288 Keport S.A.A. Advancement of Science. 



the affiiiilv is much closer than has been hitherto beheved. So nearly 

 allied indeed are many Endothiodonts to Dicynodon and Oudenodon 

 that I venture to affirm that a considerable number (jf skulls labelled 

 in our museums as Dicynodon and Oudenodon are really Endothio- 

 dont. 



Gadow in his recent work on Reptilia (190J) divides the Thero- 

 morpha into 4 sub-orders. (1) Pareiasauria, (:;) Theriodontia. (3) 

 Anomodonlia, (4) Placodontia. 



In the recent English translation of Zittel's Palasontology (190-') 

 a classification is given which does not differ very greatly from that 

 of Gadow. The Theromorpha is divided into (i) Pareia.sauria. in- 

 cluding Pareiasauridae. Pareotic])ed3e. and Diadectidse. (2) Therio- 

 dontia. including Galesauridas. Deuterosauridae and Tritylodontidaa. 

 (3) Anomodontia, and (4) Plarodontia. The Pelycosauria are re- 

 garded as a sub-order of the Rhynchocejjhalia. 



Before j)roceeding further with the consideration of the clas- 

 sification, it will be well to look at one of two of the more recent 

 obsen'ations on the structure of some of the principal tvpes. 



PROCOLOPHOX. 



This little primitive lizard-like reptile was first placed (1876) b\ 

 Owen in the Theriodontia. apparentlv less on account of any Therio- 

 dont affinities, than because he did not know where else to put it. 

 Seeley in 1878 described three imperfect skulls from Donnybrook. 

 S. Africa — the same locality as Owen's specimens came from — aJid 

 says " I see no reason to hesitate on the evidence detailed, in regard- 

 ing it [Procolophon] as a fossil Rhvnchocephalian." In his later 

 works Seelev was much more impressed bv its affinities with Pareia- 

 saurus. Lydekker though he places Procolophon in the Anomodontia 

 points out that it " Apj)ears to ])resent an approximation in several 

 points to the Rhynchocephalia. While Cope unites Procolophon 

 with Palaeohatteria, Homoeosaurus. and a few other genera to fonii 

 the order Proganosauria. 



As almost every detail of the structure of Procolophon is now 

 known it can be definitely asserted that the affinities are very much 

 more with Palaeohatteria than with any of the Therio<lonts. Anomo- 

 donts or even with Pareiasaurus. 



Procolo))hon has well developed abdominal ribs, each composed 

 of a series of pieces : the vertebrae are notochordal : the scapula is a 

 short l)r<)a(l bone without a distinct acromion: the digital formula for 

 the manus is 2. 3. 4. 5. 4. and for the j^es 2. 3. 4. 5. ?: the lower 

 jaw has a distinct coronoid element : and the pubes and ischia are 

 flat plate-like bones re.sembling those in Palaeohatteria except that 

 the pubic foramen is near the middle of the bone. As the posterior 

 cranial region of Palaeohatteria is unknown, it is impossible tO' .sav in 

 what degree il differs from that of Procfilophon. but as there is such 

 close agreement in the other i)arts of the skeleton it becomes 

 necessary to place Procolophon somewhere near to Palaeohatteria, and 

 until the cranial structure of Palaeohatteria is fullv known it will 

 perhaps be safer to make the Procolophonia a distinct sub-order. 



