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



its waters are subject. This factor undoubtedly exercises a continual selective 

 influence on the animals of the lake and it is to it, in the main, that the special 

 interest of the fauna is due. 



The methods which we adopted in determining the salinity of the water are 

 explained on p. 17. 



As is pointed out below (p. 18) our investigations were made chiefly at two 

 periods, in each of which we attempted, so far as was possible, to visit all parts of 

 the lake. Observations were, of course, made at other seasons and we have in fact, 

 in 1914 and in previous years, paid visits to the lake in practically every month ; on 

 these occasions, however, our investigations were of a more or less restricted nature, 

 concerning only a portion of the area to be investigated. 



The more comprehensive surveys effected in the two periods mentioned above 

 were made respectively in the salt- and the freshwater seasons. In the first, in which 

 our observations extended from February 12th to March 18th, the entire lake was 

 filled with water of varying but comparatively high density, while in the second, 

 from September ist to September 23rd, the water throughout a great part of the 

 system was quite fresh, owing to the floods which enter the lake at the close of the 

 monsoon. 



The charts on p 9 showing the corrected specific gravity of the water will give a 

 good idea of the enormous variation in density at these two periods. It is of course 

 improbable that they represent the maximum and minimun with any exactitude. 

 Somewhat higher specific gravities are doubtless to be found in early summer, that 

 is to say in the period immediately preceding the monsoon, and subsidiary investiga- 

 tions made in July tend to prove that this is 'the case. It is also possible that the 

 general density indicated in fig. 2 is capable of further reduction in exceptionally 

 high floods so far as the southern end of the lake is concerned. 



In giving an account of the general configuration of the lake-system, we noticed 

 that it could be divided into two parts, the main area, which comprises the bulk 

 of its waters, and the outer channel that forms the communication with the sea. 

 This division is not founded entirely on geographical considerations ; there are also 

 very marked differences in the range of density of the water in the two regions and 

 therefore notable faunistic distinctions. 1 The division is consequently based on both 

 physical and biological features of considerable importance. 



In March we found that the specific gravity of the sea, taken at a point some 



miles below the mouth (and therefore, owing to the 



Salinity of the outer channel. , . ^ ... . ... 



strong north-easterly currents that prevail along the 

 coast, uncontaminated by any discharge from the lake) was 1-0270. An additional 

 observation made a few days later just inside the mouth gave a reading of 1-02825. 

 At this period there was no appreciable outflow from the lake and the water in the 

 channel over an area extending from Barnikuda Id. to Arakhuda yielded specific 



1 The prevalence of a sandy bottom over a large part of the outer channel must also of course be 

 taken into account in considering the faunistic differences. 



