continuity in the organic world. When species are no discontinuous or sharply marked off from one anothe it by no means follows that they were sharply marke off at the time when they arose from the parent specie So far, on the other hand, as they form a continuo series, the interpretation of their origin by mutation altogether excluded. Discontinuity between existin species does not prove an origin by mutation: continuit wherever it occurs, disproves it. De Vries, in the follow ing words, accepts Prof. Bateson's statement of discor tinuity in organic nature: 'species are discontinuous and we must assume that their characters are discon tinuous also' (p. 250). The full meaning of this sentence is better sought in Prof. Bateson's original statement:

'In proportion as the transition from term to term is minima and imperceptible we may speak of the Series as being Continuous, while in proportion as there appear in it lacunæ, filled by no transitional form, we may describe it as Discontinuous.'*

Mr R. H. Lock puts the same statement in another form: "... the species riddle presents itself definitely as the problem of the existence of a series of discontinuous groups of creatures, sharply marked off the one from the other' (p. 11).

An enquiry into the experience and life-work of these authors of the dogmatic statement that species are discontinuous does not inspire confidence in their authority. Many hundreds of naturalists all over the civilised world devote their lives to systematics-to the study of the differences between species. Much of the work is superficial and poor, but much of it is extremely good. Furthermore, the standard of excellence has been rapidly rising in recent years. If some of the best of these students of systematic natural history had expressed the opinion that species are 'sharply marked off the one from the other,' or that their series is discontinuous, as defined by Prof. Bateson, we should have grounds for accepting the conclusion. We find, however, that systematists are very far from recognising the discontinuity which is so obvious to some of those who are not practically concerned with the problems of specific and sub

* Materials for the Study of Variation,' p. 15. London, 1894.

specific discrimination. Among the most exact of recent systematic research is that pursued in the zoological museum at Tring. It is of interest to learn the conclusions to which that work has led. The following statements as to the origin of species and the transition between species and sub-species are made by Rothschild

and Jordan:

.......... geographical variation is the beginning of the ramification of one species into more. From it has resulted the enormous variety of existing species, each breeding true. Geographical varieties differ in various degrees. They repreent various steps in the evolution of daughter-species. Whoever studies the distinctions of geographical varieties closely and extensively will smile at the conception of the origin of species per saltum. For he will find that in the large majority of cases the geographical distinctions are minute, and he will see further that there is a complete gradation from geographically separate varieties of a species which are very distinct from one another in colour, pattern and structure in all individuals, to geographically separate portions of a species which do not exhibit any distinctions. It is just this prevalence of minuteness in the geographical distinguishing characters which gives us the best insight into the working of evolution. Hence it is of the highest importance to demonin geographically separate portions of a species-a demonstrastrate wherever one can the prevalence of minute distinctions tion which depends entirely on the intensity and minuteness of the research of the specialist in the systematics of the re

spective group of animals.' *

The words I have italicised show that the distinctions between geographical varieties-regarded by the authors as incipient species-are by no means obvious. They are only to be perceived at all by keen and most patient in such a spirit are what we should expect-the recogenquirers. The results of systematic enquiries pursued nition of specific difference where the earlier naturalists thought they perceived specific identity.

No Evidence that De Vries' Mutations are New Species. Much baseless exultation has been indulged in by those who, in these days no less than in St Paul's, spend their

* 'Novitates Zoologica' (December 1903), x, 492.

time in nothing else but either to tell or to hear soni new thing. And in modern times, if not in ancient, tal novelty loses nothing of importance in the telling. Th we constantly read popular accounts of the new specie which de Vries has experimentally produced; and at tl; beginning of his volume mentioned in our list the thr following sentences have been printed, presumably 1 the editor, Dr D. T. MacDougal :

"The origin of species is a natural phenomenon' (Lamarcl "The origin of species is an object of inquiry' (Darwin). "The origin of species is an object of experimental invest gation' (de Vries).

In spite of the exalted position assigned to de Vrie in this crescendo of statements, there is not the slightes evidence that the Dutch professor has produced an new species, or indeed that his experimental breedin of strains from the large variations which suddenl appeared in Enothera lamarckiana has done anything but repeat, under more carefully guarded conditions and with more precise records, the process by which number: of strains of cultivated plants have been raised. If de Vries' evening-primroses are new species, so are dozens of the artificially selected breeds of animals and plants -in fact, those very forms which, still freely interbreeding with each other after centuries of human selection, prevented Huxley from feeling entire confidence in the power of natural selection to create fresh species.

