[Image of page 1]
PROCEEDINGS OF THE ZOOLOGICAL SOCIETY OF LONDON.
January 11, 1848. William Yarrell, Esq., Vice-President, in the Chair.
The following communications were read: --
1. ON THE REMAINS OF THE GIGANTIC AND PRESUMED EXTINCT WINGLESS OR TERRESTRIAL BIRDS OF NEW ZEALAND (DINORNIS AND PALAPTERYX), WITH INDICATIONS OF TWO OTHER GENERA (NOTORNIS AND NESTOR). BY PROFESSOR OWEN, F. R. S. ETC. ETC.
In this memoir (No. III.) Professor Owen confined himself to the description and comparison of the bones of the head and beak, forming part of a very extensive and valuable series collected by Mr. Walter Mantell in a deposit of volcanic sand at Waingongoro, North Island of New Zealand. After enumerating the principal bones, now in the possession of Dr. Gideon Mantell, F. R. S., by whom Prof. Owen had been kindly invited to determine and describe them, and stating the species to which the majority were referable, viz. Dinornis giganteus, D. casuarinus, D. didiformis, D. curtus, Palapteryx ingens, P. dromioides, P. geranoides, the author alluded to a form of tarsometatarsal bone, which had supported a strong back-toe, and resembled the metatarsus of the Dodo, but was shorter and thicker, as apparently belonging to the tibia of the species described in a former memoir (Zool. Trans, iii. 1843, p. 247), to the Dinornis otidiformis, but which must belong to a genus (Apterornis) distinct from both Dinornis and Palapteryx. He also stated that the collection contained many bones of seals of the genus Arctocephalus, F. Cuv., with a few bones of a dog and of the human subject: the latter had been calcined, and were probably the remains of some cannibal feast of the natives. The uncalcined bones of the seal were in the same state, brittle, absorbent, and of a yellowish brown colour, as the bones of the extinct birds, with which they were associated and appear to have been
No. CLXXX. --PROCEEDINGS OF THE ZOOL. SOC.
[Image of page 2]
coeval. Numerous fragments of the shells of more than one kind of egg, the largest surpassing in size the egg of the ostrich, had also been discovered with the bones.
In the present memoir Prof. Owen described the bones of the head and beak. They belonged to four distinct genera of Birds. The largest skull, with a very strong, broad, subelongate and subincurved beak, like an adze, was referred to the genus Dinornis. The second in size, with a beak to which that of the Emeu makes the nearest approach, was referred to Palapteryx. The third skull, with a beak like that of the Porphyria and Brachypteryx, was referred to the same family--'Rallidae'--to which those genera belong; but, through the peculiarities of the cranium, formed the type of a new genus, Notornis. The fourth form of beak was referable to the genus Nestor in the family Psittacidae.
The cranium of the Dinornis presents the family characters of great breadth, and forward inclination of the occipital region, of the vertical plane of the occipital foramen, and of the prominent and pedunculate occipital condyle; but the downward development of the basi-occipital and basisphenoid is exaggerated, as compared with the Palapteryx, the basis cranii, which is 2 1/2 inches in length, descending abruptly for the extent of 1 inch below the foramen magnum; the condyle is hemispherical as in Otis, not a quarter of a sphere as in Struthio and Palapteryx, nor, as in Didus, a transverse reniform tubercle with a median notch above. The foramen magnum is a vertical ellipse, with lateral processes encroaching upon it, as in Didus; but in this large extinct bird the upper half of the foramen is narrower and almost pointed above. In Apteryx and Palapteryx the foramen is widest transversely. The margin of the foramen magnum is broad and excavated in both Dinornis, Otis and Didus, but the upper border ends in the latter genus in a tubercle on each side.
In Didus there is a small middle supraoccipital foramen and two lateral ones, but these do not exist in Dinornis, Otis, or Palapteryx: the lateral foramina are present in Apteryx.
In the extinct genera and in Otis the supraoccipital ridge is well-marked, but denned rather by the subsidence of the occipital surface than the elevation of the ridge above the parietal one.
In no bird is the extent of surface for muscular attachment so great at the back part of the head, or so strongly marked by depressions and ridges, as in the Dinornis.
