Effect of Successive Removal upon Regeneration 503 
for present comparisons, the remainder being discarded because 
of death, escape or accidental injury. The animals were kept in 
individual numb red finger-bowls. Food consisting of pieces of 
Fundulus was supplied every day except August 17, and the water 
was changed every day. The first operations were made on 
August 15th. The order of procedure was as follows: 
a The right chela was removed in twenty individuals, and the 
left in twenty-five individuals. 
b Both right and left chela were removed on the day after the 
first molt which gave a nearly complete regeneration of the 
removed organ. In some cases this was the first, in others the 
second molt. About seven-tenths of the original length was ordi- 
narily regenerated at the time of a second removal. 
c On the day after the next succeeding molt, the animals were 
preserved in 95 per cent alcohol, and the right and left chelz were 
compared. 
The measurements used in the present comparison consist of 
the final thoracic length to the tip of the rostrum, and the final 
chelz lengths, the second regeneration on one side of the body and 
the first on the other. The lengths given for the chelz are the 
lengths of the propodites in millimeters. 
Sources of Error. Because of the method of procedure most 
of the sources of error enumerated under the other experiments 
were eliminated, but at the expense of a restriction in the applica 
tion of the results as pointed out above (p. 502). 
1. Inherent psychological difference between the two chele of an 
individual. ‘This is eliminated by the use of a large number of 
individuals with the second regeneration nearly equally distributed 
between r ght and left chela. ‘There are seventeen first and four- 
teen second regenera ions of the right chela and fourteen first and 
seventeen second regenerations of the left chela. 
2 The molting habit. The successive removal may affect the 
rates at certain parts of the regeneration process without exerting 
an influence upon the total amount regenerated during a molting 
period. For nstance, if after a first removal the whole regenera- 
tion to be accomplished during the molting period 1s practically 
completed during its first half, the effect of successive removal may 
