To sum up the whole curious case: wild silver-greys may be considered as black rabbits which become grey at an early period of life. When they are crossed with common rabbits, the offspring are said not to have blended colours, but to take after either parent; and in this respect they resemble black and albino varieties of most quadrupeds, which often transmit their colours in this same manner. When they are crossed with chinchillas, that is, with a paler sub-variety, the young are at first pure albinoes, but soon become dark-coloured in certain parts of their bodies, and are then called Himalayans. The young Himalayans, however, are sometimes at first either pale grey or completely black, in either case changing after a time to white. In a future chapter I shall advance a large body of facts showing that, when two varieties are crossed both of which differ in colour from their parent-stock, there is a strong tendency in the young to revert to the aboriginal colour; and what is very remarkable, this reversion occasionally supervenes, not before birth, but during the growth of the animal. Hence, if it could be shown that silver-greys and chinchillas were the offspring of a cross between a black and albino variety with the colours intimately blended—a supposition in itself not improbable, and supported by the circumstance of silver-greys in warrens sometimes producing creamy-white young, which ultimately become black—then all the above given paradoxical facts on the changes of colour in silver-greys and in their descendants the Himalayans would come under the law of reversion, supervening at different periods of growth and in different degrees, either to the original black or to the original albino parent-variety.

It is, also, remarkable that Himalayans, though produced so suddenly; breed true. But as, whilst young, they are albinoes, the case falls under a very general rule; albinism being well known to be strongly inherited, for instance with white mice and many other quadrupeds, and even white flowers. But why, it may be asked, do the ears, tail, nose, and feet, and no other part of the body, revert to a black colour? This apparently depends on a law, which generally holds good, namely, that characters common to many species of a genus—and this, in fact, implies long inheritance from the ancient progenitor of the genus—are found to resist variation, or to reappear if lost, more persistently than the characters which are confined to the separate species. Now, in the genus Lepus, a large majority of the species have their ears and the upper surface of the tail tinted black; but the persistence of these marks is best seen in those species which in winter become white: thus, in Scotland the L. variabilis (4/19. G.R. Waterhouse 'Natural History of Mammalia: Rodents' 1846 pages 52, 60, 105.) in its winter dress has a shade of colour on its nose, and the tips of its ears are black: in the L. tibetanus the ears are black, the upper surface of the tail greyish-black, and the soles of the feet brown: in L. glacialis the winter fur is pure white, except the soles of the feet and the points of the ears. Even in the variously-coloured fancy rabbits we may often observe a tendency in these same parts to be more darkly tinted than the rest of the body. Thus the several coloured marks on the Himalayan rabbits, as they grow old, are rendered intelligible. I may add a nearly analogous case: fancy rabbits very often have a white star on their foreheads; and the common English hare, whilst young, generally has, as I have myself observed, a similar white star on its forehead.

When variously coloured rabbits are set free in Europe, and are thus placed under their natural conditions, they generally revert to the aboriginal grey colour; this may be in part due to the tendency in all crossed animals, as lately observed, to revert to their primordial state. But this tendency does not always prevail; thus silver-grey rabbits are kept in warrens, and remain true though living almost in a state of nature; but a warren must not be stocked with both silver-greys and common rabbits; otherwise "in a few years there will be none but common greys surviving." (4/20. Delamer on 'Pigeons and Rabbits' page 114.) When rabbits run wild in foreign countries under new conditions of life, they by no means always revert to their aboriginal colour. In Jamaica the feral rabbits are described as having been "slate-coloured, deeply tinted with sprinklings of white on the neck, on the shoulders, and on the back; softening off to blue-white under the breast and belly." (4/21. Gosse 'Sojourn in Jamaica' 1851 page 441 as described by an excellent observer, Mr. R. Hill. This is the only known case in which rabbits have become feral in a hot country. They can be kept, however, at Loanda (see Livingstone 'Travels' page 407). In parts of India, as I am informed by Mr. Blyth, they breed well.) But in this tropical island the conditions were not favourable to their increase, and they never spread widely, and are now extinct, as I hear from Mr. R. Hill, owing to a great fire which occurred in the woods. Rabbits during many years have run wild in the Falkland Islands; they are abundant in certain parts, but do not spread extensively. Most of them are of the common grey colour; a few, as I am informed by Admiral Sulivan, are hare- coloured, and many are black, often with nearly symmetrical white marks on their faces. Hence, M. Lesson described the black variety as a distinct species, under the name of Lepus magellanicus, but this, as I have elsewhere shown, is an error. (4/22. Darwin 'Journal of Researches' page 193; and 'Zoology of the Voyage of the Beagle: Mammalia' page 92.) Within recent times the sealers have stocked some of the small outlying islets in the Falkland group with rabbits; and on Pebble Islet, as I hear from Admiral Sulivan, a large proportion are hare-coloured, whereas on Rabbit Islet a large proportion are of a bluish colour, which is not elsewhere seen. How the rabbits were coloured which were turned out of these islets is not known.

The rabbits which have become feral on the island of Porto Santo, near Madeira, deserve a fuller account. In 1418 or 1419, J. Gonzales Zarco (4/23. Kerr 'Collection of Voyages' volume 2 page 177: page 205 for Cada Mosto. According to a work published in Lisbon in 1717 entitled 'Historia Insulana' written by a Jesuit, the rabbits were turned out in 1420. Some authors believe that the island was discovered in 1413.) happened to have a female rabbit on board which had produced young during the voyage, and he turned them all out on the island. These animals soon increased so rapidly, that they became a nuisance, and actually caused the abandonment of the settlement. Thirty-seven years subsequently, Cada Mosto describes them as innumerable; nor is this surprising, as the island was not inhabited by any beast of prey or by any terrestrial mammal. We do not know the character of the mother-rabbit; but it was probably the common domesticated kind. The Spanish peninsula, whence Zarco sailed, is known to have abounded with the common wild species at the most remote historical period; and as these rabbits were taken on board for food, it is improbable that they should have been of any peculiar breed. That the breed was well domesticated is shown by the doe having littered during the voyage. Mr. Wollaston, at my request, brought home two of these feral rabbits in spirits of wine; and, subsequently, Mr. W. Haywood sent to me three more specimens in brine, and two alive. These seven specimens, though caught at different periods, closely resembled each other. They were full grown, as shown by the state of their bones. Although the conditions of life in Porto Santo are evidently highly favourable to rabbits, as proved by their extraordinarily rapid increase, yet they differ conspicuously in their small size from the wild English rabbit. Four English rabbits, measured from the incisors to the anus, varied between 17 and 17 3/4 inches in length; whilst two of the Porto Santo rabbits were only 14 1/2 and 15 inches in length. But the decrease in size is best shown by weight; four wild English rabbits averaged 3 pounds 5 ounces, whilst one of the Porto Santo rabbits, which had lived for four years in the Zoological Gardens, but had become thin, weighed only 1 pound 9 ounces. A fairer test is afforded by the comparison of the well-cleaned limb-bones of a Porto Santo rabbit killed on the island with the same bones of a wild English rabbit of average size, and they differed in the proportion of rather less than five to nine. So that the Porto Santo rabbits have decreased nearly three inches in length, and almost half in weight of body. (4/24. Something of the same kind has occurred on the island of Lipari, where, according to Spallanzani ('Voyage dans les deux Siciles' quoted by Godron 'De l'Espece' page 364), a countryman turned out some rabbits which multiplied prodigiously, but, says Spallanzani, "les lapins de l'ile de Lipari sont plus petits que ceux qu'on eleve en domesticite.") The head has not decreased in length proportionally with the body; and the capacity of the brain case is, as we shall hereafter see, singularly variable. I prepared four skulls, and these resembled each other more closely than do generally the skulls of wild English rabbits; but the only difference in structure which they presented was that the supra-orbital processes of the frontal bones were narrower.

In colour the Porto Santo rabbit differs considerably from the common rabbit; the upper surface is redder, and is rarely interspersed with any black or black-tipped hairs. The throat and certain parts of the under surface, instead of being pure white, are generally pale grey or leaden colour. But the most remarkable difference is in the ears and tail; I have examined many fresh English rabbits, and the large collection of skins in the British Museum from various countries, and all have the upper surface of the tail and the tips of the ears clothed with blackish-grey fur; and this is given in most works as one of the specific characters of the rabbit. Now in the seven Porto Santo rabbits the upper surface of the tail was reddish-brown, and the tips of the ears had no trace of the black edging. But here we meet with a singular circumstance: in June, 1861 I examined two of these rabbits recently sent to the Zoological Gardens, and their tails and ears were coloured as just described; but when one of their dead bodies was sent to me in February, 1865, the ears were plainly edged, and the upper surface of the tail was covered with blackish-grey fur, and the whole body was much less red; so that under the English climate this individual rabbit had recovered the proper colour of its fur in rather less than four years!

The two little Porto Santo rabbits, whilst alive in the Zoological Gardens, had a remarkably different appearance from the common kind. They were extraordinarily wild and active, so that many persons exclaimed on seeing them that they were more like large rats than rabbits. They were nocturnal to an unusual degree in their habits, and their wildness was never in the least subdued; so that the superintendent, Mr. Bartlett, assured me that he had never had a wilder animal under his charge. This is a singular fact, considering that they are descended from a domesticated breed. I was so much surprised at it, that I requested Mr. Haywood to make inquiries on the spot, whether they were much hunted by the inhabitants, or persecuted by hawks, or cats, or other animals; but this is not the case, and no cause can be assigned for their wildness. They live both on the central, higher rocky land and near the sea-cliffs, and, from being exceedingly shy and timid, seldom appear in the lower and cultivated districts. They are said to produce from four to six young at a birth, and their breeding season is in July and August. Lastly, and this is a highly remarkable fact, Mr. Bartlett could never succeed in getting these two rabbits, which were both males, to associate or breed with the females of several breeds which were repeatedly placed with them.

