No. 3-] LATERAL LINE OF AMIA. 473 
face of the post-orbital process. His description of the canal 
in the Characinide shows that it agrees closely in general 
course and position with that of Amia, beginning in both 
forms at the anterior end of the nasal, median to the nasal 
openings, and ending at the hind margin of the frontal, 
where one or more branches are sent backward into the parietal. 
The main canal in the Characinidz also agrees closely with 
that in Amia, but Sagemehl describes the suborbital part of it 
as a branch given off downward behind the eye, from the 
extreme anterior end of what he considers to be the main tem- 
poral or posterior division of the supra-orbital. 
3. Operculo-mandibular Canal.— The operculo-mandibular 
canal begins at the anterior end of the lower jaw, close to the 
middle line of the head and close to the anterior end of the 
canal of the opposite side, but without any connection what- 
ever with it. Bridge (No. 4, p. 620), Sagemehl ( No. 13, p. 183), 
Van Wijhe (No. Io, p. 288), and Shufeldt ( No. 16) all agree in’ 
stating that here the two canals are continuous; but this was 
not the case in any of the many specimens I have examined. 
Starting here as an independent line on each side of the 
head, the mandibular part of the canal runs backward along 
the lower inner margin of the ramus of the mandible, nearly to 
its hind end, where it turns upward and passes out of the man- 
dible immediately in front of and above the articular process for 
the symplectic. It then turns sharply backward and is contin- 
ued as the opercular part of the canal upward and backward 
through the pre-operculum in the curved line of that bone, 
Leaving the bone at its upper end it passes through a narrow 
strip of dermis, and joins the infra-orbital canal, near the hind 
end of the squamosal, turning forward at the point of union. 
The mandibular and opercular portions of the canal develop as 
two distinct canals, uniting later with each other to form a 
continuous line and then uniting with the main infra-orbital. 
These later connections in Amia are not always formed in other 
fishes. In Fsoxr /uctus for example, the two portions always 
remain, even in the adult, separate from each other and from 
the infra-orbital ; and in Polypterus bichir, Amturus catus, and 
Cottus gobio, although they unite to form a continuous line, they 
do not unite with the main canal. 
4. Supratemporal Cross-commissure. — The supratemporal 
