| PERCY R. LOWE: |
| On the Systematic Position of the Swifts (Suborder Cypseli) and Humming-Birds (Suborder Trochili), with special reference to their Relation to the Order Passeriformes |
|
Extract Part I. The Swifts (Cypseli) Introductory In so far as I have been able to ascertain from the literature on the subject (see „references") no very definite or convincing evidence has been brought forward to settle if possible and once and for all the position held by the Swifts and Humming-Birds in the class Aves. I might be accused of optimism if I thought that I had succeeded where others had failed; but having lately spent a good deal of time studying the anatomy of these two groups in relation to one another, and to the order Passeriformes, it occurred to me that these notes might at any rate be a useful contribution towards the solution of the problem. I do not think, however, it would serve a very useful purpose to enter here into a long review of the anatomical papers that have been written on the subject, and I shall accordingly spare the reader by limiting myself to a few brief remarks thereon. Garrod (1877), in a paper on the anatomical characters distinguishing the Swallow and the Swift, published in the 'Zoologist' and not included in his "collected papers," says the Swift is not Passerine. If we hold the view, as I do, that the Passerine birds constitute a suborder Passeres of the order Passeriformes, while the Swifts (Cypseli) constitute another suborder, this is undoubtedly true - as true in fact as that the Swallows are Passerine. But this merely makes the Swifts Passeriform, although not Passerine. Dr. Phillip Sclater (1865) says "the Swifts have no relationship whatever with the Swallows": Kitchin Parker (1889), in a paper in the 'Zoologist' on "The Systematic Position of the Swifts (Cypselidæ)," remarks that this statement is too emphatic to be true. It would not be true, he says, of the relationship of the Swifts to any Passerine bird; for they certainly lie on the Passerine border of the Picariæ, if they cannot be put even as abnormal Coracomorphæ. Shufeldt (1885, p. 914) says, "The Swifts are essentially modified Swallows, and, as the family Cypselidæ they belong in the order Passeres, next to that group." Later on (1889) he puts them in a separate order. As far as palæontological evidence goes (see below), my own view is that the Swifts (and Humming-Birds) probably specialised from the generalised Passeriform stock long before the Swallow became slightly isolated from the purely Passerine group. In the same paper, as quoted above, Kitchin Parker (1889) says. "I agree with my friend Dr. Shufeldt that the Swallow and the Swift are near akin "- a statement, however, which I, personally, think cannot be baldly made without definite qualifications; he goes on to say that "all the Passerine birds are 'Finch-jawed,' that is to say Ægithognathous. No other birds but the Swifts are so, except in an imperfect degree." By which last sentence, of course, he was referring to such generalised Plover types as the Seed-Snipe (Thinocorythidæ), and to the Turnicomorphs. In the same paper Kitchin Parker says "as in all Passeres, the Swifts and Humming-Birds have no second phalanx on the pollex, nor a third on the index." He also thinks the Swallows are very isolated in their own proper suborder of Passeres; and bearing on this notes that "among all Passeres and related types a Swallow is the only bird in which I have found a second or ungual phalanx to the pollex. "This was in the Sand-Martin (Cotyle riparia), the condition being well seen in the embryo and not quite obscured in the adult. Kitchin Parker (1878), again, in a paper on the anatomy of the Swift published in the 'Transactions of the Zoological Society,' had previously, in my opinion, come nearer than anyone else to placing it in its proper position; but, strangely enough, in his final summing up, concludes that "the Swift is formed, as it were, by a commingling of Passerine and Caprimulgine characters; but in it the latter preponderate sufficiently to exclude it from the territory of the Coracomorphæ." I have recently studied the anatomy of the Goatsucker very closely in order to familiarise myself with Caprimulgine characters. There is no shadow of doubt that the characters are almost uniformly non-Passerine (cf. list of Caprimulgine characters, with list of Passerine characters, and with those of the Swift, below). In this list of twenty or more characters there is barely one which we could regard as being indicative of Passerine affinities. In point of fact, the characters might very well pass for those of some Limicoline bird, such as a Woodcock. Yet, if we take at random half a dozen systematic lists of the class Aves emanating from as many authors, we shall find the Caprimulgidæ closely associated with the Swifts. It is true, of course, that in the general form of the skull of the two birds there are superficial resemblances doubtless due to the fact that both procure much the same, kind of food in much the same way, and for this purpose require a very wide gape to the bill. When, however, we come to make a close examination of such outstanding structures in the skull as the vomer, maxillo-palatines, lacrymals, pterygoids, and quadrates the most conspicuous differences are revealed, although it was impossible to incorporate an account of these in the above tables. So different indeed are they that, in my opinion, we may feel certain that the two forms are not in any way comparable as regards affinities; while Parker's statement that the Swift has so many Caprimulgine characters that its classification with the Coracomorphæ is precluded seems to me to be wholly inexplicable. The Caprimulgine characters as revealed by a study of the twenty-two given in the table below, to say nothing of skull-characters, are, as I have said, beyond doubt non-Passerine. If compared with those given in the table for the Swift (or the Humming-Bird) they are found, with only one or two exceptions of minor importance, to be in complete disagreement. If then these characters have any value, and I may say I have searched in vain for others, the question may be fairly put, How can the Swift and Humming-Bird be regarded any longer as being non-Passeriform? Coming to a consideration of the works of purely systematic authorities of recent or fairly recent date, I find that Swifts and Humming-Birds are invariably treated as non-Passerine birds, being referred either to an order Macrochires equivalent to Huxley's Cypselomorphæ or considered as separate families under some such order as the Coraciiformes. Moreover, in all these works the Passeres are considered either as an order in themselves or as a suborder Passeres of an order Passeriformes