1911 Encyclopædia Britannica/Bird

by Hans Friedrich Gadow

BIRD, the common English name for feathered vertebrates, members of the class Aves. The word in Old Eng. is brid and in Mid. Eng. byrd or bryd, and in early uses meant the young or nestlings only. It is partly due to this early meaning that the derivation from the root of “brood” has been usually accepted; this the New English Dictionary regards as “inadmissible.” The word does not occur in any other Teutonic language. As a generic name for the feathered vertebrates “bird” has replaced the older “fowl,” a common Teutonic word, appearing in German as Vogel. “Bird,” when it passed from its earliest meaning of “nestlings,” seems to have been applied to the smaller, and “fowl” to the larger species, a distinction which was retained by Johnson. In modern usage “fowl,” except in “wild-fowl” or “water-fowl,” is confined to domestic poultry.

The scope of the anatomical part of the following article is a general account of the structure of birds (Aves) in so far as they, as a class, differ from other vertebrates, notably reptiles and mammals, whilst features especially characteristic, peculiar or unique, have been dwelt upon at greater length so far as space permitted. References to original papers indicate further sources of information. For a comprehensive account the reader may be referred to Prof. M. Fürbringer’s enormous work Untersuchungen zur Morphologie und Systematik der Vögel, 4to., 2 vols. (1888); H. G. Bronn’s Klassen und Ordnungen des Thierreichs, vol. vi., “Aves,” Leipzig, completed 1893 by Gadow; and A. Newton’s Dictionary of Birds, London, 1896. For the history of the classification of birds see the article Ornithology, where also the more important ornithological works are mentioned. Egg, Feather (including Moult), Migration, &c., also form separate articles to which reference should be made. In this article (a) the general anatomy of birds is discussed, (b) fossil birds, (c) the geographical distribution of birds, (d) the latest classification of birds.

A. Anatomy of Birds

1. Skeleton.

Skull.—When W. K. Parker wrote the account of the skull in the article Birds for the 9th edition of the Encyclopaedia Britannica, he had still to wrestle with the general problem of the composition and evolution of the skull. That chapter of comparative anatomy (together with other anatomical details, for which see the separate articles) is now dealt with in the article Skull; here only the most avine features are alluded to, and since some of Parker’s original illustrations have been retained, the description has been shortened considerably.

One general feature of the adult bird’s skull is the almost complete disappearance of the sutures between the bones of the cranium proper, whilst another is the great movability of the whole palatal and other suspensorial apparatus. The occipital condyle (fig. 1) is a single knob, being formed almost wholly by the basioccipital, while the lateral occipitals (often perversely called exoccipitals) take but little share in it. Part of the membranous roof between the supra-occipital and parietal bones frequently remains unossified and presents in the macerated skull a pair of fontanelles. The squamosals form the posterior outer margin of the orbits and are frequently continued into two lateral downward processes across the temporal fossa. One of these, the processus orbitatis posterior, often combines with an outgrowth of the alisphenoid, and may be, e.g. in cockatoos, continued forwards to the lacrymal bone, so as to form a complete infraorbital bridge. The posterior, so-called processus Zygomaticus is very variable; in many Galli it encloses a foramen by distally joining the orbital process. The ethmoid frequently appears on the dorsal surface between the frontals. There are three periotic bones (pro-, epi-, opisth-otic). The proötic encloses between it and the lateral occipital the fenestra ovalis, into which fits the columella of the ear. The epiotic is often small, ossifies irregularly, and fuses with the supra-occipital. The opisthotic lies between the epiotic and the lateral occipital with which it ultimately fuses; in some birds, e.g. in Larus, it extends far enough to help to bound the foramen magnum. The basisphenoids are ventrally overlaid, and later on fused with, a pair of membrane bones, the basi-temporals, homologous in part with the parasphenoid of lower vertebrates. They contribute to the formation of the auditory meatus, and of the right and left carotid canals which accompany the eustachian tubes. In many birds the basisphenoids send out a pair of basipterygoid processes by which they articulate with the pterygoids. Dorso-laterally the basisphenoid is joined by the alisphenoid, which forms most of the posterior wall of the orbit. The orbito-sphenoids diverge only posteriorly, otherwise they are practically unpaired and form the median interorbital septum, which is very large in correlation with the extraordinary size of the eyeballs.

Fig. 1.—End view of skull of a Chicken three weeks old. Here the opisthotic bone appears in the occipital region, as in the adult Chelonian. (After W. K. Parker.) bo, Basi-occipital. bt, Basi-temporal. eo, Opisthotic. f,   Frontal. fm, Foramen magnum. fo, Fontanella. oc, Occipital condyle. op, Opisthotic. p,   Parietal. pf, Post-frontal. sc, Sinus canal in supra-occipital. so, Supra-occipital. sq, Squamosal. 8,   Exit of vagus nerve.

Fig. 2.—Ripe Chick’s head, 11/4 in. long. (After W. K. Parker.)

bo, Basi-occipital.	pa, Palatine.
bt, Basi-temporal.	pf, Post-frontal.
dpx, Dentary process of premaxilla.	pg, Pterygoid.
eo, Opisthotic.	pn, Prenasal cartilage.
eu, Eustachian tube.	ppx, Palatine process of pre-maxillary.
f,    Frontal.	prp, Pterygoid process of sphenoid.
fm, Foramen magnum.	qj, Quadratojugal.
j,    Jugal.	so, Supra-occipital.
l,    Lacrymal.	sq, Squamosal.
mx, Maxilla.	ty, Tympanic cavity.
mxp, Maxillo-palatine process.	v,    Vomer.
oc, Occipital condyle.	8,    Exit of vagus nerve.
	9,    Exit of hypoglossal nerve.

Fig. 3.—Skull of an old Fowl, upper view. (After W. K. Parker.)
eo, Lateral occipital.	npx, Nasal process of premaxillary.
eth, Ethmoid.	p,    Parietal.
f,    Frontal.	pf, Post-frontal.
j,    Jugal.	px, Premaxilla.
l,    Lacrymal.	qj, Quadratojugal.
n,    Nostril.	so, Supra-occipital.
np, Upper process of nasal.	sq, Squamosal.

Prefrontal bones are absent; post-frontals are possibly indicated by a frequently occurring separate centre of ossification in the post-orbital process, to which the frontals always contribute. The lacrymal is always present, and perforated by a glandular duct. Attached to it or the neighbouring frontal is often a supraorbital; infraorbitals occur also, attached to the jugal or downward process of the lacrymal. The nasals were used by A. H. Garrod to distinguish the birds as holorhinal (fig. 2) where the anterior margin of the nasal is concave, and schizorhinal where this posterior border of the outer nares is continued backwards into a slit which extends beyond the frontal processes of the premaxilla. Many birds possess a more or less well developed cross-joint in front of the frontals and lacrymals, perhaps best developed in Anseres and Psittaci. Owing to this joint the whole upper beak can be moved up and down with extra facility, according to the shoving forwards or backwards of the palato-pterygo-quadrate apparatus which moves sledge-like upon the cranial basis. The premaxilla is always unpaired, but each half has three long processes directed backwards; one fuses with the maxillary bone, another helps to form the anterior part of the palate, while the third, together with its fellow, forms the “culmen” and extends backwards to the frontals, or rather to the ethmoid which there crops up on the surface. The maxillaries (fig. 3) have besides others, a maxillo-palatine process directed inwards in a transverse horizontal direction. The palatines are long, always fused anteriorly with the premaxilla, and frequently with the maxillo-palatine processes; posteriorly they slide upon the presphenoidal rostrum, and articulate in most birds with the pterygoids; they form the greater part of the palatal roof and border the choanae or inner nares. Between these, resting vertically upon the rostrum, appears the vomer; very variable in shape and size, often reduced to a mere trace, as in the Galli, or even absent, broken up into a pair of tiny splints in Pici.

The taxonomic importance of the configurations of the palate was first pointed out by J. de Cornay. T. H. Huxley, in 1868, divided the carinate birds into Dromaeo-, Schizo-, Desmo-, and Aegithognathae, an arrangement which for many years had a considerable influence upon classification. However, subsequent additions and corrections have detracted much from its value, especially when it became understood that the above sub-orders are by no means natural groups. Dromaeognathae have a struthious palate, with a broad vomer meeting in front the broad maxillo-palatal plates, while behind it reaches the pterygoids. The only representatives are the Tinamous. Schizognathae, e.g. fowls (fig. 4), pigeons, gulls, plovers, rails and penguins, have the vomer pointed in front while the maxillo-palatines are free, leaving a fissure between the vomer and themselves. The schizognathous formation is doubtless the most primitive, and its representatives form a tolerably natural assembly. Desmognathae (fig. 5) were supposed to have the maxillo-palatines united across the middle line, either directly or by the intermediation of ossifications in the nasal septum. This is a hopeless assembly. Parker and Fürbringer have demonstrated that desmognathism has been produced in half a dozen ways, implying numerous cases of convergence without any nearer relationship than that they are all derived from some schizognathous group or other. The Aegithognathae, meant to comprise the passeres, woodpeckers and swifts, &c., are really schizognathous but with a vomer which is broadly truncated in front.

Fig. 4.—Skull of adult Fowl. This skull is unusually schizognathous, the vomer (v.) being very small, and the maxillo-palatine process (mxp) much aborted. bo, Basi-occipital. bt, Basi-temporal. eo, Lateral occipital. eu, Eustachian tube. ic, Internal carotid. j,    Jugal. l,    Lacrymal. mx, Maxilla. mxp, Maxillo-palatine process. oc, Occipital condyle. pa, Palatine. pf, Post-frontal. pg, Pterygoid. prp, Pterygoid process of sphenoid. px, Premaxilla. q,    Quadrate. qj, Quadratojugal. rbs, Rostrum of basi-sphenoid. so, Supra-occipital. v,    Vomer. 8,    Exit of vagus nerve. 9,    Exit of hypoglossal nerve. (After W. K. Parker.)

Fig. 5.—Skull of a nestling Sparrow-hawk (Accipiter nisus), palatal view. The circular space on each side of the basi-temporal (bt.) is the opening of the anterior tympanic recess. The basi-pterygoids (bpg) are mere knobs, and the common eustachian opening is seen between them. The maxillo-palatine plates (mxp) are dotted to show their spongy character. bt, Basi-temporal. bpg, Basi-pterygoid. eo, Lateral occipital. f,    Frontal. fm, Foramen magnum. j,    Jugal. l,    Lacrymal. mpg, Mesopterygoid process of W. K. Parker. mx, Maxillary. mxp, Maxillo-palatine process. op, Opisthotic. pa, Palatine. pg, Pterygoid. px, Premaxilla. pto, Prootic. q,    Quadrate. qj, Quadratojugal. sn, Nasal septum. so, Supra-occipital. ty, Tymapanic cavity. v,    Vomer. 8,    Exit of vagus nerve. 9,    Exit of hypoglossal nerve. (After W. K. Parker.)

The remainder of the appendicular skeleton (fig. 6) of the head requires little description. The maxillaries are connected with the distal anterior corner of the quadrate by the thin, splint-like jugal and quadratojugal. The quadrate is invariably a conspicuous bone and movably articulating with the cranium and by a special process with the pterygoid. The mandible is composed of several bones as in reptiles. The os articulare bears on its inner side the inner mandibular process which serves for the insertion of part of the digastric muscle or opener of the mouth; another portion of this muscle is attached to the os angulare, which frequently forms a posterior mandibular process. The greater part of the under-jaw is formed by the right and left dentaries, which in all recent birds are fused together in front. Supra-angular and coronoid splint-bones serve for the insertion of part of the temporal or masseter muscle. Additional splints rest on the inner side of the jaw. Like the crocodiles, birds possess a siphonium, i.e. a membranous, or ossified, tube which rises from a pneumatic foramen in the os articulare, on the median side of the articulation, and passes upwards between the quadrate and lateral occipital bone, opening into the cavity of the middle ear.

Fig. 6.—Skull of adult Fowl. Here the temporal fossa is bridged over by the junction of the post-frontal and squamosal processes (pf., sq.). The processes of the mandible (iap, pap) are characteristic of this type, and of the anseres.
a,    Angular of mandible.	pa, Palatine.
ar, Articular.	pap, Posterior angular process of mandible.
bt, Basi-temporal.	pe, Ethmoid.
d,    Dentary.	pf, Post-frontal.
eo, Lateral occipital.	pg, Pterygoid.
eth, Ethmoid.	ps, Pre-sphenoid.
f,    Frontal.	px, Premaxilla.
iap, Interangular process of mandible.	q,    Quadrate.
ios, Interorbital septum.	qj, Quadratojugal.
j,    Jugal.	sa, Supra-angular or coronoid.
l,    Lacrymal.	so, Supra-occipital.
mx, Maxillar.	sq, Squamosal.
n,    Nasal.	ty, Tympanic cavity.
os, Orbito-sphenoid.	v,   Vomer.
p,    Parietal.	1,   Exit of olfactory nerve.

Fig. 7.—Oshyoides of adult Fowl.
c.h, Ceratohyals (confluent).
b.h, The so-called basihyal, answering to the first basibranchial of a fish.
b.br, Basibranchial, or urohyal, answering to the rest of the basibranchial series.
c.br, e.br, together form the thyrohyal, answering to the first cerato- and epi-branchials.

The Hyoid apparatus is, in its detail, subject to many variations in accord with the very diverse uses to which the tongue of birds is put. It consists of (1) the basihyal variously called copula, or corpus linguae, or unpaired middle portion. (2) The urohyal likewise unpaired, rested ventrally on the larynx. (3) The os entoglossum originally paired, but coalescing into an arrow-headed piece, attached to the anterior end of the basihyal and lodged in the tongue proper. It is homologous with the distal ends of the ceratohyals or ventral elements of the hyoidean or second visceral arch. The dorsal or hyomandibular portion of this same arch is transformed into the auditory chain, ending in the fenestra ovalis. (4) A pair of thyrohyals, homologous with the posterior hyoid horns of mammals, i.e. third visceral or first branchial arch. As the most developed pair in birds they are commonly, although wrongly, called the hyoid horns. They articulate upon facets of the hinder outer corners of the basihyal.

The vertebrae are stereospondylous, the centrum or body and the arch being completely fused into one mass, leaving not even a neuro-central suture. The arch alone sends out processes, viz. the spinous process, the anterior and posterior oblique (commonly called pre- and post-zygapophyses), and the transverse processes. The latter articulate with the tuberculum of the corresponding rib, while the capitulum articulates by a knob on the side of the anterior end of the centrum. In the cervical region the ribs are much reduced, fused with their vertebrae and enclosing the transverse canal or foramen. When the vertebrae are free their centra articulate with each other by complicated joints, exhibiting four types. (1) Amphicoelous; each end of the centrum is concave; this, the lowest condition, is embryonic, but was retained in Archaeopteryx and in the thoracic vertebrae of Ichthyornis. (2) Procoelous, concave in front; only in the atlas, for the reception of the occipital condyle. (3) Opisthocoelous, or concave behind, only occasionally found in the thoracic region, e.g. Sphenisci. (4) Heterocoelous (fig. 8) or saddle-shaped; the anterior surface is concave in a transverse, but convex in a vertical direction, which on posterior surface shows the conditions reversed. This is the most perfect arrangement attained by the vertebral column, and is typical of, and restricted to, birds. The intervertebral joints are further complicated by the interposition of a cartilaginous or fibrous pad or ring. This pad varies much; it is morphologically the homologue of the pair of basiventral elements which by their lateral extension give origin to the corresponding ribs. Later those pads fuse with the anterior end of the centrum of the vertebra to which they belong; where the vertebral column is rendered inflexible, the disks are ossified with the centra and all trace of them is lost. Sometimes the pad is reduced to a ventral semi-ring or meniscus; it retains its largest almost original shape and size in the second vertebra, the axis or epistropheus, where it forms a separately ossifying piece which connects, and coössifies with, the odontoid process (the centrum of the atlas) and the centrum of the second vertebra. Sometimes the ventral portions of these pads form paired or unpaired little ossifications, then generally described as intercentra; such are not uncommon on the tail. The atlas is composed of three pieces; a pair of lateral elements (the right and left dorsal arch pieces) joining above the spinal cord, and a ventral piece equivalent to the first basiventral elements, i.e. serially homologous with the intervertebral pads. In the adults the atlas forms a more or less solid ring. A remnant of the chorda dorsalis and its sheath persists as the ligamentum suspensorium between the central portions of the successive vertebrae.

Fig. 8.—A cervical vertebra from the middle of the neck of a Fowl; natural size. a, Side view; b, upper view; c, lower view; pr.z, pre-zygapophyses; pt.z, post-zygapophyses.

In birds we distinguish between the following regions of the axial skeleton. (1) Cervical vertebrae, or those between the skull and the first vertebra which is connected with the sternum by a pair of complete ribs. The last 1 to 5 of these vertebrae have movable ribs which do not reach the sternum, and are called cervico-dorsals. (2) Dorsals, those which begin with the first thoracic rib, and end at the last that is not fused with the ilium. The term “lumbar” vertebrae is inapplicable to birds. (3) Pelvic, all those which are fused with the iliac portion of the pelvis, generally a considerable number. (4) Caudal, those which are not connected with the pelvis. It is to be noted that often no absolute line of demarcation can be drawn in regard to these regions, their definitions being rather convenient than morphological.

In comparison with all other vertebrates the number of neck-vertebrae of the birds is considerably increased; the lowest number, 14 to 15, is that of most Passeres and many other Coraciomorphae; the largest numbers, 20 or 21, are found in the ostrich, 23 in Cygnus olor and 25 in the black swan. Dorsal vertebrae frequently have a ventral outgrowth of the centrum; these hypapophyses may be simple vertical blades, ⊥-shaped, or paired knobs; they serve for the attachment of the thoracic origin of the longus collianticus muscle, reaching their greatest development in Sphenisci and Colymbidae. In many birds some of the thoracic vertebrae are more or less coössified, in most pigeons for instance the 15th to 17th; in most Galli the last cervical and the next three or four thoracics are coalesced, &c. The pelvic vertebrae include of course the sacrum. There are only two or three vertebrae which are equivalent to those of the reptiles; these true sacrals are situated in a level just behind the acetabulum; as a rule between these two primary sacral vertebrae issues the last of the spinal nerves which contributes to the composition of the sciadic plexus. These true sacrals alone are connected with the ilium by processes which are really equivalent to modified ribs; but the pelvis of birds extends considerably farther forwards and backwards, gradually coming into contact with other vertebrae, which in various ways send out connecting transverse processes or buttresses, and thus become pre- and post-sacral vertebrae (fig. 9). The most anterior part of the ilium often overlaps one or more short lumbar ribs and fuses with them, or even a long, complete thoracic rib. Similarly during the growth of the bird the posterior end of the ilium connects itself with the transverse processes of vertebrae which were originally free, thus transforming them from caudals into secondary post-sacrals. Individual, specific and generic variations are frequent.

