COURTSHIP AND SEX 213 



partly scarlet, while during the rest of the year he wears a 

 plain, somewhat sparrow-like suit, as the female always does. 

 An interesting point is that the transformation of feathers 

 occurs in the less essential tail-feathers much more frequently 

 than in the more important wing-feathers. But there are 

 many instances in this part of the plumage as well. Thus in 

 the extraordinarily decorative Quezal {Pharomacrus mocinno) 

 of the Guatemala forests, the male has much elongated wing- 

 coverts, as well as greatly exaggerated and very gorgeous 

 tail-coverts and a high crest. In some of the African goat- 

 suckers of the genus Macrodipteryx there is a great elongation 

 of a primary feather or of three of them ; in one species the 

 ninth primary is over two feet in length, more than twice 

 as long as the bird's body. 



C. Theories 0! Sex Dimorphism, — Darwin's Theory. — 

 As every one knows, Darwin argued that the evolution of 

 dimorphic sex-characters might be accounted for in terms of 

 selection — especially sexual selection. This has two modes : 

 the combats between rival males and the preferential mating 

 where the female chooses or seems to choose. It must be 

 admitted that Darwin's theory has to face some serious 

 difficulties, (a) What is known experimentally in regard 

 to selective breeding, e.g. Johannsen's work, does not 

 favour the view that the level of differentiation of, say, the 

 stag's antlers and the decorations of the bird of paradise 

 could be gradually raised generation after generation by 

 such selection as combats and preferential mating respec- 

 tively afford, {b) In some cases of pronounced sex di- 

 morphism there is no evidence either of preferential mating 

 or of combats, and subsidiary hypotheses have to be invoked. 

 Thus we have Giinther's suggestion that masculine characters 

 have their justification as a means of " bluffing " enemies. 

 (c) Statements in regard to serious disproportion in the 

 number of the two sexes must be taken critically, having 

 broken down in several cases. And it is plain that the value 

 of the selectionist interpretation depends largely on the 

 evidence that considerable numbers of the less attractive or 



