1902] The Holotlinrioidea of Northern Norway. 29 



bution of the drift -ice. Oetmann does so (1896), he lets the border 

 of the Arctic „subregion" coincide with that of the drift-ice during" 

 -the summer. Thus the difference between the European and the 

 American coasts will be sharply represented. Still the limit will 

 everywhere lie altogether too far to the north, passing from the 

 Newfoundland Bank (44° N.) by Jan Mayen (71° N.) as far as 

 up to the northwest of Spitzbergen (80° N.). Neither is the limit 

 of the drift-ice during the winter quite satisfactory as a limit for 

 the region of the arctic fauna. 



In the deep sea, the limit for „the cold area" (with a bottom 

 temperature under 0° C.) forms a fairly good limit for the arctic 

 fauna. But in shallow water, where the hydrographical conditions are 

 more complicated, that is, varying according to the different seasons 

 of the year etc, it is more difficult to fix a limit. Extensive in- 

 vestigations must yet be made, before we can fully give the hydro- 

 graphical causes, which produce the changes in the composition of 

 the fauna. For this reason, one must as yet, when fixing a limit, 

 begin with a very careful study of the fauna itself, and it was in 

 this way G. 0. Såes, as along ago as 1879, succeeded in fixing 

 the limit of the Arctic coast fauna in the Atlantic. 



When one wishes to decide whether the sea fauna of a certain 

 place is principally arctic or principally boreal, one should compare 

 it both with the fauna from the high arctic regions (e. g. the Sea 

 of Kara, northern Spitzbergen, northern Greenland) and with the 

 fauna from districts w^hich are undisputably boreal (e. g. Great 

 Britain, southern Norway). If the fauna on both coasts of the 

 North Atlantic be examined according to this plan, it will be found 

 that the limit fixed by G. 0. Såes is in the main correct. 



It will also be found that my foregoing classification of Holo- 

 thurians is not arbitrary. Those which I have called purely arctic 

 are widely spread over the district for the Arctic fauna, whereas they 

 are either not found at all beyonds its borders or only rarely, and 

 then principally in places which in hydrographical conditions are 

 similar to the Arctic Ocean. Those species which I have called 

 purely boreal are also either not found at all, or only in few places 

 in the Arctic regions, and then only where the Gulf Stream has 

 increased the temperature of the water. 



With regard to the four species mentioned above (page 22), 

 which are widely distributed both within the Arctic and the Boreal 

 faunistic regions, I have not felt competent to decide, whether they 



