NOTICES OF BOOKS. 57 



\ 



m 



amongst a multitude of valuable data, appertaining to certain mountain- 

 chains, or to certain groups of plants. Amongst the principal causes 

 of the difficulty of coming to any satisfactory results, is said to be the 

 limitation of the area over which one observes ; a few feet of eleva- 

 tion being analogous to many miles of latitude, all the causes that 

 interfere with absolute determinations of the polar and equatorial limits 

 of plants at the level of the sea, are proportionately increased when 

 we attempt to discover their upper and lower limits upon a mountain ; 

 to this is added, the greater facilities for accidental transport of seed- 

 lings, beyond the normal limiting line of the species, the position and 

 persistence of snow and of fogs, the varying nature of the soil within 

 a small area, and perhaps the density of the atmosphere ; also the 

 want of data as to the climate of alpine regions ; thus, in Switzerland 

 (the best-known alpine region of the world), we are told that there are 

 two points alone (Saint-Gothard and Saint-Bernard), at which the 

 monthly mean temperatures of the air at any elevation more than 600 

 metres above the level of the sea have been determined. 



The causes which limit the upward and downward spread of species 

 upon mountains, are treated of in detail, under Humidity; tempera- 

 ture of the air, of the water^ and of the upper strata of the soil, 

 especially as affected by solar and terrestrial radiation ; exposure ; 

 rarity of the air ; mineral character of the soil ; geological character 

 of the mountains ; isolation of mountains or their approximation in 

 groups ; duration of Snow ; and finally a section is given to the rela- 

 tive importance of each of these causes. Critical details of the distri- 

 bution of individual species follow. In concluding the subject of the 

 upward limitation of species, M. de CandoUe dwells at length on the 

 necessity of first studying the distribution of the same species on plains, 

 before any attempt can be made to explain their limitation on mountains. 

 Of the downward limits of species on mountains little is said ; data 

 are wanting, and the difficulty of observing is greatly increased ] very 

 few cases are adduced, and the general laws are supposed to be analo- 

 gous to those which determine their equatorial limit. A curious article 

 follows, upon the very different relations to one another in respect to 

 the elevations they inhabit, that the same species affect in different 

 mountain-systems, even when these are in the same latitude. 



From the distribution of native species M. de Candolle passes to the 

 consideration of cultivated, and here the natural cases become compli- 



TOL, VIII. 



I 



