508 REPOET OF COMMISSIONER OF FISH AND FISHERIES. [54] 



their ends. Beyond this stage I have not followed them, and it only 

 remains to suggest that the canals are further developed by growth in 

 length, in the course of which the curved tubular portions are elevated 

 and separated from the utriculus. As regards the sacculus, I have no 

 observations of value to record; this structure, as far as I can make 

 out, seems to be developed during the post-larval period. The connec- 

 tion of the auditory vesicle with the air-bladder seems also to take 

 place at a late period, for in all the forms observed b^' me the diverticu- 

 lum of the fore-gut, which gives rise to it, is quite rudimentary, even up 

 to the time when the semicircular canals of the ear have been formed. 

 It only remains for us to call attention here to the fact that both the 

 auditory and olfactory organs are less intimately connected with the evo- 

 lution of the neurula or larval nervous system than the eyes, which are 

 connected with it from the first moment of their development, seeming, 

 in fact, to be mere outgrowths of that system. The internal ears and 

 the nasal organs, on the other hand, are formed as paired involutions 

 of the epiblast, their connection with the nervous system being estab- 

 lished in a manner entirely different from that of the eyes. 



13. — The lateral sensory organs of the larval cod. 



These organs have a rather singular distribution in the young cod- 

 fish just hatched. There are five of them to be seen on either side of the 

 body, as may be noticed in Figs. 40 and 42, but on the head and on the 

 side of the body they are not placed on the middle of the side, as they 

 are on the tail. About three of them are placed on either side of the 

 tail, as may be seen in Fig. 42, at sh. A nerve filament, nf, Fig. 43, 

 passes out to each of them, and is presumably connected with the nerv- 

 ous system, but the exact relations of these nervous connections I have 

 failed to make out. The nerve fiber which passes out to the one on the 

 head, Fig. 40, sh, appears to arise from the medulla oblongata ; in the 

 one behind the pectoral fin, on the side of the body, at the base of the 

 dorsal natatory fold, the nerve going to it seems to arise from the 

 spinal cord. In both cases faint filamentous prolongations from these 

 two nerve eminences are seen to be prolonged anteriorly and posteriorly 

 in the skin. These filaments, I take it, represent the nerve of the lateral 

 line, evidence of the presence of which is seen in the serial arrangement 

 of the sensory eminences, sh, themselves. They are not nearly as numer- 

 ous as the muscular segments, a feature in which the larval cod difiers 

 greatly from the larva of Gambiisia patruelis where these sensory eleva- 

 tions correspond exactly to the number of muscular segments. This 

 segmental arrangement of the sensory eminences or nerve hills has also 

 been noticed in other larval fishes by Schulze, and is an exceedingly 

 interesting fact. In a good many other forms of larva? of osseous fishes 

 these lateral sensory eminences are not developed at all at the time of 

 hatching. This is the case with Alosa and Pomolobus. We have there- 

 fore all grades of their development in known types, from none to a few 



