JoHS. Schmidt 111 



populations there found. Dividing up the area into four parts, and 

 utilising the same symbols as above, we obtain the follovv^ing : (1) western 

 North Sea, Abe; (2) eastern North Sea, abc; (3) west Baltic region, 

 ABC ; (4) east Baltic region, ABc. On taking together three popula- 

 tions, typical as far as possible, from each of the four regions, we obtain 

 the following mean values : 



TABLE I. 



Average value for Vert. Pd iJj Pigm. Di 



I. Noith Sea 1. Western part 116-2 18 71 7-2 12-7 



„ 2. Eastern part 111-2 18-48 6-0 12-3 



II. Baltic 3. Western part 1177 19*35 8-0 14-3 



4. Eastern part 117-2 18-66 11-1 123 



On going into details, we find many discrepancies and irregularities 

 in the geographical distribution of the mean values. A certain regu- 

 larity, on the other hand, and also highly peculiar conditions are noted 

 in the shallow Danish and other fjords of the western Baltic. The 

 rule here is that the average number of vertebrae, hard rays and pig- 

 ment spots will be essentially lower up in the ^ords than outside, and 

 this applies even where the distance between the two populations only 

 amounts to some few miles. The graphs in Figs. 5, 6 show some in- 

 stances of this as regards the number of vertebrae. Table II, giving 

 the values for Manager Fjord, east coast of Jutland, Denmark, shows 

 the same also for the other characters. 



TABLE 11. 



Mariager Fjord, Denmark. 



Average value for Vert. D.^ 



Stat, 14 li7-37 9-21 



r„ 15 115-43 8-74 



J „ 16 110-99 7-21 



{_ „ 17 ■ 110-18 6-87 



„ 19 109-30 6-40 



St. 14 lies outside the jQord ; the remaining stations up in the ^ord ; 

 St. 19 at its base. 



The lower number of vertebrae in the Zoarces of the ^ords gives 

 them a physiognomy differing greatly from that of their relatives on 

 the coasts, the former being of short and stumpy build, the latter 

 slender and elongated in shape. 



