Bible Encyclopedias
Feather

1911 Encyclopedia Britannica

Search for…

Enter query below:

or

A B C D E F G H I J K L M N O P Q R S T U V W Y Z

Prev Entry
Feasts and Festivals

Next Entry
Featherstone

Resource Toolbox

Print Article

Additional Links

Dictionaries
Wilson's Bible Types
• Feathers
Webster Dictionary
• Sea Feather
• Featherness
• Feathering (2)
•   (only first 3 shown)
Encyclopedias
1911 Encyclopedia Britannica
• Feather
• Tarring and Feathering
International Standard Bible Encyclopedia
• Feathers
McClintock and Strong's Bible Encyclopedia
• Feather
• Feathering, or Foliation
• Feathers Tavern Association
Lexicons
• חֲסִידָה
• כָּנָף
• נוצָה נצָה, , נצָה

(0. Eng. fether, Ger. Feder, from an IndoEuropean root seen also in Gr. irmepov, and IIT€aOaU, to fly), a horny outgrowth of the skin of birds homologous with the scale of the reptile. The body-covering of birds is, without exception, comprised of feathers, and by this character alone birds may be distinguished from all other animals.

The most perfect form of feather is made up of a long, tapering rod, fringed on either side, for the greater part of its length, by a secondary series of slender and tapering rods forming a more or less acute angle with the central axis. This fringe is known as the vexillum or " vane " (fig. i a). The central axis is divisible into two distinct parts, - a hollow, cylindrical, transparent calamus, or " quill," the base of which is inserted into the skin, and a solid, quadrangular rhachis or " shaft " which supports the vane. At the lower end of the quill is a small hole - the lower umbilicus - through which the nutritive pulp passes during the growth of the feather: while at the upper end, where it passes into the shaft, a similar hole will be found, - the upper umbilicus - and from this the last remains of the capsules which contained the nutritive pulp may sometimes be seen protruding. If the quill is cut open a series of these capsules will be found fitting one into the other throughout the whole length of the tubular chamber.

FIG.. - Diagrams of a, Outline of a feather showing d, the relation of the barbs and barbules to the central axis e, or shaft.

b, Section across two of the barbs shown --in a, highly magnified.

c, Two barbules of the posterior series - seen only in crosssection in b. The rods comprising the lateral fringe, or vane, are known as the rami or the " barbs," and will be found, on microscopic examination, to be lath-shaped and to taper to a point. Further, each barb supports a double series of smaller outgrowths known as the radii, or "barbules "; so that each barb may be likened to a feather in miniature. These " barbules," however, differ markedly in structure on the two sides of the barb, those pointing towards the tip of the feather - the " anterior barbules " - being ribbon-shaped from the base outwards for about half their length, when they become cut up to form a series of long and very delicate hooklets (fig. I'd). On the opposite side of the barb the barbules are also ribbon-shaped for about half their length, but the ribbon is curved trough-fashion, so that the whole series of posterior barbules forms a number of deep valleys, and into these the 'hooklets are thrust so as to catch hold of the upper edges of the troughs, which are set so that the f, Feather-Barbs.

A barbule of the anterior series.

Section across the base of three anterior barbules showing attachment to barb.

A portion of the hooklet of the anterior series showing the method of interlocking with the barbules of the posterior series.

upper edge is towards the upper, and the lower edge towards the under surface of the feather. The manner in which this beautiful mechanism works may be seen in fig. i b. In one of the primary or " quill " feathers of the wing of a crane, each barb of the inner side of the vane was found to bear about 600 pairs of barbules, which would make about 800,000 barbules for the inner web of the vane alone, or more than a million for the whole feather (H. F. Gadow). It is to the agency of these hooklets alone that the closely-knit, elastic vanes of the flight feathers and, the body feathers are due. Where these hooklets are wanting the barbs do not adhere together, resulting in a loose " discontinuous " vane such as, for example, is found in the plumes of the ostrich.

Many feathers, in addition to the main axis, bear a second, generally much shorter axis, supporting a loose discontinuous vane; this shorter branch is known as the "aftershaft " and arises from the under surface of the feather. Only in the cassowary and emu among adult birds is the aftershaft as large as the main shaft.