If the history of these Porto Santo rabbits had not been known, most naturalists, on observing their much reduced size, their colour, reddish above and grey beneath, their tails and ears not tipped with black, would have ranked them as a distinct species. They would have been strongly confirmed in this view by seeing them alive in the Zoological Gardens, and hearing that they refused to couple with other rabbits. Yet this rabbit, which there can be little doubt would thus have been ranked as a distinct species, as certainly originated since the year 1420. Finally, from the three cases of the rabbits which have run wild in Porto Santo, Jamaica, and the Falkland Islands, we see that these animals do not, under new conditions of life, revert to or retain their aboriginal character, as is so generally asserted to be the case by most authors.

OSTEOLOGICAL CHARACTERS.

When we remember, on the one hand, how frequently it is stated that important parts of the structure never vary; and, on the other hand, on what small differences in the skeleton fossil species have often been founded, the variability of the skull and of some other bones in the domesticated rabbit well deserves attention. It must not be supposed that the more important differences immediately to be described strictly characterise any one breed; all that can be said is, that they are generally present in certain breeds. We should bear in mind that selection has not been applied to fix any character in the skeleton, and that the animals have not had to support themselves under uniform habits of life. We cannot account for most of the differences in the skeleton; but we shall see that the increased size of the body, due to careful nurture and continued selection, has affected the head in a particular manner. Even the elongation and lopping of the ears have influenced in a small degree the form of the whole skull. The want of exercise has apparently modified the proportional length of the limbs in comparison with that of the body.

[As a standard of comparison, I prepared skeletons of two wild rabbits from Kent, one from the Shetland Islands, and one from Antrim in Ireland. As all the bones in these four specimens from such distant localities closely resembled each other, presenting scarcely any appreciable difference, it may be concluded that the bones of the wild rabbit are generally uniform in character.

SKULL.

I have carefully examined skulls of ten large lop-eared rabbits, and of five common domestic rabbits, which latter differ from the lop-eared only in not having such large bodies or ears, yet both larger than in the wild rabbit. First for the ten lop-eared rabbits: in all these the skull is remarkably elongated in comparison with its breadth. In a wild rabbit the length was 3.15 inches, in a large fancy rabbit 4.3; whilst the breadth of the cranium enclosing the brain was in both almost exactly the same. Even by taking as the standard of comparison the widest part of the zygomatic arch, the skulls of the lop-eared are proportionally to their breadth three-quarters of an inch too long. The depth of the head has increased almost in the same proportion with the length; it is the breadth alone which has not increased. The parietal and occipital bones enclosing the brain are less arched, both in a longitudinal and transverse line, than in the wild rabbit, so that the shape of the cranium is somewhat different. The surface is rougher, less cleanly sculptured, and the lines of sutures are more prominent.

Although the skulls of the large lop-eared rabbits in comparison with those of the wild rabbit are much elongated relatively to their breadth, yet, relatively to the size of body, they are far from elongated. The lop-eared rabbits which I examined were, though not fat, more than twice as heavy as the wild specimens; but the skull was very far from being twice as long. Even if we take the fairer standard of the length of body, from the nose to the anus, the skull is not on an average as long as it ought to be by a third of an inch. In the small feral Porto Santo rabbit, on the other hand, the head relatively to the length of body is about a quarter of an inch too long.

This elongation of the skull relatively to its breadth, I find a universal character, not only with the large lop-eared rabbits, but in all the artificial breeds; as is well seen in the skull of the Angora. I was at first much surprised at the fact, and could not imagine why domestication could produce this uniform result; but the explanation seems to lie in the circumstance that during a number of generations the artificial races have been closely confined, and have had little occasion to exert either their senses, or intellect, or voluntary muscles; consequently the brain, as we shall presently more fully see, has not increased relatively with the size of body. As the brain has not increased, the bony case enclosing it has not increased, and this has evidently affected through correlation the breadth of the entire skull from end to end.

(FIGURE 6. SKULL OF WILD RABBIT, of natural size.

FIGURE 7. SKULL OF LARGE LOP-EARED RABBIT, of natural size.

FIGURE 8. PART OF ZYGOMATIC ARCH, showing the projecting end of the malar bone of the auditory meatus: of natural size. Upper figure, Wild Rabbit. Lower figure, Lop-eared, hare-coloured Rabbit.

FIGURE 9. POSTERIOR END OF SKULL, of natural size, showing the inter- parietal bone. A. Wild Rabbit. B. Feral Rabbit from island of P. Santo, near Madeira. C. Large Lop-eared Rabbit.)

In all the skulls of the large lop-eared rabbits, the supra-orbital plates or processes of the frontal bones are much broader than in the wild rabbit, and they generally project more upwards. In the zygomatic arch the posterior or projecting point of the malar-bone is broader and blunter; and in the specimen, figure 8, it is so in a remarkable degree. This point approaches nearer to the auditory meatus than in the wild rabbit, as may be best seen in figure 8; but this circumstance mainly depends on the changed direction of the meatus. The inter-parietal bone (see figure 9) differs much in shape in the several skulls; generally it is more oval, that is more extended in the line of the longitudinal axis of the skull, than in the wild rabbit. The posterior margin of "the square raised platform" (4/25. Waterhouse 'Nat. Hist. Mammalia' volume 2 page 36.) of the occiput, instead of being truncated, or projecting slightly as in the wild rabbit, is in most lop-eared rabbits pointed, as in figure 9, C. The paramastoids relatively to the size of the skull are generally much thicker than in the wild rabbit.

(FIGURE 10. OCCIPITAL FORAMEN, of natural size, in—A. Wild Rabbit; B. Large Lop-eared Rabbit.)

The occipital foramen (figure 10) presents some remarkable differences: in the wild rabbit, the lower edge between the condyles is considerably and almost angularly hollowed out, and the upper edge is deeply and squarely notched; hence the longitudinal axis exceeds the transverse axis. In the skulls of the lop-eared rabbits the transverse axis exceeds the longitudinal; for in none of these skulls was the lower edge between the condyles so deeply hollowed out; in five of them there was no upper square notch, in three there was a trace of the notch, and in two alone it was well developed. These differences in the shape of the foramen are remarkable, considering that it gives passage to so important a structure as the spinal marrow, though apparently the outline of the latter is not affected by the shape of the passage.

(FIGURE 11. SKULL, of natural size, of Half-lop Rabbit, showing the different direction of the auditory meatus on the two sides, and the consequent general distortion of the skull. The left ear of the animal (or right side of figure) lopped forwards.)

In all the skulls of the large lop-eared rabbits, the bony auditory meatus is conspicuously larger than in the wild rabbit. In a skull 4.3 inches in length, and which barely exceeded in breadth the skull of a wild rabbit (which was 3.15 inches in length), the longer diameter of the meatus was exactly twice as great. The orifice is more compressed, and its margin on the side nearest the skull stands up higher than the outer side. The whole meatus is directed more forwards. As in breeding lop-eared rabbits the length of the ears, and their consequent lopping and lying flat on the face, are the chief points of excellence, there can hardly be a doubt that the great change in the size, form, and direction of the bony meatus, relatively to this same part in the wild rabbit, is due to the continued selection of individuals having larger and larger ears. The influence of the external ear on the bony meatus is well shown in the skulls (I have examined three) of half-lops (see figure 5), in which one ear stands upright, and the other and longer ear hangs down; for in these skulls there was a plain difference in the form and direction of the bony meatus on the two sides. But it is a much more interesting fact, that the changed direction and increased size of the bony meatus have slightly affected on the same side the structure of the whole skull. I here give a drawing (figure 11) of the skull of a half-lop; and it may be observed that the suture between the parietal and frontal bones does not run strictly at right angles to the longitudinal axis of the skull; the left frontal bone projects beyond the right one; both the posterior and anterior margins of the left zygomatic arch on the side of the lopping ear stand a little in advance of the corresponding bones on the opposite side. Even the lower jaw is affected, and the condyles are not quite symmetrical, that on the left standing a little in advance of that on the right. This seems to me a remarkable case of correlation of growth. Who would have surmised that by keeping an animal during many generations under confinement, and so leading to the disuse of the muscles of the ears, and by continually selecting individuals with the longest and largest ears, he would thus indirectly have affected almost every suture in the skull and the form of the lower jaw!

In the large lop-eared rabbits the only difference in the lower jaw, in comparison with that of the wild rabbit, is that the posterior margin of the ascending ramus is broader and more inflected. The teeth in neither jaw present any difference, except that the small incisors, beneath the large ones, are proportionately a little longer. The molar teeth have increased in size proportionately with the increased width of the skull, measured across the zygomatic arch, and not proportionally with its increased length. The inner line of the sockets of the molar teeth in the upper jaw of the wild rabbit forms a perfectly straight line; but in some of the largest skulls of the lop-eared this line was plainly bowed inwards. In one specimen there was an additional molar tooth on each side of the upper jaw, between the molars and premolars; but these two teeth did not correspond in size; and as no rodent has seven molars, this is merely a monstrosity, though a curious one.

The five other skulls of common domestic rabbits, some of which approach in size the above-described largest skulls, whilst the others exceed but little those of the wild rabbit, are only worth notice as presenting a perfect gradation in all the above-specified differences between the skulls of the largest lop-eared and wild rabbits. In all, however, the supra- orbital plates are rather larger, and in all the auditory meatus is larger, in conformity with the increased size of the external ears, than in the wild rabbit. The lower notch in the occipital foramen in some was not so deep as in the wild rabbit, but in all five skulls the upper notch was well developed.

