100 THE SCOTTISH NATURALIST 



of skins are available for comparison. There is a good 

 deal of divergence of opinion as to what constitutes a species 

 or a subspecies, and no two lists, indeed hardly any two men, 

 agree on this thorny matter. In this paper we shall, as is 

 our wont, follow Dr Hartert, and include two forms which 

 have not been differentiated in the new B.O.U. list. These 

 are the Continental Goldcrest and Hedge-sparrow, which 

 in our opinion are readily distinguishable from the 

 British races. 



We propose to give a brief account of the status, in 

 Scotland, of the races dealt with, along with a description of 

 the difference between them and our Scottish breeding birds, 

 except in the case of forms which have frequently been 

 described before. To this we shall add any field notes 

 which we have made on the habits of these races. When 

 there is a big rush of immigrants from the Continent on an 

 isolated station like the Isle of May, and when Continental 

 Thrushes, Robins, etc., have been killed at the lantern, it is 

 ^ fairly safe to assume that those seen on the island next day 

 are also arrivals from overseas. 



The Redpolls are a very puzzling group ; the differences 

 between them are small, and the difificulties are increased by 

 transition forms. The Mealy Redpoll {Cardiielis linaria 

 linaria), which breeds from the far north, south to the 

 Baltic, is a fairly common winter visitor to Scotland. In 

 some seasons it occurs in very large numbers, the largest 

 immigration of recent years being in 1910. It is larger, 

 and the brown on the back is less red than in our resident 

 Lesser Redpoll {Carduelis linaria cabaret). A doubtful 

 and rather unsatisfactory form is Cardiielis linaria holboelli : 

 it nests in the Polar regions of the Old and New World, and 

 is stated to be larger, the wing longer (74-78 mm. in the 

 Mealy Redpoll, 75-81-5 mm. in Holboell's Redpoll), the bill 

 stronger and usually longer than in C, I. linaria (8 mm. in 

 the type, 9- 11 in the other). Hartert says there are transi- 

 tion forms between these two Redpolls, and that they often 

 breed in the same locality, which makes him doubtful 

 whether Holboell's Redpoll should not be accorded specific 



