1903.] C. Little — On two remarkable rain-bursts in Bengal. 2\) 



it appears to me to be an established fact that tlie depression accom- 

 panying the rainfall of the IJtli June passed over Thibet in a westerly 

 or south-westerly direction, and that- at 8 a.m. on the morning of the 

 28th, the region of Gilgit whs near the front of the advancing wave. 



Before commenting separately on the information regarding these 

 storms preserved in the meteorological records, I will again point out 

 that from whatever direction the storms entered Northern India, it was 

 not from the Bay of Bengal. For several days before and after the 

 two dates, mentioned above, weather was unusually quiet over the Bay 

 and in one respect was in striking contrast to what is usual in disturbed 

 weather. At Diamond Island the most exposed of the observatories 

 on the sea coast easterly winds of greater or less strength, are an 

 invariable accompaniment of disturbance. During the two periods of 

 disturbance the direction was westerly day after day, which would 

 indicate that weather was more probably disturbed over the south of 

 Burma, than over the Bay, that is, if there were any disturbance in that 

 region. The unusually low wind velocity at Diamond Island is 

 sufficient in itself to prove that there was disturbance, neither over the 

 Bay nor in Burma, until some days after the events under discussion. 



Part II. 



The following tables give in the form which appears to me most 

 convenient for purposes of comparison, the data for the storms in succes- 

 sion. When weather is unsettled changes at different observatoi'ies are 

 often very irregular more especially when local storms are frequent as 

 appears to have been the case on both of these occasions. I have there- 

 fore given the average change for divisions containing four to six 

 observatories or even more. The number of stations for each province 

 or division is given in the rainfall tables. 



Storm of June 30th. 



The following tables I (a) and II (a) give the pressure changes from 

 June 27th to July 5th, and the variation from the normal in Assnm, 

 Bengal Proper, and on the northern coasts of the Bay, arranged with the 

 view of showing the southerly movement of the disturbance. As I have 

 already stated the pressure change is practically useless for this purpose 

 in the case of the June storm because it extended over India with great 

 rapidity. The fall on the 28th was general and it continued in the 

 north on the 29th. The recovery began on the 30th and extended from 

 Assam and Bengal Proper to Orissa on the 1st July, practically the only 

 evidence of south-westerly movement, afforded by the table. Table 

 II (a) shows that pressure was relatively high on the 27th June, and that 



