December 7, 1905] 



NA TURE 



137 



Baluchistan, Afghanistan, Persia, Zanzibar, and Mauritius, 

 while the rainfall of Arabia, the Straits Settlements, and 

 Port Blair was generally in defect. The defect at Port 

 Blair is an illustration of the general rule that the rain- 

 fall of the Indian Sea area frequently varies inversely with 

 that of the land area. 



During the dry period 1895-1902 there was an almost 

 continuous deficiency of rain over North Bombay, Central 

 Provinces, Central India, and the Punjab. There was 

 deficient rainfall for five years in Bengal, for four years 

 in United Provinces and Madras, and for seven years in 

 east and south Punjab. The interior districts suffered 

 more than the coast, and those supplied by the Bombay 

 current more than those supplied by the Bengal current. 

 In [899 the rainfall of North Bombay was 48 per cent, 

 below the normal, and that of Rajputana and Central 

 India 31 per cent, below. For five years out of eight these 

 areas received at least 20 per cent, less rain than the 

 normal. 



The countries bordering on Indian area, and including 

 Australia and South Africa, mostly suffered from want of 

 rain. 



During the wet period 1892-4 the monsoons were re- 

 markable for the length of time over which they extended 

 and for the persistence and steadiness of the monsoon 

 conditions. In the dry period 1895-1902 the monsoons 

 were generally characterised by their shortness. In 1S96 

 and 1899, the years of drought, there was no prolonged 

 delay in the commencement of the monsoon rains, but 

 they stopped earlier than usual by three to seven weeks in 

 the case of the Bombay current, and two to six weeks in 

 the case of the Bengal current. This abrupt termination 

 of the rains had a most disastrous effect upon the crops, 

 especially in the Gangetic Plain and the Central Provinces. 

 The crops dried and withered, and famine resulted. 



The drought of 1896 was due in the United Provinces 

 to scanty rainfall throughout the whole season, whereas 

 in the Central Provinces and Berar it was due to the 

 early termination of the rains. The year 1899 was 

 characterised by the lack of heavy falls (i.e. falls of more 

 than 3 inches in twenty-four hours) over all India, and 

 especially so in the field of the Bombay current. 



The data available show that the rainfall for all the 

 countries which depend for their rain on the Indian Ocean 

 was in excess during 1S92-4, and in defect during 1S95- 

 1902. The rainfall over Russia, Turkestan, and Central 

 Asia varied from the normal in the opposite manner. 



The observations of cloud amount, relative and absolute 

 humidity and temperature, show that the curves for these 

 meteorological elements agree very closely with the curve 

 for rainfall. 



In discussing the variations of atmospheric pressure, Sir 

 John Eliot refers to the important fact that the long-period 

 variations as disclosed by barometric observations are 

 similar in direction, amount and epoch over the whole of 

 India, and gives both annual and monthly data showing 

 this. Examination of the data giving the monthly vari- 

 ation of pressure from the normal shows that there were 

 fairly long periods of continued excess or defect of 

 pressure, that there was a decided oscillation of pressure. 

 The period of oscillation is given as about two years. Sir 

 Norman Lockyer and Dr. Lockyer, in a recent paper deal- 

 ing with the rainfall of the Thames basin, refer to the 

 annual pressure variation at Bombay, and speak of a 

 3-8-year period. Sir John Eliot gives a table snowing the 

 approximate dates of the changes from excess pressure and 

 vice versd, and notes that these changes almost invariably 

 occur about the time of the change of season. 



If these oscillations were due to exchange of air between 

 the Indian Ocean and southern Asia, such as might 

 accompany the seasonal changes, then the oscillations of 

 pressure over these areas would be of similar period, but 

 of opposite phase; but comparison of the Indian data with 

 data from East Indies, China, South Africa, and Australia 

 shows that this was not generally the case during 1895- 

 1902. In 1893, when there was a deficiency of pressure 

 recorded at Batavia, Singapore, Cape Town, Perth, and 

 Adelaide, there was excess of pressure at Mauritius, 

 Zanzibar, Hong Kong, and Zika Wei, and over India, but 



NO. 1884, VOL 73] 



