30 
BULLETIN OF WISCONSIN NATURAL HISTORY SOCIETY. VOL. 5, NO. 1. 
Perhaps it may seem rash to assert that in no one of these six 
points does the mutation-theory hit the mark ; nevertheless, that 
is the conviction from which I see no escape. 
The one error that vitiates all these contentions is that of unit- 
characters, each of which is supposed to enjoy a sort of indepen- 
dent existence, although in correlation with the other units of the 
organism. Such a conception implies that every character of the 
adult organism pre-exists in some rudimentary form in the germ- 
cell from which the organism develops. These assumed rudiments 
of unit-characters must be as fixed in number and constitution in 
the fertilized egg as in the adult. ^Moreover, unless we assume 
that these units may mutate at any stage of development, we must 
suppose that all mutations destined to appear in the adult must 
originate in the very first stage of existence of the primordial 
units. Thus mutation would be carried back to the pre-mutation 
forever beyond the reach of investigation. Such conceptions, help 
us in no wise to understand the origin of species. To claim that 
w^e can actually see mutation performed, is the climax of absurdity. 
De Vries has seen offspring differing more or less constantly 
from the mother plants. These visible differences are referred to 
invisible differences in one or more of the invisible unit-characters 
conjectured to exist in the seed before germination. The initial 
differences, in which, ex hypothesi, the whole mutation is given, 
De Vries has never seen and of course never expects to see. If 
there be any such thing as mutation, as conceived by De Vries, it 
is safely beyond human ken, and unapproachable through experi- 
mental investigation. 
This theory of mutation coincides well with Bateson's doctrine 
of disconfiiiiiify in evolution and development. Discontinuity in 
evolution has all the elusiveness of mutation. The discovery of 
such a negative is tantamount to a failure to discover anything. 
A new form appears ; it differs in one or more respects from its 
parents. The difference is labelled a discontinuity, i. e., the off- 
spring is assumed to have obeyed the law of heredity in so far as 
it resembles the parental stock, and to have jumped hereditary 
bonds in so far as it departs from such resemblance. The jump is 
a break in the chain — a void where eyesight fails and all sorts of 
ghost-stories are accepted in lieu of reliable knowledge of all the 
antecedent conditions represented in the ancestry. 
The less we know of the previous history of new forms, the 
