Effect of Successive Removal upon Regeneration 495 
dence of regularity of time of the first molt after February 11 
indicates that all had molted the same number of times. Since 
there is no certainty as to the number of previous periods, the first 
molt after February 11 is called molt 1, the second molt 2, etc. 
In the data of the experiments the period during which a regenera- 
tion occurred is indicated by its bounding molts. ‘Thus period 
6—7 indicates that the Operation was made two days after the sixth 
molt and that the regeneration continued until two days after the 
seventh molt. 
The specific amount of regeneration is the length of the regener- 
ated chela divided by the final thoracic length or the length regen- 
erated per millimeter of theracic length. The specific rate is 
the length regenerated per millimeter of thoracic length per day. 
The specific rate as already stated (p. 493) is not a proper basis 
for comparison. ‘The data are classified in the tables for direct 
comparison between the first’? and third or fourth regenerations. 
Sources of Error. « Age. All the individuals belonged to 
a single brood. However, since the crayfish did no molt at the 
same rate and since comparisons are made between corresponding 
molts and not between individuals of the same age, the age factor 
was not perfectly controlled. ‘The error due to this source is, how- 
ever, probably slight since no great differences in age exist. 
2 Periodic Physiological Changes. Molting. This very im- 
portant factor was controlled as far as possible by several devices. 
a_ All operations were made two days’ after the molt. 
b ‘The molting period and not the day is used as a unit. 
c Comparisons are made between corresponding periods. 
3 Laboratory History. All individuals belonged to the same 
brood and had the same laboratory history. 
4 Changes in Rate during the Regeneration Period. Because 
of the necessity of controlling the molting factor the time period 
is not the same in compared individuals. However, since all or 
nearly all of the regeneration comes in the early part of a molting 
period, and since whole periods were used as units in the compari- 
son, this source of error is probably negligible. 
17 Second in one individual. 
