German East Africa. Portugal might retain, for her own exploitation on right lines, the large remaining province of Mozambique proper from the Lucio as far south as Quelimane, disposing by sale to the British Empire of inner Zambezia and South-East Africa. The alternative to such proceedings, and the course which should be adopted in any case in the Portuguese as well as the French colonies (if an important German grievance is to be removed), would be the abolition of all differential duties in the Customs tariff. Such differential duties exist nowhere in the German colonies, in the Indian Empire, or the Crown Colonies of Great Britain, and the present writer is certainly not one who favours their continuance even in the tariffs of the Daughter Nations.

We hear a great deal of Portuguese pride forbidding Portugal to do this and that, under a republic as under a monarchy. But this pride is based on that ignorance that still keeps the Portuguese nation under its cruel yoke, a nation of which some 70 per cent. amongst its people are unable to read and write. There is no reason whatever why Portugal should withdraw from her Asiatic possessions, which are quite manageable with her existing means, especially with the British Empire as her friend and ally. But a portion of the Portuguese dominions in Africa should be sold or leased at a fair valuation, and the proceeds be most carefully employed by the Portuguese Government in developing the resources of Portugal itself, a land, with its annectant islands, of 35,500 square miles, extraordinarily blessed by Nature and yet possessing a population which scarcely reaches the total of London and its suburbs. Even if the Portuguese sold Guinea to the French; the Congo province and North Mozambique (Ibo) to the Germans; Zambezia, Beira and Delagoa Bay to the British, they would still remain the recognised and effective rulers of an empire of 500,000 square miles, a much larger area than they actually possessed in 1870; while, in addition, they should have acquired a fund which would suffice to build a network of light railways over Portugal itself and enable that land to become the greatest fruit-producing region of Europe.

H. H. JOHNSTON.

INHERITANCE OF ACQUIRED

CHARACTERS

THEORETICAL DIFFICULTIES

I

THE facility with which animals and plants vary under the direct action of altered surroundings, both in their specific and secondary features, has been proved lately by a vast amount of observations and experiments. The forms of animals, their colour, their skin, their skeletons, all their organs, and their habits are easily modified as soon as the animal's food and the general conditions of its existence and its biological surroundings are altered. The same is true of plants, even to a still greater extent. So many striking facts have been accumulated lately in this direction that the chief interest of such researches is now to study the inner physiological and anatomical modifications which take place in the tissues under the influence of changing surroundings. The reason why the modifications so well answer in most cases the new requirements as to represent adaptations is what now chiefly interests the biologist. The study of variation and evolution is thus tending more and more to become a physiological problem. These were the points I dealt with in my previous articles on 'The Direct Action of Surroundings upon Plants and Animals.''

An important question arises, however, in connexion with all similar researches. Are the modifications of the individuals transmitted, entirely or partially, to their offspring? Even if we see that a modification produced in some individuals reappears in their descendants, are we sure that it is not produced anew in each generation? And, supposing it is inherited, will it continue to appear for some time, even though the offspring be taken back to the old surroundings?

Let us take, for instance, the modifications which Viré obtained in some crustaceans after he had transferred them from open rivers and ponds to a laboratory established in the darkness of the Paris catacombs. After a few months' stay in the darkness the eyes of

2

1 Nineteenth Century and After, July, November and December 1910. * See Nineteenth Century and After, November 1910, p. 861.

these little animals, being used no more, were atrophied, while their organs of touch and smell, chiefly used in the dark surroundings, took a rapid development. An important adaptation to new conditions of life, formerly explained by natural selection acting upon accidental congenital variations, was thus accomplished by the surroundings themselves within the individual's life. However, before we recognise in the direct action of environment a powerful factor of the evolution of new species, it must be known whether the adaptations just mentioned, or at least an increasing disposition for acquiring them, are transmitted to the next generation of crustaceans born in the catacombs? And this is what we have not yet learned from direct experiments.

The same question must be asked concerning plants taken from the plains of Middle Europe to Alpine, maritime, or desert surroundings. They rapidly acquire in their new environment the anatomical structure and the forms characteristic for Alpine, maritime, or desert plants. But are these new forms and structure inherited? And, if so, how long will the newly acquired characters last after the plant has been taken back to its old environment?

Unfortunately, the replies to these questions obtained till now by experiments are not quite clear. And yet the questions are most important. For, if characters acquired by individuals under the influence of a new environment and new habits are inheritable, then the whole problem of evolution is immensely simplified. Variation becomes the beginning of evolution, and the function of natural selection is quite comprehensible. Selection has not to increase, or to accentuate a variation; it has only to weed out those individuals which are not capable of varying rapidly enough in accordance with the new requirements. Those crustaceans whose organs of smell and touch do not develop rapidly. enough in an underground river, those plants the tissues of which do not rapidly increase their powers of assimilation in an Alpine climate perish, while those which vary in the proper direction with a sufficient rapidity survive. Nothing is thus expected from natural selection which it could not accomplish.

Darwin and his contemporaries-Herbert Spencer, Haeckel, Moritz Wagner, Huxley, and a host of practical biologists-did not doubt of the inheritance of acquired characters whenever they substantially affect the inner structure of a plant or an animal.* It was only after Darwin's death, in the years 1883-1887, that doubts were raised upon this point, especially by the Freiburg professor of entomology, A. Weismann.

See Nineteenth Century and After, July 1910.

