ROSACEA. 435 



lias only been observed in one part of Abyssinia ; the only described 

 Leucosidea is peculiar to the Cape; Bencomia is only found in the 

 Canaries and Madeira; Exochorda in Mongolia; and the two 

 known species of the genus Adenostema in California. The two 

 genera Kageneckia and Margyricarpus, each including two or three 

 species, and Polglejns, containing half a score, are peculiar to 

 the Andine region of South America. The two or three known 

 species of Stglobasitm are Australian. Trichocarga is only found in 

 Borneo and Sumatra. Next come genera with a pretty large number 

 of species, still, however, included in relatively narrow geographical 

 limits. Polglepis comes again in the category, of which genus, as we 

 have seen, half a score species are admitted, all growing in the tem- 

 perate Andine regions of Peru, Bolivia, and Columbia ; also Hirtella} 

 and Licania, including the one two score, the other upwards of fifty 

 species, confined to the warmest regions of America. From these we 

 pass to the larger genera of the family, such as Potentilla, Fragaria, 

 Geum, Riibus, Rosa, Pgrus, Cratagus, Primus, &c., all, it is found, coin- 

 ciding in the remarkable fact that their species are spread all over the 

 surface of the globe, or at least over a very large extent. Thus,- we 

 find Brambles from the north of Europe, Asia, and America to the 

 Cape, New Zealand, and the islands of the Pacific Ocean. The same 

 applies to the Potentils. Eoses, again, are found all over the northern 

 hemisphere ; while Accena is equally spread over the southern hemi- 

 sphere from the Cape to Patagonia, and even extends beyond the 

 Equator to Mexico. True, there is great diflSculty in ascertaining 

 whether some of the vulgar species of Geum, Fragaria, Poteniilla, &c., 

 have not been introduced by man into a large number of countries of 

 which they are not native. This interesting point of geographical 

 botany" has been chiefly discussed with regard to our fruit trees, which 

 belong almost exclusively to the order Posacece. It is now-a-days gene- 

 rally agreed that some are natives of Europe, others of the East, and 

 that the latter, or at least the greater part of them, have only been 

 introduced in comparatively modern times. Our varieties and races of 

 cultivated Cherries are all considered to have sprung ho\x\Prumis avium 

 McENCH and P. Cerasus L , the one spontaneous in Europe, the other 

 in the south of Caucasia, and even in the Crimea, Macedonia, and 



' The only exception in this genus is H. The- ^ A. DC, Geogr. Bot. Bais., 542, 619, 877, 



lira, which comes from Madagascar. 1102. 



PF 



9 