The skull of the Angora rabbit, like the latter five skulls, is intermediate in general proportions, and in most other characters, between those of the largest lop-eared and wild rabbits. It presents only one singular character: though considerably longer than the skull of the wild rabbit, the breadth measured within the posterior supra-orbital fissures is nearly a third less than in the wild. The skulls of the silver-grey, and chinchilla and Himalayan rabbits are more elongated than in the wild, with broader supra-orbital plates, but differ little in any other respect, excepting that the upper and lower notches of the occipital foramen are not so deep or so well developed. The skull of the Moscow rabbit scarcely differs at all from that of the wild rabbit. In the Porto Santo feral rabbits the supra-orbital plates are generally narrower and more pointed than in our wild rabbits.

As some of the largest lop-eared rabbits of which I prepared skeletons were coloured almost like hares, and as these latter animals and rabbits have, as it is affirmed, been recently crossed in France, it might be thought that some of the above-described characters had been derived from a cross at a remote period with the hare. Consequently I examined skulls of the hare, but no light could thus be thrown on the peculiarities of the skulls of the larger rabbits. It is, however, an interesting fact, as illustrating the law that varieties of one species often assume the characters of other species of the same genus, that I found, on comparing the skulls of ten species of hares in the British Museum, that they differed from each other chiefly in the very same points in which domestic rabbits vary,—namely, in general proportions, in the form and size of the supra-orbital plates, in the form of the free end of the malar bone, and in the line of suture separating the occipital and frontal bones. Moreover two eminently variable characters in the domestic rabbit, namely, the outline of the occipital foramen and the shape of the "raised platform" of the occiput, were likewise variable in two instances in the same species of hare.

VERTEBRAE.

The number is uniform in all the skeletons which I have examined, with two exceptions, namely, in one of the small feral Porto Santo rabbits and in one of the largest lop-eared kinds; both of these had as usual seven cervical, twelve dorsal with ribs, but, instead of seven lumbar, both had eight lumbar vertebrae. This is remarkable, as Gervais gives seven as the number for the whole genus Lepus. The caudal vertebrae apparently differ by two or three, but I did not attend to them, and they are difficult to count with certainty.

(FIGURE 12. ATLAS VERTEBRAE, of natural size; inferior surface viewed obliquely. Upper figure, Wild Rabbit. Lower figure, Hare-coloured, large, Lop-eared Rabbit, a, supra-median, atlantoid process; b, infra-median process.)

In the first cervical vertebra, or atlas, the anterior margin of the neural arch varies a little in wild specimens, being either nearly smooth, or furnished with a small supra-median atlantoid process; I have figured a specimen with the largest process (a) which I have seen; but it will be observed how inferior this is in size and different in shape to that in a large lop-eared rabbit. In the latter, the infra-median process (b) is also proportionally much thicker and longer. The alae are a little squarer in outline.

(FIGURE 13. THIRD CERVICAL VERTEBRAE, of natural size, of: A. Wild Rabbit; B. Hare-coloured, large, Lop-eared Rabbit, a, a, inferior surface; b, b, anterior articular surfaces.)

THIRD CERVICAL VERTEBRA.

In the wild rabbit (figure 13, A a) this vertebra, viewed on the inferior surface, has a transverse process, which is directed obliquely backwards, and consists of a single pointed bar; in the fourth vertebra this process is slightly forked in the middle. In the large lop-eared rabbits this process (B a) is forked in the third vertebra, as in the fourth of the wild rabbit. But the third cervical vertebrae of the wild and lop-eared (A b, B b) rabbits differ more conspicuously when their anterior articular surfaces are compared; for the extremities of the antero-dorsal processes in the wild rabbit are simply rounded, whilst in the lop-eared they are trifid, with a deep central pit. The canal for the spinal marrow in the lop-eared (B b) is more elongated in a transverse direction than in the wild rabbit; and the passages for the arteries are of a slightly different shape. These several differences in this vertebra seem to me well deserving attention.

FIRST DORSAL VERTEBRA.

Its neural spine varies in length in the wild rabbit; being sometimes very short, but generally more than half as long as that of the second dorsal; but I have seen it in two large lop-eared rabbits three-fourths of the length of that of the second dorsal vertebra.

(FIGURE 14. DORSAL VERTEBRAE, from sixth to tenth inclusive, of natural size, viewed laterally. A. Wild Rabbit. B. Large, Hare-coloured, so called Spanish Rabbit.)

NINTH AND TENTH DORSAL VERTEBRAE.

In the wild rabbit the neural spine of the ninth vertebra is just perceptibly thicker than that of the eighth; and the neural spine of the tenth is plainly thicker and shorter than those of all the anterior vertebrae. In the large lop-eared rabbits the neural spines of the tenth, ninth, and eighth vertebrae, and even in a slight degree that of the seventh, are very much thicker, and of somewhat different shape, in comparison with those of the wild rabbit. So that this part of the vertebral column differs considerably in appearance from the same part in the wild rabbit, and closely resembles in an interesting manner these same vertebrae in some species of hares. In the Angora, Chinchilla, and Himalayan rabbits, the neural spines of the eighth and ninth vertebrae are in a slight degree thicker than in the wild. On the other hand, in one of the feral Porto Santo rabbits, which in most of its characters deviates from the common wild rabbit, in a direction exactly opposite to that assumed by the large lop-eared rabbits, the neural spines of the ninth and tenth vertebrae were not at all larger than those of the several anterior vertebra. In this same Porto Santo specimen there was no trace in the ninth vertebra of the anterior lateral processes (see figure 14), which are plainly developed in all British wild rabbits, and still more plainly developed in the large lop-eared rabbits. In a half-wild rabbit from Sandon Park (4/26. These rabbits have run wild for a considerable time in Sandon Park, and in other places in Staffordshire and Shropshire. They originated, as I have been informed by the gamekeeper, from variously-coloured domestic rabbits which had been turned out. They vary in colour; but many are symmetrically coloured, being white with a streak along the spine, and with the ears and certain marks about the head of a blackish-grey tint. They have rather longer bodies than common rabbits), a haemal spine was moderately well developed on the under side of the twelfth dorsal vertebra, and I have seen this in no other specimen.

LUMBAR VERTEBRAE.

I have stated that in two cases there were eight instead of seven lumbar vertebrae. The third lumbar vertebrae in one skeleton of a wild British rabbit, and in one of the Porto Santo feral rabbits, had a haemal spine; whilst in four skeletons of large lop-eared rabbits, and in the Himalayan rabbit, this same vertebra had a well developed haemal spine.

PELVIS.

In four wild specimens this bone was almost absolutely identical in shape; but in several domesticated breeds shades of differences could be distinguished. In the large lop-eared rabbits, the whole upper part of the ilium is straighter, or less splayed outwards, than in the wild rabbit; and the tuberosity on the inner lip of the anterior and upper part of the ilium is proportionally more prominent.

(FIGURE 15. TERMINAL BONE OF STERNUM, of natural size, A. Wild Rabbit. B. Hare-coloured, Lop-eared Rabbit. C. Hare-coloured Spanish Rabbit. (N.B. The left-hand angle of the upper articular extremity of B was broken, and has been accidentally thus represented.))

STERNUM.

The posterior end of the posterior sternal bone in the wild rabbit (figure 15, A) is thin and slightly enlarged; in some of the large lop-eared rabbits (B) it is much more enlarged towards the extremity; whilst in other specimens (C) it keeps nearly of the same breadth from end to end, but is much thicker at the extremity.

(FIGURE 16. ACROMION OF SCAPULA, of natural size. A. Wild Rabbit. B, C, D, Large, Lop-eared Rabbits.)

SCAPULA.

The acromion sends out a rectangular bar, ending in an oblique knob, which latter in the wild rabbit (figure 16, A) varies a little in shape and size, as does the apex of the acromion in sharpness, and the part just below the rectangular bar in breadth. But the variations in these respects in the wild rabbit are very slight: whilst in the large lop-eared rabbits they are considerable. Thus in some specimens (B) the oblique terminal knob is developed into a short bar, forming an obtuse angle with the rectangular bar. In another specimen (C) these two unequal bars form nearly a straight line. The apex of the acromion varies much in breadth and sharpness, as may be seen by comparing figures B, C, and D.

LIMBS.

In these I could detect no variation; but the bones of the feet were too troublesome to compare with much care.]

I have now described all the differences in the skeletons which I have observed. It is impossible not to be struck with the high degree of variability or plasticity of many of the bones. We see how erroneous the often-repeated statement is, that only the crests of the bones which give attachment to muscles vary in shape, and that only parts of slight importance become modified under domestication. No one will say, for instance, that the occipital foramen, or the atlas, or the third cervical vertebra is a part of slight importance. If the several vertebrae of the wild and lop-eared rabbits, of which figures have been given, had been found fossil, palaeontologists would have declared without hesitation that they had belonged to distinct species.

[THE EFFECTS OF THE USE AND DISUSE OF PARTS.

In the large lop-eared rabbits the relative proportional length of the bones of the same leg, and of the front and hind legs compared with each other, have remained nearly the same as in the wild rabbit; but in weight, the bones of the hind legs apparently have not increased in due proportion with the front legs. The weight of the whole body in the large rabbits examined by me was from twice to twice and a half as great as that of the wild rabbit; and the weight of the bones of the front and hind limbs taken together (excluding the feet, on account of the difficulty of cleaning so many small bones) has increased in the large lop-eared rabbits in nearly the same proportion; consequently in due proportion to the weight of body which they have to support. If we take the length of the body as the standard of comparison, the limbs of the large rabbits have not increased in length in due proportion by one inch and a half. Again, if we take as the standard of comparison the length of the skull, which, as we have before seen, has not increased in length in due proportion to the length of body, the limbs will be found to be, proportionally with those of the wild rabbit, from half to three-quarters of an inch too short. Hence, whatever standard of comparison be taken, the limb-bones of the large lop-eared rabbits have not increased in length, though they have in weight, in full proportion to the other parts of the frame; and this, I presume, may be accounted for by the inactive life which during many generations they have spent. Nor has the scapula increased in length in due proportion to the increased length of the body.

