80 SCIENCE. 
a eycle of phenomena, of whatever kind, 
requires the initiation of the original varia- 
tion in which it began. The same is true 
of any variation thereafter, if we are to ap- 
ply the reasoning which is valid regarding 
physico-chemical laws of matter to the 
phenomena of organisms. If the variation 
be primitive, and normal, to use the word 
in the sense proposed, it is evident that 
what we call reproduction is but a pulsation 
of the phenomena of life itself, recurring in 
the precise cycles which express the equili- 
brium between a definite quantitative force 
in the individual organism itself and the 
definite quantitative forces of the total en- 
vironment in which it carries on its life 
functions. 
The preservation and perpetuation of de- 
rived characteristics in a race seem at first 
sight to be easily accounted for by the pro- 
cess which Darwin has called natural selec- 
tion. _ Variation, in this hypothesis, is sup- 
posed to occur ‘ naturally,’ by accident, or, 
as Darwin says, ‘spontaneously.’ But a 
close examination of what such a proposi- 
tion would mean in concrete facts reveals 
serious difficulties. The apparent sim- 
plicity arises from the assumption that the 
law of hereditary transmission of ancestral 
characters is a primary law of organism, 
which is violated in every case of variation. 
On such an assumption we have only to con- 
ceive of the removal of whatever may have 
occasioned the accidental or spontaneous 
disturbance in order to permit the continu- 
ing on of the normal working of heredity. 
But when we follow the hypothesis back to 
its beginning, it provides no means for ris- 
ing from the original level ofsimplicity. Each 
variation must, according to the theory, 
be a violation of the normal action of the 
organism ; hence if the organism were ad- 
justed when the variation took place the 
variation puts it out of adjustment, and we 
have no place for the action of natural 
selection. If, on the other hand, the varia- 
[N. 8. Vou. VI. No. 133. 
tion is advantageous to the economy of the 
organism, then we must assume either that 
the organism was not in perfect adjust- 
ment when it varied ; and then again the 
adjustment is accounted for without the 
action of natural selection, or else the lack 
of adjustment came from change of condi- 
tions. In this case the conditions of en- 
vironment, not natural selection, account 
for the adjustment. And there seems to 
be a still greater difficulty, viz., the ex- 
treme length of time necessary to bring 
about the changes that have taken place by 
the process. Recently Professor Poulton 
called attention to the necessity of this 
great time period (longer than the physi- 
cists or geologists are generally ready to 
allow to have been possible), in order to 
account for the results we find recorded 
in the fossil-bearing rocks, requiring at 
least 400 millions of years for the work of 
evolution.* But Mr. Poulton does not ex- 
aggerate the matter. 
Let us examine this time factor and see 
if we can imagine it to have been long 
enough. In the first place, if hereditary 
repetition be the normal law of organism, 
then Professor Poulton has made a fair es- 
timate of the ages it would take according 
to the present rate of evolution. But he 
has not taken into account all the necessi- 
ties of the theory, two of which must be 
these: First, if the exact hereditary trans- 
mission and repetition be the fundamental 
law of organisms, not only must the prog- 
ress produced by any mode of variation have 
been exceedingly slow—at first at a rate 
decreasing geometrically in proportion to 
the greater simplicity of the organism; but 
second, the theory requires that if natural 
selection consists in making variations per- 
manent, the general progress must take 
place by means of a process which in every 
particular case consists in stopping the very 
phenomena by which the progress is at- 
*SCIENCE, Vol. 54, p, 504. 
