BULLETIN OF THE AUCKLAND INSTITUTE AND MUSEUM
No. 1.
THE MOA
A STUDY OF THE DINORNITHIFORMES
BY
GILBERT ARCHEY
PUBLISHED BY ORDER OF THE COUNCIL ISSUED MAY 29, 1941
THE UNITY PRESS LIMITED
1941
*mmm -> • tc wini.uMri.st a«i «» « ymurn - t&MWMtmeatmtm
AUCKLAND INSTITUTE AND MUSEUM LIBRARY
N*£Vf 7E ALAMO COLLECTION
(RESERVE)
class No. Q£$73L-j£
#•
BULLETIN OF THE AUCKLAND INSTITUTE AND MUSEUM " i
No. 1. w
. • j. - • - v-— > *■» v . V’;A. * ^ ■ '3^
THE MOA
A STUDY OF THE DINORNITHIFORMES
BY
GILBERT ARCHEY
PUBLISHED BY ORDER OF THE COUNCIL ISSUED MAY 29, 1941
THE UNITY PRESS LIMITED 19 4 1
•44379
CONTENTS
Page
INTRODUCTION .
TYPES AND NOMENCLATURE .
SIGNIFICANT CHARACTERS AND CLASSIFICATION
DESCRIPTION OF GENERA AND SPECIES .
EGGS
SKIN AND FEATHERS .
TRACHEAL RINGS .
CLASSIFICATION .
DIAGNOSES .
DEVELOPMENT AND DISTRIBUTION OF SPECIES
PHYLOGENY OF THE DINORNITHIFORMES .
ADVENT OF THE MO A:
DISTRIBUTION OF THE PALAEOGNATHAE
DEVELOPMENT AND EXTINCTION .
SUMMARY .
PRIORITY OF NAMES PROPOSED FOR DINORNITHIFORMES
BIBLIOGRAPHY .
INDEX .
PLATES .
TABLES
5
6 9
13
73
75
76 76 76 79 81
86
89
98
100
101
113
121
137
<
INTRODUCTION.
It is just over a hundred years since Professor Richard Owen exhibited to the Zoological Society of London part of a large bone from New Zealand brought to him by Dr. John Rule, and, as the result of examining it, declared . . . so far as my skill in interpreting an osseous fragment may be credited, I am willing to risk the reputation for it on the statement that there has existed, if there does not now exist, in New Zealand, a struthious bird nearly, if not quite, equal in size to the Ostrich.”
Owen’s “osseous fragment” has been reinforced by vast quantities of skeletal remains from many sources, and his brief paper of 1840 has been followed by a voluminous litera¬ ture on the extinct giant flightless birds of this country, a literature of over four hundred contributions from one hundred and eighty authors! Owen, who gave us the foundation and framework of our understanding of the moa ; Haast, Hector and Hutton ; Lydekker, Parker and Forbes; these are the names that stand in our mind as the leaders in the keen enquiry that followed Owen’s first bold deduction. But over a century of discovery and of research and speculation has not been able to establish finality, either in fact or inference, in this fascinating field of study, for each new find or investigation, in solving some problem, has as frequently revealed new uncertainties. Nor can the present study, instigated through the discovery of some new North Island collecting fields yielding the remains of individual birds, pretend to resolve outstanding issues. It is primarily sys¬ tematic; it has endeavoured to review the material and literature of past investigations, and it includes a detailed study of the new collections and of as much of the older material, including types, as could be located in New Zealand and Great Britain. It is hoped that the changes that have been made in the names of species and their generic grouping may take us a little further towards the desired final classification, but there still remain difficulties in discerning the natural relationships underlying the variability in form and size exhibited among moa bones and skeletons; there are unsolved problems, too, in nomenclature and the fixation of types, and, apart from systematic details of this nature, there still remains room for differences in interpretation of the facts already established.
In 1930 Mr. W. H. Gregory, engineer-in-charge of the Waikaremoana hydro-electric station, discovered moa remains in the caves which occur throughout the outlet baniei- wall of Lake Waikaremoana; following this discovery a collecting excursion was arranged by Mr. F. Crossley Mappin, with Sir Carrick Robertson, Mr. A. T. Pycroft and the writer, to explore the area. Trips were made in 1930, 1931 and 1935, and in the intervening period the high limestone country of the Mangaotaki Valley, the Te Anga Valley and the hill country of Mr. Phillips’ property at Marakopa were similarly investigated, all with grati¬ fying results.
In 1932, skeletons were excavated from the sand-dunes at Doubtless Bay and pre¬ sented to the Museum by Mr. L. J. Matthews and his son, Mr. Geoffrey Matthews. It was in this area that Mr. Matthews senior, thirty years before, had discovered the two moa eggs described by the writer in 1931. Remains from these sandhills have also been presented by Mr. E. T. Frost and Major G. A. Buddie; others from Doubtless Bay and the North Cape district were secured this year by Mr. Pycroft, Mr. H. R. Jenkins, Mr. A. B. Deeming and the writer. A further series was obtained in 1933 from caves on the Mt. Arthur table-land, Nelson, by Messrs. F. Gibbs, H. Kidson, S. W. Street and Wm. C. Davies, who kindly organised the trip for my benefit and participation. During last summer Mr. J. Hodgen discovered remains in a swamp on his property at Pyramid Valley, near Waikare, North Canterbury. The site is being carefully excavated by a Canterbury
5
Museum party and, contrary to past experience in swamps, has yielded individual skele¬ tons T have had the pleasure of joining one Canterbury Museum excursion and have
had the advantage, not only of readily given facilities for
. . . •ii tv.-, t> a Director of the JVluseum, conceining
also of helpful discussions with Dr. R. A. raiia, uirecioi
them.
Altogether 50 fairly complete skeletons and over 100 partial sets of the bones of indi¬ vidual birds have been secured from these various sources. A study of this material and of the individual skeletons already in other museums in New Zealand and in the Bi ltish
Museum, has made it possible to present a detailed account of certain species and by
ascertaining the range of variation that occurs, to judge the value of certain characters as a basis for classification.
It will be seen that I am indebted to many friends who have contributed to the assembling of this important material, and I hasten to make my grateful acknowledgment to them all for their help. I am particularly beholden to Mr. Mappin and to Sir Carrick Robertson and Mr. Pycroft, whose enthusiasm, energy and good company through many seasons of assiduous search and collecting among rocks and caves and through fern, supple-jack and brambles, provided a spur, shall I say, and encouragement in pursuing this study. No less than 42 skeletons or individual sets of bones were secured through their explorations. I have made many demands upon my colleagues the directors and curators of all the New Zealand museums for facilities to examine specimens, for the loan of them, and for taking measurements which I had missed during my visits. Mr. J. Grant, Hon. Director of the Wanganui Museum, generously gave me the detailed measure¬ ments and calculated proportions of the large number of bones which he and Mr. Shepherd have recovered from the Makirikiri deposit, and Mr. A. Robertson of Wanganui invited me to inspect his considerable collection from the same site and kindly arranged all the specimens for examination and measurement.
I have to acknowledge helpful suggestions and assistance from Dr. R. S. Allan, Hon. Palaeontologist of the Canterbury Museum, and Mr. A. W. B. Powell, of our own museum, in connection with references, problems of nomenclature and fixing of types ; Mr. Powell has also drawn plates 6, 8 and 12 to 15 and all the text-figures, as well as giving me help¬ ful advice with regard to the other drawings. Sir William Benham has most obligingly written long detailed replies to my questions about Otago Museum specimens; Miss Dorothea M. A. Bate and her colleagues in the Geological Department of the British Museum have most courteously and readily made available the many specimens from their collections that I desired to examine, and I received a like cordial welcome and help from the late Lord Rothschild and Dr. Jordan at Tring. Nor must I forget to record my indebtedness to many students, including past teachers and friends, whose investigations have been freely drawn upon, particularly in the interpretative portion of this study. Finally, I have to acknowledge the generosity of the Carnegie Corporation of New York, both for the visitor’s travelling grant which enabled me to examine the material in Great Britain and for a further grant on my return for visiting the museums in our own country.
TYPES AND NOMENCLATURE.
All who have recently studied the Moa have remarked upon the uncertainty and con¬ fusion that exist in the classification of the group. This is partly due to the puzzling manner in which the species grade into one another ; but it is equally due to the fact that most of the species have been founded either on single bones, or more frequently and unhappily, on mixed bones, i.e. those not of individual birds, sometimes even on bones from more than one source.
6
Sir Richard Owen's historic studies were, of necessity, based upon material of this kind whereby he was led in some instances to associate in one species bones which have subsequently been shown to belong to more than one. He himself made certain collec¬ tions, and others have yet to be made. Further confusion, it must be said, resulted from his lack of precision in proposing species: they are sometimes indicated in the most casual manner without the nomination of types, with particularly unfortunate results when mixed bones were being examined.
Rules of nomenclature had not been formulated in Owen’s time, and their present strict application, by which alone can finality be reached, may in some cases seem to nullify Owen’s intention. This is unfortunate; but we have to accept what he actually did rather than what he intended to do. For instance, he intended that his extensive illustrated papers in the Transactions of the Zoological Society should be the means of establishing his species ; but frequently the brief advance notices in the Proceedings of the Zoological Society contain adequate descriptions and must therefore be accepted as prior designations. It may thus be necessary to correct, even to criticise certain phases of Owen’s systematic work; on the other hand one cannot refrain from paying a tribute to his masterly anatomical analyses and his skill in deducing form and function from
skeletal material.
Later workers of the last century unravelled some of the confusion, but in many cases they added to it, partly because they, also, had incomplete and mixed material for study, and partly because the workers were in different countries. Lydekker, in England, had some of Owen’s types and the large collections, for the most part of unassociated bones from several localities, in the British Museum. Haast and Hutton had extensive series, chiefly of leg-bones from the swamp deposits of Glenmark, Enfield, Kapua and Hamilton’s (the latter in Central Otago), but no early types. I can find no record that Lydekker had actually compared the British Museum material with Owen’s types. The latter were supposed to be in the Museum of the Royal College of Surgeons, but they could not be identified there when I examined this collection in 1937, * and I am constrained to make the suggestion that when Sir Richard Owen transferred to the British Museum he may possibly have taken specimens with him to continue his studies, and that the types may therefore still be in the British Museum, unrecognized, and possibly now unidentifiable. Fortunately, there are casts of some of them m the British
Museum.
Hutton, having for examination many hundreds of mixed bones from swamps and finding that, although they varied considerably in size and proportion, there was. no definite break which might distinguish the groups, established certain species on points of concentration” in the recorded dimensions, i.e. on dimensions that were strongly repre¬ sented numerically in his series. Unfortunately he also failed to designate types, and in order to embody these idealistic species I have, where possible, nominated lectotypes from among Hutton’s specimens. Nomination of lectotypes is made in the synonymy of
the different species concerned.
Parker (1895b) propounded a classification based on skull characters. It is an admir¬ able comparative study of skull anatomy, but as a systematic study suffers from two defects. The skulls were for the most part not from individual skeletons, and were assigned only interpretatively, and sometimes wrongly, to already-described species founded on leg-bones, which, having priority, cannot be ignored. Rothschild (1907) did not introduce any new criteria for the determination of species ; he did, however, give names to certain species designated by Parker as species a, sp. (3, sp. y.
*Sharpe (1891, 424 footnote) reported them missing nearly fifty years ago.
7
Oliver’s classification (1930), which was followed by Lambrecht (1933), is based on the proportionate widths of the leg-bones, and is in line with the evolutionary tiend towards increased bulk perceived by Owen (1844b, 241-3, 250, 1873a, 362), who leco^ nized also the classificatory significance of the accompanying decrease in metataisal length.
It now appears that these principles can be followed only in part j moreover the skull characters as defined by Oliver require adjustment in the light of information deiived from newly-found individual skeletons. These, it is believed, disclose the types oi skull characteristic of the genera. They also reveal that two subfamilies with widely differ¬ ing skulls, sterna and phalanges have independently embarked upon identical courses of development of shorter and heavier leg-bones. This hitherto unsuspected parallelism in evolution has, perhaps more than any other factor, obscured the relationships of the genera and species of moa. Re-arrangements will therefore be necessary both on the basis of the characters used and, as is so often tiresomely necessary in a revision, through the fixing of types and priority, according to the Rules of Nomenclature.
Because of the numerous cases in which species have been founded on more than one bone, not being those of an individual, close attention has had to be given to the fixing of types, and in this connection the correct procedure to be adopted is important. Both the present writer (1927, 156) and Oliver (1930, 35) had assumed that the procedure was to fix the first-described bone as the type specimen and to ignore the others as regards that species, but Dr. R. S. Allan has kindly drawn my attention to the undesirability and doubtful validity of this course. My special thanks are due to Dr. Allan: not only has he given me the memorandum he intended to publish on this matter, but he has also handed to me the bibliographical notes and references to the proposals of genera and species that he had brought together. These have enabled me to check my own compila¬ tion and have also drawn my attention to references that I should otherwise have over¬ looked. I have availed myself of his advice on several points in synonymy and have had his co-operation in preparing the undermentioned statement of a problem on type fixation forwarded for an opinion to the International Commission on Zoological Nomen¬ clature.
Returning to type-selection, Di. Allan s viewT is that, where a species is based upon more than one bone not being those of an individual, and where the author has failed to designate or indicate a type, all the bones described in the original publication are of equal \alue and must be tieated as co-types or syn-types until a subsequent reviser shall have nominated one of them as a lectotype and thereby given it the status of a holotype. Phis selection, once made, is not subject to change. To proceed otherwise, Dr Allan observes, is to do scant justice to earlier workers who have adopted this method, and to reverse decisions made by them after due consideration. With this one can but agree (though I may later have occasion to express a wish that in some cases the selector had
chosen other than he did), and, in the absence of rules for selecting the type of a species,
the procedure outlined above, based as it is on Rule 31 governing the selection of geno¬ types, is acceptable and will be followed here.
Arising out of this procedure a new question arose in the case of Dinornis novae- zedandiae as to whether certain proposals by Owen, made after his original proposal of that species but not in themselves a definite selection of a type, had, or had not, had the effect of determining the type specimen, and, thereby, of invalidating Lydekker’s subse¬ quent nomination of a lectotype. The case has been presented to the International Com¬ mission on Zoological Nomenclature whose opinion is, however, not yet available.
8
SIGNIFICANT CHARACTERS AND CLASSIFICATION.
Before entering upon detailed descriptions of the genera and species we should dis¬ cuss two factors that have created uncertainty in classification, (a) the form of the beak, and (b) size and proportions of leg-bones.
(a) Form of the beak.
No uncertainty as to identification arises in the case of the wide flat skulls, with broad terminally depressed beaks of the species of Dinornis (PI. 5, fig. 1), nor is there any difficulty in distinguishing either the skulls of Anomalopteryx with their large tem¬ poral fossae and characteristic narrow beak, or the much smaller, round-contoured skulls of Megalapteryx. Identification in these cases has been facilitated by the discovery of sufficient skeletons to establish the correct association of skulls with leg-bones. The un¬ certainty has arisen in the case of the heavier moas, through the occurrence in siwamp and dune deposits of broad-billed and sharp-billed skulls mingled indiscriminately with several kinds of leg-bones, whose proper relations to the skulls, and to one another, the lack of individual skeletons has made difficult to determine.
In 1870 (p. 123, pi. 10) Owen described certain unassociated crania and sharp-pointed beaks as being of D. elephantopus. From further unassociated material he referred to D. crassus , a skull having both a broad bill, and, if one can judge from the illustration (pi. 11), a collapsed or constricted antrum wall. He assigned to D. rheides a sharp-billed skull (pi. 12), (really of Em. crassus), and he proposed the name Dinornis gravis (p. 141, pi. 14) for a skull lacking a premaxilla but with a broad mandible, which, from his statement that he had here “parts of the skeleton of the same indi¬ vidual bird,” we can assume belonged to it. No question of generic grouping was involved in these descriptions.
When Haast (1874) proposed a classification of the group he regarded his Meionornis (type casuarinus) as having a narrow pointed beak, Palapteryx (elephantopus) as having the bill obtusely rounded, and Euryapteryx (type gravis) as possessing a bill “not so obtuse as the former.” Lydekker (1891) included the round-billed skulls in Emeus (“the mandible is in the shape of a wide U”) and proposed the genus Pachyornis (type elephantopus) for those with a sharp narrow beak. Parker (1895b) followed the same arrangement, and made the further observation that the broad-billed skulls (Emeus) lacked the antrum cavity.
Oliver (1930) reversed this, stating that Emeus had a pointed bill and Euryapteryx a broad one. This arrangement appeared to be right according to the then known indivi¬ dual skeletons, three only, belonging, apparently, to these genera, i.e., the broad-billed Eu. ponderosus from Riverton (Otago Museum), Eu. kuranui (Canterbury Museum) and the sharp-billed Amodeo Bay skeleton which the present writer (1927) had described as C. geranoides and Oliver referred to Emeus exilis: the broad-billed skull of the type of Em. exilis Oliver reported as being doubtfully associated with it.
Oliver’s grouping, however, ignored the long narrow beaks known to exist on large skulls, i.e. Pachyornis immanis of Parker (1894, 224), P. elephantopus of Parker (1895b, 375, pi. 60, fig. 22) and Mesopteryx sp. f3 of Parker (ibid. 378, pi. 60, figs. 20-21).* Except for the latter these narrow sharp beaks are not certainly associated with their crania, though of one of them Parker had “no doubt that they belonged to the same individual” ; moreover they were found with the crania, are appropriate in size to them and much too large for any known species of Emeus.
^Figured by Oliver as “Skull of Emeus.”
9
In the Otago Museum collection there are several perfect skulls not, however assoc, ated with skeletons, with a broad rounded bill and a collapsed ^trum (PL J, S- ■ ,
pcr contra there are twoskuOs with a cran ^ tha - * eto J ^
expanded antrum (PL 8, figs. 1 and 2). It is not certain in these cases hat the beaks belong to the crania, but again they are of the right S1ze and much too large for any
known species of Anomalopteryx or Emeus.
None of the large, sharp-beaked skulls, it will be noted, was found with its skeleton, and the only appropriately sized leg-bones with which they might have been affiliated were of the same size and proportion as those of Eu. ponderosus and Eu. kuramir Lyde t ter (1891a, 316) linked the sharp-beaked skulls with tibiae that were strongly inflected, an with certain very wide metatarsi; it happens that he was mainly right, but he had, ai that time, no real grounds for this association, for the type skeleton of elephantopus, the species on which he established his sharp-beaked genus Pachyornis, is composite, a recon¬ struction from mixed bones. From a series of small North Island individual skeletons obtained during our own collecting, and from large skeletons secured by Canterbuiy Museum from the Pyramid Valley swamp, it has now been ascertained that three types of skull can be recognized : (a) sharp-billed, as in Anomalopteryx , but rather shorter and broader than in that genus ; (b) with a narrow bill moderately rounded at the tip , and (c) with a broad, obtusely rounded bill. (Text-figs. 11a, b, c, p. 45.) These three types of skull adequately separate the genera (a) Pachyornis, (b) Emeus, and (c) Euryapteryx, and are found to be associated with different forms of sterna and also with certain types of leg-bones. The latter, however, present their own problem in classification, i.e. the difficulty, especially marked in the three genera just named, of distinguishing species or
genera by measurements of leg-bones.
Size and Proportions of Leg-bones.
The fact is, many leg-bones, taken by themselves, cannot be identified by their size and proportions as being of this or that species or genus, and sometimes even a careful study of their form will not help us. For instance, the three tibiae shown in outline on Plate 13 are almost indistinguishable from one another, yet they belong to three different genera in two separate sub-families. When, however, examination can be made of sets of leg-bones of individual birds, distinguishing characters readily present themselves in the length, and sometimes the proportionate breadth of the femur and metatarsus in relation to the length of the tibia. The three genera included on Plate 13 are An&malop- tcryx , Emeus and Euryapteryx ; but, when we come to distinguish the last, Euryapteryx, from Pachyornis, we are faced with the further difficulty that these two genera not only have the same relative lengths of the three leg-bones, but also exhibit the same range in width of bones, from moderately stout to extremely massive forms. In the North Island the species of Euryapteryx are small, and are exactly matched as to relative length and pro¬ portions of the leg-bones by like-sized species of Pachyornis, while in the South Island the species of both genera are larger and again exhibit an almost identical range of sizes and of attainment of massive proportions in their leg-bones.
At one time, having in mind the exact similarities in leg-dimensions in all these heavy footed moas, and the occurrence together, in the same sand-dune deposits, of sharp and broad-billed individuals with the same sized leg-bones, I had come to regard them as representing species exhibiting sex-dimorphism in the skull, as in Heteralocha, the New Zealand huia. The accompanying differences in the sterna, however, and slight modifica¬ tions in the form of the leg-bones, together with the retention of five outer phalanges in the sharp-beaked forms and their reduction to four in the broad-billed (and in Emeus), indicated that the birds stand related by these latter characters rather than by similari¬ ties in the leg-bones, which are to be regarded as the result of parallel development.
10
The facts just related, considered in conjunction with others mentioned below, pro¬ vide the basis of the classification set out in paragraph 8 following.
1. The acceptance of a wide range both in length and proportionate width of the leg- bones of any one species seems to be inevitable ; this clearly emerges from the dimensions of 43 individuals of Anomalopteryx didiformis , mostly from the North Island, including 15 from the one locality, Lake Waikaremoana, where the con¬ ditions indicate that they were approximately contemporary.
2. The relative length to one another of the three leg-bones is a more important criterion for classification than length and breadth of the bones. In the largest moas, Dinornis, the metatarsus is well over half the length of the tibia; in all the others it is hardly more than half as long, usually considerably less. A decrease in the length of the femur generally accompanies that of the metatarsus.
3. On the basis of the dimensions and proportions of leg-bones alone, there would, at first sight, seem to be a single line of evolutionary development towards shortei and stouter legs with increased shortening of femur and metatarsus. This appar¬ ently single line of development proves to be double and parallel : i.e. it has taken place in two groups here accorded subfamily rank. These groups are well defined by differences in the form of the skull and beak (4), by the number of phalanges on the outer toe (5) and less clearly by the form of the sternum (6).
4. The tallest moas, Dinornis, with long metatarsus, have an exceptionally broad flattened skull with a broad flat beak; the outer toe has five phalanges. These comprise the family Dinornithidae (Oliver 1930). All the others are shorter birds with metatarsi of reduced length and the skulls higher or moie rounded in sec¬ tion. The beak also is higher whether it is narrow and sharp, or broad and blunt. These are the family Anomalopterygidae (Oliver ibid.).
5. (a) Of the Anomalopterygidae three genera have the normal five phalanges in
the outer toe: they comprise the subfamily Anomalopteryginae. They also have skulls with sharp-pointed beaks and with expanded maxillary antra. (Text-fig. 10b, p. 42.) Genera Anomalopteryx, Megalapteryx and Pachyornis.
