The views and writings of Darwin have influenced in an unexpected way the nature of the work carried on by biological investigators during the past fifty years. To a great extent, whilst generally holding the doctrines he held, they have forsaken his methods of inquiry.

If animals and plants have arrived at their present state by descent with modification from simpler forms, it ought to be possible by careful searching to trace the line of ancestry; and it is this fascinating but frequently futile pursuit which has dominated the minds of many of our ablest zoologists for the last thirty years. To such an extent has this pedigree-hunting been carried that there is scarcely a group of invertebrates from which the vertebrates have not been theoretically derived; and one of the ablest of our physiologists has used his great powers in the attempt to trace the origin of the backboned animals from a spider-like creature, and has exercised his ingenuity in a plausible but unconvincing effort to equate the organs of a king-crab with those of a lamprey. This appeal to comparative anatomy and the consequent neglect of living animals and their habits are no doubt partly due to the influence of Huxley, Darwin’s most brilliant follower and exponent. He had the engineer’s way of looking at the world, and his influence was paramount in many schools. The trend which biology has taken since Darwin’s time is also partly due to a fervent belief in the recapitulation theory, according to which an animal in developing from the egg passes through phases which resemble certain stages in the past history of the ancestors of the animals. For example, there is no doubt that both birds and mammals are descended from some fish-like animal that lived in the water and breathed by gills borne on slits in the gullet, and every bird and mammal passes through a stage in which these gill-slits are present, though their function is lost, and they soon close up and disappear. In the hope, which has been but partially realized, that a knowledge of the stages through which an animal passes on its path from the ovum to the adult would throw light on the origin of the race, the attention of zoologists has been largely concentrated on details of embryology, and a mass of facts has already been accumulated which threatens to overwhelm the worker.

The two chief factors which play a part in the origin of species are heredity and variation, and until we know more about the laws which govern these factors, we cannot hope to arrive at any satisfactory criteria by which we can estimate the importance of the data accumulated for us by comparative anatomists and embryologists. Signs are not wanting that this view is beginning to be appreciated. The publication of ‘Materials for the Study of Variation’ by Mr. Bateson some years ago shows that there exists a small but active school of workers in this field; and recent congresses on hybridization give evidence that in America, on the Continent, and in Great Britain, one of the most important sides of heredity is being minutely and extensively explored. Professor Cossar Ewart’s experiments, which we shall attempt to summarize, deal with heredity and cognate matters, and although they are so far from complete that the results hitherto obtained cannot be regarded as final, they mark an important stage in the history of the subject.

Twelve years ago Professor Ewart began to collect materials for the study of the embryology of the horse, about which, owing to the costliness of the necessary investigations, very little is at present known. At the same time he determined to inquire into certain theories of heredity which have for centuries influenced the breeders of horses and cattle, and the belief in which has played a large part in the production of our more highly bred domestic animals. Foremost amongst these is the view widely held amongst breeders that a sire influences all the later progeny of a dam which has once produced a foal to him. This belief in the ‘infection of the germ,’ or ‘throwing-back’ to a previous sire, is probably an old one, possibly as old as the similar faith in maternal impressions which led Jacob to placed peeled wands before the cattle and sheep of his father-in-law Laban. The phenomenon has recently been endowed with a new name—Telegony. Since the publication of Lord Morton’s letter to Dr. W. H. Wollaston, President of the Royal Society, in 1820, it has attracted the attention, not only of practical breeders, but of theoretical men of science. The supporters of telegony, when pressed by opponents, having almost always fallen back on Lord Morton’s mare, it will be well to recall the chief incidents in the history of this classic animal.

It appears that early in last century Lord Morton was desirous of domesticating the quagga. He succeeded in obtaining a male, but, failing to procure a female, he put him to a young chestnut mare of seven-eighths Arab blood which had never been bred from before. The result was the production of a female hybrid apparently intermediate in character between the sire and the dam. A short time afterwards Lord Morton sold his mare to Sir Gore Ouseley, who bred from her by a fine black Arabian horse. The offspring of this union, examined by Lord Morton, were a two-year-old filly and a year-old colt. He describes them as having

The description of the stripes visible on their coats is careful and circumstantial, but the evidence of the nature of the mane is less convincing:

This is the classical—we might almost say the test—case of telegony: the offspring resembled not so much the sire as an earlier mate of the dam. The facts related tended to confirm the popular view, and that view is now widely spread. Arab breeders act on the belief, and it is so strongly implanted in the minds of certain English breeders that they make a point of mating their mares first with stallions having a good pedigree, so that their subsequent progeny may benefit by his influence, even though poorly-bred sires are subsequently resorted to.

