Art. L.—The Moas of the North Island of New Zealand.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 29, 1896, Unnumbered Page

Art. L.—The Moas of the North Island of New Zealand. By Captain F. W. Hutton, F.R.S., Curator of the Canterbury Museum. [Read before the Philosophical Institute of Canterbury, 4th November, 1896.] Plates XLVII. and XLVIII. The moas of the North Island are not so well known as those of the South Island. This is partly owing to the scarcity of their remains, and partly because they were the-first described. Sir R. Owen gave specific names to the bones as they were sent to him, and was obliged to fit them together by guesswork. Thus the leg-bones of his D. gracilis, D. dromæoides, and D. curtus belong, in each case, to two-distinct species, while D. didiformis and D. geranoides are made up from the bones of three different species. Both Mr. B. Lydekker and I have tried to correct these mistakes, and the differences in our nomenclature arose almost entirely from our having pursued different systems with Owen's composite species. I took the metatarsus of each as the type of the species, while Mr. Lydekker took the bone which had been, described first, no matter what it might be. Although my plan was the simpler of the two, that of Mr. Lydekker was more in conformance with the rule of priority, and, as his publication is so much better known than mine, I have now conformed to his rule, for all we want to get is a permanent and generally-recognised nomenclature. Mr. Lydekker also resuscitated Sir R. Owen's old name of Dinornis novæ-zealandiæ and applied it to all the large individuals from the North Island, remarking that “specimens belonging to more than one species were included under this name. As the first-mentioned specimen is a femur which is not absolutely characteristic, it seems best to take the second tibia [of D. ingens] as the type.”*“Catalogue of the Fossil Birds in the British Museum,” p. 224. But the femur first mentioned has a

length of 11in., and a circumference at the middle of 5 ½in., which measurements are certainly characteristic of the femur of D. struthioides; indeed, this very bone is referred by Mr. Lydekker to D. struthioides on p. 245, No. 18597, of his catalogue. If, therefore, the name is to be used at all it should be given to the smaller, not the larger, individuals of Dinornis in the North Island. But Owen employed the name only in his provisional notes, and before his paper was published he broke up his former species into three, and dropped the first-proposed name altogether. There is therefore no description of D. novæ-zealandiæ, and by the rules of zoological nomenclature the name should be abandoned. At any rate, I think that this is a case in which later authors may respect the wishes of the giver of a name without doing any harm to science. During the last four years I have been able to examine at leisure the collection of North Island bones in the Canterbury Museum, and I have also visited the Museums at Auckland, Wanganui, Wellington, and Dunedin as opportunities offered, And have to thank Mr. T. F. Cheeseman, Mr. S. H. Drew, Sir James Hector, Professor T. J. Parker, and Mr. Augustus Hamilton for the facilities they offered me for examining their collections of moa-bones. I have thus been enabled to correct several mistakes, and to clear up most of the obscure points; but we cannot expect to have the nomenclature quite fixed until the skulls and sterna of several species have been found and correctly determined. With regard to the number of species of Dinornis, the material does not exist to work them out in the same way as I did those of Kapua,* Trans. N.Z. Inst., vol. xxviii., p. 627. I have therefore taken the three South Island species as a guide in limiting the North Island species, and, consequently, I have reduced two names of my own making to the rank of synonyms. For my reasons for using the generic term Euryapteryx instead of Emeus† There is an Emea, Leidy (1848), among the Vermes, which is too near Emeus. I must refer the reader to my paper on the axial skeleton of the Dinornithidæ in Trans. N.Z. Inst., vol. xxvii, p. 158. No doubt it is difficult to decide what should be considered as a sufficient description and what is not sufficient, but I think it will be allowed that when a naturalist makes a new genus he should at least have seen the specimen he names, and that he should point out one, at least, generic character. If this is allowed Reichenbach's names will not stand. No doubt Sir Julius von Haast made mistakes, as all his successors have done; but he correctly conceived his three genera—Meionornis, Euryapteryx,