The capacity of the osseous case of the brain is a more interesting point, to which I was led to attend by finding, as previously stated, that with all domesticated rabbits the length of the skull relatively to its breadth has greatly increased in comparison with that of the wild rabbits. If we had possessed a large number of domesticated rabbits of nearly the same size with the wild rabbits, it would have been a simple task to have measured and compared the capacities of their skulls. But this is not the case: almost all the domestic breeds have larger bodies than wild rabbits, and the lop-eared kinds are more than double their weight. As a small animal has to exert its senses, intellect, and instincts equally with a large animal, we ought not by any means to expect an animal twice or thrice as large as another to have a brain of double or treble the size. (4/27. See Prof. Owen's remarks on this subject in his paper on the 'Zoological Significance of the Brain, etc., of Man, etc.' read before Brit. Association 1862: with respect to Birds see 'Proc. Zoolog. Soc.' January 11, 1848 page 8.) Now, after weighing the bodies of four wild rabbits, and of four large but not fattened lop-eared rabbits, I find that on an average the wild are to the lop-eared in weight as 1 to 2.17; in average length of body as 1 to 1.41; whilst in capacity of skull they are as 1 to 1.15. Hence we see that the capacity of the skull, and consequently the size of the brain, has increased but little, relatively to the increased size of the body; and this fact explains the narrowness of the skull relatively to its length in all domestic rabbits.

TABLE 3: MEASUREMENTS OF WILD AND SEMI-WILD RABBITS.

I. Length of Skull (inches).

II. Length of Body from Incisors to Anus (inches).

III. Weight of whole Body (pounds and ounces).

IV. Capacity of Skull measured by Small Shot (grains).

V. Capacity calculated according to Length of Skull relatively to that of No. 1 (Wild Rabbit, Kent) (grains).

VI. Difference between actual and calculated capacities of Skulls (grains).

VII. Showing how much per cent the Brain, by calculation according to the length of the Skull is too light (-) or too heavy (+), relatively to the Brain of the Wild Rabbit No. 1.

NAME OF BREED: I. II. III. IV. V. VI. VII.

WILD AND SEMI-WILD RABBITS:

1. Wild Rabbit, Kent: 3.15 17.4 3,5 972 .. .. +2

2. Wild Rabbit, Shetland Islands: 3.15 .. .. 979 .. .. +2

3. Wild Rabbit, Ireland: 3.15 .. .. 992 .. .. +2

4. Domestic rabbit, run wild, Sandon: 3.15 18.5 .. 977 .. .. +2

5. Wild, common variety, small specimen, Kent: 2.96 17.0 2,14 875 913 38 - 4

6. Wild, fawn-coloured variety, Scotland: 3.1 .. .. 918 950 32 - 3

7. Silver-grey, small specimen, Thetford warren: 2.95 15.5 2,11 938 910 28 +3

8. Feral rabbit, Porto Santo: 2.83 .. .. 893 873 20 +2 9. Feral rabbit, Porto Santo: 2.85 .. .. 756 879 123 - 16 10.Feral rabbit, Porto Santo: 2.95 .. .. 835 910 75 - 9 Average of three Porto Santo rabbits: 2.88 .. .. 828 888 60 - 7

DOMESTIC RABBITS:

11.Himalayan: 3.5 20.5 .. 963 1080 117 - 12

12.Moscow: 3.25 17.0 3,8 803 1002 199 - 24

13.Angora: 3.5 19.5 3,1 697 1080 383 - 54

14.Chinchilla: 3.65 22.0 .. 995 1126 131 - 13

15.Large lop-eared: 4.1 24.5 7,0 1065 1265 200 - 18 16.Large lop-eared: 4.1 25.0 7,13 1153 1265 112 - 9 17.Large lop-eared: 4.07 .. .. 1037 1255 218 - 21 18.Large lop-eared: 4.1 25.0 7,4 1208 1265 57 - 4 19.Large lop-eared: 4.3 .. .. 1232 1326 94 - 7 20.Large lop-eared: 4.25 .. .. 1124 1311 187 - 16 21.Large hare-coloured: 3.86 24.0 6,14 1131 1191 60 - 5 22.Average of above seven large lop-eared rabbits: 4.11 24.62 7,4 1136 1268 132 - 11

23.Hare (L. timidus) English specimen: 3.61 7,0 1315

24.Hare (L. timidus) German specimen: 3.82 7,0 1455

In the upper half of Table 3 I have given the measurements of the skull of ten wild rabbits; and in the lower half, of eleven thoroughly domesticated kinds. As these rabbits differ so greatly in size, it is necessary to have some standard by which to compare the capacities of their skulls. I have selected the length of skull as the best standard, for in the larger rabbits it has not, as already stated, increased in length so much as the body; but as the skull, like every other part, varies in length, neither it nor any other part affords a perfect standard.

In the first column of figures the extreme length of the skull is given in inches and decimals. I am aware that these measurements pretend to greater accuracy than is possible; but I have found it the least trouble to record the exact length which the compass gave. The second and third columns give the length and weight of body, whenever these observations were made. The fourth column gives the capacity of the skull by the weight of small shot with which the skulls were filled; but it is not pretended that these weights are accurate within a few grains. In the fifth column the capacity is given which the skull ought to have had by calculation, according to the length of skull, in comparison with that of the wild rabbit No. 1; in the sixth column the difference between the actual and calculated capacities, and in the seventh the percentage of increase or decrease, are given. For instance, as the wild rabbit No. 5 has a shorter and lighter body than the wild rabbit No. 1, we might have expected that its skull would have had less capacity; the actual capacity, as expressed by the weight of shot, is 875 grains, which is 97 grains less than that of the first rabbit. But comparing these two rabbits by the length of their skulls, we see that in No. 1 the skull is 3.15 inches in length, and in No. 5 2.96 inches in length; according to this ratio, the brain of No. 5 ought to have had a capacity of 913 grains of shot, which is above the actual capacity, but only by 38 grains. Or, to put the case in another way (as in column VII), the brain of this small rabbit, No. 5, for every 100 grains of weight is only 4 grains too light,—that is, it ought, according to the standard rabbit No. 1, to have been 4 per cent heavier. I have taken the rabbit No. 1 as the standard of comparison because, of the skulls having a full average length, this has the least capacity; so that it is the least favourable to the result which I wish to show, namely, that the brain in all long-domesticated rabbits has decreased in size, either actually, or relatively to the length of the head and body, in comparison with the brain of the wild rabbit. Had I taken the Irish rabbit, No. 3, as the standard, the following results would have been somewhat more striking.

Turning to Table 3: the first four wild rabbits have skulls of the same length, and these differ but little in capacity. The Sandon rabbit (No. 4) is interesting, as, though now wild, it is known to be descended from a domesticated breed, as is still shown by its peculiar colouring and longer body; nevertheless the skull has recovered its normal length and full capacity. The next three rabbits are wild, but of small size, and they all have skulls with slightly lessened capacities. The three Porto Santo feral rabbits (Nos. 8 to 10) offer a perplexing case; their bodies are greatly reduced in size, as in a lesser degree are their skulls in length and in actual capacity, in comparison with the skulls of wild English rabbits. But when we compare the capacities of the skull in the three Porto Santo rabbits, we observe a surprising difference, which does not stand in any relation to the slight difference in the length of their skulls, nor, as I believe, to any difference in the size of their bodies; but I neglected weighing separately their bodies. I can hardly suppose that the medullary matter of the brain in these three rabbits, living under similar conditions, can differ as much as is indicated by the proportional difference of capacity in their skulls; nor do I know whether it is possible that one brain may contain considerably more fluid than another. Hence I can throw no light on this case.

Looking to the lower half of Table 3, which gives the measurements of domesticated rabbits, we see that in all the capacity of the skull is less, but in very various degrees, than might have been anticipated according to the length of their skulls, relatively to that of the wild rabbit No. 1. In line 22 the average measurements of seven large lop-eared rabbits are given. Now the question arises, has the average capacity of the skull in these seven large rabbits increased as much as might have been expected from the greatly increased size of body. We may endeavour to answer this question in two ways: in the upper half of the Table we have measurements of the skulls of six small wild rabbits (Nos. 5 to 10), and we find that on an average the skulls are .18 of an inch shorter, and in capacity 91 grains less, than the average length and capacity of the three first wild rabbits on the list. The seven large lop-eared rabbits, on an average, have skulls 4.11 inches in length, and 1136 grains in capacity; so that these skulls have increased in length more than five times as much as the skulls of the six small wild rabbits have decreased in length; hence we might have expected that the skulls of the large lop-eared rabbits would have increased in capacity five times as much as the skulls of the six small rabbits have decreased in capacity; and this would have given an average increased capacity of 455 grains, whilst the real average increase is only 155 grains. Again, the large lop-eared rabbits have bodies of nearly the same weight and size as the common hare, but their heads are longer; consequently, if the lop-eared rabbits had been wild, it might have been expected that their skulls would have had nearly the same capacity as that of the skull of the hare. But this is far from being the case; for the average capacity of the two hare-skulls (Nos. 23, 24) is so much larger than the average capacity of the seven lop-eared skulls, that the latter would have to be increased 21 per cent to come up to the standard of the hare. (4/23. This standard is apparently considerably too low, for Dr. Crisp ('Proc. Zoolog. Soc.' 1861 page 86) gives 210 grains as the actual weight of the brain of a hare which weighed 7 pounds, and 125 grains as the weight of the brain of a rabbit which weighed 3 pounds 5 ounces, that is, the same weight as the rabbit No. 1 in my list. Now the contents of the skull of rabbit No. 1 in shot is in my table 972 grains; and according to Dr. Crisp's ratio of 125 to 210, the skull of the hare ought to have contained 1632 grains of shot, instead of only (in the largest hare in my table) 1455 grains.)

