THE GEANIAL NEEVES A>:i3 LATERAL SEXfSE OKGANS OF FISHES. 181 



This occurs particularly between the frontal and nasal on each side, and between 

 the upper and lower ends of the preojicrculum and the squamosal and angular 

 [=articular] respectively." In S<diiiu, according to W. K. Parker, the upper extremity 

 of the operculo-mandibular canal between the preoperculum and the squamosal is 

 supported by a small ossicle, whilst in Gadits of course the operculo-mandibular canal 

 ends blindly and is not connected with the main body canal at all. Amia agrees with 

 Salmo in that the mandibular and body canals are connected near the posterior extremity 

 of the squamosal. There arc^ two commissures between the sensory canals of Amia — an 

 infra-orbital commissure at the extremity of the snout, and a supra-temporal commissure 

 on the back. In Gadus the supra-temporal canals, which correspond jn-ecisely to the 

 commissure in Amia, are not united across the back, whilst the infra-orbital commissure 

 is similarly absent, but the supra-orbital commissure, which is very characteristic of the 

 modern Teleostean fishes, is, on the other hand, wanting in Aii/io. 



I have previously remarked (p. 132, also ]). 205) on the anomalous condition of the so-called 

 squamosal or pterotic of Gadus (PI. 21. fig. 1, Sq.), and it is very interesting to note that 

 the post-frontal or spheuotic of Aiiiia is in a very similar condition. Allis says (pp. 1.78, 

 479) : — '• The dermal portion of the post-frontal is a small bone, somewhat triangular in 

 shape, exactly tilling a large notch extending from the middle of the lateral edge of the 

 frontal to the hind edge of the bone. Its small posterior end usually fits into a notch on 

 the anterior end of the squamosal, which overlaps somewhat its lateral edge. It rests 

 directly upon the deeper post-orbital ossification, and is so closely connected with it that 

 in attempting to remove it in fresh specimens one of the bones is usually broken, and a 

 fractured surface ol)tained ; but in skeletons joroperlj^ prepared — by maceration or 

 boiling — the two bones are easily parted, leaving a clean and perfect surface of 

 separation." These facts open up questions of considerable import, especially one which 

 I have previously mentioned, as to how far we may regard the ear ossicles (sphenotic, 

 pro-otic, epiotic, pterotic, and opisthotic) as modified lateral line ossicles. In any case 

 it seems to me that the dermal portion of the post-frontal or sphenotic of Amia should 

 be kepjt perfectly distinct from that bone, and should rank as a lateral line ossicle since 

 it is developed round a portion of the infra-orbital canal. (But cp. p. 132.) The same 

 remark applies to the separate pterotic or squamosal of Gad/ts. Both bones are precisely 

 comparable to true lateral line ossicles such as the lachrymal or preoperculum. 



Before proceeding to com25are Amia and Gadus in detail we may note that the lateral 

 or body canal of the former corresponds to the condition previously described in Gadus. 

 Allis states (p. 491) : — " Towards the tail there are usually but one or two pores in a 

 scale, and the development here is of ten so greatly arrested that the lateral canal, through 

 one or more scales, is an ojjen channel." Again, on p. 504: — "On the body, and 

 particularly towards the tail, the organs retain nearly this ^^superficial] condition even in 

 the adult; but on the head tliey are later so much withdrawn from the surface that only 

 a series of minute holes indicates their position." 



In Amia there are 47 * sense oi'gans on each side of the head, counting as far back as 

 the supra-clavicle, as against 32 in Gadus. These have the following distributioit : — 



* Allis (j>. 409) says " forty," but this is doubtless a miscalculation, since the numbers he gives total 47. 



25* 



