QK YI

Nns¢g
fad » /§SoO
Vif

ILLUSTRATIONS

OF

SOUTH AMERICAN PLANTS.

BY

JOHN MIERS, FRS. F.LS., &c.

VOL...

LONDON:
H. BAILLIERE, PUBLISHER,
219, REGENT STREET.

PARIS: J. B. BAILLIERE, LIBRAIRE DE L’ACADEMIE ROYALE DE
MEDECINE, RUE DE L'SCOLE DE MEDECINE.

MDCCCL.

ILLUSTRATIONS

OF

SOUTH AMERICAN PLANTS.

BY

JOHN MIERS, F.RS. F.LS., &c.

LONDON:

H. BAILLIERE, PUBLISHER,
219, REGENT STREET.

PARIS; J. B. BAILLIERE, LIBRAIRE DE L’ACADEMIE ROYALE DE
MEDECINE, RUE DE L'ECOLE DE MEDECINE.

MDCCCXLVI.

Mc. Bot. Gariiss

ot apiienn
} as

PREFACE.

TuEsE contributions to the Botany of South America
are the results of observations made by the Author, during
several years residence, both on the Western and Eastern
shores of that vast continent. Having availed himself of the
opportunity of examining many plants in the living state,
and also preserved drawings and details of a great number,
he proposes to select from these, such as may be still undes-
cribed, or yet imperfectly understood. In the course of
these investigations, he has met with many ambiguities in
several already recorded species, and after a careful exa-
mination of dried specimens, whenever they have been
within his reach, he has sought to ascertain how far they
severally correspond with the most authorized generic cha-
racters, and how far it becomes necessary to remodel these
characters, in accordance with more recent, and more
extended observations: in this way, he hopes to have
collected a few satisfactory materials towards monographs of
several genera; these will not be offered in any regular
order, but as the subjects present themselves to his notice.
In his endeavours to carry out this plan, he has received
much kind assistance from Sir W. Hooker, and he has great
pleasure in acknowledging his obligation to that dis-
tinguished Botanist, for the liberality with which he has
afforded the use of his valuable library of reference, and the

PREFACE.

freest access to his extensive herbarium, which is_parti-
cularly rich in the botanical productions of South America.
From this fertile source, and the different herbaria in the
British Museum, as well from the materials in his own
collection, he has prepared the following contributions,
accompanied by illustrations and the details, which ac-
cording to his views, tend to exemplify the characters of
each genus referred to, The vast accumulation of plants
during the last few years, collected in all parts of the globe,
has been so great, that it becomes absolutely necessary to
define with greater accuracy the limits, not only of genera,
but of species, and where this can be accomplished upon
sufficient evidence, much will be done towards removing
the confusion that exists in many cases; but even should
the author of these Illustrations succeed in only a few in-
stances, he hopes to render some service, though it be not
very great, towards the advancement of the science of
Botany. !

A portion of the text that has first appeared, and will be
continued at different intervals in the “ London Journal of
Botany” under the title of “ Contributions to the Botany
of South America,” is here reprinted with a few alterations,
and a description in detail of the several plates offered as
“ Illustrations” of these respective subjects, will be given
at the end of each part; these will be drawn on stone by
the Author, who claims much indulgence for the many
imperfections attending his first attempt at any work of the
kind; 22;[ 4 on

Hammersmith,
July, 1846,

DESCRIPTION OF THE PLATES.

Puate I. exhibits a drawing of Salpichroa (Perizoma)
rhomboidea.

Fig. 1, is a transverse section across the lobes of the
corolla in bud, to show its mode of estivation. Fig. 2, the
corolla cut open, exhibiting the stamens and the fleshy woolly
ring at base. Fig. 3, a vertical section of the corolla and
calyx, showing the perigynous woolly ring, the fleshy hypo-
gynous disc, ovarium, style, and stigma, as well as the stamens
in their natural position. Fig. 4, the stamens seen in
three different points of view. Fig. 5, a grain of pollen,
highly magnified. Fig. 6, the ovarium. surrounded by its
fleshy disc, the style and the stigma. Fig. 7, a vertical
section of the ovarium, style and stigma. Fig. 8, a trans-
verse section of the ovarium :—all: magnified. Fig. 9, seed,
nat. size. Fig. 10, the same magnified, exhibiting the hairy
testa. Fig. 11, the same seen edgeways. Fig. 12, a sec-
tion of the seed, showing the albumen and embryo. Fig. 13,
the embryo extracted :—all magnified.

ii DESCRIPTION OF THE PLATES.
Puate II, is a representation of Dunalia Lycioides.

Fig. 1, is the flower. Fig. 2, the calyx, style and stigma.
Fig. 3, the ovarium, style and stigma. Fig. 4, the corolla
cut open :—al/ natural size. Fig. 5, the upper portion of the
corolla, magnified, showing the intermediate tooth and the
mode of estivation of its lobes. Fig. 6, the stamens seen
in different positions, together with the mode of dehiscence
of the anthers: the central portion of the filament is shown
adnate to the tube, while the margins are free to the base,
as well as the lateral setose teeth. ‘Fig. 7, the stigma, front
and side view. Fig. 8, is a transverse section of the
ovarium :—all magnified.

PuarTe III, is a drawing of Acnistus cauliflorus.

Fig. 1, the calyx. Fig. 2, section of the calyx, showing the
pistillum. Fig. 3, the corolla. Fig. 4, the corolla cut open :—
all nat. size. Fig. 5, corolla cut open:—magnified. Fig. 6,
transverse section of bud, to show the mode of estivation.
Fig. 7, articulated hairs seen on the corolla and other parts
of the plant. Fig. 8, stamens, front and side view, before
and after dehiscence. Fig. 9, transverse section of anther,
before dehiscence. Fig. 10, transverse section of anther, —
after dehiscence :—all magnified. Fig. 11,a grain of pollen :—
highly magnified. Fig. 12, ovarium, style and stigma.
Fig. 13, transverse section of the ovarium:—all magnified.
Fig. 14. seed:—nat. size. Fig. 15, seed, magnified. Fig. 16,
vertical section of seed. Fig. 17, transverse section of
seed. Fig. 18, embryo, enveloped in albumen. Fig. 19,
embryo extracted :—all magnified.

Puate IV. A, exhibits Himeranthus runcinatus.

Fig. 1, flower. Fig. 2, calyx. Fig. 3, corolla. Fig. 4,
corolla cut open, showing the hairy lines beneath the almost
sessile stamens, Fig. 5, the anthers seen in different posi-

DESCRIPTION OF THE PLATES. iii

tions, showing the enlarged fleshy connective. Fig. 6, the
calyx, ovarium, style and stigma. Fig. 7, pistillum. Fig. 8,
the articulated hairs seen in the corolla.

B, exhibits Himeranthus tridentatus.

Fig. 1, one of the lobes of the corolla. Fig. 2, the pis-
tillum.

Puate V. A, represents Himeranthus erosus.

Fig. 1, corolla, cut open. Fig. 2, stamens, seen in diffe-
rent positions. Fig. 3, pistillum.

B, drawing of Jaborosa integrifolia, copied from Sir Wm.
Hooker’s figure in the Bot. Mag. tab. 3489, showing the
much longer tube, and larger corolla, with its style far
exserted, having 5 long linear lobes.

Fig. 1, pistillum, showing its stigma, sometimes having
only three lobes. Fig. 2, stigma with five distinct lobes,
nat size. Fig. 3, sessile anther, with its apiculated connec-
tive.

Puate VI, A, is a representation of Dorystigma cau-
lescens.

Fig. 1, the corolla, cut open. Fig. 2, stamens, before
dehiscence. Fig. 3, stamens, after bursting. Mig. 4, trans-
verse section of the anther in each stage :—all magnified.
Fig. 5, a grain of pollen :—highly magnified. Fig. 6, calyx
and pistillum, nat. size. Fig. 7, a front and side view of
stigma, magnified. Fig. 8, the berry. Fig. 9, a seed :—doth
nat. size.

B, exhibits Dorystigma squarrosum.

Fig. 1, corolla, cut open. Fig. 2, stamens, before dehis-
cence. Fig. 3, stamens, after bursting. Fig. 4, calyx,
ovarium, style, and stigma, seen in front. Fig. 5, style and
stigma seen sideways. Fig.6, a berry. Fig. 7, transverse
section of a berry. Fig. 8, the dissepiment, with its central

vi DESCRIPTION OF THE PLATES.

placenta. Fig. 9, a berry seen edgeways. Fig. 10. seeds :—
all nat. size. Fig. 11, seed, magnified. Fig. 12, section of
seed. Fig. 13, embryo extracted from the albumen.

Puate VII. is a drawing of Trechonetes laciniata.

Fig. 1, the calyx. Fig. 2, lobe of the calyx. Fig. 3,
corolla. Fig. 4, corolla, cut open, showing the filaments,
entirely free to the base :—all nat. size. Fig. 5, stamens, seen
in different aspects :—magnified.. Fig. 6, the ovarium, style
and stigma :—natural size. Fig. 7, the style and stigma,
seen in two positions. Fig. 8, articulated hairs, seen espe-
cially in the flower, peduncle, and petiole :—all magnified.
Fig. 9, a grain of pollen :—highly magnified.

Puate VIII, represents a portion of a branch of Pio-
nandra (Ceratostemon) floribunda.

Fig. 1, is an expanded flower, seen from above. Fig. 2.
a side view of the same, Fig. 3. the calyx. Fig. 4, a
section of the corolla, with its stamens removed, showing
the annular fleshy ring:—all nat. size. Fig. 5, three of
the stamens, upon a portion of the fleshy ring, showing
the mode of their origin: slightly magnified. Figs. 6, 7 and
8, are stamens, seen under three points of view. Fig. 9,
transverse section across the stamen, showing the position
of the anther cells with the connective. Fig. 10, a grain of
pollen, more highly magnified. Fig. 11. ovarium, style, and
stigma, showing the fleshy lobes in the mouth of the
cavity. Fig. 12, vertical section of the style and stigma.
Fig. 13, is a transverse section of the ovarium :—all magni-
fied. Fig. 14, transverse section of the ovarium, more
highly magnified, to show the mode of placentation. Fig. 15,
a berry of Pionandra ciliata, natural size, but not yet
arrived at its full growth.

_—)

ii
ie
=
<

J Miers bthog

Printed by Hullmandel & Walton.

meranthus tridentatus.

Plate 4.

lila

en

4

- Jaborosa integrifolia.

D
o
oF
cop)

a
a
CO.
5°)
ae

5 ee et
re

wei

i

Printed by

vis

~_ Trechonates Jacimata._

ILLUSTRATIONS

oF

SOUTH AMERICAN PLANTS.

SALPICHROA.

Under this name it is proposed to class several plants that
have been hitherto arranged in Afropa, the limits of which
genus remained for a long while undefined, many species
having been referred to it, and again removed by different
botanists. Its character, as given by Professor Spenner (Gen.
Pl. Germ. p. 21, tab. 18) upon the typical species A. Bella-
donna is deficient in so far as regards all the South American
species. That offered by Prof. Endlicher (Gen. Pl. n. 3857)
has evidently been framed with the intention of embracing
the whole of these, amounting to about ten, which, however,
include two other very distinct forms; of these, four will be
classed in Salpichroa, and the remaining six under the name
of Hebecladus.* The plants before mentioned possess a calyx

* They are all remarkable for their conspicuous flowers presenting an
intermediate tooth in the plicature between the lobes of the corolla. The
generic name of Hebecladus is derived from Sn, pubes, cdadoc, ramus tener,
- in allusion to their habit, which much resembles that of Salpichroa, but is
more suffruticose. The following elements for a generic character have
been taken from the species I have examined in the dried state.
Hesxciavvs. (gen. nov.) Calyx brevis, profunde 5-partitus, laciniis ovatis

submembranaceis, 1-nerviis, venosis, persistens. Corolla infundibuli-

formis, tubo amplo calyce 2-6 plo longiori, fauce ampliato, limbo patenti-
B

2 ILLUSTRATIONS OF

that scarcely enlarges, and that is usually cleft, almost to the
base, into five linear erect segments, not a campanulate,

sinuato, 5-lobo, lobis acutis, sepissime dentibus interjectis, estivatione
basi valde plicatis. Stamina 5, imo corolle inserta, filamentis filiformi-
bus, glabris, basi dilatatis, antheris exsertis, cordato-oblongis, adnatis,
2-lobis, longitudinaliter dehiscentibus, polline albido. Ovarium subro-
tundum, glabrum, (disco nullo?), 2-loculare, placentis dissepimento
adnatis, pluri-ovulatis. Stylus simplex, exsertus. Stigma clavato-
capitatum, sub 2-lobum. Bacca globosa, parva, calyce membranacea
suffulta. Semina plurima in pulpam nidulantia, compressa, reniformia,
testa reticulata. Embryo intra albumen carnosum hamato-arcuatus,
cotyledonibus semiteretibus; radiculi tereti, inferné paulo crassiora,
duplo longiori, hilum petente. Suffrutices Americe intertropice, ramulis
subdichotomis, flexuosis, teneris ; foliis plerumque geminis, altero vix mi-
nori, ovatis, ellipticis, vel cordatis, integris, petiolatis. Inflorescentia
pedunculo solitario laterali, cernuo, floribus 1-2 vel plurimis, umbellatis,
rubris, flavis, vel rubro viridescentibus, Bacca alba, pisi magnitudine.

1, Hebecladus viridifiorus. Atropa viridiflora. H. B. K. 3, 11, tab. 196;
Caule fruticoso volubili, foliis geminis, elliptico-ovatis, subacuminatis,
integerrimis, basi in petiolum decurrentibus, utrinque (presertim subtus)
hirtellis; pedunculo bifloro ; floribus nutantibus ; corolla calyce 3-4 plo
longiori, basi externe tuberculis 5 instructa.x—Nova Granada.
Specimens of this plant exist in the herbarium of Sir William Hooker,

collected by Professor Jameson on the Western side of the Volcano of

Pichincha, at an altitude of 13,000 feet ; by Colonel Hall, in the Valley of

Lloa, who describes it as a large shrub ; and again by Professor Jameson

in Columbia (n. 195); and by Goudot, at Bogota in New Granada. The

leaves are 3 inches long, and 14 inch broad, on a petiole $ inch long; pe-
duncle 1-2 flowered, the calyx is somewhat pentagonous, and deeply cleft
into 5 thin greenish triangular segments, with ciliate margins; the corolla
is slightly pubescent ; the filaments are wholly free to the base, where: they
have a short triangular dilatation, which is ciliate, above this they are
slender and quite glabrous; the anthers are cordate at base, apiculate at —
summit.

2. Hebecladus umbellatus. Atropa umbellata. R. & P. 2, 44, tab. 181, a.;

‘ caule frutescente, flexuoso fragili; foliis subgeminis, cordato-ovatis,
subangulatis, pubescentibus ; pedunculo axillari, laterali, umbellato,
multifloro, nutante; corolla mellifera, purpurea, limbo reflexo, ciliato,
luteo, plicaturis vix dentatis ; staminibus inclusis; stylo exserto; bacca
albida, calyce patenti anffulta,—Peravize collibus circa Limam et Chan-
cay. v. s. in Herb. Hooker, Amancaés prope Limam (Mathews, n. 722),
A small shrubby plant; the leaves are not only somewhat angularly
sinuate, but have eroso-denticulate margins.

SOUTH AMERICAN PLANTS. 3

5-partite calyx, with stellately patent lobes, increasing con-
siderably in size with the fruit; it has a narrow tubular fleshy
corolla, often contracted in the mouth, not one that is cam-

3. Hebecladus biflorus. Atropa biflora, R. §& P. 2, 44, tab. 181, 0. Pubes-
cens; caule fruticoso, ramis glabris, nutantibus ; foliis ovatis, acutis,
superioribus geminatis, utringue glabris, subtus pallidis, venis promi-
nentibus ; pedunculis hirtellis, 2-floris, nutantibus; corolla purpu-
rascente, tomeutoso-pilosa, limbo luteo-viridi; staminibus exsertis,
antheris ceruleis, sagittatis; bacca depresso-rotundata, alba, calyce
patenti suffulta.—In Andibus Peruvie. v. 8. in Herb. Hooker, Obra-
jillo et Cuallay ad Vallem Cante (Mathews).

4. Hebecladus bicolor. Atropa bicolor, R. & P. 2,45; caule fruticoso;
ramis teneris, angulatis; foliis plerumque geminatis, ovatis, acutis,
angulatis, glabris; pedunculo axillari laterali, umbellato, 3-4-floro.
corolla rubicund4, limbo viridescente.—In Andibus Peruvie.

5. Hebecladus asperus. Atropa aspera. R. § P. 2,45.; caule herbaceo,
‘dichotomo; foliis geminis, ovatis, altero minori, asperis; floribus soli-
tariis, cernuis; corollé luteo-cerulea, fauce violacea, campanulata ; fila-
mentis hirsutis, violaceis; bacca alba,—Amancaés prope Limam.

6, Hebecladus intermedius, (sp. nov.); caule glabro ; foliis ovatis, angu-
lato sinuatis, margine erosis, utrinque parce pilosis, venis pulverulentis ;
umbella trifora; corolla tubulosa, floccoso-hirtella, lobis 5 sublineari-
bus, dentibus interjectis ; antheris sagittatis, longe exsertis, filamentis
gracilibus, glabris.—v. 8. in Herb. Hooker, Purruchuco | Peruvie.
(Mathews, n. 524. sub nomine Afrope aspere, R. & P.)

This plant, in the shape of its leaves, approaches A. aspera, R. & P.,
but they are by no means asperous; the flowers differ much from the
figure of that species, resembling more those of A. biflora ; the leaves are
32 inches long, and 14 inches broad, on a petiole 3 inch long ; the stem of
the umbel is § inch long, the pedicels being 10 lines, the calyx is 4 lines,
the tube of the corolla 1 inch, the border 4 lines in length.

7. Hebecladus Janceolatus, (sp. nov.); caule flexuoso, glabro, subangulari ;
foliis geminatis, altero minori, lanceolato-ellipticis, basi cuneatis, apice
acuminatis, petioloque utrinque parce molliter hirtellis; umbella in
axillis laterali, 4-flora; calycis margine floccoso ; corollz tubo rubello,

glabro, ore flavo, 5-lobo, lobis acutissimis, puberulis, dentibus tot con-

spicuis, acutis, interjectis ; staminibus inclusis ; stylo exserto, clavato.

—Columbia, v. s. in Herb. Hooker (Hartweg, n. 1301).

The leaves of this species are 3% inches long, and 14 inch broad, on a
petiole 4 inch long; the stem of the umbel is 1? inch, and the pedicels
Q lines in length. ‘The flowers are of similar size to those of H. viridiflora.
8. Hebecladus Turneri, (sp. nov.) Caule flexuoso, tereti; foliis geminis,

B 2 .

4 ILLUSTRATIONS OF

panular, veined, thin in substance, and widened in the throat ;
its filaments are short, slender, and inserted in the middle,
not in the bottom of the corolla; the anthers are linear, erect,
nearly exserted, not oval and reflexed in the middle of the
tube; its ovarium is deeply imbedded ina large coloured fleshy
disc, not wholly free, or at least supported ona small 5-lobed
ring. The stigma is clavate, almost cup-shaped, not deeply sul-
cated, 2-lobed, and reflexed. The berry is of a bright scarlet
colour, not greenish or black. The testa of the seed is rugous,
and covered with rigid hairs, not smooth and reticulated:
characters offering many well marked points of distinction
from Atropa. There is another remarkable difference be-
tween Atropa and Salpichroa ; in the one, the corolla is thin,
becoming membranaceous and unchanged in drying, while
that of the other is thick and fleshy, becoming black as it
dries, a character it possesses in common with most of the
Jaborosa group, such as Jaborosa, Himeranthus, Dorystigma,
and by Juanulloa, &c. The name is derived from cadmyé
buccinum, x%poa colour, in allusion to its pretty, trumpet-
shaped flowers, and the following are its generic characters.
Saupicuroa. Calyx persistens, sepissimé profunde 5 par-
titus, lobis linearibus acutis, fructifer haud mutatus, rarius
5-fidus. Corolla hypogyna, subcarnosa, infundibuliformis,

qualibus, ovatis, acutis, basi late rotundatis, in petiolum decurrentibus,

utrinque pilis mollissimis sparsis; umbella cernua, 6-8 flora; calycis

laciniis lanceolatis, pubescentibus ; corolla parva, glabra, aurantiaca,

lobis acuminatis, margine floccosis; staminibus brevibus, inclusis ;

stylo exserto ; bacca rubra calyce patenti suffulta.—Patria ignota.

This specimen exists in Sir W. Hooker’s Herbarium, evidently dried from
a cultivated species by Dawson Turner, Esq. without any note of its
origin or the place of its growth, The leaves are 34 inches long, and
24 inches broad ; the stem of the umbel is 4 lines, the pedicels 6 lines, the
corolla 6 lines long, the latter being of an orange colour, rather broad in
proportion. a

Species dubia.

9. Hebecladus hirtellus. Atropa hirtella, Spr. 1, 699. Ramis asperis;

foliis ovato-oblongis, acutis, scabris ; pedunculis elongatis, 1-floris,
filiformibus,—Brazil, ok

SOUTH AMERICAN PLANTS. 5

interdum tubulosa, fauce subconstricta, limbo 5-fido, lobis
lineari-oblongis, reflexis, eestivatione fer? induplicatis, mar-
ginibus floccosis, vix introflexis. Stamina 5, equalia, sub-
exserta; filamenta filiformia, glabra, medio corolle orta.
Anthere lineares, erectee imo dorsi affixee, liberee, circa
stylum conniventes, 2-loculares, loculis parallelis adnatis,
rimA longitudinali dehiscentibus. Ovarium liberum, coni-
cum, disco carnoso magno colorato suffultum, 2-loculare,
placentis centralibus, é dissepimento formatis, multiovu-
latis. Stylus simplex, filiformis, erectus, stamina excedens,
basi conicus. Stigma subcapitatum, depressum, cavum,
obsolete 2-lobum. Bacca ovalis, carnosa, 2-locularis, stylo
apiculata, Semina numerosa, in pulpam nidulantia, rhom-
boideo-rotundata, valde compressa, testa rugosa, utrinque
(preesertim versus marginem) pilis simplicibus rigidis dense
vestita, hilo marginali. Embryo intra albumen carnosum
arcuatus; cotyledonibus semiteretibus ; radicula tereti
hilum spectante.—Herbe Americe meridionalis puberule,
diffuse, ramose, subscandentes ; caule angulato, flexuoso,
suffruticoso; ramis foliis oppositis. Folia solitaria, vel
gemina, rarius terna, integra, in petiolum longum decur-
rentia. Flores solitarii, albidi, vel lutei, siccitate nigricantes,
pedunculati, demum cernui. Bacca rubra.

§ I. Eusaupicaroa. Corolla lenge tubulosa, intus imo glabra.

1. Salpichroa glandulosa. Atropa glandulosa. Hook. Bot.
Misc. 2.230. Hook. Icon. 106; caule fruticoso, suberecto;
foliis geminis, longe petiolatis, cordato-ovatis, glanduloso-
pubescentibus, seepe incano-tomentosis ; calyce pubescente,
profunde 5-partito, laciniis linearibus; corolla flava, longe
tubulosa, staminibus styloque inclusis.—Pasco Peruvie.

2. Salpichroa dependens. Atropa dependens. Hook. Icon. 107 ;
caule suffruticoso, pendente; foliis geminis, cordato-
ovatis, breviter petiolatis, fere glabris, subtus pallidis;
calyce tubuloso, tenui, pubescente, 5-fido, demum fructifero
subampliato hine fisso; corolla longe tubulosa, stylo sta-
minibusque exsertis,—Peruvia (Mathews, n. 829).

6 ILLUSTRATIONS OF

3. Salpichroa jirsuta. Atropa hirsuta, Meyen (Reise um die
Erde, 1, 466). Nees ab Esenb. (Nov. Act. 19, Suppl. 1, 389) ;
caule suffruticoso, ramosissimo, diffuso; foliis geminis,
subcordato-ovatis, longe petiolatis, calycibusque profunde

5-partitis, hirsutis; corolla tubulosa, antheris cum styli

apice exsertis.—Peruvia, circa Pisacomam, altitudine 15,000

ped.

This appears to be the same plant, of which many speci-
mens exist in the herbarium of Sir William Hooker from
various localities, viz:

Pichincha (Jameson, n. 32), Pichincha (id. n. 301), Colum-
bia (Hartweg, n. 1311). In these the stem is 4-angular,
flexuose; the leaves are solitary, geminate, and ternate, from
$ to 1 inch long, and 4 to 9 lines wide; the petiole is nearly
as long as the leaf, slender, and caniculate; the internodes
are about the length of the leaves; the flowers are solitary
and lateral; the corolla about ten lines long, tubular, of a
greenish yellow colour, glabrous, having oblong obtuse lobes,
with ‘floccose margins; the oval berry is terminated by the
persistent glabrous style. Andes of Peru (Md‘Lean), where the
leaves are more ovate, nearly glabrous above, pubescent
beneath, the margins being somewhat crenate. Andes of
Quito (Jameson, n. 125), of more stunted growth, a flexuose
stem, presenting many short knotty leafless internodes ;
leaves ternate, barely 3 to 4 lines long, 2 to 3 lines broad,
petiole 3 to 5 lines long. Bogota (Goudot), very near the last ;
the branches are somewhat more divaricate, and of a darker
reddish colour.

4. Salpichroa ramosissima, sp. nov.; caule fruticuloso, 4-gono ;
axillis annotinis nodosis; ramulis pubescentibus ; foliis
utrinque glabriusculis, geminis ternisque, equalibus, sub-
membranaceis, elliptico-oblongis, basi rotundatis, in petio-
lum gracilem decurrentibus, apice subacuminatis, margine
ciliatis ; floribus solitariis lateralibus; corolla tubulosa,
calyce 3plo longiori, fauce contracta, lobis acutiusculis, re-
flexis, margine velutinis.—Purruchuco, Peruvie. v. s. in
Herb. Hooker (Mathews, n. 1053 sub nomine Atrope ramo-
sissimee),

SOUTH AMERICAN PLANTS. 7

A very distinct species, both in the form of the leaf and the
shape of the corolla; the leaves are 1} inch long, 7 to 9 lines
broad, the petiole, which is slender and caniculate, being
4 inch long; the lobes of the calyx are narrow, almost linear,
and slightly hairy; the berry is red, of an oval shape, $ inch
long, 4 inch in diameter; the seeds are flattened, with a hairy
testa resembling that recorded in the following section,

§ Perizoma. Corolla brevis, medio et fauce constricta, intus
annulo carnoso lanato instructa.

5. Salpichroa rhomboidea. Atropa rhomboidea. Hook. Bot.
Misc. 1, 135, tab. 37.; foliis rhomboideo-ovatis, basi rotun-
datis in petiolum attenuatis, fere glabris, margine petio-
loque ciliatis, interdum valde pubescentibus,—Bonaria.
Busbeckia, sp. Mart. MSS.

I met with this species in 1825, when its details were re-

corded by drawings from the living plant; it was also col-

lected about the same time by Dr. Gillies, from whose speci-
mens Sir William Hooker gave the excellent figure and
description above quoted. I had long ago separated this
from Atropa asa distinct genus under the name of Perizoma,
and had prepared a drawing to exemplify it; but upon
examining with more attention the species of the foregoing
section that exist in Sir William Hooker’s herbarium, I was
led to the conclusion that it is better to place it as a distinct
section of Salpichroa, on account of the close resemblance of
the most essential characters of the flower and the seeds ; the
presence of the hairy perigynous ring and the different size

and shape of the tube of the corolla not —- more than a

sectional difference.

This is a weak plant trailing ayiong?! gle or on : the
ground. The stem is slender, sub-4-angular, somewhat
flexuose. The leaves are sometimes almost glabrous, often
slightly pubescent on both sides, with very short articulate
hairs, the margin and petiole being ciliated. The flowers are
usually solitary (or geminate when the axils present ternate
leaves); they are cernuous upon slender pubescent lateral

8 ILLUSTRATIONS OF

peduncles. The calyx is deeply cleft into 5 long sharply-
pointed, almost linear erect segments ; it is slightly pubescent
and persistent, it does not increase much in size nor become
patent as the fruit ripens. The corolla is short and tubular,
broad in proportion, quite white and smooth, and sensibly
contracted both in the middle where the filaments are inserted,
as well as in the mouth; the segments of the border are
narrow, acuminated and reflected; the estivation is slightly
induplicate, the inflected edges adhering by their woolly mar-
gins; near the base of the tube is seen a somewhat fleshy
ring, which is covered with long, woolly, white, articulated
hairs. The stamens are equal, short, slender, and glabrous ;
the anthers, converging around the style above the mouth of
the tube, are narrow, linear, 2-locular, with parallel adnate
cells, which burst longitudinally somewhat laterally; the pol-
len is yellow and farinaceous, and when seen in the micro-
scope, both in the dried and humid state, is globular with 3
salient obtuse equidistant points. The ovarium is pyramidal,
its base being deeply enveloped by a conspicuous fleshy
orange-coloured disc; the style is conical at base, above which
it has a broad band of long white hairs, and is quite smooth
towards the summit, which rises a little above the anthers;
the stigma is rather large, clavate, somewhat hollow, cup-
shaped, and fleshy. The berry is of an oblong form, of a
bright scarlet colour, about 5 an inch long, and 2 wide, api-
culate at the summit with the persistent base of the style; it
contains from 16 to 20 rather largé dark brown seeds, enve-
loped in pulp; these are of a roundish square form, greatly
flattened, with a small hilum on the marginal edge ; the testa
is hard, brittle, and rugous, both its surfaces, and especially
the margin, being densely set with long rigid simple hairs;
the albumen is fleshy rather compact, and encloses a nearly
annular, filiform, embryo; the radicle is about one-third the
length of the embryo, and points towards the hilum, the coty-
ledons being semiterete and arcuate.*

* A representation of the above species is shown in Tab. 1. an explana-
tion of its details being given in the description that accompanies it.

SOUTH AMERICAN PLANTS, 9

I found this plant at several places in the Pampas, at a
distance of nearly 600 miles in the interior, and afterwards
near Buenos Ayres; and although these differ very much in
appearance, I can hardly venture to designate them as dis-
tinct species; they are therefore added merely as varieties.

Var. 6. divaricata; caule flexuoso, valde divaricatim ra-
moso, foliis obovatis, minoribus, utrinque pilosiusculis.—
Pampas, ab Esquina de Medrano usque Frayle muerto, Pro-
vincie Cordovensis.

This plant is of more straggling habit, the branchlets
spreading very widely, and the leaves scarcely exceeding 4 to 3
of an inch in length, the stem being 4-angular and pu-
bescent.

var. y. pubescens.—caule 4-gono ; foliis fere ovalibus, rotun-
datis, utrinque valde pubescentibus, petiolo gracili— Pampas,
San Luiz usque Rio Quinto.—This presents a very different
appearance to the last variety; the leaves being about the
same size, but the petiole is much longer and more slender.
6. Salpichroa ciliata. Lycium ciliatum, Schl. Linn. 7, 70:—

suffruticosa, subscandens, ramis elongatis, flexuosis, ramu-

lis interdum subspinescentibus ; foliis sparsis, inequila-
teris, rotundato-ovatis, interdum subcordatis, in petiolum
attenuatis, apice brevissime acuminatis, utrinque glabris,

_ margine denticulato-ciliatis ; floribus axillaribus, solitariis,
brevissime pedunculatis, cernuis, calycis laciniis linearibus,
acutis ; corolla puberula, infundibuliformi, genitalibus ex-
sertis.—Rio Negro, Banda Oriental (Sellow).

From Professor Schlechtendahl’s description, it is evident
that this plant has a close affinity to the foregoing. The
eaves are 14 inch long, 16 to 18 lines broad; the peduncle is
14 to 2 lines long, erect, and curving downward ; the calyx
is deeply cleft into 5 linear segments, is covered with short
glandular pubescence, and the margins are ciliated with glan-
dular hairs. The corolla, somewhat longer than the calyx,
is downy externally with glandular hairs, (articulate ?), funnel-
shaped, the segments of the border are triangular, rather
obtuse, reflexed, with glandular ciliate margins; the stamens

c

10 ILLUSTRATIONS OF

and style are far exserted, the filaments being inserted above
the base of the corolla, where the tube is contracted, and
closed internally by a ring of dense villous hairs. The
berry is red, globose, and shorter than the persistent calyx,
which increases in size to the length of 5 lines. The seeds
are numerous, whitish, compressed, subreniform, but were
not ripe when seen.

DUNALIA.

Among the plants sent from Peru by Mathews is one
marked Lycium obovatum, but it is clearly not the one
figured under that name in the Flora Peruviana.* In its

* I have searched in vain for this plant in the herbarium of Ruiz and
Pavon in the British Museum, nor can I find it elsewhere. In the herba-
rium of Sir Wm. Hooker there are, however, three plants from central
America that bear much resemblance to it. It cannot be a true Lycium
on account of the estivation of its corolla which is valvato-plicative (not
imbricate) and its lobes are acute with pubescent margins (not rounded and
smooth) ; it appears to me that with some others I shall point out, these
will form a distinct group; they possess a habit widely different from
Lycium, although they are all spiny shrubs, generally with 1 or 2 violet
or crimson flowers growing out of the fascicles of rather small fleshy leaves
that cluster upon the spines; the cérolla is usually broader and more
tubular than in Acnistus or Lycium, with stamens often unequal and in-
cluded; the calyx has generally acute lobes, and is not pentagonous with
obtuse lobes as in Acnistus, They offer much resemblance in external
appearance to the species of Dunalia, above described, but they want the
intermediate tooth in the corolla and the appendiculate stamens of that
genus. I propose to call them by the name of Lycioplesium from Lycium,
and Answoy, approximatus. It may be said that they should, like the
Lyciobatos of Endlicher, form a distinct section of Lycium, but on account ~
of the estivation of the corolla, the generic character, so altered to admit
of them, would necessarily include Acnistus, Salpichroa, Chenesthes, and
Tochroma ; genera decidedly inadmissable. The following is therefore
offered as the generic character. = eegeer
LyctopLestum, (gen. nov.) Calyx ovato-campanulatus, 5-dentatus, per-

sistens. Corolla tubulosa, limbo 5-partito, estivatione lobis acutis,

valvato-plicatis, margine tomentosis. Stamina 5, corolle longitudine,
ultra basin inserta, subinzequalia, filamentis insertione villosis, vel omni-
no glabris ; antheris oblongis, 2-lobis basi cordatis longitudine dehis-
centibus, Ovarium obovatum, 2-loculare, placentis e dissepimento forma-

SOUTH AMERICAN PLANTS. hl

spiny habit it greatly resembles a Lycium or Grabowskya,
but, on examining the flowers I, found it to possess all the

essential characters of Dunalia. The genus Dunalia, founded

tis, multi-ovulatis. Stylus simplex. Stigma bilobo-capitatum. Bacca calyce

inclusa. Semina numerosa, parva, compressa, reniformia, albuminosa ;

embryo cylindricus, annularis.

Frutices Americe meridionalis glabri vel fomentosi spinosi ; folia crassius-
cula, oblonga, in petiolum basi angustata, in spinis junioribus subfasciculata ;
flores pedunculati (1-2) in quoque axilla ; corolle violacee vel rubescentes ;
bacce rubre vel aurantiace.

1, Lycioplesium obovatum. Lycium obovatum, R. & P. 2.46: tab. 183. c.—
glaberrimum ; ramis patentibus, aculeis rigidis, junioribus fasciculato-
foliosis, adultis nudis ; foliis obovatis, obtusis, crassiusculis, basi in
petiolum decurrentibus ; pedunculis solitariis, extra-axillaribus, 1-floris,
nutantibus, corollis tubulosis, limbo erecto, marginibus floccosis.—Pe-
ruvia (Tarma).

A shrub 12 feet high, very smooth, leaves including the petiole 14 inch,
7 lines broad; peduncles 7 lines long; calyx urceolately tubular, 2 lines
long, with 5 short obtuse lobes, having a small mucro at the apex;
corolla tubular, purplish violet, 9 lines long, lobes of border short,
3-angular, erect, with floccose margins; filaments glabrous, arising above
the base of the tube where it is pubescent ; anthers included within the
mouth; the style is exserted; the berry is globose, orange coloured.
2. Lycioplesium nitidum, (n. sp.) :—fruticosum, spinosum ; foliis alternis,

ovato-oblongis, glaberrimis, crassiusculis, supra lucidis, subtus pallide

glaucis, rarius in venis tomentosis, margine revolutis, apice obtusis et
minute retusis, basi cuneatis, in petiolum brevem attenuatis ; floribus
binis, pedunculis apice incrassatis, corolla rubro-violacea, staminibus
exsertis. Columbia, Pulzhum, in valleculis latere occidentali. (Jameson)

v. s. in Herb. Hooker).

The young shoots (about $ inch long) from which the leaves and
flowers spring, finally become converted into spines ; the peduncle is 7 lines
jong, and thickened at its apex; the calyx is campanular, 2 lines long,
about the same breadth, and has 5 short rounded teeth with a small mu-
cronate acumen; the corolla is smooth, tubular, 10 lines long, with 5
small erect lobes with a mucronulate apex and floccose margins ; the an-
thers are oblong, 2-lobed, cordate at base, yellowish green, affixed at
base, the filaments being inserted above the base of the tube of the
corolla, where it is densely pubescent.
3. Lycioplesium ovatum, \n. sp.):—fruticosum, ramis sub-flexuosis, spinosis,

ferrugineo-tomentosis ; foliis glaberrimis, crassiusculis, ovalibus, utrin-

que rotundatis, margine revolutis, supra lucidis, subtus pallidioribus,
breviter petiolatis ; floribus solitariis, extra-axillaribus ; corolla glabra,
c 2 .

12 ILLUSTRATIONS OF

by Prof. Kunth on a shrubby plant with much the habit of a
Witheringia, brought from the Cordillera of New Granada
by Humboldt, was placed by that learned botanist among
Cestrinee, on account of the resemblance of its flowers to
those of Cestrum, although he confesses he knew nothing of
the form of the embryo of its seed. Until this fact be ascer-
tained it remains doubtful whether it may not with equal
reason be classed in Solanew, near Salpichroa or Chenesthes,
which view is much favoured by its numerous ovules seen
upon the thickened placenta on the dissepiment; but on the
other hand it must not be forgotten that some analogy exists
between the appendiculate processes of the filaments in this
genus, and the singular projection often seen upon the fila-

rubro-flavida, tubo latiori, staminibus inclusis.—~Nova Granada, (Gou-
dot.) v. s. in Herb. Hooker.

This is a shrub 10 or 12 feet high, with leaves 11 lin. long, 7 lin. broad,
petiole 2 lin. ; corolla smooth both without and within, except at the place
of the insertion of the filaments somewhat above its base, where it is
tomentous ; the lobes of the border are oblong, obtuse, mucronulate at the
apex, and tomentous on the inflected margins ; filaments glabrous, anthers
2-lobed, oblong, yellowish green, in the mouth of the tube; berry black.
4. Lycioplesium horridum. Lycium horridum, HBK. 3.52. Lycium par-

vifolium, Wild, R. and Sch. 14.698 3—suffruticosum, spinosissimum ;

foliis fasciculatis, obovatis, obtusis, crassiusculis, nitidis ; floribus soli-
tariis, pendulis; corolla violacea, staminibus subinclusis.—Andibus

Peruvie circa Caxamarcam.

This is said to be a shrub 6 feet high, with many spreading, thick, leaf-
less branches ; branchlets an inch long, furnished with leaves at base, spi-
nose at the apex ; leaves fasciculate (4-7), petiolate, 4-6 lin, long (includ-
ing the petiole 1 lin.) 2-24 lin. broad; flowers subaxillary, peduncles 3-4
lin. long, filiform, and smooth ; calyx semiglobose, obsoletely 5-toothed,
nearly entire, smooth, with 5 small acute teeth, about 1 line long ; corolla
smooth, tubular, somewhat curved, 7 lines long, border with 5 equal, acute,
spreading lobes with ciliate pubescent margins ; stamens inserted at base
of tube, and equal in length to the corolla, filaments smooth, anthers erect,
oblong ; ovarium conical, smooth ; style filiform, smooth, rather longer
than the stamens ; stigma thickened, green,

5. Lycioplesium Meyenianum. Lycium (Grabowskya?) Meyeniana, Nees ab

Esenb. Nov, Act. 19 Suppl. 1.390. Atropa spinosa. Meyen, Reise um die

Erde, 1,416 ;—erectum, rigidum, spinosum ; foliis lanceolatis, obtusis,

SOUTH AMERICAN PEANTS. 13

ment of some species of Cestrum, which appears as if two

lateral lobes were agglutinated into one salient tooth.
The following is proposed for its generic character :

Dunauia Kunth. (char. emend.)—Caly# urceolatus sub-
pentagonus, 5-dentatus. Corolla hypogyna longe tubulosa
10-nervia, limbo 5-fido, lobis estivatione marginibus floc-
cosis vix induplicatis basi plicatis, dente mucronato erecto
interjecto. Stamina 5 corolle tubo inserta, inclusa vel
exserta, filamentis basi dilatatis et pubescentibus, linea
centrali tubo adnatis, marginibus membranaceis liberis,
superne glabris, omnino liberis, 3-partitis, laciniis latera-
libus linearibus acutis erectis, intermedio gracili antheri-
fero; antheris oblongis, 2-lobis, basifixis, longitudinaliter
dehiscentibus.—Ovarium ovatum, 2-loculare, placentis cras-
sis dissepimento adnatis, multi-ovulatis. Stylus simplex
exsertus. Stigma eargimnato-capitatum. Bacca globosa,
ealyce suffulta, 2-locularis. Semina plurima, reniformia
compressa. Embryo ignotus.

Frutices Andicoli Americe meridionalis intertropice ; foliis
alternis petiolatis, geminis vel fasciculatis, glabris vel pubes-
centibus; floribus subumbellato-fasciculatis vel solitariis,

extra-axillaribus : corollis albis vel coccineis.

1. Dunalia solanacea, HBK. 3.55. tab. 194. Dierbachia sola-
nacea Spreng. Syst. 1.676.—inermis: ramulis tomentosis ;
foliis alternis, ovato-oblongis, acuminatis, basi ineequaliter
rotundatis, supra glabris, subtus incanis, stellato tomen-

glabris ; floribus solitariis, nutantibus ; calyce late campanulato, 5-den-
~ tato, 2-plo,3-plo-ve longiore, corolla violacea, staminibus inclusis.—

Peruvia, circa Pisacomam, altit. 15,000 ped.

A handsome shrub, said much to resemble the last mentioned species,
but differing in its smaller lanceolate leaves ; branchlets covered with thick
white tomentum, 1-12 in. long, often spiny at the apex, bearing fascicles
of leaves at the base; peduncles axillary, smooth, 6 lines long; calyx
smooth, 3 lines long, with 5 short equal obtuse teeth, terminated by a
woolly cuspidate point ; corolla tubular, 15 lines long, with a slightly
spreading border, having 5 short triangular acute lobes with ciliate mar-
gins ; berry red, twice the size of a pea, globose, partly enclosed within
the calyx, which now becomes unequally 3-4 cleft. “gy

14 ILLUSTRATIONS OF

tosis; floribus umbellato-fasciculatis ; corollis stellato-

tomentosis.—Nova Granada.

The leaves are said to be about 10 inches long and 4 inches
broad, on a petiole an inch in length; the calyx does not
measure aline; the corollais white, nearly an inch long with
a narrow slender tube, the lobes of the border being ovate,
and 1-nerved; the stamens are very short, and placed in the
middle of the tube of the corolla, the lateral appendages
equalling in length the intermediate antheriferous filament ;
the style is much longer than the corolla; the berry is glo-
bose, glabrous, about the size of a pea.

2. Dunalia lycioides (sp. nov.)—fruticosa, glaberrima; ra-
mulis horrido-spinosis ; foliis fasciculatis (1-2-3), lanceola-
to-spathulatis, obtusis, in petiolum decurrentibus; floribus
(1-2) nutantibus, staminibus exsertis. — Peruvie Prov.
Canta, Tarma et Jauja. (Mathews n. 850) in herb. meo ;
etiam in herb. Hook. cum aliis Columbia (Lobb. n. 255)
et Bolivia (Pentland).

This is described to be a shrub 6 or 8 feet high. The
branches are flexuose, quite smooth with internodes scarcely
an inch distant, and a single stout, sharp pointed, divaricate
spine in each axil, 2 inches in length, the older ones being
bare and sometimes again spiny; the younger ones bearing
leaves and flowers. The leaves are smooth, fleshy, rounded
at the apex, and tapering at base into the petiole, they are 9
lines in length and 2% lines wide ; the peduncles are 4 lin.
long ; the calyx at first slightly pubescent, is urceolate, with
5 projecting ribs which terminate in as many short teeth, with
a mucronulate woolly apex. The corolla is broader and about
the length of the last species, being 10 lines long, smooth, of
a crimson colour, having a border of 5 short, rather erect
lobes, with floccose margins, and a narrow intermediate pli-
cature with tomentose edges and asmall erect tooth in the
centre. The crimson filaments are adnate by a central line
to the base of the tube of the corolla for one third of its
length, the upper part being wholly free, the lateral appen-
dages being short, acute, and only xsth part of the length of

SOUTH AMERICAN PLANTS. 15

the antheriferous portion, which is slender and subulate; the
anthers are oblong, yellow, protruding beyond the mouth of
the corolla. The fruit is unknown.*

ACNISTUS. Schott.

This genus was first proposed by Schott in 1829 (Wiener
Zeitschrift 4.1180) upon a Brazilian plant considered to be
identical with the Cestrum cauliflorum, Jacq. Hort. Schoen.
3.41 tab. 325. Another species, also confounded with it, had
been long previously known and figured by Plumier, under
the name of Belladonna frutescens (tab. 46, f. 1). The authors
of the Flora Peruviana have given a representation of a fourth
species, under the name of Lycium aggregatum, (2.45. tab.
182. f. a.); but as the characteristic features of the genus are
not delineated in the figures above quoted, nor any exact
details have, to my knowledge, yet been published, I offer the
following from my own observations. 7 st
Aontstus Schott. (char. reform).— Calyx campanulatus, sub-

5-gonus, obsolete 5-dentatus. Corolla hypogyna, infundi-

buliformis, fauce sensim ampliato, limbo 5-partito, patento,
reflexo, wstivatione lobis margine floccosis, valvato-indu-
plicativis. Stamina 5, corolle tubo supra basin inserta,
filamenta simplicia ; anthere 2-lobe, longitudinaliter dehis-

centes, seepe exsertee. Ovarium e disco calycino ortum, 2-

loculare, placentis crassis, dissepimento adnatis, pluri-ovu-

latis. Stylus simplex. Stigma capitatum, sub-bilabiatum.

- Bacca calyce suffulta, 2-locularis. Semina pauca, renifor-
mia, compressa; testa rugosa, dura. Embryo intra albu-
men carnosum, hamato-arcuatus, cotyledonibus semitere-
tibus, radicula tereti, inferne crassiori fere duplo longiori,
hilum spectante. | |

Frutices Americe tropice, foliis alternis, integris, junaoribus
aggregatis ; floribus pedunculatis, in axillis sepe annotinis,

* A figure of this species is given in Tab. 2, an explanation of its
details being given in the description that accompanies it. ee

16 ILLUSTRATIONS OF

fasciculatis, rarius in racemis terminalibus ; pedunculis
apice incrassatis.

There is very little tenable ground for maintaining this
genus, as hitherto constituted, distinct from Lycium, there
being hardly any single character that is not equally common
to both of them, excepting the hairy tuft at the base of the
filaments in the one, (and that is a very inconstant feature),
and the numerous fascicles of flowers in the cicatrices of the
fallen leaves in the other. An important distinction will, how-
ever, be found to exist in the estivation of the corolla. We
have the respectable authority of Schlechtendahl and Schott,
which has been acceded to by all succeeding botanists, that
Acnistus possesses an imbricate estivation. I cannot affirm
this statement, for in the Brazilian species upon which Schott
founded this genus, the lobes of the corolla unquestionably
adhere by their tomentous margins, which are mutually and
slightly turned in, a mode of estivation observed in many
arborescent species of Solanum, and very different from that
of true Lycium, where the lobes of the corolla offer an imbri-
cate or quincuncial zstivation. It therefore seems advisable
to unite with Acnistus, several species hitherto combined with
Lycium, forming part of the section called Anisodontia by
G. Don, and Lyciothamnos by Endlicher ; these mostly con-
sist of spineless trees or shrubs, with large leaves, having
flowers in umbellate fascicles, and I propose to confine within
the limit of Lycium proper, those shrubs, mostly with small fas-
ciculate leaves, whose branchlets terminate in spines, or have
a tendency to do so, that have only 1 or 2 flowers in each axil,
and with elements corresponding to the old generic character
exhibited by Gartner (de fructu 2.242), with the addition of
the before mentioned estivation.*

* The remaining species of Zycium in the section above alluded to, ap-
pear to me again distinct, approaching very closely to Dunalia, but as
their filaments want the lateral appendages peculiar to that genus, I pro-
pose uniting them under the name of Cuznesrues, derived from wa
dehisco, es9n¢ vestis; on account of its tubular calyx splitting by the
growth of the fruit. This genus will comprise 5 species described by Prof,

SOUTH AMERICAN PLANTS. 17

It should be remarked that the flowers in most (and I be-
lieve in all) species of Acnistus possess a very sweet smell.
1. Acnistus cauliflorus, Schott ;—foliis obovato-oblongis, utrin-

Kunth from the plants brought home from central America by Humboldt
and Bonpland, together with another hitherto undescribed that exists in
the herbarium of Sir Wm. Hooker; they are all trees or large shrubs, with
abundant foliage, growing at great elevations in the vallies of the Andes,
having generally long crimson, or orange coloured flowers of much
beauty, the corolla presenting a 5-lobed border, with 5 small teeth in the
intermediate narrow plicatures, as in Dunalia, and an unequally 5-
toothed calyx, that somewhat enlarges with the fruit, and splits as above
mentioned.

CuanestuEs.—Calyz tubulosus, inequaliter obtuse 5-dentatus, ‘sub 2-lo-
bus, demum parum auctus, lateraliter fissus, persistens. Corolla hypogyna,
nfundibuliformi-tubulosa, subincurvata, lobis 5-acutis, margine floc-
cosis, zstivatione valvato-induplicativis, basi plicatis, dentibus brevibus
interjectis. Stamina 5, subinclusa, filamentis basi adnatis, mox liberis,
gracilibus, erectis, vix exsertis ; antheris oblongis, basifixis, Ovarium
ovatum, 2-loculare. Stylus gracilis, apice incrassatus, exsertus- Stigma
clavato-bilobum. Bacca obovata, calyce hinc fisso inclusa. Semina
numerosa, in pulpo nidulantia, rugosa, reniformia, cetera ignota.

Frutices Andicoli Americe intertropice. Folia alterna, petiolata, Flores

* speciosi, coccinei, vel aurantiaci. Bacca rubra.

1. Chenesthes fuchsioides. Lycium fuchsioides. H. B. K. 3, 52. Pl, Aiquin.
tab. 42. Bot. Mag. tab. 4149. Fruticosa ; foliis obovato-oblongis, obtu-
siusculis, glabris ; umbellis extra-axillaribus, terminalibusque, sessilibus,
multifloribus ; pedicellis glabris, cernuis; calyce 2-lobo, sub 5-dentato,
lobo altero 3-dentato, vel integro; corolla coccinea, glabra, filamentis
basi villosis, dilatatis, inclusis. Quito (in vallem Llow), Hail, n. 7.
Columbia, Jameson. v. s. in Herb. Hooker.

Bonpland describes this to be a shrub 10 or 12 feet high. The leaves
are smaller than most of the other species, being only 2 in. long, and 9 lin.
wide, broader towards the top, and narrowing gradually into a petiole of
6 lin. in length, The calyx is tubular, quite glabrous, about 5 lin. long,
broadly 2-lobed, the one lobe having a single, sometimes two minute pu-
bescent teeth, the other having three minute approximate teeth, which are
downy ; the flowers, according to Bonpland, are of a “ beau rouge,” while
Colonel Hall states them to be “ orange red;’’ they are about 1 inch long,
tubular, glabrous, with a border of five somewhat erect lobes, with a small
tooth in each intermediate plicature ; the filaments are crimson, subulate,
slightly hairy below, inserted near the base of the corolla; the berry is —
pyriform (not globular), and three times the length of that figured by Bon-

D

18 ILLUSTRATIONS OF

que attenuatis, basi cuneatis, in petiolum longum subdecur-
rentibus, integris, utrinque pubescentibus, demum subgla-
bris, subtus pallidis; floribus fasciculatis, confertis, longe

pland, 9 lines long, enclosed by the enlarged calyx, which is cleft to the
base on one side; the seeds are very numerous, but too unripe to discover
the form of the embryo. ‘The plant found by Prof. Jameson in Columbia
is hardly to be distinguished from that of Col. Hall, except that in the lat-
ter, the stamens are somewhat exserted, and the calyx is divided into five
nearly equal segments, being scarcely bilabiate ; but that difference alone
can hardly make it a distinct species. Prof. Jameson says it is found abun-
dantly in the neighbourhood of villages (azogues), where it is used for
fences. The cultivated specimens described by Sir Wm. Hooker (Bot. Mag.
tab. 4149), exhibit larger and broader leaves and larger flowers, but the
calyx is exactly that as above described from Columbia.

2. Cheenesthes umbrosa. Lycium umbrosum. H. B. K. 3, 54. Fruticosa ;
ramulis hirto-pubescentibus ; foliis oblongis, acuminatis, glabriusculis
floralibus ovato-rhomboideis ; floribus umbellato-fasciculatis, lateraliter
extra-axillaribus ; corollis coccineis, tubulosis, hirtellis ; staminibus sub-
inclusis ; stigmate exserto, bilobo.—Nova Granada. Columbia (Hartweg,

“n. 1310). v. s. in Herb. Hooker.

The leaves are 3 in. long and 2¥ in, broad, the petiole being 14in. long ;
the pedicels are 14 in. long, the calyx 5 lin. the corolla 14 in, long; the
crimson filaments are adnate to the base for a length of 3 lines, where
they are downy, thence they are free, tomentous, and dilated below, smooth
and tapering gradually upwards; the style thickens considerably towards
its summit,

3. Chenesthes gesnerioides. Lycium gesnerioides. H. B. K. 3, 53. Fruti-
cosa ; foliis ovatis, oblongisve, acutis, supra fere glabris, infra pulveru-
lentis ; floribus umbellato-congestis ; calyce 5 dentato; corolla auran-
tiaca, pubescenti, filamentis pubescentibus.—Peruvia, Prov, Chachapoyas
(Mathews). v. s. in Herb. Hooker.

In this species the leaves are about the size of C. Suchsioides, the flowers
are in fascicles, with slender pedicels swelling at the summit, 14 in. long
and tomentous; the calyx is short, unequally 5-toothed, 2-lobed, the one
having sometimes 3 teeth, often truncated; the corolla of an orange-red
colour, is covered with soft, dense, yellowish down, and is 1 in. long ;
the anthers are half exserted; the style being somewhat longer, and the
stigma capitate and bilobed. | af
4. Chanesthes Zorensis. Lycium Loxense, H. B. K. 3, 83. Arborea; ramu- —

lis pubescenti-tomentosis ; foliis ovatis, acuminatis, utrinque puberulis ;

- umbellis multifioribus, subaxillaribus et terminalibus, sessilibus ; corollis
flavis ? tubulosis, pubescentibus, limbo 5-partito, lobis brevibus, dentibus

~ minimis interjectis; staminibus inclusis.—Peruvia prope Loxam.

SOUTH AMERICAN PLANTS. 19

pedunculatis; corollis pubescentibus, staminibus breviter

exsertis.—Rio de Janeiro et in Insulis Antillanis.

This species, which is widely disseminated throughout
tropical South America, is considered by Schlechtendahl and
others as identical with the three following; but as it differs in
many respects, I have kept it distinct. The leaves are more

5, Chenesthes cornifolia. Lycium cornifolium, H, B. K. 3, 54. Arborea ;
ramulis canescenti-tomentosis ; foliis subrotundato-oyvatis, subacuminatis
complicatis, supra puberulis, subtus molliter fuscescenti-tomentosis ; flo-
ribus umbellato-fasciculatis, subaxillaribus ; calyce pentagono, ineequali-
ter obtuse 5-dentato ; corollis tubulosis, flavis? hirtellis; staminibus

. inclusis.—Quito.

