1895,] SEITSORY CAKAL SYSTEM OI- PISHES. 279 



line, and then uniting with the main infra-orbital. These latei' 

 connections in Amia are not always found in other fishes ... in 

 Amiurus catus although the^^ unite to form a continuous Une, they 

 do not unite with the main canaV." 



I shall endeayoiu' to prove that in Amiurus catus the operculo- 

 mandibular branch does join \nth the main canal of the head. 



In the account given by Eamsay Wright (12. pp. 262-265) it is 

 stated that the posterior portion of the lateral canal is detached 

 from the remaining portion, which is scarcely correct, and that 

 there is " no communication between the principal canal and that 

 which is lodged in the preoperculum." 



Whether or not Professor Ramsay Wright regards the man- 

 dibular portion as distinct from the preopercular portion I am 

 unable to gather from his account. On p. 265 he speaks of " the 

 canal which is lodged in the preoperculum and mandible " as if it 

 were continuous ; but on the same page he refers to the " inde- 

 pendence of the mandibular branch," — I take this to mean the 

 operculo-mandibular branch. 



In Amiurus cat^AS there is no true lateral canal. In the posterior 

 portion of the body are a series of short isolated dermal tubes. 

 The interruptions are more frequent and the tubes shorter in the 

 posterior region than in the anterior. Posterior to the operculum 

 there is a coalescence of these short tubes, and a distinct, but very 

 fine, canal is formed. At the point wliere the lateral canal enters 

 upon the head a short backwardly directed branch is given off 

 (PL XVIII. fig. 2, ,r), which passes over the supra-clavicle {McMxir- 

 rich). Immediately after entering the skull an occipital commissure 

 (PI. XVIII. fig. 2, Ocxom.') is given off, which communicates with the 

 main canal of the opposite side ; a little further on a lateral branch 

 passes off — the operculo-mandibular branch. The main canal 

 passes forwards, giving off a short branch towards the median line, 

 and posterior to the orbit divides into supra- and sub-orbital 

 branches. The sub-orbital branch passes through a series of canal- 

 bones forming a circumorbital series, from these it is continued 

 forwards into the antorbital bone. The supra-orbital branch runs 

 above the orbit, giving off at its junction with the main canal a 

 short backwardly directed branch, which runs towards the mid- 

 dorsal line ; slightly in front of the orbit another similar branch is 

 given off. Passing to the anterior region the supra-orbital branch 

 traverses a small canal-bone (PI. XVIII. fig. 2) and terminates at 

 the base of the maxillary barblet. 



Course of the Canals and Branches. 



1. The Latehal Canal consists o£ a series of isolated dermal 

 tubes ; in the posterior portion of the body they are exceedingly 

 small, but become slightly larger in the anterior region. Each 

 tube opens to the surface by one or more pores. Slightly 

 behind the operculum the tubes anastomose with one another and 

 form a fine canal, which gives off a backwardly directed branch 

 * The italicB are mine. — W. B. 0. 



