Mr. H. Seebohm on the Ornithology of Siberia. 343 



and China, and find that in every case where the large under 

 tail-covert has not been shot away this distinction holds good. 

 This bird breeds in considerable numbers on the tundra as 

 far north as we went. There is a great variation in the 

 colours of the eggs in the same nest, some being much darker 

 than others. There are skins of this bird in the Museum at 

 St. Petersburg, collected by Baron Maydell in the Tschuski 

 Land. 



ANTHUS RICHARDI, Vieill. 



Richard's Pipit must breed in great numbers on the exten- 

 sive meadow-lands which stretch away for miles from Yen-e- 

 saisk' on the banks of the river. I found it common there 

 in the middle of August, and shot both adult birds in full 

 moult and young in first plumage. This bird has a habit of 

 hovering over the ground almost exactly like a Kestrel. 



ANTHUS TRIVIALIS (Linn.). 



I did not meet with the Tree-Pipit until I reached lat. 62 

 on my return journey. 



MOTACILLA ALBA, Linn. 



Motacilla alba, Linn. Syst. Nat. i. p. 331 (1766). 



Motacilla dukhunensis, Sykes, P. Z. S. 1832, p. 91. 



Motacilla baicalensis, Swinhoe, P. Z. S. 1871, p. 363. 



I think there can be no doubt that M. alba and M. dukhu- 

 nensis are the same species. The only difference seems 

 to be in the amount of white on the wing-coverts. M. alba 

 has dark grey or black wing-coverts, more or less broadly 

 edged with white. In M. dukhunensis the inside half of 

 each wing-covert is the same as in M. alba-, but the 

 outside half is entirely white, making the wing-coverts, as 

 they lie on the wing overlapping each other, an entirely 

 white mass. This latter form seems to be confined to Siberia 

 and India ; but as in both these countries a complete series 

 of intermediate forms occur also, we cannot consider the 

 eastern form more than a variety. The amount of white 

 on the wing-coverts of many of the species in this genus, 

 and in some of them the amount of white on the secondaries, 

 varies so much, that if we were to admit it as a specific cha- 



