- i68 — 



ist Stage Pl.VII, Fig. 1 — 14; PI. VIII, Fig. 1—2 

 (Place and time of discovery: Atlantic Ocean over deep water in May— August; ca. 265 

 specimens). 



The body compressed, almost of the shape of a Nerium-leaf; ca. 18 larval teeth in 

 each half of the upper jaw. The distance from the anus to the end of the tail varies 

 from ca. 28 — ca. 34 % of the total length. Pigment is quite wanting. 



2nd Stage I PI. VIII, Fig. 3—6 



(Place and time of discovery: Atlantic Ocean over deep water in beginning of Sep- 

 tember; 4 specimens). 



The body is still essentially of the same form as in the first stage; but the parts 

 nearest the head and the end of the tail are principally narrower than there. All or the 

 greater part of the larval teeth have fallen out. The distance from the beginning of the 

 anal fin to the end of the tail varies from ca. 45— ca. 53% of the total length. A trace 

 of internal pigment begins to appear at the end of the tail. 



3rd Stage PL VIII, Fig 7—8 



(Occurrence same as that of the 2nd stage; 3 specimens). 



The body is still high at the centre but in front and posteriorly is considerably 

 narrower than in the earlier stages, so that the head especially stands out more distinctly. 

 Larval teeth are wanting. The distance from the beginning of the anal fin to the end of 

 the tail varies from ca. 58—61 % of the total length. In addition to the internal pigment 

 at the end of the tail a trace of surface pigment has begun to appear at the base of the 

 caudal fin. 



4th Stage 2 PI. VIII, Fig. 9 (Stage 4 A) and Fig. 10 (Stage 4 B) 

 (Place of discovery as 2nd and 3rd stage, from September— November ; 2 specimens). 



1 My ind Stage is not figured or described by Grassi and Calandruccio. The Italian authors' Fig 3 

 (1S97, I.e. p. 12; see p, 150) corresponds best to my 3rd Stage, and their Fig. 4 quite agrees with my 4th 

 Stage as also their Fig. i with my ist Stage. On the other hand, I have no stage con-esponding to Grassi 

 and Calandruccio's Fig, 2 (p. n), which in regard to form stands between my 3rd and 4th Stages, but in 

 regard to the position of the anus comes nearest to my ist Stage, the postanal portion of the body (to judge 

 from the figure) being only ca. 33% of the total length. If this Italian specimen is correctly figured as regards 

 the position of the anus, there is this difference between the metamorphosis of my (Atlantic) specimens and the 

 Mediterranean ones, namely, that in the former the anus moves a considerable distance forwards before the 

 height of the body shows any great signs of reduction, whilst this is not always the case in the Mediterranean 

 specimens. Further, if we compare Grassi and Calandruccio's Figg. 2 and 3, the latter gives one the im- 

 pression of being more reliable than Fig. 2 and also corresponds better to my photographs given on PI. VIII. 



2 I am unable to understand the remarks made by Gr.^ssi and Calandruccio, 1897 p. 12 and 13. They 

 first describe 3 "semilarvae" taken at the end of 1892, which in development come between their Fig. 3 and Fig. 4 

 (thus at my 3rd stage). It is then stated that these 3 specimens were kept in the aquarium and became me- 

 tamorphosed to eels with the usual reductions, after which they began to take food and to grow. According 

 to this we may conclude, that the whole metamorphosis, i.e. passage through such stages as the Italian authors' 

 Fig. 5 (= my 5th Stage) and Fig. 6 (^ my 6th Stage) had been gone through, as according to the avail- 

 able information (from Grassi and Calandruccio, A. C. Johansen and myself) the fish take no food during all 

 this time. This does not occur before the metamorphosis and reduction are complete. Again however the authors 

 in describing the specimen represented in their Fig. 4 (nearly = my 4 th Stage) which came from the meta- 



