AN ISLAND IN THE CHILKA LAKE. 279 



[E. neriifolia) also puts forth its large and handsome leaves in June, and the various 

 cryptogamic plants revive, while the more conspicuous herbs spring up afresh. 



Types of Vegetation on the Island. 



We may now consider the types of vegetation to be noticed on the island further. 



It will be convenient to discuss them in geographical order, commencing our survey 



with the foreshore and proceeding inland. I will, however, leave those types that are 



due to the interference of man until the last, discussing what we may call the natural 



types first. 



Vegetation of the Foreshore. 



As the flood-level of the lake begins to sink in autumn with the cessation of the 

 monsoon rains a wide stretch of foreshore is gradually exposed. Its character differs 

 at different spots, but it is everyv/here either rocky, gravelly or sandy and nowhere 

 muddy. The greater part of the sheltered northern shore is sandy, while the southern 

 shore is rocky with sandy patches. The eastern shore, the most exposed, is mainly 

 composed of gravel or very coarse sand, while the western shore is rocky. 



The vegetation of the foreshore is very scanty. Indeed, on its rocky parts there 

 is at most spots, practically speaking, no vegetation at all except that of filamentous 

 algae, with Potamogeton pectinatus in softer depressions ; and these plants of course 

 die down completely as the rocks become desiccated. Just below the highest flood-level, 

 however, one or two trees have managed to establish themselves among the rocks. 

 Their roots are washed by the waters of the lake for a short period in years of high 

 flood-level only. Of such trees the only one at all common is Salvador a persica, 

 which on Barkuda is only found in such situations.^ It grows to a fair size but 

 always has its trunk distorted and of abnormal form owing to the young shoots being 

 eaten by deer. A beautifully regular tree of Barringtonia aadangula, the only one 

 on the island, grows in a small patch of rocks on the south shore, while on the same 

 side two handsome specimens of Mimusops hexandra occupy a similar position. 

 Apart from lichens on the stones, the other plants at all characteristic of the upper 

 flood-level below the zone of Pongamia to which I shall have to refer later, are two 

 species of Phyllanthus, one of which (P. ? debilis), has, though small, a bushy 

 habit and an extremely tough woody stem and tap-root. I may mention here the 

 dwarfed dom.e-shaped plants of Canthium parvifl.orum (Plate X) that grow among 

 rocks near the shore, as their peculiar form is due partly to the leaves and young 

 twigs being killed off by salt spray, and partly to the attacks of deer. These plants, 

 however, as a rule grow well above high flood-level. 



The vegetation of the sandy and gravelly foreshore is much less conspicuous, if 

 more composite, than that of marginal rocks. It consists in the earlier part of the 

 year of a few small sedges and grasses, of a single abundant herbaceous plant {Cressa 

 cretica) of wide geographical distribution, and of occasional species that attempt 

 without success to establish themselves annually. Among these is Ipomaea hiloha, 

 one of the most characteristic forms in the vegetation of sand-dunes on the Indian 



1 Except for a few dwarfed specimens growing in tlie side of a well. 