I have previously remarked that, if we had possessed many domestic rabbits of the same average size with the wild rabbit, it would have been easy to compare the capacity of their skulls. Now the Himalayan, Moscow, and Angora rabbits (Nos. 11, 12, 13 of Table 3) are only a little larger in body and have skulls only a little longer, than the wild animal, and we see that the actual capacity of their skulls is less than in the wild animal, and considerably less by calculation (column 7), according to the difference in the length of their skulls. The narrowness of the brain-case in these three rabbits could be plainly seen and proved by external measurement. The Chinchilla rabbit (No. 14) is a considerably larger animal than the wild rabbit, yet the capacity of its skull only slightly exceeds that of the wild rabbit. The Angora rabbit, No. 13, offers the most remarkable case; this animal in its pure white colour and length of silky fur bears the stamp of long domesticity. It has a considerably longer head and body than the wild rabbit, but the actual capacity of its skull is less than that of even the little wild Porto Santo rabbits. By the standard of the length of skull the capacity (see column 7) is only half of what it ought to have been! I kept this individual animal alive, and it was not unhealthy nor idiotic. This case of the Angora rabbit so much surprised me, that I repeated all the measurements and found them correct. I have also compared the capacity of the skull of the Angora with that of the wild rabbit by other standards, namely, by the length and weight of the body, and by the weight of the limb-bones; but by all these standards the brain appears to be much too small, though in a less degree when the standard of the limb- bones was used; and this latter circumstance may probably be accounted for by the limbs of this anciently domesticated breed having become much reduced in weight, from its long-continued inactive life. Hence I infer that in the Angora breed, which is said to differ from other breeds in being quieter and more social, the capacity of the skull has really undergone a remarkable amount of reduction.]

From the several facts above given,—namely, firstly, that the actual capacity of the skull in the Himalayan, Moscow, and Angora breeds, is less than in the wild rabbit, though they are in all their dimensions rather larger animals; secondly, that the capacity of the skull of the large lop- eared rabbits has not been increased in nearly the same ratio as the capacity of the skull of the smaller wild rabbits has been decreased; and thirdly, that the capacity of the skull in these same large lop-eared rabbits is very inferior to that of the hare, an animal of nearly the same size,—I conclude, notwithstanding the remarkable differences in capacity in the skulls of the small Porto Santo rabbits, and likewise in the large lop-eared kinds, that in all long-domesticated rabbits the brain has either by no means increased in due proportion with the increased length of the head and increased size of the body, or that it has actually decreased in size, relatively to what would have occurred had these animals lived in a state of nature. When we remember that rabbits, from having been domesticated and closely confined during many generations, cannot have exerted their intellect, instincts, senses, and voluntary movements, either in escaping from various dangers or in searching for food, we may conclude that their brains will have been feebly exercised, and consequently have suffered in development. We thus see that the most important and complicated organ in the whole organisation is subject to the law of decrease in size from disuse.

Finally, let us sum up the more important modifications which domestic rabbits have undergone, together with their causes as far as we can obscurely see them. By the supply of abundant and nutritious food, together with little exercise, and by the continued selection of the heaviest individuals, the weight of the larger breeds has been more than doubled. The bones of the limbs taken together have increased in weight, in due proportion with the increased weight of body, but the hind legs have increased less than the front legs; but in length they have not increased in due proportion, and this may have been caused by the want of proper exercise. With the increased size of the body the third cervical has assumed characters proper to the fourth cervical vertebra; and the eighth and ninth dorsal vertebrae have similarly assumed characters proper to the tenth and posterior vertebrae. The skull in the larger breeds has increased in length, but not in due proportion with the increased length of body; the brain has not duly increased in dimensions, or has even actually decreased, and consequently the bony case for the brain has remained narrow, and by correlation has affected the bones of the face and the entire length of the skull. The skull has thus acquired its characteristic narrowness. From unknown causes the supra-orbital process of the frontal bones and the free end of the malar bones have increased in breadth; and in the larger breeds the occipital foramen is generally much less deeply notched than in wild rabbits. Certain parts of the scapula and the terminal sternal bones have become highly variable in shape. The ears have been increased enormously in length and breadth through continued selection; their weight, conjoined probably with the disuse of their muscles, has caused them to lop downwards; and this has affected the position and form of the bony auditory meatus; and this again, by correlation, the position in a slight degree of almost every bone in the upper part of the skull, and even the position of the condyles of the lower jaw.

CHAPTER 1.V.

DOMESTIC PIGEONS.

ENUMERATION AND DESCRIPTION OF THE SEVERAL BREEDS. INDIVIDUAL VARIABILITY. VARIATIONS OF A REMARKABLE NATURE. OSTEOLOGICAL CHARACTERS: SKULL, LOWER JAW, NUMBER OF VERTEBRAE. CORRELATION OF GROWTH: TONGUE WITH BEAK; EYELIDS AND NOSTRILS WITH WATTLED SKIN. NUMBER OF WING-FEATHERS, AND LENGTH OF WING. COLOUR AND DOWN. WEBBED AND FEATHERED FEET. ON THE EFFECTS OF DISUSE. LENGTH OF FEET IN CORRELATION WITH LENGTH OF BEAK. LENGTH OF STERNUM, SCAPULA, AND FURCULUM. LENGTH OF WINGS. SUMMARY ON THE POINTS OF DIFFERENCE IN THE SEVERAL BREEDS.

I have been led to study domestic pigeons with particular care, because the evidence that all the domestic races are descended from one known source is far clearer than with any other anciently domesticated animal. Secondly, because many treatises in several languages, some of them old, have been written on the pigeon, so that we are enabled to trace the history of several breeds. And lastly, because, from causes which we can partly understand, the amount of variation has been extraordinarily great. The details will often be tediously minute; but no one who really wants to understand the progress of change in domestic animals, and especially no one who has kept pigeons and has marked the great difference between the breeds and the trueness with which most of them propagate their kind, will doubt that this minuteness is worth while. Notwithstanding the clear evidence that all the breeds are the descendants of a single species, I could not persuade myself until some years had passed that the whole amount of difference between them, had arisen since man first domesticated the wild rock-pigeon.

I have kept alive all the most distinct breeds, which I could procure in England or from the Continent; and have prepared skeletons of all. I have received skins from Persia, and a large number from India and other quarters of the world. (5/1. The Hon. C. Murray has sent me some very valuable specimens from Persia; and H.M. Consul, Mr. Keith Abbott, has given me information on the pigeons of the same country. I am deeply indebted to Sir Walter Elliot for an immense collection of skins from Madras, with much information regarding them. Mr. Blyth has freely communicated to me his stores of knowledge on this and all other related subjects. The Rajah Sir James Brooke sent me specimens from Borneo, as has H.M. Consul, Mr. Swinhoe, from Amoy in China, and Dr. Daniell from the west coast of Africa.) Since my admission into two of the London pigeon-clubs, I have received the kindest assistance from many of the most eminent amateurs. (5/2. Mr. B.P. Brent, well known for his various contributions to poultry literature, has aided me in every way during several years: so has Mr. Tegetmeier, with unwearied kindness. This latter gentleman, who is well known for his works on poultry, and who has largely bred pigeons, has looked over this and the following chapters. Mr. Bult formerly showed me his unrivalled collection of Pouters, and gave me specimens. I had access to Mr. Wicking's collection, which contained a greater assortment of kinds than could anywhere else be seen; and he has always aided me with specimens and information given in the freest manner. Mr. Haynes and Mr. Corker have given me specimens of their magnificent Carriers. To Mr. Harrison Weir I am likewise indebted. Nor must I by any means pass over the assistance received from Mr. J.M. Eaton, Mr. Baker, Mr. Evans, and Mr. J. Baily, jun., of Mount-street—to the latter gentleman I have been indebted for some valuable specimens. To all these gentlemen I beg permission to return my sincere and cordial thanks.)

The races of the Pigeon which can be distinguished, and which breed true, are very numerous. MM. Boitard and Corbie (5/3. 'Les Pigeons de Voliere et de Colombier' Paris 1824. During forty-five years the sole occupation of M. Corbie was the care of the pigeons belonging to the Duchess of Berry. Bonizzi has described a large number of coloured varieties in Italy: 'Le variazioni dei colombi Domestici' Padova 1873.) describe in detail 122 kinds; and I could add several European kinds not known to them. In India, judging from the skins sent me, there are many breeds unknown here; and Sir W. Elliot informs me that a collection imported by an Indian merchant into Madras from Cairo and Constantinople included several kinds unknown in India. I have no doubt that there exist considerably above 150 kinds which breed true and have been separately named. But of these the far greater number differ from each other only in unimportant characters. Such differences will be here entirely passed over, and I shall confine myself to the more important points of structure. That many important differences exist we shall presently see. I have looked through the magnificent collection of the Columbidae in the British Museum, and, with the exception of a few forms (such as the Didunculus, Calaenas, Goura, etc.), I do not hesitate to affirm that some domestic races of the rock-pigeon differ fully as much from each other in external characters as do the most distinct natural genera. We may look in vain through the 288 known species (5/4. 'Coup d'Oeil sur l'Ordre des Pigeons' par Prince C.L. Bonaparte, Paris 1855. This author makes 288 species, ranked under 85 genera.) for a beak so small and conical as that of the short-faced tumbler; for one so broad and short as that of the barb; for one so long, straight, and narrow, with its enormous wattles, as that of the English carrier; for an expanded upraised tail like that of the fantail; or for an oesophagus like that of the pouter. I do not for a moment pretend that the domestic races differ from each other in their whole organisation as much as the more distinct natural genera. I refer only to external characters, on which, however, it must be confessed that most genera of birds have been founded. When, in a future chapter, we discuss the principle of selection as followed by man, we shall clearly see why the differences between the domestic races are almost always confined to external, or at least to externally visible, characters.