Fig. 9.—The “sacrum” of a young Fowl; natural size, seen from below. d.l, Dorso-lumbar, s, sacral, c, caudal vertebrae.		Fig. 10.—A side view of the Chick’s sternum.

The last six or seven caudal vertebrae coalesce into the pygostyle, an upright blade which carries the rectrices. Such a pygostyle is absent in Archaeopteryx, Hesperornis, Tinami and Ratitae, but it occurs individually in old specimens of the ostrich and the kiwi. In Ichthyornis it is very small. In all the Neornithes the total number of caudal vertebrae, inclusive of those which coalesce, is reduced to at least 13.

Sternum (figs. 10 and 11).—Characteristic features of the sternum are the following. There is a well-marked processus lateralis anterior (the right and left together equivalent to the mammalian manubrium), which is the product of two or three ribs, the dorsal parts of which reduced ribs remain as cervico-dorsal ribs. Then follows the rib-bearing portion and then the processus lateralis posterior; this also is the product of ribs, consequently the right and left processes together are equivalent to the xiphoid process or xiphisternum of the mammals. The lateral process in most birds sends out an outgrowth, directed out and upwards, overlapping some of the ribs, the processus obliquus. The median and posterior extension of the body of the sternum is a direct outgrowth of the latter, therefore called meta-sternum. The anterior margin of the sternum, between the right and left anterior lateral processes receives in sockets the feet of the coracoids. Between them arises a median crest, which varies much in extent and composition, and is of considerable taxonomic value. It is represented either by a spina interna or by a spina externa, or by both, or they join to form a spina communis which is often very large and sometimes ends in a bifurcation. Eventually, when the right and left feet of the coracoids overlap each other, the anterior sternal spine contains a foramen. The keel, or carina sterni, is formed as a direct cartilaginous outgrowth of the body of the sternum, ossifying from a special centre. This keel is much reduced in the New Zealand parrot, Stringops, less in various flightless rails, in the dodo and solitaire. It is absent in the Ratitae, which from this feature have received their name, but considerable traces of a cartilaginous keel occur in the embryo of the ostrich, showing undeniably that the absence of a keel in the recent bird is not a primitive, fundamental feature. The keel has been lost, and is being lost, at various epochs and by various groups of birds. The swimming Hesperornis (see Odontornithes) was also devoid of such a structure. In many birds the spaces between the meta-sternum and the posterior processes and again the spaces between this and the oblique process are filled up by proceeding ossification and either remain as notches, or as fenestrae, or they are completely abolished so that the breastbone is turned into one solid more or less oblong plate.

Fig. 11.—Sternum of a Chick (Gallus domesticus) three days old, lower view. The cartilage is shaded and dotted, and the bony centers are light and striated.

Shoulder Girdle.—Scapula, coracoid and clavicle, meet to form the foramen triosseum, through which passes the tendon of the supracoracoideus, or subclavius muscle to the tuberculum superius of the humerus. The coracoid is one of the most characteristic bones of the bird’s skeleton. Its upper end forms the acrocoracoid process, against the inner surface of which leans the proximal portion of the clavicle. From the inner side of the neck of the coracoid arises the precoracoidal process, the remnant of the precoracoid. Only in the ostrich this element is almost typically complete, although soon fused at either end with the coracoid. Near the base of the precoracoidal process is a small foramen for the passage of the nervus supracoracoideus. In most birds the feet of the coracoids do not touch each other; in some groups they meet, in others one overlaps the other, the right lying ventrally upon the left. The scapula is sabre-shaped, and extends backwards over the ribs, lying almost parallel to the vertebral column. This is a peculiar character of all birds. The clavicles, when united, as usual, form the furcula; mostly the distal median portion is drawn out into a hypocleidium of various shape. Often it reaches the keel of the sternum, with subsequent syndosmosis or even synostosis, e.g. in the gannet. In birds of various groups the clavicles are more or less degenerated, the reduction beginning at the distal end. This condition occurs in the Ratitae as well as in the well-flying Platyrcecinae amongst parrots.

Fig. 12.—Bones of Fowl’s right wing, adult, nat. size.

h, Humerus.
r, Radius.
u, Ulnar.
r′, u′, Radial and ulnar carpal bones; with the three digits I., II., III.

The fore-limb or wing (fig. 12); highly specialized for flight, which, initiated and made possible mainly by the strong development of quill-feathers, has turned the wing into a unique organ. The humerus with its crests, ridges and processes, presents so many modifications characteristic of the various groups of birds, that its configuration alone is not only of considerable taxonomic value but that almost any genus, excepting, of course, those of Passeres, can be “spotted” by a close examination and comparison of this bone. When the wing is folded the long glenoid surface of the head of the humerus is bordered above by the tuberculum externum or superius, in the middle and below by the tuberculum medium or inferius for the insertion of the coraco-brachialis posterior muscle. From the outer tuberculum extends the large crista superior (insertion of pectoralis major and of deltoideus major muscles). The ventral portion of the neck is formed by the strong crista inferior, on the median side of which is the deep fossa subtrochanterica by which air sacs enter the humerus. On the outer side of the humerus between the head and the crista inferior is a groove lodging one of the coraco-humeral ligaments. The distal end of the humerus ends in a trochlea, with a larger knob for the ulna and a smaller oval knob for the radius. Above this knob is often present an ectepicondylar process whence arise the tendons of the ulnar and radial flexors. The radius is the straighter and more slender of the two forearm bones. Its proximal end forms a shallow cup for articulation with the outer condyle of the humerus; the distal end bears a knob which fits into the radial carpal. The ulna is curved and rather stout; it articulates with both carpal bones; the cubital quills often cause rugosities on its dorsal surface. Of wrist-bones only two remain in the adult bird; the original distal carpals coalesce with the proximal end of the metacarpals. These are reduced, in all birds, to three, but traces of the fourth have been observed in embryos. The first metacarpal is short and fuses throughout its length with the second. This and the third are much longer and fuse together at their upper and distal ends, leaving as a rule a space between the shafts. The pollex and the third finger are as a rule reduced to one phalanx each, while the index still has two. The first and second fingers frequently carry a little claw. The greatest reduction of the hand-skeleton is met with in Dromaeus and in Apteryx, which retain only the index finger. It is of importance for our understanding of the position of the Ratitae in the system, that the wing-skeleton of the ostrich and rhea is an exact repetition of that of typical flying birds; the bones are much more slender, and the muscles are considerably reduced in strength also to a lesser extent in numbers, but the total length of the wing of an ostrich or a rhea is actually and comparatively enormous. Starting with the kiwi and cassowary, people have got into the habit of confounding flightless with wingless conditions. It is absolutely certain that the wings of the Ratitae bear the strongest testimony that they are the descendants of typical flying birds.

The pelvis (fig. 13), consisting of the sacrum (already described) and the pelvic arch, namely ilium, ischium and pubis, it follows that only birds and mammals possess a pelvis proper, whilst such is entirely absent in the Amphibia and in reptiles with the exception of some of the Dinosaurs. The ventral inner margin of the preacetabular portion of the ilium is attached to the pre-sacral vertebrae, whilst the inner and dorsal margin of the postacetabular portion is attached to the primary sacral and the postsacral vertebrae. In rare cases the right and left preacetabular blades fuse with each other above the spinous processes. In front of the acetabulum a thick process of the ilium descends to meet the pubis, and a similar process behind meets the ischium. The acetabulum is completely surrounded by these three bones, but its cup always retains an open foramen; from its posterior rim arises the strong antitrochanter. The ischium and postacetabular ilium originally enclose the ischiadic notch or incisura ischiadica. This primitive condition occurs only in the Odontornithes (q.v.), Ratitae and Tinami; in all others this notch becomes converted into a foramen ischiadicum, through which pass the big stems of the ischiadic nerves and most of the blood-vessels of the hind-limb. The pubis consists of a short anterior portion (spina pubica or pectineal process, homologous with the prepubic process of Dinosaurs) and the long and slender pubis proper (equivalent to the processus lateralis pubis of most reptiles). The shaft of the pubis runs parallel with that of the ischium, with which it is connected by a short ligamentous or bony bridge; this cuts off from the long incisura pubo-ischiadica a proximal portion, the foramen obturatum, for the passage of the obturator nerve. Only in the ostrich the distal ends of the pubes meet, forming a dagger-shaped symphysis, which is curved forwards. The pectineal process is variable; it may grow entirely from the pubis, or both pubis and ilium partake of its formation, or lastly its pubic portion may be lost and the process is entirely formed by the ilium. It is largest in the Galli and some of the Cuculi, in others it is hardly indicated. It served originally for the origin of the ambiens muscle (see Muscular System below); shifting or disappearance of this muscle, of course, influences the process.

Fig. 13.—Pelvis and caudal vertebrae of adult Fowl, side view, natural size.
Il. Ilium; Is, ischium; Pb, pubis; d.l, dorso-lumbar vertebrae; Cd, caudal vertebrae; Am, acetabulum.

The Hind Limb.—The femur often possesses a well visible pneumatic foramen on the median side of the proximal end of its shaft. The inner condyle, the intercondylar sulcus, and a portion only of its outer condyle, articulate with corresponding facets of the tibia. The outer condyle articulates mainly with the fibula. There is a patella, intercalated in the tendon of the femori-tibialis or extensor cruris muscle. In Colymbus the patella is reduced to a small ossicle, its function being taken by the greatly developed pyramidal processus tibialis anterior; in Podiceps and Hesperornis the patella itself is large and pyramidal. The distal half of the fibula is very slender and normally does not reach the ankle-joint; it is attached to the peroneal ridge of the tibia. On the anterior side of the tibia, is the intercondylar sulcus, which is crossed by an oblique bridge of tendon or bone, acting as a pulley for the tendon of the extensor digitorum communis muscle. The condyles of the tibia are in reality not parts of this bone, but are the three proximal tarsalia which fuse together and with the distal end of the tibia. The distal tarsalia likewise fuse together, and then on to the upper ends of the metatarsals; the tarsale centrale remains sometimes as a separate osseous nodule, buried in the inter-articular pad. Consequently the ankle-joint of birds is absolutely cruro-tarsal and tarso-metatarsal, i.e. intertarsal, an arrangement absolutely diagnostic of birds if it did not also occur in some of the Dinosaurs. Of the metatarsals the fifth occurs as an embryonic vestige near the joint; the first is reduced to its distal portion, and is, with the hallux, shoved on to the inner and posterior side of the foot, at least in the majority of birds. The three middle metatarsals become fused together into a cannon bone; the upper part of the third middle metatarsal projects behind and forms the so-called hypotarsus, which in various ways, characteristic of the different groups of birds (with one or more sulci, grooved or perforated), acts as guiding pulley to the tendons of the flexor muscles of the toes. Normally the four toes have two, three, four and five phalanges respectively, but in Cypselus the number is reduced to three in the front toes. Reduction of the number of toes (the fifth shows no traces whatever, not even in Archaeopteryx) begins with the hallux, which is completely or partly absent in many birds; the second toe is absent in Struthio only. The short feet of the penguins are quite plantigrade, in adaptation to which habit the metatarsals lie in one plane and are incompletely co-ossified, thus presenting a pseudo-primitive condition.

Literature.—Only a mere fraction of the enormous literature dealing with the skeleton of birds can here be mentioned.

M. E. Alix Essai sur l’appareil locomoteur des oiseaux (Paris, 1874); E. Blanchard, “Recherches sur les caractères ostéologiques des oiseaux appliquées à la classification,” Ann. Sci. Nat. Ser. iv., t. xi.; W. Dames, “Über Brustbein Schulter- und Beckengürtel der Archaeopteryx,” Math. Naturw. Mitth., Berlin, vii., 1897, pp. 476-492; T. C. Eyton, Osteologia avium (London, 1858-1881), with many plates; C. Gegenbaur, Untersuch. z. vergl. Anat. d. Wirbelthiere, I. Carpus und Tarsus, II. Schultergürtel (Leipzig, 1864-1865); P. Harting, L’Appareil épisternal des oiseaux (Utrecht, 1864); T. H. Huxley, “On the Classification of Birds and on the Taxonomic Value of the Modifications of certain of the Cranial Bones...” P.Z.S., 1867; G. Jaeger, “Das Wirbelkörpergelenk der Vögel,” Sitzb. K. Ak. Wiss., Wien, xxxiii., 1858; A. Johnson, “On the Development of the Pelvic Girdle and Skeleton of the Hind-limb in the Chick,” Q.J.M.S., xxiii., 1883, pp. 399-411; K. F. Kessler, “Osteologie der Vogelfüsse,” Búll. Soc. Imp. Nat., Moscow, xiv., 1841; B. Lindsay, “On the Avian Sternum,” P.Z.S., 1885; E. Mehnert, “Entwickelung des Ospelvis der Vögel,” Morph. Jahrb., xiii., 1877; A. B. Meyer, Abbildungen von Vogel-Skeletten (Dresden, 1879); St G. Mivart, “On the Axial Skeleton of the Ostrich, Struthionidae, Pelecanidae,” Trans. Zool. Soc. viii., 1874; x., 1877; E. S. Morse, “On the Carpus and Tarsus of Birds,” Ann. Lyc. N.H., New York, x., 1874; J. S. Parker, “Observations on the Anatomy and Development of Apteryx,” Phil. Trans., 1890, pp. 1-110, 17 pls.; W. K. Parker, numerous papers in Trans. L.S., R.S. and Z.S., e.g. “Osteology of Gallinaceous Birds,” T.Z.S., v., 1863; “Rhinochetus,” ibid. vi.; “Skull of Aegithognathous Birds,” ibid., x., 1878; “Skull in the Ostrich Tribe,” Phil. Trans. vol. 156, 1866; “Skull of Common Fowl,” ibid. vol. 159, 1870; “Skull of Picidae,” T. Linn. Soc., 1875; “Monograph on the Structure and Development of the Shoulder-girdle and Sternum,” Ray Soc. London, 1868; W. P. Pycraft, “On the Morphology and Phylogeny of the Palaeognathae (Ratitae and Crypturi) and Neognathae,” Trans. Zool. Soc. xv., 1900, pp. 149-290, pis. 42-45; id. “Some points in the morphology of the Palate of the Neognathae,” T. Linn. Soc. 28, pp. 343-357, pls. 31-32; P. Suschkin, “Zur Morphologie des Vogelskelets. I. Schädel von Tinnunculus,” Mem. Soc., Moscow, xvi., 1900, pp. 1-63, pls.

2. Muscular System.

Of the muscles of the stem or axis, those of the neck and tail are well-developed and specialized, while those of the lower back are more or less reduced, or even completely degenerated owing to the rigidity of this region, brought about by the great antero-posterior extent of the pelvis.

The muscles of the limbs show a great amount of specialization, away from the fundamental reptilian and mammalian conditions. The muscles of the fore limbs are most aberrant, but at the same time more uniformly developed than those of the hinder extremities. The reasons are obvious. The whole wing is a unique modification, deeply affecting the skeletal, muscular and tegumentary structures, but fluttering, skimming, sailing, soaring are motions much more akin to one another than climbing and grasping, running, scratching, paddling and wading. The modifications of the hind-limbs are in fact many times greater (such as extremely long legs, with four, three or only two toes; very short legs, almost incapable of walking, with all four toes directed forwards, or two or one backwards, and two or more connected and therefore bound to act together, in various ways). Thus it has come to pass that the muscles of the hind limbs are, like their framework, more easily compared with those of reptiles and mammals than are the wings, whilst within the class of birds they show an enormous amount of variation in direct correlation with their manifold requirements. The only really aberrant modifications of the wing-muscles are found in the Ratitae, where they are, however, all easily explained by reduction, and in the penguins, where the wings are greatly specialized into blades for rowing with screw-like motions.

The wing of the bird is folded in a unique way, namely, the radius parallel with the humerus, and the whole wrist and hand with their ulnar side against the ulna; upper and forearm in a state of supination, the hand in that of strong abduction. Dorsal and ventral bending, even in the extended wing, is almost impossible. Consequently only a few of the original extensor muscles have been preserved, but these are much modified into very independent organs, notably the extensor metacarpi radialis longus, the ext. metac. ulnaris and the two radio- and ulnari-metacarpi muscles, all of which are inserted upon the metacarpus by means of long tendons. The chief muscular mass, arising from the sternum in the shape of a U, is the pectoralis muscle; its fibres converge into a strong tendon, which is inserted upon the greater tubercle and upper crest of the humerus, which it depresses and slightly rotates forwards during the downstroke. This great muscle covers completely the supracoracoideus, generally described as the second pectoral, or subclavius muscle, in reality homologous with the mammalian supraspinatus muscle. This arises mostly from the angle formed by the keel with the body of the sternum, passes by a strong tendon through the foramen triosseum, and is inserted upon the upper tubercle of the humeral crest, which it rotates and abducts. The extent of the origin of this muscle from the sternum, on which it leaves converging, parallel or diverging impressions, is of some taxonomic value.

From Newton’s Dictionary of Birds, by permission of A. & C. Black.

Fig. 14.—Wing muscles of a Goose. Bi, Biceps; Elast. sec., elastic vinculum and Exp.sec., expansor secundariorum; Pt.br and Pt.lg, short and long propatagial muscles; Tri, triceps.