The extension of the surface by the downward thick wedge-shaped development of the basi-occipito-sphenoidal surface, and by its lateral strong backwardly produced ridges, is quite peculiar to the Dinornis. An approach to this structure is made by Otis in the ridges that connect the sides of the flat basisphenoid 1 with the paroccipital 2 processes. In Palapteryx the basi-sphenoid is square and flat below, in Didus it presents a longitudinal channel bounded by parallel lateral ridges; the sides of the basisphenoid, which incline to these ridges,
[Image of page 3]
are slightly concave, have two perforations posteriorly, one above and a little in advance of the other, and form the anterior and internal boundary of the tympanic cavity.
In Palapteryx, as in Didus, the basioccipital descends and expands into two thick obtuse processes, from which muscles pass to the inwardly-bent angles of the jaw. Internal to these processes are two short tubercles. On each side the base of the occipital condyle in Dinornis are three small foramina; in Didus two, the outer one the largest.
In Dinornis, Otis and Didus, two foramina, the upper one for the hypoglossal nerve, the lower one for the entocarotid artery, open externally in a deep elliptic depression. The paroccipital is enormously developed in Dinornis, and sends a rough thick process from its under part to abut against the lateral basioccipital ridge, where it articulates and sometimes anchyloses with the stylohyal: in Palapteryx and Didus the paroccipital carries the posterior surface of the skull downwards and outwards in a minor degree than in Dinornis, and terminates in a curved convex thick border: its internal surface next the tympanic cavity is cellular in Didus. The eustachian outlets open, in both Dinornis and Otis, above a transverse ridge terminating the basisphenoid anteriorly: this ridge is not present in Apteryx or Palapteryx. The Palapteryx also differs from Dinornis in the higher position of the precondyloid holes and their greater separation from the carotid holes, in the minor development of the paroccipitals, the major development of the mastoids, and by the large and single oblong depression, beneath the mastoid, for the single superior condyle of the tympanic bone. In Dinornis the temporal fossa is wide and deep, in Didus narrow and deep; the alisphenoid is concave where it ascends to coalesce with the mastoid, parietal and post-frontal to form the temporal fossa: the limits of the orbitosphenoid are also obliterated by a similar confluence: in this region of the skull the 'foramen ovale' is preceded in Dinornis as in Didus by two smaller foramina, and in front of these is the great 'foramen opticum.' The parietals are very broad and short in both extinct genera; but in Dinornis there is a median rising where the sagittal suture originally ran, whilst Didus shows a depression and foramen here. The mastoid in Dinornis, as in Otis, sends down two processes, one, the tympanic process, short, --the other, or proper mastoid process, long; this coalesces with the postfrontal in Dinornis, not in Otis: the base of the mastoid has two articular cavities for the upper condyles of the tympanic bone. In Didus the outer side of the mastoid is convex, smooth, but with a slight oblique ridge; it overhangs the tympanic cavity, bending inwards, and sends a short compressed pointed mastoid process in front of the anterior articular cavity for the anterior and upper condyle of the tympanic.
The presphenoid is a deep compressed plate, thickened and rounded below; the palatines abut against it, as in Didus, where the fore-part of the pterygoids also rest in part upon the presphenoid. The frontals of Dinornis form together a broad hexagonal plate moderately convex, with the cerebral hemispheres indicated by very
[Image of page 4]
slight risings; the postfrontals form the depressed lateral angles; the anterior border is emarginate and coalesces with the nasals and pre-maxillary, without being elevated above them. In Palapteryx the frontals are more produced anteriorly before coalescing with the base of the beak. In Otis the interorbital part of the frontals is deeply and widely excavated. In Didus the frontals are broad and convex, rising singularly above the cranial ends of the nasals and premaxillary, with which they also coalesce. The supraorbital plate presents a rough notch near the fore-part, where in Dinornis there is a shallow emargination. In Dinornis there is a shallow depression with vascular grooves at the outside of the base of the postfrontal distinct from the temporal fossa: in Didus the temporal fossa extends forwards above the postfrontal and forms there a reniform depression, either for a gland, or what is less likely, for a co-extension of the origin of the temporal muscle. The postfrontal is a strong triangular obtuse process, ending freely as in Palapteryx, not joined to the mastoid as in Dinornis. The orbitosphenoids, indicated by the optic foramina, continue the roof and septum of the orbits by coalescence with the alisphenoids behind, the frontals above, the prefrontals in front, and the presphenoid below: they send a ridge upwards and outwards to the under part of the postfrontals, but do not present that singularly swollen character which is so peculiar in Didus; in which also the prefrontals form a large smooth protuberance, like a tumour, at the fore-part of the orbits, and appear on the upper surface of the cranium in front of the antorbital process of the true frontal and external to the lachrymal. The interorbital bony septum is entire in both Dinornis and Didus; but in the latter it is more than an inch in thickness and cellular, and in this respect more resembles the singular structure of the part in Apteryx. The orbits are smaller in Dinornis than in the large existing Struthionidae or in Otis, but are larger than in Apteryx. The olfactory chambers in Dinornis are less developed than in Palapteryx and Apteryx.