in association with another suborder Eurylaimi or Broadbills. After a rather intensive anatomical study of the Passeres, the Swifts, the Humming-Birds, and such other groups as the Woodpeckers and Coraciiform birds, carried out by means of my own dissections and on a rigid plan which accepted for granted not a single statement derived from the work of others, I may so far anticipate my findings by saying that I have come to the conclusion that all the evidence seems to point to the fact that the Swifts and Humming-Birds should be included in the order Passeriformes as separate suborders Cypseli and Trochili, along with the suborders Passeres, Eurylaimi, and Pici. Whether further subordinal additions to the Passeriformes will be indicated by the evidence I am not yet in a position to suggest, but hope that my present studies of the Coraciiform birds, will throw light on the problem. In any case in this present paper I only propose to deal with the Cypseli and Trochili; while discussion on the Pici and other groups will be postponed for another occasion. In the meantime, my present view of the composition of the Passeriforms based on my own dissections and independent investigations may be set forth as follows :- Order PASSERIFORMES Suborders (1) Passeres (2) Eurylaimi (3) Cypseli (4) Trochili
(a) Capitonidæ
TABLE OF CHARACTERS Order PASSERIFORMES Suborder Passeres (1) Ægithognathous palate. (2) Transpalatine processes present. (3) Sternum with a single notch on each side. (4) A bifurcate manubrium. (5) Hallux alone directed backwards. (6) A nude oil-gland. (7) Short simple colic cæca, of a very characteristic nature. (8) A characteristic Passerine pterylosis. (9) Characteristic Passerine structure of barbules at base of contour-feathers. (10) Secondaries with major under wing-coverts absent and lesser closely linked with median (PI. I. fig. 3). (11) No vinculum between Flexor longus hallucis and Flexor cummunis digitorum. (12) Rectrices twelve (ordinary rule), but variations from none to sixteen. (13) A characteristic Passeriform insertion of Tensor patagii brevis. (14) Thigh muscle formula (Garrod) A.XY. (15) Ambiens absent. (16) Deltoid muscle-scapular portion with low insertion to ectepicondylar process. A low insertion is a Passerine feature. (17) Dermo-tensor patagii (of Shufeldt) joins Tensor patagii longus. (18) No biceps slip. (19) No expansor secundariorum. (20) One carotid only present-the left. (21) Syrinx - acromyodian, mesomyodian, or tracheophone. (22) Secondary wing-feathers quinto-cubital. Suborder Eurylaimi (1) Ægithognathous palate. (2) Transpalatine processes present. (3) Sternum with a single notch on each side. (4) Manubrium not bifurcate. (5) Hallux alone directed backwards. (6) A nude oil-gland. (7) Short simple colic cæca - typically Passerine. (8) A characteristic Passerine pterylosis. (9) Characteristic Passerine structure of barbules at base of contour-feathers. (10) Secondaries with major under wing-coverts present, but undeveloped (yet much larger than in normal Passerines, where they are absolutely rudimentary if present). (11) A vinculum present (cf. Passeres). (12) Rectrices twelve. (13) A Passeriform insertion of Tensor patagii brevis. (14) Thigh muscle formula (Garrod) A.XY. (15) Ambiens muscle absent. (16) Deltoid muscle - scapular head low insertion to just above ectepicondylar process. (17) Dermo-tensor patagii (of Shufeldt) does not join Tensor patagii longus. (18) No biceps slip. (19) No expansor secundariorum. (20) One carotid only-left. (21) Syrinx tracheo-bronchial, mesomyodian. (22) Secondary wing-feathers quinto-cubital. Suborder Cypseli Cypselus apus apus (1) Ægithognathous palate (Pl. I. fig. 1). (2) Transpalatine processes present. (3) Sternum with single notch on each side filled up by bony extension. (4) Manubrium not bifurcate - but potentially so. (5) Hallux and foot Passerine in Hemiprocne and Hirundapus; Cypseline in other genera. (6) A nude oil-gland. (7) No colic cæca; a pure insect-eater, therefore presumably lost or never developed. (8) Pterylosis a specialized Passerine one (see Pl. 1. fig. 2). (9) Some Swifts (unspecialized) with Passerine structure of barbules; others without. (10) Secondaries with major under wing-coverts absent, median enlarged. Only one row of reversed under wing-coverts (Pl. II. fig. 7). (11) No vinculum present ; but arrangement differs from Passerine morphology. (12) Rectrices 10, even in Hemiprocne and Hirundapus. (13) Characteristic Passerine insertion of Tensor patagii brevis in Hemiprocne and Hirundapus, in other specialized Swifts a specialized arrangement (see Pl. III. figs. 2 & 3 for Hirundapus and Hemiprocne). (14) Thigh muscle formula A, X. (15) Ambiens absent. (16) Deltoid muscle extremely reduced, but insertion Passerine. (17) Dermo-tensor patagii in specialized Swifts feebly developed or absent, but M. pectoralis propatagialis longus present. In Hemiprocne comata, on the other hand, it is present. (18) No biceps slip, and no biceps muscle except in Hemiprocne. (19) No expansor secundariorum. (20) One carotid only: except in Cypseloides (2). (21) Syrinx-tracheo-bronchial. (22) Secondary wing-feathers quinto-cubital (I have a note stating that I find Hemiprocne comata is quinto-cubital. This is a curious condition because whenever Hemiprocne departs from the Cypseline mode it favours the Passeres. It is, however, correct.). Suborder Trochili (1) Ægithognathous palate (see notes and illustrations). (2) Transpalatine processes ill-developed - cartilaginous. (3) Sternum with no distal notch on either side - complete ossification. (4) Manubrium not bifurcate or only potentially. (5) Hallux alone directed backwards, but feet only pseudo-Passerine. (6) A nude oil-gland. (7) Cæca absent. (8) A Passerine pterylosis with slight modifications. (9) Characteristic Passerine structure of barbules at base of contour-feathers. (10) Secondaries with major under wing-coverts absent. (11) A vinculum present, although in other respects morphology Passerine. (12) Rectrices ten. (13) Insertion of Tensor patagii brevis highly specialized as in majority of Swifts, but Passerine in plan. (14) Thigh muscle formula (Garrod) A. (15) Ambiens absent. (16) Deltoid muscle - both scapular and humeral portions enormously reduced or absent. The scapular portion if present has a low insertion on or about the ectepicondylar process. (17) Dermo-tensor patagii (of Shufeldt) apparently absent. (18) No biceps slip. No Biceps humeralis! (19) No expansor secundariorum. (20) One carotid only - left. (21) Secondaries quinto-cubital. (22) Syrinx oscinine. Family CAPRIMULGIDÆ A Table of Characters to show its non-Passeriform Nature (cf. text). (1) Pseudo-desmognathus, desmognathous, or even doubly desmognathous in some genera. (2) No