in other years, notably 1896, 1898, and 1899, there was a 

 general agreement over the whole Indian Oceanic area and 

 southern Asia. According to Sir John Eliot, this was not 

 the case previous to the period under discussion. He 

 says (p. 273): — "It was shown in the memoir 1 that the 

 pressure variation at Mauritius from 1877 to 1889 presented 

 long period oscillations or variations of similar period but 

 opposite phases to the pressure variations in India," and 

 also (p. 276) " the usual relation based upon previous in- 

 vestigations is for the pressure variations in Southern Asia 

 to be of opposite character or sign to those of the Indian 

 Oceanic region." Therefore he concludes that there was 

 some great and abnormal movement of air affecting the 

 barometric pressure over half the eastern hemisphere, but 

 he has no data available to show the region where the 

 opposite variation has taken place. 



Sir Norman Lockyer and Dr. Lockyer, in their paper 2 on 

 " The Similarity of the Short Period Pressure Variations 

 over Large Areas," refer to a set of curves representing 

 the pressure variations in Bombay, Colombo, Batavia, 

 Mauritius, Perth, Adelaide, and Sydney, saying " the 

 striking similarity between these curves shows that over 

 the whole of this area, which includes both north and 

 south latitudes, the same kind of variations is in action, 

 and that therefore the whole region is intimately con- 

 nected meteorologically." 



These curves refer to the period 1874 to 190 1. Attention 

 might be directed to the fact that the term " long-period " 

 seems to be applied by Sir John Eliot to variations which, 

 when discussed by Sir Norman Lockyer, are called " short- 

 period." 



In another paper ' by Sir Norman Lockyer and Dr. 

 Lockyer two pressure curves are given, one for Bombay 

 and one for Cordoba (Argentine), which are referred to 

 thus : — " Dealing with the pressure of Cordoba during the 

 high pressure months April to September, the curve re- 

 presenting the variation from the mean from year to year 

 is exactly the inverse of the curve representing the Bombay 

 and other Indian pressures for the same months over the 

 same period of time. The cause therefore which raises the 

 mean value for the low pressure months over the Indian 

 area would appear to lower the mean value of high pressure 

 months at Cordoba simultaneously. In fact we have a 

 see-saw. ' ' 



In a further paper 1 by the same authors, the surface of 

 the globe is divided into two areas, one having the pressure 

 variations of the Indian type and the other those of the 

 Cordoba type. 



These quotations show that there is evidently a differ- 

 ence of opinion on the question of the similarity or dis- 

 similariiv of the pressure conditions of Southern Asia, 

 Australia, and Africa previous to the year 1892 ; and it is 

 quite possible that the meteorology of these regions during 

 the period, 1892-1902, was not so abnormal as Sir J. Eliot 

 suggests. 



From a discussion of the observations of variation of 

 solar radiation, as indicated by the black bulb thermometer, 

 Sir J. Eliot states that the data indicate that during 

 189 1 to 1896 or 1897 there was an excess of solar radiant 

 energy, and during 1898-1902 there was defect. 



As such a defect would diminish the supply of aqueous 

 vapour, and consequently the rainfall, accurate observations 

 of the variations in solar radiation should give an explan- 

 ation of the variations of the rainfall and air pressure. 

 Observations by means of the black bulb solar radiation 

 thermometers are, however, not considered very satis- 

 factory. 



Appendices to this important memoir give extracts from 

 various official reports on the famines of 1897 an ^ 1900 

 containing information with regard to the damage to crops 

 and cattle. A large amount of data is also given referring 

 to seasonal rainfall, rainy days, pressure, and dates of 

 commencement and termination of the monsoon rains 

 during the period discussed. Twenty-one plates of curves 

 relating to the same observations form a not unimportant 

 part of the volume which they conclude. W. M. 



1 " Indian Meuoroloeical Memoirs," \ol. vi. 

 - Roy. Soc. Proc, vol. Ix.xi., p. 134. 

 3 Roy. Soc. Proc, vol. Ixjc.. p. 502. 

 * Roy. Soc. Proc, vol. lxxiii., p. 457 



