EEPRODUCTION AND DEVELOPMENT OF THE CONGER. 35 



and while a few are mature and transparent the rest are still opaque. 

 Therefore it would not be at all surprising if the eggs of the conger 

 were transparent and buoyant when perfectly mature and ready for 

 fertilisation. This probability is made almost a certainty by the 

 observation by Voeltzkow, already cited, on the ripe ova in a speci- 

 men of Mursena found at Zanzibar. The eggs in this case were 

 perfectly transparent, and, therefore, probably after fertilisation 

 would be pelagic. 



The Italian naturalist Raffaele in his valuable paper on the 

 Pelagic Eggs and Larvse of Fishes occurring in the Gulf of Najjles,"^ 

 published in 1888, described five different kinds of pelagic eggs, 

 which all resembled one another in certain common characters, and 

 which could not be traced with certainty to the parent fish. 

 Rafltaele thinks it possible that these eggs belong to various species 

 of the eel family (Muraenidge) . He bases this suggestion on the 

 form of the body, the form of the head, and the large number of 

 muscular segments in the larvae hatched from the eggs. The eggs 

 all agree in having an extremely large perivitelline space, like that 

 of the pilchard's egg, and in the fact that the yolk is not homo- 

 geneous but made up of separate vesicles, also like that of the pilchard. 

 This similarity to the eggs of the pilchard is an important matter. 

 For the family Clupeidae is the only one among the Physostomi 

 hitherto known to include species with pelagic eggs, and these eggs 

 are distinguished from the eggs of Physoclisti by the two characters 

 above mentioned. Therefore it is in the highest degree probable 

 that Raffaele's eggs belong to some family of the Physostomi, and 

 the Muraenidae is the only family among these in Europe whose 

 eggs are not known. 



We may consider it, then, as all but proved that the eggs of the 

 Muraenidge are pelagic, and that to Raffaele belongs the credit of dis- 

 covering them. In size the eggs described by Raffaele agree very 

 well with those of the conger which I have measured. Unfortu- 

 nately he only gives the diameter of the actual ovum inside the 

 capsule in one case, in which it was r2 to 1'3 mm., scarcely larger 

 than the unripe ova of the conger measured by me before the 

 formation of the perivitelline space. The diameter of the external 

 capsule in RafEaele's eggs was 2 to 3 mm., all the five kinds, except 

 one, having a varying number of oil globules. I am not sure that 

 the egg of the conger when ripe is without oil globules, but so far 

 as I could judge it is so. In this case the egg without oil globules 

 among those described by Raffaele is probably th.at of Conger vulgaris. 

 The larvae hatched from these eggs were, as I have said, all very 

 similar. Besides the large number of body segments, they all agreed 

 * Mitt. Zool. Stat. Neap,, Bnd. viii. 



