Sr es eee ae en nee a SO SS SSeS Ve Se Se ee ES Vee ea Ses
- 7 L~ . -~ - POTTS Se

- A i

Sa, ane
hes ai.

———————

—

1 ¢ - ‘

Bern hectare tie Helier telilellieteelinteclieeddchehedldededdhhDedetdhihehibettbhihdhelhedededed tei hit iiss 7
- ‘ iad i ¥. : ae etl oe a BR ee . —, - ~~ 7 “ "77 - 27 .
a ° “4 : . Sor 4 whe
‘ if ;

/
ty
Pa

Vat ie Es * m\

a>
Pe :
ark De
Aye
Ben Fuse |

keh

a ae
Vyas) Ps
A) Yaa a

lal Trek
Pi) L

ne

%

nm en A

: ‘ :
i : on 7 : . / ’ J -- ¥ Wh .
— . « " ain ‘ +

" ~ I = an psa) - ; ah abate | ole ay * vate ue OT hee
ry eer _ ae i ‘ ‘“ ee ee awh iy an aes prea a End ' shen An ip aay a
‘? on se” ern ee wa be eee eS —— » enon | Or carder, ke erly Typha ah eae Re Fe anti ramed bila AALEN Ae LUVIN Pla WA 7THntp.1 PAdaPecty {Yeh Pups
re ~ ee my = . y —s> : as ig hah es Son nen semmammia isk Wich girth ameter 0 aul es Vleet ante tan Mecha eae Naw tome

ie a - ado

/ 2

“4o2e 4/7

Peni 2 on = a)
- bl

UNITED STATES

EXPLORING EXPEDITION.

BY AUTHORITY OF CONGRESS.

af

9g &K

G77 Us8

}@54
UNITED STATES

EXPLORING EXPEDITION.

DURING THE YEARS
1838, 1839, 1840, [841, 1842.

UNDER THE COMMAND OF

CHARLES WILKES, U.S.N.
VO lw BV,

BOTAN Y.
PHANEROGAMIA.

BY
ASA GRAY, M.D.
WITH A FOLIO ATLAS OF ONE HUNDRED PLATES.

SOY, | |
BARTS =

PHILADELPHIA:
PRINTED BY C. SHERMAN.
1854, D

SMITHSON ae
MAY 06 1988

SIBRARIED

af

Be OP sa ON XS

PHANEROGAMIA.,

Orvn RANUNCULACE &.
1. CLEMATIS, Jinn.

1. Crematis PickErineu, Sp. Nov.

C. floribus paniculatis ; foliis ternatis ; foliolis membranaceis ovatis acu-
minatis integerrimis quinguenervatis laxe venosis glabris ; pedunculis
3—-T-floris ; caudis carpellorum plumosis.

Has. Ovolau, Feejee Islands.

Climbing, glabrous, at least when old, except the loose and ample
panicles, which are minutely pubescent. Leaves ternate, without any
tendency to farther division. Leaflets thin and membranaceous, ovate,
acuminate, the larger ones conspicuously so, rounded or sometimes
slightly cordate at the base, entire, five-nerved or quintuplinerved, and
not inconspicuously veiny, glabrous, about 3 inches long, on partial
petioles of about an inch in length. Flowers not seen; the only
specimen procured being in advanced fruit. The fructiferous panicles
elongated, somewhat racemiform; the peduncles 2 inches long, 3-7-
flowered, rather longer than the pedicels. Bracts subulate, minute.
Achenia oblong, pointed, pubescent; the ¢ads silky-plumose, an inch

long.
1

2 PHANEROGAMIA.

This is doubtless a polygamo-dicecious species. Although the flowers
have fallen, they were evidently smaller than those of C. indivisa of
New Zealand, of which the sepals, in a specimen gathered by Mr. Moss-
man, are fully an inch in length, and the thin, broader, and acu-
minated leaves, and more compound panicles, further distinguish it
from that species. It has more points of resemblance with C. Jotida,
Raoul, Pl. Nouv. Zel. p. 23, t. 22; but that likewise has coriaceous
leaflets, which, moreover, are hairy underneath. I have dedicated the
species to Dr. Pickering, who, with Mr. Brackenridge, secured a fine
botanical collection in the Feejee Islands, under circumstances of great
personal risk.

9. CLEMATIS PARVIFLORA, A. Cunn.

C. parviflora, A. Cunn. Bot. N. Zeal. in Ann. & Mag. Nat. Hist. 4, p. 259; Hook. f.
Fl. N. Zeal. p. 7.

Has. Tippona, Bay of Islands, New Zealand. (The specimens
destitute both of flowers and fruit.)

3. CLEMATIS INDIVISA, Willd.

C. indivisa, Willd. Spec. Pl. 2, p. 1291; DC. Prodr. 1, p.5; A. Rich. Fl. Nouv. Zel.,
p. 228; Hook. Bot. Mag. t. 4398; Hook. f. Fl. N. Zeal. p. 6.
CO. integrifolia, Forst. Prodr. no. 321, non Linn. :

Has. Bay of Islands, New Zealand. (Foliage only; a peculiar
state, with very thin leaves; evidently growing in deep shade.)

4. CLEMATIS ARISTATA, 2. Br.

C. aristata, R. Br. in DC. Syst. 1, p. 147; Ker, in Bot. Mag. t. 2983; Hook. f. in Jour.
Bot. 2, p. 399.

Has. Sydney, New South Wales.

The foliage varies greatly in this, as in other species of the genus;
but our specimens, both in flower and fruit, belong to the typical form.

RANUNCULACES?. 8

The leaflets show manifest traces of variegation with dark-brown and

purple, as is said by Dr. Hooker to occur in Tasmanian specimens of
this species.

5. CLEMATIS PERUVIANA, DC.

C. Peruviana, DC. Syst. 1, p. 141, & Prodr. 1, p. 4.
C. sericea, Hook. Bot. Mise. 2, p. 205, var. 8., non Humb. & Bonpl.

Has. Bajos, Andes of Peru.

Although very imperfect, like too many in the interesting collection
hastily made in the Peruvian Andes, the specimen suffices to identify
the plant as the C. Peruviana, described from Dombey’s collection ; and
also as the plant gathered between Obrajillo and Culluay, by Cruck-
shanks, and named by Hooker C. sericea; but it is not the C. sericea
of Humboldt and Bonpland. Mr. Matthews and Mr. M’Lean have

also sent to the Hookerian herbarium specimens of the same species
from the Andes of Peru.

6. CLEMATIS DIOICA, Linn. ,

C. dioica, Linn. Ameen. Acad. 5, p. 398, & Spec. ed. 2, p. 765; DC. Syst. 1, p. 148;
Benth. Pl. Hartw. p. 167, no. 872.

Has. Bafios, Andes of Peru.
4
Our specimen accords with the Bogotensian plant of Hartweg, re-
ferred by Bentham to C. dioica, as well as with that gathered by Mat-
thews at Chachapoyas; and I see no valid differences between it and

the Jamaica plant.
7. CLEMATIS H@NKEANA, Prest. ?

C. caulibus scandentibus glabriusculis teretibus ; foliis quinque foliolatis ;
foliolis cordatis acuminatis integris vel apicem versus dentibus grossis

4 PHANEROGAMIA.

2-3 instructis, adultis supra glabratis subtus petiolis cum pedunculis
sepalisque molliter pubescentibus ; paniculis multyfloris.

C. Heenkeana, Presl. Rel. Henk. 2. p. 69?
Has. Obrajillo, Peru.

Although probably the Clematis Hankeana of Presl., I suspect this
plant may be no more than a less pubescent C. sericea, Humb. &
Bonpl. The stems, or rather branches, are terete, very slightly sulcate-
striate, and nearly glabrous, except the younger parts. The leaves,
even the uppermost, are pinnate, with 5 leaflets; these are ovate,
cordate at the base, acuminate, from 2 to 3 inches long, rather thin,
entire or coarsely 2—3-toothed, 3-5-nerved and veiny; the younger
clothed with a soft and fine, rather silky pubescence (like that of the
petioles, inflorescence, &c.) on both surfaces, but the upper surface soon
becomes glabrous, or nearly so. The specimen is in flower only, and
the axillary, rather densely flowered panicles are not longer than the
leaves. Flowers about as large as those of C. Virginiana. Sepals 4,
obovate-oblong, about the length of the pedicels, silky-pubescent ex-
ternally. There is no fruit; but the styles in the fertile flowers are
very silky-villous.. In the same panicle are staminate flowers with
abortive ovaries, and hermaphrodite flowers with both perfect anthers
and fertile ovaries, just as indicated by Mr. Bentham of C. diowca
(Plante Hartwegiane, p. 157), and as is the case, I believe, in most of
the related species; although some are diceciously polygamous.

2. ANEMONE, Tourn.
1. ANEMONE HELLEBORIFOLIA, DC.

A. “foliis pedatisectis, segmentis glabris subcoriaceis [vel membranaceis]
subpetiolulatis basi cuneatis trifidis, lobis serratis acutis ; involucris om-
nibus trifoliis subsessilibus ; ovariis glabris ;” acheniis levibus oblongis
via compressis gibbosis ex angulo interno breviter stipitatis stylo brevi
nudo revoluto uncinatis in capitulum globosum supra receptaculum vil-
losum aggregatis.

A, helleborifolia, DC. Syst. 1, p. 211, & Prodr. 1, p. 21.

RANUNCULACES®. 5

Has. Near Obrajillo (in fruit) and Baiios (in flower), Andes of
Peru.

To complete the character of this well-marked species, that of the
fruit only was needed, which desideratum is supplied by one of the
specimens in this collection. The achenia, like the ovaries, are per-
fectly glabrous, narrowly oblong, turgid, distinctly stipitate from the
inner angle at the base, therefore gibbous, and tipped with a short,
strongly revolute or hooked style; they form a globular head, the recep-
tacle of which is villous. The leaves of our flowering specimen, more-
over, are thin and membranaceous, and so they are in one of the speci-
mens from Matthews, in the herbarium of Sir Wm. Hooker; while in

others, as in Dombey’s plant, they are more or less coriaceous. The
flower also varies in size.

3. RANUNCULUS, Linn.
* Peruvians et Chilenses.
1. RANUNCULUS PRaMorRsUS, H. B. K.

R. premorsus, H. B. K. in DC. Syst. 1, p. 292, & Nov. Gen. & Spec. 5, p.47; Benth.
Pl. Hartw., p. 158.

Has. Baiios, Andes of Peru.

Both Cruckshanks and Matthews have gathered the same species in
the Peruvian Andes (about Pasco, &c.), and some specimens show the
preemose rhizoma.

2. RANUNCULUS BonpLanpbIANUS, H. B. K. 1. .

Var. folits etiam pagina superiore pubescenti-pilosis ; caulibus decumben-
~ tibus vel declinatis.

Has. Bajios, Andes of Peru.

6 PHANEROGAMIA,

Stems a foot long, decumbent or reclining, sparsely hirsute with long
hairs, those of the lower part of the stem spreading, those of the
peduncles appressed and more strigose. Leaves pilose-hirsute on both
surfaces; the radical and lower cauline trisected; the upper three-
parted or trifid; their divisions incised. Flowers as large as those of
R. acris. Petals 10 or 12, spatulate-obovate, entire, apparently light
yellow. Styles uncinate-recurved. Fruit not seen.

3. Ranuncuus Cuitensis, DC.

R. Chilensis, DC. Syst. 1, p. 286; Schlecht. Animad. Ranun. p. 27; Hook. &
Arn. Bot. Beech. Voy. p. 4, t. 3; Gay, Fl. Chil. 1, p. 42; Hook. f. Fl. Antare.
p. 224. |

Casalea Chilensis, St. Hil. Fl. Bras. Merid. 1, p. 9.

Has. About Valparaiso and Santiago.

Although the sepals are only three (a point unnoticed by Hooker
and Arnott); it has no further resemblance to St. Hilaire’s genus
Casalea, to which he referred it.

* * Antarctict.

4, RANUNCULUS BITERNATUS, Smith.

R. biternatus, Smith, in Rees Cycl. no. 48; DC. Syst. 1, p. 286; Deless, Ic. Sel.
1, t. 24; Hook. Ic. Pl. t. 397; Hook.f. Fl. Antare. p. 224.
R. exigiuus, D’Urv.; Gaudich. Bot. Freyc. Voy. p. 186.

Has. Orange Harbour, Fuegia.

This species has been gathered by nearly all Antarctic collectors,
from Commerson and Banks down to the most recent times. It takes
the place of the Arctic R. Lapponicus and R. pygmeus. The petals
are decidedly yellow, as stated by Hooker (not white, as described by
DeCandolle), but rather pale. The fruit was first represented by

Hooker (lcon. supr. cit.); when mature it is considerably rounder
than in his figure.

RANUNCULACEAZA, 7

* * * Novo-Zelandici et Aucklandici,
5. RANUNCULUS RIVULARIS, Banks & Sol.

Ranunculus rivularts, Banks & Solander, Mss. in Herb. Banks; DC. Syst. 1, p. 270,
& Prodr. 1, p. 34; A. Cunn., Bot. N. Zeal. in Ann. & Mag. Nat. Hist. 4, p. 257;
Hook. f. Fl. N. Zeal., p. 11.

Has. Bay of Islands, New Zealand.

The characters of this species given by DeCandolle are well com-
pleted by Cunningham. In our specimens the achenia, sometimes
12 or more in number, are manifestly compressed, and surrounded by
a sharp margin. The species is related to the preceding.

6. RANUNCULUS ACAULIS, Banks & Sol.

Ranunculus acaulis, Banks & Solander, 1. c.; DC., 1. ¢., A. Cunn., 1. ¢.; Hook. f.
Fl. Antare. p. 4. t. 2, & Fl. N. Zeal. p. 11.

Has. Bay of Islands, New Zealand.

This species is admirably illustrated by Dr. Hooker, in his An-
tarctic Flora, from Aucklandian specimens, and is again described in
his Flora of New Zealand.

7. RaNnuncuLus Hirtus, Banks & Sol.

R. hirtus, Banks & Solander, 1. c.; DC. 1. ¢. p. 289; A. Cunn. 1. c. p. 259; Hook. f.
Fl. N. Zeal. p. 9. |

Has. Bay of Islands, New Zealand.

This has smaller flowers, especially the petals, and more incised
leaves, with sharper lobes, than the following. I should suppose that
this was the plant which Richard and Cunningham referred to R.
acris, notwithstanding the small flowers. But Dr. Hooker adduces
the synonym under £. plebevus.

8 PHANEROGAMIA.

8. RANUNCULUS PLEBEIUS, Rt. Br.?

Ranunculus plebeius, A. Cunn., Bot. N. Zeal., 1. ¢. p. 258; Hook. f. Fl. N. Zeal. p.
9; an R. Br.?

Has. Bay of Islands, New Zealand.

This is certainly the R. plebeius of A. Cunningham, and a species
abundantly collected in New Zealand by Mr. Colenso and Dr. Hooker;
‘but I doubt if it be Mr. Brown’s New Holland species. It does
resemble some forms of the polymorphous R. repens; but is quite un-
like R. acres.

9, RAnUNcULUS AUCKLANDICUS, Sp. Nov.

R. sericeo-strigosus ; caule simplici e rhizomate subrepente erecto bifoliato
superne longe nudo unifloro ; foliis longe petiolatis rotundatis vel reni-
formibus trifidis sinubus clausis, segmentis cuneiformibus inciso-sub-
lobatis ; carpellis compressis immarginatis levibus breviter apiculatis
im receptaculum cylindraceum pilosiusculum acervatis.

Has. Lord Auckland Islands.

Rhizoma horizontal, short and rather thick, sending out stout
fibrous roots, and throwing up a solitary erect stem, which in the
fruiting state is from 6 to 10 inches high, rather stout, strigose-hir-
sute with appressed and stiff hairs, two-leaved below, the long and
naked swmmit or peduncle from 3 to 6 inches long. Radical and
cauline leaves similar, on long strigose-hirsute petioles, which are
dilated and partly sheathing at the base; the blade round or rounded
kidney-shaped, or the uppermost truncate or barely subcordate at the
base, both surfaces clothed equally with silky-strigose pubescence ;
they are from an inch to an inch and a half in diameter, thickish in
texture but not fleshy, three-cleft to or beyond the middle, but with
the sinuses mostly closed ; the broadly cuneiform lobes again 2—3-lobed,
or coarsely toothed and incised. The flowers not known. The ripe
carpels are ovate, compressed, not margined, apiculate with a very short

RANUNCULACES. 9

and straight subulate style, smooth and glabrous, a little more than a
line long. The cylindraceous or somewhat club-shaped and papillose
receptacle, from which the ripe achenia have fallen, is a quarter of an
inch long, and minutely hairy.

It is much to be regretted that all the specimens of this interesting
and well-marked species were past flowering, and even the fruit had
mostly fallen from the receptacle. The species is probably allied to
the curious /. pinguis, Hook. f., also an Aucklandian species, on the
one hand, as it certainly is to the Tasmanian &. scapigerus, Hook.,
on the other; but it is entirely distinct from both.

x * * * Novo-Hollandici.

10. RANUNCULUS LAPPACEUS, Smith.

FR. lappaceus, Smith, in Rees Cycl. no. 61; DC. Syst. 1, p. 286; Hook. Jour. Bot.
1, p. 243 (excl. syn. FR. plebeius, R. Br. 7).

Has. Near Sydney, New Holland.

A species very variable in foliage, with flowers as large as those of
R. acris, with which it is justly compared by Smith, or of FB. bulbosus,
and of the same deep yellow colour.

11. Ranuncuus inunpatus, &. Br. in DC.
Var. floribus majoribus ; petalis sepala plerumque duplo superantibus.

Has. Hunter’s River and Woolongong, New Holland (also, Sydney,
Bynoe and Wilberforce, Backhouse, in herb. Hook.)

Some specimens appear to have the very small flowers assigned to
the species by DeCandolle; in others they are more developed, with
petals a quarter of an inch in length; and all belong, I doubt not, to
Mr. Brown’s Ranunculus inundatus; these amphibious plants being
variable in such respects, according to the conditions under which

3

10 PHANEROGAMIA.

they grow. The smooth and turgid achenia are abruptly subulate
with a very slender and straight style.

x * * * * Sandwicenses.
i 12. Ranuncutus Hawazensis, Sp. Nov.

R. hirsutus; caule valido erecto superne ramoso ; pedunculis plurimis
corymbosis; foliis biternatisectis divisionibus primariis longe petiolu-
latis, ramealibus biternatipartitis trisectisve, segmentis omnium cuneato-
oblongis lanceolatisve incisis dentatis ; sepalis ovatis margine lato
hyalino limbatis retusis eatus hirsutis petala obovata (flava) sub-
cequantibus ; achentis levibus late semiovatis compressis stylo conspicuo
subulato apice recurvo vel rectiusculo acuminatis in capitulum subglo-

bosum aggregatis.

Has. Hawaii, Sandwich Islands, in water-runs (also collected by
Douglas).

Stems 3 feet or more in height, erect, thick, fistulous, terete, branch-
ing, clothed, as well as the branches, petioles, &c., with short and
spreading hirsute pubescence. Leaves clothed throughout with a
similar pubescence; the radical and principal cauline biternately com-
pound; the primary divisions on partial petioles of an inch to 3
inches in length; the secondary divisions narrowed at the base into
short and usually margined petioles or nearly sessile, cuneate or
oblong, 2-3-cleft; the lobes oblong or sublanceolate, incised and toothed.
Main petioles dilated and partly sheathing at the base. Upper cau-
line and rameal leaves simply ternate or trisected, with the divisions
laciniate-toothed and incised; the floral commonly undivided and
lanceolate. Peduncles an inch or 2 inches long, often somewhat
fascicled or umbellate-corymbose on the summit of a common peduncle
or branch, forming altogether an ample and many-flowered compound
corymb. Flowers smaller than those of 2. acris, although the fruits
are larger. Sepals rather persistent, hairy outside, very concave,
thin and membranaceous, somewhat coloured, several-nerved near the
centre, ovate-orbicular in outline, including the broad and delicate
hyaline margin, retuse at the summit. Petals yellow, obovate, fur-

RANUNCULACEA, 11

nished with a conspicuous two-cleft appendage on the short claw,
scarcely exceeding the sepals in the later flowers, which the speci-
mens alone furnish. <Achenia 2 lines or more in length, including
the stout and rather conspicuous subulate beak, which is more or less
recurved, in shape very broadly semiovate or dolabriform, compressed,
scarcely margined, smooth, aggregated into a globular head of half an
inch in thickness when full-grown.

This coarse and stout, pretty well-marked species may be compared
with the Mauritanian Ranunculus sericeus, Poir., figured by Delessert;
but that has the divisions of the less dissected leaves sessile, much
narrower sepals, and an oblong head of smaller and short-pointed
carpels. In ours the achenia are remarkably large for the size of the
flowers. '

13. RanuncuLus Mauviensis, Sp. Nov.

fh. parce hirsutus, nune glabratus; caule ramoso elongato decumbente ;
ramis im paniculam laxam foliosam diliquescentibus ; foliis caulinis
trisectis, divisionibus petiolulatis vel sessilibus ovatis bi-tri-lobatis, lobis
undique argute serratis subincisis, ramealibus scepius indivisis ovatis
vel oblongis subsessilibus; pedunculis brevibus subsolitartis ; floribus
parvulis; sepalis ovatis late hyalino-marginatis petala obovata (flava)
subcequantibus ; acheniis ovatis compressis marginatis loevibus stylo
brevi apiculatis in capitulum globosum aggregatis.

Has. Mountains of West Maui, Sandwich Islands.
Var. @. foliis caulinis magis divisis, divisionibus longe petiolulatis.
Has. Mountains of Kauai, Sandwich Islands.

Stem apparently attaining several feet in length, slender, much
branched above, reclining or ascending, hirsute with a scattered
pubescence, in general appearance nearly glabrous. Leaves trisected,
membranaceous, minutely and sparsely hairy, but appearing nearly
glabrous; the lower cauline on long petioles; their divisions ovate,
short-petioled or sessile, 2 or 3 inches long, sharply serrate all

12 PHANEROGAMIA.

round and more or less incised, the lateral 2—3-lobed, the middle one
trifid. Leaves of the branches sometimes divided, commonly undi-
vided and almost sessile, serrate, ovate or oblong, an inch or two
long, and becoming smaller on the ultimate flowering ramifications.
Peduncles terminal and dichotomal, an inch long, or less, one-flowered.
Flowers (all on a far advanced plant) not larger than those of £.
recurvatus ; the somewhat closed sepals and petals otherwise much as
in the preceding species. Mature achenia only a line long, ovate-lenti-
cular, compressed, margined much as in f. repens, smooth, tipped
with a small, straight or somewhat recurved, abrupt, subulate beak,
consisting of the persistent style. They form a globose head of only
a quarter of an inch in diameter—This description is from the Maui
specimen. The poor specimen from the island of Kauai, indicated as
a variety, is more hairy; the cauline leaves" more dissected, and the
divisions on elongated partial petioles; in these respects approaching
the R. Hawaiensis, although very different in the fruit, &c.

This species is readily distinguished from the preceding by its
weaker, reclining habit, less divided and much less hairy leaves,
smaller flowers, and, more definitely, by its carpels of less than half
the size, of a different shape, and abruptly tipped with a small and
short style, instead tapering gradually into a conspicuous beak. In
the fruit, as well as the foliage, it bears more resemblance to some of
the coarser North American forms of the polymorphous &. repens,
from which the small flowers with short petals should distinguish it.

4. CALTHA, Linn.

1. CALTHA SAGITTATA, Cav.

Caltha sagittata, Cay. Ic. 5, t. 414; DC. Syst. 1, p. 307; Hook. f. in Bot. Mag., t.
4056, & Fl. Antare. p. 228.

C. multicapsularis, Soland. in herb. Banks; Forst. in Trans. Linn. Soe. 8, p. 324.
Psychrophila sagittata & P. andicola, C. Gay, Fl. Chil. 1, p. 50, t. 2.

Has. Orange Harbour, Fuegia.

This curious and widely distributed Antarctic American species is

RANUNCULACE &., 13

more closely allied to true Caltha, as Dr. Hooker remarks, than to C.
appendiculata.

5 PSYCHROPHILA.

Sepala 5, crassiuscula, appendice membranacea caudato-attenuata vel
obtusa brevi aucta, marcescentia, persistentia. Petala nulla. Stamina
9-9. Ovaria 2-9, pauci-(2—-8-)ovulata. Folliculi 2-3-sperma. Semina
pendula.—Herbee nan, ceespitosce, caudicibus petiolorum vaginis latis-
simis scariosis obtectis, lamina foliorum 2-3-loba crassa, basi intus
appendicibus binis sursum erectis appressis instructa ; pedunculo bre-
vissumo unifloro ; floribus subdioicis.

CALTHA, § PsycHROPHILA, DC. Syst. 1, p. 307, excl. sp.; Hook. f. Fl. Antare. p. 228.
PsYCHROPHILA, C. Gay, Fl. Chil. 1, p. 47, excl. char. & sp. 2-3.

Distinguished from Caltha as a genus by the membranaceous appen-
dages which terminate the thickish sepals, either attenuated and as
it were caudate, as in P. appendiculata, or short and blunt as in P.
dionecefolia ; by the jew stamens (from 5 to 9); and the few (2 to 9)
and 2—8-ovulate ovaries, and 2—3-seeded follicles ; the subdiccious flowers ;
and the very different habit. It must be confessed, however, that the
New Zealand species of Caltha, recently illustrated by Dr. Hooker,
having narrow and attenuated sepals, is too nearly intermediate for
the satisfactory discrimination of the two genera. Of the curious in-
ternal appendage of the leaves, adnate to the base of the blade, we
have indeed an approach in the inflexed auricles of Caltha sagittata.
This appendage is doubtless of the same nature as that on the inner
face of the petals of Hrythroxylum, and of certain Sapindacee.

1. PSYCHROPHILA APPENDICULATA, Gay, J. c.

Caltha appendiculata, Pers. Ench. 2, p. 107; DC. Syst. 1, p. 307; Deless. Ic. Sel.
1, t. 48; Hook. f. Fl. Antare. p. 228.
C. paradoxa, Solander, in herb. Banks. ; Forst. in Linn. Trans. 8, p. 324.

Has. Orange Harbour, Fuegia.
4

IA.
14 PHANEROGAM

9. PSYCHROPHILA DIONEHFOLIA, Gay, l. ¢.

Caltha (Psychrophila) dionexfolia, Hook. f. in Lond. Jour. Bot. 2, p. 306, & Fi.
Antare. p. 229, 284.

Has. Orange Harbour, Fuegia.

Very striking, although on a diminutive scale, is the resemblance
between the blade of the leaf of this curious little plant and that of
our American Flytrap, the Dionea muscipula. Dr. Hooker’s account
and admirable figure of this species leave nothing to be added, except
the fruit. Of this one or two of our specimens show the vestiges, in
the three dehiscent, empty follicles. These were short and broad,
ovate, slightly pointed, not very different from those of P. appendicu-
lata, and had evidently borne only two seeds at most. Dr. Hooker
informs us that a small specimen of this plant was brought home by
Forster, mixed with Ozalis Magellanica, although not described by
him; and moreover that, “in the southern parts of Tierra del Fuego
it is a very common plant, covering large tracts of ground with a
carpet of deep but shining green, upon which the stellate flowers have
a very pretty appearance.”

Orv DILLENIACES.

1 DAVILLA, Vand., DC.
1. Davina RuGOSsA, Poi.

Davilla rugosa, Poir. Dict. Suppl. 2, p. 457; St. Hil. Fl. Bras. Mer. 1, p. 18.
D. Brasiliana, DC. Syst. 1, p. 405, & Prodr. 1, p. 69; Deless. Ic. Sel. 1, t. 71.

Has. In forests, near Rio Janeiro; where it is apparently common,
as it is gathered by most collectors.

2. CAPELLIA, Blume.

CAPELLENIA, Hasskarl, in Flora (Allgem. Bot. Zeit. Regensb.) 1847, p. 659 (nom.
emend. ).

1. CAPELLIA BIFLORA, Sp. Nov. (Tab. 1.)

C., foliis ovato-oblongis repando-serratis acutis obtusisve subcoriaceis con-
fertim recte-venosis, basi stipuliformi petiolum utrinque latissume margi-
nante mox sursum dissiliente ; pedunculis bifloris ; ovaris 12.

Has. Ovolau, Feejee Islands.

Tree about twenty feet in height; the branchlets glabrous, or
clothed when young with a fine deciduous pubescence, minutely ver-
rucose, annulate by the scars whence the sheathing bases of the stipu-
liform wings of the petiole have fallen. Leaves alternate, rather
coriaceous, ovate-oblong or oval, acutish, not acuminate, from 5 to

ANEROGAMIA.
16 PH

8 inches long in the specimens, repandly serrate with small teeth,
glabrous, or minutely pubescent underneath when young, prominently
straight-veined, in the manner of a Chestnut leaf, the veins OF ribs rather
close, about 20 on each side, connected by @ reticulation of minute
transverse veinlets. Petioles an inch or two 1n length, thick, dilated at
the base, winged for their whole length by a decurrent prolongation of
the lamina, or rather, by a pair of wholly adnate stipules. These
sheath the apex of the stem and enclose the bud in gemmation (there
are no bud-scales); the clasping insertion separates by a clean scar
soon after the leaf expands, and the line of separation extends from
below upwards, cutting these stipular wings away from. the petiole,
and at length they break away from the lamina, to which they are
united by an obversely wedge-shaped apex, adapted to a narrow sinus
of the blade itself. Peduncles solitary, terminal, becoming lateral and
opposite the leaves by the continuation of the stem, pubescent, longer
than the petiole, usually twojflowered, sometimes one-flowered, the
apex articulated with the flower. Sepals 5, orbicular, coriaceous,
glabrous, or the exterior minutely puberulent outside, strongly imbri-
cated in sestivation, persistent. Petals 5 (in the bud obovate),
“white,” deciduous. Stamens indefinite, inserted in several series
into the dilated base of a thickened torus, the exterior shorter:
anthers linear, longer than the filaments, adnate (introrse?), the
inner successively longer and recurving over the outer series. Ovaries
12 (or sometimes 102), uniserial, united by the greater part of the
length of the ventral face to a thickened central prolongation of the
axis. Styles terminal, distinct, long and filiform, erect with the
upper part recurved, persistent. Stigma strictly terminal, emarginate.
Ovules 10 or 12 in each ovary, occupying nearly the whole length of
the ventral suture, in a double row, horizontal, on short and thick
funiculi, globose-pyriform. Mature fruit unknown. Immature and
partly grown fruit a ring of 12, apparently follicular, ovate-oblong,
compressed carpels, coalescent by their narrow ventral face only, for
half their length or more, to the thickened but dry axis, several-
seeded. Arillus none? The fully expanded flower would probably
measure about two inches in diameter.

The genus Capellia was proposed by Blume (in his “Bijdragen tot
de Flora van Nederlandsch Indie,” Part 1, p. 5), in 1828, for a tree
found in the mountains of the Island of Nusa Kambanga. I am not

DILLENIACES. 17

aware any fuller description of it has appeared, or that any other
species have since been published (except the doubtful C.? pauciflora,
Zoll. & Moritz., from Java), although to the same genus undoubtedly
belongs No. 329 of Lobb’s collection at Singapore, in 1846, as is noted
by Planchon, in the Hookerian herbarium, as well as the two species
here characterized. The fruit of these Feejeean plants unfortunately
was not collected; but the halfgrown carpels of the present species
plainly indicate a capsular fruit. As the genus has not been figured,
I devote a plate to its illustration, even from our imperfect materials.
The anthers, unfortunately, in all the flowers I could examine, had
been in a good degree destroyed by larve. I think they open longi-
tudinally; while in the plant from Singapore, above-mentioned, the
cells open only at the top, by a pore or short slit.*

PLATE 1.—CAPELLIA BIFLORA: a branch in flower. Fig. 1. Ver-
tical section of an unopened flower. 2. A petal. 3. A separate sta-
men, enlarged. 4. The pistils divided transversely, moderately mag-
nified. 5. One of the pistils detached from the fleshy central axis,
its ovary longitudinally divided, more magnified. 6. Longitudinal
section of the halfgrown fruit; one of the carpels laid open, to show
the young seeds, moderately enlarged.

2. CAPELLIA MEMBRANIFOLIA, Sp. Nov.

C. foliis oblongis obovato-oblongisve repando-serratis membranaceis recte-
venosis costis subdistantibus ; pedunculis bi-tri-floris ; floribus parvulis ;
ovarus 8-10.

Has. Ovolau, Feejee Islands.

Of this we have only a single specimen, bearing one or two flower-
buds, which, so far as I can see, differs from the last principally in its

* While this sheet was in type I have received, through Prof. Miquel, specimens of
the original Capellia multiflora, from Java. It bears a close general resemblance to the
present species, from which, however, it is clearly distinguished by the very narrow wings
or margins of the petiole, which do not wholly clasp the stem at the insertion, as well as
by the more numerous flowers and fewer ovaries. A few of the innermost anthers are
abruptly very much longer than the others, as described by Blume. The anthers open

" oly by two pores at the apex.
5

18 PHANEROGAMIA.

thinner, membranaceous leaves (9 inches, or more in length), with
fewer and more distant primary veins, in its perhaps several flowered
peduncles (though I have noticed only three), and in the flowers being
to all appearance only half as large, and with rather fewer ovaries.
I think that the axis of the united pistils is not so much thickened.
The stamens of the flower-bud examined are in good condition. The
exterior are shorter, and perhaps imperfect. The others are succes
sively longer and more recurved: anthers oblong-lmear and linear,
shorter than the filaments, adnate-introrse; the cells opening longi-
tudinally down the margin apparently for their whole length ;—cha-
racters which would refer these Feejeean plants to Dillenia, if the fruit
would permit. To the same species I doubtfully refer a sterile branch,
with the thin leaves a foot anda half long. Perhaps these are all
mere varieties of the foregoing.

From Caldera, Mindanao, we have foliage of one or perhaps two
apparently allied Dilleniacew, in an undeterminable condition : and
from Luzon (near Manilla), a single leafy specimen of what may be
Wormia apetala, Gaud.

3. ADRASTEA, DC.

Sepala 5, persistentia, oblonga, subulato-acuminata, marginibus membra-
naceis petaloideis. Petala 5, ovalia, calyce breviora, caduca. Sta-
mina 10, equalia, conformia, persistentia: anther oblonge, apice
emarginate, filamento plano longiores, loculis longitudinaliter subin-
trorsum dehiscentibus. Ovaria 2, distincta, conica, in stylum subu-
latum attenuata: stigma terminale minutissima, Ovulum unicum, é
basi suture ventralis adscendens, anatropum. Folliculi membranacet,
intus longitudinaliter dehiscentes. Semen erectum, ovoideum, leve, ad
hilum arillo brevissimo scarioso appendiculatum.—Suffrutex lineari-
folius, floribus flavis.

Aprastna, DC. Syst. 1, p. 424, & Prodr. 1, p. 738; Endl. Gen. no. 4752.

1. ADRASTEA SALICIFOLIA, DC. le.

Has. Near Sydney, New South Wales.

DILLENIACE SX. 19

Stems a foot high from a creeping base, erect or ascending, slender,
cinerous-puberulent, soon glabrous. Leaves linear and oblong-linear,
an inch or an inch and a half in length, obtuse, mucronate, the revo-
lute margins sparingly and obsoletely callose-denticulate above the
middle, veinless; the midrib conspicuous; the upper surface glabrous;
the lower, like the short petioles and young branchlets, canescent with
a soft silky pubescence. Flowers solitary and subsessile at the extre-
mity of short leafy branchlets, small. Sepals 3 lines long, imbri-
cated in eestivation, persistent, a thickish carinate axis silky-pubescent
externally, and produced at the apex into a subulate or subaristate
acumination, the rest membranaceous and partly petaloid, yellowish.
Petals 5, imbricated in estivation, oval, yellow, rather shorter than
the calyx, early deciduous. Stamens 10, about as long as the petals,
persistent, equal in length, and symmetrically arranged. Filaments
shorter than the anthers, flat. Anthers elongated-oblong, emarginate
at the apex; the cells somewhat separated by the rather broad and
flat connective, opening longitudinally by a slightly introrse line. Ova-
ries 2, distinct, erect, somewhat obcompressed, conical, tapering regu-
larly into a short, subulate, and very acute style: stigma terminal,
simple, very minute. Ovule solitary, ascending from the base of the
ventral suture, on a very short funiculus, obovoid, anatropous. Fruit
of 2 thin and membranaceous follicles, shorter than the persistent
calyx, pointed with the persistent style, dehiscent by the ventral
suture, one-seeded. Seed filling the lower part of the follicle, ovoid,
with a smooth testa, and with a minute, thin, and scarious, orbicular
arillus, or rather a caruncle, which is apparently attached to the
micropyle.

I am not aware that anything has appeared respecting this plant
since it was characterized by DeCandolle, thirty-five years ago, from
specimens he examined in the Lambertian herbarium. I have there-
fore confirmed and completed the character (which was left in some
respects imperfect by DeCandolle), from the specimens collected in the
Exploring Expedition. This was the more necessary since the altera-
tions made by Endlicher (I know not whether from actual materials),
viz., “capsule coriacez,” and “arillo membranaceo,” do not accord
with the plant before me. For the carpels are membranaceous, as
said by DeCandolle, and the arillus, if such it be, is almost obsolete.

20 PHANEROGAMIA.

4, HIBBERTIA, Andr., DC.

1. HipsEeRTIA VOLUBILIS, Andr.
Hibbertia volubitis, Andr. Bot. Rep. t. 126; Salisb. Parad. Lond. No. 73; DC.
Syst. 1, p. 426.

Dillenia volubilis, Vent. Choix. Plant. p. 11, t. 11; Pers. Ench. 2, p. 72.
D. speciosa, Curt. Bot. Mag. t. 449, excl. syn.

Has. Near Sydney, New South Wales, apparently very common.

2. HippERTIA DENTATA, [2. Br.

Hibbertia dentata, R. Br. in DC. Syst. 1, p. 426, & Prodr. 1, p. 74; Bot. Reg. t.
282; Bot. Cab. t. 347; Bot. Mag. t. 2338.

Has. With the preceding.

3. HIBBERTIA LINEARIS, R. Br. l.c. &.
Var. @. OBTUSIFOLIA: foltis latioribus obtusis; ccet. ut in forma typica.
Hibbertia obtusifolia, DC. Syst. 1, p. 429, & Prodr. 1, p. 75, ex. char.
Has. Port Jackson and Sydney, New South Wales (a and £).

The variety, which differs from the ordinary form of the species*

only in the broader and obtuser leaves, well accords with the character
of H. obtusifolia, DC.

4, HiIBBERTIA DIFFUSA, R. Br.

H. humilis, ramosissima; ramulis puberulis; foliis cuneatis spathu-
latisve emarginatis apicem versus 2—5-dentatis nunc integris calycibus-
que glabris ; caulibus diffusis.

Hibbertia diffusa, R. Br. in DC. Syst. 1, p. 429, & Prodr. 1, p. 74.

Var. @. OBLONGATA (DC. l.c.): foliis oblongo-cuneatis seu oblanceolatis
seepius mtegris cinereo-puberulis vel glabratis ; ramis suberectis.

DILLENIACEA. 21

Var.? y. CANESCENS: foliis spathulatis oblanceolatisve integerrimis emar-
ginatis utringue cinereo-pubescentibus ; sepalis extimis dorso puberulis ;
Jloribus paullo majoribus.

Hibbertia canescens, Planchon, in Herb. Hook., non’ Sieb. ex. char.

Has. Sydney, New South Wales. Var. 8, Puen Buen; and y,
Hunter’s River, New South Wales.

The stems of the typical H. diffusa are about a span high, from a
thickened woody base, very diffuse; the branchlets minutely pubes-
cent. The leaves are wholly glabrous, except the very youngest, quite
variable in size and form; the larger an inch long and half an inch
wide, dilated-cuneiform, baneriie into a narrow base, emarginate, and
mostly coarsely and obtusely 2—5-toothed towards the summit; others
are narrowly spatulate, or cuneate, half an inch or less long, only one
or two lines wide, and usually entire. Sepals glabrous, or nearly so.
Ovaries glabrous—What I refer to var. 2 has rather more upright
branches, which, like the narrower and sometimes oblanceolate, or
lanceolate, mostly entire leaves, &c., are clothed with a fine cinereous,
but more or less deciduous pubescence. This is no doubt a mere variety
of H. diffusa. But it runs into the var. y, which is stouter and more
upright, with rather larger flowers; the leaves all entire, oblanceolate
or narrowly cuneiform-spatulate, and clothed with a more cinereous
pubescence, which is tardily somewhat deciduous. There is no dif-
ference in the flowers, except that the exterior sepals are more or less
pubescent outside. It has been gathered by Dr. Hooker and others
around Sydney, and is named H. canescens, Sieb., in the Hookerian
herbarium: but that species is characterized as having pubescent ova-
ries, and leaves pilose above and glabrous underneath.

0. HIBBERTIA FASCICULATA, R. Br. 1. ¢.

Has. Sydney, New South Wales.

6. HipspertrA CAMPHOROSMA.

H. pilosula; foluis subulato-filiformibus fasciculatis brevibus quam
6

29 PHANEROGAMIA.

flores vie longioribus; staminibus 9-10; carpellis 3 dispermis ; caule

diffuso.

Pleurandra Camphorosma, Sieb. in Spreng. Syst. Veg. Cur. Post. p. 191.
Has. Sydney, New South Wales.

A shrubby plant, a foot or less in height, with numerous, diffuse
and spreading branches; the branchlets minutely hairy. Leaves subu-
late-filiform, obtuse, narrowed at the base, mostly /fascicled, minutely
hairy, at length nearly glabrous, about a quarter or one third of an
inch long; those next the flowers scarcely longer than the calyx.
Flowers terminal, or terminating very short branchlets, thus appear-
ing lateral, not pedunculate. Sepals mucronate-cuspidate, shorter
than the dilated-obovate petals. Stamens 9 or 10 in three clusters
alternate with the three ovaries, usually four in one cluster and three
in each of the others: filaments distinct, longer than the oval anthers.
Carpels 3, glabrous, ovoid, two-seeded. Seeds ovoid; the base sur-
rounded with a small hyaline and much laciniated arillus.

This species is very closely allied to H. fasciculata ; from which it
principally, and perhaps too slightly, differs in its more diffuse habit,
rather smaller flowers, and shorter leaves; those of H. fasciculata

being from half an inch to an inch long, and much exceeding the
flowers. Both appear to be common around Sydney.

5. PLEURANDRA, Labill.

1. PLEURANDRA SCABRA, 2. Br.
Pleurandra scabra, R. Br. in DC. Syst. 1, p. 419, & Prodr. 1, p. 72.

Var. 8. foliis oblongo-ovalibus vel obovatis ; ramis diffusioribus ; ramulis
angulatis fere glabris.

Has. Sydney, New South Wales.

We have both the var. @, which has more diffuse or spreading

DILLENIACE &. 93

branches, and smoother as well as broader, oblong-oval or obovate
leaves; and the typical form, with narrowly oblong or sublinear
leaves, more narrowed at the base. In the Hookerian herbarium it is
named P. parviflora (from Sydney, Dr. Hooker, and Bass’s Straits, Mr.
Bynoe); but it has no stellate pubescence. The upper surface is sca-
brous with short, hispid hairs, which leave a broad base or scar when
they fall away. |

2. PLEURANDRA ACICULARIS, Labill.

Pleurandra acicularis, Labill. Fl. Nov. Holl. 2, p. 6, t. 144; DC. Syst. 1, p. 421,
& Prodr. 1, p. 73.

Has. Sydney, New South Wales.
The specimens accord with Labillardiere’s figure. My Tasmanian
specimens from Mr. Gunn have shorter and more crowded leaves.
3. PLEURANDRA stricta, R. Br.,in DC. l. .

Has. Port Jackson, New South Wales (where Mr. Brown first
gathered it).

Orn. MAGNOLIACES.

1. DRYMIS, Forst.
1. Drymis WInTERI, Forst.

Drymis Winteri, Forst. Char. Gen. p. 84, t. 42; Linn. f. Suppl. p. 269; Mill. Fasc.
Ic.; DC. Syst. 1, p. 442; Hook. f. Fl. Antare. p. 229.

D. punctata, Lam. Dict. 2, p. 230, & Tl. Gen. t. 494.

Winterana aromatica, Soland. in Med. Obs. 5, p. 46, t. 1.

Wintera aromatica, Murr. Syst. p. 507; Humb. & Bonpl. Pl. Aiquinoct. 1, p. 209.

Has. Orange Harbour, Fuegia.

Var. CuILensis. (Forma angustifolia, nempe foliis oblongo-lanceolatis.)

D. Chilensis, DC. Syst. 1, p. 444, & Prodr. 1, p. 78; Deless. Ic. Sel. 1, t. 88; Gay,
Fl. Chil. 1, p. 61.

Has. Chili, near Valparaiso, in deep ravines.

The full accounts that have been given of the well-known Winter’s
Bark leave nothing to be added here. Our specimen from Valparaiso
belongs to a state with wnusually narrow, oblong-lanceolate leaves. I see
little to militate against Dr. Hooker's conclusion respecting the species,
at least so far as respects DeCandolle’s D. Chilensis. But that the
same species should extend from Fuegia and Juan Fernandez to New
Granada, and even to Mexico, over 86° of latitude, is indeed remark-
able; especially when we consider that all other species of Magnoliacee
have a narrow geographical range.

Orv ANONACE A.

1. ANONA, Linn.
1, ANONA PALUSTRIS, Linn.

Anona palustris, Linn. Spec. ed. 2, p. 757? St. Hil. Fl. Bras. Mer. 1, p. 832; Mart.
Anonac. F]. Bras. p. 11.

A. Americana, etc. Pluk. Alm. t. 135, f. 1.
Has. Near Rio Janeiro, Brazil: abundant in the Mangrove swamps.

I have not at present the means of comparing the Brazilian and
West Indian specimens of the Alligator-apple, Monkey-apple, or Cork-
wood, as it is variously called; but our specimens evidently belong to
what the learned Von Martius takes for Anona palustris, and have
nearly membranaceous leaves, in shape like those of the Pear, as he
describes them.

2. ANONA squamosa, Linn., Mart. 1. c.

Has. St. Jago, Cape de Verde Islands; introduced. Dr. Pickering

records it, as an introduced tree, at the Society, Navigator, and Tonga
Islands.

The Sour-sop or Guanabo, a West Indian species, is now widely
diffused throughout the tropics. For an admirable historical account
of the cultivated species, vide Martius, Adnotationes de Historia

Anonarum Cultarum, in his Flora Brasiliensis, 1. c.
7

26 PHANEROGAMIA.

3. ANONA CHERIMOLIA, Mill., Mart. 1. ¢.
Has. Lima: cultivated. “A native in the ravines to the northward.”
A sterile shoot, with broader leaves than ordinary, of the well-

known Cherimoya, the fruit of which is so much prized in Peru; and
which is ranked as one of the three best fruits in the world.

2, ROLLINIA, S&S. Mil.
1. Roiuinra Loneiroiia, St. Ail.

Rollinia longifolia, St. Hil. Fl. Bras. 1, p. 29, t. 5; Mart. Anonac. Fl. Bras. p. 17.
Anona, dolabripetala, Raddi, Act. Soc. Mod. 16, p. 15; DC. Prodr. 1, p. 86.
A. xestropetala, Spreng. Syst. 2, p. 641, fide Mart.

Has. Near Rio Janeiro, Brazil.

2. RoLLINIA PARVIFLORA, St. Hil.

Rollinia parviflora, St. Hil. Fl. Bras. Mer. 1. c.; Mart. Anonace. 1. c. p. 19, t. 6, f. 1.

Has. Near Rio Janeiro. (A form intermediate between var. a and
2, of Martius.)

38. GUATTERIA, Ruiz & Pw.

1. Guartertia psitopus, Mart.

_ Guatteria psilopus, Mart. Anonac. Fl. Bras. p. 27, t. 7, f. 1.

Has, Organ Mountains, Brazil, in the vicinity of Rio Janeiro.

ANONACES, o7

2. GUATTERIA PALLIDA, Blume?
Guatteria pallida, Blume, Bijdr. p. 20, & Fl. Jav. Anonac. p. 97, t. 48.
Has. Bajos, near Manilla, Luzon.

A poor specimen, with young fruit, destitute of flowers, which I
can only doubtfully refer to Blume’s G. pallida, with the figure of
which it pretty well accords.

4,5 ANAXAGOREA, &. Hil.
1, ANAXAGOREA Luzonensis, Sp. Nov.

A. foliis lanceolato-oblongis utrinque angustatis promisse acuminatis con-
coloribus; staminibus conformibus; ovariis 3 levibus in stylum gra-
cilem attenuatis ; folliculis longe clavato-stipitatas.

Has. Mountains near Batfios, in the Island of Luzon. (Also in
Cuming’s Philippine collection, No. 831.)

Branches slender, glabrous, as is the whole plant. Leaves membra-
naceous, from 3 to 5 inches in length, lanceolate-oblong, or oblong,
narrowed into an acute base, conspicuously acuminate, green both
sides, not at all glaucescent underneath: petioles about 3 lines long.
Peduncles opposite the leaves, 2, or in fruit 3, lines in length, one-
flowered, furnished with a very small clasping bractlet near the
apex. Calyx of 3 ovate and obtuse sepals, united at the base, mi-
nutely ciliolate, otherwise glabrous, about one-third the length of the
corolla. Petals lanceolate-oblong, the 3 outer ones about half an
inch long; the 3 inner a little smaller. Torus not at all en-
larged. Stamens about 30, equal and similar, all perfect and with
short filaments: anthers linear, with a short obtuse tip, three or
four times the length of the filament. istils 3, slightly stipitate
in the flower-bud, smooth throughout; ovary ovoid-oblong, tapering into

28 PHANEROGAMIA.

a style as long as itself: stigma obtuse, little dilated, not discoid.
Ovules 2, collateral, erect from the base of the cell. Follicles 2 to 3,
or solitary, gibbous, minutely wrinkled or roughened, tipped with a
persistent style, which is a line and a half long, tapering gradually
into the smooth and thickened stipe, the whole together club-shaped, and
over an inch long, two-valved at the apex; the valves thickened,
corticose-coriaceous, at length splitting through the clavate-thickened
stipe. Seed often by abortion solitary, oval, compressed; the testa
chestnut-coloured, smooth and shining.

This species, which appears to be a true Anaxagorea, notwithstand-
ing the extreme fewness of its pistils, differs from A. Javanica of
Blume in its narrower leaves, not at all glaucous beneath, in its
slender styles, and especially in its uniform, wholly perfect stamens ;
in which last respect it accords with the genuine American species.
Our specimens exhibited good fruit; but there was only a single un-
opened flower to examine.

5. RICHELLA, Nov. Gen.

Calyx subtrilobus, persistens. Corolla e petalis 6 ovatis, internis dimidio
brevioribus. Torus acetabuliformis. Stamina indefinita Guatterie,
etc. Ovaria plurima, libera, biovulata: styli majusculi, intus longi-
tudinaliter stigmatosit. Ovula suture ventral: juxta basum mserta,
adscendentia, superposita. Fructus e carpellis paucis, obovoideis, sub-
carnosis (exsuccis, siccatis coriaceis), indehiscentibus, breviter stipitatis,
monospermis. Semen erectum, magnum, samarordeo-nuciforme ; nempe,
testa coriacea marginibus alato-productis. Albwmen, embryo, ete.
ordinis. Arbor Uvarise facie.

1. RICHELLA MONOSPERMA, Sp. Nov. (Tab. 2).

Has. Ovolau, Feejee Islands; not uncommon.

A tree of considerable size, with glabrous leaves, branches, &c., and
with the aspect of an Uvaria. Leaves coriaceo-membranaceous, ellip-

ANONACES. 99

tical or oblong, obtuse or rounded at the base, the apex usually with
a short acumination, feather-veined (the principal veins slender)
and minutely reticulated, from 5 to 9 inches long, and 2 or 3
wide ; the midrib minutely verrucose-roughened underneath. Petioles
half an inch long. Peduncles solitary, axillary or opposite the leaves,
naked, an inch or more in length, one-flowered. Calyx small,
expanded, slightly three-lobed, persistent. Corolla of 6 petals in two
distinct series, thick and coriaceous in the dried specimen, minutely
silky-pubescent, deciduous; the three exterior spreading, ovate, sub-
acuminate, two-thirds of an inch in length; the three ¢nner similar in
shape but only half the size, more tardily opening, not spreading.
Stamens very numerous, crowded, in many ranks, occupying the thick-
ened annular margin of the saucer-shaped torus: filaments very short;
the large connective tipped with the truncate thickened apex which
is common in this family: anther-cells linear, extrorse, opening longi-
tudinally. Pistils numerous (about 20), inserted on the concave-
depressed centre of the dilated torus: ovaries minutely hairy, linear-
oblong, angled, about the length of the stamens, abruptly contracted
into a linear, thickish style, which is as long as the ovary itself, and
canaliculate and stigmatose for nearly the whole length of its internal
Jace. Ovules 2, mserted close together, but one above the other, on the
ventral suture, very near the base of the cell, obcompressed, ascending.
Fruit of few (in the specimen 3) ripened carpels from one flower:
these are coriaceous, but were probably fleshy in the living plant,
obovate, obtuse, contracted at the base into a short and thick stipe, and
indehiscent, at least, showing no indications of dehiscence: they are an
inch and a half long and an inch in thickness, and have rather thin
walls. Seed solitary, erect, fully an inch long and almost as wide, as
long and as broad as the cell, oval or obovate in outline, almost flat
on one face, the other angled in the middle, so that the transverse
section is triangular, the margins of the smooth, coriaceo-crustaceous
(chestnut-coloured) testa produced all round into a sharp, salient, wing-
like edge, which is notched on one side at the base: the hilum small,
next the base on the angled face. Albumen very oily, ruminated in
the manner of the family. HKmbryo minute, with oblong cotyledons
and a slender radicle.

The name fichea being preoccupied, I trust I may be permitted

somewhat to modify and prolong it, in order to dedicate this well-
8

30 PHANEROGAMIA.

marked genus to William Rich, Esq., the Botanist of the Expedition in
which these collections were made. According to Blume’s arrange-
ment of the family, this genus would stand next to his Polyalthia,
from which it is at once distinguished, as from all the others, by
its remarkable winged seed.

PLATE 2.—RICHELLA MONOSPERMA: in fruit. Fig. 1. A flower, of
the natural size. 2. Vertical section of the receptacle, with the sta-
mens and pistils, enlarged. 3. A stamen, magnified. 4. A pistil,
with the ovary vertically divided to show the two superposed ovules,
magnified. 5. Fruit and enclosed seed, divided transversely, of the
natural size. 6. Back view of a detached seed, of the natural size.
7. Same, seen from the other side, and divided transversely. 8. The
same, longitudinally divided, showing the embryo. 9. Embryo de-
tached and magnified.

6. UVARIA, Linn.

1. UVARIA opoRATA, Lam.

Uvaria odorata, Lam. Dict. 1, p. 595, & Ill. t. 495, f. 1; Blume, Fl. Jav. Anonac.
p- 29, t. 9, & 14, B.; Decaisne, Herb. Timor. p. 94.

Unona odorata, Dunal, Anonac. p. 108; DC. Syst. 1, p. 492, & Prodr. 1, p. 90.

U. leptopetala, DC. Syst. 1, p. 496; Deless. Ic. Sel. 1, t. 88.

U. ligularis, Dunal, 1. c.? DC. 1. ¢.? & Uvaria ligularis, Lam,?

Cananga, &. Rumph. Herb. Amboin. t. 65 (& 667).

Has. Samoan or Navigators’ Islands; at Savaii, &c.

That this is the Uvaria odorata I cannot doubt, although the
peduncles incline to branch, and to bear a larger number of flowers
than is usually attributed to that species; on which account, as I find
no other difference of moment in the character, I suspect that U.
ligularis, Lam. is no more than a variety. The petals are remarkably
long and narrow. The fruit, that of an Uvaria, is ovoid, and about
the size of a small cultivated cherry ; and the flat seeds are minutely
porose-pitted, as described and figured by Blume. I do not know
whether the flowers were noticed to be agreeably fragrant, and
whether they are used to scent the cocoa-nut oil which the Samoans

ANONACES. 31

apply to their hair; as they are used, according to Blume, in the
Malayan and Moluccan Islands. The tree is said to attain the
height of 60 or 70 feet, and its wood to be considerably valued as
timber.

2. UVARIA AMYGDALINA, Sp. Nov.

U. glabra, foliis oblongo-lanceolatis acuminatis basi sinw parvo subcor-
datis breviter wpetiolatis fere membranaceis supra nitidis reticulatis,
venis omnibus tenuibus; pedunculis brevibus unifloris; fructibus
ovoideo-globosis obtuse apiculatis in gynophorum sessilibus oligospermis.

Has. Ovolau, Feejee Islands.

A tree; its size not recorded. Branches slender, with a minutely
rimose or somewhat verrucose bark, glabrous. Leaves 5 or 6 inches
long, by an inch and a half in width, oblong-lanceolate, acuminate, a
little narrowed towards the base, which is subcordate, with a small
and narrow sinus, rather thin in texture, coriaceo-membranaceous, gla-
brous, the upper surface shining, but both surfaces of the same green
colour, reticulated with slender veins, of which the principal ones
are scarcely larger than the veinlets; the midrib minutely verrucose-
roughened underneath. Petiole little more than a line long. Flowers
not seen. Fructiferous peduncle half an inch long, axillary, one-
flowered, at the apex bearing the vestiges of the calyx; above which
the torus is immediately dilated into a globular gynophore, bearing 8
or 9 fleshy carpels. These in an unripe state, but some of them appa-
rently full-grown, are half or two-thirds of an inch long, sessile on the
gynophore, globose-ovoid, with a short obtuse point, containing a few
(from 5 to 8) rather large and flat horizontal seeds. Albumen deeply
ruminated, as in the order.

This species, having the carpels sessile on the receptacle, belongs to
the section Asimina, as the genus is disposed by Blume. Judging
from the figures of Blume’s three Javanese species of this section, and
from an arillus having been noticed by Martius in U. Brasiliensis, I
fear it will be hardly possible to maintain Asimina as a genus.

A leafy branch apparently of a second species exists in the Feejeean
collection, but without flowers or fruit.

Orv. MYRISTICACE &.

1MYRISTICA, Linn.
1. MyristIcA CASTANEZFOLIA, Sp. Nov.

M. foliis oblongis seu oblongo-lanceolatis sensim acutis basi rotundatis
glabris subtus albidis multicostatis; jfloribus aaillaribus, masculis
amentaceo-spicatis, rhachi incrassata sceepius furcata deflexa; fructu
subsessilt oblongo tomentuloso.

Has. Ovolau, Feejee Islands, on mountains; common.

A middle-sized tree. Branchlets thick, with a rough and wrinkled
bark, glabrous; the youngest parts barely ferrugineous-puberulent.
Leaves oblong and oblong-lanceolate, large, from 8 or 10 inches to a
foot in length, and 3 or 4 inches wide, of a thick coriaceous
texture, rounded at the base, toward the apex usually tapering gra-
dually to a more or less acute point, glabrous, even when young,
conspicuously feather-veined with 20 to 30 pairs of straight and
simple ribs, which are rather prominent underneath, anastomosing
near the margin, connected with somewhat conspicuous transverse
veinlets, green above, whitish underneath. Petiole thick and stout, an
inch or an inch and a half long. Flowers axillary, or slightly supra-
axillary; the fertile not seen; the sterile spicate on a thickened rhachis,
so as apparently to form a kind of ament. At least, the specimens
(which are in an advanced state) exhibit stout and lignescent deflexed
spines, or rhachises, solitary in the axils of the leaves, or above leaf-
scars, or some of them apparently supra-axillary, simple or oftener
forked, an inch or less in length and over 2 lines in thickness, squar-
rose with strong scars which mark the insertion of braets, and appa-

MYRISTICACEZ. 33

rently of sessile flowers also, which must have been closely imbricated.
At their apex remain three or four compacted sessile flower-buds,
each subtended by an ovate bract; the bracts and three-cleft? perianth
coriaceous, ferrugineous-pubescent outside. Androecium composed of 8
or 10 nearly sessile and linear anthers, which are closely united with
each other and with a central axis. Fruit nearly sessile, solitary,
axillary, oblong, obtuse, with a minute and oblique apiculation, an
inch and a quarter long, minutely tomentose with a ferrugineous pubes-
cence, apparently splitting into two valves at maturity. Seed an
inch long, ovoid-oblong, conformed to the pericarp.

2. MYRISTICA MACROPHYLLA, Sp. Nov.

M. foliis (bipedalibus et ultra) obovato-oblongis basi angustatis glabris
multicostatis subtus albidis, venulis inconspicuis.

Has. Ovolau, Feejee Islands; on mountains.

This is mentioned in Dr. Pickering’s MSS. notes as a smaller tree
than the foregoing, but with the leaves sometimes 24 feet long. In
the specimen, which is unfortunately destitute both of flowers and
fruit, the leaves are over 2 feet in length, and of a different shape
from those of the preceding species, being nearly obovate, and tapering
gradually from above the middle to the acute base ; the summit scarcely
acute. The ribs or straight veins are about 35 pairs, and very promi-
nent underneath; but the transverse veinlets are scarcely visible on
the smooth, but whitish lower surface.

3. MyRIsTICA HYPARGYRZA, Sp. Nov.

M. foliis oblongis basi obtusis glabris pluricostatis subtus argentato-

_ albidis ; floribus axillaribus in pedunculum brevem sew brevissimum
glomeratis, masculis breviter pedicellatis; fructu ovoideo-subgloboso
tomentuloso.

Haz. Tutuila and Savaii, Samoan or Navigators’ Islands; in deep

woods. Also at Tongatabu.
9

34 PHANEROGAMIA.

“A middle-sized tree, with large leaves; the fruit of the size of a
peach, and more woolly.” Branchlets as in No. 1, but less stout, and
smoother. Leaves oblong, or the larger lanceolate-oblong, 8 to 15
inches long and from 4 to 6 wide, rather coriaceous, rounded or
obtuse at the base, slightly or abruptly acute or acuminate, green above,
silvery-white underneath, where they are conspicuously feather-veined
with 14 to 20 pairs of straight ribs, which anastomose near the
margins; the veinlets obscure in young leaves, or sometimes mani-
fest. Petioles stout, an inch or an inch and a half long. lowers
axillary; the staminate ones glomerate on the dilated apex of a very
short peduncle, apparently only 5 or 6 in number, on pedicels a
line long, and with an ovate bractlet subtending the perianth. The
latter is only a line and a half long, ferrugineous-puberulent. Androe-
cium oblong, of 8 or 10 longitudinally united anthers. Pistillate
flowers occur (along with forming fruit) only in the specimens from
Tongatabu. These have short and very thick fertile peduncles, about
half an inch long, terminated by a globular forming fruit, beneath
which are borne a considerable number of sessile female flowers, with
a subglobose trifid perianth, and a globular ovary. The fruit-bearing
peduncles of some of the Samoan specimens likewise show indications
of having borne three or more female flowers., The full-grown fruit is
globose-ovoid, minutely tomentose. Seed an inch and a quarter long,
and 10 lines wide, enclosed in a deeply laciniate mace; the albumen,
&c. as in the genus.

4, MyrIstIcA INUTILIS, Rich, in herb.

M. foliis lanceolatis acuminatis basi subobtusis pluricostatis subtus ferru-
gineo-puberulis, adultis albicantibus ; floribus supra-axillaribus, mas-
culis atgue feemineis umbellato-congestis; pedunculo brevissimo vel
subnullo; fructu oblongo puberulo.

Has. Tutuila and Savaii, Samoan Islands; abundant. Tongatabu.

“A middle-sized tree,” with slender branchlets. Leaves lanceolate,
acuminate, 6 to 10 inches long, 12 to 2% inches in width, rather thin
in texture, obtusish at the base, copiously feather-veined after the
manner of all the foregoing species, and with the connecting veinlets

MYRISTICACESA. 35

obscure, glabrous above, when young /errugineous-pubescent beneath
(as are the branchlets and flowers), soon glabrate, but with the
shining lower surface still fulvous, or at length whitish. Petioles an
inch or less in length. Flowers supra-acxillary (the male decidedly so),
umbellate on a short (1-14 line long) or almost obsolete peduncle. Sta-
minate flowers numerous: pedicels 2 to 3 lines long, with a bractlet
at their apex. Perianth 2 lines long, clavate-oblong, three-toothed.
Androecium lanceolate, acute: the anthers about 5, slender, and with
their short filaments united longitudinally into a column, which is
nearly as long as the perianth. Female flowers several or numerous
on the very short peduncle, or subsessile rhachis, pedicellate like the
male flowers; the ovoid perianth three-lobed. Fruit oblong, or elon-
gated-oblong, ferrugineous-puberulent when young, at length almost
glabrate, 14 to 14 inches in length, apiculate; the pericarp thin and
two-valved; the mace much laciniated. Seed oblong, when full-grown
an inch long and scarcely half an inch wide, with a thin crustaceous
testa. Albumen, in the specimens, inodorous and tasteless.

This needs to be compared with the M. lancifolia of Poiret; which
is described as having olive-shaped fruits, two or three together, and
leaves not unlike those of the present plant, except that they are
much smaller, as also the fruits.

5. MyYRISTICA CINEREA, Poir. ?
Myristica cinerea, Poir. Dict. Suppl. 4, p. 83? Spreng. Syst. 3, p. 65?
Has. Caldera, Mindanao, Philippine Islands.

The leaves are larger than those of M. cinerea are said to be by
Poiret, and are rounded or obtuse at the base. The flowers and fruits
were not gathered, having fallen from the globular and sessile, axil-
lary receptacles. |

Orv. MENISPERMACE /&.

1 STEPHANIA, Lour.

SrepHania, Lour. Fl. Cochin, ed. Willd. 2, p. 746; Endl. Atakt. Bot. t. 49, 50, &

Gen. p. 827.
Cryrea, Blume, Bijdr. p. 26; Decaisne, Herb. Timor. p. 95, t. 18; Wight & Arn.

Prodr. Ind. Or. 1, p. 14.
SrepHania & CLYPEA, Miers, in Ann. & Mag. Nat. Hist. Jan. 1851.

1. STEPHANIA FORSTERI.

S. glabra; foliis peltatis ovatis hinc rotundatis illine subacutis vel sub-
acuminatis integerrimis, inferioribus obliquis; pedunculis petiolo
equilongis; floribus capitato-congestis ; capitulis umbellatis ; fl. mase.
sepalis 6-8; petalis 3-4 cuneatis; putamine annulari muricato.

Menispermum Japonicum, Forst. Prodr. Fl. Ins. Austr. p. 71, non Thunb.
M. peltatum, Forst. in Herb. Lambert.
Cocculus Forsteri, DC. Syst. 1, p. 517, & Prodr. 1, p. 96; Guill. Zeph. Tait. p. 76.

Has. Tahiti, Society Islands: open places in mountain woods.
Also, without flowers or fruit, in the Samoan or Navigators’ Islands,
and Tongatabu.

Stem twining, striate-grooved, glabrous, or nearly so. Leaves excen-
trically peltate, rownded-ovate, with a rather acute or acuminate apex,
3 or 4 inches long, glabrous, paler, but scarcely glaucescent under-
neath, entire, the lower transversely oblique, the uppermost shorter
than the petioles. Pduncles supra-axillary, as long as the petioles ; the
sterile compound-umbellate, with few primary rays of an inch or two

MENISPERMACE AZ. 37

in length; the secondary rays several, a quarter of an inch long, each
terminated by a close capitate cluster of staminate flowers. These
have six or more frequently eight sepals (oblanceolate-oblong and
obtuse), and dhree, or more frequently four petals. The latter are
broadly cuneate, cucullate, thickish, distinct, shorter than the sepals.
Filament or column at length longer than the sepals, supporting a
peltate disk, which is edged with a perfectly continuous, horizontally
dehiscent, annular anther, just as figured by Decaisne in Stephania
(Clypea, Decaisne) glaucescens. Fertile flowers not seen; but the
fructiferous peduncle shows them to have been capitate at the apex of
the (simple?) rays of the umbel. Drupe with a bony annular puta-
men, which is strongly muricate with transverse processes, the sides
hollowed. Embryo slender, terete, curved almost into a ring. Albu-
men scarcely any.—The sterile specimen from Tongatabu has rounder
leaves; but is likely to belong to the same species.

The floral envelopes vary from trimerous to tetramerous in diffe-
rent flowers of the same individual, in this, and, I believe, also in
other species: the presence or absence of an outer row of sepals
(which perhaps are rather to be regarded as bractlets), in the male
flowers, is scarcely of generic importance: and the number of cells
which enter into the composition of the annular anther cannot be
ascertained in such species as this and Stephania glaucescens, although
it is probable there are only two. The genus Clypea, as newly pro-
posed by Mr. Miers, is therefore not likely to be maintained. Hence
I refer all our species to the older genus.

2. STEPHANIA GAUDICHAUDII, Sp. Nov.
S. foliis peliatis late ovatis subacuminatis basi truncatis sew ovato-sub-
cordatis subtus ramulisque junioribus molliter pubescentibus ; capitulis

umbellatis ; sepalis petalisque fl. foem. 3-4.

Has. Near Sydney, New South Wales. Also, Puen Buen: foliage
only, of a smoother form.

This is doubtless the plant mentioned by Decaisne (Herb. Timor.
p. 95) as a species of Clypea, collected by Gaudichaud at Port Jack-
10

38 PHANEROGAMIA..

son, and which Mr. Miers, who has examined a specimen in the Can-
dollean herbarium, informs me is a Clypea, as he distinguishes the
two genera. We have only the female plant, in flower; the sepals
and petals of which, however, are as frequently jour as three in
number. The leaves broadly ovate, obtusely or slightly pointed, excen-
trically peltate, either truncate or the smaller ones more or less cordate
at the base, 2 or 3 inches long, glabrous above, softly pubescent under-
neath, longer than the petioles. Peduncles of the fertile inflorescence
an inch or more in length, pubescent, bearing from 5 to 8 wmbellate
short rays, each terminated by a dense or capitate umbellet or glome-
rule. The roundish petals are thickish, only half the length of the
sepals. )

3. STEPHANIA AUSTRALIS, Miers, l. c.
Has. Hunter’s River, New South Wales. The fertile plant only.
This is very probably the species enumerated by Mr. Miers, under

this name, founded on the Cissampelos australis of A. Cunningham’s
collection.

2. CISSAMPELOS, Linn.
1. CISSAMPELOS CAPENSIS, Linn.

Has. Cape of Good Hope: common in the vicinity of Cape Town.

2. CISSAMPELOS DISCOLOR, DC’?

Var. CARDIOPHYLLA : foliis plerisque reniformi-cordatis subtus canescenti-
bus ; pedunculis floribusque hirsutissimis.

Cissampelos Pareira, Blanco, Fl. Filip. p. 815? Walp. Rel. Meyen. p. 299?

Has. Small Island, in the Sooloo Sea.

MENISPERMACE &. 39

Only the sterile plant is in the collection. Some of the leaves being
“very broadly ovate, with a truncate subcordate base,” as DeCandolle
(Syst. 1, p. 534) describes his @. discolor, from the Moluccas, I have
ventured to refer our plant, although with some misgiving, to that
species. But most of the leaves are distinctly cordate, or even reni-
form-cordate, and canescent with a fine pubescence underneath, al-
though glabrous above; and the short peduncles, as well as the crowded
and cymulose, minute flowers, are very hirsute with spreading bristly
hairs. The corolla is a minute three-lobed cup, surrounding the base
of the androecium, which consists of four subsessile, peltately-disposed
anther-lobes.

3. NEPHROICA, Mers.

Petala bifida, lobis acutis vel acuminatis.— Cetera omnino Cocculi.

Nepuroica, Miers, in Ann. & Mag. Nat. Hist. Jan. 1851.
Nepurotr, Loureiro, Fl. Cochinch. (ed. Willd.) 2, p. 691.

1. NeruroicA FERRANDIANA.

N. ramulis retrorsum hirsutis ; foliis subcoriaceis ovatis vel ovato-lanceo-
latis acuminatis nunc obtusis mucronatis basi obtusa rariusve subcor-
data 3—5-nerviis, adultis glabratis ; pedunculis pauci—plurrfloris petiolo
brevioribus.

Coceulus Ferrandianus, Gaud. Bot. Voy. Freyc. p. 477, t. 101; Walp. Rel. Meyen.
p- 268.

Has. Kaala Mountains, behind Honolulu, Oahu; Districts of Puna

and Waimea, Hawaii; also on the mountains of Kauai, Sandwich
Islands.

Stems herbaceous, slender, twining, hairy with a reflexed pubescence,
at length nearly glabrous. Leaves rather coriaceous, varying, often
on the same individual, from ovate and subcordate, sometimes obtuse
and retuse, to ovate-lanceolate, commonly acuminate, mucronate, some-

AO PHANEROGAMIA.

what narrowed and obtuse at the base, where they are 3—5-nerved,
from one to 8 inches long, entire, strigose-pubescent when young,
soon glabrous, or nearly so, except on the veins. Petioles slender,
hairy. Peduncles shorter than the petioles ; the fertile few-flowered ; the
sterile few- or many-flowered. Sterile flowers scarcely as large as those
of Cocculus Carolinus, and of the same structure, except that the petals
are longer than the stamens and sharply two-cleft at the summit; the
lobes subulate. Fruit blue, much like that of Cocculus Carolinus ; the
linear cotyledons perhaps a little broader.

It is questionable whether the two-cleft petals sufficiently distin-
euish this genus from Cocculus, in which they are usually emarginate.
But, according to Mr. Miers, it includes a very distinct group of
plants, mostly natives of India.

4. LIMACIA, Lour., Mers.
1. LIMACIA VELUTINA.

L. pube fulvo-aurea undique velutina; foltis ovali-oblongis mucronato-
acuminutis bast rotundata trinerviis supra nunc glabratis ; pedunculis
fl. ster. supra-axillaribus fasciculatis petiolo subeequilongis apice con-

Jertim 3—5-floris.
Cocculus velutinus, Wall. Cat. no. 4970, a.
Has. Singapore.

The specimen consists of a sarmentose branch of the sterile plant,
clothed all over with a shining and soft, golden-coloured, velvety pubes-
cence. Leaves 3 to 5 inches long and 2 or more in width, oval or
oblong, the apex abruptly contracted into a narrow acuminate point,
which is prolonged into a conspicuous mucronation, coriaceous and
the upper surface becoming glabrous with age, three-nerved at the
rounded base, the prominent midrib sending off 3 or 4 strong pri-
mary veins on each side. Peduncles 3 to 5 in a fascicle a little
above the axil, an inch or less in length, bearing from 3 to 5 subses-
sile flowers in a close cluster. Sepals velvety-hirsute, subtended by
2 small bractlets; the three exterior ovate, about half the length of

LARDIZABALEAZ. 4]

the three concave inner ones, which are valvate in estivation. Petals
6, shorter than the inner sepals, smooth, cuneate, truncate, the
margins involute below. Stamens 6, about the length of the petals
and embraced by them, but free: the filaments somewhat clavate,

sparsely hairy: anthers two-celled, didymous, innate. Fertile flowers
not seen.

Orv LARDIZABALES.

1, GARDIZA BAL AS Rue G Poo.
1. LARDIZABALA TRITERNATA, Ruiz & Paw.

Lardizabala triternata, Ruiz & Pav. Fl. Peruv. & Chil.; DC. Syst. 1, p. 512;
Deless. Ic. Sel. 1, t. 91; Decaisne, Lardizab. p. 188; Gay, Fl. Chil. 1, p. 70.

Has. Chili, near Valparaiso; in high ravines.

There is reason to suspect, with M. Gay, that L. triternata and L.
biternata are not distinct species. Only one species is mentioned by
Dr. Pickering, as having been noticed by him. The characters de-
rived from the foliage appear to be inconstant.

1i

Orv. BERBERIDACE &.

1 BERBERIS, Linn.

* Antarctice.
1. BERBERIS BUXIFOLIA, Lam.

Berberis buaxifolia, Lam. Ill. t. 253, f. 3; Poir. Dict. 8, p. 619; DC. Syst. 2, p. 15;
Hook. f. Fl. Antarc. p. 231, t. 87; Gay, Fl. Chil. 1, p. 91.

B. microphylla, Forst. Comm. Goett. 9, p. 29; Willd. Spec. 2, p. 228; Poir. I. c.

B. dulcis, Sweet, Hort. Brit. ser. 2, t. 100, ex. Hook. f.

Has. Orange Harbour, Fuegia: common.

This species and the next one are beautifully illustrated by Dr.
Hooker, in his excellent Flora Antarctica. The berries, which are
elobular and blue, like those of the Mahonia section of the genus, are
agreeably acid, like those of the common Barberry, and have been
used for tarts, &c.

2. BERBERIS ILICIFOLIA, Forst.

B. ilicifolia, Forst. Comm. Goett. 9, p. 28; Linn. f. Suppl. p. 210; DC. 1. ¢.; Hook.
Fl. Antare. p. 230, t. 86; Gay, Fl. Chil. 1, p. 77.
B. lagenaria, Poir. Dict. 8, p. 619.

Has. With the preceding : common.

Some of the specimens have fine ripe fruit. The blue berries,

BERBERIDACEZE. 43

as large as our huckle-berries, are ovoid-globose, and pointed with the
thick persistent style, terminated by a dilated stigma; whence they
have been said to be “lageneform.” It is the handsomest species of
the genus, on account of its deep green and shining, Holly-like leavers
and very copious, golden yellow flowers.

* * Chilenses.
3. BERBERIS EMPETRIFOLIA, Lam.

B. empetrifolia, Lam. Tl. t. 253, f.4; DC. 1.¢.; Hook. & Arn. Bot. Mise. 3, p.
186; Hook. f. Fl. Antare. p. 231; Gay, Fl. Chil. 1, p. 93.

Has. Mountains near Santiago, Chili: “abundant in the upper
part of the middle region of the Chilian Cordilleras.”

The species is not contained in the Antarctic collection, which was
made on the moister eastern side of Fuegia. It belongs to a drier
region, and extends northwards along the Chilian Andes to latitude
30°, according to Gay. Our specimens of this and the succeeding
Chilian species have neither flowers nor fruit.

4, BERBERIS CHILENSIS, Gill.

Berberis Chilensis, Gillies in Hook. Bot. Misc. 3, p. 185; Gay, Fl. Chil. 1, p. 79.
B. ilicifolia, Bert, in Mere. Chil., ex Gay.

Has. Cordilleras, near Santiago, Chili; and between Santiago and
Valparaiso.

5. BERBERIS ACTINACANTHA, Mart.

Berberis actinacantha, Mart. in Schult. Syst. Veg. 7, p. 12; Hook. & Arn. Bot.
Mise. 3, p. 185; Gay, 1. c.
B. crispa & B. horrida, Gay, Fl. Chil. 1. ¢.?

Has. Between Santiago and Valparaiso. Also on the heights be-
hind Valparaiso.

44 PHANEROGAMIA.

6. Berperis FERox, C. Gay, l. ¢.

Has. Cordilleras, near Santiago, Chili. (Not improbably this is
the same species as the obscure B. cuneata of De Candolle.)

Orv. PAPAVERACE.

1. ARGEMONE, Linn.
1. ArcemMonE Mexicana, Linn.
Has. Callao to Yanga, Peru. Valparaiso and Santiago, Chili. Sand-

wich Islands.—Originally an American plant, no doubt; but now
widely distributed over the warmer parts of the world.

Orv. FUMARIACE &.

1FUMARIA, Linn.
1. FUMARIA OFFICINALIS, Lini.

Has. Valparaiso: doubtless introduced from the Old World, to
which the genus entirely belongs. Not being mentioned in Gay’s Flora
of Chili, it is probably of recent or scanty introduction. In North
America it occurs only in cultivated or recently cultivated grounds,
and it can scarcely be said to be naturalized.

2. FUMARIA AGRARIA, Lag.

Fumaria agraria, Lagasca, Elench. p. 21; Parlatore, Mon. Fumar. p. 72.
F. media, DC.; Gay, Fl. Chil. 1, p. 104.
F. capreolata, var. y. Hook. & Arn. Bot. Beech. Voy. p. 5.

Has. Valparaiso, near the coast. Doubtless introduced from Europe; »
but now so thoroughly and extensively naturalized in Chili that, ac-
cording to M. Claude Gay, it is at first view difficult to believe that it
was not indigenous to the country.

12

Orn. CRUCIFER &.

1. MATTHIOLA, R. Br.

1. MarruioLA MApDERENSIS, Lowe.
Matthiola Maderensis, Lowe, ex Lemann, Cat. Mad. Pl. in Hook. Niger FI. p. 78.

Has. Madeira, on the coast east of Funchal. (In fruit.)

2. CHEIRANTHUS, Linn., BR. Br.
1. CHEIRANTHUS MUTABILIS, L’ Her.

Cheiranthus mutabilis, L’ Her. Stirp. 1, p. 92; Bot. Mag. t. 195; Bot. Reg. t. 1481.
Dichroanthus mutabilis, Webb. & Berth. Phytogr. Canar. 1, p. 66.

Has. Madeira; with the preceding.

3. NASTURTIUM, R&. Br.
1. NASTURTIUM OFFICINALE, 2. Br.
Has. Sea-coast, east of Funchal, Madeira.
The Water-Cress is now very widely distributed over most parts of

the world that have been colonized by Europeans. It was probably
introduced into Madeira.

CRUCIFERA. AT

2. NASTURTIUM PALUSTRE, DC“, var.

Nasturtium sylvestre, A. Rich. Fl. N. Zel. (Voy. Astrolab.) p. 309, non DC.
N. terrestre, R. Br.; Hook. f. Fl. Antare. p. 14.

Has. Bay of Islands, New Zealand.

Although the specimens are incomplete, and without ripe fruit, I
think they cannot belong to the European Nasturtium sylvestre; but
they accord very well with what I take for a variety of N. palustre in

the Northern United States, with more or less hairy stems, and with
rather shorter pods than I have ever observed in the European plant.

4, BARBAREA, R. Br.
1. BARBAREA AUSTRALIS, Hook. /.
Barbarea australis, Hook. f. Fl. N. Zeal. p. 14.
Has. Bay of Islands, New Zealand.
Our specimens are indifferent, wanting the lower leaves. The
pods are not as large as they are described by Dr. Hooker, being only

an inch long, and barely a line wide. The radicle is occasionally
oblique, as in certain Sisymbria.

5. ARABIS, Linn.
1. ARABIS ALBIDA, Séev.
Arabis albida, Stev. in DC. Prodr. 1, p. 142; Hook. Niger Fl. p. 78.

Has. On rocks, Curral, Madeira.

A8 PHANEROGAMIA.

This plant is so named by Dr. Lemann, in the Niger Flora, above
cited: but the pods are longer than in the Caucasian plant. The
specimen has old fruit only.

6. CARDAMINE, Linn.

1. CARDAMINE GERANIIFOLIA, DC.

Cardamine gerantifolia, DC. Syst. 2, p. 268; Hook. f. Fl. Antare. p. 2338, f. 88.
Sisymbrium geraniifolium, Poir. Dict. 7, p. 218.

Has. Orange Harbour, Fuegia.

The specimen has only half-grown pods, like those of Commerson’s
collection, described by Poiret and DeCandolle. Dr. Hooker’s admi-
rable figure represents the flowering state only. The seeds are not
yet known.

2. CARDAMINE CORYMBOSA, Hook. /.
Cardamine corymbosa, Hook. f. Fl. Antare. p. 6; Hook. Ic. Pl. t. 686.
Has. Orange Harbour, Fuegia.

The specimens accord in all respects with those of Dr. Hooker, who
has the plant only from Campbell’s Island, in the same latitude as
Fuegia, indeed, but 110° distant in longitude.

3. CARDAMINE GLACIALIS, DC.

C. perennis, pilosiuscula vel glabra; foliis carnosulis pinnatisectis, seg-
mentis |-3-jugis sessilibus oblongis ciliolatis cum impari rotundato seu
reniformi scepius maximo; floribus majusculis; siliquis latiuscule
linearibus confertis strictis stylo brevissimo apiculatis, replo margi-
nibus incrassatis.

CRUCIFERA. 49

Var. a. pumila; segmentis lateralibus foliorum sepius parvis, terminali
reniformt. ,

Cardamine glacialis, DC. Syst. 2, p. 264, & Prodr. 1, p. 153.

C. antiscorbutica, Banks & Soland. in herb. Banks.

C. hirsuta, var. nivalis, Hook. f. Fl. Antare. p. 232.

Sisymbrium glaciale, Forst. Comm. Geett. 9, p. 82; Poir. Dict. 7, p. 218.

Var. 3. elatior ; segmentis lateralibus foliorum oblongis terminali obovato
vie mimoribus.

Cardamine nivalis, Gillies, in Hook. Bot. Mise. 3, p. 136.

Has. Orange Harbour, Fuegia; both varieties: the latter also from
Rio Negro, Patagonia. Andes of Chili, near Santiago, on the snow
line; a dwarf form, in fruit.

Root manifestly perennial, often tuberous. Stems 2 to 4 inches, or in
the larger variety a foot high, sparsely hairy, or glabrous, at least
above, ascending or erect, leafy to the top. Leaves pinnate, petioled,
rather fleshy: leaflets ciliate, somewhat angulate-toothed or entire; the
terminal one rounded or reniform and petioled, the lateral sessile and
smaller, sometimes minute or wanting; or, in the larger form, the ter-
minal obovate and the lateral somewhat similar to it, little smaller, all
from half to three-quarters of an inch in length. Flowers 3 lines long.
Siligues numerous and much crowded on the short racemes, erect,
strict, rather broadly linear, from an inch to an inch and a half in
length, almost a line wide; the valves separating from a broadly mar-
gined replum. Seeds oval, about 12 in each cell.

T cannot think, with Dr. Hooker, that this plant is a form of the
variable and cosmopolite Cardamine hirsuta. Ours has a perennial
root, and flowers fully as large as those of C. amara; but 1 most rely
for distinguishing it upon the larger and stouter pods; these being
fully twice as broad as those of any form of the European or North
American C. hirsuta. I scarcely doubt, although my specimen of it
is incomplete, that Cardamine cordata of Barneoud, in Gay’s Flora
Chilena, 1, p. 109, is a variety of C. glacialis—The C. pratensis var.,
Fl. Chil. 1, p. 114, is the well-marked C. tenuirostris, Hook. & Arn.

Bot. Voy. Beech. 1, p. 6.
13

50 PHANEROGAMIA.

4, CARDAMINE CHILENSIS, DC.
Cardamine Chilensis, DC. Syst. 2, p. 244; Barneoud, in Gay, Fl. Chil. 1, p. 108.
Has. Between Valparaiso and Santiago, Chili.

The single specimen in the collection is remarkable for its solitary
and axillary flowers; in which, as in other respects, it accords better
with the character given in the Flora Chilena, than with that of De-
Candolle.

5. CARDAMINE LAXA, Benth.

Cardamine laxa, Benth. Pl. Hartw. p. 158 (no. 880).

Var. PUMILA: siliqguis pedicello duplo longioribus stylo brevi apiculatis.
Has. Culnai to Casa Cancha, high Andes of Peru.

Stems barely a span in length, decumbent or ascending. Segments
of the cauline leaves only 3 lines long. Siliques 8 or 9 lines long, on
pedicels only 4 or 5 lines long, nearly erect, pointed with a very short
style, broader for their length than those of C. hirsuta; the flowers
also larger. The replum has the same broad margins as in C. gla-
cialis—I think that our specimens belong to a depauperate form of
Mr. Bentham’s @. lawa.

6. CARDAMINE HIRSUTA, Linn.
Has. Hawaii, Sandwich Islands: in the district of Waimea, &c.
One of the specimens agrees with the ordinary C. hirsuta. The

other has rounder, obtusely lobed, somewhat reniform, and more

petioled segments of the leaves; and resembles the var. sylvatica (C.
sylvatica, Link).

CRUCIFERS. 51

7. CARDAMINE SARMENTOSA, Fort.

C. sarmentosa, Forst. Prodr. Fl. Ins. Austr. no. 529; DC. Syst. 2, p. 265; Hook. &
Arn. Bot. Voy. Beech. p. 59.

Has. Samoan and Society Islands; on the shore, and in waste
grounds. Dr. Pickering also noticed it in the Tonga and Feejee
Islands, around dwellings; apparently introduced. Also at Lima,
Peru !

The divergent pods are terete or nearly so, giving the plant the
appearance of a Sisymbrium; and the valves show some traces of a
midnerve. But the seeds, although turgid, have the cotyledons
accumbent. The pods are from 7 to 9 lines long, tipped with a style
of a line in length. This plant had been found only in the South
Sea Islands; but the specimen gathered at Lima clearly belongs to
this species.

7 KONIGA, &. Br.
1. Konica maritima, 2. Br.

Alyssum maritimum, Lam. Dict. 1, p. 98; DC. Syst. 2, p. 318, & Prodr. 1, p. 164.
' Lobularia maritima, Desv. Jour. Bot. 38, p. 162; Webb & Berthol. Phyt. Canar. 1,
p- 91.

Has. St. Helena. Probably introduced from Europe.

8. DRABA, Linn.
1. DRABA CRYPTANTHA, Hook. /.

D. ineano-tomentosa, depressa; caudice ramoso cceespitoso; foliis imbri-
catis obovatis obtusis concavis; racemo abbreviato paucifloro, florifero

52. PHANEROGAMIA.

inter folia suprema sessili (petalis spathulatis albis unguiculatis),
fructifero nune breviter exserto; siliculis longe pedicellatis ovatis
incanis stylo brevi apiculatis. ;

D. cryptantha, Hook. f. Fl. Antare. p. 234, adnot.

Has. Between Culnai and Casa Cancha, high Andes of Peru; in
the alpine region.

Stems depressed, numerous from a perennial root, ccespitose, the tufts
from half an inch to 2 inches long, crowded, covered with the imbri-
cated leaves. These are canescently tomentose, as is the whole plant
more or less, obovate, obtuse, concave, thickish, 2 or 3 lines long.
Raceme few-flowered, abbreviated, and almost sessile among the leaves
at the summit of the branches when in flower (the flowers small, with
spatulate and unguiculate white petals); in fruit raised, in our speci-
mens, upon a peduncle of from 2 to 4 lines in length, and therefore
more or less exserted beyond the leaves; the pedicels somewhat
umbellate, a line or a line and a half long. WSilicle ovate, canescent,
about 2 lines in length, t2pped with a very short style.

The plant described by Dr. Hooker (from Cerro Pasco, in the same
district) is in flower, and with some barely mature fruit; in which
state it deserves the name of D. cryptantha. Our specimens are not
only in fruit, but far advanced, the valves and seeds having fallen
from the pods; and in this state the peduncles are more or less ex-
serted. The species properly belongs to the second section of Dr.
Hooker’s revision of the South American Drabe; so that his first
section might be suppressed.

2. DraBa MAcLEANI, Hook. f.

D. imcano-tomentosa; caule brevissimo diviso e radice perenni; foliis
confertis obovato-oblongis obtusis; pedunculo elonguto nudo; corymbo
subcapitato; floribus parvis breviter pedicellatis albidis ; siliculis gla-
berrimis ovatis in stylum brevissimum attenuatis, valvis planis.

D. Macleani, Hook. f. Fl. Antare. p. 235, adnot.

CRUCIFERAE. 53

Var. @. STYLOSA: siliculis juxta margines parce hirtis mox glabris stylo
longiusculo superatis ; corymbo laxiflore.

Has. Above Obrajille, Andes of Peru: probably in the alpine region
(var. (.).

The specimen agrees well with that described by Dr. Hooker,
which was gathered by Mr. M’Lean in the same region, except in
the particulars above mentioned. The stems divide from the crown
of the slender perennial root, are an inch or two in length: the leaves
crowded, but spreading, canescently tomentose, somewhat hirsute, 3
lines long,.obovate-oblong, obtuse. Peduncle an inch or so in length,
naked, hirsute, bearing a small umbelliform corymb, which is scarcely
subcapitate, as the flowers are few and lax, on pedicels as long as the
calyx. Sepals a line long, loosely hirsute, as are the pedicels, shorter
than the spatulate and apparently white petals. Stlicles ovate, gla-
brous, except that the margins are beset with scattered hirsute hairs,
which are, however, deciduous, 2 or 3 lines long, tapering into a
distinct style, which is slender, and over half a line long, while in
Dr. Hooker’s plant it is very short: the valves flat.

3. DraBA BRACKENRIDGEI, Sp. Nov.

D. canescenti-pubescens ; caulibus abbreviatis e radice perenni divisis ;
foltis confertis obovato-oblongis obtusis; pedunculis elongatis nudis ;
racemo densifloro; floribus parvulis albis ; siliculis oblongis seu lanceo-
lato-ellipticis pubescentibus stylo brevissimo superatis, valvis subplanas.

Var. 8. siliculis mox glabratis; caulibus nune elongatis.

Var. y. siliculis sepius lineari-oblongis mox subglabratis ; foliis canes-
centi-tomentosis.

Has. Bafios, and above Obrajillo, Andes of Peru.

Plant canescent with a rather hirsute stellate pubescence, or, im var. y.
tomentose-canescent: the short stems branching from a perennial root,

tufted, an inch or two in length. Leuves crowded, spreading, 3 to 6
14

54 PHANEROGAMIA.

lines long, obovate-oblong, obtuse. Peduncles scapiform, leafless, from
one to 3 inches long. Raceme in flower dense and subcorymbose, in
fruit more or less elongated, with the lower pedicels rather distant,
but the upper corymbose-crowded. Flowers nearly 2 lines long:
petals white. Pedicels spreading, a line or two in length. Stdlieles
mostly 3 lines long and a line wide, pubescent with short and close
hairs, sometimes almost as canescent as the leaves, even when
mature; while in the varieties characterized above, this pubescence
is sparser and more or less deciduous: their form varies from ellip-
tical-oblong to oblong-linear; the valves nearly flat. Style very short,
but distinct. Seeds 10 to 15 in each cell.

A rather variable species, allied to D. Macleani, into which some of
the forms may possibly pass, notwithstanding their narrow and pubes-
cent pods.

4, Drapa Pickerineu, Sp. Nov.

D. tomentoso-canescens; caulibus e radice perenni divisis abbreviatis ;
foliis confertis obovatis obtusis; pedunculis elongatis nudis vel oligo-
phyllis; racemo densifloro; floribus majusculis sublonge pedicellatis
albis; pedicellis calyceque villosis; siliculis ovali-ellipticis imcano-
pubescentibus stylo longiusculo abrupte superatis ; valvis planis.

Has. Between Culnai and Obrajillo, high Andes of Peru.

Plant canescent with a close stellate tomentum, and with hirsute or
villous hairs. Stems clustered from a perennial root, an inch long: the
leaves crowded, spreading, obovate, obtuse, 4 to 6 lines long, often one or
two oblong ones on the peduncle, or at the base of the raceme. Peduncle
2 or 3 inches long. Raceme densely-flowered, an inch long; the
pedicels from 3 to 5 lines long, all approximated, or two or three of
the lower ones distant, villous, as also is the calyx. Flowers rather
larger than in the preceding species: petuls white. Silicles oval or
broadly elliptical, very obtuse, about 3 lines long and 2 lines wide,
canescently pubescent, abruptly tipped with a conspicuous style of about
half a line in length; the valves flat, very thin. Seeds 10 to 16 in
each cell.

CRUCIFERS. 55

9 TEESDALIA, R. Br
1. TrespauiA Lepipium, DC.

Has. On the summit of the Pico Ruivo, Madeira.

10. CREMOLOBUS, DC.
1. CREMOLOBUS APHANOPTERUS, Sp. Nov. (Tab. 3.)

C. annuus, parvulus, subglaber ; foliis pinnatipartitis ; racemis elon-
gatis; silicule valvis fere apteris minimis stipitem gracilem haud
excedentibus ; stylo elongato e basi crassa subulato.

Has. Andes of Peru, near Obrajillo.

Root slender, annual. Stem slender, simple or branching above,
about a span high, glabrous, or beset with a few minute simple hairs.
Leaves small (6 or 8 lines long), glabrous or nearly so, mostly sessile,
deeply pinnatifid; the divisions oblong and very obtuse. Racemes
elongated, strict, many-flowered. Pedicels 2 or 3 lines long. Flowers
very small: petals apparently white, obovate, unguiculate. Stamens
6: filaments filiform. Stipe as long as the ovary with its style; the
former consisting of a pair of rounded, scrotiform lobes or cells, sepa-
rated by the very thick axis, or dissepiment, which is continued into
the large and long style, the tapering apex of which is crowned by a
globular, undivided stigma. Silicle very small, minutely puberulent
under a lens; the rounded lobes or valves scarcely if at all above half
a line in diameter, not longer than the stipe, not rugose nor reticulated,
nearly wingless, but surrounded by a slight and wavy or interrupted
margin. Seed nearly orbicular, resupinate-pendulous, so as to make
the ascending radicle internal !

Of the genus Cremolobus, the true character of which was pointed

56 PHANEROGAMIA.

out by Mr. Brown,* five Peruvian species have been figured by Sir
William Hooker; but in none of them is the position of the embryo
shown. I have, therefore, given a figure of the present species, which
differs from Hooker’s C. pinnatifidus nearly as his C. rhomboideus
does from his C. sinuatus. It is moreover remarkable for the minute
size of the silicles, which are even smaller in our specimens than they
are delineated in the figure. |

Puate 3, A.—CREMOLOBUS APHANOPTERUS: a plant of the natural
size. Fig. 1. A flower, enlarged. 2. A petal, more enlarged. 3.
Stamens and pistil, magnified. 4. The pistil, more magnified, on its
long stipe. 5. A ripe silicle, magnified. 6. The same, with one
valve detached, and with the contained seed and embryo transversely
divided; the other valve longitudinally divided in place. 7. A seed,
magnified.

2. CREMOLOBUS PINNATIFIDUS, Hook.

C. annuus, pusillus, subglaber ; foliis pinnatifidis vel incisis spathulato-
oblongis; racemis laxis nunc elongatis; silicule valvis orbiculatis ala
lata sinuata sew integriuscula cinctis stipite gracili longioribus; stylo
e basi crassa subulato.

Cremolobus pinnatifidus, Hook. Ic. Pl. t. 100.

Var. @. INTEGRIFOLIUS: foliis indivisis aut obsolete dentatis aut integer-
rimis; racemo brevt.

Has. Andes of Peru, near Obrajillo and Baiios. §. At Baiios.

A small, slender annual, a span or less in height, nearly glabrous,
but showing some hairiness on a close inspection, as in the last species.
Leaves incisely pinnatifid, as represented in Hooker’s figure, or some-
times almost pinnately parted as in C. aphanopterus; or, in var. @.
entire or nearly entire, and oblong or spatulate. The latter is perhaps
a depauperate state, from one to 3 inches high; but its flowers are as

* Observations on the Structure and Affinities of the more remarkable Plants col-
lected by Oudney, Denham, & Clapperton, &c., p. 7.

CRUCIFERE. 57

large as in the pinnatifid forms, namely, about 2 lines long, which is
larger than those of the foregoing species. The racemes are either
rather short and few-flowered, or elongated in fruit. The mature
silicles are from 2 to 3 lines broad ; the valves considerably longer than
the stipe, surrounded with a pretty wide wing, which is either repand
or almost entire, or sometimes with the margin sinuate-toothed ; the
disk smooth or slightly rugose. Style nearly as in the foregoing, or
perhaps not quite so long, subulate or tapering upwards from a very
thick base.

The specimens of Matthews, figured by Hooker, do not exhibit any
mature fruit. I add a ripe silicle, therefore, to Plate 3. The wing
varies considerably in different states or stages of the plant; but I
think it unlikely to pass into the foregoing species. Forms of the
species gathered by Professor Jameson and Mr. McLean are also pre-
served in the Hookerian herbarium.

PuatE 3, A.—Fig. 8. CREMOLOBUS PINNATIFIDUS: a ripe silicle, of the
natural size. 9. The same, enlarged.

ll. BRAYA, Sternd. & Hopp.
1. BrayA PUSILLA.

B. multiceps; foliis fere omnibus radicalibus rosulatis lineari-spathulatis
integerrimis carnosulis preesertim ad margines hispidis ; scapo glabro
5—6-jloro; siliquis breviter tinearibus seu lineari-oblongis stylo brevis-
sumo apiculatis.

Erysimum ? pusillum, Gillies, in Hook. Bot. Misc. 3, p. 140.
Has. Andes of Chili, above Santiago; on the snow line.

A dwarf, ceespitose, and multicipital plant, with the aspect of an
alpine Draba, except as to the pods. Leaves all radical and rosulate-
tufted, or sometimes with a small one on the scape, rather thick and

fleshy, spatulate-linear, 3 to 5 lines long, sparsely hispid all over, or
15

58 PHANEROGAMIA.

when old glabrous, except the hispid-ciliate, entire margins. Scapes
in our (fruiting) specimen only half an inch long (the plant of Gillies
is larger), 5-6-flowered. Siliques corymbose, linear or oblong-linear,
nearly terete, 3 or 4 lines long, often a little curved, tipped with a
very short but distinct style; the valves strongly convex, one-nerved.
Seeds about 8 in each cell, as much biserial as in Braya alpina, as
figured by Hooker (Exot. Fl. t. 121), Cotyledons incumbent.

Only a single fruiting specimen exists in the collection, which, on
comparison, I find to be the same as the plant of Gillies characterized
by Hooker and Arnott as a doubtful Erysimum: but I believe it has
white or purplish flowers, and that its characters more nearly accord
with Braya, of which it has much the habit. To this genus, now for
the first time identified in the southern hemisphere, I should also
refer the Draba imbricatifolia of Barneoud, in Gay’s Flora Chilena,
which indeed is hardly to be distinguished from B. pusilla, except by
its slender style.*

12, SISYMBRIUM, Linn.

1. Sisymprium LEPTocARPUM, Hook. & Arn.
S. leptocarpum, Hook. & Arn. Bot. Mise. 3, p. 189; Gay, Fl. Chil. 1, p. 124.

Has. Obrajillo, Andes of Peru.

Our specimen, which is in a mature fruiting state, accords pretty
well with the S. leptocarpum of Chili. The whole plant is smooth ;
the leaves not unlike those of S. officinale: but the slender, slightly

* BRAYA IMBRICATIFOLIA : multiceps e caudice lignescente, ccespitosa ; foliis brevissimis
rosulatis imbricatis spathulatis integerrimis hispidissimis; scapo glabro nudo 4-8-
foro; siliquis linearibus stylo gracili coronatis.

Draba imbricatifolia, Barn. in Gay, Fl. Chil. 1, p. 158.
Pods narrowly linear, 4 to 6 lines long, terete, acute, pointed with a style of a line in

length: the valves obscurely one-neryed. The dissepiment is split through the middle,
as in B. alpina. Seeds about 10 in each cell, nearly uniserial. Cotyledons incumbent.

CRUCIFERE 59

curved, spreading pods are an.inch and a half or more in length, on
pedicels over half an inch long.

2. SISYMBRIUM MYRIOPHYLLUM, Humb. & Bonpl.

S. myriophyllum, Humb. & Bonpl. in DO. Syst. 2, p. 477,.& Prodr. 1, p. 194;
Benth. Pl. Hartw. p. 159; Walp. in Rel. Meyen. p. 249.
S. pimpinellifolium, Barn. in Gay, Fl. Chil. 1, p. 129.

Has. Baiios and Obrajillo, Andes of Peru.

Ours accord with Jameson’s and Hartweg’s specimens from Anti-
sana: and S. pimpinellifolium, from the Cordilleras of the northern
part of Chili is the same species.* The seeds occupy two series in
each cell as distinctly as in S. canescens (vide Gray, Gen. Am. Bor.
Or. Ill. 1, t. 64), and other species of the section Descurea (Descu-
rainia, Webb), and one specimen has the dissepiment fenestrate or
abortive, as in the following species.

3. SISYMBRIUM ATHROCARPUM, Sp. Nov. (Tab. 3).

S. cano-tomentosum; folis pinnatisectis, segmentis oblongis pinnatipar-
titis, lobis ovalibus confertis; floribus fructibusque capitato-corymbosis
arcte congestis; siliquis brevissimis oblongis tetragonis florem marces-
centem bis terve tantum superantibus ; septo late fenestrato vel evanido ;
seminibus biseriatis.

Has. Obrajillo, and from Culnai to above Casa Cancha, in the high
Andes of Peru. (Andes of Peru, Mr. McLean, in herb. Hook.)

Root apparently annual. Stems from 2 inches to a foot high,
clothed, as is the foliage, &c., with a soft hoary wool. Leaves an inch
or two in length, narrowly oblong in circumscription, pinnately dis-
sected into from 13 to 19 subsessile segments or leaflets: these are
oblong, and pinnately parted or cleft into from 9 to 15 oval, entire,

* Sisymbrium Titicacense, Walp. Rel. Meyen. p. 249, or, at least, a plant in the
Hookerian herbarium, from Titicaca, which accords with the character, is apparently
only the short-fruited form of Nuttall’s S. canescens.

60 PHANEROGAMIA.

crowded lobes. The stem branches at the summit, and bears a num-
ber of crowded corymbs, each with a great number of flowers, on pedi-
cels as long as the calyx, but so aggregated as to appear capitate, as
well in fruit as in flower; for the axis does not at all elongate with
age. Petals pale yellow, spatulate, unguiculate. These, as well as
the rest of the floral organs, are persistent, more or less withering,
around the base of the pod. Stliques very short, only 2 lines long,
oblong, acutely four-angled, only twice or thrice the length of the persistent
petals; the valves strongly carinate and one-nerved: the partition
obsolete, except at the summit, or reduced to a narrow margin.
Stigma depressed-capitate, subsessile. Seeds about 20 in each cell,
occupying two distinct series. Cotyledons incumbent.

This is a well-marked new species of the group to which the pre-
ceding and S. canescens, Nutt. belong, and which would appear to be
fully as distinct from the true Sisymbria as is the genus Braya. If
S. Sophia, S. millefolium, and S. tanacetifolium, had the seeds likewise
biserial, there would be no doubt of the propriety of adopting the
genus Descurea, Guettard, or Descwrainia, Webb & Berth. (Hugue-
ninia, Reichenb.)

Piate 3, B.—SISYMBRIUM ATHROCARPUM. Fig. 1. Upper part of a
plant. 2. The whole of a dwarf specimen: both of the natural size.
3, 4. Branched hairs of which the pubescence consists, magnified.
5. A flower, magnified. 6. A sepal. 7. A petal. 8. A flower from
which the sepals and petals have been removed. 9. Silique, with
the withering-persistent remains of the flower. 10. The replum
(placentae), with the vestiges of the almost obsolete partition. 11. A
pod transversely divided. 12. The replum, with the seeds attached.
13. A seed transversely divided. 14. Embryo detached entire —AIL
the analyses more or less magnified.

4. SISYMBRIUM SPATHULAFOLIUM, Sp. Nov.

S. humile; caulibus e radice perenni crassa pluribus adscendentibus
pilis bi-trifurcatis pubescentibus ; foliis omnibus obovato-spathulatis
carnosulis integerrimis nunc parce subdentatis glabriusculis basi
attenuatis subpetiolatis, summis flores infimos bracteantibus ; racemo

CRUCIFERAE. 61

laxifloro; siliquis glabris subulato-linearibus pedicello erectiusculo:
. longioribus stylo brevi apiculatis, valvis subcarinato-uninerviis.

Has. Obrajillo or Bafios, Andes of Peru.

Root perennial, thickish, somewhat lignescent? Stems several, 3 to
6 inches long, ascending, rather leafy, more or less hairy, with a 2-3-
Jorked pubescence. Leaves sparsely beset with similar hairs, or at length
almost glabrous, all obovate or spatulate, rather fleshy, entire, or some
of the upper ones obscurely 2—3-toothed, the radical ones 6 to 8 lines
long, and narrowed into distinct petioles of equal length; the cauline
3 or 4 lines long, narrowed at the base into a short petiole, or sub-
sessile. Raceme short and corymbose in flower, more or less elongated
and loose in fruit; some of the lower flowers usually bracteate; the
bracts, like the upper cauline leaves, rather shorter than the fructi-
ferous pedicels. Flowers a line and a half long: sepals greenish: the
petals apparently white or purplish. Stliques glabrous, 6 or 8 lines
long, ascending or slightly spreading (the pedicel mostly erect, 3 or 4
lines long), linear-subulate, nearly terete, pointed with a short style;
the valves one-nerved, scarcely connate. Dissepiment one-nerved. Seeds
8 or 10 in each cell, in a single series. Cotyledons incumbent.

In habit and foliage this plant has much the aspect of a T'hlaspr.
Its pods are those of a Sisymbriwm.

5, SISYMBRIUM? AMPLEXICAULE, Sp. Nov.

S.? glabellum; caule e basi suffruticosa ramoso ; ramis divergentibus ;
foliis ovato-oblongis sinu profunde sagittato-cordato amplexicaulibus
integerrimis ; floribus laxe racemosis ebracteatis ; petalis unguiculatis ;
siliquis immaturis patentissimis filiformibus elongatis stigmate sessilr

- superatis.

Has. Between Caball and Obrajillo, Andes of Peru.

Stems from a span to a foot high, the base apparently suffruticose,
loosely and divergently branched ; the branches slender, nearly glabrous.

Leaves glabrous or minutely pruinose, about half an inch long, ovate-
16

62 PHANEROGAMIA.

oblong or ovate-lanceolate, and clasping by a very deep and narrow
cordate sinus, entire, veinless, probably of a fleshy texture when living.
Racemes loose; the pedicels 2 or 3 lines long, not bracteate. Sepals
smooth, membranaceous, greenish, oval-oblong, rather lax. Petals
obovate-spatulate, conspicuously unguiculate (the claws about the
length of the calyx), apparently white or rose-colour, 3 lines long.
Anthers linear, sagittate. Siliques very slender, filiform, an inch or
more in length, terete, glabrous, not stipitate, widely spreading, tipped
with a depressed-capitate, nearly sessile stigma; the valves nerveless
or nearly so. Seeds numerous in a single series in each cell, oblong,
surrounded by a hyaline pellicle. None of them are sufficiently
matured to show the nature of the embryo; but, from the narrow
shape of the seed, the cotyledons are doubtless incumbent. (Stsym-
brium amplexicaule, Desf. = Sinapis amplexicaulis, DC.)

13. LEPIDIUM, Linn. R. Br.
1. Lepiprum BonAriEnse, Linn.

L. Bonariense, Linn.; DC. Syst. 2, p. 543; Gay, FI. Chil. 1, p. 164.
Thiaspi Bonariense, etc. Dill. Hort. Hlth. p. 281, t. 286, f. 370.
T. Bonariense & T. multifidum, Poir. Dict. 7, p. 548, 540.

Has. Rio Negro, North Patagonia; on sand-hills.

2. LEPIDIUM BIPINNATIFIDUM, Desv.

Lepidium bipinnatifidum, Desy. Jour. Bot. 3, p. 177; DC. Syst. 2, p. 544; Gay, FI.
Chil. 1, p. 165.

Has. Obrajillo, Peru. (Also, Quito, Antisana, &c., Jameson.)

3. LEPIDIUM RUDERALE, Linn.

Has. Puen Buen, New South Wales. Introduced from Europe.
But it was long ago gathered in New Holland by Labillardiere.

CRUCIFERA, 63

4, LEPIDIUM GRAMINIFOLIUM, Linn.

Has. Cape of Good Hope, in the vicinity of Cape Town.

5, Leprprum piscrprum, Forst.

Lepidium piscidium, Forst. Prodr. p. 46, & MS. descr. in Guill. Zeph. Tait. p. 75;
DC. Syst. 2, p. 546; Hook. & Arn. Bot. Beech. Voy. p. 59.

Has. Coral Islands of the Pacific; widely diffused. It is also
mentioned as having been gathered at the Sandwich Islands, by
Nelson, Meyen, &c.; but from thence we have only the nearly allied

6. Lepmpium OWAHIENSE, Cham. & Schlecht.

Lepidium Owahiense, Cham. & Schlecht. in Linnea, 1, p. 32; Hook. & Arn. Bot.
Beech. Voy. p. 78; Walp. Rel. Meyen. p. 250.

Has. Coast of Oahu and Hawaii, Sandwich Islands.

This is pretty well distinguished from the preceding species by its
rounded silicles (orbicular or rounded-obovate, instead of oblong-
obovate), with the stigma included in the notch. The leaves, also,
are usually of a firmer texture, the cauline ones sharply or incisely-
toothed; the base of the stem is lignescent; and the root appears to
be perennial. A specimen from the district of Waimea, Hawaii, with
more lignescent stems, has only linear or lanceolate leaves; the lower
ones, however, are wanting.

14. SENEBIERA, Por.
1. SENEBIERA DIDYMA, Pers.
Senebiera didyma, Pers. Ench. 2, p. 185; Desv. Jour. Bot. 3, p. 164.

S. pinnatifida & S. pectinata, DC. Syst. 2, p. 523, & Prodr. 1, p. 202.
Lepidium didymum, Linn. Mant. p. 92.

64 PHANEROGAMIA.

Has. Table land east of Funchal, Madeira. Waste grounds about
Honolulu, Oahu; and island of Kauai, Sandwich Islands.—For a
view of the present geographical diffusion of this species, see Dr.
Hooker's Flora Antarctica, p. 241.

15. SINAPIDENDRON, Lowe.

SINAPIDENDRON FRUTESCENS, Lowe.

Sinapidendron frutescens, Lowe, in Camb. Phil. Trans. 4, p. 36.
Sinapis frutescens, Ait. Hort. Kew. ed. 1, 2, p. 404; DC. Syst.-2, p. 623; Hook.
Bot. Misc. 1, p. 119, t. 28.

Has. On dry rocks, Corral, Madeira.

16. RAPHANUS, Linn.
1. RapHanus RApHAnistrum, Linn.
Has. Rio Negro, North Patagonia; on the coast. Bay of Islands,
New Zealand. Auckland Islands—This Wild Radish or Jointed
Charlock was doubtless carried from Europe, and probably is scarcely

naturalized at any of these places, except around the dwellings of the
colonists.

17, ISATIS, Linn.
1. Isatis TrnctoriA, Linn.

Has. Corral, Madeira; where the Dyer’s Woad is probably not
truly indigenous. |

Orv. CAPPARIDACE A

1 GYNANDROPSIS, DG.

1. GYNANDROPSIS PENTAPHYLLA, DC.

Gynandropsis pentaphylla, DC. Prodr. 1, p. 238; Guill. Zeph. Tait. p. 74; Gray,
Gen. Am. Bor. Or. 1, t. 78.

G. afinis, Blume, Bijdr. p. 51; Decaisne, Herb. Timor. p. 97.

Cleome pentaphylla, Linn.; Sims, Bot. Mag. t. 1681.

Has. Tahiti; in waste places. Now widely distributed over the
warmer parts of the world; but doubtless derived from Africa, as
indicated by Mr. Brown, in his Appendix to Denham & Clapperton’s
Narrative.

2, CLEOME, Linn. DC.
1. CLromME SANDWICENSIS, Sp. Nov.

C. herbacea, aculeolata, undique viscoso-pubescens ; foluis quinquefolio-
latis, floralibus ovatis subcordatis; pedicellis cum calyce harsuto-
viscosissimis ; siliqua glabra thecaphoro vix aut ne vix longiore.

Cleome spinosa, Hook. & Arn. Bot. Beech. Voy. p. 78; Walp. Rel. Meyen. p. 201,

non Linn.

* Tovaria pendula, Ruiz & Pav., has the ovary completely 6-7-celled, as characterized
by Hooker (Icones Plantarum, t. 664); but the ovules instead of being inserted over
the whole face of the dissepiments, as he figures and describes them, are rather borne on
thick, spongy placentz, which project from the axis into the cells.

17

66 PHANEROGAMIA.

Has. Oahu, Sandwich Islands; on the coast near Honolulu. Also
gathered by Lay and Collie, Barclay, and Meyen.

Plant herbaceous, clothed with a viscous pubescence ; the stem, petioles,
and midrib of the leaflets sparsely aculeate with small and weak
prickles; the stipular ones stronger. Cauline leaves of 5 leaflets,
which are oblong-lanceolate, one to 22 inches long, and cinereous on
both surfaces with a short and close viscid pubescence. Floral leaves
ovate, slightly cordate, small, on short petioles, shorter than the very
pubescent and somewhat setose-hispid viscous pedicels. Calyx hairy
and viscous like the pedicels. Petals apparently white, 3 lines long,
obovate-oblong, on short claws. Pod 14 to 2 inches long, about 2
lines in diameter, smooth and glabrous, terete, crowned with a thick
subsessile stigma, raised on a filiform glabrous stipe (the carpophore) of
its own length, or very little shorter.

This species is certainly to be distinguished from Cleome spinosa
(with which it has been confounded), by its pubescent leaves, and the
proportionally longer stipe; as long as the pod itself in most specimens
that I have seen, or a little shorter in the single specimen in the col-
lection of the Exploring Expedition. The pod is only half the length
of that of C. pungens.

2. CLEOME CHILENSIS, DC.

Cleome Chilensis, DC. Prodr. 1, p. 194; Gay, Fl. Chil. 1, p. 187; Deless. Ic. Sel.
3, t. 2.

Has. Above Obrajillo, Andes of Peru: common.

Accords well with the figure in Delessert’s Icones, or perhaps with
the pruinose-glandular pods a little shorter. The calyx is persistent
in one of the specimens, but deciduous in the other.

3. CLEOME DIFFUSA, Banks, in DC. 1. e.

HAs. Rio Janeiro, Brazil: common in open places.

CAPPARIDACEA. 67

4, CLEOME AFFINIS, DC. 1. c.

Has. Rio Janeiro, Brazil: collected with the preceding species.

5. CLeomE Ricuu, Sp. Nov.

C. herbacea, Jere glabra, aculeis stipularibus et petiolaribus validis com-
planatis armata; foliis 5—T-foliolatis ; foliolis lanceolatis, costa scepius
aculeata ; petalis (albis vel albidis majusculis) anguste oblongis longe
unguiculatis; ovario cum thecaphoro eodem subbreviare staminibus
multoties brevioribus.

Has. Rio Janeiro, Brazil: in open places.

Plant nearly glabrous, or the branches, pedicels, &c., minutely gra-
nulose or glandular-viscous: the stem /Aerbaceous, apparently 2 or 3
feet high, rather stout, armed with short and flattened stout stipular
prickles. Petioles elongated (5 or 6 inches long), grooved above,
armed with numerous stout recurved prickles, resembling the stipular
ones. Leaflets 5 to 7, broadly lanceolate, scareely petiolulate, about 2
inches long, usually prickly along the midrib underneath. Floral
leaves small and simple, ovate, nearly sessile. Pedicels filiform, 2
inches long. Sepals oblong-lanceolate, 3 lines long, minutely glan-
dular, like the pedicels. Petals apparently white, an inch and a
quarter in length, including the claw, which is nearly as long as the
narrowly oblong lamina. Filaments purplish, 2 inches long. Ovary
less than 2 lines long, on a stipe of only a line in length. Fruit not
seen.

This may prove to be some ill-described species; but I cannot iden-
tify it.
6. CLEOME DENDROIDES, Reem. & Schult.

Cleome dendroides, Rem. & Schult. Syst. Veg. 7, p. 28; Hook. Bot. Mag. t. 3296.
C. atropurpurea, Schott, in Schreib. Nachtr. p. 129, ex. Rem. & Schult.

68 PHANEROGAMIA.

Has. Brazil, in the vicinity of Rio Janeiro. (In flower only.)

3. POLANISIA, Raf.

1. PoLANISIA ICOSANDRA, Wight & Arn.

Polanisia icosandra, Wight & Arn. Prodr. Fl. Ind. Or. 1, p. 22.
P. viscosa, DC. Prodr. 1, p. 242, var. B. fide Wight & Arn.
Cleome icosandra & C. viscosa (excl. syn.) Linn. Spec. p. 672.

Has. Luzon, near Manilla. St. Jago, Cape de Verde Islands.
(Not enumerated by Mr. Webb in his Spicilegia Gorgonea.)

4, CRAT AVA, Linn.

CRATAVA TAPIOIDES, DC.

Has. Rio Janeiro, Brazil, near the coast.

(A specimen with
mature fruit.)

5. CAPPARIS, Linn., Endl.
1. CAPPARIS CYNOPHALLOPHORA, Linn.

Capparis cynophallophora & C. flexuosa, Linn. Spee. ed. 2, p. 721, ex. DC. Prodr.
1, p. 249.

C. cynallophora, Jacq. Stirp. Amer. p. 158, t. 98.

Has. Rio Janeiro, Brazil; along the shores.

2. CAppARIS BrasitiAna, DC. 1. ¢.

Var. 8. LONGIPES: thecaphoro (pollicari) pedicello paullo longiore.

CAPPARIDACES, 69

Has. Rio Janeiro, Brazil. (Also gathered by Gardner, No. 5355,
and by Sellow.) Distinguished from the ordinary form only by its
more elongated thecaphore.

3. CAPPARIS SANDWICHIANA, DC.

Capparis Sandwichiana, DC. Prodr. 1, p. 245; Hook. & Arn. Bot. Beech. Voy. p.
59; Gaudich. Bot. Voy. Bonite, t. 55.

Has. Sandwich Islands: on Diamond Hill, Oahu; and on the
coast of Kauai and Hawaii. (First collected by Menzies in Van-
couver’s Voyage.) Matia, Society Islands; where it was gathered in
Beechey’s Voyage.

This is distinguished from C. spinosa (which is sometimes unarmed),
by its longer petioles (only one-third shorter than the blade) and its
elongated, club-shaped fruit, 24 inches long and half an inch in thick-
ness, raised on a stipe of 3 inches in length.—The letter-press answer-
ing to the plate above-cited, is not yet published.

4, Capparis Ricu, Sp. Nov.

C.. scandens ; stipulis spinosis brevibus uncinatis interdum nullis ; foltis
ovato-oblongis nune lanceolatis seu lineart-lanceolatis subacuminatis
basi obtusis subcoriaceis, novellis cum ramulis calycibusque ferrugineo-
puberulis, adultis glabris subcoriaceis; pedicellis 5 seriatim supra-
axillaribus petiolum subequantibus; floribus parvis; staminibus cir-
citer 12.

Has. Feejee Islands; the narrow-leaved forms from the Muthuata
Mountains, and Direction Island.

Stem woody, climbing, smooth, the branches slender, minutely fer-
rugineous-pubescent when young, armed with very short and stout
hooked prickles in the place of stipules, which, however, are nearly
wanting in some of the broader-leaved specimens. Leaves very

variable, or diverse even upon different branches of the same stem,
7 18

70 PHANEROGAMIA.

24 to 4 inches long, clothed with a ferrugineous pubescence when
very young, soon glabrous, rather coriaceous, reticulated, either ovate
or oblong-ovate and somewhat acuminate, or oblong-lanceolate, varying in
some specimens into lanceolate or even linear, the base rather narrowed
but obtuse, not in the least cordate: petioles 4 to 6 lines long, of the
narrower leaves only 2 or 3 lines long, ferrugineous-pubescent. Pedi-
cels about. 5, inserted one above the other in the axil of the leaves
(supra-axillary), nearly the length of the petiole, and with the flower-
bud similarly ferrugineous-pubescent. Flower-buds not more than a
line in length, gibbous. Stamens apparently 12. Ovary on a long
stipe: placentee 4. Fruit unknown.

A species of the group Eucapparis Seriales, which should be collated
with Capparis quiniflora, DC., from the northern coast of New Hol-
land, known only by a very brief character. It cannot be C. flexuosa,
Blume, since this, according to Hasskarl (Plante Jav. Rar. p. 178)
has 38 stamens, while ours has no more than 12. For the details of
the flower I am obliged mainly to depend upon a drawing, executed
under the superintendence of Mr. Rich. Greatly as the leaves vary
in shape in the different specimens, I suppose that they all belong to
one species. At least, as they are not in flower, there are no means
of characterizing the narrow-leaved forms as a distinct species.

5. CAPPARIS AURANTIOIDES, Presi.

Capparis aurantioides, Presl. Rel. Hank. 2, p. 86; Walp. Repert. 1, p. 198.
C. nemorosa, Blanco, FI. Filip. p. 488?

Has. Philippine Islands,.at Caldera, Mindanao; and Sooloo Islands.

This also belongs to the division Seriales ; the pedicels, sometimes
five in number, being superposed above the axil.
6. CAPPARIS SEPIARIA, Linn.

Has. Small island of the Sooloo Sea. A widely diffused, Hast.
Indian species.

CAPPARIDACES. 71

7. CAPPARIS CERASIFOLIA, Sp. Nov.

C. inermis, aut stipulis spinosis fere obsoletis; foliis oblongo-lanceolatis
acumimatis basi obtusis membranaceis laxe venosis, novellis ramulisque
sericeis, adultis glabratis ; pedicellis axillaribus petiolo sublongioribus
solitariis vel scepius 2-3 e gemma subulato-perulata cum calyce the-
caphoro et ovario tomentulosis ; staminibus circiter 12.

Has. Small island of the Sooloo Sea.

Stem shrubby, glabrous, unarmed or nearly so; the minute spines-
cent stipules mostly deciduous. Leaves oblong-lanceolate, much acumi-
nate, membranaceous, obtuse at the base, silky-pubescent when young, as
well as the branchlets, at length glabrous and rather shining above
and nearly glabrous beneath; the lower surface with 5 or 6 pairs of
conspicuous primary veins: the minute veinlets reticulated. Petioles
3 or 4 lines long. Pedicels axillary, 3 to 5 lines long, one-flowered,
tomentose-pubescent, as is the calyx, stipe, and ovary, sometimes soli-
tary, more commonly 2 or 38 together from a small axillary bud
or cluster of minute scales and spinescent bracts like the stipules.
Flower-bud scarcely 2 lines long. Calyx and petals nearly as in C.
pubijlora, DC., as figured by Delessert; the petals somewhat pubes-
cent externally at the apex. Stamens not more than 12. Stipe of
the ovary as long as the pedicel. Placentz only 2. Fruit not seen.

This species is evidently related to Capparis pubiflora, DC.; from
which its unarmed stems, longer and narrower leaves, smaller flowers,
and few stamens readily distinguish it. It is the same as Cuming’s
No. 1068, from the Philippine Islands; and is allied to his No. 955,
but has longer pedicels and thinner leaves, which are much less silky-
tomentose when young.

8. CAPPARIS ODORATA, Blanco?

Capparis odorata, Blanco, FI. Filip. p. 439 ?

Has, Vicinity of Manilla, Luzon. (Without flowers or fruit.)

Orv PANGIACE &.

1. CARPOTROCHE, An.

CARPOTROCHE, Endl. Gen. no. 5066 (non Walp. Repert. 1, p. 208).
Mayna, Raddi, Pl. Nov. Bras. p. 23, f.1; Zuce. Pl. Nov. Hort. & Herb. Monac.
fasc. 2, p. 55, t. 5, 6.

1. CARPOTROCHE BRASILIENSIS.
Mayna Brasiliensis, Raddi, 1. c.; DC. Prodr. 1, p. 79; Zuce. 1. ¢.
Has. Brazil, near Rio Janeiro.

The specimen has only male flowers, and affords nothing to add to
Zuccarini’s full description and figures. Walpers has without any
reason referred to Carpotroche three species of Mayna described by
Bentham, which manifestly belong to Presl’s Lindackeria; a genus
which Mr. Bentham has shown to be most probably identical with
Mayna of Aublet.* It was Mr. Brown} who referred the present
genus to the family which Blume afterwards published under the
name of Pungiacee : but Mr. Bennett’s character of the family appears
to be framed so as to apply only to Pungiwm, Gynocardia, and Hydno-
carpus. In several particulars, none of them of any high importance
indeed, it does not accord with the character of Carpotroche.

BIXINEA.— Bixa Orellana, Linn., the tree which yields the well-
known colouring matter named Arnotto by the English, and Rocou by

* Hook. Kew Bot. Misc. and Jour. Bot. 3, p. 117.
T Zucearini, Pl. Nov. supra cit. p. 60; Bennett, Plante Javanicw Rariores, p. 190.

FLACOURTIACES. 73

the French, occurs in the collection from the Sandwich Islands: but
it is stated on the ticket to be only a cultivated tree, in gardens, in
the vicinity of Honolulu. A native of tropical America, it is now
widely distributed through the warmer parts of the world.

Orv. FLACOURTIACE&.

1. PROCKIA, P. Browne.
Procx1a, P. Browne, in Linn. Gen. no. 647; Bennett, Pl. Jav. Rar. p. 191.
Prock1A, § Procxiaria, DC. Prodr. 1, p. 260; Endl. Gen. no. 5072.
1. Prockxta Crucis, Linn.
Prockia completa, Hook. Ic. Pl. 1, t. 94; Walp. Repert. 1, p. 204.

Has. Brazil, in the Organ Mountains, near Rio Janeiro.

2, KUBLIA,. 7B.
1. Kuauia PARVIFLORA, Sp. Nov.
K. foliis oblongis acuminatis remote glanduloso-serratis bast subincequa-
libus tripli—quintuplinerviis utrinque viridibus ; panicula composita ;
floribus minimis.

Has. Brazil, in the Organ Mountains, near Rio Janeiro.

A shrub or small tree, apparently : the branchlets slender, minutely

verrucose, glabrous. Leaves oblong, or oblong-lanceolate, more or less
19

74 PHANEROGAMIA.

acuminate, remotely serrate with glandular-tipped or glandular-truncate
teeth, more or less obtuse and rather unequal at the base, which is
triplinerved or quintuplinerved (the lateral nerves fainter or marginal),
elabrous, of the same green hue both sides, thin and chartaceous in
texture, reticulate-veined, alternate, 2 or 3 inches long. Petioles 3
lines long. Stipules minute, subulate, caducous. Panicle compound,
terminal, many-flowered; its branches, with the subulate bracts, the
short pedicels, &c., minutely and softly pubescent. lowers very small ;
the flower-buds less than half a line in diameter, globular, glabrous
or nearly-so. Perianth in two distinct series, each three-parted ;
the three exterior segments (sepals) broadly ovate, obtuse, nearly
if not strictly valvate in early estivation, but their margins more or
less separated before the bud has attained its full growth, their edges
minutely ciliate; the three inner (equivalent to the petals of Banara
and Prockia) united with the exterior at the base, rather larger than
they and more ciliate, but nearly similar in texture, obovate, their
overlapping summits strongly imbricated in sestivation, yellowish
inside, spreading in anthesis. Disk none. Stamens from 15 to 20,
inserted into the base of the perianth: filaments filiform, yellowish,
twice the length of the perianth: anthers didymous. Ovary ovoid-
conical, one-celled, narrowed into a short and thickish style, which is
tipped with an obtuse and obscurely three-lobed stigma. Placentz 3,
parietal, occupying the whole length of the cell, into which they
slightly project. Ovules indefinitely numerous, in several series, ana-
tropous. Fruit unknown.

In habit as well as in floral structure, this plant is manifestly a con-
gener of Kuhlia glauca of Kunth. I ought perhaps to follow the high
authority of Mr. Bennett,* who pronounces Auhlia not to be generi-
eally distinct from <Azara. But it appears to me, that the biserial
perianth (as truly double as that of Banara), the exterior subvalvate,
the inner strongly imbricated, along with the difference in habit, sti-
pulation, and inflorescence, should serve to distinguish them.

To the genus, if preserved, belongs another Brazilian plant, namely
Ascra Brasiliensis, Schott, in Spreng. Syst. Cur. Post., p. 407, of which
I have received a specimen from Professor Von Martius. The same
species is rudely figured in the Flora Fluminensis, 5, t. 115, under
the name of Boca serrata. Kuhlia Brasiliensis, as it must be called, is

* Plants Javanice Rariores, p. 190.

FLACOURTIACE SR, 75

distinguished by its thick and coriaceous leaves and large flowers ;
the inner divisions of the perianth two lines or more in length, and
nearly equalling the stamens. The fructified ovary shows no ten-

dency to become spuriously several-celled. The Linnzan genus Trilix
is probably the same as Banara.*

3. AZARA, Ruiz & Pav.
-l. AZARA CELASTRINEA, Don.

Azara? celastrinea, Don, in Edinb. Phil. Jour. 11, p. 119; Gay, Fl. Chil. 1, p. 195.
A. Lilen & Lilenia dentata, Bertero, herb. 3

Has. Near Valparaiso, Chili: common.

2. AZARA INTEGRIFOLIA, Ruiz & Pav.

Azara integrifolia, Ruiz & Pay. Syst. p. 138, & Fl. Per. & Chil. 5, (ined.) t. 466 ;
DC. Prodr. 1, p. 262; Gay, Fl. Chil. 1, p. 199.

Has. Chili, with the preceding: also at Santiago.

Another species, probably A. serrata, Ruiz & Pav., is mentioned in
Dr. Pickering’s notes; but it is not found in the collection.

4. FLACOURTIA, LHer.

1. FLAcourTIA INERMIS, Roxb., DC.
Has. Singapore.

The specimen, in flower only, agrees perfectly with those from the
Peninsula of India, distributed by Dr. Wight.

* From a collection made by M. Claude Gay, in the province of Cusco, Peru, I have
received, under the name of Blackwellia, a male specimen of what appears to be a true
congener of Banara mollis and B. Ibaguensis, Tulasne, in Ann. Sci. Nat. 3d ser. 7, p-
288, except that it is polygamo-diccious. The abortive ovary shows five placental lines,
which, however, are not ovuliferous.

76 PHANEROGAMIA.

5. XYLOSMA, Forst.

Xytosma, G. Forst. Prodr. Fl. Ins. Austr. p. 72; Bennett, Pl. Jay. Rar. p. 191.
Mrroxyton, J. R. Forst. Char. Gen. p. 125, t. 63, non Mutis.

1. XYLOSMA SUAVEOLENS, Forst. (Tab. 4).

X. foliis ovatis crenato-subserratis chartaceis; calyce 4—5-partito, lobis
acutis.

Xylosma suaveolens, G. Forst. Prodr. no. 380; Guill. Zeph. Tait. p. 76.
Myroxylon suaveolens, J. R. Forst. 1. c. & MS. descr. in Zeph. Tait. 1. c.

Has. Tahiti, Society Islands, in forests on the mountains: also, on
Eimeo.

A small tree, or a shrub, with a reddish and sweet-scented, hard
wood, according to Forster. Branches, leaves, &., glabrous. Leaves
alternate, thin and chartaceous in texture, 2 or 3 inches long, ovate,
sometimes ovate-oblong, more or less acute or acuminate, obtuse or
acutish at the base, rather obscurely crenate-serrate, the teeth tipped
with a small callous gland, loosely reticulate-veined, the narrow mid-
rib prominent underneath, the upper surface somewhat shining:
petiole half an inch long. Flowers dioecious, small, greenish or
yellowish, in loose axillary and terminal corymbulose clusters. Pe-
duncle very short. Pedicels about 3 lines long, articulated near the
base, subtended by small ovate and scale-like bracts. Calyx deeply
4_5-cleft, or in the female flowers 4—5-parted, apparently yellowish-
green, nearly glabrous externally, silky-pubescent internally; the
lobes ovate, acute, ciliate on the margin (imbricated in estivation).
Petals none. Disk annular, hypogynous, at least in the female plant,
glandular and fleshy, the margin 8-10-lobed. Male flowers: of these
only a single one is left on the specimen. The calyx resembles that
of the fertile flowers, except that it is less deeply cleft. Stamens
indefinitely numerous: filaments subulate-filiform, twice the length of
the calyx: anthers extrorse, two-celled, didymous, not appendiculate.
Female flowers not seen, except the calyx (with the disk) persistent

FLACOURTIACES. | 77

at the base of the advanced or ripened pistil. Fruit baccate, oblong-
ovoid, 3 or 4 lines long, a little narrowed at the base, the apex pointed
with an extremely short style, which is tipped with a depressed or
somewhat peltate two-lobed persistent stigma, one-celled, with 2
parietal placentz. The ovules are from 2 to 4 on the upper part of
each placente, oval, pendulous, anatropous. Seeds 4 to 6, or by
abortion 2, one pendulous from the upper portion of each placenta,
on a thickish funiculus, pretty large, so that the two fill the cell,
obovate; the testa smooth and even, marked with a narrow raphe
and an apical chalaza. Embryo as long as the fleshy albumen,
straight: radicle cylindrical, superior: cotyledons foliaceous, cordate-
oval.

According to Forster, the fragrant wood of this tree is steeped in
cocoa-nut oil, which, thus scented, is used by the Tahitians to anoint
their hair. I find no allusion to this custom as still existing, how-
ever, in the Narrative of the Expedition, or in Dr. Pickering’s volume;
further than that, in the former, it is said that “this oil is often
scented with aromatic herbs, to be employed by the natives in anoint-
ing the hair and body.” Except our imperfect specimens, the plant
does not appear to have been collected by any botanist since Cook’s
voyage; but Guillemin has printed Forster’s manuscript description
of the plant, and Mr. Bennett has recently given, from Forster’s
specimen and drawing preserved in the Banksian herbarium, an
amended generic character. The structure of the seed and embryo,
however, are not stated; and on all accounts it is desirable to figure
the plant, even from the incomplete materials which our collection
affords.

Pirate 4, A.—XyYLOSMA SUAVEOLENS. Fig. 1. A branchlet of the
staminate plant, in flower. 2. Exterior, and 3, interior view of a
stamen, magnified. 4. Pistillate plant, in fruit. 5. Calyx and the
disk. 6. A longitudinal, and 7, a transverse section of the fruit,
with the persistent calyx. 9. A seed. 10. A longitudinal, and 11,
a transverse section of the seed.—The details are all variously mag-

nified.
20

78 PHANEROGAMIA.

2. XYLOSMA ORBICULATUM, Forst. (Tab. 4).

X. foliis ovato-orbiculatis coriaceis integerrimis; calyee breviter quadri-
Fido, lobis obtusissumis.

X. orbiculatum, G. Forst. Prodr. Fl. Ins. Austr. p. 72.
Myroxylon orbiculatum, J. R. Forst. Char. Gen. p. 126.

Has. Island of Tongatabu.

Of this species, also, the collection contains only a poor specimen,
with male flowers. The leaves, about 3 inches long, and 22 inches
wide, are rounder, of a thicker and more coriaceous texture than those
of the preceding species, and quite entive. The male inflorescence and
flowers, however, are similar, except that the calyx is not so deeply
Jour-cleft, and its rounded labes very obtuse.

PLATE 4, B.—XYLOSMA ORBICULATUM. A branch of the staminate
plant, in flower. Fig. 1. A flower, magnified. 2. Vertical section of
the same. 3. Calyx, with the disk. 4, 5. Interior and exterior
view of a stamen, more magnified.

Orv SAMYDACE ZA.

1. CASEARIA, Jacg., Endl.
§ 1. PITUMBA (Aubl. Pl. Guian. 2, t. 385), Benth. ia Hook. Jour. Bot. 4, p. 110.
1. CASEARIA OBLONGIFOLIA, Camb.
Casearia oblongifolia, Camb. in St. Hil. Fl. Bras. Mer. 2, p. 234; Mart. Herb. Bras.
no. 469, p. 272.
Bigelovia Brasiliensis, Spreng. Neue Entd. 2, p. 150, t. 2, f. 1-6.
Samyda Selloi, Spreng. Syst. Veg. 2, p. 354.

Has. Near Rio Janeiro, Brazil.

After Martius, I adduce the synonyme of Sprengel; but the plate
above cited represents the stamens as exserted, and gives no squamz
or sterile filaments. One of our specimens accords with those dis-
tributed by Martius; the others have mostly larger leaves, more in-
clined to obovate, and more cuneate at the base. It is in Salzmann’s
collection from Bahia, under the name of “ Vaccinium laurifolia.” It
belongs to the section Pitumba.

2. CASEARIA INZQUILATERA, Camb.
Casearia inequilatera, Camb. in St. Hil. Fl. Bras. Mer. 2, p. 237, t. 127.

Has. Brazil, with the preceding species.

3. CASEARIA AFFINIS, Gardn.

Casearia affinis, Gardn. in Hook. Lond. Jour. Bot. 1, p. 529.
C. parviflora, Camb. in St. Hil. Fl. Bras. 1. c., non Willd.

80 PHANEROGAMIA.

Has. Near Rio Janeiro, Brazil: probably from woods on the Cor-
covado, where Gardner found the plant.

Our specimen bears ripe fruit only, which is exactly globose, mostly
one-seeded, about 4 lines in diameter, and tipped with a short persis-
tent style, which appears to be entire.

4, CASEARIA? ACUMINATISSIMA, Sp. Nov.

C. glabra; foliis nitidis oblongo-lanceolatis longe acuminatis basi rotun-
datis vix punctatis subintegerrimis; fructu obovato-globoso trivalvi
monospermo.

Has. Sandal-wood Bay, Vanua-levu, one of the Feejee Islands.
(In fruit only.)

The specimen consists of a branch of a glabrous shrub or small tree,
with ripe fruit. The leaves are oblong-lanceolate, with a long and
slender acumination and a rounded base: they are chartaceous in tex-
ture, smooth and shining both sides, copiously feather-veined, mostly
entire, occasionally showing a few minute teeth, obscurely if at all
pellucid-punctate, 2 or 3 inches long, an inch or less in width: the
petioles a line and a half in length. Stipules minute. Flowers not
seen. They appear to have been solitary in the axils of leaves
(which have mostly fallen) of the slender axillary branchlets of the
season. Pedicels in fruit 4 or 5 lines long. Sruit obovate-globose,
naked, 4 lines long, tipped with a short and apparently entire persis-
tent style, probably fleshy and becoming coriaceous, glabrous, one-
celled, tardily three-valved from the apex, filled with a single seed.
The vestiges of numerous ovules are still visible, however, on 3
filiform, parietal placente. The solitary seed is globular, nearly
3 lines in diameter; its dark-coloured crustaceous testa covered with
a thin and fleshy pellicle; its base marked with a circular impressed
raphe; the micropyle superior, apiculate. Albumen fleshy, rather
soft. Embryo in the axis of the albumen next the micropyle,
straight, fully half the length of the albumen: radicle superior, about
the length of the flat, orbicular-cordate cotyledons.—The flowers are
needed for the certain determination of the genus.

SAMYDACES, 8]

§2. MELISTAURUM. (WMelistaurum, Forst.)—Stamina fertilia 10. Stigma sub-

sessile, crassum. Ovula in placentis pauca.
5. CASEARIA DISTICHA. (Tab. 6.)

C. foliis oblongis seu ovali-oblongis membranaceis crebre punctatis brevis-
sime petiolatis ; floribus e gemma axillari pluribus parvis subsessilibus ;
filamentis fertilibus glabris; stigmate imtegerrimo.

Melistaurum distichum, Forst. Char. Gen. p. 143, t. 72? ideo
Samyda polyandra, Willd. Spec. Pl. 2, p. 6267 et
Casearia Melistaurum, DC. Prodr. 2, p. 51?

Has. Sandal-wood Bay, Vanua-levu, one of the Feejee Islands.

A tree or- shrub (its size not recorded), glabrous throughout, with
flexuous branches. Leaves oblong, or oval-oblong, the lower inclining
to ovate, acutish or obtusely acuminate, membranaceous, entire,
thickly punctate with round and oblong pellucid dots, from 3 to 5
inches long, on petioles of about 2 lines in length: the leaves are dis-
tichously arranged. Stipules very minute. lowers very small
(barely a line in length), crowded in a sessile axillary glomerule or
capitulum, on very short pedicels, subtended by minute scaly bractlets.
Sepals 5, obovate, longer than the stamens. Stamens monadelphous
at the base: fertile ones 10, in two ranks, the five exterior shorter:
their filaments subulate and glabrous: anthers didymous. Sterile
filaments 10, interior, subulate, nearly equalling the shorter fertile
series, villous-bearded. Ovary conoidal, glabrous, one-celled, with
3 parietal placenta, each bearing 4 to 6 ovules in two series. Style
none, or a mere contraction at the apex of the ovary. Stigma. thick,
depressed-capitate, entire. Fruit not seen.—In one specimen nearly
all the flowers are in an abnormal or diseased state; when they are
enlarged, and the deformed sterile filaments are exserted in the form
of horns or callous processes. The larger flowers of the specimen
figured are more or less thus affected.

This plant is, I suspect, not specifically different from Forster's

Melistaurum distichum, from New Caledonia, of which only the generic
21

82. PHANEROGAMIA.

characters were ever published. The specimen preserved in the
British Museum, however, has larger leaves than ours. Forster sup-
posed his plant to be polygamo-dicecious, and that he had only male
flowers; which may be the case; but the flowers I have examined
in our specimen are hermaphrodite. They accord with the Casearve
of the section Pituwmba, except in the sessile, or nearly sessile, and
thickened stigma, and the fewer ovules ;—characters which may serve
to define a sixth section of the genus.

PLATE 5, A.—CASEARIA DISTICHA. Fig. 1. An expanded flower,
magnified. 2. Andrcecium displayed, magnified. 3. A portion of
the same, more magnified (inside view). 4. Exterior view of one of
the stamens. 5. The pistil, magnified.

6. CasEARIA Ricuu, Sp. Nov. (Tab. 5.)

C. foliis ovatis subacuminatis vix punctatis subcoriaceis modice ipetiolatis
distichis ; floribus minimis e gemma axillari fasciculatis subsessilibus ;
filamentis fertilibus inferne ciliatis ; stigmate subtrilobo.

Has. Ovolau, Feejee Islands.

A close congener of the foregoing species; from which it is distin-
guished by its much smaller and crowded ovate and somewhat taper-
pointed leaves (from one to 2 inches long), of a rather coriaceous tex-
ture, very obscurely punctate, and on petioles of greater proportional
length: they are equally distichous, smooth, and entire. The flowers
are apparently still smaller, from similar azwillary and sessile buds.
Their jertile filaments, moreover, are ciliate below the middle; the ovary
is more ovoid; and the depressed-capitate and sessile stigma is some-
what three-lobed.

Puate 5, B—Casearta Rica. Fig. 1. Diagram of the flower.
2. Vertical section of the flower. 3. A stamen, with a sterile fila-
ment; inside view. 4. A stamen, outside view. 5. The pistil—The
analyses all magnified.

Orv VIOLACE 4.

1. VIOLA, nnn.
* Antarctice.
1. VIoLA TRIDENTATA, Menzies.

Viola tridentata, Menzies, in Herb. Banks; DC. Prodr. 1, p. 300; Hook. f. Fi.
Antare. p. 245.

Has. Orange Harbour, Fuegia.

This remarkable Violet has recently been admirably characterized
by Dr. Hooker. The stipules are wholly adnate to the petiole, which
is thus scariously winged; the only free portion is a minute herba-
ceous tooth at the base of the blade on each side, and this is often
wanting. The leaves somewhat resemble the. leaflets of Potentilla
tridentata. The stems are suffrutescent.

2. Viota Commersoni, DC.
Viola Commersonii, DC. Prodr. 1, p. 297; Hook. f. Fl. Antare. p. 245.
Has. Orange Harbour, Fuegia.
Two. diminutive specimens of this rare. species occur in the collec-

tion, mixed with those of the succeeding. It has been found only by
Commerson, Mr. Darwin, and in this Expedition.

84 PHANEROGAMIA.

3. VioLA MAGELLANICA, Forst.

Viola Magellanica, Forst. in Comm. Gott. 9, p. 41, t. 8; DC. Prodr. 1, p. 297;
Hook. f. Fl. Antare. p. 244.

Has. Orange Harbour, Fuegia.

Some of the specimens are no larger than those of V. Commersoni ;
others have scapes fully a span in height, and leaves and flowers as
large as those of V. odorata. Iam by no means sure that the corolla
is yellow. It bears some later flowers which are much smaller than
the others.

* * Australasice.
4, VIOLA HEDERACEA, Labill.

Viola hederacea, Labill. Fl. Nov. Holl. 1, p. 66, t. 91; DC. Prodr. 1, p. 805; Hook.
Exot. Fl. 3, t. 225.
Erpetion hederuceum, petiolare, & reniforme, Don, Syst. Gard. & Bot. 1, p. 336.

Haz. Sydney, New South Wales; the form with the leaves reni-
form-truncate, or a little cuneate at the base: Woolongong; the form
with round-reniform and mostly toothed leaves. They all manifestly
belong to one polymorphous species.

5. VIOLA SIEBERIANA, Spreng.

Viola Sieberiana, Spreng. Syst. Cur. Post. p. 96.
V. spathulata, Sieber, Pl. Nov. Holl. no. 406, non Willd.
V. Sieberi, Hook. Comp. Bot. Mag. 1, p. 274; Hook. f. Jour. Bot. 2, p. 406.

Has. Near Sydney, New South Wales.

This can scarcely be confounded with any form of the preceding
species.

VIOLACES, 85

* * * Sandwicenses.
6. Viota KAvAENSIS, Sp. Nov.

V. caulibus gracilibus repentibus; foliis subcordato-rotundis coriaceis
nervosis appresse serratis, dentibus glanduliferis; stipulis sqwamaceis
ovato-lanceolatis acuwminatis setigero-dentatis; petalis ceeruleis calyce
subbrevioribus (an semper ?), inferiore haud calcarato.

Has. On mountains of Kauai, Sandwich Islands, with Drosera
longifolia.

Stems, or stolons, creeping, flexuose, rather slender, glabrous, fur-
nished with rather large, ovate or ovate-lanceolate, squamaceous and
brownish stipules, of which the upper ones are narrower and more acw-
minate ; all beset with sharp or bristle-shaped and glandular-tipped teeth.
Leaves mostly crowded at the extremity of the floriferous stolons,
rounded and slightly cordate, or somewhat reniform, but with the base
more or less cuneate-decurrent into the slender petiole, coriaceous,
thickly punctate, an inch or less in diameter, closely serrate woth
obtuse and inflexed-appressed teeth, which bear a gland at their apex,
the surface strongly nerved or nervose-veined, and glabrous in the
specimens, although said to be hairy in Dr. Pickering’s note of the
locality, &e. Peduncles 13 to 2 inches long, nearly the length of the
petioles, glabrous, furnished with a pair of linear-lanceolate, slightly
glanduliferous bractlets a little below the flower. Sepals oblong-
lanceolate, glabrous, thickened and produced at the base. “ Petals
pale blue,” according to Dr. Pickering’s notes. The flowers of the
specimens are probably late ones, with the petals smaller than when
fully developed, as they are not quite so long as the calyx: they are
oblong-spatulate; the lower one a little larger than the others, and
only inconspicuously saccate at the base; all of them smooth and
beardless. What strengthens the suspicion that these are only such
precociously fertilized and cryptopetalous flowers as are. produced by
many Violets is, that the stamens are scarcely shorter than the petals,
all with distinct and rather narrowed filaments, which are longer
than the oblong anther-cells, and similar, except that the two lower

ones are rather larger: the connective in these merely carinate on the
| 22

86 PHANEROGAMIA.

back, but not at all spurred. The style, also, is very short, thick,
and uncinate-cucullate. In this view, the species may be compared
with Viola sarmentosa, of Oregon, from which, however, it is abun-
dantly distinct. But if these are the normal flowers, the species
must be ranked with the Australian V. (Erpetion) hederacea; and it
has a similar mode of growth.

7. VioLA CHAMISSONIANA, Ging. (Tab. 6.)

V. caule fructicoso erecto; foliis oblongo-ovatis subcordatis acuminatis
serrulatis, ramealibus basi nunc acutis; stipulis ovatis lanceolatisve
acuminatis seepe denticulatis ; calcare brevi late saccato; antheris apice
subulato-appendiculatis.

Viola Chamissoniana, Gingins, in Linnea, 1, p. 408, &
V. trachelitfolia, Gingins, |. ¢. p. 409.

Has. Oahu, Sandwich Islands; on the Kaala Mountains, and in
the vicinity of Wailuka.

Stems upright, from one to 6 feet high, woody, and from one-eighth
to more than half an inch in thickness, branching above, the branches
annulate with the cicatrices of former leaves and stipules, glabrous.
Leaves ovate, or mostly oblong-ovate, and somewhat cordate, acuminate,
serrulate with rather appressed glandular-tipped teeth, glabrous, or a
little pubescent beneath when young, membranaceous or chartaceous,
veiny; the principal cauline ones from 3 to 6 inches long, mostly
subcordate, and not unlike those of Trachelium cceruleum; their
petioles one or two inches long: leaves of the flowering branches suc-
cessively smaller and shorter petioled, merely obtuse or the upper
acute at the base. Stipules scale-like in texture, brownish, small; the
lower triangular-ovate, the uppermost lanceolate, acuminate, either
entire or beset with a few lacerate or setaceous teeth, which are
usually glandular-tipped. Peduncles axillary, half an inch to an
inch long, 1-2-bracteolate above the middle; the bractlets lanceolate-
subulate. Flower 6 to 8 lines long. Sepals very little produced at
the base, linear-lanceolate; the two lower sometimes a little broader
and shorter than the others. Petals apparently light blue or violet,
oblong-spatulate, beardless, nearly equal, the lower one broadly sac-

VIOLACES. 87

cate-calearate at the base. Anthers glabrous, tipped with a slender
subulate appendage; the two lower bearing a thick and short very
obtuse spur on the back. Style sigmoid-curved, thickened and
gibbous at the apex, terminated by a unilateral truncate stigma.
Capsule ovoid-oblong, smooth. Seeds as in the genus: none were
seen with a matured embryo.

Gingins’ V. Chamissoniana, of which I have seen an authentic
specimen in the Hookerian herbarium, was evidently described from
a plant that wanted the principal cauline leaves, which are more or
less cordate. It is clearly the same species as the V. tracheliifolia of
the same author, who took the character from a drawing made by
Chamisso. The plant varies greatly in height, and in the size of its
leaves, &c. The only discrepancy between our specimens and the
description of Gingins is in the appendage of the anthers, which he
states to be obtuse, without mentioning its form; while in our plant
it is reduced to a narrow and subulate process.

Puate 5.—VIOLA CHAMISSONIANA: the summit, with the base of
the woody stem. Fig. 1. A young seedling plant. 2. Stipules, en-
larged. 8. Diagram of the flower. 4. A flower, enlarged. 5. Same,
with calyx and corolla removed. 6. Three detached petals. 7. A
back view, and 8, a front view of one of the two lower stamens, mag-
nified. 9. A back view, and 10, a front view of one of the three
upper stamens, equally magnified. 11. Pistil, with the ovary longi-
tudinally divided. 12. A seed, magnified.

2, IONIDIUM, Vent.

1. Ionmprum monoPEeTALUM, Reem. & Schult.
Tonidium monopetalum, Reem. & Schult. Syst. Veg. 5, p. 400.
Pigea monopetala & P. filiformis, DC. Prodr. 1, p. 3807.
Solea monopetala, Spreng. Syst. Veg. p. 804.
Has. Near Sydney, New South Wales.

The ample lower petal is either slightly emarginate or entire.

88 PHANEROGAMIA.

38 NOISETTIA, 7 BK.

1. Norserria LonciIroLiA, H. B. K.

Noisettia longifolia, H. B. K. Nov. Gen. & Spec. 5, p. 882, t. 499; DC. l.c.; St.
Hil. Pl. Rem. Bras. p. 385, t. 26.

N. orchidiflora, Ging. in DC. Prodr. 1, p. 290.

Viola longifolia, Poir. Dict. 8, p. 649.

V. orchidiflora, Rudge, Pl. Guian. Rar. 1, p. 111, t. 10.

lonidium longifolium & I. orchidiflorum, Rem. & Schult. Syst. Veg. 5, p. 898, 400.

Has. Organ Mountains, near Rio Janeiro, Brazil.

4. ANCHIETEA, S&. Hil.

1. ANCHIETEA SALUTARIS, St. Hil.

Anchietea salutaris, St. Hil. Pl. Us. Bras. t. 19, Pl. Rem. Bras. p. 290, & Fl. Bras.
Mer. 2, p. 141.

Noisettia pyrifolia, Mart. & Zuce. Nov. Gen. & Spec. Bras. 1, p. 24, t. 16.
Viola summa, Velloz. Fl. Flum. 8, t. 161.

Has. In the vicinity of Rio Janeiro, Brazil. (1n flower only.)

5 AMPHIRROX, Spreng.

1. AMPHIRROX LONGIFOLIA, Spreng.

Amphirroa longifolia, Spreng. Syst. Cur. Post. p. 51, & 99.
Spathularia longifolia, St. Hil. Pl. Rem. Bras. p. 318, t. 28, & Fl. Bras. 2, p. 107.

Has. Near Rio Janeiro, Brazil.

The specimen bears ripe fruit only. The capsule is ovoid-trian-

VIOLACE S&. 89

gular, or somewhat three-lobed, nearly an inch long, papillose; the
valves crustaceo-coriaceous. Seeds numerous upon each of the three
placentz, 23 lines long, obovate, with a smooth testa, marked with a
slender rhaphe and a broad apical chalaza. Embryo not seen in the
seeds examined.

6. AGATEA, Nov. Gen.

Calyx pentaphyllus subequalis, basi haud productus, deciduus. Corolle
petala 5, hypogyna, erecta, inequalia; postica lateralibus paullo
minora; anticum majus, labelliforme, spathulatum, basi dilatatuin
grbboso-saccatum. Stamina 5, diadelpha, nempe; jfilamenta brevia,
plana, antica (singula glandula carnosa aucta) et lateralia margi-
nibus connata, posticum angustius distinctum: anthercee introrsum
adnate, loculis appositis apice liberis mucronatis ; connectivo in appen-
dicem petaloideam latam producto. Ovariwm globosum, placentis
paretalibus 3 pluriovulatis. Stylus apice clavatus, subcurvatus :
stigma laterale. Fructus baccatus ?— Frutex sarmentosus; foliis
alternis oblongis subintegerrimus ramisque glabris; stipulis minimis
caducis; racemis paniculisve axillaribus multifloris; p2dicellis 2-3-
bracteolatis infra apicem articulatis ; floribus parvis viridulis.

1. AGATEA VIOLARIS, Sp. Nov. (Tab. 7.)

Var. a. foliis oblongo-lanceolatis integerrimis vel obsolete repandis pani-
culas suboeequantibus.

Var. 0. foltis majoribus ovato-oblongis nunc repando-subdentatis pant-
culas excedentibus.

Has. Feejee Islands: the var. a. from Naloa Bay. Var. @. Ovolau.

A shrubby, sarmentose, or somewhat climbing plant, entirely gla-
brous, except the inflorescence. Branches slender, leafy to the top.
Leaves alternate, thin, chartaceous in texture, smooth, in the speci-

men from Naloa Bay (var. a.) oblong-lanceolate, 3 or 4 inches in
28

90 PHANEROGAMIA.

length, and an inch or less in breadth, entire or sometimes obscurely
repand; in the specimen from Ovolau, broader and mostly oblong-
ovate, 34 to 5 inches long and 14 to 23 wide, the larger ones obso-
letely repand-toothed; all more or less acuminate, loosely pinnately
veined. Petioles half or three-quarters of an inch in length. St-
pules very minute, squamaceous, triangular, caducous. Inflorescence
from the axils of the upper leaves, and terminal, minutely tomentose-
pubescent; the flowers numerous, in compound and paniculate, or
occasionally simple racemes, which in var. a. often exceed the leaves,
but in 8. (in which, however, the flowering is less advanced) are much
shorter. Peduncles and pedicels slender; the latter 3 or 4 lines long,
recurved or drooping, subtended by a minute scale-like bract, and
furnished with 2 or 3 mostly alternate bractlets resembling the bract,
articulated a little below the flower and above the uppermost bractlet.
Oalyx small, about one-quarter of the length of the larger petal, deci-
duous after flowering, of 5 distinct sepals: these are imbricated in
estivation, not appendaged at the base, greenish, ovate or oblong-ovate,
pubescent-ciliate, somewhat unequal; the two lateral, which are exte-
rior in the bud, being a little shorter than the others. Corolla hypo-
gynous, of 5 erect and wnequal (greenish white) petals, obliquely imbri-
cated in sestivation, namely with one (of the two posterior) exterior,
and one (the anterior and saccate one) interior; in the manner of
Viola, Ionidium, &. All the petals are obscurely ciliate, and not
unguiculate; the two posterior linear-oblong, with their tips somewhat
spreading; the two lateral similar to the posterior in shape, but a litile
larger, and with the margins inclined to be involute about the middle;
the anterior one larger, and one-third longer than the lateral, hairy
inside below the middle, labelliform or panduriform; the upper portion,
or true lamina, obovate-spatulate, slightly emarginate, involute in the
bud, continuing to be involute from the middle to near the base,
which is abruptly dilated and gibbous-saccate, the sac-like projection
hemispherical externally. Stamens 5, hypogynous; the short /ila-
ments very broad and dilated, shorter than the anthers, diadelphous,
namely, all but the posterior one united by their edges into an imper-
fect ring or cup; the two lateral ones inappendiculate and nearly
smooth externally, the two anterior pubescent externally, and each
thickened on the back by an adnate, large and fleshy gland or tubercle :
filament of the posterior stamen resembling the lateral ones, but rather
narrower, smooth, and not appendiculate on the back, distenct, or con-

VIOLACE &. 9]

nected with the lateral ones at the very base only. The filaments
are continued into a membranaceous, yellowish-brown, oblong-ovate
and obtuse, entire, petaloid appendage, or production of the connective,
which much exceeds the introrsely adnate anther. © The two cells of
the anther are narrowly oblong, longitudinally dehiscent, as in the
order generally; but the apex of the anther is free, and tipped with a
(simple or double) mucro (as in several Alsodewe), which in some
cases is conspicuous, in others nearly obsolete. Ovary globose, canes-
cently pubescent, one-celled, with 3 parietal many-ovuled placenta.
Style elongated, glabrous, somewhat thickened upwards, slightly curved,
the apex more thickened and produced on one side: the stigma small
and simple, terminating the lateral projection. Ovules indefinite,
occupying several rows on each placenta, obovoid, anatropous. Fruit
unknown.

On account of the great thickness of the walls of the fertilized ova-
ries (from which the sepals have fallen) I have little doubt that the
fruit is a berry: but even if it prove to be capsular, the structure of
the stamens, with the deciduous sepals, will sufficiently distinguish
the plant from Jonidiwm, of which it has not the habit; and the very
different corolla equally separates it from Corynostylis, Mart. & Zucc.
(Calyptrion, Gingins).

As the ancients garlanded the graves of their deceased friends with
violets, so I dedicate this new genus of Violacece to the memory of
Alfred T. Agate, the Botanical Artist of the Expedition, who died
at Washington shortly after its return. The engraving is copied from
a drawing made by Mr. Agate, from the fresh plant, at the Feejee
Islands. I trust the name Agdtea will be thought sufficiently distinct
in pronunciation both from Agathéa and Agat.

The odd and saccate petal in this plant, and probably in the whole
Violet Family, is truly anterior (vide Gray, Genera Flore Am. Bor.-
Or. Illustrata, 1, p. 183), and does not become so by resupination, as
the order is characterized by Bartling, Endlicher, &c.

Prats 7.—Aaqarea viotaris. Fig. 1. Diagram of a flower. 2. A
flower, with its pedicel and bractlets. 3. Lower petal. 4. A lateral,
and 5, an upper petal. 6. A flower, with the calyx and corolla
removed. 7. The andrcecium spread open; seen from without. 8.

92 PHANEROGAMIA.

The same, seen from within. 9. The upper stamen. 10. One of the
lower stamens. 11. The pistil. 12. Transverse, and 13, longitudinal
section of the same. The details variously magnified.

7 ALSODETA, Petit-Thouars, R.Br.

1. ALsopEIA PHYsIopHoRA, Mart. & Zuce.
Alsodea physiophora, Mart. & Zuce. Nov. Gen. & Spec. Bras. 1, p. 28, t. 19.

Has. Near Rio Janeiro, Brazil. (In fruit only.)

8. ISODENDRION, Nov. Gen.

Calyx pentaphyllus, cequalis, persistens. Corolle regularis petala 5
linearr-spathulata, cequalia, longe tubuloso-conniventia, apice dilatata
patentia. Stamina 5, discreta: filamenta angusta, inappendiculata,
apice haud producta, antheram basificam nudam gerentia. Ovarium
untloculare, placentis 3 parietalibus biovulatis. Stylus elongatus, sub-
clavatus, apice decurvus, simplex: stigma punctiforme laterale. Ovula
collateralia, horizontalia. Capsula coriacea, sepalis petalisque persis-
tentibus cincta, trisperma, trivalvis. Semina obovata.—Arbuscule vel
Srutices Sandwicenses ; foliis alternis confertis; stipulis triangulart-
subulatis appressis diu persistentibus; floribus axillaribus solitariis
breviter pedicellatis parvis.

A genus evidently allied to Alsodeia, Paypayrola, Aubl. (which, as
illustrated by Tulasne, in Ann. Sci. Nat. 3d ser. 7, p. 368, appears to
differ from Alsodeia principally in the inappendiculate stamens), and
Pentaloba, Lour., which Mr. Brown long since stated to be congeneric
with Alsodeva, if it should prove to have a capsular fruit.* From these

* Prof. Arnott (in Mag. Zool. & Bot. 2, p. 543), in uniting these two genera, has
adopted the older name of Pentaloba; but the inappropriateness of this name even for
Loureiro’s own plant, in which Mr. Brown found only three placenta, should forbid its
superseding the long-admitted name of A/sodeia.

VIOLACES. 93

the present genus is distinguished by its entirely separate stamens, with
narrow filaments and normal anthers, destitute of any prolongation of
the connective, and by the unilateral stigma, which, in a flower otherwise
perfectly regular, vindicates its relationship with the genuine Violew.
In allusion to the regular flowers and arborescent habit, I have com-
pounded the name of the genus from ies, equal, dédpiv, tree, and tov, the
Violet.

1. IsopENDRION PyRIFoLIUM, Sp. Nov. (Tab. 8.)

L, foliis membranaceis ovalibus sew ovato-ellipticis crenato-serrulatis petio-
latis, junioribus subtus ramulisque pubescentibus ; stipulis sepalisque
dorso sericas margine late scariosis ; floribus pendulis,

Has. Wooded portion of the Kaala Mountains, Oahu, Sandwich
Islands.

A branching shrub, about 6 feet high; the branches rather slender,
glabrous, except the younger, which are a little pubescent, beset with
the persistent stipules, which remain long after the leaves have fallen.
Leaves alternate, approximate at the summit of the branches, membra-
naceous in texture, ovate-ellipticul or oval, obtuse or rounded at both
ends, 12 to 2 inches long, loosely pinnately veined, crenate or minutely
serrate, with small and glandular-tipped teeth, glabrous, or the lower
surface minutely pubescent when young, as are the petioles (3 to 5
lines long). Stipules geminate, triangular, acute or acuminate, ob-
lique, squamaceous or scurious, carinate with a very strong and thick
midrib which is minutely siky-pubescent externally, ciliolate on the
margins, a line and a half long; the two partly overlying each other
in the axil of the petiole, as if intrapetiolar, close-pressed to the
axis, somewhat imbricated at the summit of the branchlets, more
remote on the larger branches, persistent, or at length wearing away.
Flowers solitary and axillary, developed with the leaves. Pedicels
recurved, 2 or 3 lines long, scarcely as long as the flower, minutely
pubescent, bibracteolate in the middle; the bracts opposite, resembling
the stipules. Calyx of 5 equal and nearly distinct sepals, which are
ovate and acute, not at all produced at the base, of a texture and form

nearly the same as the stipules, somewhat carinate and minately silky-
24

94 PHANEROGAMIA.

pubescent on the back, the margins more or less scarious, quincuncially
imbricated in estivation, persistent. Corolla hypogynous, or obscurely
perigynous, regular, hypocraterimorphous; the five petals distinct, but
connivent for the greater part of their length into a cylindrical tube,
the dilated summit, or true lamina, abruptly spreading, slightly pubes-
cent externally. They are all alike in size and shape, about 4 lines
long, probably white, not appendiculate, linear-spatulate (that is, the
connivent claws, if they may be so called, are linear, and the spread-
ing lamina ovate), convolutely imbricate in sestivation, in the manner
of Violacee: generally (that is, with only one petal wholly external and
one (the anterior) internal), at length separated by the growth of the
fructified ovary, marcescent and persistent, with the calyx, around the
base of the ripe capsule. Stamens 5, hypogynous, or obscurely peri-
eynous, alternate with the petals, and much shorter than they, longer
than the ovary, all similar, entirely distinct and separate, glabrous.
Filaments subulate-linear, somewhat flattened, longer than the anthers,
entirely distinct, destitute of any dorsal glands or other appendages,
not dilated or prolonged at the apex, but bearing the anthers on their
summit. Anthers oblong, fixed by the base, introrse; the cells
approximate, joined by a narrow connective, longitudinally and some-
what introrsely dehiscent, muticous. Pollen-grains simple, spherical.
Disk none. Ovary closely sessile, ovoid, glabrous, one-celled, with 3
parietal placente. Style elongated, included in the corolla, some-
what curved, filiform, a little club-shaped at the simple apex, which
is somewhat gibbous and produced on the lower side, where it is
tipped with a minute and simple stigma. Ovules 6, namely, a single
pair on the middle of each nerviform placenta, collateral, horizontal,
anatropous. Capsule ovoid, coriaceous, about three-seeded, loculi-
cidally three-valved; the persistent style also splitting from the base
to the apex into three portions; the valves conduplicate after dehis-
cence, as in Viola, &e. Seeds obovate, smooth; the testa coriaceous,
marked with a slender rhaphe and a broad apical chalaza. Embryo
not seen.

Piate 8.—ISoDENDRION PyRIFOLIUM. Fig. 1. A pair of stipules,
persistent on a portion of the stem, enlarged. 2. One of these stipules,
detached. 3. Diagram of the flower. 4. A flower with its bractlets,
enlarged. 5. One of the petals, enlarged. 6. Stamens and pistil,
magnified. 7. Posterior view of a stamen, magnified. 8. Anterior

VIOLACES, 95

view of the same. 9. Magnified pistil, with the ovary vertically
divided, showing the ovules. 10. Apex of the style with the stigma,
more magnified. 11. The dehiscent capsule, with the persistent
floral envelopes, enlarged. 12. A seed, magnified.

