that the Trirachodon molar is related to that of the Stormberg fossil in much the same way that the molar of the pig is related to that of the opossum, but it seems impossible that the two animals can be more nearly related than an ungulate and a marsupial.

If this new form is a mammal it must be regarded as the representative of a new order, and if, as I believe, it ought to be regarded as a Cynodont, it will have to be the representative of at least a new family.

The exact horizon of the fossil is unknown. Professor Schwarz writes me that Paballon might mean any horizon from Molteno Bed to the Cave Sandstone, but from the matrix of the fossil it seems highly probable that it comes from the Red Beds. In any case it may be regarded as certain that the fossil is of Lower Jurassic age. This is interesting from the fact that no Cynodont has hitherto been got from beds younger than the Burghersdorp or Upper Triassic.

I suggest for this interesting fossil the name Tritheledon riconoi, g. et sp. nov., and the new family of which it is the type may be called the Tritheledontida.

FIG.

PLATE XXII.

x 2.

30. Side view of fragment of the skull of Tritheledon riconoi.
31. Palatal view of fragment of skull of Tritheledon riconoi. × 2.
32. Front view of ditto. x 2.

33. Restoration of front of skull of Tritheledon riconoi. Nat. size.
34. Anterior view of molar tooth of Tritheledon riconoi. × 6.
35. Palatal view of molar tooth of Tritheledon riconoi.

× 6.

36. Posterior view of posterior molar of Tritheledon riconoi.

× 6.

Fig. 28-29 Tylosaurus capensis, Broom.
Fig. 30-36 Tritheledon riconoi, Broom.

West, Newman lith.

18. On Some Points in the Structure of the Dicynodont Skull.By R. BROOM, M.A., D.Sc.

(Five Text Figs.)

THE skull on which most of the following observations are based was found on Rhenoster Fontein, a farm at the top of the Molteno Pass north of Beaufort West, and brought to the South African Museum by Dr. Kolbe. It was sent in 1902 to Prof. Seeley, and the matrix was most carefully removed under his direction by the British Museum preparators. He had most carefully examined the skull and traced out the sutures, and intended to give a full description, and had prepared for his paper a series of process blocks of drawings of the skull; but apparently he had not gone further. Doubtless other work which he thought more important, such as the study of the skull of Erythrosuchus, came to hand, and death overtook him with his work unfinished. The skull has been returned to the South African Museum, and as the illustrations have been prepared I have been asked to write a description of the skull to accompany the blocks. Though Prof. Seeley has left no manuscript, so far as I am aware, it is possible to get a good idea of his views from his determinations on the figures. In a number of points my views differ from those of Seeley, and where this is the case I shall give his views as well as my own.

Though imperfect as regards the right zygomatic arch and slightly crushed and distorted, it is the finest Dicynodont skull that has ever been obtained. Others have been got more complete but owing to the condition of the bone or matrix it has been impossible to display the bones in such perfection, and no skull has ever shown so many sutures satisfactorily.

As regards the species there is here, as with all Dicynodonts, an element of doubt. In the first place, we do not yet know whether Oudenodon is not really the female of Dicynodon. Owen, Lydekker, and the writer have all considered the question, and all come to the conclusion that probably the genera are distinct, but as time goes on doubts became greater, and at present the evidence on the two sides is about evenly balanced. Lystrosaurus is apparently tusked in both males and females, while Cistecephalus is tuskless in both