I shall not here give a repetition and summary of the results arrived
at with respect to seasonal dimorphism, but rather the general
conclusions derived from these results; and, at the same time, I may
take the opportunity of raising certain questions which have not
hitherto found expression, or have been but briefly and casually stated.
It must, in the first place, be admitted that differences of specific
value can originate through the direct action of external conditions
of life only. Of the truth of this proposition there can be no doubt,
after what has been above stated concerning the difference between
the two forms of any seasonally dimorphic species. The best proof is
furnished by the older systematists, to whom the genetic relationship
of the two forms was unknown, and who, with unprejudiced taxonomy, in
many cases indicated their distinctness by separate specific names.
This was the case with _Araschnia Levana_ and _Prorsa_, _Euchloe Belia_
and _Ausonia_, _E. Belemia_ and _Glauce_, _Plebeius Polysperchon_ and
_Amyntas_. In the presence of these facts it can scarcely be doubted
that new species can be formed in the manner indicated; and I believe
that this was and is still the case, with butterflies at least, to
a considerable extent; the more so with these insects, because the
striking colours and markings of the wings and body, being in most
cases without biological significance, are useless for the preservation
of the individual or the species, and cannot, therefore, be objects of
natural selection.
Darwin must have obtained a clear insight into this, when he attempted
to attribute the markings of butterflies to sexual and not to natural
selection. According to this view, every new colour or marking first
appears in one sex accidentally,[51] and is there fixed by being
preferred by the other sex to the older coloration. When the new
ornamentation becomes constant (in the male for example), Darwin
supposes that it becomes transferred to the female by inheritance,
either partially or completely, or not at all; so that the species,
therefore, remains more or less sexually dimorphic, or (by complete
transference) becomes again sexually monomorphic.
The admissibility of such different, and, to a certain extent,
arbitrarily limited inheritance, has already been acknowledged.
The question here concerned is, whether Darwin is correct when he
in this manner attributes the entire coloration of butterflies to
sexual selection. The origin of seasonal dimorphism appears to me to
be against this view, howsoever seductive and grand the latter may
seem. If differences as important as those which exist between the
summer and winter forms of many butterflies can be called forth by the
direct action of a changed climate, it would be extremely hazardous to
attribute great importance to sexual selection in this particular case.
The principle of sexual selection appears to me to be incontestible,
and I will not deny that it is also effective in the case of
butterflies; but I believe that as a final explanation of colour this
agency can be dispensed with, inasmuch as we see that considerable
changes of colour can occur without the influence of sexual
selection.[52]
The question now arises, how far does the transforming influence of
climate extend? When a species has become transformed by climatic
change to such an extent that its new form possesses the systematic
value of a new species, does it return to its older form by removal
to the old climatic conditions? or would it under these circumstances
become again transformed in a new manner? This question is not without
importance, inasmuch as in the first case climatic influences would be
of little value in the formation of species, and there would result
at most only a fluctuation between two extremes. In the same manner
as in seasonally dimorphic species the summer and winter forms now
alternate with each other every year, so would the forms produced by
warmth and cold then alternate in the greater periods of the earth’s
history. Other groups of animals are certainly changed by the action of
different climatic influences; but in butterflies, as I believe I have
proved, temperature plays the chief part, and as this only oscillates
between rather narrow limits, it admits of no great differences of
coloration.
The question thus suggests itself, whether species of butterflies
only oscillate between two forms, or whether climatic change, when
sufficiently great to produce variation, does not again originate a new
form. Inasmuch as the reversion experiments with seasonally dimorphic
butterflies appear to correspond with the latter view, I believe that
this must be admitted. I am of opinion that an old form never again
arises through change of climate, but always a new one; so that a
periodically recurring change of climate is alone sufficient, in the
course of a long period of time, to admit of new species arising from
one another. This, at least, may be the case with butterflies.
My views rest essentially upon theoretical considerations. It
has already been insisted upon, as results immediately from the
experiments, that temperature does not act on the physical constitution
of the individual in the same manner as acid or alkali upon litmus
paper, i.e., that one and the same individual does not produce this or
that coloration and marking according as it is exposed to warmth or
cold; but rather that climate, when it influences in a similar manner
many succeeding generations, gradually produces such a change in
the physical constitution of the species that this manifests itself
by other colours and markings. Now when this newly acquired physical
constitution, established, as we may admit, throughout a long series
of generations, is again submitted to a constant change of climate,
this influence, even if precisely similar to that which obtained during
the period of the first form of the species, cannot possibly reproduce
this first form. The nature of the external conditions may be the same,
but not so the physical constitution of the species. Just in the same
manner as a _Pieris_ (as has been already shown), a _Lycæna_, or a
_Satyrus_, produces quite different varieties under the transforming
influence of the same climate, so must the variation originating
from the transformed species of our present case after the beginning
of the primary climate be different from that primary form of the
species, although perhaps in a less degree. In other words, if only
two different climates alternated with each other during the earth’s
geological periods, every species of butterfly submitted to these
changes of climate would give rise to an endless series of different
specific forms. The difference of climate would in reality be greater
than supposed, and for any given species the climatic variation would
not only occur through the periodic shifting of the ecliptic, but also
through geological changes and the migrations of the species itself, so
that a continuous change of species must have gone on from this sole
cause of alternation of climate. When we consider that many species
elsewhere extinct have become locally preserved, and when, further, to
these we add those local forms which have arisen by the prevention of
crossing (amixia), and finally take into consideration the important
effects of sexual selection, we can no longer be astonished at the vast
numbers of species of butterflies which we now meet with on the earth.
