Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 67, 1938
This text is also available in PDF
(2 MB) Opens in new window
– 1 –

Fossil Cetacea of New Zealand II.—On Lophocephalus, a New Genus of Zeuglodont Cetacea*

[Read before the Otago Branch, September 8, 1936; received by the Editor, October 28, 1936; issued separately, June, 1937.]

This new genus is founded on the study of two skulls, or rather on the posterior part of two crania, which are manifestly not Squalodonts; both are very imperfect in that the rostrum and frontal region are lacking, but sufficient of the cranium remains to indicate (1) that the nostrils were situated far forwards, (2) that the parietal bones take a considerable share in the formation of the roof. The two parietals meet on the dorsal surface and produce a well-marked “sagittal crest” along the median line, from which each bone slopes rather steeply at an angle of 45° down to the broken latero-ventral margin; there is no sign of any great decrease in width of the cranium, which is as great at the anterior as at the posterior end of the fragment, owing to imperfection in the region of squamosal.

Of these two skulls, the larger belongs to the Department of Geology, and for the examination of this I wish to thank Professor Benson. Unfortunately we have no record of the actual locality at which either of the skulls was found, but from the character of the matrix there is no doubt that they came either from the Milburn or the Clarendon quarry, white limestone of “about the Upper Oligocene” (Marwick).

It naturally occurred to me that this skull might be that of Kekenodon but as not teeth accompany it, and as no skull is known for Kekenodon, it seems necessary to give a new name to its owner. Even the larger of the two crania appears to be too small to carry the huge teeth, 4–5 inches in length, of K. onamata.

Since it is desirable to have a distinctive name for the fossil, I have taken the sagittal crest as the foundation for the generic name (lophos being Greek for the crest carried by a helmet), while the specific name serves to recall the important and extensive geological work carried out by Emeritus Professor James Park, recently of the University of Otago.

[Footnote] * The first article of this series was published in Vol. 65 of these Transactions, and entitled “The Teeth of an Extinct Whale, Microcetus hectori, n.sp.”

[Footnote] † J. Hector, Trans. N.Z. Inst., vol. 13, p. 435, 1880.

– 2 –

Order Archaeoceti Family Zeuglodontidae

Lophocephalus parki n.gen. et n.sp.

The length of the larger cranium is 46 centimetres from posterior edge of occipital condyle to extreme end, that of the smaller 32.5 cm. Each is broken across anteriorly at approximately the same place, viz., at a short distance behind the junction of parietal with frontal bones.

As the larger cranium has the occipital region more complete than the smaller one, I take it as the type of the genus, though the smaller one has a greater relative length of parietal bone remaining, being broken across slightly nearer to the frontal.

The supra-occipital slopes upward and forward from the foramen magnum forming a large triangular shield as in Prosqualodon; it is slightly concave from side to side; its lateral margins where it meets the exoccipitals and parietals rise up so as to produce a fairly well-defined lambdoidal crest to which the hinder margins of the parietals contribute; these two ridges or crests meet anteriorly in the middle line and coalesce with the sagittal crest. In the smaller skull the shield is incomplete posteriorly, and the lambdoidal crest is also broken laterally. The length of the supra-occipital shield is 26 cm., its breadth 18 cm. The inclination of this bone is greater than that figured for those species of which the literature is available to me, viz., Zeuglodon osiris and Prozeuglodon atrox of Andrews, where, indeed, the bone is nearly vertical, so that I suspect that in the present crania there must have been a downward compression or crushing previous to fossilisation; but both of them exhibit the same slope. The exoccipital region is fairly well preserved in the larger skull; the left condyle is entire, the right one broken away; the breadth over the two condyles would be 16 cm.; the width of the condyle is 6.5, its height 8 cm. The foramen magnum has a height of 4 cm. and a width of 7 cm.

The parietal bone is of considerable extent; indeed, it is relatively enormous as compared with the bone in existing Odontocetes and in the Squalodonts. Unfortunately the lower margin is broken away and the squamosal has also gone. A portion of the parietal has also been torn away above the region of the squamosal so that the “brain” is exposed. At the extreme anterior region of the bone there are a few short longitudinal cracks, indicative of some degree of pressure.