6. Chenesthes Janceolata, (sp. nov.) Fruticosa ; ramulis cano-vel subferru-
gineo-floccosis ; foliis lanceolatis, acuminatis, supra parce pubescentibus,

_infra pallidioribus, floccoso-tomentosis, petiolo caniculato, tomentoso ;
umbellis brevibus, multifloribus ; calyce urceolato, 5-dentato, mollissime
pubescenti, pilis floccosis ; corolla subcurvata, parce puberula lobis mar-
ginibus floccosis ; antheris lineari-oblongis, subinclusis.—Paramo de
Quindui, Nova Granada (Goudot), v. s. in Herb. Hooker. 70
The leaves are 54 in. long and 14 in. broad, on a petiole 1 in. long, with

many divergent parallel veins. The stalk of the umbel does not exceed
3 lin., the calyx 3 lin., on a pedicel of 9 lin. long ; the corolla seems of a
crimson colour, the tube slightly curved, about 14 in, long and 4 lin. broad ;
the filaments are adnate to the base of the corolla for the length of 6 lin.
where they are tomentous; they then become free, are pubescent below,
slender and glabrous upwards, and of a crimson colour; the anthers are
half exserted.

locuroma, Benth.

With Sir Wm. Hooker’s kind permission I add here a new species of
Mr. Bentham’s beautiful genus Jochroma, in addition to the three species
enumerated in the Bot. Reg. 1845, tab. 20. ait TGISS SHS G Baie
4. lochroma macrocalyx (sp. nov.) 5 Suffruticosa ; foliis rhomboideo-ovatis,

utrinque molliter pubescentibus, subtus -pallidis ; floribus umbellato-

fasciculatis; calyce tubo magno, ventricoso, ore constricto, 5-dentato,
2-lobo, lobo altero 3-dentato ; corolla magna, speciosa, cyanza, hirtella,
staminibus inclusis. —Quito, in vallem Lloe (Hall). v. s. in herb.

Hooker. Loe

This is a very handsome species, with large elliptic leaves, of a some-
what rhomboidal form, broader above the middle, and tapering gradually
to the petiole, which is 1 inch long; they are 54 in, long and 3 in. broad,

be ae

20 ILLUSTRATIONS OF

elliptic, with comparatively longer petioles, the stamens less
exserted, the corolla more slender, with a proportionally longer
tube, which is pubescent, the anthers are apiculated, the style
far exserted, the peduncles glabrous, longer, and more slen-
der. A specimen in Sir William Hooker’s herbarium, from
the Island of St. Vincent’s is not distinguishable from the

Brazilian specimens.*

2. Acnistus arborescens, Schlecht. Linn. 7, 67. Atropa arbo-
rescens, Lin. Lam. Dict. 1. 396. Cestrum cauliflorum, Jacq.
Hort. Schen. 3, 41, tab. 325.—Arbusculus, ramis pubescen-
tibus, vix tomentosis; foliis oblongis, utrinque attenuatis,
breviter petiolatis, integerrimis, supra cano-pubescentibus,
demum glabris, infra albo-tomentosis: floribus umbel-
lato-fasciculatis: corolla tubo brevi, lata; staminibus
longe exsertis.—In Insulis Antillanis. (v. s.in herb, Hook.
sp.in hort. cult.)

This species is sufficiently distinct from the others in the
form of its flowers, which are well shown in Jacquin’s figure,
where the leaves are not represented longer than 3 or 4 inches,
but in Sir W. Hooker’s cultivated specimen they are 9 inches
long, upon a petiole of 1 inch, they are 43 in. broad, quite
smooth on both sides; the pedicels are slender, the corolla
quite glabrous, very short, and broader in proportion than
any other species (4 to 5 lin. in length, 2 lin. in diameter), the

minutely punctulate above, with a few scattered downy hairs on each side ;
the umbels are sessile in flower, pedunculated in fruit ; the calyx is 14in,
long, 5 lines in diameter in the mouth, and swelling in the middle to 9 lin. ;
the corolla is long, tubular, and somewhat curved, 24 to 24in, long, of a
** dark purple colour,” its border is much expanded, its 5 lobes having floc-
cose and ciliato-erose margins, with a broad intervening plicature between
each, which is a character common to the other species, Another feature
seems to belong to all the species of this genus, the upper surface of the
leaves is covered with very minute raised dots. :

A representation of this species is given in the Lond. Journ. Bot, Vol. 4,
Tab, XIII, XIV, f. 1. Corolla ; f. 2. the same laid open; f. 3. Fruit; ali
nat, size.

* This species is shown in Tab, 111, of this work, an explanation of the
generic features being given in the description that accompanies it.

SOUTH AMERICAN PLANTS, 21

calyx is smooth and membranaceous, and the stamens far

exserted.

3. Acnistus Plumieri. Belladonna frutescens, Plum. Amer.
tab. 46, f.1. Arbusculus, ramis glabris; foliis longe lan-
ceolatis undulatis, utrinque attenuatis, breviter petiolatis ;
floribus umbellato-fasciculatis, exalbidis —Antilles.

This species, hitherto confounded with the preceding, dif-
fers in the size and shape of the leaves, and in its flowers.
The leaves are from 10 to 12 in. long, and 3 in. broad, upona
petiole 9 lin, in length ; the corolla is white, with a much longer
and narrower tube ; the calyx is deeply cleft into 5 acute lobes.
4. Acnistus aggregatus. Lycium aggregatum. R. § P., 2, 45,

tab. 182, a. Cestrum campanulatum, Lam. Dict. 1, 688.
Lycium arborescens, Hook. Spreng. Syst. 1, 701. Hook.
Bot. Misc. 2, 232.—Frutex leviter tomentosus, canescens ;
foliis primum fasciculatis, demum sparsis, oblongis, utrin-
que acutis, undulatis, integerrimis, supra subglabris, subtus
incano vel flavido-tomentosis : floribus umbellato-fascicu-
latis, corolla alba, tubo brevi, lobis macula viridi notatis,
staminibus vix exsertis.—Peruvia. v. s. in herb. Hook.

The leaves are from 3 to 6 in. long, and 1} to 24 in. broad,
more elliptic: the calyx and corolla are both pubescent, the
lobes of the former being short and obtuse.

5. Acnistus Guayaquilensis, G. Don. Lycium Guayaquilense,
H. B. K. 3, 50:—foliis elliptico-oblongis, acutis, supra
parce puberulis, subtus molliter cano-tomentosis ; umbellis
extra-axillaribus, sessilibus; calyce pentagono; corolla
alba; staminibus exsertis.—Guayaquil.

In this species the leaves are described by Kunth to be
from 6 to 7 in. long, and 24 to 3 in. wide, on a petiole } in.
long; the flowers are about the size of those of Lycium Euro-
peum, and of a sweet smell, as in the preceding species.

6. Acnistus floribundus, G. Don. Lycium floribundum, H, B. K.
3, 51 :—foliis oblongis, acutis, glabris, infra pubescentibus;
umbellis extra-axillaribus, sessilibus, approximatis ; flori-
bus precedenti majoribus ; corolla alba, extus pubescenti;

92 ILLUSTRATIONS OF

staminibus exsertis.—In Andibus Peruvianis, ad Caxamar-

cam.

The leaves of this species are from 2 to 3 in. long, on a
petiole of 3 to 4 lin., the flowers somewhat smaller than those
of the species last described, are more numerous in each fas-
cicle, of a sweet smell, upon smooth pedicels, which are from
5 to 6 lin, long, and thickening towards the calyx.

7. Acnistus spathulatus, G. Don. Lycium spathulatum, R. & P.
2, 46, tab. 183, a :—glaber, ramis angulatis; foliis obovato-
spathulatis, dispersis, vel fasciculatis, in petiolum longum
decurrentibus ; umbellis subaxillaribus sessilibus; corolle
lobis acutis, pubescentibus.—Huanuco Peruvie. v. s. in
Herb. Hooker. (Mathews, n. 849).

This is described as a shrub, 8 to 10 feet high; the leaves,
including the decurrent petiole, are 5 in. long and 14 in. broad,
the pedicels are thicker above; the flowers are of a violet hue,
and the filaments pubescent at base.

8. Acnistus sideroryloides, G. Don. Atropa sideroxyloides,
Wild. in R. et Sch. 4, 686 :—foliis oblongo-ovatis, utrin-
que acutis, infra pubescentibus ; umbellis lateralibus, ses-
silibus.—Peruvia ad ripas fluv. Magdalene, eS

9, Acnistus ramiflorus, (sp. nov.) ; ramulis pallidis, glabris,
verruculosis; foliis oblongis, utrinque attenuatis; nodis
annotinis floriferis, valde approximatis; floribus plurimis,
umbellato-fasciculatis ; pedicellis gracilibus ; corolla glabra,
lobis margine tomentosis; staminibus styloque 2-lobo ex-
sertis.—In Insula San Vincenti. v. s. in Herd. Hooker.
This is a very distinct species, remarkable for its proximate

floriferous internodes, which are not more than half an inch

apart. The leaves are 83 in. long, and 34 in. broad, on a pe-
tiole 15 in. long; the flowers are numerous in each fascicle,
the pedicels being 6 to 9 lin. long: the corollais smooth, _

10. Acnistus Benthami. Lycium macrophyllum, Benth.—Caule
argenteo, foliis subfasciculatis, oblongis, utrinque euneatis,
supra glabris, subtus incano-puberulis ; floribus in axillis
annotinis fasciculatis; corolle lobis— cilatis; stamini-

SOUTH AMERICAN PLANTS. 93

bus exsertis—Mexico. v. s. in herb. Hooker (Hariweg,

n. 368).

A species nearly approaching A. spathulatus in the size of
its leaves, which are 4 in. long, and 14 in. broad, on a petiole
1 in. long :—the pedicels measure 1 inch.

11. Acnistus ambellatus. Lycium umbellatum. R. & P. 2,45,
tab. 182, 4: foliis. oblongo-lanceolatis, longe petiolatis,
subglabris, subtus pulverulentis; floribus umbellato-fasci-
culatis, pedicellis flori aqualibus, corolla violacea, lobis
acutis, filamentis glabris, inclusis.—Canta Peruvie. v. 8.
in herb. Hooker (Mathews, 1836. Provincia Chachapoyas ;
1840. Caxamarca, n. 3244).

The leaves are 33 in. long, and 1} in. broad, on a petiole of
9 lin.; the pedicels are 14 in. long; the corolla has.a tube
7 lin. and a border 4 lin.in length.

12. Acnistus cestroides. Lycium cestroides. Schlecht. Linn.
7, 70: ramulis pubescentibus, interdum spinescentibus ;
foliis late lanceolatis, utrinque acutis, glabris; floribus um-
bellato-fasciculatis; pedunculis apice incrassatis ; corolla
glabra, lobis brevibus, obtusis, margine tomentosis ; stami-
nibus ineequalibus, inclusis.—Banda Oriental, in San José

ide Uruguay. (Sedlow).

“This is said to’be'a bushy shrub, with flexuose branches,
the younger ones downy, short (some of them almost spiny
at the apex), bearing fascicles of young leaves, and almost
umbellate clusters of ‘flowers; the leaves have short petioles,
measuring altogether 2 to 2} in. long, and 8 to 10 lin. broad,
the petiole being semiterete and pulvinate at base, out of a
short tubercle; when’in flower the peduncles are 3 lines, in
fruit 5 lines long ; the calyx is 13 jin. long, having’ short lobes
with a-subulate tooth at the apex ; the corolla is 8 lin. long;
the berry is globose, red, 3 lin. diameter; the seed with its
embryo agrees with the character of that given in the typical
species.

13. Acnistus “ellipticus, Hook. ‘fil. ined.—Fruticosus ; foliis
ellipticis, utrinque attenuatis et glabris, floribus paucis, fas-

- ciculatis, pedicellis longis ; corolla extus ‘cano-pubescenti,

24 ILLUSTRATIONS OF

lobis brevissimis, staminibus inclusis, medio tubi insertis.

—Insula Galapagos (Darwin).

This is a very distinct species, with smooth elliptic leaves,
34 in. long, 13 in. broad, with a channelled petiole 4 in.
long; the flowers 3 or 4 in a fascicle, grow out of the
cicatrices of the fallen leaves; the peduncle being 14 in. long,
considerably thickened towards the summit; the calyx is

about 3 lin. long, with short, unequal, rounded teeth ; the
corolla is tubular, about 8 lines long.

Species dubia.

14. Acnistus grandiflorus. Lycium grandiflorum. Wild. in R.
et Sch. 4, 689 :—foliis ellipticis ; calycibus 3-lobis.—Ca-
raccas.

HIMERANTHUS.

The Jaborosa runcinata of Jussieu has very ‘properly been
separated from the Jaborosa integrifolia, Comm. and consti-
tuted as a new genus by Prof. Endlicher under the name of
Himeranthus (Gen. Plant. n. 3860) ; but as that distinguished
botanist probably had no opportunity of seeing the plant in
its living state, his character is incomplete in some of its
essential features. I therefore propose the following as its
true limits, and annex the details of two other species that I
met with,

Himerantuvs, Endl. (char. reform.)—Calyz 5-fidus. Corolla
hypogyna campanulato-tubulosa, limbo patenti 5-partito,
estivatione plicato. Stamina 5 nunc fauci subsessiles, nunc
e medio corollz orta, tubo hinc ad calcem lineolis totidem
utringue pilis divergentibus notato; filamenta brevissima
inferne subtenues, apice incrassati in connectivum magnum
gibbum producta; anthere bilobee, dorso affixee, longitudi-
naliter dehiscentes. Ovarium 2-loculare, placentis dissepi-
mento adnatis, multi-ovulatis, Stylus simplex, apice per-
foratus. Stigma clavatum, sub 2-5-lobum. Bacca calyce

SOUTH AMERICAN PLANTS. 95

suffulta 2-locularis. Semina plurima, reniformia. Embryo

intra albumen carnosum sub annularis.

Herbe Bonarienses, glabre inferne radicantes, superne sub ad-
scendentes ; foliis magnis, collo radicali congestis, ovatis,

_ eroso-vel sublyrato-sinuatis, petiolatis ; pedunculis Jaterali-
bus, solitariis, unifloris.

1. Himeranthus runcinatus, Endl. Jaborosa runcinata, Link.
& Otto, Ic. Select. tab. 48, Hook. Bot. Misc. 1, 348 :—foliis
oblongis, lyrato-sinuatis ; pedunculo petiolo vix longiori.—
Bonaria.

I found this plant in 1825, at Arecife, 120 miles westward
of Buenos Ayres; it seems to correspond with the Jaborosa
runcinata of Link and Otto; but as Sir Wm. Hooker justly
observes, it has no upright stem, as figured by these authors,
for the leaves and peduncles proceed at once from the collar
at the summit of the root. The plant seems to propagate
itself by running suckers which at intervals strike root into
the ground, where they form new plants. The leaves vary
from 2 to 3 inches in length, and 14 to 2 in. in breadth; the
petioles being about an inch, and the peduncles about 14 in.
long; at times they are somewhat larger. The calyx is sub-
fleshy, with 5 equal, acute, erect, persistent lobes. The co-
rolla is campanular, somewhat contracted in the mouth, the
border being 5-partite, with oblong acuminate lobes, and a
plicate estivation ; it is of a yellowish white colour, fleshy,
quite glabrous outside, and hairy at base within. The stamens
have very short filaments inserted below the mouth, expand-
ing into a thick fleshy connective, to which the 2-celled anther,
bursting longitudinally, is dorsally attached. The ovarium is
green, globular, half immersed in the fleshy torus ; the style
is simple, white, slightly curved, and is distinctly tubular at
the summit to some depth; the stigma is clavate, and indis-
tinctly 2-lobed. I met with no seed sufficiently advanced to
show any indication of the embryo, the character of which I
have stated on the authority of Endlicher.*

* A drawing of this species is shown in Tab, rv.

26 ILLUSTRATIONS OF

From the above description it will be seen that the genus
Himeranthus is very distinct from the Jaborosa of Jussieu,
founded originally upon the plant discovered by Commerson,
which is the only species known, and which has been very
accurately described and figured by Sir Wm. Hooker, (Bot.
Mag. tab, 3489), from living specimens raised in Glasgow,
from seeds sent home by Mr. Tweedie.*

2. Himeranthus erosus (n. sp.) ;—foliis subintegris, vel eroso-
sinuatis, carnosulis, petiolatis, petiolo crasso suleato; pe-
dunculo petiolo 3-plo-longiori ; corolle lobis integris, acu-
tis—In Provincia Bonariensi.

I found this plant at some considerable distance from Bue-
nos Ayres; the leaves are more than double the length of
the former species, comparatively broader, more entire, the
petiole being about 2} in. long, very fleshy, semiterete, with
slightly decurrent margins, The peduncles are from 4 to 6 in.
long; the flowers are larger, the tube of the corolla not so
much contracted in the mouth, and the segments of the bor-
der larger in proportion; the filaments are nearly half the
length of the tube of the corolla, are fixed about the middle

* ] did not find any specimen of Jaborosa during my journey, but judg-
ing from the excellent figure above mentioned and the dried specimens |
have seen, I subjoin what I consider to be an amended character of this
genus,

Jasorosa, Juss.—Calye 5-fidus, Corolla hypogyna, tubo longissimo infun-
dibuliformi, limbo plicato, 5-partito, laciniis longis, valde acuminatis
Stamina 5, corolle fauci inserta, inclusa, sessilia, filamentis ad tubum
corollz arcte concretis ; antheris dorso insertis, connectivo apiculatis,
longitudinaliter dehiscentibus. Ovarium 2-loculare, placentis dissepi-
mento insertis, multi-ovulatis. Stylus simplex longe exsertus. Stigmata
5, longe linearia, erecta. Bacca calyce suffulta 2-locularis. Semina plu- -
rima, subreniformia. Embryo ignotus.

Herba Bonariensis subacaulis, prostrata, radicans ; foliis integerrimis, oblon
gis, petiolatis ; pedunculis longis, solitariis, 1-floris,

1. Jaborosa integrifolia, Juss. Lam. Dict. 3, 189, tab. 114. Hook. Bot,
Misc. 1. 347. Bot. Mag. tab. 3489,—Bonaria.

In order to exhibit more strikingly the difference between the two genera
I have placed in juxta-position with Himeranthus erosus in Tab. v. 8B.
Sir Wm. Hooker’s representation of Jaborosa integrifolia.

SOUTH AMERICAN PLANTS. 27

above the villous lines described in the last species ; the style

is hollow at the apex for one-sixth of its length, and the stig-

ma, with three expanded, obtuse, and almost obsolete lobes,
is exserted. Specimens of this plant exist in Sir William

Hooker’s herbarium, gathered by Dr. Gillies.*

3. Himeranthus ¢ridentatus (n. sp.) ;—foliis ovatis, angulato-
sinuatis ; pedunculo petiolo 2-plo longiori; corolle lobis
oblongis, obtusis, 3-dentatis.—Bonaria.

This species I found also in the province of Buenos Ayres,
in 1826, the only specimen of which was afterwards much
destroyed, but I preserved the drawing made on the spot.t
The leaves are far more membranaceous and more entire than
the former species. The lobes of the corolla are longer,
broader, more obtuse, and 3-dentate.

DorysTIGMA.

Among the plants that I found in my last rapid journey
over the lofty chain of the Andes, in 1825, were two species,
one of which was also collected about the same time by Doc-
tor Gillies, from whose specimens it was figured and well
described by Sir William Hooker in his Bot. Misc. 1, 347,
tab. 71, under the name of Jaborosa caulescens. The differ-
ence in the stamens and the stigma, the presence of stipular
bracts, a somewhat ascending stem, and a far more rigid and
dry habit, constitute the ground upon which I propose to
separate it from Jaborosa, from which genus it differs far more
strikingly than Himeranthus. The following is offered as its
generic character, its name being derived from Aopy hasta,
and seyya on account of its lance-shaped stigma. —
DorystieMA, (gen. nov.).—Calyx profunde 5-fidus. Corolla

hypogyna, infundibuliformi-tubulosa, intus hirsuta, limbo

plicato, 5-partito. Stamina 5, corolle fauci inserta, inclusa ;
filamenta breves; anther virides, oblonge, 2-lobe, late-

* A representation of this species is given in Tab. v, a.
+ The above species is also figured in Tab. Iv. B.
E 2

- om

28 ILLUSTRATIONS OF

raliter valde compress, incurve, apice acuminate, longitu-
dinaliter dehiscentes, basi affixee. Ovarium 2-loculare, pla-
centis dissepimento adnatis, multi-ovulatis. Stylus simplex,
inclusus. Stigma magnum, crassum, stylo utrinque adna-
tum, acutum, lanceolato-obcordiforme. Bacca calyce suf-
fulta, 2-locularis. Semina plurima reniformia. Eméryo
intra albumen carnosum filiformis, annularis.

Herb Andicole, inferne radicantes, prostrate, vel subadscen-
dentes ; foliis petiolatis, subternis, lyratis, vel pinnatifido-
laciniatis, denticulatis ; pedunculis extra-axillaribus, soli-
tariis, 1-floris, bracteatis.

1. Dorystigma caulescens. Jaborosa caulescens, Hook. Bot.
Misc. loc. cit.;—caulibus plurimis; foliis suboppositis vel
ternis, lyrato-pinnatifidis, spinuloso-dentatis, petiolatis;
floribus 3-4 in quaque axilla; pedunculis brevibus ; bracteis
parvis, subulatis——In Andibus Chilensium.

Sir William Hooker’s above mentioned figure affords
so admirable a representation of this plant, and is accom-
panied by so good a description, that it is needless to
make any further remark, than that the bracts are scarcely
half an inch long, very slender, and subulately acuminate.
Excepting in their relative size, there is little difference in the
flower of this'and the following species; the mouth and seg-
ments of the corolla are densely lined within with tomentum,
the hairs being articulated ; the anthers have a distinct mucro-
nate apex, and the filaments are somewhat longer.*

2. Dorystigma squarrosum, (n. sp.) ; Jaborosa decurrens (Nob.
Trav. Chile, 2, 531) ;—foliis subternis, longe petiolatis, irre-
gulariter pinnatifido-laciniatis, laciniis eroso-denticulatis,
petiolo alato, pedunculo duplo longiori; bracteis longissi-
mis, lineari-spathulatis, pedunculo fere eequalibus.—In An-
dibus Chilensium, altitudine 12,000 ped.

This plant was found by me in January, 1825, in another

= A representation of this plant, with ample sectional details, is given in
» VE. A.

SOUTH AMERICAN PLANTS. 29

and far more elevated portion of the Cordillera, near the sum-
mit of the Cumbre; although exposed to the bleak drying
winds prevalent in that great altitude, it is larger in its general
proportions than the former species, and is remarkable for the
great difference in the length of its bracts. The root is
fusiform, and from its summit arise several stems, which are
somewhat prostrate and ascending. The leaves have a blade
about 3 in. long and 1 in. broad at the widest part, quite
smooth, light green, opaque, and more coriaceous than fleshy
in consistence; the petiole is about 3 in. long, fleshy, round
beneath, flat above, with a somewhat broad decurrent ciliated
margin ; the leaves are generally ternate, and the united bases
of the foot-stalks give a knotty form to the axils, which are
about 4 in. apart. The number of peduncles and bracts gene-
rally correspond with that of the leaves. It may be doubted
whether the slender leaflets seen in the axils should be consi-
dered as bracts or stipules, neither of which organs are usually
met with in the Solanaceous group of plants; but I have
adopted the view of Sir Wm. Hooker, who considers them as
bracts, which is justified by the circumstance of their being
always seen rather within the line of the petioles; they are
linear, slender at base, about 14 in. long, swelling at the ex-
tremity into a spathulate blade, with a long cuspidate point.
The peduncles are round, rather slender, about 1} in. long,
somewhat erect, 1-flowered. The calyx is persistent, swelling
about the torus, somewhat membranaceous above, and divided
into 5 equal, long, tapering, erect segments, furnished with
long articulate pubescence. The corolla is of a lurid cream-
colour below the border, which is white, both externally and
within, where it is covered with woolly tomentum; the tube
is funnel-shaped, rather more than } in. long, the border
being divided into 5 rather acute, expanding lobes, which are
somewhat plicate at base; the anthers are almost sessile, and
fixed by their base below the mouth of the corolla, they are
deeply 2-lobed and laterally compressed, so that they stand
out in a circular ring around the stigma, they are of a lurid

30 ILLUSTRATIONS OF

green, and burst in front somewhat laterally by a longitudinal
fissure, throwing out a yellowish-coloured pollen. The ova-
rium is globular, somewhat flattened at the summit, and
divided into 2, 3, or 5 indistinct lobes. The style is erect,
cylindrical, somewhat enlarging at its extremity. The stigma is
oblong, cordate at base, tapering, obtuse at apex, formed of 2
indistinct flattened adnate lobes, fleshy, green, and shining.
The berry is cernuous, fleshy, and 2-celled ; the seeds are flat-
tened, reniform, subrhomboidal, with a marginal hilum; the

testa is somewhat fleshy and rugous; the embryo is slender,
filiform, almost annular, and imbedded in fleshy albumen ; the

radicle, which points toward the hilum, is very long, straight
at the extremity, and bent above; the cotyledons are short,
slender, semiterete, curved, and terminate near the extremity
of the radicle.*

TRECHONZATES.

Among the few interesting plants gathered in my journey
over the Cordillera in the January, 1825, was one found upon
the eastern descent of the Cumbre, towards Las Cuevas, at an
altitude of 11,500 feet, in a very dry and arid situation, which
circumstance suggested the above generic name, from rpnxo
locus asper, vaérns incola. Specimens of the same plant have
lately been sent to this country by Mr. Bridges; and as it has
not been yet described, I now offer the result of my long
recorded observations, from which it will be seen to possess
characters very distinct from Jaborosa and its congeners.
TRECHONZTES, (gen. nov.).—Calyz 5-partitus. Corolla hypo-

gyna, late campanulata, limbo plicato, 5-partito, laciniis

latis, acutis. Stamina omnino libera, inclusa; filamenta
longa, filiformia, imo tubi adnata; anthere connectivo nullo,
_ dorso affixee, 2-lobee, rotundate, basi divaricatee, longitudi-
naliter dehiscentes. Ovarium rotundum, toro carnoso ortum,
2-loculare, placentis dissepimento adnatis, multi-ovulatis.

* A figure of this species, with full details, is given in Tab, vi. s.

SOUTH AMERICAN PLANTS. 31

Siylus filiformis, subexsertus, apice incrassatus. Stigma
capitatum, lamellis 2 magnis, refléxis, corrugatis, adnatis.
Bacca ignota.

Herbee Andicole, pubescentes, inferne radicantes; caulibus
plurimis, prostratis, vix adscendentibus ; foliis pinnatifido-
laciniatis, dentatis ; pedunculis extra-azillaribus, folio multo
brevioribus; floribus solitariis, vel fasciculatis ; bracteis
linearibus.

1. Trechoneetes aciniata. Jaborosa laciniata (Olim. Nob. Trav.
Chile, 2, 531) ;—caulibus plurimis, brevibus, subadscendenti-
bus ; foliis subcoriaceis, pinnatifido-laciniatis, lobis dentatis
acutis ; floribus subsolitariis ; pedunculis extra-axillaribus,
bractea brevi tenui subulata apice lanata instructis.—In
Andibus altissimis Chilensibus.

This curious plant is evidently allied to the Jaborosa group
of Solanee, but it has an aspect very different from the others,
both in its general pubescence and more lurid hue, as well as
by its broad, duller, campanular flowers, with stamens quite
free and insertedin the base of the corolla. The stem is fusiform,
extending horizontally in the stony soil, from which radiate
several half prostrate branching stems, whose axils are distinct
although approximate; the leaves are sometimes alternate,
sometimes subopposite; the petiole is fleshy, round below,
flattened or caniculate above, with broad decurrent margins,
and about 2 in. long ; the blade is about 2 or 2} in. long, and
1} in. broad at the extreme points ; they are deeply and some-
what pinnately laciniate, the segments being rather narrow,
somewhat parallel and roundish, the margin being sometimes
entire, with a cuspidate apex, but more generally sharply
toothed with intervening sinuosities; they are somewhat erect,
covered with long soft pubescence, the hairs being composed
of several broad articulations; the peduncles are about half
the length of the petioles, and round ; the calyx is divided into
five rather acute segments, very pubescent outside, glabrous
within ; the corolla is broadly campanulate, the tube being
sickly. pubescent on both sides; the lobes are ovate, acute,
terminated at the apex by a somewhat terete woolly spur, the

SZ ILLUSTRATIONS OF

margins being whitish, membranaceous, the more central por-
tion as well as the tube being marked with numerous purplish
brown reticulations, and a few scattered hairs; the filaments
are slender, nearly the length of the tube of the corolla, quite
free to the base, whence they originate from a small adnate.
ring; they are smooth and slightly pubescent at base, erect,
and curving downward toward the apex; the anthers, which
are thus inverted, consist of two nearly globular adnate lobes,
divaricate at base, bursting longitudinally by lateral fissures ;
they are without any sensible connective, and are fixed dor-
sally near the base, on the apex of the filament. The ovarium
is round, pubescent, and partially imbedded in the fleshy
torus, is 2-celled, with many ovules, adnate to a central en-
largement of the dissepiment ; the stigma is clavate, consist-
ing of 2 adnate corrugated lobes. At the time I collected ©
this plant, there was no indication of fruit, nor is any seen in

Mr. Bridges’s specimens, but from analogy it may be assumed

to be a berry.*

2. Trechonetes sativa (n. sp.) ;—caulibus plurimis, confertis,
subadscendentibus, pubescentibus ; foliis ternis, ineequali-
bus, oblongis, angulato-sinuatis, vel pinnatifido-laciniatis,
laciniis angulato-sinuatis, lobis mucrone glanduioso apicu-
latis, utrinque parce pilosulis; floribus plurimis, peduncu-
latis, aggregatis, pubescentibus, bracteis totidem spathulatis,
integris, subglabris.—In Andibus Provincize Tucumanensis.
v. 8. m herb. Hooker, a Tweedio lecta, n. 1253.

This plant, according to Tweedie, is cultivated in Tucuman,
where it is “used as mustard,” from which it may be inferred
that the seeds have a pungent taste, for no part of the dried
plant exhibits any such flavour. It grows in broad patches
about a foot high; the stems are covered with soft hairs, and
the axils are about 3in. apart ; the leaves, 3 in each axil, are
of unequal size, the largest being 7 in. long, including the
rather short decurrent petiole, and 4 in. wide; these are pin-
nately laciniate; the smaller ones are about 4 in. long and

* A representation of this species is given in Tab. vir.

-

SOUTH AMERICAN PLANTS. 33

24 in. broad, and are angularly sinuate. Numerous flowers
are crowded together in a fascicle on one side of each axil, the
peduncles being 9 lin. long, with a glabrous, spathulate bract,
at the base of each, from 2 to 9 lin. long. The calyx is
deeply cleft into 5 acuminate, subulate, membranaceous,
green segments, pilose, 2 lin. long. The corolla is of a dusky
green, less than half the size of that of the former species,
sparsely covered with long, soft pubescence, broadly campanu-
late, submembranaceous, veined, and deeply divided into 5
ovate, acute lobes, with undulated margins, and terminated
by a woolly rostrate apex; the stamens are included; the
filaments, entirely free, slender, glabrous, erect, and re-
curved at the summit, arise from the points of a slender,
adnate, 5-toothed ring in the base of the corolla; the anthers
are ovate, cordate, 2-lobed. The ovarium is round and smooth ;
the style short and thick; the stigma clavate, broad, and 2-
lobed.

PIoNANDRA.

Under this name I propose to found a genus comprising
some Solanaceous small trees and arborescent shrubs with
wide spreading branches, and long racemes of flowers
similar to three species that I found in the Organ Mountains
in 1829 and 1838. The Witheringia diploconos,* figured by
Von Martius in his Nov. Gen. et Sp. vol. 111. p. 76, tab. 229,
evidently belongs to this genus, the characters of which may

be thus defined.

* The genus Witheringia, according to the latest arrangement in the
Repert. Bot. of Walpers, 3.29, contains many (24) heterogeneous species,
and it appears to me that very few of those enumerated, harmonize with
the generic character as established by L’Heritier. In the herbarium of
Sir William Hooker, I can find no plant. corresponding with the typical
in the British Museum where L’Heritier’s original specimens
there are two plants marked Witheringia solanacea, both
the figure and description of the founder

of the genus. In the absence, therefore, of the typical plant, without any
good description of it, or any satisfactory, drawing of its details, without

even the knowledge of the country where the original was obtained, nor

by whom collected, it is difficult to understand the true limits of the genus.
F

species ; and
are deposited,
different, and neither answering to

34 ILLUSTRATIONS OF

PronanpRA (gen. nov.) Calyx parvus, 5-partitus, persistens.
Corolla hypogyna, tubo brevissimo, limbo amplo, 5-partito,
lobis 5 subcarnosis basi inflatis vel lanceolatis tenuioribus,
zestivatione marginibus (fere valvatis) introflexis, interdum
mucroni lineari rostratis. Stamina 5, eequalia, circa stylum
conniventes ; filamenta. breves, erecta, ex annulo plus
minusve carnoso tubo adnato orta, crassa, lata, seepissime

utrinque auriculata, nunc figura sigmoidea recurvata, nunc
rectiora, in connectivum magnum carnosum frequenter

incurvum ‘continua ; anthere 2-loculares, dorso adnate,
lobis oblongis, longitudinaliter dehiscentibus, apice cervice
gibbo, utrinque poro hiante, jugate. Ovarium superum,

- obovato-oblongum, cum stylo articulatum, 2-loculare, pla-
centis lunulatis, utrinque seminiferis, septo _adnatis, multi-
ovulatis. Stylus nunc brevis, crassus, turbinato-infundi-
buliformis, apice cavus, nunc gracilior, longus, subfusiformis,
apice incrassatus. Stigma sub—2-labiatum, intus glandulis
2 instructum. Bacca magna, pulposa, 2-locularis. Semina
numerosa, structura ignota. .

Frutices, vel arbuscule, in America meridionali intertropica
indigene, suborgyales, ramosissimt, ramis dichotomis, foliis
petiolatis. geminis, subintegris plerique cordatis, altero
minori ; racemis extra-axillaribus sepissime in dichotomia
ramulorum ; floribus pedicellatis, secundis, pedicellis articu-
latis, seepe deciduis.

Derivatio ex zwy pinguis, avnp stamen, propter antherarum con-
nectivum magnum.

I propose to separate the different species into two
sections, with curved and straight stamens,

§ I. Ceratostemon. Stamina valde curvata, crassa, apice
10 ie conniventes.

1. Pionandra floribunda (n. sp.) :—foliis 5-nerviis, fere glabris,
cordato-ovatis, apice valde attenuatis, et ciliato-serratis,
limbo integro, supra nitidis, creberrime punctulatis, subtus
glauco-pallidis, sub lente minutissime pubescentibus ;
racemis secundis, folio equalibus, corolle lobis extus apice
rostratis. Serra Organensi, Rio-Janeiro.

SOUTH AMERICAN PLANTS. 35

This is a small tree with very bushy widely spreading
branches, conspicuous for the number of its long pendent
racemes of flowers, that I met with on the ascent of the
Organ Mountains, at an elevation of nearly 3000 feet. The
branches are flexuose, somewhat dichotomous, quite smooth
and glabrous. The leaves are sharply acuminate at the apex,
where they are subserrulated and ciliate; the margin is very
entire and somewhat membranous ; they are about 44 inches
long, exclusive of the petiole, and 24 inches broad, the twin
leaf being about an inch shorter: they are quite smooth and
shining above, their surface being covered with very numerous,
minute, raised points, which are somewhat pellucid when
viewed through the light; below they are quite smooth, of a
pale glaucous green, apparently glabrous, but when examined
by a lens, are seen covered with a very fine pubescence ; they
are somewhat 5-nerved at base, the midrib and nerves being
slender and prominent. The petiole is channelled above,
slightly pubescent, and 14 inch long. The racemes of flowers,
about 3 inches long, generally spring from the stem on one
side a little above the origin of the pair of leaves, and some-
times out of the bifurcation of the branches; the pedicels all
arising from one side of the peduncle, are alternately arranged
in two rows, and articulated at a little distance from their
origin, where the greater number fall off: they are about an
inch long and glabrous. The calyx is small and fleshy,
and does not increase in size with the fruit, its segments
being acute and slightly pubescent. The corolla is of a lurid
greenish white, fleshy, and about an inch in diameter, when
fully expanded; it has a short campanular tube, which is —
distinctly saccate and obtusely 5-angular at the corners
opposite the stamens; the, border is cleft into 5 equal, ovate,
acute segments, having very woolly and slightly inflected
margins, which in the bud present an induplicate cestivation
nearly valvate, each segment has externally at its apex, a long
terete spurlike woolly process ; when the flower is fully open,
the lobes are quite patent: the corolla outside is quite
glabrous, but it is lined within with a short woolly white

F 2

836 ILLUSTRATIONS OF

. tomentum, and near the base of the tube, is seen a semi-
terete fleshy annular staminiferous ring. The stamens are
very large and conspicuous, quite glabrous, and alternate
with the lobes of the corolla; the filaments are united at
their base into a very short tube, which springs from the
annular ring alluded to; they are dilated and fleshy, quickly
expanding beneath the anthers into 2 prominent auricular
lobes, and terminate in a remarkably large fleshy gibbous con-
nective supporting the adnate anthers, which consist of 2
distinct and separate lobes, somewhat divaricate at base, and
united at their summit by a cervix, forming a bottle-necked
apex, in which two distinct pores are seen : the lobes, however,
present a decided longitudinal dehiscence by a central furrow ;
the summits of the anthers all converge round the style. The
ovarium is oval, quite smooth, 2-celled, with a broadly lunu-
late placenta arising from each side of the dissepiment,
around which numerous ovules are closely arranged. The
style is short, thick, scarcely longer than the ovarium, it is
hollow and tubular more than half way down. The stigma
is a cyathiform fleshy cup, obsoletely 2-lobed, terminating
the style, having within its mouth 2 fleshy glands. The
berry grows to a large size; the largest that 1 saw of a
kindred species which was far from mature, was oblong, and
nearly 2 inches in length, in which the seeds were not ripe
enough for examination.*

2. Pionandra pubescens (n. sp.)—Arbuscula, tota pubescens
foliis 5-nerviis, obovatis, basi obtusis, vix cordatis, apice
attenuatis integris, utrinque pubescentibus; racemis secun-
dis, folio brevioribus.—Rio de Janeiro, Serra Organensi.
A species much resembling the former; but altogether

closely pubescent, and of pallid appearance. The leaves are

more obtuse than cordate at the base, the flowers are smaller,
not quite so fleshy, but the stamens are hardly so much
curved. The fruit was, in a very young state, 17 inch long

* A representation of this species with full details is shown in
Tab. vuit.

SOUTH AMERICAN PLANTS. 37

and half an inch in diameter, but without doubt it grows to a

much larger size.

3. Pionandra ciliata (n. sp.) :—pubescens ; foliis geminis,
altero minori, lanceolatis, utrinque attenuatis, vel basi
obtusioribus, seepe ineequilateris, apice caudato-acuminatis
margine denticulatis, dense piloso-ciliatis, supra sparse pi-
losis, subtus tomentoso-pubescentibus, pilis articulatis :
racemis secundis, folio brevioribus: bacca magna,—Rio de

Janeiro, Serra Organensi.

This species is very distinct from the two former, and
approaches more in the form and size of its leaves to the
Solanum capsicoides, Mart. which also belongs to this genus.
I did not find it in flower, but from the resemblance of
its habit to those above described, I imagine it will be
found to belong to this section. The berry, though far from
mature, measures nearly 14 in length and 3 inch in diameter.
4. Pionandra fragrans. Solanum fragrans Hook. Bot.

Mag. tab. 3684.—Arborescens : foliis geminis, 5-nerviis,
inzequalibus, ovatis, majori basi sub-elliptico, altero cor-
dato, integerrimis, glaberrimis, supra lucidis, albo-punctu-
latis, infra argenteis, marginibus paulo incrassatis : racemi
floribus secundis, pedicellis inferioribus maxime elongatis :
corolla late campanulata, profunde 5-partita, staminibus
geniculatis, stigmate dilatato, concavo—Guiana. v. 8. in.
herb. Hooker.

From Sir Wm. Hooker’s excellent figure of this species, it
will be scen how closely it approaches P. floribunda in its
general habit, and in the size of its leaves, one of which only
is here cordate. The racemes are 4 inches long, arising
from the forks of the dichotomous branchlets, the pedicels
are unequal in length and articulate : the calyx is somewhat.
pentagonous, with triangular lobes and ciliate margins: the
segments of the corolla are lanceolate as in P. capsicoides.

5. Pionandra diploconos. Witheringia diploconos Mari.
Gen. et spec. 3.77, tab, 229 :—fruticosa, glabra : foliis sub 5-
nerviis, ovatis, acuminatis, basi subcordatis, integris, utrin-
que glabris; racemis alaribus, folio longioribus.—Rio Janeiro
et Tejuca.

38 ILLUSTRATIONS OF

It is only from the description of Dr. von Martius that
"I am acquainted with this species which certainly approaches
close to my P. floribunda: the leaves, however, are entirely
glabrous, and not punctulate; they are little more than half
their size, and less cordate at base than in that species. The
raceme is much longer than the leaf.. The calyx is denticu-
lated, the stamens are ventricose, but the fleshy connective
is less curved: the filaments are expanded below, but are
not auriculate: it has the fleshy perigynous ring, and the
annular disc, as well as the short tubular broad style and
stigma of P. floribunda. Its flowers are odorous, which I do
not remember to have noticed in the Organ Mountain species,
and the lobes of the corolla want the rostrate apical appen-
dage so conspicuous in that species.

6. Pionandra Gardneri (n. sp.)—fruticosa: foliis geminis,
altero minori, sub 5-nerviis, cordato-ovatis, valde acumi-
natis, utringue pilis longis mollibus articulatis, infraque
pallide tomentosis, margine ciliatis et eroso-denticulatis :
racemis é bifurcatione secundis, folio multo longioribus.—
San Caetano, Prov. Minar. General. Brazilizee.— Gardner,
No. 5041. a iis
A very distinct species remarkable for the length of its

articulate glandular hairs. It forms a shrub about 4 feet

high : its larger leaves are 3} in. long and 2 in. broad on a

petiole 14 in. in length, the smaller leaves are 23 in. long and

1} in. broad on a petiole 2 in. The raceme is 7 in. long : the

calyx is 5-angled with broadish deep segments, very pubes-

cent, and somewhat membranaceous: the corolla is very
pubescent outside, smooth within, its segments being broad

and acute: the stamens and pistillum resemble those of P

floribunda.

7. Pionandra betacea. Solanum betaceum Cav. 6.599, tab
521 :—fruticosa : ramis cauleque succulentis : foliis magnis,
ovatis, acuminatis, cordatis, baseos rotundatis, incumben-
tibus, 5-nerviis, utrinque molliter pubescentibus, margini-
bus undulatis, ciliatis, infra purpurascente-ferrugineis
racemo e bifurcatione pendulo, folio breviori; floribus
secundis—Noya Hispania. ». s. in Herb, Hooker, e

SOUTH AMERICAN PLANTS, 39

plurimis locis relata, nempe-Nova Granada, Goudot. Lima,

in hort. cult. Mc. Lean, Buenos Ayres, in hort. cult. Tweedie,

&e.

According to Cavanilles this is a shrub about 4 feet in
height. In all the specimens I have seen, the larger leaves
measure 9 or 10 in. in length, 5 in. in breadth on a petiole,

13 in. long: the smaller leaves are 53 in. long and 4 in.
broad on a petiole of 14 in.: the racemes are about 7 in. in

length: the corolla of a rosy hue has a short tube, with 5
-oblong segments reflected at tip: the 5 equal stamens are
included, the filaments are short and thick, the auricular lobes
though small and hidden by the anthers are distinct and free :
the anthers are large, curved, and approximate : the berry is
reddish, about the size of a pigeon’s egg, and 2-celled : this is
doubtless the same fruit that I saw in the markets of Lima,
where it is commonly used for cooking in lieu of the ordinary

Tomate, the flavour of which it greatly resembles. Tweedie

remarks that it is used in Buenos Ayres for the same purpose,

but not ordinarily, for I never observed it.

8. Pionandra pendula. Solanum pendulum, R. et P. Flor.
Peruv. 2.39, tab. 174, non Link :—fruticosa : foliis alternis,
obovatis, simplicibus, vel geminis, aut pinnatis, utrinque
pubescentibus, integerrimis, venosissimis, foliolis 2—6 nis,
oblique cordatis, acutissimis, impari majori: racemis depen-
dentibus, furcatis, pedicellis articulatis, sepe deciduis :
fructu magno pyriformi—Peruvia in Muia.

From the figure of Ruiz and Pavon, I do not doubt this
species “belongs to this genus; and although the leayes are
sometimes pinnate, they are often simple and cordate: the
racemes, as in the other species, generally grow out of the
bifurcation of the branchlets, and have many fleshy, secund
flowers, with articulated pedicels: the fruit is also large, 2-
celled, with lunulated placentations. .

9, Pionandra odliqua. Solanum obliquum R. et P. Fl, Per.
9.35, tab. 165, a:—glabra: foliis simplicibus, sub 5-
nerviis, obovatis, oblique cordatis, acutis, supra nitidis,
leviter pubescentibus : racemis extra-axillaribus, recurvis,

40 ILLUSTRATIONS OF

floribus duplici serie secundis, pedicellis articulatis, inferi-

oribus maxime longioribus—Peruvia ad Chinchao.

There can be no doubt from the figure and description
above cited, that this species possess all the essential charac-
ters of Pionandra.

10. Pionandra viridiflora. Solanum viridiflorum R. et P.
Fl. Per. 2.38, tab. 173, 6 :—fruticosa, villosa, caule tereti :
foliis, geminatis, altero minori, sub 5-nerviis, ovatis, acutis,
rotundato-cordatis, baseos lateribus incumbentibus, utrin-
que pilosis, pilis flavis, articulatis, infra pallidioribus ; -
racemo e bifurcatione, folio breviori, floribus secundis,
pedicellis articulatis: fructo magno, ovali—Peruvia (v. s. in
herb. Hooker alabastris nondum maturis.)

A species closely allied to P. pendula, and very near P.
betacea, but with much smaller leaves: the larger ones mea-
sure 7 inches in length, and are 53 in. broad, on a petiole 14
in.: the smaller being 43 in. long, and 3} in. broad, on a
petiole of the same length of the basal lobes 1 in. long: the
calyx is pubescent, 5-angular, and somewhat campanulate,
with 5 short lobes: the corolla is tomentous outside, with
woolly margins: the stamens are somewhat long, with a
thickened fleshy, somewhat scabrid connective: the style is
considerably swollen in the middle, and the stigma cup-
shaped: the berry is of the size of that of P. betacea.

11. Pionandra premnefolia. Solanum premnefolium. Dun.
mss :—tota pubescens: fruticosa, foliis geminis, altero
minori, 5-nerviis, cordato-ovatis: racemis folio multo
longioribus, pedicellis articulatis, deciduis.—Brazilia in
Bahiam (a Luccombe) et Prov. San. Pauli (a Bowie et Cun-
ningham lecta) v. s. in Herb. Mus. Brit.

In habit this species much resembles P. floribunda; but
the leaves and stems are covered with long close hairs: the
larger leaves are 3} to 4 in. long, and 24 to 23 in. broad, on a
petiole 1 in. to 1} in. long: the smaller leaves measure 23, by
2 in. on a petiole 1 in. long: the raceme is 4 in, long: it
probably belongs to this section.

SOUTH AMERICAN PLANTS. 41

§ 2. Eurnystemon. Stamina rectiora.

12. Pionandra capsicoides. Solanum capsicoides Mart. Flora
(BZ) 21. Biebl. 1.78:—suffruticosa ; ramulis ‘pubescenti-
bus; foliis ovato-lanceolatis, lanceolatisve, acuminatis,
interdum geminis, minori obovato, basi obtuso, majoribus
acutiusculis, vel obtusis, inequilateribus, subtus pubes-
centi-mollibus, supra pilis articulatis sparse adspersis;
pedunculis filiformibus, fructiferis deflexis.—Brasilia, Prov.
Rio de Janeiro et Min. Geraes.

I have this species from Cape Frio and Villa Ricca in the
province of Minas Geraes. Its leaves are nearly 6 inches
long and 2 inches broad; the racemes are pubescent and
about as long as the leaf, with secund flowers, the pedicels
being articulated a little above their base. The calyx is
pubescent; but the corolla is glabrous, and of much more
membranaceous structure than any of the former species
with lanceolate segments: the stamens are more slender,
nearly straight ; the filaments are united upon a somewhat
membranous perigynous ring, and although not fleshy, are
dilated and expanded in a bilobed form below the anthers:
the connective is thin, flat, tapering above; the anther cells
are turgid, burst longitudinally, and as in P. floribunda are
constricted near the summit by a collar, which is surmounted
by an emarginate globular apex, that opens by two distinct
pores. The ovarium is smaller, and together with the style
and stigma, is quite glabrous: the style is rather slender,
thickening toward the summit, and is as long as the stamens ;
the stigma, though much smaller, resembles that of the
before-mentioned species.* :
13. Pionandra divaricata. Witheringia divaricata Mart. Nov.

Gen. et Sp. 3.75, tab. 228 :—suffruticosa, tota subtiliter

pubescens : foliis ovato-lanceolatis ovatisve, acuminatis,

basi subrotundatis, racemis e. bifurcatione alaribus, sim-

* A figure of this species with details is given in Tab. Ix.
G

49 ILLUSTRATIONS OF

plicibus, paucifloris, folio superantibus.—Serra do Mar,

Braziliz.

This species very closely resembles the last, but its leaves
are not more than 4 inches long, and | inch broad, generally
much smaller, pubescent on both: sides, and nearly equi-
lateral ; the lobes of the calyx are ciliate, often serrulate ;

the filaments are dilated, and membranaceous below, expand-
ing above in a large inflated connective ; the anther cells are

straight, sub 4-gonous, ventricose and turgid, of a yellow

colour ; the style is longer than the stamens, and with the

stigma intermediate between that of the last species and of P.

fioribunda. It approaches P. ciliata, but the’ leaves are

smaller, broader in proportion, not so much attenuated at the

apex, and less pubescent. The racemes are shorter, 2 to 6

inches long, with 4—6 to 10 flowers, which have the articula-

tion of the pedicels close to the peduncle, so that when they
fall off, the stem scarcely exhibits the persistent bases of the
stalks observed in the other species.

14. Pionandra Tegore. Solanum Tegore Awdl. 212. tab. 84 ;—
frutescens, villosa: foliis inferioribus amplissimis, sinuato-
Pinnatifidis superioribus ovatis, cordatis, acutis ; racemis
secundis, e bifurcatione alaribus.—Guiana. v. s. in Herb.
Mus. Brit. et Herb. Hooker.

The upper leaves are about the size of those of P. pubes-
cens, and in like manner, as well as the stems and peduncle,
are covered with articulate hairs; they are also geminate, un-
equal, and sub 5-nerved. ‘The racemes are very short, and
secund, growing out of the bifurcation of the branches; the
pedicels are also articulated, and deciduous, characters quite
in conformity with all the species above enumerated. Aublet
describes and figures the fleshy ring at the base of the tube
of the corolla, out of which the stamens originate, which he
states to be long; straight, and approximate at the apex; the
corolla is deeply 5-partite; the fruit is a spherical berry,
about the: size of a cherry. A remarkable character is ob-
servable in this plant, in the very large size of the lower
leaves, which are about 1 foot in length, and 8 inches broad,

SOUTH AMERICAN PLANTS. 43

divided into acute lobes, by several deep incisures, sinuate
at base; they are nearly glabrous. I did not observe this
remarkable dissimilarity in the size of the upper and lower
leaves in any of the species I found in the Organ Mountains,
nor has this been noticed by the authors who have described
the other species, if we except P. pendula where some of the
leaves are simple and cordate at the base, while others are
larger, and divided into 3 or 5 segments, as shown in the
figure of Ruiz and Pavon above quoted.

15. Pionandra Hartwegii (n. sp.) ;—fruticosa : foliis geminis,
altero minori, 5-nerviis, cordato ovatis, integris, supérne
lucidis, parce pilosis, infra ferrugineo-pruinoso-punctatis ;
racemis e bifurcatione subscorpioideis, folio multo longiori-
bus.—Columbia (Hartweg, n. 1297) v. 8. m herb. Hooker.
Stems round, dark brown, subpubescent: leaves 4% in.

long, 2% in. broad, on a petiole 17 in. length, the smaller one

2% in. long, 12 in. broad, ona petiole 1 in. long; raceme

pubescent, 8} in. long; pedicels articulated, 1} to 14 in. long;

calyx small, 5-gonous, glabrous, lobes short, obtuse, with a

very small tooth in the apex; corolla with linear, lanceolate

segments, an inch long, obtuse at the apex, smooth with
floccose margins; anthers 2-celled, very long, equal, linear,
erect, somewhat scabrous, opening by an apical pore, and by

a lateral fissure in each cell: style obtuse; stigma clavate.

16. Pionandra coriacea, (n. sp.) :—fruticosa, fohiis geminis,
magnis, glabris, valde coriaceis, cordato-ovatis, 5-nerviis,
altero minori, supra punctulatis, infra pruinosis : racemis
brevibus, scorpioideis; floribus secundis, pedicellis. arti-
culatis, approximatis; corolla tubo brevi, laciniis lanceo-
latis: antheris linearibus, erectis, connectivo crasso.—
Peruvia (Mathews, 1971) v. 8. in herb. Hooker.

The leaves of this species are remarkably thick, opaque and
coriaceous, about 8 inches. long, 5 in. wide, with a petiole
2 in. long, the smaller ones 5 inches long, and 34 in. wide,
the petiole being the same length as the basal lobes.

17. Pionandra Cajanumensis. Solanum Cajanumense, H.B.K.
3.47 :—fruticosa, ramulis hirsutis: foliis solitariis, subro-

G 2

44 ILLUSTRATIONS OF

tundo-ovatis, breviter acuminatis, cordatis, integerrimis,
_hirsuto-pilosis; racemis supra-axillaribus, trifidis, floribus
unilateralibus.—Nova Granada.

The leaves are described as 7 inches long, 54 inches wide.
The stamens included, the filaments short and much dilated,
the anthers opening by pores, according to Kunth, and by
lateral fissures, according to Bonpland. The berry is 2 inches
in length.

Species DuBie.

18. Pionandra crotonifolia. Solanum crotonifolium, H, B. K.
3.30 Dun, Syn. 18:—fruticosa: foliis geminis, oblongis,
angusto-acuminatis, basi rotundatis et ineequalibus, sub-
repandis, supra canescenti-pubescentibus, subtus molliter
albo-tomentosis: racemis supra-axillaribus, seepe bifidis,
floribus unilateralibus.—Nova Granada.

The leaves are 4-5 inches long, 14 to 2 inches broad, with
stellate pubescence; the racemes are short, sometimes bifid,
and the flowers have articulated pedicels.

19. Pionandra Narensis. Solanum Narense, H. B. K. 3.31.
Dun. Syn. 18 :—fruticosa, ramis tomentosis, foliis geminis,
altero minori, ovatis, acutis, basi cordatis et ineequalibus,
supra canescenti—subtus albido—tomentosis, mollibus :
inferioribus sinuato-angulatis ; racemis lateralibus, bifidis ;
floribus unilateralibus.—Nova Granada.

The upper leaves are 5-6 inches long, 24-3 inches broad ;
the lower ones 7-8 inches long, and 5-6 inches broad; all
with stellate pubescence.

20. Pionandra trachyphylla. Solanum trachyphyllum, H. B. K.
3.31. Dun. Syn. 18 :—fruticosa; ramis tomentosis, foliis
geminis, altero-minori, oblongis, acuminatis, basi rotun-
datis, inequalibus, integerrimis, supra scabriusculis, subtus
mollissimis, cano-tomentosis, racemis lateralibus, dicho-
tomis, floribus unilateralibus.—Popayan.