The evidence of Lord Morton’s mare convinced Darwin of the existence of telegony. After a careful review of the case, he says: ‘There can be no doubt that the quagga affected the character of the offspring subsequently got by the black Arabian horse.’ Darwin, however, latterly came to the conclusion that telegony only occurred rarely, and some years before his death expressed the opinion that it was ‘a very occasional phenomenon.’ Agassiz believed in telegony. He was strongly of opinion

Romanes also believed that telegony occasionally occurred. He paid a good deal of attention to the matter, commenced experiments in the hope of settling the question, and corresponded at length on this subject with professional and amateur breeders and fanciers. The result of his investigations led him to the conclusion ‘that the phenomenon is of much less frequent occurrence than is generally supposed. Indeed, it is so rare that I doubt whether it takes place in more than 1 or 2 per cent. of cases.’ He adds that his professional correspondents regard this as an absurdly low estimate. Tegetmeier and Sutherland believe that telegony exists in dogs and other animals; and Captain Hayes, whose opinion probably coincides with that of the majority of veterinary surgeons, takes for granted that it occurs in horses. A controversy some years ago in the Contemporary Review shows us that Mr. Herbert Spencer was a firm upholder of telegony, and that he had a theory of his own as to the mode in which it is brought about.

The explanations put forward by the supporters of telegony as to the mechanism by which it is effected differ widely. It will be well to discuss them here. The view that telegony is due to the mental impression of the dam, held by Sir Everard Home and many others since his day, has nothing to support it; but the other two views, which may be termed (1) the infection hypothesis, and (2) the saturation hypothesis, demand more detailed treatment.

The infection hypothesis supposes that the reproductive organs of the mother are specifically altered or infected by bearing offspring to a previous sire. The method by which this is effected is now most commonly thought to be by a fusion or blending of some of the unused germ-cells of the first sire with the unripe ova in the ovary of the dam. Physiologists, however, regard this as very unlikely. Although at the time that the ovum of a mare is fertilized there are usually other ova almost mature, or approaching maturity, these disappear during gestation. Subsequent offspring arise from successive crops of ova, into whose composition it is most improbable that the earlier spermatozoa could enter. Further, it is known that in the EquidÆ the male germinal cells do not live long within the body of the female; they are already disintegrating eight days after insemination, and they probably lose their fertilizing power after three or four days, if not sooner; hence it is not possible for them to remain in the body during the whole of a period of gestation and to fertilize the next succeeding batch of ova.

The second theory which attempts to account for the phenomenon of telegony is termed the saturation hypothesis. In the words of Mr. Bruce Lowe, who has formulated the theory, we may say that, ‘briefly put, it means that with each mating and bearing the dam absorbs some of the nature or actual circulation of the yet unborn foal, until she eventually becomes saturated with the sire’s nature or blood, as the case may be.’ Although not very well expressed, it is obvious what the author means; and if this saturation really takes place, it accounts for a good deal more than telegony. It would affect the whole body and nature of the dam, and not only the reproductive organs, which, according to Romanes and others, are alone influenced. There is no doubt that matter can and does pass from the blood of the embryo into that of the mother—in certain classes of mammalia, at any rate. The published Report of the Fourth International Congress of Zoology, which met in 1898 at Cambridge, contains a paper by Professor Hubrecht, of Utrecht, in which he describes certain blood-corpuscles formed in the embryo which undoubtedly make their way into the maternal bloodvessels and take part in her circulation. That matter can pass from the bloodvessels of the embryo to those of the mother is further demonstrated by the experiments of M. Charrin, who showed that diphtheritic toxins injected into the embryos of a rabbit caused the death of the mother within five days, and further that a rabbit can be rendered immune by injecting anti-diphtheritic toxins into the embryos.

There is nothing in these experiments to show that the nature of the dam is radically altered; and in the EquidÆ, in which, as we have seen, the classical case of telegony occurred, there is a strong presumption against any such transference of blood-corpuscles from the embryo to the mother. Still, taking all the facts into consideration, it appears that, if telegony exists, it is more likely to be brought about by saturation than by the direct infection of the ovary; though, if the former method be accepted, telegony must be confined to the mammals and the comparatively few other animals whose young spend some time in the body of the mother and are not hatched out from eggs which have lost their connexion with the body of the mother at an early stage.

Before passing on to consider the views of those who hold that telegony does not exist and to see what light the Penycuik experiments throw on the subject, a word or two may be said about Mr. Herbert Spencer’s theory of the mode in which telegony, in which he firmly believed, is brought about. He suggested that some ‘germ-plasm’ passes from the embryo into the mother and becomes a permanent part of her body, and that this is diffused throughout her whole structure until it affects, amongst other organs, the reproductive glands. This view, which in some respects recalls the pangenesis of Darwin, is intermediate between the saturation and the infection hypotheses. Professor Ewart refers to it as ‘indirect infection.’

