52 
by jars or other forms of mechanical stimulation except when it is 
necessary to change the solution in the container. 
Since the unit that most concerns us in this paper is the time 
rec£uired by the pupil to dilate to a maximum it becomes necessary 
to explain a few points in this connection. (1) In some eyes the 
long axis is the first t'o reach a maximum length; at other times it 
is the short axis. (2) The axis first to reach a maximum length 
usually shortens while the other axis is in the process of elongating. 
It is only occasionally that both axes reach their maximum length 
simultaneously, or that the one waits upon the other. Hence it often 
becomes difficult to determine the exact moment at which the eye 
is most ^videly dilated unless the absolute area of the pupil be calcu- 
lated. This is too tedious a process, so as a rule one of the following 
methods is used: (1) The mean of the time required for the two axes 
to reach their greatest length; (2) the moment of maximum dilation 
determined by the first axis reaching its greatest length or turning 
point. Unless otherwise stated, the last method, though less accu- 
rate, is the one used throughout the paper. 
I always endeavored to choose frogs of the same variety, that 
weighed not less than 17 grams nor more than 30. The head was 
cut off just back of the eyes and the bulbi removed from their sockets 
at once. In removing the eyeball undue pressure was avoided and 
care taken not to cut the sclerotic coat. The bulbi were immedi- 
ately dropped into a container of Kinger solution, observed and 
measured. When necessary to change the solution, it was carefully 
pipetted off and a fresh solution of nearly the same temperature 
substituted. 
In spite of all precautions certain eyes will dilate and remain 
dilated in the Ringer solution for an unusual length of time. When 
this occurs they should be rejected. Furthermore, eyes from frogs 
of low vitality seem to be more susceptible to adrenalin than those 
of healthy, vigorous animals. This source of variation is the most 
difficult to overcome, and perhaps accounts for many of the varia- 
tions that occur in experiments of this nature. 
