26 THE INTRODUCTION AND [pt. i 



example, which has been much quoted in this connection, all 

 but 11 of the 387 naturalised species (115) are either weeds about 

 houses, due to cultivation, or weeds of open ground, which was 

 all but unknown in the old days of forest. Seven of these 11 are 

 only a clump or two of planted trees, and there are really only 

 tAvo cases of natural or nearly natural spread, and then only to 

 a distance of a few hundred yards, downstream, in a very steep 

 valley. 



The only instances of rapid spread in Ceylon, on land already 

 occupied by a growth of plants suited to the conditions, have 

 been in the case of a few such weeds as Tithonia and Mikania^ 

 M'hich have spread rapidly over the open ground already occu- 

 pied by weeds previously introduced. One is inclined to think 

 that this is due to the fact that such areas have not as yet 

 elaborated the best plant societies suited to their conditions, 

 and that room is still left for newcomers. On the other hand, 

 there is no evidence for rapid spread in forest, where the adjust- 

 ment of species to environment has probably been carried to 

 great lengths. 



In a few cases, new species have spread rapidly over ground 

 already occupied by herbaceous plants, like Elodea in the waters 

 of western Europe, or Spartina in the low coast lands on the 

 south of England. Here, again, one may suppose that there has 

 still been room for newcomers, especially in the case of the 

 Spartina, which is largely found on land that was submerged 

 not so very long ago. 



The enormous majority of cases of rapid spread of introduced 

 weeds are due to cutting of forest, or other serious alterations 

 of conditions ; in North America and Argentina often to cultiva- 

 tion of the soil (even if only once), leaving conditions different 

 from what they were. In St Helena, which has been much used 

 as an argument for nat\n-al selection, man introduced goats, 

 which are most destructive to vegetation, with the result that 

 there is left only a flora practically "goat-proof." Even the 

 largest trees are not safe, for the goats may destroy the smaller 

 trees, and expose them to the action of the sun and wind. [Cf. 

 also the effects of the exclusion of rabbits from a heath (36, 

 1917, p. 1).] 



Cockayne has devoted much attention to New Zealand, an 

 island in which over 550 introductions have become more or 

 less naturalised, and which has often been quoted as evidence 

 for the great superiority of introductions from the crowded flora 



