 160 miles farther south, on the high southern bank of the Jumna, has again a slightly lower rainfall, 26 inches, and a mean, temperature of 79. These two stations represent the climate of the drier western districts. The next station, Lucknow, stands in the very heart of the Gan- getic plain, about 100 miles from the Himalaya, and at an equal distance from the high ground south of the Jumna. It is an im- portant military and civil station, and the former capital of Oudh. It is warmer than any of the foregoing, except Agra, the mean annual temperature being 78, but the rainfall is higher, viz. 37 inches. Allahabad, the seat of the local government, is again farther east, at the confluence of the Ganges and Jumna. The European part of the station is on high bhangar land, and the climate is but little affected by the proximity of the two great rivers. The mean temperature is the same as that of Lucknow, 78, but the highest temperature reached in the year appears to be a little higher, viz. 116. In the winter it never sinks to the freezing-point. The average rainfall is between 37 and 38 inches. Finally, Benares, 75 miles farther east, and not far from the eastern boundary of the province, also situated on the bank of the Ganges, has a very similar climate, and a slightly higher rainfall, viz. 3 9 '6 inches. The Central Indian Plateau. The high ground that extends to the south of the Jumna and Ganges, almost from the banks of these rivers to the valleys of the Nerbudda and Sone, includes, besides the native states of the Central Indian Agency, the districts of Jhansi and Lalitpur, under the government of the North-west Provinces, and that part of the Central Provinces formerly known as the Saugor and Nerbudda territories. It forms a low tableland, sloping from south to north, and terminating in the former direc- tion, abruptly, within a few miles of the Nerbudda, in the escarpment which, in modern maps, bears the name of the Vindhyan range. The eastern half consists of sandstones and other sedimentary rocks, the successive outcrops of which form a series of parallel ranges running north-east and south-west ; while the western portion is covered with CLIMATES OF THE PLAINS 145 a sheet of volcanic rock, the northern extension of the great trap formation of the Deccan. Between them, the northern portion of the plateau, in Bandelkhand, consists of an ex- tensive exposure of crystalline rocks. The highest part of the plateau is the south-western, in Malwa, where it is more than 2000 feet above the sea, and where the productive black soil that covers the basaltic rocks yields rich crops of wheat and the opium poppy. Of the drug manufactured from this last, Malwa is one of the most productive pro- vinces in India. The climate of the Central Indian tableland is drier than that of the Gangetic plain, and except from the begin- ning of April to the setting in of the rainy season in June, the more elevated tracts are comparatively cool. In these months its surface is swept very persistently by dry hot west winds ; but, so soon as the rains set in, the temperature is greatly reduced, and although the rainfall is moderate (between 30 and 50 inches) the atmosphere is by no means so still and oppressive as in most parts of the Gangetic plain. To acclimatised Europeans, such stations as Saugor, Jubbulpore, and Indore are not unpleasant places of re- sidence during the greater part of the year. The cool season lasts about four months, from the beginning of Novem- ber to the beginning of March. Less cold than in the Punjab, it is also less damp and cloudy ; and rain is in- frequent, there being not more than one or two rainy days on an average in the month. As a rule, bright, clear, pleasant weather lasts uninterruptedly for weeks together at this season. In these cool months, the eastern part of the plateau, the tract between Jubbulpore and Benares, is cooler than the central and western districts, owing, as it would seem, to the greater prevalence of north-east winds. From November to March the difference of temperature between Jubbulpore, to the north of the Satpura range, and Nagpur, to the south, amounts to as much as 6 or 7, the distance between them being not more than 160 miles, and the difference of elevation little more than 300 feet. The meteorological statistics of six stations on the Central 146 CLIMATES AND WEATHER OF INDIA Indian plateau are given in the Appendix, in illustration of its climate, viz. Neemuch and Indore in the west, Jhansi and Saugor in the central, and Sutna and Jubbulpore in the eastern division of the area. Politically, Neemuch, Indore, and Sutna are in the native states of Gwalior, Indore, and Eewah, Jhansi in the North-west Provinces, and Saugor and Jubbulpore in the northern districts of the Central Provinces. Neemuch and Indore are respectively situated 1600 and 1800 feet above the sea, and have mean annual temperatures of 75 and 74, and an annual range of 73 and 68. The temperature of the coolest month is 62, that of the warmest, May, 87 or 88. In the winter, the temperature rarely falls below 40, and 108 and 111 are the average highest readings of the year. These are from 4 to 7 lower than those reached at stations on the Gangetic plain, farther north. In the rains, the mean morning temperatures range from 71 to 74 in different months, and those of the afternoon are 83 or 84 at Indore, 85 to 87 at Neemuch. The average annual rainfall is 33 inches at Neemuch, 36 inches at Indore, and the number of rainy days 57 at the former and 76 at the latter station. Jhansi, although more northerly than either of the above, but at a lower elevation (850 feet), being surrounded by numerous rocky hills which absorb the sun's heat by day and radiate it at night, is a much warmer station, especially in the hot season. The mean tem- perature of the year is 79, that of January 63, and that of May, the warmest month, 95. In June, the morning minimum averages 84, and the afternoon maximum 104. In. May they are respectively 81 and 107. The highest reading of the year averages 115, the lowest 43, and the highest yet recorded is 117'9. The annual rainfall is about 36 inches. Saugor is a more elevated station, 1750 feet above the sea, but in a slight depression, on the margin of a large artificial tank, which probably aids in mitigating the temperature in the hot season. The mean temperature of the year is 76 ; that of January 63, and that of May 8.9 ; therefore 6 lower than that of the same month at Jhansi. The morning minimum in June is 7 lower. The rainfall is 46 inches in the year and the average number of rainy days 70. The mean humidity of the air is only 50 per cent of saturation, and that of the driest months, April and May, only 28. Sutna is a small station on the railway between Allahabad and Jubbulpore, and except as the headquarters of the Political Agency of the eastern states of Central India, has little importance. It is warmer than Saugor from May to October, being at a lower elevation (1040 feet), but cooler in the winter and early spring, owing to the circumstances already noticed as affecting the eastern portion of the plateau. CLIMATES OF THE PLAINS 147 Finally, Jubbulpore is an important station at the junction of the East India and Great Peninsula railways, 1340 feet above the sea, and with a considerable European population. Although there are some low rocky hills round about it, it is open to the winds from all quarters, and is comparatively a cool station. The mean temperature of the year is 75, that of December 61, and that of May 90. The highest in the year varies between 106 and 113 ; the lowest between 32 and 38. The rainfall is higher than at stations to the north and west, viz. nearly 54 inches in the year ; and in July, the wettest month, it rains on an average on 22 days. In the whole year there are 80 rainy days. Here, as in most parts of Central India, the rains last from the middle of June to the middle or latter part of September, but there is little rain from November to May inclusive. In the latter month and June an occasional thunder-storm precedes the rains. Jubbulpore has a damper atmosphere than any of the places previously noticed, the mean humidity being 57 per cent of saturation. This difference manifests itself at all times of year, but especially in the cool season. It is not due to any peculiarity in the position of the observatory, which stands on an open grassy plain, not in the neighbourhood of any swamp or water expanse, and has been, ever since its establishment, under the competent and careful superintendence of Dr. Bice, now Deputy Surgeon-General of the North-west Provinces. Behar and Chutia Nagpur. Behar includes that por- tion of the Gangetic plain which lies between the North- west Provinces and Bengal ; including a considerable alluvial tract to the south of the river, which is crossed by its great southern tributary, the Sone, and a tract of high land be- tween this last and the northern half of the Gangetic delta. Towards the south, this rises gradually to the Chutia Nagpur plateau, which is virtually an extension of the belt of hilly country that stretches across India south of the Sone and Nerbudda, and is generally known as the Satpura range. At Hazaribagh and Eanchi, the level of the plateau is about 2000 feet above the sea, and has a few hills of much greater elevation, the most prominent of which is Parasnath to the east of Hazaribagh. Behar is a rich and densely -peopled province, highly cultivated and enjoying a higher average rainfall than the more westerly extension of the Gangetic plain, the part already described under the North-west Provinces and Oudh. In addition to the ordinary crops, such as wheat, rice, millet, pulses, oil seeds, etc., it is the chief seat of the indigo cultiva- 148 CLIMATES AND WEATHER OF INDIA tion, and, in Bengal, of the opium poppy, from which the drug is manufactured at the Government manufactory at Patna. Besides many smaller streams, it is traversed by three great Himalayan tributaries of the Ganges, the Gandak, the Gogra, and the Kosi, and to the south of the main river by its great Central Indian tributary the Sone ; and in most years the water supplies from these rivers and the rainfall render Behar one of the richest agricultural provinces in India. But, like the North- west Provinces, it is subject to the occasional failure of the rains, and in such seasons it has frequently been the seat of disastrous dearth and famine; the more disastrous, owing to the density of the population which its very fertility has fostered. The climate of Behar is more akin to that of the N orth-west Provinces than to that of Lower Bengal, but damper, especi- ally in the eastern districts, and to a certain extent transi- tional between the two. The winter months are cool and bracing, and the rainfall of this season is less than on the plain farther west. Hot west winds set in in March, and last into April and sometimes into May ; but in April and May light damp easterly winds blow intermittently to the north of the Ganges, more particularly in the neighbour- hood of the hills ; and thunder-storms accompanied with heavy rain, and sometimes with hail, take the place of the rainless or almost rainless dust-storms of the Upper Pro- vinces at this season. The rains set in in June, about the same time as in Lower Bengal, and last to the end of Sep- tember or the beginning of October, in which month the rainfall of the province amounts, on an average, to between 2 and 3 inches, and to upwards of 4 inches in the most easterly districts and in the neighbourhood of the northern hills. The Chutia Nagpur plateau bears much the same re- lation to Northern Behar as does that of Central India to the North-west Provinces and Oudh. Less than half a century ago it was a wild forest-clad tract thinly peopled by abo- riginal tribes ; but it includes all the valuable coalfields of Bengal, and the opening up of these, together with the ex- CLIMATES OF THE PLAINS 149 tension of the railway, and the spread of cultivation which has followed in the track of the pioneer of commerce, has now effected the denudation of all but the more remote and hilly parts of the province, and all over the eastern and northern parts of the plateau, all large timber has long since disappeared. In many parts there remains little but an occasional sacred grove of large trees, and patches of sal coppice, intermingled with trees of less value as timber, but some of which yield other valuable products, such as the flowers of the Mahowa (Bassia latifolia), largely used by the natives as food and for the distillation of a weak and nauseous spirit ; also lac gum and dye, and the cocoons of the tusseh silk moth. The field produce of Chutia Nagpur is similar to that of the highlands farther west, wheat being largely grown in the cold season. Chutia Nagpur has a more copious rainfall than most parts of Central India ; in general, but little short of 5 inches, and in some places exceeding that amount. But it is swept by dry west winds in the spring months, and hence the attempts that have been made to cultivate the tea plant in the neighbourhood of Hazaribagh have not been attended with success. The higher parts of the plateau enjoy a dry and pleasant climate ; and, in the rainy season, Hazaribagh, were it more accessible and provided with the requisite accommodation, would afford a grateful place of temporary refuge from the saturated and oppressive atmosphere of Lower Bengal. In the Appendix are given the statistics of temperature, humidity and rainfall of two stations in Northern Behar, two in Southern Behar, and one in Chutia Nagpur, viz. Durbhanga and Purneah on the alluvial plain north of the Ganges, Patna on the southern bank of the river, Gya on somewhat higher and rocky ground between the Sone and the uplands of Chutia Nagpur, and Hazaribagh on one of the highest parts of the Chutia Nagpur plateau. In point of temperature, Durbhanga and Purneah are very similar. They have a mean annual temperature of 77; in January of 62, and from April to September a nearly uniform mean temperature between 150 CLIMATES AND WEATHER OF INDIA 83 and 85, but with, a greater range in April and May than in the subsequent months. At Purneah, but not at Durbhanga, April is on the whole the warmest month, the heat of May being slightly mitigated by the damp easterly winds that are felt intermittently several weeks before the rains, and bring an occasional thunder-storm. But the highest afternoon temperatures are generally reached in May, during intervals of fine weather. The mean maximum of the year is 105, and no reading so high as 110 has been recorded at either station during the last 9 or 10 years. The mean lowest of the year is 45 at Durbhanga and 40 at Purneah ; a difference probably depend- ing on some local peculiarity of position, possibly on the fact that the Durbhanga observatory is in the immediate neighbourhood of a large tank, while that of Purneah is far from any water expanse. It may be owing to the same fact that the diurnal range of temperature at the Durbhanga observatory, at all times of the year, is several degrees less than at Purneah. The mean annual humidity, however, is the same at both places, viz. 71 per cent of saturation. The rainfall of Purneah, owing to its more easterly position, is one-fourth greater than that of Durbhanga, the former being 64^ inches, the latter 47^. As already remarked, the rainy season begins earlier and ends some weeks later than in the Upper Provinces, and the September rainfall is but little if at all less than that of July and August. From the middle of October to the end of May, at Durbhanga, the average monthly rainfall varies from O'l to 0*6 inch only ; but at Purneah the fall of April already exceeds 1 inch, and that of May averages between 3 and 4 inches, being distributed over 6 rainy days. On the mean of the year it rains on 66 days at Durbhanga, on 88 at Purneah, and in the wettest of the last 15 years there were 86 rainy days at the former and 112 at the latter. In respect of rainfall, therefore, there is a considerable difference between the eastern and western districts of Behar. Patna, which lies somewhat farther west than Durbhanga and more to the south, has a drier climate than either of the above stations and a greater diurnal range of temperature. The mean temperature of the year is 78; that of January 61, and that of May 89. In both April and May the highest ' afternoon readings average 100 or more, and the highest of the year has varied between 107 and 113 during the last 11 years. In the coldest season it has not fallen below 36. The rainfall averages nearly 43 inches in the year, which is some- what lower than that of the plain north of the Ganges. July is the wettest month, and November and December the finest. The Sep- tember rainfall is under 8 inches, that of October between 2 and 3 inches. On an average it rains on 71 days in the year. Gya, owing perhaps to its position on high ground in the neigh- bourhood of some rocky hills, is a very warm station. Its mean annual temperature is 79; that of January 64, and that of May 92, or 3 higher than that of Patna, only 60 miles farther north. The CLIMATES OF THE PLAINS 151 average afternoon temperature of this latter month is 104, and the highest of the year has varied from 109 to 116. The lowest read- ing recorded has been but little under 40. The average rainfall is only 41 inches, and its distribution throughout the year is nearly the same as at Patna, five-sixths of the whole falling between June and September and between 2 and 3 inches in October ; after \vhich follow two and a half months of almost rainless weather. In January and February the winter rains of Northern India contribute about \\ inches, after which follow March and April with dry winds and a smaller rainfall. In May the average fall is rather over 1 inch, in June over 6 inches, and the heaviest (nearly 12 inches) is in July. The average number of rainy days is 75 in the year. Hazaribagh, at 2000 feet above the sea, is much cooler, the mean temperature of the year being only 74, or 5 lower than at Gya. In December and January it is 61, the mean of the minima 50 and 51, and the lowest reading in the year varies between 39 and 46. The mean temperature of May, the hottest month, is only 85, the same as at Durbhanga, and the highest in the year varies from 102 to 109. During the spring months, hot west winds blow at Hazaribagh as they do at stations at lower levels, but, so soon as the rains set in, the after- noon temperature falls 13 or 14, and from that time forth the mean temperature of the day averages no more than 78, while the move- ment of the air across the plateau renders the station cool and pleasant, affording a great contrast to the comparatively stagnant and oppressive atmosphere of the lower plains, and especially of Bengal. The rainfall is moderate, amounting to about 13 inches in each of the months, July and August, and about 8 inches in June and September. In October the temperature falls to 74 and the morning minimum to 66, and from this time to March the climate is very pleasant and healthy. Bengal and Orissa. These provinces include the whole of the alluvial plain that skirts the northern shore of the Bay of Bengal, that of Bengal being the combined deltas of the Ganges and Brahmaputra, together with the extension of the plain up to the foot of the northern mountains and the gates of Assam ; that of Orissa the deltas of the Mahanadi and of a number of smaller rivers that drain the highlands of Chutia Nagpur. Intersected by innumerable river chan- nels and abounding in swamps, while open to the damp winds from the bay, which begin to blow on the coast as early as February and gradually penetrate farther inland with the increasing heat, the climate of Bengal is as charac- teristically damp and relaxing as that of North- Western 152 CLIMATES AND WEATHER OF INDIA India is the reverse. The dry westerly winds that play so great a part in the meteorology of the Upper Provinces are felt only occasionally and intermittently in the province of Bengal, during the spring months and chiefly in the warmest hours of the day, and even then with a reduced temperature and of a less parching character, owing to the moisture taken up from the surface over which they blow. The customary division of the year into three seasons, the cool season, the hot season, and the rains, holds good in Bengal as in the more westerly provinces, but the first is shorter and less bracing, and the heat of the second, if less intense, owing to the greater dampness of the air, is on this account, perhaps, more trying to the European consti- tution. The rains are also longer and more copious. At Calcutta, the cool weather scarcely sets in before the second week of November, and lasts only to the middle or latter part of February. For ten or twelve weeks the weather is delightful, pleasantly cool in the daytime, and in the evening sufficiently cold to make a fire agreeable, though perhaps hardly indispensable to comfort. But the damp- ness of the climate manifests itself at night in frequent fogs on the river and the low grounds around ; and in the native part of the town and the crowded suburbs, the smoke from the huts condenses the moisture of the air and enshrouds the habitations in a veil of fog that remains suspended in the still atmosphere. About the end of February the days begin to be appreciably warmer, the sun's power greater and his glare more intense, and in March it is sufficiently warm in the house to require that the punkah be set in movement once more after its three months' holiday. But the nights, or at least the hours after midnight, are still comparatively cool, and as the air is as yet not highly charged with moisture, to an acclimatised European the heat is by no means oppressive. During this month and April the rise of temperature is very rapid, and by the end of the latter month the afternoon readings of the thermometer are often as high as any recorded during the year. In December any rain is exceptional. In January and CLIMATES OF THE PLAINS 153 February it occurs less regularly and frequently than in Upper India, but there are generally two or three rainy days in these months, preceded by some days of close cloudy weather with light southerly winds ; and when the weather clears a great fall of temperature follows and a renewal of the cold season. In March and April rain becomes more frequent, but in the shape of the thunder squalls known as nor'-westers which, as a rule, come on towards the evening, and are often preceded by a dust-storm. Their general character has been described in a previous part of this work. For a graphic but perhaps somewhat exaggerated account of one of these little storms the reader may be referred to M. Eousselet's well-known work on India, where, however, the writer appears to be under the mistaken impression that he had experienced a cyclone. These squalls are more frequent in May, and sometimes usher in a day or two of continuous rain ; but, more frequently, an hour or two expends their force and they are followed by a refreshing night and a somewhat cooler day. Now and then, at intervals of some years, a cyclone may pass over Bengal in May or the early part of June, but these more formidable visitors are perhaps rather to be expected at the end than at the beginning of the monsoon, and happily, at any season, are to be reckoned as meteorological rarities in Bengal. An advantage which Calcutta enjoys in the hot season, and which is denied to places much farther inland, is the southerly breeze which, at the close of a hot day, often blows up from the wide estuary of the Hooghly, bringing some hours of agreeable relief. It does not last beyond midnight, but it mitigates the heat of the dinner hour, and renders the southern verandahs of the large Chowringhee houses pleasant places for the post-prandial lounge. At length, in the early part of June, the clouds gather more thickly, while the barometer falls to a lower point than it has reached since the beginning of the year ; and in the first or second week, heavy and continuous rain ushers in the monsoon. This first burst of the rains usually ac- companies a cyclonic storm, formed either at the head of 154 CLIMATES AND WEATHER OF INDIA the bay or over the delta itself. As has been explained in a previous part of this work, such storms are not attended with very strong winds, at least on land, though the weather may be stormy at sea ; and the barometric depression at their centre does not exceed two or three-tenths of an inch. The first onset generally carries the rains to the greater part of Bengal, and sometimes, but not always, to Behar. As a rule, the rainfall does not penetrate to the Upper Provinces till some days or even weeks later. Its im- mediate effect is a great fall of the day temperature ; and the comparative coolness, supervening on many weeks of close oppressive weather, brings a sense of relief. Bursts of rain of a similar character, alternating with sporadic showers and an occasional rainless interval, rarely lasting more than a day or two, follow in succession through July and August. The air is saturated with vapour ; vegetation grows apace, and indoors and out of doors every absorbent material reeks with moisture ; but so long as the rainfall is abundant, and the intervals of its suspension short, the climate, if not exactly pleasant, is not very oppressive nor notably unhealthy. When, however, in September, the rainless intervals become longer, and the day temperature begins to rise, while the air, still highly charged with moisture, is almost motionless, the relaxed energy of the human system fairly rebels against this further trial of its endurance, and all who are not compelled by their avocations to remain at their post hasten to escape to the temporary refuge of a hill station. September and October are thus the most trying and un- healthy season of the year ; and in Bengal it is not until the end of October or the early part of November that an appreciable fall of temperature brings relief. Such, according to the author's experience of many years, is the climate of Calcutta, and the description may serve as fairly representative of that of the greater part of Bengal. Farther north indeed, and on the higher ground to the west of the delta, the hot season is somewhat drier and its temperature rather higher ; and in Eastern Bengal the differences are of the opposite character. In Orissa the CLIMATES OF THE PLAINS 155 temperature is higher throughout the year, and in other respects the climate differs in the same manner as in West- ern Bengal. The agriculture of Bengal is determined by the watery character of its surface. Eice, which in the Upper Pro- vinces is of subordinate importance and an article of luxury, in Bengal is the universal crop and the food of the people, and two sowings are reaped during the year, viz. the Aus crop, sown in the early spring and harvested in July and August, and the Amun crop, which is sown in April, transplanted in August, and reaped in the cold season. Oil seeds and dal are also largely grown, and in Northern Bengal tobacco and jute-fibre are important crops, widely cultivated for export. Among the minor but still important articles of produce may be mentioned sugar-cane, betel pepper, and locally, the mulberry for the culture of silkworms, hemp, and cocoa- nuts, which last never grow at any great distance from the sea. The only important forests on the plains of Bengal are those of the Sunderbuns, the marshy islands along the seaward face of the delta, of which the most important tree is the Sundri, producing a tough wood, much used in carriage and cart building. In the Appendix are given the tabular statistics of Burd- wan and Berhampore, representing the western margin of the alluvial tract, the former being the more southerly. These lie in the drier portion of the province ; Calcutta and Jessore, which are more central; and Dacca and Chittagong in the eastern and damper districts. Northern Bengal is not represented by any station, but its climate may be re- garded as intermediate between that of Purneah in Behar and that of Dhubri, near the entrance of Assam. The province of Orissa is illustrated by Cuttack. The mean temperature of Bengal is nearly uniform between 77 and 78 ; that of January 65 or 66, Chittagong being a degree warmer. April and May are about equally hot, viz. 85 or 86 in the western and drier districts, and between 81 and 83 at Dacca and Chittagong. In the former, the mean temperature falls 1 or 2 on the advent of the rainy season ; at the latter it remains nearly uni- form from April to the end of September, the night temperature rising 156 CLIMATES AND WEATHER OF INDIA in the same measure as the day temperature falls, until the daily range is reduced by one-half, more or less ; and not until the middle or latter part of October does a general fall set in with the light northerly winds that succeed the rains. The highest temperature of the year occurs in April, or, more frequently, in May ; and varies between 106 and 111 at Burdwan, 105 and 113 at Berhampore, 97 and 106 at Calcutta, 99 and 108 at Jessore, 94 and 106 at Dacca, and 91 and 99 at Chittagong. In the cool season the lowest tem- perature occurs as a rule in January, sometimes in February, and varies between 44 and 51 at Burdwan and Berhampore, 45 and 52 at Calcutta, Dacca, and Chittagong, and 39 and 48 at Jessore. The diurnal range of temperature in the drier seasons of the year is less in Bengal than in the Upper Provinces, not exceeding 22 in any month at Calcutta, 23 at Chittagong, and 25 or 26 at the other stations, except Berhampore, where, in March, it averages 28. The humidity of the air averages 69 per cent of saturation at the driest and most westerly station, Burdwan ; 74 at Berhampore, 75 at Jessore, 77 at Dacca, 78 at Calcutta, and 80 at Chittagong ; and in the driest month, either February or March, is as high as 55 at Burd- wan, 57 at Berhampore, 62 at Jessore, 65 at Dacca, 69 at Calcutta, and 70 at Chittagong. Calcutta, owing probably to the low level of the site, and the proximity of a large sheet of water, the Salt Lake, which skirts its eastern suburbs, is therefore damper than most other stations in the western and central part of the delta, but the dampest climate is that of Chittagong. At the height of the rains the humidity of the air averages 87 per cent and upwards at all stations except Burdwan ; at Calcutta as much as 89 in August. The rainfall has a similar distribution, being heavier in the eastern districts and the marginal portion of the delta, and heaviest at Chitta- gong. The extremes are Berhampore and Burdwan with 55 and 58 inches respectively, and Chittagong with 106 inches. Calcutta has an annual rainfall of 65 J inches, Jessore of 68 inches, and Dacca of 74 inches. In the wettest year on record (1871) the rainfall of Calcutta amounted to over 93 inches, and in the driest (1837) to only 43j inches. The number of rainy days is less variable than the amount of the fall in different parts of the province. It amounts on an aver- age to 91 at Berhampore, 105 at Burdwan, 110 at Dacca, 118 at Calcutta, 121 at Jessore, and 122 at Chittagong. The mean temperature of Cuttack is 81, 3 or 4 warmer than Bengal ; that of May, here the warmest month, 89 ; and the highest temperature of the year has varied during the last 10 years between 106 and 118. Owing to the lightness of the spring rainfall, and the prevalence of dry westerly winds, the hot season is here, therefore, for the time, as intense as in the Upper Provinces. December is the coolest month, and the lowest winter temperatures have never fallen below 49. The mean humidity of the year is about the same as at Burdwan, but in the driest month does not average below 62, and in CLIMATES OF THE PLAINS 157 the dampest does not exceed 82. It is therefore more uniform at different seasons than in the province of Bengal. This peculiarity is probably local, owing to the position of the station, at a low level, between two rivers. The mean annual rainfall is 57J inches, dis- tributed over 106 rainy days. The rains are rather more prolonged in Orissa than in most parts of Bengal, as this province receives more of the October rainfall, when the winds become easterly in the north of the bay and the season of heavy rain is setting in on the Madras coast. On the other hand, the rainfall of the spring season is lighter than in Bengal. Assam and Cachar. This province, which, including Sylhet, lies to the east of Bengal, and to the north of Bur- mah, Manipur, and Tipperah, consists of two alluvial plains, that of the Brahmaputra and its affluents, and that of the Barak ; together with the intervening hills, different sections of which, named from the several indigenous tribes that occupy them, are distinguished on our maps as the Garo, Khasi, and Naga hills. The valley of the Brahmaputra, which constituted the former kingdom of Assam, is long and narrow, and is nearly divided into two parts, Upper and Lower Assam, by the Mekhir hills, an isolated and in- dependent group which stands out midway in the valley between the Naga hills and the Himalaya. The valley of the Barak, forming the province of Cachar and the plain of Sylhet, which opens out to the west of the former and is traversed by its two branches, the Soorma and Kus- hiyari, is much shorter, and occupies the interval between the Khasi hills on the north and the low hills of Tipperah and the Looshaie country on the south. The plain of Sylhet has long been cleared and cultivated. Many parts of it are low, and in the rainy season are flooded to a considerable depth by the torrential drainage from the hills around. Nevertheless large tracts are cultivated with rice, and the long low hill spurs that run out from the southern hills, the low hillocks that dot the plain along the bases of these and the northern hills, and the higher parts of the plain itself, together with similar tracts in Cachar, are occupied by tea gardens, established for the most part on clearings made in the forest during the last 40 years. 158 CLIMATES AND WEATHER OF INDIA In the Assam valley tea cultivation is still more extensive, but the cleared area is far exceeded by that of the virgin forest. Writing in 1872, Mr. Peal describes the aspect of the valley, as seen from the slopes of the Naga hills, as " literally a sea of jungle forest, an enormous dead level. The smallness of the area under cultivation surprised us more than anything; it did not look 1 per cent. The Potars I could easily recognise ; yet they were but little green streaks, hardly noticed in the general view. The amount of waste land is enormous." l And Col. Godwin Austen, describing the valley of the Dunsiri in the following year, says, " At 5 miles from Golaghat the forest is entered, and this is continuous to the foot of the hills for a distance of 44 miles." 2 Around the rivers which intersect the plain, tracts that are too swampy for forest are covered with tall grass. In these, rice cultivation is gradually extending, but the native population is at present too sparse to deal with more than some small portions of the available area, and thus the greater part of Assam is still in the condition of forest and swamp. As the principal seat of tea cultivation in India, and the country in which the tea plant is indigenous, still growing wild on the hills around, the climate of Assam is one of peculiar interest. Its most characteristic feature is its dampness at all seasons, in conjunction with the moderately high and comparatively equable temperature, due to the sub-tropical position of the province and to its being pro- tected from all desiccating winds by the hills which enclose and seclude it. In the cool season thick fogs cover the low grounds, often remaining undissipated till the sun has attained its noonday power. In the spring, with the in- creasing warmth, thick clouds gather daily over the valleys, and frequent showers and thunder-storms drench the dense foliage of the evergreen forest while moderating the heat ; so that, instead of a hot season preceding the rains, such as we have seen to be an universal characteristic of India 1 Journ. As. Soc. Bengal, Vol. xli. Part i. p. 10. 2 Report on Survey Operations in the Naga hills and Manipur, 1872-73. CLIMATES OF THE PLAINS 159 proper, the temperature rises gradually and uniformly as the sun becomes higher and the days longer, and reaches its highest point in July, in the middle of the rainy season. At this season, the rainfall of the Assam valley is not more copious than in Bengal, though torrential on the southern face of the Garo and Khasi hills. But while it begins much earlier it lasts till October ; and up to the end of that month, and even in November an occasional day's rain continues to stimulate vegetative growth. December is the only month in which the average rainfall is below 1 inch, and it is not till the beginning of December that the plucking of the tea bushes comes to an end for the season, and that the plant after pruning is allowed to rest till the showers of February and March bring out the first flushes, and afford materials for the manufacture of the following season. A climate so uniformly moist in a country of such a character can hardly be described as eminently favourable to health and comfort. But it has been proved by half a century of experience, that when the jungle has been cleared and due regard is had to efficient drainage and the selection of the dwelling site, Europeans of good constitution who observe the precautions indispensable in all tropical climates and lead an active life, may preserve their health in Assam as well as in other parts of India. The cool season, if less bracing than in the Punjab and the drier provinces of Northern India, is still such as an Englishman may enjoy, and if from May to October he has to endure the oppressive heat of a vapour -laden atmosphere, it is questionable whether even this is more trying than the torrid fervour of an Indian hot season with some months of damp heat to follow. And to those whose natural taste and training enable them to appreciate the beauties of nature and to feel an interest' in her works, Assam presents many attractions. No province in India affords a more exquisite presentment of nature in her native glory of hill, river, and forest. To the sportsman, indeed, the very impenetrability of the forests is an obstacle which may deter any but the most persever- ing from penetrating to the yet unvisited haunts of the 160 CLIMATES AND WEATHER OF INDIA rhinoceros and the wild buffalo. But to the naturalist the field is one of extraordinary richness. The province stands on the borderland of the Indian, Malay, and Indo-Chinese regions, and is tenanted by numerous forms of animal and vegetable life unknown in India. In describing the hills that border the north-east frontier, Col. Godwin Austen writes: 1 " We have a mingling in this direction of the Indian with the Indo-Chinese forms. Many birds, extremely rare in collections, and only represented by a few solitary examples, described by Hodgson and Blyth from Nepal, etc., were obtained here again, and a large number have yet to be identified or compared with specimens from other dis- tant quarters. Some 280 species were collected. Other families of natural history are equally rich, none more so than the insects, and interesting forms of land mollusca are numerous. The great forest of the Dunsiri swarms with insect life, the Lepidoptera conspicuous by their numbers and coloration. Leaving this and ascending the spurs of the Burrail, with the changing flora new forms are constantly appearing, and I could point to few areas where a naturalist can find more rare objects for his search than on the slopes of this range." The botany of the province is equally rich and varied. " The mountains display a rich vegetation of the most tropical forms which India produces. Anonacea are numerous, several species of MyristicecB occur, and the India-rubber fig forms large forests in some places. Calami and Plectocomia abound in the dense jungles, as well as other rare and interesting palms belonging to the genera Livistonia, Licuala, Arenga, Areca, Wallichia, etc. Oaks and chestnuts are also characteristic types, as are Guttiferce, Ternstrcemiacece, Magnoliacece, Sauraujcij and tree ferns." 2 Besides tea, which is the chief agricultural product of Assam, rice is largely grown, but, as regards the Assam valley, insufficient even for local consumption. From Sylhet, however, there is a considerable export, and among minor articles of produce the supari or betel nut, oranges, and India-rubber, the latter collected in the forests by the hill tribes around. Of the observatories now or formerly existing in the province, the climatological data afforded by those of four stations are given in the Appendix. One of these, Shillong, 1 Report on Survey Operations in the Naga hills and Manipur, 1872-73. 2 Hooker and Thomson, Introductory Essay to the Flora of India. CLIMATES OF THE PLAINS 161 on the Khasi hills, has already been noticed among the hill stations of India. The others are, Sibsagar in Upper Assam, Dhubri on the confines of Bengal opposite the en- trance of the Assam valley, and Silchar, the chief town of Cachar. The mean temperature of Sibsagar is 73, the same as that of Sialkot and Ludhiana in the Punjab : but while that of the coolest month, January, is 59, or 7 higher, that of the warmest month, here July, is 84, or 7 lower than the June temperature of the latter stations. A still greater difference is shown by the extreme tempera- tures of the year. At Sibsagar the thermometer very rarely reaches 100 ; this has occurred in only 2 years out of 10 ; and the highest reading yet recorded is 1 02, while the lowest vary between 40 and 46. The highest averages 17 or 18 lower, and the lowest 7 or 8 higher than the corresponding extremes at the Punjab stations. In like man- ner the daily range of temperature is much less, in April little more than half that of Ludhiana. All these differences are due to the dampness of the climate. This is very high. It averages 83 per cent of saturation,, and only in one month of the year is it below 80. In the cold season the air is relatively damper than in most months of the rains, and hence the frequency of fog. Cloud is common at all seasons. The rainfall of Sibsagar averages 93 inches. It is least in December, but even then, on an average, it rains one day in ten ; and in all other months, except November, more frequently. Even in March there are 14 rainy days, and on the average of the whole year 164. In 1874, a very wet year, it rained on 195 days. The rainfall of the valley above Sibsagar is still higher. Dhubri, which may be taken as in some measure representative of the climate of Lower Assam, and also of that of North-eastern Bengal, has a higher mean temperature than Sibsagar, viz. 75, and even a smaller annual and diurnal range of variation, except from February to April. January is 4 warmer, while July, which here as in Upper Assam is the warmest month, is 4 cooler. There are as yet only 4 years' registers for this station, and the highest and lowest temperatures yet recorded, both in the same year, are 101 and 45 respectively. In the driest months there is a mean difference of 22 between the morning minimum and the afternoon maximum in the shade, but in the 4 months of the rains only 8 or 9. The mean humidity is not so high as in Upper Assam, being only 78 ; but while it is much lower in the cold season, it is rather higher in the rains. The rainfall is about the same, viz. 94 inches, but there is very little from the end of October to the end of February, less in March and April than at Sib- sagar, and much more in May and June. With an equal total annual fall, the rainy days are only two-thirds as numerous as at Sibsagar. The greater part of Lower Assam being under the lee of the Garo M 1 62 CLIMATES AND WEATHER OF INDIA and Khasi hills, has a lower rainfall than either Dhubri or Sibsagar. Thus Gauhati has but 69 inches, Nowgong 79j, and Tezpur 76 inches ; but it is no doubt higher along the northern margin of the valley, at the foot of the Himalaya. There are numerous tea gardens in Lower as well as in Upper Assam, even where the rainfall is least, but they are less productive. Silchar is in the narrower part of the Barak valley, on an alluvial plain, but little above the flood level of the river. Ridges of low forest-clad hills run southwards from the river bank at intervals of 6 or 8 miles, and these intervals are occupied by plains, a great part of which are under water in the rains, and form swamps in the dry season. 