THE EEL FAMILY. 447 



that the larval form buries itself in the sand or mud and thus 

 escapes the trawl. Leptocephali kept in confinement show a 

 marked tendency to bury themselves and a liking for dark holes 

 and corners, and the young eel dug out of the sand, already 

 referred to, lends credence to this suggestion. We have also 

 seen that many of the larval sand-eels remain embedded in 

 the sand until the yolk is nearly absorbed, and this habit has 

 prevented their capture up till now. 



Another Italian naturalist, Raffaele, describes five different 

 pelagic eggs of unknown parentage which he is inclined to 

 regard as belonging to fishes of the eel-family ; they have a 

 large perivitelline space (cf. Long-rough dab) and a vesicular 

 yolk, and the larvse hatched from them have a large number of 

 muscle-segments, a peculiar head and body, and a series of 

 long teeth in the jaws. Grassi and Calandruccio claim to 

 prove the suggestion of EafFaele that these are eggs and 

 larvEe of MurwnidcB. 



One of these in particular, viz. Raffaele's undetermined 

 species No. 10, which is devoid of oil-globules, has a diameter of 

 2'7 mm., and gives birth to what Grassi calls a prse-larva with 

 long projecting teeth (Plate XX, fig. 6), and only 44 abdominal 

 muscle-segments (myomeres), he associates with the eel. After 

 twenty-four hours the larva (Plate XX, fig. 7) presents a 

 trowel-shaped lower jaw with developing teeth. A further 

 advancement is seen in Plate XX, fig. 8, which is a larva 

 one day older than the preceding — in which the remarkable 

 armature of temporary teeth is fairly developed. The anterior 

 end (Plate XIX, fig. 9) of a more advanced, though probably not 

 older larva, presents very large ear-capsules. All the foregoing 

 were hatched by Mr Williamson towards the end of August, at 

 Naples, and belong to Raffaele's species No. 8. As regards 

 general form, teeth, and other points such larvse are similar to 

 that derived from egg No. 10 of Raffaele, which Grassi con- 

 siders to be the egg of the eel. They differ, however, in the 

 number of the abdominal segments, which are fewer in the 

 larva of No. 10, and moreover the yolk of the latter has no oil- 

 globules. 



The preceding 'prse-larva' becomes, according to Grassi, a 



