THE EVOLUTION OF PLANTS. 
25 
Berberis stcno2)hyUa— Mr. C. C. Hurst informs mc that he has raised 
500 seedlings from this hybrid : 90 per cent, have come true to parent; 
the remainder prove variable, some tending to one grandparent, and 
some to the other, but not one has completely reverted, all being more or 
less intermediate. For full account see Hybrid Conference Report, 
p. 121.* 
Calceolaria rugosa x C. herbacea, Hort. — Vouched by Mons. Mottet. 
This plant is uncommon, but is used for bedding. 
Canna. — The beginning of the modern race of large-flowered dwarf 
Cannas took place, according to Bailey, in 18G3, when C. iridiflora, of 
Peru, was crossed by C. Warscewiczi, of Costa Rica. This hybrid, 
familiar to many as C. Ehemanni, was again crossed by other species. 
The existing plant, I understand, does not revert to anything else. 
Mr. Geo. Paul writes : " Some kinds have certainly a tendency to come 
near the parent from seed." 
Fuchsia, — Mr. Banks, writing to Mr. Cannell, says : " As you know, 
in years gone by I raised many thousands of seedling Fuchsia, but I 
cannot say I ever remember their reverting to parents." 
Gladiolus — G. ganda^vensis, raised from cardinalis, floribundus, and 
psittacmus, G. Lemomei (purpureo-auratus x gandavensis), G. Nan- 
ceanus (SaundersixLemoinei). — Mr. Ernest H. Krelage writes: "We 
have never observed the reversion of hybrid Gladioli to their parents." 
Hippeastrum. — The garden Amaryllis. For parents see account by 
Mr. H. Veitch in the Journal of the Society, vol. xii. p. 243. Messrs. 
Jas. Veitch & Sons write: "On one occasion our foreman, Mr. Heal, 
raised seeds of the variety ' Empress of India ' by fertilising the stigma 
of a flower with pollen of the same variety. The resulting plants were 
all sufficiently identical with the parents to be distributed under the 
same name." It may here be remarked that a hybrid may vary within 
itself and yet come true from seed. It is only when it ceases to be itself 
that it fails to come true. 
Narcissi. — The Rev. G. H. Engleheart writes me : " Broadly speak- 
ing, 1 should pronounce that hybrid Narcissi do not come true from 
.seed. As a rule the seed produces a series of forms showing considerable 
divergence, and reproducing in varying proportions the back ancestry of 
the hybrid. But I have been able to see tendencies to become fixed. 
Thus I have a very fine N. incomparabilis of my own raising, named 
^ Seagull,' a large, tall plant with fine flowers, broad white perianth, and 
open yellow cup. Seed of this not crossed gives flowers which are all to 
be called * Seagulls ' — ix. I could at once see their parentage in their 
stature, form, and colour. They come near enough to be recognised as 
'Seagulls ' or the ' Seagull ' ntrain. And I think by seeding from the flowers 
which come nearest to the 'Seagull' type one could, in a few generations, 
fix the form." This, of course, amounts to more than is necessary for 
the hybrid to come true from seed. The hybrid is N. incomparabilis 
(N. Pseudo-Narcissus xN. pocticus), and the hybrid is true from seed so 
long as that is produced. Variation must be allowed for hybrids as well 
as for other plants, and, as will be seen elsewhere, hybrid strains as well as 
* Since the above has been printed, Mr. George Paul kindly informs nie that he 
has obtained reversions. 
