106 PROCEEDINGS OF THE NATIONAL MUSEUM vol. ill 



end of each gonopod by a loose, flexible pivot joint. The shape of the 

 apodeme varies somewhat, it being long and slender in the larger bodied, 

 primitive forms (Eurelinae, Atopetholinae), and becoming shorter in 

 the smaller forms, where it may also (as in Onychelinae) be distally 

 expanded. The relatively larger size of the posterior gonopod in this 

 subfamily presumably requires, within the confines of a smaller avail- 

 able space, a greater surface for attachment of the muscles. In two 

 of the subfamilies, Eurelinae and Atopetholinae, the distal joint of the 

 posterior gonopod is undifi'erentiated ; there is no perceptable articula- 

 tion between, or remaining evidence of, coxal and postcoxal elements. 

 That the gonopod is actually 2-jointed, however, is shown in the 

 Onychelinae and Arinolinae, where an acute angle is formed, with a 

 flexible articulation at the apex (figs. 8,c-e; 10,c). In no case, however, 

 are there remnants of intersegmental muscles in the posterior gonopods 

 of atopetholids or any other group know^n to us. It is curious that this 

 primitive leg condition would be preserved in members of what, in all 

 other respects, are specialized atopetholids, and emphasizes the point 

 that no existing spiroboloid species has retained ancestral characters 

 in all of its structural features, the parts apparently evolving at 

 different rates and somewhat independently of each other in this 

 respect. 



In general appearance the posterior gonopod is short and bent 

 somewhat mesiad. In the Eurelinae it tends to develop a large hyaline 

 flange on the median side, as in Comanchelus hubrichti (fig. 6,c). In 

 the Atopetholinae this gonopod is longer, more slender, and somewhat 

 arcuate, with a thin expansion on the caudal edge in Atopetholus and 

 a distinct projecting branch there in Watichelus. In the Arinolinae 

 there is a distinct submarginal groove from the coxal attachment 

 distad to a short solenomerite (Arinolus, fig. 10,d) or a longer one 

 which actually extends distad beyond the tip of the gonopod (Piedolus). 

 In this subfamily, also, the gonopod terminates in an expanded, 

 laminate or subglobose area (figs. 10,a,c; ll,c) which is herein tenta- 

 tively referred to as the calyx. The occurrence of the gi-oove, presum- 

 ably homologous to the seminal groove of many other diplopods, is 

 obviously an evolutionary specialty not found in the other three 

 subfamilies. 



Gonopod musculature: It is felt that a satisfactory concept of 

 phylogeny and classification of the Spirobolida can be achieved only 

 by a detailed consideration of comparative morphology of the various 

 groups. Studies of the hard parts are undoubtedly a step in the right 

 direction, but lack real significance until the physiological functions 

 of the various modifications are established. These functions are to 

 a certain extent reflected by the nature of the musculature, and as a 

 contribution toward this end we include a brief account of the gonopod 

 muscles in Atopetholus angelus, a species of the tjrpical genus of the 



