The South African fossils with broad, flat, tuberculate tooth-crowns of mammalian type all from the eastern part of Cape Colony. Some of the most interesting are known from fragments, which indicate nothing but the middle region of the sail. They are apparently extremely rare. Two species, with teeth well preserved, found ten years ago by Dr. Kannemeyer, at Wonderboom, and presented to South African Museum, Cape Town, where they were brought under my notice by Peringuey. Others were found in a fragmentary condition by Mr. Alfred of Aliwal North, to the west of that town, in a bed which appeared to me be reconstructed. There is no doubt that the fossils are from the upper part of te Karroo formation, probably of Permian age, and below the Stormberg beds, in Saurischian fossils are found allied to those of the Trias of Europe. If the teeth had occurred isolated, without the means of demonstrating their rsemblance to Theriodonts, by comparison of what remains of the skull, it would have legitimate to have referred them to Mammals. There is no evidence of affinity Accept resemblances to Theriodonts, which show that the skull had pre-frontal and frontal bones, and therefore may be inferred to have had the lower jaw composite. The teeth are such as might be expected, perhaps, in a Monotreme Mammal, their interest is therefore the greater that there is no ground for suspecting them be mammalian, other than a general resemblance of the crown to the crowns of true teeth of
Ornithorhynchus.
That teeth of this type should occur in a group animals in which the shoulder-girdle and pelvis have monotreme characters, and in hich the principal limb bones are intermediate in character between Monotremes Marsupials, is evidence of a closer approximation between Mammals and Pep tiles has been manifest. And so far as I am aware the only Theriodont characters remaining to distinguish these animals from Mammals are the composite lower which is covered externally by the dentary bone along its whole length, and the resence of the pre-frontal, post-frontal, and transverse bones in the skull.
A proposed procedure for the analysis of large phytosociological data sets in the classification of South African grasslands