2o6 PETREL GROUP 



edition of Varrell's BritisJi Birds it is stated that in puffins the 

 whole foot — or rather the foot, and what is commonly called, as 

 in this work, the shank — is applied to the ground when walking, 

 but this appears to be incorrect. When they cannot find rabbit- 

 holes to save them the trouble, puffins commonly begin to excavate 

 their breeding-burrows during the latter part of May in England ; 

 the work being mainly accomplished by the cock, although incubation 

 is performed by the hen, who is fed with fish during that period by 

 her attentive partner. At the end of the burrow, which may be 

 as much as a yard in length and is frequently curved, is deposited 

 the single white or bluish-white egg. Although almost uniformly 

 coloured examples are known, the eggs, which measure from 2.15 

 to 2.7 inches in length, are usually spotted and blotched with pale 

 purple and grey, but may also .show spots and scribblings of yellowish 

 brown. Although such inelegant walkers, puffins are excellent divers 

 and strong fliers, making long excursions from and to their breeding- 

 places out to sea in search of the young fishes which, with crustaceans 

 and insects, constitute their food. At certain seasons the chief food- 

 supply appears to be formed by young herrings, but on some occasions 

 by young sand-eels or launces, of which forty-one have been taken from 

 the crop of a single bird.' Young puffins remain in their breeding- 

 holes for about three weeks after hatching, when they are enticed out 

 by their parents, and, with the latter, soon after desert the land for the 

 open sea. This account may be closed by the mention of the curious 

 fact that on September 20, 1905, three puffins were taken in Kerry by 

 rod and line, two by fly-fishing, and the third while trolling for bass. 



Storm Petrel ^'^^ petrels and their allies, constituting the order 

 (Procellaria Tubinares, present a remarkable contrast in regard 

 J . X to geographical distribution to the auks, for in place 

 of being confined to the higher latitudes of the 

 northern hemisphere, they are mainly characteristic of the southern 

 hemisphere, and have but a comparatively small number of repre- 

 sentatives on our side of the equator. From the general appearance 

 of many of them, due to adaptation to a similar mode of life, the 

 petrels (as the entire group may be collectively termed) were at one 

 time associated with the gulls and terns ; and they agree with the 

 latter and all the ordinal groups hitherto treated of in the structure 



' Lcsl ihc present writer shoulil l)e accused of error, it ni.iy he pointed out tliat in the 

 work where this statement occurs, the fish is incorrectly said to be the lancelet, which, l>y the 

 way, is not a fish at all. 



