- 166 - 



From these and other examples the conclusion can only be drawn that the different 

 characters do not vary together. Incipient reduction of the posterior part of the gut is 

 thus not always accompanied by a relatively forward position of the anus nor by a smaller 

 body height nor by fewer teeth. On the contrary all possible combinations of the devel- 

 opmental stages of the different characters occur in the same specimen. This clearly 

 indicates that all the specimens are in essentially the same stage of development, 

 and this becomes still more apparent if we compare the material with the specimens 

 taken later in the year on September 1st, all these being very different from the June 

 material as will be seen from the description p. l68 and figures PI. VIII, Fig. 3—9. 



Concerning the June material, finally, the following may be noted, (l) The larval, 

 teeth, whose partial disappearance is stated by Gkassi and Calandruccio always to be 

 the beginning of metamorphosis, do not show the least sign of being in process of 

 reduction in any one of my specimens. A few here and there were broken or had fallen 

 out (as result of damage), but on the whole the rows of teeth could be considered as 

 quite intact in all specimens in which the mouth was sufficiently open to permit of a 

 close examination. (2) Position of the anus. As mentioned, the distance from the 

 anus to the end of the tail is a little less than '/s^d of the total length. Expressed in 

 percentage of the total length it was found to vary from ca. 28 to ca. 34, but was usually 

 about 31. (3) Thickness of the gut. Two extreme cases are represented in Fig. l — 2 

 on PI. VIII. As a rule the condition was almost halfway between these extremes. 4) Fin- 

 rays. No difference in the number of fin-rays as a measure of the degree of develop- 

 ment could as mentioned above be detected. In some specimens the anterior developing 

 rays in the dorsal and anal fins were somewhat more distinct than in others, but there 

 was not much difference. (5) Size ofthe eyes. A little difference in the size of the 

 eyes could now and then be detected. 



All things considered it appears quite clear that the differences between the 265 June 

 specimens must be mainly referred to individual variation (cf. PL VII), and that they were 

 all essentially at the same stage of development', as can also be seen from 

 the measurements represented graphically on p. 163. (If some of the specimens had not 

 been full-grown, the curve would not have shown so steep a descent from the maximum 

 at 75 mm. down to the smallest occurring sizes). This stage of development in 

 which the fish no longer take food indicates the maximum point of the 

 larval development, and it is a very obvious feature which deserves special mention, 

 that the developmental series ceases so abruptly both above and below 

 (cf. Section II, Chap. 4). 



Of the earlier developmental stages, the preleptocephalic stages, we know nothi^% as 

 yet; the next (2nd) stage in development is however well-known and is characterised by 

 the approach of metamorphosis during which the body becomes reduced both in height 

 and length, as will be described later. 



(b) The appearance of the larvae in September and their metamorphosis. 



As already mentioned, we took 8 specimens of Leptocephalus brevirostris or rather 



I With exception of the specimen at 60 mm. which probably was not full-grown, as can be seen from the 

 measurements on p. 172 of the specimens in the 5th stage taken in January, several thousands of which were 

 examined without any less than 64 mm. being found. Whether the Leptocephalus at 60 mm. would ever have 

 become fullgrown and developed still further is a different matter. 



