— 164 — 



As in other species of Leptocephalus the mouth is provided with a characteristic 

 equipment of larval teeth. In the upper jaw there are the following (see Fig. 6 A) ; in 

 front there is a pair of strong, projecting grasping teeth, above which is a pair of very- 

 thin and slender teeth: behind the large grasping teeth in each half of the upper 

 jaw there are ca. 14 — 18 pointed teeth pointing obliquely (downwards), which decrease in 

 size from before backwards. As a .rule however the first 5 — 8 have a somewhat difïerent 

 character from the succeeding, being usually considerably larger. In the underjaw there 

 are in addition to the front pair of large projecting grasping teeth ca. 17 teeth in each 

 half of the jaw, the first of which is small and slender, whilst the others decrease fairly 

 evenly in size backwards. The teeth in the underjaw are often difficult to count however 

 as they are partly concealed by the margin of the upper jaw. 



The position of the anus is very far back at this stage. Thus the distance from 

 the anus to the end of the tail is somewhat less than '/s^d of the total length. 



A question of considerable importance was, whether the variation in the June material 

 of Leptocephalus brevirostris with respect to size and form might arise from individual 

 variation or from the different specimens not being in the same stage of development. 



It was therefore necessary to find some characters which might be used to measure 

 the degree of development. From Grassi and Calandruccio's investigations we know 

 that various changes occur during development from the youngest known stage,' L. bre- 

 virostris, to the elver stage. The most important changes are: (1) the larval teeth gradually 

 fall away, (2) the anus as also the anterior points of the dorsal and anal fins move forwards, 

 (3) the height of the body and the total length are reduced. It should be mentioned first 

 of all that all the specimens (with perhaps a single exception) had passed through the 

 developmental phase preceding the metamorphosis and which in biological regard may be 

 compared for example with the larval stages of the insects. This appeared from the fact 

 that the alimentary canal was constantly empty of food. I have examined about half a 

 hundred specimens in this connection from the most different localities and become con- 

 vinced of the fact. It was only in a single specimen (the smallest of 60 mm.) that the 

 posterior part of the gut contained a trace of a pulpy mass which might possibly be 

 considered as digested food, though its nature could not be determined. Otherwise the 

 digestive tract was empty or contained only some broken off epithelial cells'. It must 

 therefore be concluded from this investigation, that the fish were no longer feeding 

 at the present stage, and that the feeding and growth stage which precedes 

 the metamorphosis must therefore have been passed by practically all the 

 specimens. 



The question then arose to determine, whether the retrograde metamorphosis, which 

 results in reduction of height and length, had begun in some of the specimens, and 

 whether the variation undoubtedly present in the material, might be referred to this and 



I In order to obtain confirmation of this condition from another side I asked Cand. A. Brinkmann to 

 examine some specimens in this connection. The result of his investigation was the same negative one as mine. 



- For the sake of comparison I may mention that I have examined the digestive canal of numerous 

 pelagic Gadzt.s-yo\x-a% in all stages of development and always found that they contained food. Further the 

 condition is the same with the young of the flat-fishes, though whether these or some of them, like the eel, 

 do not feed during metamorphosis which to my mind may well be compared with that of the eel, I do not 

 know. That the completely metamorphosed young flat-fish are often smaller than the stages prccceding the 

 metamorphosis might indicate this. 



