3o6 Carl S1{0ttsberg 



varieties or small species. Perhaps Coriaria will undergo the 

 same metamorphosis, although its distribution by birds seems 

 possible; the distances between the stations of C. sarmentosa 

 are not very great, and part of the road may have been over land 

 if there is any truth in the speculations of paleogeography. 



Of the remaining genera, all of which have dry fruits, Oreo- 

 bolus possesses no special facility for dispersal unless it be the 

 small size of its fruits, which, I suppose, are carried about by birds 

 frequenting the moist heaths where Oreobolus is at home. The 

 range of O. obtusangulus in Andine America is very large. We 

 have good reason to believe that, with the aid of birds, it reached 

 the summit of Masaf uera, together with a few more Magellanic 

 species, under geographical conditions very similar to those of 

 the present. However, in Hawaii we find a distinct endemic 

 species living in utter geographical isolation. 



The species of Weinmannia have, as a rule, a restricted range 

 with pronounced endemism, but we have seen that the same 

 species, in at least two instances, is credited to islands separated 

 by wide expanses of ocean. The seeds are provided with long 

 hairs and may be carried through the air for short distances; 

 doubtless this is the normal way in which these plants are dis- 

 seminated, but it is hard to tell what exceptional conditions en- 

 abled them to cover the distances between Rapa, Tahiti, and the 

 Marquesas. 



Acaena, Uncinia, and Lagenophora are particularly well 

 adapted for epizoic dispersal. The achenes of Acaena have 

 barbed bristles, and nobody denies that the introduced A. ar- 

 gentea has spread all over Masatierra because it adheres to man's 

 clothes and to the furs of dogs, horses, and cattle. The indigenous 

 species on the summit of Masafuera is not common, and its 

 minute fruits perhaps do not attach themselves so easily to car- 



