C'est celle d'un petit carnassier dont les mâchelières ressemblent beaucoup à celles des sarigues; mais il y a dix de ces dents en série, nombre que ne montre aucun carnassier connu." (Oss. Foss. 111. 349. note). It is to be regretted that the particular data, with the exception of the number of the teeth, on which Cuvier based his opinion, were not detailed; but he must have been well aware that the grounds of his belief would be obvious, on an inspection of the fossil, to every competent anatomist: it is also to be regretted that he did not assign to the fossil a generic name, and thereby have prevented much of the reasoning founded on the supposition that he considered it to have belonged to a true Didelphys.

Mr. Owen then proceeded to describe the structure of the jaw; and he stated that having had in his possession two specimens of the Thylacotherium Prevostii, belonging to Dr. Buckland, he has no hesitation in declaring that their condition is such as to enable any anatomist, conversant with the established generalizations in comparative osteology, to pronounce therefrom not only the class, but the more restricted group of animals to which they have belonged. The specimens plainly reveal, first, a convex articular condyle; secondly, a well-defined impression of what was once a broad, thin, high, and slightly recurved, triangular, coronoid process, rising immediately anterior to the condyle, having its basis extended over the whole of the interspace between the condyle and the commencement of the molar series, and having a vertical diameter equal to that of the horizontal ramus of the jaw itself: this impression also exhibits traces of the ridge leading forwards from the condyle and the depression above it, which characterises the coronoid process of the zoophagous marsupials; thirdly, the angle of the jaw is continued to the same extent below the condyle as the coronoid process reaches above it, and its apex is continued backwards in the form of a process; fourthly, the parts above described form one continuous portion with the horizontal ramus of the jaw, neither the articular condyle nor the coronoid being distinct pieces, as in reptiles. These are the characters, Mr. Owen believes, on which Cuvier formed his opinion of the nature of the fossil; and they have arrested the attention of M. Valenciennes, in his endeavours to dissipate the doubts of M. de Blainville.1

"From the examination of a cast the latter, however, has been induced to infer that there is no trace of a convex condyle, but in place thereof an articular fissure, somewhat as in the jaws of fishes; that the teeth, instead of being imbedded in sockets, have their fangs confluent with or anchylosed to the substance of the jaws, and that the jaw itself presents evident traces of the composite structure.

"In answer to the first of these positions, Mr. Owen states that the portion of the true condyle which remains in both the specimens of Thylacotherium examined by Cuvier and M. Valenciennes, clearly shows that the condyle was convex, and not concave. It is situated a little above the level of the grinding surface of the teeth, and projects beyond the vertical line dropped from the extremity of the coronoid process, but not to the same extent as in the true Didelphys. In the specimen examined by M. Valenciennes, the condyle corresponds in position with that of the jaw of the Dasyurus rather than the Didelphys; it is convex, as in mammiferous animals, and not concave as in oviparous. The entire convex condyle exists in the specimen belonging to the other genus, Phascolotherium, now in the British Museum, but formerly in the cabinet of Mr. Broderip. Mr. Owen is of opinion that the entering angle or notch, either above or below

1

seq.

'Comptes Rendus,' 1838; Second Semestre, No. 11, Sept. 10, p. 527 et

the true articular condyle, has been mistaken for "une sorte d'échancrure articulaire, un peu comme dans les poissons.'

"The specimen of the half-jaw of the thylacothere examined by M. Valenciennes, like that which was transmitted to Cuvier, presents the inner surface to the observer, and exhibits both the orifice of the dental canal and the symphysis in a perfect state. The foramen in the fossil is situated relatively more forward than in the recent opossum and dasyure, or in the placental Insectivora, but has the same place as in the marsupial genus Hypsiprymnus. The symphysis is long and narrow, and is continued forward in the same line with the gently convex inferior margin of the jaw, which thus tapers gradually to a pointed anterior extremity, precisely as in the marsupial Insectivora. In the relative length of the symphysis, its form and position, the jaw of the Thylacotherium precisely corresponds with that of the Didelphys.