Much labour has been bestowed by A. H. Garrod and Max Fürbringer upon the investigation of the variations of the inserting tendons of the patagial muscles (fig. 14), mainly from a taxonomic point of view. The propatagialis longus muscle is composed of slips from the deltoid, pectoral, biceps and cucullaris muscles. Its strong belly originates near the shoulder joint from clavicle, coracoid and scapula. Its elastic tendon runs directly to the carpus, forming thereby the outer margin of the anterior patagium, or fold of skin between the upper and forearm, which it serves to extend, together with the propatagialis brevis muscle. This runs down the anterior and outer side of the upper arm, and is attached to the proximal tendon of the extensor metacarpi radialis longus, a little below the outer condyle of the humerus. In most birds the tendon is split into several portions, one of which is often attached to the outer side of the ulna, below the elbow joint, while others are in variable but characteristic ways connected with similar slips of the propatagialis longus. The posterior patagium, the fold between trunk and inner surface of the upper arm, is stretched by the metapatagialis muscle, which is composed of slips from the serratus, superficialis, latissimus dorsi and the expansor secundariorum muscles. This, the stretcher of the cubital quills, is a very interesting muscle. Arising as a long tendon from the sterno-scapular ligament, it passes the axilla by means of a fibrous pulley, accompanies the axillary vessels and nerves along the humerus, and is inserted by a few fleshy fibres on the base of the last two or three cubital quills. Here, alone, at the distal portion of the tendon, occur muscular fibres, but these are unstriped, belonging to the category of cutaneous muscles. We have here the interesting fact that a muscle (portion of the triceps humeri of the reptiles) has been reduced to a tendon, which in a secondary way has become connected with cutaneous muscles, which, when strongly developed, represent its belly.

The flexor digitorum sublimis muscle arises fleshy from the long elastic band which extends from the inner humeral condyle along the ventral surface of the ulna to the ulnar carpal bone, over which the tendon runs to insert itself on the radial anterior side of the first phalanx of the second digit. Owing to the elasticity of the humerocarpal band the wing remains closed without any special muscular exertion, while, when the wing is extended, this band assists in keeping it taut. The arm-muscles have been studied in an absolutely exhaustive manner by Fürbringer, who in his monumental work has tabulated and then scrutinized the chief characters of fourteen selected muscles. The results are as interesting from a morphological point of view (showing the subtle and gradual modifications of these organs in their various adaptations), as they are sparse in taxonomic value, far less satisfactory than are those of the hind-limb. He was, however, the first to show clearly that the Ratitae are the retrograde descendants of flying ancestors, that the various groups of surviving Ratitae are, as such, a polyphyletic group, and he has gone fully into the interesting question of the development and subsequent loss of the power of flight, a loss which has taken place not only in different orders of birds but also at various geological periods, and is still taking place. Very important are also the investigations which show how, for instance in such fundamentally different groups as petrels and gulls, similar bionomic conditions have produced step by step a marvellously close convergence, not only in general appearance, but even in many details of structure.

Of the muscles of the hind-limbs likewise only a few can be mentioned. The ambiens muscle, long and spindle-shaped, lying immediately beneath the skin, extending from the pectineal process or ilio-pubic spine to the knee, is the most median of the muscles of the thigh. When typically developed its long tendon passes the knee-joint, turning towards its outer side, and lastly, without being anywhere attached to the knee, it forms one of the heads of the flexor perforatus digit, ii. or iii. One of the functions of this peculiar muscle (which is similarly developed in crocodiles, but absent, or not differentiated from the ilio-tibial and ilio-femoral mass, in other vertebrates) is that its contraction helps to close the second and third toes. Too much has been made of this feature since Sir R. Owen (Cyclop. Anat. Phys. i. p. 296, 1835), following G. A. Borelli (De motu animalium, Rome, 1680), explained that birds are enabled to grasp the twig on which they rest whilst sleeping, without having to make any muscular exertion, because the weight of the body bends the knee and ankle-joints, over both of which pass the tendons of this compound muscle. There are many perching birds, e.g. all the Passeres, which do not possess this muscle at all, whilst many of those which have it fully developed, e.g. Anseres, can hardly be said to “perch.”

Garrod went so far as to divide all the birds into Homalogonatae and Anomalogonatae, according to the presence or absence of the ambiens muscle. This resulted in a failure. To appreciate this, it is sufficient to enumerate the birds without the critical muscle: Passeriformes and Coraciiformes, without exception; Ardeae and Podiceps; lastly various genera of storks, pigeons, parrots, petrels and auks. The loss has taken place, and still takes place, independently in widely different groups. It follows, first, that the absence of this muscle does not always indicate relationship; secondly that we can derive birds that are without it from a group which still possess it, but not vice versa. The absence of the ambiens muscle in all owls, which apparently use their feet in the same way as the Accipitres (all of which possess it), indicates that owls are not developed from the latter, but from a group which, like the other Coraciiformes, had already lost their muscle.

Garrod further attributed much taxonomic value to the caudilio-femoralis muscle (fig. 15). This, when fully developed, consists of two parts, but inserted by a single ribbon-like tendon upon the hinder surface of the femur, near the end of its first third; the caudal part, femoro-caudalis, expressed by Garrod by the symbol A, arises from transverse processes of the tail; the iliac part (accessoro-femoro-caudal of Garrod, with the symbol B), arises mostly from the outer surface of the postacetabular ilium. Of course this double-headed condition is the more primitive, and as such exists in most nidifugous birds, but in many of these, as well as in many nidicolous birds, either the caudal or the iliac head is absent, and in a very few (Cancroma, Dicholophus, Steatornis and some Cathartes) the whole muscle is absent. The caud-ilio flexorius (semitendinosus of most authors) arises from the transverse processes of the tail, and from the distal half of the postacetabular ilium, thence passing as a broad ribbon to the popliteal region, where it splits into two portions. One of these, broad and fleshy, is inserted upon the posterior surface of the distal third of the femur. This portion, morphologically the original, was named the “accessory semitendinosus” with the symbol Y; the other portion descends on the hinder aspect of the leg and joins the fascia of the inner femoral head of the gastrocnemius muscle. In many birds the insertion is shifted from the femur to the neck of the tibia, in which case the “accessory head” is said to be absent, a condition expressed by Garrod by the symbol X. By combining the four symbols A, B, X, Y, according to their presence or absence, Garrod got a considerable number of formulae, each of which was overruled, so to speak, by the two categories of the presence or absence of the ambiens muscle. It needs hardly to be pointed out why such a purely mechanical scheme was doomed to failure. Its author, with a considerable mathematical and mechanical bias, reckoned entirely with the quantity, not with the quality of his units, and relied almost implicitly upon his formulae. It is, however, fair to state that his system was not built entirely upon these muscular variations, but rather upon a more laborious combination of anatomical characters, which were so selected that they presumably could not stand in direct correlation with each other, notably the oil-gland, caeca, carotids, nasal bones and above all, the muscles of the thigh. He was, indeed, the first to show clearly the relationship of the heron-like birds with the Steganopodes; of stork-like birds with the American vultures; the great difference between the latter and the other birds of prey; the connexion of the gulls and auks with the plovers, and that of the sand-grouse with the pigeons—discoveries expressed in the new terms of the orders Ciconiiformes and Charadriiformes. These are instances, now well understood, that almost every organic system, even when studied by itself, may yield valuable indications as to the natural affinities of the various groups of birds. That Garrod has so very much advanced the classification of birds is ultimately due to his comprehensive anatomical knowledge and general insight.

From Newton’s Dictionary of Birds.
Fig. 15.—Left thigh-muscles of a Rail. Outer view after removal of the Il.fb, ilio-fibularis and Il.tib, ilio-tibialis.
A, Caudal.	N, Sciatic nerve.
B, Iliac portion of caud-ilio-femoralis.	Is.fm, Ischio-femoralis.
X, Caud-ilio-flexorius.	Is.fl, Ischio-fibularis.
Y, “Accessory” portion of the same.	Sart. Sartorius.
Pif, Pubischio-femoralis.

To return to these thigh muscles. The most primitive combination, ambiens and A B X Y, is the most common; next follows that of A X Y, meaning the reduction of B, i.e. the iliac portion of the caud-ilio-femoralis; A B X and B X Y are less common; A X and X Y are rare and occur only in smaller groups, as in subfamilies or genera; B X occurs only in Podiceps. But the greatest reduction, with only A remaining, is characteristic of such a heterogeneous assembly as Accipitres, Cypselidae. Trochilidae, Striges and Fregata. This fact alone is sufficient proof that these conditions, or rather reductions, have been acquired independently of the various groups. A B Y, A Y, A B, X Y and B do not occur at all, some of them for obvious reasons. Occasionally there is an instructive progressive evolution expressed in these formula; for instance Phaethon, in various other respects the lowest of the Steganopodes, has A X Y, Sula and Phalacrocorax have A X, Fregata, the most specialized of these birds, has arrived at the reduced formula A. Further, the combinations B X Y and A X Y cannot be derived from each other, but both directly from A B X Y in two different directions. Keeping this in mind, we may fairly conclude that the flamingo with B X Y points to an ancestral condition A B X Y, which is still represented by Platalea and Ibis, whilst the other storks proper have taken a different line, leading to A X Y.

Literature.—Well nigh complete lists of the enormous myological literature are contained in Fürbringer’s Untersuchungen zur Morphologie und Systematik der Vögel, and in Gadow’s vol. Vögel of Bronn’s Klassen und Ordnungen des Tierreichs. Only a few papers and works can be mentioned here, with the remark that few authors have paid attention to the all-important innervation of the muscles. A. Carlsson, Beiträge zur Kenntniss der Anatomie der Schwimmvögel; K. Svensk, Vet. Ak. Handlinger. J. G. No. 3 (1884); A. Alix, Essai sur l’appareil locomoteur des oiseaux (Paris, 1874); H. Gadow, Zur vergl. Anat. der Muskulatur des Beckens und der hinteren Gliedmasse der Ratiten, 4° (Jena, 1880); A. H. Garrod, “On Certain Muscles of the Thigh of Birds and on their value in Classification,” P.Z.S., 1873, pp. 624-644; 1874, pp. 111-123. Other papers by Garrod, 1875, pp. 339-348 (deep planter tendons); 1876, pp. 506-519 (wing-muscles of Passeres), &c.; J. G. de Man, Vergelijkende myologische en neurologische Studien over Amphibien en Vogels (Leiden, 1873), (Corvidae); A. Milne-Edwards, Recherches anatomiques et paléontologiques pour servir à l’histoire des oiseaux fossiles de la France (Paris, 1867–1868), tom. i. pls. ix.-x. (Aquila and Gallus); R. Owen, article “Aves,” Todds’ Cyclopaed. of Anat. and Phys. i. (London, 1835); “On the Anatomy of the Southern Apteryx,” Trans. Zool. Soc., iii., 1849; A. Quennerstedt, “Studier i foglarnas anatomi,” Lunds Univers. Aarsk., ix., 1872 (hind-limb of swimming birds); G. Rolleston, “On the Homologies of Certain Muscles connected with the Shoulder-joint,” Trans. Linn. Soc., xxvi., 1868; R. W. Shufeldt, The Myology of the Raven (London, 1891); M. Watson, “Report on the Anatomy of the Spheniscidae,” Challenger Reports, 1883.

3. Nervous System.

Brain.—The more characteristic features of the bird’s brain show clearly a further development of the reptilian type, not always terminal features in a direct line, but rather side-departures, sometimes even a secondary sinking to a lower level, and in almost every case in a direction away from those fundamentally reptilian lines which have led to the characters typical of, and peculiar to, the mammals.

The forebrain forms the bulk of the whole brain, but the large size of the hemispheres is due to the greater development of the basal and lateral portions (pedunculi cerebri and corpora striata), while the pallium (the portion external to the lateral ventricles) is thin, and restricted to the median side of each hemisphere. As a direct result of this undoubtedly secondary reduction of the pallium—due to the excessive preponderance of the basal and lateral parts—the corpus callosum (i.e. the transverse commissure of the right and left pallium) is in birds reduced to a narrow flat bundle of a few white fibres; it is situated immediately above and behind the much stronger anterior commissure, i.e. the connexion between the corpora striata, or chief remaining part of the hemispheres. Owing to the small size of the olfactory lobes the anterior arms of the latter commissure are wanting. There is very little grey matter in the cortex of the hemispheres, the surface of which is devoid of convolutions, mostly quite smooth; in others, for instance pigeons, fowls and birds of prey, a very slight furrow might be compared with the Sylvian fissure.

The Thalamencephalon is much reduced. The epiphysis, or pineal body, is quite as degenerate as in mammals, although still forming a long stalk as in reptiles. In birds, this stalk consists entirely of blood-vessels, which in the adult enclose no terminal vesicle, and fuse with the membranous linings of the skull. The midbrain is represented chiefly by the optic lobes, the cortex of which alone is homologous with the corpora quadragemina of the mammals. Their transverse dorsal connexion is the posterior commissure; otherwise the whole roof portion of the midbrain is reduced to a thin membrane, continuous with that which covers the Sylvian aqueduct, and this ventricle sends a lateral cavity into each optic lobe, as is the case in reptiles. The right and left lobes themselves are rent asunder (so to speak), so that they are freely visible from above, filling the corners formed by the hemispheres and the cerebellum. The latter is, in comparison with mammals; represented by its middle portion only, the vermis; in a sagittal section it shows an extremely well developed arbor vitae, produced by the transverse, repeated folding of the whole organ. In comparison with reptiles the cerebellum of birds shows high development. Forwards it covers, and has driven asunder, the optic lobes; backwards it hides the much shortened medulla oblongata.

Several futile attempts have been made to draw conclusions as to the intelligence of various birds, from comparison of the weight of the whole brain with that of the body, or the weight of the hemispheres with that of other parts of the central nervous system.

The brachial plexus is formed by four or five of the lowest cervical nerves; the last nerve of this plexus often marks the boundary of the cervical and thoracic vertebrae. The composition of the plexus varies much, not only in different species, but even individually. The most careful observations are those by Fürbringer. The serial number of these nerves depends chiefly upon the length of the neck, the extremes being represented by Cypselus (10th-14th cervical) and Cygnus (22nd-24th), the usual numbers of the common fowl being the 13th-17th nerves.

The Crural Plexus is divided into a crural, ischiadic and pubic portion. The first is generally composed of three nerves, the hindmost of which, the furcalis, issues in most birds between the last two lumbo-sacral vertebrae, and then divides, one half going to the crural, the other to the sciatic portions. The obturatorius nerve invariably comes from the two main stems of the crural. The ischiadic portion consists generally of five or six nerves, which leave the pelvis as one thick system through the ilio-ischiadic foramen. The last nerve which contributes to the ischiadic plexus leaves the spinal column in most birds either between the two primary sacral vertebrae, or just below the hindmost of them, and sends a branch to the pubic portion which is composed of post-ischiadic nerves, partly imbedded in the kidneys, and innervates the ventral muscles between the tail and pubis, together with those of the cloaca and copulatory organs.

The Sympathetic System forms a chain on either side of the vertebral column. In the region of the neck lateral strands pass through the transverse canal of the cervical vertebrae; but from the thoracic region onwards, where the cardiac branch to the heart is given off, each strand is double and the basal ganglia are successively connected with the next by a branch which runs ventrally over the capitulum of the rib, and by another which passes directly through the foramen or space formed between capitulum and tuberculum. In the pelvic region, from about the level of the posterior end of the ischiadic plexus, the strand of each side becomes single again, passing ventrally over the transverse processes. Lastly, towards the caudal region the right and left strands approach and anastomose, eventually coalescing in the mid line.

Literature.—A. Bumm, “Das Grosshirn der Vögel,” Zeitschr. wiss. Zool., 38, 1883, pp. 430-466, pls. 24-25; F. Leuret and P. Gratiolet, Anatomie comparée du système nerveux (Paris, 1839-1857), with atlas; A. Meckel, “Anatomie des Gehirns der Vögel,” in Meckel’s Archiv f. Physiol. vol. ii.; H. F. Osborn, “The Origin of the Corpus Callosum, a contribution upon the Cerebral Commissures of the Vertebrata,” Morphol. Jahrbuch, 1886, xii. pp. 223-251, pls. 13-14; M. A. Schulgin, “Lobi optici der Vögel,” Zool. Anzeig. iv. pp. 277 and 303; E. R. A. Serres, Anatomie comparée du cerveau (Paris, 1824, 4 pls.); L. Stieda, “Studien über das centrale Nervensystem der Vögel und Säugethiere,” Zeitschr. wiss. Zool. xix., 1869, pp. 1-92, pls.; J. Swan, Illustrations of the Comparative Anatomy of the Nervous System (London, 1835, 4to, with plates).

Concerning the spinal nerves and their plexus: H. v. Jhering, Das peripherische Nervensystem der Wirbeltiere (Leipzig, 1871); W. A. Haswell, “Notes on the Anatomy of Birds,” Proc. Linn. Soc. N.S.W. iii., 1879; M. Fürbringer, “Zur Lehre von den Umbildungen der Nervenplexus,” Morph. Jahrb. v., 1879, p. 358.

4. Organs of Sense.

The Eye is essentially reptilian, but in sharpness of vision, power and quickness of accommodation it surpasses that of the mammals. The eyeball, instead of being globular, resembles rather the tube of a short and thick opera-glass.

The anterior half of the sclerotic is composed of a ring of some ten to seventeen cartilaginous or bony scales which partly overlap each other. Another cartilage or ossification, the posterior sclerotic ring, occurs within the walls of the posterior portion of the cup, and surrounds, especially in the Pici and in the Passeres, the entrance of the optic nerve. The iris is in most young birds at first brown or dull-coloured, but with maturity attains often very bright tints which add considerably to the charm of the bird; sexual dimorphism is in this respect of common occurrence. The iris contains a sphincter and a dilator muscle; the former, supplied by branches from the oculomotorius nerve, is under control of the will, whilst the dilator fibres belong to the sympathetic system. When fully dilated, the pupil is round in all birds; when contracted it is usually round, rarely oval as in the fowl. From near the entrance of the optic nerve, through the original choroidal fissure, arises the much-folded pecten, deeply pigmented and very vascular, far into the vitreous humour. The number of its folds varies considerably, from three in Caprimulgus to nearly thirty in crow (Corvus). Apteryx, which since Owen has generally been stated to be devoid of such an organ, likewise possesses a pecten; its base is, however, trumpet-shaped, covers almost the whole of the optic disk, and extends nearly to the lens in the shape of a thick, densely pigmented cone, without any plications, resembling in these respects the pecten of many Lacertilia (see G. L. Johnson, Phil. Trans., 1901, p. 54). In the retina the cones prevail in numbers over the rods as in the mammals, and their tips contain, as in other Sauropsida, coloured drops of oil, mostly red or yellow. Near the posterior pole of the fundus, but somewhat excentrically placed towards the temporal or outer side, is the fovea centralis, a slight depression in the retina, composed almost entirely of cones, the spot of most acute vision. Many birds possess besides this temporal fovea a second fovea nearer the nasal side. It is supposed that the latter serves monocular, the other the binocular vision, most birds being able to converge their eyes upon one spot. Consequently the whole field of vision of these birds possesses three points where vision is most acute. It may here be remembered that of the mammalia man and monkeys alone are capable of convergence, and have a circumscribed macular area.

