420 Edward Phelps Allis jr., 



the Muraenidae there is here, as I have shown [9\ a median ethmoidal 

 latero-sensory chamber, but this chamber is not a cross-commissural 

 one, formed by the end-to-end anastomosis of two canals, one belonging 

 to either side of the head. On the contrary, it is a longitudinal 

 chamber formed by the anastomosis of two parallel longitudinal canals, 

 one on either side of the head, which touch and coalesce to form a 

 chamber which is apparently the strict homologue of the so-called 

 median canal of Garman's [39] descriptions of selachians. In certain 

 of the Batoidei this median canal is said by Grarman to be "transverse", 

 and this transverse canal may perhaps be the homologue of the 

 ganoidean cross-commissure, and not of the longitudinal median canal of 

 selachians and the Muraenidae. 



Polypterus furthermore agrees with both Amia and Lepidosteus 

 in that the latero-sensory component of the squamosal is not fused 

 with an underlying primary component: for Sagemehl [63] has shown 

 this to be true of, Amia, I [5] have called attention to it in Polyp- 

 terus, and Parker says [58, p. 448], of Lepidosteus, "I can find no 

 opterò tic' under the squamosal". The squamosal of Polypterus is, how- 

 ever, one latero-sensory ossicle shorter, anteriorly, than the squamosals 

 of Amia and Lepidosteus, for the bone lodges but one, instead of two, 

 ante-preopercular sense organs. This, it may here be stated, is a 

 characteristic of many teleosts. Scomber and Conger being the only 

 known exceptions to it. 



Furthermore, Polypterus agrees with Lepidosteus, but differs fix)m 

 Amia, in that the supraorbital canal anastomoses with the main infra- 

 orbital by its terminal instead of its penultimate primary tube, and 

 in that it has no temporal hole; and it differs from both those fishes 

 in that it has, so far as known, no spiracular sense organ. The 

 presence of this latter organ, it may be noted, is a selachian 

 characteristic. 



Ostariophysi. 

 Siluridae. 



Ameiurus nehulosus. Fig. 5 is from a drawing, made in 

 1H87 — 88, of a larva of Ameiurus, and is given simply to show that the 

 canals in tliis fish develop, in principle, exactly as they do in Amia. 



1 