and Palapteryx—and gave M. casuarinus, E. gravis, and P. elephantopus as their types. Palapteryx cannot be allowed in Haast's sense, for it had already been used differently, but the other two should, in my opinion, be retained. Anomalopteryx must also be changed, as it is preoccupied in insects, and, consequently, I propose to substitute Anomalornis for it. Formerly it was thought that the genera Pachyornis and Euryapteryx were confined to the South Island, while several species of Dinornis, Meionornis, and Anomalornis were common to both islands. But it now appears that most of the genera occur in both islands, while nearly all the species of each-island are distinct. There is as yet no proof that Meionornis lived in the North Island, and no skull or sternum of Pachyornis or of Megalapteryx have been found in the North and none of Cela in the South Island. Also, the three species Megalapteryx tenuipes, Cela curta, and Pachyornis pygmæus, are only provisionally considered as belonging to both islands, the first and third being very imperfectly known, and the evidence for the occurence of C. curta in the South Island resting solely on a single metatarsus in the British Museum, said to come from near Oamaru. The general conclusion to be drawn from this is that the two islands of New Zealand were separated from each other after the development of most of the genera, but before the development of the known species,* Except, of course, Anomalornis antiquus. and that they have not since been united. It also follows that the deposits of moa bones at Te Aute, Glenmark, Kapua, Enfield, Waikouaiti, Hamilton's, &c., are of a later date than the separation of the islands. Dinornis giganteus. Dinornis giganteus, Owen, Trans. Zool. Soc., iii., p. 244 (1844). Dinornis novæ - zealandiæ, ♀, Lydekker, Cat. Fossil Birds in B.M., p. 224. Dinornis giganteus and excelsus, Hutton, Trans. N.Z. Inst., xxiv., pp. 110 and 112. Figures.—Metatarsus, Trans. Z.S., iii. (and Ext. Birds of N.Z.), pl. 27, fig. 1; tibia, l.c., iii., pl. 45 (Ext. Birds, pl. 37), fig. 1; femur, l.c., iii., pl. 44 (Ext. Birds, p. 36). This species appears to have been rare, as it is known at present by a few leg-bones only. There is no femur nor complete tibia in the Canterbury Museum. The type is a metatarsus 468mm. in length. The tibia described by Owen has a length of 887mm., and a distal width of 101mm. The femur, which was found with the tibia, has a length of 406mm. A metatarsus and a tibia were found together near

Auckland, the lengths of which were 451mm. And 800mm. respectively. A very large tibia, found at Te Aute, was measured by the Rev. W. Colenso, and said to have a length of 954mm., and a distal width of 96mm. Two metatarsi in the Canterbury Museum, one from Te Aute the other from Wanganui, have the following measurements: Length, 514mm.; prox. width, 104mm.; mid. width, 51mm.; distal width, 140mm. These bones are more slender than those of D. maximus, and the extremities of the metatarsus and tibia are not so expanded. A pre-maxilla, belonging to this species or the next, from near Wanganui, is more pointed and more curved down at the apex than the same bone in D. maximus or D. robustus. It resembles the pre-maxilla of D. torosus, but is much larger. The following are its dimensions : Length, 113mm.; length of body, 60mm.; width of body, 57mm. I have already, in my papers on the moas of Kapua and Enfield,* Trans. N.Z. Inst., vol. xxviii., pp. 627 and 645. explained why I cannot agree with Mr. Lydekker that this species is the female of the next. Dinornis ingens. Dinornis ingens, Owen, Trans. Zool. Soc., iii., p. 247 (1844). Dinornis novæ-zealandæ, ♂ Lydekker, Cat. Foss. Birds in B.M., p. 227. Dinornis firmus and ingens, Hutton, Trans. N.Z. Inst., xxiv., pp. 114 and 116. Figures.—Metatarsus, Trans. Z.S., iii., pl. 48 (Ext. Birds, pl. 40), fig. 1; tibia, l.c. (and Ext. Birds), pls. 25 and 26, figs. 1, 2; femur, l.c. (and Ext. Birds), pl. 21, figs. 1, 2; cranium, † The skull, figured as that of D. ingens, in Trans. Zool. Soc., vii., pl. 15, and Extinct Birds, pl. 82, is very incorrect, and should be disregarded. l.c., iv., pls. 2, 3, and 24 (Ext. Birds, 52 and 53), figs. 1, 3. The type is a tibia 736mm. in length, and with a distal width of 92mm. The Rev. W. Colenso has in his possession the bones of an individual leg the lengths of which are : Metatarsus, 412mm.; tibia, 736mm.; femur, 368mm. The commonest lengths are : Metatarsus, 355mm., and tibia, 660mm. The largest metatarsus in the Canterbury Museum has the following dimensions : Length, 419mm.; prox. width, 96mm.; mid. width, 48mm.; distal width, 132mm. The smallest metatarsus is : Length, 343mm.; prox. width, 77mm.; mid. width, 38mm.; distal width, 38mm. A tibia has the following dimensions : Length, 645mm.; prox. width, 142mm.; mid. width, 45mm.; distal width, 86mm. The largest femur is: Length, 366mm.; prox. width, 124mm.; mid. width, 55mm.;

