Fossil Cetacea of New Zealand III.—The Skull and other Parts of the Skeleton of of Prosqualodon hamiltoni n.sp.
[Read before the Otago Branch, September 8, 1936; received by the Editor, October 28, 1936; issued separately, June, 1937.]
I Name this whale after the late Augustus Hamilton, at one time registrar at the University of Otago and later Director of the Dominion Museum; he was a very keen and observant field naturalist to whom this museum is indebted for much of the material which forms the subject of this and other papers on fossil Cetacea.
In a block of sandstone obtained from the now disused Caversham Quarry in 1902 Mr Hamilton noted a piece of a skull protruding and had the block conveyed to the Otago University Museum; here the matrix was carefully cleared away with the result that the greater part of the skull was revealed and various other parts of the skeleton were recovered from the surrounding debris as the process of excavation went on. Unfortunately at that time I was ignorant of the details of extinct whales, and the taxidermist was not so careful to avoid damage to teeth and parts of the skull as we should be to-day. Nevertheless, the skull proved to be nearly complete, except that the end of the rostrum was broken off before it came into our possession. The bones of the dorsal surface are fairly easily traced, though the sutures of some of them are indistinguishable; but the ventral surface has suffered a good deal of damage, possibly in the attempt to clear it from the surrounding matrix. The skull is that of an aged animal as is evidenced not only by its size but also by the fact that most of the sutures of the cranial region are obliterated.
The genus Prosqualodon of Lydekker (1891) differs only in a few points from the older genus Squalodon, viz., the anterior margin of the nasal bone overhangs slightly the nostril; the rostrum is relatively shorter and the mandible has but a short symphysis; and especially in the fact that the two fangs of the molar teeth are united for the greater part or whole of their length. Prosqualodon is known from two species occurring in the southern hemisphere, namely P. australis Lyd., from South America, and P. davidi Flynn, from Tasmania. The genus represents in the southern hemisphere the genus Squalodon of the north.
The new species differs in sundry features from each of these as will be indicated later in this article; but notably in the character of the sculpturing of the crowns of the teeth, which alone would be sufficient to differentiate it.
Abbreviated description of the Skull.*
A marked feature is the large size of the supra-occipital bone, which has a circular contour anteriorly. It is inclined forwards from the occipital foramen to meet the frontals, as in Squalodon, but in this specimen it forms an angle of about 45° with the back of the skull, which may be due to downward compression during fossilisation, if it be compared with that of P. davidi. Its surface is concave, its margin very prominent, forming a pronounced lambdoidal crest, both laterally and anteriorly; the lateral part of the crest forms a plate some 6 cm. in width overhanging the temporal fossa. Immediately in front of the crest lie the two frontal bones, each consisting as usual of the two regions, a nearly square dorsal region and a much larger moiety which constitutes the supra-orbital plate, covered by the ascending process of the maxilla, as in modern Odontocetes. The nasals are short and nearly square, marked by furrows: the anteriro margin overhangs to a slight but distinct, degree the aperture of the nostrils. The nostrils, as the figure shows, are asymmetrical. The maxilla presents a feature which puzzled me a good deal, but which I think is due to distortion due to pressure from above; its lateral margin overhangs the alveoli and teeth like the cave of a house-roof, for a space of 2 cm. on the right side and much more on the left. As this is quite a unique feature in Squalodonts and in existing Odontocetes, it cannot be natural and examination of the “mesorostral gutter” (Kellogg) shows that the roof of the rostrum has been thrust has been thrust outward, for the side wall, instead of being concave, is pushed out of the perpendicular. The squamosal is a particularly heavy bone, its outwardly directed process forming a large horizontall “shelf” ending in a thick boss of bone from which the zygomatic process arises. This “shelf” is continued forward as a “ledge” alongside the base of the cranial wall, and may be formed partly of the alisphenoid, though no suture is visible.
The under surface of the skull is very imperfect; neither palatine nor pterygoid bones remain, for no doubt during the process of cleaning these delicate bones were scraped away. The “palate” is very convex and is reminiscent rather of the Baleenids than of Odontocetes.
|P. hamiltoni||P. davidi.||P. australis|
|L. of rostrum||24 +||25.5||27|
|Width over supra-occ.||18||—||21|
|Width over over sup. orbital pl.||28||—||36|
|Width over zygoma||34||37.8||40.8|
|L. from supra-occip. to tip of rostrum||38|
[Footnote] * It is not necessary to give a detailed account since this is well known from the writings and figures of Lydekker, Cabrera and Flynn, the last being the most complete skull of the genus. Kellogg gives list of all known spp. of Squalodonts—so it is not necessary to discuss the literature.