Transpalatine processes. (3) Sternum with a single notch (single in some genera, double in other genera). (4) No bifurcate manubrium (spina externa vestigial). (5) Feet non-Passerine, long middle toe; toes definitely webbed in the angles between them. (6) A nude oil-gland, (7) Colic cæca very long - un-Passerine (In an example of C. europæus I found the cæca measured 31 mm. in length. Their distal ends were curiously dilated and they had much the same disposition as in a Wader. In a Passerine bird they are very short and closely attached to the walls of the gut. The cæca of a Hooded Crow, a far larger bird than a Nightjar, measured only 14 mm. in length, and their morphology was totally dissimilar to that seen in the cæca of a Nightjar.). (8) Pterylosis - non-Passerine. Dorsal tract forked as in a Wader. (9) Structure of barbules at base of contour-feathers non-pennaceous (Not typically Passerine, as on one side of rachis barbs seen running into pennaceous barbules quicker than on the other.). (10) Secondaries with major under wing-coverts present and functioning. Median also present. Both coverts reversed. (11) Vinculum present between deep flexors (specialized arrangement, cf. Garrod). (12) Rectrices 10. (13) Insertion of Tensor patagii brevis non-Passerine, very like a Wader; a patagial fan present. Tensor patagii longus like a Wader. (14) Thigh muscle formula of Garrod A.XY. (15) Ambiens absent. (16) Deltoid muscle - high insertion, non-Passerine. (17) Dermo-tensor patagii of Shufeldt does not join Tensor patagii longus. (18) A biceps slip present. (19) No expansor secundariorum (present in Steatornis, fide Beddard). (20) Two carotids present. (21) Basi-pterygoid processes present. (22) Syrinx-mesomyodian, Tracheo-bronchial (*Bronchial in some genera). (23) Secondaries aquinto-cubital. It will therefore be observed that in almost every character given above the Caprimulgidæ differ from those tabulated under the Passeres, Eurylaimi, Cypseli, and Trochili. How then can the Cypseli (as stated by Kitchin Parker) be said to possess so many Caprimulgine characters as to preclude their being included in the Passeriformes ? (Cf. list of Cypseline characters.) A COMPARATIVE LIST OF CHARACTERS RELATING TO PASSERES, EURYLAIMI, CYPSELI, AND TROCHILI In order to justify the above conclusions, which are pretty certain to be regarded as unwelcome and too revolutionary, it was first essential to determine the characters of the Passeres. I have accordingly jotted down twenty or more characters which seem to me to definitely nail this highest suborder of Passeriform birds to the counter: and this I have done as the result of madissections of the various types characteristic of the groups into which the suborder Passeres may be divided. These characters will be found tabulated above. I have also added a table of corresponding characters in the suborder Eurylaimi (Broadbills) and in the Caprimulgidæ (Goatsuckers) to demonstrate in the latter its non-Passerine characters. Having done this I subjected the Swifts and Humming-Birds to exactly the same test; and if a comparative examination of the tables be made I think it will probably be granted that there is a very great deal of similarity between the characters distinctive of the Passeres, on the one hand, and the Swifts and Humming-Birds, on the other; while there is no such outstanding dissimilarity as would justify the view that one of these, the Passeres, was Passeriform and the other two, the Swifts and Humming-Birds, non-Passeriform. It will be clear, in fact, to anyone who will compare these lists, or dissect a fair series of Swifts and Humming-Birds of different genera, that where dissimilarities occur they are due to the extreme degree of specialization to which both Swifts (not all) and Humming-Birds have been subjected, especially as regards the wing-bones and wing-muscles, the differing morphology of which cannot, of course, be fully indicated in a mere list of characters. I now propose to consider these Cypseline and Trochiline characters in greater detail, taking each suborder separately and also discussing later on some of the outstanding muscular and osteological features. For this purpose the following genera have been dissected by me: Cypselus apus apus, - sladenis, Hirundapus giganteus, Chætura leucopygialis, Collocalia francica spodiopyga, - vanikorensis , Cypsiurus parvus myochrous, Tachornis balassiensis infumatus, Hemiprocne comata, while other specimens in the osteological collection (Brit. Mus.) have been examined. A REVIEW OF THE CYPSELINE CHARACTERS TABULATED BEFORE; AND EVIDENCE OF THEIR PASSERIFORM NATURE In this review I enlarge on each Cypseline character given in the list (above). (1) Ægithognathous Character of Palate.-As regards the character of this there can be no dispute; and, so far as I am aware, there never has been any dispute. Moreover, it is the ægithognathism of the complete kind peculiar to the higher Passeriform suborder of birds (Passeres). As Kitchin Parker has stated (see above) no other birds but the Passeres and Swifts have this kind of palate in so perfect a degree (Pl. I. fig. 1). The importance of this character hardly requires emphasis. (2) Transpalatine Processes.-These are conspicuously well developed in the Swift. They represent a character which is almost a hall-mark among the Passeres. In the 'Zoologist' of 1889, Kitchin Parker remarks in connection with the transpalatine processes that " In Passerine birds and in the Swifts and in no other birds (italics mine) a large remnant of the old Ichthyopsidan cartilaginous palatine is developed postero-laterally to the main bony bar, becomes ossified independently, and then becomes fused with that bar and forms its projecting part or apophysis." This is even plainly visible in an adult example of a Swift (Cypselus) which I caught on a fly when fishing and subsequently dissected. I mention it as it was probably a bird of the year.
It may be noted that the Manubrium is not bifurcate in the Eurylaimidæ, which are generally accepted as Passeriform. In both the Swifts and Humming-Birds the coraco-sternal joints are specialized, more so in the latter. Hilzheimer thinks that in the Humming-Bird this specialization is associated with the rapidity of the wing-strokes. The morphology of the coracoid itself is widely different in the two groups. (5) The Feet.-As is well known, all four toes in Cypselus and other Cypseline genera are directed forwards, and a good deal of importance seems to have been attached to this Cypseline specialization; but the fact that the condition obtaining in Hemiprocne, Hirundapus, and Collocalia is typically Passerine points to the fact that the condition in Cypselus and other Cypseline genera is a secondary specialization.
