Skull.—The individual bones, especially those at the base, do not differ in shape. But the whole skull, in its proportions, outline, and relative direction of the bones, differs greatly in some of the breeds, as may be seen by comparing the figures of (A) the wild rock-pigeon, (B) the Short-faced Tumbler, (C) the English Carrier, and (D) the Bagadotten Carrier (of Neumeister), all drawn of the natural size and viewed laterally. In the Carrier, besides the elongation of the bones of the face, the space between the orbits is proportionally a little narrower than in the rock-pigeon. In the Bagadotten the upper mandible is remarkably arched, and the premaxillary bones are proportionally broader. In the Short-faced Tumbler the skull is more globular: all the bones of the face are much shortened, and the front of the skull and descending nasal bones are almost perpendicular: the maxillo-jugal arch and premaxillary bones form an almost straight line; the space between the prominent edges of the eye-orbits is depressed. In the Barb the premaxillary bones are much shortened, and their anterior portion is thicker than in the rock-pigeon, as is the lower part of the nasal bone. In two Nuns the ascending branches of the premaxillaries, near their tips, were somewhat attenuated, and in these birds, as well as in some others, for instance in the Spot, the occipital crest over the foramen was considerably more prominent than in the rock-pigeon.

In the lower jaw, the articular surface is proportionably smaller in many breeds than in the rock-pigeon; and the vertical diameter, more especially of the outer part of the articular surface, is considerably shorter. May not this be accounted for by the lessened use of the jaws, owing to nutritious food having been given during a long period to all highly improved pigeons? In Runts, Carriers, and Barbs (and in a lesser degree in several breeds), the whole side of the jaw near the articular end is bent inwards in a highly remarkable manner; and the superior margin of the ramus, beyond the middle, is reflexed in an equally remarkable manner, as may be seen in fig. 25, in comparison with the jaw of the rock-pigeon. This reflection of the upper margin of the lower jaw is plainly connected with the singularly wide gape of the mouth, as has been described in Runts, Carriers, and Barbs. The reflection is well shown in fig. 26 of the head of a Runt seen from above; here a wide open space may be observed on each side, between the edges of the lower jaw and of the premaxillary bones. In the rock-pigeon, and in several domestic breeds, the edges of the lower jaw on each side come close up to the premaxillary bones, so that no open space is left. The degree of downward curvature of the distal half of the lower jaw also differs to an extraordinary degree in some breeds, as may be seen in the drawings (fig. 27 A) of the rock-pigeon, (B) of the Short-faced Tumbler, and (C) of the Bagadotten Carrier of Neumeister. In some Runts the symphysis of the lower jaw is remarkably solid. No one would readily have believed that jaws differing in the several above-specified points so greatly could have belonged to the same species.

Vertebræ.—All the breeds have twelve cervical vertebræ.^[36] But in a Bussorah Carrier from India the twelfth vertebra carried a small rib, a quarter of an inch in length, with a perfect double articulation.

The dorsal vertebræ are always eight. In the rock-pigeon all eight bear ribs; the eighth rib being very thin, and the seventh having no process. In Pouters all the ribs are extremely broad, eight bear ribs; the eighth rib being very thin and the seventh having no process. In Pouters all the ribs are extremely broad, and, in three out of four skeletons examined by me, the eighth rib was twice or even thrice as broad as in the rock-pigeon; and the seventh pair had distinct processes. In many breeds there are only seven ribs, as in seven out of eight skeletons of various Tumblers, and in several skeletons of Fantails, Turbits and Nuns.>

In all these breeds the seventh pair was very small, and was destitute of processes, in which respect it differed from the same rib in the rock-pigeon. In one Tumbler, and in the Bussorah Carrier, even the sixth pair had no process. The hypapophysis of the second dorsal vertebra varies much in development; being sometimes (as in several, but not all Tumblers) nearly as prominent as that of the third dorsal vertebra; and the two hypapophyses together tend to form an ossified arch. The development of the arch, formed by the hypapophyses of the third and fourth dorsal vertebræ, also varies considerably, as does the size of the hypapophysis of the fifth vertebra.

The rock-pigeon has twelve sacral vertebræ; but these vary in number, relative size, and distinctness, in the different breeds. In Pouters, with their elongated bodies, there are thirteen or even fourteen, and, as we shall immediately see, an additional number of caudal vertebræ. In Runts and Carriers there is generally the proper number, namely twelve; but in one Runt, and in the Bussorah Carrier, there were only eleven. In Tumblers there are either eleven, or twelve, or thirteen sacral vertebræ.

