under surface pale rufous, everywhere streaked with dusky blackish shaft-lines, less marked on 

 the vent and under tail-coverts, the long ones of which end in dark blue or blue-black like the 

 upper tail-coverts ; sides of body and flanks washed with a little deeper rufous ; axillaries and 

 under wing-coverts clear rufous, with scarcely any shaft-lines, the edge of the wing more dis- 

 tinctly streaked with the latter; quills dusky below. Total length 8 - 4 inches, culmen - 4, 

 wing 5 - 15, tail 4"7, tarsus O'fio. 



Adult female. Similar to the male in plumage. Total length 7T> inches, culmen 035, wing 4'75, tail 4'5, 

 tarsus 055. 



Young. Duller in colour than the adult, and easily distinguished by the rufescent margins to the tips 

 of the wing-coverts and secondary quills ; rump more coarsely striped than in the old birds ; ear- 

 coverts nearly uniform sooty brown ; rufous colour on sides of hinder crown very dull and less 

 developed than in the adults. Wing 4 - 35 inches. 



Hub. Eastern Siberia, wintering in Tibet and Mongolia, and also in Assam. 



The number of species of Asiatic Mosque-Swallows lias been a subject of discussion for 

 many years, aud even now we cannot regard the present state of our knowledge with 

 any great satisfaction. After several attempts on Mr. Swinboe's part to define the 

 Chinese species, Mr. Allan Hume wrote a capital article on the Indian species of the 

 group, aud Mr. Henry Seebohm, in 1883, made a further contribution to our knowledge 

 of the subject. In 1885 we had to describe the species of the H. daurica section of 

 Swallows in the ' Catalogue of Birds,' and our conclusions principally agreed with those 

 of Mr. Seebohm. The year 1890 has been remarkable for a further exposition of the 

 Indian species, and this took place in the Natural History Museum, Avhen Mr. E. W. 

 Oates was able to lay out on the table a goodly series of specimens from the Hume 

 collection, such as had never before been available for any European naturalist to work 

 with. To this series of skins Mr. Seebohm brought his Japanese and Chinese examples, 

 and found that Mr. Oates's conclusions were correct, and they are in the main adopted 

 by Mr. Seebohm in his ' Birds of the Japanese Empire.' We have also had the 

 advantage of the loan of Mr. Seebohm's specimens, and, with some slight modifications, 

 we have adopted the opinion of the two naturalists above mentioned, though we still 

 maintain our conviction that to draw a hard-aud-fast line between the four races of 

 Oriental Mosque- Sw r allow r s is nearly impossible, so much do they grade towards each 

 other both in size and colour. 



Four races may, however, be recognized, of which two are large and two small, two 

 rufescent underneath and two whitish. Mr. Oates fixes the length of wing in the two 

 large forms, H. striolata and H. daurica, as from 49 to 5"3 inches, and in H. nipalensis 

 and II. erythropygia as from 4>5 to 4"7 inches. 



Mr. Seebohm divides the four races into two sections, relying on the coarseness 

 or fineness of the streaks on the under surface, and the presence or absence of shaft- 

 streaks to the rump-feathers. Thus in the first section of finely streaked birds he puts 