distal width, 147mm. The smallest femur is: Lengthy 325mm.; prox. width, 117mm.; mid. width, 45mm.; distal width, 132mm. This species was far more common than the last; but, like-it, it is distinguished from its allies in the South Island by greater slenderness of limb. Mr. Lydekker gives also the greater obliquity of the extensor bridge of the tibia as a character by which the North Island species can be recognised from those of the South Island, but I cannot follow him in this. Dinornis struthioides. Dinornis novæ-zealandæ (part), Owen, Pro. Zool. Soc., 1843, p. 8 (no description). Dinornis struthioides, Owen, Trans. Zool. Soc., iii., p. 244 (1844). Dinornis dromioides, Owen,. l.c., iii., p. 319 (1846), tibia and metatarsus. Dinornis: struthioides, Owen, l.c., iv., p. 141 (1853). Dinornis gracilis, Owen, l.c., iv., p. 145 (1853), tibia and metatarsus. Dinornis struthioides and gracilis, Lydekker, Cat. Fossil Birds in Brit. Mus., pp. 242 and 248 (1891). Dinornis struthioides, gracilis, and Palapteryx dromioides, Hutton, Trans. N.Z. Inst., xxiv., pp. 119–121 (1892). Figures. — Metatarsus, Trans. Zool. Soc., iii. (and Extinct Birds), pl. 27, fig. 2; pl. 48 (E.B., pl. 40), fig. 2. (dromioides); iv., pl. 41 (E.B., pl. 54), figs. 3, 4 (gracilis) : tibia, l.c., iii., pl. 47 (E.B., pl. 39), fig. 1 (dromioides); l.c., iv.,. pl. 42 (E.B., pl. 55), figs. 1, 2 (gracilis) : femur, l.c., iii. (E.B.), pl. 21, fig. 3; iv., pl. 41 (E.B., pl. 54), fig. 2 : cranium, l.c., iii., pl. 38 (E.B., pl. 16), figs. 1–4. This species varies very much in size, and possibly two may be included. The type is a metatarsus 279mm. in length. There is an almost perfect skeleton in the collection of Sir Walter Buller, and an imperfect one in the British Museum. In the Wellington Museum there are the bones of a leg, found together, which have the following lengths: Metatarsus, 302mm.; tibia, 559mm., femur, 286mm. The following are the dimensions of the largest and smallest bones in the Canterbury Museum :— — Length. Prox. Width. Mid. Width. Distal Width. Metatarsus 294mm. 79mm. 35mm. 101mm. 246mm. 61mm. 25mm. 76mm. Tibia 483mm. 134mm. 40mm. 71mm. 464mm. 91mm. 30mm. 58mm. Femur 312mm. 106mm. 43mm. 124mm. 213mm. 72mm. 30mm. 70mm.(?)

There is a pelvis in the Wellington Museum, from Waipawa, which has the length of the ilium 470mm., length of pre-acetabular portion 223mm., and breadth at the antitrochanters 203mm. In Mr. A. Hamilton's collection, from Te Aute, there is a much smaller pelvis, in which these measurements are 305mm., 140mm., and 152mm. respectively. In the Canterbury Museum there is an imperfect cranium and pre-maxilla, from Wanganui, which is 60mm. wide at the temporal fossæ, while the distance between the temporal ridges is 44mm. The pre-maxilla belonging to this skull is 78mm. in length; the length of the body is 46mm., and its width 47mm. D. struthioides is smaller and more slender than D. torosus, and the extremities of the leg-bones are not so dilated. It has been found from Whangarei in the north to Wanganui in the south. Megalapteryx tenuipes. Megalapteryx tenuipes, Lydekker, Cat. Fossil Birds in Brit. Mus., p. 251 (1891). Megalapteryx tenuipes, Hutton, Trans. N.Z. Inst., vol. xxviii., p. 635 (1896). There are in the British Museum two femora from the North Island which Mr. Lydekker refers to this or an allied species. One has a length of 255mm., while the other is slightly smaller and relatively narrower, especially at the distal extremity. The latter comes from Waingongoro, near Wanganui. There are no bones of Megalapteryx from the North Island in the Canterbury Museum, and I have not seen any in any other museum. Anomalornis* Substituted for Anomalopteryx, as that name is preoccupied (1874) in the Neuroptera. gracilis. Plata XLVII., Fig. A. (?) Dinornis didiformis, Owen, Trans. Z.S., iii., p. 245 (1844), tibia. Dinornis gracilis, Owen, l.c., iv., p. 143 (1853), femur. Anomalopteryx, sp. α, Lydekker, Cat. Fossil Birds in Brit. Mus., p. 255. Figures.—(?) Tibia, Trans. Zool. Soc., iii. (and E.B.), pls. 25, 26, figs.3, 4 (didiformis); femur, l.c., iv., pl. 41 (E.B., pl. 54), fig. 1. The type of this species is a femur from Opito, near Mercury Bay, which has a length of 278mm., a proximal width of 95mm., a mid. width of 38mm., and a distal width of 101mm. There is in the Canterbury Museum a femur, from Akitio, which closely resembles the type, but is smaller. With it I provisionally associate a tibia and metatarsus (Plate XLVII.,