The caudal vertebræ are seven in number in the rock-pigeon. In Fantails, which have their tails so largely developed, there are eight or nine, and apparently in one case ten, and they are a little longer than in the rock-pigeon, and their shape varies considerably. Pouters, also, have eight or nine caudal vertebræ. I have seen eight in a Nun and Jacobin. Tumblers, though such small birds, always have the normal number seven; as have Carriers, with one exception, in which there were only six.

The following table will serve as a summary, and will show the most remarkable deviations in the number of the vertebra and ribs which I have observed:—

	Rock Pigeon.	Pouter, from Mr. Bult.	Tumbler, Dutch Roller.	Bussorah Carrier.
Cervical Vertebræ	12	12	12	12 The 12th bore a small rib.
Dorsal Vertebræ	8	8	8	8
Dorsal Ribs	8 The 6th pair with processes, the 7th pair without a process.	8 The 6th and 7th pair with processes.	7 The 6th and 7th pair without processes.	7 The 6th and 7th pair without processes.
Sacral Vertebræ	12	14	11	11
Caudal Vertebræ	7	8 or 9	7	7
Total Vertebræ	39	42 or 43	38	38

The pelvis differs very little in any breed. The anterior margin of the ilium, however, is sometimes a little more equally rounded on both sides than in the rock-pigeon. The ischium is also frequently rather more elongated. The obturator-notch is sometimes, as in many Tumblers, less developed than in the rock-pigeon. The ridges on the ilium are very prominent in most Runts.

In the bones of the extremities I could detect no difference, except in their proportional lengths; for instance, the metatarsus in a Pouter was 1·65 inch, and in a Short-faced Tumbler only ·95 in length; and this is a greater difference than would naturally follow from their differently-sized bodies; but long legs in the Pouter, and small feet in the Tumbler, are selected points. In some Pouters the scapula is rather straighter, and in some Tumblers it is straighter, with the apex less elongated, than in the rock-pigeon: in fig. 28, the scapula of the rock-pigeon (A), and of a short-faced Tumbler (B), are given. The processes at the summit of the coracoid, which receive the extremities of the furculum, form a more perfect cavity in some Tumblers than in the rock-pigeon: in Pouters these processes are larger and differently shaped, and the exterior angle of the extremity of the coracoid, which is articulated to the sternum, is squarer.

The two arms of the furculum in Pouters diverge less, proportionally to their length, than in the rock-pigeon; and the symphysis is more solid and pointed. In Fantails the degree of divergence of the two arms varies in a remarkable manner. In fig. 29, B and C represent the furcula of two Fantails; and it will be seen that the divergence in B is rather less even than in the furculum of the short-faced, small-sized Tumbler (A), whereas the divergence in C equals that in a rock-pigeon, or in the Pouter (D), though the latter is a much larger bird. The extremities of the furculum, where articulated to the coracoids, vary considerably in outline.

In the sternum the differences in form are slight, except in the size and outline of the perforations, which, both in the larger and lesser sized breeds, are sometimes small. These perforations, also, are sometimes either nearly circular, or elongated as is often the case with Carriers. The posterior perforations occasionally are not complete, being left open posteriorly. The marginal apophyses forming the anterior perforations vary greatly in development. The degree of convexity of the posterior part of the sternum differs much, being sometimes almost perfectly flat. The manubrium is rather more prominent in some individuals than in others, and the pore immediately under it varies greatly in size.

Correlation of Growth.—By this term I mean that the whole organisation is so connected, that when one part varies, other parts vary; but which of two correlated variations ought to be looked at as the cause and which as the effect, or whether both result from some common cause, we can seldom or never tell. The point of interest for us is that, when fanciers, by the continued selection of slight variations, have largely modified one part, they often unintentionally produce other modifications. For instance, the beak is readily acted on by selection, and, with its increased or diminished length, the tongue increases or diminishes, but not in due proportion; for, in a Barb and Short-faced Tumbler, both of which have very short beaks, the tongue, taking the rock-pigeon as the standard of comparison, was proportionally not shortened enough, whilst in two Carriers and in a Runt the tongue, proportionally with the beak, was not lengthened enough, thus, in a first-rate English Carrier, in which the beak from the tip to the feathered base was exactly thrice as long as in a first-rate Short-faced Tumbler, the tongue was only a little more than twice as long. But the tongue varies in length independently of the beak: thus in a Carrier with a beak 1·2 inch in length, the tongue was ·67 in length: whilst in a Runt which equalled the Carrier in length of body and in stretch of wings from tip to tip, the beak was ·92 whilst the tongue was ·73 of an inch in length, so that the tongue was actually longer than in the carrier with its long beak. The tongue of the Runt was also very broad at the root. Of two Runts, one had its beak longer by ·23 of an inch, whilst its tongue was shorter by ·14 than in the other.

