Professor S. G. Seeley — On Ceratodus. 543' 



All branches of science must necessarily pass through three 

 stages — (1) the observation of facts; (2) the enunciation of theories 

 to explain the facts ; (3) the discovery of laws governing all. In 

 geology many facts have been gleaned, many theories have been 

 launched, and of these precious argosies, only the seaworthy, those 

 built in conformity with natural laws, will arrive in port. 



in. — On Ceratodus Kannemeyeri (Seeley). 

 By H. G. Seeley, F.R.S., Professor of Geology in King's College, London. 



^H. DANIEL RUSSOUW KANNEMEYER sent me a photo- 

 graph of a dental plate in April, 1897, to which was appended 

 the remai'k — " I conclude it to be Ceratodus, and its bones are said 

 to be lying at Kraal Fontein. I have sent the man who found it 

 to search for and fix the locality." The search was not successful; 

 and since then the rinderpest has prevented any attempt to reach 

 the locality, which is geologically at the top of the Karoo beds 

 above the Coal of the Indwe district of Cape Colony, in strata which 

 yield Zanclodont reptiles in other localities. It may therefore be 

 classed as of Triassic age, belonging to the Stormberg Series, and 

 well above the Permian rocks which have yielded Theriodonts. This 

 determination is of some interest as contributing to define, by means 

 of a well-marked Trias genus, the age of the upper limit in South 

 Africa of strata which have yielded Theriodont and Dicynodont 

 reptiles. Such a division between Trias and Permian agrees with 

 the grouping of the rocks to which I was led by a physical examina- 

 tion of the country in 1889. 



Failing for the time to obtain the other reputed remains, 

 Dr. Kannemeyer has recently sent me the tooth, which differs as 

 a species from all known types. In 1889 Mr. A. S. Woodward 

 (Ann. and Mag. Nat. Hist., Sept., p. 243) mentioned and figured 

 a small example of a tooth of Ceratodus, found at Smithfield in the 

 Orange Free State, which he named Ceratodus Capensis. It is 

 distinguished from the new form, which I associate with the name 

 of Dr. Kannemeyer, by the narx-ow rounded ridges which diverge 

 over the dental plate, as well as by the relatively broader form of the 

 plate, and by its thinness ; and C. Capensis difiers from the common 

 forms of the Muschelkalk and Rhastic beds of Europe by such 

 characters as separate it from the new African species (see Figure). 



The left superior dental plate of Ceratodus Kannemeyeri is thick 

 and flat, with a nearly vertical serrated external border made up 

 of three strong wedge-shaped denticles, with a slight convexity in 

 front and a marked convexity behind them. The plate has a thick 

 base, which is prolonged backward in a way seen in some species 

 from the Trias of Germany, but the base does not extend to the 

 inner anterior border of the dental plate, which may possibly be 

 a line of union with the corresponding plate on the opposite side of 

 the palate. 



The specimen measures 4-5 cm. from front to back. The extreme 

 transverse width over the first denticle is 3-5 cm., while over the 



