page break

The capture of a specimen of the rare Notornis mantelli in the South Island, is an event of sufficient importance to warrant a special memoir in our "Transactions," and I have therefore much pleasure, at the request of our president, in bringing before you this evening all the information I have been able to collect on the subject.

I may here mention—and do so with regret—that the specimen which I am about to describe is no longer in the colony, having been despatched by the Waitangi about three weeks ago for sale in England.

page 239

Every effort was made by Dr. Hector and others to retain possession of it for one of our local Museums, and immediately before its departure from our shores I wrote myself to the owner offering him fifty pounds for the skin, but to no purpose.

It will be interesting to watch its ultimate fate; but as there are already two fine examples in the National Collection, it will most probably find its way into one of the Continental or American Museums. Although we have failed to detain the prize, there is every reason to believe that the species still survives in the land, and that it will yet be added to the type collection in the Colonial Museum. It is a curious fact, illustrating the wide range of a bird supposed to be nearly extinct, that the three known examples have been obtained at localities nearly a hundred miles apart from each other, and over an interval of thirty-five years. As the species belongs to a gregarious family, and as the general character of its habitat is rough and inaccessible in the extreme, I think it may be fairly inferred that many yet survive to reward the future search of the Southern naturalist.

The two fine specimens now in the British Museum (supposed to be male and female) were obtained through the exertions of our former president, the Hon. Walter Mantell, after whom the bird was named. The first of these was captured alive in 1849 by a party of sealers at Duck Cove, on Resolution Island, Dusky Sound. "Perceiving the trail of a large and unknown bird on the snow, with which the ground was covered, they followed the footprints till they obtained a sight of the Notornis, which their dogs instantly pursued, and after a long chase caught alive. It ran with great speed, and upon being captured uttered loud screams, and fought and struggled violently. It was kept alive three or four days on board the schooner and then killed, and the body roasted and eaten by the crew, each partaking of the dainty, which was declared to be delicious." The second of Mr. Mantell's specimens was caught by the Maoris on Secretary Island, opposite to Deas Cove, Thompson Sound. This also was eaten, but fortunately the skin was preserved and sent to England to join the other, and (as already mentioned in my "Birds of New Zealand") these members of an expiring race, "having been carefully mounted by Mr. Bartlett, now stand side by side in the National Collection of Great Britain, and, like the remains of the Dodo in the adjoining case, daily attract the attention of thousands of eager visitors."

The third specimen to which I have specially to refer this evening, was obtained last year, on what are called the "Bare-patch Plains," on the eastern side of To Anau Lake. The circumstances of the capture were thus narrated to me by Captain Hankinson, on whose property it occurred. A man who was engaged "rabbiting" on the run, had camped on the page 240 Maruroa Flat, not far from the homestead. One day his dogs ran down a large bird, and on coming up he found it alive and unharmed. Taking the bird from the dogs he deliberately killed it, took it to his tent and hung it up to the ridge pole. On the following day the station manager (Mr. J. Connor), in making his customary round, visited the camp. The rabbiter had just struck his tent, and calling the manager's attention to the dead bird, still suspended to the ridge pole, told him he might have it. Mr. Connor, who was intelligent enough to suspect that he had found a Notornis, at once accepted the offer and took the bird home to the station, where he carefully and very successfully skinned it, preserving also all the bones of the body.

The weather had been exceptionally severe, and it is supposed that this was how the Notornis came to be found on the flats, having been driven down from the high country. The man who caught it said that it seemed quite tame, whereas Mantell's bird (as already mentioned) made a vigorous resistance on being taken.

Professor Parker having undertaken to describe the skeleton for our "Transactions," Dr. Hector invited me to undertake the same duty in regard to the skin, in order that, in default of the specimen itself, we might have on record in the colony as complete a monograph as possible of this interesting bird. I cheerfully undertook the task, and made a visit to Dunedin specially for this purpose.

On being introduced to this rara avis I experienced again the old charm that always came over me when gazing upon the two examples in the British Museum—the lingering representatives of a race co-existent in this land with the colossal Moa! Then, retiring to the Museum Library, I shut myself in with Notornis, handled my specimen with the loving tenderness of a true naturalist, sketched and measured its various parts, and made a minute description of its plumage.

Like many other New Zealand forms of an earlier period, the Notornis is the gigantic prototype of a well known genus of Swamp Hens. It is, in fact, to all appearance a huge Pukeko (Porphyrio), with feeble or aborted wings, and abbreviated toes, the feet resembling those of Tribonyx—a bird incapable of flight, but admirably adapted for running. Similar, no doubt, was the relation borne by the powerful Aptornis to our present Woodhen (Ocydromus); but in that case the prototype has disappeared, leaving only its fossil bones for the study of the scientist, and its place in nature to be filled by its existing diminutive representatives.