(b) Two genera of Anomalopterygidae have four phalanges in the outer toe (subfamily Emeinae ) i one of them, Emeus, has a narrow beak with a some¬ what rounded tip, and with the maxillary antra expanded, but not to the same extent as in the Anomalopteryginae. The other genus, Euryapteryx, has a broad beak with a rounded tip and with the maxillary antra collapsed. (Text-
fig. 10a.)
6. The sternum is short and broad or moderately broad in the Dinornithidae and Anomalopteryginae ; it is long and narrow in the Emeinae.
7. Parallel development of shorter and stouter leg-bones is especially marked as between Pachyornis (Anomalopteryginae) and Euryapteryx (Emeinae).
8. To repeat : the tall Dinornithidae, with flattened broad skulls and the normal num¬ ber of phalanges, are regarded as standing apart from the others. In the Anoma¬ lopterygidae the genera Anomalopteryx and Megalapteryx are slenderest, and in this respect, and in their normal number of phalanges, stand nearest to the Dinorni¬ thidae. ’ Pachyornis, which also retains five outer toe phalanges, exhibits increased stoutness and curvature of the leg-bones (bandy-leggedness) and stoutness of body (broad pelvis and sternum) ; this is developed to a moderate extent in the small North Island species, but to an amazing degree in the large species of the
11
South Island. The genus Emeus (four outer-toe phalanges) has diverged rom t e proportions of Anomalopteryx only moderately in the direction of shorter and stouter limbs; in skull form however it shows a slight increase in breadth and bluntness of the bill and reduction of the antrum, thus faintly foreshadowing the very broad bill and completely collapsed antrum of Euryapteryx. The latter genus has also gone further than Emeus, indeed nearly as far as the South Island species of Pachyornis, in increased breadth and massiveness of the leg-bones. These lela- tionships may be indicated graphically thus —
DRNIS ANOMALOPTERYX PACHYORNIS EMEUS EUR YAPTER V
We rely, therefore, on structure and form rather than on size, for generic characters ; but we still have to depend on differences in size for the separation of species. It will appear that all the specimens of Anomalopteryx are included in the one species, except for the fragments of much smaller bones which somewhat precariously sustain A. antiquus. Similarly, except for an exceptionally large femur and tibia on which M. benhami is founded, all the specimens of Megalapteryx are included in one species. There are no specially marked “out-sizes” in the series of North Island skeletons of Pachyornis and Euryapteryx, yet there are grounds for admitting more than one species in each genus.
In the first place there is, in the North Island skeletons of Pachyornis and Euryapteryx, a much greater total range in sizes than in those of A. didiformis, notwithstanding that they are smaller birds. For instance the range in total length of the three leg-bones in 40 specimens of A. didiformis is barely 30% above the smallest total length, whereas it is 38% in the same number of Euryapteryx skeletons, and over 40% in specimens of Pachy¬ ornis. In the second place there is, about the middle of the otherwise even sequence from the smallest to the largest Euryapteryx specimens, a small but clearly marked break both in size and proportions of the leg-bones. The very large range in sizes, together with the break in the sequence, is considered to warrant the admission of two species. A pre¬ cedent for this course may, as Dr. Falla reminds me, be found in the general recognition of Apteryx haasti and A. oweni as separate species, notwithstanding the intermediates in both size and plumage that occur. Among the smaller specimens of Euryapteryx there is an even better separated group of still larger skeletons which are recognized as a third North Island species of this genus. In Pachyornis the specimens fall into two distinct size groups, and although this may be due to there being fewer specimens available for measurement, they are accepted as representing two North Island species in this genus also.
12
In the same way in Emeus there is an even graduation of sizes in the specimens attri¬ buted to Em. crassus; but there is a distinct drop in size to the largest of the next group, which is accordingly identified with Em. huttonii. While some of the breaks in the sequence of sizes may ultimately be closed by later discovered specimens, it is con¬ sidered that in the meantime the differences that exist should be recognized and the corresponding specific names retained.
Note re Synonymy : It will be clear, from what has been said above about the leg-bones in Pachyornis and Euryapteryx, that many of the previous descriptions of material will have included specimens of both genera under one name. In order to avoid omission of refei- ences and at the same time to include under the names here adopted only indubitable references to the species concerned, I have set out, separately under the relevant species, the references that are part Pachyornis and part Euryapteryx .
DESCRIPTION OF GENERA AND SPECIES.
This section commences with a detailed examination of the skeleton of Anomalopteryx didiformis, of which I have 32 individuals of our own collecting and measurements or the published records of 11 other individual specimens. The variation of the different chaiac- ters will be recorded, and discussed as to the range which may be expected within a species ; the relative variation of skull and pelvis as to the leg-bones will be significant in this respect. A brief analysis of fifteen skeletons or partial individuals of ,M egalaptet y.i will follow, and the material available in the other four geneia, Pachyornis, Emeus, Euryapteryx and Dinornis, will be discussed in the same manner, i.e. dimensions of all individuals known and of types of the species that have been proposed will be set out as evidence for the classification proposed and as conveniently arranged records for future students. I may appear to have given over-full bibliography, synonymy of species and other references, but these details are so scattered and have proved so tedious in the compilation that I feel I should at least make them available.
In each species the leg-bones will first be discussed and the remaining characters in the order skull, vertebral column, pelvis and sternum. For brevity the tibio-tarsus and the tarso-metatarsus will be called tibia and metatarsus.
Wherever dimensions of leg-bones are given, either in tables or in the text, they will be in centimetres in the following order.
Length. Proximal Width. Middle Width. Distal Width. Girth.
17.5 5.75 3.20 7.83 8.1
32.8 18.2 44.7 46.3
The second line gives the widths as percentages of the length as used by Oliver. I have added the girth, or circumference at the middle of the shaft, not because it is in itself especially significant, but to facilitate comparison with the dimensions supplied by earlier students who not infrequently recorded length and girth only.
The measurements have all been made between uprights and, to measure widths, the bones have been placed at right angles to the direction of the measuring slide (Text fig. 1) The shorter measurements were made to tenths of a millimetre, and the pioportions calculated on this basis: this was only because the calipers gave this refinement and it was just as easy to read to two places of decimals as one on the slide-rule by which the proportions were obtained. These dimensions have been given in the tables only to the nearest millimetre, partly because recordings to two decimal places suggest a degree of accuracy that is really quite spurious when the varying shapes of the bones are con¬ sidered and also because the figures themselves, if given to two places, suggest a greater
13
Text-fig. 1. Measuring apparatus.
difference than is really significant in one or two tenths of a millimetre. I mention this because, by calculating from the measurements as now adjusted to the nearest millimetre, the results might differ slightly from the proportions entered in the table, which were, in fact, calculated from finer detail than is actually recorded.
In the systematic portion, which follows immediately, the necessary nomination of lectotypes is made in the synonymy against the appropriate citation of the species con¬ cerned.
Genus Anomalopteryx Reichenbach, 1852.
1852 . Anomalopteryx Reichenbach, Av. Syst. Vog. p. xxx. Type, by monotypy, Dinornis
didiformis Owen.
1897 . Anomalornis Hutton, Trans. N.Z. Inst., 29, p. 543. Substituted for Anomalopteryx,
supposed wrongly by Hutton to be preoccupied.
Two species: —
A. didiformis (Owen), in which are included Dinornis dromaeoides Owen, Dinornis parvus Owen, Anomalopteryx fortis Hutton and the metatarsus of Palaeocasu- arius velox Forbes.
A. antiquus Hutton, p. 29.
Anomalopteryx didiformis (Owen), 1844.
Dinornis didiformis: Owen, Trans. Zool. Soc. London vol. 3, pt. 3, p. 242, pi. 27, figs. 3-6.
Dinornis dromaeoides: Owen, ibid, P. 253, pi. 22, figs. 1, 2; pi. 23, fig. 1. Type: femur from Poverty Bay. No. f.16 in Table of Admeasurements, Owen p. 248; originally in Museum of Royal College of Surgeons. Casts in British Museum (18598, Lydekker p. 267) and Canterbury Museum.
• • Dinornis didiformis, Owen, Cat. Foss. Org. Remains (Mammalia and Aves) Mus. Roy. Coll. Surg. p. 361.
Dinornis dromaeoides : Owen, ibid, 369.
Dinornis didiformis : Haast, Trans. N.Z. Inst. 1, pp. 82, 83 (2nd ed. 1875, pp. 23, 24). .. Dinornis parvus Owen: Proc. Zool. Soc. for 1882, no. 1, p. 2 (nom. mid.)
Dinornis parvus Owen, Trans. Zool. Soc. London, vol. 11, pt. 8, 233, pis. 51-58. Type: Skeleton from Pokororo, Nelson, in the British Museum (A.3, Lydekker, p. 279).
Anomalopteryx dromaeoides : Lydekker, Cat. Foss. Birds Brit. Mus. 266. Anomalopteryx didiformis : Lydekker, ibid, 275.
Anomalopteryx parva : Lydekker, ibid, 278.
13 Anomalopteryx dromaeoides : Sharpe, Cat. Ost. Vertebr. Mus. Roy. Coll. Surg. Lon¬ don, Part III, 430.
Anomalopteryx didiformis ■ Sharpe, ibid, 432.
1844 June 5
1845
1869 May . . 1882
1883 January
1891 April 25
1891 November
14
1891 November
1892 May . .
1893 May . .
1895
1897 June
1907
1930
1933 . .
Palapteryx dromaeoides : Hutton (part, femur) N.Z. Journ. Sci. new issue, vol. 1, pt. 6, 248.
Palapteryx plenus Hutton (part, femur) ibid, 248.
Anomalopteryx didiformis • Hutton, ibid, 248.
Palapteryx dromaeoides : Hutton (part, femur) Trans. N.Z. Inst. 24, 121.
Palapteryx plenus • Hutton (part, femur), ibid, 122.
Anomalopteryx didiformis • Hutton, ibid, 123.
Anomalopteryx fortis Hutton, Trans. N.Z. Inst. 25, 9.
Founded on three metatarsi a tibia and three imperfect femora from Glenmark in the Canterbury Museum. Of the metatarsi only one can now be identified in the Museum collection: it is here selected as the type.
Anomalopteryx didiformis : Parker, Trans. Zool. Soc. vol. 13, pt. 11, 378. Anomalopteryx parva '■ Parker, ibid, 379.
Anomalornis didiformis • Hutton, Trans. N.Z. Inst. 29, 547.
Anomalornis gracilis : Hutton (not of Owen), part : tibia and metatarsus, ibid, 546. Dinornis dromaeoides ’ Rothschild, Extinct Birds, 194.
Anomalopteryx didiformis • Rothschild, ibid, 202.
Anomalopteryx parvus '• Rothschild, ibid, 202.
Palaeocasuarius velox Forbes (part) Rothschild, ibid, 220 (metatarsus).
Dinornis dromaeoides : Oliver (part), New Zealand Birds, 41.
Anomalopteryx didiformis • Oliver, ibid, 44.
Anomalopteryx parvus ■ Oliver, ibid, 45.
Dinornis dromaeoides : Lambrecht, Handbuch der Palaeornithologie 140. Anomalopteryx didiformis ‘ Lambrecht, ibid, 143.
Anomalopteryx parvus : Lambrecht, ibid, 144.
The type is a metatarsus from Poverty Bay. It should be in the Museum of the Royal College of Surgeons, but I could not identify it there. A cast of the type is in the British Museum (No. 18595; Lydekker, p. 276); its dimensions and proportions are:
17.5 5.75 3.20 7.83
= 100 32.8 18.2 44.7
An individual skeleton with a metatarsus of approximately this size and proportions has the following leg dimensions:
Mangaotaki, A.M. 156
|
Femur |
24.0 |
7.9 |
3.13 |
8.1 |
|
— 100 |
32.9 |
13.0 |
33.7 |
|
|
Tibia |
37.7 |
10.05 |
3.34 |
5.4 |
|
— 100 |
26.6 |
8.8 |
14.4 |
|
|
Metatarsus |
17.7 |
5.75 |
3.1 |
7.6 |
|
= 100 |
32.5 |
17.5 |
42.9 |
In this species the femur (PI. 1, fig. 3) is long and of moderate width, not slender as in Megalapteryx (fig. 2) ; seen from the side it is almost straight and only slightly arched upward. The muscle ridges are well developed, but less strongly than in Dinornis. On the posterior surface there are two ridges behind the medullary foramen , the outei of these lies immediately behind the foramen, the inner some distance behind. In Dinornis, they are nearly opposite each other; in Emeus and Euryapteryx they are usually confluent, but when they are, as occasionally, separate, their relative position is as in Anomalopteryx.
The tibia (PI. 2, fig. 3) is relatively stouter than in Dinornis and Megalapteryx, but less stout than in Emeus ; its outer margin is straight with the proximal third elevated as the fibular ridge; its inner margin is deflected inwardly just above the trochlear bridge, more
15
. 7 - Tina rvrrmmal end is expanded and the pro-
so than in Megalaptcry.v, less than in ^ degree the direction of the
cnemial crest is strongly deflected outward, fo g deflected
ectocnemial. Except that it is usually strmghter audits disW ® ^ proportionate
inwardly, it does not differ veiy much fro Fmeus (fig 5)
width the metatarsus (PI 3 fig. 3) stands between Megalaptery.v (fig. 2) and h ( g. )■
„.ideriw» L,er than the met.t.r.h., ,.d both bo*. are towr m pro-
portion to length of the tibia, than in species of Emeus.
The 43 skeletons or individual sets of bones that I refer ^o this^spec^es^displnp^gretd
ZSSZ Z eertain bird, ate larger
Among individuals which, for instance, may have been grouped for then posse, a long tibia, some will be found to have the femur and metatarsus corresponding y ong, in others both these bones will be relatively short; or the femur may be i long _a metatarsus short, or vice versa. This is sufficiently demons ra e y ‘ , . . ,.
femora, the tibiae and the metatarsi each in order of length independently of their indi¬ vidual association, and then drawing connecting lines between the bones o in ivi (Table 1). The considerable range in size and proportion of the three leg-bones in different individuals indicates the unreliability of a table of dimensions and proportions of a series of any one leg-bone as a basis for the establishment of a species. . The table also shows how easily mistakes can be made in attempting to group into species-sets the mixed bones from a swamp deposit.
The extent of variation in follows : —
Femur
Tibia
Metatarsus
the length of the leg-bones recorded in Table 1 is
|
Max. |
Mean. |
Min. |
Range. |
|
27.5 |
23.3 |
20.0 |
7.5 |
|
43.4 |
37.8 |
32.3 |
11.1 |
|
21.2 |
17.9 |
15.7 |
5.5 |
as
A wide range of variation is also found in the proportionate width or stoutness of the bones as the following record of maximum and minimum proportionate width (length = 100) shows. f
|
Femur. |
Tibia. |
Metatarsus. |
|||||
|
Prox. |
Mid. |
Dist. |
Prox. Mid. Dist. |
Prox. Mid. |
Dist. |
||
|
Max. 37.5 |
15.7 |
43.0 |
29.4 |
10.1 16.7 |
36.1 20.7 |
49.0 |
|
|
Mean 33.4 |
13.9 |
37.6 |
27.3 |
8.2 15.2 |
32.4 18.2 |
44.0 |
|
|
Min. 29.3 |
12.1 |
32.3 |
25.2 |
6.4 13.8 |
28.8 15.8 |
39.0 |
|
|
Another aspect of this variability is |
indicated by the |
range |
in proportionate |
width |
|||
|
exhibited by |
ten tibiae of approximately the same length |
(37.4 to 38.0 cm.) . |
|||||
|
Length. |
Proximal. |
Mid. |
Distal. |
||||
|
Max. |
• • |
= 100 |
28.9 |
9.35 |
16.4 |
||
|
Min. |
— 100 |
25.3 |
6.68 |
14.4 |
^Tables A to O, giving detailed dimensions and proportionate widths of leg-bones, are together follow¬ ing the plates; Tables 1 to 19, giving other dimensions, are distributed through the text.
tThe above maxima and minima are not proportionate widths taken from a single bone; the proximal width may be from one bird, the middle width from another and the distal from still another. Some¬ times in an individual (i.e. Tring Museum) the femur will be broad in all dimensions, with the propor¬ tions varying considerably in the other bones.
16
Table 1. A. didiformis.
Femora, tibiae and metatarsi each arranged in order of length: the lines connect the bones of individual birds.
Femur. Tibia. Metatarsus.
21.1
20.0
lo nf the variability in the relative stoutness of We may include, as a further exampl f tt ^ individuals from the same locality
the leg-bones of this species, the following detm
(e.g. Waikaremoana, A.M. 70 and 71) •
F. 25.75 = 100
A.M. 70. 8.46 3.41
32.8 13.2
T.
M.
39.6
100
19.9 100
11.67
29.4
6.41
32.2
3.30
8.3
3.34
16.8
9.54
37.0
6.07
15.3
8.24
41.4
A.M. 71. 25.15 8.88
100 35.3
39.3 = 100
18.7 = 100
11.17
28.4
6.4
34.2
3.60
14.3
3.75
9.5
3.68
19.6
9.38
37.2
6.14
15.6
8.46
45.2
It will be seen that A.M. 70 is definitely more ^le that its tibia has a more expanded proximal end , thei e a , differences between their skulls, pelves and sterna.
Thu», throughout .hi, ,eri„ ”f £ KX'taXS
arid'trregular i”b"»,on tto b" . of MivM.al,. An indWdu.l with » wide femur mu,
h.VtheS nurrower uud the „e«„,u, either wide or »nw Vj^ZtZl. U may be wide in one transverse dimension and narrow in one or both of the otheib^ should also be noted that Table A includes the dimensions of only those bones which or part of an individual ; the inclusion of dimensions of the numerous mixed bones av 1 - able would show an even finer gradation of variation.
As will be seen later (p. 20) there is considerable variation m size and proportion o ^ skull and pelvis ; but I have not found it possible to associate any one type ot skull oi pelvis with either stout or slender leg-bones. Neither do the specimens from any one locality or area exhibit anything approaching a uniformity m size or proportions such as would warrant the recognition of local subspecies or varieties. The largest and nearly the smallest individuals in my material are from one restricted area, e.g. the Waikaremoana lake-barrier, and the range is nearly as great in the specimens 10m Mangaotaki in the North Island and Mt. Arthur in the South Island.
It might be suggested that the specimens we are considering represent merely an evolutionary development over a long space of time, and that the smaller forms are of an earlier period than the larger. Time may possibly have seen an increase in size in these birds, but it did not eliminate the smaller ones, for at Waikaremoana all sizes are approximately contemporary. The slip which fissured the sandstone and formed the caves that trapped the moas is of Recent origin (Marshall, 1927), and the birds must have lived subsequent to it. They are even later than immediately after the formation of the caves, for their skeletons lie above the Gisborne pumice layer which later covered the lake-barrier and was washed into the caves. I have no means of estimating the age of the Mangaotaki specimens, which also vary greatly in size.
At all events, it is clear that, on the basis of the size and proportion of the leg-bones, we have here a species exhibiting gradual and continuous variation in all dimensions and proportions. In this extensive range of variation the type of A. didiformis occupies a middle place, while the type of A. parvus is one of the smallest. There is no break between them, and A. parvus is accordingly included here as a synonym of A. didiformis. The type of Dinornis dromacoides Owen, a femur from Poverty Bay, also falls well within the range of the femora of A. didiformis. The type itself is missing, but there are casts in the British and the Canterbury Museums, and one has only to place the cast alongside the femur of an indubitable A. didiformis to recognize that D. dromacoides is identical with it.
18
I may add that the femur which Owen (T.Z.S. 3, PI. 24), judged to be that of his D. didi- formis, is probably that of Eu. exilis. Lydekker (1891a, 273) in discussing a cast of this latter femur recognized its distinctness from didiformis.
Previous writers, in commenting on the range in sizes of the moas, have suggested that one sex may have been the larger, as the female is in Apteryx. Hutton (1892b, (H) pointed out that, if this were so, the species should occur geographically m pairs, but those recognized by him did not do so. I am unable to discern two separate groups m A. didiformis; the leg-bones vary evenly as to length, and haphazardly as to relative width; neither can I detect any association of, say, a larger or bioadei pelvis, 01 a arger skull, with either the larger or smaller leg-bones. If the sexes were of different sizes,
the size-range of the sexes overlapped.
Skull: For dimensions and proportions see Table 2.
Variability in the form and proportion is as marked in the skull as in the leg-bones, but there are sufficient constant characters to enable the skull of Anomalopteryx to be readily distinguished from that of the other genera. In general it is moderately wide, with very large temporal fossa and a long tapering sharp-pointed bill.
Viewed from the side (PL 4, fig. 3) the roof is evenly arched, sometimes with a slight eminence (single, not double) in line with, or just behind the postorbital processes ; t e premaxilla is long and curves downward to an acute tip, its lower margin also being slightly downcurved and slightly over-reaching the tip of the mandible. The posterior (paroccipital) margin of the tympanic cavity is oblique, sometimes convex, usually forming an angle with the upper margin, which curves forward and downward to merge anteriorly with the zygomatic process; sometimes, however, the outline of the cavity is an even arch slightly higher than wide. The temporal fossa is wide and deep, the tem¬ poral ridge extending from its confluence with the lambdoidal to high up on the roo , it is wider than the orbit ; the mid-temporal ridge is always present and is sometimes prominent. The post-temporal fossa is a wide convex area, and the zygomatic processes
moderately long and acute. The postorbital process, in lateral view, projects backwards
to a greater or less degree. The upper margin of the orbit is usually evenly arched, but is sometimes obtusely angular, or even right-angled where the postorbital process joins the upper margin. The preorbital (lachrymal) curves evenly outwards and downwards, a deep notch near its extremity forming the mesial wall of the lachrymal foramen w ic is completed ectad and anteriorly by the maxillo-nasal (absent from fig. 3), a narrow slip of bone extending from the nasal along the front of the lachrymal and expanding below to join the maxilla.* In a few cases the outer wall of the lachrymal foramen is com¬ pleted by further growth of the lachrymal bone itself.
In posterior aspect the skull in Anomalopteryx sometimes appears to be broader than in Emeus, but this is only because of its lesser height. The supra-forammal ridge is no much swollen (as, usually, in Emeus) and the supra-occipital crest is a narrow, distinctly raised ridge ; the supra-occipital pits are small or vague ; there is a shallow depression on each side above the paroccipital processes. The outer borders of the paroccipital pro¬ cesses may be sinuate with the lower extremity of the process somewhat pointed, or th y may curve evenly to a rounded lower margin ; generally they reach to about halt way between the levels of the condyle and the mamillar tuberosities. The latter are of moder¬ ate size, and separated by a wide arch.
In dorsal view (PI. 5, fig. 3) the single slight eminence on the roof, if evident, varies in size and position. The lambdoidal and temporal ridges are almost invariably confluent (in one skull there is a separation of 1 mm. on one side only) ; the temporal ridges may
* Parker (1895b, 383) had not seen this bone in Anomalopteryx ,
but thought it might be a separate element.