20 or 30 miles to the south, these ridges run up into the hill country of South Cachar and the Looshaie hills ; and 10 miles to the north of the river are the Naga hills. About 20 miles farther east, the valley is nearly closed by the Burrail range, which rises to 5000 feet, and other higher ranges intervene between it and Manipur. Thus situated, Silchar has a climate somewhat warmer than that of Upper Assam, in virtue of its lower latitude, but quite as rainy. Indeed the rainfall is heavier than that of Sibsagar, viz. 120 inches, and it rains nearly as often. But in the cool months fogs are less frequent. The mean temperature is 76 ; that of January 64, and that of the three months, June to August, 82. A temperature ex- ceeding 100 has been registered but once in the last 11 years ; but the average maximum of the year, which is remarkably uniform, is 99, and the minimum 45. In point of humidity the atmosphere rather resembles that of Dhubri than that of Upper Assam ; but Silchar is much damper from February to April. Central Provinces (Nagpur) and Berar. These pro- vinces lie entirely south of the Tropic of Cancer, and occupy the northern central portion of the Indian peninsula. To the south of the Sone and Nerbudda valleys a tract of hill country, in places 100 miles in breadth, known in ancient annals as the Vindhya, and on modern maps bearing the name of the Satpura range, stretches across the peninsula from the high plateau of Amarkantak to the fortress of Asirgarh, beyond which, after a short interruption, it is continued in the Eajpipla hills to the Gulf of Cambay. It separates the drainage of the Ganges and Nerbudda from that of the Tapti, the Godavery and Mahanadi, and extra- tropical from inter-tropical India. From a geological point of view, it is not a definite mountain range, presenting no uniformity of structure or rock character, but is rather a CLIMATES OF THE PLAINS 163 belt of broken highland, a series of tablelands varying from 2000 to 4000 feet in elevation, which have remained while the country to the north and south has been worn down to lower levels by long ages of denudation. To the south of this hill-belt, the former kingdom of Nagpur, now included in the Central Provinces, and the province of Berar, for- merly a portion of Hyderabad, consist of three great stretches of plain, in great part cultivated and, at the foot of the Satpuras, elevated more than 1000 feet above the sea, to- gether with several ranges of hills which divide and sur- round them ; some bare and rocky, but more generally covered with forest, or with the scrub jungle that remains from the destruction of the former forest. The three plains are that of Berar, drained by the Poorna, a tributary of the Tapti ; that of ISTagpur, traversed by the Waingunga, a tributary of the Pranhita and Godavery ; and that of Eaipur or Chhatisgarh, on the upper course of the Mahanadi. The black soil which covers these plains is very fertile, and yields abundant crops of cotton and wheat, which grain is here at its southernmost limit as a staple crop. Other produce is of the same kinds as are raised generally in Northern India. But there remain large stretches of culturable land, now or formerly covered with forest, which, in due course of time, with the increase of population, will doubtless be brought under the plough. Besides these, however, the provinces include hilly tracts of enormous extent, formerly bearing forests of valuable timber, which have been terribly devastated and, for the time being, rendered of little value, by the wasteful temporary cultiva- tion of the nomad hill tribes, and by the destructive fires which the villagers annually kindle to burn off the coarse grass, and provide a young growth for pasturage. Some three-fourths of the Central Provinces are thus nominally forest; and of this an area of 20,000 square miles is the property of Government. At present not more than 3500 miles are rigorously preserved, but this area may and doubtless will be largely extended ; and it may be expected that, in the course of some years, the forests of the Central 164 CLIMATES AND WEATHER OF INDIA Provinces will become one of the most valuable and pro- ductive sources of teak and sal timber in the peninsula. Other valuable timber trees are the Saj (Terminalia tomen- tosa), Bijesal (Pterocarpus marsupium), Shishum or Sissoo (Dalbergia latifolia), Kawa (Pentaptera arjuna), and Anjun (Hardwickia binata). In describing the climate of Nagpur, Mr. (now Sir Charles) Grant says : " The mean temperature is higher than in many parts of India, at the same height above the sea-level. But the absence of the really bracing air in the cold season of Upper India is in some measure com- pensated for by the fresh cool weather during the greater part of the monsoon ; and by tolerably cool nights in the summer months. " As in other parts of India, there are three seasons, the hot, the cold, and the rainy. The positively hot weather ordinarily commences about the first of April and lasts till the first week in June. The monsoon lasts throughout June, July, and August. At this season the climate, though full of moisture, is fresh and pleasant to the feelings. In September there are long breaks between each fall of rain, when the weather is often close and sultry, though never so much so as in the plains of the north of India at this time. October is generally sultry and unpleasant, but diversified occasionally by refreshing showers. The cold weather does not fairly set in till the middle of November. From the 15th of November to the end of February the air is generally cool and pleasant. Often, however, with the appearance of clouds, the thermometer rises as much as 7 or 8, and the climate becomes disagreeable and close. From the 15th of February the weather gets warmer, and the hot winds blow from the beginning of April till the monsoon. Rain falls during every month in the year, usually during the hot and cold season only in showers, but sometimes accompanied with violent storms. Hail falls occasionally in January, February, and the early part of March, some- times in very large stones, doing much damage to the spring crops. " The climate during the rains is considered by the poorer inhabit- ants, who are exposed to it, as more trying than the cold of the real cold weather. In July and August it is not unusual to see people sitting round a fire in the very early morning, before going out for their day's labour. The climate is certainly not unhealthy. . . . Fever is the most frequent among the epidemic diseases. The most un- healthy season is from the second week in September to the second week in December. The jungle tracts are certainly not free from malaria until the cold weather has well set in." This description applies to the plain. On the Satpuras it is much cooler. Thus, of Betul, Mr. Grant writes : CLIMATES OF THE PLAINS 165 "It is almost out of the reach of hot winds, and would no doubt be an agreeable residence in the hot season. The climate' of Betul generally, at least to Europeans, is fairly salubrious ; its height above the plains (2000 feet) and the neighbourhood of extensive forests moderate the heat of the sun and render the temperature pleasant throughout the greater part of the year. During the cold season the thermometer at night continually falls to several degrees below the freezing-point. 1 Little or no hot wind is felt before the end of April, and even then it ceases after sunset. The nights in the hot season are invariably cool and pleasant. During the monsoon the climate is very damp, and at times even cold and raw, thick clouds and mists enveloping the sky for many days together." A similar account is given of the climate of Chhindwara; that of Pachmarhi, at 3500 feet, has already been noticed among the hill stations ; and that of Seoni may be gathered from the statistics given in the Appendix. In the Appendix are given the tabulated data of four stations ; three of which, Akola, Nagpur, and Eaipur, re- present the three southern plains ; and the fourth, Seoni, the Satpura tableland. Beginning with the most easterly, Raipur, which is situated on the undulating plain of Chhatisgarh, at an elevation of 960 feet above the sea. The mean temperature is 78, the same as that of Calcutta and Burdwan in Bengal, fully 2 farther north, and 3 cooler than Cuttack in nearly the same latitude. The earlier months of the year have a higher temperature than at some of the Bengal stations, but this is compensated by the greater coolness of the raina and later months. December is the coolest month, with a mean tem- perature of 66 and a mean minimum night temperature of 54. The lowest reading of the year has varied between 41 and 51 in the last 10 years. May is the hottest month, with a mean temperature of 92, and of 105 in the afternoon, the absolute highest in the year being between 108 and 116. In the rains the mean temperature falls below 80, and that of the afternoon to 87, and the temporary rise on the cessation of the rains is only 1. The diurnal range amounts to 25 or upwards from January to May, decreases to 12 in the rains, and then increases again gradually till the end of the year. The atmo- sphere is moderately dry, the mean humidity being 59 per cent. April and May are the driest months, when the hot winds blow and the mean humidity sinks to 36 per cent, and much lower in the day- time. The rainfall is nearly 52 inches, but of this but little falls from the end of October to the end of May, the cool months, Decein- 1 This must refer to an exposed, hardly to a well-shaded, thermometer. 166 CLIMATES AND WEATHER OF INDIA her to February, being especially rainless. The average number of rainy days in the year is 76, of which 61 occur in the four months, June to September. Nagpur, although but little farther south and at a somewhat higher elevation (1025 feet), is rather warmer, the mean temperature of the year being 79. From November to May it is 1 or 2 warmer, a difference due solely to the higher temperature of the daytime, since the morning minima are as low or lower than those of Kaipur. The extreme readings of the year have varied between 112 and 118 for the maximum, and between 43 and 51 for the mini- mum. The climate is drier than that of Raipur, the mean humidity being 53 per cent of saturation, and that of April and May but 28 and 30 per cent. The rainfall also is lower, viz. 45 inches, and its distribution in the different seasons is similar. But it would seem to rain rather oftener, as the registers show an average of 84 rainy days in the year. Akola, the westernmost station, situated on the plain of Berar, is again drier. Its mean temperature is 78; that of January 68, and that of May 93. In all months the mean temperature differs but little from that of Nagpur, but the winter minima are 2 or 3 lower and the annual minimum some 5 lower. The mean humidity is but 50 per cent, and that of April, the driest month, 22 per cent ; while that of the rainy season does not average more than 74 or 75 per cent. The rainfall is much less than at the more easterly stations, not ex- ceeding 30 inches, and the number of rainy days is 66, of which 9 only occur in the seven months, November to May. Seoni, on the tableland, almost due north of Nagpur, and 2030 feet above the sea, has a mean temperature 5 lower, viz. 74; a differ- ence which holds good all through the dry season pretty equably, but is reduced to 3 in the rains. The afternoon temperature of May averages 103, the morning minimum in December 50; and while the highest readings of the year have varied between 105 and 111, the lowest have ranged between 36 and 47. The average humidity is rather higher than that of Nagpur, except in the last three months of the year ; and the rainfall is rather greater, viz. 5 1 inches, distributed over 98 rainy days, of which, *on an average, 12 occur iri the six months, November to April, and 6 in May. The West Coast of India. The Konkan and Malabar. The west coast of the peninsula and the strip of hilly and undulating country that extends below the Ghats from the Gulf of Cambay to Cape Comorin has the dampest and most uniform climate of any part of the peninsula. While open to the westerly winds from the ocean, which mitigate the more intense effect of the tropical sun and maintain the CLIMATES OF THE PLAINS 167 verdure of the land surface and of the evergreen forests that clothe the slopes of the Ghats, the country is shielded by this range from the desiccating winds of the Deccan table- land, and its surface is watered by the numerous small streams that bring down its drainage and open out into the tidal estuaries and backwaters that intersect the seaward margin of the tract. Although extending through 13 of latitude (from 8 to 21 N. lat.), the annual mean temperature of the western coast is nearly the same throughout, viz. 79 or 80. But in some other respects there are considerable differences between its northern and southern extremities. To the north of Bombay the climate of January and February, if not comparable with the cold season of Northern India or even of Bengal, is still such as to afford cool nights, while the ordinary damp heat of the daytime is tempered by northerly winds ; and as far down as the southern limits of the Bom- bay Presidency (in N. lat. 14) the season from the latter part of October to the end of May is almost rainless. But in South Canara, Malabar, and Travancore, the day tempera- tures of January and February are little if at all lower than those of the subsequent months, and in the early morning the thermometer does not fall much below 70 at places near the sea ; while spring showers precede the rains, and the autumnal rainfall lasts to the end of October. At Cochin, January and February are the only months in which the average rainfall does not exceed 1 inch, and July is cooler than January. The rainfall of the summer monsoon is heavy all along the west coast, and still heavier on the Ghats, which force the saturated current to ascend to a height of from 2000 to 7000 feet before pursuing its course across the Deccan tableland and the loftier hill groups of the Nilgiris, the Anamalais, and the Pulnis. The cooling which it under- goes in this ascent, amounting to about 1 in each 400 feet, causes the enormous precipitation on the face of the Ghats, of which Mahableshwar with 255 inches, Baura Fort with 250 inches, and Matheran with 244 inches, are examples. 1 68 CLIMATES AND WEATHER OF INDIA On the coast the fall is less. From June to October it amounts to 119 inches at Mangalore and 132 inches at Honawar, where it is heaviest ; decreasing thence north- wards to 100 inches at Eutnagiri, 73 inches at Bombay, and to only 42 inches at Surat. In the south, on the coast of Travancore, the rains set in somewhat earlier than at Bombay, generally in the latter part of May, and are heaviest in June. In Bombay they may be expected in the first or second week of June, are heaviest in July, and come to an end in October ; but in the south they last quite to the end of October, and the rainfall of this month is rather heavier than that of September. It is almost needless to remark that in such a climate the vegetation has all the luxuriance that is commonly asso- ciated with our ideas of the Tropics. The strip of low plain that borders the greater part of the coast is covered with cocoa-nuts and rice fields, and the villages of Malabar and Travancore are embowered in groves of betel nut palms and Talipots. " Cassia, pepper, and cardamoms flourish wild in the jungles, and form staple products for export. The fact that the pepper is cultivated without the screens used in other parts of India to preserve the humid atmosphere about it, is the best proof of the dampness and equability of the climate. The low valleys are richly clothed with rice fields, and the hillsides with millets and other dry crops, whilst the gorges and slopes of the loftier mountains are covered with dense and luxuriant forest." l Of the forests Sir D. Brandis writes : " The richness of the vegetation . . . may be seen in the forest tracts at the foot of the Ghats, north-east of Bombay, stretching to the neighbourhood of the Tapti river. These forests, in which the Teak is the most important tree, consist chiefly of species that shed their leaves in the dry season. The dense evergreen forests, characteristic of the wet zone, begin to the south of the Rutnagiri district. In Canara, Malabar, and Travancore they occupy large tracts at the foot of the Ghats and also on their summits and the higher hill groups, which, like the Nilgiris and Anamalais, branch off to the east of the chain. Manableshwar and the surrounding slopes are clothed with 1 Hooker and Thomson, Introductory Essay to the Flora Indica. CLIMATES OF THE PLAINS 169 evergreen forest which, some miles farther east, where the climate becomes drier, yield place to a mixed forest of deciduous trees. On the windy plateau of Mahableshwar, as well as elsewhere on the crest of the Ghats, the trees are indeed small and stunted ; but in protected depressions and valleys the thickest evergreen forest of the wet zone is luxuriantly developed, and besides Laurels, Eugenia and the wild Mango, consists of Anonacece and other characteristic tropical families. . . . The variety and richness of this evergreen forest increases to the south, and reaches its highest development on the western slopes of Coorg, the Wynaad, and the Nilgiris, opposite that part of the coast where the rainfall is highest. Extensive forests with a mean height of 200 feet are here not rare. Between the tree-boles the ground is covered not only with dense bush and herbage, in which species of Strobilanthes, an Acanthaceous genus with large beautiful flowers, plays a conspicuous part, but also with young trees whose parents protect them with their dense canopy. The trees of these forests as well as the undergrowth are all shade-loving plants, which can thrive in the deep gloom of the forest. When, on the fall of some forest giant, a gap is made in the leafy canopy, the young trees shoot up, mastering the underwood, and, striving towards the light, contend with each other for pre-eminence of growth, till the weaker yield place to the stronger. Besides the families already mentioned, the southern tract especially abounds in Guttiferce, Dipterocarpce, Meliacece, Leguminosce, Rubiacece, Euphorbiacece, and Urticacece. In many places tree-ferns, palms, and bamboos give a characteristic aspect to the forest." 1 The climate of the west coast is illustrated in the Appendix by four stations, viz. Surat in the extreme north, Bombay, Mangalore and Cochin ; and that of the summit of the Ghats by Mercara in Coorg. The gradations of tem- perature, humidity, and rainfall, experienced in passing from north to south, will be best shown by the following com- parison of the four coast stations : The mean annual temperatures of all four stations are within a degree of each other, viz. 79 or 80. But whereas that of Surat in January is 70 and in May 86, the corresponding temperatures of Bombay are 74 and 85, of Mangalore 76 and 83, and of Cochin 79 and 82 (that of April being 84). The extreme temperatures of the year show a much greater variation. At Surat they are respect- ively 109 and 48, at Bombay 95 and 61, at Mangalore 94 and 63, and at Cochin 95 and 67. The Bombay observatory being situated on a narrow point of land, surrounded on three sides by many miles of sea, has doubtless more equable temperatures than places on the coast of the mainland in the same latitude. 1 Die Beziehungen zwischen Regenfall und Wald in Indien (translation). 1 70 CLIMATES AND WEATHER OF INDIA The mean humidity of the atmosphere is comparatively high, and equable at all the stations except Surat. At this place it averages but 62 per cent of saturation, and remains between 50 and 60 from November to May. At the remaining stations it averages between 77 and 80, and never falls to 60 in any single month. At Cochin the lowest mean humidity of any month is 70 per cent. The humidity of the air, therefore, like its temperature, becomes more constant in passing from north to south, and probably this circumstance, quite as much as the rainfall and the rise of the winter temperature, deter- mines the change in the character of the vegetation noticed in the foregoing description. The average cloudiness of the sky is about the same at Bombay, Mangalore, and Cochin, viz. between four and five- tenths of the sky expanse, but only three-tenths at Surat. The distribution of the rainfall has already been noticed. The average number of rainy days in the year is 66 at Surat, 108 at Bombay, 142 at Mangalore, and 164 at Cochin. Mercara, which is situated in Coorg, only a few miles from the crest of the Ghats at an elevation 3700 feet above the sea, has a cool, equable, and pleasant climate at all times of the year. The mean temperature is only 67, and that of the warmest month, April, 72; the average afternoon maximum of this month being 85, and the highest reading recorded in each of the last 5 years within a degree or so of 90. The mean of December and January is 64, that of the early morning 55 or 57, and the mean minimum of the year 49. In the rainy season the air is almost saturated with vapour, and the station enveloped in cloud, and during the four months, June to September, it rains almost daily, and often day and night long without intermis- sion. From December to the end of March, however, there is but little rain, and in the first three months of the year the humidity is between 60 and 70 per cent of saturation. The annual rainfall averages 127 inches, of which about 106 inches falls from June to September. The Wynaad, which is the chief coffee-growing district of India, lies on the Ghat range, and at its foot, immediately to the south of Mercara, and its climate is similar but some- what warmer, owing to its lower elevation. A high rainfall, a moderately high and equable temperature, and a generally damp atmosphere, are the conditions most suitable to the plant, as will be further illustrated by the climatic features of Kandy, in Ceylon. Cinchona is also largely cultivated on the slopes and on the plateau of the Nilgiris. Indeed, the plantations are, altogether, much more extensive than those of Darjiling, and certain varieties, especially the crown- CLIMATES OF THE PLAINS 171 barks (C. qfficinalis), also C. Ledgeriana and 0. pulescens, are grown with greater success than in the Sikkim Himalaya. Khandesh, the Deccan, and Mysore. In India without the Tropic, the most arid province, Sind, is separated from the dampest, Assam, by the whole breadth of Northern India, a distance of 22 of longitude; and while the former lies in the extreme west the latter occupies the ex- treme east of the region. In the peninsula these relations are in a measure reversed, and whereas the seat of the heaviest rainfall and the dampest atmosphere is the west coast of the peninsula, the traveller has but to ascend the Ghats and strike eastward across the tableland beyond their crest and, in the course of only 30 or 40 miles, he will have passed from the evergreen forest and torrential rainfall of the Sahyadri range to the rolling plains of black soil and the flat- topped terraced hills of the Deccan, where a shady tree is among the rarest of nature's gifts, and where, not- withstanding the natural fertility of the soil, all agriculture is more or less precarious owing to the uncertainty of the scanty rainfall. The interior of the peninsula, above the Western Ghats, is a plateau, the greater part of which is elevated between one and two thousand feet above the sea. Its general slope is eastward, and the great rivers that carry off its drainage, the Godavery, the Bhirna, the Kistna, the Tungabhadra, and the Cauvery, and many of their tributaries, take their rise on the very crest of the Sahyadri range and flow eastwards to the Bay of Bengal. The northern portion, comprising nearly all that lies within the Bombay Presidency and the greater part of Hyderabad, consists of flat sheets of volcanic rock, and is thus described by the authors of the G-eological Manual of India : " The volcanic region of Central and Western India is distinguished by marked peculiarities of scenery, and the characters of the surface are widely different from those found in other parts of the Indian peninsula. Great undulating plains, divided from each other by flat- topped ranges of hills, occupy the greater part of the country, and the hillsides are marked by conspicuous terraces which may often be traced to great distances, and are due to the outcrop of the harder 172 CLIMATES AND WEATHER OF INDIA basaltic strata or of those beds that resist best the disintegrating in- fluences of exposure. . . . The vegetation of the trap area differs no less conspicuously from that which is found on other formations ; the distinction in the dry season being so marked that, especially when taken in connection with the form of the surface, it enables hills and ranges of trap to be distinguished at a distance from those composed of other rocks. The peculiarity consists in the prevalence of long grass and the paucity of large trees, and in the circumstance that almost all bushes and trees, except in the damp districts near the sea, are deciduous. The result is that the whole country, except where it is cultivated, presents during the cold season, from November to March, a uniform straw-coloured surface with but few spots of green to break the monotony ; whilst from March, when the grass is burnt, until the commencement of the rains in June, the black soil, black rocks, and blackened tree stems present a most remarkable aspect of desolation. During the rainy season, however, the country is covered with verdure, and in many parts it is very beautiful, the contrast afforded by the black rocks only serving to bring into relief the bright green tints of the foliage." The southern half of the plateau, in the Madras districts of Bellary and Anuntapur and the kingdom of Mysore, has a different aspect. This, too, consists of rolling plains, but the formation being that of the older crystalline rocks, the hills are of rounded whale-backed forms, and frequently the eminences are crowned with great tors of highly picturesque and bizarre shapes or rounded hummocks of bare rock, the result of ages of atmospheric action on rocks of varied in- ternal structure. As a general rule, the climate of the Deccan and Mysore, beyond 30 or 40 miles from the crest of the Ghats, is very dry. The driest portion of the Deccan is a strip running north and south, parallel with the Ghats, and from 50 to 80 miles to the east of them, from the foot of the Satpura range in the Tapti valley to almost the foot of the Nilgiris. As far south as the latitude of Poona, the zone of country with a rainfall below 30 inches averages not more than 100 miles in width ; but to the south of this it extends right across the plateau to the Eastern Ghats, and even beyond to the sea- coast. In the centre of this part of the plateau, to the south and south-east of Bellary, is an area of 6000 or 7000 square miles, within which the annual rainfall is below 20 CLIMATES OF THE PLAINS 173 inches ; a tract which has repeatedly suffered from famine in its severest form. In this dry tract the waste lands are covered with coarse grass or a thin bushy scrub, but in the hilly country of Sandur to the west, where the rainfall is higher, there is a flourishing forest. In the forests of Mysore, especially those of the Eastern Ghats around Kollegal, the most valuable tree is the Sandal wood (Santalum album), which is restricted to the Deccan and Mysore plateau, and to the south of latitude 16 or 17. The Teak is limited to the zone fringing the Ghats, where the rainfall exceeds 30 inches ; and it occurs also in the Sandur hills, and in the hilly country to the east of the dry zone which extends far into Hyderabad, and where the rainfall is also in excess of 30 inches. The red sanders (Pterocarpus santalinus), which yields a valuable dye-wood, is restricted to the eastern margin of the plateau, chiefly in the districts of Cuddapah and North Arcot. In Khandesh and the greater part of the Deccan dis- tricts of Bombay, north of the Kistna river, the winter and spring months are almost rainless. Land winds from the east in the earlier, and from north-west in the later months, prevail almost up to the setting in of the summer monsoon. The winter rainfall of Northern India does not as a rule penetrate so far south, and the thunder-storms of the spring months, fed by vapour from the Bay of Bengal, rarely occur in the northern part of the peninsula much farther west than the meridian of Nagpur. But farther south in Dhar- war, in Hyderabad, and also in Mysore and the Madras districts of Bellary, Anuntapur, and Cuddapah, showers occur occasionally in March and April, and more frequently in May. During the monsoon the plateau is swept by a strong steady west wind, which only occasionally slackens and gives a light rainfall ; but the rains last longer than in most parts of Northern India, and the October rainfall generally amounts to between 3 and 4 inches in the Deccan, and in Bellary and Mysore is quite as copious as that of September. The rainy season, if much less rainy than in Northern India, is cloudy, cool, and pleasant. i?4 CLIMATES AND WEATHER OF INDIA Poona is the chosen seat of the Bombay Government all through the rains ; and the great military stations Secun- derabad and Bangalore enjoy at this season a climate which is by no means oppressive, either on account of heat or excessive moisture. In the Appendix, the stations that represent this tract are Malegaon in Khandesh, 1430 feet above the sea ; Poona, at 1850 feet, distant only 30 miles from the summit of the Ghats, and on the border of the dry tract ; Sholapur, near its eastern limit ; Belgaum in Dharwar, opposite a depres- sion in the Ghats that admits the monsoon current less obstructed than in most parts of the range ; Secunderabad, Bellary, and Bangalore ; the last on a very high part of the plateau, nearly 3000 feet above the sea. The mean annual temperature of Malegaon is 76 ; that of Decem- ber, the coolest month, 66, and that of May, the warmest, 88 ; the extreme temperatures of the year being between 36 and 43 as a minimum, and between 107 and 111 as a maximum. In the first three months of the year, when the atmosphere is very dry, the diurnal range of temperature averages not less than 34 or 35, and even in the rains is as much as 14. The mean humidity is only 51 per cent of saturation, and in April, the driest month, 28 per cent. The aver- age rainfall is about 25 inches, and of this, less than 2j inches falls in the seven months from November to May. The average number of rainy days is 66, of which but 7 occur in these months. Poona, which is rather more than 2 farther south, and 400 feet higher, has a mean temperature of 78 ; in December and January of 72, and in April of 86. Since 1878, when for the first time the instruments were properly exposed, the lowest readings in the year have varied between 40 and 50, and the highest between 101 and 110. The former average 4 higher than at Malegaon, the latter 3 lower. In the rainy season, from July to September, the mean temperature is 75, and the afternoon maximum 81 in July and' 83 in the two following months. In the daytime, therefore, Poona is as cool in the rainy season as in December ; but the night tempera- tures are but little below 70 in the former, while they average but 54 in the latter. In the spring months the climate is as dry as that of Malegaon, and the mean humidity of the year is but little greater, viz. 52 per cent. The average rainfall is 28 inches, of which but 2 inches fall from November to April. Belgaum is again 2^ farther south, and 700 feet higher than Poona Its mean annual temperature is 74 ; the December tempera- ture 71, and that of April 81. During three months of the rains it CLIMATES OF THE PLAINS 175 averages only 70 or 71. The winter temperature, therefore, differs but little from that of Poona, but that of the spring and rainy season is from 3 to 5 lower. The climate is much damper and the rainfall heavier (49 inches). Showers fall in April and May on 6 or 7 days in each month, and this rainfall greatly mitigates the temperature. Moreover, all through the monsoon, a very strong west wind blows constantly through the depression in the Ghat range, opposite Belgaum ; the mean rate of movement, as shown by the anemometer, being between 20 and 30 miles an hour, night and day, in June, July, and August. These months and September are very rainy, the average number of rainy days being from 21 to 28 in each month, and even in October it rains nearly every second day. The climate is there- fore much damper than in most other parts of the Deccan, especially those to the north and east. Sholapur is on the eastern margin of the dry tract, and 1600 feet above the sea. Its mean annual temperature is 79, or 5 higher than that of Belgaum ; but the winter temperature is about the same as that of Poona and Belgaum, viz. 70 in December and 72 in January. In the hottest month, May, it is 89, and in the rainy season from 77 to 79. The lowest winter readings have ranged between 42 and 49 in the last 10 years, and the highest summer readings between 108 and 112. The atmosphere is very dry, the mean humidity of April being only 26 per cent of saturation ; and even in the rains it aver- ages only between 60 and 70 per cent. The westerly monsoon, though less strong than at Belgaum, blows very steadily, with an average movement of 10 or 12 miles an hour, and brings compara- tively little rain. The mean monthly fall of each of the three months, June to August, is between 4 and 6 inches only, but that of September is higher, when, the westerly current having slackened, easterly winds begin to set in from the Bay of Bengal ; and in October the rainfall amounts to 3^ inches. The total fall of the year is barely 30 inches, and the number of rainy days 83. / Secunderabad, which lies 160 miles farther east, at about 1800 feet above the sea, has a very similar climate, but slightly cooler. The mean temperature of the year is 78 ; that of December and Janu- ary 69 and 70 respectively, and that of May 89. The extreme temperatures of the year are nearly the same, the minimum varying between 41 and 53, the maximum between 106 and 111 in different years. The atmosphere is somewhat damper ; the mean humidity of the air being 56 ; that of the driest month 36, and that of the three months, July to September, from 72 to 75 per cent. The spring months are more showery, there being, on an average, 12 rainy days from March to May inclusive. But the monsoon rainfall is equally light, and that of September somewhat less than at Sholapur. In October it is between 3 and 4 inches. It rains on an average on 89 days in the year. Bellary is a very dry station as regards rainfall, indeed one of the 176 CLIMATES AND WEATHER OF INDIA driest in the Deccan, and is rather warmer than any of the foregoing, though elevated 1450 feet above the sea; its mean annual tempera- ture being 80. The mean of December and January is 73, that of April and May 89 and 88, and the lowest and highest readings recorded in each of the last 11 years have varied between 50 and 57 in the former case, and from 106 to 111 in the latter. Owing to the lightness of the monsoon rainfall, the mean temperature from June to September ranges a little over 80, and in October is 79. The air is much drier than at Secunderabad, except in May and June and in the last two months of the year, and especially so from July to September ; the highest mean humidity of any single month being 64 per cent. Occasional showers fall in April and May, but the average rainfall of both July and August is below 2 J inches ; and most rain falls in October, the average of this month being 4 inches. The total mean fall of the year is barely 18 inches, and it rains on an average on only 55 days. Lastly, Bangalore, a favourite climatic resort of the residents of Madras, and one of the largest military stations in Southern India, has a climate only second in attractiveness to that of the Nilgiri hills, and especially suited for raising such fruits and vegetables as thrive in the warm temperate zone. The mean temperature of the year is 73 ; that of December and January 67, and that of the warmest month, April, 80. The highest temperature yet recorded is below 100, and the lowest in the winter have varied between 46 and 56. The atmo- sphere is neither very damp nor very dry at any season. The mean humidity of the year is 66 ; that of the driest month 49, and that of the dampest 77. The rainfall is moderate and well distributed through eight months of the year. January and February are the most rainless months ; but in March there are usually one or two showers, and in April more ; and from May to October from 3 to 6 \ inches fall on an average in each month. The mean rainfall of the year is 35 J inches, distributed over 96 rainy days. The Carnatic. The Ghats, which form the eastern boundary of the great central tableland of the peninsula, are by no means so well defined or continuous as the Sahyadri range on the western margin of the Deccan and Mysore. To the north of the Godavery they rise from the plain of the Northern Circars to the level of the tableland of Jaipur, and are here well defined. But to the south of the Kistna they are represented by a crescentic system of parallel ranges, the successive outcrops of an ancient series of stratified rocks, which advance nearly to the sea-coast, some miles north of Madras ; and from this point an ill- CLIMATES OF THE PLAINS 177 defined escarpment, bounding the tableland of Mysore, trends away to the south-west, leaving a number of more or less isolated and independent hill groups, from 2000 to- 4500 feet in height, between it and the coast plain of the Carnatic. These hills, which bear various names, occupy the centre of the peninsula as far south as the Cauvery valley, in latitude 11. In this latitude, which also coin- cides with the southern limit of the outlying ridges of the Mlgiri hills, the hill country of the peninsula is broadly interrupted. A plain, traversed by the Cauvery river in its- lower course, leads up by a gentle ascent to the foot of the Mlgiris and the outlying ranges, and at Coimbatore has an elevation of 2000 feet above the sea ; and thence descends by a steeper incline to the west coast, separating these latter hills from the Anamalai and Pulni groups, which rise at a distance of from 20 to 25 miles to the south, and rival the Mlgiris in height. These link on to the Travan- core hills farther south, and carry on the line of the western highlands to the southern extremity of the peninsula. The interruption of the Ghats, opposite Coimbatore, known as the Palghat gap, exercises an important influence on the weather of the Central Carnatic on the one hand, and of the Mala- bar coast, south of Calicut, on the other, affording ah unobstructed passage to the easterly and westerly monsoons. In the introduction to their report on the geological survey of this part of the Carnatic, Messrs. King and Foote re- mark : " The south-west wind blows with great force into the Palghat Pass, which has a width of about 25 miles, diminishes a little as the mountains recede and form a funnel, out of which the wind issues. Afterwards it meets with little or nothing in the Coimbatore district to oppose it, till, east of the Cauvery, the Shevaroys and Kolamalais break its force, and receive in heavy showers a large share of its moisture then remaining. . . . Southward of the Kolamalais no ob- struction is offered to the westerly wind, owing to the absence of mountains, the nearest being the Dindigul mountains at a distance of some 40 miles ; and, in consequence, it blows very strongly across the delta of the Cauvery to the east coast. This (westerly) wind, added to the evaporation from the vast area of country laid under water in the delta, during the prevalence of freshets in the Cauvery, tends to N 178 CLIMATES AND WEATHER OF INDIA cool very greatly the intensely hot climate of the Tanjore district. The coast climate, generally, is much more damp and relaxing than that of the more inland parts of the low country, but the thermometer generally has a lower range near the sea." Of the hill groups, which occupy so much of the interior, the same authors write : " The climate of the mountain plateaux is one much more agree- able and suitable to the European constitution in point of temperature, but unfortunately all the ranges are, during the hottest season, to a greater or less degree the seat of malarious fever, giving rise to very dangerous jungle fever. None of the mountains, even the Shevaroys, are above the so-called fever range. 1 . . . April and May are gener- ally the most unsafe months, and during these the jungle regions, especially those among the lower hills, should be carefully avoided. During the cold season, however, especially in December and January, the malaria seems to be quite in abeyance ; it is also said to be so for some weeks after the heavy rains of the south-west monsoon." The width of the coast plain is very variable. To the north of the Pulicat lake the hills advance to within 30 miles of the sea. But to the south of Madras it has an average breadth of about 80 miles, of which the marginal portion, and certain tracts stretching inland along the rivers, are alluvial flats, and the remainder gently undulating and dotted with occasional little rocky hills. Very much of the area is uncultivated, especially the higher grounds. Agri- culture largely depends on artificial irrigation, and except in the deltas of the Godavery and the Cauvery, this is obtained chiefly from tanks, formed by embanking natural depressions in the surface, and fed either by the drainage of the sur- rounding country, or, in some cases, by canals taken off' from the larger rivers, which bring into them a portion of the flood waters. Some of these tanks are 4 or 5 miles across, and form a very characteristic feature of the Car- natic. The waste lands are for the most part covered with bush and scrub, with small trees. Except on the hills, there is little that can properly be termed forest ; but the coast is fringed with groves of cocoa-nut palms, and, in the neigh- bourhood of Madras, extensive plantations of Casuarinas 1 This remark does not refer to the Nilgiris, the Pulnis, or Anamalais, all of which are well above fever range. CLIMATES OF THE PLAINS 179 have been made for the supply of firewood. There is also a protected natural forest on the island, 33 miles in length, that separates the Pulicat lake from the sea. The principal roads are bordered with avenues of pipal, banyan, tamarind, mango, and other trees, and many of the camping grounds near villages are shaded by topes of large and well-grown trees of these and other species. Except wheat and barley, the chief articles of agricultural produce are similar to those raised in other parts of India. Eice is extensively cultivated, and enters largely into the food of the people ; but, as elsewhere, the millets and some inferior grains also contribute an im- portant share. The climate of the Carnatic differs in many important respects from that of other parts of India. The dry season lasts from the middle of December to the latter part of June ; but in April and May there are usually little storms accom- panied with heavy showers, locally known as the "mango showers," heavier on the hills than on the plains, but even on the latter giving an average of from 3 to 5 inches of rain in the two months. The rainfall of June is lighter than that of May. That of the whole summer monsoon, from June to October, is comparatively light, amounting to 4 or 5 inches a month in July, August, and September on the coast, but to not more than one-third that amount in the dry tract to the north of Combatore, and in the district of Tinnevelly in the extreme south. Heavier rain sets in in October, and lasts till the middle of December ; and it is on this that the cultivators chiefly depend for filling the tanks, excepting such as are supplied by feeders from the larger rivers that come down in flood during the summer monsoon. There is in the Carnatic no season that would approve itself as cool to European feelings, otherwise than in a compara- tive sense ; but, for about 6 weeks after the cessation of the late autumnal rains in December, the winds are northerly, the nights pleasant, and the heat of the day not excessive. In the spring months hot land winds blow in the interior, and sometimes down to the coast; but at Madras and other places on the coast, April is the season of the long- i8o CLIMATES AND WEATHER OF INDIA shore winds, winds which, as their name implies, blow from the south parallel with the coast, and though less hot than the land winds, are damp and very relaxing. The climate of the province is generally dry as compared with that of the west coast or of Bengal and Arakan ; but that of the interior, and of the northernmost and southern- most districts, much drier than that of the coast region from the northern extremity of the Pulicat lake to Point Cali- mere. The average annual rainfall of this latter tract, to a distance of from 10 to 20 miles inland, exceeds 40 inches. But in the northern portion of the Nellore district, the greater part of Coimbatore, and on the coast of Tin- nevelly it is below 20 inches in the year, and over much larger areas around these below 30 inches. Some of these phases of the climate are illustrated by the five stations, the meteorological statistics of which are given in the Appendix. These are Masulipatam and Mad- ras on the coast ; the first in the extreme north, on the sea face of the Kistna delta, the second on a projecting portion of the coast south of the Pulicat lake ; Coimbatore, on the dry plain that extends between the Nilgiri and Shevaroy hills, not far from the Palghat gap ; Trichinopoly, at the head of the Cauvery delta, almost due east of Coimbatore ; and Madura, 70 miles south of Trichinopoly, on the plain to the east of the Travancore hills. The mean temperature of Masulipatam is 81 ; that of December and January 74, and that of May, the hottest month, 88. In no month does the afternoon maximum average less than 83, or the night minimum less than 66 ; and in May the former is as high as 99. The highest reached in the course of the year varies between 101 and 116. The lowest readings are less variable, being between 58 and 62. Owing to the situation on the sea-coast of a great river delta, the air is never very dry, nor the range of temperature during the day very great. The latter averages 13 in the dampest months and 19 in the spring. The air is driest in June, when the humidity is 67 per cent, and dampest in October and November, when it is 78 ; the average of the year is 74. The mean rainfall of the year is about 38 inches, of which only \\ falls from December to April. In the first three months of the monsoon the monthly fall does not exceed 5 or 6 inches ; but it is somewhat greater in September, and heaviest in CLIMATES OF THE PLAINS 181 October (8j inches). In November it amounts to about 4 inches, and ceases a full month earlier than in the central and southern districts of the Carnatic. The number of rainy days has varied in the last 11 years from 61 to 101 in the year. Madras, although also situated on the sea-coast, has a drier atmo- sphere and a somewhat higher temperature than Masulipatam. Its mean temperature is 82; that of December and January 76, and that of June 88. The afternoon temperatures are lowest in December (83), but the early morning readings are somewhat lower in January, when they average 68. The lowest reached in the course of the year, since the thermometers have been properly exposed, have been between 57*6 and 62. In former years they were many degrees higher. The highest readings, which are recorded with the hot land winds gener- ally in May, have been betw r een 102 and. 113 in different years. The daily range of the thermometer is much the same as at Masulipatam. At the observatory, which is nearly 3 miles from the sea, the mean humidity is 71. The driest month is June, when it averages 61, and the dampest November, when it is 79 per cent of saturation. The mean annual rainfall is only a little under 50 inches. In the wettest year on record (1827) it amounted to 88-4 inches, and in the driest (1832) it was only 18*5 inches. From June to September there is less rain than at Masulipatam, the average of each of these three months not exceeding 4 or 5 inches. But that of October is nearly 1 1 inches, and that of November, which is the wettest month of the year, nearly 14 inches. There is usually a spell of heavy rain in the early part of December ; it is in this and the two preceding months that little cyclones frequently form off the coast, whence they pass across the Carnatic bringing a deluge of rain. From January to the end of April there is but little rain, and in May only a few thunder- storms, unless, as sometimes happens, the Carnatic is visited by a cyclone from the bay. Trichinopoly has the same mean temperature as Madras, viz. 82, but while it is 1 or 2 warmer from February to May, it is 1 or 2 cooler from June to September. Still