High and low Dimorphisin. 331 



same kind of evideuce: but I will indicate which genera cliiefly 

 show evidence of our principle. In the genera Pholidotus, Chiasogiiathiis, 

 Lamprlma, Li/canus, Odontolabis and Cladognatìms^ we do not get 

 any long continuous series of speeies graduating into one anotlier, 

 but with a very high degree of generality it is found that the larger 

 speeies are much more highly sexually dimorphic thau the smaller, 

 and there is no clear instanee of the opposite case. In individuai 

 cases we get obvious instances of one speeies being high to a smaller 

 and very closely allied low speeies. 



The genus Cìjclomatus gives a very perfect series of closely 

 similar forms passing from the large C. tarandus where the mandi- 

 bles may be as long as the body to smaller and smaller forms, until 

 in C. stryiceps the highest males bave mandibles less than half the 

 body ìength. The genera Hemisodorcus, Eurijirachelus and Dorcus 

 give similar but not such perfect series. 



AA"e bave so far mentioned ali the genera of Lucanidae without 

 Omission: the rest of the genera contain much smaller forms with 

 their secondary sexual differences very much reduced; consequently 

 in the majority of cases, it is hard to detect the presence of our 

 principle. On the whole the Lucanidae atford very definite support 

 and confirmation to the general law which was found to be of 

 application to the Tanaidae and Lamellicornia, in this case the 

 character affected being quite new, namel}', the mandibles. 



6. Conclusioii. 



In the foregoing paragraphs it has beeu showu that the diffe- 

 rentiation of the males of many speeies of Arthropods into high 

 and low forms is of wide occurrence in distantly related groups 

 which show a high degree of sexual differentiation. The high and 

 low dimorphism may be facultative, as in those cases in which the 

 low male by growth passes through au intermediate condition to 

 the high male, or definitive, as in those cases where growth ceases 

 '•n the assumption of the mature condition. The development of the 

 >econdary sexual eharacters is strictly correlated with that of the 

 primary sexual development, so that the condition of highness and 

 lowuess in the secondary sexual eharacters must stand in relation to 

 some modificatiou in the condition of the primary sexual develop- 

 ment, as defined in Section 4. 