(7) There are very typical colic cæca in the Passeres -in fact, their morphology, size, and disposition are so constant as to constitute a Passerine hall-mark. In the Swifts colic cæca are entirely absent. I have not found them even in Hemiprocne, which in some respects, as we shall see, is an intermediate Cypselo-Passerine form. It would be easy by way of explanation to point to the fact that Swifts are purely insect-eaters, live an entirely aerial existence, and therefore have lost them. This would be plausible, since in vegetable-eaters like Grouse the cæca are highly developed. Unfortunately, Swallows, which feed in just the same way as the Swifts, have them developed in the typical Passerine way, and in every other anatomical respect cannot be divorced from the suborder Passeres. (8) The Pterylosis of the Swifts is essentially Passerine, although of a low or generalized type (cf. Pl. I. fig. 2). There are obvious Cypseline modifications, such as the apterion at the back of the neck (marked with a star) and the tracts on the vertex but nothing more. Both dorsal and pectoral tracts are fundamentally unlike a non-Passerine bird. (9) The Structure of the Barbules in Barbs taken from the Base of a Contour-feather.-I have found that if in the Passeres we take a barbule from a barb springing from the basal third of a contour-feather (I have generally taken a feather from the mid-dorsal tract) we find that this barbule has a non-pennaceous or pennal-down-like structure. If, on the other hand, we take a similar barbule from a non-Passerine bird we find that its structure is pennaceous (Whether this is an invariable rule I cannot say, but I have tested it in a very large number of families and genera.). Applying the test to the Swifts, I find that in unspecialized Swifts like Hemiprocne comata the structure of the barbules is non-pennaceous or Passerine-like, whereas in a specialized Swift like Cypselus apus apus or C. apus melba it is pennaceous. On the other hand, there are some Swifts like Hemiprocne coronata which seem to be referable to neither category, but to be on the border-line. Thus on one side of the barb (distal or proximal) the barbules may be definitely non-pennaceous; on the other "just beginning" to be pennaceous, the bases of the barbules being long and well developed. I would not have included this new Passerine character (if it is one) in my list had it not been for the fact that just as Swifts exhibit specialized and unspecialized modifications of their feet, so apparently, we find the same thing as regards their feathers, or, as we shall find below, as regards their wing-bones and muscles. (10) The Morphology of the Secondary Under Wing-coverts in Passerine and non-Passerine Birds.-As is well known, the major and median under wing-coverts of the secondary wing-feathers of a bird, when pre, are reversed. I have never been able to find a bird of the suborder Passeres (and this applies also to the suborder Eurylaimi) in which the secondary major under wing-coverts were not either absent or so reduced as to be completely rudimentary or entirely functionless. In the Passeres (sensu stricto) I have discovered that the median under wing-coverts, in addition to being reversed, are closely "coupled up" with the adjoining row of lesser under wing-coverts (unreversed). This close linkage of median and lesser coverts by means of elastic ligaments and unstriated muscles in these two rows of coverts is, so far as my investigations have gone, very characteristic, if not diagnostic, of the Passeres. On dissection both median and lesser coverts are found to have a "free-existence" in the sub-epidermic tissues - that is to say, they are really epidermic structures and probably activated by non-striated muscles which either pass from one feather to another, or from one row of feathers to another row, or from the coverts to the deep fascia or aponeurotic sheath of the Flexor carpi ulnaris muscle, where the unstriated muscles get their attachment or purchase. I shall not here attempt to describe the musculature of these linked feathers in greater detail, but refer the reader to a drawing by Mr. Frohawk from a dissection I made of the condition seen in a Crow (see Pl. I. fig. 3) If now we take a non-Passerine bird we find that the general rule is for the secondary major under wing-coverts to be present and well developed (see Pl. I. fig. 4). Instead, however, of being freely movable structures attached to the subepidermal tissues a "non-Passerine" major under wing-covert is fixed and closely bound for the, whole, length of its calamus to the under side of the secondary wing-feather, though more to one side than the other according as to whether it is situated on the right or left wing. A glance at fig. 4 will also disclose the fact that the proximal end of the calamus of the coverts lies so far forward along the length of the basal end of the secondary wing-feather as to be overlapped and hidden by a muscular slip from the post-axial portion of the Flexor carpi ulnaris muscle (see Pl. I. fig. 4). Passing still further forward in company with its remex, the proximal end of the calamus of the major wing-covert, in a non-Passerine bird like a Wader (Tringa ochropus) overpasses the deep flexor muscles until its tip ends in close apposition with the ulnar to which it has fibrous attachments. Thus the origin of the major under wing-covert is sunk deep among the flexor muscles and tissues of the forearm. By these differences in morphological characters and anatomical behaviour may a major under wing-covert of a secondary wing-feather be distinguished with certainty from a median under wing-covert and so far as I know the point has not been made before. It is true that a major under wing-covert can often be distinguished with much uncertainty from a median, or a median from a major, when an attempt is made to do so without recourse to dissection. The trouble is, however, that in the case of a bird like a Swift, or a Humming-Bird, the secondary median under wing-coverts may be so specialized that they seem to have assumed the proportions and functions of the major, and even with dissection I found it difficult for some time to convince myself of their nature. In a Swift or a Humming-Bird, for instance, there is only one row of reversed under wing-coverts. They are not bound down to their respective remiges, they are not overlapped by the slips from the post-axial portion of the Flexor carpi ulnaris to the secondary remiges, nor does the proximal (lower umbilical) end of the calamus surpass the hinder edge of the post-axial portion of the Flexor carpi ulnaris or become attached to the ulnar bone. Thus in the Swift, and I may also add the Humming-Bird, the feathers of this one row of reversed under wing-coverts have a much freer existence than is the case with the major coverts in a non-Passerine bird: and I am driven to the conclusion that they are merely specialized median coverts, and that in the Swifts (and Humming-Birds) the secondary major under wing-coverts are absent as in the Passeres. This view, moreover, is supported by the discovery that the major under wing-coverts are also absent on the primaries, the only row of reversed under wing-coverts in the primaries being closely linked up with the lesser wing-coverts and therefore certainly median. If, therefore, the major under