20 
After the statements which have just been made, it can 
scarcely be said that the theory of Natural Selection is 
quite so definite or self-consistent as it is generally sup 
posed to be. It is not quite certain that he who reads it 
as he runs will be sure to read aright, or that the wayfaring 
man will not be liable to err therein. 
Now, in the first place, let me say that it is quite natural 
that there should be differences of opinion among different 
persons, and the first thing one has to remember is that 
Natural Selection may be held with various modifica 
tions. Some may believe that it is the one only method 
of the transmutation of species, others may hold that 
possibly there are many methods, among which Natural 
Selection is one. Some may follow Ray Lankester 
and Weismann, others may follow Darwin and Romanes. 
On this and other points we have modifications of the 
theory which almost amount to rival theories, and the 
attempt to reconcile the two would be a mere waste of 
time. 
In the second place, a discrepancy, if it exists, may arise 
from the fact that even an eminent specialist sometimes 
writes from the point of view of the logical exponent of a 
theory, and sometimes in the character of a careful observer 
of nature. Both statements may be correct in their own 
field. Assuming certain data as the axioms of a theory, 
a man may reason as to what the result must be : looking 
out on nature, he may honestly record what the fact 
actually is : and he may not be quite aware of the discrep 
ancy which exists between the two sets of statements. 
This may explain Mr. Darwin’s treatment of the question 
whether variations are small or large. 
Mr. Darwin asserts that Natural Selection deals only 
with slight variations ; that on the theory of Natural 
Selection the variations must be slight ; and in nature