19
„„ ivim-lv and then turn sharply outward.
form an even curve or they may con^rg nd extend laterally only slightly beyond
The post-orbital p,~ «.* n7„arr.w with a sub-.opt. tipi it. lateral
the squamosal prominences. In - . f ature appears to be constant. Scattered
margins are slightly incurved, anc liS ' . ^ skulls- but there is no occurrence
nits annarentlv pneumatic, occur above the orbit m all skulls, out
KSS, supposedly tor .rest feathers, that oe.ur m
rpi 7 fio- 2) exhibits less variation. The basitempoial platform is The ventral aspect { PI. 7, g. mamillar tuberosities rather less than
of the usual form, with the space aero ss them passes. The median
(but occasionally the same as) considerably: it may be
longitudinal depiession m e p , ’ at e;ther end. It is least developed in
scarcely discernible, or a distinct groove deepei at eitner enu. f
voung skulls and its greater depth and distinctness seem to be the outcome ot the
thicken ng of the inner wall of the eustachian grooves. The rostrum has a slight con-
sSioTat about the middle of its length, and its triangular processes are moderately
developed.
The maxillo-jugal arch, formed of the usual three parts, maxitta jugal and quadrate jugal is a nearly straight bar, exhibiting only a slight convexity m its posterior portion. Inthis respect it differs from Pachyornis (PI. 7, fig. 3) in which the maxillo-jugal arch curves markedly outwards. The expanded anterior portion of the maxilla coveis a spacious antrum cavity (PI. 5, fig. 3) which opens posteriorly, usually by a wide round aperture, though occasionally by a narrower flattened, almost slit-like passage. The anterior pai t of the maxilla is partly overlaid ventrally by a thin triangular expansion of the palatine, the two together forming a maxillo-palatine plate. The three components of the maxillo- jugal arch apparently fuse together early, for in only one of my specimens can the junction between the maxilla and jugal be seen: in this case the jugal covers the maxilla dorsally up to where it abuts against the lachrymal. The jugal also apparently covers
Table 2. Dimensions and proportions of skulls; A. didiformis. (See Note top of page 24.)
|
Auckland Museum Collection No. |
121 |
72 |
70 |
71 |
82 |
|
Length: total |
13.5 |
14.1 |
14.1 |
13.38 |
|
|
Length: paroccipital to preorbital |
|||||
|
process |
8.0 |
8.0 |
8.25 |
7.9 |
|
|
Height of cranium |
4.46 |
4.40 |
4.40 |
||
|
Width across paroccipital processes.. |
6.37 |
5.9 |
5.83 |
5.94 |
5.43 |
|
Width across squamosal prominences |
7.55 |
6.6 |
7.00 |
7.00 |
6.54 |
|
Width across temporal fossae . . |
4.38 |
4.53 |
4.57 |
4.37 |
|
|
Width between temporal ridges |
2.64 |
2.80 |
2.76 |
3.05 |
|
|
Width across post-orbital processes.. |
8.08 |
7.8 |
7.83 |
8.04 |
7.53 |
|
Width across pre-orbital processes |
4.45 |
5.04 |
5.00 |
4.74 |
|
|
Width of tympanic cavity |
1.8 |
1.99 |
1.84 |
1.85 |
|
|
Width of temporal fossa |
3.4 |
3.38 |
3.30 |
3.30 |
|
|
Width of orbit |
2.8 |
2.85 |
3.13 |
2.90 |
|
|
Space between lambdoidal and tern- |
|||||
|
poral ridges . . |
0.0 |
0.0 |
0.0 |
0.0 |
|
|
Proportions (percentage of total |
|||||
|
length) : — |
|||||
|
Height |
31.6 |
31.2 |
32.8 |
||
|
Paroccipital width . . |
43.8 |
41.3 |
42.1 |
||
|
Squamosal width |
49.0 |
49.6 |
49.7 |
49.6 |
|
|
Width across temporal fossae . . |
32.5 |
32.1 |
32.4 |
31.7 |
|
|
Post-orbital width . . |
57.8 |
55.5 |
57.0 |
57.1 |
|
|
% width between temporal ridges: |
|||||
|
temporal fossae |
52.5 |
60.2 |
61.8 |
60.4 |
69.5 |
|
% widths, squamosal: post-orbital |
93.4 |
85.0 |
89.5 |
87.1 |
87.0 |
20
the anterior two-thirds of the quadrato- jugal, but I have not been able to discern the exact form of the junction. The quadrato- jugal moiety of the arch is slightly expanded, and its extremity has an inner facet for articulation with the quadrate.
The palatine (PI. 7, fig. 2) : anteriorly the palatine expands horizontally to form a narrow triangle whose outer border fits into a corresponding groove on the ventral sur¬ face of the maxilla, the two together forming the maxillo-palatine plate mentioned above. Posteriorly it expands vertically to form a thin triangular lamina which twists mesally to junction with the oblique postero-dorsal margin of the vomer (Text fig. 2b) ; postero- cctally it effects an oblique junction with the anterior margin of the pterygoid, and pos- teromesally it sends out a small stout triangular process which fits between, and separates, the posterior ectal border of the vomer and the antero-mesal border of the pterygoid (Text fig. 2b). The vomer and pterygoid, however, join one another above this process (Text fig. 2a). The pterygoid is of irregular shape: from its junction with the vomer and palatine it passes backwards and outwards, effecting at its middle an articulation with the basi-pterygoid processes; posteriorly and ectally it articulates with the lower border of the orbital process of the quadrate. In two of the three perfect skulls I have with the vomer still in situ , its slender paired laminae are united anteriorly, where they enclose the rostrum ; in the other specimen they are separate anteriorly. Posteriorly they are separate and diverge to make junction with the palatine and pterygoid as described above.
The vomero-palato-pterygoid junction in Anomalopteryx agrees with that described by Parker (1895b) ; it also clearly confirms the close resemblance to the condition in Rhea affirmed by Pycraft (1900, p. 262), as against Parker’s interpretation. In fact, in the immature skull of Anomalopteryx we see clearly that encroachment of the palatine over the vomero-pterygoid junction (Text fig. 2a and b) which is held by Pycraft (pp. 206-7) to indicate the first stage of the evolutionary change from the primitive arrangement in Dromaeus ( Dromiceius ) towards the derived neognathous condition; it is also of significance for the possible phyletic unity of the struthious birds (see p. 84). It may be added
|
63 |
55 |
102 |
151 |
150 |
51 |
149 |
66 |
155 |
69 |
89 |
|
13.5 |
13.75 |
13.0 |
12.8 |
12.5 |
12.8 |
13.9 |
13.25 |
13.5 |
12.9 |
14.2 |
|
7.63 |
7.70 |
7.45 |
7.4 |
7.5 |
8.1 |
7.75 |
7.68 |
7.5 |
8.2 |
|
|
4.40 |
4.31 |
4.47 |
4.07 |
4.07 |
4.2 |
4.88 |
4.54 |
4.4 |
4.47 |
4.14 |
|
5.46 |
6.48 |
5.25 |
5.52 |
5.7 |
5.65 |
6.06 |
5.80 |
5.3 |
6.29 |
5.9 |
|
6.35 |
6.70 |
6.96 |
6.37 |
6.1 |
6.30 |
6.82 |
6.60 |
6.35 |
6.62 |
6.95 |
|
4.35 |
4.28 |
4.18 |
4.10 |
3.97 |
4.12 |
. 4.43 |
4.50 |
4.32 |
4.15 |
4.18 |
|
2.37 |
2.90 |
2.84 |
2.64 |
2.23 |
2.9 |
2.78 |
2.75 |
2.37 |
2.66 |
2.17 |
|
7.0 |
7.26 |
7.87 |
7.28 |
6.9 |
6.82 |
7.50 |
7.15 |
7.00 |
7.49 |
7.36 |
|
4.5 |
4.86 |
4.30 |
4.58 |
4.60 |
4.59 |
4.85 |
4.57 |
4.22 |
||
|
1.73 |
1.94 |
2.06 |
1.80 |
2.0 |
1.92 |
1.8 |
1.70 |
1.75 |
2.00 |
1.94 |
|
3.22 |
3.29 |
3.48 |
3.1 |
3.06 |
3.07 |
3.6 |
3.30 |
3.18 |
3.25 |
3.50 |
|
2.80 |
2.91 |
2.7 |
2.75 |
2.37 |
2.9 |
2.83 |
2.60 |
2.90 |
||
|
0.0 |
0.0 |
0.0 |
0.0 |
0.0 |
0.1 |
0.0 |
0.0 |
0.0 |
0.0 |
0.0 |
|
32.6 |
31.3 |
34.4 |
31.8 |
32.55 |
32.8 |
35.2 |
34.4 |
32.6 |
34.6 |
29.25 |
|
40.5 |
47.2 |
40.4 |
43.1 |
45.5 |
44.1 |
44.0 |
43.7 |
39.3 |
48.7 |
41.6 |
|
47.0 |
48.7 |
53.6 |
49.5 |
48.8 |
49.2 |
49.0 |
49.8 |
47.0 |
51.3 |
49.0 |
|
32.2 |
31.2 |
32.2 |
32.0 |
31.8 |
32.2 |
31.9 |
34.0 |
32.0 |
32.2 |
29.5 |
|
51.8 |
52.6 |
60.7 |
56.8 |
55.2 |
53.3 |
54.0 |
54.4 |
51.8 |
58.0 |
51.9 |
|
54.5 |
67.7 |
67.8 |
64.4 |
58.5 |
70.4 |
62.7 |
61.1 |
54.8 |
64.1 |
51.9 |
|
90.7 |
92.2 |
88.5 |
87.5 |
88.4 |
92.3 |
90.9 |
92.3 |
90.8 |
88.3 |
94.3 |
21
and the Anomalopterygidae, for, notwithstanding its peculiarity in having the pterygoid bi-pronged anteriorly, it exhibits the same ventral encroachment of the palatine over the junction of the vomer and pterygoid as occurs in the other palaeognathous genera.
The mandible is stout, thus according with the large temporal fossa ; seen from above the rami are straight and moderately divergent postad; the symphysis is relatively long and pointed terminally. In lateral view it first curves downward, then upwards, and downwards again at the symphysis.
The variations in the different characters have been indicated generally in the above description, and, as to dimensions and proportions, may be noted more in detail from Table 2. I may add that no variation in any one character has been found in skulls of any particular form or size. A large temporal fossa, for instance, may be present in either a large or a small skull, and greater relative breadth may also occur in a long or a short skull. Similarly in the relation between skull and leg-bones in individual skele¬ tons, it cannot be stated that taller or larger birds have larger skulls — on the contrary a tall bird may have a smaller skull than a shorter one ; neither can it be determined that broader skulls, for instance, are associated with stout leg-bones, or indeed with leg-bones of any particular type. The form and size of skull are just as varied and haphazard as to the type of leg-bones with which they are associated, as the three leg-bones are to one another in different individual skeletons. The skull on the mounted skeleton labelled A. didiformis in the Dominion Museum has a striking peculiarity in that at the junction
of the lambdoidal and temporal ridges there is a deep angular depression on the right side, and a slight depression on the left side (Text-fig. 4). Otherwise this is a typical Anonmlopteryx skull. This skeleton seems to be composite : the legs are typical of an adult of the species, the pelvis however is immature and small for the skeleton, there are eight instead of six rib-bearing thoracic vertebrae, and the sternum is typical of Text-fig. 4. A. didiformis: skull of skeleton that of Emeus . in Dominion Museum.
22
Structural elements of the skull.
Two immature skulls and one juvenile have provided the following details of the form and extent of the constituent cranial elements.
Seen from behind the disposition of the bones is as described and illustrated by Parkei (1895b, p. 380 et seq. pis. 57 and 59) . The condyle is formed medianly by a wedge-shaped portion of the basi-occipital, and dorso-ectally by the exoccipitals. The mamillar tuber¬ osities are composed of basi-occipital, exoccipital and prootic.
On the base of the skull the basioccipital extends further forward than shown by Parker, as he himself suggested that it might ; it forms the base of the skull m the prootic region as well as in the occipital and abuts on the basisphenoid and its underlying basitemporal at the posterior basicranial fontanelle.
In lateral view (Text Fig. 5) the relations between the bones differ to some extent from Parker’s description. The parietal, supraoccipital, exoccipital and prootic are as he defined them, except that the parietal extends only a short way down the side and meets the alisphenoid by a curved junction. The latter bone is irregularly shaped; it sends a broad extension postero-dorsally to meet the squamosal, and a narrow process antero- dorsally to lie under the lateral projection of the frontal which it joins to form a as for the attachment of the separately ossified post-frontal or post-orbital process. Beneath the squamosal the junction between the alisphenoid and prootic passes downward and slightly forward; at the point where the prootic and basitemporal adjoin, the ahsphemH margin is notched ; so is the basitemporal margin, and all three bones form by their recessed margins the trigeminal foramen. From this point forwards the lower margin of the alisphenoid is slightly curved and then turns abruptly upwards to effect junction with the posterior margin of the frontal, forming finally the front margin of the narrow antero-dorsal process of the alisphenoid mentioned above. At the junction of alisphenoid, frontal and basitemporal, notches in these bones form the fossa containing openings for
the orbito-nasal, the oculomotor nerves and the internal ophthalmic artery (Parkei p.
388) In front of this fossa is the optic foramen, bounded, according to Parker by the orbitosphenoid above and the presphenoid below. My own juvenile skulls are deficient here and I am unable to supplement Parker’s observations on this point, except to note that the lower margin of the optic foramen seems to be formed, at least m part if not entirely, by an upward prolongation of the basitemporal.
23
Note: In measuring skulls the total (i.e. overall) length is the basis on which p ■ portionate widths are calculated, i.e. they are recorded as percentages of the total length. There is, however, a practical disadvantage in taking this as a basis m that scu s aie so seldom obtained with the beak entire and still fused m position on the cranium. e length from the condyle to the tip of the rostrum is also frequently unobtainable because
of abrasion or breaking of the rostrum.
On the other hand the length from the posterior margin of the paroceipital (Text fig. 6, n) to the front of the preorbital (m) is almost always obtainable. In perfect
skulls it varies from 58% to 59r/c (average 58.5%) of the length in Eu. exilis and Eu. curt ns, and from 57.8 to 59.2 (average 58.5%) in A. didiformis, so it is not possible to calculate the total length precisely from the known paroccipital-preorbital length. Never¬ theless, in all those cases in which the length of a not quite perfect skull is given in the tables, I have used a calculation from the paroccipital-preorbital length as a means of checking the estimate of the total length made by comparison with perfect skulls.
Vertebrae: The vertebral column is composed of 21 cervical vertebrae, three cervico- thoracic, three thoracic, 18 pelvic and 11 caudal.
Following the atlas and axis, Nos. 3, 4 and 5, and sometimes 6, present dorsally a subquadrate outline; the neural spine is represented by a pair of small subparallel pro¬ cesses, and there are distinct hyperapophyses. These vertebrae may be termed nape cervicals or nape vertebrae. In those immediately succeeding the nape series there is an abrupt change in the dorsal outline: the anterior, or diapophysial portion is expanded, the middle region is constricted, and posteriorly the elongated diverging post-zygapo- physes are separated by a median V-shaped depression; the neural spines are higher and divergent, becoming increasingly so, but at the same time lower in the succeeding cervicals until, from the 17th or 18th, they approach one another again and increase in height, coalescing on 20 and 21 to form a single, high, median neural spine. This single neural process increases in height on the succeeding thoracic vertebrae.
Hyperapophyses, present on the nape vertebrae, may, or may not be present on 6 to 9. The “neural ridge” of Hutton (1894, p. 160) rarely appears as a continuous even ridge; normally a constriction separates the neural spine from the hyperapophysis. The latter are present up to the 24th or 25th. The interzygapophysial canal is a small pore on the nape vertebrae, but it may be occluded on, or absent from, 4, 5 or 6. Thereafter it is
24
present up to 14, 15, 16, or 17, though in one skeleton it is present only up to 10. The interzygapophysial bar which forms the outer walls of the canal increases in width and thickness on the hinder cervicals, finally occluding the canal on or about the 14th to 17th as just described.
The ventral surface of the nape cervicals shows a median hypapophysial ridge, strongly developed on the axis and 3, progressively reduced on 4 and 5, and evanescent on or absent from 6 where parial hypapophyses appear mesally on the base of each parapo- physis. The pleurapophyses at first increase in length, becoming long and styliform by the sixth; in succeeding vertebrae they progressively shorten, this being accompanied by an increase in length of the parial hypapophyses, which by the 17th have approached one another closely, uniting on the 18th to form a median hypapophysis again. Txiis median hypapophysis has, therefore, a different structural origin from tnat on the nape cervicals where it is a median projection from the hinder part of the centrum. The hypa¬ pophysis is strongly developed on the remaining cervicals; on the thoracic vertebrae it changes form again, becoming a ridge or keel on the anterior part of the 22nd and 23i d , a notch divides the keel into an anterior and posterior portion on 24 and 25, or on 25 and 26, or the posterior portion alone projects from 26, becoming bifid on 27 with the two resulting knobs standing more widely apart on the pelvic veitebiae 28 and 29.
Of the pelvic vertebrae No. 28 has freely-articulating floating ribs, while 29 and 30 have floating ribs fused to the ilia; 31 to 34 bear transverse processes which extend out to and unite with the ilia. The number of pel /ic vertebrae is eighteen (PI. 11, fig- 3), which is one more than has hitherto been recorded ; this makes it accord numerically with the condition in E. huttonii (Benham 1934, pi. 7), though there is a difference in grouping of the component series ; moreover there is sometimes an extra vertebra present.
Excluding the occasional coalescence of the last thoracic vertebra (27) with the pelvis, there are normally seven pre-acetabular vertebrae (28 to 34). These are followed by four vertebrae (not three as previously described) without transverse processes (35 to 38) ; the last bears small neural canals higher up on the side than in the others and has thus escaped notice (Text fig. 7, 38) . Occasionally it has the pleurapophysis represented by a small splinter-like process (PI. 11, fig. 3, 38). In the pelvis of A.M. 117, and that of 186, vertebra 35 has transverse processes. The remaining seven vertebrae (39 to 45) have prominent transverse processes extending obliquely to the ilium. Vertebra 46 is occasionally fused to the pelvis, but is more usually a free caudal; counting it as such there are eleven caudals (46 to 56) of which either the last two or the last three are fused together. The foregoing account is based on adult specimens; the condition des¬ cribed is even more clearly determined in immature pelves (Text-fig. 7) with the synsa- crum not yet fused to the ilia.
Text-fig. 7. A. did if or mis: pelvic vertebrae immature.
A formula, ,u«h the follo.i.ff,
in the nelvis or svnsacrum, will facilitate indication of the variations that occur t Brouninff ofS the vertebrae Thus in the formula “28-34 (35-38) 39-45” which sets out fhe normal condition in Anomalopteryx, “28-34” represents the pre-acetabular vertebrae (35-38) indicates the four acetabular vertebrae which lack transverse processes; and “39-45” the remaining posterior vertebrae with transverse processes , the addition o ( )
in brackets would indicate the fusion of the first caudal with the pe vis.
Owen (1883, p.244), in describing the pelvis of A. parvus, also recorded eighteen “sacral” vertebrae. He missed the small eleventh (No. 38) ; but a misnumbering of the succeeding centra and pleurapophyses gave him an extra centrum, one for which there were no transverse processes. The latter, on his Plate 54, were numbered 11, 12, 13, 14, 16, 17, 18; they should have been 11, 12, 13, 14, 15, 16, 17, or, according to the present finding 12 to 18. He also made a small slip in stating that the pleurapophyses of the 6th and 7th vertebrae were fused into one process, for he had already described them as similar to 4 and 5, as they appear in the illustration. Assuming the small eleventh (38) to be present, and I have never found it absent, the formula for the pelvis of Owen s type of A. parvus would be the normal one, i.e. 28-34 (35-38) 39-45. A total number of fifty-six vertebrae in the moa was also given by Hutton (1894, p. 159) ; but he included twelve caudal. His grouping was: — 1-21 cervical; 22-27 thoracic; 28-44 pelvic; 45-56 caudal. The arrangement found in the skeletons I have examined is: 1-21 cervical, 22-27 thoracic, 28-45 pelvic; 46-56 caudal.
In Anomalopteryx and in the other genera, there is sometimes an exti a pi e-acetabular vertebra. It is a true vertebra, having its own neural canals as well as transverse pro¬ cesses. The condition is not due to vertebra 27 having become fused with the pelvis, for it occurs where there is the normal number of 27 free vertebrae ; moreover the form of the ribs borne by 28, 29 and 30 is characteristic enough to identify them. In the follow¬ ing formulae I have numbered the interpolated vertebra 33A ; this will give the number 34 constantly to the large transverse process which joins the acetabulum anteriorly, and 39 and 40 to the two which unite to join it posteriorly. It will also serve to distinguish the two instances (Nos. 117 and 186 mentioned above) in which the normal 35 has developed transverse processes ; in these cases, as will be seen from the formulae, there are only three vertebrae (36-38) lacking pleurapophyses.
Pelvic formula — A.
Normal A.M. 190 A.M. 117 A.M. 187
A.M. 63, 70, 151 . .
Finally, to return interest.
didiformis.
28-34 (35-38) 39-45 28-34 (35-38) 39-45 (46) 28-35 (36-38) 39-45 (46) 28-35 (36-38) 39-45 28-33A-34 (35-38) 39-45
to the axis and atlas, the following measurements may be of
|
Auck. Mus. No. |
150 |
151 |
51 |
149 |
66 |
69 |
|
Tibia length . . |
36.4 |
36.1 |
36.0 |
35.9 |
34.9 |
34.5 |
|
Atlas: height. . |
1.91 |
2.20 |
2.18 |
1.78 |
||
|
width . . |
1.81 |
2.27 |
2.14 |
1.86 |
||
|
Axis: height. . |
2.71 |
2.6 |
2.44 |
3.14 |
3.07 |
2.56 |
|
width . . |
2.57 |
2.54 |
2.50 |
2.87 |
2.76 |
2.59 |
26
These dimensions indicate once more that, although there is considerable variation in the size of vertebrae, it does not follow that of the leg-bones. No. 51, for instance which has larger legs than 66, has a considerably smaller atlas and axis, though its skull is only slightly smaller ; curiously, although the atlas of 51 is larger than that of 69, its axis is smaller, yet they fit one another perfectly in each skeleton. There are appreciable differences in the form of the vertebrae also. It may be suggested that differences sue as are recorded above are trivial and insignificant; it is to draw attention to this tha
they have been noted.
Pelvis (PI. 9, fig. 3 ; lateral view) : The dorsal margin of the ilium presents a gentle even curve which passes anteriorly without angularity into the fiont margin, this in turn curves round strongly to the ventral border, whence project the short ribs of the 29th and 30th vertebrae. Viewed from above (PI. 10, fig. 3) the dorsal iliac margins diverge posteriorly, gently at first and then, in the acetabular region, abruptly outwai s to form the anterior margin of the escutcheon. The lateral margins of the. escutcheon are convex, converging posteriorly; there is, however, a good deal of variation here. Seen from the side (PI. 9, fig. 3) the upper margins of the escutcheon continue the even curve of the pre-acetabular iliac dorsal margin, and there is no considerable flattening m this area as in Pachyornis (fig. 4) . The width of the escutcheon is usually about 37 )o o
the total length.
