SEXUAL SELECTION— NESTING OE BIRDS. 227 



he instances are sufficient to upset any theory 

 of " protection " that might be based upon the 

 interesting facts. Now I was under the impression 

 that in mv essay I had been particularly careful to 

 deal with these exceptions to the general rule, and 

 had offered what I believed to be a plausible and 

 logical explanation of them in my cHvision : " Birds 

 in which both sexes are dull in colour, and which 

 build covered nests from motives of safety other 

 than concealment." It seems to me, however, that 

 Mr. Allen has been singularly unfortunate in the 

 selection of the instances which he is pleased to call 

 " glaring exceptions " to this rule. His first, the 

 Mandarin and Wood Ducks, Aix galeriadata and 

 Aix sponsa, of China and North America respect- 

 ively, may possibly nest in trees, either because the 

 dnuii of the female is light in colour and conse- 

 quently conspicuous ; or because the male, contrary 

 to most species of the Anatimp, keeps close company 

 with the female after she has commenced to sit 

 (Conf. IJater Birds of Nortli America, ii. p. 14). 

 Of his second exception, the birds in the genus 

 Clangula (Golden-eyes), it seems almost an incon- 

 testable fact that the paleness of the female's down 

 is the reason why the eggs are liatched in a covered 

 site. So that really these exceptions are not excep- 