2. IsopENDRION LoNeIFOLIUM, Sp. Noy. (Tab. 9.)

I. glabrum ; foliis subcoriaceis obovato-lanceolatis seu cuneato-oblongis in
petiolum angustatis subrepandis; sepalis ovatis stipulisque leevibus ;
floribus in ramos crassos brevissime pedicellatis.

Has. Kaala Mountains, Oahu, Sandwich Islands.

Stems 6 feet high, with rather few very stout branches, roughened
with the cicatrices of fallen leaves, and above with the persistent
stipules. Leaves alternate, approximate at the summit of the branches,
of a coriaceous texture, entirely glabrous, as is the whole plant, 6 to 8
inches long, 1 to 2% inches wide, obovate-lanceolate, varying from
obovate-oblong to oblong-lanceolate, somewhat acute or acuminate, or
rarely obtuse at the apex, below tapering to an acute base, veiny, and
with the veinlets reticulated, the midrib very prominent underneath ;
the margins obscurely repand-undulate, or obsoletely and sparingly
serrulate. Petioles 4 to 14 inches long. Stipules geminate, subulate-
triangular from a broad base, glabrous, carinate with a thickened axis,
and with somewhat scarious margins, appressed, persistent long after
the leaves have fallen. Flowers axillary, solitary, on very short
drooping pedicels, which are bibracteolate next the base, and only
one or 2 lines long. Calyx of 5 nearly distinct, ovate and glabrous
sepals, quincuncially imbricated in estivation, not at all produced
at the base, scarcely a line long, persistent. Corolla hypogynous or
obscurely perigynous, regular, apparently white, scarcely 3 lines long:
petals 5, linear-spatulate, connivent for two-thirds of their length into
a tube, the upper and dilated portion spreading, convolutely imbri-
cated in gestivation, as in Violew, marcescently persistent. Stamens 5,
hypogynous or obscurely. perigynous, alternate with the petals and
much shorter than they, scarcely exceeding the ovary, all similar,
entirely unconnected, glabrous, Filaments subulate-linear, flattened,
about the length of the anthers which terminate them, destitute of

96 PHANEROGAMIA.

any dorsal or terminal appendages. Anthers oblong, inappendiculate,
exactly like those of the foregoing species. Disk none. Ovary,
ovules, style, stigma, &c., nearly as in the foregoing species. Mature
fruit not seen.—The leaf-bearing and flowering branches are about
a quarter of an inch in diameter: the flowers are small and inconspi-
cuous.

PLATE 9.—IsopENDRION LONGIFOLIUM. Fig. 1. Stipules, front and
back view, enlarged. 2. Diagram of the flower. 3. A flower, en-
larged. 4. A petal, more enlarged. 5. Stamens and pistil, magni-
fied. 6. The same after the removal of the stamens. 7. Vertical
section of the ovary, magnified.

3. ISODENDRION LAURIFOLIUM, Sp. Nov.

I. glabrum; foliis coriaceis oblongo-lanceslatis subrepandis basi obtusis
brevissime petiolatis ; sepalis lanceolatis.

Has. Sandwich Islands, with the preceding species.

Branches rather stout, entirely glabrous, as is the whole plant.
Leaves coriaceous, oblong-lanceolate, somewhat narrowed at both ends,
obtuse at the base, about 3 inches long and an inch wide, sparingly and
obscurely repand-serrulate, or entire, very short petioled, shining above,
pale beneath, not unlike those of Laurus nobilis. Petioles only one
or 2 lines long, seldom longer than the stipules. Pedicels a line long,
bibracteolate. Flowers apparently similar to those of the preceding
species; except that the sepals are lanceolate and much narrower.
Fruit not seen.

Only a single and incomplete specimen of this plant was gathered.
Perhaps it will be found to vary into I. longifolium; but the branches
are much less stout, the leaves are less than half the size, are all
obtuse or rounded at the base, and very short-petioled, and the sepals
are of a different shape. As to the rest of the flower, no especial dif
ference is manifest. The stipules are similar, but apparently less per-
sistent.

VIOLACES. 97

9. MELICYTUS, Poet.

Meticytus, Forst. Char. Gen. Fl. Ins. Austr. t. 62; Hook. f. Fl. N. Zeal. p. 17.

1. MELicytus RAMIFLORUS, Forst.

Melicytus ramiflorus, Forst. 1. ¢., & Prodr. Fl. Ins. Austr. p. 70; Lam. Ill. t. 812;
DC. Prodr. 1, p. 257; A. Rich, Fl. N. Zel. p. 313; A. Cunn. in Ann. & Mag,
Nat. Hist. 4, p. 256; Hook. f. Fl. N. Zeal. p. 18.

Has. Bay of Islands, New Zealand.

Dr. Hooker has with manifest correctness referred this genus to the
order Violacew, and has increased the number of the species to five.
The dorsal appendage of the anther is inconspicuous in the present
species: it is thickened and concave at the apex.

2. MELICYTUS MACROPHYLLUS, A. Cunn.
Melicytus macrophyllus, A. Cunn. in Ann. & Mag. Nat. Hist. 1. c.; Hook. f.

Has. Bay of Islands, New Zealand. (In fruit.)

The leaves of this species, as Dr. Hooker remarks, are larger than
those of the preceding, of a deeper green, and with fewer and coarser
teeth: the somewhat obovoid berries are more than twice as large.

10. SAUVAGESTA, Jacg.

«

1. SAUVAGESIA ERECTA, Linn.

Has. Organ Mountains, near Rio Janeiro, Brazil. Common almost

throughout tropical America.
25

Orv. DROSERACE &.

1. DROSERA, Linn.
* Antarctice & Novo-Zelandicex.
1. DrosERA UNIFLORA, Willd.

D. uniflora, Willd. Enum. p. 8340; DC. Prodr. 1, p. 8317; Gaud. Bot. Freyc. Voy.
p- 1837; Hook. f. Fl. Antare. p. 245.

Has. Orange Harbour, Fuegia; growing with Psychrophila dionee-
Folia, Drapetes, Ke.

The character given by Dr. Hooker well distinguishes this from the
other one-flowered species (all belonging to high southern latitudes),
except that for “perpusilla,” the term nana should be substituted, as
the plant is rather stout. The rounded blade of the leaves and the
short peduncles distinguish it from D. Arcturi. The expanded flower
measures about five lines across.

2. Drosera CUNNINGHAM, Walp.

D. Cunninghamii, Walp. Repert. 1, p. 229.
D. intermedia, A. Cunn. Fl. Noy. Zeal. in Ann. & Mag. Nat. Hist. 4, p. 110, non
Hayne.

Has. Bay of Islands, New Zealand.

Root fascicled, fleshy. Caudex none or scarcely any. Petioles,
scapes, &c., perfectly smooth and glabrous, both compressed: the

DROSERACES. 99

former 5 or 6 inches, the latter a foot or more in length. Leaves
simply two-parted; the divisions narrowly linear, scarcely broader
than the petiole, 2 or 3 inches long, glanduliferous above, the lower
surface naked. Flowers 12 to 20, smaller than those of D. binata,
and on more slender pedicels, but, in our specimens, mostly disposed
in a forked and equally subcymose inflorescence. Seeds scobiform,
linear; the testa produced at each end to thrice the length of the
globular nucleus.

* * Novo-Hollandice.
3. DROSERA PEDATA, Pers.

Drosera pedata, Pers. Ench. 1, p. 8837; DC. Prodr. 1, p. 319.
D. dichotoma, Smith, in Rees. Cycl. no. 6.

Has. Sydney, New South Wales.

Root of thickened fasciculated fibres. Caudex none or scarcely
any. Petioles and scapes compressed, especially the latter, which are
strongly ancipital, from 10 to 15 inches long. The numerous flowers
are larger than in the preceding species, but smaller than in D.
binata; the cyme trichotomous. Seeds scobiform, minute. Leaves
two-parted, with the divisions dichotomous, or often with one division
simple and the other dichotomous, or some of the earlier ones with
the primary divisions perfectly simple, as in D. binata; from which
the ancipital scape should distinguish it. I have a specimen of the
simple-leaved form, gathered at Sydney, I believe, by Dr. Hooker.

4, DROSERA PELTATA, Smith.

Drosera peltata, Smith, Exot. Bot. t. 41, & in Rees, Cycl.; DC. Prodr. 1, p. 319,
vix Labill.

Has. Sydney, New South Wales.

The sepals are beset externally, and especially on the margins,
with glandular bristles, as D. peltata is, I believe, described by Smith.

100 PHANEROGAMIA.

But Labillardiere’s D. peltata, according to his figure and description,
and some specimens from his collection, given to me by Mr. Webb,
has the calyx glabrous and naked, or the margins barely 2—3-toothed
towards the apex. I have the same plant from Sydney, &c., named
“D. petiolaris, Sieber,” by Planchon; but D. petiolaris, R. Br., is a
very different species. The true D. peltata also comes from Van
Diemen’s Land.

5. DROSERA SPATHULATA, Labill.

Drosera spathulata, Labill. Pl. Noy. Holl. 1, p. 79, t. 106, f. 1; DC. Prodr. 1, p. 318.

Has. Sydney, New South Wales.

6. DrosrrA PYGMHA, DC.

Drosera pygmea, DC. Prodr. 1, p. 317; Hook. f. in Jour. Bot. 2, p. 407; Lehm. PI.
Preiss. 1, p. 250.

D. pusilla, R. Br. ined. ex. DC., non H. B. K.

D. micrantha, Lehm. Pugill. Pl. 8, p. 39, ex Lehm. 1. c.

Has. Near Sydney, New South Wales.

The most minute of all the species; the lamina of the peltate
radical leaves less than a line in diameter; the capillary scapes half
an inch, or sometimes an inch long, the flower minute, in our speci-
mens always solitary. Lehmann describes the plant of Preiss as
2-6-flowered. The conspicuous stipules, as Dr. Hooker remarks,
present the appearance of a silvery star, surrounding the base of the
scape.

* * * Sandwicenses.

7. DRrosEeRA LONGIFOLIA, Linn.

Drosera longifolia, Linn. Fl. Suec. no. 274, & Spec. 1, p. 282, pro parte; Fries.
Novit. Fl. Suec. ed. 2, p. 82.

D. Anglica, Huds. Fl. Angl. p. 185; DC. Prodr. 1, p. 318; Hook. Brit. Fl., & FL
Bor. Am. 1, p. 81.

TREMANDRACEA. 101

Has. Mountains of Kauai, Sandwich Islands; “from a marsh, on
the table land.”

It is interesting to receive a Drosera from the Sandwich Islands,
and especially one of the boreal species. The specimens, in their
foliage, seeds, &c., are undistinguishable from Drosera longifolia
(applying that name, as I think we should, to the D. Anglica of
Hudson), a species which, with D. rotundifolia, is found quite around
the northern hemisphere from Oregon to Kamtschatka.

Orv. TREMANDRACES.

1 TETRATHECA, Smith.

TeTraTHEcA, Smith, Specim. Bot. N. Holl. 1, t. 2; Labill. Pl. N. Holl. 1, p. 95,
t. 122, 123; R. Br. in Flinders, Voy. 2, p. 544; DC. Prodr. 1, p. 843; Endl.
Pl. Hugel. p. 7, & Gen. Pl. 5644; Steetz, in Pl. Preiss. 1, p- 211; Payer, in
Ann. Sci. Nat. ser. 3, 15, p. 351.

1. TETRATHECA JUNCEA, Smith, l. c.
Has. Vicinity of Newcastle, New South Wales.

Steetz (1. c.) remarks that, having carefully examined many species
of Tetratheca, he has never found more than one ovule in each cell of
the ovary, and Payer (I. supr. cit.), chiefly upon this authority, denies
the existence of a greater number, notwithstanding the phrase “Ova-
rium 2-loculare, loculis 1—3-ovulatis,” in Mr. Brown’s ordinal cha-
ragter. To show, if need be, that this character is not “founded
upon an error,” I may state that I find ¢wo (superposed) ovules in

26

102 PHANEROGAMIA.

each cell of the ovary of Tetratheca juncea, and T. thymifolia, and three
(the two upper collateral) in 7: affinis. M. Payer is not more fortu-
nate in his statement, that Mr. Brown characterized Tetratheca by its
tetramerous flowers and biovulate cells of the ovary, and Tremandra
by its pentamerous flowers and solitary ovules; whereas Mr. Brown
has refrained from mentioning a single distinctive character of the
two genera. It was DeCandolle who inferred that Tetratheca was
always tetramerous, and Endlicher, who assumed in his Genera Plan-
tarum, that the cells of the ovary were always biovulate, he having
previously found them so in his 7. affinis and T. setigera, Pl. Hugel,
l.c. The latter species I have not seen. In the former, the single
ovary which my specimen (from Drummond’s Swan River collection)
furnishes, plainly exhibited three ovules in each cell. Both are said
by Steetz to have solitary ovules.

2. TETRATHECA ERICIFOLIA, Smith.
T. erictfolia, Smith, Exot. Bot. p. 87, t. 20; Rudge, in Linn. Trans. 8, t. 11; DC. 1. e.
Has. Near Sydney and Newcastle, New South Wales.

Some of the specimens have the calyx and peduncles nearly gla-
brous, as the species is characterized. But in most of them they are
beset with scattered glandular-tipped bristles, as they are in other
specimens from different collectors; and their stems and leaves are
more scabrous. Dr. Steetz has remarked that several species vary in
this manner. The cells of the ovary are uniovulate.

3. TETRATHECA THYMIFOLIA, Smith.
T. thymifolia, Smith, Exot. Bot. t. 22; DC. Prodr. 1, p. 343.
Has. Vicinity of Newcastle, New South Wales.
There are two ovules in each cell of the ovary, one inserted above

the other, the lower one occupying nearly the middle of the ®ell.
The sepals in our specimens are scarcely or not at all ciliate.

Orv, POLYGALACE ®.

1. POL YGALA, Linn.

* Austro-Americane.
1. PoLYGALA STENOPHYLLA, Sp. Nov.

P. glabra; caulibus gracilibus ramisque erectis gunceis; foliis filifor-
mibus mox deciduis; racemis elongatis spicatis laxifloris; carina
cassidiformi alas calycinas subequante ample cristata seu in floribus
serotinis ecristata !

Has. Rio Negro, Northern Patagonia; on sand-hills.

Root apparently perennial. Stems slender, one to 2 feet high, her-
baceous but rigid, terete, green and rush-like, paniculately branched,
the branches also erect, filiform, glabrous, as is the whole plant. Leaves
alternate, all filiform or setiform, an inch or more in length, sparse,
the cauline ones deciduous. acemes or spikes terminating the
branches one or two inches in length, either rather densely or quite
loosely flowered. Bracts minute, ovate, scarious, caducous, about the
length of the very short pedicels, that is, less than a line long. Exte-
rior sepals ovate, obtuse, scarious or petaloid, nearly white. Wings
broadly obovate, very obtuse, nearly 2 lines long, white in the dried
specimens, as long as the carina. Corolla tinged with blue or purple:
the lateral petals narrowly oblong, equalling the carina, below adnate
to the claw for its whole length: the carina itself short and some-
what helmet-shaped, entire, surmounted with an ample multifid-fim-
briate crest, which, however, in the later flowers is reduced to a few
Jimbrie, and in some of them wholly disappears! Stamens 8. Style

104 PHANEROGAMIA.

strongly geniculate, thickened above, somewhat two-lobed at the apex ;
the upper (anterior) lobe acute, the other bearing a globular stigma.
Capsule 2 lines long, oval, not emarginate, glabrous. Seed oblong,
villous. Lobes of the caruncle linear-filiform, as long as the seed.

This is a well-marked, and apparently undescribed species; unless
it should prove to be the P. capillifolia, of unknown origin, imper-
fectly characterized by DeCandolle, from Desfontaines’ herbarium.
But that species is said to have a two-cleft capsule and an annual

root. The flowers of the present species are twice as large as those
of P. linoides, and the stems are rush-like.

2. PoLtyGaLa Cyparissias, St. Hil.
Polygala Cyparissias, St. Hil. Fl. Bras. Mer. 2, p. 15; Walp. Repert. 1, p. 239.

Has. Near Rio Janeiro, Brazil; on the beach.

3. POLYGALA PANICULATA, Linn.

Has. Near Rio Janeiro, and in the Organ Mountains, Brazil.

4. PoLYGALA LANCIFOLIA, St. Hil. ? 1. ¢.
Has. Organ Mountains, near Rio Janeiro, Brazil.

A single fragment is in the collection, which I can only doubtfully
refer to the above-named species of St. Hilaire.

5. PoLYGALA CESTRIFOLIA, St. Hil.? 1. ¢.
Has. Organ Mountains, near Rio Janeiro, Brazil.

These specimens likewise are too imperfect for certain _determi-
nation, but they best accord with the character of P. cestrifolia of St.
Hilaire. ;

POLYGALACESA 105

x x Africane.
6. PotyGaLaA GaARcINI, DC.

P. Garcini, DC. Prodr. 1, p. 8323; Eckl. & Zeyh. Enum. Pl. Afr. Austr. p. 21.
P. frutescens, etc. Burm. Pl. Rar. Afric. Dec. p. 203, t. 78, f. 3.

Has. Cape of Good Hope, near Cape Town.

7. PoLyGALA ERIOPTERA, DC.

Polygala erioptera, DC. Prodr. 1, p. 326; Deless. Ic. Sel. 8, t. 15; Webb, Spic.
Gorg. p. 1038.

Has. St. Jago, Cape de Verde Islands.

2. SALOMONTA, Lour.
1. SALoMONIA CANTONIENSIS, Lour.

Has. Singapore. Probably introduced from China.

3. COMESPERMA, Labill.
1. CoMESPERMA CONFERTA, Labill.
Comesperma conferta, Labill. Pl. Nov. Holl. 2, p. 23, t. 161; DC. Prodr. 1, p. 334.

Has. Near Sydney, New South Wales.

2. COMESPERMA VIRGATA, Labill.

Comesperma virgata, Labill. Pl. Noy. Holl. 1, p. 21, t. 159; DC. Prodr. 1, p. 334.

Has. Near Sydney and Cook’s River, New South Wales.
27

106 PHANEROGAMIA.

3. COMESPERMA ERICINA, DC. 1. c.

Has. Near Sydney, &c., New South Wales.

This has more strict stems and smaller (often more spreading)
leaves than the foregoing; of which I suspect it is merely a variety.
The lower petal is about as much emarginate as in Labillardiere’s
figure of his C. virgata. The species require a careful revision.

4, COMESPERMA NUDIUSCULA, DC. 1. ¢.

Has. Hunter’s River and Woologong, New South Wales.

A slender species, with the leaves reduced to setaceous bracts, and
those mostly deciduous.

4. MURALTIA, Necker.

1. Murata Hetster1a, DC.

Muraltia Heisteria, DC. Prodr. 1, p. 334; Eckl. & Zeyh. Enum. Pl. 8. Afr. p. 24.
Heistera pungens, Berg. Descr. Pl. Cap. p. 185.
Polygala Heisteria, Linn. Spec. Pl. ed. 2, p. 989; Curt. Bot. Mag. t. 340.

Has. Cape of Good Hope, near Cape Town.

2. Murartia mixta, Linn. f.

Has. Cape of Good Hope; with the preceding.

3. MuRALtiA DIFFUSA, Burch.
Muraltia diffusa, Burchell, in DC. Prodr. 1. c. ; Eckl. & Zeyh. Enum. Pl. S. Afr. 1. ¢.

Has. Cape of Good Hope.

POLYGALACEA. 107

5. MONNINA, Ruiz & Pav.
1. MonninaA mAcrostacuya, Ruiz & Pav.

Var. pumila; racemis breviusculis; foliis caulinis inferioribus et radi-
calibus obovatis ; fructu puberulo.

Has. Peru: common below Obrajillo.

The same form was gathered near Lima and at Purruchuca by
Matthews. It appears to differ from J. herbacea, DC. only in having
a manifest wing to the fruit. Our specimens are small, 3 to 7 inches
high, from an annual root; the radical and some of the lower cauline
leaves broadly obovate ; and the racemes are only an inch or two in length.
The wing of the fruit is glabrous or nearly so, but its disk is tomentose-
puberulent.

2. MonnINA ANGUSTIFOLIA, DC.
Monnina angustifolia, DC. Prodr. 5, p. 840; Presl, Rel. Heenk. 2, p. 103.
Has. Peru ; on a mountain ridge, between Lima and Yanga.
Plant 2 or 3 feet high, with diffusely spreading and slender
branches. Fruit 2 or 3 lines in diameter, including the broad and

membranous wing, either glabrous or minutely tomentose, deeply
emarginate at both ends.—Perhaps not distinct from I. pterocarpa.

3. MONNINA LINEARIFOLIA, Ruiz & Pav.

Monnina linearifolia, Ruiz & Pav. Fl. Peruv. Syst. 1, p. 173; DC. 1. ¢.; Hook. &
Arn. Bot. Beech. p. 10, t. 6; Gay, Fl. Chil. 1, p. 240.

Has. Chili, near Valparaiso.

108 PHANEROGAMITIA.

6 SECURIDACA, Linn.
1, SECURIDACA OVALIFOLIA, St. Hil.

Securidaca ovalifolia, St. Hil. Fl. Bras. Mer. 2, p. 68; Walp. Repert. 1, p. 246.

Has. Near Rio Janeiro, Brazil. (In fruit.)

Orv! TRIGONIACE &.

1. TRIGONIA, Audi.

Triconta, Aubl. Pl. Guian. 1, p. 390, t. 149, 150; Lam. Ill. t. 847; H. B. K.
Nov. Gen. & Spec. 5, p. 141; St. Hil. & Moquin, in Mem. Mus. 18, t. 31; Endl.

Gen. 5659.
1. TRIGONIA NIVEA, Camb.

Trigonia nivea, Cambess. in St. Hil. Fl. Bras. 2, p. 113; Walp. Repert. 1, p. 248.

Has. Organ Mountains, in the vicinity of Rio Janeiro, Brazil.

(In flower only.)

2. TRIGONIA CROTONOIDES, Camb.

Trigonia crotonoides, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 115, f. 105.
Croton eriospermum, Lam. Dict. 2, p. 211.

Has. Organ Mountains and Rio Janeiro, Brazil, the var. (3. INCANA,
Camb., and var. y. OBLONGIFOLIA, Camb., l. c.: also a form with the

oblong leaves entirely glabrous.

Orv CARYOPHYLLACE-.
Susorv. Il SILEN E &.

1 DIANTHUS, Linn.
1. DIANTHUS PROLIFER, Linn.

Has. Madeira. Also near Sydney, New South Wales; un-
doubtedly introduced from Europe.

2. SILENE, Linn.
1. SILENE STRUTHIOLOIDES, Sp. Nov. (Tab. 10.)

S. fruticosa, ramosissima ; foliis subulatis marginibus involutis rigrdis
bifariis, ramulorum sterilium confertoambricatis demum recurvo-paten-
tibus ; cyma racemoso-paniculata ; calycibus tubulosis clavatis ; petalis
cuneatis obcordato-bilobis breviter biappendiculatis ; stipite ovario duplo
longiore.

Var. a. foliis rigidioribus cum ramis et inflorescentia glaberrimis.

Var. GB. foliis tenuioribus seepius longioribus, juventis preesertim ad
margines ramulisque pl. m. viscoso-pubescentibus.

Has. On Mouna Loa, and Mouna Kea, Hawaii, in the region of
Edwardsia (8,000 feet and higher); near the crater of Lua Pele; also
28

110 PHANEROGAMIA.

in the district of Waimea, in the same island. The two forms appa-
rently grow in the same stations.

Stems about a foot high, woody, branching, often much branched,
erect or spreading, in some specimens half an inch in diameter at the
base; the bark smooth and thin, brownish. Leaves crowded, even
umbricated on the sterile branches, but not exceeding the internodes on
the upper part of the flowering shoots, fwo-ranked (but only the al-
ternate pairs exactly superposed), subulate, pungently acute, canalicu-
late and with incurved margins, nerveless, or obscurely one-nerved,
rigid, persistent, their connate bases somewhat appressed, especially
on the sterile branches, above at length spreading or recurved, half an
inch or less in length, and more rigid in the first variety, in which
they are perfectly glabrous, as is the whole plant; in the var. 3. they
are less rigid, more spreading, and, at least when young, a little pubes-
cent with fine viscous hairs (as are the branches and inflorescence),
their margins especially pubescent-ciliate; the larger cauline ones
from an inch to an inch and a quarter long; those of the inflorescence
reduced to subulate bracts. Flowers rather few (5 to 12), racemi-
Jorm-paniculate, at first spreading or drooping, erect in fruit: pedun-
cles about 3 lines long, bibracteolate in the middle and one-flowered,
or sometimes two-flowered. Calyx half an inch long, tubular, soon
becoming strongly clavate, glabrous, greenish, often tinged with purple,
five-toothed; the teeth ovate, about a line in length. Petals with a
small cuneate and two-lobed or deeply obcordate lamina, exserted about
3 lines beyond the calyx, apparently greenish-white and turning
purplish; the dilated claw crowned at its summit with a small and
fleshy two-parted appendage. Stamens 10, included: the five filaments
opposed to the petals adnate to the base of their claws. Stipe 3 lines
long, about twice the length of the cylindraceous ovary, which is three-
celled at the base. Styles 3. Capsule partly exserted, ovoid,
six-valved at the apex, many-seeded. Seeds minutely reticulate-
roughened. Kmbryo curved about half way round.

I know of no species with which this remarkable, shrubby Silene
may be compared. It belongs to the section Stphonomorpha ; but it
is altogether peculiar in its aspect. The foliage somewhat resembles
that of some of the numerous Ratllardie. of the same district, or of a
Struthiola.

CARYOPHYLLACEZ&. 111

Piate 10, A.—SILENE STRUTHIOLOIDES, of the natural size. Fig. 1.
Vertical section of a flower. 2. A petal, with the stamen adnate to
its base. 3. Pistil, with its stipe. 4. Capsule, with the persistent
calyx, petals, &c. 5. A seed. 6. The same, vertically divided,
showing the embryo, &c.—All magnified.

2. SILENE LANCEOLATA, Sp. Nov. (Tab. 10.)

S. frutescens, ramosa, glabra; caulibus gracilibus erectis; folvis lanceo-
latis acutis uninerviis bast angustata ciliatis; cyma nuda corymbi-
formi; calycibus oblongo-campanulatis, fructiferis ovoideis; petals
cuneatis integris obsolete coronatis calyce dimidio longioribus ; stipite
brevi.

Has. On the mountains of Kauai, one of the Sandwich Islands.
Also on Maui (specimen without flowers or fruit).

Stems one or 2 feet high, branching, woody, but slender ; the branches
naked below, and annulate with the approximate scars of the fallen
leaves, very leafy above, glabrous. Leaves narrowly lanceolate, acute,
or acuminate, one-nerved, sessile, narrowed towards the base, where they
are ciliate, otherwise entirely glabrous, rather coriaceous in texture,
flat, spreading, 14 to 24 inches long, about 3 lines wide; the floral
ones reduced to small subulate and ciliate bracts. Flowers numerous,
rather crowded, in a naked and pedunculate corymbiform cyme. Pedi-
cels 2 or 3 lines long. Calyx 4 lines long, glabrous, oblong or tubular-
campanulate, in fruit becoming ovoid or obovate, five-toothed. Petals
“white,” nearly twice the length of the calyx, the cuneate or obovate
lamina entire or nearly so; the crown scale-like and minute, or almost
obsolete. Stamens 10: the five filaments opposed to the petals adnate
only by their very base to the claws. Styles 3. Ovary oblong,
three-celled at the base. Stipe very short. Capsule ovoid, not ex-
serted from the calyx, six-valved at the summit. Seeds pretty
strongly muricate. Embryo curved more than half way round the
albumen.—The foliage, although narrower and only one-nerved, and
the inflorescence, are somewhat like Saponaria officinalis.

Piate 10, B—SmENE LANCEOLATA; of the natural size. Fig. 1.

112 PHANEROGAMIA.:

Capsule, with the persistent calyx and the withered petals. 2. A
seed. 3. Vertical section of the same, showing the embryo and albu-
men.—All magnified.

3. SILENE INFLATA, Smith.

Has. On rocks, east of Funchal; island of Madeira.

4, SILENE QUINQUEVULNERA, Linn.

Has. Valparaiso. Bay of Islands, New Zealand. Sydney, New
South Wales.—A South European species (including & Gallica, 8.
Anglica, and S. cerastoides, Linn.), now widely distributed and natu-
ralized on the coasts, in many of the subtropical and warmer tempe-
rate parts of the world. It has not been previously noticed in New
Zealand, where it is doubtless of very recent and probably local
introduction.

5. SILENE ANTIRRHINA, Lino.
Has. Rio Negro, Northern Patagonia; on sand-hills.
The specimen is entirely in fruit, and not in a condition for critical
comparison with the plant of the United States; but, except that the

pods are rather large, it appears to belong to this species; which,
moreover, St. Hilaire gathered at Maldonado (FI. Bras. Mer. 2, p. 164).

38. GYPSOPHILA, Linn.
1. GYPSOPHILA AUSTRALIS.
G. annua; caulibus paniculato-ramosis erectis fere e basi usque ad

‘apicem floriferis minutissime puberulis ; foliis lineartbus angustissimis
obtusiusculis, ramealibus setaceis; pedicellis jfiliformibus patentibus

CARYOPHYLLACE 4&4. 113

cum calyce turbinato glanduloso-pubescentibus ; petalis linearibus apice
emarginato-bilobis valvulisque capsule calycem paullo excedentibus ;
seminibus reticulato-rugulosis.

Dichoglottis australis, Schlecht. in Linnea, 20, p. 631.
Has. In the vicinity of Puen Buen, New South Wales.

Stems 6 to 12 inches high, from an annual root, branching in a
paniculate manner, erect, very minutely puberulent, a little glandular
towards the summit, flowering from near the base to the apex: the
pedicels nearly all becoming lateral. Leaves smooth, rather obtuse ;
the lower cauline ones narrowly linear, 6 to 8 lines long; the upper
smaller and becoming filiform o7 setaceous. Pedicels filiform, solitary,
6 to 10 lines long, spreading in fruit, pubescent with glandular-tipped
spreading hairs, which are rather deciduous. Calyx turbinate, a little
larger than in G. muralis, five-lobed at the summit (the lobes or teeth
ovate and very obtuse), glandular-pubescent. Petals perhaps a quarter
longer than the calyx, narrowly cuneate-linear (but the flowers are
all late ones: probably they were broader and more conspicuous in
the earlier blossoms), notched at the apex. Stamens 10. Styles 2,
short. Capsule obovoid, obtuse, four-valved; the valves slightly
exceeding the calyx. Seeds numerous, reticulated-rugose with oblong
areole much as in Jaubert and Spach’s figure of G. (Dichoglottis)
tubulosa, except that towards the micropyle the areolations are rather
longitudinal than transverse.

Doubtless this is the species described by Schlechtendal, and it is
the only known species of the genus from that part of the world.
But his plant was found in the southeastern part of Australia, and in
some minor points it does not altogether accord with ours, as de-
scribed above. It is a coarser plant than G. muralis, with more strict
and prolonged stems, and lateral flowers.

4. MELANDRIUM, BRedhi.

MeELANDRIUM, Reehling, Deutschl. Fl. ed. 1, p. 254; Fenzl, in Ledeb. Fl. Ross. 1,
p. 326; Endl. Gen. Suppl. 2, p. 78.
29

114 PHANEROGAMIA.

Lycunis, subgen. MeLAnprium, Braun, in Flora, 1889, 1, p. 319; Koch. FI.
Germ. & Helv. ed. 2, 1, p. 116.

1. MELANDRIUM THYSANODES.

M. caule adscendente uni-paucifloro; foliis lanceolatis radicalibusve
spathulatis marginibus ciliato-tomentosis, superioribus calyce pedun-
culisque glanduloso-pubescentibus ; calyce latissime campanulato ad
medium quinquefido petalis fere nudis brevioribus ; unguibus filamen-
tisque basi tomentosis in annulum carnosum cupulatum ab gynophoro
brevi discretum coalitis ; seminibus levibus margine lato inflato alatis.

Silene thysanodes, Fenzl, in Endl. Stirp. Nov. Dee. 4, p. 31.
Lychnis thysanodes, Hook. f. Fl. Antare. p. 246.

Has. Above Bajios and Alpamarca, in the high Andes of Peru.

Plant ceespitose, from a perennial root. Stems in our specimens all
simple and one-flowered, 3 to 6 inches high, ascending, glandular-
pubescent, especially near the summit. Leaves rather fleshy, lanceo-
late, or the radical ones often spatulate, not unlike those of J. apetalum,
except that they are shorter, more or less pubescent, or the upper
ones (like the peduncle and calyx) glandular-pubescent, all tomentose-
ciliate, especially when young. Calyx 4 lines long, broadly campanu-
late, or in fruit broadly turbinate or hemispherical, scarcely at all
striate or nerved, five-cleft to near the middle; the lobes ovate, obtuse,
strongly tomentose-ciliate. Petals about one-third longer than the
calyx; the limb short, obovate or cuneate, purplish; the crown
reduced to two minute appendages, or obsolete. Stamens 10, the five
filaments opposed to the petals adnate with the base of their claws,
all woolly-pubescent at the very base, and inserted into the edge of a cup-
shaped or cylindraceous fleshy disk, which encircles the short stipe of
the ovary, but is free from it. Styles 5. Ovary cylindraceous, one-
celled. Capsule about the length of the calyx, five-valved ; the valves
two-cleft. Seeds numerous, quadrate-reniform, with a smooth testa,
which is extended beyond the nucleus into a wide and thickish (probably
inflated) wing. Embryo coiled into almost a complete ring.

This is without much doubt the species of the Colombian Andes,

CARYOPHYLLACESA 115

described by Fenzl from specimens gathered by Mr. Hall, and by Dr.
Hooker from those gathered by Jameson. Our specimens accord with
the character given by the latter, except that they are all one-
flowered: but Fenzl, who referred his plant to Silene, describes it as
trigynous, and with the ovary three-celled, at least towards the base;
whereas ours has five styles, and the ovary is one-celled throughout.
The pod and seeds have not been previously described; nor were the
hairy bases of the filaments and of the claws of the petals noticed.
A trace of this pubescence may be observed in M. Magellanicum (or at
least in Lychnis Chilensis, Gay, Fl. Chil., which appears to be no more
than a variety of that species); but it is wholly wanting in the Arctic
M. apetalum.

Scusorv. IL. ALSINE &.

5. ARENARTIA, Linn. (excl. sp.), Fenal.
1. ARENARIA DIFFUSA, Zl.

Arenaria diffusa, Ell. Sk. Bot. Car. & Georg. 1, p. 519 (1821); Gray, Pl. Wright,
2, p. 18. |

A. nemorosa, H. B. K. Nov. Gen. & Spec. 6, p. 35 (1822); DC. Prodr. 1, p. 408.

Spergulastrum lanuginosum, Michx. Fl. Bor.-Am. 1, 275; DC. Prodr. 1, p. 421.

Micropetalon lanuginosum, Pers. Ench. Pl. 1, p. 509.

Stellaria elongata, Nutt. Gen. Pl. N. Amer. 1, p. 289; DC. Prodr. 1, p. 399.

S. lanuginosa, Torr. & Gray, Fl. N. Amer. 1, p. 187.

Has. Bafios, Obrajillo, and Culnai; Andes of Peru.

This plant, likewise gathered by Matthews in Peru, is common in
the Andes of Colombia, and extends to Jamaica, Mexico, and thence
into the Southern United States. For a critical comparison brings to
view no characters to distinguish the Peruvian and Quitensian plant
from the Arenaria diffusa of Elliott; which name I retain, being
older than that of Kunth, and more appropriate than that of Michaux,
under Spergulastrum. It is a true Avenaria, as the genus is limited
by Fenzl; the petals, when present, being entire. In specimens from

116 PHANEROGAMIA.,

Florida and New Mexico, the petals are sometimes as long as the
calyx. Kunth describes them as shorter than the calyx. In our
Peruvian specimens they are fully as long as the sepals. The pubes-
cence is variable, sometimes almost wanting. The seeds are lenti-
cular, black, very smooth and shining.

2. ARENARIA SCopULORUM, H. B. K.

Arenaria scopulorum, H. B. K. Nov. Gen. & Spec. 6, p. 831; DC. Prodr. 1, p. 411.

Has. Above Casa Cancha, in the high Andes of Peru, a little below
the snow-line.

3. ARENARIA SERPENS, H. B.K.
Arenaria serpens, H. B. K. Nov. Gen. & Spec. 6, p. 82; DC. Prodr. 1, p. 412.

Has. Bafios, Andes of Peru. (Also near Cerro Pasco, Matthews.
Andes of Quito, Jameson.)

4. ARENARIA ANDICOLA, Gill.

Arenaria andicola, Gillies, in Hook. & Arn. Bot. Misc. 3, p. 148.
A, serpylloides, Gay, Fl. Chil. 1, p. 271?

Has. High Andes of Chili, above Santiago, near the snow-line.

“Very closely allied to A. serpens, H. B. K.,” as Hooker and-Ar-
nott remark; and I suspect no more than a variety of that species.

5. ARENARIA ALPAMARCH, Sp. Nov.

A. cespitosa, nana, glabra; foliis confertis imbricatisve ovato-lanceolatis
carinatis cuspidatis rigidis basi parum ciliolatis connatis ; pedunculo
terminal unifloro foliis floreque paullo longiore; sepalis foliis simili-
bus capsulam oligospermam sex-valvem subsuperantibus ; petalis nullis ;
staminibus 5,

CARYOPHYLLACE &. 117

Has. Andes of Peru, at Alpamarca, a little below the snow-line;
also above Obrajillo and Banos. (High Andes of Peru, Matthews, in
herb. Hook.) -

Plant dwarf, ccespitose, in pulvinate, perennial tufts, of one or two
inches in height, entirely glabrous. Leaves much crowded, and more
or less imbricated, but somewhat spreading, ovate-lanceolate or oblong-
lanceolate, about 2 or 3 lines long, broadest at the connate base, coria-
ceous and rather rigid, tipped with a sharp cusp or mucronation,
strongly carinate with a thickened midnerve, and more or less two-
grooved on each side of it, the margins thickened, smooth, or very
obscurely ciliate near the base, persistent. Peduncle terminal, or by in-
novation lateral, not bracteolate, one or 2 lines, or in fruit 3 to 5 lines
long, one-flowered. Sepals coriaceous, ovate-lanceolate, nearly like the
leaves, 14 to 2 lines long, usually 5, in some flowers 4. Petals none,
or not detected. Stamens 5, or 4 in the tetramerous flowers, opposite
the sepals and shorter than they are. Capsule rather shorter than the
calyx, six-valved (or rarely four-valved) to near the base. Ovules
about 20. Seeds 5 or 6, globular, smooth.

This species is probably to be referred to Fenzl’s section Dicranilla,
although the ovules are rather numerous, and the leaves are not
closely imbricated, as in the following. The specimens are mostly in
fruit: but, although the stamens remain, I find no trace of petals.

6. ARENARIA DICRANOIDES, H. B. Kx.

Arenaria dicranoides, H. B. K. Nov. Gen. & Spee. 6, p. 834; DC. Prodr. 1, p. 413.
A. bryoides, Walp. Rel. Meyen. p. 802; an Willd.?
Lobelia bryoides, Willd. Herb., ex Schlecht. in Roem. & Schult. Syst. Veg. 5, p. 41.

Has. Above Casa Cancha, Culnai, and Alpamarca, in the high
Andes of Peru, near the snow-line.

The plant has much the aspect of Dicranum glaucum, as 1s men-
tioned by Kunth, and grows in similar cushion-like tufts. ‘The leaves
vary considerably in shape and size, even on different parts of the
same plant: those of the older stems are mostly triangular-lanceolate

and more or less acute; those of the younger branches rather ovate
30

118 PHANEROGAMIA.

and obtuse. Our larger specimens pretty well accord with those
gathered by Mr. Matthews (No. 688 in herb. Hook.), near Cerro
Pasco, which are doubtless the A. dicranoides of Kunth. Those from
Alpamarca, &c., have smaller and blunter leaves, and might be
referred to A. bryoides, Willd. (which, I notice, Walpers, who had
original specimens before him, has done with a Peruvian plant
gathered by Meyen); but I cannot draw any line between the diffe-
rent forms. Mr. Pentland collected the same species, apparently,
without flowers, at La Paz in the Bolivian Andes. The few flowers
the specimens furnish are not in good condition. They are sessile at
the apex of the branches, apetalous, and with 10 stamens. The styles
are 2, or sometimes 8. The ovary contains 5 or 6 ovules, of which
only one, perhaps, matures.—In Kunth’s detailed description for, “folia
13~2-pollicaria,” we should evidently read, 13—2 lineas longa.

6. STELLARIA, Linn.
1. STELLARIA DEBILIS, D’ Urv.

Stellaria debilis, D’Ury. in Mem. Soc. Linn. Par. 4, p. 618 ; Hook. f, Fl. Antare.
p. 250.

Has. Orange Harbour, Fuegia.

Besides the ordinary form of the species, there is a variety, appa-
rently growing in very wet places, which is more czespitose, depressed,
and with longer, crowded, narrowly linear, and gramineous leaves.

2. STELLARIA CUSPIDATA, Willd.

Stellaria cuspidata, Willd. herb., ex Schlecht. Berl. Mag. 1816; H. B. K. Nov.
Gen. & Spec. 6, p. 27; DC. Prodr. 1, p. 896; Gay, Fl. Chil. 1, p. 264.

Has. Obrajillo, Andes of Peru.

This species resembles the European S. nemorum; but the leaves
are less petioled, all the upper ones sessile or nearly so. It appears

CARYOPHYLLACEA, 119 ,
to be rather common from the Quitensian Andes to Valparaiso. Stel-

laria leptopetala, Benth. Pl. Hartw. p. 163, is probably a smaller form
of the same species.

3. STELLARIA MEDIA, Linn.

Has. Sydney, New South Wales; and everywhere, in places inha-
bited by man.

4. STELLARIA C&SPITOSA, Hook. f.

Stellaria ccespitosa, Hook. f. in Hook. Jour. Bot. 2, p. 411?

Has. Puen Buen, New South Wales.

The specimens consist of the upper part only of a tuft of floriferous
stems, in an advanced state, much more developed and _ prolonged
than in Tasmannian specimens of S. ccespitosa; of which it seems
likely to be merely a many-flowered variety.

5. STELLARIA PUNGENS, Brongn.

Stellaria pungens, Brongn. Bot. Voy. Duperr. Coq. t. 78; Hook. f. in Jour. Bot. 1. ¢.
Stellaria squarrosa, Hook. Jour. Bot. 1, p. 250.

Has. Puen Buen, New South Wales.

7% CERASTIUM, Linn.
1. CERASTIUM ARVENSE, Linn., &

Var. Furcianum: humile; foliis imbricatis ; pedicellis solitariis termi-
nalibus; floribus amplis. Hook. f.

Has. Orange Harbour, Fuegia.

120 PHANEROGAMIA.

Both the ordinary form of the species, and some very dwarf ones
occur in the collection. The latter are only one or two inches high,
with crowded or even wmbricated leaves, and solitary, short-peduncled
flowers : they accord with those referred to this species by Dr. Hooker,
as his variety Fuegianum.

2. CERASTIUM VULGATUM, Linn. (ex Fries).

Var. PeRuviANUM: caulibus superne spedicellisque glanduloso-pubes-
centibus; folus plerumque linearibus; petalis nunc calyce glanduloso
paullo longioribus nune plane nullis !

Var. ANDINUM: nanum, condensatum, vix glandulosum ; foliis oblongo-
linearibus ; petalis calyce paullo brevioribus.

Has. Bajios, Andes of Peru (var. Peruvianum): also gathered near
Quito by Prof. Jameson. Above Obrajillo and Alpamarca, on the high
Andes of Peru (var. Andinum).

The first variety well accords with my specimens of Cerastiwm
Beeringianum, Cham. & Schlecht. (which Fenzl., in Ledebour’s Flora
Rossica, reduces to a form of C. vulgatum); except that the flowers
are rather more glomerate, and most of the later ones are wholly
apetalous, as in a corresponding variety of C. viscosum. The upper
part of the slender stem and the pedicels are strongly pubescent
with spreading glandular hairs, while the calyx is beset with sessile
glands. The cauline leaves are all linear, from 6 to 10 lines long, and
only a line wide. The capsule, seeds, &c., are just as in C. rulgatum.
—The var. Andinum is, I suppose, a more dwarf and condensed form
of the same species, growing at a greater elevation. It is smaller in
all its parts and less glandular; the leaves oblong-linear. I think that
both belong to the same species, and that they cannot be satisfactorily
distinguished from C. vulgatum, as that species is now received by
Prof. Fenzl.

3. CERASTIUM viscosuM, Linn. (ex Fries).

Cerastium viscosum, Linn. Spec. ed. 1, p. 437; Curt. Fl. Lond. 2, t. 34; Fries,
Novit. Fl. Suec. p. 128; Fenzl, in Ledeb. Fl. Ross. 1, p. 404.

CARYOPHYLLACEA. 121

C. vulgatum, Linn. Herb. & Fl. Suec. (excl. syn.); Smith, Fl. Brit. 2, p. 496; DC.
Prodr. 1, p. 415.

C. glomeratum, Thuill. Fl. Par. p. 225; Mert. & Koch, Deutch. Fl.; Coss. & Germ.
Fl. Env. Par. & Atlas, t. 4, f. b.

Has. Organ Mountains, near Rio Janeiro. Waiaruru Bay, New
Zealand. Doubtless introduced from Europe.

Susorv. IIT. TILLECEBRE &.
8 SPERGULARIA, BPrs.
1. SPERGULARIA RAMOSA, Camb.

Spergularia ramosa, Cambes. in St. Hil. Fl. Bras. Mer. 2, p. 178.

Has. Rio Negro, Northern Patagonia.

The flowers and pods are as large as those of the allied Spergularia
grandis, of which it may be only a variety. It has a long and thick-
ened, somewhat woody root. ,

2. SPERGULARIA RUBRA, Presi.

Has. Woolongong, New South Wales; the var. marina (Arenaria

marina, Smith). Valparaiso, Chili: a slender, small-flowered variety ;

apparently the same as the Arenaria floribunda, of Gay’s Flora Chi-
lena.

9. SPERGULA, Linn. (excl. sp.)
1. SPERGULA ARVENSIS, Lani.

Has. Hunter’s River, New South Wales. St. Helena.—Doubtless

introduced from Europe.
31

122 PHANEROGAMIA.

10. POLYCARP AA, Lam.
1, PoLycARP#A TENUIFOLIA, DC.

Achyranthes tenutfolia, Willd. Spec. 1, p. 1196.

Mollia tenuifolia, Willd. Hort. Berol. no. 11.

Polycarpeea tenuifolia, DC. Prodr. 3, p. 374.

P. glabrifolia, 8. rutila, Fenzl, in Pl. Kotsch. It. Nub. no. 88.

Has. Arid places, St. Jago, Cape de Verde Islands.

This is not enumerated in Mr. Webb’s Spicilegia Gorgonea, nor in
the Appendix subsequently published, in Hooker’s Journal of Botany.
The specimens agree with Kotschy’s plant, above cited. The linear-
subulate leaves are too narrow to accord with the character of DeCan-
dolle’s P. glabrifolia ; but the stipules are pretty large, in which point
alone it differs from the character of Achyranthes tenuifolia, Willd.

ll. POLYCARPON, Leff.
1. PoLYCARPON TETRAPHYLLUM, Linn.

Has. Madeira. Bay of Islands, New Zealand. Near Sydney and
Puen Buen, New South Wales. Doubtless introduced from Europe.

This species is not before noticed as growing in New Zealand,
where many European weeds are rapidly becoming naturalized; nor
in New Holland, which, however, DeCandolle gives as a habitat of P.
alsinefolvum.

12. DRYMARIA, Willd.

1. Drymaria corpata, Willd.

Drymaria cordata, Willd. Herb.; Rem. & Schult. Syst. 5, p. 406; H. B. K. Nov.
Gen. & Spec. 6, p. 23; DC. Prodr. 1, p. 395.

CARYOPHYLLACE 4S. 123

od

Has. Organ Mountains, and Rio Janeiro, Brazil. Luzon, Philip-
pine Islands; in mountains near Manilla: doubtless introduced,

2. DRYMARIA ROTUNDIFOLIA, Sp. Nov.

D. glaberrima; caule erecto gracili; foliis. orbiculatis vel subrenifor-
mibus sessilibus ; stipulis obsoletis; cyma corymbosa pluriflora ; pedi-
cellis (fructiferis erectis) alaribus flore vix duplo longioribus ; sepalis
ovalibus obtusissimis petalis paullo brevioribus,

Has. Obrajillo, Andes of Peru; common.

Stems 4 to 8 inches, according to Dr. Pickering’s notes sometimes 2
feet high, from an annual root, slender, entirely glabrous, erect, more or
less branched. Leaves roundish; the upper somewhat reniform or
subcordate and broader than long, closely sessile, sometimes mucronate-
acuminate; the lower verging to obovate and more or less contracted
at the base, but sessile or nearly so; all glabrous; the cauline ones 3
lines long, by 5 or 6 wide, those of the branches smaller. Stipules
obsolete, or of one or two very minute setz on each side, deciduous.
Cymes corymbose, terminating the stem and branches, on filiform gla-
brous peduncles 6 to 10, or rarely 18 lines long, 7-15-flowered; the
small flowers rather crowded. edicels slender, glabrous, or very
minutely glandular-puberulent, all erect both in flower and fruit; the
central ones 2 to 4 lines long; the others only one or two lines long.
Bracts ovate-lanceolate, entirely scarious and glabrous, minute. Calyx
entirely glabrous, a line and a half long; the sepals oval, very obtuse,
somewhat carinate at the base, thin, inconspicuously three-nerved,
the margins narrowly scarious. Petals about one-third longer than the
calyx, narrowly cuneiform, white, two-cleft to a little below the middle ;
the lobes narrowly spatulate; the claw entire. Stamens 5, half the
length of the petals. Style three-cleft to the middle. Capsule ovoid,
chartaceous, slightly longer than the calyx, three-valved. Seeds 9 or
10, reniform ; the testa muricate with rows of small tubercles.

This species is nearly allied to Drymaria glaberrima, Bartl. in
Presl, Rel. Henk. (which the late Mr. Matthews collected at Hua-
mantanga); but that has acute or pointed leaves, and acute sepals,
while in our plant they are remarkably obtuse. :

124 PHANEROGAMIA.

3. DRYMARIA VISCIDULA, Sp. Nov.

D. caule diffuso glabro; ramulis cum inflorescentia glanduloso-pubes-
centibus ; foliis subreniformi-ovatis in petiolum contractis glabriusculis ;
stipulis setaceis; cyma corymbosa pluriflora; pedicellis (fructiferis
erectis) alaribus flore duplo longioribus ; sepalis oblongis acutis glan-
dulosis petala subceequantibus.

Has. Andes of Peru, with the preceding.

Stems diffuse, a foot long, from an annual root, slender, glabrous,
except the very summit. Leaves 3 or 4 lines long, by 4 or 5 in width,
dilated rownd-ovate or somewhat reniform, glabrous or slightly pubes-
cent, abruptly contracted into short and margined, mostly hairy petioles,
or the lowest cauline with a petiole as long as the blade. Stipules of
2 or 3 capillary interpetiolar sete on each side, often as long as the
petioles, tardily deciduous. 2¢duncles half an inch or an inch long,
erect, clothed with a short glandular and glutinous pubescence, bearing
a cyme of 5 to 15 rather crowded flowers. Bracts setaceous-subulate,
glandular. Pedicels glandular-pubescent and viscous; the central ones
3 or 4 lines long; the lateral only one or 2 lines long; all erect in
flower and fruit. Calyx a line and a half long; the sepals ovate-oblong,
or oblong-lanceolate, with conspicuously scarious-petaloid margins,
acute, or the narrowed point rather obtuse, three-nerved, glandular
externally. tals scarcely longer than the calyx, white, narrowly
cuneate, two-cleft to the middle; the lobes spatulate. Stamens 4,
nearly as long as the petals. Style three-cleft above the middle.
Capsule as long as the calyx, 12-15-seeded. Seeds obscurely tuber-
culate-roughened.

This is at once distinguished from the preceding species by its glan-
dular and viscid peduncles and calyx, its pointed sepals, its smaller
and shorter petals, and its petiolate leaves. It can hardly be a
smaller form of D. divaricata, H. B. K., which is said to have glabrous
peduncles, but viscid-pubescent pedicels; for the flowers are consi-
derably smaller than those of Holosteum umbellatum, the calyx itself
is glandular, and the peduncles and pedicels are not at all divaricate
nor reflexed.

CARYOPHYLLACES. 125

4. DRYMARIA FASCICULATA, Sp. Nov.

D. glaberrima ; caule erecto ; foliis sessilibus ovato-oblongis subtrinerviis ;
pedunculo stricto apice minute glanduloso; floribus subsessilibus sca-
rioso-bracteatis in fasciculum arcte congestis; sepalis glaberrimis ob-
longo-lanceolatis trinerviis cuspidato-acuminatis margine scariosis petala
subcequantibus.

Has. Andes of Peru, near Obrajillo.

Root annual. Stem a span high, simple or sparingly branched,
slender, erect, glabrous, as is the whole plant. Leaves only 3 or 4
pairs, much shorter than the internodes, one-nerved and with a pair
of inconspicuous lateral nerves, sessile, 5 to 9 lines long; the lower
obovate or spatulate; the upper and those of the branches oblong-ovate,
and acute or acuminate. Stipules very minute, setaceous. Peduncle
erect and strict, from half an inch to 2 inches long, naked, towards the
summit minutely viscous-glandular. Flowers crowded in a dense fascicle,
subtended by oblong-ovate, one-nerved, and cuspidate-pointed, scarious
bracts, which are little shorter than the calyx: the primary or central
flower on a pedicel of barely a line in length, and much shorter than
the calyx; the others nearly or quite sessile. Calyx 3 lines long; the
sepals perfectly glabrous, lanceolate-oblong, acutely pointed, thin, with
rather broad scarious margins, conspicuously three-nerved near the
middle. Petals 5, slightly exceeding the calyx, white, oblong-linear,
two-lobed above the middle, the lobes linear-oblong; the dilated claw
entire. Stamens 5, rather shorter than the petals. Style slender,
three-cleft above the middle; the recurved branches stigmatose.
Capsule shorter than the calyx, membranaceous, three-valved, about
twelve-seeded. Seeds circinnate, as in the genus, scabrous with very
minute points arranged in lines.

The compact fascicles of flowers, imbricated with large scarious
bracts, abundantly distinguish this from any described species of Dry-
maria, nor is there any one with which I can particularly compare it,
excepting the following; which makes some approach to it in the in-

florescence.
32

126 PHANEROGAMIA.

5. DRYMARIA MACRANTHA, Sp. Nov.

D. glabella, multicaulis ; caule laxe ramoso; foliis cordato-ovatis sessilt-
bus 3—5-nerviis crassiusculis; bracteis subulatis; eyma multiflora
conferta ; pedicellis centralibus jlore cequilongis ; sepalis ovato-oblongis
late petaloideo-marginatis ; petalis nullis.

Has. With the preceding.

Root and base of the stem not seen. Stems said to be numerous and
upright, a foot high: they are rather diffusely branched in the speci-
mens, with approximate and tumid nodes, nearly glabrous, or the
branches, like the inflorescence, puberulent. Leaves 5 to 7 lines long,
broadly ovate and somewhat cordate, acute, or the upper pointed, sessile,
glabrous or nearly so, 3—5-nerved, the larger ones five-nerved, of a rather .
thick and fleshy texture. Peduncles an inch or less in length. Cyme
10-20-flowered, rather close and fasciculate. Bracts narrowly subulate,
greenish, with narrow scarious margins below, slightly puberulent, 2
or 3 lines long. Pedieels puberulent: the primary central ones 3 to
6 lines long, usually about the length of the calyx. Sepals ovate-oblong,
4 or 5 lines long, minutely glandular-puberulent externally, slightly
three-nerved, acute, except for the broad and petaloid hyaline margins,
which are broadest near the summit. Petals none. Stamens 5. Style
short, 2-3-cleft. Capsule coriaceous, ovoid, shorter than the calyx,
three-valved, many-seeded. Seeds obliquely obovate, minutely sca-
brous. Embryo uncinate-incurved.

This is the largest-flowered species known. The flowers are twice
the size of those of the Chilian D. apetala, and a third larger than
those of D. grandiflora, Bartl.; which, moreover, is described as
having very long peduncles, and pedicels many times longer than the
calyx, the sepals obscurely one-nerved, and the petals conspicuous.

18. CORRIGIOLA, Linn.

1. CorRIGIOLA SQUAMOSA, Hook. & Arn.

Corrigiola squamosa, Hook. & Arn. Bot. Mise. 3, p. 3837; Gay, Fl. Chil. 2, p. 518.

CARYOPHYLLACE &. 127

Has. Valparaiso, Chili; abundant in sand on the sea-shore.

14. PARONYCHIA, Juss.

1. PARONYCHIA ILLECEBROIDES, Webd.

Paronychia illecebroides, Webb, Spicil. Gorg. in Hook. Niger Flora, p. 106, t. 7;
Hook. Ic. Pl. t. 756.
Herniaria illecebroides, Chr. Smith, in Tuck. Voy. p. 250, ex. Herb. Mus. Brit.

Has. St. Jago, Cape Verde Islands; on rocks near the coast.

2, ParonycutaA CuHILEeNsis, DC.

Paronychia Chilensis, DC. Prodr. 3, p. 8370; Hook. & Arn. Bot. Misc. 38, p. 337;
Gay, Fl. Chil. 2, p. 521.

Has. Near Valparaiso.

3. PARONYCHIA COMMUNIS, Camb,

Paronychia communis, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 186.
Has. Rio Negro, North Patagonia; in sandy soil.

This species is nearly allied to Puronychia Chilensis (from which
P. Coquimbensis, Gay, Fl. Chil., seems hardly distinct); and, indeed,
I can only distinguish it by its shorter leaves (only three or four lines
long) and minuter pubescence, The suffrutescent stems spread, in
great numbers, from the crown of the long and thick, perpendicular
root. The tips of the sepals are either obscurely or manifestly mucro-
nate-pointed.

4, PARoNnyCHIA BoNnARIENSIS, DC,

Paronychia Bonariensis, DC. Prodr. 8, p. 8370? Presl. Rel. Henk. 2, p. 6; Walp.
Rel. Meyen. p. 301.
P. Brasiliana, DC. in Lam. Dict. 5, p. 23? St. Hil. Fl. Bras. Mer. 2, p. 186?

’

Has. Bajos, Andes of Peru.

128 PHANEROGAMIA.

It would seem unlikely that an Andine Peruvian species should be
the same as one from Monte Video. But our specimen accords with
the character of Puronychia Bonariensis, having a smooth calyx, the
lacinie of which are tipped with slender and straight awns, &c.
Moreover, the Peruvian plant of Meyen’s collection, here referred by
Walpers, appears to have been compared with a specimen commu-
nicated by Poiret to the Willdenovian herbarium.

5. PAronycuiA ANDINA, Sp. Nov.

P. glabella; caulibus multicipitibus e radice lignescente depressis ccespi-
toso-confertis; folvis coriaceis confertissimis ovato-oblongis enerviis
cymbeformibus aristulatis stipulas vix superantibus; floribus bracteis
stipularibus occultatis capitato-congestis ; calyce extus puberulo, laciniis
oblongis late scariosis sub apice cucullato aristulatis.

Haz. Bafios; Casa Cancha to Culnai and Alpamarca, in the high
Andes of Peru. (Also gathered in the same region by Mr. M’Lean,
and at Cerro Pasco by Mr. Matthews.)

Root perennial and somewhat woody, perpendicular, 6 to 10 inches
long; the crown dividing into numerous, depressed, much crowded, and
tufted, branching stems, which are 2 or 3 inches long, a little woody
at the base, nearly glabrous, clothed with the oblong-lanceolate sca-
rious stipules, which are imbricate-crowded and much longer than the
internodes. Leaves thickish and coriaceous, much crowded, or almost
imbricated (the older ones more or less spreading), opposite, closely
sessile, scarcely longer than the conspicuous stipules, a line or a line and
a half long, ovate-oblong, more or less convex below and concave
above, so as to appear boat-shaped, or else involute, destitute of a cari-
nate midrib and of veins, minutely puberulent, or nearly glabrous,
somewhat ciliolate, conspicuously awn-pointed. Flowers subsessile,
covered by ovate and more or less connate, stipular, silvery-scarious bracts,
which exceed and hide the floral leaves, forming a capitate-crowded
inflorescence at the summit of the branches. Calyx thin, very
minutely puberulent externally under a lens, deeply five-cleft; the
lobes oblong, equal, the greater part scarious, the cucullate apex obtuse;
the midrib just below the apex produced into a very short but distinct

CARYOPHYLLACES®. 129

awn, which is hardly a quarter of the length of the lobe. Petals, or
staminodia, 5, setaceous, shorter than the sepals, a little longer than

the 5 stamens. Style two-parted; the lobes slender. Ovary, seed,
&c., as in the genus.

This appears to be a common species in that part of the Peruvian
Andes which was visited by the Naturalists of the Expedition.
It extends from Bafios, which is only 10,000 feet above the sea-level,
to the limit of vegetation above Alpamarca. It varies somewhat in
form; the specimens from the higher regions being more condensed,
and exhibiting almost the aspect of a Colobanthus. I do not find that
it has before been described, nor do I know of any species with which
it may be particularly compared.

6. PARONYCHIA RAMOSISSIMA, DC,

Paronychia (Acanthonychia) ramosissima, DC. Prodr. 3, p. 8372, & Mem. Paronye.
t. 4; St. Hil. Fl. Bras. 2, p. 188; Torr. & Gray, Fl. N. Amer. 1, p. 172.

Pentacena polycnemoides, Bartl. in Presl. Rel. Heenk. 2, p. 5, t. 49; Walp. Repert.
1, p. 261.

P. ramosissima, Hook. & Arn. Bot. Mise. 3, p. 338; Gay, Fl. Chil. 2, p. 528.
Has. Valparaiso, Chili. Rio Negro, North Patagonia.

The inequality of the sepals, unsupported by other characters,
hardly entitles this plant to the rank of a genus. The leaves are not

alternate, as is said by Fenzl (in Endlicher’s Genera), but uniformly
opposite.

Susorv, IV. SCLERANTHE &.
15s. MNIARUM, Forst.

1. MNrAruM BIFLORUM, Forst.

Mniarum biflorum, Forst. Char. Gen. 1, t. 1; R. Br. Prodr. Nov. Holl. p. 413;
' DO. Prodr. 8, p. 878; Raoul, Enum. Pl. Nov. Zel. p. 43.
33

130 PHANEROGAMIA.,

M. pedunculatum, Labill. Fl. Nov. Holl. 1, p. 8, t. 2.
Ditoca muscosa, Gertn. Fruct. & Sem. 2, p. 196, t. 126, f. 1.

Has. Waia-ruru Bay, New Zealand. (Without flowers or fruit.)

Susorpv. Vv. MOLLUGINE &.

16. MOLLUGO, Linn.
1. MOoLLUGO VERTICILLATA, Linn.

Mollugo verticillata, Linn. Spec. 1, p. 89; Torr. & Gray, Fl. N. Amer. 1, p. 176; St.
Hil. Fl. Bras. Mer. 2, p. 170; Fenzl, Mollug. in Ann. Wien. Mus. 2, p. 376.

M. dichotoma, Schrank, Pl. Rar. Hort. Monac. t. 64 (forma angustif.).

M. Schrankii & M. Berteriana, Seringe, in DC. Prodr. 1, p. 391.

Has. Rio Janeiro, Brazil, on the sea-beach; only the narrow-leaved
variety ((. linearis).

This species is found from Canada to Southern Brazil, but is seldom
here met with far from the habitations of man. Whether it had
so wide an original range is therefore doubtful. In the United States,
it always appears like an introduced plant.

2. MoLLUGO NUDICAULIS, Lam.
Mollugo nudicaulis, Lam. Dict. 4, p. 254; Seringe, in DC. Prodr. 1, p. 391; Fenzl,
Mollug. 1. c. p. 882.
M. bellidifolia, Seringe, in DC. 1. c.; Webb, Spicil. Gorg. in Hook. Nig. Fl. p. 104.

Has. St. Jago, Cape de Verde Islands; only the variety bellidifolia.

3. Mouuueo stricta, Linn.

Mollugo stricta, Linn. 1. c.; DC. Prodr. 1, p. 391; Fenzl, Mollug. 1. ¢. p. 380.
M. triphylla, Lour. Fl. Cochin. ed. 2, p. 79; DC. 1. ¢. p. 392.

CARYOPHYLLACE &. 131

Has. Feejee Islands; common ‘in cultivated grounds, at Ovolau,
Muthuata, and Sandal-wood Bay.

The specimens are intermediate between the original J. stricta of
Linnzus and the I. triphylla of Loureiro.

17. COLOBANTHUS, Baril.
1. CoLOBANTHUS MUSCOIDES, Hook. /f.
Colobanthus muscoides, Hook. f. Fl. Antare. p. 14; Walp. Repert. 5, p. 788.

Has. Auckland Islands; collected only by Dr. Holmes.

This species and the next, from their fleshy character, have quite
the aspect of Portulacacece, to which Fenzl refers the whole tribe of
Molluginee, on account of the position of their stamens. C. muscoides
grows in large and very compact, moss-like tufts.

2. COLOBANTHUS CRASSIFOLIUS, Hook. /.

Colobanthus crassifolius, Hook. f. Fl. Antarc. p. 248; Homb. & Jacquinot, Bot. Voy.

Pol. Sud. Phanerog. t. 17.
C. Quitensis & C. sayinoides, Bartl. in Presl. Rel. Haenk. 2, p. 18, t. 49 (excl. syn.

H. B. K.)
Sagina crassifolia, D’Urv. in Mem. Soc. Linn. Par. 4, p. 617; Gaud. Bot. Voy.

Freye. p. 187.

Var. 8. caulibus brevissimis dense ceespitosis; capsula matura calycem

superante.

+

Colobanthus aretioides, Gillies, in Hook. & Arn. Bot. Misc. 3, p. 836.

Has. Orange Harbour, Fuegia; where it is very abundant. Casa
Cancha, Andes of Peru. Var. @. Cordilleras above Santiago, Chili.

I see no difference between the specimen from the high Andes of
Peru, at Casa Cancha, and the more condensed states of the species

132 PHANEROGAMIA.

from Antarctic America. The Chilian specimens, which also have
very short and densely ceespitose stems, are in mature fruit, with the
pods dehiscent, and their valves considerably longer than the calyx;
they belong therefore to a variety which Dr. Hooker has indicated.
None of our plants show any trace of the four bracts on the peduncle,
which Kunth’s Sagina Quitensis is said to bear.

3. COLOBANTHUS SUBULATUS, Hook. f.

Colobanthus subulatus, Hook. f. Fl. Antare. p. 18 & 247, t. 93.
C. Benthamianus, Fenzl, in Endl. Atakt. t. 49, & Ann. Wien. Mus. 1, p. 49.
Sagina subulata, D’Urv. in Mem. Soc. Linn. Par. 4, p. 618.

Has. Orange Harbour, Fuegia; where it abounds, growing in
broad and close tufts.

This species, which is admirably figured and characterized by Dr.
Hooker, 1s more rigid and pungent than any of its congeners.

18 SCHIEDEA, Cham. & Schlecht.

Calyx quinquepartitus, persistens. Corolla nulla. Staminodia, vel
squamule petaloidee, 5, ligulata, apice bifida, imo calyci inserta,
sepalis opposita. Stamina fertilia 10, capillaria, imo calyct inserta,
gquinque sepalis alterna, quinque visdem opposita cum basi stamino-
diorum accreta. Styli 3, rarius 4-0, intus stigmatosi. Ovarium uni-
loculare ; ovulis plurimis columelle centrali affixis. Capsula trivalvis,
raro 4—9-valvis. Semina plurima, estrophiolata. Embryo annularis,
albumen farinaceum cingens.—Suffrutices Sandwicenses, oppositifolii,
exstipulati, cymulis in thyrsum interruptum digestis, vel paniculatis
patentissimis.

SoHIEDEA, Cham. & Schlecht. in Linnza, 1, p. 46; Fenzl, in Endl. Atakt. Bot. t.
14; Endl. Gen. no. 5192; Hook. Ie. Pl. t. 649, 650.
Evcrapvus, Nutt. MSS. in Herb. Hook.

The essential character of this genus inheres in the petaloid bodies
which Chamisso and Schlechtendal took for true petals, and incau-
tiously described as alternate with the sepals; whereas they stand

CARYOPHYLLACEX. 133

before them, as was first stated by Fenzl, who called them para-
stemones, but apparently did not notice their connexion with the
base of those stamens which stand before them. If these petaloid
bodies represent, as they probably do, the primary series of stamens,
the genus is not Portulacaceous, according to Fenzl’s own distinguish-
ing characters—Some years ago, Sir Wm. Hooker (in Icones Plan-
tarum, |. c.) figured a second species of the genus, of widely different
habit from the original 8. ligustrina, and likewise indicated a third
species, S. Menziestx. But the plant he took for S. digustrina is, more-
over, a fourth species, which, like S. Menzies, has strongly three-
neryed leaves, in this respect differing from the rest of the genus.
The collection of the Exploring Expedition comprises the true 8.
ligustrina, and a new, narrow-leaved species allied to it; also S. Nut-
tallii, and a well-marked new species of that group, with effuse
panicles and acute sepals.*

1. ScHIEDEA LiGUSTRINA, Cham. & Schlecht.

S. glaberrima ; panicula thyrsoidea contracta e cymulis brevibus ; folirs
oblanceolatis lanceolatisve basi atienuatis sessilibus acutatis uninerviis ;
sepalis obtusis ; seminibus muricatis.

* The species now known of this curious genus may be disposed as follows :—

+ Panicula thyrsoidea, contracta, interrupta, nuda: folia uninervia,

1. ScHIEDEA LIGUSTRINA, Cham. & Schlecht. Vide supra.

2. SCHIEDEA SPERGULINA, Sp. Noy. Vide infra, p. 135.

* * Panicula contracta, ramosa, deliquescens: folia trinervea.

8. ScurepEa Menzrestt, Hook. Ic. Pl. fol. 649, adnot.

4, ScorEDEA HooKeERI: glabra; panicula elongata laxiflora basi foliosa ; pedicellis
pedunculisque filiformibus patentibus ; foliis membranaceis lanceolatis acutis eximie trt-
nerviis nunc quinquenerviis basi supra insertionem connatam abrupte contractis ; sepalis
obtusiusculis.— S. ligustrina, Hook. |. c. adnot, non Cham. & Schlecht. (Sandwich
Islands, probably Hawaii, Menzies.)

* * * Panicula deliquescens, effusa: sepala acuminata: folia uninervia, lata.
5. ScurepEA Nutra, Hook. Vide infra, p. 137.

6. ScHIEDEA DIFFUSA, Sp. Nov. Vide infra, p. 138.
34

134 PHANEROGAMIA.

Schiedea ligustrina, Cham. & Schlecht. in Linnea, 1, p. 46; Fenzl, in Endl. Atakt.
Bot. t. 14, & Ann. Wien. Mus. 2, p. 2738.
Portulacea, Hook. & Arn. Bot. Beech. Voy. p. 188.

Has. Oahu, Sandwich Islands; common on arid plains of the
Kaala Mountains, in the district of Waianae.

Plant wholly glabrous in every part, in our specimens only a foot
high (sometimes 3 feet high), shrubby, except the flowering shoots,
upright, dichotomously much branched; the older stems 2 lines in
diameter, covered with gray bark; the branches rather strict, with
tumid internodes; the older ones nearly terete, those of the season
four-angled and more or less flattened. eaves all opposite and con-
nate at the insertion, in the manner of the order, thickish, from an
inch to 2 inches long, varying from obversely lanceolate or oblong-
spatulate to narrowly lanceolate, and from 3 to 1% lines in width,
broadest above the middle, thence gradually narrowed to the base, but
sessile, tipped with a short and abrupt or mucronate point, one-nerved,
or very obscurely triplinerved (the lateral nerves and minute veins
obsolete or hidden), but not at all three-nerved; the margins thickish,
often slightly revolute. Sometimes a few smaller leaves are fascicled
in the axils, on undeveloped branches. Inflorescence a narrow and
strict interrupted thyrsus, or contracted panicle, from 2 to 4 inches Jong,
terminating the branches, composed of four or five pairs of subsessile
or short-peduncled and many-flowered cymules; the lower pairs from
half an inch to an inch apart, the upper ones approximate. Bracts
very small, ovate or subulate, or only the lower pair foliaceous. Pedi-
cels 12 to 6 lines long, capillary, thickened at the apex. Flowers
small; the sepals of the five-parted (or occasionally 4—6-parted) calyx
only a line or a line and a half long, ovate or oblong, very obtuse,
somewhat fleshy, with scarious margins, obsoletely three-nerved.
Corolla none. Staminodia (in place of the first series of stamens)
petaloid, inserted into the base of the calyx opposite the sepals, more
or less shorter than they are, persistent, narrowly linear, white, more
or less thickened at the base, where they are hollowed into a concavity
inside, the apex deeply two-cleft; the slender lobes subulate. Fertile
stamens 10, unequal, five of them inserted into the base of the calyx
iternate with its divisions, and five opposite the divisions and inside
of the staminodia, with which the base of their filaments is coherent:
in the tetramerous flowers the stamens are 8, in the hexamerous, 12.

CARYOPHYLLACE &. 135

Filaments capillary, elongated and exserted, but of variable length,
persistent. Anthers oblong, two-celled. Ovary obovoid, one-celled,
many-ovuled. Styles 3, or sometimes 4, capillary, stigmatose down
the inner side. Ovules campylotropous, inserted on a central colu-
mella. Capsule ovoid, three-valved to the base, rarely four-valved,
a little longer than the calyx; the valves chartaceous, at length two-
toothed at the apex. Seeds several, minute, roundish, slightly com-
pressed, with a crustaceous and acutely tuberculate or muricate testa.
Embryo coiled into an almost complete ring, around the sparing fari-
naceous albumen.

Although our specimens have not been compared with those of Cha-
misso, yet the habitat, the detailed description, and a tracing which I
possess of the figure above-cited in the rare Atakta Botanica, leave no
doubt of their identity. The inflorescence has justly been compared
with that of Silene Otites, although the flowers are much smaller.
The leaves, also, are not unlike those of that plant, although more
crowded on the woody stem. ‘The original authors describe them as
triplinerved, but with the lateral veins immersed and obsolete. Fenzl
more correctly calls them one-nerved; and in our specimens there is
scarcely a trace of lateral nerves.—To this species belongs the “ Por-
tulacaceous” plant described, without a name, in the Botany of
Beechey’s Voyage, p. 188, which was received in a package of plants
from China, and therefore was supposed to have been collected in
that country; but it doubtless came from Oahu.—The plant of Men-
zies’ collection, which Hooker mentions, in Icones Plantarum, l. c.,
under the name of S. ligustrina, is a wholly different species, with a
deliquescent panicle and with conspicuously three-nerved leaves, like
those of S. Menziesti. (Vide p. 133.)

2. SCHIEDEA SPERGULINA, Sp. Nov. (Tab, 11.)

S. panicula thyrsoidea contracta e cymulis brevibus puberulis ; foliis
angustissimis filiformibus uninerviis fasciculatis ; sepalis obtusis ;
seminibus fere leevibus.

Has. On the Mountains of Kauai, one of the Sandwich Islands.

Stem shrubby, erect, apparently one or two feet high, terete, nodose,

136 PHANEROGAMIA.

much branched, the branches very leafy, glabrous. Leaves opposite,
spreading, approximate (the internodes varying from a quarter of an
inch to an inch in length) and usually with smaller ones fuscicled in
their axils, thus appearing much crowded, filiform, or nearly so, the
larger ones 2 inches long and only half a line wide, one-nerved, invo-
lute, or channelled above, mucronulate, glabrous, their bases connate
by a narrow membrane, not at all stipulate. Inflorescence a narrow
and crowded, interrupted thyrsus, much like that of Schiedea ligus-
trina, composed of from 3 to 5 pairs of many-flowered cymules; the
lower pairs mostly rather distant, and sometimes prolonged; the
upper ones approximate. Bracts of the lower pairs like the cauline
leaves; of the upper, very short, ovate-subulate. Peduncles and pedi-
cels minutely pubescent, one or two lines long. Calyx campanulate, a
line or less in length, five-parted; the sepals ovate, obtuse, rather
fleshy, but with narrow hyaline margins, nearly nerveless, very
slightly pubescent externally near the base, persistent. Staminodia
5, petaloid, inserted into the very base of the calyx, one opposite each
sepal, at first short, at length almost equalling them in length, linear-
subulate, thickened at the base, where they are concave and appa-
rently nectariferous on the inner side, persistent. Fertile stamens
10, five of them inserted into the base of the calyx alternate with its
divisions, and five inserted before the staminodia, with the base of
which they cohere. Some of the flowers are tetramerous and octan-
drous. Filaments capillary, elongated and exserted, persistent.
Anthers oblong, two-celled. Ovary ovoid, one-celled, with about 15
campylotropous ovules on a free central placenta. Styles 3, capillary,
stigmatose on the inner face. Capsule ovoid, three-valved, a little
longer than the calyx. Seeds several, minute, reniform; the festa
obscurely tuberculate or roughened, nearly smooth.

This species is a strict congener of S. ligustrina; and is remarkable
for its very narrow and fascicled leaves, not unlike those of the com-
non Spurrey.

Piate 11, A.—ScurepEA sperGuLina. Fig. 1. A flower. 2. A
sepal. 3. A staminodium, with the stamen coherent with its base.
4. A stamen of the other series. 5. Pistil, with the ovary longi-
tudinally divided. 6. Dehiscent capsule, with the persistent calyx,
&e. 7. A seed.—AIl the details more or less magnified.

CARYOPHYLLACE&. 137

3. ScuteDEA Nutra, Hook.

S. glaberrima ; panicula ampla ramosissima patentissima ; foliis breviter
petiolatis oblongis ovatis seu ovato-lanceolatis crassiusculis obscure pen-
ninerviis; sepalis ovatis acuminatis capsula brevioribus; seminibus
tuberculatis.

Schiedea Nuttallii, Hook. Ic. Pl. t. 649, 650; Walp. Repert. 5, p. 789.
Eucladus suffruticosus, Nutt. in herb. Hook.

Has. Oahu, Sandwich Islands; on rocks, near Honolulu; where it
was also collected by Nuttall, and recently by B. Seemann, in the
cruise of the British surveying ship, Herald. Also on the mountains
of the west side of Maui? (the specimen not in flower).

Plant entirely glabrous. Stems branched from the base, suffruticose,
2 or 3 feet high; the branchlets spreading, angled. Leaves of a thick
texture, opposite, 2 or 3 inches long, and 12 to 18 lines wide, or the
lower considerably larger, oblong, ovate, or the upper ovate-lanceolate,
obtuse and abruptly apiculate, the lower rounded, the upper mostly
acute at the base, short-petioled, marked with a rather prominent
midrib, but with no lateral nerves, obscurely pinnately-veined. Panicle
terminal, compound, deliquescent, ample, sometimes a foot long, diffusely
spreading ; the slender branches elongated, divergent, subtended by
small ovate-subulate bracts, or the lowest by small leaves. Pedicels
capillary, 3 to 6 lines long, glabrous. Calyx a line and a half long,
five-parted; the sepals ovate-lanceolate, acuminate, rather fleshy, with
whitish hyaline margins, persistent. Staminodia, stamens, pistil, &c.,
nearly as in the other species of the genus. Capsule ovoid-oblong,
longer than the calyx, three-valved, few-seeded. Seeds tuberculate-
roughened. Embryo curved around the farinaceous albumen into.
almost a complete ring.

This species and the next are undoubted congeners of the preceding
ones, notwithstanding the effuse paniculate cymes, and the acute
sepals. There is only a single and poor flowering specimen of S.
Nuttallii in the collection.

85

138 PHANEROGAMITA.

4, SCHIEDEA DIFFUSA, Sp. Nov.

S. glabella ; panicula effusa, pedunculis pedicellisque capillaribus pube-
rulis; caulibus diffusis reclinatis herbaceis; foliis membranaceis ob-
longo-lanceolatis acuminatis breviter petiolatis uninerviis venulosis ;
sepalis ovato-lanceolatis acuminatis capsulam plerumque 5-valvem
equantibus ; seminibus levissimis.

Has. Hawaii, Sandwich Islands; in the forests of Mouna Kea and

Mouna Loa. Also on mountains of the west side of the island of
Maui.

Stems herbaceous nearly to the base? slender, several feet in
length, terete, reclining, loosely branched, glabrous or nearly so; the
branches more or less angular, tumid at the nodes. Leaves oblong-
lanceolate, acuminate, often very much so, either rounded or acute at
the base, 3 or 4 inches long, on petioles of about 3 lines in length,
their texture thin and membranaceous, glabrous or nearly so, with a
rather conspicuous midrib, but no lateral nerves, pinnately veined by
transmitted light. Punicle compound, effuse, very loose, from 6 to 20
inches long when fully developed; the branches, peduncles, and pedi-
cels capillary, minutely pubescent, sometimes nearly glabrous, the latter
from one to two inches long in fruit. Lower bracts often foliaceous,
lanceolate or linear; the others subulate or setaceous. Calyx five-
parted, glabrous; the sepals 2 lines, or in fruit 3 lines long, ovate, or
ovate-lanceolate, much acuminate, obscurely 3-5-nerved at the base,
with narrow scarious margins. Staminodia 5, inserted into the very
base of each sepal, shorter than it, narrowly subulate, two-cleft at the
apex, persistent. Fertile stamens 10, inserted into the base of the
calyx, five alternate with the sepals, and five opposite them within
the staminodia, with the very base of which they are slightly cohe-
rent. Filaments capillary, about twice the length of the calyx.
Anthers two-celled. Ovary ovoid, one-celled, many-ovuled; the cam-
pylotropous ovules on slender funiculi, inserted on a free central
placenta. Styles mostly 5, sometimes 3, capillary, minutely downy.
Capsule ovoid, membranaceous, nearly 3 lines long, not exceeding the
persistent calyx, three-valved, or more commonly jfive-valved ; the valves

PORTULACACES. 139

membranaceous, opposite the sepals when they are equal in number.
Seeds indefinitely numerous (at least in pentamerous capsules), lenti-
cular, with a perfectly smooth and even testa. Embryo curved into
almost a complete ring around the farinaceous albumen.—The panicle,
at first terminal, often becomes lateral or alar by the development of
branches from its base, which overtop it.

Puate 11, B—Scurepea pirrusa. Fig. 8. Capsule, with the per-
sistent calyx, filaments, &. 9. A seed. 10. The seed vertically
divided, showing the embryo coiled around the albumen.—The details
all more or less magnified.

Orv. PORTULACACES.

1. PORTULACA, Journ.

1. PorTuLaca OLERACEA, Linn.
Portulaca lutea, Soland. in Forst. Pl. Esc. Ins. Austr. p. 72 (abs. char.).

Has. Clermont Tonnerre, Wake’s, Gardiner’s, Binney’s, and other
Coral Islands. Navigators’ Islands.

This is enumerated in Dr. Pickering’s list under the name of Portu-
laca lutea of Forster, which it undoubtedly is, as distinguished from
the P. oleracea; but the specimens do not furnish any tangible dis-
tinctive characters. I find no specimens from the Society Islands, but
Dr. Pickering mentions “P. oleracea (not lutea),” as an introduced
plant there, growing in cultivated grounds.

140 PHANEROGAMIA.

2. PORTULACA QUADRIFIDA, Linn.

Has. Tongatabu, Samoan, and Feejee Islands; near the coast.

3. PORTULACA HIRSUTISSIMA, Camb.
Portulaca hirsutissima, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 191, t. 114.
Has. Low grounds, near Rio J aneiro, Brazil.
Accords with the figure above-cited, except that the testa of the
seed is perfectly smooth and shining. “Flowers purple.”
4. PorTULACA PILOSA, Linn.
Portulaca pilosa, Linn.; Geertn. Fruct. 2, p. 212, t. 128, f. 4; Cambess. 1. c.
Has. On hills, near Rio Janeiro.
The seeds, in the single specimen gathered, are obscurely muricate;

while in the North American plant they are strongly muricate, as in
the figure of Geertner above cited.

5. PoRTULACA VILLOSA, Cham.
Portulaca villosa, Cham. in Linnea, 6, p. 565; Walp. Repert. 2, p. 234.

Has. Sandwich Islands: on Diamond Hill, and on the plains near
Ewa, Oahu.

“Flowers fine purple;” the petals 5 or 6 lines long, rather longer
than the sepals. It appears scarcely different from P. pilosa, except
in the seeds, which are smooth, or, when considerably magnified,
minutely marked with sinuous areolations.

PORTULACACE AZ. 14]

6. PoRTULACA SCLEROCARPA, Sp. Nov.

P. (Teretifolice) radice carnosa tuberosa; foliis subulatis fasciculis pilo-
rum paullo tongioribus; floribus sessilibus ad apicem ramorum con-
gestis (pallidis); capsula spherica coriacea juxta basim sero circum-
scissa ; seminibus leevibus.

Has. Sandwich Islands: district of Waimea, Hawaii. Also? on
sand-hills, Maui.

Root fleshy and tuberous, about half an inch in diameter, apparently
not ligneous. Stems also thickened at the base, 2 or 3 inches high,
much branched. Leaves terete, subulate, 4 lines long, crowded,
resembling those of P. pilosa, somewhat exceeding the copious tufts
of fulvous bearded hairs that clothe the nodes. Flowers sessile and
crowded at the apex of the branches, apparently smaller than in
P. pilosa, “very pale” (the particular colour not recorded). Capsule
spherical, 12 to 2 lines in diameter, with very thick walls of a firm
coriaceous or almost crustaceous texture, adherent to the calyx-tube
only at the very base, a little above which it is marked with the cir-
cum-sessile line of dehiscence. The dehiscence is extremely tardy:
the ripe pods in the specimens fall away entire from the peduncle, and
afterwards open by the separation of the short basal portion (which is
also much thickened internally) only on the application of pressure.
Placentz 8, elongated, crustaceous, bearing many seeds on short funi-
culi. Seeds smooth, shining, minutely and closely areolate under a
powerful lens.

Unless the fruit is in abnormal condition (of which there is no indi-
cation), this species is well distinguished by its spherical and almost
crustaceous capsule, which on a smaller scale somewhat resembles the
fruit of Cow Lachryma. The Hawaiian specimen is dwarf and de-
pressed; that from Maui is taller, and more like P. pilosa, although
with a tuberous root; and having no ripe fruit, I am doubtful whether

it really belongs to this species.
36

142 PHANEROGAMIA.

2, SESUVIUM, Linn.
1. Sesuvium PortuLacastrum, Linn.

Has. Rio Negro, North Patagonia. Feejee Islands; Tongatabu ;
and Wake’s Island; on the border of lagoons.

The specimens from Rio Negro show that Cambessedes (in St. Hil.
Fl. Bras. 2, p. 199) has rightly referred to this species the S. parvi-
jlorum, DC. They entirely resemble the North American plant: their
flowers are mostly as large, and on short but manifest peduncles.
The South Sea Islands specimens are usually more rigid; the stems
apparently suffrutescent at the base.

3. TALINUM, Adans.

1. Tatinum paTens, Willd.

Talinum patens, Willd. Spec. 2, p. 862; DC. Prodr. 8, p. 8357; Hook. & Arn. Bot.
Beech. Voy. p. 63; Guillem. Zeph. Tait. p. 56.

‘Has. Tahiti and Kimeo, Society Islands. “An introduced weed,
in cultivated ground.” As it is not mentioned by Forster, it has

probably been introduced, from the American continent, in more
recent times.

4, CALANDRINIA, AB K.
1. CALANDRINIA GLAUCA, Schrad.
Calandrinia glauca, Schrad. in DC. Prodr. 8, p. 359; Gay, Fl. Chil. 2, p. 487.

Has. Maritime cliffs, near Valparaiso, Chili.

PORTULACACEZ. 143

The stems are said to attain the height of three feet, and the leaves
are rather ovate or obovate than lanceolate-spatulate. The flowers
are large and red-purple.

2. CALANDRINA ARENARIA, Cham.

Has. Environs of Valparaiso. (Destitute of flowers and fruit.)

3. CALANDRINIA BIFLORA, Meyen.

Calandrinia biflora, Meyen, Riese, 1, p. 349; Walp. Rel. Meyen, p. 340.
C. acuta, Meyen, MSS. in Herb. ips. et Hook.
C. Gayana, Barneoud, in Gay, Fl. Chil. 2, p. 503?

Var. GLANDULOSA : caulibus floriferis foliisque acutis glandulosis ; sepalis
basi parce echinatis hirto-glandulosis.

Has. Andes of Chili, on the first Cordillera above Santiago.

The specimens are in fruit only, and with the capsules dehiscent.
Stems or branches 2 or 3 inches long, from a procumbent and slender,
perhaps lignescent base, slender, minutely glandular, as well as the
leaves. The latter are only from 4 to 6 lines long, rather lanceolate
than linear, acute, attenuate at the base, about a line wide; those at
the base clustered and often spatulate. Flowers terminal, solitary or
in pairs; peduncles half an inch in length. Sepals 5 or 6 lines long,
broadly ovate, obtuse, obscurely denticulate, or the inner one entire,
glandular-hispid, and beset, especially towards the base, with scattered
and stout glandular-roughened sete or echinate projections. Capsule
longer than the calyx, three-valved. Seeds (principally shed) nearly
smooth, inappendiculate.

4, CALANDRINIA PHAcospERMA, DC.
Calandrinia Phacosperma, DC. Prodr. 3, p. 359? (Phacosperma Peruviana, Haw. ?)

Has. Bajios and Obrajillo, Andes of Peru.

The specimens are too poor for exact determination. The plant is
allied to C. caulescens, H. B. K., but is much smaller in all its parts,
The subspatulate-linear leaves are sparingly hispid along the margins,

144 PHANEROGAMIA.

5, CALANDRINIA PicKERINGU, Sp. Nov.

C. glabra; radice tuberosa; caulibus simpliciusculis (tripollicaribus) ;
Soltis omnibus alternis linearibus carnosis, supremis ad bracteas mini-
mas ovato-subulatas reductis ; floribus racemosis ; pedicellis erectis flore
subequilongis ; staminibus 3.

Has. Woolongong, New South Wales.

Stems numerous from the tuberous root, rather slender, about 3
inches high, simple or forked, leafy, glabrous, as is the whole plant.
Leaves all alternate, fleshy, sessile; the radical linear-oblong; the cau-
line linear, obtuse, apparently almost terete, narrowed towards the
base, half an inch long. Raceme terminal, erect, simple or sometimes
forked, 6-10-flowered. Pedicels erect, from one to 24 lines long, oppo-
site the minute ovate subulate bracts. Flowers small, about two lines
in length. Sepals ovate, obtuse. Petals (purplish?) obovate, longer
than the calyx. Stamens 3. Ovules 15 to 20. Capsule three-valved,
longer than the persistent calyx, 10-12-seeded. Seeds compressed,
minutely muricate-roughened.

Several Australian species of Calandrinia have recently been pub-
lished: but this accords with none of them. |

5. MONTIA, Mich.
1. Montra rontana, Linn.

Has. On the high Andes of Chili above Santiago, near the snow-
line.

6. TETRAGONLIA, Linn.
1, TETRAGONIA EXPANSA, Ait.

Has. Maritime sands, near Valparaiso (where it was also gathered
by Bertero); “native.” Sydney, New South Wales.

Orn MESEMBRYANTHEMEX.

1 MESEMBRYANTHEMUM, Linn.
1. MESEMBRYANTHEMUM NoDIFLORUM, Linn.

Has. Madeira, on the coast east of Funchal.

2. MESEMBRYANTHEMUM GLAUCUM, Linn.

Has. Cape of Good Hope, in the vicinity of Cape Town.

3. MESEMBRYANTHEMUM TENUIFOLIUM, Linn.

Has. Cape of Good Hope; with the preceding.

4, MrsEMBRYANTHEMUM AUSTRALE, Aid.

Has. Bay of Islands, New Zealand (where it was also collected by
Cunningham, Raoul, &c.): the specimen is destitute both of flower

and fruit.

37

Orvp. MALVACE &.

1. PALAVA, Cav.

1. PALAVA MOSCHATA, Cav.

Palava moschata, Cav. Diss. 1, p. 41, t. 11, f. 5; Lam. Il. t. 577.
Palavia moschata, Willd. Spec. 3, p. 768; DC. Prodr. 1, p. 458.

Has. San Lorenzo, near Callao, Peru: without. flowers or fruit.

2, MALVASTRUM, Gray.

MALVASTRUM, Gray, Pl. Fendl. (Mem. Amer. Acad. 4), p. 21, & Gen. Am. Bor.
Ill. 2, p. 59.

This genus is distinguished from Malva by the capitate stigmas;
from Sida by the peritropous-ascending ovule and inferior radicle.
The species with a manifest involucel have commonly been referred

to the former, and those with a small or caducous involucel, or none
at all, to the latter genus.

1. MALvAsStRuUM PERUVIANUM.

Malva Peruviana, Linn. Spec. ed. 1, p. 688; Jacq. Hort. Vindob. t. 156; DC.
Prodr. 1, p. 485.

M. Limensis, Hook. & Arn. Bot. Mise. 3, p. 151; non Linn.?

Has. Near Lima and Obrajillo, Peru; common.

MALVACES®. 147

The small flowers are noted as “blue;” but the carpels are “exas-
perata dentibus alternis plurimis,” leaving no doubt that ours is the
Linnzan plant. It was gathered also by Dombey in the same district.

2. MALVASTRUM SCABRUM.

Malva scabra, Cav. Diss. 5, p. 281, t. 188, f. 1; DC. Prodr. 1, p. 430.
M. scoparia, Jacq. Ic. Rar. t. 139, ex DC.

Has. Obrajillo, Andes of Peru.

Differs (perhaps too slightly) from Malvastrum scoparium (Malva
scoparia, L’Her., Cav.) in the more coarsely and sharply toothed
leaves, which are not canescent, and in the shorter tubercles rather
than awns on the back of the carpels, which moreover are muticous
at the proper apex. JM. scoparium-(to which apparently belongs Sida
depressa, Benth. Bot. Voy. Sulph. p. 69), a common plant around
Lima, although not in the present collection, has a short and erect

cusp at the strongly incurved apex of the carpels, as is described by
Cavanilles.

3. MALVASTRUM PLUMOSUM.
Malva plumosa, Presl. Rel. Heenk. 2, p. 124; Walp. Repert. 1, p. 295.
Has. Obrajillo, Peru. (Peru, Matthews, in herb. Hook. No. 776.)

The carpels are destitute of any awns or processes on the back;
but they bear at the apex a pair of long and slender, woolly-plumose
awns: they are nearly membranaceous in texture, minutely downy,

not rugose nor reticulated: in dehiscence they separate into two
valves,

4, MALVASTRUM SPICATUM, Gray.

Malvastrum spicatum, Gray, Pl. Fendl. 1. c¢. p. 22.

Malva spicata, Linn. Spec. ed. 2, p. 967; Cay. Diss. 2, t. 20, f. 4; DC. Prodr. 1,
p. 450.

M. ovata, Cav. Diss. 2, p. 81, t. 20, f. 2.
M. Timorensis, DC. 1. c.; Blume, Bijdr. p. 64; Decaisne, Herb. Timor. p. 102,

148 PHANEROGAMIA.

Has. Rio Janeiro, Brazil. St. Jago, Cape de Verde Islands; doubt-
less introduced from America.

5. MALVASTRUM TRICUSPIDATUM, Gray.

Malvastrum tricuspidatum, Gray, Pl. Wright. (in Smithson. Contrib. 3) 1, p. 16.

M. carpinifolium, Gray, Pl. Fendl. p. 22, excl. syn. Sida carpinifol. & planicaulis.

Malva Coromandeliana, Linn. Spec. 2, p. 687 (Pluk. Mant. t. 334, f. 2); Swartz,
Obs. p. 262.

M. tricuspidata, Ait. Kew. ed. 2, 4, p. 210; DC. Prodr. 1, p. 430.

M. subhastata, Cav. Diss. 2, p. 72, t. 21, f. 3; St. Hil. Fl. Bras. Mer. 1, p. 214.

M. Americana, Cav. 1. c. t. 22, f. 2, vix Linn.

M. Domingensis, Spreng. in DC. Prodr. 1, p. 481.

M. Lindheimeriana, Scheele, in Linnea, 21, p. 470.

Sida carpinoides, DC. Prodr. 1, p. 461; Walp. Rel. Meyen. p. 306.

Has. Rio Janeiro. Oahu, Sandwich Islands; naturalized around
Honolulu. Manilla, Luzon.

An American species now widely diffused over the warmer parts of
the world. It would not be proper to re-establish the Linnean specific
name of Coromandeliana for a species which Linnzus himself gives
as only American, and which certainly is not indigenous, even if now
to be found, on the Coromandel coast.

6. MALVASTRUM RHIZANTHUM, Sp. Nov.

+

M. acaule, glabrum; radice fusiformi tuberosa; foliis longe petiolatis
rotundis subcordatis duplicato-crenutis; floribus in collo umbellato-
congestis; pedunculis unifloris petiolis multo brevioribus; braeteolis
mvolucelli eciliatis; fructu glabro; coccis muticis dorso ad margines
dentibus 3-4 muricatis.

Has. High Andes of Peru, above Bafios.

A strictly stemless plant, glabrous throughout, except the inside of
the calyx; the clustered foliage and numerous, crowded peduncles
growing directly from the crown of the perpendicular fusiform-tuberous
root. Petioles mostly an inch and a half long. Leaves rotund, or

MALVACEA. 149

obovate-orbicular, about three-fourths of an inch in diameter, most of
them subcordate, some of them scarcely so, 3—7-plinerved from near
the base, veiny, crenate-toothed or doubly crenate, with the coarser
incisions few, and the obtuse teeth at first bristle-pointed. Peduncles
one-flowered, umbellute-crowded, even in fruit shorter than the petioles.
Bractlets of the involucel 3, linear, glabrous, not ciliate, shorter than
the calyx. Calyx 3 lines in length, five-cleft to the middle; the lobes
ovate, rather obtuse, pilose with soft hairs on the inside, but exter-
nally glabrous. Corolla nearly twice the length of the calyx, pale
purple? (the colour not recorded). Stamineal column simple, anthe-
riferous only at the summit. Styles 12 or 13, united below the
middle: stigmas capitate. Ovary glabrous; the cells uniovulate.
Ovule peritropous-ascending; the micropyle inferior. Fruit glabrous,
depressed : carpels rather membranaceous, indehiscent, reniform, with
the compressed summit more extended than the base, muticous, that
is, wholly awnless and pointless both at the apex and dorsally, but
the angles down the back muricate with 3 or 4 soft and long teeth.
Seed reniform. Radicle inferior.

In Sir William Hooker’s herbarium are specimens of this plant
gathered at Cerro Pasco by Matthews, along with an allied but pilose
species, which is perhaps the same as the following.

7. Matvastrum Ricut, Sp. Nov.

M. acaule, pilis stellatis hirsutum; radice napiformi; foliis rotundatis
subcordatis duplicato-dentatis vel incisis; floribus in collo wmbellato-
congestis ; pedunculis unifloris petiolis brevioribus ; bracteolis involucelli
2-3 exiguis.

Has. Near Obrajillo or Baiios, Andes of Peru.

A very small, depressed, strictly stemless plant, hirsute all over with
rather soft, fulvous, stellate hairs; the clustered foliage and the pedun-
cles growing directly from the crown of the thick and fleshy, peren-
nial, turnip-shaped root. Petioles half an inch to an inch long. Leaves
roundish, some of them slightly cordate at the base, half an inch in

diameter, coarsely doubly-toothed or incised, strongly five-ribbed, thickly
38

150 PHANEROGAMIA.

hirsute both sides. Peduncles wmbellate-crowded, 3 or 4 lines long,
simple, one-flowered, naked, or with one or two bractlets, like those of
the involucel, which are 2 or 3 in number, linear-setaceous, gla-
brous or stellately ciliate, half the length of the calyx. Calyx hirsute
externally, nearly glabrous within, five-cleft nearly to the middle, 2 to
3 lines long. Corolla white or purplish, a little longer than the calyx,
hairy at the base externally. Fruit not seen.

Allied to M. rhizanthum and M. acaule; but smaller in all its parts,
and hairy.*

8. MALVASTRUM ULOPHYLLUM, Sp. Nov.

M. nanum, e radice fusiformi multiceps, acaule; foliis confertis, petiolo
dilatato stipulisque adnatis nudis, lamina reniformi-rotundata flabel-

* MALVASTRUM ACAULE (Malva acaulis, Cay. Diss. 2, t. 35), of which there are
specimens in the Hookerian herbarium, from Matthews (Huamantanga) and M’Lean,
has hairy petioles, &c., sharply toothed and incised leaves, longer and usually more than
one-flowered peduncles, and larger flowers than MM. rhizanthwm. I have not seen the
fruit, which, however, is said to be tomentose.

MALVASTRUM PARNASSIL@FOLIUM (Sida parnassiefolia, Hook Ic. Pl. t. 385), from
the Quitensian Andes, has larger flowers than M. rhizanthum, on peduncles as long as
the obsoletely toothed leaves; the calyx is hirsute, and the petioles are hispid-ciliate.

MALVASTRUM PurDI@I (Sp. Nov.): acaule, hirsutum; radice fusiformi maxima;
foliis cordato-rotundis sublobatis crenato-dentatis glabratis ; pedunculis in collo umbellato-
confertis petiolis brevioribus apice capitato-trifloris ; bracteolis involucelli 2-38 lanceolatis ;
Jructu tomentoso depresso, coccis omnino muticis—Andes of New Granada, Mr. Purdie.
A species nearly allied to M. acaule, and with flowers about the same size: but the
peduncles (which are less than an inch long, even in fruit) are terminated by a capitate
cluster of about three flowers, which are not pedicellate. The bractlets of the involucel
are glabrous, or nearly so; but the calyx is hirsute externally. Corolla purple. Carpels
about 15, almost orbicular, tomentose on the back, not muricate nor roughened.

MALVASTRUM HUMILE (Malva humilis, Gillies, in Hook. Bot. Mise. 8, p. 150), as
formerly mentioned in Plante Fendlerianz, p. 21, belongs to the section of which the
North American M. coce‘newm is the type.

Malva Belloa, Gay, Fl. Chil. 1, p. 304, t. 7, is a true Malvastrum; and so probably
are all the other really indigenous Chilian species referred to Malva in the same work,
except Malva sulphurea of Gillies, which is a Sida.

Sida pygmea, Remy, in Ann. Sci. Nat. ser. 8, 8, p. 238, from the Bolivian Andes, is
probably a Malvastrum.

MALVACEA, 151

lato-multilobata supra cano-tomentosa subtus glabella, lobis 3—7T-lobulatis
confertissimis lobulisque obovatis involutis quasi crispis; floribus soli-
tariis petiolo insidentibus sessilibus ; calyce haud involucellato corollam
subeequante; coccis 8 subulato-rostratis hirto-villosissumis.

Has. Alpine region of the Cordilleras of Peru, at Alpamarca, near
the snow-line.

Root perpendicular, fusiform, rather stout, 5 or 6 inches in length ;
the crown divided into a number of extremely short leafy caudexes ;
the whole plant above-ground forming a depressed and rather close
tuft, an inch or two in diameter, and barely an inch high. Ascend-
ing stems none. Leaves crowded on the caudexes, but not imbricated,
spreading. Petioles about 3 lines long; the lower half or more dilated
and squamaceous, being winged by the large adnate stipules: these
are rather scarious, nearly glabrous, naked (not ciliate); their free
portions broadly linear, 3 lines long, about twice the length of the free
but flattened summit of the petiole. Lamina of the leaf rotund and
slightly reniform, 7-9-nerved from the base, jlabellately T-11-lobed to
the middle, thickish in texture, canescent-tomentose above, nearly gla-
brous underneath, about 3 lines in length and 4 in width; the lobes
obovate, again lobulate (the lobulets of the lateral segments 2 or 3,
or the larger middle ones 5-7), with involute margins, very much
crowded, and overlying each other, giving the leaf a remarkably
crisped appearance. lower borne on the petiole about its middle,
sessile, small, destitute of an involucel. Calyx campanulate, minutely
tomentose, five-cleft to the middle, 3 lines in length; the lobes trian-
cular-ovate, obtuse. Corolla scarcely exceeding the calyx; the colour
unknown. Stamens and styles (8) as in Malvastrwm. Capsule about
the length of the calyx; the carpels 8, very villous with rather hirsute
hairs, thin, semiovate, pointed with a subulate beak, which is const-
derably shorter than the cell, two-valved from the apex. Ovule
ascending; the micropyle inferior. Seed reniform. Embryo arcuate :
the radicle centripetal-inferior.

This species and the succeeding ones, as well as M. Pichinchense,
and their allies (natives of the Quitensian, Peruvian, and Chilian
Andes), all bear the flower on the dilated petiole, between the stipules,
in the manner indicated by Cavanilles in his Sida Phyllanthos. They
will constitute, therefore, a well-marked section of the genus (the

152 PHANEROGAMIA.

MALVASTRA PHYLLANTHOPHORA).* The character has been noticed by
Meyen in two species from the southern part of Peru; but it seems
to have escaped observation in the most conspicuous of all, namely

* Besides those in the collection of the Exploring Expedition, above enumerated, the
following species belong to this section :—

MatvastruM Puyxuantuos. Sida Phyllanthos, Cay. Diss. 5, p. 276, t. 127, f. 4;
H. B. K. Nov. Gen. & Spec. 5, p. 264.—The hirsute carpels are pointed with a subu-
late beak, which is shorter than the cell. The radicle, although rather abruptly inflexed,
is centripetal-inferior.— Sida Borussica of Meyen, is either a variety of this species, as
Walpers suggests, or a closely allied species.

M. PicutncuHEense. Sida Pichinchensis, Humb. & Bonpl. Pl. Aig. 2, p. 115, t. 116.
—The ripe fruit (in specimens collected by Hartweg) consists of about 12 two-valved
carpels, which are hirsute, membranaceous, tapering into an erect, lanceolate-subulate,
and soon bifid beak, which is longer than the cell. Embryo rather closely coiled, but
not conduplicate; the radicle centripetal-inferior. The character of the leaves is cor-
rectly indicated by Bentham (in Plante Hartwegiane). The flower is borne on the
winged part of the petiole between the stipules, just as in the preceding species.

Matvastrum Macweant (Sp. Nov.): nanum, multiceps e caudice brevissimo crasso,
acaule; foliis confertis biternatilobatis seu tripartitis segmentis 2-3-fidis, lobis obovatis
oblongisve obtusis supra (cum calyce) albo-villosis subtus villosulis ; Jlore in petiolo inferne
stipulis adnatis alato sessili magno.—High Andes of Peru, Mr. M’Lean (in herb.
Hook.).—The plant is only about an inch high, exclusive of the flower: the leaves form-
ing a dense tuft on the thickened caudex. Petioles 2 to 4 lines long, flattened below,
and with the stipules adnate to the middle, naked. The leaves themselves are villous
with long and soft white hairs, more densely so on the upper surface, 3 or 4 lines long
and 5 or 6 in width, somewhat reniform in circumscription, three-parted, or sometimes
flabellately 4—5-parted, and with the divisions deeply 2-3-cleft or parted, obovate or
cuneiform, much crowded; the segments either entire or again 2—3-lobed; the ultimate
lobes narrowly oblong, or obovate, obtuse. The flower is very large for the size of the
plant; the petals (apparently pale purple) being an inch and a quarter in length. Calyx
villous, like the leaves. Stamens, styles, &c., as in the genus. Fruit not seen.

MALVASTRUM PEDICULARIZFOLIUM. Sida pediculariafolia, Meyen, Reise, 1, p.
460, ex Walp. Rel. Meyen. p. 308.—A species evidently related to the foregoing, but

unknown to me.

MALVASTRUM compactuM. Sida compacta, Gay, Fl. Chil. 1, p. 329, is a well-
marked Phyllanthophorous species.

MALVASTRUM ANTHEMIDIFOLIUM. Sida anthemidifolia, Remy, in Ann. Sci. Nat.
ser. 3, 6, p. 356.

There are other undescribed Chilian and Peruvian species in the Hookerian herbarium.

MALVACER, — 153

M. Pichinchense. These plants have all the floral characters of Mal-
vastrum ; the solitary ovules being ascending; the embryo arcuate, or
in some species rather arcuate-inflexed, almost as in Sida, but the
radicle is always centripetal-inferior. The flowers are not involucel-
late nor bracteate.

9. MALVASTRUM ARETIOIDES, Sp. Nov.

M. nanum, densissime ccespitosum, acaule ; foliis imbricatis parvis; petiolo
stipulis fere ad apicem adnatis alato squamaceo hispido-ciliato ; limbo
pedato 5—T-partito stipulas vix superante, segmentis confertis obo-
vatis supra concavis cano-tomentosis subtus glaberrimis ; floribus soli-
tartis basi petiolo adnatis sessilibus; calyce exinvolucellato hirsuto
corolla dimidio breviore ; ovario tementoso T-8-loculari; coccis 8 subu-
lato-rostratis ?

Has. Alpine region of the Cordilleras of Peru, at Casa Cancha.

A single small specimen, in flower, is all that I have detected in
the collection. It is scarcely above half an inch high, looking as to
the vegetation somewhat like Donatia, or the tufts of a dense and
dwarf Saxifrage or Aretia, and it evidently grows in the same cespi-
tose manner. Leaves densely wmbricated on the short caudexes; the
dilated and scale-like petioles 2 or 3 lines long, winged with the adnate
portion of the stipules almost to the summit, and ciliate with strong
hispid hairs, which also sparingly beset the lower surface : the free por-
tion of the stipules also scarious and hispid-ciliate, almost as long as the
blade of the leaf; the naked portion of the petiole very short, some-
times almost wanting. Blade of the leaf only a line or a line and a
half long, rather coriaceous, glabrous underneath, minutely canescent-
tomentose above, palmately five-parted, or more commonly pedately-parted
into three principal divisions, of which the lateral are three-cleft and
the middle one entire; the lobes obovate, entire, concave (the margins
involute), very much crowded. Flower sessile, adnate to the base of
the dilated petiole. Involucel none. Calyx hirsute, not involucellate,
three lines long, five-cleft to the middle, the lobes oblong-lanceolate.
Corolla twice the length of the calyx, “white with a bluish base,” if I

rightly identify the plant with one mentioned in Dr. Pickering’s notes:
39

154 PHANEROGAMIA.

the petals cuneiform, with the apex truncate and slightly retuse.
Stamineal column and anthers as in the genus. Styles 8, united
below: stigmas capitellate. Ovary tomentose; the cells each contain-
ing a single ascending ovule. Mature fruit not seen: the young
_ carpels subulate-beaked more or less.

10. MALVASTRUM PINNATUM.

M. nanum, subacaule, ceespitosum ; caudice multicipite crasso ; foliis con-
Sertis pinnatipartitis supra cano-tomentosis subtus glabratis, segmentis
linearibus integerrimis; floribus solitariis in medio petiolo alato his-
pido-ciliato sessilibus ; petalis oblongo-obovatis ; ovario circiter 12-locu-
lari hirsutissimo.

Sida pinnata, Cav. Ic. 5, p. 18, t. 422, f. 1; DC. Prodr. 1, p. 466.

Has. Alpine region of the Cordilleras of Peru, between Casa

Cancha and Culnai, &c.; common. -

Plant growing in depressed and apparently dense and broad caspi-
tose tufts, from a stout and thick branching caudex, only an inch high.
Leaves crowded and thickly covering the branches of the caudex, an
inch or more in length, including the dilated and spathaceous petiole;
which is 5 or 6 lines long, and combined (as in the preceding species)
with the adnate stipules, sparsely pilose externally, the margin ciliate
with hispid hairs. The stipules are adnate up to the base of the
lamina; the free portion lanceolate-subulate, hispid-ciliate. Lamina
of the leaf pinnately parted into 7 to 9 linear and entire segments, of
unequal length, nearly glabrous underneath, above canescent with a
white tomentum, the margins involute; the segments tipped with one
or more hair-like bristles. Flower pretty large and showy, sessile on
the middle of the dilated petiole, in the manner of MV. Phyllanthos, &e.,
but considerably below the point where the stipules become free.
Calyx campanulate, tomentose-pubescent, five-cleft to the middle; the
lobes triangular, acute. Corolla (purplish?) nearly thrice the length
of the calyx; the petals oblong-obovate, 10 lines long. Stamens, styles,
&c., as in the genus. Ovary very hirsute, about twelve-celled, with a
solitary ascending ovule in each cell. Fruit not seen.

nt

MALVACE&. 155

The specimens accord so well with the figure and description of the
Sida pinnata of Cavanilles (which, however, was gathered on Chim-
borazo), that they may confidently be referred to that species. True,
the flowers of S. pinnata are said to be sessile in the axils; but the
peculiarity in the insertion might readily be overlooked in this, as it
certainly has been in several other species. The leaves, also, are said
to be tomentose-canescent underneath, which is a natural oversight,
since the lower surface, if either, is the downy one in most plants.
But all the Phyllanthophorous Malvastra have the upper surface more
downy than the lower.

11. MALYAsTRUM STENOPETALUM, Sp. Nov.

M. nanum, cespitosum, acaule; foliis pinnatipartitis supra niveo-tomen-
tosis subtus glabris, segmentis tri-quinque-jugis confertis profunde tri-
fidis, lobis linearibus integerrimis vel jugi infimi bi-tri-lobulatis ; pe-
tiolo superne exalato basim versus florem sessilem inter stipulas adnatas

gerente; petalis angusto-spathulatis puniceis; ovario circ. 12-loculari
hirsuto.

Has. Alpine Cordilleras of Peru, near Casa Cancha. (Also gathered
by Mr. M’Lean.)

Plant growing in close, depressed tufts, half an inch to an inch
high, smaller in all its parts than the foregoing species; the branches
of the caudex not so stout. Leaves crowded, 6 to 8 lines long (includ-
ing the petiole) glabrous beneath, white-tomentose above with a very fine
and close tomentum, pinnately parted into 3 to 5 pairs of approximated
segments, which are again deeply trifid; the lobes linear or narrowly
oblong, entire, or those of the lowest and largest pair 2—3-lobulate ;
the margin involute. Petiole wingless and naked above; the scarious
stipules adnate towards the base only, scarcely if at all ciliate. Flower
sessile on the petiole next its base, a little below the free portion of the
stipules. Involucel none. Calyx campanulate, five-cleft to below the
middle, puberulent externally, the tube marked with five brown spots
within next the base; lobes triangular-ovate, acute, canescent-tomen-
tose inside. Petals deep red, greenish or yellowish at the base, at first
oblong-spatulate and 4 or 5 lines long, of a rather firm texture, not

156 PHANEROGAMIA.

convolute together after anthesis but remaining separate and _persis-
tent, becoming narrowly spatulate, and attaining the length of 6 or 8
lines. Stamineal column antheriferous for a considerable portion of
its upper part; the anthers apparently red or reddish. Styles 12,
red at the suminit, which is clavate-thickened: stigmas capitate, small.
Ovary hirsute, twelve-celled; the ovules, &., as in the preceding
species. Fruit not seen.

This well-marked species, although not yet met with in the collec-
tion, is so well indicated in Dr. Pickering’s notes as to enable me to
identify it with specimens in Sir William Hooker’s herbarium, ga-
thered in the same district by Mr. M’Lean, from which it is here
described. The corolla appears to have been blood-red.

12. MALVASTRUM CAVANILLESII.

M. nanum, subacaule ; caudice multicipite crasso ; foliis confertis pinnati-
partitis utrinque tomentosis, segmentis 5—7 obovatis trilobatis ; floribus
solitarius basi petioli dilatato sessilibus ; petalis obovatis ; ovario T-10-
loculart hirsuto.

Sida acaulis, Cav. Ic. 5, p. 18, t. 422, f. 2; DC. Prodr. 1, p. 466.

Has. High Cordilleras of Peru, near Casa Cancha, &c. (Also
gathered by Mr. M’Lean.)

This is another of these depressed and tufted alpine species, not
over an inch high, with branching caudexes almost as thick as those
of M. pinnatum. The leaves are smaller than in that species, tomen-
tose on both sides with a cinereous or fulvous pubescence; the very
small segments (barely a line in length) obovate and three-lobed; the
petioles less winged. The flowers also are small, and sessile at the
very base, if not truly in_the axils, of the leaves; the obovate petals
only 3 or 4 lines in length. The styles and cells of the ovary are
7, according to Cavanilles. In our plant they are 9 or 10; and the
ovary is hirsute, as in all these species. I have not seen the fruit.
Our specimens closely resemble the figure given by Cavanilles; but
the corolla does not appear to have been yellow.

MALVACES. 157

3. SIDA, Linn., Kunth.
1. SA SULPHUREA, Gray.

Sida (Pseudo-Malvastrum) sulphurea, Gray, Pl. Fendl. in Mem. Amer. Acad. 8, p. 23.
Malva sulphurea, Gillies, in Hook. Bot. Mise. 8, p. 149; Gay, Fl. Chil. 1, p. 295.

Has. Rio Negro, North Patagonia; very common.

The specimens of this prostrate, silvery-lepidote species wholly
accord with those of Gillies from Mendoza, and of C. Gay from the
interior of Chili. There are one or two setaceous or subulate bractlets,
either alternate, near the apex of the pedicel, or subtending the base of
the calyx, like an involucel. The ovule is resupinate-suspended,
and the embryo conduplicate, with the radicle superior: it is therefore
a true Sida. The colour of the flower in the recent plant is not
recorded: it 1s probably sulphur-yellow, changing to dull red and
greenish in drying, as in the Chilian plant. The species must be
closely allied to Malva leprosa of Ortega, which is unknown to me,
and which is said to come from Cuba.

2. SIDA LINIFOLIA, Cav.

Sida linifolia, Cav. Diss. 1, p. 14, t. 2, fig. 1; DC. Prodr. 1, p. 459; St. Hil. Fl.
Bras. Mer. 1, p. 180.

Has. Around Rio Janeiro, Brazil; where it is very abundant.
Ovolau, Feejee Islands. (Not mentioned in Dr. Pickering’s notes; so
that we know not if it was thought to be indigenous. It is singular

that an American species, not very widely diffused, should*have been
conveyed to this little-visited group of islands.)

3. SIDA SPINOSA, Linn.

Has. St. Jago, Cape de Verde Islands. Callao, Peru. Tongatabu.

The specimens belong to the variety with broader leaves, charac-
40

158 PHANEROGAMIA.

terized by Webb, in his Spicilegia Gorgonea. That from the coast of
Peru has more downy leaves, obtuse, and obtusely toothed; and well
accords with S. alba of Linnzeus, as described by Wight and Arnott.

4. SIDA RHOMBIFOLIA, Linn.
Var. a. carpellis apice aristatis, vel subulato-rostratis.

Var. 2. carpellis apice mucronulatis vel omnino muticis.

S. rhomboidea, Roxb. Hort. Bengh. p. 50; DC. Prodr. 1, p. 462; Wight & Arn.
Prodr. Ind. Or. p. 58; Decaisne, Herb. Timor, p. 105.

S. Hondensis, H. B. K. Nov. Gen. & Spec. 5, p. 260; DO. 1. ¢.

S. salicifolia, Forst. in herb. Mus. Par. ex Guillemin.

Has. Samoan and Feejee Islands; common: the var. a. with dis-
tinctly awned carpels—Samoan Islands, Tahiti, Luzon, Singapore, Rio
Janeiro, and Callao, Peru; the var. @. with pointless or barely mucro-
nate carpels.

This widely diffused species varies much in the shape and size of the
leaves, as well as in the smoothness or the degree of tomentose pubes-
cence of their lower surface; and the carpels are either long-awned,
awn-pointed, mucronate-beaked, or entirely blunt, without apparently
any means of drawing a line of distinction between the different forms,
as Mr. Bentham remarks. . canescens and S. alba, Cav., S. Cana-
mensis, Willd., belong to the species; and perhaps even S. retusa and
S. Philippica, make an extreme form of it.

5. Sma RETUSA, Linn.

Var. via suffruticosa ; foliis ovalibus vel obovato-oblongis parvis ; pedun-
culis folio longioribus.

Sida microphylla, Benth. in Hook. Lond. Jour. Bot. 2, p. 211, non Cav. ?
Has. Samoan and Feejee Islands, in waste places.

A small, depressed or procumbent species, with the stems scarcely

MALVACES. 159

woody at the base; the leaves only 2 to 4 lines long, oval, elliptical,
or obovate-oblong, scarcely retuse ; the peduncles nearly an inch long;
the carpels, &c., as in S. rhombifolia. It may be the little-known 8S.
microphylla; but the peduncles are much longer than the leaf.

6. Sripa acuta, Burm., DC.

Has. Tahiti, Society Islands: doubtless introduced. Shores of
Laguna, Baitos, Luzon.

To this belongs Sida stipulata, Cav., S, Balbistana, DC., and SW.
glabra of Nuttall. It has not before been recorded from the Society
Islands.

7. SIDA CARPINIFOLIA, Linn. f.

Sida carpinifolia, Linn. f. Suppl. p. 307; Cav. Diss, 5, p, 274, t. 184, f. 1; DC.
Prodr. 1, p. 461; St. Hil. Fl. Bras. 1, p. 184.

S. planicaulis, Cav. Diss. 1, p. 24, t. 8, f. 11.

S. bracteolata, DC. Prodr. 1, p. 460.

S. spireefolia, Willd.? Link. Enum. 2, p. 203? DC. Prodr. 1, p. 472?

Has. Around Rio Janeiro, Brazil; very common.

It is doubtful whether this plant grows at all in Madeira, nor have
I yet seen any specimens from the Canary Islands: but Malvastrum
tricuspidatum is naturalized in both places. The younger Linnzus
described his Sida carpinifolia, from a specimen gathered by Masson
“in the garden of the Franciscan monastery, Madeira.” The seeds
doubtless were derived from South America, probably from Brazil,
where the species is Indigenous and very abundant.

8. SIDA PANICULATA, Linn.
Sida floribunda, H. B. K. Nov. Gen. & Spec. 5, p. 258, t. 473; DC. 1. c. p. 465.

Has. Rio Janeiro, Brazil. Around Callao and Lima, Peru; very
abundant.

160 PHANEROGAMIA.

The, carpels are barely mucronate-pointed ; and they are represented
quite as much so in the figure of Sida floribunda, H. B. K., which
must be added to the synonymes adduced by DeCandolle. Sida
venusta, Walp. in Rel. Meyen. (excl..syn. Schlecht. ?) must also be the
same plant, as it is said to grow around Lima; and a Peruvian speci-
men of S. paniculata, received from Meyen under the name of Sida
atrosanguinea, exists in the Hookerian herbarium.

9. Sma Dompryana, DC.
Sida Dombeyana, DC. Prodr. 1, p. 463 ; Hook. Bot. Mise. 2, p. 209 t. 89.

Has. Lima, Peru: common in the bed of the river. Also near
Obrajillo.

This will probably have to be joined to the East Indian Sida
humilis.

10. Siva JAVENSIS, Cav.

Sida Javensis, Cav. Diss. 1, p. 10, t. 1, f. 5; DC. Prodr. 1, p. 465.
S. veronicefolia, Lam. Dict. ex Zollinger, coll.

Has. Near Manilla, Luzon.

11. Sma corpiroiia, Linn.
Has. St. Jago, Cape de Verde Islands. Near Rio Janeiro, Brazil.

To the synonymes adduced by Webb (Spicilegia Gorgonea) to this
species, viz.: S. herbacea, Cav., S. rotundifolia, Cav., S. altheifolia,
Swartz, and 8. Africana, Beauv., that of Sida multiflora, Cav., should
doubtless be added. The plant is very widely distributed through
the tropics both of the Old and of the New World. From Dr. Pick-
ering’s notes I am led to infer that it was seen around Lima (where
it was also gathered by Matthews); but it is not found in the collec-
tion.

MALVACES. 161

12. SipA FALLAX, Walp.

S. tomentosa, frutescens ; caule ramoso; foliis cordato-ovatis sew ovato-
rotundis canescenti-velutinis recte penninervus creberrime serrato-
erenatis; pedunculis axillaribus unifloris folia subcequantibus vel
superantibus ; calycis lobis ovatis acutis; coccis 6-9 (12 ex Walp.)
muticis glabellis apice obtuso rima brevi dehiscentibus.

Sida rotundifolia, Hook. & Arn. Bot. Beech. Voy. p. 79, non Cav.

Var. a. foliis (sepe parvulis) utrinque velutino-tomentosis subplicatis.

Sida fallax, Walp. Rel. Meyen. p. 306.
Anoda ovata, Meyen. Riese, 2, p. 139.

Var. 2. foliis supra minus velutinis aut subglabratis obtusis acutis vel
subacuminatis.

Has. Oahu, Sandwich Islands; on Diamond Hill, &., near Hono-
lulu (var. a.) Also gathered on Malden Island by Macrae.—The
var. 3. also from Oahu; and the Coral Islands: namely, Wake’s Island,
Gardner’s Island, and Birney’s Island. Also gathered on Oahu by
Lay and Collie, Meyen, Nuttall, Gaudichaud, Sc.

Stem more or less shrubby, at least at the base, erect, branching,
apparently 2 or 3 feet high; the branches, like the rest of the plant,
tomentose with a fine and close velvety pubescence. Leaves variable
in form, ovate, rowndish-ovate or oblong-ovate, with a rounded and
usually cordate base (the sinus small and narrow), either very obtuse,
acute, or somewhat acuminate, equally velvety-tomentose and canescent
on both sides, or, in var. 3. less downy and greener, or even almost
glabrate, above, closely crenate-toothed with rather fine and unequal
obtuse teeth, 6 to 18 and even 24 lines long, and on petioles of half
or a third the length of the blade, rather strongly pinnately-veined
with 8 to 11 pairs of straight veins on each side: in var. a. these are
so strong that the leaf appears somewhat plicate. Stipules setaceous.
Peduncles azillary, solitary, or two or three from the same axil, fili-

form, equalling or the upper exceeding the leaves in length, arti-
41

162 PHANEROGAMIA.

culated towards the apex, one-flowered. Calyx tomentose like the
leaves, five-cleft nearly to the middle, the lobes ovate or triangular,
acute or somewhat acuminate. Petals yellow, obovate, half an inch
in length, about a third longer than the calyx. Ovary minutely
pubescent. Fruit shorter than the tube of the calyx, subglobose,
nearly glabrous ; the carpels 6 to 9, or 12 (according to Walpers, and
Dr. Pickering’s notes), at length falling away separately, a line and a
half in length, slightly wrinkled or reticulated, obtuse and pointless,
dehiscent by a short transverse chink at the narrowed apex. Seed
and embryo of the genus.

A variable species, of which we have specimens well agreeing with
Walpers’ character (except that the leaves are almost always a little
cordate, and I have not noticed so many as 12 carpels), and others
which differ widely from it in the size of the leaves, &. The Sand-
wich Island plant cannot be separated from that of the Coral Islands.
I am not aware that the latter has been anywhere described.*

13. Stipa Drei, Sp. Nov.

S. cano-tomentosa ; caule basi suffrutescente? ramoso ; foliis oblongo-ovatis
subacutis basi rotundatis seu leviter subcordatis crenato-serratis velu-
tunis subtus imeanis recte penninerviis; pedunculis axillaribus 1-2-
Joris folia subeequantibus ; calycis lobis ovatis obtusiusculis 3 coccis 8
puberulis apice apiculato-bidentatis.

Has. Hawaii, Sandwich Islands; on the coast southwest of the
crater Lua Pele. (Port Byron, Rev. J. Diell.)

Plant apparently 2 or 3 feet high, with slender and spreading
branches (the base probably a little woody), softly tomentose-canescent
throughout; the branchlets and peduncles also beset with some fine
villous hairs. Leaves (the lower cauline ones wanting) 6 to 14 lines
long, on rather long petioles, oblong-ovate, acutish or obtuse, rounded or
very slightly cordate at the base, finely crenate-toothed, velvety both

* What is “Sida maura, “Link, Enum. 2, p. 204,” Endl. Bemerk. Fl. Sudseeins.
in Ann. Wien. Mus. 1, p. 182? “Ins. Marianne, Cham.’ There is no species under
this name in the work referred to.

MALVACES, 163

sides, the lower surface whitened and conspicuously pinnately veined,
the straight veins 6 or 7 on each side of the midrib. Stipules seta-
ceous. Peduncles axillary, solitary or in pairs, one-flowered or some-
times 2-flowered, 6 to 10 lines long, articulated near the apex. Flowers
yellow, nearly as large as in the foregoing species. Calyx tomentose,
five-cleft to the middle; the lobes ovate and obtuse or obtusish. Ovary
canescently pubescent. Fruit shorter than the tube of the calyx,
puberulent; the carpels usually 8, a line and a half long, somewhat
wrinkled, bidentate with two short teeth at the apex, opening between
the teeth. Seed and embryo as in the genus.

This species appears to be sufficiently distinct from both the pre-
ceding and the following. I have dedicated it to the memory of its
discoverer, the late Reverend John Diell, formerly Chaplain at Hono-
lulu. A specimen given by him to M. Gaudichaud, is preserved in
the herbarium of the Garden of Plants at Paris.

14. Sma Sertum, Nutt. ined.

S, subcanescens ; caule frutescente ramoso; ramis confertis ; foliis ovali-
bus crenulatis utrinque rotundatis basi scepius subcordatis longius
petiolatis subtus canescentibus supra glabratis; pedunculis axillaribus
unifloris folia cequantibus ; calycis lobis obtusissimis ; coccis 8~9 gla-
bellis breviter birostratis.

Sida Sertum, Nutt. ined. in Herb, Hook.
S. rotundifolia, Hook. & Arn. Bot. Beech. Voy. p. 79, var.

Has. Oahu, Sandwich Islands; on mountains behind Honolulu,
(Also gathered by Lay and Collie, Nuttall, Diell, &c.)

Apparently a low suffruticose species, with crowded and bushy
branches; which, with the petioles, peduncles, calyx, &c., are barely
canescent with a minute pubescence. Leaves an inch to an inch and
a half long, oval, or subovate-oblong, very obtuse and rounded at both
ends, usually more or less subcordate at the base, finely crenulate,
pinnately veined, but less conspicuously so than in the foregoing
species, canescent underneath with a minute and very close pubescence,
glabrous or glabrate above; the petioles two-thirds the length of the

164 PHANEROGAMIA.

blade. Stipules setaceous. Peduncles axillary and usually solitary,
capillary, one to two inches long, one-flowered. Calyx five-cleft
nearly to the middle; the lobes broadly ovate, very obtuse. Petals
yellow, nearly twice the length of the calyx, 5 or 6 lines long. Car-
pels about 8 in number, nearly glabrous, abruptly tipped with two
subulate-aristiform beaks, which are shorter than the cell, and minutely
hairy.

I am not certain whether Mr. Nuttall’s specimens in the Hookerian
herbarium under this name do not include some of Sida fallax.:
Hooker and Arnott have considered them all as forms of one poly-
morphous species. S. Meyeniana is distinguished by the sharply
toothed leaves of the same colour both sides, the sharply acuminate
sepals, and the long-beaked carpels.

15. Sipa Meyentana, Walp.

Sida Meyeniana, Walp. Rel. Meyen. p. 807, & Repert. 1, p. 94.
S. ulmifolia, Hook. & Arn. Bot. Beech. Voy. p. 79, non Cav.

Haz. Oahu, near Honolulu, Sandwich Islands. (Also gathered by
Lay & Collie, Macrae, Meyen, Gaudichaud, Barclay, &c.)

A suffruticose species, glabrous, or the branchlets and leaves when
young furfuraceous or scurfy with a sparse stellate pubescence. Sti-
pules setaceous, minutely hairy. Leaves membranaceous, green and
of nearly the same hue both sides, ovate, oblong-ovate, or rotund-
ovate, acute or acuminate, rarely obtuse, rounded or slightly subcor-
date at the base, sharply serrate, one to two inches long, or those of
the branchlets smaller. Peduncles axillary, usually solitary, shorter
than the leaves, or nearly equalling them in length. Calyx glabrous;
the lobes triangular-ovate, sharply acuminate, about half the length
of the oblique and yellow petals. Carpels 5 to 9, nearly glabrous,
two-beaked; the beaks subulate-awned, pubescent, as long as the cell
or longer.

Walpers compares this well-marked species with Sida rhombifolia,
to which it bears no particular resemblance.

MALVACES. 165

16. Smpa FIBULIFERA, Lindl.

Sida fibulifera, Lindl. in Mitch. Exped. Austr. ex Aun. Sci. Nat. ser. 2, 15, p. 58.

Has. Hunter’s River, New South Wales.

17. Sma corrvuGata, Lindl. 0. ¢.

Has. Hunter’s River and Puen Buen, New South Wales.

4. CRISTARIA, Cav.
1. CrISTARIA ECRISTATA, Sp. Nov.

CO. glabra; caule erecto paniculato; foliis caulinis 1-2-ternatisectis lacie
niatis hirtellis, ramealibus trisectis, segmentis linearibus ; coccis disco
clypeato subintegerrimo accretis apice exalatis et.am via appendiculatis.

Has. Rio Negro, North Patagonia; in sandy soil.

Root annual or biennial. Stem erect, 2 or 3 feet high, much
branched from the base, paniculate, glabrous, as is the whole plant
except the lower leaves. Cauline leaves trisected, and commonly with
lateral segments two-parted, or the larger biternately divided ; the lobes
laciniate, three-cleft, or entire, and as well as the segments narrowly
linear, sparsely beset, as is the petiole, with small and spreading
hispid-stellate hairs. The leaves of the branches and branchlets are
for the most part glabrous, simply three-divided, and the filiform-linear
segments entire; the uppermost reduced to simple filiform or subulate
bracts, subtending the loosely racemose-paniculate pedicels. The
latter are an inch or less in length, and obscurely articulated near the
apex, one-flowered. Flowers small, 2 or 3 lines long. Calyx naked,
glabrous, deeply five-cleft; the lobes ovate and obtuse. Corolla twice
the length of the calyx, apparently pale purple. Stamens, style, &c.,

42

166 PHANEROGAMIA.

as in the genus. Capsule conoidal, with a somewhat truncate and
umbilicate apex, about 3 lines high and. 3 lines broad at the base,
where it is girt by a saucer-shaped, or at length flat and shield-shaped,
membranaceous disk (larger than the calyx), to which the carpels are
adherent, and from which they tardily separate at maturity. The
scarious circular disk, which thus persists after the carpels have
fallen, is entire, or very obscurely crenulate, often margined with 12
to 16 short setaceous processes, which are a portion of a dorsal nerve
of each carpel, torn away at the separation: in the centre it bears the
columnar persistent axis, which is moderately dilated and umbilicate
at the apex. Carpels 12 to 16, separable, and at length falling away
entire, flatly compressed, broadly semiovate (the straight side ex-
ternal), membranaceous in texture; the sides reticulated, acutely two-
margined all round; the ventral edge bilamellar and scarious; the
back moderately sulcate; the apex wingless, barely pointed, rather
than appendaged, with two very small and obtuse flat processes or
teeth: the dehiscence is from the apex between the teeth, extending
downward first along the ventral, then along the dorsal suture, at
length two-valved. Seed resupinate-suspended, triangular-obovate,
compressed, conformed to the cell, which it fills. Radicle superior,
straight: cotyledons incumbent and folded.

The crest, or pair of apical wings, from which the genus Cristaria
takes its name, varies greatly in size, compared with the body of the
carpel, in different species. In the present plant it is reduced toa
minimum, or is, in fact, obsolete. Yet I do not hesitate to refer it to
this genus, with which it otherwise accords in structure, and espe-
cially in the remarkable scarious disk.

5 GAYA, ABE.

1. GAYA SUBTRILOBA, H. B. K.
Gaya subtriloba, H. B. K. Nov. Gen. & Spec. 5, p. 270, t. 476.
Has. Obrajillo, Peru.

Our specimen, a miserable one, accords well with No. 3236 of Mat-

MALVACES. 167

thews, collected at Chachapoyas, and likewise with the character and
figure given by Kunth.

6. ABUTILON, Garin.
1. ABuTILON GLAUCUM, Webb.

Abutilon glaucum, Webb, Spicil. Gorg. in Hook. Niger Flora, p. 109.
Sida glauca, Cav. Ic. 1, p. 8, t. 11; DC. Prodr. 1, p. 471.

Has. St. Jago, Cape de Verde Islands.
The synonymy and geographical distribution of this species are
fully elaborated by Mr. Webb, in the work above cited.
2. AputTILoN Inpicum, Don,

Has. Luzon, near Manilla and Banos: common in waste places.

3. ABUTILON CARNEUM, St. Hil,
Abutilon carneum, St. Hil. Fl. Bras. Mer. 1, p. 205.

Has. In the vicinity of Rio Janeiro, Brazil.

4, ABUTILON RUFINERVE, St. Ail.
Abutilon rufinerve, St. Hil. Fl. Bras. Mer. 1, p. 205, t. 42.

Has. In the vicinity of Rio Janeiro, Brazil. (In fruit only.)

5. ABUTILON ESCULENTUM, St. Hil. 1. ¢ ?

Has. Near Rio Janeiro, Brazil. (In fruit only.)

168 PHANEROGAMIA.

6. ABUTILON MOLLISSIMUM, Don.
Sida mollissima, Cay. Diss. 2, p. 49, t. 14, f. 1; DC. Prodr. 1, p. 470.

Has. Yanga, near Lima, Peru. (Also gathered at Lima, by Mr.
Cuming.)

7, ABUTILON INcANUM, Don.
Sida incana, Link. Enum. Hort. Berol. 2, p. 204; DC. Prodr. 1, p. 468.

Has. Oahu, Sandwich Islands; on Diamond Hill, and likewise on
the coast near Honolulu. (Also gathered by Chamisso, Macrae, and
Nuttall.)

Plant low, a foot or so in height, clothed all over with a very fine
and close velvety and canescent tomentum. Stems slender, diffusely
branched from a woody or suffrutescent base, leafy. Stipules very
small, filiform, deciduous. Leaves cordate-ovate, acuminate or acute,
or those of the sterile branchlets rounded-cordate, finely serrate,
almost equally canescent both sides with the very soft tomentum;
the cauline an inch to an inch and a half long, and on petioles of
fully half their length; those of the flowering branches smaller.
Peduncles mostly axillary and solitary, exceeding the leaves, about
an inch long, articulated near the apex. Flower 3 lines in length.
Calyx canescent, five-cleft to the middle, about half the length of the
yellow corolla, much shorter than the capsule, spreading in fruit.
Capsule cinereous-tomentose, short-oblong, pentacarpellary, strongly
five-lobed, truncate at the apex; the carpels barely mucronulate at
their obtuse tips, dehiscent at the apex and down the dorsal suture,
three-seeded. Seeds globular, pubescent, superposed.

This species much resembles the North American Abutilon Texense ;
but the leaves are clothed with a still softer and whiter indumentum,
the flowers are smaller, the carpels are only five, and the seeds are
minutely downy.

MALVACES. | 169

8. ABUTILON TUBULOSUM, Hook.
Abutilon tubulosum, Hook. in Mitch. Jour. Trop. Austral. p. 368, adn.

Has. Hunter’s River, New South Wales. (A fragment only was
gathered.)

7. URENA, Linn.
1. Urena topata, Linn. &

Var. RHOMBIFOLIA: foliis basi 3-5-nerviis haud cordatis ovato-rhombeis
oblongisve integris subtus cano-tomentosis ; fructu longiuscule echinato.

Urena monopetala, Lour. Fl. Coch. ed. Willd. 2, p. 508?

Var. SCABRIUSCULA : foltis subrotundis 7T-9-nerviis acutiuscule 3—5-lobatis
scabriusculis subtus albo-tomentosis, nervis cum petiolo ramisque hir-
sutis; wMvolucell hispidi segmentis linearibus calycem superantibus ;
Fructu longiuscule echinato.

Urena scabriuscula, DC. Prodr. 1, p. 441; Wight & Arn. Prodr. Ind. Or. 1, p. 46.
U. lobata, 8. tomentosa, Walp. Rel. Meyen. p. 304.
U. lobata, y. Hook. f. Fl. Nigrit. p. 226.

Has. Rio Janeiro, “probably introduced.” Tahiti. Tongatabu.
Samoan or Navigators’ Islands. Feejee Islands.—Var. rhombifolia,
Ovolau, Feejee Islands.—Var. scabriuscula, Mountains near Baiios,
Luzon. (Philippine Islands, Cuming, No. 469.)

Of this widely spread species various forms were collected, princi-
pally with rounded, cordate or ovate-subcordate, and scarcely lobed
leaves, which are either green both sides, and more or less pubescent
underneath, or with the lower surface canescently velvety-tomentose.
In all of them the segments of the involucel are lanceolate and about

the length of the calyx, and the fruit is armed with short glochidate
48

170 PHANEROGAMIA.

prickles. They pass into the first variety, in which the fruit is armed
with rather longer prickles; the leaves white-tomentose underneath,
glabrate above, only 3—5-nerved at the base, which is either acutish or
rounded, not cordate, rhombic-ovate, varying to oblong or ovate-lanceo-
late, not lobed, the larger ones doubly toothed. This might be the U.
viminea of Cavanilles; but the involucel is no longer than the calyx,
and its segments are obtuse.—The second variety has rather larger
flowers and fruit, the latter densely armed with larger prickles; the
leaves are large (3 or 4 inches in diameter) and rounded, subcordate,
7-9-nerved, flabellately 3-5-lobed; the lobes short, coarsely and spar-
ingly toothed, and finely serrulate; the upper surface roughish with
a short stellate pubescence, the lower canescently tomentose, and hirsute
on the ribs and veins; the petioles, branchlets, &c., also hirsute with
spreading hairs. The calyx and involucel are hispid or strongly hir-
sute, and the segments of the latter are narrowly linear, acute, and
longer than the calyx.

2. URENA MORIFOLIA, DC.
Urena morifolia, DC. Prodr. 1, p. 442.
of e . . . . . . . . . ® .
Var. foliis hirsutulis, superioribus sepe indivisis lineari-elongatis.
Has. Feejee Islands; in cultivated ground at Rewa.

This must, I think, belong to DeCandolle’s Urena morifolia, which
came from the Friendly Islands, although the leaves when young are
rather strigose-hirsute than tomentose, and when full grown are spar-
ingly hirsute and scarcely paler underneath. It resembles U. sinuata ;
but the leaves are larger, from 3 to 5 or more inches in length, very
deeply 3—5-lobed; the lobes, at least the three larger ones, obtusely
three-lobed and sinuate above, contracted below into an elongated and
linear base, which is sharply serrate, as are likewise the very wide
sinuses. ‘The uppermost leaves, in the single specimen, are linear-
elongated, something as those of U. reticulata are represented by
Cavanilles, and either undivided, or barely sinuate-toothed, or ob-
scurely 1-3-lobed. Involucel hirsute; the segments as long as the
calyx.

MALVACES. 171

8. MODIOLA, Mench.
1. Mopiona rEprans, St. Hil.

Modiola reptans, St. Hil. Fl. Bras. Mer. 1, p. 212, t. 48.

Has. Rio Negro, North Patagonia ;—according to Dr. Pickering’s
notes. Not found in the collection.

9 SPHAH RALCEA, S&. Mi.

1. SPHARALCEA OBTUSILOBA, Gay.

Spheeralcea obtusiloba, C, Gay, Fl. Chil. 1, p. 291. .
Malva obtusiloba, Hook. Bot. Mag. t. 2787; Hook, & Arn, Bot. Mise. 3, p. 151.

Has. Chili, near Valparaiso.

10. PAVONTA, Caw.
1. PAvonta communis, St. Hil.
Pavonia communis, St. Hil. Fl. Bras. Mer. 1, p. 224.

Has. Organ Mountains, near Rio Janeiro, Brazil.

2. Pavonia sepium, St. Hil. 7. ¢.

Has. Near Rio Janeiro; common.

3. PAVONIA HASTATA, Cav.

Pavonia hastata, Cav. Diss. 3, p. 178, t. 47, f. 2; St. Hil. Fl. Bras. 1, p. 228.

172 PHANEROGAMIA.

Has. Hunter’s River, New South Wales!

The specimens (which are in fruit) perfectly accord with the
Pawonia hastata of Southern Brazil, a species very unlikely to occur
in Australia, at some distance too from the coast. I cannot but sus-
pect some transposition of the tickets, and that the plant may really
have been gathered on the Rio Negro, or at Rio Janeiro: but there is
no mention of such a plant in Dr. Pickering’s notes on the collections
made at these places. No separate memoranda have been furnished
me respecting the Australian collection. The point must be left for
future elucidation.

ll. ABELMOSCHUS, Medik.

1. ABELMOSCHUS MoscHATUS, Meench.

Abelmoschus moschatus, Moench, Meth. p. 616; Wight & Arn. Prodr. Ind. Or. p. 53;
Guillem. Zeph. Tait. p. 72.

Hibiscus Abelmoschus, Liun.; Cav. Diss. 3, t. 62; DC. Prodr. 1, p. 452.

H. Pseudo-Abelmoschus, Blume, Bijdr. p. 70.

Has. Tahiti. Tongatabu. Samoan Islands. Feejee Islands.
Mindanao, Philippine Islands.

This is found only in cultivated grounds throughout the South Sea
Islands, where it has doubtless been widely diffused by the natives,
The leaves vary greatly in the number and depth of their lobes.

2. ABELMOSCHUS ESCULENTUS, Wight & Arn. 1. ¢.

Has. Feejee Islands, &c.; “cultivated by the natives; the leaves
seem to be principally used."—There are no specimens of this, the
well-known Ochra, preserved in the collection.

3. ABELMOSCHUS MAntHoT, Medtk.

Has. Partly naturalized in the Society, Samoan, Tonga, and Feejee

MALVACES. 173

Islands, according to Dr. Pickering. There are no specimens in the
herbarium.

12. HIBISCUS, Linn.

1. Hisiscus (GAMocHLaNA) Sturtit, Hook.

Hibiscus Sturtii, Hook. in Mitch. Jour. Trop. Austral. p. 868, adn.
Has. Hunter’s River, New South Wales.

The turbinate involucel is gamophyllous almost to the summit; on
account of which, and of the oligandrous column, this species appears
to be the type of a distinct section of the genus.

2. H1BiscUs PHYSALOIDES, Guillem. & Perr.
Hibiscus physaloides, Guillem. & Perr. Fl. Seneg. p. 52; Hook. Niger Fl. p. 227.

Has. St. J ago, Cape Verde Islands ;—where it was also gathered
by Dr. Hooker.

3. Hipsiscus Rosa-Sinensis, Linn.

Has. Feejee Islands: along banks of mountain streams at Ovolau,
where it has probably escaped from cultivation. Cultivated for the
beauty of the flowers, near their dwellings, by the natives of the
Feejee, Samoan, and Society Islands.

4. Hisiscus Aituioricus, Linn.

Hibiscus Aithiopicus, Linn. Mant. p. 258; Cav. Diss. 3, t. 61, f. 1; Eckl. & Zeyh.
Enum. Pl. Afr. Austr. p. 38.

Has. Cape of Good Hope, near Cape Town.
44

174 PHANEROGAMIA.

). Hipiscus Trionum, Linn.

Hibiscus vesicarius, Cav. Diss. 8, p. 171, t. 64, f. 2, ex Hook. f. Fl. N. Zeal.
HT. tridactylites, Lind]. in Mitch. Jour. Trop. Austral., & in Ann. Sci. Nat. ser. 2,
15, p. 57.

Has. St. Helena. Woolongong, New South Wales. Indigenous?
6. HrpiscUS HETEROPHYLLUS, Vent.
Mibiscus heterophyllus, Vent. Hort. Malmais. t. 103; DC. Prodr. 1, p. 450.

Has. Newington, New South Wales.

7. HIBISCUS BIFURCATUS, Cav.

Mibiscus bifurcatus, Cav. Diss. 8, p. 146, t. 51, f. 1; DC. Prodr. 1, p. 449; St. Hil.
Fl. Bras. Mer. 1, p. 246.

Has. Rio Janeiro, Brazil.

8. HIBISCUS DIVERSIFOLIUS, Jacq.
Hibiscus diversifolius, Jacq. Ic. Rar. 8, t. 551; DC. Prodr. 1, p. 449; Endl. Prodr.

Fl. Norf. p. 74.
I. ficulneus, Cav. Diss. 8, p. 148, t. 51, f. 2.

Has. Sandal-wood Bay, Feejee Islands; in an open marsh.

A species allied to the Hibiscus scaber, Michx., of the Southern
United States; but I think distinct from it, although the two are
united in Hooker’s Niger Flora.

9. Hisiscus Youneranus, Gaud.

Hibiscus Youngianus, Gaud. Bot. Voy. Freye. p. 91 (abs. char.); Hook. & Arn. Bot.
Beech. Voy. p. 79.

MALVACES. 17

on

Has. Oahu, Sandwich Islands; in marshes, near Honolulu.

The specimens afford nothing of importance to add to.the descrip-
tion given by Hooker and Arnott. The branches, petioles, and lower
surface of the leaves are velvety-canescent; the former armed with
short sete: rather than prickles: these are fragile and at length deci-
duous, leaving a persistent papillose base. The upper leaves are
slightly if at all lobed, and are mostly acute or acuminate. Pedun-
cles half or two-thirds of an inch long, stout, tomentose, and setose-
hispid. Involucel and calyx hispid; the former of 10 or 12 filiform
divisions, all or most of them two-lobed at the apex; the latter
strongly ten-nerved, the lobes triangular-lanceolate, in fruit closed
over the capsule. Corolla “rose-colour;” the petals 3 inches long.
Capsule canescently hispid with appressed bristles, an inch long,
rather shorter than the persistent calyx. Seeds numerous, very
smooth.

10. Hisiscus BRAcKENRIDGEI, Sp. Nov. (Tab. 12.)

H. fruticosus, glabellus; foliis longe petiolatis membranaceis rotundatis
5—T-fidis, sinubus angustis, lobis grosse dentatis; floribus breviter
pedunculatis axillaribus ad apicem caulis confertis ; involucelli phyllis
8 setaceo-subulatis integris rigidis calycem hispidissimum cequantibus ;
capsula sericeo-hispida, loculis 4—5-spermis ; seminibus lepidotis.

Has. On a mountain in the west division of Maui, one of the Sand-
wich Islands.

Stem shrubby (the height not recorded); the flowering branches
rather stout, glabrous, or when young scurfy-puberulent with the stel-
lular pubescence of this family, very leafy; the lower part tuberculate
with the approximated leafscars. Leaves crowded, on long petioles
(the longer petioles 3 or 4 inches in length, and exceeding the blade),
membranaceous in texture, nearly glabrous, rounded in outline and
subcordate, or with an angular sinus at the base, 5—7-ribbed, /five-cleft
to the middle or deeper, or sometimes seven-cleft (when the two basal
lobes are smaller); the lobes separated by acute and very narrow
sinuses, somewhat ovate in form, acute or obtuse, veiny, unequally

176 PHANEROGAMIA.

and coarsely toothed, and often more or less incised; the terminal lobe
little prolonged. The whole shape of the leaf is much like that of
the common Grape-vine: the larger ones are 4 inches, the smaller 2
inches in diameter. Stipules setaceous, caducous. Flowers solitary
im the axils of the leaves at the summit of the branches, where they
are much crowded: the peduncles, or rather pedicels (since they are
articulated at their insertion), only half or a third of an inch in
length, pubescent and sparingly hispid. Leaflets of the involucel 8,
setaceous-subulate, rigid in texture, nearly glabrous, spreading, entire,
adnate to the base of the calyx, 8 to 10 lines long, about the length
of the calyx, persistent, not at all glanduliferous. Calyx very hispid
with fulvous spreading hairs, at least when young, five-cleft to
below the middle, persistent; the short tube ten-ribbed; the lobes
lanceolate, acute, each bearing an oblong dorsal gland on its midrib
near the base. “Corolla yellow,” turning green in drying; the petals
doubtless spreading, pubescent externally, an inch and a half or two
inches long. Stamineal column antheriferous for a great part of its
length, five-toothed at the apex, straight, not longer than the corolla.
Style five-lobed only at the apex; the short lobes hirsute: stigmas
depressed-capitate. Ovary densely villous-hispid, five-celled, with 4
to 6 biseriate ovules in each cell. Capsule ovoid, closely invested by
the calyx and rather shorter than its lobes, two-thirds of an inch long,
silky-hispid with appressed hairs, loculicidally five-valved, five-celled ;
the cells 4—5-seeded. Seeds angled, minutely tomentose at the hilum;
the surface scur/y, with star-shaped scales. Embryo, &c., as in the
genus.

This is a striking and well-marked new species, apparently of the
section Ketmia, which does not resemble any other known to me.

PLATE 12.—Hisiscus BRACKENRIDGEI: in flower and fruit, of the
natural size. Fig. 1. Vertical section, through the column, pistil, &e.,
enlarged. 2. Vertical section of a capsule, with the persistent calyx,
of the natural size. 3. A seed, enlarged. 4. The embryo, more
enlarged.

11. Hisiscus ARNOTTIANUS, Gray.

| HL. fruticosus, glaberrimus ; foliis ovatis seu ovalibus subcoriaceis integer-
rimis (nune subdentatis) basi trinervatis; floribus solitariis pedun-

MALVACE &. 177

culatis ; involucelli phyllis 5—T parvis decidwis; petalis (rubris) ob-
longis basi attenuatis atque in tubum gracilem calyce cylindrico lon-
giorem coalitis; columna staminea preelonga sprthamea; capsula
polysperma.

Hibiscus Arnottianus, Gray, in herb. Hook. anno 1837.
H. Boryanus, Hook. & Arn. Bot. Beech. Voy. p. 79, non DC.

Has. Sandwich Islands; on the Kaala Mountains behind Honolulu,
Oahu; where it was also gathered by Macrae, Lay and Collie, Diell,
Barclay, &e. (Byron’s Bay, Hawaii, Macrae, Diell.)

A shrubby species, several feet in height, glabrous throughout, espe-
cially the leaves. Leaves ovate or oval, mostly obtuse, rounded or
sometimes slightly contracted at the base, where they are three-nerved,
or obscurely five-nerved, chartaceous or somewhat coriaceous in texture,
entire (rarely a little toothed, at least in Hawaiian specimens) green,
and of the same hue both sides, 2+ to 4 inches long; the petioles an
inch or an inch and a half in length. Stipules subulate, caducous.
Flowers solitary, terminating the branches, or in the uppermost axils,
peduncled; the peduncle 6 to 12 lines, or even 2 inches in length,
articulated towards the summit. Involucel very much shorter than
the calyx, of 5 to 7 lanceolate or subulate and entire leaflets, only 2
or 3 lines long, more or less deciduous. Calyx cylindrical or tubular,
two-thirds of an inch in length, membranaceous, not inflated, rather
deeply five-toothed at the apex (the teeth triangular, puberulent
within), glabrous externally, inclining to split down one side with
age, but not in a regular manner. Corolla red or deep rose-colour;
the petals oblong, or narrowly obovate, 24 inches long, apparently
ascending, narrowed at the base, and united into a slender tube, of ©
nearly an inch in length, which is exserted beyond the calyx. Stami-
neal column very long (5 or 6 inches in length), slender, erect, or
perhaps declined, the upper half exserted beyond the corolla and
copiously antheriferous: the filaments capillary, nearly an inch long,
puberulent. Style five-cleft at the apex. Stigmas capitate. Ovary
five-celled, oblong, glabrous. Ovules numerous in each cell, biserial,

glabrous.

This plant was referred by Hooker and Arnott to Hibiscus Bory-

anus, DC.; a native of the Isle of Bourbon, which is described as
45

178 PHANEROGAMIA.

having very short-pedicelled and whitish flowers, with the petals
velvety-pubescent externally, and an involucel rather longer than the
calyx. Noticing these points when I long ago examined a collection
of Sandwich Island plants, made by the late Rev. John Diell, I com-
municated a specimen to the herbarium of Sir William Hooker, under
the name of Hibiscus Arnottianus, which is here adopted. It is a very
showy species, apparently of the section Cremontia; although the
petals are expanded in anthesis. According to Hooker and Arnott,
the seeds are clothed with fulvous hairs. None of the specimens in
the collection of the Expedition show the fruit; but the ovules, after
the corolla has fallen, are glabrous.

18. PARITIUM, Adr. Juss.

1. PAritrum TILIACEUM, Adz. Juss.

Paritium tiliaceum, Adr. Juss. in St. Hil. Fl. Bras. Mer. 1, p. 256; Wight, Ic. Pl.
Ind. Or. t. 7.

Hibiscus tiliaceus, Linn.; Cav. Diss. 8, t. 55; Geertn. Fruct. t. 185; DC. Prodr. 1,
p. 454.

Has. Society, Samoan, Tonga, Feejee, and Sandwich Islands.
Luzon, near Manilla. Rio Janeiro.

A small tree. “In the deep interior woods of Savaii, Samoan
Islands, it attains the height of sixty feet.”

2. PARITIUM TRICUSPIS, Gull.

Paritium tricuspis, Don. Syst. Bot. 1, p. 485; Guill. Zeph. Tait. p. 72, cum descr.
Forst. MS.

Hibiscus tricuspis, Cav. Diss. 8, p. 152, t. 55, f. 2; DC. 1. ¢.; Hook. & Arn. Bot.
Beech. p. 60.

HI, hastatus, Linn. f. Suppl. p. 310; Forst. Prodr. Ins. Austr. p. 49.

Has. Tahiti, Society Islands; near the coast: rather rare.

MALVACE &., 179

14. THESPESIA, Correa.
1. THESPESIA POPULNEA, Correa.

Thespesia populnea, Correa, in Ann. Mus. Par. 9, p. 290, t. 8; DC. Prodr. 1, p. 455.
Hibiscus populneus, Linn.; Cav. Diss. 3, t. 56, f. 1.
HH. bacciferus, Forst. Prodr. Ins. Austr. p. 48.

Has. Society, Samoan, Tonga, and Feejee Islands. (Cultivated in
the Sandwich Islands.) Mangsi Islands.

156. GOSS YPIUM, Linn.
1. GOSSYPIUM RELIGIOSUM, Swartz.

Has. Plains of Waianae and Ewa, Oahu, Sandwich Islands.

2. GOSSYPIUM BARBADENSE, Linn.

Has. Society, Tonga, Samoan, and Feejee Islands. Coast of Oahu,
Sandwich Islands.

3. GOSSYPIUM HERRACEUM, Linn.
Has. Vanua-levu, Feejee Islands.

No one has yet satisfactorily elucidated the species of Cotton. Our
first-mentioned species has finely canescent-tomentulose leaves and
branchlets, and copper-coloured wool: from the latter character it
should be the Gossypium religiosum of Swartz, and the G. crocewm of
Hamilton. The second, with thinner leaves of various outline, has
white cotton and dark seeds; and is therefore G. nigrwm of Hamilton.
The third, with ample and deeply five-lobed leaves, has white cotton
and whitish seeds, and is accordingly the G. album of Hamilton. All
these species were doubtless introduced into the South Sea Islands.

180 PHANEROGAMIA.

16. HOHERIA, A. Cunn.

1. HowertA POPULNEA, A. Cunn.

Hoheria populnea, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 3, p. 319; Hook. Ic.
Pl. t. 565, 566; Hook. f. Fl. N. Zeal. p. 30.
HT. angustifolia, Raoul, Pl. N. Zel. p. 48, t. 26, var. fide Hook. f.

Has. Bay of Islands, New Zealand.

Dr. Hooker enumerates four varieties of this species (including the
A. angustifolia of Raoul), which, thus viewed, is very variable as to
foliage. Our collection contains only the original or typical form;
the fruit of which, I may remark, bears fully as large and long a wing
as that delineated by Raoul, in the work cited above. Mr. Cunning-
ham referred this genus to the Bombacee ; but Sir William Hooker
justly insisted on its close relationship to Sida; and recently Dr.
Hooker has referred both it and Plagianthus to the proper Malvacee.
Here they would fall into the tribe Malvew, in the arrangement which
I formerly sketched ;* where a new subtribe (PLAGIANTHE®) should,
I think, be constituted for their reception, between the Humalvee and
the Side, characterized by the simple pentadelphous column, the capi-
tate or introrse stigmata, and the solitary resupinate-suspended ovules.
This would place them next to Sidalcea, in which the stamens are
similarly united in 5 phalanges, opposite the petals; but that has the
inner series of stamens likewise developed, which does not occur in
any other genuine Malvaceous genus. If the details of Raoul’s plate
above-cited (Fig. 4) were strictly correct, Hoheria would exhibit a
character not previously recognised in Malvacee; namely, a truly

* Genera Flore Am. Bor.-Or. Illustrata, 2, p. 46.—Lawrencia, Hook. is wrongly
placed in the conspectus referred to. At least Lawrencia spicata has introrsely stigma-
tose styles; and therefore, notwithstanding its resupinate-pendulous ovules, the genus
must be referred to the Hwmalvece, next to Napea; unless, indeed, the stamens should
prove to be pentadelphous, which does not appear to be the case in the badly-preserved
flowers I have examined.—Moreover, the name Lawrencia is preoccupied by Laurencia
in the Alga, a genus received by all modern Algologists. This Malvaceous genus may
be transformed by an anagram into Wrenciala.

MALVACE X. 181

suspended ovule with the rhaphe internal. In fact, however, the
rhaphe is dorsal, and the ovule (which is truly anatropous) is accord-
ingly resupinate, as in Sida. There is a similar oversight in the
details of the plate of Dr. Hooker’s Hoheria Lyalliit (Flora of New
Zealand, Plate 11, fig. 4, 6), where the radicle is wrongly represented
by the artist as superior and dorsal._—The introrse, although terminal,
stigmas of this latter plant (those of Hoheria populnea being strictly
and conspicuously terminal, and so represented both by Raoul and Sir
William Hooker), as well as the total want of wings to the fruit,
should doubtless exclude it from Hoheria. Notwithstanding the more
numerous carpels, I scarcely doubt that it is a congener of Sida pul-
chella (Bonpl.*), Hook., and S. Zasmannica, Hook. f.; both of which
are species of Plagianthus (including <Asterotrichion), or at least differ
from P. sidoides only in having one or two more carpels.

17. PLAGIANTHUS, Forsi.

1. PLAGIANTHUS DIVARICATUS, Jorst.

Plagianthus divaricatus, Forst. Char. Gen. t. 43, & Prodr. p. 47; Hook. Bot. Mag.
t. 3271; A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 4, p. 24; Hook. f. Fl. N.
Zeal. p. 29.

Has. Bay of Islands, New Zealand. (In fruit.)

This is the type of a Malvaceous genus, in which the gyneecium is
reduced to the last degree of simplicity, P. divaricatus being mono-
carpellary.—The excellent figure in the Botanical Magazine, above-
cited, represents the stigmas of the sterile flowers: those of the fertile
flowers I find (in cultivated specimens) to be more clavate, or even
capitate. The figure of the Tasmannian Plagianthus sidoides, Hook.,
in the same work, Plate 3396 (the Asterotrichion sidoides of Klotzsch),
apparently represents the sterile flowers alone. In this also the
stigmas of the fertile flowers are more clavate and truncate, and some-

* The carpels of Sida pulchella are said by DeCandolle to be biaristate. (I have not
access to the work in which the species was originally published.) In fruiting specimens
of Gunn’s plant they are muticous.

46

182 PHANEROGAMIA.

times three or even four in number. So in Sida Tasmannica, Hook. f.,
and S. pulchella, Bonpl., Hook., the male fiowers have the styles re-
duced to two, three or four, while those of the fertile flowers are pen-
tacarpellary. As both of them equally have pentadelphous stamens,
as well as the general aspect of P. sidoides, they should manifestly be
referred to the same genus. The ovule, in all of them, is resupinate-
suspended, as in Sida.

Plagianthus betulinus, Cann. (to which Dr. Hooker refers P. urti-
cus, Cunn.), also found at the Bay of Islands, does not occur in our
collection, so far as I am aware.

Orv. STERCULIACE &.

1 ERIODENDRON, DC.
1. ERIODENDRON TRISCHISTANDRUM, Sp. Nov.

E. trunco parce aculeato ; foliolis 5-7 subintegerrimis acuminatis ; petalis
extus sericeo-lanosissimis ; staminibus pentadelphis, phalangibus sin-
gulis é filaments 3 infra medium coalitis apice monanthiferis ; antheris
anfractuosis.

Has. Gathered in a garden at Lima, Peru; the origin not recorded.

Of this remarkable species, separate leaves and fallen corollas, with
the adherent andrcecium, only were gathered, from a tree which is
said in Dr. Pickering’s notes to be “30 feet high, the trunk swelling
in the middle, and armed with a few short spines,” or prickles. The
flowers are said to be “white.” The leaves are glabrous, digitately
0-9-foliolate ; the leaflets oblong, 24 to 34 inches long, entire or ob-
scurely repand-serrulate towards the apex, which is abruptly and con-

STERCULIACEA. 183

spicuously acuminate. They resemble those of Eriodendron leian-
therum, DC., as figured by Cavanilles (under the name of Bombax
erianthos), or those of LE. Samauma of Martius, except that they are
for the most part abruptly acuminate. The petals are from 1% to 2
inches long, oblong-obovate, very thickly clothed on the outside (except
their coalescent bases) with a silky wool, the hairs mostly reversed ;
their inner face glabrous, or a little pubescent towards the apex.
Filaments coriaceous, monadelphous only at the base, where they are
coalescent with the corolla, dividing almost as soon as they become
free into five phalanges, which answer to the five stamens of LE. leian-
therum, E. Samauma, &c. These apparently alternate with the petals.
The phalanges, or united portion of the filaments, are half an inch or
more in length, linear-filiform, each dividing into 3 filaments, of about
the length of the petals, bearing single anfractuose anthers. The
anthers, having all discharged their pollen, are too old to be minutely
investigated. They resemble those of &. Samawma, however, as
figured by Martius, are equally elongated, and apparently fixed by
the middle, and continuously one-celled (the cell of course longitu-
dinally bclocellate by the septum, here more or less persistent, which
normally divides the cavity, at least in an early state); but the turns
are closer and more heaped together, which is probably owing to their
condition after anthesis. They are very unlike those of EF. leian-
therum; the corolla of which appears closely to resemble that of the
present plant, except in its greater size. The style is filiform, longer
than the stamens, and terminated by a capitate, obscurely five-lobed
stigma.

I find no indication of any published species to which this plant
can belong.* Its remarkable peculiarity, it will be seen, consists
in the division (chorisis) of each of the five filaments into three ;—
notwithstanding which I do not hesitate to refer the species to the
genus Hriodendron, modified by the needful extension of the generic
character.

Chorisia speciosa, St. Hil., is mentioned by Dr. Pickering as found

* In the Hookerian herbarium there is a specimen marked ‘ Lima, ex Herb. Ruiz &
Pavon,” and also “Loxa, 805,” which appears to be the same as our plant, and may
therefore throw some light upon the habitat of the species. Unfortunately the stamens
are destroyed by insects, which renders perfect identification impossible,

184 PHANEROGAMIA.

abundantly around Rio Janeiro; but there is no specimen in the
collection.

2 HELICTERES, Linn.
1. Heticteres ovata, Lam.
Helicteres ovata, Lam. Dict. 3, p. 88 (excl. 8. & y.); St. Hil. Fl. Bras. Mer. 1, p. 2738.

Has. Near Rio Janeiro, Brazil; common. (In fruit.)

3. HERETIERA, Dryand.

1. HERETIERA LITTORALIS, Dryand.

Heretiera littoralis, Dryand. in Ait. Hort. Kew. 8, p. 546; DO. Prodr. 1, p. 484; R.
Br. in Horsf. Pl. Jav. Rar. p. 237.

HI. Fomes, Buchanan; Willd. Spec. 4, p. 972; DC. 1. c. ex R. Br.

H. minor, Lam. Dict. 8, p. 229; DC. 1. c. ex R. Br.

Balanopteris Tothila & B. minor, Geertn. Fruct. 2, t. 98, 99.

Has. Sooloo Islands. Feejee Islands. Tongatabu: in lagoons.

Some of the leaves, from both localities, are acuminate and nearly
lanceolate; thus agreeing with the character of the second species
admitted by Mr. Brown, Heretiera lanceolata of Wallich. The large
fruit is sharply carinate at the sutures, rather than winged.

4. STERCULIA, Linn., R. Br.

1. SrercuLiA Ceramica, R. Br.

Sterculia Ceramica, R. Brown, in Horsf. Pl. Jay. Rar. p- 238.

Has. Small island, in the Sooloo Sea.

The flowers are very numerous, in compound panicles. The short

STERCULIACE &. 185

lobes of the calyx are more or less tomentulose inside, and are not
coherent at their apices. The stigmas are subsessile and connate.
Follicles furfuraceous-tomentose, short-stipitate, very turgid, an inch
and a half in length along the straight dorsal suture. Seeds 3 or 4,
cylindrical-oblong, 5 lines in length; the testa smooth and reddish,

except at the apex, where it is furfuraceous. Cotyledons adnate to
the albumen.

5 BRACHYCHITON, Schott & Endl, R. Br.
1. BRACHYCHITON POPULNEUM, FR. Br.

Brachychiton populneum, R. Brown, in Horsf. Pl. Jav. Rar. p. 234.
Pecilodermis populnea, Schott & Endl. Melet. Bot. p. 33.