The ischium, narrow anteriorly, widens posteriorly; its dorsal margin terminates slightly behind the end of the escutcheon, the ventral margin terminating some 3 cm further beyond, the terminal margin thus being oblique. The pubis is a slender curved bar slightly wider distally, terminating in line with the lower margin, of the ischium. Pubis and ischium are only moderately divergent laterally in this species; the distance across their extremities is normally about 1.08 times the width at the antitrochanters.
Variation: There is no marked variation in the form or proportion of the pelvis in A didiformis In three pelves, two of immature birds, the front margins of the escutcheon diverge more gradually from the upper iliac margin, though the escutcheon itself is no less proportionately wide than in the other specimens. The leg-bones to which, these three pelves belonged exhibit the usual degree of variation among themselves ; that is to say the individuals exhibiting this pelvic variation did not in any way form a separate variant
group.
Table 3. Dimensions and proportions of pelves of A. didiformis.
Auckland Museum No.
Length
Width at antitrochanters Width at pectineal tubercles Width of escutcheon Ischium: length width divergence
Pubis: length .
width
divergence
Proportions: length — 100
Width at antitrochanters Width of escutcheon Proportion: antitrochanter width — 100 Divergence of ischia Divergence of pubes
|
128 |
117 |
70 |
71 |
51 |
149 |
152 |
66 |
|
35.5 |
35.4 |
36.9 |
36.0 |
34.35 |
35.5 |
31.15 |
|
|
16.3 |
15.54 |
16.1 |
GO bi |
14 95 |
16.5 |
13.55 |
13.9 |
|
11.35 |
12.7 |
13.2 |
11.1 |
11.9 |
|||
|
13.0 |
12.80 |
13.6 |
14.0 |
13.25 |
12.9 |
11.5 |
|
|
16.3 |
14.5 |
16.5 |
14.7 |
12.9 |
|||
|
6.4 |
5.3 |
5.5 |
5.1 |
||||
|
17.3 |
16.0 |
17.8 |
14.0 |
14.8 |
|||
|
17.5 |
17.0 |
18.5 |
|||||
|
18.0 |
16.5 |
18.85 |
|||||
|
46.0 |
43.9 |
43.6 |
51.4 |
43.5 |
46.4 |
43.5 |
|
|
36.6 |
36.1 |
36.8 |
38.8 |
38.5 |
36.3 |
36.9 |
|
|
93.5 |
107 |
108 |
103.5 |
106.5 |
|||
|
97.3 |
111.2 |
114 |
27
Sternum. The sternum is usually very badly preserved in both cave and sand-dune skeletons, and the extremities regularly give the appearance of being abiaded, so that one may doubt whether they are complete. This applies chiefly to the tips of the median and lateral processes. On the other hand they frequently exhibit so definite a trans¬ verse truncation (PI. 12, fig. 4; cf. Owen 1883a, pi. 55) as to make one hesitate to say that parts have been lost. Bearing this in mind, one may say that, whethei the body of the sternum is wider or broader, as it may be in Anomalopteryx , it is always deeply con¬ cave ventrally, and has relatively short and broad lateral and median piocesses, the latter with a notch.* The anterior margin, viewed from the front, is stiongly cuived ventrally; from above it is slightly bowed forward medianly. The pre-costal piocesses aie usually well developed, and the notches for the articulation of the scapulo-coracoid may be distinct or vague, or a notch may be discernible on one side only. In only one of the skeletons is there preserved a bone that may be the scapulo-coracoid ; its lower portion is cylindrical and its upper flat or blade-like, but it is evenly curved throughout instead of being bent : it seems too long to be a sternal rib.
There is much variation in form and relative size in the sterna of different indivi¬ duals. While, generally, the larger sterna are associated with larger leg-bones, the con¬ trary may occur, and a narrow sternum may be associated with stout leg-bones or vice versa. There is complete diversity in the size and form of the three costal articular sur¬ faces; no two sterna show the same arrangement and considerable difference may occur on either side of the same bone. The three sternal ribs, or pairs of ribs, also exhibit some diversity, but not of any great extent.
Table 4. Dimensions of five sterna: A. didiformis.
|
Auckland Museum No. |
70 |
71 |
55 |
51 |
149 |
|
|
Tibia length |
39. G |
39.3 |
37.5 |
36.0 |
35.9 |
|
|
A. |
Breadth across the pre-costal processes . . |
15.7 |
15.0 |
14.8 |
13.5 |
12.0 |
|
B. |
Breadth at anterior end of costal border . . |
11.5 |
11.9 |
11.5 |
11.0 |
10.2 |
|
C. |
Width of base of median process, i.e. between the |
|||||
|
lateral notches |
8.3 |
7.5 |
7.7 |
8.1 |
||
|
D. |
Distance across outer ends of lateral processes . . |
16.0 |
14.5 |
17.0 |
15.0 |
16.0 |
|
E. |
Length from anterior margin to tip of median |
|||||
|
process |
15.5 |
13.5 |
15.4 |
|||
|
F. |
Length from anterior margin to xiphisternal |
|||||
|
notch . . . . . . . . . . . . |
13.5 |
13.2 |
13.2 |
|||
|
G. |
Length from anterior margin to lateral notch . . |
11.0 |
8.1 |
9.0 |
9.5 |
10.2 |
|
H. |
Length from precostal to tip of lateral process . . |
15.8 |
16.0 |
15.0 |
15.6 |
Localities: North Island: Whangarei, Coromandel, Kawhia to Te Kuiti, Rotorua (Hutton), Waikaremoana, Te Aute, Lyall Bay (Hutton), Wanganui.
South Island: Collingwood, Takaka, Nelson, North Canterbury, Hamilton Swamp (Hutton), Southland (Hutton).
Summary. Anomalopteryx didiformis may be described as a moa of intermediate size with moderately stout limbs. Its skull has a long sharp beak, a strong mandible and a deep temporal fossa for attachment of the muscle which moved it. Its pelvis is slightly arched throughout its length and relatively slender, and the sternum has the lateral and
*The “sternum of Anomalopteryx ” illustrated by Oliver (1930, 45) is that of a species of Emeus
28
median processes rather shorter and broader than the other genera; its median process is notched. The species exhibits a considerable range of variation in size and proportions of leg-bones, skull and pelvis ; but variation in the different structures are so diversely associated in different individuals that no grouping either for sex or locality can be sug¬ gested. All ranges of size and varieties of form are found in one very limited area, e.g. the barrier-wall of Lake Waikaremoar.a, where there is a time-limiting factor in the comparatively recent formation of the caves which trapped them. We have, therefore, to regard these birds as having been, in the geological sense, contemporaneous.
A. didiformis occurred throughout New Zealand (?Westland) chiefly in hilly country, though it has occasionally been found on the coast, but here possibly in association with Maori cooking places.
1891
1892 May . .
1893
1893
1907
1930
1933
Anomalopteryx antiquus Hutton, 1892.
“Avian Remains”: Forbes, Trans. N.Z. Inst., vol. 23, 366.
Anomalopteryx antiquus Hutton, Trans. N.Z. Inst., 24, 124. Founded on portions oi leg-bones, of which the larger tibia fragment may be regarded as the actual type, from an Upper Miocene or Lower Pliocene deposit at Timaru.
Anomalopteryx cintiqiia: Hutton, Trans. N.Z. Inst., 25, 14.
Anomalopteryx antiquus: Forbes, Nat. Sci., 3, 318-9.
Anomalopteryx antiquus: Rothschild, Extinct Birds, 202.
Anomalopteryx antiquus: Oliver, New Zealand Birds, 46.
Anomalopteryx antiquus : Lambrecht, Handbuch der Palaeoi nithologie, 114.
The identity or relationships of this species cannot be determined with any degree of certainty. On the measurements given by Hutton (1892, p. 125) the tibia would seem to be that of Eury. curtus, a species not known to occur in the South Island. But it must be remembered that the given length (30.5 cm.) was estimated from the proximal portion of one tibia and the distal portion of another ; if the estimate were increased by only 1 cm. the length would equal that of the smallest A. didiformis and the proportion¬ ate widths of the bone would be definitely those of Anomalopteryx. Furthermore, the metatarsus subsequently described by Hutton (1893, p. 14) has all those details of con¬ figuration that have been observed to be characteristic of A. didiformis and to distinguish its metatarsus from that of E. exilis or E. curtus (see p. 58 below). It is therefore desirable to retain the species, regarding it, as suggested by Hutton (1892b, 126) as a small species of Anomalopteryx probably ancestral to A. duhfoi ims.
1884
1885
1886
1891
1892
1907
Genus Megalapteryx Haast, 1886.
Megalapteryx : Haast, Trans. N.Z. Inst., 16, 576-7. Type Megalapteryx hectori Haast, nom. nud.
Megalapteryx : Haast, Proc. Zool. Soc., No. 35, p. 541. Nom. nud.
Megalapteryx : Haast. Trans. Zool. Soc. 12, pi. 5. p. 161. Type, by monotypy, Megalapteryx hectori Haast.
Not Mesopteryx Hutton, N.Z. Jnl. Sci. new issue, 1, no. 6, p. 248. (Although “Mcsopteryx didinus” is the only species cited, it is clear, both from the measurements given and the synonymy, as well as from Hutton’s subsequent use of the name, that he was confusing D. didinus Owen with the prior D.
huttonii Owen (r= Emeus huttonii , Q-V.)
Palaeocasuarius : Forbes, Trans. N.Z. Inst. 24, 189, nom. nud.
Palaeocasuarius : Rothschild, Extinct Birds, p. 219. Type, by original designation, Pal. haast i Rothschild (ex. Pal. haasti Forbes, nom, nud.)
29
Birds of about the same height as anTles^acute" th^0 in
-y slender, antorbitals extending widely laterad.
Maxilla with well-developed antrum cavity.
Two species are recognised, M. didinus (Owen) which includes the smaller form iwo species aie recogmseu, , , secies to be named M. benhami, at
hitherto separated as M. hectori Haast, and a 1 g P M , district
present represented only by a femur and tibia from Mt. Arthur, Nelson district.
1882 .
1883 January
1884 May
1885 .
1886 December . .
1891 April
1891 November 13
1892 May 1892 May ..
1897 Juno
1897
1907
1930
1933
Megalapteryx didinus (Owen), 1883.
Diuornis didin, ,s Owen, Proc. Zool. Soc. tor 1882, no. 36, p. 549 (nomen nudum) Dinornis didinus Owen, Trans. Zool. Soc. London, vol. 11 pt. 8 P_^7, pK 59- ^ TYPE: incomplete individual skeleton from Queenstown in British Muse
(No. A. 16 ; Lydekker, p. 277).
Megalapteryx Melon Haast, Trans. N.Z. Inst., 16, 576-7 no,,,, nud.
Megalapteryx hectori Haast, Proc. Zool. Soc. tor 1885, no. 35, p. 541 (none nndunM Megalapteryx hectori : Haast, Trans. Zool. Soc. London, vol. 12, pt. 5, p. 161, pi. 30. TYPE: leg-bones of an individual from Takaka in Nelson Museum.
Anomaloptcryx didina : Lydekker, Cat. Foss. Birds Brit. Mus., 277.
Megalapteryx ter.uipes Lydekker, ibid, 251, fig. 69A. TV PE. imperfect light tibio tarsus from, near Lake Wakatipu, in British Museum (No. 49990).
Megalapteryx hectori : Lydekker, ibid, 252.
Megalapteryx hectori : Sharpe, Cat. Ost. Vertebr. Mus. Roy. Coll. Surg. London, III, p. 42S.
in, p.
70T Mesopteryx didinus : Hutton, Trans. N.Z. Inst., 24, p. 131. Hutton here
con-
Palaeocasuarius haasti : Forbes, nom. nud. Trans. N.Z. Inst., w4, 189 Palaeocasuarius velox Forbes, nom. nud., ibid, 189.
Palaeocasuarius elegans Forbes, nom. nud., ibid, 189.
Megalapteryx tenuipes : Hutton, Trans. N.Z. Inst., 29, 546.
Anomalornis gracilis : Hutton (not of Owen), part (femur), ibid, 546.
NOT Meionornis didnus : Hutton, ibid, P- 558. Hutton again uses this name for bones which are really of E. huttonii.
Anomaloptcryx tenuipes : Andrews, Novitates Zoologicae, 188.
Megalapteryx hectori : Rothschild, “Extinct Birds , 197.
Megalapteryx hamiltoni Rothschild, ibid, 197. TYPE: left femur from Waingongoro, North Island, in British Museum, No. 32145 (Lydekker, p. 252).
Megalapteryx tenuipes : Rothschild, ibid, 198.
Megalapteryx huttoni : Rothschild (part), ibid, 198. Rothschild repeats Hutton’s confusion of M. didinus with E. huttonii.
Palaeocasuarius haasti : Rothschild, ibid, 220. Founded on leg-bones from Mani- toto, in Liverpool Museum : femur indicated as type by Rothschild, p. 219.
Palaeocasuarius velox : Rothschild, ibid, 220. Founded on leg-bones from Manitoto, in Liverpool Museum : femur indicated as type by Rothschild, p. 219.
Palaeocasuarius elegans : Rothschild, ibid, 220. Founded on leg-bones from Mani¬ toto, in Liverpool Museum : femur indicated as type by Rothschild, p. 219.
Megalapteryx didinus : Oliver, “New Zealand Birds”, 42.
Megalapteryx hectori : Oliver, ibid, 43.
Megalapteryx didinus : Lambrecht, Handbuch der Palaeornithologie, 141. Megalapteryx hectori : Lambrecht, ibid, 143.
There is less material of Megalapteryx available for study than of Anomaloptcryx, but the assembly of the dimensions of fifteen individual skeletons and of certain unassoci¬ ated leg-bones (Table B) gives the same picture as that presented by A. didiformis, i.e. of a degree of continuous variation in the sizes of each of the three bones, combined with
a diversity in the association of bones of various lengths in the different individuals. As was also found to be the case in Anomalopteryx, the full range in variation is exhibited by birds from one locality, in this case the limestone plateau which extends between Mt. Arthur and Takaka, west of Nelson ; nearly as extensive a range is exhibited by the speci¬ mens from Central Otago. This range of sizes may perhaps include two sex groups, but, if so, they cannot be defined, and the largest of the smaller sex must be bigger than the smallest of the other. From these considerations I have concluded that, as with A. didiformis, only one species should be recognized.
The dimensions (Table B) reveal the slender proportions of the leg-bones of Megalapteryx didinus; they differ considerably in form also from Anomalopteryx didiformis. The femur, seen from the side, has a distinct dorsal curvature, which is only just dis¬ cernible in A. didiformis ; the proximal face is higher in proportion to its width, and, at the distal end, the rotular cavity is narrower and deeper. The muscle ridges (PI. 1, fig. 2) are finer than in Anomalopteryx, and not much raised. They converge towards the medullarterial orfice, behind which they diverge, the inner becoming rugose and merging with the inner wall of the popliteal depression, the outer remaining fine and disappear¬ ing on the outer wall of the depression. The tibia (PI. 2, fig. 2) is not only straighter, but its proximal end is less expanded ectally, and the procnemial ridge is deflected out¬ wards much less than in Anomalopteryx. In the metatarsus (PI. 3, fig. 2) the proximal face, like that of the femur, is higher in proportion to its width ; the distal trochleae, not¬ withstanding the general slenderness of the bone, are sometimes expanded nearly to the same degree as in Anomalopteryx. The length and slenderness of the toes of this species have been remarked by Andrews (1897) ; it had a well-developed hind toe.
Skull. We have four skulls with individual skeletons of M. didinus. That of the type is covered with skin, but certain measurements have been obtained; the others aie from specimens taken from Takaka in the Tring, the Dominion and the Auckland Museums. Their dimensions and proportions are set out in Table 5, together with those of the skuil on a composite skeleton, with leg-bones in plaster, labelled A. parvus at Tring. This latter agrees so exactly with the skull of the complete individual at Tring that I am suie it is of the same species.
The characters with which the four smaller of the above five skulls (the type, the Auckland Museum, and the two Tring specimens) agree with one another and differ from skulls of Anomalopteryx didiformis are as follows: the few characters in which an observation can be taken from the type are in italics.
1. Skull slightly smaller, but of relatively greater height.
Posterior aspect.
2. Swollen supraforaminal ridge and supraoccipital crest.
3. Outer margins of paroccipital processes rounded.
4. Ventral termination of paroccipitals rounded and reaching to only a short dis¬ tance below condyle.
Lateral aspect (PL 4, fig. 2) .
5. Posterior (paroccipital) margin of tympanic cavity convex, and continuing evenly, without angular break into the dorsal margin.
31
Table 5. Skulls: Megalapteryx didinus.
Takaka
D.M.
|
Length : total |
13.2 |
|
Length : paroccipital to preorbital |
8.0 4.17 |
|
Height |
|
|
Width at paroccipital processes |
5.56 |
|
Width at squamosal prominences |
6.42 |
|
Width at temporal fossae |
4.1 |
|
Width between temporal ridges |
2.3 |
|
Width at postorbital processes Width at preorbital processes Width of tympanic cavity Width of temporal fossa . Width of orbit Space between lambdoidal and temporal |
6.78 |
|
ridges Proportions : % of length |
0.0 |
|
Height . . |
31.5 |
|
Paroccipital width |
42.2 |
|
Squamosal width |
48.6 |
|
Postorbital width |
51.3 |
|
Temporal fossae width |
31.0 |
|
% Temporal ridges : fossae |
56.1 |
|
% Squamosal : postorbital |
94.5 |
|
Nelson Tring. |
On Skeleton (composite) “parvus” Tring. |
Wakatipu TYPE didinus. |
Takaka A.M. 120 |
|
12.2 |
12.2 |
11.85 |
|
|
6.64 |
|||
|
4.2 |
4.30 |
||
|
5.5 |
5.0 |
4.90 |
|
|
6.5 |
6.15 |
5.S |
5.80 |
|
4.5 |
4.41 |
4.25 |
|
|
3.6 |
3.61 |
3.50 |
|
|
7.3 |
7.08 |
6.8 |
6.75 |
|
4.7 |
4.37 |
||
|
2.0 |
1.60 |
||
|
2.6 |
2.80 |
||
|
3.2 |
2.95 |
||
|
0.2 |
0.3 |
||
|
34.42 |
36.3 |
||
|
45.2 |
41.3 |
||
|
53.2 |
47.5 |
48.8 |
|
|
59.8 |
55.7 |
56.9 |
|
|
36.85 |
38.85 |
||
|
80.0 |
82.3 |
||
|
89.0 |
85.5 |
86.0 |
6. Zygomatic process slender and acute.
Dorsal aspect (PL 5, fig. 2) .
7. Occipital area more nearly vertical, i.e. not sloping forward as in Anomalopteryx.
8. Lambdoidal ridge passes almost straight across instead of forming a double curve
as in Anomalopteryx.
9. Temporal ridges reaching back to lambdoidals, but not extending far up on to roof of skull.
10. Slight paired eminences on roof in postorbital area.
11. Greater breadth between the supra-orbital ridges.
12. Beak with narrow nasal process, the end moderately acute and the lateral margins slightly convex, not concave as in Anomalopteryx.
Ventral aspect.
18. A median swelling instead of a depression, on basipterygoid platform.
14. Basipterygoid processes small, slender, projecting obliquely forward; in Anoma¬ lopteryx they are larger, broader, and project laterally.
15. Antorbitals extending more widely laterad. (Text fig. 8a; cf. 8b, A. didiformis) .
The maxillo-nasal, present only in A.M. 120, is a fine splinter expanding slightly distally. the lacrymal foramen is contained entirely within the lacrymal in A.M. 120 and Tring specimen, as occasionally in A. didiformis.
82
Characters common to skulls of M. didinus and A. didiformis.
i. Post-temporal “fossa” a wide convex band instead of a narrow groove.
ii. Postorbital processes project backwards.
iii. Maxilla with well developed antrum cavity.
An individual skeleton from Inangahua in the Canterbury Museum has only a much abraded calvarium from which no measurements can be obtained; the only character that can be noted is that in posterior view the lateral margins of the paroccipital pro¬ cesses are strongly convex. Another incomplete skull, of a juvenile skeleton from Mt. Arthur (A.M. 118), agrees with the above-mentioned skulls in respect to features num¬ bered 1 to 9 and 13 to 15 ; the other characters could not be observed.
Text-fig. 8a. M. didinus: skull, ventral view.
In respect to the features numbered 1 to 5, and 7 to 15, the skull on the Takaka skeleton in the Dominion Museum has the condition obtaining in Anomalopteryx, and this almost complete presentation of Anomalopterygian characters makes one wonder if it really was the skull of this particular skeleton; it is certainly exceptionally large for i . Dr. W. R. B. Oliver, Director of the Dominion Museum was, however, given to understand by the person from whom it was bought that there was nothing else in the cave fiom which it was obtained ; moreover both skull and skeleton are sub-immature.
33
, . ,, a a Mdiformis with the following slight
Vertebrae. The vertebrae resem e marging of the nape vertebrae are less
difference observed in two specime . , Quadrate’ in the immediately
constricted at the middle, i.e. the outline is ‘Urgent posteriorly ; and
succeeding cervicals the zygapop yses axe ^ gpines are larger. The ventral surfaces
in the middle cervicals (10-17) the shorter and wider, and the parial
of the centra are wider, the pleurapophyses much shorter a , to
hypapophyses stand further apart on vertebrae 6 to 10, behind wn approach one another again as in A. didiformis.
p / • (VU 0 10 11 fig 2) The pelvis is smaller and much narrower than m
ZSSSSSSZ m* - i*~ - - — * - “h
divergent caudad than in Anomalopteryn.
The vertebral formula differs slightly : there are the usual seven pre-acetabular verte- ine veiieui , . • lveg v,v gve (35-39), instead of four (35-38),
brae (28-34) ; these are followed, in six pelves, oy y ,, renresented bv
acetabular vertebrae lacking transverse processes, though 39 has them lepresentea Dy
splinter processes Nos. 40 and 41 then send transverse processes to meet on the pos¬ ted surface of the acetabulum, and the remainder, 42-45, or 46 (the first caudal) in a well ossified specimen (PI. 11, fig. 2) send their processes more obliquely to the sides of the escutcheon. In the pelvis figured the splinter-like transverse process of 39 joins dis-
tally with the process of 40.
The formula is thus 28-34 (35-39) 40-45 M. didinus
instead of 28-34 (35-38) 39-45 A. didiformis.
This means that there is an extra acetabular vertebra and that 40 and 41 (instead of 39 and 40) send processes laterally to the acetabulum, and 42-45 instead of 41-45 send them obliquely to the escutcheon. This difference is not altogether constant ; I have one pelvis with the formula of A. didiformis.