wing-coverts of the primaries are absent, a most unusual condition, it is highly improbable they are present in the secondary wing-coverts. If this view which I have taken in regard to the presence in the Swifts and Humming-Birds of median under wing-coverts and the absence of major under wing-coverts is correct, we have a most important piece of evidence pointing to their Passeriform nature. The Flexor Carpi Ulnaris Muscle. - Another point which ought to be mentioned here is that in both Cypseli and Trochili the arrangement of the slips from the post-axial division of the Flexor carpi ulnaris muscle to the under surface of the calami of the secondary wing-feathers is Passerine-like; but instead of being thin and narrow these slips are very strong and cord-like in each suborder of birds - a specialized modification of the Passerine form which it is impossible to overlook on dissecting these feather-endings of the Flexor carpi ulnaris (cf. also condition in a Wader, Pl. I. fig. 4). Finally, I might add that I have dissected the Flexor carpi ulnaris muscle and the slips it gives off to the secondary wing-feathers from both its pre-axial and post-axial divisions in a good many orders of birds and hope to publish the results at a later date. I find that the morphology of this muscle has been strangely neglected, although so important and interesting from the share it takes in the matter of flight - a subject which perhaps one may be excused for thinking might have inspired more interest in such a class as Birds. I find, too, so far as I have yet been able to discover, that in every avian order or group of birds the Flexor carpi ulnaris has its own peculiar and special morphology, so that from a taxonomic point of view an intensive study of this muscle ought to be profitable. I find, also, in a preliminary survey of the orders, that we get a condition of extreme evolutionary simplicity in the Palæognathæ and an increasing degree of specialization in the Neognathæ ending with the Passeres and higher groups. Finally, I might mention here that in Cypselus apus (and no doubt other genera,) the tendon of origin of the Flexor carpi ulnaris goes through the same kind of beautiful sling as is seen in the Passeres; but this point has not been worked out in other Cypseline genera. (11) Of all the characters which I have chosen I think the presence or absence of a vinculum between the tendons of the Flexor longus hallucis and the Flexor profundus digitorum is the one to which least importance can be attached, and this because the method of using the feet has been the subject of so much adaptation and variability. All the Passeres, probably without exception, are without a vinculum; the Eurylaimi have one. The Cypseli have no vinculum, but the details of the arrangement of the two tendons not only differ from the suborder Passeres but differ among themselves. (12) The number of rectrices (10) seems to be constant in every Cypseline genus. The ordinary rule in the suborder Passeres is twelve, but the number varies from none to sixteen. From the point of view of phylogenetic relationship, the loss of an outer pair of rectrices in the Swifts is not a character to which one would feel inclined to attach much importance, especially when one considers that the Swift is more completely aerial in its habits than any birds known and that steering is done to a great extent with the wings, and that some Passeres have only ten rectrices or less. (13) The Patagial Muscles. - In birds in general one of the principal and most characteristic muscles of the wing is the Tensor patagii brevis. In the suborder Passeres, as Garrod (1877) has pointed out, this muscle ends in a slender tendon which plunges into the substance of the Extensor carpi radialis longior muscle, and then while still maintaining its identity passes backwards almost at right angles to its former course to end on or about the external condyle of the humerus. Close above it is the glistening tendon of origin of the Extensor carpi radialis longior which springs from the ectepicondylar process of the lower end of the humerus; and this tendon page outwards to be lost in the fleshy belly of that muscle. The relations which these two tendons bear to one another are very constant and very characteristic of the suborder Passeres, although I do not think, as the result of numerous dissections, that the two tendons invariably describe such a triangular space as Garrod's figure would lead one to suppose, for they are often almost superimposed. When, however, we come to examine the same muscles in the Swifts, and I may also add the Humming-Birds, we are met with a very striking innovation. The Tensor patagii brevis has not only assumed relatively enormously large proportions, but it is entirely fleshy and rectangular in form, and runs apparently (but not actually, as we shall see) without the intervention of any tendon, such as is seen in birds of the suborder Passeres, straight on to the body of the Extensor carpi radialis longior. That is the general rule as far as I have been able to ascertain from the dissection of birds in spirit in all Cypseline genera. But there is at least one exception, viz., in Hemiprocne comata, where we find not a Cypseline arrangement but one just described as typical of the suborder Passeres. It constitutes to my mind a most interesting and striking telltale indication of the Passeriform relationships of the Swifts, for Hemiprocne is an undoubted Swift. If more evidence of a like kind were required, we can get it by examining more closely the insertion of the Tensor patagii brevis in a large specialized Swift like Hirundapus. I suggest this Swift simply because it is large and details are easier to demonstrate. Here we shall find that the fleshy end of the muscle does not really terminate in the fibres of the Extensor carpi radialis longior, but in a very narrow tendinous selvage of its own, and that this tendinous selvage at its inner end becomes free and runs in the ordinary Passerine way as a free tendon to be inserted on to the external condyle of the humerus. Moreover, it has just the same relations to the tendon of origin of the Extensor carpi radialis longior on the ectepicondylar process as obtains in a Passerine bird. Nothing could be more unlike the morphological relations of these two tendons in Cypseline genera than that obtaining in non-Passerine birds; and it is an important point, which, as far as I know, has never been noted before, and, moreover, is one which extreme specialization makes it easy to miss unless great care is taken in the dissection-under a dissecting microscope for preference (In Cypselus apus so far has specialization gone that the original condition is all but obliterated.). When we come to compare, further on, the morphology of the Cypseline humerus as exampled in the Eocene Ægialornis, as also in Hemiprocne, and the specialized Cypselus, we shall discover even more cogent reasons for considering that the Swifts have Passeriform relations. (14) Garrod's thigh muscle formula, which he used in his attempt to classify birds, was represented by the symbol AB, XY, where A = the femoro-caudal; B = the accessory femoro-caudal ; X = the semi-tendinosus ; and Y = the accessory semi-tendinosus. Birds of the suborder Passeres lack the accessory femoro-caudal, and therefore their formula is represented