286 N. ANNANDALE. 



widely distributed in India. Among the most abundant are Aerua lanata, Boerhavia 

 diffusa, Mollugo stricia, Euphorhia thymifolia, lonidiimi suffruticosum, Vernonia cmerea 

 and Oldenlandia Hcynii. The last, however, is not so abundant as at the edge of the 

 thickets. 



There are two wells on the island, dug in the rock, and also, as already stated, 

 a small pond constructed in the same manner. At first sight the vegetation 

 immediately surrounding these, where the rock is damp, looks very different from 

 that of the cleared and levelled area, but on examination most of the plants are 

 found to belong to the same species but to be much more luxuriant and to 

 grow erect. V ernonia cinerea, for example, is rarely more than three inches high on 

 the bare ground, while at the edge of a well surrounded by territory of the kind it 

 grows to a height of at least two feet. This is also so with Aerua lanata, and in less 

 degree with most of the herbs I have mentioned. The sedge Bulbostylis barbata 

 graduall}^ assumes the form pulchella as the rock on which it grows becomes damper, 

 and in all the species transitionary forms can be found between the extreme 

 depauperated and fairly normal phases. 



A curious feature of the vegetation of dry cleared areas is that while the majority 

 of its species flourish, as might be expected, in wet weather, a few [e.g. Evolvulus 

 alsinoides and Indigo/era linifolia) appear in the dry season. 



On the rocks immediately round the pond a few plants are found that do not 

 occur elsewhere on the island. The most conspicuous of these is the sedge Mariscus 

 albescens, the roots of which are submerged in the rainy season. Its flowering 

 stems grow at least three feet high. This is the onl}^ Phanerogamous plant in any 

 sense aquatic found in ihe pond. 



Behind the bungalows and sheltered by them and by a circle of large fig-trees 

 [F. bengalensis, F. infectoria, and F. retusa), a small field was prepared for agriculture 

 some years ago by the removal of all indigenous plants and probably also of most of 

 the stones. The site was chosen as that in which the soil was most likely to be 

 fertile, but owing to some difficulty about land revenue nothing was ever planted. 

 A high, dense herbaceous thicket consisting mainly of Tephrosia ptirpurea and Crota- 

 laria striata, with scattered clumps of Datura, has sprung up on this area. In the wet 

 season the Tephrosia is the dominant species, but in the cold weather it begins to 

 die down and the Crotalaria assumes the i:)redominance. Both die down almost 

 completely in the hot weather. 



This type of vegetation is also found on neglected footpaths in those parts of 

 the island from which fig-trees have recently disappeared. It can only flourish at 

 spots which are unshaded through the day. Where the soil is shaded by trees it 

 disappears abruptly and its place is taken by thickets of Glycosmis and Anona. This 

 is shown very clearly in the lower photograph reproduced on plate XI. The photo- 

 graph had to be taken at early dawn in order to secure sufficient light under the 

 large trees of Ficus infectoria seen on the right of the picture. 



A further point that may be noted as to the interference by man with 

 the natural vegetation of the island is that a small herd of spotted deer {Cervus 