Owing to the amount and gradations of difference between the several breeds, I have found it indispensable in the following classification to rank them under Groups, Races, and Sub-races; to which varieties and sub- varieties, all strictly inheriting their proper characters, must often be added. Even with the individuals of the same sub-variety, when long kept by different fanciers, different strains can sometimes be recognised. There can be no doubt that, if well-characterised forms of the several races had been found wild, all would have been ranked as distinct species, and several of them would certainly have been placed by ornithologists in distinct genera. A good classification of the various domestic breeds is extremely difficult, owing to the manner in which many of the forms graduate into each other; but it is curious how exactly the same difficulties are encountered, and the same rules have to be followed, as in the classification of any natural but difficult group of organic beings. An "artificial classification" might be followed which would present fewer difficulties than a "natural classification;" but then it would interrupt many plain affinities. Extreme forms can readily be defined; but intermediate and troublesome forms often destroy our definitions. Forms which may be called "aberrant" must sometimes be included within groups to which they do not accurately belong. Characters of all kinds must be used; but as with birds in a state of nature, those afforded by the beak are the best and most readily appreciated. It is not possible to weigh the importance of all the characters which have to be used so as to make the groups and sub-groups of equal value. Lastly, a group may contain only one race, and another and less distinctly defined group may contain several races and sub-races, and in this case it is difficult, as in the classification of natural species, to avoid placing too high a value on the number of forms which a group may contain.

In my measurements I have never trusted to the eye; and when speaking of a part being large or small, I always refer to the wild rock-pigeon (Columba livia) as the standard of comparison. The measurements are given in decimals of an inch.

(5/5. As I so often refer to the size of the C. livia, or rock-pigeon, it may be convenient to give the mean between the measurements of two wild birds, kindly sent me by Dr. Edmondstone from the Shetland Islands.

LENGTH IN INCHES:

From feathered base of beak to end of tail: 14.25 From feathered base of beak to oil-gland: 9.5 From tip of beak to end of tail: 15.02 Of tail-feathers: 4.62 From tip to tip of wing: 26.75 Of folded wing: 9.25 Beak.—Length from tip of beak to feathered base: .77 Beak.—Thickness, measured vertically at distal end of nostrils: .23 Beak.—Breadth, measured at same place: .16 Feet.—From end of middle toe (without claw) to distal end of tibia: 2.77 Feet.—From end of middle toe to end of hind toe (without claws): 2.02

WEIGHT: 14 1/4 ounces.)

(FIGURE 17. THE ROCK PIGEON, or Columba livia. The parent-form of all domesticated Pigeons. (5/6. This drawing was made from a dead bird. The six following figures were drawn with great care by Mr. Luke Wells from living birds selected by Mr. Tegetmeier. It may be confidently asserted that the characters of the six breeds which have been figured are not in the least exaggerated.))

DIAGRAM 1. DOMESTIC PIGEONS.

Columba livia or ROCK-PIGEON—

—GROUP I.—(SUB-GROUP (RACE) 1.)—German P. —Lille P.— —Dutch P. —ENGLISH POUTER.

—GROUP II.—(SUB-GROUPS (RACES) 2, 3, 4.)—Kali-Par—Bussora— —Murassa. —Dragon—ENGLISH CARRIER. —Bagadotten—Scanderoon—Pigeon Cygne—RUNT. —TRONFO. —BARB.

—GROUP III.—(SUB-GROUPS (RACES) 5, 6, 7, 8, 9.)— —Java Fantail—FANTAIL —Turbit—AFRICAN OWL. —Persian Tumbler—Lotan Tumbler—Common Tumbler—SHORT-FACED TUMBLER. —INDIAN FRILL-BACK. —JACOBIN.

—GROUP IV.—(SUB-GROUPS (RACES) 10, 11.)— —TRUMPETER. —LAUGHER. —ENGLISH FRILL-BACK. —NUN. —SPOT. —SWALLOW. —DOVECOTE PIGEON.

I will now give a brief description of all the principal breeds. Diagram 1. may aid the reader in learning their names and seeing their affinities. The rock-pigeon, or Columba livia (including under this name two or three closely-allied sub-species or geographical races, hereafter to be described), may be confidently viewed, as we shall see in the next chapter, as the common parent-form. The names in italics on the right-hand side of the page show us the most distinct breeds, or those which have undergone the greatest amount of modification. The lengths of the dotted lines rudely represent the degree of distinctness of each breed from the parent-stock, and the names placed under each other in the columns show the more or less closely connecting links. The distances of the dotted lines from each other approximately represent the amount of difference between the several breeds.

(FIGURE 18. ENGLISH POUTER.)

GROUP I.

This group includes a single race, that of the Pouters. If the most strongly marked sub-race be taken, namely, the Improved English Pouter, this is perhaps the most distinct of all domesticated pigeons.

RACE I. POUTER PIGEONS. (KROPFTAUBEN, GERMAN. GROSSES-GORGES, OR BOULANS, FRENCH.)

Oesophagus of great size, barely separated from the crop, often inflated. Body and legs elongated. Beak of moderate dimensions.

[SUB-RACE 1/I.

The improved English Pouter, when its crop is fully inflated, presents a truly astonishing appearance. The habit of slightly inflating the crop is common to all domestic pigeons, but is carried to an extreme in the Pouter. The crop does not differ, except in size, from that of other pigeons; but is less plainly separated by an oblique constriction from the oesophagus. The diameter of the upper part of the oesophagus is immense, even close up to the head. The beak in one bird which I possessed was almost completely buried when the oesophagus was fully expanded. The males, especially when excited, pout more than the females, and they glory in exercising this power. If a bird will not, to use the technical expression, "play," the fancier, as I have witnessed, by taking the beak into his mouth, blows him up like a balloon; and the bird, then puffed up with wind and pride, struts about, retaining his magnificent size as long as he can. Pouters often take flight with their crops inflated. After one of my birds had swallowed a good meal of peas and water, as he flew up in order to disgorge them and feed his nearly fledged young, I heard the peas rattling in his inflated crop as if in a bladder. When flying, they often strike the backs of their wings together, and thus make a clapping noise.

Pouters stand remarkably upright, and their bodies are thin and elongated. In connexion with this form of body, the ribs are generally broader and the vertebrae more numerous than in other breeds. From their manner of standing their legs appear longer than they really are, though, in proportion with those of C. livia, the legs and feet are actually longer. The wings appear much elongated, but by measurement, in relation to the length of body, this is not the case. The beak likewise appears longer, but it is in fact a little shorter (about .O3 of an inch), proportionally with the size of the body, and relatively to the beak of the rock-pigeon. The Pouter, though not bulky, is a large bird; I measured one which was 34 1/2 inches from tip to tip of wing, and 19 inches from tip of beak to end of tail. In a wild rock- pigeon from the Shetland Islands the same measurements gave only 28 1/4 and 14 3/4. There are many sub-varieties of the Pouter of different colours, but these I pass over.

SUB-RACE 1/II. DUTCH POUTER.

This seems to be the parent-form of our improved English Pouters. I kept a pair, but I suspect that they were not pure birds. They are smaller than English pouters, and less well developed in all their characters. Neumeister (5/7. 'Das Ganze der Taubenzucht' Weimar 1837 pl. 11 and 12.) says that the wings are crossed over the tail, and do not reach to its extremity.

SUB-RACE 1/III. THE LILLE POUTER.

I know this breed only from description. (5/8. Boitard and Corbie 'Les Pigeons' etc. page 177 pl. 6.) It approaches in general form the Dutch Pouter, but the inflated oesophagus assumes a spherical form, as if the pigeon had swallowed a large orange, which had stuck close under the beak. This inflated ball is represented as rising to a level with the crown of the head. The middle toe alone is feathered. A variety of this sub-race, called the claquant, is described by MM. Boitard and Corbie; it pouts but little, and is characterised by the habit of violently hitting its wings together over its back,—a habit which the English Pouter has in a slight degree.

SUB-RACE 1/IV. COMMON GERMAN POUTER.

I know this bird only from the figures and description given by the accurate Neumeister, one of the few writers on pigeons who, as I have found, may always be trusted. This sub-race seems considerably different. The upper part of the oesophagus is much less distended. The bird stands less upright. The feet are not feathered, and the legs and beak are shorter. In these respects there is an approach in form to the common rock- pigeon. The tail-feathers are very long, yet the tips of the closed wings extend beyond the end of the tail; and the length of the wings, from tip to tip, and of the body, is greater than in the English Pouter.]

(FIGURE 19. ENGLISH CARRIER.)

GROUP II.

This group includes three Races, namely, Carriers, Runts, and Barbs, which are manifestly allied to each other. Indeed, certain carriers and runts pass into each other by such insensible gradations that an arbitrary line has to be drawn between them. Carriers also graduate through foreign breeds into the rock-pigeon. Yet, if well-characterised Carriers and Barbs (see figures 19 and 20) had existed as wild species, no ornithologist would have placed them in the same genus with each other or with the rock-pigeon. This group may, as a general rule, be recognised by the beak being long, with the skin over the nostrils swollen and often carunculated or wattled, and with that round the eyes bare and likewise carunculated. The mouth is very wide, and the feet are large. Nevertheless the Barb, which must be classed in this same group, has a very short beak, and some runts have very little bare skin round their eyes.

RACE II. CARRIERS. (TURKISCHE TAUBEN; PIGEONS TURCS, DRAGONS.)