Weismann, to whom we owe the term telegony, came to consider the facts for and against its existence in connexion with his well-known inquiry into the inheritance of acquired characters. If telegony be true, there is no need to look further for a clear case of the inheritance of a character which has been acquired during the lifetime of the parent. The quagga-ness—if one may be permitted to use such an expression—of Lord Morton’s mare was acquired when she was put to the quagga or shortly afterwards, and was transmitted to her foals. A clearer case of a character acquired during lifetime and transmitted to offspring could not be imagined. Weismann does not absolutely deny the possibility of the existence of telegony, but he would like more evidence. In the Contemporary Review he writes: ‘I must say that to this day, and in spite of the additional cases brought forward by Spencer and Romanes, I do not consider that telegony has been proved.’ And further: ‘I should accept a case like that of Lord Morton’s mare as satisfactory evidence if it were quite certainly beyond doubt. But that is by no means the case, as Settegast has abundantly proved.’ He would, in fact, refer the case to reversion, and quotes Settegast to the effect that every horse-breeder is well aware that the cases are not rare when colts are born with stripes which recall the markings of a quagga or zebra. We shall return to this point later.

A considerable number of German breeders support the contention of Weismann that telegony is as yet unproven, and it may be pointed out that in Germany, on the whole, breeders have had a more scientific education than in England, and that in that country science is regarded with less aversion or contempt than is usually the case among so-called practical men in England. Settegast has been quoted above: neither he nor Nathusius, a leading authority on domestic cattle, has ever met with a case of telegony, and the same is true of Professor KÜhn, the late Director of the Prussian Agricultural Station at Halle. We may mention one more case of an experienced breeder who was equally sceptical—the late Sir Everett Millais, who was, as is well known, an authority of great weight in the matter of dog-breeding. He writes as follows, in a lecture entitled ‘Two Problems of Reproduction’:

It is thus evident that there was a considerable body of opinion, both practical and theoretical, for and against telegony; and that a re-investigation of the subject was urgently needed. Such a re-investigation has been begun by Professor Ewart at Penycuik. Since the clearest and most definite evidence of this throwing back to a previous sire is derived from the crossing of different species of the EquidÆ, it was desirable to repeat the experiment of Lord Morton. This is now unfortunately impossible, because the quagga is extinct. The zebra is, however, still with us, and the mating of a zebra stallion with every variety of horse, pony, and ass, and subsequently putting the dam to pure-bred sires, has been the more important part of the numerous experiments carried on in the Midlothian village some ten miles southwest of Edinburgh.

Before considering in detail the result of the experiments it will be necessary to say a few words on the question of the various species of zebra; and since, like Weismann, Professor Ewart explains certain of the phenomena attributed to telegony by reversion, it will be as well to inquire how far reversion is known amongst the EquidÆ, and what evidence we have that the ancestor of the horse was striped.

Matopo, the zebra stallion from which Professor Ewart had, some eight years ago, bred eleven zebra-hybrids from mares of various breeds and sizes, belongs to the widely distributed group of Burchell’s zebras. Many sub-species or varieties are included in this group, which, as regards the pattern of the stripes, passes—in certain varieties found in Nyassaland—into the second species, the mountain zebra, once common in South Africa. The third species is the GrÉvy’s zebra of Shoa and Somaliland; it is probably this species which attracted so much attention in the Roman amphitheatres during the third century of our era. A pair of Somali zebras were presented to the late Queen some years ago by the Emperor Menelik, and for a time were lodged in the Zoological Gardens, Regent’s Park. This species measures about fifteen hands high, is profusely striped, and stands well apart from the other two groups. It is important to note that in Professor Ewart’s opinion it is the most primitive of all the existing striped horses.

There is no direct evidence that the ancestors of horses were striped. Certain observers think that some of the scratches on the life-like etchings on bone, left us by our palÆolithic cave-dwelling ancestors, indicate such stripes; but little reliance can be placed on this. On the other hand there is much indirect evidence. Every one who has an eye for a horse, and who has travelled in Norway, is sure to have noticed the stripings, often quite conspicuous, on the dun-coloured Norwegian ponies. Colonel Poole assured Darwin that the Kathiawar horses had frequently ‘stripes on the cheeks and sides of the nose.’ Breeders are well aware that foals are often born with stripes, usually on the shoulders or legs, less frequently on the face. Such stripes as a rule disappear as the colt grows up, but can often be detected in later life for a short time after the coat has been shed; they are sometimes only visible in certain lights, and then produce somewhat the same impression as a watered silk. From the facts that more or less striped horses are found all over the Old World; that in Mexico and other parts of America the descendants of horses which were introduced by the Spaniards and which afterwards ran wild are frequently dun-coloured and show stripes; that foals are frequently striped; and that mules not uncommonly have leg and shoulder stripes, the inference is largely justified that the ancestors of all our horses were striped.