As to P. australis, neither Lydekker nor Cabrera give dimensions; I have measured the figure given by the latter (fig. 5) which is one-sixth natural size, and multiplying the measurements by six, have arrived at the figures above.
As to P. davidi, the length is taken from Flynn's account in the Geol. Mag.; the other figures from his account in Nature, but as the length given in that journal is rather greater than his later figure, it may have to be reduced slightly, but this is of little importance.
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
|P. hamiltoni||P. davidi||P. australis|
|Sculpturing of teeth||smooth long ridges||coarse reticulations with “denticles”||stout ridges|
|Rostrum||probably longer than in australis thin||rel. longer than in australis stout||relat. shorter|
|Sup. orb. plate||inclined, so that ant. process is above level of base of rostrum large nearly straight anter.|
|Sup. orb. processes||small||larger||large|
|Lambdoidal crest||circular||?||nearly straight anter.|
|Zygomatic pro.||heavier than in australis|
The tympano-periotic is in position on the right side of the skull, but that of the left side was found in the refuse resulting from clearing it; the tympanic is somewhat broken, it has the general character of a modern Odontocete; the periotic is very like that of Squalodon calvertensis Kellogg.
The teeth. Unfortunately, the crowns of those in the maxilla were broken, but five teeth were found among the debris. There is evidence also of five teeth in the maxilla, broken below the crowns, and the root are embedded in the bone; but on the left side is one tooth the root of which is most fortunately fully exposed, thus establishing the generic position of the skull in that the two fangs are united by cementum for nearly their entire length, only the tips being free. Of the five loose teeth, two have double fangs, united though the root is traversed by a groove: these are premolars. The three others have single fangs; one of them is stouter than the other two and I regard it as a canine, the two remaining being incisors. The complete teeth are 8 cm. long. The crowns of these teeth have a brown enamel, the surface of which is traversed along the entire length by a number of smooth, continuous, longitudinal ridges converging towards the apex; there is no “reticulation” such as occurs in the teeth of P. davidi nor any “denticules” or even slight prominences on these ridges.
The premolars show on one margin three, or in one case four, very low cusps which are much worn down but are evident when the tooth is looked at edge-wise, and in one of the teeth six small
oval areas are visible. The assumed canine likewise has four of these little prominences. The incisors are smooth, and in these the root is curved at an angle with the crown.
The first record of the occurrence of Prosqualodon in New Zealand is found in the memoir by Andrew (1905), where he identifies certain fragments of teeth found by him in the Milburn Quarry as belonging to Squalodon grateloupi. The specific identification is erroneous, as also is the generic, but justifiably. There is no doubt that the three teeth figured (loc. cit., pl. ix, fig. 1a, b, c, d, and f) belong to P. hamiltoni. But that drawn in fig. e belongs, as I shall show, to the new genus Lophocephalus; it has separate and divergent fangs.
The four anterior vertebrae were recovered from the debris, though the fourth is merely represented by the centrum. As hitherto the only cervical vertebra of a Prosqualodon to have been described is the atlas of P. australis (True), it is desirable to give dimensions of these so that when other vertebrae are found isolated it may be possible to identify them as those of this genus, for they are of course much smaller than those of Lophocephalus or of Kekenodon.
The atlas. This is much thicker than in modern Odontocetes such as Globicephalus, as is seen from True's figure for the South American species, in which he describes and figures a large foramen in the neural arch which he regards as that for the transmission of the first spinal nerve. There is no such foramen in the present species, nor can I detect any groove on the anterior side of the neural arch such as exists in some Cetacea. The anterior margin is entire and shows no sign of any groove. The roof “lamina” of the arch is imperfect in the median region posteriorly, but is entire laterlly on each side of the break, and should show the foramen: here the roof measures 4.5 cm. along the (left) side. The neural spine is low and possibly has lost a little piece from the tip; the distance from the upper edge to the inner surface of the arch is 2.5 cm. The hypapophysis arising from the hinder region of the floor of the arch is directed backwards as usual as a small knob. The transverse processes, upper and lower, are also only low knobs, widely separate; the lower one much less developed than the upper, the apex of which is 3 cm. from the inner surface of the arch. The posterior zygapophysis is remarkably extensive as it occupies the whole height of the posterior end of the arch and terminates in a small oval face with a transverse diameter of 2 cm. separate by a wide shallow groove from the arch itself.
