MARINE ALGAL VEGETATION 47 



It happens rather frequently that the winds are stormy and, 

 as an example, the annual percentage of storms for Stykkisholmur 

 (from */9 1845 to 31 /2 1892) may be given: N. 32, NE. 61, E. 13, 

 SE. 17, S. 44, SW. 31, W. 26, NW. 11. 



The frequency of "calm" is 10% at Grimsey and in Berufjordur 

 and the frequency of "wind" is therefore 90 % in both places; at 

 Stykkisholmur the frequency of "calm" is 12 % and that of "wind" 

 88%; in the Vestmannaeyjar the frequency of "calm" is 22% and 

 that of "wind" 78 %. 



At Thorshavn, in the Faeroes, the annual "calm" is 11 % and 

 the frequency of "wind" 89 %, somewhat the same, therefore, as 

 at Stykkisholmur and greater than in the Vestmannaeyjar. 



4. LIGHT. 



The influence of light on the distribution of the algal associa- 

 tions and on their appearance is, as is well-known, exceedingly 

 great. Without doubt most investigators assume that the main di- 

 vision of algal vegetation into a green, a brown and a red zone is 

 due to the quality of the light, but one cannot on that account 

 consider the intensity of the light to be of no importance. To what 

 extent the shades of colour in the red algae are to be regarded as 

 an adaptation to the intensity of the light or to the quality of the 

 light, I find rather difficult to decide. 



I agree with Berth old and Oltmanns in thinking that the 

 Floridece may be characterized as shade-plants in the same sense 

 that we speak of shade-vegetation in lava-clefts and in other places 

 where there is a faint light. By shade-plants I understand plants 

 which prefer feebly illuminated spots, and do not, as a rule, thrive 

 in the full light of day. In the tidal region (Part VI) the littoral 

 Floridece evidently prefer crevices and grottoes, i. e. feebly illuminated 

 places, and thus prove themselves to be shade-plants. 



I shall not enter more fully into the question of light, as I 

 have made no experiments in that connection and, moreover, the 

 subject requires to be reinvestigated (Oltmanns, 54). 



