THE CLASSIFICATION OF PLANTS 221 
difficulty is increased by the ease with which very closely 
allied forms hybridize with one another—e.g., thenumerous 
forms of willows, roses, brambles, and hawkweeds. 
One of our native geraniums is the bloody crane’s-bill 
(Geranium sanguineum). It has a diffusely - branched 
stem and large purple flowers. G. prostratum, found on 
the sands of Walney Island, Lancashire, has a dwart 
tufted stem, with small flesh-coloured flowers. Is the 
latter plant a mere variety of G. sanguineum, or is it a 
separate race? We cannot say; we must know how 
its seeds behave in another environment before we can 
decide. In the west of Ireland there grow together two 
saxifrages—Sazxifraga uwmbrosa and S. Geum. According 
to Babington, these are distinct species ; according to 
Clement Reid, they are divergent forms of the same 
plant, for he saw growing among them a nearly complete 
series of intermediate forms. 
Hybrids.—H ybrids are neither species nor true varieties. 
The hybrid being produced from a seed which is formed 
by the co-operation of two parents, with certain distinct 
characters, shares the characters of both, either latent or 
expressed, or in an intermediate form. But the seeds of 
hybrids do not breed true. Some of the descendants will 
be hybrids like the immediate parents ; others may share 
the hybrid characters in a different way ; while some will 
always revert back to the pure form of the ancestral 
types. 
The unit of classification must, then, be the distinct 
race, which breeds true to seed. According to what we 
have said, this, in general, will be the species ; but where 
the species is subdivided into varieties which breed true 
to seed, then these constant varieties will become the 
units. De Vries gave the name “ elementary species ”’ to 
varieties which breed true to seed. Inconstant variations 
have no claim to varietal rank ; they are only of biological 
interest, indicating the range of variation within the same 
species. The characters of the true variety are permanent 
and inherited. In horticulture new varieties are being 
constantly produced. In one sense these are rarely new. 
In flowers, most of them are hybrids. In this case man 
assists Nature by bringing into breeding contact two 
forms which are seldom or never found together. In 
crops they are generally the isolated forms of already 
