408 Marion L. Durbin 
The graphic method probably gives the most satisfactory repre- 
sentation of these changes in rate. A curve of rate may be con- 
structed using the abcissz as the intervals since the operation and 
the ordinates as the rates of length regeneration per day. Minot 
expresses the rate of growth as per cent of increment per day; 
that is, the amount of increase each day is divided by the weight 
of the animal at the end of the day before. These curves of regen- 
eration rates are not per cent of increment curves, because the 
i] 
ty 
Rate per day in millimeters 
° 
ts 4 6 8 11 13 16 
Days since operation 
Fig. 3 Curve of rates of regeneration for Experiment III. For explanation, see Fig. 1. 
amount of material added each day is so small in comparison 
with the length of the rest of the tadpole that the latter can be 
considered a constant. 
The curve of rate of regeneration is made up of four distinct 
regions showing the rate to have undergone four distinct kinds 
of changes. The first is the initial region of low rate; it is most 
clearly exemplified in the curve of Experiment III, Fig. 3, in 
which the first measurement was made two days after the opera- 
