20 



tured four. At the end of June, when we ceased finding 

 them, we had as many as fourteen Leptocephali Hving in 

 the glass aquaria. They were then measured one by one 

 and placed in two aquaria, which had a layer of sand on 

 the bottom. The sizes ranged from 2/^ to 2jjy inches in 

 extreme length, -^ of an inch in vertical depth, from fin to 

 fin, and about yV of an inch thick. They were flat, colour- 

 less, and perfectly transparent, the viscera, and the heart 

 and its movements, being easily seen through the skin. 



On being placed in the aquaria, the Leptocephali swam 

 swiftly round the sides, with an undulating movement, 

 like that of the sand-eel. They soon settled down to 

 their new surroundings, and quickly buried themselves in 

 the sand. The movements gone through in burrowing 

 are exactly similar to those made by the sand-eel. The 

 .head is first directed into the sand, then by a rapid back- 

 ward and forward movement of the posterior part of the 

 body, the anterior part is forced into the sand, and finally, 

 by a gliding motion, the posterior part disappears. During 

 the day-time the Leptocephali remained hidden away in 

 the sand. On the slightest disturbance of the water, such 

 as would be caused by the throwing in of food, their heads 

 -would be thrust out and a rapid survey taken to ascertain 

 •the nature and position of this disturbing element. If it 

 were food that happened to fall close to them, they would 

 seize it without coming entirely out of the sand, and 

 would then glide backwards into the burrow. If the food 

 did not fall within reach, they would not venture to pur- 

 sue it in daylight. On going into the tank-room at night, 

 when all was in darkness, and suddenly flashing a light on 

 the aquaria, it would usually be seen that the Leptocephali 

 were swimming about actively, but soon retired into the 

 sand if the light were continued. 



The Leptocephali were kept under observation for a 



