120 
L. Cockayne 
(b) Alseuosmia Banksii x linariifolia. 
Here may also be discussed A. Banksii x quercifolia and A . linarii¬ 
folia x quercifolia, since all occur together, notwithstanding A . querci¬ 
folia is of fairly wide range, and its hybrids really come into class 3. 
A. Banksii and A. linariifolia are confined to forest in the North 
Auckland Botanical District; A. quercifolia in the southern part of 
its range is invariable. Carse’s account of the species sufficiently 
proves to what an extent they cross (Trans. N.Z.Inst. 43 , p. 206, 
1911). “There are certain forms of Alseuosmia,” he states, “which a 
mere beginner can place at a glance as typical, but there are so 
many forms intermediate between the various species that hardly 
two botanists in a dozen will agree as to which species predominates 
in the particular specimen” (italics mine). Allan Cunningham de¬ 
scribed eight species, and it may well be that there are more than 
is usually allowed. According to the Manual, p. 240, “The species 
are exceedingly variable and difficult of discrimination.” 
Hybrids within the species 
Were knowledge sufficiently advanced, so that the true-breeding 
units (jordanons), which, taken together, form the ground-work of 
an aggregate species, could be recognised each by its special char¬ 
acters, and were a name given to each unit, then a true knowledge of 
these polymorphic groups (the aggregate species) might be possible, 
and the intervarietal hybrids be recognised. But, as in all floras, 
though especially in those of imperfectly botanised countries, such 
as New Zealand, variety after variety, for which there is no name, 
has its place in nature, while so many varieties as defined are them¬ 
selves aggregates, it is frequently impossible to do more than hazard 
a guess as to the status of any particular form. Such an one may 
really be a true-breeding entity, or it may be a hybrid or even an 
unfixed environmental form, but stable enough in that particular 
habitat. 
There is, in fact, no definite rule according to which varieties 
can be taxonomically constituted any more than there is any rule 
governing the making of species. Further, from time to time taxo¬ 
nomic fashion changes. At present, the tendency is to “accord 
varieties specific rank,” which action, should there be hybrids between 
them, would also change the status of the latter. 
When a study is made of aggregate species in the field, then, so 
far as polymorphy goes, they fall into several classes, and there 
appears a distinct progression from the simple with two well-defined 
