672 HENRY H. DONALDSON ~ 
are much the same (table 1) then it would follow that the size of 
the probable error of the mean would rapidly reduce as the num- 
ber of cases was increased. It is by reason of this fact that I still 
consider the successive mean values of C significant, despite the 
large probable error. 
When the data in table 3 are put in the form of a chart (1) 
where the ordinates represent the values of C on a base line of 
time in days, the relations above described are shown clearly. 
We conclude therefore from this series that the relative weight 
of the central nervous system of the leopard frog rises from the 
time that the frog appears in the spring until midsummer, arid 
297 VALUE OF C. CHART | 
2 CHICAGO 
FEB. MAR. APR. MAY. JUNE. JULY AUG. SEPT. OCT. NOV. DEC. JAN. 
Chart 1 Based on data in table 3. Determination for C, June 29, not entered. 
then falls until the frog goes into hibernation, 7.e., when the 
external temperature drops to 7°-10° C. =45°-50° F. (Torelle, 
03). It should be added however that this critical temperature 
is probably modified by latitude and tends to become lower as 
we pass from south to north in the range of the frog. 
In table 2 are given the necessary fundamental data relating 
to the several groups belonging to this lot of Chicago frogs. 
Observations on the frogs from Minnesota, 1908 
The exact locality of the frogs in this series is not known. 
They came however from Minnesota, probably from southern 
Minnesota and all from the same station. They were delivered 
in good condition in Philadelphia where they were examined. As 
