SENSE ORGANS OF AMIURUS. 265 
tion of hyomandibular. With Bodenstein I find no communication 
between the principal canal and that which is lodged in the preoper- 
culum and mandible opening with eight pores on either side. (Figs. 
5 and 6). 
From the hyomandibular articulation the canal passes forwards 
and inwards giving off the infraorbital branch which passes through 
the infraorbital chain of bones and terminates in the adnasal or 
antorbital bone, which is the most anterior of these. In its course 
the infraorbital canal first opens directly behind the eye, then by 
two pores below it and one in front, and finally by two in the same 
transverse plane behind but lateral to the anterior nasal aperture. 
The supraorbital canal may be regarded as the continuation of the 
principal canal ; immediately after giving off the infraorbital branch, 
a tube is directed backwards which opens behind the first infraor- 
bital pore, but near the middle line. From this point the canal 
inclines distinctly towards the middle line, opens by a pore in the 
plane of the eyes, by another medial to the posterior nares, and 
terminates by two pores which lie in the same sagittal plane over 
the medial division of the nasal sac. No further communication 
takes place between the supraorbital and infraorbital canals of the 
same side, nor do the supraorbital canals of opposite sides meet in 
the middle line as in Cottus. The chief departure from Wieder- 
sheim’s diagram (p. 359 /. c.) consists in the independence of the 
mandibular branch, and the absence of an anterior anastomosis of 
the infra- and supraorbital branches —features which are common to 
Amiurus and Cottus. On the other hand, Cottus differs from 
Amiurus in possessing one median and two lateral pores in the occi- 
pital commissure, and in the supraorbital branches meeting each 
other in the middle line before they give off a single backwardly- 
directed tube in place of the two noted above. 
(2), Accussory LATERAL ORGANS. 
In various Teleosts the lateral line is not an uninterrupted canal 
as in Amiurus, but may be regularly interrupted as in Hsox, two or 
more uncanaliculated scales separating those which are canaliculated. 
“ As if in compensation, however,” says Leydig’, ‘‘ additional scat- 
tered canaliculated scales are present above and below the lateral 
line, to a certain extent accessory or rudimentary lateral lines, as 
1], e. p. 33. 