Should any one be inclined to conclude, from my reversion experiments
with seasonally dimorphic butterflies, that the secondary species
when exposed to the same climate as that which produced it must
revert to the primary, he forgets that this reversion to the winter
form is nothing but a reversion--i.e., a sudden return to a primary
form through peculiar laws of inheritance--and by no means a gradual
re-acquisition of this primary form under the gradual influence of
the primary climate. Reversion to the winter form occurs also through
other influences, as, for instance, by high temperature. Reversions
of this kind, depending on laws of heredity, certainly happen with
those cases of transmutation which do not alternate with the primary
form, as in seasonal dimorphism, but which occur continuously. They
would, however probably be more quickly suppressed in such cases than
in seasonal dimorphism, where the constant alternation of the primary
and secondary forms must always maintain the tendency of the latter to
produce the former.
That the above conclusion is correct--that a secondary species, when
exposed to the external conditions under the influence of which the
primary form originated, does not again revert to the latter--is proved
by experience with plants. Botanists[53] assure us “that cultivated
races which become wild, and are thus brought back to their former
conditions of life, do not become changed into the original wild form,
but into some new one.”
A second point which appears to me to be elucidated by seasonal
dimorphism, is the origin of variability. It has already been
prominently shown that secondary forms are for the most part
considerably more variable than primary forms. From this it follows
that similar external influences either induce different changes in
the different individuals of a species, or else change all individuals
in the same manner, variability arising only from the unequal time
in which the individuals are exposed to the external influence. The
latter is undoubtedly the case, as appears from the differences which
are shown by the various individuals of a secondary form. These are
always only differences of degree and not of kind, as is perhaps most
distinctly shown by the very variable _A. Prorsa_ (summer form), in
which all the occurring variations differ only by the _Levana_ marking
being more or less absent, and, at the same time, by approximating
more or less to the pure _Prorsa_ marking; but changes in a totally
different direction never occur. It is likewise further evident, as has
been mentioned above, that allied species and genera, and even entire
families (_Pieridæ_), are changed by similar external inducing causes
in the same manner--or, better, in the same direction.
In accordance with these facts the law may be stated, that, in
butterflies at least, all the individuals of a species respond to the
same external influences by similar changes, and that, consequently,
the changes brought about by climatic influences take a fixed
direction, determined by the physical constitution of the species.
When, however, new climatic forms of butterflies, in which natural
selection is completely excluded, and the nature of the species itself
definitely determines the direction of the changes, nevertheless show
variability from the very beginning, we may venture to conclude that
every transformation of a species generally begins with a fluctuation
of its characters. But when we find the primary forms of butterflies
always far more constant, this shows that the continued crossing of the
individuals of a species to a certain extent balances the fluctuations
of form. Both facts taken together confirm the law formerly enunciated
by me,[54] that in every species a period of variability alternates
with one of (relative) constancy--the latter indicating the
culmination, and the former the beginning or end, of its development.
I here call to mind this law, because the facts which I advanced at
that time, viz., Hilgendorf’s history of the phyletic development of
the Steinheim fossil shells, having since become somewhat doubtful, one
might easily be inclined to go too far in mistrusting them and refuse
to give them any weight at all.[55]
In the essay just indicated I traced the origin of a certain class
of local forms to local isolation. I attempted to show that when a
species finds itself in an isolated district in a condition (period)
of variability, it must there necessarily acquire somewhat deviating
characters by being prevented from crossing with the individuals
of other regions, or, what comes to the same thing, a local form
must originate. This production of local forms results because
the different variations which, for the time being, constitute the
variability of the species, would always be in a different numerical
proportion in the isolated district as compared with other regions;
and further, because constancy is produced by the crossing of these
(isolated) varieties among themselves; so that the resultant of
the various components is (local) variation. If the components are
dissimilar the resultant would also be different, and thus, from a
theoretical point of view, there seems to me no obstacle in the way
of the production of such local forms by the process of ‘amixia.’ I
believe that I have further shown that numerous local forms can be
conceived to have arisen through this process of preventive crossing,
whilst they cannot be explained by the action of climatic influences.
That I do not deny the existence of true climatic forms in admitting
this principle of ‘amixia,’ as has been frequently imagined, appears
sufficiently from the treatise in question. The question arises,
however, whether climatic influences may not also originate forms
by ‘amixia’ by making a species variable. It would be difficult at
present to decide finally upon this subject. If, however, in all cases
a variation in a certain fixed direction occurred through climatic
influences, a form could not arise by ‘amixia’ from such a variability,
since the components could then produce resultants different only
in degree and not in kind. But we are not yet able to extend our
researches to such fine distinctions.
