GREGORY, PRESENT STATUS OF ORIGIN OF TETRAPODA 333 



by Pander or Traquair and seems to be known only in Megalichthnjs. In 

 this genus, Watson (1912, pp. 9-10) has noted the following remark- 

 able resemblances to the very primitive Stegocephalians Fteroplax and 

 Loxomma: 



"The Basisphenoid of Megalichthys has sometimes carotid foramina just as 

 in Loxomma. It has small but distinct basi-pterygoid processes which arc, 

 however, not provided with articulating surfaces but with sutural ones. The 

 long parasphenoid extends forward to the premaxillse as it may do in Ptero- 

 plaw. Its lateral borders are in contact with the Pterygoids, to which they 

 afford support, and the bone is connected with the roof of the skull by a fused 

 ethmoid. 



"The Pre-vomer is identical with that of 'Loxomma' in the majority of its 

 attachments, carries one large tusk and a pit for the replacing tooth. It meets 

 its fellow of the opposite side, and forms the front of the posterior naris ; it is 

 doubtful, however, if it meets the palatopterygoid. 



"The Palatopterygoid of Megalichthys is exceedingly like the palatine and 

 pterygoid of Pteroplax. They have similar relations to the basisphenoid, 

 parasphenoid and maxilla. There is the same row of small teeth parallel to 

 those of the maxilla with larger teeth inside them, and the . pterygoid is cov- 

 ered with the same shagreen of fine teeth." 



Watson also states that unlike later Stegocephali Loxomma has a single 

 median occipital cond3'le on the basioccipital, and that this condyle "ex- 

 actly resembles the end of a vertebral centrum, which it no doubt is.^' 

 In these features Loxomma therefore agrees with Ehipidistia rather than 

 with typical Stegocephali. In reference to the dentition of Loxomma 

 Watson states that : 



"The Palatine is very similar to the transverse in general character, but 

 bears two large tusks near its outer border. Each of these teeth has assor^i- 

 ated with it a shallow pit from which a tooth has been shed, and in which a 

 replacing tooth will be formed. In some cases both teeth are present at once, 

 a condition which was undoubtedly only transitory ; this curious type of tooth 

 change is very characteristic of the Stegocephalia, and is unknown elsewhere 

 except in the Crossopterygian ^ fish, where it occurs in a very typical form in 

 the vomerine tusks of Megalichthys, and no doubt in many other genera, and 

 in Lepidosteus. This occurrence seems to me a strong additional reason for 

 regarding the Tetrapoda as derived from this group of fish." 



The infolded base of the teeth in the Ehizodontidae is also strongly 

 suggestive of stegocephalian affinities, but may be only an independent 

 adaptive device for fastening the teeth to their bases. 



8 "Throughout this paper 'Crossopterygian' is used as including only the three families 

 Holoptychiidae, Rhizodontida- and Osteolepidse of S. Woodwards sub-order Rhipidistla, 

 and excluding Tarrasius, VuJacanthtts and Polyptervs." (Watson). 



t 