Of the outer eyelids, the lower alone is movable in most birds, as in reptiles, and it frequently contains a rather large saucer-shaped cartilage, the tarsus palpebralis. The margins of the lids are sometimes furnished with eyelashes, e.g. in the ostrich and in the Amazon parrots, which are vestigial feathers without barbs. During the embryonic stage the lids are fused together, and either become separated shortly before the bird is hatched, as is the case with most Nidifugae, or else the blind condition prevails for some time, in the young Nidicolae. All birds have, like most reptiles, a well-developed third lid or “nictitating membrane,” which moves from the inner canthus obliquely upwards and backwards over the cornea. The moving mechanism is a further and much higher development of that which prevails in reptiles, there being two muscles completely separate from each other. Both are supplied by the abducens nerve, together with the rectus externus muscle. One, the quadratus or bursalis muscle, arises from the hinder surface of the eyeball, and forms with its narrow margin, which is directed towards the optic nerve, a pulley for the long tendon of the pyramidalis muscle. This arises from the nasal surface of the ball, and its tendon passes into the somewhat imperfectly transparent nictitating membrane. The quadrate muscle adjusts the motion, and prevents pressure upon the optic nerve; during the state of relaxation of both muscles the nictitans withdraws through its own elasticity.

See R. Leuckart in Graefe and Saemisch’s Handbuch d. Ophthalmologie (Leipzig, 1876, vol. i. chap. 7); H. Müller, Gesammelte Schriften (Otto Becker, Leipzig, 1872), and Arch. f. Ophthalmol. iii.; Ch. Rouget, “Recherches anatomiques et physiologiques sur les appareils érectiles,” “Appareil de l’adaptation de l’œil” . . . Compt. Rend. (Paris, xlii., 1856, pp. 937-941); M. Schultze, art. “Retina,” in Stricker’s Handbuch der Gewebelehre, 1871, vol. ii.; J. R. Slonaker, “Comp. Study of the Area of Acute Vision in Vertebrates,” Journ. Morph., 1897.

Ear.—The outer opening of the ear is, with rare exceptions, concealed by feathers, which are often rather stiff, or modified into bristles. There is no other protection, but slight, imperfectly movable folds of skin arise from the outer rim. The largest ear-opening is met with in the owls, with correspondingly larger folds of skin, the function of which is less that of protection than, probably, the catching of sound. In many owls the right and left ears are asymmetrical, and this asymmetry affects the whole of the temporal region, all the bones which surround the outer and middle ear, notably the squamosal and the quadrate, so that the skull becomes lopsided, one ear being turned obliquely down, the other upwards. (For, detail see Collett, Christiania Vidensk. Forhandl., 1881, No. 3.)

Fig. 16.—Auditory “chain” of Chicken. Lateral and basal views. (After W. K. Parker).

The middle ear communicates with the mouth by the Eustachian tubes, which pass between the basisphenoid and basioccipital bones, and unite upon the ventral side of the sphenoid, a little behind its articulation with the pterygoids, where they open into the mouth cavity by a short membranous duct. The columellar apparatus, or auditory chain of ossicles (fig. 16), extending between the fenestra ovalis and the tympanic membrane or drum, consists of (1) the long and slender columella, a straight, ossified rod which fits with a disk into the fenestra ovalis; it is homologous with the stapes (m.st.), although not stirrup-shaped; (2) the extracolumellar mass. This is chiefly cartilaginous and sends out three processes: the dorsal (s.st.) is attached to the upper wall of the drum cavity; the outermost (e.st.) is fastened on to the middle of the drum membrane; the third, ventral or infracolumellar process (i.st.) is directed downwards and tapers out into a thin, partly cartilaginous, strand, which originally extended to the inner corner of the articular portion of the mandible, but on its long way comes to grief, being squeezed in between the pterygoid and quadrate. This long downward process being homologous with an almost exactly identical arrangement in the crocodile, and with the processus folii of the mammalian malleus, it follows that the whole extracolumellar mass, that between stapes and drum, is equivalent to incus and malleus of the mammalia. There is, in birds, no annulus tympanicus. Birds possess an ear-muscle which at least acts as a tensor tympani; it arises near the occipital condyle, passes through a hole into the tympanic cavity, and its tendon is, in various ways, attached to the inside of the membrane and the neighbouring extracolumellar processes.

As regards the inner ear, the endolymphatic duct ends in a closed saccus, imbedded in the dura mater of the cranial cavity. The apex of the cochlea is turned towards, and almost reaches the anterior wall of the occipital condyle; at most it makes but half a twist or turn; it possesses both Reissner’s membrane and the organ of Corti. Although the scala tympani is so rudimentary, not reaching a higher level than in most of the reptiles, and remaining far below the mammalia, birds do not only hear extremely well, but they distinguish between and “understand” pitch, notes and melodies.

See G. Breschet, Recherches anatomiques et physiologiques sur l’organe de l’audition chez les oiseaux (Paris, 1836), with Atlas; C. Hasse, various papers in Zeitschr. f. wiss. Zool. vol. xvii, and in Anatomische Studien, pts. ii. and iv. (Bresku, 1871); I. Ibsen, Atlas anatomicus auris internae (Copenhagen, 1846); G. Retzius, Das Gehörorgan der Wirbelthiere (Stockholm, 1884), ii. pp. 139-198, pls. 15-20.

Nose.—The olfactory organ is poorly developed, and it is still a question whether birds possess much power of smell; many are certainly devoid of it.

The olfactory perceptive membrane is restricted to the posterior innermost region of the nasal chamber, where it covers a slight bulging-out prominence on the nasal wall. This so-called third, upper or posterior conch is not a true conch, nor is that of the vestibulum; only the middle one forms a scroll, and this corresponds to the only one of reptiles and the lower of the mammals. The nasal cavity communicates with the mouth by the choanae or posterior nares, situated between the palatine process of the maxillary, the palatine and the vomer. The outer nares or nostrils are most variable in size and shape. In the Steganopodes they tend to become much reduced, e.g. in cormorants (Phalacrocoracidae), and especially in Sula, where the nasal slits become completely closed up, and the greater portion of the nasal cavity is also abolished, being restricted to the olfactory region with its unusually wide choanae. The nasal septum is often more or less incomplete, producing nares peniae, e.g. in the Cathartae, in the Anseres, gulls, rails and various other aquatic birds. The secretions of the mucous membrane of the nasal cavity, and a pair of naso-lacrymal glands (not to be confounded with the Harderian and the lacrymal glands), moisten and clean the chamber. The glands are variable in size and position; when very large, e.g. in plovers, they extend upon the forehead, causing deep impressions on the bones of the skull. Jacobson’s organ has been lost by the birds, apparently without a trace in the embryonic fowl, but T. J. Parker has described vestiges of the corresponding cartilages in the Apteryx (Phil. Trans., 1890).

See C. Gegenbaur, “Über die Nasenmuscheln der Vögel,” Jena Zeitschr. vii., 1873, pp. 1-21.

5. Vascular System.

The heart lies in the middle line of the body, its long axis being parallel with that of the trunk. The whole ventral surface of the pericardium is exposed when the sternum is removed. The right and left halves are completely divided by septa, no mixture of the venous and arterial blood being possible, an advance upon reptilian conditions, even the highest.

The atria are comparatively small, the walls being thin, especially those of the right, which possesses numerous muscular ridges projecting into the cavity presenting a honeycombed appearance. The interauricular septum is mostly entirely membranous; in the middle it is thinner, rather transparent, but there is no depression or fossa ovalis. The whole sinus venosus has become part of the right atrium. It receives the three great venous trunks of the body, namely the vena cava superior dextra, the vena cava superior sinistra more dorsally, and the vena cava inferior more to the right and below; the opening of the last is guarded by two prominent valves in place of the mammalian valvula Eustachii. The right ventricle occupies the ventral portion of the heart. The communication with the atrium is guarded by a valvula cardiaca dextra, which only in function represents the mammalian tricuspid; it consists of an oblique reduplication of the muscular fibres together with the endocardiac lining of the right ventricle, while the opposite wall is convex and forms neither a velum nor papillary muscles, nor chordae tendineae. The right anterior corner of the right ventricle passes into the short stem, guarded by three semi-lunar valves, which divides into the two pulmonary arteries. There are likewise two pulmonary veins, entering the left atrium by one orifice. Two or three membranous flaps, held by numerous chordae tendineae, form a true mitral valve, and allow the blood to pass through the left ostium atrioventriculare. The blood leaves the heart past three semi-lunar valves, by the right aorta, this being alone functional, a feature characteristic of, and peculiar to, birds. Remnants of the left aortic arch persist sometimes in the shape of a ligamentous strand. The aortic trunk is very short, sends off the coronary arteries and then the left aorta brachiocephalica, while the rest divides into the right brachiocephalic and the aorta descendens. Each brachiocephalic soon sends off its subclavian, while in the normal or more usual cases the rest proceeds as the carotid trunk, inclusive of the vertebral artery. But the carotids show several interesting modifications which have been examined chiefly by C. L. Nitzsch and by A. H. Garrod. (1) The right and left carotids converge towards the middle and extend up the neck, imbedded in a furrow along the ventral surface of the cervical vertebrae. This is the usual arrangement. (2) The two carotids are fused into one carotis conjuncta, imbedded in a special median osseous semicanal of the vertebrae; e.g. herons, flamingos, and some parrots. (3) There is one carotis conjuncta, but the basal portion of its original right component is obliterated, leaving a so-called c. primaria sinistra, an unfortunate name. Such Aves laevocarotidinae of Garrod are common, e.g. all the Passeriformes. (4) The reverse of the third modification, producing a c. primaria dextra in the bustard Eupodotis. In other likewise very rare cases a left, or a left and right, superficial carotids are developed and take the place of the then vanished deep or primary carotids.

Venous System.—The bird’s liver receives nearly all the blood from the stomach, gut, pancreas and spleen, as well as from the left liver itself, into the right hepatic lobe, by a right and left portal vein. The venae hepaticae magnae join the vena cava posterior and thereby form with it the vena cava inferior. The left hepatica magna receives also the umbilical vein, which persists on the visceral surface of the abdominal wall, often anastomosing with the epigastric veins. A likewise unpaired vena coccygeo-mesenterica is usually present. There is no renal portal system, excepting unimportant vestiges of such a system in the head kidneys.

Lymphatic System.—The white blood-corpuscles are produced in the follicles at the base of the intestinal villi. The lymph vessels of the tail and hinder parts of the body enter the hypogastric veins; and at the point of junction, on either side, lies a small lymph heart, which often persists until maturity. The red blood-corpuscles are invariably oval disks, with a central nucleus which causes a slight swelling; hence they are oval and biconvex.

See A. H. Garrod, “On the Carotid Arteries of Birds,” Proc. Zool. Soc., 1873, pp. 457-472; E. A. Lauth, “Mémoire sur les vaisseaux lymphatiques des oiseaux,” Ann, Sci. nat. (iii. 1824), p. 381; J. J. Mackay, “The Development of the Branchial Arterial Arches in Birds, with special reference to the Origin of the Subclavians and Carotids,” Phil. Trans. 179 B (1888), pp. 111-141; L. A. Neugebauer, “Systema venosum avium,” Nov. Act. Leopold. Carol. xxi., 1844, pp. 517-698, 15 pls.; R. Gasch, “Beiträge zur vergl. Anatomie des Herzens der Vögel und Reptilien,” Arch. f. Naturgesch., 1888.

6. Respiratory System.

The lungs are small and occupy only the dorsal portion of the thoracic cavity. There is only one right and one left lobe, each traversed through its whole length by a mesobronchium, whence arise about ten secondary bronchia; these send off radially arranged parabronchia, which end blindly near the surface. The walls of these tertiary tubes send out, in all directions, canaliculi aeriferi which, ending in slight swellings, recall the mammalian aveoli.

Highly specialized air-sacs are characteristic of all birds. They are very thin-walled membranes, very poor in blood-vessels, formed by the bulged-out pleural or peritoneal covering of the lungs, through the parabronchial tubes of which they are filled with air. Their function is not quite clear. The usual suggestion, that the warm air contained within them assists the bird in flight, balloon-like, is absurd. They assist in the extremely rapid and vigorous ventilation of the lungs, the latter being capable of but very limited expansion and contraction in birds. Exchange of gas through the walls of the air-sacs, almost devoid of blood-vessels, can at best be much restricted.

There are five pairs of larger sacs belonging to the pulmonary system:—(1) prebronchial or cervical, extending sometimes far up the neck, even into the cranial cavities; the throat-bags of the prairie fowls (Cupidonia and Pedioecetes) are a further development; (2) subbronchial or interclavicular; (3 and 4) anterior and posterior thoracic or intermediate; (5) abdominal sacs. Most of these extend through narrow apertures—foramina pneumatica—into the hollow bones, sometimes, e.g. in hornbills and screamers, into every part of the skeleton, or, in the shape of innumerable pneumatic cells, even beneath the skin. There is also a naso-pharyngeal or tympanic system of air-sacs, restricted to the head (cf. the siphonium described in connexion with the mandible), but filling also such curious organs as the frontal excrescence of Chasmorhynchus, the Brazilian bell-bird, the throat-bag of the adjutant stork, and the gular pouch of the bustard.

The trachea or windpipe is strengthened by numerous cartilaginous, often osseous, complete rings, but in the emeu several of these rings are incomplete in the medioventral line, and permit the inner lining of the trachea to bulge out into a large neck-pouch, which is used by both sexes as a resounding bag. In humming-birds and petrels the trachea is partly divided by a vertical, longitudinal, cartilaginous septum. In some of those birds which have a peculiarly harsh or trumpeting voice, the trachea is lengthened, forming loops which lie subcutaneously (capercally, curassow), or it enters and dilates the symphysis of the furcula (crested guineafowl); or, e.g. in the cranes and in the hooper swan, even the whole crest of the sternum becomes invaded by the much elongated, manifolded trachea.

The syrinx or lower larynx is the most interesting and absolutely avine modification, although absent as a voice-producing organ (probably due to retrogression) in most Ratitae, storks, turkey buzzards (Cathartes) and Steganopodes. The syrinx is a modification of the lower part of the trachea and of the adjoining bronchi. Essential are vibrating membranes between the cartilaginous framework, and next, special muscles for regulating the tension. The majority of birds possess a pair of internal tympaniform membranes forming the inner or median walls of the bronchi, which are there furnished with semi-rings only. External tympaniform membranes exist, with great variations, between the specialized one or two last tracheal and some of the first bronchial rings.

According to the position of the chief sound-producing membranes, three types of syrinx are distinguishable:—(1) Tracheo-bronchial, by far the commonest form, of which the two others are to a certain extent modifications. The essential feature is that the proximal end of the inner membranes is attached to the last pair of tracheal rings; outer tympaniform membranes exist generally between the 2nd, 3rd and 4th bronchial semi-rings. This type attains its highest development in the Oscines, but it occurs also in many other orders. (2) Syrinx bronchialis. The outer membranes are spread out between two or more successive bronchial semi-rings, a distance from the trachea which is, in typical cases, devoid of sounding membranes; some Cuculi, Caprimulgi, and some owls. (3) Syrinx trachealis. The lower portion of the trachea consists of thin membranes, about half a dozen of the rings being very thin or deficient. Inner and outer membranes may exist on the bronchi. The Tracheophonae among the Passeriformes, the possessors of this specialized although low type of syrinx, form a tolerably well-marked group, entirely neotropical. But indications of such a syrinx occur also in Pittidae, pigeons and gallinaceous birds (Gallidae), the last cases being clearly analogous.

Whilst the type of syrinx affords no help in classification, it is very different with its muscles. These—as indicated by their supply from a branch of the hypoglossal nerve, which descends on either side of the trachea—are, so to speak, a detached, now mostly independent colony of glosso-pharyngeal muscles. Omitting the paired tracheo-clavicular muscles, we restrict ourselves to the syringeal proper, those which extend between tracheal and bronchial rings. Their numbers vary from one pair to seven, and they are inserted either upon the middle portion of the bronchial semi-rings (Mesomyodi), or upon the ends of these semi-rings where these pass into the inner tympaniform membrane (Acromyodi). The former is morphologically the more primitive condition, and is found in the overwhelming majority of birds, including many Passeriformes. The acromyodian type is restricted almost entirely to the Oscines. Further, according to these muscles being inserted only upon the dorsal, or only upon the ventral, or on both ends of the semi-rings, we distinguish between an-, kat- and diacromyodi. But the distinction between such Acromyodi and the Mesomyodi is not always safe. For instance, the Tyranninae are anacromyod, while the closely allied Pipras and Cotingas are katacromyod; both these modifications can be shown to have been derived but recently from the weak meso- and oligomyodian condition which prevails in the majority of the so-called Oligomyodi. On the other hand, the diacromyodian type can have been developed only from a strong muscular basis which could split into a dorsal and a ventral mass; moreover, no Passeres are known to be intermediate between those that are diacromyodian and those that are not.

Attempts to derive the anacromyodian and the katacromyodian from the diacromyodian condition are easy on paper, but quite hopeless when hampered by the knowledge of anatomical facts and how to use them. There remains but one logical way, namely, to distinguish as follows:—(1) Passeres anisomyodi, in which the syrinx muscles are unequally inserted, either on the middle or on one end of the semi-rings, either dorsal or ventral. This type comprises the Clamatores. (2) Passeres diacromyodi, in which some of the syrinx muscles are attached to the dorsal, and some to the ventral ends, those ends being, so to say, equally treated. This type comprises the Oscines. Both types represent rather two divergent lines than successive stages, although that of the Clamatores remains at a lower level, possessing at the utmost three pairs of muscles, whilst these range in the Oscines from rarely two or three to five or seven.

