278 



SEA-BIRDS 



Fig. 51 

 Breeding distribution of the puffin, Fratercula 

 arctica, and its Pacific counterpart the horned 

 puffin, F. corniculata. Black Hnes embrace 

 probable extent of breeding range ; black dots 

 represent some known breeding-places 



(North Rona, Darling, 

 1947), or render the site 

 so porous and flimsy 

 as to be untenable after 

 a few decades, as at 

 Grassholm (Lockley, 

 1938). The compara- 

 tive safety of the burrow 

 from predators enables 

 the puffin to leave the 

 &gg or chick for long 

 periods with impunity. 

 It has been observed 

 that razorbills, nesting in 

 well-concealed crevices, 

 will also leave the tgg 

 or chick alone, but for 

 a shorter period. The 

 habit is of further signifi- 

 cance if we connect it 

 with the implications 

 of the double brood- 

 spot in these two birds: 



if the puffin and the 

 razorbill formerly laid two eggs (as the double brood-spot suggests) 

 the present delay in beginning incubation of the single ^gg in these 

 species is explicable as the survival of a once useful habit (of com- 

 mencing incubation when the clutch was complete) which no longer 

 serves its original biological function. In this connection it is interest- 

 ing to note that the black guillemot (with two brood-spots, near the 

 centre of the breast) itself may be following the same evolutionary 

 trend towards one ^gg. Frequently only one is laid, occasionally 

 three, but two is the normal clutch; but though two may be hatched 

 as a rule only one chick is reared (Armstrong, 1940). MortaUty is 

 heavy: Winn found 73.8 per cent, loss of eggs and young in forty-six 

 nests. The brooding bird seems deliberately to incubate one ^gg only — 

 Winn found that "unwanted" eggs were pushed out of some nests, 

 even when he replaced them. 



The common and Briinnich's guillemots, with the Q:gg fully exposed 

 to attack from the air if it is uncovered, never voluntary leave the Qgg 



