CLASSIFICATION OF THE EUCALYP'TS. 547 
in the same species ; in a few species, however, the leaves of the 
adult plant offer very distinctive characters, such as connate in 
L. gamophylla, hispid in £. setosa. 
Though #. amygdalina and £. regnans may be separable by 
leaf-shape alone, or £. hemiphloia and LZ. Behriana by colour, 
yet I think it will be conceded that such distinctions are not very 
reliable. 
“The venation characterist# as it often is in the lanceolate 
leaves, the specific modifications disappear in a great measure as the 
leaf gets broader. In general it would appear that the horizontal 
leaves have the two surfaces different, and the veins very divergent 
or transverse, and the vertical leaves have the surfaces similar and 
the veins oblique.” —Bentham., 
Again, the leaves of some species are more distinctly or more 
copiously pellucid-dotted than in others ; but as this feature varies 
with the age of the leaf it would be untrustworthy to employ it 
for specific distinction, except perhaps when it offers extreme 
contrasts. 
ESSENTIAL OILS. 
The drug, eucalyptus oil, is commercially obtained from the 
leaves, though it is also contained in the calyx. The oils from 
the various species of Eucalypts have the same general character, 
though there are most important differences between some of 
them, chiefly as to specific gravity and odour. In respect of the 
latter character Z. citriodora and EF. globulus offer great diver- 
gence, readily determinable by bruising the leaves: it is thus in 
a few species only that the odour of the foliage may be utilised 
for diagnostic purposes. 
PETIOLE. 
Messrs. McAlpine and Remfrey, ‘Trans. Roy. Soc. Victoria,” 
vol. 2, 1891, have suggested the employment of the characters 
afforded by transverse sections of the petiole in the classification 
of the Eucalypts. The chief characters employed are those derived 
from the epidermis, hard bast, wood with its vessels, cortical 
cavities and central canal; they have found that each of the 
thirty species analysed has distinctive characters. But the tests 
employed appear to have been too limited ; and before it can be 
admitted that such transverse sections are aids in the determina- 
tion of species they should be applied to individuals of species 
living under diversified conditions. Even otherwise this histo- 
logical system is too cumbersome in its application to be of 
diagnostic value. 
INFLORESCENCE. 
The usual condition is an umbel, but by lengthening of the axis 
passes to the panicle or corymb. ‘The transition from one to the 
other is so easy, and often exemplified on tho same tree, that it is 
