664 HENRY H. DONALDSON 
The formula for the weight.of the central nervous system was 
accordingly written 
Weight C. N.S. = (Log. W x yL)C 
where ‘weight C. N.S.’ is the combined weight of the brain and 
spinal cord in milligrams; W the body weight in grams; L the total 
length of the frog in millimeters, and C, a constant empirically 
determined. 
Corresponding results were obtained for both the bullfrog and 
leopard frog (Donaldson ’02). 
By this formula it is possible to calculate the approximate 
weight of the central nervous system of the frog from the data 
on body weight and total length, and also to show its growth. 
The observations used in the foregoing study, from which the 
formula was obtained, were taken from summer frogs (%.e., in 
the case of the bullfrogs, July and August, and in the case of the 
leopard frogs, June and July). 
In commenting on these results, I pointed out at the time that 
it was necessary to avoid several sources of observational error. 
These are represented (1) By variations in the moisture of the 
frog, and therefore only frogs that have been kept moist for some 
hours at least, should be used. (2) By loss of weight during cap- 
tivity, especially in frogs taken in the spring and early summer. 
Hence such frogs must be examined either as soon as caught or 
must be kept under special conditions or some correction must be 
made for the loss which they undergo. (38) By season; as I noted 
that both in the few spring and autumn frogs which I had exam- 
ined, the nervous system was apparently relatively lighter than 
in frogs killed during the midsummer. 
In the course of this work, the first two sources of error were 
taken into account, and corrections made where they were deemed 
necessary. Also, as just stated, the third was escaped by using 
summer frogs only. 
The difference thus found between the relative weight of the 
central nervous system in the summer andin thespringandautumn, 
appeared to me worth further examination, for unless it could 
