— 13 — 



(Denmark, England, South of France). From Table IV (1 — 3) it will be seen, that 

 the investigation of this feature has similarly failed to bring to light any such 

 difference; the variations of the averages as found are independent of the geo- 

 graphical distribution of the samples, that from the south of Europe (10.768) lying 

 between the two from the north (10.807 and 10.745). 



Altogether, the result of the present investigation of the number of branchiost- 

 egal rays must be stated as follows : it has been found, that specific differences may, 

 by statistical means, be shown to exist, though the variants occurring in each 

 species be the same, or even when the variant most frequently occurring is common 

 to both. The investigations further show, that even very small differences may 

 with perfect certainty be regarded as important, as long as the accuracy with which 

 they are determined is sufficiently great. 



With regard to the length at which the full number of branchiostegal rays is 

 present, it may be mentioned that larvae of A. vulgaris measuring 38 — 42 mm 

 appeared to have the full complement of rays (11 in all). 



4. Number of pectoral rays. 



In Report I, 1913, p. 5 — 6 mention is made of the investigation as to number 

 of pectoral rays in a couple of samples of older specimens of the European eel, 

 and one sample of an East Indian, short-finned species. No difference could be 

 determined by statistical means between the three samples, which fact is commented 

 on as follows (1. c. p. 7) »From this result we can perhaps conclude, that this 

 character is not suited to the separation of species. When species so distinct have 

 the same number of rays, it is improbable that any difference will exist in this 

 character between the more nearly related European, American, and Japanese eels, 

 but naturally nothing can be said on this point until investigations have been made«. 



The reason for investigating the East Indian species lay m the fact that there 

 was not sufficient material in the case of A. rostrata. Thanks to the frequently 

 mentioned sample of elvers from W. Gloucester, Mass., it has now been possible to 

 include the America eel in the investigations. At the same time, a new investiga- 

 tion of the pectoral rays has been made in the case of the European eel, in addi- 

 tion to which, a small quantity of material consisting of Japanese eels has likewise 

 been dealt with. In the previous investigations, (Report 1, 1913) no endeavour was 

 made to consistently compare the pectoral rays from one and the same side of the 

 fish, taking right and left separately, (cf. branchiostegal rays). And as the material 

 previously available for A. vulgaris was, in technical respects, not very satisfactory, 

 I have thought it best to make the whole investigation over again. 



The work was carried out in the following manner. The right pectoral fin of 

 each specimen was removed and stained with alizarine, cleared in xylol and 

 examined under the microscope. The staining process is frequenty attended with 

 some difficulty, especially in the case of young elvers, where the lower rays are 

 often found to be insufficiently coloured. On the other hand, older eels are some- 

 what difficult to deal with, owing to the necessity of removing the adjacent skin, 

 wherebj' the pectoral rays may be spilt lengthwise, occasioning some doubt in the 