Has. Hunter's River, New South Wales. (In flower.)

6 FIRMIANA, Marsil.
1. FirMIANA DIVERSIFOLIA, Sp. Nov. (Tab. 13.)

F, foliis cordatis integris vel apice trilobis (quandoque quinquelobis?),
lobis acuminatis ; folliculis apertis oblongo-lanceolatis utrinque obtusis
breviter stiprtatis.

Has. Ovolau (and Vanua-levu?), Feejee Islands.

A tree of considerable size. Leaves cordate, entire, or nearly so, and
acuminate, or three-lobed towards the apex, T-9-ribbed, reticulate-veined,
membranaceous, 6 to 9 inches long and 5 to 8 inches wide, glabrous,
or the under surface, with the branchlets, &c., very minutely stellate-
puberulent under a lens; the lobes short, triangular and acuminate,
the, middle one longer. Petioles 4 or 5 inches long. Flowers not
seen. Fructiferous panicles ample. Common gynophore 3 lines long,

glabrous, peltately dilated at the apex, where it bears 5 carpels on
47

186 PHANEROGAMIA.

slender but short stipes (4 or 5 lines long). The mature follicles are
all open and leaf-like, in the manner of the genus, membranaceous,
very veiny, glabrous or nearly so, oblong-lanceolate, 22 inches long, at
most an inch wide, obtuse at both ends, abruptly tipped with a small
mucroniform style; the edges not thickened towards the base, each
bearing one or 2 seeds. Seed globular, 3 lines in diameter, attached
by a short linear hilum, without any funiculus, semianatropous; the
testa, or outer integument, thin, rather fragile, and wrinkled when
dry, but apparently fleshy in the living state; the inner integument
crustaceous, although rather thin, umbilicate at the chalaza, lined
with a very thin membrane. Albumen fleshy. Embryo as long
and as broad as the albumen, nearly straight; the short radicle a
little bent towards the hilum: cotyledons nearly plane, broadly oval,
foliaceous.

The five-lobed leaves, either of a second species, or more probably of
a variety of this, destitute of flowers or fruit, were gathered at Sandal-
wood Bay, on Vanua-levu, according to Dr. Pickering’s notes. They
are not found in the collection.

PLATE 13,—FIRMIANA DIVERSIFOLIA: a branch in fruit. Fig. 1. A
seed, enlarged. 2. A longitudinal section of the same parallel with
the hilum. 3. A transverse section of the seed. 4. A longitudinal
section parallel with the cotyledons, the breadth of which is displayed.
The analyses magnified.

Orv. BUTTNERIACES.

1 SERINGIA, Gay.
1. SERINGIA PLATYPHYLLA, Gay.

Seringia platyphylla, Gay, in Mem. Mus. Par. 7, p. 442, t, 16,17; DC. Prodr. 1,
p. 488; Steetz, in Pl. Preiss. 2, p. 349,

Has. Near Sydney, New South Wales.

The five ovaries are connivent, so as to appear like a five-lobed
compound ovary; and the five styles, although separable, are mani-
festly coherent for their whole length,

2, THOMASTA, Gay.
1. THOMASIA QUERCIFOLIA, Gay, J. ¢.

Has. Near Sydney, New South Wales. (Without flowers or fruit.)

8 LASIOPETALUM, Smith.
1. LASIOPETALUM FERRUGINEUM, Smith,

Has, Near Sydney, New South Wales,

188 PHANEROGAMIA.

4, RULINGIA, BR. Br.
1. Runner PpANNoSA, R. Br.

Rulingia pannosa, R. Brown, in Bot. Mag. t. 2191; Steetz, in Pl. Preiss. 2, p. 351.
Buttneria pannosa, DC. Prodr. 1, p. 486.
B. dasyphylla, Gay, in Mem. Mus. Par. 1. c. p. 200, t. 12; DC. 1. ¢.

Has. Sydney, New South Wales. (In flower.)

5. COMMERSONIA, Forst.

1. CoMMERSONIA FRASERI, Gay.

Commersonia Fraseri, Gay, in Mem. Mus. Par. 1. c. p. 215, t. 15; Steetz, in PI.
Preiss. 2, p. 359.

Has. Sydney, New South Wales. (In fruit.)

2. COMMERSONIA ECHINATA, Forst.

Commersonia echinata, Forst. Char. Gen. p. 48, t. 22; DC. Prodr. 1, p. 486; Guill.
Zeph. Tait. p. 71.

Has. Tahiti and Eimeo, Society Islands; on mountain ridges.

The specimens accord with the detailed description published by
Guillemin from Forster’s manuscripts. The cordate-lanceolate leaves
are glabrous or merely stellular-scabrous above, and velvety-canescent
underneath. The prickles of the fruit are hirsute-pubescent.

3. COMMERSONIA PLATYPHYLLA, DC. 1. c.

Commersonia Javensis, Don, Syst. Bot. 1, p. 528; Hassk. Pl. Jav. Rar. p. 312.
C. echinata, Blume, Bijdr. p. 86.

BUTTNERIACE S&, 189

Has. Samoan and Feejee Islands. Tahiti (unless a ticket is mis-
placed). Singapore.

The leaves are broader than in the foregoing species, roughish-
pubescent (or at length glabrate) above, whitish-tomentose and more
veiny underneath. Cymes compound and many-flowered. Prickles
of the fruit very villous.

6. AYENTA, Linn.
1. AYENIA TOMENTOSA, Linn.
Has. In the vicinity of Rio Janeiro, Brazil.

This is the same as No. 5372 of Gardner’s Brazilian collection.
The leaves are elliptical or oblong, rather than ovate-subrotund. The
anthers are trilocular, as I had already remarked them to be in Ayenta
pusilla and A. microphylla (Plante Wightiane, 2, p. 24). |

7 KLEINHOVIA, Linn.
1. Kuemnuovia Hosprta, Linn.

Has. Ovolau and Vanua-levu, Feejee Islands. Upolu, Samoan
Islands. Tahiti, Society Islands.

Finding no record of this tree in Dr. Pickering’s botanical notes, I
cannot learn whether it is truly indigenous in all or any of these loca-

lities. It has not previously been recorded from the South Sea
Islands.

8. WALTHERTIA, Linn.
1. WALTHERIA AMERICANA, Linn.

Has. Rio Janeiro. Upolu, Samoan Islands. Vanua-levu and
48

190 PHANEROGAMIA.

Oneata, Feejee Islands. Hawaii and Oahu, Sandwich Islands, on the
coast.

All the specimens from the South Sea Islands, having sessile heads,
belong to the Waltheria Indica of Linnzus, which has been justly
united to W. Americana. Those from Rio and the Sandwich Islands
have the heads either pedunculate or sessile. In this, as in the fol-
lowing species (vid. Cay. Diss, infra cit. f. x.), some flowers have the
androecium less developed than others, their shorter filaments being
monadelphous nearly to the top, while in others they are united only
at the base.

2. WALTHERIA OVATA, Cav.

Waltheria ovata, Cav. Diss. 6, p. 317, t. 171, f. 1; DO. Prodr. 1, p. 498.
Has. Yaso, Cahall, and below Obrajillo, Peru.

Well distinguished by the fine and close velvety tomentum, the
broadly ovate leaves, and the pointless calyx-lobes, which are entirely
destitute of hirsute hairs.—Nearly related to this is Dr. Hooker's W.
reticulata.

3. WALTHERIA PYROLAFOLIA, Sp. Nov.

W. fruticosa ; foliis confertis coriaceis rotundis scepius utrinque retusis
serrulatis plicato-venosis supra glabris subtus canescentibus glabratis ;
glomerulis axillaribus subsessilibus plurifloris; calycis cano-sericei
lobis lanceolatis obtusis petalis glaberrimis brevioribus.

Has. Sand-hills near Wailuku, Maui, Sandwich Islands.

Stems from a foot to 3 feet high, erect, shrubby to the top, much
branched; the branches, petioles, &c., canescently villous-tomentose,
rigid. Leaves crowded on the branches, coriaceous, roundish, inclining
to orbicular-obovate, commonly more or less refuse both at the base
and apex, about an inch in length and breadth (the smaller 8 ‘or 9

BUTTNERIACES. 191

lines long, the largest an inch and a half in length and of nearly the
same width), jinely-toothed, somewhat plicately straight-veined; the
upper surface glabrous, or slightly puberulent when very young; the
lower canescent or cinereous with a fine and close pubescence, minutely
reticulated between the primary veins, becoming almost glabrous (but
pale) when old. Petioles from 3 to 6 lines long. Stipules setaceous,
early deciduous. Flowers crowded in subsessile axillary glomerules or
heads, which are not very dense, or the cluster rarely on a peduncle
which almost equals the petiole, inclined to be secund. Bracts
lanceolate, mostly obtuse, silky-villous like the calyx, which they
equal in length. Calyx canescently silky-villous with fine and soft
hairs, not at all hirsute, 3 lines long, five-cleft scarcely to the middle;
the lobes lanceolate or broadly linear, obtuse or obtusish. Petals entirely
glabrous and beardless, narrowly spatulate, connected only at the very
base, about one-third longer than the calyx. Filaments slender, half
the length of the petals, monadelphous only at the base. Anthers
oblong, two-celled, extrorse. Ovary very villous. Style filiform,
sparsely arachnoid-villous, terminated by a simple and naked truncate
stigma. Fruit utricular, hirsute-villous. Seed and embryo, as in W,
Americana.

This is a more shrubby and branching plant than W. Americana,
from which it is abundantly distinct, also, in its rounded and smooth
leaves, its broader and blunter calyx-teeth, which are not hirsute, and
its smooth petals. JI have seen no other species with which it would
particularly be compared, except Waltheria Lophanthus of Forster,
which has silky-tomentose leaves, and is figured (in Char. Gen.) as
having acute calyx-teeth, and a two-cleft style!

9 MELOCHTA, Linn,
1, Metocuta (RIEDLEIA) CORCHORIFOLIA, Linn.

Melochia corchorifolia, Linn. Spec. p. 675. (Dill. Hlth. p. 221, t. 176, f. 217.)
M. concatenata, Linn. Spee. 1. ¢.; Cay. Diss. 6, t. 175, f. 2.
Riedleia corchorifolia & R. concatenata, DC. Prodr. 1, p. 491, 492,

Has. Singapore. Bajos, Luzon.

192 PHANEROGAMIA.

2. Mrtocuta (VIsENIA) oporAta, Linn. f.

M. foliis ovatis subcordatis serratis glabris vel glabellis ; cymis ramulisque
tomentulosis ; floribus pallide purpureis; capsule hirsute coccis cus-
pidato-apiculatis ; seminibus apice truncatis exalatis.

Melochia odorata, Linn. f. Suppl. p. 302; Forst. Prodr. Fl. Ins. Austr. p. 47; Cav.
Diss. 6, p. 320, t. 173.

Riedleia odorata, DC. Prodr. 1, p. 491, excl. syp. Gmel. & Houtt.

Has. Savaii, one of the Samoan Islands; common. <A specimen
is also ticketed “Tahiti;” perhaps by some transposition, since it is
not alluded to in Dr. Pickering’s notes, nor have any other collectors
met with it in the Society Islands.

“A middle-sized tree.” Leaves ovate, more or less acuminate, sub-
cordate, or the smaller ones truncate or obtuse at the base, finely
serrate, membranaceous in texture, glabrous, except a minute pubes-
cence on the veins, or when young on the surfaces also, 3 to 5 inches
long; the petioles 2 or 3 inches in length, minutely tomentose.
Cymes axillary, compound, many-flowered, on peduncles longer than
the petioles, which, like the branchlets, calyx, &c., are minutely tomen-
tose. Petals spatulate, about twice the length of the calyx, “pale
purple.” Stamens nearly as long as the petals: filaments monadel-
phous only near the base, dilated or membranaceously winged for
one-third of their length, and nearly the whole dilated portion adnate
to the contracted base of the petal; the free upper portion filiform :
anthers oblong, two-celled. Ovary five-lobed, five-celled, very hirsute :
styles shorter than the stamens, hirsute towards the base. (Other
individuals apparently produce shorter filaments and longer styles.)
Capsule hirsute, ovoid, 3 or 4 lines long, deeply five-lobed, the 5
carpels cuspidate-pointed with the short and persistent base of the
styles, partially septicidal ; the carpels opening from the apex by both
sutures. Seeds solitary or in pairs (superposed), a line and a half
long, gibbous, with the smoothish testa conformed to the nucleus
except at the narrowly membranaceous rhaphe, and at the chalaza,
which is truncate and somewhat cellular-appendaged or thickened,

BUTTNERIACEA. 193

but entirely wingless: the inner integument crustaceous. Embryo
nearly straight.

The specimens accord with those collected by Mr. Barclay at
Tanna, one of the habitats given by Forster—This species and the
following plainly show that Visenia, Houtt. (the Glossospermum of
Wallich, and Alewrodendron of Reinwardt), cannot be generically dis-
tinguished from Melochia, and that even as a section it passes into
Riedleva.

3. Metocur1a (VISENIA) ARISTATA, Sp. Nov.

M. capsulis hirsutissimis, coccis acuminato-aristatis ; seminibus subulato-
apiculatis ; ccet. fere proecedentis.

Has. Upolu, one of the Navigators’ or Samoan Islands.

This may prove to be no more than a variety of the preceding
species, with which it nearly accords in foliage, &.: but the very hir-
sute capsules are larger, and more conspicuously awned (the hirsute
awns about 2 lines long); and the seeds bear a conspicuous subulate
appendage at the apex, in place of the broad wing of the succeeding
species. The specimen is in fruit, with only the remains of the flower
at the base of the capsule. The filaments appear to be dilated and
membranaceous for their whole length.

4. Metocutia (VisEnrA) ViTIENSIS, Sp. Nov.

M. foliis cordatis ovatis oblongisve glabris serratis; cymis pedunculisque
tomentoso-puberulis ; petalis flavis; capsule tomentoso-sericece coccis
mucronatis ; seminibus apice late alatis.

Var. 3. foliis subcoriaceis haud cordatis, superioribus basi acutis.

Has. Vanua-levu, Somu-somu, Ovolau, Oneata; and var. 3. Muthu-
ata, Feejee Islands. “A common shrub on the leeward coasts.”

This is a variable species as to foliage, &c., unless possibly two or

more are here confounded. The specimens from Ovolau, Oneata,
49

194 PHANEROGAMIA.

and Vanua-levu very much resemble those of M. odorata, and have
mostly membranaceous leaves, either ovate or oblong-ovate, and com-
monly subcordate. But, according to Dr. Pickering, the petals are
yellow ; the capsules are silky-tomentose, instead of hirsute, and merely
mucronate with the persistent base of the styles; and the seeds are
conspicuously winged at the apex; the broad and scarious wing being
as long as or longer than the body of the seed. One fruiting specimen
exhibits much larger and subulate-pointed capsules, which, however,
are found on inspection to be in a diseased state. The speclinens
from Somu-somu have broadly ovate and truly cordate leaves, of a
chartaceous texture. Those from Muthuata, here appended, with
some misgiving, as a variety, have the leaves almost coriaceous in
texture, very glabrous, broadly or narrowly oblong, or the uppermost
lanceolate, acute at the base, none of them cordate ; but the specimens
do not exhibit many cauline leaves. The flowers also appear to be
rather larger.

10. PTEROSPERMUM, Schreb,
1. PrEROSPERMUM DIVERSIFOLIUM, Blume ?

Has. Shores of Laguna, Bafios, Luzon. — Leafy branches only,
without flowers or fruit.

ll. MELHANITA, Forsh.
1. MELHANIA Leprievru, Webd.

Melhania Leprieurii, Webb, Spic. Gorgon. p. 111, t. 4,5; Hook. Ic. Pl. t. 763, 755.
Brotera Leprieurii, Guillem. & Perr. Fl. Seneg. p. 85.

Has. St. Jago, Cape de Verde Islands.

Orv. TILIACE &.

1. ENTELEA, R. By.

1. ENTELEA ARBORESCENS, R. Br.

Entelea arborescens, R. Br. in Bot. Mag. t. 2480; Hook. f. Fl. N. Zeal. p. 82.
Apeiba australis, A. Rich, Fl. Nouv. Zel. (Voy. Astrolabe) p. 301, t. 34.

Has. Bay of Islands, New Zealand. (In fruit only.)

2 CORCHORUS, Linn,

1. CorcHorvs TortiPEs, St. Hil.
Corchorus tortipes, St. Hil. Fl. Bras. Mer. 1, p. 281, t. 55.

Has. Rio Janeiro, Brazil.

2. CORCHORUS TRILOCULARIS, Linn,

Has. St. Jago, Cape de Verde Islands.

3. CORCHORUS OLIToRIUS, Linn.

Has. Shores of Laguna, Bafios, Luzon, Philippine Islands.

196 PHANEROGAMIA.

4. CorcHorus AntiIcHORUS, Roeuschel.

Antichorus depressus, Linn. Mant. p. 64; Linn. f. Fase. Pl. Rar. Hort. Ups. p. 8, t.
2; DC. Prodr. 1, p. 504.

Has. Cape de Verde Islands.

5. CORCHORUS TRIDENS, Linn.
Has. Cape de Verde Islands; where it was also gathered by the
late Dr. Vogel.
6. CoRCHORUS CAPSULARIS, Linn.

Has. Shores of Laguna, Baiios, Luzon. (In fruit.)

3 TRIUMFETTA, Plum.

1. TRIUMFETTA ERIOCARPA, St. Hil.

Triumfetta eriocarpa, St. Hil. Fl. Bras. Mer. 1, p- 288.

Has. Rio Janeiro, and in the Organ Mountains, Brazil.

2. TRIUMFETTA oBscuRA, St. Hil. 1. ¢.

Has. With the foregoing. (Not in flower.)

3, 'TRIUMFETTA RHOMBOIDEA, Jacq.

Triumfetta rhomboidea, Jacq. Stirp. Amer. p. 147, t. 90? Lindl. Collect. Bot. t. 29.

TILIACES. 197

Has. Near Obrajillo, in the lower Andes of Peru.

This is manifestly the plant figured by Lindley: but it may not be
the same as the Caribbean plant.

4. TRIUMFETTA ANNUA, Linn.

Triumfetta annua, Linn.; Bot. Mag. t. 2296; Walp. Rel. Meyen. p. 310.
T. Indica, Lam. Dict. 3, p. 420? DC. Prodr. 1, p. 508?
T. trilocularis, “‘ Roxb. ex Horn. Suppl. p. 140;” DC. 1. «.

Has. Vicinity of Manilla, Luzon. (The same as Cuming’s No.
1462, from the Philippine Islands.)

5. 'TRIUMFETTA PROCUMBENS, Forst.

Triumfetta procumbens, Forst. Prodr. Fl. Ins. Austr. p. 85; Hook. & Arn. Bot.
Beech. Voy. p. 60; Guillem. Zeph. Tait. p. 71.
f. Fabreana, Gaud. Bot. Voy. Freyc. p. 478, t. 102.

Has. Eimeo and Metia, Society Islands. Savaii and Manua,
Samoan Islands. Ovolau, Feejee Islands. Rurik, Gardner's, and
Binney’s Islands. A maritime species.

The MS. description of Forster, now printed by Guillemin, with
the character published by Hooker and Arnott, above-cited, give a
good account of this widely diffused South Sea species, to which the
T. Fabreana of Gaudichaud doubtless also belongs. In character it is
intermediate between DeCandolle’s two sections. The globose and
strongly echinate fruit varies considerably in size.

4. GREWTIA, Juss.

1. Grewia Mattococca, Linn. f.

Mallococca crenata, Forst. Char. Gen. p. 77, t. 39, & Nov. Act. Ups. 3, p. 185.
Grewia Mallococca, Linn. f. Suppl. p. 409; Forst. Prodr. p. 62; Cav. Ic. t. 309;

Guillem. Zeph. Tait. p. 70, cum descr. Forst.
50

198 PHANEROGAMIA.

G. tilicefolia, A. Rich. Sert. Astrolab. introd. p. 9, non Vahl.
G. Richardiana, Walp. Repert. Bot. 1, p. 363.
G. Amicorum, Steud. Nomencl. Bot.

Has. Tongatabu, Friendly Islands. Vanua-levu, Feejee Islands.

The petals are ovate-lanceolate and acute, the circular nectary at
their base surrounded by a dense villous ring; the nearly entire margin
of the turbinate disk or torus equally villous. Stigma two-parted,
the divisions somewhat two-lobed or laciniate. The leaves vary from
ovate-oblong to oblong-lanceolate, and are barely subcordate; their
upper surface is either slightly scabrous or smooth. The margins of
the sepals, as in the following species, are induplicate in estivation.

2. GREWIA PERSICaFOLIA, Sp. Nov.

G. glabra; foliis oblongo-lanceolatis acuminatis subserratis basi obtusa
tantum trinerviis laxe venosis membranaceis; pedunculis axillaribus
solitarts petiolo sublongioribus bi-trifloris ; pedicellis pedunculo cequi-
longis ; petalis minimis? toro apice villoso-barbato ; fructu hirsutulo.

Has. Ovolau, Feejee Islands.

Shrub glabrous, or the branchlets, petioles, &c., slightly puberulent.
Leaves oblong-lanceolate, or some of them oblong, from 3 to 6 inches in
length, and one or 2 inches in width, acuminate, usually tapering gra-
dually from above the middle to the apex, finely and mostly obscurely or
repandly serrate with obtuse teeth, thin and membranaceous in texture,
of the same green hue both sides, pinnately veined with 4 or 5 pri-
mary veins on each side, besides a rather stronger pair at the obtuse
base, therefore three-nerved at the base only; the veins connected by
rather loose and inconspicuous transverse and reticulated veinlets.
Petioles 4 to 6 lines long. Peduncles axillary and solitary, half an
inch long, slender, 2-3-/lowered ; the pedicels as long as the peduncle ;
the bractlets, if any, caducous. The sepals, petals, &., have fallen
or have been destroyed. There are vestiges, however, of linear and
three-nerved sepals, smaller than those of G. Mallococca, and of very
small? petals. The top-shaped torus‘is villous-bearded at the undulate

TILIACE &. 199

margin, but less so than in the foregoing species. Ovary densely
hirsute-villous. Stigma not seen. Fruit sparsely hairy, much as in
G. Mallococea, four-lobed.

Manifestly related to Grewia glabra, Blume, or at least to No. 1016
of Zollinger’s Java collection, which accords with Blume’s character ;
but that has more decidedly three-nerved leaves, with more conspi-
cuous transverse veinlets, and less hairy fruit. G. levigata, Vahl.
has glabrous fruit and longer peduncles.

3. GREWIA PRUNIFOLIA, Sp. Nov.

G. glabra; foliis ovato-oblongis sew oblongo-lanceolatis acuminatis serru-
latis bast rotundata trinerviis supra nitidis; pedunculis axiltaribus
uni—trifloris ; petalis ovalibus obtusis, nectarto linea pubescente cincto ;
toro apice pubescente ; stigmate crasso peltato; fructu hirsutulo.

Has. Ovolau and Muthuata, Feejee Islands; “‘a common shrub on
the leeward coasts.”

Shrub glabrous, except a minute pilose pubescence on the slender
branchlets, petioles, &c. Leaves ovate-oblong, ovate, or on young
branches oblong-lanceolate, acuminate, from one to 2 or 3 inches in
length, therefore small for this genus, finely and rather sharply serru-
late, chartaceous rather than membranaceous in texture, three-nerved
barely at the rounded base, and loosely pinnately veined, as in the pre-
ceding species, glabrous on both sides, shining above. Petioles 3 or 4
lines long. Peduncles axillary, 1-3-flowered, short. Bractlets minute
and caducous. Flowers smaller than those of G. Mallococca. Sepals
canescent externally, obscurely five-nerved. Petals oval, obtuse, about
a line and a half long, shorter than the stamens, canescent externally
towards the base, the lower half occupied by the roundish nectary,
which is surrounded by a finely pubescent line. Torus turbinate, finely
pubescent at its apex. Ovary hirsute. Style thickish, columnar, ter-
minated by a large and thick peltate stigma, which appears to be
entire, but when macerated is seen to be fringed or dissected into
many minute and densely matted lobes. Fruit smaller than in G.
Mallococea, sparsely hirsute, four-lobed.

200 PHANEROGAMIA,

From Fulanga, Feejee Islands, there is a single fruiting specimen,
possibly of a different species, with larger, crenately toothed leaves.

59. DICLIDOCARPUS, Nov. Gen.

Flores polygamo-dioici? Calyx tribracteolatus, pentaphyllus; sepalis
crassis cestivatione valvatis. Petala 5, inappendiculata, cestivatione
wmbricata. Discus hypogynus, annularis, crenatus. Stamina creber-
ruma, distincta (fl. fert. pauciuscula effocta?): anther biloculares.
Ovarium sessile, oblongum, biloculare, pilis parcis cireumdatum, stig-
mate sessilr retuso coronatum, multiovulatum, fl. masc. effeetum seepis-
sime exovulatum. Capsula latissime obcordato-rhomboidea, bilocularis,
disseprmento contrarie compressa, marginibus alata, ab apice loculicide
bivalvis, polysperma. Semina lenticulari-globosa, hine impressa, mar-
gine pilis prelongis crinita; testa fragili laxa. Embryo albumine
carnoso via brevior ; cotyledonibus orbiculatis planis radicula cequi-
longis.— Arbor, foliis ovalibus integerrimis, stipulis caducis, floribus
m cymulis axillaribus parvis.

1, Diciiocarpus Ricuu, Sp. Nov. (Tab. 14.)
Has. Ovolau, and Vanua-levu, at Sandal-wood Bay, Feejee Islands.

“Tree about 40 feet high.” Branchlets, &., pulverulent with a
minute stellular pubescence, terete, annulate with the scars of the
fallen stipules. Leaves alternate, oval, entire, obtuse, or usually with
a minute and abrupt acumination, rounded or rarely slightly subcor-
date at the base, 2 to 33 inches long, and 13 to 23 inches wide, char-
taceous, prominently feather-veined, three-ribbed at the base, the
principal veins connected by transverse veinlets, green, scarcely paler
underneath, glabrous to the naked eye, under a lens finely dotted
with a minute stellular pubescence, especially beneath and on the
midrib and veins; the latter more or less bearded in their axils.
Petioles about half an inch long. Stipules oblong, acute, caducous.
Flowers polygamo-diacious, or perhaps monceciously polygamous, very
small, crowded in small axillary cymes; the inflorescence and exte-
rior of the calyx tomentose-puberulent. Sterile jlowers. Peduncle

TILIACE A. 901

shorter than the petiole, twice or thrice dichotomous; the forks sub-
tended by a pair of oblong bracts. Pedicels articulated, not longer
than the calyx. Calyx calyculate with an involucel of 3 bractlets, glo-
bose in the bud; the sepals very thick and coriaceous, triangular-ovate,
valvate in sestivation, deciduous. Letals 5, not appendiculate, hypo-
gynous, broadly obovate, minutely veiny (only seen in the buds), im-
bricated in estivation. Stamens very numerous, crowded in a globular
mass, inserted on the flat summit of a depressed receptacle, which
is not at all elevated, but its margin is slightly produced into an
obscurely five-crenulate border or hypogynous disk : filaments destinct,
not longer than the ovary: anthers two-celled, linear-oblong ; the cells
opening longitudinally. Ovary sterile, oblong, sessile, glabrous, some-
what compressed, surrounded by some long hirsute hairs, which are
more or less intermixed with the inner stamens (nearly as in Jun-
tingia), and are almost as long as their filaments, two-celled, destitute
of ovules: style none: stigma sessile, retuse or two-lobed. Fertile
flowers scarcely known: a single apparently hermaphrodite flower
exhibited a nearly similar calyx, corolla, &., fewer stamens, and a
broader ovary, with the cells containing many minute ovules. The
truly fertile, pistillate flowers probably unknown. Inflorescence of
the fruiting specimen nearly as in the sterile, but with longer pedun-
cles. Capsule dilated-rhomboidal and emarginate, or somewhat obcor-
date, two-celled, compressed contrary to the partition, puberulent, between
ligneous and coriaceous in texture; the rather turgid body of the fruit
extended all round, except at the base, into a firm wing of 3 or 4 lines
in width: it is loculicidully dehiscent at maturity through the wing
from the apex downwards, thus opening like a bivalve shell. The
length of the pod is about an inch; the width about one-fourth larger.
Seeds very numerous in each cell, horizontal, lenticular-globose, a line
broad, on a short funiculus, which separates by a clean scar from the
placenta, and remains attached to the excised or subcordate hilum :
the testa thin and fragile, cellular, in the dried specimens loose, and
as it were inflated and detached from the smooth and ovoid crusta-
ceous tegmen, except on the dorsal face of the seed, where it is
reduced to an adnate pellicle, or apparently disappears abruptly,
leaving a broadly oval depression on that side (fig. 18, 20): 1t is
naked on both sides, but the rounded edges are throughout beset wath
a continuous tuft of very long and soft hairs. Embryo axile, straight,

nearly as long as the fleshy albumen ; the cotyledons flat, nearly orbi-
51

202 PHANEROGAMIA.

cular, thickish: radicle about the length of the cotyledons, pointing
to the hilum.

The pod of this remarkable new Tiliaceous genus bears no slight
resemblance to some bivalve shell; which has suggested the name,
composed of dais, a double door, and xapric, fruit. The specific name
is in honour of the Botanist of the Expedition, Mr. Rich, who had
indicated it as a new genus. The drawings from which the engraving
is executed, were in part made under his superintendence, by the
lamented Mr. Agate.

Puate 14.—Diciipocarpus Ricuu. Fig. 1. Branchlet of the sterile
plant with flower buds, of the natural size. 2. A fruit-bearing
branchlet, of the natural size. 8. Portion of the lower surface of a
leaf, magnified, to show the minute stellular pubescence. 4. A
cymule from fig. 1, magnified. 5. Diagram (transverse section of the
flower with its three bractlets). 6. Vertical section of a flower-bud
of a sterile flower. 7. The calyculus of three bractlets, detached. 8.
Flower-bud with the calyculus and calyx removed, one of the sepals
also shown. 9. A petal from the same. 10. The globular cluster of
stamens, seen after the petals are removed. 11. Vertical section of
the pistil of a perfect? flower, and the receptacle, with two of the
stamens and some of the hairs that surround the ovary left in place.
12. A (fertile?) pistil, with the receptacle and some of the hypo-
gynous hairs. 13. A sterile pistil, &¢.—AlIl these details variously
magnified. 14. A dehiscent capsule, of the natural size. 15. Trans-
verse section of a capsule. 16. One of the valves of the capsule,
seen from within, with the seeds. 17. A seed with its persistent
funiculus seen on the ventral side, enlarged. 18. Dorsal view of the
same. 19. Vertical section of the same. 20. A seed transversely
divided. 21. The embryo, magnified.

6 ELAOCARPUS, Linn.

1. Hizocarpus (Monocera) cyaneus, Sims.

Lileocarpus cyancus, Sims, Bot. Mag. t. 1737; DC. Prodr. 1, p. 519.
E. reticulatus, Smith; Ker. Bot. Reg. t. 657.

TILIACES. 903

Has. Vicinity of Newington, New South Wales. (In flower.)

2. Enmocarpus (MOoNOCERA) PETIOLATUS.
Monocera petiolata, Jack, Malay. Misc. & in Hook. Bot. Mise. 2, p. 86.
Var, 8. petiolis brevioribus (haud ultra bipollicaribus).

Has. Singapore. (A form with shorter petioles than ordinary.)

3. Enaocarpus (Monocera) Grirritai, Wight.

Monocera Grifithii, Wight, Ill. Ind. Bot. 1, p. 84; Walp. Repert. 1, p. 365.
Monocera paniculata, Wall. Cat,

Has, Singapore.

4, ELZOCARPUS (ACRONODIA?) LAURIFOLIUS, Sp. Nov.

E. glaber ; foliis coriaceis oblongis subserratis utrinque acutis vel subacu-
minatis supra nitidis; racemis axillaribus brevibus paucifloris ; flori-
bus diclinis, masculis perianthio pentamero ; petalis lacimatis ; antheris
via apiculatis plurimis,

Has. Feejee Islands.

The single specimen of this plant which exists in the collection
(without any indication of its particular locality) is a stout branchlet
of a shrub or tree, entirely glabrous, with some short and few-flowered
racemes from the axils of the leaves of the preceding year (which have
principally fallen), bearing unexpanded flowers, of which those exa-
mined are staminate only, showing no trace of a pistil. From its dicl-
nous flowers it may he allied to Blume’s genus Acronodia, which this
author describes as dicecious, and of which he knew only the stami-
nate plant (although, indeed, Hasskarl, in Pl. Jav. Rar. p. 323, has
recently characterized the flowers as hermaphrodite) : but, otherwise,
it appears not different from Hlaocarpus. The leaves are crowded on

204 PHANEROGAMIA.

the branch, alternate, coriaceous, oblong, 2 to nearly 4 inches long, an
inch or more in width, remotely and rather obscurely serrate, acute or
tapering to both ends, the base truly acute, the apex acuminate with a
blunt point, pinnately veined, and with reticulated veinlets, paler
beneath, shining above. Petioles half an inch long. Racemes simple;
the peduncle longer than the petiole. Pedicels 3 lines long. Flower-
buds globular-ovoid. Sepals 5. Petals 5, small, ovate, laciniate into
5 or 7 acute lobes, glabrous (except a few minute silky hairs exter-
nally); their margins induplicate in sstivation and folded round a
cluster of the stamens. Receptacle pilose. Stamens more than 20:
filaments very short: anthers linear, sparsely pilose with long hairs,
slightly apiculate or mucronate. Lying loose with the specimen is an
oval dry drupe, of 4 or 5 lines in length, one-celled, one-seeded, its
surface pubescent under a lens.

0. ELMOCARPUS CASSINOIDES, Sp. Nov.

E. glaber ; foliis coriaceo-chartaceis obovatis denticulatis leviter penni-
nervous; racemis axillaribus; drupa pyriformt.

Has. Tonga or Friendly Islands. Sandal-wood Bay, Feejee Islands.

The specimens of this shrub or tree bear ripe fruit only: the flowers
are unknown. Those from the two localities, as ticketed, are so
exactly alike that they might have been taken from the same stem,
and, since the habitats are not to be verified from Dr. Pickering’s
notes, one or the other may be considered doubtful. The branches,
foliage, &., are perfectly glabrous. Leaves between coriaceous and
chartaceous in texture, very smooth, obovate, obtuse, about 2 inches
long and an inch and a quarter wide, sparingly denticulate or serru-
late above the middle, tapering at the base into a petiole of 3 or 4
lines in length, both sides coloured alike, pinnately veined with 8 or 4
pairs of slender and inconspicuous veins. Racemes (or perhaps short
and leafy flower-branches) azillary from the axils of the leaves of the
preceding year, apparently as long as the leaves, several-flowered.
Drupe pear-shaped, 7 lines long, on pedicels of 2 lines in length,
smooth, with a thin pulp, apparently blue or purple when fresh : puta-
men very thick, bony, one-celled, one-seeded. Seed and embryo as in
the genus.

TILIACE &. 205

6. ELZOCARPUS PYRiFORMIS, Sp. Noy.

E. subglaber ; foliis membranaceis ellipticis utrinque obtusissimis serru-
latis perspicue penninervis ; racemis axillaribus ; drupa pyriformi.

Has. Sandal-wood Bay, Vanua-levu, Feejee Islands.

“Tree with a rough bark.” Branchlets and petioles minutely
pubescent. Leaves crowded, alternate, glabrous, or when young
minutely puberulent on the prominent midrib, membranaceous, nearly
of the same hue both sides, elliptical, rounded and very obtuse at both
ends, rather sharply and copiously ser7wlate quite to the base, 2 to 22
inches long, about an inch and a half wide, prominently feather-veined ;
the primary veins 5 or 6 pairs, slender but salient underneath, occa-
sionally with a gland in their axils, and more commonly with one in
their forks towards the margin of the leaf. Petioles 12 to 3 lines
long. Flowers unknown; the specimens bearing only ripe fruit: they
were evidently borne in short axillary racemes, from axils of the leaves
of the preceding year. Drupes pear-shaped, 6 to 8 lines long, on pedi-
cels of barely 2 lines in length, nearly smooth, with a thin pulp, and
a very thick and bony putamen, one-celled, one-seeded. Seed and
embryo as in the genus.

This is nearly related to the foregoing species, and has a similar
fruit; but the leaves are different. Both must rank as species of
Eleocarpus until the flowers are known.

7. E.aocarpus (BeytHEA) Biripus, Hook. & Arn.

Eleocarpus bifidus, Hook. & Arn. Bot. Beech. Voy. p. 110, t. 24.
Beythea bifida, Endl.; Walp. Repert. 1, p. 365, & 5, p. 121.

Has. Kaala Mountains, behind Honolulu, Oahu, Sandwich Islands.
(Also gathered by Macrae, Lay & Collie, Nuttall, &c.)

Following Arnott’s suggestion, Endlicher established a separate
52

206 PHANEROGAMIA.

genus for this species; but apparently on insufficient grounds. All
gradations are now known to occur in species of Hlaocarpus between
laciniate-dissected, 2-3-lobed, and even quite entire petals. The
ovules, indeed, are said by Endlicher to be numerous; but there are
only 4, or possibly 6, in each of the two cells. The anthers are as in
Lileocarpus proper, except that the valves open to a greater extent.
The style, however, although undivided and two-grooved, is bipartible.
The drupe, although ovoid when young, is nearly globose when ripe;
the putamen, which is not very thick, is one-celled and one-seeded.
Embryo not seen,

Orn TERNSTREMIACE 2.

1. DRAYTONIA, Nov. Gen.

Calyx ebracteolatus, quinquepartitus, ime basi ovarii tantum aceretus,
persisiens; sepalis inequalibus cestivatione imbricatis. Petala 5,
obovata, cestivatione convoluta vel convoluto-imbricata. Stamina plu-
roma; filamentis basi dilatatis breviter monadelphis: anther bilocu-
lares, dorso affixce incumbentes, loculis apice rima introrsa hiantibus.
Ovarium triloculare (rarius 4—5-loculare): stylus unicus: stigma
obtuse trilobum. Ovula in placentis incrassatis, e loculorum angulo
centralt prominentibus plurima, anatropa. Capsula subcarnosa, tri-
locularis (rarius 4—5-locularis), apice loculicide trivalvis ? loculis poly-
spermis. Semina reticulato-scrobiculata. Embryo in axi albuminis
carnosi, ejusdem dimidio brevior, subcylindricus ; cotyledonibus brevi-
bus semiteretibus.— Arbuscula Sauraujx facie.

TERNSTREMIACES. 207.

1. DRAYTONIA RUBICUNDA, Sp. Nov. (Tab. 15.)
Has. Banks of shaded mountain streams, Ovolau, Feejee Islands.

Shrub or small tree 10 to 20 feet high; the branchlets, petioles,
veins of the leaves, &c., thickly beset with ferrugineous chaffy scurf
when young, otherwise glabrous. Leaves alternate, crowded towards
the ends of the branches, chartaceous in texture, oblong, more or less
acute at both ends, 4 to 7 inches long, and 1% to 2% inches wide,
acutely serrulate, pinnately veined from a strong midrib, and under-
neath reticulated between the veins; the areolz of the veinlets oblong
and transverse. Petioles an inch or more in length. Leaf-buds
naked: vernation of the leaves conduplicate. Peduncles axillary, as
long as the petioles, or shorter, bearing a several-flowered corymb-like
cyme. Pedicels minutely bibracteolate about the middle, half an
inch long. Calyx not bracteolate, deeply jfive-parted, its very base
adnate to the base of the ovary, otherwise free, persistent: the sepals
somewhat chartaceous, unequal (the two exterior smaller), roundish-
ovate, concave, quincuncially imbricated in cestivation. Petals 94,
obovate, entire, longer than the sepals, 3 or 4 lines in length, “red,”
distinct, subperigynous, convolute, or oftener convolute-imbricated
(one petal being exterior as in Sawrawa) in estivation, tardily deci-
duous. Stamens numerous (40 or more), inserted with the petals, but
not adnate to them, scarcely half the length of the corolla: filaments
subulate, their dilated bases monadelphous into a short ring: anthers
oblong, two-celled, fixed by their back near the base to the apex of the
filament, therefore incumbent, destitute of a manifest connective; the
cells slightly recurved, separate at the apex, opening from the apex
downward by a long introrse chink, extending quite to the middle of
the cell. Pollen-grains simple, oval. There is no manifest disk.
Ovary three-celled (rarely 4-5-celled), globular. Style single, columnar,
longer than the stamens, terminated by a subcapitate three-lobed
stigma. Placente 3, thickened and fleshy, projecting from the axis
into the cavity of the cells; the surface thickly covered by the nwme-
rous anatropous ovules. Capsule globose, invested by the persistent
calyx, about 3 lines in diameter, glabrous, reddish, three-celled (rarely
4-5-celled), with their dissepiments rather fleshy in texture and

208 PHANEROGAMIA.

inclining to baccate, but apparently opening Joculicidally at the apex
at full maturity: the cells nearly filled by the large and fleshy or
pulpy placentse. Seeds very numerous and crowded on the surface of
the placenta, obovate or globular, somewhat angled by mutual pres-
sure; the testa very strongly reticulate-pitted, conformed to the nucleus:
inner integument thin. ZHmbryo nearly cylindrical in the axis of the
fleshy albumen, occupying about half its length, straight: radicle next
the hilum: cotyledons short, semiterete.

The tissue of the placente is replete with acicular crystals (ra-
phides): but I have not observed them in the seed-coats, where,
according to Mr. Bennet, they abound in Saurayja.

This quite “ornamental plant” is nearly allied to the genus Saw-
rawja, principally differing from it, indeed, in the union of its styles
into one. The prevailingly trimerous pistil, the cohesion of the very
base of the calyx with the base of the ovary, and the want of a hypo-
gynous disk, are subsidiary characters ; which decide us to establish a
distinct genus for our plant, even if we do not thus separate Malacho-
dendron from Stuartia. The discovery of a plant so exactly like
Saurauja, but with united styles, is particularly interesting from its
bearing on the affinities of that genus, which has always been con-
sidered as a somewhat paradoxical member of the order Ternstre-
miacee, Dr. Lindley having even proposed its transference to the
Dilleniacee ;* and Dr. Planchon+ having indicated some striking
points of resemblance with Clethra. The first-mentioned view, at
least, which was probably suggested by the distinct styles of Saurauja,
along with a certain similarity in the foliage, is set aside by the plant
now made known.

With much satisfaction I dedicate this genus to J oseph Drayton,
Esq., the principal of the scientific artists of the Expedition, of no
small attainments in natural history, especially in Conchology, to
whose pencil and superintendence the illustrations of the whole inver-
tebrate Zoology of the Expedition owe their high perfection.

Puate 15.—DRAYTONIA RUBICUNDA: a branch, of the natural size,
with flowers and fruit. Fig. 1. Portion of the deciduous chaffy scurf,

* The Vegetable Kingdom, in loco.
{ In Hooker’s London Journal of Botany, 5, p. 2538.

TERNSTR@MIACEZ. 209

magnified. 2. Diagram of the flower, in a transverse section. 3.
Flower, seen from below. 4. The same, seen from above. 5. Verti-
cal section of the pistil, with the base of the calyx, corolla, &. 6.
A portion of the ring of stamens. 7. Apex of the style and stigma,
more magnified than in fig. 5. 8. Fruit, surrounded by the persistent
calyx. 9. Transverse section of the fruit. 10. A seed. 11. Vertical
section of the same. 12. Embryo, detached.—All the dissections
variously magnified. |

2, TERNSTREMIA, Muti.

1. TernstRemiA BRASILIENSIS, Camb.
Ternstremia Brasiliensis, Cambess. in St. Hil. Fl. Bras. Mer. 1, p. 298, t. 59.

Has. Near Rio Janeiro, Brazil; the var. MINOR.

3. EHURYA, Thunb.
1. Eurya? Sanpwicensis, Sp. Nov.

E. ramulis ultimis strigillosis ; foliis ellipticis oblongisve basi subcordatis
creberrime serrulatis reticulatis ; floribus (masculis) in axillis solitaris
nutantibus 10—1d-andris.

Has. Sandwich Islands: on mountains, behind the town of Hono-
lulu, Oahu: also in forests, on the side of Mouna Kea, Hawaii.
(Gathered at Oahu by Macrae and Gaudichaud.)

Apparently a large shrub, much branched; the branches crowded
with leaves; the ultimate branchlets strigosely hairy. Leaves elliptical
or oblong, coriaceous, varying from one to 22 inches in length, and
from half an inch to an inch in width, mostly obtuse or rounded at
the apex, more or less cordate at the base (the smaller slightly so),
closely serrulate throughout with inflexed mucronulate teeth, nume-

rously and closely feather-veined, and with the veinlets finely reticu-
58

210 PHANEROGAMIA.

lated underneath, glabrous, or the midrib strigillose underneath, on
distinct petioles of only a line and a half in length, or sometimes
sessile and the leaf appearing as if partly clasping. Only one kind of
flowers seen: these appear as if hermaphrodite, but their ovary is
destitute of ovules. The lowers are solitary in the axils of the much-
crowded leaves, on ebracteolate pedicels of from 3 to 5 lines long,
mostly nodding, large for the genus, being 3 or 4 lines in length.
Calyx scarcely puberulent, brownish, subtended by two small bract-
lets at the base: sepals 5, unequal (the exterior smaller), coriaceous,
orbicular, imbricated, persistent. Corolla pale yellow, hypogynous,
nearly twice as long as the calyx; the petals 5, united at the base,
imbricated, obovate, thickish. Stamens 10 or 15, hypogynous, or
very slightly if in the least adnate to the base of the corolla: fila-
ments distinct, short, subulate: anthers oblong-linear, mucronulate,
longer than the filaments. Ovary (sterile) three-celled; the cells not
ovuliferous: styles as long as the stamens, united at the base, or sepa-
rate, or two of them united: stigma subcapitate.

Until the fertile flowers or the fruit are known, we cannot be
certain that this is a true Hurya; but it probably is so.

2. EuryA VitiEnsis, Sp. Nov.

E. glaberrima ; foliis lanceolato-ellipticis oblongisve utrinque acuminatis
serrulatis supra nitidis; floribus plerisque geminis, masculisve (10-
andris) fasciculatis; sepalis orbiculatis; stylis 3 vel 4 brevissimis
Jere discretis ; fructu globoso.

Has. On the summit of a mountain of Ovolau, one of the Feejee
Islands, at an altitude of 2,000 feet.

Shrub (or small tree?) very glabrous, even to the ultimate branch-
lets; or these slightly and minutely pubescent, when very young, in
the specimens with male flowers. Leaves lanceolate-elliptical, oblong,
or oblong-lanceolate, rather crowded on the angled branches, tapering
or acuminate at both ends, coriaceous, serrulate with callous teeth, 14
to 2 inches long, 8 to 10 lines wide, smooth and shining above, dull
but scarcely paler underneath: petioles 2 lines long. Flowers dios

TERNSTR@MIACEA. 211

cious, very small (the buds about a line in diameter), axillary; the
fertile ones mostly in pairs, rarely in threes, sometimes solitary; the
staminate often in clusters of 3 or 4 or more in the same axils. Pedi-
cels of the fertile flowers 12 to 2 lines in length ; of the sterile shorter,
not so long as the flowers. Calyx minutely 1—2-bracteolate; the 5
sepals orbicular, imbricated. Corolla yellowish, rather longer than
the calyx; the petals 5, oval or obovate, a little united at the base.
Staminate flowers with an abortive rudiment in place of a pistil,
decandrous. Filaments distinct, free or nearly so from the corolla.
Anthers oblong, scarcely as long as the filament, obtuse, very minutely
apiculate under a lens. Pistillate flowers destitute of any rudiments
of stamens. Ovary globose, glabrous, four-celled, sometimes three-
celled, with numerous downwardly imbricated ovules in each cell.
Styles 3 or 4, very short, nearly distinct to the base: stigmas introrse,
subcapitate. Fruit globose, barely a line in diameter, a little longer
than the persistent calyx, dry, crustaceous, indehiscent; the cells
several-seeded.

This species in foliage resembles Lurya nitida of Korthals (in Ver-
hand. Naturl. Gesch. Nederl. Overzeesch. Bezitt. p. 115, t. 17), from
Borneo; but that has cuspidate-pointed anthers, an ovoid and pointed
ovary with much fewer ovules, a larger fruit, and longer styles, which
are united for more than half their length,

Subgen. EURYODES.—Jlores masculi pentandri, rartus hexandri. (Samoénses, flo
ribus subsessilibus solitariis. )
3. Hurya (Euryopes) Pickerine, Sp. Nov.

E. glaberrima; foliis obovatis seu ovalibus basi acutis crebre serratis
coriaceis supra nitidis.

Has. Tutuila, one of the Samoan or Navigators’ Islands; on moun-
tains, at an elevation of about 2,000 feet.

Shrub “4 to 10 feet high,” very glabrous; the ultimate branchlets
scarcely if at all pubescent, even when young. Leaves obovate, oblong-
obovate, or oval, acute at the base, obtuse or obtusely pointed, closely

212 ' PHANEROGAMIA.

serrate with inflexed callous teeth, 2 to 24 inches long, thick and
corvaceous, the upper surface shining, much resembling those of Thea.
Flowers solitary in the axils, subsessile, small and inconspicuous; the
sterile pentandrous. Sepals round-ovate, mucronulate, coriaceous.
Petals broadly ovate. Fruit globose, dry, indehiscent, smooth, much
larger than the calyx, 2 lines in diameter, 4—5-celled. Seeds several
in each cell, globular-reniform, with a cellular scrobiculate-reticulated
testa. Embryo curved, in the axis of fleshy albumen.

The specimens of this and the subjoined species afford only scanty
materials for elucidating their floral characters. This, however, fur-
nishes some mature fruit; the other has male flowers. Both appear
to differ from Hurya only in the reduction of the stamens to five, or at
most to six. This single character probably should not suffice to sepa-
rate them from that genus, several species of which are only decan-
drous.

4, Kurya (Euryopes) Ricuu, Sp. Nov.

E. foliis oblongo-lanceolatis serrulatis, novellis subtus (preecipue ad
costam) ramulisque ultimis pilosulis.

Has. Upolu, one of the Samoan or Navigators’ Islands.

A shrub, much like the preceding; but the young branchlets and
the lower surface of the young leaves are pilose with slender and
silky fine hairs, which are mostly persistent on the midrib. The
leaves, too, are much narrower, being oblong-lanceolate, 2 to 24 inches
in length by 6 to 10 lines in width, more finely serrulate, less coria-
ceous, and the upper surface not so shining. Flower-buds silky-
pubescent externally, solitary (rarely, perhaps, in pairs) and almost
sessile in the axils of the leaves, a line and a half in length; the
sepals and petals broadly ovate. Stamens 5 (or in one instance 6),
wholly free from the corolla: filaments very short: anthers linear-
sagittate, mucronate. A mere subulate rudiment takes the place of
the pistil in the staminate flowers. The fertile flowers have not been
noticed.

TERNSTREMIACEA. 213

4, PLOIARIUM, Korth.

1. PLOIARIUM ELEGANS, Korth.

Ploiarium elegans, Korthals, Verhand. Naturl. Geschied. Nederl. Overzeesch. Bezitt.
Bot. p. 135, t. 25.

Has. Singapore: in fruit. Perhaps not indigenous. But no memo-
randa are extant relating to the Singapore and Philippine collections.

0» LAPLACEA, HBX.

1. LAPLACEA SEMISERRATA, Camb.

Laplacea semiserrata, Cambess. in St. Hil. Fl. Bras. Mer. 1, p. 800.
Heemocharis semiserrata, Mart. & Zuce. Nov. Gen. & Spec. Bras. 1, p. 107 t. 67.

Has. Organ Mountains, near Rio Janeiro, Brazil.

6 CALPANDRA, Blume.
1. CALPANDRA LANCEOLATA, Blume.
Calpandra lanceolata, Blume, Bijdr. p. 178; Korthals, Verhand. Nederl. p. 148, t. 31.
Has. Mountains near Baiios, Luzon, Philippine Islands.

The specimens, which consist of leafy branches, bearing flower-buds,
with a portion of a half-grown fruit, so perfectly accord with the
excellent figure and analyses given by Korthals, as to leave no doubt
of their belonging to the same species. The genus manifestly belongs
to the Ternstroemiacee.

54

914 PHANEROGAMIA.

a

7.2 ARISTOTELIA, Ler.
1. AristoteLiA Maqut, L’ Her.

Has. Chili, near Valparaiso. (A fragment in flower only in the
collection.)

Orv GUTTIFER &.

1 ARRUDEA, Camb.
1. ARRUDEA CLUSIOIDES, Camb.

A, foliis obovatis ; stigmatibus radiantibus 8; capsula octoloculari octo-
valvi, loculis 3—4-spermis.

Arrudea clusioides, Cambess. in St. Hil. Fl. Bras. Mer. 1, p. 319, t. 66.

Has. Organ Mountains, near Rio Janeiro, Brazil.

A small tree, with wholly the aspect of Clusia, except in the greater
number of (imbricated) parts in the calyx and corolla; the obovate
leaves about 4 inches long. In respect to the flowers, there is nothing
to add to the figure and description of Cambessedes, above cited,
except that there is no distinct style, at least in the fertile flowers.

GUTTIFER &. 915

Calyx persistent. Petals tardily deciduous. Ovary eight-celled, eight-
lobed, crowned with as many radiated stigmas: these are distinct,
tongue-shaped, large, strongly reflexed, and of a fleshy texture: the
cells, in specimens which are considerably enlarged after flowering,
are all destitute of ovules, as was the case in St. Hilaire’s specimens.
The flowers are undoubtedly polygamous. We have, however, a ripe
pod, from which the character may be completed, as to the fruit
and seeds. Capsule eight-celled, globose-ovoid, an inch in length,
septicidally etght-valved ; the valves margined by thin dissepiments,
separating from the thickened axis which bears the 8 persistent and
projecting +placentz, cartilaginous (the exocarp fleshy when fresh *),
each tipped on the back, just below the summit, with the vestiges of
the reflexed stigma. Seeds 3 or 4 in each cell, in a single series,
enclosed in pulpy matter, cylindrical-oblong, slightly arcuate, 3 lines
in length; the testa chartaceous, loose; the inner integument mem-
branaceous, lineate with resinous vitte, conformed to the oblong-linear
nucleus, which consists of an embryo, the parts of which are not
readily distinguishable.

The genus was founded on a specimen of the present plant, with
flowers which, although apparently hermaphrodite and perfect, were
found to have the cells of the ovary destitute of ovules. The fruit is
only now made known. Meanwhile a second species, gathered by
Splitgerber and Focke, in Surinam, and briefly indicated by the former,
under the name of Arrudea purpurea, has been fully described by
Miquel, in Linnzxa, 18, p. 229-232, who has recast the character of
the genus from the new materials. A. purpurea is described as
having a pentamerous pistil (with 5 stigmas and 69 cells to the ovary),
and numerous seeds, occupying several series in each cell of the fleshy
five-valved capsule. Hence Prof. Miquel inclines to doubt whether
A. clusicides has truly .an octomerous ovary, with few-ovuled cells
(the latter character having been inserted on the strength of a MS.
note made by St. Hilaire in Brazil). I am enabled to confirm both
these points. I have not seen the ovules, however; but they can
hardly be numerous, since there are only three, or at most four, seeds
in each cell of the single pod which the present collection affords.
The calyx in A. purpurea is said to be deciduous: in A. clusiovdes it
persists at the base of the capsule; or at least the exterior scaly
envelopes are persistent.

216 PHANEROGAMIA.

2. CLUSIA, Linn.
1. CLUSIA LANCEOLATA, Camb.
Clusia lanceolata, Cambess. in St. Hil. Fl. Bras. Mer. 1, p. 318.

Has. Near Rio Janeiro, Brazil. (In fruit.)

2. CLUSIA GANABARICA, Casaretto.

Clusia Ganabarica, Casaretto, Nov. Stirp. Bras. Dec. no. 63; Walp. Repert. 5, p. 144.

Has. Near Rio Janeiro. (In flower.)

3. CLUSIA RUPICOLA, Casaretto, 1. c.?

Has. Organ Mountains, Brazil. (In flower.)*

8. DISCOSTIGMA, Hassk,

Discostiama, Hasskarl, Cat. Hort. Bogor. no. 212, & Pl. Jav. Rar. p. 276; Endl.
Gen. Suppl. 3, p. 95.

1. Discostiema Vitrensz, Sp. Nov. (Tab. 16.)

D. foliis subaveniis ; floribus (feeminis) in axillis solitariis geminisve ;
pedicellis bibracteolatis.

* Clusia sessilis, Hook. & Arn. Bot. Beech. Voy. p. 80 (non Forst.), from the Sand-
wich Islands, is a Rutaceous plant, to be described under that order, in this work.
Moreover the original Clusia sessilis of Forster, from Tongatabu, appears not to be a real
Clusia.

GUTTIFERZ. 217

Has. Ovolau, Feejee Islands; at the altitude of 1500 feet above
the level of the sea.

“Tree 35 feet high,” glabrous, with terete branches; the slender
branchlets somewhat angled. Leaves opposite, approximate, short-
petioled, oblong or oblong-lanceolate, acuminate at both ends, 2 to 23
inches long, about two-thirds of an inch wide, coriaceous, both sides
smooth and nearly similar in aspect, not shining, appearing nearly
veinless: the base is acute and decurrent into the short petiole; the
apex is conspicuously acuminate, but with a rounded point. The
only flowers known are fertile, probably only female ones, from which
the floral envelopes are deciduous: they are awillary and solitary,
rarely in pairs, on pedicels of 3 lines in length, which are bibracteolute
near the middle; the bractlets opposite and minute. The hypo-
gynous torus appears to have borne very few sepals or other floral
envelopes, and probably no stamens. Ovary ovoid, glabrous, fleshy,
very obtuse, the apex covered with a very broad and depressed, disk-
shaped, fleshy stigma, which is entire or nearly so; within it is two-
celled, and with a single, oblong, amphitropous ovule in each cell;
the micropyle apparently inferior. Fruit not seen, probably it is
baccate.

Of the original species, Discostigma caudatum, as of ours, only the
fertile flowers are known: these are said to be in small and lateral
or axillary racemes. Even the floral envelopes of our plant are
unknown; but there is hardly a doubt that it is a genuine species of
Dhscostigma.

PuatE 16, A—Discostiema Vitiense: fertile plant, with forming

fruit. Fig. 1. Fertilized pistil, enlarged. 2. Vertical section of the
same. |

4, GARCINIA, Linn.
1. GarcintA MAnGostana, Linn.

Has. Mangsi Islands. (Foliage only collected.)
55

218 PHANEROGAMIA.

0 CALOPHYLLUM, Lun.
1. CALopHYLLUM InopHyLium, Linn,

Has. Society, Navigators’, Friendly, Feejee, and Mangsi Islands:
very common.

2. CALOPHYLLUM SPECTABILE, Willd,

Has. Navigators’, Friendly, Feejee, and Mangsi Islands. (Foliage
only, except the specimens from the Feejee Islands; these are in
flower and fruit.)

There are also the following undeterminable Guttifere in the col-
lection, namely: 1. A Garcinia? with male flower-buds only, from
Muthuata Bay, Feejee Islands. 2. Possibly a second species of Gar-
cinia, foliage only, from mountains near Baiios, Luzon. 3. A third
plant, foliage only, from the Samoan or Navigators’ Islands, the genus
of which cannot be guessed.

Orv. HYPERICACE &.

1. VISMIA, Véelloz.

1. Vismia Hinaru, Gardn. ined.

Vismia Guianensis, St. Hil. Fl. Bras, Mer. 1, p. 827, non Chois, nec Hypericum
Guianense, Aubl. .

Has. Brazil, near Rio Janeiro, and on the Organ Mountains.

2 HYPERICUM, Linn.

1. HYPERICUM GRANDIFOLIUM, Choisy.

Androsemum Webbianum, Spach. Suit, Buff. 5, p. 418; Webb, Phytogr. Canar. 1,
p. 50, t. 4, E. ;

Has. Funchal and Curral, Madeira.

2. HyYpErRicuM FLORIBUNDUM, Ait.
Webbia floribunda, Spach. 1. c.; Webb, Phytogr. Canar. 1, p. 46, t. 4, B.

Has. On the coast, east of Funchal, Madeira.

3. HyprricuM GLANDULOSUM, At.

Hypericum glandulosum, Ait. Hort. Kew.; Webb. Phytogr. Canar. 1, p. 44, t. 3.

220 PHANEROGAMIA.

Has. On rocks, east of Funchal, Madeira; where it is abundant.

4, HYPERICUM PERFORATUM, Linn.
Has. Valley west of Curral and east of Funchal, Madeira. (Two
forms of the species.)
5. Hypericum numirusum, Linn.

Has. On rocks, east of Funchal, Madeira.

6. Hypericum BRASILIENSE, Choisy.

Hypericum Brasiliense, Choisy, in DC. Prodr. 1, p. 547; St. Hil. Fl. Bras. 1, p. 835.

Has. Organ Mountains, Brazil.

7. Hypericum JAPonicum, @. HUMIFUSUM, Hook. f.

Hypericum Japonicum, Thunb. var. 8. humifusum, Hook. f. Fl. N. Zeal. p. 87.
H. pusillum, Choisy, in DC. Prodr. 1, p. 549; A. Gunn. Prodr. Fl. N. Zeal.
Ascyrum humifusum, Labill. Fl. N. Holl. 2, p. 33, t. 175.

Has. Waia-ruru Bay, and Bay of Islands, New Zealand.

8. HyPERICUM GRAMINEUM, Forst.

Hypericum gramineum, Forst. Prodr. p. 58; Labill. Sert. Austro-Cal. p. 53, t. 53;
Hook. f. Fl. N. Zeal. p. 36.

H. involutum, Choisy, in DC. Prodr. 1, p. 549.

Ascyrum involutum, Labill. Fl. N. Holl. 2, p. 82, t. 174.

Has. New South Wales, near Sydney, Hunter’s River, &.; both
broad and narrow-leaved forms.

Orv. PITTOSPORACEZ.

1. CITRIOBATUS, A. Cunn.

1. CITRIOBATUS MULTIFLORUS, A. Cunn.

Citriobatus multiflorus, A. Cunningham, ex Loud. Hort. Brit. Suppl.; Putterl. Syn.
Pittosp. p. 4.

Has. Puen Buen, New South Wales. (In fruit.)

2, BURSARIA, Cav.
1. BuRSARIA SPINOSA, Cav.

Bursaria spinosa, Cav. Ic. 4, p. 30, t. 8350; Sims, Bot. Mag. t. 1767; Putterl. Syn.
Pittosp. p. 19.

Has. Sydney, Puen Buen, &c., New South Wales.

2. BURSARIA DIOSMOIDES, Putterl. 1. c.

Has. Near Sydney, New South Wales.

3 BILLARDIERA, Smith.

1. BILLARDIERA SCANDENS, Smith.

Billardiera scandens, Smith, Specim. Bot. N. Holl. (Exot. Bot.) t. 1; Sims, Bot.

Mag. t. 801.
ia 56

222 PHANEROGAMIA.

Has. Near Sydney, New South Wales; not uncommon.

2. BILLARDIERA MUTABILIS, Salish.
Billardiera mutabitis, Salisb. Parad. Lond. 1, t. 48; Sims, Bot. Mag. t. 1818.

Has. Puen Buen, &c., New South Wales.

4. PITTOSPORUM, Soland.
* Australasica.
1. PrrrosporuM UNDULATUM, Vent.

Pittosporum undulatum, Vent. Hort. Cels. t. 76; Andr. Bot. Rep. t. 883; Ker. Bot.
Reg. t. 16; Putterl. Syn. Pittosp. p. 6.

Has. Sydney, New South Wales. (In fruit.)

2. PITTOSPORUM ACACIOIDES, A. Cunn.

Pittosporum acactoides, A. Cunn. Spec. Bot. N. Zeal. 1. ¢. p. 109, adn.

Haz. Near Woolongong, New South Wales. (In fruit only.)

x * Novo-Zelandica.
3. PITTOSPORUM TENUIFOLIUM, Banks & Soland.

Pittosporum tenutfolium, Banks & Soland. in Gertn. Fruct. 1, p. 86, t. 59 ; A. Cunn.
Spec. Bot. N. Zeal. in Ann. Nat. Hist. 4, p. 107; Putterl. 1. ¢. p- 13; Hook. f.
Fl. N. Zeal. p. 21.

Trichilia monophylla, A. Rich. Fl. N. Zel. p. 360, t. 84.

Has. Bay of Islands, New Zealand. (With mature fruit.)

PITTOSPORACEA. 993

4, PITTosPoRUM CoRNIFOLIUM, A. Cunn.

Pittosporum cornifolium, A. Cunn. in Bot. Mag. t. 3161, & Spec. Bot. N. Zeal. 1. ¢.;
Putterl, Syn. Pittosp. p. 14; Hook. f. Fl. N. Zeal. p. 23.

Has. Bay of Islands, New Zealand. (In fruit.)

5. PITTOSPORUM CRASSIFOLIUM, Banks & Soland.

Pittosporum crassifolium, Banks & Soland. ined.; A. Gunn. Spec. Bot. N. Zeal. 1. ¢.
4, p. 106; Putterl. Syn. Pittosp. p. 12; Hook. f. Fl. N. Zeal. 1, p. 23.

Has. Waia-ruru Bay, New Zealand. (In fruit.)

6. PITTOSPORUM UMBELLATUM, Banks & Soland,

Pittosporum umbellatum, Banks & Soland. in Geertn. Fruct. 1, p. 286, t.59; A. Cunn.
l.c.; Putterl. 1. c. (excl. syn. P. eugenioides) ; Hook. f. Fl. N. Zeal. p. 28.

Has. Bay of Islands, Tippona, Waia-ruru Bay, New Zealand.
Putterlick’s P. microcarpum, and not P. umbellatum, evidently is the
P. eugenwoides of A. Cunningham,
7. PITTOSPORUM PIMELIOIDES, R. Cunn.

Pittosporum pimelioides, R. Ounn. in. A. Cunn. Spec. Bot. N. Zeal. 1. c.; Putterl.
l. c.; Hook. f. 1. ¢.

Has. Bay of Islands, New Zealand. (In fruit.)

x * * Polynesica,
8. PITTOSPORUM ARBORESCENS, Rich, in herb.

P. glabrum ; foliis chartaceis obovato-oblongis seu oblongis obtusis basi
acutis longiuscule petiolatis margine subundulatis ; pedunculis glome-
rato-multifloris ; calyce 4—6-fido; petalis coalitis; capsula levi sub-

224 PHANEROGAMIA.

globosa bacciformi bivalvi polysperma, valvis crassissimis dorso sul-
catis ; seminibus lenticularibus pallide purpureis.

Has. Tongatabu. Also Feejee Islands.

“Tree 25 feet high,” glabrous throughout, except a minute and
deciduous pubescence on the inflorescence. Leaves chartaceows in
texture, alternate, occasionally approximated as if in whorls, obovate-
oblong, oblong, or sometimes narrowly elliptical-oblong, 3 or 4 inches
in length, one or 2 inches in width, obtuse at the apex, acute at the
base, tapering into a petiole of half an inch or more in length, smooth,
scarcely shining above, paler and dull beneath, rather conspicuously
feather-veined, the margin minutely undulate. ' Inflorescence terminal
and in the upper axils. Peduncles many—flowered, short; the flowers
glomerate in crowded cymules, small; the pedicels not more than a
line and a half long, subtended by minute ovate-subulate bracts.
Calyx 4—5-cleft to the middle, not longer than the pedicel, glabrous;
the lobes ovate, obtuse. tals 4 or 5, linear, more or less coalescent
into a tube (especially about the middle), their obtuse apices imbri-
cated in estivation, 3 or 4 lines long. Stamens 4 or 5. Filaments
subulate, short. Anthers introrse, linear-sagittate. Pistil glabrous :
style short: stigma truncate, obsoletely two-lobed. Ovary sessile, im-
perfectly two-celled, many-ovuled. Oapsule fleshy or berry-like, “ orange-
coloured,” smooth, and glabrous (the surface minutely wrinkled in
the dried state), globular, half an inch long, two-valved, many-seeded :
the valves very thick, woody-coriaceous when dry, grooved on the back
opposite the placenta, Seeds sessile and two-ranked on each valve,
angled-lenticular, barely 2 lines in diameter; the smooth and purplish
testa thin, not shining. Albumen and embryo as in the genus.

The materials consist of one poor specimen with flowers, chiefly
unexpanded, and one with dehiscent fruit from Tongatabu; with a
similar fruiting specimen from the Feejee Islands, the particular
habitat not recorded. The inflorescence is like that of the following
species.

9. PirrosporuM Ricuu, Sp. Nov.

P. glabrum ; foliis coriaceis lanceolato-oblongis utrinque acutis vel acu-
tiusculis; pedunculis multifloris, floribus glomerato-cymosis; calyce

PITTOSPORACE S&. 925

obtuse quinquefido ; capsula subglobosa leviuscula bivalvi polysperma,
valvis corvaceo-lignosis dorso convexis ; seminibus ovalibus atris.

Has. Vanua-levu, one of the Feejee Islands.

Apparently a small tree, glabrous throughout, except the inflorescence
which is minutely pubescent. Leaves coriaceous, but not very thick,
alternate, crowded towards the end of the branches, of nearly the
same green hue both sides, not shining, lanceolate-oblong, acute at both
ends, or the apex sometimes obtuse, tapering into a petiole of half an
inch or less in length, entire, inconspicuously feather-veined, 24 to 34
inches long, an inch to 1¢ wide. Peduncles many-flowered, axillary
and terminal, 6 to 12 lines long; the flowers cymose-clustered, small,
on very short pedicels, subtended by minute bracts. Calyx nearly
glabrous, scarcely a line in length, obtusely five-cleft; the lobes very
short, roundish. Corolla in bud barely 14 lines long; the petals
oblong, imbricated, scarcely if at all coalescent. Stamens, pistil, &c.,
nearly as in the foregoing species. Capsule globular, apparently not
baccate nor fleshy, smoothish (the surface minutely wrinkled), two-
valved, many-seeded ; the thick valves rather woody in texture, about
8 lines long and 10 lines wide, not grooved but convex on the back.
Seeds oval, turgid, sessile, fully 2 lines long, with a rather fleshy,
black and shining, smooth testa.

I have combined a fruiting specimen (from Vanua-levu) with two
flowering specimens (the locality of which is not recorded), as in all
probability belonging to the same species : and there is another imper-
fect specimen with young fruit, which is most probably to be referred
here. The expanded flowers I have not seen. The species differs

from the foregoing principally in the characters given in the specific
phrase.

10. Prrrosporum BrackENRIDGEI, Sp. Nov. (Tab. 17.)

P. glabrum; folits coriaceis ovali-oblongis sew ellipticis obtusis basi
g ’
acutis; pedunculis (fructiferis) binis ternisve terminalibus unifloris
petiolum haud superantibus; capsula subglobosa leviuscula bivalvi
polysperma, valvis coriaceo-lignosis dorso convexis ; seminibus subglo-
bosis atris.
57

226 PHANEROGAMIA.

Has. Muthuata, one of the smaller Feejee Islands. (In fruit only.)

“A small tree, 15 to 20 feet high,” glabrous throughout. Leaves
coriaceous in texture, but not thick, alternate, much crowded at the
summit of the branches, oval-oblong or elliptical, obtuse, or a few of
them obtusely somewhat pointed, all acute at the base, smooth, scarcely
shining above, dull, but little paler underneath, copiously feather-
veined, entire, 34 to 5 inches long, 12 to 2 inches wide; the petioles
from half to two-thirds of an inch in length. Flowers not seen: the
specimens all with mature fruit. Peduneles (in fruit) 2 or 3 together,
terminal, simple, apparently only one-flowered, 6 or 7 lines long, not
exceeding the petiole. Capsule globular, apparently a little flattened
parallel with the two valves, smoothish, but minutely tuberculate-
wrinkled under a lens, apiculate with the short persistent style, very
many-seeded ;. the valves thick and rather woody, 9 or 10 lines in length
and breadth, convex on the back. Seeds obovoid-globose, in several
series, with a smooth and black testa.

Piate 17, A.—Pitrrosporum BRACKENRIDGEI: branch in fruit. Fig.
1. Transverse section of a capsule. 2. A seed, enlarged.

11. Prrrosporum tosrroipEs, Sp. Nov. (Tab. 17.)

P. glabrum ; foliis coriaceis elliptico-obovatis obtusissimis subretusis basi
acutis ; pedunculis (fructiferis) terminalibus solitariis unifloris petio-
lum bis superantibus; capsula globoso-subtetragona levi polysperma
bivalvi, valvis crassissimis coriaceo-lignosis dorso convexis; seminibus
subglobosis atris.

Has. Somu-somu, Feejee Islands. (In fruit.)

Apparently a small tree, glabrous, except the peduncle, which is
somewhat pubescent. Leaves coriaceous, somewhat shining above,
scarcely paler underneath, elliptical-obovate, 2 or 3 inches long, 1+ to
nearly 2 inches wide, with a very obtuse and rounded or slightly retuse
apex, acute at the base, but not so tapering as in P. Tobira, entire,
copiously but not prominently feather-veined. Flowers not seen.
Fructiferous peduncle terminal, and to all appearance solitary and only

PITTOSPORACE &. 997
one-flowered, an inch long, twice the length of the petioles. Capsule
smooth or nearly so, globular, but somewhat quadrangular, having
rather acute angles at the dehiscent sutures, the placental sutures
obtuse; the valves thick and woody, 9 lines in length and breadth,
many-seeded. Seeds globular, a line and a half in diameter; the testa
very smooth, d/ack and shining.

This species bears more resemblance to the preceding than to any
other; from which it is abundantly distinguished by its smaller and
rounded leaves (that may be likened to those of P. Tobira, although
larger and not so cuneate), and solitary, longer peduncles. It should
be compared with P. uniflorum of Putterlick, which is briefly cha-
racterized from a cultivated specimen, of uncertain origin.

Piate 17, B.—PitrosporuM ToBIROIDES: in fruit. Fig. 1. .Trans-
verse section of a capsule. 2. A seed, enlarged. 38. Vertical section
of the same.

12. Pirrosporum PickERinGu, Sp. Nov.

P. glabrum; foltis coriaceis oblongis vel oblongo-lanceolatis utrinque
acuminatis margine undulatis ; pedunculis terminalibus in umbellam
digestis gracilibus apice umbellatim plurifloris; calyce abbreviate
quinquefido; petalis discretis; ovario ovoideo quinqueloculart basi
attenuato substipitato, loculis sexovulatis ; stylo brevissimo,

Has. Feejee Islands. (In flower only.)

Apparently a small tree, entirely glabrous, except the inflorescence,
which is minutely pubescent. Leaves coriaceous, alternate, but mostly
crowded as it were in whorls at intervals on the flowering branches,
oblong or oblong-lanceolate, acuminate at both ends, from 4 to 6 inches
in length, and 12 to 2 inches in width, on petioles of about an inch
in length, smooth and of the same hue both sides, rather conspi-
cuously feather-veined, the margins undulate, often as if repand-cre-
nulate. Inflorescence terminal: peduncles 6 or 7 in number, wmbellate,
subtended by as many leaves of the ultimate false whorl, slender,
unequal, from one to 2 inches in length, naked, wmbellately 3—7-flowered

I28 PHANEROGAMIA.

at the apex: pedicels 4 or 5 lines long, subtended by small subulate
bracts. Calyx short, barely a line in length, five-cleft scarcely to the
middle; the lobes obtuse. Corolla 3 or 4 lines long; the petals dis.
tinct, oblong-linear. Filaments linear-subulate, thrice the length of
the sagittate and mucronulate anthers. Ovary ovoid or fusiform,
pubescent (the pubescence apparently deciduous), five-celled, narrowed
at the base, so as to be slightly stipitate, tapering at the apex into a
very short style (of not more than half a line in length): stigma
obscurely five-lobed. Ovules 6 in each cell, crowded in a single series
at the middle portion of the axis, on short but distinct funiculi.
Fruit not seen.

This would belong to the section Senacia of Blume (Mus. Bot.
Lugd.-Bat. 1, p. 159); but it is manifestly different from any described
species.

13. PirrospoRUM RHYTIDOCARPUM, Sp. Nov. (Tab. 18.)

P. glabrum; foliis coriaceis oblanceolatis seu obovato-oblongis breviter
acuminatis bast attenuatis ; floribus in umbellulas sessiles terminales
congestis; calyce abbreviato quinquesepalo; petalis coalitis; ovario
tomentuloso haud stipitato; stylo longiusculo; capsula oblonga lig-
nosa crassissima tuberato-rugosissima bivalvi polysperma; seminibus
COMpPYESSIS.

Has. Ovolau, Muthuata, and north coast of Viti-levu, Feejee
Islands.

Three or four forms, or varieties, occur in the collection, most of
them in fruit, probably all of them referable to one species, which is
remarkable for its warty-tuberculate capsule. It is “a shrub, or small
tree, from 6 to 20 feet high,” glabrous, except the inflorescence, which
is pubescent. Leaves coriaceous, green, and of nearly the same hue
both sides, oblanceolate, varying to obovate-lanceolate and obovate-
oblong, acute or acuminate with a short point, usually tapering from
the middle to a slender base, decurrent into the petiole (which is an
inch or 13 inches long), feather-veined, with entire or slightly undu-
late margins, from 4 to 7 inches in length, and 14 to 2 inches in

PITTOSPORACES. 229

width: they are either scattered or falsely whorled, as is common in
the genus. lowers terminal, or in the axils of the whorled leaves
which terminate the flowering branches, in one or more small and
sessile umbels or fascicles, which are shorter than the petioles. Pedi-
cels 2 or 3 lines long, with subulate bracts at their base. Calyx of
five ovate or oblong distinct sepals, a line and a half long. Corolla
half an inch long; the petals coalescent above the middle, forming a
tube, their free and obtuse summits spreading. Filaments subulate,
more than twice the length of the sagittate mucronulate anthers.
Ovary not stipitate, nor contracted at the base, linear-oblong, minutely
tomentose, pointed with the rather slender style (12 lines long): stigma
two-lobed. Ovules very numerous on the two placente, sessile.
Capsule oblong, 14 inches in length, very thick and woody, minutely
tomentose when young, and then more or less two-edged or crested,
the surface excessively tuberculate-rugose or warty, two-valved, many-
seeded. Seeds compressed vertically, 22 to 3 lines in diameter, with a
pale and nearly smooth testa, not shining, almost destitute of funiculi.

Piate 18,.—PirrosporuM RuHyTIpocARPUM. Fig. 1. Branch, with
flowers and young fruit. 2. Another form, with young fruit. 3. A
large-leaved form, with dehiscent capsules. 4. A flower, enlarged.
5. The connected petals displayed. 6. A stamen, seen from the out-
side. 7. The same, seen from within. 8. Pistil, with a stamen and
the receptacle. 9. Transverse section of the ovary. 10. Pistil, with
the ovary longitudinally divided. 11. Transverse section of a cap-
sule. 12. A seed. 13. The same, divided, to show the embryo, &c.
—All the analyses enlarged, except Fig. 11.

% *« * * Sandwicensia.
14. Prirrosporum GLABRUM, Hook. & Arn.

P. undique glabrum; foliis tenuiter coriaceis levibus obovato-oblongis
oblanceolatisve basi atienuates apice obtusis vel subacuminatis ; pedun-
culis terminalibus nunc lateralibus subracemoso-paucifloris; calyce
quinquepartito ovario pedicellisque glaberrimis ; capsula levi subglo-

-bosa compressiuscula bivalvi polysperma, valvis coriaceis; seminibus
atropurpureis levibus.

58

230 PHANEROGAMIA.

Pittosporum glabrum, Hook. & Arn. Bot. Beech. Voy. p. 110, vix Putterl. Pittosp.

Has. On the Kaala Mountains, and behind Honolulu, Oahu, Sand-
wich Islands.

A tall shrub, or small tree, glabrous throughout, even the inflores-
cence. Leaves coriaceous, but mostly rather thin, scattered or approxi-
mate in false whorls, from 2 to 5 inches long, the widest part an inch
to an inch and a half in breadth, obovate-oblong, or oblanceolate, with
a tapering base, the apex either obtuse, rounded, or more or less acumi-
nate, very smooth even when young, of nearly the same hue both
sides, entire, inconspicuously veined: petioles from 6 to 12 lines long.
Peduncles terminal, or lateral (by the yearly growth of the branch),
and below the leaves, spreading or pendulous, glabrous, half an inch
or more in length, racemosely or corymbosely jew-(5—-6)-flowered at
the apex. Pedicels, 3 lineslong. Calyx jive-parted, very short, 12 lines
in diameter, wholly glabrous; the segments broadly ovate, acute.
Corolla 6 lines long, apparently white: the petals linear-spatulate,
connivent, and at their upper part coalescent into a tube; the dila-
ted and spreading tips about 2 lines in length. Filaments filiform, 3
or 4 lines long: anthers linear-sagittate, mucronate. Ovary entirely
glabrous, sessile, somewhat quadrangular, terminated by a columnar
style as long as itself; the stigma truncate. Capsule subglobose, but
somewhat compressed parallel with the two valves, glabrous, smooth, or
the surface minutely roughish, two-celled, many-seeded ; the valves
coriaceous, not very thick, about an inch in length and 9 lines broad.
Seeds dark purple, closely packed in two rows in each cell, vertically
compressed, 2 or 3 lines in diameter; the testa smooth (Plate 19, Fig.
12), scarcely if at all shining.

There are specimens in the collection, from the mountains behind
Honolulu (in fruit only) which, with the foliage and the smooth seeds
of P. glabrum, have thicker-walled and minutely tomentose pods, the
surface of which is more or less tubercular-roughened. These, with
some other specimens, seem to be intermediate between P. glabrum
and the following species, although the ordinary states of the
two would appear to be abundantly different. It remains to be
seen, however, whether the character taken from the seeds will prove
constant.

PITTOSPORACE &. 231

15. PiTTOSPORUM TERMINALIOIDES, Planchon, in Herb. Hook.

P. foliis coriaceis oblongo-obovatis obtusis basi attenuatis subtus (pre-
sertim junioribus) cum inflorescentia brevissima pauciflora calyce
guinquesepalo et ovario tomentoso pubescentibus, pube scepius ferru-
ginea; seminibus scabro-rugosis :—ceet. Jere praecedentis.

Pittosporum glabrum, Putterl. Synops. Pittosp. p. 11, pro parte, non Hook. & Arn.

Var. 8. foliis magnis etiam adultis subtus bracteisque awrato-tomentosis
scepius oblongo-oblanceolatis et breviter acuminatis.

Var.? y. foliis parvulis confertis etiam junioribus glabris.

Has. Hawaii, Sandwich Islands; on the coast southeast of the
crater of Lua Pele and Puna. On Mouna Loa at an elevation of
7,000 feet. (@. Kaala Mountains, Oahu; also on mountains of Kauai.
y.? District of Waimea, Hawaii; and Kaala Mountains, Oahu. (Also
collected by Douglas.)

Apparently a large shrub, with stout branches; the branchlets
often pubescent. Leaves coriaceous, usually thick and firm, crowded
at the end of the branches, oblong-obovate, or somewhat cuneiform,
with the apex mostly rounded, obtuse or even retuse, always tapering
at the base, conspicuously feather-veined and reticulated, glabrous
and shining above, clothed underneath, at least when young, with a
tomentose pubescence, which is usually rusty or ferrugineous, from 3 to
5) or even 8 inches long, by one to 2, or even 3 in width, on petioles
of 6 to 12 lines in length; its margins entire and revolute. Inflores-
cence very short, usually lateral and below the leaves of the season,
few-flowered, tomentose or pubescent; the peduncles only 2 or 3 lines
long; the pedicels very short, sometimes solitary in the axils of the
leaves. Calyx pubescent, of 5 wholly distinct sepals. Corolla, stamens,
&c., nearly as in the foregoing species, but the ovary is densely tomen-
tose or pubescent. Capsule tomentulose or glabrate, two-valved, consi-
derably compressed parallel to the valves, many-seeded; the valves
coriaceous, pretty thick, at length flat after dehiscence, and then

232 PHANEROGAMIA.

almost orbicular in outline, 9 or 10 lines in diameter, slightly grooved
on the back, especially next the base, where they are sometimes a
little emarginate, the surface even or slightly roughened. Seeds very
numerous, closely packed in two rows in each cell, vertically more or
less compressed, 2 to 2% lines in diameter; the purple or brownish
testa rugose-roughened (Plate 19, Fig. 13), usually dull—Of the
various forms of this species which occur in the collection, one, the
var. @., has the leaves mostly acuminate, 5 or 6 inches in length and
proportionally narrow (oblong-lanceolate), the lower surface, as well as
the branchlets, bracts, &c., clothed, even at maturity, with a dense
brownish-yellow tomentum. In the typical forms the down is seldom
so yellow or so persistent. The specimens from Mouna Loa and from
the district of Puna have smaller leaves, of very thick texture, and
their young capsules are densely tomentose. Another variety, if such
it be, the var. y., has usually smaller leaves, only 12 to 3 inches in
length, much crowded at the summit of the branches, and entirely
glabrous, even when young; but the inflorescence and calyx are more
or less tomentose-pubescent. It must be left for a more complete
suite of specimens, in fruit as well as in flower, to determine the true
limits of this apparently polymorphous species.

16. PirTosPORUM CONFERTIFLORUM, Sp. Nov. (Tab. 19.)

P. foliis crasse corvaceis obovatis supra glabris subtus ramisque novellis
lanato-tomentosis ; umbella terminali subsessili confertiflora; calyce
quinquesepalo majusculo pedicellisque tomentosissimis; petalis in
tubum calyce vix duplo longiorem subcoalitis ; capsula ovoideo-globosa
subcompressa, tomentosa, tuberato-rugosa bivalvi polysperma, valvis
crassis lignosis; seminibus atropurpureis rugulosis.

Has. At the base of the crater of Haleakala, east division of
Maui, Sandwich Islands.

“Tree 20 feet high, with odoriferous flowers.” Branches very stout,
leafy, woolly when young. Leaves very thick and coriaceous, alternate,
but crowded on the branches, sometimes imperfectly whorled, obovate,
or obovate-oblong, obtuse, or often with a short acumination, tapering
at the base into a short petiole, the nascent ones woolly both sides,
when full-grown very glabrous above and somewhat shining, densely

PITTOSPORACE SA, 233

tomentose underneath with fulvous wool, which is more or less deci-
duous with age, 3 or 4 inches in length, 14 to 2 inches wide towards
the apex, strongly and copiously feather-veined, the upper surface
finely bullate-reticulated; the margins entire and revolute. Flowers
very numerous and crowded in a dense, terminal and nearly sessile
umbel, large for the genus. Pedicels half an inch long, thick, very
woolly, as are the bracts, &. Calyx very woolly externally; the sepals
5, distinct, orbicular-ovate, obtuse, imbricated, 3 lines long. Corolla
glabrous, probably white (but the colour not recorded); the petals of
a thickish texture, connivent, and above somewhat cohering, into a tube
of 5 or 6 lines in length, their broadly ovate summits 3 or 4 lines
long, forming a spreading limb, imbricated or sometimes convolute in
zstivation. Stamens nearly as long as tube of the corolla: filaments
filiform-subulate: anthers linear-sagittate, apiculate. Pistil 6 lines
long. Ovary sessile, oblong, tomentose, incompletely two-celled; the
cells many-ovuled. Style thick, glabrous, about as long as the ovary,
terminated by a truncate undivided stigma. Capsule globular-ovoid,
somewhat flattened parallel with the valves, tomentose, becoming gla-
brate with age, two-celled, two-valved, many-seeded ; the valves thick
and woody, tubercular-wrinkled and rough externally, nearly an inch
in length and breadth. Seeds dark purple, compressed and angled,
closely packed in two rows in each cell; the testa dull, minutely
tuberculate-rugose: funiculi scarcely any.

The flowers of this remarkable species are among the largest of the
genus: they are said to be odorous. If of a good colour, the species
would be a very desirable one in cultivation.

PLATE 19.—PITTOSPORUM CONFERTIFLORUM: in flower and in fruit, of
the natural size. Fig. 1. A flower, with its pedicel and bract, of the
natural size. 2. Diagram of a flower. 3. Corolla, detached and laid
open, enlarged. 4. Stamens and pistil, equally enlarged. 5. Pistil,
transversely divided, more enlarged. 6. Longitudinal section of the
same. 7. An ovule, more magnified. 8. A valve of the capsule, of
the natural size; inside view, showing the seeds, &. 9. Transverse
section of a dehiscent capsule. 10. A seed, enlarged. 11. Vertical
section of the same, showing the minute embryo in the base of the
albumen. 12. Seed of PirrosporuM GLABRUM, enlarged. 13. Seed of

PITTOSPORUM TERMINALIOIDES, equally enlarged.
59

Or. AURANTIACE SZ.

1 ATALANTIA, Correa.
1. ATALANTIA MONOPHYLLA, DC.

Haz. On a small island in the Sooloo Sea. (Flowers occasionally
trimerous.)

2, TRIPHASIA, Lour.
1. TRIPHASIA MONOPHYLLA, DC.

Has. On mountains, in the neighbourhood of Bajios, Luzon.
(Without flowers or fruit.)

2. TRIPHASIA TRIFOLIATA, DC.

Has. In the vicinity of Manilla, Luzon.

3. SCLEROSTYLIS, Blume.
1. SCLEROSTYLIS ATALANTIOIDES, Wight & Arn. 2
Has. Mangsi Islands, in the Sooloo Sea.

This is the same as No. 991 of Cuming’s Philippine Collection. I
am unable satisfactorily to determine it.

AURANTIACEA. 206

4. MURRAYA, Kenig.

1. Murraya Exotica, Linn.

Has. Philippine Islands, near Manilla, Luzon.

0. MICROMELUM, Blume.

1. MicromELuM puBEscENs, Blume.
Micromelum pubescens, Blume, Bijdr. p. 138.
Has. Mangsi Islands, in the Sooloo Sea. (In fruit only.)

The leaflets in this, as in Javan specimens, are often acutely acu-
minate, and from 9 to 13 in number. They are glabrate, or perfectly
glabrous, except when young. The fruit is oval and about 4 lines

long.

2. MICROMELUM GLABRESCENS, Benth.

Micromelum glabrescens, Benth. in Hook. Lond. Jour. Bot. 2, p. 218.
Limonia minuta, Forst. Prodr. Fl. Ins. Austr. p. 33; DC. Prodr. 1, p. 536.

Has. Friendly, Samoan or Navigators’, and Feejee Islands: com-
mon.

“A shrub 4 to 10 feet high.’—This is very closely related to
Blume’s IM. pubescens ; from which it is to be distinguished mainly by
its smaller flowers (only a line and a half in length), and its oblong
Jruit, which when full grown is half an inch in length. Mr. Bentham
seems not to have been aware that it is the Limonia minuta of Forster,
who found it at the Friendly Islands.

236 PHANEROGAMIA.

6. CITRUS, Linn.
1. Crrrus Decumana, Linn.

Has. Tongatabu, and Feejee Islands. Introduced and cultivated.
(The Shaddock.) |

Another species, mentioned by Dr. Pickering under the name of
Citrus torosa (a name which I do not find anywhere published), is
said to be introduced into the Samoan, Tonga, and Feejee Islands.
There are no specimens in the collection.

Orv. MELIACES.

1. AGLATA, Lour.
1. AGLAIA oporata, Lour.

Aglaia odorata, Lour. Fl. Cochinch. 1, p. 216; DC. Prodr. 1, p. 537; Hook. & Arn.
Bot. Beech. Voy. p. 174, t. 24.

Has. Singapore. (In flower only, with the ovary sterile.)

2. AGLAIA SAMOENSIS, Sp. Nov.
A. foliolis 5-11 oblongo-lanceolatis acuminatis basi rotundatis vel acutius-
culis subtus ramulisque minutissime Jerrugineo-lepidotis glabratis ;

paniculis compositis ; fructu immaturo biloculari.

Haz. Tutuila, one of the Samoan or N avigators’ Islands.

MELIACE &. 237

Shrub or tree with the branchlets, &c., rusty with a fine reddish scurf,
becoming smooth with age. Leaves pinnate, 6 to 15 inches long,
including the petiole, alternate, occasionally almost opposite. Leaflets
5 to 11, oblong-lanceolate, from 5 to 5 inches in length, acuminate,
either obtuse, rounded, or acutish at the base, petiolulate, somewhat
repand, rather membranaceous in texture, glabrate or, at least when
young, rusty underneath with a minute ferruginous scurf. Panacles com-
pound, ample, axillary, very many-flowered, clothed with a reddish
scurf, as is the calyx also. Flower-buds less than a line in diameter,
globose. Petals orbicular-obovate, concave or connivent, of a thickish
texture, imbricated in estivation, much longer than the rusty-pubes-
cent calyx. Stamineal tube obovoid-turbinate, with scarcely any
contracted base, the border obscurely five-toothed, with 5 sessile,
nearly included anthers alternate with the teeth. Ovary two-celled?
apparently sterile in the flowers examined. Immature fruit obovoid,
half an inch in length, two-celled; only one cell containing a seed in
the single fruit contained in the collection. The immature seed oblong,
apparently destitute of arillus, but with a fleshy integument.

3. AGLAIA (MILNEA) EDULIS.
Milnea edulis, Roxb. Fl. Ind. 1, p. 637, & ed. Wall. 2, p. 430.
Has. Samoan and Feejee Islands. (In flower only.)

Probably the “complete, fleshy arillus, covering the seed” in Milnea
is only a fleshy integument of the seed; which would abolish the
only character left to distinguish Milnea from Aglaia. The petals in
the present species are imbricated (not convolute) in estivation.

4, AGLAIA? BASIPHYLLA, Sp. Nev.

A. ramulis junioribus petiolis costisque subtus pube rufa furfuraceo-
hirsutis ; foliolis 5-7 elongato-oblongis membranaceis petiolulatis, in-
fimis minoribus scepissime ime basi petioli adproximatis; floribus
axillaribus glomeratis subsessilibus.

Has. Ovolau, Feejee Islands.
60

238 PHANEROGAMIA.

The specimens of this tree or shrub, having young fruit only, are
not in a state for determination. I can only conjecture that the plant
may be an Aglaia. The branchlets, petioles, and the midrib of the
leaflets underneath, at least when young, are thickly clothed, not only
with the scurf which is common in the genus, but with a stellate hir-
sute pubescence, of a rusty-red colour. The leaflets, 5 to 7 in number,
are elongated-oblong, membranaceous, glabrous, mostly obtuse at both
ends, rather long-petiolulate, the larger, especially the superior ones,
from 3 to 6 inches in length; the lowest pair usually closely approxi-
mate to the base of the petiole, so as to appear like a pair of leafy sti-
pules; they are smaller and shorter than the others, an inch or two
in length, often subcordate at the base. A short spur in the axils
bears apparently a small number of flowers, in a nearly sessile cluster.
The young fruits are cylindraceous or club-shaped, 5 lines long, ferru-
ginous-tomentose, incompletely two-celled, the base surrounded by the
five-toothed, ferruginous-hirsute, persistent calyx.

T'wo other plants, apparently congeners of the last, occur in the
collection from the Feejee Islands, in a state too imperfect for identi-
fication or description; the specimens having only one or two half
grown (one-celled, or at first two-celled) fruits. They do not accord
with the brief characters of any of Blume’s species of Aglaia.

2, HARTIGHSEA, A. Juss., ex parte.

1. HARTIGHSEA SPECTABILIS, A. Juss.

Hartighsea spectabilis, A. Juss. Meliac. (in Mem. Mus. Par. 9,) p. 76 & 111;
Hook. Ic. Pl. t. 615, 616; Hook. f. Fl. N. Zeal. p. 39.
Trichilia spectabilis, Forst. Prodr. p. 83; A. Rich. Fl. N. Zel. p. 306.

Has. Bay of Islands, New Zealand. (With young flower-buds
only.)

Dr. Hooker has overlooked Mr. Bennet’s remark (in Pl. Jav. Rar.
p. 170), that this is not a true species of Hartighsea, but differs from
the typical species “in the entire want of cohesion between the petals
and the stamineal tube, and consequently of the petals inter se (not-

MELIACES. 239

withstanding the description given by Forster), and in several other
characters of minor importance.” He has moreover conformed the
generic character to the exception; the phrase “ petala bast cum tubo
stamineo obscure coalita,” being inapplicable to the typical H. Fraseriana,
in which the cohesion extends to the middle of the stamineal tube,
according to the figure in M. de Jussieu’s Memoir.* I have not seen
that species, nor does its fruit appear to be known; but except in
its shorter tubular disk, it seems to differ in no essential particular
from the older genus Didymochiton, to which I should confidently
refer H. Forsteri, Juss., along with a species in the present collection.
If this view were adopted, the name of Hartighsea might be retained
for H. spectabilis and H. Billardieri.

3. DIDYMOCHITON, Blume.

DipyMmocuitTon (v. Didymocheton), Blume, Bijdr. p. 177; Juss. Meliac. |. ¢. p. 78, t.
16, f. 14.
HarTIGHSE, Spec. Juss. Meliac. p. 76.

1. Dipymocuiron Rican, Sp. Nov. (Tab. 20.)

D. glabratum, foliolis 5-8-jugis oblongis seu lanceolato-oblongis subacu-
minatis bast rotundato-ineequilateris ; paniculis compositis thyrsoideis
racemifloris; floribus scepius tetrameris; columna staminea et disco
glabro tubulart minutissime crenulatis; ovario tri—quadriloculart,
loculis unrovulatis.

Has. Feejee Islands: Vanua-levu, at Sandalwood Bay; Somu-
somu; Nukulau.

A tree of considerable size, the younger parts softly canescently
puberulent, soon glabrate. Petioles a foot or more in length, velvety-
puberulent or glabrate, sometimes sparingly pubescent along the upper

* The genus was not named after G. L. Hartig, who was, moreover, a German, not a
‘‘ French” author; but in honour of a Dutch navigator, whose name, though not given by
Jussieu, must have been Hartighs or Hartighse.

240 PHANEROGAMIA.

side. Leaflets 5 to 8 pairs, sometimes with and sometimes without a
terminal leaflet, oblong, or oblong-lanceolate, 3 or 4 inches in length, or
the lowermost 1% to 2 inches in length, and from an inch to 12 inches
in breadth, slightly acuminate, entire or slightly repand, wnequal at the
buse, the upper side rounded, the lower oblique, often subcordate; the
midrib above and the partial petiole (3 or 4 lines long) tomentulose ;
the surface glabrous, except usually a bearded pubescence in the axils
of the primary veins underneath. Panicles compound, axillary, many-
flowered, thyrsoid, the flowers racemose on its spreading branches, most
frequently teframerous. Pedicels extremely short. Calyx half a line
long, tomentulose, cupulate, 4—5-toothed. Corolla “white,” 3 lines
long; the petals spatulate-linear, slightly pubescent externally, cohe-
rent with the stamineal tube to the middle, valvate in estivation.
Stamineal tube nearly as long as the corolla, cylindrical, a little hairy
outside, its orifice minutely multicrenulate, not setigerous. Anthers 8
in the tetramerous, 10 in the pentamerous flowers, sessile within the
orifice of the tube, oblong-linear, muticous. Hypogynous disk tubular,
between lageniform and urceolate, glabrous, sheathing the ovary and
the base of the style, its orifice minutely 8-10-crenulate. Ovary
ovoid, very villous, 3-4-celled: style filiform, hairy below, as long as
the stamineal tube; the stigma dilated and disk-shaped. Ovules soli-
tary in each cell, amphitropous-ascending; the micropyle superior.
Immature fruit ovoid, 3-4-celled, probably baccate, 3-4-seeded, or by
abortion 1—2-seeded.

From the character given by Jussieu, and the detailed description
reproduced by him from Forster’s manuscript, I should have taken
this plant for the Trichilia alliacea of Forster, the Hartighsea Forsteri,
A. Juss.; were it not that the “interior nectary” (the tubular disk)
in that plant, is said by Forster to be villous, and the exterior, or sta-
mineal tube, according to Jussieu, has its crenatures setigerous. At
least Forster's plant (which came from an island in nearly the same
longitude, only three degrees farther south) must be a close congener
of ours, which appears to me to belong to Blume’s genus Didymo-
chiton, notwithstanding the gamophyllous calyx, the rather shorter
tube of the disk, and the solitary ovules.

Pate 20.—Dipymocuiron Ricui: flowering branch, of the natural
size. Fig. 1. A flower. 2. Flower, with the calyx removed and

MELIACEA, 241

corolla, the adnate stamineal column and the tubular disk laid open.
3. An anther, seen from within. 4. The same, seen from without.
5. Vertical section of an unexpanded flower. 6. Transverse section
of the ovary.—All these details variously magnified. 7. Portion of
inflorescence with unripe fruits, of the natural size. 8. Vertical, and
9, transverse section of an unripe fruit.

2. DIDYMOCHITON GAUDICHAUDIANUM, A. Juss., 1. ¢.?

Has. Mangsi Islands, in the Sooloo Sea. (Without flowers or ripe
fruit.)

4, DYSOXYLON, Blume.
1. Dysoxyton? SAMOENSE, Sp. Nov.

D. foliolis 6-9-jugis cum vel absque impari oppositis oblongis acuminatis
basi obtusis vel rotundatis; paniculis multifloris; fructibus immaturis
obovatis.

Has. Manua and Tutuila, Samoan or Navigators’ Islands.

This brief character is taken from two specimens, one with flower-
buds and developing leaves; the other with full-grown, ample leaves,
and immature fruit. It is not absolutely certain, therefore, that the
two belong to the same species. From these unsatisfactory materials,
I can only conjecture that the plant belongs to Blume’s genus Dysoxy-
lon. It cannot be a Hartighsea nor a Didymochiton, however those
genera be limited; for the petals are decidedly imbricated in estiva-
tion, as well as free from the stamineal tube.—The size of the tree is
not recorded, nor is the wood said to have any odour. The branches
and leaves are glabrous, except a minute and cinereous pubescence
when very young, which is more or less persistent on the petioles
and the inflorescence. Petiole with the rhachis one or two feet long.
Leaflets 6 to9 pairs, sometimes with a much smaller terminal one, oftener
abruptly pinnate: they are opposite, oblong, acuminate, from 5 to 7

61

242 ‘ PHANEROGAMIA.

inches long when full grown, membranaceous, glabrous, mostly
rounded and slightly oblique at the base, on partial petioles of 2 or 3
lines in length; or the terminal one when present with a stalk an
inch long. Punicles many-flowered, axillary, much shorter than the
leaves, compressed. Flower-buds crowded on the extremity of the
stout branchlets of the panicle, on very short pedicels. Calyx silky-
puberulent, 2 lines in diameter, of a thick and firm texture, four-cleft ;
the lobes orbiculate, imbricated. The unexpanded corolla scarcely
exceeding the calyx; the petals orbicular or broadly ovate, silky-
pubescent outside, imbricated in sestivation, free from and longer than
the glabrous, cup-shaped or urceolate, eight-toothed stamineal tube,
which bears 8 sessile and nearly included anthers, alternate with its
lobes. Hypogynous disk annular or cup-shaped, not so high as the
four-celled ovary. Style short: stigma peltate and disk-shaped.
Ovules in pairs in each cell. Lmmature fruit obovate, nearly an inch
long, 3-4-celled, of a rather soft texture, it is uncertain whether
baccate or capsular.

5 MOSCHOXYLUM, A. Juss.

1. Moscuoxytum RicHARDIANUM, A. Juss.

Moschoxylum Richardianum, A. Juss. Meliac. p. 86; Walp. Repert. 1, p. 433.
Trichilia Richardiana, A. Juss, in St. Hil. Fl. Bras. Mer. 2, p. 78.

Has. Brazil, near Rio Janeiro.

2. MoscHoxYLuM ELEGANS, A. Juss.

Moschoxylum elegans, A. Juss. Meliac. 1. ¢. p. 87, t. 19, f. 19.
Trichilia eleyans, A. Juss, in St. Hil. Fl. Bras. Mer. 2, p. 79, t. 98.

Has. Brazil; with the preceding species: both apparently not un-
common.

MELIACES. 943

6. GUAREA, Lunn. .
1. GUAREA PURGANS, A. Juss.

Guarea purgans, A. Juss. Meliac. 1. c. & in St. Hil. Fl. Bras. Mer. 2, p. 83.

Has. Brazil, near Rio Janeiro.

2. GUAREA TUBERCULATA, A. Juss.

Guarea tuberculata, A. Juss. Meliac. 1. c. & in St. Hil. Fl. Bras. Mer. 2, p. 83, t. 100.

Has. Brazil: with the preceding species.

7. XYLOCARPUS, A. Juss.

1. XyLocarPus GRANATUM, Keenig.
Xylocarpus Granatum, Kenig; Roxb. Fl. Ind. 2, p. 240; A. Juss. Meliac. p.

92, t. 20, f. 22.
Carapa Moluccensis, Lam. Dict. 1, p. 621; DC. Prodr. 1, p. 626.

Has. Feejee, Tonga, and Samoan or Navigators’ Islands: on the

coast.

2. XYLOCARPUS OBOVATUS, A. Juss, J. c.?

Carapa obovata, Blume, Bijdr. p. 179?

Has. Feejee Islands: in fruit only. Possibly only a variety of the
preceding species.

Imperfect fruiting specimens of an apparently Meliaceous tree like-
wise occur in the collection, which I cannot determine. The tree is

|
244 PHANEROGAMIA.

said to occur in cultivated ground only, planted by the natives, at
Tongatabu, Savai, one of the Navigators’ Islands, and Tahiti. It is
mentioned by Dr. Pickering as the Mallea Rothu, A. Juss. (which I
observe A. Richard enumerates as collected by Lesson at Tongatabu) :
but it cannot be that plant, nor of the same genus.

x*, MELIACHIS AFFINE.

8. VAV AA, Benth.

‘ Calyx 4-T-fidus, persistens; lobis triangulari-ovatis cestivatione leviter
umbricatis. Petala lobis calycis numero equalia, hypogyna, ligulato-
oblonga, utringue sericeo-puberula, cestivatione leviter imbricata, deci-
dua. Stamina 12-20, incequilonga: filamenta basi glabra in tubum
disco hypogyno cupuliformi carnoso adnatum monadelpha, superne
imtus hirsutissima: antherce introrse, biloculares, loculis longitudi-
naliter dehiscentibus. Ovarium intra discum sessile, 3-4-loculare, in
stylum columnarem extensum: stigma crassum, peltatum, obscure 3—-4-
radiatum. Ovula in loculis gemina, angulo centrali affixa, collate-
ralia, adscendens, subamphitropa, micropyle supera. Bacca globosa,
abortu 1—4-sperma.—Arbuscula glabella; foliis simplicibus obovato-
oblongis alternis obsoletissime punctatis; pedunculis ex azillis supe-
rioribus multifloris ; floribus cymosis.

Vavama, Benth. in Hook. Lond. Jour. Bot. 2, p. 212.

1. VAy#A Amicorum, Benth. l.c. (Tab. 16.)

Has. Muthuata and Vanua-levu, Feejee Islands. Also, Tongatabu,
Friendly Islands. (Vavao, Friendly Islands, Mr. Barclay.)

Apparently a small tree (the size not recorded), glabrous or glabrate,
except the nascent parts (which are silky-pubescent) and the flowers.
The bark is said to be bitter, like that of Canella. Leaves alternate,
crowded towards the end of the branches, simple, entire, obovate-oblong,
obtuse, or retuse, or sometimes obscurely acuminate, occasionally
cuneate-obovate, always acute at the base, tapering into a petiole of

MELIACES. 245

eal

an inch or less in length, chartaceous in texture, pinnately veined
with 6 or 7 pairs of rather prominent primary veins, connected by
minute, reticulated veinlets, obscurely and minutely punctate with pel-
lucid dots: they are from 3 to 6 inches long and 12 to 3 inches wide.
Stipules apparently none; what were taken for linear-lanceolate and
villous deciduous stipules being apparently nascent leaves or perule.
Peduneles solitary in the axils of the upper leaves, shorter than they,
naked below, many-jlowered at the summit; the flowers in a compound
cyme; its branches puberulent, with ovate-subulate bracts and bract-
lets. Calyx 4—7-cleft below the middle, free, puberulent, persistent ;
the lobes triangular-ovate, lightly imbricated in cstivation. Petals as
many as the lobes of the calyx and alternate with them, hypogynous,
lagulate-oblong, of a thickish texture, clothed with a fine and close
silky pubescence on both sides, lightly imbricated in cestivation, recurved-
spreading from an erect base in anthesis, twice or thrice the length of
the calyx, 2 or 3 lines long, deciduous. Stamens 12 to 20, or perhaps
21, usually 12 in tetramerous flowers, and about 15 in pentamerous
flowers, therefore normally thrice the number of the petals, more or
less unequal in length, hypogynous, free from the petals, shorter and
less deciduous than they: filaments flat, linear, glabrous towurds the
base, where they are monadelphous, and also coherent with the whole
exterior face of an annular, cup-shaped, and fleshy, hypogynous disk,
above distinct, but closely approximate in a single series, very hirsute
towards the summit; the long and dense beard mostly borne on the
inner face: anthers oval, emarginate at both ends, two-celled, fixed by
the sinus at their base to the narrowed apex of the filament, introrse ;
the cells opening longitudinally for their whole length. Disk sur-
rounding the ovary and equalling it in length, rather thin, minutely
pubescent inside. Ovary 3—4-celled, free, ovoid, densely hirsute-pubes-
cent, closely sessile within the disk; the cells small in proportion to
the thick and fleshy parietes. Style columnar, rather thick, as long
as the stamens, pubescent below: stigma peltate or disk-shaped, thick,
undivided, the summit obscurely 3-4-radiate. Ovules 2 in each cell,
attached to the inner angle near the base, ascending, collateral, between
orthotropous and amphitropous, the micropyle superior. The ripe fruit
is unknown: the young state, however, manifestly indicates it as a
fleshy or pulpy, spherical berry, of small size, stipate by the persistent
calyx, with small cells, some, or perhaps all but one, of which (and a
single seed ?) may be suppressed in ripening.
62

246 PHANEROGAMIA.

The ripe fruit and seeds are still wanting to the full illustration of
this remarkable plant. Until the latter are known, the affinity of the
genus cannot be positively made out. Mr. Bentham thought it allied
to Leionanthes of Jack, itself a genus of very obscure affinities; but,
having subsequently examined specimens of the latter, he no longer
recognises the relationship. Mr. Rich, the Botanist of the Expe-
dition, in a note annexed to the specimens, suggests its relationship
with Canella (another genus of doubtful place), guided probably by
some similarity in sensible properties, and by the number and union
of the stamens. There may be, indeed, a real, although not imme-
diate, affinity between these two genera, notwithstanding the ex-
trorsely adnate anthers and the very different ovules of Canella. I
suspect, however, that Vavea is scarcely to be separated from the
Meliacee, from which its floral characters differ only in the stamens
being (at least for the most part) above twice the number of the
petals, and in the lesser union of their filaments. In the leaves of
Quivisia, if I mistake not, traces of similar dots in the leaves may be
discerned.

The specimen which I find figured (to which I have merely caused
new and full details to be added), represents a form of the plant with
larger leaves and stouter branchlets than that described by Mr.
Bentham. Other specimens of the present collection have slender
branchlets, and smaller leaves (only two or three inches in length),
with a little sparse pubescence remaining on their lower surface; but
there seems to be no specific difference between them. In both forms
the sepals and petals vary from four to seven.

Pirate 16, B—Vav#A AmicoruM: flowering branch of a large-
leaved form, of the natural size. Fig. 1. Diagram of a (pentamerous)
flower. 2. An expanded (pentamerous) flower. 38. Pistil, with the
androecium laid open and displayed. 4. A stamen, seen from within.
5. Exterior view of the same. 6. Vertical section of a flower, dividing
the ovary, and displaying the ovules, &. 7. Unripe fruit, with the
persistent calyx. 8. Transverse section of the same.—The analyses
variously magnified.

SAPINDACES. OAT

Orv. SAPINDACE &.

1 CARDIOSPERMUM, Linn.

1. CARDIOSPERMUM MICROCARPUM, H. B.K.
Has. St. Jago, Cape de Verde Islands. Tahiti and Matia, Society
Islands; in wild situations. Feejee Islands: apparently introduced.

Gardner’s and Birney’s Islands.

A species now as widely diffused as C. Halicacabum, from which it
differs chiefly in its much smaller and turbinate fruit; but perhaps
only as a variety, as Blume, in Rumphia, 3, p. 184 (cited by Walpers)
regards it.

2. CARDIOSPERMUM LOXENSE, HZ. B. K.

Has. Peru, near Lima. (Apparently referable to this species; but
with the mature pods perfectly glabrous.)

2, SERJANIA, Plumier.
1. SERJANIA CUSPIDATA, Camb.

Serjania cuspidata, Cambess. in St. Hil. Fl. Bras. Mer. 1, p. 356.
Paullinia Guarumima, Velloz. Fl. Flum. 4, t. 39.

Has. Brazil: in hedges near Rio Janeiro. (In flower only.)

2. SERJANIA COMMUNIS, Camb.? I. ¢.

Has. Brazil, in the Organ Mountains, near Rio Janeiro. (In
flower only.)

248 PHANEROGAMITIA.

3. PAULLINIA, Linn.
1. PAULLINIA MULTIFLORA, Camb.
Paullinia multiflora, Cambess. in St. Hil. Fl. Bras. Mer. 1, p. 379.

Has. Brazil, in the Organ Mountains. (In flower only.)

At least this is the plant of Martius, distributed under the above
name; but in some points it does not accord with the character given
by Cambessedes.

2. PAULLINIA WEINMANNIAFOLIA, Mart.

Paullinia weinmanniefolia, Mart.! in Flora, 20, Beibl. 2, p. 91; Walp. Repert. 2,
p. 414.
P. discolor, Gardn.! in Hook. Lond. Jour. Bot. 1, p. 337.

Has. Brazil, near Rio Janeiro. (In flower only.)

3. PAULLINIA RUBIGINOSA, Cumb, ?
Paullinia rubiginosa, Cambess. in St. Hil. Fl. Bras. Mer. 1, p. 8371?

Has. Brazil, in the Organ Mountains.
The specimen (in flower only) nearly accords with the character of

P. rubiginosa. The sparse teeth of the leaflets are tipped with a
slender and setaceous cusp.

4, PAULLINIA FALCATA, Gardn.

Pauilinia falcata, Gardn. in Hook. Lond. Jour. Bot. 1, p. 580.

Has. Brazil, in the vicinity of Rio Janeiro.

SAPINDACEAE. 249

Except that the racemes are many-flowered, and the wings of the
fruit free from the style, I should take this for P. melic/folia, Juss.,
with the figure of which the foliage closely agrees.

5. PAvuLiintA BARBADENSIS, Jacq. ?

Has. Brazil, near Rio Janeiro. (In flower only.)

4, SCHMIDELIA, Linn.
1. ScHMIDELIA L&vIs, Camb.
Schmidelia levis, Cambess. in St. Hil. Fil. Bras. Mer. 1, p. 382.

Has. Brazil, in the vicinity of Rio Janeiro.

2. SCHMIDELIA RACEMOSA, Linn.

Has. Island in the Sooloo Sea. (In flower only.)

This is most likely the Allophyllus ternatus (Schmidelia Cochinchi-
nensis, DC.), and the Gemella trifoliata (Aporetica Gemella, DC.) of
Loureiro. JS. serrata, DC., is perhaps the same species, as Wight and
Arnott suspect.

3. SCHMIDELIA OBOVATA, Sp. Nov.

S. glabra; foliis trifoliolatis ; foliolis obovatis membranaceis subdentatis ;
petiolis angulatis ; racemis compositis; petalis filamentisque longissime
barbatis.

Schmidelia glabra, Benth. in Hook. Lond. Jour. Bot. 2, p. 218, non Ornitrophe
glabra, Roxb.

Has. Upolu and Savai, Samoan Islands. Mangsi Islands, in the

Sooloo Sea.
63

250 PHANEROGAMIA.

A branching shrub, about 12 feet high, glabrous, except a minute
and deciduous pubescence on young parts. Stipules none. Leaves
trifoliolate, on angled petioles of 2 or 3 inches in length. Leaflets obo-
vate, ample, sometimes 5 or 6 inches in length and 3 or 4 in width,
irregularly and mostly obscurely toothed, membranaceous in texture,
rounded or acute and tapering at the base, more or less petiolulate;
the apex obtuse, rounded, or sometimes obtusely acuminate. Pedun-
cles shorter than the petiole. Racemes compound, or at least trifid,
rarely simple, densely-flowered, shorter than the leaves. Flowers
polygamous, about a line in diameter, on pedicels of a line or more in
length. Sepals 4, orbicular-obovate, white. Petals 4, much shorter
than the sepals, deltoid-ovate, unguiculate, with the margins involute-
appendaged at the base, within densely bearded with very long, villose
hairs. Glands of the disk 4, scale-like, two of them larger. Stamens
7 or 8, shorter than the sepals: filaments sparsely bearded with long
hairs, like those of the petals; the didymous anthers also sparingly
pilose. Ovary didymous, only one of the lobes ripening into a globose
berry, of the size of a small pea.

This species indeed resembles the East Indian & glabra ( Ornitrophe
glabra, Roxb. in Wall. Cat.), to which Bentham referred specimens
from New Guinea, New Ireland, and Amboyna; but that has simple
racemes, terete petioles, &. To the present species apparently
belongs Cuming’s No. 1502, from the Philippine Islands. Our plant
does not appear to accord with the characters of any of the species of
Allophyllus cited by Walpers from Blume’s Rumphia; a work which
is not now accessible to me.

5 MOULINSIA, Camé.

1. Movxinsia RuBiGiNosA, Don.

Sapindus rubiginosus, Roxb. Pl. Corom. 1, t. 62 (ex sp. Hort. Cale.) ; DC. Prodr. 1,
p. 608; Wight & Arn. Prodr. Ind. Or. 1, p. 112.

S. fraxinifolius, DC. Prodr. 1, p. 608?

Moulinsia cupanioides, Cambess. Sapind. in Mem. Mus. Par. l. e. p. 27, t. 2?

Has. Caldera, Mindanao, one of the Philippine Islands.

SAPINDACES. 951

The inflorescence of the specimen is undeveloped: but it is plainly
the same as the plant cultivated in the Calcutta Botanic Garden,
under the name of Sapindus rubiginosus, Roxb. In this, the appen-
dage of the petals is bearded with such long and dense woolly hairs
as to render it questionable whether it can be the same as the species
figured by Cambessedes, above cited.

6 HEMIGYROSA, Blume.

1. Hemicyrosa Prerrotteti, blume.
Hemigyrosa Perrottetii, Blame, Rumphia, 8, p. 165, fide Walp. Ann. 2, p. 212, ex char.
Has. Philippine Islands: near Banos, Luzon.

The specimens (in flower only) well accord with Blume’s character,
as cited by Walpers. The fifth petal is destitute of the two-parted,
very villous appendage, and is rather smaller than the others, which
accord in character with those of Hrvoglossum.

7. SAPINDUS, Lann.
1. Saprnpus? Virrensis, Sp. Nov.

S.? foliis glabris; rhachi aptera; foliolis tri-quadryugis summisve
bijugis ovato-oblongis paullo obliquis subacumimatis breviter petiolu-
latis; paniculis terminalibus amplis decompositis; sepalis imcequa-
libus ; petalis ovatis cucullatis unguiculatis haud appendiculatis inferne
ciliatis ; filamentis T-8 inferne villosis.

Has. Rewa and Ovolau, Feejee Islands: common on the leeward
coasts.

A tree “30 feet high,” glabrous, except the inflorescence, which is
clothed with a fine rusty tomentum; the bark of the branchlets warty.

252 PHANEROGAMIA.

Leaves with a wingless rhachis, abruptly pinnate. Leaflets 3 or 4 pairs,
or the uppermost fewer, opposite or alternate, ovate-oblong, or lanceolate-
oblong, tapering to a more or less acumimate or acute point, 3 to 5
inches long, 124 to 2 inches wide, somewhat oblique, entire, chartaceous
in texture, veiny, green and rather shining above, dull but little paler
underneath, rounded at the base, or contracted into a short partial
petiole (of 2 lines in length). Panicle terminal, large, and decom-
pound; the flowers thickly clustered along its racemose branches,
polygamo-dicecious. Pedicels very short. Calyx petaloid, apparently
yellowish-white, 2 lines or more in diameter; the 5 sepals unequal,
distinct, roundish-obovate, concave, imbricated in sestivation, glabrous,
with ciliolate margins; the 2 exterior only half as large as the others.
Petals 5, ovate, or deltoid-ovate, rather shorter than the larger sepals,
white, cucullate, destitute of an appendage, unguiculate ; the claw rather
long, and, as well as the base of the lamina, cil/zate with villous hairs.
Disk complete, crateriform, 7—-8-toothed. Filaments 7-8, villous below
the nuddle, at length longer than the calyx in the sterile flowers,
which alone occur in the collection. Abortive ovary three-lobed,
three-celled, with a short and undivided style. Fruit not seen: the
genus therefore doubtful.

At Kimeo, Society Islands, was collected the foliage apparently of a
Sapindus, with a winged rhachis, and 6 to 8 narrowly oblong leaflets.
It may be the Sapindus saponaria of Forster, from Easter Island, of
which nothing farther is known.

8. CUPANIA, Plumier.

* Oceanice.
1. Cupanta? (Miscuocarpus) FALCATA, Sp. Nov.

C. foltiis 4-5-foliolatis glabris ; foliolis lanceolatis basi rotundatis sensim
acuminatis falcatis; racemis spiciformibus densifloris pubescentibus ;
petals calycem parvum adcequantibus hastato-trilobis intus inappen-
diculatis, lobis lateralibus incurvis margine villoso-barbatis; stami-
nibus glabris; capsula obovoidea subtrigona glabra haud stipitata tri-
loculare. |

SAPINDACE &.

i)
or
SU)

Has. Ovolau and Somu-somu, Feejee Islands: in forests.

“Tree 20 feet high;” the branches and foliage glabrous, except the
young shoots, which are minutely pubescent. Leaves alternate, desti-
tute of stipules, with petioles 2 to 4 inches long. Leaflets 4 to 8, com-
monly 6, lanceolate, with a rounded base, gradually acuminate, falcate,
entire, rather pale, of the same hue both sides, veiny, chartaceous in
texture, 3 to 6 inches long, 9 to 18 lines wide; the partial footstalks
2 to 4 lines long. Racemes densely-flowered and spike-like, axillary
and subterminal, mostly simple, short-peduncled, pubescent, about the
length of the petioles. Pedicels articulated with the rhachis, about
14 lines long, subtended by a minute deciduous bract. Flowers poly-
gamo-dicecious. Calyx small, a line in diameter, five-cleft; the lobes
ovate-triangular, greenish and rather herbaceous in texture, minutely
pubescent externally, valvate or nearly so in estivation; the bud
opening very early. Petals 5, very small when the calyx opens, at
length as long as the sepals, unappendaged within, consisting of a trans-
versely dilated hastate-three-lobed limb, raised on a slender claw, the
lateral lobes involute and villous-bearded on the margin. Stamens 8,
glabrous, purplish, inserted within the glabrous, crenate, and com-
plete, fleshy disk: filaments equal, in the sterile flowers filiform and
thrice the length of the calyx, in the fertile flowers scarcely exserted :
anthers oblong. Ovary sessile, ovoid, pubescent, three-celled, with a
single ascending ovule in cell: style none: stigma small, three-lobed.
Capsule obovoid or turbinate, obscurely three-angled, glabrous, not stypi-
tate, nor with a tapering base, 4 or 5 lines long, very obtuse, pointless,
three-celled, loculicidally three-valved; the valves coriaceo-crustaceous,
thick, smooth externally, densely villous inside. Seeds not seen,
having all fallen from the open capsules.

This plant is a manifest congener of Cupania apetala, Labill. Sert.
Austro-Cal. (judging from the figure), which must belong to the genus
Mischocarpus of Blume; but I know not whether this author, in his
Rumphia, has so referred it. From Walpers, Ann. Bot. 2, p. 216, I
learn that Blume has there characterized a species, Mischocarpus
fucescens, with minute and caducous petals. If the lobes of the calyx
are valvate or nearly so in sstivation in Blume’s species, as they are
in ours, the genus Mischocarpus would appear to be sufficiently distinct

from Cupania.
64

254 PHANEROGAMITIA.,

2, CUPANIA (DIMEREZA) RHOIFOLIA, Sp. Nov.

C.. foluis 8-12-foliolatis glabris ; foliolis oblongo-lanceolatis subacuminatis
basi scepius acutis subtus glaucescentibus; paniculis puberulis laxi-
Jloris; petalis orbiculatis exwnguiculatis intus squamula bipartita
villoso-barbata auctis; filamentis inferne pilosis; capsula profunde
triloba.

Has. Ovolau, Feejee Islands.

“Tree about 30 feet high,” glabrous, except the young branchlets,
&c., which are puberulent. Petiole with the rhachis 3 to 6 inches
long, slender. Leaflets from 8 to 12, or rarely fewer, lanceolate or
oblong-lanceolate, more or less acuminate, mostly acute at the base, or
tapering into a short partial footstalk, green and rather shining above,
with the midrib and veins tinged with purple, glaucescent underneath,
chartaceous in texture, usually alternate, from 14 to 3 inches long,
and from 6 to 12 lines wide. Punicles loosely flowered, nearly equal-
ling the leaves, somewhat compound, axillary and _ subterminal,
minutely puberulent. Pedicels a line long. Flowers polygamous.
Calyx petaloid (white?); the 5 orbicular-obovate sepals rather un-
equal, imbricated in sstivation, glabrous, a line long. Petals 5, orbi-
cular, shorter than the sepals, sparingly ciliate, not unguiculate, rather
longer than their deeply two-parted appendage or squamula, which is
villous-bearded on the inside, and especially on the margins, each lobe
bearing a gland at its tip. Disk complete, fleshy, undulate. Stamens
in the sterile flowers 8; the filaments sparsely hairy below, exserted,
slender, bearing oval anthers. Fertile flowers not seen. Capsule
deeply three-lobed, glabrous, abruptly stipitate (the stipe about a line
and a half long); the lobes wing-like, wider than long (4 or 5 lines
in breadth), chartaceo-coriaceous in texture, loculicidal, glabrous
within. Seeds subgloboge, 3 lines in diameter, enclosed in a fleshy
arillus.

This species is evidently related to Dimereza glauca, Labill. (the

Cupania glauca of Cambessedes) ; the structure of the flowers being
wholly similar.

SAPINDACE £&. 955

3. CUPANIA BRACKENRIDGEI, Sp. Nov.

C. glabella; foliis 6-8-foliolatis ; foliolis oblongis vel sublanceolatis sub-
acuminatis basi attenuatis crebre penninervirs subtus badits ; paniculis
compositis folio brevioribus ; capsula compressa abovato-cuneata trun-
cata bilocular.

Has. Ovolau, Feejee Islands.

A shrub or small tree, nearly glabrous; the young shoots, petioles,
&e., clothed with a minute rusty scurf rather than pubescence.
Petioles angled, much dilated at their insertion. Leaflets 6 or 8,
varying from oblong or ovate-oblong to broadly lanceolate, more or less
acuminate, tapering at the base into a short and margined partial
petiole, 24 to 4 inches long, chartaceous, glabrous or nearly so, entire,
thickly feather-veined (the veins conspicuous and more or less reticu-
lated), the upper surface pale green, the lower tinged of a brownish or
chestnut hue. Flowers not seen. Fruiting panicles compound, shorter
than the leaves, mostly axillary, loose. Capsules compressed, obovate-
wedge-shaped, approaching to obcordate, but the apex is only truncate,
and tipped with a persistent style, 5 lines in length and breadth,
acute at the base, not at all stipitate, clothed with a very minute
rusty scurf, two-celled, crustaceous in texture; the whole interior
lined with a dense coating of bristly hairs. Seeds (immature) soli-
tary, ascending from near the base of the cell, showing only a rudi-
mentary arillus.

4, CUPANIA LEPTOBOTRYS, Sp. Nov.

C. glabella; foliolis 14-20 laneeolato-oblongis acuminatis basi rotundatis
longiuscule petiolulatis glabris concoloribus; paniculis gracillimis
racemifloris ramosis folio cequilongis cum petiolis gunioribus tomen-
tulosis; floribus polygamo-monoicis ; calyce quinquepartito persistente
petalis squameformibus multoties majoribus ; antheris 10 subsessilibus ;
capsula obcordato-triloba stipitata.

Has. Ovolau, Feejee Islands.

256 PHANEROGAMIA.

A shrub “10 to 12 feet high,” nearly glabrous, except the young
shoots, inflorescence, and petioles, which are tomentulose with a close
and soft, more or less deciduous, rusty down. Leaves, including the
petiole, one or 2 feet long, usually abruptly pinnate. Leaflets 14 to 20,
opposite or alternate, /anceolate-oblong, acuminate, rounded at the base,
entire, from 3 to 7 inches long, 12 to 2 inches wide, membranaceous,
loosely veiny, glabrous, of the same green hue both sides; the slender
partial petioles half an inch or more in length. Piunicles very long and
slender, usually equalling the leaves, axillary, sparingly branched, loosely
flowered; the flowers racemose, on pedicels of 2 or 3 lines in length,
polygamo-moneecious. Calyx minutely ferruginous-tomentose exter-
nally, five-parted, 3 lines broad when expanded; the lobes orbicular-
obovate, concave, nearly equal, imbricated in estivation, persistent at
the base of the unripe fruit. Petals minute and scale-like, many times
shorter than the calyx, shorter than the almost sessile oblong anthers of
the male flowers, not unguiculate, glabrous, except a beard on the
minutely bi-appendiculate base within. Stamens 10, at least in the
male flowers (the female flowers not seen), inserted within the small
and thin disk. Capsule obcordate-three-lobed, three-celled, tipped with
a very short style in the sinus, tomentulose with a ferrugineous pubes-
cence, o: a corky texture, somewhat wrinkled when dry, nearly an
inch long, including the stipe of 2 lines in length, the lobes somewhat
carinate, loculicidal; the valves densely pubescent within. Seeds
solitary, ascending from near the base of the cell, enclosed in a mem-
branaceous arillus.

5. CUPANIA LENTISCIFOLIA, Pers.

Cupania lentiscifolia, Pers. Ench. 1, p. 413; DC. Prodr. 1, p. 614; Cambess.
Sapind. 1. ¢.

Guioa lentiscifolia, Cav. Ic. Pl. 4, p. 49, t. 378.

Guaiacum dubium, Forst. Prodr. Fl. Ins. Austr. p. 32.

Has. Tongatabu, Friendly Islands. Upolu, Navigators’ Islands.

The specimen from Tongatabu has neither flowers nor fruit: that
from the Samoan Islands has a few flower-buds. If I mistake not it
is likewise the Guaiacum dubium of Forster, whose specimen, in the
herbarium of the British Museum, is in fruit only.

SAPINDACES. 957

At Upolu, Navigators’ Islands, leafy specimens were collected of an
apparently Sapindaceous shrub, without flowers or fruit; the leaves
with about five pairs of ovate-lanceolate and acuminate leaflets, of a
thin texture, smooth and green on both sides.

* x Philippenses.
6. Cupania? Ricuu, Sp. Nov.

C.? ramulis pubescentibus; foliis paripinnatis 2-3-jugis; foliolis ova-
libus membranaceis glabris concoloribus ; paniculis in axillis fascicu-
latis, fructiferis petiolum cequantibus; capsula subcompressa obovata
glabra biloculari haud angulata apice retusa.

Has. Near Caldera, Mindanao, Philippine Islands. (In fruit only.)

For the want of flowers the genus of this plant cannot be ascer-
tained; but the fruit accords with that of several species of Cupania.
The stem is glabrous; but the young branches, &c., are pubescent.
Leaflets 2 or 3 pairs, oval, more or less acuminate, rounded at the base,
on very short partial petioles, glabrous, membranaceous in texture,
green and of the same hue both sides, very veiny, from 4 to 7 inches
long. Fruiting panicles fascicled im the axils, not exceeding the petiole,
pubescent, apparently many-flowered. The vestiges of a minute five-
parted calyx are distinguishable at the base of the fruit. This is a
smooth and glabrous, moderately compressed, obovoid, two-celled capsule,
8 or 9 lines in length, with a stipe a line and a half long, retuse at the
apex, and apiculate with a short style, the edges rather acute, but not
angled nor produced, loculicidal; the valves crustaceous, and with a
thin and separable cartilaginous endocarp, which is tomentose inside.
Only unripe seeds occur: these are solitary in each cell, ascending,
inserted above the base, with a fleshy arillus forming around the
hilum.

There is another Sapindaceous plant from Caldera, with male
flowers only, its inflorescence similar to the foregoing species, and
with very large bijugate leaflets. It is undeterminable from the

present materials.
65

958 PHANEROGAMIA.

x * * Novo-Hollandice.
7. CUPANIA ANACARDIDOIDES, A. Rich.
Cupania anacardidoides, A. Rich. Bot. Voy. Astrolab. 2, p. 33, t. 13.

Has. Puen Buen, New South Wales. (In fruit.)

8. CUPANIA SUBCINEREA.

C. foliolis 2-6 oblongo-lanceolatis utringue subacuminatis subsessilibus
parce serratis crebre penninerviis supra nitidis subtus ramulis petio-
lisque cinereo-puberulis ; paniculis folio brevioribus ; petalis minimis ;
capsula 2—3-loculart 2—3-loba glabra.

Sapindus cinereus, A. Cunningh. in Herb. Hook.
Has. Hunter’s River, and Puen Buen, New South Wales.

A shrub or small tree. Leaves abruptly pinnate: the leaflets vary-
ing from one to 3 pairs, oblong-lanceolate, or when only a single pair
oblong or oval, rather acuminate at both ends, scarcely petiolulate,
coarsely and sparingly serrate, especially above the middle, rarely
entire, from 2 to 4 inches long, 9 to 15 lines wide, chartaceous in
texture, conspicuously and thickly feather-veined; the upper surface
glabrous and shining; the lower minutely cinereous-pubescent, as are
the petioles, branchlets, and inflorescence, at first brownish in hue, at
least in the dried specimens, at length pale. Flowers very small, in
axillary panicles, which are shorter than the leaves. Calyx five-cleft,
minutely pubescent. Petals minute, much smaller than the lobes of
the calyx, apparently inappendiculate. Stamens 10? with very short
filaments. Capsule 2-3-celled, 2-3-lobed, or tricoccous, at first appa-
rently fleshy, at length loculicidal, glabrous, not at all stipitate, sub-
tended at the base by the minute persistent calyx; the lobes rounded;
the valves glabrous inside. Seed globose, invested with a fleshy
arillus.

SAPINDACES. 259

This species was collected by Mr. Cunningham at Moreton Bay. I
am not aware that it has been anywhere published; and our materials
for describing it are imperfect. There is already a Cupania cinerea.

x x * * Brasilienses.
9. CUPANIA ANACARDIAFOLIA, Gardn.
Cupania anacardiefolia, Gardn. in Hook. Lond. Jour. Bot. 2, p. 338.

Has. Brazil; in the Organ Mountains, not far from Rio Janeiro.
(In fruit only.)

9 NEPHELIUM, Linn.
1. NEPHELIUM PINNATUM, Camb.

Nephelium pinnatum, Cambess. Mem. Sapind. (in Mem. Mus. 18), p. 30.
Pometia pinnata, Forst. Char. Gen. t. 54, & Prodr. Fl. Ins. Austr. p. 74.
Aporetica pinnata, DO. Prodr. 1, p. 610.

Has. Tongatabu, and Feejee Islands; planted around dwellings.
(Fruit sweet and edible.)

10. OTOPHORA, Blume.
1. OtopHorA BuLANco1, Blume.

Otophora Blancot, Blume, Rumphia, 3, p. 142, in obs. ex Walp. Ann. Bot. 2, p. 222.

Sapindus baccatus, Blanco, Fl. Filip. p. 290.
Otolepis nigrescens, Turczan. in Bull. Soc. Nat. Mose. 1848, & Flora, 33, p. 7087

Has. Bafios, near Manilla, Luzon.

The specimen bears only unexpanded flower-buds. On the whole

260 PHANEROGAMIA.

it so nearly accords with the character of Turczaninow’s Ofolepis
nigrescens, founded on Cuming’s No. 1922 (which I have not seen)
from Luzon, that I suppose it to be the same species. But the genus
is plainly identical with Blume’s Otophora, which is probably the
earlier published, and with a considerable number of species. This
name, therefore, I adopt, although too like Otiophora, Zucc. The
stipules, if such they be, of our plant are over an inch in length.

ll. ALECTRYON, Geritn.
1. ALECTRYON EXCELSUM, DC.

Alectryon excelsum, DC. Prodr. 1, p. 617; Hook. Ic. Pl. t. 570; Hook. f. Fl. N.
Zeal. p. 88.

Has. Waia-ruru Bay, &., New Zealand. (In fruit.)

12. DODON ZA, Linn.
1. Dopon#A viscosa, Linn.

Has. Rio Janeiro, Brazil (the form named D. Brasiliensis by
Schlechtendal). Peru, between Lima and Yaso (near the D. Schie-
deana, Schlecht.). Bay of Islands, New Zealand. (Leaves only.)
Tahiti and Matia, Society Islands. (Several forms: one, from Matia,
has large and mostly retuse leaves, and the fruit more deeply notched
at both ends than usual. Another, in fruit, has some of the capsules
surmounted by a long exserted style, of 3 or 4 lines in length! as in
the flowers of the D. viscosa, Cay. Ic. t. 327 (D. Jamaicensis, DC.) ;
while other fruits of the same panicle have the usual short style, not
exserted beyond the sinus of the wing.) Tutuila, Samoan Islands.
Sooloo Islands. Caldera, Mindanao. (Large-fruited forms, mostly with
obtuse leaves, referable to D. Burmanniana, DC.) Feejee Islands.
Sandwich Islands: several widely different forms. On the coast of
Hawaii and on the mountains behind Honolulu, Oahu, some ordinary

SAPINDACEAZ. 261

states of the species were collected; at the latter locality also the
(probably abnormal) form with elongated styles, some of them 5 or 6
lines in length! Similar specimens were gathered at Oahu by Gaudi-
chaud, in the Voyage of the Bonite.

2. DopoNHA SPATHULATA, Smith.

Dodonea spathulata, Smith, in Rees’ Cycl. no. 2; DC. Prodr. 1, p. 616.

Has. Sandwich Islands. On the mountains of Hawaii, near the
crater Lua Pele, and on Mouna Loa, at the elevation of 8,000 feet.
Mountains of Maui, on the eastern side (a narrow-leaved variety).

Although it would be difficult to point out any absolute characters,
yet I cannot but regard the specimens before me as belonging to a
species different from D. viscosa; and I presume they are rightly
referred to Smith’s D. spathulata, which is said to be smaller than the
former. The leaves, in our numerous specimens, vary from an inch
to at most 2 inches in length, and from spatulate-obovate to narrowly
oblanceolate, the apex mucronate, the base tapering into a short
petiole; their texture dry and rather coriaceous; the surface often
somewhat furfuraceous, probably from the exsiccation of the viscous
exudation. The racemes or corymbs are simple, short, and few-
flowered. The orbicular fruits are pretty broadly winged, half an
inch or more in diameter, glabrous or nearly so, with scarcely any
notch at the base, but with a deep and very broad one at the apex.
The style is usually very short; but in one instance I find it elon-
gated, as in the peculiar state of D. viscosa, mentioned above.

3. DopONHA ERIOCARPA, Smith, l. ¢.?
Has. Sandwich Islands. District of Waimea, Hawaii.

To this I doubtfully refer two imperfect specimens, entirely des-
titute of fruit, but with minutely hairy branchlets, &c., which I
should have taken as a form of D. viscosa. None of our specimens
from the interior of the Sandwich Islands exhibit a pubescent or

hairy fruit.
66

262 PHANEROGAMIA.

4. DopONZA TRIQUETRA, Andr.
Dodonea triquetra, Andr. Bot. Rep. t. 280; DC. Prodr. 1, p. 617.
Has. Sydney, New South Wales. Ovolau, Feejee Islands.

Most of the styles, in a flowering specimen from Sydney, are pro-
longed to the length of half an inch, many times exceeding the rest
of the flower. This is doubtless an abnormal state. The plant from
the Feejee Islands appears to be the same.

5. Dopon#A CUNEATA, Smith.
Dodonea cuneata, Smith, 1.c.; Rudge, in Trans. Linn. Soc. 11, p. 296, t. 19, non DC.?

Has. Near Sydney, New South Wales.

The specimens, in fruit only, have the pedicels fully as long as the
capsules ; while in DeCandolle’s plant the flowers are said to be sub-
sessile. Our plant accords very well with Rudge’s figure. It occurs
in Cunningham’s collections under the name of D. con/ferta.

6. DopoN#A NERIUFOLIA, A. Cunn. ined.
Has. Sydney, New South Wales.
This is the plant so named by A. Cunningham in Hooker’s herba-

rium. I do not find it anywhere published; and our specimens, with
male flowers only, are not sufficient for characterizing it.

7. DopoN#A CALYCINA, A. Cunn. ined.

Has. With the preceding species.

Apparently a narrowly linear variety of D. calycina, A. Cunning-
ham, in Hooker’s herbarium, which is probably still unpublished.

Orv. MALPIGHIACE &.

1 BYRSONIMA, Rich.
1. ByRsoNIMA SERICEA, DC.

Byrsonima sericea, DC. Prodr. 1, p. 580; A. Juss. Monogr. Malp. 2, p. 19.
B. Brasiliensis, Griseb. in Linnea, 18, p. 255, ex A. Juss.

Has. Brazil, near Rio Janeiro; common on the coast.

2. BUNCHOSIA, Rich.

1. Buncnosta Martiana, A. Juss.

~ Bunchosia Martiana, A. Juss. Monogr. Malp. 2, p. 80.
Malacmea fluminensis, Griseb. in Linnza, 18, p. 248.
Malpighia maritima, Velloz. Fl. Flum. 4, t. 173.

Has. Brazil, near Rio Janeiro, or in the Organ Mountains.

2. BuncHosiA ARMENIACA, Lich.

Bunchosia Armeniaca, Rich. in Mem. Mus. 18, p. 481; A. Juss. 1. ¢. p. 78.
Malpighia Armeniaca, Cav. Diss. p. 410, t. 238.

Has. Peru. Gathered in a garden at Lima, with immature fruit,
which resembles a young apricot.

264 PHANEROGAMIA.

3 THRYALLIS, Mart.
1. THRYALLIS ROTUNDIFOLIA, Sp. Nov.

T. folus orbiculatis subtus farinoso-incanis ; racemis cymoso-paniculatis
folia vie superantibus ; coccis acutissime cristatis.

Has. Brazil, near Rio Janeiro. (Also in the Brazilian collection
of Pohl, No. 1539.)

A species intermediate in some respects between T. brachystachys
and 7. latifolia; but distinguished from both by its orbicular leaves,
which are not in the least cordate at the base; they are rounded at
both ends, or retuse at the apex, 12 to 2 inches in length and breadth,
glabrous above, or stellately puberulent only when very young,
beneath whitened with a close farinose tomentum: there are usually a
pair of hemispherical glands at the base of the midrib, and often one
terminating its apex. Petiole 3 lines long. Stipules minute, cadu-
cous. Inflorescence and young branches minutely tomentose or fari-
nose. Ltacemes cymose-paniculate at the end of the branches, scarcely
exceeding the leaves. Sepals occasionally 6, ferrugineous-tomentose; in
flower, ovate-rotund and barely 3 lines long; in fruit, becoming lan-
ceolate-oblong and 5 or 6 lines in length. Petals as in -T. brachy-
stachys. Stamens 10, or occasionally 12. Cocci sharply crested, more
so than in T. latifolia, and the sides less strongly varicose; otherwise
nearly similar. They appear externally to be fertile and perfect; but
the cavity in those examined is nearly filled with the fibrous-spong
growth from the endocarp, or placenta, which was noticed by Martius
and Jussieu in 7. latifolia.

4. STIGMAPHYLLON, A. Juss.
1. STIGMAPHYLLON VITIFOLIUM, A. Juss.

Stigmaphylion vitifolium, A. Juss. in St. Hil. Fl. Bras. Mer. 3, p. 50, & Monogr.
Malp. 2, p. 105.

MALPIGHIACES. 265
Has. Brazil; common in the Organ Mountains, near Rio Janeiro.
Some of the leaves are slightly, and others not at all lobed; thus

rendering it probable that S. Gaudichaudianum, A. Juss. is only a
form of this speciés.

2. STIGMAPHYLLON CILIATUM, A. Juss.

Stigmaphyllon ciliatum, A. Juss. in St. Hil. Fl. Bras. Mer. 3, p. 49, & Monogr. 1. ¢.

Has. Near Rio Janeiro. (In flower only.)

3. STIGMAPHYLLON TOMENTOSUM, A. Juss. 1. ©.

Has. Near Rio Janeiro: common on rocks along the shore.

5 BANISTERIA, Linn, A. Juss.

1. BANISTERIA CAMPESTRIS, A. Juss.

Banisteria campestris, A. Juss. in St. Hil. Fl. Bras. 3, p. 36, t. 168, & Monogr.
Malp. 2, p. 145.

Has. Brazil; in the Organ Mountains.

6. PHEIXOTOA, A. Juss.
1. PErxoToA GLABRA, A. Juss.

Peixotoa glabra, A. Juss. in St. Hil. Fl. Bras. Mer. 3, p. 60, t. 172, & Monogr.
Malp. 2, p. 174, t. 18.

Has. Brazil; near Rio Janeiro, or probably in the Organ Moun-
tains.
67

266 PHANEROGAMIA.

7% HETEROPTERYS, Kunth, A. Juss.

1. HETEROPTERYS METALLOCHROA, A. Juss.

Heteropterys metallochroa, A. Juss. Monogr. Malp. 2, p. 195.

Has. Brazil; near Rio Janeiro.

2. HETEROPTERYS SERICEA, A. Juss.

Heteropterys sericea, A. Juss. in St. Hil. Fl. Bras. Mer. 3, p. 29, & Monogr. Malp. 1.c.
Banisteria sericea, Cav. Diss. p. 429, t. 258.

Has. Brazil; near Rio Janeiro.

3. HETEROPTERYS NITIDA, Kunth.

Heteropterys (Heteropteris) nitida, Kunth, in H. B. K. Nov. Gen. & Spec. 5, p. 163;
A. Juss. |. ¢.
Banisteria nitida, Lam. Dict. 1, p. 369; Cav. Diss. t. 244.

Has. Brazil; near Rio Janeiro, and in the Organ Mountains: two
forms.
4. HETEROPTERYS CHRYSOPHYLLA, Kunth, 1. ¢.

Banisteria chrysophylla, Lam. Dict. 1, p. 368; Cav. Diss. p. 422, t. 245.
B. monoptera, Velloz. Fl. Flum. 4, p. 158.

Has. Brazil; near Rio Janeiro.

5. HETEROPTERYS VENOSA, Grised.

Heteropterys venosa, Griseb. in Linnea, 18, p. 230; A. Juss. Monogr. Malp. 2, p. 217.
H. laurifolia, Gardn. in herb. Hook.; non A. Juss.

MALPIGHIACE ZA, 267

Has. Brazil; in the Organ Mountains, and also near Rio Janeiro.

6. HETEROPTERYS MEGAPTERA, A. Juss. 1. c.?

Has. Brazil; with the preceding. In flower only.

8. HIPTAGE, Gerin.
1. Hiprace JAvAnicA, Blume?
Var. crista dorsali fructus in alam alis lateralibus minorem evoluta.
Has. Somu-somu and Naiau, Feejee Islands.

The diagnosis of the species of Hiptage is certainly difficult, as M.
de Jussieu remarks. The present specimens accord very well with
the imperfect character of H. Javanica; having oval-oblong or ovate
leaves (some of them moderately acuminate, but others very obtuse),
and flowers (apparently whitish) only half as large as those of 7.
Madablota; but the fruit, which, according to Jussieu, is nearly desti-
tute of a crest in Blume’s plant, in ours has the crest extended into a
wing fully half as large as the lateral wings, and similar in shape. It
exactly resembles that of H. Madablota, as figured by Cavanilles at
fig. f (Cav. Diss. t. 263), the proper wings, too, being narrower than
they are represented by Jussieu (Monogr. Malp. t. 16). Probably the
want of a crest or dorsal wing is no more constant in H. Javanica,
should the small flowers, &c., prove adequately to distinguish it,
than in the original species. The petals are fimbriate, and the tenth
stamen much longer and larger than the others, in our specimens;
just as in H. Madabdlota.

2. HipTaGE MyRtirotia, Sp. Nov. (Tab. 21.)

H. foliis lanceolato-ellipticis oblongisve corvaceis mitidis ; racemis pauci-
floris; petalis roseis; staminibus 5 anticis majoribus subcequilongis

268 PHANEROGAMIA.

arcuatis; alis fructus obovatis brevibus, crista dorsali quoque in alam
producta.

Var. a. foliis 1-2 poll. longis crassiusculis rigidis haud perspicue venosis.
(Tas. 21, A, B.)

Var. 2. foliis 3-4 poll. longis tenuioribus magis venosis. (Tas. 21, C.)
Has. Ovolau and Vanua-levu, Feejee Islands.

A climbing shrub; glabrous, except the youngest shoots and espe-
cially the inflorescence, which are more or less beset with fine and
close-pressed Malpighiaceous hairs; the bark of the branchlets thickly
verrucose, like the other species, with lenticels or small warts. Leaves
lanceolate-elliptical or oblong, coriaceous, more or less acute at both ends,
but scarcely acuminate, often obtuse at the base, sometimes elliptical
and very obtuse at both ends, shining, especially the upper surface,
pretty thick and rigid and only one or 2 inches long in var. a., the
veinlets inconspicuous; in var. 8. rather thinner and veiny, 3 or 4
inches long, dotted with a few glands along the margin on the under
side: the petiole 2 lines long. acemes few-(4—8-)flowered, axillary
and terminal, short-peduncled, sometimes paniculate by the abortion
of the upper rameal leaves. Calyx as in the preceding; the gland on
the posterior sepal very large and decurrent on the slender pedicel.
Petals very concave, rose-coloured or light purpel, at least on the inner
side, minutely silky-pubescent outside, orbicular, less fringed than
those of H. Madablota, but with the margins more or less erose and
ciliate, especially those of the exterior petal (which strongly overlaps
the others in the bud), 5 or 6 lines in length. Stamens 10, the fila-
ments of the 5 anterior and larger ones nearly equal in length and
thickness, incurved, at length exserted; 2 or 3 of the others equally
stout but shorter; the remainder still shorter and more slender.
Style elongated and circinnate: no abortive styles seen. Ovary
deeply three-lobed, three-celled, silky-hirsute. Fruit four-winged ; the
dorsal crest being produced into a wing of about 3 lines in width, and
fully half as large as the lateral ones, the terminal or upright wing
larger, broadly obovate, obtuse or retuse, 7 to 9 lines in length,
minutely strigose-puberulent. Seed somewhat twisted, as well as
reniform-incurved. Cotyledons very unequal; the inner one much
smaller.

MALPIGHIACESA. 269

I am unable properly to compare this with H. laurifolia, which,
from Jussieu’s description, appears to have very similar stamens.

Puate 21.—Hrperace myrtirotia. A. Branch of var. a. in flower:
B. the same, in fruit. Fig. 1. Diagram of the flower. 2. A flower-
bud. 3. A detached exterior petal. 4. Flower-bud, with the petals
and limb of the calyx removed, showing the proportion of the
stamens, &c. 5. An anther, inside view. 6. The same, seen from
the outside. 7. Flower, with the petals and stamens removed,
showing the pistil. 8. Stigma and apex of the style, at anthesis.
9. Vertical section of the flower, through the ovary. 10. Dorsal view
of a carpel, in fruit. 11. The same, with a lateral wing cut away,
exposing the seed, and showing the persistent style. 12. A seed.
13. Vertical section through the embryo. 14. Embryo detached.—All
the details, except Fig. 10, more or less magnified.—C. A branch of
the var. @. of the natural size.

9 TETRAPTERYS, Caw. A. Juss.
1. TETRAPTERYS ROTUNDIFOLIA, A. Juss.

Tetrapterys rotundifolia, A. Juss. in St. Hil. Fl. Bras. Mer. 3, p. 6, t. 161, &
Monogr. Malp. 2, p. 274, t. 18.

Has. Brazil, near Rio Janeiro. (In fruit.)

2. TETRAPTERYS PUNCTULATA, A. Juss. l. ¢.
Has. Brazil; in the Organ Mountains, near Rio Janeiro.

The specimen, in flower only, accords with the plant of Sellow
(Hirea glabra, Spreng.), of Pohl, &c. The leaves are minutely pel-
lucid-punctate, but show no black dots.

3. TETRAPTERYS MOGORIIFOLIA, A. Juss. I. ©.

Has. Brazil, near Rio Janeiro: apparently the var. platyptera.
68

270 PHANEROGAMITIA.

4, TETRAPTERYS GUILLEMINIANA, A. Juss. l. ¢.

Has. Brazil; with the preceding species.

10. HIR AA, Jacg.
1. HirnmA GAUDICHAUDIANA, A. Juss.
Has. Organ Mountains, Brazil. (In flower only.)

In the collection from Rio Janeiro there is a specimen, too incom-
plete for determination, which much resembles Hirwa rigida, A. Juss.,
but apparently does not belong to that species.

Orn. ERYTHROXYLACEZ.

1. ERYTHROXYLON, PC Browne.
1. ERyrHroxyLon PuLcuRum, St. Hil.

Erythroxylon pulchrum, St. Hil. Fl. Bras. Mer. 2, p. 94; Mart. Erythrox. (in Ab-
handl. Bayer. Akad. 3) p. 100.

Has. Brazil, near Rio Janeiro. (In fruit.)

2. ERYTHROXYLON NiTIDUM, Spreng.

Erythroaylon nitidum, Spreng. Syst. 2, p. 390; Mart. Erythrox. p. 111.
E. deciduum & E. nanum, St. Hil. Fl. Bras. Mer. 2, p- 95, 97.

Has. With the preceding. (The bractlets in a diseased condition.)

Orv. VITACE 2.

1. CISSUS, Linn.
1. Cissus QUINQUEFOLIA, Sims.

Cissus quinguefolia, Sims, Bot. Mag. t. 2448; St. Hil. Fl. Bras. Mer. 1, p. 346;
non Pursh,

Has. Brazil, near Rio Janeiro.

In foliage this plant bears a striking resemblance to our Virginia
Creeper (Ampelopsis quinquefolia) ; but the flowers are those of Cissus,
being tetramerous, and with a fleshy disk surrounding the ovary.

2. CISSUS SYLVATICA, Camb,
Cissus sylvatica, Cambess. in St. Hil. Fl. Bras. Mer. 1, p. 345.
Has. With the preceding. (A broad-leaved variety; the same as

Gardner’s No. 344.)

3. CISSUS DEFICIENS, Hook. & Arn.
Cissus deficiens, Hook. & Arn. in Bot. Misc. 8, p. 160; Gay, Fl. Chil. 1, p. 877.

Has. Chili; in ravines, near Valparaiso. (Probably not distinct
from C. striata, to which Bertero referred it.)

4, Cissus Antarctica, Vent.
Cissus Antarctica, Vent, Choix. t. 21; DC. Prodr. 1, p. 629.

Has. New South Wales; in the vicinity of Woolongong.

972 PHANEROGAMIA.

5, Cissus HYPOGLAUCA, Sp. Nov.

C. glabra; foliis digitato-quinquefoliclatis ; foliolis ovalibus oblongisve
apicem versus subserratis acuminatis petiolulatis subcoriaceis supra
lucidis subtus albo-glaucis ; cymis compositis.

Has. New South Wales, near Sydney.

A glabrous vine, with rather stout branches. Petioles and pedun-
cles 12 to 2 inches long. Leaflets 5, digitate, conspicuously petiolulate
(the partial petioles 6 to 12 lines long, except those of the two lateral
leaflets, which are 3 or 4 lines long), oval or oblong, 3 or 4 inches in
length, abruptly acuminate, sparingly serrate with a few small and
sharp teeth above the middle, rounded or obtuse at the base, rather
coriaceous in texture, reticulated-veiny, bright green and rather shining
above, glaucous-white underneath. Cymes compound, ample, rather
loosely flowered. Calyx truncate. Petals 4 or 5. Disk fleshy,
slightly 4—5-lobed, closely surrounding the ovary.

It is smgular that so conspicuous a plant should have escaped

notice in the neighbourhood of Sydney; but I do not find the species
described, nor does it occur in the Hookerian herbarium.

6. Cissus GENICULATA, Blume?

Cissus geniculata, Blume, Bijdr. p. 184?

Has. Luzon, at Baiios, near Manilla. Feejee Islands, at Sandal-
wood Bay. (From both habitats without flower or fruit, and there-
fore undeterminable.)

7. Cissus Vitrensis, Sp. Nov.

C. glabra; foliis trifoliolatis; foliolis subcarnosis ovatis seu ovalibus
2 e 2
petiolulatis acuminatis dentato-serratis ; cymis brevissime pedunculatis.

onal

VITACES. 273

Has. Feejee Islands. At Sandalwood Bay, Vanua-levu, with the
preceding, according to the ticket annexed to the specimens: but in
Dr. Pickering’s manuscript notes both are said to have been gathered
on Ovolau.

Only a wretched specimen occurs in the collection, with fruit, but
without flowers; which I am forced thus to characterize, being unable
to identify it with any described species. It is glabrous and trifoliolate,
with rather fleshy, ovate or oval, and acuminate leaflets, from 2 to 3
inches long and from one and a half to 2 inches wide, sparingly and
coarsely toothed, petiolulate. The fruiting cymes are very short-pedun-
cled: the berries are obovoid, in their dried state, and nearly half an
inch in length.

8. CissuS ACUMINATA, Sp. Nov.

C. glabella; foliis pedato-quinquefoliolatis nunc fere trifoliolatis ; foliolis
membranaceis ovatis seu ovali-oblongis bast acutis apice eximie acumi-
natis inequaliter serratis; pedunculis gracilibus laxifloris.

Has. Ovolau, Feejee Islands.

A rather slender and nearly glabrous vine, with pedately five-foliolate
leaves ; or sometimes one or both the lateral leaflets are undivided.
Petioles slender, one and a half to 3 inches long. Leaflets membra-
naceous in texture, one and a half to 2% inches long, ovate or oval-
oblong, inclining to rhomboidal, with an acute base, and a long-acumi-
nate apex, irregularly serrate, the teeth mucronate-tipped: the middle
leaflet is borne on a partial petiole of half an inch in length; the
lateral are less petiolulate. eduncle slender, 2 inches long; the cyme
small and loosely flowered. Flower-buds less than a line in length.
Fruit not seen.

There is in the collection, besides, an imperfect specimen, which I
cannot identify, of a Cissus with pedate leaves, which is ticketed as
from the Bay of Islands. But no other collectors, so far as I am

aware, have met with any plant of this family in New Zealand.
69

274 PHANEROGAMIA.

2. LEE A, Linn.
1. LEEA sAmBUCINA, Linn.

Leea Staphylea, Roxb.; Wall. Cat.; Wight, Ic. Pl. Ind. Or. t. 78.
LL. Manillensis, Walp. Rel. Meyen. p. 314.

Has. Philippine Islands: Luzon and Mindanao. Feejee Islands:
Ovolau and Muthuata; common, under several forms.

Orn RHAMNACE &.

l. VENTILAGO, Gertn.
1. VentiLAco? Viriensrs, Sp. Nov.

V. glaberrima ; jfoliis ovato-oblongis obtuse acuminatis nitidis venulis
tenuibus transversis eximie reticulatis ; floribus longiuscule pedicellatis.

Has. Feejee Islands; on the top of a mountain in Muthuata.

A very glabrous, shrubby plant, climbing by its somewhat cirrhose
young branches. Leaves ovate-oblong, varying to ovate-lanceolate,
obtusely acuminate, shining, entire, or obscurely undulate-repand,
rounded at the base, from 2 to 4 inches long, chartaceous in texture,
loosely pinnately veined; the primary veins about 6 on each side,
curved; the fine veinlets beautifully reticulated in narrow transverse
meshes. Petioles 2 or 3 lines long. Stipules minute, subulate.
Racemes axillary, elongated, simple or panicled, leafless, glabrous;

RHAMNACE &. 275

the rhachis filiform, bearing interrupted fascicles of flowers on rather
slender pedicels (about 2 lines long). Calyx flat, glabrous, a line and
a half in diameter after expansion, rather thin. Stamens not longer
than the convolute petals. Style short, flattish, emarginate, in the
oldest flowers showing some disposition to enlarge. Fruit wholly
unknown: so that the genus cannot be positively determined.

2. ZIZYPHUS, Tourn.
1. Zizyeuus Jususa, Lam.

Has. St. Jago, Cape de Verde Islands. Indigenous ?

38. CONDALIA, Caz.
1. CoNDALIA LINEATA, Sp. Nov.

©. glabra; ramis divaricatis spinosis ; foliis spathulatis vel obovatis
mucronatis crasso-coriaceis supra aventis subtus nervoso-lineatis ; drupa
oblonga.

Has. Rio Negro, North Patagonia.

A much branched, spinose shrub, entirely glabrous throughout; the
branches divaricate, alternate. Leaves alternate, or mostly fascicled on
alternate spurs, spatulate, obovate, or oblong-obovate, slightly petioled,
from 2 to 5 lines in length, mucronate, or those of young branches con-
spicuously and abruptly cuspidate-acuminate, of a thick and coriaceous
texture, entire, the upper surface veinless, the lower lineate with 3 or +
broadish and impressed nerve-like veins on each side of the midrib,
which is scarcely more prominent than the veins; the intervening
spaces narrow. Flowers not seen, except the occasional vestiges of 5
whitish lobes of the calyx, the tube of which girts the base of the
fruit, and of as many stamens alternate with these and nearly equal-
ling them in length. Petals apparently wanting. Fructiferous pedi-

276 PHANEROGAMLA.

cels solitary or geminate, 3 lines long. Drupe oblong, fully 3 lines in
length, with a thin, dry pulp, and a bony, one-celled putamen.

This is without doubt a congener of Condalia microphylla, with
which it accords in habit. But its leaves are much thicker, and con-
spicuously lineate underneath with parallel and approximate nerve-
like veins; and the fruit is of a different shape.

4, COLLETIA, Commers.
1. CoLLeTIA spinosa, Lam.

Has. Chili. Lagunas and Valparaiso: a form with short and ex-
cessively branched spines (but with the filaments exserted). Near
Santiago: a form with stout and simple spines, and the whole surface
softly puberulent.

2. COLLETIA ULICINA, Gill. & Hook.

Colletia ulicina, Gill. & Hook. Bot. Mise. 1, p. 155, t. 44; Gay, Fl. Chil. 2, p. 33.

Has. Near Valparaiso. (Destitute of flowers and fruit.) Perhaps
the specimen belongs to a slender form of C. spinosa.

0 RETANILLA, Brongn.
1, Reranitia Epuepra, Brongn.

Has. Chili; in the vicinity of Valparaiso.

2, RETANILLA TRINERVIA, Hook. & Arn.

Retanilla trinervia, Hook. & Arn. Bot. Mise. 3, p. 174.
Trevoa trinervia, Gill. & Hook. Bot. Mise. 1, p. 159; Gay, Fl. Chil. 2, p. 24.

Has. Chili, near Valparaiso. (In flower.)

RHAMNACES, 277

6. COLUBRINA, Richard.
1. Cotusrina Asratica, Brongn.

Colubrina Asiatica, Brongn. Rhamn. in Ann. Sci. Nat. 10, p. 368; Guillem. Zeph.
Tait. p. 68.

Ceanothus Asiaticus, Linn.; Cav. Ic. 5, t. 440; DC. Prodr. 2, p. 30.

C. capsularis, Forst. Prodr. Fl. Ins. Austr. p. 18.

Has. Tahiti, Matia, and Kimeo, Society Islands: abundant in deep
woods near the coast. Manua, Samoan Islands. Feejee Islands;
common. Sooloo Islands.

At Sandalwood Bay, Feejee Islands, a variety was collected with
smaller and narrower leaves; otherwise apparently not different from
the ordinary form of the species.

7. ALPHITONTA, Retssek.

Calyx planiusculus, tubo subturbinato, limbo 5-partito, lobis intus apicem
versus cristato-appendiculatis. Petala 5, unguiculata, convoluta, drsct
carnosi 10-crenulati plani calycis tubum replentis et ovario adherentis,
margint inserta. Stamina cum petalis inserta, tisdem cequilonga :
filamenta filiformia: antheree biloculares, loculis longitudinaliter dehis-
centibus basi acuminatis. Ovarium disco tectum, cum tubo calycis
connatum, biloculare, raro triloculare: styli 2, raro 3, breves. Drupa
globosa, baccata, calycis circumscisst cupula adnata basi imvoluta ;
sarcocarpio serius pl. m. exsucco et rupto; pyrenis 2 (raro 3) cocct-
formibus crustaceis intus dehiscentibus, demum subbivalvibus atque sece-
dentibus, semina scepius relinquentibus in arillo? membranaceo tn-
clusa. Semina obovata, plano-convexiuscula, testa cornea nitida.
Embryo rectus, albumine carnoso sew granuloso subequilongus ; cotyle-
donibus ovalibus foliaceis planis——Arbores Oceanicee et Australasice ;
foliis coriaceis paralelle penninerviis subtus dealbatis, stipulis minutis
caducis; floribus terminalibus axillaribusque cymosis.

AtpHitontiA, Reissek, in Endl. Gen. Pl. p. 1098, no. 5729.
70

278 PHANEROGAMIA.

A genus (the floral characters of which have not before been indi-
cated), with nearly the foliage and venation of Frangula, and the
flowers of Colubrina; distinguished by the very strongly salient crest
of the inner face of the sepals, by the anther-cells pointed at their
base, and by the membrane that encloses the seeds. I doubt if this
membrane is truly an arillus; but our materials are not sufficient for
the complete investigation of its nature. Although marked with
what seems like a rhaphe, it cannot be the testa or any proper inte-
gument of the seed, as Fenzl (in Pl. Hugel.) took it to be, for it has
no connexion with the corneous seed-coat at the chalaza, but only at
the hilum. The chalazal end appears to be slightly open, as would
be the case with an arillus. The cocci of the Australian species (A.
excelsa), the flowers of which are still unknown, often fall away after
dehiscence, just as in A, zizyphoides, so as to leave the seeds attached
to the persistent cupulate base, and enclosed in their fragile arillus?
Perhaps this membrane is a separable lining of the coccus.

1. ALPHITONIA ZIZYPHOIDES. (Tab. 22.)

A, foliis oblongo-lanceolatis sew ovato-lanceolatis acuminatis glabratis
supra nitidis sublus albidis venulisque inter venas primarias rectius-
culas eximie reticulatis.

Lhamnus zizyphoides, Soland. in Forst. Prodr. Ins. Austr. p. 90 (absq. char.) ;
Spreng. Syst. 1, p. 768; DC. Prodr. 2, p. 27.

Re. incana, Roxb.; Spreng. Syst. Veg. Cur. Post. p. 86? fide spec. Hort. Calcutt.

Pomaderris zizyphoides, Hook. & Arn. Bot. Beech. Voy. p. 61.

Has. Tongatabu. Samoan or Navigators’ Islands. Ovolau, Feejee
Islands. (Tahiti, Barclay, &c.)

“Tree 20 to 30 feet high, the trunk often a foot in diameter.”
Young branchlets, petioles, the midrib of the leaves, and the inflores-
cence clothed with a minute, more or less deciduous, ferrugineous
tomentum. Leaves oblong-lanceolate or ovate-lanceolate, acuminate,
usually with a gradually tapering point, from 3 to 6 inches long,
entire, or obsoletely repand-serrulate, thin but coriaceous in texture,
soon glabrate both sides, the upper very smooth and shining, the
lower whitish or minutely canescent, with the straightish primary veins

RHAMNACES | 979

(12 to 14 pairs) rather prominent and ferrugineous, and the inter-
mediate veinlets closely and conspicuously reticulated. Petioles an inch
long. Stipules minute, subulate, caducous. Cymes axillary and ter-
minal, much shorter than the leaves, many-flowered, the branches
somewhat unilateral. Pedicels about the length of the calyx.
Bracts minute. Calyx nearly flat when expanded, 2 to 3 lines in
diameter, with a very short and somewhat turbinate tube, clothed with
a whitish or ferrugineous tomentum externally; the limb five-parted ;
lobes triangular-ovate, three-nerved by transmitted light, valvate in
estivation, plane, except near the apex inside, which is abruptly
appendiculate with a salient fleshy crest. Disk flat, fleshy, filling the
throat of the calyx, and adherent to the surface of the ovary, which it
covers, its border obscurely 10-crenulate; its centre a little hairy at
the point of junction with the styles. Petals 5, unguiculate, convolute
and cucullate, obovate, or at length spatulate, entire, or barely retuse,
greenish or white, nearly equalling the lobes of the calyx in length,
inserted on the margin of the disk. Stamens 5, inserted with the petals
and enclosed by them, nearly equalling them in length: filaments
filiform: anthers two-celled, didymous; the cells opening longitudinally,
parailel, conspicuously pointed at their base, in a manner not observed
in any other genus of the order. Ovary two-celled, rarely three-celled,
coherent with the calyx-tube: styles short, distinct or nearly so: stigmas
simple. Ovules solitary, erect, anatropous; the rhaphe ventral.
Fruit a globose, baccate drupe, 4 or 5 lines in diameter, girt at the base
with the persistent, circumscissile tube of the calyx, which forms a kind
of cupule. Nucules 2, or sometimes 3, in a thin pulp, crustaceous,
plano-convex, dehiscent down the ventral suture after the drying of
the pulp, and partly down the dorsal suture also, thus imperfectly
two-valved, often falling away so as to leave the seeds borne on the
persistent cupulate base, enclosed in its loose, chestnut-brown, shining,
chartaceo-membranaceous, fragile arillus, if such it be: this is fissile
down the inner side, and is adherent to the seed only at the hilum.
The seed is plano-convex, with no evident rhaphe: the smooth testa
crustaceous or bony. Hmbryo straight, nearly the length of the thin
and fleshy albumen. Cotyledons oval, foliaceous, flat. Radicle short,
inferior, not perceptibly incurved.