less than at Madras is there any cool season, properly so called. Even in December and January the afternoon readings of the thermometer average 85 and 87, but the nights are comparatively cool at this season, though the thermometer rarely sinks below 60 on the coldest nights in the year. May is the hottest month, when the mean temperature is 88 ; but although the average maximum of the afternoon is 102, no reading higher than 108 has been recorded, at all events during the last 12 years. In June the temperature falls with the setting in of the strong west wind through the Palghat gap, as already described, and this brings cloudy skies ; but the atmosphere remains dry and there is but little rain. It is very dry, indeed, at all times, except in the last three months of the year. The mean humidity is 63 per cent, and in April, the driest month, only 54 per cent. Even in June 1 82 CLIMATES AND WEATHER OF INDIA and July it is only 57 per cent, but in November it rises to 76 per cent. The mean annual rainfall is 37 inches. Up to the latter part of April but little falls, but at the end of this month and in May thunder-storms are frequent, and the average rainfall of the latter month is nearly 4 inches. June and July again are dry months, there being on an average only six rainy days in the two months. But rain becomes more frequent in August and September, and reaches its maximum in October, in which month there are about 11 rainy days. November and the first half of December are also more or less rainy, but the rainfall is lighter than at Madras. Coimbatore also has a very dry climate, but being situated at a high level in the neighbourhood of lofty hills and within the influence of the Palghat gap, it is by no means a very hot station for the lati- tude. The mean temperature is 78, the same as that of Calcutta, and 4 lower than that of Trichinopoly and Madras, and in no month of the year does it deviate more than 5 above or 4 below this mean. In December and January it is 74, in April 83, and from June to September it fluctuates between 76 and 78. But the diurnal range is as much as 27 in February and March, and 17 in the dampest months of the year. The lowest readings of the year have varied between 54 and 61 in different years, and the highest between 99 and 104 in the shade. The mean humidity is 66 per cent, and that of February, the driest month, 52 per cent. From this time it rises gradually to August, when it is 73 per cent, and in October, the dampest month, it does not exceed 75. There is less cloud than at Trichinopoly, and the rainfall is much lighter, viz. little above 21 inches in the year. Of this, showers in April and May con- tribute rather more than 4 inches. Then, from June to September, although there are from 6 to 10 rainy days in the month, the average monthly rainfall is barely \\ inches ; in October and November the total fall amounts to 9 or 10 inches, and after the middle of December there is but little rain till the April showers. Although, as already stated, the rainfall is only about 21 inches, it rains on an average on 85 days in the year. Madura is a warmer station, having the same mean temperature as Trichinopoly and Madras (82). In December and January it is 77, and in April and May 86. On the coolest night of the year it rarely falls to 60, and the mean minimum of the three months, December to February, is between 68 and 70. In April and May the afternoon maximum readings average 101, but the highest re- corded temperatures of the last 12 years have not been higher than between 104 and 107. From February to May inclusive the daily range averages 25. The mean humidity of the year is 65, and in the driest months, April, June, and July, it falls to 59. In the last three months of the year it averages from 73 to 75. The rainfall is much higher than that of Coimbatore or than that of the plains to the south-east, viz. 35 inches in the year ; and of CLIMATES OF THE PLAINS 183 this, nearly 5 inches fall in April and May, about 9 inches altogether in August and September, 9 inches in October, and 7J inches in the two final months of the year. Although the rainfall is thus 60 per cent heavier than at Coimbatore, it rains less frequently, the average number of rainy days being only 59 in the year. Ceylon. In the course of the foregoing pages we have seen that, in passing from north to south, the different seasons of the year become less and less contrasted, the temperatures of the winter and summer months more and more equal, and at length, in the south-west of the peninsula, on the coast of Travancore, beyond the alterna- tion of a wet and a dry season, there is little to distinguish the different months of the year. In Ceylon, and especially in its south-west province, we have these essentially tropical, or rather equatorial, features of climate in their strongest development. At Point de Galle the difference of the highest and lowest readings of the thermometer in the course of the year amounts, on an average, to no more than 19, and in some ye^rs has not exceeded 17, a difference less than half that of a single day and night at certain seasons in the Punjab. But, despite the moderate dimensions of the island (less than 300 miles in length and 140 in width), the climate of Ceylon is by no means the same in its several parts. Its east and west coasts, if not presenting so great a contrast as the dry plains of Tinnevelly and the evergreen forests of Travancore on the Indian mainland, exhibit differences of the same kind. The heavy rainfall of the summer monsoon is restricted to the south-west of the island, while the eastern provinces receive their most copious rain in November and December. At this latter season the rain is not restricted to the eastern half of the island; it extends not only to the hills, but also to the south-west province ; and in such measure that, both at Galle and Colombo, the average fall of October and Novem- ber is as heavy as that of May, and considerably heavier than that of June and those months when the summer monsoon blows most strongly on the coasts of India. The south-west province, comprising the generally rugged 1 84 CLIMATES AND WEATHER OF INDIA country between the central hills and the coast from Galle to Colombo, is that in which frequent and copious rain and constant damp heat foster the luxuriant vegetation that has made Ceylon proverbial for its beauty among the beautiful islands of the tropics. But on the north-west coast, opposite to India, and also on the south-east coast, the rainfall is comparatively light, and a jungle of thorny acacias, euphorbias, and deciduous trees, such as characterise the drier parts of the Carnatic, replace the dense evergreen forests of the more humid tract. The climate of Ceylon has been frequently described, and by no one more fully than Sir Emerson Tennant, from whose work the following passages are extracted in a much abbreviated form : " The island is seldom visited by hurricanes, and the breeze, nnlike the hot and arid winds of Coromandel and the Deccan, is always more or less refreshing. The range of the thermometer exhibits no violent changes, and never indicates a temperature insupportably high. The mean, on an annual average, scarcely exceeds 80 at Colombo, though in exceptional years it has risen to 86. But at no period of the day are dangerous results to be apprehended from exposure to the sun, and except during parts of the months of March and April, there is no season when moderate exercise is not practicable and agreeable. For half the year, from October to May, the prevailing winds are from the north-east, and during the remaining months the south-west monsoon blows steadily from the great Indian Ocean. The former is subject to many local variations and intervals of calm. But the latter, after the first violence of its onset is abated, becomes more nearly uniform throughout the period of its prevalence, and presents the character of an on-shore breeze, extending over a prodigious expanse of sea and land. " There are, of course, abnormal seasons with higher ranges of tem- perature, heavier rains, or droughts of long continuance, but such extremes are exceptional and rare. Great atmospheric changes occur only at two opposite periods of the year, and so gradual is their approach that the climate is monotonous, and one longs to see again the falling of the leaf to diversify the sameness of perennial verdure. The line is faint that divides the seasons. No period of the year is divested of its seedtime and its harvest in some part of the island, and the ripe fruit hangs on the same branches that are garlanded with opening buds." The following more detailed description is given of the climate of Colombo : CLIMATES OF THE PLAINS 185 " At the opening of the year the north-east monsoon, which sets in two months previously, is nearly in mid-career. It reaches Ceylon comparatively dry, and its general effects are parching and disagreeable. This character is increased as the sun recedes towards its most southern declination, and the wind acquires a more direct draught from the north ; passing over the Indian peninsula and almost totally divested of humidity, it blows down the western coast of the island and is known there by the name of the ' along-shore wind.' For a time its influence is uncomfortable and its effects injurious to both health and vegetation. It warps and rends furniture, dries up the surface of the earth, and withers the delicate verdure. These characteristics, however, subside towards the end of January, when the wind becomes somewhat variable with a westerly tendency and occasional showers, and the heat of the day is then partially compensated by the greater freshness of the night. " February is dry and hot during the day, but the nights are cloudless and cool. Eain is rare, and when it occurs it falls in dashes, succeeded by damp and sultry calms. The wind is unsteady, and shifts from north-east to north-west, sometimes failing entirely between noon and twilight. The difference of temperature between day and night is frequently as great as 15 or .20. " In March the heat continues to increase. The day becomes oppressive, the nights unrefreshing, the grass is withered and brown, the earth hard and cleft, the lakes shrunk to shallows, and the rivers evaporated to dryness. Europeans now escape from the low country and betake themselves to the shade of the forests adjoining the coffee plantations on the hills, or to the still higher sanitarium of Newara Eliya. The winds, when any are perceptible, are faint and unsteady with a still increasing westerly tendency ; partial showers sometimes fall, and thunder begins to mutter towards sunset. " April is by far the most oppressive portion of the year for those who remain at the sea-level of the island. A mirage fills the hollows with mimic water. The heat in close apartments becomes extreme, and every living creature flies to the shade from the suffocating glare of mid-day. At length the sea exhibits symptoms of an approaching change ; a ground swell sets in from the west, and the breeze towards sunset brings clouds and grateful showers. "May is signalised by the great event of the change of the mon- soon and all the grand phenomena which accompany its approach. About the middle of the month, but frequently earlier, the sultry suspense is broken by the arrival of the wished-for change. As the monsoon draws near the days become more overcast and hot, banks of clouds rise over the ocean to the west. At last the sudden lightnings flash among the hills and sheet through the clouds that overhang the sea, and with a crash of thunder the monsoon bursts over the thirsty land, not in showers or partial torrents, but in a wide deluge that in the course of a few hours overtops the river banks and spreads in inundations over every level plain. 1 86 CLIMATES AND WEATHER OF INDIA "The extreme heat of the previous month becomes modified in June ; the winds continue steadily to blow from the south-west, and frequent showers, accompanied with lightning and thunder, serve still further to diffuse coolness throughout the atmosphere and verdure over the earth. "July resembles to a great extent the month that precedes it, except that in all particulars the season is more moderate, showers are less frequent, less wind and less absolute heat. In August the weather is charming, notwithstanding a slight increase of heat. The same atmospheric condition continues throughout September, but towards its close the sea breeze becomes unsteady and clouds begin to collect, symptomatic of the approaching change to the north-east monsoon. The nights are always clear and delightfully cool. Rain is sometimes abundant. " October is more unsettled, the wind veering towards the north with rain pretty frequent. November sees the close of the south-west monsoon and the arrival of the north-eastern. In the early part of the month the wind visits nearly every point of the compass, but shows a marked predilection for the north. The great change is heralded as before by oppressive calms, lurid skies, vivid lightning, bursts of thunder and tumultuous rain. But at this change of the monsoon the atmospheric disturbance is less striking than in May, the previous temperature is lower, the moisture of the air is more reduced, and the change is less agreeably perceptible from the southern breeze to the dry and parching wind from the north. " In December the wind setting in steadily from the north-east brings with it light but frequent rains from the Bay of Bengal. The morning and the afternoon are again enjoyable, but at night every lattice that faces the north is cautiously closed against the treacherous ' along-shore wind.' " With respect to the climate of Kandy, Sir Emerson Tennant writes : " Kandy from its position shares in the climate of the western coast, but from the frequency of the mountain showers, and its situa- tion at an elevation of upwards of 1600 feet above the level of the sea, it enjoys a much cooler temperature. It differs from the low country in one particular which is very striking the early period of the day at which the maximum heat is attained. This, at Colombo, is generally between two and three o'clock in the afternoon, whereas at Kandy the thermometer shows the highest temperature of the day between ten and eleven o'clock in the morning. In Kandy the nights are so cool that it is seldom, that warm covering can be altogether dis- pensed with." And of Trincomalee : CLIMATES OF THE PLAINS 187 " At Trincomalee the climate bears a general resemblance to that of the Indian peninsula south of Madras ; showers are frequent but light, and the rain throughout the year does not exceed forty inches. 1 "With moist winds and plentiful dew this sustains a vigorous vegeta- tion near the coast, but in the interior it would be insufficient for the culture of grain were not the water husbanded in tanks ; and the bulk of the population are, for this reason, settled along the banks of the great rivers. " In the extreme north of the island, the peninsula of Jaffna and the vast plains of Neura Kalawa and the Wanny form a third climatic division, which, from the geological structure and the peculiar configuration of the district, differs essentially from the rest of Ceylon. This region, which is destitute of mountains, is undulating in a very slight degree ; the dry and parching north-east wind desiccates the soil in its passage, and the sandy plains are covered with a low and scanty vegetation, chiefly fed by the night dews and whatever moisture is brought by the on-shore wind. The total rain of the year does not exceed thirty inches, 2 and the inhabitants live in the frequent appre- hension of droughts and famines. These conditions attain their utmost manifestation in the extreme north, and in the Jaffna peninsula ; there, the temperature is the highest in the island, and owing to the humidity of the situation and the entire absence of hills it is but little affected by the changes of the monsoons. The soil, except in particular spots, is porous and sandy, formed from the detritus of the coral rocks which it overlays. It is subject to droughts, sometimes of a whole year's continuance, and rain, when it falls, is so speedily absorbed, that it renders but slight service to cultivation, which is entirely carried on by means of tanks and artificial irrigation." The agricultural and horticultural products of Ceylon are even more varied than those of most parts of India. There is perhaps no plant native to the tropics, and but few sub-tropical forms of vegetation, that may not be raised either on the low country or the hills ; and the actual ex- ports, in addition to tea, coffee, and cinchona, which are (or were) the chief produce of the plantations, include the cocoa- bean, cocoa-nuts, india-rubber, cardamoms, cinnamon, cloves, nutmegs, pepper, and many other spices, vanilla, betel nuts, and most of the agricultural productions of India. Coffee, which, twenty years ago, was the great staple article of ex- 1 Becent registers show that this is much underestimated. The rainfall of Trincomalee is over 60 inches. But it holds good of the west coast in the latitude of Trincomalee. 2 This again is underestimated. The average rainfall of Jaffna is 49 inches, and that of Manaar 40 inches, as the result of 16 years' registers. 1 88 CLIMATES AND WEATHER OF INDIA port, has suffered so severely from the attacks of a fungus, known as the leaf-disease, that from 1877 to 1883 the exports sank from 850,000 to 250,000 hundredweights, and the average produce from 4J to less than 1-J- hundred- weights per acre. Many of the estates, formerly devoted exclusively to coffee, have, in consequence, been either abandoned to jungle, or partly planted with tea, cinchona, or cocoa ; and at the present time the cultivation of tea, more especially, is advancing by leaps and bounds, in such measure that Ceylon is already a formidable rival of Assam in the production of this important staple. Those who may desire to obtain detailed information on these and kindred subjects may be referred to Ferguson's Ceylon Handbook and Directory, a most -valuable compendium of general information relating to the colony. The detailed climatic statistics of three stations are given in the Appendix, viz. Trincomalee, on the north-east coast, representing the drier, but not the driest portion of the island ; and that which, like the Carnatic, receives its chief rainfall during the transition from the south-west to the north-east monsoon. Kandy, which may be taken as re- presentative of the climate of the lower hill region, the chief seat of tea and coffee culture, for which it enjoys the great advantage of almost constant humidity, a very equable tem- perature, and a copious rainfall at two opposite seasons. And finally, Colombo, which is the seat of government, the commercial capital and chief port of the island. The mean temperature of Trincomalee is the same as that- of Madras and the southern Carnatic, viz. 82, but the annual range is less ; the temperature of December and January being only 4 below the mean, and that of the four months, April to July, 3 above it. The mini- mum night temperatures are even more uniform, averaging between 74 and 77 throughout the year, and never falling below 65. The day maximum averages 83 in the coolest and 93 or 94 in the warmest months, and reaches to between 97 and 102 only on the hottest days of the year. The mean humidity of the observatory is 72 per cent, in November 82 per cent of saturation, and in July, here the driest month of the year, 65. From May to September, that is, all through the south-west monsoon, it averages below 70, owing to the position of the station on the lee coast of the island, so that the south- CLIMATES OF THE PLAINS 189 west monsoon here blows . as a land wind. The mean annual rainfall is 62 inches ; of which two-thirds fall in the four months, October to January, and only one-seventh from March to July inclusive. There are, on an average, 104 rainy days in the year, of which also two-thirds occur in the four months, October to January. Kandy has a still more uniform climate, with a mean temperature of only 76. That of January is 74, and that of the three warmest months, March to May, 79. The minimum night temperatures range from 68 to 73 in different months, and the maximum day tempera- tures from 80 to 87. The highest temperature yet recorded is be- low 95, and the lowest 54. The mean humidity is 77 per cent of saturation, varying from 69 in April, the driest month, to 81 in June and 80 in July and also from October to December. The average rainfall is 84j inches, and this is more evenly distributed through the year than is the case on either coast of the island. October and November have on an average more than any other months, and February and March the least. But in no month does the average fall exceed 12 inches or fall below 2j inches, and excepting in Febru- ary and March, it rains on an average on from 10 to upwards of 20 days in each month. The average number of rainy days in the year is 185. Colombo is 5 warmer than Kandy, but the annual range of tempera- ture is even less. The mean of the year is 81, but that of the warm- est months is only 2 higher, and that of the coolest 2 lower than this mean. On the average of 12 years, the difference of the highest and lowest readings in the year is but 25, and that of the highest and lowest yet recorded only 5 greater. In February alone, the range of the thermometer is nearly as great as in the whole course of the year. The humidity of the atmosphere is high, viz. 78 per cent of saturation, and even in the driest month, February, it averages 73 per cent ; but it occasionally falls much below this during the heat of the day, with a north wind. The average rainfall is 87 inches, of which between 12 and 13 inches falls in each of the three months May, October, and November, and less than 2 inches in February. In most other months except January, the fall is little if at all below 5 inches. There are fewer rainy days than at Kandy, viz. 159 in the year on an average. Burmah. Of the climate of Burmah I am able to give a detailed account only in so far as relates to the provinces which have been under British rule since 1825 and 1852 respectively, viz. Arakan, Tenasserim and Pegu. In the case of the recently-annexed province of Ava, it is known generally that the climate is drier and the annual range of temperature greater than in the older maritime provinces,, but the only statistical data at present available are. the IQO CLIMATES AND WEATHER OF INDIA registers of nine months of 1 8 7 9 at Mandalay, recorded by the late Mr. E. Shaw, and those of two months of 1886. An abstract of most of these is given in the Appendix. The observatories which have now been established at Bhamo and Kindat, in addition to that of Mandalay, will, in the course of a few years, afford a better idea of the characteristic climatic features of the newly -acquired territory. The position and configuration of Arakan and Tenasserim, on the west coast of the peninsula, with hill ranges running parallel with the coast, expose them to the influence of the south-west monsoon of the Bay of Bengal, in the same manner and as fully as are the Konkan and Malabar to that of the Arabian Sea ; and with a similar result, viz. an excessive rainfall from June to September. In Arakan, however, this rainfall is more prolonged than on the west coast of the Konkan in the same latitudes. The southerly monsoon continues to blow in the eastern half of the bay for some weeks after it has ceased in the western half and in India generally, and the stormy weather which frequently prevails on the bay in October and the beginning of November is always accompanied with southerly winds, which discharge a good deal of rain on the hills of Arakan, as well as on those of the more southerly portion of the Malay peninsula. In Pegu, which lies wholly east of Arakan, and to lee- ward of the coast range, the sea face of the Irawadi delta shares the heavy rainfall of the coast generally ; and the hill ranges to the east of the delta, the Pegu Yoma, and that which separates the Sitang from the Salwin, also have a high rainfall ; but the Irawadi valley above Prome is much drier, and this character becomes more and more marked as one ascends the river to Mandalay. The valley of the Irawadi consists of plains intersected by low isolated ranges of hills, running north and south ; and these, at certain places, hem in the main stream of the river, forcing it to flow in a narrow but deep channel. On the west, the valley is bounded by the Arakan Yoma, CLIMATES OF THE PLAINS 191 which commences at Cape Negrais, and consists of a belt of parallel ridges, one of which rises to 4000 feet in about lat. 18, but sinks again to about 1000 feet at the Pegu frontier. Farther north, it gradually increases in elevation as well as in width; and in latitude 22 30' rises in the Blue Moun- tain to a height of 7100 feet. The extensive hill tract covered by this range and its off-shoots between the pro- vinces of Ava on the east and Arakan arid Chittagong on the coast, is occupied by native hill tribes, and covered with forest. But few parts of it have ever been visited by Europeans. To the east of the Irawadi the Pegu Yoma, a range not more than 2000 feet in height, intervenes between that river and the Sitang, and contains some valuable teak forests. Another and much loftier range separates the latter from the Salwin, and its northern extension becomes a broad series of parallel ridges occupied by the Karens, and east of Bhamo by the Kakhyens. The forests of these hills and those of the independent Shan states east of the Salwin have for many years past furnished the chief supplies of teak timber to the ports of Rangoon and Moulmein. The forests of Burmah and the neighbouring states contain many other valuable timber trees, among which Padouk is already in some demand both in India and Europe. The chief agricultural export of Burmah is rice. The whole of Burmah is swept by the summer monsoon from the Bay of Bengal, which impinges obliquely on the coast and, as already observed, discharges a very heavy rain- fall on the seaward face of the coast ranges and on the marginal portion of the Irawadi delta. On the plains of this river it prevails as a south wind, its southerly direction being probably determined by that of the numerous hill ranges ; but it is also probable that it passes over the crest of the higher ranges with its normal south-west direction, blowing towards the plains of China and the slopes of the eastern Himalaya. Ava, which lies under the lee of the Arakan coast range, has a comparatively dry climate and a vegetation which, according to Hooker and Thomson, re- 192 CLIMATES AND WEATHER OF INDIA sembles that of the Carnatic ; but the northern part of the province beyond Bhamo is damper, owing to the loftier mountains and the more irregular surface of the country, and the vegetation is very similar to that of Assam. In the winter months the winds are light and northerly ; up to the end of January, chiefly north-east at Thyet Myo and in the Irawadi delta, and north-west at Tounghoo. At the end of January they become north-westerly and gradually back to south-west with the increasing heat of the land. From December to April but little rain falls in any part of Burmah, with the partial exception of the most southern districts of Tenasserim ; but in May rain becomes frequent, and during the succeeding four months a rainless day occurs but rarely, save only on the plains of the Irawadi north of the delta. During this season the atmosphere is saturated with damp, and at Rangoon and Moulmein the temperature is moderate, not exceeding an average of 83 or 85 at the warmest time of day, and sinking to 74 or 76 at night. In Ava, however, where the rainfall is light, the temperature of this season is much higher, sometimes reaching to nearly 100 in the shade, to judge from the Mandalay registers of 1879. A very marked feature of the climate of Pegu, and one which probably goes far to explain its notorious unhealthi- ness, is the great change of temperature during the 24 hours, combined with the prevailing dampness of the atmosphere. At stations in the interior, such as Thyet Myo and Tounghoo, the difference between the thermometer readings at sunrise and in the afternoon, in the earlier months of the year, is as grdkt as in the driest parts of the Punjab, and some 10 or 12 greater than in Bengal, although the atmosphere is not drier than in the latter province. As all the meteorological observations have been recorded under exactly similar conditions at all stations, there is no reason whatever to question the validity of this comparison. Abstracts of the chief climatological data of seven stations are given in the Appendix, viz. Akyab, on the coast CLIMATES OF THE PLAINS 193 of Arakan ; Mandalay (for 1 1 months only), as repre- senting Ava ; Thyet Myo, in the driest portion of Pegu, 1 1 miles from the northern frontier of the province ; Tounghoo, in the Sitang valley, east of the Irawadi and the Pegu Yoma ; Eangoon, in the eastern corner of the Irawadi delta, and 25 miles from the embouchure of the eastern branch of the river ; Moulmein, on the Sal win, 24 miles above where it enters the Gulf of Martaban ; and Mergui, on the sea coast of Tenasserim, near the southern extremity of the province. The annual mean temperature of Akyab is 79, or 2 higher than that of Chittagong, on the same coast but 160 miles farther north. That of January, which month, owing to the fall of the night tem- perature is 3 cooler than December, is 69, and that of April and May 84. In July and August the mean temperature sinks to 81, the maximum of the day falling from 92 to 84 or 86. The mean minimum temperature in January is 59, and the lowest in the year varies from 47 to 55. The highest yet recorded is below 100, and has ranged between 93 and 99 in the past 13 years. In Feb- ruary the diurnal range averages 25, but at the height of the rainy season sinks to 7 or 8. The mean humidity of the station, which is situated on a narrow point of land between the sea and the broad estuary of the Koladyne river, is 80 per cent of saturation, falling to 70 in February and rising to 89 in July and August. The mean annual rainfall is 196 inches, but from December to March inclusive the fall is insignificant, and that of April under 2 inches. In May, however, rain is frequent, amounting to between 12 and 13 inches on an average on 14 rainy days, and the average fall in each of the months June and July is over 50 inches. In August the fall is lighter but not less frequent, and even in September it rains on two days out of three, and in October on more than one in three. The average number of rainy days in the year is 140. For Mandalay we have complete registers for only 11 months, and relating to two different years, and rainfall registers for 17 months, including two monsoons. From the former, a mean tempera- ture of 81 has been computed, but this is probably rather higher than the true normal average. That of January 1879 was 71, and that of April 91. The mean minimum of January was 59, and 52-9 was the lowest reading recorded. In April and May the mean of the maxima was over 100, and the highest reading registered 106-1. The winter temperature was considerably higher than that of places in the same latitude in India, and only in January a little lower than that of Cuttack, lj farther south. After April the mean temperatures were about the same as those of Cuttack, 194 CLIMATES AND WEATHER OF INDIA but the rainfall being lighter the afternoon temperatures were higher at Mandalay in July and August. The mean atmospheric humidity of the year was 64 per cent of saturation, and that of March 43. September was the dampest month, when it rose to 77 per cent, and September and November (October being wanting) were the dampest months in the year. The rainfall, as far as can be deduced from the 17 months' registers, is only 27 J inches, and in no month apparently does it rain more frequently than about one day in three. The first three months of the year were rainless. Thyet Myo has a mean temperature of 79, being therefore ap- parently cooler than Mandalay, although it lies more than 2 farther south ; but this must be considered as doubtful. The mean January temperature is 68, and that of April, the warmest month, 87. In January the mean of the morning minima is 54, and the lowest read- ings of each year have varied between 40 and 49. In April the mean of the afternoon maxima is 103, and 110-2 is the highest read- ing yet recorded. In the first three months of the year the average daily range is from 31 to 36, which is as great as in the driest months in the driest part of the Punjab, and must be the more trying to health that the atmosphere is by no means so dry as that of the latter province. The mean humidity of the year is 72, and that of the driest months, March and April, 56 per cent. In the rainy season (July and August) it is as high as 84. The rainfall averages about 45 J inches in the year, and the four months, December to March, are almost rainless. As in other parts of Burmah, the rainfall of May is considerable, viz. 5 inches ; but during the summer monsoon it is only 7 or 8 inches in the month, and as much as 5 inches in October. It rains on an average on 107 days in the year. Tounghoo is slightly cooler than Thyet Myo, the mean temperature being 7 8. The temperature of January is, however, relatively higher, viz. 70, and that of April lower, viz. 85. The mean minimum of the mornings in the former months is 57, and the lowest of the year between 47 and 54. In April the afternoon maxima average 100, the highest varying from 101 to 107 in different years. The range during the day in the first three months is not quite so great as at Thyet Myo, but still very considerable, viz. from 29 to 32 on an average, and in the rainy season it amounts to 12 or 15. The mean humidity is 76 per cent of saturation, and in March not below 61 per cent. In July and August it averages 87 and 88 per cent. The mean annual rainfall is 78 inches, scarcely any of which falls in the first three months of the year. But the rainfall of May amounts to over 6 inches, and the rains continue till quite the end of October. The average number of rainy days in the year is 135. Eangoon has a damper and more equable climate. The mean tem- perature is the same as that of Akyab and Thyet Myo, viz. 79 ; but the temperature of the winter months is higher and that of the rainy season lower, so that there is a difference of only 3 between the mean CLIMATES OF THE PLAINS 195 temperature of January and those of July, August, and September. Indeed the afternoon maxima are lower from June to September than in any other months of the year. In January the mean temperature is 75 ; that of the morning minima 64, and that of the afternoon maxima 88. Even here, on the margin of the Irawadi delta, within a few miles of the sea, and surrounded by broad rivers, the diurnal range is between 20 and 30 in the first four months of the year. In April, the mean temperature is 84, the afternoon maxima average 98 and they often exceed 100. But heavy rain begins in May, rapidly bringing down the heat of the daytime, and from June to September the mean temperature is only 78 or 79, the afternoon maxima 85 or 86 on an average, and the diurnal range only 9. The mean humidity of the station is 78, that of February 62, and that of July and August 90 and 91 per cent. The mean annual rainfall is but little below 100 inches, and of this less than one inch falls from December to March inclusive. From June to September, on the other hand, a rainless day is rare, and in October there are about 14 rainy days. The average number in the year is 153. Moulmein has a very similar climate in point of temperature and general humidity, but a much higher rainfall. The winter tempera- tures are about the same as at Rangoon, but from April to September they are about 1 lower, and on the hottest days in the year the ther- mometer rarely reaches 100. The diurnal range of temperature is either the same or rather less at the different seasons, and the humidity of the atmosphere less in the winter months, and about the same in the rains. From December to March the weather is equally rain- less, but in the remaining eight months the rainfall averages not less than 188 inches. In May it is as much as 19 inches, and in July and August there is an average of nearly an inch and a half a day. In these months the atmosphere is always near saturation. Mergui, the most southerly station in Burinah, resembles places on the south-west of Ceylon in the equability of its temperature through- out the year, but the mean is 3 lower, viz. 78. July and December are the coolest months, and have the same mean temperature, viz. 76, but while the average maximum of the day is 88 in December, in July it is only 84. April, the warmest month, has a mean tempera- ture of 81, and an afternoon maximum of 92. In most years the thermometer never sinks to 60 on the coolest morning, and it has never risen to 100 on the hottest days during the last 9 years. The mean humidity is high, viz. 82 per cent, and in no month does it fall below 73 ; while in August and September, which are here the dampest months, it rises to 90 per cent of saturation. The dry season is shorter and less dry than anywhere to the north. December and January are the only months in which the rainfall averages less than an inch, but in the summer monsoon the fall is less heavy than at Moulmein, and the mean rainfall of the year is about 160 inches. It rains on an average on 168 days. . -*- 196 CLIMATES AND WEATHER OF INDIA Andaman and Nicobar Settlements. The climate of these islands, which have been occupied as convict settle- ments since 1858, is almost equatorial in its uniformity and in many respects similar to that of Tenasserim. The Andamans are hilly, the hills being for the most part only a few hundreds of feet in height, and covered with forest, except where clearings have been made around the settlement. Mangrove swamps extend in places along the sea margin, and outside is a fringe of coral reef. The JSTicobars, as far as they are known, are similar ; but owing apparently to the magnesian character of certain of the rocks, there are large tracts, the dense clay soil of which is unfavourable to vegeta- tion, and these constitute grassy plains, bare of forest. The Mcobar settlement of Nancowry has a much worse reputa- tion than that of the Andamans for health, and this appears to be due to the different characters of the sites selected. On the latter islands clearings have been made in the once unbroken forest. The soil is porous, and admits of good drainage ; but in the former, the grass land has been occupied, perhaps on account of its openness, but the stiff clay, which does not support vegetation, is impervious to drainage, and an obstinate form of fever is very prevalent among the settlers. Both settlements are warmer than the Tenasserim mainland in the same latitude, the mean temperature of Port Blair and Nancowry being 80, while that of Moulmein and Mergui is only 78. This excess may possibly be due to both the former having an eastern aspect, and to their being therefore on the lee-side of the islands in the south-west monsoon. There is but little variation during the year ; March and April are the warmest months, with mean temperatures of 82 and 83 respectively, and a mean daily maximum of 92 at Port Blair and 89 at Nancowry. A reading of 9 9 '5 has been once re- corded at Nancowry, but at Port Blair 96'4 is the highest, and 94 or 95 on an average, the highest in the year. The lowest yet re- corded at Port Blair is 66, and at Nancowry 70*2. The average extreme range of temperature in the course of the year is only 26 at the former, and 22 at the latter. The diurnal range of temperature at Nancowry never averages more than 10 or 11 in any month, and varies but little at different seasons. But at Port Blair it is as much as 14 or 15 in the driest months, February, March, and April. Port Blair is the damper station of the two ; the mean WEATHER AND WEATHER REPORTS 197 humidity being 83, while that of Nancowry is only 79 per cent of saturation ; and this difference appears to hold good in all months. The rainfall of Port Blair is rather heavier, especially during the south-west monsoon; but Nancowry has more rain from November to April inclusive. At both stations the monsoon rains set in in May, and the rainfall of that month is little less than that of June. The number of rainy days is about the same at both stations, and amounts to nearly half the days in the year ; but during the summer monsoon, while at Port Blair there are only 5 or 6 rainless days in the month, there are about twice as many at Nancowry. WEATHER AND WEATHER EEPORTS. Climate and Weather. The distinction of climate and weather in the language of everyday life is too familiar to need more than a cursory reminder of the sense in which the terms are here employed. By the former are meant the average conditions of heat and cold, damp and dryness, and the like, characteristic of a place or country, in so far as they vary regularly with the succession of the seasons. The latter has reference to the apparently irregular, though not always minor, atmospheric changes that take place from day to day. Climate depends on the latitude, on the height above sea-level, on the distance from the sea, on the form and slope of the land surface, on the nature of .the soil and its vegetation, and a number of other circumstances, for the most part permanent, and generally easy to ascertain. The vicissitudes of the weather stand in very different case ; and although of late years much has been done to determine fixed relations between weather changes and those great movements of the atmosphere that immediately give rise to them, the ulterior causes that produce these movements can be indicated only in very general terms. In great part they would appear to be beyond the range of our field of observation, and can be arrived at, if ever, only by a slow process of inference. Anticyclones and Cyclonic Depressions. The modern practice of weather forecasting is based on the known mutual relations of those tracts of relatively high and low 198 CLIMATES AND WEATHER OF INDIA barometric pressure that have been described at page 24 of this work ; and which are respectively the source and goal of the surface winds. The former are generally termed anticyclones, the latter, barometric depressions, or cyclonic systems, or simply " cyclones " ; though this extension of the latter term, from the hurricanes of the tropics (to which alone it was originally applied), to all barometric depressions, is attended with some inconvenience. As has been ex- plained at page 46, the terms "cyclone" and "anticyclone" have reference to the course of the winds around the centres of low and high pressure respectively ; which courses are different in the northern and southern hemi- spheres, though constant in the same hemisphere. In the northern hemisphere, with which we are now concerned, the winds blow spirally outwards from an anticyclone, the curve of the spiral being with the movement of the clock hands, or " clockwise " ; and they blow spirally inwards to a cyclonic centre against the clock movement, or counter- clockwise. Each of these systems has its characteristic type of weather. In an anticyclone, as a rule, in all parts of the world and always in India, the weather is fine, and either the skies are cloudless, or the cloud, if any, is high ; except that in damp regions, such as Assam, a high baro- meter is sometimes accompanied with a ground fog. In and around a cyclonic system the weather is more or less unsettled and cloudy, stormy or rainy, according to the season and the intensity of the depression. A partial ex- ception to this rule occurs indeed in the Indus valley region throughout the summer monsoon, in so far that, where the barometer is lowest, the atmosphere is hot and oppressive, but the weather remains fine, and rain is restricted, as a rule, to the outskirts of the depression, where the barometer is somewhat higher. Owing to other causes, a clear sky and a calm atmosphere are also sometimes met with in the very centre of a cyclone. The general prevalence of fine dry weather in the Indian cold season and of rainy and stormy weather during the summer monsoon, is primarily due to the fact that, in the WEATHER AND WEATHER REPORTS 199 former, Northern India is generally, for weeks together, a region of high barometer, and in the latter of low baro- meter; and it is in virtue of the comparative stability of these opposite conditions that the opposite characters of the two seasons are so marked. The anticyclone over Northern India in the cold season is, however, not quite stable. Now and then a barometric depression partly or entirely displaces it ; and it is on such occasions that we have the cold weather rains of Upper India. How these depressions originate is at present by no means clear. In some cases they seem to be formed in North-western India, or even farther south. In others they probably reach India from the west; though, in the absence of weather reports from any of the countries that lie in that direction (beyond Quetta) it is impossible at present to verify this. It is, however, certain that when these cold weather depressions travel from the place of their first appearance, they always move in some easterly direction, and thus it is that a spell of rain in the early months of the year begins, as a rule, in North-western India, and on subsequent days extends to the North-west Provinces, and sometimes, though more rarely, to Bengal. In the rainy season they are formed either in some part of the permanent depression then exist- ing in Northern India, or more frequently, on its outskirts, in the prolongation of its axis over the north of the Bay of Bengal ; and their subsequent course is between west and north-west. In the intervals between the monsoons they originate over the seas around India, and more often on the Bay of Bengal than on the Arabian Sea. The direction in which they travel is then more variable, but still most frequently to the north-west or west-north-west. It is to such travelling depressions as these, which, in their more intense form constitute cyclones (in the original sense of the term), and in their less intense form what I have termed cyclonic storms, that the more striking changes of weather are due. They are always accompanied with heavy rain ; on the mountains in the winter and spring months with snow ; and frequently with stormy winds. 