Table 6. Pelvis: Megalapteryx didinus.
|
A.M. 164. |
Tring. |
A.M. 115. |
A.M. 165. |
A.M. 166. |
|
|
Length |
33.0 |
32.0 |
30.0 |
32.5 |
32.0 |
|
Width at antitrochanters |
14.6 |
14.0 |
12.9 |
15.3 |
12.3 |
|
Width at pectineal tubercles . . |
9.58 |
10.1 |
10.7 |
11.8 |
9.5 |
|
Width of escutcheon |
12.0 |
10.2 |
11.8 |
10.6 |
|
|
Ischium length |
14.5 |
12.0 |
14.3 |
12.3 |
|
|
Ischium height |
5.25 |
3.65 |
4.8 |
3.6 |
|
|
Ischium divergence |
14.5 |
13.7 |
16.7 |
11.7 |
|
|
Pubis length |
16.0 |
15.5 |
15.5 |
||
|
Pubis height |
• • |
1.5 |
|||
|
Pubis divergence . |
14.5 |
17.2 |
|||
|
Proportions : |
|||||
|
% Width at antitrochanters : length 44.2 |
43.7 |
43.0 |
47.0 |
38.1 |
|
|
Width at escutcheon : length |
36.4 |
34.0 |
36.4 |
33.2 |
|
|
Ischium divergence : width |
at |
||||
|
antitrochanters |
99.3 |
106 |
109 |
95.1 |
|
|
Pubis divergence : width at anti- |
|||||
|
trochanters |
99.3 |
133 |
34
Sternum (PI. 12, fig. 3). This may be narrow or broad: the front margin seen from above, is straight and the pre-costal processes project more laterally, or more vertically, in broad or narrow sterna respectively; the narrow sterna are more deeply convex ven- trally than the broad ones. The lateral processes are more slender, and extend further behind the median processes than in Anomalopteryx. A median notch is presen m our sterna ; a fifth lacks it, but I am not certain that the median process is entire m this case. The coracoid articular facets are deep in one Auckland Museum sternum, and also m t e Tring and Dominion Museum specimens; they are not developed in two otheis. e ace for the third sternal rib is sometimes separated from the other two. Andrews (IX J ) does not mention the scapulo-coracoid ; it is present in A.M. 120.
Distribution. M. didinus had previously been recorded only from the South Island, but leg-bones have recently been secured fr >m the Makirikiri swamp, north o anganui. It seems to have been much more restricted in range and numbers than A. didiformis, bu it was not uncommon on the Takaka tableland and, apparently, in Western Otago, was possibly a high country species.
Megalapteryx benhami n. sp.
The fer- ur and tibia on which this species is founded are from a cave on the Mt. Arthur table-land, Nelson. They were found together by the writer, hut ; as they were lying among mixed bones (though of other genera) it can only be inferred that they were of the same individual bird. They exceed in size the largest Anomalopteryx bones, are considerably larger than M. didinus. They are relatively rather stouter than t bones of M. didinus, approaching Anomalopteryx in this respect; but e presence o narrow rotular cavity and fine, clearly defined muscle ridges in the femur, an a scarcely deflected procnemial ridge in the tibia, indicates that the species is o j. ega.ap ci \ a • femur, although deficient in the middle of the shaft, better exhibits the charac¬
ters and is therefore designated the actual type. It is a
with the name of Sir William Benham, F.R.S., whose studies of individual skeleto moa have materially assisted this review.
The dimensions and proportions of the type femur, and of the tibia which probably belonged to it, are: —
|
Femur : |
29.3 = 100 |
9.2 31.4 |
3.9 13.3 |
9.3 31.7 |
12.5 42.7 |
|
Tibia : |
45.4 = 100 |
10.9 24.0 |
3.9 8.6 |
6.3 13.87 |
10.6 23.35 |
1891
Genus Pachyornis Lydekker, 1891.
Pachyor„is Lydekker, Cat. Foss. Birds Brit. Mas., 361. Type, by original designa- tion, Dinornis elephant opus Oven.
The species of Pachyornis differ from those of Anomaloptaryx and having the femur and metatarsus shorter m re ^ *°"ku° a smaller temporal fossa
£ “nd« ^ and temporal ridges
*» “ With> ^-diverging
lateral processes.
35
Four species- two, in the North Island, smaller than Anomaloptcryx didiforwis and not rplative size" is indicated by length of leg-bones as follows:—
|
P. elephantopus (Owen) |
max. min. |
|
P. pygmaeus (Hutton) |
max. min. |
|
P. mappini n. sp. |
max. min. |
|
P. oweni (Haast) |
max. min. |
|
Femur. |
Tibia. |
Metatarsus. |
|
|
32.9 |
59.7 |
25.5 |
|
|
29.3 |
45.7 |
21.0 |
|
|
• |
40.0 (estimate) |
18.0 |
|
|
• |
34.0 (estimate) |
15.6 |
|
|
19.4 |
33.2 |
15.5 |
|
|
16.1 |
27.8 |
12.5 |
|
|
14.3 |
26.4 |
11.6 |
|
|
13.5 |
22.5 |
10.1 |
1856 July 30
1858 September 28
1870 .
1875 July ..
1891 April 25
1891 .
1892 April
1892 . .
1893 . .
1894
1895 1907
'1930
1933
Pachyornis elephantopus (Owen), 1856.
Dinornis elephantopus Owen, Proc. Zool. Soc. tor 1S56, pt. 24, p. 54. Founded upon a skeleton in British Museum (Lvdekker, p. 322) made up from mixed bones of several individuals found buried in sandhills at Awamoa, Oamaiu Point. TYPE: No single bone having as yet been selected as the type, I here designate the left metatarsus of this skeleton as such: it was figured by Owen (see next citation).
Dinornis elephantopus: Owen, Trans. Zool. Soc., vol. 4, pt. 5, p. 149, pis. 43, fig. 1 (femur), 47, fig. 5 (tibia) and 44, fig. 1 (metatarsus, LECTOTYPE).
Dinornis elephantopus: Owen, ibid. PP. 159-64, pis. 46-47 (composite skeleton).
Dinornis elephantopus: Owen, Trans. Zool. Soc., vol. 7, pt. 2, p. 123, pi. 10 (skull).
Dinornis elephantopus var. major Hutton, Trans. N.Z. Inst., 7, p. 274; Table A, oppo¬ site p. 278. Founded on three femora, tibiae and six metatarsi from Hamilton Swamp, Otago, in Otago Museum. LECTOTYPE (here designated) metatarsus from the above specimens with the following dimensions: —
23.65 11.7 7.5 14.8 19.0
Pachyornis elephantopus : Lydekker, Cat. Foss. Birds Brit. Mus., 321.
Pachyornis immanis Lydekker, Cat. Foss. Birds Brit. Mus., 343. TYPE: A meta¬ tarsus from the South Island in the British Museum (A. 168).
Euryapteryx ponderosus Hutton, N.Z. Jnl., Sci., new iss., vol. 1, no. 6, 249. Aveiage measurements of leg-bones, type not indicated. LECTOTYPE here designated, metatarsus from Hamilton Swamp in Otago Museum with the following dimensions:
20.95 10.0 5.8 12.3 14.8
Pachyornis rothschildi Lydekker, Proc. Zool. Soc. for 1891, no. 33, 479-482. Types: associated right femur, and the two tibiae and metatarsi; locality unknown. Tring Museum.
Euryapteryx elephantopus: Hutton, Trans. N.Z. Inst., 24, 135 (part; Type No. 1).
Euryapteryx ponderosus : Hutton, ibid., 137, part.
Pachyornis inhabilis Hutton, Trans. N.Z. Inst., 25, 11. TYPE: Incomplete individual skeleton from unknown locality, “probably somewhere in Canterbury” in Canterbury Museum (No. 9.2.23).
Pachyornis valgus Hutton, Trans. N.Z. Inst., 25, 12. Types: A pair of tibiae from Enfield, in Canterbury Museum, but only one now identifiable.
“Pachyornis immanis?” : Parker, Trans. N.Z. Inst. 26, pp. 224, 225 (skull, probably of this species).
Pachyornis elephantopus: Parker, Trans. Zool. Soc., 13, p. 375, pi. 60, figs. 22. Skull, (pi. 8, fig. 3, of this paper).
Pachyornis immanis: Rothschild, “Extinct Birds,” 215.
Pachyornis rothschildi: Rothschild, ibid., 215.
Pachyornis inhabilis: Rothschild, ibid., 216.
Pachyornis valgus: Rothschild, ibid., 216.
Dinornis novae zealandiae: Oliver (part), “N.Z. Birds,” pp. 39-41.
Euryapteryx irnmanus: Oliver, “New Zealand Birds,” 52
Euryapteryx immanis: Lambrecht, Handbuch der Palaeornithologie, 150.
36
The following are part Euryapteryx gravis.
I860
1874 June
1874 July ..
1875
1891 November 13
1891 November . .
1892 May 1890 June
1907
1930
1933
Dinornis elephant opus: Haast, Trans. N.Z. Inst., vol. 1, 85.
Dinornis erassus Haast, Trans. N.Z. Inst., 1, 86, 87 (No. 16).
Palapteryx elephant o pus : Haast, Trans. N.Z. Inst., vol. 6, 427.
Palapteryx elepliantopus : Haast, The Ibis (3), 4, 209.
Dinornis erassus var. major Hutton, Trans. N.Z. Inst., vol. 7, 276-7, Table A, opposite p. 278. Founded on numerous leg-bones not now distinguishable from the following.
Dinornis elephant opus : Hutton, ibid., 276-7, Table A.
Dinornis erassus: Hutton (part, femora and metatarsi maxima) ibid., 276-7, Table A. Dinornis gravis: Hutton (part, maxima; includes types of Euryapteryx ponder osus Hutton) ibid., 277, Table A.
Pachyornis elephantopus : Sharpe, Cat. Ost. Vertebr., Mus. Roy. Coll. Slug., London. III., p. 436.
Euryapteryx elephantopus : Hutton, N.Z. Jnl. Sci., new issue, vol. 1, No. 6, p. 249. Euryapteryx gravis: Hutton, N.Z. Journ. Sci., new issue, vol. 1, No. 6, 249. Euryapteryx gravis: Hutton, Trans. N.Z. Inst., 24, 138.
Euryapteryx gravis: Hutton, Trans. N.Z. Inst., 28, 638, 647.
Pachyornis immanis: Hutton, Trans. N.Z. Inst. 28, 642, bones from Kapua. Ibid. p. 647 — bones from Enfield.
Euryapteryx ponder osa: Hutton, Trans. N.Z. Inst. 28, 638; bones from Kapua. Pachyornis elephantopus : Hutton, ibid., 641.
Pachyornis inhabilis: Hutton, ibid., 642.
Euryapteryx ponderosa: Hutton, ibid., 647; bones from Enfield.
Pachyornis elephantopus : Hutton, ibid., 647.
Pachyornis inhabilis: Hutton, ibid., 648.
Pachyornis elephantopus : Rothschild, Extinct Biids, 214.
Pachyornis pondcrosus : Rothschild, ibid., 216.
Euryapteryx elephantopus : Oliver, N.Z. Birds, 51.
Euryapteryx pondcrosus : Oliver, ibid., 52.
Euryapteryx elephantopus: Lambrecht, Handbuch der Palaeornithologie, 150. Euryapteryx pondcrosus : Lambrecht, ibid., 150.
This species was founded upon a skeleton made up of mixed bones, but, as Owen in his description referred especially to the size of the metatarsus (1858a, p. 55), and as it most distinctively exhibits the character of the species, it is appropriate to select it as the type In fact, the terms in which Owen discussed it (p. 58) might perhaps be regarded as a designation of it as the type. “I had hitherto regarded the metatarse of the Dinornis erassus as presenting the most extraordinary form and proportions of all the restored species of huge wingless birds of New Zealand; but it is strikingly surpassed in robustness and in great relative breadth and thickness by the same bone of the present species, which chiefly on that account I have proposed to name elephantopus”
Not many individual skeletons or sets of leg bones have been secured, and only two with the skull. On the other hand, numerous mixed bones have been recovered from swamps, from among which series might be arranged of, say, the tibia, which would seem to indicate varieties or subspecies. But the few individual skeletons we alrea<-*y have are sufficient to indicate that the femora and metatarsi would not conform to the grouping on tibiae, and that small tibiae are sometimes associated with larger femora and metatarsi than those belonging to much larger tibiae (Table C.). A considerable range of sizes is therefore included under the name elephantopus. P. immanis was separ¬ ated by Lydekker as an extremely massive form ; but if this is to be recognized it will be
known as P. major (Hutton) .
Hutton (1875) clearly established this name when he said of the discoveries in the Hamilton Swamp: “The excavations have certainly brought to light a variety of D. elephantopus , larger and more exaggerated than any yet recorded by Prof essor Owen or by Dr. Haast.” The accompanying dimensions in his Table A opposite p. 278 indicate a
37
, in the metatarsus for which
bird with the same ™ass^® m®ta^rSa J^posed the name immanis. The lectotype that Lydekker subsequently (1891, p. 3 ) 1 l bones in the Otago Museum is
I have nominated from among the Hamilton swamp bones
slightly shorter than Lydekker s type, but is q . . ,
Oliver (p. 41) included Pachyorms^HscMH m D, norms -»ova ^ ^ ^ M the length and distal width of t e 1 ia. e measurements ; their form, however
SSS.S51 who* 0h.r.ctoi,tic inflected
^ ’ • • h-Viq fvnp iprr bones of Pachyornis nihabihs and P. valgus.
tibia appears again m the type leg-bones oi i auiyu
Thi. fe.tor. provide, . mfe " b^T.ttojoi.SToogh \Z
ssirrcfii •< »• - “ -s:i
„„c, h.,.l ridg, .hove in P- S
tion of the trochlea in Eu. gravis (PI. 14, ct. tigs, la ana j ,. „ t , d
the proximal end is also usually higher in cUphantopus In the smal ler North s
species the actual inward inflexion of the tibia is as great ^ur^pte^x « “
but a strongly-developed inner flange near the extensor bridge and the higher ^eicon
dylar ridge of the metatarsus in Pachyornis form even better means of distinction f o
Euryapteryx (Text-figs. 9a, 9b) .
Fig. 9a.
Text-fig. 9. P. mappini (9a);
I do not think that a smaller variety (ponderosns) of P. elephantopiis will need to be recognized. In case it should, it may be as well to mention that the basis of ponderosus, according to the original synonymy, was bones from Hamilton Swamp described by Hut¬ ton (1875) as D. gravis. These included an associated tibia and metatarsus, the only two Hamilton Swamp bones that Hutton (p. 275) could say were those of an individual bird. These would have been appropriate for selection as the type of ponderosus; but they can¬ not now be recognized. The bone selected as lectotype is the Hamilton metatarsus in Otago Museum nearest in size to the dimensions of the above-mentioned metatarsus ; it happens to have a high intercondylar ridge and is accordingly to be included with Pachyornis (elephantopiis) rather than Euryapteryx (gravis).
Skull. In identifying large South Island skulls, particularly from among mixed swamp material, the commonest need will be to determine whether a cranium separated from its beak is of P. elephantopus or Euryapteryx gravis. Comparative details bearing on this point will therefore be as desirable as differences between the species of Pachyornis itself.
38
The description which follows is taken from the skull of the recently-found Pyramid Valley specimen, xxB, in the Canterbury Museum; the figures (Plate 8), however, which were drawn and processed some time ago, are from other sources. The only other skull of this species definitely found with its skeleton is that of the type of P achy arms inhabihs, also in the Canterbury Museum.
In occipital view the skull has a well-arched roof; the supra-foraminal ridge not being swollen, the supra-occipital median crest and the supra-occipital fossae remain well- defined. In these respects P. elephant opus resembles Eu. gravis and differs from the small North Island species of its own genus. The paroccipital depressions are wide and shal¬ low. The outer border of the paroccipitals is sinuate ; the processes extend doWn to about half way between the levels of the condyle and the mamillar tuberosities ; the latter are large, separated by a distinct arch, and are rather more outstanding than in Eu. gravis.
In dorsal view the lambdoidal and temporal ridges are separated, and the anterior and posterior lambdoidal ridges enclose a flat triangular area. The temporal ridges encroach upon the occiput and curve backward in a characteristic sweep towards the lambdoida ridges (PL 8, fig. 1), while in Eu. gravis (PI. 6, fig. 3) they extend upward rather than backward, and to a much less extent. The temporal area of the roof is flattened m the Pyramid Valley skull, but is arched or has a double tumidity in other specimens.
In lateral view the posterior, paroccipital, border of the tympanic fossa is convex and curves into the upper border (PI. 8, fig. 2), whereas in Eu. gravis (PI. 6, fig. 1) the hinder margin is less curved and meets the upper at less than a right angle; m both species the upper border of the fossa is a lobed overhanging ledge. The postorbital process is moder¬ ately wide, in other skulls it is narrower; in Eu.gravis it is usually very Wide. The long acute premaxilla has a large median septum which has a thin semi-transparent central
area; in Eu. grains the septum is short and opaque.
In general the skulls of P. elephantopus exhibit their differences from Eu grains m the pre-lambdoidal area; in the occipital area the resemblances are more marked than the differences. In the small North Island species of Pachyorms, however, the occipital region resembles that in Emeus and differs markedly from the small Euryapteryx species which have the occipital characters of the large species.
Sternum and Pelvis. The sternum in three individual skeletons is broad with widely diverging lateral processes. The pelvis also is broad, but does not exhibit such mar e differences from Eu. gravis as appear between the small North Island species o ac lyorms
and Euryapteryx.
Distribution. P. elephantopus is known so far from Canterbury and Otago only. _ The heaviest forms (i.e. the types of major and immemis) were from °tago I these are peculiar in that they appear to have acquired a second coating of bone as though y
an overgrowth.
1891 November
|
1892 |
May . . |
|
1895 |
• • • • |
|
1897 |
June |
|
1907 |
• • • • |
|
1930 |
• • • • |
|
1933 |
• • • • |
Pcichyornis py£ma.eus (Hutton), 1891.
irvaPterw pygmaeus Hutton, N.Z. Jnl. Sci., new issue, vol. 1, No. 6, p. 249. Founded of average measurement, of leg-bones, of Which a pair of metatarsi from Takalta in Nelson Museum were subsequently (Hutton 1892b, 139) selected
as the types.
uryapteryx pygmaeus: Hutton, Tians. N.Z. Inst., , , 61 figs 20 21
•csoMerv » species 0 : Parker, Trans. Zool. Soc. 13, pt. 11, p. 378, pi. 61, tigs. 20, 21.
aZnus pygmaeus: Hutton, Trans. N.Z Inst. 29 555.
aehyornis pygmaeus: Rothschild, “Extinct Birds 217
,, . j,.. c. Oliver, “New Zealand Birds, 53.
uryapteryx p g • Lambrecht Handbuch der Palaeornithologie, 152.
uryapteryx pygmaeus: naniDrecui,
39
Table 7. Dimension of skulls of Pachyomis elephant opus , P. pygmaeus; Eurapteryx gravis.
c
°3
Eb
§ ’ds snouig
» ‘ds Mtooq sliding ^
'H Sbus ‘P^I ,J ■(is sn3luU ^
•adAj, 'suzdaB 'g
•muvunq ‘g ladXj,^.
O
(M9>I.n3j L 'ds H
sliding) judyuvcf sliding ^
•U019I931S I'BnpiAipni <s\ ■snsoudpuo4 'g TtO^iaATy; o
'ICO -luooe -s 'H dd smis J
1'0 -mooa -s H Id Smis ^
CO ■<* O t- D3 o 05 o o ©
CO
LD
ca H ^ M ® ^ 'I ^
do6o6a5't^t'T}1
esC5C<l-^OOr-J'^00 o
^MNlNMOOindHHCOM L M lO M C OO H ®
C^*
CO
co
co oo
co co
CO 00
o (D 00 N ^ W lo id t- ld H °°
ld h< _ „
N H O O N
tO t- LD H 00
LO ID ID
tSWNMfflHNCi '^501O'i('c000Ti,H
TH<T30t-OOCOLDOC01D CO LD ID CO t— I 00 rH l 03 ® 00
CO lfl CO ® O rl ffl t-
t-t-uocoopt^oO;^
CO t— CO
co oo as
t~. 03 t— H< O LD OO N ffl
rjitOt-lC'^OOCOClNN®®1^
CO CD CO
t— liHi-ICOTHCOOOCDOt^OO
U5 O t-
00 03 ID CO 03 © CD
|
CO |
00 |
tH |
oo |
uo |
|||||
|
00 |
<cq |
oo |
CD |
00 |
CO |
ID |
rH |
LD |
|
|
CO |
00 |
ai |
O |
n |
CD |
t- |
LO |
H |
00 |
|
tH |
rH |
||||||||
|
lO |
LD |
||||||||
|
o |
©q |
CO |
o |
O |
cq |
O |
o |
||
|
00 |
05 |
o |
ID |
00 |
LD |
oi |
|||
|
rH |
rH |
O rH t— I O rf t—
LD ID t- r- I ID 03 OO
CO ID CO CD r-l t— I ID
ID 03
C\J CD O t>“ LD tH CO O ID CO
NCqCijlDOOffl^t-fflO CO ID OO CD T— I O CD
OS
c-
Ol
OS
to
?A X
‘dT'iL •£/ ‘ds XiUdtfosdft
tH tH
H 00
C-- <35 CO OO T-t O CO tH CO 03 CO p O 00. CD ID OS LD
^idi>^^®ifiN'NMMeonieooo«!®oio
CO ID CO co <M r-l ID
00
O
H~.
•S»
■oa
’OX Id
‘qOL8I ‘U9A\0 •pu'Bisi ’S
•X68X ‘.i9q^d •PPUUH
'IM O lX9Id T ■pieun[j^
O
ID U5
CO as
CO 00
tH as
|
ID |
ID |
LD |
t>- |
03 |
CO |
o |
tr-; |
H |
||
|
ID |
CD |
OO |
ID |
LD |
os |
03 |
03 |
CO CO |
||
|
tH |
CO |
CD |
p |
o |
to |
|||||
|
CD |
tH |
ci |
LD |
CO |
CO |
|||||
|
'H OO |
CA |
oq (M |
03 CD |
CO o |
o CD |
t> |
H |
CO p |
H |
H |
|
ID |
CD |
ci |
LD |
H |
as |
CD |
03 |
03 |
CO |
o CO |
ta CD LD H Tj! CO
O ID t- N LD
as
o
r~
ID
as
K*“i
rO
m
-j— »
<D
a
ao
m
aj
a>
l=H
. CO
. .
IS)
'Zj
& *£
h-> 5-1
. be a)
. P< •Ei 03
0 co co _£} •— i 5-1 c«
p -up o o
• • • J® o • co . be • *1-1 TO ^ C3
•^c! * • ® Cl • 0 CO . rj . fe »|H O •“ 3
aja co hn0M W ^>0003
-§ g § “ | g
. | ts a . s s ^ s ^ . g •■ g S 1 ** | g
• e ? S - sp|2S|?g • • 1 1 1 •& § 3 i I
sis «“ls“Qgo •■ ~ - § |i| ..