There are several different kinds of feathers - contour feathers, semiplumes, down-feathers, filoplumes and powder-down. Contour feathers, as their name implies, are those which form the contour or outline of the body, and are all that can generally be seen. Those which form the " flight feathers " of the wing, and the tail feathers, are the most perfectly developed. Semiplumes are degenerate contour feathers. The down-feathers are generally completely hidden by the contour feathers: they form in many birds, such as gulls and ducks, a thick underclothing comparable to the under-fur of mammals such as the seals. In all cases they are of a loose, soft, " fluffy " structure, the barbs being of great length and slenderness, while the barbules are often long and provided with knob-like thickenings answering to the hooklets of the more perfectly developed contour feathers; these thickenings help to " felt " the separate down-feathers together, the barbs of one down-feather interlocking with those of its neighbour. Down-feathers differ from semiplumes both in their relation to contour feathers and in that they do not possess a main axis, all the barbs arising from a common centre.

Filoplumes are degenerate structures having a superficial resemblance to hairs, but they always bear a minute vane at the tip. They occur in all birds, in clusters of varying number, about the bases of contour feathers. In some birds they attain a great length, and may project beyond the contour feathers, sometimes forming conspicuous white patches, as for example in the necks of cormorants. In their early stages of development they often possess a large aftershaft made up of a number of barbs, but these quickly disappear, leaving only the degenerate main shaft. The eyelashes and bristles round the mouth found in many birds appear to be akin to filoplumes.

Powder-down feathers are degenerate down-feathers which appear to secrete a dry, waxy kind of powder. This powder rapidly disintegrates and becomes distributed over the plumage, adding thereto a quite peculiar bloom. In birds of the heron tribe powder-down feathers have reached a high degree of development, forming large patches in the breast and thighs, while in some hawks, and in the parrots, these mysterious feathers are scattered singly over the greater part of the body.

The nature of the covering of nestling birds is of a more complex character than has hitherto been suspected. The majority of young birds, as is well known, either down . emerge from the egg clothed in down-feathers, or they develop these within a day or two afterwards. But this covering, though superficially similar in all, may, as a matter of fact, differ widely in its constitution, even in closely related forms, while only in a very few species can the complete history of these feathers be made out.

The brown or tawny owl (Syrnium aluco) is one of these. At hatching, the young of this species is thickly clad in white, woolly down-feathers, of the character known as umbelliformthat is to say, the central axis or main shaft is wanting, so that the barbs all start from a common centre. These feathers occupy the position of the ultimate contour feathers. They are shortly replaced by a second down-like covering, superficially resembling, and generally regarded as, true down. But they differ in that their barbs spring from a central axis as in typical contour feathers. Feathers of this last description indeed have now made their appearance in the shape of the " flight " or quill feathers (remiges) and of the tail feathers. This plumage is worn until the autumn, when the downy feathers give place to the characteristic adult plumage. The down feathers which appear at hatching-time are known as pre-pennae, or pre-plumulae, as the case may be; the first generation of pre-pennae, in the case of the tawny owl for example, is made up of protoptyles, while the succeeding plumage is made up of mesoptyles, and these in turn give place to the teleoptyles or adult feathers. The two forms of nestling plumage - pre-pennae and pre-plumulaemay be collectively called " neossoptyles," a term coined by H. F. Gadow to distinguish the plumage of the nestling from that of the adult - the " teleoptyle " plumage.

As a rule the nestling develops but one of these generations of neossoptyles, and this generally answers to the mesoptyle plumage, though this is of a degenerate type. In some birds, as in the Megapodes, the " protoptyle " or first of these two generations of pre-pennae is developed and shed while the chick is yet in the shell, so that at hatching the mesoptyle plumage is well developed. But in the majority of birds, probably, the mesoptyle plumage only is developed, while the earlier, and apparently more degenerate, dress is suppressed. In the penguins both of these nestling plumages are developed, but the mesoptyle dress has degenerated so that umbelliform feathers now take the place of feathers having a central axis.

The Anatidae show traces of the earlier, first generation of feathers in one or two species only, e.g. Cloephaga rubidiceps. In all the remaining species mesoptyles only occur. And this is true also of the game-birds. In both the Tinamous, the ducktribe and the game-birds this mesoptyle plumage shows, in different species, every gradation between feathers having a well-developed main shaft and aftershaft, and those which are mere umbelliform tufts.