by the symbol A, XY. In the specialized Swifts the only muscle of the group present is the femoro-caudal and therefore their formula is A. In Hemiprocne I find the semi-tendinosus muscle is present in a very degenerate condition. The same thing obtains in the case of the Humming-Birds-that is to say, the muscle formula is A. It would be tempting to explain this deficiency in the group of muscles under consideration by the reflection that the loss has been sustained through simple disuse of these particular thigh-muscles consequent upon the fact that both Swifts and Humming-Birds are very special aeronauts. But Humming-Birds spend a lot of their time perching; and Swallows, which gain their living entirely in the air, have a Passerine formula A, XY. It is true that neither Swifts nor Humming-Birds resort to the ground-nor Swallows, except on rare occasions. (15) In both Swifts and Humming-Birds the ambiens muscle is absent. They therefore belong to Garrod's anomalgonatous group, which includes the Passeres, Pici, etc. (16) The Deltoid muscle in the more specialized group of Swifts is extremely reduced and may be missed unless carefully looked for. The humeral portion, if present at all, consists of a few fascicles only. The scapular portion reduced to a narrow ribbon-like band has the same origin and low insertion as in true Passeres, a point which seems of some importance in view of the fact that the insertion of the Deltoid in Passerine and non-Passerine birds is so different. In Passeres I have always found the scapular portion of the Deltoid with a low insertion - that is to say, on or about the ectepicondylar process of the humerus. In non-Passerines like the Coraciiformes it has a high insertion on the humerus. In Hemiprocne comata, although the Deltoid is not so well developed as in Passerines, yet it is so well developed that any reduction might be missed unless looked for. It has a Passerine insertion. (17) I could not find Shufeldt's Dermo-tensor patagii (the Pars propatagialis musculi cucullaris of Fürbringer and Gadow), which he describes in the Raven, and which I find characteristic of the Oscines, in either the Swift or the Humming-Bird. In Hemiprocne, a Swift of an intermediate type, it is, however, present. Shufeldt notes that he could not find it in Tyrannus tyrannus. I can confirm this in other mesomyodian Passeres, and this may prove to be a general rule. In the Acromyodian Passeres it always seems to join the tendon of the Tensor patagii longus and sometimes even to take its place. (18) There is no biceps slip in any Swift known to me, nor is the biceps muscle itself present in any Swift except Hemiprocne comata. The absence of the biceps muscle in the Swifts seems to have been entirely overlooked. The extreme reduction of the Deltoid and the absence of the biceps is interesting in view of the peculiar Cypseline method of using the wings. On the other hand, the Coraco-humeralis, all three portions of the Triceps, the Tensor patagii longus and brevis, the Extensor communis digitorum, and the Pectoralis secundus are among the most obvious of the enormously developed flying muscles. The tendon of this latter muscle is immensely strong. (19) I found no Expansor secundariorum muscle in any Swift or Humming-Bird I have dissected. This peculiar little muscle is not found in the Passeres, Eurylaimi, or Pici, but when we come in my second paper to the Cuculiformes and Coraciiformes we shall see it is invariably present. (20) As in the Passeres, Eurylaimi, and Pici only one carotid artery, the left, is found in the Swifts - and I may add in the Humming-Birds (Trochili). SOME ADDITIONAL NOTES ON THE ANATOMY AND PALEONTOLOGY OF THE SWIFTS It will be clear then, I think, to anyone who compares the lists of characters given above that where dissimilarities from the Passeres occur in the Swifts these are due in almost every case to the extreme degree of specialization to which most of the Swifts have been subjected. But it is not possible in a bare list of characters to give any adequate idea of these specialities. There have, for instance, been remarkable changes in the morphology of the humerus and in the relative lengths of the wing-bones generally. Everyone knows that the humerus in the adult Swift (Cypselus apus) and also in the Trochili is remarkably shortened and wide, while the hand is remalengthened. In respect to this specialization no other bird in the Class, except perhaps the Penguin, comes anywhere near the Swift or Humming-Bird. But Zehnter (1890) has shown that this has been the result of secondary modifications. Thus in the eight days' old embryo Swift the relative measurements were as follows: - Humerus 1; Radius 0.86; Manus 1.70; while following various intermediate phases leading up to full growth we finally get Humerus 1 ; Radius 1.44; Manus 3.47. On the other hand, if we compare the humerus of the Eocene fossil Swift Ægialornis with those of the more specialized of the present-day Swifts, as for example Cypselus, Hirundapus, Collocalia, and others, we find in Ægialornis a long-shafted almost normally Passerine humerus with only the beginnings of the Cypseline specializations which are so obvious in the exaggerated bony projections and shortened shaft of Cypselus and other genera. As regards other Cypseline specialities to be noted in Ægialornis, it is curious to observe that, although the hand of the present-day Swift has apparently increased so much in length in relation to the arm and forearm, yet if we compare its bony constituents with those of the Eocene Ægialornis we find there is so much similarity in actual length, form, and detail as to amount almost to actual identity. The metacarpal bones, for instance, do not seem to have altered in the very least. I have carefully compared a metacarpal of this Eocene Swift with that of Cypselus. The similarities are astonishing. It is as if the bones of the hand had been, so to speak, fashioned in the self-same mould for countless generations and unrealizable millions of years. On the other hand, the arm-bones (humeri) of the present-day Swift and Ægialornis are, as different as they well could be. It is interesting to note, too, that the curious gull-like specialization of the proximal phalanx of the index-finger in Cypselus was also present in Ægialornis of the Eocene and in such a degree of similitude that it would be difficult to distinguish between the two. It was this gull-like speciality which probably led Lydekker to provisionally place Ægialornis close to the Gulls (Cf. 