296 THE VARIATION OF ANIMALS IN NATURE 



come under the present category (with the usual qualification 

 that some apparently useless structures may later be found 

 to have a function). Admittedly, experimental evidence is 

 required to prove that a structure has no significance as an 

 ornament, but, though this evidence is usually lacking, we can 

 scarcely, therefore, assume that all sorts of apparently very 

 trivial male characters are adaptive. In the wasp Trypoxylon 

 palliditarse, for instance, the male differs from the female as 

 follows (besides rather smaller size and different genitalia) : the 

 proportions of the antennal segments, especially apically, are 

 different ; the clypeus has an outstanding lamella with two 

 small teeth on each side of it ; the stipes of the maxillae bears 

 a large angular tubercle, the mid-coxae are set further apart 

 and the mesosternum is more angularly emarginate between 

 them ; the posterior margin of the metasternum is more 

 deeply emarginate ; the antero-dorsal margin of the hind 

 tibiae bears a dense row of short spines ; the first abdominal 

 sternite bears a long recurved hook ; the second and third 

 abdominal sternites are basally impressed. None of these 

 structures appears at all likely to be correlated with courtship 

 or mating, except possibly the modification of the thoracic 

 sternites, which may enable the male to fit more closely to the 

 convex dorsum of the female. 



We are aware of only one or two cases in which actual 

 experiment has shown that secondary sexual characters are 

 apparently without function. Lutz (191 1) removed the tarsal 

 comb in a male Drosophila and found that mating was in no 

 way impeded. The tarsal comb is found in the males of cer- 

 tain species, for which it is an important diagnostic character. 

 Mayer (1900) and Mayer and Soule (1906) showed that wing- 

 colour had no influence on the mating of certain Saturniid and 

 Lymantriid moths, in which the males and females differ 

 markedly in colour. Painting of the wings scarlet, etc., or 

 providing the females with male wings, has no effect on the 

 percentage of successful matings. 



It is impossible to estimate what percentage of secondary 

 sexual characters would have to be classed as apparently 

 useless ; it would certainly be very high and would include a 

 large number of specific characters. The sporadic distribution 

 of such structures is just as marked as in the case of grasping 

 organs. 



(d) Of recent years more and more weight has been placed 



