1920.] Birds of the Canary Islands. 535 



When we turn to the island of Teiierife we find tlie zones 

 of vegetation much more clearly defined — moreover, more 

 attention has been paid to the botany o£ Tenerife than to 

 any other member of tlse group, and we have, therefore, 

 ample material upon which to base our deductions. 



Humboldt recos;nised four distinct zones of vegetation ; 

 Dr. Christ })rererred three great zones ; Guppy divided the 

 vegetation into six belts. Lastly, we have the thoroughly 

 up to date and clearly expressed opinions of Dr. J. H. Salter, 

 whose treatise on the Regional Distribution of the Native 

 Flora in Tenerife appeared as part vii. of vol. Ixii. of the 

 Manchester Memoirs (1918). 



Dr. Salter, with Dr. Christ, recognises three main regions 

 — each of which he subdivides into more or less clearly 

 defined zones. With Dr. Salterns conclusions, in so far 

 as I have studied the regional distribution in Tenerife, 

 I entirely concur ; moreover, from personal observation I 

 can endorse his remark that the zones can only be charac- 

 terized in general terms and that they are separated by no 

 hard and fast lines — the range of most species extending 

 higher upon tlie southern side of the central ridge than 

 upon its northern slope. Strict demarcation of the zones 

 is quite impossible; bearing this in mind, let us examine 

 the vegetation belts propounded by Dr. Salter : — 



A. The Coast Ret/Ion and Lower Slopes from sea-level to 



2100 ft., comprising (on the southern coast*) : — 



1. Foresliore. 



2. Desert (stony, rocky, or black sand). 



3. Orchards, plantations, vineyards (banana, tomato, 



orange, vine), extending to about 1300 ft. 



4. Cultivated lands (wheat, potatoes, lupins, broad beans), 



extending to about 20G0ft. 



B. The Cloud Region, 2300-6500 ft. 



1. The Monte Verde, from 2400-1000 ft., characterized 

 by woods and thickets of evergreens (far fi'om being 

 a continuous belt). 



* Upon the northeru slopes the desert strip is much reduced or 

 altogether absent. 



SER. XI. VOL. II. 2 O 



