1891.] D. Prain— r^e Vegetation of the Coco Group. 365 



as well as physiographically the Coco Group forms an integral part of the 

 Andamans. Farther, it shows that of the possibly predominating elements 

 in their flora, the Indo-Chinese element, as a whole, is slightly weaker 

 even in that part of the Andamans nearest to Burma than is an element 

 indicating a Tenasserim influence and an element indicating a Malayan 

 influence. This seems strange when we recollect that not only do the 

 Cocos form that part of the Andamans nearest to Barma but that there is a 

 shallow ridge, at times raised into islands, along the line between the 

 Cocos and the nearest point on the Burmese mainland, whereas Tenas- 

 serim is at the opposite side of a deep sea, while Malaya is separated from 

 the opposite extremity of the Andaman group by a much greater distance 

 and by much deeper straits than Burma is from the area under discussion. 

 In order, if possible, to account for this peculiarity of distribution, it 

 becomes necessary to discuss the probable origin of the flora of the group. 

 The first step in such an inquiry is to ascertain the species in a flora 

 that may possibly have been introduced and that do not therefore 

 necessarily postulate for an isolated locality such as the Cocos any former 

 connection with neighbouring land. It is, of course, evident that if a 

 previous land connection be shewn to be necessary to explain the pre- 

 sence of any species in the islands this same land connection would 

 sufficiently explain the presence of most of the species that occur there 

 without requiring the suggestion of any extraneous means of introduc- 

 tion. But until all the possibilities of introduction by means of physical 

 agencies now at work under existing physiographical conditions are com- 

 pletely exhausted, we are not at liberty to assume the existence of dis- 

 similar physiographical conditions or a different application of the present 

 physical agencies. 



There is, however, always great difficulty in deciding absolutely 

 what species are indigenous and what species are introduced in any 

 locality, and here no species will be considered "indigenous" for which it 

 is possible to suggest in the remotest fashion any means of introduction. 

 At the risk thei-efore of including among introduced species many that 

 are probably quite entitled to be termed indigenous, the possibilities are 

 discussed under the headings of the various active introducing ao-encies. 

 As this involves a use of the terms " indigenous " and " introduced " 

 somewhat different from the sense in which they are generally accepted, 

 it seems better that the possibly introduced species be spoken of as 

 "migrant"; and the certainly indigenous residuary species termed 

 " remanent," many of the " migrant " species being doubtless perfectly 

 " indigenous " in the generally received sense. 



Even within the group of " migrant " Species difficulties often arise 

 owing to certain species being assisted in one way from island to island 



