1899] ON THE STUD Y OF PLANT ASSOCIATIONS i 1 1 



cliraats" (1807, p. 14). In his classical essay (1807), where he 

 treats of the vegetation of Chimborazo, he shows how the differences in 

 the plant-life at different elevations are dependent upon corresponding- 

 differences in temperature, humidity, barometric pressure, etc. 



A. Pyr. de Candolle (1820) recognised the importance of noting 

 all these facts of association. In his scheme for a flora of the valley 

 of the Lake of Geneva he indicates the following amongst the facts 

 which should be ascertained : — 



"5°, La station, c'est-a-dire, la nature speciale des localites 

 relativement a la nature du sol, a son inclinaison, a son arrosement ; 

 la nature et la temperature des eaux ; l'eclairement ou l'obscurite, etc. 



6°, Ses varietes locales, c'est-a-dire, les changemens de formes 

 observees dans les limites adoptees et leurs rapports s'il en est avec 

 les localites. 



8°, L' indication du degre de sa rarete" ou de son abondance, le role 

 qu'elle joue comme support, appui, ou ennemi des autres vegetaux, la 

 designation de la plante comme eparse ou sociale ; enfin, l'indication 

 des especes avec lesquelles elle a coutume a vivre." 



This attempt to include all the observations on the distribution of 

 the species in a flora has, so far as we know, not been carried out, 

 since it was soon recognised, that many of the facts could from their 

 nature be more conveniently recorded and studied apart from the 

 " flora." 



Grisebach (1838) more fully organised the methods for describing 

 the physiognomy of the landscape, and for this purpose introduced the 

 new term " pnanzengeographische Formation." A plant " formation," 

 according to him, is a group of plants, such as a wood or a meadow, 

 which forms a distinct and complete feature of the landscape. It may 

 be characterised by one social species (a pine wood), by a group of 

 similar social species (a pasture), or by an aggregate of species of 

 manifold organisation, but having one general and conspicuous 

 characteristic (an alpine meadow where all the plants are perennial 

 herbs). 



Thurmann (1849, p. 22) distinguishes between the flora and the 

 vegetation of a country ; after discussing the scope of each method of 

 study he sums up as follows : " La Flore s'entend surtout du nombre 

 des formes vegetales distinctes qu'on y observe, la Vegetation de leurs 

 proportions et de leur association." He and other contemporaneous 

 students of the relations between the plants and the subjacent soil 

 (such as Unger, von Mohl, Sendtner, etc.) were well aware of the 

 fact that slight differences in the soil conditions give corresponding 

 differences in the frequency of the species growing there, and they 

 employed a series of terms from social to isolated to indicate this. 



Alph. de Candolle (1855) attempted to answer the question 

 why some species are social and others not. This, according to him, 

 is due to two causes : — 



