A NEW AND ANNEOTANT TYPE OP CHILOPOD. 439 



merit of drawing attention to the undeniable relationship 

 subsisting between the ScutigeridfB and the Lithobiidse on 

 the one hand, and the Scolopendridfe and Geophilidee on 

 the other, it appears to me to be open to the serious 

 objection of obscuring the value of the great structural 

 differences which separate the Scutigerid^ not only from 

 the Lithobiidas, but from the Scolopendridae and Geophilidse 

 as well, and of ignoring the many and important features 

 in which the LithobiidfB resemble the Scolopendridae.^ 



The characters Avhioh justify the association of the 

 Lithobiidee with the Scutigeridae are the following : — the 

 presence of fifteen pairs of legs in the adult, of seven 

 pairs in the newly hatched young, the rest being added 

 with successive moults ; the presence of stigmata upon the 

 same somites ; the completeness of the penultimate and ante- 

 penultimate segments of the toxicognaths ; the absence of all 

 trace of "sub-segments" on the somites; the large size of 

 the coxae of the legs, and the freedom of those of the 

 posterior pairs, and the presence of powerful gonopods in the 

 female — characters which are not found in the other families. 



But although these similarities attest the derivation of 

 the Scutigeridae from Chilopoda of the Lithobioid type, 

 there can surely be no two opinions as to the extent to 

 which they are outweighed when placed in the balance 

 against the many and deep-seated characteristics in which 

 Scutigera differs from all other existing centipedes. 



My views on this point were expressed in the classification 



' No doubt Haase found coiilirniation for liis views in tlie discovery of the 

 genus Ceniiatobius — a genus differing from all the Lithobiidse known to 

 him, and approaching the ScutigeridtE in its longer, "funiculate" antennae, longer 

 and subdivided tarsi, absence of ooxal pores, longer female gonopods, and 

 slightly more dorsal stigmata. There is no doubt, however, that Cermatobius 

 is very closely related to Lamyctes and its allies. For example, although 

 typically four or five in number in these forms, the coxal pores are reduced 

 to one on each coxa in Haasiella and in Henicops, as typified by 

 maculatus, tiie anterior tarsi consist of three segments, and the posterior 

 pairs of six — facts which largely discount the value of the absence of coxal 

 pores and tlie rich tarsal segnientatiun in Cermatobius. 