200 CLIMATES AND WEATHER OF INDIA At sea they are more violent than on land, and their characteristic features on Indian seas will be the subject of a special section of this work. When such a storm has appeared in any part of the Indian weather report system, the general experience already gained of the storm tracks of the different seasons renders a certain amount of weather prevision possible ; but such prevision is better grounded when the storm has already begun to travel, and the direction to which it tends is shown by the reports of two or more successive days. How the weather is affected by the passage of these storms at different seasons will be described presently. But before doing so, it is necessary to notice another class / ^ / \ / Lahore x \ 5 t "X, 7* N S 1 Allahabad *^ / \ / 2 9-3 Calcutta s^ ? ~^ / ~-^ Nagpur x, \ 28-7 29-8 ^ 4 29-8 FIG. 10. BAROMKTRIC VARIATIONS IN MARCH 1887. of barometric oscillations, the cause and origin of which are even more obscure than are those of cyclonic storms. Barometric Surges. These oscillations have already been shortly noticed at page 22. If the readings of the barometer at 8 A.M., as now published in the Simla weather reports, be plotted on ruled paper for say a dozen stations in different parts of India, and if this be repeated day by day for a month, a connecting line being run through all the readings of the same station, as in Fig. 10, it is found that these lines show certain oscillations common to all or nearly all the stations, and apparently simultaneous, but generally greater in amount at the northern than at the more southern stations. These oscillations are evidently independent of those produced by the passage of cyclonic WEATHER AND WEATHER REPORTS 201 storms, which are much more local and not simultaneous at distant places. They are irregular in period, each oscil- lation lasting from 3 or 4 days to a fortnight. As was remarked in a former place, they do not necessarily accom- pany or betoken any important change of weather, but they are apparently not without influence on the weather, since there is more tendency to cloud and the formation of summer squalls in the hot season and of cyclonic storms in the rainy season, when the barometer is below the mean, than when it is above it. Mr. Eliot has remarked, with regard to them, that their coincidence over so large an area " points to the prob- ability of their being produced by the alternate expansion and contraction of the atmosphere." l In the absence, how- ever, of any fuller explanation of this view, it seems hardly reconcilable with the fact that they are as great or probably greater in higher than in lower latitudes, that is to say, where the alternate expansion and contraction of the atmosphere are of less magnitude than in India. They occur in the temperate zone as well as in the tropics, though perhaps their occurrence is there more masked by the greater barometric changes arising from other causes. Mr. Abercromby has recognised them in his valuable work on Weather, and has termed them "surges," the name here adopted, re- marking that they have a decided tendency to travel from west to east, or from south-west to north-east ; and further, that while the passage of a surge exercises a moderate in- fluence on the characteristic weather of any isobaric shape (cyclonic depressions, anticyclones, etc.), it exercises a very powerful influence on the formation of new systems. He states also that in the South Indian Ocean, during the period of the north-west monsoon (December to March), "hurricanes almost invariably form during the downward period of the surge, and practical forecasters, like Meldrum at the Mauritius, are always specially on the lookout for signs of serious bad weather, whenever there is the slightest symp- tom of a non-diurnal diminution of pressure." To a certain extent this is in accordance with Indian experience. 1 Report on the Meteorology of India in 1877, p. 48. 202 CLIMATES AND WEATHER OF INDIA In the Simla weather reports, the passage of these surges is shown by the column that gives the change of the barometer in the preceding 24 hours. If, as sometimes happens, the whole or the large majority of the stations in all parts of India show a rise or fall, this is due to the passage of a surge wave. As a rule, however, the surge is associated with changes of a more local character, and these may add to or detract from the general change, or even locally neutralise it. It is, of course, impossible so to analyse the total change as to assign to each element its exact value. Weather of the Cold Season. The most settled weather of the year in Northern India is from the cessa- tion of the rains in September or October to the latter part of December. It is but seldom that the serenity of the climate of the greater part of extra-tropical India is dis- turbed by storms, either from the Bay of Bengal or of more northerly origin, during these two or three months. The barometer is high in the north-west ; and light north- westerly winds or calms prevail down the Gangetic plain, becoming cooler as the year draws towards its close. Storms are, indeed, still formed in the south of the bay and often cross the peninsula, their tracks becoming more southerly as the season advances, until, by the end of December or a few days later, they cease to appear in Indian seas. The first cyclonic disturbances producing the winter rains of Northern India generally make their appearance in the latter part of December or the beginning of January, and they are repeated at intervals in this and the following months. One or two examples will show, better than any general description, the changes that precede and accompany this cold weather rain- fall. The first instance is one in which the disturbance apparently originated in India. On reference to Figure 3 on page 25, which shows the average distribution of pressure in the mont% of January, it may be observed that on the west coast of India the isobars all make a loop northward, indicating a trough of low pressure. In January 1883 this trough was more than usually developed, and on several occasions in the early part of the month extended northward, encroaching on the region of high pressure that, at this season, usually occupies North-western WEATHER AND WEATHER REPORTS 203 India. This state of things is shown in Fig. 11, which is the weather chart of 10 A.M. on the 4th January. The wind arrows show that, in conformity with this anomalous barometric feature, light southerly winds prevailed at several stations in the west of India, and easterly winds in Kathiawar and Rajputana, blowing cyclonically around the FIG. 11. WEATHER CHART FOR 10 A.M., 4th January 1883. loop of the isobars, and bringing damp oppressive weather to Western India. Up to the 9th, only a few light showers fell on the most nor- therly hills of the Punjab and at their foot. But on this day the de- FIG. 12. WEATHER CHART FOR 10 A.M., 9th January 1883. pression extended so as to occupy the whole of Central and Western India, as shown on the weather chart for 10 A.M. of the 9th (Fig. 12). Damp and warm southerly and easterly winds now prevailed over the whole of Central and Northern India, and on this and the following day rain fell over the greater part of the North-west Provinces, the Eastern Punjab, Central India, and Chutia Nagpur, up to the confines 204 CLIMATES AND WEATHER OF INDIA of Bengal, and snow on the North-west Himalaya. This rainfall tract is indicated by the shaded portion of the chart. When the rain ceased, on the 10th, a -band of high pressure was established from the Indus valley across Rajputana and Central India, and cool west and north-west winds prevailed for a day or two in the Punjab and North-west Provinces, as usually happens after falls of rain at this season. But this did not last long, and later on the trough again extended with a repetition of the rain. The second instance is of a different kind and more typical of the storms of the cold season, although in some respects remarkable and, in one, unprecedented. It occurred towards the end, viz. between the 22d and 28th of the same month, January 1883. FIG. 13. WEATHER CHART FOR 10 A.M., 25th January 1883. On the 22d and 23d the barometer fell rapidly over the Indus valley to a minimum in Upper Sind, Rajputana, and the Punjab, and on the 23d and 24th rain set in over the whole of these provinces. On the 25th a well-defined cyclonic depression appeared in the Punjab, as represented in Fig. 13, which is the weather chart for 10 A.M. of that day ; while in Bengal, Assam, and Burmah the pressure rose from 0*1 to 0*15 inch. A high barometric gradient was therefore established from east to west over Northern India, and steady southerly winds blew over the whole of India. Up to the forenoon of the 25th, the rain, which had ceased in Sind and Western Rajputana, extended east- wards over Guzerat, Eastern Rajputana, Central India, the whole of the Punjab, and the Gangetic plain as far as Cawnpore and the east of Oudh. Heavy snow fell on the Himalaya, and continued without in- termission till the 27th, and on the latter day extended to the high plain at the foot of the Himalaya above the Salt Range, so that at Rawalpindi, on the nigh b t of the 27th, the ground was covered to WEATHER AND WEATHER REPORTS 205 a depth of 4 inches. No such occurrence has been recorded either previously or since at so low a level. On the 26th, 27th, and up to the 28th, a barometric minimum continued to occupy the Punjab, the pressure rising, however, from a minimum of 29'65 to 30 inches, while the rain continued extending eastward, the weather clearing up in the west ; so that, on the afternoon of the 27th and forenoon of the 28th, rain was restricted to Bengal, Assam, the Himalaya, and the Upper Punjab. Probably, therefore, the main disturbance had moved east- wards over Tibet. On the 28th a spell of cold weather with north-west winds set in over Northern India, the temperature falling for several days, and the lowest temperatures of the year were registered in the first week of February : at Simla and Darjiling the lowest on record. In all prob- ability this cold was the effect of the enormous snowfall which, as mentioned above, descended to lower levels than has been recorded on FIG. 14. WEATHER CHART FOR 10 A.M., 26th February 1881. any other occasion. A similar but less intense spell of cold weather invariably follows a storm of this character, and the distribution of pressure and the winds which accompany it are so characteristic, that the weather chart on one such occasion is almost exactly reproduced on any other. Fig. 14 represents an occurrence of this kind on the 26th February 1881, and in their main features that of the day which followed the storm of January 1883 and that published in the Simla weather report of the 22d February 1888 are almost exactly similar. These instances will suffice to illustrate the chief vicissi- tudes of the weather of the cold season in Northern India. Warm close weather with light southerly winds precedes the storm. Then follows rain ; and if the disturbance is one of those that first appear in the Indus valley and thence 206 CLIMATES AND WEATHER OF INDIA travel eastward, the rain, which always falls to the east and north of the depression, also travels eastward. Unless the course of the storm is much more to the south than is usually the case, more or less heavy snow falls on the Himalaya ; and when the storm has passed away, the baro- meter rises rapidly in the north-west and cold weather FIG. 15. AVERAGE BAROMETRIC AND WIND CHART FOR MAY. follows, with north-west winds in the upper provinces and easterly in the peninsula ; the winds blowing clockwise around the anticyclone which is then formed over North- western India. If, however, the depression be formed over Western or Central India it is sometimes more persistent, and produces little or no snow on the Himalaya ; and in such cases it is often not followed by cold weather. Weather of the Hot Season. The prevalent weather WEATHER AND WEATHER REPORTS 207 of the hot season is determined by the heating of the land surface and the consequent reduction of atmospheric pressure in the interior of the country below that of the surrounding seas. The distribution of pressure thus resulting is illustrated in Fig. 1 5, which represents the normal barometric and wind chart for the month of May. The general arrangement of the isobars in this chart resembles that in the July chart (Fig. 4, p. 26), a broad trough of low pressure running obliquely across Northern India, with the isobars of higher value surrounding it concentrically on the east and south. But the greater intervals between these and the higher value of that encircling the region of minimum pressure indicate that the gradients are less steep, and therefore, as has been explained at page 25, the winds, shown in the Figure by arrows of varying length in proportion to their steadiness of direction, are less strong and steady. In the peninsula, too, there is a marked difference. In the May chart they form a southerly loop, conforming generally to the form of the land, which is not the case in July. It indicates that the pressure in the interior of the peninsula is lower than on either coast : the winds, in accordance with the cyclonic law, are therefore from south on the east coast and from between west and north-west to the west of the loop. These latter are dry winds, and accordingly, through- out Western India, from Sind as far south as Belgaum, it is very seldom that there is even a slight shower all through the spring months. Farther south, however, they have more the character of sea winds, and on the east coast, where they are from the south, they are decidedly sea winds. But at this season the winds are light over the sea, and the air drawn from the Bay of Bengal, though damper than that of Western India, is by no means so damp as it afterwards becomes in the stronger current of the summer monsoon. In the south and east of the peninsula, in Bengal, Assam, and Burmah, some rain falls in the spring months, but chiefly in the form of short-lived afternoon thunder-storms, and not continuously for a day or two, except on the comparatively infrequent occasions of 208 CLIMATES AND WEATHER OF INDIA cyclones being formed over the bay and travelling thence to the land. In North-western India, as in the west of the peninsula, the chief winds of the hot season are land winds from between west and north-west, but less steadily in May than in March and April. They become less steady as the barometer continues falling in the Punjab, Sind, and Western Kajputana, but they occur at intervals up to the setting in of the rains and sometimes even during the rainy season itself. Meanwhile the sea winds, which are pre- dominant in Bengal, gradually penetrate farther inland along the face of the Himalaya and the northern margin of the barometric depression of the Ganges valley. Accordingly, as westerly or easterly winds have for the time the mastery, so the hot season of Northern India alternates between dry heat, scorching but not intolerable, and the less intense but far more oppressive damp heat, which is only occasionally relieved for a few hours by a thunder -shower or simply a dust-storm. These little storms do not originate with any great local disturbance of the barometer. They occur when the baro- meter is low, and are more frequent on the east and north of the tract of lowest pressure, along the foot of the Hima- laya, in Bengal and especially in Assam, than elsewhere ; that is to say, where southerly and easterly winds pre- dominate. Everywhere they are most frequent in the neighbourhood of hills. The characteristic features of these storms, and the relation of the nor'-westers of Bengal to the dust-storms of North-western India, have already been de- scribed in the section on storms in the first part of this work. 1 Cyclones scarcely ever occur in March, and seldom in April, on Indian seas north of the Equator. In the former month there are but two recorded instances in the Bay of Bengal, and in April I find but nine in considerably more than a century. All these were formed in the south of the bay, and but two reached the shores of Bengal ; both in the last 1 Page 81. WEATHER AND WEATHER REPORTS 209 week of the month. The remainder either passed westward across the coast of Madras, or from the Andaman sea north- ward over Pegu. In May, storms of this class are more common. Including those of all degrees of severity, six have been recorded by Mr. Eliot in the 10 years, 1877-1886. Of these, two passed westward across the Madras coast \ one on a west by north course from Balasore Bay to Central India, the usual track of monsoon storms ; one northward across the Sunderbuns and over Jessore ; one, originating to- the west of the Andamans, reached the Arakan coast north of Akyab ; and one from the south-west of the bay travelled north-westward to the mouths of the Kistna and Godavery, and then turning to north-east, skirted the Eastern Ghats of the Northern Circars, and finally reached Behar. Of the May storms sufficiently severe to have attracted notice in years anterior to 1875, and which have passed over the land of some part of India and its dependencies, the following are extracted from the list published in the 1877 volume of the Journal of the Asiatic Society of Bengal : 1787. Coringa. Very destructive storm wave. 1811. Madras. Great loss of ship- ping. 1820. Madras, across peninsula. 1823. Balasore. 1827. Madras. 1830. Felt slightly at Calcutta. 1832. Gangetic delta. Destructive storm-wave. 1833. Mouth of Hooghly. Storm wave. 1834. Kyaukpyu. Hence it appears that, in May, cyclones may be experi- enced on any part of the coasts of the Bay of Bengal, but that the coast of Madras is especially subject to them at this season. See also Fig. 26 (p. 231,) and the list of May cyclones in Appendix II. On the west coast of India, cyclones are altogether less common; but according to a list drawn up by Mr. F. Chambers, p 1840. Orissa coast. 1841. Madras. 1843. Ongole. 1844. Noakhally and Chittagong. 1849. Chittagong. 1851. Madras. 1852. Sunderbuns, east of Calcutta.. 1858. Across peninsula. Also Chittagong and Dacca. 1869. Across Bengal. 1872. Madras. 1874. Madras. 210 CLIMATES AND WEATHER OF INDIA they would seem to be relatively more so in the spring months than at the close of the monsoon. He enumerates two in March, nine in April, and thirteen in May, out of a total of 74 storms. But exception being made of those that have crossed the peninsula from the Bay of Bengal, and which became greatly weakened in the course of their transit, they have been felt only on the coast. I do not find any instance of a storm travelling inland from the west coast of India at any time of year. Weather of the Summer Monsoon. The transition from the hot season to the rains is gradual only in Assam, and to a less extent in Bengal and Arakan. In Western, North-western, and Central India, where land winds pre- vail, more or less, all through the hot season, the change is rapid ; a few days only of light damp winds and calms being the forerunner of the monsoon. In some seasons, a day or two of rainy weather occurs in these provinces about a fort- night before the monsoon sets in permanently, and is called the " chhoti barsat " or " little rains." It is, however, by no means a regular phenomenon. When it occurs, it is generally the result of an early cyclonic storm, similar to those which are frequent during the monsoon, and is due to an early and short-lived invasion of the monsoon. It is followed by a re-establishment of the land winds ; a sequence that sometimes happens also in the middle of the rainy season, but this latter is then called " a break in the rains." In Bengal, the average date for the setting in of the rains is the second week in June. The barometer falls steadily for three or four days beforehand, till it approaches its annual minimum ; and the rainfall often sets in with a small cyclonic storm, giving squally weather at the head of the bay. This sometimes carries the rain at once up to Behar, and even farther to the north-west, but as a rule it takes from a week to a fortnight to extend to the North- west Provinces and the Eastern Punjab. On the Bombay coast, the rains usually set in in the first week of June, a few days earlier than in Bengal, and quickly extend to the Central Provinces, and afterwards WEATHER AND WEATHER REPORTS 211 to Eajputana and Central India. On the Malabar coast they begin some days earlier than in Bombay, but do not extend very copiously beyond the Western Ghats ; and in Mysore, the ceded districts of Madras, the Deccan and Hyderabad, more rain falls when the strength of the mon- soon to Northern India relaxes, than when the interior plateau of the peninsula is swept by a strong current from the west coast. In Pegu and Arakan the rainy season sets in a week or so earlier than in Bengal, on an average, but not invariably. The change, that takes place in the general distribution of the atmospheric pressure and winds on the setting in of the monsoon, will be best understood by comparing the May chart on a preceding page (Fig. 15) with that for July (Fig. 4, p. 26). From May to June or July, the barometer falls O'l or 0'15 inch lower in Northern India, and rises 0'05 inch on the Travancore coast ; and the southern loop of the isobars, in the peninsula in May, disappears in July, these lines running obliquely across the peninsula from coast to coast. The western branch of the monsoon blows steadily and strongly across the peninsula as far north as the Satpuras, as is in- dicated by the lengths of the wind arrows on Fig. 4 ; less strongly over the Central Indian plateau up to the Ganges valley. In the latter and the Punjab, east winds predominate; but the whole region enclosed by the isobar 29*5" is one of barometric depression and generally light winds. Over the eastern half of this depression, and in Bengal and Assam, much rain falls without much local barometric disturbance, especially during the low stage of a surge ; but heavier falls occur at intervals along definite belts of country, which are traversed by cyclonic storms, formed either over the head of the bay, or locally in some part of the permanent depression. The tracks of these storms are, with rare exceptions, between west and north-west, and they succeed one another at longer or shorter intervals all through the monsoon ; reaching in some cases only as far as Central India or the North-west Provinces, but also occasionally to the Punjab, Sind, or Kathiawar ; whence, in some rare instances, they pass out 212 CLIMATES AND WEATHER OF INDIA to the north of the Arabian Sea, with increasing intensity, and give rise to stormy weather on the Gwadar coast. These cyclonic storms are so important a feature of the rainy season, and influence so greatly the distribution of the FIG. 16. WEATHER CHART FOR 10 A.M., Sth September 1882. rainfall, that a more detailed notice of one or two instances will much facilitate a due comprehension of the character- istic weather of this season. Figs. 16 to 19 represent four stages of a storm that travelled from the Bay of Bengal to the Punjab and Sind at the close of the rainy FIG. 17. WEATHER CHART FOR 10 A.M., 10th September 1882. season of 1882. It was formed over the head of the bay, off the Orissa coast, on the 6th and 7th September, reached the Punjab on the 14th, and finally disappeared in Upper Sind on the 18th, having thus lasted at least 12 days. It brought heavy rain to Orissa on the 7th and Sth, while in Eastern and Northern Bengal but little fell ; as is usually WEATHER AND WEATHER REPORTS 213 the case when the centre of the storm lies so far to the south. But in the Central Provinces there was heavy general rain from the 8th to the llth ; in Central India on the llth and 12th ; in Northern Raj- putana up to the 13th ; and in the Punjab between the 10th and 15th. On the other hand, the North-west Provinces, Oudh, and Behar, which lay at a distance to the north of the track, had little or no FIG. 18. WEATHER CHART FOR 10 A.M., llth September 1882. rain ; and in Bombay there was a break of about a week from the 13th to the 20th, and the same in Hyderabad. The rainfall of this period or the greater part of it was therefore concentrated in the storm. The barometric depression in this storm, at least in its earlier FIG. 19. WEATHER CHART FOR 10 A.M., 14th September 1882. stages, was somewhat greater than usual, and its duration was more prolonged. Figs. 20 to 23 represent four stages of a storm that followed an almost due westward track, at the beginning of the monsoon of the following year, 1883. A full description of it, illustrated by charts of the weather for each day, was published by Mr. Eliot in the 53d volume of the Journal of the Asiatic Society of Bengal It originated 214 CLIMATES AND WEATHER OF INDIA on the 27th June, between 20 and 21 north lat, and 89 and 90 east long., and during the next two days moved very slowly (at between 2 and 5 miles an hour) on a north-west course to Balasore Bay, crossing the coast on the evening of the 29th. While still over the bay, the south-west and south winds of the south and east quadrants had the force of a strong gale (10 to 11 on Beaufort's scale) ; while to the Fio. 20. WEATHER CHART FOR 10 A.M., 29th June 1883. north and west, the north-easterly and westerly winds were only moderate to strong (3 to 5) ; the barometric depression at the centre amounted to somewhat less than half an inch. On reaching the land its course became almost due west, and this FIG. 21. WEATHER CHART FOR 10 A.M., 30th June 1883. direction was maintained across the whole width of the peninsula, at a rate of about 15 miles an hour during the greater part of the dis- tance, but reduced to not more than 8 miles an hour while traversing the hilly country that intervenes between Orissa and Sambalpur. Its intensity also, as indicated by the depression of the barometer at its centre, was reduced by nearly two-thirds in this part of its course ; but it recruited its strength, on being fed by the monsoon current WEATHER AND WEATHER REPORTS 215 from the Bombay coast, and after passing out to the Arabian Sea, it regained its original cyclonic force. The centre passed north of Cut- tack and Sambalpur, close to Seoni, Indore and Bhuj ; after which, on the morning of the 3d July, it left the land, travelling out to the south of Kurrachee. During the transit, the winds were strongest and the rainfall heaviest to the south of the central track. To the north FIG. 22. WEATHER CHART FOR 10 A.M., 2d July 1883. of it, in equal proximity to the centre, the rainfall was not more than one- fourth as heavy. The strength of the wind in different parts of the storm was very variable ; where strongest it did not exceed 43 miles an hour on the average of 24 hours, and at very few places amounted FIG. 23. WEATHER CHART FOR 10 A.M., 3d July 1883. to 25 miles. At one station it appears to have been as high as 64 miles an hour on the average of six hours, but this is the only instance of an inland station where it had the force of a strong gale until the storm again approached the sea. There do not appear to have been, in this storm, any of those exceptionally high rainfalls of 20 or 30 inches in the 24 hours that have been registered on certain other similar occasions. The highest 216 CLIMATES AND WEATHER OF INDIA noticed by Mr. Eliot are 10 '05 inches on the 3d July at Rajkote, and 9*44 inches on the 29th June at Pooree. But there was an average of 6' 15 inches over the whole Pooree subdivision of the Cuttack dis- trict, 4'5 inches over that of Cuttack. and 2'66 inches over that of Balasore on the latter of these days ; and of 10'56 inches, 9 '6 inches, and 7 '6 8 inches respectively on the same subdivisions in the three days 28th to 30th June. In the Central Provinces, an average of 6-13 inches fell on the Nagpur subdivision, 5 '7 inches on that of Bhandara, 5 -19 on that of Chhindwara, and 6'26 on that of Nimar in the space of two days. Falls of equal and even greater amount are not in- frequent on the Satpura range in cyclonic storms at this season, and at intervals of 2 or 3 years they produce heavy floods in the Tapti and Nerbudda rivers, sometimes causing great destruction. Arrangements have been made by the Meteorological Department to telegraph such occurrences to the districts on the lower course of these rivers, and thus to prepare them in some measure to meet the threatened catastrophe. A storm, which followed a track almost identical with the above, occurred in September 18*78. All these were comparatively long-lived storms ; the majority do not travel beyond Central India or the North-west Provinces ; and, after giving those provinces 1 or 2 days of heavy rain, they disappear. At the beginning and end of the monsoon, and also during seasons of drought in Upper India, when there is a prolonged suspension of the rains, and westerly land winds hold sway in Eajputana, Central India, and the western half of the Gangetic plain, storms formed over the head of the bay sometimes take a more northerly course, across Bengal. Such was the case with four storms in succession in August 1877, the year of the last great dearth in the North-west Provinces; and one in August 1880, a month which was also one of prolonged drought and, both in the anomalous distribution of pressure and the unseasonable prevalence of land winds in Upper India, closely resembled the same month of 1877. In May and the first half of June, in September and October, the usual course taken by cyclones from the head of the bay is northerly. WEATHER AND WEATHER REPORTS 217 Temporary interruptions of the rains, during which the monsoon falls weak, and ceases to penetrate Upper India, occur in all years, but, as a rule, do not last many days. But in the years just mentioned, and in 1883, they lasted several weeks, and west and north-west winds, similar to the characteristic winds of the hot season, blew all over the Upper Provinces, in Eajputana, and Central India and, even, for a time, into Berar and the Deccan. On all such occasions a very characteristic feature appears in the isobars of the weather chart, which is often referred to in the reports as " a shoulder of high pressure " in Western India. It consists in the greater northern protrusion of the north- ward bend of the isobars in Eajputana and Central India, and the somewhat greater obliquity of their slope from west to east. It is not always so marked as to be readily seized by an unaccustomed eye, unless by comparison with a nor- mal chart, or with one of a day of well-distributed and normal rainfall ; but though small, it is always significant. It means that the barometric pressure in Western India is higher than usual, relatively to that in the east of the country, in Bengal and Chutia Nagpur, and that the winds of this part of India are more northerly than is usual at the time of year. Now, it is this slight northing of the westerly winds that distinguishes the dry current coming from Baluchistan and Afghanistan from the rain-bearing monsoon of the Arabian Sea, and therefore indicates the replacement of the latter by the former. Generally this is associated with another feature of equally unfavourable import, viz. the northward displacement of the axis of the low pressure trough of Northern India to near the foot of th6 hills. This means that the easterly winds from the bay of Bengal, which prevail only to the north of that axis, fail to penetrate as usual up the Ganges valley, and, therefore, that the eastern branch of the monsoon is withheld from the Upper Provinces. The accompanying chart of the 9th August 1882 (Fig. 24) is a good illustration of these ab- normal features. On this occasion there was a prolonged break in the rains from the 5th to the 20th of the month, 218 CLIMATES AND WEATHER OF INDIA during the whole of which time the charts exhibited these peculiarities of the isobars. It was observed on a previous page that the serenity of the weather of a great part of Northern India is rarely dis- turbed during the 2 or 3 months that follow on the cessa- tion of the rains in September or October. This is strictly true of Upper India, but somewhat less so of Bengal. In this province and Assam, the rains last well into October, and both in this and the early part of the following month severe cyclones from the bay sometimes reach the province. Indeed some of the most destructive storms on record have occurred in October and November, although several years FIG. 24. WEATHER CHART FOR 10 A.M., 9th August 1882. may pass without such a visitation. But October is often a stormy month in the bay, and even though the hurricane core of the storm may not reach its northern shores, no great disturbance occurs in any part of the bay that does not make itself felt in Lower Bengal by close damp weather and cloudy skies. The signs and prognostics of these storms will, however, be more fitly noticed in the next section of this work. When the monsoon has ceased to blow to Upper India, that branch which comes from the bay is still directed for a few weeks towards the north-east of the peninsula ; but feebly and with frequent intermission. Mr. Eliot records nine cyclonic storms in the months of September and October, and two in November, in the 10 years 1877-1886, that WEATHER AND WEATHER REPORTS 219 crossed the coast between the Chilka lake and the mouths of the Godaveiy. Of these, one only traversed the whole width of the peninsula. The remainder either broke up against the Eastern Ghats or shortly beyond, or were diverted northward towards the Central Provinces or Chutia Nagpur. Hyderabad and the Deccan districts of Bombay enjoy, therefore, greater immunity from storms of this character than either Northern India or the south of the peninsula. In the Northern Circars, at the foot of the Eastern Ghats, the September or October rainfall is the heaviest of the year; but in Bustar, Hyderabad, and the Deccan, the rainfall of this season is small, and so soon as the westerly branch of the monsoon ceases to predominate, it is replaced by light north-easterly winds, with little or no rain. About the middle of October or somewhat later, the heavy rainfall of the year sets in on the Madras coast. As in Northern India in its rainy season, so also here, much of this rain is attended with but little local barometric disturb- ance. But little cyclones and cyclonic storms are frequently formed over the south-west of the bay, and travel to the Madras coast, and though small, are sometimes locally de- structive. The more severe cyclones of this month and November usually originate at a greater distance from land, either half-way across the bay, or still farther east, in the neighbourhood of the Andamans and Nicobars ; and al- though the majority of them take a course about west by north towards the Madras coast, they occasionally travel northward up the bay, and reach the shores of Bengal. The disastrous Calcutta cyclone of the 5th October 1864, that of the 1st November 1867, and the Backergunj cyclone of the 1st November 1876, are well-remembered instances of such storms. The process of formation, and the subsequent movements of those that originate in the south-west of the bay, can generally be watched and traced from day to day by the indications of the barometers and the wind changes at Trincomalee, Negapatam, and Madras. These stations being at no great distance from the storm 220 CLIMATES AND WEATHER OF INDIA cradle, the barometer falls steadily during their incubation ; and the wind directions, taken in conjunction with the relative rates of fall of the barometers at the different stations, afford a fair indication of the storm's position. Warnings are now sent to the coast ports from the Calcutta office, the meteorological reporter for Bengal having special Walker & Boutall se- FIG. 25. SIMLA. WEATHER CHART OF THE 26TH JULY 1888, 8 A.M. charge of the storm warning system for all the coasts of the bay. Weather Reports. The Indian daily weather reports having been frequently referred to in the foregoing pages, and many of the daily charts given in illustration of certain characteristic states of weather, the signification and utility of these reports and charts will be fairly well understood by any one who has studied the preceding descriptions of WEATHER AND WEATHER REPORTS 221 the weather, and the explanation of its constituent elements as set forth in the first part of the work. It may, there- fore, suffice in this place to recapitulate in due order the tabular statistics of one of these reports, with references to those pages of this work in which each of them is ex- plained ; and, in conclusion, to give one more example of a chart, casually selected from the recent issues, drawing attention to the chief points worthy of notice. I must suppose that the reader has one of these reports before him, the sheet being too large for reproduction on these pages. The chart which illustrates it is given on a reduced scale in Fig. 25. The report for the 26th July 1888 gives telegraphic reports from 96 stations, despatched shortly after 8 A.M. Those of three other stations which are deficient, together with all other deficiencies in the reports of the month, are published in a supplementary issue at the end of the month. Col. i shows the readings of the barometer at 8 A.M., corrected to the Calcutta standard (p. 21) and to the temperature of freezing (p. 21), and except in the case of hill stations, to its equivalent value at the sea-level (p. 23). These readings are the data from which the isobars, showing the distribution of atmospheric pressure at 8 A.M. of the 26th July, are laid down on the chart, Fig. 25 (see p. 24). Col. 2 gives the changes of the barometer since the same hour on the preceding day, when, therefore, as regards the regular diurnal varia- tion, it was in the same phase. In this report, nearly all the readings show a rise, greatest at the most northerly stations, Sialkot, Kawal- pindi, and Peshawar, where it exceeds 0*2 inch ; and least on the west coast of the Konkan, where it amounts to only 0*02 or 0*03 inch. This is an instance of a surge (pp. 22, 200). Its effect has been to diminish the barometric gradient (p. 24) from the west coast to the Punjab and Sind, which determines the strength of the western branch of the monsoon towards Upper India ; since the barometric difference is reduced by nearly 0'2 inch. But the tract around Allahabad has not shared in the general rise of pressure ; in this direction, therefore, the gradient is slightly increased, and therewith the strength of the winds towards it. Cols. 3 and 4 give the wind direction at 8 A.M., and its mean rate of movement per hour during the preceding 24 hours. These are in- dicated on the chart by arrows, and the number of the feathers (from 1 to 7) show the strength or velocity of the wind according to a scale printed below the chart. A calm is represented by a cross. Cols. 5 to 9 give the temperature of the air at 8 A.M., the highest and lowest temperatures of the previous 24 hours, as recorded by self- 222 CLIMATES AND WEATHER OF INDIA registering thermometers ; the mean of the two, as representing ap- proximately the mean temperature of the 24 hours ; and the difference of this mean temperature and that given in the report of the previous day. These changes are not very great in the rainy season, and they depend very much on rain, the air being cooled by rain, and becom- ing warmer again on its cessation. Cols. 10 and 11 show the relative humidity of the air (p. 46 ) at 8 A.M., and its increase or decrease since the same hour of the preced- ing day. As the air becomes drier, so does the prospect of rain diminish, and vice versa. As a general rule, the humidity increases when the temperature falls and decreases when the latter rises. Col. 12 gives the estimated cloud proportion (p. 61) at 8 A.M., on the scale to 1 0. indicates a cloudless sky, and 1 one entirely overcast. Cols. 13 to 16 are the rainfall data viz. 1st, the rain measured at 8 A.M., that had collected in the gauge during the previous 24 hours ; 2d, the total quantity registered at the station since the 1st June, (i.e. approximately that of the summer monsoon up to date) ; 3d, the average of the same period deduced from the registers of past years ; and 4th, the excess or deficiency of the actual fall as compared with this average. The rainfall of the 24 hours at any station is shown on the chart by a small circle over the position of the station, with an inscribed figure if the fall has exceeded half an inch. Col. 17 contains a single word or short phrase descriptive of the general character of the weather during the previous 24 hours. The chart gives a general view of the state of the weather as regards the three important elements barometric pressure, winds, and rainfall. Temperature, humidity, and cloud proportion may also be represented on a chart, but they are omitted to avoid overcrowding and confusion, and any features of importance are noticed in the remarks on the first page of the report. If we compare the chart (Fig. 25) with the normal chart for July (Fig. 4, p. 26) an important difference appears in Northern India. The trough of low pressure which, in the normal chart, extends obliquely across Northern India, occupying nearly the whole of the Ganges basin as far east as Behar and the confines of Lower Bengal, is divided, in Fig. 25, by a shoulder of higher pressure occupying the western half of the North-west Provinces and Oudh ; and its eastern half is extended so as to include nearly the whole of Behar and Bengal. Within this are two small cyclonic systems, indicated by the wind arrows, the one with its centre between Allahabad and Benares, tolerably definite, and further indicated by a small oval isobar ; the other less distinct, but apparently centring between Berhampore and Purneah, and indicated only by the wind arrows. In both, the barometric depression is very slight, amounting to only a few hundredths of an inch, as is frequently the case in these local depressions in the rains ; but their influence is shown, not only by the winds, but STORMS OF INDIAN SEAS 223 also by the concentration of the rainfall of Northern India around them. The more easterly of the two first appeared on the chart of the 23d, two days previously, and was short-lived, since it had entirely disappeared from that of the 27th, having probably amalgamated with the western depression. This latter was much more durable. It originated about the 18th, off the coast of the Sunderbuns, and in the course of eight days had travelled very slowly to the position, between Allahabad and Benares, shown in Fig. 25. It lasted some ten days longer, moving slowly westward to between Agra and Jhansi, still concentrating around it the greater part of the rainfall of Northern India. Simultaneously with its advance from the coast, was formed the shoulder of high pressure noticed above, viz. on the 23d ; and, on this day, a break of the rains set in in the Punjab and lasted until the 28th, when the former began to yield to the encroachment of the advancing storm. During this time, north-west winds prevailed at Delhi, Agra, and Jeypore, and a strong south-west wind blew over Lower Sind and Guzerat, as shown by the six feathered arrows, which indicate a mean rate of over 26 miles an hour. The greater part of this current appears to have been diverted eastward across Rajputana, to feed the rainfall around the depression. On the 28th, another very slight depression, similar to that of the 18th, appeared over Bengal, and like the former, moved westward and amalgamated with its fore- runner on the 1st August, with the effect of recruiting its strength. It finally disappeared between the 3d and 4th. During the fortnight or 1 6 days that these changes were in progress, the strength of the western branch of the monsoon varied considerably, accordingly as the depression in the North-west Provinces became enfeebled or recruited its strength. The Bengal branch was weak throughout after the 22d, and was chiefly directed towards Arakan. THE STORMS OF INDIAN SEAS Piddington's Work. At a time when the science of storms was in its infancy, and long before, even in the western homes of science, there existed any official organisation for collecting information on the meteorology of the ocean, the spontaneous and patient labours of Henry Piddington laid the foundation of our knowledge of the storms of Indian seas. For this work, in so far as regards the collection of marine observations and the inter- pretation of their teachings in language that appealed to the minds of seamen, his own early seafaring experience gave 224 CLIMATES AND WEATHER OF INDIA him an advantage which he turned to the best account ; and his name is deservedly held in honour as that of a pioneer of science, and a benefactor whose claims to the gratitude of at least two generations of sailors in Indian seas stand unquestioned and universally recognised. The practical rules laid down by him for the guidance of seamen have undoubtedly been the means of saving many a ship from destruction, and if these rules are in some respects defective, and his conception of the nature of cyclonic storms erroneous, it must be borne in mind that, in his day, the whole science of meteorology was so little advanced, that even had he lived less remote from the centres of scientific activity, it would not have been possible for him to bring the results of his work into harmony with the physical laws of the atmosphere as we now know them. That the winds blow in circles around a storm centre, as taught by Reid and Piddington, is, as we now know, only a rough approxima- tion to the actual facts, even in high latitudes, and in low latitudes is considerably wide of the truth; and practical rules based on that assumption may, therefore, seriously mislead in tropical seas ; still, the work done by Piddington in ascertaining the directions in which storms usually travel, and in describing their premonitory indica- tions, is of very great value, even admitting, as we must do, that his preconceived idea that the winds blow in circles sometimes led him astray in his conclusions. 1 * In the charts that illustrate Piddington's later Memoirs, the observed wind directions are not shown ; but as the logs that furnish the data are given in full in the text, it is easy for any one to lay them down for himself, and he will then find, not only that they afford ample evidence that in these storms, as in all those since observed, the wind really blew in spirals, but that, this being admitted, many of the difficulties which Mr. Piddington met with, and which obliged him sometimes to suppose that two or more storms existed simultaneously in near proximity to each other, disappear. His fourteenth and fifteenth Memoirs give good instances of this. And in his first Memoir, in which he deduces a west-south-west track for the storm of the 3d to 5th June, 1839 (a track of which I find no other example in all the records before me), it is obvious that, when once it is admitted that the winds blow in spirals, the course of the storm was in a northerly direction, indeed almost due north, not an uncommon course at that time of the year. It must be evident to any one who goes carefully through Piddington's STORMS OF INDIAN SEAS 225 The general characters of cyclones have already been described briefly in the first part of this work. In the present section, which is written for the practical guidance of seamen more especially, it will be necessary to notice them more in detail, with reference to their local characters and their relative frequency and behaviour at different seasons ; and in the first place, I shall deal with the storms of the Bay of Bengal. 