SEPTEMBER 21, 1899 | 
even in members of the same family, z.e. in the offspring of the 
same parents. It is these differences to which we apply the 
term variatzen. The immense importance of the study of 
variations may be judged from the fact that, according to the 
generally received evolution theory of Darwin, it is to them that 
the whole of the variety of living and extinct organisms is due. 
Without variation there could have been no progress, no 
volution in the structure of organisms. If offspring had always 
exactly resembled their parents and presented no points of 
difference, each succeeding generation would have resembled 
those previously existing, and no change, whether backwards 
or forwards, could have occurred. This phenomenon of genetic 
wariation forms the bedrock upon which all theories of evolution 
must rest, and it is only bya study of variations, of their nature 
and cause, that we can ever hope to obtain any real insight into 
the actual way in which evolution has taken place. Notwith- 
standing its importance, the subject is one which has scarcely 
received from zoologists the attention which it merits. 
Though much has been written on the causes of variation, too 
little attention has of late years been paid to the phznomenon. 
Since the publication of Darwin's great work on the ‘* Variation 
of Animals and Plants under Domestication,” there have been 
but few books of first-rate importance dealing with the subject. 
The most important of these is Mr. William Bateson’s work, 
entitled ‘* Materials for the Study of Variation.” I have no 
hesitation in saying that I regard this work as a most important 
contribution to the literature of the evolution theory. In it at- 
tention is called, with that emphasis which the subject demands, 
to the supreme importance of the actual study of variation to the 
evolutionist, and a systematic attempt is made to classify vari- 
ations as they occur in nature. In preparing this book Mr. 
Bateson has performed a very real service to zoology, not the 
least part of which is that he has made a most effective protest 
against that looseness of speculative reasoning which, since the 
publication of the ‘‘ Origin of Species,” has marred the pages 
of so many zoological writers. 
‘The variations of organisms may be grouped under two heads, 
according to their nature and source: (1) There are those vari- 
ations which appear to have no relation to the external condi- 
tions, for they take place when these remain unchanged, e.g. in 
members of the same litter ; they are inherent in the constitution 
of the individual. These we shall call constitutional variations, 
or as their appearance seems nearly always to be connected with 
reproduction, they may be called geve/zc (congenital, blastogenic) 
zariations. (2) The second kind of variations are those which 
are caused by the direct action of external conditions. These 
variations constitute the so-called acguéred characters. 
My first object is to consider these two kinds of variations, 
their nature, their causes, and their results on subsequent gener- 
ations, and te inquire whether there are any fundamental 
differences between them. In this connection it is of particular 
importance that we should inquire whether acquired modifi- 
cations are transmitted in reproduction. As is well known, there 
are two schools of thought holding directly opposite views as to 
this matter. The one of these schools—the so-called Lamarckian 
school—holds that they may be transmitted as such in reproduc- 
tion; the other school, on the other hand, maintains that 
acquired modifications affect only the individual concerned, and 
are not handed on as such in reproduction. That the decision 
of the matter is not only theoretically important, but also 
practically, is evident, for upon it depends the answer to the 
question whether mental or other facilities acquired by the 
Jaborious exercise of the individual are ever transmitted to the 
offspring—whether the facility which the individual acquires in 
eesisting temptation makes it any easier for the offspring to do 
the same, whether the effects of education are cumulative in 
successive generations. To put the matter as Francis Galton 
has put it, Is nature stronger than nurture, or nurture than 
nature ? 
We have then two kinds of variation to consider: (1) genetic 
wariation, (2) acquired modification. It is the former of these— 
namely, genetic variation—with which I wish primarily to deal. 
Let us examine more fully the mode of its occurrence. 
Genetic Variation. 
Organised beings present, as you are aware, two main kinds 
of reproduction, the sexual and the asexual.. These two kinds 
of reproduction present certain differences, of which the most 
important, and the only one which concerns us now, is- the fact 
that genetic variation is essentially associated with sexual repro- 
NO. 1560, VOL. 60] 
NATURE 
593 
duction, and is rarely, if ever, found in asexual reproduction. 