With the increased or diminished length of the beak the length of the slit forming the external orifice of the nostrils varies, but not in due proportion, for, taking the rock-pigeon as the standard, the orifice in a Short-faced Tumbler was not shortened in due proportion with its very short beak. On the other hand (and this could not have been anticipated), the orifice in three English Carriers, in the Bagadotten Carrier, and in a Runt (pigeon cygne), was longer by above the tenth of an inch than would follow from the length of the beak proportionally with that of the rock-pigeon. In one Carrier the orifice of the nostrils was thrice as long as in the rock-pigeon, though in body and length of beak this bird was not nearly double the size of the rock-pigeon. This greatly increased length of the orifice of the nostrils seems to stand partly in correlation with the enlargement of the wattled skin on the upper mandible and over the nostrils; and this is a character which is selected by fanciers. So again, the broad, naked, and wattled skin round the eyes of Carriers and Barbs is a selected character; and in obvious correlation with this, the eyelids, measured longitudinally, are proportionally more than double the length of those of the rock-pigeon.

The great difference (see fig. 27) in the curvature of the lower jaw in the rock-pigeon, the Tumbler, and Bagadotten Carrier, stands in obvious relation to the curvature of the upper jaw, and more especially to the angle formed by the maxillo-jugal arch with the premaxillary bones. But in Carriers, Runts, and Barbs the singular reflexion of the upper margin of the middle part of the lower jaw (see fig. 25) is not strictly correlated with the width or divergence (as may be clearly seen in fig. 26) of the premaxillary bones, but with the breadth of the horny and soft parts of the upper mandible, which are always overlapped by the edges of the lower mandible.

In Pouters, the elongation of the body is a selected character, and the ribs, as we have seen, have generally become very broad, with the seventh pair furnished with processes; the sacral and caudal vertebræ have been augmented in number; the sternum has likewise increased in length (but not in the depth of the crest) by ·4 of an inch more than would follow from the greater bulk of the body in comparison with that of the rock-pigeon. In Fantails, the length and number of the caudal vertebræ have increased. Hence, during the gradual progress of variation and selection, the internal bony framework and the external shape of the body have been, to a certain extent, modified in a correlated manner.

Although the wings and tail often vary in length independently of each other, it is scarcely possible to doubt that they generally tend to become elongated or shortened in correlation. This is well seen in Jacobins, and still more plainly in Runts, some varieties of which have their wings and tail of great length, whilst others have both very short. With Jacobins, the remarkable length of the tail and wing-feathers is not a character which is intentionally selected by fanciers; but fanciers have been trying for centuries, at least since the year 1600, to increase the length of the reversed feathers on the neck, so that the hood may more completely enclose the head; and it may be suspected that the increased length of the wing and tail-feathers stand in correlation with the increased length of the neck-feathers. Short-faced Tumblers have short wings in nearly due proportion with the reduced size of their bodies; but it is remarkable, seeing that the number of the primary wing-feathers is a constant character in most birds, that these Tumblers generally have only nine instead of ten primaries. I have myself observed this in eight birds; and the Original Columbarian Society^[37] reduced the standard for Bald-head Tumblers from ten to nine white flight-feathers, thinking it unfair that a bird which had only nine feathers should be disqualified for a prize because it had not ten white flight-feathers. On the other hand, in Carriers and Runts, which have large bodies and long wings, eleven primary feathers have occasionally been observed.

Mr. Tegetmeier has informed me of a curious and inexplicable case of correlation, namely, that young pigeons of all breeds which when mature become white, yellow, silver (i.e., extremely pale blue), or dun-coloured, are born almost naked; whereas pigeons of other colours are born well-clothed with down. Mr. Esquilant, however, has observed that young dun Carriers are not so bare as young dun Barbs and Tumblers. Mr. Tegetmeier has seen two young birds in the same nest, produced from differently coloured parents, which differed greatly in the degree to which they were at first clothed with down.

I have observed another case of correlation which at first sight appears quite inexplicable, but on which, as we shall see in a future chapter, some light can be thrown by the law of homologous parts varying in the same manner. The case is, that, when the feet are much feathered, the roots of the feathers are connected by a web of skin, and apparently in correlation with this the two outer toes become connected for a considerable space by skin. I have observed this in very many specimens of Pouters, Trumpeters, Swallows, Roller-tumblers (likewise observed in this breed by Mr. Brent), and in a lesser degree in other feather-footed pigeons.