'Catalogue of British Birds (Brit. Mus.),' 1891, pp. 182-184, where a number of bones from the Phosphorites of Quercy are preserved.) (Laridæ). There seems, on the contrary, to be no sort of doubt whatever that Ægialornis was a generalized Swift beginning to specialize on the lines of the present-day Cypselus. I have figured its humerus along with that of the Swift Cypselus apus and the Swallow (Hirundo rustica) because it must be a very rare thing, at any rate in the avian world, to have actual ocular proof of such a continuous specialization from the more or less generalized to the ultra-specialized over such a vast amount of time. THE ECTEPICONDYLAR PROCESS We may also observe another remarkable fact which I do not think has been noted before, viz., that in Ægialornis the ectepicondylar process has only just begun, as it were, to creep up the outer side of the shaft of the humerus from its normal Passerine position, whereas in Cypselus apus it has attained a position of extreme specialization, close under the exaggerated spine-like projection which gives attachment to the great Pectoral muscle. Yet it is clear that the humerus of Ægialornis, although so comparatively little specialized from the Passerine humerus, has nevertheless all the Cypseline characters present and was presumably potentially capable of further development. At the same time we are apparently justified in concluding that if we were in a position to demonstrate the humerus of its progenitors we should find these Cypseline characters still less marked and its purely Passerine - or, at any rate, its more generalized - characters more obvious. Moreover, an even more interesting fact has apparently been overlooked in the Swifts. For if we examine the humerus of Hemiprocne comata, an undoubted Swift, we find that this same ectepicondylar process occupies a position on the shaft of the humerus which is almost exactly intermediate between the position which it occupies in Cypselus and all the more specialized Swifts, on the one hand, and Ægialornis and the Passerines, on the other (Since this paper was read and accepted for publication by the Zoological Society in April 1938, I have received from Prof. Claude Gaillard of Lyon a copy of his "Contribution à l'Étude des Oiseaux Fossiles," published in 1938 in the 'Archives du Muséum d'Histoire Naturelle de Lyon,' tome XV., in which he describes and figures a new species of the fossil genus Cypselavus under the name Cypselavus intermedius (Cf. picture 1) The species is based on a humerus in a perfect state of preservation discovered by M. le Professeur Viret in the Aquitanien limestone of Chavroches (Allier). These limestones, I am given to understand are referred to the top of the Upper Oligocene. On the figure given by Prof. Gaillard can be seen the ectepicondylar process occupying a position on the diaphysis of the humerus almost exactly intermediate between the position in Ægialornis gallicus of the Eocene and Cypselus apus of Recent times, so that nothing could be more opportune from the point of view of my paper and contention as to the taxonomic position of the Swifts than the discovery and description of this species, which in the matter of the movement of the ectepicondylar process up the shaft of the humerus represents an actual example of a halfway stage of evolution between the condition seen in the Eocene Ægialornis and the present-day Cypselus.- P. R. L.). SOME ADDITIONAL MYOLOGICAL NOTES. The ectepicondylar process when present - or, perhaps I should rather say, when obvious - in Passerine birds is situated just above the external condyle of the humerus and it gives origin to the Extensor metacarpi radians longior muscle. This muscle runs along the pre-axial or radial border of the forearm and is attached to the distal end of the first metacarpal. It is, in Passerines, and indeed in almost all birds, one of the principal extensors of the hand upon the forearm; but in consequence of the tendency in the Swifts (seen as far back as in Ægialornis of the Eocene) for this bony process to travel up the outer side of the shaft of the humerus, and to become stronger and more developed as it goes, a far greater leverage and pull can be exerted in the direction of extension in the more and more specialized Swifts than is possible in the Passeres. This extension of the long hand upon the forearm is one of the most characteristic features of the Swift in flight. The extensor communis digitorum muscle is another specialized muscle. In some Swifts and all Humming-Birds its fleshy part is rounded like a racket, while its enormously strong tendon represents the long handle. I have already described the morphology of the patagial muscles above and have pointed out that in Hemiprocne comata we get a purely Passerine morphology, while in the more specialized Swifts we find a degree of specialization which is very remarkable indeed. I also pointed out that Hemiprocne comata was the only Swift in which I had found the Biceps muscle present and a quite well-developed Deltoid (both parts). All other Swifts have this latter muscle extremely reduced. The tendon of the Pectoralis secundus, the chief elevator of the wing, is relatively to the size of the humerus enormously large and strong, out of all proportion to any Pectoralis secundus I have ever dissected. The three portions of the Triceps muscle, the great extensor of the forearm, are enormously developed. There is also another curious little muscle situated on the inner side of the chestwall, viz., the Triangularis sterni. In both Swifts and Humming-Birds this muscle is relatively to the same muscle in other birds very large indeed. THE PERONEAL MUSCLES The morphology of the Peroneal Muscles in the Swifts, as also in the Humming-Birds, as compared with that existing in the Passeres and Swallows, is too interesting to be passed over and moreover has a direct bearing on the subject of this paper. As ChalmersMitchell (1913) has already shown, in a paper dealing with the peroneal muscles in birds, both the Peroneus longus and the P. brevis in the Passeres are present and well developed. In the Cypselidae, on the other hand, the P. longus is, as he says, absent, but the P. brevis is strongly developed and functional. This applies also to the Humming-Birds. I have confirmed these statements in Cypselus, Hirundapus, Cypsiurus (Tachornis), Chætura, Collocalia, and several Humming-Birds. In Hirundo rustica (the Swallow) and Chelidon urbica (the House-Martin), however, the Peroneus longus is evidently in process of disappearing; for, while it has the usual broad superficial origin, it has very little deep or fascial origin and the slip to the tendon of the flexor perforatus is absent. The P. brevis, too, while of the usual Passerine type, is, I think, relatively weaker than in either Passerines or Swifts. I should therefore infer that the Cypselines and Trochilines have lost the P. longus through specialization or loss of function and the Hirundines are in process of doing so. In view, therefore, of the more or less general Passerine morphology seen in the wing-muscles of Hemiprocne comata (an undoubted Swift), I was rather surprised, if not a little disappointed, to find that the status of the peroneal muscles was purely Cypseline. But at any rate it is interesting to find in a single avian entity a Passerine morphology in the wing and a Cypseline in the leg. Another interesting comparative myological condition relates to the morphology of the post-acetabular portion of the Ilio-tibialis muscles of the thigh in the Passeres, Hirundines, and Cypselines. In the Passeres it is well-developed, in the two latter it is absent. Function, therefore, seems to have played a large part in differentiating the anatomy of these groups. And so we might go on finding specialization in almost every part of the Swift's body and one outstanding form, Hemiprocne comata, representing an almost perfect intermediate between the Passeres and the adaptations of the more specialized Cypselines. CONCLUSION On the anatomical evidence, therefore, derived from the fossil Ægialornis and from the specialized and non-specialized Swifts of the present day, we are driven to the conclusion