1. THE BAY OF BENGAL Storm Seasons in the Bay of Bengal. The following table of cyclones of the Bay of Bengal was drawn up from the records available up to the end of 18*76. At that time little was known of the less severe cyclonic storms of the summer monsoon, and, with a few exceptions, they may be considered as omitted from the list, having failed to attract prominent notice. As will presently be seen, when they are included, the summer monsoon, instead of being compara- tively free from storms, is found to be the stormiest time of year ; but except that the bay is swept by the strong south- west gale that feeds them, they are restricted to the north of the bay and to the coasts of Orissa and Bengal. NUMBER OF RECORDED CYCLONES ON THE BAY OF BENGAL UP TO 1876 January . . .2 February . . . March . . .2 April . . .9 May '. . .21 June . . .10 July ... 3 August . 4 September . . 6 October . . .31 November . .18 December. . . 9 Total, 115 storms. February is the only month in the year in which no storm Memoirs on the Law of Storms that he never sought to evade or ignore the difficulties and apparent inconsistencies that presented themselves in his evidence. But his method was faulty, and of this he was unaware. Having assumed the truth of the view that the winds blow in circles, he determined the position of the storm or storms on that assumption, often on quite insuf- ficient evidence ; and if one storm centre would not fit in with the wind observations, he assumed two, and regarded the result as proving the truth of the original assumption. He seems never to have had the alternative of spiral winds present in his mind, and although he admits the occasional in- curvature of the winds, he regards it as exceptional and irregular. Q 226 CLIMATES AND WEATHER OF INDIA has been recorded in any part of the bay. The two January storms occurred in the first week of the month, in the ex- treme south of the bay, one passing over Pondicherry, the other over Trincomalee and the southernmost districts of the peninsula. Both of the March storms occurred in the last week of the month on the Madras coast. It may therefore be said that, during the two and a half months from the 7th January to the 26th March, storms were, so far, unknown in the bay ; and as appears from Mr. Meldrum's investiga- tions, this is precisely the season when they are most fre- quent in the South Indian Ocean. All the April storms of the above list, with two excep- tions, both in the last week of the month, all the December storms and those of the last half of November, were entirely restricted to the southern half of the bay. But later ex- perience has shown that the north of the bay does not enjoy complete immunity from stormy weather up to the early days of December; since, in 1883, an ill -defined cyclonic disturbance of no great severity, and producing but a slight depression of the barometer, was formed at the end of November in the south-east of the bay, and traversed its entire length to the Eastern Sunderbuns and Cachar. And again, in 1885, a storm formed off the coast of Ceylon on the 14th and 15th November, after moving towards the Madras coast, turned away to the north-east, and reached the Arakan coast, near Akyab, on the 23d. The cyclonic storms of the winter and early spring, that first appear in North-western India and travel eastward (p. 87), very rarely reach to the Bay of Bengal. There has been, however, one recent instance of such a storm, the only instance of the kind on record. At the end of January 1888, a storm crossed India, on an east-south-east course, from Baluchistan to Gopalpur, and thence passed across the bay to Akyab. According to our present experience, then, with the single exception just noticed, the northern half of the bay is always quite free from storms from the end of the first week in December to nearly the end of April, an STORMS OF INDIAN SEAS 227 interval of about four and a half months, and they are very rare after the middle of November. In March and April strong south-west winds, sometimes amounting to a moderate gale, blow at the Sand-heads and in the north-east of the bay, but they are not cyclonic at sea. In May and the first half of June, and in October and November, cyclones may occur in any part of the bay (see Figs. 26 to 28); but in the first half of May the large majority are restricted to the southern half; in the second half of the month, though formed in the middle or south of the bay, they more frequently travel to the coasts of Orissa and Bengal. In the first half of October, they are about twice as frequent on the coasts of Bengal and Orissa as on those of Madras ; in the second half these relative propor- tions are reversed; and in November storms are comparatively rare in the north of the bay. In June, one cyclone has been recorded at Madras and one at Akyab. All others have affected Bengal and Orissa and their coasts. And the cyclonic storms of the summer monsoon have all been formed either in the north of the bay (generally between False Point and Chittagong) or over the adjacent land. Hence, for two months at the beginning of the storm season, viz. in March, April, and the first half of May, and again for two and a half months at its close, viz. from the middle of October to the beginning of January, the southern half of the bay is chiefly or exclusively the stormy region. The intervening months, including the whole of the summer monsoon, are the stormy season of the north of the bay. The relative frequency of storms in these latter months will be better shown by the following table, drawn up from the lists prepared by Mr. J. Eliot, of the storms recorded on the Indian weather charts from 1877 to 1886. It includes all the severe cyclones and the milder or smaller cyclonic storms of those years, excepting, perhaps, some whirls of small magnitude that have been promptly dissipated, and those formed over the land which have not been felt as cyclonic in any part of the bay, though, for the time being, they have increased the strength of the monsoon. May . . ~ . 6 September . . .11 October July . ". August . 18 .17 November December 228 CLIMATES AND WEATHER OF INDIA CYCLONES AND CYCLONIC STORMS ON THE BAY OF BENGAL IN THE TEN YEARS 1877 TO 1886, FROM MAY TO DECEMBER INCLUSIVE . 17 10 . . .16 . - . 4 Total, 99 storms. Place of Formation, Tracks, and Rate of Travelling. So much with regard to the seasons at which the seaman may be prepared to encounter stormy weather in the north and south of the bay respectively. I shall now notice, in more detail, the parts of the bay in which storms originate, the directions in which they move, and their rate of pro- gression. We shall see that, while courses between west- north-west and north-west are much more common than others, exceptions sometimes occur in which they deviate very considerably from these most usual directions ; but even the exceptions do not surpass certain limits, and with the single exception of the January 1888 storm, noticed on page 226, all storms yet recorded have travelled towards some point of the compass comprised between west and north-east. The following results are mainly deduced from the work of Mr. John Eliot, who has made a special study of the storms of the Bay of Bengal, under far more advantageous circumstances than were ever at Mr. Piddington's command. The data are more extensive than any that have yet been brought to bear on the questions at issue. Not to mention his elaborate and detailed Memoirs, and those of Messrs. Willson, Pedler, and others, describing the life history of several of the more striking and important cyclones of late years, Mr. Eliot has given lists of all the recorded storms, of all degrees of severity, that have occurred during the last ten years, together with charts of their tracks. All these, in addition to Mr. Piddington's Memoirs, have been consulted in drawing up the following summary. Lists of all the storms referred to and their principal data, summarised in the following tables, are given in Appendix II. at the end of this STORMS OF INDIAN SEAS 229 work, and Figs. 26, 27, and 28 are charts of the tracks in the three months May, October, and November. In order to exhibit in the most convenient form those results that are of most importance to seamen, I have sum- marised them in four tables, showing for each month from May to December the following data : 1. The number of storms that have originated in each square of 4 of latitude by 4 of longitude. 2. The number that have been experienced in each square of 4 X 4. This includes, in addition to the cases of the first table, all those in which the vortex of a storm has traversed any portion of a square. 3. The directions of the storm tracks. In this table storms that, in the course of their passage across the bay, have changed their direction through two or more points, appear in more than one column. 4. The rates at which storms have travelled across the bay. Here again the same storm sometimes appears under different headings. Probably in most, if not in all cases, a storm moves forward from its place of origin at first slowly, and advances more rapidly as it approaches the land ; but it is only in a small number of instances that the available data show such an increase as to double or treble the initial rate thus determined. 2 3 CLIMATES AND WEATHER OF INDIA TABULAR SUMMARY OF RECORDED STORMS ON THE BAY OF BENGAL, ACCORD- ING TO THEIR PLACE OF ORIGIN, AND THE PARTS OF THE BAY TRAVERSED BY THE CENTRAL VORTEX. STORMS ORIGINATED STORMS EXPERIENCED BE- j BETWEEN LATITUDES TWEEN LATITUDES Months. || CO 5 CO *$ o jj B 3 3 3 s _0 f i o 1 3 o CO 3 o i tf p CO 1 3 o o s -'" 17 15 20 8 21 34 101 78 66 12 1 12 5 3 2 1 6 20 38 28 20 5 4 1 1 1 5 17 31 44 24 6 3*7 ins. 0-5 0-7 07 1-2 9-0 5-9 7-0 5-3 1-8 0-2 2-0 2 14 30 21 8 3 1 7 12 6 1 5 2 i 2 1 1 1 1 3 ... 13 years . 385 129 135 83 26 9 4 1 9-0 ins. RAINFALL, EVAPORATION, AND WIND 261 LAHORE (12 years) INCHES. Below \. JtoJ. itol. 1 to 2. 2 to 3. 3 to 5. 5 to n. Highest. January 24 4 7 1 1-5 ins. February . 21 7 4 3 i ... 2-2 March 21 6 5 1 1-1 April . 28 7 4 1 ... 1-2 May . 18 12 2 2 3-0 June . 23 7 8 6 1 3-5 July . 22 20 19 15 5 4 2 7-2 August 35 8 14 10 5 2 1 51 September . 18 6 6 4 ... 1 ... 57 October 9 1 2 1 2-0 November . 6 1 2 ... 0-8 December . 7 4 4 ... 0-8 12 years . 232 83 77 41 14 8 3 7'2 ins. NAGPUR INCHES. Below i. Jtoi. itol. Ito2. 2 to 3. 3 to 5. 5 to 7i. January 11 3 5 1 1-0 ins. February . 9 5 1 1 ... 1-5 March 17 2 4 1 1-8 April . 22 2 3 1 ... 1-1 May . 34 9 4 1 i ...~-~ 3-9 June . 98 44 24 27 7 5 1 7-2 July . 122 38 54 37 18 6 5 6-2 August 120 35 38 33 10 2 1 6"2 September . 94 34 28 22 9 2 2 7-2 October 33 9 10 12 1 2 4-9 November . 9 2 1 2 I 27 December . 10 ... 6 2 ... ... 1-8 13 years . 579 183 178 140 46 18 9 7'2 ins. 262 CLIMATES AND WEATHER OF INDIA BOMBAY INCHEF L r+f o 1 Hn 3 r+* 3 r* (N a 3 11-5 4th September 1879. Chandbally 20-3 16th September 1879. Cox's Bazar 157 9th July 1886. Furreedpore 14-3 5th July 1877. Chittagong 14-0 3d June 1886. 10-3 9th August 1877. Bengal . . - Mozufferpore 12-5 17th June 1883. Rangamatia 12-2 8th July 1886. Balasore 12-0 12th July 1877. Nattore 11-1 6th July 1877. Kooshtea . ll'l 20th August 1877. Mymensing 10-8 21st August 1877. Calcutta . 12 + ? 21st August 1844. 12-0 1st May 1835. / Cherrapunji 40-8 14th June 1876. Assam . . . \ Jowai 40-1 llth September 1877. Central India \ Rewah I Sutna 30-4 22-1 16th June 1882. j > Rajputana . Abu . 13-0 2d September 1884. Central Provinces / Jubbulpore 1 Sambalpur 12-5 11-0 29th June 1877. 9th August 1879. Berar . /Chikalda . I Makhla . 12-0 11-0 7th September 1884. j j ( Joongshaie 19-1 5th August 1866. 13-1 6th August 1866. Sind . { Doorbaji . 20-0 4th August 1866. 14-0 5th August 1866. I Hyderabad 10-2 6th August 1866. Cutch . Mandvi 14-8 12th July 1881. f Bombay 161 18th June 1886. jj 15-3 27th June 1869. 14-6 19th June 1877. ? 11-3 3d August 1881. 10-3 15th August 1885. Bombay 1 Rajkote . 10-1 10-1 3d June 1882. 3d July 1883. Baura 15-5 17th June 1886, Mahableshwar 13-0 18th June 1886. il 10-8 22d July 1886, Honawar . 11-3 17th June 1886. iRutnagiri . '' 10-6 22d June 1877. (Madras . 20-6 21st October 1846. Madras i ;; : 18-0 13-0 24th October 1857. 18th May 1877. I I! 12-1 9th May 1827. In all these instances, the day which forms the time unit is the 24 hours intervening between the fixed hours at which the rainfall is regularly measured, and the quantities may, RAINFALL, EVAPORATION, AND WIND 269 in some cases, have been greater in an interval of 24 hours beginning and ending otherwise. The following instances of falls exceeding 14 inches in two, and 18 inches in three days, are subject to the same remark, and are so far from being exhaustive that they can be regarded merely as a few illustrations of excessive falls. RAINFALLS OF 14 INCHES AND UPWARDS IN 2 DAYS PROVINCE. STATIONS. INCHES. DATES. 'Delhi 22-4 8th-9th August 1875. Gurgaon . 19-5 ?> Rohtak 14-5 5 Punjab . Jullundur . 15-0 lOth-llth August 1875. Ludhiana . 15-5 4th-5th July 1880. Amritsar . 14'5 9th-10th July 1881. Hoshiarpur 14-2 2d-3d September 1877. ' Nagina 41-0 17th-18th September 1880. Dhanipur . 391 Najibabad . 38-6 Hurdwar . 31-5 Jauli Jansath 30-6 N.W. Provinces . < Bijnor NainiTal . 24-8 21'4 Roorkee 187 Mozuffernagar . 17-1 Ranikhet . 15-7 Dehra Dun 18-5 8th-9th August 1885. s.Jhansi 15-0 23d-24th July 1869. Bengal Assam Mymensing Cherrapunji Jowai . 16-9 63-6 52-2 21st-22d July 1877. 14th-15th June 1876. llth-12th September 1877. Central Provinces Pachmarhi 14-0 23d-24th July 1884. Sind . Cutch i Doorbaji . \ Joongshaie Mandvi 34-0 32-2 20-5 4th-5th August 1866. 5th-6th August 1866. 12th-13th July 1885. 'Bombay 22-6 18th-19th June 1886. ) > * 20-4 19th-20th June 1877. Bombay 5 j * 14-4 14-4 14th-15th August 1885. 2d-3d June 1882. Mahableshwar . 20-6 22d-23d July 1886. < Rutnagiri . 17'8 22d-23d June 1877. Madras Madras 19-2 17th-18th May 1877. 270 CLIMATES AND WEATHER OF INDIA RAINFALLS OF 18 INCHES AND UPWARDS IN 3 DAYS PROVINCE. STATIONS. INCHES. DATES. Punjab . - . Sialkot . 18'8 9th-llth August 1875. Assam . . Sind . Cherrapunji Doorbaji . 79-0 36-0 12th-14th June 1876. 4th-6th August 1866. ( Drol ' . 33-3 llth- 14th July 1881. Kathiawar . < Nawanagar 28-9 * >5 ( Morvi . 25-5 J) ( Rutnagiri 23-9 21st-23d June 1877. Bombay I Bombay . 28-6 22-3 17th-19th June 1886. 18th-20th June 1877. ^ Karvvar 20-3 15th-17th June 1886. Madras Negapatam 25-0 14th-17th November 1885. Imperfect as these lists undoubtedly are, the above instances of remarkably heavy falls suffice to show how large a quantity of rain, within an interval of two or three days, has sometimes to be reckoned with by engineers in providing for flood drainage. Applying to these cases the law, already found, of the relative frequency of different amounts, it may safely be concluded that the probability of minor but still excessive falls is greater than in the inverse pro- portion of the quantities. The instances of Bombay and Madras, the registers of which have been better examined than those of other stations, show that such falls may be expected at irregular intervals, averaging from 3 or 4 to 1 years ; in the case of the more arid provinces, per- haps somewhat more. But there is probably no part of the country that can be regarded as absolutely exempt from such visitations. It has been mentioned, in the earlier part of this work, that these heavy falls occur only in cyclones and cyclonic storms, most frequently the latter ; and their frequency at any given place is likely, therefore, to vary with the liability of that place to come within the track of such storms. Now the charts of the storm tracks of the 10 years, 1877 to 1886, drawn up by Mr. Eliot, show that the cyclonic storms of the summer monsoon most frequently traverse India in a west -north -west direction from the Orissa coast ; and accordingly, the greater number of the RAINFALL, EVAPORATION, AND WIND 271 falls enumerated in the foregoing lists, including some of the heaviest, have occurred in this belt of maximum storm frequency. The Satpura range, the greater part of which lies within it, is notoriously subject to heavy falls in the monsoon, producing such floods in the Tapti and Nerbudda as were followed by disastrous consequences in 1864 and 1866, noticed by Mr. A. C. Howden in the Proceedings of the Institution of Civil Engineers?- and those of 1877 and 1883. The central part of the peninsula, including the Deccan and Hyderabad, are less frequently visited by cyclonic storms than the Satpura region and extra-tropical India on the north, or the Carnatic on the south ; but in the later months of the monsoon they are sometimes reached by storms from the Bay of Bengal, and the absence from the lists of any instances of rainfalls exceeding the adopted limits is probably due rather to the want of complete data than to the fact of their non-occurrence. The extent of country simultaneously affected by these heavy rainfalls is sometimes very great, usually some thousands of square miles. In the case of the remarkable floods of the 17th and 18th September 1880, in Eohilkhand and the adjacent divisions of Meerut and Kumaon in the North-west Provinces, an average of more than 10 inches in the two days was measured over the greater part of the districts Saharunpore, Mozuffernagar, Meerut, Bijnor, Mora- dabad, Pilibhit, and Kumaon, a total area of not less than 10,000 square miles. This was doubtless quite exceptional, but rather in the heaviness of the fall than in the extent of its horizontal distribution. On the 14th and 15th July 1882, an average fall of 5 inches took place over the greater part of the Central Provinces, south of the Nerbudda and Sone valleys, an area of some 30,000 square miles; and in his Report on the Meteorology of India in 1877, Mr. Eliot states that from the 6th to the 9th October, the average rainfall of the whole of the North-west Provinces and Oudh amounted to 3^ 1 Op. cit., vol. xxvii., session 1867-68. 272 CLIMATES AND WEATHER OF INDIA inches in the four days, the greater part of which fell on the 7th and 8th. The area of these provinces is upwards of 111,000 square miles. Special Hourly Rainfalls. With rare exceptions, it is the practice at Indian observatories to measure the rainfall once only in the 24 hours, and hence the majority of the registers afford no information of its heaviness within shorter periods, and the available statistics on this head are very scanty. A few casually-recorded instances will, however, serve to afford at least some idea of what occasionally occurs in India. The heaviest fall of short duration, within my own re- collection, at Calcutta, was one of 4 inches in an hour and a quarter. But a heavier and more prolonged fall took place on the llth May 1835, when 12 inches were regis- tered within 3 hours, viz. from midnight to 3 A.M. According to Mr. C. Chambers, the greatest amount registered at Bombay in one hour, during the 28 years, 1847 to 1874, is 4'22 inches, viz. between 3 and 4 A.M., on the 12th June 1847. And on the 12th September 1872, 7*2 inches fell in the 2 hours between 10 P.M. and mid- night. Other falls of 3 '07 inches, 3 inches, and five between 2 and 3 inches within an hour, are mentioned by Mr. Chambers as having occurred during the same 28 years. At Madras, 17 inches fell in 12 hours on the 21st October 1846, 6 -2 2 inches in 5 hours .on the 20th Novem- ber 1856, and 12-21 inches in 12 hours on the 24th October 1857. These instances are quoted from Mr. Lowis Jackson's Hydraulic Manual. Other instances mentioned in the same work are 1*2 inches in a quarter of an hour, 1 inch in 35 minutes, 1/6 inches in 40 minutes, and 2 -4 9 inches in 2^- hours at Bangalore ; these occurred respect- ively in September 1860, May 1859, and May and Sep- tember 1861. In Mr. Howden's notice of floods in the Nerbudda valley, in the Proc. lust. Civil Engineers, is mentioned a fall of 10-5 inches in 18 hours at Hurda, on the G. I. P. Rail- way, on the 5th August 1864, and in a letter of Mr. G. P. RAINFALL, EVAPORATION, AND WIND 273 Bidder, published among the remarks on this paper, an instance is quoted of a fall of 3 inches per hour at a station in Sind in 1866. Mr. Binnie also, in a paper on the Ambajhari tank at Nagpur, in the same serial, gives an instance of a fall of 3'55 inches in 45 minutes at Nagpur on the 6th October 1872, and another of 3*5 inches in Ij hours on the 10th August of the same year ; also two other instances of falls lasting respectively 20 and 30 minutes, at rates of between 2 and 3 inches in the hour, all occurring in the same monsoon. From these instances it appears that falls of from 3 to 4 inches in the hour are not rare in India, though but few are recorded at one and the same station ; and, further, that such falls sometimes last for 2 or 3 hours in succession. Proportion of Surface Drainage to Rainfall. The proportion of the rainfall which flows off the ground surface, and either fills the rivers or is available for tank storage, is a subject which concerns the physical geographer and the engineer rather than the meteorologist ; but a knowledge of this proportion is so indispensable to the engineer, in order that he may turn to practical account the information set forth in the foregoing pages, that it seems desirable to supple- ment the rainfall statistics with a brief notice of such ob- servations as have been recorded in India, and in other countries which, like India, are subject to the annual alter- nation of dry and rainy seasons. The various formulae used by engineers for expressing the relations of flood drainage to rainfall need not here be referred to. Some of the elementary facts by which their validity must be tested will suffice for the purpose of the present work. In planning the Vehar reservoir at Tanna for the water supply of Bombay, Mr. Conybeare states 1 that it was assumed that six-tenths of the total rainfall on the catchment area of 3918 acres would be avail- able for tank storage, and that the rainfall would be at least equal to that of the low-lying station Tanna, which he estimated at 124 inches. Nearly the whole of this falls during the four months of the monsoon. 1 Proc. Inst. Civil Engineers, vol. xvii., session 1857-58. T 274 CLIMATES AND WEATHER OF INDIA The mean rainfall of Tanna was here overestimated ; it is really less than 100 inches ; nor does it appear to be higher, if indeed so high, on the hills that feed the Vehar tank. In other respects, the estimate has been better justified by experience. According to Mr. Russell Aitken, 1 the proportion rim off and stored has been as follows. In 1865, when the recorded rainfall was 89 inches, it amounted to 70 per cent ; in 1870, when the former was only 65 inches, to 75 per cent ; and in 1871, a very dry year, when the rainfall was only 39 inches, to 50 per cent. The mean of these is a little over 60 per cent. In the case of the Ashti tank, near Pandharpur, on the Bhima river, in the dry climate of the Deccan, it was estimated that the "run off" would be one-fourth of the average monsoon rainfall, taken at 23 inches. But actual measurements, during the monsoon of 1881, gave an average storage of only one-sixth of the total of 18 inches, and the proportion varied, in different months, from one-five-hundredth to one-third. In July, when on only one day the fall exceeded half an inch, the proportion was one-five-hundredth ; whereas, in Septem- ber, when there were 13 days on which the rainfall was from half an inch to over 2 inches, the proportion was one-fifth ; and in October, on the first 5 days of which 2*75 inches fell, one-third. 2 At Nagpur, observations were made by Mr. Binnie on the dis- charge of the Ambajhari drainage area (4224 acres) in the monsoons of 1869 and 1872. He found that, at the beginning of the monsoon, a fall so heavy as 2^ inches within 1 hour and 20 minutes gave no appreciable surface drainage, the whole being absorbed by the de- siccated soil ; while, in September, 98 per cent of an almost identical fall (2-2 inches in the same interval of time) flowed into the tank in the space of 2 hours and 50 minutes. The total results of the several months were as follow : 10 per cent. 35 44 26-5 4-7 per cent. 22-7 55-8 74-4 39-4 40 The rain of October 1872 was preceded by an interval of dry weather, 1 Proc. Inst. Civil Engineers, "Discussion on Nagpur Waterworks," vol. xxxix., session 1874-75. 2 Idem, "Burke on the Ashti Tank," vol. Ixxvi., session 1883-84, Part ii. 1869. June and July 12-76 inches August 9-61 September . 7-41 Totals of season 29-78 1872. June . . 6'77 inches July . 12-70 August 11-82 September . . 7-99 October 4-37 Totals of season 43-65 RAINFALL, EVAPORATION, AND WIND 275 allowing of the partial desiccation of the ground ; and to this fact was doubtless due the decline of the drainage percentage from 74*4 to 39-4. All the above observations have reference to places where the geological formation is a basaltic or other trap- pean rock, covered with a highly absorbent layer of regur or black soil, which becomes cracked and crumbling after long drought. Evidently, under these conditions, the drain- age proportion varies from to nearly the whole of the rainfall, according to the condition of the soil ; and the proportion collected in any season depends chiefly on the continuity of the rain, and only in a subordinate degree on the total quantity that falls. The high average of the Vehar tank is probably due partly to the greater raininess of the site, and partly to the higher slopes of the catchment area, which is situated among hills, while those of the Ashti and Ambajhari tanks are on the plains of the Deccan and Nagpur. The experience of engineers in New South Wales and Cape Colony may profitably be taken into consideration in connection with the above Indian statistics, inasmuch as, while there is much similarity in the climates, it has reference to larger areas. The Proc. Inst. Civil Engineers contain two papers bearing on the present subject, one by Mr. Coghlan on the relation of the rainfall to the drainage of the Nepean and Cataract rivers, in the neighbourhood of Sydney, and one by Mr. Tripp on that of the river Buffalo, at the Cape of Good Hope. From these the following data are ex- tracted : The area of the Nepean basin is about 284 square miles, that of the Cataract about 70 square miles. They lie to the west of the Illawarra or Mittagong range, at an average elevation of 1200 feet above sea- level, and are, for the most part, uninhabitable. Numerous swamps, chiefly in the Cataract basin, retain the water, and serve to equalise the flow of the rivers, but do not occupy a very large proportion of the area. The remainder is a rugged plateau, intersected by steep and broken ridges. The whole country is barren, and the surface soil poor, resting directly on sandstone. Vegetation is stunted, though the scrub is in many places thick, and in the valleys the gum trees attain goodly proportions. The annual rainfall varies from 75 to 34 inches, the average 276 CLIMATES AND WEATHER OF INDIA being 54 inches at the Cataract, and 44j inches at the Nepean. It has been found that the average proportion between rainfall and dis- charge in the case of the Cataract is 44-7 per cent, while in that of the Nepean it is 44'1 per cent for an extended period, including every condition of flood and drought. The average of freshets is 54 per cent of the rainfall, while, after the ground has been saturated with previous rain, the volume discharged is nearly equal to the volume of rain fall- ing. In dry seasons, on the other hand, the average yield of the streams is not more than 10 or 12 per cent. The quantity of water carried off is, however, not directly proportional to the rainfall, but the greater the quantity of rain falling within a given time, the greater is the percentage of that rain carried off by the streams. The area of that portion of the drainage basin of the Buffalo river, dealt with in Mr. Tripp's paper, is about 105 square miles, and the surface is covered with forest and bush. Owing to this circumstance, unlike many South African rivers, the Buffalo has a small perennial flow, sustained by more or less permanent springs, and varying from 87 cubic feet to 2880 cubic feet per minute, with an average of 933*8 cubic feet. The flood discharge, due to surface drainage, varies from 600 to 285,000 cubic feet per minute above the ordinary flow, and the total ordinary flow is to the flood discharge as about 1 to 2j. The mean rainfall of the drainage basin, on the average of three years, was about 27 inches; but these seasons were dry and abnormal. Its annual distribution was very irregular ; an almost complete drought in one winter being followed by heavy spring rains, while, in the follow- ing year, the chief fall occurred in the winter ; and, in both the second and third years, the springs and summers were very dry. The highest rainfall in any one day was 2 inches, and there were two falls of over 3 inches in 48 hours. Three-eighths of the whole rainfall was contri- buted by falls under half an inch, and the remainder equally divided between falls below and above 1 inch. Under these circumstances, the proportion of the rainfall flowing off the ground from all sources, during the whole period, varied from 0'019 to 0'715 in different months ; the total result being 0'225, or between one-fifth and one- fourth of the rainfall. Evaporation from Water Surfaces. The most trust- worthy and indeed the only practically useful observations on the evaporation of water surfaces are such as are afforded by the same authorities whose works have been quoted in the preceding paragraphs, and to these therefore our atten- tion may be restricted. 1 1 For ascertaining the rate of evaporation of an extensive sea surface, a datum much required by meteorologists and physical geographers, it may be hoped that some day the unique advantages afforded by the Suez Canal and Bitter Lakes will be turned to account. The evaporation of this great sea- RAINFALL, EVAPORATION, AND WIND 277 In the case of the Vehar tank, 2 square miles in area, Mr. Cony- beare found that, during the first dry season, after the completion of the dam, the surface was not lowered more than 6 inches in the month, and he observes that this loss was principally due to leakage through a temporary plug, closing the orifice of a large pipe. In 1874, 16 years after the completion of the work, Mr. Russell Aitken stated, as the result of subsequent experience, that the leakage, together with evaporation from the lake, did not exceed 5 inches per month, and he calculated the leakage from the various dams at Ij inches ; so that the evaporation during the dry season would be about 3j inches a month or an average of 0*116 inch per day. The Ashti tank in the dry Deccan climate (how dry is shown by the figures for Sholapur given below) has an area of 1412 acres. During the six months, December 1880 to May 1881, when no water was drawn off from the tank, the loss by evaporation, absorption, and leakage was as follows in each month. I add for comparison the mean temperature, humidity, and wind movement of the same months, as registered at the Government observatory at Sholapur, less than 40 miles distant, and under similar climatal and topographical conditions. Reduction of Surface Level . Temperature. SHOLAPUR. Humidity. Wind Daily. December 1880 . 0'12 foot 71'4Fahr. 46 per cent. 140 miles. January 1881 . 0-40 707 41 169 February ,, . 0-65 75-6 34 170 March ,, . 0-65 ,, 82-2 ,, 27 154 ,, April , , . 0-95 88'6 28 154 55 May . 1-02 90-4 ,, 30 250 5J Total . 379 Mr. Burke gives no estimate of how much of this total loss of 379 feet may be attributed to leakage, etc. But assuming that the whole was evaporated, it would give an average for this very dry, and in the later months, hot climate, of 0'37 inch per day as the maximum that could possibly have occurred. Careful experiments on a neighbouring tank with an area of 100 acres water surface, gave a total loss from water .surface is made good almost entirely by the influx from the Mediter- ranean and the Red Sea, through two canals of uniform section, the rainfall being very small ; and it only needs a constant gauge of the flux and re- flux in these two channels, say for a twelvemonth, to ascertain the net excess of the flow over the ebb. Arrangements might certainly be made to obtain this by self-registering current gauges, so placed as not to interfere with the canal traffic ; and with one or two rain gauges at Lake Timseh and the Bitter Lakes to register the small rainfall, and a set of thermometers, hygrometer, and anemometer, to record the temperature, humidity, and wind movement, we should have all the data requisite provided at little expense. The subject is well worthy the attention of the British Association or the Academic des Sciences. 278 CLIMATES AND WEATHER OF INDIA October to May, both months inclusive, of 4-41 feet, an average of 0'29 inch per day. At the Ambajhari tank, Nagpur, Mr. Binnie registered the fall of the water surface, foot by foot, from the 10th October 1872 to the 9th June 1873, and deducting 200,000 cubic feet daily, for the estimated " draw off," loss by leakage, etc., found the residual loss due to evapora- tion as follows. The mean temperature and humidity of each period are given in parallel columns. Evaporation. Temperature. Humidity. 10th Oct. to 14th Nov. '. 0'19 in. per day. 74 '5 Fahr. 77 per cent. 14th Nov. to 24th Dec. . 015 70'9 , 80 24th Dec. to 4th Feb. . 0'13 4th Feb. to 16th March . 0'14 16th March to 15th April 0'23 15th April to 7th May . 0'37 7th May to 9th June . 0'19 69-3 82 77'2 63 887 51 917 38 92-6 , 63 Adding to the total loss 3 inches of rainfall, registered during the period, the general average for the whole season was 0'198, or say 0'2 inch per day. The validity of this result depends, of course, on the accuracy of the estimated reduction of the water level from other causes, but it is to be observed that the allowance made for this reduc- tion is only 46 per cent of the whole, the total fall of level averaging 0-0289 feet or 0-347 inch per day, so that it cannot be very greatly in excess. Moreover, some observations conducted by Mr. Culcheth, near Ajmere in Rajputana, with evaporation gauges, partly on land and partly floated on a tank, gave a result almost identical with that com- puted by Mr. Binnie. Allowing for differences of climate, the results of these several observations are sufficiently consistent, and it may be concluded with some confidence that the loss by evapora- tion of large fresh-water tanks in India varies from rather more than one-ninth of an inch daily, on the average of the dry season, in the comparatively damp climate of Bombay, to about one-third of an inch as a possible maximum in the very dry hot climate of the Deccan ; probably less. Some measurements of the evaporation of a large reser- voir (Van Wyk's Vley), in the driest part of Cape Colony, reported by Mr. J. G. Gamble, hydraulic engineer to the Cape Government, gave an annual average of 6 feet, and others near Port Elizabeth, in the south-eastern district, one of 3 feet. The former would correspond to an average of one-fifth of an RAINFALL, EVAPORATION, AND WIND 279 inch, the latter one-tenth of an inch per day, agreeing there- fore well with the Indian results. Rainfall on Indian River Basins. A knowledge of the average quantity of rain falling annually on the drainage hasin of a river, although of less immediate importance to the engineer than is that of flood-producing rainfalls, is nevertheless a fundamental datum in questions of water supply ; and not less so in many problems that concern the physical geographer. Hitherto, as far as I am aware, no one has attempted to frame such an estimate in the case of Indian rivers, and that now put forward can be regarded only as a first attempt, aiming at an approximative result, and subject to indefinite future correction, as the available registers of the rainfall become more numerous, and come to represent longer periods of time. In the basins of the southern tributaries of the Ganges and Jumna, and of those rivers of the Bengal and Bay of Bengal drainage systems that take their rise in Central India and the Satpura highlands, or on the Deccan plateau at some distance from its western margin, the inequalities of the mean annual rainfall in different parts of the area are comparatively moderate, and provided the rain gauge stations are evenly distributed, and so situated as fairly to represent the country around, the error of the average of all their registers, as representing 'the average rainfall of the whole area, is probably not much larger than that of the mean annual rainfall of each individual station, which, as a general rule, is about 1 inch. But in the case of rivers, which, like the Kistna and its tributaries, rise on the crest of the Western Ghats, where the annual rainfall is 200 inches and upwards, and then traverse a broad belt of country where it barely exceeds one-tenth of that amount, there is very considerable difficulty in so selecting the stations that each portion of the basin may be represented in proportion to its area ; and an approximation to within 2 or 3 inches, for the general average of the whole, is the utmost that can perhaps be hoped for. I do not include in these statistics any of the Hima- 2 8o CLIMATES AND WEATHER OF INDIA layan rivers. In their case, or at least that of all the larger streams which bring down the waste of the melting glaciers, to the difficulty arising from the great inequalities of the rainfall on the face of the Himalaya, and on the Punjab and Gangetic plains at its foot, is added that of assigning any probable value to the precipitation of the mountains beyond their outermost spurs ; a difficulty which is all the greater, since the snowfall on the high peaks and glaciers may be very different from the rain and snowfall at the few valley stations for which we have registers. Instead, then, of attempting to give an estimate involving so much con- jecture as to be almost valueless, I have deemed it better to omit these rivers. I have taken for the fundamental data for the computa- tion, the rainfall averages up to 1886, and have corrected those of the minor stations accordingly. I think that the adopted values of the general average of the river basins are rather under than over the true amounts. I begin with the rivers of the Central Indian tableland, that drain into the main arterial system of the Jumna and Ganges ; then follow with those of Western Bengal and Orissa, and the chief river systems of the eastern slope of the peninsula ; and conclude with the great rivers that debouch into the Gulf of Cambay and the Gulf of Cutch. Rivers of the Central Indian Tableland. The chief rivers of this plateau, omitting some minor but still con- siderable streams, are, in order from . west to east, the Chumbul with its great tributary the Banas, the Sind, the Betwa, the Cane, and the Sone. The first four drain Eastern Eajputana, the states of the Central India Agency, and por- tions of the North-west Provinces,Saugor, and Dumoh; and all but the Sind take their rise on the southern margin of the tableland. The Banas, and eventually the Chumbul, receives most of the drainage of the eastern slope of the Aravalis as far south as Udaipur, the driest portion of the whole plateau. The rainfall of this region is little over 20 inches on its western margin, and it increases southwards and eastwards to over 40 and locally over 50 inches in RAINFALL, EVAPORATION, AND WIND 281 the extreme south and east of the tableland. The Sone alone rises at one of the highest points of the Satpura system, where the annual rainfall exceeds 50 inches, and some of its southern feeders in hills of equal elevation, where it exceeds 60 inches, but its main course lies through a country where, as far as the somewhat scanty data show, it is between 40 and 50 inches, and its final portion traverses South Behar, where the fall scarcely exceeds 40 inches. The averages of these five basins I make respectively as follows : Chumbul and Banas ... 28 inches. Sind 34 ,, Betwa 39 ,, Cane 42 Sone 49 ,, Rivers of Western Bengal. These rivers drain the highlands of the Sonthal country and the northern portion of Chutia Nagpur. They all empty themselves into the Bhagirathi arm of the Ganges and the Hooghly. Excepting two small tracts drained by the Damooda and the Adjie, the region has an annual rainfall above 50, and in part above 6 inches. In order from north to south they are Mor ..'... 54 inches. Adjie 52 Damooda 50 Roopnarain . . . . . 59 ,, Cossye ...... 56 Rivers of Orissa. Of these, the Soobunrika and Brah- minee drain the southern part of Chutia Nagpur, all of which is hilly country with a rainfall of 50 inches and upwards. The northern tributaries of the Mahanadi, a much more im- portant stream, take their rise near the .head waters of the Nerbudda and Sone, in hills where the annual fall is from 50 to over 60 inches, and the chief southern tributary, the Tel, in Bustar, where it also exceeds 60 inches, and in places perhaps reaches 70 inches ; but the western feeders traverse the plain of Chhattisgarh, where the fall is, for the most part, between 40 and 50 inches. The general averages are 282 CLIMATES AND WEATHER OF INDIA Soobunrika . ; . . "'. '= '. . 57 inches. Brahminee . . , . 56 ,, Mahanadi . .. . '. . . 53 Rivers of the Deccan. The two great rivers which dis- charge their waters through contiguous deltas, midway down the eastern coast of the peninsula, together bring down nearly the whole drainage of the Deccan tableland. The head waters of all their western feeders lie close under the crest of the Western Ghats, and the torrential rainfall of this hill range fills their channels in June and July, many weeks before rain has fallen in abundance on the rolling plains and flat-topped hills of the Mahratta country and Hyderabad. The northern tributaries of the Godavery, the Paingunga, Wardha and Waingunga, which, united in the Pranhita, form a junction with the Godavery at Sironcha, are of sufficient importance to be treated as independent streams. So also the Bhima, the great northern, and the Tungabhadra the chief southern tributary of the Kistna. Of the former, the Wardha and Waingunga rise in the Sat- pura range, where the rainfall generally varies from 45 to 60 inches, and locally, as at Pachmarhi, exceeds *70 inches. The subsequent course of the Waingunga is wholly through plains with a rainfall of 45 inches and upwards; but the Wardha, skirting the drier plain of Berar farther west, drains a considerable area, where the average is not more than 35 inches. The basin of the Paingunga, which river drains the southern slopes of the Indhyadri range, except its lower part, lies wholly in a region where the annual rainfall is below 40 inches. The main stream of the Godavery rises near Nasik, under the lee of the Ghat range, but it receives the drainage of only a small extent of this range, and for upwards of 100 miles to the eastward, traverses a plain where the fall varies between 20 and 30 inches. Beyond this, all the northern portion of the Hyderabad state, which is drained partly by the main river and partly by its tributary the Manjira, has a rainfall be- tween 30 and 40 inches. The Godavery joins the Pranhita at Sironcha. Below this point, the only important additions RAINFALL, EVAPORATION, AND WIND 283 to its volume are from the hill country of Bustar to the north of the main stream ; of the rainfall of this tract nothing is known, except for a few places about the head waters of the Indravati, where it amounts to between 60 and VO inches. I do not therefore attempt any estimate of the drainage of the Godavery below Sironcha. The Bhima drains about 100 miles of the eastern face of the Ghat range, and its subsequent course to its junction with the Kistna, a distance of more than 250 miles in a direct line, lies entirely through a country where the rain- fall is below 30 inches, in places below 20. The Kistna, above this junction, receives the drainage of more than 250 miles of the Ghats, and then traverses nearly 200 miles of the same dry tract of the Deccan ; and the Tungabhadra, its great southern feeder, drains about 150 miles of the Ghats, and then, also for about 200 miles, crosses the same plain and the equally dry and even drier districts of Nor- thern Mysore and Bellary. The lower course of the Kistna, below the junction of the Tungabhadra near Kurnool, is partly through a country where it is below 30 inches, but in the Nullamalai hill tract below Kurnool, and in that traversed in the last 80 miles of its course, the rainfall ex- ceeds this amount. The estimated averages of these several rivers are as follow : Godavery and tributaries . Kistna and tri- butaries Waingunga . 52 in. \ "Wardha . 40 ,, > Pranhita and tributaries 41 in. Paingunga . 39 ,, ) Godavery and Manjira, above Sironclia . . 37 ,, Bhima and Sina . 49 in. Kistna above junction 59 Whole Kistna basin Tungabhadra . . 43 , Kistna below junction 33 j Rivers of the Carnatic. The very dry tract in the north-east of Mysore and the rocky hills between Bellary and Cuddapah are drained by the Pennaur, which flows past Nellore to the Coromandel coast ; the remainder of Mysore, to the south of this river basin and that of the Tungabhadra, by the Palar, which enters the sea near Chingleput ; the Panar, which debouches at Cuddalore ; and the Cauvery. Of 284 CLIMATES AND WEATHER OF INDIA these rivers, the last only has its sources under the crest of the Western Ghats. The others take their rise in the hilly but not very elevated country of Eastern Mysore, where the rainfall is about 3 inches or somewhat less, and then descending the Eastern Ghats, cross the Carnatic plain, the Palar and Panar receiving a considerable accession from the isolated hill groups that intervene between the Ghats and the coast plain, and which, to judge from the rainfall register of Yercaud on the Shevaroys, have a rainfall in some places exceeding 60 inches. A smaller river, the Vellaur, which enters the sea at Porto Novo, also brings down part of the drainage of these hills. The Cauvery alone receives a con- siderable flood drainage from the Western Ghat range during the summer monsoon. It then traverses the com- paratively dry tableland of Southern Mysore, and descend- ing to a lower level through the Baramahal, where the rainfall exceeds 30 inches, crosses another dry tract lying to the west of the hill groups above mentioned. Finally, passing Trichinopoly, it spreads out its delta in the fertile district of Tanjore, which it irrigates with its flood waters, and itself receives a rainfall of over 40 inches, chiefly at a later season. To the south of the Cauvery the only stream of im- portance is the Vigay, which drains the eastern slopes of the Pulni hills, where the rainfall is probably about the same as on the Nilgiris, which they rival in height. In the lower hills around Madura, which also, in part, drain into the Vigay, the rainfall is over 30 inches; but the plain beyond, intervening between these hills and the coast, is for the most part very dry, and, like the Carnatic, receives its chief rainfall towards the end of the year. The annual average around the lower course of the river is between 20 and 30 inches, and very precarious. The mean rainfall on the drainage basins of these several rivers has been computed as follows : Pennaur . . . .26 inches. Palar . ., . . 