18 



8 



and abundant birds throughout the country, congregating in lai'ge flocks, and breeding wherever 

 they can find suitable spots. They are most partial to large buildings, such as churches, pagodas, 

 mosques, tombs, and the like, frequently entering verandahs of inhabited houses, and building 

 in the cornices. Holes in walls of cities or towns, too, are favourite places ; and in some parts of 

 the country they prefer holes in wells, especially, I think, in the west of India, the Deccan, 

 &c. In default of such spots, they will breed in crevices and cavities of rocks, caverns, and 

 sea-side cliffs; and I have often noticed that they are particularly partial to rocky cliffs by 

 waterfalls. The celebrated falls of Gairsoppa are tenanted by thousands of Blue Pigeons, which 

 here associate with the large Alpine Swift. It is more rare in forest countries generally than in 

 the open country. It extends from Ceylon throughout India to the Himalayas, and also to 

 Assam, Sylhet, and Burrnah. It is doubtful if it occurs in Afghanistan, or in other parts of 

 Central Asia. These Pigeons are held in favour by most natives, and almost venerated by some ; 

 and if they build in the house of a native, he considers it a most fortunate omen. They are, 

 however, very destructive to grain, assembling in large flocks in the cold weather; and in general 

 the natives do not object to their being shot. They are undoubtedly the origin of most of the 

 domestic Pigeons of India." How far south it ranges I cannot say with any degree of certainty. 

 Mr. Holdsworth cites two localities in Ceylon which this Pigeon inhabits — Pigeon Island, near 

 Trincomalie, and Berberyn, not far from Galle ; and Layard mentions their having been killed 

 about fifty miles inland from Trincomalie. There is also in the late Lord Tweeddale's collection 

 a specimen from Java which is very dark in colour, and has blotches of blackish slate on the 

 lesser wing-coverts, the rump being greyish white. In the British Museum there is, however, 

 one example from India which agrees closely with this bird ; and, judging from Professor 

 Schlegel's description, examples from Japan must somewhat resemble it. In China, according 

 to Pere David, both the present species and Columba rupestris are found. He says that he 

 found the Rock-Dove in Northern China, and observed large numbers inhabiting caverns at 

 great altitudes in the Tsinling Mountain ; and it is said to be common in Japan. In Dauria 

 and Southern Siberia generally Columba livia is replaced by what appears to me to be a 

 fairly good species, Columba rupestris, Bp., which is certainly the bird referred to by Pallas 

 (Zoogr. Eosso-As. i. p. 560) as variety § of Columba osnas, under the name of Columba rupestris. 

 This bird may always be recognized by having a broad subterminal white band across the tail, 

 being otherwise coloured as ordinary British examples of Columba livia. The bird figured by 

 Henderson and Hume (Lahore to Yark. pi. 31) as Columba eversmanni, Bp., appears to me to be 

 fairly distinct. There are two specimens in the late Lord Tweeddale's collection which are 

 distinguishable in having the rump bluish white, the crown and breast richly tinged with 

 vinaceous, and the upper wing-bar very indistinct, being merely a small patch or two small patches 

 of black. In one specimen, which was obtained by Severtzoff, and is labelled by him Columba 

 bracliyura (fusca, Pall.), the back is also tinged with vinaceous. 



Essentially a rock-hunting species, the name by which this bird is usually known is a very 

 appropriate one. With us in Great Britain it is almost exclusively found on the rock-bound 

 portions of our coasts, whence they make excursions to the cultivated portions of the country 

 in search of food, often collecting in large flocks in the winter season. It feeds on grain of 



