The Juvenile Plumage of Some Birds and an Interpretation of Its Nature.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 72, 1942-43, Page 6

The Juvenile Plumage of Some Birds and an Interpretation of Its Nature. By E. Percival, Canterbury University College, Christchurch, New Zealand. [Read before Royal Society New Zealand, Canterbury Branch, June 7, 1939; received by the Editor, February 19, 1942; issued separately March, 1942.] Kerr (1919, p. 74) and Plate (1922, p. 267) have concluded that the feather succession, from natal plumage onwards, consists of the appearance of a series of parts of a whole; as the former puts it, “the down feather and the definitive feathers which succeed it in the series of moults, are all portions of a single greatly elongated and basally growing structure—the first down feather being its tip, and the succeeding feathers being successive portions of it. The moult consists not in the shedding of the whole feather, but merely in the breaking off of its projecting portion.” Newton (1893–96, p. 248) refers to the persistently growing rectrices of the males of a Japanese breed of fowl, the moulting of which is said to be artificially prevented. The fact that continuity can be determined between the first and second downs and the juvenal feathers of the Royal Albatross and Penguins, and between old and new feathers in moulting Penguins, sufficiently reinforces the above quotation and the matter hardly calls for further support. There are, however, discussions of the phylogeny of feathers as a whole and of the nature of nestling feathers which make necessary some reconsideration of the growth and the nature of the structures involved. Pycraft (1907, p. 11), writing about the nestling down of Penguins, says, “it has been suggested that these down feathers are really a part of the actual contour feather.” Further, he says, “But there can be no doubt that the neossoptiles represent distinct feather generations,” and again, “the bases of the rami of the first become welded on to the tips of those of the second (down).” Ewart (1921, p. 633 et seq.) discusses the origin and history of feathers and concludes, basing his conclusions largely on a consideration of the juvenile covering of Penguins, that protoptiles, derived from simple filaments, in some cases acquired the chief characteristics of true feathers. He believes that mesoptiles were interpolated or introduced to cope with an ice age, and that later they were in many cases superseded by true feathers. 1. The Nature of the Early Coats. The Emu. Ewart (1921, text-figure 11) has figured what he calls “the first three generations” of the Emu's plumage, a diagram presumably based on the material which is photographed on plate vii, figs. 26 and 27 of his paper. These two figures were apparently from small specimens and appear to be magnified about two diameters. Material from four distinct birds, two alive and bred in Wellington Zoological Park, and two preserved ones, one in the Otago Museum, Dunedin,

and one in the Dominion Museum, Wellington, shows no sign of “three generations” up to 10 months of age. Ewart's material was seven months old. The specimens examined in the present case include a sample about six weeks old, one seven months old, one about nine months old and one about ten months old. Through this series can be traced the wearing down and disappearance of the protoptile. In all cases examined, the tip of the succeeding aftershaft broke away from the calamus of the protoptile so that there was not formed a compound hyporachis. The calamus of the protoptile was found to contain a pith which, in 70% alcohol, gave the appearance of “cones” to the number of four. In spite of Ewart's view (op. cit., p. 628) that the protoptile of the Emu has no calamus, there does not seem any justification for believing that the structure is not a true calamus. At about ten months of age the young Emu produced a calamus in the feather follicle, the inferior umbilicus being near to closure. Thus, two feather generations were completed at about this age, the second having the character of feathers which would be moulted by dropping from the follicles, not lost by being broken from the tips of the next succeeding generation. The relation between protoptile and this second generation in the Emu is similar to that between protoptile and mesoptile in Ducks and Swans (see, e.g., those preceding dorsal contour feathers). It is to be concluded that the second generation of the Emu is mesoptile (which Pycraft, 1907, p. 13, regarded as a probability), and is completed at about 10 months of age, while in a Mallard–New Zealand Grey Duck cross the mesoptile of the dorsal contour feather was completed in about 40 days. Penguins. The double-down of Penguins was described by Pycraft (1907) and Clark (1906). As was mentioned above, Pycraft considered that the rami of the first down had become “welded” to those of the second. Ewart (op. cit., p. 638) suggested that the mesoptile may have been produced by lengthening of protoptile barbs and by new formations. His text-figure 14 gives a good impression of the relations between the two sections. This figure shows a feature which is found in the White-flippered Penguin (Eudyptula albosignata), namely, the absence of barbules from the region representing the junction between first and second downs, or, as they are called, “protoptile” and “mesoptile.” Ewart describes this region of junction as not a true protoptile calamus. In the present study, specimens of the double down of the King Penguin (Aptenodytes patagonica) were examined. The numbers of filaments in each section of four examples were, in first and second down respectively, 15 and 15, 26 and 28, 26 and 26, 28 and 28. In the second case there was evidence that the extra two barbs in the second down had broken ends, as they were blunt. Double downs from a White-flippered Penguin (Plate 2. figs. 2a, 2b, 2c) had each a tuft of 10 or 11 filaments attached to a growing second down consisting of a very much larger number of filaments, most of which