Beak elongated, narrow, pointed; eyes surrounded by much naked, generally carunculated, skin; neck and body elongated.

[SUB-RACE 2/I. THE ENGLISH CARRIER.

[This is a fine bird, of large size, close feathered, generally dark- coloured, with an elongated neck. The beak is attenuated and of wonderful length: in one specimen it was 1.4 inch in length from the feathered base to the tip; therefore nearly twice as long as that of the rock-pigeon, which measured only .77. Whenever I compare proportionally any part in the carrier and rock-pigeon, I take the length of the body from the base of the beak to the end of the tail as the standard of comparison; and according to this standard, the beak in one Carrier was nearly half an inch longer than in the rock-pigeon. The upper mandible is often slightly arched. The tongue is very long. The development of the carunculated skin or wattle round the eyes, over the nostrils, and on the lower mandible, is prodigious. The eyelids, measured longitudinally, were in some specimens exactly twice as long as in the rock-pigeon. The external orifice or furrow of the nostrils was also twice as long. The open mouth in its widest part was in one case .75 of an inch in width, whereas in the rock-pigeon it is only about .4 of an inch. This great width of mouth is shown in the skeleton by the reflexed edges of the ramus of the lower jaw. The head is flat on the summit and narrow between the orbits. The feet are large and coarse; the length, as measured from end of hind toe to end of middle toe (without the claws), was in two specimens 2.6 inches; and this, proportionally with the rock-pigeon, is an excess of nearly a quarter of an inch. One very fine Carrier measured 31 1/2 inches from tip to tip of wing. Birds of this sub-race are too valuable to be flown as carriers.]

SUB-RACE 2/II. DRAGONS; PERSIAN CARRIERS.

[The English Dragon differs from the improved English Carrier in being smaller in all its dimensions, and in having less wattle round the eyes and over the nostrils, and none on the lower mandible. Sir W. Elliot sent me from Madras a Bagdad Carrier (sometimes called khandesi), the name of which shows its Persian origin: it would be considered here a very poor Dragon; the body was of the size of the rock-pigeon, with the beak a little longer, namely, 1 inch from the tip to the feathered base. The skin round the eyes was only slightly wattled, whilst that over the nostrils was fairly wattled. The Hon. C. Murray, also, sent me two Carriers direct from Persia; these had nearly the same character as the Madras bird, being about as large as the rock-pigeon, but the beak in one specimen was as much as 1.15 in length; the skin over the nostrils was only moderately, and that round the eyes scarcely at all wattled.]

SUB-RACE 2/III. BAGADOTTEN-TAUBEN OF NEUMEISTER (PAVDOTTEN- OR HOCKER- TAUBEN).

[I owe to the kindness of Mr. Baily, jun., a dead specimen of this singular breed imported from Germany. It is certainly allied to the Runts; nevertheless, from its close affinity with Carriers, it will be convenient here to describe it. The beak is long, and is hooked or bowed downwards in a highly remarkable manner, as will be seen in figure 24.D. when I treat of the skeleton. The eyes are surrounded by a wide space of bright red skin, which, as well as that over the nostrils, is moderately wattled. The breast-bone is remarkably protuberant, being abruptly bowed outwards. The feet and tarsi are of great length, larger than in first-rate English Carriers. The whole bird is of large size, but in proportion to the size of the body the feathers of the wing and tail are short; a wild rock-pigeon, of considerably less size, had tail-feathers 4.6 inches in length, whereas in the large Bagadotten these feathers were scarcely over 4.1 inches in length. Riedel (5/9. 'Die Taubenzucht' Ulm 1824 s. 42.) remarks that it is a very silent bird.]

SUB-RACE 2/IV. BUSSORAH CARRIER.

[Two specimens were sent me by Sir W. Elliot from Madras, one in spirits and the other skinned. The name shows its Persian origin. It is much valued in India, and is considered as a distinct breed from the Bagdad Carrier, which forms my second sub-race. At first I suspected that these two sub- races might have been recently formed by crosses with other breeds, though the estimation in which they are held renders this improbable; but in a Persian treatise (5/10. This treatise was written by Sayzid Mohammed Musari, who died in 1770: I owe to the great kindness of Sir W. Elliot a translation of this curious treatise.), believed to have been written about 100 years ago, the Bagdad and Bussorah breeds are described as distinct. The Bussorah Carrier is of about the same size as the wild rock-pigeon. The shape of the beak, with some little carunculated skin over the nostrils,— the much elongated eyelids,—the broad mouth measured internally,—the narrow head,—the feet proportionally a little longer than in the rock- pigeon,—and the general appearance, all show that this bird is an undoubted Carrier; yet in one specimen the beak was of exactly the same length as in the rock-pigeon. In the other specimen the beak (as well as the opening of the nostrils) was only a very little longer, viz., by .08 of an inch. Although there was a considerable space of bare and slightly carunculated skin round the eyes, that over the nostrils was only in a slight degree rugose. Sir W. Elliot informs me that in the living bird the eye seems remarkably large and prominent, and the same fact is noticed in the Persian treatise; but the bony orbit is barely larger than that in the rock-pigeon.

Amongst the several breeds sent to me from Madras by Sir W. Elliot there is a pair of the Kali Par, black birds with the beak slightly elongated, with the skin over the nostrils rather full, and with a little naked skin round the eyes. This breed seems more closely allied to the Carrier than to any other breed, being nearly intermediate between the Bussorah Carrier and the rock-pigeon.

The names applied in different parts of Europe and in India to the several kinds of Carriers all point to Persia or the surrounding countries as the source of this Race. And it deserves especial notice that, even if we neglect the Kali Par as of doubtful origin, we get a series broken by very small steps, from the rock-pigeon, through the Bussorah, which sometimes has a beak not at all longer than that of the rock-pigeon and with the naked skin round the eyes and over the nostrils very slightly swollen and carunculated, through the Bagdad sub-race and Dragons, to our improved English Carriers, which present so marvellous a difference from the rock- pigeon or Columba livia.]

RACE III. RUNTS. (SCANDEROONS: DIE FLORENTINER TAUBEN AND HINKELTAUBEN OF NEUMEISTER; PIGEON BAGADAIS, PIGEON ROMAIN.)

Beak long, massive; body of great size.

[Inextricable confusion reigns in the classification, affinities, and naming of Runts. Several characters which are generally pretty constant in other pigeons, such as the length of the wings, tail, legs, and neck, and the amount of naked skin round the eyes, are excessively variable in Runts. When the naked skin over the nostrils and round the eyes is considerably developed and wattled, and when the size of body is not very great, Runts graduate in so insensible a manner into Carriers, that the distinction is quite arbitrary. This fact is likewise shown by the names given to them in different parts of Europe. Nevertheless, taking the most distinct forms, at least five sub-races (some of them including well-marked varieties) can be distinguished, which differ in such important points of structure, that they would be considered as good species in a state of nature.]

SUB-RACE 3/I. SCANDEROON OF ENGLISH WRITERS (DIE FLORENTINER AND HINKELTAUBEN OF NEUMEISTER).

[Birds of this sub-race, of which I kept one alive and have since seen two others, differ from the Bagadotten of Neumeister only in not having the beak nearly so much curved downwards, and in the naked skin round the eyes and over the nostrils being hardly at all wattled. Nevertheless I have felt myself compelled to place the Bagadotten in Race II., or that of the Carriers, and the present bird in Race III., or that of the Runts. The Scanderoon has a very short, narrow, and elevated tail; wings extremely short, so that the first primary feathers were not longer than those of a small tumbler pigeon! Neck long, much bowed; breast-bone prominent. Beak long, being 1.15 inch from tip to feathered base; vertically thick; slightly curved downwards. The skin over the nostrils swollen, not wattled; naked skin round the eyes, broad, slightly carunculated. Legs long; feet very large. Skin of neck bright red, often showing a naked medial line, with a naked red patch at the distal end of the radius of the wing. My bird, as measured from the base of the beak to the root of the tail, was fully 2 inches longer than the rock-pigeon; yet the tail itself was only 4 inches in length, whereas in the rock-pigeon, which is a much smaller bird, the tail is 4 5/8 inches in length.

The Hinkel- or Florentiner Taube of Neumeister (Table 13 figure 1) agrees with the above description in all the specified characters (for the beak is not mentioned), except that Neumeister expressly says that the neck is short, whereas in my Scanderoon it was remarkably long and bowed; so that the Hinkel forms a well-marked variety.]

SUB-RACE 3/II. PIGEON CYGNE AND PIGEON BAGADAIS OF BOITARD AND CORBIE (SCANDEROON OF FRENCH WRITERS).

[I kept two of these birds alive, imported from France. They differed from the first sub-race or true Scanderoon in the much greater length of the wing and tail, in the beak not being so long, and in the skin about the head being more carunculated. The skin of the neck is red; but the naked patches on the wings are absent. One of my birds measured 38 1/2 inches from tip to tip of wing. By taking the length of the body as the standard of comparison, the two wings were no less than 5 inches longer than those of the rock-pigeon! The tail was 6 1/4 inches in length, and therefore 2 1/4 inches longer than that of the Scanderoon,—a bird of nearly the same size. The beak is longer, thicker, and broader than in the rock-pigeon, proportionally with the size of body. The eyelids, nostrils, and internal gape of mouth are all proportionally very large, as in Carriers. The foot, from the end of the middle to end of hind toe, was actually 2.85 inches in length, which is an excess of .32 of an inch over the foot of the rock- pigeon, proportionally to the relative size of the two birds.]

SUB-RACE 3/III. SPANISH AND ROMAN RUNTS.

