OF WASHINGTON. 91 



apods, but are, as it were, larviform off-shoots from the insect 

 phylum. In other words, we may compare the diplopods with 

 caterpillars and other larvae, and may seriously undertake the 

 study necessary to determine the reality of what have been sup 

 posed to be merely superficial similarities, such as the form of 

 the cephalic sclerites, the barbed hairs and the repugnatorial 

 pores. Polyxenus looks enough like a caterpillar, and its large 

 fossil relative Palceocampa would have been even more strongly 

 suggestive of such an affinity.* Moreover, a hexapod origin for 

 the Diplopoda would explain the fact that the diplopod larvas are 

 hatched with the six anterior legs, the remaining pairs being 

 attached to rings intercalated behind the genital segment. The 

 anamorphous Chilopoda are hatched with seven pairs of legs, but 

 the others are added by intercalation in front of the genital seg 

 ment, which thus appears near the posterior end of the body in 

 the one group and near the anterior in the other. 



In this way it is possible to bridge the chasm which seemed to 

 so profoundly separate the Progoneata (Diplopoda, Symphyla, 

 and Pauropoda) from the Opisthogoneata (Hexapoda and Chilo- 



* The barbed hairs of the larvaa of the Merocheta, and the bristles of the 

 Coelocheta, Monocheta and Colobognatha support the view that the soft- 

 bodied, hairy Polyxenus is the most primitive of existing diplopod types. 

 It may also be said that the skeletons of the different orders of Diplopoda 

 are too diverse to.be rationally explained by descent from a single hard- 

 bodied type. In the Merocheta the segmental rings are solid and com 

 plete, without even traces of sutures to represent pleural or ventral plates. 

 In the Coelocheta, Monocheta, and Colobognatha the ventral plates 

 are free ; in the Diplocheta, Zygocheta, and Anocheta they are adnate, but 

 are distinct by sutures. The pleurae are free in the Oniscomorpha and 

 Limacomorpha and in the Siphonotidoe ; adnate in the Polyzonidse and 

 remaining Colobognatha, and in the Monocheta and Anocheta; no traces 

 of pleural elements have been reported in the Diplocheta, Zygocheta, 

 Coelocheta and Merocheta. Finally, in the order Anocheta, the dorsal 

 part of the segmental ring is composed of three transverse bands, a con 

 dition perhaps paralleled only in the larvae of the saw-flies. From the 

 entomological standpoint these differences would be thought very grave; 

 indeed, they maybe said to be altogether too grave for explanation by evo 

 lutionary changes in parts already hardened. If, however, we think of them 

 as independent acquisitions of firm armor by soft-skinned animals we have 

 ample analogies in other groups. This interpretation does not, of course, 

 decrease the actual diversity, but it enables us to credit the evidence 

 of the otherwise great similarity of structure and function among the 

 Diplopoda. It permits Polyxenus to be more closely associated with the 

 other Diplopoda, and brings the Diplopoda, as a group, closer to the 

 Symphyla and to the Hexapoda. 



