The red-fleshed orange, on the other hand,
fails to reproduce itself. Gallesio also observed that the seeds of
several other singular varieties all reproduced trees having a peculiar
physiognomy, but partly resembling their parent-forms. I can adduce
another case: the myrtle-leaved orange is ranked by all authors as a
variety, but is very distinct in general aspect: in my father's
greenhouse, during many years, it rarely yielded any seed, but at last
produced one; and a tree thus raised was identical with the
parent-form.

Another and more serious difficulty in determining the rank of the
several forms is that, according to Gallesio,[630] they largely
intercross without {336} artificial aid; thus he positively states that
seeds taken from lemon-trees (_C. lemonum_) growing mingled with the
citron (_C. medica_), which is generally considered as a distinct
species, produced a graduated series of varieties between these two
forms. Again, an Adam's apple was produced from the seed of a sweet
orange, which grew close to lemons and citrons. But such facts hardly
aid us in determining whether to rank these forms as species or
varieties; for it is now known that undoubted species of Verbascum,
Cistus, Primula, Salix, &c., frequently cross in a state of nature. If
indeed it were proved that plants of the orange tribe raised from these
crosses were even partially sterile, it would be a strong argument in
favour of their rank as species. Gallesio asserts that this is the
case; but he does not distinguish between sterility from hybridism and
from the effects of culture; and he almost destroys the force of this
statement by another,[631] namely, that when he impregnated the flowers
of the common orange with the pollen taken from undoubted _varieties_
of the orange, monstrous fruits were produced, which included "little
pulp, and had no seeds, or imperfect seeds."

In this tribe of plants we meet with instances of two highly remarkable
facts in vegetable physiology: Gallesio[632] impregnated an orange with
pollen from a lemon, and the fruit borne on the mother tree had a
raised stripe of peel like that of a lemon both in colour and taste,
but the pulp was like that of an orange and included only imperfect
seeds. The possibility of pollen from one variety or species directly
affecting the fruit p