210 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



the plants that are actually flowering at the same time, it soon becomes 

 apparent that whilst it is easy to collect representatives of different 

 genera, the different species of a genus which are flowering together 

 are in the great majority of instances not very closely related. More- 

 over, where this does not hold good, the related species are commonly 

 restricted to different types of soil or habitat, and so do not really 

 come into strenuous competition at all. One has only to think of the 

 succession, in time of flowering, of allied species of Umbelliferae, Com- 

 positae, Caryophyllaceae, Saxifragaceae, &c., and at the same time to 

 recall how definitely the contemporaneously flowering allied species 

 respectively of Heath, Rhododendron, Primula, Gentian, &c., are 

 segregated by soil preference or habitat. This point is especially 

 brought out by a consideration of the calcareous and siliceous species of 

 many nearly related alpines, e.g. Achillea moschata and A. atrata, which 

 in one valley are commonly severally confined to the non-calcareou 

 and calcareous soils, though in the absence of one of them the other 

 seems able to occupy both sorts of soil. It is of course a matter of 

 common horticultural experience that a wild plant, when introduced 

 into cultivation, will often flourish in soil and under other conditions 

 very different from those to which it is restricted in its natural state. 



I take it these facts, which anyone can easily observe for himself, 

 really mean that the change involved in the production of the 

 new variety, mutant, or whatever we may call it, is one which 

 affects its chemical processes. Just how this change will take 

 effect cannot be predicted in any particular instance, and it is 

 evident that it might operate in a variety of different ways, each, 

 ceteris paribus, representing some advantage to the new form. Any 

 change, however, that lessens the active and actual competition must 

 tell in favour of the individuals concerned, whether this depends on a 

 slowing down or hurrying up of the nutritive processes, so as to shift 

 the time at which the different races are competing for the same food, 

 or whether it depends on a more fundamental change, such as would 

 result from an alteration in the proportion of the food constituents 

 actually required. I have little doubt that it is in changes of this 

 sort that the origin of the so-called " replacing species " of alpine and 

 other plants is to be sought, but each case needs to be studied by itself 

 in order to unravel its tangled history. It is evident that a segregation, 

 whether in time of flowering, for example, or in locality, provides the 

 best opportunities for the persistence of new races and species as they 

 come into existence. This will be equally true whether we recognize or 

 whether we refuse to admit the swamping influence of intercrossing. 



I have already pointed out that the vegetation of most places is 

 a mixed one. The exceptions to this depend upon special conditions of 

 soil, or other physical circumstances, to which only a hmited number 

 of species are fully adapted. These plants are consequently able to 

 exclude others from gaining a foothold, even if the latter, supposing 

 they were introduced, could manage to persist. 



The chemical processes of some of the constituent species of mixed 



