666 E. PH. ALLIS, 



beneath the thin skin of the fish, and not "on the dorsal surface of 

 the dermo-sphenotic of Bridge"; an anterior ossicle, or the anterior 

 portion only of that ossicle, alone lying dorsal to the posterior portion 

 of the above mentioned bone. This anterior ossicle of the series lies 

 directly and closely upon the so-called dermo-sphenotic, the posterior 

 ossicles of the series being separated by a considerable space, tilled 

 with connective tissue, from the underlying arm of the bone CK 



This series of lateral canal ossicles varies in each one of my 

 several specimens, and also on the two sides of each specimen. What 

 would seem to be the primitive arrangement was found on one side 

 only of one of my specimens, and is shown on the right side of the 

 accompanying figure (Fig. 1, PI. 28). Here there were five ossicles 

 in the chain ; and a primary tube, here the trunk of a dendritic 

 system, arose from the canal between the first and last ossicles of the 

 series and the bones C^ and C- respectively, and also between each 

 two ossicles of the series excepting only the first and second. The 

 tube that probably primarily had its origin from the canal between 

 these two last-named ossicles arose at about the middle of the first 

 ossicle of the series, from its lateral aspect. It was enclosed in bone 

 for a considerable part of its length, differing in this from the other 

 tubes, all of which were simple dermal tubes without bony envelope. 

 Opposite the point where this tube probably arose, that is, between 

 the first two ossicles of the series, from the mesial aspect of the canal, 

 a branch canal arises and extends mesially nearly to the middle line 

 of the head, where it ends in a large dendritic system. Whether 

 there is here but one terminal system, or one such system and a 

 closely adjoining penultimate one, could not be determined, but as 

 the branch canal is enclosed in but two ossicles there is probably 

 here but a single system. A second or third system, as the case may 

 be, arises from the branch canal soon after it leaves the first one of 

 the two enclosing ossicles. It may be directed, in diÖ'erent specimens, 

 either forward or backward, and clearly indicates that we have here 

 to do with a commissural canal that contains at least two sense or- 

 gans. CoLLiNGE does not consider this canal as a commissure, but 

 this is probably because he would limit that term to a full cross- 

 commissure, which this canal certainly is not. 



The two or more sense organs of the above described commissure 

 are probably innervated by a supratemporal branch of the nervus 

 lineae lateralis; for Collinge says (p. 519) that he could find no 