The feet of the smaller and larger breeds are of course much smaller or larger than those of the rock-pigeon; but the scutellæ or scales covering the toes and tarsi have not only decreased or increased in size, but likewise in number. To give a single instance, I have counted eight scutellæ on the hind toe of a Runt, and only five on that of a Short-faced Tumbler. With birds in a state of nature the number of the scutellæ on the feet is usually a constant character. The length of the feet and the length of the beak apparently stand in correlation; but as disuse apparently has affected the size of the feet, this case may come under the following discussion.

On the Effects of Disuse.—In the following discussion on the relative proportions of the feet, sternum, furculum, scapulæ, and wings, I may premise, in order to give some confidence to the reader, that all my measurements were made in the same manner, and that they were made without the least intention of applying them to the following purpose.

Table I.
Pigeons with their beaks generally shorter than that of the Rock-pigeon, proportionally to the size of their bodies.

Name of Breed.	Actual length of Feet	Difference between actual and calculated length of feet, in proportion to length of feet and size of body in the Rock-pigeon.
Wild rock-pigeon (mean measurement)	2·02	Too short by	Too long by
Short-faced Tumbler, blad-head	1·57	0·11	—
Short-faced Tumbler, almond	1·60	0·16	—
Tumbler, red magpie	1·75	0·19	—
Tumbler, red common (by standard to end of tail)	1·85	0·07	—
Tumbler, common bald-head	1·85	0·18	—
Tumbler, roller	1·80	0·06	—
Turbit	1·75	0·17	—
Turbit	1·80	0·01	—
Turbit	1·84	0·15	—
Jacobin	1·90	0·02	—
Trumpeter, white	2·02	0·06	—
Trumpeter, mottled	1·95	0·18	—
Fantail (by standard to end of tail)	1·85	0·15	—
Fantail (by standard to end of tail)	1·95	0·15	—
Fantail crested va. (by standard to end of tail)	1·95	0·0	0·0
Indian Frill-back (by standard to end of tail)	1·80	0·19	—
English Frill-back	2·10	0·03	—
Nun	1·82	0·02	—
Laugher	1·65	0·16	—
Barb	2·00	0·03	—
Barb	2·00	—	0·03
Spot	1·90	0·02	—
Spot	1·90	0·07	—
Swallow, red	1·85	0·18	—
Swallow, blue	2·00	—	0·03
Pouter	2·42	—	0·11
Pouter, German	2·30	—	0·09
Bussorah Carrier	2·17	—	0·09
Number of specimens	28	22	5

I measured most of the birds which came into my possession, from the feathered base of the beak (the length of beak itself being so variable) to the end of the tail, and to the oil-gland, but unfortunately (except in a few cases) not to the root of the tail; I measured each bird from the extreme tip to tip of wing; and the length of the terminal folded part of the wing, from the extremity of the primaries to the joint of the radius. I measured the feet without the claws, from the end of the middle toe to the end of the hind toe; and the tarsus and middle toe together. I have taken in every case the mean measurement of two wild rock-pigeons from the Shetland Islands, as the standard of comparison. The following table shows the actual length of the feet in each bird; and the difference between the length which the feet ought to have had according to the size of body of each, in comparison with the size of body and length of feet of the rock-pigeon, calculated (with a few specified exceptions) by the standard of the length of the body from the base of the beak to the oil-gland. I have preferred this standard, owing to the variability of the length of tail. But I have made similar calculations, taking as the standard the length from tip to tip of wing, and likewise in most cases from the base of the beak to the end of the tail; and the result has always been closely similar. To give an example: the first bird in the table, being a Short-faced Tumbler, is much smaller than the rock-pigeon, and would naturally have shorter feet; but it is found on calculation to have feet too short by ·11 of an inch, in comparison with the feet of the rock-pigeon, relatively to the size of the body in these two birds, as measured from the base of beak to the oil-gland. So again, when this same Tumbler and the rock-pigeon were compared by the length of their wings, or by the extreme length of their bodies, the feet of the Tumbler were likewise found to be too short in very nearly the same proportion. I am well aware that the measurements pretend to greater accuracy than is possible, but it was less trouble to write down the actual measurements given by the compasses in each case than an approximation.

Table II.
Pigeons with their beaks longer than that of the Rock-pigeon, proportionally to the size of their bodies.