36 ,, Panar . 38 RAINFALL, EVAPORATION, AND WIND 285 Vellaur . . . .40 inches. Cauvery . . . 44 ,, Vigay . . . 32 The Tapti and Nerbudda. These two rivers, together, receive the whole drainage of the Satpura range, except that of its eastern extremity, which is shared by the Sone, the Mahanadi, and the Godavery. The Nerbudda carries down little beyond this and the rainfall of the long narrow plain through which it flows, since the Central Indian tableland, which rises a mile or two to the north of the main stream, contributes nothing of importance to its volume. The rain- fall of the Satpura range is locally very variable. In places it is upwards of 6 or 7 inches, but most of the stations at 2000 feet or less have a fall of about 5 inches or somewhat under. The eastern end of the Nerbudda valley has as much as 50 inches, but the rainfall diminishes to between 30 and 40 inches in the western half, below Khandwa, Still farther west, the Malwa Ghats, to the north of the river, have a rainfall between 40 and 50 inches, and some part of this drains into the Nerbudda; but the monsoon of the Arabian Sea, which brings its torrential rainfall to the Western Ghats south of the Tapti, reaches only occasionally so far north, and under ordinary circumstances is more or less diluted and intermingled with the drier currents of the Indus region. The Tapti and its main tributary the Poorna have their sources in the Gawilgarh hills, where the rainfall is nearly as high as at Pachmarhi. But after leaving these hills they flow through the high plains of Berar and Khandesh, in a great part of which the annual average is below 30 inches, receiving from the south only the scanty drainage of a tract of the northern Deccan, where the fall is of even less amount. Only the last 80 miles of its course, where it skirts the northern extremity of the Sahyadri or Western Ghat range, and crosses the coast plain of Surat, is through a wetter country. The rainfall here amounts to between 40 and 50 inches, and is doubtless higher on the hills that hem in its channel. 286 CLIMATES AND WEATHER OF INDIA Both the Nerbudda and Tapti are subject to occasional Hoods of great destructiveness, arising from cyclonic rainfalls on the Satpuras, such as have already been referred to in the earlier part of this section. The mean rainfalls of their basins are respectively as follow : Nerbudda . . . . 46 inches. Tapti . . . " . 39 The Looni. This river carries off the uncertain and intermittent drainage of the west face of the Aravali range to the north of Mount Abu ; and its main course lies through the arid plain of Western Kajputana, where rain at any season is very precarious, and the averages of many years vary between 10 and 15 inches. The soil of the country around the lower course of the Looni is strongly impregnated with salt, and in the dry season the water of the river, as described by Mr. W. T. Blanford, is not merely brackish, but decidedly salt ; whence the name of the river. Looni . . . .18 inches. Wind Pressure. The pressure of the wind, as an element to be considered in providing for due stability in bridges and other structures, is usually so low in India as to be practically negligible. It has been mentioned in the former part of this work that, except in some few parts of the country during the height of the summer monsoon, the wind is too light to allow of the introduction of windmills as a source of motive power, with any prospect of success. But there are few parts of the country that are not liable to occasional storms, either the temporary squalls of the hot season, or the more lasting and destructive cyclones ; and therefore the pressure of the wind on the vertical surfaces of structures cannot be left entirely out of account, though the occasions on which it may put their stability to the proof are rare and may not occur more than once in many years. There exist very few continuous registers of wind pressures for India. The best and longest available to me is that of the Calcutta observatory, which extends over the 10 years RAINFALL, EVAPORATION, AND WIND 287 from April 1867 to March 1877. The instrument used was a self-registering Osier's pressure gauge, exposing a pressure plate of 1 square foot of surface to the direct action of the wind. It was verified more than once, and the indications may therefore be accepted with confidence, in so far as such an instrument is capable of affording trustworthy data. It was set up at an elevation of 73 feet above the ground, on the old office of the Surveyor-General, in Park Street, Chowringhee. During these 10 years the following pressures were registered, omitting all below 10 Ibs. to the square foot : 10 to 19-9 Ibs. ... 34 days. 20 to 29-9 . . 7 30 to 39-9 . . 6 40 to 49-9 ,, . . 3 50 Ibs. . . . 1 this last being the highest the instrument was capable of registering. This was recorded in the cyclone of the 9th June 1869, a by no means severe storm as experienced at Calcutta, and one that caused but little damage. The lowest barometer reading at Calcutta was 2 8 '7 13 inches. In the great cyclone of 1864, which swept the port of its shipping and the city of its trees, and was one of the most destruc- tive storms on record, the pressure gauge was blown away when it had recorded 3 3 Ibs. to the square foot ; and, at the time of that of the 1st and 2d November 1867, it was dis- mounted and under repair. There is therefore no register of the maximum pressure in a great destructive storm. But it must certainly considerably exceed 50 Ibs. It cannot be computed directly from the mean recorded velocity of the wind, because this is by no means constant ; the wind comes in violent intermittent gusts, the pressure oscillating from about 1 Ibs., during the lulls, to its maximum. Nearly all the remaining instances above tabulated occurred in nor'-westers, and rarely more than one gust of the maximum pressure, lasting a few seconds only, has been registered in the same storrn. At Bombay an Osier's pressure gauge was in operation 288 CLIMATES AND WEATHER OF INDIA from 1847 to 1865. Mr. Chambers states that the maximum pressure of the wind recorded by it was " in the hurricane of the 1st and 2d November 1854, and exceeded 2 9 '5 Ibs., on attaining which the instrument became de- ranged." He estimates its maximum force at 50 Ibs. A pressure of 35 Ibs. in the same storm was recorded with a Lind's wind gauge. These are the only instrumental records of wind pressure, known to me, that relate to Indian storms. In the Midna- pore and Burdwan cyclone of the 16th October 1874, it is recorded in Mr. Willson's report, that a railway train of twenty-five carriages, standing on the line at Kanoo junction, was blown over. Mr. Willson obtained the dimensions and weights of the several carriages composing it, and con- sidering them as empty carriages (which they were not), computed the minimum pressure requisite to overturn each vehicle separately. This he found to range from 34'1 to 5 5 '6 Ibs. per square foot. But he remarks that the rails at the place of the occurrence are about 6 feet above the level of the surrounding country, and the lifting force of the wind, under the carriages, which was not considered in the calculation, probably helped to upset them. APPENDICES APPENDIX I THE following tables give the mean and extreme daily and annual temperatures, and their daily, monthly, and annual range ; the humidity, cloud proportion, rainfall (quantity and frequency), and the mean monthly barometric pressure and daily range, at 92 stations, selected from those of the Indian Government observatories. With one or two exceptions the mean values are taken from the tables in the Official Report for 1885, neglecting fractions of Fahrenheit degrees, hundredths of the rainfall inch, and thousandths of the barometric inch. The mean temperature -extremes and ranges, the mean barometric range (diurnal), and the mean number of rainy days have been computed specially for this work from the published reports of the Meteorological Offices of India and Bengal. The monthly rainfall data, those of the wettest and driest years, and the annual extreme temperatures have been corrected up to 1886. The elevations are those of the barometer cisterns above mean sea level. CLIMATIC TABLES 291 1. QUETTA Elevation 5501 feet TEMPERATURE. p RAINFALL. BAROMETER. H M. Range. o 1 || ll * | ea c oS 1 03 1 1 || ft % V January 40 53 29 24 48 65 4-1 1-6 7 24-68 08 February 40 52 31 21 46 66 4'3 1-9 7 64 07 March . 50 64 40 24 47 59 4'9 2'4 8 63 07 April . 58 72 46 26 47 52 3-9 1-3 8 59 07 May 66 83 53 30 47 42 2-2 0-5 3 55 06 June 74 79 59 31 42 41 1-2 o-i 1 44 07 July . 77 92 65 27 43 47 1-7 0-8 2 40 07 August . 75 91 63 28 42 43 1-7 0-6 2 45 08 September . 67 86 52 34 50 37 0-7 0-2 1 55 08 October 56 75 40 35 53 41 0-5 o-i ... 69 08 November 46 64 30 34 52 49 0-9 73 08 December 41 57 28 29 43 54 3'2 0-4 i 71 08 Year . 58 49 2-4 9'9 40 24-59 Mean highest temperature of year 99 lowest ,, 15 , , annual range of temperature 84 Highest recorded reading (9 years) 103 (1878) Lowest 10-8 (1878) Absolute range of temperature . 92 '2 Rainfall of wettest year (9 years) 21 -6" (1885) driest 4 -2" (1879) Rainy days in wettest year (8 years) 73 (1885) driest 18 (1880) 2. LEH Elevation 11,503 feet TEMPERATURE. . p RAINFALL. BAROMETER. M | M. Range. Q S3 rt i * 03 C j 00 es n: P 0} a *j %* c o 1 ID S B| January 18 35 11 25 44 61 6-4 0-2 4 19-64 07 February 19 35 9 26 48 61 67 0-2 5 58 08 March . 31 46 20 26 50 56 6-5 0-2 2 67 08 April . 41 57 31 26 43 42 6'4 o-i 1 67 08 May 47 64 36 27 44 40 6-5 o-i 2 68 08 June 56 74 44 30 50 37 4-9 0'2 1 64 09 July . 62 80 51 29 44 44 51 0-5 3 60 10 August 60 80 50 30 45 48 5-1 0'4 3 63 11 September . 52 72 42 30 46 44 4-0 0'2 1 69 11 October 40 59 30 29 47 41 3-9 0'5 73 10 November 30 47 20 27 43 50 4-0 72 09 December 23 39 13 26 41 58 5-3 o-i 2 70 08 40 49 5-4 27 24 19-66 Mean highest temperature of year 90 ,, lowest ,, -4 annual range of temperature 94 Highest recorded reading (10 years) 9* (1876) Lowest -17 (1878) Absolute range of temperature . 110 Rainfall of wettest year (11 years) 5-4" (1879) driest ,, 0'4" (1876) Rainy days in wettest year (8 years) 43 (1886) driest 8 (1878) 292 APPENDIX I 3. MURREE Elevation 6344 feet TEMPERATURE. z Q RAINFALL. BAROMETER. n 3 o M. Range. 1 a II 1 Month. 1 1 8 ! I ^ | C January 39 50 36 14 34 57 57 2-8 4 23-88 04 February 39 48 34 14 34 61 5-4 3-4 5 84 04 March . 49 59 44 15 40 53 6-1 37 8 86 04 April . May . 57 65 68 75 51 58 17 17 40 41 51 47 5'5 4-4 4-3 3-8 10 9 83 79 03 03 June 71 83 65 17 42 48 37 2-4 7 72 04 July . 68 78 63 15 34 72 6-2 11-0 14 70 04 August . 67 76 62 14 28 80 6-1 14'0 18 74 04 September 65 74 59 15 30 69 3-6 6-1 9 83 04 October 58 69 53 16 34 46 2-2 2-2 4 92 04 November 49 | 60 45 15 33 44 2-9 1-7 3 93 04 December 43 54 40 14 33 49 4-5 1-2 1 92 04 Year 56 56 4-7 56-6 92 23-83 Mean hishest temperature of year 93 lowest ,, 24 ,, annual range of temperature 69 Highest recorded reading (10 years) 98 -7(1880) Lowest 16T (1886) Absolute range of temperature . 82 Rainfall of wettest year (12 years) 71 -8" (1885) driest 39 -1" (1880) Rainydays in wettest year(ll years) 123(1885) driest 74(1880) 4. SIMLA Elevation 7048 feet TEMPERATURE. | g RAINFALL. BAROMETER. C3 n S M. Range. Q I g* SS i c a* jd | >, 'o3 I 1 p B, a 1 .b> si 1 ft 1 a January 41 54 38 16 36 58 5-0 2'8 3 23-29 06 February 41 53 36 17 34 56 5'0 2-7 5 24 05 March . 50 63 45 18 40 53 5-0 3-0 6 27 05 April . 58 72 52 20 38 51 4-6 2'8 6 26 05 May . 64 77 57 20 39 49 41 47 9 22 05 June 67 79 61 18 36 64 6-1 7'9 10 15 05 July . 64 73 61 12 22 88 8-5 19-3 21 12 05 August . 63 71 60 11 20 91 8'6 18-1 22 16 05 September . 61 72 58 14 23 82 6-2 6-0 12 25 05 October 56 68 51 17 28 53 1-0 1*4 2 33 06 November 49 61 43 18 31 50 1-5 0-3 1 32 06 December 45 57 40 17 30 47 3-5 1-1 2 . -31 06 Year . 55 62 4'9 701 99 23-24 Mean highest temperature of year 88 lowest 25 annual range of temperature 63 Highest recorded reading (11 years) 94 -4 (1 879) Lowest 19-7 (1883) Absolute range of temperature . 74 '7 Rainfall of wettest year (24 years) 94 -9" (1864) driest 52-1" (1867) Rainydays in wettest year (12 years) 136 (1884-1885) Rainy days in driest year (12 years) 74 (1877) CLIMATIC TABLES 293 CHAKRATA Elevation 7052 feet TEMPERATURE. , g RAINFALL. BAROMETER. B 3 M. Range. 1 gg II | I 1 d B 1 1 02 i Q 1 ft ft Jj & January 42 51 35 16 35 63 4-5 2-3 5 23-26 05 February . 43 51 35 16 34 63 4-8 27 7 22 05 March . 51 62 44 18 40 53 4-2 3-2 6 25 05 April . , May . 60 64 70 73 50 55 19 18 39 37 46 51 3-7 3-8 1-7 2-9 6 10 24 20 04 04 June 67 75 60 15 31 66 5'2 7-5 12 13 05 July . 64 70 60 10 21 91 8-7 17-2 23 11 05 August . 64 70 59 10 19 91 8-6 157 24 15 05 September . 63 69 58 12 21 83 6-7 5-6 12 23 05 October 58 66 51 15 28 59 1-8 07 2 30 05 November 51 59 43 16 29 56 1-5 0-2 1 31 05 December 46 55 39 16 31 52 3-3 1-1 2 30 05 Year 56 65 4-7 60-8 110 23-22 Mean highest temperature of year 85 lowest ,, 24 , , annual range of temperature 61 Highest recorded reading (11 yrs.)89'6 (1877) Lowest 18-7 (1876) Absolute range of temperature . 70 '9 Rainfall of wettest year (18 years) 80'7" (1885) driest 42'2"(1869) Rainy days in wettest year (12 years) 127 (1885) ,, driest 95(1877) . 6. RANIKHET Elevation 6069 feet TEMPERATURE. g Q t> RAINFALL. BAROMETER. p n s M. Range. o 1 II II \>> ,5 ft 1 1 a OJ > fi 1 J?&> | 2 1 v H 1 January 46 55 42 13 32 61 3-8 1-9 3 24-12 06 February 48 55 40 15 33 60 4'6 2-0 5 08 06 March . 57 65 49 16 37 52 3-8 2-2 5 09 06 April . 65 73 56 17 34 45 3-4 1-3 4 06 06 May . 68 76 59 17 34 52 3-8 3'0 10 .'02 05 June 71 78 64 14 28 64 5'6 6'0 11 23-94 05 July . 68 74 63 11 20 85 8-7 12-7 21 93 05 August . 67 73 62 11 19 86 8-5 11.6 20 96 06 September 66 73 61 12 22 80 6-1 6-0 11 24-03 06 October 61 69 54 15 25 61 1-6 1-1 2 11 06 November 55 62 47 15 27 57 1-2 0-3 1 15 06 December 49 57 43 14 28 56 2-6 1-0 2 14 06 Year 60 63 4-5 49-1 95 24-05 Mean highest temperature of year 86 ,, lowest 30 ,, annual range of temperature 56 Highest recorded reading (11 years) 87'8 (1876) Lowest ,, ,, 26 (1877) Absolute range of temperature . 61 '8 Rainfall of wettest year (16 years) 61 -2" (1874) driest ,, ,, 26 -4" (1873) Rainy days in wettest year (12 years) 107 (1884) ,, driest ,, 83(1877) 294 APPENDIX I , 7. DAEJI LING Elevation (Old Observatory) 6912 feet TEMPERATURE. 3 ft p RAINFALL. BAROMETER. p B 6 M. Range. 1 || Il f 1 1 i HH I 1 . a 'rt % eg Q a 3 ^ January 41 50 35 15 27 78 5-7 07 2 23-40 08 February 44 53 37 16 30 78 5-9 1-2 4 38 07 March. 50 61 44 17 32 71 5-3 2-4 5 38 07 April . May . . 56 59 67 68 50 53 17 15 29 27 77 84 6-1 7'2 3-7 7-1 11 18 37 35 08 07 June . 62 69 58 11 21 91 87 24-1 24 28 06 July . . 63 70 59 11 19 92 8-7 30-5 28 28 06 August 63 70 59 11 19 92 8-5 26-0 26 32 07 September . 61 69 57 12 21 91 81 17-8 22 38 08 October 67 66 51 15 26 81 5-2 6-4 7 44 08 November . 50 "60 43 17 28 75 4-4 0-2 1 47 08 December . 44 54 37 17 27 73 4-3 0-2 1 45 08 Year . 54 82 6-5 120-3 149 23-38 Mean highest temperature of year 78 , lowest 30 i Absolute range of temperature . 58-2 I Rainfall of wettest year (21 years) 160-9" (1S79) ,, annual range of temperature 48 ,, driest ,, 77 -2" (1873) Highest recorded reading (12 years) 84'2 (1877) Rainy days in wettest year (14 years) 171 (1870) Lowest 26 (1874) t ,, driest 124(1878) , 8. DARJILING Elevation (New Observatory) 7421 feet (4 years) TEMPERATURE. j RAINFALL. BAROMETER. p B S M. Range. Q c 9 3d 1-2 , ,4 I i ^0 i ss gg j? 8 hH a e o & 1 % January 40 45 36 9 21 83 4-3 0-7 3 23-00 08 February 39 45 33 12 29 82 4-6 1-3 5 22-92 08 March . 47 56 43 13 27 76 4-7 1-7 4 96 08 April . 53 62 47 15 29 75 6-0 5-3 9 93 08 May . . 54 62 49 13 27 88 7-2 77 18 92 08 June . 56 65 55 10 20 92 8-4 28-4 21 87 07 July . 61 67 57 10 16 94 8-6 28-5 27 86 06 August 61 66 56 10 16 94 9-0 28-5 28 89 07 September . 59 65 55 10 19 94 8-1 16-9 21 99 07 October 54 61 49 12 24 87 5-9 7-5 6 23-05 07 November . 48 56 43 13 23 79 37 o-i 04 08 December . 43 50 38 12 23 81 3-8 0-5 "i 02 08 Year . 51 85 127-1 143 22-95 Mean highest temperature of year 72 ,, lowest ,, 25 annual range of temperature 47 Highest recorded reading (5 years) 72-1 Lowest ,, 19 -9 (1883) Absolute range of temperature . 52 '2 Rainfall of wettest year (5 years) 144'2' 7 (1882) ,, driest ,, 107-5" (1884) Rainy days in wettest year 151 (1885) driest ,, 124 (1883) CLIMATIC TABLES 295 9. SHILLONG Elevation 4792 feet (4 years) TEMPERATURE. s Q P RAINFALL. B H s M. Range. O 1 II II >, i | I 1 m a ! January .... February 51 54 62 64 42 46 20 18 31 30 70 65 2'5 3'4 0-4 0-8 i 4 March .... 62 71 54 17 32 59 3'1 2-0 5 April .... 65 74 58 16 29 65 5'2 37 11 May .... 68 76 63 13 22 77 6'6 10-0 21 June .... 69 76 65 11 18 84 7-8 17-0 24 July .... 69 76 66 10 16 87 8-6 14-0 22 August .... 69 76 65 11 17 88 8-5 14-4 23 September 67 74 64 10 20 89 8-1 15'4 22 October .... 63 72 58 14 25 86 6-2 6-2 13 November 57 66 49 17 30 76 3'4 1-0 3 December 51 61 42 19 31 75 3'2 0-4 1 Year .... 62 77 5'6 85-3 150 Mean highest temperature of year 82 ,, lowest 34 ,, annual range of temperature 48 Highest recorded reading (4 years) 84 (1872) Low 32-5 (1872) Absolute range of temperature . 51-5 Rainfall of wettest year (20 years) 121 -2" (1867) driest ,, 53-6" (1873) Rainy days in wettest year (18 years) 191 (1883, 1884) Rainy days in driest year (18 years) 109 (186S) 10. PACHMARHI Elevation 3528 feet TEMPERATURE. M d t> RAINFALL. BAROMETER. D W s M. Range. o 1 3 a ss II > | d 1 I o5 d I a M 8> *i P 1 jj a January . 58 72 47 25 41 52 2-2 0-5 2 26-55 09 February 62 75 51 24 42 43 2-1 0-3 2 52 09 March 72 85 60 25 44 32 2-1 0-4 1 49 10 April 79 92 69 23 36 26 2-2 0-3 2 42 10 May 83 94 74 20 34 32 2-8 0-6 4 34 09 June 78 86 72 14 30 65 6-6 10-8 14 24 08 July 71 76 67 9 18 89 87 28-8 25 21 07 August 70 75 67 8 16 90 8'4 18-2 23 26 07 September . October 70 67 77 78 66 61 11 17 20 32 83 61 7'3 2'9 15-1 1-9 17 5 34 48 09 09 November . 60 73 50 23 36 53 1-9 0-4 1 54 09 December 56 70 45 25 38 54 2-0 07 2 56 09 Year . 69 57 4-1 78-0 98 41 Mean highest temperature of year 100 lowest ,, 35 annual range of temperature 65 Highest recorded reading(ll yrs.) 102'4 (1881) Lowest 30 (1878) Absolute range of temperature . 72 '4 Rainfall of wettest year (16 years) 106-1" (1S71) driest 62-0" (1880, 1885) Rainy days in wettest year (11 years) 115 (1878) driest 70(1876) 29 6 APPENDIX I 11. CHIKALDA Elevation 3656 feet TEMPERATURE. g Q p RAINFALL. BAROMETER. P K s M. Range. . SB cJ . m i, c c3 r^tO S 01 08 S3 "3 | 1 1 Q ^ 3 2| p s * S January 64 73 56 17 31 52 1-7 0-5 i 26-43 10 February 67 77 59 18 35 43 1-3 o-i i 39 10 March . 75 86 67 19 35 34 1-8 0-4 i 37 11 April . 81 92 72 20 31 27 2-0 o-i 2 29 11 May . . 82 94 73 21 33 35 2'4 0-5 3 22 10 June 75 84 69 15 30 69 6-4 11-9 15 13 08 July . ... 69 74 66 8 19 91 8-9 17-8 24 11 08 August . 68 73 65 8 15 92 8-7 16-6 22 14 08 September . 68 74 65 9 19 89 7-5 12-3 19 21 09 October 69 77 64 13 22 63 3-0 4-6 6 35 09 November 65 74 59 15 25 55 1-9 0-5 2 40 10 December 62 72 55 17 27 53 1-9 1-3 2 43 10 Year . 70 59 4-0 66-6 98 26-29 Mean highest temperature of year 99 lowest 46 , , a nnual range of temperature 53 Highest recorded reading (11 years) 103 (1876) Lowest 39-5 (1880) Absolute range of temperature . 63 '5 Rainfall of wettest year (15 years) 97'1" (1867) ,, driest 39-3" (1866) Rainy days in wettest year (11 yrs.) 121 (1884) driest 80(1876) 12. MOUNT ABTJ Elevation 3945 feet TEMPERATURE. s p P RAINFALL. BAROMETER. p W s M. Range. - a ^ rf i-s . 5 rt 1 cS e fl S 1 a 1 Q S 1 January 58 68 52 16 34 39 2-6 0-2 1 26-15 07 February 60 69 68 16 36 34 3-0 0'4 1 10 07 March . 69 79 62 17 36 30 2-9 o-i 1 08 08 April . 75 85 67 18 32 29 2-6 1 03 08 May . 79 89 71 18 33 35 2-0 1-0 3 25-96 07 June 75 84 69 15 27 65 5'5 51 8 86 06 July . 70 75 66 9 25 85 8-7 22-2 22 82 06 August . 68 73 65 8 20 87 8-4 22-5 22 86 06 September . 69 76 65 11 21 76 6-4 91 15 96 07 October 70 79 64 15 27 45 2-4 2-1 2 26-09 07 November 64 74 57 17 28 36 1-1 0-2 1 13 08 December 59 70 53 17 30 41 1-9 0-2 1 14 07 Year . 68 50 4-0 63-1 78 26-02 Mean highest temperature of year 96 ,, lowest ,, 39 , , annual range of temperature 57 Highest recorded reading (10 years) 101 (1881) Lowest ,, 32-8 (1878) Absolute range of temperature . 68*2 Rainfall of wettest year (27 years) 123" (1862 driest 19'2" (1877 Rainy days in wettest year (8 years) 88 (1884 ,, driest ,, 66 (1885 CLIMATIC TABLES 297 13. OOTACAMUND Elevation 7252 feet (13 Months) TEMPERATURE. j RAINFALL. BAROMETER. W M. Range. o S ! || 1 g i T r. B fi 1 If 1 o a 1 January . , 48 66 35 31 44 55 1-9 0-5 1 23'20 06 February 51 68 39 29 42 61 4'2 0-2 22 07 March . 55 71 44 27 45 46 3-3 1-2 1 23 08 April . 58 72 51 21 29 64 5-6 3-9 11 20 08 May 59 71 53 18 31 70 6'4 6-2 16 16 09 June 56 65 53 12 23 84 87 6-0 17 13 06 July . 55 62 52 10 20 84 9-2 5'6 23 14 05 August . 55 63 51 12 24 82 8-6 4'2 19 14 07 September . 55 63 50 13 25 79 8-0 3-7 13 16 07 October 54 64 50 14 24 89 8-7 9-8 22 21 09 November 53 63 47 16 35 90 67 2'9 16 20 07 December 51 64 42 22 35 71 4'9 1-6 4 18 07 Year 55 71 6 "3 45-8 143 23-18 Highest recorded temperature . . . 77'3 I Rainfall of wettest year (10 years) 58'4" (1830) Lowest ... 25-3 driest ,, ,, 33'7" (1867) Range during year 52 | 1 The rainfall is the mean of 11 years. 14. WELLINGTON Elevation 6200 feet TEMPERATURE. s Q P RAINFALL. BAROMETER. 3 M. Range. o !>> 1 II J| jgi | d 8 i 9 1 1 ,5? a n '5 Q 1 I 1 January 55 66 45 21 34 70 3'3 0'8 3 24-26 07 February 58 70 46 24 37 59 2'8 0-3 1 26 08 March . 63 74 53 21 31 60 2'9 2-0 6 27 08 April . 65 76 56 20 26 63 4'4 2-9 7 23 08 May 66 76 59 17 26 69 5'6 4-1 14 20 07 June 63 72 58 14 22 75 7'4 3-6 15 16 06 July . . 63 71 58 13 22 73 7-5 3'2 14 16 06 August . 62 71 57 14 22 77 7'4 4'0 16 17 07 September 62 71 56 15 25 75 6-9 47 13 19 07 October . j 61 68 55 13 22 84 7'3 9'8 18 21 08 November 58 66 53 13 28 84 7-1 8-5 16 24 08 December 56 65 48 17 34 78 5'6 4-1 10 26 07 Year 61 72 5-7 48-0 133 22 Mean highest temperature of year 80 lowest ,, ,,37 Mean annual range of temperature 43 Highest recorded reading (7 years) S0'9 (1886) Lowest 34 '2 (1885) Absolute range of temperature . 46'7 Rainfall of wettest year (10 years) 63 -91" (1885) driest ,, 40'77"(1S76) Rain y days in wettest year ,, 158 (1880) driest 75 (1876) APPENDIX I 15. NEWARA ELIYA Elevation 6240 feet TEMPERATURE. kj Q p RAINFALL. BAROMETER. D H s M. Range. O 1 Id && ll >, ft I d rf >> a 1 3? Ic 1 & a 1 JanuaTy 57 67 48 19 31 81 5-0 5-6 11 24-01 ? February 57 69 45 24 38 73 3-8 2-5 6 02 1 March . 59 70 46 24 36 73 4-0 3-0 8 03 ! April . May . . 60 61 71 70 48 52 23 18 33 30 78 83 5-0 57 5-6 8'2 12 14 01 23-99 I ? June 59 65 53 12 22 88 8-0 14-4 23 98 ? July . . 58 64 52 12 23 88 7'5 13-7 22 98 ? August '. 59 65 52 13 23 86 7'2 9-4 22 98 ? September . 59 66 52 14 24 86 6'6 8-9 18 99 ? October 59 67 51 16 26 87 6'4 10-5 23 24-00 ? November 59 67 50 17 28 85 5'8 9'2 19 02 ! December 58 66 49 17 30 83 5-6 7-9 17 01 ? Year 59 83 5-9 98-9 195 24-00 ? Mean highest temperature of year 77 lowest 35 ,, annual range of temperature 42 Highest recorded reading (11 years) 79 (1877) Lowest (9 years) 32 (1883) Absolute range of temperature . 47 Rainfall of wettest year (17 years) 129 -3" (18S2) driest 75 "4" (1870) Rainy days in wettest year (11 years) 229 (1877) driest 177 (1SS1) 16. PESHAWAR Elevation 1110 feet TEMPERATURE. g p RAINFALL. BAROMETER. D 3 M. Range. . ,_; X II || * | 1 c g ,3 ! 3 ~f 2 o > January 50 65 39 26 40 62 4-5 1-6 3 28-96 09 February 52 66 40 26 43 57 4-1 1-2 3 91 10 March . 62 76 51 25 47 57 4-8 1-8 ' 4 83 10 April 70 85 60 25 46 55 4-5 2-0 5 69 10 May . 80 96 68 28 48 43 3-0 07 3 56 10 June 89 105 76 29 47 39 2-2 0-3 1 39 10 July . 89 103 78 25 41 51 2-6 17 3 37 11 August . . . 87 100 78 22 38 59 27 2-0^ 3 43 11 September 81 95 70 25 39 56 17 0-8 2 59 11 October 71 86 57 29 46 48 1-5 0'2 1 79 10 November 58 76 44 32 50 55 1-9 0'6 2 91 10 December 51 67 38 29 44 62 3-3 0-6 2 97 10 Year 70 ! 54 3-1 13-5 32 2870 Mean highest temperature of year 115 ,, lowest ,, 29 annual range of temperature 86 Highest recorded reading (10 years) 119 (1880) Lowest 24-7 (1879) Absolute range of temperature . 94 '3 Rainfall of wettest year (27 years) 27'9" (1877) driest 5'0" (1880) Rainy days in wettest year (12 years) 51 (1885) ,, driest 19 (1875) CLIMATIC TABLES 299 17. RAWALPINDI Elevation 1652 feet TEMPERATURE. g . d RAINFALL. BAROMETER. W s M. Range. o 3 * Is S S3 g | a 8 a 03 03 1 g, a 1 2f 1 o S a S January 49 64 38 26 40 72 4-3 2'4 5 28-37 08 February 52 64 40 24 41 70 4-2 2-0 5 31 08 March . 61 76 50 26 47 64 4-3 1-9 5 25 08 April 71 86 59 27 49 58 3-8 2-3 6 13 08 May . 81 96 67 29 49 51 2-8 1-6 5 02 08 June 89 103 74 29 48 49 2-1 17 4 27-88 09 July . 87 98 76 22 42 65 3-9 7'4 9 86 10 August . 84 95 75 20 35 72 37 7-3 11 94 09 September 80 94 69 25 38 67 2-0 3-2 6 28-06 08 October 70 87 56 31 48 56 1-2 0-6 3 23 08 November 57 76 42 34 50 61 1-8 0-9 2 34 08 December 50 67 36 31 43 67 3-0 1-1 3 39 08 Year 69 63 31 32-4 64 28-15 Mean highest temperature of year 114 ,, lowest ,, 29 ,, annual range of temperature 85 Highest recorded reading (11 years) 118 (1880) Lowest ' 23-9 (1878) Absolute range of temperature . 94'1 Rainfall of wettest year (26 years) 46" (1875) driest 20'3" (1880) Rainy days in wettest year (12 years) 76 (1885) driest 43(1878) 18. SIALKOTE Elevation 830 feet ' TEMPERATURE. g . i RAINFALL. BAROMETER. M 3 M. Range. 1 1 |.s 1 d 5 || X gg |ijS 3 -g 1 fi S ftr* 3 ft 1 >s January 52 67 43 24 38 70 4-6 1-4 2 29-21 08 February 56 70 45 25 41 68 4-3 1-8 4 14 08 March . 66 82 55 27 51 58 4-5 1-9 3 05 08 April . 77 93 66 27 49 43 3-9 1-6 3 28-92 08 May . 85 101 73 28 47 38 3-0 1-2 3 81 08 June 91 105 81 24 44 44 2-4 3'2 4 66 09 July . . 87 98 80 18 40 66 4-8 11-6 9 66 08 August . 85 96 78 18 34 71 4-7 91 9 73 09 September 83 95 74 21 36 66 27 3'2 4 85 09 October 75 92 62 30 45 53 1-1 0-6 1 29-03 09 November 62 80 49 31 47 56 1-9 0-4 1 15 08 December 53 70 42 28 40 69 3-1 0-8 1 20 08 Year 73 59 3-4 36-8 44 28-95 Mean highest temperature of year 117 lowest ,,34 ,, annual range of temperature 83 Highest recorded reading (10 yrs. ) 121 '3 (1877) Lowest (9 years) 31 -6 (1878) Absolute range of temperature . 89 '7 Rainfall of wettest year (27 years) 54-9" (1862) driest 20'2" (1880) Rainy days in wettest year (12 years) 59 (1884) driest 35 (1876, 1880) 300 APPENDIX I * 19. LAHORE Elevation 732 feet TEMPERATURE. x g RAINFALL. BAROMETER. o q PH M. Range. i Id O ^ ^ c rl jf 02 1 I"! * 1 c 9 1 * January 54 68 43 25 39 60 3-2 07 2 29-34 08 February 59 71 46 25 32 57 3-4 1-1 3 28 08 March . 69 83 57 26 48 48 3-1 1-1 2 18 08 April . 81 95 66 29 48 37 2-5 0-6 3 05 09 May . 88 102 73 29 48 33 2'4 0-9 3 28-92 08 June 93 107 81 26 45 37 2-9 1-8 3 "77 09 July . 89 99 81 18 39 58 4-2 7-4 7 78 09 August . 88 98 80 18 32 61 37 4-6 6 85 09 September 85 97 75 22 34 55 17 2-4 4 97 09 October . 77 93 62 31 46 46 0-8 0-6 1 29-15 08 November 64 81 49 32 48 47 1-2 0-2 1 30 08 December 55 71 43 28 41 56 2-0 0-5 2 35 08 Year 75 50 2-6 21-9 37 29-08 Mean highest temperature of year 117 ,, lowest ,, 34 ,, annual range of temperature 83 Highestrecordedreading(ll yrs.)120-3 (1879) Lowest 29 "8 (1878)! Absolute range of temperature . 90 -5 Rainfall of wettest;year (27 years) 37'8"(1875) driest ,, 8*7" (1871) Rainy days in wettest year (12 yrs.) 47 (1881) driest 26 (1875) . 20. LTJDHIANA Elevation 812 feet TEMPERATURE. a RAINFALL. BAROMETER. b W s M. Range. 1 Id ll & . . D 1 1 t| & ^* ^ &4 " 1 s rt O M PH Q i s g January . ' 52 69 43 26 40 65 37 1-3 3 29-23 10 February 57 71 46 25 43 64 3-6 1-2 3 17 09 March . 68 84 57 27 48 55 3-8 1-6 3 07 09 April 78 96 66 30 48 50 2-8 0-9 2 28-94 10 May . . 85 103 72 31 47 47 2-2 1-2 3 83 09 June 91 107 80 27 45 49 2-9 2'3 4 69 10 July . 86 99 80 19 36 68 57 97 9 69 10 August . 86 96 80 16 32 70 5-2 6-5 7 76 10 September 83 95 75 20 31 67 2-8 4-3 4 88 10 October 75 92 64 28 43 54 0-8 1-2 1 29-05 09 November 63 80 50 30 46 52 1-3 o-i 18 09 December 54 71 43 28 41 61 2-6 0-9 "i 24 09 Year 73 59 3-1 31-2 40 28-98 Mean highest temperature of year 116 ,, lowest ,, ,, 35 ,, annual range of temperature 81 Highest recorded reading (11 yrs.) 118'3 (1881) Lowest (10 years) 32 1(1878) Absolute range of temperature 86-3 Rainfall of wettest year (27 years) 46'2" (1872) driest 15'6"(1876) Rainy days in wettest year (12 years) 55 (1885) ., driest 24 (1876) i A reading of 24 is recorded in December 1879, but it appears to be of very doubtful validity. CLIMATIC TABLES 301 , 21. DELHI Elevation 718 feet TEMPERATURE. j Q p RAINFALL. BAROMETER. a w s o M. Range. I I d |.s . rd c 03 i d 'gs ^ '"* S ^ i i W ft jg O A* 1 a January 59 71 48 23 38 57 3-4 i-o 2 29-32 10 February 62 75 52 23 42 50 3'0 0-5 2 26 10 March . 74 88 62 26 49 43 3-1 07 2 16 11 April . 84 99 72 27 44 33 27 0-4 1 02 12 May 89 104 78 26 44 39 2-2 07 3 28-92 11 June 93 105 84 21 40 48 3-6 3-4 4 78 11 July . 87 95 81 14 28 68 6-0 8-5 10 78 09 August . 86 94 80 14 25 68 57 6-9 9 85 10 September . 84 94 77 17 27 65 3-5 4-5 5 96 11 October 79 92 68 24 38 49 0-8 0-5 1 29-14 10 November 68 83 56 27 43 45 0-7 o-i 27 10 December 60 74 49 25 37 52 1-8 0-4 1 33 10 Year 77 51 3-0 27-6 40 29-07 Mean highest temperature of year 116 lowest ,, ,, 40 annual range of temperature 76 Highest recorded reading (11 yrs.) 117 -6 (187 Lowest 36-3 (1878) Absolute range of temperature . 81 '3 Rainfall of wettest year (36 years) 43 '5" (1860) driest S'l" (1868) Rainy days in wettest year (12 years) 56 (1885) driest 27 (1877) - 22. SIRSA Elevation 662 feet TEMPERATURE. D o RAINFALL. BAROMETER. w i_i o M. Range. a n* X SsJ 03 M d 3 S) S a 0> OS a* I f 1 a d ft i fij ft a s * January 56 71 42 29 46 55 3-8 0-7 2 29-39 09 February 60 75 46 29 49 49 4-0 0-3 1 33 10 March . 71 88 57 31 55 44 4-3 0-5 1 22 10 April . 82 100 67 33 49 36 3'6 0-4 2 08 10 May . 88 105 75 30 49 37 2'9 0'6 3 28-97 10 June 93 107 83 24 43 42 3'5 2-3 3 83 10 July . 89 100 80 20 36 60 5'4 37 7 82 10 August . 88 99 80 19 30 60 5'4 3'6 5 89 10 September October 85 78 97 95 75 63 24 32 37 49 56 39 3-1 1-0 2-0 0-3 5 1 29-02 20 10 10 November 65 84 48 36 52 40 1-2 33 09 December 57 75 41 34 46 51 2-6 0-3 "i 39 09 Year 76 47 3-4 147 31 29-12 Mean highest temperature of year 116 lowest ,,34 , , annual range of temperature 82 Highest recorded reading Qlyrt.) 118-3 (1879) -Lowest (9 years) 29'9 (1878) Absolute range of temperature . 88 -4 Rainfall of wettest year (33 years) 24 '6" (1863) driest G'2" (1880) Rainy days in wettest year (12 years) 49 (1885) driest 15 (1880) 3 02 APPENDIX I , 23. DERA ISHMAIL KHAN Elevation 573 feet TEMPERATURE. g p H Q 1 RAINFALL. BAROMETER. M. Range. 1 II c ,g . c 1 !"! a i o M a p * fi * January 52 70 41 29 45 55 3-1 0-4 2 29-49 09 February 56 72 44 28 46 50 2-9 0-7 3 44 09 March . 67 83 55 28 50 49 3'2 0-9 3 34 10 April . 77 92 65 27 48 45 2-9 0-8 3 20 10 May . 87 101 74 27 49 39 1-8 0-4 2 07 10 June 93 107 80 27 44 44 1-6 0-6 2 28-90 10 July . . j 91 103 82 21 41 59 2-3 1-8 4 89 09 August. . 90 101 81 20 36 62 2-0 1-6 3 95 09 September . ; 86 100 75 25 39 56 1-0 0-6 2 29-10 09 October . 75 93 62 31 48 46 0-5 o-i 1 29 09 November 62 81 47 34 52 48 1-4 o-i 1 44 08 December 54 72 40 32 46 52 2-3 0-3 1 51 09 Year 74 50 .2-1 8'3 27 29-22 Mean highest temperature of year 117 ,, lowest ,, 31 annual range of temperature 86 Highest recorded reading (11 yrs.)12r5 (1882) Lowest 26 (1878) Absolute range of temperature . 95-5 Rainfall of wettest year (24 years) 16'2" (1878) driest ,, 4 '4" (1880) Rainy days in wettest year (12 years) 35 (1884-1885) Rainy days in driest year (12 years) 12 (1880) | 24. MOOLTAN Elevation 420 feet TEMPERATURE. s o g RAINFALL. BAROMETER. B | M. Range. >< a! S3 | o5 oj d sf c M SS S^ '3 cu i i p Pp> P s January 54 70 43 27 42 58 2-6 0'4 1 29-64 10 February 58 74 46 28 45 52 2'6 0-3 1 59 10 March . 70 85 58 27 50 50 2'4 0-5 1 48 11 April . May . 80 89 96 104 66 76 30 28 49 48 45 45 1-8 1-3 0-3 0-5 1 1 34 20 12 10 June 94 107 82 25 40 51 1-0 0-4 2 05 09 July . 92 103 83 20 34 59 2-3 2-2 3 03 09 August . 89 100 82 18 31 64 2-2 1-3 3 10 09 September . 87 99 77 22 34 62 1-1 0'8 1 24 10 October 77 94 64 30 45 55 0'3 o-i 44 09 November 66 82 52 30 48 53 07 o-i 59 10 December 56 73 43 30 44 56 1-8 0-3 1 66 10 Year 76 54 17 7'2 15 29-36 Mean highest temperature of year 114 ,, lowest ,, ,, 34 annual range of temperature 80 Highest recorded reading(ll yrs.) 118-4 (1876) Lowest 29 (1878) Absolute range of temperature . 89'4 Rainfall of wettest year (25 years) 15'5" (1877) driest 1-9"(1871) Rainy days in wettest year (12 years) 22 (1884) ,, ,, driest .. 9 (1875, 1883) CLIMATIC TABLES 303 , 25. JACOBABAD Elevation 186 feet TEMPERATURE. * d P RAINFALL. BAROMETER. o W M. Range. o d rt X* e . . 4 ti i | * II SS 1" 1 i ri a c3 ft m | & ^ January 57 74 43 31 48 46 2-4 0-2 1 29-89 12 February 62 77 48 29 51 39 3-2 0-2 2 82 12 March . 74 90 60 30 54 41 3-3 0-3 1 70 13 April . 83 99 68 31 51 38 3-1 0-2 1 56 13 May . 91 108 76 32 50 36 1-4 o-i 1 42 12 June 96 111 83 28 44 42 1-1 o-i ... 26 12 July 94 107 83 24 37 53 2-2 1-4 3 24 12 August . 91 103 81 22 33 58 2-4 1-4 4 33 11 September . 88 101 76 25 39 55 0-9 0-3 47 11 October . | 78 96 63 33 51 46 0-4 ... 67 11 November 65 84 50 34 52 45 0-9 0*1 82 11 December 58 75 43 32 47 48 1-8 o-i 90 12 Year 78 46 1-9 4'4 13 29-59 Mean highest temperature of year 118 lowest 32 Mean annual range of temperature 86 Highest recorded reading (9 yrs.) 120-9 (1882) Lowest ,, 29-2 (1881) Absolute range of temperature . 91-7 Rainfall of wettest year (26 years) 12-1" (1869) driest 0-7" (1881) Rainy days in wettest year (9 years) 35 (1878) ., driest ,, 5(1880) 26. HYDERABAD Elevation 94 feeb TEMPERATURE. g d RAINFALL. BAROMETER. P W s M. Range. rt" H s _ O2 C >> ai 1 *3 II f o 1 1 1 ft 1 11 January 63 77 51 26 45 50 1-9 0-2 1 29-96 12 February 66 80 54 26 48 43 2-6 o-i 1 90 12 March . 78 93 64 29 51 42 1-8 o-i ... 78 12 April . 85 101 71 30 44 43 1-3 0-2 66 11 May . 91 106 78 28 44 45 0-7 o-i "i 53 11 June . 91 103 81 22 34 53 1-5 0-4 i 37 09 July . 88 99 80 19 33 61 2-9 2-8 5 36 09 August . 86 95 78 17 29 65 3-1 3-2 5 44 08 September . 86 97 77 20 31 63 17 0-8 2 58 09 October 83 97 71 26 42 46 0-3 ... 75 10 November 72 87 59 28 46 48 0-6 6-1 ... 89 10 December 64 78 52 26 40 48 1-1 97 11 Year 79 51 1-6 8 : b 16 29-68 Mean highest temperature of year 115 ,, lowest ,, 40 annual range of temperature 75 Highest recorded reading (8 years) 121 (1886) Lowest 38-3 (1878) Absolute range of temperature . 82-7 Rainfall of wettest year (21 years) 20'2" driest 1-9" Rainy days in wettest year (9 years) 30 driest 5 1869) 1868) 1878) 1880) 304 APPENDIX I 27. KURRACHEE Elevation 49 feet TEMPKRATURE. S d p RAINFALL. BAROMETER. p M s o M. Range. ^,a3 1 u x eS a* >, 1 1 I m n >> a i ~ & ! 3 $ a a January 65 77 54 23 37 57 2-6 0-6 1 30-02 11 February 68 79 57 22 37 59 3-1 0-3 2 29-98 10 March . 76 86 67 19 39 67 3-3 0-2 1 88 10 April . 80 89 72 17 32 68 ! 2-3 0-2 1 79 09 May . 85 93 79 14 29 74 2-2 O'l 65 08 June . 87 93 82 11 20 74 4'5 0-2 1 52 07 July . 84 90 81 9 18 78 7-1 3-1 6 49 07 August . 82 88 79 9 16 78 6-8 1-8 4 57 .07| September . 82 88 77 11 19 76 4-5 0-9 2 j -69 08 October 80 91 72 19 35 65 1-0 o-i 86 09 November 72 86 62 24 40 56 1-2 o-i "i 96 10 December 67 79 56 23 35 57 2-1 0-2 30-03 10 Year 77 67 3-4 7-8 19 29-79 Mean highest temperature of year 107 lowest 45 annual range of temperature 62 Highest recorded reading (11 yrs.) 117 -6 (1879) Lowest (10 years) 41 (1886) Absolute range of temperature . 76'6 Rainfall of wettest year (31 years) 28 -0" (1869) ,, driest 0-5" (1871) Rainy days in wettest year (12 years) 30 (1878) ,, ,, driest -ir^ 5 " 9 (1880) 28. BICEANJR Elevation 744 feet TEMPERATURE. 1 1 RAINFALL. . BAROMETER. M. Range. Sw rt C? rf CO a ~f 1 03 x'x f "e a g g d a X a i a a January 61 74 49 25 38 37 3-2 0-3 2 29-30 10 February 64 76 53 23 42 39 3-8 0-2 2 23 10 March . 77 90 66 24 48 36 3-9 o-i 1 12 10 April . May . ; 87 93 98 106 76 82 22 24 40 41 32 34 3'2 2-8 0-2 1-6 1 4 00 28-87 11 10 June 95 106 86 20 35 42 2-8 1-8 4 75 10 July . 90 100 83 17 33 55 5-6 3'9 8 72 09 August . 87 97 80 17 27 61 6-0 27 7 80 10 September . 87 97 79 18 27 58 4-0 11 4 92 10 October 84 96 73 23 36 41 1-5 o-i 1 29-10 10 November 71 83 59 24 40 35 0-8 o-i 1 23 10 December 62 75 52 23 35 35 2-4 1 29 10 Year 80 42 3-3 12 : 5 36 29-03 Mean highest temperature of year 114 lowest 41 J annual range of temperature 73" Highest recorded reading(8years)117-4 (1879) Lowest ,, (7 years) 35 (1878) Absolute range of temperature . 82 -4 Rainfall of wettest year (9 years) 17'9" (1878) driest 7 -7" (1885) Rainy days in wettest year (8 years) 43 (1879, 1881) Rainy days in driest year (8 years) 30 (1883) CLIMATIC TABLES 305 29. JEYPORE Elevation 1431 feet TEMPERATURE. ^ d P RAINFALL. BAROMETER. W o M. Range. 5 >, o5 S3 Id c t 1 & 1 J?f * Is SS *c3 g 1 1 ft 1 oj 1 * g January 61 75 49 26 43 53 2-5 0-3 2 28-57 10 February 64 78 50 28 48 45 2-7 0-2 1 54 11 March . 76 89 61 28 51 39 2-8 0-2 2 45 11 April . 85 99 69 30 45 28 2-4 0-2 1 35 11 May 90 104 75 29 46 35 2-6 0-7 5 25 11 June 91 102 80 22 36 52 4-2 3-3 6 13 10 July . 84 91 77 14 28 77 7-3 9-0 15 12 09 August . 82 91 75 16 25 73 6-6 6-1 12 19 09 September . 83 92 73 19 30 67 4-1 3-1 8 29 10 October 78 94 64 30 42 44 1-6 0-2 1 46 10 November 69 84 52 32 46 42 0-7 o-i 1 56 10 December 62 77 48 29 40 52 1-8 0-4 2 60 10 Year 77 51 3-3 23-8 56 28-38 Mean highest temperature of year 112 lowest ,, ,, 38 ,, annual range of temperature 74 Highest recorded reading(6years)115-S (1886) Lowest 34 -5 (1886) Absolute range of temperature . 81 '3 Rainfall of wettest year (18 years) 35-9" (1879) driest 10'6" (1868) Rainy days in wettest year (10 years) 72 (1885) driest 33(1880) 30. AJMERE Elevation 1611 feet TEMPERATURE. g d P RAINFALL. BAROMETER. W S M. Range. a l X rt *"* . SL g" ?> 1 Is Is 1 1 1 1 c i i 1 S 2J January 58 75 45 30 47 51 2-2 0-2 1 28-42 10 February 61 78 48 30 51 46 2-3 0-3 2 37 10 March . 72 90 58 32 53 40 2'4 0-4 1 29 10 April . 83 98 69 29 47 35 2-1 o-i 1 19 10 May . 89 103 77 26 45 39 1-8 0-7 4 08 10 June 88 101 80 21 39 49 3'5 2-5 4 27-98 09 July . 82 93 77 16 30 69 6'3 6-9 13 95 08 August . 80 89 75 14 25 74 6-1 7-3 11 28-02 08 September . 81 91 72 19 31 68 3-5 3'4 7 11 09 October 75 93 63 30 43 50 0-9 0'3 1 30 09 November 66 84 49 35 49 45 0-8 o-i 1 39 09 December 59 77 44 33 46 50 1-4 0-3 1 44 10 Year 74 51 2-8 22-5 47 28-21 Mean highest temperature of year 112 lowest ,, ,, 34 , , annual range of temperature 78 Highest recorded reading (10 yrs.) 119-2 (1879) Lowest 301 (1878) Absolute range of temperature . 89 '1 Rainfall of wettest year (24 years) 36 -4" (1875) driest 9'3" (1868) Rainy days in wettest year (11 years) 55 (1875, 1881) Rainy days in driest year (11 years) 35 (1877) X APPENDIX I 31. DEES A Elevation 465 feet TEMPERATURE. s Q b RAINFALL. BAROMETER. W s M. Range. Id 8-S ,~ S i a S > 5 * 3 9 i i H ^K fi * s January 67 82 50 32 46 38 2-0 o-i 29-55 12 February 71 84 54 30 49 31 2'2 0-2 "i 50 12 March . 81 95 64 31 48 31 2'3 O'l ... 43 13 April . 88 101 71 30 43 28 1-7 o-i i 33 13 May . m 92 105 78 27 42 40 1-4 0-2 i 23 13 June 90 101 81 20 32 53 4'5 2-2 4 13 12 July . 83 91 78 13 28 74 7'8 9-8 13 10 10 August . . , 82 89 76 13 21 73 7'5 8-5 12 18 10 September . October 82 80 91 93 74 66 17 27 27 39 69 44 5'2 1-6 3'3 0-8 8 1 28 42 11 11 November 74 89 57 32 43 35 1-0 o-i ... 51 11 December 69 84 52 32 43 38 1-6 01 1 55 12 Year 80 46 3'2 25-5 42 29-35 Mean highest temperature of year 112 lowest 40 annual range of temperature 72 Highest recorded reading (11 yrs.) 118'6'(1886) Lowest 34-l(lS80) Absolute range of temperature . 84 '5 Rainfall of wettest year (28 years) 51'4" (1884) driest 11 -1" (1877) Rainy days in wettest year (12 years) 65 (1881) driest 20(1877) 32. DEHRA Elevation 2232 feet TEMPERATURE. s - d RAINFALL. BAROMETER. 1 3 M. Range. o I 11 g.a 1 | 1 1 S t d t! January 55 66 45 21 34 I 3-9 2'2 3 2775 08 February 57 69 46 23 38 ?. 4-1 1-9 4 70 08 March . 66 79 55 24 42 ? 3-5 1-4 4 64 08 April 77 88 64 24 42 ? 3-0 0-7 2 55 09 May . 82 93 69 24 40 ? 3-0 1-6 6 47 09 June 84 94 75 19 37 ? 4-3 8-6 10 35 08 July . . 80 86 75 11 23 ? 7'7 25-0 22 35 07 August . . 79 85 74 11 20 ? 7'4 24-8 22 41 08 September 78 85 71 14 23 ? 4-8 9-3 13 50 09 October 71 82 61 21 32 ? 1-0 0-8 3 65 09 November 62 74 51 23 34 ? 1-1 o-i 1 74 09 December 56 68 45 23 33 ? 2-5 0-6 2 77 09 Year i 71 ? 3'9 77-0 92 27-57 Mean highest temperature of year 105 lowest 36 annual range of temperature 69 Highest recorded reading (10 years) 106'8 (1876, 1879) Lowest recorded reading (10 yrs.)33'9 (1876) Absolute range of temperature . 72'9 Rainfall of wettest year (38 yrs.) 119-9" (1885) driest 35 -I" (1848) Rainy days in wettest year(ll years) 114 (1875) driest 60(1877) CLIMATIC TABLES 307 33. ROORKEE Elevation 887 feet TEMPERATURE. a dj b RAINFALL. BAROMETER. p n 1 M. Range. O . 3 * Sa nl "5 d c3 ?8> I M as Z* > 'a rC i 1 d 1 S 1 ft 1 a a January 56 70 44 26 41 64 3-3 2-0 3 29-12 10 February 60 74 47 27 44 60 3'4 1-4 4 06 09 March . 70 85 56 29 50 49 2-5 i-o 3 28-97 10 April 82 97 66 31 49 34 2-3 0-4 1 85 11 May . 88 101 73 28 45 36 1-9 1-2 3 75 11 June 90 102 79 23 42 51 3-3 5-1 7 62 11 July . 85 93 78 15 28 76 6-4 12-5 15 63 09 August . 84 92 77 15 25 76 6-2 12-3 14 69 10 September 83 93 74 19 30 72 4-2 5-1 7 80 11 October 75 89 62 27 43 59 0-8 0-6 1 96 11 November 64 81 49 34 44 56 0-8 0-2 1 29-09 10 December 57 72 43 29 40 62 2-2 0-4 2 14 10 Year 75 58 3-1 42-2 61 28-89 Mean highest temperature of year 113 ,, lowest ,, 35 ,, annual range of temperature 78 Highest recorded reading(10yrs.) 115 -5 (1878) Lowest 30-9 (1879) Absolute range of temperature . 84'6 Rainfall of wettest year (33 years) 74'3" (1853) driest 24-1" (1877) Rainy days in wettest year (18 years) 88 (1874) .driest 26(1877) - 34. MEERTJT Elevation 737 feet TEMPERATURE. g g RAINFALL. BAROMETER. D w s M. Range. o CS 3d ert C C/3 g * bO i a II i | i i c 1 3 21 fi * a January 57 71 44 27 42 63 2-9 i-o 2 29-28 10 February 62 75 47 28 45 57 2-7 0-7 2 23 09 March . 73 88 58 30 52 50 2-6 0-7 2 12 10 April . 1 . May . 83 89 98 101 68 74 30 27 45 43 40 41 1-9 0-4 0-8 1 3 00 28-89 10 11 June 92 102 82 20 38 48 3-2 3-6 5 76 11 July . 86 94 80 14 27 71 6-7 9-2 11 77 09 August . 85 93 79 14 24 73 6-3 7'2 11 83 10 September . 83 93 76 17 28 68 4-1 4-0 6 94 10 October 76 91 63 28 42 56 0-7 0-5 1 29-12 10 November 65 82 50 32 45 54 0-6 o-i ... 25 10 December 58 73 44 29 40 62 1-7 0-3 1 30 10 Year 76 57 2-9 28-5 45 29-04 Mean highest temperature of year 112 lowest 37 annual range of temperature 75 Highest recorded reading (9 years) 115 (1880) T "~ 4 - 33-5 (1878) Lowest Absolute range of temperature . 81-5 Rainfall of wettest year (39 years) 45-3" (1880) driest 15'6" (1868) Rainy days in wettest year (11 yrs.) 62 (1884) driest 29 (1876, 1877) 308 APPENDIX I 35. AGRA Elevation 555 feet TEMPERATURE. a p RAINFALL. BAROMETER. P W s M. Range. o d |x p c < . 03 i a ~ to 1 X* is 1 1 1 I a 1 11 January . 1 60 74 48 26 40 56 2-0 0'5 2 29-49 11 February . i 65 78 52 26 46 47 2-4 0-3 1 43 11 March . .77 92 63 29 51 38 1-9 0-2 2 33 12 April . 88 101 73 28 44 29 1-5 0-2 1 19 12 May . 94 106 80 26 42 32 1-3 0-7 3 08 12 June 95 105 85 20 38 44 3'6 2-9 5 28-96 11 July . 87 94 81 13 26 72 6'4 9-8 13 96 10 August. . j 85 93 80 13 24 73 6'3 6-7 10 29-03 10 September . | 84 October . 80 94 93 77 69 17 24 27 ! 67 40 48 3-9 0-8 4-3 0-4 7 1 14 32 11 11 November . j 70 84 55 29 44 43 0-5 44 11 December . \ 62 75 48 27 39 53 1-3 0-2 "i 50 11 Year . 79 50 27 26-2 46 29-24 Mean highest temperature of year 116 lowest ,, 40 ,, annual range of temperature 76 Highest recorded reading (lOjrs.) 120 -3 (1878) Lowest 36 -3 (1878) Absolute range of temperature . 84 Rainfall of wettest year (40 years) 46'5" (1873) ,, driest ,, ,, lO'O" (1877) Rainy days in wettest year (14 yrs.) 61 (1879) driest 19(1877) - 36. LTJCKNOW Elevation 369 feet TEMPERATURE. S 1 RAINFALL. BAROMETER. D s i M. Range. , c |g i . aj ?> s > s 3 gs f o C 1 hH P S p| E ... 1- January 61 74 46 28 42 59 2-7 0'8 2 29-68 11 February 66 78 50 28 47 52 3-0 0-3 2 61 11 March . 77 91 61 30 52 43 2-7 0-3 1 50 11 April . 87 101 71 30 47 35 2-1 o-i 1 37 11 May . 92 104 76 28 44 42 2-0 0-9 2 27 12 June 92 103 82 21 38 54 4-7 5-0 7 15 11 July . 86 93 80 13 26 74 7-7 10-8 16 15 10 August . 85 92 79 13 23 77 7-4 10-4 15 22 10 September 85 93 77 16 27 72 5-4 7-1 9 31 11 October 79 91 65 26 42 60 1-5 1-4 1 50 11 November 69 83 52 31 44 53 0-8 63 11 December 61 75 45 30 41 58 1-8 0-5 1 69 11 Year 78 57 3-5 37-6 57 29-42 Mean highest temperature of year 114 lowest ,,38 annual range of temperature 76 Iighestrecordedreading(10yrs.) 117-3 (1878) ..owest 34 -4 (1878) Absolute range of temperature . 82'9 Rainfall of wettest year (20 years) 64-9" (1871) ,, driest ,, ,, 14-4" (1877) Rainy days in wettest year (14 years) 77 (1874) driest 27(1876) CLIMATIC TABLES 37. ALLAHABAD Elevation 307 feet TEMPERATURE. g Q > RAINFALL. BAROMETER. b W s M. Range. O a g* s 03 ~ o5 >i d .5? to 3 S3 >* c 03 d S q 1 I fi| 1 o a S 3 January 61 74 48 26 40 68 1-8 0-8 2 29-74 12 February 66 80 51 29 47 58 2-1 0-4 2 67 12 March . 78 93 62 31 51 43 1-8 0-4 1 56 13 April . 88 104 72 32 47 33 1-4 0-2 1 43 13 May . 92 106 78 28 44 40 1-6 0-3 2 33 12 June 91 104 83 21 38 55 4-5 4-6 7 20 11 July . 85 93 80 13 25 81 7'6 ii:9 17 21 10 August . 84 91 79 12 21 82 7'3 9-6 18 28 11 September . 83 92 77 15 24 80 5-2 67 11 38 12 October 78 89 67 22 38 69 1-6 2-3 3 56 11 November 68 82 54 28 41 64 0-9 0-2 68 11 December 61 74 47 27 39 69 1-6 0-2 1 75 11 Year 78 62 3-1 37-6 65 29-48 Mean highest temperature of year 116 ,, lowest ,, 40 ,, annual range of temperature 76 Highest recordedreading(10yrs.) 117-2 (1884) Lowest 36 (1878) Absolute range of temperature . 81 -2 Rainfall of wettest year (39 years) 61-0" (1854) ,, driest ,, 15-7" (1864) Rainy days in wettest year (14 years) 80 (1 879) ,, driest 31(1877) 38. BENARES Elevation 267 feet TEMPERATURE. g RAINFALL. BAROMETER. 5 W 1 M. Range. o 3 u l-s . 4 5? | ??c 3 as * >, a OJ rt s fl m p <3 1 i S 3 January February 61 66 75 81 48 51 27 30 41 47 62 54 2-1 2'3 07 0-5 2 2 2978 ' 71 12 12 March . 77 93 62 31 51 41 2'1 0-4 2 61 13 April . 87 103 72 31 45 34 1-7 0'2 1 48 13 May 91 105 79 26 42 40 1-9 0'5 2 37 12 June 91 102 83 19 35 58 5-0 5'0 7 25 11 July . 