INLAND FISHERIES. 71 



generate if a portion of the disc is absent, and that the rate of 

 regeneration, like the rate of growth, in normal specimens, is 

 dependent upon the nourishment, for, while the new growth in 

 these cars Avas only 7 mm. in six weeks, those which were well fed 

 at Kickemuit gained a new growth of from 13 to 18 mm. in the 

 same time. See page 65. 



Other experiments were tried, to determine what regeneration 

 would take place if the whole starfish were cut through in various 

 ways, while the arms were left in place. It will be seen that the 

 results were not always the same. 



In the summer of 1897 several stars about 2^ inches in length 

 were cut through so as to leave three arms and part of the disc on 

 one piece, and two arms and i^art of the disc on the other. The 

 smaller pieces perished, but the larger ones lived for several ^veeks 

 and showed no sign of regeneration. All but one Avere destroyed 

 by other starfishes, which got into their compartment of the car 

 by accident. The remaining fragment, consisting of three arms 

 and part of the disc, lived several months after the operation and 

 did not regenerate. 



On May May 11, 1898, several stars about one inch long were 

 cut as in Fig. 27. One arm was pulled ofi' and the disc then cut 

 in two, leaving two arms with a part of disc and madreporic plate 

 on one piece (=A), and two arms and part of disc on the other 

 piece ( = B.) The fate of the single arm has been already con- 

 sidered, page 70. On June 9 all the pieces were alive. In the 

 piece marked A (/. e. having madreporic plate) a trace of a new arm 

 arm was visible on the side toward the lost arm, but in no other 

 place. The pieces marked B showed no regeneration at all. On 

 June 25, six and a half weeks after the operation, the condition 

 w^as as follows : All these parts of specimen 1 were alive. The 

 fate of the single arm is mentioned on the previous page. The 

 piece (A) with madreporic plate is sketched from the lower side 

 in Fig. 28. Two arms were well started, and one minute arm was 

 growing out between them. In the other piece (B) of this speci- 

 men, the wound was completely healed, but there was no visible 