The leaves are 5-6 inches long, and 2} to 23 inches broad,
with stellate pubescence.

SOUTH AMERICAN PLANTS. 45

21. Pionandra flagrans. Solanum flagrans. Tenore, Ann. Sc-
Nat. 13.381 ,—arborescens : foliis geminis utrinque glabris,
oblongo-lanceolatis, acutis, integerrimis, racemis secundi-
floris, extra-foliaceis, pedunculis incrassatis.— Brasilia.

It is probable that under a careful revision of the exten-
sive Linnean genus Solanwm, which is greatly required, many
other species will be found to come within the limits of
Pionandra. tis with much doubt I have placed here the
last four species which are only known from the short des-
criptions quoted; when examined with more attention they
may probably be found to belong to another group, many
species of which I have observed in the splendid herbarium
of Sir Wm. Hooker; I allude to such as approach S. conicum
R. & P. Flor. Peruv. tab. 172, 6. Many of these have simple,
others pinnate leaves, sometimes smooth, often with stellate
tomentum; they have racemes either simple or scorpioid,
often dichotomously branched, either extra-axillary or grow-
ing out of the bifurcation of the branches, they have very
long narrow coriaceous anthers, and a peculiar form of style;
a section of this same group, numerous in species, have their
stems and petioles aculeate, such as S. torvum loc. supra cit.
tab. 175 a. S. lanatum, tab. 174 b, S. incarceratum, tab. 176.
The present arrangement of the genus Solanum comprising
upwards of 500 species, is certainly very defective, the form
of the leaves, offering very unsatisfactory, and uncertain
characters on which to found any subdivisions; far better
elements will be found to exist in the floral characters as-
sisted by the particular habit of the several species; much
therefore may be expected from the distinguished author of
the well-known monograph on Solanum, who has undertaken
the arrangement of the nat. ord. Solanacez, for the forth-
coming volume of the Prodromus of De Candolle.

SorREMA.

The following details of Nolana paradoxa, Lindl., were 7
made as far back as 1823, and notwithstanding several
species of Nolana have already been figured at different

46 ILLUSTRATIONS OF

times, I am not aware that the carpological characters of the
order Nolanacee have hitherto been illustrated. The plant
in question, which I found near the sea-shore at Concon,
the place of my residence in Chile, is now called by Dr.
Lindley, Sorema paradoxa, in a very interesting paper which
he has given on the divisions of this order in the Botanical
Register for Sept. 1844, tab. 46. Although much additional
knowledge has thus been afforded, the real limits of Noda-
nacee are not yet fully defined, and the true place of its
arrangement in the Natural system not yet quite agreed on.
Dr. Lindley, in the last edition of his “ Natural System,”
p- 229, places it near Convolvulacee, with which it accords in
its expanded funnel-shaped plicated corolla. Others have
combined it with Borraginee, with which it agrees in having
a plicated corolla, included stamens, and distinct nuts. Prof.
Endlicher, in his Genera Plantarum, p- 655, following nearly
the views of Dr. Lindley, places it as a sub-order, or rather
as an aberrant group “* Convolvulaceis affinia”’ After a careful
examination of its relations, I venture to suggest for it a
distinct place in the system, at the beginning of the class
Tubuliflore of Endl., immediately following the Borraginee, in
the Nuculifere of that eminent Botanist, so that intermediate
with Convolvulacee, the Nolanacee will thus retain their close
affinity towards Solanacee, for it is especially with Petunia,
&c., that they agree in their convoluted and deeply plicated
corolla with unequal included stamens, and not less with
many others among Solanee in their geminate or fasciculate
leaves and general habit ; and while they also accord in the
annular filiform shape of their embryo, enveloped in albumen,
and in the position of the radicle, they differ from the whole
of that order in the origin and development of their distinct
carpels, for the ovules of Solanacee are invariably attached
to the dissepiment of a 2-celled or imperfectly 4-celled ova-
rium. With Borraginee, on the other hand, they agree in
the gynobasic insertion of their distinct ovaria upon a fleshy
lobed disc, and in their separate nuts, with a single seed in
each cell, perforated at the base, but whether the areolar

SOUTH AMERICAN PLANTS. 47

process, which I have shown to exist in all the Nolanacee,
possesses any direct resemblance, in its nature and origin,
to the salient “strophiole (Celomphala, Schrad.)” that is
seen attached to the perforated nuts of many Borragem, ac-
cording to the descriptions of the late Prof. Spenner (Nees
Gen. Plant, tom. 2, tab. 69—73), it is not now necessary to
determine, it being sufficient for our present purpose to indi-
cate the fact, and to add that they differ from the whole of
that family, in the form and position of their embryo, as well
as in habit and inflorescence. With Convolvulacee, as Prof.
Lindley remarks, many analogies exist, but they differ in
their simple, not imbricate calyx, in their distinct ovaria and
the important character of their embryo. I shall presently
attempt to show that the hitherto anomalous genus Gra-
bowskya, is referrible to a position between Borraginee and
Nolanee, and Dichondree will then form an excellent con-
necting link between Nolanee and Convolvulee, to the former
of which this small group has a very close affinity, on account
of the almost gynobasic origin of its nearly distinct carpels,
and also because its embryo is really cyclical, notwithstanding
that the, cotyledons, at their extremity, are bent back ina
sigmoid form, after having completed more than an entire
helix, somewhat in the manner of the embryo of Convolvulus,
but it is to be remarked, that although the cotyledons of
Dichondra are broader than the radicle, and more foliaceous
than those of Nolanee, they are simply parallel, and have not
their margins crumpled and conduplicated, as in the true
Convolvulee. Doctor Lindley observes (Nat. Syst. 230) that
<if we attend to the embryo, they will stand among Convol-
vulacee, if to the carpels, among Nolanacee : upon the whole
the latter must be accounted of the most importance, and
‘consequently it is with Nolanacee that I arrange them.” I
find on examining the seed of Dichondra repens, that the
utricle falls away from its receptacular attachment, showing
a distinctly round perforate aperture at base, and on the
receptacle are to be seen opposite the opening, two distinct
prominences, corresponding to what, by analogy, may be

48 ILLUSTRATIONS OF

considered as of a similar nature to the areolar processes
observed in Nolanee and Grabowskya. The reasons above
offered will therefore probably justify the position I have
ventured to assign for Nolanacee in the general system, so
that without violating the connexion already established by
the authority of the most distinguished Botanists, between
Dichondree and Convolvulee, they will stand after Borraginee,
and before Convolvulee through the intermedium of Dichon-
dree. .
The new genera of Nolanacee, proposed as before alluded to
by Doctor Lindley, although not so well distinguished by
characters derived from the inflorescence, are nevertheless
well marked by a distinct and peculiar habit, aided by diffe-
rences of structure in the fruit. Most of the species com-
prised under Nolana and Sorema, are succulent prostrate
plants with broad fleshy leaves; in the latter genus, the
leaves are geminate, the inner one petioled, the outer one
sessile, with one of its margins decurrent on the stem. The
species included by Alona, Dolia and Aplocarya, are mostly
erect plants with a shrubby habit, and approximate or fas-
ciculated linear leaves, many of them being densely covered
with tomentum; the flowers of the two last mentioned
genera are proportionately very small, those of the others
presenting large campanulate flowers, resembling those of
Convolvulus. It is worthy of remark, that all the plants
of this order, grow either within reach of the humidity of
the sea, or in inland tracts where the soil is impregnated with
particles of salt or natron. The genus in which the plant
about to be described is placed may be thus defined.
Sorema Lindl.—Calye tubuloso-campanulatus, 5-angulatus,
imo toro adnatus, limbo 5-partito, lobis erectis acuminatis
obtusiusculis, persistens. Corolla hypogyna, infundibuli-
formis, limbo amplo, campanulato, plicato, obsolete 5-lobo,
lobis brevissimis, emarginato-mucronatis. Stamina 5, in-
eequalia, inclusa; filamenta erecta, breves, imo corollz
_ orta, basi pilosa; anthere basifixe, 2-lobe, rotundate.
Discus hypogynus, carnosus, margine libero. Ovaria 20, ad

PLATE IX.

Exhibits a portion of Pionandra (Euthystemon) capsicoides.
Fig. 1, the calyx; Fig. 2, a portion of the corolla, with three
of the stamens, to show their position and mode of origin:
all nat. size; Fig. 3, a segment of the corolla; Fig. 4,
stamens, seen in different positions: all magnified; Fig. 5,
pistillum: nat. size; Fig. 6, style and stigma: magnified ;
Fig. 7, section of the same.

PLATE X.

Shows a portion of Sorema paradowa. Fig. 1, the calyx ;
Fig. 2. the same, partly cut away to show the relative size
and position of the entire pistillum ; Fig. 3, corolla cut open;
Fig. 4, the stamens: all nat. size; Fig. 5, the stamens
magnified, and seen in different positions, before and after
dehiscence; Fig. 6, a portion of the style and stigma ;
Fig. 7, fleshy 5-lobed disc upon the torus (with the calyx
cut away), together with the included ovaria, the style, and
stigma; Fig. 8, vertical section of the same, with the ovaria
removed to show the central gynobase and points of attach-
ment of the ovaria: all more or less magnified ; Fig. 9,
Drupes: nat. size; Fig. 10, the same magnified, with a single
cell in each ; Fig. 11, a drupe, with a nut having three cells ;
Fig. 12, a vertical section of a one-celled nut; Fig. 13, front,
side-view, and section of the operculum that closes the
aperture leading into each cell; Fig. 14, seed removed, seen
in front; Fig. 15, the same viewed edgeways ; Fig. 16, trans-
verse section of the same; Fig. 17, vertical section of ditto,
showing the position of the embryo within the albumen.
Fig. 18, the embryo removed : all magnified.

PLATE XI.

Shows Alibrexia rupicola. Fig. 1, is a flower; Fig. 2, the
same, with the calyx thrown back to show the stipitate base
of corolla, and the line of its circumscissure; Fig. 3, the
corolla cut open; Fig. 4, stamens before and after dehis-
cence; Fig. 5, the transverse section of an anther; Fig. 6,
ovaria upon the gynobase; Fig. 7, calyx, induvial remains of
corolla, disc, and ovaria, seen from above ; Fig. 8, vertical
section of the same, showing the stipitate remains of the
corolla after it falls off, the mode of insertion of the ovules
and style upon the gynobase; Fig. 9, the induvial base of
the corolla; Fig. 10, the style and stigma: all magnified ;
Fig. 11, the ripe drupes, with the persistent calyx : nat. size ;
Fig. 12, the same: magnified; Fig. 13, the same, with the
drupes removed; Fig. 14, a 2-celled nut, seen from its base ;
Fig. 15, a single-celled nut, seen sideways; Fig. 16, a
vertical section of ditto; Fig. 17, the nucleus removed, with
the strophiole; Fig. 18, vertical section of albumen and
embryo; Fig. 19, hairs of the pubescence: all much mag-
nified, .

PLATE XII.

Is a representation of Dolia vermiculata. Fig. 1, the
flower: nat. size; Fig, 2, the same: magnified; Fig. 3, the
calyx: nat. size; Fig. 4, the same: magnified; Fig. 5, the
corolla cut open; Fig. 6, a stamen before and after dehis-
cence; Fig. 7, the entire pistil, with the calyx thrown back
to show the stipitate 5-lobed disc: all magnified ; Fig. 8, the
fruit: nat. size; Fig. 9, a nut seen in three points of view;
Fig. 10, section of nut; Fig. 11, operculum; Fig. 12, seed
removed from nut, with strophiole and operculum attached ;
Fig. 13, albumen; Fig. 14, longitudinal section of the same;
Fig. 15, embryo removed ; Fig. 16, a leaf, seen in front and
edgeways: all magnified.

hs

§
og
.o

r

ea i

Plate ll.

”

Ahbrexia rupicola. :

Plate 12.

J.Miers lithog. Printed by Hullmandel & Walton.

Doha vermiculata.

PLATE XIII.

Shows a drawing of Grabowskya obtusa. Fig. 1, is the
flower; Fig. 2, the corolla: nat size; Fig. 3, the corolla, cut
open and magnified; Fig. 4, the calyx; Fig. 5, the pistillum:
both nat. size; Fig. 6, the same magnified, the stigma being
seen both in front and edgeways; Fig. 7, the stamens: nat,
size; Fig. 8, the same: magnified; Fig. 9, the calyx; Fig. 10,
section of calyx and ovarium: doth magnified; Fig. 11, lon-
gitudinal section of ovarium; Fig. 12, transverse section of
ditto : both more highly magnified ; Fig. 13, the fruit; Fig. 14,
the berry, removed from the persistent calyx; Fig. 15, the
same, with half of the sarcocarp removed ; Fig. 16, the nut:
all nat. size; Fig. 17, the nut, seen both inside and edge-
ways; Fig. 18, transverse and longitudinal sections of ditto :
all magnified; Fig. 19, seed, seen in front and edgeways ;
Fig. 20, the same, with half of the testa removed, showing
the albumen; Fig. 21, longitudinal section of albumen and
embryo; Fig. 22, transverse section of ditto; Mg. 23, em-
bryo extracted: all more highly magnified.

PLATE XIV.

Exhibits Metternichia princeps, Fig. 1, the calyx; Fig. 2,
the corolla, with the upper part removed and thrown back,
and the lower portion of the tube cut open to show the
stamens and mode of their insertion: all nat. size; Fig. 8,
the anthers; Fig. 4, transverse section of the same; Fig. 5,
anther after dehiscence: all magnified; Fig. 6, a grain of
pollen: highly magnified; Fig. 7, longitudinal section of
calyx, and of the lower portion of tube of corolla, with the
enclosed pistil : nat. size; Fig. 8, a section of the ovarium ;
Fig. 9, the same, with the ovules removed, showing the
dissepiment and the points of their insertion upon it; Fig,
10, stigma, seen in two points of view: all magnified ; Fig.
il, capsule, after dehiscence; Mig. 12, transverse section
of the same, showing the dissepiment, with eight seeds in
each cell; Fig. 13, seeds, seen in three points of view;
Fig. 14, transverse section of the same to show its 3-winged
margins; Fig. 15, longitudinal section of the same; Fig. 16,
albumen; Fig. 17, longitudinal section of the same; Fig. 18,
embryo removed, seen in front and edgeways: all nat. size.

PLATE XV.

Is Sessea stipulata. Fig. 1, a flower and bract; Fig. 2,
the calyx ; Fig. 3, a flower, with half of the calyx removed to
show the stipitate base of the corolla: all nat. size; Fig. 4,
corolla cut open: magnified; Fig. 5, a stamen, showing the

_scabrid anther; more magnified ; Fig. 6, pollen grains : highly

magnified; Fig. 7, stellate and stipitate hairs of the pubes-

cence; Fig. 8, section of calyx, and induvial remains of
corolla to show the enclosed stipitate pistil: all magnified ; _
Fig. 9, capsule, with persistent calyx: nat. size; Fig. 10, the

same: magnified; Fig. 11, the same, with the calyx cut open

to show the stipitate capsule ; Fig. 12, the two valves of the

capsule, with the dissepiment separated, seen edgeways ;

Fig. 13, transverse section of capsule; Fig. 14, the dissepi-

ment, with the points of attachment of the seeds; Fig. 15,

seeds; Fig. 16, a transverse section of a seed, showing its

3-winged margins ; Fig. 17, the albumen removed : ali magni-

fied; Fig. 18, a longitudinal section of the same; Fig. 19,

the embryo extracted, seen in front and edgeways: all still

more magnified.

PLATE XVI.

Gives a view of Cestrum Organense. Fig.1, a flower
expanded, with its peculiar bract ; Fig. 2, the calyx; Fig. 3,
the corolla removed, showing the induplicate mode of
its wstivation; Fig. 4, the same, cut open: all nat. size;
Fig. 5, the stamens: majnified; Fig. 6, grains of pollen:
very highly magnified ; Fig. 7, the pistil; Fig. 8, stigma, with
a portion of the style; Fig. 9, a section ‘of the calyx, with
the induvial remain of the corolla, enclosing the ovarium ;
Fig. 10, longitudinal section of the same, showing the stipi-
tate ovarium, and the mode of attachment of its ovules:
all magnified ; Fig. 11, berry, with the persistent calyx ;
Fig. 12, longitudinal section of the berry, with the seeds
enclosed ; Fig. 13, seeds, seen edgeways and on the ventral
face, showing the hilum in the centre; Fig. 14, albumen;
Fig. 15, section of the same, with the enclosed embryo ;

Fig. 16, embryo removed, seen in front and edgeways :
all nat. size,

RRL IO a Rit

Plate 16.

J. Miers lithog.

Printed by Hullmandel & Walton.

*

ganense.

Cestrum ean

~

oo
oe

SOUTH AMERICAN PLANTS. js 49

40 distincta, supra discum pluri-serialiter disposita, 1-ovu-
lata.* Stylus centralis, breviusculus, pentagonus. Stigma
clavatum, 5-lobum. Drupe totidem ; nux angulata, endo-
carpio crasso, textura coriacea, spongiosa, 1-6-locularis,
loculis 1-spermis, basi perforatis, apertura omnino clausa,
operculo a semine demum secedente. Embryo filiformis
intra albumen carnosam cyclicus, cotyledonibus semitere-
tibus, radicula ad hilum spectante.—Herbee Chilenses annue
prostrate carnosule, floribus speciosis Convolvulaceis.

1. Sorema paradoxa, Lindl. Bot. Reg. 1844, tab. 46. No-
lana paradoxa Lindl. (non Hook.) Bot. Reg. tab. 865.—
prostrata, pubescens: foliis geminatis, ovatis, obtusis,
spathulatis, altero subsessili, subdecurrenti; pedunculo
axillari, folii longitudine; corolle limbo amplo, campanu-
lato, ceeruleo, fauce albo.—Concon, Chile.

This plant is so well known, that it needs no particular
description: all that is worthy of mention is the structure of
its fruit, which I do not think has been yet sufficiently detailed.
The peculiarity of this genus as established by Dr. Lindley is,
that its many carpels are all perfected into an equal number of
1-seeded drupes; but I have observed sometimes in the above
species, though it rarely happens, that 2-3 nuts are combined
into one, which is then 2-3 celled, each cell having a single seed.
In habit it approaches the typical genus Nolana. The nut is
smooth, unequally rhomboidal, sharp-angular, of a spongy co-
riaceous consistence, the place of its basal attachment being
marked by a small round cicatrice, being the area of a hard cy-
indrical operculum that closes the channel leading into the
included.cell. The seed which fills the cavity is reniform and
compressed, the testa is yellow, reticulate, and crisp, the inner
integument is a very thin transparent membrane enclosing
the albumen, which is white, hard and fleshy ; the embryo is
white, filiform, bent in a somewhat spiral form, the radicle

® Although the disc is partially adnate, it is not so distinctly stipitate
as in Dolia and Alibrewia, for the lobed margin is quite free, and the
corolla arises below it upon the line of junction of the base of the disc with

the calyx and torus as shown in fig, 7 and 8, Plate 10.
H

Mo. Bot. Garcsa
1S.

50 ILLUSTRATIONS OF

pointing towards the hilum, or aperture of the cell: the

cotyledons, which together are somewhat more slender than

the radicle, are about the same length.* :

var. 8. atriplicifolia :—procumbens, subpubescens 5 foliis
spathulatis, radicalibus majoribus, forma Atriplicis hor-
tensis; calyce campanulato, lobis ovato-lanceolatis, acutis ;
corolle tubo intus flavo, fauce albo, limbo amplo, ceruleo.

Sorema atriplicifolia Lindl. Nolana atriplicifolia D. Don.

Sweet, Fl. Gard. n. ser. tab. 305:—Peruvia? an potius

Chile?

The figure given in the work above cited, exhibits the
cauline leaves to be about 12 in. long, and 1 in. broad, they
are wavy, fleshy, broadly oblong, obtuse, subulate at base,
ona very broad decurrent petiole, one of the margins of
which is continuous along the angle of the stem. This variety
is known only from cultivated specimens, raised from seeds
said to have been obtained from Peru, and [ have not been
able to learn of its existence in any herbarium; but I have
just seen in a living state a cultivated species that corres-
ponds with the figure last quoted, from which it appears to
me only a more luxuriant form of Sorema paradoxa. I
therefore, consider them, to be identical, and the stated
place of origin of the cuitivated variety, (Peru), to have been
mistaken for that of Chile.

2. Sorema litoralis (n. sp.)—herbacea, prostrata; radice fusi-
formi, ramulis plurimis, e collo radiatis 3 foliis radicalibus
majoribus, longe petiolatis, cordato-ovatis, obtusis, cau-
linis geminis, inequalibus, obovatis, obtusis, uno sessili,
altero subspathulato, late petiolato, decurrenti: floribus
solitariis, corolla ampla, cerulea.—Chile, Valparaiso, v.
8, in herb. meo (Mathews i) in herb. Hooker, ( Cuming. n. 627,
Bridges n. 327.)

This plant grows sparingly on a sandy beach, within reach
of the spray of the sea: its tap root descends to a depth of 6
or 8 inches: its branches, springing from the neck, spread
along the sand in all directions; the radical leaves have a
narrow petiole, about an inch long, and barely a line broad,

* A figure of this Species is given in Phare 10.

SOUTH AMERICAN PLANTS. 51

the blade being nearly 10 lines long, and 8 lines broad: the
cauline leaves are only half that size, in unequal pairs, the
outer one being sessile, and decurrent on the stem as in the
preceding species, the younger leaves, peduncle, and calyx
are pubescent, but the older leaves and stem are glabrous :
the calyx is 5-angular, 5 lines long, divided half way down
into 5 triangular erect teeth: the corolla resembles that of

S. paradoxa: the nuts are about 16, with sharp angles, each

1-celled.

3. Sorema acuminata (n. sp.) :—fruticulosa, prostrata ;
caulibus ramosis, angulatis ; foliis geminis, pubescentibus,
lanceolatis, lineari-acuminatis, oblique sessilibus, margine
exteriori decurrenti: floribus axillaribus, solitariis, coeruleis ;
nuculis distinctis 35, parvis, foveolatis.—Chile ad Concep-
cionem : (v. s. in Herb. Hooker, n. 1322.)

The stem is almost 4-angular; the leaves are nearly am-
plexicaul at base, where they are fixed obliquely on the
stem, the lower edge being decurrent ; they taper gradually
upwards, and are linearly acuminate, are about 14 in. long,
and 3 lin. broad at base; the younger ones are pubescent,
but they soon become glabrous; the peduncle is compressed,
6-7 lin. in length, and together with the calyxis covered with
long, soft pubescence; the calycine tube is turbinate, pen-
tagonous, 4 lin. long, with five equal erect, triangular, acumi-
nate lobes of equal length; the corolla is about the size of
that of S. paradowa, with a broad campanulate border of a
blue colour. The nuts are scarcely as large as rape seeds,
black, deeply foveolated with very sharp angles; on one
receptacle I found 35 distinct nuts, all 1-celled.

4. Sorema lanceolata (n. sp.) :—herbacea, prostrata, incano-
pubescens ; caule subangulato ; foliis geminis lanceolatis
semiamplexicaulibus, basi oblique adnatis, hinc decurren-
tibus; floribus in axillis solitariis, speciosis, ceruleis.—
Chile ad Coquimbo. v. s. in herb. Hooker, (Cuming,
n. 856.)

The whole plant is furnished with incanous pubescence ;
the younger leaves, peduncles and stems, are ciliated and

H 2

oa ILLUSTRATIONS OF

covered with very thick articulate hairs; the leaves are
somewhat spathulate, lanceolate, oblique at base, and decur-
rent on the stem, as in the former species, they are 1¥ in.
long, 4-6 lin. broad ; the peduncle is compressed, 13 in. long.
the calyx is campanulate, 5-angled, 6 lin. long; the teeth being
half that length, and lanceolate ; the corolla is 13 in. long, and

much resembles that of S. paradoxa. .

5. Sorema longifolia. Alona longifolia Lindl. loc. cit :—her-
bacea, prostrata; caule crasso; foliis geminis, lineari-
lanceolatis, subspathulatis, alato-petiolatis in caulem hinc
decurrentibus, parce et molliter pubescentibus ; floribus
speciosis solitariis, axillaribus, ceruleis.—Chile ad Co-
quimbo. v. s. in herb. Hooker, (Cuming. n. 887).

This is evidently a succulent prostrate plant, with a fleshy
stem 3 lin. in diameter, and with axils 1 in. apart ; the leaves
are 3 in. long, } in. broad, spathulate and decurrent ; the
pedicel is 14 in. long; the calyx is altogether 9 lin. long, the
linear segments measuring 5 lin.; the corolla is 13 in. long,
with a broad campanular blue border as in S. paradoxa.
Doctor Lindley mentions having found in one receptacle 7
drupes, viz: 1-4-celled, and 6-1-celled, in all ten cells: in
the one I examined I found 4 nuts, each 6-celled, 1-3-celled,
and 8-1-celled, in all 13 nuts, with 35 cells; it is worthy of
remark that all the seeds do not produce a perfect embryo.
6. Sorema linearis (n. sp.) ‘—herbacea, glanduloso-pilosa,

demum subglabra; ramulis angulatis ; foliis linearibus,

obtusis, hine decurrentibus ; floribus solitariis, axillaribus.

—Chile ad Conceptionem. v. s. in herb, Hooker. (Bridges,

n. 1323).

This is probably a procumbent plant; the younger leaves
are covered with dense glandular tomentum 3 they are decur-
rent on the stem as in the preceding species, 11 in, long,
2-24 lin. broad ; the peduncle is 3 to 3 in. long; the calyx is
short, 5-angular, with lanceolate segments, altogether 5 lin.
long, and covered with soft pubescence ; the corolla is of the
same shape, but smaller than that of S. paradoxa, and in the
dried state is of a yellow colour. In one case, I found 2 of

SOUTH AMERICAN PLANTS. 53

the nuts 3-celled, 3-2-celled, 15-1-celled, in all 6 nuts with
27 cells; in another instance I observed 1-4-celled, 1-3-celled,
3-2-celled, and 5-1-celled, in all 10 nuts with 18 cells.

From the above details it may be inferred that as the nuts
differ so constantly in their number, and as in each nut the
number of cells is so uncertain, differing even in the same
plant, this feature can no longer be considered a good generic
character. I have examined the plants above described.
with much attention, and cannot perceive any mark to dis-
tinguish Alona from Sorema, except that in the former the
species are all erect plants with woody stems, and fascicu-
late, terete, or 3-gonous leaves, while those of the latter are
herbaceous, prostrate plants, with geminate broad, fleshy
leaves, which in every case appear decurrent on the stem.
I have had no opportunity of examining more than one out
of the 5 species of Nolana enumerated by Dr. Lindley, and
that has not enabled me to appreciate the distinction between
that genus, Sorema, and Alona. In Nolana the species are all
succulent prostrate plants, mostly with geminate leaves,
which are both. petiolated, and are not decurrent on the
stem, as in Sorema ; in all the 3 genera the flowers closely
resemble each other; in Nolana tenella, Lindl., I found 5
nuts, which were either 1-3-4 or 5-celled; if no difference
then can be detected in the flower or the seeds, habit alone
remains to draw a line of distinction between them, and a
question arises whether habit alone will be considered suffi-
cient to separate these plants into 3 genera. Should they
all verge into Nolana, this genus might then with propriety
be divided into 3 sections :—l. Eunolana, comprising the
5 species alluded to; 2. Sorema, containing the 6 species
above enumerated; and 3. Alona, embracing 8 species, viz :
1. A. celestis, 2. A. rostrata, 3. A. obtusa, 4. A. glandulosa,
5. A. carnosa, and 6. A. baccata of Dr: Lindley, together
with two new species described below. It is to be hoped
that some Botanist, possessing the means of examining these
plants, if possible in the living state, will observe whether
any tangible and constant characters exist between them,
or whether from the similarity of their structure, they should

54 ILLUSTRATIONS OF

all become referrible to Nolana as above suggested ; but in

the mean time it is not unfair to presume, from the indica-

tions alluded to, that some good generic differences may yet
be discovered, when the plants have been more carefully
examined.

7. Alona ericifolia (n. sp.) :—fruticulosa, glanduloso-pube-
scens, ramulis sub-dichotomis; foliis fasciculatis confertis
linearibus, margine revolutis et tunc teretibus 3 floribus
speciosis ceruleis ; calyce tomentoso, tubo 5-gono, lobis
erectis, lineari-acuminatis; corolle limbo amplo campanu-
lato; nucibus paucis, magnis, baccatis, plurilocularibus.—
Chile ad Conceptionem. v. s. in herb. Hooker. (Bridges,
n. 1325).

This is apparently a low-growing suffruticose branching
plant, distinguished by its numerous close fascicles of narrow
linear leaves, which are about 2 in. long. and 2 a lin. wide, .
somewhat broader towards the apex, the margins being
rolled back on the mid-rib, so as to assume a perfectly terete
form, they are covered with dense short, glandular tomen-
tum. The flowers are about the size and shape of those of
Sorema paradoxa. The calyx is funnel-shaped, about 3 inch
long. the acuminate lobes being about one third of its
length, and somewhat curved outwards, as in 4, celestis, and
A. rostrata,

8. Alona microphylla (n. sp.) :—fruticulosa, ramulis tortuo-
sis nodosis ; foliis parvis, fasciculatis, confertis, spathulato-
oblongis, carnosulis, viscidulo-pubescentibus 3 floribus
solitariis, mediocribus, calyce campanulato, ad medium
5-partito, lobis late triangularibus, pubescenti, pilis glan-
dulosis, aliisque articulatis 3 corolla pubescenti, limbo amplo
campanulato, staminibus styloque exsertis.—Chile ad
Conceptionem. ». s. in herb. Hooker. (Bridges, n. 1330),
This is another low growing suffruticose Species, with very

much the habit of some of the small-leaved Lyciums. The

stem and lower portion of the branchlets are tortuous, bare,

and knotty; the leayes are close, about 3 lin. long, Ito 1

lin, broad, spathulate, nerveless, fleshy, and covered with

short, viscid, glandular hairs, The peduncles are ciliate,

SOUTH AMERICAN PLANTS. 55

1 in. long; the tube of the calyx is 2 lines in length, as
well as in diameter, having 5 equal, broad, triangular, erect

lobes, 2 lin. long; the corolla is 1 in. long, broadly campanu-
late, with 5 rounded lobes.

Douia, Lindl.

This genus was proposed by Prof. Lindley for a plant
brought by Mr. Cuming from Chile, and which I obtained
many years ago from Dr. Miller, of H.M.S. Dublin, who
collected it in Concepcion. Although unquestionably be-
longing to Nolanee, it has more the habit and inflorescence
of a Fabiana, from the flower of which it is scarcely distin-
guishable. The following is offered as a more extended
generic character than that given by its distinguished
author.

Doutta Lindl.—Calyz persistens, tubulosus, limbo 5-partito,
lobis lineari-acuminatis, carnosulis, obtusiusculis. Corolla
hypogyna, fere hypocrateriformis, tubo ore ampliato, limbo
ad basin 5-fisso, lobis brevibus rotundatis apice vix mucro-
nulatis. Stamina 5, inequalia, inclusa, rarius exserta ;
filamenta erecta, medio corolle inserta, filiformia; anthere
basifixee, 2-lobe, lobis rotundatis, longitudinaliter dehis-
centibus. Discus hypogynus, carnosus, substipitatus, mar-
gine 5-lobo libero gynobasin cingens. Ovaria 8-10, coadu-
nata, l-ovulata. Stylus centralis, filiformis. Stigma clava-
tum. Drupe totidem, carnose, demum sicce, vernicose ;
nux ovalis, 1-6-locularis, basi operculo clausa. Semen ut
in congeneribus.

Fruticuli Chilenses, erecti, ramosissimi; ramulis brevibus

. flecuosis, interdum cottoneo-floccosis ; foliis fasciculatis,
minimis, spathulatis, carnosulis, pilosis ; floribus parvis,
solitariis, terminalibus, v. axillaribus.

1. Dolia vermiculata Lindl. :—ramis niveo-cottoneis ; foliis
brevissimis, spathulatis, rotundatis, crassis; calycis dentibus
carnosis, sub-recurvis, tubo corolle multo brevioribus.—
Chile ad Conceptionem. (Cuming, n. 893, Bridges, n. 1336.

56 ILLUSTRATIONS OF

This is an erect low-growing shrub, with slender woody
stems, and numerous short flexuose branchlets, which are
densely covered with long white cottony hairs; the leaves are
fasciculated, linear, spathulate with fleshy rounded summits,
pubescent, scarcely 1 line in length ; the flowers are terminal,
solitary, and erect, hardly more than 4 inch long, and 14-2
lines diam.; the calyx is 1} line long; 5 to 8 small drupes
become matured in each calyx, in the specimen I examined I
found 6, of which one was 2-celled, and five were 1-celled,
with a single seed in each cell; the nuts are ovoid, rounded,
somewhat angular, the basal point of attachment being
small, and the opening into each cell marked by a round scar
or areola, as in Sorema.*

2. Doxta salsoloides, Lindl.—ramis calvis junioribus pube
brevissima sparsis ; foliis fasciculatis linearibus fere glabris,
calycis dentibus linearibus, obtusis, subpubescentibus,
tubo corolle fere equalibus.—Chile. (Macrae).

The leaves of this species are 4 lines long, } line broad,
slightly ciliated, or exhibiting under a lens a few scattered
articulated hairs; the peduncle is about the length of the
leaves; the calyx is about 3 lines long, divided half way
down into 5 segments, which are linear, obtuse, fleshy, and
sparsely covered with short pubescence ; the corolla is about
4 lines long.

3. Douta clavata (n. sp.).—omnino calva ; foliis fasciculatis,
carnosulis, lineari-spathulatis, imo pulvinatis ; calycis den-
tibus linearibus, obtusis, tubo corolle dimidio brevioribus ;
staminibus exsertis, filamentis basi sericeis.—Chile ad
Conceptionem. v. 8. in herb. Hooker. (Bridges, n. 1324.)
This species has much the aspect of the last, but the leaves

are broader, spathulate and rounded, 3 lines long, and 1 line

wide, they are quite glabrous and fleshy; the peduncle is of
the — length as the leaves ; the calyx about 2 lines long,
is divided halfway down, its segments being linear, obtuse,

* A figure of this species is shown in Plate 11, together with sectional

SOUTH AMERICAN PLANTS. 57

and somewhat thickened at the apex; the corolla is about
6 lines long, with oblong reflected segments, the stamens
being exserted, the filaments arising about the middle of the
tube, from as many dense velvety tufts, above which they
are glabrous, as is likewise the style; the ten ovaries are
arranged in 2 series on a conical receptacle, the margin of
the surrounding disc being erect and obsoletely lobed.

3. Dolia leptophylla, (n. sp.) :—fruticulosa, tota pubescens,
ramulis teneris fusco-cottoneis foliis fasciculatis, lineari-
teretibus, incurvis; calyce parvo, lobis triangularibus,
attenuatis: corolla cerulea, tubo inferne gracili, superne
ampliato, campanulato, lobis parvis rotundatis; nucibus
rotundatis, nigris, rugosis.—Peruvia. v. s. in herb. Hooker.
Cuming, n. 956.

The above specimen is small, apparently a portion of an
erect low-growing shrubby plant. The leaves are barely
4 an inch long and half a line broad, covered with dense
grey tomentum: the peduncle is scarcely 2 lines long; the .
calyx is small, only 13 to 2 lines in length, with triangular
erect teeth; the corolla, about 8 lines long, is very slender at
base, spreading above in a bell-shaped tube, with a 5-lobed
margin. The seeds are black, shining, rounded, covered with
rugous prominences: on one receptacle I observed 1 of the
nuts to be 6-celled, 1-4-celled, 1-3-celled, 2-1-celled, in
all 5 nuts with 15 cells.

4. Dolia dara (n. sp.)—fruticulosa, tomentosa, ramulis graci-
libus laxis ; foliis sparsis, lineari-spathulatis, acutis, glan-
duloso-pubescentibus ; floribus axillaribus, parvis. Canta?
Peruviee.—v. s. in herb. Hooker.

This specimen, or rather fragment of one, was sent from
Peru by Mathews, where he states it to have been obtained
out ofa collection made by Ruiz and Pavon, then existing in
Lima. It has quite the habit of the species above described,
but the axils are nearly an inch apart, the leaves in pairs,
being $ in. long, and barely a line in width, they are tomen-
tous and fleshy; the flowers are solitary in each axil, the
peduncle is only 1 line, and the calyx, deeply cleft into 5

I

58 ILLUSTRATIONS OF

linear acute segments, is scarcely more than a line in length.
The tube of the corolla, which is slender below, swells con-
siderably above.

ALIBREXIA.

Under this name I propose a new genus for a prostrate
plant that I found growing upon the rocks in the Caleta of
Concon in Chile, in the year 1823, where it was constantly
exposed to the spray of the sea, whence its name, from
adiPpekd, mare madefacio. It differs from Nolana and Sorema
by having 10 carpels supported upon a distinctly stipitate
disc quite free from the calyx, by a more tubular corolla
with a border cleft to the base into 5 very small rounded
reflexed lobes as in Dolia, and by its drupes with rounded
oval nuts umbilicate at base, the perforation not being wholly
filled up by a woody operculum or strophiole-like process,
and by the constant adhesion of this process to the testa.
It differs from Nolana and Alona, in having a somewhat
fleshy corolla, with a small 5-lobed border, not one that is
broad, deep and campanular. From Dolia it is distinguished
by its herbaceous, fleshy and prostrate habit, not being suf-
fruticose with a decidedly woody erect stem: by its stamens
arising from the base of the corolla, not simply fixed in the
middle of the tube: by its calyx being cleft nearly to the

base, by the greater number of its ovaries, by its nuts being
_ quite rounded, and narrowed at base to a slight ring around
the areolar cicatrice, It differs from Aplocarya (a genus
hardly distinct from Dolia) in the want. of the very con-
spicuous, large, cicatrized base of its drupes, to which a
portion of the withered receptacle and dise often remains
attached, in having 10 distinct ovaria, and a more infundi-
buliform and less hypocrateriform corolla, and by its stamens
not being exserted. It differs from all others by its ramified
and stellate, not simply articulate pubescence. As in Dolia
and Aplocarya, the tube of the corolla is quite free, both from
the fleshy dise and the calyx, but in Alibrevia it falls off by
a horizontal line parallel with the disc, leaving it surrounded

SOUTH AMERICAN PLANTS. 59

by a hollow cup. The following is the outline of its generic

character.

ALIBREXIA (gen. nov.)—Calyx persistens, utrinque dense
tomentosus, 5-partitus, lobis lmearibus vel 3-angularibus
obtusiusculis erectis. Corolla hypogyna, inferne tubulosa,
superne tubuloso-campanulata, limbo ad basin 5-partito,
laciniis parvis rotundatis reflexis. Stamina 5, ineequalia,
inclusa ; filamenta imo tubi orta, basi villosa, hinc subulata
graciles, ineequales, tubo corolle breviores: anthere basi-
fixe, oblonga, 4 sulcate, 2-loculares, longitudinaliter de-
hiscentes. Discus hypogynus breviter stipitatus, patellifor-
mis, margine crasso, 5-lobo, 10-crenato, gynobasin centralem
cingens. Ovaria 10, distincta, circa gynobasin conicam
biserialiter aggregata, et angulo interno affixa, !-ovulata.
Stylus centralis, 5-striatus, longitudine staminum. Stigma
clavatum, 5-lobum. Drupe (alteris abortivis vel coalitis)
5-8, distinctee, carnosule, demum sicce; nur ‘ovoideo-
rotundata, subossea, 1-4 locularis, loculis 1-spermis, basi
imminuto 1-4-foveolato, fovea operculo semi-clausa. Semen

- solitarium, reniforme, compressum, testa tenui operculo
persistente affixa. Embryo filiformis, intra albumen carno-
sum cyclicus, cotyledonibus semiteretibus, radicula ad hilum
spectante.

Plante suffruticulose Chilenses, prostrate, succose, in saxis
mare adspersis incole ; caulibus ramosis, basi subligneis ;
ramulis succosis; foliis alternis, sub-confertis, lineari-spathu-
latis, carnosis, velutino-tomentosis, pilis ramoso-articulatis,
vel stellatis ; floribus parvis, avillaribus, pedunculatis.

1. ALIBREXIA rupicola :—prostrata; foliis lineari-spathu-
latis, confertis, tomentosis: floribus solitariis, axillaribus,
pedunculo calyceque utrinque tomentoso, calycis laciniis
linearibus, corolla parce pubescenti, violascenti-albida.—
Chile, ad Concon.

The plant spreads itself in a dense mass upon the surface
of the rock to which it is attached, is frequently washed by

the surf, and constantly exposed to the spray of the sea.
12 |

60 ILLUSTRATIONS OF

The leaves are small, linear, spathulate, with a rounded apex,
fleshy, and covered on both sides with dense, short, dark grey
tomentum : they are about 8 lines long, and barely a line in
width: they are attached to the stem by a small glabrous
pulvinate gland: the peduncle is slender, half an inch long;
the calyx is 2 lines long, cleft nearly to the base, with narrow
linear segments tapering upwards, it is tomentous both
within and without. The corolla is barely half an inch long,
slender within the calyx, it swells above in a somewhat
campanular form, is of a pale lilac colour, rather fleshy,
and slightly pubescent outside; the border is narrow, and
divided to its base into 5 short, rounded, reflected lobes,
having at the apex a minute toothlet: the stamens are
wholly included, and are somewhat unequal in length, the
filaments slender, tapering, and glabrous, arise out of dense
hairy tufts in the base of the corolla. The disc is distinctly
stipitate, cup-shaped, and quite free both from the corolla
and calyx, its border nearly erect, is formed of 5 confluent
lobes, with a margin divided into 10 distinct crenatures ; the
gynobase arises in a conical form within the centre of this
cup, the intermediate cavity being filled by the carpels which
are arranged in 2 series, and are attached by a ventral, and
almost basal point to the gynobase: from the centre of this
arises the style, which is columnar, 5-grooved, glabrous, and
surmounted by a 5-lobed, hollow, clavate stigma. The
drupes are small and fleshy, enclosing an ovoid rounded nut,
which is of a more woody texture than most of its congeners:
this is usually 1-celled, sometimes 2-celled, the base of each
cell being marked by an areolar cicatrix, which is partly
hollow, the bottom of the cavity being filled up by the stro-
phiole-like process that remains attached to the testa of the
included seed. The testa, smooth, somewhat membrana-
ceous, and of reticular texture, encloses the albumen, which
is fleshy, and not very copious: in this is imbedded the fili-
form embryo, whose semiterete cotyledons are bent round in
a nearly circular form, while the terete radicle, which is only
slightly curved, and somewhat thickened towards its extre-

SOUTH AMERICAN PLANTS. 61

mity, terminates at a point close to the attachment of the

process before mentioned.*

2. ALIBREXIA tomentosa. Alona tomentosa, Lindl. Bot. Reg.
1844, sub. tab. 46.—prostrata: foliis lineari-oblongis, spa-
thulatis, confertis, incano-tomentosis; floribus solitariis,
calyce tomento aurantiaco utrinque vestito, laciniis 3-an-
gularibus, erectis, corolla pubescenti, alba.—Chile, Valpa-
raiso.—(v. s. in herb. Hooker. Cuming, n. 121. 241.
Bridges, n. 481. 328.)

This species also grows on maritime rocks, and is distin-
guished by its longer, broader (in proportion to their length)
and more incanous leaves, by its corolla not quite so fleshy
and whiter. The leaves are sometimes 13 lines long, and
nearly 2 lines broad, densely covered with short white branch-
ing almost stellate hairs, the base of the petiole is enlarged,
and adheres to the stem by a concave pulvinate gland, which
is almost glabrous, and much more conspicuous than in the
last mentioned species: the calyx is covered, within and
without, by a dense orange coloured tomentum, and its
lobes are broader and more triangular.

3, AnIBREx1A? revoluta. Nolana revoluta R. & P. 2. 8.
tab. 113. (male depicta). Alona revoluta Lindl. Bot. Mag.

1844, p. 46.—prostrata, incana, stellato-tomentosa, ramulis
plurimis succosis; foliis geminis, lanceolatis, spathulatis,
carnosulis, margine revolutis; floribus solitariis, axillari-
bus, ceruleis.—Peruvia, v. s. in herb. Hooker: locis maritimis

Lurin, Mathews, n. 836—837. Cuming, n. 1068.

This plant grows on the sandy beach at Lurin in the
harbour of Callao, and also in the province of Camana,
whence it was sent to Ruiz and Pavon by Tafalla, together
with the drawing above cited, which affords a very imperfect
representation of the flower: it is quite prostrate, with many
short, slender, radiant branchlets which are woody towards
the base, fleshy towards the extremity: the leaves are about
1 or 14 inch long, and about two lines broad, the peduncle is

* A representation of this plant with full details is shown in Plate 12.

62 ILLUSTRATIONS OF

about 4 in. long; the calyx is campanular about 4 lines long
with 5 equal, 3-angular lobes ; the corolla is tubular below,
somewhat swollen at base around the disc, above it swells
into a somewhat campanular form, with 5 short revolute
lobes, it is about an inch long, of a bluish violet colour,
pubescent and apparently not marked with the radiate
nervures so conspicuous in Nolana and Sorema. The con-
tracted portion of the tube of the corolla is pubescent within,
whence the stamens arise, the filaments are dilated, tapering
upwards, smooth, and unequal, the anthers are oval,: bluish,
and included within the mouth: the style is of equal length.
The disc has a 10-lobed border, and supports 8 to 12 ovaria.
The drupes are fleshy, and vary in size. I found in one case
five nuts, each 3-celled, and seven 1-celled—in all twelve
nuts with 22 cells: in another instance, I observed three
nuts each 4-celled, two 3-celled, one 2-celled, and two
1-celled, in all eight nuts with 22 cells. The whole plant
' is densely covered with short greyish tomentum, the hairs of
which, when magnified, appear sometimes articulate, but
most generally stellate and stipitate, a form of pubescence
peculiar to this genus. I confess, however, that I feel some
hesitation in referring this plant here, as its corolla more
nearly approaches that of Sorema and Alona'in size and
colour, but in its general aspect, peculiar habit, the size and
shape of its leaves densely covered with remarkable tomen-
tum, as well as in the form of its nuts, it greatly resembles
the two preceding species.

GRABOWSKYA.

This genus was founded by Prof. Schlechtendahl (Linn. 7.
72) upon the Ehretia halimifolia of L’Heretier (Stirp. p. 45.
tab. 23). By Linneus and succeeding Botanists, it was as-
signed to Lycium, without doubt on account of the similarity
of its flowers and habit to that genus. Schlechtendahl, for
the same reasons, preserved his new genus Grabowskya
among Solanacee, but Nees von Esenbeck restored it to
Ehretiacee, because of its unilocular 4-celled ovarium, be.

SOUTH AMERICAN PLANTS. 63

coming a nut, a character much at variance with the bilocular
ovarium with its many seeded placentation on the dissepi-
ment, which is the constant attribute of Solanaceae. Dr.
Arnott (Linn. 11. 484) who added 2 new species, supported
the views of Schlechtendahl in assigning it a place among
Solanacee, on account of its curved embryo, a view also
maintained by Doctor Lindley, who figured a species in the
Botanical Register tab. 1985, under the name of G. Boer-
haavifolium, which I have designated under that of G. Lindleyi.
Finally, however, Prof. Endlicher in his Genera Plantarum,
No. 3745 has again placed it in the albuminous section of
Ehretiacea, a disposition that can hardly be supported, when
it is remembered that these have an embryo, either straight,
or but slightly curved, broad foliaceous cotyledons, and a
small superior radicle, and that they all possess moreover
a totally dissimilar habit; Grabowskya, on the other hand,
has a long, slender, filamentous, and cyclical embryo, with
semiterete cotyledons, as long as, and even more slender than
the radicle, which points to the base. My own observations
lead me to differ somewhat from the views of these distin-
guished Botanists, and to consider it, as stated in p. 47, rather
as forming a subtribe of Nolanacee, the reasons for which
will presently be shown.

Having examined both G. duplicata and G, obtusa in my
last journey across the Cordillera in 1825, I offer the follow-
ing as an amended character of this genus.

Gnrapowskya Schlect.—Calye parvus campanulatus, nunc
5-partitus, nunc subinteger, margine mucronibus 5 subu-
latis extus instructus. Corolla hypogyna, infundibulifor-

mis, limbo 5-partito, laciniis patenti-reflexis, zstivatione

imbricatis. Stamina 5, prope corolle basin inserta, exserta,
filamentis gracilibus, basi villosis, antheris ovatis, bilobis,
basi divaricatis, dorso affixis, longitudine dehiscentibus.

Ovaria 2, adnata, e disco carnoso orta, obovata, singulo

2-loculare, ovulis in loculis solitariis, erectis, angulo interno

basali affixis. Stylus simplex. Stigma clavatum, ‘com-
pressum, sub-bilobum. Drupa baccata, calyce parum aucto

64

ILLUSTRATIONS OF

suffulta, 2-pyrena, pyrenis osseis, 2-locularibus, loculis 1-
spermis, basi perforatis. Semina oblongo-obovata, com-
pressa, facie subplana, dorso convexa, testa imo in stro-
phiolam carnosam aucta, apertura basali pertensa. Embryo
filiformis, intra albumen carnosam cyclicus, radicula tereti
ad hilum spectante, subrecta, cotyledonibus semiteretibus,
arcuatis.

Frutices Andicoli vel Bonariensis ramosissimi, spinis axilla-

ribus alternis, Lycii habitu: folia alterna, solitaria, vel
gemina, aut fasciculata, petiolata: flores pedunculati, soli-
tarti vel parce racemosi, aut axillis approximatis paniculam
terminalem simulantes.

. Grabowskya Boerhaavifolia Schlect. loc. cit. Ehretia

halimifolia L’Herit. loc. cit. Lycium Boerhaavifolium,
Linn. (non Lindl.) Lycium heterophyllum Murray, in
Comment. Gott. 6. tab. 2.—foliis alternis, petiolatis, utrin-
que attenuatis, petiolo gracili: panicula corymbosa ter-
minali ex ultimis turionibus, pedicellis imo bracteatis,
bractea parva lineari acuta: calyce 5-partito, laciniis
subulatis, simplicibus, acutis.—Peruvia, v. s. in herb.
Hooker. ou pias |

This character is drawn from the description of L’ Heritier,

which I have compared with a specimen in the herbarium of
Sir W. Hooker.

2. Grabowskya duplicata, Arn. loc. cit : Hook. Bot. Mag. tab.

3841: Ehretia duplicata, Nees ab Esenb.—foliis longe pe-
tiolatis, orbiculari-obovatis, basi cuneatim attenuatis, apice
acute ac breviter acuminatis, calyce campanulato, ore
subintegro, membranaceo, mucronibus 5, subulatis, infra
Marginem extus notato, maturescenti fructu parum aucto,

tune dentibus quasi biserialibus, interiori obtuso, exteriori

subulato crasso, multoties longiori—Esquina de Medrano,

Provincize Cordovensis (a Bonaria 400 m. p. intervallo) mihi
detecta: Bonaria (Gillies et Tweedie).
This species has been very faithfully delineated by Sir W.

Hooker as above cited, and is remarkable for the peculiar
form of its calyx. | ets

SOUTH AMERICAN PLANTS. 65

3. Grabowskya obtusa, Arn. loc. cit: G. Boerhaavifolia
Schlect. Linn. 7-72. Ehretia halimifolia, Nees ab Esenb:
—foliis breviuscule petiolatis, cuneatis, obovato-oblongis,
obtusis, calyce 5-partito, laciniis ovalibus, obtusis.—In
Cordilleris Andium detexi, altitudine 6000 ped. A Men-
doze desertis retulitque Gillies, altit. 2600 ped.—Verna-
cule Una del Tigre.

In the form of its calyx and general appearance, this nearly
approaches the original species. It is a low growing shrub,
with very spinous fiexuose branches almost denuded of
leaves, the stems being round, smooth, and pallid: the
spines generally longer than the internodes are evidently
young abortive branchlets, for they often bear leaves, and
most frequently flowers, sometimes lengthening and becoming
flexuose and prickly: they grow out above the petiolar in-
sertion of each leaf, and there appears on either side of —
every spine, a young branch, bearing copious alternate leaves ;
these mostly soon die away, leaving cicatrices on both sides.
The leaves are alternate, oblong, almost orbicular at the apex,
where there is a slight mucro, they are cuneate at base,
terminating in a slender petiole, entire on the margin, and
of a pallid glaucous green on both sides. The calyx is
tubular, campanulate, somewhat 5-angled at base, the border
is divided into 5 short, obtusely angular, erect, very fleshy
teeth. -The corolla is of a lurid white colour, quite glabrous
outside, the tube is slender at base, gradually swelling above,
the border is divided into 5 rounded, obovate segments,
which overlap in estivation: inside it is smooth at base, but
from one fourth of its length, where the stamens are inserted,
to a little below the mouth, it is covered with white woolly
pubescence. The stamens rise above the mouth half the
length of the tube, the filaments are slender, glabrous above,
but in the lower half within the tube, they are very pubes-
cent: the anthers are divaricate at base, and apiculate at the
summit, The ovarium is small, obovate, green, smooth, 4-
locular, but at a very early period, the existence of 2 distinct
bilocular carpels is manifested, a single erect ovule arising

K

4

66 ILLUSTRATIONS OF

from the base of each cell. The style is erect, simple, some-
what shorter than the stamens. The stigma is clavate, green,
with 2 compressed rugose lips. The fruit is a berry with
very little pulp, inclosing 2 hard obovate nuts, flat within,
rounded outside, each having at the base 2 distinct aper-
tures, which on the inner side extend some way upwards,
outside they are separated by a short spine ; in this aperture
may be seen the strophiole of the seed, by which it receives
its nourishment from the fleshy support of the nut: the
testa is of a dark green hue, oblong, compressed, smooth,
tapering below, exhibiting on the inner flattened side, the
before-mentioned protuberant prolongation of the testa: the
endopleura is a very thin membrane covering a hard fleshy
albumen which encloses the embryo: this is amphitropous
and filiform ; the radicle which points to the base is terete,

a little swollen below ; the cotyledons are incumbent, sharply

curved at their origin, becoming somewhat straight towards

the extremity which closely approaches the end of the
radicle.

4. Grabowskya Lindleyi : G. Boerhaavifolia, Lindl. Bot. Reg.
tab, 1985 :—parce spinosa, frondosa: foliis ovatis,” apice
acutis, basi in petiolum longum cuneatis: floribus paucis,
corolla alba, fauci viridi-venosa, limbo subviolacea.—Rio
Grande, Braziliee meridionalis. (Sellow).

This appears to be a more bushy, and far less spiny
species than any of the others, the foliage seems dense, the
leaves more elliptic, and the purplish flowers few in each
axil, while in the Peruvian species, with which it has been
confounded, the flowers are white, and crowded in almost
terminal corymbs.

Dr. Walpers (Repert. Bot. Syst. 3. 113) adds 2 other
species, but there appears no reason for placing the first
(G. disticha. Meyen.) in this genus, since the fruit is un-
known, and its characters agree quite as well with Lycium.
The other (G. Meyeniana Nees. Atropa spinosa, Meyen.)
is the plant I have described under the name of Lycioplesium
Meyenianum (ante p. 13.) I am not acquainted with the

SOUTH AMERICAN PLANTS. 67

Triguiera of Cav. which may probably belong to this
tribe.