As development proceeds and the contour feathers make their appearance they thrust the mesoptyle feathers out of their follicles - the pockets in the skin in which they were rooted - and these will often be found adhering to the tips of the contour feathers for many weeks after the bird has left the nest. This occurs because the development of the contour feather begins before that of the mesoptyles has completed.

The plumage in nestling birds is still futher complicated by the fact that it may be almost, or entirely, composed of pre-plumulae; that is to say, of down-feathers which are later succeeded by adult down feathers. This is the case among the accipitrine birds for example, and thereby it differs entirely from that of the owls, which develop neither pre-plumulae nor adult down. The cormorants are, so far as is known, the only birds which have a nestling plumage composed entirely of pre-plumulae.

In variety and brilliancy the colours of birds are not surpassed by those of any other group of animals. Yet the pigments to which these colours are due are but few in number, while a large number of the most resplendent hues are produced by structural peculiarities of the feathers. colourless horny surface of the feathers, and hence are known as subjective or optical colours.

The principal colour pigments are (a) melanin pigments, derived possibly from the haemoglobin of the blood, but more probably from the blood plasma, and (b) lipochrome or " fat " pigments, which are regarded as reserve products; though in the case of birds it is exceedingly doubtful whether they have this significance.

The melanin pigments (zoomelanin) occur in the form of granules and give rise to the black, brown and grey tones; or they may combine with those of the lipochrome series.

The lipochrome pigments (zoonerythrin and zooxanthin) tend to be diffused throughout the substance of the feather, and give rise respectively to the red and yellow colours.

In addition to these must be reckoned turacin, a reddishpurple pigment consisting of the same elements as zoomelanin, but remarkable for the fact that it contains from 5 to 8% of copper, which can be extracted by a weak alkaline solution, such as ammonia, and with the addition of acetic acid it can be filtered off as a metallic red or blue powder. The presence of metallic copper is indicated by the green flame of these red feathers when burnt. Turacin was discovered by Sir A. H. Church in the quill-feathers of the wings of Touracoes or " plantain eaters." These feathers, he showed, lose their colour after they have become wet, but regain it on drying. But turacin is not, as was supposed, confined to the feathers of the plantain eaters, since it has been obtained from a cuckoo, Dasylophus superciliosus. What effect food may have on colour in birds in a wild state we have no means of knowing, but it is significant that flamingoes and linnets in confinement never regain their bright hues after their first moult in captivity. If cayenne pepper be mixed with the food of certain strains of canaries, from the time the birds are hatched onwards, the yellow colour of the feathers becomes intensified, till it takes on a deep orange hue. Bullfinches, if fed on hemp-seed, turn black. According to Darwin, the natives of the Amazonian region feed the common green parrot on the fat of large Siluroid fishes, and as a result the feathers become beautifully variegated with red and yellow. Similarly, in the Malay Archipelago, the natives of Gilolo change the colours of another parrot.

With but rare exceptions bright colours are confined to the exposed portions of the plumage, but in some of the Bustards the down is of a bright pink colour.

Structural colours include all metallic or prismatic colours, blue, green, white, some yellows, and, in part, glossy black. In metallic feathers the radii (barbules) are modified Structural in various was frequently to form flattened, colours. ways, c l Y ? over lapping plates or tiles, while the surfaces of the plates are either smooth, finely striated or pitted. But, save only in the case of white feathers, beneath this colourless, glazed outer coat there is always a layer of pigment.

The only green pigment known to occur in feathers is turacoverdin, found in the feathers of the plantain eaters; it contains a relatively large amount of iron, but no copper. In all other cases the green colour of feathers is due to yellow, orange or greyish-brown pigment occurring with a special superstructure consisting of narrow ridges, as in some parrots and pittas (antthrushes), or the surface of the barbs and barbules is smooth and transparent, while between it and the pigment there exists a layer of small polygonal, colourless bodies having highly refractory, and often striated, surfaces.

Blue is unknown as a pigment in feathers. Blue feathers contain only orange or brownish pigment (Gadow), the blue colour being caused by the combination of pigment corpuscles and colourless striated polygonal bodies, as in green feathers.

While in many birds the coloration takes the form either of a uniform hue or of bands and patches of colour more or less brilliant, in others the coloration is sombre, and made up of dark longitudinal stripes or transverse bars on a lighter ground. The latter is the more primitive, and there seems good reason to believe that longitudinal stripes preceded transverse bars. This is indicated by the fact that the nestlings of the more primitive groups are longitudinally striped, and that young hawks in their first plumage are so striped, while the adults are barred.

