ON CERTAIN CROSSOPTERTGIANS. 1251 



cavity, but that the medial parts of the lateral and ventral walls 

 of the occipital region consisted of cartilage. Expressed in 

 another way : the ossification in the occipital region probably 

 occupied only the external parts of the region, while the inner 

 ones were cartilaginous. The basioccipital has a rather con- 

 siderable extension forward, and actually lies with more than 

 its anterior half in the labyrinth region. There cannot, how- 

 ever, be any doubt of its homology. The lateral occipital is 

 strikingly high, extending from the basioccipital below nearly 

 to the dermal cranial roof dorsally (Bryant, 1919, text-fig. 3), 

 thus partly occupying the position of the supi-aoccipital, which 

 acconling to Bryant is totally lacking. The lateral occipital is 

 said by Bryant to be perforated by a fine canal, but as he does 

 not describe the course of this canal, one cannot decide with full 

 certainty whether it is a canal for a dorsal branch from the 

 n. lineae lateralis or the vagus canal, although the latter alternative 

 seems to be the more probable one. If however, the canal 

 should reveal itself to have transmitted a dorsal branch of the 

 n. lineaa lateralis, the vagus nerve must have perfoi-ated the 

 lateral wall of the primordial neurocranium immediately anterior 

 of the lateral occipital. 



In either lateral wall of the labyrinth region of the fish, Bryant, 

 1919, p. 18, describes five ossifications. One of these ossifications 

 he calls the pro-otic, another the opisthotic. The three remaining 

 ones he considers it impossible to identify, and accordingly he 

 does not name them. 



The description which Bryant gives of the pro-otic is not 

 sufficient to make fully clea,r the shape and actual extension of 

 the bone. As I understand his account, the pro-otic is a rather 

 low bone (c/. his text-fig. 5 a and pi. ix. fig. 3), which does not by 

 far extend to the cranial roof dorsally. It is pei'forated by a 

 canal, one of the openings of which is situated approximately at 

 the middle of the length of tlie bone just at the boundary 

 between its ventral and lateral surfaces or perhaps on the 

 ventral surface, though close to the lateral edge {cf. Bryant, 

 1919, text-fig. 4; pi. ix. fig. 3; pi. xiv. figs. 4^5). ^The canal 

 is considered by Bryant to have transmitted the n. facialis (he 

 evidently means the truncus hj^oideo-mandibularis of this nerve), 

 but he gives no sta.tements either as to its course nor as to 

 the position of its other opening, and the facts so far known do 

 not seem to support this view. As far as can be judged from 

 the conditions in recent fishes, the canal seems most probably 

 to have been developed for the carotis externa on its way 

 upwards from the cranial basis to approximately the height of 

 the trigeminus and facialis exits {cf. AUis, 1897, pp. 497-500; 

 1908, pp. 219-221 ; 1909, pp. 51-53, 185-187; 1911 «, pp. 284- 

 287; 1914; 1919 a; Allen, 1905, pp. 51-56; Danforth, 1912, 

 pp. 435-442 ; Stensici, 1922). The truncus hyoideo-mandibularis 

 ought not to have run so strongly downwards that it can be 

 thought to have pierced the canal. 



