of Westralian Plants. 
229 
produced one million seeds, while the latter’s sixty-seven thousand flowers 
produced only four hundred seeds. This surely gives the Eucalypt an 
enormous advantage over the Banksia. It took only seventy-five inflo¬ 
rescences to produce the four hundred Banksia seeds, while the Eucalypt 
required eighty-four thousand to produce its million. The advantage seems 
not so great; but it must be understood that the Banksia inflorescence is 
enormous compared with that of the Eucalypt. I think these estimates, 
somewhat rough though they are, fairly represent the advantage in seed 
production that Eucalypts possess over Grevilleoid Proteaceae. Surely 
the immensely greater number of its seeds should favour variability in 
Eucalyptus, and even without variation should give it immensely greater 
power to seize new territory and increase its numbers. 
I have been struck by the number of instances I have seen in which 
several or many small flowers are gathered together into inflorescences 
resembling, and perhaps often behaving as, one large flower. Its prevalence 
suggests that there is some special advantage in such an inflorescence. It is 
seldom, I think, an economy of tissue or a reduction of transpiring surface. 
A study of the genus Eucalyptus , however, suggests an advantage of 
considerable importance. The large solitary flower is found, for example, 
in E. macrocarpa , and E. loxophleba , Benth., will serve for an instance of the 
anthoidal inflorescence of many small flowers. In the first-named species 
of 3,000 sq. mm. of flower surface adapted for the disposal or receipt 
of pollen only the central square millimetre is stigmatic, whereas the 
inflorescence of E. loxophleba presents 12 sq. mm. of stigmatic surface 
evenly distributed amongst 1,000 sq. mm. of anthers. As there is nothing 
in the structure of the flowers to compel a pollinator to brush a central 
stigma with a part of its body which has been pollen-dusted, it seems to 
me certain that a flower of the latter species has a far better chance of 
pollination than one of the former, and I think it is not unlikely that it also 
has a better chance of cross-pollination. I have no exact data ; but my 
experience of the species has given me the impression that the flowers 
of E, loxophleba are far more fertile than those of E. macrocarpa. It is 
possible, however, that failure to fruit is not always due to failure to secure 
pollination. The solitary-flowered Eucalypts, I believe without exception, 
are few, well-marked, and restricted in range. The umbel-bearing species 
are numerous, usually very variable, and they often range over a wide area 
of territory. It seems possible that the form of inflorescence is contributory 
to these effects. Banksia attenuata has an inflorescence presenting many 
stigmas well distributed over a very large area, yet it is a very shy seeder. 
This tends to emphasize the fact that floral structure is but one factor in 
the production of seed. Banksia , however, differs from Eucalyptus', its 
pollen-bearing surface is not very much larger than its stigmatic surface, 
and its pollen easily rubs off. 
