THE DEOCAN AND SOUTB MAHRATTA COUNTRY. 379 



the species as I watched them closely for several days, being- 

 unable to shoot them on account of their being in barracks, 

 and there is no other Martin for which they could have been 

 mistaken. In all probability they were in course of migration 

 from the Nilgiris, where they are known to occur. I believe 

 I saw it also at Singhur some years ago when I first arrived in 

 India, but cannot be quite certain, as I did not take the same 

 interest in ornithology in those days that I do now. It is 

 easily recognised by its short, forked tail, broad white rump 

 band, glossy blue black upper surface and snow white lower 

 surface ; legs, feet, and toes feathered white. 



98. — Cypsellus melba, Lin. The Alpine Swift. 



Cold weather visitant, as far as my experience goes. Com- 

 mon all along the Ghats, frequenting high cliffs, in which it 

 roosts, and descending into the surrounding plains during the 

 day time to feed. In Belgaum, throughout the cold weather, 

 large flocks may be seen every morning passing over the station 

 between 6 and 8 a.m. in an easterly direction, and returning 

 again in a westerly direction to the hills to roost just before 

 dusk in the evening. Mr. Davidson says that it is a per- 

 manent resident at Satara, and he thinks breeds there on the 

 cliffs. This may be the case as it is known to breed on the 

 Nilgiris, and, if so, it may also breed at Singhur and on other 

 high cliffs along the Ghat range, but as yet the fact has not 

 been proved. 



100.— Cypsellus affinis, J. R Or. The Common 

 Indian Swift. 

 Permanent resident. Common throughout the region. 



? 101.— Cypsellus leuconyx, Blyth. The White- 

 clawed Swift. 



Entered in Mr. Fairbanks list of Mahratti species, but on 

 what authority, unless Jerdon's, I do not know. No other 

 collector has recorded it, neither have I myself ever come across 

 a specimen ; I have therefore entered it as doubtful. If it does 

 occur, which appears to be the case according to Jerdon, it 

 must be a rare bird. Jerdon says, u a rare species. I obtained 

 one specimen on the western part of the Deccan and several 

 in Malabar, where it frequents rocky hills/' I have no other 

 record of its occurrence. 



102.— Cypsellus batassiensis, J. E. Gr. The Palm 

 Swift. 



Permanent resident. Common in some localities where there 

 are groups of palmyra, areca or cocoanut palm trees ; in other 



