152 ON THE SKULL AND DENTITION OF TRITYLODON LONGAVUS. 
Dr. Exon said that, as Curator of the Museum at Bloemfontein, 
many very interesting forms came under his notice. The specimen 
was found with Batrachian fossils of Triassic genera, and the matrix 
was quite similar in each case. In association with the species 
was a Ceratodus, a form found in the Triassic rocks of India. The 
fossils were found within a radius of 50 or 60 miles. 
Prof. T. Ruperr Jones said that Messrs. Wylie, Dunn, and Green 
had separated the Upper beds of the Triassic Series in South 
Africa from the Karoo beds under the name of the Stormberg beds. 
The Karoo beds contain Dicynodont remains. In the Stormberg 
beds reptilian remains occur, but without Dicynodonts. The fossil 
appeared to have come from the Stormberg beds, and was therefore 
within the great Triassic formation of South Africa. Palzoniscoid 
fishes had been found in the Karoo beds, and an allied form in the 
Stormberg beds, and these may possibly indicate a Permian age 
for the beds; but the balance of the evidence was in favour of 
the Triassic age of the fossil. 
Mr. Branrorp said that in Africa, Australia, and India a great 
series of beds, without marine fossils, had been found, the age of 
which was chiefly based on the uncertain evidence of plants and 
land or freshwater vertebrates. He thought it unwise to call the 
fossil in question Triassic. The beds in India which contain Cera- 
todus are now considered post-[riassic, and the genus is still living. 
He deprecated too hasty conclusions as to the age of the beds from 
the evidence of plants and terrestrial fossils. 
The Avrnor said all agreed with him that the skull is Mamma- 
lian. As tothe question ‘whether it is Marsupial or not, the size of 
the brain, so far as it could be determined, pointed to Marsupial 
analogies. The characters of the lacrymal foramen pointed in the 
opposite direction. The double incisors were paralleled in certain 
Rodents (Leporide) end gaone the Marsupials in Phascolomys; in 
Macropodide they are ° = : ; but in most of that order they are more 
numerous than in Placentals. The molars, by the number of the 
“true” ones, reminded one rather of the marsupial than of the 
placental forms. The nearest analogies to these molars in pattern 
of grinding-surface were found in the extinct genera Microlestes and 
Stercognathus. 