Name of Breed.	Actual length of Feet	Difference between actual and calculated length of feet, in proportion to length of feet and size of body in the Rock-pigeon.
Wild rock-pigeon (mean measurement)	2·02	Too short by	Too long by
Short-faced Tumbler, bald-head	1·57	0·11	—
Carrier	2·60	—	0·31
Carrier	2·60	—	0·25
Carrier	2·40	—	0·21
Carrier Dragon	2·25	—	0·06
Bagadotten Carrier	2·80	—	0·56
Scanderoon, white	2·80	—	0·37
Scanderoon, Pigeon cygne	2·85	—	0·29
Runt	2·75	—	0·27
Number of specimens	8	—	8

In these two tables (Tables I and II) we see in the first column the actual length of the feet in thirty-six birds belonging to various breeds, and in the two other columns we see by how much the feet are too short or too long, according to the size of bird, in comparison with the rock-pigeon. In the first table twenty-two specimens have their feet too short, on an average by a little above the tenth of an inch (viz. ·107); and five specimens have their feet on an average a very little too long, namely, by ·07 of an inch. But some of these latter cases can be explained; for instance, with Pouters the legs and feet are selected for length, and thus any natural tendency to a diminution in the length of the feet will have been counteracted. In the Swallow and Barb, when the calculation was made on any standard of comparison besides the one used (viz. length of body from base of beak to oil-gland), the feet were found to be too small.

In the second table we have eight birds, with their beaks much longer than in the rock-pigeon, both actually and proportionally with the size of body, and their feet are in an equally marked manner longer, namely, in proportion, on an average by ·29 of an inch. I should here state that in Table I there are a few partial exceptions to the beak being proportionally shorter than in the rock-pigeon: thus the beak of the English Frill-back is just perceptibly longer, and that of the Bussorah Carrier of the same length or slightly longer, than in the rock-pigeon. The beaks of Spots, Swallows, and Laughers are only a very little shorter, or of the same proportional length, but slenderer. Nevertheless, these two tables, taken conjointly, indicate pretty plainly some kind of correlation between the length of the beak and the size of the feet. Breeders of cattle and horses believe that there is an analogous connection between the length of the limbs and head; they assert that a race-horse with the head of a dray-horse, or a grey-hound with the head of a bulldog, would be a monstrous production. As fancy pigeons are generally kept in small aviaries, and are abundantly supplied with food, they must walk about much less than the wild rock-pigeon; and it may be admitted as highly probable that the reduction in the size of the feet in the twenty-two birds in the first table has been caused by disuse,^[38] and that this reduction has acted by correlation on the beaks of the great majority of the birds in Table I. When, on the other hand, the beak has been much elongated by the continued selection of successive slight increments of length, the feet by correlation have likewise become much elongated in comparison with those of the wild rock-pigeon, notwithstanding their lessened use.

As I had taken measures from the end of the middle toe to the heel of the tarsus in the rock-pigeon and in the above thirty-six birds, I have made calculations analogous with those above given, and the result is the same— namely, that in the short-beaked breeds, with equally few exceptions as in the former case, the middle toe conjointly with the tarsus has decreased in length; whereas in the long-beaked breeds it has increased in length, though not quite so uniformly as in the former case, for the leg, in some varieties of the Runt varies much in length.

As fancy pigeons are generally confined in aviaries of moderate size, and as even when not confined they do not search for their own food, they must during many generations have used their wings incomparably less than the wild rock-pigeon. Hence it seemed to me probable that all the parts of the skeleton subservient to flight would be found to be reduced in size. With respect to the sternum, I have carefully measured its extreme length in twelve birds of different breeds, and in two wild rock-pigeons from the Shetland Islands. For the proportional comparison I have tried three standards of measurement, with all twelve birds namely, the length from the base of the beak to the oil-gland, to the end of the tail, and from the extreme tip to tip of wings. The result has been in each case nearly the same, the sternum being invariably found to be shorter than in the wild rock-pigeon. I will give only a single table, as calculated by the standard from the base of the beak to the oil-gland; for the result in this case is nearly the mean between the results obtained by the two other standards.

Length of Sternum.

Name of Breed	Actual Length. Inches	Too short by
Wild Rock-pigeon	2·55	—
Wild Rock-pigeon	2·55	—
Pied Scanderoon	2·80	0·60
Bagadotten Carrier	2·80	0·17
Dragon	2·45	0·41
Carrier	2·75	0·35
Short-faced Tumbler	2·05	0·28
Barb	2·35	0·34
Nun	2·27	0·15
German Pouter	2·36	0·54
Jacobin	2·33	0·22
English Frill-back	2·40	0·43
Swallow	2·45	0·17

This table shows that in these twelve breeds the sternum is of an average one-third of an inch (exactly ·332) shorter than in the rock-pigeon, proportionally with the size of their bodies; so that the sternum has been reduced by between one-seventh and one-eighth of its entire length; and this is a considerable reduction.

