694 



SKOTTSBERO 



Other cases it is real, and we have to find out the reason, perhaps the direction 

 of the current, or a difference in water temperature. 



If Plate's opinion of the oceanic nature of our islands is correct, they must 

 have received their entire fauna and flora from more or less distant coasts. 

 Central Chile is certainly the nearest, but because PlyVTE found a greater dif- 

 ference between Juan Fernandez and the opposite coast, and more resemblance 

 to the littoral fauna of Peru, he assumed the insular fauna to have come from 

 Peru and North Chile. Judging from the present hydrographical conditions it 

 seems unlikely that the region Peru-Central Chile can have been the main source; 

 migration across or against the Peru Current offers considerable difficulties. The 

 passage from Callao to Juan P'ernandez runs against the current, at least most 

 of the way [vide Plate's observations on local changes of direction near the 

 islands), and going west from Valparaiso we find a very marked rise of temperature. 

 It was shown above that the American element in the marine flora is small, and 

 that the microthermic PhacopJiyceae, which constitute a dominant feature in the 

 vegetation of Central Chile, do not reappear at the islands. The natural path 

 of migration is the West Wind Drift and the Peruvian Current (in the wide 

 sense of GUNTllEk), and this route was probably followed by a majority of Sub- 

 cosmopolitan and Subtropical algae (gr. i and 2), and certainly suggests itself 

 for the Pacific and Subantarctic species (gr. 5 and 7). Many of these species 

 reached both islands and mainland, other species made halt at the islands (comp. 

 for instance GlossospJwra with Splachnidiiini). Travelling by the West Wind 

 Drift, species at home in Australian waters followed the Peru Oceanic Current 

 from about 40° S or were in any case unable to establish themselves in the 

 region of upwelling cold water, which acted as a barrier, and thence were 

 excluded from the mainland or, in some instances, found suitable living condi- 

 tions in Peru and North Chile, where inshore temperatures are about the same 

 as at Juan Fernandez and Desventuradas. 



It is also possible that the species of the American group (3), which are 

 common to Juan P"ernandez and Peru — Chile, but not found elsewhere (with the 

 exception of Pterosiplioiia dendroidea which reappears in California) share the 

 history of the West Wind Drift flora. The species belonging to group 4, 

 California + Juan Fernandez, are small microepiphytes not readily lending them- 

 selves to a discussion on distribution, except Codiuni nvilaterale and PlocaniiuDi 

 pacificuiii. The situation is much the same with regard to algae common to 

 the Pacific coast of both Americas but jumping the Tropics and not found at 

 Juan Fernandez (the Macrocystis type of distribution). Macrocysiis in Subantarctic- 

 circumpolar and its North Pacific area is, I suppose, regarded as an expansion 

 of its main range, but it should be kept in mind that its closest relatives, 

 Nereocystis and Pelagopliycus, are North Pacific. The current charts do not 

 suggest a direct route of migration from Peru to California or vice versa. 



The Endemic element presents the usual difficulties; it is there and nowhere 

 else. The usual way of looking at these things is to say that endemism is the 

 result of long isolation under altered conditions which are, directly or indirectly, 

 responsible for the evolution of new species and genera, a procedure generally 