The hypothesis of reversion has recently been called in question, and no doubt the term has been much abused. Animals and plants have been said to revert to some remote ancestor when they have varied in some particular, and this variation has then been described as a primitive character possessed by the ancestor; thus there has been much arguing in and about a vicious circle. But the fact that a term has been illogically applied does not destroy the existence of that which the term signifies, and there can be no doubt that reversion exists. That it exists in the EquidÆ is shown by the following proofs: (1) The ancestors of the horse had four premolar teeth in the upper jaw; the modern horse has lost, or is losing, the first of these, and as a rule has only three. When the first is present—the so-called wolf-tooth—it is small, and soon disappears. Zebras usually retain the ancestral number. A few years ago Professor Ewart had a Shetland pony in which the first premolar was relatively nearly as large as it is in hipparion, one of the supposed ancestors of the horse. (2) There is no doubt that the horse is descended through three-toed ancestors from five-toed ancestors. All trace of the latter condition is now lost in development, but an embryo horse six weeks old has three toes as completely formed as those of a rhinoceros. The outer toes then begin to dwindle, and the newly-born foal supports itself on its central digit alone; but horses are occasionally born with two digits, each encased in a hoof, and at very rare intervals with three. CÆsar’s favourite horse was polydactylous, and so was Alexander’s Bucephalus. Major Waddell, in his book on the Himalayas, refers to a creamy fawn-coloured pony, which ‘had a black stripe down the spine ... broad black stripes over the shoulders, flanks, and legs, and dappled spots over the haunches.’ Many other instances might be quoted, but enough has been said to show that reversion is found in the EquidÆ, as in other families of animals.

We now pass to the experiments made at Penycuik in crossing the zebra Matopo with various mares of different breeds.

1. Matopo was first mated with Mulatto, one of Lord

Arthur Cecil’s black West Highland ponies. The result was the hybrid Romulus, which on the whole, both in mental disposition and bodily form, took more after his father than his mother. His striping was even more marked than that of his sire. He had a semi-erect mane, which was shed annually. The pattern of the markings, on both body and face, resemble the stripes on a Somali zebra—which, as we have seen, is regarded by Professor Ewart as the most primitive type—more than they resembled that of any of Burchell’s zebras. The profuse striping is a point of difference between this hybrid and Lord Morton’s. The quagga-hybrid was less striped than many dun-coloured horses (see illustration).

The mother Mulatto was next mated with a highly-bred grey Arab horse, Benazrek. The offspring agreed in all respects with ordinary foals; it had, however, a certain number of indistinct stripes which could only be detected in certain lights. The stripes were not nearly so clear as in a foal bred by Mr. Darwin from a cross-bred bay mare and a thoroughbred horse, and they disappeared entirely in about five months.

Mulatto has produced a third foal to Loch Corrie, a sire belonging to the Isle of Rum group of West Highland ponies, and closely resembling its mate. This foal was about as much striped as its immediate predecessor. In both cases the pattern of the stripe differed not only from that of Matopo, the previous sire, but from that of the hybrid Romulus. These two foals seem to lend some support to telegony; but the evidence which might be drawn from the second of them is destroyed by the fact that the sire, Loch Corrie, has produced foals from two West Highland mares, one brown and one black, and each of these foals has as many and as well-marked stripes as the foal of Mulatto.

2. Four attempts were made to cross the zebra with Shetland ponies: only one succeeded. The hybrid was a smaller edition of Romulus. The dam Nora had been bred from before, and had produced by a black Shetland pony a foal of a dun colour which was markedly striped. After the birth of the hybrid she was put to a bay Welsh pony; the resulting foal had only the faintest indication of stripes, which soon disappeared. It is a remarkable fact that Nora’s foals were more striped before she had been mated with the zebra than afterwards.

3. Five Iceland ponies were mated with Matopo, of whom one produced, in 1897, a dark-coloured hybrid. The dam, Tundra, was a yellow and white skewbald, which had previously produced a light bay foal to a stallion of its own breed. Her third foal (1898) was fathered by a bay Shetland pony, and in coloration closely resembled its dam. There was no hint of infection in this case. In 1899 Professor Ewart bred from this mare, by Matopo, a zebra-hybrid of a creamy fawn colour, and so primitive in its markings that he believes it to stand in much the same relation to horses, zebras, and asses as the blue-rock does to the various breeds of pigeons (see illustration).

4. Two Irish mares, both bays, produced hybrids by Matopo, and subsequently bore pure-bred foals. One of the latter was by a thoroughbred horse, the other by a hackney pony. The foals were without stripes, and showed no kind of indication that their mother had ever been mated with a zebra.