I have also measured in twenty-one birds, including the above dozen, the prominence of the crest of the sternum relatively to its length, independently of the size of the body. In two of the twenty-one birds the crest was prominent in the same relative degree as in the rock-pigeon; in seven it was more prominent; but in five out of these seven, namely, in a Fantail, two Scanderoons, and two English Carriers, this greater prominence may to a certain extent be explained, as a prominent breast is admired and selected by fanciers; in the remaining twelve birds the prominence was less. Hence it follows that the crest exhibits a slight, though uncertain, tendency to be reduced in prominence in a greater degree than does the length of the sternum relatively to the size of body, in comparison with the rock-pigeon.

I have measured the length of the scapula in nine different large and small-sized breeds, and in all the scapula is proportionally shorter (taking the same standard as before) than in the wild rock-pigeon. The reduction in length on an average is very nearly one-fifth of an inch, or about one-ninth of the length of the scapula in the rock-pigeon.

The arms of the furcula in all the specimens which I compared, diverged less, proportionally with the size of body, than in the rock-pigeon; and the whole furculum was proportionally shorter. Thus in a Runt, which measured from tip to tip of wings 38½ inches, the furculum was only a very little longer (with the arms hardly more divergent) than in a rock-pigeon which measured from tip to tip 26½ inches. In a Barb, which in all its measurements was a little larger than the same rock-pigeon, the furculum was a quarter of an inch shorter. In a Pouter, the furculum had not been lengthened proportionally with the increased length of the body. In a Short-faced Tumbler, which measured from tip to tip of wings 24 inches, therefore only 2½ inches less than the rock-pigeon, the furculum was barely two-thirds of the length of that of the rock-pigeon.

We thus clearly see that the sternum, scapula, and furculum are all reduced in proportional length; but when we turn to the wings we find what at first appears a wholly different and unexpected result. I may here remark that I have not picked out specimens, but have used every measurement made by me. Taking the length from the base of beak to the end of the tail as the standard of comparison, I find that, out of thirty-five birds of various breeds, twenty-five have wings of greater, and ten have them of less proportional length, than in the rock-pigeon. But from the frequently correlated length of the tail and wing-feathers, it is better to take as the standard of comparison the length from the base of the beak to the oil-gland; and by this standard, out of twenty-six of the same birds which had been thus measured, twenty-one had wings too long, and only five had them too short. In the twenty-one birds the wings exceeded in length those of the rock-pigeon, on an average, by 1-1/3 inch; whilst in the five birds they were less in length by only ·8 of an inch. As I was much surprised that the wings of closely confined birds should thus so frequently have been increased in length, it occurred to me that it might be solely due to the greater length of the wing-feathers; for this certainly is the case with the Jacobin, which has wings of unusual length. As in almost every case I had measured the folded wings, I subtracted the length of this terminal part from that of the expanded wings, and thus I obtained, with a moderate degree of accuracy, the length of the wings from the ends of the two radii, answering from wrist to wrist in our arms. The wings, thus measured in the same twenty-five birds, now gave a widely different result; for they were proportionally with those of the rock-pigeon too short in seventeen birds, and in only eight too long. Of these eight birds, five were long-beaked,^[39] and this fact perhaps indicates that there is some correlation of the length of the beak with the length of the bones of the wings, in the same manner as with that of the feet and tarsi. The shortening of the humerus and radius in the seventeen birds may probably be attributed to disuse, as in the case of the scapula and furculum to which the wing-bones are attached;—the lengthening of the wing-feathers, and consequently the expansion of the wings from tip to tip, being, on the other hand, as completely independent of use and disuse as is the growth of the hair or wool on our long-haired dogs or long-woolled sheep.

To sum up: we may confidently admit that the length of the sternum, and frequently the prominence of its crest, the length of the scapula and furculum, have all been reduced in size in comparison with the same parts in the rock-pigeon. And I presume that this may be attributed to disuse or lessened exercise. The wings, as measured from the ends of the radii, have likewise been generally reduced in length; but, owing to the increased growth of the wing-feathers, the wings, from tip to tip, are commonly longer than in the rock-pigeon. The feet, as well as the tarsi conjointly with the middle toe, have likewise in most cases become reduced; and this it is probable has been caused by their lessened use; but the existence of some sort of correlation between the feet and beak is shown more plainly than the effects of disuse. We have also some faint indication of a similar correlation between the main bones of the wing and the beak.

