CH^TODERMA. 285 



rior, most movable part of the body, as well as the great spines of 

 the posterior end, played the leading part. These spines are so 

 placed that in the contracted animal they converge backward, but 

 in the expanded condition they diverge, pointing obliquely back- 

 ward and laterally. When the animal expands these prickles must 

 accordingly grip into the side walls of the burrow and thus lend 

 support to the posterior end, preventing any movement upward of 

 this end. Consequently, with each elongation the anterior end is 

 pushed forward a distance equal to the difference in length between 

 the extended and the contracted animal. In a great contraction 

 the anterior part of the prothorax is swollen to a thick bulb, 

 whereby the anterior part of the animal is apparently wedged in the 

 burrow, the small spines of that end, which are directed obliquely 

 to the side, affording insufficient support. The great spines of the 

 hinder end at the same time become loosened from the walls of the 

 burrow. "With each contraction the hinder end draws itself forward 

 without change in the position of the anterior end. 



" I have never observed an animal which had burrowed deeply 

 in this manner, come up in the same burrow in which it had de- 

 scended. In order again to reach the resting position the animal 

 must first bore upward and on reaching the surface, again bore 

 downward. It describes, therefore, during its wanderings from the 

 first resting position to the second, the curve represented in pi. 41, 

 fig. 26. The animal proceeds a short distance on the surface, in- 

 deed, it sometimes crawls several inches, before it again bores down- 

 ward. This is a very slow procedure and attended with considerable 

 difficulty, the hinder part of the body swinging now to right, now 

 to left, by means of alternating expansion and contraction. Gen- 

 erally, particularly on a slightly uneven surface, Chsetoderma makes 

 wholly irregular tracks ; on even surfaces, however, and when the 

 animal crawls straight forward, we obtain that peculiar regular 

 appearance which the subjoined drawing (pi. 41, fig. 27) presents, 

 which might easily convey to the paleontologist the idea of a plant 

 impression. 



" I have never seen Chcetoderma performing swimming move- 

 ments, nor has it seemed able to crawl up the walls of the aquarium. 

 It is wholly adapted to life in the slime bottom, and the knowledge of 

 this circumstance is of importance for the proper understanding of 

 the organization of the animal in reference to its relationships with 

 allied forms. Chcetoderma does not devour sand or slime as many 