were free distally and were pointed, showing no signs of having been broken off. As far as mere arrangement goes, Ewart's text figure 14 falls between the King Penguin and the White-flippered Penguin. Ewart's text-figures 3 and 4 (Ringed Penguin) and 14 (Adelie Penguin) show no barbules at the junction of first and second down and between second down and definitive feather. A similar condition occurs in the White-flippered Penguin, but in the King Penguin barbules are continuous between the first and second downs. As indicated by Ewart (op. cit., p. 630), the junction of the first and second down of Penguins is formed by a constricting, short band of epitrichium, the early separated barbs lying side by side within. When this band of epitrichium is cut and removed, the barbs fall apart and form a single tuft diverging from the base of the second down (Plate 2, figs. 1a and 1b). Sometimes barbs remained gummed together and had the appearance of being fused, but were easily separated. Ewart's text-figures 3 and 4, of the Ringed Penguin feather, show a remarkable structure which he interprets as a complex mesoptile aftershaft “connected with the shaft as well as with the aftershaft of the true feather.” He does not say to what extent this complexity was present in his material or whether only a single feather was examined. Without an opportunity to review the material on which the figures are based it is not possible to discover to what extent gumming of filaments, if any, was responsible for that arrangement of barbs. Adhesions of a similar odd nature were seen in down attached to teleoptiles of the Royal Albatross (Diomedea epomorpha) but the parts were separated easily even four days after being taken from the bird, when drying had already taken place. The Royal Albatross. A feature was present in the “calamus” of the second down of the body contour feather which resembled the condition in the Penguin's double down. On a seven months' old bird the junction between second down and teleoptile of a thigh contour feather was marked by a broad band of epitrichium which could be pushed to and fro quite easily over the enclosed structures (Plate 3, fig. 3c). This band was quite tough, more so than that of the Penguin. The barbs enclosed formed another cylinder deeply marked in ridges which connected the barbs of the teleoptile with those of the second down. A small portion of this ridged inner cylinder carried no barbules on the ridges, but more distally, and still enclosed by epitrichium, barbules were present. A similar condition was found in the connection between second down and teleoptile near the base of the bird's neck. A Museum specimen of the same species, with age given as one month, had double down of coverts and remiges built as above, as well as what may be neossoptiles of plumulae or filoplumulae, and from which the epitrichial bands appeared to be in process of slipping, perhaps due to preening. The junctions between first and second down in some cases showed continuous barbules and, in others,

gaps in the distribution of these appendages. In all these cases examined, the barbs were free inside the epitrichial sheath exactly as was seen in the King Penguin. In several examples, but occurring irregularly, the barbs in the region of the “calamus” formed a wavy bundle which, after soaking in water, was readily dissected into its separate barbs. These latter seemed to be gummed by means of dried lymph or a similar coagulable fluid. On the neck of this younger bird, the first-second down junction was in some cases formed by a calamus-like section which was somewhat transparent but contained some pith. This part formed an unsplit cylinder of feather material enclosed by an adherent sheath of epitrichium. In specimens of this kind, when splitting of the “calamus” took place, there were two or more groups of barbs, each attached to its own segment of “calamus.” The final splitting resulted in more or less continuity between one barb of second down and a barb of first down. Occasionally two or more of one section would be attached to a barb from the other section, showing that splitting of the “calamus” might be irregular. Contour feathers from the base of the neck and bearing second down had barbules continuously throughout, the difference between teleoptile and second down sections being marked by a sudden change in the form of the barbules (see Plate 4, fig. 4, Pachyptila turtur). More distally there was no indication of the region of junction of first and second down, suggesting that this had been formed by a band of eptrichium binding a bundle of separate barbs in a similar manner to that seen in the double down of the King Penguin. Down from the undercoat of the neck of the one month old Albatross contained examples which resembled mesoptiles in that they had a rachis-like structure, but of an unusual form. The pinnate structure was irregular and the barbs branched dichotomously, forming a feather much unlike a mesoptile as found in the Duck or Swan. The rachis-like structure could have been formed by fusion of barb rudiments during development. They are not regarded as true mesoptiles. The first down on the head of the Albatross seems to break off as a tuft, the second down forming a velvety surface. On the neck and elsewhere the first down, by splitting of the “calamus,” remains as filaments continuous with the second down filaments. These first down barbs are gradually preened away, so that in the one month old bird much of the first coat is lost. In this double down of the Royal Albatross some feathers possessed the same number of first down barbs as of second, while others had more second than first, suggesting either an increase in the former or a decrease in the latter. The excess second down barbs had pointed ends and were free of the “calamus.” The same applies to the relation between teleoptile and second down. Similar arrangements have been noted in Penguins. A dorsal contour feather of Pachyptila turtur, in the fledgling stage, showed an absence of barbules at the first-second down junction and a sharp discontinuity in length of barbule at the second-down-teleoptile junction. The barbs of the aftershaft of the teleoptile