5. Although Professor Ewart experimented with seven English thoroughbred mares and an Arab, he only succeeded in one case. The mare produced twin hybrids, one of which, unfortunately, died immediately after birth. In the summer of 1899 the same mare produced a foal to a thoroughbred chestnut; ‘neither

in make, colour, nor action’ does it in any way resemble a zebra or a zebra-hybrid.

6. A bay mare which had been in foal to Matopo for some months miscarried. Here—if there is anything in the direct infection theory—the unused germ-cells of the zebra had a better chance than usual of reaching the ova from which future offspring are to arise, yet neither of the two foals which this mare subsequently produced to a thoroughbred horse ‘in any way suggests a zebra.’

The above is the record of the successful experiments which have been tried at Penycuik, with a view of throwing light on the existence of telegony in the EquidÆ. Experiments have also been made with other animals, such as rabbits, dogs, pigeons, fowls, and ducks. Space allows us to quote but one. Six white doe rabbits, all of which had borne pure white offspring to white bucks, were crossed with wild brown rabbits. The result was forty-two young rabbits, all of a bluish-black colour, which in a very short time turned to a brown. These, at the time of writing, were about half grown, and Professor Ewart tells us that it was almost impossible to distinguish them from a full-blooded wild rabbit kept in the same enclosure. The half-breeds, however, were tamer and slightly lighter in colour. The mother does next bred with white bucks again, and in every case bred true. The pure white young showed no trace of throwing back to a previous sire.

A phenomenon somewhat similar to telegony, and one which seems at present quite unexplained, is that a hen which has been crossed with a cock of another breed often lays eggs whose shell is no longer like that of its own breed, but in colour, and frequently in texture, resembles that of the breed with which it has been crossed. Mr. Bulman has recorded a case of this in the pages of ‘Natural Science.’ Some Orpington fowls which laid eggs of a buff tint were allowed to run loose in a large yard with fowls of various breeds. After a few months they were confined in separate pens again, and for several weeks afterwards they continued to lay white eggs. There seems to be no doubt of the existence of this curious phenomenon; it is mentioned by Gadow in his volume on ‘Birds,’ in Bronn’s ‘Thierreich,’ by Nathusius in the Journal fÜr Ornithologie, and in Newton’s ‘Dictionary of Birds.’ When one calls to mind that the shell is deposited by a special shell-gland which is in no way connected with the ovary, but is a part of the quite distinct oviduct, and that the change in the colour of the eggshell must be caused by some change brought about in this gland by cross-fertilization, we begin to recognize how mysterious and inexplicable are many of the problems which affect breeding.

Throughout his account of his experiments Professor Ewart is extremely cautious in claiming to prove anything, but we think he has justified his claim to have shown that telegony by no means always occurs, as many breeders believe. His experiments so far support the view of Continental mule-breeders that telegony, if it takes place, occurs very seldom. But the experiments are not complete, and it is much to be hoped that they may be continued. If it should subsequently appear that out of fifty pure-bred foals from dams which have been previously mated with the zebra no single instance of telegony be found, the doctrine may surely be neglected by breeders; and if in the experiments which are now being carried out with various other mammals and birds telegony does not occur, the doctrine may be relegated to what the Americans would term the ‘dumping-ground’ of old superstitions. The present state of the matter may be summed up in the Professor’s own words: ‘The experiments, as far as they have gone, afford no evidence in support of the telegony hypothesis.’ Nothing has occurred which is not explicable on the theory of reversion.

Partly owing to a certain doubt or distrust which has recently been expressed as to the existence of reversion, and no doubt partly because it is reasonable to hold that the phenomena of telegony may all be referred to reversion, Professor Ewart has made some direct experiments on this subject. Darwin, Tegetmeier, and many others have made numerous breeding experiments on pigeons, with the result that we may say that the crossing of extreme forms usually tends to reversion in the offspring. The ancestor of the domestic pigeon is known with tolerable certainty to have been the blue-rock pigeon, Columba livia. By crossing a male barb-fantail and a female barb-spot Darwin produced a bird ‘which was hardly distinguishable from the wild Shetland species’ of blue-rock. In his description of this experiment, Darwin, as Weismann points out, confines himself chiefly to the coloration: he does not inquire how far reversion also appears in the structure of the bird. This question has been answered by one of Professor Ewart’s many experiments with pigeons. He crossed a white fantail cock with the offspring of an owl and an archangel. The fantail was pure white, with thirty feathers in its tail, and was so prepotent as to produce white offspring when mated with blue pouters. The owl-archangel was more of an owl than an archangel. One of the young of this complex pair had the coloration of the Shetland rock pigeon, which has a white croup and the wings in front of the bars a uniform blue; the other resembled the Indian rock pigeon in having a blue croup and the front part of the wings chequered. In this second bird there was complete reversion as to colour, and in the first, wherever measurements were possible, there was practically complete reversion also as to form. ‘In its measurements it is relatively almost identical with a typical Shetland blue-rock.’ The tail feathers are twelve in number, and show but the faintest indications of any colour-inheritance from their immediate parents. An additional point of interest is that in disposition this bird seems wilder and more shy than the domesticated breeds usually are. It is vigorous and hardy, and is much admired by the fanciers.

