November 3, 1904] 



NATURE 



capital," and so forth. If that interpretation is correct, 

 then there has been no such drought in South Africa in 

 the years stated. 



This is proved by the accompanying table. It shows the 

 average rainfall over each of the twenty rainfall districts 

 of South Africa, during each year, in percentages of the 

 means. These means have been computed for i6o stations 

 having long records of twenty years, more or less, and are 

 fully given and explained in my " Introduction to the Study 

 of South African Rainfall." The information from which 

 they are derived is open to all who take the trouble to look 

 for it in the annual reports of the Cape Meteorological 

 Commission. 



The great mortality among cattle and stock can be ex- 

 plained without assuming that there has been a prolonged 

 drought. In farming matters we live from hand to mouth. 

 Farmers of the Karroo prefer to pray for rain rather than 

 take the trouble to store it up when it comes. Therefore, if 

 the rain is short in the late summer, and late in coming 

 in the next spring, they have no reserve to fall back upon, 

 and their cattle die. One year's drought kills off the stock 

 almost as surely as fifty years' would. For instance, there 

 was great loss of stock i'n iSgy. Yet what were the facts 

 of rainfall? At my station, where the annual mean is about 

 185 inches, the fall in December, 1896, was 8-42 inches ; 

 in the whole of 1897 it was 8-85 inches, and in January, 



Percentages of Rainfall in the Various Districts of South Africa during the Years 1891 to 1902. 



It is pretty plain that the area of winter rains, including 

 the west coast and Cape Peninsula, was short of rain in 

 1896 ; that 1897 was a dry year over the area of summer 

 rains, which comprises the greater part of South Africa; 

 and that the south coast and adjacent districts, where the 

 rainfall is fairly uniform throughout the year, had a dry 

 year in 1899, and one not very wet in 1895. The area of 

 summer rains, being so much greater than the rest, of 

 course sets the tone of the mean rainfall of the whole 

 country, making 1897 a dry year on the whole, and 1891 a 

 very wet year. 



There seem to be dry areas somewhere or other in pretty 

 well every year. For example, the rainfall was short in the 

 western part of the area of summer rains in 1902, although 

 the fall was good enough further east. It was short over 

 the east-central Karroo and south-east in 1899 in sympathy 

 with the dryness of the south in that year. Even in 1891 

 there was a short fall over an extensive region. 



I fancy that the impression of unusual dryness over South 

 Africa in recent years arises from the misleading mean 

 values used by the Meteorological Commission for com- 

 parative purposes. These are taken from Buchan's rather 

 futile " Rainfall of South Africa," and average fully two 

 inches (equal to perhaps 10 per cent.) too great. Buchan 

 used only the rainfall of the ten years 1885-94 in construct- 

 ing his results, and therefore got inflated averages in con- 

 sequence of the heavy rainfall of 1891 ; whence the rainfalls 

 of recent years are made to appear minus as compared with 

 what is called the mean, whereas, as compared with the 

 better means of longer periods, they would be often ^lus, 



NO. 1827, VOL. 71] 



1898, it was 843 inches. Thus there was a drought during 

 1897, many cattle died, and there was much praying for 

 rain. The year 1903 was probably almost the same as 

 1897, the fall at Kimberley being only some 65 per cent, 

 of the mean, whereas the fall during the last half of 1902 

 was good, and during the first half of 1904 excellent. But 

 with the exception of these years there has been nothing 

 that can properly be called drought, in the sense of Sir J. 

 Eliot's address, over any extended region of South Africa 

 within the past fifteen years at least. Thus there is nothing 

 to justify the statement that we have been under the same 

 influence as that which set up the prolonged drought in 

 Australia and the dry years in India. J. R. Sutton. 



I TRUST to your courtesy to give my reply to Mr. Sutton's 

 criticisms on certain portions of my address at the recent 

 British Association meeting. 



My address was in part based on an investigation I have 

 had on hand for nearly two years, and which will be shortly 

 published as a paper in the Indian Meteorological Memoirs. 

 In that will be found a statement of the chief features of 

 the meteorology of South Africa during the period 1892-1902. 

 It is confessedly based upon very imperfect information — ■ 

 partly derived from newspaper reports, partly from data in 

 certain meteorological reports received from Cape Town by 

 the Calcutta Meteorological Office, and partly from data 

 obtained from Mr. Hutchins, Conservator of Forests, 

 Cape Colony, with whom I have been in correspondence 

 for many years on the meteorology of South Africa and 



