The Craspedosomatidae of North America. 35 



First two pairs of legs of male short and slender, the next five pairs longer 

 and much crassate. 



Anterior male legs hispid on the interior face of the last joint. 



Ninth pair of legs of male four-jointed and clawed, the basal joint large and 

 thick, inferiorly deeply sinuate; second joint slender, cylindrical; distal joints 

 short, the apical conic. The basal joints embrace a bifid lamina (see fig. 6). 



Tenth and eleventh legs of male with large coxal apertures; the coxae of 

 the eleventh legs with a conical process at base. 



Pedigerous laminae with a vertical carina. Segments of adult 30. 



The two known species of this genus inhabit caves, and have 

 been collected in Virginia, Kentucky and Indiana, They are not, 

 however, to be looked upon as cave animals in the strictest sense 

 of the word, as the}- are provided with eyes. The e^^es are less 

 numerous and less prominent than in the other large species, and 

 seem sometimes not to be fully pigmented. That the number has 

 been reduced by cave life is an unsafe inference. The Polydes- 

 mid8e,have no ej^es and j'^et very few are cave species. 



The legs and antennae are proportionally the longest of any 

 Diplopoda known to us. That this is due to cave life cannot be 

 maintained, for such thoroughly cavernicolous species as Sco- 

 terpes copei and Zygonojnis whitei do not have longer antennae 

 than some of the open-air forms. 



The proximal three or four pectinate lamellse of P. cavernarum 

 have the apices of the spines bifid. We have not made suflSci- 

 ently numerous dissections to be confident that this is a constant 

 character, but its occurrence is certainl}^ worthy of note, for such 

 a condition does not seem to have been recorded. 



The sculpture of the dorsal surface is also characteristic, none 

 of the subsequent genera being roughened except b}^ very fine 

 areolation, excepting also, of course, the six setigerous tubercles. 

 The carinse also are somewhat squarer and higher up, making the 

 dorsal surface flatter than in other American genera, although 

 the body as a whole is not depressed. 



In a description of this genus, purporting to be drawn from 

 specimens sent by Dr. A. S. Packard, Jr., Latzel * says : " Die 6 

 Borsten und die zugehdrigen Wartzchen fehlen ganz.'''' The 

 bristles must have been rubbed off in sending, as is usually the 

 case, though some of the bristles of the posterior segments can 

 nearly always be found. 



* Myr. d. Ost.-Ung. Mon. Bd. II. pp. 64 and 214. 