The nasal bones in Dinornis and Otis converge where they over-lap the prefrontal (ethmoide, Cuv.) in order to join the frontal and include that end of the nasal process of the premaxillary, which is on a lower plane; and, as they advance, they pass beneath that process, coalesce with it and with each other, and terminate in Dinornis in a point. In Didus the nasals also anchylose with the frontal, where they are separated by the nasal process of the premaxillary, as indicated by the two longitudinal fissures, which, commencing behind at 2 lines distance from the outer border of the anchylosed base of the beak, gain that border at 1 inch 9 lines distance from the frontal, and thus indicate the proportions of the base formed by the anchylosed nasals: the fissure can also be traced as in Dinornis, bending inwards upon the under surface of the nasal process of premaxillary, to about 3 inches from the frontal, when the fissure returns back, inclining to the median line, and meets its fellow there. All the outer part of the median stem or base of the beak defined by these linear furrows I regard as the nasals, which thus support the nasal process of the premaxillary.
[Image of page 5]
This process is a broad transversely arched plate, where it joins the maxillary processes to form the anterior or rostral part of the premaxillary; the extent of which, anterior to the external nostrils, is inches, the whole length of the premaxillary being 4 1/2 inches. Its breadth at the middle is rather more than an inch; the depth of the upper bony beak gradually decreases from its base where it is 1 inch 9 lines, to its apex where it is less than 1 line, but retains a breadth of 8 lines, the edge appearing to have been truncate or very slightly rounded off: the whole upper beak being gently arched to this terminal edge resembles the cooper's adze (doloire, Fr.). The palatal surface is broad, very slightly excavated, and bounded laterally by well-defined alveolar ridges: the palatal nostril commences anteriorly 1 inch 10 lines from the anterior border of the premaxillary. In Didus the nasal process of the premaxillary presents an elliptic transverse section where it quits the maxillary processes, and diminishes in depth as it retrogrades, becoming depressed and broad where it rests upon and divides the nasals to anchylose with the frontal. Where the nasal and maxillary processes diverge, there is a deep groove externally terminating in a canal directed forwards into the rostral part or body of the premaxillary; this part is subincurved, pointed, rough and with irregular vascular perforations, with a sharp inferior border on each side, and a more concave palatal surface than in Dinornis. The long and slender palatines of Dinornis coalesce behind with the vomer and in front with the mamillaries; they are concave below, particularly at their back part, by the downward extension there of their inner border. In Didus the palatines arch outwards from their posterior attachments, are broad and smooth mesially with a sharp crenate edge above; a thin, outwardly smooth, convex ridge is directed outwards and downwards, and a more angular ridge is directed downwards with an obtuse apex: a groove divides this from the outer ridge: the upper and outer ridge extends to the maxillary; the lower ridge subsides before it reaches the maxillary. The palatines form the boundaries of the naso-palatine aperture, and approximate each other at both their ends, but do not meet. There is a fossa at the outer and near the back part of each palatine, where there is a rough concavity; the rest of the outer surface is convex lengthwise, concave vertically. The boundaries of the maxillary are more readily traceable in Didus than in Dinornis; but they have coalesced in both, with the palatine, malar and lachrymal behind, and with the maxillary process of the premaxillary in front: the maxillary in Didus forms a compressed longitudinal plate of bone with thick rounded borders above and below, and almost touches its fellow, leaving a deep narrow chink between the nasal fossa above and the palate below, closed by the palatal membrane.