In Prozeuglodon and other members of the family the cranium is much constricted in the anterior region of the parietals; but in the present specimen there is not sufficient of the cranium to show this. In the smaller cranium in which relatively and absolutely a greater length of this bone is preserved, the former of these injuries is absent. In this smaller cranium the length of the bone on the right side along the ventral broken margin is 26 cm. measured from its junction with the exoccipital; its height from the top of the sagittal crest to this lower margin is 13.5 cm. along the surface of the bone. The roof of the cranium is at the anterior end very thick,

– 3 –

being as much as 4 cm. from the top of the crest to the inner surface of the bone, constituting, one would imagine, sufficient protection against crushing in of the roof. The surface of the parietal is smooth, slightly concave from above downwards, and there is no sign of the suture separating it from the frontal.

On the underside of the smaller cranium there is a piece of the mandible amongst other fragments of bone and some molluscan shells, one being a Pecten, all embedded in matrix. The mandible is broken, the fragment measures 14 cm. in length with a breadth of 8 cm.; the exposed surface is convex and smooth.

Chief dimensions in centimetres.
Smaller cranium. Larger cranium.
Length 32.5 46
Breadth 19 28—ventrally across broken edges of parietals.
Height 15 14.5 at junction of sagittal with lambdoidal crest.
Sagittal crest 14.5 13 measured from the lambdoidal crest.
Supra-occipital 12.5 26
Supra-occipital breadth 10 18 lambdoidal crest broken.
Exoccipital condyles, width over the two, 16.
Locality: Otago.
Geological horizon: Upper Oligocene.

A comparison of this cranium with the skull of such a Zeuglodont as P. atrox* indicates that the skull of Lophocephalus parki must have been somewhere about three feet in length. By measuring the figure of the Egyptian species, which is one-third of the natural size and multiplying therefore by three, I find that the complete skull was 60 cm. in length, of which the occipital and parietal region is 13.5 cm.; therefore the length of the rostrum plus the frontals was 46.5 cm. The sagittal crest measured from the lambdoidal crest is 9.9 cm.; hence the total length of the skull is about six times the length of this crest. In Lophocephalus parki the sagittal crest, which of course is imperfect, is 13 cm. long, hence the length of the entire skull would have been about 84 cm. Consequently even the smaller of the two would have been longer than P. atrox.

In addition to these crania there are certain other bones which I attribute to the species.

Premaxilla (Mus. regist.: no. C.'34.8).

A large heavy bone or rather a piece of matrix of limestone enclosed in a thin layer of bone which has scaled away in places. I do not know whether the bony material has been infiltrated by lime, but I presume that is so, as it is unlikely that the bone of this size would be hollow. The bone is an elongated rectangle broken at the ends, somewhat triangular in section; one surface which I take to be the upper or dorsal is traversed by four longer or shorter furrows,

[Footnote] * C. W. Andrews, “Tertiary Vertebrata of the Fayoum, Egypt,” Brit. Mus. Catalogue, 1906.

– 4 –

sometimes distinct grooves, then flattening out; they are, no doubt, artificial due to pressure and partial fracture. The bone is slightly curved lengthwise, the convex surface being the grooved surface. The other or ventral is feebly concave. The third side of the tri|angular section is “mesial” where it abutted upon the “mesorostral gutter.” One margin is nearly straight. This I take to be the mesial margin where it met perhaps the nasal bone; the other irregular, being chipped at intervals.

Dimension of premaxilla (cms.).
Length, mesial edge 38
Width, greatest 11.5
Width at each end 9
Thickness, greatest 5
Thickness, outer edge 1.5

As to the interpretation of this bone the apparent hollowness: the matrix enveloped in this layer of bone, led me to think at first that it was part of a mandible, but there is no trace of teeth or sockets, and in Zeuglodonts the teeth extend the whole length of the jaw bone. But the grooving on the one surface makes it more likely to be a bone from the upper surface of the skull, so it may be either a nasal or premaxilla. It does not fit against the end of either of the two crania, and it was found at a different date and time, though it comes from Milburn.

Comparing it with Prozeuglodon atrox, with skull of 60 cm. length, the premaxilla is 27 cm. and the nasal 16.5 cm., and as the measurements of the skull of Lophocephalus indicate a longer skull there is no incompatibility in regarding this bone as a pre|maxilla.


There is no record as to the provenance of this, but the matrix resembles the Clarendon rock in which teeth occur (p. 6). This mandible has the condyle entire, but the coronoid is lacking, though it seems to be concealed in the matrix, for the edge of the bone is traceable into this. There is no sign of teeth nor of alveoli, so presumably much of the anterior region is missing. The outer face has a rounded ridge along its middle starting from a slight but distinct knob just in front of the rounded condyle. The bone rests upon the rock so that its inner face is not visible, but on another mandible which is free this face is smooth. The upper and lower margins are entire except for a short distance along the latter edge.

