— 165 — 



thus indicate different stages of development in the different specimens. Our Leptocephali 

 appear to vary not only in total length (smallest 60, largest 88 mm.) but also in the 

 height of the body and thickness of the gut, especially the posterior part, which first 

 disappears when metamorphosis begins. Further there was an obvious though quite small 

 difference in the position of the anus, which in some specimens lay slightly further for- 

 wards than in others. In order to find a basis for comparison I divided the specimens 

 into three groups with respect to the height of the body and thickness of the gut. Thus 

 a distinction was made between specimens in which the body-form was broad — medium 

 narrow, and in regard to the thickness of the gut between thick — medium — thin. 



The following characters were now examined in different specimens from the most 

 different localities and depths, viz. ( l ) body-height, (2) thickness of gut, {3) position of 

 anus, and (4) number of teeth in upper jaw'. 



The accompanying table shows the result for a number of specimens. 



I I first thought it might be possible to use the number of fin-rays a.s a criterion for the degree of devel- 

 opment and a number of specimens were therefore counted (by Cand. Strubberg) both of the June material 

 and of the partly metamorphosed specimens of September ist. The few numbers given here show at once 

 however, that the number of fin-rays can hardly be used as we can find half-metamorphosed specimens in the 

 September material with considerably fewer rays than June specimens. 



Leptocephalas brevirosiris taken by the "Thor" in the Atlantic Ocean 