Width over prezygapophyses (condylar facets): 12 cm.; height: 8 cm.
Width over the postzygophyses: 10.75 cm.; height: 5 cm.
The axis. This vertebra has not been described for either of the species of this genus nor I think for Squalodon. The neural arch has been somewhat depressed during the process of fossilisation, and the posterior face of the centrum is crushed in slightly. The spine arises from its whole length, its base being 4.8 cm. long; its
posterior dorsal region is broken, but only a small piece is missing; its upper edge is 7.3 cm. from floor of the canal. The odontoid process is conical, with a length of 2 cm. rising from a circular base, 4.1 cm. in diameter. This is in contrast to the feeble process in modern Odontocetes. On each side of it the prezygapophysis is a vertically extended flat area with a height of 6 cm., its upper 3 cm. being on the arch; this portion is 3 cm. in diameter. The neural arch is 3.5 cm. across at the level of the transverse process, the roof is 4 cm. in length. The transverse process is single, rises quite low down on the side of the centrum and is directed outwards and slightly backwards. It measures 4.5 cm. from its tip along the hinder margin and has a thickness dorso-ventrally of 2.5 cm. There is no trace of an upper process, and therefore no vertebrarterial canal. A small hypapophysis springs from the hinder end of the centrum.
The third vertebra. This has suffered a good deal of distortion, especially at the anterior face of the centrum, owing to the flattening of the neural arch, which is cracked in its anterior region. The epiphysis is lacking at this end, but is present on the posterior face. The transverse process springs from a point low down at the side and is directed outwards. It is somewhat flattened antero-posteriorly. It seems to be complete on the left side, but that on the right side has its tip broken. There is no evidence of an upper transverse process and therefore no vertebrarterial canal. The process is 4.3 cm. long with dorso-ventral diameter at the base of 2.6 cm.
The fourth vertebra. Only the centrum remains with the base of the neural arch; it lacks both epiphyses.
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
|Atlas||Axis||3rd vert.||4th vert.|
|Height to top of neural spine||9.7||12.3||7.5|
|Width over tips trans. proc.||12.3||14.3||14|
|Length||—||5 (incl odont.)||2.5||2|
|Vertical diam. poster.||4.6||4.5||5|
|Horizontal diam. poster.||6.3||6.7||6.5|
|Length of lamina||4.5||4||2|
|Length of pedicle (side)||5.25||2.3||1.75||1|
|Length of floor||3.25||—||—|
|Neural canal; trans. diam.||5.5||4||4.5|
|Neural canal; vert. diam.||5||distorted||distorted|
|Transverse process, length.||—||4.3||4.3|
The scapula has not been described for any Squalodont, so far as I have been able to ascertain. In the present whale the right scapula was excavated from the matrix. It still lies on the rock with its inner surface upwards; but the rock has been cleared away from below the acromion and glenoid fossa in order to study the contours. It is more like that of the Baleenids than of modern Odontocetes (a) in the small size of the coracoid process, and (b) in the relation of the acromion to the blade.
The surface of the bone near the supra-scapular border has become split away from the bone of the outer surface, leaving this on the matrix. The supra-scapular border itself is a good deal broken so that the real dimensions are greater than the figures given below. The anterior or coracoidal border has a rounded edge and forms a gentle curve to the coracoid process. The acromion springs from the margin though it lies at a slightly different plane from the blade-bone, there being a drop of about one centimetre; it is a broad plate with parallel upper and lower margins, its end is damaged though its length is measurable. The glenoid or posterior border takes a sweeping curve, convex outwards, then straightening out to reach the glenoid fossa. The coracoid process is a mere rounded knob with the upper surface slightly rubbed, though not sufficiently to conceal its size. The glenoid fossa is oval and quite shallow.
|Greatest width from glenoid to sup.-sc. border||25|
|Greatest height from glenoid to sup.-sc. border||17|
|Distance upper edge from sup.-sc. border||7|
|Distance, lower edge from coracoid||5|
Locality: Caversham Quarry, Dunedin.
Geological horizon: Upper Oligocene (Hutchinsonian).