[I am not sure that I am right in placing these Runts in a distinct sub- race; yet, if we take well-characterised birds, there can be no doubt of the propriety of the separation. They are heavy, massive birds, with shorter necks, legs, and beaks than in the foregoing races. The skin over the nostrils is swollen, but not carunculated; the naked skin round the eyes is not very wide, and only slightly carunculated; and I have seen a fine so-called Spanish Runt with hardly any naked skin round the eyes. Of the two varieties to be seen in England, one, which is the rarer, has very long wings and tail, and agrees pretty closely with the last sub-race; the other, with shorter wings and tail, is apparently the Pigeon romain ordinaire of Boitard and Corbie. These Runts are apt to tremble like Fantails. They are bad flyers. A few years ago Mr. Gulliver (5/11. 'Poultry Chronicle' volume 2 page 573.) exhibited a Runt which weighed 1 pound 14 ounces; and, as I am informed by Mr. Tegetmeier, two Runts from the south of France were lately exhibited at the Crystal Palace, each of which weighed 2 pounds 2 1/2 ounces. A very fine rock-pigeon from the Shetland Islands weighed only 14 1/2 ounces.]

SUB-RACE 3/IV. TRONFO OF ALDROVANDI (LEGHORN RUNT?).

[In Aldrovandi's work published in 1600 there is a coarse woodcut of a great Italian pigeon, with an elevated tail, short legs, massive body, and with the beak short and thick. I had imagined that this latter character so abnormal in the group, was merely a false representation from bad drawing; but Moore, in his work published in 1735, says that he possessed a Leghorn Runt of which "the beak was very short for so large a bird." In other respects Moore's bird resembled the first sub-race or Scanderoon, for it had a long bowed neck, long legs, short beak, and elevated tail, and not much wattle about the head. So that Aldrovandi's and Moore's birds must have formed distinct varieties, both of which seem to be now extinct in Europe. Sir W. Elliot, however, informs me that he has seen in Madras a short-beaked Runt imported from Cairo.]

SUB-RACE 3/V. MURASSA (ADORNED PIGEON) OF MADRAS.

[Skins of these handsome chequered birds were sent me from Madras by Sir W. Elliot. They are rather larger than the largest rock-pigeon, with longer and more massive beaks. The skin over the nostrils is rather full and very slightly carunculated, and they have some naked skin round the eyes; feet large. This breed is intermediate between the rock-pigeon and a very poor variety of Runt or Carrier.

From these several descriptions we see that with Runts, as with Carriers, we have a fine gradation from the rock-pigeon (with the Tronfo diverging as a distinct branch) to our largest and most massive Runts. But the chain of affinities, and many points of resemblance, between Runts and carriers, make me believe that these two races have not descended by independent lines from the rock-pigeon, but from some common parent, as represented in the Table, which had already acquired a moderately long beak with slightly swollen skin over the nostrils, and with some slightly carunculated naked skin round the eyes.]

(FIGURE 20. ENGLISH BARB.)

RACE IV. BARBS. (INDISCHE TAUBEN; PIGEONS POLONAIS.)

Beak short, broad, deep; naked skin round the eyes, broad and carunculated; skin over nostrils slightly swollen.

[Misled by the extraordinary shortness and form of the beak, I did not at first perceive the near affinity of this Race to that of Carriers until the fact was pointed out to me by Mr. Brent. Subsequently, after examining the Bussorah Carrier, I saw that no very great amount of modification would be requisite to convert it into a Barb. This view of the affinity of Barbs to Carriers is supported by the analogical difference between the short and long-beaked Runts; and still more strongly by the fact, that, young Barbs and Dragons, within 24 hours after being hatched, resemble each other much more closely than do young pigeons of other and equally distinct breeds. At this early age, the length of beak, the swollen skin over the rather open nostrils, the gape of the mouth, and the size of the feet, are the same in both; although these parts afterwards become widely different. We thus see that embryology (as the comparison of very young animals may perhaps be called) comes into play in the classification of domestic varieties, as with species in a state of nature.

Fanciers, with some truth, compare the head and beak of the Barb to that of a bullfinch. The Barb, if found in a state of nature would certainly have been placed in a new genus formed for its reception. The body is a little larger than that of the rock-pigeon, but the beak is more than .2 of an inch shorter; although shorter, it is both vertically and horizontally thicker. From the outward flexure of the rami of the lower jaw, the mouth internally is very broad, in the proportion of .6 to .4 to that of the rock-pigeon. The whole head is broad. The skin over the nostril is swollen, but not carunculated, except slightly in first-rate birds when old; whilst the naked skin round the eye is broad and much carunculated. It is sometimes so much developed, that a bird belonging to Mr. Harrison Weir could hardly see to pick up food from the ground. The eyelids in one specimen were nearly twice as long as those of the rock-pigeon. The feet are coarse and strong, but proportionally rather shorter than in the rock- pigeon. The plumage is generally dark and uniform. Barbs, in short, may be called short-beaked Carriers, bearing the same relation to Carriers that the Tronfo of Aldrovandi does to the common Runt.]

GROUP III.

This group is artificial, and includes a heterogeneous collection of distinct forms. It may be defined by the beak, in well-characterised specimens of the several races, being shorter than in the rock-pigeon, and by the skin round the eyes not being much developed.

(FIGURE 21. ENGLISH FANTAIL.)

RACE V.-FANTAILS.

SUB-RACE 5/I. EUROPEAN FANTAILS (PFAUENTAUBEN; TREMBLEURS).

Tail expanded, directed upwards, formed of many feathers; oil-gland aborted; body and beak rather short.

[The normal number of tail-feathers in the genus Columba is 12; but Fantails have from only 12 (as has been asserted) up to, according to MM. Boitard and Corbie, 42. I have counted in one of my own birds 33, and at Calcutta Mr. Blyth (5/12. 'Annals and Mag. of Nat. History' volume 19 1847 page 105.) has counted in an IMPERFECT tail 34 feathers. In Madras, as I am informed by Sir W. Elliot, 32 is the standard number; but in England number is much less valued than the position and expansion of the tail. The feathers are arranged in an irregular double row; their permanent fanlike expansion and their upward direction are more remarkable characters than their increased number. The tail is capable of the same movements as in other pigeons, and can be depressed so as to sweep the ground. It arises from a more expanded basis than in other pigeons; and in three skeletons there were one or two extra coccygeal vertebrae. I have examined many specimens of various colours from different countries, and there was no trace of the oil-gland; this is a curious case of abortion. (5/13. This gland occurs in most birds; but Nitzsch (in his 'Pterylographie' 1840 page 55) states that it is absent in two species of Columba, in several species of Psittacus, in some species of Otis, and in most or all birds of the Ostrich family. It can hardly be an accidental coincidence that the two species of Columba, which are destitute of an oil-gland, have an unusual number of tail-feathers, namely 16, and in this respect resemble Fantails.) The neck is thin and bowed backwards. The breast is broad and protuberant. The feet are small. The carriage of the bird is very different from that of other pigeons; in good birds the head touches the tail-feathers, which consequently often become crumpled. They habitually tremble much: and their necks have an extraordinary, apparently convulsive, backward and forward movement. Good birds walk in a singular manner, as if their small feet were stiff. Owing to their large tails, they fly badly on a windy day. The dark- coloured varieties are generally larger than white Fantails.

Although between the best and common Fantails, now existing in England, there is a vast difference in the position and size of the tail, in the carriage of the head and neck, in the convulsive movements of the neck, in the manner of walking, and in the breadth of the breast, the differences so graduate away, that it is impossible to make more than one sub-race. Moore, however, an excellent old authority (5/14. See the two excellent editions published by Mr. J.M. Eaton in 1852 and 1858 entitled 'A Treatise on Fancy Pigeons.') says, that in 1735 there were two sorts of broad-tailed shakers (i.e. Fantails), "one having a neck much longer and more slender than the other;" and I am informed by Mr. B.P. Brent, that there is an existing German Fantail with a thicker and shorter beak.]

SUB-RACE 5/II. JAVA FANTAIL.

[Mr. Swinhoe sent me from Amoy, in China, the skin of a Fantail belonging to a breed known to have been imported from Java. It was coloured in a peculiar manner, unlike any European Fantail; and, for a Fantail, had a remarkably short beak. Although a good bird of the kind, it had only 14 tail-feathers; but Mr. Swinhoe has counted in other birds of this breed from 18 to 24 tail-feathers. From a rough sketch sent to me, it is evident that the tail is not so much expanded or so much upraised as in even second-rate European Fantails. The bird shakes its neck like our Fantails. It had a well-developed oil-gland. Fantails were known in India, as We shall hereafter see, before the year 1600; and we may suspect that in the Java Fantail we see the breed in its earlier and less improved condition.]

(FIGURE 22. AFRICAN OWL.)

RACE VI. TURBIT AND OWL. (MOVENTAUBEN; PIGEONS A CRAVATE.)

Feathers divergent along the front of the neck and breast; beak very short, vertically rather thick; oesophagus somewhat enlarged.

Turbits and Owls differ from each other slightly in the shape of the head; the former have a crest, and the beak is differently curved; but they may be here conveniently grouped together. These pretty birds, some of which are very small, can be recognised at once by the feathers irregularly diverging, like a frill, along the front of the neck, in the same manner, but in a less degree, as along the back of the neck in the Jacobin. They have the remarkable habit of continually and momentarily inflating the upper part of the oesophagus, which causes a movement in the frill. When the oesophagus of a dead bird is inflated, it is seen to be larger than in other breeds, and not so distinctly separated from the crop. The Pouter inflates both its true crop and oesophagus; the Turbit inflates in a much less degree the oesophagus alone. The beak of the Turbit is very short, being .28 of an inch shorter than that of the rock-pigeon, proportionally with the size of their bodies; and in some owls brought by Mr. E. Vernon Harcourt from Tunis, it was even shorter. The beak is vertically thicker, and perhaps a little broader, in proportion to that of the rock-pigeon.