92 H. G. SMITH. 
between the salient points, both botanically and chemically, 
appeared to point to important functional characteristics, 
so that it has been possible to suggest, with perhaps more 
than a degree of probability, the line of descent followed 
during the evolution of this most extensive Australian 
genus. 
The comparative constancy of chemical constituents in 
the products of individual species—essential oil, astringent. 
exudation, tannin, etc., suggests the idea that each species 
acts as if it were a chemical factory, manufacturing par- 
ticular chemical constituents, under natural conditions, 
according to a specific formula. What these conditions of 
formation are it should be our endeavour to determine. It. 
does not matter bow extensive the distribution of the 
species, the chemical constituents are still in agreement, 
and this has been found to be the case with members of the 
same species growing both in Tasmania and on the main- 
land, districts hundreds of miles apart. These agreeing 
results seem tosuggest that the establishment of the species. 
itself is largely governed by the chemical requirements of 
the tree, particularly in its earlier stages. If the con- 
ditions of growth still remain in conformity with original 
factors the species continue, generation after generation 
to repeat, in every detail, both the botanical and chemical 
features characteristic of the tree. This has been well 
demonstrated with several Hucalyptus species common to 
both Australia and Tasmania, which were evidently quite 
definitely fixed long before Tasmania was separated from 
the mainland, a period sufficiently long for some species. 
peculiar to the island to have definitely established them- 
selves. Available evidence also indicates that a few 
species have found conditions of separation not so congenial, | 
the struggle to overcome certain adverse conditions having 
brought about changes not difficult to trace. Here we 