Another bird whose wild ancestor is known with a high degree of certainty is the barn-door fowl. It has sprung from the jungle fowl, Gallus bankiva, and less remotely from the game fowl. Hence, if fowls of different breeds are crossed, the offspring, should reversion occur, ought to resemble either the jungle fowl or their less remote ancestors, the game fowl. A dark red-breasted bantam was crossed with an Indian game Dorking; of the nine chickens which resulted, six resembled Dorkings, and three in both form and colour resembled game birds. Two of the three grew up, and the only visible trace of their parentage was a double comb inherited from their cross-bred father. Here again the reversion does not stop at the colour and form, but extends to disposition; the birds are very shy, and fly about like wild birds. The above are but two instances out of many which might be quoted from the Penycuik experiments; they are, however, unusually clear cases, and should do something to restore confidence amongst recent doubters of reversion.

An animal is said to be prepotent when it strongly impresses its own peculiarities of form, colour, temperament, etc., on its offspring. In the above-mentioned experiment with pigeons the owl had been prepotent over the archangel in the mother of the offspring which showed such marked reversion. There is no factor in breeding of more importance than prepotency, and none which it is more difficult to estimate. The term is necessarily a relative one, and, further, it may affect some characters and not others. Often it must go undetected, as in the case of the leader of a herd of wild cattle, who may be highly prepotent, but whose prepotency, unless he is mated with members of another herd displaying different characters, may pass unnoticed. Breeders claim to be able to produce cattle so prepotent that they will produce their like however mated. A well-known dealer in highly-bred ponies used to boast that he had a filly so prepotent that, though she were sent to the best Clydesdale stallion in Scotland, she would throw a colt showing no cart-horse blood. Prepotency is usually obtained by inbreeding, which up to a certain point fixes the character of a race, and in all cases tends to check variation and reversion—the Jews, for instance, as a race are strongly prepotent—but there is no doubt that it may also arise as a sport, and this is probably its more usual origin in a state of nature. Professor Ewart, however, believes that inbreeding is much commoner among wild animals than has usually been conceded, and he holds the opinion that the prepotency so induced has played a considerable part in the origin of species. This, if true, would to some extent take the place of Romanes’ ‘physiological selection’; for Romanes also thought that, though of great importance, variation and natural selection were insufficient to account for the origin of species without some factor which would help to mitigate the swamping effect of intercrossing—some such agency as the fences of modern farms and cattle-ranches—without which the famous cattle breeds of the world would soon disappear in a general ‘regression towards mediocrity.’

In inbreeding the great difficulty of the breeder is to know when to stop. Carried too far it undoubtedly leads to degeneracy. In the ‘Domesticated Animals of Great Britain,’ Lowe records the case of a gentleman who inbred foxhounds to such an extent that ‘the race actually became monstrous and perished.’ Hogs, if too closely inbred, grow hair instead of bristles; their legs become short and unable to support the body; and not only is their fertility diminished, but the mothers cannot nourish the young. That infertility is induced by inbreeding is further shown by some experiments of Ritzema Bos with rats. From seven rats of one family and an unrelated male he continued inbreeding for a period of some six years, and bred about thirty generations. The average of the numbers in each litter fell from 7½ in 1887 to 4⁷/₁₂ in 1891 and 3? in 1892. Further, the offspring of inbred parents are usually weak. Sir Everett Millais estimated that 60 to 70 per cent. of inbred dogs attacked by distemper were carried off.

On the other hand, inbreeding often succeeds even when carried to what the ordinary man would consider excess. The ‘Herd-book’ contains the following case in point. The bull Bolingbroke and the cow Phoenix were more closely related to one another than half-brother is to half-sister. They were mated, and produced the bull Favourite. Favourite was then coupled with his dam, and produced the cow Young Phoenix; he was then coupled with his daughter Young Phoenix, and the world-famed Comet was the result. Professor Ewart tells us that if there was little crossing in the production of Comet, there was still less in that of Clarissa, the mother of the celebrated Restless. An instance of the faith in close inbreeding which exists in the minds of breeders occurred in a letter which the Field published in 1898, in which the writer stated he had heard ‘Mr. Joseph Osborne, the ablest authority living on English

thoroughbreds, declare that you cannot now get too much of Birdcatcher.’