fig. A) of Anomalornis found in a cave at Waipu, north of Auckland. The dimensions of these three bones are as follows:— — Length. Prox. Width. Mid. Width. Distal Width. Metatarsus 203mm. 66mm. 36mm. 91mm. Tibia 419mm. 117mm. 36mm. 55mm. Femur 257mm. ?mm. 32mm. 80mm. No doubt the femur belonging to the leg from the Waipu Cave would be larger, as its mid. width would be about 36mm. The tibia is specifically identical with that figured by Owen, in 1844, as belonging to D. didiformis, but as that name goes with the metatarsus, which belongs to a smaller, species, we fall back on the femur as giving the name. Mr. Lydekker is of opinion that the tibia figured by Owen may belong to A. dromæoides, but I think it is too large. No doubt it belongs to one or other of these species. There is in the Canterbury Museum a pelvis from Pohui, Hawke's Bay, presented by Mr. H. Hill, F.G.S., which I am inclined to refer to this species. It is a typical Anomalornis in shape, and. has the following dimensions : Length of the ilium, 355mm.; length of pre-acetabular portion, 198mm; height of pre-acetabular portion of ilium, 83mm.; width at the antitrochanters, 160mm. The lower margin of the posterior portion of the ilium descends as a sharp ridge. The lower surfaces of the centra of the three pre-sacral vertebrae are hollowed longitudinally. Anomalornis didiformis. Dinornis didiformis, Owen, Trans. Zool. Soc., iii., p. 242 (1844), metatarsus. Anomalopteryx didiformis * Not Meionornis didiformis, Haast, which is M. didinus. (part), Lydekker, Cat. Foss. Birds in B.M., p. 276 (1891). Anomalopteryx didiformis (part), Button, Trans. N.Z. Inst., xxiv., p. 123 (1892). Figures.—Metatarsus, Trans. Zool. Soc., iii. (and E.B), pl. 27, figs. 3–6: (?)pelvis, l.c. (and E.B.), pl. 19, fig. 3; pl. 20, fig. 4: cranium, l.c., iii., pl. 39 (E.B., pl. 31), figs. 4–6 (dromæides). The type is a metatarsus from Poverty Bay, which (from the figure) has the following dimensions : Length : 173mm.; prox. width, 55mm.; mid. width, 31mm.; distal width, 76mm. In the Canterbury Museum there is a set of the three bones of the leg from a cave near Whangarei, the

metatarsus of which corresponds closely with Owen's figure and description. The following are their dimensions :— — Length. Prox. Width. Mid. Width. Distal Width. Metatarsus 167mm. 57mm. 30mm. 71mm. Tibia 335mm 94mm. 30mm. 49mm. Femur 230mm. 76mm. 30mm. 74mm. I have no doubt that these three bones belonged to the same individual. There is in the Museum collection a pair of femora from Akitio, east coast of Wellington, which have the following dimensions: Length, 228mm.; prox. width, 76mm.; mid. width, 33mm. There is in the Auckland Museum a cranium, and another in the Canterbury Museum, from the same cave near Whangarei, and it is probable that one or other of them belongs to the leg-bones in the Canterbury Museum. They closely resemble the skull of A. parvus, but are flatter, and the temporal ridges do not extend so far over the roof of the cranium. The best-preserved one has the optic foramina closer together than in A. parvus, and there is no horizontal shelf of bone below them, and no pre-sphenoidal fossa. Also the three foramina of the lacerate fossa are all united, and communicate with the optic foramen. The following are the only dimensions which can be taken : Width at squamosals, 65mm.; width at temporal fossæ, 42mm.; distance between temporal ridges, 34mm.; height of cranium, 41mm.; distance between optic foramina, 7mm. The other cranium is of about the same size, but more imperfect. So far as can be seen, it differs from it only in the absence of a mid-temporal ridge, which is probably due to age or sex. The pelvis described by Owen (as p. 5, Trans. Zool. Soc., iii., p. 256) probably belongs here, as also does another in Mr. A. Hamilton's collection from Te Aute, which measures 318mm. in length, with a width of 146mm. at the antitrochanters. These differ from the pelvis of A. parvus in having the lower surfaces of the centra of the sacral vertebræ longitudinally ridged, as in Meionornis Another pelvis, from the Hawke's Bay district, presented by Mr. H. Hill, F.G.S., has the following dimensions: Length of the ilium, about 300mm.; of pre-acetabular portion, 152mm.; height of pre-acetabular portion of ilium, 76mm.; width at the antitrochanters, 140mm. This pelvis differs from the typical pelvis in Anomalornis parvus in having the lower margin of the posterior portion of the ilium flattened out, as in Euryapteryx. But that it is an

