110 D. Prain — Fauna of N&rcondam avJ Barren Island. [April, 



species is the Andamanese form of the widely distributed Tree-Hal 

 [Mus rattus var. andamanensis.)* 



In Narcondam at dusk a frugivorous bat was observed flitting from 

 tree to tree, and in Barren Island the writer came upon an example of 

 T&umphii, on the outer aspect of the outer cone and about half a 

 mile to the north of the lauding place at the hot-spring, which was the 

 home of a colony of these creatures. The writer was able to reach a 

 point on the cliff almost overlooking the tree and where he was not 

 more than twenty-five yards away from the bats. He was able to see 

 that they were of a very dark brown (almost black) colour but, having 

 no fowling-piece, he could not secure a specimen. f It is not improbable 

 that, as in the case of the rats, the bats on the two islands belong to 

 the same species. 



Goats have more than once been landed on both islands. On Nar- 

 condam there were however none to be seen and when one recollects 

 that for a considerable portion of the year the island is absolutely water- 

 less the fact is not surprising. And those formerly landed on Barren 

 Island must also have perished, for a few years ago, Col. Tucker, then in 

 charge of the Andamans, found it necessary to land five more. During 

 our visit we several times saw three goats in company and as we dis- 

 covered among the lava blocks the skeletons of two others it is not im- 

 possible that we were thus able to account for the whole five. The three 

 seen together were all adults, perhaps therefore they have either not 

 begun to breed or their young have been unable to survive. A well- 

 trodden foot-path across the lava causeway near the landing-place 

 shows that the goats come daily, at least during the dry season, to drink 

 at the only spring the island possesses, hot-water charged with over 200 

 grains per gallon of saline matter.J Each morning during the writer's 



* Catal. of Mammalia in the Indian Museum, pt. ii, p. 63. 



f Hud a specimen been shot it probably could not have been secured, for a direct 

 descent from the writer's position to the foot of the tree was impracticable and to 

 reach the spot where the specimen must have fallen would have entailed : — 1, a return 

 to the landing-place (two hours work to begin with) ; 2, rowing half a mile to the mouth 

 of the gulley in which the tree stood ; and i grant ing that t be sea-month of the proper 

 galley was hit off and, if it was, that a landing could have been effected — neither 

 event being at all a necessary consequence,) 3, a search for the particular tree in- 

 volving a climb of over 800. feet through a particularly difficult jungle with much 

 the same prospeofcof ultimate suooess thai there would be in the proverbial search 

 for " a needle in a haystack." But this should only wh<i t he appetite of tho true 

 collector for a visit to tho islands ; it will however show that the geologist or 

 botanist who visits the place with but a limited time at his disposal, must confine his 

 attention to hit own subject and can do bat dilettante work in any other direction. 

 J Memoirs of the Qeol. Survey "/India, vol. xxi, p. 277. 