This way of using the characters of the syrinx for the classification of the Passeriformes seems simple, but it took a long time to accomplish. Joh. Müller introduced the terms Polymyodi and Tracheaphones, Huxley that of Oligomyodi; Müller himself had, moreover, pointed out the more important characters of the mode of insertion, but it was Garrod who invented the corresponding terms of Acro- and Mesomyodi (= Tracheophones + Oligomyodi). (For further historical detail, see Ornithology). After W. A. Forbes had investigated such important genera as Philepitta and Xenicus, P. L. Sclater, A. Newton and R. B. Sharpe divided the Passeres respectively into Oscines, Oligomyodae, Tracheophonae and Pseudoscines (= Suboscines); Oligomyodae, Tracheophonae and Acromyodae; Oscines, Oligomyodae, Tracheophonae and Atrichiidae. Ignoring the fact that some Oligomyodae are meso- and others acromyodian, they tried to combine two irreconcilable principles, namely, mere numbers against quality.

Bibliography.—M. Baer, “Beitr. z. Kenntniss d. Atemwerkzeuge bei den Vögeln,” Zeitschr. wiss. Zool. lxi. 1896, pp. 420-498; Campana, Physiologie de la respiration chez les oiseaux. Anatomie de l’appareil pneumatique ... (Paris, 1875); A. H. Garrod, “Major Divisions of Passerine Birds (syrinx, &c.),” P.Z.S., 1876, pp. 506-519; and “On the Conformation of the Thoracic Extremity of the Trachea in the Class Aves,” P.Z.S., 1879, pp. 357-380; J. Müller, Stimmorgane der Passerinen, Müller’s Arch. (1847); and Abh. Akad. Wiss. (Berlin, 1845-1847), translation by F. J. Bell, Oxford, 1878; H. Strasser, “Luftsäcke der Vögel,” Morph. Jahrb. iii., 1877, pp. 179-227; C. Wunderlich, “Unterer Kehlkopf der Vögel,” Nov. Act. Leop. Carol., 1884; Ph. C. Sappey, Recherches sur l’appareil respiratoire des oiseaux (Paris, 1847); W. A. Forbes, “Contributions to the Anatomy of Passerine Birds (syrinx),” P.Z.S., 1880, pp. 380-386, 387-391; 1881, pp. 435-737; 1882, pp. 544-546, 569-571; W. Yarrell, “Observations on the tracheae of Birds,” Trans. Linn. Soc., 1827, pp. 378-391.

7. Digestive System.

For a general account of the digestive organs, see Alimentary Canal. Here only a few peculiar features may be mentioned.

The young pigeons are fed by both parents with a peculiar stuff, the product of the strongly proliferating epithelial cells of the crop, which cells undergo a cheese-like fatty degeneration, and mixed with mucus, perhaps also with the proventricular juice, make up a milk-like fluid. Should the young die or be removed during this period, the parents are liable to die, suffering severely from the turgid congestion of the hypertrophied walls of the crop.

The male of the hornbills, Bucerotinae, feeds his mate, which is imprisoned, or walled-up in a hollow tree, during the whole time of incubation, by regorging his food. This bolus is surrounded, as by a bag, by the cast-up lining of the gizzard. Since this process is repeated for many days the habitual reaction of the stomach well-nigh exhausts the male. A graphic account of this is given in Livingstone’s travels.

The hoactzin, Opisthocomus, feeds to a great extent upon the leaves of the aroid Montrichardia or Caladium arborescens. The crop is modified into a large and very rugose triturating apparatus, while the gizzard, thereby relieved of its function, is reduced to the utmost. The large and heavy crop has caused a unique modification of the sternal apparatus. The keel is pushed back to the distal third of the sternum, whilst the original anterior margin of the keel is correspondingly elongated, and the furcula fused with the rostral portion.

In the ostrich, Struthio, the craze of overloading the stomach with pebbles which, when triturated into sand, are not voided, has brought about a dislocation, so that the enormously widened and stretched space between proventriculus and gizzard forms a bag, directed downwards, whilst the gizzard itself with part of the duodenum is rotated round its axis to more than 100°. A similar rotation and dislocation occurs in various petrels, in correlation with the indigestible sepia-bills, &c., which these birds swallow in great quantities. In Plotus, the snakebird, the pyloric chamber of the stomach is beset with a mass of hair-like stiff filaments which permit nothing but fluid to pass into the duodenum. The gizzard of various birds which are addicted to eating hairy caterpillars, e.g. Cuculus canorus and trogons, is often lined with the broken-off hairs of these caterpillars, which, penetrating the cuticle, assume a regular spiral arrangement, due to the rotatory motion of the muscles of the gizzard.

8. Cloaca and Genital Organs.

The cloaca is divided by transverse circular folds, which project from its inner walls, into three successive chambers. The innermost, the coprodaeum, is an oval dilatation of the end of the rectum, and attains its greatest size in those birds whose faeces are very fluid; it serves entirely as the temporary receptacle of the faeces and the urine. The next chamber, the urodaeum, is small, and receives in its dorso-lateral wall the ureters and the genital ducts; above and below this chamber is closed by circular folds, the lower of which, towards the ventral side, passes into the coating of the copulatory organ when such is present. The urodaeum serves only as a passage, the urine being mixed with the faeces in the chamber above. The third or outermost chamber, the proctodaeum, is closed externally by the sphincter ani; the orifice is quite circular. It lodges the copulatory organ, and on its dorsal wall lies the bursa Fabricii, an organ peculiar to birds. It is most developed in the young of both sexes, is of unknown function, and becomes more or less obliterated in the adult. Only in the ostrich it remains throughout life, being specialized into a large receptacle for the urine, an absolutely unique arrangement. A true urinary bladder, i.e. a ventral dilatation of the urodaeum, is absent in all birds. It is significant that the whole type of their cloaca much resembles that of the Crocodilia and Chelonia, in opposition to that of the Lacertilia.

The penis, and its much reduced vestige of the female, is developed from the ventral wall of the proctodaeum. It occurs in two different forms. In the Ratitae, except Rhea, it consists mainly of a right and left united half (corpora fibrosa), with a deep longitudinal furrow on the dorsal side, and much resembles the same organ in crocodiles and tortoises. It is protruded and retracted by special muscles which are partly attached to the ventral, distal end of the ilium. Another type exists in Rhea and in the Anseriformes, greatly specialized by being spirally twisted and partly reversible like the finger of a glove. This is mainly due to the greater development of an unpaired, median portion, analogous to the mammalian corpus spongiosum, which is much less prominent in the Ratitae; the muscles of this type are derived solely from the anal sphincter. In other Carinatae, e.g. tinamous and storks, the penis is very much smaller and simpler, with every appearance of a degenerated organ. In the great majority of birds it has disappeared completely and the primitive way of everting the cloaca is resorted to.

Both right and left testes are functional. They become greatly enlarged in the breeding season; in the sparrow, for instance, from the size of a mustard seed to that of a small cherry. The vas deferens descends with many undulations down the lateral side of the ureter of the same side, and opens upon a small papilla into the urodaeum. Extraordinary increase in length during the breeding season causes the vasa deferentia in some of the African weaver-birds to protrude, or to bulge out the cloacal walls beyond the vent. The spermatozoa exhibit many differences in shape, size and proportions, in the various groups of birds. They have been studied minutely by E. Ballowitz.

Only the left ovary becomes functional, with rare individual exceptions. Both present the appearance of diminutive clusters of grapes, at the anterior end of the kidneys, close to the suprarenal bodies, separated from each other by the descending aorta and by the vena cava where this is formed by the right and left vena iliaca communis. During the breeding season many more eggs are developed than reach maturity, amounting in most birds to several dozens. Those germs which do not ripen during the season undergo a process of resorption, and in the winter the whole ovary dwindles to often a diminutive size. In young birds both oviducts are almost equal in size, but the right soon degenerates into an insignificant strand. During every laying season the left duct increases enormously by new formation of its component fibres. For instance, in the fowl its volume increases about fifty-fold, growing from some 6 in. in length and scarcely one line in width to more than 2 ft. in length and ½ in. in thickness. The upper, wide opening of the duct is attached by elastic, peritoneal lamellae to the hinder margin of the left lung; the middle portion of the duct is glandular and thick-walled, for the deposition of the albumen; it is connected by a short, constricted “isthmus” (where the shell-membrane is formed) with a dilated “uterus” in which the egg receives its calcareous shell and eventual pigmentation.

Bibliography.—A. v. Brunn, Rückbildung nicht ausgestossener Eierstockseier, Henle Festschrift (Bonn, 1882); E. Ballowitz, “Die Spermatozoen der Vögel,” Arch. Mikr. Anat. xxxii., 1888, pls. 14-18; M. Sacchi, “Contribuzione all’ istiologia del ovidotto dei sauropsidi,” Att. Soc. Ital., Milano, vol. xxx.; W. A. Forbes, “On the Bursa Fabricii in Birds,” P.Z.S., 1877, pp. 304-318; H. Gadow, “Remarks on the Cloaca and on the Copulatory Organs of the Amniota.” Phil. Trans., 1887, pp. 5-37, pls. 2-5; Martin Saint Ange, “Étude de l’appareil reproducteur dans les cinq classes d’animaux vertébres,” Mem. Ac. Soc., Paris, xiv., 1856; E. Retterer, “Contribution à l’étude du cloaque et de la bourse de Fabricius,” Robin’s Journ. de l’anat. et physiol., 1885, pp. 369-454, pls. 17-19.

B. Fossil Birds

Much had naturally been expected from the study of fossil birds, but, so far as the making of classifications is concerned, they have proved rather a source of perplexities. So long as the characters of new fossils are only of specific and generic value, it is mostly possible to assign the birds to their proper place, but when these characters indicate new families or orders, for instance Hesperornithes, Ichthyornithes, Palaelodi, their owners are put outside the more tersely constructed classifications applicable to modern birds. It is no exaggeration to say that the genus, often even the species, can be determined from almost any recent bone, but in the case of Miocene, and still more, of Eocene fossils, we have often to deal with strange families, which either represent an extinct side branch, or which connect several recent groups with each other. Our artificially-established classifications collapse whilst we gain further insight into the mutual affinities of the existing groups. Of course this must be so if evolution is true. But it also follows that, if every extinct and recent bird were known, neither species, nor genera, nor families, nor orders could be defined. We should be able to construct the pedigree of every group, in other words, the gigantic natural system, but there would be no classification. Much light has also been thrown by fossil birds upon the study of geographical distribution. The key to the distribution of recent groups lies in that of the extinct forms. Not only have many absolutely new families been discovered, but many kinds of modern birds are now known to have existed also in countries which they are now extinct. There were, for instance, trogons, secretary-birds, parrots, and other now Ethiopian forms in Miocene France. Ostriches, undistinguishable from Struthio, have been found in Samos and in the Sivalik Hills.

The proper study of fossil birds may be said to have begun with A. Milne-Edwards, whose magnificent Oiseaux fossiles de la France was published from 1867 to 1871. This work deals chiefly with mid-Tertiary forms. A new impetus was given by O. C. Marsh, who, after 1870, discovered a great number of bird remains in the Cretaceous strata of North America. The most important result is the proof that, until the end of the Cretaceous epoch, most, if not all, birds were still possessed of teeth (see Odontornithes).

The oldest known bird is the Archaeopteryx (q.v.), of the upper Oolite in Bavaria. The imprints in the enormously older new red sandstone or Lower Trias of Connecticut, and originally named Ornithichnites, belong to Dinosaurian Reptiles.

A wide gap separates Archaeopteryx from the next order of fossil birds of the Cretaceous epoch, and, since freshwater deposits of that age are rare, bird remains are uncommon. Many bones formerly referred to birds have since proved to belong to Pterodactyls, e.g. Cimoliornis from the English Chalk. But in 1858 were discerned in the Upper Greensand of Cambridgeshire remains which are now known as Enaliornis. W. Dames has described bones from the Chalk of southern Sweden under the name of Scaniornis, probably allied to Palaelodus. From the Cretaceous rocks of North America a large number of birds have been described by O. C. Marsh. Of these the most interesting are Ichthyornis (= Graculavus) and Hesperornis, from the Cretaceous shales of Kansas. They were placed by Marsh in a distinct subclass of birds, Odontornithes (q.v.). Probably all birds of Cretaceous age were still possessed of teeth. Baptornis, another of Marsh’s genera, seems to be allied to Enaliornis, Palaeotringa and Talmatornis, were by him referred to Limicoline and Passerine birds. Laornis from the Cretaceous marls of New Jersey was as large as a swan.

Fig. 17.—Remains of head of Odontopteryx, from the original in the British Museum; side view; natural size.

Fig. 18.—Remains of head of Odontopteryx, seen from above.

The lower Eocene has furnished a greater number of bird bones. Some of the largest are those of Gastornis, with three species from France, Belgium and England. Much difference of opinion obtains as to the affinities of these birds, which were far larger than an ostrich; they were undoubtedly incapable of flight and there are indications of teeth in the upper jaw. Provisionally this genus has been grouped with the Ratitae, which at any rate are a heterogenous assembly. Sir R. Owen’s Dasornis, of the London Clay, known from an imperfect cranium, and E. D. Cope’s Diatryma of New Mexico, based upon a gigantic metatarsus, may also belong there. The London Clay of South England has likewise supplied some long upper arm bones, Argillornis. The most remarkable specimen is a skull, Odontopteryx toliapicus (figs. 17, 18); the edges of the jaws were serrated like those of certain tortoises. The character of this skull and the compound rhamphotheca (known by the imprints left upon the jaws) indicate affinities with the Steganopodes. Remnants of a heron-like bird, Proherodius, of a gull-like creature, Halcyornis, a raptorial Lithornis; and a supposed Passerine from Glarus in Switzerland, called Protornis = Osteornis, complete the list.

The upper Eocene has yielded many birds, most of which are at least close forerunners of recent genera, the differentiation into the leading orders and families being already well marked, e.g. Gallinaceous birds, stork- and crane-like waders, rails, birds of prey, cormorants, &c. Especially numerous bones have been found in the Paris basin, chiefly described by G. Cuvier, F. L. P. Gervais, E. Blanchard, and above all by A. Milne-Edwards, and in the equivalent beds of Hampshire. Others have been discovered in Wyoming; a giant penguin, Palaeeudyptes, is known from New Zealand, and Palaeospheniscus from Patagonia. The Miocene has yielded by far the greatest number of bird-bones, including even eggs and imprints of feathers. For instance, from the lower Miocene beds of Allier and Puy-de-Dôme Milne-Edwards has described about 50 species. Of these Palaelodus was an ancestral flamingo, but with shorter legs; Limnatornis is referred to the hoopoes. The existing genera include Anas, Aquila, Bubo, Columba, Cypselus, Lanius, Picus, Phalacrocorax, Sula, &c. Very interesting is the fact that Serpentarius, Psittacus and Trogon are amongst this list of birds, which are now restricted to the tropics. A similarly mixed avifauna has been found in the mid-Miocene beds of various other parts of France, Germany and Italy. In Colorado and New Mexico Marsh has detected bones of Meleagris, Puffinus, Sula and Uria, all existing genera; but the first is especially suggestive, since it is one of the most characteristic forms of the New World.

Here may be interpolated a short account of the very peculiar avifauna found in the Tertiary strata of Santa Cruz in Patagonia. Instead of the age of lower Eocene, as had been stated originally, these beds are not older than mid-Miocene, and not a few of the bones are of a much younger, even latest Tertiary date. Discovered, and partly described, by F. Ameghino, the bones have been sumptuously monographed by F. P. Moreno and A. Mercerat, who proposed for them the name of Stereornithes, a new order of birds, mostly gigantic in size, and said to combine the characters of Anseres, Herodiones and Accipitres. But the whole mass of bones is in hopeless disorder, apparently without any record of association. At any rate, the “Stereornithes,” accepted as such in Bronn’s Thierreich, and in Newton’s Dictionary of Birds, had to be dissolved as an unnatural, haphazard assembly. Many of these birds, to judge from the enormous size of their hind-limbs, were undoubtedly flightless, e.g. Brontornis, and remind us of the Eocene Gastornis of Europe. Phororhacos, the most extraordinary of all, belongs to the Gruiformes, perhaps also Pelecyornis and Liornis. On the other hand, the late Tertiary Dryornis is a member of the Cathartae or American vultures, and Mesembriornis, likewise of late Tertiary date, is a close forerunner of the recent genus Rhea.

Pliocene remains are less numerous than those of the Miocene. From Pikermi in Greece is known a Gallus, a Phasianus and a large Grus. From Samos a large stork, Amphipelargus, and a typical Struthio; from the Sivalik Hills on the southern flanks of the Himalayas also an ostrich, and another Ratite with three toes, Hypselornis, as well as Leptoptilus, Pelecanus and Phalacrocorax. The fossil egg of a struthious bird, Struthiolithus, has been found near Cherson, south Russia, and in north China. The Suffolk Crag has yielded the unmistakable bones of an albatross, Diomedea.

Most Pleistocene birds are generically, even specifically, identical with recent forms; some, however, have become extinct, or they have become exterminated by man. A great number of birds’ bones have been found in caves, and among them some bearing marks of human workmanship. In France we have a large and extinct crane, Grus primigenia, but more interesting are the numerous relics of two species, the concomitants even now of the reindeer, which were abundant in that country at the period when this beast flourished there, and have followed it in its northward retreat. These are the snowy owl, Nyctea scandiaca, and the willow-grouse, Lagopus albus. A gigantic swan, Cygnus falconeri, is known from the Zebug cavern in Malta. From caves of Minas Geraes in Brazil, O. Winge has determined at least 126 species, of which nearly all still survive in the country. Kitchen-middens of England, Ireland and Denmark reveal the existence of the capercally, Tetrao urogallus, and of the great auk or gare-fowl, Alca impennis; both species long since vanished from those countries. In the fens of East Anglia have been found two humeri, one of them immature, of a true Pelecanus, a bird now no longer inhabiting middle Europe.