The tympanic bone of the Dinornis has more a triangular than a quadrate form by reason of the unusually large size of its inferior condyle, which forms its base: the orbital process is a compressed subrhomboidal plate: the lower condyle is not so extended inferiorly in the Bustard (Otis); its upper condyle is bifid, as in Dinornis. In
[Image of page 6]
Palapteryx it is single, as in Apteryx. In Didus the tympanic bone is subquadrate with the four angles produced, and the upper and hinder one bifurcate, forming the bifid condyle for the mastoid articulation: in Dinornis the mastoid condyle is also double, with a linear strip of bone between; and behind this the pneumatic foramen, where also similar foramina are situated in Didus in this extinct bird, the orbital process, forming the anterior angle, is compressed and truncate: the outer surface of the bone is smooth and convex vertically; the inner surface is traversed by a sharp concave ridge extending from the inner division of the upper condyle to the anterior part of the inner and lower angle: the anterior division of the inner surface is concave, the posterior one is concave vertically, convex transversely. The antero-posterior extent of the condyle for the lower jaw is little, but greatest at its outer part, where it rests upon the shallow reniform outer division of the concave articular part of the lower jaw: the inner, more ridge-like part of the condyle sinks into a deeper transversely extended depression of the same articular concavity. The tympanic of the Dinornis chiefly differs in the great extension, upwards and backwards, of the broad and undivided inferior condyle: there is also an articular surface, on its outer side, for the mastoid process (not present in Otis) and another small one on the inner side for the pterygoid; besides the lower and outer cup for the end of the slender zygoma (squamosal).
The inner angle of the expanded articular end of the lower jaw of Dinornis ends by a short obtuse process. In Otis and Didus it forms a strong trihedral process, the anterior and posterior facets meeting a transverse ridge below, which is continued into a compressed plate forming the posterior angle of the jaw. The posterior surface is smooth and slightly concave, semioval in Dinornis, deeper and subtriangular in Didus.
The outer part of the articular end of the mandible is smooth and convex in Dinornis: in Didus a masseteric ridge is continued downwards and forwards from the outer overhanging border of the articular cavity to the back and lower angle of the dentary piece, defining, with the posterior border of the dentary, a concave, slightly pitted surface. The surangular in Dinornis has a short and low thick coronoid ridge, external to which there is a rough oval surface. In Didus the surangular developes a very small coronoid process, and its fore-part is deeply notched: a deeper and more angular notch divides the surangular from the angular piece. This notch receives the lower fork of the dentary on the outside, and the end of the splenial at the inner side. These notches do not exist in Dinornis: the surangular, angular and articular pieces have coalesced together in both the extinct birds. Where they join the posterior forks of the dentary piece, a long narrow vacuity is left, which in Dinornis is almost divided by a broad bar of bone extending upwards from the angular, but which does not quite touch the surangular. In Didus the upper fork of the dentary joins the upper and fore part of the surangular; the notch between the hinder forks of the dentary bounds anteriorly the narrow elliptic vacuity, 15 millimeters long by 3 milli-
[Image of page 7]
meters deep. A notch also extends forwards, and divides outwardly the symphysial from the ramal part of the dentary: this notch or hole does not exist in Dinornis.
The parts of the bones of the beak referred to Palapteryx consist of the anterior end of the premaxillary and of the symphysis and part of both rami of the mandible. The premaxillary, by the proximity of the external nostrils to its apex, and by the nasal grooves continued thither on each side from the anterior boundary of the nostrils, resembles that of the large existing Struthionidae, and the Emeu more especially by the slenderness of the nasal process of the premaxillary and the angle at which it rises from the broad and flat maxillary processes. The end of the beak was, however, more obtuse than in the Emeu, and the short symphysis of the lower jaw is more deeply excavated above: it presents, however, the two parallel longitudinal grooves on its under part, as in the Emeu and Ostrich. The lower jaw appears from the remains of one ramus to have been 5 inches or 5 1/2 inches in length, and to have been broader and deeper than in the Ostrich or Emeu: and the cranium by its greater breadth behind, its less depth, its vertical foramen magnum and prominent occipital condyle, the lower position of the basisphenoidal platform, and the marked angle which it forms with the almost vertical basi-occipital, concurs with the beak in establishing the generic distinction of the great bird to which it belonged. As the characters which were adduced in a former memoir (Zool. Trans, iii. p. 327) to separate those bones of the extremities that by their more slender proportions approximated the Struthionidae and, by the indication of a small back-toe, the Apteryx more particularly, from other bones of corresponding size but more robust proportions and devoid of a back-toe, --led to the former being assigned to the genus Palapteryx, and the latter to Dinornis proper; --so the characters which, in the first of the skulls described in the present memoir, show a departure from the struthious type, and in the second skull an approach thereto, clearly indicate the propriety of assigning the one to the genus Dinornis and the other to the genus Palapteryx. The total length of the skull referred to Palapteryx geranoides is 6 inches at least; the breadth of the cranium 2 1/2 inches: the bird probably equalled the Emeu in size.
