262 



The lachrymal bone overlaps externally the ventral edge of this part 

 of the premaxillary. 



The premaxillary of Polypterus thus seems certainly to contain 

 the homologue not only of the premaxillary bone of Amia, without its 

 so-called posterior process, but also the separate, antorbital bone of 

 Amia, and the lateral part of the ethmoid of that fish. This I have 

 already had occasion to suggest in two of my earlier works (Nos. 3 

 and 4), The anterior end of the antorbital of Amia rests upon the 

 dorsal surface of the tooth-bearing part of the premaxillary, and there 

 abuts against, and is firmly bound to, the lateral end of the ethmoid. 

 The posterior portion of its inferior edge abuts against the lachrymal. 

 Immediately in front of the lachrymal the inferior edge of the bone 

 is bound by tissue to the anterior portion of the maxillary (No. 3)^ 

 but it does not give articulation to the latter bone, that function 

 being fulfilled by the premaxillary and vomer, and, more particularly, 

 by a separate bone of primary origin, the septomaxillary of Bridge's 

 (No. 8) and Sagemehl's (No. 36) descriptions. The hind end of the 

 antorbital bone of Amia does not reach the frontal, and this hind 

 part of the bone, in this fish, is traversed, its full length, by a part 

 of the infraorbital lateral canal. The posterior portion of the pre- 

 maxillary of Polypterus thus has no exact homologue in Amia, and 

 it may, perhaps, represent some such bone as the separate dermal 

 prefrontal of Bridge's descriptions of Amia, a bone that I, however, 

 have never been able to find in the fish (No. 1, p. 478). 



The palatine process of the premaxillary bone of Polypterus is 

 not found in Amia, at least as an appreciably developed structure, 

 but the vomer of Amia has much the relations to the premaxillary 

 of the fish that the palatine process of the premaxillary of Polypterus 

 has to the tooth-bearing part of that bone of that fish. Between the 

 mesial edges of the two vomers of Amia, at the points where they 

 join the premaxillaries, there is a little open space in which a part 

 of the ventral surface of the chondrocranium is exposed (No. 3, Fig. 5), 

 exactly as there is, in Polypterus, between the mesial edges of the 

 palatine processes of the premaxillaries. Moreover, in old and large 

 specimens of Amia, the vomers and premaxillaries may be found in 

 process of anchylosis (No. 3, p. 451). That all this is sufficient to 

 warrant the definite assumption of an homology of the vomers of 

 Amia with the palatine processes of the premaxillaries of Polypterus 

 I am not prepared to assert, for I find no descriptions of Teleosts 

 sufficiently complete to control it. It seems to me, however, almost 

 certain, and it is worthy of note, that Sagemehl, in his careful de- 