Anomalornis is shown by the compressed centrum of the anterior vertebra, and by its anterior pneumatic, foramina not descending below the rib-facet. The lower surfaces of the centra of the three pre-sacral vertebræ are hollowed longitudinally, as in the last species. Anomalornis oweni. Dinornis oweni, Haast, Trans. Zool. Soc., xii., p. 171 (1886). Anomalopteryx oweni, Lydekker, Cat. Fossil Birds in Brit. Mus., p. 280 (1891). Cela curta (part),. Hutton, Trans. N.Z. Inst, xxiv., p. 127 (1892). Figures.—Limb-bones, skull, and two vertebrae, Trans. Zool. Soc., xii., pl. 31; pelvis,. Trans. Zool. Soc., xii., pl. 32. The type skeleton is in the Auckland Museum. The following are the measurements of the largest and smallest of the leg-bones in the Canterbury Museum—the smallest are rather larger than the types :— — Length. Prox. Width. Mid. Width. Distal Width. Metatarsus 128mm. 44mm. 24mm. 58mm. 117mm. 40mm. 21mm. 49mm. Tibia 285mm. 75mm. 23mm. 39mm 259mm. 70mm. 20mm. 36mm. Femur 157mm. 50mm. 23mm. 52mm. 150mm. 50mm. 21mm. 49mm. The femur has not the characters of the, typical Anomalornis, but is short and straight, with a rather deep popliteal depression. The tibia has the distal extremity much expanded inwards, as in A. parvus. The metatarsus has deep lateral pits on the middle trochlea, and generally on the anterior surface at its base. The skull is very like that of A. didiformis, but is rather smaller. The lacerate fossa is distinct from the optic foramen, and there appears to have been a pre-sphenoidal shelf below the optic foramen; but this part of the skull is much damaged. The width at the squamosals is 62mm.; width at the temporal fossæ, 42mm.; and the distance between the temporal ridges is 29mm. The height of the cranium is 33mm. Distance between optic foramina, 6mm.; length of body of pre-maxilla, 20mm.; width of the same, 18mm.; greatest height of the mandible, 14mm.; length of mandibular symphysis, 6mm.; width of the same, 8mm. The width of the pelvis at the antitrochanters is (from the figure) 102mm. The whole of the posterior portion-is absent,

but, judging from the figure, the centra of the ribless sacral vertebrae have their lower surfaces rounded, and with a rudimentary longitudinal keel. The sternum is not known. Cela* As Cela of Möhring is a synonym of Casuarius, the name may perhaps be allowed to stand here. carta. Plate XLVII., Fig. B. Dinornis curtus, Owen, Trans. Zool. Soc., iii., p. 325 (1846). Anomalopteryx curta, Lydekker, Cat. Fossil Birds in Brit. Mus., p. 281. Cela curta, Hutton, Trans. N.Z. Inst., xxiv., p. 127. Figures.—Metatarsus, Trans. Zool. Soc., iii., pl. 48 (E.B., pl. 40), fig. 6; vii., pl. 44 (E.B., pl. 87), figs. 7–10: tibia, l.c., iii., pl. 47 (E.B., pl. 39), figs. 3–5: femur, l.c., v., pl. 55 (E.B., pl. 68), figs. 5, 6 (geranoides): cranium, l.c., iv., pl. 24 (E.B., pl. 53), fig. 5; xiii., pls. 61, 62, Mesopteryx, sp. α. This was a common species all over the North Island. The metatarsus figured by Owen in “Extinct Birds of New Zealand,” pl. 87, f. 7, is from a cave fourteen miles distant front Oamaru. All other bones are from the North Island. The type is a tibia from Waingongoro, near Wanganui, which has a length of about 292mm. The following are the dimensions of the largest and smallest leg-bones in the Canterbury Museum:— — Length. Prox. Width. Mid. Width. Distal Width. Metatarsus 138mm. 46mm. 26 mm. 58mm. 127mm. 41mm. 24mm. 54mm. Tibia 310mm. 81mm. 25mm. 41mm. 294mm. 76mm. 24mm. 41mm. Femur 170mm. 64mm. 26mm. ? 162mm. 56mm. 25mm. 55mm. The pelvis is like that of Meionornis casuarinus, but much smaller, and the centra of the sacral vertebræ are compressed. The length of the ilium is 203mm.-228mm.; that of the pre-acetabular portion is 94mm.-107mm. The height of the ilium before the acetabulum is 63mm.—70mm.; and the width at the antitrochanters is 120mm.-127mm. The sternum resembles that of M. casuarinus, but there are no pneumatic depressions in the antero-lateral corners. The width below the costal border is about 68mm. I refer to this species the skull described by Professor

