MILLIPEDS — HOFFMAN 185 



clarus from several localties in western Louisiana, and recorded the 

 color pattern of the species. 



To the present time, therefore, ten specific names have been based 

 upon specimens referable to the genus Pachydesmus, with characters 

 of the male gonopods forming the sole basis for recognition. However, 

 since published illustrations of the gonopods have been made from a 

 diversity of aspects, rarely any two being the same, differences leading 

 to the separation of presumed new species have been much more 

 apparent than real, as the drawings made for this study will reveal. 

 I believe that the genus contains but two species, P. clarus Chamberlin 

 and P. crassicutis Wood, most of the other names representing merely 

 geographic races of the latter. 



Taxonomic Characters 



Although its species can readily be distinguished by nonsexual 

 characters from the species of other genera occurring in the same 

 region, the genus Pachydesmus is diagnosed primarily upon the con- 

 figuration of the male gonopods. The unusual secondary process of 

 the telopodite, which appears to be derived from the strongly chitinized 

 tibiotarsal region rather than from the prefemur, occurs in both of 

 the species and serves as a unifying character although the body 

 structure differs more between P. clarus and P. crassicutis than be- 

 tween species of various distinct genera in other sections of the 

 Xystodesmidae. Although there can scarcely be any doubt that the 

 two are congeneric, their dissimilarities indicate a very remote period 

 of divergence and subsequent isolation. 



Although members of the Crassicutis Group are immediately recog- 

 nizable by their large size and laterally attenuated collum, it is difficult 

 to formulate a generic diagnosis not based on gonopod structure since 

 females of clarus do not present any unusual features and are not very 

 similar to those of crassicutis. A peculiarity of females of both 

 species, however, is the rather large cephalomesial lobe of the pleuron 

 of the 3d segment (discussed further on in this section), which may 

 prove to be constant and stable for generic recognition. Once a 

 specimen is placed in Pachydesmus, the matter of its identification 

 becomes rather simple. 



Size of body: Insofar as can be determined with the limited 

 material at hand, variation in the size of mature specimens appears to 

 be correlated with populations recognizable on the basis of structural 

 features. Figure 2 indicates the range in length (to the nearest milli- 

 meter) of specimens of the recognized forms of the genus. Apparently 

 an over-all range of about 10 mm. is to be expected for each, and 

 although some cases of overlap occur, certain forms can be separated 

 solely on the basis of length, and this variable should clearly be helpful 



