INTRODUCTORY ESSAY. vii 
not so absolutely as the former), but they, on the contrary, consist of comparatively exceedingly well- 
marked genera and species. Melanthacea and Scrophularinea , on the other hand, are not limitable 
as Orders, and contain very many differently constructed groups j but their genera, and to a great 
extent their species also, are well-marked and limitable. The circumstance of a group being cither 
isolated or having complex relations, is hence no indication of its members having the Bime characters. 
Again, as with species, so with genera and orders, we find that upon the whole those are the 
best limited which consist of plants of complex floral structure : the Orders of Dicotyledons are 
better limited than those of Monocotyledons, and the genera of Dichlamyderc than those of Achla- 
mydeae.* 
Now my object in dwelling on this parallelism between the characteristics of individuals in 
relation to species, of species in relation to genera, and of genera in relation to Orders, is because I 
consider (Introd. Essay to Fl. N. Z.) that it is to the extinction of species and genera that we arc 
indebted for our means of resolving plants into limitable genera and orders. This view is now, I 
believe, generally admitted, even by those who still regard species as the immutable units of the 
Vegetable Creation ; and it therefore now remains to be seen how far we are warranted in extending 
it to the limitation of species by the elimination of their varieties through natural causes.f 
6. The evidence of variability thus deduced from a rapid general survey of the prominent facts 
elicited from a study of the principles of classification, are to a certain extent tested by the behaviour 
of plants under cultivation, which operates either by hastening the processes of Nature (in rapidly 
inducing variation), or by effecting a prolepsis or anticipation of those processes (in producing sports 
i. e. better marked varieties, without graduated stages), or by placing the plant in conditions to which 
it would never have been exposed in the ordinary course of natural events, and which eventually either 
kill it or give origin to a series of varieties which might otherwise have never existed.} 
* There are too many exceptions to this to admit of our concluding at once that it is attributable to any 
simple and uniform law of variation ; but it may be explained by assuming that the degree or amount of variation is 
differently manifested at different epochs in the history of the group. Thus, if a genus is numerically increasing, and 
consequently running into varieties, it will present a group of species with complex relations inter se ; if, on the con 
trary, it is numerically decreasing, such decrease must lead to the extinction of some varieties, and hence result in the 
better limitation of the remainder. The application of this assumption to the fact of the best limited groups being 
most prevalent among the higher classes (i. e. among those most complicated in their organization), would at first 
sight appear an argument against progression, were it not for the consideration that the higher tribes of plants have 
in another respect proved themselves superior, in that they have not only far surpassed the lower in number of genera 
and species, but in individuals, and also in bulk and stature. And lastly, as all the highest orders of plants contain 
numerous species and often genera of as simple organization as any of the lower orders are, it follows that that phy- 
sical superiority which i> manifested in greater extent of variation, in better securing a succession of race, in more 
rapid multiplication of individuals, and even in increase of bulk, is in some senses of a higher order than that repre- 
sented by mere c ion of organ. 
f It follows as a corollary to the proposition (That species, etc., are naturally rendered limitable by the destruc- 
tion of varieties), that there must be some intimate relation between the rate of increase and the duration of wn ^ 
(or other groups of species) on the one hand, and the limitability of their species on the other. T 
consists of a midtitude of inimitable forms, we may argue with much plausibilitv that it is on th 
no intermediates have as yet been destroyed, and that the birth of individuals and the production of new forms Ts pro- 
ceeding at a greater proportional rate than in an equally large genus of which the species are limitable. 
% My friend Mr. Wallace treats of animals under domestication, not only as if they were in very different 
physical conditions from those in a state of nature, inasmuch as every sense and faculty is continually fully exer- 
cised and strengthened by wild animals, whilst certain of these lie dormant in the domesticated, but as 'if they were 