"In addition, however, to these proofs of the mammiferous nature of the Stonesfield remains, and in part of their having belonged to Marsupialia, Mr. Owen stated that the jaws exhibit a character hitherto unnoticed by the able anatomists who have written respecting them, but which, if co-existent with a convex condyle, would serve to prove the marsupial nature of a fossil, though all the teeth were wanting.

"In recent marsupials the angle of the jaw is elongated and bent inwards in the form of a process, varying in shape and development in different genera. In looking, therefore, directly upon the inferior margin of the marsupial jaw, we see, in place of the edge of a vertical plate of bone, a more or less flattened triangular surface or plate of bone extended between the external ridge and the internal process or inflected angle. In the opossum this process is triangular and trihedral, and directed inwards with the point slightly curved upwards and extended backwards, in which direction it is more produced in the small than in the large species of Didelphys.

"Now, if the process from the angle of the jaw in the Stonesfield fossil had been simply continued backwards, it would have resembled the jaw of an ordinary placental carnivorous or insectivorous mammal; but in both specimens of Thylacotherium, the half-jaws of which exhibit their inner or mesial surfaces, this process presents a fractured outline, evidently proving that when entire it must have been produced inwards or mesially, as in the opossum.

"Mr. Owen then described in great detail the structure of the teeth, and showed, in reply to M. de Blainville's second objection, that they are not confluent with the jaw, but are separated from it at their base by a layer of matter of a distinct colour from the teeth or the jaw, but evidently of the same nature as the matrix; and secondly, that the teeth cannot be considered as presenting an uniform, compressed, tricuspid structure, and being all of one kind, as M. de Blainville states, but must be divided into two series as regards their composition. Five, if not six, of the posterior teeth are quinque-cuspidate, and are molares veri; some of the molares spurii are tricuspid and some bicuspid, as in the opossums. An interesting result of this examination is the observation that the five cusps of the tuberculate molars are not arranged, as had been supposed, in the same line, but in two pairs placed transversely to the axis of the jaw, with the fifth cusp anterior, exactly as in the Didelphys, and totally different from the structure of the molars in any of the Phoca, to which these very small Mammalia have been compared and in reference to this comparison, Mr. Owen again calls attention to the value of the character of the process continued from the angle of the jaw, in the fossils, as strongly contradistinguishing them from the Phocide, in none of the species of which is the angle of the jaw so produced. The Thylacotherium differs from the genus Didelphys in the greater

:

number of its molars, and from every ferine quadruped known at the time when Cuvier formed his opinion respecting the nature of the fossil. This difference in the number of the molar teeth, which Cuvier urged as evidence of the generic distinction of the Stonesfield mammiferous fossils, has since been regarded as one of the proofs of their saurian nature; but the exceptions by excess to the number seven, assigned by M. de Blainville to the molar teeth in each ramus of the lower jaw of the insectivorous Mammalia, are well established and have been long known. The insectivorous chrysochlore, in the order Fere, has eight molars in each ramus of the lower jaw; the insectivorous armadillos have not fewer; and in one subgenus (Priodon) there are more than twenty molar teeth on each side of the lower jaw. The dental formulæ of the carnivorous Cetacea, again, demonstrate the fallacy of the argument against the mammiferous character of the Thylacotherium founded upon the number of its molar teeth. From the occurrence of the above exceptions in recent placental Mammalia, the example of a like excess in the number of molar teeth in the marsupial fossil ought rather to have led to the expectation of the discovery of a similar case among existing marsupials; and such an addition to our zoological catalogues has, in fact, been recently made. In the Australian quadruped described by Mr. Waterhouse under the name of Myrmecobius, an approximation towards the dentition of the Thylacotherium is exemplified, not only in the number of the molar teeth, which is nine on each side of the lower jaw in the Myrmecobius, but also in their relative size, structure, and disposition. Lastly, with respect to the dentition, Mr. Owen says it must be obvious to all who inspect the fossil, and compare it with the jaw of a small Didelphys, that contrary to the assertion of M. de Blainville, the teeth and their fangs are arranged with as much regularity in one as in the other, and that no argument of the saurian nature of the fossil can be founded on this part of its structure.