T. J. Parker, F.R.S., under the name of Mesopteryx, sp. α,* Trans. Zool. Soc. of London, vol. xiii., p. 378. from Te Aute (Plate XLVII., fig. B.). This skull resembles that of Meionornis casuarinus in—(1) Its arched cranial roof; (2) the shape of the temporal fossæ and ridges; (3) the uniformly-rounded orbits; (4) the small mammillar tuberosities; and (5) the slender-pointed mandible. But it differs in (6) the supra-, foraminal ridge not projecting beyond the occipital condyle;. (7) in the par-occipital processes being-short and rounded; (8). in the margin of the tympanic cavity being evenly curved; (9) in the zygomatic processes sloping more forward and outward; (10) in the broader posterior temporal fossæ; (11) in the presence of a sulcus immediately in front of the ridge separating: the temporal fossa from the optic cavity; (12) in the pre-sphenoidal fossa extending-considerably in advance of the optic foramen; and (13) in the broad articular cup and small, posterior angular process of the mandible. In all these points the skull approaches those of Anomalornis or Pachyornis, especially the latter. There is another, specimen of what appears to be the same or a closely-related species in the Auckland Museum, from a cave near Whangarei. It is more imperfect than the one from Te Aute, and has no mandible. The two skulls are alike in size, and in the shape of the par-occipital processes, the tympanic cavity, the zygomatic processes, and the posterior temporal fossæ; and also in the shape of the temporal fossæ and presence of a sulcus in front of the ridge separating the temporal fossa from the optic cavity. The pre-sphenoidal fossa cannot be compared. However, the Whangarei specimen differs from that from Te Auto in many small particulars. The occipital condyle does not project so much, and is overhung by the supra-occipital ridge, as in Meionornis. The cranial roof is more-convex, both longitudinally and transversely; the mammillar tuberosities are better developed; the area between the temporal and lambdoidal ridges is rather narrower; the post-orbital processes are broader; the optic foramina are larger and rounder; and the three foramina of the lacerate fossa are larger. This cranium has the basi-pterygoid processes preserved; they are very slender, and project downward more than in any other species. No doubt the two skulls belong to the same genus. The dimensions of the Te Aute skull are given by Professor, Parker. The following are those of the Whangarei skull : Length of basis cranii, 26mm.; length of roof of cranium, 65mm.; width at squamosals, 51mm.; at temporal fossæ, 41mm.; at postorbital processes, 62mm.; distance between

the temporal ridges, 36mm.; height of the skull, 38mm.; distance between optic foramina, 7mm. It is, I have no doubt, specifically identical with that figured by Owen in “Extinct Birds of New Zealand,” pl. 53, fig. 5. The Whangarei skull is nearer to Meionornis than is that from Te Aute; but the curved margin of the tympanic cavity and the sulcus close to the posterior margin of the optic cavity are found in both, but never occur in Meionornis. If to these we add the large pre-sphenoidal fossa and the shape of the articular cup of the mandible, seen in the Te Aute specimen, we have sufficient reasons for placing these skulls in a different genus from Meionornis, or, at any rate, in a distinct sub-genus of Meionornis. The mandible is quite unique among the moas, combining the slender rami of Meionornis with the broad articular cup of Anomalornis and Pachyornis. The photograph (Plate XLVII., fig. B) shows the articular cup well, but the point of the beak looks more rounded than it really is, owing to distortion by pressure when wet. The differences between the vertebræ of Cela and Meionornis have been pointed out by me in the Trans. N.Z. Inst., vol. xxvii., p. 166. There is in the Museum collection a femur from Tolago Bay, which shows the pneumatic foramen very plainly. Euryapteryx exilis, sp. nov. Plate XLVIII., Fig. C. Figures.—Femur, Trans. Zool. Soc., iii. (and E.B.), pl. 24, figs. 1–3 (didiformis) : pelvis, l.c. iii. (and E.B.), pl. 19, fig. 2 (didiformis); pl. 20, figs. 2, 3 (didiformis). The type of this species is a nearly-complete skeleton in the Wanganuí Museum. It appears to have been common in the southern parts of the North Island, but; its bones have been confounded partly with those of A. didiformis and partly with those of C. curta. Indeed, the slimness of the legs would have prevented them being put into the genus Euryapteryx if they had not been accompanied by the skull. The following are the dimensions of the leg-bones of the type specimen:— — Length. Prox. Width. Mid. Width. Distal Width. Metatarsus 140mm. 55mm. 30mm. 63mm. Tibia 338mm. 101mm. 28mm. 44mm. Femur 210mm. 72mm. 30mm. 74mm. But the species evidently attained a larger size, for there are in the Canterbury Museum bones belonging to this species of the following sizes :—