were not continued as second down, and almost all of the barbs on the shaft were continued as second down barbs (Plate 4, fig. 4). The double down and its relation with the definitive feather in the Albatross show, except in the doubtful case of the underdown from the neck, a complete absence of rachis and hyporachis. The rachis of the teleoptile extends not at all to the second down, and as far as has been observed, the barbs of the aftershaft are not continued as down filaments. The Black Swan (Chenopis strata). The nature of the neossoptiles of this species is generally similar to that of the Mallard as described by Ewart (op. cit.). Emphasis must be laid on the mixture of neossoptiles, in both Ducks and Swans, consisting of a clear succession of protoptile, mesoptile, teleoptile, with rachis running throughout and a true calamus between the sections, and a double down built on a plan similar to that found on the neck of the Royal Albatross, where was a calamus-like junction of unsplit material with adherent epitrichium, and no rachis in either section of down. The first down, in this type, appeared to fall as a tuft by the breaking away from the proximal side of the calamus of the barbs of the second down. The head bore a double down which was clearly protoptile and mesoptile. Ewart (op. cit., pp. 613–618), in dealing with the wing-quill mesoptiles of the Mallard, refers to the suppression of mesoptile in Anseres and Galli, and figures (text figure 5), from the Chinese Goose, the relations between protoptile, mesoptile, and teleoptile. An ulnar remex from the Black Swan showed what may be a vestige of mesoptile, quite absent from other specimens examined, the protoptile being well developed and having a short hyporachis with two barbs. Generally in these remiges, as well as in ventral contour feathers, there was a strap-like and perforated portion running in from the protoptile calamus and terminating freely behind the teleoptile. This may represent the highly modified teleoptile aftershaft as figured by Ewart (see above). The Woodhen. Out of nine remex feathers from a chick of a Weka or Woodhen (Gallirallus australis) all of which had well-developed protoptiles each with a long rachis, one showed a structure which may be identified as a vestigial mesoptile. The structure resembles that on the remex of the Black Swan, mentioned above, and is to be interpreted according to Ewart's text figure 5. Ewart's view that the mesoptile has become reduced or is gone would seem to be applicable beyond Anseres. Reference will be made later to his inclusion of Galli in this consideration. The general juvenile plumage of the Weka, apart from that of the remiges and rectrices, consists of a single down similar to that of the domestic Fowl, with barbs and barbules. There is a connection between the tuft and the teleoptile consisting of an upsplit cylinder covered with epitrichium. This cylinder splits longitudinally as in the Tern and others.

The Pukeko or Swamphen. Porphyrio melanonotus has neossoptiles similar to those of Gallirallus. A feature of this juvenile plumage is that the young down feathers are enclosed in epitrichial sheaths which remain for a long time. A young male, 9 inches high to the crown of the head, still bore sheaths on neossoptiles on the front of the head and on remex protoptiles. The White-fronted Tern (Sterna striata). In this bird is a single coat of nestling down consisting of filaments with barbules, based on a more or less incompletely split tube with adherent epitrichium and continuous with the tips of most of the barbs of the teleoptile. The splitting of the connecting section (“calamus”) is irregular so that several down barbs may be attached to one teleoptile barb. No rachis was found in the down. The Domestic Fowl (Gallus gallus). Here the nestling down resembles that of the Tern in its relations with the teleoptile. On the neck the down breaks away entirely, as a tuft, from the juvenal feather, but generally elsewhere the “calamus” splits longitudinally as in the tern. The condition in the Pheasant and Californian Quail resembles that of the domestic Fowl. The Domestic Pigeon (Columbia livia). The nestling down resembles that of the Fowl except in its lack of barbules. A tail quill from a young fantail Pigeon showed the tips of the teleoptile barbs tapering to join the down “calamus.” The barbules of this tapering part became rapidly shorter and finally disappeared, leaving a bare termination. The short, tapering portion broke away and was dropped with the liberated tuft of down, so that a relatively uniform distal border was left on the teleoptile. This portion agrees in position with what Ewart (op. cit., pl. x, fig. 38) calls vestigial mesoptile. It is in fact a portion of the teleoptile (Pl. 4, fig. 6). Passerine Birds. These, as is well known, offer a final expression of the modification of nestling down. Not only is there a marked reduction of filaments in each tuft, but there is a marked reduction in the extent of down-bearing areas, such nestlings as those of Blackbird, Starling and Chaffinch having only vestiges of down tracts. The English Sparrow is quite naked until the juvenal coat appears, although Dwight (1900) says that it has a natal down of a mouse grey colour. The natal down of Starlings and Blackbirds is shed as tufts, the barbs of the teleoptile breaking away, one by one, from the unsplit “calamus,” which is still encased by epitrichium. Morphologically, the relations are similar to those of the Pigeon and the Fowl. 2. The Use of Terminology. The names applied to the various coats and parts of coats of juvenile birds give the impression that homologous structures are dealt with throughout. At the same time, references are made in literature to views which appear to conflict with such an impression. Pycraft has already mentioned the suggestion that the nestling-down feathers of Penguins “are really a part of the actual contour feather,”

