56 



THE HISTORY OF THE PELYCOSAURIA, WITH 



mosal by Credner is the prosquamosal ; the true squamosal must have been free, and con- 

 nected with the parietal processes. 



" In Sphmnodon the maxillary forms the lower boundary of the orbit ; in Palceohatteria 

 and Dimetrodon, the jugal excludes the maxillary from the orbit, The vertebra with the 

 well-developed intercentra, the ribs with the double articulations, can only be compared 

 with those of the Ehynchocephalia and Proganosauria (Palceohatteriidoe) . The presence 

 of a free central bone in the tarsus of the Pelycosauria is an original character, which is 

 shared only by the PalczohaMerUdce and Prolerosauridw ; but in the Pdmohatteriida, 

 tarsals 1 and 5 arc; free, in Proterosaurus and Dimetrodon they are muted, to support 

 metatarsal 4 and metatarsal 5. The humerus of Dimetrodon can be directly reduced to 

 that of Sphcenodon. The entepicondylar foramen is well developed in both; the ectepi- 

 condylar for amen of Sphcenodon is represented by a very distinct ectepieondylar groove in 

 Dimetrodon. 



"The specialization of the Pelycosauria consists in the enormous development of the 

 neural spines of the dorsal vertebrae, and in the reduction of the upper part of the quad- 

 rate and its nearly complete inclosure by the squamosal, prosquamosal and quadratojugal. 

 It, is quite evident, that the Pelycosauria with the two temporal arches and the specialized 

 neural spines cannot be the ancestors of Mammals; they represent a, specialized side 

 branch of a, line leading from the Proganosa/aria to the llkyn/'hocephalia, which becomes 

 extinct in the Permian. 



"The Mammals have a, single temporal (zygomatic) arch; the posterior nares are 

 placed far behind, and are roofed over by the maxillary and pterygoid plates; the quad- 

 rate is completely co-ossified with the squamosal and quadratojugal ; the occipital condyle 

 is double, the entepicondylar foramen is present, in all the generalized forms. The 

 ancestors of Mammals must show the same condition. 



"Seeley 45 has combined a, number of Pcrmo-triassic Peplilia from South Africa into 

 an order which Ik; calls Oom/phodontia. These Reptiles are : Trilylodon Owen (always 

 so far considered a Mammal), J)iademA>don Seeley, Gompkognathm Seeley, Microgompko- 

 don Seeley, and Trirachodon Seeley. 



"In Oqmphognathus we have a double occipital condyle; the posterior nares are 

 placed far behind and are roofed over by the maxillary and pterygoid plates, and there 

 is an entepicondylar foramen. The quadrate seems to be of the reduced form ; a condi- 

 tion we see also in the closely related Oynognalhus. 



"These forms look very much like Mammals and could possibly be ancestral to 

 them. We must suppose that the condition of the palate we see in the Mammalia and 

 Gomphodontia, has been developed from a type we find among the Phynehocephalia. 

 The Crocodilia, where we have a similar palate as in Mammals, show us, how such a type 