– 4 –
Picture icon

Fig. 1.—Photograph of dorsal surface with left side of the larger cranium. × ½.
Fig. 2.—Photograph of side of the same (right side). × ½.

br.: exposed part of brain outlined by dotted line. cr.: Sagittal crest. exoc.: exoceipital condyle of left side. la.: lambdoidal crest. pa.: parietal. 8.occ.: supra-occipital.

– 4 –
Picture icon

Fig. 3.—Outline of skull of Prozeuglodon atrox with cranial region of Lophocephalus shaded to indicate the relative size of the completed skull. × ½. fr.: frontal bone; ju.: jugal; la.: lambdoidal crest; max.: maxilla; na.: nasal; nos.: nostril; pa.: parietal; pmx.: premaxilla; sag.: Sagittal crest; sq.: squamosal; zy.: zygoma.

– 4 –
Picture icon

Fig. 4.—Mandible—outer surface (× ½) with end view A. co.: condyle; ma.: matrix covering the coronoid.
Fig. 5.—Photograph of piece of Clarendon rock (Block A) with four teeth in Situ. × ½.
Fig. 6.—The crown of the middle tooth (nat, size) showing characteristic sculpturing of interrupted ridges. The root has been added from the tooth on the right of above figure.
Fig. 7.—Photograph of another piece of rock (Block C) to show the separate and divergent fangs, well seen on the right, where the impression of the two fangs are seen: (a) cementum broken from side of one fang. (b) impression of the second fang. c. d, broken fangs of the middle tooth. (Xat. size.)

– 5 –
Dimensions (cms.).
Length 33.5
Height 14
Height at broken end 5.5
Thickness at broken end 3.5

Caudal vertebra.

From Milton limestone. (Mus. regist.: no. C.'09.5.) The centrum alone is preserved and is much compressed laterally and distorted as well as being somewhat water-or air-worn. There is no clear evidence of the bases of neural arch nor of transverse processes, hence it is probably from the hinder end of the series.

Dimensions (cms.).
Length 11.5
Height of end 10.5
Width (compressed) 6.5

Certain fragments.

In the course of my investigations I received in 1936 a number of lumps of limestone from Waimate, kindly sent to me by Mr. J. A. Hurst, formerly in the Department of Geology. The material contains many fragments of Cetacean bones, terribly mixed together, higgledy-piggledy, before preservation; and this confusion was rendered still greater by the modern method of quarrying by means of explosive; so that it has been impossible to find sufficient fragments to enable one to identify the parts.* But there are three cases of vertebrae from the caudal region that have been split in two, so that measurements can be made; one has a length of 10 cm., with the terminal faces of the centrum 12.5 cm. or perhaps slightly more in diameter. A second centrum is 7 cm. in length, with a diameter of 10 cm., the ends being nearly circular.

These vertebrae are closely comparable in dimensions with the single centrum from Milburn, which is 11.5 cm. long and has a distorted diameter of 6.5 cm. We may, I think, safely identify them as belonging to Lophocephalus. Moreover, amongst the broken bones are two pieces which fit together and are, I believe, parts of the mandible of this genus.

In four of the blocks there are portions of teeth; in one the imperfect crowns are visible, and in another the broken roots. The roots of the molars are not united as they are in Prosqualodon, but are separate, rather widely divergent. The crowns so far as they are exposed are partly embedded in the matrix; they are about 2 cm. in height, and there is evidence of 4 sharp cusps, sharper than in the Clarendon teeth, but this may be due to their youth: they are about the same dimensions. Most of the enamel has been broken away so that the character of the sculpturing cannot be fully made out, but in the impression left by one tooth in the matrix, it seems to be similar to that of the Clarendon teeth. In two or three cases

[Footnote] * Amongst these bones are extremely well-preserved fragments of Palaeeudyptes, tibia and femur being recognisable.

– 6 –

I excavated the rock so as to trace the fangs for some distance to confirm the statement as to their character. Unfortunately the teeth are so much fractured and so imperfect that it is impossible to give drawings or their dimensions; hence I must rely on the Clarendon teeth for a fuller account of the teeth of Lophocephalus.

The material from Clarendon consists of three pieecs of rock containing teeth exposed in various ways.

Block A bears the imprint of a bone traversed by a shallow wide groove, which I regard as portion of a maxilla. This measures 17.6 cm. in length and has a width of 2.3 cm. in the middle. It carries a series of six teeth, of which the foremost is the upper part of a broken root with two fangs united for the short distance exposed. This I take for a premolar, as its antero-posterior length is only 1.5 cm. in contrast to the 2.2 cm. of the other teeth which may be regarded as molars.