and Duerden (1911, p. 5, footnote) says, “It has recently been shown that in many birds the barbs of the new feathers are directly continuous with the barbs of the down feathers, no real break occurring between the two. For this reason some writers consider that the down feathers do not represent a distinct plumage, but are to be looked upon as the modified tip of the true feather (the definitive feather).” The consideration of the phylogeny of feathers seems to have led to the view expressed by Ewart (op. cit., p. 636), that the phyletic sequence was first the filament, such as is found in many newly hatched birds (e.g., Emu, Penguin, Royal Albatross), next, hair-like barbs (compare with the nestling down of the Pigeon), followed by a “protoptile” of barbs provided with barbules, as in the Penguin. Ewart says (p. 637), “a series of links connect the relatively simple umbelliform protoptiles of Penguins with the highly specialised protoptiles of Ducks and Emus.” He believed that the evolution of birds took place in arid conditions followed by a glacial period, and that the latter had set in before wing-quills and other true feathers were evolved. This involved the interpolation of a furlike coat exemplified by the “mesoptiles” of Penguins. The facts presented above in part 1 suggest that juvenile plumages as a whole have undergone some modifications, to various extents, since the birds arose. At this stage it is not necessary to inquire closely into the origin of neossoptiles, but rather to determine homologies by a comparative study. The Ducks and Swans present neossoptiles having a resemblance in the fact that there are continuous first and second “down” feathers possessing in each part a rachis which is structurally continuous with that of the definitive feather. The same applies to the down on the remiges and rectrices of the Weka and the Pukeko. The first two feather generations of the Emu have similar relations. There appears to be no reason to doubt that in different birds comparable parts of these structures are homologous and that, when in most complete succession, they are protoptile and mesoptile followed by some sort of teleoptile (cf. Newton, op. cit., p. 243). In this succession the barbs of one section, particularly in the Emu, are independent of the barbs of another, there being a more or less pinnate series of them along the rachis. In some other birds, however, the situation appears to be very different. Those species dealt with in this paper include examples in which neossoptiles are very much unlike the above. The Penguin presents such a case in one of its most obvious forms. Here, from the early appearance of the second down, as noted in Eudyptula, there is continuity of individual barbs from first down to second, and on the removal of the constricting epitrichial sheath the barbs can be separated without further breaking of tissue. Further, in the King Penguin there is a continuous series of barbules along the barb of the double down. There is no rachis or hyporachis in the double down, these being restricted to the teleoptile. The junction between second down and teleoptile of Eudyptula and Megadyptes appears to undergo delayed splitting, to separate the barbs.

