Hybridism in the New Zealand Flora 121 
varieties, by way of species with more and more varieties, up to 
aggregates so complex that it becomes impossible to recognise the 
true-breeding varieties, so great are the number of intergrading 
forms. In other cases the varieties may be clear-cut 1 , so that inter¬ 
mediates when they do occur can be recognised as such, or varieties 
as taxonomically defined may be merely groups of related forms 
connected by intermediates to other so-called “ varieties ” of a similar 
character. Thus in the simpler cases where the varieties are really 
such ( i.e . true-breeding entities), hybrids between them can be re¬ 
cognised, but, where all is confusion, their presence can only be 
assumed by analogy. An account of a few cases may make the 
foregoing remarks clearer. 
Mesembryanthemum australe is an easily : recognised species which 
is confined to rocky stations on the coast. In certain localities two 
well-marked varieties occur, one with the usual pink flowers and 
reddish leaves, and the other with white flowers and pale-green 
leaves. Where there are numerous individuals, as on new stony 
ground, resulting from disturbance of a stony beach, there may be 
abundance of both varieties together with a good many forms inter¬ 
mediate in colour of flower and leaf. So, too, in the case of Rumex 
flexuosus there is a variety—the most common in fact—with dark- 
brown leaves. But in certain localities under the same ecological 
conditions along with the above there is a green-leaved form, and, 
when this happens there are also present plants intermediate in 
colour, which must be considered hybrids. Similarly, in the case of 
Senecio bellidioides, there are plants with tomentum and without 
such on the undersurface of the leaf and between these extremes are 
plants showing every degree in the amount of tomentum. 
On the other hand there are many cases where no hybrids have 
been noted between varieties of similar character to the above. 
Coprosma Petriei is made up of two varieties, the one with large 
port-wine coloured drupes and the other with drupes faintly stained 
with blue, but nowhere have intermediates been seen. Neither have 
intermediates been noted between the var. of Cyathodes acerosa with 
1 Epilobium pedunculare is treated in the Manual, p. 180, as a var. of 
E. nummularyfolium of which three other vars. are recognised. But, in point 
of fact, as I have pointed out (Trans. N.Z.Inst. 50 , pp. 171-73) the latter 
is probably invariable, while I have shown that aggregate E. pedunculare is 
made up of a number of invariable varieties of wide range to three of which 
I have given varietal names. The var. nerterioides also is abundantly distinct 
from E. nummularifolium, but it can be taxonomically united with var. 
minimum, which also is a well-marked, certainly true-breeding group, dis¬ 
tinguished by its very short capsule and peduncle. 
