264 



vein of Allen's descriptions; but the median descending branch of 

 Ameiurus is not found. This venous cross-commissure is apparently 

 not described by Mc Kenzie, and I find, in my larva, certain other 

 variations from his descriptions. According to McKenzie the cranial 

 cavity is drained by a pair of veins that arise in the nasal sac and 

 that, passing backward the full length of the skull, issue "with the 

 rami laterales trigemini, through the supraoccipital". I find the nasal 

 sac drained, in part at least, by a branch of the orbito-nasal vein 

 which issues from the cranial cavity with the orbito-nasal artery, ex- 

 actly as in Scorpaena; and I find a large vein issuing from the cranial 

 cavity with the trigemino-facialis nerves to join the jugular vein, also 

 exactly as in Scorpaena. What the intracranial connections of these 

 veins are I did not trace. 



In still one further point relating to this same region of the head, 

 that is, in the manner of innervation of its eye muscles, Ameiurus is 

 said to present an arrangement that is uuteleostean in character. 

 According to Workman ('00) the eye muscles are innervated in Amei- 

 urus exactly as they are in Amia (Allis, '97); and this manner of 

 innervation differs from that in all other Teleosts that have been pro- 

 perly investigated in this respect (Herrick, '99; Workman, '00; 

 Corning, '00; Allis, '03; Allis, 1). In Lepidosteus (Allis, 1) the 

 eye muscles are also innervated as in Amia, and as Corning ('00) 

 finds them also so innervated in Acipenser, this would seem to be a 

 ganoidean characteristic. In two specimens of Polypterus that I have 

 examined, one in sections and the other in dissections, I find the innerv- 

 ation differing from that in Amia in that the inferior branch of the 

 oculomotorius passes dorsal instead of ventral to the rectus inferior; 

 my interpretation of Van Wijhe's descriptions of this fish, in my work 

 on Amia ('97), thus being incorrect. In this course of the oculomo- 

 torius dorsal instead of ventral to the rectus inferior, Polypterus diöers 

 from all other vertebrates that I find described, excepting only Petfo- 

 myzon, and as, in Petromyzon, the rectus inferior is innervated by the 

 abducens and not by the oculomotorius, the muscles in the two fishes 

 are certainly not homologous. Although this position of the rectus 

 inferior relative to the oculomotorius, if found to be constant, is de- 

 finitely characteristic of Polypterus, it could easily be derived from, 

 or give origin to, either the ganoidean or the teleostean arrangement. 

 For if, starting from the arrangement found in Polypterus, the rectus 

 inferior should, by shifting its point of origin, acquire an origin dorsal 

 to the inferior branch of the oculomotorius by traversing that nerve 

 posterior to the point of origin of the branch that innervates the 