— Length. Prox. Width. Mid. Width. Distal Width. Metatarsus 155mm. 54mm. 32mm. 68mm. Tibia 368mm. 100mm. 32mm. 53mm. Femur 213mm. 76mm. 33mm. 81mm. This metatarsus and femur are from Wanganui; the tibia from Akitio, on the east coast of Wellington Province. The bones resemble those of E. crassa, but the tibia is more convex on the anterior surface. The skull (Plate XLVIII., fig. C) resembles that of E. crassa in miniature, but with some differences. The occipital condyle does not project beyond the par-occipital processes. The plane of the occipital foramen is inclined backwards. The occipital crest, the lambdoidal and posterior lambdoidal ridges, the supra-foraminal ridges, and the par-occipital processes are like those of E. crassa. The cranial roof is much vaulted, evenly arched from side to side, and with only a slight frontal rising, such as is conspicuous in other species of Euryapteryx. Also, the flat area between the temporal and lambdoidal ridges is narrower than in other species. The margin of the orbit is evenly curved. The basi-occipital is shorter than in E. crassa, and the mammillar tuberosities are but slightly developed; the basi-pterygoid processes are broad. The rostrum, maxillo-jugal arch, palatines, and pterygoids are missing. The tympanic cavity is as in E. crassa, and the zygomatic process is bifid. The posterior temporal fossa is narrow, as it is in all species of Euryapteryx: The temporal fossæ are relatively broader than in E. crassa, and are shaped as in Meionornis, but they resemble those of Emeus, sp. β, in Professor Parker's memoir “On the Cranial Osteology of the Dinornithidæ.” * Trans. Zcol. Soc. of London, vol. xiii., p. 373. The optic foramina are large and wide apart for the genus. The supra-orbital ledge makes almost a right angle with the inner portion of the orbital roof, instead of gradually sloping into it, as is usual. The quadrate is like that of E. crassa, but the orbital process is more slender. The pre-maxilla is short and broad; the apex has been abraded, but, judging from the shape of the mandible, it was no doubt rounded, The mandible is much deflexed, broadly U-shaped at the apex—more so than in E. crassa—and the symphysis is broad and low. The articulating-cup of the ramus is like that of E. crassa, but the external posterior angular process is not so prominent. The following are the measurements in the order of Professor Parker's paper, already

quoted: Length of skull, 92mm.(?); length of basis cranii, 27mm.; length of roof of cranium, 64mm.; width at paroccipital processes, 46mm.; width at squamosals, 54mm.; width at temporal fossæ, 40mm.; width at post-orbitals, 62mm.; distance between temporal ridges, 35mm.-; height of cranium, 38mm.; width of tympanic cavity, 17mm.; width of temporal fossa, 19mm.; width of orbit, 24mm.; distance between optic foramina, 10mm.; greatest length of pre-maxilla, 57mm. (?); length of body, 16mm.(?); width of body, 27mm.; length of mandibular ramus, 85mm.; greatest height of mandible, 14mm.; least height of mandible, 8mm.; length of mandibular symphysis, 10mm. (?); width of mandibular symphysis, 26mm. The skull undoubtedly belongs to Euryapteryx; but in the shape of the temporal fossæ, in the small mammillar tuberosities, and in the small posterior angular processes of the mandible it approaches the skull of Cela. On the other hand, the distance between the optic foramina and the narrowness of the flat area between the temporal and lambdoidal ridges are characters which connect it with Anomalornis. The cervical vertebræ have neural ridges as in Meionornis; the atlas and axis are missing. The thoracic vertebræ resemble those of other species of Euryapteryx, but there are only two pairs of sternal ribs, which belong to vertebras Nos. 24 and 25. No. 28 is anchylosed to the pelvis, showing that the bird is an old one. There are seventy tracheal rings preserved with the skeleton, and about twenty more were found, but were too brittle to be moved. The upper rings are of the slender elliptical pattern, while the lower ones are of the thick, rough, tube-like pattern, one pattern passing, into the other about half-way down. The sternum agrees with that of other species of Euryapteryx, but there are slight saucer-shaped coracoidal depressions. The costal processes are large and horizontal. The following are its dimensions: Length of body, 83mm.; total length, 152mm.; width of body, 100mm.; width across costal processes, 152mm. The pelvis resembles that of other species of Euryapteryx The centre of the ilium is slightly in front of the centre of the acetabulum. The length of the pre-acetabular portion of the ilium is 127mm. (?), and its depth is 76mm. The width at the antitrochanters is 165mm. Pachyornis rothschildi. Pachyornis rothschildi, Lydekker, Pro. Zool. Soc., 1891, p. 279. Figures.—Leg-bones, P.Z.S., 1891, pl. 38. Described from a set of leg-bones supposed to belong to the same individual; locality not known, but thought by Mr.

Lydekker to be from the North Island. The following dimensions are given :— — Length (about) Mid. Width. Distal Width. Metatarsus 216mm. 53mm. 127mm. Tibia 559mm. 75mm Femur 269mm. 124mm. No other part of the skeleton is known. The shafts of the tibiæ are very much curved, but this may be an individual deformity. There are in the Christchurch Museum some leg-bones from Te Aute, which I refer to this species. They have the following dimensions :— — Length. Prox. Width. Mid. Width. Distal Width. Metatarsus 176mm. 76mm 46mm 99mm. Tibia 483mm. 140mm. 41mm. 71mm. Femur 259mm. 100mm. 46mm. 112mm. Although these bones are smaller than the types and the metatarsal trochleæ are not so expanded, still they have the same relatively long tibia which distinguishes this species from P. inhabilis, but they are not curved like the type. Other bones, rather larger than these, were also found at Te Aute, and are now in the collection of Mr. A. Hamilton. Pachyornis pygmæus. Plate XLVIII., Fig. D. Dinornis geranoides, Owen, Trans. Zool. Soc., v., p. 395 (1866). Anomalopteryx(?) geranoides, Lydekker, Cat. Fossil Birds in Brit. Mus., p. 288 (1891). Cela geranoides, Hutton, Trans. N.Z. Inst., xxiv., p. 126 (1892). Euryapteryx pygmæus, Hutton, Trans. N.Z. Inst., xxiv., p. 139 (1892). Figures.—Metatarsus, Trans. Zool. Soc., v., pl. 67 (E. B., pl. 70), figs. 5, 6; femur, Trans. N.Z. Inst., xxvii., pl. 9. A very rare species, the type of which is a metatarsus found near Nelson, There is also in the British Museum a metatarsus from a cave near Oamaru; all other known specimens are from the North Island. These are more slender than the type, and may perhaps belong to a distinct species. A metatarsus from Wanganui, in the Canterbury Museum, has the following dimensions: Length, 152mm.; proximal

