the foramen magnum. Since this is the only occurrence 

 of such a bone in amphibians, it is suggested that it is a 

 secondary ossification or calcification of the cartilaginous 

 area observed here in other fossil forms. 



The parasphenoid extended high up on the sides of the 

 otic region nearly reaching the fenestra vestibuli, which 

 marks the line of union of prootic and opisthotic — these two 

 bones are indistinguishably fused and therefore only pre- 

 sumed to be present. The anterior margin of the prootic 

 forms the posterior wall of a large prootic foramen. This 

 foramen has been described as a notch between oticoccipital 

 and ethmosphenoid division of the endocranium (as in the 

 crossopterygian fish); however, ventrally there was no in- 

 dication of such divisions— the basisphenoid region was fused 

 with the parasphenoid and the otic capsule. The basisphen- 

 oid area had lateral basipterygoid processes and extended 

 forward as the interorbital septum below and the lateral 

 and ventral walls of the anterior part of the cranium above, 

 that is, as the orbitosphenoid. Marking the Hmit between 

 the basisphenoid and orbitosphenoid was a large orbital 

 fenestra. This endocranium was solidly roofed except for a 

 parietal foramen. 



The endocrania of Edops craigi (Pennsylvanian) and 

 Eryops megahcephalus (Lower Permian) are better known but 

 of a somewhat different plan. Both of these are large species, 

 the first with a head length of about two feet, and the 

 second, one and a half feet. In these the otic region is con- 

 tinuous with the ethmosphenoid region. The details of struc- 

 ture suggest some specialization but contribute little to the 

 problem at hand. 



The oldest amphibians known are from the uppermost 

 Devonian or the Lower Mississippian. These are the ichth- 

 yostegids and acanthostegids from Greenland, and several 

 fragmentary types known as Elpistostege, Otocratia, and 

 Loxonuna. 



The icthyostegids are well known in terms of the dermal 

 head skeleton but the endocranium is not yet fully described. 

 Ichthyostega (Figure 4-25) is of moderate size with a head 

 length of about eight inches. It lacks the intertemporal bone 

 but has a small preopercle and subopercle. The sensory 

 canals are embedded in the bone or lie so superficially that 

 even grooves are not formed. The external naris is peculiar 

 in its marginal position. The palate has minimal interptery- 

 goid fontanelles and the parasphenoid is limited to the 

 rostral portion. The lower jaw was encased in dermal bones; 

 there were three coronoids and an ossified Meckel's carti- 

 lage. The hyoid arch is not known nor is the state of the 

 branchial skeleton— presumably they were cartilaginous. 

 The endocranium has been described as having ethmos- 

 phenoid and oticoccipital divisions like that of a crossop- 

 terygian. 



If this endocranium is two parted then the ichthyostegid 

 must be viewed as distinct from other early amphibians. In 

 terms of their dermal shield, there is some support for this 

 view but not enough to lead to the necessary conclusion that 



an amphibian type of structure was developed at least twice. 

 Such a polyphyletic origin is not warranted in terms of the 

 facts at hand. 



Seymouriamorphs and origin of repfiles The seymouria- 

 morphs are Upper Pennsylvanian and Permian fossil forms 

 considered reptiles by some and amphibians by others. Since 

 well-differentiated reptiles were already present at the time 

 these animals occurred, they can be viewed as relicts, if in- 

 deed their line was ancestral to the reptiles. 



The head skeleton of Seymouna, an early Permian form, 

 has been described in detail (Figure 4-26). This head skele- 

 ton is quite similar to those already described, and thus it 

 would be diflncult to ascribe to it any definitely reptilian 

 features. The lacrimal canal, the transverse posterior flange 

 of the pterygoid, and the form of the sella turcica and dor- 

 sum sellae suggest the reptile. Basically, this is an amphibian 

 skull; evidence of reptilian affinity is better marked in 

 other parts of the skeletal system. 



Earliest reptiles 



The first reptiles in the fossil record show a bewildering 

 amount of variation in the few details known. It is generally 

 believed that there are two basic lines of Lower Permian 

 forms: the diadectomorphs and the captorhinomorphs. 



These lines differ in such features as the loss or great re- 

 duction of the otic notch and loose connection of the roof 

 and cheek in the captorhinomorphs. The palate of the dia- 

 dectomorph generally lacked interpterygoid fenestrae and 

 the pterygoid was sutured to the basis cranii rather than 

 articulated with a large basipterygoid process. The dia- 

 dectomorph had a well-ossified endocranium with a trigem- 

 inal foramen, whereas the captorhinomorph had the antero- 

 lateral cranial wall, anterior to and above the trigeminal, 

 widely open. 



The two lines agreed in having the posterior part of the 

 roof reduced in size with the result that the postparietals 

 which were fused at the midline were moved, more or less, 

 onto the occipital surface. Modification of the roof was ac- 

 companied by loss of the intertemporal and sometimes loss 

 of the supratemporal or tabular. These changes were less 

 marked in the diadectomorphs. Both groups also agreed in 

 the reduction in the number of the coronoids and splenials to 

 one of each, and in the presence of a small Meckelian fenestra. 



The earliest known reptiles are hard to place in either of 

 these groups (Figure 4-27). Cephalerpelon, from the Upper 

 Pennsylvanian of Mazon Creek, Illinois, is possibly a cap- 

 torhinomorph but so little of its skull is known that this 

 must be largely conjectural. Petrolacosaurus is also of Upper 

 Pennsylvanian age but is extraordinarily lizard-like in its 

 greatly reduced premaxilla and in the form of the palate 

 and its articulation with basipterygoid process. It could be 

 considered a captorhinomorph but has also been compared 

 with the Eosuchia. 



88 • HEAD SKELETON OF OTHER TETRAPODS AND CHOANATES 



