ORIGIN OF ADAPTATIONS AND OF SPECIES 241 
mulation of such advantageous variations as already exist. 
Variation, then, is the basis of natural selection. Though the 
question of the origin of variations is still unsettled, the fact 
of their occurrence in a manner sufficient for the purposes of 
natural selection is beyond dispute. No two individuals of a 
species are ever exactly alike in structure or behavior, and 
their differences furnish the material for the operation of 
natural selection. . 
Two classes of variations are distinguished on the basis of 
the amount of variation: (1) continuous (individual) varia- 
tions, of small extent, intergrading with one another and with 
the typical form; and (2) discontinuous variations (sports), 
or considerable and isolated departures from the normal con- 
dition. Furthermore, variations of either class are adaptive 
or unadaptive, the latter kind being either harmful or simply 
neutral. 
Origin of Adaptive Variations.—Natural selection, as 
was said, does not begin to-operate until useful variations are 
already in existence; and the origin of these primary adaptive 
variations is a question quite distinct from that of their sub- 
sequent preservation and accumulation by natural selection. 
That all adaptive variations are due to the response of pro- 
toplasm to environmental influences (using the term “ envi- 
ronment ”’ in its widest sense), it goes without saying. These 
variations are, however, either direct or indirect. Direct 
variations, appearing first in the soma, or body, of the organ- 
ism, are termed somatogenic; indirect variations, apparently 
spontaneous, and due immediately to the germ cells, are termed 
blastogenic. Weismann places somatogenic variations, ac- 
cording to their origin, into three categories: (1) imyuries, 
(2) functional variations, and (3) variations depending on 
the so-called “influences of environment,” these influences 
being mainly climatic. These three kinds will receive brief 
consideration. 
Injuries.—There appears to be no good evidence that in- 
juries or mutilations can be transmitted. Nearly all the ex- 
17 
