290 Mr. R. I. Pocock on the 



he so, for in Glomeris the mandible is without the cardo and 

 the gnathochilarium without the promentum, and the number 

 of segments in Glomeris and Polydesmus is less than in any 

 other Chilognath. From this latter fact it seems likely that 

 the ancestral Chilognath was possessed of but few segments, 

 an idea to which the existence of but few segments in larval 

 forms lends great weight. And as bearing upon the same 

 subject it is perhaps worthy of remark that Polyxenus, which 

 in the palpiform character of its second pair of gnathites, and 

 questionably in the absence of foramina repugnatoria, 

 resembles, I believe, the ancestral Diplopod, also possesses a 

 small number of segments. 



Polydesmus then more nearly resembles the ancestor of the 

 Chilognatha than does any other genus of the Helmintho- 

 morpha, and Lysiopetalum in the conversion of but one pair 

 of appendages into copulatory organs resembles Polydesmus. 

 But important as this one particular is as a sign of atfinity, 

 it is outweighed by the many points of resemblance between 

 Lysiopetalum and lulus. I have therefore associated the 

 Lysiopetalida^ with the lulidee, Polyzonidee, and Chordeumidse 

 in the suborder luloidea. 



At the same time, however^ it must be borne in mind that 

 Lysiopetalum is intermediate between Polydesmus and lulus^ 

 being more highly specialized than the former and less highly 

 than the latter. 



The conversion of both pairs of appendages of the seventh 

 segment into copulatory organs shows close relationship 

 between Chordeuma, Polyzonium^ and lulus — the Polyzonidee, 

 as M. de Saussure long ago suggested, appearing to be but 

 degraded lulidce, and the Chordeumidge only differing from 

 the lulidas in the absence of the foramina repugnatoria, in 

 the smaller size of the first segment, and in the possession of 

 a smaller number of somites. 



To sum up : Polyxenus in the possession of a small number 

 of segments and in the pediform character of its second pair 

 of gnathites shows comparatively but little specialization, and 

 presumably therefore but little differentiation from the an- 

 cestor of the Diplopoda. The fusion of the second pair of 

 gnathites into a plate, the gnathochilarium, characterized the 

 ancestral Chilognath, which was further distinguished by the 

 possession of tufted trachese (?), by the absence of the man- 

 dibular cardo and of the promentum in the gnathochilarium, 

 and showed resemblance to the ancestral Diplopod by the 

 presence of but few body -somites and by the absence of copu- 

 latory feet. From this Protochilognath sprang the Onisco- 

 morpha and the Helminthomorpha. The former, undergoing 