In other words, whereas the offspring resulting from asexual re- 
production as a rule exactly resemble the pirent, they are always 
different from the parents in sexual reproduction. I am aware 
that I am treading on disputed ground. You will observe that 
I do not make the assertion that asexually produced offspring 
always exactly resemble the parent, and never present genetic 
variations. To say that would be going to» far in the present 
state of our knowledge. Therefore [ have put the matter less 
strongly, and merely assert that whereas asexual reproduction is 
on the whole characterised by identity between the offspring and 
the parent, sexual reproduction is al ways characterised by differ- 
ences more or less marked between the two; and I reserve the 
question as to whether genetic variations are ever found in 
asexual reproduction for later consideration. 
This modified form of the statement will, I think, be admitted 
on all hands, but before going on I will illustrate my meaning 
by reference to actual examples. 
Asexual reproduction is a phenomenon comparatively rare in 
the animal kingdom, and when it does occur it is exceedingly 
difficult to investigate from this particular point of view. In the 
vegetable kindom, on the other hand, it is quite common. All, 
or almost all, plants possess this power, and in a very great 
many of them the result of its exercise can be fully followed out, 
and contrasted with that of sexual reproduction. Let us follow 
it out in the potato-plant. The potato can and does normally 
propagate itself asexually by means of its underground tubers. 
As you will know, if you take one of these and plant it, it gives 
rise to a plant exactly resembling the parent. If the tuber (seed 
as it is sometimes erroneously called) be that of the Magnum 
Bonun, it gives rise to a plant with foliage, flowers and tubers 
of the Magnum Bonum variety ; if it be of the Snowdrop, the 
foliage, flowers, habit and tubers are totally different from the 
Magnum Bonum, and are easily identified as Snowdrops. In 
this way a favourable variety of potato can be reproduced to 
almost any extent with all its peculiarities of earliness or late- 
ness, pastiness or mealiness, power of resisting disease and so 
forth. By asexual reproduction the exact facsimile of the 
parent may always be obtained, provided the conditions remain 
the same. 
Now let us turn to the results of sexual reproduction—the 
seeds, z.e. the real seeds, which as you know are produced in 
the flowers, are the means by which sexual reproduction is 
effected. They are produced in great quantity by most plants, 
and when placed in the ground under the proper conditions they 
germinate and produce plants. But these plants do not resemble 
the parent. Try the seed of the Magnum Bonum potato, and 
raise plants from it. Do you think that any of them will be the 
Magnum Bonum with all its properties of keeping, resisting 
disease, and so forth? Nota bit of it. The probability is, that 
not one of your seedling plants will exactly reproduce the 
parents; they will all be different. Again, take the apple; if 
you sow the seed of the Blenheim Orange and raise young 
apple-trees, you will not get a Blenheim Orange. © All your 
plants will be different, and probably not one will give you 
apples with the peculiar excellence of the parent. If you want 
to propagate your Blenheim Orange and increase the number of 
your trees, you must proceed by grafting or by striking cuttings, 
which are the methods by which such a tree may be asex- 
ually reproduced. And so on. Examples might be multiplied 
indefinitely. Every horticulturist knows that variety characterises 
seedlings, ¢.e. sexual offspring, whereas identity is found-in slips, 
grafts and offsets, ¢.¢. in asexual offspring ; and that if you want 
to get a new plant you must sow seeds, while if you want to 
increase your stock of an old one you must strike cuttings, plant 
tubers or proceed in some analogous manner. 
An apparent exception to this rule is afforded by so-called 
bud variation, but it is not certain that this is really an exception. 
In so far as these bud variations are not of the nature of 
acquired variations produced by a change of external conditions, 
and disappearing as soon as the old conditions are renewed, 
they are probably stages in the growth and development of the 
organism. That is to say, they are of the same nature as those 
peculiarities in animals which appear at a particular time of life, 
such as a single lock of hair of a different colour from the rest 
of the hair,! the change in colour of hair with growth,” the 
appearance of insanity or of epilepsy at a particularage. There 
1 Darwin, ‘ Variation,” vol. i. p. 449. x ! } 
2 Asan example I may refer to the Himalayan rabbit ; Darwin, “‘ Varia- 
tion,” vol. i. p. 114. : 
