June 3, 1897] 



NATURE 



"3 



The preceding discussion has shown the chief features of the 

 south-west monsoon of 1896, which made it so disastrous a 

 season for the staple food crops in India. 



It is natural to inquire whether the deficiency of rainfall in 

 India accompanied a greater determination of the monsoon 

 currents towards, and hence heavier rainfall in, the remaining 

 land areas of the monsoon region. 



The chief land areas which obtain, in general, moderate to 

 heavy rains from the south-west monsoon winds (the continua- 

 tion of the south-east trades of the Indian Seas) are : (i) India ; 

 (2) Burma, Siam and the Malay Peninsula ; (3) the equatorial 

 lake region of Central and East Africa and the Abyssinian 

 highlands. 



In nine years out of ten, when the rainfall is in marked defect 

 in Northern India, it is more or less in excess in Burma ; and 

 the monsoon of 1896 was no exception to the rule, as is shown 

 by the following : — 



The very limited information I have received indicates that the 

 precipitation was also in excess in Siam. 



Information of the rainfall in the third region of summer 

 monsoon rainfall has not yet reached India. I am, however, 

 informed that the curve showing the flood level of the Nile at 

 the gauge stations at Assuan and at the Barrage, near Cairo, at 

 the head of the Delta, for the past year was most unusual. 

 Instead of having one maximum in September, as is usually the 

 case, it presented two well-marked maxima, the first being the 

 ordinary and absolute maximum in September, and the second 

 an abnormal and secondary maximum in November, apparently 

 due to heavy rainfall in October in Abyssinia. Heavy rain is 

 reported to have fallen in the Upper or White Nile Basin in 

 November. The defection of the humid current indicated by 

 this fact is confirmed by the unusual and abnormal easting of 

 the winds in October and November at the Seychelles and 

 Zanzibar. This heavy rainfall hence occurred in the African 

 region at the time when the retreating south-west monsoon was 

 giving unusually scanty rain in Southern India and the Deccan. 



It is also probable that the rainfall in the sea area passed over 

 by the monsoon currents in their advance towards India may 

 have been greater than usual. Mr. Blanford, in his monograph 

 on the rainfall of India, has laid down the following principle : 

 " that it is not when the monsoon current is blowing steadily 

 that rain is most probable, but when it is defected from its 

 normal direction by some local irregularity of pressure." 



It is not possible at the present time to obtain direct evidence 

 on this point. The rainfall of the Seychelles and Zanzibar 

 during the period June to September of the past five years has, 

 on the M'hole, varied inversely to that of the same period in 

 India, thus slightly in favour of the assumption that the rainfall 

 in the sea area was somewhat larger than usual during the past 

 two years, and more especially in 1895. 



So far as can be judged from the limited data available, it is 

 almost certain that the distribution of the rainfall in the whole 

 monsoon land region has not been compensatory, and hence 

 that the precipitation has, from some unknown conditions or 

 actions, been below the normal to a considerable extent, prob- 

 ably more than 5 per cent, of the average of the whole area. It 

 is not possible to infer whether this deficiency has been com- 

 pensated by a heavier precipitation over the equatorial belt of 

 the Indian Ocean and over the Arabian Sea and Bay of Bengal. 

 The data for the insular stations of Port Blair, Minicoy, Amini 

 Devi and the Seychelles are certainly opposed to this supposi- 

 tion. 



It is hence almost certain that the deficient rainfall in India 

 in the past year, 1896, is only a phase of conditions and actions 

 extending over a much larger area. 



The variations in the amount and distribution of rainfall in 

 India during the past five years, 1892-6, cannot be explained by 



NO. 1440, VOL. 56] 



the local meteorological conditions, either antecedent to or 

 during the monsoon, and are, in fact, of such a character as to 

 indicate that they are probably mainly due to corresponding 

 variations in the strength of the south-east trades, of which the 

 south-west monsoon circulation is the northward extension from 

 June to October. As the strength of a steady horizontal air 

 movement is directly related to the pressure gradients, the varia- 

 tions of the latter, which are approximately known from the 

 pressure data of the Observatories at Mauritius, Zanzibar, the 

 Seychelles, and Colombo, may be employed to indicate roughly 

 the variations in the strength of the south-east trades from 

 1892-6. 



The following gives the mean differences of pressure between 

 Mauritius and Zanzibar, Seychelles and Colombo (i) for the 

 period June to August, and (2) for the period June to October 

 of each year from 189 1 to 1896 : — 



The preceding data establish conclusively that the mean pres- 

 sure differences or gradients in the south-east trades region 

 during the south-west monsoon periods of the past five years, 

 1892-96, exhibit a similar law of variation to that of the rainfall 

 of -India. The gradients or pressure differences were above the 

 normal in 1892, 1893, and 1894, reaching their maximum 

 normal in 1893, and considerably below their mean value in 

 1895. The figures seem to indicate that the maximum and 

 minimum pressure effect precede by some months the maximum 

 and minimum rainfall in India, a very important indication if 

 further experience should show it is a general rule and not a 

 casual coincidence. 



The variations of gradient almost certainly indicate corre- 

 sponding variations in the strength of the south-east trades. 

 That there are variations in the strength of the south-east 

 trades has been suggested by Mr. W. E. Hutchins, Conservator 

 of Forests to the Cape Government, in explanation of the 

 variations of rainfall from year to year in the Cape Colony. In 

 a letter (written in April 1896), which I received from him in 

 June, and sent for publication to several of the Indian news- 

 papers, he says : " The absence of the south-east trades in the 

 Cape throughout the past summer {i.e. November 1895 'o 

 March or April 1896) has brought abnormal summer rains to the 

 Cape Peninsula, and drought, scarcity and locusts to the bulk of 

 the continent, where the usual summer rains have failed. Are 



