17 [Vol. xxxvi. 



Now we come to the question of colour. Here, at first, 

 we are bewildered by the variation, but at last certain rules 

 begin to dawn on us. Thus, colour tends to disappear when 

 breeding is carried on in the dark, and colour is accompanied 

 by light. Many eggs are more or less protectively coloured ; 

 in others the necessary protection is obtained by means of 

 down or nesting material. Here again, then, we find struc- 

 ture and habits directly connected with colour in some cases. 

 A bird whose eyes are constructed for night- work will natu- 

 rally tend to nest in a dark hole, and its eggs will probably 

 be white. The extraordinary range of variation in Cuckoos' 

 eggs is the inevitable result of their parasitic habits on 

 different species. But I have said enough to show that it is 

 impossible to regard the variation in these respects as acci- 

 dental, and therefore negligible. We may not understand 

 the working of these laws, and at times one may be in oppo- 

 sition to another,in which case that of the greater importance 

 to the preservation of the species will prevail ; and this will, 

 I think, explain the apparent anomalies which undoubtedly 

 do exist. 



If you took a representative collection of eggs of S. African 

 birds, and showed them to a man who was well acquainted 

 with British birds' eggs alone, he would be able to select with 

 some certainty the eggs of the Birds of Prey, Owls, Pigeons, 

 Herons, Ducks, Waders, Goatsuckers, Grebes, Cormorants, 

 and many of the Passeres, such as Sparrows, Shrikes, Chats, 

 Swallows, etc., but with the eggs of species which have no near 

 allies on our list, such as the Fin-foot, the Hammer-kop, and 

 the Weavers, he would naturally be at a loss. But the fact 

 remains that he would be able to tell almost as much about 

 the birds from a study of the eggs as from seeing the skins ! 



The fact is that in many orders the eggs are so constant 

 to type as to be recognizable at a glance. It is almost 

 impossible to mistake a Petrel's egg. Those of the Owls 

 and Pigeons are almost equally characteristic. Among the 

 the Accipitres we find two main types, but all conform 

 more or less to one or the other. As a rule the Limicoline 

 egg is unmistakable. On the other hand, among the Passeres, 

 and even in a family like the Muscicapidee of Dr. Hartert (or 



