903.] C. Little — On two remarkable rain-bursts in Bengal, 53 



the Darjeeling Hills. It broke up on reaching the Himalayas, causing 

 heavy rainfall in the eastern part of the range. A slight residual 

 depression moved eastward into Assam. 



It will be seen from the above extracts and remarks that the storms 

 from the Bay during the monsoon season fall clearly into the four classes 

 I have formed. Two in period (A) moved northwards; two in period 

 (B) moved in the usual west-north-westerly course, but recurved over 

 Central India towards the Kumaon Himalayas ; three in period (0) 

 moved west-north-westward, and without recurving passed over Guzerat 

 and other parts of Western India where rain was much needed ; and 

 that the two storms in period (Z>) moved into Bengal; thus showing 

 that, whatever the determining cause of the line of advance of these 

 storms may be, it was in September becoming similar to what obtained 

 in period {A) that is at the beginning of the season. 



The question therefore is what cause would be sufficient to account 

 for the change of motion in its various degrees shown by these storms 

 of the past monsoon season. I know of only one, and that is the westerly 

 wind overhead which is believed to cease before the monsoon commences, 

 but which may continue in the higher levels after it has ceased near the 

 ground. I was watching this wind very carefully last year, and believe 

 it still existed over Lower Bengal as late as the last week of June, 

 because typical nor 'westers occurred about that time. I believe also 

 that the north-westerly wind returned earlier than usual at the end of 

 the season, and was stronger than usual or in some other way differed 

 from what it is in ordinary years, and my reason for thinking so is that 

 nor'westers occurred in October in Western Bengal, a most unusual 

 event. 



Another question is, why should the two storms from Central Asia, 

 which I have discussed in the second part of this paper, influence that 

 westerly current. I am unable to say why it should be so, but I think 

 there can be no doubt, but that seasonal currents are often materially 

 altered by what for want of a better word I will call the shock of a 

 storm. The Rangoon cyclone early in May supplies an example of such 

 a change. An examination of weather charts for April and May last 

 year will show that the wind directions on the Burma coast were north- 

 erly up to the occurrence of that storm and that afterwards they had 

 generally the south-westerly direction usual in the monsoon season. 



The permanent change in the wind system on the Burma coast then 

 produced is none the less instructive, because the north-westerly winds 

 in April and May are believed to be a continuation of the very current 

 which appears to me to have so much to do with the advent of the 

 monsoon in different parts of North ern India, and in reasoning that the 



