High and low Dimorpliisra. 321 



the total leugth are uot given, the authors state (Footnote p. 589): 

 "There is nevertheless no doubt that the ratio of the length of the 

 forceps to the total length is higher in high males than in low". 



E. Xijlotnipes gideon and the Lamellicornia. The dimorphism 

 in Xylotrupes gideon has been described in Bateson & Brindley's 

 paper already referred to. The curve representìng the relative 

 freqiiency of high and low males was markedly bimodal. 



In the whole order of Lamellicornia, wherever highly differen- 

 tiated secondary sexual characters occur, the essential feature of 

 high and low dimorphism can be seen, i. e. the association of highly 

 developed secondary sexual characters with a largo size and their 

 reduction with a decrease in size among the males of the species 

 that exhibit the pheuomeuon. 



F. Pectinicornia. The character bere chiefly affected is the 

 mandible of the male. The essence of high and low dimorphism is 

 bere also apparent in ali the species showing sexual differentiation. 

 The curve, however, obtained by Bateson & Brindley for a sample 

 of Lucanus cervus was unimodal ^ 



4. Tlie correlatioii betweeu the secondary sexual characters 

 aud the primary sexual character. 



I am unable to enter very widely into this highly important 

 aspect of the question without encroaching on a subject which forms 

 the chief part of a work that I hope soon to publish. It is however 

 well kuown from the facts of ordinary and parasitic castration, and 

 of ovariotomy, and from observations on animals with abnormal 

 genitalia that a dose correlation exists between the primary and 

 secondary sexual characters. I may also revert again to the facts 

 described for luachus scorpio (see page 316), where it was found that 

 the middle males with fiat female-like chelae always had reduced 

 testes in which spermatogenesis was not active, as is also the case 

 with low males in winter, when they too bave fiat chelae and small 



* Certain cases of high and low dimorphism have been suggested by 

 GiARD (3] as due to the effects of parasitic castration. Although this may be 

 the case in certain instances , it is hardly possible that a phenomenon of such 

 Wide distribution associated with particular stages in growth in totally different 

 groups of animals can be always due to a parasitic afFection. This is certainly 

 not the case in the facultative dimorphism of Inaclms scorpio, and it can hardly 

 be doubted that we are dealing with essentially the same kind of fact there 

 as in definitive dimorphism. 



