IO8 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. 



and the peripheral figures become prominent conical tubercles. Each of 

 the areas has one plate, or sometimes two, in which this modification is 

 extreme. All of the genera have a row of marginal plates around the 

 borders of the carapace, which also display a modification of the sculp- 

 ture ; these plates will be more particularly described in connection with 

 Propalceohoplophorus. 



3. The tail-sheath, which is likewise of very uniform character in 

 nearly all of the genera, consists of two quite distinct portions, of which 

 the anterior one is composed of four or five freely movable rings, each 

 ring having two rows of plates and overlapping the one behind it ; in this 

 region the ornamentation is very similar to that of the carapace ; in the 

 posterior part of the sheath the rings, though still imbricating, are much 

 more closely fitted and less movable, and the plates, which are smaller, 

 have no definite sculpture, merely a punctate surface. The number of 

 rings composing [this posterior part of the tail-sheath appears not to 

 exceed four or five, and the shape of the mass differs somewhat in the 

 various genera in which this structure is known ; in Propalceohoplophorus 

 it is heavy and club-like, tapering but little and ending abruptly, the tube 

 being closed by a small irregular plate; in Cochlops it is similar, but 

 more tapered and with thinner plates ; in Asterostemma it is elongate 

 and much more tapering, ending in a point, and made up of obscurely 

 marked rings, each consisting of a single row of plates. This difference 

 between Asterostemma and the other genera is not without importance, 

 for in the former may be observed a distinct approximation to the structure 

 of the armadillos. No coossification occurs between the various plates of 

 the tail-sheath. 



4. In the cephalic shield there is much more variety, each genus appear- 

 ing to have its own characteristic form. However, this statement cannot 

 yet be positively made, for the head-shield of Metopotoxus is very imper- 

 fectly known and that of Asterostemma not at all. In Propalczohoplo- 

 phorus the component plates of the shield are all separate and quite thin 

 and most of them are of relatively large size ; the ornamentation is simi- 

 lar to that of the carapace, but the demarcating grooves are extremely 

 shallow, and the surface of the plates is quite smooth and finely punctate. 

 In Cochlops the plates, likewise separate, are more numerous, and rela- 

 tively smaller and thicker ; the pattern is made by much deeper grooves, 

 and the surface of the plates is decidedly rougher and more coarsely 