Pirate 22, A.—ALPHITONIA ZIZYPHOIDES: summit of a flowering
branch. Fig. 1. A flower, enlarged. 2. Vertical section of the same.

28) PHANEROGAMIA.

ol

3, A stamen, seen from within, more magnified. 4. Branchlet, with
fruit, of the natural size: one of the cupules, from which the cocci
have fallen away, showing a seed, enclosed in its arillus, still attached.
5. Fruit, with one of the cocci entire, and covered with the remains
of the pulp; the other fallen, leaving the seed in its arillus, magnified.
6. One of the dehiscent cocci, equally magnified. 7. Seed, enclosed
in its arillus, more magnified. 8. Same, with the arillus split down
anteriorly. 9. Vertical section of the seed and embryo, magnified.

2. ALPHITONIA FRANGULOIDES, Sp. Nov. (Tab. 22.)

A. folits ovato-oblongis seu oblongo-lanceolatis cuspidato-acuminatis supra
nitidis subtus primum rufo-tomentulosis deinde glabratis albescentibus
iter venas primarias crebras rectas reticulatis; alabastris albo-tomen-
tosis.

Var.? @. ostusa: folits ellipticis apiculatis vel mucronulatis nunc retusis
subtus albidioribus.

Has. Sandalwood Bay, &., Vanua-levu, Feejee Islands. Var.? £.
Tongatabu : in fruit only.

The Feejee plant is said to be a shrub, for the most part. Although
nearly allied to the preceding species, it appears to differ specifically
in its smaller leaves (varying from one and a half to 3 inches in
length, either ovate-oblong or ovate-lanceolate), which taper to an acute
point, are tipped with a slender mucronate cusp, and when young
are clothed with a rusty-red tomentum underneath: when this disap-
pears a whitish but almost glabrous surface is left: while the upper
surface is very smooth and shining. The primary veins are closer
and straighter than in A. zizyphoides; the fine reticulation of the
lower surface is similar. The small and many-flowered cymes are ter-
minal, or from the uppermost axils. The /lower-buds are white-tomen-
tose, while their pedicels are ferrugineous. The flowers, although
rather smaller, accord in structure with those of A. zizyphoides, except
that the crest of the inside of the sepals is still more salient. The
drupe, seeds, &., are similar in structure, but rather smaller—The
specimens from Tongatabu (in fruit only) here doubtfully appended

RHAMNACES. 98]

a

to this species, are remarkable for their elliptical and obtuse leaves,
with their apex barely apiculate or mucronate, but sometimes even
retuse: the lower surface is not so rufous when young, and is whiter
when old: the drupes are just as in the Feejee plant. The plant
may prove to be a distinct species; but it is more likely to pass into
A. franguloides.

PuaTE 22, B.—ALPHITONIA FRANGULOIDES: a flowering branch, of
the natural size. Fig. 1. A flower-bud. 2. An expanded flower.
3. A lobe of the calyx, seen from within. 4. A petal, seen from the
outside. 5. A petal and stamen, seen from within. 6. Lateral
view of a petal and stamen. 7. Vertical section of the ovary, &c.
8. Transverse section of the ovary. The details, magnified.

8 POMADERRIS, Labill.

1. PoOMADERRIS ERICHFOLIA, Hook.
Pomaderris ericeefolia, Hook. Jour. Bot. 1, p. 257; Hook. Fl. N. Zeal. p. 46.

Has. Bay of Islands, New Zealand.

2. POMADERRIS LANIGERA, Sims.

Pomaderris lanigera, Sims, Bot. Mag. t. 1823; DC. Prodr. 2, p. 33.

Has. Near Sydney, New South Wales.

3. POMADERRIS MULTIFLORA, Sieb.

Pomaderris multiflora & P. discolor, Sieb. Pl. Exsic. N. Holl. ex Fenzl, Pl. Hugel.
p. 21.

P. malifolia, Sieb. ex Steud. Nomen. Bot. 2, p. 379.

P. discolor, DC. Prodr. 2, p. 33, excl. syn.

Has. Near Sydney, New South Wales. (Foliage only.)
71

282 | PHANEROGAMIA.

9 GOUANTA, Jacg.
1. Govania Rica, Sp. Nov.

G. ramulis puberulis glabratis ; foliis glabris subcordato-oblongis acumi-
natis subdenticulatis; racemis elongatis; coccis orbiculatis utringue
emarginatis leviter alatis extus disco pilosulis.

Has. Vanua-levu, one of the Feejee Islands.

The specimen, which is in fruit only, I cannot identify with any
described species. The branch is glabrous or glabrate below; above‘
like the short fructiferous branchlets, minutely pubescent. Petioles 3
to 9 lines long, puberulent. Leaves glabrous, except some minute
hairs on the veins of the lower surface, 14 to 24 inches in length,
chartaceous in texture, of the same hue both sides, subcordate-oblong,
acuminate, the acumination denticulate with a few minute and glan-
dular teeth, one or two of which also occasionally occur near the
base, otherwise entire; the veins pinnate and nearly simple, except
the basal pair, which are larger and branched. Stipules obsolete.
Racemes elongated, spicate, terminating short axillary branches and
panicled at the summit of the stem; the rhachis pubescent. Flowers
not seen. Fruit mostly dicarpellary, flat, very short-pedicelled : the
carpels orbicular, 3 or 4 lines in diameter, rather broader than long,
emarginate at both ends, moderately winged, the wing thickish; the
body minutely hairy externally with a ferrugineous pubescence.
There are no tendrils on the specimen.

2. GOUANIA DENTICULATA, Smith.
Gouania denticulata, Smith, in Rees, Cycl. ; DO. Prodr. 2, p. 39.

Has. Ovolau, Feejee Islands.

An imperfect flowering specimen, which accords pretty well with

RHAMNACES. 283

the character of G. denticulata; so that it may be provisionally re-
ferred to that species. It is probably the same as No. 1578 of
Cuming’s Philippine collection.

3. GOUANIA VITIFOLIA, Sp. Nov.

G. foliis membranaceis ovatis rotundisve cordatis subacuminatis creber-
rime crenato-dentatis ramulisque glabratis; stipulis ovatis parvis ;
spicis brevibus calycibusque rufo-tomentosis ; coccis orbiculatis alatis
utrinque emarginatis.

Has. Sandwich Islands: on dry hills, in the district of Waianai,
Oahu.

Branches shrubby, scandent by involute tendrils, glabrous. Branch-
lets, petioles, &c., minutely ferrugineous-pubescent, at least when
young. Leaves membranaceous, one and a half to 2 inches, or more
in length, one to 2 inches wide, ovate or roundish, cordate, with a
narrow sinus, somewhat acuminate, except the earlier ones, which are
very obtuse, closely crenate-toothed, the obtuse teeth furnished with
glandular tips, glabrous above, softly and minutely pubescent, but
soon glabrate underneath, triple-nerved from the base, and with
straight and approximate primary veins. Petioles 3 to 6 lines long.
Stipules ovate, acute, a line long. Spikes short, or at length equalling
the leaves, axillary and terminal, dense, especially the upper part,
which bears the sterile flowers; the rhachis, calyxes, and minute oval
bracts clothed with a dense and fine ferrugineous or reddish tomentum.
Lobes of the disk emarginate. Fruits very short-pedicelled, mostly
tricarpellary, when young ferrugineous-pubescent, obovoid, and en-
tirely wingless; but the mature carpels moderately winged, orbicular
in outline, slightly emarginate at both ends, glabrate, about 5 lines in

diameter.

This species considerably resembles G. tilie/folia, especially in its
inflorescence; but the stems are glabrous, the leaves (which resemble
those of a Vitis or a Tilia on a small scale) are closely beset with
even and rather coarse teeth; the stipules are shorter and broader; ~

and the mature carpels are winged.

284 PHANEROGAMIA.

4, GOUANIA? ORBICULARIS, Walp.

G. ramis erectis ecirrhosis; ramulis petiolisque sericeo-pubescentibus ;
Joliis orbiculatis integerrimis coriaceis glabratis penninerviis ; cymulis
axillaribus confertis petiolum haud superantibus; fructibus bi-tri-
alatis.

Gouanta integrifolia, Meyen, Reise, 2, p. 156, non Lam.
G. (sphalmate Gossania) orbicularis, Walp. Rel. Meyen. p. 323.

Has. Sandwich Islands: in dry and rocky situations, near Waianai,
Oahu.

An upright, shrubby plant, destitute of tendrils; the older parts
glabrate; the young branchlets, petioles, cc., silky-pubescent. Stipules
subulate, caducous. Leaves crowded towards the summit of the
branches, alternate, orbicular or nearly so, sometimes retuse, entire,
pinnately-veined even from the base, very pale both sides, nearly gla-
brous or glabrate, except a minute pubescence on the veins, coriaceous,
but rather thin, about an inch in diameter; the slender petiole almost
of the same length. Flowers moneecio-polygamous, axillary, in
short-peduncled and crowded, small cymes, not exceeding the petiole.
Bracts subulate, opposite or verticillate, silky-pubescent, as is the
whole inflorescence. Pedicels one or two lines long. Flower-buds
ovate, acuminate, minutely silky-pubescent; the edges of the thin,
triangular-ovate and acuminate, plane calyx-lobes scarcely reduplicate
in xstivation. Petals spatulate, convolute, thin and delicate, shorter
than the calyx-lobes, as long as the stamens. Anthers two-celled.
Disk as in Gouania, except that it is scarcely at all lobed or angled.
Styles in the fertile flowers 2 or 3, distinct nearly to the base.
Immature fruit narrowly two—three-winged, glabrous; the 2 or 8 cells
each containing a flat seed.

This can scarcely be a true Gouania, having such a different habit
and inflorescence (the flowers are not glomerate-spicate, as described
by Walpers, but cymulose); still it cannot be separated from the
genus until the ripe fruit is known.

RHAMNACES, 285

10. REISSEKIA, nd.
1. REISSEKIA CORDIFOLIA.

Gouania cordifolia, Raddi, Mem. Pl. Brasil. add. p. 16; DC. Prodr. 2, p. 39.
G. smilacina, Smith, in Rees, Cycl.; DC. 1. ¢.

Has. Brazil, in the vicinity of Rio Janeiro; where it is very
common.

1l. WALPERSIA, Reissek, in Endl.
1. WALPERSIA STIPULARIS, Reissek, 1. c.
Phylica stipularis, Linn. Mant. p. 208; Wendl. Coll. t. 832; DC. Prodr. 2, p. 35.

Has. Cape of Good Hope, in the vicinity of Cape Town.

2. WALPERSIA CAPITATA, Presi.
Trichocephalus spicatus, Eckl. & Zeyh. Enum. Pl. Afr. Austr. p. 130.

Has. With the foregoing.

12. PHYLICA, Linn.
1. Puyuica Bicotor, Linn.

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.

72

Orv. HIPPOCRATEACE .

1 HIPPOCRATEA, Linn.

1. HrprpocRATEA MICRANTHA, Camb, ?

Hippocratea micrantha, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 103 7

Has. Organ Mountains, near Rio Janeiro, Brazil.

2, ANTHODON, Ruiz & Pav.
1. ANTHODON UNDULATUM, Mart.

Anthodon undulatum, Mart. in Schult. Mant. Syst. 1, p. 253; DC. Prodr. 1, p. 569.
Salacia undulata, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 107.
Tontelea undulata, Endl. in Walp. Repert. 1, p. 400.

Has. Rio Janeiro, Brazil.

38 SALACIA, Linn.

1. SALACLA MACROPHYLLA, Blume.

Salacia macrophylla, Blume, Bijdr. p. 221, ex pl. Zolling. no. 524, a.

Has. Near Caldera, Mindanao, Philippine Islands.

The specimen, like that with which it has been compared in Zol-

CELASTRACEZ. 287

linger’s collection, is destitute both of flowers and fruit. But a loose
capsule which occurs in the same sheet would appear to belong to a
Celastraceous plant.

Orv. CELASTRACE A.

1 PUTTERLICKIA, Fndi.

1. PurrerLicKIA PYRACANTHA, End.

Putterlickia pyracantha, Endl. (Gen. Pl. p. 1086) in Walp. Repert. 1, p. 528.
Celastrus pyracanthus, Linn.; DC. Prodr. 2, p. 8; Eckl. & Zeyh. Enum. Pl. Afr.
Austr. p. 119.

Has. Cape of Good Hope, in the vicinity of Cape Town.

29. CATHA, Foérsk., Webb.
1. Carua Virrensis, Sp. Nov. (Tab. 23.)

C. inermis; foliis ovalibus crenulatis basi in petiolum brevem attenuatis ;
cymis brevibus multifloris; calycis lobis fimbriato-ciliatis ; petalrs
denticulatis ; capillis inter stamina nullis ; styls in fl. fert. 3 inferne
connatis; capsula subglobosa; seminibus ovalibus basi arillo carun-
culeformi glabro stipatis.

Has. Ovolau, Feejee Islands.
A shrub, apparently of considerable size, with slender and diverging

erayish branches, unarmed, glabrous throughout. Leaves oval, crenu-
late, or crenate-serrate with small teeth, obtuse or obtusely somewhat

288 PHANEROGAMIA.

acuminate, contracted at the base into a short petiole (of 3 or 4 lines in
length), chartaceous in texture, dull, veiny, from 2 to 34 inches long:
they are scattered on vigorous shoots, and more or less fascicled on
short spurs. Cymes many-flowered, axillary or sometimes terminal,
loose, much shorter than the leaves, either nearly sessile, or on a
common peduncle which does not exceed the petiole. Pedicels capil-
lary, 3 to 5 lines long. Flowers polygamous; the expanded corolla
3 lines in diameter. Calyx minute, deeply five-cleft; the lobes
greenish-white, obtuse and roundish, ciliate-fringed with bristly hairs.
Petals obovate or spatulate-oblong, very minutely denticulate, spreading,
sessile by a thickish base, white, tardily deciduous. Stamens 5; in
the sterile flowers with the subulate filaments as long as the petals,
their cordate anthers obtuse (or sometimes mucronate) : in the fertile
flowers much shorter, and with smaller and often imperfect anthers.
There are no “capilli between the stamens, in pairs before the petals,”
as is said by Forster to be the case in his Célastrus crenatus. Disk
fleshy, orbicular, almost entire, perigynous, in the early state covering
the ovary and adherent to it. Ovary in the sterile flowers effete, but
usually three-celled and ovuliferous, tipped with a short style and a
three-lobed stigma; in the fertile flowers ovoid, three-celled, and with
3 slender styles which are more or less united below the middle, their
diverging summits stigmatose on their inner face. Ovules 2 in each
cell, erect from the base, anatropous, sessile, without any trace of a
cupule or arillus. Fruit a nearly globose capsule, 3 lines in diameter,
not lobed nor triquetrous, three-celled, three-valved, three-seeded:
the valves coriaceous, bearing the dissepiment on the middle; to the
base of which the seed is attached. Seed oval or oblong, its base cupu-
late with a short and irregular, caruncle-like, fleshy arillus, which is
entirely glabrous: testa coriaceous, with a thin external pellicle.
Embryo nearly the length of the fleshy albumen: cotyledons oval,
flat: radicle short, inferior.

This species is obviously related to Catha crenata, the Celastrus
crenatus of Forster, from the Marquesas (and Society?) Islands, with
the description of which, reproduced by Guillemin (Zeph. Tait. p. 69)
from Forster’s manuscript, I have endeavoured to contrast it. Our
plant has smaller and less coriaceous leaves, smaller pods, and a red-
dish and glabrous (not woolly and white) cup-shaped arillus to the
oblong seed; the flowers are very much smaller than those of Prunus

CELASTRACE S&. 289

Padus, and are destitute, as already remarked, of the ten capillary
bodies mentioned by Forster as interposed in pairs between the
stamens.—That an erroneous character has been ascribed to the
arillus of Catha by Endlicher has already been noticed by Webb, who,
in his elaborate Phytographia Canariensis, has well illustrated Catha
cassinoutes,

Prate 23.—Catua Viriensis: in flower and fruit. Fig. 1. A
sterile? flower. 2. Vertical section of the same. 3. A petal. 4. A
stamen. 9. Vertical section of a fertile flower. 6. Transverse section
of the ovary. 7. A dehiscent capsule. 8. A seed, with its arillus.
9. Transverse section of the seed. 10. Vertical section of the seed
and its short arillus—AIl the details magnified.

da CELASTRUS, Linn.
1. CeLastrus Ricuu, Sp. Nov.

C. imermis, glaberrima; foliis oblongis utrinque subacutis crenulatis
supra lucidis brevissime petiolatis; racemo terminali paucifloro.

Has. Vanua-levu, Feejee Islands.

The specimens are in fruit only: but the terminal raceme and the
complete arillus, as well as the habit, show the plant to be a genuine
Celastrus. It bears much resemblance to the East Indian C. panicu-
latus; but the few-flowered racemes, the branchlets, &c., are perfectly
glabrous, and the leaves are oblong or elliptical, more or less acute at
both ends, or only slightly acuminate, shining above, rather indistinctly
crenate, and on very short petioles (one or two lines in length). The
stems would appear to be sarmentose; the slender branchlets are
thickly warty-dotted; the leaves of a chartaceous texture, 2 or 3
inches long. Capsule globose, obscurely three-lobed, 5 lines in dia-
meter, three-celled, three-valved, stipate at the base by the persistent
calyx (the lobes of which apparently are not ciliate); the cells two-
seeded. Arillus thin and fleshy in the dried plant, enclosing the seed,

apparently orange-coloured; the seed purple.
73

290 PHANEROGAMIA.

4, MAYTENUS, Fewill.

1. MAytENus ostusiFouius, Mart.

Maytenus obtustfolius, Mart. Herb. Bras. & in Flora, 24 (1842), p. 88.

Has. Brazil, near Rio Janeiro. (In fruit.)

2. Maytenus Brasiuiensis, Mart. 1. ec.

Has. Brazil; with the preceding.

3. Maytenus Maceuuanicus, Hook. f.

Maytenus Magellanicus, Hook. f. Fl. Antare. p. 254.
Cassine Magellanica, Lam. Ill. n. 2590, & Dict. Suppl. 2, p. 130.
Celastrus Magellanicus, DO. Prodr. 2, p. 8; Hook. Ic. Pl. t. 537.

Has. Orange Harbour, Tierra del Fuego; in woods, rather rare.

4, Maytenus Cuitensis, DC.

Maytenus Chilensis, DC. Prodr. 2, p.9; Gay, Fl. Chil. 2, p. 7.
Maiten, Feuill, Obs. 8, p. 39, t. 27.

Has. Chili; abundant near Valparaiso.

ind

0 PERROTTETIA, HBX.

Char. emend. et auctus. Flores diwcio-polygami. Calyx quinquefidus
vel quinquepartitus, persistens ; lobis subovatis cestivatione imbricatis,
Petala 5, sub margine disci perigyni integerrimi orbicularis inserta,

CELASTRACES, 291

ovata, acuta, cestivatione valvata? seu (in P. Sandwicensi) imbricata,
persistentia. Stamina 5, cum petalis inserta, tisdem alterna et bre-
viora, vel in fl. masc. duplo longiora: anther: didyma, incumbentes,
biloculares, mutice, loculis longitudinaliter dehiscentibus. Ovarium
liberum (fl. mase. sterile), biloculare, loculis biovulatis: stylus brevis
vel nullus: stigma bifidum. Ovula collateralia, e basi loculi erecta,
anatropa (rhaphe in P. Sandwicensi fere dorsali!). Bacca parva,
depresso-globosa, bilocularis, 2—4-sperma. Semina exarillata, obo-
voidea, testa crustacea multicostata. (Embryo ignotus.)\—Frutices
mermes, glabelli ; foliis alternis oblongis gland uloso-serratis vel denticu-
latis; stipulis deciduis; paniculis axillaribus thyrsoideis parvifloris.

PrerrorreTiA, H. B. K. Nov. Gen. & Spee. 7, p. 78, t. 622; Endl. Gen. p. 1089.

1. PERROTTETIA SANDWICENSIS, Sp. Noy. (Tab. 24.)

P. foliis ovato-oblongis glanduloso-serratis ; stipulis minutis caducis ;
paniculis compositis ; floribus pedicellatis viridulis ; petalis triangulari-
ovatis lanuloso-ciliatis calycem vix superantibus ; testa seminum costis
transversis exsculpta.

Has. Sandwich Islands: on mountains behind Honolulu, Oahu
(where it was also gathered by Gaudichaud): also Hawaii; along the
margin of forests.

A shrub, or small tree, nearly glabrous ; the younger branchlets and
the lower surface of the nascent leaves more or less pubescent. Leaves
alternate, ovate-oblong, abruptly somewhat acuminate, either obtuse or
acute at the base, serrate with small and glandular-tipped teeth, rather
chartaceous in texture, deciduous, pinnately-veined, somewhat shining
above, paler beneath, 23 to 32 inches long; the petiole from half an
inch to an inch in length. Stipules minute, caducous. Flowers small
(a line or a line and a half in diameter, when expanded), polygamo-
dicecious, greenish, pedicellate, very numerous, in compound thyrsoid
panicles from the axils of the leaves, especially of those near the base
of the branches of the season: the panicles are mostly shorter than
the leaves; the puberulent common peduncle usually shorter than the
petiole; its branches divaricate : pedicels crowded or slightly fascicled,

999 PHANEROGAMIA.,

a line or a line and a half long, subtended by a minute bract, articu-
lated in the middle. Calyx five-parted, persistent, externally slightly
puberulent; the lobes ovate-lanceolate, thickish, rather acute, imbri-
cated in estivation. Petals 5, triangular-ovate, acute, greenish-white,
sessile by a broad base, inserted under the entire margin of the orbicular,
flat, or somewhat concave, perigynous disk, which fills the bottom of
the calyx and surrounds the base of the ovary; they are decidedly
imbricuted in cestivation, scarcely exceeding the calyx, plane, obscurely
one-nerved, their thin margins ciliate, persistent. Stamens 5, inserted
on the margin of the disk, alternate with the petals: filaments in the
male flowers filiform and twice the length of the petals; in the female
flowers very short (about half the length of the petals), persistent :
anthers didymous, incumbent, fixed near the middle, destitute of a
manifest connective, tiwo-celled, pointless ; the ovoid cells opening longi-
tudinally ; those of the fertile flowers smaller and scarcely pollini-
ferous. Ovary free, ovoid, in the male flowers abortive and sterile; in
the fertile flowers two-celled ; the style very short and thick, terminated
by a spreading two-lobed stigma. Ovules 2 in each cell, collateral,
obovoid-globose, anutropous, erect from the base of the cell, where they
are inserted on a placental projection from the axis, so that their oppo-
site and very thick rhaphes become nearly dorsal, as respects the axis
of the ovary. The fruit, not perfectly mature, appears to be a
rather dry berry, depressed-globose, slightly 2—4-lobed, according to
the number of seeds perfected; in the specimens scarcely exceeding a
line in diameter, stipate at the base by the persistent calyx, corolla,
&e., two-celled; the cells two-seeded, or by abortion one-seeded. Seeds
obovoid, oblique, large for the size of the fruit, with a short and
caruncle-like rhaphe, which is exterior or dorsal: the thick and erus-
taceous testa sculptured with numerous transverse ribs and grooves.
Arillus none. The testa, although plump and well-formed, is empty
in all the numerous seeds examined: so that the characters of the
albumen, if there be any, and of the embryo, are unknown.

The general characters of this plant so nearly accord with those of
Perrottetia Quinduensis that I refer it, without much hesitation, to
that little-known genus. Since it is obvious from Kunth’s figures,
that what he doubtfully terms pyrene or ossicula ave seeds, there
remains no important generic difference between that plant and our
own, except the estivation of the petals; which in the latter are

CELASTRACES. 293

quincuncially imbricated, while, according to Kunth, they are valvate
in P. Quinduensis. This would seem to be the case in two flowers
which have been obligingly furnished me by M. Tulasne, taken from
an original specimen in the Paris Museum; but they are much too
old to decide this point. Even if this be so, it would not be proper
to establish a new genus for the Hawaiian plant, unless the flowers of
the Peruvian species should prove to be truly hermaphrodite (which
I doubt), and the rhaphe normally ventral. As species, they are
abundantly distinguished by the subsessile flowers, the conspicuous sti-
pules, and the larger petals of P. Quinduensis, which are not at all
ciliate, and by the longitudinal sculpturing of its bony seeds.* These,
in the specimens of Humboldt and Bonpland (the only ones known),
are hollow and destitute of embryo, just as are those of our Hawaiian
plant; probably from not having been fertilized by the pollen of male
flowers. The internal structure of the seed is therefore a desidera-
tum; and until this is known, the affinities of the genus cannot be
positively determined. Most probably, however, it belongs to the
Celastracece, notwithstanding the total absence of any arillus.

According to a note on the ticket of Gaudichaud’s specimen, this
.shrub is called Oraye or Olaye by the Hawaiians.

Piate 24.—PERROTTETIA SANDWICENSIS. Fig. 1. A branch of the
fertile plant, in flower and fruit, of the natural size. 2. Flowering
branch of the male plant. 3. A small portion of the inflorescence of
the fertile plant, enlarged. 4. Diagram of the flower. 5. A sterile
flower. 6. A fertile flower, expanded. 7. A pistil, from the latter.
8. Vertical section of a flower through the pistil and disk. 9. Trans-
verse section of the base of the ovary, showing the position of the
ovules. 10. An ovule. 11. A fruit. 12. Vertical section of the
same. 13. A seed.—All the details magnified.

* M. Tulasne informs me that, if he is not mistaken, the ovary of P. Quinduensis is
sometimes four-celled, with a single ovule in each cell.

74

Orv. AQUIFOLIACE SZ.

1. CASSINE, Linn.

1. Casstne Mavrocenta, Linn.

Cassine Maurocenia, Linn. Spec. Pl. p. 885 (Dill. Hort. Elth. t. 121); Hook. Ic.
Pl. t. 552.

Has. Cape of Good Hope, near Cape Town. (Without flowers or
fruit.)

2. ILEX, Linn.

1. nex CAnARIENSIS, Poir.

Tex Canariensis, Poir. Dict. Suppl. 3, p. 67; DC. Prodr. 2, p. 14; Webb. Phytogr.
Canar. 2, p. 137, t. 69.

Has. Madeira.

2. Inex Paraacuayrensis, Lamb.

Ilex Paraguayensis, Immb, Pin. t. 2; Spreng. Syst. Cur. Post. p. 48; Hook. Lond.
Jour. Bot. 1, p. 35, t. 1, 8.

I, Paraguariensis, St. Hil. in Mem. Mus. 8, p. 351, & Voy. Brés. 1, p. 273; DC.
Prodr. 2, p. 15.

1. Maté, St. Hil. Hist. Pl. Remar. Brés. & Parag. 1, p. 41.

Has. Organ Mountains, Brazil; where it was also collected by the
late Dr. Gardner.

AQUIFOLIACES. 295

The specimens (with young fruit) accord with the var. @. of Sir
William Hooker's article, above-cited, to which the reader is referred
for a full account of this plant, the celebrated Paraguay Tea. It is
used in Paraguay, &c., in much the same manner as was the Yapon
(a related species of J/ex) by the aborigines of North Carolina.

3. Inex Vitiensts, Sp. Nov. (Tab. 25.)

I. foliis ovalibus subacuminatis integerrimis tenuiter coriaceis; floribus
abortu divicis in cymulas breviter pedunculatas dispositis, masculis
46-meris, feemineis scepissime 8-meris; pyrenis 8 dorso profunde
sulcatis,

Has. Sandalwood Bay, Vanua-levu, Feejee Islands.

A shrub with ash-gray bark, glabrous throughout, except a very
slight pubescence on the pedicels and calyx. Leaves oval, one and a
half to 2 inches long, abruptly and obtusely somewhat acuminate,
entire, thin, but coriaceous in texture, mostly rounded at the base, dull,
rather obscurely veined; the petioles 3 or 4 lines long. Peduncles
axillary, 3 or 4 lines in length, bearing a small cyme, very much
shorter than the leaves; in the sterile plant more or less compound
and with numerous flowers; in the fertile simple and mostly few-
flowered. Flowers dicecious by abortion, barely a line in length; the
male flowers 4—6-merous, more commonly hexamerous, with a rather
deeply cleft calyx, and an abortive pistil; the petals united only at
the base, longer than the stamens. Corolla and stamens of the female
flowers not seen; the persistent calyx obscurely eight-toothed, at
least in most cases. Drupe (immature) a line long, obtusely apicu-
late; the 8 nucules deeply excavated on the back.

This plant, which is probably related to Blume’s [lea cymosa, would
be a Prinos, except that the nucules of the drupe are deeply grooved
on the back. The number of parts in the flower is far too variable
for a generic character, when not accompanied by other distinctions.
I believe, therefore, that the genus Prinos must be merged in J/ex.

Puate 25, 4.—Ivex Vittensis; with immature fruit. Fig. 1. Bud
of a sterile flower. 2. Corolla of the same, laid open, and stamens.

296 PHANEROGAMIA.

3. Anterior view of a stamen. 4. The calyx and abortive pistil. 5. ;
Immature drupe. 6. Same, with the upper part of the sarcocarp re-
moved. 7. Transverse section of a drupe.—The details magnified.

3. BYRONIA, Endl.

Flores polygami. Calyx 3-4-lobus, lobis seepius wregulariter crenulatis.
Corolla 5-9-partita. Stamina ime corolle inserta, ejusdem lobis
numero equalia rariusve dupla; antheris fl. fert. plus minusve effoetis.
Ovarium 12-18-loculare, fl. ster. imperfectum, stigmate sessili late
discoideo radiato centro umbilicato coronatum. Ovula in loculis soli-
taria, ex apice pendula; rhaphe dorsali. Drupa baccata, 12-18-
pyrena; pyrens fibroso-cartilagineis. Embryo in apice albuminis
carnost minimus.—Arbuscule Sandwicenses et Tuitenses, foliis coria-
ceis lucidis, floribus axillaribus cymosis.

A genus, now strengthened by a second species, differing from Ilex
(including Prinos) in the petals being more numerous than the lobes
of the calyx, and the cells of the ovary of double their number. The
number of the stamens, also, is sometimes doubled in the sterile flowers.

1. Byronta Sanpwicensis, Endl. (Tab. 26.)
B. foliis ovalibus seu ellipticis ; ovarii loculis pyrenisque fructus 12-18.

Byronia Sandwicensis, Endl. in Ann. Wien. Mus. 1, p. 184, & Gen. p. 1098.
Ilex? anomala, Hook. & Arn. Bot. Beech. Voy. p. 111, t. 25.

Has. Sandwich Islands. Kaala Mountains, Oahu. Forests on
Mouna Kea, and near the crater of Lua Pele, &c., Hawaii. (Also,
gathered by Macre, Lay & Collie, &c.)

A small tree, or sometimes a shrub, glabrous throughout. Leaves
coriaceous, shining above, alternate, oval or elliptical, mostly obtuse or
rounded at both ends, 23 to 34 inches long, sometimes 2 inches Wide,
reticulate-veiny, entire, or occasionally serrulate on vigorous shoots;
the petioles from half an inch to an inch long. Flowers polygamous,

AQUIFOLIACE &. 297

numerous, in trichotomous awillary cymes: the peduncle an inch or
more in length, apparently two-edged; its branches divaricate, sub-
tended by a pair of small oblong bracts: pedicels 2 or 3 lines long.
Calyx small, persistent, usually four-lobed, sometimes three-lobed
according to Hooker and Arnott; the lobes broad and rounded, imbri-
cated in zstivation, crenulately 3-4-toothed in the fertile flowers, entire
or nearly so in the sterile, of a thick and firm texture. Corolla hypo-
gynous, 6-9-parted, more commonly eight-parted, rotate-spreading,
deciduous; the lobes rounded, imbricated in estivation, one and a
half to 2 lines long. Stamens as many as the lobes of the corolla and
alternate with them, or rarely one or two more, inserted on its short tube:
filaments subulate, flattened, about half the length of the lobes of the
corolla: anthers didymous, two-celled, introrse; the cells opening
longitudinally, in many of the fertile flowers smaller and effete,
Hypogynous disk none. Ovary closely sessile in the calyx, globular,
12-18-celled ; in the sterile flowers smaller and imperfect. Stigma
closely sessile, broad, discoid, radiate with 12 to 18 lines, depressed
in the centre, persistent. Ovules solitary in each cell, suspended from
tts apex, anatropous; the rhaphe dorsal. Drupe baccate, spherical,
12-18-grooved when immature or dry, 3 or 4 lines in diameter, black,
. containing from 12 to 18 separable pyrene or nucules, which are of a
fibrous-cartilaginous texture, smooth, and moderately grooved on the
back. Seed filling the cell, with a thin testa. Embryo minute at the
apex of the fleshy albumen: radicle superior.

PLaTE 26.—Byronta SANDWICENSIS: a plant in flower and fruit.
Fig. 1. A sterile tlower, with 6 petals and stamens. 2. Corolla of 8
petals and stamens, displayed. 38. Calyx and ovary of the same. 4.
Interior, and 5, an exterior view of a stamen. 6. A fertile flower. 7.
Vertical section of the same. 8. An ovule detached. 9. Transverse
section of the ovary. 10. Drupe, with the persistent calyx. 11.
Transverse section of the same. 12. Vertical section of a drupe.

2. Byronia Tairensis, Sp. Nov. (Tab. 25.)

B. foliis spathulato-oblongis seu lanceolato-oblongis ; staminibus fl. masc.
quandoque petalis numero duplis; ovarii loculis pyrenisque 10-12.

Has. Mountains of Tahiti, Society Islands.
75

298 PHANEROGAMIA.

Apparently an arborescent species, nearly related to the preceding;
from which it is distinguished by its less reticulated and narrower,
spatulate-oblong, or lanceolate-oblong leaves (from 2% to 4 inches in
length, and an inch or rarely more in breadth), the base usually
tapering into the short petiole; and by the fewer (only 10 or 12) cells
to the ovary and fruit. The corolla is occasionally five-parted, but
more frequently six-parted, the calyx in both cases being only four-
lobed. In all the fertile flowers examined the (sterile?) stamens are
uniformly of the same number as the divisions of the corolla; but in
some sterile flowers they are more numerous, usually twice as many !

The inflorescence is commonly more developed, and with more
numerous flowers than in the specimen delineated in the plate, at
least in the sterile plant; in one specimen of which the peduncles
are two inches long, and the trichotomous cyme equals the leaves.

PLATE 25, B—Byronta TAairensis: a branch with sterile flower-
buds, and another in fruit. Fig. 1. Unexpanded sterile flower. 2.
Sterile flower with the hexamerous corolla and stamens displayed :
the calyx and pistil underneath. 8. A seven-cleft corolla of a sterile
flower, with 12 stamens displayed. 4. A drupe. 5. The same,
transversely divided. 6. Vertical section of a drupe, showing a section
of one seed, and its embryo.

4. VILLARESIA, Ruiz & Pav.
1. VILLARESIA MUCRONATA, Ruiz & Pav.

Villaresia mucronata, Ruiz & Pav. Fl. Per. & Chil. 3, p. 9, t. 231; A. Juss. in Ann.
Sci. Nat. 25, p. 14, t. 5, £. 2; Gay, Fl. Chil. 2, p. 18.
Citronella mucronata, Don. in Edinb. New Phil. Jour. 13, p. 248.

Has. Chili, near Valparaiso. (Foliage only.)

Orv. OLACACE &.

Suporp. l ICACINE &.
1 PLEUROPETALON, Blume.

Char. paullo auctus, nempe: Flores hermaphroditi, polygami, vel abortu
monoici. Calyx aut parvus aut mediocris. Ceet. ut in opere infra cit.

PLEUROPETALON, Blume, Mus. Bot. Lugd.-Bat. p. 248, non Hook. f.

1. PrevropeTALon SAMmoENSE, Sp. Nov. (Tab. 27.)

P. foliis late ovatis; thyrso virgato racemiformi puberulo ; floribus her-
maphroditis ad apicem pedunculorum partialium breviwm glomeratis ;
sepalis rotundatis ciliates.

Haz. Samoan or Navigators’ Islands: the particular habitat not
recorded.

Shrub glabrous, except the inflorescence, which is minutely pubescent,
at least when young. Leaves broadly ovate, 4 or 5 inches long, and 3
inches or more in width, on petioles of half an inch in length, alter
nate, often oblique, slightly acute or pointed, entire, rather coriaceous
in texture, loosely veined and reticulated, the lower surface dull, the
upper rather shining. Inflorescence terminal, pedunculate; the flowers
apparently all hermaphrodite, clustered on the apex of very short pedun-
cles (3 lines long), which are crowded on the elongated rhachis; form-
ing a dense, virgate, raceme-like thyrsus of 6 inches or more in length.
Bracts subulate and minute. Bractlets ovate, scarious, minute, ciliate.
Flower-buds nearly sessile, cylindraceous, 23 lines long. Calyx one-

300 PHANEROGAMIA.

third of the length of the corolla, of 5 distinct, nearly orbicular-ovate
and scariously margined, imbricated sepals, their edges hirsutely
ciliate. Corolla hypogynous, deciduous perhaps before expansion,
glabrous; the 5 linear-oblong and somewhat spatulate petals conni-
vent, slightly united towards the base by means of the stamens, their
margins thin and lightly imbricated in estivation, except near the
summit, where the estivation is valvate-induplicate, the margins
being strongly inflexed; their inner face bears an elevated and fleshy
central crest or ridge, which is abruptly thinner and three-forked
above the middle, the lateral divisions divergent and evanescent near
the margins, the middle one prolonged to the apex. Stamens 4,
hypogynous, almost as long as the petals, alternate with them, and
occupying the space between their salient ridges: filaments elongated,
subulate, thickish and fleshy, glabrous, below lightly coherent with
the petals: anthers cordate-oblong, two-celled, very obtuse or retuse ;
the cells acutish at the base and thence distinct up to the middle,
where the anther is introrsely attached to the acute apex of the fila-
ment, longitudinally dehiscent. Pollen spherical, simple. There are
no sterile stamens. Hypogynous disk obsolete, or none. Ovary free,
closely sessile in the persistent calyx, oblong-ovoid, one-celled, nearly
symmetrical, marked with a slight salient ridge on one (the anterior)
side, tapering into the style, which is slender, subulate-filiform,
straight or nearly so, and central, longer than the ovary, lightly
grooved for its whole length on one (the posterior) side, and termi-
nated by a subcapitate emarginate stigma. Within, the ovary is desti-
tute of incomplete partitions or of any abortive cells; but there is a
slight and obtuse longitudinal ridge occupying the whole length of the
posterior or placental side of the cell. Ovules 2, collateral, suspended
from the summit of the cell on the posterior side, anatropous, appa-
rently furnished with a single proper coat, which however is adherent
to the nucleus. Fruit unknown.

Although all the flowers examined are hermaphrodite, yet, as con-
siderable diversity in the length of the style is noticed, the plant may
prove to be polygamous. In that case it would accord in all essential
respects with the characters of Blume’s genus Pleuropetalon; a genus
manifestly allied to Jcacina and Rhaphiostylis (Planch. in Hook. Niger
Flora). The ovary is not gibbous, as in the latter genus, and the
slender groove of the style in our plant extends for its whole length.

OLACACE &. 201

The exstivation of the corolla at once distinguishes it from all other
Icacineous genera, except the Bursinopetalum of Wight, which also
has an imbricative sestivation, according to Mr. Miers. On which
account, indeed, this acute author, in his profound and elaborate
memoirs on the Olacinew, Icacinee, &c.,* refers both Bursinopetalum
and Blume’s Pleuropetalon (which he supposes to be one and the same
genus) to the order Aquifoliacee, next to the anomalous genus Villa-
resia. But, from the examination of one or two young flower-buds of
Bursinopetalum arboreum, I should say, with Dr. Wight, that the
estivation of the thick petals was of the valvate kind, as truly so as
in Clematis. Their thick and abrupt edges appear to be apposite
throughout, next the base without any modification, while higher up
a thin margin is induplicate, and at the apex these thin margins,
more strongly induplicate, form the deeply inflexed apical appendage.
In Pleuropetalon, at least in our species (for I have not seen that of
Blume), the imbricative and the induplicate-valvular modes of sesti-
vation are truly combined in the same organ; the thin edges of the
petals overlapping each other, in the normal quincuncial manner,
except near the apex, where the inflexion begins: here their margins
abruptly become induplicate, as is represented in the accompanying
figures, bringing this portion of the estivation into the valvular cate-
gory. ‘The estivation of the corolla seldom affords an ordinal cha-
racter free from all exceptions; and the modification in these two
genera would seem wholly insufficient to exclude them from the Ola-
caceee, or Icacinucee, unless supported by other characters; whereas,
in fact, they appear fully to accord in all other respects with the
characters assigned by Mr. Miers to his lcacinacee. Pleuropetalon at
least, with its simple pistil, one-celled and biovulate ovary, and
slender style, would surely seem much out of place in Aquifoliacee.
As to the generic difference between Bursinopetalum and Pleuro-
petalon, now confirmed by a second species, apparently there can be
little question. Bursinopetalum has a merely five-lobed calyx, with
its tube adnate to the ovary, which contains a single ovule, and is
surmounted by a very short and conical style. Plewropetalon has a
calyx of five distinct sepals, entirely free from the biovulate ovary,
and the style is long and filiform. Its fruit unfortunately is unknown.

* In the Annals and Magazine of Natural History, New Series, vols. 8, 9, 10;

especially vol. 8, p. 169, and 9, p. 223.
76

302 PHANEROGAMIA.

Whether the Jcacinee should form a distinct order, with affinities
in some respects different from those of the Olucacee, as is ably argued
by Mr. Miers, or whether they may still be arranged as a suborder
of Olacacece, is a question which I am not competent to decide. It is
merely more convenient on the present occasion to adhere to the
older view.

The Pleuropetalum of Dr. Hooker is several years older than the
Pleuropetalon of Blume; but, since the name of the former genus has
already been changed by Moquin-Tandon, on account of its inap-
propriateness (the plant being apetalous), the latter should be con-
tinued.

PLATE 27.—PLEUROPETALON SAMOENSE: in flower. Fig. 1. A branch
of the inflorescence, or peduncle, with unexpanded flowers. 2. Trans-
verse section of the flower-bud through the calyx: 3, of the same,
above the calyx, showing the slightly imbricated xstivation of the
corolla: 4, of the corolla near the apex, the wstivation there indupli-
cate. 9. A petal, seén from within. 6. A stamen, seen from within.
7. The same, seen from without. 8. Vertical section of a flower-bud,
through the ovary, showing the two suspended ovules, &c. 9. The
pistil, seen from the side which is directed towards the axis of the
inflorescence: the calyx torn away from the torus. 10. The same,
transversely divided. The details magnified.

2 RHYTIDANDRA, Nov. Gen.

Flores hermaphroditi. Calyx parvulus; tubo cum ovario connato ;
limbo cupulari truncato, margine 6—7-denticulato. Corolle epigyne
petala 6-7, lineariu, conniventia, cestivatione valvata. Stamina 6-7,
petals alterna, cum visdem inserta, libera: filamenta brevissima, intus
barbata: anthere lineares, introrsum adnate, dithecee, quadrilocellate,
locellis transversim annulato-rugosis vel cameratis. Discus epigynus
scutelliformis. Ovarium inferum, uniloculare, uniovulatum 3 ovulo

ex apice loculi parvi suspenso. Stylus elongatus, sulcatus, bifidus,
lobis scepius apice bi-tridentatis; stigmatibus terminalibus parvis.
Fructus ignotus—Frutex sarmentosus, foliis ovatis obl iquis, pedunculis
axillaribus cymulam paucifloram gerentibus.

OLACACES, 303

1. Ruyrmpanpra VitrEnsis, Sp. Nov. (Tab. 28.)
Has. Feejee Islands. (The particular habitat is not recorded.)

Only a single specimen, with unexpanded flowers, occurs in the
collection. The plant appears to be a shrub, with slender and sar-
mentose branches, glabrous, or the young parts more or less cinereous-
puberulent. Leaves alternate, petioled, ovate, oblique, unequal at the
base, acuminate, entire, or obscurely repand, 3 or 4 inches long and
one and a half to 2 inches wide, membranaceous, sparingly feather-
veined ; the veins connected by transverse veinlets. Stipules none.
FPeduncles axillary, longer than the petioles (half an inch or more in
length), bearing a few-(7-12-)flowered small cyme. Pedicels short,
minutely unibracteolate at the apex, where they are articulated with
the flower. Unexpanded flowers 3 lines long, minutely silky-canes-
cent, hermaphrodite. Calyx small, about a line or a line and a half
long, including its turbinate tube, which coheres with the ovary through-
out; the free summit or limb cup-shaped, half a line long, trwneate, the
border obscurely 6—T-toothed. Petals 6 or 7, inserted at the junction of
the free part of the calyx with the summit of the ovary, without the
intervention of any manifest perigynous disk, linear, valvate in cesti-
vation, apparently distinct in anthesis, and deciduous, of a thickish
texture, glabrous, and destitute of any appendage inside. Stamens as
many as the petals and alternate with them, inserted with them, and
equalling them in length, not cohering with their bases. Filaments
extremely short, bearded inside, continued into a smooth and unappen-
daged connective; to which the four cells of the long and linear anther
are imtrorsely adnate. These slender cells, or Jocelli, are transversely
constricted at short and irregular intervals (nearly as in most species
of Gomphia), so as to appear more or less necklace-shaped or cham-
bered: their dehiscence is not obvious. Style elongated (about the
length of the petals), central, glabrous, somewhat angled or grooved,
two-cleft above the middle; the divisions filiform and somewhat flat-
tened, 2—3-toothed or cleft at the apex; the stigmas terminal, small,
and simple. Ovary turbinate, one-celled, with no vestiges of other
cells or of incomplete partitions, not eccentric, fleshy, wholly invested
by the adnate calyx-tube, except the truncate summit, which is covered

304 PHANEROGAMIA.

by a thin and flat disk, perforated in the centre by the base of the
style; the central and very small cell nearly filled by the solitary
and anatropous ovule, which is suspended from its apex, without the
intervention of any perceptible placental column, whether free or
adnate. The ovule appears to be invested by a single coat; but this
18 uncertain.

Although this plant appears to belong to the Olacacee (taking that
order in the most extended sense), I know of no genus with which it
may be particularly compared. If rightly referred to this order, it
must be viewed as a genus whose affinity tends toward Styracacee
rather than Santalacee. The adherence of the calyx to the ovary,
which to all appearance is perfect and immediate, is at variance with
Mr. Miers’ diagnoses both of Olacaceew and Icacinacew. In the points
in which these two groups differ, our plant accords with the latter,
except that there is only a solitary ovule, suspended from the very
apex of the cell. The elongated, quadrilocellate anthers, with their
cells annular-constricted at short intervals, giving them a wrinkled
appearance, are so different from those of any related plant known to
me, that I have derived from the latter character the generic name,
viz.: from piris, pirivos, a wrinkle or fold, and ap, kdpa, botanically used
for stamei.

PLATE 28.—RuytipaAnpRaA Vitrensis. Fig. 1. A flower-bud, with
the short pedicel and a portion of the axis. 2. The corolla and
stamens, displayed. 38. Vertical section of a flower-bud, showing the
solitary ovule, &. 4. A stamen, inside view. 5. The same, seen
from the outside. 6. A transverse section of the same, inside view,
more magnified. 7. Pistil, with the limb of the calyx removed,
showing the flat epigynous disk. 8. Transverse section of the ovary.
—All the details magnified.

Susorp. II OLACINE &.
3. XIMENIA, Plumier.
1. XimentA Americana, Linn.

Has. Brazil, near Rio Janeiro. (Fruit oval, or oblong.)

OLACACE®. 305

2. XIMENIA ELLIPTICA, Forst.

Aimenia elliptica, Forst. Prodr. Fl. Ins. Austr. p. 27; Labill. Sert. Austro-Cal. p.
34, t. 87; DC. Prodr. 1, p. 533.

Has. Samoan or Navigators’ and Feejee Islands: very common on
the coast.

The fruit, “of the size of a green-gage plum, and sweet-scented,”
is spherical, as it is figured by Labillardiere. Otherwise the plant
too closely resembles Ximenia Americana, to which, indeed, Decaisne

(in Herb. Timor.) has joined it. The branches are either unarmed
or spiny.

4, HEISTERIA, Linn.
1. Hersterta Rapprana, Benth.
Heisteria Raddiana, Benth. in Hook. Niger Flora, p. 258, in not.

Has. Brazil, in the vicinity of Rio Janeiro. (In fruit only.)

5 OLAX, Linn.
1. OLAX IMBRICATA, Roxb.
Olax imbricata, Roxb. Fl. Ind. 1, p. 169; DC. Prodr. 1, p. 532.

Has. Philippine Islands: on the shores of Laguna, Bafios, near
Manilla, Luzon.
77

Orv CORIARIACES.

1 CORIARIA, Miss. Linn.
1. CortariA RuScrIFoLIA, Linn.
Has. Bay of Islands, New Zealand: common. (In fruit.)

The New Zealand plant, Coriaria sarmentosa of Forster, Dr. Hooker
has joined, apparently with good reason, to the Linnean C. ruscifolia,
of Chili and Peru. The baccate fructiferous perianth yields a pala-
table purple juice, which is much liked by the natives, and from which
a kind of wine may be made: but the seeds are poisonous, as they are
likewise in C. myrtifolia, of the south of urope.

Orv. STACKHOUSIACES®.

1 STACKHOUSIA, Smith.
1. STACKHOUSIA SPATHULATA, Sieber.

Stackhousia spathulata, Sieber, in Spreng. Syst. Veg. Cur. Post. p. 124.
S. muricata, Lindl. Bot. Reg. sub t. 1917?

Has. New South Wales, near Sydney and Puen Buen. Also,
Cook’s River ; a form with the rugose-reticulated cocci somewhat muri-
cate, and the leaves broader.

VIVIANACES, 307

2. STACKHOUSIA LINARIIFOLIA, A. Cunn. (ined. ?)

Has. Hunter’s River and Puen Buen, New South Wales. (In fruit.)

Orv SURIANACE &.

1. SURIANA, Plumier.
1. SURIANA MARITIMA, Linn.

Has. Society Islands. Clermont Tonnere, Raraka, King’s, Wil- -
son’s, and the Coral Islands generally. Mangsi Islands.

Orv VIVIANACE

jl. VIVIANIA, Caw.

1. VIVIANIA MARIFOLIA, Cav.

Viviania marifolia, Cav. in Ann. Cienc. Nat. 7, p. 211, t. 49; Gay, Fl. Chil.-1,
p- 397.

Has. Andes of Chili, above Santiago.

Orn GERANIACE &.

1. GERANIUM, Linn.
§ 1. GERANIUM verum.

1. GERANIUM DISSECTUM, Linn.

Has. Rio Negro, North Patagonia (nearly the G. Putagonicum,
Hook. f. Fl. Antare. p. 252). Chili, from Valparaiso to the Andes
above Santiago. Bay of Islands, New Zealand (the var. retrorsum
and var. glabratum, Hook. f.). New South Wales (G. piloswm, For-
ster, Xc.).

I adopt Dr. Hooker’s view, in referring the Geranium pilosum and
G. retrorsum of Forster, as well as his own G. Patagonicum, to the

Kuropean G. dissectum. At least, sufficient characters for distinguish-
ing them have not been detected.

2. GERANIUM POTENTILLOIDES, L’ Her.

Geranium potentilloides, L’ Her. in DC. Prodr. 1, p. 6839; Hook. f. Fl. N. Zeal. p. 40.
G. microphyllum, Hook. f. Fl. Antarc. p. 8, t. 5; Raoul, Pl. N. Zel. p. 47.

Has. Bay of Islands, New Zealand. (A form between the var.
macrophyllum and. var. debile, Hook. f. 1. c.)

3. GERANIUM bDirrusuM, H. B. K.

Geranium difusum, H. B. K. Nov. Gen. & Spec. 5, p. 230; DC. Prodr. 1, p. 639.

~_ = see

GERANIACES. 309

Has. Bafos, Andes of Peru. (Also (not in flower), island of San
Lorenzo, near Callao).

This species has been collected in Peru by Matthews, M’Lean, &c.,
and near Quito by Jameson and Hall. If the specimen from the
island of San Lorenzo truly belongs to the species, it has a thick and
fleshy, fusiform, perennial roet. The leaves are softly hairy, above
as well as below. On small specimens and on ultimate branches they
are as small as described by Kunth (7 or 8 lines broad); but the
larger ones are an inch in diameter. Only the lower peduncles are
shorter than the petioles; the upper are longer than the leaves.
Petals 3 or 4 lines long, exceeding the calyx. Carpels sparingly
pubescent. Seed smooth and even, or more obscurely reticulated
under a lens than in G. dissectum. From the character, it would
seem that G. Berterianum of Colla may be the same plant.

4, GERANIUM SESSILIFLORUM, Cav.

Geranium sessiliflorum, Cay. Diss. 4, p. 198, t. 77; DC. Prodr. 1, p. 639; Walp. Rel.
Meyen. p. 315; Hook. f. Fl. Antarc. p. 252.

Has. Obrajillo, Bafios, and Alpamarea, Andes of Peru. (Also
gathered by M’Lean and Matthews, fide herb. Hook.)

5. GERANIUM CRASSIPES, Hook. tn Herb.

G. nanum, ecespitosum, acaule ; caudice crasso multicipite stipulis imbri-
catis obtecto; foliis cum petiolis undique cano-sericeis 5—T-partitis,
segmentis oblongis obovatisve confertis integerrimis vel bilobatis ; pedun-
culis brevissimis unifloris; sepalis sericeis muticis ; coccis pubescen-
tibus; seminibus levibus.

Has. Obrajillo, Andes of Peru. (Cerro Pasco, Matthews, in herb.
Hook.)

Plant dwarf and stemless, forming dense ceespitose tufts, of about an

inch in height; the caudex thick, branching, the upper part elothed
78

310 PHANEROGAMITA.

with the large and scarious, dark chestnut-coloured wmbricated stipules,
which are adnate to the base of the petioles. Leaves much crowded,
canescently-silky all over, as well as the petioles, roundish in outline,
5—7-parted ; the segments oblong or narrowly obovate, acute, not over 3
lines in length, crowded, entire, or sometimes, especially the middle
one, two-lobed. Peduncles one-flowered, shorter than the leaves, among
which they are partly hidden, silky-tomentose. Calyx in fruit as
Iong as the peduncle; the sepals silky like the leaves, oblong-ovate,
rather obtuse, pointless, nearly 3 lines long. Petals not seen. The
persistent filaments nearly distinct, subulate from a broad base, as
long as the calyx, a little hairy externally. Beak of the fruit 5 lines
long, canescently pubescent, as are the carpels. Seeds with a smooth
and even testa.

In the Peruvian and Quitensian Andes, there are some other, still
undescribed species, allied to this, which I ‘have not the means of
elucidating.

6. GERANIUM MULTIPARTITUM, Benth. ?
Geranium muiltipartitum, Benth. Pl. Hartw. p. 166?
Has. Obrajillo; with the preceding species.

This species differs from the foregoing principally ‘in its ‘finer, ‘but
‘still canescent pubescence, its narrower stipules, and longer petioles,
the segments of the leaves again 2—5-cleft or parted; the sepals nar-
rower and less hairy. The single small specimen, is in fruit.

§ 2. NEUROPHYLLODES.—Frutices vel arbuscule Sandwicenses; pedunculis seepe
multifloris ; staminibus omnino vel fere discretis; foliis omnibus alternis cuneatis ovali-
busve nervosis !

A remarkable group of plants, peculiar to the Sandwich Islands,
whence we have four species, only one of which has been published.
They are all shrubby plants, except possibly the first, and one of
them is truly arborescent, with a trunk of three inches or more in
diameter! They are further remarkable for their wedge-shaped or

GERANIACES. 211

oval and parallel-nerved leaves, all of which are alternate; and the
peduncles in some of them bear many-flowered cymes. The flowers
and fruit, however, are just those of true Geraniums, except, perhaps,
that the filaments are very slightly, if at all, united at the base.
Their aspect, although peculiar, would lead one to refer them to
Pelargonium rather than to Geranium: but their flawers are perfectly
regular (except in G. arborewm) and symmetrical, and there is no trace
of a spur or other downward prolongation of the calyx.—The first
species, G. multiflorum, shows least of the peculiarities of the section
in its foliage, which is thin, and with the nerves more branched and
inosculating towards their summit than in the succeeding species.
Its branches, moreover, are scarcely woody.

7. GERANIUM MULTIFLORUM, Sp. Nov. (Tab. 29.)

G. fruticosum? molliter pubescens; foliis membranaceis obovato-rotun-
datis grosse dentatis (basi subcuneata integerrima) utrinque viridibus ;
pedunculis elongatis cymam lacam multifloram gerentibus; sepalis
mucronatis.

Has. District of Waimea, Hawaii, Sandwich Islands.

The specimen is-a single branch; and the notes upon the plant do
not record the height which it attains, nor whether the stem is
woody. But it probably is so, as the branch, although fistulous, is
somewhat lignescent. It appears to have been broken from a decum-
bent stem. The young branchlets, peduncles, leaves, &c., are softly
pubescent with fine and mostly spreading hairs; which on the young
leaves are somewhat silky. Leaves alternate, approximate, membra-
naceous in texture, roundish-obovate, an inch and a half long, by an
inch or more in width, coarsely toothed, except the more or less cuneate
base, which is entire, green on both-sides, T-9-nerved from the base;
the nerves more or less forked, and connected by anastomosing vein-
lets. Petioles about half an inch long, pubescent, terete. Stipules
setaceous-subulate from a dilated and connate-clasping base, minutely
hairy or ciliate, scarious, brownish, 4 or 5 lines long; their bases only
adnate to the base of the petiole. duncle becoming lateral by the
evolution of a branch, elongated (3 inches in length), bibracteolate in

312 PHANEROGAMIA.

the middle and at its summit, where it branches into a repeatedly
trichotomous, open and loose, many-flowered cyme. Bracts and bract=-
lets ovate-lanceolate and subulate, scarious. Internodes of the cyme
from 5 to 12 lines long: the pedicels from one to 3 lines long, erect,
both in flower and in fruit. Sepals pubescent, ovate-oblong, mucro-
nate, pubescent, with slightly scarious margins, 3 to 4 lines long.
Petals obovate, entire, glabrous, a little longer than the calyx, equal
and similar; their colour undetermined. Glands opposite the sepals,.
woolly. Stamens 10, all fertile and similar: filaments distinct, their
dilated bases obscurely hairy; the five exterior ones a little shorter :.
anthers, &c., as‘in the genus. Ovary and beak pubescent; the latter
8 or 9 lines long. Styles little produced beyond the apex of the beak,
above which their bases only are united; in fruit separating elastically
from the base upwards, in the manner of the genus, glabrous inside.
Cocci minutely pubescent. Seed with a smooth and even testa.

PLATE 29, A.—GERANIUM MULTIFLORUM: a flowering branch, of the
natural size. Fig. 1. A sepal; inside view. 2. A petal. 3. Flower,
from which the calyx and corolla have been removed. 4. Vertical
section of a flower through the ovary, &c.—All the analyses mode-
rately magnified.

8. GERANIUM CUNEATUM, Hook. (Tab. 29.)
G. fruticosum ; foliis coriaceis euneatis apice scepius truncato 3—5-den-
tatis 0—1-nervatis ; pedunculis plurifloris vel abortw paucifloris ; sepalis

mucronulatis ; petalis albis.

Var. a. Mrnziesit: foliis utrinque glabris seu glabellis viridibus. (Tab.
29, B.)

_ Geranium cuneatum, Hook. Ic. Pl. t. 198.

Var. @ HYPOLEUCUM: foliis subtus sepalisque argenteo-incanis nitentibus.
(Tab. 29, C.)

Var. y. HOLOLEUCUM : foliis utringue sepalisque argenteo-incanis niten~
tibus. (Tab. 29, D.)

GERANTIACE &, 313

Has. Hawaii, Maui, and Kauai, Sandwich Islands; on the moun-
tains. a. Aleala, Hawaii; a small form. @. Mouna Loa; a common
shrub from the crater of Lua Pele to the elevation of 8,000 feet.
Mountains of Kauai; a very small form, without flowers. y. Mouna
Kea; also Mouna Loa, with the preceding form; abundant through
the region of Hdwardsia. East part of Maui, on the banks of the
crater Haleakala.

A. shrub, one or two feet or more in height, much branched; the
foliage very variable as to size and pubescence. Branches glabrate,
or the younger ones cinereous, at least in vars. @. and y., thickly
beset with the scaly and persistent stipules. These are subulate from
their dilated and connate bases, as in the foregoing species. Leaves
crowded on the branchlets, coriaceous in texture, from 5 to 12, or even
18 lines long, and from 3 to 6, or 9 lines wide, wedge-shaped, with the
apex truncate, or very obtuse, and 5-5-toothed, tapering to an acute
base, the sides entire, strongly 5-T-nerved ; the nerves parallel, simple,
or very sparingly forked at their apex. Their surface is either gla-
brous or glabrate (very minutely pubescent under a lens), and green,
as in the var. a.; or silvery-white, with a fine silky pubescence under-
neath, while the upper surface is green and glabrous or nearly so, as
in var. 3.; or else silvery-whitened and shining both sides, as in var. y.
Petioles 2 or 3 lines in length. Peduncles terminal or opposite the
leaves, shorter than the leaves, or often elongated, cymosely several-
(9-12-) flowered, or by abortion 1-3-flowered ; the peduncles or branches
of the cyme being pluribracteate. Bracts subulate, small. Pedicels
3 to 8 lines long. Sepals 2 or 3 lines long, ovate or oblong-ovate,
barely mucronulate, nearly glabrous in the smooth-leaved form, in
the others silky-canescent. Petals white, obovate, 4 or 5 lines in
length. Hypogynous glands minute, hairy. Stamens as in G. multi-
jlorum; but the filaments rather more hairy at their dilated base.
Ovary and beak minutely pubescent. Styles little prolonged beyond
the beak, their summits distinct; in fruit recurved from the base,
glabrous inside. Seed smooth and glabrous, very minutely striate-
reticulated under a lens.

The specimen gathered by Menzies, in Vancouver’s voyage, and
figured by Hooker, resembles the more common form of our var.

hololeucum, except that the leaves are wholly glabrate. The speci-
79

314 PHANEROGAMIA.:

mens of our glabrous variety are smaller in all their parts, and were
apparently gathered at a greater elevation; they have short and few-
flowered peduncles, and are wholly in fruit. The form of var. hypo-
leucum from Kauai apparently came from a great elevation: the
plants are only a span high, naked, as.if they had grown in moss; the
branches tipped with a-eluster of small leaves, and destitute of flower
and fruit. The var. hololeucum usually exhibits the largest leaves and
a fuller cyme of flowers.

Piate 29. B, C, D.—GrERANIUM cUNEATUM.—B. Var. MeEnzizsmr:
a dwarf state, in fruit. Fig. 5. Seed, magnified. 6. The embryo,
detached and cut across, magnified —C. Var. HYPOLEUCUM: in flower.
Fig. 7. Sepal. 8. Petal. 9. A filament.—The dissections enlarged.
—D. Var. HOLOLEUOUM, of the natural size.

9, GERANIUM OVATIFOLIUM, Sp. Noy. (Tab. 30:)

G. fruticosum ; foliis chartaceis longe petiolatis ovatis acutis argute ser-
ratis (bast rotundata tantum integerrima) T—-11-nervatis supra seepius
glabratis subtus sericeo-canescentibus ; pedunculis paucifloris ; sepalis
mucronulatis; petalis albis venis purpureis pictis.

Has. East division of Maui, Sandwich Islands, on the north bank
-of the crater Haleakala.

Shrub apparently of 2 or 3 or several feet in height, with spreading
branches, which are thickly scarred, or the branchlets clothed with
the conspicuous, connate, subulate-pointed, scaly, persistent stipules,
which are like those of the two preceding species. Leaves chartaceous,
or nearly membranaceous, or in a stunted form coriaceous, ovate, more
or less acute, rather finely but sharply serrate with appressed mucro-
nate teeth, except towards the obtuse or rounded base, which is entire,
from one to 23 inches long, 7—ll-nerved ; the nerves sparingly-forked
or branched and inosculating; the wpper surface glabrate and green,
or rarely canescent when young; the lower more or less whitened
with a fine silky or silvery pubescence. Petioles slender, pubescent,
from half an inch to I inches in length. Peduncles opposite the
leaves, or terminal, an inch or so in length, usually forked, /ew-
(3-7-)flowered, or by abortion one-flowered; the peduncle or its

GERANIACEZ. 315

branches pluribracteate. Flowers rather larger than those of G.
cuneatum ; the sepals mucronulate, ovate-oblong, sometimes tinged with
purple. “Petals white, with purple veins.” Stamens, styles, &c.,
nearly as in the preceding species. Seed minutely striate-reticulated
under a lens.

PLATE 30.—GERANIUM OVATIFOLIUM: branches, of the natural size.
Fig. 1. A sepal, inside view. 2. A petal. 3. Flower, from which
the calyx and corolla have been removed. 4. Calyx and fruit. 5.
A seed.—All the details enlarged. |

10. GERANIUM ARBOREUM, Sp. Nov. (Tab. 31:)

G. arborescens, parce pubescens ; foliis chartaceis vel membranaceis ovatis
seu ovalibus obtusis plerumque subcordatis argute serratis T-nervatis
utrinque concoloribus ; pedunculis folio brevioribus wni-bifloris ; sepalis
aristato-acuminatis; petalis rubris “ superioribus cucullatis;” stylis
coalitis ultra rostrum longe productis. |

Has. Eastern part of the island of Maui, Sandwich Islands; near
the upper border of the forest, at about the elevation of 6000 feet.

An arborescent species ; the: trunk: becoming 4 inches or more in dia-
meter, and attaining the height of 6 to 12 feet! Branchlets squarrose
with the persistent stipules; which are like those of the preceding
species, when young pubescent with fine -and soft hairs, as are the
leaves, peduncles, calyx, &e. Leaves chartaceous or membranaceous in
texture, ovate, or sometimes nearly oval, obtuse, rounded, and usually
subcordate atthe base, sharply serrate almost down to the base with
mucronate teeth, about seven-nerved (the nerves sparingly branching,
especially the marginal ones), nearly of the same hue both sides,
softly pubescent, more or less glabrate with age, an inch to 14 inches
long, the larger ones over an inch in width. Petioles from 3 to 8 lines
long. Peduncles 1—2-flowered, perhaps sometimes three-flowered, : ter-
minating the branches, or opposite the leaves, from half an inch to an
inch in length, with one or more pairs of subulate bracts. Flowers
larger than in any of the cognate species, not unlike those of a Pelar-
gonium, but destitute of all traces of an adnate spur, and the stamens

316 PHANEROGAMIA.

all alike and fertile. Sepals oblong or oblong-lanceolate, abruptly
awn-pointed, half an inch long. “ Petals red,” obovate-spatulate, 8 or
9 lines long, equal and similar in shape, but, according to Dr. Picker-
ing, the three upper are more upright and cucuillate, the’ lower more
spreading. Hypogynous glands inconspicuous. Filaments 10, similar,
and all antheriferous, capillary, persistent, longer than the sepals after
anthesis, sparsely hairy, more densely so at their dilated base. Ovary
villous. Styles hairy as far as they are united with the slender beak
(7 or 8 lines long in fruit), then glabrous and prolonged for half an
inch as a filiform compound style, their separate and stigmatose apices
short. Carpels hairy in fruit. Seed glabrous, minutely reticulated
under a lens.

To this most remarkable species belongs a portion of a stem in the
museum of the Exploring Expedition, marked by Mr. Brackenridge,
“Stem of a Pelargonium, from the mountains of Maui.” This stem
is three inches in diameter, with a smooth and thin bark: the wood
is pretty hard, compact, and close-grained : it exhibits about fourteen
indistinct concentric layers, surrounding a small pith, and traversed
by rather conspicuous medullary rays. Its intimate structure does
not present any marked peculiarity.

The plant would be a curious and striking one in cultivation. The
flowers are said by the collectors, Dr. Pickering and Mr. Brackenridge,
to be quite showy ;—more so, doubtless, than they would appear to
be from our figure; the scanty flowering specimens in the collection
not being in a very good state. Nor does the plate distinctly represent
the irregularity of the corolla; this peculiarity (which is not shared
by the other species of the group) having been made known to us by
Dr. Pickering, only since this sheet was in type. Notwithstanding
their unequal direction, the petals appear to be all alike in shape and
size: nor does the species partake of any other characters of Pelargo-
niwm, except, perhaps, in the general habit and foliage.

PLATE 31.—GERANIUM ARBOREUM: branchlets, of the natural size,
in flower and fruit. Fig. 1. A sepal; inside view. 2. A petal. 3.
Flower, vertically divided through the ovary. 4. Stamens, seen an-
teriorly and posteriorly. 5. Pistil, in fruit, with the persistent sta-
mens and calyx. 6. A seed. 7. Transverse section of the same.
—The details variously magnified.

GERANIACES. - 317

2. ERODIUM, Ler.
1. Eropium cicutartum, Leman.

Has. Rio Negro, North Patagonia. Obrajillo and Bafios, Andes of
Peru. A minute form, apparently of this widely diffused species,
was gathered on the high Chilian Andes, above Santiago.

2. ERODIUM MALACHOIDES, Walld.

Has. Obrajillo, Andes of Peru. Doubtless introduced from Europe.

3. PELARGONIUM, P#er.

1. PELARGONIUM CLANDESTINUM, L’ Her.
Pelargonium clandestinum, U’ Her. ined.; A. Cunn. in Ann. Nat. Hist. 8, p. 117;

Hook. f. Fl. N. Zeal. p. 41.

P. Acugnaticum, Thouars, FI. Tristan d’Acugn. p. 44, t. 13 (1811); DC. Prodr. 1,
p- 660.
P. anceps, Ait. Hort. Kew. 2, p. 40?

Fias. Bay of Islands, New Zealand.

2. PELARGONIUM AUSTRALE, Willd.

Pelargonium australe, Willd. Spec. 8, p. 675; DC. Prodr. 1, 654; Hook. Jour. Bot.
2, p. 415.

Has. Sydney, Woolongong, &c., New South Wales: several forms.

3. PELARGONIUM CUCULLATUM, Ait.

Has. Cape of Good Hope; in the immediate vicinity of Cape Town.
80

318 PHANEROGAMIA.

4, PELARGONIUM MYRRHIFOLIUM, Ait.

‘Has. Cape of Good Hope; in the immediate vicinity of Cape Town.

Orv TROP ZOLACE &.

1. TROP ZOLUM, Linn.
1. Trop“oLtum-masus, Linn.

Has. Around Lima and Callao, Peru: common in waste grounds.

2. TROPAOLUM TUBEROSUM, Ruiz & Pav.

‘Tropeolum tuberosum, Ruiz.& Pay. .Fl. Per. 3, p. 77, t. 314, f. 6; Hook. ‘Bot.
Mag. t. 8714, & Ic. Pl. t. 653.

‘Has. High Andes of Peru; very common from Obrajillo to Culnai.
‘Also, cultivated below Culnai. “ Root.edible, even in.a raw. state.”

3. TROPOLUM TRICOLOR, Lindl.

Tropeeolum tricolor, Lindl. Bot. Reg. t. 1935; Gay, Fl. Chil. 1, p. 411.
T. tricolorum, Sweet, Brit. Fl. Gard. t. 270 ; "Hook. & Arn. Bot. Beech. "yey: p. 14;
Hook. Bot. Mag. t. 3169.

Has. Ravines, near Valparaiso: without flowers or fruit.—Similar
sterile-specimens occur in the Sandwich Islands collection, from Puna,
Hawaii: probably from a cultivated plant.

Orv. LINACE &.

1. LINUM, Linn.

1. Linum macrat, Benth,

Linum Macrei, Benth. in Bot. Reg. sub no. 1826; Planch. in Hook. Lond. Jour.
Bot. 7, p. 490.
LL. aquilinum, Molina, Hist. Nat. Chil.? DC. Prodr. 1, p. 126?

Has. Near Valparaiso, Chili.

In these specimens, the sepals have smooth and even margins, and
the styles are united to the very summit. The stipular glands are
occasionally present in these, as also in the original specimens of L.
Macree.

2. LINUM POLYGALOIDES, Planch.

Linum polygaloides, Planch. in Hook. Lond. Jour. Bot. 7, p. 490.
L. oligophyllum, Presl, Rel. Heenk. 2, p. 1, haud Willd. ex Planch.

Has. Obrajillo, Andes of Peru.

3. Linum monoGynum, forst.

Linum monogynum, Forst. Prodr. Fl. Ins. Austr. p. 23; Hook. Bot. Mag. t. 3574;
Hook. f. Fl. N. Zeal. p. 28.

Has. Bay of Islands, Waya-ruru Bay, New Zealand.

320’ PHANEROGAMITA. 4

4, LinuM MARGINALE, A. Ouin:

Linum marginale, A. Cunn. in herb. Hook.; Planch. in Hook. Lond. Jour. Bot. 7,
p. 169.

Has. Sydney, New South Wales.

The root certainly appears to be that of a merely annual plant.

®. Linum ArricaAnum, Linn.

Has. Cape of Good Hope, in the vicinity of Cape Town.

Orv. OXALIDACE 2.

¥. OXALIS, Linn.

* Cosmopolitane.
1. Oxais cornicuata, Linn.

Haz. Organ Mountains, Brazil. Lima, Peru (a very hairy form).
Tahiti. Samoan Islands. Oahu and Hawaii, Sandwich Islands.
Bay of Islands, New Zealand ;—the varieties which answer to var. y.
microphylla, Hook. f. (the O. exilis and O. microphylla, A. Cunn., and
O. reptans, Forster) ; var. 6. ciliifera, Hook. f. (O. cilitfera, A. Cunn.) ;
and var. e. crassifolia (O. crassifolia, A. Cunn.). New South Wales;
several forms, among them the O. microphylla, Poir. Baiios, Luzon.
St. Helena. Cape of Good Hope.

OXALIDACEA, 29]

* * Brasilienses.
2. OXALIS FRUTICOSA, Raddi.

Oxalis fruticosa, Raddi, Mem. Bras. Add. p. 22; DC. Prodr. 1, p. 690; St. Hil.
Fl. Bras. 1, p. 116.

Has. Near Rio Janeiro; abundant.

3. OXALIS BARRELIERI, Jacq.
Oxalis Barrelieri, Jacq. Oxal. no. 4, t. 3 (Barr. Ic. t. 1169); DOC. Prodr. 1, p. 690.

Has. Rio Janeiro.

4, OXALIS HEDYSAROIDES, H. B. K.

Oxalis hedysaroides, H. B. K. Nov. Gen. & Spec. 5, p. 247; DC. Prodr. 1, p. 691;
Zuce. Monogr. Amer. Oxal. p. 50.

Has. Rio Janeiro, and on the Organ Mountains, Brazil.

5. OXALIS MARTIANA, Zuce.

Oxalis Martiana, Zucc. Monogr. Amer. Oxal. p. 20, & Nachtr. p. 27; Hook. Bot.
Mag. t. 3938.

O. urbica, St. Hil. Fl. Bras. Mer. 1, p. 126.

O. floribunda, Link & Otto, Abbild. Gewisch. 1, p. 19, t. 10.

O. lasiopetala, Zuce. |. c.? Hook. Bot. Mag. t. 3932.

O. bipunctata, Graham, in Bot. Mag. t. 2781.

.

Has. Roadsides, near Rio Janeiro: common.

There are also specimens gathered at Hilo, Hawaii, Sandwich
Islands, marked by Mr. Brackenridge as probably not indigenous :

doubtless-it was introduced from Brazil.
81

322 PHANEROGAMIA.

* * * Peruviane et Chilenses.
\

6. OXALIS ELEGANS, A, B.K.
Oxalis elegans, H. B. K. Nov. Gen. & Spec. 5, p. 182, t. 466; Zucc. Monogr. 1. c.
Has. Near Lima, Peru;—where the same form, with smaller
flowers than in the figure above cited, was gathered by Matthews.
7. OXALIS LOBATA, Sims.
Oxalis lobata, Sims, Bot. Mag. t. 2886; Zuce. 1. c. p. 25; Gay, Fl. Chil. 1, p. 426.

Has. Dry grounds around Valparaiso, Chili; common.

8. OXALIS LINEATA, Gill.

Oxalis lineata, Gillies, in Hook. Bot. Mise. 8, p. 162; Gay,. Fl. Chil..1, p. 489.
O. polyantha, Walp. Rel. Meyen. p. 819, ex char.

Has. High Andes, above Santiago, ChiliiaAn alpine state of the

. species, with smaller leaves and creeping rootstocks ; not.in. flower.
9. OXALIS GEMINATA, Hook. & Arn.

Oxalis geminata, Hook. & Arn.’ Bot. Misc. 3, p. 165; Gay, Fl. Chil. 1, p. 457.

-Has. High Andes, above Santiago, Chili. (Not in flower.)

10. OxALis pyemma, Sp. Nov.

‘O. parva: vel minima, e radice annua? subcaulescens, parce hirsutula;
stipulis conspicuis scariosis petiolo ad medium usque adnatis apice
libero fimbriatis; foliis trifoliolatis; foliolis obcordatis; pedunculis

OXALIDACE A. 393

unifloris folio cequilongis medio tribracteolatis ; bracteolis angustissime
linearibus; flore minimo; petalis integris luteis; ovarii loculis uni-
ovulatis ; seminibus rugosis.

Has. Near Baiios, and Culnai, on the high Andes of Peru.

A very small or minute plant, half an inch to an inch high, branch-
ing, from a slender and apparently annual root, into a number of very
short stems or rootstocks, which are scaly with persistent stipules,
minutely and sparsely hairy under a lens, especially the leaves. Sti-
pules conspicuous and large for the size of the plant, 4 or 5 lines long,
scarious, very thin, adnate to the lower half of the petiole, their free
apex fimbriately cleft into setaceous or linear lobes. The free portion
of the petiole filiform, scarcely margined or dilated. Leajlets 3, obcor-
date, from one to 22 lines long, thickish, areolate-punctate, nearly
sessile. Peduncles one-flowered, filiform, about the length of the leaves,
bibracteolate near the middle with a pair of scarious, narrowly linear
bractlets, which are entire, or nearly so. Flower minute, at most 14
lines in length. Sepals oblong, obtuse, glabrous. Petals obovate,
entire, or barely retuse, “ yellow,” scarcely exceeding the sepals, appa-
rently more or less united above the base. Filaments glabrous.
Styles short, glabrous. Stigmas capitate. Cells of the ovary apparently
one-ovuled. Seed solitary in each cell of the globular-ovoid membra-
naceous capsule, ovoid; the integument transversely rugose.

This is doubtless related to the Bolivian Oxalis parvula of Remy
(perhaps O. nubigena of Walpers); which is more strictly acaulescent,
glabrous, with white and obcordate petals, and the cells of the ovary
many-ovuled. .

11. Oxatis PickErineu, Sp. Nov.

O. caulescens e radice fusiformi lignescente, viscoso-pubescens ; stipulis
nullis; folits trifoliolatis ; foliolis parvis obovato-cuneatis emarginatis
glabratis petiolulatis; pedunculis folium superantibus apice bifidis
cymoso-plurifloris ; pedicellis bracteolis lanceolatis brevioribus ; sepalis
ovato-acuminatis capsula oligosperma longioribus ; filamentis stylisque
viscoso-barbatis.

324 PHANEROGAMITIA.

Has. Between Caballeros and Obrajillo, and near Banos, Andes of
Peru.

Stems 3 or 4 inches high, very numerous from the long and: fusi-
form woody root, ascending, branching, viscous-pubescent, as are the
slender petioles, peduncles, &c. Stipules none. Leaflets 3, very small
(only a line and a half in length), distinctly petiolulate, thickish, pale,
becoming glabrous, obovate and wedge-shaped, the apex notched, or
slightly obcordate. Peduncles axillary, longer than the leafstalks, an
inch or more in length, naked, two-cle/t at the apex, cymosely several-
jlowered: the two branches usually undivided; each node bearing a
pair of lanceolate bractlets, of about a line in length, which subtend a
single flower. Pedicels shorter than the bractlets. . Sepals ovate or
ovate-lanceolate, gradually acuminate, 2 to 3 lines long, much shorter
than the petals, the colour of which cannot be determined. Filaments
thickened upwards, exceeding the styles, and like them bearded with
viscous hairs; the longer ones often toothed near the base. Stigmas
capitate. Capsule ovoid, shorter than the calyx, very few-seeded. Seeds
ovoid, rugose.

* * * * Antarctice.

12. Oxatis MAGELLANICA, Forst.

Oxalis Mayellanica, Forst. in Comm. Geett. 9, p. 33; Hook. f. Fl. Antare. p. 253, &
Fl. N. Zeal. p. 42, t. 18.

O. lactea, Hook. Jour. Bot. 2, p. 416, ex Hook. f.

O. cataracte, A. Cunn.; Hook. Ic. Pl. t. 418, ex Hook. f.

Has. Orange Harbour, Fuegia.

There are two forms in the collection: one of them, in fruit, well
agrees with Dr. Hooker’s figure, from New Zealand specimens: the
other is smaller in all its parts, and more delicate. The seeds are
obscurely ribbed longitudinally, but not transversely wrinkled.

x * * x * Capenses.

13. OxA.is nirTA, Linn.

Has. Cape of Good Hope, in the vicinity of Cape Town.

ZYGOPHYLLACEA. 325

14. OXALIS PURPUREA, Jacg. —

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.

15. OXALIS FILICAULIS, Jacq.

Has. Cape of Good Hope; with the preceding.

16. OXALIS VERSICOLOR, Linn.

Has. With the three preceding species.

Orv. ZYGOPHYLLACE &.

1. TRIBULUS, Tourn.
1. TRIBULUS CISTOIDES, Linn.

Tribulus cistoides, Linn. Spec. Pl. p. 887; Jacq. Hort. Scheenb. t. 103; Gray, Gen.
Am. Bor. Ill. 2, p. 116, t. 145.

Has. St. Jago, Cape de Verdes. Sandwich Islands, near Honolulu,
Oahu. Doubtless introduced from South America.

Endlicher, in his Synopsis Fl. Ins. Austr. (in: the Annals of the
Vienna Museum), has inadvertently referred this species to Scopoli’s
genus Kallstreemia, restored by him: but it is a true Tribulus having
a pentacoccous fruit, as is stated by DeCandolle, on the authority of
Kunth. The spines of the fruit vary greatly in size.

82

326 PHANEROGAMIA.

2, FAGONIA, Tourn.
1. FAGONIA ASPERA, Gay.
Fagonia aspera, C. Gay, Fl. Chil. 1, p. 470.

Has. Peru, in uplands, near Yanga. (A very small, diffuse form,
appressed to the ground.) :

38. LARREA, Cav.
1. LARREA DIVARICATA, Cav.

Larrea divaricata, Cav. Ic. Pl. 6, t. 560, f. 1; Hook. & Arn. Bot. Misc. 3, p. 166;
Gay, Fl. Chil. 1, p. 473.

Has. Rio Negro, North Patagonia: where it is a common bush on
the sandy plains.

This is the species to which our northern Larrea Mexicana 1s most
related. The leaves are by no means villous, as they are described ;
but the younger ones are more or less silky-pubescent; the older are
glabrate. .

On. RUTACEA (DIOSMEA, RB. Br).

It is abundantly evident, from the comparison of Huodia, Melicope,
Acronychia, &c. with Zanthowylum on the one hand, and with the
true Diosmee on the other, that the characters by which Endlicher
distinguished his Zunthoxylee from his Diosmee are of no value what-
ever; and that the view of the family long ago suggested by Brown
should be adopted; retaining, however, the name of Hutacece. Al-
though the Rue and its nearest allies are “not calculated to give a
clear idea of the order, from the usual structure and habit of which
they deviate in some important points ;” yet the name in use for the
group cannot now be readily discarded, any more than in the analo-
gous case of the Huphorbiaceee.

1RUTA, Lin.
1. Ruta Bracteosa, DC.

Has. Near Valparaiso, Chili. Doubtless introduced from the south
of Europe.

2. AGATHOSMA, Willd.
1. AGaTHosMA SPICATA, Licht.

Agathosma spicata, Licht. in Rom. & Schult. Syst. Veg. 5, p. 447; DC. Prodr. 1, p.
715, sub Diosma.

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.

328 PHANEROGAMIA.

3. ACMADENTA, Bartl. & Wendl.
1. ACMADENIA JUNIPERINA, Bartl. & Wendl.
Acmadenia juniperina, Bartl. & Wendl. Beitr. Diosm. 1, p. 61.

Has. Cape of Good Hope; common in the immediate vicinity of
Cape Town.

4. DIOSMA, Berg.
1. Diosma succuLeNntTA, Willd.

Has. Cape of Good Hope; common, with the preceding.

9 COLEONEMA, Baril. & Wendl.
1. CoLEONEMA PULCHRUM, Hook.
Coleonema pulchrum, Hook. Bot. Mag. t. 3340; Eckl. & Zeyh. Enum. p. 106.
Has. Cape of Good Hope; with the preceding species.
6. CORREA, Smith.
1. CoRREA ALBA, Andr.
Correa, alba, Andr. Bot. Rep. t. 18; Vent. Hort. Malm. t. 18; DO. Prodr. 1, p. 719.

Has. Hunter's River, New South Wales.

RUTACE &. 399

2. CORREA VIRENS, Smith.

' Correa virens, Smith, Exot. Bot. 2, p. 25, t. 72; DC. Prodr. 1, p. 719.
C. viridiflora, Andr. Bot. Rep. t. 486.
C. refleca, Pers. Ench. 1, p. 419; Labill. Fl. Nov. Holl. 2, p. 120.
Mazeutoxeron reflecum, Labill. Voy. a la recherche de La Perouse, 2, p. 66, t. 19.

Has. New South Wales; with the preceding, and, like it, not in
flower.

7 PHILOTHECA, Rudge.

1. PHILOTHECA AUSTRALIS, Rudge.
Philotheca australis, Rudge, in Linn. Trans. 11, p. 298; DC. Prodr. 1, p. 721.

Has. New South Wales, in the vicinity of Sydney, and on Hun-
ter’s River.

2. PHInoTHECA REICHENBACHIANA, Sieber.

Philotheca Reichenbachiana, Sieber, Pl. Nov. Holl. Exsic. no. 308; Spreng. Syst.
Veg. Cur. Post. p. 2538; Reichenb. Ic. Bot. Exot. t. 200.

Has. Sydney, New South Wales.

8. CROWEA, Smith.
1. CRowEA SALIGNA, Andr.

Crowea saligna, Andry. Bot. Rep. t. 79; Vent. Hort. Malm. t. 7; DC. Prodr. 1, p.
720; Reichenb. Fl. Exot. t. 152.

Has. Sydney, New South Wales.
88

330 PHANEROGAMIA.

9, ERIOSTEMON, Smith.
1. ERIostEMON BUXIFOLIUM, Smith.

Has. Sydney, New South Wales: nearly the var. a. of De Candolle.

2. ERIOSTEMON VERRUCOSUM, A. Rich.

Eriostemon verrucosum, A. Rich. Voy. Astrolab. 2, p. 74, t. 26.

Has. New South Wales; with the preceding.

3. ERIOSTEMON SALICIFOLIUM, Smith.

Has. New South Wales; with the preceding.

4, ERIOSTEMON CUSPIDATUM, A. Cunn.

Eriostemon cuspidatum, A. Cunn. in Hook. Jour. Bot. 1, p. 254; Lodd. Bot. Cab.
t. 1247.

Has. In the vicinity of Sydney, New South Wales.

10. BORONIA, Smith.
1. BoroNIA SERRULATA, Smith.

Has. In the vicinity of Sydney, New South Wales.

2. BoRoNIA PARVIFLORA, Smith.

Has. New South Wales; probably from the vicinity of Sydney.

Hap.

Has.

HAs.

Has.

Has.

Has.

RUTACES. 99)

3. BORONIA POLYGALAEFOLIA, Smith.

In the vicinity of Sydney, New South Wales. (In fruit.)

4, BoRONIA LEDIFOLIA, Gay.

New South Wales; probably from near Sydney.

5. BoRONIA TETRANDRA, Labill,

Sydney, New South Wales.

6. BorontA PINNATA, Smith.

Sydney, New South Wales,

ll. ZIERIA, Smith.

l. ZiERIA LANCEOLATA, R. Br.

Near Sydney, and along Hunter’s River, New South Wales.

2. ZIERIA PAUCIFLORA, Smith.

Sydney, and on Hunter’s River, New South Wales.

12, PILOCARPUS, Wahl.

1. PILocARPUS SPICATA, S¢. Hil.

Pilocarpus spicata, St. Hil. Hist. Pl. Rem. Bras. t. 16, & Fl. Bras. Mer. 1, p. 83.

Has.

Near Rio Janiero, Brazil. (Imperfect specimens.)

332 PHANEROGAMIA.

13. EUODIA, FPorst.

1. Evopra HorTENSIS, Forst.

Euodia hortensis, Forst. Char. Gen. p. 14, t. 7; DC. Prodr. 1, p. 724; Adr. Juss.
Mem. Rutae. t. 22, f. 28.

Fagara Euodia, Tain, f. Suppl. p. 125; G. Forst. Prodr. Fl. Ins. Austr. p. 10.
Zanthozylum varians, Benth. in Hook. Lond. Jour. Bot. 2, p. 215.

Has. Tutuila, one of the Navigators’ or Samoan Islands: in culti-
vated ground. Also at Rewa, Viti-levu, one of the Feejee Islands,
in similar situations. (Feejee Islands, Barclay.)—It appears also to
have been seen at Tongatabu, where it was found by Forster.

Like the following (of which specimens were also collected by
Forster), the leaves are occasionally simple, especially the upper ones.
Dr. Pickering, in his manuscript notes, remarks that they are “said
to be used for scenting cocoa-nut oil.”

2. Evopia toneiroiia, A. Rich.
Luodia longifolia, A. Rich. Bot. Voy. Astrolab. 2, p. 61, t. 22.
Has. Savaii, Navigators’ or Samoan Islands: also Feejee Islands.

The leaves are either simple, trifoliolate, or bifoliolate, elongated-
lanceolate or linear, from 6 to 12 inches in length, and half an inch
to an inch in width, with undulate margins. The flowers are in long

and slender panicles, more simple and slender than those of E. horten-
sis, to which it is closely related.

3. HKvopta? prupAcga, Labill.
Huodia drupacea, Labill. Sert. Austr.-Cal. p. 78, t. 74?

Has. Muthuata, Feejee Islands.

The specimens as to foliage and inflorescence exactly accord with

RUTACES. 399

Labillardiere’s figure; but they bear only half-grown flower-buds,
sufficing however to show that the flowers are tetrandrous and appa-
rently perfect. What Labillardiere takes for the putamen of the
drupe, is the testa of the seed, as is evident from his figure-—The
leaves of our plant are occasionally simple.

14. ACRONYCHIA, Foret.

This genus, of which our collection considerably extends the
number of the species, differs from Huodia in the double number of
stamens and the gynobasic disk; from Melicope in the valvate esti-
vation of the corolla; and from both in the complete union of the
carpels into a compound ovary, ripening into. a drupaceous or fleshy-
capsular fruit, which when dehiscent is loculicidal. The flowers are
polygamous in all our species, and probably in the whole genus, as in
Fuodia, Ptelea, &c.

x Diversifolie.
1. ACRONYCHIA HETEROPHYLLA, Sp. Nov. (Tab. 32.)

A, glabra; foliis membranaceis amplis, .aliis integris alus trifoliolatis,
foliolisque ovalibus oblongisve utrinque acutis; cymis multifloris pani-
culatis compositis breviter pedunculatis ; floribus polygamis, foemineis
stylo gracili, stigmate subquadrilobo, gynophoro cano-pubescente ova-
rio fere equilongo; fructu subgloboso loculicide quadrivalvr, loculis
dispermis.

Has. Tutuila, one of the Samoan or Navigators’ Islands.

Shrub, or small tree, glabrous throughout, with much the habit of
Melicope ternata; but the leaves more ample, some of them simple and
entire, while others are ¢rifoliolate on the same branches, membra-
naceous in texture, pale and dull, opposite; the simple leaves oval or
oval-oblong, more or less acute at both ends, or sometimes obtuse,
loosely feather-veined, thickly punctate with rather opaque dots, 3 to

84

334 PHANEROGAMIA.

5 inches in length, 2 inches or more in width, on petioles half an inch
or an inch long, which are not perceptibly articulated with the blade:
the compound leaves have similar digitate leaflets, only more oblong
and commonly larger, from 4 to 7 inches in length, often cuneate-
narrowed, or the lateral ones oblique at the base, on partial petioles
of 4 to 12 lines long; the common petiole often 3 inches long.
Peduneles axillary, sometimes in pairs, shorter than the petiole, bear-
ing a thyrsoid or paniculate, many-flowered, compound cyme, which
seldom exceeds the petioles. Bracts and bractlets minute. Pedicels
fascicled, a line and a half long. lowers diceciously or perhaps
moneeciously polygamous; the male with a small and sterile ovary ;
the female with short and effete stamens. Flower-buds oblong, about
the length of the pedicels. Calyx deeply four-cleft, somewhat glan-
dular; the lobes broadly triangular, imbricated in estivation, persis-
tent. Corolla thrice the length of the calyx; the petals narrowly
oblong, obtuse, with a minute inflexed point, valvate in sstivation,
deciduous. Stamens 8, in the male flowers as long as the petals:
filaments subulate-filiform, glabrous, somewhat glandular-dotted ; the
4 alternate ones rather shorter than the others: anthers oval, cordate,
or emarginate at both ends, two-celled. Disk or gynophore nearly as
tall as the ovary it supports, but narrower, densely canescent or tomen-
tose. Ovary (of the fertile flowers) globular, glabrous, four-celled,
surmounted by a slender style, which is commonly twice or thrice the
length of the ovary: stigma somewhat four-lobed, thick. The sterile
pistil bears a short and more or less rudimentary style and stigma.
Ovules 2 in each cell, between amphitropous and anatropous, pendu-
lous; the micropyle superior. uit capsular, 3 lines in diameter,
globular, slightly four-lobed (the carpels united quite to the summit),
four-celled, loculicidally four-valved from the apex: the chartaceous
endocarp separating from the thin exocarp; the cells two-seeded, rarely
by abortion one-seeded. Seeds hanging by the slender funiculus after
escaping from the cell, oval, scarcely 2 lines long, amphitropous, with
a small hilum: the testa very black, smooth and shining, somewhat
drupaceous, as in Zanthoxylum. Embryo as long as the albumen:
cotyledons oval, flat: radicle short, slightly bent.

PLATE 32.—ACRONYCHIA DIVERSIFOLIA: a branch, in flower and
fruit, of the natural size. Fig. 1. An opening sterile flower. 2. An

(3)

expanded sterile flower. 3. A posterivr, and, 4, anterior view of a

RUTACEAE, 335

stamen from Fig. 1. 5. Section of a sterile flower, showing the abor-
tive pistil. 6. A fertile flower. 7. Vertical section of the same,
showing the gynophore, ovules, &c. 8. Capsule opening. 9. The
same, more dehiscent. 10. A seed, with its funiculus. 11. Vertical
section of a seed parallel with the cotyledons, showing the whole
embryo. 12. Transverse section of a seed.—The details variously
magnified.

* * Simplicifolie.
2. ACRONYCHIA PETIOLARIS, Sp. Nov. (Tab. 33.)

A, glabra; foliis chartaceis oblongis utrinque acutis longe petiolatis ;
pedunculis brevissimis paucifloris; fructu ovoideo apiculato suberoso-
lignescente tomentuloso demum quadrivalvi? basi calyce petalisque
persistentibus extus cinerers stipato,

Has. Muthuata, one of the smaller Feejee Islands.

A tree or shrub (the size unknown), glabrous throughout. Leaves
chartaceous in texture, pellucid-punctate, opposite, oblong or elongated-
oblong, acute or acutish at both ends, entire, or the margins somewhat
undulate, 3 to 5 inches long, 14 to 2 inches wide, bright green and
rather shining above, scarcely pale underneath, inconspicuously
feather-veined and reticulated, long-petioled (the petioles about 12
inches in length). Peduneles axillary, very short, 4 or 5 lines long in
fruit, apparently simple and very /ew-jlowered. Flowers not seen,
except the persistent calyx, corolla, and filaments under the base of
the fruit, to which they are closely appressed. Both the calyx and
corolla are cinereous externally with a close pubescence. The former
is four-lobed; the lobes short and obtuse: the latter of 4 oval or
oblong petals of thickish texture, 2 lines long, twice the length of the
calyx. Persistent filaments 8, flattened, dilated-subulate, ciliate below
the middle, occasionally still bearing small two-celled anthers. Fruit
ovoid, slightly pointed, 6 to 8 lines long, minutely tomentose or canes-
cent, four-celled, with either one or two seeds in each cell; the very
thick walls between corky and woody in texture, but the rind probably
fleshy, so as to be more or less drupaceous when fresh: the endocarp

336 PHANEROGAMIA.

chartaceous. In one specimen the apex has separated into four
points, so that it is perhaps four-valved at complete maturity : if so it
would seem to be septicidal. The fruit is closely sessile in the calyx;
but its base is marked with the vestige of a very short gynophore,
like that of A. Richii. Seeds amphitropous, large, when solitary 3
lines long and filling the cell, somewhat club-shaped, the lower, cha-
lazal end being prolonged and tapering; when two in each cell, the
upper is ascending and the lower, probably imperfect one, is descend-
ing: testa rather drupaceous, with a thin and somewhat dull exterior
pellicle, and a very thick and crustaceous inner portion. Nucleus
small in proportion to the size of the seed, consisting of a sparing
fleshy albumen, and a straight embryo of nearly its length. Cotyle-
dons oblong. MRadicle superior.

This is doubtless a true congener of Acronychia levis, Forst., of
New Caledonia, and of A. Endlicheri, Schott (Vepris simplicifolia,
Endl.) of Norfolk Island. So far as can be judged from the brief
character, the species bears more resemblance to Blume’s A. arborea
than to any other.

PuatE 33, A.—ACRONYCHIA PETIOLARIS: in fruit. Fig. 1. A mar-
cescent persistent flower, taken from the base of a young fruit, en-
larged. 2. Fruit, of the natural size, with the persistent remains of
the flower at its base. 3. Transverse section of the same. 4. Ver-
tical section of the same. 5. A seed, magnified. 6. Vertical section
of the same, showing the embryo, &c.

3. ACRONYCHIA Ricuu, Sp. Nov. (Tab. 33.)

A, glabra; foliis subcoriaceis oblongis sew oblanceolato-oblongis obtusius-
culis basi attenuatis; pedunculis brevibus paucifloris; floribus sub-
polygamis ; stylo ovario equilongo vel breviore ; stigmate quadrilobo ;
gynophoro pubescente ; fructu depresso-globuloso loculicide quadrivalvi,
loculis monospermis.

Has. Samoan or Navigators’ Islands.

Shrub or tree (the size not recorded), glabrous, except a minute

RUTACEZ., 337

pubescence on young shoots and inflorescence. Leaves rather coria-
ceous, opposite, crowded on the branches, varying from oblong-lanceo-
late to oblong, tapering at the base into a short petiole, more or less
obtuse at the apex, entire, sparsely feather-veined, dull, inconspi-
cuously but thickly pellucid-punctate, 2 or 22 inches long, from 7 to
12 lines wide. Peduncles axillary, longer or shorter than the petiole,
sometimes very short (varying from 2 to 5 lines in length), cymosely
Jfew-lowered. Pedicels a line long, bracteolate at the base. Flowers
a line and a half long, polygamous; some flowers having large anthers
on filaments as long as the petals, and a nearly sessile stigma, while
others have a longer style and shorter stamens, with smaller and
apparently effete anthers. Calyx puberulent, glandular, four-cleft;
the lobes triangular. Petals 4, much longer than the calyx, valvate
in estivation, oblong-lanceolate, with a minute inflexed point, gla-
brous, spreading, deciduous. Stamens 8: filaments glandular-dotted,
either filiform and as long as the petals, and with didymous fertile
anthers, or only as long as the ovary, and mostly with effete anthers.
Disk or gynobase pubescent, four-lobed. Ovary globular, glabrous,
four-celled, with 2 amphitropous-ascending ovules in each cell: style
rather short, sometimes as long as the ovary, in some truly fertile
flowers considerably shorter, terminated by a thickened /four-lobed
stigma. The staminate flowers examined have a similar, although
smaller ovary, which contains one or two apparently well-formed
ovules in each cell, and is terminated by a small, four-lobed stigma.
Fruit globular, depressed, not in the least pointed, but rather umbili-
cate at the summit (the style wholly deciduous), 2 lines in diameter,
four-celled, four-seeded (or by abortion two-seeded), loculicidally four-
valved from the summit: exocarp rather thin, coriaceous, and glan-
dular, as in Zanthoxylum: endocarp thin, chartaceous. Seed solitary
in each cell, ovoid, with a crustaceous, black and shining testa, and a
linear lateral hilum; the micropyle superior. Embryo straight, in
the axis of fleshy albumen: cotyledons oval, flat: radicle very short.

Piate 33, B—AcronycoutA Ricuu. Fig. 1. Branchlet of a fertile
plant, in flower. 2. Branchlet, in fruit. 3. Bud of a sterile? flower.
4. A petal, from the same. 5. The same, with the petals removed.
6. Vertical section through the ovary of the same. 7. Fertile flower,
from which the petals, &c., have fallen. 8. Vertical section of a

similar flower, showing also two of the sterile? stamens. 9. A fruit,
85

338 PHANEROGAMIA.

with the sutures opening. 10. Vertical section of the same, and of
one of the seeds, showing the embryo. 11. A seed detached.—All
the analyses more or less magnified.

4, ACRONYCHIA RETUSA, Sp. Nov. (Tab. 34.)

A. subglabra; foliis membranaceis obovatis ovalibusve retusis basi sub-
acutis ; pedunculis petiolo vix longioribus; cymis plurifloris ; floribus
polygamis ; stigmate peltato-quadrilobo subsessili.

Has. Samoan or Navigators’ Islands.

A shrub or tree (the size not recorded), glabrous, except a slight
pubescence on the young shoots, &. Leaves thin and membranaceous,
opposite, from 2 to 32 inches long, obovate or oval, with a more or less
acute base, retuse at the apex, entire, or obscurely undulate, sparingly
feather-veined, pale green and dull, nearly of the same hue both sides,
thickly punctate with pellucid dots; the base obscurely articulated
with the apex of the petiole, which is from 5 to 8 lines long. Pedun-
cles axillary, equalling or slightly exceeding the petioles in length,
bearing rather numerous flowers in a small cyme. Bracts and bractlets
minute, subulate. Pedicels only a line long. Flowers diceciously
polygamous; the male (with very small flower-buds) having a sterile
ovary, with an effete and less thickened stigma; the female (flower-
buds a line and a half long) having effete anthers: otherwise nearly
similar. Calyx glandular-dotted, four-cleft to the middle; the short
lobes ovate and obtuse, or in the male flowers acutish. Petals 4,
white or whitish, valvate in estivation, with a minute inflexed point
at the apex, ovate-oblong, glandular-dotted, glabrous or nearly so.
Stamens 8: filaments short, the alternate ones shorter, glabrous but
obscurely glandular, sparingly ciliate towards the base: anthers cor-
date-didymous. Ovary ovoid, obscurely pubescent, at length covered
with pulverulent glands, raised on a short and thick disk or gynobase,
four-celled, crowned with a large and purple, somewhat peliate and
Sour-lobed, subsessile stigma. Ovules 2 in each cell, amphitropous,
ascending. Fruit unknown.

This species is evidently a congener of Acronychia Endlicheri,

RUTACES. 339

Schott (the Vepris simplicifolia of Endlicher), and doubtless also of
the original A. levis, notwithstanding the want of a style.

Puiate 34, A.—AcRONYCHIA RETUSA. Fig. 1. Branchlet of the sterile
plant, with flower-buds. 2. Branchlet of the fertile plant. 3. Sterile
flower-bud and pedicel. 4. Vertical section of the same. 5. A petal,
from the same. 6. Stamens, from the same. 7. Sterile pistil. 8.
Transverse section of the same. 9. Stamens and pistil of a fertile
flower (from Fig. 2). 10. Vertical section of the same.—The analyses
variously magnified.

15. PELEA, Nov. Gen.

Flores polygami. Calyx quadripartitus, cestivatione imbricatus, cito
deciduus. Petala 4, cestivatione valvata, mox decidua. Stamina 8:
filamenta subulata vel complanata, fl. fert. breviora sepiusque antheris
(sagittatis) cassis donata. Discus hypogynus brevissimus, integer, ser
octocrenulatus. Ovariuwm quadriloculare (leculis petalis oppositis),
guadrilobum, sceepius umbilicatum: stylus centralis: stigma quadri-
lobum. Ovula in loculis gemina, hemitropa, adscendentia. Capsulu
guadripartita, stellariformis (coccis divaricatis), loculicida; endo-
carpio chartaceo ab exocarpio coriaceo seu lignescente solubili. Semina
in loculis scepissime bina, ovoidea, testa nitente drupacea. Limbryo
intra albumen curnosum rectus; cotyledonibus ovalibus; radicula
supera.—Arbores Sandwicenses (necnon Sumoenses), inermes, odoratee ;
foliis simplicibus integerrimis oppositis sew verticillatis cortacers punc-
tatis venosissimis, venis in venulam intramarginalem pl. m. confluen-
tibus ; floribus axillaribus.

A genus of several species of trees, natives of the Sandwich Islands,
one of which has been inadvertently taken for a Clusia, while
another, known many years since by the fruit alone, was referred by
Adrien de Jussieu and Gaudichaud to the South American genus
Brunellia, Ruiz & Pav. But all the species of that obscure genus are
apetalous, and have a persistent calyx, with a valvate estivation, as
well as perfectly distinct pistils, with subulate styles. Their leaves
also are said not to be punctate, and to be furnished with small

340 PHANEROGAMIA.

stipules, of which there is no trace in our plants. The true affinity
of our genus is manifestly with Acronychia and Melicope; from both
of which it differs in habit, in the deciduous calyx, &c.; from the
former likewise in the deeply four-lobed ovary and fruit; from the
latter in the valvate sestivation of the corolla, and in the union of the
carpels by their bases to a greater or less extent. The last-named
character, however, may be invalidated by an ambiguous species from
the Samoan Islands, which I am obliged for the present to append to
this genus, since the valvate sestivation of the corolla excludes it from
Melicope, and the separation of the ovaries, from Acronychia.*

As this is a characteristic and important genus of the Hawaiian
Islands, and especially of the forest-clad sides of the volcanoes of her
fabled abode, I dedicate it to the Hawaiian goddess Pele.

* Species Sandwicenses, vere.
I, PELEA cLust#roLiA. (Tab. 35.)

P. glaberrima ; foliis ter-quaternatim verticillatis vel oppositis obovatis

* The diagnoses of these allied genera may be expressed as follows :—

1. Astivatio corolle valvaris. Flores polygami.

1. EUODIA. Stamina 4. Ovaria 4, discreta, stylis tantum connexa, in disco urceo-
lari inclusa. Folliculi bivalves.

2. ACRONYCHIA. Stamina 8. Ovaria in unicum quadriloculare omnino coalita.
Fructus aut indehiscens, aut loculicide quadrivalvis.

3. PELEA. Stamina 8. Ovaria in unicum quadrilobum coalita. Capsula quadri-
partita, stellata, tetracocea.

2. ™Aistivatio corolle imbricativa. Ovaria discreta.

4. MELICOPE. Flores hermaphroditi.(?) Stamina 8. Ovaria 4, stylis coalita, disco
brevissimo insidentia. Cocci 4, bivalves.

5. PITAVIA. Flores polygamo-dioici. Stamina 8. Ovaria 4, gynophoro satis elon-
gato insidentia. Styli superne coaliti. Baccse indehiscens.

6. ZANTHOXYLUM. Flores dioici. Stamina 3-5. Ovaria 1-5, gynophoro brevi
insidentia. Folliculi carnosi semibivalves.

RUTACES 34]

petiolatis ; floribus in axillis fasciculatis, brevissime pedicellatis ; calycis
lobis ovatis membranaceis petalis plus dimidio brevioribus ; stylo ovario
glabro longiore; capsula obtuse quadriloba.

Clusia sessilis, Hook. & Arn. Bot. Beech. Voy. p. 80, non Forst.

Var. @. capsula multum majore, valvis crassis lignescentibus. (Tas.
34, fig. 10.)

Var. y. folwis tenuioribus cuneato-oblanceolatis ; capsula parvula.

Has. Sandwich Islands: mountains behind Honolulu, Oahu (where
it was also gathered by Macrae, Lay and Collie, Gaudichaud, &c.) :
Hawaii, in the district of Puna, and on the margins of the Bullock
Plains on Mouna Kea. Var. @. Kaala Mountains, Oahu. Var. y.
Mouna Kea, Hawaii.

Tree about 25 feet high, very glabrous, except the nascent leaves in
the naked leaf-buds. Leaves verticillate in fours, or sometimes in threes,
or simply opposite, especially the lower ones, coriaceous, obovate or
obovate-oblong, obtuse or retuse, more or less acute at the base, entire,
pale, with the strong midrib rather dark underneath, either dull or
rather shining, punctate, but the dots seldom translucent, on account
of the thickness and opacity of the full-grown leaf, closely and
copiously feather-veined; the primary veins parallel and somewhat
transverse, reticulated with intermediate veinlets, and at their extre-
mity confluent into a distinct intramarginal vein: they are crowded on
the branches, from 2 to 3 inches long, 8 to 20 lines wide, distinctly
petioled ; the petioles varying from two lines to an inch in length.
Their texture and ordinary hue in the dried specimens, as well as
the numerous straight and parallel veins, give the foliage slightly the
appearance of that of a Clusia. Flowers in auillary subsessile clusters,
or sessile cymules; the very short pedicels becoming 2 to 4 lines long
in fruit, naked. Calyx four-parted ; the lobes ovate, membranaceous, of
nearly the same yellowish hue as the corolla, obtuse, imbricated in
cestivation, scarcely a line and a half long, deciduous. Petals 4, valvate
in sstivation, with an inflexed small appendage at the thickened
apex, in the manner of Acronychia, ovate in the bud, becoming ovate-

lanceolate and fully 3 lines in length, honey-yellow or wax-coloured,
86

342 PHANEROGAMIA.

thickish, glandular-dotted, deciduous. I have seen only fertile flowers,
which appear as if hermaphrodite; but the anthers contain no good
pollen. Stamens 8, shorter than the petals: filaments dilated, linear-
lanceolate, glabrous: anthers sagittate, mucronate, adnate-introrse.
Hypogynous disk gynobasic, very short, annular, entire. Ovary gla-
brous, depressed-globular, four-lobed, four-celled, the four carpels united
to the umbilicate apex; which bears a columnar central style that
exceeds the ovary in length: stigmas 4, thick and short. Ovules 2 in
each cell, probably collateral at an early period, ascending, hemi-
tvopous; the micropyle superior. Capsule tetracoccous, rather deeply
Jour-lobed, with the lobes obtuse and rounded, diverging so as to be
star-shaped, half an inch or rather more in diameter, loculicidal, so
that the cocci become two-valved; the exocarp thick and coriaceous
when dry, somewhat wrinkled or reticulated; the thin and papery
endocarp separating after dehiscence, glabrous within. Seeds 2 in
each cell, or sometimes by abortion solitary, black and shining like
those of Zanthoxylum, ovoid, 12 to 2 lines long; the integument
drupaceous ; the exterior pellicle being thin and fragile, and connected
by a sparing dry pulp with the inner very thick and crustaceous
portion: hilum ventral, linear. Embryo straight, nearly as long as
the fleshy albumen: cotyledons oval, flat: radicle superior, slender.

The variety @., of which there is only an imperfect specimen, with
ripe fruit, appears to differ merely in the much lurger capsule (an
inch in diameter), with very thick and lignescent walls.

The variety y., from Mouna Kea, is distinguished from the other
Hawaiian specimens (of which the foliage is mostly greener than in
those from Oahu) only by the longer and narrower leaves (3 or 4
inches in length), with a more tapering base, and of thinner texture.
It probably grew in a closer forest.

All the specimens I have seen are in fruit, save one from Macrae’s
collection; from which the (fertile) flowers are here described and
illustrated.

PLATE 50,—PELEA CLUSI@FOLIA: in flower and in fruit. Fig. 1, A
flower-bud. 2. A petal, from the same, seen from within. 3. A
stamen, from the same, seen from within. 4. An expanded (fertile)
flower. 5. Vertical section of the same, showing the ovules, &c. 6.
Pistil, showing the disk, &. 7. A seed. 8. Vertical section of the
same. 9. Embryo, detached.—The analyses all variously magnified.

RUTACE £&. 843

2. PELEA AURICULZFOLIA, Sp. Nov. (Tab. 36.)

P. glabra; foliis ternis oblongo-spathulatis bast auriculatis sessilibus ;
floribus ‘fasciculatis ad azxillas foliorum delapsorum secus caulem
virgatum brevissime pedicellatis ; capsula quadripartita.

Has. Forests of Hawaii, on the flank of Mouna Kea.

The specimen, taken from “an upright, nearly simple shrub,” bears
only a little fruit, and a few fertile ovaries, from which the perianth,
stamens, &c., have fallen. The virgate stem is very leafy above; and
the flowers have been produced lower down, in small fascicles, from
the axils of earlier leaves, now fallen. Plant glabrous throughout.
Leaves verticillate in threes, coriaceous, pale, oblong-spatulate, obtuse,
auriculate at the base, sessile, from 3 to 5 inches long, veined and
dotted nearly as in the preceding species; the midrib salient under-
neath. Ovary more deeply lobed than in P. clusie/folia, being united
only at the base: the style has mostly fallen. Capsule deeply fow-
parted ; the cocci oval-oblong, otherwise similar, as apparently are the
seeds, to those of the foregoing species.

PLate 36.—PELEA AURICULAHFOLIA: of the natural size. Fig. 1.
Ovary, with the receptacle, &., magnified. 2. Vertical section of the
same, more magnified, showing the ovules, Xc.

3. PELEA OBLONGIFOLIA, Sp. Nov.

P. glabra; foliis oppositis oblongis sew ovalibus petiolatis ; pedunculis
(A. fert.) in axiilis solitariis uni—bifloris petiolum adequantibus ; cap-
sula quadriloba, coceis subcarmatis.

Has. Hawaii, in the district of Puna, at an elevation of 800 or
1,000 feet; near the crater Lua Pele: also in the forest on the margin
of Bullock Plains, Mouna Kea. Kauai? (foliage only). Mountains
behind Honolulu, Oahu.

To this species I refer a number of individuals from different
habitats, none of them with flowers, and few with mature fruit.

344 PHANEROGAMIA.

Some of them approach P. Sandwicensis in appearance, but are gla-
brous (the nascent foliage only minutely pubescent), the (fertile)
peduncles are one-flowered, or rarely two-flowered ; the glabrous cocet
appear to be more or less carinate at the sutures, and their endocarp is
elabrous inside. Leaves simply opposite in all the specimens, oblong,
or occasionally oval, sometimes rather contracted towards the base,
more commonly obtuse or even retuse at both ends, coriaceous or
ehartaceo-coriaceous in texture, somewhat shining, and of the same
green hue both sides, punctate, copiously and rather conspicuously
feather-veined, but minutely reticulated; the primary veins almost
transverse, more or less confluent into a submarginal vein: they vary
from 2 to 4 inches in length; their petioles are commonly half or three-
quarters of an inch long. Peduneles (of the fertile flowers only seen)
solitary in the axils of the leaves, usually equalling the petioles, some-
times longer, bearing one or two pairs of small subulate, bracts. Fer-
tilized ovary glabrous, four-lobed, girt at the base by an eight-crenulate
disk. Style not seen. Capsule deeply four-lobed, an inch or two-
thirds of an inch in diameter; the cocci transversely oblong and di-
varicate, or in some specimens decurved, so that the vertical section
would be kidney-shaped: in these the carpels appear to be more
united, and their long diameter is vertical. Seed-coat drupaceous,
shining.

4. PELEA ROTUNDIFOLIA, Sp. Nov. (Tab. 37.)

P. glabra; foliis orbiculatis sessilibus valde reticulatis; floribus cymu-
losis ; calycis lobis ovatis petala subequantibus ; stylo ovario puberulo
breviore.

Has. Oahu, on mountains behind Honolulu.

Only a single specimen of this well-marked species occurs in the
collection; and that bears a cluster of unexpanded, apparently her-
maphrodite flowers. Leaves orbicular or nearly so, about 23 inches in
length and breadth, sessile, coriaceous, rather shining, and of the same
hue both sides, very veiny and reticulated, especially underneath, gla-
brous, rather obscurely punctate. Flowers several, in a small and
short-peduncled, somewhat racemose cyme in the axils of the upper-

RUTACE &. 345

most leaves. Bracts and bractlets opposite, minute, ovate-subulate,
puberulent. Pedicels rather longer than the flower buds, which are
only a line and a half in length. Calyx deeply four-cleft, imbricated
in estivation; the ovate lobes nearly equalling the petals when they
first open. The latter are triangular-ovate, valvate in estivation,
with an inflexed tip. Stamens 8: filaments subulate, thick, not
longer than the ovary: anthers sagittate, mucronulate, introrsely
adnate; the connective, especially of the alternate stamens, glandular-
dotted. Hypogynous disk small, annular, eight-crenulate. Ovary
puberulent, subglobose, of 4 carpels, which are united at the base, and
also by the short style: stigmas 4, capitate. Ovules 2 in each cell,
hemitropous, ascending, or the upper one nearly centripetal. Fruit
unknown.

Puate 87, A.—PELEA ROTUNDIFOLIA: summit of a flowering branch.
Fig. 1. A flower-bud, with its bracteolate pedicel. 2. The flower,
opened. 3. A stamen, seen externally. 4. The same, seen from
within. 5. Pistil, with its disk. 6. Vertical section of the same.—
All the analyses more or less magnified.

5. Perea SANDWIcENSIS. (Tab. 37.)

P. ramis novellis cum inflorescentia cymosa 3-9-flora hirsutulo-tomen-
tosis; foliis oppositis ovalibus oblongisve supra glaberrimis subtus pree-
sertim valde reticulatis petiolatis; calycis lobis ovato-rotundis petalis
dimidio brevioribus; stylo fl. fert. ovario longiore; capsula profunde
quadriloba tomentulosa, coccis ovalibus haud carinatis, endocarpio
intus puberulo !

Brunellia Sandwicensis, Gaud. Bot. Freye. Voy. p. 98, sine deser.; Hook. & Arn.
Bot. Beech. Voy. p. 80.

Has. Mountains behind Honolulu, Oahu (where it was also ga-
thered by Gaudichaud, Macrae, Lay and Collie). Mountains of the
western part of Maui.

The specimens in the collection being very imperfect ones and
without ripe fruit, our principal figure and character are taken from a
specimen of the fertile plant, gathered by Mr. Macrae. The species

87

346 PHANEROGAMIA.

is apparently a tree, of considerable size, glabrous at maturity, or
nearly so; but the new branchlets, inflorescence, &c., tomentose with a
rather hirsute pubescence. Leaves opposite, oval or oblong, 2% to 44
inches long by 12 to 22 in width, thick and coriaceous, very glabrous
above, more or less puberulent or glabrate beneath, and when young
pubescent on the salient and thick midrib, very veiny and reticulated,
the reticulations especially conspicuous underneath, punctate; the stout
and lignescent petioles 7 to 14 lines long. Cymes axillary, short-
peduncled, 3—9-flowered : pedicels short, annulate by the broad scars of
the ovate-subulate bracts. Flowers a line and a half long. Calyz-
lobes round-ovate, half the length of the petals. Stamens in the male
flowers nearly as long as the petals, with sagittate anthers; in the
fertile flowers examined, with shorter filaments and rather smaller
anthers. Disk crenulately eight-lobed, nearly glabrous. Style in the
fertile flowers slender, longer than the tomentose and deeply four-lobed
ovary, early deciduous; in the sterile flowers shorter than the ovary.
which is similar but smaller: stigma four-lobed. Ovules, &c., as in
the other species. Capsule finely tomentose, even at maturity scarcely
above half an inch in diameter, deeply four-lobed, star-shaped; the
cocci oval, not flattened laterally nor with sharp or carinate sutures,
The papery endocarp is finely pubescent inside, as was noticed by
Adrien de Jussieu, in his Memoir on Rutacee.

PLATE 37, B.—PELEA SANDWICENSIS: summit of a fruiting branch,
of the natural size. Fig. 1. A hermaphrodite-sterile flower. 2. A
hermaphrodite-fertile flower. 3. Vertical section of a capsule—The
details enlarged.

6. PELEA VOLCANICA, Sp. Nov. (Tab. 38.) .

P. ramis junioribus petiolis et inflorescentia cymuloso-paniculata hirsuto-
tomentosis; foliis oppositis ovalibus longe petiolatis majusculis gla-
bratis; calycis lobis ovatis petalis plus dimidio brevioribus; stylo
gracily ovario tomentoso cequilongo; capsula (sesquipollicari) glabra
quadriloba, coccis recurvis carinatis.

Has. In forests, on Mouna Kea, near the Bullock Plains, &c.,
Hawaii.

RUTACE &. 347

“A tree, 40 feet high; the trunk a foot and a half in diameter :”
young branches with the petioles, &e., tomentose with a hirsute pubescence,
which disappears with age. Leaves opposite, oval, 4 to 6 inches long
and 2 to 34 wide, mostly obtuse at both ends, coriaceous, but not so
thick nor so strikingly reticulated as in the foregoing species, gla-
brous above, glabrate beneath, the strong and salient midrib more or
less pubescent, punctate with more or less translucent dots, not
shining: the petioles long (12 to 2 inches) and stout, apparently
lignescent. Peduncles axillary and solitary, half an inch to an inch
and a half long, tomentose-pubescent, branching into a trichotomous,
loosely several-flowered, paniculate cyme. Bracts subulate, caducous.
Pedicels stout, a line or two in length. The apparently hermaph-
rodite flowers only are known. Lobes of the calyx ovate, mucronu-
late, canescently pubescent, imbricated in sestivation, early caducous,
as in the preceding species, less than half the length of the ovate-
lanceolate petals. The latter are about 3 lines long, glabrous, valvate
in ewstivation, early deciduous. Stamens 8, nearly as long as the
petals, or in some flowers shorter (and imperfect ?): filaments subu-
late: anthers oblong, retuse, introrse. Disk short and thick, some-
what crenulate, minutely tomentose. Ovary tomentose, four-lobed
from the summit, four-celled, its umbilicate apex bearing a slender
style as long as the ovary itself, from which it early falls away entire.
Stigma four-lobed; the lobes oblong. In our fruiting specimen only
one fruit matures on each cyme, which therefore appears as if borne
on a simple and articulated axillary peduncle, Capsule larger than
in any other known species, being an inch and a half in diameter when
fully formed ; but one or two, and sometimes three of the lobes or cocci
are often abortive or infertile: cocci glabrous, somewhat lignescent,
united in the axis, but recurved, so-that a vertical section would be
kidney-shaped, more or less flattened laterally, acute or carinate at the
sutures; the papery endocarp glabrous inside. Seed ovoid, with a black
and shining, drupaceous testa. Embryo nearly the length and breadth
of the albumen. Cotyledons oval: radicle very short, superior.

Of most of the species of this genus here characterized complete
materials are still needed. There are indications of still others in the
Sandwich Islands, unless these plants vary greatly in their forms,
which (as is justly remarked by Bory and Dr. Hooker), is apt to be
the case in insular floras.

$48 PHANEROGAMIA.

Puate 38.—PELEA VOLCANICA: branches, of the natural size, in
flower and in fruit. Fig. 1. Diagram of the flower. 2. Expanded
hermaphrodite flower. 3. Vertical section of the same. 4. A stamen.
5. Fructified ovary, with the disk, pedicel, and bractlets. 6. Vertical
section of the seed through the hilum.—The details variously mag-
nified.

* * Species Samoensis, dubia.
7. PELEA? LucipaA, Sp. Nov. (Tab. 34.)

P.? glaberrima,; foliis oppositis ovalibus seu ovali-oblongis subcoriaceis
supra lucidis creberrime penninerviis; cymis longe pedunculatis
multifloris; ovarus fere discretis stylo connexis.

Has. Mountains of Tutuila, one of the Samoan or Navigators’
Islands.

A shrub or tree, glabrous throughout. Leaves opposite, approxi-
mate, oval and slightly pointed in the male specimen, oval-oblong and
more acumiate in the female specimen, obtuse or acutish at the base,
entire, 3 or 4 inches long, about 2 inches wide, rather coriaceous in
texture, bright green and shining above, slightly pale underneath,
thickly pellucid-punctate; the strong midrib sending off along its
whole length a great number of transverse and close, pinnate veins,
which are very conspicuous on the upper surface, reticulated towards
the margins. Petioles about half an inch long. Peduncles elongated,
one or two inches in length, bearing a small, loosely flowered cyme or
panicle. Bracts and bractlets minute, subulate. Pedicels 14 to 2
lines long, fascicled, articulated. Flowers diceciously polygamous;
the male having a sterile gyneeium; the female more or less effete
anthers. Calyx sprinkled with glands, deeply four-cleft, deciduous ;
the lobes ovate, obtuse, imbricated in eestivation. Petals 4, deciduous,
valvate in eestivation, lanceolate-oblong, or ovate-oblong in the male
flowers, acutish, with a minute inflexed point, glandular and dotted.
Stamens 8, shorter than the petals, the alternate ones still shorter:
filaments glabrous, glandular; in the effete stamens of the fertile
flowers slightly ciliate. Ovaries 4, glabrous, glandular, supported on

RUTACES, 349

a small and short, somewhat eight-lobed, gynobasic disk, closely con-
nivent, as if forming a four-lobed compound ovary, but distinct; their
styles united into one, but at length separable? Stigma four-lobed.
The styles are either longer or shorter than the ovary, even in dif
ferent flowers of the same cyme. Ovules geminate from near the
base of the cell, ascending, the micropyle superior. Ovary of the
male flowers four-lobed, crowned with a small sessile stigma, and con-
taining only abortive ovules. Fruit not seen.

We have only imperfect and scanty materials for characterizing
this plant, the genus of which must for the present remain uncertain.
Apparently it differs from Melicope only in the valvate sstivation of
the corolla and the deciduous calyx, and from <Acronychia in the
uncombined ovaries. Since it is nearly in these same particulars that
FPelea is distinguished from these two genera, the plant may most
naturally be referred to the present genus, with which it pretty well
accords in habit.

Piate 34, B—PrLEa? tucipA. Fig. 1. Branch of the sterile plant.
2. Portion of a branch of the fertile plant. 38. Flower-bud and pedi-
cel of the sterile plant. 4. Expanded flower of the same. 5. Calyx
and abortive pistil of the same. 6. Flower-bud, with the lobes of the
calyx and petals removed. 7. Stamens of the same. 8. A fertile
flower (from fig. 2). 9. Vertical section of the same. 10. The pistil
shortly after anthesis; the ovaries connected only by their common
style-—The details variously magnified.

16. MELICOPE, Forst.

The estivation of the corolla appears not to have been determined
either in Melicope ternata or M. simplex. Some flower-buds of the two
species, kindly communicated by Sir William Hooker for the purpose,
enable me to state that the wstivation is imbricative. In M. ternata
the thin edges of the petals slightly but distinctly overlap: in JL
simplex they are more strongly imbricated. There are some indica-
tions that the flowers are polygamous, rather than truly hermaphro-
dite. The testa of the seed is not “coriaceous,” but crustaceous ; and

the radicle of the embryo is superior, as in all the allied genera.
88

350 PHANEROGAMIA.

I refer to this genus, with some confidence, several simple-leaved
Sandwich Island species, of which our materials are very imperfect.
Two are in fruit; of two others only a flower or two is known; and
the specimens of probably a fifth species afford neither flowers nor
fruit.

* Novo-Zelandice.
1. MELICOPE TERNATA, Forst.

Melicope ternata, Forst. Char. Gen. t. 28, & Prodr. Fl. Ins. Austr. p. 166; DO.
Prodr. 1, p. 723; A. Rich. Voy. Astrolab. 1, p. 293; A. Cunn. Bot. N. Zeal.
in Ann. Nat. Hist. 3, p. 815; Hook. Ic. Pl. t. 603; Hook. f. Fl. N. Zeal. p. 48.

Entoganum levigatum, Banks & Soland. in Herb. Banks; Gertn. Fruct. 1, p. 331,
t. 68.

Has. Tippona, New Zealand. (In fruit only.)

x x Sandwicenses; foliis simplicibus integerrimis, petiolis anguste marginatis.
2. MELICOPE CINEREA, Sp. Nov. (Tab. 39.)

M. foliis lanceolato-oblongis basi rotundato-retusis pallidis utrinque subtus
presertim ramulisque junioribus puberulo-tomentellis; pedunculis
axillaribus paucifloris ; floribus subracemosis extus canescentibus.

Has. Oahu; district of Waianae; in a ravine of the Kaala Moun-
tains.

The specimens, apparently of an arborescent plant (the size not
recorded), were just developing the inflorescence, and furnished only
two well-formed (apparently hermaphrodite) flower-buds for exami-
nation. Branches glabrous, marked with strong approximate leaf-
scars; the younger shoots canescent with a fine and soft tomentum.
Leaves opposite, petioled, narrowly oblong, entire, more or less obtuse,
the rounded base retuse, coriaceous in texture, but rather thin, whitish
or pale, puberulent with a fine and soft tomentum when young, of which
the closer portion is persistent on the lower surface, the upper gla-

RUTACE SA, 351

brate, pellucid-punctate, pinnately veined, with the veinlets slightly
reticulated, 4 or 5 inches long, 14 to 20 lines broad, not articulated
with the petiole; which is 9 or 10 lines long, narrowly channeled
above, and with the sides sharply but narrowly margined. Peduncles
axillary, solitary, short, the inflorescence as far as developed not
longer than the petiole, canescent, bearing pairs of opposite, oblong-
ovate, small bracts, from the axils of which are short pedicels (making
a short raceme) or perhaps branches, developing several flowers.
Calyx four-parted; the ovate sepals, like the petals, canescent externally,
glabrous within, imbricated in estivation. Petals 4, ovate, a little
longer than the calyx, 2 lines in length, thickish, imbricated in eesti-
vation, the narrow and thin edges overlapping. Stamens 8, shorter
than the petals: filaments thickish, subulate, glandular-dotted on the
back; the four opposite the sepals longer than the others: anthers
oblong, emarginate at both ends, two-celled, incumbent. Hypogynous
disk ecrateriform, eight-lobed, the lobes alternate with the stamens.
The gynecium appears at first view to consist of a four-lobed and
four-celled umbilicate ovary, but on examination it is found that the
four rounded ovaries are connected only by means of the short style.
Stigmas 4, capitate. Ovules geminate, superposed, hemitropous ; the
micropyle superior. Fruit wholly unknown.

Prats 39, 4.—Me.icope CINEREA: a branch, of the natural size.
Fig. 1. A flower-bud. 2. Transverse section of the same (diagram).
3. Vertical section through the ovaries, &. 4. Pistil, with the hypo-
gynous disk.— All the details magnified.

3. MELICOPE BARBIGERA, Sp. Nov. (Tab. 39.)

MV. foliis oblongis utrinque obtusis, adultis viridibus supra glabris subtus
secus costum villoso-barbatis; pedunculis axillaribus uni-trifloris pe-
tiolo brevioribus ; folliculis lenticulari-ovoiders.

Has. On the mountains of Kauai, Sandwich Islands.

The specimens, apparently of an arborescent plant, are in fruit, and
with some forming inflorescence; but there are no flower-buds suffi-
ciently developed for examination. Manifestly it is a species nearly
related to the preceding, of which we possess the flowers alone, while

392 PHANEROGAMIA.

these are wanting in this. The young branchlets, the nascent foliage,
and the inflorescence are similarly cinereous-canescent: but the adult
leaves (oblong and obtuse at both ends) are green both sides, wholly
glabrous above, and beneath densely villous-bearded with very long
hairs along the midrib, with some scattered and more deciduous ones
over the lower surface. They are coriaceous, but thin, pellucid-pune-
tate, the dots black by reflected light: in the specimens they are all
curved; the upper surface convex, the lower concave. Petioles
about an inch long, similarly margined, longer than the Sruiting
peduncle, which appears to have been 1-3-flowered at the apex. In
one instance a partly persistent calyx is seen; the sepals ovate,
obtuse, canescent externally. The vestiges of the hypogynous disk
and of barren ovaries accord with those of the preceding species.
The fruit consists of 4, or by abortion fewer, dry, glabrous, coriaceous,
somewhat; veiny and -glandular,-lenticulur-ovoid follicles, which are
divaricate, sessile, half an inch in length, dehiscent down the inner
suture, at length perhaps two-valved, containing one or two black
seeds, which resemble those of JL ternata.

Puate 39, B—MELICOPE BARBIGERA: a branch, in fruit, of the
natural size. Fig. 5. Section of a carpel and seed. 6. A seed.—
Both enlarged.

4, MELICOPE SPATHULATA, Sp. Nov.

M. glabra; foliis elongatis spathulato-oblongis seu oblanceolatis obtusis
basi acutis; pedunculis axillaribus bi-trifloris; sepalis petalisque
glabris orbiculatis.

Has. On the mountains of Kauai, Sandwich Islands.

The single specimen of this species, bearing two flower-buds only,
belongs to an apparently sarmentose shrub. There is, however, no
record of the size or habit of the plant. The slender branch is com-
pressed-angled, glabrous, as is the foliage, &c. Leaves opposite, coria-
ceous, elongated (from 3-to 6 inches long, and one or two broad),
spatulate-oblong, or the smaller ones oblanceolate, obtuse, entire, tapering
to an acute base, pale green and rather dull both sides, inconspicuously

RUTACES, 353

feather-veined, pellucid-punctate, the copious dots black by reflected
light: petiole 6 to 8 lines long, slightly margined. From one axil, in
the specimen, proceeds a short bibracteate peduncle, only 2 lines long,
when it is trichotomous; the one-flowered pedicels 3 or 4.lines long,
bibracteolate in the middle. Flower-bud 3 lines long, globular; the
calyx rather longer than the still undeveloped corolla: sepals gla-
brous, orbicular, thickish, with very thin margins, strongly imbricated
in eestivation, slightly ciliate, doubtless persistent. Petals in the not
full-grown bud, orbicular, glabrous, strongly imbricated in eestivation ;
their minute apexes however are inflexed and more or less compli-
cated together, or coherent. Stamens 8, short: anthers adnate,
introrse; the connective glandular on the back. Ovaries apparently
distinct or nearly so, glabrous.

5. MeELicope? ELLIPTICA, Sp. Nov.

M. glabra; foliis ellipticis utrinque obtusis coriaceis subtus grosse reticur
latis; pedunculis brevissimis axillaribus ; folliculis ovoiders cum calyce
persistente tomentulosis.

Has. Kaala Mountains; in the district of Waianae, Oahu.

The specimens are two small branches of an apparently arborescent
plant, in fruit, and with mere vestiges of flowers. Leaves opposite, .
glabrous, as is the stem, &c., elliptical, obtuse at both ends, about 2
inches long and little over an inch wide, coriaceous, with some-
what revolute entire margins, punctate, feather-veined, sparingly and
coarsely reticulated, especially wnderneath; the petioles 3 or 4 lines
long, slightly margined. Peduneles very short, either solitary in the
axils and 1—3-flowered, or somewhat fascicled. Calyx tomentulose ;
the sepals broadly ovate, coriaceous, with thin margins, persistent at
the base of the fruit, to which they are appressed. The vestiges of
petals seen in the remains of a single flower are oblong, canescent
externally, longer than the calyx. Style rather shorter than the
globular distinct ovaries, villous below, deciduous: stigma four-lobed.
Follicles ovoid, cinereous-tomentose, 5 lines long, turgid, coriaceous, at
length two-valved or nearly so: the separable papery endocarp gla-

brous within. Seeds as in the genus.
89

304 PHANEROGAMIA.

The carpels of this ambiguous and doubtful species resemble those
of Pelea Sandwicensis; but the endocarp is smooth within; and the
calyx 1s perfectly persistent, while in the plant above mentioned it is
very deciduous. The estivation of the corolla is requisite for the
determination of the genus. The withered petals examined appeared
to have their edges thin towards the base, but thicker and abrupt near
the apex.

6. Meticorpe? GRANDIFOLIA, Sp. Nov.

M. glabella; foliis magnis obovato-lanceolatis sew spathulato-oblongis
membranaceis basi attenuatis subtus pl. m. pubescentibus.

Has. In forests of Mouna Kea, Hawaii.

Only leafy shoots occur in the collection, gathered, according to Dr.
Pickering’s notes, from “‘a tree 20 feet high, with the trunk 8 or 10
inches in diameter at the base.” The still young and membranaceous
leaves are from 8 to 12 inches long, 2 or 3 inches wide above the
middle, obovate-lanceolate or spatulate-oblong, tapering into a short
petiole, sparsely veiny, thickly pellucid-punctate, glabrous or glabrate
above, mostly pubescent underneath with rather sparse hairs. Neither
flowers nor fruit occur in the collection. The fruit is said to be “ soli-
tary and axillary.”

17. ZANTHOXYLUM, Colden.

1. ZANTHOXYLUM KAUAENSE, Sp. Nov.

4. merme, glabrum ; folws alternis pinnatis 3—5-foliolatis ; foliolis coria-
ceis ovalibus integerrinus haud punctatis; paniculis axillaribus com-
positis; fructibus stipitatis.

Has. On the mountains of Kauai, one of the Sandwich Islands.

The specimens bear mature fruit only, and are insufficient for the
positive determination of the genus. They were gathered from a tree

RUTACES. 355

of considerable size; the sensible properties of which are not recorded.
No aroma or pungent taste 1s perceptible in the dried bark of the
branches, which are unarmed, and no dots in the foliage, which is very
coriaceous. Leaves alternate, glabrous, as are all the parts, punnately
3—5-foliolate; the petiole and rhachis terete. Leaflets oval, obtuse,
more or less unequal at the base, except the terminal one, which is
long-petiolulate, the others short-petiolulate, one and a half to two
and a half inches long, very opaque, not shining, of the same hue both
sides, veiny, entire. Panicles or cymes axillary, shorter than the
leaves, short-peduncled, apparently compound and many-flowered : but
in the specimens only a few dehiscent carpels remain. There are
some traces of at least two carpels to each flower, only one of which
has ripened. The follicles are stipitate (stipe 2 lines long), short and
turgid, 4 to 5 lines long, somewhat wrinkled and punctate, two-valved
from the apex; the endocarp not readily if at all separating from the
exocarp. Seed solitary, filling the cell, oval, with a black and
shining, somewhat drupaceous testa of great thickness. HEmbryo
nearly of the length and breadth of the albumen: cotyledons round-
oval: radicle superior.

Orv. SIMARUBACE &.

1 AMARORIA, Nov. Gen.

Flores monoici vel dioict. Masc.—Sepala 6, an semper? Petala nulla.
Stamina numero sepalorum, wsdem opposita: anther subsessiles.
Discus carnosus, profunde trifidus, lobis bifidis. Foem.—Sepala 4-5,
parva, persistens. Petala 4-5, linearia, carinata, reflexo-patentia.
Rudimenta staminum petalis numero dupla, minima, sub disco meras-
sato 8—10-crenato imserta. Ovarium simplex, ovoideum, uniloculare,
uniovulatum, vertice stigmate sessilli maximo depresso reniformi
crasso obtectum. Ovulum sub apice loculi appensum, subanatropum.
Drupa sicca, nuciformis, ovoidea, subcompressa, epicarpio tenui, puta-
mine osseo. Semen loculum wmplens, amphitropum, exalbumimosum.
Cotyledones ovales, plane: radicula brevissima, supera.—Arbuscula
Soulamez amare facie, foliis longius petiolatis.

1. AMARORIA SOULAMEOIDES, Sp. Nov. (Tab. 40.)
Has. Mountains of Muthuata, Feejee Islands.

Apparently a small tree, with stout and simple branches, scarred
by the large and approximate cicatrices of the fallen leaves; the
young shoots and nascent leaves minutely silky-pubescent; otherwise
glabrous. Leaves alternate, crowded, almost exactly like those of
Soulamea amara, except that they have longer petioles (from 2 to 3
inches in length) and are rather narrower and less contracted towards
the base, elliptical-oblong or elongated-oblong, acute or obtuse, the
base usually acute, 5 or 6 inches long, 2 or 24 inches wide, coriaceous,
entire, not punctate, copiously feather-veined; the primary veins

SIMARUBACES. 357

straight, about 20 pairs, decurrent into an indistinct and waved
intramarginal vein. Stipules none. Peduncles axillary, bearing a
several-flowered, small, and racemose panicle, shorter than the petiole.
Flowers either monecious or dicecious (from the state of the specimens
it is impossible to ascertain which), very small, about a line and a
half long, on short pedicels. Bracts subulate, minute. Male flowers
(known only from the analyses on the plate, made under Mr. Rich’s
superintendence). Calyx of 6 ovate sepals. Petals wanting. Stamens
6, opposite the sepals: anthers subsessile, ovoid-didymous, two-celled,
probably introrse; the cells opening longitudinally. Disk fleshy,
large, deeply three-cleft; the segments two-lobed, obtuse. No rudi-
ments of a pistil. Female flowers. Sepals 4 or 5, oblong-ovate,
minute, in anthesis much shorter than the pistil, erect, persistent,
furnished with a small globular gland at the base on the inside.
Petals 4 or 5, hypogynous, narrowly linear, carinate, at length reflexed-
spreading, a line or a line and a half long, deciduous or marcescent.
Rudiments of stamens 8 or 10, opposite and alternate with the petals,
hypogynous, corresponding with the crenate sinuses of the disk,
which they slightly exceed in length, minute, thickened above, but
not antheriferous. Disk very thick and fleshy, somewhat saucer-
shaped, the edge 8—10-crenate, or 4—5-lobed, with the lobes emarginate.
Ovary simple, sessile in the disk, longer than the petals, fleshy, ovoid,
scarcely at all compressed, one-celled, one-ovuled, its apex covered by
the very large and fleshy, depressed, kidney-shaped, sessile stigma.
Ovule borne on one side of the cell below its summit, between amphi-
tropous and anatropous; the micropyle superior. Fruit a dry drupe
or nut, ovoid, slightly compressed, scarcely margined, an inch in length,
slightly pointed, indehiscent; the epicarp thin and crustaceo-coria-
ceous, smooth; the putamen bony. Seed filling the cell, amphitropous,
oval, with a membranaceous testa, destitute of albumen. Embryo con-
formed to the seed: cotyledons oval, flat, rather fleshy: radicle very
short, partly included, superior.

The drupe and the wood are very bitter, perhaps as much so in
Soulamea, which was named by Rumphius, from this quality, Rea
Amaroris. In allusion both to this sensible property, and to the
Rumphian name of its nearest ally, I have chosen the name of the
genus, from amaror, bitterness. For our plant is evidently gene-

rically distinct from Soulamea, as well by its 4—6-merous (instead of
90

358 PHANEROGAMIA.

trimerous) floral envelopes, and the completely diclinous (instead of
moncecio-polygamous) flowers, as by its monocarpellary ovary, and its
thickened, nut-like fruit. I regret that I am wholly unable to con-
firm the characters of the male flowers; the few buds which existed
on the specimens having been destroyed in the preparation of the
analyses given upon the plate, which was prepared and engraved
under the superintendence of Mr. Rich. I have merely caused some
further details of the ovary and fruit to be added or corrected. (Fig.
12-15.)

PLATE 40.—AMARORIA SOULAMEOIDES; a branch of the female plant,
of the natural size. Fig. 1. Expanding bud of a male flower. 2.
Same, with the sepals and one of the stamens removed. 3. Male
flower, with the stamens removed, showing the lobed disk. 4, 5.
Anthers. 6. Portion of the inflorescence, with female flowers. 7.
Female flower. 8. A petal. 9. A sepal. 10. An abortive stamen.
11. Female flower, with the pistil removed, and a portion longitudi-
nally cut away. 12. Vertical section of a pistil and receptacle, show-
ing the ovule-—All these figures variously magnified. 13. Fruit, of
the natural size. 14. Vertical section of the same, showing the seed.
15. The embryo, of the natural size.

Orv OCHNACE &.

1. GOMPHIA, Schreb.
1. GOMPHIA LINEARIS, Sp. Nov. (Tab. 41.)

G. ramosissima, glaberrima; ramis gracillimis; foliis linearibus utrin-
que acutiusculis brevissime petiolatis subintegerrimis subaveniis ; sti-
pulis subulatis caducis; paniculis terminalibus paucifloris; petalis
calyce vix longioribus ; antheris levibus ; gynophoro ovario longiore.

Has. Organ Mountains, near Rio Janeiro, Brazil; in deep shady
woods.

Shrub 5 to 10 feet high, entirely glabrous, much branched, the branches
very slender, and perhaps somewhat climbing or sarmentose, very
leafy; the youngest branchlets minutely puberulent. Leaves linear,
or linear-lanceolate, often falcate, mucronulate, one to two inches long,
one to 3 lines wide, tapering more or less to the apex, which, like the
base is rather acute, on very short petioles (of barely a line in length),
entire, or sometimes obscurely serrulate above the middle, not shining,
almost veinless to the naked eye, under a lens showing rather sparse
and indistinct, oblique, simple, primary veins. Stipules subulate,
rather longer than the petioles, scarious, striate, distinct, early decidu-
ous for the most part. Panicles terminal, small, loosely few-flowered ;
the branches divaricate. Bracts minute, caducous. Pedicels one to 3
lines long, very glabrous. Flowers 3 or 4 lines in diameter, yellow;
the buds ovoid and very obtuse. Sepals thin, elliptical-oblong, very
obtuse, deciduous. Petals rather longer than the calyx, obovate.
Anthers smooth and even, oblong-linear, sessile. Gynophore longer than
the pentacarpellary ovary, columnar. Style filiform, elongated. Stigma

360 PHANEROGAMIA.

simple, acute. Ovule arising from a little above the base of the cell,
ascending, between anatropous and amphitropous, straight ; the micro-
pyle inferior. Fruit not seen.

Allied in some respects to Gomphia stipulacea and G. salicifolia, but
a truly distinct, and apparently entirely new species, with narrower
leaves than any other known Gomphia.

PLATE 41.—GoMPHIA LINEARIS : a branch, of the natural size. Fig. 1.
A leaf, somewhat enlarged, showing the venation and the stipules.
2. Diagram of the flower. 3. An expanded flower. 4. A sepal. 5.
A petal. 6,7. Anthers. 8. Pistil, with the gynophore and pedicel.
9. Vertical section of the same.—The details enlarged.

2. GOMPHIA STIPULACEA, Planch.

Ochna stipulata, Velloz. Fl. Flum. 5, t. 91.
Gomphia stipulacea, Planch. in Hook. Lond. Jour. Bot. 6, p. 6.
G. iteoides, Erhard, in Flora (Bot. Zeit.), 1849, p. 2517

Has. Organ Mountains, Brazil.

The specimen resembles Gardner’s No. 5692; but the flowers are
decandrous; and the leaves (of which the larger are almost three
inches long) are not three-toothed, but barely emarginate at the apex.

3. GOMPHIA PARVIFLORA, DC.

Gomphia parviflora, DC. in Ann. Mus. 17, p. 420, t. 16; Planch. l.c.; Erhard, 1. ¢.
Ochna Jabotapita, Velloz. Fl. Flum. 5, t. 90.

Has. Vicinity of Rio Janeiro, Brazil.

4, GOMPHIA ZMULA, Pohl.
Gomphia emula, Pohl, Pl. Bras. Ic. 2, p. 180, t. 182, ex tab.; Erhard, 1. c. p. 249.

Has. Organ Mountains, Brazil. (With young fruit.)

OCHNACESR, 361

5. GOMPHIA OLIVFoRMIS, St. Hil.

Gomphia oliveformis, St. Hil. Fl. Bras. Mer. 1, p. 67; Planch. in Hook. Lond. Jour.
Bot. 6, p. 11.

Has. Rio Janeiro, and on the Organ Mountains, Brazil.

2 BRACKENRIDGEA, Nov. Gen.

Calyx persistens. Antherce (lceves) longitudinaliter dehiscentes. Stigma
leviter quinquelobum. Ovulum circa processum e fundo ovarit assur-
gentem curvatum, hyppocrepicum. Semen angustum circinnatum.
Embryo semini conformis, gracilis; cotyledonibus anguste linearibus ;
radicula centripeto-infera. — Flores umbellato-fasciculatr.. — Cetera

Gomphie.

The plant from which this character is derived appears to be neatly
distinguished from Gomphia, by the longitudinal dehiscence of the
anthers, the nearly annular ovule and seed, curved around a large pro-
jection into the cell of the ovary (in the manner of Menispermum), which
arises from its inner angle near the base, and by the slender embryo,
with narrowly linear cotyledons ; to which may be added, as subsidiary
characters, the entirely persistent calyx, and the radiately somewhat
five-lobed stigma. The Gomphia? Hookeri, of Planchon, in Hook.
Lond. Jour. Bot. 6, p. 8, from Penang, is most probably a second
species of the genus; and some other Gomphice of the Old World
make some approach to it in the curvature of the ovule, and have per-
sistent calyxes; but the embryo and anthers of Gomphia, so far as
known, are different.

The name selected for this genus is intended to commemorate the
important scientific services of Mr. William D. Brackenridge, the
Assistant Botanist of the Expedition, through whose indefatigable
zeal and industry this botanical collection was principally made, of

which he has himself elaborated the Ferns and the allied orders.
91

362 PHANEROGAMIA.

1. BRACKENRIDGEA NITIDA, Sp. Noy. (Tab. 42.)

B. glaberrima,; jfoliis oblongis lanceolatisve nitidis; pedicellis in um-
bellas sessiles axillares et terminales multifloras fasciculatis cum rhachi
brevissima articulatis ; floribus flavidis ; filamentis brevissimis.

Has. Sandalwood Bay, Vanua-levu, one of the Feejee Islands.

A shrub, or small tree, 15 or 20 feet high, entirely glabrous through-
out; the bark of the branchlets mostly wrinkled or corrugated.
Leaves alternate, varying from oblong to lanceolate, 3 to 6 inches long
and one or two wide; the broader ones more or less obtuse; the nar-
rower commonly acute at both ends, on very short and thickened
petioles (the bark of which becomes corrugated like that of the
branchlets), coriaceous, entire, very smooth and shining, especially
above, finely and very copiously feather-veined, the veins obliquely
transverse, except one or two usually of the lower ones on each side,
which are procurrent within the margin and parallel to it, and conti-
nued to the apex of the leaf, inosculating with the ascending extre-
mities of the finer transverse veins. Stipules small, subulate-seta-
ceous, and with two lateral cusps about the middle, their bases united
within the petiole, as in the Gomphice of the Old World, caducous.
Pedicels aggregated in sessile and many-flowered umbels or fascicles, in
the axils of the leaves, or sometimes terminal on very short spurs or tuber-
cular bases, which are often clustered, and clothed with minute and
scarious, deciduous scales: the pedicels are 6 or 8 lines long, at least
in fruit, articulated at their insertion, filiform. Calyx of 5 broadly
oval and chartaceo-membranaceous sepals, imbricated in sstivation,
yellowish, spreading, persistent in fruit. Petals 5, oval-obovate,
apparently yellow, about as long as the calyx, hypogynous, imbricated
in estivation, deciduous. Stamens 10, hypogynous, inserted on the
summit of a short and thick gynobase, outside of a narrow disk or
margin which surrounds the base of the ovary: filaments very short,
persistent: anthers articulated with the filament, innate, linear-
oblong, smooth (not wrinkled), two-celled, with a rather broad con-
nective, the cells opening longitudinally for their whole length at the
margins. Pollen-grains spherical, simple. Ovary five-parted on the

OCHNACE &. 363

summit of the short gynobase; the lobes connected only at. their
insertion around the base of the columnar five-grooved style, laterally
compressed, smooth, one-ovuled. Stigma truncate, radiately some-
what five-lobed. Ovule linear, arising from the base of the cell out-
side of the thick, ascending protuberance that partly fills the cavity,
around which it is curved nearly into the shape of a horse-shoe: it is
anatropous, with a ventral rhaphe; the micropyle next the hilum.
Fruit of 5 baccate drupes, somewhat compressed, borne on the margin
of the enlarged and depressed, disk-shaped, fleshy gynobase; the
sarcocarp very thin; the endocarp coriaceous, coherent with the
edges of the internal projection, so that a transverse section at the
middle shows four small cells, two of them occupied by the seed, the
lateral ones empty. Seed linear, conformed to the cell, coiled nearly
into a ring; its testa membranaceous. Albumen none. Embryo
conformed to the seed, nearly annular: the thin cotyledons narrowly
linear; the slender radicle centripetal-inferior.

PLATE 42.—BRACKENRIDGEA NITIDA : a branch, of the natural size,
in flower and in fruit. Fig. 1. Stipules, with part of a young leaf.
2. Unopened flower. 3. Expanded flower. 4, 5. Stamens. 6. Pistil,
with the gynobase and persistent filaments. 7. Vertical section of the
same, showing the curved ovules in place, &c. 8. Immature fruit,
with the persistent calyx. 9. Mature fruit, with the calyx. 10. Ver-
tical section of one of the drupes, showing the seed. 11. Trans-
verse section of the same. 12. The embryo detached—The details
variously enlarged.

Orv. ANACARDIACE .

1 MANGIFERA, Linn.
1. Maneirera Inpica, Linn.

Has. Near Manilla. (Only foliage of the Mango was collected.)

2 ANACARDIUM, Rottb.
1, ANACARDIUM OCCIDENTALE, Linn.

Has. Near Rio Janeiro, Brazil. ‘Only seen in a cultivated state.”

38. ONCOCARPUS, Nov. Gen.

Flores dioici (sew monoici?). Calyx cupularis, quinquedentatus. Petala
5, hypogyna, oblonga, patentissima, cestivatione valvata. Masc.—
Stamina 5, circa torum hemisphericum hirsutum inserta. Gyncecium
nullum. Foem.—Stamina nulla? Ovarium pyramidatum, basi
quinquelobum, stigmate sessili truncato terminatum, uniloculare ?
Drupa depressa, difformis, torosa vel sublobata, toro incrassato carnoso
obconico insidens; putamine osseo, sinuoso-multilobato, uniloculari,
monospermo. Semen exalbuminosum, loculo sinuoso conforme ; testa
tenui. Embryo transversus; cotyledonibus carnosis lobatis ; radicula
brevissima.—Arbor simplicifolia, Semecarpi facie et inflorescentia.

ANACARDIACES. 36

ony ]

1. Oncocarrus Vitrensis, Sp. Nov. (Tab. 43.)

Has. Feejee Islands. In a jungle at Rewa, Viti-levu. Also in
the Sandalwood district of Vanua-levu; on hillsides.

Tree 30 or 40 feet high; the stout branches, foliage, &c., entirely
glabrous. Leaves simple, alternate, of a thick and coriaceous texture,
obovate-oblong, very obtuse, sometimes retuse, entire, or somewhat
undulate or repand, more or less acute at the base, from 4 to 9 inches
long, and from 2 to 44 inches wide, on stout petioles of 6 to 18 lines
in length, smooth, strongly pinnately-veined with 9 or 10 pairs of
primary veins which, like the thickened midrib, are very prominent
underneath, and inosculate near the margins; the veinlets extremely
copious and finely reticulated. Stipules none. Flowers small, very
numerous, in a compound terminal panicle, which is minutely pubes-
cent. Pedicels very short. Bracts ovate-subulate, minute. Male
flowers. Calyx short, cup-shaped, five-toothed, puberulent. Petals 5,
hypogynous, oblong, acutish, puberulent outside, valvate in cestivation,
with a minute inflexed apex, widely spreading, a line and a half long,
deciduous. Stamens 5, alternate with the petals, and inserted with
them around the depressed, or hemispherical and hirsute torus that
occupies the centre of the flower: filaments filiform-subulate, rather
shorter than the petals: anthers oblong, emarginate at both ends,
two-celled, attached to the apex of the filament a little above the
notch at the base, introrse; the cells opening longitudinally. Pistil
none. Female flowers known only from a sketch made by the bota-
nical draughtsman of the Expedition (copied in Fig. 5, 6 of the
Plate), which represents the calyx and corolla as in the male flowers,
no stamens, and a conical or pyramidal ovary, with 5 obtuse lobes at the
base, alternate with the petals, occupying the centre of the flower;
the stigma sessile and truncate: its interior structure is not repre-
sented; but it is probably one-celled. Fruit a depressed, and as it were
deformed, knobby drupe, borne on a short and thickened, obconical, fleshy
torus (3 or 4 lines long), which is much smaller than the pericarp :
sarcocarp fleshy: putamen bony, depressed, irregularly sinuate-lobed
(the lobes about 8) one-celled. Seed solitary, conformed to the sinuose

cell, with a very thin integument: albumen none. Embryo transverse,
- 92

366 PHANEROGAMIA.

conformed to the cell: cotyledons irregularly lobed, fleshy, partly con-
solidated : radicle extremely short, centrifugal.

The tree is said to be poisonous, producing inflammation of the
skin, with eruption, like Rhus venenata, or Anacardium and Seme-
carpus. From Semecarpus our plant generically differs in the valvate
estivation of the corolla, the smaller fructiferous receptacle, and the
drupaceous, depressed, knobby or tuberculate fruit (whence the name,
from éyx0s, a tubercle or swelling, and xaprés, fruit), with an exceedingly
sinuous and lobed putamen, to which the seed is conformed. Perhaps
our plant is more nearly related to Mr. Bentham’s genus Cyrtosper-
mum, from the Amazon: but that has decandrous and hermaphrodite
flowers, an ovoid drupe, and the putamen and cotyledons not lobed.

PLaTteE 43.—Oncocarpus Vitiensis: the male plant, in flower.
Fig. 1. Diagram of a male flower. 2. The flower, expanded. 3, 4.
Stamens. (5. Unexpanded female flower. 6. The same, expanded,
From a sketch by Mr. Agate.) 7. Branch of the panicle, with fruit,
of the natural size. 8. Vertical section of a young drupe and its
fleshy torus. 9. Transverse section of the same. 10, 11. Putamen,
from two different drupes, one seen from above, the other from below.
12, 13. Embryo, of the natural size—The details enlarged.

4. BUCHANANIA, Roz.
1. BucHANANIA FLORIDA, Schauer. (Tab. 44.)

B. glabra; foliis obovato-oblongis obtusis retusisve in petiolum attenuatis ;
paniculis ramosis laxifloris glabellis ; pistillis in disco cupulari multi-
dentato semiimmersis.

Buchanania florida, Schauer, in Reliq. Meyen. p. 481.
Has. Vanua-levu, one of the Feejee Islands.

A shrub, or tree, with stout and thick branches, towards the
extremity of which the leaves are crowded, glabrous, or the nascent
branchlets pubescent. Leaves obovate-oblong, 5 or 6 inches long and
2 or more wide, very obtuse, sometimes retuse, mostly acute at the base,

ANACARDIACES. 367

coriaceous, dull; the stout petiole half or two-thirds of an inch in
length. Punicles in the axils of the leaves at the summit of the
branch, branching, loosely many-flowered, slightly pubescent under a
lens, exceeding the leaves. Pedicels shorter than the flowers. Calyx
elabrous, persistent, five-cleft; the lobes broadly ovate, obtuse, slightly
if at all ciliate, probably imbricated in sestivation. Petals 5, scarcely
2 lines in length, ovate-oblong, recurved-spreading, deciduous. Sta-
mens 10, rather shorter than the petals, and inserted with them:
filaments subulate: anthers oblong, somewhat sagittate. Disk cupu-
late, exceeding the calyx, with the base of which it is scarcely cohe-
rent, thick and fleshy, the thinner edge minutely many-toothed, Pistals
usually 5, half immersed in the closely encircling disk, slightly hairy,
one (or sometimes 2 or 3) of them obliquely ovate and ovuliferous ;
the others fusiform and effete. Styles scarcely any: stigma terminal,
thickened, very obtuse. Only the immature fruit is known; in which
two or three of the ovaries sometimes enlarge and become more or less
coherent at the base: probably only one of them is fertile.

The plant here described well accords with the character and an
imperfect original specimen of Schauer’s B. florida, from Manilla, ex-
cept that the flowers are larger, and the lobes of the calyx show
scarcely a trace of minute ciliate hairs—From the ticket, I perceive
that Mr. Rich supposed our plant to be the Rhus atrum of Forster, a
New Caledonian species, which has obovate and coriaceous leaves,
indeed, but bears its (probably pentandrous) flowers in a dense, spi-
cate-crowded inflorescence, much shorter than the leaves.

Puate 44, 4.—BuCHANANIA FLORIDA: a branch in flower, of the
natural size. Fig. 1. A flower. 2. A petal. 3, 4. Stamens. 0.
Vertical section of a flower. 6. Pistils, with the enclosing disk. 7.
The fertile and a sterile pistil; the former with the ovary vertically

divided, to show the ovule-—The analyses all magnified.

5. RHUS, Linn.
1. Ruvs SIMARUB&FOLIA, Sp. Nov. (Tab. 44.)

R. glabra; foliis pinnatis, rhachi superne marginata: Joliolis 7-9

368 PHANEROGAMITIA.

oblongis obtusis basi ineequilateris integerrimis supra nitidis subtus
pallidis subsessilibus; paniculis axillaribus laxifloris folio paullo bre-
vioribus ; floribus albis.

Has. Muthuata, one of the Feejee Islands.

Shrub (the size not recorded) glabrous throughout; the branchlets
somewhat warty or dotted, when young reddish, as are the petioles.
Leaves pinnate. Petiole and rhachis 5 or 6 inches long, the latter
angled and becoming narrowly margined between the upper pairs of
leaflets, and more so below the terminal leaflet. Leaflets T-9, oblong
or narrowly elliptical, one or two inches in length, obtuse, nearly
sessile, all but the terminal one strongly oblique or wneguilateral at the
base, entire, of a rather firm texture (evidently deciduous), copiously
feather-veined ; the veins rather indistinct on the shining upper sur-
Jace, but more conspicuous on the pale lower surface. Panicle axillary,
rather shorter than the leaf, loosely flowered; the flowers numerous, on
pedicels of one or two lines in length, probably polygamous. Calyx
five-lobed; the lobes very short and obtuse. Petals white, oval-obo-
vate, a line and a half in length, veiny; with a midrib which is
rather prominent on the upper surface and a little hairy towards the
base. Stamens 5, shorter than the petals, inserted around the fleshy
ten-grooved or crenate disk. Ovary glabrous, one-celled, with 3
thickened subsessile stigmas. Fruit not seen. .

This species belongs to the section Swmac, and to the group which
comprises 2. Copallina. The leaflets, although thinner and with a
different venation, bear some resemblance to those of Simaruba
glauca, whence the specific name.

PLATE 44, B—RuUS SIMARUBAFOLIA: a flowering branch, of the
natural size. Fig. 1. A flower. 2. A petal. 3. Flower, with the
petals removed. 4. Pistil and disk. 5. Vertical section of the pistil.
—The analyses magnified.

2. Raus Tarrensis, Gull.

Rhus Taitensis, Guillem. Zeph. Tait. p. 67, & in Ann. Sci. Nat. ser. 2, 7, p. 361.

ANACARDIACES. 369

Has. Tahiti: in mountain forests, sometimes in scattered groves.
Common also at Tutuila, Navigators’ Islands, and at Tongatabu.

I find no Tahitian specimens in the collection; but those from the
Tonga and Samoan Islands (both in fruit only) accord very well with
the character of &. Tuitensis, except that in the Samoan specimen
the leaflets are occasionally even 6 inches long, and some of them
acuminate. The pinne are often fewer than 10 pairs. The drupe is
black, smooth and shining, 2 or 3 lines in diameter.

3. Ruus SANDWICENSIS, Sp. Nov.

R. ramulis novellis velutino-tomentosis; foliis pinnatis; folrolis 5-9
(rarius 3) ovatis oblongisve subacuminatis serratis subsessilibus supra
glabratis subtus cum rhachi immarginata paniculaque terminalt com-
posita amplissima tenuiter tomentosis ; floribus parvis.

Has. Sandwich Islands: Oahu, in the mountains behind Honolulu
(where it was collected by the late Rev. J. Diell); and in forests near
Hilo, Hawaii. (Byron’s Bay, Hawaii, Macrae. Oahu, Gaudichaud.)

Shrub 4 to 10 feet high, with much the habit of R. typhina; the
stout branches warty, smooth, or when developing velvety-tomentose ;
the marginless petioles, inflorescence, &c., tomentose with a fine and soft
pubescence. Leaves pinnately 5-9-foliolate, or the upper ones some-
times trifoliolate. Leaflets oval or oblong, more or less acute or slightly
acuminate, ample, from 2 to 6 inches long, and from an inch to 4
inches wide, more or less serrate, almost sessile, copiously feather-
veined, with 12 to 20 pairs of straight veins, which are prominent on
the downy under surface, the upper face nearly glabrous. The leaves
considerably resemble those of a Walnut or Butternut. Punicle ter-
minal, very larye, and compound (sometimes with smaller ones in the
upper axils), loose and open, often a foot in diameter, very many-
flowered; the flowers small (a line and a half in diameter), apparently
yellowish-white, racemosely crowded on the ultimate ramifications.
Calyx nearly glabrous, deeply five-cleft; the lobes oblong-ovate,
minutely ciliate, not half the length of the oval and minutely ciliolate
petals. Stamens 5, rather shorter than the petals. Styles 3, clavate.

Fruit not seen.
93

370 ' PHANEROGAMITA.

This very distinct and still unpublished species (apparently allied
to LR. juglandifolia of New Grenada) was sent to me many years
since by the late Mr. Diell, when I communicated a specimen to the
herbarium of Sir William Hooker, under the name here adopted.

4. Ruus tucipa, Linn.

Has. Cape of Good Hope, in the vicinity of Cape Town. (Foliage
only.)

6 LITHRAA, Mers.
1. Lirar#a caustica, Hook. & Arn.

Lithrea caustica, Hook. & Arn. Bot. Misc. 3, p. 175.

LI. venenosa, Miers, Trav. in Chil. 2, p. 529; Gay, Fl. Chil. 2, p. 44.
Laurus caustica, Molina, Chil. p. 151; Willd. Spec. Pl. 2, p. 479.
Rhus? caustica, Hook. & Arn. Bot. Beech. Voy. p. 15, t. 7.

Has. Chili, from Valparaiso to the base of the Cordilleras near
Santiago. Both the smooth and the downy forms; the latter more
common near the coast.

7. DUVAUA, Kunth.
1. DuUVAUA DEPENDENS, DC.
Has. Near Valparaiso, Chili.

Besides the narrow-leaved form of the species, which answers to
Amyris polygama, Cav. Ic. 3, t. 239, and the var. latifolia (D. lati-
folia, Gillies), which is considered to be only a state of the species
with broad and mostly toothed leaves, there is a variety like the
latter, but with very downy leaves and branchlets.

Duvaua? dentata, DC. Prodr. 2, p. 74 (Schinus dentatus, Andr. Bot.
Rep. t. 620), is wholly founded on a cultivated plant, supposed, on
mere hearsay evidence, to come from Hawaii. But I have seen

ANACARDIACE &. 371

nothing like the plant figured, in any collection made at the Sandwich
Islands.

8. SCHINUS, Linn.
1. SCHINUS TEREBINTHIFOLIUS, Raddi.

Has. Brazil; common around Rio Janeiro, and in the Organ
Mountains.

2. Scarnus Motte, Linn.

Has. Peru; very abundant around Lima.

9 MAURIA, HBX.

1. MAurRIA MULTIFLORA, Mart.

Mauria multifiora, Mart. Herb. Fl. Bras. no. 1274; Benth. in Hook. Kew Jour. Bot.
4, p. 8.

Has. Brazil, near Rio Janeiro.

For the probable synonymy of this species, see Mr. Bentham’s
annotations, in the work here cited.

10? JULIANIA, Schlecht.

JULIANIA, Schlecht. in Linnea, 17 (1843), p. 746, non Llave & Lex.
HYPOPTERYGIUM, Schlecht. in Linnza, |. c. p. 635, non Bridel.

1. Junrania Huaucut, Sp. Nov.
J. fructu recto equilatero (tripollicart et ultra).

Two leafless branches are in the collection, from “the vicinity of
Yanga, Peru,” with nothing besides a terminal fascicle of immature,

372 PHANEROGAMIA.

pendent, samaroid fruits, of a peculiar structure, the pericarp being
borne on the apex of a long and linear wing (2% inches in length and
half an inch wide), its base narrowed into a short peduncle. The
fruit may be likened to that of an Ash inverted. The structure of
the pericarp itself, which is involved in the apex of the wing, cannot
be made out in the specimens, farther than that it is dry, and appears
to be by abortion one-celled and one-seeded. A specimen of the same
plant, in nearly the same condition, and collected in the same district,
is preserved in the herbarium of Sir William Hooker, where it is
referred to the genus Juliana of Schlechtendal. It was collected by
Mr. Matthews, who sent with it the following note :—*Small tree,
covering the sides of the base of the Cuesta of Purruchuca, province
of Canta, Peru; April, 1831.” Another specimen, from Lima,
McLean, is ticketed “ Huaucui of the natives: a dicecious shrub,
seldom seen with leaves, and always black as if burned or blasted.”
Not having at the time collated the materials in the Hookerian her-
barium with the character and description of Schlechtendal’s Juliana
(or Hypopterygium) adstringens, of Mexico, I cannot say whether our
plant is correctly referred to that genus. It can hardly be the same
species, since the wing of the fruit is straight and equal-sided, and an
inch longer than that of the Mexican plant is said to be.

