PT.i.cH. iij THE DISPERSAL OF PLANTS ii 



necessary that a species, genus, or tribe should arise upon one 

 spot only, or even in one region only. They consider that the 

 same thing may arise independently in different places, very 

 rarely indeed the species, more often the genus, tribe, or family, 

 either from the same species by the same road (as would probably 

 be the case with the origin of a species in this wav), or from 

 different species, which all made the necessary changes to place 

 them in the same genus or tribe (cf. 116, p. 446). This sup- 

 position would unquestionably get rid of some of the difficulties 

 of explaining many cases of discontinuous distribution, where 

 the same species, genus or tribe appears in widely separated 

 regions. 



Whatever view has been held as to origin, however, it seems 

 to have been generally taken for granted that except in so far 

 as they have been prevented by actual barriers, such as seas, 

 ranges of mountains, sudden changes of climate from one dis- 

 trict to the next, and the like, species have spread over the 

 whole area to which they are suited, i.e. where they can grow 

 and reproduce in spite of any adverse conditions to which they 

 may be subject. In other words, it seems to have been assumed 

 that the distribution about the world of the species now existing 

 therein is largely a closed chapter, except in so far as man by 

 his various activities may alter it. Why this idea of finality 

 should have sjirung up is not quite so easy to decide, unless it 

 has been that people take for granted that in nature dispersal 

 of plants is rapid^, and it is one of the objects of the present work 

 to show that we are still dealing here with open questions. 



It is clear, however, that the large areas now occupied by 

 many sj^ecies must almost always, if not always, be due to 

 spreading from others originally much smaller, and a careful 

 study of the ways in which this dispersal may be effected must 

 form a necessary preliminary to the study of geographical dis- 

 tribution in general. It is of course obvious that, as a rule, a 

 plant once established will not move again, but its seeds, or 

 detached portions of itself (or sometimes, as in the case of runners, 

 connected portions), maj'^ in A^arious ways be carried to a distance 



^ People see a dandelion scattering seed over a large area, or notic-e tin- 

 rapid spread of a new weed in the garden, and are apt to reason that this 

 sort of thing is always going on with all species, while at the same time 

 they forget that most, if not almost all, seeds dropped upon ground already 

 fully occupied by plants, fail to grow, even if they germinate. One may see 

 the same clump of traveller's-joy, for example, occupy the same place 

 without spreading, for a whole lifetime. 