See, for instance, the quotations in point given by Herbert Spencer from Darwin's works, in Nature, li. 414.

A naturalist accustomed to rely upon the experimental method would be disposed to think that some decisive experiments proving the non-transmission of acquired characters were made about that time. But there was nothing of the sort, and up till now we are waiting in vain for such experiments. True, that Weismann announced in 1888 that he had clipped the tails of some twelve to fifteen white mice for five generations, but had obtained no tailless mice; nor had he noticed any tendency towards a shortening of the tails. Later on he extended his experiment to twentytwo generations, and came to the same negative result.' Cope and Rosenthal experimenting on mice, and Ritzema on mice and rats, came to the same conclusion. We may thus take as granted that a superficial mutilation, such as the clipping of the tails of mice (which were left to breed a few weeks after their birth), is not inherited.

8

But this was known long since. Darwin, who had studied with his usual carefulness the experience of the breeders concerning the cutting of the tails in certain breeds of sheep, and of ears in dogs, came long ago to the conclusion that a part of an organ may be removed during several successive generations, and if the operation be not followed by disease, the lost part reappears in the offspring.' But, of course, he did not consider the non-inheritance of superficial mutilations as an argument against the hereditary transmission of acquired characters. In fact, there are many reasons why the absence of a tail or a digit has little chance of being transmitted one of them being, as suggested by Professor Nussbaum, that the young embryo would promptly regenerate its missing portion.' 10

On the other side, in conformity with the just-mentioned views of Darwin, when Professor Brown-Séquard studied the physio

5 The obligation to give the proof lies with the Lamarckians, who believe in the transmission,'- -we are told sometimes by English anti-Lamarckians. But this attitude, not unfrequent in law courts, is certainly not that of men of science. If Darwin were still among us he surely would have begun long ago a series of well-planned experiments to test a statement so fundamental for the theory of evolution.

The Supposed Transmission of Mutilations,' in Essays upon Heredity and Kindred Biological Problems (Oxford, 1891, 2nd edition), i. 444-445.

'Vorträge über die Descendenztheorie (2nd edition : Jena, 1904), ii. 56.

8 For an examination of all the cases in point, see Prof. J. Arthur Thomson's capital work, Heredity (London, 1898), pp. 224-225. Contradictory experiments were once mentioned in an American report, but nothing more was heard of them. • The Variation of Animals and Plants under Domestication (2nd edition, 1899) ii. 391.

10 Die Vererbung erworbener Eigenschaften, 1903, quoted by Prof. Th. H. Morgan in his capital work, Experimental Zoology (New York: 1907), p. 57. The long-continued inheritance of a part which has been removed during many generations is even not an argument against the Pangenesis hypothesis, for gemmules formerly derived from the part are multiplied and transmitted from generation to generation.' (Darwin's Variation, ii. 392.)

VOL. LXXI-No. 421

K K

logical effects of more serious lesions, he found that these effects were transmitted; and his experiments were confirmed later on by Romanes. With guinea-pigs, an injury made to the spinal cord or to the sciatic nerve provoked epilepsy, or at least a dis position for it; and in a number of cases this disposition was transmitted to the offspring. Otherwise epilepsy never appeared among the guinea-pigs bred in very large numbers by BrownSéquard-not even among the individuals he had operated upon, but in a different way. The possibility of explaining an inherited disposition for epilepsy as a bacterial infection, which took place during the operation, is thus excluded. Several other results, due to other lesions, were also inherited. Altogether, these experiments were conducted so carefully, during a long succession of years, that the opinion which has prevailed among specialists is, that they really prove the hereditary transmission of certain abnormal states of different organs, provoked by certain lesions."

II

If Weismann's experimental contribution to the question of inheritance of acquired characters was worthless, his critical revision of the whole subject was, on the contrary, very valuable.12 Besides, Weismann gave a new interest to the whole matter under discussion by bringing forward an ingenious hypothesis of heredity, in defence of which he wrote quite a number of essays and books, always in a form both attractive and accessible to a large circle of readers.

It is evident that a discussion about the part played by environment in the evolution of organic beings ought never to have been made to depend upon our preferences for this or that hypothesis of heredity. It is the hypothesis of heredity which must be built up so as to explain the facts established by our knowledge of evolution; and this is what Darwin did when he worked out his Pangenesis hypothesis. But if we had a hypothesis of heredity,

11 Prof. T. H. Morgan, who in his earlier work, Evolution and Adaptation (1903), had made restrictions, concerning the value of these experiments, wrote as follows in his later work, Experimental Zoology (New York, 1907), p. 54: 'I have given somewhat fully these remarkable results of Brown-Séquard because the experiments appear to have been carried out with such a care, and the results are given in such detail, that it seems that they must be accepted as establishing the inheritance of acquired characters.'-Weismann's reply (Vorträge über die Descendenztheorie, 2nd edition, ii. 57, 58) consists in repeating that, 'probably' a microbe infection took place. He says moreover that once the lesion itself was not inherited, the experiments are no proof of an inheritance of an 'acquired character.' But why the inheritance of an anatomical modification of portions of operated nerves, resulting in the offspring in the same morbid phenomena as those that were provoked in the parents, should not be considered as the transmission of an acquired character, remains unexplained.

12 An excellent review of Weismann's critical contribution in this field will be found in J. Arthur Thomson's Heredity.