The evidence given above, in regard to the carpological
character of Grabowskya, taken into consideration with what
I have advanced on a former occasion respecting Nolanacee
(p. 47), renders it clear that this genus cannot appertain
either to Borraginee, or to Nolanee, although it is to this
order that it bears the greatest affinity, its position being
manifestly between them; with the tribe Borragee it agrees
in the gynobasic origin of its ovaria, and in having a fruit
with 2 bilocular nuts, and in the adhesion of the style to
the axis of the adnate ovaria, but the form and position of
the embryo in copious albumen, independent of its glabrous
and totally different habit, forbids any positive connexion
with it. On the other hand, the difference between it and
Nolanee is not great: it agrees with them in the form and
position of the embryo enveloped in albumen, in its 2-locular
nuts, which that tribe often possesses, but it differs in the
small and regular number of its ovaria, which in Nolanee
are constantly more numerous, always distinct, and never con-
fluent with the style as in Grabowskya; the estivation of
the corolla is also deeply plicate in the one, and imbricate
in the other; it agrees, however, as before shown, both
with some genera of Borragee, and all Nolanacee, in having
the cells of its nuts perforated at base, through which a stro-
phiole subtends that connects the testa immediately with the
gynobasie disc that supports the ovaria. Upon the whole it
appears to offer the closest alliance to Nolanacee, which order
I therefore propose dividing into 2 distinct subtribes, viz :—

NoLANACE,

1. Grabowskyee. Corolla estivatione imbricata. Ovaria 2,
biloculares, stylo unico centrali adnata. Nuces 2, bilo-
culares, loculis 1-spermis, imo perforatis. Semen basi
strophiolo instructum. Embryo albumine amplo filiformis,
annulari-arcuatus.

2, Nolanee. Corolla estivatione contortu-plicata. Ovaria

K 2

68 >» ILLUSTRATIONS OF

plurima, stylo unico centrali, distincta. Nuces plurime,

1-6 loculares, loculis 1-spermis, imo perforatis. Semen

basi strophiolo vario instructum. Embryo albumine amplo

filiformis, spiralis.

Thus it is seen that Grabowskyee stand in relation to
Nolanee, in the same position that Dichondree do to Convol-
vulee, and although the former, in respect to Solanee, are
placed at the two extreme points of the class Tubulifiore
Endl., it cannot be denied that these Orders offer many ana-
logies common to each other, for Grabowskya has quite the
habit and inflorescence of Lycium, and Nolana is not very
dissimilar in habit from Physalis, and other Solanaceous
plants ; still the carpological characters of the Nolanacee
seems so very distinct, verging evidently towards Ehretiacee,
that there appears to me ample reason for justifying the ar-
rangement above proposed, which also. offers the advantage
of conciliating the very opposite views of our most distin-
guished Botanists in regard to these plants.

METTERNICHIA.

This genus was proposed by Dr. Mikan for a beautiful
shrub, with large white flowers from the neighbourhood of
Rio de Janeiro, and by him dedicated to Prince Metternich,
the liberal patron of the many scientific expeditions that
have so greatly extended our knowledge of the Natural
History of Brazil. It comprises only a single species, of
which a very good figure was published by Dr. Mikan in his
Delect. Flor. et Faun. Braz. 3, tab. 1, but as that work
is extremely scarce, and the details of the seeds not quite
perfect, I offer the results of my own observations.

+ Merrernicuta, Mik.— Calyx campanulatus, tubo inflato,

Ineequaliter 4-5-6-fido, laciniis 2-3 seepe duplo majoribus

-) foltiformibus, persistens. Corolla hypogyna, infundibulifor-
mis, tubo superne amplo, limbi expansi, wstivatione pro-
funde plicati, laciniis 5, brevibus, eequalibus, crenulato-un-
dulatis. . Stamina 5, corolle contractione inserta, inclusa,

SOUTH AMERICAN . PLANTS. 69

2 breviora; filamenta filiformia, basi dilatata, imo pube-
scentes, superne glabra; anthere 9-loculares, basifixe,
longitudinaliter dehiscentes. Ovarium sessile, 2-loculare,
ovulis in loculis utrinque 8, in septi faciem juxta basin ad
nervos 2 parallelos, 2-seriatim dispositis, adscendentibus,
imbricatis, inferioribus exterioribus. Stylus simplex. Stigma
2-lamellatum, laminis, crassiusculis margine revolutis. Cap-
sula coriacea, lignosa, cylindracea, 9-locularis, apice septi-
frage 2-valvis, dissepimento libero, valvis parallelis demum
semi-bifidis. Semina 2-4 in quoque loculo, e dissepimenti
basi adscendentia, longe linearia, utrinque acuminato-at-
tenuata, margine ala membranacea cincta, ventre convexa,
dorso carinata, hilo basali, testa chartacea ad endopleuram
spongiosam adnata. Embryo in axi albuminis carnosi tenui
rectus, cotyledonibus linearibus, radiculam cylindricam in-
feram equantibus. Arbor Brasiliensis frondosa, foliis al-
ternis breviter petiolatis, ellipticis, penninervis, valde reti-
culatis, integerrimis, nitidis, floribus terminalibus, subso-
litariis, vel plurimis subracemose fasciculatis, ebracteatis,
speciosis, albis.

1. Metternichia princeps, Mik. loc. cit. tab. 1. Char. ut.

-supra.—Rio de Janeiro, v. v.

This is a handsome tree, or shrub, with copious foliage of
bright evergreen leaves and large conspicuous flowers. It is
altogether perfectly glabrous. The leaves are about 3-4
inches long, and 1% to 1} in. broad, upon a slender chan-
nelled petiole of 4 lines; they are thin, and of a somewhat
membranaceous texture, the margin being slightly revolute ;
they have a reticulated yenation, and both surfaces under a
strong lens appear minutely punctulated. The calyx is very
thin, membranous, light green, very reticulately nerved, and
unevenly cleft into 4-5-6 obtuse leaf-like segments of various
length ; it is generally half an inch, sometimes nearly 1 in.
jong. The corolla is white, 24 in. long, 1 in. broad in the
mouth, and 2 in. across the border: the tube is slender at
base, for about half an inch, when it suddenly swells into a
long and somewhat campanular form, marked with many

70 ILLUSTRATIONS OF

longitudinal veins: the border is obtusely 5-lobed, and
crenulated on the margin, in estivation it is deeply plicated,
the inflected portion being externally covered with soft down,
which soon entirely disappears; the stamens are the length
of the corolla, the filaments are fixed near the base in the
contraction of the tube, which is the only part that is
pubescent, they are dilated at the base, and ciliated, but are
tapering and glabrous upwards, two of them are somewhat
shorter than the others, the anthers are ovate, 2-lobed, some-
what cordate at base, where they are fixed in the sinus to the
apex of the filament. The pollen grains are globular, obsoletely
3-gonous, or rather having 3 indistinct mammiform projections
at equidistant points, with alternating converging lines.
The ovarium is slightly stipitate, obovate, sericeous, 2-celled,
with 8 ovules in each cell, arising from the base of the disse-
piment, and affixed in 2 series near 2 elevated parallel nerves
that extend from the base to the apex of the dissepiment.
The style is slender, glabrous, as long as the corolla; the
stigma is bilamellated, consisting of two short fleshy lobes,
with reflected margins, the inner faces being glutinous and
subpapillose. The capsule is 14 in. long, 4 lines diameter,
supported by the persistent calyx, the valves are thick and
coriaceous, each being divided half way down the middle;
the dissepiment contracts and becomes free as the seeds
ripen, sometimes only two or four of which are perfected
in each cell, the entire number being seldom matured. The
seeds are linear, winged, with a dorsal keel, which is sometimes
double; the endopleura adheres to the lining of the cell,
which is filled by the albumen: this is cylindrical, 9 lines
long, + line broad, straight, thin, and fleshy, and encloses an
embryo of the same form, with semiterete cotyledons
of similar thickness, and about the length of the inferior
radicle.*

From the above deseription, no doubt can. any longer
exist as to the true position of Metternichia, which has

* A figure of this species, eres a ample sectional details is given in
Plate 14 of this work.

SOUTH AMERICAN PLANTS. 71

hitherto been considered doubtful. It was placed by Dr.
Lindley in Solanee, near Solandra: others have arranged it
in Bignoniacee, and ultimately, on the authority of Fenzl,
Dr. Endlicher has doubtfully assigned it a position in the
subtribe Vestiee, among Solanacee. Its true place is cer-
tainly next to Sessea, with which it agrees in its 2-locular
capsule, with the dissepiment parallel to the valves, which
are both nearly cleft to the base, in its erect seeds, fixed by
their base to the bottom of the dissepiment, and in having a
lengthened straight embryo, with linear cotyledons, in which
respects it differs from Vestia, Cestrum, and their congeners.
As those two genera exhibit such distinct characters, I pro-
pose to arrange them under a separate section, to be called
Metternichiee. On examining many of the Solanaceous
genera, I have found some among the Nicotianee, that
possess very distinct characters, very closely approaching the
Cestrinee, on which account I have ventured to arrange
Fabiana, and Nierembergia, in juxtaposition with them,
placing these two genera, because of their capsular fruit, as a
separate subtribe, under the name of Fabianee : as in Ces-
trinee, they offer, among other features, a bilamellar stigma,
and a decidedly stipitate ovarium. I also propose to remove
Vestia from the Cestrinee, and to place it in Nicotianee,
which subtribe 1 have arranged in the Rectembryee, for they
mostly present an embryo as straight as that of Cestrum,
together with similar small ovoid cotyledons. The attachment
of the seeds in Cestrinee is by a ventral hilum, but in Nicotiana,
and Petunia, the seeds are quite oval, so that it is impossible
to say whether the hilum be ventral or marginal : among the
Curvembryee, on the contrary, the seeds are reniform, and
compressed, and the hilum always marginal. In a general
review of the natural order, Solanacee, I would theretore

propose the following arrangement and division :

SoLANACE.

Sunorpo L on Embryo fere rectus. Stigma sub 2-lamellatum, intus papulosum, Ovarium sepissime stipitatum,
rarius sessile.
Teisus 1. Merrernicate®. Embryo elongatus, omnino rectus, teres. Capsula 2-locularis, valvis semifissis : semina pauca, triquetra,
alata, hilo basali imo dissepimenti affixa.
Embryo cotyledonibus teretibus, radiculam equantibus =. : ee . . 1. Metternichia.
a sf fo fe brevioribus : : ; . waste ae Se Seteea,
Trinus 2. Cestrinzx. Embryo fere rectus, cotyledonibus parvis, ovatis, compressis. Fructus baccatus; semina angulata, hilo
ventrali, dissepimenti medio suspensa, Ovaringh stipitatum.
Bacca 2-locularis, seminibus paucis : ¥ 4 , : ‘ é . 3. Cestrum.
Trisus 3, Fastangea. Embryo fere rectus, subteres, cotyledonibus compressis. Fructus capsularis, 2-locularis, valvis a dissepimento
demum solutis. Ovarium stipitatum. Stigma 2-lamellatum,
Capsula valvis 2 profunde 2-fidis, marginibus suturalibus . ‘ - - ‘ . 4. Nierembergia.
“- »» apice 2-fidis, ie: introflexis . ; ; i . 5. Fabiana.
Trisus 4. Nicotianem. Embryo fere rectus, vel levissime arcuatus, subteres, cotyledonibus brevibus subclavatis. Capsula
2-locularis, valvis a dissepimento demum solutis. Ovarium sessile, plerumque glandula hypogyna suffultum. Stigma

capitato-2-lobum.

Capsula valvis 2 profunde 2-fidis, dehiscentia septifragali . ; : : : . 6. Vestia.
» » integris pe vix septicidali [s . . ; . 7. Petunia.
” ” apice 2-fidis, is profunde septicidali, stamina inclusa, corol, regul, . 8 Nicotiana.
” ” ~ F i i ae » exserta, ,, irreg. . 9, Lehmannia.

Pr pe ae pe pa i a % 2 » inclusa, ,, obliqua. 10. Sairanthus.

Susorpo II, Curvemsryex. Embryo in formam semicircularem, annularem, vel spiralem curvatus. Ovarium sessile vel disco
insidens. Stigma capitato-2-lobum, rarissime in laciniis 5 divisum. Semina plurima, hilo submarginali.

Trisus 5. Datura. Fructus pseudo-4-locularis.

Capsula muricata.. ; 2 : ‘ < ‘ , : F . 11. Datura.
» Glabra : ; : ; oe, ‘ : ‘ P . 12. Dictyocalye.
_ Bacca pulposa ‘ : ; ; ‘ : : “ ; e - 13. Solandra.
Trisus 6. Hyoscyamea. Capsula circumscissa,
Operculo 2-loculari . ; , 4 . d ‘ ° P F . 14. Hyoscyamus.
» 4-valvi, mucronato : j ‘ 3 ; é 5 F - , 15. Anisodus.
» l-loculari™ : ‘ : . 16. Scopolia,

‘Parpus 7. Soranza. Bacca 2-locularis. Semina reniformia, hilo marginali.

ol

JO SNOILVULSNTII

SOUTH AMERICAN PLANTS. 73

Before concluding these remarks, I will offer some further
observations upon the genera constituting the suborder Rec-
tembryee, among which I have noticed a feature hitherto
unobserved, and although not universal among all the genera, it
offers in most cases a very decided character; I allude to
the position of the ovarium upon a well-marked and distinct
columnar support. In Petunia and Nicotiana the ovarium
is sessile upon an annular disc, which surrounds and conceals
this support: in Petunia this disc is 2-lobed, in Nicotiana
it is generally 4-lobed. In Metternichia itis sessile,* without
any glandular appendage. Among those possessing a stipitate
ovarium, Fubiana has 2 oblong erect free glands, fixed on
the column, while in Veséia an annular ring invests the
stipes and supports the ovarium. In Nierembergia, Sessea, and
Cestrum, (including Habrothamnus), the column is simple,
quite free, and generally without any glandular appendage,
or at least one that is not always easily perceptible. In
Sessea, Cestrum, Vestia, Nierembergia, and Petunia, the tube
of the corolla is circumscissile, leaving a persistent cupuli-
form base that, in some cases, nearly incloses the ovarium,
a character I have before pointed out, as existing in some of
the Nolanacee.

I have observed also in the Rectembryee, that the surface
of the testa in the seeds affords a variable character: in
Metternichia and Sessea, it is chartaceous; in Cestrum, mi-
nutely reticulated and favose ; in Vestia, broadly rugose,
with fine transverse strie; in Nierembergia, with polished
prominent ruge ; and in Fabiana, it is nearly smooth, with
almost imperceptible, longitudinal, rugose strie. In Petunia
it is divided into large, equal, hexagonal hollows, separated
by simple ridges, while in Nicotiana these ridges are waving,
crenulate, or even sometimes cristate when viewed by a
powerful lens.

The pollen grains of Metternichia and Sessea are spherical,
with 3 rounded mammiform equidistant points, and 3 inter-
mediate convergent lines; those of Cestrum, Fabiana and

* This, however, is only apparently so, as the column is of the same

thickness as the ovarium.
L

74 ILLUSTRATIONS OF

Nierembergia resemble each other, being in the shape of
a compressed sphere, somewhat 3-gonous, with a mammi-
form projection at each angle; those of Nicotiana and Pe-
tunia are of a rounded oblong cylindrical form, with 3
longitudinal grooves, that of Vestia being similar, only of a
nearly spherical shape.

I have endeavoured to detect some generic distinctions
between Cestrum and Habrothamnus ; but after careful in-
vestigation, I cannot discover any differential characters
either in the flower or the seed: the calyx of the latter
is similar to that of many species of Cestrum; the corolla
is of the same form, its lobes having in like manner a
conduplicate estivation, 7.e. their edges being turned in
on each side for about one sixth of their breadth, and
adhering to the adjoining lobes by their woolly surfaces ; the
Stamens are quite similar, as is likewise the style and 2-
lamellar stigma; the ovarium, in like manner, is supported
upon a short and somewhat glandular column, which when
the corolla falls, is invested by its induviate base. I have
also examined a capsule yet immature, in which, as in
Cestrum, the seeds are few, rounded externally, angular
within, and attached from a ventral hilum to the thickened
placentation of the dissepiment, each by a short ligular
strap, and we have the authority of Schlectendahl (Linnea,
8, 251) to show, that the form of its embryo enveloped
in albumen is straight, with small round flattened co-
tyledons; characters which are all exactly those of Cestrum.
In habit, there is certainly a difference that enables us,
at a glance, to distinguish the one from the other, but the real
amountof differenceis confined to the peculiarkind of articulate
pubescence, and the crimson colour of the calyx and corolla;
but it must be remembered, that several species of Cestrum
from Central America, present a calyx and corolla of a deep
orange colour, bordermg on crimson. If pubescence then
be the only tangible character, that can hardly be sufficient
to maintain Habrothamnus, for in such case, those species of
Cestrum with stellate hairs, especially those that are hardly
distinguishable from Sessea in external appearance and in

SOUTH AMERICAN PLANTS. 75

their flowers, ought also to form a distinct genus, which no
one would venture to propose: upon the whole, therefore,
I would recommend that the genus Habrothamnus be sup-
pressed, and its species arranged as a particular section
of Cestrum under that appellation.*

I may here avail myself of the opportunity of observing
that the genus Laureria of Schlechtendahl, placed by some
authors among the Nicotianee is beyond doubt, identical with
the Juanulloa of Ruiz and Pavon: it ought, therefore to
be placed among the Solanee, where also Marckea of
Richard, and Nectouxia of Kunth, should be arranged, as
they better agree with many of that tribe in habit, and
as nothing is known of the character of their seeds to warrant
their being placed among the Reetembryee.

I have excluded Dunalia from the Cestrinee in the above
distribution, because nothing is known of the nature of its
embryo, and for other reasons before pointed out, (ante
p. 13.)

In external aspect there is a remarkable resemblance
between some species of Petunia, Nierembergia and Salpi-
glossis, so much so, that several able Botanists have referred
such plants to these three different genera; but there exists
in the latter, certain decided characters that cannot be con-
founded with the two former: I do not allude so much to
the didynamous stamens, with a fifth sterile filament, as
to the more important one of its deeply reniform seeds,
and completely annular, and sometimes even spiral embryo,
while in the two former genera the embryo is nearly straight.
But there is also another well-marked difference in the esti-
vation of the corolla, a character of the utmost importance,

* The, three known species of Habrothamnus thus forming a distinct

section of Cestrum would be arranged as follows :

§ HapRoTHAMNUS.

Habrothamnus fasciculatus, Endl.) |

—— tomentosus, Bth.)

__— corymbosus, Endl.)
L 2

Cestrum fasciculatum (
——— Benthami. (
—— Endlicheri. (

76 ILLUSTRATIONS OF

and one that will probably be found to constitute an unerring
line of distinction between the Solanacee and Scrophu-
larinee, orders so closely allied, that our most able
Botanists are yet often undecided as to which of the two,
certain plants ought to be referred. This is a desideratum
of some moment, and one that appears to me worthy
of being pursued: it seems to me possible to reconcile
all the discrepancies that at first view stand in the way
of such a line of demarcation: indeed, there exist many
singular analogies between some plants of the Salpiglossidee,
and other Scrophularineous genera, and some of tribes of
the Solanacee as above defined, which will probably assist
us in this inquiry, but I will reserve to a future opportunity
the exposition of my sentiments on this subject, in order
to have more ample means of examining the characters of
these plants, and of availing myself of the assistance that
will be afforded by the extensive views, and vast collection of
facts, made by the learned author of the monograph of

the Scrophularinee, in the forthcoming volume of the Pro-
dromus.

SESsEA.

In order to confirm what I have before stated respecting
the affinities of Metternichia, 1 will offer a few observations
upon the genus Sessea of Ruiz and Pavon, two species
of which are only recorded, both being well figured by
the authors of the Flora Peruviana; but as I have also
had an opportunity of examining two other undescribed
species existing in the splendid herbarium of Sir William
Hooker, and as the characters of the seed have not been
yet sufficiently well determined, I venture to propose here
an amended definition of the genus and a description of
the four species.

Sessea, R.& P. (Char, emend.)—Calyx tubulosus, 5-den-
tatus. Corolla hypogyna, infundibuliformi-tubulosa, limbi
5-fidi erecto-patenti laciniis ovatis, estivatione condu-
plicatis. Stamina 5, tubi medio corolle inserta, inclusa ;

SOUTH AMERICAN PLANTS. 77

anthere 2-lobee, profunde cordate, basifixe, lobis ovatis,
longitudinaliter dehiscentibus. Ovarium ovatum, breviter
stipitatum, cyatho (corolle reliquo) cireumsessum, 2-locu-
lare, ovulis in loculis utrinque, in septi juxta basin 2-
serialiter dispositis, adscendentibus, imbricatis, inferioribus
exterioribus. Stylus simplex. Stigma inequaliter 2-la-
mellatum, lobo inferiori breviori, subdeflexo, altero erecto.
Capsula coriacea, cylindrica, subarcuata, calyce stipata,
2-locularis, apice septifrage 2-valvis, dissepimento libero,
valvis parallelis, demum semibifidis. Semina pauca, im-
bricata, lineari-oblonga, compressa, margine ala membra-
nacea cincta, ventre convexa, dorso carinata. Embryo in
axi albuminis carnosi tenui rectus, cotyledonibus lineari-
bus, subcompressis, radiculam cylindraceam inferam medio
eequantibus.—Frutices Peruviani ef Nova Granatenses,
foliis alternis, petiolatis, oblongis, pseudo-stipulatis, inte-
gerrimis, infra sepissime tomentosis ; racemis terminalibus,
paniculatis, calyce corollaque sepius stellato-tomentosis.

1. Sessea stipulata, R. & P. Flor. Peruv. 2,9, tab. 115 6.—
Frutex foetidus, orgyalis; foltis lanceolatis, acuminatis,
cordatis, supra glabriusculis, subtus stellato-tomentosis,
ferrugineo-incanis ; pseudo-stipulis magnis, auriculeformi-
bus, gibbosis, deciduis, racemis alaribus terminalibusque,

_ pedicellis corymbosis.— Peruvia, v. s.—in herb. meo.

The branches are glabrous, compressed, and finely striated.
The leaves are about 44 in. long, 2} broad, on a petiole
9 lin. long, channelled and tomentous, the internodes being
about 14 in. apart. The stem of the axile raceme exhibits
sometimes a few small leaves, but it is generally bare for
the length of 3 inches, when it spreads into a close branch-
ing tomentous panicle of flowers ; the terminal peduncles are
shorter, closer, and more branched, forming a thyrsoid
panicle; the pedicels have each a linear, very acute, leaf-like
bract, cuneate at base, and tapering into a long slender
petiole. The calyx is 3 lin. long, tubular, 5-toothed; the
tube of the corolla above the calyx is funnel-shaped, about
7 lin. long, and the border is divided to the mouth into five

78 ILLUSTRATIONS OF

equal oblong segments of the length of 2 lines, with deep
inflected estivation ; both the calyx and corolla are smooth
nside, and are covered on the outside with dense stellate
tomentum. The stamens are included, the filaments rise from
about the middle of the tube, out of as many prominent
longitudinal ridges, which are retrorsely pilose, they are
quite smooth above; the anthers are deeply cordate, 2-lobed,
fixed on the apex of the filament, the lobes being oval
and slightly scabrid. The pollen grains are spherical
3-grooved, with alternating convergent lines. The ovarium
is ovate, very sericeous, surmounted on a distinct glabrous
stipes, which is surrounded by a short tubular cup, being
a persistent portion of the corolla, which here breaks off
by a horizontal line as I have described in Alibrewia, a feature
I find to be apparent in nearly all the species hitherto placed
in the Cestrinee that I have examined. The style is columnar,
erect, suleated, slightly but distinctly scabrid in its entire
length, and is longer than the corolla; the stigma is bilabiate,
the upper lip being erect and longer than the other, which is
somewhat declined. The capsule is curved, cylindrical,
4-grooved, about double the length of the persistent calyx,
smooth, bursting by four nearly equal fissures that extend
half way down. They usually contain two or four seeds
in each cell, imbricate in two series; they are oblong, com-
pressed, winged and keeled, much in the same manner
as in Metternichia, only smaller and broader in proportion ;
the cavity containing the embryo is about 2 lines long, $line
broad, the inner membrane remains attached as a lining
to the cavity, the albumen, which entirely fills the cell,
is thin and fleshy, and encloses an embryo of the same
cylindrical shape ; this is nearly straight, slightly sigmoid in
one direction, and a little curved in the other, the cotyledons
are not broader than the radicle, are semiterete and some-
what compressed, and are about half the length of the
inferior radicle.*

2. Sessea dependens, R. & P., Flor. Peruv. 2,9, tab. 116.—

® This species, with details, is shown in Plate 15.

SOUTH AMERICAN PLANTS. 7

Arbor 5-orgyalis: ramis pendulis: foliis sparsis petiolatis

cordato-oblongis, acutis, integerrimis, subtus valde pul-

verulentis ; pseudo-stipulis nullis: racemis longissimis,

terminalibus, dependentibus, floribus ternis sessilibus.—

Tarma, Peruvie, v. s. in Herb. Mus. Brit. Wi

The specimens collected by Dombey, and Ruiz and Pavon,
existing in the British Museum, greatly resemble the figure
given in the Flora Peruviana: the stem is flexuose, much com-
pressed, and finally glabrous; the leaves are smooth above,
pulverulent beneath, with soft stellate hairs; the capsules
are twice the length of those of the former species, and half
enclosed in the persistent calyx, which is finally quite
glabrous.

3. Sessea vestita. Cestrum vestitum, Hook. Icon. pl. tab.
381.—Frutex, ramulis compressis demum glabris: foliis
oblongis, acutis, basi obtusis, petiolatis, nervis sub 8-jugis,
venisque reticulatis superne impressis, supra glabris, subtus
petiolis ramulis corymbisque tomento incano vel subfulvo
dense vestitis, pilis apice stellatis vel articulatis et plumosis ;
pseudo-stipulis nullis ; corymbis terminalibus densifloris ;
corolle tubo gracili fulvo-tomentoso; ovario tomento
stellato dense vestito; capsula glabra subincurva.—Nova
Granada, v. -s. in Herb. Hooker. Paramo de Quindii,
(Goudot.) Quito, Pichincha et Tambillo, (Jameson, n. 185
et 67.)

This is described as a tree of middle size, growing at
an elevation of about 9000 feet above the level of the
sea. The leaves are from 3} to 4% in. long, 14 to 2 in.
broad, on a petiole 6 lines in length; they are shining above,
quite glabrous, with impressed reticulate veins, which give
the older leaves a very rugous appearance. The flowers,
peduncles and branchlets are covered with very dense
incanous or fulvous down, the hairs of which are stellate
at the apex, but on the under side of the leaves, this
tomentum consists of hairs composed of many short joints,
which are stellate at each articulation, so that they appear
elegantly plumose. The corymbs are much branched and
densely flowered, the whole being covered with close fulvous

80 ILLUSTRATIONS OF

tomentum; the calyx is obconical, truncated, with five promi-
nent nervures, and as many almost obsolete teeth, and is 13
line long; the corolla is 8 lines long, with a very slender
funnel-shaped tube, nearly smooth within, the border consists
of five short erect lobes, which have a conduplicate esti-
vation ; the stamens arise somewhat below the middle of the
tube, are a little geniculate, and slighty pubescent at base,
slender, straight, and glabrous above, and recurved at the
apex ; the anthers are included, with ovate adnate lobes,
which are attached at a dorsal point to the apex of the
filament. The ovarium is fixed on a distinct glabrous
column equal to its own length; it is oval, densely covered
with white stellate hairs, and is somewhat umbilicate at
the apex, out of which rises a long slender glabrous style,
somewhat longer than the stamens; the stigma is 2-lamel-
late. The capsule is cylindrical, 4-grooved, somewhat in-
curved, and half invested by the persistent calyx, is 6 lines
long and 1 line diameter, splits into two valves, which
are 2-fid at the apex ; the seeds are small and winged, as in
the first described species. :

4. Sessea corymbosa, (n. sp.)—Frutex, omnino glaberrimus,
ramulis subcompressis; foliis approximatis, petiolatis,
cuneato-oblongis, supra lucidis, subtus pallidioribus, glan-
dulis minutis creberrimis albis notatis, pinnato-nerviis,
nervis primariis divaricatis, 16-20 jugis cum alteris in-
termediis brevioribus, reticulato-venosis, pseudo-stipulis
nullis: racemis corymbosis terminalibus, floribus sessili-
-bus.—Nova Granada, v. s. in Herb. Hooker, Bogota ad
Barro Blanco, (Goudot), S. corymbosa, MSS.

This species differs from all the preceding in being wholly
free from any pubescence, except externally on the inflected
portion of the lobes of the corolla. The leaves are smooth
above, of a clear dead green colour; the nerves and reticu-
lated veins are prominent ; beneath, they have a somewhat
ferrugineous hue, but under a lens the lower surface is
very closely covered with minute white glandular raised dots ;
their length is 5 inches, their breadth 14 to 2 inches, upon a
petiole } in. long, which is thick and rounded below, some-

SOUTH AMERICAN PLANTS. 81

what flattened above, and of a glaucous hue. The flowers
are sessile, and arranged ina branching few-flowered corymb ;
the calyx is 3 lines long, the corolla 8 lines in length, funnel-
shaped, about 2 lines in diameter, with five rather short,
erect teeth; the filaments, of unequal length, are fixed below
the middle of the tube, have a few recurved hairs at base,
but are glabrous above ; the ovarium, which rises on a short
glandular column, is glabrous, as well as the long, slender,
and slightly incurved style; the stigma is somewhat larger,
2-lipped, and inclined; the capsule is shorter and thicker
than in the preceding species, being half invested by the
persistent calyx, about 6 lines long, 2 lines in diameter, and
somewhat curved; there are two seeds matured in each
cell, which are broadly winged, with a slender longitudinal
keel. The form of the embryo corresponds with that of the
first described species.

CESTRUM.

During my residence in Rio de Janeiro, I found in its
neighbourhood and in the Organ Mountains, several species
of Cestrum; and as I do not find them recorded, I will here
offer a short diagnosis of their characters. In order to explain
the views before proposed, in regard to the distribution of —
the Rectembryee, and as a term of comparison between this
genus and others included in this sub-order, I will first give a
somewhat amended character, in accordance with observa-
tions made upon several species that I examined in the living
state. Mi St
Crstrum, Linn. (char. emendat.)—Calyz parvus, tubulosus,

5-fidus, vel 5-dentatus, persistens. Corolla hypogyna, in-

fundibuliformis, tubo elongato, superne ampliato, limbo
5-partito, patenti vel revoluto, eestivatione conduplicativo.

Stamina 5, medio corolle tubo inserta, inclusa, filamentis

simplicibus, vel basi dente auctis, antheris longitudinaliter

dehiscentibus. Ovarium ovatum, breviter stipitatum, sti-
pite vix glanduloso cyatho (corolle reliquo) circumdato,
M

82 ILLUSTRATIONS OF

2-loculare, dissepimento medio placentifero, ovulis paucis,
spermadermis ligulatis suspensis. Stylus simplex. Stigma
sub-2-lobum, vel concavo-capitatum. Bacca calyce suf-
fulta, 2-locularis, vel abortu 1-locularis. Semina pauca,
compressa, dorso rotundata, facie interna angulata, hilo
ventrali. Embryo in axi albuminis carnosi rectus, coty-
ledonibus parvis, orbiculatis, compressis, radicula tereti
infera.—Frutices Americe tropice, foliis alternis, solitaris
rarius geminis, integerrimis, interdum pseudo-stipulatis ;
floribus racemosis ; racemis bracteatis, axillarihus, elongatis,
vel in corymbum, spicam, aut fasciculam abbreviatis ; floribus
sepe suaveolentibus, corollis luteis, rarius viridescenti-albi-
dis, aurantiacis vel rubris; baccis nigris aut violascen-
tibus,

The following are the new species above alluded to.

1. Cestrum Organense (n. sp.)—glaberrimum, ramulis tereti-
bus ; foliis eilipticis, utrinque acuminatis, submembranaceis,
in petiolum basi tumidum subdecurrentibus; racemis axil-
laribus, floribus subcapitatis, bracteis foliosis magnis oblon-
gis sessilibus, corolle fere longitudine; staminibus insertione
barbatis, edentulis—Rio de Janeiro, Serra Organensi,
v. Uv. erg 1S
This species approaches Cestrum bracteatum, (Link et

Otto, Pl. Select. 1. tab. 6. Bot. Mag. 2974), but its leaves

are neither lanceolate nor pubescent, nor have they any

pseudo-stipules: they are 5 in. long, 13 broad, on a petiole

6 to 8 lines in length. The peduncles of the racemes are

from 2 to 3 in. long: the bracts are 6 lines long, 4 lines broad,

with ciliate margins; the calyx is glabrous, tubular, 1 line
long, with 5 short, erect, ciliolate teeth : the corolla with a very

slender tube, swollen below the mouth, is 9 lines long, with a

rotate border, the margins of the lobes externally are tomen-

tous for one-third of their breadth on each side, which are
turned in during estivation, so that they adhere together by
these broad woolly surfaces: the included stamens. are in-
serted above the middle of the tube, and arise out of as many
tufts of spreading hairs, the anthers are rounded, 2-lobed,

SOUTH AMERICAN PLANTS. 83

and somewhat scabrid: the pollen is compressed, 3-gonous,
with a small rounded lobe at each angle, from which 3 con-
vergent lines meet in the centre: the ovarium is supported
upon a short, distinct, but somewhat slender column, which
in many species is somewhat glandular, and this is sur-
rounded by the persistent base of the corolla, which inva-
riably breaks off as it falls, by a horizontal line, leaving a
cylindrical cup: the ovarium is divided into 2 cells by a dis-
sepiment, which presents in its centre a thickened placenta-
tion, and upon this are suspended, by a short ligular strap,
upon their ventral faces, a small number of ovaria in each cell :
the style is filamentous, glabrous, and of the length of the
tube of the corolla: the stigma is capitate, being formed of
2 lamellar lobes, whose inner surfaces are covered with a
thick green glutinous exudation: the berry is oval, scarcely
fleshy, and each cell presents 4 ripened seeds: these are
somewhat obovate, rounded on the outside, and angular on
the inner face, a little above the centre of which is the hilum,
from which the seed is suspended by its short ligular strap:
the testa is of a dark green colour, of very soft texture when
fully ripe: the albumen is soft and fleshy, in the centre of
which is seen the erect embryo, which is white, and as in
most other Cestrinee, lays longitudinally across the hilum,
the cotyledons are small, flat, and rounded, two-thirds of the
length and twice the breadth of the terete radicle, which
points towards the bottom of the seed.*

2. Cestrum Janceolatum (n. sp.)—ramulis teretibus virgatis,
foliis elongato-lanceolatis, acuminatis; margine subundu-
latis, revolutis, basi rotundatis, apice mucronatis, utrin-
que glabris, breviter petiolatis; racemis axillaribus, brevi-
bus, paucifloris.—Rio de Janeiro, Serra Organensi. v. v.
The leaves are 4 in. long, 8 lines broad, ona caniculate

petiole, 3 lines in length: they have no pseudo-stipules. The

racemes are 9 lines long, which I found only in seed: the
* A representation of this species, with detailed sections, is shown in

Plate 16.
mM 2

84 ILLUSTRATIONS OF

berry is oblong, contracted below, supported on a calyx appa-

rently but little enlarged, and contains 2 or 3 seeds in each

cell; the seeds correspond with those of the last species.

3. Cestrum collinum (n. sp.)—ramulis compressis junioribus
angulatis, tomentosis: foliis lanceolatis, basi rotundatis,
apice acuminatis, apiculatis, supra nitidis, subtus stellato-
pulverulentis, petiolatis : paniculis terminalibus, vel racemis
axillaribus, abbreviatis, subumbellato-capitatis, pedunculis
calycibusque tomentosis, baccis pisiformibus.—Rio de Ja-
neiro, Corcovado. v. v.

The leaves of this species are 5 in. long, and 1} in. broad,
on a petiole 5 lines long, which is swollen at base.

4, Cestrum Corcovadense (n. sp.)—ramulis teretibus: foliis
ellipticis, utrinque sub-attenuatis, glaberrimis, petiolo gra-
cili basi tumido, pseudo-stipulis sessilibus, inzquilateris :
racemis axillaribus brevissimis, floribus subcapitatis, sta-
minibus edentulis, inclusis, insertione villosis: baccis pisi-
formibus.—Rio de Janeiro, Monte Corcovado. »v. v.

The leaves are of thin texture, with numerous reticulate
veins, 8 in. long, 33 in. broad, on a petiole | in. long: the
pseudo-stipules are 9 lin. long, 4 lin. broad: the racemes are
only 9 lines long, the corolla with a very smooth tube is
9 lines long, the lobes of the border being deeply turned in,
where in estivation the margins adhere by these broad
tomentous edges, the stamens are one-third the length of the
corolla.

5. Cestrum mucronatum (n. sp.)—ramulis subcompressis ;
foliis petiolatis, exstipulatis, ellipticis, utrinque glabris et
viridescentibus, apice valde attenuatis, calloque apiculatis,
marginibus subrevolutis et crassioribus; racemis axillari-
bus, geminis, compositis ; baccis oblangit em Eeagiitees Prov.
Rio de Janeiro, ad Freixal. v. v.

The leaves of this species are 4 in. long and 13 in. broad,
on a.canicular petiole, 6 lines long; the racemes are from

13 to 35 im. long; the calyx is campanular, 5-toothed, with
ciliate margins,

SOUTH AMERICAN PLANTS. 85

6. Cestrum montanum (n. sp.)—glaberrimum, ramulis tereti-
bus, virgatis, superne subangulatis ; foliis lineari-lanceola-
tis, valde acuminatis, petiolatis, pseudo-stipulis foliosis ;
racemis axillaribus, compositis, pedunculo, pedicellisque
compressis, apice tumidis, bracteis linearibus, caducissimis,
staminibus inclusis, filamentis tenuibus, dente parvo mu-
nitis.—Brasilia, in montibus Organensibus. v. v.

The leaves are 4} in. long and 11 lines broad, with 14 pairs
of divaricate nervures, on a slender petiole, 6 lin. long; the
pseudo-stipules are spathulately ovate, 5 lines long, 3 lin. ~
broad ; the racemes are 2 in. long, and the pedicels 3 lines.
The calyx is tubular, 1 line long, 5-toothed, with ciliate
margins, the tube of the corolla is slender, 8 lin. long, swollen
at the mouth, the border is expanded with lanceolate lobes
and tomentous plicate edges, the filaments, one-third of its
length, being inserted above the middle of the tube; the
stigma is cup-shaped, hollow, somewhat 2-lobed, and
exserted; the bracts are 2 lin. long, and very soon fall
away.

7. Cestrum coriaceum (n. sp.)—ramulis teretibus, furfuraceis ;
foliis lanceolato-ellipticis, coriaceis, superne nitidis, subtus
glaucis, rugosis, marginibus subrevolutis, petiolo crasso,
caniculato, pseudo-stipulis nullis, floribus paucis axillaribus,
vix racemosis, subfasciculatis, bracteis linearibus, stamini-
bus edentulis, inclusis, medio corollz insertis, filamentis
glabris, basi hirsutis.—Minas Geraes Brasilie. v. s. in herb.
meo, et Herb. Hooker, (Gardner, n. 1786, et Claussen.)
This is very distinct from C. corymbosum in its much

larger and more coriaceous leaves and longer petiole, in its

axillary flowers, in the insertion of the stamens, and other
characters. It differs also from C. aurifolium in the size of
its leaves, and its edentulate stamens. The leaves are very
thick and opaque, 9 in. long and 33 in. broad, ona petiole

9 lin. long. The peduncle of the raceme is barely 3 lines in

length, few-flowered, with linear bracts, 1 line long. The

calyx is campanulate, glabrous, 5-toothed, 14 line long, the

86 ILTUSTRATIONS OF

tube of the corolla is funnel-shaped, smooth, 6 lines long, the
induplicate margins of the lobes being tomentous in estiva-
tion, but afterwards quite smooth.

FABIANA.

This genus was first proposed by the authors of the Flora
Peruv. for a small sea-side shrub of almost ericaceous habit,
with erect terminal solitary flowers. Another species,
likewise of Chilean growth, is enumerated in the Botany of
Beechy’s Voyage; and a third is recorded by St. Hilaire,
from the southernmost province of Brazil. To these I have
now toadda fourth species, which I collected in the Paramillo
of Mendoza,in 1825. This genus by most authors has been
placed in the sub-tribe Nicotianee ; Lindley, however, in his
Introd. Nat. Syst. p. 296, has arranged it in Cestrinee, I
believe, on the authority of Bentham; I can fully confirm
this disposition, for reasons founded on its carpological cha-
racters, (ante, p. 147). Having seen the typical species, as
well as the plant above alluded to, in the living state, and
collected it in seed, 1 am provided with data sufficient
to justify the following reformation of its generic features.
Fasiana, R. et P. Prod. 22, tad. 34.— Calyx tubulosus, in-

sequaliter quinquefidus, lobis linearibus obtusis, persistens.

Corolla hypogyna, subhypocrateriformis, tubo sensim am-

pliato, limbo plicato, reflexo, breviter 5-lobo. Stamina 5,

imo corolle inserta, imclusa, inequilonga, apice incurva,

antheris cordato-2-lobis, longitudinaliter dehiscentibus.

Ovarium breviter stipitatum, glandulis 2 hypogynis liberis
- Carnosis munitum, multiovulatum, placentis sub-lamellatis

dissepimento adnatis. Stylus simplex, apice incrassato-

curvatus. Stigma 2-lobum, obliquum, lamellis incrassatis.

Capsula calyce stipata, 2-locularis, septicido-2-valvis, valvis

apice 2-fidis, margine utrinque septiferis, introflexis,

columna placentari compressa, libera. Semina plurima,
minima, ovata, facie interna angulata, hilo ventrali. Hm-

SOUTH AMERICAN PLANTS. 87

bryo intra albumen carnosum fere rectus, cotyledonibus

oblongis, compressis, radicula infera, tereti, vix latioribus,

et duplo brevioribus.—Fruticuli Austro-Americani, viscost,
vel resinosi; foliis alternis, sparsis, vel imbricatis ; pedun-
culis, extra-axillaribus, vel terminalibus, solitariis, unifloris.

1. Fabiana imbricata, R. et P. Flor. Per. 2. 12, tab. 122 c.
Hooker, Icon. Pl. tab. 340. Lind. Bot. Reg. n.s.12. tab. 59.
—Chile in paludosis maritimis, vernacule Pichi. v. v.

2. Fabiana viscosa, Hook. et Arn. loc. cit. 36.—Chile ad
Barasca, vernacule Pichinilla.

3. Fabiana thymifolia, St. Hil. Mem. Mus. 12, 317, tab. 9.
Pl. Remarg. Bres. 220, tab. 20.—Rio Grande do Sul, Bra-
zilie. te

4. Fabiana denudata (nv. sp.)—fruticosa, resinoso-glutinosa,
virgato-ramulosa, ramulis flexuosis angulatis fere aphyllis :
foliis valde deciduis, alternis, minimis, lineari-spathulatis,
obtusis, carnosis : floribus terminalibus, solitariis, bractea-
tis, breviter pedunculatis, erectis—In Andibus Mendocinis
(Paramillo). v. v.

This is a dry, arid-looking shrub, that I found growing on
the Paramille of Mendoza, in January, 1825; it is about 3
or 4 feet high, almost leafless, with numerous, erect, virgate,
very flexuose branchlets, which are angular, resinous, and

somewhat glutinous, with internodes about din. apart; the
leaves are extremely deciduous, very minute, linear, scarcely
more than a line in length, and extremely narrow. The
flowers are smaller than those of F. imbricata, solitary,
erect and terminal, upon the younger branchlets. The
calyx is tubular, 3 lines long, with 5 erect linear teeth
one third of its length; the corolla is tubular, slender,
a little swollen at base, rather funnel-shaped in the
mouth, 9 lines long, of a yellowish-white colour, quite
glabrous, with a 5-lobed border, the segments being short,
rounded, and reflected. The filaments are slender, arising
from the base of the corolla, to which they adhere for a short
distance, being somewhat dilated at base, they are nearly the
ength of the tube, 3 being somewhat shorter, they are sud-

88 ILLUSTRATIONS OF

denly reflected a little below the apex, and terminate in a
short pulvinate connective, to which the 2 anther lobes are
attached, these being somewhat divaricate at base. The pollen
(like that of Cestrum Organense), is compressed, trigonous,
with a small rounded lobe at each angle, from which 3 con-
vergent lines meet in the centre. The ovarium is obovate,
shining, stipitate, and invested at base by a free fleshy disc,
consisting of 2 broad lobes, arising from the stipes : (in F. im-
bricata these lobes are narrow, opposite, quite separate, and
erect). The style is filiform, as long as the stamens, incurved
towards the summit, where it is considerably thickened ; the
stigma is thick, clavate, oblique, with two rather reflected
lamellar lips. The capsule is cylindrical, about 5 lines long,
and 13 line diameter, invested at base by the persistent calyx ;
it splits into 2 valves, the margins of which are deeply intro-
flected, and the summit of each valve is cleft about one-
fourth of its length: the placentary column is placed trans-
versely with the valves, and shows a number of small projec-
tions arranged in longitudinal series, which have been the
points of attachment of the seeds. The seeds are numerous
and small, oval, rounded on the dorsal side, and angular on
the ventral face, the hilum being here situated a little above
the middle. The embryo placed in the centre of fleshy albu-
men is linear, almost straight, or very slightly sigmoid, the
radicle is terete, with cotyledons one-fourth of its length,
which are equal to it in breadth, and slightly compressed.*

{ may here remark, that the Fabiana lanuginosa, Hook.
Arn. Bot. Beechy’s Voyage 35, is the Dolia vermiculata, Lindl.
(ante p. 55). As the plants collected in that expe-
dition offered no specimen in seed, there was every reason
for considering it to be a Fabiana, the flower and the habit of
which it so much resembles, rather than the type of a new
genus among the Nolanacee.

mg representation of this plant with full generic details is shown in
Plate 17.

PLATE XVII. .

Exhibits a drawing of Fabiana denudata. Fig. 1, Calyx
and pistil; Fig. 2, corolla, both nat. size; Fig. 3, calyx with
bractiform leaflet; Fig. 4, corolla; Fig. 5, calyx, one half
removed, to show the stipitate ovarium upon its glandular
disc, with its style and stigma; Fig. 6, corolla cut open;
Fig. 7, stamens before bursting; Fig. 8, stamens after dehis-
cence; Mg. 9, pollen; Fig. 10, stigma; Fig. 11, capsule,
with persistent calyx; Fig. 12, capsule with half of calyx
removed, to show the persistent gland and stipes; Fig. 13,
- transverse section of capsule, with seeds removed from one
of the cells, to show placenta; Fig. 14, capsule after dehis-
cence; Fig. 15, valves separated, to show inflected margins ;
Fig 16, placente, seen front and sideway; Fig. 17, seeds in
different positions: Fig. 18, longitudinal section of seed ;
Fig. 19, embryo: all much magnified.

PLATE XVIII.

A presents a figure of Nierembergia rivularis. Fig. 1, the
corolla in bud; Fig. 2, transverse section of ditto, to show
the mode of estivation; Fig. 3, border expanded, seen from
above: all nat. size; Fig. 4, stamens in bud, removed to
show the mode in which they are embraced by the stigma;

Fig. 5, stamens seen in front and from behind, before and
after dehiscence ; Fig. 6, stigma seen sideways ; Fig. 7, stigma
seen in front, with a portion of the style removed, to show
the position of the stigma, in regard to the cells of the ova-
rium; Fig. 8, stigma seen from behind: all magnified.

B shows Nierembergia hippomanica. Fig. 1, a flower, nat.
size; Fig. 2, the calyx; Fig. 3, the corolla; Fig. 4, the co-
rolla cut open; Fig. 5, the stamens embraced by the stigma:
all magnified ; Fig. 6, the anthers, before and after dehiscence:
more highly magnified; Fig. 7, the calyx half removed, to
show the pistil, with the base of ovarium encircled by the
persistent cup of the corolla; Fig. 8, the pistil with the
stigma expanded: all magnified; Fig. 9, capsule with the
persistent calyx: nat, size; Fig. 10, ditto, magnified; Fig. 11,
the capsule removed, to show its manner of dehiscence;
Fig. 12, the persistent cup of the corolla removed; Fig. 13,
a transverse section of the capsule: all magnified; Fig. 14,
some of the seeds: nat. size; Fig. 15, a seed: magnified;
Fig. 16. a seed seen sideways; Fig. 17, a longitudinal section
of ditto; Fig. 18, the embryo extracted from the albumen
and seen sideways; Fig. 19, front view of ditto 3 Fig. 20, a
leaf: all much magnified,

PLATE XIX.

A is Mierembergia stricta.
B is Nierembergia rigida. Fig. 1, a calyx; Fig. 2,
a corolla; Fig. 3, a corolla cut open; Fig. 4, the stamens

removed; Fig. 5, the pistil seen sideways and in front: al/

natural size.

PLATE XX.

A exhibits Nierembergia linifolia. Fig. 1, the corolla cut
open; Fig. 2, the pistil, with the stamens removed and em-
braced by the stigma; Fig. 3, the stamens; Fig. 4, the pistil
with the stigma expanded: all nat. size.

B shows Nierembergia anomala. Fig. 1, a flower: nat.
size; Fig. 2, a flower cut open: magnified; Fig. 3, the pistil,
showing the stigma embracing the stamens; Fig. 4, stamens
seen in different positions before dehiscence: more highly

magnified ; Fig. 5, the pistil separated, and seen in front and

sideways : magnified.

PLATE XXI.

Shows Vestia lycioides. Fig. 1, @ flower; Fig. 2, the

calyx; Fig. 3, the corolla cut open; Fig. 4, a section through
the lobes of the corolla in bud, to show the mode of zstiva-

tion: all nat. size; Fig. 5, an anther, before and after dehis-

cence: magnified; Fig. 6, grains of pollen, highly magnified :

Fig. 7, the calyx with the front portion removed, showing
the stipitate pistillum : nat. size; Fig. 8, the stigma; Fig. 9,
a longitudinal section of the lower portion of the calyx,
showing the half of the persistent stipitate remains of the
corolla, and the ovarium seated on its stipitate 4-lobed
glandular disc: both magnified : Fig. 10, the capsule after
dehiscence, inverted by its persistent calyx ; Fig. 11, a trans-
verse section of ditto; Fig. 12, the dissepiment with its
thickened placenta bearing the seeds, detached from the
valves ; Fig. 13, seeds: all nat. size; Fig. 14, a longitudinal
section of a seed in two directions; Fig. 15, the embryo
extracted from the albumen, seen in front and sideways, and
a case of most rare occurrence with three cotyledons: all
much magnified.

PLATE XXII.

Gives a figure of Nicotiana cirrhoides. Fig. 1, the corolla
cut open; Fig. 2, the calyx and pistil; Fig. 3, ditto with the
calyx cut open and thrown back: all nat. size; Fig. 4, thé
stigma; Fig. 5, the ovarium seated on its 4-lobed glandular
disc, with the calyx cut away: both magnified; Fig. 6, the
capsule after dehiscence, invested by its persistent calyx ;
Fig. 7, seeds: both nat. size; Fig. 8, a seed, seen side and
edgeways: magnified; Fig. 9, longitudinal section of ditto ;
Fig. 10, section of albumen; Fig. 11, the embryo extracted ;
more highly magnified. q

a
igs

Plate |

P

Printed by Hullmandel & Walton —

I.Miers lithog.

Fabiana denudata.

"60
td
fH
cae)
nal
a
fas)
A)
‘ang
ary
tL)
foe ht
Pal

J.Miers lithog. : . Printed by Hulimandel & Walton.

ae Nierembergia rigida,

Plate 20. :

“Peinted by Hullmandel & Walton

oe 1 Miers lihog.

Mierember

gla. anomala.

sa ee

oe ee

Nierember

Plate 21.

‘Printed by Hullm andel & Walton

J.Miers lithog.

og
ro
i)
ws
Ay

Frinted by Hulimande} & Walon — :

aac 8 IRD se.

SOUTH AMERICAN PLANTS. 89

NIEREMBERGIA.

I have already offered some reasons (p. 71) for showing why
this genus, hitherto considered as belonging to Nicotianee,
should be separated from that sub-tribe. Several species are
now cultivated in England, so that it is not requisite to enter
into any particular details, but it is essential to define the
exact limits of this genus, especially as much resemblance
exists between several of its species and some kinds of Petu-
nia and Nicotiana, which often renders it difficult, from mere
external form, to determine to which of these three genera
they should be referred. It is, however, distinguished by
having the divisions of the calyx usually acute and rigid,
not foliaceous ; the tube of the corolla is usually very elon-
gated, and more slender than in the Petunoid section of Nico- —
tiana, though it sometimes approaches in form to some of
the smaller flowering Petunias; the stamens are generally
inserted in the mouth of the corolla, rarely in the middle of
the tube; but its most prominent characteristic, is that of the
peculiar form of the stigma, which is in the shape of an in-
verted crescent, with its horns curved round in front, embrac-
ing the anthers within its grasp; the capsule also differs from
that of the two genera above-mentioned, by having its valves
scarcely, if at all, introflexed. I have already pointed out the
difference in its stipitate ovarium, and in the form of its
pollen grains, in which respects, and in its distinctly bilamel-
lar stigma, it affords a close approximation to Fadiana, In
order to remove the doubts that may exist in regard to the
several species of this genus, I will offer a short enumeration
of each, with the more striking characters, as far as I have
been able to observe them, and will add some new species
that I met with in 1825, together with a few others, yet
undescribed. that exist in the herbarium of Sir Wm. Hooker ;
I propose to amend the generic character in accordance with
these observations, in the following manner.
Nrerembercia, R. et P.— Calyx tubulosus, 10-costatus, semi

5-fidus, laciniis linearibus, subineequalibus, acuminatis. Co-

N

90 ILLUSTRATIONS OF

rolla hypogyna, tubulosa, tubo gracili, elongato, rarius sub-
infundibuliformi, ore campanulato, limbo amplo, expanso,
breviter 5-lobo, lobis rotundatis, zstivatione plicato. Sta-
mina 5, corolle fauci, rarius medio inserta, exserta, raris-
sime inclusa, ineequilonga, circa stylum conniventia; an-
theris longitudinaliter dehiscentibus, stigmate circumplexis.
Ovarium breviter stipitatum, 2-loculare, placentis dissepi-
mento adnatis, multiovulatis, eyatho (corolle reliquo)
demum circumdatum. Stylus simplex. Stigma sub-
laterale, oblunato-bilamellatum, lamellis reflexis. Capsula
calyce persistente tecta, 2-locularis, septicide 2-valvis,
valvis introflexis demum 2-partitis, dissepimento placen-
tari libero. Semina plurima, minima, ovata, facie interna
angulata, hilo ventrali. Hméryo intra albumen carnosum
fere rectus, cotyledonibus oblongis, compressis, radicula
infera tereti vix latioribus, et longitudine equantibus.—
Herbee Austro-Americane procumbentes vel radicantes, viz
suffruticulose, foliis alternis solitariis v. geminis integerri-
mis, floribus extra-axillaribus vel oppositifoliis, solitariis,
subsessilibus, albis vel violascentibus.

1. Nierembergia repens, R. et P. Fl. Per. 2, 13, tab. 123 c.—
caule basi repente, ramulis erectis filiformibus ; foliis sub-
fasciculatis, oblongis, pilosulis ; floribus subsessilibus, ca-
lyce curvato, tubuloso, lobis obtusiusculis.—Chile ad Con-
ceptionem.—v. s. in Herb. Hook. (Cuming. n. 135.)

The leaves in the specimens I have seen are ovate, about

5 lines long, 3 lines broad, on a short petiole of 2 lines, they

are thinly covered with short pubescence; the calyx is about

3 lines long, also pubescent; the corolla is white, with a long

slender tube, about an inch in length, and a broad campanu-

late border, nearly 9 lines diameter, divided into 5 rather
short, rounded lobes, each of which is marked with 3 purple
longitudinal lines, the throat being of a yellowish colour.

2. Nierembergia spathulata, H. B, K. nov. gen. sp. 3, 8.—
eaule lignoso repente, fibrillis tuberoso-nodosis 3 Tamulis
erectis filiformibus puberulis; foliis solitariis vel geminis
elliptico-spathulatis, longe petiolatis, glabris margine cilio-
latis, junioribus lanuginoso-pubescentibus; calyce subsessili

SOUTH AMERICAN PLANTS. 91

curvato, tubuloso, pulverulento-tomentoso, laciniis obtusis.

— Nova Granada, ad Santa Fe de Bogota et Loxam; altit.

6-8000 ped. v. s. in Herb. Hook. (Bogota, Goudot.)

