48 C. Little — On two remarkable rain-hursts in Bengal. [No. 2, 



The change of temperature shows that the disturbance which bad 

 begun at 4 a.m. in Calcutta, reached the Pilot Brig between 8 and 10 

 a.m., and the column giving wind force shows that nothing more than a 

 moderate breeze was experienced. The increase of cloud began about 

 8 a.m., and the sky was more or less overcast during the day. 



Part III. 



In the preceding, which I have called Part II, I have considered 

 only the weather changes, as they are indicated chiefly by the 8 a.m. ob- 

 servations from day to day during the period of disturbance. These 

 are of sufficient interest to justify their separate consideration. Bat 

 the two storms, which in what follows I shall represent by the dates 

 June 30th and August 11th, appear to me to have caused a change so 

 striking in the atmospheric conditions over Northern India, that those 

 dates become punctuation marks in the monsoon season of 1902. The 

 expression " punctuation marks " inadequately conveys my full mean- 

 ing, and I would perhaps indicate more clearly the importance of the 

 changes which then took place if I say that new chapters begin with 

 those dates. It is impossible in the space which I now have at my dis- 

 posal to go fully into the wider question which I am attempting to open 

 out, even if I had the material ready. But I will indicate briefly the 

 general run of the argument in order to form a line of connection with 

 some future effort in this direction. 



A study of the monsoon season of 1902 falls naturally into four 

 periods : — 



A — From the beginning up to the end of June, that is until the 



first Himalayan storm occurred. 

 B — From the 30th June to the 11th August, that is, from the first 



Himalayan storm up to the beginning of the second. 

 C — The three to four weeks which follow the 11th August, and 

 during which the ' remarkable series of storms ' moved 

 from the Bay of Bengal north-westward to the extreme 

 west of India. 

 D — The remaining part of the season, which I consider began 

 with the storm which early in September broke up over 

 the south-west of the Province instead of moving west- 

 ward as the various members of the ' remarkable series ' 

 did. 

 During each of these periods we have a well-defined behaviour 

 of the cyclonic storms, and a well-defined distribution of rainfall. Also 

 the connection between the line of advance of the storms and the 

 prevalence of monsoon conditions is so striking that the study of the 



