ATOPETHOLID MILLIPEDS — HOFFIVIAN AND ORCUTT 141 



Discussion: This species is known only from females and may 

 well be, as Loomis (1949) suggested, actually referable to the related 

 genus Watichelus, which occurs in the same area. The lateral end 

 of the collum is strongly bent caudoventrad. Final allocation must 

 await the study of conspecific males from Mountain Spring. 



Onychelinae, new subfamily 



As presently known, this group consists of two small genera that 

 occupy rather distant regions: Onychelus in southern California and 

 adjacent Arizona, and Saussurobolus in the southern end of the Mexi- 

 can Plateau. The diagnostic characters of the subfamily are set forth 

 in the preceding key, to which may be added the general remarks 

 that the group appears to be a rather disjunct one, particularly in 

 the form of the anterior gonopods. Very Httle is known about the 

 structural details of the Mexican species, but Onychelus at least appears 

 to be modified for an arenaceous habitat because of having more and 

 longer setae on the sides of the legs than the other known genera of 

 the family — in effect a fringe that is conceivably an aid in locomotion 

 under loose soil. The marginal setae of the anal valves are also 

 longer and more closely set than in the other species. 



In Onychelus the anterior legs of the males are provided with long 

 slender claws, and the coxal lobes are but weakly produced. Un- 

 fortunatel}^ nothing is loiown about the tarsal claws in the genus 

 Saussurobolus, but the coxal lobes of the third leg pair in that genus 

 are rather long and conspicuous. The similarity of the two genera 

 is best noted by comparing the basal structure of the gonopods, par- 

 ticularly the distinctly incurved sternal apodemes. The sternite itself 

 is much more strongly produced distad than in any other atopetholid 

 genus, approximating the type characteristic of most Rhinocricidae. 

 The apodemes of the posterior gonopods are short and strongly ex- 

 panded distally in both genera, the telopodites taking the form of a 

 falciform blade, longer and more slender in Onychelus. 



The very great distance between the two regions inhabited by spe- 

 cies of this subfamily indicates that the intervening area has not been 

 explored for its milliped fauna and that a large number of onychelids 

 surely remains to be discovered in northern and central Mexico. 

 Partly on the basis of geographic probabiHty, we refer the inadequately 

 described Arinolus zacatecus (Chamberlin, 1947b) to this group of 

 genera, although it is clear enough that the species is not a member 

 of the Arinolinae. 



