G4 DR. W. G. RIDEWOOD ON THE CRANIAL [May 3, 



bones arise independently of the teeth, it is not unreasonable to 

 assume that the pi-esence of teeth on these bones indicates a more 

 primitive condition than the absence of teeth — unless, of course, 

 the teeth are of secondary, and not primary origin. A general 

 review of the skull in Teleostean fishes certainly goes to show that, 

 although no broad conclusions can be drawn from the presence or 

 absence of teeth on the vomei-, a dentigerous parasphenoid is 

 habitually associated with other features which may be regarded 

 as archaic. Of the genera examined, Elops, Megalops, Alhula, 

 A7'apawia, Osteoglossum, Notopterus, Hyodon, and the Mormy- 

 ridse * have teeth on the parasphenoid. 



There is probably in a considerable number of the lower 

 Teleosteans the half of a vertebral centrum fused with the basi- 

 occipital and exoccipitals to form the hollow-cone articular surface 

 at the back of the cranium, so that the cranial articulation really 

 takes place between this half-centrum and the first free centrum. 

 Such a condition has long been known to exist in Amia, in which 

 the corresponding neural arcfi persists ; and Shufeldt (Rep. U.S. 

 Com. Fish. 1883 (1885), p. 816) has shown that in Megalops the 

 suture between the half-centrum and the exoccipitals and basi- 

 occipital is readily recognisable. When this half-centrum is 

 removed, the transverse end of the basicranial axis presents a 

 rough surface, with a suture of the shape of an inverted T or Y 

 separating the basioccipital below and the right and left ex- 

 occipitals above. In all of the forms at present under considera- 

 tion, the upper part of the suture between the half-centrum and 

 the exoccipitals is recognisable in the floor of the foramen magnum, 

 although the suture may no longer be visible down the side of the 

 half -centrum ; and in these cases the half-centrum can only be 

 removed by the application of some force, and the inverted T 

 suture above mentioned be brought into view. 



In the Salmonidse, in Hyodon and in Heterotis, although there 

 is not a very definite articulation between the first centrum and 

 the basioccipital and exoccipitals, and only a very slight range of 

 movement is possible between these, the centrum separates quite 

 readily, and exposes the triiudiate or invei-ted T suture, as already 

 figured by Gegenbaur (Festschr. A. v. Kblliker, Leipzig, 1887, 

 pl.i. fig. 9) and" Parker (Phil. Tra,ns. vol. 163. 1873 (1874), pi. viii. 

 fig. 8). The condition thus approximates to that which is so 

 commonly met with in the Acanthopterygian fishes, where the 

 cranial articulation is definitely between the basioccipital and 

 exoccipitals in front and the first vertebra behind. 



In Arapaima the occipital articulation is further complicated 

 by the fact that the first complete centrum, although free from 

 the half-centrvim in front of it, is not capable of free play, since 

 its lower portion sends forward a pair of long, stout processes 

 which are rigidly united by oblique, jagged sutures with the back 

 of the parasphenoid. 



* The parasplienoidal teeth are rudimentary in Mormyro;ps and absent in 

 Gymnarclius. 