There is also evidence to show that the evolution of brilliant plumage began with the males, and has, in many cases, been more or less perfectly acquired by the females, and also by the young, as for example in the kingfishers, where parents and offspring wear the same livery. Often, where the parents are alike in plumage, the young wear a different and duller livery, as in the case of the common starling (Sturnus vulgaris). But where the female differs from the male in coloration the young resemble the female parent.

The physiological explanation of complete disappearance of pigment in adult life, e.g. gannet, is not yet apparent.

At least once annually birds renew their feathers completely by a process known as a moult. Until the new feathers have attained at least half their full length they are invested moulting, in a soft sheath, and, as development proceeds, the sheath breaks up from the tip of the feather downwards, so that for a time the new feathers have almost a brush-like appearance. Generally this replacement takes place gradually, new and old feathers occurring side by side, and on this account it is not always possible to see whether a moult is proceeding without raising the old feathers.

The " quill " feathers of the wing and tail are renewed in pairs, so that flight is little, if at all, impaired, the change taking place in the wing from the region of the wrist inwards, as to the primaries, and from the body outwards, towards the tip of the wing, as to the secondaries. In certain birds, however, as in the duck tribe and the rails, for example, all the quill-feathers of the wing are shed at once, so that for some time flight is impossible.

In the penguins this simultaneous method of moulting is carried still further. That is to say, the old feathers covering the body are not replaced gradually, but en masse. This method of ecdysis is, however, still further remarkable in that the old feathers do not drop out, to be succeeded by spine-like stumps which, later, split at the tip, liberating the barbs of the new feathers. They are, on the contrary, thrust out upon the tips of the new feathers, the barbs of which are never enclosed within an envelope such as that just described. When their growth has practically completed, and not till then, the old feathers are removed in large patches by the aid of the bird's beak; exposing thereby a perfectly developed plumage. In the cassowary, and emeu, the old feathers similarly adhere for a time to the tips of the new; but in these birds the feathers are moulted singly as in other birds.

Some birds moult twice within the year, the additional moult taking place in the spring, as in the case of the " warblers " (Sylviidae) and Limicolae, for example. But when this is the case the spring moult is only partial, since the quill feathers of the wings and the tail feathers are not renewed.

At this spring moult a special " nuptial " plumage is often assumed, as for example in many of the Limicolae, e.g. godwits, knots, dunlin, ruff.

The sequel to this habit of assuming a nuptial dress is an interesting one. Briefly, this plumage, at first assumed at the mating period by the males only, and doffed soon after the young appear, has become retained for longer and longer periods, so that the succeeding plumage, often conspicuously dull compared with the nuptial dress, is worn only for a few weeks, instead of many months, as in the case of many of the ducks, for example; wherein the males, as soon as the young are hatched, assume what C. Waterton has aptly called an " eclipse " dress. This, instead of being worn till the following spring, as in the waders, is shed again in the autumn and replaced by what answers to the waders' " nuptial " dress. In the game-birds but a trace of this " eclipse " plumage remains; and this, apparently, only in jungle-fowl, the common grey partridge (Perdix cinerea) and the blackcock (Lyrurus), in whose case the head and neck for a short period following the breeding season are clothed only by dull feathers. Further, this more highly developed plumage becomes transferred, first to the female, then to the young, so that, in many groups, the dull phase of plumage is entirely eliminated.

But the assumption at the breeding season of a conspicuously brilliant plumage is not always due to a moult. In many birds, notably many Passerines, this change is brought about by shedding the tips of the feathers, which are of a duller hue than the rest of the feather. In this way the bright rose pink of the linnet's breast, the blue and black head of the chaffinch, and the black throat and chestnut-and-black markings of the back of the sparrow, are assumed - to mention but a few instances. These birds moult but once a year, in the autumn, when the new feathers have broad brown fringes; as the spring advances these drop off, and with them the barbicels from the barbules of the upper surface of the feather, thus revealing the hidden tints.

According to some authorities, however, some birds acquire a change of colour without a moult by the ascent of pigment from the base of the feather. The black head assumed by many gulls in the spring is, for example, said to be gained in this way. There is, however, not only no good evidence in support of the contention, but the whole structure of the feather is against the probability of any such change taking place.

