18 



EINAR LEA. 



[REP. OF THE "MICHAEL SARS" NORTH 



larvae have not yet attained lull length, and that two of them 

 were found over great depths, bear out Schmidt's con- 

 clusions in regard to the immigration of these larvae to- 

 wards shallower waters in course of their development. 



6. Leptocephalus Cong-ri mystacis. 



Grassi asserts in his work on "The Reproduction and 

 Metamorphosis of the common Eel (Anguilla vulgaris)", 

 Proc. Roy. Soc, vol. LV, 1896, that the larvae of Con- 

 gromuraena mystax are identical with those described 

 under the names of Leptocephalus Haeckell, Yarrelli, 

 Bibroni, Gegenbauri, Kollikeri, and stenops (in part), in 



Fig. 12. Head of larva no. 1. 10 /i. 



the descriptions of which the number of segments is not 

 stated. The long list of synonyms is a characteristic 

 proof of the difficulty in determining larval Apodes when 

 no account is taken of the number of segments. Grassi 

 does not mention this point, and I should certainly have 

 added another name to the list of synonyms, had not 

 Schmidt in one of his recent works told us that Conger 



Fig. 13. Head of larva no. 4. 10 /i. 



mystax has about 140 vertebrae. This information, and 

 the excellent photograph in Schmidt's paper (29) enable 

 me to describe 20 larvae taken by the "Michael Sars" under 

 their right name. 



The larvae in question, one of which is reproduced 

 in fig. 1 pi. Ill, differ greatly in size (from 31 to 113 mm 

 long) and represent various stages of development; nine- 

 teen of them, not having rays on all the interspinous 

 elements in the fins, are not yet full-grown, while one 

 has begun its metamorphosis. The smallest larva (no. 1 

 in the table) is pronouncedly a prelarva, for, as far as 



can be seen, there are no indications of interspinous 

 elements in the fins (the tip of the tail is wanting, so 

 that nothing can be said of the development of the caudal 

 fin), and its nostrils are not separated. In a somewhat 

 larger larva they are almost separated, and in most of 

 the others they are well separated. 



The pigmentation is in accordance with former de- 

 scriptions, for we find spots along the alimentary canal 

 and at the bases of both the anal and caudal fins, and 

 and as a rule there are also a few spots directly in front 

 of the curve on the alimentary canal where it leaves 

 the head. 



Figs. 12, 13 and 14, representing the heads of the 

 specimens numbered 1, 4 and 20 in the table, show how 

 the smaller individuals differ from the larger in the deve- 

 lopment of the nostrils, and in the relative size and 

 number of the teeth. 



The results of the measurements of each specimen 1 ) 

 are given in the table on p. 19. 



Fig. 14. Head af larva no. 20. 



No. 11 is the larva which has begun its metamor- 

 phosis, as shown: 1) because its height is small relatively 

 to its length, 2 ) 2) because of the strong development of 

 the vertical fins, and 3) because the anus is placed 

 farther forward, and there are consequently fewer preanal 

 segments. On the other hand it is evident that its meta- 

 morphosis is not very far advanced, for the teeth are 

 intact and the dorsal fin is not farther forward than in 

 the others. Excluding no. 1 1 the table does not indicate 

 any distinct variation in character, except that the teeth 

 are fewer in the smaller larvae than in the larger, and 

 that the nostrils of the larvae numbered 1, 2, 3 and 4 

 are not separated. 



The number of muscle-segments, however, varies 

 between 132 and 147, and for a while this made me 

 suspect that there were several species among the 20 



1 ) No. 21 in the table is described in a concluding note. 



2 ) The larva being quite deformed, I have not been able to measure the height exactly. 