Until a very recent epoch there flourished in Madagascar huge birds referable to the Ratitae, e.g. Aepyornis maximus, which laid enormous eggs, and not unnaturally recalls the mythical “roc” that figures so largely in Arabian tales. New Zealand has also yielded many flightless birds, notably the numerous species and genera of Dinornithidae, some of which survived into the 19th century (see Moa); Pseudapteryx allied to the Kiwi; Cnemiornis, a big, flightless goose; Aptornis and Notornis, flightless rails; and Harpagornis, a truly gigantic bird of prey with tremendous wings and talons.

From a tracing by M. A. Milne-Edwards of the original drawing in a MS. Journal kept during Wolphart Harmanszoon’s voyage to Mauritius (A.D. 1601–1602), penes H. Schlegel (Proc. Zool. Soc. 1875, p. 350). Reduced.

Fig. 19.—Extinct Crested Parrot of Mauritius
(Lophopsittacus mauritianus).

It is, of course, quite impossible, in a survey of extinct birds, to divide them into those which are bona fide fossil, sub-fossil, recently extirpated and partially exterminated. Nor is it possible, except in a few cases, to decide whether they have come to an end through the agency of man or through so-called natural causes. Like other creatures birds have come, some to flourish and stay, others to die out.

Fig. 20.—Mandible of Aphanapteryx, side view. (From the original in the Museum
of Zoology of the University of Cambridge.)

Mauritius is famous for the dodo, killed off by man; there was also a curiously crested parrot, Lophopsittacus (fig. 19). In the Mare aux Songes have been found the bones of another parrot, of ducks, pigeons, rails, herons, geese and of a dwarf darter, Plotus nanus, all sub-fossil, now extinct. Very interesting is Aphanapteryx (fig. 20), a long-billed, flightless rail, practically the same as Erythromachus of Rodriguez and Diaphorapteryx of Chatham Island. Réunion possessed the peculiar starling, Fregilupus. Rodriguez was inhabited by Pezophaps, the solitaire, Necropsittacus and Palaeornis exsul, which is now probably extinct. The Antilles tell a similar tale. The great auk, once common on the British coasts, those of Denmark, the east coast of North America, then restricted to those of Newfoundland, Greenland and Iceland, has been killed by man, and the same fate has overtaken the Labrador duck, the Phillip Island parrot, Nestor productus, and the large cormorant of Bering Island, Phalacrocorax perspicillatus; and how long will the flightless cormorant, Ph. harrisi of the Galapagos, survive its quite recent discovery?

Fig. 21.—Pied Duck (Somateria labradora), male and female.
(From specimens in the British Museum. Reduced.)

Authorities.—A. Milne-Edwards, Recherches anatomiques et paléontologiques pour servir à l’histoire des oiseaux fossiles de la France (Paris, 1867–1868); F. P. Moreno and A. Mercerat, Catalogo de los Pajaros fosiles de la Republica Argentina. Anales Mus. La Plata, 1891, 21 pls.; O. C. Marsh, Odontornithes: A monograph of the Extinct Toothed Birds of North America (New Haven, Conn., 1880); R. Lydekker, article “Fossil Birds,” in A. Newton’s Dictionary of Birds (London, 1893); Cat. Foss. Birds, Brit. Museum, 1891; K. v. Zittel, Handbuch der Paläontologie, i. 3 (1887–1890); C. W. Andrews, “On the Extinct Birds of Patagonia,” Tr. Zool. Soc. xv., 1899, pp. 55-86, pls. 14-17.

C. Geographical Distribution

The study of the extinct organisms of any country leads to a proper appreciation of its existing flora and fauna; while, on the other hand, a due consideration of the plants and animals which may predominate within its bounds cannot fail to throw more or less light on the changes it has in the course of ages undergone. That is to say, the distribution of forms in time is a subject so much connected with the distribution of forms in space, that the one can hardly be separated from the other. Granting this is a general truth, it must yet be acknowledged as a special fact, that in fossil birds we have as yet but scanty means of arriving at any precise results which will justify bold generalization in the matter of avine distribution. Remains of extinct birds are, compared with those of other classes of vertebrates, exceedingly scarce, and these have been found in very few, widely separated countries. The great problems involved in the study of geographical distribution must therefore be based mainly upon the other classes, both vertebrate and invertebrate, which, moreover, enjoy less great facilities of locomotion than the birds.

Yet it so happens that the great zoogeographical regions of the world, now more or less generally accepted, have been based upon the distribution of birds. The whole subject was properly introduced by Treviranus,^[1] who in his large philosophical work devotes considerable space to the “geographical distribution of animals.” Next we have to mention F. Tiedemann,^[2] the Heidelberg anatomist, who has been generally ignored, although he surpassed many a recent zoogeographer by the wide view he took of the problem; in fact he was the first to connect distribution with environmental or bionomic factors; e.g. the remark on p. 481 of his work that “the countries of the East Indian flora have no kinds of birds in common with America which are vegetable feeders.” L. K. Schmarda^[3] divided the land into twenty-one realms, characterizing these mainly by their birds. P. L. Sclater^[4] was the first to divide the world into a few great “regions,” the Palaearctic, Ethiopian, Indian and Australian forming one group, the “Old World” (Palaeogaea); and the Nearctic and Neotropical forming a second, the New World (Neogaea). Birds being of all animals most particularly adapted for extended and rapid locomotion, it became necessary for him to eliminate from his consideration those groups, be they small or large, which are of more or less universal occurrence, and to ground his results on what was at that time commonly known as the order Insessores or Passeres, comprehending the orders now differentiated as Passeriformes, Coraciiformes and Cuculiformes, in other words the mass of arboreal birds. His six main divisions—practically adopted by A. R. Wallace^[5] in his epoch-making work—are excellent, taken separately. They express the main complexes of land with their dependencies in well-chosen terms; for instance the “Neotropical region” stands short for South and Central America with the Antilles.

But these six divisions of Sclater and Wallace are not all equivalent, only some are of primary importance; they require co- and sub-ordination. This most important advance was made by T. H. Huxley.^[6] Some of the “regions” have now to be called subregions, e.g. the Nearctic and the Palaearctic. The reduction of the Oriental to a subregion, with consequent “provincial” rank of its main subdivisions, will probably be objected to, but these are matters of taste and prejudice. Above all it should be borne in mind that nearly all the last subdivisions or provinces are of very little real value and most of them are inapplicable to other classes of animals.

Besides some occasional references in the text, only a few more of the general works dealing with the distribution of birds can here be mentioned. Especial attention has to be drawn to the article “Geographical Distribution,” in Newton’s Dictionary of Birds. See also A. Heilprin, The Geographical and Zoological Distribution of Animals (New York, 1887); W. Marshall and A. Reichenow, two maps with much detail, although badly arranged, in Berghaus’ Physikalischer Atlas, pt. vi. (Atlas d. Thierverbreitung), (Gotha, 1887); A. Reichenow, “Die Begrenzung zoogeographischer Regionen vom ornithologischen Standpunkte,” Zoolog. Jahrb. iii., 1888, pp. 671-704, pl. xxvi.; E. L. Trouessart, La Géographie zoologique. (Paris, 1890).

The scheme adopted in the following account stands as follows:—

(A) Austrogaea or I. Australian Region		New Zealand subregion.     Australian”   Papuan”
(B) Neogaea or II. Neotropical Region		Antillean”   Columbian”   Patagonian”
(C) Arctogaea	III. Holarctic Region	Nearctic”   Palaearctic”
	IV Palaeotropical Region	Ethiopian”   Oriental”

In the following account the characterization of the various regions and subregions has to a very great extent been adopted from Newton’s article in his Dictionary of Birds, and from the chapter on distribution in the article on “Birds” in the Encyclopaedia Britannica, 9th edition. This applies especially to those instances in which the members of families, genera and species are mentioned. The families are those which are enumerated in Garow’s classification. The numbers of genera and species of birds are, of course, a matter of personal inclination. If we take a moderate computation the number of recent species may be taken at 10,000-11,000.^[7] Dr R. B. Sharpe increases their number to about 15,000 in the New Hand-List of Birds, published by the British Museum. In the first two volumes fossil birds, occasionally based upon a fragmentary bone only, are also included.

(A) Austrogaea, the Australian region in the wider sense, with the Papuan, Australian and New Zealand subregions, including also Polynesia. We may here quote Newton (Encyclopaedia Britannica, 9th ed., “Birds,” p. 738) on the remarkable differences between this region and the rest of the Old World:—“The prevalent zoological features of any Region are of two kinds—negative and positive. It is therefore just as much the business of the zoogeographer, who wishes to arrive at the truth, to ascertain what groups of animals are wanting in any particular locality (altogether independently of its extent) as to determine those which are forthcoming there. Of course, in the former case it would be absurd to regard as a physical feature of any great value the absence from a district of groups which do not occur except in its immediate neighbourhood; but when we find that certain groups, though abounding in some part of the vicinity, either suddenly cease from appearing or appear only in very reduced numbers, and occasionally in abnormal forms, the fact obviously has an important bearing. Now, mere geographical considerations, taken from the situation and configuration of the islands of the so-called Indian or Malay Archipelago, would indicate that they extended in an unbroken series from the shores of the Strait of Malacca to the southern coast of New Guinea, which confronts that of north Australia in Torres Strait, or even farther to the eastward. Indeed, the very name Australasia, often applied to this part of the world, would induce the belief that all the countless islands, be they large or small—and some of them are among the largest on the globe—were but a southern prolongation of the mainland of Asia. But so far from this being the case a very definite barrier is interposed. A strait, some 15 m. or so in width, and separating the two fertile but otherwise insignificant islands of Bali and Lombok, makes such a frontier as can hardly be shown to exist elsewhere. The former of these two islands belongs to the Indian Region, the latter to the Australian, and between them there is absolutely no true transition—that is, no species are common to both which cannot be easily accounted for by the various accidents and migrations that in the course of time must have tended to mingle the productions of islands so close to one another. The faunas of the two are as absolutely distinct as those of South America and Africa, and it is only because they are separated by a narrow strait instead of the broad Atlantic that they have become so slightly connected by the interchange of a few species and genera.

“Now, first, of the forms of birds which are prevalent throughout the Indian Region, but are entirely wanting in the Australian, we have at once the bulbuls (Ixidae), very characteristic of most parts of Africa and Asia, including the sub-group Phyllornithinae, which is peculiar to the Indian Region; the widely-spread families of barbets (Megalaeminae) and vultures (Vulturidae); and the pheasants (Phasianidae), which attain so great a development in various parts of the Asiatic continent and islands that there must their home be regarded as fixed. Some naturalists would add the finches (Fringillidae), rightly if we assume that the Ploceidae or weavers constitute a separate family. Then, of forms which are but weakly represented, we have the otherwise abundant thrushes (Turdidae), and, above all, the woodpeckers (Picidae), of which only very few species, out of 400, just cross the boundary and occur in Lombok, Celebes or the Moluccas, but are unknown elsewhere in the region.”

But the Australian region is also remarkable for its ornithic singularity. All the existing Ratitae (with the exception of the ostriches of Africa and South America, belonging to the genera Struthio and Rhea, and comprising at most but five species) are found in Austrogaea and nowhere else. Of the Passeres the honey-suckers (Meliphagidae) are most characteristic, and, abounding in genera and species, extend to almost every part of the region, yet only one species of Ptilotis oversteps its limits, crossing the sea from Lombok to Bali. Other peculiar families are much more confined. But the positive characteristics of the region as a whole are not its peculiar forms alone; there are at least four families which, being feebly represented elsewhere, here attain the maximum of development. Such are the thick-headed shrikes (Pachycephalidae), the caterpillar-eaters (Campephagidae), the flower-peckers (Dicaeidae), and the swallow-flycatchers (Artamidae). Besides these, three or perhaps four groups, though widely distributed throughout the world, arrive in the Australian region at their culmination, presenting an abundance of most varied forms. These are the weaver-birds (Ploceidae), and the moreporks (Podargidae), but especially the kingfishers (Alcedinidae) and the pigeons (Columbidae), the species belonging to the two last obtaining in this region a degree of prominence and beauty which is elsewhere unequalled.

The boundaries of the subregions are not well defined.

Fig. 22.—Extinct Phillip-Island Parrot (Nestor productus).
(From specimen in the British Museum. Reduced.)

The New Zealand Subregion, considered by Professors Newton and Huxley and various other zoogeographers as deserving the rank of a region, is, and to all appearance has long been, more isolated than any other portion of the globe. Besides the three larger islands numerous satellites belong to the subregion, as Lord Howe, Norfolk and Kermadec islands, with the Chatham, Auckland and Macquarie groups. The main affinities of the avifauna are, of course, Australian. The most extraordinary feature is unquestionably the former existence of the gigantic Dinornithes or moas (q.v.) and, another family of Ratitae, the weird-looking kiwis or Apteryges, which are totally unlike any other existing birds. Of other peculiar genera it will suffice to mention only the more remarkable. Rallidae present the very noteworthy woodhens, Ocydromus, and the takahe, Notornis, which is almost extinct. The widely-spread plovers, Charadriidae, have two not less singular generic developments, Thinornis, and the extraordinary wrybill, Anarhynchus. There is an owl, type of the genus Sceloglaux. Of parrots, Stringops, the kakapo or owl-parrot, is certainly peculiar, while Nestor constitutes a peculiar subfamily of the brush-tongued parrots or Trichoglossidae. Xenicus and Acanthositta form a little family of truly mesomyodean Passeres Clamatores. Of the Meliphagidae the genera Prosthemadera, Pogonornis and Anthornis are peculiar. The starlings, Sturnidae, are represented by Callaeas, Creadion and the very abnormal Heterolocha. The gallinaceous birds are represented by a quail, Coturnix novae zealandiae, now exterminated. A large flightless goose, Cnemiornis, allied to the Australian Cereopsis, and the gigantic rapacious Harpagornis, have died out recently, with the moas. In all, there is a wonderful amount of specialization, though perhaps in a very straight line from generalized forms; but the affinity to Australian or Polynesian types is in many cases clearly traceable, and it cannot be supposed but that these last are of cognate origin with those of New Zealand. A very long period of isolation must have been required to produce the differences so manifestly to be observed, but a few forms seem at rare intervals to have immigrated, and this immigration would appear to be kept up to our own day, as shown by the instance of Zosterops lateralis, which is said to have lately made its first appearance, and to have established itself in the country, as well as by the fact of two cuckoos, the widely-ranging Eudynamis taitensis and Chrysococcyx lucidus, which are annual visitors.

Polynesia forms, of course, part of Austrogaea. Its extent is so vast that it necessarily contains some peculiar, outlying forms, so to say forgotten, which in their long-continued isolation have specialized themselves. For instance, the kagu (Rhinochetus) of New Caledonia, a queerly specialized form with Gruine affinities pointing only to South America. The toothbilled pigeon (Didunculus) is restricted to Samoa. Most interesting is the avifauna of the Sandwich islands; entirely devoid of Psittaci and of Coraciiformes, these islands show an extraordinary development of its peculiar family Drepanidae, which are probably of South or Central American descent. Acrulocercus is a Meliphagine, and a peculiar genus. There are a raven (Corvus), a coot (Fulica), the well-known Sandwich island goose (Bernicla sandvicensis), now very commonly domesticated in Europe; and some flycatchers and thrushlike birds.

The Australian Subregion comprises Australia and Tasmania. In the north it is influenced, of course, by its proximity to Papuasia, whence there is a considerable admixture of genera which do not proceed beyond the tropics, and of these Casuarius is a striking example. The Cape York peninsula practically belongs to Papuasia. As a whole, Australia is rich in parrots, of which it has several very peculiar forms, but Picarians in old-fashioned parlance, of all sorts—certain kingfishers excepted—are few in number, and the pigeons are also comparatively scarce, no doubt because of the many arboreal predaceous marsupials. The continent, however, possesses the two important genera of the Pseudoscines, namely the lyre-birds (Menura) and the scrub-birds (Atrichia). Among the more curious forms of other land-birds may be especially mentioned the Megapodiidae, Lipoa and Talegallus, the rail Tribonyx and Pedionomus, which represents the otherwise palaeotropical Turnices in Australia. The presence of bustards (Eupodotis) is a curious example of interrupted distribution, since none other of the Otididae are found nearer than India. The Ratitae are represented by two species of emeu (Dromaeus), besides the cassowary of Cape York peninsula, and the extinct Dromornis and Genyornis with its enormous skull.

The Papuan Subregion, chiefly New Guinea with its dependencies, the Timor group of islands, the Moluccas and Celebes. On the whole its avifauna presents some very remarkable features. Its most distinctive characteristic is the presence of the birds of paradise, which are almost peculiar to it; for, granting that the bower-birds, Chlamydodera and others, of Australia, belong to the same family, they are far less highly specialized than the beautiful and extraordinary forms which are found, within very restricted limits, in the various islands of the subregion. Another chief feature is the extraordinary development of the cassowaries, the richness and specialization of the kingfishers, parrots, pigeons, honey-suckers and some remarkable flycatchers. It has several marked deficiencies compared with Australia, among which are the babblers (Timeliidae), weaver birds (Ploceidae), the Platycercinae among parrots, diurnal birds of prey and the emeus. As a whole, the birds of Papua are remarkable for their brilliance of plumage, or their metallic colouring. The birds of paradise, the racquet-tailed kingfishers, Tanysiptera, the largest and smallest of parrots, Calyptdrhynchiis and Nasiterna, and the great crowned pigeons, Goura, are very characteristic; and so are the various Megapodes.

(B) Neogaea, or the Neotropical region.—Excepting towards the north, where, in Mexico, it meets, and inosculates with the Nearctic subregion, the boundaries of the Neotropical region are simple enough to trace, comprehending as it does the whole of South America and all Central America; besides including the Falkland islands to the south-east and the Galapagos under the equator to the west, as well as the Antilles or West India islands up to the Florida channel.

