1879.'] NATURAL SCIENCES OF PHILADELPHIA. 153 



nerve-buttons, and in which terminates a nerve fibre. He corrobo- 

 rated Leydig's statements regarding the existence of a little mass 

 of viscid mucous or jelly-like matter resting on each disk, and also 

 regarding the positions of the disks, i.e., as generally occurring 

 under small bridges of bone in the canals of the head, and as 

 occurring in every scale of the lateral line. However, the speci- 

 mens exhibited by him showed a result entirely different from that 

 of Leydig as regards the distribution of the nerves. Leydig 

 pictures the disk as composed of two distinct areas, a dark or 

 less translucent central portion, and a lighter peripheral portion. 

 This appearance is, indeed, simulated in the fresh preparation, as 

 the disk is somewhat transparent, and allows the insertion of the 

 nerve fibre to be seen directly through it. No such appearance, 

 however, is presented in specimens treated by osmic acid. A 

 dense, arborescent plexus of nerve-fibres comes into view, and the 

 distinction into two areas entirely disappears. 



The size of the entering nerve as compared with the size of the 

 disk is relatively very great, so that when the dense plexus of 

 nerves makes its appearance it strikes one as though the bulk of 

 the disk were nerve matter. Indeed, besides a large number of 

 capillaries it contains only a small amount of connective tissue. 



Owing to the want of the proper material, Dr. Dercum had not 

 been able to confirm the observations of F. E. Schulze on young 

 fishes. However, transverse sections of the disks macerated in 

 osmic acid and teased, yielded essentially the same results as re- 

 gards the structure of the epithelium. Tins appears to consist of 

 two kinds of cells, one long and cylindroid, the other small and 

 globose or pyriform, and having long outrunners penetrating the 

 subjacent connective tissue. These outrunners are probably 

 continuous with nerve-fibres, which they resemble. The drawings 

 representing the connections of the nerves with the cells, given by 

 F. E. Schulze for the " nerve-hills" of young fishes, are, therefore, 

 very probably correct, but the material at hand did not permit an 

 absolute decision of the point. The hairs of these perceptive cells 

 were readily distinguished, but were generally broken and muti- 

 lated, owing doubtless to the reagents and teasing. 



As is well knOwn, the canals of the head are generally provided 

 with bony supports, which form grooves, and which are at intervals 

 generally bridged over by bone. Thus certain membranous inter- 

 spaces of greater or less size are produced. In some fishes, as 

 Centropomus undecimalis, the bon}' bridges are more or less 

 wanting, so that a comparatively large expanse of membrane is 

 formed stretched between the two walls of the canal. These 

 membranes are, of course, composed of two elements. They 

 consist, first, of the delicate connective tissue and flattened epi- 

 thelium belonging to the canal, and, secondly, of the dense layer 

 of connective tissue and epithelium belonging to the skin. The 

 two layers can be readily separated by a careful dissection. Each 

 11 



