ISOLATION 149 



are alike. Again, many insects, just prior to» mating, form 

 swarms of one sex only : the attraction here cannot be strictly 

 sexual, although it is a preliminary to mating. While we know 

 little in detail as to the influence of recognition marks, we 

 can see that any tendency to form aggregations will lead to 

 some degree of isolation. We cannot yet say whether such 

 recognition marks often come to differ in the early stages of 

 species evolution. In many animals with more than one 

 colour-form the various types all interbreed (cf. Elton, 1927, 

 p. 182 et seq. ; Richards, 1927). Probably recognition marks 

 grade insensibly over into what must be classed as epigamic 

 characters, but the former would include stimuli not acting 

 only during a brief period before mating. 



II (ii). Differences in epigamic characters. 



The enormous mass of data concerning the epigamic 

 characters of animals is not very helpful from the present 

 point of view. An examination of the literature shows that 

 the greatest number of papers describe the morphology of 

 epigamic structures ; a less number describe the mating 

 behaviour, including the use of such structures ; still fewer 

 provide any evidence as to the significance in isolation of 

 specific differences in epigamic structure and behaviour. 



It is well established that species do very often differ in 

 secondary sexual characters. In only a small fraction of the 

 total number of species have these differences been shown to 

 have a significance in mating. In many cases (e.g. Saturniid 

 moths (Mayer, 1900) ) the characters are probably only indi- 

 cators of important differences in metabolism. In other cases 

 the female may have been modified in connection with her 

 maternal duties (development of brood-pouches, etc.). 



Where the sexual characters are known to play a part in 

 courtship their exact significance is nearly always doubtful. 

 There is not enough experimental work to prove that particular 

 structures or types of behaviour are actually essential if the 

 male is to be successful. Usually the most that is known is 

 that some conspicuous structure is exhibited in a provocative 

 way during courtship. We may give a few examples in which 

 the significance of epigamic structures or behaviour is fairly 

 certain. 



Sturtevant (191 5) has shown that the wing- waving of male 



