300 DESCRIPTIONS AND 



are often seen and therefore referred to. In the beginning of the 

 chapter on Shrubs, pages 455 to 459, are some remarks on the con- 

 siderations which influence a choice of shrubs (some of which apply 

 equally to trees), to which the reader's attention is invited. 



ORDER OF ARRANGEMENT. It is extremely difficult to follow 

 any system for the classification of trees and shrubs that will 

 greatly facilitate the reader in finding readily what he wishes to 

 read of, or that will save him constant references to an index. 

 Botanical classifications, when thoroughly made, require quite too 

 much familiarity with botany to give them any value to the mass 

 of readers who know only the a, b, c's of the science ; yet they 

 must, after all, be the ground-work of the most convenient arrange- 

 ment for descriptions. Though the same botanical family often 

 the same species has plants of every variety of size, from ground- 

 lings to lofty trees, which differ from each other in their larger 

 characteristics as much as from some members of other families 

 with which they have little botanical connection, yet, in general, 

 it will be found that grouping by botanical relationship brings together 

 those which resemble each other in the greatest number of particulars. 



To classify trees and shrubs by their sizes, would separate 

 family groups, and scatter them promiscuously among each other, 

 while in all respects but size, their similarity of traits make it most 

 easy to describe them by families. Take the oaks, for instance. 

 The different species are numbered by hundreds, all having some 

 marks of consanguinity in their general appearance, but quite 

 diverse in forms and sizes. The immense variety of species of the 

 first differ still more among themselves ; varying in size from lofty 

 trees to pigmy shrubs. If we class them with evergreen trees 

 according to their varying sizes, they would become sadly mixed 

 among the pines, junipers, arbor-vitaes, yews, and a score of newer 

 evergreen families. If classified by forms alone, the same confusion 

 would arise. It is best therefore to keep botanical family groups 

 together. All oaks, for example, large and small, are described 

 consecutively under the head of THE OAK ; and as most of them 

 are trees, they are described under the general head of DECIDUOUS 

 TREES ; though there are varieties which are really shrubs only. 