85 93 80 13 25 80 7'6 12-8 19 26 10 August 84 91 78 13 21 82 7'4 107 17 31 10 September 83 92 78 14 25 78 5'5 6-5 12 42 11 October 78 90 68 22 38 66 2'3 2-1 3 59 11 November 68 83 55 28 41 60 1-1 o-i 73 11 December 61 75 48 27 39 63 1-6 o-i "i 80 11 Year 78 60 3'4 39-6 68 29-53 Mean highest temperature of year 114 lowest ,,40 , , annual range of temperature 74 Highest record ed reading (10 yrs. ) 1 1 7 '3 (1878, 1884) Lowest 36-4 (1878) Absolute range of temperature 80*9 Rainfall of wettest year (39 years) 65 -3" (1874) ,, driest 19'4" (1864) Rainy days in wettest year (18 years) 91 (1874) driest 43(1868) 3io APPENDIX I , 39. NEEMTTCH Elevation 1639 feet TEMPERATURE. j d RAINFALL. BAROMETER. W 3 M. Range. o c 9 I d cS S j j & 1 ? s? m ss && i c I ft ^ J January ; 62 78 49 29 45 40 2-2 o-i 1 28-37 12 February ; 65 80 51 29 49 35 2-9 0-2 1 32 12 March . ; 76 92 62 30 50 30 2-5 o-i 1 25 12 April . 84 99 70 29 42 28 2-1 o-i 16 12 May . . 88 103 76 27 42 31 2-2 0-5 "3 06 11 June . . 86 98 77 21 38 50 5-0 3-9 8 27-96 10 July . 79 87 74 13 26 74 8'1 11-2 15 94 08 August . 78 85 72 13 20 76 7-8 10-4 14 28-01 09 September . 78 87 71 16 26 72 5'9 5-5 11 09 10 October 76 90 65 25 37 47 2-1 1-0 2 26 10 November 68 83 54 29 42 35 07 ... 34 11 December 63 77 49 28 41 41 17 : 2 1 38 11 Year 75 47 3-6 33-2 57 28-18 Mean highest temperature of year 111 lowest ,, ,,38 annual range of temperature 73 Highest recorded reading (9 yrs.) 113-8 (1886) Lowest 31 '2 (1886) Absolute range of temperature . 82 '6 Rainfall of wettest year (20 years) 48 -6" (1864) driest 16-7" (1868) Rainy days in wettest year (8 years) 68 (1884) driest 45(1880) 40. INDORE Elevation 1823 feet TEMPERATURE. 1 I RAINFALL. BAROMETER. 1 II JL 1 ll 'a 1 i i A ft 8 l| ft i * ^ January 63 76 51 25 40 49 07 0-5 1 29-17 10 February 68 80 54 26 45 41 07 0-2 1 12 10 March . 79 93 65 28 49 34 0-3 0'4 1 03 11 April . 89 102 73 29 41 29 0-4 01 1 28-90 12 May . 95 107 81 26 41 31 0-6 0-3 2 79 11 June 93 104 84 20 38 47 17 4-0 7 68 10 July . 84 92 79 13 37 74 4-3 13-6 15 67 09 August . 83 90 78 12 22 76 3-9 10-5 14 74 09 September . 83 91 76 15 27 70 2-0 5-2 9 84 09 October 81 92 68 24 36 48 0-3 0-8 1 29-01 09 November 73 84 57 27 41 39 0-3 ... 12 09 December 65 77 50 27 39 45 0-3 : 2 i 18 09 Year 79 49 1-3 35-8 53 28-94 Mean highest temperature of year 115 lowest ,, 43 ,, annual range of temperature 72 Highest recorded reading (10 yrs. ) 117 '9 (1878) Lowest 40-7 (1878) Absolute range of temperature . 77'2 Rainfall of wettest year (27 yrs.) 53-8" (1881) driest ,, 13-3" (1868) Rainy days in wettest year (11 yrs.) 75 (1884-1885) Rainy days in driest year (11 yrs.) 24 (1877) x 42. SAUGOR Elevation 1769 feet TEMPERATURE. s g RAINFALL. BAROMETER. D M 3 M. Range. o i 03 ^ gd t cc t>l d * S) a as 1 sl lS j>> '3 rC 1 1 I a Js ft 1 *i ft * % s January 62 78 48 30 46 59 2-2 0-6 2 28-67 12 February ' . 66 82 52 30 50 50 2-3 0-5 1 62 12 March . 76 93 62 31 51 38 2-4 0'5 1 54 13 April . 85 101 71 30 43 29 2-2 0-2 1 44 14 May . 90 104 78 26 41 30 3'3 0-5 4 34 13 June . . 86 97 79 18 37 59 6-5 8-5 13 23 11 July . . 79 86 75 11 23 81 87 18-6 22 22 09 August . 78 85 74 11 19 82 8-2 13-8 19 29 10 September . 79 87 73 14 23 79 6-4 8-3 12 36 11 October . 74 87 64 23 39 65 2-7 1-5 3 53 11 November 66 81 52 29 44 56 17. 0'3 1 65 11 December 61 76 46 30 45 58 1-9 0-3 1 68 12 Year . 75 57 4-0 53-6 80 28-46 Mean highest temperature of year 111 ,, lowest 36 annual range of temperature 75 Highest recorded reading (10 yrs.) 112-5 (1883) Lowest ,, 32 (1879 Absolute range of temperature . 80 -5 Rainfall of wettest year (43 yrs.) 94-8" (1884) driest 28'S" (1868) Rainy days in wettest year (13 yrs.) 107 (1885) driest 69 (1873) CLIMATIC TABLES 3*3 45. PATNA Elevation 183 feet TEMPERATURE. s d P RAINFALL. BAROMETER. P o M. Range. I || I| j- c a cj i i t| X SS X^ o S 1 SI ftpS ft X January 61 73 49 24 36 67 2-4 07 2 29-87 13 February 66 78 52 26 43 58 2-6 0-5 2 80 18 March . 78 92 64 28 47 46 2-9 0-3 2 68 13 April . 87 101 73 28 42 41 2-2 0-3 1 56 14 May 89 100 77 23 41 55 2-8 1-6 4 47 14 June 88 98 81 17 32 68 6'2 7-1 10 35 12 July . 85 91 80 11 23 81 8-0 11-0 17 36 10 August . 84 90 80 10 19 82 7-9 10-1 16 42 11 September . 84 91 79 12 21 80 6-9 7-9 12 52 12 October 80 88 72 16 31 70 3-6 2-9 4 68 11 November 70 82 59 23 37 61 1-7 0-2 82 12 December 62 74 50 24 35 65 2-0 0-2 1 88 12 Year 78 65 4-1 42-8 71 29-62 Mean highest temperature of year 110 ,, lowest 42 ,, annual range of temperature 68 Highest recorded reading (9 yrs.) 112'4 (1884) Lowest (10 yrs.) 36-4 (1878) Absolute range of temperature . 76 Rainfall of wettest year (35 yrs.) 65-5" (1886) driest 21'8" (1843) Rainy days in wettest year (18 yrs.) 87 (1880) driest ,, 51 (1882) 46. DURBHANGA Elevation 166 feet TEMPERATURE. s Q RAINFALL. BAROMETER. W S M. Range. b^oJ a SM 1 c CO g> a 1 as | i a 3 fi| fi i s January February March . 62 65 76 72 76 87 52 53 63 20 23 24 30 36 41 71 63 54 1-3 1-8 1-6 0-4 0-5 0-3 2 1 29-88 81 69 13 13 13 April May . 84 85 96 95 71 75 25 20 39 36 50 65 1-3 2-1 0'6 2-3 2 5 57 51 14 13 June 85 92 79 13 27 77 4'8 7'8 10 38 11 July . 84 89 80 9 18 83 6'4 11-8 14 38 10 August . September October 84 84 79 89 88 86 79 79 73 10 9 13 20 19 24 84 83 77 6'2 5-7 3'2 10-7 9-9 2-8 16 11 3 44 54 71 11 12 12 November 71 81 62 19 31 69 0'9 o-i ... 82 12 December 63 74 54 20 31 75 1-2 0-2 1 89 12 Year 77 71 3-0 47-4 66 29-64 Mean highest temperature of year 105 ,, lowest ,, ,,45 ,, annual range of temperature 60 Highest recorded reading (8 years) 107 '4 (1879) .Lowest (9 years) 38-3 (1878) Absolute range of temperature . 69*1 Rainfall of wettest year (16 years) 78-7" (1871) driest 23-9" (1873) Rainy days in wettest year (15 years) 86 (1871) driest 50 (1872, 1873) APPENDIX 1 47. PURNEAH Elevation 125 feet TEMPERATURE. M p 1 RAINFALL. BAROMETER. 1 M s M. Range. O 1 11 a a f 1 1 1 to ! d > 1 ft i s January , 62 75 48 27 39 67 1-3 0-3 i 29-91 13 February 66 80 51 29 47 61 1-3 0-6 2 84 13 March . 76 91 61 30 48 55 1-6 0-4 1 73 14 April 84 98 69 29 44 55 1-6 1-3 3 61 14 May . , 83 95 73 22 40 69 3-5 3-3 6 56 13 June 84 93 77 16 30 79 6'0 11-2 13 44 11 July . ' . 84 91 79 12 20 85 7'2 14-3 21 43 10 August . 84 90 79 11 21 85 7'2 14-5 21 48 11 September 83 90 78 12 22 83 6-1 14-4 14 59 12 October 79 88 71 17 31 77 3'4 4-0 5 74 12 November 71 83 59 24 36 69 1-0 o-i 84 12 December 63 77 50 27 38 70 1-2 0-2 "i 91 13 Year 77 71 3-5 64'6 88 29-67 Mean highest temperature of year 105 lowest ,,40 ,, annual range of temperature 65 Highestrecordedreading(9years)109-7(1879) Lowest ,, (10 years) 36 -9 (1878) Absolute range of temperature . 72 -8 Rainfall of wettest year (16 years) 131 -4" (1879) driest ,, 38-4" (1883) Rainy days in wettest year (15 years) 112 (1874) driest 68(1872) 48. GYA Elevation 375 feet TEMPERATURE. .j P RAINFALL. BAROMETER. D M S o M. Range. 1 is |s ~ a B a 1 1 t| S ^s S^ jj 1 1 ft S ft p-j ft S ^ January 64 76 51 25 38 54 1-8 07 2 29-68 11 February 69 81 55 26 43 45 2-4 0-5 2 60 11 March . 81 94 66 28 46 37 2-2 0-4 2 50 12 April . 90 103 74 29 43 37 1-6 0-3 1 38 13 May . 92 104 78 26 42 49 2-5 1-2 4 29 13 June . , 89 100 80 20 36 63 6-6 6-2 10 17 11 July . 84 92 79 13 24 79 8'4 11-9 19 18 10 August . 84 91 79 12 20 81 8'5 9-9 17 24 10 September 84 91 78 13 22 78 7-0 7-0 12 33 11 October 80 89 71 18 32 65 3-4 2-5 4 50 10 November 72 83 59 24 38 55 1-7 0-2 1 63 10 December 64 76 51 25 36 57 1-8 0-2 1 69 11 Year 79 58 4-0 41-0 75 29-43 Mean highest temperature of year 113 ,, lowest ,, ,, 43 annual range of temperature 70 Highest recorded reading (11 yrs.)116-2(1878) Lowest 39 -6 (1878) Absolute range of temperature . 76 -6 Rainfall of wettest year (27 years) 67 -4" (1886) driest ,, 20'6"(1854) Rainy days in wettest year (16 yrs.) 99 (1885) driest 54(1873) CLIMATIC TABLES - 49. HAZARIBAGH Elevation 2007 feet TEMPERATURE. p' RAINFALL. BAROMETER. D w S . M. Range. J . 1 II 3i 1 ,0 I d c3 d ci 3 03 i d a IJ January 61 73 51 22 36 50 2-2 0-4 2 27-98 10 February 65 77 54 23 39 43 2-2 0-8 3 93 10 March . 75 88 64 24 43 35 2-5 07 3 87 10 April . 84 97 72 25 39 30 2-6 0-4 2 77 11 May . 85 98 74 24 39 42 3-3 1-6 7 69 10 June 82 92 76 16 32 65 7-3 8-3 15 58 08 July . 78 85 74 11 20 85 9-0 12-6 23 57 08 August . 78 84 73 11 18 85 8-6 12-7 22 62 09 September 78 85 72 13 20 82 7-7 8'0 17 71 10 October 74 83 66 17 28 65 4-1 3-4 7 86 09 November 67 77 57 20 31 53 2-1 0-3 1 97 09 December 61 71 50 21 33 49 2-0 0-2 1 28-00 10 Year 74 57 4-5 49-4 103 27-80 Mean highest temperature of year 106 lowest 43 ,, annual range of temperature 63 Highest recorded reading (10 yrs.)10e -3" (1884) Lowest 39 "2 (1878) Absolute range of temperature . 70-1 Rainfall of wettest year (24 years) 62-2" (1885) driest 35-1" (1872) Rainy days in wettest year (19 yrs.) 134 (1874) driest 73 (1869) 50. BERHAMPORE Elevation 66 feet TEMPERATURE. M d P RAINFALL. BAROMETER. p W s M. Range. o >>o> 1 II II j>> | 1 i 8 hH i 1 ?&c 1 2 1 S a January 65 78 53 25 36 69 2-1 0-4 i 29-97 13 February March . 69 78 82 94 56 66 26 28 42 47 64 57 2-1 2'8 1-0 1-0 2 2 90 79 14 15 April . 85 101 74 27 42 59 3-2 1-9 4 68 15 May 85 96 76 20 40 71 5'3 4-8 9 60 14 June 84 93 78 15 27 83 8-2 97 14 48 11 July . 83 90 79 11 20 86 8-8 10-3 18 49 10 August . September . October 83 83 80 89 89 88 79 78 74 10 11 14 18 20 24 87 86 78 8-8 8-2 4-8 10-8 9'8 5'3 19 15 6 55 64 79 11 12 12 November 73 82 64 18 32 73 2-5 0-3 91 13 December 66 77 55 22 33 72 2-0 o-i 1 97 12 Year 78 74 4-9 55-4 91 29-73 Mean highest temperature of year 109 lowest ,, ,,46 annual range of temperature 63 Highest recorded reading (1 1 yrs.) 1 1 2 -4 (1S76) Lowest 43-1 (1876, 1886) Absolute range of temperature . 69-3 Rainfall of wettest year (31 years) 73" (1861) driest 31 -5" (1873) Rainy days in wettest year (18 yrs.) 113 (1871) driest 72(1874) APPENDIX I , 51. BURDWAN Elevation 99 feet TEMPERATURE. 5 o 1 S RAINFALL. BAROMETER. M. Range. 1 || II >, | d S3 03 fl i | || 1 1 & 0} January 66 79 55 24 37 64 17 0-3 i 29-95 14 February 70 83 58 25 42 58 2-1 0-8 3 88 14 March . 80 94 69 25 44 55 2'2 1-2 3 76 14 April . 86 100 75 25 39 55 2'5 2'4 5 65 15 May . 85 97 77 20 36 67 4'0 5-0 10 58 13 June 84 94 78 16 29 78 7'4 9'8 16 47 11 July . 84 90 79 11 19 83 8'2 12-3 20 46 10 August . 83 89 79 10 17 84 8-2 12-4 23 52 11 September . 83 90 79 11 18 83 7-2 8-2 16 61 12 October 80 88 75 13 25 76 4-3 4-6 6 76 11 November 73 83 64 19 31 66 2-6 0-5 1 88 12 December 66 77 56 21 34 64 2'0 0-2 1 94 13 Year 78 69 4-4 577 105 29-71 Mean highest temperature of year 108 lowest 47 ,, annual range of temperature 61 Highest recorded read ing (llyrs.) 110'7 (1879, 1885) Lowest recorded reading 44 '2 (1878, 1886) Absolute range of temperature . 66'5 Rainfall of wettest year (26 years) 99-3" (1861) driest 40-5" (1870) Rainy days in wettest year(16 yrs.) 131 (1S80) driest 80(1873) 52. CALCUTTA Elevation 21 feet TEMPERATURE. D LOUD. RAINFALL. BAROMETER. M. Range. 1 M i! ! | 1 1 4 c I 1 Q 4* PH January 65 77 55 22 33 71 1-4 0-4 2 30-02 13 February 70 82 61 21 37 69 2-2 1-0 3 29-95 "13 March . 79 91 70 21 38 69 2-2 1-3 3 86 14 April . May 85 85 96 94 76 77 20 17 33 30 71 76 2-8 4-4 2-3 5-6 5 10 75 66 14 12 June 84 91 79 12 25 84 7-0 1V8 18 55 10 July . 83 88 78 10 17 87 8-1 13-0 24 54 09 August 82 87 78 9 16 89 8'2 13-9 24 60 10 September . 82 87 78 9 17 88 7-1 10-0 18 68 n October ' . 80 87 75 19, 33 83 4-9, 5'4 8 83 11 November 72 82 64 18 29 74 2'4 0-6 2 96 12 December 65 76 56 20 31 72 2-0 0'3 1 30-02 13 Year : . 78 78 4-3 65-6 118 29-78 Mean highest temperature of year 102 lowest ,, 48 annual range of temperature 54 Highest recorded reading (8 years) 105 '3 (1879, 1885) Lowest recorded reading (8 years) 45 (1878) Absolute range of temperature . 60 '3 Rainfall of wettest year (57 years) 93 -3" (1S71) driest 43'6" (1837) Rainy days in wettest year (34 yrs.) 154 (1861) driest ,, 72(1853) CLIMATIC TABLES 317 , 53. JESSORE Elevation 33 feet TEMPERATURE. g P W Q 5 RAINFALL. BAROMETER. M. Range. d rt 8jj M . j dj g ^0 S 1 e o 1 (3 c3 ft 3 *! January 65 78 53 25 37 68 1-7 0-5 l 30-00 12 February 70 83 57 26 45 63 2-1 0-9 3 29-94 13 March . 79 93 69 24 44 62 2-8 2-0 5 84 14 April . 83 97 75 22 39 67 3'5 4-7 8 74 14 May . 84 94 76 18 32 76 5-1 7'5 12 67 13 June 83 91 79 12 23 84 7-8 13-5 18 55 10 July . 83 89 79 10 18 87 8-4 11-5 23 54 10 August 82 89 79 10 17 87 7-9 11-6 23 60 10 September . 82 89 78 11 19 86 7-5 9-1 17 69 11 October 80 89 75 14 25 81 4-7 5-6 8 83 11 November 73 83 64 19 33 72 2-4 0-9 2 95 11 December 66 78 55 23 35 70 17 0-2 1 30-00 12 Year 77 75 4-6 68'0 121 29-78 Mean highest temperature of year 104 ,, lowest ,, ,, 44 ,, annual range of temperature 60 Highest recorded reading(10 yrs.) 107'5 (1879) Lowest ,, ,, 39-4 (1884) Absolute range of temperature . 68 !" Rainfall of wettest year (26 yrs.) 100-4" (1861) driest 40'2" (1851) Rainy days in wettest year (18 yrs.) 131 (1880) driest 90(1869) 54. DACCA Elevation 22 feet TEMPERATURE. 9 | RA INFALL. BAROMETER. p w S o M. Range. c *' PH . CO a >a ? \ 1 ll |s "I 1 5 1 p S I| P S s S January 66 78 55 23 34 69 1-6 0-3 1 30-01 12 February 71 82 57 25 39 65 2-0 1-1 3 29-95 13 March . 79 90 69 21 38 68 3-0 2-5 4 86 13 April . 83 93 74 19 34 73 3-8 5'8 8 76 13 May . 83 92 75 17 29 80 5'3 9-2 12 70 11 June 83 90 78 12 20 87 7-6 13-3 17 58 09 July . 83 90 79 11 16 87 7-9 12-8 21 57 09 August . 83 89 80 9 16 87 7'8 12-4 20 62 09 September . 83 88 79 9 18 86 6-9 10-2 15 71 11 October 81 88 75 13 23 81 4-1 5-2 6 84 11 November 75 84 66 18 30 72 2-0 0-7 2 95 11 December 68 78 57 21 30 71 1-4 0-2 1 30-01 12 Year 78 77 4-5 737 110 29-80 Mean highest temperature of year 100 lowest ,, 48 annual range of temperature 52 Highest recorded reading (10 yrs. ) 105-6 (1 879) Lowest 45 "2 (1878) Absolute range of temperature . 60 '4 Rainfall of wettest year (32 yrs.) 112 -6" (1880) driest 52-2" (1874) Rainy days in Avettest year (18 yrs.) 140 (1874) driest (1872) APPENDIX I * 55. CHITTAGONG Elevation 87 feet TEMPERATURE. g g RAINFALL. BAROMETER. W s M. Range. o a 8 l| g.g ^ j j 9 i i ft i S^ 3 1 i * , m P M a January 67 78 55 23 34 72 1-3 0-4 i 29-95 11 February . 71 81 58 23 36 70 1-8 1-2 2 91 11 March 77 86 67 19 34 74 2-9 1-9 4 83 11 April . 81 89 73 16 26 77 4-0 4-6 6 75 11 May . . 82 89 75 14 27 80 5-0 9"2 12 67 10 June . 81 86 76 10 19 86 7-4 23'8 19 57 09 July . 81 86 76 10 18 87 7'5 22-2 24 56 08 August 81 86 76 10 16 87 7'2 20-5 23 60 -09 September . 81 ! 87 76 11 18 86 6-5 14-1 18 68 11 October 80 86 73 13 23 83 4-4 5-7 9 79 11 November . 74 83 66 17 29 79 2-7 1-6 3 88 11 December . 68 78 58 20 31 75 2-1 0-6 1 94 i -11 Year . 77 80 4-4 105-8 122 29-76 i Mean highest temperature of year 94 lowest ,, 48 ,, annual range of temperature 46 Highest recorded reading (10 yrs.) 98-7 (1878) Lowest 45 '4 (1878) Absolute range of temperature . 53'3 Rainfall of wettest year (26 yrs.) 1541" (186S) driest 83-1" (1874) Rainy days in wettest year (19 yrs.) 146 (1881) ,, driest 99(1867) , 56. CUTTACK Elevation 80 feet TEMPERATURE. s Q io RAINFALL. BAROMETER. fi s M. Range. o i S % |.s . . 5 4 ? c > 3 K * * s * S 1 I 1 ft ' 1 ! January 72 85 60 25 38 65 1-7 0-4 1 29-96 13 Februar} 7 76 90 65 25 40 63 1-9 0'6 2 89 14 March . 83 98 73 25 39 62 2-3 1-1 2 81 15 April . 88 102 78 24 37 62 3-0 1-5 3 70 14 May . . 89 100 79 21 35 65 3-8 3'2 8 61 13 June . . . 86 95 80 15 30 74 6-6 107 15 50 10 July . 83 90 78 12 21 81 7'5 12-6 22 50 09 August 83 90 78 12 20 81 7-0 11-2 21 55 10 September . 83 90 78 12 19 82 6-3 9-8 19 64 11 October 81 90 74 16 28 75 4-2 5-8 10 78 11 November . 75 85 65 20 33 68 2-3 1-0 2 91 11 December . 70 82 58 24 34 66 17 0-5 1 97 12 Year . 81 70 4-0 58-4 106 2974 Mean highest temperature of year 110 ,, lowest ,, ,, 51 annual range of temperature 59 Highest recorded reading (10 yrs.) 117 '7 (1879) Lowest 49-1 (1879) Absolute range of temperature . 68-6 Rainfall of wettest year (23 years) 91'9" (1875) driest 41-1" (1877) Rainy days in wettest year (18 years) 124 (1880) driest ,, 80(1876) CLIMATIC TABLES 319 57. DHUBRI Elevation 115 feet TEMPERATURE. d q" p RAINFALL. BAROMETER. B s M. Range. 1 Id g.S ,5 . a I it d f| 1 1 1 $ p * 3 January 63 74 54 20 29 77 1-5 0-3 1 29-94 15 February 65 76 54 22 35 65 1-7 0'5 1 88 15 March . 75 87 65 22 39 61 2-4 1-9 3 77 16 April 79 89 71 18 32 68 2-9 5-1 8 67 15 May 78 86 73 13 29 80 4'6 15'6 16 61 13 June 80 86 77 9 20 87 7-1 26-4 21 50 12 July . 82 88 79 9 17 85 6'6 15'4 18 48 11 August . 81 87 78 9 16 86 7'3 11-4 18 53 12 September 80 85 77 8 19 88 6-6 12-4 15 63 13 October 78 84 73 11 21 81 2-2 4-3 6 78 13 November 71 79 63 16 25 75 1-3 86 13 December 64 74 56 18 27 78 1-2 b-2 "l 93 14 Year 75 78 3-8 93-5 108 29-71 Mean highest temperature of year 98 lowest 48 annual range of temperature 50 Highest recorded reading (5 yrs.) 100'S (1885) Lowest ,, 45-3 (1886) Absolute range of temperature . 55 '5 Rainfall of wettest year (14 yrs.) 156 -3" (1878) ,, driest ,, 69" (1877, 1881) Rainy days in wettest year (8 years) 129 (1881) driest 81 (1873) 58. SIBSAGAR Elevation 333 feet TEMPERATURE. g p b RAINFALL. BAROMETER. D B s M. Range. O 1 II siS >a '3 1 d 3 1 B ! 1 > p| Q & a January 59 70 49 21 32 85 5-4 I'l 7 2974 14 February 62 73 53 20 36 81 6'2 2-2 9 67 14 March . 69 79 59 20 37 79 6'4 4'4 14 59 14 April . 74 82 66 16 30 81 7'3 9-9 18 50 14 May . 78 86 71 15 30 82 8'2 11-0 19 43 13 June . 83 90 76 14 25 83 8-6 14-1 20 31 12 July . 84 91 78 13 23 83 8-8 15-6 22 28 13 August . 83 90 78 12 23 84 87 16-0 21 33 13 September . 82 89 76 13 24 85 8-2 11-7 18 43 14 October 77 85 70 15 29 85 6-9 5-2 10 57 14 November 69 79 60 19 31 84 4-8 1-3 3 67 14 December 61 72 51 21 31 85 4-6 0-6 3 73 15 Year 73 83 7-0 93-1 164 29-52 Mean highest temperature of year 99 lowest 42 ,, annual range of temperature 57 Highest recorded reading (10 yrs.) 102-1 (1883) Lowest 40 (1878) Absolute range of temperature . 62-1 Rainfall of wettest year (30 yrs.) 121-7" (1S71) driest ,, 72 "2" (1866) Rainy days in wettest year (17 yrs.) 195 (1874) ,, driest ,, ,, 137 (1870) 320 APPENDIX 1 59. SILCHAR Elevation 104 feet TEMPERATURE. j d D RAINFALL. BAROMETER. W 3 M. Range. o || |.S . i, 1 ti a ~ ^ ^ "a s p a Op* ft a a a January 64 77 52 25 35 75 2-7 0-6 2 29-94 13 February 67 79 55 24 38 70 3-1 2-6 6 89 13 March 73 85 63 22 36 72 4-0 7-9 12 81 14 April . 78 87 69 18 30 76 5-1 13-0 17 72 14 May . 80 87 72 15 30 81 6-1" 15-7 19 65 13 June . 82 89 76 13 24 85 7'5 19-1 22 54 11 July . 82 89 77 12 21 85 7-6 20-6 25 53 12 August 82 89 77 12 22 86 7-6 18-2 25 58 13 September . 82 89 76 13 24 84 6-9 14-2 19 66 13 October 80 88 72 16 28 81 4'6 6'4 9 78 13 November . 73 84 64 20 32 77 3-2 1-0 2 88 12 December . 66 78 55 23 36 76 2-8 0-7 1 94 13 Year . 76 79 5-1 120-0 159 29-74 Mean highest temperature of year 99 i Absolute range of temperature . 58-8 lowest 45 Rainfall of wettest year (29 yrs.) 188-2" (1S66) ,, annual range of temperature 54 ,, driest ,, ,, 65 -3" (1863) Highestrecordedreading(llyrs.)101-S(1886) Rainy days in wettest year (18 yrs.) 187 (1874) Lowest 43 (1880) I driest 137 (1882) 60. SEONI Elevation 2030 feet TEMPERATURE. g P RAINFALL. BAROMETER. & M 3 M. Range. o 1 II = j5 ts 1 | d oS GO d A ft Q 1 > ! o & I January 64 80 51 29 42 52 3-6 0-7 2 27-96 February 68 84 55 29 43 45 3-7 0-9 3 92 March . 77 94 64 30 48 34 3-4 0-5 2 86 April . 84 100 71 29 41 31 3-7 0-6 2 78 May . 87 103 76 27 40 34 3-5 0-9 6 69 June 82 94 75 19 34 62 4-1 10-1 16 59 July . 76 85 73 12 22 81 4-7 15-0 23 57 August . 76 84 72 12 20 80 4-4 11-0 21 62 September . 76 86 71 15 24 76 3-9 8-2 16 70 October 73 86 65 21 36 59 3'9 1-9 4 85 November . 66 81 55 26 40 50 3-7 0-4 1 94 December 63 77 50 27 40 51 3-6 07 2 98 Year 74 55 3-8 50-9 98 27-79 Mean highest temperature of year 109 ,, lowest ,, 41 * ,, annual range of temperature 68 Highest recorded reading (9 yrs. ) 1 10 -6 Lowest .. 36-4 Absolute range of temperature . 74'2 Rainfall of wettest year (27 years) 70 -0" (1884) driest 31 -4" (1868) Rainydays in wettest year(13 yrs.) 130 (1885) driest 80(1873) CLIMATIC TABLES 321 . 61. KAIPUR Elevation 960 feet TEMPERATURE. j d t> RAINFALL. BAROMETER. s n s M. Range. d rt "* c . o5 d I^H" S) 1 11 X* f 1 1 1 1-1 1 " 1 January February 67 71 81 86 55 61 26 25 39 42 54 49 1-3 1-2 0-3 0-3 1 1 29-05 28-99 12 13 March . 80 97 69 28 45 40 1-6 0-6 1 90 14 April . May . 88 92 103 105 76 80 27 25 40 41 36 36 2-2 3-2 0-6 0-9 2 4 79 69 14 14 June 86 97 78 19 38 63 6-3 10-4 12 60 11 July . 79 87 75 12 24 82 8-1 14-8 18 60 10 August . 79 87 75 12 20 82 7-6 12-1 18 66 10 September . 80 88 75 13 21 79 6-4 7-7 13 73 11 October 77 88 68 20 32 67 3-0 2-2 4 90 12 November 70 82 60 22 39 59 1-9 0-8 1 29-00 11 December 66 78 54 24 36 56 1-5 0-3 1 05 12 Year 78 59 37 51'0 76 28-83 Mean highest temperature of year 112 ,, lowest ,, 46 , , annual range of temperature 66 Highest recorded reading (10 yrs.) 115 -6 (1878) Lowest 41 -4 (1883) Absolute range of temperature . 74'2 Rainfall of wettest year (21 yrs.) 85 -9" (1884) driest 35 -2" (1873) Rainy days in wettest year (13 yrs.) 96 (1885) driest 61(1873) ' 62. NAGPUR Elevation 1025 feet TEMPERATURE. g . g RAINFALL. BAROMETER. P W 3 M. Range. a 1 X |.s . CO d 03 T?> 3 ss |S 3 "3 o 1 fl fl * fijj g 1 January 69 83 55 28 42 51 2-2 0-6 1 28-97 14 February 73 89 59 30 46 42 2-0 0-4 2 91 14 March . 82 99 67 32 49 32 2-5 0-6 2 83 15 April . 89 105 75 30 42 28 3-1 0-5 2 73 15 May . 93 108 80 28 42 30 4-1 0-8 4 64 14 June 86 98 78 20 38 60 7-1 8-8 15 56 11 July . 79 88 75 13 25 80 8'6 13-3 20 56 10 August . 79 88 75 13 23 78 8-3 8-9 18 61 11 September 79 89 73 16 24 76 7'4 7-8 15 67 12 October 77 90 68 22 35 60 4-0 2-3 3 82 12 November 71 84 59 25 39 52 2-6 0-4 1 94 12 December 67 80 54 26 38 52 2-4 0-5 1 98 12 Year 79 53 4-5 44-9 84 2877 Mean highest temperature of year 115 ,, lowest ,, ,, 46 ,, annual range of temperature 69 Highest recorded reading (10 yrs.) 117'7 (1883, 1885) Lowest recorded reading (10 yrs.) 43'1 (1883) Absolute range of temperature . 7.4-6 Rainfall of wettest year (39 years) 65 -3" (1831) driest 25 -5" (1868) Rainy days in wettest year (13 yrs.) 119 (1884) ,, driest 61(1876) 3 22 APPENDIX I 63. AKOLA Elevation 930 feet TEMPERATURE. % g RAINFALL. BAROMETER. H | M. Range. ^ 1 13 ll '5 ! i i 3 rt P 1 l| p % ^ a January 68 85 53 32 47 47 1-3 0-5 1 29-07 14 February 73 90 57 33 50 36 1-5 o-i 1 01 14 March . 82 99 66 33 51 28 1-2 0-4 1 28-94 15 April . 89 105 74 31 45 22 1-3 o-i 1 84 14 May . 93 107 81 26 40 27 1-9 0-2 2 76 14 June 86 98 78 20 37 56 6-2 5-2 11 69 12 July . 80 89 74 15 26 74 8-5 7'8 17 69 10 August. * 79 88 73 15 24 74 8-0 6-7 13 74 11 September . 78 88 72 16 26 75 6'4 5-8 12 81 12 October 77 90 66 24 39 61 2-5 2-2 4 94 12 November 70 86 56 30 45 51 1-3 0-4 1 29-03 13 December 66 82 51 31 43 51 1-5 0-8 2 08 13 Year 78 50 3-5 30-2 66 28-89 Mean highest temperature of year 113 lowest 41 ,, annual range of temperature 72 Highest recorded reading (10 yrs.)115'5 (1881) Lowest 36-7 (1878) Absolute range of temperature . 78 '8 Rainfall of wettest year (25 yrs.) 48-9" (1884) driest ,, ,, 10'4" (1866) Rainy days in wettest year (13 yrs.) 77 (1878) driest 47(1876) > 64. STJRAT Elevation 36 feet TEMPERATURE. s s W CLOUD. RAINFALL. BAROMETER. 1 Id S^ 11 M. Range. 3 ! & | SJ ^ | o a I 3 a January February March . April . May . June July . August . September . October November December Year 70 72 79 84 86 85 81 81 81 80 75 71 79 87 89 96 100 98 94 88 88 88 91 90 86 56 58 65 72 77 79 78 77 76 70 62 58 31 31 31 28 21 15 10 11 12 21 28 28 43 49 48 41 37 27 21 18 21 33 39 41 53 50 51 50 59 68 80 79 79 67 55 55 62 0-8 0-8 0-9 1-0 1-6 5'4 8-0 6-8 6-0 2-4 07 1-1 3-0 29-98 94 88 80 73 63 61 67 74 85 92 97 29-81 12 13 13 13 12 10 08 09 11 11 11 12 0-2 9'4 13-8 9'8 7-3 1-4 o-i o-i 42-1 i 11 20 17 12 4 1 66 Mean highest temperature of year 109 lowest ,, ,, 48 , , annual range of temperature 61 Highest recorded reading (9 yrs.) 113'2 (1881) Lowest (8 yrs.) 45'3 (1879) Absolute range of temperature . 67'9 Rainfall of wettest year (18 yrs.) 89'3" (1878) driest 16'7" (1877) Rainy days in wettest year (8 yrs.) 84 (1878) ,, driest ,, 48(1880) CLIMATIC TABLES 323 . 65. BOMBAY Elevation 37 feet TEMPERATURE. g d RAINFALL. BAROMETER. b w a M. Range. o i i * c . ^ 03 i d rt i^wi p O S S^ ^^ ^ a o i rt p m Pp3 & a * January 74 82 68 14 24 70 1-5 o-i i 29-95 12 February 75 82 69 13 26 69 1-3 i 92 12 March . 79 85 74 11 22 73 1-8 88 12 April . May . 82 85 88 90 77 80 11 10 17 15 75 75 2-3 4-1 0-5 2 81 77 11 10 June 83 88 80 8 17 82 7-9 20-8 20 67 08 July . 81 85 77 8 13 87 9-1 24-7 29 67 06 August . 80 84 77 7 13 87 8-8 15-1 26 73 08 September . 80 84 76 8 13 86 7-5 10-8 21 78 09 October 81 87 76 11 18 81 4-3 1-8 7 84 11 November 80 86 73 13 22 71 2-2 0-5 1 91 11 December 76 84 70 14 23 70 1-8 o-i 95 12 Year 80 77 4-4 74-4 108 29-82 Mean highest temperature of year 95 lowest 61 , , annual range of temperature 34 Highest recorded reading(37 yrs.) 100'2(1857) Lowest 53-3 (1847) Absolute range of temperature . 46 -9 Rainfall of wettest year (40 yrs.) 114-9" (1849) driest ,, 40'6" (1871) Rainy days in wettest year (11 yrs.) 123 (1883) driest 82(1876) , 66. MANGALOEE Elevation 52 feet TEMPERATURE. .. RAINFALL. BAROMETER. W S M. Range. 1 Id l| ^ d d 1 1 f | 1 o g rt p * * January 76 88 69 19 30 62 1-5 0-2 29-91 12 February 78 87 71 16 26 69 1-4 o-i 89 13 March . 81 89 76 13 19 73 2-0 o-i 87 13 April . 83 91 79 12 19 73 2'8 2-0 "3 81 11 May . 83 90 79 11 19 74 4'4 8-1 8 78 09 June . 78 85 75 10 17 86 7-9 37-8 27 78 07 July . 77 83 74 9 15 89 77 37-9 30 79 07 August 77 83 74 9 14 89 7'3 23-1 28 80 08 September . 77 83 74 9 14 87 6-5 11-3 22 84 09 October 78 85 74 11 17 83 5-7 8-0 15 84 10 November . 78 87 73 14 25 76 4-0 1-9 7 84 11 December . 77 88 71 17 26 67 2-6 0-5 2 88 11 Year . 79 77 4-5 131-0 142 29-84 Mean highest temperature of year 94 lowest ,, ,, 63 ,, annual range of temperature 31 Highest recorded reading (7 years) 97 '4 Lowest ,, (6 years) 62-2 (1882, 1884) Absolute range of temperature . 35 '2 Rainfall of wettest year (28 yrs.) 182'3" (1878) driest 95'5" (1881) Rainy days in wettest year (9 years) 156 (1880) driest 1^2(1877) 324 APPENDIX I < 67. MERCAHA Elevation 3695 feet TEMPERATURE. s Q RAINFALL. BAROMETER. p n s M. Range. d 1 3 Ij JJ . 0! s 1 > g S^ S^ rs 48 o 1-1 * jq PH Q s January 64 75 55 20 30 68 27 0-3 1 26-31 10 February . 67 81 57 24 36 60 2-5 31 11 March 70 84 61 23 32 64 3-6 VI "4 29 11 April . 72 85 64 21 29 73 5-2 2-2 8 . 25 10 May . 70 81 65 16 26 80 6'8 6-1 13 23 08 June . 66 72 64 8 18 93 9-6 25-8 26 21 06 July . 64 69 62 7 14 97 9-9 42-1 29 22 06 August 64 70 62 8 16 95 9-5 25-7 27 22 07 September . 64 72 62 10 19 93 9-2 12-3 23 25 08 October 66 75 62 13 22 88 7'9 8-0 17 26 09 November . 65 74 61 13 24 84 6-9 2-6 10 28 09 December . 64 74 57 17 26 77 4-8 0-6 3 30 09 Year . 67 81 6-5 126-8 161 26-29 Mean highest temperature of year 90 lowest 49 ,, annual range of temperature 41 Highest recorded reading (6 years) 91 -1 (1S86) Lowest 47-6 (1884) Absolute range of temperature . 43 '5 Rainfall of wettest year (23 years) 227 '!" (1882) driest 94'2" (1875) Rainy days in wettest year (7 years) 184 (1880) driest ,, 148 (1882) , 68. COCHIN Elevation 11 feet TEMPERATURE. g' Q RAINFALL. BAROMETER. W ,J M. Range. 1 II ll 1 I 1 i I 1 If M S * January 79 89 71 18 25 70 2-2 0-9 1 29-93 13 February . 80 90 73 17 24 72 2-4 07 2 92 15 March 83 91 76 15 21 75 3-1 2-1 5 91 14 April . 84 91 78 13 20 77 4-0 4-4 8 86 13 May . 82 89 77 12 21 81 5-7 127 17 84 11 June . 78 85 74 11 17 88 7-4 30-7 27 86 09 July . 77 83 74 9 16 88 7'2 22-7 27 87 09 August 79 83 74 9 16 87 6-2 12-4 22 88 10 September . 78 84 74 10 15 86 5'5 9-4 19 90 11 October 79 8.6 74 12 17 84 5'9 12-1 19 89 12 November . 79 87 74 13 20 81 4'8 5-1 12 90 13 December . 79 88 72 16 23 75 3'3 1-9 5 92 13 Year . 80 80 4-8 115-1 164 29-89 Mean highest temperature of year 95' ,, lowest ,, 67 annual range of temperature 28 Highest recorded reading(10 yrs.) 99*5 (1879) Lowest 63-6 (1876) Absolute range of temperature . 35 -9 Rainfall of wettest year (23 yrs.) 227'1" (1882) driest 94'2" (1875) Rainy days in wettest year (11 yrs.) 189 (1880) driest 122(1876) CLIMATIC TABLES 325 69. MALAGAON Elevation 1430 feet TEMPERATURE, g . Q p RAINFALL. [BAROMETER. M S M. Range. O I I Se Sri tf. 5 a 3 i | f D & SS ss j 1 S i H a ftpS * a January 68 86 52 34 49 41 1-7 0-2 1 28-56 14 February 72 90 55 35 51 38 1-3 0'2 1 51 14 March . 80 98 64 34 52 31 1-9 46 14 April . 85 102 70 32 45 28 2-1 : 3 "i 37 13 May 88 104 75 29 41 39 2-2 0-7 2 30 13 June 83 95 75 20 33 63 5'7 5-0 10 22 11 July . 79 87 73 14 23 72 7'8 4-2 15 22 09 August . 78 86 72 14 23 72 7'4 4-3 13 27 09 September . 77 86 70 16 26 72 6-8 6-4 15 34 11 October 76 89 65 24 38 57 3-3 2-4 6 45 12 November 70 86 56 30 43 47 2-0 0-4 1 53 12 December 66 83 51 32 45 47 2'1 0'6 1 56 13 Year 76 51 37 24-7 66 28-40 Mean highest temperature of year 109 lowest 40 ,, annual range of temperature 69 Highest recorded reading (8 yrs.) 110'S (1883) Lowest ,, 36-1 (1879) Absolute range of temperature . 74 '7 Rainfall of wettest year (17 yrs.) 38-4" (1884) driest 15'7"(1876) Rainy days in wettest year (8 yrs.) 81 (1884) driest 48(1885) 70. POONA Elevation 1849 feet TEMPERATURE. g p P RAINFALL. BAROMETER. t> W I M. Range. ri Sri 03 fl M i d ! 1 O & SS i tf> ,ri a 1 d 8 5 ^ 1 i Q o g % a January- 72 86 54 32 43 41 1-8 0-2 1 28-13 14 February March . 76 83 90 98 56 64 34 34 50 47 33 29 1-7 2-5 : 2 "i 09 05 14 14 April . 86 101 69 32 43 31 2-4 0-6 2 27-99 14 May . 85 100 71 29 41 42 3-0 1-6 2 94 12 June 79 89 72 17 27 69 7'6 5-6 13 88 09 July . 75 81 70 11 19 79 9-0 6-6 21 87 07 August 75 83 69 14 22 79 87 4-1 20 91 08 September . October 75 78 83 87 68 66 15 21 23 33 77 58 8-1 4-9 4-3 4-1 15 8 97 28-05 10 12 November 75 85 59 26 41 46 2-8 0-8 2 10 13 December 72 83 54 29 41 41 27 0-2 1 14 13 Year 78 52 4-6 28-3 86 28-01 Mean highest temperature of year 106 lowest ,,44 annual range of temperature 62 Highest recorded reading (8 yrs.) 109'2 (1886) Lowest 40-8 (1881) Absolute range of temperature . 68 '4 Rainfall of wettest year (43 years) 56 "9" (1861) ,, driest 14-2" (1844) Rainy days in wettest year (11 yrs.) 94 (1883) driest 71(1876) 326 APPENDIX I 71. SHOLAPUR Elevation 1590 feet TEMPERATURE. S jj RAINFALL. BAROMETER. & s M. Range. o jj a ix" cS C to !> d r^?> 1 aa a* 1 ,15 a 1 i a fi 1 "3 g 0(2 January 72 87 58 29 42 42 2-2 28-40 13 February 77 94 62 32 43 34 1-9 O'l i 35 14 March . 83 100 68 32 43 29 2-5 0-3 i 30 14 April . 88 105 74 31 39 26 3'2 0-7 3 21 15 May 89 104 76 28 39 34 4-0 1-2 4 15 14 June 82 95 72 23 35 56 7-1 4-6 14 12 12 July . 79 90 71 19 28 65 8-1 4-3 16 12 10 August 78 89 71 18 27 66 8-1 6-0 17 15 11 September . 77 88 70 18 26 70 8'3 7'5 14 20 12 October 77 89 68 21 32 60 5-9 37 8 29 12 November . 73 87 63 24 39 53 37 07 3 35 12 December 70 85 58 27 39 49 3-0 0-4 2 39 12 Year 79 49 4-8 29-5 83 28-25 Mean highest temperature of year 110 lowest ,,47 ,, annual range of temperature 63 Highest recorded reading (8 yrs.) 112'1 (1884) Lowest 42-8 (1884) Absolute range of temperature . 69 -3 Rainfall of wettest year (32 yrs.) 68 -0" (1878) driest 10'6" (1876) Rainy days in wettest year (8 yrs.) 99 (1878) driest 58(1881) 72. BELGATJM Elevation 2550 feet TEMPERATURE. g Q t> RAINFALL. BAROMETER. c> W s M. Range. o C ^ ** C SL d > S la |s >* % 5 o 1 1 3 >> 1 1 21 M S S m January 72 83 58 25 36 47 1-7 o-i 27-44 13 February 76 89 60 29 41 39 1-5 41 13 March . 80 95 65 30 49 41 2-1 : 5 3 38 14 April . 81 97 67 30 40 50 2-6 2-0 6 32 13 May . 80 94 68 26 38 59 3-4 2-8 7 29 12 June 74 82 68 14 26 80 77 9-3 21 25 08 July . 71 76 67 9 16 89 8-8 15-2 28 25 07 August . 70 77 66 11 18 88 8'5 9-0 25 27 08 September . 71 78 65 13 21 86 77 3-7 22 32 10 October 74 83 65 18 27 70 5-5 4-7 13 36 12 November 73 82 62 20 31 57 37 1-2 5 40 12 December 71 82 59 23 33 50 2-9 0-3 3 43 12 Year 74 63 47 48-8 133 27-34 Mean highest temperature of year 102 ,, lowest ,, ,, 51 annual range of temperature 51 Highest recorded reading (10 yrs.) 109 (1879) Lowest 47-1 (1878) Absolute range of temperature . 61 '9 Rainfall of wettest year (42 yrs.) 73-4" (1882) driest 35'2" (1880) Rainy days in wettest year (11 yrs.) 174 (1883) driest 91(1876) CLIMATIC TABLES 327 , 73. SECUNDERABAD Elevation 1787 feet TEMPERATURE. g d RAINFALL. BAROMETER. p 3 M. Range. o . | g M 1-3 j o5 03 >> 3 03 a| i ^ |s ; c8 g I ~ M Q 03 a ojj Q i! * a January 70 84 57 27 37 54 1-7 0-3 1 28-19 14 February 76 91 62 29 40 45 1-4 0-2 15 15 March . 82 98 69 29 40 39 1-7 0-7 "3 10 15 April . 87 102 75 27 40 36 1-9 0-7 4 03 15 May . 89 103 78 25 39 37 2-6 1-4 5 27-95 15 June 82 94 74 20 34 58 5-4 3-7 12 91 13 July . 77 86 71 15 26 72 6-7 6-0 17 91 11 August . 77 86 70 16 25 72 6-0 57 18 94 12 September . 76 86 70 16 24 75 5'9 5'2 15 98 13 October 76 87 67 20 31 66 4-0 3-3 8 28-07 13 November 72 83 61 22 35 60 2-9 0-8 4 15 14 December 69 81 56 25 37 57 2-2 0-3 2 19 14 Year 78 56 3-5 28-3 89 28-05 Mean highest temperature of year 109 lowest ,, 48 ,, annual range of temperature 61" Highest recorded reading (9 yrs.) 110-4 (1883) Lowest 41 -7 (1879) Absolute range of temperature . 6S'7 Rainfall of wettest year (44 yrs.) 44'6" (1878) driest 12'9" (1859) Rainy days in wettest year (11 yrs.) 116 (1883) driest ., 57(1876) 74. BELLARY Elevation 1455 feet TEMPERATURE. g ' RAINFALL. BAROMETER. M s M. Range. o a |g' G j 02 g, 4 :;?> 05 ! % ^ e 1 a Q( 1 I a January 73 88 61 27 38 50 2-3 o-i 28-48 14 February 77 94 65 29 41 37 1-5 ... 44 15 March . 85 101 72 29 40 33 2-3 0'6 i 39 15 April . 89 104 77 27 37 35 3-1 0'8 3 32 15 May . 88 103, 77 26 38 42 4-8 1^8 5 27 15 June 83 95 76 19 30 58 7'4 1-8 6 26 12 July . . 81 92 75 17 27 62 8'0 1-3 8 27 11 August 81 92 74 18 26 63 7'4 2-3 8 29 12 September . 80 91 73 18 27 64 7'3 3-7 9 32 13 October 79 90 71 19 30 64 6'3 3-9* 9 38 14 November 75 87 66 21 34 61 4-7 1-0 5 44 14 December 73 86 61 25 35 57 3-4 0*3 1 48 14 Year 80 52 4-9 17-6 55 28-36 Mean highest temperature of year 108 lowest 54 , , annual range of temperature 54 Highest recorded reading(ll yrs.) 110-3 (1882) Lowest (9 yrs.) 50'8 (1879) Absolute range of temperature . 59 -5 Rainfall of wettest year (33 yrs.) 30 '2" (1885) driest 7 -2" (1876) Rainy days in wettest year (11 yrs.) 69 (1879) driest 33(1876) 328 APPENDIX I , 75. BANGALORE Elevation 2981 feet TEMPERATURE. ri 6 p RAINFALL. BAROMETER. p H s M. Range. c i II 11 |J i g g a i i I S > 1 1 i 1 a January 67 79 56 23 33 60 2-8 0-2 1 27-04 12 February 72 85 59 26 36 52 2-3 o-i 02 13 March . 77 90 64 26 37 49 2-6 0-6 "2 26-99 13 April . May . 80 79 93 92 69 69 24 23 32 33 52 61 3-5 4-5 1-3 5-0 3 10 94 90 13 12 June . . 74 85 67 18 27 73 57 3'2 13 88 10 July . ,. 72 83 66 17 24 77 6-2 4-0 14 89 09 August 72 83 66 17 24 77 6-1 5-9 17 90 11 September . 72 82 65 17 25 76 5-9 6'3 13 93 12 October 72 82 65 17 24 75 57 6-4 13 96 12 November 70 79 62 17 27 73 4-9 1-9 7 27-00 12 December 67 78 59 19 29 68 4-0 07 3 04 12 Year 73 66 4-5 35-6 96 26-96 Mean highest temperature of year 98 lowest ,, ,, 51 annual range of temperature 47 Highest recorded reading (10 yrs.) 99'4 (1882) Lowest 45-8 (1877) Absolute range of temperature . 53'6 Rainfall of wettest year (51 yrs.) 56'7" (1874) driest 16'0" (1838) Rainy days in wettest year (11 yrs.) 149 (1878) driest 73(1876) 76. MASULIPATAM Elevation 10 feet TEMPERATURE. s d RAINFALL. BAROMETER. K M. Hange. o I II II f 1 1 1 j! i 1 || S January 74 83 66 17 27 73 3-3 0-3 i 30-01 12 February . . 77 87 68 19 29 74 3'2 o-i i 29-96 12 March . 81 92 73 19 30 74 3-5 0-3 i 90 13 April . 85 95 77 18 30 75 3-9 o-i i 81 13 May . 88 99 81 18 36 72 4-8 17 4 71 13 June 87 98 81 17 34 67 6-6 4'4 8 65 12 July . 84 94 79 15 27 72 6-9 5'6 15 67 12 August 84 92 78 14 25 74 6-6 6-0 16 70 12 September . 83 91 78 13 23 76 6-3 6-5 16 74 13 October 81 89 76 13 24 78 5-5 8'8 12 84 12 November , 77 85 71 14 24 78 5*1 4'0 6 94 12 December 74 83 67 16 24 75 4-2 07 2 30-00 12 Year 81 74 5'0 38-5 83 29-83 Mean highest temperature of year 109 lowest 59 ,, annual range of temperature 50 Highest recorded reading (11 yrs.) 115-9 (1878) Lowest (10 yrs.) 58-1 (1883) Absolute range of temperature . 57 '8 Rainfall of wettest year (24 yrs.) 69'4" (1878) driest 25'5" (1863) Rainy days in wettest year (12 yrs.) 101 (1875) driest 61 (1876) CLIMATIC TABLES 329 - 77. MADRAS Elevation 22 feet TEMPERATURE. si p P RAINFALL. BAROMETER. p o H 3 o M. Range. II 1-3 1 Month. I 1 3 1 1 i! January 76 85 68 17 26 72 4-1 1-0 3 29'99 12 February 77 87 68 19 28 71 2-8 0-3 1 97 13 March . 81 90 72 18 29 73 2-5 0-4 1 91 14 April . May . 85 87 93 98 77 81 16 17 27 32 72 67 2-9 3-9 0-6 2-2 1 3 82 74 13 13 June 88 99 81 18 31 61 6-5 21 10 70 12 July . 86 97 79 18 28 64 7-1 3-8 14 72 12 August . 85 95 77 18 27 69 6-4 4-4 14 75 13 September . 84 94 77 17 28 70 6-2 47 11 77 13 October 81 89 75 14 25 77 6-2 10-8 14 84 13 November 78 85 72 13 25 79 6-3 13-7 14 92 12 December 76 83 70 13 23 77 5-4 5-1 9 98 11 Year 82 71 5-0 49-1 95 29-84 Mean highest temperature of year 108 ,, lowest 60 annual range of temperature 48 Highest recorded reading(27 yrs.)112'9 (1880) Lowest 57-6 (1876) Absolute range of temperature . 55 '3 Rainfall of wettest year (74 yrs.) 88-4" (1827) driest 18 "5" (1832) Rainy days in wettest year (26 yrs.) 119 (1847) .. driest 73 (1876) , 78. TRICHINOPOLY Elevation 275 feet TEMPERATURE. g p RAINFALL. BAROMETER. n 1 M. Range. o t>, o> 1 id ss II >f i i d i p h- 1 | i ?> 3 i fi i 3 a January February . March . 76 78 83 87 92 98 67 68 73 20 24 25 30 34 36 68 62 58 4-2 2-9 3-1 i-o 0-5 0-7 1 "i 29-71 70 66 14 16 16 April . May . 87 88 101 102 78 78 23 24 33 33 54 56 4-4 5-6 1-8 3-8 i 5 57 52 16 15 June 86 99 78 21 29 57 7'4 1-3 3 50 12 July . 85 98 78 20 27 57 7'7 2-2 3 51 12 August . September October 84 83 81 96 95 91 76 75 74 20 20 17 28 27 26 61 63 73 7-8 7-4 7'8 4-4 5-3 7-8 7 8 11 53 55 59 13 14 14 November 79 87 71 16 26 76 7'4 5-2 9 65 13 December 76 85 69 16 27 74 5-9 3-1 6 70 13 Year 82 63 6-0 37-1 55 29-60 Mean highest temperature of year 106 lowest 60 ,. annual range of temperature 46 Highest recorded reading (12 yrs. ) 107 '9 (1876, 1881) Lowest recorded reading (12 yrs.) 56 -6 (1884) Absolute range of temperature . 51 -3 Rainfall of wettest year (33 yrs.) 95'3" (1863) driest 17 -1" (1845) Rainy days in wettest year (11 years) 70 (1877) ,, driest 31(1876) 330 APPENDIX I 79. COIMBATORE Elevation 1348 feet TEMPERATURE. . Q p RAINFALL. BAROMETER. t> W s o M. Range. 1 11 Is i Month. 1 1 fl 1 1 i! January 74 87 64 23 34 59 3-4 0-3 1 28-61 15 February 77 92 65 27 37 52 2'4 o-i 1 58 17 March . 81 97 70 27 36 54 2-9 0-6 3 55 17 April . 83 98 74 24 32 58 4'2 1-8 4 49 16 May . 81 95 74 21 32 65 5-4 2-6 10 45 14 June 78 90 72 18 28 71 6'6 1-8 9 44 10 July . 77 89 71 18 25 72 6'5 1-3 9 46 09 August 77 88 71 17 25 73 6'1 1-2 8 47 11 September 77 90 71 19 26 71 5-7 1-2 6 49 13 October 77 88 71 17 26 75 6-3 57 15 51 15 November 76 86 69 17 27 74 6-1 3-4 13 56 14 December 74 85 66 19 31 68 4-9 1-1 6 59 14 Year 78 66 5-0 21-1 85 28-52 Mean highest temperature of year 102 lowest 56 annual range of temperature 46 Highestrecordedreading(llyrs.) 103'8 (1879) Lowest 54 (1876) Absolute range of temperature . 49 '8 Rainfall of wettest year (24 yrs.) 29-0" (1880) driest 12 -6" (1867) Rainy days in wettest year (11 years) 118 (1882) driest 55(1876) 80. MADURA Elevation 448 feet TEMPERATURE. * | RAINFALL. BAROMETER. B 3 M. Range." 1 1 S .s 4 q i I f> S ss ^S 'e 1 1 1 M Q PH ft g S s January 77 88 68 20 29 67 4-2 07 i 29-50 16 February 79 93 68 25 33 61 3-9 0-4 i 48 17 March . 83 97 72 25 35 60 4-0 0'6 i 44 17 April . * 86 101 76 25 33 59 4-6 2'0 2 37 17 May . 86 101 76 25 33 61 5-1 2'8 5 32 15 June 85 99 76 23 30 59 5'5 1-6 3 31 13 July . 84 99 76 23 30 59 5-3 1-7 3 33 12 August . 84 97 75 22 30 63 5-3 4-7 7 34 13 September . 83 97 75 22 29 63 5-0 4-5 8 36 14 October 81 92 73 19 27 73 5-4 8-7 14 40 14 November 79 87 72 15 25 75 5-6 5-1 9 44 15 December 77 86 70 16 25 74 5-1 2-2 5 48 15 Year . 82 65 4-9 35-0 59 29-40 Mean highest temperature of year 105 ,, lowest 62 annual range of temperature 43 Highest recorded reading (12yrs.) 106 '7 (1880) Lowest 59-8 (1876) Absolute range of temperature . 46 -9 Rainfall of wettest year (24 yrs.) 54-0" * (1877) driest ,, 20-7" (1876) Rainy days in wettest year (11 yrs.) 75 (1885) ,, ,, driest ,, 42 (1876) 1 A much higher rainfall, viz. 73*0 inches, is given in the tables for 1853, but as a register kept in the^suburb of Pasumalai by the missionaries gives only 31 inches in the same year I cannot accept the former as valid. CLIMATIC TABLES 81. TRINCOMALEE Elevation 175 feet TEMPERATURE. s g RAINFALL. BAROMETER. p W 3 M. Range. o a 8 * 11 gri ' "Is 5 i a 8 of rf >> a 1 > 2| ft a 3 January 78 83 74 9 15 77 5-8 6'2 10 29-88 10 February 80 85 75 10 17 74 4-5 2-4 3 87 11 March . 82 88 75 13 20 75 4-1 1-3 4 84 12 April . 85 92 77 15 23 70 4-2 1-6 4 77 12 May 85 94 77 17 26 67 5-1 2'2 6 71 11 June 85 93 76 17 25 67 6-2 1-9 4 70 09 July . 85 93 76 17 24 65 57 2'2 2 71 10 August . 84 93 75 18 25 68 5-7 4-2 9 72 10 September . 83 92 75 17 24 69 57 4-6 7 74 12 October 81 89 75 14 23 74 6-0 8'9 16 77 12 November 79 85 74 11 18 82 6-2 13-1 19 82 11 December 78 83 74 9 15 81 6-4 13-2 20 85 11 Year 82 72 5-5 61-8 104 2978 Mean highest temperature of year 99 ,, lowest 68 ,, annual range of temperature 31 Highest recorded reading (11 yrs.) 101 '4 (1886) Lowest 65 -3 (1876) Absolute range of temperature . 36-1 Rainfall of wettest year (16 yrs.) 78'5" (1883) driest 48 '5" (1879) Rainy days in wettest year (12 yrs.) 128 (1879) driest 78 (1884) ' 82. KANDY Elevation 1696 feet. TEMPERATURE. g ej RAINFALL. BAROMETER. p M 3 M Range. I 3 c ^ ^ . . * 1 i tl * 1 o a 1 g fi^ January 74 80 69 11 21 76 5-1 5-0 10 28-22 10 February 76 82 68 14 24 71 4-3 2-6 4 21 12 March . 79 86 71 15 23 69 4-8 3'1 9 20 12 April . 79 87 71 16 22 74 5-3 6-9 14 16 11 May . 79 86 73 13 20 76 5-7 67 13 14 09 June 77 84 71 13 17 81 6-8 9-1 21 15 08 July . 76 80 71 9 16 80 6-5 7'8 21 16 08 August 76 81 71 10 19 79 6-5 6'2 19 16 09 September . October 76 76 83 82 70 .69 13 13 21 21 79 80 6-3 6-4 5-8 11-3 16 23 18 18 10 10 November 75 82 68 14 22 80 5'9 11-6 18 19 10 December 75 81 68 13 23 80 5'9 8'4 17 20 10 Year 76 77 5'8 84-5 185 28-18 Mean highest temperature of year 90 ,, lowest ,,60 annual range of temperature 30 Highest recorded reading (5 yrs.i) 94-6 (1886) Lowest 54 (1886) Absolute range of temperature . 40*6 Rainfall of wettest year (17 yrs.) Ill -I" (1882) driest 67" (1884) Rainy days in wettest year (12 years) 213 (1877) driest 160(1884) i From 1876 to 1882 the maxima and minima temperatures recorded at Kandy are evidently untrustworthy, owing apparently to want of proper exposure. These have therefore been rejected. According to Sir Emerson Tennant's description, the diurnal range of temperature in the cooler months is much greater than that shown by the above table. 332 APPENDIX I v 83. COLOMBO Elevation 40 feet TEMPERATURE. s Q b RAINFALL. BAROMETER. P s M. Range. | u 1 *&& ll >* ^ | 1 3 a 1 it Q 1 1 s January 79 86 72 14 21 74 4-9 3-0 6 29-88 12 February 80 87 73 14 22 73 4-2 17 4 88 13 March . 82 88 75 13 18 76 47 5-5 9 87 13 April . 83 89 77 12 18 77 5'6 8'8 13 82 12 May. . . 83 87 78 9 18 79 6-8 13-2 20 81 10 June 82 85 78 7 14 81 7'6 8-2 17 82 08 July . . 81 85 77 8 14 80 6-8 5-5 12 83 08 August 81 84 77 7 13 80 7-0 4-5 13 84 09 September . 81 85 77 8 13 80 6-9 4-9 14 86 11 October 81 85 76 9 15 80 6-8 12-9 21 86 11 November 80 85 74 11 16 79 6-2 127 17 86 11 December 80 85 73 12 19 76 5-9 6-4 13 87 12 Year 81 78 6-1 87-3 159 29-85 Mean highest temperature of year 93 j Absolute range of temperature . 30 lowest ,, 68 Rainfall of wettest year (17 yrs.) 139-7" (1878) annual range of temperature 25 driest ,, ,, 57'0"(1874) Highest recorded reading (12 yrs.)95-S (1885) j Rainy days in wettest year (12 years) 195 (1885) Lowest 65-8 (1880) I driest ,, 125(1876) 84. AKYAB Elevation 20 feet TEMPERATURE. 1 d t> RAINFALL. BAROMETER. M. Range. 1 II li 1 1 | I 1 1 I || a o A January- 69 82 59 23 34 72 1-6 O'l ... 30-00 11 February . 73 85 60 25 37 70 1-1 0-2 ... 29-95 12 March 79 89 68 21 34 73 1-9 0-5 1 90 12 April . 84 92 75 17 26 74 3-0 1-6 2 83 12 May . . 84 91 77 14 25 78 5'3 12-2 14 76 10 June . 82 86 77 9 18 87 7-0 51-6 27 68 09 July . 81 84 77 7 15 89 8-5 51'0 29 67 08 August 81 85 77 8 15 89 8-0 38-6 28 71 09 September . 82 87 77 10 16 86 6-9 23-0 22 76 11 October 81 88 76 12 19 83 4-8 12-4 12 85 11 November . 78 85 71 14 25 81 3'2 3-9 3 94 10 December . 72 82 64 18 30 77 2-5 0-6 2 99 11 Year . 79 80 4-5 1957 140 29-84 Mean highest temperature of year 96 ,, lowest ,, 51 annual range of temperature 45 Highest recorded reading(12years)99'l (1882) Lowest 47-3 (1878) Absolute range of temperature . 51 -8 Rainfall of wettest year (28 yrs.) 254-1" (1850) driest 142" (1874) Rainy days in wettest year (17 years) 156 (1879) driest 118(1869) CLIMATIC TABLES 333 ^ 85. MANDALAY Elevation 250 (?) feet TEMPERATURE. s Mean. CLOUD. RAINFALL. BAROMETER. 1 II ll M. Range. 22 i i , 1 a o s January February . March April . May . June . July . August September . October . November . December . Year 71 74 83 91 90 86 84 82 83 [81] 75 72 81? 84 90 98 102 101 94 93 92 90 59 60 69 81 82 79 79 78 78 25 30 29 21 19 15 14 14 12 33 42 43 35 34 28 27 21 22 64 54 43 49 53 66 67 73 77 [76] 75 69 64? 8'4 91 8'6 37 1-4 : 8 4-6* 5-1* 2-9* 2-4* 4-7* 4-2 2-6 0-2 27-5 "4 12* 11* 10* 10* 10* 6 7 2 72 2975 70 59 50 48 45 44 48 50 [29-61] 72 76 29-58? 15 16 17 19 17 14 14 13 14 12 14 84 84 68 62 16 22 26 30 Highest recorded temperature 106-1 (1879) Lowest 52-9 (1879) Annual range 53-2 N.B. The values for October in brackets are interpolated. * For the five months May to September, the rainfall data are the means of the two years 1879 and 1886. 86. THYET MYO Elevation 134 feet TEMPERATURE. s o RAINFALL. BAROMETER. p a i M. Range. 5 * d B* & s r^H* tJD 8 cS 8= "3 i 1 a i i i 3| Q 1 . s s January 68 85 54 31 46 68 1-8 29-85 14 February 73 92 56 36 48 58 1-0 ... 79 15 March . 82 100 66 34 48 56 1-2 o-i 73 16 April . 87 103 75 28 38 56 2'5 0-7 "2 64 16 May . 86 99 78 21 34 67 4-8 5-3 10 61 14 June . 81 90 76 14 23 82 6-4 7-9 20 58 10 July . 80 89 76 13 19 84 6-6 8-0 22 57 11 August 80 89 76 13 18 84 6-3 8-5 20 59 11 September . 81 90 76 14 20 82 5-9 7-8 18 65 12 October 81 90 75 15 22 80 5-3 4-9 10 73 12 November . 76 87 69 18 31 78 3-8 2-3 4 80 12 December . 71 84 60 24 36 72 3-0 1 85 13 Year 79 72 4-0 45-5 107 2970 Mean highest temperature of year 108 lowest ,, ,, 45 ,, annual range of temperature 63 Highest recorded reading (7 yrs.) 110-2 (1881) Lowest ,, 40-3 (1878) Absolute range of temperature . 69*9 Rainfall of wettest year (17 years) 62-6" (1877) driest ,, ,, 33 -6" (1884) Rainy days in wettest year (9 years) 133 (1879) driest 106(1878) 334 APPENDIX I 87. TOUNGHOO Elevation 181 feet TEMPERATURE. g ft p RAINFALL. BAROMETER. W s M. Range. O d 8 I d |.s j . c 03 >, 5 >> ' i ss f 1 i 1 8 S sj January 70 86 57 29 39 70 0-9 29-83 13 February 74 91 59 32 43 64 0-6 0-2 ... 76 14 March 80 98 67 31 44 61 0-5 70 16 April . 85 100 74 26 35 62 1-8 1-5 "5 63 17 May . 83 97 76 21 32 70 4-7 6-6 12 59 14 June . . i 79 88 75 13 21 84 77 13-4 24 57 10 July .. . 78 87 75 12 19 87 8-0 17-5 26 56 10 August 78 87 75 12 19 88 8-2 18-1 27 58 10 September . 80 89 75 14 19 85 67 11-8 22 62 12 October 80 90 75 15 20 82 4-3 7-4 14 69 12 November . 77 87 71 16 28 80 2-8 1-4 4 76 11 December . 