Other material from Milburn of Upper Oligocene age (it is, according to Marwick in a letter, “not securely placed, but somewhere from Ototaran Hutchinsonian”).
A mid-lumbar vertebra.
(Reg. no. C. 13.2). Both transverse processes are broken. They arise from about half-way down the side of the centrum. The neural arch is broken away across its base; its base is 3.7 cm. in length, and the width of the canal is only 1.25 cm. Haemal processes are well marked and separate, there is a space of 4 cm. across the pair.
|Distance from neural arch||3|
(Reg. no. C. 20.12). Five of these isolated, together with two smaller ones embedded in the matrix alongside the largest of them. They all have epiphyses fused to the centrum.
The smallest has a length of only 2 cm. with diameter of 4 cm. As none of these show any trace either of transverse process or of neural arch they are evidently some of the most posterior caudals. Several other caudal vertebrae from Milburn are in the collection.
Intracranial casts of brain.
Further relics of Prosqualodon are the intracranial casts of the brain. Two of these casts are in the Museum. The larger one shows the general shape with even casts of the blood vessels coursing in the pia mater, especially well shown along each side of the hemispheres. That this one does belong to Prosqualodon is rendered very probable by the fact that I had a wax cast made of the fossil, and though owing to the damaged condition of the skull details are obscure, yet the size of this wax cast is closely similar to that of the stone cast.
There is no record as to the provenance of this cast.
A smaller cast obtained from Milburn in 1934 may come from a smaller individual or from some other whale. It is not sufficiently well preserved for any details to be made out.
Other museums have some relics of this whale, but I intend to publish a list of these in a later paper.
Andrew, A., 1905. On the Geology of the Clarendon Phosphate Deposits, Trans. N.Z. Inst., vol. 38, p. 447.
Flynn, T., 1920. Squalodont Remains from the Tertiary Strata in Tasmania, Nature, vol. 106, p. 406.
— 1923. A Whale of Bygone Days, Austral. Mus. Mag., vol. 1, no. 9, p. 266. (A popular account.)
— 1932. A New Species of Fossil Cetacean from Tasmania, Geol. Mag., vol. 69, p. 327.
Kellogg, R., 1923. Description of Two Squalodonts recently discovered in the Calvert Cliffs, Maryland; and notes on the Shark-toothed Cetacea, Proc. U.S. Nat. Mus., vol. 62, art. 16, p. 39.
Lydekker, R., 1891. Cetacean Skulls from Patagonia, Annales Mus. d. l. Plata, Palaeontolog. Argentina, vol. 2, art. 2, p. 8.
— 1923. On a Skull of a Shark-toothed Dolphin from Patagonia, Proc. Zool. Soc., p. 919.
Cabrera, A., 1926. Cetaceos Fossiles de l Museo de la Plata, Revista d. Mus. d. l. Plata, vol. 29, p. 363.
True, F. W., 1909. A New Genus Fossil Cetacean from the Santa Cruz Tertiary, Patagonia; and description of the mandible and vertebrae of Prosqualodon, Smithsonian Miscell. Publications, vol. 52, p. 441, p. 45.
Fig. 11.—The atlas vertebra seen from the left side. (× 1).
Fig. 4.—The root of a molar tooth found in situ on the left side of the skull. (× 1).
Fig. 13.—The axis vertebra, from the right side, showing the odontoid process. Photograph (× 1).
Fig. 5.—A premolar with root entire. Photograph (rather more than nat. size).
cr.: broken crown of molar tooth. [ unclear: ] hypophysis. od.: odontold process of axis. [ unclear: ] postzygapophysis. [ unclear: ] root of molar tooth. tr. and [ unclear: ] upper and lower transverse processes.
Fig. 6.—A premolar with fractured root. 6A.—The crown of this premolar viewed from the edge to show the small worn cusps. (× 1).
Fig. 7.—A nearly complete incisor. (× 1).
Fig. 8.—Another incisor, the crown broken and the root distorted. (× 1).
Fig. 9.—The probable canine tooth. (× 1).
Fig. 14.—The scapula (× ½). Along the supra-scapular border flakes of bone have been splintered away from the exposed surface. The stone has been removed from below the acromion and glenoid, but the rest of the scapula is still resting on the matrix, though it is not indicated in the drawing. [ unclear: ] acromion. [ unclear: ] coracoid process. [ unclear: ] glenoid fossa. [ unclear: ] bone from which flakes have been split off.]