The condition of the Royal Albatross is similar to the above, no rachis or hyporachis except in juvenal feathers, and, except in the head region, a more or less complete separation of barbs within the eptrichial constriction, and generally, on the trunk, a continuity of first and second down barbs with those of the teleoptile. As regards the relation of the three plumages to one another, however, there is almost a complete range of variation between one set of barbs continuous throughout and merely constricted by a loose epitrichial girdle, and that of distinct sets of barbs separated (as well as united) by unsplit, calamus-like junctions. These junctions between first and second downs on the head and neck bear a close likeness to the “calamus” of nestling down of higher birds, such as the Tern, the Fowl, the Pigeon, etc. The condition in these last-named birds is to be interpreted readily by reference to the Albatross and the Penguin. The substance of the last two paragraphs is very much in agreement with the sense of the quotation of the footnote from Duerden (see above), and we are left with the conception of two main categories of neossoptiles, namely, those having a rachis and pinnate barbs, preceding the teleoptiles, and those having no rachis but being clearly modified portions of teleoptile barbs. The neossoptiles of the Emu are in the first category, those of the Anserines and Rails are a mixture, those of the Albatrosses, chiefly if not entirely, are in the second, and those of the Fowls, Pigeons, Passerines are in the second. The clear distinction between the neossoptiles of Emu and in part of Anserines on the one hand and of Penguins, Albatrosses, Galli, Pigeons, Passerines, on the other, makes necessary a reconsideration of the use of the terms protoptile and mesoptile. Newton (op. cit., p. 243) states, “Neossoptiles are characterised by (1) a very short calamus, (2) an insignificant or ill-defined rachis, if there be one at all, (3) the almost universal absence of cilia, (4) long and slender rami, and (5) absence of an aftershaft, except in Dromaeus.” Pycraft (1907, p. 11) says that nestling down feathers “present very different grades of perfection, such for example as may be seen in the umbelliform tufts of loose, woolly down of, say, an owl, the semi-plumose type of the Galli and Anseres, and the strongly pennaceous type of some Tinami.” He concludes that “the full sequence is represented (1) by neossoptyles, composed of (A) pre-pennae, divisible into α-protoptyles and β-mesoptyles; (B) pre-plumulae, and (2) teleoptyles or definitive feathers.” On p. 12 he says, “This second generation of feathers we may call provisionally mesoptyles.’ ” It is seen, then, that the terms “protoptyle” and “mesoptyle,” used by Pycraft, apply to generations of feathers without any particular reference to their structure. Newton appreciated the varied structure of his neossoptiles without any great emphasis on the succession, if, indeed, he were aware of it. Ewart elaborates the knowledge of structure and succession and considers all neossoptiles to be fundamentally homologous, and uses the terms “protoptile” and “mesoptile” to lead to that conclusion. In the present study, however, it has been shown that the nestling down of Albatrosses and Penguins consists of modified teleoptile barbs, and that the structure

of the neossoptiles of Galli, Passeres, and more or less of the nestling down of Ralli and Anseres may be similarly interpreted. It is also clear that this kind of nestling down is fundamentally different from the first plumage of the Emu and from that part of the juvenile plumage of the Duck and the Swan which possesses a rachis. To clarify the issue it is here proposed to restrict the term protoptile to apply to the primitively first, rachial neossoptile, and to restrict the term mesoptile to apply to the primitively second, rachial neossoptile. These restrictions make necessary new terms to cover the different modifications of precocious teleoptile barbs. The term haploptile is here proposed to designate the single non-rachial neossoptile formed by precocious elongation and appearance of the barbs of teleoptiles. The term diploptile is proposed to designate the double, non-rachial neossoptile formed by precocious elongation and appearance of the barbs of teleoptiles. The term pseudo-protoptile is proposed to designate that part of the double, non-rachial neossoptile which appears first in the succession. The term pseudo-mesoptile is proposed to designate that part of the double, non-rachial neossoptile which appears second in the succession. For the purpose of exactitude, the terms protoptile and mesoptile, as used above, are applied to the rachial neossoptiles of Anseres, in particular, Chenopis strata. The term haploptile is applied to the non-rachial neossoptile of Galli, in particular, Gallus gallus. The term diploptile is applied to the non-rachial neossoptile of Spheniscidae, in particular, Eudyptula albosignata. The foregoing application of terms makes necessary a further classification, viz.: that of the use of the word Calamus. This word is particularly applied to the proximal, hollow, transparent, unsplit part of the adult feather, which is normally held in the follicle. It is made up of median and lateral material, i.e., material which is continuous directly with rachis and barbs and hyporachis (if present) of the preceding feather. Obviously, then, the term calamus cannot apply to the proximal part of a haploptile or to a part of a diploptile since there is no median material in it or in the preceding neossoptile; it consists of modified parts of teleoptile barbs. The term pseudo-calamus is here proposed to apply to the proximal part of a haploptile, or of a pseudo-protoptile, or of a pseudo-mesoptile, which, by its relations, contains only lateral material, continuous with the barbs of the neossoptile preceding it and with the barbs of the feather succeeding it. The presence or absence of cones is no criterion of the nature of a true or false calamus. It will be noted that these neossoptiles derived by modification of teleoptile parts vary in constitution, some containing material continuous with barbs of shaft and aftershaft, as in the Penguin, others consisting only of barbs derived from the shaft, as in Pigeons and higher birds, and in Albatrosses. 3. Some Phyletic Considerations. The differences of origin and structure of protoptile and mesoptile on the one hand and pseudo-protoptile and pseudo-mesoptile on the other make necessary a review of Ewart's conclusions that

