458 KINSHIP AND ADAPTATION 
a single plant. These three forms were found to come true 
to seed and when crossed the seedlings were like one parent 
or the other; or occasionally other new forms were produced, 
with or without crossing, which were as unlike the parent 
form as were the two new forms first discovered. In the 
course of seven generations several such new forms appeared, 
numbering in individuals about eight hundred out of a total 
of about fifty thousand plants raised. Similar forms also 
appeared in the field. All these new forms were distinct 
types differing from one another in several particulars such 
as shape and color of leaves, flowers or fruit, and annual, 
biennial, or perennial habit; and although growing together 
they did not intergrade. They were thus as sharply definable 
as species are, and they could be regarded as true species 
except for the fact that their differences distinguished parent 
from offspring. De Vries regarded them as elementary species 
equivalent to what botanists have been calling well-marked 
varieties, races, or subspecies, and as representing the abrupt 
changes through the accumulation of which species and higher 
groups arise. He calls variations of this sort mutations, the 
new forms being termed mutants. Hence we may designate 
as mutationism this new theory of evolutionary change. 
Mutationism supposes that from time to time, especially 
under the influence of changed conditions, some of the in- 
dividuals of a species bear offspring distinctly different from 
the parent, often in several particulars, and that the set of 
peculiarities thus suddenly arising is completely hereditary 
even when individuals of the new form are crossed with 
other forms of the same species. That is to say, the progeny 
from parents of two different mutations are like one or the 
other parent, and never intermediate as is often the case with 
crosses between fluctuating variations. If mutations again 
mutate and we have mutants of the second or higher degree, 
their hybrids may be wholly like one parent or the other, or 
may be apparently intermediate from having inherited one 
or more peculiarities from one parent and the rest from the 
other; but such cases of apparently intermediate forms show 
new combinations of elements rather than elements of an 
1 Mu-ta’tion < L. mutatio, a changing. f 