The leaves are nearly glabrous, or slightly ciliolate, the
blade elliptic, 6 lines long, 3 lines broad, on a petiole of 6
lines. The calyx is 4 lines long, the tube of the corolla 8 lines
long, the border being more tubular and narrower in the
mouth than the former species, and much less in diameter.
3. Nierembergia calycina, Hook. Bot. Mag. tab. 3371.—caule

repente, filiformi, ramulis procumbentibus, foliisque alternis

v. geminis, obovatis, petiolatis, hirto-pubescentibus, pilis

canis divaricatis ; pedunculis solitariis, brevibus, extra-axil-

laribus; calyce magno, tubuloso-campanulato, laciniis obo-
vatis, foliaceis, expansis; corolle tubo longissimo.—In

Andibus, prope Mendozam. v. s. in Herb. Hooker.

This species is very distinct from the two former, the axils
are more distant, the leaves generally in pairs, obovate, sud-
denly acuminate at each end, 9 lines long, 5-6 lines broad, on
ashort petiole of 2 lines, they are hirsutulate, with long, white,
spreading, adpressed hairs ; the peduncles 9 to 11 lines long;
the calyx about 9 lines long, is tubular at base, for the length
of 3 lines, the lobes being broad, foliaceous, obovato-spathu-
late, slightly apiculate, with a row of long spreading hairs on
each side of the mid-rib; the tube of the corolla is extremely
slender, 2 inches long, the border broadly campanulate, 14 in.
diameter, with 5 founded lobes, the throat and tube being of
a yellowish hue.

4, Nierembergia rivularis (n. sp.)—caule repente, filiformi,
ramulis procumbentibus, foliisque longissime petiolatis,
spathulato-ovatis glabris; floribus solitariis subsessilibus,
oppositifoliis ; calyce curvato, laciniis oblongis, subobtusis,
erectis; corolle tubo valde elongato, glabro.— Buenos
Ayres. v. 0. -

I found this plant on the grassy banks of the Rio de la
Plata, the prostrate branches creeping among the grass, above
which rise its pretty white flowers. The leaves are alternate,

more rarely geminate, quite glabrous, the blade is from 9 to
n 2

92 ILLUSTRATIONS OF

12 lines long, 4 to 6 lines broad, elliptic, somewhat obtuse
at summit, cuneate at base, on a slender petiole 13 long. The
flower is solitary upon a short horizontal peduncle, 1 line
long; the calyx is curved and slightly pubescent, 10-nerved,
4 lines long, with unequal, oblong and somewhat obtuse seg-
ments; the corolla is glabrous, with a very slender tube,
23 inches long, } line thick, its border is broadly campanulate,
nearly an inch in diameter, with 5 short rounded lobes; the
stamens are exserted, fixed in the extreme edge of the tube,
quite smooth. .

5. Nierembergia hippomanica (n. sp.)—Trav. Chile. 2, 532.—
Planta scabrido-pilosa, caule lignoso, humifuso, radicante ;
ramulis brevibus adscendentibus; foliis confertis, spathu-
lato-linearibus, acuminatis, calloso-mucronatis; floribus,
parvis, subpaniculatis; calyce brevi, 10-carinato, lobis
linearibus, patulis; corollz tubo gracili, brevi, calyce 2-plo
longiori, glanduloso-pubescenti, limbo late campanulato,
lobis rotundatis.—Provincias Argentinas, vernacule Chu-
chu. v. v.

a. foliis confertis, lanceolato-linearibus, valde approxima-

tis, scabrido-pilosis. In Prov. Cordove et San Ludov.

B. foliis glaucis, glabriusculis, brevibus, anguste linearibus.

Achiras,

This is a low-growing species, disseminated in certain
pasturage districts within the Provinces of San Luiz
and Cordova, and is said by the Gauchos to be very
poisonous to horses. The root is much thicker, more tor-
tuous and woody than the former species, creeping along
the surface of the ground, and throwing up several woody
erect branching stems, 4 to 6 inches high, which are clothed
with close-set foliage. The leaves are 4 lines long, ¢ lin.
broad, linear, and spathulate, covered with dense scabrid
hairs. The peduncles are 1 line long, the calyx 2 lines, the
tube of the corolla 3 to 4 lines long, with a broad campanu-

late border, 5 lines diameter, and is of a white colour, tinged
with a rosy hue.* :

ace representation of this species ig shown in Plate 18 of this

SOUTH AMERICAN PLANTS. 93

6. Nierembergia ericoides (n. sp.)—caule adscendente, ramu-
lis erectis virgatis; foliis brevibus sub-fasciculatis, breviter
linearibus, obtusis, carnosulis, scabrido-pilosis, calyce 10-
costato, costis hirsutulis.—Banda oriental. (Tweedie, v. s.
in Herb. Hooker).

This species approaches the last, but is very different in
habit, being much branched, with straight virgate stems, and
growing to the height of 12 to 18 inches, whereas the other
seldom exceeds 3 to 6 inches in height, is tortuous and
greatly stunted. The leaves are fasciculate, much more fleshy,
more scabrid, and the axils are more diffuse. The flowers
are covered with simple, not glandular pubescence, are much
larger, and the tube of the corolla is longer in proportion to
the calyx; the leaves are sessile, about 3 lin. long, and barely
a line broad; the calyx is covered with long hairs, and is 2
to 3 lines long, on a peduncle of 2 lines; the tube of the
corolla is 5 lines long, and the broadly campanulate border is
from 6 to 8 lines in diameter, and of a white colour; the
capsule is small and ovate. .

7. Nierembergia gracilis, Hook. Bot. Mag. tab. 3108. Sweet.
Fi. Gard. 2 ser. tab. 172 ;—caulibus strictis, erectis, her-
baceis, foliisque spathulato-linearibus, pubescentibus, axillis
superioribus alternis, inferioribus subverticillatis, internodiis
brevibus ; floribus in ramulis junioribus terminalibus, ca-
lyce tubuloso 10-costato, inter costas membranaceo, laci-
niis linearibus, obtusis, corollz tubo calyce multo longiori,
limbi campanulati lobis rotundatis. — Bonaria et Prov.
Argent. (Pampas.) v. s. in Herb. Hooker.

This is a far more erect and slender species than N. hippo-
manica, its stems being of a less woody texture: its leaves
are not fasciculate as in the two preceding species, but are
simple, very linear, about 7 lines long, and 1 line broad,
sometimes smaller. It is a well known plant in our gardens,
where, however, its habit is still more slender, its stems
weaker, and more prostrate than in its native growth.

8. Nierembergia filicaulis, Lindl. Bot. Reg. tab. 1649. Bot.
Mag. tab. 3370. Sweet. Fl. Gard. 2 ser. tab. 243 ;—caule
herbaceo, erecto, filiformi, ramulis subflexuosis foliisque

94 ILLUSTRATIONS OF

lineari-lanceolatis, vagis, glabris ; pedunculis oppositifoliis,

solitariis, calycis laciniis acuminatis, corolla tubo gracili

calyce vix longiori, filamentisque glanduloso-pubescentibus,
limbo magno, eampanulato, lilacino, fauce luteo.— Bon aria,

v. 8. in Herb, Hooker. Rio Parana, (Tweedie). Entre-Rios,

(Tweedie) n. 284. Rio Grande do Sul (Isabelle).

Its native place is erroneously stated in most botanical
works, to be Mexico. The leaves are 12 to 14 lines long,
1} line broad, somewhat spathulate at base, a little obtuse at
summit, with rather distant internodes, The calyx is 10-

nerved, the nervures being covered wit

h short pruinose
fomentum, is 6 lines long, cleft half way down into 5

subequal, linear, obtuse segments. The tube of the corolla
to the extremity of the calyx is slender, above which, for
an equal length, it swells into a broadly campanular form,
the border. about 14 inch diameter, being expanded and
divided for half its breadth into 5 subequal short rounded
lobes ; it .is slightly pubescent outside, of a lilac colour
marked with purplish lines.

9. Nicrembergia angustifolia, HBK. Nov. Gen. 3, 9, tab. 198:
= —caulibus herbaceis, ceespitosis, erectiusculis ; foliis ses-

Be silibus, lineari-lanceolatis, glabris, pedunculis oppositifoliis,
_ Solitariis ; calycis laciniis longe linearibus, acutis, corolle

tubo elongato; glabro, limbo campanulato, lobis brevissi-
mis, rotundatis.—Mexico:

This species much resembles the last in habit, the leaves

8 lines long, 1 lin. broad, are Spathulate at the base. The
peduncle is 2 lines, the calyx 5 lines,

10, Nierembergia aristata, Sweet,

Fl. Gard. 2 ser. tab. 2553
. glabriuscula, caulibus lignosis,

erectis, valde ramosis, gra-

SOUTH AMERICAN PLANTS. 95

cile filiformibus: foliis spathulato linearibus, obtusiusculis,
calloso-mucronatis, calyce urceolato tubuloso, 10-costato,
inter costas membranaceis, laciniis zequalibus lanceolato-
acuminatis, calloso-aristatis, subpatulis; corolle tubo
gracili, glanduloso-pubescenti, limbo late campanulato,
lobis amplis rotundatis; capsula ovata, calyce aucta in-

clusa.—Provincias Argentinas. Uv. v.

I met with this species at Encrucijada, in the Province of
San Luiz, and in Buenos Ayres, in which last place it was
also found by Tweedie, from whom the seeds cultivated in Eng-
land were obtained ; from the former locality, the branches
are stronger, and more ligneous ; but shorter, more branched,
and more herbaceous from the latter. The leaves are 8 tol
inch long, 1 line broad, almost glabrous, the internodes mea-
suring 4 lines: the peduncle is of the length of the axil,
4 lines, opposite the leaf, and flexuosely divaricate with the
stem, which is, however, quite straight below; the calyx,
almost glabrous, is 4 lines long, cleft half way down into 5
spreading rigid lobes, the tube of the corolla is extremely
slender, 6 lines long, the border being about § inch diameter,
it is white, with longitudinal striated purple lines. The cap-
sule is about 4 lines long and sub-4-valved.

11. Nierembergia stricta, (n. sp.) — glabriuscula, caulibus
simplicibus, lignosis, erectis, gracillimis, strictis: foliis spa-

_ thulato-linearibus, acuminatis, calloso-mucronatis, eveniis,
utrinque glabris, margine sub-ciliatis: pedunculis oppositi-
foliis ; calyce 10-costato glabro, costis ciliatis, laciniis spa-
thulato-lanceolatis, aristatis, erectis, inaqualibus, tubo
3plo-longioribus, corolle tubo valde elongato, glanduloso-
pubescenti, calyce 3-4plo longiori, limbo late campanulato,
lobis brevissimis rotundatis.—Buenos Ayres, v. s. in herb.
meo.

- This plant may be only a variety of N. aristata, but it cer-
tainly differs from it in having very slender simple stems
with more distant internodes, and only 1 or 2 terminal
flowers; its calyx has shorter, broader, and more erect seg-
ments, and the corolla is larger, with a far more elongated

96 ILLUSTRATIONS OF

tube: the leaves are 1 inch to 14, or even 20 lines long,

13-2 lines broad, the internodes being 5 to ?) lines distant ;

the peduncle is 3 lines, the calyx 7 lines long, obsoletely 10-

ribbed, the segments are 5 lines in length, the space between

the ribs not being white and membranaceous as in the former
species: the tube of the corolla is extremely slender, 14-2 in.
long, the border measuring 10 lines in diameter.*

12. Nierembergia rigida (n. sp.) Trav. Chile. 2.532 ;—glaber-
rima, caulibus ramosis erectis, valde flexuosis, filiformibus,
striatulis: foliis angustissime linearibus, 3-costatis, mucro-
nato-aristatis: pedunculis solitariis, brevibus, oppositifoliis,
calyce 10-costato, laciniis linearibus, rigidis, aristatis, tubo
3-plo longioribus ; corolle tubo gracili, pubescenti, limbo
late campanulato, lobis rotundatis.—In Prov. Argentinis et
Andibus Mendosinis: Rio Negro (Patagonica), Tweedie,
v. v. et 8. in herb. Hook.

I found this very distinct species at Punta de Agua and
Achiras, in the Province of Cordova, about 550 miles from
Buenos Ayres: it was met with in the Pampas by Dr.
Gillies, who also collected it in the Cordillera of Mendoza.
The stems, which arise from a long slender ligneous root,
are from 6 to 8 inches high, very filiform, extremely flexuose,
with internodes 3 lines distant: the very rigid leaves are 15 lines
long, not more than one sixth of a line broad, and somewhat
falcate and spreading: the peduncle is 3 lines in length ; the
calyx, 6 or 7 lines long, is furnished with very short and
almost glandular white pubescence, the tube being only 2
lines, has 10 very prominent ribs, and 5 rigid linear segments,
nearly a line broad, tapering gradually to a sharp point, each
3-nerved : the tube of the corolla is 9 lines long, with a
white campanular border, nearly an inch in diameter, having
5 oblong rounded lobes, with purple longitudinal lines ; the

stamens, fixed in the mouth of the tube, are slightly pu-
bescent.+

* A figure of this species is given in Plate 19 of this Work.
+ This species is represented in Plate 19.

SOUTH AMERICAN PLANTS. 97

13. Nierembergia pinifolia (n. sp.);— caule adscendente
lignoso, inferne denudato, ramulisque delapsu foliorum,
cicatricosis, apice foliiferis; foliis imbricato-confertis,
ereetis, linearibus, 3-costatis, apice miucronatis, caly-
ceque rigido 10-costato glanduloso-pubescentibus ; corolla
violascenti.—Brasilia meridionali, (Prov. Rio Grande do
Sul.) v. s. in herb. Hooker.

The leaves are closely imbricated at the summit of the
branchlets, nearly erect, and closely tmbricated, 11 lines
long, } line broad, the calyx is 8 lines long, cleft half way
down into 5 equal lanceolate rigid segments, the tube of the
corolla is scarcely longer than the calyx, and the campanular
border 3 inch diameter, of a lilac hue, is marked with longi-
tudinal purple lines.

14, Nierembergia pulchella (n. sp.);—caulibus adscendentibus;
foliis lineari-lanceolatis, utrinque acutis, margine apiceque
incrassatis, subpubescentibus: calyce subinflato, 10-nervi,
pilosulo ; corolla profunde violacea, puberula, tubo gracili,
calyce paulo longiori, limbo late campanulato, staminibus
ineequalibus.—Cordillera de los Andes, ora orientali. v. s.
in herb. Hook. (Gillies MSS. N. pulchella.)

The leaves are 11 lines long, on a petiole of 2 lines, and
1} line broad; the calyx is 5 lines long, cleft half way into 5
linear, acuminate, erect segments, it swells in fruit to the
length of 7 lines, the tube becoming nearly globular, 2} lines
in diameter, the tube of the corolla is 6 lines in length, the
border campanulate, 5 lines in diameter, and of a deep purple
colour. .

15. Nierembergia cerulea (n sp.) ;—caulibus adscendentibus,
pubescentibus; foliis parvis, anguste linearibus, fere fasci-
culatis, erectiusculis, falcatis, demum glabris; corolla vio-
lacea.—Punta del Sauce, Prov. Cordovee. v. s. in herd.
Hooker. (Gillies MSS. N. cerulea.)

The leaves are 6-7 lines long. 3 lin. broad, the calyx 3
lines long, cleft half way in 5 equal, linear, somewhat reflexed
segments, on a peduncle of 2 lines: the tube of the corolla

ny)

98 ILLUSTRATIONS OF

is slender, 6 lines long, with a border broadly campanulate,

8 lines in diameter, divided into 5 short rounded segments.

16, Nierembergia pubescens, Spr.—N. graveolens, St. Hil. Pl.
rem. Bres. 221, tab. 21, A.—caule lignoso, diffuso : foliis
oblongo-linearibus, utrinque acuminatis, glanduloso-pilo-
siusculis.—Montevideo.

The above figure shows the leaves to be 10 lines long, 2 lines
broad, the calyx 7 lines long, cleft half way into 5 lanceolate
erect segments, the tube of the corolla 10 lines long and
slender, the border broadly campanulate, 6 lines diameter.
17. Nierembergia linifolia (n. sp.) ;— glanduloso-scabrida,

caule lignoso, e basi ramuloso, ramis virgatis, rectiusculis ;

foliis sessilibus, lanceolato-linearibus, apice callosis ; pe-
dunculis oppositifoliis, bracteatis, calyce brevi 2-3-plo lon-
gioribus ; istius lobis linearibus aristatis; corolle tubo in-
fundibuliformi, limbo parvo expanso, 5-lobo: staminibus

inclusis.—In Proy. Argentinis. v. v.

a, internodiis longioribus, foliis majoribus pallidis, erectis,
floribus subflexuoso racemosis.—Mendosa, in herb. meo et
Hook. Petunia (Mendozinensis, Gillies MSS.)

8. internodiis brevioribus, foliis divaricatis, floribus subsoli-
tariis.—Prov. Cordove. In herb. meo et Hook. (Nicotiana
linoides, Gillies MSS.)

Both varieties of this species, which I found in the
places quoted, in 1826, bear much the appearance of a Linum
in habit. It is distinct from all others belonging to the same
genus, in the greater length of its peduncle, in its corolla
having a much thicker and more infundibuliform tube, with
its stamens included altogether within its mouth, and a
narrow border cleft to its base, into oblong reflected seg-
ments, in which respects it more nearly resembles Fabiana.
In the var. a, the leaves are 4 to 9 lines long, 1 line broad,
the internodes 6 to 10 lines distant : in var. 8. the leaves are
4-6 lines long, Z to 1 line broad, the internodes 3 lines apart :
the peduncle is 7 lines long; the calyx 3 lines, cleft deeply into
5 linear erect segments ; the tube of the corolla is 6 lines long,

SOUTH AMERICAN PLANTS. 99

slender at base, swelling gradually towards the mouth, where

it is 13 line broad, the border only 3 lines in diameter, con-

sisting of 5 short expanded oblong rounded lobes.*

18. Nierembergia anomala (n. sp.) ;—glabriuscula, suffruticu-
losa: caulibus plurimis ramosis adscendentibus : foliis radi-
calibus longissime, caulinis breviter petiolatis, oblongo-
lanceolatis, utrinque attenuatis, crassiusculis, eveniis, fere
glabris, sparse pilosiusculis, junioribus linearibus, floribus-
que glanduloso-pilosis, pilis sepissime scabridis patentibus
dense tectis; floribus paucis, longe pedunculatis; corolle
tubo infundibuliformi, calyce fere duplo longiori, fauce
amplo, limbi 5-fidi lobis parvis rotundatis expansis.—Pro-
vincias Argentinas. v. v. in Prov. Cordove. v.s. in Herd.
Hook. Prov. Cordovee (Gillies, Nicotiana breviflora, MSS.) —
Monte Video et Cordova (Tweedie, No. 1122).—Texas ad
San Felipe (Drummond, 3rd Coll. 245).—Chile ad Quillota
(Bertero. sec. Colla.)

I first saw this plant in May 1826, at Frayle Muerto and
Zanjon in the province of Cordova, at a distance of 360 miles
from Buenos Ayres, and it was afterwards found by Gillies
and Tweedie in the same province, but their specimens are of
more slender growth, the stems more virgate and herbaceous,
the leaves more distant, narrower, more pubescent and glau-
cous than the plants I met with, which in habit closely resem-
ble those collected in Texas. The root is repent, knotty
and woody, throwing out several erect shoots at intervals,
which are from 6 inches to a foot in height : the radical leaves
have a slender petiole as long as the blade, being altogether 23
inches long, and 4 lines broad, the cauline leaves have a petiole
scarcely a line in length, and are from 10 to 16 lines long,
about the length of the internodes, and are 2 to 5 lines broad,
veinless, erect, and together with the stem, are almost gla-
brous, the younger leaves and branches are, however, covered
more or less densely, with short, rigid, spreading, and glutinous

* A representation of this species is shown in Plate 20.
o 2

100 ILLUSTRATIONS OF

hairs, the floral leaves being linear, narrow, scarcely more
than 4 lines long. The flowers are axillary, the peduncle
being 8 lines long, lengthening in fruit to 18 lines: the calyx
as well as the peduncle is densely covered with short very
spreading rigid hairs, somewhat glutinous, it is 3 lines long,
tubular, 10-nerved, and divided half way down into 5 une-
qual, thickened, linear, obtuse, erect segments ; the corolla is
of a whitish, sometimes of a yellowish colour, slightly pubes-
cent outside, 5 lines long, quite infundibuliform, marked
with 15 longitudinal purplish veins, is enlarged in the mouth,
and has a very narrow border of 5 short spreading rounded
lobes: the filaments are dilated, fixed in the middle of the
tube: the ovarium is oblong and smooth: the style is erect, —
smooth, as long as the corolla: the stigma is lunulate, or
deeply reniform, expanded and embracing the anthers within
its encircling rounded lobes: the capsule is smooth, and of
the length of the incanescent persistent calyx.*

This plant, it may be presumed, is widely disseminated,
for I can discover no difference between the specimens from
Texas, and those I found in the Pampas of Buenos Ayres,
either in their inflorescence, their leaves, or their habit, except
that from the latter place the pubescence is short, rigid,
widely spreading, and viscid at the tips, while from the
former, it is longer, soft, adpressed, and quite free from
glandular viscidity ; but this is not sufficient to constitute a
specific difference.

The plant collected at Quillota in Chile and described and
figured by Colla, (Memorie di Torino, 38, 135, tab. 45), as
my Petunia viscosa,t is evidently the same species.

* This species with sectional details is exhibited in Plate 20.
t The plant enumerated by me under this name (Trav. Chile, 2. 531) is

that subsequently named by Prof, Graham, Nicotiana (Petunioides) acumi-
nata,

SOUTH AMERICAN PLANTS. 101

SPECIES DUBIA.

19. Nierembergia? nana. Nicotiana nana, Lindl. Hort.
Trans. 6,92. Bot. Reg. tab. 833.—Planta acaulis, czespi-
tosa; foliis confertis, lanceolatis, longe spathulatis, pilosis,
ficribus multo longioribus; pedunculis 1 floribus, calyce
turbinato, glutinoso-piloso, lobis obtusis, corolle tubo
calyce 2-plo longiori, pilosulo.—Oregon, Rocky Mountains,
v. 8. in Herb. Hook.

This plant in no way resembles any species of Nicotiana,
and as stated by Dr. Lindley, it is difficult to account for its
having been confounded, with N. multivalvis; in habit it
greatly resembles the four first species of Nierembergia above
enumerated, but a better knowledge of the structure of its
flower, is wanting to point out its true position. The only
specimen I have seen, exists in the herbarium of Sir Wm.
Hooker, and that is entirely destitute of inflorescence.

VESTIA.

This genus consists only of a single well-known species,
long since described by many authors, and figured in the
Flora Peruviana, under the name of Periphragmos fetidus.
I find, however, many of these recorded features at variance
with what I have seen, more especially in the figure alluded
to, where the structure of the capsule and the seed, is very
inaccurately given. The following is an outline of its generic
character, according to my own observations.

Vestia Wild. (Char. emendat.)—Calyxr campanulato-tubu-
losus, breviter 5-dentatus, dentibus apiculatis, demum
auctus et capsulam suffulciens. Corolla hypogyna, infun-
dibuliformi-tubulosa, basi demum circumscissa, limbi 5-fidi
lobis eequalibus ovatis, zstivatione conduplicatis. Stamina
5, paulo supra basin in contractionem corolle adnata, ex-
serta; filamenta e glandula crassa dense barbata orta,

102 ILLUSTRATIONS OF

basi dilatata, superne filiformia, glabra; anthere cordate,
ovate, basifixe, longitudinaliter dehiscentes. Ovarium
ovatum, breviter stipitatum, disco annulari crenato striato
sessum, cyatho persistente (corolle reliquo) demum clau-
sum, 2-loculare, ovulis plurimis, dissepimento utrinque
adnatis. Stylus simplex, exsertus. Stigma incrassatum,
lamellis 2 carnosis agglutinatis. Capsula ovata, 2-locularis,
2-valvis, valvis fere 2-fidis, dissepimento demum libero
medio placentifero parallelis, incrassato. Semina plurima,
angulato-ovata, sinu ventrali excavata, hinc supra medium
hilo parvo notata, testa tuberculata, striatulis transversis
reticulatis signata. Embryo in axi albuminis carnosi fere
omnino rectus, cotyledonibus 2, rarissime 3, parvis, sub-
compressis, radicula tereti infera sub-latioribus, et 3-plo
brevioribus.—Frutex Chilensis erectus ramosus, Cestri

Sacies ; foliis alternis, petiolatis, integerrimis, glabris ; floribus

e summis alis subracemosis, pedicellis medio articulatis et

hine bractea foliosa decidua instructis, nutantibus.

1. Vestia lycioides, Wild. Cantua ligustrifolia, Juss. Cantua
foetida, Pers. Periphragmos fctidus, R. et P. Flor. Peruv.
2. 7. tab. 132:—foliis obovatis vel lanceolato-ellipticis,
basi in petiolum brevem decurrentibus, carnosis, eveniis,
glaberrimis ; corolla viridi-flava, limbo subviolaceo.—Chile,
(Valparaiso et Concepcion.)

This shrub grows in the neighbourhood of the saline lake
of Bucalemu, about sixty miles to the southward of Valpa-
raiso, on the opposite side of the river Maypu, where it is
called Porotillos. It abounds also in Concepcion, where
according to Ruiz and Pavon, it is known by the name of
Guevil guevil. It is much branched, with dense foliage : the
leaves being somewhat fleshy, veinless, about 13—2 inches
long, 4 in. broad, tapering into a short petiole, the upper
surface is smooth, channelled in the midrib, which is promi-
nent below. The flowers appear on the termination of small
branchlets, the pedicel is articulated in the middle, where a
small lanceolate leaf-like deciduous bract is seen, and it
becomes deflexed, giving a pendent position to the flower.

SOUTH AMERICAN PLANTS. 103

The calyx is 4 lines long, somewhat campanulate, glabrous,
purplish green, with a long woolly apex to each broad short
tooth. The corolla is tubular, contracted at base within the
calyx, expanding above into a somewhat funnel-shaped tube,
ofa greenish yellow hue, about 1% inch long, and 3 lines in
diameter, quite glabrous, the border is cleft to the mouth into
5 equal oblong lobes, the margins being broadly conduplicate
and tomentous in estivation, but glabrous when fully ex-
panded : the base of the tube is somewhat pubescent within,
thickened, and exhibiting 5 raised fleshy ribs, surmounted by
a thick broad gland, which is tufted with long white hairs,
from these arise the filaments, dilated at their origin, glabrous,
tapering upwards into a slender form, and exserted: the
anthers are of a lurid green, deeply cordate, of 2 oval lobes,
fixed in the sinus on the apex of the filament. The ovarium
is wholly enclosed in the persistent cup-shaped base of the
corolla, which here always falls off by a horizontal line: it is
supported on a small 5-lobed disc, which is shortly stipitate,
and quite free. The capsule is oval, smooth, LO lines long,
5 lines diameter, supported by the cupuliform enlarged calyx,
is divided nearly to the base into 4 valves, the thickened dis-
sepiment becoming free, to which the seeds are attached by a
number of projecting fleshy seminiferous plates. The seeds
are ovate, somewhat angular, with a ventral excavation, in
which the hilum is seen a little above the middle. The testa
is tuberculated, with intervening small transverse striz, and
encloses an oval fleshy albumen, in the centre of which the
embryo is placed, which is erect and almost straight, but
which viewed in front is very little sigmoid, and sideways
slightly curved inwards: the radicle is thick and terete, the
cotyledons being somewhat broader, slightly compressed and
fleshy: I have sometimes found 3 cotyledons : they are about
1 the length of the embryo.*

* A representation of this species with sectional details is given in
Plate 21

104 ILLUSTRATIONS OF

NICOTIANA.

I have very few observations beyond what has already
been remarked upon the well known genus, my chief object
being to describe here a new and singular species that I found
in Chile in 1822. I will, however, avail myself of the oppor-
tunity of tracing an outline of its generic character, in accord-
ance with the views before stated.

Nicotiana, Linn.—Calyx tubulosus, semi-5-fidus. Corolla
hypogyna, infundibuliformis, limbo plicato 5-lobo. Sta-
mina 5, inequilonga corolle, tubo infra medium inserta,
inclusa : anthere longitudinaliter dehiscentes: pollen oblon-
gum, longitudinaliter 3-suleatum. Ovarium sessile, disco
annulari obsolete 4-lobo stipatum, 2-loculare, placentis
linea dorsali dissepimento adnatis, multiovulatis. Stylus
simplex. Stigma subpattelliforme, intus glandulis 2 mag-
nis instructum. Capsula calyce persistente tecta, 2-locu-
laris, imo septicide 2-valvis, valvis demum 2-fidis, placenta
centrali denique solutis. Semina plurima, minima, oblonga,
imo ad faciem ventralem hilo rostrato notata, testa reti-
culato-foveolata, costis intermediis crenulatis. Embryo in
axi albuminis carnosi fere rectus, vel leviter incurvus, teres,
radicula infera cotyledonibus subclavatis duplo longiori.—
Herbe interdum suffrutescentes, sepissime glutinoso-pilose
in America tropica copiose, parcius in Asia orientali cali-
diori crescentes : foliis alternis, integerrimis ; floribus termi-
nalibus, racemosis, aut paniculatis, albidis, virescentibus vel
purpurascentibus.

Nicotiana (Petunioides) cirrhoides (n. sp.) Petunia cirrhoides,
Nob. Trav. 2. p. 531 3—herbacea, erecta, glanduloso-viscida,
pilisque brevibus articulatis vestita: foliis lanceolatis, basi
in petiolum longum attenuatis, margine undulatis, acu-
mine in appendicem gracilem cirrhiformem apice spathu-
latum attenuatis, summis linearibus longissime et tenuiter
apiculatis, floralibus angustissime linearibus: floribus ter-
minalibus paniculatis, calyce campanulato, 5-nervi, denti-

SOUTH AMERICAN PLANTS. 105

bus 3-angularibus inequaliter et longissime apiculatis ; co-

rolla cylindracea, sordide albida, tubo 15-nervi, subglabro,
_ basi coarctato, calyce 5-plo longiori, limbo fere rotato,
_ obsolete 5-dentato, dentibus aagustis longissime cuspidatis.

—Chile ad Concon. ». »v.

This plant which I found in Chile in 1821, bears much
analogy to N. acuminata, Grah. and is distinguished for the
very singular elongation of the midrib of the leaves, and of
the nervures of the calyx and corolla. I do not remember
the size of the radicular leaves, but those of the stem measure
9 inches, including 1 inch for the petiole, 5 inches for the
blade, and 3 inches for its cirrhiform apical extension : they
are 9 lines broad with a very undulatory margin, and are
covered.on both sides (like the remainder of the plant) with
short glutinous articulated pubescence, the upper leaves
become gradually shorter, narrower and linearly lanceolate
with reflected margins, while the floral leaves or bracts are
14 lines long, quite linear and terete, being reduced to a
mere slender midrib covered with glutinous pubescence. The
panicle is nearly a foot in length, branching at the axil of
each flower ; each pedicel is about 4 lines long; the somewhat
campanular tube of the calyx is 2 lines, and the mucronate
teeth are from 2 to 3 lines in length; the corolla is nearly
2 inches long, and barely 3 lines diameter in the mouth, the*
border is narrow, with 5 short obtuse teeth, having slender
cuspidate nerval extensions of 14 line in length; one of the
stamens is shorter than the other 4, which are almost in didy-
namous pairs, the filaments are slender and glabrous, arising
from a little above the base of the tube: the ovarium is gla-
brous, the style slender, and the stigma has the peculiar
form of this genus, the capsule is about 5 lines in length,
invested by the persistent calyx, is 2-valved, and in other
respects resembles that of the other species of this genus.
The seeds are small, roundish, subangular, the hilum being
seen on a prominent beak that projects from the lower end

of the ventral face, the surface of the testa is divided into
P

106 ILLUSTRATIONS OF

numerous deep hexagonal areoles, separated by crenate undu-
lating ridges, a character more or less conspicuous, and com-
mon to the seeds of all the species of this genus that I have
examined; the embryo enveloped in copious albumen is
cylindrical, the radical which points to the base of the seed,
not far from the hilum, is slightly curved at its union
with the cotyledons, which are somewhat clavate and half its
length.*

I take this opportunity of indicating the suggestion, judg-
ing from specimens existing in Sir Wm. Hooker’s Herbarium
that the Nicotiana quadrivalvis, Pursh. and the N. multivalvis,
Lindl. both from the western coast of America, may belong to
the new genus established by Dr. Hooker, under the name of
Dictyocalyz, upon a plant found by Mr. Darwin in the island
of Gallapagos ; but they require examination.

The 4th section of Nicotiana proposed by G. Don, Dict.
4. 466, in order to comprise the above species, under the
title of Polydiclia must therefore be suppressed; the other
species contained in this section (with what reason I cannot
ascertain, as its seed is unknown) is certainly a plant of very
opposite character, and appears to me for the reasons given
in p. 101 to belong rather to Nierembergia. The N. solani-
folia of Dr. Walpers, placed by him in this division (Rep. Bot.
‘4, 12) has since been referred to the section Rustice.

PETUNIA.

The limits of this genus, as I have before observed, are not
very definite, so that some species by different able botanists
have been confounded with Nicotiana, Nierembergia, and Sal-~
piglossis. To Nicotiana a near approach is manifest, the
most striking distinction being seen in the valves of its cap-
sules, which are entire, while in the former they are cleft half
way, or sometimes nearly to the base, but in Petunia though

* A figure of this plant with sectional details is represented in,
Plate 22,

SOUTH AMERICAN PLANTS. 107

the capsule is generally of more membranaceous texture,

notched and entire at the apex, it is sometimes more coria-

ceous, when by the least pressure, the valves often split as in

Nicotiana : there is hardly, however, so great an introflexion

in the lower portion of the valves. The corolla of Petunia is

often somewhat oblique, and marked with several branching
and reticulate veins on one side, as in Salpiglossis, corres-
ponding with the shorter pair of stamens.

In Nicotiana the inflorescence is always in terminal panicles,
while in Petunia, the flowers are invariably axillary and soli-
tary, on a peduncle which is generally deflexed in fruit: the
ovarium as in Nicotiana, is always supported upon a 2-lobed,
not a 4-lobed disc, which conceals the short columnar sup-
port: the lobes of the disc correspond with the sutures. The
approach to Salpiglossis* is shown in the didynamous charac-
ter of its 4 longer stamens, which are inflected towards each
other in 2 unequal pairs, while the fifth always shorter, though
generally fertile, is sometimes sterile, and is placed between
the longer pair. The style is likewise declinate at its apex,
and the stigma though it bears much analogy in its structure
to that of Nicotiana, is doubled into a somewhat bilabiate
form. The difference between this genus and Mierembergia
has been already pointed out (p. 89). The following is
offered as an emended generic character.

Petunia: Juss. — Calyx tubulosus, 10-nervis, 5-partitus,
laciniis spathulatis subfoliaceis. Corolla hypogyna, infun-
dibuliformis, vel subhypocrateriformis, tubo cylindraceo
vel ventricoso, limbo patenti ineequaliter 5-lobo, estiva.

* I can yet hardly venture to propose the separation of Lycium and
Petunia from the Solanacee, although more than half persuaded of the
propriety of doing so; but if in a more advanced stage of these inquiries,
more forcible reasons present themselves, I hope not to incur the charge
of inconsistency for this recommendation, and for having in the preceding
contributions followed the views of all former botanists in regard to the
classification of these two genera in the Natural System. It will be seen,
however, from what has been here demonstrated, that Petunia is more
closely allied to Salpiglossis than has been generally supposed; perhaps
the Callibrachoa of Llarve and Lexarve (which I have not yet seen) will be
found to be hardly distinct from Petunia.

108 ILLUSTRATIONS OF

tione irregulariter obvoluto-conduplicata,* subgibbosa. Sta+
mina 5, medio corolle tubo inserta, inequilonga, inclusa:
Anthere 2-lobe, imo profunde cordate, lobis ovatis longi-
tudinaliter dehiscentibus: Pollen oblongum, longitudina-
liter 3-suleatum. Ovarium subsessile, disco carnoso sub-2-
lobo stipatum, 2-loculare, placentis centralibus dissepi-
mento adnatis, multiovulatis. Stylus apice compressus,
incrassatus, subincurvus. Stigma obliquum, compressum,
truncato-2-lobum. Capsula calyce persistente testa, 2-locu-

_laris, imo septicido-2-valvis, valvis indivisis placenta cen-
tralidemum solutis. Semine plurima, mimuta, subspherica
vel ovata, hilo ventrali, testa reticulato-foveolata, costis
intermediis elevatis. Embryo in axi albuminis carnosi,
apice subcurvatus, radicula infera tereti fere recta cotyle-
donibus parvis oyatis subcompressis 3-plo longiori.—Suf-
frutices vel herbe Austro-Americani subviscosi ; foliis alter-
mis, integerrimis, floralibus geminis, pedunculis extra-axil-
laribus, solitariis, \-floris, demum deflexis ; corolla albida vel
violacea,

1. Petunia nyctaginiflora, Juss. Ann. Mus. 2. 215. t. 47. f. 2.
— Bot. Mag. t. 2552. Sweet Fi. Gard. 2.119. Tratt. tab. 72.
—Nicotiana axillaris, Lam. Ill, 2237 ;—herbacea diffusa,
villoso-glandulosa, viscida: foliis inferioribus _alternis,
ovato-oblongis, obtusis, pubescentibus, basi in petiolum
attenuatis ; foliis floralibus sessilibus, cordato-ovatis, oppo-
sitis: pedunculis sub 3-floris, folia superantibus: tubo
corolle calycem multo excedente, limbo lato, obtuso.—
Rio de la Plata.

This plant is so well known in our gardens, that it is need-
less to add any particular description.

* A mode of estivation (as seen in P. violacea) to approach near Sal-
piglossis ; its definition may perhaps be thus extended, nempe; lobis om.
nibus medio conduplicatis, plicaturis utrinque versus superiorem torsis,
hujusce omnino interioris marginibus in flexuris contiguis utringue obvo-
latis, unici lateralis aut interdum inferioris omnino exterioris marginibus
imbricatis, ceterorum torsive imbricatis et vicissim obvolutis, sed duorum
inferiorum dimidiis proximis gibboso-inflatis, inferne voluto-plicatis,
superne autem simpliciter imbricatis: this complex mode of zstivation is
shown in Plate 24,

SOUTH AMERICAN PLANTS. 109

2. Petunia propingua (n. sp.) ;—herbacea, viscido-pilosa, cau-
libus plurimis prostratis, demum adscendentibus: foliis
lanceolatis, acutis, basi in petiolum longum spathulatis,
floralibus consimilibus, oppositis: pedunculis solitariis
elongatis, calycis laciniis ineequalibus, linearibus, erectis,
tubo coroll calycem multo excedente, limbo campanulato,
5-lobo, lobis brevibus, rotundatis.—Buenos Ayres. v. v.
This plant is very near the former, and probably may

be only a variety, but it is certainly different in ap-

pearance, and in the shape of its floral leaves: it seldom ex-
ceeds a foot in height, and is less pubescent, the stems being
rounded, and slightly striate: the leaves are smaller, being

(inclusive of the lengthened petiole) only 12 lines long and 3

lines broad: the pedicels are 2 inches long, the calyx, 6 lines,

divided half way down into 5 unequal, linear, obtuse, erect,
and not foliaceous lobes: the tube of the corolla is 2 inches
long, slender, cylindrical in the lower portion, funnel-shaped
above, swelling into a bell-shaped mouth, the lobes of the
border being short and rounded, the whole is of a dull white
colour, slightly pubescent, and marked with 5 longitudinal
purplish lines: the stamens, which are included, are of
unequal length, one being shorter than the rest : the capsule
is little more than half the size of that of the former species.

3. Petunia violacea, Lindl. Bot. Reg. tab. 1626. Hook. Bot.
Mag. t. 3556. Salpiglossis integrifolia, Hook. Bot. Mag. t.
3113. Nierembergia pheenicea, Sweet. Fi. Gard. 2, t. 193 ;

_ —prostrata, pilis viscidis vestita, v. lanuginosa; foliis
ovatis, acutis, basi in petiolum attenuatis; floribus axil-
laribus, pedunculatis ; calycis laciniis lineari-spathulatis, ob-
tusis, foliaceis; corolle magne ventricose roseo-purpure®
limbo irregulari, lobis ovatis acutis.—Bonaria.

This plant, now so extensively cultivated in our gardens
on account of its great beauty, scarcely requires any observa-
tion, but it may be remarked, that in the native species the
leaves are much smaller than in the cultivated specimens.
Several plants in Sir William Hooker’s herbarium, collected
near the Rio Parana by Tweedie, which appear of the same

110 ILLUSTRATIONS OF

species, have leaves not more than 8 lines long by 2 lines

broad, while others from the Uruguay have leaves as large as

those of our gardens, measuring about 1} in. long, including
the petiole, and 43 lines broad.

4. Petunia intermedia, Don. Nierembergia intermedia, Grah,
Ed. Ph. Journ. 14, 175, Sweet Fl. Gard. 3, t. 237. Salpi-
glossis linearis, Hook. Bot. Mag. t. 3256 ;—herbacea, pros-
trata, pilis glanduloso-viscidis undique tecta ; foliis lineari-
bus, obtusis, dense pubescentibus ; corolle tubo vix calyce
longiori, luteo, striato, fauce ventricoso, limbi laciniis
emarginatis, puniceis; stylo clavato; capsula costata.—
Bonaria.—v. s. in herb. Hooker.

The leaves are from 4 to 8 lines in length, 1 line broad,
and somewhat spathulate at base.

5. Petunia elegans (n. sp.) ;—suffruticosa, pilis articulatis vis-
cidis undique dense tecta: caulibus adscendentibus, de-
lapsu foliolum basi cicatrisatis, ramosis, superne foliosis ;
foliis confertis, linearibus, obtusis, margine revolutis, sub-
tus canescentibus, in petiolum brevissimum cito angus-
tatis, demum deflexis; floribus axillaribus pedunculatis,
corolle roses tubo ventricoso calycem 3-plo excedente.—
Minas Geraes, Brasiliz.—v. s. in herb. meo et Hook.

var. 6, ramis erectis, elatioribus, virgatis, foliis erectis, ad-
pressis.—Minas Geraes, (Claussen) in herb. Hook. ;
This is a very distinct species, the stem and branches

being decidedly woody and of hard texture; the leaves are

from 6 to 10 lines long, 1-1} line broad, densely covered
with short viscid pubescence, which is incanescently tomen-
tous beneath. The peduncles are 6 lines long, and become
deflexed on the fall of the corolla: the calyx is 3-4 lines
long, cleft into 5 unequal linear obtuse teeth; the ‘corolla
is about 9 lines long, 6 lines broad across the mouth, the
tube being contracted below, where it is yellowish, above it,
is funnel-shaped and yentricose, of a deep rose colour, and
veined, the border being divided into 5 ovate rounded lobes ;
the stamens are unequal, 4 of the anthers approximate in
didynamous pairs, while the fifth is intermediate between the
longer pair, and is much shorter: the ovarium is obovate

SOUTH AMERICAN PLANTS. lll

smooth, the supporting column being enveloped by a
fleshy annular 2-lobed disc, which invests the base of the
ovarium: the style is compressed, 2-grooved, and declinate
at summit; the stigma is flattened, truncated, somewhat
tubular, with an erect edge, containing within its cavity 2
globular, papillose, viscid glands. The capsule is small,
obovate, smooth, and enclosed within the persistent calyx,
2-celled, 2-valved, each valve being entire, notched at the apex,
with sutural, scarcely introflexed margins. The seeds are
oblong, rounded, with the hilum on its ventral face, dark
brown, reticulately areolar ; the embryo, enclosed in fleshy al-
bumen, is terete, and slightly curved at the junction of the
cotyledons with the radicle, the latter being twice the length
of the former, which are slightly compressed, somewhat
broader, and ovate.*

6. Petunia parviflora, Juss. Ann. Mus. 2, 216, t. 47, f. 1.
Nicotiana parviflora, Lehm. Gen. Nicot. Mon. 20. Lin-
dernia Montevidensis, Spr. Syst. 2,769 ;—herbacea, pros-
trata, villoso-glandulosa, viscida, radice humifusa, caulibus
plurimis ramosis; foliis subfasciculatis, lineari-oblongis,
obtusis, crassiusculis; pedunculo brevissimo ; calycis la-
ciniis ineequalibus, spathulatis; corolle tubo calycem fere
-duplo excedente, fauce intus prominente, limbi segmentis
obtusis.— Bonaria. v. v.

I found this plant growing in moist places in the neigh-
bourhood of Buenos Ayres; its root is suffruticose, spreading
into many branching prostrate stems, which are of purplish
hue, rugous and spreading at the axils ; the leaves are often
fasciculate, linear, obtuse, spathulate, rather fleshy and vein-
less, covered with many rugous tubercles, and beset with
short, close, downy, viscid hairs ; they are 4 to 6 lines long,
and 1 line broad; a fleshy, raised gland is seen on each side
of the petiole in the usual place of stipules; the peduncle is
short, axillary, 1 line long ; the calyx is 3 lines in length, and
cleft nearly to the base into 5 unequal, spathulate, obtuse, erect
segments, which are fleshy and covered with glandular hairs ;

* A representation of this species is given in Plate 24.

112 ILLUSTRATIONS OF

the tube of the corolla is nearly cylindrical, 5 lines long,
yellow, and marked outside with 5 purplish lines, the border
is cleft into 5 rather equal roundish segments, each with a
purplish central line and a mucronulate apex, it is of a lilac
colour, and somewhat oblique, the upper lobe being more erect,
and the lower ones somewhat more reflected, the throat, of a
deep lilac hue, is compressed and hollowed imward on the front
side, in a somewhat personate form ; the stamens are unequal,
the 5 filaments arising from the contraction above the base
of the tube, are white and compressed, the short sterile
stamen is seen below the ringent compression of the tube,
between the front and longer pair, while the upper pair is
somewhat shorter, the fertile anthers, which approach each
other by the inflection of the filaments, are 2-celled, ovate,
the 5th sterile one being smaller, erect and castrate. The
evarium is oblong and ‘smooth, 2-celled, laterally grooved on
each side, and seated on a somewhat 2-lobed, annular, fleshy
disc, the lobes corresponding with the grooves and the
edges of the dissepiment; the style is slender, greenish,
2-grooved and compressed at the summit, where it is de-
flected towards the upper pair of anthersin the mouth of the
tube, the stigma is somewhat clavate, or rather 2-labiate,
with a thin expanded margin, enclosing 2 large papillose
viscid glands. The capsule is small, obovate, smooth, en-
elosed within the persistent calyx, 2-celled, 2-valved, with a
notch in the apex of each valve, where it splits half way
down by pressure, The seeds are attached by a point on
their ventral face to the thickened portion of the dissepiment
that becomes free and is parallel with the valves : they are
numerous, light brown, and oval, the surface being divided
into rather large, deep areoles, separated by straight prominent
ridges. The embryo is cylindrical, slightly curved, the radicle
is somewhat thick in proportion to its length, the cotyledons,
scarcely broader than the radicle and half its length, are some-
what compressed and ovate.*

* A figure of this species, with sectional details, is represented in Plate
23 of this work. :

SOUTH AMERICAN PLANTS. 113

7. Petunia ovalifolia (n. sp.) ;—suffruticosa, subglabra ; foliis
obovatis, erectis, utrinque scabrido-punctatis, margine sca-
brido-ciliatis, sessilibus, rachi prominenti basi valde tumido,
supra venis impressis, subtus evanidis; pauciflora, pe-
dunculo axillari folio dimidio longiore, corolle tubo calycem
5-partitum vix excedente, limbo expanso dense purpureo,
interdum pallidiori, lobis ovalibus emarginatis.—Rio de la
Plata et Brasilia meridionali.—v. s. in herb. Hooker. Banda
oriental ad Coloniam, (Tweedie, n. 286). Viamonte, Prov.
Rio Grande, (rupibus siccis, Tweedie, n. 287). Rio Grande,
(Isabelle).

This is a plant differing in its aspect from all other species,
it has a woody repent root, with erect branches about 6 inches
in height, the internodes being shorter than the leaves,
which are sessile, erect, cuneately ovate, 7 lines long, and
nearly 4 lines broad, the midrib is prominent below, termi-
nating on the stem in a tumid pulvinate swelling, they are
covered on both sides with short, rigid, spreading pubescence,
the margins ciliate; the pubescence of the peduncle and
calyx is similar, but often terminating in a viscid gland;
the peduncle is 9 lines long, the calyx 5 lines, cleft half-way
into 5 angular, somewhat oblong and obtuse segments ;
the tube of the corolla is 6 lines long, swelling towards the
mouth, with a broad spreading border nearly 3 of an inch
in diameter, divided into 5 oblong emarginate lobes of a
deepish purple colour: the capsule is smooth, obovate,
shorter than the persistent calyx that encloses it, the valves
being entire, with a notch in the apex of each.*

8. Petunia viscidula.—Nierembergia viscidula, HBK. 7, 205 ;
—herbacea, procumbens, tenuissime viscidulo-hirtella,
foliis subsessilibus, lanceolatis, vel oblongo-spathulatis ;
floribus solitariis, axillaribus, breviter pedunculatis, corolle
tubo calycis laciniis foliaceis spathulatis subeequante.—
Mexico, in hort. cult.

Nothing seems to be known of this plant beyond the

i A figure of this plant is given in Plate 24.
Q

114 ILLUSTRATIONS OF

description of Prof. Kunth, above referred to, from which it evi-
dently appears to belong to this genus, rather than to Merem-
bergia. The branches are prostrate, about one foot long; the
leaves are somewhat thick and membranaceous, six to seven lines
long, and one and a half to two and a half broad ; the peduncles
are one line long, viscidly pubescent ; the calyx has a short tube,
with obtuse, lanceolately spathulate, unequal, spreading segments ;
the corolla, of a bluish colour, has a funnel-shaped tube as long
as the calyx, pubescent, its limb with five roundish lobes; the
stamens are unequal, included; the ovarium is supported by a
small annular disc ; the style is filiform, slightly bent, as long as
the stamens ; the stigma two-lobed and papillose; the capsule,
enveloped by the persistent calyx, is two lines long, with two
entire submembranaceous valves; the seeds are minute, angular,
reticulate, with a two-lobed embryo enclosed in copious albumen.*

Stendel states that Mierembergia graveolens of St. Hilaire,
which is the NV. pubescens of Sprengel, is identical with this
species ; but their descriptions do not accord.

* Since the above was in type, it has been kindly suggested to me by Mr. Bentham,
that this species is probably identical with the Callibrachoa procumbens, Ll. and Lex,
the Salpiglossis prostrata, Hook. Arn. I am glad to find what I hinted in regard to this
plant (see note p. 107), thus soon confirmed by so learned an authority ; all the in-
formation I could glean on the subject, was derived from the short generic character,
published in Walp. Rep. vol. iii. p. 958, its specific details being omitted in the enu-
meration in p. 178 of that work. On stating my impression to Sir Wm. Hooker, he
kindly took much trouble to search for his Californian specimens, but in vain, as, un-
fortunately, they had been misplaced in his Herbarium.

[It is nearly two years since the above note appeared in print, and it is now only
(Feb. 1848,) that I have been able to see, for the first time, a specimen of the genus
Callibrachoa, from Loredo, Mexico, upon the borders of Texas; this has served to
confirm the anticipation before expressed, as, upon examination, I find this, the only
described species of the genus, is in no way distinct from Petunia parviflora, which is
fully described in a previous page (p. 111). To this last-mentioned species, must now,
therefore, be added, as synonyms, Salpiglossis prostrata, Hook, and Arn., and Calli-
brachoa procumbens, Lav. et Lex. DC. Prod. vol. ix. 462,

T avail myself of this opportunity of observing, that I have since had an opportunity
of examining living specimens of Petunia and Nierembergia, and feel convinced that
these genera are so very intimately allied to Salpiglossis, that they must all be referred
to the same Natural Order, for, independently of their somewhat bilabiate corolla, with
unequal and almost didynamous stamens, of which full details are exhibited in plates

SOUTH AMERICAN PLANTS. 115

SCLEROPHYLAX.

The plant upon which this genus is proposed to be established,
was found by me during my rapid journeys across the Pampas,
from Mendoza to Buenos Ayres, in 1825 and 1826, but I could
not examine its details until 1827, when I was first able to observe
the results of the present analysis. It is of a prostrate, succulent
habit, resembling much that of a Zetragonia, more especially as
the drupaceous covering of the seed becomes ligneous and spi-
nescent, owing to the enlargement and tumescence of the calyx,
which finally encloses the capsule. My attention having again
lately been directed to this anomalous plant by Sir William
Hooker, at the suggestion of Prof. Arnott, who had noticed it in
the collection of Doctor Gillies, I was induced to examine the
specimens existing in the Herbarium of the former distinguished
botanist, which I found to constitute two other species, distinct
from that of my own collection. These plants are certainly very
curious in their structure, and cannot be referred to any known
natural order. Their leaves are geminate, as in the Nolanacea,
and they resemble in their fleshy and prostrate habit, many of the
plants of that family, with which also the structure of their flowers
corresponds, although these are very small and inconspicuous,
approaching in size and form to those of Petunia parviflora, which
18, 28, and 24, the peculiar imbricate mode of estivation of the flower, will be seen
to be similar to that of Sa/piglossis, and therefore quite different from the induplicato-
‘valvate manner of priefloration, existing universally throughout all the Solanaceae, if
we except Hyoscyamus and Iycium. This latter genus, it appears to me, should also
be referred to Scrophulariacee, rather than to Solanacee. In Capraria, especially
the Xwaresia biflora of the Flora Peruviana, we have an instance of a regular gamope-
talous corolla, with five equal stamens, as in Zycium. So much, indeed, does that
plant resemble this last mentioned genus, that Sir William Hooker (Bot. Misc.
vol. ii. p. 231) referred it doubtfully to Witheringia, remarking that its twiggy, glabrous
stem, and long, narrow, lanceolate leaves, gave it much the appearance of a Lycium.
The baccate fruit of this genus cannot be held as a reason fur its exclusion, as several
similar instances occur among the Scrophulariacee ; viz.—Duboista, Halleria, Leu-
cocarpus, and Teedia. Hemiphragma, a genus closely allied to Capraria, has also a
baccate fruit, although subsequently dehiscent, and many other similar cases of this
kind occur among the Scrophulariacee, We have also in Verbascum, a nearly regular
five-lobed corolla, with imbricate eestivation, and with five stamens, generally all fer-

tile, and for the most part equal, as in Lycium.] : f
Q

116 ILLUSTRATIONS OF

I have described on a former occasion, p. 111. plate 24; for the
tubular corolla is, in like manner, ventricose on one side, with a
somewhat five-lobed, companulate and slightly bilabiate border,
and it offers quite the induplicato-valvate estivation of the Ces-
tracea and Solanacee. The calyx has a very short, fleshy tube
with five unequal, erect. segments, two of them being reduced to
the size of small teeth, while the other three are more or less half
the length of the corolla; in two species these are foliaceous and
singularly veined, in the other they are subulate and fleshy ; the
tube of the calyx enlarges and becomes intumescent and bony, as
the fruit advances to maturity; and in the last-mentioned instance
the calycine lobes in like manner are at length converted into
spines. The structure of the seed, however, is quite at variance
with that of the Nodanacee, approaching nearer that of the Myo-
poracee ox Ehretiacea, for the drupaceous calyx encloses an in-
dehiscent, two-celled carcerule, with a single seed suspended from
the summit of each cell, the almost straight and inverted embryo
being nearly the length of its fleshy albumen, and having a small
superior radicle with two oblong, compressed cotyledons. I have
called the genus ScLEROPHYLAX, from oxAnpos, durus and dvda€, car-
cer, because of the manner in which the seed-vessel becomes incar-
cerated by the singular enlargemont and bony intumescence of the
calyx. The following generic character will explain its structure.