Feathers correspond with the scales of reptiles rather than with the hairs of mammals, as is shown by their development. They make their first appearance in the developing chick at about the sixth day of incubation, in the shape of small papillae. In section each papilla is found to be made up of a cluster of dermal cells - that is to say, of cells of the deeper layer of the skin - capped by cells of the epidermis. These last form a single superficial layer of flattened cells - the epitrichium- overlaying the cells of the Malpighian layer, which are cylindrical in shape and rapidly increase to form several layers. As development proceeds the papillae assume a cone-shape with its apex directed backwards, while the base of this cone sinks down into the skin, or rather is carried down by the growth of the Malpighian cells, so that the cone is now sunk in a deep pit. Thereby these Malpighian cells become divided into two portions: (I) those taking part in the formation of the walls of the pit or " feather follicle," and (2) those enclosed within the cone. These last surround the central mass or core formed by the dermis. This mass constitutes the nutritive pulp for the development of the growing feather, and is highly vascular. The cells of the Malpighian layer within the cone now become differentiated into three layers. (i) An inner, extremely thin, forming a delicate sheath for the pulp, and found in the fully developed feather in the form of a series of hollow, transparent caps enclosed within the calamus; (2) a thick layer which forms the feather itself; and (3) a thin layer which forms the investing sheath of the feather. It is this sheath which gives the curious spinecovered character to many nestling birds and birds in moult. As growth proceeds the cells of this middle layer arrange themselves in longitudinal rows to form the barbs, while the barbules are formed by a secondary splitting. At their bases these rudimentary barbs meet to form the calamus. Finally the tips of the barbs break through the investing sheath and the fully formed down-feather emerges.

A part of the pulp and Malpighian cells remains over after the complete growth of the down-feather, and from this succeeding generations of feathers are developed. The process of this development differs from that just outlined chiefly in this: that of the longitudinal rows which in the down-feather form the barbs, two on the dorsal and two on the ventral aspect of the interior of the cylinder become stronger than the rest, combining to form the mainand after-shaft respectively. The remainder of the rods form the barbs and barbules as in the down-feather.

The reproductive power of the feather follicle appears to be almost inexhaustible, since it is not diminished appreciably by age, nor restricted to definite moulting periods, as is shown by the cruel and now obsolete custom of plucking geese alive, no less than three times annually, for the sake of their feathers. The growth of the feathers is, however, certainly affected by the general health of the bird, mal-nutrition causing the appearance of peculiar transverse Vshaped grooves, at more or less regular intervals, along the whole length of the feather. These are known as " hunger-marks," a name given by falconers, to whom this defect was well known.

It would seem that while the feather germ may be artificially stimulated to produce three successive generations of feathers within a year, it may, on the other hand, be induced artificially to maintain a continuous activity extending over long periods. That is to say, the normal quiescent period, and periodic moult, may be suspended, so that the feather maintains a steady and continuous growth till it attains a length of several feet. The only known instance of this kind is that furnished by a domesticated breed of jungle-fowl known as the " Japanese long-tailed fowls " or as " Yokohamas." In this breed the upper tail coverts are in some way, as yet unknown to Europeans, induced to go on growing until they have attained a length of from 12 to 18 or even 20 ft. ! In this abnormal growth the " hackles " of the lower part of the back also share, though they do not attain a similar length.

The feathers of birds are not uniformly distributed over the body, but grow only along certain definite tracts known as pterylae, leaving bare spaces or apteria. These pterylae differ considerably in their conformation in different groups of birds, and hence are of service in systematic ornithology.

The principal pterylae are as follows: (i) The head tract (pt. capitis), which embraces the head only.

(2) The spinal tract (pi. spinalis), which extends the whole length of the vertical column. It is one of the most variable in its modifications, especially in so far as the region from the base of the neck to the tail is concerned. In its simplest form it runs down the back in the form of a band of almost uniform width, but generally it expands considerably in the lumbar region, as in Passeres. Frequently it is divided into two portions; an upper, terminating in the region of the middle of the back in a fork, and a lower, which commences either as a fork, e.g. plover, barbet, or as a median band, e.g. swallow. Very commonly the dorsal region of this tract encloses a more or less extensive featherless space (apterion), e.g. swift, auk. While, as a rule, the dorsal region of this tract is relatively narrow, it is in some of great breadth, e.g. grebe, pigeon, coly.

