SENSORY STIMULATION OF THE OYSTER, OSTREA YIRGINICA, BY CHEMICALS 251 
sodium hydroxide and potassium hydroxide. The alkalies are often thought to be 
stimuli for the common chemical sense rather than for taste. 
The results are presented graphically and each point represents the average value 
of from 10 to 20 latent period determinations. Five minutes were allowed to elapse 
between succeeding tests to prevent possible fatigue of the sensory endings. A single 
complete series, involving tests for enough concentrations to produce a relatively 
complete curve, was made within two or three successive days and under conditions 
as nearly constant as possible. The temperature of the water during a series varied 
over approximately 3° C. and was in general close to 20° C. It is not known at pres- 
ent how great is the error due to temperature fluctuations, but the probable error in 
any group of tests with a single concentration did not exceed ±0.05 second, except 
in certain cases which will be mentioned later. 
CUMARIN 
In general the latent period values were not more than about five seconds, for 
above this the reaction was so weak as to be indistinguishable from the frequent 
slight movements of the 
tentacles. However, in 
the case of cumarin the 
reaction was sharp and 
clear up to about 14 
seconds, although the 
probable error increased 
to about ±0.3 second. 
In Figure 1 the results 
of a series with one 
specimen are presented. 
It is unnecessary to give 
more for, aside from a 
slight individual varia- 
tion, the other series 
agree perfectly with this 
one. Concentrations 
from 0.0004 to 0.011 per 
cent were tested, and the points as shown in the figure fall with reasonable ac- 
curacy into a hyperbolic curve. Crozier (1918, 1918a) obtained similar curves in 
studies on the stimulation of the earthworm by acids and alkalies. The relationship 
between concentration (C) and latent period (T) is what would be expected for a 
simple monomolecular chemical reaction; namely, CxT = K. 
On the same graph the data are given logarithmically, as logarithmic concentra- 
tion (abcissse) and logarithmic 1/latent period (ordinates). The points fall close to 
a straight line with the exception of that referring to the highest concentration. 
Whether the latent period at high concentrations actually reaches a limit and changes 
the form of the logarithmic curve or whether the error is too high to permit accurate 
determination is not quite clear. The latter possibility, however, is very likely, for 
at high concentrations the latent period is extremely short and the reaction involves 
not only the tentacles but also the adjacent portion of the mantle. 