• -rH >J • . • H J Cd 0 ,H -P ,H CIS • tn rH Curri Q W f_2 CD
Oh $_ ,H O L +J ,1] -H d £
■3 » o 8 a ^ A 3 O rt r ” S ..ft ^
c3^c'303c^'ti|l>
o S - § ’Sgsi a^ss-g-sggsi
«a<ai<0^§S»2§oga
M H hi, S ^ 5 ^ ^ ^ ^ ^ ^ ^ S 0 o o* 0 C3 0)
a a a a-r'bS’S'S'a'a'a'a'a'SWt/sE-'PHKiP-ipH ^
0 S 0 0 <p r1 £ r1 r* r" r1 r1 ir a.
40
% temporal
Having founded this species on average measurements, Hutton subsequently designated a pair of metatarsi in the Nelson Museum from Takaka as the types.
Dimensions of types: —
|
Right |
15.5 |
6.8 |
3.9 |
8.3 |
10.0 |
|
= 100 |
43.7 |
25.4 |
53.3 |
64.3 |
|
|
Left |
15.6 |
6.8 |
3.9 |
8.4 |
10.0 |
|
= 100 |
43.6 |
25.3 |
53.8 |
64.2 |
These bones have the raised intercondylar ridge characteristic of P achy onus and repre¬ sent a South Island species much smaller than' P. elephant opus. They are of the same length as the metatarsi of the type of the next species, P. mappini of the North Island, but are much heavier. Similarly a larger metatarsus in the Canterbury Museum from an unknown (South Island) s'wamp locality, the only other leg-bone I can confidently identify with this species, reveals a stout heavily built bird (see Table C) . I have also included in P. pygmaeus the skull which Parker (1895b, pi. 60, figs. 20, 21) designated Mesopteryx sp./k It is considerably smaller than other South Island Pachyornis skulls, and, by comparison with the proportion “tibia length: cranium length” noted in individual skeletons of the genus, would have had a tibia about 39 cm. and a metatarsus 18 or 19 cm. in length. In other words, it is appropriate in size to the larger metatarsus just described. The skull is also from the same locality as the types of pygmaeus, the Takaka tableland west of Nelson city. The leg bones from Karamu, near Hamilton (North Island) identified as P. pygmaeus by the writer in 1927 belong to the next species.
Pachyornis mappini n. sp.
C-ela geranoides: (Lydekker) ; Archey, Trans. N.Z. Inst. 58, 151, pis. 18, 19. Pachyornis pygmaeus: Archey, Trans. N.Z. Inst., 58, pts. 1 and 2, p. 156.
references include part Eury apteryx exilis.
Cel a geranoides: Hutton, N.Z. Journ. Sci. new issue, vol. 1, No. 6, 248.
Cela geranoides: Hutton, Trans. N.Z. Inst., vol. 24, 126.
Cela geranoides : Rothschild, Extinct Biids, 206.
Emeus exilis: Oliver, New Zealand Birds, 49.
Emeus exilis: Lambrecht, Handbuch der Palaeornithologie, 148.
The type of this species is an almost complete skeleton discovered in 1933 by Mi. 1. Crossley Mappin in a cave at Mangaotaki. It is No. 124 in the Auckland Museum^ collec¬ tion. It has been a valuable specimen, for it has been the chief means of confirming for the smaller North Island specimens of Pachyornis the essential characters of the genus. I have much pleasure in associating Mr. Mappin’s name with the species in acknow^e g- ment of the many opportunities he has provided for exploring likely “moa country, and of his own keenness in search of specimens and patient care in recovering them rom
caves.
The type, which is the largest North Island specimen 'we have of the genus, was a bird of about the same height as the type of P. pygmaeus, but much lighter ; the Karamu specimen (Archey, 1927) was more robust, but it also was not as heavy as P. pygmaeus. As has already been mentioned, this species, and the still smaller P. <m>em, do not differ from their size-fellows in Euryapteryx by displaying greater inward inflexion of the lower end of the tibia; instead, they have a projecting flange, not found in Euryapteryx, oppo¬ site the osseous bridge for the extensor tendon (Text-figs 9a 9b). The higher proximal intercondylar ridge of the metatarsus in Pachyornis is a better-developed diffeience in these small North Island species than in the larger South Island birds.
41
1927 August 15 . . 1927 .
The following
1891 November . .
1892 May
1907 November 12
1930 .
1933 . . • • • •
The type, a.so two smaller skeletons “»i£™
excellently preserved skulls. Typica ' ; where they exhibit a distinct
teryx sp. fT (i.e. P. except m the occiptal reg ^ ^ Qf the skulls of
swelling of the supra-foramma margi • ’ , tpe sparp beak widening more
The, di«.r from the ** «< ZZ-i „Jl» (PL 7. 3, ,nd
. ,WM
Euryapteryx exilis. Pachyornis mappini.
Text-fig. 10. Eu. exilis (10a); P. mappini (10b): beak and maxillary antrum.
the type and the Amodeo Bay skull exhibit the same backward sweep of the temporal ridge as in “Mesopteryx sp. (3,” but this is not so marked in the smaller skulls. A.M. 84 and 85. The maxillo-nasal is a long slender splinter-bone lying close-pressed to the front margin of the antorbital (PI. 4, fig. 4), and expanding below where it fits against the maxilla. Comparing skeletons of equal size in Pachyornis mappini and Euryapteryx exilis , the skulls of the former are larger (Table 8) .
The sternum (PI. 12, fig. 5) has a broad straight anterior margin and widely diverging lateral processes. In all five skeletons I have the coracoid pits are developed (faintly in A.M. 150) and in four of them there is a small scapulo-coracoid ; A.M. 150 from which it is missing is a very incomplete skeleton.
Pelvis. The pelvis in P. mappini is considerably flatter dorso-ventrally and relatively wider than in Anomalop teryx, and has widely diverging ischia and pubes. Viewed from the side (PI. 9, fig. 4) the dorsal margin is very slightly arched ; the front margin which emerges from the dorsal margin by an unbroken curve forms a sharp angle with the ventral iliac border. Seen from above (PI. 10, fig. 4) the anterior portions of the ilia show wide concave expansions of subquadrate outline. Immediately behind the pro¬ jecting ribs belonging to vertebrae 29 and 30, the margins are constricted, forming a narrow waist. The acetabular region is wide and the escutcheon is also wide, with its sub-parallel lateral margins. The proportionate width at the antitrochanters is on the aver-
42
age slightly greater than in Eu. exilis or Anomalopteryx ; the proportionate width of the escutcheon is definitely greater, as also are the posterior divergences of the ischia and pubes.
The dorsal view of the escutcheon in P. mappini differs from Anomalopteryx in that while in the latter its front margin is formed by the abruptly diverging iliac crests and the lateral margins extend widest anteriorly, in P. mappini the iliac margins diverge more gradually and the escutcheon is widest further aft. This greater width posteriorly in the ilium of P. mappini is seen more markedly on the ventral side (PI. 11, fig. 4).
Table 8. Dimensions of skulls, Pachyornis mappini and Euryapteryx
it exuis.
Auckland Museum No.
Total length
Length : parocc. to preorbital Height
Width at parocc. processes Width at squam. prominences Width at temporal fossae Width between temp, ridges Width at postorbitals . .
Width at pre-orbitals . .
Width of tympanic cavity Width of temp, fossa . .
Width of orbit % squam. width : length % width at temp. foss. : length % postorbital width : length . .
% temp, ridges width : fossae width % parocc.-preorb. length : total length
|
Pachyornis mappini. |
Euryapteryx exilis. |
Pachyornis mappini. |
Euryapteryx exilis. |
|
124 |
3 |
85 |
360 |
|
12.00 |
10.70 |
11.2 |
10.70* |
|
7.00 |
6.15 |
6.35 |
5.90 |
|
4.00 |
3.67 |
4.20 |
3.90 |
|
5.30 |
5.06 |
5.20 |
5.15 |
|
6.30 |
5.60 |
6.08 |
5.65 |
|
4.56 |
3.90 |
4.30 |
3.80 |
|
3.86 |
3.26 |
3.50 |
3.15 |
|
6.77 |
6.28 |
6.90 |
6.20 |
|
3.56 |
3.24 |
3.80 |
3.20 |
|
1.75 |
1.72 |
1.90 |
1.65 |
|
2.65 |
1.78 |
2.35 |
1.85 |
|
2.75 |
2.54 |
2.40 |
2.70 |
|
52.5 |
52.3 |
54.3 |
52.8 |
|
38.0 |
36.4 |
38.4 |
35.5 |
|
56.4 |
58.7 |
61.6 |
58.0 |
|
84.6 |
83.5 |
81.4 |
83.0 |
|
58.3 |
57.5 |
56.7 |
58.0* |
Table 9. Dimensions of Pelvis : Pachyornis mappini and Euryapteryx exilis.
Auckland Museum No
Length
Width at antitrochanters Width at pectineal tubercles Width of escutcheon . .
Ischium length Ischium height Ischium divergence Proportion % — •
Width at antitrochanters : length Width of escutcheon : length Ischium divergence: antitrochanter width
hi
|
124 |
3 |
160 |
|
31.9 |
31.0 |
29.0 |
|
15.9 |
15.2 |
15.0 |
|
12.7 |
11.3 |
• |
|
14.4 |
13.0 |
11.1 |
|
16.2 |
16.0 |
15.5 |
|
5.3 |
4.0 |
3.5 |
|
22.0 |
20.2 |
18.0 |
|
49.2 |
# |
51.7 |
|
45.1 |
. |
• |
|
138 |
132 |
120 |
43
Table 10. Dimensions of Sternum: P. mappini and Eu. cxilis.
Auckland Museum No.
A.
B.
C. B.
E.
F.
G.
Breadth across pre-costal processes . . Breadth at anterior end of costal border Width at base of median process Distance between outer ends of lateral processes
Length: anterior margin to tip of median process
Length: anterior margin to lateral notch Length: anterior margin to tip of lateral process
|
P. mappini. |
Eu. cxilis. |
P. mappini. |
Eu. cxilis. |
P. mappini. |
Eu. cxilis. |
|
Type. 124 |
Type. Wang. M. |
150?. |
3 |
84 |
5 |
|
14.5 |
13.9 |
14.5 |
13.5 |
11.3 |
10.0 |
|
12.5 |
10.2 |
11.0 |
11.0 |
9.5 |
7.8 |
|
9.5 |
6.8 |
7.5 |
6.5 |
7.0 |
5.0 |
|
21.0 |
12.8 |
19.0 |
15.5 |
18.5 |
14.0 |
|
14.7 |
14.8 |
11.5 |
13.5 |
# |
11.2 |
|
7.6 |
• |
5.2 |
8.0 |
5.4 |
7.0 |
|
14.8 |
7.8 |
13.0 |
16.0 |
13.0 |
14.5 |
Distribution: North Island generally. Skeletons have been obtained from sand dunes at Doubtless Bay and Coromandel, from caves at Mangaotaki and Waikaremoana, and leg-bones and a skull from the Makirikiri swamp deposit near Wanganui.
Pachyornis oweni (Haast), 1886.
Dinornis ozveni Haast, Proc. Zool. Soc. for 1S85, no. 31, 482. Nomcn nudum.
Dinornis oweni Haast, Trans. Zool. Soc. 12, pt. 5, 171-182, pis. 31-32. Type: by original designation, incomplete skeleton from near Whangarei; in Auckland Museum, A.M. 384.
The following references include part Euryapteryx curt us.
1891 April 25
1891 November 13
1891 November . .
1892 May
1893 .
1897 .
1907 .
1930 .
1933 .
Anomalopteryx oweni: Lydekker, Cat. Foss. Birds, Brit. Mus. 280.
Anomaloptcryx curta: Lydekker, ibid, 281.
Anomalopteryx curta: Sharpe, Cat. Ost. Vertebr. Mus. Roy. Coll. Surg. London, III., 433.
Ccla curt us: Hutton, N.Z. Journ. Sci., new issue, vol. 1, no. 6, 248.
Ccla curtus: Hutton, Trans. N.Z. Inst., 24, 127.
Anomaloptcryx curta: Parker, Trans. N.Z. Inst. 25, 2.
Anomalornis ozveni: Hutton, Trans. N.Z. Inst., 29, 549.
Ccla curtus: Hutton, ibid, 550.
Ccla curtus: Rothschild, Extinct Birds, 205.
Ccla ozveni: Rothschild, ibid, 206.
Anomaloptcryx curtus: Oliver, New Zealand Birds, 46.
Anomaloptcryx ozveni: Oliver, ibid, 46.
Anomaloptcryx curtus: Lambrecht, Handbuch der Palaeornithologie, 145. Anomaloptcryx ozveni: Lambrecht, ibid, 145.
Of this species, the smallest of the moas, I have only three incomplete skeletons, including the type, and some sets of leg-bones (Table D). The localities are Pataua, near Whangarei; Tom Bowling and Doubtless Bays, in North Auckland; sand-dunes at Wai- kawau, Coromandel; and Westmere, a suburb of Auckland. They include quite slender bones, and one, a tibia from unkncAvn locality, that is exceptionally stout. P. ozveni is the size-fellow of Euryapteryx curtus, but is rather smaller.
44
Genus Emeus Reichenbach, 1852 *
1S52
1852
1874 June
1874 July . . 1891
Emeus Reichenbach, Av. Syst. Vo'g., p. xxx. Type, by monotypy, Dinornis crassus Owen.
Syornis Reichenbach, ibid. P- xxx. Type, Dinornis casuarinus Owen.
Meionornis Haast, Trans. N.Z. Inst., vol. 6, 426. Type, here selected, Dinornis casuarinus Owen.
Meionornis Haast, Ibis (3), 4, 212.
Mesopteryx Hutton, N.Z. Jnl. Sci. new issue, vol. 1, no. 6, p. 248. Type, by mono¬ typy, Dinornis huttonii Owen (referred to by Hutton in error as D. didinus Owen).
The limb bones of the species of Emeus are stouter than those of Anomplopteryx, but less stout than in Euryapteryx. There are only four phalanges, including the claw, in the outer toe. The skull has a smaller temporal fossa, the temporal ridges reaching neither back to the lambdoidal ridge nor far up on the roof. The beak is less narrow than in Anomalopteryx, and rounded at the tip (Text-fig. 11b). The pelvis is flat with widely diverging ischia and pubes.
Text-fig. 11.
Euryapteryx exilis.
Two species: separated as to length of leg bones as follows
|
Femur. |
Tibia. |
Metatarsus. |
||
|
E. crassus (Owen) |
maximum minimum |
29.4 26.5 |
52.0 43.7 |
24 20.0 |
|
E. huttonii (Owen) |
maximum minimum |
24.4 22.0 |
39.7 35.8 |
18.7 16.3 |
*The following prior generic names approach g nousm. As i read the rules,
Eumaea, Geyer 1834, Lep.; Eumaeus , Koch ^ 1843 • slmyllarity; if however Article 35 be held
only specific names are to be rejected . for ’ 1852.
to apply to generic names Emeus Reichenbach will De repiaceu uy oyotms
45
Emeus crassus (Owen), 1846.
184C July
1846 December 28
1848 April 22 1850 or 1852
1869 May . .
1870 January 1875
1891 April 25 1891 November 13
1891 November . .
1892 May
1895 October
1896 June
1897 June
1906 June
1907 .
1930
1933
. iQ/ifi nt 14 n 46. Founded on a Dinornis crassus Owen, Proc. Zoo . °c. ° ^ Iglan’L LECTOTYPE: meta¬ femur and metatarsus from ai the present place of deposition
tarsus, nominated by Lydek er . ■ and the Auckland Museum
(A.M. 298). .... ,7 Founded on two femora, a tibia and a meta-
Dinornis casuarmus Owen Y, . tibia, nominated by Lydekker (1891a,
tarsus from Waikouaiti. LE M „ p»ov c0ll Surgeons,
257); its present plaee ol deposition is unknown, Mus. Roy. Loll.
or British Museum. ooo i *n ^o- 2 (tibia:
• . Owen Trans Zool. Soc. vol. 3, pt. 4, p. 322, pi. 4/, ng. ^ O-iuxi.
Dinornis casuarmus: Owen, irans . auu ptatarsus).
LECTOTYPE), pi. 46 (femur), pi. 48, fig. 3 ( t?pt0TYPF
^ „ ooe r>i 48 ties 4 and 5 (metatarsus, LECTOTYPE,
Dinornis crassus: Owen, ibid, P- 325, pi. 48, figs.
figure reversed). 5> p. 357 (Skull of Aptornis).
Non Dinornis casuarmus: Owen, nans,
Emeus crassus: Reichenbach, Nat. Syst. Vogel, p. xxx.
Syornis casuarinus : Reichenbaeh, ibid, P- xxx.
Dinornis casuarinus: Haast, Trans. N.Z. Inst., 1, 82, no. 1 no. 2 (2nd edn., 1875, Dinornis rheides: Owen, Trans. Zool. Soc. 7, pt. 2, 132, pi. 12 (s ).
Dinornis casuarinus: Hutton, Trans. N.Z. Inst 7, 275, skeleton
Dinornis crassus: Hutton (part: femur mm., tibia max. and min., metatarsus ..
Trans. N.Z. Inst. 7, Table A, opp. p. 278.
, • t vrfp.Wpr Cat Foss Birds Brit. Mus. 25 1.
Anomaloptcryx casuarma: LydeKKer, e<u. rub&.
Fmcui crassus: Lydekker, ibid, 307.
. • sharne Cat Ost Vertebr. Mus. Roy. Coll. Surg. London,
Anomaloptcryx casuarma: bfiarpe, Ear. usu
Emeus crassus: Sharpe, ibid, 435.
Syornis crassus: Hutton, N.Z. Journ. Sci„ new issue, vol. 1, no. 6, 249. Syornis casuarinus: Hutton, ibid, 249.
Syornis crassus: Hutton, Trans. N.Z. Inst. 24, 132.
Syornis casuarmus : Hutton, ibid, 133.
Mesopteryx casuarma: Parker, Trans. Zool. Soc. vol. 13, pi. 11. Skull Meionornis casuarinus: Hutton, Trans. N.Z. Inst., 28, pp. 636, 646. Euryapteryx crassus: Hutton, ibid, PP. 638, 647.
Meionornis casuarinus : Hutton, Trans. N.Z. Inst., 29, 558.
Emeus crassus: Hutton, Trans. N.Z. Inst., 38, 66.
(?) Ccla rheides: Rothschild, Extinct Birds, 207.
Ccla casuarinus : Rothschild, ibid, 207.
Emeus crassus: Rothschild, ibid, 209.
Emeus crassus: Oliver, N.Z. Birds, 48.
Emeus casuarinus: Oliver, ibid, 48.
Emeus crassus: Lambrecht, Handbuch der Palaeornithologie, 146. Emeus casuarinus : Lambrecht, ibid, 147.
The priority of the name crassus over casuarinus is established by page-precedence. In the Proceedings of the Zoological Society for 1846 the name Din. crassus appears on p. 46 without description. On p. 47 the name Din. casuarinus appears with mention only of a metatarsus having a feeble depression indicating a back toe ; but in the next paragraph we find crassus again, this time supported by a statement as to the size and proportions of a metatarsus relative to that of the ostrich. If this be not considered a sufficient description, the precedence of crassus is further established in the Table of Admeasure¬ ments which extends across pages 48 and 49; here dimensions of the femur of Din. crassus are given on p. 48, and of the metatarsus on p. 49, both being in tabular precedence to the dimensions of Din. casuarinus, which are also given on p. 49. The first designations of single bones as the type of each species were made by Lydekker (1891, p. 257, p. 307) and, although his selections of the metatarsus for crassus and the tibia for casuarmus do not facilitate comparison, they must be accepted.
46
Until last year no complete skeleton of Emeus crassus had been secured ; the leg bones were known, but the only fairly complete skeletons obtained had lacked a skull. Conse¬ quently certain comparisons that have been made of skulls with the “skull of crassus ” or the “skull of casuarinus” have been valueless. Through the gratifying results of the Pyramid Valley swamp excavations by the Canterbury Museum excursions I am now able to provide details from twelve individual skeletons of this species, six of them with well- preserved skulls.
The dimensions recorded (Table F) present us with a repetition of what was observed in A. did if or in is and other species; i.e. a considerable range of size-variation in skele¬ tons from one locality, together with an indiscriminate association, as to relative length, in the three leg bones of different individuals. The measurements of the designated types of crassus and casuarinus stand close together in this series, so Lydekker’s selection of the metatarsus as the type of the former and the tibia for the latter has at least resulted in the two species being merged. If Owen’s cotype femur of casuarinus 'had been fixed as the type, as Oliver regarded it, a smaller slender species, or subspecies might have been admitted, though the femur is not a good means of establishing a species. Oliver (1930) considered the two species as being only doubtfully separate, and I think that the recorded dimensions indicate the propriety of uniting them.
Lydekker (p. 257) considered casuarinus to be close to A. didiformis; it is true that the largest leg bones of didiformis approach the smallest of casuarinus in length : in relative width, however, especially of the metatarsus, and, as will be seen, m the form of the skull the generic differences between these two forms are clearly maintained. Hutton and others have experienced difficulty in separating bones of Em. crassus from those of “elephantopus” from mixed swamp material ; but, as will be seen below, this difficulty
lies rather between Em. crassus and Eu. gravis.
There is a degree of variation in the pelves of these specimens, and considerable diversity in their sterna; but the skulls are fairly uniform, all having a narrow beak
with well-rounded tip.
Skulls of Pyramid Valley skeletons: Canterbury Museum VIII B, VIII C, IX A (Text fig. 12), X B, XIII D, XIII E. (Table 13.)
Except for small details to be noted, all six skulls have the same general, form. In occipital view the paroccipital processes have a sinuous outer border and terminate below in a rounded point at about the level of, or just short of, the .mam, liar 1 t^0B,^e is a latter are small eminences 1.64 to 1.78 cm. apart, separated by ^a low -arch. Theie distinct depression or pit on each side, above and mesad of the paroccipi al processes nvvt-ficr 12hl The upper margin of the foramen, which is swollen m VIII C, X il, ana XIII D extends obliquely outwards to form a ridge across the paroccipital processes: m X i the paroccipital portion is tom -
S I! Will £ anpraforamin,.. W
the skull in the post-frontal region, and lesser eminences m the pre-frontal region are a feature of all six skulls The lambdoidal and temporal ridges are separated by from fifto 1 08 cm and the narrow space between them extends outwards to form a flat space above the’ squamosal prominences. The beak is long and tapering, with a narrow rounded tip; the lateral margins are slightly convex.