The skull which indicates the third genus of apparently extinct bird (Notornis) measures 4 1/2 inches in length, and the cranium is 1 inch 8 lines in breadth. The bones of the beak closely resemble in form and structure those of the Purple Coot (Porphyrio), but the occiput is relatively broader, and more inclined forwards as it ascends: the plane of the occipital condyle is vertical, and the basioccipital extends further below the occipital condyle, though less so than in Palapteryx. In these characters the Brachypteryx or Short-winged Rail of New Zealand more resembles Notornis. The articular surface of the tympanic is divided, as in Dinornis and Otis, into two subcircular cups. The parietal region is singularly flat, the temporal fossae unusually long, well-defined by ridges extending from the paroccipital to the postfrontal. In the comparatively small Porphyrio and Bra-
[Image of page 8]
chypteryx, in which, as in all small birds, the cerebral hemispheres, as requiring a certain bulk for their functions, do not decrease in the ratio of the size of the body, the upper surface of the cranium is raised by the hemispheres beneath into a smooth convexity.
The Notornis is a large modified form of the same natural family of the Grallae as the Porphyrio and Brachypteryx, and from the form of its sternum it must have been, like the latter peculiar bird of New Zealand, deprived of the power of flight.
The fourth genus of bird indicated by portions of the skull in Mr. Walter Mantell's collection was referable to the family of Parrots (Psittacidae), and amongst these to the genus Nestor. The bony portion of the upper beak--the only part of the skull preserved--by its deep, subcompressed, curved and pointed form, its seeming solidity, pierced by small subcircular nostrils close to its base, attested the family character; whilst the proportional length as compared with the depth, the narrow upper surface to where it suddenly expands above the nostrils to join the cranium, the absence of the notch on the under border, the very narrow elongated triangular palatal surface, with the median linear notch at its base, --all demonstrate that in this characteristic part of the skull the New Zealand bird represented by it most resembled the genus Nestor, --a singular nocturnal Parrot at present only known as a denizen of that island.
Thus then it appears that the indications of two genera, with several species of terrestrial birds of large or gigantic size, deduced in the Author's former Memoir (Part II.) from bones of the legs, are most fully and satisfactorily confirmed by the evidence of the subsequently received bones of the head and beak.
The form and structure of these characteristic parts in one of the genera (Dinornis) are so peculiar, that the author does not refer the genus to any known natural family of birds. Its location in the order Struthionidae implies little more than an arrested development of wings, and an exaggerated development of legs, organized for progression on dry land.
As, however, there are strictly aquatic forms of birds deprived, by a low development and special modification of the wings, of the power of flight, so also there are, in other natural groups of birds, aberrant forms similarly debarred from the privilege and enjoyment of the characteristic kind and field of locomotion of their class. Apart from the true Struthionidae, we have an instance of this in the Brachypteryx or modified Rail of New Zealand; the Dodo is a second instance, whether it be regarded as an aberrant Vulture or a modified Pigeon, according to the views entertained by Mr. Gould and supported, with new arguments, by Mr. Strickland, before the British Association at Oxford, and which will be fully elucidated in the forthcoming work on the extinct flightless birds of the Mauritius and neighbouring isles, which Mr. Strickland is about to publish in conjunction with Dr. Melville.
With regard to the natural group or family of birds to which the Dinornis, with its adze-like bill and crocodiloid cranium, may be referable, the author pointed out several marks of resemblance in the
[Image of page 9]
skeleton of the Bustard to the Dinornis, which are not presented by the skeletons of the true Struthionidae. But he also dwelt upon the peculiar characters of the Dinornis, distinguishing it from the Otidae, and indicating it to form a distinct family-type in the order of Grallae.
With regard to the peculiar form of beak in Dinornis, reference was made to the deductions in the former memoirs, "from the unusual strength of the neck," that the Dinornis would be found to have a beak applicable "to a more laborious task than the mere plucking of seeds, fruits or herbage;" and that "the robust proportions of the cervical vertebrae, especially of their spinous processes, may have been the foundation of those forces by which the beak was associated with the feet in the labour of dislodging the farinaceous roots of the ferns that grow in characteristic abundance in New Zealand."
For this labour the beak of the Dinornis, formed after the model of the adze or pick-axe, seems peculiarly adapted, and the singular development in both breadth and depth of the occipital part of the cranium, with its strongly marked ridges, processes and muscular depressions, is precisely calculated for the adequate attachment of the muscular masses arising from the cervical vertebrae.
The second form of cranium and beak, referred to the genus Palapteryx, indicates that genus to be a member of the true Struthionidae, and by its affinities to have been intermediate between Dromaius and Apteryx.