11? CORYNOCARPUS, Forst.
1, CoRYNOCARPUS LEVIGATA, Forst.

Corynocarpus levigata, Forst. Prodr. Fl. Ins. Austr. p. 19; Hook. Bot. Mag. t.
4397; Hook. f. Fl. N. Zeal. p. 48.

Has. Bay of Islands, New Zealand,
The specimens consist of foliage only.—A genus of very doubtful

affinity, which Dr. Hooker has recently appended to the Anacar-
diacece.

Orv. BURSERACE &.

1. CANARIUM, Linn.
1. Canartum VitienssE, Sp. Nov.

C. glaberrimum ; stipulis subulatis parvis caducis; foliolis 5-7 oblongo-
ellipticis utringue obtusis nunc subacuminatis nitidis ; paniculis axil-
laribus paucifloris ; pedicellis (semipollicaribus et ultra) clavatis.

Var. @. foliolis 5-9 sceepius apice vel utrinque plus minus acuminatis.
Has. Feejee Islands: on Muthuata, at an elevation of 2,000 feet.

A shrub or tree, perfectly glabrous in every part; the leaflets 5 or
7, smooth and shining, mostly obtuse or obtusish at both ends, some-
times rather acuminate (or the terminal one acute at the base), coria-
ceous, about 3 inches long and 2 in width, veiny. Neither the
flowers nor the ripe fruit are known; the specimens bearing only
immature drupes (less than an inch long), in small, evidently /ew-
flowered, axillary (or slightly supra-axillary) panicles or racemes,
which are shorter than the leaves. Fructiferous pedicels half an inch
or more in length, stout, clavate, bearing the persistent, three-angled,
spreading calyx (about 5 lines in diameter), which surrounds the
acute base of the fusiform, somewhat triangular, sharp-pointed, gla-
brous, immature drupe: the latter contains a single forming seed.—
The variety has more numerous and thinner, as well as acuminate
leaflets; but it is probably merely a state of the species, growing in a
more shaded situation.

The species appears to be abundantly distinct from any of those
94

374 PHANEROGAMIA.

recently characterized by Blume, and to bear most resemblance to
Bentham’s Canarium asperum, from New Guinea.

2. CANARIUM (PimeLA) Luzonicum.

Pimela Luzonica, Blume, Mus. Bot. Lugd.-Bat. p. 220?
Canarium album, Blanco, FI. Filip. p. 793, non Reeusch, ex Blume, 1. ¢.

Has. Vicinity of Baiios, Luzon. (An imperfect fruiting specimen.)

2. EVIA, Comm.
1. Evra puucis, Comm., Blume.

Evia dulcis, Commerson Mss. ex Juss.; Blume, Mus. Bot. Lugd.-Bat. p. 233.
Chrysomelon pomiferum, Forst. Mss. in Mus. Bot. Par.

Spondias dulcis, Forst. Prodr. p. 34; DO. Prodr. 2, p. 75; Guill. Zeph. Tait. p. 68.
Spondias Cytherea, Sonn. Itin. 2, p. 222, t. 123; Geertn. Fruct. t. 103.

Poupartia dulcis, Blume, Bijdr. p. 1161.

Has. Tahiti and Eimeo, Society Islands: common on the coast
and in the interior. Also cultivated at the Tonga and other islands.
The Vi apple, a well-known fruit.

3 DRACONTOMELON, Blume.

1. DRACONTOMELON SYLVESTRE, Blume.

Dracontomelon sylvestre, Blume, Mus. Bot. Lugd.-Bat. p. 281.
Pomum Draconis sylvestre, Rumph. Herb. Amboin. 1, p. 159, t. 59.

Has. Planted at Rewa, Feejee Islands.

This genus Blume distinguishes from Spondias by the imbricative
wstivation of the calyx, the united styles, and the remarkably de-
pressed, five-angled, and as if star-shaped, unarmed putamen of the

CONNARACE AZ. 8375

fruit. The specimen in the collection is a fruiting one: the foliage
answers well to the character of D. sylvestre, Blume: the fruit nearly
accords with the figure of Blume’s D. mangiferum. It is said by Dr.
Pickering (who mentions the plant under the name of Canariwm?
in his volume on the Races of Man, p. 336) to be sour and edible.

Orv, CONNARACE &.

1. ROUREA, Audi.
1. RovurEA HETEROPHYLLA, Planch.
Rourea heterophylla, Planchon, Prodr. Connar. in Linnea, 23, p. 419.
Has. Vanua-levu, Feejee Islands. (In fruit only.)

The plant characterized by Planchon, from Cuming’s Philippine
collection, is in flower only. Ours, which appears to be the same, has
the fruit. The young pods are usually solitary, oblong, glabrous,
scarcely more than 3 lines long, and not twice the length of the
calyx; but when full grown they are 6 or 7 lines long.

2, CONNARUS, Lann., Planch.
1. Connarus Pickerineu, Sp. Nov. (Tab. 45.)

C. ramulis foliisque junioribus pube ferruginea decidua tomentosis ;
foliolis 5-7 (rarius 3) lanceolato-oblongis nunc ovato-oblongis subacu-

376 PHANEROGAMIA.

minatis basi obtusis vel rotundatis, adultis glaberrimis, venis utringque
5-6 subtus prominentibus, rete tenwi transverso; paniculis folio multo
brevioribus calycibusque rufo-tomentosis ; pedicellis brevissimis ; petalis
extus tomentosis glanduloso-punctatis calyce triplo longwribus; folli-
culis obovatis turgidis tomento detersibilt primum vestitis, stipite petalis
longvore.

Has. Feejee Islands: on Ovolau, Rewa, and Vanua-levu.

Shrub apparently of considerable size; the younger branchlets with
the developing leaves, &c., tomentose with a rusty or reddish pubescence,
which is soon deciduous, except from the inflorescence. Leaves alter-
nate, pinnate, with 5 to 7 leaflets, or rarely only 3. The leaflets are
of a chartaceous texture, entirely glabrous when full grown, not shining,
from 3 to 62 inches long and 1! to 2 inches wide, varying from lan-
ceolate-oblong to ovate-oblong, most of them slightly acuminate, obtuse
or rounded at the base, on partial petioles of 3 lines in length: the
primary veins 5 or 6 pairs, prominent underneath, connected by
slender veinlets which form ¢ransverse reticulations. Vestiges of the
rusty pubescence often persist on the midrib underneath. Punicles
shorter than the leaves or even the petioles, axillary and terminal,
many-flowered, clothed with reddish wool. Pedicels very short. Calyx
a line and a half long, deeply five-cleft, woolly, like the inflorescence.
Petals spatulate, 5 lines long, more or less tomentose outside, and
sprinkled with orange-coloured glandular dots. Stamens 10; those of
both series bearing fertile anthers. Ovary and style ferrugineous-
pubescent. Pod obovate, turgid, an inch in length, excluding the
stipe, or contracted base (which is 3 lines long, exceeding the marces-
cent petals), and 7 or 8 lines wide, thick and coriaceous, clothed with a
rusty down, which rubs off at maturity, two-valved. Seed solitary,
conformed to the cell, attached above its base; its lower part embraced
by a lobed and somewhat lacerate arillus. Cotyledons fleshy, plano-
convex, conformed to the seed. Radicle short, superior.

This species appears to be most nearly related to Connarus muta-
bilis of Blume, a native of the Indian Archipelago; but it has more
tomentose branchlets, &c., larger and smoother leaflets, pubescent
petals, the stamens all with fertile anthers, and larger pods, the
surface of which is not transversely striate.

CONNARACE AZ. 377

Prats 45.—Connarus PicKERINGII: in flower and in fruit. Fig. 1.
A flower, enlarged. 2. A petal, more magnified. 3. Stamens and
pistil, magnified. 4. Pistil, with the ovary divided longitudinally, to
show the ovules, magnified. 5. An ovule, more magnified. 6. Seed,
with a valve of the pod, of the natural size. 7. The embryo, of the
natural size.

2. CONNARUS CrMmosus, Planch.

Connarus cymosus, Planchon, Prodr. Connar. in Linnza. 23, p. 430.

Has. Brazil, in the Organ Mountains, near Rio Janeiro. (In
flower only.)

3. ConNARUS Bryricui, Planch. 1. c.
Var.? foliolis bi-quadrijugis cum impari subito acuminatis venosioribus.
Has. Brazil; with the preceding.

The specimen (with young fruit only) well accords with an au-
thentic specimen of Planchon’s 0. Beyrichii, except that the leaflets,
instead of three in number, are from two to four pairs, abruptly
pointed, and less coriaceous, but more veiny.

38. BERNARDINIA, Planch.

1. BERNARDINIA FLUMINENSIS, Planch.
Bernardinia Fluminensis, Planchon, Prodr. Connar. in Linnea, 23, p. 413.
Has. Near Rio Janeiro, Brazil.

The specimen bears young fruit only: it appears to be identical
with the plant upon which Planchon established his genus Bernar-

dinia, in memory of Bernardin de St. Pierre.
95

Orv. LEGUMINOS

Susorv. I. PAPILIONACE &.
1 PODALYRIA, Lam.
1. PopALYRIA BIFLORA, Lam.

Podalyria biflora, Lam. Dict. 5, p. 445; Walpers, in Linnea, 13, p. 457.
P. subbijlora, DC. Prodr. 2, p. 102.

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.

2, CALLISTACHYS, Vent.
1. CALLISTACHYS LANCEOLATA, Vent.
Callistachys lanceolata, Vent. Hort. Malm. t. 115; DC. Prodr. 2, p. 104.

Has. Near Sydney, New South Wales.

2. CALLISTACHYS SPARSA, A. Cunn.

Callistachys sparsa, A. Cunn. ex Benth. Comm. Legum. in Ann. Wein. Mus. p. 5.
Oxylobium? Pultenece, DC. Prodr. 2, p. 104.

Has. Hunter’s River, New South Wales ;—where it was discovered
by Allan Cunningham.

LEGUMINOSE. 379

3. OX YLOBIUM, Andr.
1. OxyLopium? HAMULOSUM, Benth. ined.

O. folus confertis acerosis apice hamato-recurvis minute hispidulis ; ramis
apice paucifloris ; leguminibus calycibusque hirsutis.

Has. Hunter’s River, New South Wales.

The specimen, in fruit only, is that of a Heath-like, suffruticose
plant, a span or two in height; the branches clothed with a fine and
appressed pubescence. Leaves crowded, alternate, or often imperfectly
verticillate or opposite, acerose, with revolute margins, the acute apex
hooked, the surface minutely hispid under a lens, or at length glabrate :
they vary from 4 to 6 lines in length. The corymb is simple and /ew-
flowered at the end of the branches. Lobes of the fructiferous calyx
lanceolate from a broad base, acuminate, hirsute. Legume hirsute,
ovoid-oblong, pointed, half an inch long, including the persistent style,
without any partitions. The flowers are unknown; but the fruit is
that of an Oxylobium.—lt has not been identified in any other collection.

4. PODOLOBIUM, R. &r.
1. PopoLopium TRILOBATUM, R. Br.

Has. Near Sydney, and Cook’s River, New South Wales.

2. PopoLoBIuM OBOVATUM, Sp. Nov.

P. foliis obovatis subrotundisve retusis integerrimis subtus ramulisque
hirtello-puberis ; ovario hirsutissimo breviter stipitato.

Has. Near Sydney, New South Wales.

Apparently a procumbent or climbing plant, and most resembling

380 PHANEROGAMIA.

the P. scandens of DeCandolle, which Bentham (in Comm. Legum.
l.c. p. 6) has removed to Oxylobium, on account of its “sessile ovary
and numerous ovules.” But the ovary, in the single specimen of that
species which I possess, appears to be substipitate, and to contain only
6 or 7 ovules. The present plant has thicker and broader, obovate or
roundish leaves, mostly retuse, although mucronulate, all opposite and
entire; their lower surface, like the young branches, pubescent with a fine
and rather hirsute down. The flowers appear to be rather larger than
in P. scandens; the lobes of the calyx are less obtuse; the ovary, as
in that species, is very densely hirsute, and is raised on a manifest
although short stipe: and the ovules are only 6 or 7. Hence the
plant seems to be a Podolobium. The fruit is unknown.

5. CHOROZEMA, Labill.
1. CHorozEMA GomMPHOCARPUM, Benth. ined.
Has. New South Wales, in the vicinity of Newcastle.
The specimens accord with the plant from Bidwill’s Bay, so named
by Mr. Bentham, the characters of which are perhaps already pub-

lished. The species is a somewhat anomalous one, with rather the
habit of Oxylobium.

6 GOMPHOLOBIUM, Smith.
1. GOMPHOLOBIUM BARBIGERUM, DC.
Has. Near Newcastle, New South Wales. (Flowers considerably
larger than those of the following species.)
2. GOMPHOLOBIUM GRANDIFLORUM, Simith.

Has. Sydney, New South Wales. (In fruit.)

LEGUMINOSA, 39]

3. GOMPHOLOBIUM GLAUCESCENS, A. Cunn.

Gompholobium glaucescens, A. Cunn. in Field, N. S. Wales, p. 8346; Benth. Comm.

Legum. 1. c.
G. grandiflorum, B. setifolium, DC. Prodr. 2, p. 105.

Has. With the preceding. (Foliage only.)

4, GOMPHOLOBIUM LATIFOLIUM, Smith ?
Has. Sydney, New South Wales. (Foliage only, and hardly in a

determinable condition.)

5. GOMPHOLOBIUM VIRGATUM, Sieber.

Gompholobium virgatum, Sieber, Pl. Exsic. N. Holl. no. 8360; DC. Prodr. 2, p. 105.

Has. Sydney, New South Wales.

6. GOMPHOLOBIUM UNCINATUM, A. Cunn. ex Benth.

Has. New South Wales; with the preceding species.

7. GOMPHOLOBIUM GLABRATUM, DC.

Has. New South Wales; probably with the preceding species.

8. GOMPHOLOBIUM PINNATUM, Smith.

Has. Near Sydney, New South Wales. (In fruit only.)

The root appears to be that of an annual plant, as DeCandolle sus-

pected.
96

382 PHANEROGAMIA.

7. JACKSONIA, R. Br.
1. JACKSONIA SCOPARIA, R. Br.
Jacksonia scoparia, R. Br. in Hort. Kew. 3, p. 12; Benth. Comm. Legum. 1. ¢. p. 10.

Has. New South Wales, in the neighbourhood of Sydney.

8. DAVIESTIA, Smith.
1. DAVIESIA ACICULARIS, Smith.

Daviesia acicularis, Smith, in Linn. Trans. 9, p. 258; DC. Prodr. 2, p. 114.
Has. New South Wales; in the vicinity of Sydney and Woolon-
gong.
2. DAVIESIA ULICINA, Smith, 1. c.

Has. New South Wales, near Sydney and along Hunter’s River.

3. DAVIESIA CORYMBOSA, Smith, 1. ¢.

Has. New South Wales; probably from the neighbourhood of
Sydney.

9 VIMINARIA, Smith.
1. VIMINARIA DENUDATA, Smith.

Has. Woolongong, New South Wales. (In flower and fruit.)

LEGUMINOSS&. 393

10. PHYLLOTA, DC, Benth.
1. PHYLLOTA ASPERA, Benth.

Phyllota aspera, Benth. Comm. Legum. in Ann. Wien. Mus. l. ¢. p. 18.
Pulteneea aspera, Sieber, Pl. Exsice. N. Holl.; DC. Prodr. 2, p. 118.

Has. Sydney, New South Wales.

2. PuytioTA BAvERI, Benth. l. ¢.

Has. Sydney and Hunter’s River, New South Wales.

11. AOTUS, Smith.
1. Aotus vitLosa, Smith, DC.

Has. In the vicinity of Sydney, New South Wales.

12, DILLWYNIA, Smith.
1. DILLWYNIA ERICEHFOLIA, Smith.

Has. New South Wales, near Sydney; where it appears to abound.

2. DILLWYNIA PEDUNCULARIS, Benth.
Dillwynia peduncularis, Benth. Comm. Legum. in Ann. Wien. Mus. p. 14.

Has. Hunter’s River, New South Wales. (In fruit.)

384 PHANEROGAMIA.

3. DILLWYNIA PARVIFOLIA, R. Br.

Has. In the vicinity of Newington, New South Wales.

4, DILLWYNIA GLABERRIMA, Smith.

Has. Along Hunter’s River, New South Wales.

5. DILLWYNIA TENUIFOLIA, Sieber ?

Has. In the vicinity of Sydney, New South Wales.

138. SPADOSTYLES, Benth.
1. SpapostyLes SreBEr1, Benth.

Spadostyles Sieber’, Benth. Comm. Legum. in Ann. Wien. Mus. p. 17.
Pultencea euchila, DC. Prodr. 2, p. 112.

Has. Near Sydney, New South Wales. (Flowering specimens
only.)

14. PULTEN AA, Smith.
1. PULTENHZA DAPHNOIDES, Smith.

Has. Near Sydney, New South Wales; apparently abundant.

2. PULTENMA LINOPHYLLA, Smith.

Haz. New South Wales; with the preceding species.

LEGUMINOSZ. 385

3. PULTENZA RETUSA, Smith.

Has. New South Wales; with the two preceding species.

4, PULTENZA POLIFOLIA, A. Cunn.

Has. New South Wales: probably from the immediate vicinity of
Sydney. A variety with larger leaves, &c.

5. PULTENAA STIPULARIS, Smith.

Has. In the vicinity of Sydney, New South Wales.

6. PULTENHA ELLIPTICA, Smith.

Has. Near Sydney, New South Wales; very abundant.

7. PULTENZA BILOBA, R. Br.

Has. Near Sydney, New South Wales.

8. PULTENZA VILLOSA, Smith.

Has. In the neighbourhood of Sydney, New South Wales.

15. MIRBELIA, Smith.
1. MiIrRBELIA RETICULATA,’ Smith.

Has. Sydney and Hunter’s River, New South Wales. (In fruit.)
97

386

HAB.

Has.

Has.

HAs.

Has.

Has.

PHANEROGAMIA.

16. HOVEA, R. Br.
1. Hovea toneiroiia, R. Br.

In the vicinity of Sydney, New South Wales. (In fruit.)

2. HovEA LANCEOLATA, Sims.

Hunter's River, New South Wales. (In fruit.)

17. PLATYLOBIUM, Smith.
1. PLATYLOBIUM FoRMOSsUM, Smith.

In the vicinity of Sydney, New South Wales.

2. PLATYLOBIUM PARVIFLORUM, Smith.

New South Wales; probably with the preceding species.

18. BOSSI AA, Vent.
1. Bossl#A SCOLOPENDRA, Smith.

New South Wales; probably from the vicinity of Sydney.

2. BossL@A HETEROPHYLLA, Vent.

In the vicinity of Sydney, New South Wales.

HAs.

Has.

Has.

Has.

Has.

Has.

LEGUMINOSAZA. 387

3. BossLH#A MICROPHYLLA, Smith.

Sydney, New South Wales; with the preceding species.

19. RAFNIA, Thunb.
1. RAFNIA AXILLARIS, Thunb.

Cape of Good Hope, in the immediate vicinity of Cape Town.

20. BORBONIA, Linn.
1. BoRBONIA BARBATA, Lam.

Cape of Good Hope, in the immediate vicinity of Cape Town.

2. BoRBONIA LANCEOLATA, Linn.

Cape of Good Hope; with the preceding species.

3. BoRBONIA CORDATA, Linn.

Cape of Good Hope; with the two preceding species.

21. PRIESTLEYA, DC.
1. PRIESTLEYA SERICEA, DC.

Cape of Good Hope, in the immediate vicinity of Cape Town.

388 PHANEROGAMIA.

22. AMPHITHALEA, Eckl. & Zeyher.

1. AMPHITHALEA DENSA, Eckl. & Zeyher.

Amphithalea densa, Eckl. & Zeyh. Enum. Pl. Afr. Austr. p. 167; Benth. in Lond.
Jour. Bot. 2, p. 451.

Has. Cape of Good Hope.

23. HALLIA, Thunb.
1. Hara virgata, Thunb.

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.

2. HALLIA corDATA, Thunb.

| Has. Cape of Good Hope; with the foregoing species.

3. HALLIA IMBRICATA, Thunb.

Haz. Cape of Good Hope; with the two preceding species.

24. CROTALARIA, Linn.

1. Croratarta SENEGALENSIS, Bacle.

Crotalaria Senegalensis, Bacle, in DC. Prodr. 2, p. 183; Webb, Spicil. Gorg. in
Hook. Niger Fl. p. 116.

Has. St. Jago, Cape de Verde Islands.

LEGUMINOSAE. 389

2. CROTALARIA MICROPHYLLA, Vahl.

Crotalaria microphylla, Vahl, Symb. 1, p. 52; Webb, Spicil. Gorg. 1. ¢. p. 117.

Has. With the preceding. (The form named C. trigonelloides by
Mr. Webb.)

3. CROTALARIA STIPULARIA, Desv.

Crotalaria stipularia, Desy. Jour. Bot. 2, p. 76; DC. Prodr. 2, p. 124.
C. sagittalis, Velloz. Fl. Flum. 7, t. 111, non Linn.

Has. Rio Janeiro, Brazil. (In flower only.)

4, CROTALARIA STRIATA, DC.

Crotalaria striata, DC. 1.c.; Benth. in Lond. Jour. Bot. 2, p. 586, & Pl. Jungh.
p- 207.

Has. Rio Janeiro. Probably introduced into Brazil from Africa or
the East Indies. |

5. CROTALARIA VITELLINA, Jer.

Crotalaria vitellina, Ker. Bot. Reg. t. 447; DC. Prodr. 2, p. 132.

y

Has. Rio Janeiro, Brazil.

The Brazilian collection contains an imperfect and indeterminable
specimen of another species, of the same group as C. vitellina.

6. CROTALARIA INCANA, Linn.

Crotalaria incana, Linn. ; DC. Prodr. 2, p. 182; Benth. in Lond. Jour. Bot. 2, p. 587.

Has. Peru, in the immediate vicinity of Callao.
98

390 PHANEROGAMIA.

7. CROTALARIA VERRUCOSA, Linn.

Crotalaria verrucosa, Linn.; DC. Prodr. 2, p. 125; Benth. in Lond. Jour. Bot. 2,
p. 560; Wight, Ic. Pl. Ind. Or. t. 200.

Has. Tahiti, Society Islands; “common in some places near the

coast :” doubtless introduced; perhaps recently, as it is not mentioned
in Guillemin’s Zephyritis Taitensis. Bafios, near Manilla, Luzon.

8. CROTALARIA FULVA, Loud.

Crotalaria fulva, Roxb. Fl. Ind. 3, p. 266; Wight & Arn. Prodr. Fl. Ind. Or. 1,
p. 183.

Has. Singapore.

9. CROTALARIA CALYCINA, Schrank.

Crotalaria calycina, Schrank, Pl. Rar. Hort. Monac. t. 12; Benth. 1. c. p. 564.

CO. Roxburghiana, DC. Prodr. 2, p. 129.

O. anthylloides, Don, Prodr. Fl. Nep. p. 241; Wight & Arn. Prodr. Fl. Ind. Or. 1,
p. 181.

Has. Caldera, Mindanao, one of the Philippine Islands.

10. CROTALARIA SESSILIFLORA, Linn.

Crotalaria sessiliflora, Linn.; Benth. in Lond. Jour. Bot. 2, p. 555.
C. anthylloides, Lam. Dict. 2, p. 195, ex Benth. 1. ¢.

Has. Luzon; on Mount Majaijai.

11. CROTALARTIA QUINQUEFOLIA, Linn.

Has. Feejee Islands. “Emboa.” “Probably introduced.”

LEGUMINOSS. 39]

25. LUPINUS, Tourn.
1. Lupinus microcARPuS, Sims.

Lupinus microcarpus, Sims, Bot. Mag. t. 2418; Agardh, Syn. Gen. Lup. p. 2.

Has. Chili, near Valparaiso. (A form less hairy than usual.)

2. Lupinus Linpteyanus, Agardh.

Lupinus Lindleyanus, Agardh, Syn. Gen. Lup. p. 9.

Has. Peru, below Obrajillo.

A dwarf form of the species, a span to a foot high, more hirsute
than usual, or villous with long and spreading hairs, some of them on
the upper face of the leaflets. The flowers are said to be “ pale blue,
with a white carina.”

3. LUPINUS NUBIGENUS, Kunth.

Lupinus nubigenus, Kunth, Pl. Legum. p. 174, t. 50; H. B. K. Nov. Gen. & Spec-
6, p. 480; Hook. Bot. Mise. 2, p. 217; Agardh, l.c. p. 21.

Has. Andes of Peru, above Baiios. (In flower only.)

Our specimens are identical with Hooker’s plant, above-cited, which
was gathered in the same district by Mr. Cruckshanks. They accord
so nearly with specimens from Antisana that I cannot but refer them
to Kunth’s L. nubdigenus: but the stem is more developed, often a
span high; the short-pedunculate, very thick and dense, cylindrical
flowering spike is from 5 to 7 inches long; the upper lip of the calyx
is very deeply two-cleft, the lower tridenticulate, or more decidedly
three-toothed at the apex. Still the plant is not large enough to agree
with the character of L. alopecuroides, and the ovary is only four-
ovuled. ‘The leaflets, from 9 to 11 in number, are an inch and a half
in length. “ Flowers light blue.”

392 PHANEROGAMIA.

4, LUPINUS ToMENTOsUS, DC.
Lupinus tomentosus, DC. Prodr. 2, p. 409; Agardh, Syn. Gen. Lup. p. 34, ex char.

Has. Bafios and Culnai, Andes of Peru. Also collected by Mr.
M’Lean. (Nearly allied to the next.)

5. LuPINUS PANICULATUS, Desv.
Lupinus paniculatus, Desv. in Lam. Dict. 8, p. 625; Agardh, Syn. Gen. Lup. p. 35.

Has. Obrajillo and Bafios, Andes of Peru.

6. LupINus PRostRATUS, Agardh.

Lupinus prostratus, Agardh, Syn. Gen. Lup. p. 39.

Has. Culnai to Casa-cancha, in the high pampas or pasture region

of the Andes of Peru. (In fruit only.) It was collected by Mr. Mat-
thews in the same district.

7. Lupinus Pickerineu, Sp. Nov.

L. suffruticulosus, depressus, undique sericeus; caulibus brevissimis ;
petiolis gracilibus foliolis 5-7 oblanceolatis stipulisque multoties longio-
ribus; pedunculis recurvis apice capitato-plurifloris ; calyce subsessili
ebracteolato, labio superiore bipartito, inferiore integro.

Has. Bafios and Culnai, in the high Andes of Peru.

A dwarf and depressed, multicipital species, silky all over with a
rather hirsute appressed pubescence; the very short stems suffruticose
and thickish, branching directly from the crown of the long and deep
root, ceespitose-crowded, only 2 or 3 inches long. Leaflets 5 to 7,
oblanceolate, hirsute-silky both sides, only 3 to 5 lines long, very much

LEGUMINOS &. 393

shorter than the slender petioles; which are from an inch to 2 inches
in length. Stipules short, adnate to the base of the petiole, triangular-
subulate. Peduncles recurved or procumbent, thickish, naked, an inch
long, bearing 8 to 12 flowers in a head, composed of 2 or 3 closely
approximate verticils, at length forming a short spike. Bracts shorter
than the flowers, lanceolate, acute, deciduous. Calyx subsessile, not
bracteolate, densely: silky-hirsute; the upper lip two-parted, the lower
entire. Corolla 3 lines long; the colour not manifest. Ovary hirsute,
three-ovuled. Fruit not seen.—The species belongs to Agardh’s tribe
Microphylh.

8. Lupinus CRUCKSHANKSI, Sp. Nov.

L. nanus e radice perenni? gracili, subacaulis, crinito-hirsutissimus ;
foliolis T-11 lineari-lanceolatis superne nunc glabratis petiolo multo
brevioribus ; spica pluriflora foliis breviore ; calyce ebracteolato, labio
superiore bipartito, inferiore tridenticulato.

Has. Alpamarca; high Andes of Peru. (Cerro Pasco, Cruck-
shanks, in herb. Hook.)

A dwarf and nearly stemless species, 3 inches high, from a simple
and tapering, probably perennial root, all over very hirsute with long
and shaggy fulvous hairs. Petioles about an inch and a half in
length ; with the broad and acuminate-awned stipules adnate to their
base. Leaflets 7 to 11, linear-lanceolate, 6 to 8 lines long, much shorter
than the petioles, their upper surface becoming glabrate with age. Pe-
duncle (in flower) very short. Spike many-flowered, about an inch
long, rather dense, shorter than the leaves. Flowers subsessile, half an
inch long, exceeding the ovate and acuminate-awned bracts. The
corolla is said to be “ purplish” in Mr. Cruckshanks’ specimen. Calyx
not bracteolate, hirsute; the upper lip two-parted, the lower tridenticulate,
or truncate. Ovary linear, hirsute, with 3 or 4 ovules, Fruit not
seen.

9. Lupinus TrERmis, Forsk. ?
Has. Puen Buen, New South Wales. In fruit only. Doubtless

introduced from Europe.
99

394 PHANEROGAMIA.

26. ASPALATHUS, Linn.

1. ASPALATHUS CHENOPODA, Linn.

Aspalathus chenopoda, Linn. Spec. Pl. p. 711; Benth. in Lond. Jour. Bot. 7, .

p- 610.

Has. Cape of Good Hope, near Cape Town.

2. ASPALATHUS SPICATA, T'hunb.

Has. Cape of Good Hope; very abundant near Cape Town.

3. ASPALATHUS CANESCENS, Linn.

Aspalathus canescens, Linn. Mant. p. 262; Benth. in Lond. Jour. Bot. 7, p. 625.

Haz. With the preceding species.

4, ASPALATHUS SPINOSA, Linn.

Has. With the preceding species; very common.

5. ASPALATHUS ACUMINATA, Lam.
Aspalathus acuminata, Lam. Dict. 1, p. 287; Benth. in Lond. Jour. Bot. 7, p. 650.

Has. With the preceding.

6. ASPALATHUS MICROPHYLLA, DC.
Aspalathus microphylla, DC. Prodr. 2, p. 143; Benth. in Lond. Jour. Bot. 1. ¢.

Has. With the preceding.

LEGUMINOS&. 395

7. ASPALATHUS GALIOIDES, Linn.
Aspalathus galioides, Linn. Mant. p. 260; Benth. in Lond. Jour. Bot. 7, p. 647.

Has. Cape of Good Hope, with the preceding.

8. ASPALATHUS UNIFLORA, Linn.

Aspalathus uniflora, Linn. ex Benth. in Lond. Jour. Bot. 1. ¢. p. 613.
A. cymbeeformis, DC. Prodr. 2, p. 140.

Has. With the preceding.

9, ASPALATHUS CAPILLARIS, Benth.

Aspalathus capillarts, Benth. in Hook. Lond. Jour. Bot. 7. p. 653.
Ononis capillaris, Thunb. Fl. Cap. p. 585; DC. Prodr. 2, p. 166.

Has. Cape of Good Hope, in the vicinity of Cape Town.

27. ULEX, Lin.
1. Utex Evropaus, Linn.

Has. St. Helena: introduced, without doubt, from Europe.

28. SPARTIUM, Linn, DC.
1. SPARTIUM JUNCEUM, Lann.

Has. Peru: introduced from Europe. “ Abundantly naturalized
from Callao to Obrajillo.”

396 PHANEROGAMIA.

29. SAROTHAMNUS, Wamin.

1. SAROTHAMNUS SCOPARIUS, Wimm.

Spartium scoparium, Linn. Spee. Pl. ed. 2, p. 996; Fl. Dan. t. 313; Schk. Handb.
t. 196.
Cytisus scoparius, Link; DC. Prodr. 2, p. 154.

Has. Corral, and on the summit of the Pico Ruivo, Madeira.—The
common Broom. |

30. GENISTA, Linn., Lam.

1. Genista (TELine) Maprrensis, Webb.

Genista Maderensis, Webb, ex Lemann, Cat. Pl. Mader. in Hook. Niger Fl. p. 79.
Teline Maderensis, Webb & Berth. Phyt. Canar. 2, p. 37.

Has. Madeira; on mountains northeast of Funchal. (Very closely
resembling G. candicans.)

2. GENISTA virGATA, DC.

Genista virgata, DC. Prodr. 2, p. 149.
Spartium virgatum, Ait. Hort. Kew, 3, p. 11.
Cytisus tener, Jacq. Ic. Rar. t. 147.

Has. Madeira; on dry rocks, near Corral, &c.

3. GENISTA Cumineu, Hook. & Arn.

Genista Cumingii, Hook. & Arn. Bot. Mise. 3, p. 178; Gay, Fl. Chil. 2, p. 56.

Has. Andes of Chili, above Santiago; where it is very common.

LEGUMINOS &. 397

4, GENISTA GAYANA.
Genista wmbellata, Clos, in Gay, Fl. Chil. 2, p. 56, non Poir.

Has. High Andes of Chili, above Santiago, near the line of per-
petual snow.

Our specimens accord with the character of the G. wmbellata of the
Flora Chilena, the name of which must be changed, as it 1s borne by
a much older species; nor is that name appropriate for a plant which
has solitary and sessile flowers. It forms dense tufts, only a few
inches high; the stems imbricated throughout with the sheathing
bases of the small, trifid, rigid, and spinescent leaves. The pod
is ovate-lenticular, about three lines long, very smooth and shining,
one-seeded.

31. MEDICAGQO, Linn.
1. Mepicaco sativa, Linn.
Has. Rio Negro, North Patagonia: undoubtedly introduced from
Kurope.
2. MepicaGo LUPULINA, Linn.
Has. Chili; between Valparaiso and Santiago: doubtless intro-
duced.
8. MEDICAGO DENTICULATA, Willd.
Has. Rio Negro, North Patagonia: introduced with cattle. Now

pretty widely diffused through all the temperate regions of the New

World that were colonized from the south of Europe.
100

398 PHANEROGAMIA.

32. MELILOTUS, Tourn.
1. MELILOTUS PARVIFLORA, Desf.
Has. Rio Negro, North Patagonia. Obrajillo, Peru. Introduced

from Europe. Now abundant in all the temperate and subtropical
parts of the New World that were colonized by Portugal and Spain.

338. TRIFOLIUM, Tourn.
1. TRIFOLIUM FRAGIFERUM, Linn.

Has. Madeira; on the coast east of Funchal.

2. TRIFOLIUM FILIFORME, Linn.
Has. Bay of Islands, New Zealand. Probably of recent and slight

introduction, as it is not mentioned by Dr. Hooker, in his Flora of
New Zealand.

3. Trirotium Peruvianum, Vogel.
Trifolium Peruvianum, Vogel, in Reliq. Meyen. p. 12.

Has. Near Baiios, Andes of Peru.

4. Trirotium Marruewsu, Sp. Nov.

L. laxe hirsutum; caulibus procumbentibus; foliolis obovato-cuneatis
retusis argute serratis glabratis; stipulis membranaceis scariosis ;

LEGUMINOS &. 399

pedunculis folio triplo longioribus; floribus plurimis in wmbellam
congestis denique reflexis; dentibus calycis villosult subulato-acumo-
natis tubo longioribus ; legumine pilosulo trispermo.

Has. Near Obrajillo, Andes of Peru. (Chachapoyas, Matthews,
in herb. Hook.)

A species belonging, like the foregoing, to the group of which 7.
repens is the type, most allied apparently to TJ. obcordatum, Desv., of
Buenos Ayres (to which probably belongs 7. polymorphum of Seringe,
in DC., if not of Poiret); but it has narrower and barely retuse, not
obcordate, and sharply serrulate, obovate-wedgeform leaflets. From 7.
Peruvianum it is abundantly distinguished by its many-flowered,
umbelliform heads, nearly as large as those of TJ. repens, on peduncles
of thrice the length of the leaves (from 5 to 7 inches long), and by its
elongated procumbent stems, a foot or more in length, and perhaps
creeping. All the plant is softly and loosely hirsute; the leaflets how-
ever at length glabrate, or their upper surface glabrous. Petioles an
inch and a half in length. Stipules membranaceous, acuminate, 4 or 5
lines long. Pedicels reflexed after anthesis, a line and a half in length.
Flowers 4 or 5 lines long. Calyx sparsely hairy; the subulate-acum-
nate teeth longer than the tube, nearly equal, shorter than the scarious-
persistent corolla. Legume oblong, sessile, very obtuse at both ends,
sparingly and minutely pubescent, three-seeded, as long as the persistent
corolla. Seeds nearly orbicular, flat.

34. LOTUS, Linn.
1. Lotus @Laucus, Az.

Haz. Madeira; very common on the coast, near Funchal.

2. Lotus ANGUSTISSIMUS, Lann., var.
Lotus divaricatus, Soland. in Herb. Banks, ex Lemann, non Boiss.

Has. Madeira; at Santa Anna.

400 PHANEROGAMIA.

3. Lotus uLicinosus, Schkulhr ?

Has. Madeira, in company with the preceding species.

4, Lotus corNnIcuLAtus, Linn.

Has. Hunter’s River, New South Wales. (The Australian form
of the species, which appears not to differ essentially from the Euro-

pean plant.)
5. Lorus Fraser, Benth. ined.
Has. Hunter’s River, New South Wales.
The flowers are scarlet, according to the late Mr. Fraser (in herb.

Benth.), who long since collected this still unpublished species, in the
same region.

6. Lotus CANDOLLEANUS, Sweet.

Lotus Candolleanus, Sweet, ex Steud. Nomencl. Bot. ed. 2.
L. australis, 8. angustifoliola, DC. Prodr. 2, p. 212.

Has. Puen Buen, New South Wales.

30. HOSACKIA, Dougl.

1. HosacxiA Macrat, Benth.

Hosackia Macre, Benth. in Bot. Reg. sub. t. 1257.
Lotus (Microlotus) Macrei, Benth. in Linn. Trans. 17, p. 367.

Has. Chili: in ravines and on hills, around Valparaiso.

LEGUMINOS &. 401

36. DALEA, Linn.
1. Darea Onosrycuis, DC.
Dalea Onobrychis, DC. Prodr. 2, p. 247.

Has. Andes of Peru, below Obrajillo: a pubescent form of the
plant.

Both this and the nearly allied D. Mutisiz, Kunth, were gathered
by Matthews in the province of Chachapoyas, Peru. I cannot clearly
distinguish the two species, except by the calyx. WD. Mutisii has the
plumose-villous and subulate-aristiform teeth of the calyx longer than
the glabrous tube. In D. Onobrychis, the subulate teeth of the calyx
are shorter than the equally villous tube.

2. DALEA CYLINDRICA, Hook.
Dalea cylindrica, Hook. Bot. Mise. 2, p. 218.

Has. Obrajillo, Andes of Peru.

Described by Hooker from more developed specimens than ours,
with elongated spikes, gathered by Mr. Cruckshanks in the same dis-
trict, “ Valley of Canta.” Mr. M’Lean, however, found it with spikes
only an inch long, like those of our plant. Mr. Matthews likewise
collected it at Purrochuca.

37. PSORALEA, Jinn.
x Maderenses.
1. PsorALEA DENTATA, DC.

Psoralea dentata, DO. Prodr. 2, p. 221.
P. Americana, Linn. Spec. Pl. ed. 2, p. 1075; Jacq. Hort. Scheenb. t. 227.
101

402 PHANEROGAMIA.

Has. Madeira; on the coast at Funchal.—Not a native of America,
wherefore the Linnzean name was changed by DeCandolle.
2. PSORALEA BITUMINOSA, Linn.

Has. With the preceding species.

x * Capenses.

3. PSORALEA PINNATA, Linn.

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.
4, PSORALEA APHYLLA, Linn.

Has. With the preceding species.

5. PSORALEA DECUMBENS, A2t.

Has. With the preceding species.

x * * Australiane.

6. PsSoRALEA PATENS, Lindl.

Psoralea patens, Lindl. in Mitch. Exped. Austral. & Ann. Sci. Nat. ser. 2, 15, p. 59.
Has. Hunter’s River, New South Wales. Also gathered by Fraser
in the interior of New Holland.
7. PSORALEA TENAX, Lindl. 1. c.

Has. Hunter’s River and Puen Buen, New South Wales.

LEGUMINOSE. 403

x * * * Austro-Americane.
8. PSORALEA GLANDULOSA, Lin.

HAs. Valparaiso ; in valleys near the coast.—“ A shrub, from 6 to
10 feet high.”

9. PsoRALEA PUBESCENS, Bald.
Psoralea pubescens, Balb. in Pers. Ench. 2, p. 847; DC. Prodr. 2, p. 220.

Has. Peru: below Obrajillo; rare.—‘“‘A coarse, shrubby plant,
about 6 feet high.” This has been gathered above Lima by Mat-
thews, and near Loxa by Seemann. The specimens agree well with
the cultivated plant, which was doubtless of Peruvian origin. JP.
lasiostachys, Vogel, Rel. Meyen. p. 13, so far as can be judged from
the character, would seem to be hardly distinct from this species.

38. INDIGOFERA, Linn.
1. INDIGOFERA TINCTORIA, Linn.
Has. St. Jago, Cape de Verde Islands. Caldera, Mindanao, Philip-
pine Islands.
2. INDIGOFERA ANIL, Linn.

Has. Brazil, around Rio Janeiro; common. Tahiti, Society
Islands: “evidently introduced.” Sandwich Islands; on the hills
behind Honolulu, Oahu.

3. INDIGOFERA TRUXILLENSIS, A. B. 1,
Indigofera Truxillensis, H. B. K. Nov. Gen. & Spec. 6, p. 406.

Has. Peru: inthe dry bed of the river near Callao.

404 PHANEROGAMIA.

4, INDIGOFERA OBRAJILLENSIS, Sp. Nov.

L herbacea, cinereo-strigosa; caule decumbente ; folvis pinnatis ; foliolis
6-8-jugis obovatis oblongisve mucronatis utrinque canescenti-strigosis
supra subglabratis ; racemis breviter pedunculatis folio brevioribus ;
mucrone antherarum nudo; legumine deflexo lineari recto canescenti-

puberulo.
Has. Obrajillo, Peru; abundant. (Also gathered by Mr. M’Lean.)

Apparently an herbaceous perennial, cinereous throughout with the
close strigose pubescence of the related species. Stems decumbent or
spreading, a foot or more in length. Stipules subulate-awned, 5 lines
long, or the lower broader and shorter, brownish, strigose-pubescent
externally, rather persistent. Leaves pinnate, very short-petioled.
Leaflets 6 to 8 pairs, obovate or oblong, from 4 to 8 lines in length and
about 3 lines wide, mucronate, some of them rather conspicuously so,
canescently strigose both sides, at least when young, the upper surface
becoming more or less glabrate with age, veinless, slightly petiolulate.
Peduncles (half an inch to an inch or rather more in length), with the
short and dense spike or raceme, shorter than the leaf. Pedicels less
than half a line long. Bracts subulate, as long as the calyx. Calyx
strigose-hirsute; the lobes subulate-acuminate, longer than the tube.
“Corolla scarlet, showy :” vexillum 4 or 5 lines long, a little hairy
outside. Anthers uniform, strongly mucronate; the mucro naked (not
bearing a tuft of short bristles). Legume deflexed, linear, an inch and
a half long, straight, canescently puberulent, pointed, turgid, several-
seeded.

Our specimens are in flower only: the fruit is described from a
plant in the Hookerian herbarium, gathered by Mr. M’Lean. The
species is manifestly allied to /. tephrosioides, H. B. K.; but the leaflets
are broader, strigose-canescent both sides, and destitute of the fine
parallel veins which are delineated in Kunth’s figure of that species,
nor have they a long and awn-like mucronation; the peduncles are
shorter; and there is no trace of a tuft of bristles surmounting the
cusp of the anthers. It needs, however, to be more critically com-
pared with that species,

LEGUMINOSS. 405

5. INDIGOFERA viscosa, Lam.

Indigofera viscosa, Lam. Dict. 3, p. 247; DC. Prodr. 2, 227; Webb. Spicil. Gorg.
p. 121.
I. glutinosa, Perr. in DC. 1. ¢. non Vahl.

Has. St. Jago, Cape de Verde Islands: very abundant in dry
places.

6. INDIGOFERA LINEARIS, DC.

Has. Cape de Verde Islands; with the preceding species.

7. INDIGOFERA AUSTRALIS, Willd.

Has. Hunter's River, New South Wales.—A large form: also what
appears to be a much smaller and more slender variety: both in fruit

only.

8. INDIGOFERA ADESMIAFOLIA, Sp. Nov.

L fruticosa, ramosissima, glabra; foliis pinnatis ; rhachi complanata
articulata ; foliis 6-T-jugis parvis obovatis emarginatis vel fere obcor-
datis: nedunculis paucifloris; legumine lineari recto tetragono pleto-

3 2

spermo.
Has. Hunter’s River, New South Wales.

A shrubby plant, apparently erect, very much branched, glabrous
throughout or very nearly so; the branches slender. Stipules minute
and deciduous. Leaves pinnate, with 6 or 7 pairs of obovate, emargt-
nate, nearly obcordate leaflets, which are only a line or a line and a half
in length, of a thickish texture, smooth on both sides. Petiole and
rhachis flattened and somewhat dilated, an inch and a half long; artv-
culated at the insertion of the leaflets, where it is apparently nodose-

glandular. Peduncles mostly shorter than the leaves, apparently /ew-
. 102

406 PHANEROGAMIA.

flowered: but the specimen bears ripe fruit only. Legume linear,
straight, four-sided, an inch long, glabrous, several-(9-10-) seeded.

I have not seen this remarkable species in any other collection, nor
do I know any with which it may be particularly compared.

9. INDIGOFERA FILIFORMIS, Thunb.

Has. Cape of Good Hope, in the vicinity of Cape Town.

10. INDIGOFERA CORTACEA, Ait.

Has. Cape of Good Hope; with the foregoing species.

39. GLYCYRHIZA, Tourn.
1. GLYCYRHIZA ASTRAGALINA, Gillies.

G. foliolis oblongo-linearibus ; racemis laxifloris folia superantibus ; legu-
mine inerme leevi subglanduloso pseudo-lomentaceo moniliformi ; semi-
nibus fere globosis.

Glycyrhiza astragalina, Gillies, in Hook. Bot. Mise. 3, p. 183.

Has. Rio Negro, North Patagonia; common. (Also gathered at
Port Desire, by Mr. Darwin and Captain Middleton; Port St. Helena,
Captain King; and Bahia Blanca, Mr. Tweedie; specimens of the
latter, like our own, with fruit.)

Roots or creeping rootstocks with the taste of Liquorice. Stems a
span to a foot high, branching, and, like the whole plant, minutely
glutinous-glandular, otherwise glabrous. Stipules subulate, deciduous.
Leaflets 9 to 13, oblong-linear, retuse and mucronulate, thickish, 6 to
10 lines long. Racemes loosely-flowered, peduncled, usually exceeding
the leaves. Calyx campanulate, rather gibbous. “Corolla purple,”

LEGUMINOSE. 407

resembling that of the other species, 3 lines long. Stamens diadel-
phous. Ovules usually 3. Legume smooth, minutely glandular, moni-
liform, so constricted between the joints as to appear lomentaceous,
commonly linear, and of three turgid, one-seeded joints, when it is 8 or
10 lines long, sometimes shortened and of one or two joints, not in the
least prickly or hispid, tardily if at all dehiscent. Seeds subglobose,
with a small hilum.

The fruit of this plant has not before been described. Its lomen-
taceous form is peculiar, and the seeds are nearly spherical ;—cha-
racters, however, insufficient to separate it from Glycyrhiza, of which
it is the sole known representative in the southern hemisphere.

40. TEPHROSIA, Rrs.
1. TEPHROSIA BRACTEOLATA. Guill. & Perr.
Tephrosia bracteolata, Guill. & Perr. Fl. Seneg. p. 194; Webb. Spic. Gorg. p. 121.

Has. St. Jago, Cape de Verde Islands. (A diminutive specimen.)

2. TEPHROSIA PISCATORIA, L7s.
Tephrosia piscatoria, Pers. Ench. 2, p. 329; DC. Prodr. 2, p. 252; Guill. Zeph.
Tait. p. 62.
Galega littoralis, Forst. Prodr. Fl. Ins. Austr. p. 52, non Linn.
Has. Sandwich, Society, Samoan, Tonga, Feejee, and Coral
Islands: “intrusive.” Also Luzon, near Manilla.

3. TEPHROSIA HIRTA, Hamild.

Tephrosia hirta, Hamilt. in Trans. Linn. Soc. 13, p. 5467

Has. Singapore.

408 PHANEROGAMIA.

4, TrepHrostA BAvERI, Benth. ined.

T. herbacea, glabella; caulibus decumbentibus; stipulis aristato-subu-
latis; foliolis 3-4yugis lato-linearibus obtusissimis mucronatis basi
acutiusculis subtus striguloso-puberulis ; racemis paucifloris folio sub-
longioribus ; legumine lineart puberulo apice arcuato-incurvo.

Has. Hunter’s River, New South Wales. (New South Wales,
Ferd. Bauer.)

A low, herbaceous species, almost glabrous; with branching, ascend-
ing or decumbent stems from a perennial root; the younger parts more
or less puberulent. Stipules subulate-awned, 2 lines long, persistent.
Leaflets T or 9, broadly linear, 7 to 12 lines long and about 2 lines
wide, very obtuse, mucronate, acutish at the base, glabrous above,
minutely strigose-puberulent underneath; the straight veins not con-
spicuous. Racemes few-flowered, rather exceeding the leaves ; the flowers
scattered. Calyx-teeth subulate-setaceous. Legume linear, an inch
and a half long, 2 lines wide, incurved towards the apex, compressed,
8—10-seeded, puberulent.

5. TrepurostaA CAPENsIs, [e7's.

Has. Cape of Good Hope, in the vicinity of Cape Town.

41. AGATI, Rheede.
1, AGATI GRANDIFLORA, Desv.
Has. Oahu, Sandwich Islands. “Cultivated, but not naturalized.”
The A. coccinea, now considered to be no more than a variety of A.

grandiflora, seems not to have been met with at the Society Islands,
where it was originally collected by Forster.

LEGUMINOSA. 409

9. Agatt tomeNTosA, Nutt. in Herb, Hook. (Tab. 46.)

A. sericeo-tomentosa, demum glabrescens; caulibus reclinatis ; foliolis
8-12-iugis supra glabris; calyce cyathiformi ; leguminibus teretius-
culis torulosis folia cequantibus apice longe rostratis.

Sesbania tomentosa, Hook. & Arn. Bot. Beech. Voy. p. 286, excl. loco natali.

Has. Sandwich Islands. District of Waianae, Oahu. Coast of
Hawaii, east of the crater Lua Pele. Also collected, principally on
Oahu, by Lay and Collie, Douglas, Gaudichaud, Nuttall, &.

A woody plant, with decumbent stems, “from 15 to 20 feet in
length ;’ the branches, foliage, &., stlky-tomentose when young, but
glabrate with age. Leaves abruptly pinnate, from 4 to 7 inches long,
very short-petioled. Leajlets 8 to 12 pairs, oblong-elliptical, from 7 to
14 lines long, 3 or 4 lines wide, pale, minutely black-dotted and gla-
brous above, minutely silky-pubescent underneath even when old,
retuse, usually more decidedly mucronate than those of A. grandi-
flora, otherwise very similar, except as to the pubescence; the rhachis
pubescent. Stipules minute, caducous. Inflorescence as in A. grandt-
flora: the peduncles short, about three-flowered. Pedicels half an
inch to an inch long, mostly longer than the peduncle. Bracts and
bractlets minute, caducous. Flowers one-third or less than half the
size of those of A. grandiflora, and with a proportionally shorter
cyathiform calyx, which, like the pedicels, &c., is minutely silky-
pubescent; the broad teeth subulate-pointed. Corolla “ yellow or
red, as brightly coloured as in A. grandiflora ;” the moderately falcate
carina and ale from an inch to an inch and a quarter long: vexillum
orbicular, emarginate, reflexed, when spread out little shorter than
the carina. Stamens and pistil nearly as in A. grandiflora. Legume
nearly terete, at least when mature, although compressed when young,
5 or 6 inches long, 14 to 2 lines wide, glabrous, usually torulose, and
occasionally constricted at intervals, slightly stipitate in the nearly
persistent calyx, conspicuously rostrately pointed, multilocellate, de-
flexed, as long as the leaves. Seeds from 8 to 20, oblong, compressed,
shining.

103

410 PHANEROGAMIA.

This is ‘a manifest congener of Agati grandiflora; but the abbre-
viated calyx leaves scarcely any positive character to distinguish the
genus from Sesbania. The specimen described by Hooker and Arnott,
in the Botany of Beechey’s Voyage, as from Acapulco, was doubtless
collected at the Sandwich Islands, and accidentally misplaced among
Mexican plants.

PLATE 46.—AGATI TOMENTOSA: a branch, in flower and with young
fruit, of the natural size. Fig. 1. A corolla, displayed, somewhat
enlarged. 2. Stamens and pistil, more magnified. 38. The calyx and
pistil, magnified. 4. Mature pods, of the natural size. 65. Part of
one of the valves of the pod, with the seeds in place. 6. Embryo,
enlarged.

42, SWAINSONA, Salasb.
1. SWAINSONA GALEGIFOLIA, R. Br.

Swainsona galegifolia, R. Brown, in Ait. Hort. Kew. ed. 2, 4, p. 8326; DC. Prodr.
2, p. 271.

Has. Hunter’s River, New South Wales. (In flower.)

2. SWAINSONA CORONILLZFOLIA, Salisb.

Swainsona coronillefolia, Salisb. Parad. Lond. no. 28; Sims, Bot. Mag. t. 1725.

Has. Hunter's River, New South Wales. (In flower and with
fruit.)

3. SWAINSONA MICROPHYLLA, Sp. Nov.

y . ° e . . « he

S. glabella, suffruticulosa ; foliolis 6-12-jugis obcordato-cuneatis ; racemis

multifloris folia superantibus; leguminibus deflexis ovoideis coriaceis
glabris estipitatis rostro brevi abrupto incurvo apiculatis.

LEGUMINOS &. All

Has. In the district of Hunter’s River, and Puen Buen, New South
Wales.

Plant a span or more in height, almost glabrous to the naked eye,
under a lens minutely strigose-puberulent. Stems a little woody at the
base, slender, erect, branched. Stipules subulate, minute. Leaves
pinnate. Leaflets 6 to 12 pairs, obcordate, or cuneate, and deeply
emarginate, from a line to a line and a half (seldom 2 lines) in
length. Racemes longer than the leaves, 5 or 6 inches long in fruit,
including the peduncle, many-flowered. Flowers only 3 lines long.
Pedicel recurved, longer than the bract, shorter than the nearly gla-
brous calyx; the teeth of the latter much shorter than the tube.
Vexillum not callose. Ovary nearly glabrous. Legume ovoid, deflexed
glabrous, coriaceous in texture, very turgid, 5 lines long, sessile in the
calyx (not stipitate), abruptly pointed with a short incurved beak, tew-
seeded.

4, SWAINSONA MONTICOLA, Benth. ined,
Has. Near Puen Buen, New South Wales.

This is the Lessertia monticola of Richard Cunningham’s collection,
and is an undescribed Stwainsona. Our collection does not furnish
adequate materials for characterizing it.

43. CARMICH ALIA, AR. Br.

1. CARMICHALIA AUSTRALIS, £2. Br.

Carmichelia australis, R. Brown, in Bot. Reg. t. 92; Hook. f. Fl. N. Zeal. p. 50.
C. Cunninghamii, Raoul, Pl. N. Zel. p. 29, t. 28, B.

Bossicea scolopendria, A. Rich. Bot. Voy. Astrolab. p. 346, non R. Brown.

Lotus arboreus, Forst. Prodr. Fl. Ins. Austr. p. 92.

Has. Bay of Islands, New Zealand. (In fruit only: the var. gran-
_diflora, Hook. f.)

412 PHANEROGAMIA.

44. ASTRAGALUS, Linn., Benth.

AstraGcaLus & Paaca, Linn. (excl. sp.), DC., Endl., ete.

1. ASTRAGALUS DISTANS, Sp. Noy.

A. cinereo-pubescens ; radice annua; caulibus a bast ramosis gracilibus ;
stipulis parvis a petiolo liberis inter se subconnatis; foliolis 9-13
angustissime linearibus obtusis; pedunculo folium superante laxe
multifloro; dentibus calycis subulatis tubo paullo brevioribus ; corolla
ochroleuca; legumine deflexo ovoideo-oblongo cinereo-pubescente apice
acuto incurvo polyspermo uniloculari, suturis utrisque leviter intro-

fleais.
Var.? B. legumine majore recto, suturis vix intrusis.
Has. Rio Negro, North Patagonia; on sand-hills.

Root annual. Stems branched from the base, slender, diffusely
spreading or ascending, a foot or more in length, flexuous, cinereous,
as is the whole plant, with a fine strigulose pubescence. Stipules
small, scarious, ovate-subulate, free from the petiole, more or less united
with each other on the opposite side of the stem. Leaves 12 to 2
inches long, including the short petiole. Leaflets 9 to 13, very nar-
rowly linear, half an inch to an inch long and only half a Ime wide,
plane, or the margins involute when young, obtuse, the upper surface
becoming almost glabrous with age. Peduncles axillary, exceeding the -
leaves, 2 or 3 inches long, bearing a loose, many-flowered, spike-like
raceme, of an inch or two in length. Pedicels shorter than the calyx,
recurved, as long as the subulate bractlets. Flowers 3 or 4 lines
long. Calyx oblong-campanulate, hirsute-pubescent with whitish
and a few brownish hairs; its subulate teeth rather shorter than the
tube, about half the length of the apparently yellowish-white corolla.
Legume deflexed, ovoid-oblong, not more than 5 lines long, 2 lines wide,
nearly terete, abruptly acute, not at all stipitate nor tapering at the
base, cinereous-pubescent like the stems, more or less incurved at the

LEGUMINOSZ. 413

apex, many-seeded, one-celled, but with both sutures projecting a little
into the cell; the dorsal one not externally appearing introflexed, but
the section shows a distinct infolding of the coriaceous endocarp.

The specimens of the variety, if such it be, are in advanced fruit,
and show the root. Their stems, foliage, &c., are similar to those
of the species to which it is annexed, only rather more glabrate.
But the /egumes are larger, from 5 to 7 lines long, and straight or
nearly so, less pointed; the sutures very slightly introfleced; but the
projection of the dorsal one is still observable. Additional specimens
are requisite to determine whether they belong to a distinct species.

Specimens of a nearly allied species, if not the very same, exist in
the Hookerian herbarium, from Uruguay. This adds another to the
numerous species, of various sections, which so invalidate the distinc-
tion between Phaca and Astragalus as to compel their union, although
the vast number of known species would render the separation
extremely desirable, if that were longer possible. Mr. Bentham has
already combined the two genera in his as yet unpublished account
of the Indian <Astragali; and I have followed his example in the
second part of Plante Wrightianz.

2. ASTRAGALUS (PHACA) OCHROLEUCUS.
Phaca ochroleuca, Hook. & Arn. Bot. Misc. 8, p. 186; Gay, Fl. Chil. 2, p. 95.

Has. Chili, near Valparaiso. (In fruit only.)

3. ASTRAGALUS (PHACA) CANESCENS.
~ Phaca canescens, Hook. & Arn. Bot. Mise. 8, p. 185; Gay, Fl. Chil. 2, p. 100.
Has. Chili; with the preceding. (Without flowers or fruit.)

This specific name may be retained, since the A. canescens of
Solander, published by Mr. Lowe, has been ascertained by Mr. Webb
to be identical with A. brachyceras, Ledeb., and no more than a variety

of A. hamosus.
104

414 PHANEROGAMIA.

4. ASTRAGALUS PROCUMBENS, Hook. & Arn. l. c.

Has. Chili; with the preceding species. (Without flowers or fruit.)

5. ASTRAGALUS GARBANCILLO, Cav.

Astragalus Garbancillo, Cay. Ic. Pl. 1, t. 85; DC. Prodr. 2, p. 283; Hook. Bot.
Misc. 2, p. 214.

Has. High Andes of Peru, at Bafios; abundant. (In flower, with-
out fruit.)

The stems are herbaceous, and the calyx black with a dark pubes-
cence, as in Cruckshanks’ plant, described by Hooker: but the leaflets
are not bidentate, and seldom even retuse. Except in being less
hairy, it seems not distinct from A. unifultus, L’Her. If I rightly
identify the plant, the flowers are said in Dr. Pickering’s notes to he
cream-colour.

6. AstrAGALUS (PHAca) Rica, Sp. Nov.

A. hirsuto-canescens, humilis; caulibus e basi suffrutescente adscenden-
tibus; stipulis oppositifoliis coalitis ; foliolis 15-21 oblongo-linearibus ;
pedunculo folium adequante; spica brevi pauciflora; dentibus calycis
nigro-hirsuti subulatis tubo cequilongis; leqgumine ovoideo chartaceo
inflato glabello acumine subincurvo apiculato uniloculari polyspermo,
suturis haud intrusis.

Has. Andes of Peru, below Obrajillo.

A low, canescently hirsute plant, with many slender stems, ascending
from «@ slightly woody base, a span or less in height. Stipules united
opposite the leaf, small. Leaves about an inch and a half long, in-
cluding the short petiole. Leaflets 15 to 21, oblong-linear or linear,
mostly obtuse, 3 lines long, both sides clothed, like the stems, with
whitish hirsute pubescence. Peduncle as long as the leaf, bearing a

LEGUMINOSAE. 415

close and short spike, of from 6 to 12 flowers. Pedicels much shorter
than the calyx, about the length of the subulate bracts. Calyx clothed
with black hairs, the subulate teeth as long as the campanulate tube.
Corolla probably purple or white, 3 lines long. Legume ovoid, inflated,
of a papery texture, half an inch in length, 3 or 4 lines in transverse
diameter, narrower in the direction of the sutures, nearly glabrous,
under a lens sparsely and minutely hairy, closely sessile in the calyx,
pointed, the acumination somewhat incurved, many-seeded, strictly one-
celled ; the sutures not projecting into the cavity of the cell.

7. AsTRAGALUS PickERINGI, Sp. Nov.

A. multiceps, subacaulis, depressus ; stipulis vaginantibus imbricatis ;
foliis cano-villosis ; foliolis 19-23 obovato-oblongis emarginatis ; pedun-
culo petiolo breviore capitatim 3—-4-floro ; calycis dentibus triangulart-

~ subulatis cequalibus tubo cylindraceo dimidio brevioribus; ovario
tomentoso substipitato quinque-ovulato.

Has. High Andes of Peru, between Casa Cancha and Culnai.
(Also Cerro Pasco, Matthews.)

A depressed, and multicipital, nearly stemless species; the caudexes
or short stems ceespitose, an inch or two in length, from a perennial
or somewhat lignescent root, clothed with the sheathing stipules, which
are closely imbricated on the upper, leafy portion. The stipules are
scarious, 2 or 3 lines long, more or less silky-villous, united on the
side opposite the leaf almost to their summits, and behind the petiole
to a less extent, inclosing the base of the latter, with which it is
adnate only at the very base. Leaves canescently villous, as are the
peduncles, calyx, &c., 14 to 2 inches long, including the petiole of
about one-third the length. Leaflets 19 to 23, obovate-oblong and
emarginate, or almost obcordate, silky-villous both sides, somewhat
complicate, 2 or 3 lines long, approximate, but not closely compacted.
Flowers 3 or 4 together in a kind of head, on a peduncle shorter than
the petiole, which is at first nearly included in the stipular sheaths, at
length exserted to the length of 3 or 4 lines: the separate flowers
fully half an inch long, nearly sessile, subtended by ovate and mem-
branaceous bracts of scarcely one-fourth their length. Calyx cylin-

416 PHANEROGAMIA.

draceous, more than half the length of the (purplish?) corolla, silky
with white and a few dark hairs; the teeth subulate-triangular, equal,
only half the length of the tube. Ovary tomentose, elongated-oblong,
one-celled, but with some introflexion of the dorsal suture, short-
stipitate, about five-ovuled. Fruit not seen.

Although allied to A. geminiflorus of Humboldt and Bonpland,
from the Quitensian Andes, this species is distinguished by its pro-
portionally larger leaves, with less crowded, broader, and emarginate
leaflets; its flowers three or four together, and raised on a short but
manifest peduncle; its longer and regularly five-toothed calyx, &c.
The number of ovules in that species is not mentioned.

8. ASTRAGALUS BRACKENRIDGE, Sp. Nov.

A, subacaulis vel procumbens; stipulis vaginantibus ramorum novel-
lorum imbricatis; foliis canescenti-villosis; foliolis 21-27 oblongis
obtusis supra glabellis ; spica 6-10-flora breviter pedunculata petiolum
adequante; calycis nigro-pubescentis dentibus subulatis tubo cam-
panulaty subeequilongis ; legumine ovoideo-triangulari apiculato dorso
sulcato biloculart, loculis monospermis.

Has. High Andes of Peru, near Bafios.

Plant with the same aspect and mode of growth as the preceding
species, unless that the caudexes or short procumbent stems are more
elongated (the proper root not seen): these are apparently numerous
in a crowded tuft, and are terminated by a dense cluster of the leafy
shoots of the season, which are about an inch in length, and imbri-
cated with sheathing stipules, like those of the foregoing species. Leaves
canescently villous with finer and more spreading hairs, 13 to 2 inches
long, including the petiole. Leaflets 21 to 27, narrowly oblong, obtuse,
smoothish above, mostly conduplicate, 2 to 3 lines long, not very
crowded. lowers 6 to 10, in a spike, on a short but distinet peduncle,
together about the length of the petiole. Calyx blackish-pubescent, a line
and a half in length, subtended by a shorter and similarly pubescent
bract; its broadly subulate teeth nearly as long as the campanulate. tube.
Corolla apparently purple or white, 3 lines long. Ovary silky-pubes-

LEGUMINOSA. 417

cent, 2-3-ovuled. Legume triangular-ovoid, closely sessile in the
calyx, 3 lines long, deeply grooved on the dorsal side, thence appearing
somewhat didymous, not grooved on the ventral face, minutely pounted,
slightly pubescent, completely ¢wo-celled by the introflexion of the
dorsal suture; the cells one-seeded. Seeds large for the size of the cell.

The much smaller flowers with a shorter calyx, and the entire
leaflets, at once distinguish this species from the foregoing, which
probably bears a different fruit.

9. AsTRAGALUS ALPAMARCA, Sp. Noy.

A. pygmeus, ceespitoso-acaulis; stipulis vaginantibus ; foliis confertis-
simis minimis; foliolis 11-21 ovalibus emarginatis obcordatisve villo-
sulis seu glabellis ; floribus solitariis binisve sessilibus ; calycis dentibus
triangulato-subulatis tubo campanulato paullo brevioribus; ovario
biloculari, loculis uniovilatis.

Has. Near Alpamarca, on the high Andes of Peru.

A very dwarf, densely coespitose, and stemless species, forming a
matted tuft, only an inch or two in height, from lignescent caudexes ;
the condensed branches covered with the smoothish sheathing stipules,
and with densely crowded, minute leaves; the latter only a quarter or
half an inch in length, including the petiole. Leaflets 11 to 21, oval
and notched at the apex, or obcordate, from half a line to scarcely a
line in length, of a rather fleshy texture, conduplicate, the upper
surface mostly glabrous, the lower more or less villous with loose hairs,
or some of them glabrate. Flowers solitary or in pairs, sessile or
nearly so at the end of the branches, small. Calya-tube campanulate,
a line and a half long, sparsely hairy, rather longer than the triangular-
subulate-teeth; the latter spreading. “Corolla blue,” according to Dr.
Pickering’s notes: vexillum fully twice the length of the calyx,
rounded-obovate, retuse, very large in proportion to the other petals.
Stamens diadelphous from the middle downwards. Ovary two-celled,
silky-villous, with a single ovule in each cell. Fruit not seen.—An
imperfect specimen, from the same habitat, has larger and less con-

gested leaves, and is perhaps a more developed state of the species.
105

418 PHANEROGAMIA.

This species is closely allied to A. uniflorus, DC. and A. gemini-
florus, Humb. & Bonpl.; but is much smaller in all its parts, has
different leaflets, &c. It appears to be equally distinct from the A.
Peruvianus, A. minimus, and A. pusillus of Vogel, which are described
as having the tube of the calyx three or four times the length of the
teeth.

10. ASTRAGALUS ALIENUS, Sp. Nov.

A. humilis, fruticulosus, multiceps, cinereo-pubescens ; stipulis vaginan-
tibus; foliis confertis; petiolis induratis persistentibus ; foliolis multi-
jugis ellipticis retusis ; floribus geminis subsessilibus ; calycis dentibus
tubo oblongo-campanulato suwbdimidio brevioribus; ovario quinque-
ovulato uniloculari.

Has. High Andes of Peru; on the summit of the ridge at Baiios;
rare.

Plant /ow (3 or 4 inches high), rigid, fruticulose, scarcely caules-
cent; the thick and woody root dividing into a tuft of caudex-like
branches, which below are beset with the persistent stipules, above
thickly armed with spinescent and persistent indurated petioles, from
which the leaflets have fallen. Leaves crowded, cinereous-pubescent,
an inch long. Stipules sheathing, much less united on the petiolar
side, where they are adnate to the base of the petiole only, but united
on the opposite side nearly to their apex, pubescent like the leaves.
Leaflets im many pairs (12 to 15), elliptical, retuse, usually complicate,
about a line and a half in length, the upper surface less pubescent.
Flowers in pairs at the apex of the stem, or in the upper axils, almost
sessile, half an inch long. Calyx oblong-campanulate, pubescent with
short dark-coloured hairs; the triangular-subulate teeth nearly half the
length of the tube. Corolla twice the length of the calyx, probably
purple. Ovary oblong-linear, silky-villous, one-celled, five-ovuled, some-
what stipitate. Legume unknown.

This would appear to belong to the same group as several of the
foregoing species; but it exhibits the spinescent persistent stipules of
the Zragacanthe,—in which respect it is unlike any other known

species of the New World.

LEGUMINOSAE. 419

45. VICIA, Linn.
1. Victa Setxor, Vogel.
Vicia Selloi, Vogel, in Linnea, 13, p. 833; Walp. Repert. 1, p. 717.

Has. Banks of the Rio Negro, North Patagonia.

This is without doubt the plant described by Vogel under this
name. Perhaps it is not distinct from V. graminea of Smith (a
species omitted in DeCandolle’s Prodromus) : but that is said to have
three pairs of leaflets and very small flowers.

2. VictA NIGRICANS, Hook. & Arn.

Vicia nigricans, Hook. & Arn. Bot. Beech. Voy. 1, p. 20, & Bot. Mise. 3, p. 195.

Has. Chili, near Valparaiso.

3. VICIA BIDENTATA, Hook.

Vicia bidentata, Hook. Bot. Mise. 2, p. 215.
V. depauperata, Clos, in Gay, Fl. Chil. 2, p. 182%

Has. Peru, near Callao; among grass, in rather dry places.
‘“‘ Flowers blue.”

4, Victa Martruewsi, Sp. Nov.

V. parce molliter pubescens; caulibus debilibus; fohiolis 6-12 anguste
oblongis venosis obtusissimis vel truncato-retusis mucronulatis ; stipulis
semisagittatis lobatis; pedunculis quadrifloris folium cequantibus ;
calycis dentibus subulatis tubo eequilongis, superioribus parum mino-
ribus; leguminibus glabris.

420 PHANEROGAMITIA.

Has. Andes of Peru, near Bafios and Obrajillo. (Chachapoyas
and Purruchuca, Andes of Peru, Matthews.)

A low and slender species, with weak stems, sparsely and softly
pubescent throughout, or the foliage at length glabrate. Leaflets 6 to
12, narrowly oblong or elliptical, varying to linear, from 7 to 10 lines
long, 22 to 4 lines wide, of a thin and membranaceous texture,
copiously veiny, very obtuse or mostly truncate and slightly retuse,
mucronulate. ‘Tendrils commonly pinnate; the lowest pair close to
the stipules. Stipules semisagittate, or the upper broader and semi-
cordate, sharply and incisely lobed or 2—3-cleft, 2 or 3 lines long, per-
sistent. Peduncles about 2 inches long, equalling the leaf, pubescent,
fourflowered. Flowers about half an inch in length, recurved, on
pedicels of barely a line in length. Calyx campanulate, pubescent,
or else glabrate with age, almost equally five-cleft; the subulate teeth
as long as the tube, the two upper ones slightly smaller. Corolla appa-
rently purple. Ovary linear, somewhat stipitate, glabrous or nearly
glabrous. Style densely bearded from the apex nearly to the
middle. Immature legume glabrous, somewhat sabre-shaped, fully an
inch long.

This can hardly pass into V. Andicola, H. B. K.; which has much
narrower and strongly mucronate leaflets, entire and narrower sti-
pules, the style bearded at the summit only, and larger pods.

o. VictA Menzizsu, Spreng.

V. glabella; caulibus angulatis prelongis; stipulis supervoribus semi-
sagiltatis acuminatis, inferioribus flabellatis pectinato-dentatis, dentibus
longe subulatis; foliolis 8-12 (magnis) ovato-oblongis obtusis mem-
branaceis; pedunculis folio brevioribus apice plurifloris; floribus
(maximis) purpurascentibus longiuscule pedicellatis ; calycis dentibus
acutissime subulatis tubo cequilongis.

Vicia Menziesii, Spreng. Syst. Veg. 3, p. 267.
V. grandiflora, Smith, in Rees’ Cyel. no. 7, non Seop.

Haz. Hawaii, Sandwich Islands: in forests of Mouna Kea: without

LEGUMINOS &A. 491

flowers or fruit. (Discovered by Menzies, on Mouna Loa, at the
upper edge of the forest: also gathered by Macrae, on Mouna Kea;
in flower.)

A nearly glabrous species, of most remarkable size. Stems elon-
gated, “climbing, tangled among shrubs,” angled, minutely pubescent
or glabrous, very leafy. Stipules foliaceous, pectinately dentate with
long and subulate teeth, especially the lower ones, which are rounded or
flabellate and half an inch or more in diameter; the upper half-sagit-
tate and acuminate. Leaflets 8 to 12, often 2 inches in length, ovate-
oblong, obtuse, membranaceous, veiny, mucronulate, commonly alter-
nate; the lowest near the stem. Tendrils compound. Peduncles
shorter than the leaves, 2 or 3 inches long, 6-9-flowered ; the pedicels
slender, approximate, 3 to 5 lines long. Flowers very large, from an
inch to an inch and a half in length, pale purple. Calyx-teeth as long
as the tube, 3 lines long, narrow, very acutely subulate, the lower one
a little longer than the others. Vexillum recurved, ecallose. Style
filiform, minutely pubescent all round above the middle. Stigma
terminal, minute. Legume unknown.

This is much the largest-flowered species of the genus known;
and the leaflets are proportionally ample. The character and descrip-
tion, as to the flowers, are entirely drawn from a specimen gathered
by the late Mr. Macrae. The species is omitted in DeCandolle’s
Prodromus.

46. LATHYRUS, Linn.
1. Latayrus sessiuirotius, Hook. & Arn.

Lathyrus sessilifolius, Hook. & Arn. Bot. Beech. Voy. p. 20, & Bot. Mise. 3, p. 197.
L. epetiolaris, Clos, in Gay, Fl. Chil. 2, p. 146.

Has. Chili: in ravines near Valparaiso.

This name has the priority in publication, I believe, over the

homonymous L. sessilifolius of Tenore.
106

492 PHANEROGAMIA.

2. LATHYRUS SERICEUS, Lam.

Lathyrus sericeus, Lam. Dict. 2, p. 708 ; DC. Prodr. 2, p. 369?
L. tomentosus, Vogel, in Linnea, 13, p. 24, an Lam.?

Has. Rio Negro, North Patagonia; where it is very abundant.

Without doubt this is the L. sericeus of Lamarck, who described a
small and undeveloped state of the plant. Vogel has completed the
character, under the name of L. tomentosus; and perhaps the plant so
named by Lamarck is a less silky and somewhat scandent state of the
same species, with longer leaflets and smaller stipules. The latter
vary greatly in size, sometimes nearly equalling the leaflets (which
are from half an inch to an inch in length), and always much exceed-
ing the petiole. The tendril is mostly short and simple. The linear,
very silky-tomentose pods are nearly 2 inches long. The flowers are
said to be “light blue.”

3. LATHYRUS PUBESCENS, Hook. & Arn.

Lathyrus pubescens, Hook. & Arn. Bot. Beech. Voy. p. 21, & Bot. Misc. 8, p. 198;
Gay, Fl. Chil. 2, p. 148.

Has. Andes of Peru, near Obrajillo.

Less pubescent than Chilian specimens; but in other respects
nearly the same. It was likewise gathered by Mr. Matthews, in
Peru; where Mr. M’Lean also collected what appears to be a nearly
glabrous form of the same species, with broader leaflets.

47, ORMOCARPUM, Beawe.

1. ORMOCARPUM SENNOIDES, DC.

Ormocarpum sennoides, DC. Prodr. 2, p. 315; Wight, Ic. Pl. Ind. Or. t. 297.

Has. Feejee Islands; on the leeward coasts. (In flower only.)

LEGUMINOSE 423

48. CHA TOCALYX, DC.

1. CHATOCALYX LONGIFLORUS, Benth.
Cheetocalyx longiflorus, Benth. ined. in Herb. Imp. Vindob.
Has. Brazil, in the vicinity of Rio Janeiro.

A miserable specimen, in flower only, and insufficient for charac-
terizing the species. The leaflets appear to be only five, rather
smaller than those of C. Vincentina, and somewhat pubescent on both
sides. The calyx has a truncate orifice (as in the second section of
Vogel’s Rhadinocarpus), bearing the setaceous teeth, which are almost
as long as the campanulate tube.

49, PHYLACIUM, Bennett.

1. PHYLACIUM BRACTEOSUM, Bennett.
Phylacium bracteosum, Bennett, Pl. Jay. Rar. p. 159, t. 33.

Has. Majaijai Mountains, Luzon, Philippine Islands.

50. ZORNIA, Gmelin.

1. ZoRNIA ANGUSTIFOLIA, Smath.

Zornia angustifolia, Smith, in Rees’ Cycl.; DC. Prodr. 2, p. 316; Webb, Spicil.
Gorg. in Hook. Niger Fl. p. 121.
Hedysarum diphyllum, var. a. Linn. Syst. excl. syn.

Has. St. Jago, Cape de Verde Islands.—An Oriental species, now
widely diffused.

424 PHANEROGAMIA.

2. ZORNIA RETICULATA, Smith.

Zornia reticulata, Smith, 1. ¢ ; DC. Prodr. 2, p. 216.
Hedysarum diphyllum, var. 8. Linn.; Swartz, Obs. Bot. p. 28.

Has. Brazil, in the vicinity of Rio Janeiro.

These two species, together constituting the Hedysarum diphyllum
of Linnzeus, are probably to be reunited, under the name of Z.
diphylla, Pers., as is done by Vogel, and recently by Bentham (in
Kew Jour. Bot. 4, p. 45). One belongs to the Old, the other to the
New World; but the characters that distinguish them are not of much
value.

51. STYLOSANTHES, Linn.
1. STYLOSANTHES VIScosA, Swartz.

Has. Brazil, around Rio Janeiro; where it is very common.

2. STYLOSANTHES GUIANENSIS, Swartz.

Has. Brazil; with the preceding species.

62. ARACHIS, Linn.
1. ARACHIS HyPOG.mA, Linn.

Has. Singapore: probably escaped from cultivation.—It is uncer-
tain whether the Pea-nut is a native of Africa or of Equinoctial
America. Mr. Bentham inclines to the latter opinion, adducing the
fact that the other known species of the genus, five in number, are
all Brazilian.

LEGUMINOSAE. 425

538. ADESMIA, DC.
1. ADESMIA FILIPES, Sp. Nov.

A. annua; caulibus e basi ramosis gracilibus erectis glandulosis ; stipulis
subulatis ; foliolis 5-6jugis minimis ovalibus utringue sericeo-cinereis ;
racemis terminalibus elongatis sparsifloris; pedicellis filiformibus
erectis flore longioribus scepius hispidulo-glandulosis; calyce pubes-
cente, lobis lineari-subulatis tubo longioribus ; lomento 5—6-articulato
setis plumosis confertis villosissumo.

Has. Rio Negro, North Patagonia; on sand-hills.

Root annual. Stems branched from the base, erect, from a span to a
foot high, very slender, hispidulous-glandular and viscid, also minutely
pubescent. Stipules subulate, small, nearly distinct. Leaves abruptly
pinnate, an inch or more in length, including the petiole; the latter
sparsely and slightly hispidulous-glandular. Leaflets 5 or 6 pairs,
very small, from a line to a line and a half in length, oval, obtuse, not
mucronate, grayish with a silky pubescence both sides. Racemes ter-
minal, elongated (from 2 to 6 inches long), sparsely flowered. Bracts
very minute, persistent. Pedicels filiform, erect, from 5 to 8 lines long,
sparingly and minutely glandular-hispidulous. Flowers 3 lines long.
Calyx pubescent, not glandular, half the length of the corolla; the
lobes linear-subulate, longer than the tube, rather blunt. Corolla yellow;
the carina and wings nearly equalling the vexillum, which is gla-
brous, except a sparse and minute pubescence on the outside near the
summit. Stamens asin the genus. Style filiform, elongated. Loment
straight, linear, of 5 or 6 joints, each a line and a half in diameter,
villous with the long plumose bristles which thickly beset their faces.

2. ADESMIA HISPIDULA, DC.
Adesmia hispidula, DC. Prodr. 2, p. 319, & Mem. Legum. p. 307, t. 48.
Var. @. SUBNUDA: articulis lomenti medio setis nudis disco brevioribus

echinulatis rariusve inermibus ; foliolis retusis.
107

426 PHANEROGAMIA.

Var. y. PLUMOSA: articulis lomenti (infimo excepto) setis elongatis vere
plumosis crinitis ; foliolis retusis.

Has. Obrajillo, Andes of Peru: both varieties.

These are, I believe, two extreme forms of the species which
DeCandolle has illustrated; one of them having only very short and
glabrous setee on the joints of the pod, and even these sometimes obso-
lete; the other having the copious sete elongated and conspicuously
plumose. The leaflets in both are retuse. A specimen in the Hooke-
rian herbarium, gathered by Matthews, No. 540, connects our var. y.
with the original form of the species.

3. ADESMIA BALSAMICA, Bert.

Adesmia balsamica, Bertero, in Mem. Turin. 39, p. 59, t. 10; Hook. & Arn. Bot.
Beech. Voy. p. 20, & Bot. Misc. 3, p. 192; Gay, Fl. Chil. 2, p. 180.
Mimosa balsamica, Molina, Hist. Chil.

Has. Valparaiso, Chili.

The foliage and branchlets are covered with glands, which yield an
agreeable aromatic balsam.

4, ADESMIA MICROPHYLLA, Hook. & Arn.

Adesmia microphylla, Hook. & Arn. Bot. Beech. Voy. p. 19, t. 9; Lodd. Bot. Cab.
t. 1962.

Has. Valparaiso, Chili; abundant on hills near the sea.

0. ADESMIA HORRIDA, Gillies.
Adesmia horrida, Gillies in Hook. Bot. Misc. 8, p. 191; Gay, Fl. Chil. 2, p. 204.

Has. Andes of Chili, above Santiago. (Branches without flowers
or fruit.)

LEGUMINOSE. 427

Adesmia Echinus of Presl, Symb. Bot. 2, p. 14, t. 61, is probably
only a form of A. horrida, with the loment reduced to a single joint.

6. ADESMIA TRIUGA, Gillies.

Adesmia trijuga, Gillies; Hook. & Arn. Bot. Misc. 3, p. 191; Gay, Fl. Chil. 2,
p- 200.

Has. High Andes of Chili: on La Dessa, above Santiago, near the
snow-line. (In fruit: a form with rather large leaflets.)

54. STREPTODESMIA, Nov. Gen.

Calyx persistens, quinquenervis, quinquefidus, tubo campanulato, lacinis
subeequalibus. Corolla Adesmize, sed emarcida persistens. Stamina
(10, libera) Adesmix. Ovariwm 4—-G-ovulatum: stylus filiformis,
adscendens. Lomentum corolla marcescente inelusum, sutura carinalt
excisum, 3-6-articulatum ; articulis subglobosis, a sutura vexillart con-
tinua filiformi stylifera mox contorta persistente secedentibus, bivalvi-
bus, monospermis, valvulis levibus membranaceis. Semina subglobosa.
—Suffrutex intricato-ramosissimus, canescens ; ramulis spinescentibus ;
foliis abrupte pinnatis paucijugis ; racemis paniculatis brevibus, rhacht
spinescente persistente ; corolla lutea.

1. SrREPTODESMIA CANESCENS, Sp. Nov. (Tab. 47.)
Has. Rio Negro, North Patagonia; on sand-hills; common.

A rigid, shrubby plant, a foot or two in height, canescent with a fine
and close silky pubescence, intricately much branched ; the branchlets
spinescent, slender, glabrate. Stipules minute, silky. Leaves abruptly
pinnate, or perhaps sometimes with a terminal leaflet, half an inch
long, including the petiole, often fascicled on short spurs. Leaflets 3
pairs, or sometimes 2 or 4 pairs, crowded, oblong or oval, acutish or
obtuse, not retuse, minute (a line or a line and a half in length),

498 PHANEROGAMIA.

thickish, finely silky-canescent both sides. Flowers crowded in short
panicled racemes at the summit of the branches; the rhachis persistent
and terminating in a spinescent point. Bracts minute. Pedicels
slender, 3 to 6 lines long, often longer than the nodding flowers,
Bractlets none. Calyx persistent, canescently pubescent, 3 lines long,
fiwe-cleft to the middle; the tube campanulate, or at first somewhat
turbinate, five-nerved ; the nerves thickened and extending to the apex
of the linear-subulate, nearly equal lobes (the two upper ones slightly
exceeding the others). Corolla yellow, marcescently persistent, a little
exceeding the lobes of the calyx, incurved in the manner of Adesmia:
vexillum dilated-obovate, complicate, at length upwardly spreading,
glabrous, marked with a puberulent spot near the summit of the claw
on the inner side: ale cuneate-oblong, oblique, nearly equalling the
vexillum in length, the lamina minutely bearded on the lower edge
near the base: carina obtuse, incurved, its petals bearded on the
lower edge up to the point of junction. Stamens 10, abruptly in-
curved above the middle: filaments distinct, or slightly coalescent at
the base, flattened, in 2 series: anthers uniform, oval. Ovary linear,
flattish, minutely pubescent, containing from 4 ¢o 6 ovules. Style fili-
form, elongated, geniculate, ascending: stigma minute, terminal,
naked. Loment included in the marcescent corolla, and partly covered
by the calyx, irregularly or at length spirally twisted, of 3 to 6 pube-
rulent, turgid, or globular joints, which separate at maturity from each
other and from the continuous, filiform, persistent ventral suture: the
latter coils spirally or assumes a zigzag forin, and is tipped with the
deflexed persistent style: the joints are promptly two-valved, one-
seeded ; the valves membranaceous, even, and destitute of reticulations,
or very nearly so. Seeds globular, obscurely kidney-shaped, smooth.
Embryo incurved: cotyledons very thick.

Mr. Brown and Mr. Bennett* have called attention to “ the variety
and singularity of the modes adopted in the different subdivisions of
the Linnean genus Hedysarum for the protection of the pod and its
contents, during their progress to maturity.” To the various modes
which they have enumerated the present plant affords a curious
addition; the protection being here afforded by the persistence of the
corolla, as well as the calyx, until the seeds are matured and shed,
and the retention of the pod within its enclosure by the twisting or

* Horsfield, Plantes Javanicee Rariores, p. 157.

LEGUMINOSAE. 429

spiral coiling of the semi-replum, instead of the alternate flexion of
the loment observed in Smithia* and Lourea. These peculiarities,
along with the separation of the globular articles from the persistent
semi-replum at maturity, and their prompt bivalvular dehiscence, will
surely warrant the establishment of a distinct genus for this plant,
notwithstanding its close resemblance to Adesmza in other respects.

The name, compounded of scperris, twisted, and dona, a bond, 18
expressive of a characteristic of the fruit.

Piate 47.—STREPTODESMIA CANESCENS: a branch, of the natural size.
Fig. 1. A portion of the inflorescence, with mature fruit, enclosed by
the persistent calyx and corolla. 2. Calyx. 3. Corolla displayed.
4, Stamens and pistil. 5. Pistil. 6. An immature loment. 7. A
mature loment. 8. A mature loment, with the articles separating
from the persistent and twisted ventral suture. 9. A seed. 10. Lon-
gitudinal section of the same.—The details variously enlarged.

55. ZA SCHYNOMENEH, Linn.

1. AdgcHYNOMENE RuDIS, Benth.
Aischynomene rudis, Benth. Pl. Hartweg. p. 116, no. 649.

Has. “Rio Janeiro, Brazil,” according to the ticket: but, perhaps,
from Peru, where alone this species has hitherto been found; the
original specimens having been gathered near Lima by Cuming, and
at Guayaquil by Hartweg.

* It may here be noted that Smithia spicata, Spreng. Neue Enid. 2, p. 160, truly
belongs to that genus, as now received by Bentham (in Pl. Jungh.): it is a species of
his section Kotschya (genus Kotschya, Endl.), nearly allied to the S. strigosa from Mada-
gascar, and with flowers of about the same size, the fructiferous calyx not more than 3
lines in length. The lobes of the latter are, however, very obtuse ; the bractlets oblong-
faleate and barely acute, approximate to the reflexed calyx; the spikes rather shorter
than the leaves; and the rigid leaflets fully as long as those of S. Kotschyz, but nar-
rower. The original specimens in the herbarium of Professor Torrey (from whom
Sprengel received his materials) are said to have been collected in the’ “‘ West Indies,
probably Guadalupe,” by Mr. Perrin. The plant, however, was most probably brought

from Eastern Africa.
108

430 PHANEROGAMIA.

2. AXSCHYNOMENE CILIATA, Vogel.
Aischynomene ciliata, Vogel, in Linnea, 12, p. 84.

Has. Brazil; common in the vicinity of Rio Janeiro: growing in
marshes.

3. AUSCHYNOMENE SENSITIVA, Swartz.

Aischynomene sensitiva, Swartz, Fl. Ind. Oce. 3, p. 1276; DC. Prodr. 3, p. 320.

Has. Brazil; with the preceding species.

56. URARIA, Desv.

1. Urarta prota, Desv.

Uraria picta, Desv. Jour. Bot. 8, p. 122; Wight, Ic. Pl. Ind. Or. t. 411.
U. linearis, Hassk. Pl. Jav. Rar. p. 349, fide Benth. in Pl. Jungh. p. 213.

Has. Near Caldera, Mindanao, one of the Philippine Islands.

2. URARIA LAGOPOIDES, DC.

=>

Uraria lagopoides, DC. Prodr. 3, p. 824; Wight. Ic. Pl. Ind. Or. t. 289.
Hedysarum lagopodioides, Linn.? Forst. Prodr. F]. Ins. Austr. p. Ol.
H. lagopoides, Burm. Fl. Ind. p. 68, t. 53, f. 2.

Has. Feejee Islands; on the leeward coast of Vanua-levu. Samoan
Islands; common on Savaii, in cultivated grounds. Also near Cal-
dera, Philippine Islands; a variety with narrowly oblong leaflets, the
same as No. 1873 of Cuming’s Philippine collection. Forster found
it at New Caledonia.

LEGUMINOS &. 43]

57. DENDROLOBIUM, Wight & Arn., Benth.

Desmopium, § Evprsmopium, DC. Prodr. 2, p. 325, excl. sp. no. 3.
Desmoprum, subgen. DenpDRoLoBIUM, Wight & Arn. Prodr. Fl. Ind. Or. 1, p. 2238.
DeNDROLOBIUM, Benth. in Pl. Jungh. Jav. fase. 2, p. 21.

1. DENDROLOBIUM UMBELLATUM, Wight & Arn. l. ¢.

Has. Small islands in the Sooloo Sea. Ovolau, Feejee Islands.
Tongatabu. Manua, Upolu, and Savaii, Navigators’ Islands.

“A mostly littoral shrub, 6 to 20 feet high, with white flowers.”

2. DENDROLOBIUM AUSTRALE, Benth. 1. c.

Hedysarum umbellatum, Forst. Prodr. Fl. Ins. Austr. p. 51? non Linn.
H. australe, Willd. Spec. Pl. 8, p. 1183.
Desmodium australe, DC. Prodr. 2, p. 326.

Has. Nukulau, Feejee Islands.
The specimen, in fruit only, accords both with Willdenow’s and

DeCandolle’s character, and has much smaller leaves than the fore-
going species.

58. PHYLLODIUM, Desv., Benth.

1. PHYLLODIUM PULCHELLUM, Desv.

Phyllodium pulchellum, Desv. Jour. Bot. 3, p. 123, t. 5; Benth. in Kew Jour. Bot.
4, p. 46.

432 PHANEROGAMIA.

Dicerma pulchellum, DC. Mem. Legum. & Prodr. 2, p. 8389; Wight, Ic. Pl. Ind.
t. 418.

Has. Philippine Islands: near Bafios, Luzon.

59. DESMODIUM, Desv., Benth.
1. Desmoprum (SAGoTIA) TRIFLORUM, DC.

Desmodium triflorum, DC. Prodr. 2, p. 3834; Wight & Arn. Prodr. 1, p. 229, var. 6.
D. (Sagotia) triflorum, Benth. in Pl. Jungh. 2, p. 221.

Nicolsonia reptans, Meisn. in Linnea, 21, p. 260.

Sagotia triflora, Walp. & Duchass. in Linnea, 23, p. 738, & Ann. Bot. 2, p.. 413.

Has. Luzon, Philippine Islands. A widely diffused tropical species.

2. Desmopium (NICOLSONIA) BARBATUM, Benth.

Desmodium (Nicolsonia) barbatum, Benth. in Pl. Jungh. fase. 2, p. 224, non Wall.
Hedysarum barbatum, Linn. Spec. Pl. ed. 2, p. 1055; Swartz, Obs. p. 287.
Nicolsonia barbata, Cayennensis, & venustula, DC. Prodr. 2, p. 825.

Has. Rio Janeiro, Brazil.

3. Desmopium (Nicotsonra) poLycarruM, DC., Benth. l. c.

Has. Society Islands; Tahiti, Matia, and Eimeo: on dry hills,
near the coast. Samoan Islands. Tongatabu. Feejee Islands:
common at most of the stations visited, near the shores.

The copious synonymy of this widely diffused species is collated in
Wight & Arnott’s Prodromus, and lately corrected by Bentham, in
his account of the Leguminose of Junghuhn’s collection (Pl. Jungh.
fasc. 2). Among the synonyms are D. heterocarpum, DC. (the Hedy-
sarum heterocarpum of Linneeus and of Forster), D. purpureum, Hook.
& Arn., and D. nervosum of Vogel.

LEGUMINOS&. 433

4, Desmopium (NicotsoniA) capitatum, DC., Benth. l.c.

Hedysarum capitatum, Burm. Fl. Ind. p. 167, t. 54, f. 1.
Pseudarthria capitata, Hassk. Pl. Jay. Rar. p. 390.

Has. Baiios, Luzon, Philippine Islands. (With ripe fruit.)

5. Desmopium (HetrrotomaA) Ganceticum, DC, Benth. 1. c.

Has. Luzon; with the preceding species.

6. Desmoprum (HETEROLOMA) SANDWICENSE, . Meyer.

Desmodium Sandwicense, E. Meyer, Ind. Sem. Hort. Regiomont. 1850, & in
Linnea, 24, p. 230.

Has. Oahu, Sandwich Islands; on the hills behind Honolulu.
(Also gathered by Seemann.)

This is described from specimens raised from seed, which was
gathered at Oahu by Diedrichson. Our specimen is in fruit, with
longer pods than is described by E. Meyer, often as much as nine-
jointed. Dr. Pickering mentions a flowering specimen, which is not
found in the collection. Seemann’s specimen is in flower. This is
the only species known from the Sandwich Islands; for D. Chamis-
sonis of Vogel came from Luzon, not from Oahu, as was erroneously
stated by Walpers.

7. Desmoprum (Hrrrrotoma) Limense, Hook.
Desmodium Limense, Hook. Bot. Misc. 2, p. 215; Walp. Repert. 1, p. 743.
Has. Callao, Peru.

Two other species are mentioned in Dr. Pickering’s notes as having
109

434 PHANEROGAMIA.

been found at Callao and Lima: one of them appears to be D. Peru-
vianum, Vogel, in Rel. Meyen. p. 30; but neither have been identified
in the herbarium of the Expedition.

8. Dresmopium (HETEROLOMA) INCANUM, DC.

Desmodium incanum & D. ancistrocarpum, DC. Prodr. 2, p. 382.
Has. Rio Janeiro and Organ Mountains, Brazil. The typical form

of the species, and likewise the D. ancistrocarpum, which appears to
be no more than a variety.

9. Dresmoprium (HETEROLOMA) AXILLARE, DC.
Desmodium axillare (& D. reptans), DC. Prodr. 2, p. 833; Benth. Bot. Sulph. p. 82.

Has. Rio Janeiro, Brazil: the var. villosum, Benth.

10. Desmopium (HETEROLOMA) oBovATuUM, Vogel.
Desmodium obovatum, Vogel, in Linnza, 12, p. 106, & Rel. Meyen. p. 30.

Has. Organ Mountains, Brazil. (Without flowers or fruit.)

11. Desmoprum (HETEROLOMA) BRACHYPODUM, Sp. Nov.

D. erectum, glabellum; stipulis persistentibus striatis e basi ovata vel
subcordata acuminatis subaristatis ; foliis trifoliolatis ; foliolis oblongis
obtusis laxe venosis ; racemo virgato terminali elongato; bracteis e basi
lata subulatis deciduis; pedicellis geminis perbrevibus post anthesin
deflexis ; lomenti articulis 4-6 pubero-scabris parvulis.

Has. Sydney and Hunter’s River, New South Wales. (Port
Curtis, Australia, Macgillivray, in Herb. Hook.)

LEGUMINOSE. 435

A rather tall, herbaceous, apparently erect, nearly glabrous species
of the section Heteroloma, division Strobilacea, Benth. in Pl. Jungh.
Branches, &c., minutely scabrous-puberulent. Stipules persistent,
striate, rather rigid, with a dilated ovate or subcordate base, contracted
into an awn-pointed acumination, in all 4 lines long. Stipels seta-
ceous. Leaflets 8, oblong, or ovate-oblong, very obtuse, loosely veined,
and somewhat reticulated, chartaceo-membranaceous, 12 to 3 inches
long, glabrous and smooth above, slightly puberulent and a little paler
underneath. Racemes terminal, at least the principal one, strict and
virgate, at length elongated, the stronger ones in fruit becoming a foot
or more in length; the flowers small (2 or 3 lines long), rather
crowded. Bracts subulate from a broad base, striate like the stipules,
about 3 lines long, deciduous. Pedicels in pairs under each bract,
barely a line long, deflexed after anthesis, and remaining so in fruit.
Calyx-lobes triangular-subulate. Loment of 4 to 6 joints, the upper
suture slightly, the lower deeply sinuate, not stipitate; the joints
obliquely oval, or at length semioval, from 12 to 2 lines long, scabrous-
puberulent.

12. Desmopium (HetTErotoma?) Guynianum, Benth, ined.

Has. Near Sydney, New South Wales. (Also gathered by Mr.
Backhouse, in Eastern Australia, and by Mr. Gunn in Van Diemen’s
Land.)

The character of the species has not yet been published by Mr.
Bentham; and our specimens, having no good fruit, are insufficient.

13. Desmoprum Scorriurus, Desv.

Has. Tahiti, Society Islands —Probably introduced from America.

14. Desmoprum (CHALARIUM) TorRTUOSUM, DC.

Has. St. Jago, Cape de Verde Islands.—Chiefly a tropical Ame-
rican species.

436 PHANEROGAMIA.

15. Desmoprum (PopocARPIUM) LEPTOPUS, Sp. Nov.

D. glabriusculum; foliolis ovato-lanceolatis, terminali subrhombeo acu-
minato; racemo laxo subsimplici; pedicellis articulo longioribus ;
lomento scabro-pubescente stipite longo gracili; articulis longe cuneato-
semirhombeis latitudine triplo longioribus.

Desmodium (Podocarpium) leptopus, Gray, ined.; Benth. in Pl. Jungh. fase. 2, p.
226, in adn.

Has. Luzon, Philippine Islands; on mountains near Baiios.

A slender, nearly glabrous species, of the section to which belongs
our North American D. acuminatum, D. pauciflorum, &. Stem with
two or three scattered leaves, a foot or two in height. Stipules and
bracts caducous. Leaflets 3, ovate-lanceolate, or the terminal one somewhat
rhomboidal, acuminate, membranaceous, 2 or 3 inches long, scarcely
paler underneath. Peduncle terminal, filiform, 6 or 8 inches long,
bearing a nearly sumple and lax raceme of few flowers. Pedicels in
pairs, 6 to 11 lines long. Stamens monadelphous. Loment on a /fili-
form stipe as long as the pedicel, of 2 scabrous-pubescent, cuneate and
semirhomboidal joints, each nearly half an inch long and 2 lines wide.

60. ALYSICARPUS, Necker.
1. ALYSICARPUS VAGINALIS, DC.

Has. St. Jago, Cape de Verde Islands.

61. ABRUS, Linz.
1. ABRUS PRECATORIUS, Linn.

Has. Feejee Islands, Tongatabu, Samoan, and Society Islands.
Introduced.

LEGUMINOSA. 437

62. NEUROCARPUM, Desv.

1. NEUROCARPUM CAJANIFOLIUM, Presl.

Neurocarpum cajanifolium, Presl, Symb. Bot. 1, p. 17, t. 9; Benth. Comm. Legum.
Lotus Fluminensis, Velloz. Fl. Flum. 7, t. 152.

Has. Rio Janeiro, Brazil.

63. CENTROSEMA, DC, Benth.
1. CenTROSEMA PLuMIERI, Benth. 1. c.

Has. Rio Janeiro, Brazil; very common.

2. CENTROSEMA DECUMBENS, Mart., Benth. l. c.

Has. Near Rio Janeiro, Brazil. (Also, like the preceding, a West
Indian species.)

64. KENNEDYA, Vent.
1. KENNEDYA RUBICUNDA, Vent.

Has. Sydney and Hunter’s River, New South Wales. Also,
Woolongong: a variety (in fruit only) with the leaves remarkably

silky underneath.
2. KENNEDYA (ZICHYA) COCCINEA, Vent.

Has. New South Wales, in the neighbourhood of Sydney.
110

4388 PHANEROGAMIA.

6. HARDENBERGIA, Benth.

1. HARDENBERGIA OVATA, Benth.

Hardenbergia ovata, Benth. Pl. Hugel. p. 40, & Comm. Legum. |. c. p. 60.
Kennedya ovata, Sims, Bot. Mag. t. 2169; DC. Prodr. 2, p. 384.

Has. Hunter’s River, New South Wales. (In fruit.)

The allied HA. monophylla, which must have been noticed in the
vicinity of Sydney, does not occur in the collection.

66. LEPTOCYAMUS, Benth.
1. LEPTOCYAMUS MICROPHYLLUS, Benth.

Leptocyamus microphyllus, Benth. in Linn. Trans. 18, p. 209; Walp. Repert. 1,
p. 798.

Has. Sydney, New South Wales.

2. LEPTOCYAMUS ELONGATUS, Benth. 1. c.

Has. Puen Buen and Hunter’s River, New South Wales.

67. GALACTIA, P. Browne.
1. GALACTIA HISPIDULA, Benth.

Galactia hispidula, Benth. in Linnea, 22, p. 514; Walp. Ann. Syst. Bot. 2, p. 422.

Has. Near Rio Janeiro, Brazil; “rare.”

LEGUMINOS &. 439

68. CLEOBULIA, Mart.

Char. fruct. Legumen oblongum, plano-compressum, coriaceum, exa-
latum, suturis haud incrassatis. Semina Dioclez. P

1. CLEOBULIA MULTIFLORA, Mart.
Cleobulia multiflora, Mart. in Herb. Bras.; Benth. Comm. Legum. l.c. p. 67.
Has. Brazil; in the Organ Mountains, near Rio Janeiro.

The specimen bears only the fruit, which was not before known,
on which account the characters are now given. The legumes appear
to be essentially those of Cratylia, while the seeds are those of a
Dioclea. The former are 2 or 24 inches long, three-fourths of an inch
wide, much compressed, obtuse at both ends, not stipitate, of a coria-
ceous texture, coated with a rusty down, two-valved, within inter-
cepted between the seeds by a soft cellular tissue. The sutures are
not perceptibly thickened, nor are there any traces of wings. Seeds
5 or 6, transverse, 3 lines long, oval, not much compressed, chestnut-
brown, with darker spots, nearly the whole length of one side occu-
pied by the linear hilum, from which the slender strophiole separates.

69. DIOCLEA, ZB K.

1. DiocLeA VIoLACEA, Mart.

Dioclea violacea, Mart. in Herb.; Benth. Comm. Legum. |. c. p. 69.
Dolichos altissimus, Velloz. Fl. Flum. 7, t. 134.

Has. Hawaii, Sandwich Islands; on the coast near Hilo: also on
Kauai. (Byron’s Bay, Macrae.)

Probably introduced into the Sandwich Islands, as it is a native of

440 PHANEROGAMIA.

Brazil, although cultivated in Mauritius, &. The pods are 3 or 4
inches long, 14 to 2 inches wide, flat, glabrate when mature, about
two-seeded, conspicuously tricarinate at the ventral suture. Seeds
nearly orbicular, compressed, 9 or 10 lines in diameter, with a bony
testa; the linear hilum extending more than half way round the cir-
cumference of the seed.

70. CANAVALIA, DC.

1. CANAVALIA OBTUSIFOLIA, DC.
Canavalia obtusifolia, miniata, rosea, & rutilans, DC. Prodr. 2, p. 404, ex Benth.

Has. On the sandy shores of Upolu, Samoan Islands, and of
Ovolau, &., Feejee Islands; common.

2. CANAVALIA TURGIDA, Graham.
Canavalia turgida, Graham, in Wall. Cat. no. 5534.

Has. Mangsi Islands. Feejee and Samoan Islands. Tongatabu.
Tahiti: on the coast.

This species is closely allied to C. gladiata, and is probably only a
variety of it, with somewhat turgid and short pods, of 4 or 5 inches
in length and about an inch and a half in width. Dr. Wight, in his
Icones Pl. Ind. Or. t. 753, figures C. gladiata with pods only six
inches long.

3, CANAVALIA SERICEA, Sp. Nov.

C. sericeo-tomentosa ; foliolis obovato-rotundis retusis supra demum gla-
bratis subtus eximie sericeis ; racemis plurifloris folia subcequantibus ;
calycibus glabriusculis ; leguminibus tomentulosis triplo quadruplove
longioribus quam latis.

LEGUMINOSS. 4A]

Has. Feejee Islands: found at Rewa, Ovolau, and Direction Island.

A species of Hucanavalia, section Malochia ; with rather stout, pro-
cumbent or somewhat twining, softly tomentose stems and branches.
Stipules ovate-subulate, caducous. Leaflets rounded-obovate, retuse or
emarginate, thickish, 3 or 4 inches long, 23 to 3% inches broad,
minutely silky pubescent, but at length glabrate above, very silky under-
neath with a thick and somewhat silvery pubescence; the petioles,
inflorescence, &c., downy. Racemes several-flowered, short-peduncled,
nearly as long as the leaves. Flower an inch and a half in length.
Calyx almost glabrous; the large upper lip emarginate-two-lobed; the
short lower lip three-cleft; its lobes broadly triangular, acute, of equal
length. Ovary canescently tomentose. Legume tomentulose, oblong .
or oblong-linear, 3 or 4 inches long and an inch wide, straight, com-
pressed, raised on a stipe of half an inch in length. Seeds 3 or 4.

4, CANAVALIA GALEATA, Gaud.

C. subpubescens, mox glabrata ; foliolis ovato-ellipticis acuminatis, adultis
superne vel undique glabris; racemis 3-10-floris ; calycis labio supe-
riore subbilobo recurvo, inferiore trifido, lobis lateralibus ovatis obtusis,
infimo triangulato subacuto; corolla purpurea; leguminibus lineari-
oblongis quadruplo longioribus quam latis.

Dolichos galeatus, Gaud. Bot. Freyc. Voy. p. 486, t. 115.
Canavalia galeata, Gaud. 1. c. adn.; Vogel, in Linnza, 10, p. 584.
C. Gaudichaudii, Endl. Syn. Fl. Ins. Austr. in Ann, Wien. Mus. 1, p. 186.

Var. 3. PUBESCENS: foliolis etiam adultis subtus pubescentibus.
Canavalia pubescens, Hook. & Arn. Bot. Beech. Voy. p. 81; Vogel, in Linnza, L.¢.
Has. Sandwich Islands: near the coast at Waianae, Oahu; also on
Kauai. (Collected by Gaudichaud, Chamisso, Hinds; the pubescent
variety by Gaudichaud, Chamisso, Lay & Collie, Douglas, &c.)

Stem slender, twining, the young shoots and the leaves more or less
pubescent when young, soon glabrate or nearly glabrous in most cases.
11

4492 PHANEROGAMIA.

Stipules minute, ovate-subulate, caducous. Leaflets ovate-elliptical or
oblong-ovate, more or less acuminate, rounded at the base, membra-
naceous, 3 to 5 inches long, green and glabrate both sides, or (espe-
cially in var. (.) pubescent underneath, as are the partial petioles,
Racemes 3-10-flowered, rather shorter than the leaves. Flowers an
inch and a half long. Upper lip of the calyx recurved in flower, very
large even for the genus, strongly emarginate or somewhat two-lobed,
the lobes rounded; the short lower lip three-cleft, its lateral lobes ovate
or triangular-ovate and obtuse; the middle one more triangular and
acutish or acute. Corolla purple, apparently dark reddish-purple; the
obovate and emarginate vexillum rather longer than the wings and
keel. Androecium and pistil almost straight. Ovary silky-canescent,
many-ovuled. Legume linear-oblong, 4 or 5 inches long in the speci-
men, an inch or more in width, straight, compressed, glabrous, when
half grown minutely silky, slightly tricarinate at the ventral suture,
apiculate, short-stipitate. Seeds 6 or 7.

In the dried specimens the lower lobes of the calyx often appear to
be lanceolate and acute, as they are described and figured by Gaudi-
chaud. But this is owing to the involution of their margins: when
explanate they are seen to be as above described, and as in OC. pubes-
cens, Hook. & Arn. 1. ¢., which is evidently no more than a downy-
leaved variety of this species. These lobes are somewhat variable in
shape. In the specimen from Kauai they are unusually broad and
blunt, and the upper lip is more decidedly lobed; but no other diffe-
rence is remarked. In Gaudichaud’s figure, the flowers are repre-
sented considerably too large——The specimen collected by Chamisso
must have borne shorter pods than ordinary, as they are said by
Vogel to be only two inches and a half long, and of more than half
that width:

71, MUCUNA, Aduns.

1. Mucuna gigantea, DC.

Mucuna gigantea, DO. Prodr. 2, p. 405; Wight, in Hook. Bot. Mise. 2, p. 307,
Suppl. t. 14; Guill. Zeph. Tait. p. 65.

LEGUMINOSE. 443

Has. Feejee Islands; on Ovolau and Vanualevu. Samoan
Islands; on Tutuila, Savaii, &c., “with a stem 3 inches in diameter,
spreading over the tallest trees.” Hawaii, Sandwich Islands; in the
district of Puna, at a watering place, near Kaimo: a form with a
elabrate pod. Introduced? “Flowers green.”

The MV. altissima enumerated in Vogel’s account of Chamisso’s col-
lection, in Oahu, is probably this species. That of Hooker & Arnott,
in Beechey’s Voyage, as to the specimens, is Strongylodon ruber.

2. Mucuna vRENS, DC. l. ¢.

Dolichos urens, Linn.; Jacq. Stirp. Amer. t. 182, f. 84. (Plum. Amer. t. 107.)

Has. Hawaii, Sandwich Islands, in the district of Waimea, near
the coast (in flower and fruit): probably introduced. (Also gathered
by Gaudichaud, in the Voyage of the Bonite.) Tahiti, in the interior
forest; where Dr. Pickering noted a species with alate sutures and
oblique ridges, probably IZ wrens; but there is no such specimen in
the collection.

3. MUCUNA PLATYPHYLLA, Sp. Nov.

M. fulvo-pubescens ; foliolis magnis rotundis apiculatis subtus reticulatis
supra mox glabris; floribus cymosis viridulis ; leguminibus ovalibus
plano-compressis transverse lamellosis, junioribus hispidissimis.

Has. Ovolau and Rewa, Feejee Islands.

Plant with a stout and apparently tall-climbing, woody stem ; the
branches, ample foliage, &c., clothed throughout with a soft rusty or
fulvous pubescence. Petioles 5 or 6 inches long. Leaflets roundish,
5 or 6 inches in length, the terminal one often broader than long,
the lateral ones obliquely rotund-ovate, all abruptly apiculate with a
small acumination, tomentose with rusty pubescence and conspi-
cuously reticulated underneath, early glabrous above. Stipels seta-
ceous. Peduncle axillary, shorter than the petiole, bearing many

444 PHANEROGAMIA.

flowers in a compound cyme or a cymose cluster of short racemes.
Pedicels 3 to 7 inches long. Calyx canescent, and hispid with a few
scattered bristles; the lower lobe triangular-lanceolate, twice the
length of the ovate and obtuse lateral ones, which are shorter than
the ovate upper lip. Corolla “green,” as large as in M. monosperma
(2 inches long). Legume oval, flat, 4 inches long and 2 inches wide,
and still immature, very hispid with rusty and short prurient bristles,
which are soon deciduous; each membranaceous valve produced at the
sutures into a narrow wing-like margin, and its face strongly lamellate
with obliquely transverse and nearly straight salient plates. Seeds
apparently about 3.

722. ERYTHRINA, Linn.
1. ERYTHRINA MACROPHYLLA, DC. ?
Has. Rio Janeiro, Brazil. (Terminal leaflets only, 12 to 15 inches
long.)
2. Eryturina Inpica, Lam.

Has. Feejee Islands. Tongatabu. Samoan and Society Islands.
“Introduced, abundant: often cultivated by the natives.”

3. ERYTHRINA MONOSPERMA, Gaud.

Erythrina monosperma, Gaud. Bot. Freye. Voy. p. 486, t. 114; Hook. & Arn. Bot.
Beech. Voy. p. 81.

Has. Oahu, Sandwich Islands; on the Kaala Mountains, in the
district of Waianae. (In fruit.)

The specific name is badly chosen, since the torose, at length dehis-
cent pod frequently matures two seeds, and perhaps more, as the
ovary has as many as five ovules. The leaflets are for the most part

LEGUMINOSE. 445

dilated rhomboidal rather than ovate-reniform, as in Gaudichaud’s
original specimens and figure.

4, ERYTHRINA OVALIFOLIA, Roxb.

Erythrina ovalifolia, Roxb. Fl. Ind. 3, p. 254; Wight, Ic. Pl. Ind. Or. 1, t. 247.

Has. Near Manilla, Luzon. (Foliage only.)

73. STRONGYLODON, Vogel.

Char. auctus. Calyx campanulatus, truncatus sew 4—5-dentatus, denti-
bus obtusissimis vel obsoletis. Vewxilluwm ovato-oblongum, acutum,
demum recurvo-patentissimum, basi breviter unguiculatum et mem-
branula infleca appendiculatum, intus bicallosum. Ale subfalcate,
vevillo et carina multo minores. Carina gamopetala, falcata, ros-
trata, vexillo equilonga. Stamina 10, diadelpha: anthere fere uni-
formes. Ovarium stipitatum, uni—pauciovulatum: stylus capillaris,
longissimus, stigmate subcapitato penicillato terminatum. (Fructus
adhuc ignotus.)—Frutices vel suffrutices, caulibus gracilibus volu-
bilibus scandentes, inermes, glaberrime; foliis pinnatim trifoliolats
stipellatis ; racemis elongatis multifloris, pedicellis gracilibus ad nodos
fasciculatis ; floribus rubris.

Strongylodon ruber, the typical species of this genus, is distin-
guished from Erythrina (Micropteryx) by its appendiculate and con-
spicuously bicallose veawillum, its obtusely four-lobed calyx, with the
upper lobe notched, its wni(—bi-)ovulate ovary, its setaceous stipels, and
its voluble, Phaseoloid habit. From Spatholobus, which has a similar
habit, it differs in the form of the calyx, in the acute as well as
appendiculate and bicallose vexillum, and in the uniovulate, long-
stalked ovary. From Mucuna it differs in the calyx, vexillum, the
uniform anthers, the stigma, ovary, &c. The fruit is still unknown.
The ovary, although often uniovulate, as characterized by Vogel, I
find as frequently to contain two ovules. I therefore venture to

annex to the genus a plant from Luzon, of similar habit and cha-
112

446 PHANEROGAMIA.

racter, except that the calyx is obscurely toothed, and the ovary
contains several ovules (accumulated in the middle of the cell); the
legume of which is equally unknown. If this be admissible, we shall
have a genus differing from Spatholobus in the characters above-
mentioned, from Hrythrina (including Micropteryx, &c.) in the form of
the corolla, in its stipels, and especially in habit, and from both in
the appendiculate and bicallose vexillum.

1. StRONGYLODON RUBER, Vogel. (Tab. 48.)

S. foliolis ovatis membranaceis ; racemis folio duplo longioribus ; pedi-
cellis flore cequilongis; calyce basi bibracteolato manifeste quadri-
quinqguedentato, dentibus obtusissimis; carina falcato-incurva ; ovario
uni—biovulato.

Strongylodon ruber, Vogel, in Linnza, 10, p. 585.
Mucuna altissima, Hook. & Arn. Bot. Beech. Voy. p. 81, quoad specimen.
Harpxtropis speciosus, Nutt. in Herb. Hook.

Has. Sandwich Islands: on the Kaala Mountains behind Hono-
lulu, Oahu: in forests on the side of Mouna Loa, Hawaii. (Also
gathered by Chamisso, Macrae, Diell, Barclay, Nuttall, &.) Ovolau,
Feejee Islands: in forests.

“ A woody, twining vine,” with slender herbaceous branches, gla-
brous throughout. Stipules ovate, obtuse, striate, free, persistent.
Petioles 3 or 4 inches long. Leaves pinnately trifoliolate. Stipels
subulate-setaceous, rather persistent. Leaflets ovate, varying from
roundish to oblong-ovate, more or less acuminate, 23 to 5 inches long,
membranaceous in texture, green, both sides of nearly the same hue,
triplinerved, the minuter veinlets reticulated : petiolules 3 lines long,
exceeding the stipels. Racemes axillary, at least twice the length of
the leaves, 10 to 18 inches in length, including the peduncle, ascend-
ing; the rhachis multinodal, each node fasciculately about three-
flowered, or rarely bearing a partial peduncle which is three-flowered
at its apex, so that the raceme becomes somewhat paniculate. Bracts
minute? caducous. Pedicels an inch long, equalling the developed
flower in length, filiform, naked, bearing a pair of ovate bractleis, of

LEGUMINOS &. 447

about a line in length, which immediately subtend the calyx, and are
early caducous. Calyx campanulate, 3 lines long, truncate or even,
manifestly four-toothed with broad and very obtuse teeth, the upper
one broader, but not otherwise larger, and emarginately two-lobed, so
that the calyx appears five-toothed. Corolla papilionaceous, red,
apparently bright purplish red: petals thin. Vexillum ovate-oblong,
acute, an inch or an inch and a quarter in length, complicate and
more or less recurved-falcate in the bud, at length spreading or
recurved; the base raised on a very short claw (a line and a half
long), appendiculate on each margin by a strongly inflexed membrane ;
the face callose as if by an adherent membrane, which is denticulate
at the upper edge, and produced on each side into a free and salient
auricle or lamella. Ale small, less than half the length of the vexil-
lum, the lamina obliquely obovate and falcate, very obtuse, 4 or 5
lines long, on a claw of about 3 lines long, appressed to the carina
and partly adherent to its face. Carina about the length of the vexil-
lum, faleate and incurved, complicate, acuminately beaked, unguiculate,
its two petals cohering by their anterior edges for the whole length
more or less firmly. Stamens 10, diadelphous; the tenth filament
wholly distinct, not geniculate; the others united to the middle :
anthers oblong, similar, or five alternate ones a little shorter and sub-
cordate. Ovary raised on a stipe which is at length longer than itself
and as long as the calyx, glabrous, compressed, one-ovuled, or more
frequently two-ovuled; the ovules then superposed. Style capillary,
very long, exceeding the stamens, gradually incurved with the carina:
the stigma terminal, capitellate, and minutely penicillate. Legume
unknown.

The specimen from the Feejee Islands, a very imperfect one, has
shorter racemes and ovate-oblong leaflets. The latter, however, are
said by Vogel to be “oblong” in Chamisso’s specimens; while in all
of ours from the Sandwich Islands they are ovate-rotund. No other
difference is noted,

PrAtn 48.—STRONGYLODON RUBER: portion of a flowering specimen,
of the natural size. Fig. 1. A young flower-bud, with the bractlets.
2. Calyx, in anthesis. 3. Corolla displayed. 4. Stamens and pistil,
displayed. 5. Pistil and its stipe; the ovary longitudinally divided,
to show the ovules.—The details variously enlarged,

448 PHANEROGAMIA.

9. SrRONGYLODON MACROBOTRYS, Sp. Nov. (Tab. 49.)

S. foliolis oblongis; racemis pendentibus? longissumis ; pedicellis flore
brevioribus; calyce truncato obsolete quadri—quinquedentato ; carina
leviter incurva; ovario quadri—sex-ovulato.

Hap. Tn the mountains near Bafios, Luzon, Philippine Islands,

Probably a woody vine, with herbaceous and twining branchlets,
glabrous throughout. There is only a short leafy branch, with the
leaflets disarticulated, and a flowering raceme in the collection, unac-
companied by notes. Stipules caducous, or a depressed gland in their
place. Leaves pinnately trifoliolate. Stipels setaceous, deciduous.
Leaflets oblong or ovate-oblong, acuminate, green both sides, somewhat
coriaceous in texture, triplinerved and veiny, about 4 inches long;
the terminal one slightly cuneate, the others rounded at the base.
Raceme very long, probably axillary and pendent, 42 feet long, in-
cluding the short peduncle, above which it bears a series of alternate,
small and salient nodes, each bearing a fascicle of from 3 to 5
flowers, on pedicels of an inch and a half in length. Bracts and
bractlets not seen (either caducous or obsolete). Calyx campanulate,
truncate, 4 or 5 lines long, the border obscurely 4—5-toothed or repand.
Corolla papilionaceous, reddish or purple. Vexillum ovate-oblong,
acute, of a rather thickish texture, an inch and a half long, many-
striate, complicate in the bud, and slightly falcate-recurved, at length
expanding and spreading, at the base subunguiculate, and the mar-
gins appendiculate; the face callose and bilamellate nearly as in 4,
ruber, Ale oblong, falcate, short-unguiculate, obtuse, nearly half the
length of the vexillum, partly adherent to the base of the falcate, taper-
ing-beaked, moderately incurved carina, which is as long as the vexil-
lum, its two petals coherent by their anterior edge for nearly or quite
their whole length. Stamens and style, as in the foregoing species.
Anthers linear-oblong, uniform or nearly so. Stigma capitellate-trun-
cate, somewhat expanded or indusiform, the margin fringed with
minute hairs. Ovary puberulent, oblong, compressed, raised on a stipe
as long as itself, with from 4 to 6 ovules crowded about the middle of
the cell, probably only one or two of them fertile. Fruit unknown.

LEGUMINOS4E. 449

Piate 49.—StTRONGYLODON MACROBOTRYS: in flower, of the natural
size. Fig. 1. The calyx. 2. Corolla displayed. 38. The carina.
4. Stamens and pistil, enlarged. 5. Pistil, more enlarged; the ovary
divided, showing the ovules.

74. PHASEOLUS, Linn.
1. PHASEOLUS ROSTRATUS, Wall.

Phaseolus rostratus, Wall. Pl. Asiat. Rar. 1, p. 50, t. 63; Wight, Ic. Pl. Ind. t. 34.
P. ameenus, Soland. ex Forst. Prodr. Ins. Austr. no. 533, abs. char.

Has. Tahiti, Society Islands; common on the coast. Tutuila, &c.,
Samoan Islands. Ovolau, &c., Feejee Islands. On Diamond Hill,
Oahu, Sandwich Islands. Probably introduced into the South Sea
Islands.

2. PHASEOLUS TRUXILLENSIS, H. B. EK.

Phaseolus Truxillensis, H. B. K. Nov. Gen. & Spec. 6, p. 451; DC. Prodr. 2, p. 391.
P. Cumingii, Benth. Comm. Legum. (in Ann. Wien. Mus.) p. 75.

Has. Callao, Peru: on the sea-coast. A glabrate and a villous-
pubescent form.

3. Paasrotus (StropHostyLes) Muneo, Linn.

Has. Samoan and Feejee Islands. (Only fragments, apparently
referable to this species.)

4, Puaseotus (StropHostyLes) Scuorru, Benth.
Phaseolus (Strophostyles) Schottti, Benth. Comm. Legum. in Ann. Wien. Mus. p. 75.

Has. Organ Mountains, near Rio Janeiro, Brazil.
118

450 PHANEROGAMITIA.

5. PHASEOLUS (MAcRopTILIUM) MARITIMUS, Benth.

Phaseolus (Macroptilium) maritimus, Benth. Comm. Legum. 1. c. p. 77.
Lotus maritimus, Velloz. Fl. Flum. 7, t. 1338.

Has. Near Rio Janeiro, Brazil: on the sands of the coast.

6. PHASEOLUS VESTITUS, Hook.
Phaseolus vestitus, Hook. Bot. Misc. 2, p. 216; Benth. Comm. Legum. 1. ec.

Has. Near Callao, Peru. (Also gathered by Mr. Cuming.)

75. VIGNA, Savi, Benth.
1. ViGNA BRACHYSTACHYS, Benth.
Vigna brachystachys, Benth. Bot. Voy. Sulph. p. 8.

Has. Callao, Peru. A narrow-leaved variety of the species, which
was also gathered by Cuming. 7

2. Viana OAHUENSIS, Vogel.

Vigna Oahuensis, Vogel, in Linnzxa, 10, p. 585; Walp. Repert. 1, p. 779.
Vigna villosa, Hook. & Arn. Bot. Beech. Voy. p. 81, non Savi.

Has. Oahu, Sandwich Islands; on the Kaala Mountains (where it

was likewise gathered by Chamisso, Douglas, &.). Also on Kauai
and Maui.

Our specimens are imperfect, and want the fruit: and I have not
the means of properly comparing this with the related species. The
upper lobe of the calyx is decidedly emarginate.

LEGUMINOSAE. 451

3. Viena SAnpwicensis, Sp. Nov. (Tab. 50.)

V. radice tuberosa? eduli; caulibus volubilibus foliisque hirsutulo-puben-
tibus; stipulis stipellisque subulatis parvis; foliolis linearibus seu
lineari-lanceolatis mucronulatis utrinque reticulatis ; pedunculis folio
brevioribus bi-quadrifloris; calycis dentibus subcequilongis, supremo
ovato obtusissimo subemarginato, inferioribus 3 subulatis; carina
rectiuscula; legumine elongato-lineart recto glabrato compresso.

Has. Sandwich Islands: Hawaii, on Mouna Loa, six miles from
the crater Lua Pele: Kauai, on mountains: Maui, on mountains in
the western division of the island.

¢ Root edible,” according to Dr. Pickering’s notes; therefore probably
tuberous. Stems twining, slender, retrorsely and minutely hirsute-
pubescent, as are the petioles and peduncles; and the leaves are beset
on both sides with a similar pubescence. Stipules and stipels subulate,
minute. Leaflets linear or lUinear-lanceolate, 14 to 2 inches long, and
from one to 4 lines wide, acutish or obtuse, mucronulate, of a rather
firm texture, and of the same hue both sides, rather conspicuously
reticulated with transverse veinlets, abrupt or retuse at the base, short-
petiolulate. Peduncles shorter than the leaj, an inch or less in length,
2-4-flowered. Pedicels opposite in pairs, 3 or 4 lines long. Flower
almost an inch in length. Calyx deeply four-toothed, subtended by a
small deciduous bractlet; the teeth nearly equal in length, fully as
long as the tube; the uppermost ovate, very obtuse, or slightly emar-
ginate; the others subulate. Corolla “ yellow or yellowish :” yexillum
on a very short claw, obovate, entire, with two semilunar callosities
and two short and auriculate deflexed basal appendages: ale nar-
rowly oblong, straightish, shorter than the carina, furnished with a
spur-like appendage at the base of the lamina. Carina straightish
or merely arcuate, as long as the vexillum, somewhat beaked ; its
petals partly united, similar, each with a minute inflexed auricle at
the base of the lamina. Stamens diadelphous: the free filament not
geniculate or appendaged at the base. Ovary hirsute, many-ovuled.
Style filiform, slightly incurved, much longer than the ovary, the
upper part not dilated, and only slightly flattened on the back, the

452 PHANEROGAMIA.

vexillar face densely and strongly bearded for a considerable length
below the stigma, which is lateral, but almost apical, since only a
minute lamellar appendage projects beyond it. Legume linear, elon-
gated, almost 3 inches long, 3 lines wide, manifestly flattened, straight,
or nearly so, glabrate, somewhat torulose, intercepted between the
seeds. These are 8 or 10 in number, oblong-oval, compressed, chest-
nut-brown, shining; the oblong hilum covered with an inconspicuous
strophiole.

Without doubt, this well-marked species belongs to Vigna, as the
genus is received by Mr. Bentham, notwithstanding the compressed
legume and the straightish keel.

Piate 50.—VienA SANDWICENSIS: of the natural size. Fig. 1.
Corolla displayed. 2. Stamens and style. 38. Calyx and pistil. 4.
Upper part of the style and the stigma, more magnified. 5. A seed.
6. Embryo.—The analyses more or less magnified.

4, VIGNA LUTEA.

Dolichos luteus, Swartz, Fl. Ind. Oce. 8, p. 1246? DC. Prodr. 2, p. 898? Hook. &
Arn. Bot. Beech. Voy. p. 81.

D. luteolus ? Guillem. Zeph. Tait. p. 64, an Linn. ?

Scytalis anomala, Vogel, Rel. Meyen. p. 83? et

Vigna anomala, Walp. Repert. 1, p. 779?

Has. Sandwich Islands: on the shores at Oahu; and on mountains
of the west part of Maui. Society Islands, at Tahiti and Eimeo; on
the coast. (Probably introduced.) Mangsi Islands.

The seeds are dark chestnut-brown.—Perhaps this passes into J.
Sinensis ; but the leaflets are mostly roundish and very obtuse, espe-
cially when it grows on the sea-shore. I am doubtful whether it is
the Scytalis anomala of Vogel, which also came from Oahu.

In the collection from Sydney, New South Wales, is a flowerless
specimen, apparently of a Vigna, near the South African V. hirta,
and the Australian V. lanceolata, but not the same. It was also
gathered by Col. Mitchell.

LEGUMINOSE. 453

76. DOLICHOS, Linn., Benth.
1. Doxicnos FALCATUS, [Klein.
Dolichos falcatus, Klein, in Willd. Spec. Pl. 3, p. 1047; DC. Prodr. 2, p. 398.

Has. Shores of Laguna, at Bajios, Luzon.

77. LABLAB, Adans.
1. LABLAB VULGARIS, Save.

“Has. Feejee and Sandwich Islands: introduced and partially
naturalized.

78. FAGELIA, Necker.
1. Fagenia pituminosa, DC.

Has. Cape of Good Hope, in the neighbourhood of Cape Town.

79. CAJANUS, DC.
1. Casanus Inpicus, Spreng.

Cajanus Indicus, Spreng. Syst. 3, p. 248; Wight & Arn. Prodr. Ind. Or. 1, p. 256.
C. flavus & C. bicolor, DC. Prodr. 2, p. 406.

Haz. Luzon. Sandwich Islands: “cultivated.” Rio Janeiro,

Brazil. Introduced from India, to which the Pigeon Pea is indigenous.
114

454 PHANEROGAMIA.

80. RHYNCHOSIA, Lour.
1. Ruyncnosia mintma, DC.

Has. St. Jago, Cape de Verde Islands. Doubtless introduced from
America.

81. FLEMINGIA, Roxb.

1. FLEMINGIA STROBILIFERA, 22. Br.

Flemingia strobilifera, R. Br. in Ait. Hort. Kew. 4, p. 350; DC. Prodr. 2, p. 351;
Wight, Ic. Pl. Ind. Or. t. 267.

Has. Philippine Islands; Luzon, near Manilla and Bajios.

82. HECASTOPHYLLUM, P. Browne, Kunth.
1. HecastopHyLttum Brownel, Pers.

Has. Rio Janeiro, Brazil; common along the shores of the bay.

83. MACH ZRIUM, Pers.

1. Macuarium Ancustirotium, Vogel.

Macherium angustifolium, Vogel, in Linnea, 11, p. 193; Benth. Comm. Legum.
in Ann. Wein. Mus. 1. ¢. p. 33.

Has. Rio Janeiro, Brazil. (Without flowers or fruit.)

LEGUMINOSAE. 455

92. MAcuarium acutirotium, Vogel.

Macheriwm acutifolium, Vogel, in Linnea, 11, p. 187, non Mart. in Benth. Comm. 1. c.
M. muticum, Benth. Comm. Legum. 1. ¢. p. 36.

Has. Rio Janeiro and Organ Mountains, Brazil. (In flower.)

3. MAcHARIUM OBLONGIFOLIUM, Vogel.
Macheriwm oblongifolium, Vogel, in Benth. Comm. Legum. ]. ¢. p. 36.

Has. Organ Mountains, Brazil. (In fruit.)

4, MACH&RIUM INCORRUPTIBILE, Benth.

Macherium incorruptibile, Benth. Comm. Legum. 1. ¢. p. 37.
Nissolia incorruptibile, Velloz. Fl. Flum. 7, t. 82.

Has. Rio Janeiro, Brazil. (In fruit.)

From the account of Dr. Allemfo, in the Transactions of the Vel-
losian Society of Rio Janeiro, as noticed by Mr. Bentham (in the
Kew Journal of Botany, for September, 1853), it at length appears
to be settled that the rosewood of the cabinet-makers is furnished by
this tree and some other species of Macherium.

84. PONGAMIA, Lam.

1. PongamiA GLABRA, Vent. (Tab. 53.)

Pongamia glabra, Vent. Hort. Malm. t. 28; Wight & Arn. Prodr. Fl. Ind. Or. 1,
p. 262; Wight, Ic. Pl. Ind. Or. t. 99; Benth. in Pl. Jungh. fase. 2, p. 251.
Pongamia, Lam. Ill. Gen. t. 603, f. 1.

Has. Feejee Islands: on the mountains of Muthuata; in flower:

456 PHANEROGAMIA.

Vanua-levu, &c.; in fruit. Mindanao, Philippine Islands, near Cal-
dera.—The specimens from the Feejee Islands belong to. the variety
with rather narrow leaflets: in that from Caldera (in fruit only) the
leaflets are broader.

PuatE 53.—PoNGAMIA GLABRA: a flowering branch, from the Feejee
plant. Fig. 1. Flower and pedicel. 2. Corolla displayed. 3. Sta-
mens and pistil. 4. Pistil and ovules—The analyses enlarged.

85. MILLETTIA, Wight & Arn.

Miuterria, Wight & Arn. Prodr. Fl. Ind. 1, p. 263; Benth. in Pl. Jungh. p. 249.

1. Mittertia? Luzonensis, Sp. Nov.

M. scandens? glabrata; foliolis 7 ovatis seu ovalibus acuminatis mem-
branaceis supra glabris subtus glaucis; racemis axillaribus sessilibus
confertifloris folio brevioribus ; pedicellis solitariis vel geminis calyce
truncato subcequilongis ; vexillo glabro ecalloso; ovario glaberrimo.

Has. Shores of Laguna, Bafios, Luzon.

There is only a miserable flowering specimen of this plant in the
collection, which, from the floral characters, must belong to Millettia,
as the genus is defined by Mr. Bentham (in Pl. Jungh.), since the
vexillum is not auriculate, and the stamens are diadelphous: but it
does not accord with any described species. The plant is apparently
a clumbing or sarmentose, stout, shrubby vine, glabrous, except a
minute pubescence on the young parts. Leaves pinnate. Lea/lets 7,
ovate, or nearly oval, acuminate, about 24 inches long, pale or glaucous
underneath, membranaceous. eacemes axillary, shorter than the leaf,
sessile, being floriferous quite to the base. Pedicels crowded, either
single or in pairs on the nodes of the rhachis, about the length of the
truncate calyx, 3 lines long. Corolla 7 or 8 lines long: vewillwm gla-'
brous, not callose at the base, not auricled, oval or roundish. Carina
very obtuse; the apex incurved. Ovary glabrous, linear, several-
ovuled. Fruit unknown. |

LEGUMINOSE. 457

86. DERRIS, Lour.
1. Derris uLicinosa, Benth.

Derris uliginosa, Benth. in Pl. Jungh. fase. 2, p. 252.

Pongamia uliginosa, DO. Prodr. 2, p. 416; Wight & Arn. Prodr. Ind. Or. 1, p. 262.

P. religiosa (sphalm.), Wight, in Hook. Bot. Misc. 3, p. 301, & in Suppl. t. 41,
sub nom. P. triphylla.

Has. Philippine Islands, near Caldera, Mindanao. Feejee Islands,
at Rewa, &. (In fruit.) Tongatabu. (In fruit.)

The specimens from Tongatabu frequently have seven leaflets,
instead of five, or even three, but apparently do not otherwise differ
from the Indian D. uliginosa. In the figure cited above, the tenth -
stamen is represented as distinct; but in the description it is properly
said to be united with the sheath of filaments except at the base and
summit.

87. DALBERGIA, Linn. f

DaLBerGiA, Linn. f. Suppl. p. 52; Benth. Pl. Jungh. fase. 2, p. 254.

1. DALBERGIA MULTIJUGA, Graham.
Dalbergia multijuga, Graham, in Wall. Cat. no. 5865, non H. Meyer.

Has. Singapore—A specimen without flowers or fruit, probably
belonging to this species, but glabrous or nearly so.

2. DALBERGIA CASSIOIDES, Wall.

Dalbergia cassioides, Wall. Cat. no. 5861; Benth. Pl. Jungh. fasc. 2, p. 254, in
adnot.

Has. Philippine Islands, at Caldera, Mindanao. (Without flowers

or fruit.)
115

458 PHANEROGAMIA.

3. DALBERGIA TORTA, Graham.
Dalbergia (Selenolobium) torta, Graham, in Wall. Cat. no. 5789; Benth. 1. c.
Has. Feejee Islands: the particular habitat not recorded.

The specimens, in flower only, accord with one in Griffith’s col-
lection from Malacca, so named by Mr. Bentham.

4. DALBERGIA (TRIPTOLEMEA) VARIABILIS, Vogel.
Dalbergia variabilis, Vogel, in Linnzea, 11, p. 196.
Triptolemea latifolia, ovata, montana, &c., Mart. in Benth. Comm. Legum. p. 38.

Pterocarpus frutescens, Velloz. Fl. Flum. 7, t. 96.

Has. Brazil, at Rio Janeiro (the Triptolemea montana, Mart.), and
Organ Mountains (the 7! ovata, Mart.).

5. DALBERGIA (‘TRIPTOLEMEA) MYRIANTHA, Benth.

Triptolemea myriantha, Mart. in Benth. Comm. Legum. 1. c. p. 88.

Has. Organ Mountains, Brazil. (In flower only.)

88. EDWARDSIA, Salisb.
1. EDWARDSIA GRANDIFLORA, Salish.
Edwardsia grandiflora, Salisb. in Linn. Trans. 9, p. 209; DO. Prodr. 2, P- 97; A.
Rich. Bot. Voy. Astrolab. 1, p. 244; Hook. f. Fl. N. Zeal. p. 02.
Sophora tetraptera, Linn. f.; Forst. Prodi: p. 82; Curt. Bot. Mag. t. 167.

Var. @. MICROPHYLLA (Hook. f. 1. c.): foliolis parvis.

Edwardsia microphylla, Salisb. 1. c.; DC. 1.c.; Bot. Mag. t. 1442; A. Rich. |. .
Sophora microphylla, Ait.; Jacq. Hort. Schoenb. t. 269.

LEGUMINOSAE. 459

Has. Bay of Islands, New Zealand: in fruit; with foliage of the
smaller-leaved state, which, as Mr. Bentham remarks (in Hook. f. FI.
N. Zeal.) is not to be definitely distinguished, even as a variety, from
the larger-leaved form of the species.

2. EDWARDSIA CHRYSOPHYLLA, Salish.

Edwardsia chrysophylla, Salisb. in Linn. Trans. 9, p. 802, t. 26, f. 1; Ker. Bot.
Reg. t. 788; DC. l.c.

Var. G. GLABRATA: foliis calycibusque glabratis ; floribus minoribus.

Has. Sandwich Islands (where it was first collected by Menzies) ;
on the mountains above the forest zone. Mouna Kea, Hawaii, occu-
pying a zone reaching to an elevation of 11,000 feet; and Mouna Loa,
above the crater Lua Pele. Also on the mountains of Kauai; and of
the eastern part of Maui, on the banks of the crater Haleakala.
Var. @. Hawaii, 1,000 feet above Puna.

The Mamani of the natives is a small tree, 20 or 30 feet high,
somewhat variable in its foliage, pubescence, &c.; the silkiness of the
lower face of the leaflets being sometimes of a golden or tawny hue,
and sometimes almost silvery. In the variety @. this pubescence is
shorter as well as more scanty, and the calyx is still more glabrate ;
but I perceive no farther difference, except that the flowers are rather
smaller. In all, the upper surface of the leaflets becomes glabrous
and more or less shining. The yellow flowers are not quite an inch
in length. The broad vexillum is recurved, as noted by Mr. Ben-
tham in this and in two East Indian species. The stamens are as long
as the keel, but not exserted. Ovary tomentose. Legume about 4
inches long, 6—9-seeded, more or less constricted between the seeds,
quadrate by the 4 narrow wings, which, with the whole corky epicarp,
separate in a somewhat bivalvular manner at full maturity, when the
chartaceous and moniliform endocarp becomes more or less two-valved.
Seeds oval, somewhat compressed, chestnut-colour.

The unpublished species of Edwardsia, from the Sandwich Islands,
mentioned by Mr. Bentham, in Dr. Hooker’s New Zealand Flora,

460 PHANEROGAMIA.

cited above, is our var. 8. I believe it is only a state of EL. chryso-
phylla, occurring at a lower elevation than the ordinary form of the
species. .

89. SOPHORA, Lunn, FR. Br.
1. SoPHORA MACROCARPA, Smith.

Sophora macrocarpa, Smith, in Rees, Cycl. no. 6; DC. Prodr. 2, p. 96; Benth. 1. c.
Edwardsia Chilensis, Miers, Trav. Chil.; Hook. in Bot. Reg. t. 1798, & Bot. Mise.
3, p. 177; Gay, Fl. Chil. 2, p. 215.

Has. Chili; common in the neighbourhood of Valparaiso.

2. SOPHORA TOMENTOSA, Linn.

Has. On the coast, at Rio Janeiro, Brazil: also at the Society,
Friendly, Feejee, and Mangsi Islands.

90. VIRGILIA, Lam.
1. Vire@it1aA CAPENSIS, Lam.

Has. Cape of Good Hope, in the vicinity of Cape Town.

91. GOURLIEA, Gillies.
1. GOURLIEA DECORTICANS, Gillies.
Gourliea decorticans, Gillies, in Hook. Bot. Misc. 3, p. 208, t. 106.

Has. Rio Negro, North Patagonia. (Without flowers or fruit: the
inflorescence in a diseased state.)

LEGUMINOS &. A461

Susorv,. LL CHSALPINIE &.

92. GUILANDINA, Juss.
1. GuitanpiIna Bonpwuc, Att.

Has. On the coast at Rio Janeiro, and of the Friendly, Navigators’,
Feejee, and Sandwich Islands.

Two forms, noted by Dr. Pickering as species, were observed in the
South Sea Islands, one of them with tall and climbing stems.

93. CHSALPINIA, Linn.
1. CHSALPINIA SAPPAN, Linn.

Has. Small island in the Sooloo Sea. (In fruit.)

2. CHSALPINIA GLABRATA, H. B. KX.

Cesalpinia glabrata, H. B. K. Nov. Gen. & Spec. 6, p. 826; DC. Prodr. 2, p. 482.
C. corymbosa, Benth. Pl. Hartw. p. 117.

Has. Peru; between Lima and Obrajillo.

Apparently intermediate between the C. glabrata and C. corymbosa,
which may with probable reason be united.

3, CRSALPINIA (POINCIANA) PULCHERRIMA, Swartz.

Has. Rio Janeiro, Brazil. Tahiti, Society Islands. “Cultivated.”
Also cultivated at the Friendly and Navigators’ Islands.

Another species was noticed in the Organ Mountains, Brazil; but

there are no specimens in the collection.
116

462 PHANEROGAMIA.

94. COULTERIA, AB. XK.
1. CovuLtTEeRIA TINCTORIA, H. B. K.
Has. The specimen is ticketed “Tahiti?” But it was probably

gathered in Peru. There is, moreover, a sterile branch in the collec-
tion from near Lima, either of this species or of C. horrida.

9. HOFFMANSEGGIA, Cav.

1. HorrManseceia viscosa, Hook. & Arn.

Hoffmanseggia viscosa, Hook. & Arn. Bot. Misc. 3, p. 209; Benth. in Gray, Pl.
Wright. 1, p. 57.
Larrea viscosa, Ruiz & Pay. Fl. Peruv. t. 377 (ined.).

Has. Peru; on a mountain ridge near Yanga.

96. CASSIA, Linn.
1. Cassia ALATA, Linn.

Has. Shores of Laguna, Baiios, Luzon, Philippine Islands.

2. CASSIA OCCIDENTALIS, Linn. &
Var. 8. SoPHERA: foliolis numerosioribus (legumine turgido breviore.)
Cassia Sophera, Linn, Wight & Arn., etc.

Has. St. Jago, Cape de Verde Islands. Upolu, Samoan Islands:

LEGUMINOS&Z. 463

“introduced, and as yet rare.” Var. 8. Feejee Islands. (A form
with more numerous leaflets, in which respect C. Sophera alone differs
from C. occidentalis, and with shorter and rather turgid pods, only
two inches long.)

8. CASSIA TOMENTOSA, Linn. f.

Has. Hunter’s River, New South Wales. Probably introduced ?

4, CASSIA OBLONGIFOLIA, Vogel.
Cassia oblongifolia, Vogel, Syn. Cass. p. 23, & in Linnea, 11, p. 666.

Has. Brazil, near Rio Janeiro: in sand, near the coast,

5. CASSLA OBTUSIFOLIA, Linn.

Has. St. Jago, Cape de Verde Islands. Probably introduced.

6. Cassra GAupicHAupi, Hook. & Arn.

Cassia Gaudichaudii, Hook. & Arn. Bot. Beech. Voy. p. 81; Vogel, Syn. Cass. p. 26.
CO. montana, Nutt. ined. in Herb. Hook.

Has. Mountains behind Honolulu, Oahu, Sandwich Islands. (Also
gathered by Gaudichaud, Lay & Collie, Nuttall, &c.)

This shrubby species is pretty well characterized in both the works
above cited. When the leaflets are reduced to three pairs, the gland
is not between the lowest, but on the petiole at some distance below:
this gland, at first clavate, as described by Vogel, at length becomes
filiform and nearly 2 lines long. The leaflets vary from two-thirds of
an inch to 2+ inches in length, and are soon glabrate: so are the pods
in most cases, although sometimes finely pubescent, even at maturity.
They are broadly linear and very flat, like those of C. glauca, mem-
branaceous, acuminate or abruptly pointed at both ends, somewhat

464 PHANEROGAMIA.

stipitate, varying from 13 to 42 inches long, from 4 to 6 lines in
width. Seeds from 8 to 10, flat (parallel with the valves), nearly
orbicular, blackish.

7. Cassta @LAuCcA, Lam.

Has. Feejee Islands; on Ovolau and Oneata.

8. CASSIA MULTIUGA, Rich.

Cassia multijuga, Rich. in Act. Soc. Par. p. 188; Vogel, in Linnza, 15, p. 69.
C. calliantha, Meyer, Prim. Esseq. p. 169; Vogel, Syn. Cass. p. 31.

Has. Brazil, near Rio Janeiro, and in the Organ Mountains.

9, CASSIA LATOPETIOLATA, Dombey ?
Cassia latopetiolata, Dombey, ex Vogel, Syn. Cass. p. 29?

Has. Andes of Peru, near Obrajillo. (The specimen too imperfect
for proper determination.)

10. CASSIA COLUTEOIDES, Colladon.
Cassia coluteoides, Colladon, Mon. Cass. p. 102, t. 12; Vogel, Syn. Cass. p. 42, & in
Linnea, 15, p. 72.
C. flexuosa, Bertero, in Not. Pl. Rar. Genev. 7, p. 34, t. 6, non Burm.

C. Candolleana, Vogel, Syn. Cass. p. 42.
C. frondosa, Hook. & Arn. Bot. Beech. Voy. p. 22, 75, & Bot. Misc. 8, p. 210.

Has. Chili, near Valparaiso.

11. Cassta uirsuta, Linn. /.

Has. Peru, between Lima and Obrajillo. (In fruit.)

LEGUMINOS &. 465

12. CASSIA ZYGOPHYLLA, Benth.
Cassia zygophylla, Benth. in Mitch. Jour. Trop. Austral. p. 258.

Has. New South Wales: in the interior. (A nearly glabrous
variety : probably not distinct from C. eremophila, Cunningham, and of
Brown in Sturt. Exped. Centr. Austral.)

13. CASSIA ACIPHYLLA, Benth. ined.

Has. Hunter's River, New South Wales; in fruit: accompanied
by a flowering specimen, which is noted as having been presented by
Mr. Cunningham.

A species allied to C. artemisioides, Gaud., C. Sturtii, R. Br., &.
The flowering specimen is canescently sitions: with the leaves soon
glabrate. Leaflets 4 to 8 pairs, linear, about an inch long and a line
or two in width, coriaceous, with revolute margins, and a sharp cuspi-
date point: a subulate gland interposed between most of the pairs:
the rhachis terminated by a subulate process. Peduncle two-flowered,
about the length of the leaves. Corolla yellow. Legume linear, flat,
scarcely stipitate, from 3 to 5 inches long, falcate, obtuse, glabrous.

14. CASSIA AUSTRALIS, Sims.

Cassia australis, Sims, Bot. Mag. t. 2676; Bot. Reg. t. 1322; Vogel. Syn. Cass. l. c.
CO. umbellata, Reichenb. Ic. Exot. t. 206.

Has. Newington, New South Wales.

15. Cassta Apoucouita, Awbl.

Cassia Apoucouita, Aubl. Pl. Guian. 1, p. 379, t. 116; Vogel, in Linnea, 15, p. 72.
OC. ramiflora, Vogel, Syn. Cass. p. 48.
C. ensiformis, Velloz. Fl. Flum. 4, p. 79.

Has. On the Corcovado, near Rio Janeiro, Brazil.
117

466 PHANEROGAMIA.

16. Cassta RAMOSA, Vogel.
Cassia ramosa, Vogel, Syn. Cass. p. 55, & in Linnea, 15, p. 72.

Has. Brazil, in the immediate neighbourhood of Rio Janeiro.

17. CASSIA ROTUNDIFOLIA, Pers.

Cassia rotundifolia, Pers. Ench. 1, p. 456; Vogel, Syn. Cass. p. 57.
C. bifoliolata, DC.; Colladon, Mon. Cass. p. 120, t. 9.

¢

Has. Brazil; with the preceding.

18. CAssiA CHAMACRISTA, Linn. y. BRASILIENSIS, Vogel.
Cassia Chamecrista, Linn., var. y. Brasiliensis, Vogel, Syn. Cass. p. 63. An spec.?

Has. Brazil, in the vicinity of Rio Janeiro.

19. CasstA PAvontana, Don.

Cassia Pavoniana, Don, Syst. Gard. & Bot. 2, p. 447.
C. calycioides ? Hook. Bot. Mise. 2, p. 217.

Has. Peru, near Callao and Lima.

Probably, as Vogel suggests, a mere variety of C. Chameecrista.

20. CasstA LESCHENAULTIANA, DC.

Has. Singapore.

From Manilla there is also the foliage of an undetermined species,

apparently the same as No. 968 of Cuming’s Philippine Islands’ col-
lection.

LEGUMINOSS. 467

97. SWARTZIA, Willd.
1. SWARTZIA GRANDIFLORA, Willd.
Swartzia grandiflora, Willd. Spec. Pl. 2, p. 1219; DC. Prodr. 2, p. 422.

Has. Rio Janeiro, Brazil. Indigenous?

2. SWARTZIA ELEGANS, Schott.
Swartzia elegans, Schott. in Spreng. Syst. Veg. Cur. Post. p. 407.

Has. Organ Mountains, Brazil.

98. AFZELIA, Smith.

1, Arzerza suuGa. (Tab. 51.)

A. glabra; foliolis bijugis (summisve raro unijugis) ovatis nitidis ; pani-
culis corymbosis; bracteis bracteolisque parvis caducis ; vexillo bre-
vissime unguiculato, alis aut parvulis aut nullis ; staminibus fertilibus
3, sterilibus T anantheris minimis ; ovario sessilt ; leguminibus oblongis
planis mono—pentaspermis demum bivaloibus ; seminibus compressis
hilo nudis.

Macrolobium bijugum, Colebr. in Trans. Linn. Soc. 5, p. 12.

Outea bijuga, DC. Prodr. 2, p. 511.

Intsia 2. Amboinensis, DO. Prodr. 2, p. 509? (Metrosideros Amboinensis, Rumpb.
Amboin. 8, p. 21, t. 10), ex Benth.

Has. Feejee Islands: Ovolau and Somu-somu, near the coast.

“A spreading tree, 50 feet high,” glabrous throughout, except a
minute and cinereous pubescence on the inflorescence and flower-buds.

468 PHANEROGAMIA.

Leaves abruptly pinnate. Petioles with the rhachis an inch to 2
inches in length. Stipules not seen. Leaflets 2 pairs, or the upper-
most reduced to a single pair (or rarely 3 pairs, according to Dr.
Pickering’s memoranda), ovate, either obtuse, acutish, or somewhat
acuminate, coriaceous, somewhat Jucid, and of nearly the same hue
both sides, 3 or 4 inches long, veiny, on partial petioles of 2 or 3 lines
in length. Panicle terminal, small, consisting of short and corymbose
racemes, Only flower-buds occur on the specimens: the largest of
these are but 2 or 3 lines long, obovoid, on pedicels of about the same
length: these are articulated with the rhachis, subtended by a small
ovate bract, and furnished about the middle with a pair of similar
bractlets, that do not enclose even the half-grown bud: the bracts and
bractlets both caducous. Calyx four-parted; the segments ovate,
obtuse, nearly equal. Veaillwm in the bud roundish, plicate, auricu-
late at the base, and with a short claw, at this period not longer than
the auricles, the margins undulate: its position is opposite the poste-
rior sepal, and its size in the bud is about the same. Lateral petals,
or alee much smaller, about half the length of the vexillum, semi-ovate
with the upper edge straight, raised on a short and slightly pubescent
claw: but in the greater number of flower-buds examined the ale are
altogether wanting, as are the anterior petals in all of them. Fertile
stamens 3, being the anterior and two lateral of the outer series, on
filiform and at length elongated slightly hairy filaments: anthers
oblong, fixed by the middle, two-celled; the cells opening longitu-
dinally. Sterile stamens 7 (or sometimes fewer), reduced to very small
and naked. filaments, one of them interposed between each fertile sta-
men, the others belonging to the posterior side of the flower. Ovary
sessile or nearly so, compressed, semi-ovate, silky-pubescent, containing
from 0 to 7 roundish ovules in a single series. Style filiform, spirally
involute in the bud: stigma terminal, subcapitate. Legume oblong,
flat, coriaceous, glabrous, from 5 to 8 inches long, about 2 inches wide,
at length two-valved; the valves externally reticulated, within fur-
nished with narrow transverse projections, which intercept the cavity
between the seeds. Seeds from 2 to 5, or sometimes solitary, compressed,
orbicular, an inch or more in diameter, not avillate nor appendaged at
the small hilum; the bony but thin testa smooth, involved in a
covering of furfuraceous matter; which when fresh is probably juicy.

Cotyledons flat and thin, orbicular, enclosing the short and straight
retracted radicle.

LEGUMINOS2. 469

“The Outea byuga, DC. is certainly closely allied to the African
Afzelia, Anthonota, Intsia, and Berlinia, and also to the American
EHperua, Parivoa, Outea, and Vouapa: but if these are kept distinct,
Outea is certainly the one of them all it is farthest from in habit
and in the pod. I am not acquainted with the Intsia Madagasca-
riensis; but the L Amboinensis is probably a congener of, if not the
same species as Outea bijuga, and I think, as far ‘as we know both
Intsia and Outea bijuga, cannot well be generically separated from
Afzelia. Intsia Madagascariensis is said to have unequally pinnate
leaves; but the allied genera Hperua and Parivoa appear to have
them both abruptly and unequally pinnate on the same plant.” Ben-
tham, in litt.

PLATE 51.—AFzELIA BIUGA: a branch, with fruit and flower-buds,
of the natural size. Fig. 1. A flower-bud, with its pedicel and bract-
lets. 2. Diagram of the flower. 38. An opened flower-bud. 4.
Petals, from the same. 5,6. Fertile stamens. 7. Pistil. 8. Same,
with the ovary divided longitudinally—All these details magnified.
9. A legume, of the natural size, with one valve removed. 10. A
seed divided.

99. BAUHINIA, Linn., Benth.
1. BAuninia Brasitiensis, Vogel.
Bauhinia Brasiliensis, Vogel, in Linnxa, 18, p. 302.

Has. Brazil; common in the Organ Mountains, and near Rio
Janeiro.

2. BAUHINIA GRANDIFLORA, Juss.
Bauhinia grandiflora, Juss. in Poir. Dict. Suppl. 1, p. 600; DC. Prodr. 2, p. 513.
Has. Peru, in the vicinity of Lima. “The tree perhaps not there

indigenous.”
118

470 PHANEROGAMIA.

100. PILIOSTIGMA, Hochst.

Prrzosti¢ma, Hochst. in Flora, 1846, p. 598; Benth. in Pl. Jungh. fase. 2, p. 261.

1. Pittostiama ActpuM, Benth. 1. c.

Bauhinia acida, Reinw. in Korth. Verhand. Nat. Gesch. Kruidk. p. 86.
B. purpurea, DC. Prodr. 2, p. 515, non Linn.
Casparea castrata, Hassk. Pl. Jav. Rar. 1, p. 412.

Has. Philippine Islands: in mountains, near Bafios, Luzon.

101. SCHNELLA, Raddi.

1. SCHNELLA MACRosTACHYA, Ltaddt.

Schnella macrostachya, Raddi, Pl. Bras. Addend. in Mem. Moden. 18, p. 411.
Bauhinia ungulata, Linn. Spec. Pl. p. 535, ex Spreng.
B. Raddiana, Bongard, in Mem. Acad. Petrop. (1836), t. 3?

Has. Near Rio Janeiro, Brazil. (Foliage only.)

There is also foliage of what is probably another species resembling
S. splendens, but with almost lanceolate and acuminate leaflets.

102. CYNOMETRA, Linn.
1. CYNOMETRA GRANDIFLORA, Sp. Nov. (Tab. 52.)
C. glaberrima ; foliolis bijugis ovalibus oblongisve inequilateris scepis-

sume emarginatis coriaceis; racemis corymbosis confertifloris ; stami-
nibus 21-32; ovario levi stylo breviore.

LEGUMINOSA, 47]

Has. Ovolau and Vanua-levu, Feejee Islands.

“A spreading tree, about 40 feet high; the branchlets verrucose,
and, with the leaves, flowers, &c., entirely glabrous. Stipules linear,
very caducous. Leaflets two pairs, of equal size and shape, oval or
oblong, 2 or 3 inches in length, more or less wnequal-sided, either
obtuse or more or less acuminate, but the apex almost always nar-
rowly emarginate, coriaceous in texture, dull or pale, sometimes a little
shining above, inconspicuously feather-veined and reticulated. Petiole
somewhat woody. Buds of the inflorescence (sessile in the axils and
terminal) and also the leaf-buds, large and scaly, two-thirds of an inch
long, ovoid-fusiform ; the scales coriaceous, orbicular or ovate, finely
striate, minutely rusty-pubescent externally, regularly imbricated,
deciduous before the flowers expand; the uppermost passing into
spatulate and linear and chartaceous deciduous bracts, which are as
long as the pedicels. Racemes short and corymb-like, nearly sessile,
often as long as the leaves; the lowers (apparently flesh-colour)
crowded, on pedicels of about an inch in length. At the base of each
is a pair of subulate, stipule-like bractlets, which are sparsely hispid
on the midrib. Tube of the calyx turbinate, fleshy, free from the
substipitate base of the ovary, which it closely encloses; the lobes 4,
petaloid, oblong, naked and glabrous, half an inch long. Petals 4,
spatulate-lanceolate, as long as the lobes of the calyx, inserted on its
throat. Stamens 21 to 32, inserted with the petals: filaments fili-
form, an inch in length: anthers oval-oblong, subcordate, with a
rather fleshy connective; the two cells longitudinally dehiscent for
their whole length. Ovary flat, oblong, somewhat falcate, smooth and
glabrous, or slightly pubescent towards the apex, two-ovuled, tapering
into a slender, filiform, and -straightish style, which is terminated by
a depressed discoid stigma. Mature fruit not seen.

There are two or three forms of this species in the collection, vary-
ing principally in the breadth of the leaflets. One, from Ovolau
(Plate 52, C), has broadly oval and rather thinner leaflets, somewhat
shining above, and may possibly belong to a different species; but it
is not in flower, and its large scaly buds are leaf-buds only.

Puate 52,—CyYNOMETRA GRANDIFLORA: A, B. Flowering specimens,
two forms, of the natural size. C. Branchlet of a variety with

472 PHANEROGAMIA.

broader and thinner leaflets, and large scaly leaf-buds. Fig. 1. A
detached flower from B, with its pedicel, bract, and bractlets. 2. A
bractlet, magnified. 3. Calyx and pistil. 4. Vertical section of a
flower, enlarged. 5, 6. Stamens, enlarged; the former seen poste-
riorly, the latter anteriorly.

2. CYNOMETRA FALCATA, Sp. Nov.

C. foliolis unyjugis glaberrimis ovato-lanceolatis obliquis falcatis coria-
ceis; floribus sine pedunculo; fasciculatis decandris ovario lunato
pubescente stylo recurvo longiore.

Has. Feejee Islands, at Ba, on the western coast of Viti-levu.

Branches slender, minutely verrucose. Petioles at most 2 lines in
length, bearing a single pair of leaflets. These are ovate-lanceolate,
oblique, and falcate, gradually acuminate, with the apex rather blunt,
2 or 8 inches long, coriaceous, very glabrous, and somewhat shining
both sides, minutely feather-veined. The scaly flower-buds are ovoid,
axillary, only 2 or 3 lines long; the scales orbicular and striate. In
the solitary specimen most of these are undeveloped: but the single
undeveloped flower in the axil of the scales is seen to be laterally
enveloped by a pair of diaphanous and ciliate-fringed bractlets; and
the sepals appear to be obscurely puberulent, but not tipped with a
tuft of hairs. They enclose only 10 stamens, or sometimes perhaps
fewer. The scaly buds that have expanded display a sessile Jascicle,
or rather a small corymb (the axis 2 or 3 lines long), bearing a number
of pubescent pedicels (2 or 3 lines long), from which all but the ovary
has fallen. The ovary is lunate, tomentulose-pubescent, a line and a
half long, and rather longer than the recurved style by which it is
pointed. Fruit unknown.

This species belongs to Wight & Arnott’s first subgenus, or Cyno-
metra proper, although the sepals are not “tipped with a tuft of
hairs.” I have not the means of comparing it with the C. rami-
flora, of Linnzus. The flowers are very small, as compared with

the preceding species, and apparently much smaller than those of @
polyandra.

LEGUMINOSS. 473

3. CYNOMETRA INAQUIFOLIA, Sp. Nov.

©. foliolis bijugis coriaceis glabris reticulatis valde incequilateris ineequa-
libus, duobus inferioribus ovalibus quam superioribus oblongis multo
minoribus.

Has. Philippine Islands, near Bafios, Luzon. (Also collected by
Mr. Cuming: no. 1297.)

The specimen exhibits the foliage only, with a young scaly bud.
Leaflets 2 pairs, coriaceous, glabrous, pale beneath, lucid above, loosely
reticulate-veined, obtuse or emarginate, very unequal-sided, especially
the terminal pair, which also is very much larger than the lower pair:
the two upper are oblong, 6 or 7 inches long, and almost dimidiate ;
the lower are oval, 2 to 24 inches long, about the length of the
rhachis with the petiole. Fruit and flowers unknown.

Susorv. LI]. MIMOSE &.
103. ENTADA, Adans.
1. ENTADA SCANDENS, Benth.
Mimosa scandens, Linn. Spec. Pl. ed. 2, p. 1501; Swartz, Obs. p. 389.
Entada scandens, Benth. in Hook. Jour. Bot. 4, p. 382.
E. Gigalobium, Purscetha, monostachya, & Adenanthera, DC. Mem. Legum. &

Prodr. 2, p. 425.

Has. Samoan and Feejee Islands; climbing over tall trees: appa-
rently indigenous. (Remarkable for the extraordinary size of its pods.)

104. PIPTADENTA, Benth.
1. PrerapEenta communis, Benth.

Piptadenia communis, Benth. in Hook. Jour. Bot. 4, p. 3387.
119

474 PHANEROGAMIA.

Has. Brazil, near Rio Janeiro. (The leaves of one specimen are
in a monstrous state, probably from the sting of insects.)

2. PIPTADENIA MACROCARPA, Benth. 1. c.

Has. Brazil; with the preceding species.

105. ALGAROBIA, Benth.

ALGAROBIA, Benth. Pl. Hartw. p. 18; Torr. & Gray, Fl. N. Amer. 1, p. 399; Gray,
Pl. Lindh. 1, p. 85, & Pl. Wright. 1, p. 60.
Prosopis, § ALGAROBIA, Benth. Mimos. in Hook. Jour. Bot. 4, p. 348.

1. ALGAROBIA JULIFLORA.

Mimosa juliflora, Swartz, Prodr. Fl. Ind. Oce. p. 85.
Prosopis juliflora, DC. Prodr. 2, p. 447; Benth. in Hook. Jour. Bot. 1. ¢.
#~. affinis, Spreng. Syst. Veg. 2, p. 826.

Has. Rio Negro, North Patagonia; common on the plains. In
fruit: with foliage also of a small-leaved form, the Prosopis dulcis,
y. australis, Benth. 1. ¢.

The pods as well as the foliage so closely resemble those of our
Texan and Mexican Mezquite (A. glandulosa, Torr. & Gray) as to
render it most probable that we have one polymorphous species,
extending from Arkansas to Patagonia, as Mr. Bentham suggests.
In either case, the plant before us must bear the name of juliflora.
The fruit shows well the character of the genus, namely, the separa-
tion of the endocarp into a series of detached and closed joints, or
pips, each containing a seed.

106. STROMBOCARPA (Benth. sub Prosop.).

Prosopis, § SrromBocarPA, Benth. Mimos. in Hook. Jour. Bot. 4, p. 351, excl. sp.
StromBocarpa, Gray, Pl. Wright. 1, p. 60.

LEGUMINOSAE. 475

1. STROMBOCARPA STROMBULIFERA.

Mimosa strombulifera, Lam. Dict. 1, p. 15.
Acacia strombulifera, Willd. Spec. 4, p. 1055; DC. Prodr. 2, p. 455.
Prosopis strombulifera, Benth. Mimos. 1. c.

Has. Rio Negro, North Patagonia. (In flower only.)
This and S. reptans, take the place in the pampas of the southern
temperate zone on this continent that S. pubescens, S. cinerascens, and

S. odorata do in the arid plains of Texas, New Mexico, and the inte-
rior of California.

107. NEPTUNIA, Louwr., Benth,
1. NEPTUNIA PUBESCENS, Benth.
Neptunia pubescens, Benth, Mimos. in Hook. Jour. Bot. 4, p. 364.

Has. Peru; in the vicinity of Callao.

2. NEPTUNIA GRACILIS, Benth. 1. 6.

Has. New South Wales; in the district of Hunter’s River.

108. DESMANTHUS, Willd., Benth.

1. DESMANTHUS DEPRESSUS, Kunth.

Desmanthus depressus, Kunth, in Willd. Spec. Pl. 4, p. 1546, & Pl. Legum. p. 119,
t. 85.

Has. Peru; in the vicinity of Callao and Lima. (In fruit.)

476 PHANEROGAMIA.

109. MIMOSA, Linn., Willd.
1. Mruosa ALBIDA, Kunth.

Mimosa albida, Kunth, Mimos. t. 1; DC. Prodr. 2, p. 426; Benth. Mimos. in Hook,
Jour. Bot. 4, p. 860.

