Iss” NATURAL SCIENCE [September 
killed were taken at random without selection. Thus Darwin’s 
system contained its own refutation within itself. A later school 
of evolutionists have maintained that the effects of habits or 
conditions on the individual are not inherited, and therefore not 
cumulative. According to this view, only those variations are 
hereditary which arise in the germ, in the internal constitution of 
the egg; such variations are supposed to be numerous and to take 
place in all possible directions, and natural selection is supposed to 
pick out from among them those which are advantageous and so 
accumulate them. I do not propose here to discuss the various 
theories of heredity. The question of the possibility of the trans- 
mission of acquired characters, or the determination of congenital 
modifications by the direct influence of conditions is a very important 
one, and has been much discussed. But I wish to draw attention 
to a mode of considering the subject which is generally neglected, 
namely, the inductive method. The doctrine of evolution is an in- 
duction from the facts of zoology; in my opinion conclusions con- 
cerning the method and the causes of evolution can also be obtained 
as inductions from a sufficiently wide survey of the phenomena. 
Every one will admit without hesitation that all variations must 
be due to causes. But, according to the selectionists, hereditary 
variations have no primary and essential relation to the require- 
ments of life. Such variations occur in all or many directions 
indefinitely, and they are so diverse that by the survival of a few 
individuals out of the many that are generated, the complicated 
adaptations which we know have been gradually produced. It is as 
though we conceived of a table being produced by the process of 
selecting from a large stock of pieces of wood of all shapes and 
sizes, those which were of the shape and size required, and joining 
them together; and not by the usual process of sawing and planing 
the various parts into shape out of a stock of planks all originally 
similar. Thus selection preserves and combines the variations which 
are most advantageous under the given conditions, but the relation 
between the structure and the outer world has no hereditary effect 
in moulding or shaping the structure. Romanes maintained with 
much truth that natural selection was a theory only of the origin of 
adaptations, and not necessarily of the origin of species, but it is 
further necessary to realise that it originates adaptations only in the 
sense of preserving and combining the variations or modifications 
which occur, and which happen to be advantageous. It may be 
said to combine only in the sense of causing different variations in the 
parents to be transmitted together to the offspring, and of allowing 
new variations to occur only in the individuals which have survived. 
Now it is possible by actual observation to ascertain what 
evidence there is, that variations which might by natural selection 
