440 Edward Phelps AUis jr., 



The preopercular canal lies along-, and is attached by membrane 

 to the outer edge of the preoperculum. 



No mandibular canal was found. 



Several short lines of what are apparently small pit organs were 

 found, two of them, in the occipital region of the head, being especially 

 marked. One of these two latter lines, the only one shown in the figures, 

 was attached by a vertical band of membrane to that part of the hind edge 

 of the skull that is formed by the parietal, in exactly the same manner 

 that the ossicles that overlie the frontal and prefrontal bones are attached 

 by membrane to those bones. This line of pit organs would seem to 

 be the homologue of the middle head line of pit organs of Amia, and 

 the fact that it is attached to the hind edge of the parietal, while 

 the ossicles of the supratemporal commissure are not so attached, 

 might perhaps be taken to indicate that the canal that is found in 

 the other Cyprinidae and in the Characinidae, is the middle head line 

 and not the supratemporal canal. I consider, on the contrary, that 

 the presence of these two lines in Moxostoma, the one as a line of 

 pit organs and the other as a canal line, establishes quite unquestionably 

 that the canal line found in the Cyprinidae and the Characinidae is the 

 snpratemporal commissure and not the middle head line. 



The latero-sensory ossicles of Moxostoma thus none of them, ex- 

 cepting only in certain parts of the infraorbital series, form integral 

 parts of the regular cranial bones. They lie, in every case, in the 

 dermis, superficial to the bones of which they usually form a part, 

 and this position was considered by Sagemehl \64, p. 38] as a second- 

 ary one, as was also the presence of the thick covering dermis on 

 the dorsal surface of the cranium. But the reasons given by that 

 author in support of this conclusion certainly do not in any way 

 apply to the bones of the infraorbital series of Moxostoma. These 

 bones seem to indicate, on the contrary, that Moxostoma presents, in 

 its latero-sensory canals, a primary condition, or at least a Holocepha- 

 lan one, the latero-sensory ossicles not yet having fused with under- 

 lying membrane bones. A further, apparently also primary, or at 

 least Holocephalan condition, is the large number of sense-organs in 

 each of the canal lines, 