As a final, and not unimportant result of these investigations, I may
once more insist that dissimilar influences, when they alternatingly
affect a long series of originally similar generations in regularly
recurring change, only modify the generations concerned, and not
intermediate ones. Or, more briefly, cyclically acting causes of
change produce cyclically recurring changes: under their influence
series of monomorphic generations become formed into a cycle of di- or
polymorphic generations.
There is no occasion to return here to the immediate evidence and
proof of the foregoing law. In the latter, however, is comprised the
question--is not the cycle of generations produced by cyclical heredity
ultimately equivalent to Darwin and Haeckel’s homochronic heredity
which forms the ontogenetic stages into a cycle? It is possible
that from this point, in the future, the nature of the processes of
heredity, which are still so obscure, may be penetrated into, and both
phenomena traced to the same cause, as can now be only surmised but not
clearly perceived.
Finally, the most general, and in so far chief result of these
investigations, appears to me to lie in the conclusion, which may
be thus formulated:--A species is only caused to change through the
influence of changing external conditions of life, this change being
in a fixed direction which entirely depends on the physical nature of
the varying organism, and is different in different species, or even in
the two sexes of the same species.
I am so little disposed to speak in favour of an unknown transforming
power that I may here again insist that the transformation of a
species only partly depends upon external influences, and partly on
the specific constitution of the particular form. I designate this
constitution ‘specific,’ inasmuch as it responds to the same inciting
cause in a manner different to the constitution of another species.
We can generally form a clear conception why this should be the case;
for not only is there in another species a different kind of latent
vital activity, but each species has also a different developmental
history. It must be admitted that, from the earliest period of the
formation of an organism, and throughout all its intermediate stages,
properties which have become established, such as growth, nutrition,
or tendency to development, have been transferred to the species now
existing, each of which bears these tendencies in itself to a certain
extent. It is these innate tendencies which determine the external
and internal appearance of the species at every period of its life,
and which, by their reaction to external factors, represent the life
of the individual as well as that of the species. Since the sum of
these inherited tendencies must vary more or less in every species,
not only is the different external appearance of species as well as
their physiological and biological diversity thus explained, but it
necessarily follows therefrom, that different species must respond
differently to those external causes which tend to produce a change in
their form.
Now, this last conclusion is equivalent to the statement that every
species, through its physical constitution, (in the sense defined) is
impressed with certain fixed powers of variation, which are evidently
extraordinarily numerous in the case of each species, but are not
unlimited; they permit of a wide range for the action of natural
selection, but they also limit its functions, since they certainly
restrain the course of development, however wide the latter may be. I
have elsewhere previously insisted[56] that too little is ascribed to
the part played by the physical constitution of species in the history
of their transformation, when the course of this transformation is
attributed entirely to external conditions. Darwin certainly admits
the importance of this factor, but only so far as it concerns the
individual variation, the nature of which appears to him to depend on
the physical constitution of the species. I believe, however, that in
this directive influence lies the precise reason why, under the most
favourable external circumstances, a bird can never become transformed
into a mammal--or, to express myself generally, why, from a given
starting-point, the development of a particular species cannot now
attain, even under the most favourable external conditions, any desired
goal; and why, from this starting-point, given courses of development,
even when of considerable latitude, must be restricted, just as a ball
rolling down a hill is diverted by a fixed obstacle in a direction
determined by the position of the latter, and depending on the
direction of motion and the velocity at the moment of being diverted.
In this sense I agree with Askenasy’s “fixed” direction of variation;
but not if another new physical force directing variation itself
is thereby intended.[57] The explanation of the phenomena does not
appear to me to require such an admission, and, if unnecessary, it
is certainly not legitimate. According to my view, transmutation by
purely internal causes is not to be entertained. If we could absolutely
suspend the changes of the external conditions of life, existing
species would remain stationary. The action of external inciting
causes, in the widest sense of the word, is alone able to produce
modifications; and even the never-failing “individual variations,”
together with the inherited dissimilarity of constitution, appear
to me to depend upon unlike external influences, the inherited
constitution itself being dissimilar because the individuals have been
at all times exposed to somewhat varying external influences.
A change arising from purely internal causes seems to me above all
quite untenable, because I cannot imagine how the same material
substratum of physical constitution of a species can be transferred to
the succeeding generation as two opposing tendencies. Yet this must
be the case if the direction of development transferred by heredity
is to be regarded as the ultimate ground both of the similarity
and dissimilarity to the ancestors. All changes, from the least to
the greatest, appear to me to depend ultimately only on external
influences; they are the response of the organism to external inciting
causes. It is evident that this response must be different when a
physical constitution of a different nature is affected by the same
inciting cause, and upon this, according to my view, depends the great
importance of these constitutional differences.
If, under “heredity,” we comprise the totality of inheritance--that
is to say, the physical constitution of a species at any time, and
therefore the restricted and, in the foregoing sense, pre-determined
power of variation, whilst under “adaptation” we comprehend the direct
and indirect response of this physical constitution to the changes in
the conditions of life, I can agree with Haeckel’s mode of expression,
and with him trace the transformation of species to the two factors of
heredity and adaptation.