So far as is known, no direct investigations have been made to test how far inbreeding may be carried in the EquidÆ; but, on the other hand, the breeding of racehorses may perhaps be looked upon as a gigantic experiment in this direction. Our English thoroughbreds can be traced back to a few imported sires—the Byerly Turk, imported in 1689; the Darley Arabian, in 1710; and the Godolphin Arabian, in 1730. Since then, by careful breeding and nutrition, they have increased on an average some 8 or 9 inches in height. There is, however, a widely-spread impression that at present there is a marked deterioration in the staying power and in the general ‘fitness’ of the racer. The falling off is further shown by a fact commented on by Sir Walter Gilbey—viz., ‘the smallness of the percentage of even tolerably successful horses out of a prodigious number bred at an enormous outlay.’ In support of this he quotes a sentence from the Times (December 27, 1897), referring to a sale in which thirty-two yearlings had been sold for 51,250 guineas.

If, then, it is true that the English racehorse is on the down grade, what steps should be taken to arrest this descent? Sir Everett Millais restored a pack of basset hounds by crossing them with a bloodhound, the original forefather of bassets. The resulting pups were bassets in form, but not quite bassets in colour; when, however, these cross-breeds were mated with bassets the majority of the pups turned out to be perfect bassets both in shape and coloration. This indicates that one way to rejuvenate the racehorse would be to have recourse to a new importation of the best Arab mares that the plains of Arabia can produce. Breeders hesitate to adopt this course, because their present breed is not only larger, but, over very short distances, fleeter than its forefathers. The shortening of the course in recent years is probably a further sign of the degeneracy of our present racers. Were new blood introduced, and more three-or four-mile races instituted, we should doubtless soon have a return to the champion form of bygone days. Another method would be to import some of the racers of Australia or New Zealand, and cross them with the home product. Different surroundings, food, etc., soon influence the constitution, and this being so, it would be advisable to select those horses of pure descent which have been longest subjected to these altered conditions. Thus the chance of reversion occurring would be increased.

It has been noticed more than once in the preceding pages that a young animal showing reversion is strong and vigorous. It is the belief of dog-breeders that those members of an inbred litter which show reversion are the strongest and best. Similarly, experience shows that if an inbred sire and dam produce a dun-coloured striped foal it almost always turns out well. Reversion is accompanied by a rejuvenescence; it is as though the young animal had appeared at an earlier period in the life-history of the race, before the race had undergone those changes in the way of deterioration which so often accompany inbreeding.

Wild animals are frequently thought to be prepotent over tame ones, but of the eleven zebra-hybrids bred at Penycuik only two took markedly after their sire, the zebra Matopo.[2] There are other experiments recounted which tell the other way, and at present this matter remains in a state of considerable uncertainty. Further experiment may probably show that though in most cases the oldest type is likely to prevail, the offspring may take after the most inbred of its parents. The matter is not altogether as simple as the above statements would imply. For instance, a sport is often strongly prepotent. Standfuss’s experiments in hybridizing butterflies tend to show this, and Mr. Galton even looks upon prepotency as a sport or an aberrant variation. These butterfly experiments also indicate that the male is usually prepotent over the female; but so many questions of nutrition, the maturity of the germ-cells, etc., enter into these intricate problems that it is exceedingly difficult to disentangle the several factors which play a part in the constitution of every living being.

Some years ago it used to be taught that species are infertile inter se; nowadays it almost seems that we are giving up the idea of species altogether. No two naturalists take precisely the same view of what constitutes a species, and no one has succeeded in defining shortly and clearly what a species is. The intersterility test has broken down; the common goose and the Chinese goose, the common duck and the pintail duck, various species of pheasant, the ox of Europe and the American bison or the Indian zebu, not only breed together, but yield hybrids which are themselves fertile; and the same is true of many plants. Why the hybrids of EquidÆ should prove sterile is not clear.

This article must not close without a word or two more about the zebra-hybrids. It is mentioned above that only two out of the eleven which have already been born took strongly after their father. This is no proof that the wilder animal is not prepotent. Recent experiments in hybridizing echinoderms, star-fish, seaurchins, etc., show that the hybrid tends to resemble that species whose germ-cells are most nearly approaching maturity; and thus the nutrition of the germ-cell is but another thread in that complex tangle of heredity which must not be overlooked in attempting to estimate the part played by prepotency and reversion.

Those who have seen the young hybrids playing about in the fields at Penycuik must agree that they are the most charming and compactly built little animals possible; ‘marvellous steeds, striped as a melon is, all black and white,’ as the poet has it. Of Romulus, the eldest of the herd, Professor Ewart says:

The writer can fully confirm all the praise Professor Ewart lavishes on his pets; in truth Romulus has been well described as a ‘bonnie colt with rare quality of bone ... and with the dainty step and dignity of the zebra.’ Remus, the offspring of the Irish mare, was from the first more friendly than his half-brother; he objected less to the process of weaning, and promised to be the handsomest and fleetest of the existing hybrids.