“the protoptiles of Penguins represent a second stage in the evolution of true feathers,” and that “a series of links connect the relatively simple umbelliform protoptiles of Penguins with the highly specialised protoptiles of Ducks and Emus.” If these conclusions are well based it should be possible to obtain some kind of comparative evidence of the building up of mesoptile and protoptile with rachis and barbs from structures consisting only of barbs and barbules or of barbs only. The material presented in the foregoing suggests that, concerning pseudo-protoptile and pseudo-mesoptile, there is a solid basis of fact to support the view that the barbule-less condition of Pigeons and Passerines represents a modification of the condition like that of Fowl, Quail, Terns, where barbules are present, these latter having many features in common with the neossoptiles on the head of the Albatross, where there is a well-established, unsplit pseudo-calamus. A comparison of this down with the rest of the down of the young Albatross establishes an essential similarity of structure of the pseudo-protoptiles and of the pseudo-mesoptiles of all those birds under consideration. In the present study of the cygnet of the Black Swan or the duckling of a cross between Mallard and Grey Duck (Anas superciliosa), nothing has been seen which shows a transition from pseudo-protoptile or pseudo-mesoptile to protoptile or mesoptile. It has been noted by Ewart and confirmed in the present work that in Anseres the mesoptile is found vestigial in or absent from certain neossoptiles, and a similar condition has been noted in Gallirallus. The conclusion is, then, that there is to be seen a tendency towards the disappearance of true neossoptiles. The alleged suppression of mesoptile in the Galli, referred to by Ewart (see above), must be re-examined in the present light, if suppression in fact takes place, and regarded as a reduction of a part or a whole of a pseudo-mesoptile. It can only be analogous to the suppression of a mesoptile in the Anseres. The tail quill of the young Pigeon, described earlier, with the peculiar terminations to the distal barbs of the teleoptile, may be an example of degenerating pseudo-mesoptile or may be a specialisation which has no reference to such a structure. The question of the possible suppression of pseudo-protoptile or pseudo-mesoptile immediately brings up the further question of the relation of the single down of Passerines, Fowls, Pigeons, Rails and others to the double down of Penguins and Albatrosses. Nothing emerges from the literature on the subject which justifies a conclusion that the single down is derived by a suppression of one or other of the parts of double down. The structure of the nestling down of, and its relation to, the tail quill of the Pigeon appears slender evidence on which to build a precedent double down. (See table on next page.) It is advisable to suspend judgment on the question of the phyletic relation of these two kinds of down if we reject the view that the Pigeon has a vestige of a pseudo-mesoptile, since there appears to hand no information that those birds possessing haplop-tiles do show traces of unequivocal pseudo-mesoptiles. If those birds bearing diploptiles and haploptiles have lost protoptile and mesoptile and have replaced them by modification of the teleoptile, the question arises as to whether the single down or the double down appeared first.

Table Showing Distribution of Neossoptiles in Birds Dealt With in this Paper. Group. Genus. Protoptile-mesoptile succession only. Protoptile-mesoptile succession mixed with Diploptile. Haploptile with remnants of Protoptile-mesoptile succession. Diploptile. Haploptile. No Neossoptile. Ratitae Dromacus + Struthio + Anseres Anas + Chenopis + Rallidae Gallirallus + Porphyrio + Spheniscidae Eudyptula + Aptenodytes + Procellariidae Diomedea + Pachyptila + Galli Gallus + Phasianus + Callipepla + Columbidae Columba + Passeres Sturnus + Turdus + Fringilla + Passer +

1. Aptenodytes patagonica (King Penguin). (a) Double down showing part of pseudo-piotoptile. The epitrichial band is some distance proximal to the place of breaking. (b) Double down with epitichial band removed showing independent and continuous barbs. The barbules are shown to be continuous. 2. Eudyptula albosignata (White-flippered Penguin). (a) Pseudo-protoptile continuous with pseudo-mesoptile. (b) Pseudo-protoptile lost, teleoptile appearing. (c) Pseudo-mesoptile and teleoptile. The pseudo-calamus is split and the barbs of one are continuous with those of the other.

3 Diomedca epomoipha (Royal Albatross). (a) 1 month old. Dorsal double down with pseudo-calamus. (b) 1 month old Dorsal double down with open connevion between pseudo-protoptile and pseudo-mesoptile. The junctions show no barbules. (c) 7 months old. Thugh contour bearing pseudo-mesoptile, The pseudo-calamus consists of a loose epitrichial sheath and an almost completely split cylinder of feather substance. (d-g) 1 month old. Double downs from hand showing slipping off of epitrichial sheath and opening out of split cylinder connecting both parts.

4. Pachyptila turtur (Fairy Prion). Dorsal contour, with open double down, showing relations between pseudo-protoptile, pseudo-mesoptile and teleoptile. 5. Po mclanonotus (Pukeko). (a) Remes from young male, 9th. Protoptile bearing epitiuoal sheath (b) Dorsa! haploptile bearing sheath. 6. Fantail Pigeon. Rectrix showing breaking off of tips of barbs and their falling away attached to pseudo-calamus of haploptile.

