401 



says, it is of very rare or casual occurrence, but has been met with in Hudson's Bay and North 

 Carolina. 



It is somewhat remarkable that the breeding-haunts of this Goose are unknown. It doubtless 

 breeds in Greenland, and also, as would appear from Mr. Eaton's notes, in Spitzbergen ; but, so 

 far as I can ascertain, there is no authentic account of its nidification, and its egg is still a 

 desideratum in collections. Some time ago Mr. Collett sent me two eggs which he believes to be 

 those of this species ; but positive proof is wanting. He wrote (Orn. N. Norw. p. 93) respecting 

 these eggs as follows : — " The occurrence of this species, almost sporadical as it is, in the Arctic 

 regions of Greenland, Iceland, Spitzbergen, and Novaya Zemlya renders it not improbable that 

 single pairs, retarded perhaps at first on their passage, remain behind and breed on the outer- 

 most islands lying off the northern coast of Norway ;" and this opinion derives additional support 

 from the following facts: — "In 1870 I procured from Borgevser, a well-known Fuglevser and 

 Fiskevser, facing the Arctic Ocean, on one of the most northern of the Lofoten Islands, in 

 68° 15' N. lat., through the kindness of Mr. Irgens, the proprietor, two eggs, taken that summer, 

 of a species of Goose smaller in size than those of any of the genus that I had previously 

 examined. In both the length was 07 millims., the breadth being 45 and 46 millims. respec- 

 tively. The shape was cylindrical much rounded at the ends ; the shell was hard, with large 

 pores, in colour shining white with an almost imperceptible tinge of yellowish. The size of 

 these eggs (for they were scarcely bigger than normal eggs of Mergus serrator) induced me to 

 record them (Vid. Selsk. Forh. 1871, p. 59) as those of Anser erythropus, albeit the Borgevser 

 preserve, owing to its westerly position and diverse local reasons, did not apparently fulfil the 

 conditions of the occurrence of this species on the Scandinavian peninsula ; but since then several 

 important circumstances have led me to doubt the correctness of this assumption 



" Mr. Irgens has kindly supplied me with the following data bearing on this question, viz.: — 

 'A pair of Geese with white cheeks, but having the rest of the plumage and feet dark, arrive 

 regularly every season at Borgevser about the 1st of May, a fortnight later than the common 

 Grey Goose. We call it Fjeldskarv, as it has a slight resemblance to a Skarv (Cormorant) or 

 Finmark Goose. It is exceedingly shy in its habits, and difficult to approach. This year (1872) 

 it had deposited its first egg, which was taken, by the 9th of May, whereupon it moved off to a 

 neighbouring islet. It builds a nest composed of moss and straws, sometimes on the narrow 

 ledges of the rocks, and sometimes in a sheltered locality under stones or isolated rocky masses. 

 The full complement of eggs is five. This is the only pair that breed at Borgevser ; and they 

 have been regular visitants for some years past, though never allowed to hatch their first brood.' " 

 To this I may add that Mr. Collett has lately informed me that on his last visit to Northern 

 Norway he convinced himself that the pair of Geese which visit Borgevser are really Bernicla 

 leucopsis ; and on comparing the two eggs with my series of eggs of Anser erythropus they are 

 most certainly not those of this latter species, and there is every probability, therefore, that they 

 really are those of the Bernacle Goose. 



One or two of the later authors on ornithology have endeavoured to resuscitate Scopoli's 

 generic title of Branta for the present species and its allies Bernicla brenta and Bernicla rufi- 

 collis ; but I cannot see the propriety of so doing, and fully agree with Messrs. Sclater and Salvin 

 {vide P. Z. S. 1876, p. 361) that Branta of Scopoli is an artificial group composed of species 



