266 



to the conditions actually found in Ameiurus. And it would seem as 

 if this traversing of the nerves by the muscle must have taken place 

 as the growing muscle sought its point of attachment on the skull, 

 rather than after that attachment had been acquired. The same re- 

 sult could of course have been arrived at by a shifting of the point 

 of origin of the inferior muscle of Polypterus ; but the arrangement in 

 Polypterus would seem to be less primitive than that in Amia. 



The so-called pseudobranch of Ameiurus can now be considered. 

 This glandular structure on the internal carotid was first described 

 by McKenzie ('84), though it is referred to by Wright ('84, p. 373), 

 as a pseudobranch, in an earlier article in the same publication. A 

 thymus gland is also figured by Wright and described by Mc Kenzie 

 in the same works. The thymus lies practically in its accustomed 

 place, as compared with other Teleosts, while the pseudobranch is 

 greatly displaced, lying far forward, in the hind portion of the orbit, 

 and dorsal, instead of posterior and ventral to the posterior division 

 of the adductor arcus palatini. 



But this difierence of position, alone, is certainly not of vital im- 

 portance. Of much greater importance are certain suppositions that 

 must be made if the organ is to be considered as a mandibular pseudo- 

 branch, namely: that it had, before its practical abortion, either sent 

 an important blood current upward into the internal carotid artery 

 through the persisting dorsal end of the efferent mandibular artery or 

 received such a current from that carotid artery ; that it had received 

 an important afferent current from the internal carotid through a se- 

 condarily established connection with that artery, as in Gadus and 

 Esox; or that it had received such current by way of a commissural 

 connection with the efferent hyoidean artery. And either one of these 

 suppositions seems opposed to what might be expected to have existed 

 in any Ganoid or Teleost that could be an immediate ancestor of this 

 fish. Furthermore, if it is assumed that the connection with the in- 

 ternal carotid was through the persisting dorsal portion of the efferent 

 mandibular artery, the pseudobranch must certainly have travelled up- 

 ward along that artery until it reached the internal carotid. But the 

 usual course of the efferent pseudobranchial artery in Teleosts, up to 

 the point where it must originally have emptied into the internal ca- 

 rotid, is, so far as can be judged from the very unsatisfactory de- 

 scriptions, at first forward ventral to the adductor arcus palatini, and 

 then forward and upward through that muscle, at its origin, or over its 

 anterior edge. The pseudobranch must accordingly have first followed 

 this same course, if it followed the efferent mandibular artery, and 



