688 Memoirs of the Indian Museum. [Voi,. V, 



density is seen in either sections I or 3 so the condition is obviously local, and it 

 seems to me that the most probable explanation is that there is in the neighbourhood 

 of this station an under-water spring. If this be so, the spring appears to be an 

 intermittent one, for no trace of it is visible in the results obtained in August, 1919. 

 We now find that in the channel between Cherriakuda and Barkuda Islands there is 

 an area of higher salinity and raised temperature. The sample was taken at 12-20 

 p.m. on the 17th April, 1920, when a strong breeze was blowing across the bay from 

 the S.S.W. and it seems to me that we have here corroborative evidence of the view 

 put forward by me above that a strong breeze from this direction in addition to 

 causing a surface outflow from the bay also sets up a counter-current flowing along 

 the bottom into the bay from the shallow area enclosed by Cherriakuda, Barkuda, 

 and Samal Islands. 



A sample of water sent to me by Dr. Annandale in June, 1920, showed that yet 

 another great change had taken place in the density of the water. This example 

 was taken at 10-30 a.m. at the end of the jetty on Barkuda Island and at the time a 

 strong breeze was blowing from the S.S.W. Unfortunately, no record of the temper- 

 ature was taken simultaneously, but the density of the sample at standard tempera- 

 ture was 9-66 so that it is obvious that there had been a great increase in the sali- 

 nity of the water in the lake since April. 



A comparison of my observations in 1919 and 1920 with those made by Annan- 

 dale and Kemp in 1914, shows clearly that the conditions existing in Rambha Bay 

 during these two periods were vastly different. At the commencement of my work 

 in August, 191 9, my results agree very fairly well with those obtained five years pre- 

 viously ; but between August, 1919, and June, 1920, changes occurred in the Lake that 

 were quite unlike anything observed in 1914. I have already given Annandale and 

 Kemp's account of what, judging from their experience, is the normal sequence of 

 events in the lake during successive seasons of the year, and a comparison with my 

 results show that the abnormal features present during the end of 1919 and early part 

 of 1920 are : (1) the stead} 7 and progressive diminution in the density of the lake-water 

 long after the close of the monsoon season, so that in December the density was only 

 a little over 2*0; (2) the steady fall in the level of the lake up to and possibly even 

 beyond April, 1920 ; and (3) the subsequent rapid rise in salinity in May-June, 1920. 

 These changes in the density and salinity of the water in the lake produced a very 

 noticeable effect on the fauna, which Dr. Annandale has dealt with in the second 

 part of this paper, and we are here merely concerned with the changes themselves. 



As I have already mentioned the height of the lake was steadily rising through- 

 out the whole period of my survey in August, 1919, and for some weeks afterwards. 

 As Annandale and Kemp have pointed out, by far the most potent factor in producing 

 changes of level and reduction of density is the monsoon flood- water brought down by 

 the Mahanaddi river-system and poured into the north end of the lake, and as a result 

 of this influx the water in the greater part of the north-east end of the lake becomes 

 fresh or almost fresh. According to their view (loc. cit., p. 8) "the great volume 



