LITTNEAIT SOCIETY OF LONDOIT. 21 



I have drawn up the appended Tabular List (pp. 22-23) to show 

 the Geographic Distribution of the Characteristic Vegetation o£ 

 the Soondreebun : this list, containing 69 species, is of course only 

 a portion, perhaps not one-sixth, of the plants which may be 

 collected wild in the Soondreebun. 1 drew up this list to include 

 those species which I can collect in the Soondreebun, but cannot 

 collect in the Bengal Plain for 100 miles outside the Soondreebun. 

 The Bengal Plain Plora is very well known to me ; I think 

 few of my 69 characteristic Soondreebun plants will hereafter be 

 found in Bengal out of the Soondreebun. I have discarded 

 many species, either because of their being doubtfully indigenous 

 in Bengal or often planted elsewhere ; or, more often, because 

 they had closely allied forms in various other countries which 

 might be regarded as different species, varieties, or geographic 

 races. For my method of treatment, only such species are of 

 value as are well-defined and easily determined in herbaria. 



The flora of the Bengal Plain may comprise 600 species, a very 

 weedy set, of which possibly three-fourths are to be found within 

 the Soondreebun area. If I had included all the species found 

 in the Soodreebun, the quantity of this weedy, widely scattered 

 Flora is so great that m my tables it would overlay and com- 

 pletely obscure the distribution of the characteristic plants. The 

 present way of dealing with the affinities of a Flora proves, in the 

 case of the Soondreebun, successful from my point of view, i. e. it 

 destroys all anomalies in geographic distribution. When I com- 

 menced this paper, I thought the plan of selecting the characteristic 

 plants and examining the distribution of these only in neigh- 

 bouring countries was novel. But I find it is hardly so ; at least 

 it has been frequently employed on a small scale, as by the 

 French botanists in studying the distribution of the Sea-coast 

 Flora of France. 



For the 69 selected characteristic species I have tabulated 

 all the examples in the Kew Herbarium, in which collection the 

 flora of the coasts is generally more fully represented than that 

 of the interior. This Flora being essentially a muddy sea-coast 

 flora, extends along the shores of the warmer parts of Asia, 

 Polynesia, and Africa. After trying, not very successfully, to 

 exhibit it by a map, I have constructed the appended Table, which 

 rej)resents the distribution with tolerable completeness, though 

 it represents that distribution in longitude only. The vertical 

 lines are lines of longitude which I have drawn at the points 

 where i found the best marked " breaks " in the distribution. 

 The five central columns represent the distribution in India — 

 i. e. the central column, the Soondreebun; the first column east, 

 includes Chittagong, Arracan, Pegu ; the second one east, Malay 

 Peninsula ; the first column west, the Coromandel coast ; the 

 second column west, the Malabar coa^t. The Malay column 

 turther east necessarily (from the linear character of the tabu- 

 lation) throws South China, Malaya, and North Australia into 

 one ; and the East African column includes the coast from 

 Abyssinia to Natal together with Mascareuia. 