The palaeontology of the groups containing the species under consideration presents an interesting picture. Lambrecht (1933) leads to the conclusion that the Penguins had attained their specialisation by Miocene times. Gigantornis, placed in the Procellariiformes, is given as occurring possibly in the Eocene. Palaeotringa and other Telmatoformes extend back to the Eocene. Palaeophasianus, among the Galli, is recorded from the Eocene, while other pheasants extend back to the Miocene. The Ralliformes are given as extending back to the Eocene, as are the Anserine and Passerine birds. Thus, all those groups of birds under examination, which at present bear haploptiles and diploptiles, are reported from the early Tertiary, so that it is evident that the origin of these forms of neossoptiles must be sought in earlier times, perhaps in the Cretaceous. It is possible that a haploptile could become a diploptile, similar to that of a Penguin, by the persistence of an epitrichial band situated about halfway along the length. Such a constriction has been observed, not on a haploptile, but on the protoptile of an ulnar remex of a young Black Swan. It is not yet possible to be certain whether it was in its original position or had been caused to slide distally from the region of the calamus by preening. That the latter may be the case is suggested by the fact that the constricting bands of the diploptiles of remiges of the young Royal Albatross were in various positions, from the pseudo-calamus distally, until they had in some cases been lost and the double down had the form of an unconstricted tuft (pl. 3, figs. 3d-g). By retention of the band and a delay in the separation of the constricted barbs a pseudo-calamus similar to that of a contour diploptile could be produced. The contention of Ewart (op. cit., p. 631), that the development of the nestling feather of the Galli in a filament is in favour of its being protoptile, can now be regarded as somewhat beside the point. The appearance of an epitrichial sheath on early neossoptiles of different kinds, as in the Pukeko (pl. 4, figs. 5a and b), is to be regarded more as a fact of development than as one of phyletic significance. Feathers are ensheathed during development, teleoptiles included, and emerge sooner or later from the covering either by extension of barbs or by disintegration of sheath. It is to be remarked that so many of those birds whose neossoptiles develop from filaments are precocious to a greater or less extent, or, on account of exposure, benefit by a rapidly appearing covering. It is possible to regard the double-down as an adaptation bound up with a long nestling life in rather difficult conditions. Penguins and Albatrosses are known to undergo long periods of parental care chiefly in the open, although the young of Eudyptula albosignata, living in holes on the shore of Banks Peninsula, New Zealand, left the nest, fully fledged, at the end of about two months from hatching. Eudyptula is stated by Pycraft (1898, p. 981) to appear “to represent the least specialised form of the whole group (Impennes), and probably lies nearest the ancestral stock.” It is conceivable that the thick nestling down of the more southerly-living Penguins is a further development of a juvenile coat similar to that of Eudyptula. In this genus the first down does not completely cover the skin, a clear

feather pattern being visible, but when the second down emerges the skin is no longer visible, partly, no doubt, because each feather possesses more barbs than did the pseudo-protoptile and partly because the individual barbs are ultimately longer. This more effective covering by the second down occurs in spite of the rapid increase in the surface area of the body. The Aftershaft. The wide distribution through the birds of an aftershaft, either with or without a hyporachis, would lead to the conclusion that it is an ancient structure. Its occurrence in the neossoptiles of Ratitae, Anseres, Tinamidae, Rallidae, indicates its existence in juvenile birds since the earliest times. The absence of an aftershaft or a hyporachis from some Ratites may be regarded as due to loss. It has long been known that the adult coat of Apteryx has no aftershaft, but contrary to earlier statements (see, for instance, Pycraft, 1901, p. 163, p. XLV, fig. 61) a remnant of an aftershaft, consisting of about ten barbs carrying barbules similar to those elsewhere on the feather, has been found on feathers of juveniles of Apteryx oweni, A. mantelli and A. haasti. There is no hyporachis, the barbs being rather uniformly distributed round the lower edge of the superior umbilicus. Such an arrangement is absent from adult feathers. The barbs of the aftershaft were observed on one skin to be parallel to the skin, providing a felt-like layer beneath the rest of the plumage. The nature of the juvenile feathers of Apteryx is apparently still in doubt, particularly since there has been found no trace of connection between them and the young adult feathers (Pycraft, 1901, p. 163). According to Pycraft's account of a downy nestling of A. haasti, the prepennae fall away before the pennae have differentiated to any extent. Evidently, then, the early feather succession in these birds has undergone considerable specialisation, not only in the character of the adult feathers, but in the nature of the neossoptiles, by reduction of the aftershaft and the discontinuous production of the early feathers. Conclusions. After consideration of the foregoing facts it may be concluded that:— 1. In the history of Birds, neossoptile plumage has undergone much change. 2. In the coat of some of the Ratitae, and in some parts of the coat of Anseriformes and Rallidae, the definitive feather of the young adult is preceded by true protoptile and mesoptile or by protoptile and vestigial mesoptile, or by true protoptile only. Where the mesoptile is absent in the Neognathae, it is to be regarded as lost. 3. Protoptile, mesoptile and teleoptile form a series of structures produced in the same follicle. Each has a median rachis bearing lateral barbs. The rachis of one is morphologically continuous with that of another; the barbs of one are in the same series as those of the other but are separated from them by the calamus.

