682 



PROCEEDINGS OF THE NATIONAL MUSEUM. 



are formed in a very different manner in th3 various species. In 

 foliacem they consist of a row of chitin joints so short as to appear 

 spherical or slightly elliptical (Claus). In amerlcanus they consist of 

 two concentric rows of chitin rods, each of which is shaped like the 

 letter J (fig. 13). 



The rods in the outer row are much smaller than those in the inner 

 one, and thev do not reach nearly to the edge of the membrane. In 

 megalops the rays are made up of a series of trough or gutter shaped 

 plates overlapping one another like shingles (fig. 14). 



The base of the membrane is strengthened by three chitin rings, the 

 basal one of which is large enough to hold its shape perfectly against 

 the contraction of the muscles. 



The terminal one serves as a support for the ravs and to it their 

 proximal ends are attached. This border contains two sets of muscles, 

 circular and radial, by which it can be elevated till it assumes the 

 same direction as the cylindrical base and simply forms a hollow ter- 

 minal portion to the latter, or it can be depressed till it stands out at 

 right angles to the base like the brim of a hat. It often assumes the 



first position on fixation, but the last 

 one is the normal posture in the liv- 

 ing animal. Its membraneous struc- 

 ture, strengthened by the chitin rays, 

 with the flexibility produced by the 

 joints in the latter, and the serrated 

 edge, enable it to fit down snugly to 

 the rough surface of the fish's body 

 and produce a tight joint. The lumen of the base is nearly filled by 

 four large muscles arranged in pairs, approximately right and left 

 and anterior and posterior. By the contraction of these muscles a 

 partial vacuum is formed inside the base and the flexible border is 

 securely fastened to the surface on which it rests. The simple 

 relaxation of the muscles restores the original lumen and the disk is 

 detached. 



By relaxing its hold with one disk and carrying it forward while 

 the other remains fastened the ardmal moves about with surprising- 

 rapidity. Not only is Thorell wrong when he says (1864) that these 

 appendages are " used exclusively as fixing organs," but it seems 

 highly probable that they have been modified into this disk form for 

 the express purpose of functioning as locomotor organs. And while 

 the function of fixation may, and probably does, still take precedence, 

 yet that of locomotion becomes a close second. The relative size of 

 these disks varies greatly; in some species {fuRdull, latus) it reaches 

 nearly a third the width of the carapace, and be it remembered these 

 are forms in which the carapace is orbicular. In others (Idtimuda^ 

 megalops) it is not more than a sixth or a seventh of that width. 



Fig. 14.— Border of sucking disk in Argu 



l.US MEGALOPS. 



