40 
eels, their study became of rnuch greater interest, but it must be 
confessed, that the older descriptions of the Leptocephali are so 
imperfect and unsatisfactory, that a certain determination of the 
species from them is impossible; naturally, any attempt to discuss 
the distribution of the species on such a basis is still more im¬ 
possible. It may be said that in addition to a good figure, the 
main thin g required in a careful description is the exact number 
of my orne res. Gr assi and Calandruccio pointed out that the 
number of myomeres in the Leptocephali corresponds to the number 
of vertebræ in the parent form. This faet I have fiequently con- 
firrned and we thus have a rueaiis of referring a Leptocephalus to 
its parent species as well as of distinguishing between the different 
Leptocephali, even when in other characters there is apparently no 
difference. 
Tliough Del age showed that L. morrisii is the larva of Gonger 
vulgaris and Gr as si and Calandruccio, on confirming this, 
added that several Leptocephali known from the Straits of Messina 
under various names are the larvae of C. mystax and C. balearicus, 
yet. up to the present time, figures and descriptions of these thiee 
Leptocephali wliich would enable us to distinguish them from neaily 
related forms have not been published. I have found it necessary, 
therefore, to study the question anew from the very beginning in 
the only way possible, namely, by counting the vertebræ in the 
larval and adult forms. In this way I have been able to determine 
the larval stages occurring inside the territory investigated, which 
includes the Mediterranean and the North Atlantic north of ca. 
20° N. L. 
The full-grown larvae of the Conger-species mentioned differ 
in regard to size. C. balearicus has the largest larva (ca. 20 cm.), 
C. mystax the smallest (ca. 13 cm.), whilst the larva of C. vulgaris 
may attain a length of ca. 16 cm. All three species have long 
ribbon-like bodies with the anus far behiud near the beginning of 
the tail, nearest in C. balearicus, furthest away in C. vulgaris 
(see figs. 1—6). 