The interest attaching to Notornis has been greatly enhanced by the discovery that the white Swamp Hen, of Norfolk Island, belongs to the same genus, as this has an important bearing on the study of geographic distribution.

page 241

The characters of the genus Notornis were first determined by Professor Owen, in 1848, from certain fossil remains collected by Mr. Mantell in the North Island of New Zealand, and consisting of the skull, beaks, humerus, sternum, and other parts of the skeleton of a large brevipennate Bail. The sagacity with which the learned professor had interpreted these bones, and the absolute correctness of his prevision, were exemplified in the discovery which enabled Mr. Gould, in 1850, to communicate to the Zoological Society the complete generic characters of the bird, already known to science as Notornis mantelli, Owen. In illustration of these, Mr. Gould furnished to the society a coloured sketch of the head of Notornis, in his usual artistic style; and at a later period he published, in the supplement to his "Birds of Australia," a full-sized drawing of the bird. These plates are very beautiful, but on a close comparison with the specimen to which these notes more especially refer, I find that some of the minor features have been overlooked by the artist, or sacrificed to pictorial effect. In the following descriptive notes, I have, therefore, deemed it best to record the characters (generic as well as specific) with some minuteness of detail.

The bill is somewhat shorter than the head, greatly compressed on the sides, and much arched above, the culmen having a convex or rounded aspect, with a uniform width of ? of an inch from above the nostrils to within half an inch of the tip, when it rapidly diminishes, terminating in a rounded point. Where it merges into the frontal shield, the culmen is ? of an inch in width. Gould has somewhat exaggerated in his drawings the angle of declination towards the corners of the mouth, also the serrated edge of the upper mandible. In this specimen there is only the slightest indication of pectination. The cutting edges of both mandibles are sharp to the touch. The horny covering of the bill rises on the forehead to a line with the posterior angle of the eye, forming a depressed frontal shield (not arched as in the drawing). Nostrils oval, placed in a depression near the base of the bill, and forming an oblique opening, nearly twice as large as shown in Gould's sketch of the head (Proc. Zool. Soc.). Wings short, rounded, and slightly concave; ample in appearance, but useless for purposes of flight; first quill shortest, second half an inch shorter than third; third fourth and fifth longest and about equal; sixth scarcely shorter than fifth. On examining the wing-feathers they are found to be feeble and pliant, the outer webs being almost as broad as the inner. The tail-feathers are likewise soft and pliant, with disunited filaments, much worn at the tips. The tarsi are long, strong, and well proportioned to the bird; longer than the toes (exclusive of claws), rounded in form, and armed in front with fourteen more or less broad, regular, transverse scutellæ, forming an effective shield; on the middle toe there are twenty-three transverse scales, page 242 all very regular, but narrowed at the joints; on the inner toe fifteen, and on the outer toe twenty-one. On the hind too there are five scales. The claws are strong, thick, not much arched, rather sharp on the edges, but with blunted points, especially on the hind toe. The palate is deeply grooved.

Head and upper part of neck very dark blue, changing according to the light into brownish-black on the crown and nape, brighter on the cheeks and sides, and passing into dark purplish blue on the lower part of the neck; the whole of the back, rump, and upper tail-coverts rich olive-green, varied more or less, and particularly on the shoulders, with dull verditer green, the feathers shading off into that colour at the tips, the general olive hue, however, predominating towards the sides of the body; foreneck, breast, sides of the body, and inner portion of flanks beautiful purplish-blue; the lengthened pectoral plumes which overlap the sides and the outer portion of flanks vivid purplish-blue, mixed and varied, especially on the former, with verditer green; abdomen, thighs, and vent dull indigo or bluish-black, more or less mixed with brown; under tail-coverts pure white. The general upper surface of the wings is a rich mixture of blue and verditer green, very difficult to express exactly in words, the combination having something of the effect, in certain lights, of lapis lazuli.

On a close examination of the larger coverts it is found that they are marked transversely with numerous delicate rays of a darker purplish blue, adding much to the beauty of the plumage. On the lesser coverts this rayed character although present is less conspicuous, and the olive hue is more pronounced, while on the scapulars it becomes predominant, resembling the plumage of the back. The outer edges of the wings and the tertial plumes very rich purplish-blue or obscurely rayed with green. The outer primaries are blue on their outer webs, but this rapidly changes to dull sea-green, which colour prevails on both webs of the secondaries, only washed with a brighter tint on the outer vane. This colour deepens again into olive on the inner secondaries and their coverts, thus harmonizing with the plumage of the back. The under surface of the quills is uniform blackish-brown, and the shafts are white towards the base; the axillary plumes and the larger inner coverts are of the same colour tipped on their outer aspect with blue, and the smaller coverts, which are of very soft texture, are entirely blue. The tail-feathers are dark olive mixed with verditer green on the upper surface and changing to dull olive-brown, with lighter shafts, on their under surface.

