8. Phillip-ea. Calyx bracteolate, tubular, 4-5-
toothed. Corolla ringent, upper lip 2- under 3-lobed,
spreading.
9. Epiphegus. Flowers polygamous. Calyx bracteolate,
urceolate, 5-toothed. Corolla bilabiate, upper
lip entire, under 3-lobed.
IV . I I yoban che eje.
10. Hyobanche. Calyx bracteolate, deeply 5-
cleft. Corolla ringent, upper lip long, entire, under
small, obscurely 3-toothed. Anthers deflex, pendulous,
1-celled!
11. Campbellia. Calyx bracteolate, tubular, 5-
toothed. Corolla sub-bilabiate, 5-lobed. Anthers,
deflex, pendulous, one-celled! opening by a pore at
the apex.
12. Christisonia. Calyx tubular, 5-toothed.
Corolla infundibuliform, sub-bilabiate. Anthers 2-
celled, one sterile, subulate. Placentas free, re-
yolute.
13. Harveya. Calyx inflato-campanulate, 5-lobed.
Corolla tubular, sub-bilabiate, 5-lobed. Anthers 2-
celled, one sterile, subulate. Ovary 2-celled, with
2 fleshy placentae in each.
14. A ulaya. Calyx tubular, 5-cleft Corolla
tubular, 5-lobed. Anthers 2-celled, one sterile, subulate.
Ovary 2-celled, with a single axillary, placenta
in each. [Obs. Mr. Bentham remarks of this
genus, “placentae in diversis speciebus magis minus
ve bilobae,” which seems- to indicate that it is correctly
referable to this order.]
P. S. After the above was written, I received
the 11th Vol. of De Candolle’s Prodromus containing
the article Orobanchacea by M. Reuter. On looking
over it, perhaps rather hurriedly, I do not observe
any thing tending to invalidate the views I
have ventured to advance, except with regard to the
genera of my section Orobancheece which, if I
rightly understand, he considers have all deearpel-
lary ovaries, while I suppose there are as many
carpels as placentas, each placenta being formed by
the union of the edges of 2 carpels the same as
occurs in most other ovaries having parietal placentae.
Nor can I see upon what grounds we are
to adopt other views with respect to this family. I
must certainly admit that it is unusual for the same
species and even the same individual to furnish in
so many flowers, examples of 4-5 and 6 carpels to
the ovary, which I find in Orobanche. But I believe
it is equally rare to find similar variations in the
number of placentae to each carpel, and for the simple
reason that the one is, with very few exceptions,
dependent on the other, the carpellary margins only,
except in these few instances, being placentiferous.
One circumstance, to which he seems to have paid
much attention, merits notice, namely, the position
of the lobes of the stigma in relation to the floral
axis, which he finds right and left in some, anterior
and posterior in others, implying that in the former
the placenta are anterior and posterior, and right and
left in the latter. In Orobanche they are right and
left, and in Conopholis and Anoplanthus anterior and
posterior. How are these differences to be accounted
for? I am unable to say, but their existence
goes far to show that, so far as our information on
that point of structure yet extends, little benefit is
likely to result from its study as an ordinal character:
nowever valuable it may prove as a generic
one. The fact, however, of the stigma being simply
two-lobed, seems to throw much doubt on the correctness
of the views I have been led to take as to
the plurality of carpels, but does not altogether invalidate
them, as each lobe may be formed from the
union of two adjoining carpels, but it seems more in
accordance with analogy, as regards the rest of the
order, to suppose that each carpel has two placentae
placed à little within the margin. But that theory
will not account for cases, o f which I have seen
many, in which 5 placentae occur. Were they constantly
in pairs 2 -4 -6 that explanation might be admitted,
but in cases where an odd one occurs,, it
cannot be accounted for on that principle though
easily explained on the principle I have supposed,
viz. “plurality of carpels with the placentae formed
in the usual way along the line of union of each
pair.” This view is further supported by the fact,
that I have occasionally observed à tendency to
dilatation, or spreading to the right and left, of the
margins of the placentae in Orobanche. But on this
question further observation is required, and for the
present enough has been said to call attention to the
subject.
1420. Cystanche lutea (Link and Hoffmans.
C. tubulosa, R. W. in Icon. Philipoea lutea, Desf.),
scape simple, fleshy, sulcated : bracts opaque, ovato-
lanceolate, substriated, longer than the calyx : calyx
campanulate, lobes ovate, obtuse: corolla narrow at
the base, tubular, arched outwards, dilated at the
throat, 5-cleft, lobes ovato-rotundate equal, spreading:
stamens hairy at the base: anthers large, ob-
solately mucronate : stigma capitate, emarginate.
(Reuter in D. C. Prod.l
“Scape furrowed, thick and fleshy, bracts elongated,
acuminate, amplexicaul at the base and, like
the shorter bracteoles, downy, translucent at the
edges, and sometimes obscurely toothed. Calyx £
the length of the corolla, its segments rounded, often
obscurely crenate. Corolla bent outwards from the
middle, lower-half tubular, erect, upper-half bellshaped,
inclining outwards; throat very wide with
two dimples anteriorly; limb slightly 5-cleft, with
equal, rounded, turned down lobes ; bottom of the
tube with the insertions of the stamens densely
woolly. Anthers apiculate, all cohering by the
woolly hairs fringing the pollen clefts. Style with
a clubbed, compressed, nodding tip and a somewhat
bilob ed stigmatic surface.
“Scind, in loose, sandy soil, on roots of Salsolas—-
grasses, and Calotropis Hamütonii, See. P . caloiro-
pidis, (Edgeworth) ?
“A fine species, varying from 6 inches to 6 feet,
from the point of attachment to the apex of spike.
General colour yellow with an occasional tinge of
purple: colour of flowers generally yellow, with a
tinge of purple before expansion; or sometimes
muslin white with two yellow streaks. Carpels
each with two biseriate placentas, when à third is
present (which is rare) it is situated anteriorly. Capsule
about an inch long with numerous seed, like
coarse gunpowder.” Stocks’ MSS. description which
accompanied the drawing of Fig. 1420-6ts.
A comparison of the placentation shown in transverse
sections of the ovary of this, with that of No.
1353, which presents a placentary structure, quite in
( * )