F. B. SUMNER AND H. H. COLLINS. 



over, the act, when it does occur, 

 is so sudden that one is com- 

 monly left wondering how it hap- 

 pened. But it was watched 

 rather carefully in a number of 

 instances, in order that the move- 

 ments might be noted. In some 

 cases, the body seemed to undergo 

 a rapid whirling motion, com- 

 parable with that of a rope tied 

 at one end to a post and held 

 by the hand at the other. The 

 snout and tail, in this comparison, 

 correspond to the points of attach- 

 ment, the hind quarters of the body 

 to the whirling central portion of 

 the rope. A number of individuals 

 which did not actually detach the 

 tail were found to undergo less vio- 

 lent gyrations of the same sort. 

 In other instances, however, there 

 seemed to be an actual torsion of 

 the appendage, resulting from a ro- 

 tation of the body. In yet another, 

 \vhere the tail was held down upon 

 a cement floor, the detachment was 

 effected by a sudden leap. 



In eight cases, in which the de- 

 tached portions of the appendage 

 were carefully examined, it was 

 found that the fracture invariably 

 occurred across one of the caudal 

 vertebrae, commonly near its center. 

 The tail is likely to be broken at 

 almost any point in its length. The 

 detached portion carries aw r ay with 

 it a group of tendons of such length 

 as to indicate that these have their 

 points of origin within the body 



