In 1888, in the ‘Phil. Trans,' B, vol. 179, I described some parts of the skeleton of Theriodesmus phylarchus . Its geological horizon was then unknown. In the following year I visited the locality where it was found, at Klipfontein, near Fraserburg, in Cape Colony, on the summit of the volcanic series of rocks the escarpment of which forms the Neiuwveldt range. As the base of the Karroo rocks at Prince Albert is fully sixty miles south, and the same strata dip towards the north, over all that distance, with some minor undulations in the southern part of the area, while the elevation of the surface of the country augments northward, Klipfontein holds a high position in the deposits of Permian age which are comprised in the Karroo formation. I had the advantage of the company of the late Mr. Thomas Bain, who conducted me to the farm of two brothers, E. and O. Erasmus, where the Theriodesmus was collected. But after an interval of more than twelve years, the effects of atmospheric forces in breaking and removing the surface rocks made it impossible to discover additional remains of that animal. We carefully explored the banks and bed of a dried stream which drains into the Orange River, where I met with fragments of two species of the ganoid fish, Atherstonia , which have been regarded by Mr. A. Smith Woodward as new species, but found nothing resembling Theriodesmus . Along that stream I found and collected fragments of two skeletons which appear to me to be referable to Theriodonts, though their position in the group is not certain. These fossils show the association of limb bones and vertebrae; and with them I found fragments of skulls, which may have belonged to the same animals; but the association should be made with caution, since it is unsafe to put together disconnected fragments of skeletons which may have been associated by transport, unless there is corroborative evidence that they belong to the same type of organization. These small animals in the character of the humerus are Marsupial in type. This character being associated with carnivorous dentition among existing animals, led me to anticipate the discovery of a skull showing canine teeth. Such dentition though poorly preserved, I found near by, in the bed of the river, associated with the back of the skull.