T , , , ■ nw fi„ 12a) the crestal tumidity is marked. The hinder (parocci¬
pital) marg noTthe tympanic cavity is gently curved or almost straight. The temporal S irX^tly widUan
«fo^S^ ^ front of the tympanic cavity lying between the inferior
47
Text-fig. 12. E. crassus : skull (C.M. ix A.). Text-fig. 13. E. crassus : tracheal loop.
a, lateral view; b, occipital view.
and posterior temporal ridges) is a narrow groove as in the other species of Emeus and in Euryapteryx. In Anomalopteryx and Megalapteryx it is a broad space sometimes even slightly convex. The zygomatic process is long and narrow. The postorbital process extends outwards and bends straight down, or only slightly backward or forward : it ends in a blunt tip: in VIII C the tip (which seems to be a separate centre of ossification) is missing, and the resulting broader abbreviated termination is notched. The front margin of the process is sinuous and the upper margin of the orbit is curved ; the hinder and upper margins of the orbit do not form an angle.
No maxillo-nasal process was found in any of these skulls, but, as the vomer, palatine and maxillo- jugal bones were always found separate, the maxillo-nasals may have been lost in the peat. The lachrymal foramen is an open notch: probably the missing maxillo-nasal closed it. The premaxilla is long, tapering and acute in lateral view, though more rounded and with slightly convex sides when seen from above: it is, how¬ ever, neither so narrow nor so acute at the tip as in Anomalopteryx. The maxillo-pala- tine has been secured in only two of these skulls, in both cases with an expanded antrum opening posteriorly by a circular passage ; the same condition obtains in a skull, obviously of this species, from Enfield in the Dominion Museum. The cavity is, however, not so expanded as in the skulls of Pachyornis map pint and P. oweni, nor is the opening so wide. The mandible is slightly depressed distally and the tip is rounded. In ventral view the eustachian tube is wide, deep and straight, though not so wide as in Euryapteryx.
In general the skull is relatively low in proportion to its width ; the temporal fossae are small, with limited surface for muscular attachment. The mandible is correspond¬ ingly slender.
Vertebrae: (C.M. xiii D. $ with egg) . The axis is similar to that of A. didiformis except that the hypapophysis is thicker and only its lower edge is keel-like. Nape vertebrae: Nos. 3 to 6 are subquadrate in dorsal view, with bifid neural spine and well-developed
48
hyperapophyses. Cervical: No. 6 is of the normal form of the cervicals with narrower posteriorly-diverging post-zygapophyses. The neural spines are elongate on these cer¬ vicals, but do not extend on to the post-zygapophyses ; hyperapophyses are not developed. The neural spines approach one another again in the middle cervicals, increasing in height, and on 21 there is a high bifid spine; 22 to 24 are missing.
Table 13. Skulls : E. crassus.
Total length
Length: parocc. to pre-orbital Length : basi-rostrum Length: condyle-rostrum Height
Width at parocc. processes Width at squam. prom. . .
Width at temp, fossae . .
Width between temp, ridges Width at post-orbitals Width at pre-orbitals Width o f tymp. cav.
Width of temp, fossa Width of orbit % Height: length % Width at squam. prom.: length % Width at temp, foss.: length % Width at post-orbitals: length % Width between temp, ridges: fossae % Width at squam. prom.: parocc. width % Width at post-orbitals width
Distance between lambd ridges
% length parocc. -preorb
squam. prom, and temporal : total length
|
05 |
05 |
05 |
05 |
05 |
05 |
|
|
CO |
CO |
CO |
CO |
CO |
CO |
|
|
^ K • -o* S- © |
05 . th |
Sp |
So |
Sh |
05 rH |
Sw •»— f |
|
2 S-* |
S * |
§ :p |
• •rH s ► |
s'- |
• »rH |
|
|
§ 3(2 |
b |
b |
b |
b |
b |
b |
|
13.2 |
13.7 |
13.6 |
13.4 |
13.0 |
13.0 |
12.0 |
|
7.66 |
7.37 |
7.48 |
7.35 |
7.3 |
7.08 |
|
|
- . |
8.5 |
8.75 |
9.3 |
8.57 |
8.13 |
|
|
— |
9.52 |
9.73 |
10.1 |
9.44 |
9.03 |
|
|
4. 1-4. 3 |
4.51 |
4.5 |
4.54 |
4.36 |
4.14 |
4.07 |
|
4. 5-5. 5 |
5.52 |
5.37 |
5.17 |
5.72 |
5.22 |
5.20 |
|
5. 8-6. 7 |
6.83 |
6.64 |
6.64 |
6.55 |
6.46 |
6.40 |
|
4. 3-5.0 |
5.13 |
4.94 |
4.90 |
4.78 |
4.8 |
4.70 |
|
3. 9-4. 9 |
4.70 |
4.68 |
4.50 |
4.37 |
4.16 |
4.32 |
|
6. 5-7. 5 |
7.92 |
7.82 |
7.43 |
7.47 |
7.1 |
7.18 |
|
3.7-4. 2 |
4.25 |
4.36 |
4.40 |
4.25 |
4.11 |
4.35 |
|
1. 7-2.1 |
2.20 |
1.90 |
2.03 |
1.94 |
1.6 |
1.76 |
|
1.3-1. 7 |
2.30 |
1.64 |
1.52 |
1.74 |
2.1 |
1.86 |
|
2. 5-3.1 |
3.08 |
2.90 |
2.93 |
2.75 |
2.83 |
2.72 |
|
32.9 |
33.1 |
33.8 |
33.6 |
31.8 |
34.0 |
|
|
50.0 |
48.8 |
50.0 |
50.4 |
49.5 |
53.3 |
|
|
37.5 |
36.3 |
37.6 |
36.8 |
36.9 |
39.2 |
|
|
57.8 |
57.4 |
55.5 |
57.5 |
54.5 |
59.8 |
|
|
93.5 |
94.8 |
90.0 |
91.3 |
86.6 |
92.0 |
|
|
123.8 |
123.7 |
128.5 |
114.5 |
124 |
125.3 |
|
|
116.0 |
117.7 |
112.0 |
114.0 |
110 |
112.3 |
|
|
0.65 |
0.90 |
0.86 |
1.10 |
0.87 |
1.08 |
|
|
55.9 |
54.2 |
62.5 |
56.5 |
56.2 |
59.0 |
On the ventral surfaces the median hypapophysial ridge is reduced on 5 and 6, and parial hypapophyses appear on 7. These, as in Anomalopteryx, increase m size on suc¬ ceeding vertebrae, first separating and then approaching again to form, 20, a sing broad process which becomes higher, narrower and longer on 21 and 22 The ^P eu ap - physes have posterior processes from 6; they become longer and styhform i in the suc¬ ceeding five vertebrae, then broadening and shortening on the remaining cvosT styliform pleurapophyses in E. crassus are shorter and blunter than in either P. mappnn or A didiforL In the two dorsal vertebrae present (26 and 27) the
is divided by an arch into an anterior and a posterior portion, the atter being slig y more prominent. All the vertebrae of C.M. xiii D are considerably larger than the cor¬ responding vertebrae of A.M. 124 (P. mappini) and 51 (A. didifonms). The cervical ver brae are also wider in proportion to height, especially m the posterior vertebrae.
Ribs: The uncinate processes present are long and narrow. Tracheal iAngs^ There lire a few small slender rings with this skeleton, but most are thick, i.e. from 0.4 to 1.5 cm. high, with the walls up to 2 mm. thick. Twenty-seven thick rings have been fused into a narrow loop. (Text-fig. 13.) This also occurs m other skeletons.
49
„ , variation in the form and proportion of the
Sternum: There is a surprising degr® geen from above, the anterior margin is sternum in different individua s o g arg well.developed, sometimes broad, some-
slightly concave; the precostal P™: anterior margin the body of the sternum
times narrow. Immediately posten ■- depressions, deeper in some sterna than in exhibits a pair of antero-lateral cone & Behind thege the rest 0f the body of the
others, and marked by large pits o behind the level of the costal border,
sternum is evenly concave th® ufs The lateral margins are concave, or constricted,
It is much more concave than m t . • border; the postero-lateral processes
- sr — - - - as stout or as
convex as in Anoinaloptery.x.
• 1 -n rr cmH mav be broad or narrow; in the perfect sterna its
The median process is lo”«- ^ 0f the ten sterna examined, three
end falls short of the ends o a d notch and the other four are imper-
definitely have no median notch, t considerable variation as in Anomalopteryx ;
<«•>■ extending nbliguely the
“IIS" *!! front .1 tide » deep jit, «nd in front of the pit , targe, £«* trnter, « »rr.w band for the eee.nd rib; in front of thin '
The sterna of C.M. xiii F and xiii G (Text-fig. 14) are unusually long and narrow, and deeply concave, especially F. The median and lateral processes are long and slender and their ends bend down ventrally ; this, however, may be due to warping m drying. Both have a median notch. No. xiii F has an extra costal facet behind the depression that normally occurs just behind the third facet; this extra facet also has a depression behind it. A similar condition was noted in M. didinus (PL 12, fig. 3). The condition o the coracoid facet varies; it may be distinct, shallow or absent; or present on one side and absent from the other.
Pelvis: The pelvis of E. crassus is longer and more massive than in Anomalopteryx oi Megalaptcryx. The pre-acetabular centra are short and wide, and usually 35 has transverse processes ; the acetabular portion of the column is also broad and compressed post-axially though there are generally four vertebrae included therein. The post-acetabular region is very broad and strongly braced with transverse processes extending from 39 (fre¬ quently 38) to 46, and, occasionally, 47.
50
The usual formula is 28-35 (36-38) 39-46 as in Dinornis. There is no interpolated pre-acetabular vertebra with transverse processes; but in one case there is 35a lacking a process, and occasionally there are only 28-34 pre-acetabular vertebrae. In only one pelvis did 39 lack transverse processes. C.M. xiii D and xiii F have 47 (first caudal) fused to the pelvis. These variations are recorded in the formulae as follows:
|
• • • Vlll |
B. |
28-35 |
(35a-38) |
39-46 |
|
viii |
C. |
28-34 |
(35-38) |
39-46 |
|
ix |
A- j |
; 28-35 |
(36-38) |
39-46 |
|
1 |
i 28-34 |
(35-38) |
39-46 |
|
|
X |
B. , |
' 28-35 |
(36-39) |
40-46 |
|
< |
i 28-35 |
(36-38) |
39-46 |
|
|
viii |
D. |
" 28-35 |
(36-37) |
38-47 |
|
xiii |
F. < |
f 28-35 |
(36-38) |
39-47 |
|
< |
1 28-35 |
(36-37) |
38-47 |
28-35
(36-39) 40-47
Table 14.
Emeus crassus: dimensions of pelvis.
C.M. C.M. C.M. C.M. C.M. ? viii B. viii C. x B. xiii F. xiii D.
Length
Width at antitrochanters
Width at pectineal tubercles
Width of escutcheon
Height of pre-acetabular portion
Ischium length
Ischium height
Ischium divergence
Pubis length
Pubis height
Pubis divergence
Proportions % —
Width at antitrochanters: length.. Width of escutcheon: length Ischium divergence: antitrochanter width
|
41.1 |
41.9 |
37.6 |
40.6 |
44.4 |
|
21.9 |
22.2 |
22.4 |
23.0 |
22.9 |
|
15.5 |
15.5 |
15.5 |
17.0 |
16.7 |
|
19.8 |
17.5 |
17.2 |
20.3 |
18.9 |
|
12.5 |
11.8 |
12.0 |
• — |
12.5 |
|
18.5 |
17.5 |
15.0 |
18.5 |
22.0 |
|
5.5 |
4.5 |
6.8 |
7.0 |
6.0 |
|
- - - |
24.7 |
23.0 |
24.25 |
25.4 |
|
- - |
21.7 |
22.5 |
27.0 |
28.0 |
|
- - |
2.1 |
2.5 |
3.0 |
3.5 |
|
- - |
29.5 |
26.5 |
26.5 |
27.4 |
|
53.2 |
52.9 |
59.5 |
— • |
52.1 |
|
— • |
— • |
45.8 |
- - |
• - * |
|
111 |
. |
. - . |
112.8 |
Table 15. Emeus crassus: dimensions of sternum.
Widths A B C
D
Lengths E V G H
Depth of notch
viii B. viii C. ix A. xiii
F. xiii G. xiii D.
xiii E.
|
17.8 |
17.7 |
|
12.3 |
13.5 |
|
10.5 |
10.8 |
|
19.5 |
19.5 |
|
16.5 |
17.0 |
|
9.5 |
9.1 |
|
18.2 |
17.3 |
|
13.0 |
— • |
|
3.5 |
nil |
|
17.5 |
16.9 |
19.55 |
19.85 |
18.6 |
|
13.0 |
12.1 |
15.0 |
16.20 |
14.4 |
|
10.8 |
8.0 |
10.5 |
11.1 |
12.0 |
|
— |
18.0 |
19.0 |
21.3 |
• — |
|
- - - |
20.5 |
23.5 |
21.9 |
• — |
|
9.5 |
10.0 |
23.5 |
24.3 |
— |
|
- - |
23.0 |
10.5 |
10.5 |
14.3 |
|
15.5 |
19.5 |
21.1 |
— |
— |
|
9 |
1.0 |
2.4 |
none |
Distribution ■ South Island: Glenmark, Pyramid Valley, Enfield, Kapua, Kia Ora, Wai- kouiti, Hamilton Swamp. The only North Island bones referable to E. crassus are a skull from Maryborough and a tibia from Te Aute in the Dominion Museum, and a tibia
from Te Aute in the Canterbury Museum.
51
|
1869 |
May |
• • |
|
1875 |
July . . |
• • |
|
1879 |
. . |
• • |
|
1892 |
May |
• • |
|
1893 |
May |
• • |
|
1895 |
October |
• • |
|
1896 |
Juno |
|
|
1896 |
Juno |
• • |
|
1897 |
June |
• ♦ |
|
1907 |
November |
12 |
|
1930 |
. . |
• • |
|
1933 |
• • • • |
• • |
|
1934 |
August 20 |
• • |
Emeus huttonii (Owen), 1879.
• j-jvt • ■ Haast Trans. N.Z. Inst. 1, PP- 82, 83 (2nd ed. 18/5, pp. 23, 24). Dinorms didiforms. Haast, iia
dUKformit: Hutton. Trans. N.Z. Inst PP* •
Leg-bones from Table A, minimum, femur and metatarsus.
D!nornisTu«onii Owen, Extinct Birds of N.Z., 430. Founded on above leg-bones (D. didiformis: Hutton 1875): lectotype nominated below.
,. j. TTntton (nart) , Trans. N.Z. Inst. 24, 129.
E^uryapteryx com pacta Hutton, Trans. N.Z. Inst. 25, 11. Type: a tibia from Enfield, in Canterbury Museum.
Mesopteryx didina: Hutton, ibid, _ o7o Dis gi fies 39 and
Mcsoftcnx species 7 Parker, Trans. Zool. Soc 13, pt U, P- a?8, pis. 01 0
44 pi. 62, fig. 54. Type: skull on a “skeleton of M . didina m Ctago Museum
(cf. Benham 1934, p. 92).
. . TTntton Trans. N.Z. Inst., 28, pp. G36, 642. Bones from Kapua.
Mcionornis didZis: Hutton! Trans. N.Z. Inst. 28, pp. 646, 648. Bones from Enfield. Meionornis didinus: Hutton, Trans. N.Z. Inst 29
Megalapteryx huttonii: Rothschild (part). Extinct Birds, 199, pi. 41.
Emeus huttonii: Oliver, "New Zealand Birds,” 49.
Fmcut huttonii- Lambrecht, Handbuch der Palaeornithologie, 147. eZus huttonii: Benham, Trans. Roy. Soc. N.Z., 64, 87, pis. 5-8.
Founded on mixed leg-bones from Hamilton Swamp, Otago (Hutton, 1875) . Lectotype, hereby nominated, a right metatarsus with the following dimensions (Otago Museum) :
17.0 6.3 3.7 8.0 10.0
_ 100 38.0 18.0 45.0 55.2
The indefinite manner of the proposal of this species has been the cause of considerable confusion.
Hutton (1875) gave the dimensions of two femora, seven tibiae and six metatarsi from Hamilton Swamp, Central Otago, under the name Dinornis didiformis. He remarked, “The bones I have arranged under D. didiformis belong possibly to a new species. The tibia is well marked and quite distinct,* but the femur and metatarsus that I have asso¬ ciated with it pass almost into D. casuarinus , but are rather smaller.’’
Owen (1879), after quoting Hutton’s remarks as above, states, “Possibly the Dinornis of the South Island, with the tibia characteristic of the D. didiformis of the North Island, may need to be noted, for the convenience of naming the bones, as Dinornis huttonii."
Hutton (1892, and again in 1896 and 1897) confused huttonii with didinus (Megalap¬ teryx)]' and Rothschild (1907, as Megalapteryx huttonii) perpetuated this misunderstand¬ ing. Lydekker made no mention whatever of huttonii.
Oliver (1930, p. 49), in recounting how E. huttonii had been proposed, states that Owen selected the tibio-tarsus as the type, but this was not the meaning of Owen’s phrase, “with the tibia characteristic of the D. didiformis of the North Island.” Owen was pro¬ posing the species on the basis of Hutton’s earlier description; having no specimens, he could not do otherwise. Hutton’s observations were to the effect that, while the tibia was characteristic of D. didiformis, the femur and metatarsus were different; they were stouter, like D. casuarinus, but smaller than that species. Now this is exactly what E. huttonii is: a species with a tibia hardly distinguishable from that of Anomalopteryx didi¬ formis, but with shorter and stouter femur and metatarsus ; besides, if we were to regard the tibia as the type we could not satisfactorily separate E. huttonii from A. didiformis.
* Obviously Hutton means “quite distinct as didiformis ” This is important. fBenham (1934) has discussed this adequately.
At present there are, in the Otago Museum, only three tibiae from Hamilton Swamp with dimensions approximating to those given by Hutton, and they might be either didifonnis or luiUonii; but the femora and metatarsi are, as Hutton said, much stouter, and this is the real basis of Owen’s proposal. The selection just made of one of these metatarsi as the type is, therefore, in accordance with the original indication of the characters regarded as distinguishing the species. Benham (1934, p. 93) noted a dis¬ crepancy in the dimensions from Hutton for this species as quoted by Oliver and him¬ self. The explanation is that Oliver gave Hutton’s maximum figures, while Benham recorded the mean.
Table E, giving the dimensions of E. huttonii, reveals the paucity of individual skele¬ tons of this species, and I have considered it advisable to add to the table average measurements recorded by Hutton from Hamilton. Hutton’s smallest femur and meta¬ tarsus of his “ casuarimis” Hamilton (1875) also fall here. The Wakapatu specimen des¬ cribed in gratifying detail by Sir William Benham (1934) is the only complete skeleton known of this species, for the Hamilton Swamp mounted specimen, also in the Otago Museum is, Sir William notes, very doubtfully of a single individual. A partial skeleton in good condition has been recovered from Pyramid Valley swamp.
It has already been noted that the skull of the Wakapatu skeleton is small, but, as Benham points out, its shortness is due to the shortness of the beak, and “it may be a few millimetres short of its actual length.” I would suggest 10.5 cm. as its approximate true length. The beak is of the moderately rounded type noted in E. crassus; the maxillo- palatine is missing, so the condition of the antrum cannot be stated ; but in an exactly similar skull of unknown locality in the Dominion Museum the antrum is expanded, as it is in the two E. crassus skulls from Pyramid Valley that have the maxillopalatine preserved. The Wakapatu skeleton has a narrow sternum and typical Emeine pelvis with a flat escut¬ cheon and with ischia and pubes exhibiting considerable divergence. The pelvic formula, according to Benham, is 28-35 (36-37) 38-46: it is not possible to ascertain, now that the skeleton has been mounted, whether the small vertebra no. 38, with its neural canals set high up and difficult to observe at any time, is present as it is in all other pelves I have examined, but, assuming that it is, the formula would be 28-35 (36-38) 39-47. This supposi¬ tion is supported by a comparison of the disposition of the transverse processes of the post- acetabular vertebrae with the arrangement in the Pyramid Valley skeletons of E. crassus (p. 51) .
Distribution: Swamp localities in Canterbury and Otago and sand-dunes at Wakapatu.
Genus Eury apteryx Haast, 1874.
Type, herein selected, Dinomis gravis Owen.
1874 June . . Euryapteryx Haast, Trans. N.Z. Inst. 6, p. 427.
1874 July . . . . Euryapteryx : Haast, Ibis, ser. 3, vol. 4, p. 213.
Note: Ccla. used by Reichenbach for Dinomis curtus, was not the proposal of a new genus; it was merely assigning D. curtus to Mohring’s pre-Linnean genus Cda Gesl. Vog. 4, 43, 1752; (cf. Hutton 1895a, 158).
Lydekker stated (p. 298) of his species Emeus gravipes that it was “the type species of Euryapteryx but this species was not included under the generic name at the time of its original publication and cannot therefore be nominated as the type. Of the two species included in the genus by Haast, one, rheides, is indeterminate; and I there¬ fore select the other, Dinomis gravis Owen, as the genotype.
53
, V nf Parhvornis in the sizes, proportions and
The species of Euryapteryx dup L in occurring as a North Island group
relative lengths of the leg-bones a ' ’ group of large massive species,
of small relatively slender forms an a .o ^ & broad> roUnd-tipped beak
They differ from all other geneia . 42) and they share with Emeus the
’^te^uTaid^' reduction" of the phalanges of the outer toe to four, including the claw joint.
• no , i \r TaianH «neoies are recognized, but all the South Island
spedmOTsTrtlncMed "in one species of which an occasional example has been obtained in the south-eastern portion of the North Island.
1870 January
|
1872 |
. . |
|
1873 |
May . . |
|
1874 |
June |
|
1874 |
July . . |
|
1879 |
• • • • |
|
1891 |
April 25 |
|
1891 |
April 25 |
1891 November 13 1895 October
1898
190G
1907
1910 . .
1930 ..
1933
Euryapteryx gravis (Owen), 1870.
Dinornis gravis Owen, Trans. Zool. Soc., vol. 7, pt. 2, p. 141, pi. 14. Type a skull (possibly of an individual skeleton) from the “Kahamin (= Kakanm) River, Otago at til at time in the possession of the Baroness Burdett Coutts.
Dinornis gravis: Owen, Proc. Zool. Soc. for 1872, No. 38 6°5
Dinar,, is gravis: Owen, Trans. Zool. Soc., vol. 8, pi. 6, PP- -61-380. pis. 58 61.
Euryapteryx gravis: Haast, Trans. N.Z. Inst. 0, 42
Euryapteryx gravis: Haast, “The Ibis,” ser. 3, voL 4 No 15, 209.
Dinornis gravis: Owen, “Extinct Birds of New Zealand, pp. 34.-364, pis. 41,
42 42a
Emeus gravipes Lydekker, Cat. Foss. Birds, British Museum, p. 297 Type: Meta¬ tarsus from Kakanui River, Otago, in British Museum (No. A. 1591).
Dinornis gravis: Lydekker. Cat. Foss. Birds. Brit. Mus., 298 (note under Emeus
gravipes).
Emeus crassus: Lydekker (part, skull), ibid.. P- 311.
Emeus gravipes: Sharpe, Cat. Ost. Vertebr. Mus. Roy. Coll. Surg, London, III. , 434 Emeus sp. Barker. Trans. Zool. Soc., 13, 379, pi. 61. Skull from Shag Point, m Otago Museum.
