AN ISLAND IN THE CHILKA LAKE., 281 



commonly in any other part of the island. The Pongamia trees in this association 

 provide a remarkable contrast to those of the Nim and Crataeva in that they are 

 almost always entirely free from creepers, only an occasional spray of Derris 

 stretching out over them from an adjacent tree of one of the other species. This 

 appears to be due partly to their spreading habit and dense foliage, which prevents 

 other plants from springing up in their shade. It cannot, however, be the only 

 explanation, as the woody stems of both of the common Leguminous creepers 

 often extend along the sand for great distances before they grow upright. I once 

 traced a stem of the kind on the surface for 75 paces on the sand outside the Pongamia 

 trees. It ran parallel to them and then mounted on a Nim . 



On the more exposed eastern shore of the island the ring of Pongamia is more 

 frequently broken by intrusions of Crataeva and Nim, while in the more stony 

 parts of the margin of the lake trees of Salvadora persica are not uncommonly found 

 in the same situation. Where the rock is quite solid Pongamia glabra cannot 

 establish itself. 



In years of high flood-level the roots and lower branches of the trees are often 

 completely submerged for several months in brackish water. In these circum- 

 stances the submerged branches die, but the rest of the tree apparently does not 

 suffer. As the floods retire they leave the dead branches festooned with plants of 

 Potamogeton pectinatus and filamentous algae. 



There is an almost complete absence of herbaceous or bushy undergrowth in the 

 denser parts of this zone, but at a few spots little thickets of Datura fastuosa 

 establish themselves on its outer edge in the dry season. As the floods rise and the 

 plants are sprayed with brackish water, their young shoots undergo a peculiar 

 modification and become densely covered with very small, deeply incised leaves. At 

 other places a line of Jatropha grows up in a similar situation. Its leaves are always 

 small and distorted, without being densely packed together as in the Datura ; but 

 this dwarfing and distortion of the leaves of Jatropha is a common phenomenon all 

 over the island, and also in many parts of Southern India in which no brackish 

 water is present. 



Vegetation of sandy areas within the Pongamia Zone. 



Towards the eastern end of the island and near the northern shore, inside the 

 Pongamia zone, a distinct association of shrubs and trees occupies a considerable area 

 of sandy and gravelly soil. Indeed, this association can be traced all along the north 

 side, but it merges gradually into the one that I will describe after it, the most 

 conspicuous feature being the gradual substitution of Weihea ceylonica for Glycosmis 

 pentaphylla in the sandy ground, and the entire absence therefrom of trees of the 

 genus Ficus. In the area now to be considered the vegetation is not so dense as in 

 some other parts of the island and consists mainly of small, more or less isolated 

 clumps of bushes {W . ceylonica) and large trees. Crataeva religiosa here reaches a larger 

 size than in the Pongamia zone and shares with the Nim the preponderance among 

 the trees. Odina Wodier is also common and a few trees of Albizzia odoratissima 