The three molars as shown in the figure 5 are followed by two other crowns which were concealed by a piece of matrix I removed; hence there are apparently five molar teeth.

The crown of the fully exposed teeth—exposed by a lucky fracture of the stone in their plane—presents the usual character of the shark-toothed whales or Squalodonts; that is, the triangular crown has on each of its two edges a series of low cusps. The enamel is yellowish rather than brown and presents a corrugated surface especially near the base, produced by a number of discontinuous longitudinal ridges, some coarser, others finer, filling up the gaps between the others, giving rise to a sculpturing totally different from the simpler type of Prosqualodon hamiltoni and more like that of P. davidi but less coarse than Flynn's photograph indicates and without the little spines. Each of the sloping edges is produced into low rounded cusps, which, like the apex, is worn down to a small platform. In the first molar there are four such cusps on the anterior and only three on the posterior edge; in the following teeth, four on each edge: in the third molar they are more pronounced, longer, and more deeply separated from each other than in the previous teeth, the apex is here broken across just above the upper cusp. The two most posterior teeth, not shown in the photograph, become smaller, the last the smallest.

A seventh tooth, of a different series, either of the lower or upper, lies obliquely below the penultimate tooth of the upper series; the crown is embedded, the apex directed forwards.

The roots, as far as they remain in this block, are two-fanged and connected at any rate for a little distance from the crown by cementum which is as usual traversed by a groove, but they appear to be separate lower down in the middle tooth.

Block B is 19 cm. in length and bears evidence of six teeth in series also, of which, however, only the broken roots project from the rock, the crowns being embedded. In front of these roots there is a conical pit left by the crown of what is probably the canine. Longitudinal ridges leave their impress, and there are no cusps. At the other end, which may be taken as the posterior end of the block, is the hollow left by the upper part of a crown showing clear evidence

– 7 –

of cusps and evidently a molar. Of the four broken roots the first or most anterior has a single fang, as has also the second. The third has apparently a single fang, but I am rather doubtful. The fourth has definitely a double fang—the two are united for the very short exposed portion.

I read the series as canine, 3 premolars, and 2 molars. Between the third and fourth is the impress of a portion of a crown of an upper series, if this be the lower; and behind the fourth is a long, narrow, curved furrow crossing the series. It shows no sign of sculpture, and I feel doubtful as to whether it is the impress of a tooth or that of one fang of an incisor.

In Block C, whose length is but 11 cm., there is preserved (a) the impress of a crown of a molar with two clearly separate fangs, (b) the crown of a second molar likewise with separate fangs, though they are broken across, and (c) the impress of two long, curved, separate and divergent fangs of almost the entire length.

I do not know, though I suspect, that these three pieces of rock were found at the same time; whether they are parts of the same animal or not I cannot say; they do not fit with one another in any way. There is evidence then of 5 molars (in A), 3 premolars (in B) and a canine (in B), and that the root fangs were separate and divergent (in C).

I think I am justified in associating these teeth with those from Waimate, and attributing them to Lophocephalus parki.

Of the four fragments of teeth figured, though not described, by Andrew* one of them (fig. 1e) has widely divergent fangs and lacks a crown and much of the fang; the longest of the fangs measures 5.5 cm., with a width of 3.5 cm., the greatest width being in the upper part. This may be attributed to the same whale.


The Zeuglodonts hitherto recorded have come from rocks of Eocene age, whereas these relics were found in those of Oligocene, even in the upper part of that period, and were in New Zealand contemporary with the Squalodonts, which elsewhere, so far as I have been able to ascertain, occur in the Miocene.

I am not in a position to decide whether the Zeuglodonts are or are not Cetacean. This has been denied by such capable zoologists as Abel, D'Arcy Thompson and others, and affirmed by equally competent judges. I have therefore followed the usual practice amongst zoological systematists of placing them as Archaeoceti, linking the rest of the Cetacea with the terrestrial Creodonts, with which they agree in their dental formula, but from which they differ in the existence of an elongated rostrum, a thoroughly characteristic Cetacean structure, except that the nostril is not basally situated as in the Odontoceti, but is much nearer the end of the rostrum; while the cranial region is similar to that of the terrestrial mammals, and shows no “telescoping” of the bones except that the ascending process of the maxilla overlaps the anterior portion of the frontal.

[Footnote] * A. Andrew, “The Geology of the Clarendon Phosphate Deposits,” Trans. N.Z. Inst., vol. 38, 1905.