Has, Peru; in the vicinity of Callao and Lima.

2. Mimosa VELLoziANna, Mart.

Mimosa Velloziana, Mart. Herb. Fl. Bras. p. 185; Benth. Mimos. 1. ¢. p. 361.
M. viva, Velloz. Fl. Flum. 11, t. 33, non Linn.

Haz. Organ Mountains, Brazil, in the neighbourhood of Rio
Janeiro; very common.
3. Mimosa pupica, Linn.
Has. Tahiti and Tongatabu: where the Sensitive Plant occurs only
as “an introduced weed, around dwellings.”
4, Mimosa SEPIARIA, Benth.
Mimosa sepiaria, Benth. Mimos. in Hook. Jour. Bot. 4, p. 395.

Has. Brazil, in the vicinity of Rio Janeiro; very abundant.

5. Mimosa ELLIPTICA, Benth. 1. c.

Has. Brazil; with the preceding species.

LEGUMINOSS. 477

6. Mimosa Giuuiesu, Benth.

Mimosa Gilliesii, Benth. Mimos. in Hook. Jour. Bot. 4, p. 413.
Prosopis globosa, Gillies, in Hook. Bot. Misc. 3, p. 205.

Has. Rio Negro, North Patagonia: on the sandy plains.

This and M. ephedroides, the other species of the section, appear to
be very closely allied. The pods, as well as the spinescent Broom-
like branchlets, of our plant are quite glabrous; but the capitula are
peduncled, and there are a few leaflets. Flowers not seen. The ripe
pods, with the replum, break up into indehiscent articles.

110. LEUC ANA, Benth.
1. Levcana Forster, Benth.

Leucena Forster’, Benth. Mimos. in Hook. Lond. Jour. Bot. 5, p. 94.
Mimosa glandulosa, Soland. in Forst. Prodr. Fl. Ins. Austr. p. 92.
Acacia insularum, Guillem. Zeph. Tait. p. 66.

Has. Point Venus, Tahiti. Vanua-levu, Feejee Islands. Tongatabu.

There is nothing to add to Guillemin’s detailed description of this
plant (which is rightly referred to Leucena by Bentham), except that
the mature pods in our specimens are plane.

111. ACACIA, Linn., Benth.
x PHYLLODINEH Australasice.
1. Acacia Brownu, Steud.

Acacia Brownti, Steud. Nomencl. Bot. ed. 1, p. 2; DC. Prodr. 2, p. 449; Benth.

Mimos. in Hook. Lond. Jour. Bot. 1, p. 338.
120

478 PHANEROGAMIA.

Has. New South Wales, in the neighbourhood of Sydney.

2. ACACIA JUNIPERINA, Welld.

Acacia juniperina, Willd. Spec. Pl. 4, p. 1049; Vent. Hort. Malm. t. 46; Benth. 1. c.
A. echinula, DC. Prodr. 2, p. 449.

Haz. New South Wales, near Sydney.

3. AcAcIA Bauer, Benth.

Acacia Bauert, Benth. Mimos. in Hook. Lond. Jour. Bot. 1, p. 344.

Has. New South Wales, near Woolongong, &c.

4, ACACIA HISPIDULA, Willd.

Acacia hispidula, Willd. Spec. Pl. 4, p. 1054; Lodd. Bot. Cab. t. 823; Hook. Ic.
Pl. t. 161.

Has. New South Wales; abundant in the neighbourhood of
Sydney.

5. ACACIA VESTITA, Ker.
Acacia vestita, Ker, Bot. Reg. t. 698; Benth. Mimos. 1. c. p. 348.

Has. New South Wales.

6. ACACIA CULTRIFORMIS, A. Cunn.

Acacia cultriformis, A. Cunn. in Don, Gard. Dict. 2, p. 406; Hook. Ic. Pl. t. 170.

Has. Hunter’s River, New South Wales.

LEGUMINOS 4. 479

7. ACACIA MYRTIFOLIA, Willd.

Acacia myrtifolia, Willd. Spec. Pl. 4, p. 1054; Lodd. Bot. Cab. t. 772; Benth, l.c.

Has. New South Wales, near Sydney.

- 8. ACACIA SUAVEOLENS, Willd.

Has. New South Wales, near Sydney and Hunter’s River: with a
narrower-leaved form, perhaps A. angustifolia, Wendl.

9, ACACIA LINIFOLIA, Willd.

Has. New South Wales, in the vicinity of Sydney, &c.

10. ACACIA PROMINENS, A. Cunn. var. LATIFOLIA, Benth. ined.

Has. New South Wales. (A. Cunninghams)

11. AcAcIA ELONGATA, Sieber, in DC.

Has. New South Wales. (A. Cunningham.)

12. ACACIA MELANOXYLON, It. Br.

Has. New South Wales, near Sydney, Newington, &.: very
common.

18. ACACIA LINEARIS, Sums.

Has. Sydney, New South Wales; where it appears to be common.

480 PHANEROGAMIA.

14. AcAcrA FLoRIBUNDA, Willd.

Has. New South Wales (the particular station not recorded).

15. ACACIA LONGIFOLIA, Willd.

Has. New South Wales, near Newington, &. (Also St. Helena:
doubtless from a cultivated plant.) Principally a broad-leaved form.

16. ACACIA GLAUCESCENS, Willd.

Has. New South Wales (A. Cunningham); with a shorter and
broader-leaved variety.

x * PHYLLODINE® Oceanice.

17. AcAcIA Koa.

A. glabra; ramulis viz’ angulatis; phyllodiis elongatis falcatis lanceo-
latis basi vel utrinque angustatis coriaceis multinerviis hine inde
folio bipinnato terminatis ; foliolis oblongis emarginatis 12-libjugis ;
pedunculis solitariis vel brevissime racemosis; capitulis dense multi-
Jloris ; calyce breviter dentato, dentibus flavo-tomentosis subglandulosis ;
legumine lato-lineari plano glabro, suturis angustissime marginatis,

Acacia heterophylla, Hook. & Arn. Bot. Beech. Voy. p. 81; Benth. Mimos. 1. c. p.
368, excl. syn.

Has. Sandwich Islands: Oahu; on the mountains behind Hono-
lulu: Hawaii; on the sides of Mouna Loa and Mouna Kea.

A large tree; glabrous, or the nascent parts puberulent; the
branches terete: branchlets slightly angled, not striate. Phyllodia
Jalcate, lanceolate, coriaceous, rarely oval-oblong, 4 to 6 inches long,
varying from 3 or 4 lines to an inch or more in breadth, narrowed at

LEGUMINOSS. 481

the base, and often tapering towards the apex, which is either acute or
obtuse, often sphacelate and rounded; the smooth surface is striate
with many nerves, of which 4 or 5 are stronger and more conspicuous.
On younger plants especially, some of the phyllodia bear a bipinnate
leaf, of few pinne; the leaflets 12 to 15 pairs, oblong, emargimate,
crowded. Peduncles solitary or fascicled in the axils, or often several
and somewhat racemose on a short rhachis, about half an inch long,
bearing a dense many-flowered head, of 4 lines in diameter. Calyx
turbinate, glabrous, except the rounded and very short teeth, 5 in
number, which are densely bearded or tomentose with a yellow and
somewhat glandular pubescence. Petals 5, oblong-lanceolate, glabrous,
more or less united, about one-third longer than the calyx, half the
length of the stamens. Ovary pubescent, or when sterile glabrous,
subsessile, oblong. Legume broadly linear, straight or slightly falcate,
3 to 6 inches long, two-thirds or three-fourths of an inch broad, gla-
brous, flat, obtuse at both ends, about twelve-seeded ; the thin valves
transversely venulose; the sutures not thickened, furnished with an
acute but very narrow margin.

This, the Kou of the natives, is one of the largest and most impor-
tant timber trees of the Sandwich Islands: from its trunks the canoes
of the Hawaiians are made. Mr. Bentham’s var.? /atifolia is only a
state of the species. Our specimens are wholly destitute of leaflets.

Gaudichaud, who had the advantage of seeing both growing in
their native countries, pronounced the tree of the Sandwich Islands
identical with the original A. heterophylla, Willd. (Mimosa heterophylla,
Lam. excl. 3.) of Isle Bourbon, which moreover has long been culti-
vated in botanic gardens. Hooker and Arnott, adopted this view
(and even referred to it the A. lawrifolia, Willd.), in which they were
followed by Bentham, who inadvertently overlooked the Bourbon
plant, and cited no habitat except the Sandwich Islands. In distin-
euishing the two trees, peculiar to these most widely separated
stations, perhaps I incur the charge of being influenced by geogra-
phical considerations rather than botanical characters: for neither
the flowers nor the phyllodia of the two differ very much; and, for
lack of proper materials, I cannot complete the comparison, possessing
no leaflets of the present plant, and no pods of the original A. hetero-
phylla. I feel confident, however, that sufficient differences will be

found.
121

482 PHANEROGAMIA.

18. ACACIA LAURIFOLIA, Welld.

Acacia laurifolia, Willd. Spec. Pl. 4, p. 1053; Labill. Sert. Austro-Cal. p. 68, t. 68 ;
Benth. Mimos. 1. c. p. 369.

Mimosa Mangium, Forst. Prodr. Fl. Ins. Austr. p. 75, non Linn.

M. simplicifolia, Linn. f. Suppl. p. 436.

Has. Friendly, Navigators’ or Samoan, and Feejee Islands: along
the shores.

The phyllodia vary from 2 to 5 inches in length, and from 12 to 3
inches in width. There is likewise a form from Vanua-levu, Feejee
Islands, with narrower and mostly pointed phyllodia. The pods are
sharply margined, as described by Mr. Bentham; leaving no doubt
that the species is distinct from the foregoing, to which Hooker and
Arnott united it.

19. AcactA Ricuu, Sp. Noy. (Tab. 53.)

A. glaberrima; ramulis angulatis; phyllodiis lanceolatis acuminatis
subfalcatis chartaceo-coriaceis multinerviis, apice scepius hamato;
pedunculis fasciculatis capitulo parvo multifloro longioribus; calyce
dentato; leguinine oblongo plano glabro, marginibus acutis angustis-
suis.

Has. Feejee Islands; common in barrens, at Sandalwood Bay,
Vanua-levu, and Naloa.

“Tree 20 or 30 feet high, with the habit of a Willow,” entirely gla-
brous; the branchlets angled, slender. Leaves (phyllodia) lanceolate,
from 12 to 3 inches long and 23 to 6 lines wide, tapering tec an acute
apex, which is often hooked or curved, slightly falcate, chartaceo-coria-
ceous in texture, sessile, thickened, but not glanduliferous at the base,
closely many-nerved, and with minute connecting veinlets. Peduncles
Jascicled in the axils, from 8 to 8 together, filiform, 3 or 4 lines in
length, bearing single small and many-(16—20-) flowered capitula, of
only 2 lines in diameter (including the stamens). Calyx glabrous,

LEGUMINOS &A 483

obtusely five-toothed. Petals oblong, thrice the length of the calyx.
Ovary glabrous. Legume oblong, perhaps also oblong-linear, flat, gla-
brous, 7 or 8 lines wide; the valves thin, reticulated; the margins
acute but much less salient than in A. laurifolia.

Prats 53, B—Acacia Rica: a flowering branch, of the natural
size. Fig. 1. A leaf, capitula, &., enlarged. 2. A flower, magnified.
3. Posterior view of the same. 4. A petal, magnified. 5. A stamen,
magnified. 6. Pistil, equally magnified, the ovary divided. 7. Legume.
8. A seed, magnified. 9. Vertical section of the same.

» * * BoTRYCEPHALE (Australasice).
20, ACACIA ELATA, A. Cunn.
Acacia elata, A. Cunn.; Benth. Mimos. in Hook. Lond. Jour. Bot. 1, p. 388.

Has. New South Wales, near Newington.

21. ACACIA SPECTABILIS, A. Cunn., l. ¢.

Has. New South Wales. (A. Cunningham.)

22. ACACIA DISCOLOR, Willd.

Has. Sydney, New South Wales.

23. ACACIA DECURRENS, Willd.

Has. New South Wales: probably from the vicinity of Sydney.

24. ACACIA MOLLISSIMA, Willd.

Has. New South Wales, near Sydney and Hunter's River :—a
variety (stenoloba) with narrower pods than ordinary.

484 PHANEROGAMIA.

* * * * GUMMIFERZ.

25. ACACIA PELLACANTHA, Vogel.

Acacia pellacantha, Vogel, in Rel. Meyen. p. 45; Benth. Mimos. in Hook. Lond.
Jour. Bot. 1, p. 390.

Has. Peru; around Callao and Lima.

26. AcactA FarRNESIANA, Willd.

Has. St. Jago, Cape de Verde Islands. Tahiti, Society Islands.
Bafios, Luzon.—Doubtless introduced, into all these stations, from
America.

27. Acacta CAVENIA, Hook. & Arn.
Has. Chili; common near Valparaiso.—Probably, as Mr. Bentham
intimates, not distinct from the preceding species.

28. ACACIA ALBIDA, Delile.

Has. St. Jago, Cape de Verde Islands.

* x x * * VULGARES (Americana).
29, ACACIA LACERANS, Benth.
Acacia lacerans, Benth. Mimos. in Hook. Lond. Jour. Bot. 1, p. 511.

Has. Brazil, near Rio Janeiro.

30. ACACIA PTERIDIFOLIA, Benth. 1. c.

Has. Brazil, in the vicinity of Rio J aneiro.

LEGUMINOSEZ. 485

112, PITHECOLOBIUM, Mart., Benth.
1. Prruecotopium DULCE, Benth.
Pithecolobium dulce, Benth. Mimos. in Hook. Lond. Jour. Bot. 3, p. 199.

Has. Luzon, Philippine Islands; introduced from Tropical America.

2. PirnecoLopium CiyPEARIA, Benth, 1. c.

Has. Singapore.

113. SERIANTHES, Benth.

1, SERIANTHES MYRIADENIA, Planchon, ined.
Acacia myriadenia, Bertero, in Guillem. Zeph. Tait. p. 65, excl. syn.

Has. Eimeo and Tahiti, Society Islands: in forests. Somu-somu,
&c., Feejee Islands: on the banks of streams.

The specimens bear only fruit, and a few young flower-buds. The
ripe pod accords with the character given by Bentham for 8. grandi-
flora, except that it is not bullate, but finely wrinkled transversely,
and ferrugineous. The flowers are noted by Bentham, in Hook.
Lond. Jour. Bot. 5, p. 108. In Dr. Pickering’s notes this is said to
be a large tree at Eimeo, with the foliage of Zumarindus: at the
Feejee Islands, a spreading tree, 40 feet high.

9, SERIANTHES VITIENSIS, Sp. Nov.

S. foliolis ellipticis emarginatis utrinque letevirentibus supra lucidis
122

486 PHANEROGAMIA.

penninerviis ; glandulis petiolaribus nullis; legumine tenwi-lignoso
tomentello, valvis laxe venosis.

Has. Feejee Islands, at Sandalwood Bay, Vanua-levu, Xe.

This is nearly related to S. grandiflora, Benth.: but the pod is
thinner; the pinne and the leaflets are rather fewer; there are no
petiolar glands, and none between the pinne; but they occasionally
occur between the uppermost leaflets. The leaflets are bright green
both sides, shining above, glabrous, scarcely oblique, elliptical, emar-
ginate, about half an inch long, pinnately veined. Flowers not seen.
Legume oblong, 5 or 6 inches in length and 2 inches wide, thin
and plane, twisted, minutely tomentose with a rusty down, scarcely
thickened at the margins; the thin but woody valves strongly cohe-
rent between the seeds, loosely veined with a few salient and narrow
veins, which branch into slender reticulated veinlets. Seeds trans-
verse, oblong, about 8 lines in length.

114. INGA, Plumier, Mart.
1. INGA SEMIALATA, Mart.

Inga semialata, Mart. Herb. Fl. Bras. p. 114; Benth. Mimos. in Hook. Lond. Jour.
Bot. 4, p. 588.
Mimosa semialata, Velloz. Fl. Flum. 11, t. 5.

Has. Brazil; in the Organ Mountains.

2. Inca Fevrmnuet, DC.

Has. From a garden at Lima (without flowers or fruit) : cultivated
under the name of Pacai.

The foliage accords with the figure of the Pucai in Feuillée, except
that the leaflets are rather obtuse at both ends. There are no mate-
rials for clearing up this obscure species.

LEGUMINOS&. 487

3. Inca CuminciANA, Benth. 1. c.
Has. Peru; abundant between Lima and Obrajillo.
The leaflets are either obtuse and pointless, or mucronate, or
caudate-acuminate.
4, InGA AFrFinis, DC.

Inga affinis, DC. Prodr. 2, p. 483; Benth. Mimos. in Hook. Lond. Jour. Bot. 4,
p. 619.
I. dulcis, & I. Velloziania, Mart. Herb. Fl. Bras. p. 118.

Has. Organ Mountains, Brazil.

115. AFFONSEA, S&. Hil.
1. AFFONSEA BULLATA, Benth.
Affonsea bullata, Benth. Mimos. in Hook. Lond. Jour. Bot. 5, p. 77.

Has. Brazil; near Rio Janeiro. (With flower-buds only.)

Orv. CHRYSOBALANACE &.

1 PARINARIUM, Juss.

Parinarium (Parinari, Aubl.), Juss. Gen. p. 8342; DC. Prodr. 2, p. 526; Benth.
in Hook. Jour. Bot. 2, p. 211, & Niger Fl. p. 335.

1. Parinartum (PETROCARYA) INSULARUM, Sp. Nov. (Tab. 54.)

P. foliis lanceolato-ovatis seu oblongo-lanceolatis basi subcordatis vel
obtusissimis supra nitidis subtus incanis; stipulis lanceolato-subulatis
petiolo eglanduloso duplo longioribus; floribus cymoso-paniculatis ;
staminibus fertilibus 6-8, sterilibus 2-4; drupa biloculari scepius
disperma.

Has. Feejee Islands: Sandalwood Bay, Vanua-levu, and Ambau.
Samoan or Navigators’ Islands: the habitat not recorded.

A “tree 50 feet high, with spreading branches; the trunk 24 feet
in diameter.” Branchlets tomentose, at length glabrate; the older
ones very warty. Leaves lanceolate-ovate, varying to oblong-lanceolate,
usually acuminate, subcordate or very obtuse at the base, 3 to 5, and on
sterile shoots 6 to 8 inches long, thickish, copiously feather-veined,
and beneath with rather conspicuously reticulated veinlets, the
nascent ones clothed on both sides with a rusty or whitish wool,
the upper surface soon glabrous and shining, the lower canescent or
whitened with a fine and close pubescence. Stipules lanceolate-subu-
late or linear, fuscous, about half an inch long, twice the length of the
petiole, which, as well as the base of the leaf, is glandless. Inflores-
cence terminal, cymose-paniculate, tomentose, only seen with unde-

CHRYSOBALANACE &. 489

veloped flower-buds, when it is shorter than the adjacent leaf.
Flowers bibracteolate, small; the buds only 2 lines in length. Tube
of the calyx short, gibbous, villous inside; the 5 lobes ovate, acute.
Petals 5, not longer than the calyx, ciliolate, 2 or 3 of them (on the
side next the ovary) ovate, the others oblong-lanceolate. Stamens
apparently varying from 8 to 12, of which from 6 to 8 are fertile, and
2 to 4, on the side remote from the ovary, sterile; the latter being
reduced to small and naked tooth-like processes or filaments: anthers
didymous, two-celled. Ovary coherent to the gibbous side of the
calyx-tube, two-celled, villous, as is the base of the basilar style.
Ovule solitary in each cell, anatropous; the rhaphe on the side next
the dissepiment. Fruit an ovoid drupe, of nearly 2 inches in length,
slightly flattened laterally; the epicarp smooth; the putamen very
thick and woody, with its surface little roughened, two-celled, and
usually two-seeded: but no well-formed and mature seeds are found in
the fruits preserved in the collection.

In a drawing made by the Botanical Artist of the Expedition, the
flower is represented as having fourteen fertile stamens, occupying a
complete circle, and no sterile filaments. But all the young flower-
buds I could venture to dissect were as represented in the above
description and the accompanying analyses.

The species is, without doubt, nearly related to P. Sumatranum,
Benth. (Pétrocarya Sumatrana, Jack); which is said to have the
elliptical-oblong leaves acute at the base, and seven fertile stamens on
one side of the flower, and as many abortive ones on the other.

Puate 54, B.—PARINARIUM INSULARUM: a branch, with flower-buds,
and the summit of another, with a fruit, of the natural size. Fig. 1.
An opened flower-bud. 2. Vertical section of the same. 3. One of
the anterior petals. 4. One of the posterior petals. 5, 6. Stamens,
seen anteriorly and posteriorly—These details variously enlarged.
7. Transverse, and 8, longitudinal section of the drupe, of the natural
size.

2. Partnartum? MarGarata, Sp. Nov. (Tab. 54.)
P. glabrum ; foliis oblongis basi acutis eglandulosis ; drupa septo tenur

evanido demum scepe uniloculart; cotyledonibus conferruminatis.
123

490) PHANEROGAMIA.

Has. Feejee Islands; in the Sandalwood district of Vanua-levu.
(In fruit only.)

This is recorded in Dr. Pickering’s notes under its native name of
“ Margarata.” He states that the fruit is “indehiscent, two-celled,”
and that it is “used to blacken the face and hair.” The specimens
consist of a leafy branch, entirely glabrous, with oblong, mostly acu-
minate, ample leaves (from 7 to 10 inches long), acute at the base,
glandless, rather thin, dull, sparsely feather-veined, the veins con-
nected by inconspicuous reticulated veinlets; the petiole about half
an inch long; with which are two separate, unripe drupes. The
latter are oval, slightly flattened laterally, an inch and a half or more
in length, and an inch and a quarter in width: one of them exhibits
a thin and evanescent partition: the other has become one-celled by
the obliteration of this partition, and contains a single seed, in which
the unripe and partly decayed or injured cotyledons are seen to be
strongly conferruminate, thus appearing not unlike the albumen of a
Nutmeg.

Piate 54, A.—PaArinarium? MArcarata. Fig. 1. A leaf, with a
part of the branch. 2. A drupe. 3. Transverse section of the same
and of the contained seed. 4. Vertical section of an empty and infer-
tile drupe, showing the thin partition.—All of the natural size.

3. PARINARIUM (SARCOSTEGIA) LAURINUM, Sp. Nov. (Tab. 55.)

P. ramis gunioribus subsericeis ; foliis oblongis acuminatis basi rotundata
biglandulosis glabris lucidis; stipulis linearibus caducis; jfloribus
racemosis ; calyce infundibuliformi subobliquo, ore ceequali, lobis obo-
vatis obtusis petalis brevioribus; staminibus fertilibus circiter 15,
anticis elongatis, sterilibus T-10 brevissimis dentiformibus.

Has. Samoan or Navigators’ Islands: the particular habitat not
identified.

Apparently a tree, with glabrate branches; the young branchlets
and inflorescence silky-pubescent. Leaves oblong, inclining to lanceo-
late, more or less acuminate, rounded at the base, where there is a pair

CHRYSOBALANACES. AQ]

of glands attached to the base of the midrib: they are glabrous, 6 to
8 inches long and 2 or 3 inches wide, rather coriaceous and /ucid,
especially the upper surface, copiously feather-veined; the veins 10
to 12 pairs, divergent, and connected by an elaborate network of
veinlets. Petiole only 2 or 3 lines in length. Stipules linear, 5 or 6
lines long, very caducous. Flowers crowded in axillary or terminal,
short-peduncled, sometimes geminate racemes, which are shorter than
the leaves, more or less spreading. Bracts oblong-linear, coriaceous,
caducous. Pedicels articulated with the axis, 12 to 22 lines long,
bibracteolate near the summit; the bractlets oblong-linear, longer
than the pedicel, shorter than the calyx-tube, deciduous. Calyx
externally silky-canescent; the tube funnelform or club-shaped, a little
oblique, 3 or 4 lines long, not thickened nor fleshy; the orifice equal,
the interior villous with reflexed hairs, in the manner of the genus;
the lobes obovate or oblong, obtuse, shorter than the obovate-orbicular
and deciduous petals; the latter nearly equal, convolute in estivation,
except the posterior, which is wholly external in most cases. Stamens
monadelphous at the very base (or raised on a narrow annular disk),
inserted on the throat of the calyx; the 14 or 15 anterior fertile, with
elongated and filiform filaments (involute in sestivation); the posterior
7 to 10, reduced to as many rigid and short teeth. Ovary two-celled,
with a single erect ovule in each cell, very villous, its long stipe
laterally adnate to the anterior side of the calyx-tube quite up to
the throat; the free side densely bearded with deflexed villous hairs.
Style basal, filiform, as long as the stamens, excessively villous at the
base, the upper part glabrous, the apex tipped with a simple stigma,
Fruit not seen.

This species is manifestly related to P. Jackianum, Benth. (Petro-
carya excelsa, Jack), and to P. Grifithianum, Benth.; but it has not
the fleshy calyx which Bentham assigns to his section Sarcostegia.
Having a racemose inflorescence and short pedicels, this species 18 con-
venient for determining the position of the several floral organs in
relation to the bract and axis. We find the odd sepal to be anterior,
with a slight obliquity: it nearly, but not exactly, subtends the ovary,
as in Leguminose, being between that and the bract, but a little to the
right of the median line (vide Fig. 4): thus essentially confirming the
view indicated by Mr. Bentham, in Hooker’s Journal of Botany, 2,
p. 211.

492 PHANEROGAMIA.

Puate 55.—PARINARIUM LAURINUM: a flowering branch, of the
natural size. Fig. 1. Stipules, and base of a leaf, showing the glands.
2. Flower-bud, with the bract and two bractlets. 8. An opening
flower, with the bractlets. 4. Diagram of the flower, with the bract
and bractlets. 5. A petal, from Fig. 8. 6. A flower from which the
petals have fallen. 7. Anthers. 8. Vertical section of the calyx,
ovary, &c. 9. Longitudinal section through the partition.—All the
details enlarged.

2 HIRTELLA, Linn.
1. HIRTELLA OBLONGIFOLIA, DC.

Has. Brazil, near Rio Janeiro. (Perhaps a variety of H. Ameri-
cana.) The flowers are tetrandrous.

There is besides a branchlet of a Chrysobalanaceous plant, from the
Organ Mountains, near Rio Janeiro, probably a Licania, which I
cannot identify, and the forming flower-buds are too young for inves-
tigation.

Orv ROSACE &.

1. QUILLATIA, Molina.
1. QUILLAIA SAPONARIA, Molina.

Quillaia saponaria, Molina, Hist. Nat. Chil. p. 187; Hook. & Arn. Bot. Mise. 38,
p. 804; Gay, Fl. Chil. 2, p. 274.

Q. Moline & Q. Smegmadermos, DC. Prodr. 2, p. 547.

Has. Chili; common from Valparaiso to the Cordilleras above
Santiago.

2, KAGENECKIA, Ruz & Pov.

, 1. KAGENECKIA OBLONGA, Ruiz & Pav.

Kageneckia oblonga, Ruiz & Pav. Fl. Peruv. & Chil. Prodr. t. 87; Hook. & Arn.
Bot. Misc. 3, p. 304.

K. crategoides, Don, in Edinb. Phil. Jour. ex Hook. & Arn.

Has. Chili: in ravines; common from Valparaiso to the lower
Cordilleras.

2. KAGENECKIA LANCEOLATA, Ruiz & Pav. l. e.

Has. Peru; “between Yaso and Obrajillo; rare.”

(Specimens not
found in the collection.)

124

494

HAB.

Has.

Has.

HAs.

Has.

Has.

Has.

Has.

PHANEROGAMIA.

3. CLIFFORTIA, Linn.
1. CLIFFORTIA ILICIFOLIA, Linn.

Cape of Good Hope, in the immediate vicinity of Cape Town.

2. CLIFFORTIA RUSCIFOLIA, Jann.

Cape of Good Hope; with the preceding species.

3. CLIFFORTIA ODORATA, Linn. f.

Cape of Good Hope; with the preceding species.
4, CLIFFORTIA STROBILIFERA, Linn.
Cape of Good Hope; with the preceding species.

5. CLIFFORTIA SARMENTOSA, Linn.

Cape of Good Hope: on the sea-shore.

6. CLIFFORTIA TRIFOLIOLATA, Linn.

Cape of Good Hope, in the vicinity of Cape Town.

7. CLIFFORTIA DENTATA, Willd.

Cape of Good Hope; with the preceding species.

8. CLIFFORTIA FALCATA, Linn. f.

Cape of Good Hope; with the preceding species.

ROSACES. 495

4, MARGYRICARPUS, Ruiz & Pav.
1. MARGYRICARPUS SETOSUS, Ruiz & Pav.

Has. Rio Negro, North Patagonia; common on the sand-hills.
Chili, near Valparaiso.

5. TETRAGLOCHIN, BPepp.

1. TETRAGLOCHIN STRICTUM, Poepp.

Tetraglochin strictum, Poepp. Fragm. Syn. p. 26; Gay, Fl. Chil. 2, p. 281.
Margyricarpus alatus, Gillies, in Hook. Bot. Misc. 3, p. 3065.

Has. High Andes of Peru, above Bajos.

The specimen, although destitute of flowers and fruit, appears to
be the same as the Chilian plant.

6. AC ANA, Linn.

* Austro-Americane.,
1. Aca&naA puMILA, Vahl.
Acena pumila, Vahl, Enum. 1, p. 298; Hook. f. Fl. Antare. p. 264, t. 105.

Has. Orange Harbour, Fuegia. (In flower and fruit.)

2. Acm#NA Maceinanica, Vahl.

Acena Magellanica, Vahl, Enum. 1, p. 207; Hook. f. Fl. Antare. p. 267.
Ancistrum Magellanicum, Lam. Ill. Gen. t. 22, f. 2; Poir. Suppl. 1, p. 346.

Has. Orange Harbour, Fuegia. (In flower.)

496 PHANEROGAMIA.

3. ACHENA OVALIFOLIA, Ruiz & Pav.

Acena ovalifolia, Ruiz & Pav. Fl. Per. & Chil. 1, p. 67, t. 103; Hook. f. 1. ¢.
Ancistrum repens, Vent. Hort. Cels. t. 5.

Has. Orange Harbour, Fuegia.

4, ACHNA ADSCENDENS, Vahl.

Aceena adscendens, Vahl, Enum. 1, p. 207; DC. Prodr. 2, p. 593; Hook. f. Fl.
Antarc. p. 268, t. 106.
Ancistrum humile, Pers. Ench. 1, p. 30.

Has. Orange Harbour, Fuegia. (In flower and fruit.)

). Acana Antarctica, Hook. /.

A. pumiia; caulibus brevibus procumbentibus ; foliis confertis ; Joliolis
3-Ajugis late oblongis grosse serratis medio canaliculatis subtus ad
costam venasque prominentes et supra preesertim villoso-sericeis ; pedun-
culo scapiformi ; capitulo globoso; calycibus glabris; fructu obconico
aristis 4 brevibus apice glochidiatis armato.

Acena Antarctica, Hook. f. Fl. Antare. p. 269.
Has. Orange Harbour, Fuegia. (In fruit.)

This rare species was gathered on Staten Land by Menzies, and
Hermite Island, Cape Horn, by Dr. Hooker; in both cases with the
flowers and fruit fallen. Our specimens afford the fruit; but the
stigma and stamens are still unknown. It is a small plant, with lig-
nescent and short procumbent stems, clothed with the glabrous petiolar
stipules, and with crowded leaves. Leaflets 3 or 4 pairs, with an odd
one, 2 or 3 lines in length, subcoriaceous, elliptical or broadly oblong,
coarsely serrate with obtuse teeth, channelled above in the middle; the
upper surface silky with dense villous hairs, of a silvery-fulvous hue,

ROSACES 497
partially glabrate with age, the lower similarly silky on the very
prominent midrib and straight veins, in the intervals puberulent or
glabrate. Peduncle scape-like, naked, slightly pubescent, in fruit 2
inches long, bearing a globose head of scarcely more than 2 lines
in diameter. Bractlets obovate-spatulate, sparingly pilose-ciliate at
the apex. Sruiting calyx glabrous, obconical; its small lobes oval,
thickish, shorter than the four spine-like short awns, which are
' minutely glochidiate at the apex, and hardly as long as the fruit.

6. ACHNA MYRIOPHYLLA, Lindl.
Acena myriophylla, Lindl. Bot. Reg. sub. t. 1271; Hook. & Arn. in Bot. Mise.

Has. Chili; in the Andes above Santiago. (Without flowers or
fruit.) —Probably only a form of A. pinnatifida.

7. ACHNA TRIFIDA, Ruiz & Pav.
Acena trifida, Ruiz & Pav. Fl. Per. & Chil. 1, p. 67, t. 104; Gay, Fl. Chil. 2, p. 292.
Has. Chili; common near Valparaiso.
To this belongs the Californian plant, referred to A. pinnatifida by
Hooker & Arnott, and Torrey & Gray.
8. ACHNA SPLENDENS, Hook. & Arn.
Aceena splendens, Hook. & Arn. in Bot. Mise. 8, p. 886; Gay, Fl. Chil. 2, p. 292.
Has. Chili; on the first Cordillera above Santiago. (The flowers
and fruit fallen.)
9, ACHNA LAPPACEA, Ruiz & Pav.

Aceena lappacea, Ruiz & Pay. Fl. Per. & Chil. 1, p. 66, t. 108.
125

498 PHANEROGAMITIA.

Has. High Andes of Peru, between Obrajillo, Culnai, and Bajos.
(Also gathered by Matthews and M’Lean.)

There are fragments apparently of another species from the Andes
of Peru, which is indeterminable, for the want of flowers and fruit.

* * Novo-Zelandice & Novo-Hollandice.
10. Acmna ovina, A. Cunn.

Aceena ovina, A. Cunn. in Field, N. 8. Wales; Benth. Pl. Hugel. p. 48.

Has. Hunter's River, New South Wales. (Also collected by
Labillardiere.)

11. Acmna SANGuUISORBA, VaAl.

Acena Sanguisorbe, Vahl. Enum. 1, p. 294; DO. 1. ¢.; Hook. f. Fl. N. Zeal. p. 54.
Ancistrum Sanguisorbe, Linn. f. Suppl. p. 189.

A. anserinifolium, Forst. Char. Gen. t. 2.

A. diandrum, Forst. Prodr. Fl. Ins. Austr. p. 10.

Has. Bay of Islands, New Zealand. Sydney, New South Wales.

* x * Sandwicenses.
12, AczNA Exicua, Sp. Nov.

A. foliis in caudicibus multicipitibus rosulatis confertissimis ; foliolis
7-8jugis minimis confertis ovatis oblongisve seepius bi—tridentatis seu
bilobis coriaceis supra bullatis venosis glaberrimis subtus canescentibus ;
scapo gracilt parce bracteato; spica parva cylindrica ; calycis lobis
4—5 oblongis, tubo ovoideo undique setis glochidiatis armato; stami-
nibus 2; stigmate patelleeformi fimbriato.

Has. Sandwich Islands: on the table-land of the mountains of
Kauai, in a marsh.

ROSACE &. 499

A diminutive species, with short branching caudexes, crowned with
a very dense rosulate cluster of leaves, each half an inch or an inch long,
including the petiole, which is winged for nearly its whole length by
the adnate and scarious, brownish, glabrous stipules. Leaflets 7 or 8
pairs, crowded, very small, about a line in length, ovate or oblong, com-
monly two-lobed or two-toothed, sometimes three-toothed, otherwise entire,
rather coriaceous in texture, the convex upper surface bullate, reticu-
lately veiny under a lens, glabrous and somewhat shining; the lower
surface concave, whitened with a very minute close-pressed pubescence,
and with stouter hairs on the veins and midrib; the latter somewhat
projecting into a bearded point. Scape slender, 4 to 6 inches high,
minutely pubescent, leafless, beset with a few subulate bracts, which
are chiefly opposite: from most of them in the collection the flowers
and fruit have wholly fallen; but one or two, in an effete state, show
the inflorescence to be a small, cylindrical spike, not more than half
an inch in length. Flowers minute, crowded, subtended by subulate
or linear bracts. Calyx-lobes 4 or 5, oblong, glabrous; the ovoid tube
covered with slender bristles (as in Huaccena), which are minutely
retrorsely barbed at the apex, otherwise smooth. The flowers are pro-
bably hermaphrodite: at least, in a spent flower, the remains of two
filaments were detected, and also a sessile, dilated and depressed, fim-
briate stigma, Anthers not seen. Mature fruit unknown.

Incomplete as the specimens are, they suffice to show that this is a
new and peculiar species of Acena. The leaflets are not larger than
the leaves of many a Jungermannia, which, indeed, they considerably
resemble, except in their rigidity.

7 ALCHEMILLA, Tourn.
1. ALCHEMILLA TRIPARTITA, Ruiz & Pav.

Alchemilla tripartita, Ruiz & Pay. Fl. Per. & Chil. I, p. 68; Hook. Bot. Mise. 2,
p. 218.

Has. Andes of Peru, near Bafios, (Also gathered by Cruck-
shanks, M’Lean, &c.)

500 PHANEROGAMIA.

2. ALCHEMILLA PINNATA, Ruiz & Pav.? l.c.

Has. Andes of Peru; with the preceding. (Without flowers or
fruit.)

8. POTENTILLA, Lunn.
1. PoTENTILLA PROCUMBENS, Sibth.

Potentilla procumbens, Sibth.; Koch; Lehm. Nov. Recen. Potentill. (1851), p. 71.
P. Tormentilla, var. nemoralis, Seringe, in DC. Prodr. 2, p. 574.

Has. Madeira.

9 FRAGARIA, Linn.
1. FRAGARIA vEscA, Linn.

Has. Madeira. Probably indigenous.

2. FRAGARIA CHILENSIS, Hhrh.

Has. Sandwich Islands: Hawaii, in moist places, on the edges of
the crater Lua Pele, &.; on Mouna Kea; and in the district of
Waimea. (Specimens were gathered in the bullock-pit in which

the unfortunate Douglas lost his life.) Also on the mountains of
Maui.

This species was not gathered in Chili, where it appears to be spon-
taneous only along the coast. It is probably truly indigenous to
—Chiloe; as it is on the coast of California and Oregon, and apparently
in the Sandwich Islands. The leaflets in the specimens from the
Sandwich Islands are thick and rigid, with their lower surface and
the stalks very silky-villous.

ROSACE &. 501

10. GEUM, Linn.

1. Geum MaGELLANICcUM, Commers.

Geum Magellanicum, Commerson, ex Pers. Ench. 2, p. 57; Hook. f. Fl. Antare.

p. 262, & Fl. N. Zeal. p. 55.
G. ranunculoides, Seringe, in Mem. Soc. Genev., & Prodr. 2, p. 551.

Has. Bay of Islands, New Zealand.

To this species Dr. Hooker refers the G. ranunculoides, G. Chi-
loense or G. Chilense, G. Quillyon, and G. intermedium of authors, and
likewise our North American G. strictum, Ait.,—apparently with
reason: but in that case the plant should bear the latter name. He
also strongly inclines to unite the whole of them to the European G.

urbanum.

2. GEUM PARVIFLORUM, Commers.

Geum parviflorum, Commerson, ex Smith, in Rees, Cycl.; DC. Prodr. 2, p. 552;
Hook. f. Fl. Antare. p. 268, & Fl. N. Zeal. p. 56.

G. involucratum, Juss. ex Pers. Ench. 2, p. 57; DC. l.c. p. 504.

Sieversia albiflora, Hook. f. Fl. Antare. p. 1, t. 7?

Has. Orange Harbour, Fuegia.

11. RUBUS, Linn.
1. Rusus ros#FoLius, Sinith.
Rubus rosefolius, Smith, Ic. Pl. Ined. p. 60, t. 60; Seringe, in DO. Prodr. 2, p. 556.
Has. Luzon, near Baiios. Obrajillo, Peru. The latter doubtless

a cultivated plant, as the petals are multiplied.
126

502 PHANEROGAMIA.

2. Rusus EGLanteriA, Tratt.
Rubus Eglanteria, Trattinick, Ros. 3, p. 9; Seringe, in DC. Prodr. 2, p. 556.

Has. New South Wales, near Sydney.

3. Rusus prnnatus, Willd.

Has. St. Helena: probably an introduced plant.

4. RUBUS AUSTRALIS, Forst.

Rubus australis, Forst. Prodr. Fl. Ins. Austr. p. 40; DC. Prodr. 2, p. 559; Hook. f.
Fl. N. Zeal. p. 53, t. 14.
R. schmidelioides & R. cissoides, A. Cunn. Bot. Fl. N. Zeal., ex Hook. f.

Has. Bay of Islands, New Zealand: various forms, either glabrous
or pubescent; the leaflets ovate, lanceolate, or linear. Sydney and
Hunter's River, New South Wales.

d. Rupus mAcropopus, Seringe.

Rubus macropodus, Seringe, in DC. Prodr. 2, p. do7.

Has. Sydney and Hunter's River, New South Wales. Also “ Bay
of Islands, New Zealand ;” but I suspect there is some misplacement
in the tickets here, since no other collector has ever brought it from

New Zealand, nor does Dr. Hooker mention it in his Flora of that
country.

6. Rubus Frruticosus, Linn.

Has. Madeira: common.

ROSACES. 503

7. Rusus Bereu, Cham. & Schlecht. ?

Rubus Bergit, Cham. & Schlecht. in Linnea, 2, p.16; Eckl. & Zeyh. Enum. p. 262.
R. heterophyllus, H. Meyer, non Willd.

Has. Cape of Good Hope.

8. RuBus URTICAFOLIUS, Por.
Rubus urticefolius, Poir. Dict. 6, p. 246; Seringe, in DC. Prodr. 2, p. 563.

Has. Peru, below Obrajillo. Brazil, from Rio Janeiro to the Organ
Mountains.

9. Rusus GEOIDES, Smith.

Rubus geoides, Smith, Ic. Pl. Ined. t. 19; Hook. Ic. Pl. t. 495; Hook. f. Fl. Antare.
p- 268.

Dalibarda geoides, Pers.; DC. Prodr. 2, p. 568.

Has. Orange Harbour. (With mature fruit.)

10. Rusus rucosus, Smith.

Rubus rugosus, Smith, in Rees, Cycl.; Wight & Arn. Prodr. Fl. Ind. Or. 1, p. 290.

Has. Singapore. Luzon. New South Wales.

To this species is referred the R. alcecefolius, R. reflexus, and R.
Hamiltonianus of authors, as well as &. Molluccanus of Roxburgh.
Probably it is also the Linnzean plant of the latter name.

11. Rusus tittaceus, Smith, 0. c.

Has. Feejee Islands; at Ovolau and Rewa: “a climbing plant.”

504 PHANEROGAMIA.

12. Rusus Hawaiensis, Sp. Nov. (Tab. 56.)

R. tomentoso-pubescens ; caule fruticoso erecto setoso-aculeolato seu inermt ;
stipulis filiformibus ; foliis trifoliolatis; foliolis ovatis scepius acumt-
natis pinnatifido-ncisis vel duplicato-dentatis (dentibus acuminatis)
membranaceis supra glabratis, terminali longe petiolulato; pedunculo
unt—trifloro ; calyce tubo setoso, lobis ovatis integerrimis longe acumi-
natis petala ovata rubra cequantibus; fructu ovoideo; ovariis parce
glanduloso-hispidulis demum glabratis.

Var. 3. foliis ramulisque glabratis vel glabellis.

Has. Sandwich Islands: in the forests and the bullock plains on
Mouna Loa and Mouna Kea, Hawaii (where it was discovered by
Menzies). Var. @. District of Waimea and Hilo, Hawaii; and in
the mountains of Kauai. (Also collected, in fruit, by Gaudichaud.)

A shrubby species, the erect stems apparently of considerable height:
a short truncheon of a stem in the collection is almost two inches in
diameter! The younger branches, leaves, &c., are tomentose-pubescent,
apparently not glandular nor viscous, flexuose, unarmed or frequently
aculeolate with small and straight bristly prickles, at length glabrate.
Stipules filiform or setaceous, 3 or 4 lines long. Leaves trifoliolate.
Leaflets ovate, acuminate or acute, with an obtuse or truncate base,
pinnatifid-incised or very deeply doubly toothed (the close teeth sharply
pointed), membranaceous, strongly pinnately veined, the lower surface
softly and usually densely tomentose, the upper glabrate with age; the
lateral leaflets slightly petiolulate, 13 to 2 inches long; the terminal
one long-petiolulate (the stalk half or two-thirds of an inch long), 24
to 3 inches in length, occasionally subcordate. Peduncles 1-3-flowered,
axillary and terminal, solitary or in pairs, short; when branched the
pedicels an inch or less in length, tomentose, unarmed. Calyx
flattish, more or less tomentose externally; the short tube sparingly
setose-prickly ; the lobes ovate, entire, 6 or 8 lines long, tapering into a
slender and long acumination, as long as the broadly ovate petals,
which are of a red or bright purple colour, and apparently as showy
as those of R. odoratus. Stamens and pistils very numerous. Recep-
tacle rather hairy. Ovaries sparsely hispid with a few glandular-

ROSACES. 505

tipped bristles: but the drupaceous mature carpels glabrous, tipped
with slender styles. Fruit ovoid, half an inch in length and breadth,
apparently red and edible: it is plainly that of a Raspberry, the
carpels separable from the conical receptacle.

The variety, above indicated, appears to differ only in the less
pubescent shoots, leafstalks, &c., and in the soon glabrate leaflets, &c.
Perhaps it is a state of the species growing in deep shade.

This striking species is most related to the FR. spectubilis, Pursh, of
Northwest America, which the smoother form considerably resembles.
The long-acuminate sepals, equalling the broad petals in length, and
the greater prickliness, as well as the usually downy leaves, Xc., serve
to distinguish them. The wood of the stem is compact, with broad
and conspicuous medullary rays, but no obvious annular layers. It is
singular that so conspicuous a plant should have hitherto escaped
notice. I believe it was gathered, however, by the late Mr. Macrae.

Pirate 56.—Rusus Hawatensis: a flowering branch and fruit, of
the natural size. Fig. 1. Section of a flower. 2. A pistil. 3. Ovary,
vertically divided. 4. Putamen. 5. Vertical section of a carpel, in
fruit. 6. Vertical section of a seed. 7. Embryo.—The details vari-
ously enlarged.

13. Rusus Macrar, Sp. Nov. (Tab. 57.)

R. tomentoso-pubescens ; caulibus procumbentibus setoso-aculeolatis ; sti-
pulis subulatis ; folirs caulinis trifoliolatis, ramulorum floriferorum
trilobatis ; foliolis rotundis ovatisve duplicato-dentatis (dentibus mucro-
natis) nunc incisis subcoriaceis subtus molliter canescenti-tomentosis
supra glabratis, terminali scepissime petiolulato ; pedunculis unifloris ;
calyce inermi quinquepartito, lobis oblongis laciniato-dentatis petala
(pallida) obovatu scepius emarginata adcequantibus ; ovariis superne
hirsutis eglandulosis.

Has. Sandwich Islands: Hawaii, on Mouna Kea (where it was
first collected by Mr. Macrae): Kast Maui, on the banks of the crater
Haleakala. |

A well-marked, tomentose-pubescent species, with reclined or prostrate
127

506 PHANEROGAMIA.

stems and ascending flowering branches, armed with weak and slender
prickles, apparently suffruticose. Stipules subulate, rather large,
sometimes almost setaceous. Cauline leaves trifoliolate, or those of
the short flower-bearing branches simple and three-lobed. Leaflets
rotund or ovate, either obtuse or acute, 14 to 2 inches long, thickly
doubly toothed (the teeth mucronate and short), often incised or obscurely
lobed, rather coriaceous in texture, canescently tomentose-pubescent under-
neath, glabrate above; the lateral ones slightly, the terminal manifestly,
petiolulate, or sometimes almost sessile. Peduncles one-flowered, usually
axillary, setose-prickly, apparently recurved after anthesis, an inch to
2 inches long, the lower ones shorter than the leaves. Flowers large,
fully an inch in diameter. Calyx five-parted almost to the base,
tomentose-pubescent, not armed nor glandular; the sepals oblong,
acute or sometimes obtuse, or the innermost pointed, rather foliaceous,
becoming three-quarters of an inch long after flowering, the edges
strongly serrate or laciniate-toothed from the middle to the apex.
Petals obovate, often deeply notched at the apex, apparently white (the
colour not recorded by the collectors), as long as the calyx. Stamens
and pistils numerous. Ovaries hirsute, especially towards the summit,
with soft and glandless hairs, which persist on the ovoid unripe fruit.

This species is remarkable for the serrated sepals; a character
which, although variable in degree, is presented by all the specimens
seen. ‘he petals are all imperfect in our flowering specimens. They
are restored in the plate from a flower of a plant which was gathered
long ago by Mr. Macrae. The flowering branch represented in the
left-hand figure is also from one of his specimens.

Piate 57.—Ruzsus Macrai: a procumbent stem and flowering

branches, of the natural size. Fig. 1. A petal. 2. Vertical section
through the receptacle, &. 3. A pistil—The details magnified.

12, CHAMAMELES, Lindl.

1. CHAMMMELES CoRIACEA, Lind.

Chameemeles coriacea, Lindl. in Trans. Linn. Soc. 18, p. 104, t. 11; DC. Prodr.

ROSACES. 507

Has. Madeira; on rocks of the coast at Funchal. (Without
flowers or fruit.)

13. HESPEROMELES, Lindi.
1. HESPEROMELES CUNEATA, Lind.
Hesperomeles cuneata, Lind]. Bot. Reg. sub. t. 1956, in adnot.

Has. High Andes of Peru; common from Obrajillo to Culnai.

14. OSTEOMELES, Lind.
1. OSTEOMELES ANTHYLLIDIFOLIA, Lindl.
Osteomeles anthyllidifolia, Lindl. in Trans. Linn. Soc. 13, p. 98, t. 8.
Has. Sandwich Islands: Oahu, on the plains near Ewa, and on
the coast towards Pearl River: Hawaii, in the district of Waimea,

and on Mouna Loa. (Found by all collectors since the islands were
first botanized upon by Nelson, in Cook’s last voyage.)

The fruit is said to be “ white and sweet-tasted.”

Orv MYRTACE &.

1 BARRINGTONIA, Forst.

1. BARRINGTONIA SPECIOSA, Linn. f.

Barringtonia speciosa, Linn. f. Suppl. p. 312; Forst. Prodr. p. 47; Guill. Zeph.
Tait. p. 59, cum descr. Forst.; Wight, Ic. Pl. Ind. Or. 2, t. 547.
B. Butonica, Forst. Char. Gen. t. 38.

Has. Society and Samoan Islands; common on the coast. Feejee
Islands; “submaritime and not very common.” Mangsi Islands.

2. BARRINGTONIA EXCELSA, Blume ?

Barringtonia excelsa, Blume, Bijdr. p. 1097? Benth. in Hook. Lond. Jour. Bot. 2,

p. 221.
B. racemosa, Gaud. Bot. Voy. Freye. p. 483, t. 107?

Has. Ovolau, and Sandalwood Bay, Feejee Islands. (In flower,
and with young fruit.)

This accords very well with the figure of Gaudichaud’s B. racemosa :
whether it is the plant of Blume under that name, or truly his B.
excelsa, | cannot determine.

3. BARRINGTONIA SAMOENSIS, Sp. Nov.

B. folvis membranacers oblongo-cuneatis oblanceolatisve acuminatis obsolete
serrulatis im petiolum brevem attenuatis; racemo gracili elongato

MYRTACES&. 509

nutante; floribus parvis; pedicellis alabastro duplo triplove longio-
ribus ; calyce quadrilobo.

Var. 3. foliis majoribus dilatatis obovato-cuneatis ; pedicellis alabastro
haud longioribus.

Has. Upolu and Tutuila, Samoan or Navigators’ Islands. 8. Upolu.

A tree, the size not recorded, glabrous throughout. Leaves mem-
branaceous, from 7 to 10, or in var. 8. from 10 to 18 inches in length,
oblong with a cuneate base, or oblanceolate, or in var. @. obovate-cuneate,
obsoletely serrulate, more or less acuminate, tapering into a short petiole.
Raceme terminal and simple, a foot or two in length, slender, nodding,
rather closely-flowered. SF lower-buds 3 lines in diameter, shorter by
twice or thrice than the pedicel: but in var. 3. the pedicels are only 3
or 4 lines long. Calyx four-cleft, sometimes two of the sinuses deeper
than the others. Petals 5 or perhaps 6 lines long. “Stamens red,”
probably an inch long when expanded. Fruit not seen.

The species of Barringtonia are by no means well settled; but this
is different from any one I have seen.

2. EUGENIA, WMich., Linn.

Kueentia, Mich., Linn., Juss., St. Hil., Wight.

EvucEnta (excl. sp. Chilens. & EH. Pimenta), JamBosa, AomenA, & Syzyarum, DC.

EuGenta, GELPKEA, STRONGYLOCALYX, JAMBosA, CLAVIMyRTUS, & MicRoJAM-
BosA, Blume.

Sytiysium, Meyen & Schauer.

Although strongly predisposed to receive the Candollean genera
above cited, yet even the comparatively small amount of materials
before me furnishes such complete gradations between them all, as to
leave no doubt that Dr. Wight was justified in combining them.
Neither the shape of the tube or of the limb of the calyx, nor the
difference between an operculately deciduous and an expanding
corolla, can be relied upon for definite characters. The recent, very

careful labours of Blume (in Mus. Bot. Lugd.-Bat.) evidently lead to
128

510 PHANEROGAMIA.

the same conclusion, unless we still more largely increase the number
of genera, and abandon all absolute distinctions. A series of such
groups may be more appropriately regarded as sections of a genus.
The old genus Hugenia may thus be more simply and definitely
characterized by its thick and large, more or less conferruminate
cotyledons, including the usually short and retracted radicle; the two
or three pluriovulate cells to the ovary, but the fruit only 1—2-seeded ;
the stamens not collected in phalanges. The last-named character
barely distinguishes Caryophyllus. As Jossinia is characterized by
Blume, it would also fall into Eugenia, except for the thick testa of
the seed; but according to Wight it has foliaceous cotyledons. The
seeds of the original species, however, do not yet appear to be known.

§ 1. JAMBOSA, Rumph.— Calyx fauce ultra ovarium producta, lobis rotundatis eesti-
vatione manifeste imbricatis. Petala in anthesin expansa.—F lores seepissime magni
vel majuscult.

1. Eucenta (JAmBosaA) Maraccensis, Linn.

Eugenia Malaccensis, Linn. Spec. Pl. p. 672; Wight, Ill. Ind. Bot. 2, t. 98.
Jambosa domestica, Rumph. Herb. Amb. 1, p. 127, t. 37; Blume, Mus. Bot. p. 91.
J. Malaccensis, DC. Prodr. 3, p. 287, et Auct.

Has. Feejee, Samoan, Society, and Sandwich Islands: introduced.
“Cultivated and intrusive.” Pickering.

The imperfect specimen from Tahiti, like that gathered by M.
Meerenhout and mentioned by Guillemin (in Zeph. Tait. p. 59),
would seem to belong to a different species, having oblong-lanceolate,
acuminate, almost membranaceous, and pellucid-punctate leaves. But
Dr. Pickering, who was familiarly acquainted with the Red Rose-
Apple in different habitats, appears not to have distinguished it.

2. EvcEnta (JamBosa) Rica, Sp. Nov. (Tab. 58.)
E. ramulis ultimis quadrangulatis scepius argute marginatis; foliis

brevi-petiolatis ovatis oblongisve obtusis basi subcordatis coriaceis
venosis (venis infra marginem laxe arcuato-anastomosantibus) impel-

MYRTACE &. 511

lucidis opacis ; cyma pluriflora ; pedunculis partialibus subgracilibus ;
calycis tubo turbinato, lobis 4 subequalibus.— Variat: foliis maximis
ovatis fere pedalibus, seu spithameis, seu anguste oblongis nune aci-
tiusculis 5—9-pollicaribus.

Has. Feejee Islands: at Ovolau, Rewa, Somu-somu, Muthuata, &c.

A “tree 20 feet high,” glabrous, apparently varying considerably in
the size and shape of the leaves. Ultimate branches more or less four-
angled and sharply margined by 2 salient lines decurrent from the
base of each petiole: these are sometimes so prominent as to form
wings, while on other branches of the same individual they are obsv-
lete or wanting. Leaves coriaceous in texture, and dull, ovate, or more
commonly oblong, varying from 5 inches to nearly a foot in length,
and from 2 to 6 inches in width, obtuse, cordate at the base by a small
sinus, rather sparsely feather-veined and sparingly reticulated, the
veins rather obscurely anastomosing within the margin, opaque, only the
thinner states showing pellucid dots: the petiole 3 to 6 lines long.
Cymes several—many-flowered, terminal, or sometimes axillary; the
common peduncle from one to two inches long, articulated, trichoto-
mous, with the rather slender partial peduncles usually again divided,
about an inch long; the ultimate divisions terminated either by soli-
tary, or by three sessile, articulated flowers. Tube of the calyx tur-
binate, half an inch or less in length; the short lobes 4, nearly equal,
rounded, apparently not enlarged in fruit. Petals probably rose-colour.
Stamens and style from an inch to an inch and a half long. Ovary
two-celled. Ovules numerous. Fruit not collected.

Puate 58.—Evegenta (Jamposa) Ricui: a flowering branch, of the
natural size. Fig. 1. Vertical section of a flower, enlarged. 2.

Transverse section of the ovary.

3. EvGenra (JAMBOSA) QUADRANGULATA, Sp. Nov.

E. ramulis ultimis tetragonis, angulis argute marginatis alatisve ; Foliis
brevissime ~petiolatis oblongo-lanceolatis acutis vel acuminatis basi
obtusis supra lucidis chartaceis pellucido-punctatis crebre penninervues ;

512 PHANEROGAMIA.

cyma multiflora; pedunculi communi partialibusque brevissimis,
Sructiferis incrassatis ; calycis tubo turbinato, limbo quadrifido ; fructu
obovoideo-urceolato.

Has. Ovolau, Feejee Islands; common in woods.

This is said to be a “shrub, with a weak stem, 10 to 15 feet high;
the young stems square.” The ultimate branchlets are sharply four-
angled, the angles strongly margined, or often winged. Leaves charta-
ceous in texture, rather lucid above, glabrous, as is the whole plant,
oblong-lanceolate, with a more or less pointed apex and an obtuse base,
on a very short petiole (of about 2 lines in length): they are 7 to 9
lines long, 2 or 3 wide, pellucid-punctate by transmitted light, rather
closely feather-veined: the veins running into a submarginal vein.
There are only fruiting specimens in the collection, or enlarged
ovaries from which the petals, stamens, and style have entirely fallen.
The cyme is terminal and solitary, many-flowered and apparently
dense, very short-peduncled, its primary, secondary, and_ tertiary
branches all very short (2 or 3 lines in length) and much thickened.
The immature fructiferous calyx is turbinate and somewhat clavate,
8 or 9 lines long, terete, smooth and even, with a short four-cleft
limb; the lobes coriaceous, rounded. Ovary two-celled. Ripe fruit
obsvoid and urn-shaped, an inch in length, “red,” crowned with the
erect calyx-lobes, one-celled, one-seeded. Cotyledons thickened, cor-
neous, united.

This should be compared with Jambosa pterocaulis of Korthals,
which I know only by the brief phrase copied by Walpers, where the
leaves are said to be cordate at the base.

4, KuGEnra (JAMBOSA) NEUROCALYX, Sp. Nov. (Tab. 59.)

E. ramulis teretibus; foliis brevissime petiolatis elongato-oblongis seu
elliptico-lanceolatis obtusissimis basi subcordatis chartaceis crebre penni-
nerviis; floribus in capitulum sessile congestis ; calycis tubo cylin-
draceo pluricostato, linbo ampliato fructifero crateriformi quadrilobo,

Has. Feejee Islands, at Muthuata and Ovolau: «“ planted near
houses.”

MYRTACE A. 513

The specimens of this remarkable species are past flowering and
with immature fruit. Branches terete, glabrous, like the whole plant.
Leaves chartaceous in texture, some of them obscurely pellucid-punc-
tate, rather lucid above, elongated-oblong or elliptical-lanceolate, 6 to 9
inches long, about 2 inches wide, very obtuse, subcordate at the base, on
a very short petiole (of only a line and a half in length), closely
feather-veined ; the primary veins confluent into a lateral vein within
the margins. lowers 6 to 10, or perhaps more, large, sessile, aggre-
gated into a terminal, or sometimes axillary, sessile capitulum. Calyx-
tube after anthesis 6 or 7 lines long, or later 8 or 9 lines long, cylin-
draceous, obtuse at the base, ribbed with 10 or 12 salient nerves, above
dilated into an expanded crateriform limb, which becomes an inch in
diameter, is lined with a broad and depressed disk, and cleft into 4
spreading persistent lobes. Petals and stamens not seen. The per-
sistent style 2 to 22 inches long. Stigma minute. Ovules numerous
in each cell.

PLATE 09.—EHUGENIA (JAMBOSA) NEUROCALYX : branches, past flower-
ing, of the natural size. Fig. 1. Vertical section of the calyx and
ovary. 2. Transverse section of the ovary, enlarged.

d. HUGENIA (JAMBOSA) GRACILIPES, Sp. Nov.

£. ramulis gracilibus teretibus; foliis subsessilibus lanceolato-oblongis
subacuminatis basi parum cordatis chartaceis pellucido-punctatis
penninerviis; pedunculis terminalibus filiformibus folia superantibus
uni—trifloris ; calycis tubo turbinato basi acuto, lobis 4 cequalibus.

Has. Feejee Islands, at Sandalwood Bay and Ovolau.

Apparently a shrub, with slender, terete branches, glabrous. Leaves
chartaceous or nearly membranaceous in texture, opposite, danceolate-
oblong, somewhat acuminate, slightly cordate at the base, subsessile (the
petiole when manifest only a line long), 24 to 4% inches long, 12 to
18 lines wide, somewhat lucid above, more or less manifestly pellucid-
punctate, closely feather-veined, the veims confluent into an indistinct
infra-marginal vein. Peduncles terminal, filiform, exceeding the leaves

(2 or 3 inches long), one-flowered, or commonly di-trichotomous, with
129

514 - PHANEROGAMIA.

the pedicels an inch or two in length, minutely bracteolate at the
articulation with the acute base of the turbinate calyx-tube, which is
only 5 lines long after anthesis. Calyx-lobes 4, equal, rounded,
barely 2 lines long. Stamens 8 lines long. Style after anthesis an
inch and a half long. Ovary two-celled; the cells many-ovuled.
Fruit unknown.

This is apparently related to Jambosa hypericifolia, DC., from Java;
which, however, is said to have large flowers, on peduncles only an
inch long and shorter than the leaves; the latter 4 to 7 inches long,
and not dotted.

6. EKuGENIA (JAMBOSA) AUSTRALIS, Wendl.

Eugenia australis, Wendl.; Link, Enum. Hort. Berol. 2, p. 28; Colla, Hort. Ripul.
App. 1, t. 8.

EL. myrtifolia, Sims, Bot. Mag. t. 2230; Lodd. Bot. Cab. t. 525.

Jambosa australis, DC. Prodr. 3, p. 286.

Has. New South Wales.

This is mentioned by Blume as apparently making a transition
from Jambosa to his Clavimyrtus.

§ 2. EUGENTA, DC.—Calyx tubo subrotundo, Jfauce ultra ovarium vix aut ne vix
producta, limbo 4-5-partito, lobis sepius leviter imbricatis. Petala in anthesin
expansa. Testa seminum tenuis.—Flores parvuli.

* Oceanice.

7. EUGENIA RARIFLORA, Benth. (Tab. 60.)

E, foliis coriaceis ovalibus late ovatisque breviter petiolatis glabris sub-
venosis supra nitidis marginibus subrevolutis 3 pedicellis unifloris
solitarius vel geminis (nune quaternis quimsve ad apicem ramorum)
ramulisque appresse puberulis folio brevioribus 3 flore bibracteolato ;
calycis tubo globuloso-obconico sericeo-pubente limbo wmequaliter quadri-
partito subequilongo, lobis rotundatis petalis brevioribus; bacca sub-

MYRTACES, 515

globosa.— Variat: foliis crasso-coriaceis basi rotundatis (ut in spec.
Benth. descr.), seu ovali-orbiculatis basi acutis apice obtusissimis
sepius emarginatis, sew tenuoribus ovalibus oblongisve nune obtuse
acuminatis.

Eugenia rariflora, Benth. in Hook. Lond. Jour. Bot. 2, p. 221.
Jossinia cotinifolia, Hook. & Arn. Bot. Beech. Voy. p. 62; Guill. Zeph. Tait. p. 58,
excl. syn.

HAs. Feejee Islands (where it was gathered by Hinds and Barclay) :
common on the coasts. Upolu, Samoan Islands. Tahiti, Society:
Islands. |

Shrub 4 to 6 feet high; the young branchlets and the nascent
foliage puberulent with an appressed and fine silky pubescence, gla-
brate when old. Leaves coriaceous, sometimes thick, oftener rather
thin, glabrous, or occasionally with a fine pubescence on the midrib
beneath, shining above, paler and dull underneath, oval or broadly
ovate, sometimes verging either to obovate or to oblong, either rounded
or acute at the base, obtuse, obtusely somewhat pointed or sometimes
notched at the apex, rather veiny, the thinner states pellucid-punctate,
the thicker with more or less revolute margins, 2 or 3 inches long,
and from one to 24 inches wide; the petiole 2 or 3 lines in length.
Pedicels one-flowered, solitary or in pairs in the axils of the leaves, or
sometimes 4 or 5 together at the summit of the branches, slender, 6 to 14
lines long, puberulent, bibracteolate at the apex, and these articulated
with the flower. Bractlets subulate, small. Culyax-tube between glo-
bose-obovoid and obconical, silky-pubescent when young; the limb four-
parted almost to the ovary; the lobes fully as long as the tube, rotund,
strongly imbricated in estivation (two exterior and two interior)
unequal, becoming oval or oblong with age. Petals white, oval or
oblong, 4 lines long, somewhat exceeding the lobes of the calyx,
expanding, deciduous. Stamens very numerous, as long as the petals,
uniformly inserted over a broad and flat puberulent disk, free. Style
single: stigma entire. Ovary two-celled. Ovules numerous, crowded
on axile placents, angled by mutual pressure, amphitropous. Berry
subglobose, glabrous, apparently rather dry, crowned by the persistent
lobes of the calyx, from half an inch to three-fourths of an inch in
diameter, usually two-celled and two-seeded, when it is often some-
what didymous in the dried state and broader than long, or by abor-

516 PHANEROGAMIA.

tion one-celled and one-seeded. Seed globose, large, filling the cell,
with a very thin and membranous testa. Cotyledons very thick and
fleshy, conferruminate; the very short radicle hidden.

This appears to be a pretty widely diffused, and somewhat variable,
Oceanic species, which may have been formerly described under some
other name. It has been taken for a Jossinia by Hooker and Arnott,
&c., and it accords with Blume’s character of that genus, except that
the seed has a very thin testa. If the fruit of the genuine Jossinve
of Mauritius, &c., be as it is characterized by Blume (in Mus. Bot.
Lugd.-Bat. p. 119), that genus is indeed near to Hugenia, and not to
be definitely distinguished by either of the characters relied upon.
For the lobes of the calyx are imbricated in estivation in various
degrees in numerous American Hugenie, and in some of them the
stamens are inserted over a broad disk.—The plate, representing (not
particularly well) only one form of this species (that with thick and
rounded or emarginate leaves), was not made under my superinten-
dence. I have merely added some sections of the fruit.

Puate 60, A—EvuGENIA RARIFLORA (mal.). Fig. 1. A flower-bud,
magnified. 2. The same, with part of the calyx and disk cut away,
and the petals with most of the stamens removed. 3. The ovary,

transversely divided, magnified. 4. Fruit, of the natural size. 5. A
two-celled, and 6, a one-celled fruit and seed, transversely divided.

x * Brasilienses. (HUGENIZ VER. )

8. EUGENIA UNIFLORA, Linn.

Eugenia uniflora, Linn. Spec. Pl. ed. 1, p. 470; St. Hil. Fl. Bras. Mer. 2, p. 348.
EL. Michelii, Lam, Dict. 3, p. 203; DC. Prodr. 3, p. 263; Blume, Mus. Bot. p. 86.

Has. Rio Janeiro, Brazil; common along the shore of the bay.

9. EUGENIA NEMORALIS, DC. ? 1. ¢.

Has. Organ Mountains, Brazil.

MYRTACE #&. 517

10. EuGEentA BrAsILiensis, Lam.

Eugenia Brasiliensis, Lam. 1. c.; Cambess. in St. Hil. Fl. Bras. 2, p. 354, t. 152;
Hook. Bot. Mag. t. 4526.

Has. Rio Janeiro: along the shore.

11. EUGENIA ROTUNDIFOLIA, Casaretto ?

Eugenia rotundifolia, Casaretto, Stirp. Nov. Bras. Dec. 4, p. 40?

Has. Rio Janeiro; with the preceding. (In fruit only.)

12.. EUGENIA XANTHOXYLOIDES, Camb.
Eugenia xanthoxyloides, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 866, ex char.

Has. Near Rio Janeiro. (In young fruit.)

13. EUGENIA TENELLA, DC.

Has. Organ Mountains, Brazil, in the vicinity of Rio Janeiro.

14. EuGEntA SALZMANNI, Benth.

Eugenia Salzmanni, Benth. in Hook. Jour. Bot. 2, p. 319.

Has. Organ Mountains, Brazil.

15. EvuGENIA ARENARIA, Camb.

Eugenia arenaria, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 349?

Has. Rio Janeiro, Brazil.
180

518 PHANEROGAMIA.

There are besides incomplete and indeterminable specimens of two
or three more Eugenie in the Brazilian collection.

§ 3. SYZYGIUM, Gertn., excl. spec. (Syzygium & Acmena, DC., Endl. Syllysium,
Meyen & Schauer.) —Calyx tubo turbinato obovato vel clavato, Sauce ultra ovarium
modice producta, margine repando-truncato vel subquadrilobo. Petala in calyptram
pl. m. coalita, sub anthest dejecta, rariusve expansa sed caduca, interdum abortiva.
Seminis testa tenuis.

Only a slight artificial character separates Acmena of DeCandolle
from Syzygium, and that an inconstant one in certain species, which
cast off their petals indifferently either before or at the time of expan-
sion. If Acmena be retained as a subgenus, therefore, it were doubt-
less better to refer to it, as Wight has done, species which throw
off their corolla unexpanded, but which have the elongated calyx-
tube. Acmena would thus constitute a group between Clavimyrtus
and Syzygium, but passing by continuous gradations into the latter.

* Novo-Hollandice ; petalis parvis in anthesin expansis. (AcMENA, DC.)

16. Eveenta (AcMENA) Suita, Powr.

Eugenia Smithit, Poir. Dict. Suppl. 8, p. 127.

E, elliptica, Smith, in Trans. Linn. Soc. 3, p. 281, non Lam.
Metrosideros floribunda, Smith, 1. c. p. 267; Vent. Hort. Malm. t. 75.
Acmena floribunda, DC. Prodr. 3, p. 262.

Has. Sydney, New South Wales. (In flower.)

* * Novo-Zelandice ; floribus forte polygamis, petalis nonnunguam expansis.
17. Eucrenta (Syzycium) Maire, A. Cunn.

Eugenia Maire, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 8, p. 115; Hook. f. FI.
N. Zeal. p. 71.

Has. Bay of Islands, New Zealand.

The specimens, in flower only, appear to have sterile ovaries,

MYRTACES. 519

although they contain a few ovules. The calyx is nearly as in the
foregoing: its margin is obscurely five-toothed. The petals some-
times expand and fall separately, but more commonly are thrown off
together in anthesis, although slightly if at all coalescent.

x * * Sandwicenses; an petalis expansis ?
18. Eugenta (Syzyeium) SANDWICENSIS, Sp. Nov.

E. ramulis ultimis tetragonis, angulis argute marginatis ; foliis obovatis
oblongisve basi cuneatis apice rotundatis vel obtusis pellucido-punctatis
subtus pallidioribus venis tenuibus crebre penninerviis reticulatis ;
pedunculis axillaribus folio brevioribus ; cyma trichotoma plurifiora ;
pedicellis brevissimis; calyce turbinato, limbo brevissimo quadrilobo
staminibus pauciusculis via breviore ; bacca globosa.

Var. 2. foltis minoribus nunc obovato-ellipticis aut ovalibus.

Has. Oahu, Sandwich Islands; on the mountains behind Hono-
lulu: both forms. (Also gathered by Gaudichaud, Macrae, &c.)
Island of Kauai: foliage only.

The typical form (of which alone the fruit was collected) is said to
be a tree, of 20 feet in height; while what I take for merely a small-
leaved variety of the same is mentioned, if I rightly identify it, as a
low shrub. Branches, as well as the whole plant, glabrous; the
leafy branchlets four-sided, with sharply margined angles. Leaves
coriaceous, not very thick, obovate or oblong, cuneate at the base, 3 to
4% inches long, or in the variety inclining to obovate-elliptical or oval,
and only 14 to 2 inches long, all very obtuse or rounded at the summit,
green above, somewhat pale underneath, pellucid-pwnctate, coprously
feather-veined with fine veins, connected by reticulated veinlets on the
under surface, and within the margins confluent into the usual false
vein. Petiole 3 or 4 lines long. Peduncles axillary, solitary, rather
slender, angled, shorter than the leaf, bearing a small, once or twice
trichotomous cyme, of several or numerous flowers, on very short
pedicels. Flower-buds a line and a half long, subclavate or twrbinate ;
the limb of the calyx very short, but four-lobed, the sinuses extending,

520 PHANEROGAMIA.

however, nearly to the edge of the subcampanulate disk. Petals 4,
not over a line long, probably expanding in anthesis, caducous.
Stamens rather few, apparently about 20, little longer than the lobes
of the calyx, less than a line long, inserted on the edge of the disk:
filaments glandular-dotted. Style shorter than the stamens. Ovary
two-celled, with 10 or more ovules in each cell. Berry globular, “as
large as a cherry,” containing one or two seeds, which, as likewise
the embryo, accord with those of Eugenia.

This is one of the connecting forms between Eugenia, Acmena and
Syzygium; with the habit rather of the latter.

The collection from Oahu contains the foliage only of another
Myrtaceous plant, probably a Hugenia.

x x * * Malaysiane.
19. Evucenta (Syzyeium?) ZEYLANICA.
Myrtus Zeylanica, Linn. Spec. Pl. p. 472 (Herm. Hort. Lugd. t. 435).
Syzygium Zeylanicum, DO. Prodr. 3, p. 260; Wight & Arn. Prodr. Ind. p. 330;

Wight, Ic. Pl. Ind. Or. 1, t. 73.
Acmena? parviflora, DC. Prodr. 38, p. 462.

Has. Singapore.

The limb of the calyx is “evidently five-lobed,” as DeCandolle
remarks; but the corolla appears to fall off as a lid. It would
perhaps fall into Blume’s genus Microjambosa.

20. Eugenia (Syzycrum) BenTHamt.

Syzygium nitidum, Benth. in Hook. Lond. Jour. Bot. 2, p. 221.

Has. Mangsi Islands, in the Sooloo Archipelago. (Also? Singa-
pore: foliage only.)

There are 8 or 10 ovules in each cell of the ovary. In Mr. Ben-
tham’s plant, the cells are said to contain only two or three ovules,

MYRTACE A. 521

* * * x * Polynesice.
} Flores parvi vel parvuli.
21. Eugenta (SyzYGIuM) INOPHYLLOIDES, Sp. Nov.

E. foliis ovalibus oblongisve basi acutis apice subito in acumen obtusum
productis coriaceis supra nitidis venis subtransversis tenuibus creber-
rime lineato-penninervits reticulatis in venam margint parallelam
confluentibus ; cymis parvis terminalibus foliis breviortbus conferti-
floris; calycis margine repando-quadridentato.

Has. Tutuila, one of the Samoan or Navigators’ Islands; common
on a wooded ridge, at the elevation of 500 feet.

“A small tree,” in foliage, &c., considerably resembling EF. (Syzy-
gium, DC.) inophylla, Roxb. Branchlets slender, slightly angled, gla-
brous, as is the whole plant. Leaves oval, oblong or elliptical, 2 or 3
inches long, 15 to 18 lines wide, very abruptly acuminate into a
narrow but obtuse apex of 2 or 3 lines in length, acute or abruptly
contracted at the base into a petiole of only 2 or 3 lines long, coria-
ceous, shining on the upper surface, the lower dull, nervosely and
very closely feather-veined with fine and nearly transverse parallel veins,
which are likewise more or less reticulated in narrow areole, and
just within the margin confluent into a rather conspicuous false
vein. Cymes terminal, corymbose, short-peduncled, scarcely well
developed in the specimens (which bear only flower-buds), but shorter
than the leaves, the divisions all very short; the flowers crowded,
short-pedicellate; barely a line and a half long. Calyx turbinate,
with a minutely but distinctly repand-four-toothed margin. The
inflorescence is too young to ascertain whether the corolla falls off
like a lid. Ovules several in each cell.

22. EuGENIA (SyzyGiuM) BRACKENRIDGEI, Sp. Nov. (Tab. 61.)

S. ramis validis; foltis oblongo-cuneatis vel obovatis obtusissimis in
petiolum attenwatis crasso-coriaceis supra lucidulis creberrime penni-
131

529 PHANEROGAMLIA.

nerviis, venis in venam intramarginalem decurrentibus, venulis reti-
culatis; cyma terminali sessili fastigiata, divisionibus repitito-tricho-
tomis brevibus incrassatis apice glomerulifloris; calycis margie
repando-quadrilobo.

Has. Feejee Islands: on Ovolau, at the elevation of 1,500 feet;
also lower.

A tree, the larger form 30 feet high, glabrous throughout; the
flowering branches stout, obscurely angled. Leaves varying in size in
different specimens, from 2 to 6 inches, or on sterile shoots even 9
inches in length, thick and coriaceous, somewhat shining above, dull
and rather paler underneath, punctate, obovate with a cuneate base,
oblong-cuneate, or oblong-oblanceolate, very obtuse or even retuse at the
apex, contracted at the base into a petiole half an inch long; the
midrib stout: from it proceed from 25 to 50 slender but rather con-
spicuous straight veins on each side, which are decurrent into a direct
intramarginal vein, and numerous interposed veinlets, which are
more or less reticulated. Cyme terminal and sessile, many-flowered,
dense, flat-topped, repeatedly trichotomous, its ramifications all thickened
and angled; the primary and secondary rays about half an inch long;
the others much shorter, usually very short, articulated, terminated
by several sessile flowers im a cluster. Bracts and bractlets very
short, thick, round-ovate, rather persistent. Unexpanded flower-buds
(which alone occur on the specimens) a quarter of an inch long,
fleshy, glabrous, turbinate. Limb of the calyx repandly jfour-lobed ;
the lobes very short and broad indeed, but distinct, rounded, two of
them exterior and two interior in exstivation. After anthesis the
short calyx-lobes probably become obsolete, as in S. nitidwm and other
species. Corolla forming a short umbonate operculum; the petals
separable, but evidently falling away in connexion without expand-
ing. Stamens very numerous, inflexed in estivation, in the manner
of the genus, into the deep throat of the calyx. Style subulate:
stigma obtuse. Ovary two-celled, with very thick walls. Ovules
numerous (12 or more) in each cell. Fruit unknown.

Puate 61, A—Eueenia (Syzycrum) BRACKENRIDGE: branches, of
the smaller and the largest-leaved forms, of the natural size. Fig. 1.
Flower-bud, with the operculate corolla detached. 2. Vertical section

MYRTACE &. 523

ad

of the same. 3. Transverse section of the ovary.—The details en-
larged.

23. EucEnia (SYZYGIUM) CONFERTIFLORA, Sp. Nov. (Tab. 61.)

E. foliis obovatis vel suboblongis retusis basi in petiolum brevem angus-
tatis coriaceis pallidis tenwiter penninerviis, venis vix reticulatis im
venam intramarginalem confluentibus ; cymis terminalibus et in axillis
supremis compositis fastigiatis densifloris folia via superantibus,
divisionibus primariis gracilibus via angulatis, ultimis brevibus apice
glomerulifloris; floribus parvis; ealycis margine quadrilobo.

Has. Ovolau, Feejee Islands.

Apparently a shrub or small tree, glabrous, with rather slender
and terete, gray branches; the new shoots slightly ferrugineous.
Leaves crowded, 14 to 24 inches long, obovate, varying to oblong, but
all narrowed at the base into a short petiole, obtuse or mostly retuse at
the apex, coriaceous but thin, dull and pale both sides, or the upper
at first somewhat lucid, scarcely punctate, rather closely feather-
veined, the veins parallel, almost transverse, slender, little reticulated,
confluent into a false vein within the margin. Cymes terminal and in
the uppermost axils, compound, fastigiate, dense, more or less peduncled,
little exceeding the leaves, its principal divisions slender, slightly angled,
not over half an inch long, minutely lepidote and rusty; the succes-
sive ones shorter; the ultimate ones with the subsessile and small flowers
densely clustered at their apex. Bracts very small, ovate, obtuse,
None of the flowers are expanded in the specimens. Flower-buds a
line and a half long. Calyx turbinate, slightly produced beyond the
ovary; the thin margin distinctly four-lobed ; the lobes short and
rounded. Corolla convex in the bud, perhaps falling away as a lid;
but the petals are readily separable. Ovary two-celled; the cells
with about 15 ovules. Fruit unknown.

Puate 61, B—Evcenta (SyzyGiuM) CONFERTIFLORA: a branch with
flower-buds, of the natural size. Fig. 1. Flower-bud, with the oper-
culate corolla detached. 2. Vertical section of the same. 3. Trans-
verse section of the ovary.—The details enlarged.

524 PHANEROGAMIA.

24, Evueenta (Syzycium) EFFusA, Sp. Nov.

E. foliis obovatis subretusis basi acutis breviter petiolatis coriaceis opacis,
venis haud perspicuis; cymis terminalibus decompositis corymbosis
diffusis laxe multifloris; pedunculis folia superantibus, primaris et
secundariis gracilibus acute tetragonis ; calycis margine repando.

Has. Sandalwood Bay, Vanua-levu, Feejee Islands.

Probably a small tree (the size not recorded), glabrous throughout,
with slender ash-gray branches: the ultimate branchlets somewhat
four-angled, very leafy. Leaves obovate, or elliptical-obovate, an inch
and a half long, very obtuse, somewhat retuse, all narrowed and acute
at the base, short-petioled, coriaceous but not very thick, opaque, dull,
punctate; the midrib conspicuous; the veins scarcely visible; a very
obscure false vein close to the revolute margin. Cymes terminal,
decompound, 3 or 4 inches long, trichotomous, diffusely spreading,
loosely many-flowered: peduncles or primary branches of the inflores-
cence 1% to 2 inches long, slender, as also their successive spreading
divisions, «cutely four-angled; the ultimate pedicels a line or two in
length. Bractlets minute, caducous. Flower-buds a line long: most
of the specimens are past anthesis. Calyx turbinate; the free margin
merely repand, dilated after flowering. Corolla hemispherical in the
bud, falling off unopened. Stamens rather short. Ovules 8 to 10 in
each cell. Fruit not seen.

25. Eugenia (Syzyeium) Amicorum, Sp. Nov. (Tab. 62.)

E. folus oblongo-lanceolatis utrinque subacuminatis subcoriaceis opacis
ereberrime ac tenuiter penninervis reticulatis; cymis terminalibus
decompositis effusis multifloris folia equantibus ; pedunculis primariis
et partialibus gracilibus compressis; floribus pedicellatis; alabastris
subglobosis ; calycis margine repando ; fructu depresso-globoso.

Has. Tongatabu (in flower). Feejee Islands (in fruit).

A tree or shrub (the size not recorded), glabrous throughout, with

MYRTACES. 525

slender, terete branches and branchlets. Leaves oblong-lanceolate,
subacuminate at both ends, or sometimes obtuse, 3 to 5 inches long,
rather coriaceous, usually dull and pale both sides, opaque, scarcely
punctate, closely feather-veined, the reticulated veins slender but conspi-
cuous, especially underneath, oblique, the intramarginal false vein
rather obscure. Petiole 2 to 5 lines long. Cymes terminal, decompound,
effusely many-flowered, equalling the leaves, somewhat paniculate; the
primary and partial peduncles slender, compressed, about an inch long ;
the ultimate and penultimate divisions umbellately fascicled: pedicels
one or 2 lines long. Bracts and bractlets very caducous, probably
minute. J lower-buds globular, scarcely a line in diameter. Calyx
with a repandly four-lobed margin. Petals small, cohering in a lid,
which falls off in anthesis. Stamens only a line long. Ovules about
7 in each cell. Fruit a depressed-globose berry, 3 lines in diameter,
one-celled, one-seeded; the margin of the calyx truncate, not pro-
duced. Seed globose; the thick cotyledons peltately attached to the
slender included radicle.

This may be the Hugenia? paniculata, mentioned but not charac-
terized by Forster (Prodr. p. 90, not of Lam.), so far as can be judged
from some notes on an imperfect original specimen, although that has

nearly sessile leaves. The Syzygium paniculatum of Gertner, as Mr.

Bennett obligingly informs me, is founded neither on Forster's speci-
mens, nor on a species from Isle Bourbon, as DeCandolle supposed,
but on a widely different plant, of uncertain genus, collected at
Botany Bay, by Banks and Solander.

PLate 62.—EKueenta (Syzyerum) Amicorum: a flowering branch,
of the natural size. Fig. 1. Flower, the operculate corolla detached.
2. Vertical section of a flower-bud. 3. Transverse section of the
ovary. 4. Fruit, of the natural size. 5. Transverse section of the
same, and of the embryo. 6. The embryo. 7. Same, with the coty-
ledons separated.—The analyses magnified.

26. EuGENIA (SyzYGIUM) RUBESCENS, Sp. Nov. (Tab. 63.)

E. foliis oblongis sew lanceolato-ellipticis utrinque acuminatis subcoriaceis
opacis crebre ac tenuiter penninerviis ; cymis terminalibus paniculato-
decompositis patentibus folia multo superantibus ; pedunculis primariis

132

526 PHANEROGAMIA.

et partialibus gracilibus teretiusculis; calyce clavato-turbinato rubes-
cente, margine tenui repando-subyuadrilobo.

Has. Ovolau, Feejee Islands; in woods, at the height of 1,500 feet.

Probably a tree (the size not recorded), glabrous throughout, with
slender terete branches and branchlets. Leaves oblong or elliptical-
lanceolate, more or less acuminate at both ends, about 3 inches long by
12 to 20 lines in width, rather coriaceous, opaque, pale beneath, dull
above, slightly punctate, closely feather-veined as in the preceding
species, but the slender veins less oblique and reticulated; the intra-
marginal false vein inconspicuous. Petiole 3 or 4 lines long. Cyme
terminal, decompound, paniculate, spreading, very many-flowered, about
6 inches in length and breadth; the peduncles and their divaricate
divisions slender, almost terete; the pedicels slender, one-flowered, or
oftener three-flowered at their apex. Bracts and bractlets minute,
very caducous. Calyx turbinate-club-shaped (dilated at the summit,
the tube tapering down to the articulation at its pedicelliform base),
tinged with red or purple, 24 lines long, produced beyond the ovary
into a campanulate cup; the thin margin repandly four-lobed ; the lobes
very short, at length deciduous. Petals flesh-colour, small, lightly
coherent into a lid which falls off without expanding. Stamens flesh-
colour, nearly 3 lines long. Ovary two-celled. Ovules 8 or 10 in each
cell. Fruit not seen.

In the form of the calyx, as well as in foliage and habit, this species
bears no slight resemblance to Eugenia Smithii, the Acmena floribunda
of DeCandolle, with which it is certainly congeneric, although the
petals are lightly coherent and are thrown off without expanding.

PLATE 63.—EuGENIA (SYZYGIUM) RUBESCENS: flowering branch, of
the natural size. Fig. 1. Flower-bud. 2. Expanded flower, with the
corolla, 3, detached. 4. Vertical section of a flower. 5. Transverse
section of the ovary.—The details magnified.

27. EKuGENIA (SyZyGium) coRyNocARPA, Sp. Nov. (Tab. 64.)

E. folis elliptico-lanceolatis oblongisve subacuminatis basi acutis breviter

MYRTACE &. 527

petiolatis fere membranaceis laxe penninervits, venis intra marginem
arcuato-anastomosantibus ; cymis paniculatis decompositis folia supe-
rantibus ; pedunculis partialibus gracilibus divaricatis ; calyce clavato,
limbo subintegro; fructu clavato-fusiformi ; semine oblongo.

Has. Feejee, Samoan or Navigators’ (and Society?) Islands.
(Probably through some mistake one of the specimens is ticketed,
“ Tahiti.”)

Apparently a tree (the size not recorded), with slender and spread-
ing terete branches, glabrous throughout. Leaves nearly membra-
naceous in texture, dull both sides, scarcely paler beneath, elliptical-
lanceolate or oblong, more or less acuminate, acute at the base, about 4
inches long and an inch and a half wide, on a short petiole of a line
or two in length, somewhat pellucid-punctate, loosely feather-veined ;
the veins rather conspicuous, sparingly reticulated underneath, mostly
confluent into an undulate false vein a little distance within the
margin. Cymes terminal and sometimes axillary, ample, decompound,
paniculate, diffuse, 5 to 8 inches in length, peduncled; the slender
partial peduncles divaricately spreading, terete, many-flowered; the
ultimate divisions 3—5-flowered. Bracts and bractlets minute, cadu-
cous. Flowers subsessile on the ultimate ramifications, divergent ;
the flower-buds club-shaped, 2 to 3 lines long, including the pedicel-
like base. Calyx clavate; the summit produced beyond the ovary
and subglobose in the bud, dilated after anthesis; the margin trun-
eate, nearly entire, or obscurely repand-four-toothed, the broad teeth
at length deciduous. Corolla of 4 rounded small petals, coherent in
a lid, which falls off entire in anthesis. Stamens numerous, inserted
just within the edge of the calyx, not collected in phalanges, inflexed
in westivation, 2 to 3 lines long. Style as long as the stamens: stigma
simple. Ovary two-celled, with 8 or 10 ovules in each cell. Frwit
fleshy, becoming dry? clavate or spindle-shaped, almost an inch long,
83 or 4 lines wide, one-celled, one-seeded. Seed oblong, with a very thin
testa. Cotyledons oblong, fleshy, not very thick, not conferruminate,
but sinuously embracing each other, peltately attached near their
middle to the slender and straight, included, superior radicle. The
fruit as well as the ovary is inodorous and tasteless.

This ambiguous plant has the embryo as well as the elongated

528 PHANEROGAMIA.

calyx-tube of Caryophyllus, but the other characters are those of
Syzygium, although the fruit is more elongated than usual. It is
probably a congener of Blume’s three Javan species referred to Caryo-
phyllus in DeCandolle’s Prodromus, although the flower-buds are
insipid and inodorous: but it cannot be generically separated from
the preceding species and its near allies.

Pate 64,—EvGENtA (SYZYGIUM) CORYNOCARPA: branches in flower
and in fruit, of the natural size. Fig. 1. Vertical section of a flower-
bud. 2. A flower, with the corolla separating as a lid. 3. Trans-
verse section of the ovary. 4. A fruit. 5. Vertical section of the
same, showing the seed, &c. 6. Transverse section of a fruit and
seed. 7. Embryo. 8. The same, with the cotyledons separated.—
The details magnified.

+t Flores majores; limbo calycis magis quadrilobo, lobis estivatione imbricatis demum
sepius deciduis. (Transitus ad Microjambosam, Blume.)

28. Kucrnta (Syzye@rum) cLust@ro.iA, Sp. Nov. (Tab. 65.)

EL. foliis obovatis seu ovalibus basi cuneatis apice rotundatis nune retusis
chartaceis creberrime penninerviis reticulatis supra lucidis ; cymis ter-
minalibus decompositis paniculato-corymbosis multifloris ; pedunculis
teretibus; alabastris obovoideo-turbinatis carnosis ; calycis margine
tenur repando-quadrilobo.

Has. Samoan Islands, Tutuila and Savaii: on rocks near the sea.

Apparently a tree of considerable size, glabrous throughout, with
terete branchlets. Leaves obovate or inclining to oval, with a cuneate
base, rounded and often retuse at the apex, sometimes rather apiculate,
4 or 5 inches long, 24 to 8 broad, chartaceous or somewhat coriaceous,
green both sides, dull beneath, Zuecid above, pellucid-punctate, conspi-
cuously feather-veined with very close and slender somewhat transverse
veins, which run almost straight to near the margin, where they unite
with a slender intramarginal vein; the veinlets reticulated. Petiole
4 or 5 lines long. Cymes apparently collected in an ample compound
corymbose panicle, of 8 or 10 inches in length; the peduncles and their

MYRTACE &. 529

~_

ramifications ¢erefe, spreading, rather stout: pedicels very short.
Flowers very numerous ; the buds obovoid-turbinate, thick, 4 lines long,
including the pedicelliform base. Calyx very fleshy, produced beyond
the ovary; the truncate thin margin repandly somewhat four-lobed ;
the lobes short and broad, doubtless deciduous. Corolla rather large,
falling off as a lid in anthesis. Stamens very numerous, 5 or 6 lines
long. Ovary two-celled; the cells several-ovuled. Fruit not seen.

This is certainly a Syzygium, and a well-marked and _ peculiar
species, with its lineately-veined leaves not unlike ‘those of a Clusia
in appearance. A sterile shoot perhaps of this species was gathered
at Tongatabu.

Puate 65.—EuGENIA (SYZYGIUM) CLUSLEZFOLIA: a branch, in flower,
of the natural size. Fig. 1. A flower-bud, with the corolla becoming
detached. 2. Vertical section of a flower. 3. Transverse section of
the ovary.—The details magnified.

29. Eugenra (Syzyeium?) TurviLensis, Sp. Nov.

E. foliis oblongis seu ellipticis utrinque subobtusis chartaceis crebre penni-
nervis, venis intra marginem in venam falsam nune duplicem con-
Jluentibus ; cyma terminali pluriflora foliis breviore; alabastris (par-
vulis) haud pedicellatis obovatis ; calycis limbo breviter quadrilobo.

Has. Tutuila, one of the Navigators’ or Samoan Islands.

There is only a miserable specimen in the collection, with unex-
panded flowers. Branches terete. Leaves 5 to 7 inches long, 2 or 3
inches wide, chartaceous in texture, glabrous, as is the whole plant,
somewhat lucid, oblong or elliptical, obtuse at both ends, or abruptly
somewhat contracted at the apex into a very obtuse and short acumi-
nation, closely feather-veined ; the veins straight, sparingly reticulated,
connected within the margin by a false vein, and often with another
less distinct one close to the margin. Petioles 4 or 5 lines long.
Cyme terminal, small, rather simple, raised on a peduncle nearly an
inch long, much shorter than the leaves. lowers sessile in threes or
fives at the extremity of the divergent partial peduncles. /lower-

buds obovate, with an acute base, only 3 lines long; the limb of the
133

530 PHANEROGAMIA.

calyx rather distinctly four-lobed; the lobes short and broad, imbri-
cated in estivation. The corolla apparently falls off without expand-
ing. The flowers are small, but the foliage appears like that of a
Jambosa.

30. Evgenia (SyzyGium?) SAVAIENSIS, Sp. Nov.

E. foliis membranaceis oblongis basi acutis apice acuminatis laxe penni-
nerviis, venis intra marginem arcuato-anastomosantibus ; cyinis corym-
bosis multifloris; pedunculis partialibus subangulatis, ultimis brevis-
simis incrassatis ; alabastris obovato-clavatis (semipollicaribus) ; calycis
margine repando-quadrilobo.

Has. Savaii, one of the Samoan or Navigators’ Islands.

Probably a tree of considerable size; glabrous throughout. Leaves
membranaceous, green both sides, dull, 7 to 9 inches long and 3 or 4
wide in the middle, oblong-lanceolate or oblong, conspicuously and
acutely acuminate, the base acute or somewhat acuminate, loosely
feather-veined ; the veins oblique, somewhat branched, arcuate and
anastomosed some distance within the margin, but not forming a dis-
tinct intramarginal vein. Petiole an inch or more in length, slender.
Cyme corymbose, many-flowered, peduncled, shorter than the leaves, its
thickish branches somewhat two-edged ; the ultimate divisions very short,
bearing about 3 subsessile flowers. Flower-buds obovate-clavate, fully
half an inch long, including the acute base, above 3 lines in diameter.
Calyx produced considerably beyond the ovary; the margin repandly
Jour-lobed ; the lobes very short and broad, probably deciduous from
the fruit. Petals forming a convex lid in the bud, readily detached
from the base, and probably falling away together in anthesis; but
they are only slightly coherent. Stamens very numerous, two-thirds

-of an inch long. Fruit not seen.

The specimen of this ambiguous plant is a very imperfect one.
Perhaps the petals expand in anthesis. This and the foregoing
species have the habit of Jambosa, and should perhaps be referred to
Micrqjambosa ; but the lobes of the calyx are not more manifest than
in many Syzygia.

MYRTACE ZA, 531

31. EuGEnta (SYZYGIUM?) BREVIFOLIA, Sp. Nov.

E. foliis parvis ovato-oblongis obtusis basi rotundatis subsessilibus sub-
coriaceis punctatis subtus leviter penninerviis; cymis terminalibus
parvis; pedunculis ramulisque tetragonis ; bacca (magna) ovoidea.

Has. On the mountains of Tutuila, one of the Samoan Islands, at
the elevation of 2,500 feet.

There is only a single, fruiting branch of this plant in the collec-
tion; apparently from a shrub. Branches very leafy, glabrous, as is
the whole specimen; the slender branchlets, as well as the short and
few cymose peduncles, quadrangular, the angles acute. Leaves small,
an inch or less in length, evate-oblong, obtuse, or slightly and obtusely
pointed, rounded at the base, subsessile (the petioles only half a line
long), coriaceous in texture, dull, pellucid-punctate by transmitted
light, lightly feather-veined beneath. Flowers not seen. Mature fruit
ovoid, baccate, large for the foliage, half an inch long and almost of
that diameter near the base, narrowed at the summit, which is deeply
umbilicate; the calyx-teeth probably short, but deciduous: within
one-celled, two-seeded. Cotyledons orbiculate, fleshy, peltate.

There are in the collection undeterminable sterile branches of
several Myrtacece, probably species of Eugenia, from the Samoan,
Friendly, Feejee, and Mangsi Islands, and from Luzon.

8. MYRCIA, DC
1. Myrcta exsucca, Mart.
Myrcia exsucca, Mart. in DC. Prodr. 3, p. 247; St. Hil. Fl. Bras. 2, p. 316.

Has. Organ Mountains, Brazil. (Foliage only.)

2. MyrcrA PUBESCENS, Mart. 1. c.

Has. Brazil, in the neighbourhood of Rio Janeiro.

532 PHANEROGAMIA.

3. Myrcta ELuIpticA, Gardn.
Myrcia elliptica, Gardn. in Hook. Lond. Jour. Bot. 2, p. 352.

Has. Brazil, in the Organ Mountains, near Rio Janeiro.

4. Myrcia spH@RocarPA, DC. 1. c.

Has. Rio Janeiro, Brazil: also from the Organ Mountains, appa-
rently a variety with broader and very obtuse leaves.

5. MyYRcIA FERRUGINEA, DC. 1. c.

Has. Organ Mountains, and in the vicinity of Rio Janeiro, Brazil.

6. Myrcta prunirouia, DC. 1. c.

Has. Organ Mountains, Brazil. (In fruit.)

7. MyRcIA ELLIPTICIFOLIA, Camb.
Myrcia ellipticifolia, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 312.

Has. Organ Mountains, Brazil.

8. Myrcra Pseupo-Min1, DC. 1. ¢.

Has. Organ Mountains, or near Rio Janeiro, Brazil.

9. Myrcra rAMULOSA, DC. 1. ¢.

Has. Near Rio Janeiro. (In flower only.)

MYRTACES, 533

10. Myrcta Ertopus, Mart.

Myrcia Eriopus, Mart. in DC. Prodr. 3, p. 255.

Has. Organ Mountains, near Rio Janeiro, Brazil. (In fruit.)

11. Myrcta sprcrasiiis, DC.

Myrcia spectabilis, DC. Prodr. 3, p. 248; Cambess. in St. Hil. Fl. Bras. 2, p. 299.

Haz. Organ Mountains, Brazil. (With undeveloped flowers only.)

12. Myrcia Brownrana, Gardn.

Myrcia Browniana, Gardn. in Hook. Lond. Sour. Bot. 2, p. 354.

Has. Organ Mountains, Brazil: in forests.

Gardner collected this striking species in flower only. Our speci-
mens have only the full-grown fruit. This is globose, nearly two-
thirds of an inch in diameter, containing two proportionally large
seeds, with a thin and chestnut-coloured testa: the embryo as in the

genus.

13. MyRciA SUBSERICEA, Sp. Nov.

M. foliis breviter petiolatis oblongo-lanceolatis sensim acutissime acumi-
nates transverse lineato-multinervatis supra glabris subtus ramulisque
pube appressisima argenteo-sericeis nitentibus ; pedunculis petiolum
superantibus paucifloris; floribus sessilibus bracteatis ; calyce bracteis-
que argenteo-sericets,

Has. Rio Janeiro, Brazil. (Foliage, with peduncles, from which
the flowers have fallen.) Also gathered by Macrae, with unexpanded

flower-buds, and by Pohl (No. 1047).
134

534 PHANEROGAMIA.

This plant apparently has been gathered by several collectors, but
I do not find it anywhere described. It is remarkable for the fine
satin-like and silvery sheen of the lower face of its leaves, as in some
Malpighiacew. These are 6 or 7 inches long by 12 to 2 in width, on
petioles of 3 lines in length, oblong-lanceolate, acute at the base, and
tapering gradually into a slender and very acute point at the apex,
chartaceous, pellucid-punctate, transversely and closely nervosely
veined (like a Chrysophyllum); the veins confluent into a marginal
vein. Both sides are silvery-silky in the nascent foliage; but the
upper face is soon glabrous and bright green. Peduncles axillary,
half an inch to an inch long, bearing several sessile flowers apparently
in a small condensed cyme, the silky buds subtended by silky oblong-
lanceolate bracts. Expanded flowers not seen. The obovoid calyx
however has a five-cleft limb, the lobes imbricated in exstivation; the
throat silky inside. Ovary two-celled, with 2 ovules in each cell.

14. Myrcra rostrata, DC. ?

Myrcia rostrata, DC. Prodr. 3, p. 255? Cambess. in St. Hil. Fl. Bras. 2, p. 320.

Has. Organ Mountains, Brazil.

15. Myrcta seprartaA, DC? lc.
Has. Organ Mountains, Brazil.

There are specimens of two or three more species, probably of this
genus, but not in a state for determination.

4, MARLIEREA, &. Hil.
1. MARLIEREA SUAVEOLENS, St. Hil.
Marlierea suaveolens, St. Hil. Fl. Bras. Mer. 2, p. 874, t. 156.

Has. Organ Mountains, Brazil.

MYRTACE &. 535

The specimens, with unexpanded flower-buds, exactly accord with
the description and figure above-cited, only the leaves are mostly
larger.

2. MARLIEREA EXCORTICATA, Mart.
Marlierea excorticata, Mart. Herb. Bras. no. 59, & in Flora, 20, Beibl. p. 88.

Has. Organ Mountains, Brazil.

This, which is also Pohl’s No. 1083, has the branchlets glabrous
(not ferrugineous), and the leaves are mostly broader, and obtuser
(or rounded, instead of tapering) at the base, than in the plant of
Martius, of which it appears to be only a marked variety. The
peduncle and the branches of the trichotomous cyme are compressed.
The flower-buds are unexpanded.

5. LUMA, Nov. Gen.

Calyx tubo turbinato vel subgloboso; limbo quadripartito (rarissime
quinquepartito), lobis cestivatione wmbricatis. Petala, stamina, ete.
Kumyrti. Ovarium triloculare, raro biloculare; placentis in locu-
lorum angulo centrali pluriovulatis. Bacca subglobosa, calycis limbo
coronata, bi-trilocularis (dissepimentis quandoque evanidis), oligo—
pleiosperma. Semina reniformi-rotundata sew lenticularia; testa
membranacea a nucleo libera. Embryo cyclico-curvatus: cotyledones
sat magne, ovales vel orbiculate, radicula preelonga semicincte vel fere
circumdate, aut plane subcarnose radiculee accumbentes, aut foliacece
varioque modo contortuplicate.—Frutices vel arbores Chilenses, fra-
grantes; foliis coriaceis; pedunculis axillaribus uni—plurifioris ;
petalis albis.

The species here associated have been variously referred, some of
them to Hugenia, others to Myrtus: but they evidently belong to
neither, as the full details of Plate 66, elaborated with extreme care
and ability by Mr. Sprague, will abundantly show. From Myrtus,

536 PHANEROGAMIA.

with which they accord in the long and curved radicle, they are dis-
tinguished by the ample cotyledons, and the membranaceous testa of
the seed (unless, indeed, Lwma Temu and its immediate allies should
prove to have a hard seed-coat): from Eugenia they are still more
widely distinguished by the number of the seeds (from 3 to 10
usually maturing), by the elongated radicle coiled partly or nearly
round the broad and thin cotyledons; which, although rather fleshy
when plane, and strikingly folded and crumpled together in certain
species, are not at all united. The cotyledons when contortuplicate
much resemble those of Myrcia ; but the ovules are indefinitely nume-
rous in each cell. For the genus, thus defined, I have adopted one of
the native names of a Myrtaceous tree (Myrtus Luma of Molina),
which in all probability belongs to this group, and which has the
merit of being both short and euphonious.*

*

1. Luma Cuexen. (Tab. 66.)

L. foliis ovatis ovali-rotundis seu ovato-oblongis apiculato-acuminatis
subcoriaceis concoloribus subtus penninerviis glabris vel glabratis,
novellis margine costaque subtus cum ramulis pubescentibus ; pedun-
culis folium cequantibus vel superantibus uni—tri-(nune quinque-) floris,
jlore intermedio subsessili; bracteolis minutis caducissimis; calycis
lobis rotundatis petalisque 4; cotyledonibus planis.

* Besides the species comprised in the present collection, or mentioned in the text,
the following, which I possess in flower only, probably belong to the genus :—

Eugenia planipes, Hook. & Arn. Bot. Mise. 3, p. 323, from Childe and Valdivia;
which is allied to Z. Temu.

Myrtus multiflora, Juss., DC. Prodr. 8, p. 240; or, at least a plant from Gay’s
Chilian collection wrongly named “ Lugenia multiflora, Hook.,”’ which accords with
DeCandolle’s character, and is pentamerous! If it belong here it is the only penta-
merous species known to me.

Eugenia leptospermoides, DO. Prodr. 8, p. 266; or, at least Gay’s plant referred by
Barneoud to that species ; which should have been compared with the originals, at hand
in two Parisian herbaria, instead of being guessed at.

Eugenia Gayana, Barneoud, in Gay, FI. Chil. 2, p. 390, a handsome, small-leaved
species.

Also, Eugenia Bridgesii, Hook. & Arn. Bot. Misc. 3, p. 322, which I do not possess,
evidently belongs to the emu group.

Eugenia Cumingii, E. ovata, and E. Selkirkii, and Myrtus? Fernandeziana, Hook,
& Arn. 1. ¢. are unknown to me; as also are Myrtus Coquimbensis, and Eugenia mari-
ttma of Barneoud, in Gay’s Flora Chilena.

MYRTACEZ. 537

Myrtus folio subrotundo vulgd Cheken, Feuill. Obs. 8, p. 45, t. 82.

M. Cheken, Spreng. Syst. 2, p. 85, excl. syn. (“ Hugenia Cheken, Molina,” non est. )
Eugenia Chekan & E. ? apiculata, DC. Prodr. 3, p. 276, 278.

Myrtus Luma, Molina, Stor. Nat. Chil. p. 178, 8352? haud Gay.

Var. a. foliis ovalebus oblongisve utringue acutis; pedunculis plerisque
unifloris.

Eugenia Cheken, Hook. & Arn. Bot. Beech. Voy. p. 56, & Bot. Misc. 8, p. 320;
Gay, Fl. Chil. 2, p. 290.

Myrtus Luma, Schauer, in Rel. Meyen. p. 332 ;—itaque

M. Moline, Gay, Fl. Chil. 2, p. 381.

Var. 8. foliis late ovatis rotundisve basi aut acutis aut obtusis vel rotun-
datis apice subito cuspidatis; radicula (an semper?) cotyledonibus
vix longiore.

Myrtus Luma, Spreng. Syst. Veg. 2, p. 484?
Eugenia apiculata, Hook. & Arn. in Bot. Misc. 3, p. 821; Gay, Fl. Chil. 1. c.

(Var. y. foliis obovato-oblongis plerumque cuspidato-acuminatis.

Eugenia affinis, Gillies, in Hook. & Arn. Bot. Misc. 3, p. 321.
£. apiculata, var. Hook. f. adn. in Fl. Antare. p. 277.)

(Var. 0. foliis ovalibus via cuspidatis ramulisque glaberrimis.
Eugenia Gilliesii, Hook. & Arn. in Bot. Mise. 3, p. 820.)

Has. Chili, near Valparaiso (var. a.) ; and on a mountain 12 miles
in the interior, var. (.

A shrub or small tree; the branchlets more or less pubescent, often
with rusty hairs. Leaves varying from ovate and roundish-oval to
ovate-oblong, sometimes rather obovate, short-petioled, in different
forms either acute at both ends or rounded at the base, the apex some-
times rounded, but commonly tipped with a sharp cuspidate-acuminate
point: they are green both sides, smooth, or glabrate, or with a few
hairs, especially on the midrib underneath and on the margins when
young, in texture only slightly coriaceous, occasionally thicker, opaque,

but somewhat lucid, the midrib salient, the pinnate veins rather
185

538 PHANEROGAMIA.

manifest underneath, and confluent into an intramarginal vein: they
vary from 5 or 6 to 15 lines in length, and from 3 or 4 to 12 in
width. Peduncles equalling or often exceeding the leaves when one-
Howered, usually more elongated when ¢three-(rarely five-)/lowered ; in
the latter case the main peduncle, an inch or so in length and slender,
is flattened (as in the other several-flowered species); the intermediate
flower sessile or nearly so; the others on pedicels of 3 to 7 lines in
length. Bractlets under the flower minute and very caducous. Flowers
nearly as large as those of the Common Myrtle. Calyx almost gla-
brous; the tube obovoid-turbinate; the lobes 4, rownded, imbricated
in estivation, mostly a little ciliolate, as are the four rounded-obovate,
deciduous petals. Stamens very numerous: filaments filiform:
anthers oval. Style filiform: stigma simple. Ovary usually three-
celled, sometimes two-celled, with the placente in the axis. Ovules
numerous in two ranks in each cell, hemitropous, slightly reniform.
Berry globose, pulpy, 5 or 6 lines in diameter, apparently purple,
crowned with the persistent lobes of the calyx, ripening from 3 or 4
to 9 or even 12 seeds. Seeds 1% to 2 lines in diameter, compressed,
orbiculate, often excised at the hilum; the testa membranaceous,
chestnut-coloured, smooth, not adnate to the embryo. Radicle very
long, cylindrical or somewhat club-shaped, incurved. Cotyledons
broadly oval, accumbent to the semiannular radicle, fleshy but rather
thin and flat, plane or sometimes slightly curved laterally, not cohe-
ring with each other or at all conferruminate, shorter than the radicle.
In all the seeds examined of var. 2. the radicle is truncate, as if the
extremity were cut off nearly in a line with the summit of the
cotyledons, instead of continuing so as to half surround them or more.
Taken with the difference in foliage, this would indicate the Eugenia
apiculata, Hook. & Arn. as a distinct species; but I suspect that the
peculiarity is a casual one.

The plants here brought together pretty clearly belong (with the
possible exception indicated above) to one somewhat variable species,
which is doubtless the Myrtus Cheken of Feuillée, who first made
the plant known. Probably it likewise includes the Myrtus Luma
of Molina, to which Dr. Hooker’s Eugenia apiculata var. Arnyan
[Arrayan ?] would very well correspond, having broadly oval leaves
aud one-flowered peduncles. The only question is as to its size, and
the nature of the wood: the Luma according to Molina being a tree

MYRTACES. 539

frequently of forty feet in height, and its wood much valued for the
use of the coach-makers. Gay, indeed, asserts that the true Luma
(his Myrtus Luma) is a shrub, with ovate-lanceolate leaves, a native
of the south of Chili; but to this, with reason, he cites the synonyme
of. Molina with a mark of doubt. I have not seen his specimens of
the plant in question. Schauer’s Myrtus Luma, cited above, is from
the vicinity of Valparaiso, and must needs belong to our first variety,
although with unusually small leaves. Sprengel’s J Luma answers
to our var. (.

Piate 66, A.—LuMA CHEKEN, var. a.: branches, in flower and in
fruit, of the natural size. Fig. 1. An unexpanded flower. 2. The
same, expanded. 3. Astamen. 4. Vertical section of the ovary, &c.
5. Transverse section of a three-celled ovary. 6. Similar section of
a two-celled ovary. 7. A fruit. 8. Vertical section of the same,
showing some of the seeds. 9. A seed. 10. Vertical section of the
same and the contained embryo. 11. Transverse section of the same.
12. The embryo detached.—The details variously magnified.—B.
A branch of var. @., in fruit, of the natural size. Fig. 13. A seed,
magnified. 14. Embryo, from the same, magnified.

2. Luma TEmu.

L. foliis ovalibus ellipticisve nune obovatis obtusissimis coriaceis glabris
subtus pallidis obsolete penninerviis, novellis ramulisque ferrugineo-
puberulis; pedunculis (solitariis ternis quaternisve) folia cequantibus
apice umbellatim 3—T-floris, flore intermedio sessili, ceeteris pedicellatis,
pedicellis duobus majoribus seepe trifloris; bracteolis subulatis persis-
tentibus ; calycis lobis rotundatis demum reflexis petalisque 4.

Eugenia Temu, Hook. & Arn. Bot. Beech. Voy. p. 56, & in Hook. Bot. Mise. 3, p. 322.
E. multiflora, Hook. & Arn. in Bot. Mise. l.¢., ex char. (var. foliis apice subat-
tenuatis); Gay, l.c¢.?

Has. Chili, near Valparaiso, &c.; common.

Of this pretty well-marked species, as of the following, I have seen
no matured fruit: but the seeds and embryo will most likely be found

540 PHANEROGAMIA.

to accord with the present genus. The leaves are oval or elliptical,
sometimes inclining to obovate or rotund, usually very obtuse, short-
petioled, coriaceous in texture, dull and very opaque, an inch to an
inch and a half in length, 8 to 12 lines wide, whitish and obscurely
Feather-veined beneath, glabrous or glabrate at maturity, the young ones,
like the branchlets, ferrugineous-pubescent, the pubescence fine and
close. Peduncles solitary or even 3 or 4 from the same axil, slender,
flattish, as long as the leaves, or sometimes longer, only the depaupe-
rate ones one-flowered, commonly 3—7-flowered in a kind of umbel, the
central flower sessile, the others on pedicels of 2 to 4 lines long, or the
stronger pair again trifid at the apex, forming a 7—-9-flowered cyme.
Bracilets at the forking and at the base of each flower subulate, persis-
tent. Flowers apparently smaller than in the foregoing species, but
of the same structure. Tube of the calyx turbinate, ferrugineous,
puberulent; the lobes rounded, reflewed after anthesis. Ovary three-
celled; the cells containing numerous ovules.

3. LumMA CRUCKSHANKSII.

Eugenia Oruckshanksti, Hook. & Arn. in Bot. Mise. 8, p. 321.
Has. Chili, near Valparaiso.

Of this species we have in the collection only a fragment, destitute
both of good flowers and of fruit. It appears to be nearly allied to
the preceding species. The seeds, which are said to have a crusta-
ceous testa, and the cotyledons, said to be conferruminate, should be
re-examined.

4. LUMA STENOPHYLLA.

Eugenia stenophylla, Hook. & Arn. Bot. Misc. 8, p. 822; Gay, Fl. Chil. 2, p. 395.
Myrtus Gudilla, Colla, in Mem. Acad. Turin. 87, p. 66, t. 15.

Has. Chili; along the banks of the river near Valparaiso.

This species, with the allied Z. (Hugenia, Hook. & Arn.) planipes,
belongs to the same group as L. Jemu. I have not seen the fruit,
which is said by Gay to be a three-celled and few-seeded berry. The

MYRTACES. 541

peduncles, occasionally geminate, are either one-flowered, or two-
flowered (the intermediate flower wanting), or more commonly three-
flowered, as described. Sometimes the inflorescence becomes terminal
and somewhat racemose on the branches, by the reduction of the
upper leaves to bracts. The conspicuous linear-subulate bractlets
are persistent. The leaves occasionally become alternate-— The
name imposed by Colla is of the same date as that of Hooker and
Arnott.

d. Luma optusa. (Tab. 66.)

L, foliis ovato-ellipticis vel subrotundis utrinque obtusis coriaceis planis
punctatis glabris subtus pallidis uninerviis aveniis, petiolis ramulisque
junioribus rufo-pubentibus; pedunculis solitariis wnifloris folium
aquantibus ; bracteolis persistentibus oblongis folraceis lobos 4 calycis
oblongo-ovales patentes adcequantibus ; cotyledonibus contortuplicatis.

Eugenia obtusa, DC. Prodr. 3, p. 266?
Myrtus Raran, Colla, in Mem. Acad. Turin. 37, p. 66, ex char.

Has. Chili; on hills or in ravines, near Valparaiso.

A low shrub, apparently only a foot or two in height; the branches
very leafy; the younger branchlets ferrugineous with a close and fine
pubescence, of which the distinct petioles (scarcely a line long),
peduncle and calyx-tube partake, but even the nascent foliage shows
only slight traces of it. The leaves, however, are somewhat ferru-
gineous in hue underneath when young: they are thick and coria-
ceous, flat (the margins not revolute), glabrous and punctate both sides,
green above, pale beneath, where they are one-nerved with a rather
conspicuous midrib, but wholly veinless; in shape they are ovate-ellip-
tical, varying to roundish or to oblong, very obtuse at both ends, from 3
to 7 lines long, and from 2 to 4 lines broad. Pedunecles solitary, axil-
lary, one-flowered, about the length of the leaf, at the apex furnished
with two persistent bractlets, which are larger in proportion than in
other species, linear-spatulate or oblong, foliaceous, equalling the
flower, or at least the 4 spreading, oblong-oval lobes of the calyx.
_ Flowers, only seen after the fall of the petals and stamens, apparently

twice or thrice the size of those of L. ferruginea. Ovary many-
136

542 PHANEROGAMIA.

ovuled. Berry globular, about 3 lines in diameter, crowned with the
conspicuous calyx-lobes, three-celled; the cells usually ripening 4 or 5
pretty large and turgid-lenticular seeds. Testa thin and membra-
naceous. Radicle long, semiannular. Cotyledons large, orbicular,
nearly foliaceous (as in Myrcia), strongly and apparently variously
contortuplicate, in a manner best shown by the illustrations given on
plate 66.

This species seems not to have been noticed by Hooker and Arnott;
but it is without much doubt the Myrtus Raran of Colla, described
from Bertero’s specimens. Moreover, it accords so well with the cha-
racter of DeCandolle’s Hugenia obtusa, that, knowing there is no little
confusion between the Peruvian and Chilian specimens of Dombey’s
collection, I venture to adopt this specific name, rather than the
aboriginal appellation. It belongs to the group that comprises the
two following species, both of which probably have thin and contor-
tuplicate cotyledons.

PLATE 66, D.—LuMA optusa: a fruiting branchlet, of the natural
size. Fig. 15. A seed, magnified. 16. Vertical section of the same,
showing the contortuplication of the embryo. 17. A less mature
embryo detached, magnified.

6. LUMA FERRUGINEA.

Eugenia ferruginea, Hook. & Arn. in Bot. Mise. 3, p. 319.
Myrtus rufa, Colla, in Mem. Acad. Turin. 37, p. 66.

Has. Chili; in ravines, near Valparaiso.

A very small-leaved and small-flowered species, well characterized
by both authors cited above; who notice that even the midrib of the
leaves is not apparent. The fruit in all our specimens has been a
nidus for insects; so that I have not seen the seeds.

7. Luma corrE@Fouia. (Tab. 66.)

Eugenia correefolia, Hook. & Arn. in Bot. Mise. 8, p. 319.

MYRTACES. 543

Has. Chili; on a mountain ridge, about twelve miles south of Val-
paraiso.

The flowering plant is well described by Hooker and Arnott. The
flowers are larger than in any of the foregoing species. The persis-
tent bractlets are sometimes connate with the base of the calyx.
The berry is obovoid, 5 or 6 lines long, ripening 3 or more seeds.
Cotyledons strongly complicate and contortuplicate, thin and _ folia-
ceous. Radicle truncate at the end (as in L. Cheken @.), very long,
nearly encircling the cotyledons. The character of the embryo is
best shown by the analyses given on plate 66.

PLATE 66, B—LuUMA CORREMFOLIA. Fig. 18. Vertical section of a
fruit (the parts of the flower persistent) and of two contained seeds.
19. Embryo, detached. 20. Transverse section of the same, cutting
twice through the radicle.-—Magnified.

6. MYRTUS, Tourn.

§ 1. EUMYRTUS.—Filamenta indefinita filiformia: anthere didyme medio affixe.

Ovarium bi-triloculare, placentis ex anyulo interno loculorum projectis.

1. Myrtus communis, Linn.

Has. Madeira. A narrow-leaved form: nearly the variety Lusv-
tanica.

2. Myrrus BULLATA, Banks & Soland.

Myrtus bullata, Banks & Solander, Ic. & Ms.; A. Cunn. Bot. N. Zeal. in Ann. Nat.
Hist. 3, p. 115; Hook. Ic. Pl. t. 557; Hook. f. Fl. N. Zeal. p. 70.

Has. Bay of Islands, New Zealand. (In fruit.)
I do not possess the flowers. The fruit and seeds, which are those

of a true Myrtus, are well represented by Hooker, in the plate above
cited.

544 PHANEROGAMIA.

§ 2. LEANTRIA. (Leantria, Soland. ex Forst.)—Filamenta suhdefinita vel pauctus-
cula: anthere didyme basi vel medio affixe. Ovarium imperfecte bi-triloculare,
placentis inter se liberis.—Frutices vel suffrutices, foliis Vaccinii sew Oxycocci.

3. Myrtus PEDUNCULATA, Hook. /.
Myrtus pedunculata, Hook. f. in Hook. Ic. Pl. t. 629, & Fl. N. Zeal. p. 71.

Has. New Zealand.

There are in the collection some sterile leafy stems, perhaps be-
longing to a state of this species; of which I possess no specimens.

4. Myrrus Nummuuaria, Poir.

Myrtus Nummularia, Poir. Dict. 4, p. 407; Gaud. in Ann. Sci. Nat. 5, p. 106, t. 2;
DC. Prodr. 8, p. 238; Hook. f. Fl. Antare. p. 276.

Haz. Orange Harbour, Fuegia. Very common; creeping over the
ground like a Craneberry.

The ovary I find to be incompletely three-celled, in the manner of
the Quitensian IM phylicoides and M. oxycoccoides of Bentham (and
probably the other allied Andine species): at least the dissepiments,
if they meet, do not cohere in the axis of the cell, except at the top
and bottom. The numerous seeds are less curved than in true
Myrtus ; but the embryo is the same.

To a third section of the genus* belongs Myrtus Ugni of Molina,

* §3. UGNI. (Ugni, Turcz. in Flora, 81, p. 711, excl. char. seminum !\— Filamenta
brevia, nune complanata, indefinita, exteriora majora: anthere oblonge, apiculate,
introrsum adnate. Ovarium perfecte bi-triloculare; loculis septo spurio e dorso
carpellorum orto pl. m. bilocellatis. Embryo Myrti!

Myrrus Uant, Molina, DC. ete. Hugeni Ugni, Hook. & Arn. Bot. Mise. 3, p. 318,
where, through some mistake, the flowers are said to be quadrifid, and the peduncles
shorter than the leaf. To this clearly belongs the plate as well as the description of

MYRTACE &. 545

which has short and flattened filaments, bearing adnate anthers, a
completely bi-trilocular ovary, and with the cells becoming more or
less perfectly bilocellate by the growth of a false partition from the
back of each cell to meet the prominent placentx, much as in Rhodo-
myrtus (the character of which is somewhat weakened by this) ;—in
fact differing from Humyrtus just as true Vacciniwm and the section
Vitis-Idea differ from the section Cyanococcus, Gray, Chlor. Bor.-Am.
p- 53. Turezaninow (loc. supra cit.), remarking these characters
(except that he does not notice the dorsal septum that meets each
projecting placenta), has founded on the flowers of this plant his genus
Ugni. But his character of the embryo must have been derived from
some other plant, or else it is incorrect: for the seed and hippocre-
pical embryo in my specimens of Myrtus Ugni, from C. Gay’s Chilian
collection, are just as in Myrtus nummularia, &c.; the testa is almost
bony and reniform, the radicle is long and slender, and the cotyledons
are short, narrow, and semicylindrical, not at all conferruminate.
Turezaninow may have had the seed of some species of Luma (vide
p. 535): yet in these the radicle is not short, but remarkably long,
nor are the cotyledons truly conferruminate, nor the testa hard and
crustaceous, in any species that I have examined.

{ Species dubia.
5. Myrrus? renurroura, Smith.
Myrius tenuifolia, Smith, in Trans. Linn. Soc. 3, p. 280; DC. Prodr. 3, p. 241.

Has. Near Sydney, New South Wales.

This has scarcely been noticed, so far as I am aware, since the time
*

Feuillée, the figure being doubtless reduced in size. For the leaves are represented of
barely 3 lines in length, while in the description they are said to resemble those of the
Myrtle of Tarentum, and to be 7 or 8 lines long. A true congener, undoubtedly, is the

Myrrus CanpoLii, Barneoud, in Gay, Fl. Chil. 2, p. 382, from the province of
Childe; which I possess with undeveloped flowers only. It is pentamerous, and has the
same short, oblong or sagittate and apiculate anthers, and short filaments as M. Uyni, the
filaments not exceeding the anthers in length in the full-grown flower-buds; but they
are not dilated nor flattened. It therefore further invalidates the character on which

Turczaninow most relies to distinguish his genus Uynt.
187

546 PHANEROGAMIA.

of Smith, who has well described the foliage, &. Our specimen bore
only one or two small fruits, not larger than a pepper-corn, containing .
2 or 8 roundish-reniform, not strongly incurved seeds; the embryo as
in Myrtus or Rhodomyrtus. Perhaps it might be referred to the latter
genus, although the leaves are strictly one-nerved; but, as well as I
can ascertain, the ovary appears to have been completely two-celled,
and without any evident dorsal partitions.

7 RHODOMYRTUS, DC.

Myrtus sect. Roopomyrtus, DC. Prodr. 3, p. 241.
Ruopomyrtus, Benth. in Hook. Lond. Jour. Bot. 2, p. 220; Wight, Spicil. Nielgh.
1, p. 59, t. 71; Blume, Mus. Bot. Lugd.-Bat. p. 76.

1. Ruopomyrtus ToMENTOSA, Wight, 1. ¢.

Myrtus tomentosa, Ait.; Sims, Bot. Mag. t. 250; Wight, Ic. Pl. Ind. Or. t. 522, &
Ill. Ind. Bot. 2, t. 97.

Has. Singapore.

The true internal structure of the ovary was first indicated by
Mr. Bentham, in the London Journal of Botany, above-cited. Per-

haps the genus may be again reduced to a section of Myrtus, of which
it has the embryo.

8. RHODAMNIA, Jack.

*
Rwopamni, Jack, in Malay. Mise. & Hook. Comp. Bot. Mag. 1, p. 153; Blume,
Mus. Bot. Lugd.-Bat. p. 78.

Monoxora, Wight, Ill. Ind. Bot. 2, t. 122, & Ic. Pl. Ind. Or. 2. t. 524.

1. RHopAMNIA CINEREA, Jack. 1. c.

Monoxora spectabilis, Wight, 1. c.; Benth. in Hook. Lond. Jour. Bot. 2, p. 219, non
Rhodam. spectab., Blume ?

Myrtus smilacifolia, Wall. Cat. no. 8629.

MYRTACE &. 547

Has. Singapore: several forms, including the varieties concolor and
laxiflora of Blume.

2. RHODAMNIA TRINERVIA, Blume, 1. c.

Myrtus trinervia, Smith, in Trans. Linn. Soe. 3, p. 280.
Eugenia? trinervia, DC. Prodr. 3, p. 279.
Monoxora rubescens, Benth. in Hook. Lond. Jour. Bot. 2, p. 219.

Has. Sydney, New South Wales: with a smoother and smaller-
leaved variety.

9, NELITRIS, Gerin.
1. NELITRIS FRUTICOSA. (Tab. 60.)

N. foliis ovatis sew ovato-lanceolatis acuminatis ; pedunculis axillaribus
unifloris folio multum brevioribus; limbo calycis quinquedentato ;
Fructu decaspermo.

Decaspermum fruticosum, Forst. Char. Gen. p. 74, t. 37. (Deser. in Guill. Zeph.
Tait. p. 58.)

Psidium Decaspermum, Linn. f. Suppl. p. 252; Forst. Prodr. Fl. Ins. Austr. p. 39.

Nelitris Jambosella, DC. Prodr. 3, p. 231, quoad pl. Forst., non Geertn.

Jossinia, aff. J. cassinoidi, Guill. Zeph. Tait. l.c.?

Has. Tonga Islands: the specimens without flowers or fruit.
Feejee Islands: with fruit. Not in the collection from Tahiti (where
alone it was gathered by Forster), but perhaps noticed there by Dr.
Pickering, as, “Myrtus? A shrub in mountain forests; rare: the
fruit red, about the size of a cherry.”

No flowers of this plant were collected. The foliage so closely
resembles that of the following species as to suggest that the diffe-
rence in inflorescence might be merely sexual, were it not that there
are fertile specimens of the latter with cymose-paniculate flowers.
Nor can it be the N. Jambosella, Geertn., of Ceylon, which has long

548 PHANEROGAMIA.

and slender peduncles, equalling the leaves; while in our plant they
do not exceed the petiole, and are only 3 or 4 lines in length, as they
are described by Forster; with whose detailed description (published
by Guillemin) the scanty specimens well accord.

The structure of the ovary in Nelitris is explained by Bentham (in
Lond. Jour. Bot. 2, p. 220), but it appears not to have been rightly
understood by Blume, who, in Mus. Bot. Lugd.-Bat. p. 72, has given
a detailed generic character, and described ten species of the Indian
Archipelago.

Puate 60, D.—NELirris rruTicosaA. Fig. 1. Unripe fruit, with its
peduncle and a pair of leaves, of the natural size. 2. Transverse
section of the fruit, magnified.

2. Nevirris Viriensis, Sp. Nov. (Tab. 60.)

N. foliis ovatis sew ovato-lanceolatis acuminatis basi acutis vel attenwatis
glabratis supra lucidis venis obliquis obsolete penninerviis, junioribus
ramulisque tenuiter sericeo-pubescentibus; cymis laxifloris foliolosis
Solio equilongis ; calycis tubo cano-sericeo, lobis 5 ovatis obtusis ; mase.
ovario ubortivo.

Var.? 6. jfolus basi obtusis; floribus hermaphroditis. ovario primum
guinqueloculari, loculis deinde subbilocellatis.

Has. Ovolau and Muthuata, Feejee Islands: “a common shrub in
Open grounds.”

There are here combined, with some misgiving, male specimens of
the plant delineated on Plate 60, B, which have the ovate or ovate-
lanceolate leaves more or less tapering to both ends, and a specimen with
hermaphrodite flowers (the var. 3.?), the leaves of which are mostly
broader and rounded or obtuse at the base. No other difference of any
consequence is noticed. The branches are slender; the younger
branchlets and nascent foliage are silky with a fine and slender pubes-
cence, which disappears with age, except on the midrib beneath or
some of the lower surface of the leaves. The latter are 14 to 2

MYRTACES, 549

inches long, opaque, lucid above, obscurely and obliquely feather-veined,
on petioles of 2 or 3 lines in length. Flowers in axillary and ter-
minal, many-flowered, paniculate, loose cymes, as long as the subtending
leaf; the principal branches bracteate with small leaves, which on the
slender and silky-pubescent pedicels are reduced to linear or subulate,
mostly alternate bractlets. Flower-buds a line and a half in dia-
meter. Calyx-twbe in hermaphrodite flowers obovoid, silky-canescent
and silvery; the lobes 5, ovate, obtuse, often rather unequal. Petals
orbicular, white, somewhat ciliate. Stamens, &c., as in the genus.
Flowers of the male specimens destitute of a style and of an ovary
(the apex of the pedicel merely clavate-thickened): those of the her-
maphrodite plant with a style as long as the stamens, tipped with a
depressed-capitate stigma, and with a five-celled ovary ; each cell, at the
time of anthesis, showing a projection from its back (in the manner
of Vaccinium, § Cyanococcus), which probably soon divides the cavity
into two locelli. Ovules 3 or 4 in each proper cell, incurved. Fruit
not seen.

This needs to be compared, especially the sterile plant here de-
scribed, with Blume’s JN. laxiflora of New Guinea.

Prate 60, B—NEtirris Vitiensis: a flowering branch of the male
plant, of the natural size. Fig. 1. A flower, with the pedicel, bract,
bractlets, &e. 2. A bract, detached. 3. Calyx, from which the
stamens and petals have fallen. 4. A petal. 5. A stamen.—C. Fig.
6. A flower of the hermaphrodite plant, the petals and stamens
removed. 7. A flower-bud of the same; the ovary transversely
divided. 8. Vertical section of the ovary.—The details more or less
magnified.

10. CAMPOMANESTA, Ruz & Pw.
1. CAMPOMANESIA CERASOIDES.
Psidium cerasoides, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 290.

Has. Brazil, near Rio Janeiro. (In fruit.)
138

550 PHANEROGAMIA.

ll. PSIDIUM, Linn.
1. Psipium Guasava, Linn.

Psidium Guajava, Linn. Spec. ed. 1, p. 470; Raddi, Mem. Pl. Bras. p. 2; Blume,
Mus. Bot. Lugd.-Bat. p. 71.

P. pyriferum & P. pomiferum, Linn. Spec. Pl. ed. 2, et Auct.

Has. Eimeo and Tahiti, Society Islands: “recently introduced,
but naturalized and spreading rapidly.” Sandwich Islands: intro-
duced. The Guaiva was also noticed as largely naturalized around
Rio Janeiro.

2. Psiplum PoLYCARPON, Lamb.

Psidium polycarpon, Lambert, in Trans. Linn. Soc. 11, p. 231, t. 17; DC. Prodr. 3,
p. 235.

Has. Rio Janeiro, Brazil. (Most probably no more than a variety
of the common Guaiva.)

12, SONNERATIA, Linn. f

1. SoNNERATIA AcIDA, Linn. f.

Aubletia caseolaris, Gertn. Fruct. 1, p. 329, t. 78.

Has. Luzon, in the vicinity of Manilla. (In fruit.)

13. BACKHOUSIA, Hook. & Harv.

Backuovusta, Hook. & Harv. in Hook. Bot. Mag. t. 4133.

MYRTACE GZ. 551

1. BACKHOUSIA RIPARIA.

Backhousia myrtifolia & B. riparia, Hook. & Hary. in Hook. Bot. Mag. 1. ¢.
Eugenia riparia, A. Cunn. ined.

Has. Hunter’s River, New South Wales. Also specimens pre-
sented by Mr. Cunningham.

The ripe fruit of this remarkable plant is not yet known. Judging
from the ovary, it is probably dry and indehiscent. Our specimens
manifestly connect the two species. I adopt, accordingly, the trivial
name imposed by Cunningham, the discoverer of the plant,

14. ACICALYPTUS, Nov. Gen.

Calyx subuleformis, acute tetragonus, clausus; apice subulato-rostrato,
operculiformi, sub anthesi circumscisse deciduo; fauce ultra ovarvum
longe producta. Petala 4, in operculum leviter cohcerentia, sub anthest
dejecta. Stamina plurima, discreta, margini calycis tubv inserta:
filamenta filiformia: anthere biloculares, loculis ovalibus. Stylus
filiformis: stigma obtusum. Ovarium biloculare, dissepimento tenur.
Ovula in loculis 8-10, anatropa, subcurvata? (Fructus ignotus, ut
videter carnosus indehiscens.)—Arbor vel arbuscula; foliis oppositis
ovatis penninerviis impunctatis ; floribus cymosis terminalibus.

1. AcicALyprus myRtTorpEs, Sp. Nov. (Tab. 67.)

Has. Feejee Islands: on the mountains of Muthuata, at an eleva-
tion of 2,000 feet.

Apparently a tree (the size not recorded), glabrous throughout; the
branchlets corymbose, terete, rather slender, nodose. Leaves opposite,
approximate, 14 to 2 inches in length, with a petiole of 4 or 5 lines
long, elliptical or ovate, acuminate (the tip rather obtuse), acute at

552 PHANEROGAMIA.

the base, subcoriaceous, dull beneath, deep green and shining above,
not punctate, copiously feather-veined ; the slender but rather conspi-
cuous veins sparingly reticulated, confluent into an undulate intra-
marginal false vein. Cymes terminal, trichotomous, compound,
corymbose, many-flowered, somewhat exceeding the leaves; its divi-
sions slightly angled; the pedicels very short, articulated with the
flower. Bracts and bractlets caducous and not seen. Flower-buds
half an inch long, 12 to 2 lines wide in the middle, thence tapering
moderately to the base, and upwardly narrowed into a sharp and
slender beak, the whole between spindle-shaped and awi-shaped, acutely
Jour-angled, and below the middle marked with 4 intermediate and
less salient angles or nerves, very smooth: the continuous summit,
representing the closed limb of the calyx, separates, a little below the
base of the beak, by an even circumscissile line, and falls as a lid; but
there is no external mark to indicate the separation before anthesis.
Petals 4, small, inserted on the margin of the persistent tube of the
calyx by a broad base, rounded, hooded, lightly cohering into a lid (in
the manner of Syzygium), and falling off together when the calyptra
of the calyx is detached. Stamens very numerous, inserted on the
very edge of the calyx-tube, inflexed in the manner of the family
before anthesis and received into the throat of the calyx, which is as
long as the portion adnate to the ovary: filaments filiform, distinct,
about 3 lines long: anthers two-celled; the cells oval, fixed by the
middle, longitudinally dehiscent. Style filiform, as long as the throat
of the calyx: stigma obtuse. Ovary two-celled, with thick and fleshy
walls, the dissepiment very thin, probably obliterated in the fruit.
Ovules 8 to 10 in each cell, somewhat ascending, anatropous, more or
less curved. Fruit not seen: from the texture of the calyx it is pro-
bably fleshy and indehiscent.

Should this remarkable plant prove to have a fleshy fruit, as is
most likely, it will rank next to Calyptranthes, from which the subu-
late and quadrangular calyx and the operculate corolla sufficiently
distinguish it. If the fruit be capsular, the genus will be distin-
guished from Eucalyptus by the two-celled ovary, and the corolla of
lightly coherent or separable petals, as well as by the foliage, inflo-
rescence, and the form of the calyx. The name, compounded of az,
a point or edge, and xaiuzrds, a covering, indicates the affinities and one
of the characters of the genus.

MYRTACEA, 553

Puate 67.—ACICALYPTUS MYRTOIDES: a flowering branch, of the
natural size. Fig. 1. A flower-bud, enlarged. 2. Vertical section of
the same. 3. Flower-bud, with the lid detaching. 4. Flower, with
the ovary divided transversely.—The details enlarged.

1s. HUCALYPTUS, Ler.

1. Evcatyrrus supunata, A. Cunn.

Eucalyptus subulata, A. Cunn.; Schauer, in Walp. Repert. 2, p. 924.
E. rostrata, Schlecht. in Linnea, 29, p. 655? non Cav.

Has. New South Wales. (Probably given by Mr. Cunningham.)

The abrupt beak of the operculum varies from 14 to 3 lines long.

2. EUCALYPTUS MARGINATA, Smith ?

Eucalyptus marginata, Smith, in Trans. Linn. Soc. 8, p. 802?

Has. Near Sydney, New South Wales. (Imperfect specimens.)

3. HUCALYPTUS ACERVULA, Sieber.

Eucalyptus acervula, Sieber, Pl. Exsic. N. Holl. no. 469; DC. Prodr. 8, p. 217.

Haz. New South Wales; probably in the vicinity of Sydney.

4, KUCALYPTUS VIRGATA, Steber, 1. c.

Has. Hunter’s River, New South Wales.

5. EUCALYPTUS RADIATA, Steber.

Eucalyptus radiata, Sieber, 1. c.; DC. Prodr. 3, p. 218, & Mem. Myrt. t. 7.

Has. Vicinity of Hunter’s River, New South Wales.
139

554 PHANEROGAMIA.

6. EUCALYPTUS BOTRYOIDES, Smith.

Eucalyptus botryoides, Smith, in Trans. Linn. Soc. 3, p. 287.
E. platypodos, Cav. Ic. Pl. 4, p. 23, t. 341.
E. pallens, DC. Prodr. 3, p. 219?

Has. Near Sydney, New South Wales.

7. EUCALYPTUS MINIATA, A. Cunm.
Has. Hunter’s River, New South Wales. (Foliage only.
9 fo) ev

This accords with a fruiting specimen in the Hookerian herbarium,
so named by Allan Cunningham, likewise from Hunter’s River, but it
scarcely agrees with the character given by Schauer, in Walp.
Repert. 2, p. 925.

8. EUCALYPTUS PANICULATA, Smith ?

Eucalyptus paniculata, Smith, in Trans. Linn. Soe. 3, p. 287?

Has. Sydney, New South Wales.

t Species valde dubia.

9, EUCALYPTUS MULTIFLORA, Rich, in Herb.

L.? folus subalternis petiolatis oblongis acuminatis basi acutis cequilateris
concoloribus laxe penninerviis venulosis costa venisque primarus subtus
prominentibus; cymis paniculatis multifloris ; pedunculis compresso-
angulatis ; capsulis subglobosis.

Ifa. Near Caldera, Mindanao, one of the Philippine Islands.

The specimens consist of a leafy shoot, and of a leafless branch

MYRTACE &. 555

with the inflorescence of the previous season, bearing the persistent
capsules. The latter show what appears to be the line of circum-
scissile dehiscence; otherwise there are no evident grounds for refer-
ring the plant to Hucalyptus. The leaves are not phyllodineous, and
apparently not vertical: they are unequally alternate, oblong, acumi-
nate, or at least acute at both ends, 4 or 5 inches long and 14 or 2
inches wide, on petioles of half an inch in length, equilateral, char-
taceous, thickly pellucid-punctate, dull and of the same hue both. sides,
loosely feather-veined, the primary veins and the midrib prominent
underneath, but impressed above; the veinlets minutely reticulated.
Branchlets, especially the fructiferous ones, somewhat angled. The
flowers appear to have been in naked, terminal and axillary, panicu-
late cymes; the peduncles, &c., compressed-angled, many-flowered ; the
pedicels umbellately fascicled in threes and fives, as long as the cap-
sules. The latter are globular, 2 lines in diameter, the summit, above
the line from which the limb of the calyx has fallen, convex; there
four-valved ; within four-celled; each cell containing a large placenta,
which has evidently borne numerous seeds. These, however, have
all been shed.—I thus record the plant, under the name given by Mr.
Rich in the collection, since Blume has published one or two Euca-
lypti from the Molluccas and other Malayan Islands, to which this
plant may be related.

16. ANGOPHORA, Cav.
1. ANGOPHORA CORDIFOLIA, Cav.

Angophora cordifolia, Cav. Ic. Pl. 4, p. 21, t. 388; DC. Prodr. 3, p. 222.

Has. Sydney, New South Wales.

2. ANGOPHORA LANCEOLATA, Cav.
Angophora lanceolata, Cav. Ic. Pl. 4, p. 22, t. 888; DC. le.

Var. @. nispma, A. Cunn.: /foliis lanceolatis sensim acuminatis ; pedun-
culis calycibusque setoso-hispidis.

556 PHANEROGAMIA.
Var. y. ANGUSTIFOLIA : foliis linearibus ; pedunculis calycibusque glabris.

Has. New South Wales, Hunter's River: in fruit. Var. @. Wool-
ongong and near Sydney. Var. y. Newington: in flower.

The specimens of var. a. (except that the leaves are chiefly oppo-
site) exactly accord with the figure of Cavanilles, which also repre-
sents the plant with mature fruit. Those of var. @. with similar
lanceolate-acuminate leaves, have the inflorescence and calyx somewhat
hoary with a fine pubescence, and sparsely setose with reddish, appa-
rently glandular bristles, which are however deciduous when the fruit
is mature. The var. y. differs in its narrower, nearly linear leaves (3
or 4 inches long and 3 to 5 lines wide, acuminate at both ends), and
in having no bristles and scarcely any pubescence on the inflorescence
and calyx ;—in this respect agreeing with the character of A. lanceo-
lata. The species is apparently polymorphous, and includes DeCan-
dolle’s A, intermedia also. A remarkable form of it occurs in a collec-
tion made at Moreton Bay, by Mrs. Mallard, kindly communicated
by my excellent friend, N. B. Ward, Esq., of London. On the same
individual, some of the lower leaves are lanceolate, gradually acumi-
nate, and with a subsessile cordate-auriculate base; the rest are
petioled, most of them lanceolate, but some are elliptical-oblong, as
A, intermedia is characterized, like which the peduncles are more or
less setose-hispid. The venation of the leaves, the flowers, &c., accord
with the present species, not with A. cordifolia.

17. SYNCARPIA, TZenore.
1. SYNCARPIA LAURIFOLIA, Zenore.

Syncarpia laurifolia, Tenore, Ind. Sem. Hort. Neopol. 1839, & in Mem. Soc, Moden,
22 (1840), p. 7, t. 1.

Metrosideros glomeruliflora, Smith, in Trans. Linn. Soc. 8, p. 268; DC. Prodr. 3;
p» 225,

Tristania allens, A. Cunn.; DC. Prodr. 3, p. 210.

Kampizia albens, Nees, Diss. in Act. Acad. Nat. Cur. 18, Suppl. 1, t. 1, 2.

Has. New South Wales, near Sydney and Cook’s River. “'Tur-

pentine-tree” of the Colonists.

MYRTACE ZA. 557

18.5 METROSIDEROS, Rumph., R. Br.

* Novo-Zelandice et Aucklandice.

1. MrrtrosipEeros Lucipa, Menzies.

Metrosideros lucida, Menzies, in Herb. Hook.; Smith, in Rees, Cycl.; A. Rich. FI.
N. Zel. p. 8383; A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 3, p. 114; Hook. f.
Fl. Antare. p. 12, & Fl. N. Zeal. p. 67.

M. umbellata, Cav. Ic. Pl. 4, p. 20, t. 8387; DC. Prodr. 3, p. 225, excl. patria.

Agalmanthus umbellatus, Hombr. & Jacquin. Voy. Astrol. & Zel. Bot. t. 1.

Melaleuca lucida, Forst. Prodr. Fl. Ins. Austr. p. 88.

Has. Lord Auckland Islands: where it is the principal timber tree.

2. METROSIDEROS FLORIDA, Smith.
Metrosideros florida, Smith, in Trans. Linn. Soc. 3, p. 269; DC. l.¢c.; A. Cunn. 1. c.;
Hook. f. Fl. N. Zeal. p. 66, t. 15.
Melaleuca florida, Forst. Prodr. Fl. Ins. Austr. p. 37.
Leptospermum scandens, Forst. Char. Gen. t. 36.

Has. Bay of Islands, New Zealand.

3. METROSIDEROS RoBUSTA, A. Cunn.

Metrosideros robusta, A. Cunn. Bot. N. Zeal. 1. c.; Hook. f. Fl. N. Zeal. p. 68, t. 17.
M. florida, Hook. Bot. Mag. t. 4471, non Smith.

Has. Bay of Islands, New Zealand.

4, METROSIDEROS TOMENTOSA, A. Rich.

Metrosideros tomentosa, A. Rich. Fl. N. Zel. p. 336, t. 37; A. Cunn. 1. ¢.; Hook.
Bot. Mag. t. 4488; Hook. f. 1. ¢.

Has. Bay of Islands, Tippona, &c., New Zealand. (In fruit: the

ordinary, and a glabrate form.)
140

558 PHANEROGAMIA.

5. METROSIDEROS SCANDENS, Banks & Solander.

Metrosideros scandens, Banks & Soland. MS. & Ic. ined.; Geertn. Fruct. 1, p. 172, t.
34, f. 10; Hook. f. Fl. N. Zeal. p. 69.

M. perforata, A. Rich. Fl. N. Zel. p. 334.

M. buxifolia, A. Cunn. 1. c. p. 111; Hook. Bot. Mag. t. 4515.

Melaleuca perforata, Forst. Prodr. Fl. Ins. Austr. 37.

Leptospermum perforatum, Forst. Char. Gen. p. 72.

Has. Bay of Islands, New Zealand.

6. METROSIDEROS HYPERICIFOLIA, A. Cunn.

Metrosideros hypericifolia, A. Cunn. 1. c.; Hook. f. Fl. N. Zeal. p. 68, t. 16.

Has. Bay of Islands, New Zealand.

x «x Polynesicee (Taitensis et Vitiensis).
7. MerTROSIDEROS COLLINA. (Tab. 68.)

Leptospermum collinum, J. R. Forst. Char. Gen. Ins. Austr. p. 72, t. 88, f. mp.
Melaleuca cestuosa, G. Forst. Prodr. Fl. Ins. Austr. p. 38.

Var. a. VILLOSA: ramulis cum inflorescentia foliisque junioribus (ovatis
ovalibus seu ellipticis) tomentoso-villosis.

Melaleuca villosa, Linn. f. Suppl. p. 342.

Metrosideros spectabilis, Geertn. Fruct. 1, p. 172, t. 84, f. 9?

M. villosa, Smith, in Trans. Linn. Soc. 3, p. 268; DO. Prodr. 3, p. 224; Hook. &
Arn. Bot. Beech. Voy. p. 63; Guill. Zeph. Tait. p. 57.

Var. @. GLABERRIMA: ramulis foliis calycibusque glabris; pedunculis
nune puberulis.

Metrosideros villosa, var. glaberrima, Bertero, ex Guill. 1. ¢.

M. diffusa, Hook. & Arn. Bot. Beech. Voy. p. 63? non Smith.

MYRTACES. 559

Var. y. VITIENSIS; ramulis folusque ellipticis oblongisve glabris; in-
Hlorescentia calycibusque plus minus cano-sericeis; floribus nunc sub-

pedicellatis. (Tab. 68.)

Has. Var. a. and @. Tahiti and Eimeo, Society Islands; common
on ridges. y. Feejee Islands; Ovolau, Muthuata, and Sandalwood
Bay; at the elevation of from 100 to 2,000 feet.—A WMetrosideros,
mentioned by Dr. Pickering, as perhaps that of the Feejees but not
that of Tahiti, was noticed on the mountains of Tutuila, one of the
Samoan Islands, at an elevation of 2,000 feet; where it forms “a
spreading tree, 30 feet high, with the trunk a foot in diameter; the
leaves smooth; the flowers scarlet, but not very showy.” There are
no specimens of it in the collection.

The M. villosa of Smith, with the young leaves, branchlets, and espe-
cially the inflorescence canescent with a villous-tomentose down, passes
so completely into glabrate and truly glabrous forms that it becomes
advisable to restore the earliest specific name, that of the elder
Forster. One of the intermediate varieties is doubtless the JL diffusa
of Hooker and Arnott, not of Smith; the latter being a New Zealand
species. The leaves vary from ovate or oval to elliptical, or even ovate-
lanceolate, either rounded at both ends or acutish, or the base occa-
sionally subcordate; the copious pinnated veins manifest on both
sides but slender, minutely reticulated. Petioles 12 to nearly 3 lines
long. Cymes sometimes axillary towards the summit of the branches,
usually terminal and in pairs, as described by Smith, but often soli-
tary or else three together at the summit. lowers sessile or nearly
so, in threes or fives, at the summit of the partial peduncles; which
distinguishes the plant from the Sandwich Island species. Stamens
red. Ovary three-celled, enclosed in the tube of the calyx, to which
its lower half is adnate. The mature capsule projects one-half
beyond the crateriform calyx (the lobes of which are tardily deci-
duous), to which barely its base remains adherent, and not very
firmly.

From the above I am unable specifically to distinguish the speci-
mens from the Feejee Islands; var. y. They mostly have rather
narrower, elliptical or oblong, or even lanceolate-elliptical leaves, more
or less narrowed at both ends, in some specimens, however, as broad
as in the ordinary Tahitian forms, rarely inclining to obovate, all of

560 PHANEROGAMIA.

them entirely glabrous, as are the branchlets. The inflorescence, how-
ever, and the calyx are densely silky-canescent or silky-pubescent ; and
the flowers are slightly pedicellate, especially the central ones of each
cluster. Stamens red. Fruit not seen.

PLATE 68.—METROSIDEROS COLLINA, var. y. VITIENSIS: a branch of
the natural size. Fig. 1. Vertical section of a flower. 2. Flower,
with the petals and stamens detached. 38. Transverse section of the
ovary. 4. Ovules.—The details variously magnified.

x * * Sandwicenses.
8. METROSIDEROS LuTEA, Sp. Nov. (Tab. 69.)

M. foltis ovalr-ellipticis utrinque scepius rotundatis modice petiolatis
tenurter venosis subtus junioribus preesertim canescenti-tomentulosis ;
calycibus cum inflorescentia cano-tomentosis ; floribus subsessilibus vel
brevissime pedicellatis ; petalis staminibusque luteis.

Has. Hawaii, Sandwich Islands; in the vicinity of Hilo.

Apparently a tree of considerable size; the branchlets nearly
terete, the younger ones only hoary with a fine pubescence. Leaves
oval, or broadly elliptical, rarely verging to ovate, rounded at both ends,
often retuse, sometimes slightly subcordate, coriaceous, 1+ to 23 inches
long, glabrous or early glabrate above, hoary with a minute canescent
tomentum underneath, which is very tardily deciduous, closely feather-
veined; the veins slender but perspicuous, reticulated, the basal ones
produced into an intramarginal false vein. Letiole rather conspicuous,
2% to 4 lines long. Cymes small, solitary or in pairs at the apex of
the branches, not exceeding the leaves, very short-peduncled. Bracts
caducous. lowers subsessile or very short-pedicelled (the pedicel less
than a line, or rarely a line and a half long, oftener scarcely any),
usually in threes at the apex of the partial peduncles. Calyx densely
canescent-tomentose, as also the inflorescence, about 2% lines long, cam-
panulate-turbinate, five-lobed; the lobes very obtuse. tals tomen-
tulose externally, “yellow, as well as the stamens,’ apparently pale.
Filaments and style nearly an inch long. Ovary nearly as in I.

MYRTACES. 561

polymorpha; the free summit tomentulose. Mature capsule not
seen.

This appears to be distinguished from all other Hawaiian species
by its yellow and subsessile flowers; in the latter character as well as
in aspect approaching the Tahitian I. collina. Should these charac-
ters prove inconstant, it will add another to the numerous and wide
variations of I. polymorpha.

PLATE 69, B.—METROSIDEROS LUTEA: a flowering branch, of the
natural size. Fig. 1. Flower vertically divided through the ovary.
2. Calyx and pistil, divided transversely, showing the cells of the
ovary. 3. Ovules.—The analyses magnified.

9. Merrosiperos rugosa, Sp. Nov. (Tab. 69.)

M. ramulis quadrangulatis ; foliis orbiculatis brevissime petiolatis bullato-
rugosis penninervis, venis validis in pagina inferiort scepius ferru-
gineo-tomentosa prominentibus ; inflorescentia calycibusque tomentosis ;
Hloribus subsessilibus rubris.

Has. Oahu, Sandwich Islands; on the mountains behind Honolulu.
(Also gathered by Gaudichaud, in the voyage of the Bonite.)

Apparently a small tree or shrub, with quadrangular branchlets ;
only the ultimate ones tomentose. Leaves orbicular, about an inch in
diameter, thick and coriaceous, strikingly bullate-rugose above, where
the strong pinnate veins are impressed, while wnderneath they are very
prominent, as well as the thick costa; the upper surface is minutely
pubescent when young, but soon glabrate; the lower tomentose with a
thick and close, nearly persistent, tawny or ferrugineous wool: petioles
very short, but manifest, a line or a line and a half in length. Cymes
small, solitary or in pairs at the summit of the branches; the pedun-
cles and their divisions short and stout, whitish-tomentose, the whole
subtended by rather conspicuous and coriaceous bud-scales, which are
somewhat persistent after the evolution of the cyme. Bractlets as long
as the calyx, oval, tomentose, somewhat ferrugineous, soon deciduous.

Pedicels very short or scarcely any. Flowers about as large as in M/.
141

562 PHANEROGAMIA.

polymorpha. Calyx very tomentose externally with whitish wool, tur-
binate, five-lobed. Petals and stamens deep red; the former a little
pubescent externally. Ovary deeply immersed in the bottom of the
calyx, three-celled, its summit only free. Fruit unknown.

This is surely distinct from JZ polymorpha, however polymorphous
that may be. The stout veins of the leaves, so salient underneath,
the bullate upper surface, and the rusty or ochre-coloured wool
of the lower surface, along with the subsessile flowers, at once dis-
tinguish it. Few specimens were gathered. Those of Gaudichaud’'s
collection are, I believe, destitute both of flowers and fruit.

PLATE 69, A.—METROSIDEROS RUGOSA: a branch, in flower, of the
natural size. Fig. 1. Vertical section of a flower, enlarged. 2.
Ovules, magnified.—The analyses magnified. °

10. METROSIDEROS POLYMORPHA, Gaud.

Metrosideros polymorpha, Gaud. Bot. Voy. Freyc. p. 99, & 482, t. 108, 109; Hook.
& Arn. Bot. Beech. Voy. p. 82; Schauer, in Rel. Meyen. p. 331.

Var. a. foliis subrotundo-ellipticis vel orbiculatis basi cordatis brevissime
petiolatis tomento tenut canescentibus, ramulorum confertissimis vel
umbricatis scepe sessilibus ; cymis thyrsoideis multifloris calycibusque
(parvults) means.

Var. @. foltis crasso-coriaceis orbiculatis seu rotundo-ellipticis basi cor-
datis vel retusis breviter petiolatis supra mox glabris subtus cum
inflorescentia calycibusque (majusculis) tomento crasso cano vel fulvo
lanosissimis.— Ludit, foliis demum glabratis.

Var. y. folits crasso-coriaceis subrotundis vel ovato-ellipticis basi subcor-
datis retusisve breviter petiolatis utringue cum inflorescentia calycibus-
que (parvulis) glabris.

Var. 6. jfoliis coriaceis glabris modice petiolatis modo ovalibus modo
anguste oblongis basi acutis vel obtusis; cymis laxijloris calycibusque
(majusculis) tomento denso villoso incanis.

MYRTACES. 563

Var. «. foliis ovalibus lato-ellipticis oblongisve basi rotundatis obtusis vel
acutiusculis modice petiolatis glabris ; inflorescentia calycibusque pube
tenui canescenti-puberulis demum glabratis.

Var. ¢. foliis tenuiter coriaceis anguste oblongis seu lanceolatis acutius-
. . ’ . s y . . . .
culis basi in petiolum breviusculum vel longiusculum attenuatis glabris ;
cymis multifloris calycibusque (parvulis) glabris vel puberulis,

Has. Sandwich Islands; abundant; gathered by all collectors from
Nelson, in Cook’s third voyage, and Menzies, to the present time.
Var. a. Oahu, in the mountains behind Honolulu. . Hawaii; both
near the coast and on Mouna Loa, &c., to the elevation of 8,000 or
9,000 feet. Maui, at the crater of Haleakala. Mountains of Kauai:
some forms connecting this with succeeding varieties. y. Mouna
Loa, Hawaii, at an elevation of 8,000 feet. 8. Hawaii, on Mouna
Kea, and in the district of Puna. A form allied to it from the
mountains of Kauai. ¢. Oahu, in the mountains behind Honolulu
and Waianae. Hawaii, near the coast, in the districts of Waimea and
Puna. ¢. Oahu, in the mountains behind Honolulu, &c.

Although furnished with a fuller suite of specimens, probably, than
ever have been brought together before, I am equally unable with
preceding botanists to distinguish any of these extremely varied forms
into separate species. The first and the last, as different as two
species could well be conceived to be, are yet connected by a series of
intermediate states, of which the more prominent are characterized
above as varieties. Var. a. one extreme form, is remarkable for its
small, rownded-cordate and subsessile, or very short-petioled leaves,
crowded or even imbricated on the branchlets, from an inch to half an
inch in length, finely canescent-tomentose or cinereous underneath, early
glabrate above, coriaceous, but seldom very thick: the cymes incline
to be thyrsoid and prolonged; the flowers rather small; the calyx and
peduncles densely white-tomentose. This passes into var. @. (which
includes Gaudichaud’s tab. 108), with very thick, rounded leaves (an
inch to an inch and a half long), mostly heart-shaped at the base, and
more distinctly petioled ; the upper surface at first pubescent, but soon
glabrous; the lower very densely woolly-tomentose, as well as the
crowded cymes and larger calyxes. The wool inclines to rub off from
the older leaves. With this is connected an early glabrate state,

564 PHANEROGAMIA.

passing into others with oval or elliptical leaves (sometimes 2 inches
long) from Kauai. Var. y. is a glabrous, or perhaps early glabrate
state of 3.; the specimens with young fruit only. Var. 6. has thinner
and glabrous leaves, either oval or oblong, often acute at the base, rather
conspicuously petioled ; but the loose 2 CYMES, calyx, &e., are very densely
villous-tomentose. Var. e. embraces a variety of specimens, with
glabrous and distinctly jpetioled leaves varying from broadly oval to
narrowly oblong, either thick or thinnish; the inflorescence and calyx
minutely canescent or pubescent, but early glabrate. This passes
insensibly into var. ¢. the smoothest and most slender form, with the
narrowest and thinnest leaves (narrowly oblong and lanceolate, acutish,
with the base tapering into a petiole of about 3 lines long); the cymes
are loose, and the flowers as small as in var. a.—All have red and
pedicelled flowers; the pedicels 12 to 2 lines long. The veins of the
leaves are slender, more or less reticulated, sometimes with a pair or
two of stronger ascending ones from the base. Rarely the limb of
the calyx is six-lobed; the lobes very obtuse. The upper half of the
three-celled ovary is free from the tube of the calyx. The globular
capsule, which projects somewhat beyond the calyx, becomes nearly
free from the tube which encircles it.—This species, in some of its
forms at least, becomes a large tree, the Ohea of the Hawaians.

In the collection from Tahiti, there is a single imperfect specimen
of what appears to be the J. polymorpha, var. ., having pedicelled
flowers and lanceolate leaves, glabrous throughout. Whether truly
of this species must be left for future investigators to determine.

11. Metrosmperos Macropus, Hook. Gd Arn. (Tab. 70.)

M. glabra; foliis longe petiolatis ; petiolo lamina ovata seu ovato-oblonga
paullo vel dimidio breviore; cyma confertijlora; floribus (rubellis ?)
pedicellatis; bracteis bracteolisque ovatis et ovato-lanceolatis magnis
involucrantibus deciduis ; capsula calycis tubo fere inclusa ad medium
usque libera.

Metrosideros Macropus, Hook. & Arn. Bot. Beech. Voy. p. 83.

Haz. Oahu, Sandwich Islands; on the mountains behind Honolulu.
(Also gathered by Macrae, Lay & Collie, Gaudichaud, and Seemann.)

MYRTACE &. 565

A tree, probably of considerable size, glabrous throughout; the
branchlets somewhat angled. Leaves ovate or ovate-oblong, sometimes
rhomboid-ovate, coriaceous, rather dull, acutish, either rounded, obtuse,
or acute at the base, obscurely punctate, copiously feather-veined ;
the veins oblique, very slender but rather conspicuous, equally so on
both sides, sparsely branched, connected by minute reticulations,
terminating in an inframarginal vein: petioles elongated, sometimes
almost as long as the blade, commonly half or one-third its length.
Cymes terminal, usually geminate, subsessile, many-flowered, crowded,
evolved from a large scaly bud, the scales of which remain persistent
for some time, as ovate or oblong pointed bracts, of a coriaceo-char-
taceous texture, half an inch long; the pedicels (13 to 2, or in fruit 3
lines long) subtended by similar but smaller ovate-lanceolate bractlets,
which are early deciduous, as in the other species. Flowers rather
larger than in the largest-flowered forms of I. polymorpha; the calyx
nearly similar, glabrous or barely puberulent. Petals and stamens
apparently pale or flesh-colour. Ovary three-celled, the glabrous
summit free barely to the middle. Capsule nearly included in the
turbinate tube of the calyx (of which the spreading lobes are persis-
tent), free to the middle, adnate below, three-valved, many-seeded.
Seeds fusiform-subulate, not much pointed.

The characters of this species would appear to be amply sufficient :
but forms may occur between it and I. polymorpha var. «. The
flowering specimen figured is from the collection of Mr. Seemann.

Puate 70, B—MeEtrosiperos Macropus: branches in flower and
fruit. Fig. 1. Vertical section of a flower. 2. Ovary, transversely
divided. 3. Capsule, dehiscent in the calyx. 4. Seeds. 5. Embryo.
—The details magnified.

19. KUNZEA, Reichenbd.

1. Kunzea corirouia, Reichenb.

Kunzea corifolia, Reichenb. Consp. Reg. Veg. p. 175; Schauer, in Pl. Preiss. 1, p. 124.
Metrosideros corifolia, Vent. Hort. Malm. t. 46; DC. Prodr. 3, p. 225.
142

566 PHANEROGAMIA.

Has. In the neighbourhood of Sydney, New South Wales.

2. KunzEa ScHAUERI, Lehm.

Kunzea Schaueri, Lehm. Pl. Preiss. 1, p. 124.
Metrosideros capitata, Smith, in Trans. Linn. Soe. 3, p. 273; DC. 1.c.
Melaleuca eriocephala, Sieber, Pl. N. Holl. Exsic. no. 322.

Has. New South Wales; with the foregoing species.

20. CALLISTEMON, R. Br.
1. CALLISTEMON PINIFOLIUM, DC.
Callistemon pinifolium, DC. Prodr. 8, p. 223; Hook. Bot. Mag. t. 3980.

Has. Hunter’s River, New South Wales. (In fruit.)

2. CALLISTEMON LANCEOLATUM, DC. 1. c.

Has. New South Wales; probably from near Sydney.

21. MELALEUCA, Linn.
1. MELALEUCA STYPHELIOIDES, Smith.
Melaleuca styphelioides, Smith, in Trans. Linn. Soc. 3, p. 275; DC. Prodr. 8, p. 212.

Has. New South Wales: probably from the vicinity of Sydney.

2. MELALEUCA FERREA, A. Cunn., ined.

Has. New South Wales; the locality not recorded. Probably

MYRTACEZ. 567
given, as were some other specimens, by A. Cunningham, as it

accords with specimens of his M ferrea, “the Iron-wood of Moreton
Bay,” in the Hookerian Herbarium.

The species resembles I. gemistifolia, Smith; but the leaves are
only one-nerved and obscurely punctate.
3. MeLateuca Frasert, Hook. ?
Melaleuca Fraseri, Hook. Bot. Mag. t. 3210?

Has. Hunter’s River, New South Wales.

4, MELALEUCA Noposa, Smith, l. c.

Has. Hunter's River, New South Wales.

5. MELALEUCA ERUBESCENS, Otto ?

Melaleuca erubescens, Otto, ex DC. Prodr. 3, p. 214?

Haz. New South Wales; where it has also been gathered by
Fraser and Cunningham. Probably a variety of M. erubescens, with
more filiform and coarsely dotted leaves.

6. MELALEUCA THYMIFOLIA, Smith.

Melaleuca thymifolia, Smith, Exot. Bot. 1, t. 836; DC. Prodr. 3, p. 214.

Has. Hunter's River and near Sydney, New South Wales.

7. MELALEUCA LINARIFOLIA, Simith, l. c.

Has. New South Wales. (The specimens gathered by A. Cun-
ningham.)

568 PHANEROGAMIA.

8. MELALEUCA SQUARROSA, Smith, 1. c.

Melaleuca squarrosa, Smith, in Trans. Linn. Soc. 6, p. 300; Labill. Pl. N. Holl. 2,
t. 169; DC. l.c.
M. myrtifolia, Vent. Hort. Malm. t. 47.

Has. In the vicinity of Sydney, New South Wales.

22. TRISTANIA, R&R. Br.

1. TRISTANIA NERIIFOLIA, R. Br.
Tristania nertifolia, R. Br. in Ait. Hort. Kew. 4, p. 417; DC. Prodr. 3, p. 210.
Has. Newington, New South Wales.

The leaves are sometimes alternate, and the peduncles three-five-
flowered.

2. TRISTANIA LAURINA, R. Br. 1. c.

Tristania laurina, R. Br. 1. c.; DC. Prodr. 3, p. 210.

Has. Newington, New South Wales.

23. FABRICIA, Gertn.
1. FABRICIA MYRTIFOLIA, Geertn.
Fabricia myrtifolia, Geertn. Fruct. 1, p. 175, t. 85; Sims, Bot. Mag. t. 1304.

Has. Hunter's River, New South Wales. (In fruit.)

MYRTACE SZ. 569

24. LEPTOSPERMUM, Forse.
1. LEPTOSPERMUM ERICOIDES, A. Rich.
Leptospermum ericoides, A. Rich. Fl. N. Zel. p. 388; Hook. f. Fl. N. Zeal. p. 70.