Owing to the comparatively scanty number of harmful mammalian types, the birds play a considerable part in this large region, and some authorities consider its avifauna the richest in the world. The entire number of species amounts to about 3600. Of these 2000, or a good deal more than half, belong to the order Passeriformes. But the characteristic nature of the avifauna is more clearly brought out when we learn that of the 2000 species just mentioned only about 1070 belong to the higher suborder of Oscines, that means to say, nearly one-half belong to the lower suborder Clamatores. This is a state of things which exists nowhere else; for except in Australia, where a few indigenous and peculiar low non-Oscines are found, and in the Nearctic country, whither one family of Clamatores, viz. the Tyrannidae, has evidently been led by the geographical continuity of its soil with that of the Neotropical region, such forms do not occur elsewhere. Accordingly their disproportionate prevalence in South America points unerringly to the lower rank of the avifauna of the region as a whole, and therefore to the propriety of putting it next in order to that of the Australian region, the general fauna of which is admittedly the lowest in the world. Huxley has urged with his wonted perspicuity the alliance of these two regions as Notogaea, basing his opinion, besides other weighty evidence, in great measure on the evidence afforded by the two main sections of the Galli, viz. the Peristeropodes and the Alectoropodes, the former composed of the families Megapodiidae, almost wholly Australian, and the Cracidae, entirely Neotropical. (Cf. P.Z.S., 1868. pp. 294-319.)

Leaving, however, this matter as in some degree hypothetical, we have as genera, families, or perhaps even larger groups, a great many very remarkable forms which are characteristic of, or peculiar to, the Neotropical region in part, if not as a whole. Of families we find twenty-three, or maybe more, absolutely restricted thereto, besides at least eight which, being peculiar to the New World, extend their range into the Nearctic region, but are there so feebly developed that their origin may be safely ascribed to the southern portion of America. First in point of importance comes the extraordinarily beautiful family of humming-birds (Trochilidae), with nearly 150 genera (of which only three occur in the Nearctic region) and more than 400 species. Then the tyrants (Tyrannidae), with more than seventy genera (ten of which range into the northern region), and over 300 species. To these follow the tanagers (Tanagridae), with upwards of forty genera (only one of which crosses the border), and about 300 species; the piculules (Dendrocolaptidae), with as many genera, and over 200 species; the ant-thrushes, (Formicariidae), with more than thirty genera, and nearly 200 species; together with other groups which, if not so large as those just named, are yet just as well defined, and possibly more significant, namely, the tapaculos (Pteroptochidae), the toucans (Rhamphastidae), the jacamars (Galbulidae), the motmots (Monotidae), the todies (Todidae), the trumpeters (Psophiidae), and the screamers (Palamedeidae); besides such isolated forms as the seriema (Cariama), and the sun-bittern (Eurypyga).

The nature of the South American avifauna will perhaps become still more evident if we arrange the characteristic members as follows:—

1. Birds which are restricted to, probably indigenous of the region: Rhea; Palamedea and Chauna, the screamers; Tinami; Psophia, Dicholophus, Eurypyga, Heliornis of the Gruiform assembly; Thinocorys and Attagis; Cracidae; Opisthocomus; of parrots Ara and Conurus with their allies; Monotidae, incl. Todus; Steatornis; Galbulinae and Bucconinae; Rhamphastidae; Formicariidae, Pteroptochidae, and of the Tyrannidae the Cotinginae.

2. Birds which are indigenous, but extend far into North America: Cathartae, Trochilidae, Tyrannidae.

3. Birds which are originally immigrants from North America: Podicipedidae, with the flightless Centropelma on Lake Titicaca; Ceryle, the only genus of kingfishers in the New World; all the Oscines.

More or less cosmopolitan groups like herons, Falconidae, Anseres, Columbae, &c., and circumtropical families like Parridae, Trogonidae, Capitonidae, are to be excluded from these lists as indifferent. The differences between the Neotropical avifauna and that of North America are fundamental and prove the independence or superior value of the Neotropical region as one of the principal realms.

It is difficult to subdivide the Neotropical region into subregions; the best suggestion is that of Newton: Antillean, with the exception of the islands of Trinidad and Tobago, as well as those which lie on the northern coast of South America; Patagonian, including Chile and part of Peru; Columbian, comprising the rest of the continent and also Central America.

The Antillean Subregion is in many respects one of the most suggestive and interesting, comparatively small though it be. For narrow as are the channels between Cuba and the opposite coast of Central America, between the Bahamas and Florida, and between Grenada and Tobago, the fauna of the Antillean chain, instead of being a mixture of that of the almost contiguous countries, differs much from all, and exhibits in some groups a degree of speciality which may be not unfitly compared with that of oceanic islands. Except such as are of coral formation, the Antilles are hilly, not to say mountainous, their summits rising in places to an elevation of 8000 ft., and nearly all, prior to their occupation by Europeans, were covered with luxuriant forest, which, assisting in the collection and condensation of the clouds brought by the trade winds, ensured its own vitality by precipitating frequent and long-continued rains upon the fertile soil. Under such conditions we might expect to find an extremely plentiful animal population, one as rich as that which inhabits the same latitudes in Central America, not many degrees farther to the west; but no instance perhaps can be cited which shows more strikingly the difference between a continental and an insular fauna, since, making every allowance for the ravages of cultivation by civilized man, the contrary is the case, and possibly no area of land so highly favoured by nature is so poorly furnished with the higher forms of animal life. Here, as over so large a portion of the Australian region, we find birds constituting the supreme class—the scarcity of mammals being accounted for in some measure as a normal effect of insularity.

There is one peculiar subfamily, Todinae, represented by only four species of Todus. We note the absence of Ratitae, Tinami, Cracidae, Rhamphastidae, and any of those gruiform genera which are so characteristic of the continent. There is no family of birds common to the Nearctic area and the Antillean subregion without occurring also in other parts of the Neotropical region, a fact which proves its affinity to the latter.

The Patagonian Subregion, most extratropical, is naturally devoid of a good many typically tropical birds, or these are but poorly represented, for instance Caerebidae, Mniotiltidae, Tanagridae, Vireonidae. On the other hand some of the most characteristic features of the whole region are here well represented, e.g. Rhea, Tinami, Chauna, Dicholophus, Attagis, Pteroptochidae, and indeed therein we find some of the best evidence of the antiquity of its population, both recent and extinct (cf. the numerous fossils of the Santa Cruz formation), and also the nearest resemblance to the fauna of Austrogaea.

(C) Arctogaea is Huxley’s well-chosen term for all the rest of the world (including the Nearctic, Palaearctic, Indian and Ethiopian regions of P. L. Sclater) in opposition to Notogaea. Faunistically, although not geographically, the Nearctic and Palaearctic areas must form the two subdivisions of one great unit, for which the “Holarctic region” is now the generally accepted term.

The Holarctic Region, comprising North America and the extratropical mass of land of the Old World, may from an ornithological point of view be characterized by the Colymbi, Alcidae, Gallidae or Alectoropodous Galli, and the Oscines, which have here reached their highest development; while Ratitae, Tinami, Psittaci, and non-Oscine Passeres (with the exception of Tyrannidae extending into North America and Conurus carolinensis) are absent.

Nearctic Subregion.—The close affinity of North America with the Palaearctic avifauna becomes at once apparent if we exclude those groups of birds which we have good reason to believe have their original home in the Neotropical region, notably numerous Tyrannidae, humming-birds and the turkey-buzzards.

The following groups may be mentioned as characteristic and typically American, and, since we consider them as comparatively recent immigrants into the Neotropical region, as originally peculiar to the Nearctic area: Mniotiltidae, Vireonidae, Icteridae, Meleagris and various Tetraoninae. Restricted to and peculiar to the sub-region is only the little Oscine family of Chamaeidae, restricted to the coast district of California. “More than one-third of the genera of Nearctic birds are common also to the Palaearctic subregion. If we take the number of Nearctic species at 700, which is perhaps an exaggeration, and that of the Palaearctic at 850, we find that, exclusive stragglers, there are about 120 common to the two areas. Nearly 20 more are properly Palaearctic, but occasionally occur in America, and about 50 are Nearctic, which from time to time stray to Europe or Asia. This, however, is by no means the only point of resemblance. Of many genera, the so-called species found in the New World are represented in the Old by forms so like them that often none but an expert can distinguish them, and of such representative ‘species’ about 80 might be enumerated” (Newton, Dict. Birds, p. 335).

Of the many attempts to subdivide the Nearctic subregion, the same authority favours that of Dr S. F. Baird, who distinguishes between Canadian, Alleghanian, Middle or Missourian, Californian and Alaskan provinces. Dr Hart Merriam takes the broad point of view “that the whole of extratropical North America consists of but two primary life regions, a Boreal region, which is circumpolar, and a Sonoran or Mexican tableland region which is unique.” The first of these supports Newton’s contention of the essential unity of the Nearctic and Palaearctic areas. In any case the various Nearctic subdivisions completely merge into each other, just as is to be expected from the physical configuration and other bionomic conditions of the Nort American continent.

The Palaearctic Subregion is, broadly speaking, Europe and Asia, with the exception of India and China. The propriety of comprehending this enormous tract in one zoological “region” was first shown by Dr P. L. Sclater, and as regards the distribution of most classes of animals there have been few to doubt that it is an extremely natural one. Not indeed altogether so homogeneous as the Nearctic area, it presents, however, even at its extreme points, no very striking difference between the bulk of its birds. Though Japan is far removed from western Europe, and though a few generic forms and still fewer families inhabit the one without also frequenting the other, yet there is a most astonishing similarity in a large portion of their respective birds. In some cases the closest examination has failed to detect any distinction that may be called specific between the members of their avifauna; but in most it is possible to discover just sufficient difference to warrant a separation of the subjects. Nevertheless, it is clear that in Japan we have, as it were, a repetition of some of our most familiar species—the redbreast and the hedge-sparrow, for example—slightly modified in plumage or otherwise, so as to furnish instances of the most accurate representation, e.g. Cyanopica cooki of Portugal and Spain, and C. cyana of Amoorland and Japan.

Like the Nearctic the Palaearctic subregion seems to possess but one single peculiar family of land birds, the Panuridae, represented by the beautiful species known to Englishmen as the bearded tit-mouse, Panurus biarmicus. The entire number of Palaearctic families are, according to Newton, 67, and of the genera 323. Of these 128 are common to the Nearctic subregion. Species of 51 more seem to occur as true natives within the Ethiopian and Indian regions, and besides these 18 appear to be common to the Ethiopian without being found in the Indian, and no fewer than 71 to the Indian without occurring in the Ethiopian. To compare the Palaearctic genera with those of the Australian and Neotropical regions would be simply a waste of time, for the points of resemblance are extremely few, and such as they are they lead to nothing. It will therefore be seen from the above that next to the Nearctic are the Palaearctic has a much greater affinity to any other, a fact which might be expected from geographical considerations.

Having shown this much we have next to deal with the peculiarities of the vast Palaearctic subregion. At the lowest computation 37 genera seem to be peculiar to it, though it is certain that species of several are regularly wont to wander beyond its limits in winter seeking a southern climate. Of the peculiar genera only a few examples may be mentioned: Eurynorhynchus, the spoon-billed sandpiper of Siberia; Syrrhaptes, the sandgrouse of central Asia; Musicicapa of Europe.

We distinguish between a Siberian, Mongolian, Mediterranean and European province, none of which can be well defined. The islands of the Canaries, Madeira and the Azores belong to the Mediterranean province, and offer some peculiarities of great interest. The Azores have been monographed by F. D. Godman (Nat. Hist. of the Azores or Western Islands, London, 1870). There is a general tendency among these insular birds to vary more or less from their continental representatives, and this is especially shown by the former having always darker plumage and stronger bills and legs. In one instance the variation is so excessive that it fully justifies the establishment of a specific distinction. This is the case of the bullfinch of the more western of these islands (Pyrrhula murina), the male of which, instead of the ruddy breast of its well-known congener (P. vulgaris), has that part of a sober mouse-colour. A similar sombre hue distinguishes the peculiar chaffinch of the Canary Islands (Fringilla teydea), but to these islands as well as the Azores and Madeiras there belongs in common another chaffinch (F. tintillon) which, though very nearly allied to that of Mauritania (F. spodogenia) is perfectly recognizable, and not found elsewhere. Madeira has also its peculiar golden-crested wren (Regulus maderensis), and its peculiar pigeon (Columba trocaz), while two allied forms of the latter (C. laurivora and C. bollii) are found only in the Canaries. Further on this subject we must not go; we can only state that Godman has shown good reason for declaring that the avifauna of all these islands is the effect of colonization extending over a long period of years, and going on now.

Palaeotropical Region.—Much can be said in favour of combining the mostly tropical portion of the great mass of land of the Old World (excluding, of course, Austrogaea or the Australian region) into one region, for which Oscar Drude’s well-chosen term “palaeotropical” has been adopted (cf. Bronn’s Thierreich, System Part. p. 296, 1893). This region naturally comprises the African and Indian areas, conformably to be called subregions.

Both subregions possess, besides others, the following characteristic birds: Ratitae, viz. Struthio in Africa and Arabia, fossil also in the Sivalik Hills, and Aepyornithidae in Madagascar; Pittidae, Bucerotinae and Upupinae, of which Upupa itself in India, Madagascar and Africa; Coraciidae; Pycnonotidae or bulbuls; Trogonidae, of which the Asiatic genera are the less specialized in opposition to the Neotropical forms; Vulturidae; Leptoptilus, Anastomus and Ciconia among the storks; Pteroclidae; Treroninae among pigeons. Of other families which, however, extend their range more or less far into the Australian realm, may be mentioned Otididae, the bustards; Meropidae or bee-eaters; Muscicapidae or flycatchers; Sturnidae or starlings.

The Ethiopian Subregion comprises the whole of Africa and Madagascar, except the Barbary States, but including Arabia; in the north-east the subregion melts into the Palaearctic between its limits still farther to the eastwards, through Beluchistan and even beyond the Indus.

So large a portion of the Ethiopian subregion lies between the tropics that no surprise need be expressed at the richness of its fauna relatively to that of the last two subregions we have considered. Between fifty and sixty so-called families of land birds alone are found within its limits, and of them at least nine are peculiar; the typical genera of which are Buphaga, Euryceros, Philepitta, Musophaga, Irrisor, Leptosoma, Colius, Serpentarius, Struthio, Aepyornis. It is singular that only the first three of them belong to the order Passeriformes, a proportion which is not maintained in any other tropical region. The number of peculiar genera, besides those just mentioned, is too great for them to be named here; some of the most remarkable on the continent are: Balaeniceps, the whale-headed heron; Balaearica, the crowned crane; Podica, finfoot; Numida and allied genera of guinea fowls.

The natural division of the subregion is that into an African and a Madagascar province. Subdivision of the continental portion is beset with great difficulties, and none of the numerous attempts have proved long-lived. The forest-clad basin of the Congo, with the coastal districts of the bay of Guinea, seem to form one domain in opposition to the rest.

The Malagasy province comprises, besides Madagascar, the Mascarene, Comoro and Seyehelle islands. It may be safely deemed the most peculiar area of the earth’s surface, while from the richness and multifariousness of its animal, and especially of its ornithic population, New Zealand cannot be compared with it. In A. Grandidier’s magnificent Histoire physique, naturelle et politique de Madagascar, vol. xii. (Paris, 1875–1884), are enumerated 238 species as belonging to the island, of which 129 are peculiar to it, and among those are no fewer than 35 peculiar genera. Euryceros of the Oscines, and Philepitta of the Clamatores, are remarkable enough to form the types of Passeriform families, and Mesites half-way between Galli and Gruiformes is of prime importance. The Passerine Falculia, with its recently extinguished allies Fregilupus and Necropsar of the Mascarenes; the Coraciine Brachypteracias, Atelornis and Geobiastes, are very abundant, while Heliodilus is an owl belonging to that subfamily which is otherwise represented only by the widely-spread barn owl, Strix flammea. Lastly must be noted the extinct tall Ratite species of Aepyornis with its several fancy genera. But, as Newton charmingly puts it (Dict Birds, p. 353), the avifauna of Madagascar is not entirely composed of such singularities as these. We have homely genera, even among the true Passeres, occurring there—such as Alauda, Acrocephalus, Motacilla and Pratincola, while the Cisticola madagascariensis is only distinguishable from the well-known fan-tailed warbler, C. schoenicola of Europe, Africa and India by its rather darker coloration. But there are also species, though not Passerine, which are absolutely identical with those of Britain, the barn owl, common quail, pigmy rail, and little grebe or dabchick, all of them common and apparently resident in the island. Mauritius had the dodo (q.v.), Lophopsittacus and Aphanapteryx. Rodriguez had the solitaire, Necropsittacus and Necropsar. Bourbon or Réunion had Fregilupus.

Fig. 23.—Extinct Starling of Réunion (Fregilupus varius),
adapted from figures by Daubenton, Levaillant and others.
Reduced.

Some of the Malagasy avifauna is certainly ancient, aboriginal, and even points to India; other forms indicate clearly their African origin; while, lastly, such strikingly characteristic Indo-African birds as hornbills are unaccountably absent.

The Oriental Subregion comprises all the countries and numerous islands between the Palaearctic and Australian areas; it possesses upwards of seventy families, of which, however, only one is peculiar, but this family, the Eurylaemidae or broadbills, is of great importance since it represents all the Subclamatores. Of the many characteristic birds may be mentioned Pycnonotidae or bulbuls, of which the Phyllornithinae are peculiar, Campephagidae or cuckoo shrikes, Dicruridae or drongos, Nectariniidae or sunbirds; pheasants, together with Pavo and Gallus. Some of the similarities to the Ethiopian and the great differences from the Australian avifauna have already been pointed out. Naturally no line whatever can be drawn between the Oriental and the Palaearctic subregions, and many otherwise essentially Indo-Malayan families extend far into the Australian realm, far across Wallace’s line, whilst the reverse takes place to a much more moderate extent. Certainly the Oriental area, in spite of its considerable size, cannot possibly claim the standing of a primary region. It is a continuation of the great Arctogaea into the tropics.

Following H. J. Elwes we subdivide the whole subregion into a Himalo-Chinese, Indian and Malayan province. These divisions had the approval of W. T. Blanford, who proposed the terms Cis- and Trans-gangetic for the two first. The Himalo-Chinese or Trans-gangetic province shows the characteristics of its avifauna also far away to the eastward in Formosa, Hainan and Cochin China, and again in a lesser degree to the southward in the mountains of Malacca and Sumatra. Indo-China is especially rich in Eurylaemidae, China proper and the Himalayas in pheasants.

The Indian or Cisgangetic province is the least rich of the three so far as peculiar genera are concerned.

