430 Mr. H. Day on the 



small but distinct l)apipterv2:oi(l processes wliicli are, however, 

 not provided with articulating surfaces but with sutural ones. 

 The long parasplienoid extends forwards to the premaxillae 

 as it may do in Pteroplax, and its lateral borders are in 

 contact with the pterygoids to which they afford support." 



Here, again, the para-bnsi.*]dieiioid elements haveapparently 

 the characteri.stic form with which we are dealing; and once 

 having assumed that in Loxomma the posterior expansion is 

 basisphenoid, the assumption is also made in the case of a 

 similar expansion with processes and foramina in Mega- 

 Jidithys. That the assumption is correct in the case of 

 Lo.vomma there can be little doubt, from a consideration of 

 primitive Keptilian structures. 



Bones o£ the basi-cranial axis in fishes are rarely preserved, 

 however; in fact, the present specimens, so far as can be 

 ascertained, are almost unique in their perfect preservation. 

 Even the examples found by Fritsch lacked the side-processes 

 in every case, judging by his figures ; but, nevertheless, 

 the specimens were well enough preserved to cause Fritsch 

 on several occasions to coninient on their remarkable resem- 

 blance to the parasplienoid of the Branchiosauria, by which, 

 presumably, he would mean the compound para-basisphenoid 

 as described by Watson for Loxomma and Pteroplax. 



One definite fact seems to emerge from a consideration of 

 the above points, and that is, that certain primitive fossil 

 fishes of the Ganoid group, certain of the primitive fossil 

 Amphibia — the Branchiosauria in particular, — and both 

 living and fossil lieptilia, exhibit a bone on the ventral 

 surface of the cranium, possessing the same general form, the 

 same relationships, and all agreeing, as will be gathered from 

 the above, in representing in each group a compound para- 

 basisphenoid, the parasphenoid in each case having a 

 broadened posterior extension underlying the basisphenoid. 

 Megaliclithys stands as a type for the fishes, Loxomma and 

 Pteroplax for the early Amphibia, OpJithalmosaiirus and 

 Tricleidus for the fossil, and the Common Snake and 

 8phenodon for the recent reptiles. 



The present specimens make the case even more conclusive, 

 since in A and C we have the pituitary region well marked 

 out as already described. Now the pituitary body always 

 comes into relationship with that part of the cartilaginous 

 basis cranii which later ossifies to form the basisphenoid. 

 Hence, in the present specimens we must assume that in A, 

 in the pituitary region, we have the actual dorsal surface of 

 tlie basisphenoid preserved. The depression for the pituitary 

 body does not affect the ventral surface of specimen B, 



