424 SMITHSONIAN MISCELLANEOUS COLLECTIONS [vo. 48 
membrane. There were counted about forty acts of respiration in 
the first minutes, or the twofold movements of raising and lowering 
of the hyoid. But it now remains to account for the permanent 
closure of the opercles. The fish, meanwhile, being in the vessel, 
beat from time to.time the pectoral and ventral fins, as though it 
sought to lie down flat. 
I therefore placed some fine sea sand and water into a large recep- 
tacle, into which I put the fish. On feeling the sand with the 
paired fins (for its eyes could not see it) the fish buried itself 
in it by means of lateral movements of the tail aided by the paired 
fins, leaving in sight only the eyes and the upper portion of the 
mouth opening which formed a narrow and elongated cleft in the 
even sand layer. Farther back, toward the posterior part of the 
head, two round holes were visible on the sand, of which I shall 
speak shortly. 
In this position, the lower jaw, of which only the extreme edge 
protruded, remained completely motionless, and the slight move- 
ments which we noted while the fish was in the midst of the water, 
and which were passive, had been stopped by the obstacle furnished 
by the sand which is not so easily displaced as the former element. 
Moreover, it follows that the jaw cannot move without the fine 
sand-grains, which cover it to its edge, falling into it. The jaw 
might move slightly in the direction of closure against the upper 
jaw, but the act of approaching would not have been followed by a 
withdrawal or return to the original position, because the sand 
would push itself more and more against the jaw and finish by 
closing it completely. This same jaw is so little open that its fringe 
almost touches the upper lip, thus preventing the entrance of foreign 
bodies into the mouth, such as small crustaceans, which pass over 
the surface of the bottom and which the fish could not in any way 
keep off. Besides, as we have seen, it is necessary that the mouth 
should be slightly open to accomplish respiration. The sublingual 
membrane beat more slowly than when the fish was in the midst 
of the water. As soon as it buried itself in the sand, respiration 
stopped for a moment. Moreover, the fish entered the sand with 
the mouth shut, and on opening it on the surface of the bottom, 
the small quantity of sand which covered the mouth, fell into it 
and was then expelled again through the branchial apertures. The 
eyes, which were almast on a level with the horizontal surface of 
the head, would remain buried, if the fish did not previously sink 
them into their cavities by raising the suborbital; in fact, the fish 
makes them protrude above the surface of the sand by lowering 