The bill has lost its original colour through being dried. On the frontal plate and along the basal edges of both mandibles it appears to have been dark red, fading outwards. The culmen still has traces of its original pinky page 248 colour; but the sides of both mandibles, in the present condition of the specimen, are reddish horn colour, fading to whitish horn along the cutting edges. The tarsi and toes appear to have been originally light-red, having now faded to a transparent reddish-brown, paler on the toes. Claws dull brown, lighter towards the tips.

The texture and general appearance of the plumage on the head, neck, and under parts generally, is very similar to that of the Pukeko (Porphyria melanotics), although the latter bird lacks the produced bright-coloured pectoral plumes which overlap the sides of the body, under the wings, in Notornis. The plumage of the back is very long and thick, but at the same time soft and somewhat silky to the touch, being evidently adapted to haunts where the bird is constantly subject to drippings from wet herbage. On moving this plumage with the hand it is found that the basal portion, comprising more than two-thirds of the feathers, is of a uniform blackish-brown, whereas the basal plumage on the other parts of the body is dark grey. The plumage of the head and neck is short and close, as in Porphyrio, the feathers having a soft texture. The whole of the upper surface has a slight sheen upon it (amounting almost to a glint on the tips of the shoulder-plumage), and the bright hues of colour on the back and wings change slightly under different lights. The plumage covering the flanks and overlapping the thighs is dense and long, while its brilliant blue and green colours contrast strongly with the olive plumage of the back and rump. When looked at in front, with the wings closed in against the body, the purplish vivid blue already described is very conspicuous. The carpal spur is shaped like the claw of the hind toe, but is less arched. It is nearly one-eighth of an inch thick at the base, and is dark brown, fading into horn-colour at the tip.

Measurements.—Approximate length (measuring from tip of bill, following its curvature, and from the forehead to the end of the tail) 24.5 inches wing, from flexure, 10; from humerus to flexure 3.75; carpal spur .4; tail (to extreme tips) 4.75; bare part of tibia 1; tarsus 3.5; middle toe 3, its claw 1.1; inner toe 2.2, its claw 1; outer toe 2.4, its claw .8; hind toe .75, its claw .75. Bill, from posterior edge of frontal plate to tip of upper mandible, 3.4; from gape along edge of upper mandible 2.5; along edge of lower mandible 2.25; greatest width of bill, measuring across from the summit of the arch, or culmen, to the junction of the rami, 2.

Observations.—Taken altogether, the specimen is a very fine one—probably an adult female. The plumage is somewhat worn, the primaries and tail-feathers having their webs more or less abraded on their outer edges and tips. The edges and sides of the mandibles are considerably worn, indicating a fully adult state. The claws of the toes, and particularly that of page 244 the hind toe, appear to be much blunted by use. The colours of the plumage generally are brighter than in the supposed female specimen in the British Museum, but they are, I think, less brilliant on the whole than in the British Museum male: notably there is an entire absence of the well defined terminal margins of verditer green on the wing-coverts which form crescentic bands in the type specimen. There are, however, as mentioned above, different blending shades of green and blue on the plumage of the wings, which impart to it a very beautiful appearance. My recollection of the ♀ specimen in the British Museum collection is that it has these crescentic markings far less conspicuous than in the male.

Note.—There appears to have been originally very little colour in the beak except on and below the frontal shield and along the basal edges of both mandibles. The legs are in much the same condition as that presented by the legs in a dried Pukeko skin, the colours having faded out. But there is enough colour left in the tarsi to show that the legs and feet were originally, as described above, a light (probably pinkish) red. The skin is much stretched by unskilful treatment after being removed from the body; but I have allowed for the stretching in taking the measurements given above.

I remarked to Professor Parker, on first taking up the specimen, that the legs appeared to be more attenuated than in the British Museum examples, and the measurements which I afterwards made, as given above, prove that the toes are somewhat longer proportionately to the size of the bird, which is altogether slightly larger than the type specimen described in my "Birds of New Zealand." The frontal shield is, however, somewhat smaller, being just one inch across in its widest part, and ascending barely half an inch from the base of the culmen. It has a corrugated, shrivelled appearance in the dried specimen, and from the sides of the bill, at its base, the cuticle is inclined to peel off. The skin (in the dried state) is very tough, having the appearance and consistency of fine leather.

Hab.—South-west portion of South Island. As already mentioned the first recorded specimen (in 1849) was obtained on Resolution Island, the second, nearly three years later, on Secretary Island, in Thompson Sound, and the third, which has formed the subject of this paper (in December, 1879), on the eastern side of Te Anau Lake. Taking these three localities as marking the points of a triangle describing the ascertained limits of its occurrence, we have before us the present range of Notornis over a considerable area of very broken and rugged country. As its fossil remains testify, its ancient range was far more extensive, including the North Island, and in prehistoric times probably reaching much further.