Emeus sp. (3 Parker. Trans. Zool. Soc. 13, p. 379. Skull on (composite) skeleton named E. gravis from Glenmark, in Canteibuiy Museum.
Euryapteryx ponderosa: Hamilton, Trans. N.Z. Inst. o0, 445.
Emeus crassus: Hutton, Trans. N.Z. Inst., 38, 66 (type of Eu. kuranui Oliver). Emeus parkeri Rothschild, “Extinct Birds,” p. 210. TYPE: skull named Emeus sp. 7 Parker, 1895, p. 379.
Emeus booth i Rothschild, “Extinct Birds,” p. 210. Type: Skull of Emeus sp. a Parker, 1895.
Emeus haasti Rothschild: “Extinct Birds,” p. 210. Type: Skull of Emeus sp. ft Parker, 1895.
Emeus gravipes: Rothschild, Extinct Birds, 210.
Euryapteryx crassa: Benham. Trans. N.Z. Inst., 42, 354.
Euryapteryx kuranui: Oliver, “N.Z. Birds,” 52. TYPE: skeleton from Castle Point (North Island) in Canterbury Museum.
Euryapteryx gravipes : Oliver, New Zealand Birds, 53.
Euryapteryx kuranui: Lambrecht, j bid., 151.
Euryapteryx gravipes: Lambrecht, Handbuch der Palaeornithologie, 152.
The references to the figures of the mandible of D. gravis in Plate 14 of Trans. Zool. Soc. 7 and Plate 81 of Owen’s “Extinct Birds” are very muddled both in the text and the Description of Plates; but a careful reading of the descriptions and of the compari¬ sons with crassus and rheides makes it clear that, in each of the above two publications, figure 5 is of rheides and fig. 6 is of gravis, and that the latter has a broad blunt beak.
There is a point to be discussed with regard to this species. Dinornis gravis was origin¬ ally proposed by Owen (1870, p. 141) for a skull, and subsequently (1878, p. 361) leg- bones were described as being of the species. Lydekker assumed that the leg-bones were not those of an individual; he included the skull under Emeus crassus (p. 307, 311), and designated the metatarsus as the type of a new species, Emeus gravipes.
54
Owen, however, when describing the skull, had said, “Many characteristic parts of the skeleton of the same individual bird were obtained by William Fenwick, Esq., at the Kahamin (= Kakanui) River, Middle Island, New Zealand, and were presented by that gentleman to Miss A. Burdett Coutts. They were confided to me by that lady for deter¬ mination in 1867 ; and the grounds on which I came to the conclusion that they represented a species not previously recognized may be communicated at a future period to the Zoological Society. ... I give a description of the skull of the new species in the present Memoir.” Owen intended to state the specific characters of the new species later, and this he purported to do when, in 1873, he described the leg-bones. Of the latter he said (p. 379), “The specimens of Dinornis gravis above described and figured were dis¬ covered in the bed of the Kakamai (= Kakanui) River, South Island, by Wm. Fenwick, Esq. I am indebted to the kindness of the Baroness Burdett Coutts for the loan of the
specimens.”
The skull and leg-bones would therefore appear to be those of one and the same indi¬ vidual, and Lydekker’s proposal of a new name for the metatarsus would seem to have been unnecessary. Yet, notwithstanding Owen’s explicit statements quoted above, there is still reason, in the nature of the bones themselves, to doubt their individual associa¬ tion. The femur is much longer in proportion to the tibia and metatarsus than in any undoubtedly individual skeleton of Envy apteryx, and, judging from Owen’s figures, its distal end is too wide to articulate with the tibia ; moreover, its proportions accord with those of Emeus crassus (cf. xiii. F Pyramid Valley, Table F), and not with those of
Euryaptcryx.
It might be considered that with this doubt before us we should continue to retain Euryaptcryx gravis (Owen) for the skull and Eu. gravipes (Lyd.) for the metatarsus. It is not necessary, however, to prove the individual association of skull and metatarsus; it is sufficient that their specific identity be established and this is satisfactorily confirmed by means of individual skeletons in the Otago and Canterbuiy Museums.
These show that same association of broad-beaked skulls, narrow sterna and four- jointed outer toes with characteristic form of leg-bones in large South Island birds as is found in much smaller North Island forms. The tibia is less inflected than in same sized species of Pacliyornis, the metatarsus is usually, though not invariably, less constnctec at the middle, has a lower proximal intercondylar ridge and a less abruptly projecting
middle distal trochlea.
The metatarsus (type of Eu, gravipes Lyd.) is now in the British Museum (No A. 1591). having been acquired in 1923 from the estate of the Baroness Burdett Coutts, Miss Dorothea M. A. Bate has informed me in a recent letter that the following bones came with it, and are entered in the catalogue as “one of the type specimens :
A. 1592 — tibio-tarsus, figured Owen, T.Z.S. 8, pi. 59, figs. 1-3.
A. 1593 _ right femur, figured op. cit. pis. 60 and 61.
A. 1594 — pelvis, described op. cit. p. 369.
Unfortunately I was not aware of the status of A.1591 when I measured it, and thus did not enquire as to whether other bones were associated with it.
c, ...» In occipital view (PI. 6, fig. 2) the skull has a well arched vault; the supra-
xst* .»■>«» •• i» ~ “e w“J
the supra-occipital fossae being consequently better defined. The distance between anterior and posterior lambdoidal ridges varies ; there is a depression on each side above and mesad of the paroccipitals, but it is not deep and pit-like as in E. crassus. The par-
55
occipital processes have convex or sinuate outer borders and extend downwards to a varying degree ; the mamillar tuberosities also vary in s,ze and stance apart, but are usually less prominent than in Pack, elephaniopus.
In lateral view (pi. 6, fig. D the posterior border of the tympanic «ivity is oblique and may be straight, sinuate, or slightly convex; it meets the uppei bmdei at a light angle or less. The upper border projects outwards, forming a ledge above the tympanic "Sy anteriorly it curves evenly forward on to the zygomatic process, whicn is fairly acuteand varies in length. The space between temporal and lambdoidal ridges is from 5 to 10 mm wide; it continues outward to form a broad flat space above the squamosal prominence.' The post-temporal fossa is relatively rather wider and deeper than in Emeus The post-orbital processes are usually proportionately wider than in Emeus,
with the outer portion projecting straight down vertically or slightly forward. The
margin of the orbit is either an even arch, sinuate, or, rarely, forming a wide ang e (Emeus sp. y Parker). The orbits face more to the front than m Emeus. The lachrymal foramen is formed as usual, by a notch in the antorbital closed anteriorly by the maxi o- nasal In dorsal view the wide spread of the post-orbitals and the forward aspect ot the orbital margin are readily apparent ; the anterior portion of the temporal fossa is definitely overhung by the temporal ridge. Sometimes there is a double tumidity on the
roof ; usually it is a single slight eminence.
Certain skulls with their own beak (from Shag Point in Otago Museum; Emeus sp. a Lydekker in British Museum; and E. kuranui Oliver in Canterbury Museum) have the premaxilla and mandible short, broad and widely rounded terminally. In the Riverton specimen (pi. 6) the premaxilla is strongly downcurved and the tip abruptly tiuncated, the mandible is very stout and more depressed terminally than in Emeus. These broad¬ billed skulls have a characteristic maxillo-palatal structure ; the palatals are only slightly curved, the maxillo- jugal is stout and nearly straight, meeting the palatal at an acute angle, and there is no antrum cavity.
Pelvis: The pelvis in Eu, gravis is larger, relatively broader and more massive than in Em, crassus. Its pre-acetabular portion, though short, is not especially splayed laterally, but it is strongly expanded at the acetabulum and antitrochanters. The width at the antitrochanters is nearly half the total length. The pre-acetabular dorsal iliac margin is convex ; posteriorly the ilia diverge abruptly, curving outwards to almost right angles with the long axis to form the anterior margins of the escutcheon. The latter is flat, short and very broad, its lateral margins are convex and converge strongly pos¬ teriorly. The vertical laminae or sides of the escutcheon are strongly deflected inwards (they slope outwards in Dinornis). The isehia are generally, but not always, strongly divergent posteriorly. As with the other genera the formula of the vertebral elements comprising the pelvis varies, i.e. :
|
28-35 |
(36-38) |
39-47 |
|
28-34 |
(35-39) |
40-46' |
|
28-35 |
(36-38) |
39-?47 |
|
28-35 |
(36-38) |
39-? |
Distribution: Eu. gravis, like Pack, elephaniopus, was common in Canterbury and Otago; it has also been obtained on Stewart Island (Benham 1910) and I have a set of leg-bones from Mt. Arthur tableland, near Nelson. E. kuranui Oliver might perhaps have been regarded as a more slender North Island form, but it is matched in slenderness by the Stewart Island specimen (Table H) and the only other North Island specimen is a quite stout metatarsus from Portland Island in the Hawke’s Bay Museum.
56
1848 April 13
1848 April 22
1866 July 6
1891
1891 April 25
1927 August
Eury apteryx geranoides (Owen), 1848.
Palapteryx geranoides Owen. Proc. Zool. Soc. for 1848, p. 1, nomen nudum.
Palapteryx geranoides Owen. Ibid., P- 7, nomen nudum. Dimensions given of a skull, but insufficient to identify it except by reference to the paper next cited. Palapteryx geranoides Owen. Trans. Zool. Soc., Vol. 3, pt. 5, p. 361, pi. 54, figs. 1-5. Non Dinornis geranoides: Owen, Trans. Zool, Soc. vol. 5, pt. 5, pp. 400-402, pi. 65, figs. 5-6 (femur of Eu. exilis); pi. 67, figs. 5-6 (metatarsus of Eu. ex His). Anomaloptcryx dromacoidcs : Lydekker (part), Cat. Foss. Birds Brit. Mus., 268 (B.M. right metatarsus, 21793).
Non Anomaloptcryx geranoides Lydekker, Cat. Foss. Birds Brit. Mus. 288. (Eu. exilis). Dinornis expunctus Archey. Trans. N.Z. Inst., 58, 152.
The name geranoides has been a source of confusion through its having been given originally to a skull which could not satisfactorily be affiliated with any particular leg- bones. Those of the next species, Eu. exilis , have from time to time been attributed to it, but the several individual skeletons we now have of exilis show that the geranoides skull is much too large for that species. It is, however, definitely of Euryapteryx. Finality as to its relationship must await the discovery of an individual skeleton; we have, how¬ ever, from coastal dune-areas, a few sets of leg-bones (Table G) of Euryapteryx (one accompanied by a large broad beak) considerably larger than those of exilis, but much smaller than gravis. They are indeed appropriate in size to the skull of geranoides; more¬ over another Euryapteryx tibia (B.M. 21793; Lydekker, p. 268) of the same dimensions was found at Te Rangatapu, the dune deposit that yielded the type skull. There seems to be good reason for associating the calvarim and leg-bones together and for regarding Eu. geranoides (Owen) as the largest of the North Island group of three small species of
Euryapteryx.
Localities : North Island — Te Rangapatu, Doubtless Bay, Tom Bowling Bay.
1866 July 6 1891 April 25
1891 November 13 1895 October 1897 June
Euryapteryx exilis Hutton 1897.
Dinornis geranoides: Owen, Trans. Zool. Soc. vol. 5, pt. 5, pp. 401-2, p. 67, figs. 5 and 6. (B.M. 21706, Lydekker, p. 289).
Anomalopteryx (?) geranoides Lydekker, Cat. Foss. Birds Brit. Mus. 288, fig. 65C, p. 317. Founded on four tibiae from Te Rangatapu, of which that first men¬ tioned, No. 21789x, is here nominated as the TYPE.
Anomalopteryx geranoides: Sharpe, Cat. Ost. Vertebr. Mus. Roy. Coll. Surg., 634.
Mesoptcryx sp. a Parker, Trans. Zool. Soc., 13, pt. 11, 378, pi. 61, figs. 28, 41.
Euryapteryx exilis Hutton, Trans. N.Z. Inst., 29. 652, pi. 48, fig. C. TYPE: Skele- ton from Wangaehu, in Wanganui Museum.
See under Pachyornis lnappini for references part niappim.
The retention of geranoides Owen as the name for the last species occasions the rejec¬ tion of Anomalopteryx geranoides Lydekker for this, which is also a species of Euryapteryx. The next, and only other, name available is Euryapteryx exilis Hutton, which is accord- ingly adopted. The type of Eu. exilis is a skeleton from Wangaehu in the Wanganui Museum. Dr. Oliver, who discussed the doubtful association of the skull with this ske e- ton, and at that time (1930, 49) concluded that the cranium might have belonged to it but not the beak, has since drawn my attention to the fact that the skull at present, on the skeleton is not the one figured by Hutton (1897, pi. 47). The latter was a typical Euryapteryx skull with a broad beak apparently firmly fused in position.
Description :
This species provides as good an example as Anomalopteryx did, f omits of the undesira¬ bility of using dimensions and proportions of a series of only one of the leg-bones in e er- mining species. The Tables of Measurements A and H show that the tibiae of these two
57
, ntlv. moreover, although tibiae of exilis species differ but little and by no ““M ®°“ J ’ t the middle of the shaft and more
5is**Ti — ns.
in exilis are immediately apparent.
The average lengths of femur and metatarsus in these two spemes, expresse percentages of the length of the tibia, are T m.
c ... 59.6 =100 44'3
E'e " " " ' mi =100 45.4
A. didiformis . .
Eu. exilis, like P . mappini, was thus a shorter, rela^e1^ s flange' andby bird than A. didiformis. It differed from P, tarsug> Somewhat similar differences the lower proximal intercondylar rising o i a' didiformis i.e. at the proximal
separate the tibia and metatarsus of Eu. ex is and h (15b)
end (Text-fig. 15) the inter-condylar ridge (t.r.) is higher in a.
Text-fig. 15.
E. exilis (a, c) ; A. didiformis (b, d) :
metatarsi, proximal and distal ends.
and anteriorly it expands and rises to form a swelling on the front of the bone. The distal trochleae are much larger in E. exilis, the inter-trochlear spaces being thereby much reduced. Seen from above, the middle trochlea . is rather shorter m A. didiformis, and the inner and outer trochleae have each a more distinct groove on then distal face. ’ The depression on the ectal surface of both inner and outer . trochlea is deeper and subcircular in A. didiformis (Text-fig. 15d) and is more of a wld^ shallow groove in E. exilis (Text-fig. 15c), and, viewed from the side the lower margin of the shaft in E. exilis curves gently into the hinder border of the inner trochlea, which is somewhat angular and not evenly curved below. In A. didiformis the shaft curves down moie quickly and the lower margin of the trochlea is rounded.
Skull. Comparing the skulls of this species and of Eu. curtus with those of like-sized Pachyornis skeletons, the Enryapteryx skulls are definitely smaller (Table 8, p. 43). Differ¬ ences in form are the lack of a pronounced supraforaminal swelling in Enryapteryx, in which also the much smaller temporal fossa extends upwards instead of backward as m Pachyornis; the wide, round-tipped beak and characteristic maxillo palatal structure m exMs and curtus have already been noted as characters defining the genus Enryapteryx.
In lateral view the dorsal outline in exilis usually shows an eminence above the post¬ orbitals, and a lesser one above the pre-orbitals: these, if present, are very slight in Anomalopteryx, but may be present in Pachyornis (PI. 4, fig. 4). The lambdoidal and tem-
58
poral ridges are separated by a space of varying width (3.4 to 7.5 mm.), which
continues outwards on to the squamosal prominence as a flat area bounded antero- laterally by a well-developed posterior temporal ridge which separates it from the nai- row, concave post-temporal fossa. (In Anomalopteryx the confluence of the lidges
restricts this supra-squamosal area which the very low posterior temporal ridge scaicely separates from the post-temporal fossa). The posterior and upper margins of the tym¬ panic cavity usually form an angle.
The temporal fossa is very small in E. exilis ; it is smaller than the orbit. Its
lesser encroachment on the roof is indicated by the range of percentages distance
between temporal ridges : width between temporal fossae,” from 81.4 to 88.7 (cf. 52 to 70 in A. didiformis). The mid-temporal ridge, so prominent in A. didiformis, is se om indicated in E . exilis. The post-orbitals project straight down or slightly forward, as m P achy amis, instead of backward in Anomalopteryx. Maxillo-nasals are not preseived in any of the skulls of exilis or curtus; but they are all incomplete specimens from sand-hills, and this small bone would have been one of the first to be lost. It will be recalled that it was
well-developed, but discrete, in Eu. grains.
Vertebral Column. The vertebrae in Eu. exilis agree with those of P. mappini in being narrower than in A. didiformis. The nape cervicals have the usual subquadrate . dorsa outline; but 3 and 4 are narrower than in A. didiformis , as are 5 and 6 also, except m com¬ parison with those individuals in A. didiformis, in which 5 and 6 assume the form of tne ordinary cervicals (i.e. with long, narrow post-zygapophyses) . Otherwise there is the same kind of variation between individuals : i.e. the change from double to single neural spines mav occur on either 19 or 20, or the change from parial to median hypapophyses on 18, . , or 20. In the pelvic vertebrae 28, 29, and sometimes 30, the paired submedian hypapo- physial projections stand farther apart than in Anomalopteryx, and hypapophyses may
also appear vaguely on 40 to 44.
The grouping of the pelvic vertebrae is normally as in Anomalopteryx, with variations as indicated in the following formulae .
TYPE : E. exilis
AM. 3
A.M. 160
A.M. 83
|
28-34 |
(35-38) |
39 — broken |
|
28-34 |
(35-37) |
38-45 |
|
r 28-34 |
(35-38) |
39-46 |
|
1. 28-35 |
(36-38) |
39-46 |
|
28-35 |
(36-38) |
39-45 (46) |
The pelvis in Eu. exilis is shorter than in A. didiformis and relatively wider Its dorsal iliac margin (PI. 15, fig. 2a) is intermediate in curvature between that o /. < ' 'foi s • 9 fig 3) and P mappini (PI. 15, fig. la) ; the front margin is deeper and more round y curved than in the other two. Seen from above (PL 15, fig. 2b) the ilia -teriorly a e ag-ain intermediate between A. didiformis (PI. 10, fig. 3) and P. mappini (PL 15, fig. lb) m the degree of horizontal splaying of the lower front margin .and the constriction m fron of the acetabulum. The dorsal iliac margins diverge poster uorly, . ^duafiy m ^ «
as in P. mappini, not abruptly as in A 2c) t^ fu SSSSll
centrfin frTtTthfacetebulum form a shorter, broader mass than in P niappini and 1 didiformis ■ the post acetabular column is narrower, with concave longitudinal curvatuie as in A didiformis, instead of broad and straight or with slightly convex longitu in curvature (P map ini) The ischia and ilia diverge posteriorly more than m A. didiformis and less than in P. mappini. The pelvis of Eu. curtus differs from that of Eu. exilis m
size only.
59
The Sternum in Eu. exilis (PI. 12, fig. 6) is narrow, its fron t margin 13 5 sl J ’
the lateral processes show only moderate divergence and the median i prow ^ * t
notched. In only one is there a depression, on one side only which represent^the
coracoid facet; no scapulo-coracoid has been found so far, but have been by no means complete.
Distribution.
North Island: Wangaehu (Type), Te Rangatapu, Doubtless Bay.
18 4 G July
1846
1852
1866
Eurvapteryx curtus (Owen), 1846.
Dinornis curtus Owen, Proc. Zool. Soc. London, Ft. 14, p. 48. Type: by subsequent designation (Lydekker 1891a, 281) tibia from the North Island; its piesent place of deposition is unknown; probably British Museum or Museum of Royal
College of Surgeons. , _ ...
. Dinornis curtus Owen, Trans. Zool. Soc. London, vol. 3, pt. 4, p. -25, pi. . o figs. 3, , o.
Ccla curtus Reichenbach, Nat. Syst. Vogel, p. xxx.
. Dinornis geranoides: Owen (part: femur) Trans. Zool. Soc. 5, p. 400, pi. 60, figs. 5,
(B.M. 21781, Lydekker p. 283).
. Dinornis curtus : Owen, Trans. Zool. Soc. 7, pt. 5, p. 353, pi. A, figs, (metatarsus).
See under Pachyornis ozueni for references part Eu. curtus , part I .
December
1871 January
river# m
Eu. curtus (Table I) is separated by size only from Eu. exilis (Table H) . With the ex¬ ception of the type, which is from an unspecified North Island locality, ail the specimens are from Doubtless Bay. The following table indicates the dimensions and proportions of the skull wherein Eu. curtus and P. ozueni differ, i.e. the longer skull in P. ozueni, its greater preorbital width, the greater width and depth of the temporal fossae and the rather less width of the orbit.
Table 12. Skulls of P. ozueni and Eu. curtus.
|
P. |
oweni. |
Eu. |
curtus. |
|||
|
Whangarei TYPE. |
Doubtless Bay |
Doubtless Bay |
Doubtless Bay |
Doubtless Bay |
Doubtless Bay |
|
|
Auckland Museum No. |
384 |
150a |
5 |
364 |
180 |
4 |
|
Total length |
10.5* |
10.45 |
10.15 |
10.00 |
9.75 |
9.60 |
|
Length: parocc. to preorbital .. |
• |
5.90 |
5.93 |
5.85 |
5.70 |
5.55 |
|
Height |
3.90 |
3.76 |
3.70 |
3.60 |
3.60 |
3.37 |
|
Width at parocc. proc. . . |
. |
4.60 |
4.80 |
4.50 |
4.55 |
4.36 |
|
Width at squam. prom. . . |
5.50 |
5.40 |
5.58 |
5.55 |
5.37 |
5.23 |
|
Width at temp, fossae . . |
4.15 |
3.82 |
3.95 |
3.95 |
3.80 |
3.60 |
|
Width between temp, ridges . . |
2.95 |
3.28 |
3.37 |
3.32 |
3.30 |
3.14 |
|
Width at post-orbitals . . |
6.20 |
6.16 |
6.28 |
6.25 |
5.75 |
5.60 |
|
Width at pre-orbitals |
• |
3.17 |
3.00 |
3.05 |
3.05 |
2.85 |
|
Width of tympanic cavity |
• |
1.60 |
1.44 |
1.55 |
1.34 |
1.44 |
|
Width of temp, fossa . |
• |
2.44 |
1.86 |
1.65 |
1.97 |
1.72 |
|
Width of orbit |
• |
2.47 |
2.50 |
2.50 |
2.54 |
2.26 |
|
% Squam. width : length |
• |
51.6 |
54.9 |
55.5 |
55.0 |
55.5 |
|
% Width at temp, foss.: length |
• |
36.5 |
38.9 |
39.5 |
38.9 |
37.5 |
|
% Post-orbital width: length .. |
• |
58.9 |
61.8 |
62.5 |
58.9 |
58.3 |
|
% Temp, ridges width: fossae width. . |
. |
85.8 |
85.3 |
96.0 |
86.8 |
87.2 |
|
% Length parocc. |