The Notornis is a struthious or brevipennate form of the Rallidae, intermediate between Porphyrio and Brachypteryx. The remains of the beaks of the Psittaceous bird are not distinguishable generically from those of the genus Nestor of New Zealand.
Thus, observed Prof. Owen, "those concordances in the geographical distribution of existing and recently extinct forms of the warm-blooded vertebrate classes which are illustrated by the remains of Elephants, Rhinoceroses, Hippopotamuses, Hyaenas, large Bovines and Cervines, in the pleistocene deposits of Asia and Europe, --by the absence of these and the presence of gigantic extinct Sloths, Armadillos and Anteaters, in the coeval deposits of South America, and of huge fossil Kangaroos, Wombats and Dasyures in the bone-caves and freshwater deposits of Australia, --have received new and striking elucidations from the repeated discovery, in the cavernous fissures, turbaries, and river-beds of New Zealand, of the remains of gigantic forms of birds allied to those small species, Apteryx and Brachypteryx, which constituted the highest representatives of the warm-blooded classes in the island, until the advent of Man led to the introduction of its present terrestrial mammals."
The author in conclusion repeated his acknowledgments to Dr. Mantell for the prompt accordance of the privilege of examining and describing these rare and interesting remains and expressed his high sense of the scientific value of the labours by which that eminent geologist's intelligent and enterprising son, Mr. Walter Man-
[Image of page 10]
tell, had made so great an addition to the materials for developing the natural history of New Zealand.
The memoir was accompanied with numerous drawings of the specimens described, which will form plates 52-56 of the third volume of the 'Transactions.'
On the conclusion of Professor Owen's communication, Dr. Mantell expressed his opinion, that although the specimens formerly sent to this country were obtained from the beds of rivers and mountain-streams, and were regarded by the gentlemen who collected them as of very recent date, in reality they belonged to a period of as high antiquity, in relation to the surface-soil of New Zealand, as the diluvium containing bones of the Irish Elk, Mammoth, &c. to that of England. He observed that Mr. Colenso, Mr. Taylor, and Mr. Williams, who sent to England the bones figured and described by Professor Owen in the 'Zoological Transactions,' vol. iii., agree in this remarkable fact, that in some places, where the loamy marl in which their specimens were found was observed in situ, it was covered by several feet of strata of marine and freshwater sand, gravel and silt. The bones collected by Mr. Walter Mantell, among which were the crania and mandibles that formed the subject of Professor Owen's present communication, were all found imbedded in a loose pure sand, formed in a great measure of magnetic iron and minute crystals of augite and hornblende, the detritus of volcanic rocks. This sand has filled all the cavities and cancelli of the bones, but is not in any instance consolidated together: hence the bones are in the most beautiful state of preservation, and the most delicate processes entire. Dr. Mantell conceives that this bed of volcanic sand is a continuation of the deposit of sandy loam which occurs at the embouchures of the rivers along the west and east coasts of the North Island, in the localities that yielded the bones sent over by Mr. Williams and Mr. Taylor; and that in the higher regions of the same river-valleys, the detritus brought down by the mountain-streams from the volcanic chain whence they originate, is unmixed with the clay and silt of the lower alluvial tracts; for all the streams in these parts of the North Island rise from the lofty ridges of Mount Egmont and Tongariro. Dr. Mantell alluded to the fact, that along the sea-coasts and on the banks of the rivers Eritonga, Waiho, &c, there are horizontal terraces of boulders of trap-rocks fifty feet high; and that the small rocky islands of trachyte off the coast bear marks of wave-action to the height of 100 feet above the present sea-level. He mentioned other facts of a like nature in confirmation of his opinion, that since the Moas existed the surface of the country has been elevated many feet above the level of the sea, and that the present rivers and mountain-streams are flowing through channels cut into the ossiferous deposits; in like manner as the rivers of Auvergne flow through the newer tertiary marls and limestones containing bones of Mammalia, and those of England through the diluvial clay and loam in which
[Image of page 11]
are imbedded the remains of the large extinct Pachyderms, the Rhinoceros, Mammoth, &c. He deemed it probable that the last of the race of Moas were destroyed by the earliest inhabitants of New Zealand, as the Dodo was finally extirpated by the Dutch colonists of the Mauritius, and the Irish Elk by the early British or, Celtic tribes; but he considered it evident that the bone-deposit was in the progress of accumulation ages ere man inhabited the country.