ScLEROPHYLAX. (gen. nov.).—Calyz 5-partitus, tubo 5-gono
brevissimo, laciniis 2 vel 3, elongatis, triquetris, subulatis, aut
interdum expansis, foliaceis, subcarnosis, alteris brevibus, fructifer
auctus. Corolla hypogyna, gamopetala, tubo infundibuliformi,
sub faucem contractam superne ventricoso, limbo brevi, 5-plicato,
subcampanulato, sub-bilabiato, labio superiore 3-lobato, inferiore
2-lobato, lobis omnibus sequalibus, brevibus, obtusis, sestivatione
induplicato-valvatis. Stamina 5, tubo corolle inserta, inclusa:
Jilamenta inequalia, gracilia, paulo dilatata, uno breviore, alteris
subequalibus, apice incurvato-declinatis: anthere 2-loculares,
ovate, basi cordate, in sinu affixee, connectivo nullo, rima longi-
tudinali extus dehiscentes. . Pollen ovatum, longitudinaliter
3-sulcatum. Ovarium superum ovatum, conicum, 2-loculare :
ovula in loculis solitaria, apici appensa, anatropa. Stylus fili-

SOUTH AMERICAN PLANTS. 117

formis, longitudine staminum, apice inflexus. Stigma sublaterale
subliguliformi-capitatum. Fructus e calyce incrassato et demum
indurato nucumentaceus, lignosus, turbinatus, 5-gonus, vertice
depresso, et stylo perforato, angulis inzequaliter elongatis, spinis-
que 2-8 longis interdum terminatis. Carcerula omnino inclusa,
libera, chartacea, indehiscens, 2-locularis, loculis monospermis
‘(ano antico, altero postico). Semen inversum, obovatum, ex
apice pendulum : testa tenuis, chalaza apicali rapheque longi-
tudinali sublaterali notata : embryo in axi aléwminis carnosi paulo
incurvatus, cotyledonibus oblongis, compressis, crassiusculis, radi-
cula brevi tereti supera, 2-plo latioribus, et 3-plo longioribus. Her-
bee prostrate, Americe intertropice indigene ; caulibus plurimis,
angulatis, flecuosis, divaricatim ramosis : foliis geminatis, spathu-
lato-oblongis, cum petiolo continuis; floribus azillaribus binis,
(1 pracociore], insertione petiolorum fere sessilibus, ebracteatis ;
fractibus deflewis, plerwmque ad awillas deformatim concretis, nodos
tumescentes et spinescentes formantibus.

1. Sclerophylax spinescens : prostrata, caulibus humifusis, dif-
fusis, nodis spinescentibus : foliis geminatis spathulato-oblongis,
subcarnosulis, glaberrimis, eveniis, margine tenul integris ; flori-
bus folio florifero brevioribus, omnino glabris, calycis lobis tri-
quetro-subulatis, carnosulis, angulis membranaceis demum spi-
nescentibus.—Arroyuelo de San Jost, Prov. Cordove, in uligi-
nosis salitrosis. v. 0.

This species was found by me in the locality above quoted,
growing abundantly on the margin of saline swamps, and is pro-
bably diffused over the Pampas in similar situations, as I find in

tomously ramifying at each axil, spread out to the length of about
eighteen inches ; the stems are angular and herbaceous ; the leaves,
including the petioles, are about the length of the internodes, one
and a quarter, to one and a half, sometimes two inches ; they are
oblong, scarcely acute at the apex, tapering towards the base into
a petiole of the length of the blade, which is three lines broad ;

118 ILLUSTRATIONS OF

they are somewhat fleshy, the main rachis, as well as a very few
nerves, are remarkably tortuous, and are quite veinless, or, at
least, the veins are so deeply immersed as not to be visible. The
leaves of the younger axils, at the period of flowering, are scarcely
longer than six lines, and the flowers do not exceed three lines in
length. The fruit, which is rarely free, is quite turbinate, with a
thin fleshy pericarp, investing a hard bony nut of similar form, four
lines long, depressed and disciform at the summit, the angles
being terminated by sharp spines, of which three are erect, and
nearly as long as the body of the fruit: this encloses a small oval
carcerule, or indehiscent, two-celled, chartaceous capsule; the
single suspended seed which fills each cell, is two lines long, and
is very slightly incurved, tapering to the summit. The most
remarkable anomaly attached to this plant, is the spiny intumes-
cence of the axils : this is nearly a constant character, and is only
wanting in the few instances where the short peduncle of the
flowers and fruit remain perfect and free; in most cases, owing
probably to the operation of insects, the two nuts formed at each
node, become deformed and absorbed into the axil, which, in con-
sequence, swells, and forms a many-spined, salient, knotty, and
prickly jomt. On opening these, I have always found the grub
of an insect, which has generally destroyed one of the seeds. This
character is not singular, for Chamisso and Schlechtendahl de-
scribe a plant from Mexico (Gongylocarpus rubricaulis, Linn.
5.558), belonging to the Onagracee, where the drupaceous fruit
in like manner, and probably from a similar cause, becomes con-
crete with’each axil, which hence assumes a swollen and deformed —
appearance,*

2. Sclerophylax Arnottii : nana, prostrata, ramulis paucis
brevibus: foliis spathulatis, sub-3-lobatis, lobis lateralibus rotun-
datis, subdeltoideis, apice obtusiusculis, mucronulatis, fere eveniis,
margine membranaceis, basi in petiolum longum linearem attenu-
atis: floribus parvis, calycis lobis inzequaliter foliaceis, in fructu
persistentibus, nervis 3 parallelis retrorsum anastomosantibus

* A figure of this plant, with full generic details, is shown in Plate 25.

SOUTH AMERICAN PLANTS. 119

notatis, corolleque tubo brevi, ventricoso, imocoarctato, sca-
brido-pilosis—San Juan, Prov. Argentin-—v. s. in Herb Hook.
(Gilhes.)

This is very similar in habit to the former species, but from
the solitary specimen I have seen, it appears altogether more
diminutive, the branches extending only three or four inches
in length: the leaves, however, are larger in proportion ; the
blade is broadest at the base, contracted in the middle, and
terminates in a narrow, obtuse, and mucronulate apex; at base,
it tapers gradually into a narrow linear petiole of equal length,
being altogether one inch long, and three and a half lines broad :
they are thick and fleshy, and without any apparent venation ;
the flowers are five or six lines long; the corolla is broader in
proportion, and, as well as the calyx, is covered with short rigid
jointed pubescence: the calycine segments are broad, foliaceous,
oblong, pointed, three of them being half as long as the corolla ;
they are very distinctly veined, with three almost parallel nervures,
connected together by several retrorsely branching veins: the tube
thickens, as in the preceding species, into a hardened nut-like
body, which, in like manner, becomes conglomerated with the
axillary node; it is crowned with its persistent foliaceous lobes,
which, however, do not become spinescent, as in the former
species. The structure of the flower, the stigma, and the seed,
exactly resemble that of S. spinescens, except that the stamens
are in some degree shorter, the anthers scarcely rising above the
middle of the tube of the corolla.*

3. Sclerophylax Giddiesii: planta rigidior, prostrata, caulibus
crassioribus, angulatis, flexuosis, nodis valde tumidis: foltis
geminatis, spathulato-rhomboideis, nervosis, utrinque glaberrimis,
nitidis, nervis venisque prominentibus, in petiolum brevem latum
attenuatis : floribus folio florifero longioribus, calyce 5-gono, tubo
brevissimo, lobis insequalibus, foliaceis, lineari-lanceolatis, fruc-
tifero valde aucto: corolle tubo paulo ventricoso, glabro, genitali-

* This species is figured in Plate 26 a.

120 ILLUSTRATIONS OF

bus duplo longiore, limbo 5-lobo expansiore sub-bilabiato.—Rio
Diamante, Prov. Mendoze Argentin.—v. s. in Herb. Hooker
(Gillies).

This plant is very distinct in its habit from the two former
species, the stem being much thicker, far more flexuose and
angular, with more distant internodes, the petiole and part of the
blade of the leaves, together with the ripening fruit, being often
confluent with the axils, which are much more swollen, the
petioles in such cases becoming confluent with, and their margins
decurrent on, the angles of the stems; the petiole is shorter and
broader than in either of the former species. The leaves, including
the petiole, are nine lines long, and three lines broad; and unlike
the two former species, they are marked with distinct nerves and
veins, which are especially prominent below. The tube of the
calyx is short, but its border is divided into five large, broad,
foliaceous leaflets, which are somewhat unequal in length, two of
them being one-third of the length of the flower. The corolla is
far more slender and infundibuliform than in the two preceding
species, and is altogether seven lines long, the tube being quite
glabrous, and rather ventricose above ; its border somewhat bila-
biate, is divided into five equal, short, obtuse lobes. The stamens
are unequal in length, the two longer ones scarcely reaching the
middle of the tube of the corolla, and the fifth shortest is not
declinate at the apex as the four others. The fruit, in every
instance I have seen, becomes enclosed in the tumescent axil ; the
tube of the calyx enlarges, and becomes converted into a hardened
ligneous covering, which is crowned by its persistent foliaceous
lobes : the seed in its structure differs in no way from that of the
two preceding species.*

The plants just described, cannot be referred satisfactorily to
any known natural order. They resemble Nolanacee, Ehretiacea,
Convolvulacee, and Solanacea, in their tubular corolla, with five
included stamens, and more especially the latter in the indupli-

* A drawing of this species is shown in Plate 26 8.

SOUTH AMERICAN PLANTS, 121

cated ‘estivation of its border, but they differ from all these
families, by having a two-celled ovarium, with a solitary ovule
suspended from the summit of each cell, and in having a nearly
straight embryo, with superior radicle. The approach to Nolanacee
is more evident, by their being in like manner prostrate or strag-
gling succulent plants, growing in saline moist places, by their
geminate, spathulate, fleshy leaves, with expanded petioles, one of
which is always inserted laterally a little higher than the other
upon the stem, to the salient angle of which one of their edges
is generally decursively continuous; they have also a solitary
flower at the origin of each petiole. They differ, however, from
that order, in having a single two-celled pistillum, not distinct
ovaria, for Nolana, and most of its congeners, have always several
distinct gynobasic carpels, generally one-celled, but some of these
are often united without regularity into two or many-celled nuts,
which, in such cases, never present more than a single ovule in
each cell. Grabowskya, which I have referred with some hesita-
tion to Nolanacee, but which probably represents the type of a
distinct suborder, exhibits a similar tendency to form spines at
the axils, and presents also a single pistillum, terminated by a
lengthened style, and two two-celled nuts, each with a solitary
ovule, but here, as in the true No/anee, the embryo is nearly
annular, with the radicle pointing to the basal hilum. The group
of plants in question appears to differ from Nolanacea, exactly as
the Myoporacee are held distinct from Verbenacee; viz., by
having a somewhat bilabiate corolla, and a superior, instead of an
inferior, radicle. From the Serophulariacee they are distinguished
by a very different zestivation of their corolla, and more particu-
larly by a totally different structure of the ovarium and seed, in
which latter respect they also differ from the Solanacea, notwith-
standing that they much resemble this order in the shape and
estivation of the corolla. They certainly approach, in many
respects, to the Myoporacee, (especially through Disoon and
Nesogenes with their bi-locular, 1-ovulate ovaria), with which
Order they agree, in their somewhat bilabiate corolla, and m
having suspended ovules and albuminous seeds with a straight
R

122 ILLUSTRATIONS OF

embryo and superior radicle ; but they differ in the estivation
of the corolla, in possessing five, instead of four stamens, in their
anthers being two-celled, with longitudinal dehiscence, in their
leaves being geminate, not opposite, and in their fleshy herbaceous
habit, not having ligneous erect stems.

To the Stilbacee they also appear to offer some approach, on
account of their tubular calyx with unequal teeth, their funnel-
shaped corolla with a bilabiate border, having an induplicate
estivation, and a superior two-celled ovarium, with a single ovule
in each cell: but this is erect, not suspended. They have also a
slender capsule enclosed in the persistent calyx, and although it is
two-celled, and monospermous in each cell, the seed is erect, and
the embryo has an inferior, not a superior, radicle. They differ in
many other respects, and are altogether extremely different in habit.

.There are many analogous points of structure common to Zrapa
and Selerophylax that should not be lost sight of. In the former,
the calyx, though only half inferior, enlarges in like manner in
fructification, entirely grows over the ovarium, and finally becomes
enlarged and lignescent, the lobes being also converted into spines.
The corolla, although consisting of distinct petals, offers a plicato-
valvate estivation. The ovarium is two-celled, with a single ovule
suspended in each cell. Here, however, the analogy ceases, for
in Trapa, by the abortion of one of the ovules, the fruit becomes
one-celled, with a single exalbuminous seed, and although the
radicle is superior, the embryo, from the diminutive suppression
of one of its lobes, becomes pseudo-monocotyledonous, added to
which, the habit of the plant is quite distinct, and its alliance
very remote,

To Tetragonia, as I have before observed, there is certainly much
apparent resemblance, but it is altogether external, for notwithstand-
ing the similarity of its habit, and the spiny intumescence of its
fruit, there exists no analogy whatever in the structure of the
flower, or of its seed, to that of Sclerophylax.

On a former occasion (see page 67,) I have endeavoured to
trace the relationship of the Borraginee to the Convolvulacee,
through the intermedium of Nolanacea and the Dichondree,

SOUTH AMERICAN PLANTS. 123

on account of the gynobasic insertion of the carpels, but
the transition is now more distinctly visible and gradual,
through the medium of the Hhretiacea, this new group, and the
Nolanacee.

This affinity of Sclerophylar (having suspended ovules) with
the Nolanacee and Borraginee, (having gynobasic carpels and
erect ovules,) it must be confessed, does not, at first sight, appear
so evident as will be seen on further enquiry. In this con-
sideration, one feature should be constantly borne in mind, I
mean that of the relative position and mode of attachment of the
ovules: in most cases analogous to the present one, (i. e. where
the radicle of the embryo points towards the hilum,) these may
vary either in having a superior pomt of suspension, an axile
attachment, or a basic origin,—differences that really amount to
little else than the relative height of the poimt of adhesion of the
carpels, or that terminal summit of the gynobase, where its
nourishing vessels, proceeding from the torus, penetrate the walls
of the ovaria, and which can always be distinguished from the
fertilizing vessels proceeding from the style. These several con-
ditions. have been ably explained by M. Aug. de St. Hilaire, in
his admirable paper on the gynobase (Mem. Mus. 10, p. 131.)
Following up this view of the case, there will not be found so
great an amount of discrepancy in the structure of the seed of
Sclerophylax, and that of the various genera included in the
orders above mentioned ; for, in examining the dissepiment of the
seed of this genus, the gynobasic vessels (as might be expected)
are seen as a distinct rachis along its central axis, terminating in
_ the point of suspension of the ovules, and presenting an instance
somewhat analogous to that which St. Hilaire calls an elevated
~ gynobase. In Nolanacee and Borraginacea, where generally there
exists, on the contrary, a very depressed gynobase, it is the style
that is seen in an analogous position, as a rachis in the central
axis of the carpels, in consequence of the ovaries having an en-
tirely basic attachment : in these two extreme cases, the embryo
is alike seen in the axis of the albuminous seed, with the radicle
directed to the point of its attachment. Even in the Order Bor-

R2

124 ILLUSTRATIONS OF

raginacee, where, in most instances, the gynobasic point of union
of the carpels is generally on the level with the gynophorus itself,
several instances occur, (in Asperugo, for instance,) where the
apical point of the gynobase is mid-way, or near the summit of the
axile line of juncture of the carpels, at which point they are in
fact pendulous. In other cases again, this point is at the very
summit of the carpels, as in Mattia, Pectocarya, and others of
the tribe Cynoglossea, where the ovaries, at first pendulous, at
length, after development, exhibit their carpels in an absolutely
centrifugal position upon the summit of the gynobasic point of
their attachment.

It is evident, from the foregoing facts, that Sclerophylax
cannot be referred to any known Natural Order, and it is, there-
fore, essential to find some place for it in the system. Some
objections may be made to the establishment of a distinct order
upon a solitary genus ; but we have at present no less than twelve
natural families among phanerogamous plants, each based upon
a single genus. Even Nolana was in a similar position, with
only seven species, when the family of the Nolanacee was first
proposed in 1833.: the subsequent collections of Cuming and
Bridges have increased the number of genera to six, and the
amount of species to thirty. Under these circumstances, I have
less hesitation in offering the genus under consideration, as the
type of a distinct family, under the name of Sclerophylacea; and
accordingly, I proceed to suggest the position it will probably
occupy in the natural system, an inference derived from the com-
parison of its leading characters with those of the various families
to which it can claim the smallest relation.

In the following tabular view, the various orders there enume-
rated, which form a very natural circle, bound together by many
common ties, are placed in juxtaposition according to the number
of the stamens, the zstivation of the corolla, the number and direc-
tion of the ovules, and the relative position of the embryo. This
selection of characters may not be the most appropriate with a view
to methodical arrangement, and is not offered with any such inten-
tion ; but it answers our present purpose of determining, by such

SOUTH AMERICAN PLANTS. . 125

artificial means, the most fitting position in the system for Sc/ero-
phylax, which on account of its apparently anomalous structure,
does not at first sight fall into any distinct place, and can hardly
be attached as a suborder to any of the families here enumerated.
This table, founded upon such artificial characters, appears to
indicate by a gradual transition, a chain, nearly as perfect as any
linear distribution, based upon more methodical principles, can be
expected to exhibit, and certainly it does not materially differ
from the most approved arrangement after the method of Jussieu.

ILLUSTRATIONS OF

126

Mowers with a gamopetalous hypogynous corolla, and one or more superior ovaries with
one is always posterior, the other anterior with respect to the axis of inflorescence.

1, 2, or 4 cells in each, never 3 or 5; placentee never parietal ; when the cells are 2,
They consist of the class Nuculifera, and part of Tubifora, End,

Orders. Stamens. €stivation of corolla. Ovules. to A cers he ov ap
Q imbri :
Labiatze } 4 aes. inegats ee CY : } erect inferior =} inferior
Borraginacese * . . , . ‘ . .
a 7 ) border below plicate, . q
Cordiaceze ath, | : above, cortortu-im! imbricate . ° ; a : Z
border, below plicate : > superior superior
Le & } @ © conduplicate and plat valat > ——-1 in each cell —, |
Ehretiaceze ‘ : : t :
Sclerophylacese , } € 3 lobes, induplicato-valvate : a ae ea ae ‘
Grabowskyes : “ 7 Ns eee
tube oeboptheate- Adios often condu i
Nolanacese , ; DS on cate—margins always contortu-im- ’ . . ,
Pid bricate ' ; :
ichondrese . seg , ; : ; : : ;
uae 2 5 Juplicat duplicat 1 or 2 in each cell |
lobes valvate inferior
Convolvulacere : @ & XA and twisted, or eeduplionto-valvate : erect
Erycibese F i and straight . : } 4 é a single cell : 5 ‘
Cuscutaceze i } €: almost valvate . ' ' et} 2 in each cell ‘ . ‘ .
Stilbaceee . | ‘ : : ; : } lineachcell 4 . ‘ es es
- inferior 4
Cestracese : 4 Oe eS induplicato-valvate or plicato-valvate a ; ‘ ' .
Solanaceze , : J 3 f ‘ : ‘ : ‘. . '
» centripetal
Nicotianeze 4 : : ; | horizontal . ‘
5X plicato-imbricate several in each cell
Salpiglossidese J , : : 5 . . .
Scrophulariacese 7 i ‘ : ‘ ‘ , : hee Z : ae
Selaginaceze ‘ , : : ’ ‘ é : . ‘
Verbenacese , 4 { y imbricate @ : ; } 1 in each cal ; Ee cae : inferior
; : > suspended
Avicenniese ‘ : . 4 ’ * . ’
Myoporaceze J : : Be 1, 2, or 4 in each cell : } — } superior

SOUTH AMERICAN PLANTS. 127

CYPHOCARPUS.

The discovery of a plant possessed of many abnormal charac-
ters, is always more interesting to the Botanist, than the detection
of a new genus, marked by features that only serve to fill up an
ordinary link in the chain of some well-recognized family. The
plant under consideration will be seen to be extremely anomalous
and curious in its structure. It was collected in Chile by Bridges,
and exists in the Herbarium of Sir William Hooker, who, with
his accustomed liberality, had the kindness to offer it to me for
examination. It evidently belongs to the class Lpicorollia, or
rather the Campanulee of Jussieu, according with the Campanu-
lacee, Lobeliacea, Goodenoviacee, Cyphiacea, and Stylidiacee, in
having an epigynous corolla and stamens alternate with its lobes :
the insertion of the stamens, however, is not epigynous, as in all
these families, but decidedly perigynous, originating in the middle
of the tube of the corolla. It corresponds also with the four last-
mentioned orders, in the corolla having an irregular border, but
it is not divided into distinct petals: its tube is not cleft on one
side to the base; nor are the stamens in any degree syngenesious,
as always occurs, at least, in the Lobehacea. From the Goode-
noviacee, it differs in the eestivation of the corolla; for, in that
order, the broadly-winged margins of each lobe respectively are
involutely imbricated upon one another,* while in Cyphocarpus
the margins are irrespectively induplicate with those of the con-
tiguous lobes and valvate with them: these lobes, too, are of one
equally thin membranaceous texture, not thickened in the middle
as if another narrower petal were glued upon the back ; it must

* This is a distinction deserving of some notice. Endlicher, in his character of
the Goodenoviacee, (Gen. Pl. p. 506) defines this by saying “‘lobis estivatione
induplicatis,” which conveys a very incorrect notion of this peculiar manner of prie-
floration, especially if we confine that expression to the limit given to it by Prof.
Lindley in his Intr. Bot. 411, fig. 6. Mr. Robert Brown, who founded the order,
expresses this feature in far more exact terms, viz., “lateribus estivatione indupli-
catis” (Prod. 573) ; but it appears to me, it would be still more correctly defined by
the following amplification: “ marginibus stivatione inter se inyoluto-plicatis,
plicaturis valyatim clausis.”

128 ILLUSTRATIONS OF

not be forgotten, however, that the hooded portion of the upper
lip of Cyphocarpus, more or less partakes of this character. In
regard to wstivation, the approach to the Lodeliacee and the
Campanulacea, is equally evident, in which latter family, although
replicately valvate in Speeularia ke , it is more generally plicately
valvate, as in Campanula €"}, a form sometimes scarcely distin-
guishable from the induplicato-valvate € 3mode of estivation seen in
Cyphocarpus. In the structure of its ovarium, it resembles at the
period of its first growth, that usually seen in most of the genera of
the Campanai alliance : it is two-celled, with numerous ascending
ovules arranged about the axis, on each side of a narrow central pla-
centary line ; but the dissepiment consists of an extremely delicate
membrane, which at an early stage begins to shrink from the walls
of the ovarium, and soon evanesces entirely, leaving a unilocular
cell, with a linear, central, free placenta, about which the
ovules are crowded, and become perfected. This placenta is
very narrow, and although thicker than the dissepiment, is still
membranaceous, being marked by six very fine parallel ovuligerous
nerves, arranged in threes, and leaving a broader intermediate
space, which is sometimes, but not always, cleft for a short dis-
tance in the middle: this shows an evident tendency towards the
placentation of the Lysipomee, especially through the genus
Hypsela, of Presl. I am not aware of the existence of a similar
structure in any genus of this alliance. It differs also from all —
the orders before mentioned, in the peculiar form of its corolla,
which is quite monopetalous and bilabiate, one of the lips of its
border being galeate, with winged margins, and surmounted bya
single terminal, delicate, oblong lobe, while the other lip is fur-
nished internally with a prominent ringent palate, and has four
distinct, terminal, oblong lobes, of delicate texture, like that of
the other lip; these five lobes have all the same common indupli-
cate estivation. The style is quite glabrous, and declinate at the
summit, and the stigma is deficient of the singular indusium of
the Goodenoviacee, although it has a few external setose hairs, as
in the Campanulacea ; it is subsequently glabrous, bilabiate, with
fleshy reflexed lobes, and a small gland in the sinus; indeed, it

PLATE XXIII.

Exhibits Petunia parviflora. Fig. 1, a flower. Fig. 2, the
corolla, seen in front. Fig. 3, ditto, seen sideways, to show
its half-formed palate. Fig. 4, the border, seen from above.
Fig. 5, a transverse section of the tube across the palate,
showing the stamens. Fig. 6, the corolla, cut open; all
magnified. Fig. 7, one of the four longer stamens with
declinate anthers, before and after dehiscence. Fig. 8, the
shorter fifth stamen, with erect anther. Fig. 9, the stigma,
with a portion of the style, seen sideways and from behind.
Fig. 10, the ovarium, with the calyx cut open to show the
supporting glandular disc; all more highly magnified. Fig. 11,
the capsule, with persistent calyx; natural size. Fig. 12,
the capsule, magnified. Fig. 13,aseed. Fig. 14, a longitu-
dinal section of ditto. ig. 15, the embryo extracted, seen
in both directions ; all highly magnified.

PLATE XXIV,

A, is Petunia elegans. Fig. 1, the calyx. Fig. 2, the corolla ;
both natural size. Fig. 3, the corolla, cut open; magnified.
Fig. 4, the stamens, before and after dehiscence. Fig. 5, the
stigma, with a portion of the style. Fig. 6, the ovarium
seated on its glandular disc. Fig. 7, longitudinal section of
ditto; all more highly magnified. Fig. 8, the capsule, with
its persistent calyx; natural size. Fig. 9, the capsule after

dehiscence; magnified. Fig. 10, a seed. Fig. 11, vertical
section of ditto. Fig. 12, the embryo extracted, seen in
both directions ; all more highly magnified.

B, is Petunia ovalifolia. Fig. 13, is the capsule after
dehiscence ; magnified.

PLATE XXV.

Shows Sclerophylax spinescens.

Fig. 1, a portion of a branchlet, with two flowering axils.
Fig. 2, a single flower ; both natural size. Fig. 3, a flower
before expansion. Fig. 4, the calyx. Fig. 5, the corolla
expanded. Fig. 6, ditto, cut open. Fig. 7, stamens, before
and after dehiscence. Fig. 8, the pistil; all magnified.
Fig. 9, the stigma, with a portion of the style; more highly
magnified. Fig. 10, an axil, showing the manner in which
the seed vessels of the two flowers become embedded in its
fleshy substance. Fig. 11, the persistent calyx, when not
absorbed in the axillary node; both natural size. Fig. 12,
two views of the enlarged and persistent calyx, entirely
enclosing the seed-vessel, with a portion of the style
exserted. Fig. 13, vertical section of ditto. Fig. 14, the
carcerule. ig. 15, vertical section of ditto, showing a seed
suspended from near the summit of each cell. Fig. 16,
upper half of the carcerule, cut away to show the seeds, each
suspended by its respective funiculus. Fig. 17, ditto, with
the seeds removed. Fig. 18, the seeds seen in two direc-
tions. Fig. 19, vertical section of ditto. Fig. 20, embryo,
seen in two positions, with the albumen taken away; all

magnified. Fig. 21, a leaf, showing its flexuose nervures;
natural size.

PLATE XXVI.

A, is Sclerophylax Arnottii.

Fig. 1, a flower in bud, natural size. Fig. 2, the calyx; .
magnified. Fig. 3, one of the segments of the calyx, showing
its venation; more highly magnified. Fig. 4, the corolla
before expansion. Jig. 5, ditto, cut open; both magnified.
Fig. 6, one of the hairs of the corolla; very highly magnified.
Fig. 7, a portion of the stamens, before and after dehiscence.
Fig. 8, the stigma and part of the style; both magnified.
Fig. 9, transverse section of the carcerule. ig. 10, vertical
section of ditto, showing the seed suspended in each cell ;
both magnified.

B, is Sclerophylax Gilliesi.

Fig. 1, is one of the axils, with its geminate flowers ; natural
size. Fig. 2, ditto, magnified. Fig. 3, the calyx. Fig. 4.
portion of a stamen, both much magnified, Fig. 5, pollen
grains, seen in different directions; very highly magnified.
Fig. 6, the stigma, with a portion of the style, J/g. 7, an
axil, with the two seed-vessels imbedded in its fleshy sub-
stance. Fig. 8,a carcerule extracted. Fig. 9, a seed ; all
natural size. Fig. 10, vertical section of carcerule, with the
seeds removed. Fig. 11, horizontal section of ditto. Fig. 12,
a seed with its funiculus in two positions. Fig. 13, embryo
extracted, seen in front and sideways ; all magnified.

PLATE XXVII.

A, isa young plant of Cyphocarpus -rigescens, to show its
nearly radical leaves, which soon disappear. 8B, is the same
plant at maturity, having lost all its leaves.

Fig. 1, is a flower in bud with its three bracts; natural
size. Fig. 2, ditto, magnified. Fig. 3, the corolla of ditto
expanded. Fig. 4, ditto cut open, to show its internal
structure; more highly magnified. Fig. 5, pistillam with its
front portion removed, showing the dissepiment beginning
to separate from the sides of the ovarium. ig. 6, the epi-
gynous disc removed from ditto; both magnified. Fig. 7,
the stigma before the expansion of the flower. Fig. 8, ditto
after its impregnation; both more highly magnified. Fig. 9,
transverse section of ovarium. Fig. 10, longitudinal section
of ditto, with the calycine lobes removed, showing the
epigynous disc surrounding the base of the style. Fig. 11,
ditto with the persistent calycine lobes and corolla, with the
dissepiment contracted, and the central placenta divided
into two parallel ribbands. Fig. 12, ditto with the dissepi-
ment contracted to a single narrow central line of placenta-
tion, the more usual occurrence, all magnified. Fig. 13, a
seed. Fig. 14, vertical section of ditto. Fig. 15, the embryo
extracted from the albumen, and seen in two directions; ail
more highly magnified.

PLATE XXVIII,

A, isa branchlet of Salpichroma hirsuta, with its flower
and its berry; natural size.

B, is a portion of Salpichroma diffusa,

C, ditto of Salpichroma tristis,

Plate 23.

I.Miers lithog. 2 Printed by Hullmandel & Walton.

Petunia parviflora.

PU Ee migra tint pee ent

Plate 24.

Printed by Hullmandel & Walton.

J.Miers hthog.

Petunia elegans.

to

Petun

Plate 29

12

Printed by Hullmandel & Walton.

i Miers hthog.

é
a
oO
fo ee
oO
eee
ord
Oy
cay
Ps
fey
—
=
5
@
ae
om)

Plate 26.

en

ee
ys

3

| o Plate 28.

:
*
e ° =
4 +
‘ ” i
+ ‘
* ee
*
é \s ¥

%

- Salpichroma trists.

a
Pie

eae
ee
: said

ee

SOUTH AMERICAN PLANTS. 129

greatly resembles that of Petunia, and is much like the develop-
ment of the stigma, which I have sometimes seen in the Chile
variety of Wahtenbergia linarioides. Tt has an entire, small,
annular, fleshy, epigynous ring, surrounding the base of the style,
as in the Lobeliacee. Its seeds are neither lenticulate, nor
winged, but oval and striated, with a somewhat scrobiculate and
reticulated testa. Its general habit is very peculiar, being some-
what herbaceous, of an arid appearance when dried, with small radical
rigid leaves, having sharp spinose teeth, while its cauline leaves
are ternate, involucrating, and surrounding the base of a solitary
sessile flower in each alternate axil, the two lateral ones being
actually inserted upon the ovarium; these resemble in form the
persistent segments of the calyx, being linear and rigid, with a few
somewhat retrorse teeth on the margin, which are hard and
spinescent, and sometimes double. In the ascendant position of
its ovules, and in the form and direction of the embryo, it resem-
bles all the other orders of the Campanal alliance.

It must be evident from the above facts, that the affinity of
Cyphocarpus is unquestionably. with the class of the Campanulinee,
but it cannot obtain a tenable place in any of the five orders com-
posing that class,* for which reason I would rather suggest the
propriety of giving it a distinct station, and making it the type of
an aberrant group, of which, probably, many others remain to be
discovered, or may now, perhaps, be found in existing herbaria.
It certainly borders closely upon Campanulacea, through Prisma-
tocarpus ; upon Lobeliacea,t through Grammatotheca, Clintonia,

* Tfin any place, it would certainly stand as a third tribe of the Campanulacee,
but in an instance like the above, where a plant osculates closely upon several different
orders and cannot be arranged in any one of them, without breaking down the few
limits of demarcation between very natural families, it appears to me less objectionable
to classify it under a distinct title, as a separate group, than to force it into an un-
natural position. This genus may therefore remain for the present, as the nucleus of
a suborder, attached to the class Campanulinee, after the example of the Spheno-
cleacee, until other analogous plants be detected, that may claim for the Cypho-
carpacee its due place, as a recognized family in the Natural System.

+ I have noticed in many of the Cape species of Lodedza a very distinctly gibbous
palate, similar to that described in Cyphocarpus ; but strange to say, I can find
nowhere, either in the descriptions, or in the figures of any botanical work, any

7 s

130 ILLUSTRATIONS OF

and Lysopomia; upon Cyphiacea, through the genus Cyphiella
of Presl, which has a gamopetalous corolla; and upon Goodeno-
viacee, through the section Ochrosanthes of Goodenia.

The generic name of Cyphocarpus, now proposed for this plant,
is derived from kudos, iucurvus, and xapros, fructus, on account of
the gibbous form of its enlarged capsular fruit.

The following is an outline of its generic character :—

Crenocarpus. (gen. nov.)—Cualyx oblongus, ovario adnatus,
limbo supero, persistenti, breviter tubuloso, profunde 5-fido,
laciniis erectis, subinzequalibus, linearibus, retrorsim mucronato-
dentatis, rigidis, fructifer demum auctus. Corolla persistens, in-
sertione epigyna, longe tubulosa, tubo cylindrico pentagono,
angulis hispidulis, limbo bilabiato tubo duplo breviore; labio
superiore galeato, textura crassiori, (excepto nervo dorsali) glabro,
colorato, marginibus alatis tenuibus, lobo unico oblongo superato ;
labio inferiore, imo in palatum gibbosum plicato, plicis 3 linearibus
intus prominentibus, apice usque ad medium in lobis 4 oblongis
partito ; lobis omnibus 5 textura delicatula eequalibus, zstivatione
induplicatis. Stamina 5, squalia, inclusa, supra medium tubi
inserta, limbi laciniis alterna: filamenta gracilia, dilatata, medio
nervulo centrali barbata: anthere lineares, filamentorum longitu-
dinis, basifixee, 2-loculares, loculis collateralibus, margine rima
longitudinali dehiscentibus. Po//en globosum, simplex. Ovarium
inferum, cylindraceum, sub-5-gonum, membranaceum, 2-loculare,
loculis uno antico, altero postico, dissepimento membranaceo,
tenuissimo, medio placentifero, a parietibus cito soluto et evanido,
placenta centrali tune omnino libera, revere deinde uniloculare :
ovulis plurimis adscendentibus. S¢y/us filiformis, tubo corolla
paulo longiusculus, omnino glaber, apice sub galea reflexus, basi
annulo brevi integro carnoso cinctus. Stigma capitato-bilobum,
lobis crassis, in alabastro clausis et extus setosis, demum reflexis,
glabris, in sinu glandula centrali viscosa instructum. Capsuda cylin-
drica, conica, striata, postice gibboso-ventricosa, corolla calycisque
lacintis foliaceis persistentibus coronata, unilocularis, subfolli-
allusion to the existence of so prominent a feature. I have also observed in some

species of Lobelia, that the insertion of the stamens is decidedly perigynous, that is to

say, upon the tube of the corolla, a little above its base, not epigynous, as gene
described. 7

SOUTH AMERICAN PLANTS. 131

‘cularis, vel sutura longitudinali postice dehiscens, placenta nunc
omnino soluta, in teniam angustissimam centralem liberam semi-
nigeram (rarius medio fissam) cum stylo persistenti continua.
Semina plurima (circiter 40), patentia vel suspicientia, breviter
stipitellata, ovata; ¢eséa longitudinaliter costata, reticulato-scro-
biculata, apice chalaza subobsoleta notata ; albumen carnosum :
embryo axilis, teres, fere orthotropus, radicula terete, infera, hilo
spectanti, cotyledonibus ovalibus paulo latioribus, ultoties
longiore.

Herba Chilensis rigida, per totam scabrido-pilosula, caulibus
perpaucis, e collo ramosis, erectis. Folia fere radicalia, oblonga,
acuminata, basi in petiolum decurrentia, enervia, grosse spinoso-
dentata: folia caulina, terna, equalia, sessilia, quarum 2 lateralia
(bractese) e asi ovarii utrinque orta, rigida, linearia, spinoso-den-
tata, florem solitarium sessilem involucrantia, persistentia ; caulibus
tunc in inflorescentiam quasi spicatam redactis. —

1. Cyphocarpus rigescens: foliis radicalibus oblongis, grosse
dentatis, dentibus mucronato-spinulosis rachi marginibusque carti-
lagineis, rigidis, in petiolum decurrentibus, mox caducis, caulinis
bracteisque consimilibus linearibus, laciniisque calycinis runcinato-
dentatis, rigidissimis, persistentibus ; ramulis subspicatis, subflex-
uosis, e basi ortis, adscendentibus. Chile (Coquimbo): v. s. in herb.
Hooker et Mus. Brit. (Bridges, n. 1293.)

This curious plant seems to be quite herbaceous in its habit,
although of arid and harsh appearance : its root is long, slender,
and tapering: it branches from towards its base into a few nearly
erect, somewhat flexuose floriferous stems about a foot high, bear-
ing a single flower in each axil. The radical leaves, including the
petiole, are eleven lines long, and three broad: the floral leaves
and bracts are nine lines long, and about a line broad: the caly-
cine leaflets in flower, are four lines long, and scarcely a line
broad, but they increase in length to six lines upon the ripened

and enlarged capsule : the inferior ovarium is three lines, and the

superior corolla six lines long ; this is persistent, although the

border becomes shrivelled ; it is, apparently, of a bluish hne, but

the upper galeate lip is of a deep crimson colour, and the palate

of the opposite lip seems of a roseate tinge, judging at least from
$2 :

132 ILLUSTRATIONS OF

the appearance of the flower when moistened after being dried ;
externally it is quite smooth in bud, but the flower, at maturity,
is covered with a very short, dense, echinate, rigid pubescence,
with which, indeed, the whole plant, under the lens, will be found
to be more or less invested: the crimson galeate lip of the corolla,
with the exception of the dorsal nerve, is, however, quite glabrous.

DORYSTIGMA.

I am indebted to the kindness of Dr. Lindley for allowing
me to examine and define several of the following Solanaceous
plants, and I take this opportunity (April, 1848,) of repeating
my obligations to Sir William Hooker for his liberal and kind -
permission to describe the many following new species which,
during the last twelve months, I have found in his rich and ex-
tensive Herbarium.

In the Jaborosa group, and belonging to the genus above
mentioned, whose elements were defined in a former portion of
this work, p. 27, I have now to add a third species.

3. Dorystigma crispum, n. sp.; caulibus plurimis, ceespitosis ;
foliis subfasciculatis, glaberrimis, carnosulis, irregulariter pimnati-
fido-laciniatis, in petiolum longum alatum decurrentibus, laciniis
latis brevibus, mucronato et sinuoso-dentatis, uninerviis, events,
sinubus crispato-undulatis, margine subrevolutis; floribus cum
foliis in collum fasciculatis, bracteis parvis, subulatis; corolla
extus imo glabra, superne pubescente, intus fauce lanuginosa, limbi
laciniis oblongis, obtusis, staminibus fere exsertis. — Bolivia
v. 8. in herb. Lindley. (Bridges, 1846.)

This plant has very much the habit of the two species formerly
described, the leaves much resembling those of Dorystigma
squarrosum, (see p. 28, Plate 6), being nine lines broad, the
petiole is one inch and a quarter, the blade one inch and three
quarters, altogether three inches long; the peduncles six lines,
and the corolla six lines in length.

SALPICHROMA.
In order to harmonize better with the names of the two ap-

SOUTH AMERICAN PLANTS. 133

proximate genera, Jochroma and Pecilochroma, I propose to sub-
stitute that of Salpichroma for Salpichroa, the genus described in
the beginning of this work p. 1. The plants from Columbia
‘and New Grenada, there alluded to in p. 6, I now find to
belong to a new species, very distinct from Dr. Meyen’s Atropa
hirsuta, of which I have since seen an original specimen. The
number of species, to which T have still to add another, will there-
fore stand as follows.

§. Ensalpichroma 1. Salpichroma glandulosa, loc. cit.

” dependens, ib.

hirsuta, infra descrip.
ramosissima, loc. cit.
diffusa, n. sp. infra descr.
tristis, n. sp. infra descr.
rhomboidea, loc. cit.
ciliata, ib.

3. Salpichroma Airsuéa. Atropa hirsuta, Meyen. (Riese em
die Brde, vol. i. p. 466. Nees ab Esenb. Nov. Act. 19. Suppl.
1. p. 389): caule suffruticoso, ramosissimo, diffuso, ramulis mol-
liter hirsutis; foliis alternis vel in turionibus fasciculatis, ovatis, basi
obtusis, ineequalibus, apice acutiusculis, utrinque pilis articulatis
sparse hirsutis, longissime petiolatis, petiolo complanato, tenuis-
simo, ciliato, limbo 8 vel 4-plo longiore 5 pedunculo capillari pe-
tiolo breviore ; calyce hirsuto, profunde 5-partito, laciniis lineari-
subulatis, erectis ; corolla longe tubulosa, extus pilosula, tubo imo
gracili, superne infundibuliformi, calyce 4-plo longiore et petiolo
sequilongo, antheris styloque sub-exsertis ; bacca ovalis, calyce
persistente suffulta.—Peruvia, circa Pisacomam, altit. 15,000 ped.
__y. 3. in herb. Hooker (Atropa hirsuta, Dr. Meyen).

The specimen above referred to, being named by Dr. Meyen
himself, leaves no doubt as to the identity of the species, so that
+» accordance with it, I have given the above amended diagnosis.
~ Nees v. Esenbeck describes the plant as being much branched
but the specimen here referred to, consists only of a small single
branchlet, which is slender, with alternate distant leaves, the blade

being about nine lines long, and six lines broad, the petiole almost

”

29

§. Perizoma

Por ee eee ee
x

134 ILLUSTRATIONS OF

filiform, measuring sixteen lines, and the peduncle eight lines,
the calyx four lines, the tube of the corolla sixteen lines, with five
reflexed, short, ovate segments of one line and a half; the berry
is about seven lines long and four lines diameter.*

5. Salpichroma diffusa, (n. sp.) : caule suffruticoso, ramosissimo,
divaricato-flexuoso ; foliis geminis, ovatis, basi obtusis, apice sub-
acutis, utrinque pilis articulatis hirsutulis, margine floecoso, petiolo
dilatato limbo breviore ; floribus solitariis, breviter pedunculatis,
calyce 5-partito, hirsuto, laciniis linearibus; corolla subbrevi,
infundibuliformi, tubo nullo modo gracili, ore subcoarctato, calyce
2-plo, aut vix 3-plo longiore, extus pubescente, limbi laciniis
oblongis, obtusiusculis, reflexis, margine ciliatis, genitalibus in-
clusis—America occidentalis intertropica—v. s. in herb. Hook.
Nova Grenada (Bogota, Goudot), Quito (Lloa, Jameson, No. 301),
(Pichincha, Jameson, No. 82), Andibus Peruvianis, (Me Lean.)

The above named plants are those which I had referred, on the
occasion before quoted, to the species last described, they will,
however, be seen to be evidently different: their leaves measure
nine lines in length, seven lines in breadth, the petiole being
about four lines, the peduncle and calyx each three lines, the tube
of the corolla six lines, and its border about two lines.t

6. Salpichroma “ristis, (un. sp.) : humilis, suffruticosa, ramis —
flexuosis, subdichotomis, nudis, striato-rugosis, ramulis tenuis-
simis, brevibus; foliis geminatis minoribus, obovatis, apice sub-
acutis, basi obtusatis, in petiolum planum caniculatum decurren-
tibus, carnosulis, eveniis, utrinque glanduloso-pubescentibus ;
floribus solitariis, pedicellatis, nutantibus: calyce subglabro pro-
funde 5-partito, laciniis lineari-subulatis, acutis ; corolla tubulosa,
tubo imo latiore glabro, staminibus styloque inclusis glabris.
—Quito. v. s. in herb. Hook. (Andibus Peruvianis, Me Lean.)
(Andibus Quitensibus, Jameson, No. 125.)

‘Having seen other specimens of the above plant, I am now
enabled to offer it as a very distinct species. It appears to be a
very diminutive shrub, of stunted Alpine growth, with short. tor-

* This species is delineated in Plate’28, A.

+ This plant is figured in Plate 28, B.

SOUTH AMERICAN PLANTS. 135

tuose knotty branches, and only a few inches in height ; it throws
out a few leaf-bearing branchlets as slender as the petioles, from
half an inch to an inch in length, each exhibiting about three
pairs of geminate leaflets, giving them much the appearance of
being a pinnated leaf: the leaflets are two lines and a quarter
long, and one line and a half broad, upon a channelled, flattened
petiole, one line and three quarters in length: the calyx is cleft
nearly to the base, into five, equal, narrow, subulate segments,
two lines long: the tube of the corolla is about five lines long,
one line and a half in diameter, with five short triangular reflexed
teeth : the stamens arise from the middle of the tube, and are not
exserted : the ovarium is conico-ovate, seated upon a thick, fleshy
ring: the style is somewhat curved at the apex, and thickened
towards the stigma, which is hollow, with an obsoletely bilobed
margin. Both the leaves and flowers become quite black in
drying, a peculiarity noticed upon a former occasion in other
species of this genus: the bark of th@ woody branches is greyish,
finely shagreened with raised dots.*

LYCIOPLESIUM.

To this genus, proposed in this work (Note) p. 10, I have
now to add another species.

6. Lycioplesium fasciculatum, (n. sp.) : spinosum, ramulis sub-
flexuosis, fere rugosis ; foliis alternis, vel in axillas fasciculatis,
oblongo-spathulatis, glabris, carnosulis, nervis pinnatis immersis,
basi in petiolum subbrevem spathulatis, apice obtusis ; floribus in
medio spinaram binis, ‘vel ex apice cum foliis plurimis enatis,
verticillato-fasciculatis, pedunculis calyeeque subpubescentibus ;
corolle tubo brevi, summo campanulato, extus pubescente, limbo
profunde 5-partito, lobis expansis, margine albido-floccosis, stami- |
nibus styloque exsertis.—Bolivia (Bridges Collect. 1846.)

‘This shrub very much resembles in habit the five species for-
merly described ; the spines are from six to nine lines long, the
leaves (including a petiole of three lines) are one inch and a

ee For a figure of this species see Plate 28, C.

136 ILLUSTRATIONS OF

quarter long, and five lines wide, they are of a bright green colour,
somewhat thick and fleshy, smooth on both sides, and above are
quite polished ; the peduncles are half an inch long, the calyx is
two lines, the corolla, including the lobes of the border, is from
six to eight lines in length.*

DUNALIA.

On a preceding occasion p. 13, I offered an amended cha-
racter of this genus, founded upon the observations made upon
a new species very remarkable in its habit, which is there
described (p. 14), and figured in Plate 2. Since then, in the
fine herbarium of Sir William Hooker, which is enriched with
the collections of almost every South American traveller, I have
seen a specimen of the typical species, D. Solanacea, H.B.K.
of which an excellent figure is given by Professor Kunth in the
Nov. Gen. et Sp. tab. 194; but in this instance, the whole
plant is not almost glabrous, as is there represented: on the
contrary, the stem, the petiole, and the under side of the leaves,
are covered with stellate tomentum, which is also seen in the
hervures of their upper surface; the flowers, in like manner, are
densely clothed with similar tomentum. I find, too, that the tube
of the corolla is not so slender, nor is the border so deeply cleft
as there shown, being more sinuated with shorter and more obtuse
lobes, approaching more the form seen in D, Lyecioides (loc. cit.)
The difference in habit of these two species is very remarkable,
and from their external appearance, in one case, the peculiar
pubescence, its large leaves, its spineless branches, its dense fascicle
of flowers, offer so great a contrast to the general habit of the
other, that no one would pronounce them to belong to the same
_ genus. I have now to add three new species, two very spinose,
from Bolivia, and one, almost spineless, from Mexico, the latter
being remarkable for the greater size of its corolla. It might,
indeed, be easily mistaken for a species of Tochroma, were it not
for its appendiculate filaments and smaller calyx.

* This species, with sectional details, is shown in Plate 29,

SOUTH AMERICAN PLANTS. 137

An examination of Dunalia acnistoides will show how very in-
timately Dunalia is allied to Acnistus. In the latter genus, the
filaments are generally flattened below the middle, and gradually
expanded towards the point of insertion, and if we conceive the
dilated margins to become split, or torn away from the central
portion, we should find an Acnistus, thus, at once, converted into
a Dunalia: there appears to me, indeed, no other difference
between this and the typical species, where the flowers are
numerously aggregated, and Acnistus; in the other spinescent
species, where the flowers are few or solitary, the dissimilitude in
habit is very remarkable. On this account it will probably be
desirable to divide Dunalia into two sections :—Ilst. ConrErti-
FLOR, containing 1. D. solanoides; 2. D. acnistoides; and
2nd. PaucirLor, containing 3. D. lycioides, 4. D. brachyacantha,
5. D. senticosa, and 6. D. ramiflora, enumerated below.

§. CoNFERTIFLORA.

2. Dunalia aenistoides, (n. sp.) : inermis, ramis striatis, glaber-
rimis ; foliis alternis, (floriferis geminis vel ternis,) elliptico-
oblongis, acutiusculis, imo in petiolum longum gracilem canicula-
tum attenuatis, utrinque glaberrimis, supra glanduloso-pruinosis,
subtus pallide glaucis, rachi prominente nervisque pinnatis ruben-
tibus: floribus in axillis superioribus plurimis (circiter 20), fas-
ciculato-aggregatis, petiolo equilongis, pedunculis filiformibus
calyceque glabris, corolle tubo glabro calyce 4-plo longiore, lobis
brevibus, extus tomentosis; staminibus inclusis, infra medium
insertis, appendicibus filamento glabro tertio brevioribus, imoque ~
tubi pubescentibus ; stylo glabro vix exserto.—Huanaco, Peruvie.
v. 8. in herb. meo (Mathews, No. 849, “ Lycium spathulatum”
dicta). ce

This plant so exactly resembles an Aenistus, and possesses
so little the appearance of a Dunalia, that I did not doubt the
correctness of Mathews’s decision when on a former occasion
I referred it to Acnistus spathulatus (ante. p. 22.). Although
much resembling in habit the Lycium spathulatum of the Flora
Peruviana, its violet coloured flowers are far more numerous,

i ;

138 ILLUSTRATIONS OF

and considerably smaller than in that species. Its leaves are three
inches long, and one and a quarter broad, on a slender caniculate
petiole three quarters of an inch long; the peduncle is about six
lines, the calyx one line in length, tubular, obsoletely five-toothed,
the corolla is four lines long, slender at base, slightly infundibuli-
form above, with lobes somewhat expanded, tomentose outside,
and on the margin, half a line long and broad, without any interme-
diate tooth in each sinus: the filaments are one line and a quarter,
the appendices two-thirds of a line, and the anthers half a line
long.

§. Paucrrtor2.

4. Dunalia brachyacantha, (n. sp.) : fruticosa, spinosa, glaber-
rima, ramis vix flexuosis, spmmis nudis, brevibus ; foliis in axillis
fasciculatis, in turionibus alternis, oblongis, in petiolum elongatum
tenuem spathulatis, obtusis, utrinque glabris, supra lucidis, subtus
flavescenti-pallidis, pinnato-nervosis, marginibus subrevolutis ;
floribus sub-ternis, pedunculis 1-floris, gracilibus, calyce glabro,
_ eurte tubuloso, membranaceo, 5-nervio, breviter 5-dentato ;
corolla violacea, longe tubulosa, limbo angusto, breviter 5-lobo,
marginibus floccosis, lobis triangularibus, apice callosis, dentibus
rotundatis glabris in sinubus interjectis: staminibus inclusis.
—Bolivia. v. s. in herb. Lindley (Bridges Coll., 1846).

This species, although very distinct from D. Jycioides, much
resembles it in its spinescent and glabrous habit ; it has straighter
branches, much shorter spines, and larger leaves: its stem is
smooth, angular, and is marked with many small verrucose spots :
its spines are only four lines long, its leaves, exclusive of the
petiole, are two inches and a half long, and one inch broad, the
petiole measuring seven-eighths of an inch: the peduncle is nine
lines long, the calyx being two lines in length, and one line and
a half in diameter: the stamens arise from a contraction of the
tube, a little above its base, and are adnate to it by their central
nerve for the length of two lines, leaving the winged margins quite
free; from this point they become altogether detached and trifid,
the filament being capillary, and four or five lines long, the appen-

SOUTH AMERICAN PLANTS. 139

dages, which form a continuance of the winged margins, being
subulate, scarcely a line in length, and erect. The style is much
longer than the stamens, equalling the length of the corolla, and is
thickened towards the apex. The berry not yet ripe (as seen in
the specimen quoted), is three lines in diameter, supported on
the persistent membranaceous calyx. I regret that the seeds were
not sufficiently matured to determine the form of the embryo.

5. Dunalia sezticosa, (n. sp.) : ramis spinosis, tortuosis, vix
flexuosis, substriatis, rugosis : foliis parvis, oblongis, in petiolum
brevem spathulatis, obtusis, glabris, carnosulis, utrinque pallide
virescentibus : floribus solitariis, vel binis, uno pracociore ; calyce
brevi, 5-gono, mucronato-dentato ; corolla violacea, longe tubu-
losa, limbo versus apicem pubescente, breviter sinuato-5-lobo,
lobis 3-angularibus, callosis, margine tomentosis, dentibus longis
-interjectis ; staminibus inclusis, ineequalibus.—Bolivia. v. s. in
herb. Lindley. (Bridges, anno 1846.)

The spines in this species are one inch in length, the leaves
(including a short petiole of two lines) are one inch long, and
three lines broad; the peduncle is thickened at the apex, and five
lines long; the calyx is one line and a half in length, and dia-
meter; the corolla is an inch long, and its tube two lines in
diameter, the stamens are included, two of them being rather
longer than the others, the lower half of the filaments adhering by
a central nerve from the base to nearly half the length of the tube
of the corolla, the two free-winged margins of which are also ter-
minated by long subulate teeth, a little more than a line long, as
in both the former species, and in D. Lycioides, the anthers are
also basi-fixed, and of a purplish colour : those of the two longer
stamens are within the mouth of the corolla, the others a little
below: the style is included, and of the length of the stamens.

5. Dunalia ramiflora, (n. sp.) : fruticosa, obsolete spinosa, ramis
striatis, glaberrimis ; foliis apice ramorum fasciculatis, aut ex
axillis annotinis solitariis, vel geminis, oblongis, in petiolum elon-
gatum caniculatum tenuem spathulatis, obtusiusculis, utringue |
glabris, subtus pallidioribus, margine sub-revolutis, nervis flexuosis;
floribus axillaribus, presertim in annotinis solitariis, rarius binis ;

2

140 ILLUSTRATIONS OF

pedunculis gracilibus, 1-floris ; calyce brevi, campanulato, 5-nervio,
dentibus 5, obtusis ; corolla majuscula, virescente, tubo elongato,
infundibuliformi, glabro, intus imo tomentoso, limbo brevissimo,
campanulato, ciliato, 5-angulato, angulis acutis, dentibus brevibus
obtusis glabris interjectis, staminibus inequalibus, tubo multo
brevioribus ; stylo longe exserto.—Mexico. vv. s. i herb. Hook.
(Galeotti, No. 1145, Vera Cruz, in uliginosis alt. 500 ped.)

