14 



consideration and liable to be more or less denuded of water 

 periodically, and (b) a western division, carrying depths of 3 

 to i\ feet over the greater part of its area. Marginal shallows 

 carrying one foot and less of water are greatly in evidence 

 along the south coast, from Sunnambukulam to Mankodu and 

 thence to Annamalaicheri ; also from off the latter village 

 southwards to Pulicat North Island. All along these sections 

 of the lakeshore, light-draft fishing boats have to be moored 

 to poles driven into the mud at distances of from \ to \ mile 

 from land on account of this extreme shallowness of water. 



The shaded sketch plan appended shows graphically these 

 several points. 



The levels described above represent the lake under normal 

 conditions ; during the monsoon season, from October to 

 December, when two-thirds of the annual rainfall occurs, the 

 water level tends to rise considerably and whenever a closure 

 of the bar coincides, the flood level may rise several feet above 

 the normal. Conversely, if the bar remains closed during an 

 exceptionally dry season, April to June, the lake acts as a huge 

 evaporating basin and has been known to fall 3|- feet below 

 mean sea level (May 1391); reference to the sketch plan 

 will show how greatly reduced the water surface of the lake 

 becomes at such a time, the shrinkage entailing the drying up 

 of between 60 per cent, and 70 per cent, of its normal area. 



Unlike the majority of bars on this coast south of the 

 Suvarnamuki river, that at Pulicat does not close annually. 

 Hitherto its closure has been on an average once in five years. 

 The year of closure usually occurs w r hen the bar has shifted 

 to the extreme north end of its run and is stopped from further 

 progress in that direction by increase in the height of the 

 beach. 



Whenever this occurs a considerable head of flood water is 

 required during the north-east monsoon to scour deeply the 

 channel of the bar or to form a new breach at the south end 

 of the run. With a bar at the extreme north end and weak 

 monsoon floods, conditions which appear to coincide approxi- 

 mately every fifth year, a closure of the bar is certain in the 

 ensuing hot weather. 



Within recent years the bar closed in 1890, 1895 and 

 1905. 



An outline of the history of the first of these closures may 

 be considered as typical of the series ; as Kussell, in his 

 " History of the Buckingham Canal Project," pages 26-27, 

 gives an excellent account of the consequences, I prefer to 



