484 D. W. CUTLER. 
the end of the experiments. In no case was there any discrepancy 
between the two results. 
In the next column are given the details of the scale growth for the 
periods between the examinations. First there is the number of sclerites 
formed during each period, followed by the maximum and minimum 
breadth of the sclerites. Finally I have given the increase in length of 
the animals during the periods and their general condition. 
Figures of partial scale curves are reproduced, the portion figured 
being that part of the scale produced during successive periods of experi- 
mental conditions. The micrometer unit employed was the same as 
that for the curves in Part I. 
CONTROL FISH. 
The control tank, Table II, shows how uniform was the growth of 
both scale and fish during the two periods of the experiment. With the 
exception of those fish whose condition was not very good the increase 
in length from January to May is remarkably constant, as is also the 
number of sclerites formed. Also the maximum breadth is between 
0-9 and 1-1 and the minimum between 0°8 and 0-6. Fig. 5, which repre- 
sents the scale curves of some of these fish, is interesting in showing how 
closely the curve follows the temperature changes. They all start with 
a relative breadth for the sclerites of 0-9 or 0:8, which falls to a 
minimum in about the middle of the curve, then gradually rises to 
either 0-9 or 1-1 When we remember that the temperature in January 
was 11-8° C. falling to 9-7° C, in March and gradually rising to 13° C, by 
May, these curves seem very significant. 
The growth during the period from May, 1916, to October, 1916, is 
equally interesting. The increase in length is a little greater than during 
the previous portion of the experiment, as would be expected. As regards 
the sclerites we note that the maxima and minima are greater in accord- 
ance with the higher temperature during these months. The scale curves 
also follow very closely expectation, if we assume that temperature 
is the directive agent in the production of wide or narrow sclerites. 
In the curves of scales figured for the period, January to October, I 
have drawn a line through the portion where the growth up to May ended. 
This procedure I have followed in all the scale curves reproduced in the 
fioures. 
HOT AND COLD TANKS. 
At the commencement of the experiment great trouble was encountered 
in keeping the animals in the hot tank in good condition, and several 
times the experiment had to be restarted owing to the fish either dying 
or getting into extremely poor condition. This was possibly due to the 
daily rise and tall of the temperature caused by the imperfect apparatus 