“With respect to M. de Blainville's assertion that the jaw is compound, Mr. Owen stated, that the indication of this structure near the lower margin of the jaw of the Thylacotherium is not a true suture, but a vascular groove, similar to that which characterises the lower jaw of Didelphys, opossum, and some of the large species of Sorex."

"Dec. 9, 1838.-A paper on the Phascolotherium, being the second part of the " Description of the remains of marsupial Mammalia, from the Stonesfield slate," by Richard Owen, Esq., F.G.S., was read.

“Mr. Owen first gave a brief summary of the characters of the Thylacotherium, described in the first part of the memoir, and which he conceives fully prove the mammiferous nature of that fossil. He stated that the remains of the split condyles in the specimen demonstrate their original convex form, which is diametrically opposite to that which characterises the same part in all reptiles and all ovipara;-that the size, figure and position of the coronoid process are such as were never yet witnessed in any except a zoophagous mammal endowed with a temporal muscle sufficiently developed to demand so extensive an attachment for working a powerful carnivorous jaw ;-that the teeth, composed of dense ivory, with crowns covered with a thick coat of enamel, are everywhere distinct from the substance of the jaw, but have two fangs deeply embedded in it;-that these teeth, which belong to the molar series, are of two kinds; the hinder being bristled with five cusps, four of which are placed in pairs transversely across the crown of the teeth, and the anterior or false molars, having a different form, and only two or three cusps-characters never yet found united in the teeth of any other than a zoophagous mammiferous quadruped;-that the general form of the jaw corresponds with the preceding more essential indications of its mammiferous nature. Fully impressed with the value of these characters, as determining the class to which the fossils belonged, Mr. Owen stated

that he had sought in the next place for secondary characters which might reveal the group of Mammalia to which the remains could be assigned, and he had found in the modification of the angle of the jaw, combined with the form, structure, and proportions of the teeth, sufficient evidence to induce him to believe that the Thylacotherium was a marsupial quadruped.

"Mr. Owen then recapitulated the objections against the mammiferous nature of the thylacotherian jaws, from their supposed imperfect state, and repeated his former assertion, that they are in a condition to allow of these characters being fully ascertained: he next reviewed, first, the differences of opinion with respect to the actual structure of the jaw; and secondly, to the interpretation of admitted appearances.

"1. As respects the structure. It has been asserted that the jaws must belong to cold-blooded Vertebrata, because the articular surface is in the form of an entering angle; to which Mr. Owen replies that the articular surface is supported on a convex condyle, which is met with in no other class of Vertebrata except in the Mammalia. Again, it is asserted that the teeth are all of an uniform structure, as in certain reptiles; but, on reference to the fossils, Mr. Owen states it will be found that such is not the case, and that the actual difference in the structure of the teeth strongly supports the mammiferous theory of the fossils.

“2. With respect to the argument founded on an interpretation of structure which really exists, the author showed that the Thylacotherium having eleven molars on each side of the lower jaw, is no objection to its mammiferous nature, because among the placental Carnivora, the Canis Megalotis has constantly one more grinder on each side of the lower jaw than the usual number; because the Chrysochlore among the Insectivora has also eight instead of seven molars in each ramus of the lower jaw; and the Myrmecobius, among the Marsupialia, has nine molars on each side of the lower jaw; and because some of the insectivorous armadillos and zoophagous Cetacea offer still more numerous and reptile-like teeth, with all the true and essential characters of the mammiferous class. The objection to the false molars having two fangs Mr. Owen showed was futile, as the greater number of the spurious molars in every genus of the placental Fere have two fangs, and the whole of them in the Marsupialia. If the ascending ramus in the Stonesfield jaws had been absent, and with it the evidence of their mammiferous nature afforded by the condyloid, coronoid and angular processes, Mr. Owen stated that he conceived the teeth alone would have given sufficient proof, especially in their double fangs, that the fossils do belong to the highest class of animals.

