70 



NATURAL HISTORY OF BIRDS. 



question represents quite a tj-pe of its own, being nearly gallinaceous in form, and ro- 

 niarkalile for an exceedingly short gonys. That this form is said to have only one 

 carotid, while the other Alcida? have two, is perhaps of less moment, though we must 

 remember that the grebes were similarly specialized. 



The sea-dove is truly hyperborean in its breeding liabits, being found in incrediblo 

 numbers along the island shores of the western Arctic Ocean. Nordenskjold gives a 

 very graphic descri])tion, from wliieh we e.vtract the following. The rotge occurs only 

 sparingly off the southern part of Novaja Zemlja, and does not, so far as I know, 

 breed there. The situation of the land is too southerly, the accumulations of stones 



.♦'■li^ij'y'; ;"'!'!■;, 



Fio. 31.— Una ringvia, spectacled guillemot. 



along the sides of the mountains too inconsiderable, for the thriving of this little bird. 

 But on Spitzbergen it occurs in incredible numbers, and breeds in the talus a hundred 

 to two hundred metres high, which frost and weathering have formed at several places 

 on the steep slopes of the coast mountain sides. These stone heaps form the palace of 

 the king-auk, richer in rooms and halls than any other in the wiile round world. If 

 one climbs up among the stones, he sees at intervals actual clouds of fowl s\i(ldenly 

 emerge from the ground, either to swarm round in the air or else to fly out to sea, and 

 at the same time, those that remain make their presence underground known by an 



