THE ISLANDS OF THE BAY OF BENGAL. 177 



is so dirty that the yellow patch is quite invisible, and the wings 

 are so smeared with gum, that the wing patch looks duller 

 and darker than it should be in magnirostris. So that I am 

 not perfectly certain that the Coco's bird is identical with that 

 from the Andamans and Burmah, however I think it is. I 

 should add that the bills of the Andaman birds run somewhat 

 larger than those from Thyet Myo. A fine male from the 

 latter locality has the bill measured as before, 1/4 and 0*85. 



I may be wrong in uniting the Burmese and Andamanese 

 birds, but they are precisely similar in color, and only differ 

 a little in the size of the bills. 



Davison tells me that " this species is tolerably abundant at 

 the Andamans. It keeps in parties of from five to twenty or 

 more, frequenting the tops of the highest trees, keeping up a 

 continual screeching, even when feeding; their note is very 

 distinct from both the other species occurring at the Andamans, 

 viz., P. affinis and P. fasciatus, and when once heard can 

 never be mistaken for either of these. Beyond Viper Island, 

 on the way to Dunnyleaf Creek, is a roosting place of these 

 parrots. This is in the mangroves bordering the entrance to one 

 of the smaller creeks ; the birds have evidently used this place 

 for years, and have made a large portion of the tops of the 

 mangroves look as if they were kept trimmed with the most 

 scrupulous care. I was informed by some native convicts that 

 this was the common roosting place of all three species of 

 parrots; but I watched the birds, returning home one evening 

 for nearly an hour, during which time some 50 or 60 parties 

 of 5 or 6 to 30 or 40 must have passed, and certainly the only 

 ones that I saw consisted unmistakably of the present species. 

 I have continually seen P. affords in small parties settling for 

 the night in moderately low bushy trees in different places 

 about Port Blair, and, on the other hand, I have as often seen 

 P. magnirostris assemble in parties in some large tree just 

 before dusk, and after making a tremendous noise for some 

 time, fly off in the direction of Dunnyleaf Creek. I did not 

 observe this on any of the Nicobar Islands ; but it occurs on 

 both the Cocos and Table Island." 



This species is a permanent resident in the islands ; we 

 obtained specimens from December to April, and have had speci- 

 mens sent to us, which were killed from June to September. 



148. — Palceornis torquatus, Bodd. (0.) 



This species was introduced by Colonel Tytler, but it has entirely 

 disappeared. I do not therefore include it in our list. Not only did 



