The Latero-Scnsovy Canals and E elated Bones in Fishes. 493 



Coccosteus [76 and 77], excepting that in both these lati er fishes the 

 mesial one of the two independent extrascapulars of Dinichthys has 

 fused either with the lateral extrascapular or with the parietal, both 

 of which fusions are found in living fishes. Leaving- the lateral extra- 

 scapular, the main canal, in all three fislies, is said to traverse the so- 

 called marginal and postorbital bones of Traquair's descriptions, both 

 of which bones must be squamosals, or one of them a squamosal and 

 the other a squamoso-postfrontal. In the posterior bone of the two the 

 canal is joined by the dorsal end of what seems to be a preopercular 

 canal; and in the anterior one, it turns downward to encircle the eye, 

 being joined, at or near the bend, by the groove that I have just 

 above proposed to consider as the homologue of the middle head line 

 of pit organs of Amia. At the level of the lower edge of the orbit 

 a branch is sent backward from the suborbital groove, this branch 

 apparently being the homologue of the horizontal cheek line of pit 

 organs of Amia. The groove that appears to be the homologue of the 

 anterior head line of pit organs of Amia, and the anterior continuation 

 of which is the supraorbital canal, begins on the bone that I have 

 above proposed to consider as a parietal, the frontal of Newberry and 

 Huxley [48], and runs forward onto a bone considered by Huxley as 

 a prefrontal and by Traquair as a preorbital. Because of its relations 

 to the canal I should consider this latter bone as a frontal, a nasal 

 bone perhaps lying anterior to it. The two frontal bones, as thus 

 identified, have closely the position of the frontal bones of Acipenser, 

 and they enclose between their anterior ends a median ethmoid, as 

 they may do also in Acipenser. In Coccosteus the hind end of the 

 supraorbital groove joins the groove of the middle head line, exactly 

 as, in Salmo namaycush, the anterior head line of pit organs joins the 

 middle head line. 



In Homosteus, the sensory grooves, as described by Traquair [76], 

 are much less complete than in Coccosteus, Dinichthys and Titanichthys, 

 the supratemporal and supraorbital grooves both being wanting. 



In Pterichthys and Bothriolepis the grooves agree closely with 

 those of Homosteus, if it be assumed that the middle head lines of the 

 latter fish have anastomosed, in the two former fishes, in the mid-dorsal 



