Frpruary 18, 1904] 
NATURE 
367 
theory, but before considering them let us see how Hering’s 
theory affects our ideas of variation. 
In the first place it gives an explanation of the definite 
variation which we see in the development of non-adaptive 
or useless characters. A variation, once started, would in 
the future have a tendency to be reproduced, and this 
tendency would get stronger and stronger as the memory 
is reinforced by repetition, and when once established the 
variation would be quite definite. New variations may be 
indefinite, but they must either die out or become definite ; 
and we see by Hering’s theory why useless characters may 
be as constant as useful ones, for constancy depends upon the 
number of repetitions and not on the nature of the variation 
or on the reason for its survival. This includes, of course, 
use-inheritance, for according to the theory, when an organ 
is constantly exercised the memories of the component parts 
are strengthened, and in the next generation the organ is 
reproduced better developed than in the last. It is the same 
with instincts; they are the inherited modifications of mental 
operations, while a structural development is due to the in- 
herited modifications of physical operations. When an 
organ is not used the memories of the parts are weakened, 
and in the next generation the organ is reproduced in a 
more feeble condition, until at last it is not developed at 
all, the memory of the operation having been lost. The 
process is exactly the same as the gradual loss of an instinct 
from disuse; both are due to forgetfulness. 
With regard to the action of external causes, Hering says 
that each generation endows its germ with some characters 
acquired during life. But we cannot suppose that adapt- 
ations to new circumstances are directly produced by the 
action of the surrounding conditions. For example, the 
fur in many animals gets thicker in cold climates and some 
plants get spiny coverings in dry climates. These cannot 
have been directly produced by the action of the climate, but 
must be due to the action of the protoplasm resisting the 
climate. Dry air could not directly produce the spines on 
a plant any more than it could produce the water-pouches 
in the stomach of a camel. Neither could feeding on nectar 
have produced the honey-bag of a bee, for it would be absurd 
to suppose that sucking liquid through a tube could cause 
a projection to grow out of it. We might as well say that 
rain and wind build houses or that snowstorms make 
great-coats as to suppose that the action of external in- 
fluences made the cell-wall or the thick fur. Evidently it 
is the living protoplasm which originates these adaptations 
to protect itself from the rough elements or to prevent itself 
from being poisoned or starved. But how variations 
originate, whether they be intelligent and purposive, or 
whether they be blind, haphazard gropings after some 
change when the protoplasm feels uncomfortable, Hering’s 
theory does not tell us. 
There are other facts connected with variation which are 
explained by Hering’s theory. As the germ contains two 
different memories, derived from its two parents, these may 
clash and antagonise each other, and so allow an older but 
dormant memory to be stimulated into activity. This is 
atavism. Or degraded characters which have suffered from 
disuse can, on a renewal of the old stimulus, again be re- 
callea, as we see in proteus, which gets dark in colour 
when kept in the light. Prepotency can also be explained 
on the supposition that the germ of one parent has stronger 
memories than that of the other; and the reproduction of 
lost parts may perhaps be due to the memory of the remain- 
ing portions trying to replace the lost portion. In the same 
way we see that mutilations could not produce degener- 
ation or the loss of a part, no matter for how many gener- 
ations they may be carried on, because the part develops 
and the stimulus has been given before the part is removed. 
Again, the fact that variations appear at an earlier stage 
in the offspring than in the parent may be taken as evidence 
that they are due to an excited memory which anticipates 
events. But I do not see how Hering’s theory can explain 
the infertility of hybrids. Conflicting memories might lead 
to inaction, but I cannot see why these conflicting memories 
should arise until the time had come to differentiate the 
embryo into the form of one or other of the parent species. 
This would give rise not to sterility, but to abortion, while 
it is thought that the foetus generally perishes at an early 
stage of development. 
NO. 1790, VOL. 69] 
Now let us consider the obstacles to believing in Hering’s 
theory. 
In the first place it may be objected that it is impossible 
to suppose that the small ovum, or still smaller spermato- 
zoid, could contain all the memories necessary for building 
up the adult organism. This is an objection which applies 
to all hypotheses except epigenesis, and it is of considerable 
weight. However, the capacity of the germ-cells for 
storing up memories is not unlimited. It is only very few 
indeed of the impressions stored in the brain that are also 
registered in the germ-cells, and this, I think, is favourable 
evidence. 
Next we have the difficulty of understanding the trans- 
mission of variations from different parts of the body to the 
germ-cells. This difficulty also is not peculiar to Hering’s 
theory, but is common to all, and however difficult it may 
be to understand, we know that, with instincts, it is a fact. 
Lyarwin certainly said that it was an error to suppose that 
instincts were inherited habits, for they were due to natural 
selection. Romanes, following him, said that some in- 
stincts owed their origin to natural selection, while others 
were inherited habits. But natural selection, as Darwin also 
often said, cannot originate anything. It can only develop 
characters which are transmitted, and if habits—which are 
only mental variations—were not transmitted, natural 
selection could not develop them. These mental variations 
must have been transmitted by some physical process from 
the brain to the germ-cells, and adaptations of all kinds 
must in like manner have been transmitted, or there would 
have been no progress in the animal and vegetable 
kingdoms. E 
For instincts in animals must have been acquired either 
by inheritance or by imitation, and we have only to select 
instances where imitation is impossible to prove that in- 
stincts are inherited. For example, when a newly born 
baby cries, it is not imitating anyone in the room. It is 
repeating what its father and mother did in similar 
circumstances. It is the same with breathing. This must 
have commenced as a semi-conscious act which quickly 
passed into a habit and then became instinctive. When 
the crying of babies first began I do not know, but breath- 
ing has been instinctive ever since the Carboniferous period. 
Millions of generations, one after the other, have performed 
the operation, and it is now out of our power to stop it. 
Again, young fish never see their parents, yet they. follow 
their habits, as also do young cuckoos and many insects. 
But I need not multiply examples; these are sufficient to 
prove that instincts are transmitted. If instincts are 
transmitted it must be through physical modifications made 
in the brain, and if this is the case there can be no doubt 
but that other physical modifications, not in the brain, can 
be transmitted also. ; 
Prof. Hering says that the nervous system, which collects 
impressions from all parts of the body and transmits them 
to the brain, transmits them also to the germ-cells. But 
in plants and in animals without a nervous system the 
protoplasm itself must do the work, and it is therefore 
possible that the nervous system may not be used for this 
purpose in the higher animals. This is a question for 
future biologists to solve. But whatever the explanation 
may be, we must recognise as a fact that variations in 
external characters influence the germ-cells, and that the 
germ-cells reproduce these variations. If we call the 
analogous process in the brain memory, we must either 
apn!~ the same term to the process in the germ or invent a 
new one. 
Now we come to the last great difficulty, that of 
believing mind and memory to exist in the tissues of 
animals and plants. The best way of examining this 
difficulty is to ask ourselves What we mean by life? and 
How we recognise living matter? 
As everyone knows, we recognise its presence by certain 
movements which are distinguished, without much diffi- 
culty, from movements due solely to physical energy. A 
bird flying through the air is alive, as also is a seed if, when 
placed under certain conditions, it commences to grow. 
Assimilation, or feeding, is the basis of all these movements. 
It supplies the materials for growth and the energy 
necessary for the movements. 
This process of assimilation is only found in protoplasm, 
