OLIVEE ON THE ATLAKTIS HYPOTHESIS. 163 



there are, having corresponding representative species, in 

 Europe, over 48 per cent., identical species 27 per cent. 



"West North America, about 37 „ „ 20 „ 



East „ „ „ 61 „ „ 23 „ 



Moreover of 353 extra- European genera in the Northern States 130 

 are common to East Temperate Asia, while but 87 occur in AVest 

 North America. About 90 genera are represented in North America 

 and Japan which are absent from Europe, and of these 65 do not 

 occur in extra-tropical "Western North America,* where farther, no 

 order is represented wanting in the Northern States of East America, 

 excepting Grarryaceae and Hydroleaceae, and these both occur in the 

 Southern States. 



The Atlantic Islands Flora.f — The American element in the 

 Flora of the Atlantic Islands is very subordinate, while Mediterranean, 

 with a proportion of peculiar or Macaronesian types, greatly pre- 

 dominate, the former very remarkably. The African element is, as 

 is well known, singularly deficient. Prof. Heer points out as a trace 

 of the connection which he conceives to have existed formerly 

 between these Islands and the New "World, the American genera 

 Clethra, Bystropogon, Cedronella and Oreodaphne, species of which 

 occur in the Madeiras and Canaries. With regard to these, however, 

 it may be observed that Clethra is not exclusively a New "World type. 

 One species is Japanese and one or two grow in the Philippines and 

 Borneo. The present focus of the genus appears to be in South 

 America. But one species, C arborea, grows in the Atlantic Islands, 

 in some of which it is very abundant. As to Bystropogon, Messrs. 

 "Webb and Berthelot limit the genus to Macaronesian species, 

 excluding the section Mintliostacliys, under which Mr. Bentham 

 groups the American forms in the Prodromus. At best, this genus, 

 as Ml-. Bentham informs me, scarcely differs in technical characters 

 from Mentha, though the habit of the island species is very different. 

 Of Cedronella, another Labiate plant, but one species grows in the 

 Islands, G. canariensis. Hasskarl describes a species from Japan. 

 Oreodaphne prevails in the West Indian Islands and South America; 

 it is unrepresented in the States ; species occur, also, in South Africa, 

 Madagascar and Mauritius. Recently Mr. Mann, botanical collector 

 to the Royal Gardens at Kew, has sent to Sir W. J. Hooker two or 

 three species of Laurel from the mainland of West Tropical Africa, 

 wdiich, although not yet determined, form a connecting link, relieving 

 the isolation of the Atlantic species, both of Oreodaphne and Fersea. 

 I have examined the Elora of Webb and Berthelot and can find 

 scarcely any evidence to add to that noted by Prof. Heer. The 

 genus Messerschnidtia, limited by these authors to Canarian and 



* Gray, 1. c. 441. 

 t For some excellent obsei-vations upon the general relations of the Madeira and 

 Canary Flora, see a paper by Sir C. J. F. Bunbury, Linn. Proc. Bot. i. 1. 