"In reply to the objections founded on the double fangs of the Basilosaurus, Mr. Owen said that the characters of that fossil not having been fully given, it is doubtful to what class the animal belonged; and, in answer to the opinion that certain sharks have double fangs, he explained that the widely bifurcate basis supporting the tooth of the shark, is no part of the actual tooth, but true bone and ossified parts of the jaw itself, to which the tooth is anchylosed at one part, and the ligaments of connexion attached at the other. The form, depth and position of the sockets of the teeth in the Thylacotherium are precisely similar to those in the small opossums. The colour of the fossils, Mr. Owen said, could be no objection to those acquainted with the diversity in this respect, which obtains in the fossil remains of Mammalia. Lastly, with respect to the Thylacotherium, the author stated that the only trace of compound structure is a mere vascular groove running along its lower margin, and that a similar structure is present in the corresponding part of the lower jaw of some species of opossum, of the wombat, of the Balana antarctica, and of the Myrmecobius, though the groove does not reach so far forward in this animal; and that a similar

groove is present near the lower margin, but on the outer side of the jaw, in the Sorex Indicus.

:

"Description of the half jaw of the Phascolotherium.—This fossil is a right ramus of the lower jaw, having its internal or mesial surface exposed. It once formed the chief ornament of the private collection of Mr. Broderip, by whom it has since been liberally presented to the British Museum. It was described by Mr. Broderip in the 'Zoological Journal,' and its distinction from the Thylacotherium clearly pointed out. The condyle of the jaw is entire, standing in bold relief, and presents the same form and degree of convexity as in the genera Didelphys and Dasyurus. In its being on a level with the molar teeth, it corresponds with the marsupial genera Dasyurus and Thylacynus, as well as with the placental Zoophaga. The general form and proportions of the coronoid process closely resemble those in zoophagous marsupials; but in the depth and form of the entering notch between the process and the condyle, it corresponds most closely with the Thylacynus. Judging from the fractured surface of the inwardly reflected angle, that part had an extended oblique base, similar to the inflected angle of the Thylacynus. In the Phascolotherium, the flattened inferior surface of the jaw, external to the fractured inflected angle, inclines outwards at an obtuse angle with the plane of the ascending ramus, and not at an acute angle, as in the Thylacyne and Dasyurus; but this difference is not one which approximates the fossil in question to any of the placental Zoophaga on the contrary, it is in the marsupial genus Phascolomys where a precisely similar relation of the inferior flattened base to the elevated plate of the ascending ramus of the jaw is manifested. In the position of the dental foramen, the phascolothere, like the thylacothere, differs from all zoophagous marsupials and the placental Fera; but in the Hypsiprymnus and Phascolomys, marsupial Herbivora, the orifice of the dental canal is situated, as in the Stonesfield fossils, very near the vertical line dropped from the last molar teeth. The form of the symphysis, in the Phascolotherium, cannot be truly determined; but Mr. Owen is of opinion that it resembles the symphysis of the Didelphys more than that of the Dasyurus or Thylacynus. "Mr. Owen agrees with Mr. Broderip in assigning four incisors to each ramus of the lower jaw of the Phascolotherium, as in the Didelphys; but in their scattered arrangement they resemble the incisors of the Myrmecobius. In the relative extent of the alveolar ridge occupied by the grinders, and in the proportions of the grinders to each other, especially the small size of the hindermost molar, the Phascolotherium resembles the Myrmecobius more than it does the opossum, Dasyurus, or Thylacynus; but in the form of the crown, the molars of the fossil resemble the Thylacynus more closely than any other genus of marsupials. In the number of the grinders the Phascolotherium resembles the opossum and Thylacynus, having four true and three false in each maxillary ramus; but the molares veri of the fossil differ from those of the opossum and Thylacotherium in wanting a pointed tuberele on the inner side of the middle large tubercle, and in the same transverse line with it, the place being occupied by a ridge which extends along the inner side of the base of the crown of the true molars, and projects beyond the anterior and posterior smaller cusps, giving the quinquecuspid appearance to the crown of the tooth. This ridge, which, in Phascolotherium, represents the inner cusps of the true molars in Didelphys and Thylacotherium, is wanting in Thylacynus, in which the true molars are more simple than in Phascolotherium, though hardly less distinguishable from the false molars. In the second true molar of Phascolotherium, the internal ridge is also obsolete at the base of the middle cusp, and this tooth presents a close resemblance to the corresponding tooth in Thylacynus; but in the Thylacynus the two posterior molars increase in size, while in Phascolothe