Has. Bay of Islands, New Zealand.

2. LEPTOSPERMUM SCOPARIUM, Porst.

Leptospermum scoparium, Forst. Char. Gen. p. 72, t. 836; Smith, in Trans. Linn.
Soc. 8, p. 262; DC. Prodr. 3, p. 227; A. Rich. l.c.; Hook. f. lc.

Has. Bay of Islands, New Zealand. Near Sydney, New South
Wales. (Several varieties.)

3. LEPTOSPERMUM LANIGERUM, Ait.

Leptospermum lanigerum, Ait. Hort. Kew. 2, p. 156; Lodd. Bot. Cab. t. 1192; DC.
Prodr. 3, p. 227.

Has. Sydney, New South Wales. (A form with glabrate leaves.)

4, LEPTOSPERMUM MYRTIFOLIUM, Sieber.
Leptospermum myrtifolium, Sieber, Pl. N. Holl. Exsic. no. 314; DC. lc. p. 228.

Has. Near Sydney, New South Wales.

5. LEPTOSPERMUM FLAVESCENS, Smith ?
Leptospermum flavescens, Smith, in Trans. Linn. Soe. 3, p. 262? DC. 1.c.?

Has. Hunter’s River, New South Wales. (An imperfect fragment.)
143

570 PHANEROGAMIA.

6. LEPTOSPERMUM JUNIPERINUM, Smith, 1. c.

Has. Vicinity of Sydney and Hunter's River, New South Wales.

7. LEPTOSPERMUM ARACHNOIDEUM, Smith. 1. c.

Has. New South Wales; in the vicinity of Sydney.

25. BASCKIA, Linn.
1. Ba&cKIA LINIFOLIA, Rudge.
Beckia linifolia, Rudge, in Trans. Linn. Soc. 8, p. 297, t. 12; DC. Prodr. 3, p. 229.

Has. Sydney, New South Wales.

2. B&CKIA LEPTOCAULIS, Hook.

Beckia leptocaulis, Hook. Ic. Pl. t. 298.
Has. New South Wales: probably near Sydney.
This is intermediate in character between Beckia and Schidio-

myrtus, which, as well as Harmogia and Euryomyrtus, will probably
be restored to Beckia.

26. SCHIDIOMYRTUS. Schauer.

1. SCHIDIOMYRTUS DIOSMAFOLIA, Schauer.

Schidiomyrtus diosmezfolia, Schauer, in Linnza, 17, p. 237, & Walp. Repert. 2, p. 920.
Beckia diosmifolia, Rudge, in Trans. Linn. Soc. 8, p. 298, t. 18; DOC. lc.

MYRTACE AZ, 5

Has. New South Wales; in the neighbourhood of Sydney.

2. SCHIDIOMYRTUS CRENULATA, Schauer, 1. ¢.

Beckia crenulata, DC. Prodr. 3, p. 230.
Jungia imbricata, Geertn. Fruct. 1, p. 175, t. 35.

Has. In the neighbourhood of Sydney, New South Wales.

27. HURYOMYRTUS, Schauer.

1. Evryomyrtus pirrusa, Schauer, 1. c.

Beckia diffusa, Sieber, in DC. Prodr. 3, p. 230.
B. afinis & B. prostrata, Hook. f. Ic. Pl. t. 284, ex Schauer.

Has. Hunter’s River, New South Wales.

28. HARMOGIA, Schauer.
1. HARMOGIA DENSIFOLIA, Schauer.

Harmogia densifolia, Schauer, in Linnea, 17, p. 238; Walp. Repert. 5, p. 735.
Beckia densifolia, Smith, in Trans. Linn. Soc. 3, p, 260; DC. l.c.

Has. Newington, New South Wales.

29. DARWINIA, Rudge.

1. DARWINIA FASCICULARIS, Rudge.

Darwinia fascicularis, Rudge, in Trans. Linn. Soc. 11, p. 299, t. 22; Schauer, Myrt.

Xerocarp. in Act. Nat. Cur. 19, Suppl. 2, p. 188, t. 2.

Has. Sydney, New South Wales,

572 PHANEROGAMIA.

2. DARWINIA TAXIFOLIA, A. Cunn.

Darwinia taxifolia, A. Cunn. in Field. Mem. N. 8. Wales, p. 152.
D. laxifolia, Schauer, Myrt..Xerocarp. 1. c. p. 190.

Has. Near Sydney, New South Wales.

The name is written laxifolia by Schauer, who opines that this was
the word intended, and that it became taxifolia by a typographical
error.

30. CALYCOTHRIX (male Caryrrix), Labill.

1. CALYCOTHRIX SCABRA, DC.

Calythrix scabra, DC. Prodr. 3, p. 208, & Mem. Myrt. t. 1.
Calycothrix scabra, Schauer, Myrt. Xerocarp. 1. ¢. p. 248.

Has. Near Sydney, New South Wales.

2. CALYCOTHRIX GLABRA, A. Br.

Calythrix glabra, R. Br. in Bot. Reg. t. 409; DC. 1. ¢.
Calytria tetragona, Labill. Pl. N. Holl. 2, p. 8, t. 146 (mala), fide Schauer.
Calycothrix Billardiert, Schauer, Myrt. Xerocarp. 1. c.

Has. Near Sydney, New South Wales.

Schauer drops the original trivial name, on account of its inappro-
priateness: but, in that case, the name imposed by Brown and
adopted by DeCandolle should be continued.

Orv. MELASTOMACE &.

Susorpn. lL MEMECYLE &.

1 MEMECYLON, Linn.
1. Memecyton VITIENSE, Sp. Nov.

M. ramulis subteretibus ; foliis uninerviis aveniis oblongis obtusis scepiusve
in acumen obtusum vel retusum productis basi in petiolum brevius-
culum attenuatis siccatis subtus flavidis; pedunculis solitariis brevis-
simis plurifloris; pedicellis nudis fructu globoso seu ovoideo levi
dimidio brevioribus.

Var.? @. foliis latioribus nunc ovalibus apice rotundatis subtus haud
Jlavescentibus.

Has. Feejee Islands: Muthuata and Ovolau, at the elevation of
1,000 to 1,500 feet.

The specimens are all in fruit only: they are said to belong to a
shrub, of 8 or 10 feet in height. Branchlets nearly terete. Leaves
oblong, 14 to 2% inches long, from 7 lines to an inch wide, usually
tapering more or less to both ends, the apex into a short and obtuse or
retuse acumination, the base acutely narrowed into a petiole of 2 or 3
lines in length, sometimes the apex is directly obtuse; the texture
coriaceous; the midrib prominent; the veins obsolete, or a few trans-
verse ones barely perceptible; the colour above fuscous, beneath yel-
lowish.. Peduncles solitary in the axils of the leaves, rarely in pairs,
very short (a line or two, rarely 3 lines long), nearly terete, evidently

several-flowered; the pedicels ebracteolate, 14 or 2 lines long in fruit, in
144

574 PHANEROGAMIA.

a simple fascicle or cymule. Fruit globose, or in the narrowest-leaved
specimens ovoid (but these probably less mature), not at all depressed,
apparently dry, smooth and even, 4 lines in diameter; the minute limb
of the calyx apparently truncate. Cotyledons apparently foliaceous
and much contortuplicate.

The var.? 3. may not improbably belong to another species, which
cannot be characterized without better materials. The leaves are
larger and broader, sometimes oval and rounded at the apex, mostly
oblong, not drying yellowish underneath, or scarcely so. They closely
resemble those of J. edule, of which it may, perhaps, prove a variety;
but the more simple and shorter fruiting inflorescence occurs in the
axils of the leaves of the season.

Several species of Memecylon from Cuming’s collection were charac-
terized by Presl, in his Epimeliz Botanicse, which bears the date of
1849, but seems not to have been published until much later. This
publication appears to have been known to Blume, who, late in 1851,
published characters of many additional species, in his Museum
Botanicum Lugduno-Batavum. In the Annales des Sciences Natu-
relles for 1852 (in a part not published, however, until 1853), Naudin
has elaborated the species contained in the herbarium of the Paris
Museum, without being aware either of Presl’s or of Blume’s publica-
tions. There is a great confusion of names in consequence.

2. Mrmrcyron CALpERENSE, Sp. Nov. (Tab. 71.)

M. ramulis teretiusculis; foliis brevissime petiolatis oblongo-lanceolatis
sensim acuminatis bast rotundatis vel obtusis penninerviis, venis infra
marginem arcuato-conjunctis; pedunculis plerumque binis vel ternis
e nodis ramorum jam foliis denudatorum ortis subangulatis 3-5-radi-
atis umbellato-multifloris; pedicellis flore eequilongis; calycis limbo
truncato wtegerrimo, fauce lamellis 8 septulata.

Has. Near Caldera, Mindanao; one of the Philippine Islands.
Branches slender; the branchlets nearly terete, or obscurely angled.

Leaves very short- petidled (the petioles only a line and a half in
length), oblong-lanceolate, gradually acuminate into a slender point, 4

MELASTOMACES. 575

to 6 inches long, 14 to 22 inches wide, rounded, or often somewhat
contracted, but obtuse at the base, membranaceo-chartaceous in texture,
dull, a little paler underneath, pinnately-veined from a rather promi-
nent midrib, the considerably manifest veins arcuately confluent within
the margin into a sinuous false vein. Inflorescence in the specimens
all below the foliage of the season, from nodes the leaves of which
have fallen. Peduncles mostly 2 or 3 together, somewhat angular, half
an inch long, or rarely shorter, dividing into 3 to 5 rays, each bearing
a many-flowered umbel, or the stronger ones proliferous. Pedicels
about the length of the flower-bud, half a line long, after anthesis
becoming a line or more in length, not bracteolate. Calyx turbinate,
with a truncate and entire border, within furnished with 8 very salient
radiating lamelle, forming as many deep cells in which the inflexed
anthers lie before anthesis, and with as many intermediate, slightly
salient ribs, to the summit of which the stamens are attached. Corolla
small, perhaps caducous before expansion; the petals 4, very broad.
Anthers oblong, on short filaments, the cells nearly straight and
parallel, the connective produced below into a thickish, but flat,
oblong-lanceolate, acutish appendage, which exceeds the cells in length,
and in the bud is superior, almost equalling the unexpanded petals.
Style filiform: stigma terminal, minute. Ovary one-celled, with a
slight trace of several parietal projections. Ovules 12, sometimes 8,
oblique, scarcely reniform, on stout funiculi. Fruit not seen.

I cannot identify this with any of the numerous Malayan species
which have recently been published. It appears to be allied to J.
paniculatum of Jack, and MZ. acuminatissimum of Blume. Particularly
conspicuous in this species are the radiating lamellee within the calyx,
well indicated by Loureiro in the character of Scutula, but not men-
tioned by DeCandolle, who seems in other respects to have misappre-
hended Loureiro’s description.

Piate 71.—Mermecyton CALDERENSE: a branch, in flower, of the
natural size. Fig. 1. A flower-bud. 2. Vertical section of the same.
3. A petal from the bud. 4, 5, 6. Anterior, posterior, and lateral
views of a stamen, in the inverted position it occupies in the bud.
7. A flower, after the corolla and stamens have fallen. 8. Vertical
section of the same. 9. An ovule from the same. 10. Transverse
section of an ovary and its 12 ovules——The details magnified.

576 PHANEROGAMIA.

Susorn Il MELASTOME &.

(MELASTOMES, AsTRONIED, & KiBEssie&, Naudin.*)
2, ASTRONIA, Blume.

AsTrRonIA, Blume, Bijdr. p. 1080, Rumphia, 1, p. 20, t. 6, 7, & Mus. Bot. Lugd.
p- 9; Naudin, in Ann. Sci. Nat. ser. 3, 18, p. 257.

1, ASTRONIA FRATERNA, Sp. Nov. (Tab. 72.)

A, glaberrima; foliis longe petiolatis oblongis acuminatis basi acutis
manifeste triplinerviis (proetermisso utroque nervo tenui submargi-
nalt); eymis corymbosis ; pedicellis gracilibus; calycis (fructiferi)
limbo persistente aperto quinquedentato ; ovario quinqueloculari ; pla-
centis ex uma columella ortis.

Has. Society Islands: on mountain ridges of Tahiti and Eimeo.

Except the Melustoma Tuitense, this is the only Melastomaceous
plant in the collection made at the Society Islands, or mentioned in
Dr. Pickering’s manuscript notes. It was naturally taken for the
Melastoma glabra of Forster (the Astronia Forsteri, Naudin), with
which indeed, if truly distinct, it must have been confounded. That
species, however (of which I have seen original specimens in the
British Museum, and possess recent ones gathered by Mr. Bidwill),
has a calyx the limb of which is perfectly closed, and separates in
anthesis by a transverse circumscission, in the form of a calyptra.
In all the present specimens (which unfortunately are only fructi-
ferous ones) the whole limb of the calyx is persistent, open, and rather
“eeply divided into 5 (or rarely 6), pretty regular, broad and rounded

* M. Naudin’s primary divisions of the order are neatly characterized ; but the first
three are too closely related, and connected by too many known eresaaitiona to entitle
them to the rank of mibordets

MELASTOMACEA, 577

teeth or lobes. Whether this distinction is a constant one further
observation must determine. I should scarcely venture to rely upon
it, were it not strengthened by a difference in the venation of the
leaves; which, in A. Forsteri are rather obscurely three-nerved, the
lateral nerves near the margin and inconspicuous; while ours are
manifestly triple-ribbed from just above the base, the lateral ribs
almost as strong as the middle one, and at some distance within the
margin, which moreover is provided with a slender and inconspicuous
nerve. The shrub is said to be about 5 feet high. The leaves have
a yellowish tinge when dry, especially underneath; they are oblong,
abruptly acuminate, from 12 to 33 inches long, more or less acute at
the base, on slender petioles of 8 to 12 lines long, glabrous, as is the
whole plant, minutely dotted underneath, apparently of a chartaceous
texture. Cymes corymbose, terminal, many-flowered; the bracts all
deciduous. Fructiferous pedicels slender, about 3 lines long. Tube of
the fructiferous calyx globular, and crowned with the conspicuous
limb, together 22 lines long, smooth. Ovary five-celled. Placentee
inserted at the very base of the columella, oblong, scarcely dilated
upwards, ovuliferous throughout. Style, &., not seen. Capsule, as
in other species, not properly dehiscent, but the dry epicarp decays or
falls away in fragments, and the thin endocarp splits into numerous
pieces, leaving a frame of 10 or more, simple or forked, at length
stellately spreading nerves, surrounding the placente and columella.

Puate 72, A.—ASTRONIA FRATERNA: & branchlet. Fig. 1. Fructi-
ferous calyx, enlarged. 2. Vertical section of the same.

2. AstrontA PickErinen, Sp. Nov. (Tab. 72.)

A, glabra; foliis longe petiolatis ovalibus oblongisve utrinque subacutis
nunc basi obtusis 3-5-nerviis; cyma corymbosa composita laxiflora ;
pedicellis gracilibus; alabastris subglobosis; calycis limbo clauso
demum aperto ultra ovarium longiuscule producto, margine in den-
tibus brevissimis 9-20 irregulariter jfisso; antherarum connectivo
deorsum gibberoso; ovario quinqueloculari; placentis e fundo loeu-
lorum ortis. |

Var. a. SAmoEnsis: foliis siccatis viridulis, inflorescentia ramulisque
novellis ferrugineo-puberulis.
145

578 PHANEROGAMIA

Var. @. VitiENSIS: foliis in sicco flavescentibus ; inflorescentia glabriore ;
pedicellis paullo brevioribus.

Has. Samoan or Navigators’ Islands (var. «.); in the mountains of
Tutuila. (§. Feejee Islands; on Ovolau, at an elevation of about
1,000 feet.

As well as can be judged from the incomplete specimens, these two
forms appear to belong to the same species. The Samoan plant has
the leaves greenish when dry, without any trace of yellow; and the
inflorescence shows a minute and scurfy, ferrugineous pubescence,
which is soon deciduous; its pedicels are 2 or 3 lines long. The
Feejee specimen has a yellowish hue in the foliage; and the inflores-
cence (in fruit) shows scarcely a trace of the ferrugineous scurf.
Otherwise no difference is discernible: but we have no flowers or buds
of the latter, to compare with the unexpanded flower-buds of the
former. The floral characters given, therefore, belong only to the
Samoan plant.—A shrub or small tree, the var. @. said to be 20 feet
high; the branches terete and nodose. Leaves glabrous, oval or oblong,
acutish at both ends, or the base often obtuse, or even rounded, char-
taceous in texture, dull, 4 or 5 inches long, 14 to 24 wide, three-ribbed
from the base, besides an intramarginal nerve on each side, which is
often so conspicuous as to make the leaves /ive-ribbed ; the ribs con-
nected by rather prominent, transverse veins, from which proceed less
evident reticulated veinlets. Petioles elongated (from 12 to 18 lines
long), slender. Cyme compound, corymbose, loosely many-flowered,
terminal, trichotomous; its branches angular, nodose. Bracts cadu-
cous. Ledicels slender. Flower-buds globular, or obovoid-globose ;
those by nomeans full grown 2 lines in diameter, closed, the fleshy
calyx showing no trace of lobes. But in specimens past anthesis the
calyx is found to be open, and campanulate in form, produced con-
spicuously beyond the ovary, with the truncate margin irregularly
cleft into 9 to 20 short teeth. Petals 5, short and broad in the bud,
convolute in estivation, said to be “ white, and as long as the calyx,”
in Dr. Pickering’s notes. Stamens 10; the filaments very short
in the undeveloped flower-buds: anthers dolabriform, notched at the
base, where the salient connective is produced into a strong gibbosity.
Style in the bud columnar, as long as the calyx, terminated with the
depressed-capitate and entire stigma of the genus: after anthesis it
becomes much exserted, filiform, and half an inch long. Ovary jfive-

MELASTOMACE S. 579

celled; the cells at first depressed ; the placente borne at the very base
of the cells at a little distance from the axis, directed obliquely
upwards and outwards, flattened, dilated upwards, notched at the
summit, covered with innumerable linear-oblong ovules, which are
marked with a conspicuous brown rhaphe. Seeds not seen, nor
mature fruit; the immature fruit 3 lines in diameter. But there are
vestiges of old fruit, like that of the foregoing species; showing that
the pod, at first fleshy, at length dry, is apparently irregularly rup-
tured; the epicarp at length falling away, and leaving a fibrous
framework of numerous persistent nerves, within which is the. thin
lacerated lining of the cells, and the persistent thickened placente.

Puiate 72, B.—AsrroniA PIcKERINGII: a branch of the var. 3. past
flowering, of the natural size. Fig. 1. Vertical section of a flower-
bud of var. a. 2, 3. Stamens from the same. 4. Calyx and style of
var. a., after flowering. 5. Vertical section of the same. 6. Trans-
verse section of the ovary. 7. A placenta, detached. 8. Ovules.—
The details enlarged.

3. ASTRONIA CONFERTIFLORA, Sp. Nov.

A, foliis longe petiolatis ovalibus obtusis basi rotundatis 3—5-nerviis, costis
subtus ramulisque junioribus cum cyma composita corymbosa conferti-
flora rufo-pubescentibus ; floribus in ramulis ultimis capitato-congestis
subsessilibus; calyce glandulis rufis consperso, limbo ultra ovarium
breviter producto, margine irregulariter 8-10-crenato; ovario tri-
quadriloculart; placentis e fundo loculorum ortis.

Has. Feejee Islands: in the Sandal-wood district of Vanua levu ;
on the banks of streams.

We have this in fruit only. If rightly identified with a plant
recorded in Dr. Pickering’s notes, it is a shrub, 8 feet high. Branch-
lets stout, terete, nodose, glabrate; the ultimate ones somewhat quad-
rangular, and when young clothed with a thick and scurfy reddish-
brown pubescence, which soon disappears. Leaves oval, obtuse at both
ends, or more rounded at the base, 4 to 6 inches long, 22 to 33 wide,
chartaceous, rather bright green, glabrous, except the three principal
ribs, which when young are ferrugineous-pubescent underneath ; these

580 PHANEROGAMIA.

spring from the base of the blade, or rarely from just within it; there
are a pair of more slender, somewhat wavy, intramarginal nerves,
which if counted make the leaf five-ribbed: the petioles about an
inch and a half long. Oyme terminal, ample, corymbose, compound,
repeatedly trichotomous; the divisions stout, somewhat quadrangular,
Serrugineous-pubescent ; the ultimate peduncles bearing numerous, capt-
tate-crowded, sessile or nearly sessile flowers, forming together a very
compact inflorescence. Bracts caducous. Flower-buds, petals, sta-
mens and style not seen; the specimen being wholly in the fruiting
state. Fruiting calyx globular, rather depressed, and open at the top,
little produced beyond the flat summit of the ovary, the truncate margin
minutely and irregularly crenate-dentate into 8 or 10 rounded teeth ;
the surface sprinkled with reddish-brown, glandular dots. Capsule
depressed-globose, about 2 lines in diameter, three- or four-celled, with
a dilated clavate-semilunar strictly basilar placenta in each cell; the
thin epicarp breaking away irregularly, the thin endocarp splitting
longitudinally into 6 or 8 valves and falling away, leaving 6 or 8
strong persistent nerves, surrounding the dilated placentae. Seeds
innumerable, thickly covering the placente, subclavate, or dolabri-
form, angled by mutual pressure, with a brown lateral rhaphe.

Compared with an authentic but imperfect specimen of A. macro-
phylla, Blume, and with Naudin’s description, this species is well
marked by its very obtuse leaves, its dense and corymbose cyme,
with sessile and congested flowers, the many-toothed edge of the
calyx, the 3—4-celled ovary, and the thickish, angled (not acicular)
seeds. ‘The leaves, moreover, are not distinctly triplinerved, as in
that species.

4, Astronia? suBcoRDATA, Sp. Nov.
A.? foliis longe petiolatis ovalibus vel subovatis breviter acuminatis
obtusisve basi cordatis glubratis triplinerviis ; petiolis ramulisque

junioribus cum cyma corymbosa ferrugineo-hirtis demum glabratis.

Has. Upolu, one of the Samoan or Navigators’ Islands: in the
mountains near Apia.

The specimens, apparently of a tree or large shrub, have shed the

MELASTOMACE 4S. 581

flowers and fruit from the inflorescence; but the plant is probably a
congener of the foregoing species. The branches are very stout,
terete, nodose; the younger parts hirsute with a dense and coarse
ferrugineous pubescence, as are the elongated petioles (2 or 3 inches
long); and the nascent foliage is ferrugineous with a similar, but
shorter pubescence. Leaves oval, sometimes inclining to ovate, glabrate,
chartaceous, from 5 to 8 inches long and 3 to 5 inches broad, either
obtuse or slightly acuminate, cordate at the base, but not strongly so,
triple-ribbed from near the base, and with one or two pairs of sinuous
intramarginal nerves, transversely veined, and with evident coarsely
reticulated veinlets. Cyme corymbose, trichotomous, compound, scarcely
exceeding the petioles; the ramifications ferrugineous-hirsute, angled,
evidently many-flowered.

Some rufous-hirsute foliage of another Melastomaceous plant, per-
haps of this genus, occurs in the Samoan collection, from the moun-
tains of Upolu. Also, under the name of Melastoma? cereacea, Dr.
Pickering, in his notes, mentions a shrub, from the mountains of
Tutuila, occuring at the elevation of about 2,000 feet, with broad,
five-ribbed and smooth leaves, a smooth and five-cleft calyx, and
with white, wax-like petals: this is probably an Astronia; but I do
not find indications of any corresponding specimens in the collection.
There are some indeterminable fragments of other species from the
Feejee Islands.

3. ASTRONIDIUM, Nov. Gen.

Flores tetrameri. Calycis tubus hemisphericus, nudus ; limbus repando-
quadridentatus seu incequaliter pluri-(6-8)-denticulatus. Petala 4.
Stamina 8, equalia: filamenta complanata, subulata, brevia: anthere
oblongo-lineares, antice rima duplicit longitudinuli dehiscentes ; con-
nectivo dorsali carnoso angusto bast calcarato. Stylus filiformis:
stigma minutum, simplicissimum. Ovarium omnino adnatum, tri-
quadriloculare: placente e fundo loculorum exorte, superne dilatate,
compresscee. Capsula globosa. Semina numerosissima.— Arbuscula
glabra, micrantha, microcarpa, facie Astroniz.

The plant in question differs from Astronia, it would seem gene-
146

582 PHANEROGAMIA.

rically, in its tetramerous flowers, calcarate anthers, and minute
stigma.

1. ASTRONIDIUM PARVIFLORUM, Sp. Nov. (Tab. 72.)

Haz. Ovolau and Ambau, Feejee Islands: along the banks of
streams.

A small tree, about 20 feet high, glabrous, or the nascent parts
minutely ferrugineous-pubescent. Branches terete; the ultimate
branchlets somewhat angled. Leaves oblong or elliptical, opposite,
obtuse or acutish at both ends, or the uppermost slightly and bluntly
acuminate, 3 or 4 inches long, from an inch to 2 inches wide, char-
taceous in texture, drying of a fuscous hue, smooth, three-ribbed, and
with a pair of submarginal nerves; the ribs connected by transverse
veins, which are rather conspicuous underneath, and reticulated with
minute veinlets. Petioles from half an inch to an inch long. Cyme
terminal, decompound, paniculate-corymbose, trichotomous; the rami-
fications compressed-angular, subtended by small foliaceous bracts,
which are early deciduous; the ultimate ramifications 3—7-flowered
at the summit. Flowers smail (a line and a half long), very nume-
rous, crowded, on pedicels of a line or less in length. Calyx nal:ed,
between hemispherical and campanulate; the free portion or limb
almost as long as the proper tube, the margin repandly four-toothed
(the teeth broad and obtuse), or irregularly 6—8-toothed, nearly in the
manner of Astronia confertiflora, &c. Petals 4, small (white ?), con-
volute in sstivation, probably caducous. Stamens 8, equal and
similar, inserted on the calyx at its junction with the summit of the
ovary: filaments flattened and dilated, subulate, the apex inflexed in
eestivation in the manner of the order, about the length of the limb
of the calyx: anthers oblong-linear, as long as the filament; the con-
nective narrow, but fleshy, produced downwardly into a basal spur of
half the length of the filament; its anterior face occupied by the
two narrowly linear and opposite anther-cells, which are extended
and curved a little round the obtuse apex, each dehiscent by a lon-
gitudinal line from top to bottom. Style filiform, terminated by a
small and simple (not in the least capitellate or dilated) stigma.
Ovary 3-4-celled, depressed-globose, entirely adnate to the tube of the
calyx. Placentee rising from the very base of the cells, wholly distinct

MELASTOMACES?. 583

from the axis, dilated and flattened upwards, broadly wedge-shaped,
emarginate at the summit, thickly covered with minute oblong
ovules. Capsule gibbose, slightly depressed, 14 to 2 lines in diameter,
smooth, with rather thin walls, 83—4-celled; the dehiscence not known.
Seeds exceedingly numerous, fillmg the cells, oval or obovate, angled
by mutual pressure, with a thin and lax testa. Embryo not seen.

Piate 72, C—ASTRONIDIUM PARVIFLORUM: a flowering branch, of
the natural size. Fig. 1. Portion of the inflorescence and fruit, of
the natural size. 2. Two flower-buds, and a central flower past
anthesis, the petals and stamens fallen. 5. A petal. 4, 5. Stamens,
from a flower-bud. 6. Vertical section of the calyx and ovary, after
flowering. 7. Placenta, viewed posteriorly. 8. A fruit, vertically
divided. 9. Transverse section of a fruit. 10. A seed—The
analyses magnified. '

4. PLEIOCHITON, Naudhn, Mss.

Flores pentamert, involucrati. Involucrum generale tri-quadriphyllum,

singuli floris bi-triphyllum. Calycis tubus turbinatus; dentes 5
duplicati; exteriores subulati, cum interioribus brevioribus membra-
naceis obtusissimis inferne connati. Petala 5, ovata, acuta. Stamina
10, equalia, aut via non cequalia: antheree lineari-subulate, apice
subrecurvee, poro unico tenuissimo aperte ; connectivo non producto
nec appendiculato. Stylus filiformis: stigma acutum. Ovariwm
ovoideum, liberum, apice verticillo setarum coronatum, quadri—quin-
queloculare. Placente axiles. Ovula subglobosa innumera.—TLrutex ?
vel arbor fere glabra; ramis validis ad nodos setoso-hispidis ; folivs
ovalibus crasse coriaceis ; inflorescentia terminali ; floribus cum bracteas
Soliaceis involucrantibus capitato-congestis.

1. Pretocurron crassiroiia, Naudin, Mss. (Tab. 73.)

Has. In the Organ Mountains, Brazil (according to the tickets
which accompany the specimen). |

A nearly glabrous shrub or tree; with terete stout branches, bristly-
hispid at the nodes; the ultimate branchlets obscurely angular,

584 PHANEROGAMIA.

sparsely hispid when young, soon glabrate. Leaves oval, nearly
obtuse, rounded and sometimes emarginate at the base, entire, thick
and coriaceous in texture, perhaps somewhat fleshy in the living
plant, dull, about 3 inches long and 2 inches wide, strongly three-
ribbed, and with a pair of less conspicuous nerves near the margin;
the ribs prominent underneath and at first furnished with a few
scattered hispid hairs; the veins immersed and obscure: petioles
stout, from 2 to 4 lines in length. Peduncles terminating the branches,
about 6 or 9 lines long, sparsely hispid when young, usually
once or twice trichotomous, the divisions subtended with ovate or
oblong, leaf-like, closely sessile bracts; each partial peduncle (4 to 9
lines long) terminated with a several-flowered capitate glomerule, or a
cluster of 3 sessile glomerules, composed of thickish and foliaceous
mvolucral and involucellate bracts, subtending and enclosing the sessile
flowers. Bracts of the involucre 3 or 4, broadly oval, very obtuse,
shightly tinged with purple; of the involucel 2, or sometimes 3,
similar to those of the common involucre, but narrower and some-
what carinate below the middle, as long as or longer than the flower
they embrace. Calyx 3 lines long, glabrous; the tube turbinate, some-
what five-angled, lightly ten-nerved; the limb five-cleft; the teeth
double; the exterior subulate from a broad base, thick, very acute,
nearly a line and a half long, at first setose-pointed, and often fur-
nished with a stout bristle in the intervening sinuses or near their
margin: the interior teeth membranaceous, considerably shorter, very
obtuse, connate with the exterior to above the middle. Petals 5, ovate,
acute, exceeding the outer calyx-lobes, purple? Stamens 10, equal
and similar, or very nearly so: filaments filiform, naked, as long as
the anthers: the latter linear-subulate, with the apex a little recurved,
opening by a minute terminal pore: the connective wholly destitute of
any appendage or protuberance. Style filiform, as long as the stamens:
stigma acute, punctiform. Ovary ovoid, free from the calyx, except
the very base, glabrous and naked, but with a crown of several (8 or
10) strong bristles surrounding the base of the style, 4—5-celled; with
as many placente projecting from the axis. Ovules very numerous,
minute, roundish, amphitropous or campylotropous? Fruit not seen.

I am indebted to M. Naudin of Paris, the distinguished recent
monographer of this order, for the investigation and name of this
remarkable plant, which appears so unlike any known American
Melastomacea, as to suggest the doubt whether the specimen may not

MELASTOMACES&. 585
have found its way into the Brazilian collection through some mis-
chance. Its aspect, indeed, is more like that of some Oriental or
Oceanic species; but I find no record that can apply to it in Dr.
Pickering’s MS. notes of the insular collections. M. Naudin considers
the plant to be the type of a new genus, of his tribe Miconiales, if
truly American, to stand near Clidemia (from which it does not
widely differ in its floral characters) ; or to be appended to the subtribe
Dissocheetee, in case it should prove to be of Oceanic origin. The
name alludes to the congested and multibracteate or involucrate
inflorescence, which forms a striking character in the plant.

PiateE 73,—PLEIOCHITON CRASSIFOLIA: a flowering branch, of the
natural size. Fig. 1. A general involucre, displayed. 2. Involucel
and the enclosed flower-bud. 3. An unopened flower. 4. Vertical
section of the same. 95. A petal. 6. Anterior, and 7, posterior view
of a stamen. 8. Pistil, detached. 9. Transverse section of a five-
celled, and 10, of a four-celled ovary. 11. An ovule.—The details
variously magnified.

9. HENRIETTEA, DC.
1. HenrRiETTEA succosa, DC.

Henriettea succosa, DC. Prodr. 8, p. 104; Naudin, in Ann. Sci. Nat. ser. 3, 18, p. 104.
H., Brasiliensis, Casaretto, Noy. Stirp. Bras. p. 85; Walp. Repert. 5, p. 716.

Has. Organ Mountains, near Rio Janeiro, Brazil.

6. LEANDRA, Raddi.
1. Leanpra vILLosa, DC.

Has. Rio Janeiro, and in the Organ Mountains, Brazil.

2. LEANDRA? ANGUSTIFOLIA, DC.

Leandra? angustifolia, DC. Prodr. 3, p. 154; Naudin, l.c. p 90.
147

586 PHANEROGAMIA.

Has. Organ Mountains, Brazil; in the neighbourhood of Rio
Janeiro.

There is only a poor specimen, with one or two very young flower-
buds, which furnish little towards completing Naudin’s description.
There are 6 triangular and acuminate small petals, and 12 stamens,
the connective thickened or gibbous at the base posteriorly. Only

the hexamerous flowers exclude it from Clidemia, as restricted by
Naudin.

7 CLIDEMIA, Don, Naudin.
1. Cripem1A Nranea, DC.

Clidemia? Nianga, DC. Prodr. 8, p. 163; Gardn. in Hook. Lond. Jour. Bot. 2, p. 344.
C. stenopetala & C. longibarbis, DC. 1. ¢.?

Has. Brazil, near Rio. Janeiro: Organ Mountains; a later state,
the flowers or fruits fallen, and with thinner leaves.
2. CLIDEMIA BOTRYOPHORA, Naudin.
Clidemia botryophora, Naudin, in Ann. Sci. Nat. ser. 8, 17, p. 350,

Has. Organ Mountains, Brazil.

3. CLIDEMIA LEPTOSTACHYA, Gardner.

Clidemia leptostachya, Gardner, in Hook. Lond. Jour. Bot. 1, p. 172.
C. salicifolia, Naudin, in Ann. Sci. Nat. 1. c. p. 864, ex char.

Has. Brazil; in the vicinity of Rio Janeiro.
This is nearly allied to C. amygdaloides (a common plant on the

Corcovado, where Gardner gathered both species), as Gardner remarks,
and most probably is no more than a narrow-leaved variety of it.

MELASTOMACE®. 587

4, CLIDEMIA ALTHMOIDES, Nawdin.

Clidemia althceoides, Naudin, in Ann. Sci. Nat. ser. 8, 17, p. 563?

Has. Brazil; in the Organ Mountains, near Rio Janeiro.

An imperfect specimen; with the leaves more tomentose under-
neath, and the fruit larger than C. altheoides is described : otherwise
it accords too well to be separated. A withered remaining petal is
lanceolate and very acute. The soft and hirsute, dense, rufous

pubescence of the branchlets is reflexed. The flowers are conspi-
cuously pedicelled; else the plant might, perhaps, have been referred

to C. retropila, DC.
5, CLIDEMIA MARGINATA, DC.

Clidemia marginata, DC. Prodr. 8, p. 156; Naudin, l.c. p. 362.

Has. Near Rio Janeiro, Brazil: and a smoother form from the
Organ Mountains.

6. CLIDEMIA CONFERTIFLORA, DC. /. c.

Has. Near Rio Janeiro, Brazil: in forests.

7. CLIDEMIA SCANDENS, Garduer.

Clidemia scandens, Gardner, in Hook. Lond. Jour. Bot. 2, p. 34.

Has. Organ Mountains, Brazil. (Perhaps the C. Epibaterium, DC.)

8. CLIDEMIA DISPAR, Gardner, /. c.

Has. Organ Mountains, Brazil.

588 PHANEROGAMIA.

This is well characterized by Gardner; but the name is not well
chosen, since the leaves of each pair, although sometimes very
unequal, are not rarely of about the same size. The flowers are
much congested on the branches of the thyrsus, and densely hirsute
externally. Calyx-teeth short and triangular, rather blunt; the throat
- 10-ribbed inside. Petals oblong-ovate, or at first broadly ovate,
pointed. The narrowed apex of the three-celled ovary is free. The
species needs to be compared with C. Carassana, DC.

9. CLIDEMIA DASYTRICHA, Sp. Nov.

C. foliis ovalibus seu ovato-oblongis acuminatis basi obtusis subdenticu-
latis quinquenerviis supra glabratis subtus pube simplici brevi indutis,
costis cum petiolis brevibus ramulisque subtetragonis setis debilibus
patentibus visdem hispidulis creberrime vestitis; panicula terminali ;
jloribus in ramulis patentibus glomerato-congestis ; calyce hispidulo,
dentibus subulatis recurvis ; petalis lanceolatis acuminatissimis.

Has. Organ Mountains, Brazil.

We have only a rather imperfect specimen of this apparently well-
marked and undescribed species, which belongs to Naudin’s section
Kuchdemia. The stem is probably shrubby or frutescent. The
somewhat quadrangular ultimate branches, like the short petioles (4 to
6 lines long) and the principal ribs of the leaves underneath, are
thickly clothed, at least when young, with fulvous, rather coarse and
long, but weak and spreading hairs, which are themselves minutely
hispid or barbed for their whole length. This pubescence becomes
matted and somewhat furfuraceous, and probably is detached when
old. Leaves ample, from 5 to 7 inches long, equal or unequal in the
pairs, oval or ovate-oblong, acuminate, minutely denticulate, obtuse at the
base, membranaceous in texture, green and early glabrate above, or
minutely setulose-scabrous, and with some simple, short, and stiff
bristles persistent on the ribs, softly pubescent with short and simple
crisped hairs underneath, five-ribbed ; the lateral ribs near the margin
and weaker, the three stronger ones often originating just above the
base: the transverse veins very numerous, and connected by similar
veinlets. Panicle terminal, shorter than the leaves, furfuraceous-

MELASTOMACE GE. 589

tomentose; its branches spreading, the ultimate divisions terminated
by small and dense glomerules of sessile or subsessile flowers. Calya
scarcely a line in diameter, hispid with short and spreading hairs,
campanulate; the 5 teeth subulate and recurved, furnished with short
and rounded, adnate, internal appendages. Petals 5, lanceolate and
very sharply acuminate, small. Anthers 10, oblong. Ovary three-
celled, hispid at the base of the style. Fruit not seen.

10. CLIDEMIA CHATOCALYX, Sp. Nov. (Tab. 74.)

C. ramulis gracillimis teretibus, gunioribus pube stellata furfuraceo-
puberulis; foliis (cujusque jugi seepe incequalibus) longe petiolatis
membranacers ovatis seu lanceolato-ovatis acuminatis subcordatis quin-
guenervis glabris setoso-ciliatis; paniculis terminalibus paucifloris
laxis; pedunculis pedicellisque setoso-hispidis ; calyce setosissimo,
dentibus aristeformibus tubo campanulato longioribus basi cum appen-
dicibus internis lanceolatis parvis coalitis; petalis subulato-acumina-
tissimis ; ovario troloculart.

Has. Brazil; in the Organ Mountains.

Stem herbaceous? Branchlets slender, terete, diffuse; the older
glabrate; the younger ones clothed with a minute and furfuraceous
stellular pubescence. Leaves (of the same pair often unequal, sometimes
equal), long-petioled (the slender petioles about an inch long, puberu-
lent like the branchlets), thin and membranaceous, ovate or lanceolate,
ovate, acuminate, subcordate with a narrow sinus, about 2 inches long,
green both sides, glabrous or nearly so, except the 6 ribs, which are
puberulent underneath and often hispid-bearded next their origin ;
the innermost arising a little above the base; the margin entire,
ciliate with long and spreading bristles. Punicle terminal, or becoming
lateral by innovation, small, short-peduncled, loose, few-flowered; the
slender and spreading peduncles and pedicels strongly hispid with long
and spreading bristles, and stellately puberulent. Bracts linear,
setose-ciliate. Calyx strongly setose-hispid; the twbe campanulate or
somewhat hemispherical, a line and a half long; its truncate margin
bearing 5 awn-shaped teeth, of 2 (or in fruit 3) lines long, spreading,

sparsely setose, each with a small and fragile internal appendage semi-
y > g
148

590 PHANEROGAMIA.

adnate to its base. Petals 5, narrow, subulate-acuminate, rather shorter
than the calyx-teeth. Anthers linear-oblong, narrowed upwards.
Style filiform, elongated: stigma minute, simple. Ovary three-celled.
Fruit dry, globular, setose, 2 lines in diameter, three-celled; the par-
titions very thin. Seeds very numerous, obovoid, angled, with a
smooth testa, marked on one side by a broad and dark brown rhaphe.

PLATE 74.—CLIDEMIA CHHTOCALYX: a branch, of the natural size.
Fig. 1. Vertical section of a flower. 2. A petal. 38. A stamen. 4.
Fruit (with the persistent calyx-lobes), divided transversely. 5. A
seed.—The details variously magnified.

11. CLIDEMIA OocARPA, Sp. Nov.

C. ramulis compresso-angulatis glabris; foliis ovali-oblongis utringue
acuminatis in petiolum brevem attenuatis serrulatis guintupli-septupli-
nervous supra glabellis subtus hirtellis; panicula terminali multiflora
glabrata; floribus cymulosis subsessilibus; calyce urceolato parce
hirtello, dentibus subulatis tubo paullo brevioribus, appendicibus obso-
letis; petalis oblongis acuminatis; fructu ovoideo triloculari.

Has. Brazil; in the Organ Mountains.

There is only an imperfect, fruiting specimen, with one or two
withered petals and stamens remaining to certify the genus. The
species is apparently a new one. The branch is rather stout, appa-
rently herbaceous, and, as well as the young branchlets, glabrous, com-
pressed-angled. Leaves ample, 4 to 6 inches long and 2 or 8 wide,
rather membranaceous, oval-oblong, acuminate at both ends, tapering at
the base into a very short petiole, serrulate, strigose when young but
soon nearly glabrate above, minutely hirsute beneath with a simple
pubescence; the ribs 5 to 7, springing in pairs along the narrowed
base, the uppermost pair more remote and stronger; the transverse
veins and veinlets very numerous and conspicuous underneath.
Panicle terminal, becoming alar by innovation, open, sessile; its
branches glabrous, except a few small bristles; the flowers cymulose,
almost sessile, at least the central ones. Bractlets setaceous, minute.
Calyx urceolate, sparsely and minutely hispid; the 5 teeth subulate, 14

MELASTOMACES®. 591

to 2 lines long, rather shorter than the tube, at least after anthesis,
spreading; the internal appendages obsolete. The withered petals
oblong, sharply acummate. Anthers oblong, narrowed upwards,
opening by a terminal pore, inappendiculate. Fruit (immature) 3
lines long, ovoid, contracted between the apex and the persistent
calyx-teeth, three-celled, very many-seeded.

12. Cuipem1A BRACKENRIDGEI, Sp. Nov.

C. ramulis gracilibus subangulatis petiolisque minutim lanulosis ; foliis
(cujusque jugt imeequalibus) oblongo-lanceolatis acuminatis basi acu-
tiusculis integerrumis trinerviis membranaceis glabris; panicula ter-
minali pauciflora laxa; floribus pedicellatis; calyce fulvo-pubente,
dentibus triangulatis inappendiculatis tubo brevioribus ; petalis e basi
lata subulato-acuminatis ; ovario triloculari.

Has. Brazil; in the Organ Mountains.

Of this also we have only a single small specimen; a slender branch,
which is somewhat angled, and the upper part, as well as the petioles,
inflorescence, &c., clothed with a fine lanulose pubescence, which falls
off with age. Leaves oblong-lanceolate, or narrowly oblong, conspi-
cuously acuminate, rather acute at the base, entire, not ciliate, glabrous,
except the 3 ribs, which are minutely puberulent underneath, veiny,
and with an obscure submarginal nerve, membranaceous, pale beneath.
The two of each pair are unequal, one of them 4 or 5, the other 12
to 24 inches long. Panicle terminal, small and simple, loosely few-
flowered; the flowers pedicelled. Calyx fulvous-pubescent, 12 lines
long; the 5 triangular teeth shorter than the somewhat turbinate tube,
and shorter than the triangular-subulate and sharply acuminate petals ;
the internal appendages wanting. Anthers oblong, narrowed upwards,
opening by a terminal pore. Ovary three-celled. Fruit not seen.

13. CiipEemMIA (OXYMERIS) PULCHRA.
Oxymeris pulchra, Chamisso, fide spec. ex Herb. Berol.

Has. Brazil; in the Organ Mountains.

592, PHANEROGAMIA.

The imperfect specimen, without petals or stamens, accords very
well with a fruiting one from Sellow’s collection, distributed from the
Royal Berlin herbarium, under the name of “ Oxymeris pulchra,
Cham.;” a species which I find nowhere published. Our specimen
has the ribs of the leaves more bearded underneath, and similar
bristly hairs occur on the petioles, and sparingly on the branchlets.
It is related to C. alpestris, Gardn., but distinct. Perhaps it may not
be distinct from Oxymeris ciliata of Martius; which, it may be
remarked, cannot bear the name of Clidemia Marti, given it by
Naudin (in Ann. Sci. Nat. p. 375), since he has already a C. Mar-
tirana in the same volume, p. 340.

14. CuipemrA (OxyMerISs) PickEertncu, Sp. Nov.

C. glaberrima, herbacea? diffuse ramosa; ramulis gracilibus angulatis ;
folits membranaceis ovalibus sew obovatis utringue acuminatis petio-
latis integerrimis longe supra basim triplinervis et nervo utrinque
marginali donatis; panicula terminali parva laxiflora; alabastris
acuminatis; calyce turbinato, dentibus brevissimis callosis; petalis e
basi lata subulato-acuminatissimis ; antheris obtusissimis.

Has. Organ Mountains, Brazil.

A slender flowering branch, of an apparently herbaceous and diffuse
plant, entirely glabrous, except some minute resinous-like atoms on
the angular branchlets and other new parts. Leaves membranaceous,
obovate or oval, 2 or 3 inches long, abruptly and conspicuously acwm-
nate, tapering at the base into a petiole of 5 to 10 lines in length, dull,
paler beneath, very entire, triplinerved above the base a considerable
distance, and with a slender submarginal nerve on each side; the
transverse veins not prominent: the leaves of each pair somewhat
unequal. Panicle terminal, small and simple, with slender spreading
branches, loosely-flowered ; usually three flowers at the extremity, the
middle one subsessile, the lateral short-peduncled. Bracts minute,
subulate. Flower-buds 2% lines long, fusiform, much acwminate.
Calyx turbinate, rather oblong, slightly puberulent, the truncate
margin furnished with 5 very short callous teeth, apparently destitute
of internal appendages. Petals subulately long-acuminate from a broad

MELASTOMACES. 593

base, white, a line and a half long: at their insertion within are two
or three slender setz to each, as long as the subulate filaments.
Anthers narrowly oblong, not narrowed upwards, very obtuse. Style
slender: stigma minute. Ovary three-celled, many-ovuled. Fruit
not seen.

This species is doubtless allied to Naudin’s Clidemia acutiflora,
also from the Organ Mountains; but on comparison it was seen to be
distinct.

There remains an imperfect fruiting specimen, also gathered in the
Organ Mountains, with a five-celled ovary and aristiform calyx-teeth,
probably of this genus; but I cannot determine, nor venture to cha-
racterize it.

8 STAPHIDIUM, Naudin.

1. STAPHIDIUM BISERRATUM, Naudin.

Staphidium biserratum, Naudin, in Ann. Sci. Nat. ser. 3, 17, p. 305.
Clidemia biserrata, bullosa, umbonata, & forte spicata, DC. Prodr. 3, p. 158.

Has. Vicinity of Rio Janeiro, Brazil.

2. SraPHIpDIUM PAUCIFLORUM, Naudin, l. c.
Clidemia pauciflora & C. crenata, DC. Prodr. 8, p. 157.

Has. Brazil; with the preceding species.

9. MICONIA, Ruiz & Pav., Naudin.

Miconta, CREMANIUM, DipLocHITA, CHAINOPLEURA, JUCUNDA, GRAFFENRIEDA,
DecarPHE, ANGuSTINEA, Harricia, & GLOSSOCENTRUM, Auctorum, fide

Naudin, 1. c.
149

594 - PHANEROGA MIA.

1. Miconta (Jucunpa) sTAMINEA, DC.
Melastoma staminea, Desv. in Lam. Dict. 4, p. 53.
Miconia staminea, DC. Prodr. 3, p. 187; Naudin, in Ann. Sci. Nat. ser. 3, 16, p. 123.

Graffenrieda jucunda, Mart. & Zuce. Nov. Gen. & Spec. 3, t. 276, forma angustifolia.
Jucunda Lhotskyana & J. Martiana, Cham. in Linnea, 9, p. 456.

Has. Brazil, near Rio Janeiro, and in the Organ Mountains.

2. MIconrIA BRUNNEA, DC.
Miconia brunnea, DC. Prodr. 3, p. 184; Naudin, 1. c. p. 200.

Has. Brazil, near Rio Janeiro; or in the Organ Mountains.

3. MIconraA CINERASCENS, Miquel ?

Miconia cinerascens, Miquel, in Linnea, 22, p. 543 ?
Has. Brazil; in the Organ Mountains.

The plant before me belongs to Eumiconia, Paniculares § 1, a, of
Naudin’s arrangement. It is apparently the same as no. 1296 of
Pohl’s collection, communicated by the Vienna herbarium to that of
Hooker. Except that the pubescence is at first of a rusty or ferru-
gineous hue, and the calyx is scarcely costate, it accords so well with
Miquel’s character of M. cinerascens that it would not be proper to
establish a new species upon it.

4. MicontIa DIvVARIcATA, Gardner.
Miconia divaricata, Gardner, in Hook. Lond. Jour. Bot. 2, p. 345.

Has. Organ Mountains, Brazil.

The branches of the panicle are scarcely divergent in these speci-
mens.

MELASTOMACES®. 595

5. MIcoNnIA FASCICULATA, Gardner.
Miconia fasciculata, Gardner, in Hook. Lond. Jour. Bot. 1, p. 538.
Has. Brazil; near Rio Janeiro, probably on the Corcovado, and
Organ Mountains.
6. MICONIA DEPAUPERATA, Gardner.
Miconia depauperata, Gardner, in Hook. Lond. Jour. Bot. 2, p. 346.
Has. Organ Mountains, Brazil. (Foliage only; which accords
with that of Gardner’s plant.)
7. MicontA SELLOWIANA, Naudin.

Miconia (Amblyarrhena) Sellowiana, Naudin, in Ann. Sci. Nat. 1. c. p. 206.
Cremanium Sellowianum, Cham. in Herb. Berol.

Has. Rio Janeiro and Organ Mountains, Brazil.

8. MiconrA PALUDOSA, Naudin, U. c.

Cremanium paludosum, Gardner, in Hook. Lond. Jour. Bot. 2, p. 347.

Has. Brazil; in the Organ Mountains.

9. MiconrA PRASINA, DC.

Miconia prasina, DC. Prodr. 8, p. 188; Naudin, in Ann. Sci. Nat. 1c. p. 172.
M. attenuata, afinis, & sepiaria, DC. 1. ¢. fide Naudin.

Has. Organ Mountains, Brazil.

596 PHANEROGAMIA.

10. MicontA (CHANANTHERA) CHAINOPLEUROIDES, Naudin, 1. c.

Cremanium chenopleuroides, Gardner, in Hook. Lond. Jour. Bot. 2, p. 349.
Miconia late-crenata, Naudin, in Ann. Sci. Nat. 1. ¢. p. 239, ex char. (excl. syn. DC. 7)

Has. Organ Mountains, Brazil. (The inflorescence mostly in a
diseased state.)

A few other specimens, probably of this genus, occur in the collec-
tion, in a state too imperfect for determination.

10. HUBERIA, DC
1. Huerta ovaAuiroiia, DC.

Huberia ovalifolia, DC. Prodr. 3, p. 167; Naudin, in Ann. Sci. Nat. ser. 3, 15, p.
336, & 16, t. 25, f. 1.

HT. resinosa, Presl, Symb. Bot. 1, p. 58, t. 37.

Has. Rio Janeiro, Brazil.

ll. BERTOLONIA, Raddi.
1. BERTOLONIA NYMPHZAFOLIA, DC.

Bertolonia nympheefolia, Raddi, Pl. Brasil. Add.; DC. Prodr. 3, p. 113; Naudin,
in Ann. Sci. Nat. ].c. p. 318.

Rhexia nympheifolia, Kunth. in Bonpl. Rhex. t. 53.

Has. Organ Mountains, Brazil. (In fruit.)

2. BertotoniA Leuzeana, DC. 1. ¢.

Rhexia Leuzeana, Bonpl. Rhex. p. 144, t. 54, 55.

Has. Rio Janeiro, Brazil. (A state with rounder leaves.)

MELASTOMACEA. 597

12, ANPLECTRUM.+

APLECTRUM, Blume, in Flora, 1831, p. 502, & Mus. Bot. Lugd.-Bat. p. 37; Naudin,
in Ann. Sci. Nat. ser. 3, 15, p. 3038, t. 15; non Nutt. (1818).

1. ANPLECTRUM? OVALIFOLIUM, Sp. Nov.

A, fere glaberrimum; foliis late ovalibus sew ovato-rotundis breviter
acuminatis petiolates submembranaceis tripli-quintuplinerviis ; cymis
pulverulento-puberulis folio brevioribus; calycis dentibus brevissimis
callosis, denticulis externis nullis; petalis ovatis acutis; antheris 8
homomorphis.

Aplectrum ? ovalifolium, Naudin, in scheda.
Has. Feejee Islands; at Sandal-wood Bay, Vanua-levu.

Shrub, or small tree, glabrous, except a very minute pulverulent
pubescence on the nascent parts; the branches terete, nodose. Leaves
nearly membranaceous, about 2 inches long, or one of each pair
smaller, conspicuously petioled (the petiole half an inch long), broadly
oval or rounded-ovate, short-acuminate, rounded but not cordate at the
base, dull, paler beneath, very entire, triple-ribbed from near the base,
and with a pair of lateral, nearly basal nerves, which approach the
margin. Petioles unconnected by a transverse membrane. Cyme
terminal and in the upper axils, small, minutely puberulent, sometimes
paniculate, shorter than the leaves, several-flowered. Bracts linear.
Flower-buds a line or a line and a half in length, short-pedicelled.
Calyx turbinate; the limb with 4 very short and obtuse callous teeth,
and destitute of exterior appendages. Petals 4, ovate and acute in the
bud. Stamens 8: anthers received in estivation into as many narrow
canals left between the calyx and the wall of the ovary, oblong, with

* A slight change in the privative may render Blume’s name, Aplectrum, sufficiently
different from the earlier Aplectrun of Nuttall, which is apparently a good genus, of

Orchidaceee, allied to Corallorhiza.
150

598 PHANEROGAMIA.

the connective produced posteriorly into a short and thick tubercle,
almost spur-like, apparently all similar and equally polliniferous;
but they have been examined only in young buds. Style filiform,
short: stigma minute, punctiform. Ovary four-celled, with thick and
multiovulate placente, affixed to the middle of the columella. Fruit
not seen.

M. Naudin, who has obligingly examined the incomplete specimens
of the collection, inclines to refer the plant to Blume’s genus Aplec-
trum; notwithstanding that the anthers are probably all similar
and perfect, and the calyx is not adnate to the whole surface of the
ovary.

13. MEDINILLA, Gaud.
1. MEDINILLA HETEROPHYLLA, Sp. Nov. (Tab. 75.)

M. alte scandens; ramis teretibus ad nodos scepius radiciferis; foliis
cwjusque jugi valde disparibus quintuplinerviis, majore ovato seu ovato-
oblongo subcordato subacuminato, petiolo ejus folium alterum cordato-
rotundum adeequante vel superante; racemis paniculatis elongatis ;
bracteis verticillatis bracteolisque obovatis magnis petaloideis albis ;
Jloribus tetrameris; calycis limbo fere integerrimo; antheris basi bre-
viter tricalcaratis.

Has. Feejee Islands; in forests of Ovolau.

Stem shrubby, thick, climbing over tall trees; the branches terete,
slender, commonly rooting copiously at the nodes, nearly glabrous, as is
the whole plant. Leaves exceedingly unequal and dissimilar, one of
each pair usually almost abortive; the larger one ovate or ovate-oblong,
subcordate, acute or slightly acuminate, quintuple-ribbed or almost five-
ribbed, besides an obscure pair of intramarginal nerves, from 23 to
4% inches long, and with its slender petiole half an inch to an inch
long: the smaller leaf is rounded-cordate, short-petioled or subsessile,
either acutish, obtuse, or emarginate, 3—5-ribbed, from 3 to 9 lines in
diameter, shorter than or barely equalling the petiole of the larger

MELASTOMACES®. 599

leaf. In consistence the leaves appear to have been a little fleshy,
but thin. The inflorescence is developed from the leafless nodes of
older branches, in the form of an ample panicle, composed of drooping
elongated racemes, which attain the length of a foot or more; their
closely approximate nodes each adorned with a pair, or more com-
monly a whorl of 3 or 4 obovate and petaloid (white) bracts. These
are sessile, and persistent for some time after flowering on the nodose
rhachis, half an inch long, about the length of the pedicels; which
are not articulated in the middle. Flower subtended by a pair or
whorl of bractlets resembling the bracts, but smaller and rounder, about
the length of the calyx and embracing it. Calyx nearly 3 lines long,
tinged with violet-colour, urceolate, the tube somewhat quadrangular;
the dinb membranaceous, with a truncate, entire, at length repandly
sinuate or obscurely four-lobed margin, externally marked with 4
small callosities. Petals 4, nearly 3 lines long, obovate-cuneiform,
somewhat insequilateral, retuse, and with a minute glandular tip,
rose-colour. Stamens 8, equal: filaments slender: anthers subulate,
rose-coloured, minutely three-spurred at the base; that is, the connec-
tive bearing 2 (yellow) assurgent spur-like processes anteriorly, and
one posteriorly, which is similar but decurved. Style filiform ;
stizma minute, punctiform. Ovary wholly adherent to the calyx-
tube, surmounted by a cup-shaped membranous disk, which surrounds
the base of the style, four-celled; the thick axile placents: covered
with innumerable ovules. Berry globular-ovoid, a quarter of an inch
in diameter, many-seeded, purple? Seeds semi-obovoid, with the
inner face concave, or helmet-shaped; the rhaphe large and project-
ing. Embryo somewhat oblique.

A striking and very distinct species, remarkable as well for its long
racemes, conspicuously adorned with white bracts, as for the great
inequality of the leaves. Its inflorescence would seem to resemble
that of Dactyliota, Blume; a genus which perhaps should be reunited
to Medinilla.

Puate 75.—MeEpINILLA HETEROPHYLLA: a branch, of the natural
size. Fig. 1. A flower, with its bractlets. 2. Vertical section of an
unexpanded flower. 3. Calyx. 4. A petal. 5,6. Stamens. 7. A
fruit. 8. Transverse section of the same. 9, 10. Seeds. 11. Ver-
tical section of a seed.—The details magnified.

600 PHANEROGAMIA.

2. MEDINILLA RHODOCHL&NA, Sp. Nov.

M. subscandens; ramulis teretibus hine wide radicantibus ; foliis cujus-
que jug incequalibus homomorphis ovato-oblongis acuminatis quintu-
plinerviis basi scepe obliquis acutis vel obtusis petiolatis puberulis ;
racemis axillaribus ; bracteis bracteolisque dilatatis rubris.

Has. Feejee Islands; on the mountains of Ovolau, at the altitude
of 2,000 feet.

“

Of this the collection comprises only one or two very incomplete,
fruiting specimens, with the vestiges of the inflorescence and bracts ;
which appear to belong to a species mentioned in Dr. Pickering’s
notes (in connexion with the foregoing), as a “half-climbing plant,
with a weak stem, 20 or 30 feet long; and with red bracts.” The
branchlets are terete, thickened at the nodes, some of which bear
rootlets, their young summits ferrugineous-pubescent. Leaves of each
pur unequal in size, but otherwise nearly similar, the larger 3 or 4,
the sma.ler one or 2 inches long, ovate-oblong with an acute and often
oblique, sometimes obtuse base, the apex usually acuminate, qguintuple-
ribbed, puberulent, especially beneath, with rusty scurf, thickish in
texture; the sparse transverse veins rather conspicuous. Petioles 4
to 8 lines long. Racemes axillary, perhaps paniculate, apparently
shorter than the leaves. Bracts and bractlets opposite or ternate,
apparently oval or rounded and longer than the pedicels, petaloid,
“ved.” Berry many-seeded. Seeds somewhat helmet-shaped, exca-
vated on the ventral face; the smooth testa minutely punctate.

14. DISSOCH ATA, Blume.
1. Dissocuata Cumineu, Naudin ?
Dissocheta Cumingii, Naudin, in Ann. Sci. Nat. ser. 8, 15, p. 75?

Has. Luzon; in the mountains, near Bafios. (Without flowers or
fruit.)

MELASTOMACEA. 601

15. SPENNERA, Mart.

1. SPENNERA PALUDOSA, Mart., DC.

Has. Rio Janeiro and Organ Mountains, Brazil: very common.

16. TETRAMERIS, Naudin.
1. TETRAMERIS MArtiAnA, Naudin.

Tetramerts Martiana, Naudin, in Ann. Sci. Nat. 1. ¢. 14, p. 121.
Arthostemma Martiustanum, DC. Prodr. 3, p. 187.

Has. Brazil; in the Organ Mountains.

17. MELASTOMA, Burm.

1. Mexastoma Tartense, DC.

Melastoma Taitense, DC. Prodr. 8, p. 144; Guill. Zeph. Tait. p. 60; Naudin, in
Ann. Sci. Nat. ser. 3, 18, p. 275.
M. Malabathrica, Forst. Prodr. Fl. Ins. Austr. p. 33, & descr. in Guill. Zeph. Tait. 1. c.

Has. Tahiti, Society Islands; on dry ridges.

“ Flowers white ;” as is also said to be the case by Forster.

2. Metastoma ViTIENSE, Naudin, l. ¢.

Has. Feejee Islands; Malolo, Ovolau, &c.; both near the coast and
on mountains. Samoan Islands; Savaii and Tutuila.

The Feejee specimens are mostly smaller-leaved than those charac-
| 151

602 PHANEROGAMIA.

terized by Naudin: the Samoan form is intermediate, as to the calyx,
&c., between the former and IM. Taitense; to which both are probably
to be referred as mere varieties.

3. MELASTOMA POLYANTHUM, Blume ?
Has. Ovolau, Feejee Islands, near the coast.

The specimen is insufficient for determination. The leaves are
strigose above, but softly appressed-pubescent underneath. The
calyx-teeth are narrow, acute, and as long as the tube.

4. MELASTOMA FASCICULARE, Naudin ? 1. ¢.

Haz. Luzon, Philippine Islands, near Manilla. (Specimen too
imperfect for satisfactory determination.)

5. MELASTOMA ADPRESSUM, Wail. Cat.
Melastoma longifoliwm, Naudin, in Ann. Sci. Nat. 1. ¢. p. 298, ex char.

Has. Singapore; near the town.

18. PLEROMA, Don.

PLEROMA, Don, in Mem. Wern. Soe. 4, p. 293; DC. Prodr. 3, p. 151, excl. spp.
prior.; Benth. in Hook. Jour. Bot. 1, p. 171, & 2, p. 288 (1840); Gardner, in
Hook. Lond. Jour. Bot. 2, p. 8348; Lindl.

Lastanpra, DC. Prodr. 8, p. 127; Mart. & Zuce. Nov. Gen. & Spec. 3, p. 98;
Naudin, in Ann. Sci. Nat. ser. 8, 18, p. 127.

It is a pity that Naudin, in his excellent elaboration of Melasto-
macece, should have employed the name of Lasiandra for this genus;
since that of Pleroma has not ouly the absolute priority in publi-
cation, but was also adopted for the genus by Bentham (ten years

MELASTOMACEA. 603

before Naudin’s papers appeared), on the occasion of his showing the
identity of the two genera; and in this he has been followed by
Gardner, Lindley, &c. Irrespective of the rule of priority even,
Don’s name is the more appropriate; since, being nearly unmeaning,
it is unobjectionable; while that of Lasiandra is descriptive of only a
portion of the species.

1. PLERoMA FontTANESIANUM, Gardner.

Pleroma Fontanesianum, Gardner, in Hook. Lond. Jour. Bot. 1, p. 171.
Lasiandra Fontanesiana, DC. Prodr. 3, p. 180; Naudin, 1. c. p. 150.
Rhexia Fontanesti & Rk. Langsdorfiana, Bonpl. Rhex. t. 51 & 56.

Has. Rio Janeiro, and in the Organ Mountains, Brazil.

2. PLEROMA ARGENTEUM.
Lasiandra argentea, DC. Prodr. 3, p. 181; Naudin, 1. c. p. 148.

Has. Near Rio Janeiro, and in the Organ Mountains, Brazil.

3. PLEROMA ADENOSTEMON.
Lasiandra adenostemon, DC. Prodr. 8, p. 180; Naudin, l.c. p. 189.

Has. Organ Mountains, Brazil.

4, PLEROMA SPOLIATUM.

Lasiandra spoliata (& L. phalacrostemon), Naudin, in Ann. Sci. Nat. 1. ¢. p. 137.

Has. Organ Mountains, Brazil.

5. PLERoMA GAUDICHAUDIANUM.

Lasiandra Gaudichaudiana, DO. Prodr. 3, p. 127; Naudin, lc. p. 136,

Has. Near Rio Janeiro, Brazil.

604 PHANEROGAMIA.

6. PLreromA KuNnTHIANUM, Gardner, var.

Pleroma Kunthianum, Gardner, in Lond. Jour. Bot. 2, p. 851; Paxt. Mag. Bot. t. 125.
Lasiandra Kunthiana, DC. Prodr. 38, p. 128?

Has. Organ Mountains, Brazil.

7. PLEROMA GRACILE.

Rhexia gracilis, Kunth, in Bonpl. Rhex. t. 52. °
Cheetogastra gracilis & C. fraterna, DO. Prodr. 3, p. 183.
Lasiandra gracilis, Naudin, in Ann. Sci. Nat. 1. c. p. 128.

Has. Organ Mountains, and near Rio Janeiro, Brazil.

19. RHYNCHANTHERA, DO.

1. RHYNCHANTHERA SCHRANKIANA, DC.

Rhynchanthera Schrankiana (cum R. dichotoma & pentanthera), DC. Prodr. 8,
Naudin, |. c. 12, p. 214.

Has. Organ Mountains, Brazil.

There remains in the collection the undeterminable foliage of
several Melastomacee from the Philippine Islands, of one from the
Samoan, and one from the Feejee Islands.

Orv. LYTHRACE .

1. PEMPHIS, Porst.
1. PEMPHIS ACIDULA, Forst.

Pemphis acidula, Forst. Char. Gen. t. 834; DC. Prodr. 3, p. 89.
Lythrum Pemphis, Linn. f. Suppl. p. 249; Forst. Prodr. p. 86; Lam. Ill. t. 408.

Has. Sooloo Islands; Wake Island, Raraka, Bellinghausen, and
small coral islands generally.

2. AMMANNIA, Lina.

1. AMMANNIA LATIFOLIA, Linn.

Ammannia latifolia, Linn. Spec. Pl. ed. 1, p. 119; Lam. Il. t. 77; Torr. & Gray,
Fl. N. Amer. 1, p. 480.

Has. Peru; in the bed of the river at Callao.

3. LYTHRUM, Linn.
1. Lyturum Sauicaria, Linn.

Has. Woolongong, New South Wales. Indigenous?
152

606 PHANEROGAMIA.

2. LytHrum Hyssopiroiia, Linn.

Has. New South Wales, near Sydney. Chili, near Valparaiso.
Introduced from Europe ?

3. LytHruM MARITIMUM, HZ. B. K.
Has. Peru: in the river-bed at Callao and Lima. Sandwich

Islands; on the coast of Oahu, near Honolulu, and Hawai, in the
district of Waimea.

4. CUPHEA, Jacg.
1. Cupnza Batsamona, Cham. & Schlecht.

Cuphea Balsamona, Cham. & Schlecht. in Linnea, 2, p. 368.
Has. Brazil, near Rio Janeiro, and in the Organ Mountains.
(With a hispid variety.)
2. CUPHEA INGRATA, Cham. & Schlecht. 1. c.

Has. Organ Mountains, Brazil. (The var. C. Bonariensis, Gillies.)

3. CUPHEA LIGUSTRINA, Cham. & Schlecht. 1. c.

Has. Brazil; with the preceding.

4. CUPHEA FLAVA, Spreng., DC.

Has. Vicinity of Rio Janeiro, Brazil.

Orn. RHIZOPHORACEH &.

Susorpvp. l. LEGNOTIDE &.
1 GYNOTROCHES, Blume.

GynorrocuEs, Blume, Bijdr. p. 218, & Mus. Bot. Lugd.-Bat. p. 126.
. DryYpropeTALUM, Arn. in Ann. Nat. Hist. 1, p. 872.

1. GYNOTROCHES RETICULATA, Sp. Nov.

G. ramulis novellis stipulisque pube minutissima cinereis ; foliis oblongis
acumine brevi obtusissimo terminatis utrinque reticulatis, venis promi-
nulis scabriusculis; pedicellis petiolum adcequantibus ; stigmate bre-
vissime tri-quadriradiato.

Has. Singapore.

This is the same as No. 8457 of Wallich’s Catalogue, from Penang
and Singapore, and is perhaps not specifically distinct from G. Drypto-
petalum, Blume (Dryptopetalum corvaceum, Arn.), of which I have
only an imperfect specimen from Griflith’s Malacca collection for
comparison. But in our plant the young branches, stipules, &ec., are
minutely cinereous-puberulent ; the oblong leaves have a shorter and
very blunt acwmination, are more reticulated on both sides, duller, and
with the prominent veins and veinlets scabrous to the touch; the
pedicels (about 3 lines long, very numerous in a dense fascicle) as
long as the petiole. The style, moreover, is terminated by a depressed
slightly four-lobed, or sometimes three-lobed stigma; the lobes extremely

608 PHANEROGAMIA.

short and obtuse. The petals are wanting in the specimen. Persis-
tent stamens 8: filaments subulate-filiform. Ovary, &c., apparently
as in Blume’s figure of G. axillaris.

2 HAPLOPETALON, Nov. Gen.

Calyx profunde quadrifidus; lobis triangulatis cestivatione valvatis.
Petala 4, obovata, calyce inserta, fere exwnguiculata, integerrima,
carinata, cestivatione involuta, decidua. Stamina 16-20, brevissima,
uniserialia, margint disci perigynt tenuis inserta: anthere ovales,
introrse, biloculares, longitudinaliter dehiscentes, filamentis subulatis
cequilonge. Stylus brevis, apice quadrifidus, raro quinquesidus ; lobis
linearibus demum patentibus apice stigmatosis. Ovarium depressum,
culycis tubo (mediante disci) semiadnatum, uniloculare. Ovula 8,
raro 10, anatropa, e columna centrali geminatim appensa. Frutex
Vitiensis; foliis Carallise; stipulis interpetiolaribus caducis; pedi-
cellis in awillis laxe fasciculatis.

1. Haptoprtaton Ricuu, Sp. Nov. (Tab. 76.)
Has. Feejee Islands: at Sandalwood Bay, Vanua-levu.

Shrub (or small tree, the height not recorded) glabrous, except the
young shoots; which are hirsute with a fine, rusty, deciduous pubes-
cence. Leaves opposite, chartaceo-coriaceous, somewhat shining,
oblong or obovate-elliptical, obscurely serrulate above the middle,
obtusely acuminate or obtuse, tapering at the base into a petiole (of
4 to 6 lines in length), pinnately-veined and irregularly reticulated,
very obscurely pellucid-punctate, from 22 to 5 inches long, and one
or 2 inches wide. Stipules interpetiolar, like those of the family, con-
volute in vernation, fuscous, half an inch long, caducous. Flowers
in simple and sessile axillary fascicles, which are at first enclosed by
an involucre of two or three fuscous, early caducous scales. Pedicels
lax, 7-10 in a cluster, about 2 lines long, not articulated in the
middle. Flower-buds naked, a line or a line and a half in diameter,
globose—four-angled. Calyx deeply four-cleft; the lobes thickish,

RHIZOPHORACE &. 609

broadly triangular, minutely pubescent at the very tip, otherwise
vlabrous, valvate in estivation. Petals 4, involute in estivation, obovate,
or somewhat cuneate, scarcely if at all unguiculate, retuse at the apex,
thickish in texture, entire, carinate, and externally pubescent on the
keel, not longer than the lobes of the calyx, and inserted in their
sinuses under the thin edge of the perigynous disk, deciduous.
Stamens 16 to 20, shorter than the petals, wniserial, inserted on the
thin and slightly free margin of the perigynous disk that lines the tube
of the calyx, so as to appear somewhat monadelphous: jilaments very
short, subulate, inflexed in eestivation; the alternate ones a little
shorter than the others: anthers oval, somewhat didymous, fixed near
their base, «trorse, two-celled, the cells longitudinally dehiscent. Style
short, as long as the stamens, somewhat four-grooved, four-clef/t or
rarely five-cleft at the apex; the lobes linear, flattish, obtuse, cruciately
spreading in anthesis, stigmatose and a little thickened at the apex.
Ovary depressed, its free summit pubescent, four-rayed, entirely desti-
tute of any epigynous disk or ring, the lower part adnate to the flat-
tened base of the calyx (when fructified probably becoming wholly
or nearly superior), one-celled, apparently without even a trace of
rudimentary partitions. Ovules 8, or rarely 10, anatropous, radiately
attached in pairs to a central columella which reaches to the base
of the style; their rhaphes opposite, at length becoming pendulous.
Fruit not seen.

This plant, of which the fruit is still a desideratum, is manifestly
related to Gynotroches and Cassipourea, and still more to Carallia, one
species of which has a half free ovary, and two have merely crenu-
late petals. The tetramerous flowers, with short stamens, of more
than twice the number of the perfectly entire and fleshy petals, along
with the difference in the inflorescence, may, with our present know-
ledge, be held to constitute a sufficient generic distinction. The name
which I have chosen, in reference to its entire petals, indicates an
obvious difference between it and all its allies.

A memorandum attached to the specimens had apprised me that
Mr. Rich, the Botanist of the Expedition, regarding this as a new
genus, had assigned to it the name of Agatea, in compliment to the
late Mr. Agate, the Botanical draughtsman of the Expedition. A
former but too hasty examination having led me to refer this plant
to an old genus of Legnotidee (which must be my apology for the

15

610 PHANEROGAMIA.

seeming neglect of a manuscript name), I dedicated a genus of Vio-
lace (p. 89) to the memory of Mr. Agate, which unfortunately is
already published. This is the less to be regretted, however, since
the present genus is not so strongly marked in its characters as is
desirable for this purpose, and it may not improbably be merged in
Carallia.

Prate 76.—HaptoreraAton Ricui: a flowering branch, of the
natural size. Fig. 1. Stipules. 2. Diagram of the flower. 3. A
flower-bud. 4. Vertical section of the same. 5. An expanded flower.
6. Vertical section of the same. 7. Stamens, with a portion of the
disk. 8. A pair of ovules—The details variously magnified.

3. CROSSOSTYLIS, Furst.

Calyx profunde quadrifidus, quadrangularis; lobis triangulatis cesti-
vatione valvatis. Petula 4, calyce inserta, ligulata, unguiculata, apice
truncato tri-quinquedentata, carinata, cestivatione involuta, decidua.
Stamina fertilia circiter 20, margini libera disci perigyni inserta :
jilamenta elongata, subulato-filiformia, basi dilatata subconnata intus
glandula subglobosa instructa: anther: ovoidece, introrse, biloculares,
longitudinaliter dehiscentes. Filamenta sterilia circ. 20, fertilibus
alterna, wsdem dimidio breviora, subulato-linearia, villosissima.
Stylus filiformis, elongatus, apice radiato-12-fidus ; lobis filiformibus
apice stigmatosis in phalanges 4 basi plus minus coadunatis. Ova-
rium depressum, calycis tubo semiadnatum, vertice libero demum™ton-
vecum 12-radiatum, 12-loculare. Ovula in loculis gemina, anatropa, e
columna centrali appensa. (Bacca hemispherica, multistriata, supera,
untlocularis, polysperma, ex Forst.)—Arbuscula glabra; foliis oppo-
sitis obovatis integerrimis vel subserrulatis; stipulis interpetiolaribus
caducis; pedunculis axillaribus brevibus apice bi-trifloris rariusve
quadrifloris ; floribus pedicellatis.

1, CROSSOSTYLIS BIFLORA, Forst. (Tab. 77.)

Crossostylis biflora, Forst. Char. Gen. p. 87, t. 44, Prodr. p. 49, & Ms. ed. in Guill.
Zeph. Tait. p. 60.

RHIZOPHORACE &. 611

Haz. Mountains of Tutuila, one of the Samoan or Navigators’
Islands.

A small tree, glabrous throughout, with stout, terete, nodose
branches. Leaves opposite, obovate, 4 to 7 inches long, and 2 or 3
wide, obtuse or slightly acuminate, cuneate or contracted at the base
into a petiole about half an inch in length, loosely feather-veined, and
with the veinlets reticulated, chartaceous in texture, somewhat
shining, of the same hue both sides, very thickly and obscurely
pellucid-punctate by transmitted light. Stipules iterpetiolar, con-
volute, fuscous, caducous, resembling those of the Legnotidece gene-
rally. Peduncles axillary, solitary, shorter than the petioles (3 or 4
lines long), naked, bearing 2 or 3, rarely 4, one-flowered (or occa-
sionally three-flowered) pedicels articulated with their apex and of
about the same length, subtended by small caducous bracts? Flower
4 or 5 lines long when developed, nodding. Calyx quadrangular,
fleshy, deeply four-cleft; the very short tube somewhat turbinate, the
broadly triangular lobes valvate in estivation, glabrous, persistent.
Petals 4, inserted on the throat of the calyx just within the sinuses,
longer than its lobes, oblong or at length ligulate, tapering gradually
into a claw, 3 lines long, minutely and softly pubescent externally,
except the thin margins, glabrous inside, carinate (“ white,” Forster),
truncate at the apex, where it is erosely three—ive-toothed, with
the teeth subulate and unequal, the middle one usually longer and
setaceous, in cestivation involute or almost conduplicate. Stamens
usually 20, inserted on the slightly free margin of a fleshy perigynous
disk which lines the tube of the calyx: filaments slender, as long as
the petals, filiform from a subulate and dilated base, where they are
slightly concreted into a ring, along with as many alternate sterile fila-
ments; the latter subulate-linear, flat, and somewhat petaloid, nearly
naked below, above very hirsute-villous, especially on the inner face,
nearly as long as the calyx-lobes, about half the length of the antheri-
ferous filaments, which are glabrous, and each furnished with a large
and conspicuous, glabrous, globular gland at tts base inside. Anthers
ovoid, introrse, two-celled, the cells opening longitudinally. Style
slender, as long as the stamens, jiliform, minutely striate; the apex
abruptly cut into a radiated fringe of mostly 12 short and filiform
lobes, stigmatose at their apex, and more or less manifestly collected
or at the base united into 4 phalanges, but by no means to the extent

612 PHANEROGAMIA.

represented in Forster's rude figures. Ovary strongly depressed, its
lower part adnate to the broad base of the calya:-tube, the free summit at
first flat, radiately 12-striate, after anthesis becoming convex, 12-celled
within, the thin dissepiments apparently evanescent after anthesis.
Ovules 2 in each cell, collaterally inserted on a central axis or column,
anatropous, at first horizontal, and with the upward growth of the
ovary after anthesis becoming pendulous. Fruit not seen in the
collection. According to Forster it is a hemispherical, many-striate,
superior, one-celled berry, containing several globular seeds,

Our collectors did not meet with Crossostylis at the Society Islands,
nor indeed has it been detected since the time of Forster. It
appears, indeed, from the habitat, cited in Guillemin’s Zephyritis
Taitensis, that Forster did not collect it on Tahiti; but on the out-
lying island of Razatea. Our specimens (unfortunately incomplete
ones, with only some very young flower-buds and a few detached
flowers) all came from Tutuila, one of the Navigators’ Islands: but
they appear to belong to Forster's species, judging from the detailed
description reproduced by Guillemin, above-cited, and from the speci-
men of Lambert’s herbarium, now in that of the British Museum,
which, however, consists of the foliage alone. But the toothed apex
of the petals is not mentioned by Forster.

Crossostylis has always been one of the “genera incertz sedis.”
As such it was placed at the end of his Genera Plantarum by Jussieu,
who, however, with prophetic insight arranged it next to Cassipourea !
and conjectured their affinity with the Salicariee. DeCandolle ap-
pended it, with much doubt, to the Myrtacew, in which he was fol-
lowed by Endlicher. I am not aware that any later conjecture has
been hazarded in respect to it; except that Dr. Pickering, in his manu-
script memoranda, queries whether it may not rather belong to the
Philadelphiew. It is now abundantly evident, however, that we have
in Crossostylis an accession to that small group of genera which Mr.
Brown* long ago sagaciously appended to the Rhizophoree, and noted
as indicating a passage on the one hand to Salicarvece, on the other to
Cunoniacee ; and which, adopting the name of Legnotidee from Bart-
ling,t Dr. Blume has recently proposed as a separate natural order.t{

* Obs. Bot. Congo, p. 18.
+ Ordines Naturales Plantarum, p. 318.
{ Museum Botanicum Lugduno-Batavum, p. 126.—Apparently without valid reasons ;

RHIZOPHORACES. 613

Even the tendency to have lobed, toothed, or setigerous petals, which
so generally prevails in Lhizophoracee, is indicated in the summit of
the petals of Crossostylis. It should be mentioned that the evident
resemblance of Crossostylis to the preceding genus had not escaped the
notice of Mr. Rich, who, however, was not aware of the relationship
of either with the Rhizophoracee.

Piate 77.—CROsSOSTYLIS BIFLORA: foliage and inflorescence, with
very young flower-buds. Fig. 1. Young flower-bud, enlarged. 2.
Flower, of the natural size. 3. The same, enlarged. 4. Vertical
section of the same. 95. Interior, and 6, exterior view of a petal,
much enlarged. 7. Stamens, with the glands and sterile filaments.
8. Ovary, &e. 9. Transverse section of the ovary. 10. Pistil, after
anthesis, the calyx, &c., cut away. 11. Ovary of the same, partly
cut away, to show the ovules. 12. Pistil, the ovary vertically di-
vided.—The details magnified.

Susorv. Ll. RHIZOPHORE &.

4, RHIZOPHORA, Linn.
1. RaizopHoraA Manate, Linn.

Rhizophora Mangle, Linn.; Jacq. Stirp. Amer. p. 141, t. 89; Geertn. Fruct. 1, t.
45; Lam. Ill. t. 396; DO. Prodr. 8, p. 32; Blume, Mus. Bot. Lugd.-Bat. p. 132.
Candela Americana, etc., Catesh. Car. 2, p. 63, t. 63.

Has. Brazil; in salt marshes, near Rio Janeiro. (MMangrove.)

2. RuizoPpHORA MUCRONATA, Lam.

Rhizophora mucronata, Lam. Dict. 6, p. 189, & Ill. t. 396, f. 2; DC. 1l.c.; Wight,
Ill. Ind. Bot. p. 209, & Ic. t. 238.
R. Candelaria, DC. Prodr. 3, p. 32 (ex parte) ; Wight & Arn. Prodr. 1, p. 310.

for the peculiar economy of the germination of the embryo of the Mangroves is a
special, maritime, not an ordinal character, and one which in other cases (as in Aroi-
dece) belongs to families which generally have albuminous seeds, as Mr. Brown has re-
marked.

154

614 PHANEROGAMIA.

Has. Feejee, Tonga, and Samoan or Navigators’ Islands: in
lagoons, &c. A specimen is also ticketed, “Tahiti,” perhaps erro-
neously: as there is no mention of a Mangrove in Dr. Pickering’s
memoranda of the plants found at the Society Islands:.—The mucro
of the leaf is not always present.

5 BRUGUIERA, Lam.
1. Bruguiera Rumpuu, Blume.

Bruguiera Rumphii, Blume, Mus. Bot. Lugd.-Bat. p. 188, ex char.
B. gymnorhiza, Lam. et Auct. ex parte.
Mangium celsum, Rumph. Amb. 8, t. 68, ex Blume.

Has. Feejee, Tonga, and Samoan Islands: in lagoons, &c., along
with Mangroves.

The specimens plainly belong to Blume’s B. Rumphit; but whether
they are sufficiently distinct from B. gymnorhiza is not so evident.
The fruit is said to be eaten at the Feejee Islands.

Or. COMBRETACE &.

1 LUMNITZERA, Willd.

1. LuMNITZERA COccINEA, Wight & Arn.

Iumnitzera coccinea, Wight & Arn. Prodr. Fl. Ind. Or. p. 316, adn.

L. purpurea, Presl, Repert. Bot. 1, p. 155.

Laguncularia purpurea (in tab. coccinea), Gaud. Bot. Voy. Freye. p. 481, t. 4.
Pyrranthus littoreus, Jack, Malay. Misc. 2, p. 57; Wall. Cat. no. 4018.

Has. Feejee Islands; Rewa, Sandalwood Bay, &c.: on the coast
among Mangroves. Also Tonga Islands.

2 LAGUNCULARIA, Gerin.
1. LAGUNCULARIA RACEMOSA, Geertn.

Has. Brazil: in salt marshes near Rio Janeiro.

38. TERMINALIA, Linn.
1. TeRMINALIA CATAPPA, Linn.
Has. Mangsi Islands, in the Sooloo Sea. Samoan Islands. Fee-

jee and Tonga Islands (principally the var. subcordata); planted near
dwellings.

616 PHANEROGAMIA.

2. TrrminatiA Motuccana, Lam.

Terminalia Moluccana, Lam. Dict. 1, p. 349? Blume, Bijdr. p. 643; DC. Prodr. 3,
p- 11; Presl, Rel. Henk. 2, p. 21.

Has. Ovolau and Oneata, Feejee Islands. (In fruit.)

This has entirely glabrous and short-petioled, roundish-obovate
leaves, and very turgid fruit; which in one specimen is obtusely
angled, in the other, ovoid-globose and not at all angled.

3. TERMINALIA GLABRATA, Forst. ?

Terminalia glabrata, Forst. Prodr. Fl. Ins. Austr. p. 74, & Ms. deser. in Guill.
Zeph. Tait. p. 61?

Has. Feejee Islands. (Without fruit.)

Perhaps not distinct from the preceding. Forster’s characters of
the drupe may have been taken from the immature fruit.

From “a small island in the Sooloo Sea” is an imperfect fruiting
specimen of a species allied to the above, the immature drupes
oblong-ovoid, terete, acute at both ends.

4. TERMINALIA Rica, Sp. Nov.

T. foliis lanceolato-ellipticis acuminatis basi acutis eglandulosis glabratis ;
spicis floribusque sericeo-tomentosis; fructw valde immaturo elliptico
compresso.

Has. Samoan or Navigators’ Islands: on the coast of Upolu.
This, I presume, cannot be Forster’s 7: glabrata (the leaves of

which are said to be sometimes ovate-lanceolate, sometimes obovate) ;
since the common peduncle, rhachis, and flowers are silky-tomentose,

VOCHYSIACES. 617

instead of glabrous. Leaves lanceolate-elliptical, 24 to 34 inches long,
abruptly acuminate, acute at the base, glandless, glabrate, when young
sparingly and minutely pubescent, especially on the midrib under-
neath, chartaceous in texture, entire, or repand-undulate, prominently
feather-veined, somewhat lucid above: petioles 7 to 9 lines long,
glandless. Flowers sessile. The very immature frwit elliptical, com-
pressed, apiculate, canescent, half an inch long.

Orv. VOCHYSIACE &.

1 VOCHYSIA, Juss.
1. Vocuysta TucANORUM, Mart. & Zuce.
Vochysia tucanorum, Mart. & Zuce. Nov. Gen. & Spec. Bras. 1, p. 142, t. 85.
Has. Brazil, near Rio Janeiro; common on the Corcovado.
This is the V. emarginata of Pohl; but scarcely of Poiret, whose

plant is described as having merely opposite leaves, according to Vahl
only two inches in length.

156

Orv. ONAGRACE A.

Susorn. L @NOTHERE &.
1 JUSSIAA, Linn.

1. Jussr#a Peruviana, Linn.
Has. Peru; abundant in wet grounds from Callao to Yanga.

Flowers in the specimens tetramerous: hence it is J. macrocarpa,
H. B. K., a species already reduced to J. Peruviana by Bentham, in
Bot. Sulph. p. 92.

2. JUSSLHA PILOSA, H. B.A.

Jussicea pilosa, H. B. K. Nov. Gen. & Spec. 6, p. 101, t. 532; DC. Prodr. 3, p. 53.
Has. Brazil; in the Organ Mountains: also near Rio Janeiro: a
smoother form.
3. JUSSIMA REPENS, Linn.

Has. Callao and Lima, Peru. Valparaiso, Chili. In water.

4. JUSSLE@A ERECTA, Linn.

Has. Brazil; in the vicinity of Rio Janeiro.

ONAGRACEG. 619
5. JUSSIMA LONGIFOLIA, DO,
Jussicea longifolia, DC. Pl. Rar. Hort. Genev. 1, p. 12, t. 4, & Prodr. 3, p. 56.

Has. Brazil; in the vicinity of Rio Janeiro.

6. JusstmA ocToFILA, DC.
Jussiea octofila, DC. Prodr. 8, p. 57; Benth. Bot. Voy. Sulph. p. 92.
Has. Peru, near Lima. Also at Callao: a nearly glabrous variety.
The latter variety, in a still more glabrate form, likewise occurs
in the Sandwich Islands, at Oahu (but apparently is not in our
‘ collection); and Walpers, in Rel. Meyen. p. 326, has referred it to

J. octonervia, Lam.; but the ovary tapers at the base into a slender

pedicel, and is usually bracteolate. Probably the two species may be
united.

2. LUDWIGIA, Linn.
1. Lupwicta JUSSLEOIDES, Lam.

Has. Caldera, Mindanao, one of the Philippine Islands.

3. GNOTHERA, Linn.
1. GHNOTHERA MOLLISSIMA, Linn.
(nothera mollissima, Linn. (Dill. Hlth. t. 219, f. 286); DC. Prodr. 3, p. 48.

Has. Rio Negro, North Patagonia: on sand-hills.

620 PHANEROGAMIA.

2. CENOTHERA ODORATA, Jacq.

Enothera odorata, Jacq. Ic. Rar. 3, t. 456; Bot. Mag. t. 2403 ; Hook. Exot. Fl. t. 183.
. stricta, Ledeb.; Spach, Monogr. Onogr. p. 23; Gay, Fl. Chil. 2, p. 333.

Has. Chili; between Valparaiso and Santiago. Peru, near Lima.

3. CENOTHERA ALBICANS, Lam.

Gnothera albicans, Lam. Dict. 4, p. 552, & Ill. Gen. t. 270, f. 2; Spach, 1. e.
CE. nocturna, Jacq. Ie. Rar. 3, t. 455, ex Spach.
GH. prostrata, Ruiz & Pay. Fl. Peruv. & Chil. t. 315, ex Spach.

Has. Peru; on hills not far from Lima (a procumbent form with
deeply toothed leaves); and Batfios.

4, CHNOTHERA MULTICAULIS, Ruiz & Pav.

Gnothera multicaulis, Ruiz & Pav. Fl. Peruy. & Chil. 8, p. 80, t. 817; DC. Prodr.
3, p. Ol.

Has. Obrajillo, Andes of Peru.

5, CENOTHERA ACAULIS, Ouv.

Has. Chili, near Valparaiso. (With the caulescent state: (i.
grandiflora, Ruiz & Pav.)

6. GinorHEerA (GopETIA) TENUIFOLIA, Cav.

Enothera tenuifolia, Cav. Ic. 4, p. 67, t. 397; DC. Prodr. 3, p. 48.
Godetia tenuifolia, Spach, Monogr. Onogr. p. 70.

Has. Chili; between Valparaiso and Santiago.

ONAGRACE &. 621

4. MONTINIA, Linn.
1. Montinia Acris, Linn.

Has. Cape of Good Hope, near Cape Town. (With ripe fruit only.)

5. EPILOBIUM, Linn.
1. Eprtosium pArvirLorum, Schreb.

Has. Madeira; on the southern coast.

2. EPILOBIUM TETRAGONUM, Linn.

Has. Madeira, west of Funchal. Orange Harbour, Fuegia: the
var. Antarcticum, Hook. f. Andes of Chili (2. pedicellare, Presl?).
Andes of Peru, between Bafios and Culnai: along water-courses ;
nearly L. Bonplandianum, H. B. K.

For an enlarged view of the varying forms and world-wide dis-
tribution of this species, see Dr. Hooker's remarks in the Flora
Antarctica, p. 270.

3. EPILoBIUM NIVALE, Meyen.

Epilobium nivale, Meyen, Riese, 1, p. 815; Walp. Rel. Meyen. p. 327.

Has. High Andes of Peru, at Casa Cancha, &c.: procumbent in
broad tufts, in the clefts of rocks.

This accords with the character of FE. nivale, which, however,

Meyen collected on the high Andes of Chili. It is suffruticulose, and
156

622 PHANEROGAMIA.

more leafy than #. alpinum; the fleshy leaves when dry of a dark
colour; and the flowers are pedicellate.

4, EPILOBIUM NUMMULARI&FOLIUM, A. Cunn.

Epilobium nummulariefolium, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 3, p. 31;
Hook. f. Fl. N. Zeal. p. 57.
E. pedunculare & E. nerteroides, A. Cunn. |. c.

Has. Bay of Islands, New Zealand.

5. EPpmLOBIUM MICROPHYLLUM, A. Rich.

Epilobium microphyllum, A. Rich. Fl. N. Zel. p. 325, t. 36; A. Cunn. l.c.;
Hook. f. 1. ¢.

Has. Bay of Islands, New Zealand.

6. EPILopiuM ALSINOIDES, A. Cunn.
Epilobium alsinoides, thymifolium, & atriplictfolium, A. Cunn. |. ¢. ex Hook. f. 1. ¢.

Has. Bay of Islands, New Zealand :—the form called £. atriplicr-
folium by Cunningham.

7. EPILOBIUM GLABELLUM, Forst.
Epilobium glabellum, Forst. Prodr. Fl. Ins. Austr. p. 27; A. Rich. 1. c.; Hook. f. 1. ¢.

Has. Bay of Islands, New Zealand.

8. HPILOBIUM PUBENS, A. Rich.

Epilobium pubens, A. Rich. Fl. N. Zel. p. 329, t. 86, f. 1; A. Cunn. 1. ¢.; Hook. f. Fl.
N. Zeal. p. 58.

ONAGRACEA. 623

Has. In the neighbourhood of the Bay of Islands, New Zealand.

9, EPILOBIUM PALLIDIFLORUM, Solander,

Epilobium pallidiflorum, Solander, ex A. Cunn. l.c. p. 34; Hook. f. Fl. N. Zeal.
p- 59.
E. micranthum, Hook. f. in Hook. Ic. Pl. t. 297. 4

Has. Bay of Islands, New Zealand.

10. Eprtoprum suncEum, Solander.

Epilobium junceum, Solander, in Forst. Prodr. p. 90; A. Cunn. l.c.; Hook. f. 1. ¢.
E. cinereum, A. Rich. Fl. N. Zel. p. 820; A. Cunn. 1. .
E. incanum, hirtigerum, & virgatum, A. Cunn. 1. ¢. fide Hook. f.

Has. Bay of Islands, New Zealand. Hunter’s River, New South
Wales.

11. Erriosrum BILLARDIERIANUM, Seringe.

Epilobium Billardierianum, Seringe, in DC. Prodr. 3, p. 41.
E. Billardiert, Hook. f. Fl. N. Zeal. p. 61.

Has. Hunter’s River, New South Wales.

6. FUCHSIA, Plumer.
1. FucHSIA INTEGRIFOLIA, Camb.

Fuchsia integrifolia, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 273 ; Hook. Bot. Mag.
t. 3948.

Has. Brazil, in the Organ Mountains.

624 PHANEROGAMIA.

2. FUCHSIA AFFINIS, Camb. 1. ¢.

Has. With the preceding; of which it is probably only a variety.

3. FUCHSIA COCCINEA, Ait., var. MACROSTEMA.

Has. Chili: in ravines above Valparaiso:—the var. macrostema,

Hook. f. (Specimens of the typical form of this well-known Fuchsia
were gathered at Madeira, from a cultivated plant.)

4, Fucusia LycioweEs, Andr.

Fuchsia lycioides, Andr. Bot. Rep. t. 120; Sims, Bot. Mag. t. 1024.
F’. rosea, Ruiz & Pav. Fl. Peruy. & Chil. 8, p. 88.

Has, Chili: on dry hills above Valparaiso.

5. Fucusra excorticata, Linn. f.

Fuchsia excorticata, Linn. f. Suppl. p. 217; Lindl. Bot. Reg. t. 957; DC. Prodr. 3,
p. 89; Hook. f. Fl. N. Zeal. p. 56.

Skinnera excorticata, Forst. Prodr. Fl. Ins. Austr. p. 27.

Has. Bay of Islands, New Zealand. (Without either flowers or
fruit.)

Susorn, Il. HALORAGES.
7% HALORAGIS, Forst.

Hatoraais, Forst. Char. Gen. t. 31; R. Br. in Flinders’ Voy. App. p. 550; Endl.
Atakt. t. 15, & Gen. Pl. no. 6133; Hook. f. Fl. N. Zeal. p. 63.

CercopiA, Murr. in Comn. Geett. 8, t. 1; Gertn. Fruct. 1, p. 164, t. 82, & Auct.

Gonrtocarpus, Kenig, in Ann. Bot. 1, p. 546, & Auct.

Gonocarpus, Thunb. Fl. Japon. p. 5, t. 15; Geertn. Fruct. t. 225.

ONAGRACE &. 625

1. HaAoracis ALATA, Jacq.

Haloragis alata, Jacq. Ic. Rar. 1, t. 69; Hook. f. Fl. N. Zeal. p. 62.

H. Tetragonia, LW’ Her. Stirp. Nov. 1, p. 82.

H. Cercodia, Ait. Hort. Kew. 2, p. 37.

Cercodea erecta, Murr. 1. c.; DC. Prodr. 3, p. 67; A. Rich. Fl. N. Zel. p. 324.
C. erecta & C. alternifolia, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 3, p. 29.

Has. Bay of Islands, New Zealand.

This species is also found at Juan Fernandez; and has therefore a
geographical range which is surpassed only by H. micrantha.

2. HALoRAGIS SERRA, Brongn.
Haloragis Serra, Brongn. in Bot. Voy. Duperr. t. 699, sine deser.
Has. New South Wales; on Hunter’s River.

Our specimen, although too young to show the fruit, well accords
with Brongniart’s figure above-cited (of which no character is yet
published); but the teeth of the leaves are still sharper and longer.

3. HALORAGIS TEUCROIDES.

Goniocarpus teucrotdes, DC. Prodr. 3, p. 66.
Haloragis elata, Hook. f. in Lond. Jour. Bot. 6, p. 475? non A. Cunn.

Has. New South Wales; in the vicinity of “Sydney and Cook’s
River.

4. HALORAGIS TETRAGYNA, Hook. f.

58; DC. Prodr. 3, p. 66.

Gonocarpus tetragyna, Labill. Pl. N. Holl. p. t.
N. Zeal. p. 62.

39,

Haloragis (Goniocarpus) tetragyna, Hook. f. Fl.

Has. Bay of Islands, New Zealand. Sydney, New South Wales.
157

626 PHANEROGAMIA.

5. HALORAGIS DEPRESSA, Walp.

Haloragis depressa, Walp. Repert. 2, p. 99; Hook. f. Fl. N. Zeal. p. 63.
Goniocarpus depressus, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 3, p. 30.
G. serpyllifolia, Hook. f. in Hook. Ic. Pl. t. 290.

Has. Bay of Islands, New Zealand.

6. HALORAGIS MICRANTHA, /. Br.

Haloragis micrantha, R. Br. 1. ¢.; Sieb. & Zuce. Fam. Nat. Fl. Jap. (in Abhand.
Acad. Monac.), 1, p. 25.

Gonocarpus micranthus, Thunb. Fl. Jap. p. 69, t. 15.

Gontocarpus micranthus, Koenig, Ann. Bot. 1, p. 546, t. 12, f.5; DC. Prodr. 3, p. 66.

G. microcarpus, Thibaud, ex DC. 1. ¢. fide char.

G. citriodorus, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 1. ¢.

Haloragis tenella, Brongu. Bot. Voy. Duperr. t. 68 (sine deser.); Hook. f. Fl. N.
Zeal. p. 63 (non Goniocarpus tenellus, DC.).

Has. New South Wales, near Sydney and Hunter's River.

Dr. Hooker has identified his New Zealand H. tenella with the
Australian plant figured under that name (not Goniocarpus tenellus,*
as Hooker inadvertently cites it), but not yet described, by Brong-
niart, and has also detected it on the Khassya Mountains of Eastern
India. Iam enabled, by the possession of a Japanese specimen, still
further to extend this enormous range, and to pronounce it the same
as Thunberg’s Gonocarpus micranthus. Between the Australian and
the Japanese plants I cannot find the slightest difference.

7. HALORAGIS ELATA, A. Cun.
Haloragis elata, A. Cunn. ex Fenzl, in Enum. Pl. Hugel. p. 45, adn.
Has. New South Wales; on Hunter’s River.

* Goniocarpus tenellus, DC. Prodr. I. c. is a very different species, and probably only
a variety of H. tetragyna.

ONAGRACES. 627

8. HALORAGIS PINNATIFIDA, Sp. Nov.

H. caule erecto angulato foliisque pinnatipartitis hirtello-scabridis, seg-
mentis 5-7 angusto-linearibus mucronatis; floribus spicatis folioso-
bracteatis polygamo-monoicis, foeminis apetalis; ovario subgloboso
scabro.

Has. New South Wales; on Hunter’s River.

Stem erect, a foot high, sparingly branched, angled; the angles
minutely scabrous with a very short bristly pubescence, otherwise gla-
brous. Leaves opposite and alternate, 12 to 2 inches, or the rameal
half an inch long, all pinnately parted into 5 or 7 narrowly linear seg-
ments, which are a quarter or half an inch, sometimes even an inch
long, mucronate-pointed, scabrous, especially the edges, with a minute
hispidity ; the lower pair near the base; the terminal segment usually
longer than the others; all about the width of the rhachis (about half
aline). Flowers in terminal, lax, and somewhat interrupted spvkes,
the lower ones conspicuously /oliose-bracteate (the bracts resembling
the cauline leaves, but reduced in size), polygamo-moneecious, sub-
sessile, or the female flowers very short-pedicelled. Male flowers 2
lines long. Petals 4, oblong, complicate-cucullate, twice the length of
the lanceolate-triangular lobes of the calyx; the keel not appendaged,
minutely hispid-scabrous. Stamens 8. Barren stigmas 4, oblong-
conical, naked. Female flowers smaller than the male and apetalous,
occasionally with stamens: the 4 stigmas large, sessile, densely
penicillate-hispid, as long as the calyx-lobes. Ovary after anthesis
subglobose, scabrous, even, not ribbed, four-celled. Mature fruit not
seen.

This needs to be compared with Lindley’s Haloragis aspera,
which is known only by a brief character; and it is apparently still
more closely related to H. heterophylla, Brongn. That, however, is
well distinguished from the present species by the hood-like petals
with a hispid-fimbriate keel, and the pedicellate stigmas, as well as
by the foliage.

628 PHANEROGAMIA.

9, HALoraGis FILirormis, Sp. Nov.

H. scabro-hirtella; caulibus e basi ramosis strictis gracillimis; foliis
jiliformibus, inferioribus oppositis digitato-tri—quinquesectis, superio-
ribus floralibusque alternis inequaliter trifidis vel integerrimis ; flori-
bus laxe spicatis monoicis, masculis octandris, foemineis apetalis ?
ovario subgloboso scabro.

Has. New South Wales; in the vicinity of Hunter’s River.

Stems branched from the base, strict, very slender, terete, and ob-
scurely angled, from a span to near a foot high, very leafy, the
younger parts especially minutely scabrous-hispid. Leaves filiform,
scabrous-hispid like the stems; the lower opposite, half or three-quarters
of an inch in length, including the undivided base or petiole, which
is rather shorter than the 3 to 5 digitate segments; the upper gra-
dually becoming smaller and alternate, and passing into bracts,
variously trifid, sometimes with the terminal lobe longer, or mostly
entire. Spike very slender, loosely flowered. Flowers barely a line
long, subsessile, monecious: barren ones octandrous ; the 4 petals oval,
complicate-carinate, minutely hispid on the back, barely twice the
length of the triangular-lanceolate lobes of the calyx; the stigmas
rudimentary: female flowers apparently apetalous, with 4 thick and
sessile hispid stigmas. Ovary subglobose, scabrous, even, four-celled.
Mature fruit not seen.

This is abundantly distinct from any species known to me; and
evidently is not the Goniocarpus trifidus of Nees, which is dicecious
and tetrandrous, and with the lower leaves verticillate, &c.

8 MYRIOPHYLLUM, VJaiil.
1. MyrioPHYLLUM PROSERPINACOIDES, Gillies.
Myriophyllum proserpinacoides, Gillies, in Hook. Bot. Misc. 3, p. 313.

Has. Chili; near Valparaiso.

ONAGRACE &. 629

2. MyYRIOPHYLLUM VERRUCOsUM, Lindl.
Myriophyllum verrucosum, Lindl. in Mitch. Jour. Trop. Austral. p. 884.

Has. Hunter's River, New South Wales. (Submersed stems only,
without flowers or fruit.)

9 GUNNERA, Linn.
1. GunnERA CHILENsIS, Lam.

Gunnera Chilensis, Lam. Dict. 8, p. 61, & Ill. Gen. t. 801, f. 1; Bennett, Pl. Jav.
Rar. p. 75; Hook. f. Fl. Antare. p. 273; Gay, Fl. Chil. 2, p. 368.

G. scabra, Ruiz & Pav. Fl. Peruy. & Chil. 1, p. 29, t. 44.

Panke, Feuill. Obs. 2, p. 741, t. 30.

Has. Chili, near Valparaiso; in valleys along the coast.

Without doubt Dr. Hooker has rightly interpreted Mr. Brown’s
views (to which Mr. Bennett, in his article on Gunnera, above-cited,
gave no clue), in referring this genus to the Haloragee. His former
oversight in respect to the embryo is corrected in FI. N. Zeal. p. 65:
the exception mentioned, that the leaves are alternate in Gunnera,
may likewise be omitted, since they are alternate in Proserpinaca,
Meionectes, and in some species of Haloragis and Myriophyllum.

2. GUNNERA PETALOIDEA, Gaud. (Tab. 78, 79.)

G. hermaphrodita; foliis reniformi-rotundatis obsolete lobatis margine
subdentatis reticulato-rugosis subtus petiolisque hispidulis ; spicis sub-
sessilibus elongatis bracteis angustissime linearibus subtensis in rhachin
prelongam confertis ; petalis cucullatis.

Var. 8. Kauarnse: fere glabra; bracteis brevibus.
158

630 PHANEROGAMIA.

Has. Sandwich Islands. (Hawaii? Gaudichaud.) a. Maui: on
the mountains of the western division of the island, growing on the
face of high mural precipices. (In flower and with young fruit.)
@. Kauai. (Foliage and mature fruit.)

Plant apparently as large and conspicuous in its foliage as G. Ohi-
lensis, and with much more elongated and lax inflorescence. Leaves
rounded-reniform, 14 or 2 feet in diameter when full grown, obscurely
lobed, and with a more or less toothed margin, pedately ribbed, very
veiny and reticulated, the reticulations usually more or less bullate or
rugose, glabrous above, except some minute hairs on the ribs and
veins, beneath hirsute, especially on the veins and ribs, as is the stout
petiole ; in the specimen from Kauai, var. 3., nearly glabrous through-
out. The petiole in both forms is roughened with some sparse and
inconspicuous muricate points, of which traces are also observed on
the principal ribs. Inflorescence (with the short scapoid peduncle) at
least 2 or 3 feet long; the spikes crowded, but lax and spreading,
subsessile, 3 to 6 inches long; the main rhachis hirsute, in var. £2.
glabrate or glabrous. Bracts narrowly linear, 6 to 9 lines long and
barely a line wide; in var. @. apparently ovate or oblong, and only 3
or 4 lines long. Flowers sessile, crowded, and somewhat glomerate
on the rhachis of the spikes, not bracteolate, glabrous. Tube of the
calyx adnate, globular; the lobes 2, anterior and posterior, persistent,
broadly ovate-triangular, thick, the thin edge denticulate at the broad
or truncate apex. tals 2, cucullate, enclosing the stamens before
anthesis, broadly obovate or cuneate, retuse, obscurely glandular on
the back, thickish in texture, alternate with the lobes of the calyx,
twice or thrice their length, epigynous, tardily deciduous. Stamens
2, epigynous, opposite the petals: filaments very short: anthers
large, almost a line long, two-celled, emarginate at each end, some-
what didymous, fixed by the base, introrse, the cells opening longi-
tudinally. Pollen four-lobed, or of 4 united grains. Stigmas 2, oppo-
site the stamens, and nearly twice their length, linear-subulate,
hispid, slightly united at the base. Ovary one-celled, with a single
anatropous ovule, suspended from the summit of the cell. Drupes
ovoid-globose, apparently red or purple, 13 or 2 lines long, tipped
with the short and incurved lobes of the calyx, the tube of which
forms the fleshy sarcocarp: putamen small in proportion, acheniform,
lenticular, and somewhat quadrangular or trigonous, smooth, crusta-

ONAGRACE AZ. 631

ceous. Seed conformed to the putamen; the testa very thin and
delicate. Embryo minute, near the hilar extremity of the fleshy and
oily albumen, cordate, the radicle superior.

The smooth variety, of which we have only a full-grown leaf and
mature fruit, requires farther investigation ; but it will probably prove
to be only a form of Gaudichaud’s G. petaloidea. I am unable to com-
pare the species with the G. bracteata, from Juan Fernandez.

PLATE 78.—GUNNERA PETALOIDEA. Fig. 1. A young leaf. 2. A
full-grown leaf. 3. Summit of the inflorescence, in flower. 4. A
small portion of the inflorescence, in fruit— All of the natural size.

Pratt 79, A.—Fig. 1. Diagram of a flower. 2. A flower, viewed
anteriorly. 3. The same, with the petals and stamens removed, and
the ovary vertically divided. 4. A petal. 5. A stamen. 6. Pollen.
7. Transverse section of a drupe. 8. Vertical section of a drupe,
seed, &c., showing the embryo in place. 9. Putamen, detached. 10.
Embryo, detached.—Variously magnified.

3. GuNNERA (MisanpraA) MAGELianica, Lam.

Gunnera Magellanica, Lam. Dict. 3, p. 61, & Ill. Gen. t. 801, f. 2; Hook. f. Fl.
Antare. p. 274.

G. Falklandica, Hook. Ic. Pl. t. 489.

Misandra Magellanica, Commerson, in Juss. Gen. p. 405; Gaud. Bot. Freyc. Voy.
p. 502.

Has. Orange Harbour, Fuegia; common.
Various forms were collected, both in flower and in fruit, either
hairy or almost glabrous; the scapes varying from 3 lines to as many

inches in length; the petioles from half an inch to 10 inches long;
the blade of the leaves from 8 lines to 4% inches in diameter.

4, GUNNERA (Misanpra) LopaTA, Hook. f. (Tab. 79.)

Gunnera (Misandra) lobata, Hook. f. Fl. Antarc. p. 279.
Dysemone lobata, Banks & Solander, Ic. & Ms. in Herb. Banks.

632 PHANEROGAMIA.

Has. Fuegia; on mountains at Orange Harbour: somewhat rare.

A small and slender plant; the sterile stems creeping and rooting,
glabrous, squamose; the flowering ones shorter and thicker root-
stocks; dicecious: the female plant apparently smaller than the
males. Petioles 3 to 15 lines long, strigosely hirsute with erect
reddish hairs. Leaves orbicular or round-reniform, 3 to 9 lines in
diameter, 5--7-lobed, or on sterile shoots crenately 9-15-lobed, sub-
coriaceous, glabrous, sometimes ciliate, the lobes roundish, entire.
Stipular scales ferrugineous. Spikes simple, few-flowered; the male
raised on a slender scape of 6 to 18 lines in length; the female short
and subsessile. Male flowers 8 to 12, crowded, or at length sparse,
obscurely pedicellate or nearly sessile on the rhachis, not bracteate.
Calyx of 2 filiform-linear glabrous sepals, placed anterior and poste-
rior, much smaller than the naked anthers, which they at no period
enclose, glandular at the tip. Petals none. Stamens 2, alternate
with the sepals: anthers broadly oval, somewhat didymous, two-
celled, nearly a line long, twice the length of their filaments, turgid,
mucronulate, opening longitudinally along the margins. Female
flowers 8 or 10, capitate, ebracteate, sessile. Tube of the calyx sub-
globose, adnate to the ovary; the lobes 2, short and obtuse, oval,
often unequal, sometimes lateral as respects the axis of inflorescence,
persistent. Petals wanting. Stigmas 2, subulate, elongated, 2 or 3
lines in length, purple, hispid, alternate with the lobes of the calyx,
slightly united at the base. Ovary one-celled. Ovule anatropous,
suspended. Fruit not seen. (The fruit excepted, our specimens
enable me to complete the characters of this rare species of an inte-
resting genus.)

PLATE 79, B—GUNNERA LOBATA: male and female plants, and a
sterile shoot, of the natural size. Fig. 1. Portion of the male inflo-
rescence, magnified. 2. Female inflorescence, magnified. 3. Flower,
from the same, more magnified; the ovary divided transversely. 4.
Flower, with the ovary vertically divided.

Orv. LOASACE A.

1 MENTZELILA, Linn.

1. MentzetA SCABRA, H. B.K.

Mentzelia scabra, H. B. K. Nov. Gen. & Spec. 6, p. 120; DC. Prodr. 3, p. 343.
M. aspera, Hook. Bot. Misc. 2, p. 220, non Linn.

Has. Peru; between Caballeros and Obrajillo.

The stamens are about 75 in number; five of them with dilated
filaments.

2. LOASA, Adans.
1. LoASaA INCANA, Graham.

Loasa incana, Graham, in Hdinb. Phil. Jour. 1830; Hook. Bot. Mag. t. 3048, &
Bot. Misc. 2, p. 220.
L. atriplicifolia, Pres], Symb. Bot. 1, p. 61, t. 39.

Haz. Peru; common from Yaso to Bajios.

9, LOASA GRANDIFLORA, Lam.

Loasa grandiflora, Lam. Dict. 8, p. 580; Juss. in Ann. Mus. 5, p. 26, t. 4, f. 2.

Has. Andes of Peru, at Batis.
159

634 PHANEROGAMIA.

3. LoaAsSA ALDUNATEA, Gay.
Loasa Aldunatea, Gay, Fl. Chil. 2, p. 442.

Has. Chili; probably in the Andes, above Santiago. (The ticket
lost.)

4, LOASA HETEROPHYLLA, Hook. & Arn.

Loasa heterophylla, Hook. & Arn. in Bot. Misc. 8, p. 328; Gay, 1. c.

Has. Chili, near Valparaiso.

5. LOASA PINNATIFIDA, Gillies.

Loasa pinnatifida, Gillies, ex Arn. in Edinb. Jour. Geogr. Sci. 1, p. 275; Hook. &
Arn. l.c.

Has. Andes of Chili, above Santiago, near the snow-line.

d& CAILOPHORA, Presi.
1. CAIOPHORA CORONATA, Hook. & Arn.

Catophora coronata, Hook. & Arn. Bot. Misc. 8, p. 327; Gay, Fl. Chil. 2, p. 487.
C. absinthiifolia, Presl, Rel. Heenk. 2, p. 274.
Loasa coronata, Gillies, ex Arn. in Edinb. Jour. Geogr. Sci. 1, p. 273 (1831).

Has. Andes of Chili, above Santiago.

2. CAIOPHORA CARDUIFOLIA, Presi, 1. c.

Has. Andes of Peru, near Bafios, &. (Also collected by Mat-
thews, M’Lean, &c.)

MALESHERBIACES. 635

3. CAIOPHORA CONTORTA, Presi, 1. ¢.

Loasa contorta, Lam. Dict. 3, p. 579, & Ill. Gen. t. 426; Juss. in Ann. Mus. 5, t.
3, f. 1; DC. Prodr. 8, p. 340; Hook. Bot. Misc. 2, p. 220.

Has. Andes of Peru, near Obrajillo, &c.

Or. MALESHERBIACE/.

1 MALESHERBIA, Ruz & Pa.
1. MALESHERBIA THYRSIFLORA, Ruiz & Pav.

Malesherbia thyrsiflora, Ruiz & Pav. Fl. Peruv. & Chil. 3, t. 254; DC. Prodr. 3, p.
338 ; Hook. Bot. Misc. 2, p. 220.
Gynopleura tubulosa, Cav. Ic. Pl. 4, p. 72, t. 375.

Has. Peru; from Yaso to Obrajillo.

Orv. TURNERACE &.

1. TURNERA, Plumier. |

1. TURNERA CUNEIFORMIS, Juss.

Turnera cunetformis, Juss. in Poir. Dict. 8, p. 142; DO. Prodr. 3, p. 346; Cambess.
in St. Hil. Fl. Bras. Mer. 2, p. 218. |

Has. Brazil; very common around Rio Janeiro. (Scarcely dis-

tinct from 7. ulmifolia.)

2. TURNERA SALICIFOLIA, Camb.
Turnera salictfolia, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 227.

Has. Brazil: in the vicinity of Rio Janeiro; not uncommon in
forests.

Orvp. PASSIFLORACE.

1. PASSIFLORA, Linn.

1. PASsIFLORA (TETRAPATHZA) TETRANDRA, Banks & Soland.

Passijlora tetrandra, Banks & Soland. Ic. ined.; DC. Prodr. 3, p. 323; Hook. f. FI.
N. Zeal. p. 73.
Tetrapathea australis, Raoul, Pl. N. Zel. p. 27, t. 27.

Has. Waia-ruru Bay, New Zealand. (In fruit.)

2. PASSIFLORA LITTORALIS, H. B. Kt.

Passiflora littoralis, H. B. K. Nov. Gen. & Spec. 2, p. 188; Hook. Bot. Mise. 2,
p. 219.

Has. Peru, at Callao and Lima.
The glands of the petiole are sessile or nearly so, as noted by
Hooker.
3. PASSIFLORA ALBIDA, Jer.
Passiflora albida, Ker, Bot. Reg. t. 667; DC. Prodr. 3, p. 328.

Has. Brazil, near Rio Janeiro.
160

638 PHANEROGAMIA.

4, PASSIFLORA ELLIPTICA, Gardn.
Passiflora elliptica, Gardn. in Hook. Lond. Jour. Bot. 1, p. 1738.

Has. Brazil; on the Corcovado, near Rio Janeiro.

5, PASSIFLORA PILIGERA, Gardn. 1. e.

Passiflora bilobata, Velloz. Fl. Flum. 9, t. 78, fide Gardn.

Has. Brazil, near Rio Janeiro.

6. PASSIFLORA FaTIDA, Linn.

Passiflora fetida, Linn.; Cav. Diss. 10, t. 289; Sims, Bot. Mag. t. 2619.
P. hibiscifolia, Lam. Dict. 5, p. 89; DC. Prodr. 8, p. 331.

Has. Peru; on the coast, near Callao. Singapore: doubtless
introduced from America.

There is also in the collection the foliage of a species of Pussijlora,

to me unknown, gathered near Baiios, Luzon, and on an island in the
Sooloo Sea.

2, DISEMMA, Labill.
1. DisemMA HERBERTIANA, DC.

Disemma Herbertiana, DC. Prodr. 3, p. 332.
Passiftora Herbertiana, Ker. Bot. Reg. t. 737.

Has. New South Wales: probably from the vicinity of Sydney.

PASSIFLORACES. 639

38. TACSONTA, Juss.

1. TACSONIA TRIFOLIATA, Juss.

Tacsonia trifoliata, Juss. in Ann. Mus. 6, p. 893; DC. Prodr. 3, p. 334; Hook.
Bot. Misc. 2, p. 219.
Passiflora trifoliata, Cay. Ic. Pl. 5, p. 16, t. 427.

Has. Andes of Peru; between Obrajillo and Culnai; also near
Banos.

A species with showy red flowers. Both DeCandolle and Hooker,
above cited, have omitted all reference to the figure of Cavanilles.

2. TACSONIA PEDUNCULARIS, Juss. l. c.
Passiflora peduncularis, Cav. Ic. Pl. 5, p. 15, t. 426.

Has. Andes of Peru, near Bajios. (Also gathered by Matthews,
M’Lean, &c.)

There is likewise some foliage (without flowers or fruit) of a third
species, from the same district, with small and trisected leaves;
which I cannot identify. It is apparently allied to a Tacsonia in the
Peruvian collection of Matthews, of which the characters are sub-

joined.*

* TACSONIA GRACILENS (Sp. Nov.): glabra; caule gracili; stipulis setacets ; foliis
parvis tripartitis trisectisve, segmentis (6-12 lin. longis) oblongis obtusis penninervirs
obsolete serrulatis, serraturis subglandulosis, petiolo versus apicem uni—-triglandulosis ;
involucro e bracteis 8 linearibus parvis discretis ; floribus (pollicaribus) solitar its ; Fructu
ovali.—Huanuco, Andes of Peru, Matthews: no. 915, in herb. Hook.—The single
flower did not permit me to determine if it presented the structure of the section Diste-

phana.

Orv PAPAYACE &.

1 CARICA, Linn.
1. Carica Papaya, Linn.

Has. Tahiti; on Point Venus. Cultivated in the Society, as also
in the Samoan, Tonga, Feejee, and Sandwich Islands; where the
fruit of the Papau is a favourite esculent.

2. CARICA CANDICANS, Sp. Nov.

C. caule parce ramoso; foliis ovatis subcordatis integris nunc repandis
vel angulato-dentatis e basi trinerviis subtus adpresse cano-tomentosis.

Has. Peru: in ravines of the Amancaes Mountains, between Lima
and Obrajillo.

“Trunk a foot in diameter, 6 to 10 feet high, with a few coarse
branches: leaves entire, white-tomentose beneath.” — Pickering, Ms.
This is very different from any Carica published or known in the
herbaria I have consulted; unless it should prove to be a state of C.
pyriformis, Willd. (but not of Hook. & Arn. and Gay), founded on
the Papaya ramosa, etc. of Feuillée, which he found in a garden at
Lima; but which has palmately nine-parted leaves, and there is no
mention of pubescence or wool on their lower surface. The leaves of
our plant are ovate or ovate-rotund, with a usually slight subcordate
base, and an obtuse or pointed apex, the margin entire or merely
repand, or in a single and oblong leaf sinwately-dentate into 3 or 4

CUCURBITACEA. 641

coarse teeth on each side; the base three-ribbed, sometimes with an
additional pair of basal veins; the ribs branching into coarse veins
and reticulated veinlets. The younger leaves, of recent branches, are
only 13 or 2 inches long; the largest gathered is 9 inches long; and
this equally retains on its lower surface the soft and fine, appressed,
whitish tomentum, which, however, may be rubbed off; the upper
surface glabrous. The male inflorescence consists of small and
crowded many-flowered cymes, on peduncles of an inch or two in
length: the pedicels scarcely any. Calyx minute, seven-toothed, or
sometimes five-toothed. Corolla salver-shaped; the tube half an
inch in length; the oblong-linear lobes 3 lines long, varying from 5
to 7 in number. Stamens twice as many as the lobes of the corolla,
distinct: anthers linear-oblong; those alternate with the lobes on
filaments of about their own length; those opposite them subsessile.
Pistil rudimentary. Female flowers not gathered. But there is a
loose fructified ovary, an inch and a half long, ovoid, pointed, and
with a narrowed base.

Oro. CUCURBITACE A.

1 MELOTHRIA, Linn.
1. MetoruriA FLUMINENSIS, Gardn.
Melothria Fluminensis, Gardn. in Hook. Lond. Jour. Bot. 2, p. 173.

Has. Brazil; common around Rio Janeiro.

2. MeLoruriA SAMOENSIS, Sp. Nov.

M. glabra; foliis deltoideo-cordatis acuminatis repandis denticulatis ;
floribus dioicis 2 in axillis solitariis ; pedunculis filiformibus elongatis ;
161

642 PHANEROGAMIA.

corolle lobis integerrimis; calycis tubo ultra ovarium fusiforme pro-
ducto; bacca oblonga.

- Has. Samoan or Navigators’ Islands; common on Tutuila and
Upolu.

Stem slender, glabrous, as is the whole plant, angled, creeping.
Leaves deltoid-cordate, sometimes with the broad sinus rounded, some-
times approaching to deltoid-hastate, from 2 to 84 inches long, and 2
or 3 inches broad at the base, acuminate, membranaceous, nearly
smooth, the margins denticulate, towards the base often repand, or
obscurely angulate-toothed. Petioles about an inch long. Tendrils
simple. Flowers apparently dicecious (at least the two sexes occupy
different branches), “white,” solitary in the axils of the leaves, on
jiliform peduncles, of which the fertile are an inch or more, the sterile
2 or 3 inches in length. Calyx of the male flowers campanulate,
with 5 setaceous teeth at the sinuses of the five-parted corolla; the
divisions of which are oblong, or two of them lanceolate, 2 or 3 lines
long, perfectly entire, and glabrous. Stamens 3, shorter than the
petals: filaments short, inserted on the tube of the calyx, distinct :
anthers two-celled; the cells straight or nearly so, borne on the
margin of the dilated, subcuneate-oval, membranaceous connective,
the truncate summit of which is slightly and very obtusely apiculate :
one of the three stamens has a narrower, barely oblong connective,
but is otherwise similar. A short, obscurely three-lobed gland takes
the place of the style. Calyx of the female flowers 6 or 7 lines long,
including the ovary (which is 3 or 4 lines long); the tube produced
beyond the fusiform ovary into a short and slender beak, which is
abruptly dilated into a cylindraceous-campanulate throat; the lobes
very minute. Corolla as in the male flower. Style as long as the
calyx: stigma dilated, three-lobed. Berry oblong, nearly an inch in
length, not pointed, many-seeded. Seeds flat, smooth, about 2 lines
long.

2, KARIVIA, Arn.

Karryia, Arn. in Hook. Jour. Bot. 3, p. 275; Wight, Ic. Pl. 2, t. 502.

CUCURBITACESR 643

1. Karivia Samoensis, Sp. Nov.

K. dioica, glabra; cirrhis simplicibus; foliis cordatis acuminatis denti-
culatis modice petiolatis; floribus masculis umbellatis vel corymbosis e
pedunculo brevissimo, foemineis solitariis ternisve, pedunculo bacca
ovoidea polysperma breviore; seminibus leevibus compressis.

Var. @. VITIENSIS: j/oliis supra scabridis; pedunculis fructiferis brevis-
sums.

Has. Samoan Islands; on Manua and Upolu. Var.? 8. Ovolau,
Feejee Islands.

A slender, extensively climbing, apparently diwcious, glabrous vine,
not unlike A. umbellata, Arn., in habit, not glaucescent. Tendrils
simple, Leaves cordate, with rounded lobes and a deep sinus, more or
less acuminate, denticulate, sometimes a little repand-toothed, membra-
naceous, smooth both sides, from 2 to 4 inches long, rather conspi-
cuously petioled: the petiole of the larger leaves an inch or two in
length and exceeding the sinus; those of the smaller upper leaves
only half an inch long, and scarcely equalling the sinus. Flowers
axillary: only the sterile ones occur in the specimens, and very
sparingly. Male flowers in small wmbels or corymbs; the common
peduncle short or wanting; the pedicels 2 or 3 lines long. Calyx
hemispherical or slightly campanulate, with 5 short and subulate
teeth. Petals 5, ovate, acute, inserted by a broad base on the
summit of the calyx-tube, scarcely exceeding its lobes. Stamens 3,
distinct, inserted in the base of the calyx, included: anthers linear-
oblong, somewhat sagittate, longer than the filaments, two-celled,
similar, or perhaps one of them a little narrower than the others:
the cells straight, linear, more or less pubescent at the junction with
the narrow connective inside. A short, three-lobed gland takes the
place of the gynecium. Female flowers not seen. Berry ovoid,
resembling that of a Melothria, half an inch long, on a peduncle or 4
or 5, or sometimes only 2 lines long, many-seeded. Seeds closely
packed in 3 vertical rows, ovate, compressed, apparently flat, smooth
and even, whitish, with an abrupt but not margined edge.

644 PHANEROGAMIA.

A very imperfect fruiting specimen, from the Feejee Islands, appa-
rently belongs to this species; although the leaves are somewhat sca-
brous above, and the fructiferous peduncles very short.

There are female specimens apparently of Kurivia umbellata in the
collection, ticketed “Bay of Islands:”’ but there is probably some
mistake about the habitat, as neither Cunningham, Raoul, nor Hooker
have noticed the plant in New Zealand.

3. LUFFA, Cav.
1. Lurra insuLarum, Sp. Nov.

L. caule sulcato glabro; foliis rotundis quinquelobatis vel quinquangu-
latis sinu profundo cordatis margine denticulatis seu repando-dentatis ;
racemis masculis elongatis; staminibus discretis ; calycis lobis fl. foem.
subtus glanduliferis ; fructu oblongo levi haud sulcato bipollicari.

Has. Samoan or Navigators’ Islands; on Savaii and Upolu: climb-
ing trees. Tongatabu, Friendly Islands. Muthuata, Feejee Islands.

This appears to be a truly indigenous and undescribed species, with
rounder and less dentate leaves than L. fetida and L. acutangula,
and a very small and even fruit. It is perhaps dicecious. Stem and
branches glabrous, sometimes scabrous, strongly grooved. Leaves
rotund, sometimes inclining to reniform, 3 to 5 inches in diameter,
deeply cordate with a narrow or closed sinus, five-angled, or sometimes
rather deeply jive-lobed, scabrous both sides, or sometimes almost
smooth ; the lobes or angles rounded, or the terminal one occasionally
acuminate and a little prolonged, denticulate, or barely repand-toothed.
Tendrils two-cleft, or sometimes three-cleft. Male racemes elongated
(a span to a foot long, including the peduncle). Pedicels shorter
than the calyx; the lobes of which are ovate-lanceolate and gland-
less; the petals (“ yellow,” according to Dr. Pickering’s notes, but
seeming to have been white) obovate, entire, an inch or more in
length. Stamens 5, distinct: filaments short: anthers sinuous-con-
torted, with remote convolutions. Female flowers solitary in the

CUCURBITACEAE, 645

axils, short peduncled: the lanceolate-acuminate calyx-lobes under-
neath beset with large glands. Fruit oblong, obtuse at both ends,
smooth, not grooved nor ribbed, only 2 inches long, with a thin and
yellowish rind when dry, sparing fibrose-reticulated within. Seeds
oval, compressed, but tumid, brown, with a nearly smooth surface;
the edge very slightly margined. |

4. TRICHOSANTHES, Linn. Arn.
1. TricHosaNTHES (INVOLUCRARIA) QUINQUANGULATA, Sp. Nov.

T. glabra; folis (magnis) cordato-rotundis sinu profundo angulato-
quinquelobis, lobis acuminatis subintegerrimis; cirrhis quinquefidis ;
Hloribus masculis racemosis ; bracteis foliaceis magnis obovatis dentatis
pedicello gracili brevioribus ; calycis segmentis lineari-lanceolatis pin-
natrfido-lacimiatis petala late cuneata longe fimbriata subequantibus.

Has. Mangsi Islands, in the Sooloo Sea.

Apparently an extensively climbing vine; with an angled stem,
smooth and glabrous, or nearly so, as is the whole plant. Leaves
ample, 7 to 9 inches in diameter, thin and membranaceous, three-
ribbed from the base, the lateral ribs forked, subrotund, cordate with a
deep and narrow sinus, angularly five-lobed ; the lobes broadly trian-
gular, 12 to 2, or the terminal one 3 inches long, acuminate, nearly
entire, barely repand, or the basal lobes usually with a few coarse
teeth posteriorly. Tendrils five-cleft, or four-cleft. Male peduncle
axillary, longer than the petiole, strongly angled, bearing a many-
flowered raceme; the flowers subtended, and as it were involucrated,
with large, obovate, glandless, more or less toothed, foliaceous bracts, of
half an inch or an inch in length, but shorter than the slender pedicels
of the developed flowers, which are from an inch to an inch and a
half in length. Tube of the (male) calyx clavate-turbinate, half an
inch long; its 5 segments linear-lanceolate, almost an inch in length,
remotely pinnatifid-laciniate, nearly the length of the apparently white,
broadly cuneate petals. These are retuse, with a central apiculation,

the broad summit on each side laciniated into a copious, long, and
162

646 PHANEROGAMIA.

delicate fringe. Filaments 3, short: anthers united into a cylindrical
mass. Female flowers and fruit not seen.

This is evidently a congener of 7. palmata, Roxb., and of 7. tricus-

pidata, Lour.; from which the leaves and the long pedicels distin-
guish it. I cannot identify it with any described species.

5 MOMORDICA, Linn.
1. Momorpica CHARAntTIA, Linn.
Has. Luzon; at Banos, Laguna, &. Point Venus, Tahiti: doubt-

less recently introduced.

2. Momorpica SENEGALENSIS, Lam.
Momordica Senegalensis, Lam. Dict. 4, p. 239; DC. Prodr. 8, p. 311.

Has. St. Jago, Cape de Verde Islands. Rio Janeiro, Brazil.

6 CUCUMIS, Linn.

1. CucuMIS PUBESCENS, Willd.

Cucumis pubescens, Willd. Spec. Pl. 4, p. 614; DC. Prodr. 3, p. 301; Wight & Arn.
Prodr. Fl. Ind. Or. 1, p. 342.

Var.? 8. fructu subgloboso mox glabrato (via pollicari).
Var.? y. fructu glaberrimo ovali (sesquipollicari).
Cucumis bicirrha, Forst. Mss. ex Guill. Zeph. Tait. p. 56?

Has. St. Jago, Cape de Verde Islands: in wild and rocky situa-

CUCURBITACE€Z. 647

tions. (Without fruit.) Var.? 3. Feejee Islands; on Muthuata, &c.
Var.? y. Tahiti and Matia, Society Islands; and Tongatabu.

I find nothing to distinguish the specimens of the South Sea
Islands from C. pubescens, except the smooth fruit; which, in the
form collected at the Feejee Islands still shows traces of the pubes-
cence that clothes the ovary: the fruit appears to be globular and
scarcely an inch in diameter. In the Tahitian form, the fruit is
larger, oval, fully an inch and a half long, smooth and even, with no
trace of pubescence. ‘The short lobes of the leaves are only minutely
denticulate. This would seem to be Forster’s Cucumis bicirrha (of
which no specimens are known to be extant); but the tendrils are all
simple.

7 CYCLANTHERA, Schrad.
1. CycLANTHERA MatrHewsil, Arn.

C. subglabra; caule gracillimo; foliis pedatim quinquepartitis sew tri-
partitis segmentis lateralibus subbilobis, intermedio productiore, omnt-
bus oblongis margine repandis ; cirrhis bifidis ; racemo masculo gracili
sumplicissimo; pedunculo floris foeminet fructu oblique ovato oligo-
spermo dimidio breviore.

Cyclanthera Matthewsii, Arn. in Hook. Jour. Bot. 3, p. 280, absque char.
Has. Peru, in the vicinity of Obrajillo.

A very slender plant, glabrous, or nearly so: with pedately five-
parted leaves, or sometimes only three-parted, with the lateral divisions
slightly two-lobed or nearly entire; the segments oblong, obtuse or
slightly pointed, about an inch long, with vepand or obscurely crenu-
late margins; the middle one longer than the others. TZendrils two-
cleft, elongated. Male flowers in a simple and slender raceme, about
the length of the leaves; the flowers barely a line in diameter, on
pedicels half a line long, which are not clustered. Female flowers
solitary in the same axils with the male flowers, on a peduncle of
only a line and a half long. Fruit obliquely ovate, 5 or 6 lines long,

648 PHANEROGAMIA.

thickly beset with soft and barbless prickles, about five-sceded. Seeds
coarsely tuberculate.

8. SICYOS, Linn.

§ 1. EUSICYOS.—Fructus compressus vel turgidus, haud angulatus, setis retrorsum
scabris armatus, rariusve inermis; pericarpio tenui. Antheroe 8-5, sinuose, capt-
tato-connate.

1. Sicyos AUSTRALIS, Endl.

Sicyos australis, Endl. Prodr. Fl. Norf. p. 67.
S. angulatus, Forgt. Prodr. Ins. Austr. p. 68; Hook. f. Fl. N. Zeal. p. 72, non Linn.
S. Fretensis, Hook. f. in Lond. Jour. Bot. 6, p. 478.

Has. Bay of Islands, New Zealand. New South Wales: probably
near Sydney.

Both the male flowers and the fruit are less than half the size of
those of S. angulatus, being no larger than in S. parviflorus; which,
however, has a different foliage.

2. SICYOS LA&VIS, Sp. Nov.

S. glaber ; foliis cordatis brevissime petiolatis vix lobatis ; racemo masculo
sumplict brevi; fructu ovato glaberrimo inermi.

Has. Andes of Peru, in the vicinity of Obrajillo.

Only a single and small specimen of this plant occurs in the collec-
tion, with male flowers and with immature fruit. The stem is
slender, almost filiform, glabrous, as is the whole plant, except a slight
hairiness on the younger foliage. Leaves cordate with a deep sinus,
which exceeds the comparatively very short petiole, so that in the
specimen they appear somewhat as if clasping; they are 13 to 2
inches long (the petiole 2 or 3 lines long), ovate in outline, cuspidate-
acuminate, nearly entire, or slightly five-lobed or angulate-toothed,

CUCURBITACES®. 649

denticulate, thin and membranaceous, sparsely roughened on both
sides with small papille. Flowers moncecious; both kinds often pro-
duced from the same axils. Male raceme simple, short and rather few-
flowered, with the peduncle scarcely equalling the leaves; the flowers
only 2 lines in diameter, of the same structure as in 8. parviflorus:
the united filaments forming a filiform column nearly as long as the
corolla. Anthers apparently only 3, connate into a sinuous mass.
Fertile peduncle 2 lines long, probably bearing a number of female
flowers in a head, as there are 3 or 4 fructified ovaries or young
fruits: these are ovate, very glabrous, wholly unarmed and naked,
pointed but not beaked, somewhat compressed, 2 or 3 lines long.
Ovule and the half-grown seed solitary from the summit of the single
cell.

§2. SICYOCARYA.—Fructus ovato-pyramidatus vel oblongus, 4—-6-angulatus (raris-
sime trigueter), tnermis, pl. m. rostratus; pericarpio incrassato. Anthere 2-5,
sinuosce, bast connate ; connectivo anyusto.

Not being disposed to found a genus upon the form and texture of
the fruit, I characterize the following Hawaiian species under a dis-
tinct section or subgenus of Sicyos. One of them, S. pachycarpus, has
already been published: but its mature fruit was not known. The
second has the same habit as well as structure. The third, S. cucu-
merinus, would never by itself be taken for a Sicyos; yet it appears
to be a strict congener of the two others. Sicyos triqueter of Mogino
& Sesse, known to DeCandolle only by a drawing, appears also to
belong to this group. The filaments, as in true Sicyos, are perfectly
united to the summit; the cells of the anthers, which are only
slightly sinuous, are separated by a somewhat dilated and plane,
although narrow connective; indicating that the anomalous S. mon-
tanus, of Poeppig and Endlicher, should perhaps be retained as a
third section, rather than excluded to form the type of a new Sicyoid
genus.* The fruit of the latter is said to be obtusely pentagonal, but it
is not so delineated in the plate.

* §3. SICYOPSIS.—Fructus obovatus, turgidus, inermis, hirsutus, infra apicem
obtusissimum dentibus calycinis subulatis deflexis coronatus; “ pericarpio baccato.”’
Columna staminum extremo apice trifida. Antheree 8 libere, seu 5 triadelphee; connec-
tivo dilatato plano utringue emarginato,

S. monranus, Popp. & Endl. Nov. Gen. & Spee. 2, p. 53, t. 172.
163

650 PHANEROGAMIA.

3. Sicyos (Stcyocarya) pacHycaRPuS. (Tab. 80.)

S. foliis cordato-rotundis angulato-sublobatis denticulatis subtus papilloso-
scabridis seu novellis hispidulis; paniculis masculis subsimplicibus ;
Fructu ovato-pyramidato 5—-6-quetro glabrato, uniore tenurter rostrato.

Sicyos pachycarpus, Hook. & Arn. Bot. Beech. Voy. p. 83.

Has. Oahu, Sandwich Islands; on the Kaala Mountains, in the
district of Waianae. (Also found by Lay & Collie, Gaudichaud, &c.)
Maui; on the mountains in the western part of that island.

An herbaceous vine (the root unknown), with much the habit of
S. angulatus. Stems slender, strongly angled, nearly glabrous or gla-
brate, sometimes at first beset with slender and glandular-tipped
hairs. Leaves thin and membranaceous, 3 to 5 inches in diameter,
rounded and cordate-angled, sometimes inclining to reniform and very
obscurely lobed, more commonly angularly jive-lobed (or seven-lobed ?) ;
the lobes very short and triangular, and the sinuses very broad, the
terminal lobe largest; the margin barely denticulate; the upper sur-
face glabrous, or sparsely papillose-scabrous; the lower sometimes
hispid-pubescent when young, soon glabrate, or more or less papillose-
scabrous. Petioles one or 2 inches long, glabrate or glandular-pubes-
cent. Tendrils 2—3-cleft. Flowers of both kinds usually from the
same axils; the male flowers small (the buds only a line in diameter),
in racemose, rather small, and stmple panicles, on peduncles of one to
3 inches long, which, with the short pedicels, are glandular-pubescent.
Our specimens, however, do not well exhibit the male inflorescence.
Perianth, as in the genus, accrescent after expanding, becoming nearly
3 lines in diameter, at length rotate, five-cleft to the middle, with 5
minute and subulate calyx-teeth at the sinuses; the lobes of the
corolla triangular-ovate, sparingly pubescent externally. Anthers 5,
or sometimes only 4 or 3, collected in a head at the summit of the
slender column formed of the united filaments, distinct but sessile,
oblong, two-celled, the cells parallel and margining a narrow (or at
first roundish) connective, or else one of them one-celled, sinuous, but
only moderately so, and capable of being straightened when soaked.
Female flowers numerous in a capitulum, terminating a short peduncle

CUCURBITACE SA. 651

of a line or two in length, or in fruit becoming 3 to 6 lines long.
Perianth pubescent; the tube produced beyond the fusiform-ovate
ovary into a narrow neck or short beak, the dilated-limb like that of
the male flowers but smaller. Stigmas 3, clavate-capitate, subsessile.
Ovary one-celled, with a single suspended ovule. Fruit ovate-pyra-
midal, several usually ripening in the head, closely sessile on the
summit of the short common peduncle, 5 to 7 lines long, when young
puberulent, at length glabrate, wholly unarmed and smooth, five-six-
angled, the angles salient and sharp, pointed with a slender beak of 2
or 3 lines in length, which usually withers away before maturity :
in the dry state the fruit appears nut-like, the pericarp being thick
and hard, nearly crustaceous, one-celled, one-seeded. Seed obovate,
turgid; the testa roughened, thick and coriaceous. Embryo amygda-
loid: the cotyledons thicker than in 8. angulatus. Radicle short,
superior.

PLATE 80.—Sicyos (SICYOCARYA) PACHYCARPUS: with flowers and
fruit, of the natural size. Fig. 1. A male flower-bud. 2. Androecium.
3. Lateral view of one of the anthers. 4. An anther spread out, to
show the connective. 5. A female flower. 6. A fruit, of the natural
size. 7. A longitudinal section of the same. 8. A transverse section
of the same. 9. A seed. 10. Vertical section of the same.—The
details magnified, except Figs. 6-8.

4. Sicyos (SIcYOCARYA) MACROPHYLLUS, Sp. Nov. (Tab. 81.)

S. folits magnis cordato-rotundis tri—quinquelobatis argute denticulatis
subtus puberulis; paniculis masculis umbellato-compositis longe pedun-
culatis, pedicellis filiformibus; fructu ovato 5~-6-angulato glabrato
rostrato.

Has. Hawaii, Sandwich Islands; in the forests of Mouna Kea,
towards their upper margin.

An herbaceous vine, probably tall-climbing; the strongly angled
stems, like the petioles, &., somewhat pubescent, or glabrate. Leaves
rounded-cordate, 3-5-lobed, either slightly or deeply, in the latter case
with acute sinuses, large, 7 to 10 inches in diameter, membranaceous,

652 PHANEROGAMIA.

glabrous or nearly so above, rather softly puberulent underneath, the
short and fine pubescence at length nearly deciduous, not papillose-
scabrous, the margin coarsely and sharply denticulate with mucroni-
form callous-tipped teeth. Petioles 3 inches long. Tendrils three-
cleft. Male and female inflorescence usually from the same axils.
Male panicles wmbellate (4 or 5 together) at the summit of a deeply
grooved long and slender peduncle (of 5 or 7 inches in length), glan-
dular-puberulent, as well as the exterior of the flower-buds: the fili-
Jorm pedicels often fascicled, 4 or 5 lines long. Perianth with a turbi-
nate tube, five-cleft above the middle, or at length the accrescent and
triangular-ovate lobes of the corolla longer than the tube, greenish-
white, becoming 5 lines in diameter when expanded; the calyx-teeth
short and subulate. Androecium as in the foregoing species, except
that the anthers (either 3 or 5) are themselves more or less syngene-
siously united, often somewhat twisted. Fertile flowers not seen.
Fruit ovate, 5-6b-angled, nearly glabrous, wholly unarmed, in an
immature state conspicuously beaked, in size and character closely
resembling that of the preceding species. The fruits are clustered on
the summit of a peduncle which is 2 inches in length.

Perhaps intermediate states will occur between this and S. pachy-
carpus. Our specimens of the two are not sufficiently complete to
remove all doubt. But this would seem to differ specifically in the
foliage and inflorescence, notwithstanding the similarity of the fruit.

PLATE 81.—Sicyos (SICYOCARYA) MACROPHYLLUS: portion of branches,
of the natural size. Fig. 1. A male flower-bud. 2. A male flower.
3. Andreecium. 4. Transverse section of the same. 5. Young fruit,
of the natural size. 6. Vertical section of the same.—The details
magnified, except Figs. 5 and 6.

5. Sroyos (SicyocARYA) CUCUMERINUS, Sp. Nov. (Tab. 82.)
S. glaber; foliis late cordatis integris denticulatis ; paniculis masculis
breviter pedunculatis sepe trifidis racemosis ; fructu oblongo 5-6-angu-

lato glaberrimo.,

Var. 2. foliis triangulari-cordatis promisse acuminatis.

CUCURBITACES. 653

Var. y. foliis pedatilobis, nempe fere tripartitis, segmentis lateralibus
plus minus bilobis.

Has. Hawaii, Sandwich Islands; in the upper part of the forests
of Mouna Kea.

The three specimens (none of them very perfect), apparently from
different stations, but in the same belt of mountain forest, differ in
their foliage. In that assumed as the typical form, the leaves are
broadly cordate, with a deep and narrow sinus, occasionally almost
reniform, mostly acuminate, wndivided, and showing scarcely any
tendency to become lobed, 32 to 64 inches in diameter, glabrous, as
are the angled stems and even the inflorescence, membranaceous, but
not very thin; the margin sharply denticulate with mucroniform
callous-pointed teeth. The var. 3. presents a triangular-cordate leaf,
5 inches long and 3 wide at the base, where it somewhat inclines to
have lateral lobes, the apex long-acwminate: the petiole somewhat
pubescent. The var. y. exhibits smaller leaves, pedately lobed, being
all three-parted, with narrow sinuses, the lateral divisions more or less
two-lobed. The male inflorescence, in the best developed specimens,
consists of 3 compound and lax racemose panicles, on an axillary
short pedunele (of an inch or two in length): pedicels filiform, 2 or 3
lines long, glabrous. Perianth of the male flowers glabrous, five-cleft
to the middle, with minute calyx-teeth at the sinuses externally,
when expanded nearly rotate and 3 lines in diameter. Anthers 6 or 6,
somewhat twisted, nearly as in S. pachycarpus. Perianth of the
fertile flowers not seen. Female inflorescence at the summit of a
peduncle from the same axil with the male panicle, and sometimes 2
inches long. The fructified ovaries about 3 together, oblong, pointed
with a tapering beak, glabrous, one-celled, with a single suspended
ovule. Fruit oblong, nearly an inch in length, 4 or 5 lines in dia-
meter, apparently somewhat fleshy and drupaceous when fresh, 5—6-
angled, but less sharply so than in the preceding species, smooth and
glabrous, the thick pericarp in the dry state coriaceo-crustaceous, one-
celled, one-seeded, not beaked. Seed with a smooth and rather thin
testa: cotyledons oblong, thickish.

PLATE 82.—Sicyos (SIcyocARYA) CUCUMERINUS: portions of the

three forms. Fig. 1. Flower-bud, from the var. @. 2. Flower, from
164

654 PHANEROGAMIA.

the same. 3. Androecium. 4. Vertical section of a fructified ovary.
—These variously magnified. 5. Transverse, and 6, longitudinal
section of the fruit from var. a., of the natural size. 7. The seed.
8. Embryo.

There are several undeterminable specimens in the collection. One
of them, consisting of foliage only, from the mountains of Tahiti,
probably belongs to one of the four still undescribed species of Cucur-
bita of Solander, enumerated by Forster at the end of his Prodromus.
Two others, from Luzon, are apparently species of Bryonia, but too
imperfect for determination.

Lagenaria vulgaris was found in cultivation at all the larger Pacific
Islands.

Cucurbita Aurantia? the “ Arroro” of the Tahitians, is said to
grow in mountain woods at Tahiti. Probably it is one of the species
enumerated by Solander. There are no specimens in the collection.

Orv BEGONIACE &.

1. BEGONIA, Linn.

x Brasilienses.
1. BEGONIA LUXURIANS, Scheidw.

Begonia luxurians, Scheidw. in Dietr. & Ott. Allgem. Gartenz. 1848, p. 181; Walp.
Ann. Bot. 2, p. 651.

Has. Organ Mountains, Brazil: where it was also collected by
Gardner (No. 607), and by Pohl.

The late Dr. Gardner says of this species: “This is one of the most
handsome of the genus I have yet met with. It grows in rather wet
places in the virgin forests, and reaches sometimes to the height of
fourteen or fifteen feet. April, 1837.”—Gardn. in herb. Hook. Only

BEGONIACE &. 655

the male plant is known. If rightly referred to the B. luxurians,
briefly characterized by Scheidweiler, it is a smoother form of that
species, the leaves, &c., not being hispid. Some of the numerous
leaflets, moreover, are bisected or trisected. The (male) flowers are
small, and crowded in a compound corymbose panicle, which is very
long-peduncled.

2. BEGONIA VITIFOLIA, Schote.

Begonia vitifolia, Schott, in Spreng. Syst. Veg. Cur. Post. p. 407; Walp. Repert.
2, p. 210, excl. syn. Hook.
B. truncata, Velloz. Fl. Flum. 10, p. 47.

Has. Organ Mountains, Brazil; where it was also gathered by

Gardner.

3. BEGONIA RAMENTACEA, Facton.
Begonia ramentacea, Paxton, Mag. Bot. 11, p. 73, cum ic.
Has. Near Rio Janeiro.

A large and coarse species, apparently not uncommon near Rio
Janeiro, where it was gathered by Pohl, and by Gardner (No. 608).
I refer our plant to this species from the appropriateness of the name
to the ramentaceous petioles; the work in which B. ramentacea is
figured not being within my reach.

4. BEGONIA VAGINANS, Velloz.
Begonia vaginans, Velloz. Fl. Flum. 10, t. 37; Walp. Repert. 2, p. 217.

Has. Near Rio Janeiro.

5, BEGONIA CASTANEEFOLIA, Dietr. & Otto.

Begonia castaneerfolia, Dietr. & Otto, Allgem. Gartenz. 4, p. 356; Walp. Repert. 1. c.
B. patens, Griseb. in Herb. Hook.

656 PHANEROGAMIA.

Has. Near Rio Janeiro. (Evidently allied to B. confertiflora, of
Gardner).

6. BecontA HooKERIANa, Gardn.

Begonia Hookeriana, Gardn. in Hook. Lond. Jour. Bot. 4, p. 135.

Has. Organ Mountains, Brazil.

7. BEGONIA RETICULATA, Gardn. l. ec.

Has. Brazil; near Rio Janeiro. (Foliage, without flowers or
fruit.)

8. BEGONIA CoccINEA, Hook.
Begonia coccinea, Hook. Bot. Mag. t. 3990; Walp. Repert. 2, p. 214.

Has. Near Rio Janeiro, or in the Organ Mountains, Brazil.

9. BecontaA Fiscuert, Dietr. & Otto.

Begonia Fischeri, Dietr. & Otto, Allgem. Gartenz. 4, p. 8354; Hook. Bot. Mag. t. 3532 ;
Walp. Repert. 1. c.

Has. Organ Mountains, Brazil. (The same as Gardner’s No. 5590,
and Pohl’s No. 1824.)
10. Breconta HASTATA, Velloz.
Begonia hastata, Velloz. Fl. Flum. 10, t. 54?

Has. Organ Mountains, near Rio Janeiro. (A very imperfect
specimen.)

BEGONIACES. 657

11. BrGonta EREcTA, Velloz.
Begonia erecta, Velloz. Fl. Flum. 10, t. 43?

Has. Organ Mountains, Brazil.

This might be the B. bidentata of Raddi, in Spreng. Syst. Veg.;
but the two wings of the fruit, although broad and equal, are not
bidentate.

12. Beconta vittosa, Lindl.
Begonia villosa, Lindl. Bot. Reg. t. 1252; Walp. Repert. 2, p. 214.
Has. Rio Janeiro.
There are indeterminable specimens of two more species in the

Brazilian collection.

x x Peruviane (Tuberose).
13. BEGONIA OCTOPETALA, L’ Her.
Begonia octopetala, L’ Her. Stirp. Nov. p. 101; Hook. Bot. Mag. t. 3559.

Haz. Amancaes Mountains, near Lima. (Without flowers or fruit.)

14. BEGoNIA GERANIIFOLIA, Hook.
Begonia gerantifolia, Hook. Bot. Mag. t. 8387; Walp. Repert. 2, p. 208.
Has. Amancaes Mountains, near Lima.

The stems, a span or more in height, spring from a round tuber,

and are often terminated with flowers while the leaves are still
165

658 PHANEROGAMIA.

unfolding. The broadly triangular capsules bear three upwardly di-
lated wings, one of which is longer than the others, but narrow, and
ascending.

x x * Philippenses.

15. BEGONIA REPENS, Blume.

Begonia repens, Blume, Enum. Pl. Jav. 1, p. 95; Walp. Repert. 2, p. 208.

Has. Luzon; on mountains in the vicinity of Bajios.

16. Brconra Cuminer, Sp. Nov.

B. erecta, perennis? glabrata; foliis dimidiato-cordatis vel semisagittatis
acuminatis grosse dentutis margine denticulatis ; fructu subcequaliter
trialato, alis completis.

Has. Majaijai Mountains, Luzon.

A very imperfect fruiting specimen, which on comparison proved
to belong to the same species as No. 1897 of Cuming’s Philippine
collection, which so far as I can determine is undescribed. The plant
is probably erect, from a perennial root: it is glabrate; but the young
parts, especially the petioles and the veins of the leaves beneath, are
more or less ferrugineous-hirsute. Leaves long-petioled, elongated,
half-cordate or half-sagittate, with the basal portion rounded, above
tapering into a slender acumination, unequally and coarsely-toothed
around the whole outer margin and above the middle of the other
edge, also denticulate, 5 or 6 inches long, an inch or two in width,
paler beneath, punctate. Capsule elliptical, 7 or 8 lines long, very
smooth, almost equally three-winged ; the wings complete from the base
to the apex, even, 2 or 3 lines broad.

17. Breconta aQuata, Sp. Nov.

B. repens? glaberrima; ramis gracilibus; foliis fere ceequilateris ovato-
oblongis acuminatis parce dentatis basi obtusis trinervatis breviter

CACTACEA. 659

petiolatis tenwibus ; stipulis lanceolatis petiolum adequantibus 3 pedun-
culis brevibus axillaribus uni—paucifloris ; fructu equaliter trialato,
alis completis.

Has. Luzon; on mountains near Bafios.

Stems slender, occasionally throwing out rootlets at the nodes,
therefore probably creeping; the branches and whole plant glabrous.
Leaves nearly equal-sided, small (an inch or more in length), thin,
ovate-oblong, coarsely and sharply toothed above the middle, acuminate,
sparingly feather-veined, pale, rounded or obtuse at the base, where
they are three-nerved or indistinctly five-nerved. Stipules lanceolate,
pointed, scarious, deciduous, 3 lines long, as long as the petiole.
Peduncles axillary, very short, forked, the filiform divisions or pedicels
shorter than the leaves, one—three-flowered. Divisions of the perianth
2, rounded, barely 2 lines long. Capsule orbicular in outline, equally
three-winged ; the wings complete from the base to the apex, even, a
line and a half wide.

Orv. CACTACE &.

1. MAMILLARIA, Haw.

1. MAMILLARIA FLAVESCENS, DC.

“M. caulibus pluribus ovato-cylindraceis e radice crassa ; aculeis setacers
rectis brunneis; floribus verticalibus (ultrapollicaribus) flavis ; sepalis
lineari-oblongis obtusis patulis reflexisve circa 20; petalis fere 20
ovato-oblongis subobtusis ; stigmatibus 9 radiatis.” Engel. Ms.

Mamillaria flavescens, DO. Prodr. 8, p. 459; Pfeiff. Enum. p. 10?
M. prolifera, Haw. Synop. p. 177, ex Salm-Dyck.

Has. Near Lima, Peru.

660 PHANEROGAMIA.

There are no specimens in the collection; but there is a good
drawing, made from the living plant by the late Mr. Agate; from
which Dr. Engelmann has drawn up the characters given above, and
the subjoined description and remarks.— Stems several from the same
very thick root, or proliferous at the base, 23 to 3 inches high, an
inch and a half or less in diameter, ovotd-cylindrical ; the setaceous
straight prickles half an inch in length, brown. Flowers upright from
the summit of the stems, 14 lines long, 9 lines in diameter; the
spreading sepals about 20 in number, linear-oblong, obtuse, yellowish :
petals about the same number, ovate-oblong, obtusish, yellow. Style
half an inch long: stigmas 9, radiate.

“M. flavescens is one of the very few species coming from tropical
South America. The descriptions which I find in different works
agree tolerably well with our plant; though the stems are said to be
proliferous towards the summit, the spines are generally lighter-
coloured, and the yellow flowers appear in a ring around the top.”

2, CEREUS, Aaw.
1. Cereus CHILENSIS, Colla, var. poLyconus, Salm-Dyck.

Cereus Chilensis, Colla, Pl. Hort. Ripul. App. 2, p. 842; Pfeiff. Enum. p. 86; Salm-
Dyck, Cact. Hort. Dyck. p. 44 & 199.

Has. Chili; common from near Valparaiso to the Cordilleras,
according to Dr. Pickering’s memoranda. A drawing only in the
collection.

“This species evidently belongs to the section Velutini of Prince
Salm.: stout and erect Cerei, with broad and interrupted ribs, very
narrow grooves, and covered with a more or less velvety epidermis ;—
all natives of Chili and adjacent countries. This appears to have 13
or 14 ribs [14 or 15, Pickering], more than any species of this sec-
tion, with the exception of the above-named form, which has 15.
The tubercles of the ribs are 9 lines long, 6 lines wide, and bear
towards their upper end a whitish areola, of 3 lines in diameter,

CACTACE SA. 661

armed with 12 to 16 stout, almost conical, straight spines, the 3 to 5
inner ones the stoutest, 4 to 6 or 6 to 8 lines long. The grooves
between the ribs are very narrow, acute and not deep. Of no species
of this section do the flowers appear to be known. The following
description of the flower figured by Mr. Agate, therefore, supplies a
desideratum.—Flowers 5 inches long, sessile above the areole, and
in the transverse incision between the tubercles; the tube curved
upwards, over 3 inches long, gradually enlarging above; the limb 2
inches in diameter; ovary subglobose, horizontal, 9 or 10 lines long,
densely imbricated with nearly a hundred sepaloid scales; sepals
about 100, imbricated on the tube, the lower triangular, the upper
successively becoming larger and lanceolate, greenish, with the apex
somewhat reddish; about 15 of the uppermost ovate-lanceolate, acute,
tipped with light reddish-brown; petals 15 to 18, ovate, mucronate,
4 or 5 lines wide, white.” Engelmann, Ms.

There are no specimens of Cactacee in the collection, except an
imperfect sterile branchlet of some Opuntia from Rio Negro, North
Patagonia, which is undeterminable. In his manuscript list, Dr.
Pickering enumerates twelve species as having been noticed at Rio
Janeiro; viz. one Melocactus, five Cerei, one Epiphyllum, one Opuntia,
a Pereskia, and three species of Rhipsalis. At Rio Negro, three
Opuntice, a Cereus, and a doubtful Mamillaria. In Chili, besides the
Cereus Chilensis, two Hchinocacti and an Opuntia are mentioned. In
the tropical region of Peru, from Lima to the Cordilleras, a Melocactus,
four or five Cerei, and as many Opuntia are noted; and three or four
others are mentioned as growing in the western Cordillera region,
from Obrajillo to Culnai. Among these, one of doubtful genus :
having a slender and upright trunk, 8 or 10 feet high, with spreading
branches at the summit, distant clusters of long spines, and large,
dry, and stony seeds. A second species resembling the last, but
lower (8 inches high) and less woolly, with smaller fruit, is mentioned
as occurring in the pampas or pasture region of the Cordilleras, two
and a half learues above Culnai.

166

Orv GROSSULACE &.

1. RIBES, Linn.

1. Riggs MAacEeLuanicum, Poir.
Ribes Magellanicum, Poir. Suppl. 2, p. 856; Hook. f. Fl. Antarc. p. 269.
Has. Orange Harbour, Fuegia; in woods near the coast.
“A tall shrub, with black fruit,” which is said by Dr. Hooker to

have a very agreeable flavour. A species of the same group with R.
mgrum and R. Hudsonianum of the northern regions.

2. RIBES GLANDULOSUM, Ruiz & Pav.

Ribes glandulosum, Ruiz & Pay. Fl. Per. & Chil. t. 233, f. 6; DC. Prodr. 3, p. 481.

Has. Chili; south of Valparaiso: also on the lower Cordilleras.

3. Rises punctatum, Ruiz & Pav. l. c.

Has. Chili; near Valparaiso. (Foliage only.)

4, RIBES ALBIFLORUM, Ruiz & Pav. l.e.
Has. Andes of Peru, above Baiios. (In fruit.)

Dr. Pickering mentions two other Peruvian species, one found neat
Obrajillo, the other at Culnai: but there are no specimens of them.

Orv SAXIFRAGACE A

1 ESCALLONIA, Mutis.
1. ESCALLONIA SERRATA, Smith.

Escallonia serrata, Smith, Ic. Pl. Ined. t. 31; DC. Prodr. 4, p. 3; Hook. Ic. Pl. t.
540; Hook. f. Fl. Antare. p. 279.

Has. Orange Harbour, Fuegia; where it is a very abundant shrub.

* Mr. Brown, in indicating the Cunoniacece and Escalloniew as separate orders (in
Flinders, Voy., and in Frankl. Narr.), remarked, that they were to be distinguished
from Saxifragec by their habit, rather than by any important characters of fructifica-
tion: and the remark still holds true. M. Alphonse DeCandolle (in Monogr. Cam-
panul. p. 91, 1830), after seeking in vain for any floral characters whatever, proposed to
consider them as tribes of Saxifragacec ;—a view which was adopted in the Prodromus,
by his father, and also by Endlicher, the latter merely raising four of the tribes of De-
Candolle to the rank of orders, and appending Bauera to the Cunoniee, as Brown had
suggested. In the Flora of North America (1840), another suborder was added for
Philadelphus, a genus which, although usually associated with two real Hydrangea-
ceous genera (Decwmaria and Deutzia), was generally considered to be the type of a
distinct order, and placed in the vicinity of Myrtacew. I am not aware that this view
has been followed by any botanist, except Dr. Hooker, who in his Flora Antarctica
(1847), under Cornidia, distinctly refers Philadelphus to the “class Saxifragee.”’
There is nothing to distinguish Philadelphus from the Hydrangiee except the convolute
gstivation of the petals, which, although peculiar, is surely not of subordinal conse-
quence. Whatever value might have been assigned to it, as separating Philadelphus
from the Saxifragacee, is now reduced to insignificance by the convolute-imbricated
gstivation of the petals of Jamesia, Torr. & Gray (vide Pl. Fendl. p. 55), their imbri-
cated estivation in Fendlera, Engelm. & Gray (Pl. Wright. 1, p. 77, & 2, p. 64),
and the almost free ovary in Carpenteria, Torr. Pl. Fremont. p. 12, t. 7, which in
other respects scarcely differs from Philadelphus itself. It would seem, therefore, that
the order Saxifragacee should embrace five tribes, nearly as adopted by DeCandolle,
‘characterized solely by their organs of vegetation, namely :—the Escalloniee ; woody
plants, with alternate, simple leaves, destitute of stipules ; the Cunoniew, with opposite

664 PHANEROGAMIA.

2. HSCALLONIA RUBRA, Pers.

Escallonia rubra, Pers. Ench. 1, p. 234; Hook. Bot. Mag. t. 2890; DO. Ll. c.
Stereoxylon rubrum, Ruiz & Pay. Fl. Per. & Chil. 3, t. 236.

Has. Chili; in ravines above Valparaiso.

3. ESCALLONIA ILLINITA, Presi.

Eiscallonia illinita, Presl, Rel. Heenk. 2, p. 49, t. 59; Hook. & Arn. Bot. Misc. 35
p. 343; Lindl. Bot. Beg. t. 1900; Remy, in Gay, Fl. Chil. 3, p. 60.
Li. Mertensiana, Ruprecht, in sched. ex Herb. Acad. Petrop.

Has. Chili; near Valparaiso, and on the first Cordilleras above
Santiago. (Both a. and @. Hook. & Arn. 1. c.)

4, ESCALLONIA REVOLUTA, Pers.

Escallonia revoluta, Pers. 1.c.; DO. l.c.; Hook. & Arn. Bot. Mise. 3, p. 341;
Remy, in Gay, Fl. Chil. 8, p. 55.

L. affinis, Ruprecht, 1. c. (var. parvifolia minus hirsuta).

Stereoxylon revolutum, Ruiz & Pav. 1. ¢.

Has. Chili; near Valparaiso: with some sterile branches of the

(simple or compound) leaves and interpetiolar stipules; the Baueree, with opposite
and compound, sessile, exstipulate leaves; the Hydrangiee, with opposite and simple,
exstipulate leaves; and the Samxifragee; herbs, with alternate or opposite, simple or
compound leaves, rarely with manifest stipules. More or less united and distinct
styles, and a partly or wholly free or adnate calyx occur in each of these tribes.
Moreover Cardiandra, Sieb. & Zuce., a strictly Hydrangeaceous genus, has alternate
leaves. Dr. Lindley indeed takes a different view, which, however, will probably be
reconsidered. He not only gives to all these groups (except Bauerew, to which he inad-
vertently attributes whorled leaves and porous anthers) the rank of distinct orders, but
even excludes from his Saxifragal alliance (singularly characterized as having little or no
albumen in the seeds!) the Escalloniee and Philadelphece, retaining moreover in the latter
Deutzia, although it exactly accords with his diagnosis of Hydrangeacee and Decumaria,
which is more closely related to Schizophragma,—a genus retained by him in Hydran-
geacee, along with Broussaisia, notwithstanding their consolidated styles, as also is
Curdiandra, in spite of its alternate leaves,

SAXIFRAGACEA, 665

small-leaved and less pubescent form, which is the EF affinis of
Ruprecht.

5. EScALLONIA ARGUTA, Presi.

Escallonia arguta, Presl. Rel. Henk. 2, p. 48, t. 58; Hook. & Arn. Bot. Mise. 1. c.
E. leucantha, Remy, in Gay, Fl. Chil. 3, p. 53?

Has. Chili; between Valparaiso and Santiago, and on the middle
Cordilleras.

6. ESCALLONIA PULVERULENTA, Pers.

Escallonia pulverulenta, Pers. 1. c.; DC. Prodr. 4, p. 5; Hook. & Arn. 1. ¢.
Stereoxylum pulverulentum, Ruiz & Pav. Fl. Per. & Chil. 3, p. 15, t. 237.

Has. Chili; near Valparaiso, and Cordilleras, near Casa Blanca.

Some specimens approach the £. Berteriana, DC.; which (with £.
resiniflua, Walp.) probably is only a glabrate and often resiniferous

form of this species, with the leaves mostly acute at the base. It
has the same very short styles.

2 CARPODETUS, Pst.

1. CARPODETUS SERRATUS, Forst.

Carpodetus serratus, Forst. Char. Gen. t. 17, & Prodr. p. 48; A. Rich. Fl. N. Zel.
p- 8366; A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 8, p. 247; Hook. Ic. PI. t.
564; Fenzl, in Regensbh. Denkschr. 3, p. 155, t. 1; Hook. f. Fl. N. Zeal. p. 78.

