674 



occur either on a few islands or on a single island only, and that 

 this circumstance can only be due to oecasional and irregular im- 

 migration. If the species immigraled across a land conncction, we 

 might wonder why so large a number, as has been recorded, 

 has reached only the southernmost part of the area, viz. some 27 

 species or Vio (see Ostenfeld, p. 105). The southern islands do not 

 diller so much from the others that the reason should be sought 

 for there. Judging from the present circumstances it seems most 

 natural to assume that these islands are geographically the nearcsl 

 to receive the seeds and piants transported from Scotland by the 

 agency of wind and birds 1 . 



Plant immigration across the sea. 



The result I have arrived at is consequently, that it is almost 

 certain that the flora has not immigrated across land. Hence, it 

 could have immigrated across the sea only, and I may point out 

 that many eminent investigators including Hooker, Darwin, Alph. 

 de Candolle, Wallace and Hemsley are of opinion that »oceanic 

 transport« is the principal and most probable means by which oceanic 

 islands have been stocked with piants. 



That a flora can migrate even across vast tracts of sea is 

 according to my opinion proved by the faet that there does not 

 exist a single oceanic island, which is destitute of plant-life, and 

 many of them have unquestionably never formed parts of a con- 

 linent. In my »Grønlands Vegetation« (1. c. p. 201) I have referred 

 to Jan May en, this volcanic island, situated 450 kilometres from 

 Greenland, 550 from Iceland, and 965 from Spitzbergen, surrounded 

 on all sides by vast ocean depths (1000 — 2000 fathoms), the bottom 

 samples of which are, as far as has been ascertained by Boggild, 

 all deep sea deposits. After the Amdrup Expedition, in 1900, had 

 paid a short visit to the island, the list of its vascular piants was 

 enriched by 33%, so that it now consists of 39 vascular piants, 

 almost all of which have a wide arctic distribution-. Jan Mayen's 



1 That there is. and has been, some sort of connection between leeland and 

 the Færoes, as well as between Iceland and Greenland, is beyond doubt. The 

 faet that certain species (see Ostenfeld, p. 109) such as Alchimilla fåeroénsis are 

 common both in the Færoes and in east Iceland, and that Carc.v cryptocarpa 

 which is common in Iceland also occurs, though rarely, in the Færoes, scem to 

 bear witness of a recent immigration, and the nearest lying parts of the country 

 have been first reached by the immigrants. 



2 See C. Kruuse in Botanisk Tidsskrift, XXIV, p. 297; and Meddelelser om 

 Grønland, XXVII, p. 361. 