The Malayan province comprising the Malay islands, besides the Malay peninsula, and the very remarkable Philippines, possess an extraordinary number of peculiar and interesting genera.

The influence of the Australian realm is indicated by a Megapode in Celebes, another in Borneo and Labuan, and a third in the Nicobar islands (which, however, like the Andamans, belong to the Indian province), but there are no cockatoos, these keeping strictly to the other side of Wallace’s line, whence we started on this survey of the world’s avifauna.

D. Classification of Birds

Fürbringer’s great work, published in the year 1888 by the Natura Artis Magistra Society of Amsterdam, enabled Gadow not only to continue for the next five years the same lines of morphological research, but also further to investigate those questions which were still left in abeyance or seemed to require renewed study. The resulting “classification is based on the examination, mostly autoptic, of a far greater number of characters than any that had preceded it; moreover, they were chosen in a different way, discernment being exercised in sifting and weighing them, so as to determine, so far as possible, the relative value of each, according as that value may vary in different groups, and not to produce a mere mechanical ‘key’ after the fashion become of late years so common” (Newton’s Dictionary of Birds, Introduction, p. 103). It is not the quantity but the quality of the anatomical and bionomic characters which determines their taxonomic value, and a few fundamental characters are better indications of the affinities of given groups of birds than a great number of agreements if these can be shown to be cases of isomorphism or heterophyletic, convergent analogy. Nature possesses three great educational or developmental schools—terrestrial, aquatic and aerial life. Each of these affords animal, vegetable or mixed diet. Animal diet implies the greatest variety with regard to locality and the modes of procuring the food. Each of these schools impresses its pupils, in the case of the birds, with its own stamp, but there are many combinations, since in the course of phyletic development many a group of birds has exchanged one school for another. Originally terrestrial groups have taken to an entirely aquatic life, and vice versâ; others, originally endowed with the power of flight, have become, or are transforming themselves into, absolutely cursorial forms; some members of one group live entirely on seeds, while others have become fierce fishers, and so forth. Only by the most careful inquiry into their history can their relationship or pedigree be unravelled. A statement may now be given of Gadow’s classification of birds, in which the extinct forms have been intercalated so far as possible. The few characters assigned to the various groups are sufficiently diagnostic when taken together, although they are not always those upon which the classification has been established:—

Class AVES

I. Sub-class Archaeornithes.—The three fingers and their metacarpals remain separate, each with a claw. Well-developed remiges. Both jaws with alveolar teeth. Amphicoelous. Caudal vertebrae more than thirteen, without a pygostyle, but with about twelve pairs of rectrices. Archaeopteryx, A. lithographica, s. macroura, two specimens from the upper Oolite of Solenhofen, Bavaria.

II. Sub-class Neornithes.—Metacarpals fused. Second finger the longest. Not more than thirteen caudal vertebrae.

I. Division Ratitae.—Terrestrial, flightless. Without sternal keel. Quadrate bone with single proximal knob. Without pygostyle. Coracoid and scapula fused. Compound rhamphotheca. Adult without apteria. With copulatory organ. A collective polyphyletic or heterogeneous group, originally cosmopolitan; with certainty existing since the Miocene.

1. Order Struthiones.—With pubic symphysis. Two toes only, third and fourth. Struthio, ostrich, Pliocene of Samos and of north-west India, now Africa and Arabia.

2. Order Rheae.—With long ischiadic symphysis. Three toes. Mesembriornis, Miocene or Pliocene of Argentina. Rhea, South America.

3. Order Casuarii.—Three toes. Aftershaft as long as the other half. Casuarius and Dromaeus, Australian. Hypselornis, Pliocene of Sivalik Hills.

4. Order Apteryges.—Four toes. Bill long and slender. Apteryx, New Zealand.

5. Order Dinornithes.—Three or four toes. Bill short. Anterior limbs extremely reduced. Dinornis, numerous species, recently extinct, New Zealand.

6. Order Aepyornithes.—Aepyornis, recently extinct, Madagascar.

To the Ratitae belong possibly also the imperfectly known Diatryma, Eocene of New Mexico, Gastornis and Dasornis, Eocene of Europe, Genyornis, Pleistocene of Australia.

II. Division Odontolcae.—Marine, flightless, without sternal keel. Upper and lower jaws with teeth in furrows. Cretaceous epoch. Enaliornis, England, vertebrae chiefly biconcave; Hesperornis, North America, vertebrae heterocoelous.

III. Division Carinatae.—With keeled sternum.

1. Order Ichthyornithes.—Power of flight well developed. Vertebrae still amphicoelous. With small pygostyle. Incisura ischiadica. With alveolar teeth. Cretaceous of Kansas. Ichthyornis, Apatornis.

2. Order Colymbiformes.—Plantigrade, nidifugous, aquatic. All toes webbed, fourth largest, hallux short; metatarsus laterally compressed; tibia with high, pyramidal crest. Bill straight, pointed, with simple sheath.

Sub-order 1. Colymbi, Divers. Front toes completely webbed. Holarctic. Colymbus.

Sub-order 2. Podicipedes, Grebes. Toes lobated. Cosmopolitan.

3. Order Sphenisciformes.—Nidicolous, marine. Flightless, wings transformed into rowing paddles. Sphenisci, penguins. Antarctic and southern temperate coasts. Since the Eocene.

4. Order Procellariiformes.—Well flying, pelagic, nidicolous. Hallux absent or vestigial. Rhamphotheca compound. Cosmopolitan. Tubinares, petrels and albatrosses.

5. Order Ciconiiformes.—Swimmers or waders. Desmognathous, without basipterygoid processes; with one pair of sterno-tracheal muscles.

Sub-order 1. Steganopodes.—Well flying, aquatic, nidicolous; with all the four toes webbed together. Rhamphotheca compound; cosmopolitan. Phaëthon, tropic-bird; Sula, gannet; Phalacrocorax, cormorant and Plotus, snake-bird; Fregata, frigate-bird; Pelecanus. Here also Pelagornis, Miocene of France; Argillornis and probably Odontopteryx from the London Clay.

Sub-order 2. Ardeae.—Piscivorous, nidicolous, waders; with complicated hypotarsus and with long cervical apteria. Ardeidae, cosmopolitan; including Cancroma, Neotropical, Balaeniceps, Scopidae, Ethiopian. Proherodius, Eocene of England.

Sub-order 3. Ciconiae.—Zoophagous, nidicolous, waders; with simple hypotarsus and without cervical apteria. Cosmopolitan. Ciconiidae, storks. Ibidae, ibises and spoonbills. Propelargus, Oligocene.

Sub-order 4. Phoenicopteri.—Flamingos. Nidifugous, waders; with simple hypotarsus and without cervical apteria. Front toes completely webbed; hallux very short or absent; feed chiefly on small aquatic invertebrates. Phoenicopterus, cosmopolitan. Oligocene Elornis and, allied, Palaelodus.

6. Order Anseriformes.—Desmognathous, nidifugous; with two pairs of sterno-tracheal muscles, with complete basipterygoid processes and with a penis.

Sub-order 1. Palamedeae.—Screamers. Ribs without uncinate processes. Hypotarsus simple. Neotropical. Chauna, Palamedea.

Sub-order 2. Anseres.—Family Anatidae. Hypotarsus complex. Anser, Anas, Cygnus, since Miocene. Cnemiornis, Pleistocene, New Zealand, flightless.

7. Order Falconiformes.—Birds of prey. Carnivorous, desmognathous, nidicolous, without functional caeca. Terrestrial, aerial.

Sub-order 1. Cathartae.—American vultures. With nares perviae. Cathartes, turkey buzzards, Sarcorhamphus gryphus, condor Gypagus papa, king vulture.

Sub-order 2. Accipitres.—With nares imperviae. Serpentariidae, secretary-bird, Ethiopian; Miocene, France. Vulturidae, Old World vultures, excluding Australia. Falconidae, cosmopolitan, since the Eocene. Harpagornis, Pleistocene, New Zealand; Lithornis, Eocene, England. Pandionidae, ospreys or fish hawks, cosmopolitan.

8. Order Tinamiformes.—Nidifugous, with incisura ischiadica, without pygostyle. Herbivorous, terrestrial, neotropical. Crypturi, tinamous.

9. Order Galliformes.—Schizognathous, herbivorous, terrestrial. With ten functional remiges. With strong spinae sterni.

Sub-order 1. Mesites.—Without basipterygoid processes, and with large spina interna. Mesites, Madagascar.

Sub-order 2. Turnices.—Hemipodes or button-quails. Nidifugous; vomer large; sternum without processus obliqui. Hallux absent or vestigial. Old World. Turnix, Pedionomus.

Sub-order 3. Galli.—With large spina communis, and with large processus obliqui. Hallux functional. Megapodiidae, Australian region. Cracidae, curassows and guans, neotropical. Gallidae, cosmopolitan.

Sub-order 4. Opisthocomi.—Arboreal, with long spina externa; without basipterygoid processes. Opisthocomus hoatzin, Guiana, Venezuela and Amazon countries.

10. Order Gruiformes. Legs of the wading type. Without basipterygoid processes. Without spina interna. Nidifugous. Essentially Schizognathous. Rallidae, cosmopolitan, since Oligocene. Rallus, Fulica, Ocydromus, &c., Gallinula nesiotis, Tristan d’Acunha, flightless. Notornis, New Zealand, flightless, nearly extinct. Aptornis, New Zealand, flightless, extinct. Aphanapteryx (Mauritius) = Erythromachus (Rodriguez) = Diaphorapteryx (Chatham Island), flightless and recently extinct. Gypsornis, upper Eocene, France. Gruidae, cranes, cosmopolitan, allied Phororhacos, Tertiary of Argentina. Dicholophidae, cariamas, neotropical. Otididae, bustards, Old World. Rhinochetidae, kagus, New Caledonia. Eurypygidae, sun-bittern, neotropical. Heliornithidae, finfoots, tropical.

11. Order Charadriiformes.—Schizognathous. With eleven remiges, of which the terminal very short. Aquinto-cubital. Spinae sterni short, separate.

Sub-order 1. Limicolae.—Nidifugous, without spina interna sterni. Hypotarsus complicated. Charadriidae, plovers. Chionididae, sheath-bill. Glareolidae, wading swallows and coursers. Thinocorythidae, seed-snipes. Oedicnemididae, thick-knees. Parridae.

Sub-order 2. Lari.—Aquatic, vomer complete. Without basipterygoid processes. Front toes webbed; hallux small or absent. Large supraorbital glands. Since Miocene. Laridae, gulls, cosmopolitan. Alcidae, auks, northern half of periarctic region.

Sub-order 3. Pterocles.—Sand-grouse. Nidifugous. Vomer vestigial. With large crop and caeca. Hallux vestigial or absent since Oligocene. Africa to India, and Siberia. Pterocles and Syrrhaptes.

Sub-order 4. Columbae.—Pigeons. Nidicolous. Vomer vestigial. With large crop, vestigial caeca. Columbidae, cosmopolitan, since Miocene. Dididae, flightless, recently extinct. Didus, dodo, Mauritius. Pezophaps, solitaire, Rodriguez.

12. Order Cuculiformes.—Desmognathous, nidicolous; zygodactylous, or with the outer toe reversible.

Sub-order 1. Cuculi.—Cuckoos. Quinto-cubital. Cuculidae, cosmopolitan. Musophagidae, plantain-eaters and touracos, Ethiopian since Miocene.

Sub-order 2. Psittaci.—Parrots. Zygodactylous; aquinto-cubital. Cosmopolitan, chiefly tropical. Trichoglossidae, lories, Austro-Malayan. Nestor, New Zealand. Cyclopsittacus, Eos, Lorius, &c. Psittacidae, tongue smooth, incl. Stringops.

13. Order Coraciiformes.—Nidicolous. Nares imperviae, holorhinal. Downs restricted to the apteria or absent. Thirteen to fifteen cervical vertebrae. Mostly desmognathous. Deep plantar tendons connected with each other.

Sub-order 1. Coraciae.—Either (1) with long spina externa sterni, Coraciidae, rollers, Old World. Momotidae, neotropical, motmots and todies. Alcedinidae, kingfishers, cosmopolitan or (2) with long spina communis. Meropidae, bee-eaters, Old World. Upupidae, Upupinae, hoopoes: palaearctic and palaeotropical. Bucerotinae, hornbills, palaeotropical; Irrisorinae, woodhoopoes, Ethiopian.

Sub-order 2. Striges.—Owls. Outer toe reversible. Schizognathous. Long caeca. Flexor tendons normal. Hypotarsus simple. Cosmopolitan.

Sub-order. 3. Caprimulgi.—Nightjars. Nocturnal. With gaping mouth. Ten remiges and ten rectrices. Spinae sterni vestigial. Caeca functional. Steatornithidae, Steatornis, oil-bird or guacharo, South America. Podargidae, Australasian, Caprimulgidae, cosmopolitan.

Sub-order 4. Cypseli.—Tenth terminal remex the longest. With short spinae sterni. Without caeca. Cypselidae, swifts, cosmopolitan. Trochilidae, humming-birds, American.

Sub-order 5. Colii.—Mouse-birds. First and fourth toes reversible. Ethiopian.

Sub-order 6. Trogones.—Trogons. Heterodactyle, first and second toes directed forwards, third and fourth backwards. Tropical. Trogon gallicus, Miocene of France.

Sub-order 7. Pici.—Zygodactylous. Tendon of the flexor hallucis longus muscle sending a strong vinculum to that of the flexor profundus muscle, the tendon of which goes to the third toe only. Galbulidae, puff-birds and jacamars, neotropical. Capitonidae, barbets, tropical. Rhamphastidae, toucans, neotropical. Picidae, woodpeckers, cosmopolitan, excepting Madagascar and Australian region.

14. Order Passeriformes.—Nidicolous. Aegithognathous, without basipterygoid processes. Spina externa sterni large, spina interna absent. Quinto-cubital, toes normal. Apparently since the upper Eocene.

Sub-order 1. Passeres Anisomyodae.—Syrinx muscles entirely lateral or attached to the dorsal or ventral corners of the bronchial semi-rings, (1) Subclamatores. Deep plantar tendons connected by a vinculum. Eurylaemidae, broadbills, Indian and Indo-Malayan. (2) Clamatores. Deep flexor tendons not connected. Pittidae, palaeotropical. Xenicidae, New Zealand. Tyrannidae, American, Formicariidae, Pteroptochidae, neotropical.

Sub-order 2. Passeres Diacromyodae.—Syrinx muscles of either side attached to the dorsal and ventral corners of the rings. Hallux strong, with a large claw, (1) Suboscines with Menura, lyre-bird, and Atrichia, scrub-bird, in Australia. (2) Oscines, the true singing-birds, with more than 5000 recent species, are mostly divided into some thirty “families,” few of which can be defined.

The fourteen orders of the Carinatae are further congregated into four “Legions”:—

 I. COLYMBOMORPHAE ＝ Ichthyornithes＋Colymbiformes＋Sphenisciformes＋Procellariiformes.

 II. PELARGOMORPHAE ＝ Ciconiiformes＋Anseriformes＋Falconiformes.

III. ALECTOROMORPHAE ＝ Tinamiformes＋Galliformes＋Gruiformes＋Charadriiformes.

IV. CORACIOMORPHAE ＝ Cuculiformes＋Coraciiformes＋Passeriformes.

These four legions are again combined into two “Brigades,” the first of which comprises the first and second legions, while the second brigade contains the third and fourth legions.

Thus the whole classification becomes a rounded-off phylogenetic system, which, at least in its broad outlines, seems to approach the natural system, the ideal goal of the scientific ornithologist. The main branches of the resultant “tree” may be rendered as follows:—

The Odontolcae seem to be an early specialized offshoot of the Colymbo-Pelargomorphous brigade, while the Ratitae represent a number of side branches of early Alectoromorphae. The Ratitae branched off, probably during the Eocene period, from that still indifferent stock which gave rise to the Tinami＋Galli＋Gruiformes, when the members of this stock were still in possession of those archaic characters which distinguish Ratitae from Carinatae. It follows that new groups of Ratitae can no longer be developed since there are no Carinatae living which still retain so many low characters, e.g. configuration of the palate, precoracoid, pelvis, intestinal convolutions, copulatory organ, &c. Loss of the keel is co-ordinated with the power of using the forelimbs for locomotion; although a “Ratite” character, it is not sufficient to turn a Notornis, Cnemiornis or Stringops, not even a Phororhacos into a member of the Ratitae.

Another branch of the Alectoromorphae, in particular of the Galliformes, when these were still scarcely separated from the Gruiformes, especially rail-like birds, leads through Opisthocomi to the Cuculiformes. These are, again in an ascending direction, connected with the Coraciiformes, out of which have arisen the Passeriformes, and these have blossomed into the Oscines, which, as the apotheosis of bird life, have conquered the whole inhabitable world.  (H. F. G.) 

---

1. Treviranus, Biologie oder Philosophie der lebenden Natur, vol. ii. cap. 4, § 2 (Göttingen, 1803).
2. F. Tiedemann, Anatomie und Naturgeschichte der Vögel, vol. ii §§ 127-255 (Heidelberg, 1814).
3. L. K. Schmarda, Die geographische Verbreitung der Thiere (Wien, 1853).
4. P. L. Sclater on the general geographical distribution of the members of the class “Aves,” 2. Linn. Soc. ii. pp. 130-145, 1858.
5. A. R. Wallace, The Geographical Distribution of Animals, with a study of the Relations of Living and Extinct Faunas as elucidating the Past Changes of the Earth’s Surface, 2 vols. (London, 1876).
6. T. H. Huxley, “On the Classification and Distribution of the Alectoromorphae,” P.Z.S., 1868, pp. 313-319.
7. The following old-fashioned rough computation may serve as an indication of the relative size of the orders and suborders of recent birds:—
   
   

   Ratitae Colymbiformes Sphenisciformes Procellariiformes  Ciconiiformes Anseriformes Falconiformes Tinamiformes Galliformes

   20 20 15 90 150 150 360 40 370

   Gruiformes   Charadriiformes   Cuculiformes   Coraciiformes   Passeres Clamatores   Passeres Oscines Total about

   250 650 600 1600 1000 5000 ——— 10,300

   (incl. Columbae 350) (incl. Psittaci 400) (incl. Trochili and Pici)  species