no 



NATURE 



[June 3, 1^97 



points out that although these naked-eye observations are sub- 

 ject to a more considerable personal error than if they could 

 have been made with any kind of instrument, yet the determina- 

 tion of the position of this phenomenon depends to a great deal 

 on the experience of the observer. The best method he suggests 

 for its location is to estimate the distance of its centre from the 

 nearest fixed star, and then to identify the star by a chart or an 

 equatorial pointing ; by this means he can ordinarily determine 

 the centre to within one degree. Any attempt, he says, to locate 

 its position by tracing the outlines among the stars can only lead 

 the observer to very crude results. Prof. Barnard's observations 

 have indicated that the Gegenschein seems subject to periodical 

 changes in the first parts of October and April ; but the weather 

 this year did not permit of a corroboration of this statement. 

 The present observations show that this phenomenon has a 

 north latitude as formerly, and that there is still a tendency to 

 lag in longitude ; that is, its longitude is not exactly i8o° 

 greater than that of the sun. The following summary shows 

 the results for the last fourteen years : — 



Diff. Long. 

 Dates. Lat. (A-0) No. Obs. 



It may be stated that the difference in longitude for the last 

 four years given above is rather high, owing to the fact that the 

 three observations of- 1894 made the difference greater than 

 180°. Those for 1895 and 1897 gave a value considerably less 

 than 180°. 



PERIODIC VARIATIONS OF RAINFALL IN 

 INDIA. 



npHE meteorology of India during the past five years, 1892-96, 

 has been characterised by the largest and most marked 

 periodic variation of rainfall during the past fifty years at least, 

 and probably for a much longer period. The late Mr. Blanford's 

 monograph on the rainfall of India (page 15) gives a brief state- 

 ment of the method he adopted for ascertaining the average 

 rainfall of India, year by year, and the normal average. 

 According to this method the mean normal annual rainfall of 

 India is, very approximately, 41 inches. The following gives a 

 comparison of the actual mean rainfall of India with the normal 

 mean for each year from 1875 to 1896, determined by Mr. 

 Blanford's method. 



The preceding data indicate that from 1876 to 1891 the annua 

 rainfall of India varied somewhat irregularly and by less than 



NO. 1440, VOL. 56] 



ID per cent, from the normal with one exception, viz. during 

 the period 1876-78. The meteorological features of this period 

 were very remarkable {vide Nature, vol. xxi., page 477, "On 

 the barometric see-saw between Russia and India in the sun- 

 spot cycle," by II. F. Blanford), the most noteworthy being 

 the drought and famine in Madras in 1876, and the scanty 

 rainfall in North-western India in 1877, which gave rise to con- 

 siderable suffering. Mr. Blanford subsequently traced the cause 

 of the deficient rainfall in North-western India in 1877 to con- 

 ditions and actions resulting from excessive snowfall in the 

 Himalayan region during the winter of 1876-77. 



The causes of the drought in Mysore and the Madras and 

 Bombay Deccan in 1876, have as yet not been fully ascertained. 

 The snowfall of the winter of 1875-6 was unusually scanty in the 

 North-western Himalayas, although there was a heavy local fall 

 in March and April in a part of the Kashmir Himalayas. The 

 Madras drought and famine was probably due to more general 

 actions and conditions than the winter Himalayan snowfall. 



The present meteorological period of abnormal conditions and 

 rainfall it will, however, be seen is not only more prolonged, 

 but is accompanied by much larger variations in the amount and 

 distribution of the rainfall than were experienced in the period of 

 1876-78. It is, moreover, noteworthy that the cycle or period 

 commenced with excessive rain, continued for a period of three 

 years, and culminating in the year 1893 in an average excess of 

 9-07 inches, or 22 per cent, of the normal fall. This excess 

 rainfall in India in 1893 was equal to the amount of water 

 necessary to supply its largest canal, the Ganges canal, 300 feet 

 wide, 10 feet deep, for a hundred years. This comparison gives 

 a feeble estimate of the surplus water precipitated over India in 

 1893, in consequence of the special meteorological conditions 

 and actions of the period. 



In order to understand the causes of the partial failure of the 

 crops over at least two-thirds of India in the year 1896, it is 

 necessary to bear in mind the more prominent features of its 

 meteorology, that differ very largely from the meteorological 

 conditions in European countries. The following is a brief 

 statement of the chief features of the two monsoon periods in 

 India. 



The year in India may be broadly divided into two seasons or 

 monsoons — viz. the north-east monsoon and the south-west 

 monsoon. These names are derived from the direction of the 

 winds prevailing in the Arabian Sea and Bay of Bengal during 

 the two periods. They are inapplicable over the greater part of 

 India, where the winds are from directions nearly opposite to 

 those indicated by the names of the seasons, and are chiefly 

 determined by the axial directions of the local river valleys. 

 Thus the winds in South Bengal are from south-east, and in 

 Bihar from east, during the south-west monsoon, and are from 

 the opposite directions in the north-east monsoon. It would 

 hence be more appropriate to call the two seasons in India the 

 dry monsoon, and the wet monsoon from their most characteristic 

 features. 



The dry monsoon or season usually commences in Novem- 

 ber or December, and continues until May. Winds of land 

 origin prevail more or less steadily in the interior, and hence 

 the period is usually marked by great dryness of the air and little 

 or no rain. The first three months of this period (December to 

 February), characterised by a comparatively low temperature, 

 are known as the cool weather season ; and the second three 

 months (March to May), when the temperature increases rapidly, 

 and culminates in a period of excessive heat in May, as the hot 

 weather season. During the cold weather season, shallow 

 depressions of large extent, the majority of which form in 

 Persia, enter India from Baluchistan and traverse Northern 

 India from west to east, distributing light rain in the Indo- 

 Gangetic plains and heavy snow over the Western Himalayas. 

 The severity of the hot weather season is occasionally relieved 

 by the occurrence of series of thunderstorms and duststorms, 

 which cool the air for brief periods. Over nearly the whole of 

 the interior of India the cold weather and hot weather dis- 

 turbances occupy a very small portion of the period, and the 

 characteristic features of the dry season are persistent dry 

 weather, with clear skies and large diurnal range of temper- 

 ature. 



The chief crops in Northern and Central India during this 

 period are wheat, barley, linseed, &c., and irrigation either 

 from canals, tanks, or wells is essential in almost all districts for 

 their successful cultivation. If the summer rains cease much 

 earlier than usual, it is not possible to plough and sow the higher 