width, 59mm.; mid. width, 32mm.; and distal width, 76mm. I have not seen a tibia which I can refer to this species. Mr. Lydekker gives the average length at 342mm., and distal width 55mm. There are three femora in our collection, all from Te Aute. The largest is: Length, 221mm.; prox. width, 82mm.; mid. width, 38mm.; distal width, 84mm. The smallest is: Length, 201mm.; prox. width, 75mm.; mid. width, 34mm.; distal width, 72mm. There is also a small pelvis from Te Aute which has the characteristics of Pachyornis, and which, therefore, I refer to this species (Plate XLVIII., fig. D). The middle of the ilium is situated at the posterior margin of the acetabulum; the pre-acetabular ridge, formed by the ilia, is straight axially; the posterior portion of the ilium is very broad and flat; the ventral surfaces of the centra of the sacral vertebræ are strongly curved transversely, but are not so narrow as in P. elephantopus. The following are its dimensions: Length of ilium, 305mm.; length of the pre-acetabular portion of the ilium, 127mm., and its depth, 76mm.; width at the antitrochanters, 159mm. Incertæ sedis. Dinornis dromæoides, Owen, Trans. Zool. Soc., iii., p. 250, pl. 22, figs. 1, 2 (1844). Anomalopteryx dromæoides, Lydekker, l.c., p. 266. Professor C. Stewart has been kind enough to send me a cast of the type femur of this species, which is preserved in the museum of the Royal College of Surgeons. This specimen, which is slightly abraded at the proximal end, has the following dimensions: Length, 235mm.; proximal width, 84mm.; mid. width, 35mm.; distal width, 84mm. There is in the Canterbury Museum collection a femur from Te Aute which has the same size and shape, but in which the popliteal depression is deeper. These femora are not those of a true Anomalornis, but might belong to a species of Cela or Meionornis. It is just possible that the tibia and metatarsus, here included provisionally as A. gracilis, may belong to D. dromæoides, but they have the characters of Anomalornis and not those of Meionornis. There is in the Auckland Museum a very imperfect pelvis, which may also belong to D. dromæoides; its width at the antitrochanters is 184mm. Palapteryx geranoides, Owen, Trans. Zool. Soc., iii., p. 361 (1848), pl. 54 (E.B., pl. 45), figs. 1–7. Anomalopteryx(?) geranoides, Lydekker, l.c., p. 290. This species is founded oh a skull, and is therefore not comparable with the others, which are founded on leg-bones. Mr. Lydekker proposes to take either the metatarsus or the tibia described by Owen under the same name—Pachyornis

pygmæus of this paper—as the type, but this is against the rule followed in other cases, and cannot, I think, be allowed. He also places the cranium only in his A. geranoides, stating that the pre-maxilla and mandible belong to M. casuarinus. But Owen's figures do not bear this out, for both pre-maxilla and mandible are represented as belonging to Anomalornis, and are not at all like the same bones in M. casuarinus; and as no other bones of M. casuarinus have been found in the North Island, I hesitate to accept Mr. Lydekker's determination. The cranium and the mandible resemble a good deal those which I attribute to A. fortis; but the North Island cranium is flatter, and has larger temporal fossæ. I therefore think that Owen was right in placing these bones together, but that in his restoration he has made the mandible too long. Probably this skull belongs to D. dromæides or to A. gracilis, but we must await some lucky-discovery before this can be proved. Explanation of Plates XLVII. And XLVIII. Plate XLVII. Fig. A. Metatarsus of Anomalornis gracilis, from a cave at Waipu. Fig. B. Cranium and mandible of Cela curta, from the swamp at Te Aute. The mandible is distorted by pressure when wet; the apex is really as pointed as in M. didinus. Plate XLVIII. Fig. C. Skull of Euryapteryx exilis, from the sandhills near Wanganui. Fig. D. Pelvis of Pachyornis pygmæus, from the swamp at Te Aute.