378 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
In Bouvier’s key this species will run down to edwardsii, but it is 
quite different from that species in the form and arrangement of th 
tegumentary papillae, and also possesses a larger number of legs in th 
female (33-34 pairs), edwardsti having only 30-32, rarely 33 in this sex. 
Were it not for the presence of incomplete body folds the struct 
of the integument would place it close to brasiliensis, although the 
specimens of braszliensis figured by Bouvier have the primary papillae 
placed farther apart and the accessory ones ascend to the ridges of the 
folds. The structure of the fourth and fifth legs is as in brasiliensis 
also, with the nephridial tubercle entirely free from the third band of 
the creeping pad. It appears impossible, therefore, to associate the 
present form closely with any described species, although it undoubt- 
edly falls near edwardsi while showing a striking similarity to brasi- 
hensis. Pervpatus brasiliensis was described from Santarem, but ac- 
cording to Bouvier probably extends to Panama; P. edwardsw oceurs 
from French Guiana to Colombia. The present species is not closely 
related to any of the forms recently described by Fuhrmann (Zool. 
nz., 1913, 42, p. 241-248) from Colombia. 
PERIPATUS (EPIPERIPATUS) EDWARDSI Blanchard. 
Ann. sci. nat. Zool., 1847, sér. 3, 8, p. 140. 
Of this widely distributed species there are nine specimens (Coll. 
Univ. Mich. and M. C. Z. No. 241-244), from the Cincinnati Coffee 
Plantation, near Santa Marta, Colombia. 
As can be seen from the accompanying illustrations, there is con- 
siderable variation in the distribution of the integumentary papillae, 
particularly in the accessory ones. These latter vary much both in 
numbers and in distribution on the body folds, but the variation does 
not exceed that already observed by Bouvier in this species. Al 
have the nephridial tubercles of the fourth and fifth legs free from the 
third creeping pad and the fourth pad is continuous. The numbel 
of legs is indicated in Table 2. 
In the smaller specimens there is a marked difference in the size 0) 
the primary papillae which are alternately large with well-developec 
terminal cylinder and small with reduced cylinder. In these spec 
mens the accessory papillae are sparse and restricted almost exclu: 
sively to the edges of the folds (Plate 2, fig. 7-8). The larger example: 
(Plate 2, fig. 6, 9) have the primary papillae of more nearly equal sizt 
and similar form and the accessory ones ascend more commonly 0! 
the sides of the folds although always remaining away from the ridges 
