164 0. Little — The Himalayan summer storm of Sept. 24th, 1903. [No. 4, 



connection between the condition of the upper strata of the atmosphere 

 and the rainfall distribution, and was taking careful note of all avail- 

 able information. Mr. Eliot in Bengal was at the same time explaining 

 all changes by the ground-level observations, in fact by pressm'e varia- 

 tions. The year 1882 was in many of its features similar to 1903. 

 There was the same remarkable series of cyclonic storms in July caus- 

 ing serious misgiving to the authorities in Bengal ; there was in both 

 years a change of the same kind early in August, and subsequently 

 sufficient rainfall ; there was a similar cyclonic storm at the end of the 

 season, though not quite so late in 1882 as the first week of October. 

 That is, all the more striking features of the monsoon season of 1882 

 have been repeated in 1903 (after an interval which, it may be noted, 

 is suspiciously like two sunspot cycles). It is hardly necessary to repeat 

 that these past few years have shown that the system on which weather 

 forecasting has lately been worked has failed : but what cannot be too 

 frequently repeated is that these past two years have supplied valuable 

 evidence as to the reason for the failure, and that that evidence is in its 

 more important aspects similar to what was before students of Meteo- 

 rology in India in the early eighties. The failure has been due to the 

 exclusive reliance placed on ground-level observations. If the line of 

 investigation suggested by Mr. Blanford twenty years ago had been 

 followed up, we would probably now have been able to take full advan- 

 tage of the information to be extracted from the varied conditions of 

 the series of years between 1897 and the present year. 



In 1882, Mr. Blanford was pointing out that other information 

 than what is supplied in the registers of the observatories is necessary 

 to solve the weather problem in India. In 1903, Sir J. Eliot after 

 twenty years more study of registers and without any assistance from 

 an investigation of the upper strata, records his opinion that the 

 character of the weather in India is determined in the southern 

 seas. 



The inference appears to me to be obvious. If we are to make 

 headway with the problem of weather forecasting the lines must be 

 taken up where they were laid down 16 years ago by Mr. Blanford. 



It may be asked what means have been available for following out 

 the investigation of the upper strata outlined twenty years ago. The 

 answer may be, that meteorologists in Europe and America have been 

 steadily developing and improving such methods, and that a consider- 

 able amount of information has been collected regarding the upper 

 strata over those regions. That being so, the next question is what 

 success in weather forecasting has followed this extension of the field 

 of observation — a question to which a negative answer must be given 