On the whole the hybrids are unusually hardy; at the time of writing only two have been lost—one, a twin, which died almost as soon as it was born, and

another which lived some three months and then succumbed. It is only fair to say that the dam of the latter, who was only three years old when the hybrid was born, had been much weakened by attacks of the strongylus worm, and that she was the victim of close inbreeding. Both the zebras and the hybrids which have been under observation at Penycuik show a remarkable capacity for recovering from wounds. Accidental injuries heal with great rapidity. On one occasion the surviving twin was discovered with a flap of skin some five inches long hanging down over the front of the left fetlock. The skin was stitched into its place again, during which operation the little hybrid fought desperately, and cried piteously; but it soon recovered, the wound healed, and now scarcely a scar remains. There was no lameness and no swelling either at the fetlock or above the knee. Some time ago four hybrid colts and three ordinary foals were attacked by that scourge of the stable, the strongylus worm. One of the latter died and another was reduced almost to a skeleton: the hybrids, though obviously affected, suffered much less than the others, and soon recovered. It is further noticeable that the hybrids suffer less from colds and other slight ailments than the mares and horses amongst which they live. Thus it seems that Colonel Lugard’s hope has to some extent proved true. Some years ago, when administering British East Africa, he strongly recommended the breeding of zebra mules from both the horse and the donkey, believing that they would prove exceptionally hardy and possibly impervious to the tsetse fly. So far as Professor’s Ewart’s experiments go, the first part of the forecast has proved correct. Unfortunately, the latter half has not been justified.

The much dreaded tsetse fly, which has interfered so seriously with the colonization of whole tracts of South Africa, is now known not to be the direct cause of the disease which follows its puncture, but to be the means by which the organism which causes the disease is introduced into the body. In this respect the tsetse fly resembles the malarial mosquito. It is not thought that the organism—a hÆmatozoon—passes through any of the stages of its life-history within the body of the fly, but that the proboscis of that insect merely acts like an inoculating needle. An answer to the important question, Are zebra-hybrids fly-proof or not? has been attempted. Professor Ewart generously allowed an experiment to be tried on two of his hybrids, which were inoculated with the hÆmatozoon, supplied from the Pathological Laboratory at Cambridge. The result was unfortunate, for, although the hybrids resisted the disease far longer than a mare which was also inoculated as a control experiment, both ultimately succumbed.

There is no doubt that it is a comparatively easy matter to breed these hybrids, and that they are not only extremely attractive animals to the eye, but hardy and vigorous, possessed of great staying powers, and promising to be capable of severe work. It is recognized that one of the gravest difficulties which the Indian Army Corps has to contend with is the paucity of mules, both for transport and mountain-battery work; and at the time of the South African War a Commission was busily employed purchasing mules both in Italy and in Texas, and elsewhere. Should these hybrids turn out as well as they at present promise, they may fill a want which is acutely felt by those responsible for the conduct of our frequent ‘small wars,’ and, if bred largely in East Africa, may, as Colonel Lugard suggested, prove a source of wealth and revenue in the future.

We have hitherto said little or nothing about the book itself with which we have been dealing. The larger part consists of three articles reprinted from the Veterinarian and one from the Zoologist; but the more recent and more important half is the General Introduction, covering a hundred pages, in which Professor Ewart sums up the results of his experiments. The form of the work necessarily involves a good deal of repetition, but in so complex a subject this is on the whole rather an advantage than otherwise. Professor Ewart’s style is clear, and his pages abound in apposite illustrations. The book cannot fail to attract both the man of science and the practical breeder.

From what we have said it is evident that the Penycuik experiments are of the highest interest both practical and theoretical, and the public spirit and self-devotion shown by the Edinburgh professor in carrying them out cannot be too widely recognized. The expense of feeding and housing some thirty to forty horses, asses, and zebras is very great, and the initial expenditure in erecting stables, buying land and fencing it, is also considerable. It is, perhaps, not too much to hope that some public body may be willing to undertake at least a part of the burden. The Zoological Society of London possesses, not only the necessary establishment required, including a well-trained staff, but it also has facilities for obtaining all kinds of animals which are far greater than those of any private individual. We hope that the day is not far distant when experiments of this kind will be systematically carried on under the direction of the authorities who control the Gardens in Regent’s Park. Probably such experiments would have better prospects of success at a farm in the country than in London, and there is much to be said for such an experimental farm under the management of a body like the Zoological Society. Apart from the more strictly scientific use to which it might be put, it would serve as a convenient sanatorium for those animals which cannot stand the fogs and damp of London.