that Swifts are simply specialized Passeriformes. They cannot, however, be placed with the Passeres, and both Swifts and Humming-Birds seem to me to belong to the Passerine past and to be of low evolutionary order, the suborder Passeres being perhaps more modem. I should therefore place the Swifts in a suborder Cypseli of the order Passeriformes, the Humming-Birds (not published here) being placed in a suborder Trochili of the same order. ... SOME BIBLIOGRAPHICAL REFERENCES RELATING TO THE POSITION OF THE SWIFTS AND HUMMING-BIRDS Bates, G. L. (1918): The Ibis, pp. 529-583. The Reversed Under Wing-coverts of Birds Beddard, Frank (1890): The Structure and Classification of Birds Gadow, Hans (1893-96): Newton's Dictionary of Birds, p. 617. The Arrangement of the Deep Plantar Tendons in the Trochilidæ Gadow, Hans (1895): The Ibis, p. 299. Gadow's Retraction of the above Description Gaillard, Claude (1908): Les Oiseaux des Phosphorites du Quercy. Annales de l'Université de Lyon, fasc. 23, fig. 10, pl. iii Gaillard, Claude (1938): Archives du Muséum d'Histoire Naturelle de Lyon, tome xv. Contribution à l'Étude des Oiseaux Fossiles GARROD, A. H. (1875): Proc. Zool. Soc. London, p. 344. On the deep Flexor Tendons of Cypselus alpinus GARROD, A. H. (1877): The Zoologist, pp. 217-220. On the Anatomical Characters distinguishing the Swallow and the Swift GARROD, A. H. (1877): Proc. Zool. Soc. London, p. 356. On some Anatomical Peculiarities which bear upon the Major Divisions of the Passerine Birds Goodchild, J. G. (1890): Proc. Roy. Phys. Soc. Edin. vol. x. pp. 317-332. The Cubital Coverts of the Euornithæ in relation to Taxonomy Goodchild, J. G. (1886): Proc. Zool. Soc. London, pp. 184-203. Observations on the Dispositions of the Cubital Coverts in Birds Gunther, A. C. L. (1867): Phil. Trans. p. 599, pl. xxvi. fig. 2. Articulation of the Pterygoids with the vomers which occurs in Hatteria - normal in young Birds Huxley, Thos. H. (1867): Proc. Zool. Soc. London, p. 452. The Classification of Birds Lambrecht, Kálmán (1933): Handbuch der Palaeornithologie (Berlin), pp. 620-624 - Order Micropodiformes Lowe, P. R. (1931): Proc. Zool. Soc. London, pt. ii. pp. 457-459. On the Anatomy of Pseudocalyptomena and the Occurrence of Broadbills in Africa Lucas, Frederic (1886): The Auk, pp. 444-451. The Affinities of Chætura Lucas, Frederic (1889): The Auk, pp. 8-13. The main Divisions of the Swifts Lucas, Frederic (1893): The Ibis, pp. 365-371. Swifts and Humming-Birds Lucas, Frederic (1895): The Auk, pp. 155-157-Additional Characters of the Macropterygidæ Lucas, Frederic (1895): The Ibis, pp. 298-299. The deep Plantars in the Trochilidæ Lucas, Frederic (1896): The Auk, pp. 82-83. The Deltoid Muscle in the Swifts Lydekker, R. (1891): Catalogue of the Fossil Birds in the British Museum, London, p. 183, fig. 41. Ægialornis gallicus, nov. gen. et sp., referred to Laridæ Milne-Edwards (1891): 2e Congrès Ornithologique de Budapest, p. 66. Tachyornis hirundo. Sur les Oiseaux des dépôts eocènes de phosphate de chaux. In this paper the author demonstrates the affinity of Ægialornis gallicus Lydekker to the Swifts (Cypselidæ) with the Vomers which occurs in Ha normal in young Birds Mitchell, P. Chalmers (1913): Proc. Zool. Soc. Lond. pp. 1039-1072. The Peroneal Muscles in Birds Parker, W. Kitchin (1876): Trans. Linn. Soc. London, Zool. vol. i. p. 116. On the Structure and Development of the Bird's Skull Parker, W. Kitchin (1889): The Zoologist, vol. xiii. pp. 91-95. On the Systematic Position of the Swifts Parker, W. Kitchin (1878): Trans. Zool. Soc. vol. x. pp. 289-348. On Ægithognathous Birds, pt. i Sclater, P.L. (1865): Proc. Zool. Soc. London, pp. 593-617. Notes on the Genera and Species of Cypselidæ Seebohm, Henry (1890): The Ibis, pp. 202-205. An Attempt to diagnose the Subclass Coraciiformes and the Orders, Suborders, and Families comprised therein. (Deep Plantar Tendons of Humming-Birds.) Shufeldt, R.W. (1885): Proc. Zool. Soc. London, pp. 886-915, pls. lviii.-lxi. Contributions to the Comparative Osteology of the Trochilidæ, Caprimulgidæ, and Cypselidæ Shufeldt, R.W. (1886): Proc. Zool. Soc. London, pp. 501-503. Additional Notes upon the Anatomy of the Trochilidæ, Caprimulgidæ, and Cypselidæ Shufeldt, R.W. (1889): Linn. Soc. Journ., Zool. (London), vol. xx. pp. 299-394, pls. xvii.- xxiv. Studies of the Macrochires, Morphological and otherwise, with a View of indicating their Relationship and defining their several Positions in the System Shufeldt, R.W. (1893): The Ibis, pp. 84-100. Comparative Notes on the Swifts and Humming-Birds Shufeldt, R.W. (1894): The Ibis, p. 32. On the Taxonomy of the Swifts and Humming-Birds - a Rejoinder Thompson, D'Arcy. (1901): Proc. Zool. Soc. London, vol. i. pp. 311-324. On the Pterylosis of the Giant Humming-Bird (Patagonia gigas) Stewart (1891): R. B. Sharpe's Review of recent Attempts to classify Birds, see footnote, p. 81. Considers that the palatine morphology of the Trochili is a modification of the ægithognathous palate Zehnter, L. (1890): Arch. für Naturg. lvi. p. 189. Beiträge z. Entwicklung von Cypselus melba (Transl. 'Ibis,' 1890, p.196). On the Development of the Feet of Cypselus melba PLATE I Fig. 1. Basal view of skull of Swift (Cypselus apus), to show ægithognathous character. pr.pal., pre-palatines; vo., vomer; mx.pal., maxillo-palatines; i.p.a., interpalatine spur; trs. pal., transpalatine processes; pt., pterygoid Fig. 2. To show dorsal and ventral feather tracts in the Swift (Cypselus apus). A, dorsal; B, ventral. Fig. 3. Portion of under surface of left wing of a Passerine bird (Cornus cornix) (Hooded Crow), dissected to show (a) absence of major under wing-coverts; (b) median and lesser under wing-coverts linked together; (c) various unstriated muscles and ligaments activating and connecting the coverts and the remiges (sec-rem.); (d) two secondary remiges passing forwards to the ulnar ; (e) slips * passing backwards from the aponeurosis of the pre-axial division of the Flexor carpi ulnaris muscle to be attached to the under surface of the secondaries; (f ) finer and more tendinous slips (.) passing backwards from the post-axial division of the Flexor carpi ulnaris. Sec.Rem.., secondary remex; Med., secondary under wing-covert (reversed); Les., lesser under wing-covert (unreversed); Flex. carpi uln. (a), pre-axial division of Flexor carpi ulnaris; Flex. carpi uln. (b), post-axial division of Flexor carpi ulnaris; *, slips from pre-axial division of Flexor carpi ulnaris; (.), slip from post-axial division of Flexor carpi ulnaris. Fig. 4. Portion of under surface of left wing of a non-Passerine bird (Tringa ochrophus), dissected to show (a) presence of major under wing-coverts (M.U.C.) (median and lesser under wing-coverts not shown by artist); (b) ligamentous bands connecting major coverts and secondary remiges; (c) two secondary remiges (sec. rem.) passing forward to the ulnar (upper portion of figure slightly diagrammatic; (d) slips * passing backwards from the aponeurosis of the pre-axial division of the Flexor carpi ulnaris muscle to be attached to the under surface of the secondaries; (e) triangular-shaped muscle slips (.) from the post-axial division of the Flexor carpi ulnaris to be attached to the under surface of the secondaries. |
|
( Printed 1938 in the Transactionsof the Zoological Society of London.)
|
| © APUSLife 1997, No. 0414 |