Summary on the Points of Difference between the several Domestic Races, and between the individual Birds.—The beak, together with the bones of the face, differ remarkably in length, breadth, shape, and curvature. The skull differs in shape, and greatly in the angle formed by the union of the pre-maxillary, nasal, and maxillo-jugal bones. The curvature of the lower jaw and the reflection of its upper margin, as well as the gape of the mouth, differ in a highly remarkable manner. The tongue varies much in length, both independently and in correlation with the length of the beak. The development of the naked, wattled skin over the nostrils and round the eyes varies in an extreme degree. The eyelids and the external orifices of the nostrils vary in length, and are to a certain extent correlated with the degree of development of the wattle. The size and form of the œsophagus and crop, and their capacity for inflation, differ immensely. The length of the neck varies. With the varying shape of the body, the breadth and number of the ribs, the presence of processes, the number of the sacral vertebræ, and the length of the sternum, all vary. The number and size of the coccygeal vertebræ vary, apparently in correlation with the increased size of the tail. The size and shape of the perforations in the sternum, and the size and divergence of the arms of the furculum, differ. The oil-gland varies in development, and is sometimes quite aborted. The direction and length of certain feathers have been much modified, as in the hood of the Jacobin and the frill of the Turbit. The wing and tail-feathers generally vary in length together, but sometimes independently of each other and of the size of the body. The number and position of the tail-feather vary to an unparalleled degree. The primary and secondary wing-feathers occasionally vary in number, apparently in correlation with the length of the wing. The length of the leg and the size of the feet, and, in connection with the latter, the number of the scutellæ, all vary. A web of skin sometimes connects the bases of the two inner toes, and almost invariably the two outer toes when the feet are feathered.

The size of the body differs greatly: a Runt has been known to weigh more than five times as much as a Short-faced Tumbler. The eggs differ in size and shape. According to Parmentier,^[40] some races use much straw in building their nests, and others use little; but I cannot hear of any recent corroboration of this statement. The length of time required for hatching the eggs is uniform in all the breeds. The period at which the characteristic plumage of some breeds is acquired, and at which certain changes of colour supervene, differs. The degree to which the young birds are clothed with down when first hatched is different, and is correlated in a singular manner with the colour of the plumage. The manner of flight, and certain inherited movements, such as clapping the wings, tumbling either in the air or on the ground, and the manner of courting the female, present the most singular differences. In disposition the several races differ. Some races are very silent; others coo in a highly peculiar manner.

Although many different races have kept true in character during several centuries, as we shall hereafter more fully see, yet there is far more individual variability in the most constant breeds than in birds in a state of nature. There is hardly any exception to the rule that those characters vary most which are now most valued and attended to by fanciers, and which consequently are now being improved by continued selection. This is indirectly admitted by fanciers when they complain that it is much more difficult to breed high fancy pigeons up to the proper standard of excellence than the so-called toy pigeons, which differ from each other merely in colour; for particular colours when once acquired are not liable to continued improvement or augmentation. Some characters become attached, from quite unknown causes, more strongly to the male than to the female sex; so that we have in certain races, a tendency towards the appearance of secondary sexual characters,^[41] of which the aboriginal rock-pigeon displays not a trace.

REFERENCES

[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.

[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.

[3] ‘Les Pigeons de Volière et de Colombier’ Paris 1824. During forty-five years the sole occupation of M. Corbié 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.

[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.

[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.

	Inches
From feathered base of beak to end of tail:	14·25
From feathered base of beak to oil-gland:	9·50
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
Thickness, measured vertically at distal end of nostrils:	·23
Breadth, measured at same place:	·16
Feet:
From end of middle toe (without claw) to distal end of tibia:	2·77
From end of middle toe to end of hind toe (without claws):	2·02
Weight: 14-1/4 ounces.

[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.

[7] ‘Das Ganze der Taubenzucht:’ Weimar, 1837, pl. 11 and 12.

[8] Boitard and Corbié, ‘Les Pigeons,’ etc., p. 177, pl. 6.

[9] ‘Die Taubenzucht,’ Ulm, 1824, s. 42.

[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.

[11] ‘Poultry Chronicle,’ vol. 2, p. 573.

[12] ‘Annals and Mag. of Nat. History,’ vol. xix, 1847, p. 105.

[13] This gland occurs in most birds; but Nitzsch (in his ‘Pterylographie,’ 1840, p. 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.

[14] See the two excellent editions published by Mr. J. M. Eaton in 1852 and 1858, entitled ‘A Treatise on Fancy Pigeons.’