(3) The ventral tract (pt. ventralis), which presents almost as many variations as the spinal tract.

In its simplest form it runs from the throat backwards in the form of a median band as far as the base of the neck where it divides, sending a branch to each side of the breast. This branch commonly again divides into a short, broad outer branch which lodges the " flank " feathers, and a long, narrow, inner branch which runs backwards to join its fellow of the opposite side in front of the cloacal aperture. This branch lodges the abdominal feathers. The median space which divides the inner branches of the tract may be continued forwards as far as the middle of the neck, or even up to the throat, e.g. plover. Only in a few cases is the neck continuously covered by the fusion of the dorsal and ventral tracts, e.g. flamingo, Anseres, Ciconidae, Pygopodes.

For convenience sake the cervical portions of the spinal and ventral tracts are generally regarded as separate tracts, the pt. colli dorsalis and pt. colli ventralis respectively.

(4) The humeral tract (pt. humeralis), which gives rise to the " scapular " feathers.

(5) The femoral tract (pt. femoralis), which forms an oblique band across the thigh.

(6) The crural tract (pt. cruralis), which clothes the rest of the leg.

(7) The tail tract (pt. caudalis), including the tail feathers and their coverts; and (8) The wing tract (pt. alaris). The wing tract presents many peculiar features. Each segment - arm, forearm and hand - bears feathers essential to flight, and these are divided into remiges, or " quill " feathers, and tectrices, or " coverts." The remiges of the arm, more commonly described as " tertiaries," are, technically, collectively known as the parapteron and hypopteron, and are composed respectively of long, quill-like feathers forming a double series, the former arranged along the upper, and the latter along the lower aspect of the humerus. They serve to fill up the gap which, in long-winged birds, would otherwise occur during flight between the quill-feathers of the forearm and the body, a gap which would make flight impossible. In short-winged birds these two series are extremely reduced.

The remiges range in number from 16, as in humming-birds, to 48 as in the albatross, according, in short, to the length of the wing. But these numerical differences depend, in flying birds, rather upon the length of the forearm, since the quills of the hand never exceed 12 and never fall below 10, though the tenth may be reduced to a mere vestige.

The quills of the forearm are known as " secondaries," those of the hand as " primaries." The former are attached by their bases at relatively wide distances apart to the ulna, while the primaries are crowded close together and attached to the skeleton of the hand. The six or seven which rest upon the fused metacarpals 11. - III. are known as " metacarpals." The next succeeding feather is borne by the phalanx of digit III. and hence is known as the addigital. Phalanx i. of digit II. always supports two quills, the "middigitals," while the remaining feathers - one or two - are borne by the last phalanx of digit II. and are known as pre-digitals, while the whole series of primaries are known as the metacarpo-digitals.

In their relation one to another the remiges, it must be noted, are always so placed that they overlap one another, the free edge of each, when the wing is seen from its upper surface, being turned towards the tip of the wing. Thus, in flight, the air passes through the wing as it is raised, while in the downstroke the feathers are forced together to form a homogeneous surface.

Birds which fly much have the outer primaries of great length, giving the wing a pointed shape, as in swifts, while in species which fly but little, or frequent thickets, the outer primaries are very short, giving the wing a rounded appearance. This adaptation to environment is commonly lost sight of by taxonomers, who not infrequently use the form of the wing as a factor in classification.

The tectrices, or covert feathers of the wing, are arranged in several series, decreasing in size from behind forwards. The number of rows on the dorsal aspect and the method of their overlap, afford characters of general importance in classification.

FIG. 2. - Pterylosis of the plover.

The first row of the series is formed by the major coverts; these, like the primaries, have their free-edges directed towards the tip of the wing, and hence are said to have a distal overlap. The next row is formed by the median coverts. These, on the forearm, commonly overlap as to the outer half of the row distally, and as to the inner half proximally. On the hand this series is incomplete. Beyond the median are four or five rows of coverts known as the minor coverts. These may have either a proximal or a distal overlap. The remaining rows of small feathers are known as the marginal coverts, and they always have a distal overlap.

The three or four large quill-like feathers borne by the thumb form what is known as the "bastard-wing," ala spuria. The coverts of the under follow an Arrangement similar to that of the upper surface, but the minor coverts are commonly but feebly developed, leaving a more or less bare space which is covered by the great elongation of the marginal series.