72 85 63 22 32 78 2-1 0-2 1 80 11 Year 78 76 4-0 78-1 135 29-67 Mean highest temperature of year 104 lowest ,, 51 ,, annual range of temperature 53 Highest recorded reading (9 yrs.) 106'5 (1881) Lowest ,, ,, 47'1 (1883, 1884) Absolute range of temperature . 59 '4 Rainfall of wettest year (17 yrs.) 101 -3" (1871) ,, driest ,, 61" (1873, 1874) Rainy days in wettest year (9 yrs.) 154 (1884) ,, driest 91(1878) 88. RANGOON Elevation 41 feet TEMPERATURE. 53 I RAINFALL. BAROMETER. M. Range. o 1 II W f 1 d d a 1 1 If 1 1 8 January 75 88 64 24 35 66 2'4 0-2 1 29-96 13 February . 77 93 65 28 - 38 62 2'2 0'] 1 91 14 March . 81 97 71 26 37 64 2'2 o-i 87 15 April . 84 98 76 22 31 68 3'6 1-8 'i 80 15 May . 83 93 77 16 27 76 6'5 10-9 16 75 12 June 79 86 77 9 18 88 8'8 18-4 28 73 09 July . 78 85 76 9 16 90 9-0 21-3 29 73 08 August . 78 85 76 9 16 91 9-0 18.6 27 75 09 September . 78 85 76 9 16 89 8-3 16-0 26 79 11 October 80 88 76 12 18 85 6-2 8-1 14 84 12 November 78 87 73 14 24 80 4-5 3-4 6 89 11 December 76 87 68 19 29 74 3-2 o-i 1 94 12 Year 79 78 5-5 99-0 153 29-83 Mean highest temperature of year 104 ,, lowest 58 annual range of temperature 46 Highest recorded reading (11 yrs.)106'7 (1877) Lowest . 55-7 (1883) Absolute range of temperature . 51 Rainfall of wettest year (17 yrs.) 143-4" (1871) driest ,, ,, 69'1"(1S74) Rainy days in wettest year (10 yrs.) 161 (1880, 1882) Rainy days in driest year (10 yrs.).132 (1878) CLIMATIC TABLES 335 ). MOULMEIN Elevation 94 feet TEMPERATURE. g Q RAINFALL. BAROMETER. ffi 3 M. Range. 1 g M gd ^ | rt . i i 1 >?& '? 5 S^ S^ i i 3 I ~ H S fitf 1 M m January 75 89 64 25 34 63 0-8 29-89 12 February 77 92 66 26 35 63 0-7 o-'i 84 12 March . 81 94 72 22 31 66 0-6 o-i ... 81 13 April . 83 95 76 19 27 69 2-4 3-0 7 75 13 May . 82 91 76 15 24 76 5'0 19-7 18 70 12 June 78 84 75 9 18 89 8'2 38-4 28 68 09 July . 77 83 74 9 15 90 8'3 43'9 29 68 08 August . 77 83 74 9 18 91 8-2 43-0 28 69 09 September . 78 84 74 10 18 87 6-3 30-3 27 72 11 October 80 88 74 14 20 81 3-2 8-4 15 76 12 November 78 88 72 16 26 74 17 1-5 5 80 11 December 76 87 67 20 30 70 1-3 o-i 1 85 11 Year 78 77 3-9 188-5 158 2977 Mean highest temperature of year 99 lowest 58 ,, annual range of temperature 41 Highest recorded reading (10 yrs.) 101 -5 (1878) Lowest 54-8 (1883) Absolute range of temperature . 46'7 Rainfall of wettest year (37 yrs.) 279" (1865) driest ., 134'1"1 (1878) Rainy days in wettest year (10 yrs.) 175 (1879) ,, driest ,, ,, 135(1878) i Only 124-2 inches was recorded in 1850, and 124*3 inches in 1852 ; but up to 1865 no rain whatever was registered in the live dry months November to March, and the registers are probably incomplete. 90. MEKGUI Elevation 96 feet TEMPERATURE. g - g RAINFALL. BAROMETER. p n 3 M. Range. 1 11 gd 8 ll i 1 a 03 S 1 II January 77 89 68 21 30 73 27 0-5 2 29-87 11 February . 79 90 71 19 26 75 31 1-4 3 86 12 March 80 92 73 19 25 77 3-8 2-3 6 83 12 April . 81 92 74 18 24 79 5-4 57 8 78 12 May . 80 90 75 15 23 83 7-3 15-9 16 75 10 June . 77 85 73 12 20 89 91 29-8 25 74 09 July . 76 84 73 11 18 89 9-0 31-6 28 74 08 August 77 84 73 11 18 90 8-7 281 26 75 09 September . 77 85 73 12 17 90 8-7 26-3 25 78 10 October 78 87 73 14 19 86 7'3 13-5 20 80 11 November . 77 87 71 16 24 80 5-3 3-5 7 82 11 December . 76 88 69 19 27 74 2-8 0-5 2 85 10 Year . 78 82 61 1591 168 29-80 Mean highest temperature of year 97 lowest ,, 61 ,, annual range of temperature 36 Highest recorded reading (9 years) 98-8 (1878) Lowest 57 (1882) Absolute range of temperature . 41 '8 Rainfall of wettest year (22 years) 213" (1882) ,, driest ,, 132" (1875) Rainy days in wettest year (9 years) 188 (1878) driest 149(1885) 336 APPENDIX I 91. PORT BLAIR Elevation 61 feet TEMPERATURE. g p hH i RAINFALL. BAROMETER. M. Range. o . 1 CH ij II 1* >f '3 1 1 H g rf a >s 1 it Q $ 8 a January 79 86 75 11 18 77- 3-4 0-9 2 29-90 10 February 80 88 74 14 20 78 2-8 1-3 2 89 11 March 82 91 76 15 21 78 2-9 0-4 1 87 11 April . 83 92 78 14 20 78 4-1 2'4 5 80 11 May . . 81 89 78 11 19 84 5-0 15-9 20 75 09 June . 81 86 78 8 16 86 7'0 17-9 25 73 08 July . . 80 86 77 9 15 86 7-1 16-5 26 74 07 August 80 85 77 8 15 87 6'9 15-2 25 75 08 September . 79 85 76 9 15 88 6-9 19-6 25 78 09 October 80 87 77 10 15 87 5'8 11-8 22 81 10 November . 80 87 77 10 16 84 5-0 9-5 15 84 10 December . 79 86 76 10 18 81 4-5 5-3 9 88 10 Year . 80 83 5-1 1167 177 29-81 Mean highest temperature of year 95 lowest 69 annual range of temperature 26 Highest recorded reading (12 yrs.) 96'4 (1878, 1885) Lowest recorded reading (12 yrs.) 66 (1875) Absolute range of temperature . 30 '4 Rainfall of wettest year (19 yrs.) 149-8" (1S69) driest lOO'l" (1871) Rainy days in wettest year (14 yrs.) 206 (1882) driest ,, 158 (1885) 92. NANCO WRY Elevation 81 feet TEMPERATURE. d RAINFALL. BAROMETER. B 2 M. Range. Ss* l-s . ro 3 ~> m * 8 c 1 1 j m cS C Q a January 80 86 77 9 13 75 5'2 2'9 8 29-86 09 February . 81 87 78 9 15 74 47 1-3 4 86 10 March 82 89 78 11 15 74 4'6 2-1 5 84 10 April . 83 89 79 10 17 76 5'3 5-3 9 80 10 May . . 81 88 79 9 17 79 6'6 12-1 18 77 09 June . 80 87 78 9 16 82 7-3 127 22 76 08 July . 80 86 77 9 16 82 6-9 12-6 20 77 07 August 79 86 77 9 16 82 6-9 12-6 19 78 08 September . 79 87 77 10 16 81 6-9 10-6 19 80 09 October 79 86 76 10 17 84 6-8 13'5 21 81 09 November . 79 86 76 10 16 82 6-5 11-8 18 82 09 December . 79 86 76 10 15 79 5-8 12-2 15 84 09 Year . 80 79 6-1 1097 178 29-81 Mean highest temperature of year 94 lowest ,, ,, 72 annual range of temperature 22 Highest recorded reading (10 yrs.) 99-5 (1884) Lowest , (12 yrs.) 70'2 (1875) Absolute range of temperature . 29 '3 Rainfall of wettest year (13 yrs.) 141 -3" (1882) driest 79'3" (1877) Rainy days in wettest year (12 yrs.) 227 (1879) driest ,, 155(1874) STORMS 337 OiOOOOOiOiOOO OOOiOOOiOOCC5OOo ' 1 1 1 II If OS COCOOOOOOOOOO5OOOOO5 OOOOOOOOOOOOOOOO C r I C3 OS r-H O ooooooooosO3 _ ooojqs o o os os, OOOO OOOOCX)OOO50i OOOiOO O OS CO 00 OS OO OS 00 o I I I I O OS OOOOOiOO (Mo o rH 1 rH i-H | r-H(M(M(N(No o (Ni-H | | | | | 00 i-H 1 1 1 00 w ^ o _ i-H(MOSOiO5O5COOOOSO5COGOGOOiCOo O5OOCOOSOiCOO5COCOCO 3 o 1 o' o 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 ego 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 o 1 H O CQOJOSGOOSCOCOCOOSOIGOGOCOOOCO COOOGOC5O5GOO5GQGOOO C.OO^OO?--kOG CO CO OO CO rH i I rH . ... Z-, ..',-, li ^Q ^ O O OS CO Srt* ScOCOO5C75O C^fvi-^^Sco^QO.^-^ ^^r-l .<*>*>. 1^X>.t^J>.J>.CO .CO5QCOOOgcOCO Ilt ^ OO Oi rH llililiilliillliiifiilllilU Q STORMS 34i M ^_GrJCO^S M 4^ 1-1*5 315111 ^ Miles per Hour. (N XC<^ * o COO CO 10 o to : o : J-H o : o o : oo : O 00 rHOO t^o 10 CO -*co , 00 t CNrHN OilOCO COT* 00 *^0 ll H o CM : rH co oj b 1 o> 1 o >> t^ rH 3 S 1 O5 rH 1 3 W e I 1O , CO OO CO (N t^ TH O . : t-co- 1>- T 1 r-l i 1 i 1 CO i 1 oos CO(N ^CO t-t^ 00 rH^ (NOS 0000 r-l** : o^ : orn : rno : (Nt*. : rn^: co^: r^oo: oo co ^(M OCO CO(N OO CN l> ^O C^l XO Oi "^ 3 o co : o - : rnco : O t^ OS CO OS OO r 1 O B r-i o : o co : o ^n : I-H (N : i i TH o : oco : r-i t : 1-1 co : ,0 OS O !>. r-l CO CO O OO - rH CO 00^* i-H t^ o co : o co : o (N : CM o o i i o : i i os : i i o : i i ^ : OS CO ^, ^ ^ ^ ^ ^ CO. : 1-1 o : Oo : o co : rH CO I o >o : 4* CO O CO -* CN rH OO 1O -* rH 10 OS OS rH CO OS CO & r*.s : -* CO O -*CN : - CN os : OS rH CN CN OS 000 I l-H "* 00 os co O iO JO -J* CO O *P OS oo o : . a 1-3 O CO CN rH CN OS CO CN CN s* OO^rH CO rH O O t-^ CN CO 10 00 CO > o oo * OS CO CO CO CO TJH IO CO 1O rH CO IO O rH - CO IO rH OS CN ** rH 1 i-H 10 : rH I-* : rH t^ ; o -* : o co : CN os : rH J>. : X CO I o CN : o - CO * * CO O CO jfc^j^ CO J>. CNrH 1 rH-* : rH co : rH OS : rH CO : o CN : CN co : CN rH ; o ce.2 III III III III III 3 * a co ci .a Hoslliarpur, 30 years . a rH CO o CN I ^3 & Umballa, 35 to 37 years CO co o 3 P CO CO o CO QD Mussoorie, 23 to 35 yrs. Delira Dun, 37 to 39 yrs. MONTHLY RAINFALL TABLES 345 . 10 CO O T^H (^ CO 00 <& CM CO TH - a 1-3 CO rH O rH rH CO rH co co : XO 1>- O 7 1 co t : CN oo : xo xo : i i c3 O rH CN CO CO O O xp co oo : CN CN oo -^ o rj< : O xp OS J>- o CN : o 4n : o AM : j CO O CO O -* CN "* rH CO rH CO t CN 1O rH OS 1 rH^ : rHCO : o CM : o CN : o CN : o I-H : o rH : o o : s CN CN U<* O^H t-xo 00 XO t CO - I o rj< : o co : o co : o . rHt^ rH XO 1 o I-H : o xo : : o : o ^H : o I-H : o o : o o : (M XO CO CO CO I-H O I-H rH I-H XO CM XO * rH CO I CM rH : o <*< : co t^ : I-H - t>. rH CM 1^ CO CO rH t^ XO OS rH XO CO rH !>. Tfl OQ * i-H O rH CO 00 O I 1 XO CO O COrHO CO CO O I-H rn^ : CO T* O In 00 CO CO CO CO CO XO CO CM OO O i-H -t^ rH CO t^ CO rH rH OO 1 00 coo OS CO O O CO O rH (N 14 *O ^f rH CM O t^ I-H rH (M CM CO O co o : . t^ OS l^ OS OS O CO xo xo CO CO XO OO xo J>- OS CO i-H O xo OO rH (N CO rH O rH CO CO t^- CO S$* co t^ o OS I-H O CM O (N 3 CO I I s MONTHLY RAINFALL TABLES 347 CO CO rH CO CN CO -^ CN - < VO a rH CO rH OS rH 05 8 03 cot- 1 o o : : o : : o : & M O o rH : o o : O rH I " o 7-H : J T^I o CN rH 05 CN OS rH t^ CO CO t 3 o vo : : o : o i-i : o o : O r-l '. : o : o CN : CO CO CN t~* ^ OS CO rH CXI CO CO OS vo ^ o CN : o rH : o CN : o CN : o rH : o CN : o CN : CN CO rH CN rH CO CO rH^ T}1 VO t-t- c3 o I-H : o CN : O rH : o I-H : o o : o CN : o co : ls.a III III la. a 1 Is IrlS III lls p CN o g CO 7-H 3 i E CN 4 VO 1 Neemuch, 19 to 20 yrs. 2 o OS i I E c3 tt) O CO 1 348 APPENDIX III CN O CN r i-l O5 .CO CO (N 10 -* CO O rH O CO M p < 0,00, -* QO < i-l (N s^^ CO CO CO . CO CO W5 O5 (N O GO rH XO t^ CO O OS rH 05 CO CO CO OS rH CO CO ' 1 38= ^ K1 -^ rH , . CO t^ t>- o rH -^ OO - CO (N O rH t^ Tj< * o ^ o CO CO t^ CO >-S o cs .S ESS lls gss ld.s s^ is.s ^ss g^.s s^g IS.s sss i^.s ^^^ x n iS OT oJ o >> o (M o ^~t CO rH O t^ >t A O >> t^ (N o i a 1 & Miclnapore, 19 to 22 yrs. MONTHLY RAINFALL TABLES 349 cj rH OO CO O CN r-t CO IO IO CO OS CO CO O n o o : o co : o CN : o oo : o j>- : o co : o CN : . COi- CO CO >- CO CO ** O 10 "HH CO CO CO i o CN : o 10 : o 10 : rH IO : rH CO I rH CO I rH t^ : CO O -HH CO CN 10 CN t-t-10 OO CN CO rH OS ^ CN ^H CN o O iO t^ : ioco : lOrHO IO O O 10 CO O COt^CN 10 CO rH . GO O 1^ O CO O CN t^t^ i 1 CO IO CO CO O CO CO ^H t^ t^ CN i OS OS CN rH 00-* O^MCO ss 10 . OS OS IO rH^ rH CN "co 1 CO IO O O CN CO rH CN OS OS CO CO CO CO CN AH 10 rH CN 10 O CO SS"' CN^lO rH^ O O CN 00 O OO rH CO O OS O CO 10 OS rH CO CO xh CO O rH O OO rH rH CN ^H CO CO CM CN t>- IO O CO i 1 rH p hj O OS CN CO O IO rH CN CNC010 83*" rH CN OS (N O rH CO rH IO rH t^ rH CO x>- t^ o OO CO O CO CO IO 00 CO t^ t^ rH CO CO CO CO rH CO CO p OS rH CO CN rH CO CO rH CN CO OO ^ rH CN CO OS IO CN ^ O CO rH CO 10 CO TH CO s* . OO O t^ CO CN rH CN t^ rH CN CO CO CN IO CO -*CO^ CO 03 ^ rH TH 10 CO O rH OS !>.->* r-t OS rH rH CO OO O CO CN IO g rH CO rH OS ^H rH CN OS OS CO CO OO CN IO CO IO OO i - to OS IO IO 1 rHt^ : cNt- : rH "HH o : CN rH XO I OS CO IO OS OS CO ,cj O O CO rH IO IO OS IO rH rH t^ CO CO # 00 1 rH x* : rHCO : rH rH t^ : rH CO : iO CN rH r 1 ^ oo : 1 1 AH o : O CO AH CO ! O rH AH cb i CN CN rH CO I CO **! o co : IO IO CO CN IO O CN CN CO CN 4* TtH CN ^OS CO l>- ^^ * CN rH CN o co : o CN : O rH I o 10 : o co : rH CO I rH TjH : 1 g.S III III 1 Is s ^ . lls III -HH CO rH CO r a 2 Jr Calcutta, 58 years I CO IO CO p Chittagong, 28 to 32 yrs. Cuttack, 25 to 30 years Dibrugarh, 23 to 26 yrs. Sibsagar, 30 to 32 years 350 APPENDIX III CO CO CN US ^ ^ t-DO CO CO t>- l-H CO 05 o rH : O T-1 I o T-H : o i>. : o - o o us : : o : I-H t : T-H TJ< : I-H co : o co : o us : ,-H CO CO US rH CN CO CO -* T* US CO ^ OO CO CO US CO i co oo : T-H x* US O CO O O CO (M l-H CO O O I-H TJ< : I-H us : T-H CO l-H TH (M T* ^ US CO CN (M O5 O CO O5 CO CO C5 CO O CO i OO OO O T-H CN O5 CO SS 10 rH CO "* I-H t^ US CN "* t^ t^ o CO CO O | CN 00 US -* 00 CO rt* CO CO (M CO O CO O5 TH CN TjH O5 OO rH Tfl 9 l-H O5 -* T-H T 1 l-H Ttf CO l-H (N * O US rH CO CO O5 OO T-H (N CO ^ rH i-H rH rH (N CO CO O rH CO t^ CO US ^^^ CO CO CO (M CO CO CO O OO 00 O CO CO 00 i-s CN CO Tt< i-H CN US CO T* rH CO TH O5 "tf I-H CN O (M O5 CM US US CO O5 CN Tj< **- CO CO US rH CO flj O5 CN t^ TJH rH CO O O CO rH t^ O rH US o co : O ^fri '. J OO Tfl TH rHCN-* t^ OO OO O *> T 1 OrH rH PH HH CN CO CN US I US CO CN rH US rH (N CO US O CO <* ** I-H CO * OO CM rH CN a OrH i K o co : j US CO 05 US OrH O5 i-H TUt^ H US O5 1 - OJ>- OO CO rH CO 1O CO 1>- O5 CO i O 05 rH rH CM iO 05 CM rH CO *"rH 1^ CO xo CM : o T-H : o co : I O5 O O CM CO CM rH t^ CO O5 IO OO t^ "^ O ^f O5 CO CO to 3 CM rH CO rH CO 00 05 rH CO CM CO CM rH CM CO O5 rH rH CO ^ rH rH {>. ; rH rH rH '. rH 1C CO rH 00 OO O 00 O5 O CO CO CO CO IO CO TH **! rH p XO rH CM rH CO 00 ^H CM CM kO rH T^ OO IO rH CO CO^hrH ^CN 1 ^ ^ : co co : rH . 10 OJ rH 00 IO O * CO O5 CO CM OO CM OO I-H t^ CM CN 1 CO OO O I-H O ^M CO ScM*" OO CM CM CM 10 o : o o : o CN : . o 10 C005 . O5 t- 0005 _ CM CO rH CM rH CM 1 o co : O rH : o CM : o \o : o o : ^ o T-H : S3 a 3 o I-H : J CO CO CO CO CO 1O O5 | ^^ o ^^ CM OO 1 : o : o CM : o co : o co : o rH : " o rH : o Tf : J CM O ^HCN CO CO CO CO Ttl O CO CO CMO . 1 o co : o CM : o co : o co : o CM : o CN : O rH ; . rH rH CO CO CO CO ^0 rH CD CM 05 CO CO g o rH : o rH : o I-H : o co : o rH : o I-H : O rH I CO CO 10 o CO IO CO 00 10 O (NO CO * 1 o CM : o Tf : o co : o "* : o co : o rH : o CM : III III fl . O C3 .H III S S.S* J5; 1 Id Hoshungabad, 28 to 33 years . Pachmarlli, 16 years . 2 g o CM CN 1 1 o O5 CO 1 1 1 >* to CO *** CM ^0 O <^ IO "^ CN J 4 1 Z "^ i-a Kurrachee, 31 years . 352 APPENDIX III CN to rH CO rH -^1 CN rHrH CN O TH !>. : o : : o : o o : o o : : rH : o TH : o CM : o TH : > TH r rH Ol CN rH i 1 O CO 1C O TH T-i CN i o o : o o : o TH : rH I : o : o o : O rH I rH t>. I i _ i CO t~- CO 01 rH CO CO (^ CO TH 00 CO 00 TH o : o : o o : o TH : rH T}< : co o : TH TH o : TH 01 : io co : 00 rH CO i^ CN CO rH CO CO Tj< CO CN Ol CO CO CO CO 10 CO Ol TH CM I o TH : co to : i i TH CO O ^coo CN CN CN rH CO O rH TH TH CO CO t^ CO CN TH CN CO CN CO t^ rH 1o : CO 05 O rH Ol t^ rH GS> O 1C CO CN TH J>- CO TH CO O O rH TH CO CN CO CO CO CO CO CO rH CO (N CO 1C CO O CO CO TH O - TH rH t^. IO 1-5 CM - Pi 3 Ha o co : CN CM : 000 ^fr^ TH iO Ol CN rH O Ol CO CN TH CO TH CN CN 10 rH t^ O 01 T* rH i i T-l O CN TH T*f CO ^9 . TH J>. O t-~ O CN TH t^ 1 o o : o T-H : o * : o CM : o co : o o : o co : TH t^ : t (N TH I-l O . TH rH CN TH TH TH t^ o r-i : o rH : o o : : o : : o : o CN : o o : o t^ : Us ^gs 3 K" a o ce .3 ^SS la.s sss Is'-d ^^^3 S r^ S !ls ^S^ f&a s^^ la.s ^s Us sss ^H * o i ! 03 i t>j -* (N 5 CO I 1 -l-T QQ i >> o CO 3 Ol 1 1 cf a 1 ' g i !>> o t^ 3 Ol CO I >* CO 1 1 Mahableshwar, 31 to 32 years . . f . MONTHLY RAINFALL TABLES 353 cj T-H CO Tfl CM CMt^ <*0 CO OS CO "^H * OS fi o T-H : o co : orH : O^H : o CM : o co : o CM : T-H OO 10*. OO T-H t > * t> CM^ OO "* -rH CM coo o m co T* co : : 000 1 i . CO i-H 00 cot- CO OS O CO * !>. CO i-H CM OS rH O CO 0< 1 O OS T-H CO 1 1 TJ< o : t>. o o rH CM CO OS O in co : rH CO OS O a OS t^. i-H o oo * i-H O rH O O ^H T* CO t^OSO (M rH CO 1 OS CO CO ^ CO O ^ CM O CO OS O l-H I-H OSg^CM m i^ I-H rH rH rH O J> CO CO CO to os co co o (M (N rH O CO OO CO OS rH Tf* T 1 OS CO CO ^rH - O rH *>. m CM CM i-H OO rl : : : o o : II-H : o I-H : : o : o - co o CM rH OS rH OS co co l-H O 1 C CN I o co : o in i i o '. o o : o * : o CM : |S^ c^S i|| .s 2^B III 1 CM 3 CO CM o CO CN Q : 1 i ^ >> CO 1 -3 & PH CO Belgaum, 34 to 35 years Secunderabad, 44 years 1 s i I 02 2 A 354 APPENDIX III I 1>- 00 o io : 1O OO o 10 : CO CO CO O b . ^f OO rH rH CN iO CO t^ J>- OS CO 1 CO CO rH rH CO i-H O I-H rH CN rH t^JOrH CO CN rH rH CO I-H } 05 CN 00 VO "^ O CN O co oo : CN *! i 1 rH CO O 00 CO 00 O rH O O5 rH CO OS rH CO CN CN cb : "s i-s -* I-H CO CO CN CN O O I-H rH CO CO O rH O Cl rH 4)H O OS rH CN rH rH 1^ CN Cp CN iO CO rH CO CO H P "-5 CN CO CO CO 00 O OS O 00 O 00 CO 00 CN 0^ ^brH ^ CN O I 1 ^ CO 00 ^ oo b oo cp rH 10 I 1 p O rH IO O I-H cp ^* o CN : rH CO IO CO 00 S CO CO CN co oo b rH cs : OO CO rH iO : d CO (N CNCN o ' i Q CN CO i-H ^ "* O OS 1O I-H iO 00 CN I-H -tf : cNt- : CN o b ^ b CN : o o : o o : o^ : 1 CO -^ o cb : b o : rH 1^ CO O I-H co r b i>. : CO O b CN : co <*< b * : cp co rH O5 i-H O CO IO IO CO I-H rH CO 05 i o rn : o CN : : o : o 10 : o co : o rH : : o : CN OS rH CO CO 1>- CN CO CN CO CD rH O 1 o CN : o I-H : o co : o CN : o co : O rH I o CN : iii 1 S.g III sM S as >S III 3 * d 8 CB.S i i 1 o i '. 1 i CN to B CN o3 2 & '1 a Eajahmundry, 25 years Masulipatam, 24 years 1 CO CO 1 1 MONTHLY RAINFALL TABLES 355 !>. OO rH 00 rH rH rH OQ (M 1O (N (N 10 01 01 t- o -^ : 10 CM : (M co CN : rH rH -<* : - CO rH *3 i-i CO -^f O rH CO rH O -lO O HS - Ol (N CO CO (M O Ol ^ CO <* i ( > CO ^ . : o * : o 10 : 1 Is" ass 1 S.S sss 3 c - Sal ls.a sss is & ~ . s^s Is. a s^s 1 g.a r^^S Is. a ^^^ OT 1 ^t CO } 'd 3 o a 1 >* ** t~~ 43 OS s Trichinopoly, 33 to 35 years Coimbatore, 24 to 26 years Madura, 26 years Tuticorin, 24 years Mangalore, 28 to 34 yrs. * ^ _0. o <* & 1 O 356 APPENDIX III O5 t O *^ CO CO * t^.-* * O5 O CO T*1 CO CO t^ I OS VO -* CO 8 eo t>. OO -Jf VO Tt< rH CO OS 10 O rH^ ^ CO O (N 00 C* CO CO O5 rH t^ ^ o oo * O5 CO rH CM vo CO (M 05 (N rH i 1 rH !>. t>. rH rH CO t>- rH rn CO ^ O rH rH t^ CO O I I *; 00 rH CO "* -^ 05 CO O 00 "* rj) O CO O5 OO CO CO 00 CO 1 co o : (N O O ^~ O OO rH rH CO OO O> CO "tf t^05 -^ rH rH CO CO \ . IT5 IO - CO t->. rH CO i o vo (N CO rH (M O CO rH IO OO CO CO rH O5 VO rH rtl i 1 CO 5 rH oo : rH CO O 00 O O CM O5 CO rH rH rH !>. CO XO O5 CO t^ CO O rH CO rH l^ rH CO ;>, CO -* 00 (N O5 CO *^ CO O CO rH O CO CO 05 CO O O5 . C cj .5 s^s Us sss Il5 BSS Us sss 3 * c o a .3 SSS >> co rH 4* Batticaloa, 16 to 17 yrs. Colombo, 17 years 1 t> t^ f M 1 >* rH CO 3 O5 CO 1 ^ *< 1 t^ o" >, m 1 H o >> t^ cT o 1 1 MONTHLY RAINFALL TABLES 357 rH^H rH OS rH CO rH O cot^ CM <* CM i p O rH I o o : o o : O rH I xo xo : (M -* CM rH CO O 00 *tf CO xot^ CO OO xo ^< 00 O rH o CM co : co xo : I-H o : CN os : OS O CM CM rH t^ CO rH CM rH t^ . rH CM CO XO !>. rH CM . -tf XO CO O OS OS CO ^ O 1^ CO XO CO OS OS CO O xo 1 OS O OS rH CO co xo r^ rH CM CO xo CO CO rH CO OS (M CO xo rH OS OS CM rH CM rH O **- "* rH rH -J-i 00 O cot^xo O CM OS OS rH O CM CO XO CO O CM 1 (M xo (M CO "* CM OO O3 OS rH CM CO CM CO rH 00 ^ CO CM XO -^ t^ rH CO (M t^ CO rH rH OO J>- l-~ CO *>. -^ OS -^ CO 000 XO XO rH CO CO rH 1-9 CO XO rH CO XO rH rH XO t>- CM CO CO OO O Til CO CM CO t^ OO CO -^ CO rH CM CM CO Tt< rH CM as CO 00 ^-*XO -HH O -* OS xo XO OS CO XO to XO CO p i-s O ^H CO CO XO rH 00 CM rH rH CO rH OO XO rH CO CO CM OS (M CO CO CM 55 -t^* OS rH W r-| CM rH CM O * rH CM oj OS O Tfl O rH rH rH CO OS CM rH t~ooos s^^* CO XO CO CO XO rH r-t CO o 2 OS XO 00 r^j XO CO rJH r-l CM Q rH CO CO J CO XO oo os o o O CM ^ CM CO CO ^ CM o : rH rH^ : co os : * os : rH 3 XO rH O rH March. CO rH CM 1 rH XO b b : rH fr rH ! 1>- OO b * : g -^CO rH rfl CO CM 00 b X XO rH rH CO rH CO CO CO CO CO o co : O rH : O rH : o i>- : rH oo : rH XO : CM CM OS CO CM t^ OS OS OS ^ CO 1 : o : O rH : : o : o CM : o CM : CM OO O lis o cs .a Is. a III Us III 1 I CO rH I Moulmein, 37 years . I o 3 t OS rH f pq PH Nancowry, 13 to 14 yrs. APPENDIX IV DIRECTIONS FOR THE USE OF THE STORM SIGNALS AT PORTS AROUND THE BAY OF BENGAL, ISSUED BY THE GOVERN- MENT OF INDIA 1. The storm signals for use at ports around the Bay of Bengal (excepting at the stations on the Hooghly River) will be of two kinds for the daytime : No. 1. A ball as bad weather signal. 2. A drum as storm signal with corresponding lamp signals for night, viz. 3. Three lamps suspended on the same line one above the other (in a vertical series) as bad weather signal. ,, 4. Two lamps similarly suspended as storm signal. 2. The ball or three lamps will be hoisted to indicate that a cyclone is in existence on the bay without reference to its position and probable course. 3. The drum or two lamps will be hoisted to indicate that the storm is probably approaching the port. 4. The ball or three lamps will be hoisted on the receipt of telegraphic information from the Calcutta Meteorological Office. 5. When the ball or three lamps is hoisted, port officers must rely on their own judgment whether the drum or two lamps should be hoisted or not, and if so, when. If possible, later information will be telegraphed from the Calcutta Office, but they must not reckon on receiving such. The indications of the barometer and of the direction and force of the wind, together with the general aspect of the weather, must be their chief guides. Some general memoranda are hereto appended to assist them in forming a judgment. MEMORANDA ON STORM PROGNOSTICS 1. Cyclones have been known to occur in the Bay of Bengal STORM SIGNAL NOTIFICATIONS 359 in all months of the year except February, but not between the middle of January and the last week in March ; and from the middle of December to the middle of April such cyclones as are on record have been restricted to the southern half of the bay. 2. Cyclones that reach the west (Indian) coast of the bay, follow a course between west by north and north-north-west. On the approach of a cyclone, the wind is generally at first from north-east. If it strengthens from this direction, becoming more squally while the barometer falls rapidly, it is probably approach- ing the port. But if the wind backs beyond north, the barometer falling moderately, it is passing to the east, and the centre bears about east when the wind is about north-west or north-north- west. If, on the other hand, the wind veers towards east, with a moderate fall of the barometer, the cyclone has crossed the coast line to the south of the station by the time the wind is from east. 3. Cyclones that reach the Arakan coast follow a course between north and north-east. On the approach of a cyclone, the wind is generally from north-east or east-north-east. If it strengthens from between these directions, becoming more easterly with a rapidly falling barometer, it is probably approach- ing the port. But if the wind veers to south-east with a moderate fall of the barometer, the. storm is passing up the bay, and the centre bears about due west when the wind is at south- east or south-south-east. If the wind backs to north, with a similar change in the barometer, the storm centre has reached the coast to the south of the station. 4. Cyclones that reach the coasts of Pegu and Moulmein are formed in the Andaman sea, and follow a course between north and north-east. The indications of the wind and barometer may be interpreted as in the previous paragraph. 5. In judging of the direction of the storm centre, it must be borne in mind that the winds do not blow round the centre in a circle, but spirally inwards. When the centre is still at some distance, it bears about 12 points to the right, if the observer faces the wind. As it approaches nearer, it bears about 10 points to the right. 6. In judging of the fall of the barometer, it must be borne in mind that the barometer falls regularly about a tenth of an inch from 9.30 A.M. to 4.30 P.M., and rather less from 10 P.M. to 3.30 A.M., and rises by similar amounts from 4 to 9.30 A.M., and from 5 to 10 P.M. These changes, which are regular, must be allowed for. APPENDIX V NOTIFICATIONS OF THE GOVERNMENT OF BENGAL RELATIVE TO STORM SIGNALS FOR THE PORTS OF BENGAL The 4:th August 1887. The following arrangements for weather signals for the approaches to the Port of Calcutta, and for the port itself, have been sanctioned by the Lieutenant- Go vernor, and are published for general information, in super- session of those published in the notification, dated the 5th August 1882, at pages 678-80 of the Calcutta Gazette of the 9th August 1882 : Weather Signals for the approaches to the Port of Calcutta, and for the Port of Calcutta 1. When a severe cyclonic storm (probably accompanied by a storm-wave) is believed to be advancing to the Bengal coast and across Bengal, one or other of the following signals will be hoisted at the flagstaff near the Lighthouse in Saugor Island, at the flagstaff near the telegraph office, Mud Point, and at the flagstaff near the telegraph office, Diamond Harbour, and also at the following positions in the port, viz. the flagstaff upon the Port Commissioners' Office, on a flagstaff at Shalimar Point, Seebpore, opposite the Government Dockyard, Kidderpore, and on a flagstaff at the telegraph office, Budge Budge, on receipt of instructions telegraphed by the Meteorological Reporter to the Government of Bengal : A ball indicates the existence of a cyclonic storm of undeter- mined intensity and magnitude in the Bay of Bengal, which will either certainly cross the coast to the south of a line be- tween Chittagong and False Point, or which may approach the Bengal coast, but is as yet too far distant to enable its line of advance to be determined. A cone indicates the early probable passage northward, to the eastward of Saugor Island and west of Chittagong, of the vortex of a cyclonic storm of great intensity and magnitude. It is STORM SIGNAL NOTIFICATIONS 361 advisable that sailing vessels, with or without steam, and deep- laden or slow steaming steam-vessels, should not proceed to sea, but remain in the river till the storm has reached the coast and passed inland. The wind at the mouth of the Hooghly will probably haul from north-east through north to north-west, etc. An inverted cone indicates the early probable passage north- ward, to the west of Saugor Island and north of False Point, of the vortex of a cyclonic storm of great intensity and magnitude. No vessels should go to sea, and masters and pilots of vessels outward-bound should be guided by the appearance of the weather and height of the barometer in deciding whether it is advisable to proceed below Diamond Harbour or Mud Point. The wind at the mouth of the Hooghly will probably veer from north-east through east to south-east, etc. A drum indicates the approach of a cyclonic storm of great intensity and magnitude to the mouth of the Hooghly, and which will probably advance to Calcutta. Masters and pilots in charge of vessels are cautioned not to put to sea from Saugor Island, or to proceed down from Diamond Harbour, and they should make their vessels as snug and secure as possible. The masters of vessels in the port should take the special precautions for safety laid down in the Port Eules. There will probably be a storm-wave, but it should be care- fully remembered that its height and destructive effect will de- pend quite as much upon the state and character of the tide when the cyclonic centre reaches the coast, as upon the depression at the centre, or the intensity and extent of the storm. 2. When a small cyclonic storm (usually not accompanied by 'a storm-wave) is believed to have formed in the bay, and to be advancing towards the coast, one or other of the following signals will be hoisted, for the information of pilots and mariners, at the staff near the Lighthouse at Saugor Island, at the flagstaff at Mud Point, and at the flagstaff near the telegraph office, Diamond Harbour, and also at the following stations in the port, viz. on a flagstaff upon the Port Commissioners' Office, on a flagstaff at Shalimar Point, Seebpore, and on a flagstaff at the telegraph office, Budge Budge, on the receipt of instructions telegraphed from the Meteorological Office, Calcutta, by the Meteorological Reporter to the Government of Bengal : Two cones, the upper one inverted, indicate the existence of a cyclonic storm of small extent in the Bay of Bengal, which will probably reach and cross the coast of the bay south of a line joining Chittagong and False Point. 362 APPENDIX V Two cones, the lower one inverted, indicate the existence of a cyclonic storm of small extent, which will probably or certainly reach and cross the coast of the bay north of a line joining Chittagong and False Point, but the probable path of the vortex of which cannot be determined more exactly from the land observations. A ball below a cone indicates the probable passage northwards, to the east of Saugor Island and west of Chittagong, of a cyclonic storm of small extent and intensity, of the kind which usually forms in the rainy season. Vessels may proceed to sea if the height of the barometer, state of the sea and weather are such as to lead masters and pilots to infer that there is no danger. The wind at the mouth of the Hooghly will probably haul from north-east through north to north-west, etc. A ball below an inverted cone indicates the probable passage northwards, to the west of Saugor Island and north of False Point, of a cyclonic storm of small extent and intensity of the kind which usually forms in the rainy season. The wind at the mouth of the Hooghly will probably veer from north-east through east to south-east, etc. As these easterly winds will raise a heavy swell and produce a strong westerly set in the channel at the Sandheads, it is advisable that none but fast steamers in light trim should put to sea, and those only if the weather appearances and state of the sea, etc., are not too unfavourable. A ball below a drum indicates the approach towards Saugor roads of a cyclonic vortex of small extent, of the kind which forms during the rainy season. It is advisable that no vessels except fast steamers in light trim should put to sea until the wind direction and force, the state of weather and sea, and the rise of the barometer, indicate that the storm has either broken up or passed inland. Cyclonic storms in the Bay of Bengal of small extent, it should be remembered, sometimes blow with hurricane force and raise a high sea near their centres. 3. The following storm warning signals will be hoisted during the prevalence of cyclonic storms in the Bay of Bengal at the signalling stations on the Hooghly previously mentioned, viz. Saugor Island, Mud Point, and Diamond Harbour, and at the following signalling stations in the port, viz. on a flagstaff upon the Port Commissioners' Office, on a flagstaff at Shalimar Point, Seebpore, and on a flagstaff at the telegraph office, Budge Budge, between sunset and sunrise : Three lights exhibited on the flagstaff at equal distances in- dicate the existence of a cyclonic storm in the north of the bay. STORM SIGNAL NOTIFICATIONS 363 Two lights, in a vertical line, indicate the early approach of a cyclonic storm to the Bengal coast, or to the Port of Calcutta if the signal is hoisted in the port. The following arrangements for weather signals have been sanctioned for the Ports of Orissa, viz. Pooree, False Point, Balasore, and Chandbally by the Lieutenant-Governor, and are published for general information : Signals for the Orissa Ports. Storm signals will be hoisted at the ports named above by the port officers on the receipt of instructions from the Bengal Meteorological Office, Calcutta, to give notice to the shipping and the general public at these ports of the approach of a cyclone (which may or may not, according to the state of the tides and other conditions, be accompanied by a destructive storm-wave). The signals for the Port of Pooree will be hoisted at the flagstaff on the beach ; for the Port of False Point at the flagstaff at Hookeytollah ; for the Port of Balasore at the flagstaff near the Custom House ; and for the Port of Chandbally at the flagstaff near the Custom House. The following are the signals which will be used at the Orissa Ports, and are identical with those in use at the Burmah and Madras Ports : A ball indicates that a cyclone has formed in the centre or north of the bay, and that it is advancing in some direction between north and west, and will probably strike and cross the coast of Orissa. A drum indicates that a cyclone is in the immediate neigh- bourhood of the coast of Orissa, and will shortly cross that coast. (It may or may not be accompanied by a destructive storm- wave. The height of the storm-wave depends partly upon the extent of the cyclone, the barometric depression at the centre, and the state of the tides at the time when the cyclone reaches the coast.) Three lights in a vertical line indicate that a cyclone has formed in the centre and north of the bay, and that it is advancing in some direction between north and west, and will probably strike and cross the coast of Orissa. Two lights in a vertical line indicate that a cyclone is in the immediate neighbourhood of the coast of Orissa, and will shortly cross that coast. (It may or may not be accompanied by a destructive storm- wave.) The following arrangements for weather signals for the Port 364 APPENDIX V of Chittagong have been sanctioned by the Lieutenant-Governor, and are published for general information : Signals for the Port of Chittagong. The following storm signals will be used within the limits of the Port of Chittagong to give notice to the shipping and general public of the early approach of a cyclone with its attendant storm- wave. They will be hoisted on the flagstaff at the Sudder Ghat and on the yardarm of the flagstaff at Jooldia on the receipt of instructions from the Meteorological Office : A loll indicates that a cyclone, of which the centre is in the north-east of the bay, is advancing towards the mouth of the River Karnafuli, and will probably advance towards Chittagong. A drum indicates that a cyclone, which may or may not be accompanied by a destructive storm-wave, is approaching Chitta- gong. (It should be remembered that the force and destructive effect of a storm-wave will depend very considerably upon the state of the tides at the time when the storm centre strikes the Chittagong coast.) Three lights in a vertical line indicate that a cyclone, of which the centre is in the north-east angle of the bay, is advancing towards the mouth of the river, and will probably advance towards Chittagong. Two lights in a vertical line indicate the early approach of a cyclonic storm to the Port of Chittagong. INDEX ABU, 10, 18, 73, 74, 268; climate, 117, 296 ; monthly rainfall, 347. Agra, 18, 73, 74 ; climate, 144, 308 ; monthly rainfall, 345. Ajinere, 18, 31, 73, 74, 278 ; climate, 139, 305 ; monthly rainfall, 347. Akola, climate, 166, 322 ; monthly rainfall, 351. Akyab, 73, 75, 209, 226, 227, 254, 337, 341 ; climate, 193, 332 ; monthly rainfall, 356. Allahabad, 5, 18, 61, 73, 74, 77, 267 ; climate, 144, 309 ; monthly rainfall, 346. Ambajhari tank, 274, 278. Amraoti, 10, 73, 75. Andamans, 209, 219, 245 ; climate, 196, 336 ; rainfall, 357. Anticyclones, 44, 197. Arabian Sea, cyclones, 86, 87, 88, 215, 246 ; storm tracks, 248. Arakan, rainfall, 69 ; humidity, 50 ; climate, 189 ; storms, 359. Ashti tank, 274, 277. Assam, fogs, 55 ; rainfall, 69, 268, 269 ; humidity, 50 ; climate, 157. BALUCHISTAN, 50 ; see also Quetta. Bangalore, 18, 31, 73, 75, 272 ; climate, 176, 328 ; monthly rainfall, 354. Bannu, monthly rainfall, 343. Bareilly, 9, 31, 267 ; monthly rainfall, 345. Barometer, teachings, 19 ; annual fluctuation, 19 ; diurnal fluctuation, 20; changes with weather, 20, 21, 197, 236 ; temperature correction, 21 ; reduction to sea level, 23 ; re- lation to winds, 43 ; in storms, 236. Barometric charts, 24. Barometric depressions, 22, 26. Barometric gradients, 24. Barometric surges, 200. Batticaloa, monthly rainfall, 356. Bay Islands, 50 ; see also Andamans, Port Blair, Nancowry. Bay of Bengal, storms, 225, 337 ; storm tracks, 228, 337 ; storm sig- . nals, 252, 358, 360. Beaufort's numbers, 33. Behar, humidity, 50 ; rainfall, 69 ; climate, 147. Belgaum, 18, 31, 73, 75 ; climate, 174, 326 ; monthly rainfall, 353. Bellary, 18, 73, 75 ; climate, 175, 327 ; monthly rainfall, 354. Benares, 31, 73, 74; climate, 144, 309 ; monthly rainfall, 346. Bengal, cloud, 5 ; humidity, 50 ; rain- fall, 69, 268, 269 ; climate, 151. Berar, humidity, 50 ; rainfall, 69, 268 ; climate, 162. Berhampore, 18, 73, 74 ; climate, 156, 315 ; monthly rainfall, 349. Betul, climate, 165. Bhagulpore, 73, 74 ; monthly rainfall, 348. Bickanir, climate, 138, 304 ; monthly rainfall, 346. Bimlipatam, 73, 75, 340. Bombay, 18, 31, 35, 73, 75, 250; V climate, 169, 323 ; storm signals, 255 ; daily rainfall, 262 ; excessive rainfall, 268, 269, 270; monthly rainfall, 352. Bombay coast, storms, 248 ; storm signals, 255. Buldana, 31. Burdwan, 237, 288 ; climate, 156, 316 ; V monthly rainfall, 348. Burmah climate, 189. CACHAR, climate, 157 ; see also Assam.^X Calcutta, 3, 5, 12, 18, 31, 35, 60, 73, 74, 219, 236, 238, 240 ; climate, 152, 156, 316 ; daily rainfall, 260 ; excessive 3 66 INDEX rainfall, 266, 268, 272 ; monthly rainfall, 349 ; storm signals, 253, 360. Cape Colony, 276, 278. Carnatic, humidity, 50 ; rainfall, 69 ; climate, 176. Cawnpore, monthly rainfall, 346. Central India, humidity, 50 ; rainfall, 69, 268 ; climate, 144. Central Provinces, rainfall, 69, 268, 269, 271 ; climate, 162. Ceylon, humidity, 50 ; climate, 183 ; rainfall, 356. Chakrata, 9 ; climate, 106, 293. Chamba, 9. Cherrapunji, 73, 74, 264, 268, 269, 270. Chikalda, 10, 268 ; climate, 116, 296. Chittagong, 18, 73, 74, 209, 339, 340, 341 ; climate, 155, 318 ; monthly rainfall, 349 ; storm signals, 363. Chutia Nagpur, humidity, 50 ; rainfall, 69 ; climate, 147. Chybassa, monthly rainfall, 348. Cinchona cultivation, 111, 170. Cirrus cloud, 58, 241 ; movement, 60. Climates of India, 95 ; hill stations, 96 ; plains, 126. Clouds, forms and nomenclature, 56 ; classification, 58 ; teachings, 59, 241 ; estimation, 61 ; distribution, 5, 62. Cochin, 18, 73, 75 ; climate, 169, 324 ; monthly rainfall, 355. Coimbatore, 10 ; climate, 182, 330 ; monthly rainfall, 355. Cold season in Punjab, 129 ; in North- west Provinces, 141 ; in Bengal, 152 ; weather, 202. Colombo, 73, 75 ; climate, 185, 189, 332 ; monthly rainfall, 356. Coorg, climate, 170 ; see also Mercara. Cuddalore, 73, 75. Cuddapah, 73, 75 ; monthly rainfall, 355. Cutch, excessive rainfalls, 268, 269. Cuttack, 18, 73, 74 ; climate, 156, 318 ; monthly rainfall, 349. Cyclones, 44, 84, 197, 209, 219, 223 ; Bay of Bengal, 225, 337 ; prognos- tics, 240, 358 ; Arabian Sea, 246 ; signals, 252, 358, 360. DACCA, 18, 73, 74 ; climate, 156, 317 ; monthly rainfall, 349. Dadu (wind), 36. Daily rainfall, 75, 259, 265. Darjiling, 10, 18, 73, 74 ; climate, 108, 294 ; monthly rainfall, 348. Deccan, rainfall, 69, 271 ; climate, 171. Deesa, 10, 18, 31, 73, 74; climate, 139, 306 ; monthly rainfall, 352. Dehra, climate, 143, 306 ; excessive rainfall, 267, 269 ; monthly rainfall, 344. Delhi, 18, 72, 74, 77, 267, 269 ; climate, 132, 301 ; monthly rainfall, 344. Dera Ishmail Khan, 18, 72, 74 ; climate, 133, 302 ; monthly rainfall, 343. Dhubri, 10 ; climate, 166, 319; monthly rainfall, 350. Dhulia, monthly rainfall, 73, 75, 353. Diurnal, variation of temperature, 7 ; barometer, 20 ; winds, 34 ; humidity, 48. Drainage, proportion to rainfall, 273. Droughts, 80. Durbhanga, climate, 149, 313. Dust-storms, 81. EVAPORATION in India, 276. Excessive rainfalls, 77, 265, 272. FAMINES in North-west Provinces, 141 ; see also Droughts. GALLE, 14, 15, 18, 73, 75. Gangetic plain, climate, 139. Gauhati, monthly rainfall, 350. Goa, monthly rainfall, 353. Gorakhpur, monthly rainfall. 73, 74, 346. Guzerat, humidity, 50 ; rainfall, 69. Gya, climate, 150, 314 ; monthly rain- fall, 348. HAZARIBAGH, 18, 73, 74; climate, 151, 315 ; monthly rainfall, 348. Haze, 53. Hill stations, solar intensity, 2 ; ther- mometric range, 8, 13 ; climates, 96. Himalaya, humidity, 50. Hoshiarpur, monthly rainfall, 344. Hoshungabad, thermometer range, 10 ; monthly rainfall, 351. Hot season in Punjab, 127 ; North- west Provinces, 142 ; Calcutta, 153. Hot winds, 35, 48, 127, 141. Hourly rainfalls, excessive, 272. Humidity, 46 ; diurnal variation, 48 ; relation to winds, 49 ; geographical distribution, 50 ; annual variation, 51. Hyderabad (Deccan), humidity, 50 ; rainfall, 69 ; see also Secunderabad. Hyderabad (Sind), rainfall, 72, 74, 266, 268, 352 ; climate, 135, 303. INDEX 367 INCURVATURE of winds in cyclones, 233. Indore, rainfall, 73, 74, 347 ; climate, 146, 310. Insolation,. 2. Isobars, 24. JACOBABAD, 9, 18, 31, 72, 74; climate, 135, 303 ; monthly rainfall, 350. Jaffna, monthly rainfall, 356. Jalna, rainfall, 73, 75. Jalpaiguri, 73, 74. N Jessore, climate, 156, 317. Jeypore, climate, 139, 305 ; monthly rainfall, 347. Jhansi, rainfall, 73, 74, 346 ; climate, 146, 311. Jodhpur, rainfall, 73, 74, 347. Jubbulpore, 18, 73, 74 ; climate, 147, 312 ; monthly rainfall, 351. KAILANG, 9. Kandy, climate, 186, 189, 331; monthly rainfall, 356. Kangra, monthly rainfall, 342. Karwar, 73, 74. Khandesh, humidity, 50 ; rainfall, 69 ; climate, 171. Konkan, humidity, 50 ; rainfall, 69 ; climate, 166. Kurrachee, 18, 31, 72, 74 ; climate, 136, 304 ; monthly rainfall, 351. LADAK, humidity, 50 ; climate, 99. Lahore, 3, 9, 12, 18, 31, 72, 74 ; climate, 131, 300 ; daily rainfall, 261 ; monthly rainfall, 343. Lanauli, monthly rainfall, 352. Land winds, 49. Leh, 9, 18, 72, 74 ; climate, 99, 291. Lucknow, 18, 73, 74 ; climate, 144, 308 ; daily rainfall, 260 ; monthly rainfall, 345. Ludhiana, climate, 132, 300 ; monthly rainfall, 344. MADRAS, 3, 5, 12, 18, 31, 73, 75, 200. Madras, storms, 209, 219, 226, 227, 245, 337, 340, 341, 359 ; climate, 181, 329; daily rainfall, 262 ; exces- sive rainfall, 268, 269, 270 ; monthly rainfall, 355. Madura, 18, 73, 75 ; climate, 182, 330 ; monthly rainfall, 355. Mahableshwar, rainfall, 73, 75, 268, 269, 352 ; climate, 118. Malabar, humidity, 50 ; rainfall, 69 ; climate, 166. Malegaon, climate, 174, 325. Mandalay, climate, 192, 333. Mangalore, rainfall, 73, 74, 355 ; climate, 169, 323. Masulipatam, 73, 75, 254 ; climate, 180, 328 ; monthly rainfall, 354. Matheran, 73, 75. Meerut, 18, 73, 74, 267 ; climate, 144, 307 ; monthly rainfall, 345. Mercara, climate, 170, 324 ; monthly rainfall, 354. Mergui, climate, 195, 335. Midnapore, monthly rainfall, 348. \j Mongpoo cinchona garden, climate, 111. Monsoons, 36. Mooltan, 18, 73, 74 ; climate, 133, 302 ; monthly rainfall, 343. Moradabad, monthly rainfall, 345. Morar, monthly rainfall, 347. Moulmein, 18, 73, 75, 245 ; climate, 195, 335 ; monthly rainfall, 357. Murree, 3, 9, 18, 72, 74 ; climate, 101, 292 ; monthly rainfall, 342. Mussoorie, monthly rainfall, 344. Mysore, humidity, 50 ; rainfall, 69, 354. NAGPUR, 18, 31, 73, 75, 274, 278 ; climate, 166, 321; daily rainfall, 261 ; monthly rainfall, 351. Nairn Tal, rainfall, 267, 269, 345. Nancowry, 196 ; climate, 196, 336 ; monthly rainfall, 357. Neemuch, 18, 73, 74 ; climate, 146, 310 ; monthly rainfall, 347. Negapatam, 31, 245, 254, 337, 340, 341. Nepal, 50. Nerbudda Valley, 50, 272. Newara Eliya, climate, 123, 298. New South Wales, surface drainage, 275. Nicobar settlements, climate, 196. Nilgiris, climate, 119. Northern Circars, humidity, 50 ; rain- fall, 69. Nor' -westers, 82. North-west Provinces, cloud, 5 ; humi- dity, 50 ; rainfall, 69, 267,269, 271 ; droughts and famines, 81, 141 ; climate, 139. OOTACAMUND, 10 ; climate, 119, 297. Orissa, humidity, 50 ; rainfall, 69 ; climate, 151. PACHMARHI, 10, 18, 73, 75, 269 ; cli- mate, 114,295 ; monthly rainfall, 351. Patna, 18, 31, 73, 74; climate, 150, 313 ; monthly rainfall, 347. Pegu, humidity, 50 ; rainfall, 69 ; climate, 189, 192. 368 INDEX Peshawar, 31, 72, 74 ; climate, 130, 298 ; monthly rainfall, 342. Piddington on storms, 223. Poona, 18, 31, 73, 75 ; climate, 174, 325 ; monthly rainfall, 353. [ Port Blair, 31 ; climate, 196, 336 ; * monthly rainfall, 357. Pressure, atmospheric, distribution, 22 ; relation to winds, 43. Pressure of wind, 33, 2S6. Punjab, cloud, 5 ; rainfall, 69, 267, 269, 270 ; climate, 126. Purneah, climate, 150, 314 ; excessive rainfall, 77, 268 ; monthly rainfall, 348. QUETTA, 9, 18, 72, 74 ; climate, 97, 291 ; monthly rainfall, 342. RAINFALL, contrasts, 63 ; distribution, 65 : tables, 69, 72, 260, 267, 269, 291, 342 ; heaviness, 75, 258, 265, 272 ; variability, 78 ; cyclical varia- tion, 80 ; relation to drainage, 273 ; on river basins, 279. Rainfalls, special daily, 265 ; hourly, 272. Raipur, climate, 165, 321 ; monthly rainfall, 351. Rajamundry, rainfall, 73, 75, 354. Rajkot, monthly rainfall, 352. Rajputana, humidity, 50 ; rainfall, 69 ; climate, 136 ; excessive rainfall, 268. Rangoon, 18, 31, 73, 75, 245 ; climate, 194, 334 ; monthly rainfall, 357. Ranikhet, 9, 267, 269 ; climate, 107, 293 ; monthly rainfall, 345. Rawalpindi, 18, 72, 74, 204 ; climate, 131, 299 ; monthly rainfall, 343. River basins, rainfall, 279. Roorkee, rainfall, 72, 74, 345 ; climate, 107, 293. Rules for avoiding cyclones, 241. ST. AUBIN'S, 5. Sambalpur, 73, 74. Saugor, 18, 73, 74 ; climate, 146, 311 ; monthly rainfall, 350. Secunderabad, 18, 31, 73, 75 ; climate, 175, 327 ; monthly rainfall, 353. Seoni, climate, 166, 320. Shahpur, monthly rainfall, 343. Shillong, 10 ; climate, 112, 295 ; monthly rainfall, 350. Sholapur, 18, 73, 75, 277; climate, 179, 326 ; monthly rainfall, 353. Shoulder of high pressure, 217. Shwegyin, monthly rainfall, 357. Sialkot, 18, 72, 74, 270 ; climate, 131, 299 ; monthly rainfall, 343. Sibsagar, 18, 73, 74 ; climate, 161, 319 ; monthly rainfall, 349. Sikkim, humidity, 50 ; see also Dar- Silchar, 31, 73, 74 ; climate, 162, 320 ; monthly rainfall, 350. Simla, 9, 18, 72, 74 ; climate, 103, 292 ; monthly rainfall, 342. Sind, humidity, 50 ; rainfall, 69 ; climate, 133 ; excessive rainfalls, 266, 268, 269, 273. Sirsa, climate, 133, 301 ; monthly rainfall, 344. Storms, 81, 202, 208, 211 ; Bay of Bengal, 225, 337 ; Arabian Sea, 246. Storm centre, bearing, 233. Storm signals, 252, 358, 360. Storm tracks, northern India, 86, 205, 212, 216, 270 ; Bay of Bengal, 209, 228, 337 ; Arabian Sea, 248. Summer monsoon, 39, 210. Sunshine, temperature, 2 ; duration, 4. Sun thermometer, 1. Surat, climate, 169, 322 ; monthly rainfall, 352. Surges, barometric, 200. Sylhet, monthly rainfall, 350. TABLES, solar heat, 3 ; duration of sunshine, 5 ; daily range of tempera- ture, 9 ; mean extreme temperatures and ranges, 18 ; daily wind move- ment, 31 ; mean humidity of pro- vinces, 50 ; rainfall of provinces, 69 ; rainfall of stations, 72 ; rainy days, 74 ; excessive rainfalls, 69, 267 ; storms, 225, 228, 230, 246, 248, 337 ; daily rainfall, 260 ; general climate, 291 ; monthly rainfall, 342. Tanks, water supply, 273 ; evaporation, 276. Tanna, monthly rainfall, 352. Tavoy, monthly rainfall, 357. Tea cultivation, 110, 143, 149, 158, 188. Temperature in the sun, 2 ; air, 6 ; diurnal variations, 7 ; changes with weather, 10 ; annual variation and range, 12, 15; general tables, 291. Tenasserim, humidity, 50 ; rainfall, 69 ; climate, 189. Thunder-storms, 81. Thyet Myo, 18, 31, 73, 75 ; climate, 194, 333 ; monthly rainfall, 356. Tornadoes, 88. Tounghoo, 18, 73, 75 ; climate, 194, 324 ; monthly rainfall, 356. Travancore, climate, 166. INDEX 369 Trichinopoly, 18, 73, 75 ; climate, 181, 329 ; monthly rainfall, 355. Trincomalee, 73, 75 ; climate, 187, 188, 331. Tuticorin, monthly rainfall, 355. UMBALLA, rainfall, 72, 74, 344. VEHAR tank, 273, 277. Vizianagram, monthly rainfall, 354. WARNINGS of storms, 252, 358, 360. Weather, 197 ; cold season, 202 ; hot season, 206 ; rainy season, 210 ; around cyclones, 239. Weather reports, 220. Wellington, 10, 18, 73, 75 ; climate, 122, 297. West coast of India, climate, 166 ; storms, 246. Wind pressure, 33 ; at Calcutta and Bombay, 286. Winds, 30 ; daily movement, 31 ; diurnal variation, 34 ; annual vari- ation, 36 ; measurement, 32 ; pressure, 33, 286 ; relations to barometer, 43 ; in cyclones, 44, 223 ; relation to humidity, 49 ; in hot season, 207 ; incurvature in cyclones, 223. Wynaad, climate, 170. THE END Printed by R. & R. CLARK, Edinburgh. UNIVERSITY OF CAl^ORNLK