26 PROCEEDINGS OF THE NATIONAL MUSEUM vol. in 



tion of the species. Although the chances that Attems' material is 

 conspecific with the type specimen of Polydesmus bilineatus seem very 

 slight, I refrain from suggesting a new name until the Brazilian species 

 is restudied and compared with the considerable number of new species 

 recently described by Chamberhn (1941) from the adjacent Depart- 

 ment of Loreto, Peru. 



According to Attems' description, the pleural spines of his species 

 form a row along the caudal margin of the segments all the way from 

 sterna to level of insertion of the paranota, in contrast to the strictly 

 basal cluster in Platyrhacus acanthopleurus . There appear also to be 

 differences in the shape of the gonopod prefemur, the position of the 

 ozopores, and the development of transverse tuberculation. Probably 

 a closer study of Attems' material will provide numerous other minor 

 differences. I do not doubt that the two are closely related, but the 

 gi-eat difference in localities alone virtuall}^ precludes their identity. 

 I have been unable to reconcile acanthopleurus with any of the Peruvian 

 species described by Chamberlin in 1941. 



Genus Nyssodesrnus Cook 



Nyssodesnnis Cook, 1896b, p. 53. — Hoflfman, 1953b, p. 253. 



Type species: Nyssodesmus alboalatus Cook, by original designation. 



Range : Central America, from western Panama to Nicaragua. 



Nyssodesmus was proposed as a new genus on the basis of a speci- 

 men from Nicaragua in which the male gonopods are "nearly straight 

 for about two-thirds of their length, and are then bent toward the 

 animal at nearly a right angle." In his summary of the milli- 

 peds of Central America Chamberhn (1922) described several addi- 

 tional species conforming to this general configuration, but also used 

 the name Nyssodesmus to include species that are now considered 

 to belong to Platyrhacus. Although Attems (1938) rejected Nysso- 

 desrnus as a nomen nudum, he substituted in its place the equally 

 vague name Tirodesmus, which had been proposed in Cook's 1896b 

 paper. Attems' concept of Tirodesmus embraced members of what 

 I consider four valid genera as well as Tirodesmus in its strict sense 

 {ioY Jimhriatus Peters). The first subsequent proposal to resurrect 

 Nyssodesmus as a valid genus as originally conceived by Cook was 

 made in one of my earlier papers on Platyrhacus (1953b), in which 

 I allocated the following names to the genus : 



N. alboalatus Cook (1896b) 

 A^. tristani Pocock (1909) 

 N. antius Chamberlin (1914) 

 N. mimus Chamberlin (1922) 

 N. hivirgalus Carl (1902) 

 N. linionensis Attems (1900) 



