This deep sea that surrounds the islands everywhere but on the north, shows that, so far as need be taken into account for present purposes, they have never been connected with the Malay Peninsula or Sumatra—a condition that is further shown by the almost total absence of any members of the Malayan fauna—although they may at one time have been a prolongation of the Arakan Hills.
"It cannot, however, be asserted that this latter theory of connection derives, primâ facie, much support from a consideration of their fauna; and if they ever were in uninterrupted communication with the Arakan Hills it must apparently have been at an immensely distant period, for not only are all the most characteristic species of the Arakan Hills, as we now find them, absent from the islands, but the latter exhibit a great number of distinct and peculiar forms, constituting, where the ornis is concerned, considerably more than one-third the number known."—Hume, Stray Feathers, vol. ii.
From the above details, it is to be inferred that not only have the Nicobars—if ever in connection with the mainland—been longest separated, but that they have also been disconnected among themselves for a great extent of time. At a later period the Andamans were cut off from the continent, and the process by which they have been broken up into islands is—except in the cases of Narkondam and Barren Island—comparatively recent. This theory is fully borne out by the greatly localised nature of the fauna, nearly every island possessing its own peculiar species of terrestrial mammals.
Mammals.
The mammalian fauna of the Andamans and Nicobars is now known to consist of 35 positively identified species, 1 sub-species, and 4 others whose status is still doubtful.
Of this total of 40 animals, 19 are found in the former (if we leave out a dugong, which, though at present reported from the Andamans, will certainly be found to occur in the Nicobars), 22 in the latter. Only two species are common to both groups, and both these are bats—Pteropus nicobaricus, a wide-flying species found also in the Malay Peninsula and Java, and P. vampyrus—of which further knowledge will doubtless show that each group possesses its own variety.
To the Andamans 12 species are peculiar, the others being Mus musculus; Felis chaus, whose identification is doubtful; 4 bats; and a monkey, Macacus coininus, in all probability introduced.
The Nicobars possess 14 peculiar species and 1 sub-species, and the remaining members are Mus alexandrinus, and 6 bats.
Not only is the peculiarity marked among the terrestrial, but among the winged animals, which form so large a part of the fauna; also, of the 7 bats occurring in the Andamans, 3 are endemic, while the same is the case with 5 of the 11 in the Nicobars.
Thus it is to be noted that in the Andamans all the 11 terrestrial mammals—except M. musculus, M. coininus (introduced?) and the doubtful F. chaus—are peculiar, and also 3 out of 7 bats; while in the Nicobars, only 1 species—M. alexandrinus—of 10 terrestrial is other than endemic, and of the 11 bats 5 (nearly half) are peculiar. Remarkable as is the state of things with regard to the terrestrial, it is equally notable where the flying mammals are concerned.