This very distinct species is enumerated among Galeotti’s
Mexican plants (Enum. Acad. Reg. Brux. tom. xii. Bull. No. 2.)
under the name of “Nicotiana plumbaginifolia ? Wild,” and is said
to be found also near Jalisco, in the Province of Guadalaxara, at
an elevation of from 3,000 to 5,000 feet, and at Juquila, near the
coast at Oaxaca, on the borders of the Pacific, at the same eleva-
tion. The specimen consists of a simple, erect, and nearly straight
stem, with internodes of three quarters of an inch distant; these
mostly exhibit large cicatrices of the fallen leaves of the previous
year, and above these arise, generally, a pair of recent leaves, and a
solitary pendent flower: at the termination of the branch, the axils
become closer, the leaves and flowers more fasciculated: only a
single rectangular spine is here seen, which is half an inch in length :
the leaves are quite spathulate, one inch and three quarters long,
in addition to the caniculate petiole of half an inch in length, into
which they are gradually attenuated ; they are six lines and a half
wide at the broadest part near the summit, are quite glabrous,
marked with about five pairs of nervures, which are remarkably
flexuose: the peduncles are one inch and a quarter, to one inch
and a half long, very slender, but thickening towards the apex,
and quite glabrous; the calyx is small, campanular, two lines
long, membranaceous, with five prominent nerves, and five short
rounded teeth, marked on the edge with a marginal nerve; the
corolla is one inch and a half long, contracted for about three
lines at the base, and thence slightly infundibuliform, spreading
into a short campanular mouth tomentose outside, with a pent-
angular ciliate border, the angles being acute, and exhibiting in
_ the plicature of the sinus, a prominent, glabrous, rounded tooth ;
the filaments arising from the upper part of the contraction of the

SOUTH AMERICAN PLANTS. 141

tube, are unequal, varying from six to nine lines in length, the
lateral appendices are scarcely more than two lines long; the
number of stamens I have found to be eight in one instance,
and four only in another, with a sterile fifth, but these, no
doubt, are the result of irregularity; the lower part of the
filaments are very woolly for about three lines in length, above
which they are slender, terete, and glabrous, the anthers are erect,
the ovarium is small, and the style, almost capillary, thickening
slightly towards the apex, is from two to six lines longer than the
corolla.

TOCHROMA.

A very pretty Solanaceous shrub with long purple flowers, now
well known in our gardens, was first noticed by Mr. Bentham,
and was selected by that distinguished botanist as the type of a
new genus, under the name of Jochroma tubulosa: with this, he
at the same time, associated two other species, and I subsequently
added another, evidently congeneric with these two plants last-men-
tioned, I. macrocalyx, Hook. Aj. op. p.19, and Lond. J. Bot. iv.
p. 309, where an excellent figure of this was at the same time
kindly contributed by Sir William Hooker. At the period when
I described the plant last alluded to, I had not seen the Lochroma
tubulosa, Benth., or I should then have hardly ventured to pro-
pose the genus Chenesthes, for the plants there described under
that name. By the kindness of Dr. Lindley, I was furnished,
last year, with a living specimen of Jochroma tubulosa, in flower
and in fruit, and am now therefore prepared to compar ‘he relation
of this typical species with other analogous plants. Subsequent
observations upon this group have led me to the conclusion, that
all the plants which I formerly associated under the name of
Chenesthes, differ but little from the typical species last alluded
to, being only distinguished by an occasional splitting of the
persistent calyx in fruit, and by their flowers being always scarlet
or of a deep orange colour, instead of a dark purple : they have
all, the same long, tubular corolla, spreading very little im the
mouth into a very short campanular border, which is almost entire,

-

142 ILLUSTRATIONS OF

and furnished with five very short teeth: the stamens and pis-
tillum are all alike in structure, and I perceive no difference in
the fruit or seed. Chanesthes, therefore, as a genus, must verge
into that of Jochroma, a name that ill accords with a scarlet
corolla, but one that must remain on the score of priority. I
suggest, however, the propriety of dividing the genus mto two
sections, one Jochroma proper, with a purple or greenish corolla,
the other Chenesthes, with red and orange flowers. To both these
sections [ will here add several new species, proposing, hereafter,
to illustrate by appropriate figures, the structure of each section
respectively. The three plants first alluded to, I propose to sepa-
rate from Jochroma, under the name of Cleochroma, for the reasons
stated under that head (p. 147.) Dr. Walpers (Repert. vol vi.
p- 629) refers Jochroma to the tribe Cestrinez, and in a note
(2bid, 620) says it hardly differs, as a genus, from Cestrum. This
statement I cannot in any degree confirm ; on the contrary, after
a careful analysis, on which the following generic character is
founded, it will be seen that Jochroma most unquestionably
belongs to the tribe Solanea.

-Tocnroma, Bth. (Bot. Reg. vol. xxx. tab. 20.) Calyx ovato-
tubulosus, subinflatus, submembranaceus, 5-dentatus, dentibus
inequalibus, interdum fere obsoletis, demum parum auctus, per-
sistens, et in fructus grossificatione seepe lateraliter hinc fissus.
Corolla tubulosa, calyce 4-6-plo longior, medio subincurva et
subdilatata, limbo brevissimo, vix éxpanso, estivatione plicato,
margine pene integro, floccoso, dentibus 5, minimis, rotundatis, et
alteris 5 quasi obsoletis in sinubus intermediis notata. Stamina 5,
subinclusa ; filamenta teretia, paulo supra basin tubi inserta, imo
crassiuscula, tomentosa, superne gracilia, glabra; authera oblonge,
2-loculares, imo paulo discrete, in sinu basifixe, loculis parallelis
coadunatis rima externa longitudinaliter dehiscentibus. Ovarium
obovatum, imo disco annulari fere obsoleto cinctum, 2-loculare,
ovulis plurimis, dissepimento incrassato, utrinque affixis. Stylus fili-
formis, apice paulo incrassatus, seepissime exsertus. Stigma clavato-
capitatum, emarginato-2-lobum. Bacca calyce membranaceo vesi-
ciforme, interdum hinc fisso, inclusa, 2-locularis. Semina numerosa,

SOUTH AMERICAN PLANTS, 143

compressa, reniformi-rhomboidea, in pulpam tenuem nidulantia,
testa scrobiculata. Embryo intra albumen carnosum fere annu-
laris, filiformis, cotyledonibus semiteretibus, radicula paulo curvata,
infera, ab hilo laterali declinante, eequilongis.

Suffrutices Americe intertropice: indigene ; folia alterna, petio-
lata, elliptica, integra: flores rarius axillares, bini, vel sepissime
é ramulo novello cymulam umbelliformam, primum terminalem, mox
lateralem simulantes ; pedicelli wniflori, elongati; corolle longe,
speciosa. 3

§ I. Tocuroma vera: corolla dense purpurea.

1. Tochroma ¢udulosa, Bth. Bot. Reg. vol. xxxi. tab. 20.
Habrothamnus cyaneus, Lind?. Bot. Reg. vol. xxx. Mise. p. 72:
ramulis junioribus incano-pulverulentis : foliis. ellipticis, utrinque
acuminatis, subacutis, imo in petiolum decurrentibus, pulveru-
lentis, superne demum parce pubescentibus, cymula 6-8-flore,
calyce inflato, corolla profunde purpurea.—Loxa, in Andibus
Ecuadorensibus. (Hartweg). vv. v. cult. et sic. in herb. Hooker.
(Loxa, Seemann, n. 883.)

This is described as a shrub, from four to six feet high. The
leaves are three inches and a half long, one inch and three quarters
broad, upon a petiole one inch in length. From six to eight
flowers spring out of the apex of the branch, which subsequently
increasing, leaves the fascicle finally axillary; the peduncle is ten
to fourteen lines long, the calyx is four lines long, and the corolla,
of a deep rich purple colour, is one inch and a quarter long, and
two lines and a half in diameter, somewhat narrow in the mouth
and base, the border very short, somewhat cup-shaped, being only
four lines in diameter, when fully expanded : its margin is almost
entire, tomentose, with five extremely short, almost obsolete,
- rounded teeth. ‘The berry is oval, five lines long, three lines in
diameter, enclosed in the scarcely enlarged ventricose, membrana-
ceous calyx, and contains a number of small, flattened, rhomboidal
seeds.

2. Tochroma Jongipes, (n. sp.) : ramulis glabris ; foliis ellipticis,
utrinque acuminatis, longe petiolatis, undique glaberrimis, subtus

144 ILLUSTRATIONS OF

pallidioribus, margine subrevolutis; floribus speciosis, fasciculatis,
longissime pedunculatis, glabris, pedunculo apice incrassato,
corollz tubo elongato, limbo brevissimo, subeampanulato, margine
tomentoso, dentibus 5 minimis rotundatis cum alteris in sinubus
notato, genitalibus exsertis; bacca oblonga, calyce persistente
lateraliter fisso cincta, et duplo longiore.—Kcuador. v. s. in herb.
Hook. in Vallem Llow (Jameson).

This plant has very much the habit of the preceding species,
but it is altogether devoid of any pubescence. Its leaves are four
inches and a half long, one inch and three quarters broad, upon a
petiole from one inch to one inch and a half long; the peduncles
are two inches and a quarter to two inches and three quarters
long, thickening towards the apex; the calyx is tubular, quite
smooth, unequally five-toothed, four lines long, and. two lines and
a half in diameter; the corolla, apparently purple, is one inch and
a half long, three lines in diameter in the middle, somewhat con-
tracted below, and in the mouth, terminating in a short cup-
shaped, almost entire border, as in the last species, with five dis-
tant, small, rounded teeth, and with another short intermediate
-tooth in each sinus. The berry, apparently not quite ripe, is eight
lines long, three lines in diameter, invested by the persistent calyx
three lines long.*

§ IL. Cuanustuzs: calyce in fructu lateraliter fisso: corolla
coccinea vel aurantiaca.

The characters of the species before enumerated, are here revised
upon more extended specimens.

3. Iochroma fuchsioides. Cheenesthes fuchsioides, Nod. hu.
op. in Annot. iv. p. 17. Liycium fuchsioides. H. B. K. Nov.
Gen. vol. iii. p. 52. Plant. Aiquin. vol. i. p. 147. tab. 42. Bot.
Mag. tab. 4149. Frutex sesquiorgyalis: foliis obovato-oblongis,
in petiolum gracilem attenuatis, obtusiusculis, subfasciculatis, gla-
berrimis: floribus umbellato-fasciculatis, axillaribus, terminali-
busque, pedicellis glabris, elongatis, cernuis ; calyce subgloboso,
margine brevissime inzequaliter 5-dentato; corolla tubulosa, coc-

* A figure of this species, with sectional details, will be shown in Plate 30.

SOUTH AMERICAN PLANTS. 145

cinea, glabra, intus imo pubescente, staminibus inclusis, filamentis
gracilibus imo incrassatis et tomentosis ; bacca rubra, ovata, calyce
aucto lateraliter fisso inclusa.—Columbia. v. s. in herd. Hook.
Quito, in Vallem Llow (Hall, No.7.) Cuenca, Nove Grenade
(Jameson). Loxa, regni Ecuadorensis (Seemann, a. 882.)
Columbia, (Lobd.)

To the details of this species (huj. op. p. 17.) little more
need be added. The corolla exhibits five short teeth with
other intervening ones in the plicature of each sinus, as in
I. ¢ubulosa: the filaments are considerably thickened and densely
tomentose at base: the berry is oblong, and very pointed, and is
invested by the enlarged calyx, which splits on one side to the base.
‘The form of the embryo is similar to that of the species just
referred to.

4. Iochroma wmbrosa. Chznesthes umbrosa, Nod. (huj. op.
p- 18). Lycium umbrosum, H.B.K. vol. iii. p. 54. Arbor
biorgyalis, ramulis angulatis, hirto-pubescentibus ; foliis oblongis,
acuminatis, glabriusculis, floralibus ovato-rhomboideis; floribus
umbellato-fasciculatis, pedicellis elongatis ; calyce pilosulo, sub-
campanulato, dentibus 5, inequalibus, obtusiusculis; corolla
tubulosa, pilosa, coccinea? calyce 6-plo longiore, limbo brevi, sub-
campanulato, margine ciliato dentibus 5 et alteris in sinubus fere
obsoletis, donato staminibus inclusis, filamentis filiformibus, glabris,
imo incrassatis et tomentosis.—Columbia. v. s. a herb. Hook.
(Hartweg, 1310.) |

To the details before given (huj. op. p. 18), I have now
only to observe that the border of the corolla is very short, with
teeth somewhat larger than in I. tudu/osa; the filaments are in-
serted a little above the middle of the tube, thickened at base, and
densely tomentose for one third of their length, more slender
and glabrous above, and shorter than the corolla.

5. Iochroma gesnerioides. Cheenesthes gesnerioides, Nod. (doc.
cit. p. 19.) Lycium gesnerioides, H.B.X. vol. iii. p. 53: ramulis
cano-tomentosis, foliis ovatis, oblongisve, acutis, superne demum
fere glabris, infra pulverulentis; floribus umbellato-fasciculatis ;
calyce 5-dentato; corolla tubulosa, aurantiaca, limbo subbrevi,

U ,

146 ILLUSTRATIONS OF

campanulato, sinuato-5-lobo, angulis acutis, filamentis imo tomen-
tosis, antheris subexsertis.—Peruvia. v. s. in herb. Hook. (Prov.
Chachapoyas, Mathews.)

It may be observed in addition to what was formerly remarked
upon this species, that the pentangular border of the corolla is
more distinctly cleft than in any other species, and exhibits a ten-
dency of form towards that of Cleochroma: the berry is equal in
size to that of the species just mentioned, and is almost enclosed
by a persistent calyx of very sumilar form, sometimes cleft
irregularly.

6. Iochroma Loxensis. Cheenesthes Loxensis, Nob. (oc. cit.
p- 18.) Lycium Loxense, H.B.K. — Loxa, regni Ecuadorensis.

7. Tochroma cornifolia. Cheenesthes cornifolia, Nod. (loc. cit.
p. 19.) Lycium cornifolium, H.B.K.—Quito.

8. Iochroma /anceolata. Cheenesthes lanceolata, Nod. (doc. crt.
p. 19.) : fruticosa, ramulis subferrugineo-floccosis : foliis lanceo-
latis vel oblongis, valde acuminatis, supra parce pubescentibus,
infra _pallidioribus, floccoso-tomentosis, petiolo caniculato tomen-
-toso; floribus plurimis, umbellato-fasciculatis, calyce tubuloso,
subinflato, dentibus 5, ineequalibus, obtusiusculis; corolla tubu-
losa, flavescente, calyce 4-plo longiore, limbo brevissimo 5-dentato.
—Kcuador. v. s. in herb. Hook. Quindiu (Goudot), idem (Purdie).
—Columbia (Seemann).

Respecting this species, in addition to my former remarks (/oe.
cit.), it may only be observed, that the corolla in shape and size,
also resembles that of Jochroma tubulosa, and were it not for the
colour of its flowers, which are said to be of a pale yellow, some
of the specimens might almost be mistaken for that species. The
berry, nearly altogether enclosed by the enlarged calyx, which splits
‘on one side, also resembles that of the plant just mentioned. In
some cases, the leaves are less lanceolate than in the specimen
which I first saw and formerly referred to; they are sometimes
‘much .acuminated at each extremity, six inches long, and three
inches broad, upon a petiole one inch in length.*

* A figure of this species will be given in Plate 31.

SOUTH AMERICAN PLANTS. 147

CLEOCHROMA.

The plant with long, dark, purple flowers which I described
under the name of Jochroma macrocalyx, Hook. (ante p. 19.),
was, without hesitation, referred to that genus, on account of its
being evidently congeneric with the Jochroma calycima, Bth.
Since then, as I have just mentioned p. 141, I have had an oppor-
tunity of examining the typical species lochroma tubulosa, Bth.,
which I had not seen at the period referred to, and have indicated
the reasons for associating Chenesthes with that genus; but at
the same time it appears to me, that not only Z. macrocalyx, but
also I. calycina, Bth., and I. grandiflora, Bth., should be detached
from it, and retained as a separate group, for which I propose the
name of Cleocroma, from ‘eos, prestantia, xpopa, color, on account
of their large, handsome, purple flowers. The differences between
it and Jochroma, which I will now proceed to point out, appear
sufficient to warrant its assuming the rank of a distinct genus,
but should it be thought otherwise, it may take its station as a
third section of Jochroma: the differences between them are cer-
tainly much greater than those which separate Physalis and Sa-
racha. In Cleochroma the calyx is generally very large, much
more so in proportion than in Jochroma, increasing even during
the development of the flower, becoming sometimes nearly half the
length of its long, tubular corolla, and swelling in the middle to a
much larger diameter: it is in like manner persistent, and at
length wholly encloses a berry of considerable size. The corolla
is, in like manner, quite tubular, and also somewhat swollen in the
middle, but the border is very considerably larger, more expanded,
and deeply divided into five distinct segments, while in Jochroma,
the border is very narrow, but little expanded, and almost entire:
The contrast between the corolla of all the species of Jochroma
and that of Cleochroma grandiflora, with its large. aure-blue
flowers, with the mouth broadly expanded into a campanular form,
and deeply cleft into five acute lobes, is very remarkable, In
Tochroma (includmg Chanesthes), the filaments of the stamens are
always more or less terete, and thickened towards the base, this

2u2

148 ILLUSTRATIONS OF

lower portion being always densely tomentose, while the upper
half is glabrous: in Cleochroma, on the contrary, the filaments
are very thin, dilated, and membranaceous, especially the lower
moiety, which is quite glabrous, or only sometimes slightly
pubescent on the margins: their insertion is near the base in
Tochroma, while in Cleochroma, although adnate below, they be-
come free only a little below the middle of the tube of the corolla,
which is pubescent thence to the base, while the filaments remain
more or less glabrous. Even in the dried specimen, the remains
of the thin annular disc surrounding the base of the ovarium may
be seen in Jochroma, but I have not been able to distinguish it in
that of Cleochroma. In Cleochroma the berry is larger, the seeds
being apparently imbedded in a greater quantity of pulp, the
embryo is less curved, and the cotyledons much shorter in pro-
portion, forming even less than one third of its whole length,.
while in Tochroma, they are equal in length to the radicle. In
the seeds of Jochroma and Chenesthes, the hilum is seen laterally
in the sinus of the margin, where it is scarcely distinguishable by
any particular mark, but in Cleochroma macrocalyz, I have noticed,
in every instance, that the hilum is distinctly perforated through
the testa, which is of thinner texture than in the seeds of ochroma.
Crxocuroma, gen. nov.—Calyz tubulosus, medio ventricosus,
ore subconstrictus, ineequaliter 5-dentatus, reticulatus, persistens,
et seepius augescens. Corolla tubulosa, tubo medio subdilatato,
calyce 2-plo, rarius 6-plo longiore, limbo conspicuo, campanulato,
5-partito, lobis acutis, estivatione plicata. Stamina 5, inclusa ;
Jilamenta dilatata, tenuia, glabra, corolle tubo imo adnata, infra
medium libera; anthere oblong, 2-loculares, loculis parallelis,
connectivo dorsali adnatis, basi paulo cordatis, in sinu affixis,
longitudinaliter dehiscentibus.. Ovarium obovatum, 2-loculare,
ovulis plurimis in dissepimento incrassato utrinque affixis. Sty/us
filiformis, apice incrassatus. Stigma capitato-bilobum. Bacca
magna, ovata, calyce inflata inclusa, 2-locularis. Semina nume-
Tosa, compressa, reniformi-rhomboidea, in pulpam copiosam nidu-
lantia, testa scrobiculata, hilo in sinu laterali perforate. Zmbryo
intra albumen carnosum fere semiannularis, filiformis, cotyledo-

SOUTH AMERICAN PLANTS. 149

nibus semiteretibus, radicula incurvata, infera, ab hilo declinante,
duplo, 3-plove brevioribus.—Suffrutices Heuadorenses, folia alterna
petiotata, flores speciosi, purpurascentes, sub-wmbellati, pedicellis
elongatis, unifloris.

1. Cleochroma macrocalyx. Iochroma macrocalyx, Hook. (ante
p- 19:) Lond. J. Bot. iv. p. 339, tab. 13-14: foliis rhomboideo-
ovatis, utrinque molliter pubescentibus, subtus pallidis: floribus
umbellato-fasciculatis: calyce tubo magno, ventricoso, 5-dentato,
dentibus inzequalibus, corolla magna, speciosa, tubo calyce 2-3-plo
longiore, hirtella, violacea, staminibus inclusis, filamentis dilatatis,
glabris, nervo longitudinali notatis,imo margine ciliatis—Quito, in
vallem Lloz altit. 9,500 ped. v. s. in herb. Hook. (Hall.)

I have little to add to the details of this species given in the
place above quoted, except that. of the observations made by Col.
Hall, that “the calyx and corolla are of a deep indigo blue.”

2. Cleochroma ecalycina. Iochroma calycina, Bth. Bot. Reg.
1831. su Tab. 20: ramulis angulatis, verrucosis, pallide floc-
coso-pulverulentis ; foliis deflexis, oblongo-lanceolatis, minute
ruguloso-punctatis, aspero et incano-pulverulentis, inferne flavidis
et araneoso-pulverulentis, petiolo valido, caniculato, imo crassiore ;
floribus fasciculato-congestis, calyce magno, medio inflato, demum
augescente, hine fisso; corolla tubulosa, cyanea, floccoso-pubes-
cente, limbo expanso, 5-partito, genitalibus inclusis, filamentis
dilatatis, tenuibus, nervo centrali notatis, glabris, imo margine
ciliatis. Columbia. v. s. in herb. Hook. (Hartweg. 2. 1312.)

This plant has a very peculiar appearance ; the leaves are turned
down by the deflexion of the petiole, and are remarkable for the
numerous close, almost scabrid spots of pulverulent hairs, and for
the yellowish glandular pruinose down, that covers the under
surface: they are six inches long, and two inches and a quarter
broad, on a petiole of three quarters of an inch : the pedicels are
one inch long, swelling upwards, the calyx, at first small and cylin-
drical, afterwards swells and acquires, before the ripening of the
fruit, a length of one inch and a half, and is dilated below to the
diameter of balf an inch, remaining contracted in the mouth, so
that by the growth of the included berry, it becomes ruptured on

150 ILLUSTRATIONS OF

one side towards the summit: the tube of the corolla is more
slender than the former species, and is one inch and a half long.

3. Cleochroma grandiflora. Tochroma grandiflora, Bth. (loc. cit.):
fruticosa, ramulis angulato-compressis, striatis, junioribus floccoso-
tomentosis; foliis late ovatis, basi rotundatis, ad petiolum tenuem
breviter et abrupte attenuatis, apice acuminatis, supra pulveru-
lento- tomentosis, subtus pallidioribus et. molliter pubescentibus,
penninerviis, nervis divaricatis ; floribus apice ramulorum fascicu-
latis, pendulis, pedunculis elongatis calyceque brevi demum ampli-
ato molliter pubescentibus, corolle infundibuliformis tubo longo,
pubescente, fauce sub-campanulato, limbo” 5-lobo, lobis amplis,
triangularibus, staminibus imo ortis, fere inclusis, filamentis
omnino glaberrimis.—In Andibus Peruvianis regno Ecuadorensi
conterminis. v. s. in herb. Lindley. (Lobb. n. 316.) a herb. Hook.
(Hartweg. 814.)

This plant is quite distinct from any of the other species; the —
leaves have ten or twelve pairs of nerves, diverging nearly at right
angles with the mid-rib: they are three inches and a half long,
two inches broad, with a caniculate petiole ten lines in length;
the umbels, arising with a few leaves from the summit of the
young branchlets, which are scarcely longer than an inch, are from
six to eight flowered: the flowers are pendulous from a somewhat
slender peduncle, twenty-two lines long ; the calyx in its florescent
state, is only four lines long, and three lines in diameter, but it
increases considerably in size with the fruit: the tube of the
corolla, which is cylindrical, is one inch to one inch and a half
long, and one line and three quarters in diameter, spreads sud-
denly into a somewhat campanulate border, one inch to one inch
and a half in diameter, and is divided into five, oblong, acute, some-
what expanded lobes ; it is described as being of an “ azure blue”
colour.* The tube is quite glabrous, even at the base, where,
in the other two species, it is somewhat pubescent.

* This plant, with sectional details, is represented in Plate 32.

SOUTH AMERICAN PLANTS. 151

HEBECLADUS.

To this genus, which I proposed on a former occasion (ante
p. 1), [have to add the following new species.

9. Hebecladus mollis, (n. sp.): -caule subherbaceo, flexuoso,
dichotomo, hirtello, subangulari: foliis geminatis oblongis, basi
obtusatis, apice acuminatis, irregulariter et grosse sinuato-serratis,
utrinque molliter hirtellis, pilis articulatis, pedunculo axillari, vel e
dichotomia orto, gracili, molliter piloso, 2-floro, folio subsequilongo,
corolla glabra, lutea, genitalibus inclusis—Nova Grenada—~. s. in
herb. Hook. (Goudot, Plages de Combayma.)

This plant has very much the habit of H. asperus, but the
leaves are deeply sinuate, almost lobed, and covered with long,
soft, articulated hairs. The leaves are two inches long, one inch
and a quarter broad, with a petiole half an inch in length; the
peduncle measures one inch and a quarter, the pedicels half an
inch, the calyx a quarter of an inch, the corolla three quarters of
an inch, with a campanulate pentangular border.*

10. Hebecladus granulosus, (n. sp.): caule suffruticoso, flexuoso,
dichotomo, angulato, fusco-tomentello : foliis solitariis, ovatis,
attenuatis, supra furfurosis, vel glanduloso-asperulis, junioribus
hirtellis, pilis articulatis, subtus pilosulis: pedunculis solitariis,
e dichotomiis axillaribus, pilosis, petiolo equilongis, 2-floris, flori-
bus pedicellatis, calyce parvulo, molliter piloso, corolla lutea glabra,
margine tomentello, genitalibus inclusis.—Nova Grenada, v. s. i”
herb. Hook. (Goudot, locis frigidis inter Ibague et Cartago).

This species approaches H. /anceolatus, but the leaves are
smaller, and broader in proportion to their length; they are two
inches long, one inch broad, on a petiole three-eighths of an inch
in length: the peduncle is scarcely one line long, the pedicels,
very tomentose, are four lines ; the calyx two lines 3 the corolla
almost glabrous, tubular below, campanular above, is five lines
long, exclusive of its spreading border of five triangular segments
with tomentose margins, two lines long.

11. Hebecladus sinuosus, (n. sp.): caule angulato, striato,

* A representation of this species, with details, will be seen in Plate 33.

152 ILLUSTRATIONS OF

molliter piloso; foliis alternis, vel geminatis, altero subminon,
oblongis, grosse sinuato-dentatis, lobis obtusiusculis, . utrinque
pilis articulatis molliter hirsutis, margine ciliatis, rachi nervisque
prominulis, imo in petiolum elongatum anguste decurrentibus ;
pedunculo bifloro, petiolo 3-plo breviore, pedicellis zequilongis,
calyceque dense pilosis, corolla fere glabra, sicco lutea, limbi lobis
acutis, staminibus vix exsertis——Peruvia, Prov. Chachapoyas.
v. 8. i herb. meo (Mathews).

This species corresponds much in habit with the figure of
H. biflorus (Atropa biflora) of the Flora Peruviana, but it is alto-
gether covered with soft articulated down, and the leaves are
larger, more sinuosely lobed, and with a much longer petiole.
The leaves are four inches and a quarter long, by two inches and
a half wide, the petiole being one inch and a half long; the
peduncle measures only four lines, the pedicels are of the same
length, the calyx three lines, and the corolla, tubular below, five-
nerved, smooth, with a five-lobed expanded border, altogether six
lines long. It differs from H. mol/is, in having much smaller
leaves, in being less hirsute, with an infinitely shorter inflorescence.

I propose to distinguish under this name a very distinct group
of Solanaceous plants, all natives of. the Vallies of the Andes of
intertropical America. The type is the Saracha punctata of the
Flora Peruviana. They are distinguished from that genus in
being frutescent shrubs or trees, not herbaceous plants, in their
leaves being generally thick, fleshy, shining, and more or less
destitute of pubescence, and their much larger corolla, not rotate, —
but decidedly campanulate, of much thicker consistence, often
fleshy, and generally marked with beautiful spots, whence the
derivation of its name, from souidos, variegatus, xpopa, color. It
is distinguished from Hebecladus and Iochroma, by its much
smaller, glabrous, fleshy leaves, by its campanulate corolla, with
an expanded pentangular border, not tubular and five-lobed, as in
those genera: from Cleochroma it differs in the form of its corolla,
and in its calyx not becoming considerably enlarged with the

SOUTH AMERICAN PLANTS. 153

fruit. From Lycioplesiwm, to which mm many of its species it
approaches greatly in habit, and in the peculiar appearance of its
leaves, it differs by its being destitute of spines, by its larger,
broader, and more campanulate corolla.

Pacrtocnroma: gen. nov. — Calyx turbinatus, ore valde
coarctatus et in dentibus 5 brevissimis approximatis desinens,
tubo subcoriaceo, colorato, ineequaliter in fissuras 1+2-3 abrum-
pens, persistens et non augescens. Corolla speciosa, campanulata,
imo in tubum brevem contracta, plus minusve crassiuscula, seepis-
sime ornatim maculata, limbo expanso, sinuato-5-lobo, estivatione
plicato. Stamina 5, imo corollz inserta, inelusa : jilamenta tenuia,
erecta, colorata: axthere oblonge, 2-lobee, lobis parallele adnatis,
intus longitudinaliter dehiscentibus. Ovarian obovatum, 2-locu-
lare. Stylus longitudine staminum, gracilis. Stigma clavato- —
bilobum. Bacca pisiformis, calyce suffulta, 2-locularis. Semina
plurima, ceetera ignota. |

Frutices euadorenses et Peruviani glabri : folia integerrima
oblonga vel spathulato-ovata, breviter petiolata, crassiuscula, venis
immersis: flores axillares, solitarii, vel bini, interdum plurimi,
pedunculis \-floris elongatis, apice incrassatis, coloratis : corolla
aurantiaca, pulcherrime maculata, vel rubicunda : baceee pisi-
Sormes, rubra.

1. Peecilochroma punctate. Saracha punctata, R. & P. (Fl.
Peruv. vol. ii. p. 42. tab 1784:) suffruticosa: ramulis teretibus,
fusco-coloratis, glabris, junioribus pulveralentis : foliis solitariis,
rarius geminis, ovato-oblongis, venosissimis, supra glabris, subtus
pulverulentis: floribus ad summum ramorum fasciculatim aggre-
_ gatis, nutantibus, pedunculis 6-7, elongatis, unifloris, apice incras-
satis; calyce in dentibus rotundatis rumpente; corolla magna,
late campanulata, limbo sinuato-5-angulato, extus pulverulenta,
intus luteo-purpurascente, punctis' purpureis maculata : genitalibus
inclusis, glabris.—Ad Muna, Tambo, Portachuelo, et Obrajilla, in
Andibus Peruvianis. —

The above plant, referred by Ruiz and Pavon to Saracha, un-
questionably differs from all other species of that genus, which
are generally herbaceous, straggling plants, and very pubescent,

x

154 ILLUSTRATIONS OF

with a smaller and very rotate corolla of much thinner texture.
The leaves from the figure above quoted, are two inches and a
quarter long, one inch and five-eighths broad, with a petiole about
three lines in length; the peduncle is about one inch and a
quarter long: the corolla is one inch in length, and one inch and
a half broad across the margin.

2. Pecilochroma frondosa (n. sp.): suffruticosa, ramulis sub-
compressis, angulato-striatis, angulis ex axillis decurrentibus,
rugulosis, glabris, valde foliosis: foliis subfasciculatis, ellipticis,
utrinque attenuatis, subtenuibus, supra glabris, subtus parce fulvo-
tomentosis, penninerviis, rachi nervisque subtus rubescentibus,
margine revolutis : floribus ex apice turiorum juniorum fasciculato-
aggregatis, fasciculis foliosis, pedunculo uniflore, glabro, apice
incrassato, longitudine floris nutantis ; calycis colorati dentibus
5 brevibus rotundatis ; corolla speciosa, campanulata, extus fulvo-
pulverulenta. —Prov. Chachapoyas Peruvie. v. s. in herb. meo
(Mathews).

_ Although intermediate with the foregoing and following
species, it is manifestly distinct from both. Its leaves are three
inches and a quarter long by one inch and a quarter broad, with
a petiole three quarters of an inch long. About six or eight
flowers are closely aggregated on the very short, young branchlets,
and are mixed with young leaves: the peduncles are nine lines
long, and are much thickened at the apex: the calyx is short,
tubular, smooth, and with the peduncle, is of a dark, red colour,
its margin being membranaceous, and unequally split into five,
short, rounded teeth. The corolla is one inch long, and nine or
ten lines in diameter on the ciliate margin, which is sinuately
five-angular, very slightly pulverulent, and nearly glabrous _ont-
side; almost smooth within the mouth, but pubescent in the
lower and more contracted portion: the filaments are slightly
pubescent, with long, spreading, articulate hairs, are somewhat
unequal in length, scarcely more than half the length of the
corolla, and are slightly dilated at base. The ovarium and style
are glabrous, the latter being the length of the stamens: the
stigma is clavately bilobed.

SOUTH AMERICAN PLANTS. 155

3. Peecilochroma guttata (n. sp.): suffruticosa, ramulis angu-
loso-striatis, subverruculosis, omnino glabris: foliis solitariis,
rarius geminis, obovatis, apice breviter et repente attenuatis, imo
subcuneatis, crasso-coriaceis, utrinque (etiam junioribus) glaberri-
mis, et eveniis, supra lucido-viridibus; subtus luteo-pallidis, mar-
gine revolutis, breviter petiolatis: floribus 8-9 ad apicem ramo-
rum fasciculato-aggregatis, nutantibus, pedunculis unifloris, apice
incrassatis, flore paulo longioribus, calyce glabro, corolla speciosa,
campanulata, limbo sinuato-5-angulato, extus pulverulento-tomen-
tosa, intus subglabra, punctis purpureis maculata, et imo pubes-
cente, ovario tomentoso.—Peruvia. v. 8. in herb. meo. (Mathews,
No. 1151. sub nomine Saracha punctata, R. & P.)

Judging from the details and figure in the Flora Peruviana, this
plant is certainly specifically distinct from the first described
species to which Mathews referred it: The spots in the corolla
are not distinguishable in the dried state, and they are probably
more or less common to all the species of this genus: its leaves
are small, fleshy, with a total absence of all pubescence, and of
any apparent venation, are more ovate, much smaller, with a
comparatively longer petiole than in P. punctata; in fact, they
more resemble those of the genus Lycioplesium : in the specimen
I possess, they measure one inch and five-eighths in length, and
seven-eighths of an inch in breadth, with a petiole one inch and a
quarter long; they are thick, fleshy, polished above, below of a
pale greenish colour, with a prominent reddish mid-rib, and about
five pairs of spreading, slightly promiment nerves. The pedun-
cles are nearly one inch long, and nodding, being much thickened
towards the apex: the corolla is of the same length, and seven-
eighths of an inch in diameter across the mouth; it is less cam-
panulate below, and the lobes of its border more acute, with a
rounder intervening sinus than in P. punctatus; the margin is
ciliately tomentose, outside it is covered with short, yellowish
tomentum, inside it is nearly smooth, except towards the base,
where it is very pubescent; the calyx is quite glabrous, with
roundish, unequal, and membranaceous lobes, five longitudinal
nerves, one in the middle of each lobe, terminating im as many

156 ILLUSTRATIONS OF

short tomentose mucronate teeth. The ovarium is obovate and
tomentose; the style and stigma are quite glabrous, and together
with the stamens, are about three-fourths the length of the
corolla; the stigma is clavately bilobed.

4. Peecilochroma maculata (n.sp.): fruticosa, ramulis juniori-
bus floccoso-tomentosis, adultis glabris, cortice rimoso-verruculoso : —
foliis alternis, vel geminis, oblongis, basi cuneatis, breviter petio-
latis, crassiusculis, margine revolutis, supra lucidis, nervis pinnatis
impressis, tomentellis, subtus fulvo-tomentosis ; floribus axilla-
ribus, solitariis, vel geminis, aut ad apicem ramulorum novellorum
fasciculatim-aggregatis: pedunculis elongatis, calyceque’ brevi
5-dentato glabris: corolla speciosa, imo breviter tubulosa, cito
late campanulata, flava, maculata, utrinque pulverulento-pubes-
cente, limbo sinuato, 5-angulato, genitalibus inclusis, glabris ;
bacca globosa, pisiformi, calyce persistente suffulta. — In Andibus
Peruvie. v. s. in herb. Lindley. (Lobb. n. 152 e 388.)

This is a very handsome species. The leaves are two inches.
long, one inch and an eighth broad, with a petiole four lines in
length; the peduncle, which is considerably thickened at the
apex, is one inch long, and drooping; the corolla is large and
handsome, being one inch and a quarter in length, and the same
in diameter across. the border; it is described as being “ yellow
spotted.” The berry is small, about the size of a pea, and sup-
ported on the persistent calyx, which does not increase in size.

5. Peecilochroma Lobbiana (n. sp.) : suffruticosa, ramis juni-
oribus cupreo-floccosis, adultis nigrescentibus, ramulis divaricatis :
foliis ellipticis, utrinqne acutis, apice seepe obtusis, margine revo-
lutis, utrinque glaberrimis, supra nitidis, nervis pinnatis impressis,
subtus pallide virescentibus, rachi prominente rubello, petiolo
brevi, glabro, tenui, caniculato; floribus speciosis, axillaribus,
binis, rarius ternis, pedunculo apice incrassato folii longitudine,
calyceque brevi 5-dentato glabris ; corolla imo coarctata, deinde
campanulata, sicco aurantiaca, extus pubescente, intus glabra, et
versus basin leviter pubescente, limbo. sinuato, 5-angulato : geni-
talibus: inclusis glabris—In Andibus Pernxig, v. & im herb,
Lindley. (Lobb. n. 389.) ee.

SOUTH AMERICAN PLANTS, 157

This species, in the appearance of its leaves, has very much the
nabit of the genus Lycioplesium, but the flowers are much larger,
and more showy. ‘The leaves are one inch and a half long, ten
lines broad, with a petiole three lines in length: the peduncle is
one inch and a quarter long, drooping, slender at base, thickened
at its summit; the calyx is two lines long, three lines broad,
somewhat pentangular, and five-nerved, the teeth being short,
and rounded, with a mucronate apex: the corolla is large and
handsome, one inch and an eighth in length, and one inch and a
quarter in diameter across the border: the stamens are glabrous,
three quarters the length of the corolla, the style is somewhat
longer, glabrous, slender, and the stigma clavately bilobed *.

6. Peecilochroma Lindeniana (nov. sp.) : suffruticosa, ramis
rugosis, striatis, glabris: foliis cuneato-oblongis, in petiolum
brevem attenuatis, apice obtusis, sub-emarginatis, margine revo-
lutis, utrinque glaberrimis, crassis, supra nitidis, nervis impressis
eveniis, subtus luteo-pallescentibus, rachi nervisque prominenti-
bus; floribus axillaribus solitariis nutantibus, pedunculo folio
florifero longiore, apice incrassato, corolla speciosa, campanulata,
aurantiaca, extus pulverulenta, margine floccoso 5-angulato: geni-
talibus vix inclusis,—Ecuador.—v. s. in herb. Hooker. (Linden,
a. 489.)

It possesses a habit very similar to the species before described,
its leaves are one inch and an eighth long, five-eighths of an inch
wide, tapering, with a very short petiole two lines in length: the
axils are approximate, scarcely more than nine lines apart. ‘The
peduncle is nine lines long, the dark-red fleshy calyx splitting
irregularly into three unequal triangular mucronate lobes with
membranaceous edges, is three-eighths of an inch long, the corolla
is large, broadly campanulate, one inch and a quarter long, one
inch in diameter in the mouth, the somewhat expanded, pent-
angular border, measuring one inch and a half in diameter.

7. Peecilochroma Quitensis. Lycium Quitense, Hook. Teon. 723:
fruticosa, glabra ; foliis ellipticis, obovatisve, obtusis, subcoriaceis,
margine revolutis, breviter petiolatis, utrmque glaberrimis, supra

* The above plant is represented in Plate 34.
Y

158 ILLUSTRATIONS.

lucidis, subtus flavo-pallescentibus; floribus axillaribus, binis,
nutantibus, pedunculo folio fere zquilongo, calyce imo coarctato, -
brevissime 5-dentato, cito irregulariter 2-3-fisso, corolla glabra,
infundibuliformi-campanulata, limbo patente, sub-5-lobo : genita-
libus corollz squilongis, glabris.—In Andibus Quitensibus—
v. 8. in herb. Hooker. (Jameson, n. 200.) .

The drawing above quoted gives an excellent representation of
this species. I observe, however, that when the corolla is fully
grown, it is more campanulate, and the border is more pentangular
than is there figured, where it is seen in its half plicated state
before full expansion; in that state the plicatures of the sinus
bear somewhat the appearance of intermediate teeth, but these in
reality do not exist. The leaves are of a bright, shining green,
one inch and three quarters long, three quarters of an inch broad,
with a fleshy channelled petiole of two lines in length: above, the
nervures are wholly immersed in the fleshy parenchyma; below,
they are seen much spreading, and with the prominent midrib of
a reddish colour. The peduncles, nearly as long as the leaves,
are nodding, and are considerably thickened above: the calyx is
fleshy, three lines long, with five short, obtuse, mucronate teeth,
and its membranous margin is often split irregularly nearly to the
base: the corolla appears of a dark orange or crimson hue, rather
thick in its texture, smooth below, but slightly pubescent above
outside: within the mouth it is smooth, but below it is pubescent :
it is three quarters of an inch long, and measures three quarters
of an inch across the mouth when fully expanded. The whole
plant, especially in the shape and texture of the leaves, as well as
in the appearance of the flowers, approaches very closely some
species of the genus Lycioplesium; but it is not spinose, and the
structure of the calyx and corolla determines its place.

APPENDIX.

Observations upon several genera hitherto placed in Solanacez,
and upon others intermediate between that family and the Scro-
phulariacez.

My attention during the last few years having been directed to
the study of the Solanaceae, I have given the results of this in-
quiry in a series of memoirs in the ‘ Lond. Journ. Bot.,’ vols. iv.,
v. and vii., which are reprinted in this work, where delineations
are offered of the peculiar features of each genus. Having at
length completed the analysis of the remaining genera of this
order, the results will be given in a succeeding volume; but in
order to explain my views in regard to that family, the following
observations are necessary.

Following the track I had marked out as the basis of these
investigations, which has been chiefly to satisfy myself by careful
analysis of the true limits that serve to separate different genera,
I have encountered a number of facts which are very difficult to
reconcile with our present distribution of the Solanaceae, and
which have induced me to carry this inquiry much further than
was at first contemplated. These results having been published
at intervals, as they presented themselves, the order in which
they have appeared is necessarily imperfect in a systematic point
of view ; but as my principal object has been to arrive at truth,

I expect some degree of indulgence, for what may appear as de-
“i ca . plan. I have alluded to the

fects of arrangement and want of
increasing number of novel cases that have offered themselves
during this inquiry, which render it difficult to decide whether

certain genera should be classed in Solanacee or in Scrophula-
riacee, as these natural orders are at present considered ; and in
consequence of the accumulation of these anomalies, it appears
at length necessarily expedient to draw a more certain line of
distinction between these two important natural orders. This
difficulty is not new in the history of the science, for nearly forty

162 APPENDIX.

years ago it did not escape the acute penetration of our distin-
guished countryman Mr. Robert Brown, who then suggested
the plan of avoiding it by the establishment of an intermediate
family*. Another of the great botanists of our time, Mr. Ben-
tham, who has made the Scrophulariacee one of the chief objects
of his study, and to whom we are indebted for the admirable
monograph of that order in the 10th volume of the ‘ Prodromus’
of DeCandolle, published only two years ago, although evidently
aware of this necessity, has never carried it into execution: the
tribe of the Salpiglossidee, which he placed at the head of the Scro-
phulariacee, was manifestly framed under a point of view bearing
toward this end ; and in the addenda to the same volume of the
‘ Prodromus,’ p. 595, he offers some remarks upon what I had
previously hinted, respecting the separation of the genus Lycium
from the Solanacee (huj. op. p. 107).

The establishment of the Salpiglossidee in the manner just
mentioned, has however in no degree removed the objections be-
fore existing, and from the facts which I shall now have to com-
municate, these exceptions will be seen increased to a manifold
amount, for it is now evident that a considerable number of ge-
nera, hitherto placed in Solanaceae, possess a regular corolla, with
a 5-lobed border, offering an imbricate estivation, contrary to the
usual structure of the order, and although possessing five stamens,
one is often smaller, and sometimes sterile, showing an evident
tendency towards the structure of the Scrophulariacee ; and thus,
besides Lycium and some of the genera of the Salpiglossidew, we
have now Petunia, Nierembergia, Solandra, Juanulloa, Marckea,
Hyoscyamus, Atropa, Mandragora, Nicandra, Anisodus, &e. &e.,
forming too important a number of exceptional cases to be passed
over in neglect. Having lately examined with much care the
structure of most of these genera, I am now better prepared to
carry out the views, which I hinted at three years ago, in an
earlier stage of this inquiry (Auj. op. p. 76), where I suggested
the propriety of associating these dissident genera in a distinct
and intermediate tribe or family.

I therefore now propose definitely to confine the Solanacee as

* Solanaceae, “a Scrophularinis distinguuntur preci b

ee, arinit pue embryone ar-
cuato vel spirali et corollz zstivatione plicata, floribusque cookies regu-
laribus isostemonibus. Hinc genera corolla non plicata et simul embryone
fee vel -orergige vel cum iis corolla imbricata, embryone leviter arcuato,

min di amis in . . . . “syne
ile eee oe propria sectione disponenda, futuri ordinis
: From the state of our knowledge at that time, it is evident that these allu-
Re were intended to apply principally to the Verbascee, which by Jussieu,
gia aan most ding b ts were classed among Solanee, but
be ea y may be referred with additional force to the instances alluded

APPENDIX. 163

nearly as possible within the limits prescribed by Mr. Robert
Brown in his ‘ Prod.’ (doc. cit.), viz. to those genera with a mo-
nopetalous corolla, with a 5-, rarely 4-partite border, even in ex-
ceptional cases nearly regular and equal, the borders of whose
lobes are always valvate or induplicato-valvate in estivation ;
epipetalous stamens, alternate with and equal to the number of
the lobes, the fifth being seldom shorter and still more rarely
sterile, anthers always bursting by longitudinal slits or pores ;
an ovarium most generally 2-celled, rarely 3- to 5-locular, with a
simple style, a bilobed or clavate stigma often hollow; a fruit
either capsular or baccate, and albuminous seeds with a terete
embryo, straight, and more or less curved in a nearly annular
form, or somewhat spiral, the radicle in all cases pointing to the
basal angle of the seed, and turned away to some short distance
from the hilum, which is generally lateral and marginal, rarely
almost basal.

The Serophulariacee 1 would also propose should be confined
to those genera that possess a tubular corolla more or less curved
and irregular, with a 4- or 5-partite border generally unequal
and bilabiate, the lobes rarely equal, but in every case with a de-
cidedly imbricate zstivation ; stamens 2 or 4, didynamous, seldom
with a fifth, which is very rarely fertile, often only rudimentary :
an ovarium, most generally bilocular ; a simple style, with a stigma
more or less bilabiate or bilobed ; the fruit almost always cap-
sular (in very few instances baccate), 2-locular, rarely more-celled,
bursting in various ways, with central placente adnate to the
dissepiment, and an embryo enveloped in albumen but little
curved, generally with the radicle pointing to a basal hilum*; in
one solitary instance (Campylanthus) the embryo is however peri-
spherically curved. In this very natural family, although the
floral leaves are often alternate, the cauline leaves are most gene-
rally opposite, which occurs only accidentally in Solanacee, and
the origin of the inflorescence is strictly axillary. Thus limited,
they form a very distinct natural order.

The intermediate group, which I now propose as a suborder,
under the name of Afropinee, or as a new order, under that of
Atropacee, will consist of genera having a tubular persistent
calyx, more or less deeply divided, a hypogynous tubular corolla,
with the tube more or less plicated in bud, and with a border
generally divided into 5 lobes slightly unequal, but which are

* According to Mr. Bentham’s authority, DeCand. ‘ Prod.’ x. p. 186, and
a statement positively affirmed by most botanists, but one which, it appears
to me, must be received with some modification ; for in the age truth-
ful analyses of the genera figured by Nees v. Esenbeck, ‘Gen, Pl. Germ.,
the radicle is shown as in So/anacee, not pointing directly to the hilum. See
plates of Erinus, Veronica, Wulfenia, Odontites, Euphrasia, Bartsia, Pedi-
cularis, and Alectorolophus ( Rhinanthus).

164 APPENDIX.

always either imbricately disposed in estivation or arranged
under some modification between that form and the plicate, but
never valvate, the margins of each lobe being constantly free
from those of the adjoining ones; they have generally 5 fertile
epipetalous stamens, alternate with the lobes, with one of them
sometimes a little shorter, 1 or 3 being very rarely sterile ; an-
thers bilobed, with the lobes parallel, bursting longitudinally at
the margin, one of these lobes being sometimes sterile ; the ova-
rium 2-celled, rarely 5-locular, with ovules generally ascending,
attached to fleshy placentz which are adnate to the dissepiments,
as in Solanacee and Scrophulariacea, a simple style, and a bilobed
stigma often of a very peculiar form ; the fruit is either baccate
or capsular, the seeds generally reniform or compressed, with a
lateral hilum ; the embryo, placed in albumen, is either straight
or more or less curved, sometimes perispherically or spirally.
They are plants with much the habit of the Solanaceae, with
alternate, simple or geminate leaves, many of them possessed of
powerfully medicinal properties.

They offer the peculiarity, distinct from Serophulariacee, and
similar to that of the Solanacee, in having the origin of the in-
florescence always somewhat extra-axillary and lateral in regard
to the insertion of the petiole. I propose to arrange them in the
following manner :—

ATROPINEE OR ATROPACER.

Tribe 1. Nicotianee. Corolla with an elongated fun- )
nel-shaped tube, often more or less hypocrateri-
form, with 5 nearly equal lobes, which are con-
duplicate and then twisted in estivation, as in
Convolvulus : stamens 5, one frequently shorter ; | Nicotiana,
anthers 2-lobed, lobes almost free, medifixed, | Lehmannia,
and without connective, bursting laterally along ' Sairanthus,
the outer edge: capsule 2-locular with bifid | Polyclidia.
valves, the margins of which are somewhat sep-
ticidal, and slightly inflexed at base: seeds with
a short terete embryo somewhat incurved or
slightly arcuate. J

Tribe 2. Daturee. Corolla with an elongated fun-)
nel-shaped tube, having a 5-angular expanded
border with a contorted complicated zstivation,
as in the Nicotianee : 5 equal stamens; anthers
hiro “rg Lc laterally adnate, dorsally | Datura,
attached to a fleshy connective, and bursting }Ceratocaulis,
longitudinally in front : fruit sub-baccate or cap- Srideiepsens
sular, 2-celled above, 4-celled below, with the
fleshy placente adnate to the middle of the dis-
sepiment : seeds with a nearly annular curved
terete embryo. J

APPENDIX.

Tribe 3. Duboisiee. Corolla with a tube either sone)
gated and ventricose above, or short and rotate,
with a 5-lobed border, the lobes being diversely
volutive in zstivation : 5 equal stamens or 4 di-
dynamous with the rudiment of a fifth ; anthers

rounded, cordate, always extrorse, either 2- | Dubdoisia,

celled, with the cells confluent at the apex, or

165

unilocular with a hippocrepiform line of de- r Aathocercis,
hiscence, and gaping transversely as in Verbas- | Anthotroche.

cum: ovarium 2-locular, with numerous ovules
affixed to thickened placentz adnate to the dis-
sepiment: fruit either baccate or capsular, 2-
valved, with septicidal dehiscence: terete em-
bryo in albumen, slightly curved.

Tribe 4. Schizanthee. Corolla deeply cleft into alge |
ral irregular divisions, with a somewhat reci-
procative zstivation : stamens 5, of which 3 are

sterile ; style erect, with a small fistulose stigma, Ockicuisthes:

slightly swollen below, its contracted entire mar-
gin filled with a globose viscous gland: capsule
2-celled, 4-valved, seeds with a terete hemicy-
clically arcuate embryo.

Tribe 5. Salpiglossidee. Corolla more or less ven-
tricose above, sometimes contracted in the
mouth, the border being divided into 5 nearly
equal regular segments, one of them always
somewhat larger and more erect, their zstiva-
tion being reciprocative (see p. 172): stamens 4,
didynamous, sometimes with the rudiment of a

J

Salpiglossis,
Pteroglossis,

fifth ; anthers 2-lobed, lobes divaricate at base, >Leptoglossis,

connected at apex by intervening filament, one
of the lobes being sometimes reduced to a small
lateral dehiscent gland: style winged at its apex
or expanded into a remarkable tongue-shaped
process, which is stigmatose at its emargina-
ture: fruit capsular, 2-locular, 2-valved : embryo
slightly curved, much more so in Salpiglossis. J

Tribe 6. Petuniee. Corolla with an elongated tube,
sometimes hypocrateriform, seldom with the ru-.
diment of a palate, the border being divided
into 5 nearly equal, rounded and emarginated
lobes, their wstivation in Petunia being replica-
tive (see p. 173), in Nierembergia, replicative
at the base of the lobes, with a perfectly quin-
cuncial imbrication at their summits ; stamens 5, r
one of which is shorter, 2 longest; anthers 2- |

Petunia,

lobed, divaricate at base, without connective :
stigma expanded into a remarkably tongue-
shaped form, emarginate at its apex, in Nierem- |
bergia embracing the anthers : capsule and seed

as in Salpiglossidea. :

Browallia.

Nierembergia.

166 APPENDIX.

Tribe 7. Hyoscyamee. Corolla tubular, more or ome
expanded in the mouth in a campanular form,
with the border divided into 5 equal rounded
lobes: stamens 5, equal ; anthers 2-lobed, affixed
to a narrow dorsal connective above, free below, | Hyoscyamus,
and bursting longitudinally in front near the | Scopolia (?),
margin: ovarium 2-celled, and singularly sur- + Physochlena(?),
mounted by a fleshy epigynous gland, which is {| Cacabus,
either small and stylobasic, or else enveloping | Thinogeton.
the upper moiety of the ovarium: fruit an ex-
succous berry, which sometimes bursts by a cir-
cumscissile line on the margin of the gland : em-
bryo terete, annular, and somewhat spiral. “J

Tribe 8. Atropee. Corolla tubular, more or less)
campanular, with a border divided into 5 equal | Atropa,
rounded lobes, which are imbricate in exstiva- | Nicandra,
tion : stamens 5, equal ; anthers ovate, 2-lobed, Cliocarpus,
lobes laterally adnate, reversed in Atropa by the { Anisodus,
deflexion of the filaments : fruit baccate, 2- or | Mandragora,

5-celled, fleshy, often somewhat exsuccous : em- | Lycium.
bryo terete, nearly perispherical.