4. In many birds the definitive feather of the young adult is not preceded by a protoptile or a mesoptile but by early elongations, variously elaborated, of only the barbs of the definitive feather. The number of barbs thus extended may become so few as finally to produce no nestling down, as in the House Sparrow, Passer domesticus. 5. The growth of nestling down at first in a sheath of epitrichium, which disintegrates and leaves the barbs free, is no criterion by which to define a protoptile. 6. The appearance of protoptile, mesoptile and definitive feather (teleoptile), in that sequence, is to be considered as historically precedent to the appearance of nestling down formed as modified extensions of the barbs of the definitive feather, and the nestling downs of Penguins and Pigeons are therefore not to be regarded as representing stages of phyletic significance in the history of feathers. 7. The double down of Penguins and Albatrosses, and similar structures in other birds, and the single down of many birds are not homologous with protoptile and mesoptile of Ducks and Swans and should be given other designations. The terms suggested are as follows: single down to be called haploptile, double down to be called diploptile, the parts of the diploptile to be called pseudo-protoptile and pseudo-mesoptile. 8. The term “calamus” can be applied only to that part lying between protoptile and mesoptile, between mesoptile and teleoptile and below the teleoptile. The part lying between haploptile and teleoptile, between pseudo-protoptile and pseudo-mesoptile and between pseudo-mesoptile and teleoptile, is pseudo-calamus. 9. Those birds which lack protoptile and mesoptile may be regarded as having lost them. Haploptile and diploptile functionally replace those structures. 10. No satisfactory evidence comes forward to show clearly the relation between the haploptile and the parts of a diploptile. Summary. The general structure of the juvenile plumage of a number of birds is examined. Two main categories are found, namely, a succession consisting of protoptile, mesoptile, and teleoptile or a modification of it, such as reduction or loss of mesoptile, and an absence of protoptile and mesoptile with usually a functional replacement by precocious elongation of barbs of the teleoptile. In some birds the elongated teleoptile barbs form a single nestling-down or haploptile, in other birds those barbs form a double nestling down or diploptile, in which latter case the first down is designated pseudo-protoptile and the second down is called pseudo-mesoptile. It is considered that the protoptile, mesoptile, teleoptile succession is primitive. The presence of an epitrichial covering on a newly hatched nestling-down feather is no criterion of the nature of that feather, a protoptile or a haploptile or a pseudo-protoptile may possess such a sheath. A reduced aftershaft of barbules has been found on feathers of juvenile Apteryx spp.

Acknowledgments. Indebtedness is acknowledged to Dr. R. A. Falla, Director of the Canterbury Museum, who gave free access to and use of the collection of bird material in his care, also to Mr. E. F. Stead, of Christchurch, Professor B. J. Marples and Mr. L. E. Richdale, of Dunedin, and to the Curators of the Zoological Parks of Auckland and Wellington, all of whom provided valuable material. List of References. Clark, 1906. The Ibis (referred to by Ewart, p. 613). Duerden, 1911. Experiments with Ostriches, XVI, Agri. Jour., Union of South Africa. Dwight, 1900. Sequence of Plumage Moults of Passerine Birds of N.Y., Ann. N.Y. Acad. Sci., Vol. XIII. Ewart, 1921. Nestling Feathers of the Mallard, Proc. Zool. Soc. Kerr, 1919. Text-book of Embryology, Vol. II. London. Lambrecht, 1933. Handbuch der Palaeornithologie, Berlin. Newton, 1893–1896. Dictionary of Birds. London. Plate, 1922. Allgemeine Zoologie u. Abstammungslehre, Erst. Teil. Jena. Pycraft, 1898. Osteology of the Impennes, Proc. Zool. Soc. —– 1901. Morphology and Phylogeny of Palaeognathae and Neognathae, Trans. Zool. Soc., Vol. XV. —– 1907. Nestling Down of Penguins, Nat. Antarctic Exp., 1901–4, Vol. ii.