[15] English translation, by F. Gladwin, 4th edition, vol. i. The habit of the Lotan is also described in the Persian treatise before alluded to, published about 100 years ago: at this date the Lotans were generally white and crested as at present. Mr. Blyth describes these birds in ‘Annals and Mag. of Nat. Hist.,’ vol. xiv., 1847, p. 104; he says that they “may be seen at any of the Calcutta bird-dealers.”

[16] ‘Journal of Horticulture,’ Oct. 22, 1861, p. 76.

[17] See the account of the House-tumblers kept at Glasgow, in the ‘Cottage Gardener,’ 1858, p. 285. Also Mr. Brent’s paper, ‘Journal of Horticulture,’ 1861, p. 76.

[18] J. M. Eaton, ‘Treatise on Pigeons,’ 1852, p. 9.

[19] J. M. Eaton, ‘Treatise,’ edit. 1858, p. 76.

[20] Neumeister, ‘Taubenzucht,’ tab. 4. fig. i.

[21] Riedel, ‘Die Taubenzucht,’ 1824, s. 26. Bechstein, ‘Naturgeschichte Deutschlands,’ Band iv. s. 36, 1795.

[22] Willughby’s ‘Ornithology,’ edited by Ray.

[23] J. M. Eaton’s edition (1858) of Moore, p. 98.

[24] Pigeon pattu plongeur. ‘Les Pigeons,’ etc., p. 165.

[25] ‘Naturgeschichte Deutschlands,’ Band iv. s. 47.

[26] Mr. W. B. Tegetmeier, ‘Journal of Horticulture,’ Jan. 20, 1863, p. 58.

[27] ‘Coup-d’œil sur L’Ordre des Pigeons,’ par C. L. Bonaparte (‘Comptes Rendus’), 1854-55. Mr. Blyth, in ‘Annals of Nat. Hist.,’ vol. xix., 1847, p. 41, mentions, as a very singular fact, “that of the two species of Ectopistes, which are nearly allied to each other, one should have fourteen tail-feathers, while the other, the passenger pigeon of North America, should possess but the usual number—twelve.”

[28] Described and figured in the ‘Poultry Chronicle,’ vol. iii., 1855, p. 82.

[29] ‘The Pigeon Book,’ by Mr. B. P. Brent, 1859, p. 41.

[30] ‘Die staarhälsige Taube. Das Ganze, etc.,’ s. 21, tab. i. fig. 4.

[31] ‘A Treatise on the Almond-Tumbler,’ by J. M. Eaton, 1852, p. 8, et passim.

[32] ‘A Treatise, etc.,’ p. 10.

[33] Boitard and Corbié ‘Les Pigeons,’ etc., 1824, p. 173.

[34] ‘Le Pigeon Voyageur Belge,’ 1865, p. 87. I have given in my ‘Descent of Man’ (6th edit. p. 466) some curious cases, on the authority of Mr. Tegetmeier, of silver-coloured (i.e. very pale blue) birds being generally females, and of the ease with which a race thus characterised could be produced. Bonizzi (see ‘Variazioni dei Columbi domestici:’ Padova, 1873) states that certain coloured spots are often different in the two sexes, and the certain tints are commoner in females than in male pigeons.

[35] Prof. A. Newton (‘Proc. Zoolog. Soc.,’ 1865, p. 716) remarks that he knows no species which present any remarkable sexual distinction; but Mr. Wallace informs me, that in the sub-family of the Treronidæ the sexes often differ considerably in colour. See also on sexual differences in the Columbidæ, Gould, ‘Handbook to the Birds of Australia,’ vol. ii., pp. 109-149.

[36] I am not sure that I have designated the different kinds of vertebræ correctly: but I observe that different anatomists follow in this respect different rules, and, as I use the same terms in the comparison of all the skeletons, this, I hope, will not signify.

[37] J. M. Eaton’s ‘Treatise,’ edit. 1858, p. 78.

[38] In an analogous, but converse, manner, certain natural groups of the Columbidæ, from being more terrestrial in their habits than other allied groups, have larger feet. See Prince Bonaparte’s ‘Coup d’œil sur l’Ordre des Pigeons.’

[39] It perhaps deserves notice that besides these five birds two of the eight were Barbs, which, as I have shown, must be classed in the same group with the long-beaked Carriers and Runts. Barbs may properly be called short-beaked Carriers. It would, therefore, appear as if, during the reduction of their beaks, their wings had retained a little of that excess of length which is characteristic of their nearest relations and progenitors.

[40] Temminck, ‘Hist. Nat. Gén. des Pigeons et des Gallinacés,’ tom. i., 1813, p. 170.

[41] This term was used by John Hunter for such differences in structure between the males and females, as are not directly connected with the act of reproduction, as the tail of the peacock, the horns of deer, etc.