356 FISHES 



less definite than that of the larvae. In 1888, as already men- 

 tioned, Raffaele described five different kinds of eggs obtained 

 at Naples, which, from the structure of the larvae hatched from 

 them, he considered to belong to species of the eel family. 

 These eggs were large for pelagic eggs, and most of them pos- 

 sessed oil-globules ; one kind however was without an oil-globule, 

 and this one was considered by Grassi and Calandruccio to be 

 that of the common eel ; this opinion has not been confirmed, 

 for the structure of the immature eggs in the roe of the conger 

 and eel indicates that the eggs when shed are provided with an 

 oil-globule. No eggs similar to those of Raffaele have been yet 

 taken in the Atlantic on the European side, but in 1900 some 

 pelagic ova were taken about thirty miles from the American 

 coast in July which hatched into Leptocephali and were con- 

 sidered to belong probably to the conger ; these eggs had a 

 single large oil-globule. These eggs appear to have been taken 

 at the surface, but as such captures are so rare it seems pro- 

 bable that, as Schmidt believes, the eggs of Anguillidas are not 

 only shed at the bottom at great depths, but usually remain 

 there till hatched, suspended in the water ; Schmidt has shown 

 that the eggs of Argentina develop in this condition, and he 

 calls such eggs bathypelagic. Two Leptocephali were also 

 taken by the Americans off the coast of the United States and 

 identified by Eigenmann and Kennedy as belonging to the 

 American fresh-water eel, which, however, is distinct from the 

 European species and is called Anguilla chrysippa. 



The ages to which eels attain before migrating to the sea in 

 order to spawn have been investigated by Gemzoe at the 

 Danish Biological Station. Petersen's statistical method by 

 which numbers of specimens are separated into age-groups 

 whose ages differ by one year, was found to be applicable only 

 to the younger eels, because in consequence of variation in the 

 rate of growth the annual groups in the larger specimens gradu- 

 ate into each other and can no longer be distinguished. The 

 age of the mature eels was ascertained by counting the number 

 of annual zones on the scales, the otoliths which exhibit such 

 distinct lines of growth in plaice and other fishes being small 

 and useless for age-determination in the eel. The scales of 

 the eel are rudimentary and completely embedded in the skin, 

 but like those of other fishes they exhibit on the outer side con- 



