20 



Altogether, the investigations have shown, that the stock of Zoarces at each 

 locaHly may be characterised by its peculiar numerical position in respect of such 



Fig. 2. No. of vertebrae in 2 samples of Zoarces viviparus. Grapliical view. 

 The upper curve represents a sample from the Baltic (Bornholm), the lower one from the North Sea 



(Esbjerg). 

 The figures along the vertical line indicate no. of vertebrae. A horizontal interval of 30 mm repres- 

 ents 25 specimens. 



features, dilTering therein from the slock found on other, often closely adjacent 

 grounds. 



With regard to the combination of characters also, Zoarces is found to 

 dilïer from the samples of European eels examined. We have already seen, on 

 p. 15 — 17, that all the samples of A. onlgaris examined were found to be alike in all 

 the characters investigated (No. of vertebrae, anal, pectoral, branchiostegal and caudal 

 rays). In the case of Zoarces, all kinds of combinations may occur, as will be seen 

 from the following table. 



Two samples which are entirely identical as regards number of vertebrae may 

 prove to be altogether different in respect of one or more of the other 

 characters concerned; it is therefore not sufficient to investigate a single 

 character when endeavouring to determine how far two stocks of Zoarces are 

 different or not. 



We have seen in the foregoing, that the difference between the average number 

 of vertebrae in samples from Bornholm and others from Esbjerg was about 6.5 

 (i. e. as between 117.848 and 111.323). Even these two figures, however, do not 

 represent the extreme values. These are^ as far as the investigations hitherto have 



