Influence of Habitat on Plant Habit. 133 



of the Sahara and the alluvium of the Nile ; Lindmann has also 

 pointed out that it is very common at Cadiz (1), Meigen found 

 this type prevalent in the (Hiilian Desert (2), &c. 



But when we find a plant becoming a rosette in a habitat of 

 great exposure only and not taking on this habit when it is not 

 subjected to great transpiration, then it is better evidence of the 

 effect of habitat. 



I found that Diplotaxis 3, 5, 6, and Thlaspi 3, 12, 13, 14, 15, 

 16, 19, all have internodes in sheltered places but become rosettes 

 in exposed habitats. Wiesner was able to form internodes in the 

 Shepherd's Purse, Capsella by growing the plant in moist air, but 

 this was not possible with the Dandelion (3, 4). This is what we 

 should expect if the habit is at once a result of habitat, but may 

 become fixed by heredity if long- enough continued. 



In order to see the effect of rocks more clearly I re-examined 

 the orders named to find how many other species weie noted as 

 growing on rocks. There were 14 so described. (See table 2.) 



Table II. 

 OTHER PLANTS GROWING ON " ROCKS." 



Farsetia, 1, 2, 3 

 Sinapis, 4 ... 

 Fumaria, 27 

 Iberis, 18 ... 

 Euromodendrou 

 Matthiola, 7 



Very woolly 

 More hairy 

 Fleshy leaves 

 Fleshy leaves 

 Ericoid shrub 

 Very woody 



Turritis | (Not specially 



Arabis, 1, 2, 3, 4, 5j (Xerophytes 



The Turritis and Arabis spp. are probably not xerophytes, but 

 all the others show the characteristic modifications of a dry 

 climate v. Tschirch (5), Volkens (G), Henslow (7), &c. 



I also tabulated the number of woolly or very hairy plants, 

 and found that 21 out of the 230 species could be fairly included. 



As shown in Table 4, li grow in such dry countries as Greece, 

 Syria, Spain, and Algiers, 4 are true desert forms, 6 prefer rocks 

 or stony places, one is a seaside form, and the last is an Alpine 

 species of which I have no further details. 



