7] RATE OF REGENERATION— ZELENY 



INTRODUCTION. 



The present studies of the factors controlling rate of regeneration 

 are a continuation of previous work on the subject. An advance in 

 knowledge concerning certain of the factors has made possible an exten- 

 sion of the experimental analysis of others. The present studies are 

 therefore closely related. In fact in several cases a single series of indi- 

 viduals has been of value in connection with more than a single factor. 

 The definite determinations of the effect of level of the cut and of the 

 change in rate during the regeneration cycle have been of particular 

 value. 



The precautions taken to meet the demands of the experiments are 

 not discussed in detail because they have already been given in previous 

 papers. The frog tadpoles (when they can be used) are in all respects 

 more suitable than salamander larvae. When collected late in the fall 

 they can be kept at a fairly constant size and the results obtained under 

 these conditions are not complicated with growth phenomena. They 

 have proved to be remarkably uniform in several series. The salamander 

 larvae on the other hand vary in rate of regeneration from day to day. 

 The factors involved in this fluctuation were not discovered and could 

 not be remedied but may be related in some way to the fact that these 

 animals require living active food and the feeding reactions are therefore 

 more complicated than in frog tadpoles and more subject to disturbance. 



In regard to certain factors, such as the degree of injury, in which 

 expected differences in rate are slight the writer has felt that he might 

 be biased in making the measurements and in a number of cases this 

 work was therefore delegated to a person who had no preconceptions 

 concerning the result. 



In making averages elimination of individual cases is avoided 

 except for a few very aberrant values. Such exceptional values are in 

 every case however included in the tables. In many cases where only 

 slight differences are to be expected several different kinds of comparisons 

 are made so as to bring out the correct relation as completely as possible. 



As in the past all data obtained by the writer on the particular 

 factors in question are given. The practice of selective elimination would 

 be dangerous because of the large value of factors not at present under 

 experimental control. 