Tribe 9. Solandree. Corolla generally with an elon-))
gated, straight, rarely a short tube, in no degree
plicated in bud, border 5-cleft into more or less
rounded equal lobes : 5 equal stamens, generally | Solandra,
epipetalous, but sometimes arising from the out- | Marckea,
side of a free ring, attached to the base of the >Juanulloa,
corolla; anthers oblong, 2-celled, cells parallel Sarcophysa,
and adnate upon a dorsal connective, and burst- | Ectozoma.
ing longitudinally in front : fruit a fleshy 2-locu-
lar berry, and seeds with a nearly straight terete
embryo, with a lax testa, as in the Cestrinee. y

Tribe 10. Brunsfelsiee. Corolla with a more or less)
elongated tube, somewhat ventricose below the
contracted mouth, border divided into 5 nearly
equal segments, their zstivation being decidedly
imbricative (unknown in Heteranthia) : stamens | B Isi
didynamous, somewhat inflected at the apex, gh minis
with one pair shorter; anthers unilocular and (17 @"etscea,
hippocrepiform, as in the Verbascee and the Du- | //@teranthia.
boisiee : style slender: stigma small, bilobed,

_and simply clavate, or with the lobes somewhat
gaping : fruit either capsular or baccate, with a

nearly straight embryo. ve
The Solanaceae, Atropacee and Scrophulariacee, as here de-

fined, evidently constitute an alliance, bound together by very
striking and peculiar ¢ aracters, distinguishable in the structure
of their corolla and ovarium, but more especially in that of their

APPENDIX. 167

fruit, which is most generally 2-celled, with many seeds fixed to
thickened placentze adnate to the dissepiment, and having a
terete embryo, more or less curved, with an inferior radicle, cha-
racters that are common to the whole of this large group. So
gradual is the transition from one link to another of this chain,
that it is difficult to discover any decided break in their conti.
nuity, but notwithstanding this, they form too large an assem-
blage to constitute one single family. The Solanaceae, as distin-
guished from the Scrophulariacee in general, exhibit characters
sufficiently marked, but the difficulty lies with the large interme-
diate group above indicated, that equally partake of the features
of both these extremes. I am quite averse to the practice of
multiplying unnecessarily the amount of natural orders beyond
the smallest possible number: it is not therefore any idle no-
tion of proposing a new family that leads now to this sugges-
tion, which would defeat its own object unless supported by
facts, and urged by the necessity of the case; but it is the desire
of grappling with a formidable obstacle, that would otherwise
prevent us from establishing any decided limits between these
two great families. If this difficulty presented itself to me in so
prominent a degree three years ago (Lond. Journ. Bot. v. 188,
note), when I first noticed the anomaly in Lycium, and suggested
its separation from Solanacee on that account, with how much
more force must this discrepancy present itself, when the ex-
ceptionable cases now amount to so extensive an accumulation
in point of number! The zstivation of the corolla has hitherto
been considered to form an unerring line of demarcation between
the Solanacee and Scrophulariacee, but if we place in the former
family a large proportion of genera possessing an imbricate zsti-
vation, and offering frequently — anisomerous flowers (cha-
racters peculiar to the last-mentioned order), we lose at once the
only valid features that can serve to discriminate the boundaries
of these great families. It is clear that the intermediate group
here proposed to be collected together can only be disposed of
in three modes: they must be associated either with the So/a-
nacee, or be attached to the Scrophulariacee, or else they must
remain as a distinct family. In the first case, the Solanacee would
be then divided into two suborders : 1. the Solaninee, having a
corolla with valvate estivation ; and 2. Atropinee, with imbricate
wstivation. In the second case we should associate, 1; Atropinee,
with flowers nearly isomerous ; and 2. Scrophularinee, with ani-
somerous flowers. In either of these two cases we find that in-
consistency to a great extent would be unavoidable ; for in the
former instance we admit a large circle of exceptions to the only
leading characteristic mark of the order ; and in the second case

we include a considerable number of genera, nearly = ina

168 APPENDIX.

family whose principal feature is to possess anisomerous flowers ;
but in the third case we avoid these difficulties and ensure con-
sistency, preserving at the same time the peculiar characteristic
features both of the Solanacee and Scrophulariacee : we should
then have thus, 1. Solanaceae, offering isomerous flowers with a
valvate or induplicato-valvate zestivation ; 2. Atropacee, isomerous
flowers, or nearly so, with imbricate or a peculiar zestivation ; and
3. Scrophulariacee, anisomerous flowers with imbricate estiva-
tion. In any of the three modes of distribution above indicated,
it matters little which we adopt, in regard to the absolute ar-
rangement of the various genera, for in every case they remain
alike, in exactly the same linear order of position. The value of
the Atropacee, as a distinct order, must now rest entirely on its
own intrinsic merits: its adoption seems the only course by
which a large amount of inconsistency can be removed, and it
appears to me a far less objectionable plan to call up a new
family, than to destroy the great landmarks that serve to discri-
minate the limits of two of the most natural families in the
system.

Having shown the arrangement proposed for the distribution
of the Atropacee, I must offer the following explanation. The
division into the suborders Rectembryee and Curvembryee, as
proposed by Endlicher, and followed by me in the arrangement
of the Solanacee formerly given in ‘ Lond. Journ. Bot.’ v. 148,
offers by far too inconstant and doubtful a character to be main-
tained there, or be adopted here; for among the Salpiglossidee,
some species of Petunia possess an embryo nearly straight, and
more curved in others, while in Salpiglossis it is often spiral
bent into more than a complete gyration. I have preferred rather
to follow the estivation of the corolla, as it gradually verges from
the plicato-valvate of the Solanacee into the imbricate mode of
the Scrophulariacee : thus in the tribes Nicotianee and Daturee
we have the contorto-conduplicate, a form by no means valvate,
but the first departure from it : in the Duboisiee we have another
advance, where the lobes of the border are seemingly valvate, but
on examination their margins will be found convolutely inflected,
a form which I have named volutive: in the Salpiglossidee it
assumes the next step here denominated reciprocative : in the
Petuniee we have again another degree, which is only a modifi-
cation of the imbricative, and which I have termed replicative :
and finally, in the Hyoscyamee, Atropee, Solandree and Bruns-
Felsiea, it becomes decidedly imbricative and quincuncial, as in
the Serophulariacee, with which natural order the latter tribe most
closely osculates. In the Atropee the amount of imbrication is
small in extent ; in the genera Brunsfelsia and Solandra it is ex-
cessive im amount, the lobes wholly enveloping one another in

.

APPENDIX. 169

succession. I proceed now to add a few remarks upon each tribe
separately.

1. Nicotianee.—The estivation of the corolla in this tribe, as
has been just remarked, is by no means valvate, or induplicato-
valvate, as in the Solanaceae, the lobes of its border being on the
contrary conduplicate, that is to say, the sides are turned inwards,
and each lobe is thus folded separately on its inner face, along
the central nervure, the sides closely pressed together, the mar-
gins being quite free from those of the adjoining lobes, and thus
plicated, they all possess a spirally twisted inclination in the bud.
This approaches the estivation of the Salpiglossidee, to which
tribe they offer a still nearer affinity in having the fifth stamen
very often shorter, with the other four somewhat didynamous.
It is for these reasons that I have removed the Nicotianee from
the Solanacee, where I formerly placed them.

2. Daturee.—With this very natural group Solandra has been
associated by most botanists, but it evidently possesses a very
different relationship. The Daturee are remarkable for their
large showy flowers, and they all present an zstivation similar to
that of the Nicotianee, only more decidedly contortive and quite
distinct from the valvate preefloration of the Solanacee. Brug-
mansia I consider as most decidedly distinct generically from
Datura, with which it is associated by most botanists, differing
im many points of structure, and forming arborescent shrubs,
sometimes even tall trees, with long pendent trumpet-shaped
flowers of an unusually large size.

3. Duboisiee.—The genera composing this very distinct group
were partly included by Mr. Bentham (Prodr. DeCand. x. 191)
in his Salpiglossidee; these are Duboisia and Anthocercis, to
which Prof. Endlicher added Anthotroche, a genus which by the
former has been referred to Solanacee. In proposing to alter
the decisions of so distinguished a botanist as Mr. Bentham, who,
from the accuracy of his observations and the solidity of his con-
clusions, stands deservedly as one of the first botanists of our
time, it becomes necessary that I should offer some extremely
valid reasons for the changes now suggested, and accordingly I
will offer a few remarks on each genus in succession.

a. Duboisia appears to me to have no relation with any genus
belonging to the Scrophulariacee. Its only species was originally
described by Mr. Brown in his ‘ Prodr.’ p. 448, who placed it,
together with Anthocercis, in a second section of Solanee. The
habit of this plant, as well as the structure of its flowers, are
there stated to agree with those of Myoporum, whence it derived _
its specific name: the figure given of this plant by Endlicher in
his ‘ Iconographia,’ pl. 77, sufficiently agrees with other Myo-
poraceous plants there designed. On examining a specimen of

170 APPENDIX.

the same plant in Sir Wm. Hooker’s herbarium, I noticed one
very important character that has been quite overlooked by all
preceding observers: the anthers are here decidedly extrorse,
instead of the usual introrse direction before assigned to them.
This circumstance brings Dudoisia in close connexion with the
two following genera, and at once removes them from the tribe of
the Salpiglossidee.

8. Anthocercis.—I was glad to avail myself of the opportu-
nity of investigating the structure of the flowers in this genus
from a plant in the living state of 4. viscosa. It agrees with
the figure given by Endlicher in his ‘ Iconographia,’ tab. 68, of
A, littorea, with the exception of the very important feature of
the structure of the anthers, which, as in the preceding genus,
offer the very distinct peculiarity of being affixed extrorsely just
above the sinus upon the filament, so that the lines of dehiscence
are towards the tube of the corolla, not introrsely towards the
centre of the flower, as appears represented in the plate above
referred to. The estivation of the corolla in
Anthocercis viscosa is also very peculiar: at
first sight it would be said to be induplicato-
valvate, but upon more careful examination it {¥/
will be observed that each lobe of the border is (¥¥'
distinctly supervolute, one of its edges being | §
rolled inwards and overlapped by its opposite 9 ¥/
edge ; these are not all turned in one direction, (ug
two being dextrorsely, and the other three coiled ng
up alternately in a smistrorse order. This mode \¥J f
of wstivation is certainly extremely unusual and {|
peculiar, approaching that observed in the Goode. ©
noviacee, on which on a former occasion (Lond.
Journ. Bot. vii. p.59) I have made some observa.
tions. There exists between them this difference,
that here each lobe is longitudinally and super-
volutely coiled round upon itself, in a somewhat
spiral form, while in Goodenia the winged margins are respec-
tively folded back over one another, upon the plane of the cen-
tral portion of each segment. I have also examined in the dried
state the flowers of A. littorea, A. albicans, A. Tasmanica and
A. scabrella, and they all appear to offer the same kind of zsti-
vation and similarly extrorse anthers, so that these appear to be
constant characters. It is worthy of remark, that the peculiar
smell of the leaves and flowers of Anthocercis viscosa resembles
that of the Myoporacee, and that its pedicels are bibracteated,
which is also a feature in that family ; but its extra-axillary pe-
duncles, the estivation of its corolla, the position of its stamens,
its bilocular ovarium with numerous ovules attached to a thick-

APPENDIX. 171

ened placentiferous dissepiment, its many-seeded capsular fruit,
and its slightly curved embryo with an inferior radicle, are cha-
racters quite opposed to its admission into that family. Nor can
these be made to harmonize either with the Scrophulariacee or
Solanaceae, to the latter of which they offer a nearer affinity.
These characters are sufficiently prominent and distinct, and de-
mand a more attentive investigation.

y. Anthotroche.—This genus was placed by Prof. Endlicher
in Scrophulariacee, among the Salpiglossidee, but it has been
since excluded from the order by Mr. Bentham, and referred to
Solanacee (DeCand. Prodr. x. p- 586). It appears to me how-
ever to have as little relation with the one as with the other of these
families. Upon examining a specimen belonging to this genus
from Swan River, I find that in the structure of its anthers it
agrees entirely with that just described as existing in Duboisia ;
this consists of one reniform unilocular cell, fixed extrorsely on
the filament, and dehiscing on the exterior face by one hippocre-
pical suture. Here the tube of the corolla is short and straight,
and the border is divided into five regular lobes, which are ro.
tately expanded ; the stamens are 5 and equal. The ovarium
has an epigynous prominent stylobasic gland as in Cacabus, ana-
logous to that of Hyoscyamus.

Respecting the Duboisiee it only remains to be observed, that
the main points of distinction between it and the other tribes
with which it is here associated, will be found to exist in the ex-
trorse direction of the anthers and the singular zstivation of the
corolla, peculiarities which, although very remarkable, are not
of themselves of sufficient importance to claim for the plants that
compose it the rank of a separate family, but they constitute a
very distinct tribe of the Atropacee. It will consist of two sec-
tions: 1. Euduboisiee, with baccate fruit, and 2. Anthotrochee,
with capsular fruit, comprising Anthocercis and Anthotroche. It
corresponds with the other tribes of the Atropacee in the ori-
gin of the floral peduncles being lateral with respect to the point
of insertion of the petiole.

4, Schizanthee—The genus Schizanthus, from the lateral
extra-axillary insertion of its pedicels and other characters, ap-
pears evidently to belong to the Atropacee rather than to the
Scrophulariacee, but it does not accord with any of the tribes
above noticed. It differs from them in the structure of its an-
thers, which consist of two parallel cells, quite distinct and sepa-
rated from one another, but conjoined by a broad membranaceous
connective, upon which they are dorsally attached: it. possesses
five stamens, of which three are quite anantherous and rudimental ;
the corolla is deeply cleft into numerous unequal segments which
have an imbricate wstivation. Its stigma approaches the form

172 APPENDIX.

of that of Heteranthia : its fruit is capsular as in the Salpiglos-
sidee, and its seeds contain a terete embryo, curved in an almost
spiral form. Its leaves are always alternate and deeply pinnati-
sected, showing an approach to Salpiglossis and Pteroglossis.
The abortion of three of its stamens is an irregularity of which
we find a parallel case in Janthe, which only differs in that re-
spect from Verbascum ; and the deeply laciniated divisions of its
corolla is another abnormal feature, but this may be considered
only as a separation of the lobes of the corolla at each smus, or
a return to its five normal divisions, with a still farther cleavage
of each lobe, by an extension in an excessive degree of the inci-
sions commenced in the emarginatures of all the lobes of the
border in Salpiglossis, which thus shows a tendency towards the
laciniated form of the corolla of Schizanthus.

5. Salpiglossidee—I have ventured to remove this tribe wholly
from the Serophulariacee for the reasons that will be here fully
explained, and as these are founded upon facts in great measure
new, I may confidently expect that such an arrangement will
meet with the concurrence of the author of the able monograph
of this last-mentioned family, who in detailing the characters of
the tribe in question, as given in the Prodr. DeCand. x. p. 190,
goes the length of saying, “ subordo Nolanaceis capsularibus arcte
affinis, et forte melius eis adsociandus.” I propose however to
remove from it several of the genera there associated. They form
an extremely natural group, distinguished by the very peculiar
estivation of their corolla, their didynamous stamens, or where a
_ fifth occurs it is invariably sterile, and they are especially conspi-
cuous for the remarkable dilatation of the stigma, which at once
signalizes them from the others. Their place is manifestly
among the Afropacee, with which they agree in having the ori-
gin of the pedicels always somewhat lateral in regard to the floral
leaflet or bract, not decidedly axillary, as in the Scrophulariacee.
They are all herbaceous plants, generally clothed with viscid
glandular pubescence, and the campanular portion of the tube o.
the corolla is plicated in estivation ; but the lobes of its border
are first conduplicate, with the margins always free from those
of the contiguous lobes, and twisted inwards in a peculiar man-
ner, for which I have proposed the term reciprocative*, a con-
dition intermediate between the induplicato-valvate estivation
of the Solanacee and the imbricate preefloration of the Scrophu-
lariacee ; in order to render this more evident, the accompanying

* It may he thus defined: Astivatio reciprocativa, i. e. lobi superioris
exterioris marginibus utrinque induplicatis, loborum alterorum simpliciter
condu licatis, 2 sinistralibus dextrorsim, 2 dextralibus sinistrorsim torsive
convolutis, marginibus sese applicitis et a contiguis liberis postice spectan-
tibus, plicaturis antice inclinantibus,

APPENDIX. 173

figure is given in the margin; fig. 1 being the
corolla viewed sideways; fig. 2, ditto seen in
front; fig. 3, ditto seen from above. I have added
to this group a new genus, Péeroglossis, founded
upon a plant collected in the north of Chile by
Bridges (his No. 1889). In Salpiglossis the two
broadly expanded lips of the stigma appear al-
most confluent into a tongue-shaped process,
while in the other genera they are more or less
distinctly separated and 2-lipped, especially in
Leptoglossis and Browallia ; but 1 Pteroglossis
one of the lips appears altogether wanting, or
reduced to a small prominent gland. ;

6. Petuniee—The genera which I have separated from the
Solanacee to form this tribe, approach the Salpiglossidee most
closely in habit and in the general structure of their flowers and
seeds, and moreover partake of their peculiar feature, the great
dilatation of their stigma: the broadly expanded lips of this or-
gan appear however more or less soldered into a tongue-shaped
process, as in Sadpiglossis, which singularly embraces the con-
nate anthers in NMerembergia*. They differ notwithstanding
from the Salpiglossidee in the pe-
culiar complex estivation of their
corolla: that of Mierembergia,
being figured in plate 18 A. fig.2
of the ‘Illustration of South Amer.
Plants, will require no further
explanation : the figure of that of
Petunia was omitted in plate 23
of that work, and its description
was mostobscurely given in ‘Lond.
Journ. Bot.’ v: p. 18 (in a note), .
owing to several omissions and
transposals of words in the hurry
of the last moment of the monthly
publication of that journal. In
order to remedy this omission, a
delineation of the estivation + of
Petunia violacea is now given in the margin; fig. 1 being the
corolla seen in front ; fig. 2, the same viewed sideways ; fig. 3, a
transverse section made across the line aa; fig. 4, ditto ditto

across b b.
* See Ill, South Amer. Plants, pl. 18. A. fig. 4, B. fig. 5, and pl. 20. fig. 3.
+ It may be thus more simply defined : ’stivatio replicativa, i. e. lobis
omnibus subconduplicatis, superioris interioris marginibus revolutis, altero-
rum plicaturis postice torsis, marginibus cum contiguis quincuncialiter late
imbricatis, margine altero hinc revoluto. 3
A

174 APPENDIX.

7. Hyoscyamee.—This forms a very natural tribe, remarkable
for the very singular epigynous gland, hitherto I believe new in.
the history of vegetable physiology, the origin and nature of
which it is desirable to ascertain. It cannot bear any analogy
with the true disc, which is always hypogynous in the superior
ovarium and epigynous in the inferior germen, and which is ge-
nerally admitted, by botanists to be little more than a confluent
whorl of abortive stamens. In Cacabus it assumes the form of
an enlargement of the base of the style, but that it exists here as
a distinct organ is proved by the swelling seen within the matured
fruit, in the summit of the cavity of the cells. In Thinogeton it is
considerably larger, where it appears as a coriaceous thickening
of the chartaceous covering that forms the upper portion of its
dry berry. It is however most distinctly developed in Hyoscya-
mus, even in the young ovarium, in the form of a fleshy external
gland, which covers more than the superior moiety of the entire
germen, and on making a longitudinal section it is seen di-
stinctly adnate upon the true endocarpium : it forms therefore a
very good discriminating character of this tribe. The cause of the
opercular dehiscence of the fruit in Hyoscyamus is thus readily
accounted for, because while the lower half of the pericarpial
covering remains thin and membranaceous, the opercular portion
becomes hard and coriaceous, from the indurescence of the glan-
dular covering above-mentioned *. I have placed doubtfully in

: dl

* Although in the above case it is easy to trace the cause of the opercular
dehiscence of the fruit, the same is not so readily accounted for in other cases ;
in Anagallis for example. In this last-mentioned instance, a distinct zonal
line may be seen in the thin pericarpial covering before the ripening of the
fruit, and it is along this that the membranaceous capsule afterwards bursts,
by a clean circumscissure. This zonal line however bears no relation to the
longitudinal true nervures, which may be distinctly traced in the pericarpial
covering, and which, extending from the style to the base, may be referred
to the midribs and marginal junctions of the original carpellary leaves: but
what is the nature of the line which traverses these nervures at right angles
across all the carpellary Jeaves? This is difficult to be accounted for, unless
we imagine it to arise from a cause somewhat analogous to the case of Hyo-
scyamus, only that instead of the line being the marginal limit of an epigy-
nous gland, it may be the edge of an original elementary hypogynous disc,
which by its subsequent — and attenuation becomes hardly distin-
guishable from the rest of the SNe ea On examining this pericarpial
covering, about the period of the fall of the corolla, this zonal line is seen
more transparent than the rest of its substance, and not opake, as is ob-
servable in the regular longitudinal nervures which may then be readily
traced ; at this period however, and even in the younger state of the ova-
rium, before this zonal line becomes distinguishable, the lower: half of the
pericarpial membrane is decidedly of a more greenish hue than the upper
moiety. This appears to me the only theory on which we can account for
the dehiscence of the capsule in Anagallis, but in suggesting it, I confess
that I could not discern the fact of the original existence and ultimate at-
tenuation of such a disc as I have imagined. Although, generally speaking,

APPENDIX. 175

this tribe, Scopolia, Physochlena, Thinogeton and Cacabus, genera
which offer a striking affinity to one another in their most essen-
tial characters, and there can be little doubt that they all form
a portion of one very distinct group. These characters coincide
for the most part with those of Hyoscyamus, and the only con-
sideration wanting to complete their affinity is the zstivation of
their corolla. The funnel-shaped and almost entire border of
the corolla in those genera would almost necessarily imply the
regular plicature of its campanular portion, but it is probable
that at the same time the lobes in estivation may be somewhat
imbricate, as is distinctly observable in Nierembergia and Petu-
ma. It is impossible to determine this question from dried spe-
cimens, and it can only be ascertained from the examination of
living plants. Should the estivation be found, on the contrary,
to be entirely induplicato-valvate, these four genera would not
belong to Atropacee, but must be referred to Svlanacee, where
they would naturally find their place as a capsular tribe preceding
the Jaborosee.

8. Atropee.—This very distinct group is distinguishable from
the other tribes by its baccate fruit, and its ovary devoid of a
fleshy epigynous gland. The first four genera possess a perennial
root, with numerous deciduous herbaceous stems, large showy
flowers, and a somewhat shrubby habit, with dense foliage and
large leaves. Lycium, on the contrary, is a straggling shrub with
woody stems, and frequently with spinous branches: its flowers
are small. These differences are only generic, and do not offer
sufficient reasons for separating the latter genus as a tribe distinct
from the others.

no apparent hypogynous disc is to be seen among the Primulacea, it is oc-
casionally discernible, but I believe only in those genera where the capsule
bursts into valves by the longitudinal carpellary nervures, as in Lysimachia,
of which genus Nees v. Esenb. in his ‘ Gen. Pl. Fl. Germ.’ says distinctly,
“ Germen liberum basi disco annuliformi cinctum.”’ This view of the case,
though quite hypothetical, is rendered still more probable by the facts ob-
servable in the capsule of Plantago which offers a membranaceous pyxi-
dium very similar to that of Anagaillis, At an early period the future trans-
verse line of dehiscence is discernible in the ovarium, as in Anagallis, but it
is then more approximate to the base, proving that the growth of its lower
portion is afterwards more considerable than the upper part; as it advances
towards maturity the zonal line becomes more marked, the upper portion
of the pericarpial covering being of a deeper green hue and more opake,
while the lower moiety is distinctly hyaline and transparent, and of more
slender texture; on becoming ripe, the greater indurescence of the upper
half, by desiccation, is still more evident, facts which lead to the only reason-
able conclusion, that the upper portion of the ovarium is covered by a very
thin epigynous glandular covering, as in Hyoscyamus, but too thin to be
readily detected in parts of such very slender texture: that it does exist, is
however proved by the circumstance of that part of the pericarpial covering
being always less pervious to light, when viewed under the microscope, than
the lower moiety. so
A

176 APPENDIX.

9. Solandree.—These form a very natural group, being all
suffruticose, mostly subscandent plants, with large leaves and
generally showy flowers. I have been enabled to obtain very
satisfactory elements of the little-known genera Juanulloa and
Marckea, besides those of two new genera. They bear a some-
what similar position among the Atropacee that the Metterni-
chiee hold among the Solanacea, and the analogy in the struc-
ture of the seeds of Marckea and Metternichia is sufficiently re-
markable.

10. Brunsfelsiee.—This group, consisting of some of the plants
placed by Mr. Bentham in his Salpiglossidea, is distinguishable
from that tribe as above limited by the absence of the remark-
able dilatation of the stigma : it will comprise the genera Bruns-
felsia, Franciscea and Heteranthia: the latter much resembles
Browallia in its habit, but it accords with the two former genera
in the structure of its anthers, which are unilocular, and curved
in the shape of a horseshoe round a fleshy globular connective,
that in great part enters into and nearly fills the cavity of the
cell, as in the Verbascee. I have here considered Franciscea as
distinct from Brunsfelsia, which Mr. Bentham (in DeCand. Prodr.
x. p. 198) combined together under one genus. In Brunsfelsia
however the corolla is always of a yellowish colour, the tube is
considerably longer and narrower in proportion, and the fruit
consists of a large fleshy drupe inclosing a putamen which is
quite indehiscent. In Franciscea the flowers are always of a
purplish or violet colour, with a much shorter tube and an
oblique rotate border: the fruit is generally capsular, and rarely
somewhat baccate ; but when this occurs, I have noticed in the
dried specimens, that as the fleshy sareocarp covering the puta-
men dries into the form of a coriaceous integument, both split
into four divisions at the apex, in a valvular form, as in the cap-
sular species. In Brunsfelsia the style is very long and slender,
quite erect at the apex, and terminated by a small clavate stigma
which is bilobed, its equal concave lobes being filled with a ball
of grumous matter. In Franciscea the style is considerably en-
larged and incurved at its summit, which is terminated by a
much larger bilobed gaping stigma, the lower lobe being some-
what smaller, and it exhibits in its sinus a globe of viscous mat-
ter, seen only in the living state. In Heteranthia the style is
far exserted, and is terminated at its slender and somewhat in-
curved apex by an almost obsolete fistulose stigma. The spe-
cies of Brunsfelsia attain the size of large trees, 20 fect in height,
while on the contrary those of Franciscea do not exceed the size
of bushes, which are seldom more than 3 or 4 feet high. Hete-
ranthia, on the other hand, is a small repent perennial plant,

with short ascending branches, terminated by a racemose inflo-
rescence. —

APPENDIX. 177

Having now reviewed in succession the different genera com-
posing the Salpiglossidee of Bentham, with the exception of
Schwenkia, it is necessary to offer a few words upon that genus,
the true affinity of which for many years puzzled the sagacity of
botanists. i

Linneus had the penetration first to point out its affinity
with the Solanee, an opinion which has been. since quite disre-
garded. It was afterwards considered as belonging to Primu-
lacee, on account of the insertion of its stamens opposite to the
lobes of the corolla. By Nees v. Esenbeck and Martius it was
subsequently referred to Serophulariacee (Nov. Act. xi. p. 47) ;
but a note was added by Martius pointing out the greater proba-
bility of its affinity to Acanthacee, because of the fissure of the
apex of the dissepiment, a character which I have not observed
in the genus. This indication has not been adopted by others,
certainly not by Nees, who in his monograph on this last-men-
tioned family (DeCand. Prodr. vol. x1.) does not allude in any
way to Schwenkia in relation to it. Mr. Bentham was the first
to explain the apparent anomaly of the position of the stamens
in regard to the lobes of the corolla, and to demonstrate that the
intermediate glands seen in most of the species constituted the
true normal lobes of the border, and that the stamens were con-
sequently alternate, and not opposite to its lobes. It was there-
fore placed by that able botanist next Browallia, a position that
appears to me hardly satisfactory, on account of the valvate zsti-
vation of the lobes of its corolla, and because its anthers consist
of two distinct cells fixed on the apex of a dilated membranaceous
filament. For these reasons, I would suggest its nearer affinity to
Fabiana, with which it possesses many characters in common:
the cristate projection of the placentz from the middle of the dis-
sepiment, and the insertion of the ovules in distinct linear series
as described by Martius (loc. cit.), quite correspond with the
figure I have given of the placentation of Fabiana (Ill. 8. Am.
PL. tab. 17), Schwenkia however is a genus that requires more
careful examination.

Having thus indicated those genera which I propose to sepa-
rate from the Solanacea, it is desirable to exhibit the arrange-
ment of the remainder that will hence constitute that family.
There is a considerable alteration in the view now offered, from
that given on a former occasion, as since that time most of the
genera have been more attentively examined, and their characters
more accurately ascertained. I intend therefore in the sequel to
present a description of the outlines, all now completed, of such
of the genera as have not yet been delineated, enumerating at the
same time the several species composing them (with the excep-
tipn of those of Solanum, Capsicum, Physalis and a few others),

178 APPENDIX.

to which will be subjoined a review of the several new genera
that have presented themselves in the course of this inquiry. To
these details will be added the description of such of the genera
of the Atropacee as have not yet been described by me, and the
whole will offer a large accumulation of novel facts, that probably
may serve to facilitate the labours of the able botanist now en-
gaged in a monograph of this large family, which has hitherto
been so little studied.

In these investigations I have been carried far beyond the line
originally intended, having been tempted to proceed by the
abundant materials that have presented themselves to my notice,
principally derived from the rich herbarium of Sir Wm. Hooker,
to whose kind liberality I am mainly indebted for the opportu-
nity of bringing to light so large an accumulation of new facts.
The following synopsis will be sufficient to exhibit the proposed
arrangement without farther explanations.

SOLANACER.
Tribus 1. Merrernicuiz# (char. Lond, Journ.
Bot. v. 148). Fructus capsularis, embryo
tefes; rectus. ee eee 1. Metternichia.
2. Sessea.
Tribus 2, Cesrrrnex (char. loc. cit.). Id. id... 3. Cestrum.
Tribus 3. Fasrane# (char. loc. cit.). Fructus
capsularis, embryo paullulo incurvatus, fere
niches nik 35; (HOSTS. 1s Saat eee ees Fabiana.
. Vestia.
. Schwenkia ?

Rap

Trisus 4. Jasorose&. Corolla tubo elongato
siccatione nigrescens: fructus baccatus 2-
locularis, embryo teres, fere annularis.... 7. Jaborosa.

8. Dorystigma.
9. Himeranthus.

10. Trechonetes.

11. Salpichroma.

: 12. Nectouxia.

Tribus 5. Iocurome. Corolla tubo elongato,

limbo 5-fido plus duplo longiore : antherz

longitudinaliter dehiscentes: calyx fructife-

rus vix auctus baccam 2-locularem sufful-

_ ciens vel arcte cingens : embryo teres, fere

I ee ua as 13. Lochroma.
14. Cleochroma.
15. Lycioplesium.
16. Pecilochroma.
17. Hebecladus.
18. Dunalia.
19. Acnistus.
20. Phrodus.

APPENDIX. 179

Tribus 6. Puysatezx. Corolla tubo brevi, limbo
campanulato 5-angulato vel 5-partito: an-
there longitudinaliter dehiscentes: calyx
fructiferus valde auctus et vesicarius : fruc- :
tus baccatus, embryo teres, fere annularis.. 21. Physalis.
22. Larnaz.
23. Margaranthus.
24. Withania.
; 25. Hypnoticum.
Tribus 7. Wirnerincex. Corolla tubo brevi,
limbo 5-fido vix excedente : antherz longitu-
dinaliter dehiscentes : calyx fructiferus vix
auctus, baccam 2-locularem suffulciens, vel
eam arcte vestiens : embryo teres, spiraliter
Curvatus «45: faded 9 iii oe aeetalece a af .. 26. Witheringia.
27. Capsicum.
28. Brachistus.
29. Saracha.
30. Discopodium.
31. Puneera.
32. Aureliana.
33. Sichlera.
Tribus 8. Soranra#. Anthere apice 2-porose,
vel in tubum connate, intus dehiscentes :
fructus baccatus, 2- raro pluri-locularis : em-

bryo teres, spiraliter arcuatus ......+--- 34. Solanum.
35. Cyphomandra.

36. Triguera.
37. Lycopersicum.

Versascra.—The suggestions of our learned countryman
offered in his ‘ Prodr. Fl. Nov. Holl.,’ which I have cited in a
former page (in a note, ante, p. 162), were evidently intended,
in the state of our knowledge at that time, to apply principally
to the Verbascee, which by Jussieu, Linneus, an other emi-
nent botanists had been classed among the Solanec. Bart-
ling afterwards was the first to arrange the Verbascee as a di-
stinct tribe among the Scrophulariacee, and Nees v. Esenbeck,
acting upon the suggestion of Mr. Brown, proposed the Ver-
bascine as a distinct family, mtermediate between Solanee and
Scrophularine (Trans. Linn. Soe. xvii. p. 78). The principal
reasons that have induced all subsequent botanists to adopt the
suggestion of Bartling, have been the imbricate estivation of the
corolla, and the frequent suppression of some of the stamens,
which have been considered paramount to the many other not
less important considerations that tended to show the near ap-
proximation of the Verbascee to the Solanee ; but these objec-
tions, fatal as they were to the admission of this tribe mto the
latter family, do not apply to their connexion with the Atropacee,

180 APPENDIX.

with which group they exhibit beyond all doubt a very close
alliance. This is manifest in their general habit, their alternate
leaves with glutinous pubescence, their fetid smell, their power-
fully narcotic and other medicinal qualities, which are so charac-
teristic of the Solanacee and Atropacee: to these may be added
the particular structure of their stamens, which have their an-
thers of a somewhat lunar form, and quite unilocular, curved
round a large clavate termination of the filament, with an almost
globular expansion of their connective, within the cell, that serves
as the polliniferous receptacle, a character pomted out by Nees
as being foreign to the Solanee and rare among the Scrophula-
ring, and as claiming for them a distinct station in the system.
On the other hand it should be borne in mind, that this peculiar
character exists also in the genus Scrophularia itself, the flowers
of which exhibit often declinate anthers and barbate filaments,
together with a fifth sterile stamen, a feature rare in the Scro-
phulariacee, and one that tends to show a very close connexion
of this genus with the Verbasceew, with which tribe it had been
before associated by all preceding botanists, until Mr. Bentham,
in his admirable monograph of the order, has placed it among
the Chelonee (DeCand. Prodr. x. 299). In most of the genera
of this last-mentioned tribe, the anthers are formed constantly,
I believe, of two distinct and divaricate cells, affixed at their apex
on the slender summit of the filament, and quite wanting of the
fleshy connective so manifest in Scrophularia and the Verbascee.
Whatever may be determined in regard to the proper place of the
Verbascee in the system, it is manifest that it is not by the
number of the stamens that we can fix the limit between the Afro-
pacee and Scrophulariacee: thus it is impossible to separate
Celsia from Verbascum, and it would be equally as admissible to
include Celsia with its didynamous stamens, or Janthe with its
single pair, in Atropacee, as it is to place Verbascum, with its
regular pentandrous flowers, in Serophulariacee: such disere-
pancies cannot fail to occur in many solitary points of osculation
between the genera of different tribes, in all our artificial modes
of the classification of plants. We have also other instances not
less strikingly contrary to the ordinary rule in the Xuaresia bi-
flora of the ‘Flora Peruviana, which has a regular 5-partite co-
rolla and 5 alternate equal stamens: this plant Mr. Bentham

uh gly considers to be a true species of Capraria, a genus
decidedly Serophulariaceous ; and in like manner the Bacopa of
Aublet with its 5 equal stamens offers another exception, but
here the plant has opposite leaves, and possesses so precisely the
habit and general features of Herpestes, that its position must
without doubt be fixed contiguous to that genus. The same rule
will apply to another anomalous case instanced by Mr. Bentham

APPENDIX. 181

in the genus Campylanthus, the seeds of which have a perisphe-
rically-curved embryo, a character that by itself would place it
in Atropacee ; but that distinguished botanist fixes its position
among Scrophulariacee, on account of the form of its corolla and
of its anthers, notwithstanding, as he observes, that it bears little
analogy with any other genus contiguous to it. The principal
reason however that appears to me to give the Verbascee the pre-
ference of a place among the Scrophulariacee is the truly axillary
origin of the floral peduncles, a character that in all such doubt-
ful cases may be employed as a decisive line of demarcation be-
tween that order and the Atropacee. The position of the Ver-
bascee should then appear at the head of the Scrophulariacee,
occupying the place of a suborder in the manner of the Salpi-
glossidee of Bentham (DeCand. Prodr. x. p. 190), where they
would serve as a connecting link of the closest affinity between
these two families.

Rerz1a.—This anomalous genus* has never yet found a cer-
tain or satisfactory place in the system, and its position must
remain problematical until the structure of its fruit and seed be
more accurately investigated. By many botanists it has been
placed in Convolvulacee ; others have indicated its relation. to
Apocynee ; some have again referred it to Polemomacee, with
which it certainly offers no affinity ; and Bartling proposed for it
a new natural order, under the name of Retziacee, but this stands
upon too insufficient grounds. Endlicher places Retziacee as a
doubtfal order after Solanacee, and Dr. Lindley arranges the
genus Retzia among Solanacee, after Sessea. Tn the form of its
calyx and of its corolla, the number and position of its stamens,
its bilocular ovarium with placente attached to the dissepiment,
the structure of its capsule and of its seeds as far as they are
known, offer characters strictly conformable with those of Sola-
nacee ; but it would now rather fall among the Atropacea, on ac- —
count of the estivation of its corolla, which is said by Endlicher
and Lindley to be imbricate, and not valvate : the form of its
embryo, which on the authority of Brown (Prodr. 482) is terete
and straight, necessarily, if it were admitted into this family,
would point to its situation as a, tribe near the Solandree.

Thus far every feature appears 10 conformity with such an ar-

* Rerzia, Thunb. Calye tubular, 5-fid, lobes lanceolate, somewhat unequal.
Corolla tubular, eidaigated., straight, tube in no degree plicated, pote of
5 short equal lobes imbricated in estivation. Stamens 5, equal, near hae’
sile in mouth of tube, alternate with its lobes, filaments extremely = ort,
Anthers oblong, cordate, 2-celled, cells parallel, bursting longitudinally in
front. Ovarium oblong, seated on a fleshy gland, 2-locular, ovules Ase
placentz adnate to the dissepiment? Style longer than corolla, fi oes
Stigma very short, bifid, with divaricate linear segments. Capsule bi
cate, bil . bivalved. Seeds several. Embryo ee

182 APPENDIX.

rangement, but one objection presents itself which renders this
conclusion somewhat unsatisfactory, and that is the peculiar habit
of the only well-recognized species, Retzia spicata, which is dif-
ferent from that of any Solanaceous or Atropaceous plant. Here
the leaves are verticillate in fours, and the flowers are solitary
and sessile in each axil, being supported by two bracts similar in
size and shape to the lobes of the calyx. The genus Solenostigma
of Klotzsch, founded upon one of Zeyher’s African plants, and
supposed to be identical with Retzia, was placed by that botanist
in Stilbacee ; but the name would imply that the stigma is there
hollow and tubular, while in Retzia it consists of two small linear
divaricate segments ; hence it is probable that Klotzsch’s plant is
very different from that of Thunberg. I may here observe how-
ever, that this fact does not of itself invalidate their mutual affi-
nity, for in the vast genus Solanwn we meet with different spe-
cies, some with a hollow tubular stigma, and others with bifid
linear segments, exactly similar to the stigma of Retzia. The
Polemonium campanuloides and P. roelloides of Thunberg have
been referred to Retzia by Sprengel, G. Don and Dr. Walpers ;
these plants have both alternate leaves, and if really species of
that genus, they would tend to remove the doubts above expressed
in regard to the place of Retzia in this natural order. Willdenow
states (Syst. i. 887) that the two species last alluded to, cannot
belong to Polemonium, which has a trifid stigma; and he adds,
that P. campanuloides has a bifid stigma as in Retzia. The Con-
volvulus cenotheroides (Linn. fil.) is also said to be another species
of this genus. The only facts wanting to confirm its place in the
system are the position of its ovules and the structure of its seeds.
Dr. Lindley, who has examined its ovarium, has observed that
its ovules are very few, two (or four ?) in each cell, articulated
with and suspended from the dissepiment by a large thickened
funiculus, a character not at all conformable with the Atropacee
or Solanacee, and one that would seem to remove this genus
nearer to the Bruniacee, with which Retzia will be found to pos-
sess many similar characters. For the present therefore we must
hesitate in attaching Retzia to the Atropacee.

_ The genus Lonchostoma of Wikstrém, placed by most botanists
in Retziacee, offers, I find, many characters in common with Bru-
niacee : its sepals are united at the base by a membranaceous tube
which closely invests the ovarium, if not almost adnate with it :
they are surrounded by bracts of equal size : it resembles Graven-
horstia in having its petals combined into a funnel-shaped tube
with a 5-partite border, the lobes of which are carinate and eon-
volutely imbricate in estivation ; the anthers, cordate at base, are
nearly sessile in the mouth ; the style is divided halfway down and
terminated by clavate stigmata ; the ovarium, 2-celled, appears

APPENDIX. 183

under the microscope to be composed of two distinct, though
connate carpels ; the ovules are few, horizontally attached, or
somewhat pendulous from narrow axile placentz attached to the
twofold dissepiment. These are characters that seem to corre-
spond in great measure with the Bruniacee, with which the habit
of Lonchostoma does not ill accord. These are merely hasty
indications, as it would be foreign to the object of the present
investigation to pursue such inquiries farther.

END OF THE FIRST VOLUME,

PLATE XXIX.

Exhibits a portion of a branch of Lycioplesium fasciculatum.

Fig. 1, is a flower complete. Fig. 2, is the calyx and
peduncle. Fig. 3, the corolla. cut open to show the insertion
of the stamens; all nat. size. Fig. 4, stamens seen in dif-
ferent. positions, before and after dehiscence. Fig. 5, the entire
pistil with the calyx cut open and thrown back, to show the
ovarium ; both magnified.

PLATE XXX.

Shows lochroma longipes.

Fig. 1, is the corolla. Fig. 2, ditto, cut open to show the
stamens. Fig. 3, anther in different positions, before and after
dehiscence. Fig. 4, calyx, style and stigma. Fig. 5, ditto,
with the calyx thrown back, to show the ovarium with its
basal gland: all nat. size. Fig. 6, stigma and a portion of
the style ; magnified. Fig. 7, the berry of lochroma tubulosa,
enclosed within, the persistent calyx. Fig. 8, ditto, removed
from the calyx. Fig. 9, transverse section of ditto. Fig. 10,
a seed; all nat. size. Fig. 11, longitudinal section of a seed.
Fig. 12, the embryo, removed from the albumen; both mag-

nified.

PLATE XXXI.

Represents [ochroma (Chenesthes) lanceolata.

Fig. 1, is the calyx and peduncle. Fig. 2, the corolla.
Fig. 8, ditto, cut open to show the stamens. Fig. 4, a stamen ;
all nat. size. Fig. 5, an anther, before and after dehiscence.
Fig. 6, the pistil. Fig. 7, the berry, enclosed in the persistent
calyx, which splits on one side. ig. 8, transverse section of
the berry. Fig. 9, a seed; all nat. size. Fig. 10, ditto,
magnified. Fig. 11, longitudinal section of nucleus of ditto.
Fig. 12, embryo removed from the albumen; both equally

magnified.

PLATE XXXII.

Shows a portion of a branch of Cleochroma grandiflora.

Fig. 1, the corolla cut open, to show the position of the
stamens. ig. 2, the stamens, seen in different positions,
before and after dehiscence. Fig. 8, the calyx cut open, to
display the ovarium surmounted by the style and stigma.
Fig. 4, the berry enclosed within the enlarged calyx, which
splits on one side. Fig. 5, transverse section of the berry ;
all nat. size. Fig. 6, longitudinal section of a seed; much
magnified.

PLATE XXXIII.

Is a representation of Hebecladus mollis.

Fig. 1, is the calyx with its peduncle. Fig. 2, the corolla.
Fig. 3, ditto cut open, showing the transparent areolce at base,
and the position of three of the stamens. Fig. 4, the ‘stamens
removed. Fig. 5, the pistillum. Fig. 6, a berry, supported
by its persistent calyx. Fig. 7, transverse section of a berry.
Fig. 8, a seed; all nat. size. Fiy. 9, longitudinal section of
a seed; much magnified.

PLATE XXXIV.

Show part of a branch of Pecilochroma Lobbiana.
Fig. 1, is the corolla, cut open to show the position of the

stamens. Fig. 2, the stamens. Fig. 3, the calyx, style and
stigma; all nat. size.

Plate

ea ae eres

lage

of I

Details

Plate 31.

Bane oe

J

Prated by Hullmandel & Walton.

~s hthog

J.Mier

Ima \.U

ATC

|
‘53

Bs Miers ithog.

diflora.

jy
g
4:
OD
wo
a
a
as
ae 5
o
—
ae

Plate 33

J) Wie

Plate 34.

.

jana.

=

Peecilochroma Lobb

J Miers lithog.

INDEX

TO

THE FIRST VOLUME

THE SYNONYMS ARE IN ITALICS.

f+

PAGE
Acnistus aguvepebas; MSS. 6.5. 21
.. arborescens, Schl...... 20
Benthami, Miers...... 2%
cauliflorus,Schott.PL.111. 17
Cestroides, Miers...... 23
ellipticus, Hook. fil.... 23
floribundus, G. Don.. 21
grandiflorus? Willd.. 24
Guayaquilensis, G. Don. 21
Plumicri, Miers........ 21
sideroxyloides, G. pe 22
spathulatus, G. Don.. 22
ramiflorus, Miers...... 22
umbellatus, Miers. . 23
Alibrexia revoluta, Miers........ 61
a rupicola, Miers, Pu. x1.. 59
tomentosa, Miers. . 61
Alona ericifolia, Miers.......... 54
.. microphylla, Miers ..... . &
Alona longifolia, Lendl.......... 52
.. revoluta, Lindl.......- mer ee |
.. tomentosa, Lindl.......... 61
Atropa arborescens, Schlect....... 20
.. aspera, R, & P...... 2 3
.. bicolor, — . 4 are a ae - 3
.. biflora, Rv & P.. 2... ee 3
a it dens, Hook. .......-. 5
.- glandulosa, Hook. . eee
.. Airsuta, oleh pes gee e 133
.. rhomboidea, Hook........+ 7
oa sideroxyloides, ra. a

. spinosa, Me rs ewer ee ee ee
.. umbellata, R&P. Me. 8
.. viridifiora, H. B. We... 3

PAGE

ATROPACH ME. fe ew 166
ATROPBM. oF. sts Vee es es 166, 175
B.

Belladonna frutescens, Plum.... - 21
BRUNSFELSIEA.......-.0-- 166, 176

Cc. °

Callibrachoa procumbens, ne et tag |
Cantua fetida, Pers.. Jie ee108
ligustrifolia, Justis, 102
hea Benthami, Miers........ 75
collinum, Miers......... 84
.. Corcovadense, Miers.. 84
coriaceum, Miers.. ..... 85
.<- Endlicheri, Miers...:.... 75
.. feeniculatum, Miers. . 75
.. - lanceolatum, Micts..; read S 83
.. montanum, Miers. ..... 85
.. mucronatum, Miers...... 84
organense, Miers, PL.Xvi. 82
Cestrum campanulatum, oe 85
.. vestitum, Hook. . «- 19
Bp ie 3 Seapaerangl epee rar ne, ame ot.
en cornifolia, Miers. 19, 146
fuchsioides, Miers.. 17, 144
ae gesnerioides,Miers.. 18, 145

lanceolata, Miers, Pu.
mxxEiis sh. Fe h ag; 146

és Lowensis, Miers... 18, 146

umbrosa, Miers.... 18, 145

Cleochroma calycina, Miers...... 149
grandiflora, Miers, Pu.

“i . 150

il INDEX. .

PAGE
Cleochroma macrocalyx, Miers.. 149

Cyphocarpus rigescens, Miers, Pi.
BEVIN ins ics ae > |

D.

DAPTUBE 3. sco cx oe 164, 169
Dierbachia solanacea, Spr........ 13
Dolia clavata, Miers............ 56
lawn, Metis oie es 57
leptophylla, Miers........ 57
salsoloides, Lindl. ....... 56

-. vermiculata, Lindl. Pu. x11. 55
Dorystigma caulescens, omnes Px.

Vive ow estes 28

crispum, Miers, Sous

squarrosum, Miers,

SIOBOISIE AM, 6. occ cicss nics 105 109
Dunalia acnistoides, Miers......
.. brachyacantha, Miers....

+- _ lycioides, Miers, Pu. 11.. 14
ramiflora, Miers,.......
senticosa, Miers.... . .
solanacea, H. B. K.... 13

E.
Ehretia \duplicata, Nees.......... 64
-. halemifolia, L’Her.. . . 64
—“ ING6B) 352 254 265

og
Fabiana denudata, Miers, 4 XVII. 87
o» Mnpricate. cs OF... 87
-. thymifolia, St. “an 87

viscosa, Hook & Arn... 87
FaBIANES...

ee

G.

Grabowskya Boerhaavifolia, Schl.. 64
ee duplicata, Arn... . 64

es Lindleyi, Miers..... 66
obtusa, Arn. PL. x111, 65
Grabowskya, Boerhaavifolia, Lindl. 66
«« disticha, Meyer..... 66

ws Meyeniana, Nees. . 12, 66

H.

Habrothamnus cyaneus, Lindl ... .. 143
ey SS os ea:

ws fasciculatus, Endl... 75

‘ tomentosus, Bth..... 75
Hebecladus asperus, Miers...... 3
bicolor, Miers...... 3

biflorus, Miers ..... 3

granulosus, Miers... 151

hirtellus, Miers... . 4
intermedius, Miers.. 3

PAGE
a lanceolatus, Miers... 3
mollis, Miers, PLATE

Es ak kin 6 8 151
iG sinuosus, Miers.. 151
me Turneri, Miers...... 3
umbellatus, Miers... 2
viridiflorus, Mises 2
Himeranthus erosus, Miers, Pu.v. 26
ae runcinatus, Endl. Pu.
WViinebewwaa ewes 25
tridentatus, Miers,
PEATE: B06 65556 27
HYOSCYAMEA.... 0600. 00 166, 174
-
Iochroma cornifolia, Miers.... 19, 146
.. fuchsioides, Miers.. 17, 144
gesnerioides, Miers.. 18, 145

-» _ lanceolata, Miers, PLATE

19, 146
longipes, Miers, Pu. xxx. 143
Loxensis, Miers. . 18, 146

+. tubulosa, Bth.. eas

+» umbrosa, Miers. . ce 17, 144
Iochroma calycina, Bth...... 19, 149
+. grandiflora, Bth.... ... 150
macrocalyx, Bth. . mean 2G, 149
IOOMROMEBMBas53..is50u5 tones 178

J.
Jaborosa integrifolia, Juss, Pu. v.. 26
Jaborosa caulescens, Hook........ 28
vs Gecurrems; Mier. 2c. s..< 28
-. ,.dacimtata, .Miers....5... 31
: runcinata, Link.... .. 25
JABOBOURM as. vs0s dan cease Ue

L.

Lindernia Montevidensis, Spr...... 111
LOnchostonnn sisse sss <0.0563 nace ss 208
Lycioplesium fasciculatum, Miers

PAM SETS... TSS
os horridum, Miers... 12
“6 Meyenianum, Miers. 12
os nitidum, Miers.... 11
fee obovatum, Miers.. 11

Lock : ee tee iL
cium atum. 7s 8ee ee ip
me erborescens, Hook... 21

- 17, 144
gesnertoides, H. B. K.. 18, 145

ze te thane Willd...... 24

Guayaquilense, H. B. K..
heterophyllum, Murr......
horridum, H. B. K.....
. Lowense, H. B. K....

.. macrophyllum, Bth.....
.. Meyenianum, Nees ab E..
.. obovatum, R. & P......
.. parvifolium, Willd......-
S Qustense; A 00K css ee
spathulatum, R. & P....

Lycium

INDEX.

umbrosum, H. B. K.. 18, 145
.. umbellatum, R. & P.... 23
M.

Metternichia princeps, Mik.Pu.x1v. 69
METTERNICHIES,......-++2.+- 178
N.

Nicotiana (Petunioides) cirrhoides,
Miers, PLATE XXII... 104
Nicotiana awillaris, Lam....... 62108
: nana, Lind. «.6i-ss0.s' 101
.. parviflora, Lehm... . . 111
plumbaginiflora, Willd.. 140
NICOTIANES........ pieces: 104; 109
Nierembergia angustifolia, H.B.K. 94
Me anomala, Miers, Pu.
p &. Sern eee i see are eam 2,
aristata,Sweet...... 94
iS coerulea, Gill...... 97
ae calycina, Hook..... 91
is ericoides, Miers.... 93
Se filicaulis, Lindl..... 93
bi cilis, Hook...... 93
se ippomanica, Miers,
PLATE XVIII.... 92
. linifolia, Miers, Px.
SO is eee. 08
is ?nana, Miers.. ... 101
si pinifolia, Miers.... 97
$8 pubescens, Spr..... 98
as pulchella, Gill...... 97
oe repens, R. & P.... 90
: rigida, Miers, Pu.x1x 96
es rivularis, Miers, Pu.
MVE ei i ice 91
spathulata, H. B. K. 90
és stricta, Miers,Pu.xtx 95
Nierembergia graveolens, St. Hil.. 114
oe intermedia, Grah.... 110
. pheenicea, Sweet.... 109
. pubescens, Spr...+-- 114
5 viscidula, H, B. K.. 113
Nolana atriplicifolia, Don.... ... 50
.. paradoxa, Lindl........-. 49
wee Bok Pics.

ee

ef

eee

ll
P.

PAGE
Periphragmos fetidus, R. & P.... 102
Petunia elegans, Miers, Pu. xxiv. 110
.- intermedia, Don..:..... 110
. ‘nyctaginiflora, Juss...... 108
.. ovalifolia, Miers, Pu.xxrv. 113
.. parviflora, Juss, Pu. xx111. 111
.. propinqua, Miers........ 109
; violacea, Lindl.... ... . 109
.» Viscidula, Miers.... .... 113
Prvrontnme. 6. 6600332. 5. 166,478
PHYSADB Mees. 0 ceibtee es save TOS 5
Pionandra betacea, Miers......-. 38
-. cajanumensis, Miers... 43

ot capsicoides, Miers, Pu.
Be ei acces OL
a ciliata, Miers..... ... 37
oe coriacea, Miers.. ....- 43
“ ? crotonifolia, Miers.... 44
.. diploconos, Miers..... 37
“s divaricata, Miers.... . 41
.. ?flagrans, Miers....- OG

ah floribunda, Miers, Pu.
VII. she: Pesos 34
. fragrans, Miers........ 37
: Gardneri, Miers ..... 38
ee Hartwegii, Miers.... . 43
és ? Narensis Miers...... 44
ws obliqua, Miers........ 39
-. pendula, Miers.......- 39
.. premnifolia, Miers.... 40
.. pubescens, Miers.... . 36
. Tegore, Miers......-. 42
. ?trachyphylla, Miers.. 44
.. viridiflora, Miers....2. 49
Peecilochroma frondosa, Miers.... 154
oe ttata, Miers..... 155
= indeniana, Miers.. 157

Se Lobbiana, Miers, Pu.
eres. e157

maculata, Miers.... 156
punctata, Miers.. . 153
Quitensis, Miers... 157 .

R.

WOM eee ok eek eeeeee . 181
Ss.

Salpichroma ciliata, Miers...... 9, 133

dependens, Miers.. 5, 133
diffusa, Miers, PLATE
KXViEBs. 4c. as 194
eS Miers.. 5, 133
irsuta, Miers, PLATE
RXVI11. A... -... 6, 133
ramosissima, Miers 6, 133

1V

PAGE

Salpichroma rhomboidea, Miers,
PLATE I. 2. waves ds 198

tristis, Miers, PLATE
ERVIUy Gisvews- »-. 134
SALPIGLOSSIDES........ ws = 172
Selpishoonie integrifolia, Hook.... 109
: linearis, Hook....... 110

; prostrata, Hook, Arn. 114
Sarracha punctata, R. & P...... 153
SCHIZANTHES +000 eve-ces 165, 171
Sclerophylax Arnottii, Miers, Pi.

BEVE cis cacszec val
os Gilliesii, Miers, Pu.

ERViwwicicy« ++». FID

xt spinescens, Miers, PL.
SAV. perssc eee. 117
Sessea corymbosa, Miers. . 80
-- dependens, R. & aS re as &,
stipuinta,. By. & Pen jecess, 77
-. vVestita, Miers, PLate xv. 79
SOLANACE AGS; siseneviiys oss. 178
SOLANDRES........ wanbee 100, 170
BOM NEM cid si chvecdaes> ssc. 279
Solanum betaceum, CBV is oo cs ete oe
-. ¢cajanumensis, H. B. K.. 43
++ capsicoides, Mart........ 41
$e owueyotet Be: kh,
-. flagrans, Tenore........ 45
Jragens; TOOK. ia <5 OT

INDEX.

Solanum Narense, H. B. K......
obliquum, R. & P.,....
-. pendulum, R. & P......
premnefolium, Dun......
Tegart; AGM secs...
er a H, Bik. .
viridifiorum, R. & P....
Sorema acuminata, ‘ies:
~«...Janceolata;.:Miers.i.i....

+... \ifeatis, Miere:o.cec3...

.<:' Tittoralis, <Miersex.cx...

-» longifolia, Miers .......

-» paradoxa, Lindl. PLaTe x.

Sorema atriplicifolia, Lindl......
T.

Trechoncetes laciniata, —" PG:

VIL. et tae
a sativa, ‘Miers. . cae
Vv.
VERBASCE AB. 55 i Resi A Sis
Vestia Lycioides, Willd, PL, xx1..
Ww.

Witheringia diploconos, Mart. ... .
a divaricata, Mart......
WITHERINGIES. ..

asse er ee ae ee

179
102