384 



bulletin: museum of comparative zoology. 



quite broad anteriorly and more slender than the others posteriorly, 

 hut the approximation of the anterior pairs of legs shows this to be 

 due to a contraction of the anterior half of the animal. There is very 

 little A'ariation in length as can be seen from the following measure- 

 ments of all four female specimens. 



Legs. The number of pairs of legs varies from 36 to 37 in the female, 

 two having the former and two the latter number. In the male, as 

 shown by two specimens there are 34 pairs. Each leg has five well- 

 de^'eloped creeping pads, the fifth usually considerably narrower 



than the fourth and occasionally 

 somewhat reduced in length, al- 

 though usually almost as long as 

 the fourth. The first is broader 

 and the three following of about 

 equal width as a rule, although 

 frequently the fourth is dis- 

 tinctly narrower and sometimes 

 scarcely wider than the fifth. 

 The nephridial tubercles of the 

 fourth and fifth pairs of legs 

 (Fig. 1) are nearly included in 

 the third creeping pad which is 

 very deeply emarginate, but not 

 completely di\'ided by them; 

 the fourth pad is distinctly 

 emarginate, but not deeply ex- 

 cavated along its distal edge 

 opposite the tubercle. There 

 are four pedal papillae on each leg, ver>- rarely five in three of the 

 four specimens, while the fourth shows on a considerable number of 

 the anterior legs five well-developed papillae. The fifth papilla is 

 rarely as large as the others, and .sometimes occurs between the pair 

 on the anterior or posterior edge of the foot, or either above or below 

 them. Bouxier lias regarded the presence of extra papillae on certain 

 legs as a specific character, but it is evident at least in the ca.se of the 



Fit? 

 pliridi 



1. — I'erifxitus peruviaiiiis Brues. Ne- 

 il liihcri-les of Die fourlli pair of lofjs. 