In thus maintaining that it is an exaggeration to say that de Vries has produced new species, we are very far from undervaluing the interest of these investigations. It is of the highest importance that the achievements of the horticulturist should be repeated under the exacting conditions imposed by modern scientific research.

De Vries' definition of a biological experiment is given in the following words (p. 430):

Experiments are a repetition of things occurring in nature with the conditions so guarded and so closely followed that it is possible to make a clear analysis of facts and their causes, it being rightfully assumed that the laws are the same in both cases.'

Now, when a new species arises, the thing occurring in

nature' is the survival, increase, and final maintenance of
a certain average number of individuals forming an inter-
breeding community, and bearing characters differing
in kind or degree from other assemblages of individuals.
Whether these differences are usually large or small, and
the origin of species discontinuous or continuous, has
been already considered on pp. 8, 13-15.

f

1

It is of the essence of a species, whether new or old,
that it should be capable of withstanding the competition
of nature; and, until de Vries has shown that his strains
of Enthera possess this power, it is premature to speak
of them as species. Some of them were found by de

Tries in a field at Hilversum, in Holland, growing in
small numbers side by side with the parent, Enothera
lamarckiana, itself a garden escape.
It would be a
legitimate experiment in studying the origin of species
to increase the numbers of some new form by artificial
restriction of competition, and then turn it loose, thus
exposing to natural conditions immense numbers of indi-
viduals, and giving nature as it were increased oppor-
tunities for the rapid creation of a new species. But in
any such experiments it would be more satisfactory to
employ forms which have sprung from species of which
the history is thoroughly known, rather than off-shoots
from the problematical Enothera lamarckiana, as yet
untraced with any certainty in the wild state.

To raise fresh strains under artificial conditions affords
far less evidence as to the mode of origin of new species
than to study the actual succession of species in time as
revealed in the records of palæontology, or the changes
of varied degrees which become manifest when we
investigate their geographical distribution in space. In
both these departments of knowledge we have before us
the modifications which have actually occurred under the
conditions of living nature. In de Vries' interesting and
important experiments the most essential element in the
problem of the origin of species is, ex hypothesi, excluded.
His method can never demonstrate what forms would
have been selected by nature; palæontology shows us
what she has selected, geographical distribution what she

is selecting.

Vol. 211.-No. 420.

Paleontology and Evolution by Mutation.

The bearing of paleontology upon the hypothesis of evolution by mutation is of even higher importance than might be gathered from the last paragraph; for the fundamental question whether the course of evolution has been continuous or discontinuous must ultimately receive a decisive answer from this science. Although it is too much to expect that the detailed fossil record will ever be complete over a long piece of geological history, a great deal is known of the gradual modification of species in the successive strata of a single formation or an important section of a formation. During a recent visit to America, my friend Prof. H. F. Osborn showed me a series of mammalian skulls from the Lower Tertiary beds of western North America—a selection from the vast stores of splendid material in the American Museum of Natural History, New York. Arranged in the order of succession in time, as determined by the strata from which the fossils had been taken, one series showed the most gradual and continuous modification of the characters of teeth, another a similarly continuous evolution of horns. Evidence of this kind does not, of course, enable us to decide upon the causes of evolution; but it is conclusive proof that mutation, in de Vries' sense, has not occurred in the cases referred to.

Concordant testimony is given by Prof. W. B. Scott, of Princeton University, in his article in the Cambridge Darwin memorial volume. Speaking of the evolution of the horse in the Lower Tertiary deposits of North America he states (p. 190):

'Each one of the different Eocene and Oligocene horizons has its characteristic genus of horses, showing a slow, steady progress in a definite direction, all parts of the structure participating in the advance. It is not necessary to follow each of these successive steps of change, but it should be emphasised that the changes are gradual and uninterrupted.' Prof. Osborn and Prof. Scott consider that such gradual and continuous change is inconsistent with an interpretation based on the theory of natural selection; but, assuming natural selection to have been the motive cause, what other evidence could we expect to find?