330 Geoffrey Smith 



Now the species withiu the limits of each geniis are very closely 

 allied, beiug dittercntiated bv eoiistaut biit detailed ditit'erences in 

 marking- and strueture, and they show this remarkable gradation 

 in the degree of devehipnieut of the seeondary sexual charaeter 

 correlated with the total si/e. 



Sufficieut evidence lias now been produced to show that several 

 widely separated genera of the Lamellicornia follow the rule which 

 was postulated to meet the supposition that the priuciple of high 

 and low diniori)hism within the limits of a species can be extended 

 to a progressive difterentiation in a group of closely allied species. 

 It ma}' be suspected that we have merely selected favourable and 

 suppressed unfavourable evidence. This is not the case: the only 

 instance in which an apparent contradiction to the rule is found is 

 in the genus Goliatlms, where the smaller species have not sutfered 

 a disproportionate decrease in the seeondary sexual characters, Nor 

 in the larger species can I find any evidence of the occurrence of 

 high and low dimorphism within the limits of a species, so that 

 even this single instance does not definitely contradict us. 



C. Pectinico mia. Lacordaire divides the Pectinicornia into 

 Lucanidae and Passalidae, but the Passalidae do not exhibit any 

 high degree of sexual dimorphism, so I have neglected them. The 

 Lucanidae are nearly all highly sexually dimorphic, the characters 

 afifected being the mandibles, and to a less extent the head, prothorax 

 and legs. The mandibles of the male are often hugely developed, 

 those of the feniale beiug always much amaller: and througbout the 

 whole group the phenomenon of highness and lowne^s is of (juite 

 universal occurrence. Tiiat this factor has been an important, if not 

 the most important, factor in the evolution of this group, is I thiuk 

 thoroughly borne out, but except in one or two genera we do not 

 meet with quite such striking series as in the Lamellicornia, because 

 long series of very similar species do not occur. We are also troub- 

 led by the great qualitative differences which exist bctween closely 

 allied species, and even between individuals of the same species. 

 'I'lius iu certain species of the genus Odo/ttolabis, e. g. 0. Brooleanus, 

 I have ob.served in three similarly sized males three (piite dificrent 

 types of mandible. This kind of Variation is evidently a kind of 

 high and low polymorphism, but it is not at all strictly coincident 

 with variations in size. 



I do not iiitend going into the same detail with these bcetles 

 as with the Lamellicorns, since it would simply mcan repeating the 