One noteworthy fact about the coverts of the under side of the wing is that all save the major and median coverts have what answers to the dorsal surfaces of the feather turned towards the body, and what answers to the ventral surface of the feather turned towards the under surface of the wing. In the major and median coverts, however, the ventral surfaces of these feathers are turned ventralwards, that is to say, in the extended wing they, like the remiges, have the ventral surfaces turned downwards or towards the body in the closed wing.

But the most remarkable fact in connexion with the pterylosis of the wing is the fact that in all, save the Passerine and Galliform types, and some few other isolated exceptions, the secondary series of remiges appears always to lack the fifth remex, counting from the wrist inwards, inasmuch as, when such wings are examined, there is always found, in the place of the fifth remex, a pair of major coverts only, while throughout the rest of the series each such pair of coverts embraces a quill.

This extraordinary fact was first discovered by the French naturalist Z. Gerbe, and was later rediscovered by R. S. Wray. Neither of these, however, was able to offer any explanation thereof. This, however, has since been attempted, simultaneously, by P. C. Mitchell and W. P. Pycraft. The former has aptly coined the word diastataxic to denote the gap in the series, and eutaxic to denote such wings as have an uninterrupted series of quills. While both authors agree that there is no evidence of any loss in the number of the quills in diastataxic wings, they differ in the interpretation as to which of the two conditions is the more primitive and the means by which the gap has been brought about.

According to Mitchell the diastataxic is the more primitive condition, and he has conclusively shown a way in which diastataxic wings may become eutaxic. Pycraft on the other hand contends that the diastataxic wing has been derived from the eutaxic type, and has produced evidence showing, on the one hand, the method by which this transition is effected, and on the other that by which the diastataxic wing may again recover the eutaxic condition, though in this last particular the evidence adduced by Mitchell is much more complete. The matter is, however, one of considerable difficulty, but is well worth further investigation.

The wings of struthious birds differ from those of the Carinatae, just described, in many ways. All are degenerate and quite useless as organs of flight. In some cases indeed they have become reduced to mere vestiges.

Those of the ostrich and Rhea are the least degraded.

In the ostrich ankylosis has prevented the flexion of the hand at the wrist joint so that the quills - primaries and secondaries - form an unbroken series of about forty in number. Of these sixteen belong to the primary or metacarpo-digital series, a number exceeding that of any other bird. What the significance of this may be with regard to the primitive wing it is impossible to say at present. The coverts, in their disposition, bear a general resemblance to those of Carinate wings; but they differ on account of the great length of the feathers and the absence of any definite overlap.

The wing of the South American Rhea more nearly resembles that of flying birds since the hand can be flexed at the wrist joint, and the primaries are twelve in number, as in grebes, and some storks, for example.

The coverts, as in the African ostrich, are remarkable for their great length, those representing the major series being as long as the remiges, a fact probably due to the shortening of the latter. They are not, however, arranged in quincunx, as is the rule among the Carinatae, but in parallel, transverse rows, in which respect they resemble the owls.

In both ostrich and Rhea, as well as in all the other struthious birds, the under surface of the wing is entirely bare.

The wing of the cassowary, emeu and apteryx has undergone complete degeneration; so much so that only a vestige of the hand remains.

Remiges in the cassowary are represented by a few spine-like shafts - three primaries and two secondaries. These are really hypertrophied calami. This is shown by the fact that in the nestling these remiges have a normal calamus, rhachis and vane; but as development proceeds the rhachis with its vane sloughs off, while the calamus becomes enormously lengthened and solid.

In the emeu the wing is less atrophied than in the cassowary, but is not yet completely degenerate. Altogether seventeen remiges are represented, of which seven correspond to primaries. Since, however, these feathers have each an aftershaft as long as the main shaft - like the rest of the body feathers - it may be that they answer not to remiges, but to major coverts.

The wing of apteryx, like that of the cassowary, has become extremely reduced. The remiges are thirteen in number, four of which answer to primaries. These feathers are specially interesting, inasmuch as they retain throughout life a stage corresponding to that seen in the very young cassowary, the calamus being greatly swollen, and supporting a very degenerate rhachis and vane.

The penguins afford another object-lesson in degeneration of this kind. Here the wing has become transformed into a paddle, clothed on both sides with a covering of small, close-set feathers. A pollex is wanting, as in the cassowary, emeu and apteryx, while it is impossible to say whether remiges are represented or not.