Has. Bay of Islands, New Zealand; on the banks of streams.
(In fruit.)

The details of this plant are well illustrated by Fenzl. Carpodetus,
long viewed as an anomalous genus, has of late been referred to the

Escalloniee by Lindley, Endlicher, and Dr. Hooker, apparently with
good reason.

167

666 PHANEROGAMIA.

38 QUINTINIA, Alps. DC.
1. QUINTINIA SERRATA, A. Cunn.

Quintinia serrata, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 2, p. 256; Hook. Ie.
Pl. t. 558; Hook. f. Fl. N. Zeal. p. 78.

Has. New Zealand: in dry woods at the Bay of Islands. (In fruit.)

4. CALLICOMA, Andr.
1. CALLICOMA SERRATIFOLIA, Andr.
Callicoma serratifolia, Andr. Bot. Rep. t. 566 ; Sims, Bot. Mag. t. 1811; DC. Prodr.

Has. New South Wales; in the neighbourhood of Sydney.

9 CERATOPETALUM, Smith.
1. CERATOPETALUM GUMMIFERUM, Smith.
Ceratopetalum gummiferum, Smith, N. Holl. 1, p. 9, t. 3; DC. Prodr. 4, p. 18.

Haz. New South Wales, near Sydney.

6. SPIRHANTHEMUM, Nov. Gen.

Flores polygamo-dioici vel hermaphroditi. Calyx quadri-quinquefidus,
estivatione valvatus, persistens. Corolla nulla. Stamina & vel 10,
aimee basi calycis inserta, fere hypogyna: filamenta filiform ia, fl. mase.
exserta, hermaphrodito-fert. calyce haud longiora: anthere didyme,

SAXIFRAGACES. 667

biloculares, longitudinaliter dehiscentes. Squamule disci hypogynce
fi. masc. 4 vel 5, subcoalite, apice dentate; fl. fert. 8 vel 10, stami-
nibus alternee, sepe emarginate, Ovaria maris nulla 3; fl. foem.
4 vel 5, discreta, libera, calycis lobis alterna, ovoideo-fusiformia, in
stylos attenuata: stigmata terminalia subcapitata. Ovula gemina
collateralia, vel solitaria, pendula, subanatropa. Folliculi 4-5, com-
presst, cartilaginet vel coriacet, intus longitudinaliter dehiscentes,
mono-dispermt. Semina oblonga, compressa; testa membranacea aut
superne aut utringue alato-producta. Embryo subcylindricus, albu-
mine carnoso parum brevior; cotyledonibus oblongis planiusculis
radicula cylindrica supera subdimidio brevioribus.— Frutices vel
arbuscule Polynesice; foliis oppositis sew verticillatis simplicibus ;
stipulis interpetiolaribus deciduis; floribus parvis paniculatis; pedi-
cellis articulatis.

This curious genus, founded on two species, one from the Samoan,
the other from the Feejee Islands, would appear to have more resem-
blance to Aphanopetalum, Endl., than to any other. It is remarkably
distinguished from all known Cunoniaceous genera by its wholly
separate pistils, of the same number as the sepals. Although very
different in habit and in foliage, it accordingly exhibits the floral
characters of Crassulaceew,—an order which was already difficult
enough to be distinguished from Saaifragacee, on account of the
syncarpous ovaries of Diamorpha and Penthorum.* In this respect it
also imitates Rosaceew (from which the opposite leaves and albumi-
nous seeds exclude it); the pistils simulating those of a Spirca, as
closely as the foliage and inflorescence of the species of the Saxifra-
geous genus Aséi/be simulate the Spirwa Aruncus. The generic name
alludes to this resemblance,

1. SprR#ANTHEMUM SAMOENSE, Sp. Nov. (Tab. 83.)

S. ramis pubescentibus; foliis oppositis ovalibus subacuminatis basi
rotundatis serrulatis insigniter penninerviis; paniculis folia exceden-
tibus; folliculis dispermis; seminibus utrinque appendiculatis.

* Tt has even similar hypogynous scales, the want of which in Sax//ragacee is sug-

gested by Lindley (Vegetable Kingdom, p. 344) apparently as a reason for not admit-
ing the Crassulacee into the Saxifragal alliance.

668 PHANEROGAMIA.

Has. Samoan or Navigators’ Islands (Tutuila?): on mountains at
the elevation of 2,500 feet.

A shrub or tree; with nodose, hirsutely pubescent branches. Leaves
opposite, oval, or elliptical, simple, membranaceous or chartaceous in
texture, 24 to 5 inches long, from one to 3 inches wide, on hairy
petioles from half an inch to an inch and a half in length, rownded at
the base, somewhat pointed or acute at the apex, sevrulate with sharp
teeth, prominently pinnately veined with 9 to 11 pairs of veins, gla-
brate, except the midrib and veins underneath. Stipules inter-
petiolar, oblong, obtuse, entire, membranaceous, sparsely hairy outside,
half an inch long, caducous. Flowers polygamo-dicecious, very small,
greenish, in ample, compound and loose, pubescent panicles usually
exceeding the leaves. These arise from the upper axils, or are some-
times terminal. Peduncles longer than the petioles. Principal
bracts lanceolate or linear, opposite, or rarely quaternate, deciduous.
Pedicels crowded, sometimes fascicled, a line or two in length, articu-
lated in the middle. Flower-buds a line long. Calyx more or less
pubescent outside, either four-cleft or five-cleft, or rarely even six-
cleft; the segments ovate-triangular, valvate in estivation. Corolla
none. Stamens twice as many as the segments of the calyx, and
inserted upon its very base, almost or entirely hypogynous, distinct :
filaments filiform, glabrous, in the sterile flowers twice the length of
the calyx, in the fertile flowers rather shorter than the calyx, persis-
tent: anthers didymous, two-celled; the cells opening lengthwise.
Glands of the disk hypozynous and scale-like; in the sterile flowers
as many as the lobes of the calyx, oblong, toothed at the apex, more
or less coalescent in the centre of the flower; in the fertile flowers as
many as the stamens and alternate with them, nearly half the length
of the ovaries, cuneate-oblong, flat, truncate at the apex and usually
emarginate. Gynecium abortive or entirely wanting in the sterile
flowers; in the fertile consisting of as many perfectly distinct and
free pistils as there are lobes to the calyx. Ovaries ovoid-fusiform,
pubescent, alternate with the calyx-lobes, tapering into a short persis-
tent style, which is tipped with an obtuse somewhat capitate stigma.
Ovules 2, collateral, pendulous from near the middle of the ventral
suture, almost anatropous: the chalazal extremity considerably ex-
tended or appendaged. Follicles two-seeded, not seen mature, when
unripe twice the length of the persistent calyx, silky-pubescent, some-

SAXIFRAGACESR. 669

what compressed. Seeds pendulous, compressed, appendaged at both
ends; the thin testa of the micropylar extremity being extended into
a thin and obtuse wing; the lower or chalazal end produced into a
subulate appendage of equal or greater length. Kmbryo not seen.

PLATE 83, A.—SPIRHANTHEMUM SAMOENSE: a branch, with fertile
flowers, of the natural size. Fig. 1. Portion of male inflorescence, of
the natural size. 2. Male flower. 3. Vertical section of the same.
4. One of the glands of the disk, separated. 5. Fertile flower.
6. The same, with a part of the calyx cut away. 7. Vertical section
of a fertile flower, showing the ovules, &. 8. An ovule detached.
9. A fertilized ovule. 10. An unripe seed.—The details magnified.

2. SPIRAANTHEMUM VitrensE, Sp. Nov. (Tab. 83.)

S. glabrum; foliis oppositis et verticillatis obovato-ellipticis oblongisve
obtusis basi attenuatis integerrimis puucivenosis paniculas eaceden-
tibus ; folliculis monospermis ; semine superne alato.

Has. Feejee Islands; at Sandalwood Bay, Vanua-levu; and Mu-
thuata, at an elevation of 1,500 feet.

Shrub or small tree (the size not recorded), glabrous, or the slender
branchlets and intlorescence obscurely puberulent when young. Sti-
pules not seen, probably small and very caducous. Leaves opposite
and often verticillate in threes or fours, or even in fives, about 2
inches in length, obovate-elliptical, oblong, or oval, thin, but of a coria-
ceous texture, dull, obtuse, entire, contracted at the base into a petiole of
3 to 6 lines in length, sparingly pinnately veined ; the veins only 4 or
0 pairs, connected by minute reticulated veinlets. Flowers minute,
humerous, in rather small and cymose, axillary panicles, which are
shorter than the leaves. Peduncles longer than the petioles. Bracts
small, deciduous. Pedicels crowded, often fascicled, less than a line
long, articulated at the base. Only one kind of flowers seen, which
appear to be hermaphrodite. Flower-buds barely half a line in dia-
meter, globose. Calyx glabrous, deeply cleft or parted into 4, 5, or
rarely 6 triangular-ovate segments, valvate in estivation. Corolla

none. Stamens twice as many as the segments of the calyx, inserted
168

670 PHANEROGAMIA.

into its very base, nearly hypogynous, distinct: filaments filiform,
about the length of the calyx: anthers didymous, incumbent, two-
celled, destitute of a connective; the cells opening lengthwise. Pol-
len-grains globose, simple, not very abundant. Glands of the disk
hypogynous, as many as the stamens and alternate with them, scale-
like, cuneate or oblong, fleshy, truncate or emarginate, sometimes
slightly united? Pistils as many as the segments of the calyx and
alternate with them, sparingly hairy, ovoid, entirely free and sepa-
rate, tapering into a persistent style, which is tipped by a capitate-
truncate stigma. Ovule solitary, affixed to the ventral suture near
the middle, pendulous, between amphitropous and anatropous. Folli-
cles one-seeded, almost glabrous, thrice the length of the persistent
calyx, compressed, semiovate, sometimes slightly lunate, pointed by
the style, coriaceo-cartilaginous in texture, a line and a half long, de-
hiscent down the ventral suture, from which a filiform edge often
separates. Seed winged at the apex (the micropylar end), filling the
cell, oblong, compressed, the thin and membranaceous testa not reticu-
lated, below conformed to the oval nucleus, not at all produced at the
base or chalaza, above extended into the thin and obscurely reticu-
lated wing. Embryo straight, in the axis of rather copious fleshy
albumen, occupying the greater part of its length, somewhat eylin-
drical, but the oblong cotyledons a little broader and flattened, and
about half the length of the superior radicle.

PLATE 83, B.—Sprramanturmum VitrENsE: a branch, of the natural
size. Fig. 1. Diagram of the flower. 2. Flower. 3. Vertical section
of the same. 4. Portion of the stamens and interposed glands of the
disk, 5. Fruit, with the persistent calyx. 6. A follicle detached.
7. Vertical section of a follicle. 8. A dehiscent follicle. 9. A seed.
10. Vertical section of a seed. 11. Embryo detached.—The details
magnified.

7. WEINMANNIA, Linn.

WEINMANNIA, Linn.; Juss.; H. B. K.; Hook. f. Fl. N. Zeal. p. 79.
LerosPermuM, Don, in Edinb. New Phil. Jour. 9, p. 85 & 91; Endl. Gen. p- 818.
AcKAMA, A. Cunn. in Ann. Nat. Hist, 2, p. 358; Endl. 1. .

ARNOLDIA, Blume, Bijdr. p. 868; DC.; Endl. 1. ¢.; Hassk. Pl. Jay. Rar. p. 825.

SAXIFRAGACEZ. 671

* Novo-Zelandice.
1. WEINMANNIA SyLvicoLa, Banks & Soland.

Weinmannia sylvicola, Banks & Soland. ined.; Hook. f. Fl. N. Zeal. p. 79.

W. sylvicola, fuschioides, & hetulina, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 2,
p. 822.

Has. Bay of Islands, New Zealand.

The three forms, considered as species by Cunningham, occur in
the collection. There are no specimens absolutely referable to W.
racemosa, Linn. f.: but several approach it so closely as to render it
probable, as Dr. Hooker suggests, that all belong to one species;
which in that case should bear the last-mentioned name.—Dr. Hooker
has shown that the character of Don’s genus Leiospermum rests on an
erroneous observation, so far as respects the New Zealand species. I
may extend the remark to the species from the Society Islands, and
also to the Ackama. roscfolia.

2. WEINMANNIA (ACKAMA) ROS&FOLIA. (Tab. 84.)

W. arborea; ramulis petiolisque velutino-pubescentibus ; foliolis 3-8-jugis
cum impart ellipticis argute serratis; paniculis decompositis paten-
tibus; floribus subsessilibus minimis pentameris; petalis lineari-spa-
thulatis calycem vix superantibus ; capsula ovata turgida hirta ; semi-
nibus ovoideis undique crinitis.

Ackama rosefolia, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist. 2, p. 8358; Hook f.
Fl. N. Zeal. p. 79,

Has. Bay of Islands, New Zealand.

Tree 30 or 40 feet high; the young branchlets, petioles, &c., cine-
reous or fulvous with a soft velvety pubescence. Stipules interpetiolar,
ovate, caducous. Petioles with the elongated rhachis from 2 to 10
inches long, terete. Leaflets 3 to 8 pairs and an odd one, elliptical or
oblong, from one to 2% inches in length, or the lower smaller and

672 PHANEROGAMIA.

roundish, sharply serrate, subsessile, nearly membranaceous in texture,
pinnately veined, pubescent when young, at length glabrate, except
the midrib and veins underneath. Flowers minute (two-thirds of a
line across), sessile on the slender ramifications of the ample, com-
pound or decompound spreading panicles, which are axillary, or else
terminate short axillary branchlets. Calyx glabrous, five-parted; the
lobes triangular-ovate (valvate in estivation, according to Dr. Hooker),
persistent. Petals 5, linear-spatulate, scarcely exceeding the calyz,
deciduous. Stamens 10, inserted into the base of the calyx outside
of the deeply ten-lobed disk, alternate with its lobes: filaments fili-
form, exserted; those opposite the petals longer than the others:
anthers didymous, two-celled, opening longitudinally. Ovary free,
densely hirsute, two-celled, many-ovuled: styles 2, distinct, filiform :
stigmas minute, terminal. Ovules in several series, amphitropous.
Capsule ovate, turgid, free, subtended at the base by the small and
persistent calyx (which is now raised on an extremely short pedicel),
hairy, 2 or 2% lines long, two-celled, septicidal, the valves pointed by
the persistent styles, at length two-cleft above. Seeds numerous or
several in each cell, nearly half a line long, turgid-ovoid, slightly
apiculate at each end, hemitropous, the short rhaphe slender: testa
membranaceous, not reticulated, conformed to the nucleus, rather
sparsely beset all over with very long and lax hairs. Embryo nearly
the length of the sparing albumen: radicle cylindraceous: cotyledons
broadly oval, thickish, about the length of the radicle.

Our specimens bear ripe fruit only, with an abundance of well-
formed seeds, apparently in better condition than any previously
examined. They are not “smooth and rostrate,”’ as stated by End-
licher, although there is a minute apiculation at each end, nor are the
perfect ones “extremely minute, and with a lax reticulated testa,” as
described by Dr. Hooker. Abortive seeds often occur to which this
description will nearly apply, but the matured ones are not very
small, the testa is conformed to the nucleus, not at all reticulated,
and beset with very long hairs, like those of Weinmannia generally.
Dr. Hooker remarks that Ackama is hardly distinct from Weinmunnia,
except in the form of the petals. He assigns, however, another cha-
racter, namely a valvate calyx, and perhaps correctly: the few flowers
that I possess are too far advanced to verify the point. But even if
it be so, this character, along with the narrow petals and the panicu-

SAXIFRAGACEA. 673

late inflorescence, will hardly serve to distinguish it as more than a
subgenus of Weinmannia.

Prats 84.—WEINMANNIA (ACKAMA) ROSAFOLIA: a branch, in fruit,
of the natural size. Fig. 1. Flowers (a portion of the inflorescence),
from a specimen gathered by Cunningham, of twice the natural size.
2. Two flowers, from the same, magnified. 3. Vertical section of a
flower. 4. A petal. 5. Stamens. 6. Transverse section of the
ovary. 7. Ovules. 8. Capsule and persistent calyx. 9. Transverse
section of a capsule. 10. Capsule, dehiscent. 11. A seed. 12. Ver-
tical section of the same. 13. Transverse section of a seed through
the cotyledons. 14. Embryo.—The details variously magnified.

* * Polynesice.
3. WEINMANNIA PARVIFLORA, Forst. (Tab. 85.)

W. arborescens ; ramulis pedunculisque pubescentibus ; foliis simplicibus
ovatis oblongisve glabris calloso-serratis; racemis ternis vel subpluribus
paniculatis; floribus parvis tetrameris; glandulis disci 8 lineari-
oblongis calyce vix brevioribus ; stylis brevissimis ; capsula fere glabra ;
seminibus oblongis utringue comosis.—Variat; foliis crasso-coriaceis
ovatis seu ellipticis breviter vel brevissime petiolatis; et foliis tenuio-
ribus oblongis vel elongato-oblongis grosse sinuato-dentatis utrinque
acutis vel acuminatis modice petiolatis.

Weinmannia parviflora, Forst. Prodr. Fl. Ins. Austr. p. 29; Willd. Spec. Pl. 2,
p. 438.
Leiospermum parvifloruwm, Don, in Edinb. Phil. Jour. 1. ¢.; Guill. Zeph. Tait. p. 55.

Has. Tahiti: on mountain ridges. (Elizabeth Island, Cuming.)

A shrub or small tree; the young branches and inflorescence pubes-
cent or cinereous-hirsute. Leaves all simple, on petioles of from 14 to
3 lines long, ovate or oblong, from an inch to 2 or 3 inches long,
serrate with callous and incurved teeth, which are sometimes acute,
sometimes very obtuse and separated by large sinuate sinuses, gla-
brous, or the midrib at first pubescent underneath, thick and coria-

ceous, or sometimes rather thin, in the more elongated forms mostly
169

674 PHANEROGAMIA.

acute or pointed at both ends. Stipules rather persistent, oval, as
long as the petioles. Rucemes terminal in threes, or several and pant-
cled, erect, from one to 3 inches long. Pedicels crowded or fascicled,
a line and a half long; the minute bracts deciduous. J lowers small,
a line and a half long, including the exserted stamens, tetramerous.
Lobes of the calyx ovate, half the length of the elliptical petals, in
fruit deciduous. Glands of the disk 8, alternating with the stamens,
erect, linear-oblong, fleshy, truncate, little shorter than the calyz.
Ovary ovoid, obtuse, strigose-puberulent, abruptly tipped with the
very short styles. Capsule nearly glabrous, ovoid, obtuse, not 2 lines
long, septicidal, few-seeded. Seeds oblong, anatropous; the thin testa
not reticulated, conformed to the nucleus in the manner of the genus,
appendaged with a loose tuft of hairs at each end.

The characters given by Forster and by Willdenow differ conside-
rably, but each accords with different specimens in our collection; the
species being almost as polymorphous in foliage as is the allied W. syl-
vicola of New Zealand. Don assigned the habitat of New Zealand to
this species; and his description is not a good one: but the original
specimen before him was from Tahiti, and is only in flower. If he
had ripe seeds to examine, the tufts of hairs could hardly escape
notice.

PLATE 85, A.—WEINMANNIA PARVIFLORA. Fig. 1. A flower. 2.
Vertical section of the same. 3. A capsule with the persistent styles.
4, Aseed. 5. Vertical section of the same.—Magnified.

4. WEINMANNIA AFFINIS, Sp. Nov.

W. glabra; foliis simplicibus oblongis seu ovalibus calloso-serratis basi in
petiolum brevem contractis rarissime bifoliolatis, foliolis in petiolo
superne marginato sessilibus; racemis geminis ternisve corymboso-
paniculatis ; “floribus roseis ;” calyce deciduo; stylis capsula ovoidea
glabella dimidio brevioribus ; seminibus ovalibus utrinque comosis.

Var. 2. foliis trifoliolatis ; foliolis basi attenuatis plerumque petiolulatis.

Has. Feejee Islands: on the mountains of Ovolau, at the altitude
of 1,200 feet. (Samoan Islands ?)

SAXIFRAGACESR. 675

A shrub or small tree, glabrous, except a fine microscopic pubes-
cence on the rhachis of the inflorescence and the pedicels. Leaves
mostly simple, oblong, from 12% to 4 inches long, an inch or more in
width, sometimes elliptical or oval, rather coriaceous, usually punctate
with brown dots underneath, perfectly glabrous, obtuse or obtusely
acuminate, obtusely callose-serrate, abruptly narrowed into a short
margined petiole (of 3 to 5 lines in length), apparently not lucid. In
one instance the petiole, becoming two-thirds of an inch in length
and broadly margined above, bears a pair of sessile leaflets, otherwise
resembling the leaves, articulated with its summit. In var. @., of
which there is only an imperfect specimen, all the leaflets are in threes
and more or less petiolulate. Stipules deciduous, only seen on a shoot
of var. @., where they are oblong-ovate. “Flowers rose-colour,”
according to Dr. Pickering’s memoranda, if rightly identified, but all
the specimens are in fruit. Macemes terminal and from the upper
axils, usually an pairs or threes on a flattened common peduncle, and
accumulated so as to form a kind of corymb, 2 or 3 inches long in
fruit, very dense; the pedicels crowded or fascicled, only a line long.
Calyx deciduous from the fruit, four-parted. Capsule ovoid, obtuse,
12 to 2 lines long, minutely puberulent under a lens, but glabrous to
the naked eye, two-valved, few-seeded, twice the length of the short
styles. Seeds oval, glabrous, except the conspicuous tu/t of long hairs
at each end.

This species (of which the flowers are still a desideratum) is allied
on the one hand to W. parviflora, which has still shorter styles and
more pubescence; on the other to the New Zealand species, which
have elongated styles and (at least W. sylvicola) the perfect seeds
hairy all over, though most so at the extremities—There is a mise-
rable fruiting specimen, ticketed as from the Samoan Islands (perhaps
by some transposition), which I am uncertain whether to refer to the
present species, or to W. parviflora. Its styles, however, are rather
longer than in the plant of Tahiti.

5. Weinmannia Rica, Sp. Nov. (Tab. 85.)

W. arborescens; ramulis junioribus petiolis costisque subtus pube brevi
velutinis cinereisve; foliolis uni-quadrijugis cum impari oblongis

676 PHANEROGAMIA.

acuminatis subserratis glabris; stipulis orbiculatis integerrimis ; ra-
cemis geminis vel ternis confertis; floribus tetrameris parvis ; petalis
obovatis calyce persistente duplo longioribus; stylis gracilibus cap-
sula di-tetrasperma paullo brevioribus; seminibus oblongis utrinque
comosis.

Has. Feejee Islands: on mountains at Sandalwood Bay, Vanua-
levu, at the altitude of 2,000 feet.

An arborescent species; the younger branches, petioles, and midrib of
the leaflets underneath clothed with a soft and jine, cinereous-velvety
pubescence, which is to a good degree persistent. Leaves all pinnate.
Leaflets from 3 to 9 on the same branch oblong or elliptical, on sterile
shoots verging to oblong-lanceolate, serrate with small teeth, acuminate,
glabrous (except the midrib), inconspicuously veiny, lucid above, not
punctate, acute or pointed at the base, sessile or slightly petiolulate,
14 to 2% inches long. Petiole terete, not margined. Stipules orbi-
cular, entire, half an inch in diameter, tardily deciduous. Inflores-
cence minutely pubescent. The specimens all] in fruit, but with some
vestiges of the floral organs remaining. LRacemes in pairs or threes
from the uppermost axils, and terminal, numerous and crowded so as
to appear paniculate-corymbose, very densely-flowered, 2 or 3 inches
long in fruit; the pedicels barely a line long. Calyx glabrous, four-
parted, persistent in fruit. Petals obovate, twice the length of the calyx,
about a quarter of a line long, white. Glands of the disk 8, oblong-
linear, alternate with the stamens. Ovary minutely pubescent, two-
celled: ovules 2 to 6 (usually 4) in each cell, anatropous. Capsule
ovoid-fusiform, minutely pubescent under a lens, barely a line and a
half in length; the slender styles fully a line long. Seeds solitary or
often in pairs in each cell, oblong, anatropous, marked with a slender
complete, rhaphe, comose at each end: the testa otherwise glabrous or
nearly so at maturity, not reticulated. Embryo clavate-cylindrical,

nearly as long as the albumen. Cotyledons thick, oblong, shorter
than the radicle.

This species is closely related to Weinmannia Arnoldia, viz., the
Arnoldia pinnata of Blume (of which I possess a flowering specimen),
and perhaps scarcely less so to W. Blumet (Arnoldia heterophylla,
Blume). Hasskarl, in his Pl. Jav. Rariores, p. 324, has justly re-

SAXIFRAGACESA. 677

marked that the ovules in Arnoldia are not solitary, but several in
each cell. The seeds are not described, but they will doubtless accord,
as A. pinnata does in all other respects, with Weinmannia. Perhaps
that species and the present, in which the seeds are abortive in many
specimens, are polygamo-dicecious, in the manner of Astilbe, &c.

Piate 85, B—Weinmannia Ricu: branchlets, of the natural size.
Fig. 1. A flower; the ovary already fructified. 2. A petal. 3.
Vertical section of fig. 1. 4. Calyx, with the glands of the disk. 5.
Dehiscent capsule. 6. Transverse section of the capsule, before
dehiscence. 7. A half-grown seed. 8. A mature seed. 9. Ver-
tical section of the same. 10. Embryo.—The details variously mag-
nified.

6. WEINMANNIA SPIRAOIDES, Sp. Noy.

W. arborea, hirto-pubescens ; foliolis bijugis cum impari oblongo-lanceo-
latis seu elliptico-oblongis grosse serratis utrinque acutis; stipulis orbi-
culatis serratis.

Has. Feejee Islands; on Ovolau, at the altitude of 500 feet.

“A small tree;” of which only a sterile branch was collected.
This, as well as the petioles and costa beneath, is hirsute-pubescent.
Leaves all pinnately quinquefoliolate, somewhat resembling those of
Spirea sorbifolia, membranaceous. Leaflets oblong-lanceolate, acute at
both ends, subsessile, an inch or an inch and a half in length, or
the terminal larger, oblong-elliptical and petiolulate, all sharply and
coarsely serrate, pubescent when young, the upper surface glabrate.
Stipules orbicular, serrate, 4 or 5 lines broad, rather persistent.

7. WEINMANNIA SAMOENSIS, Sp. Nov.

W. fruticosa; ramulis petiolisque hirto-pubescentibus ; foliolis uni—tri-
jugis cum impart lanceolatis acuminatis obsolete serratis glabris ;
racemis scepissime ternis; calyce quadripartito deciduo; capsula
glabra stylis brevissimis apiculata ; seminibus utringue comosis.

170

678 PHANEROGAMIA.

Has. Samoan or Navigators’ Islands; on the mountains of Tutuila,
at the altitude of 2,500 feet.

The specimens only bear mature and dehiscent fruit. They are
said to belong to a shrub, of 6 or 10 feet in height. The slender
branchlets and petioles are hirsute-pubescent, as are the main peduncles,
and there are traces of a slight hairiness on the midrib of the leaflets
underneath. The stipules have fallen. Leaves all pinnately com-
pound, and 5—7-foliolate or sometimes trifoliolate; the rhachis terete
below, but somewhat margined between the upper pairs of leaflets.
Leaflets lanceolate, 2 or 3 inches long, 5 to 9 lines wide, more or less
acuminate, obsoletely serrate, sessile, or the terminal one somewhat
petiolulate, glabrous, dotted underneath, chartaceous in texture, rather
dull. Peduncles terminal, short, usually bearing the virgate racemes
in threes, sometimes in fives. These in fruit are 2 or 3 inches long,
slender but dense, and spreading, puberulent. Pedicels crowded and
fascicled, a line long. The vestiges of the flower show a four-parted
calyx, deciduous from the fruit, and 8 slender glands of the disk.
Petals and stamens not seen. Capsule glabrous or nearly so, ovoid, a
line and a half long, apiculate with 2 very short styles (of barely half
a line in length), septicidal; the valves at length falling away from
the persistent axis, which is dilated above into a thin lanceolate
appendage. Seeds oblong, several in each cell, comose at each end.

8. GEISSOILS, Labill.

Char. auctus. Calyx profunde quadripartitus, demum decidwus, esti-
vatione valvatus. Corolla nulla. Stamina 10-15 (vel 16 2), Gast
disci hypogynt inserta: filamenta filiformia, elongata: anthere
didymee, biloculares, loculis longitudinaliter dehiscentibus. Ovarium
liberum, conoideo-oblongum, biloculare: ovula plurima, biseriata, in
placentis bipartitis sursum imbricata, anatropa. Styli 2, filiformes,
bast connatt, decidui: stigmata simplicia terminalia. Capsula
lineari-oblonga, coriacea, bilocularis, ab apice septicide bivalvis.
Semina indefinita, complanata, adscendentia, imbricata; testa mem-
branacea reticulata sursum alata. Embryo in axi albuminis carnosi
rectus, eusdem fere longitudine ; cotyledonibus ovalibus subfoliaceis ;

SAXIFRAGACE ZS. 679

radicula infera.—Arbores Polynesice ; trunco crasso; stipulis inter-
petiolaribus ; foliis digitatis tri-quinquefoliolatis ; floribus purpureis ;
racemis laterulibus.

Getssors, Labill. Sert. Austro-Cal. p. 50, t. 50; Endl. Gen. p. 819.

1, GEISSOIS TERNATA, Sp. Nov. (Tab. 86.)
G. foliis trifoliolatis ; ovario glabro; staminibus scepissime 14 vel 15.

Has. Feejee Islands; common on the mountains of Muthuata and
Ovolau, between the altitudes of 500 and 2,000 feet.

“Tree 30 or 40 feet high, with a trunk of 2 feet in diameter;” the
branches terete, nodose; the branchlets petioles, &c., minutely silky-
puberulent when young. Leuves trifoliolate, opposite, on petioles from
half an inch to an inch long. Leaflets oblong or obovate-elliptical,
obtuse, acutish, or a little pointed at both ends, entire, glabrous, char-
taceous in texture in the dried specimens, from 2 to 5 inches long,
pinnately veined: the partial petioles 3 to 5 lines in length. Stipules
interpetiolar, canescently hirsute in the bud, in this state apparently
not costate, caducous, all the developed ones having disappeared.
Racemes lateral, solitary from the axils above the scars whence the
leaves of the previous season have fallen, or even from older wood,
spreading, many-flowered. Pedicels 2 or 3 lines long, scattered,
subtended by small and subulate caducous bractlets. Flower-buds
nearly 3 lines long, ovoid, glabrous. Calyx four-parted nearly to the
base; the segments triangular-oblong, acutish, thickish, valvate in
eestivation, spreading in anthesis, purplish-red, somewhat pubescent
inside, tardily deciduous. Corolla none. Stamens 14 or 15 in most
specimens examined, sometimes 12, or perhaps fewer, inserted at the
base of a thickened and entire gynobasic disk just at its junction
with the base or origin of the calyx, therefore obscurely perigynous :
filaments filiform, thickish towards the base, half an inch long, thrice
the length of the calyx, deep red, deciduous: anthers didymous,
fixed by the middle, incumbent, two-celled, destitute of a manifest
connective; the cells opening longitudinally. Grains of pollen glo-
bose, simple. Ovary free, conoidal-oblong, glabrous, terete, two-

680 PHANEROGAMIA.

celled; the axile placentze occupying the whole length of the cell,
two-lobed; the divergent lobes each bearing a dense row of ascending,
anatropous, compressed, and upwardly winged ovules. Styles 2,
filiform, equalling the filaments and of the same colour, somewhat
united at the base, deciduous: stigmas terminal, simple, obtuse or
subcapitate. Capsule linear-oblong, often curved, an inch or less in
length, terete, or obscurely compressed contrary to the dissepiment,
two-grooved, not ribbed, minutely apiculate, two-celled, septicidally
two-valved from the apex ; the valves coriaceous, at length separating
from the bipartible placenta. Seeds numerous in each cell, 2 lines
long, flat, oblong, upwardly imbricated; the testa membranaceous,
reticulated-cellular, loose, extended especially above into a broad
wing much larger than the nucleus, which is oblique, as respects the
axis of the wing. Embryo nearly the length of the fleshy albumen,
and occupying its axis. Cotyledons oval, flat, nearly foliaceous,
longer than the inferior radicle.

This is evidently a very close congener of Geissois racemosa, Labill.,
and it may prove to be no more than a form of that species. But
the leaves are all trifoliolate; the flowers are brightly coloured and
showy; the ovary glabrous; the capsule is not compressed nor ribbed
as in Labillardiere’s figure ; and the stamens, in all the flowers I have
examined, are more than ten, the number assigned by Labillardiere
to the New Caledonian species. Ten is an anomalous number for a
flower with a tetraphyllous calyx, and requires confirmation. In a
coloured drawing of the present species, made from the recent plant
by the late Mr. Agate, the equally anomalous number of nine stamens
is given to each flower, but I have found none with less than twelve
in the dried specimens. The seeds show no traces of the gummy
matter mentioned by Labillardiere—The genus is rightly placed by
Endlicher next to Belangera.

PLATE 86.—GEISSOIS TERNATA: in flower and fruit, of the natural
size. Fig. 1. Transverse section of a flower-bud. 2. Expanded
flower. 3. Anthers. 4. Pistil and disk. 5. Vertical section of the
same. 6. Anovule. 7. Transverse section of a capsule. 8. Dehis-
cent capsule. 9. A seed. 10. The same, with the nucleus divided
to show the embryo. 11. Transverse section through the embryo.—
The details variously magnified.

SAXIFRAGACEA. 681

9 BELANGERA, Cam.

1. BELANGERA SPECIOSA, Oamb.

Belangera speciosa, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 206, t. 117; DC. Prodr.
4, p. 11.

Polystemon pentaphyllus, Don, in Edinb. New Phil. Jour. 9, p. 95.
Lamanonia ternata, Velloz. Fl. Flum. 5, t. 104.

Has. Brazil; not uncommon near Rio Janeiro, and on the Organ
Mountains.

10. BAUERA, Kenn.

1. BAUERA RUBIOIDES, Andr.

Bauera rubioides, Andr. Bot. Rep. t. 198; Vent. Hort. Malm. t. 96; DC. Prodr.
4, p. 18.

B. rubicfolia, Salisb. in Ann. Bot. 1, p. 514, t. 10.

Has. New South Wales, near Sydney.

2. BAUERA MICROPHYLLA, Sieber.
Bauera microphylla, Sieber, P]. Exsic. N. Holl. no. 286; Seringe, in DC. 1. ¢.

Has. With the preceding species.

1l. BROUSSAISIA, Gaud.

Char. emend. et auct. Flores polygamo-dicici. Mas.—Calyx bres,

liber, quinquepartitus, lobis triangulatis. sew ovato-lanceolatis. Petala
171

682 PHANEROGAMIA.

5, imo calyci inserta, eodem majore, oblonga, cestivatione valvata,
acumine tnfleco, decidua. Stamina 10, cum petalis inserta: fila-
menta crassa subulata: anther ovate, biloculares, longitudinaliter
dehiscentes. Pistillum fere floris foeminei nisi stigma imperfectum,
ovula abortiva. Foem.—Calyx tubo ovoideo cum ovario connato,
limbo quinquedentato supero vel semisupero, persistente. Petala parva,
squamejformia, persistentia. Stamina nulla? Ovarium ovoideum,
quingueloculare: placentce crassce, e loculorum angulo centrali (ubi
via coalite) porrecte, bipartite, multiovulate. Stylus brevis vel sub-
nullus: stigma crassum, radiato-quinquelobum. Fructus baccatus,
pleiospermus. Semina horizontalia, oblonga, anatropa; testa mem-
branacea parce striato-reticulata nucleo conformi. Embryo in awi
albuminis carnost cylindricus, eodem dimidio brevior; cotyledonibus
brevissumis.—Arbuscule Sandwicenses ; ramis validis; foliis oppositis
seu verticillatis exstipulatis oblongis calloso-serratis, petiolis basi incras-
satis; cymis terminalibus sessilibus multifloris.

Brovssalsia, Gaud. Bot. Voy. Freyc. p. 479, t. 69 (pl. mase. tantum); DC. Prodr.
4, p. 17.

The materials in the present collection fortunately enable me
nearly to complete the characters of this interesting Hydrangeaceous
genus, and to demonstrate its real relationship to Adamia of Java
and Nepaul; which appeared very doubtful while Broussaisia was
supposed to have hypogynous stamens as well as a free ovary. But
what Gaudichaud, and subsequently Hooker and Arnott (the only
botanists who have hitherto examined Broussaisia) took for herma-
phrodite flowers are in fact male flowers; the ovary, although seem-
ingly well-formed, containing only abortive ovules. Figure 7 of
Gaudichaud’s plate, above-cited, does indeed represent a great number
of apparently good ovules in each cell; but his character, “ovula 1
vel 6 (constanter 62) in quolibet loculo,’” shows that in the flowers he
examined, as in our own, the ovules were for the most part rudimen-
tary. Those few even that are somewhat developed are doubtless
infertile. Moreover, the stamens and petals are perigynous, the base
of the calyx being adnate to the very base of the ovary. The actual
female flowers, indeed, are still desiderata: but we possess very
young fruit still bearing apparently all the organs of the flower
(unless there were stamens which have fallen away, leaving no

SAXIFRAGACES®. 683

trace), and with the calyx adnate nearly to the summit of the ovary,
as in the other genera of this tribe.*

There is something in the foliage, and in the strigose pubescence of
the young parts in Broussaisia that reminds one of Sauranja and
Draytonia (p. 206): the placentz and the seeds are much alike, and
the former abound in both with acicular raphides. The stamens and
petals of Draytonia, moreover, are slightly perigynous. Although
the Saurajee are doubtless not to be approximated to the Hydrangiee
on such grounds, yet they appear to be quite as much related to them
as to the Dilleniacee, or even to Clethra.

1. Broussatsia ARGUTA, Gaud. (Tab. 87.)

B. foliis oppositis obovato-oblongis ; dentibus calycis fructifert oblongis
superis stylo manifesto fere vel paullo brevioribus.

Broussaisia arguta, Gaud. l.c. (& Bot. Voy. Bonite, t. 9, f. 11, 127); DC. l.c.;
Hook, & Arn. Bot. Beech. Voy. p. 84.

Has. Sandwich Islands; Oahu; common on the mountains behind
Honolulu.

A large shrub, or small tree, with stout branches marked with very
large leaf-scars, when young hirsute, as is the inflorescence, with
strigose hairs, at length glabrate; the pith large. Leaves opposite,
obovate-oblong, 4 to 6 inches long, usually pointed with a slight acumi-
nation, closely serrate with fine and incurved callous teeth, tapering
at the base into the petiole, nearly coriaceous, thickly feather-veined
from the stout midrib, and the veins connected by a multitude of
transverse veinlets, glabrous above, the midrib and veins beneath
strigosely hirsute, especially when young. Petiole 6 to 18 lines long,
hirsute when young, stout, margined, channeled above, remarkably

* Just as this manuscript was about to pass into the printer’s hands, Gaudichaud’s
plate of Broussaisia pellucida, tab, 8, in the Atlas of the Voyage de la Bonite, fell
under my notice ;—showing that this botanist had ascertained the real characters of
Broussaisia, on his second visit to the Sandwich Islands. No letterpress of the phane-
rogamic botany of this work has been published, so far as I am aware, up to the time of
M. Gaudichaud’s recent death.

684. PHANEROGAMIA.

dilated and tumid at the insertion, the scar left by its fall 3 lines or
more in diameter. Stipules none. Leaf-buds naked. Inflorescence
terminal, forming a nearly sessile and crowded compound cyme, its
lower primary divisions subtended by rather large foliaceous bracts,
the larger about an inch in length, and resembling the leaves, but
mostly sessile. Pedicels from one to 3 lines long. Bractlets minute
and caducous. Flowers polygamo-dicecious. Male flowers.—Flower-
buds between 2 and 3 lines in diameter, globular. Calyx five-parted;
its divisions triangular or triangular-lanceolate, acute, equal, distant
in estivation in the full-grown flower-bud, much smaller than the
petals, and about half their length. Petals 5, distinct, ovate or
oblong, thickish, glabrous or nearly so, apparently greenish-white,
about 3 lines long and widely spreading when expanded, inserted by
a broad base into the base of the calyx, valvate in estivation, and
with a conspicuous inflexed acumination. Stamens 10, inserted with
the petals, nearly equal: filaments about the length of the corolla,
subulate, rather stout, dilated at the base, distinct: anthers ovate or
subcordate, two-celled, fixed by the base, somewhat introrse; the
cells opening longitudinally. Pollen-grains simple, globular, smooth.
Ovary ovoid or globose-ovate, free from the calyx except its very
base, contracted at the apex into a very short and thick style, which
is terminated by a depressed, undivided, apparently imperfect stigma;
within five-celled; the thick and spongy placente meeting but scarcely
coalescing in the centre, and bearing on their posterior faces numerous
minute rudiments of ovules, a few of which are more or less deve-
loped, but apparently sterile. Female flowers seen only in a fructified
state. Calyx-tube ovoid, connate with the ovary nearly or quite to
its summit; the limb divided into 5 oblong superior teeth of about a
line in length and persistent. Within and alternating with these
teeth are 5 shorter, but otherwise nearly similar, thickish, equally
persistent, scale-like lobes, which evidently answer to the petals:
whether they were more developed in anthesis is doubtful: they seem
to be complete. No vestiges of stamens or marks of their insertion are
to be found. Style conspicuous, in the fruit somewhat exceeding the
calyx-teeth, a line and a half long, columnar, terminated by a large,
obtusely and radiately five-lobed stigma. Fruit fleshy or baccate,
glabrous, smooth, globular, 3 lines in diameter (its colour not re-
corded), conspicuously pointed with the persistent style and stigma,
and crowned with the teeth of the calyx and the alternate petals or

SAXIFRAGACEA. 685

their vestiges, inferior, only its pointed apex free from the calyx; the
pericarp five-celled, although imperfectly so, as the retroflexed pla-
centz are scarcely if at all united in the axis at any stage: dissepi-
ments thin: cells alternate with the lobes of the calyx. Placentee
large, fleshy, two-parted, strongly reflexed into the cells, which they
almost fill (the seeds thus brought into contact with the pericarp),
their tissue replete with acicular rhaphides. The pericarp is lined
with a favose layer of tissue composed of elongated-linear and some-
what separable cells, giving it a striated appearance :* this also con-
tains an abundance of rhaphides. Seeds very numerous, horizontal,
a quarter of a line long, brownish, oblong, anatropous; the testa
membranaceous, coarsely striate-reticulated (the areole linear and
longitudinal), thin, conformed to the nucleus; the rhaphe inconspi-
cuous. Embryo cylindrical, next the hilum, half the length of the
seed, in the axis of fleshy albumen: cotyledons very small.

PLate 87.—BRoOUSSAISIA ARGUTA: a branch of the male plant, in
flower, and of the female plant, in fruit. Fig. 1. A male flower-bud.
2. Transverse section of the same. 3. An expanded male flower.
4. Vertical section of the same. 5. A fructified female flower. 6.
Vertical section of the same. 7. A fruit, with the parts of the flower
persistent. 8. Transverse section of the same. 9. Portion of the
pericarp, placenta, &c., more magnified. 10. A seed, detached. 11.
Vertical section of the same. 12. Embryo.—The details variously
magnified.

2. BROUSSAISIA PELLUCIDA, Gaud.

B. foliis ternato-verticillatis oblongo-lanceolatis oblongisve elongatis ; den-
tibus calycis fructiferi brevissimis semisuperis ; stylo brevissimo conico
seu nullo,

Broussaisia pellucida, Gaud. Bot. Voy. Bonite, t. 9 (excel. fig. 11, 12), sine deser.

Has. Hawaii, Sandwich Islands, in the district of Puna; probably
in woods,

* This structure is the same as Hydrangea, Schizophragma, Decumaria, Philadel-

phus, ke. Vid. Torrey & Gray, Flora of North America, 1, p. 593.
172

686 PHANEROGAMIA.

The Hawatian plant is merely mentioned in Dr. Pickering’s notes,
as if identical with that of Oahu. The only Hawaiian specimen in
the collection,—a fertile one with half-grown fruit,—differs from those
of Oahu, however, in having elongated-oblong leaves (6 to 8 inches in
length), all of them verticillate in threes. The fructiferous cyme is
more open; the ovoid ovary 1s free at its summit; the calyx being
adnate only to a little above the middle, its inconspicuous limb there
divided into 5, triangular and obtuse, appressed, very short teeth,
which are hardly discernible without a lens. Within and alternate
with these are as many similar teeth, of equal length, which repre-
sent the petals. Of stamens no vestiges are perceived. There is
properly no style; but the tapering apex of the fructified ovary is
contracted a little below the stigma; the latter accords with the pre-
ceding species, as do the placentze, &. Unless, therefore, Broussaisia
arguta is more variable than could be inferred without evidence, the
specimen before us must belong to a second species.—A recent com-
parison of our specimen with Gaudichaud’s plate, cited above, leaves
no doubt that it belongs to his Broussaisia pellucida, although the
plant figured by him has longer and narrower, and (judging from the
name) apparently much thinner leaves than ours, probably from
growing in deep shade. Further observation must determine whether
the characters assigned are constant, so as to specifically distinguish
the plant from B. arguta. This certainly cannot be the case if Gaudi-
chaud’s figures 11 & 12, representing the fruit with a columnar style,
really belong to his B. pellucida. The young fruit in our specimen
answers to that of his figure B.

12. CHRYSOSPLENIUM, Tourn.
1. CHRYSOSPLENIUM MACRANTHUM, Hook.

Chrysosplenium macranthum, Hook. in Lond. Jour. Bot. 1, p. 458, t. 16; Hook. f.
Fl. Antare. p. 281.
Hydrocotyle glechomoides, A. Rich, Mon. Hydrocot. t. 58, f. 17 (DC. Prodr. 4, p. 70)?

Has. Orange Harbour, Fuegia; common on the coast.

A tender, herbaceous plant, a foot high: the leaves apparently all

SAXIFRAGACES. 687

opposite.—This striking species is well illustrated by Sir William
Hooker. Although so lately published, it was gathered long ago by
Banks and Solander (and figured, under the name of @. elevatum,
according to Dr. Hooker), and probably also by Commerson. For, on
having my attention called to the point by a note of Dr. Pickering’s,
I think it is without much doubt the Hydrocotyle glechomoides of
Achille Richard, although this author’s account does not altogether
agree with the plant.

13. SAXIFRAGA, Linn.

1. Saxirraca MaGELLanica, Poir.

Sazxifraga Magellanica, Poir. Dict. 6, p. 686; DC. Prodr. 4, p. 25; Sternb. Rev.
Sax. t. 11, f. 1.
S. exarata, var. Hook. f. Fl. Antare. p. 280.

Haz. Orange Harbour, Fuegia; on the tops of mountains.
The collection comprises all three varieties of this Saxifrage;
which Dr. Hooker unites with the northern S. exarata,
2. SAXIFRAGA BICUSPIDATA, Hook. /.
Saxifraga bicuspidata, Hook. f. Fl. Antarct. p. 281, t. 97.
Has. Orange Harbour, Fuegia; on the tops of mountains; of rare
occurrence.
3. SAXIFRAGA PERUVIANA, Sternb.

Saxifraga Peruviana, Sternb. Rev. Sax. p. 55, t. 22; DC. Prodr. 4, p. 33.
S. Bonplandii, Don, Monogr. Sax. in Linn. Trans. 13, p. 481.
NS. stellata, Pav.; Don. 1. c. p. 480.

Has. High Andes of Peru; on rocks, above Casa Cancha, Alpa-
marca, &c. |

688 PHANEROGAMIA.

14. DONATIA, Forst.
1. DoNATIA FASCICULARIS, Forst.

Donatia fascicularis, Forst. Char. Gen. p. 10, t. 5; Hook. f. Fl. Antare. p. 281.
D. Magellanica, Lam. Ill. Gen. 1, p. 217, t. 51; DC. Prodr. 4, p. 53; Hook. Ic.

Pl. t. 16.
Polycarpon Magellanicum, Linn. f. Suppl. p. 115; Forst. in Comm. Gott. 9, p. 23, t. 3.

Has. Orange Harbour, Fuegia; common, forming broad and de-

pressed, dense patches.

Orv. CRASSULACE &.

1. TILLAA, Mich.
1. TILLHA VERTICILLARIS, DC.

Tillcea verticillaris, DC. Prodr. 8, p. 882; Hook. f. Fl. N. Zeal. p. 75.
T. muscosa ? Forst. Prodr. Fl. Ins. Austr. p. 11; A. Rich. Fl. N. Zel. p. 322.

Has. New South Wales, near Sydney and Hunter’s River. Also

Lord Auckland Islands (unless the specimens were misplaced).
2. TILLMA CONNATA, Ruiz & Pav.
Tillea connata, Ruiz & Pav. Fl. Per. & Chil. 1, p. 70, t. 106.
Has. Andes of Peru; common on walls, &., at and above Obrajillo.

“ Plant an inch or two in height.” The small specimen apparently

CRASSULACEA. 689

belongs to an erect plant; with ovate-oblong, obtuse and’ pointless
leaves; the floral ones as long as the fructiferous pedicels; which
however do not exceed the calyx in length. So that the Species is
probably distinct from 7. rubescens, H.B.K. Petals shorter than the
ovate and somewhat pointed sepals. Carpels two-seeded.

3. TILL@A (BULLIARDA) MoscHata, DC.

Crassula moschata, Forst. in Comm. Geett. 9, p. 26.

Bulliarda Magellanica, DC. in Bull. Philomath. no. 49.

B. moschata, D’Urv. in Mem. Soc. Linn. Par. 4, p. 618; Gaud. Bot. Freye. Voy. p.
138; Hook. f. Fl. Antare. p. 15 & 278.

Tillea moschata, DC. Prodr. 8, p. 882; Hook. Ic. Pl. t. 535 > Hook. f. Fl. N. Zeal.
p. 79.

Has. Orange Harbour, Fuegia; on rocks along the coast. Lord
Auckland Islands.

2. TETRAPHYLE, Eckl. & Zeyh.
1. TETRAPHYLE muscosA, Eckl. & Zeyh.

Tetraphyle muscosa, Hckl. & Zeyh. Enum. Pl. Afr. Austr. p. 294.
Crassula muscosa, Linn. Spec. Pl. ed. 2, p. 405, fide Eckl. & Zeyh.

Has. Cape of Good Hope, near Cape Town.

3. CRASSULA, Linn., Haw.

1. CRASSULA TETRAGONA, Linn.

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.

2. CRASSULA SCABRA, Linn.

Has. Cape of Good Hope, in the vicinity of Cape Town.
178

690 PHANEROGAMIA.

3. CRASSULA SQUAMULOSA, Willd.

Has. Cape of Good Hope; with the preceding species.

4, CRASSULA PELLUCIDA, Linn.

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.

5. CRASSULA CILIATA, Linn.

Has. Cape of Good Hope, near Cape Town.

4, ROCHEA, DGC.
1. RocHEA opoRATISsIMA, DC.

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.

5 KALANCHOE, Adans.
1. KALANCHOE BRASILIENSIS, Camb.
Kalanchoe Brasiliensis, Cambess. in St. Hil. Fl. Bras. Mer. 2, p. 196.

Has. Rio Janeiro; common on rocks along the shore of the bay.

6 BRYOPHYLLUM, Salisb.
1. BryopHyLLuM CALYcINUM, Salisb.

Has. Philippine Islands, in the vicinity of Bajios, Luzon.

CRASSULACESA. 69]

7. UMBILICUS, Da
1. Umsinicus Penputinus, DC.

Has. Madeira; on the summit of Pico Ruivo.

8. SEDUM, Linn.
1. SEepuM NupUM, Ait.

Has. Madeira; along the coast east of Funchal.

2. SepuM PERUVIANUM, Sp. Nov.

S. glabrum, pumilum; caulibus adscendentibus ; foltis sessilibus ovatis
oblongisve planis ; cyma pauciflora subracemiformi ; petalis (ut videtur
albis) ovatis acuminatis.

Has. Andes of Peru, from Obrajillo to Culnai; on rocks.

The specimens in the collection are extremely meagre and imper-
fect. There is also one from the same region in the Hookerian her-
barium, gathered probably by Matthews or M’Lean. The root is
most likely annual. Stems ascending, low (3 or 4 inches high), fleshy,
glabrous, as is the whole plant. Cauline leaves sessile, scattered, ovate
or oblong, plane, but very fleshy, small. Flowers few, racemose-
cymose; the terminal one sessile, the others short-pedicelled. Sepals
5, ovate-lanceolate, somewhat carinate. Petals ovate or broadly oval-
obovate, acuminate, nearly 8 lines long, one-third longer than the
calyx, carinate, one-nerved, scarious-persistent, apparently white.
Stamens 10. Carpels ovate, many-seeded.

Orv. UMBELLIFER &.

1 HYDROCOTYLE, Tourn.
1. HyprocotyLe interruPtTA, Mul.

Has. Sandwich Islands: Oahu; in marshes, near Honolulu.
Hawaii, in the district of Waimea.

This was also collected in Beechey’s voyage. It accords in all
respects with the plant of the United States: but the fruit, although
not emarginate, is by no means subattenuate at the base. I have seen
no specimens from the Pacific coast of North America: an allied
Species, more nearly related to Z vulgaris however, is found in Cali-
fornia and Western Mexico.

2. HyprocotyLe mopesta, Cham. & Schlecht.

Has. Peru, near Callao. A monstrous state; the umbels foliose-
proliferous.

3. HyprocoryLe Bonariensis, Lam.

Has. Brazil, near Rio Janeiro. Callao, Peru. The many-flowered
form: H. multiflora, Ruiz & Pav.

4. HYDROCOTYLE QUINQUELOBA, Ruiz & Pav.

Has. Brazil; in the Organ Mountains.

UMBELLIFERZ. 693

5. HYDROCOTYLE LEUCOCEPHALA, Cham. & Schlecht.
Has. Brazil, near Rio Janeiro. Also, Savaii, one of the Samoan
Islands.
6. HyprocoryLe LAXIFLORA, DO.

Has. New South Wales; in the neighbourhood of Hunter’s River.

7. Hyprocoryie Asratica, Linn.

Has. Luzon, near Manilla. Hunter's River, New South Wales.
Ovolau, Feejee Islands; around houses. Tongatabu. Savaii, one of
the Samoan Islands; “only seen around houses.” Bay of Islands,
New Zealand (the ordinary plant, and a diminutive state, with the
leaves only 4 or 5 lines in diameter). Brazil, near Rio Janeiro.

8. HypROCOTYLE RANUNCULOIDES, Linn. f.
Has. Chili, near Valparaiso. Without flowers or fruit. Appa-
rently also the HZ. batrachwides, DC.

9. HypROCOTYLE PLEBEIA, 2. Br.

Has. New South Wales; in the vicinity of Hunter’s River.

10. HyprocoTyLe Hirta, 2. Br.

Has. New South Wales, near Sydney.

11. HyprocotyLe ELONGATA, A. Cunn.
Hydrocotyle loader A. Cunn. Bot. N. Zeal.; Hock. f. Fl. N. Zeal. p. 84.

Has. Bay of Islands, New Zealand.
174

694 PHANEROGAMIA.

12. HypRocoTYLE MICROPHYLLA, A. Cunn. I. c.

Has. Bay of Islands, New Zealand.

13. HyprocoryLE PEDUNCULARIS, FR. Br.

Has. New South Wales, near Sydney. (The var. y. Hook. f.)

2, DIDISCUS, DC.
1. Dipiscus ALBIrLorus, DC.

Has. New South Wales; near Sydney, and on Hunter’s River.

3 TRACHYMENE, Rudge.
1. TRACHYMENE OVALIS, DC.
Haz. New South Wales, near Sydney: also (a variety with the
stem almost glabrous) on Cook’s River.
2. TRACHYMENE MYRTIFOLIA, Sieber.
Has. New South Wales; with the preceding species, which pro-

bably is only a variety of this.

3. TRACHYMENE LANCEOLATA, Rudge.

Trachymene lanceolata, Rudge, in Linn. Trans. 10, p. 300; DC. 1. ¢.; Hook. Bot.
Mag. t. 3334.
Azorella lanceolata, Labill. Fl. N. Holl. 1, p. 74, t. 99.

UMBELLIFERA. 695

Has. New South Wales; in the vicinity of Sydney and New-

ington.

4, ‘TRACHYMENE LINEARIS, Spreng.

Trachymene linearis, Spreng. ; DC. Prodr. 4, p. 78.
Azorella linearifolia, Cay. Ic. Pl. 5, p. 485.

Has. New South Wales; in the vicinity of Sydney.

5. 'TRACHYMENE ERICOIDES, Sieber.

Has. Sydney, Newington, and Hunter’s River, New South Wales.
The stems are mostly puberulent, and the leaves often puberulent-

scabrous. The species apparently includes both Z subvelutina and
f, tenuis, of DeCandolle.

4, ASTROTRICHE, DGC.

1. ASTROTRICHE LATIFOLIA, Benth.
Astrotriche latifolia, Benth. Pl. Hugel. p. 55.

Has. New South Wales; probably from the neighbourhood of
Sydney or Hunter's River.

5 XANTHOSIA, Rudge.
1. XantHosiA PiLosa, Rudge.

Aanthosia pilosa, Rudge, in Linn. Trans. 5, p. 861; Benth. Pl. Hugel. p. 55.
X. hirsuta, DC. Prodr. 4, p. 74.

Has. New South Wales; near Sydney and Woolongong: one form

696 PHANEROGAMIA.

in foliage approaching Leucolena pannosa, Benth. 1.c.; the other
remarkably barbate-villous.

2. XANTHOSIA MONTANA, Sieber, in DC. 1. c.

Has. New South Wales; in the neighbourhood of Sydney.

3. XANTHOSIA TRIDENTATA, DC. 1. ¢.

Has. Near Sydney, New South Wales.

6 BOWLESIA, Ruiz & Pav.
1. BowLesta TROPAOLIFOLIA, Gillies.
Bowlesia tropeolifolia, Gillies, in Hook. Bot. Mise. 1, p. 325; Gay, FI. Chil. 3, p. 75.

Has. Chili; on the Cordilleras above Santiago. Near B. tenera.*

2. BowLEsta LoBATA, Ruiz & Pav.

Bowlesia lobata, Ruiz & Pav. Fl. Per. & Chil. t. 251; DC. Prodr. 4, p. 75.
B. acutangula, Benth. Pl. Hartw. p. 186.

Has. Andes of Peru, near Obrajillo. (Lobes of the leaves acutish. }

3. BowLEsiA PALMATA, Ruiz & Pav. l.c.

Var. B. fructu nune nudato parce stelligero, nunc setis glochidatis his-
pido !

* Elsneria crategifolia, Walp. in Rel. Meyen. p. 846, t. 7, appears to be the Bow-

lesia dichotoma, Popp., to which B. multiradiata, Colla, and B. elata, Clos, may also
belong.

UMBELLIFERSZ. 697

Has. Andes of Peru, in the neighbourhood of Obrajillo.

This appears to be the B. palmata of Ruiz and Pavon. In one
specimen the fruit is mostly glabrate, when young sparingly beset
with sessile stellate tufts of short hairs: occasionally one or more of
these tufts are raised on a short pedicel or bristle; while in another,
otherwise similar specimen the fruits are rather copiously beset with
such glochidiate bristles, often of considerable length. From the forks
of the stem, there is occasionally produced an abnormal fusiform
body, appearing like a diseased fruit, either subsessile or long-pedun-
cled, 4 or 5 lines long, and armed all over with long glochidiate
bristles. .

7. AZORELLA, Lam.

AZORELLA, Lam. IIl. Gen. t. 484, f. 1; Cav. Ic. t. 482, f. 2; DC. Prodr. 4, p. 77.

Botax, Commers.; Juss. Gen. p. 226; DC. Prodr. 2, p. 78; Spreng. Umb. p.

Fracosa, Ruiz & Pav. Fl. Peruv. & Chil. Prodr. p. 43.

Cuamitis, Banks & Soland. ined.; Geertn. Fruct. 1, p. 94, t. 22.

Pecropuytum, H. B. K. Noy. Gen. & Spec. 5, p. 28, t. 425.

Pozoa, subgen. Scutzertema, Hook. Fl. Antare. p. 15, t. 11, & Fl. N. Zeal. p. 89,
t. 18.

oo
ew

It appears to me that Sprengel was quite right in uniting Bolax
and Azorellu, both names probably imposed by Commerson, but the
latter first published. The types of the two genera were also viewed
as one (Chamitis) by Banks and Solander, as also by Cavanilles.
Their habit is the same: A. Selago, indeed, closely resembles Bolax
glebaria; and the difference in their fruit is perhaps not greater than
between some species of -Azorella proper. Moreover, an Andine
Chilian species (A. madreporica), having the thickened margins of the
mericarps more or less recurved, often as much so as in some fruits
of Bolax glebaria, manifestly combines the two genera. The entire
absence of teeth to the. calyx will barely distinguish Pectophytum,
H. B.K., as a section of this genus; the limb being obsolete in Bolax.
Dr. Hooker has fully noticed the similarity of the fruit of his Pozoa,
subgen. Schizeilema to Azorella, and has until lately retained A. Ranun-
culus in this genus: in finally excluding it, and referring it, along

175

698 PHANEROGAMIA.

with two kindred species, to Pozoa, he has been guided principally by
habit; yet in this these plants do not widely differ from some Peru-
vian Fragose. Moreover if Pozoa itself may not be distinguished by
its striking crateriform gamophyllous involucre, and its monoecious
flowers, it may also fall into Azorella.

§ 1. SCHIZEILEMA. (Pozoa, subgen. Schizeilema, Hook. f. 1. ¢.)
1. AzorELLA Ranuncutvus, D' Urv.

Azorella Ranunculus, D’Urv. in Mem. Soc. Linn. Par. 4, p. 614; Gaud. Bot. Voy.
Freye. p. 186; DC. Prodr. 4, p. 77; Hook. f. Fl. Antare. p. 285, t. 98.
Pozoa (Schizeilema) Ranunculus, Hook. f. Fl. N. Zeal. p. 85, adn.

Has. Orange Harbour, Fuegia.

The mature fruit is more didymous than in Dr. Hooker’s otherwise
perfect figure. The other species of the group are A. reniformis, of
the Auckland and Campbell Islands, and A. trifoliolata, Hook. f. (in
tab. supra cit.), of New Zealand. In habit they much resemble Hydro-
cotyle.

§2. CHAMITIS. (Chamitis, Banks & Soland. excl. spee., Gaertn. Azorella, Lam., DC.)

2. AZORELLA FILAMENTOSA, Lam.
Azorella filamentosa, Lam. Dict. 1, p. 344, & Ill. Gen. t. 189; DC. Prodr. 4, p. 77;

Hook. Ic. Pl. t. 451; Hook. f. Fl. Antare. p. 283.
Chamitis inteyrifolia (Banks & Soland.) Geertn. Fruct. 1, p. 94, t. 22.

Has. Orange Harbour, Fuegia; on the mountains.

3. AZORELLA LYCOPODIOIDES, Gaud.

Azorella lycopodioides, Gaud. in Ann. Sci. Nat. 5, p. 105, t.3; DO. l.c.; Hook f.
Fl. Antare. p. 284.

Has. Orange Harbour, Fuegia; on the tops of mountains.

UMBELLIFERSE. 699

4. AzorELLa SELAGO, Hook. f.

Azorella Selago, Hook. f. Fl. Antare. p. 284, t. 99.

Has. Orange Harbour, Fuegia; on the tops of mountains.

5. AZORELLA MONANTHOS, Clos?

Azorella monanthos, Clos, in Gay, Fl. Chil. 8, p. 79?

Haz. On the Andes of Chili, above Santiago. (Without flowers or
fruit.)

6. AZORELLA APODA, Sp. Nov.

A. dense ceespitosa; foliis arctissime imbricatis basi vaginantibus apice
subspatulato crassis triangulato-subulatis pungenti-mucronulatis glabris,
floralibus basi filamentosis; wumbella bi-quadriflora sessili; floribus
Fructibusque subsessilibus folia haud superantibus ; calyce glaberrimo,
dentibus brevissimis obtusis; mericarpiis utriculosis dorso subcom-
pressis.

Has. Andes of Chili, above Santiago, not far from the snow-line.

The plant forms very compact tufts, from an inch to 3 inches high ;
the stems (2 or 3 lines thick) covered with very densely imbricated
leaves, with appressed sheathing bases; their upper and slightly dilated
portion, or limb, triangular-subulate, or oblong, 12 or 2 lines in length,
glabrous, very thick, mostly obtuse, but pungently mucronulate ; those
next the flowers filamentose at the base inside. Involucre of 2 to 4
small and scarious leaves, their dilated bases somewhat united and
filamentose. Umbel sessile even in fruit, of 2 to 4 rather large
flowers; the pedicels very short, not so long as the ovary, so that the
flowers are not protruded beyond the leaves, but barely equal them in
length. Calyax-tube wholly glabrous, utricular; the 5 teeth very short
and obtuse, persistent. Petals oblong, plane. Stylopodium large.

700 PHANEROGAMIA.

Styles elongated. Only one flower of each umbel usually fructifies.
Fruit not exceeding the leaves at the summit of the shoot, turgid,
globular, but dorsally somewhat compressed, 2 lines long, wtricular (as
in A. Gilliesit), the entirely loose and membranaceous epicarp smooth
and free from ribs or markings: on detaching the membrane, how-
ever, the five filiform ribs are seen on the body of the compressed meri-
carp, three on the back and one on each margin. Vittee none.

This can hardly be any state of Azorella ceespitosa of Cavanilles,
which is described and figured as having, at least in the fruiting state,
a conspicuously exserted peduncle and slender pedicels, as well as
squarrose-spreading leaves. It is much more like the Antarctic plant,
however, as described by Dr. Hooker under this name; but that is
said to have the calyx-tube hairy; and Dr. Hooker would scarcely fail
to mention the utricular character, which is manifest even in the im-
mature fruit. Moreover the leaves of our plant are appressed, rather
than spreading.

§ 3. BOLAX. (Bolax, Commerson, DC., Hook. f.)
7. AZORELLA MADREPORICA, Clos.

A. caulibus multicipitibus in ceespitem densum depressum coherentibus ;
petiolis densissime imbricatis appressis oblongis planis aut truncatis
aut superioribus lamina brevissima tripartita intus filamentoso-barbata
donatis, in ramis sterilibus petiolis nunc elongatis patentibus lamina
bis trifida majoribus ; umbella subsessili 2-5-flora ; involucro 4— d-phyllo
scarioso pedicellis cequilongo; floribus fructibusque ultra folia vix
exsertis ; calyce glubro, dentibus oblongis persistentibus ; fructu quad-
rilobo, mericarpiis obovato-orbiculatis plano-compressis marginibus
merassatis pl. m. recurvis inde dorso concavis.

Azorella madreporica, Clos, in Gay, Fl. Chil. 3, p. 79.
Has. Chili; on the high Andes, above Santiago, near the snow-line.

Plant growing in extremely dense and depressed tufts, consisting of
numerous compactly matted stems, an inch or two in length, from a

UMBELLIFERS. 701

thick and perpendicular root, clothed with dilated and appressed,
densely imbricated petioles. Many of these are destitute of lamina:
towards the summit of the stems, however, they usually bear a very
small and three-parted lamina, which is bearded inside, at least when
young, either sparsely or copiously, with long and filamentous bristles ;
its segments ovate and entire. From the summit of some of these
stems, when sterile, arise more or less open shoots, bearing spreading
leaves with longer and narrower petioles, terminated by a lamina of
one or two lines in diameter, its three lobes again three-cleft. Uimbel
subsessile and included among the leaves at the summit of the fertile
stems, 2—5-flowered; the flowers and fruits on short pedicels, but
raised only slightly above the level of the dense mass of leaves.
Involucre of 4 or 5 small and scarious lanceolate leaflets, unequal,
scarcely if at all united at the base, about the length of the pedicels.
Calyx glabrous; the teeth oblong, persistent. Fruit four-lobed; the
mericarps obovate-orbicular, or oval, a line and a half long, nearly
plano-compressed, except that the broad and thickened margins are not
unfrequently more or less recurved, so as to render the mericarp con-
cave on the back and leave wide sinuses at the commissures. The
substance of the pericarp is thick and corky, destitute of vitte, with
no utricular epicarp, the 5 slender ribs usually evident. Not unfre-
quently the fruit has three mericarps; each of just the form of those
of Bolax glebaria, except that they are more compressed. The plant
exudes a resin as copiously as the Balsam Bog and Laretia.

Azorella bolacina, Clos, 1. ¢., of which I have not the fruit, is appa-
rently nearly allied to the present species. Its leaves are sometimes
entire. May not such a state be the A. ceespitosa of Cavanilles ?

- §&, AZORELLA GLEBARIA.

Bolax glebaria, Commerson, in Juss. Gen. p. 266; Gaud. in Ann. Sci. Nat. 5, p.
104, t. 8, f. 2, & Bot. Freye. Voy. p. 186; DC. Prodr. 4, p. 78; Hook. Ic. Pl.
t. 492; Hook. f. Fl. Antare. p. 285.

B. yummifera & B. complicata, Spreng. Prodr. Umb. p. 9.

Hydrocotyle gunmifera, Lam. Dict. 8, p. 156.

Azorella tricuspidata, Lam. Ul. Gen. t. 189, f. 2.

A. ccespitosa, Vahl, Symb. 3, p. 48; Willd. Spec. Pl. 1, p. 1865, ex parte, non Cav.

Has. Orange Harbour, Fuegia; common. (Both the entire and

the trifid-leaved forms.)
176

702 PHANEROGAMIA.

The habit and mode of growth, as well as the synonymy of the
remarkable Balsam-bog are fully elucidated by Dr. Hovker. The
fruit is well figured by Sir William Hooker.

§4. PECTOPHYTUM. (Pectophytum, H. B. K., DC.)
9. AZORELLA DIAPENSIOIDES, Sp. Nov.

A. dense ceespitosa, nana, glabra; foliis imbricatis confertissimis, lamina
patente oblonga integerrima petiolo dilatato basi amplexicauli breviore ;
umbella 3-T-flora sessili inter folia summa; involucro scarioso 6—9-
jido ; floribus brevissime pedicellatis ; fructibus dorso subcompressis.

Has. On the crest of the Andes of Peru, near Casa Cancha and
Alpamarea. (High Andes of Peru, M’Lean, in herb. Hooker.)

A dwarf and depressed, densely tufted species, with somewhat the
mode of growth and foliage of Diapensia Lapponica, but on a smaller
scale, an inch or two in height, glabrous throughout, except a very few
long hairs at the base of the leaves. Leaves imbricated, dense; the
scale-like petiolar portion triangular-dilated, 3 to 5 lines long, ap-
pressed and sheathing at the base, narrowed upwards, then a little
dilated into an oblong and entire, spreading lamina, of a thick and
firm texture, 12 or 2 lines long, sometimes becoming linear, obtuse,
somewhat mucronate. Umbel sessile in the crown of upper leaves, by
which it is exceeded in length, both in flower and fruit. Involucre
scarious, 6—9-cleft, usually unequally so; the segments oblong, obtuse,
not ciliate. Flowers 3 to 7, on very short pedicels, apparently all
perfect. Limb of the calyx obsolete and truncate. Petals apparently
greenish-white, oblong, plane. Fruit barely a line and a half long,
globular, but moderately dorsally compressed, not utricular ; the meri-
carps obscurely five-ribbed.

In Dr. Pickering’s notes, a species allied to this is mentioned, from
the same district, with the leaves entire, lanceolate and densely
tomentose above. Two other species, thought to be A. (Fragosa)
multifida, and A. (Fragosa) cladorhiza, are said to be common at
Baiios, and from Casa Cancha to Culnai. But no corresponding speci-
mens are found in the collection.

UMB
ELLIFER AZ. 703

8. MULINUM, Bes.

1. Mutinum protirerum, Pers.
Mulinum protiferum, Pers. Ench. 1, p. 309; DC. Prodr. 4, p. 79; Hook. Bot. Mise.
1, p. 827.
Selinum proliferum, Cav. Ic. Pl. 5, p. 58, t. 486, f. 1.

Has. Rio Negro, North Patagonia: common on the plains.

The specimens are in fruit, the umbels not proliferous, most of
them shorter than the leaves, which are larger and their slender
divisions longer than in the figure of Cavanilles.

2. Muxinum spinosum, Pers.

Mulinum spinosum, Pers. Ench. 1, p. 8309; DC. Hook. Bot. Mise. 1, p. 327.
Selinum spinosum, Cav. Ic. Pl. 5, p. 59, t. 487, f. 1.

Has. Andes of Chili, above Santiago. (In fruit.)

3. Muinum cunEATUM, Hook. & Arn.

Mulinum cuneatum, Hook. & Arn. Bot. Beech. Voy. p. 26, & Bot. Mise. 3, p. 347.
Fragosa spinosa, Ruiz & Pav. Fl. Per. & Chil. 8, p. 27; DC. Prodr. 4, p. 76.

Has. Chili; on the highlands above Valparaiso. (Without flowers
or fruit, the latter still unknown.)

4, Muinum vticinum, Gill. & Hook.

Mulinum ulicinum, Gillies & Hook. Bot. Misc. 1, p. 328, t. 64, & 3, p. 347.
M. Echinus, Presl, in DC. Prodr. 3, p. 79?

Has. Andes of Chili, near the snow-line. (A doubtful specimen.)

704 PHANEROGAMIA.

9 LARETIA, Gil. & Hook.
1. LARETIA ACAULIS, Gill. & Hook.

Laretia acaulis, Gillies & Hook. in Bot. Misc. 1, p. 829, t. 65; Hook. & Arn.
Bot. Mise. 3, p. 847; Gay, Fl. Chil. 8, p. 106.
Selinum acaule, Cay. Ic. Pl. 5, p. 59, t. 487, f. 2.

Has. High Andes of Chili, above Santiago, near the snow-line.

The specimens have ripe fruit, which very well accords with
Hooker's figure, except that the ribs are not so salient. The plant

forms dense tufts, “in the manner of Diapensia,” and exudes a copious
resin.

10. ASTERISCIUM, Cham. & Schlecht.
1. Asteriscium CHILENSE, Cham. & Schlecht.

Asieriscium Chilense, Cham. & Schlecht. in Linnea, 1, p. 254, t. 5, f. 1; Hook. Bot.
Misc. 1, p. 332, t. 67; Gay, Fl. Chil. 3, p. 100.
A. Chilense & A. Peppigiti, DC. Prodr. 4, p. 82.

Has. Chili: in maritime sands; also on hills, in the neighbourhood
of Valparaiso.

11. ACTINOTUS, Labitl.

1. Actinotus Hetrantut, Labill.

Actinotus Helianthi, Labill. Pl. N. Holl. 1, p. 67, t. 92; DC. Prodr. 4, p. 83.
Eriocalia major, Smith, Exot. Bot. 2, p. 37, t. 78.

Has. Sydney and Hunter’s River, New South Wales.

UMBELLIFERA, 705

2. AcTINoTUS minor, DC. 1. ¢.

Eriocalia minor, Smith, Exot. Bot. p. 88, t. 79.

Has. New South Wales; in the neighbourhood of Sydney.

12, SANICULA, Tourn.
1. SanicuLa Sanpwicensis, Sp. Nov. (Tab. 88.)

S. foliis ambitu rotundatis profunde palmati-3-5—partitis, segmentis
bi-trifidis incisisque, lobis oblongis vel fol. superiorum lanceolatis
argute inciso-serratis, serraturis aristulatis; wmbellis tri-quinque-
radiatis scepe proliferis ; floribus masculis breviter pedicellatis.

Has, Sandwich Islands: Hawaii, on Mouna Loa, at the elevation
of 8,000 feet: East Maui; on the mountains.

Root perpendicular, fusiform, 6 to 10 inches long. Stem a foot or
two in height, angular, glabrous, as is the whole plant. Leaves
rotund in circumscription, about 2 inches in diameter, palmately 3-5-
parted and almost divided; the cuneate segments 2—3-cleft and incised
or laciniate, sometimes five-cleft; the lobes oblong or of the upper
leaves lanceolate or oblong-linear, more or less incised, very sharply
and incisely serrate, with aristulate teeth. Petioles scarious-dilated and
clasping at the base; the uppermost and rameal leaves sessile, and
passing into the two-leaved involucres. Uinbels of 3 to 5 rays, which
are longer than the involucels, and frequently proliferous. Umbellets
22 lines in diameter, exceeding the oblong-lanceolate and entire divi-
sions of the involucel, globular, densely many-flowered. Flowers
yellow, the male flowers short-pedicelled and exterior; the female
flowers sessile; their filiform styles exserted and recurved. Fruit (of
which little occurs on the specimens) ovate, 2 lines long, echinate
throughout with long and stout hooked prickles.

This interesting addition to the very few Umbellifere of the Sand-
177

706 PHANEROGAMIA.

wich Islands has not been found, so far as I know, by any other
collectors. It is allied to S. Menziesti, of Oregon and California; but
its leaves are much more dissected and of a firmer texture.

PLATE 88.—SANICULA SANDWICENSIS: of the natural size. Fig. 1.
A sterile flower. 2. Petal, from the same. 3. Stamen. 4. Vertical

section of a fertile flower. 5. Fruit. 6. Vertical section of a meri«
carp and seed. 7. Embryo.—The details magnified.

13. ERYNGIUM, Tourn.

1. Eryneium coronatum, Hook. & Arn.

Eryngium coronatum, Hook. & Arn. in Bot. Misc. 8, p. 350.

Has. Rio Negro, North Patagonia: procumbent, on the river-banks.

2. ERYNGIUM EBRACTEATUM, Lam.

Has. Rio Negro, North Patagonia; with the preceding species.

3. ERYNGIUM PANICULATUM, Cav.

Has. Chili (the var. Chilense, DC.): a common and conspicuous
plant around Valparaiso.

14. APIUM, Linn., Hofim.
1. APIUM AUSTRALE, Petit-Thouars.

Apium australe, Petit-Thouars, Fl. Trist. d’ Acunh. p. 43; Hook. f. Fl. N. Zeal. p. 86.
A. prostratum, Labill. Pl. N. Holl. 1, p. 76, t. 103; Vent. Hort. Malm. t. 81.

A. Chilense, Hook. & Arn. in Bot. Mise. 8, p. 353.

A. graveolens, DO. quoad pl. Antarc.; Hook. f. Fl. Antare. p. 287.

UMBELLIFERGS. 707

Has. Rio Negro, North Patagonia. Orange Harbour, Fuegia;
abundant on the coast. Bay of Islands, New Zealand. Also (the
var. 3., Hook. f.) Hunter’s River, New South Wales.

Different as the specimens of the var. 8. (the Petroselinwm pros-
tratum, DC.) appear to be from the var. a. (which represents in the
southern hemisphere the Celery of Northern Europe), Dr. Hooker
states that he has seen the two growing from the same stem, in Tas-
mania. Dr. Hooker speaks of its excellence both as a salad and a
pot-herb in the Antarctic regions. Our naturalists do not refer to its
qualities or use. The fruiting specimens all have very thick ribs to
the fruit, which is larger than in A. graveolens. No other positive
distinction has been noted.

2. APIUM GRAVEOLENS, Linn.

Has. Rio Janeiro, Brazil; among rocks on the shore. Doubtless
escaped from cultivation.

156. PETROSELINUM, Aofin.
1. PETROSELINUM PEREGRINUM, Lag.
Petroselinum peregrinum, Lag.; Koch, Umb. p. 128; DC. Prodr. 4, p. 102.

Has. Madeira; on rocks at Corral.

16. HELOSCIADIUM, Koch,
1. HELoscraDIUM LEPTOPHYLLUM, DC.
Has. Brazil, near Rio Janeiro. Peru, around Obrajillo. New

South Wales, near Sydney and Hunter’s River; where probably it is
not indigenous.

708 PHANEROGAMIA.

2. HELOSCIADIUM LACINIATUM, DC.

Has. Andes of Peru, above Obrajillo.
Depauperate specimens, too near the preceding species; but the

cauline leaves are all long-petioled, and the lobes less slender.

3. HELOSCIADIUM RANUNCULIFOLIUM, DC.

Helosciadium ranunculifolium, DO. Prodr. 4, p. 105? Hook. & Arn. in Bot. Mise.
3, p. 854.

A. lateriflorum, Hook. & Arn. Bot. Beech. Voy. p. 26, non Koch.

Has. Chili: in high ravines, above Valparaiso.

17. AMMI, Tourn.
1. Ammi masus, Linn.

Has. Madeira; on the coast east of Funchal.

2. Amoi Visnaca, Lam.

Has. Around Valparaiso, Chili. Doubtless introduced from Europe.

18. BUPLEURUM, Journ.

1. BupLevrum sauictrouium, Solander.

Haz, Madeira; on the coast, near Funchal.

UMBELLIFERA. 709

19. LICHTENSTEINIA, Cham. & Schlecht.
1. LICHTENSTEINIA LACERA, Cham. & Schlecht.
Lichtensteinia lacera, Cham. & Schlecht. in Linnea, 1, p. 894; DC. Prodr. 4, p. 135.

Has. Cape of Good Hope, in the vicinity of Cape Town.

2. LICHTENSTEINIA PYRETHRIFOLIA, Cham. & Schlecht. 1. c.?

Has. Cape of Good Hope, near Cape Town. (In fruit.)

20. FG@NICULUM, Adans.
1. F@NICULUM VULGARE, Geertn.

Has. Madeira; on the coast, near Funchal. (Fennel.)

21. LIGUSTICUM, Linn.
1. Ligusticum Pansin, Bert. in DC.

Has. Chili, near Valparaiso. (With immature fruit.)

22, ANISOTOME, Hook.f
1. ANISOTOME LATIFOLIA, Hook. /f.

Anisotome latifolia, Hook. f. Fl. Antarc. p. 16, t. 8.
178

710 PHANEROGAMITA.

Has. Lord Auckland Islands.
This most striking and characteristic Umbelliferous plant (of which

only the foliage occurs in our collection) is admirably illustrated by
Dr. Hooker.

2. ANISOTOME ROSHFOLIA, Hook.

Anisotome (Eustylis) rosefolia, Hook. f. Fl. N. Zeal. p. 90.
Angelica ? rosefolia, Hook. Ic. Pl. t. 581.

Has. New Zealand; at Waia-ruru Bay. (As a genus, the name of
Eustylis is preoccupied in Iridec.)

23. CRITHMUM, Zourn.
1, CritHMUM MARITIMUM, Linn.

Has. Madeira; on the coast, near San Vicente.

24. PEUCEDANUM, Linn.

1. PEUCEDANUM CAPILLACEUM, Thunb.

Peucedanum capillaceum, Thunb. FI. Cap. p. 257; DC. Prodr. 4, p. 178.

Haz. Cape of Good Hope, near Cape Town. (With immature
fruit.)

25. BUBON, Linn.
1. Buson Gatpanum, Linn.

Has. Cape of Good Hope, in the immediate vicinity of Cape Town.

UMBELLIFERAE. 711

26. DAUCUS, Tourn.
1. Daucus Carota, Linn.

Has. Bay of Islands, New Zealand. (Probably strayed from culti-
vation, and not naturalized. It is not mentioned by Dr. Hooker.)

2. Daucus pusiLius, Micha.

Daucus pusillus, Michx. Fl. Bor.-Am. 1, p. 164; Torr. & Gray, Fl. N. Amer. 1,
p- 636.

Has. Rio Negro, North Patagonia; on sand-hills. Hawaii, Sand-
wich Islands; on the bullock plains of Mouna Kea, and in the district
of Waimea: probably introduced with cattle.

The specimens are all of the large and scabrous-hirsute form which
occurs in California (the var. scaber, Torr. & Gray, l.c.); the stout
stems a foot or two in height, the many-rayed and dense umber
scarcely exceeding the involucre. It belongs to the section Carota,
and is surely distinct from the following.

3. Daucus BRACHIATUS, Sieber.

Daucus brachiatus, Sieber, Pl. Exsic. N. Holl. no. 115; DC. Prodr. 4, p. 214; Benth.
Pl. Hugel. p. 56; Bunge, in Pl. Preiss. 1, p. 295; Hook. f. Fl. N. Zeal. p. 91.

D. montanus, Willd. in Schult. Syst. Veg. 6, p. 482.

D. australis, Pepp.; DC. 1. ¢.

D. Montevidensis & D. toriloides, DC. 1. ¢.

Scandix glochidiata, Labill. Pl. N. Holl. 1, p. 75, t. 102.

Has. Peru, around Obrajillo. Bay of Islands, New Zealand.
Sydney and Hunter's River, New South Wales.

With Bentham and Bunge, I cannot distinguish the Australian
from the South American plant. The latter extends north to Mexico.

112 PHANEROGAMIA.

27. TORILIS, Adans.

1. ToriLis Noposa, Geertn.

Has. Island of St. Helena. Doubtless introduced from Europe.

283. OREOMYRRHIS, Endl.

1. OREOMYRRHIS ANDICOLA, Endl.

Oreomyrrhis andicola, Endl.; Hook. f. Fl. Antare. p. 288, t. 101.
Jaldasia andicola, Lagasca, in DC. Mem. Umb. t. 2, & Prodr. 4, p. 229.
CU. cherophylloides, Lag. Ameen. Nat. p.99. (C. chwrophyllea, Lag. in DC. 1. ¢.)
Myrrhis andicola, H. B. K. Nov. Gen. & Spec. 5, p. 18, t. 419.
Azorella daucoides, D’Urv. in Mem. Soe. Linn. Par. 4, p. 613; DC. 1. c. p. 77.

Has. High Andes of Peru; at Casa Cancha and Culnai.

The specimen belongs to a pubescent form of the species, like that
of Mr. M’Lean gathered in the same region, and has the leaflets of
the involucel entire, as in Humboldt’s and Dr. Hooker's plants.
Similar specimens in other collections, however, have the leaflets of
the involucre incised or sparingly lobed, and therefore doubtless belong
to Caldasia cherophylloides of Lagasca, which cannot be specifically
distinct. The fruit in all is ovate-oblong or nearly ovate.

2. OREOMYRRHIS LASIOPETALA.
O. undique canescenti-pubescens ; involucellorum foliolis multifidis flores
superantibus; petalis extus pubescenti-hirsutis; fructibus oblongis
cinereo-tomentosis pedicello longioribus.

Caldusia lasiopetala, Lagasea, in DC. Prodr. 4, p. 229.

Has, Andes of Peru, at Bajios.

UMBELLIFERA. 713

This resembles the preceding species, and may possibly pass into
it: but it is more cinereous or canescently pubescent throughout; the
leaflets of the involucel are dissected, either pinnately parted or pal-
mately trisected, with the segments pinnatifid and incised, exceeding
the rays when in flower, although not equalling the mature, oblong,
cinereous-tomentose fruit. The latter is 24 or almost 3 lines long, and
commonly twice as long as the pedicels. The flowering stems or

peduncles are thick and fistulous at the base. Petals hirsute-pubescent
externally.

29. OSMORHIZA, Raf
1. OsmoruizA Curtensis, Hook. & Arn.

Osmorhiza Chilensis, Hook. & Arn. Bot. Beech. Voy. p. 26, & Bot. Mise. 3, p. 855;
Hook. f. Fl. Antare. p. 288.

O. Berterti, DO. Prodr. 4, p. 232; Gay, FI. Chil. 3, p. 143.

Cherophyllum Chilense, Poir. Dict. 5, p. 105.

Has. Orange Harbour, Fuegia. (Much resembling the 0. brevi-
stylis of North America, except that it has neither involucre nor
involucel.)

30. HERMAS, Linn.
1. Hermas vittosa, Thunb.

Has. Cape of Good Hope, in the vicinity of Cape Town.

There is an Umbelliferous plant of large size, apparently resembling
an Angelica or Pustinaca, growing on the mountains of Kauai, one of
the Sandwich Islands. Only a leaf (pinnate, with 7 large ovate |
leaflets) and the inflorescence of the preceding year, from which all,

the fruit has fallen, were collected. The carpophore is 4 lines long.
179

Orv ARALIACE &.

1. CUSSONIA, Thunb.
1. CUSSONIA THYRSIFLORA, Thunb.
Cussonia thyrsiflora, Thunb. in Act. Nov. Ups. 8, p. 212, t. 12; Jacq. Eclog. t. 61.

Has. Cape of Good Hope, in the vicinity of Cape Town.

2 STILBOCARPA, Hook.f

1. SrmtpocaRPa POLARIS, Hook. f.

Aralia polaris, Hombr. & Jacquinot, Bot. Phaner. Voy. Pol. Sud. t. 2, sine descr. ;
Hook. f. Fl. Antare. p. 95, & in Hook. Ic. Pl. t. 747.
Stillocarpa polaris, Hook. f. Fl. N. Zeal. p. 95, sub Aralia.

Has. Lord Auckland Islands; collected by Dr. Holmes. (Foliage
only.)

Dr. Hooker, who has admirably described this most striking Ant-
arctic plant, recognises it as “certainly of a different genus from
Aralia proper,” but hesitates fully to separate it before the order shall
have undergone the revision which it so greatly needs. He doubts
whether the petals are really valvate in xstivation,—a doubt which
may be readily solved, since in specimens with flower-buds, collected
and kindly furnished by himself, they are seen to be decidedly imbri-
cated. This character, added to the acetabuliform fruit, and the
Styles fewer than the petals, must surely exclude the plant from
Aralia, however that genus be limited.

ARALIACES. 715

8. ARA LIA, Zinn.

The genuine Aralie (from which Dimorphanthus, Miquel, does not
differ) are all more or less monceciously polygamous; the sterile
flowers having the pistil imperfect, and their abortive short styles
united. The styles of the fertile flowers are also united at the base
in A. racemosa, Linn., of which A. edulis, Sieb. & Zuce., the Dimor-
phanthus edulis of Miquel, is a strict congener; as D. elatus, Miquel,
seems to be of Aralia spinosa. The A. racemosa has traces of stipules,
or stipular appendages, at the base of the petiole, which are not found
in the other North American species.

§ 1. SCHABFFLERA, Forst.—Arborescens ; foliis manifeste stipulatis diyitatis ; umbel-

lulis composito-racemosis ; stylis 5-10.
1. ARALIA SCHAFFLERA, Spreng.

Aralia Schefflera, Spreng. Pug. Pl. 1, p. 28; DC. Prodr. 4, p. 258; A. Rich. Fl. N.
Zel. p. 283; Hook. f. Fl. Antare. p. 95, t. 22.
Schefilera digitata, Forst. Char. Gen. t. 23; Lam. Ill. Gen. t. 221.

Has. New Zealand; abundant at the Bay of Islands.

The styles appear to me to be perfectly distinct (although slightly
elevated on a broad stylopodium), and so, indeed, they are repre-
sented in Dr. Hooker’s excellent figure-—To Schefflera probably
belongs Aralia Quinduense, H. B. K., having stipules, which furnish
its leading character, but do not exist in Aralia proper, except as
mere vestiges in one or two species. As to the styles, they are
reduced to five in the following truly allied species.

2. Arata Vittensis, Sp. Nov. (Tab. 89.)

A. arborescens, inermis, glabra; foliis stipulatis longe petiolatis digitats ;
foliolis 7-9 longe petiolulatis cuneato-oblongis subito acuminatis sub-

716 PHANEROGAMIA.

serratis, lateralibus parvis; umbellis confertis pedunculatis in race-
mum compositum magnum paniculeformem digestis; stylis pyre-
nisque 5.

Has. Ovolau, Feejee Islands; at the elevation of 1,000 feet above
the sea.

A small tree, “20 feet high,” wnarmed, glabrous, or the nascent
parts cinereous-pubescent. Leaves alternate, digitately 7—-9-foliolate.
Petioles 4 inches or more in length; the base slightly connected with
a pair of pretty large, oblong-triangular, acute, clasping, membra-
naceous stipules. Leaflets cuneate-oblong, or cuneate-obovate, some-
times inclining to elliptical, abruptly acuminate, obscurely serrate,
membranaceous, 3 or 4 inches long and on partial petioles of about
an inch long, or the lateral ones much smaller. Inflorescence axillary,
compound-racemose ; the 8—10-flowered wmbels (on peduncles of 3 or 4
lines long) crowded on the whole length of the elongated primary
divisions, forming a large and extremely many-flowered spreading
panicle. Bracts small, deciduous. Ultimate pedicels a line and a
half long. Only the fruit is known; which is a /five-celled drupe,
strongly five-angled when dry, only a line and a half in diameter,
marked near the summit by the truncate edge of the calyx, crowned
with 5 short and wholly distinct divergent styles (their stigmas ter-
minal): pyrenze coriaceous. Seed suspended. Embryo small, next
the hilum.

Puate 89.—ARaALIA (Sco@rrtera) VITIENSIS : branch, in fruit, of
the natural size. Fig. 1. A fruit, enlarged. 2. A transverse, and 3,
a vertical section of the same.

4. PANAX, Linn.

1. Panax Fruticosum, Linn.

Has. Philippine Islands; near Baiios, Luzon. Also Samoan and
Feejee Islands: introduced, and planted around houses. (Foliage
only.)

ARALIACE S. "17

2. PANAX SAMBUCIFOLIUM, Sieber.
Panazx sambucifolium, Sieber, Pl. Exsic. N. Holl. 2, no. 256; DC. Prodr. 4, p. 255.

Has. New South Wales; probably gathered in the neighbourhood
of Sydney.

3. Panax SAMOENSE, Sp. Nov.

P. fruticosum, inerme, glabrum ; foliis simpliciter imparipinnatis ; folio-
lis 11-195 ovato-lanceolatis acuminatis basi rotundatis vel subcordatis
membranaceis integerrimis ; umbellis bis terve compositis corymbosis ;
stylis 2; fructibus plano-compressis.

Has. Samoan or Navigators’ Islands; on Savaii and Tutuila.

This is said to form a shrub, from 6 to 12 feet in height, and is
glabrous and unarmed. The alternate leaves are apparently all semply
pinnate; the stout petiole and rhachis a foot or two in length.
Leaflets 11 to 15, ovate-lanceolate or oblong-lanceolate, acuminate, entire,
rounded or obscurely cordate at the base, 5 to 8 inches long, on foot-
stalks of half an inch long, membranaceous, sparingly feather-veined.
Inflorescence terminal, ample, corymbosely umbellate-compound, or
decompound; the ultimate umbelets 10-20-flowered; the pedicels
filiform. Flowers dicecious; the male flowers destitute of styles.
Fertile flowers with 2 short styles. Fruit flat, apparently even when
ripe, 4 lines in diameter, orbicular, slightly emarginate at both ends,
perhaps nearly dry.

4, PANAX ARBOREUM, Jorst.

Panax arboreum, Forst. Prodr. Fl. Ins. Austr. p. 75; A. Cunn. Bot. N. Zeal. in
Aun. Nat. Hist. 2, p. 213; Hook. Lond. Jour. Bot. 2,p. 421, t. 11; Hook. f. Fl.

N. Zeal. p. 94.

Has. New Zealand; at the Bay of Islands.
180

718 PHANEROGAMIA.

5. PANAX SIMPLEX, Forst.

Panax simplex, Forst. 1.¢.; A. Rich. Fl. N. Zel. p. 280, t. 31; Hook. f. l.c. & FI.
Antare. p. 19, t. 12.

Has. Lord Auckland Islands.

Some of the leaves are strictly opposite, especially the last of each
years growth.—This and the foregoing species are perhaps to be
associated with the following, which, for the present, I have referred
to Hedera.

5. HEDERA, Tourn., Linn.

To Hedera rather than to Aralia, as the genera are at present cha-
racterized, I should refer the following, along with some other species
with more or less united short styles, which surely are not real con-
geners of the original, North American Aralie. The Hawaiian
species, varying as they do from di-trigynous to pentagynous, unless
excluded, would wholly confound the already arbitrary limits between
Aralia and Panazx.

* Novo-Zelandice (foliis raro oppositis).

1. HEDERA CRASSIFOLIA.

Aralia crassifolia, Banks & Soland.; A. Gunn. Bot. N. Zeal. in Ann. Nat. Hist. 2,
p. 214; Hook. Ic. Pl. t. 588, 584; Hook. f. Fl. N. Zeal. p- 96.
Xylophylla longifolia, Banks & Soland. Mss. & Ic.

Has. Bay of Islands, New Zealand.

A small tree, with extremely polymorphous, simple or trifoliolate,

rigid foliage. The united styles are well represented in Sir William
Hooker’s figure.

ARALIACES. 719

2. HrEpERA LESSont.

Panax? Lessoni, DC. Prodr. 4, p. 253.
Cussonia Tessonii, A. Rich. Fl. N. Zel. p. 285, t. 32; A. Cunn. Bot. N. Zeal. 1. e.
Aralia Lessoni, Hook. f. Fl. N. Zeal. p. 96.

Has. Tippona, New Zealand.

The short styles are united nearly or quite to the summit; the five
stigmas at length radiate-spreading.

* x Sandwicenses (foliis plerisque oppositis).
3. HeperA Gaupicuaupr. (Tab. 90.)

H. arborescens; foluis plerisque oppositis quinquefoliolatis passim tri-
Joliolatis longe petiolulatis ovalibus sew ellipticis cartilagineo-serratis
serrulatisve ; umbellulis in paniculam thyrsiformem confertis ; stigma-
tibus pyrenisque 3-5,

Aralia trigyna, Gaud. Bot. Freyc. Voy. p. 474, t. 98.
Panax? Gaudichaudi, DC. Prodr. 4, p. 253; Hook. & Arn. Bot. Beech. Voy. p. 84,

Var. @. foliolis sepius 3 subovatis vix denticulatis seu integerrims,
Panax? ovatum, Hook. & Arn. Bot. Beech. Voy. p. 84.

Has. Sandwich Islands: Hawaii; in forests, at the base of Mouna
Loa and Mouna Kea: Oahu; on the Kaala Mountains. (First col-
lected by Menzies.) Var. 3. Oahu, and the western part of Maui.

An arborescent or shrubby species, glabrous, unarmed, branched.
Leaves opposite in all the specimens, exstipulate, digitate. Leaflets 5,
or occasionally 3, oval or elliptical, obtuse or rounded at both ends, 2
to 32 lines long, chartaceous or coriaceous, not thickened, rather
sparsely serrate or serrulate with incurved cartilagineous teeth, thickly
feather-veined, reticulated, the lateral ones smaller, all on partial

720 PHANEROGAMIA.

footstalks of 6 to 15 lines long. Common petiole 2 or 3 inches long.
Peduncle terminal, short; bearing an inflorescence shorter than the
leaves, consisting of a large number of small and short-peduncled,
several-flowered umbels, crowded in a thyrsoid panicle. Bracts and
bractlets caducous. Pedicels about a line and a half long, dilated
into a 3—5-angled or toothed small calyculus at the articulation with
the flower. Flower-buds 2 lines long, glabrous: the flowers poly-
gamous, or mostly hermaphrodite. Calyx with an obpyramidal
three-angled or occasionally 4—5-angled tube; the limb truncate, very
short, repand. Petals 5, ovate-triangular, valvate in estivation,
expanding in anthesis, early deciduous. Stamens 5, not longer than
the petals: anthers two-celled, incumbent. Ovary three-celled or
four-celled by abortion (the cells more or less unsymmetrical), rarely
two-celled, more commonly five-celled, with a solitary anatropous
ovule suspended from the summit of each cell. Style in the sterile
flowers abortive, in the hermaphrodite flowers none, besides the short
and conical stylopodium, bearing 3, 4, or often 5 (sometimes 2?)
oblong, at first connivent, in fruit divergent, introrse stigmas. Drupe
as large as a small pea, fleshy, 3—5-angled in the dried state, contain-
ing, 3, 4, or 5 pyrene, of a coriaceous texture. Albumen corneous-
fleshy. Hmbryo minute, next the hilum, cordate.

The var. @., which is undoubtedly the Punax? ovatum of Hooker
and Arnott (who had seen only foliage), is evidently a state of the
species with only three leaflets, which are usually subovate and barely
denticulate or entire.

Piate 90.—HEDERA GAUDICHAUDI: var. a., in flower, and var. @.,
in fruit, of the natural size. Fig. 1. An unexpanded flower, with its
calyculus and pedicel. 2. Vertical section of an expanded flower.
3. Flower (the petals and stamens fallen), with 3 stigmas. 4.
Transverse section of the same. 5. Transverse section of a nearly
symmetrically three-celled ovary. 6. A drupe. 7. Transverse sec-
tion of the same. 8. A pyrena. 9. Vertical section of the same,
showing the minute embryo. 10. Embryo, detached—The details

variously magnified.

4, HEDERA PLATYPHYLLA. (Tab. 91.)

f. fruticosa ; foliis plerisque oppositis trifoliolatis longissime petiolulatis

ARALIACE SR. 721

transversim rhomboideo-dilatatis latioribus quam longis subito acumi-
natis miegerrimis, margine incrassuto; umbellulis laxe cymoso-panicu-
latis; stigmatibus cum ovarii loculis 22-5.

Panax? platyphyllum, Hook. & Arn. Bot. Beech. Voy. p. 84.

Has. Oahu; on the mountains behind Honolulu, at the elevation
of 2,000 feet. (Also gathered by Gaudichaud, in the Voyage of the
Bonite.)

A shrub, glabrous throughout, unarmed. Leaves opposite in all the
specimens, on slender petioles of 3 or 4 inches in length, exstipulate,
digitately ¢trifoliolate. Leaflets transversely dilated, somewhat rhom-
boidal, from an inch to 2 inches long, but 14 to nearly 3 inches wide,
abruptly acwminate or apiculate, rarely emarginate or retuse, entire,
the margin thickened, chartaceous or slightly coriaceous in texture,
somewhat lucid, of the same hue both sides, copiously feather-veined ;
the veins slender, parallel and straight, connected by minute reticu-
lated veinlets: the partial footstalks very long, about 2 inches in
length. Umbels few-flowered, in a simple or trichotomous and loose
cymose panicle, which is shorter than the leaves. Pedicels one or 2
lines long, bearing a small calyculus at the articulation with the
flower. Flowers nearly 3 lines long; those of the specimens herma-
phrodite. Calyx-tube obconical; the short limb repand-truncate,
denticulate. Petals 5, triangular-ovate, valvate in sestivation, a line
and a half long, expanding in anthesis, early deciduous. Stamens 6,
as in the preceding species. Ovary jive-celled, or sometimes four-
celled, in our specimens. Styles none. Stigmas sessile on a short
stylopodium, 2, according to Hooker & Arnott, but in our specimens
always 4 or 5. Ovule with the rhaphe internal, as in the family.
Fruit not seen.

*

Prats 91.—Hepera PLATYPHYLLA: a branch, in flower, of the
natural size. Fig. 1. An unexpanded flower, with its small calyculus.
2. A vertical section of an opening flower. 3, 4. Stamens. 5. An
ovule. 6. Transverse section of a four-celled ovary. 7. Similar
transverse section of a five-celled ovary.—The details variously mag-

nified.
181

722 PHANEROGAMIA.

6. PARATROPIA, Blume, DC.
1. PARATROPIA SAMOENSIS, Sp. Novy.

P. arborescens? foliis longe petiolatis digitatis; foliolis 5-T membra-
naceis oblongis seu cuneato-oblongis utringue acuminatis argute mucro-
nato-serrulatis glabris, lateralibus parvis; panicula composita ; umbel-
lulis racemosis paucifloris; staminibus 9d.

Has. Upolu, one of the Samoan or Navigators’ Islands.

The imperfect flowering specimen belongs apparently to an unarmed
and arborescent plant. Leaves alternate, digitately 5—T-foliolate. Sti-
pules none. tioles elongated, 6 to 9 inches long, somewhat pubes-
cent when young. Leaflets oblong, oblong-lanceolate, or cuneate-
oblong, acuminate at both ends, on short partial footstalks, sharply and
mucronately serrulate with appressed teeth, very thin and membra-
naceous, glabrous, except a few hairs on the midrib and veins when
young, the middle ones 5 or 6 inches long, the lateral ones succes-
sively smaller. fanicle compound, termimal, shorter than the leaves,
somewhat pubescent. Flowers minute; the flower-buds scarcely half
a line in diameter, 5 to 8 together in small wimbels, which are racemose
on the branches of the inflorescence. Calyx truncate. Petals 5,
apparently expanding in anthesis. Stamens 5. Stigmas 4 or 5, very
short, scarcely emerging from the centre of the convex or conical
disk. Fruit unknown.

2. PARATROPIA? MULTIJUGA, Sp. Nov.

P. arborea? glabra; foliis impari-pinnatis multijugis, rhachi nodosa
bipedal et ultra; foliolis elliptico-oblongis integerrimis breviter petio-
lulatis bast subcordatis sinu clauso; racemo composito longissimo ;

umbeltulis paucifloris ; staminibus 5.

Has. Feejee Islands; at Sandalwood-Bay, Vanua levu.

ARALIACEZ. 723

The specimen consists of a detached leaf and a compound raceme
(with the flower-buds not full-grown) of an apparently arborescent
plant, which is doubtless a congener of Paratropia nodosa, DC., but is
at once distinguished by the entire leaflets. The leaves are glabrous,
very large, the terete nodose rhachis 2 feet long or more, and the
petiole 6 or 8 inches long, pinnate. Leaflets elliptical-oblong, 9 pairs
with an edd one, slightly coriaceous, about 6 inches long, green both
sides, entire, short-petiolulate, the base slightly cordate, with the sinus
closed. Inflorescence as long as the leaf, compound-racemose, the
primary divisions bracteate, bearing a large number of small and few-
flowered umbels on short and bracteate peduncles. Flowers minute,
pentandrous. Stigmas sessile.

There is in the collection some indeterminable digitate foliage of
what is probably an undescribed Paratropia, from Ovolau, Feejee
Islands, and of another from Luzon.

7. REYNOLDSIA, Nov. Gen.

Flores polygami. Calyx basi nudus; tubo cum ovario connato; limbo
brevissimo integerrimo vel subrepando. FPetala 8-10, epigyna, val-
vata, apice in calyptre formam coalita, sub anthesi dejecta. Stamina
8-10, cum petalis inserta, iisdem alterna: filamenta brevia: anthere
lineares. Ovarium inferum, 8-18-loculare: stylus nullus vel sub-
nullus: stigma indivisum, 8-18-radiatum. Ovula in loculis solitaria,
suspensa, anatropa. Drupa baccata, globosa, 8-18-pyrena ; pyrenis
cartilagineis. Embryo in apice albuminis dense carnosi minutus ;
radicula supera cylindrica —Arbores insularum Pacifici, glabre,
inermes, exstipulate; foliis simpliciter pinnatis scepissime triyugis cum
impari, foliolis subdentatis ; umbellis racemisve compositis paniculatis

laxifloris.

Whether certain species referred by Blume and DeCandolle to
Sciodaphyllum, on account of their calyptraeform corolla, really belong
to the present genus, I have not the means to determine. But the
perfectly consolidated stigmas, the larger number of pyrenx, the lax
and open inflorescence, and the pinnate leaves, taken together forbid

724 PHANEROGAMIA.

the combination of the two species here described with the American
Sciodaphylla. From Brassaia, Endl., they are excluded by the want
of an involucel at the base of the calyx, and by the wholly inferior
ovary, as well as by the pinnate and exstipulate leaves——I dedicate
the genus to J. N. Reynolds, Esq., who merits this commemoration
for the unflagging zeal with which he urged upon our Government
the project of the South Sea Exploring Expedition, and also for having
made, under trying circumstances, an interesting collection of dried
plants in Southern Chili, many years ago.

1. Reynotpsta SANDWIcENSIS, Sp. Nov. (Tab. 92.)

fi. foliolis subcordatis; corolla clausa oblonga; stigmate 8-10-radiato ;
drupa 8—10-pyrena.

Has. Sandwich Islands: in a ravine near Waianae, Oahu.

“Tree 30 feet high,” glabrous, or the nascent parts slightly pubes-
cent. Stipules none. Leaves alternate, simply pinnate, mostly of 7
leaflets. Petiole slender, angled, dilated at the insertion. Leaflets
subcordate, broadly ovate, or somewhat deltoid, either obtuse, acutish,
or somewhat acuminate, sparingly toothed especially towards the
base, thin, pinnately veined, 14 to 2 inches long; the partial foot-
stalks 6 or 9 lines in length. Inflorescence terminal or lateral, form-
ing an open compound or decompound panicle; the pedicels (3 to 5
lines long) racemose along its divisions, as well as umbellate at their
extremity. Bracts minute, caducous. Sterile and fertile flowers
intermixed in the same inflorescence, and apparently similar, except
that the stigma is imperfect or nearly obliterated in the sterile
flowers. Calyx naked (not involucellate at the base); the tube tur-
binate or obconical, wholly adnate to the ovary; the limb very short,
truncate and entire, or slightly undulate. Corolla in the bud oblong,
or cylindraceous, epigynous, 3 lines in length, somewhat angular,
composed of usually 10 linear petals, which are valvate in estivation,
and with their inflexed tips more strongly coalescent, separating
from the insertion in anthesis, and falling away together as a calyptra.
Stamens as many as the petals and alternate with them, epigynous,
deciduous: filaments subulate, short: anthers linear, longer than the

"ARALIACEAR - 725

filaments, fixed near the middle, incumbent, two-celled; the cells
opening longitudinally. Style none. Stylopodium in the sterile
flowers small or obsolete; in the fertile conical, and terminated by
the undivided, depressed, somewhat 8-10-rayed stigma. Ovary 8-10-
celled, wholly inferior, in the sterile flowers more or less inane; the
fertile with a solitary anatropous ovule suspended from the summit
of each cell; the rhaphe ventral. Fruit a globose drupe, of 8 or 10
pyrene in a pulpy sarcocarp, 3 or 4 lines in diameter, tipped with the
conical stylopodium and persistent stigma. Pyrene compressed,
smooth, of a firm or cartilagineous texture, filled by the seed. Testa
very thin. Hmbryo minute at the apex of the dense fleshy albumen:
radicle superior, cylindrical, twice the length of the oval thickened
cotyledons.—The leaves in the specimen collected are all young and
evidently developed at the same time as the flowers: they probably
enlarge and acquire a firmer texture.

PLATE 92.—REYNOLDSIA SANDWICENSIS: a branch, of the natural
size. Fig. 1. An unexpanded flower. 2. Same, with the corolla
detached, taking with it the stamens. 3. Vertical section of an
unexpanded flower. 4,5. Stamens. 6. Ovule. 7. Transverse sec-
tion of a ten-celled ovary. 8. A mature drupe. 9. Transverse sec-
tion of the same. 10. Vertical section of a drupe, showing the
embryo in one seed. 11. Embryo, detached.—The details variously
magnified.

2. REYNOLDSIA PLEIOSPERMA, Sp. Nov. (Tab. 93.)

R. foliolis ovatis seu lanceolato-oblongis; corolla clausa conica vel
ovoidea; stigmate 15-18-radiato; drupa 15-18-pyrena.

Has. Forests in the interior of Savaii, one of the Samoan or Navi-
gators’ Islands.

Our materials consist of two imperfect specimens, here associated ;
but they may hereafter prove to be distinct: one, with broadly ovate
leaflets and portions of a compound umbel with flower-buds, which are
apparently all sterile; the other, with oblong or lanceolate-oblong

leaflets, and with mature fruit only. The size of the tree or shrub
182

726 PHANEROGAMIA.

not ascertained. Leaflets 7, or sometimes 5, glabrous, as is the whole
plant, 2 to 4 or 5 inches long, rounded or oblique at the base, taper-
ing to an acute or acuminate apex, usually beset with 3 or 4 salient
and callous or glandular-tipped teeth on each margin; the partial
footstalks 2 to 5 lines long. Stipules none. Flowers apparently sub-
dicecious, in compound and panicled umbels, which in the fertile
plant, at least, are racemose-proliferous. Pedicels of the sterile
flowers 8 or 9 lines long; of the fruit 3 to 5 lines long. Calyx
naked at the base, with a short and truncate, obscurely repand
- border. Corolla in the bud conical or ovoid, 2 lines long, falling off
without expanding; the 8 or 10 valvate petals cohering with each
other, especially at the inflexed tips. Stamens 8 to 10, as in the fore-
going species, except that the anthers are only linear-oblong. Cells
of the sterile ovary numerous, empty: sterile stigma obscurely many-
rayed, crowning the very short stylopodium. Fertile flowers not seen.
Fruit a wholly inferior, depressed-globose, baccate drupe, of the size
of a pea, abruptly pointed by a very short style or narrow stylo-
podium, which bears a depressed, and 15-18-radiate stigma ; the cartila-
gineous and smooth pyrene 15 to 18, or sometimes perhaps even 20
in number. Seed suspended. Embryo minute, at the apex of the
hard fleshy albumen: radicle superior.

PLATE 93.—REYNOLDSIA PLEIOSPERMA: foliage, portion of inflores-
cence, unexpanded sterile flowers, and also fruit from a different speci-
men, of the natural size. Fig. 1. A flower-bud, the corolla detached
as a calyptra, taking with it the stamens. 2. A separate petal,
inside view. 3,4. Stamens. 5. Vertical section of a sterile pistil.
6. A drupe. 7. Transverse section of the same. 8. A drupe, verti-
cally divided.—The details variously magnified.

8. GASTONTA, Commers.
1. GastonrA? Oanvensis, Sp. Nov.
G.? folus pinnato-5-9-foliolatis; foliolis carnosis oblongis seu ovalibus

obtusis integerrimis ; umbellis compositis cymosis ; petalis 53 stamt-
nibus 10; ovario sterili quadri-quinqueloculari.

ARALIACES 7927

Has. Oahu, Sandwich Islands; on the mountains behind Honolulu,
at the elevation of 2,000 feet.

The collection contains only a single specimen of this evidently
arborescent plant, with staminate but sterile flowers only. Until
additional materials are obtained, it may be provisionally appended
to Gastonia, merely because it accords with the Mauritian species in
having twice as many stamens as petals, and also pinnate leayes.—
Branches stout, marked with very large leaf-scars. Leaves pinnately
5—9-foliolate, glabrous, as is the whole plant. Stipules none. Leaflets .
fleshy and thick, 2 to 34 inches long, oblong or oval, inclining to obo-
vate, obtuse, closely but inconspicuously feather-veined, mostly acute
at the base, on a short partial footstalk. Umbels compound, cymose,
crowded at the summit of a compressed-angled common peduncle.
Bracts, if any, caducous. Calyx-tube (of the sterile flowers) clavate,
4—5-angled; the limb very short, truncate, repand. Petals 5, trian-
gular-lanceolate, valvate in eestivation, expanding in anthesis, but
caducous. Stamens 10: filaments subulate, short: anthers two-
celled, oblong, opening longitudinally. « Sterile ovary 4—5-celled, inane,
crowned with a very short stylopodium and an abortive entire stigma.
Fertile flowers and fruit not seen.

The collection of M. Gaudichaud, made at Oahu in the Voyage of
the Bonite, contains some pinnately compound foliage of another
Araliaceous tree, probably a congener of the above, but hardly of the
same species.

9. TETRAPLASANDRA, Nov. Gen.

Flores polygami? Calyx tubo hemispheerico cum ovarto connato ; limbo
brevissimo truncato vie denticulato. Petala 7-8, epigyna, valvata,
leviter calyptratim cohcerentia, caduca. Stamina cum petalis unserta,
jisdem numero quadrupla, nempe 28 v. 32, uniseriata: Jilamenta
brevia: antherce oblonge, subsagittate. Ovariwm T-10-loculare:
stylus nullus: stigma indivisum, obsolete T-10-radiatum, stylopodio
brevi conico impositum. Ovula in loculis solitaria, suspensa, ana-
tropa. Drupa baccata, 8-10-pyrena; pyrenis coriacers. (Embryo

728 PHANEROGAMIA.

haud visus.)—Arbor procera, inermis; foliis exstipulatis pinnatis
\-7-foliolatis subtus incanis ; uwmbellis decompositis paniculatis.

1, TurrarLasanpra Hawatensis, Sp. Nov. (Tab. 94.)
Has. Hawaii, Sandwich Islands; in the district of Puna.

“A fine tree, 30 or 40 feet high, with the trunk a foot in diameter,”
‘unarmed. Branchlets, with the leaf-stalks, inflorescence, and the
exterior of the flowers canescent with a close and soft tomentum,
glabrate with age. Leaves alternate, destitute of stipules, large (the
stout petiole with the rhachis a foot or two in length), pinnately 5-7-
foliolate. Leaflets oblong or elliptical, from 4 to 7 inches long, and 2
inches or more in width, obtuse at both ends or acute at the base,
entire, coriaceous, glabrous above, densely canescent-tomentose under-
neath, pinnately veined from a stout and salient midrib, petiolulate.
Peduncle terminal, stout, bearing an ample and open panicle of com-
pound or decompound umbels. Bracts caducous. Peduncles and
pedicels articulated; the latter about an inch long. Flowers probably
polygamous, large for this order; the buds nearly half an inch long
and 4 or 5 lines in diameter. Calyx naked at the base (not involu-
cellate); the tube hemispherical, adnate; the truncate limb very short,
entire, or obscurely repand, persistent. Corolla epigynous, ovoid-
hemispherical in the bud, 3 or 4 lines long, of 7 or 8 triangular-
oblong valvate petals, lightly cohering, especially by their inflexed tips,
sometimes apparently separating, but usually caducous in the form of
a calyptra. Stamens 28 or 32, four times as many as the petals, and
inserted with them, in a single series, on the margin of the epigynous
disk : filaments short, filiform, about the length of the anthers, which
are sagittate-oblong, fixed by the middle, incumbent, two-celled; the
cells opening longitudinally. Ovary 7-10-celled; the apex crowned
with a short and conical stylopodium, bearing a depressed, undivided,
obscurely T-10-rayed stigma. Ovules solitary and suspended from the
summit of each cell, anatropous. Fruit a globular baccate drupe, 5
lines in diameter, containing 7 to 10 compressed and coriaceous
pyrene, Seed suspended. Albumen fleshy. Embryo not seen.

This remarkable plant differs from all other known Araliacee by

ARALIACESA. 729

having four times as many stamens as petals (whence the name, com-
pounded of ‘Tetpiinhdstos, four-fold, and &épa, used for stamens), excepting
the succeeding genus, in which they are indefinitely numerous.

PLATE 94.—TETRAPLASANDRA HAWAIENSIS: portion of inflorescence
and foliage, of the natural size. Fig. 1. A flower, with the cohering
petals detached. 2. The same, with the stamens also fallen away.
3. A petal, inside view. 4, 5. Stamens. 6. Vertical section of an
ovary. 7. Transverse section of an ovary. 8. A drupe, of the
natural size. 9. Vertical section of a drupe. 10. A pyrena, trans-
versely divided.—The details (except Fig. 8) variously magnified.

10. PLERANDBRA, Nov. Gen.

Flores polygamo-monoici vel dioict? Masc.— Calyx tubo turbinato cum
ovario connato; limbo brevissimo post anthesin repando-undulato.
Petala 42 epigyna, oblonga, cestivatione valvata, mou decidua. Sta-
mina indefinita, epigyna, pluriserialia: filamenta filiformia: antheree
oblonge. Ovariwm 14-15-loculare: ovula in loculits solitaria, parva,
suspensa, seepius abortiva vel nulla. Stigma truncatum, obsolete
multiradiatum, stylopodio conico impositum. Foem. tgnoti.—Arbor
20-pedalis, macrophylla; foliis digitatis 9-foliolatis; umbels com-
positis.

1. PuerAnpra Prcxerinet, Sp. Nov. (Tab. 99.)
Has. Ovolau, Feejee Islands; at the elevation of 1,500 feet.

A small tree, 15 or 20 feet high, according to Dr. Pickering’s notes,
“with very large leaves as well as compound umbels,” unarmed?
glabrous. Leaves digitate, the stout and fistulous petioles 2 feet or
more in length. Leaflets 9, oblong or oblong-obovate, entire, acute
at the base, membranaceous, pinnately veined, a foot and a half long
in the specimen, probably often much larger, the partial footstalks 2
or 8 inches long. Rays of the compound umbel large and stout:
the ultimate umbels many-rayed, the rays an inch and a half long.

183

730 PHANEROGAMIA.

Involucel apparently wanting. Male flowers large: the tube of the
calyx (seen only after anthesis) turbinate, adnate to the ovary, 3 or 4
lines long; its limb very short, truncate, undivided, but repandly
undulate, sometimes appearing to be 4—-5-lobed. Petals apparently 4
(perhaps 5), epigynous, 6 or 7 lines long, thickish, oblong, valvate in
estwation, perhaps expanding in anthesis, but early deciduous. Sta-
mens indefinite, as many as 60 or 70, distinct, inserted with the petals,
occupying more than one series; the filaments filiform and apparently
as long as the corolla, deciduous: anthers oblong, somewhat sagittate,
two-celled, fixed near the middle, incumbent. Ovary fleshy, turbi-
nate, fourteen-celled, or sometimes fifteen-celled, with complete but
thin dissepiments. Ovules solitary, suspended, anatropous, small, and
apparently sterile, or often wholly wanting. Stylopodium conical, large,
a line and a half long, tipped with a truncate and entire, but obscurely
14-15-radiate stigma. Female flowers and fruit (the latter noted by
Dr. Pickering as large) not collected.

It is much to be regretted that fuller materials of this most inte-
resting plant were not preserved. In the collection there is only a
single compound leaf, and one ray of the compound umbel, bearing an
umbellet of nearly 30 flowers, which are past anthesis, the filaments
and a petal or two only remaining on some of them. The enlarging
and fleshy ovary has the appearance of being fructified, but seldom
contains even a sterile ovule. The scanty materials, therefore, barely
suffice to show that we have in this plant a truly polyandrous Ara-
hacea! To this character the name chosen for the genus alludes.

Puats 95.—PLERANDRA PICKERINGI: a leaf and an ultimate ray of
the compound umbel, of the natural size. Fig. 1. Vertical section of
a flower. 2. A petal. 38,4. Anthers. 5. Flower, from which the
petals and stamens have fallen. 6. Transverse section of the ovary.
—The details enlarged.

ll. BOTRYODENDRUM, nd.

BorryoDENDRUM, Endl. Prodr. Fl. Ins. Norf. p- 62, & Gen. Pl. p. 795; Hook. f.
Fl. N. Zeal. p. 97.

ARALIACES. 731

1. BorryopEnpRUM TAITENSE, Guill. (Tab. 96.)

B. foliis lanceolatis utrinque acutis chartaceis ad apicem ramorum con-
fertis; thyrso capitulorum erecto.

Botryodendrum Taitense, Guill. Zeph. Tait. p. 55, sine char.
B. cerberoides & B. lancifolium, Rich, in Herb.

Has. Tahiti, Society Islands; in mountain forests.

The specimens consist, 1st, of a branch, with narrow, almost linear-
lanceolate leaves, and a portion of inflorescence, with fruit, ticketed
B. lancifolium, N. Sp., Tahiti, by Mr. Rich; and 2d, of one with
larger and broader, but otherwise similar leaves, with some detached
fruitful inflorescence, and male and female flowers, imperfectly pre-
served, ticketed B. cerberoides, N. Sp., Samoa, by Mr. Rich. But,
on referring to Dr. Pickering’s memoranda, I find evidence that
both specimens were gathered at Tahiti; and they appear to be mere
varieties of one and the same species. I regret that the state of the
materials does not allow me properly to illustrate the genus (of
which no figure has yet been published), nor to compare this species
with Endlicher’s B. angustifoliwm of Norfolk Island, to which it
appears nearly allied. Our plant is stated to be a shrub, from 8 to
12 feet high. The leaves are chartaceous in texture, glabrous, as is
the whole plant, lanceolate, acute at both ends, entire, alternate,
crowded at the summit of the branches; in the narrow-leaved form
only 7 or 8 inches long and three-fourths of an inch wide, on petioles
of 2 inches in length, the apex tapering to a very acute point; in
the other, they are from 8 inches to a foot long (in the notes of Dr.
Pickering said to be 18 inches long), 2 or 22 inches wide, inclining
to elongated-elliptical, more or less acute or pointed. The flowers,
as far as can be ascertained, accord with Endlicher’s character, except
that the short lobes of the calyx in the female flowers are valvate in
sestivation, and the divisions of the male perianth and the stamens
apparently only 3. Stigmas 6 ‘to 8, linear, recurved. Drupes glo-
bose, baccate, a third of an inch in diameter, capitate-clustered on
the short branches of the apparently erect and thyrsoid inflorescence.
Pyrenze 8, compressed, coriaceous or cartilagineous. Albumen between

732 PHANEROGAMIA.,

fleshy and horny. Embryo minute, next the hilum: radicle cylin-
drical, longer than the cotyledons, superior.

According to Dr. Hooker, the petiole is articulated with the lamina
in the New Zealand species: but there is no trace of such articu-
lation in this and the following species. _

PLATE 96.—BoTRYODENDRUM T'AITENSE: fruiting branches of the
narrow and the broader-leaved forms, of the natural size. Fig. 1. A
staminate flower. 2. A pistillate flower. 3. Vertical section of the
same. 4. Transverse section of a drupe. 5, Vertical section of a
drupe, and of one of the contained seeds. 6. Embryo.—The details
variously enlarged.

2. BoTRYODENDRUM MACROPHYLLUM, Jtich, in Herb. (Plate 97.)

B, foliis obovato-lanceolatis basi attenuatis membranaceis ad apicem
ramorum confertis.

Has. Samoan or Navigators’ Islands: in forests. Also, according
to Dr. Pickering’s notes, at Tongatabu, Friendly Islands, where it
was likewise gathered by Mr. Matthews.

This is said to be “a simple shrub, from 10 to 25 feet high.”
Whether it is really distinct from the preceding species cannot be
satisfactorily determined from the present materials, which consist
of foliage, some badly preserved fertile flowers, a detached portion of
male inflorescence (which perhaps belongs to B. Tuitense), and
mature fruit. The Jeaves are much larger than those of the former
species, being from 2 to 23 feet in length, and 3 to 5, or towards the
apex 6 inches wide, obovate-lanceolate or oblanceolate in form, rather
abruptly acuminate, tapering downwards to the acute base, membra-
naceous when dry, perhaps rather succulent when fresh, on petioles of
3 to 5 inches in length, crowded at the summit of the stem or
branches. The capitula of fertile flowers are sessile along an appa-
rently simple rhachis. Teeth or lobes of the calyx of the fertile
flowers 6 or 8, irregular, or becoming so after opening, valvate in
gestivation, shorter than the 8 recurved stigmas. Ovary eight-celled.

ARALIACEA. 733

Ovule solitary in each cell, suspended from its summit, anatropous;
the rhaphe ventral. Fruit as in B. Twitense. Pyrenz 8, lunate.
Embryo not seen.

PLATE 97.—BoTRYODENDRUM MACROPHYLLUM: foliage with flowers
and fruit, of the natural size. Fig. 1. Vertical section of a fertile
flower-bud. 2. Vertical section of a fertile flower. 3. Summit of a
fertile flower after anthesis, showing the calyx (the stigmas removed).
4. The same, with the stigmas. 5. Transverse, and 6, longitudinal

section of a drupe. 7. A detached pyrena.—The details variously
enlarged.

ARALIACEA ?—On high land, near Waianae, Oahu, some branches
with nascent foliage and young flower-buds were gathered from “an
Araliaceous tree, twenty feet high,” of which, for want of sufficient
materials the genus cannot be made out. The following data which
the specimens furnish will serve to indicate the tree to future col-
lectors in the Sandwich Islands.

Branchlets slender (for an Araliacea), with a wrinkled and warty
bark, nodose with the scars of former petioles; the bark not pungent
to the taste. Young foliage, &., glabrous. Stipules none. Leaves
alternate, on slender petioles which are slightly dilated at the base.
Leaves either trifoliolate or pinnately quinquefoliolate; the deve-
loping leaflets ovate or obovate, petiolulate, entire or undulate,
mostly less than an inch long. Male flowers somewhat umbellate or
fascicled on the divisions of a small paniculate inflorescence, which
arises from the base of the leafy shoot of the season: the buds half a
line in diameter, short-pedicelled. Calyx four-lobed, or sometimes
five-lobed. Corolla globular in the bud, the 4 or 5 petals moderately
imbricated in sestivation. Stamens 4 or 5, alternate with the petals:
filaments short: anthers oblong, two-celled, incumbent. Ovary abor-
tive, so that the calyx appears as if inferior: a minute stylopodium
occupies the place of the style or stigmas. Fertile flowers entirely

unknown.
184

Orv CORNACE SZ.

Il GRISELINIA, Forst.

1. GRISELINIA LUCIDA, Forst.

Griselinia lucida, Forst. Prodr. Fl. Ins. Austr. p. 75; A. Cunn. Bot. N. Zeal. in
Aun. Nat. Hist. 3, p. 261; Hook. f. Fl. N. Zeal. p. 98.

G. littoralis, Raoul, Fl. N. Zel. p. 22, t. 19 (forma latifolia).

Scopolia lucida, Forst. Char. Gen. t. 70.

Pukateria littoralis, Raoul, in Ann. Sci. Nat. ser. 8, 2, p. 120.

Has. Bay of Islands, New Zealand.

2. COROKIA, A. Cunn.

1. CoROKIA BUDDLEOIDES, A. Cunn.

Corokia buddleoides, A. Cunn. Bot. N. Zeal. in Ann. Nat. Hist, 3, p. 249; Hook.
Ic. Pl. t. 424; Hook. f. Fl. N. Zeal. p. 98.

Has. New Zealand; at the Bay of Islands.

Orv. BRUNIACEA.

1. BRUNIA, Linn.
1. Brunta NoprFtora, Linn.

Has. Cape of Good Hope, in the vicinity of Cape Town.

Orv. LORANTHACE A.

1 LORANTHUS, Linn.

* Brasilienses.
1. Lorantuus (Psirracantuus) picHRoos, Mart.

Loranthus dichroos, Mart. in Schult. Syst. Veg. 7, p. 122 j DC. Prodr. 4, p. 122.
Psittacanthus dichroos, Mart. in Flora (Regensb. Bot. Zeit.) 1830, 1, p. 108; Don,
Syst. Gard. & Bot. 3, p. 415.

Has. Brazil; in the neighbourhood of Rio Janeiro, and Organ
Mountains.

* Although widely separated in the sequence of orders which it is still convenient to
follow, the Olacacew, Loranthacee, and Santalacec, as is now well known, are closely
related, and belong to one class or alliance.

736 PHANEROGAMIA.

2. LoRANTHUS (PsITTACANTHUS) JAcQUINI, DC.

Loranthus Jacquini, DC. Prodr. 4, p. 308.

L. Americanus, Jacq. Stirp. Amer. p. 97, t. 67; Cham. & Schlecht. in Linnzea, 1,
p. 210.

Psittacanthus Jacquini, Don, 1. ¢.

Has. Organ Mountains, and in the neighbourhood of Rio Janeiro,
Brazil.

3. LORANTHUS (STRUTHANTHUS) MARGINATUS, Lam.

Loranthus marginatus, Lam. Dict. 8, p. 596; Schult. Syst. Veg. 7, p. 187; DC.
Prodr. 4, p. 289.
L. (Struthanthus) citricola, Mart. 1. ¢.

Has. Organ Mountains, probably not far from Rio Janeiro, Brazil.

* * Chilenses.

4. LORANTHUS (STRUTHANTHUS) HETEROPHYLLUS, Ruiz & Pav.

Loranthus heterophyllus, Ruiz & Pay. Fl. Per. & Chil. 8, p. 48, t. 273; DC. Prodr.
4, p. 8312; Hook. & Arn. in Bot. Misc. 3, p. 359; Gay, Fl. Chil. 3, p. 157.

L. buxifolius, Cham. & Schlecht. in Linnea, 1, p. 207; DO. 1. ¢.

LL. Eschscholizianus, Mart. in Schult. Syst. Veg. 7, p. 117.

Struthanthus Eschscholizianus, Mart. in Flora, 1. c.

Has. Chili, near Valparaiso.

5. LoRANTHUS (TRISTERIX) TETRANDRUS, Ruiz & Pav.

Loranthus tetrandrus, Ruiz & Pav. Fl. Per. & Chil. 8, p. 46, t. 275; DC. Prodr. 4
p. 807; Hook. & Arn. 1. ¢.
Tristerix tetrandrus, Mart. in Flora, 1. c. p. 109; Don, 1. c.

)

LORANTHACEGA. 737

Has. Chili; abundant from Valparaiso to the Cordilleras, near
Santiago.

6. LorANtuus (TRISTERIX) APHYLLUS, Miers.

Loranthus aphyllus, Miers, ex Bert.; DC. Prodr. 4, p. 307; Hook. & Arn. in Bot.
Mise. 3, p. 357.

ZL. Cactorum, Hook. & Arn. in Bot. Beech. Voy. p. 25.
Tristerix aphyllus, Don, Syst. Gard. & Bot. 3, p. 418.

Has. Near Santiago, Chili; growing on Cereus Chilensis ; according
to Dr. Pickering’s notes. No specimens are found in the collection.

x * * Novo-Zelandict.
7. LORANTHUS (DENDROPTHOE) TETRAPETALUS, Linn. f.

Loranthus tetrapetalus, Linn. f. Suppl. p. 217; Forst. Prodr. p. 25; Hook. f. Fl. N.
Zeal. p. 99.

Has. Bay of Islands, New Zealand. (Foliage only.)

x * * * Polynesict.

8. Lorantuus (DENDROPTHOE) ForsreRranvs, Schult.

Loranthus Forsterianus, Schult. Syst. Veg. 7, p. 114; DC. Prodr. 4, p. 295; Guill.
Zeph. Tait. p. 54.

L. Stelis, Forst. Prodr. Fl. Ins. Austr. p. 25, & deser. ex Ms. in Zeph. Tait. 1. ¢.

Dendropthoe Forsterianus, Don, Syst. Gard. & Bot. 3, p. 419.

Has. Tahiti and Eimeo, Society Islands.

The specimens (in flower only) furnish nothing of importance to
add to the description given by Schultes, and to that of Forster, which
lias been reproduced by Guillemin. In one specimen the leaves
incline to be obovate and obtuse. The flowers are frequently hexa-

merous.
185

738 PHANEROGAMIA.,

9. Lorantuus (DENDROPTHOE) INSULARUM, Sp. Nov. (Tab. 98.)

L. glaber ; ramis teretibus; foliis oppositis petiolatis ovatis obtusis sub-
quintuplinervits vix venosis ; pedunculis axillaribus brevibus racemoso-
plurifloris; pedicellis trifloris; floribus (sesqui-bipollicaribus) hexa-
heptameris.

Has. Feejee Islands; at Vanua-levu and Rewa. Samoan Islands;
on Tutuila and Savaii. Tongatabu.

There are various forms in the collection; but they appear all to
belong to one species, which is nearly related to LZ. Forsterianus, but
distinguished by its larger and usually rounder leaves and larger
flowers, either hexamerous or heptamerous. It is parasitic on Ino-
carpus and other trees. Branches terete, glabrous, as is the whole
plant. Leaves opposite, coriaceous, ovate, varying to roundish or to
oblong, obtuse, often rounded at the apex, but sometimes narrowed,
either rounded at the base, or abruptly contracted into a petiole of 3
to 6 lines in length, 2 to 32 inches long, 14 to 2 inches wide, rather
dull, the midrib sending off 3 or 4 rather inconspicuous ascending
veins on each side, mostly towards its base, so that the leaf appears
quintuplinerved; the veinlets very obscure or wanting. Peduncles
asillary, or mostly from the old wood whence the leaves have fallen,
short (about half an inch long, the rhachis about the same length or
longer), horizontal, terete, racemosely many-flowered; the pedicels
opposite, 3 or 4 pairs with a terminal one, a line and a half long,
recurved, threeflowered. Flowers hermaphrodite, pendulous, sessile
on the apex of the pedicel, or the two lateral ones slightly pedicel-
late; each subtended at the base of the cylindraceous ovary by a short
and broad, half-clasping, ovate bractlet. Limb of the calyx half a
line long, truncate, slightly repand. Corolla red, or often yellow
towards the base, linear-filiform in the bud, 14 or 2 inches long, of 6
or 7 narrowly linear valvate petals, which at first are connivent or
cohering into an angled tube, the limb (3 lines long) spreading in
anthesis, at length separating quite to the base, and deciduous. Fila-
ments 6 or 7, equal, connate to the face of the petal up to the
spreading portion which constitutes the limb; the free portion about

LORANTHACES?. 739

the length of the narrowly linear and mucronulate anther, which is
fixed by its base. Style as long as the stamens: stigma slightly
capitate, entire. Fruit not seen.

PiaTe 98.—LoRANTHUS INSULARUM: from the Feejee Islands; a
branch, of the natural size. Fig. 1. Pistil, with the calyx, and two
petals with their stamens, enlarged. 2. An anther, more magnified.

* * * * * Australasie?/.
10. Lorantuus (DENDROPTHOE) MAYTENIFOLIUS, Sp. Nov. (Tab. 99.)

L. glaber ; vamis teretibus; foliis oppositis breviter petiolatis ovali-obo-
vatis vel subrotundis scepius retusis opacis obsolete tripli-quintupli-
nerviis; pedunculis terminalibus brevissimis 6-12-floris ; floribus bre-
vissime pedicellatis penta-hexameris; calycis limbo cupulart ovario
paullo breviore; antheris linearibus.

Has. Woolongong, New South Wales.

A. glubrous parasitic shrub, with gray terete branches. Leaves oppo-
site, oval, obovate or roundish, often retuse, narrowed at the base into a
petiole of about 3 lines in length, dull and opaque, obscurely tripli-
nerved or quintuplinerved, the midrib also obscure, except near the
base; the veins not apparent, the texture apparently fleshy-coria-
ceous, 14 to 2 inches in length. Peduncles terminal, or in the upper
axils, very short, not exceeding the petioles, usually dichotomous, the
very short divisions 3-6-flowered. Flowers cymulose, very short-pedi-
celled, or the central one of each cluster sessile, hermaphrodite.
Bractlet oblong at the base of the short cylindraceous ovary. Limb
of the calyx cupulate, almost as long as the ovary, entire, or obscurely
repand. Corolla an inch and a half long, slender, of 5 or 6 narrow
petals connate into a tube, at length separable. Anthers linear,
scarcely thicker than the filament, and about the length of its free
portion, 3 lines long, apiculate. Style straight. Stigma subcapitate.

Pate 99.—LOoRANTHUS MAYTENIFOLIUS: a flowering branch, of the
natural size. Fig. 1. Details of the flower, enlarged. 2. Anther,

more enlarged.

740 PHANEROGAMIA.

11. Loranraus (Psirracantaus ?) CELASTROIDES, Sieber. (Tab. 100.)

L. glaber ; ramis teretibus ; foliis oppositis obovato-oblongis seu ellipticis
basi attenuatis breviter petiolatis obtusissimis fere aveniis ; pedunculis
axillaribus vel ramulos breves bifoliatos desinentibus brevibus cymult-
Joris; floribus breviter pedicellatis pentameris unibracteolatis ; antheris
ovali-oblongis dorso-fixis versatilibus.

Loranthus celastroides, Sieb. Pl. N. Holl. Exsic. no. 244 (sine fl. & fruct.); Schult.
Syst. 7, p. 163; DC. Prodr. 4, p. 818.

Has. Woolongong, New South Wales; and near Sydney.

The specimens bear flowers, which have not before been described.
The plant is glabrous, except an extremely minute pubescence on the
peduncles and nascent parts. Branches terete, nodose. Leaves oppo-
site, obovate, oblong, or elliptical, with a narrowed base, contracted into
a very short petiole, 1$ to 2 inches long, very obtuse, dull, thick and
fleshy-coriaceous in texture, nearly veinless, even the midrib incon-
spicuous except towards the base. Peduncles axillary, or more com-
monly terminating short and two-leaved axillary branchlets, 2 to 5 lines
long, cymosely several-flowered. Flowers im threes, the lateral short-
pedicelled, the intermediate one sessile, or sometimes all pedicelled,
each subtended by a small ovate bractlet, recurved, pentamerous,
Ovary ovoid. Calyx-tube short, coroniform, truncate, puberulent on
the edge, which is entire or obscurely denticulate, at length some-
times 4—5-toothed or lobed. Corolla apparently red or purple, an
inch and a quarter in length, curved in the bud, and the apex cla-
vate-thickened; the slender petals connivent into a tube, but sepa-
rating after anthesis. Filaments free down almost to the middle:
anthers oval, or short-oblong, emarginate at both ends, fixed by the
middle, versatile. Style filiform, as long as the stamens: stigma
minute, subcapitate. Fruit not seen.

Pate 100.—Loranruvs CELASTROIDES: a flowering branch, of the

natural size. Fig. 1. Details of the flower, enlarged. 2, 3, Anthers,
more magnified.

LORANTHACEA. 741

12. Lorantuus (PsiTTAcANTHUS?) EUCALYPTOIDES, DC. I. c.

Loranthus eucalyptifolius, Sieber, Pl. Nov. Holl. Exsic. no. 242; Schult. Syst. Veg.
7, p. 168, non H. B. K.

Has. New South Wales, near Sydney.

Of this species, as of the preceding, neither the flowers nor the
fruit were known. They occur on our specimen, and are almost
exactly like those of L. celastroides. Moreover, some of the leaves,
instead of elongated-lanceolate and somewhat falcate, are elongated-
oblong, very obtuse, and less than two inches in length ;—raising the
question, whether it may not be an extraordinary form of the fore-
going species.

13. LoranTHus (DENDROPTHOE) PENDULUS, Steber.

Loranthus pendulus, Sieber, Pl. Nov. Holl. Exsic. no. 241; DO. Prodr. 1. ¢. & Mem.
Loranth. t. 1.

Has. New South Wales; in the vicinity of Sydney.

14. Lorantaus (DENDROPTHOE) CUNNINGHAMII

Loranthus nutans, A. Cunn. in Mitch. Jour. Trop. Austral. p. 158, in adnot., non
Spreng.

Has. New South Wales, near Sydney.

15. Lorantuus (DENDROPTHOE) PHILIPPENSIS, Cham. & Schlecht.

Loranthus Philippensis, Cham. & Schlecht. in Linnea, 1, p. 204; DC. Prodr. 4,
p. 302.

»

Has. Luzon; in the mountains near Baiios.
186

742 PHANEROGAMIA.

2. TU PEIA, Cham. & Schlecht.

1. Tupera ANTARCTICA, Cham. & Schlecht.

Tupeia antarctica, Cham. & Schlecht. in Linnea, 3, p. 203; Hook. f. Fl. N. Zeal.
p- 101, t. 26.

T. Cunninghamii & T. pubigera, Miquel, in Linnea, 18, p. 85.

Viscum antarcticum, Forst. Prodr. Fl. Ins. Austr. p. 70; A. Rich. Fl. N. Zel. p. 369.

V. pubigerum, A. Cunn. Bot. N. Zeal.

Has. Bay of Islands, New Zealand. (In fruit.)

3. PHORADENDRON, Nutt.

PHORADENDRON, Nutt. in Jour. Acad. Philad. n. ser. 1, p. 185; Engelm. in Gray,
Pl. Fendl. p. 58, adnot.
Viscum, sect. 2, DC. Prodr. 4, p. 279.

1. PHORADENDRON SCHOTTII.
Viscum Schoittit, Pohl, in DC. Prodr. 4, p. 281.

Has. Brazil; on the Organ Mountains.

Only female plants of this and of the two following species were
collected: consequently I cannot verify the essential generic cha-
racter. In all of them the ovary is crowned with a three-lobed
perianth, as in many Visca.

2, PHORADENDRON PEPERICARPUM, Sp. Nov.

P. ramis teretibus ; foliis ovalibus ovato-oblongisve obtusis basi attenuatis
breviter petiolatis crassis tri-quinquenerviis aveniis ; spicis in axillis
gemini, fructiferis folio triplo” quadruplove brevioribus, vaginulis
bifidis ; baccis globosis.

LORANTHACEA. 743

Has. Brazil; on the Organ Mountains. (Also gathered by Gard-
ner, No. 5478.)

There are only fruiting specimens of this species; which is appa-
rently related to Pohl’s Viscum crassifolium. The plant is large,
with long and terete branches and branchlets, glabrous. Leaves 3 or 4
inches long, rather narrowly oval or ovate oblong, obtuse, narrowed at
the base into a petiole of about 3 lines in length, thick, evidently 3—5-
nerved, otherwise veiniess. Fertile spikes in pairs, or sometimes in
threes, in the axils of the leaves, slender, in fruit an inch or an inch
and a half long; the many-flowered joints 8 lines long; the short
sheaths two-cleft, very obtuse. Berries spherical, about a line and a
half in diameter, but immature, their bases separately immersed in
hollows of the rhachiy,

3. PHORADENDRON CHRYSOCLADON, Sp. Nov.

P. ramulis compressis; ramis teretibus auratis; foliis ovatis scepius
acuminatis brevissime petiolatis crassis subtus obsolete tripli-quintupli-
nerviis costa prominula supra nitidulis reticulato-venosis; spicis in
axillis solitariis ternisve, fructiferis folio paullo brevioribus pauci-arti-
culatis, articulis truncato-bilobis.

Has. Brazil; near Rio Janeiro.

Apparently a widely diffused species; as it appears to be the same
as one gathered at Bahia by Salzmann, at Piauhy by Gardner (No.
2884), and in British Guiana by Schomburgk (No. 554, of Ist coll.).
It is evidently allied to Visewm flavens, Swartz; but the venation is
different, and the spikes are proportionately longer. We have only
the female plant. Branches terete, tinged with golden yellow; the
young branchlets compressed or ancipital. Leaves 2 or 3 inches long,
ovate or oval-oblong, mostly acuminate, often conspicuously so, abruptly
contracted at the base into a very short petiole, thick and coriaceous,
often yellowish, obscurely triplinerved or quintuplinerved underneath,
the narrow costa prominent, above lucid and conspicuously reticulate-

744 PHANEROGAMIA.

veny. Fertile spikes solitary or in threes in the axils, 14 to 2 inches
long when fully developed, of 3 or 4 elongated and many-flowered
joints; the sheaths truncately two-lobed. Female flowers sunk in
excavations of the rhachis. Perianth three-lobed. Young fruit glo-

bular.

4, VISCUM, Tourn.
1. Viscum ARTICULATUM, Burm.

Viscum articulatum, Burm. FI. Ind. p. 811; DC. Prodr. 4, p. 284.
V. compressum, Poir. Dict. Suppl. 2, p. 861; Decaisne, Herb. Timor. p. 87; Blume,

Fl. Jav. Loranth. t. 26.

Has. Feejee Islands; Ovolau, Vanua-levu, ‘&c.: parasitic on IJno-
carpus. Also, a remarkably condensed form, in the mountains of

Muthuata.

2. VIscUM MONILIFORME, Bluse.

Viscum moniliforme, Blume, Bijdr. p. 667, & Fl. Jav. l.c. t. 25; DO. le; Cham.
& Schlecht. in Linnza, 3, p. 202; Wight, Ic. Pl. Ind. Or. t. 1018, 1019.

V. opuntioides, Forst. Prodr. Fl. Ins. Austr. p. 70, non Linn.

V. platycaulon, Bertero, ined.

Has. Tahiti, Society Islands (mostly with narrow articulations).
Sandwich Islands: on Acacia Koa, and other trees :—the articulations
varying in different specimens from a line and a half to an inch in

breadth!

No male flowers were collected; but those of V. articulatum are
figured by Blume (under the name of V. compressum, Poir.) with
adnate and multiporose anthers. The form of the articulations is
exceedingly various in different specimens; in some states they seem
to be almost terete: others appear to pass into the preceding species.
Their elucidation requires better materials than the present collection
affords.

LORANTHACES, 745

2 MYZODENDRON, Banks & Soland.

1. MyzopENDRoN PuNcTULATUM, Banks & Soland.

Misodendron (sphalm.) punctulatum, DC. Prodr. 4, p. 286, & Mem. Loranth. t. 11, 12.
Myzodendron (Gymnophyton) punctulatum (Banks & Soland. Mss.), Hook. f. FI.
Antare. p. 289, t. 102, 104, 106.

Has. Orange Harbour, Fuegia. “Very common and conspicuous
on the Antarctic Beech.”

2. MyzopENpRON BRACHYSTACHIUM, DC.

Myzodendron (Misodendron) brachystachium, DO. Mem. Loranth. p. 14, t. 12, f. 1,
& Prodr. l.c.; Hook. f. Fl. Antare. p. 297, t. 103, 105, 106, 107.

Has. Orange Harbour, Fuegia; on Fugus Antarctica and F. betu-
loides.

The botanical history, structure and affinities, as well as the ana-
tomy and parasitism of Myzodendron, have been most ably and elabo-
rately illustrated by Dr. Hooker, in the work above cited. I have
only to add that I. oblongifolium, DC. 1.¢.; Deless. Ic. Sel. 3, t. 80;
Poepp. & Endl. Nov. Gen. & Spec. t. 2; Hook. f. l.c., is not to be
distinguished from Jf. brachystachium by the foliage (our specimens
combining the two on the same individuals), but only by the greater
length of the plumose sete of the fruit, a character not likely to be
constant; so that this species will doubtless be suppressed.

ADDITIONS AND CORRECTIONS.

To p. 74, lowest line, after Kuhlia Brasiliensis, add: (Banara Velloztt,
Gardn. in Hook. Lond. Jour. Bot. 2, p. 331.)

To p. 87, add: 2. IonrpIum sETIGERUM, St. Ail.

Tonidium setigerum, St. Hil. Fl. Rem. Bres. p. 297, t. 27, & Fl. Bras. Mer. 2, p. 142.
Has. Brazil; in the Organ Mountains, near Rio Janeiro.

To p. 138. ScurepEA pirFuSA.— This was also collected by Mr.
Macrae, on Mouna Kea, in June, 1825. His specimen has a less
effuse panicle, as well as shorter and entirely glabrous pedicels.

Page 206, line 2 from bottom, for ejusdem read. eodem.

Page 291. Perrottetia Sandwicensis was first collected by Mr. Nelson,
in Cook’s third voyage.

INDEX.

SYNONYMES, AND THE NAMES OF GENERA AND SPECIES INCIDENTALLY MENTIONED, ARE IN ITALIC.

A. Acacia juniperina, 478.
Koa, 480.
Abelmoschus esculentus, 172. lacerans, 484.
Manihot, 172. laurifolia, 482.
moschatus, 172. linearis, 479.
Abrus precatorius, 436. linifolia, 479.
Abutilon carneum, 167. longifolia, 480.
esculentum, 167. melanoxylon, 479.
glaucum, 167. mollissima, 483.
incanum, 168. myriadenia, 485.
Indicum, 167. myrtifolia, 479.
mollissimum, 168. pellacantha, 484.
rufinerve, 167. prominens, 479.
tubulosum, 169. pteridifolia, 484.
Acacia albida, 484. Richii, 482.
qi angustifolia, 479. Achyranthes tenuifolia, 122.
i Baueri, 478. Acicalyptus myrtoides, 551.
2 Brownii, 477. Ackama roscefolia, 671.
4 Cavenia, 484. Acena adscendens, 496.
cultriformis, 478. Antarctica, 496.
decurrens, 483. exigua, 498.
discolor, 483. lappacea, 497.
echinula, 478. Magellanica, 495.
elata, 483. myriophylla, 497.
elongata, 479. ovalifolia, 496.
Farnesiana, 484. ovina, 498.
floribunda, 480. pinnatifida, 497.
glaucescens, 480. pumila, 499.
heterophylla, 480. Sanguisorbee, 498.
hispidula, 478. spectabilis, 483.

insularum, 477. splendens, 497.

748

Aceena suaveolens, 479.
trifida, 497.
vestita, 478.
Acmadenia juniperina, 328.
Acmena floribunda, 518, 526.
parviflora, 520.
Actinotus Helianthi, 704.
minor, 705.
Acronychia arborea, 336.

Endlichert, 336, 338.

heterophylla, 333.
levis, 836, 339.
petiolaris, 335.
retusa, 3388.
Richii, 836.
Adesmia balsamica, 426.
Echinus, 427.
filipes, 425.
hispidula, 425.
horrida, 426.
microphylla, 426.
trijuga, 427.
Adrastea salicifolia, 18.
Kschynomene ciliata, 430.
rudis, 429.
sensitiva, 430.
Affonsea bullata, 487.
Afzelia bijuga, 467.

Madagascariensis, 469.

Agalmanthus umbellatus, 557.
Agatea, 609.
Agatea violaris, 89.
Agathosma spicata, 327.
Agati coccinea, 408.
grandiflora, 408.
tomentosa, 409.
Aglaia basiphylla, 237.
edulis, 237.
odorata, 236.
Samoensis, 236.
Alchemilla pinnata, 500.
tripartita, 499.
Alectryon excelsum, 260.
Aleurodendron, 198.
Algarobia glandulosa, 474.
juliflora, 474.
Allophyllus ternatus, 249.

INDEX.

Alphitonia franguloides, 280.
zizyphoides, 278.

Alsinez, 115.

Alsodeia physiophora, 92.

Alysicarpus vaginalis, 486.

Alyssum maritimum, 51.

Amaroria soulameoides, 356.

Amblyarrhena, 595.

Ammannia latifolia, 605.

Ammi majus, 708.

Visnaga, 708.
Amphirrox longifolia, 88.
Amphithalea densa, 388.
Anacardiacesxe, 364.
Anacardium occidentale, 364.
Anaxagorea Luzonensis, 27.
Anchietea salutaris, 88.
Ancistrum anserinifolium, 498.

diandrum, 498.
humile, 496.
Magellanicum, 495.
repens, 496.
Sanguisorbe, 498.

Androsemum Webbianum, 219.

Anemone helleborifolia, 4.

Angelica rosefolia, 710.

Angophora, cordifolia, 555.

intermedia, 556.

lanceolata, 555.
Angustinea, 593.
Anisotome latifolia, 709.

roseefolia, 710.

Anoda ovata, 161.

Anona Americana, 25.
Cherimolia, 26.
dolabripetala, 26.
palustris, 25.
squamosa, 25.
xestropetala, 26.

Anonacex, 25.

Anplectrum ovalifolium, 597.

Anthodon undulatum, 286.

Anthonota, 469.

Antichorus depressus, 196.

Aotus villosa, 883.

Apeiba australis, 195.

Apium australe, 706.

Apium Chilense, 706.
graveolens, 707.
_graveolens, 706.
prostratum, 706.
Aplectrum, 597.
Aporetica Gemella, 249.
pinnata, 259.
Aquifoliaceze, 294.
Arabis albida, 47.
Arachis hypogeea, 424.
Aralia crassifolia, 718.
edulis, 715.
Lessoni, 719.
polaris, 714.
Quinduense, 715.
racemosa, 715.
Scheefflera, 715.
spinosa, 715.
trigyna, 719.
Vitiensis, 715.
Araliacea?, 733.
Araliacese, 714.
Arenaria Alpamarce, 116.
andicola, 116.
bryoides, 117.
dicranoides, 117.
diffusa, 115.
jloribunda, 121.
nemorosa, 115.
scopulorum, 116.
serpens, 116.
serpylloides, 116.
Argemone Mexicana, 44.
Aristotelia Maqui, 214.
Arnoldia, 670, 676.
Arrudea clusioides, 214.
purpurea, 215.

Arthostemma Martiusianum, 601.

Ascyrum humifusum, 220.
involutum, 220).

Asimina, 31.

Aspalathus acuminata, 394.
canescens, 394.
capillaris, 395.
cymbeformis, 395.
galioides, 395.
microphylla, 394.

INDEX. 749

188

Aspalathus spicata, 894.
spinosa, 394.
uniflora, 395.
Asteriscium Chilense, 704.
Peeppigiti, 704.
Asterotrichion sidoides, 181.
Astragalus alienus, 418.
Alpamarcee, 417.
Brackenridgei, 416.
canescens, 413.
distans, 412.
Garbancillo, 414.
geminiflorus, 418.
minimus, 418.
ochroleucus, 413.
Pickeringii, 415.
procumbens, 414.
pusillus, 418.
Richii, 414.
uniflorus, 418.
untfultus, 414.

Astronia confertiflora, 579.
Forsteri, 577.
fraterna, 576.
Pickeringii, 577.
subcordata, 580.

Astronidium parviflorum, 582.

Astroniee, 576.

Astrotriche latifolia, 695.

Atalantia monophylla, 234.

Aubletia caseolaris, 550.

Aurantiaceze, 234.

Ayenia microphylla, 189.
pusilla, 189.

tomentosa, 189.

Azara celastrinea, 75.

integrifolia, 75.
Lilen, 75.
serrata, 79.

Azorella apoda, 699.
bolacina, 701.
cespitosa, 700, TOL.
daucoides, 712.
diapensioides, 702.
filamentosa, 698.
eladorhiza, 702.
glebaria, 701.

750

Azorella lanceolata, 694.
lineartfolia, 695.
lycopodioides, 698.
madreporica, 700.
monanthos, 699.
multifida, 702.
Ranunculus, 698.
reniformis, 698.
Selago, 699.
tricuspidata, T01.
trifoliolata, 698.

B.

Backhousia myrtifolia, 551.
riparia, O51.

Beckia affinis, 571.
crenulate, 570.
densifolia, 571.
diffusa, 571.
diosmifolia, 570.
leptocaulis, 570.
linifolia, 570.

Balanopteris minor, 184.

Tothila, 184.

Banara, 75.

Vellozii, T46.

Banisteria campestris, 265.

monoptera, 266.
nitida, 266.
sericea, 256.
Barbarea australis, 47.
Barringtonia Butoniea, 508.
excelsa, 508.
racemosa, 508.
Samoensis, 508.
speciosa, 508.
Bauera microphylla, 681.
rubicfolia, 681.
rubioides, 681.

Bauerece, 664.

Bauhinia acidu, 470.
Brasiliensis, 469.
grandiflera, 469.
purpurea, 470.
Raddiana, 470.
ungulata, 470.

INDEX.

Begonia sequata, 658.
castanezefolia, 655.
coccinea, 656.
Cumingii, 658.
erecta, 657.
Fischeri, 656.
geraniifolia, 657.
hastata, 656.
Hookeriana, 656.
luxurians, 654.
octopetala, 657.
patens, 655.
ramentacea, 655.
repens, 656.
reticulata, 656.
truncata, 655.
vaginans, 655.
villosa, 657.

Begoniaceze, 654.

Belangera speciosa, 681.

Berberidacex, 42.

Berberis actinacantha, 43.
buxifolia, 42.
Chilensis, 43.
crispa, 43.
dulcis, 42.
cuneata, 44.
empetrifolia, 43.
ferox, 42.
horrida, 48.
ilicifolia, 42.
lagenaria, 42.
microphylla, 42.

| Berlinia, 469.
| Bernardinia Fluminensis, 377.
| Bertolonia Leuzeana, 596.

nympheeeefolia, 596.

| Beythea bifida, 205.

Bigelovia Brasiliensis, 79.

| Billardiera mutabilis, 222.

scandens, 221.
Bixa Orellana, 72:
Bixinex, 72.
Boca serrata, 74.
Bolax, 669, 700.
complicata, 701.
glebaria, 701.

Bolax gummifera, 701.
Borbonia barbata, 387.
cordata, 387.
lanceolata, 387.
Boronia ledifolia, 331.
parviflora, 330.
pinnata, 331.
polygaleefolia, 331.
serrullata, 330.
tetrandra, 331.
elata, 696.
lobata, 696.
multiradiata, 696.
palmata, 696.
tropeeolifolia, 696.
Bossiza heterophylla, 386.
scolopendra, 386.
scolopendria, 411.
microphylla, 387.
Botryodendrum cerberoides, 731.
lancifolium, 734.

macrophyllum, 782.

Taitense, 731.
Bowlesia acutangula, 696.
Brachychiton populneum, 185.
Brackenridgea nitida, 362.
Braya imbricatifolia, 58.
pusilla, 57.

Brotera Leprieurti, 194.
Broussaisia arguta, 683.

pellucida, 685.
Bruguiera gymnorhiza, 614.

Rumphii, 614.

Brunellia Sandwicensis, 839, 345.
Brunia nodiflora, 735.
Bruniacez, 735.
Bryophyllum calycinum, 690.
Bubon Galbanum, 710.
Buchanania florida, 366.
Bulliarda Magellanica, 687.

moschata, 689.
Bunchosia Armeniaca, 263.

Martiana, 263.
Bupleurum salicifolium, 708.
Bursaria diosmoides, 221.

spinosa, 221.

Bursinopetalum, 301.

INDEX. 751

Buttneria dasyphylla, 188.
pannosa, 188.
Buttneriacese, 187.
Byronia Sandwicensis, 296.
Taitensis, 297.
Byrsonima Brasiliensis, 263.
sericea, 263.

C.

Cactacese, 659.

Ceesalpinese, 461.

Ceesalpinia corymbosa, 461.
glabrata, 461.
pulcherrima, 461.
Sappan, 461.

Caiophora absinthiifolia, 634.
contorta, 635.
coronata, 634.

Cajanus bicolor, 553.

flavus, 453.
Indicus, 453.
Calandrinia acuta, 143.
arenaria, 143.
biflora, 1438.
Gayana, 148.
glauca, 142.
Phacosperma, 143.
Pickeringii, 144.
Caldasia andicola, 712.
cherophyllea, 712.
cherophylloides, 712.
Callicoma serratifolia, 666.
Callistachys lanceolata, 378.
sparsa, 378.
Callistemon lancéolatum, 566.
pinifolium, 566.
Calophyllum Inophyllum, 218.
spectabile, 218.
Calpandra lanceolata, 215.
Caltha appendiculata, 13.
dionecefolia, 14.
multicapsularis, 12.
paradoxa, 13.
sagittata, 12.
Calycothrix Billardieri, 572.
glabra, 572.

752 INDEX.

Calycothrix scabra, 572. Cardamine pratensis, 49.

Calythrix glabra, 572. sarmentosa, 51.
scabra, 572. sylvatica, 50.
tetragona, 572. tenuirostris, 49.
Campomanesia cerasoides, 549. Cardiandra, 664.
Cananga, 30. Cardiospermum Halicacabum, 247.
Canarium album, 374. Loxense, 247.
Luzonicum, 374. microcarpum, 247.

Vitiense, 373.
Canavalia galeata, 441.

Gaudichaudit, 441.

gladiata, 440.
mintata, 440.
obtusifolia, 440.
pubescens, 441.
rosea, 440.
rutilans, 440.
sericea, 440.
turgida, 440.

Candela Americana, 618.

Capellenia, 15.

Capellia biflora, 15.
membranifolia, 17.
multiflora, 17.

Capparidacez, 65.

Capparis aurantioides, 70.
Brasiliensis, 68.
cerasifolia, 71.
cynallophora, 68.

cynophallophora, 68.

nemorosa, 70.
odorata, 71.
Richii, 69.
Sandwicensis, 70.
sepiaria, 70.

Carallia, 609.

Carapa Moluccensis, 243.
obovata, 243.

Cardamine antiscorbutica, 49.

Chilensis, 50.
cordata, 41.
corymboza, 48.
geraniifolia, 48.
glacialis, 48.
hirsuta, 50.
laxa, 50.

nivalis, 49.

Carica candicans, 640.
Papaya, 640.
Carmicheelia australis, 411.

Cunninghamii, 411.

Carpenteria, 663.
Carpodetus serratus, 665.
Carpotroche Brasiliensis, 72.
Caryophyllacezx, 109.
Casalea Chilensis, 6.
Casearia acuminatissima, 80.
affinis, 79.
disticha, 81.
inzequilatera, 79.
Melistaurum, 81.
oblongifolia, 79.
parviflora, 79.
Richii, 82.
Casparea castrata, 470.
Cassia aciphylla, 465.
alata, 462.
Apoucouita, 465.
australis, 465.
bifoliolata, 466.
calycioides, 466.
Candolleana, 464.
Chameecrista, 466.
calliantha, 464.
coluteoides, 464.
ensiformis, 465.
eremophila, 465.
Jlexuosa, 464.
Jrondosa, 464.
Gaudichaudti, 463.
glauca, 464.
hirsuta, 464.
latopetiolata, 464.
Leschenaultiana, 466.
montana, 463.
oblongifolia, 463.

Cassia obtusifolia, 463.
occidentalis, 462.
ramifiora, 465.
ramosa, 466.
rotundifolia, 466.
Sophera, 463.
umbellata, 465.
tomentosa, 463.
zygophylla, 465.

Cassine Maurocenia, 294.

Magellanica, 290.

Catha cassinoides, 289.
crenata, 288.
Vitiensis, 287.

Ceanothus Asiaticus, 277.

capsularis, 277.

Celastraceze, 287.

Celastrus Magellanicus, 290.
crenatus, 288.
pyracanthus, 287.
Richii, 289.

Centrosema decumbens, 437.

Plumieri, 437.

Cerastium arvense, 119.

Beeringianwm, 120.
glomeratum, 121.
viscosum, 120.
vulgatum, 121.

Ceratopetalum gummiferum, 666.

Cercodea alternifolia, 625.
erecta, 625.
Cereus Chilensis, 660.
Cheenanthera, 596.
Chenopleura, 593.
Cherophyllum Chilense, 713.
Cheetocalyx longiflorus, 423.
Cheetogastra fraterna, 604.
gracilis, 604.
Chalarium, 435.
Chamzemeles coriacea, 506.
Chamitis integrifolia, 698.
Cheiranthus mutabilis, 46.
Chorisia speciosa, 183.
Chorozema gomphocarpum, 380.
Chrysobalanaceze, 488.
Chrysomelon pomiferum, 374.
Chrysosplenium elevatum, 687.

INDEX.

189

Chrysosplenium macranthum, 686.

Cissampelos australis, 38.
Capensis, 38.
discolor, 38.
Pareira, 38.
Cissus Antarctica, 271.
deficiens, 271.
acuminata, 273.
geniculata, 272.
hypoglauca, 272.
quinquefolia, 271.
sylvatica, 271.
Vitiensis, 272.
Citriobatus multiflorus, 221.
Citronella mucronata, 297.
Citrus Decumana, 236.
torosa, 236.
Clavimyrtus, 509, 514.
Clematis aristata, 2.
dioica, 3.
Heenkeana, 3.
indivisa, 2.
integrifolia, 2.
parviflora, 2.
Peruyviana, 3.
Pickeringii, 1.
sericea, 3.
Cleobulia multiflora, 459.
Cleome affinis, 67.
atropurpurea, 67.
Chilensis, 66.
dendroides, 67.
diffusa, 66.
icosandra, 68.
pentaphylla, 65.
Richii, 67.
Sandwicensis, 65.
spinosa, 69.
viscosa, 68.
Clidemia acutiflora, 593.
alpestris, 592.
althaoides, 587.
amygdaloides, 586.
biserrata, 593.
botryophora, 586.
Brackenridgei, 591.
bullosa, 593.

753

75d INDEX.

Clidemia Carassana, 588. Colobanthus subulatus, 132.
cheetocalyx, 589. Colubrina Asiatica, 277.
confertiflora, 587. Combretacez, 615.
crenata, 598. Comesperma conferta, 105.
dasytricha, 588. ericina, 106.
dispar, 587. nudiuscula, 106.
Epibaterium, 587. virgata, 105.
leptostachya, 586. Commersonia echinata, 188.
longibarbis, 586. Fraseri, 188.
marginata, 587. Javensis, 188.
Martii, 592. platyphylla, 188.
Martiana, 592. Condalia lineata, 275.
Nianga, 586. Connaracez, 375.
oocarpa, 590. Connarus Beyrichii, 377.
pauciflora, 598. cymosus, 377.
Pickeringii, 592. Pickeringii, 375.
pulchra, 591. Corchorus Antichorus, 196.
retropila, 587. capsularis, 196.
scandens, 587. olitorius, 195.
spicata, 593. tortipes, 195.
stenopetala, 586. trilocularis, 195.
umbonata, 598. tridens, 196.

Cliffortia dentata, 494. Coriaria ruscifolia, 306.
falcata, 494. sarmentosa, 306,
ilicifolia, 494. Coriariaceze, 306.
odorata, 494. Cornaceze, 734.
ruscifolia, 494. Cornidia, 6638.
sarmentosa, 494. : Corokia buddleoides, 734.
strobilifera, 494. Corrigiola squamosa, 126.
trifoliolata, 494. Corynocarpus leevigata, 372.

Clusia Ganabarica, 216. Correa alba, 328.

lanceolata, 216. reflexa, 329.
rupicola, 216. virens, 329.
sessilis, 216, 341. viridiflora, 329.

Clypea, 36, 37. Coulteria tinctoria, 462.

Cocculus Ferrandianus, 39. Crassula ciliata, 690.
Forsteri,.36. moschata, 689.
velutinus, 40. muscosa, 689.

Coleonema pulchrum, 328. pellucida, 690.

Colletia spinosa, 276. scabra, 689.

ulicina, 276. tetragona, 689.

Colobanthus aretioides, 131. squamulosa, 690.

Benthamianus, 132. Crassulaceze, 688.

crassifolius, 131. Cratzeva tapioides, 68.

muscoides, 131. Cremanium chenopleuroides, 596.
Quitensis, 131. Sellovianum, 595.

saginoides, 181. Cremolobus aphanopterus, 55.

Cremolobus pinnatifidus, 56.
Cristaria ecristata, 165.
Crithmum maritimum, 710.
Crossostylis biflora, 610.
Crotalaria anthylloides, 390.
ealycina, 390.
fulva, 390.
incana, 389.
microphylla, 389.
quinquefolia, 390.

Roxburghiana, 890.

sagittalis, 389.
Senegalensis, 388.
sessiliflora, 390.
stipularia, 389.
striata, 389.
trigonelloides, 389.
verrucosa, 390.
vitellina, 389.

Croton ertospermum, 108.

Crowea saligna, 329.

Cruciferee, 46.

Cucumis bicirrha, 646.
pubescens, 646.

Cucurbitacex, 641.

Cunoniacee, 663.

Cunoniec, 663.

Cupania anacardizefolia, 259.
anacardioides, 258.
Brackenridgei, 255.
falcata, 252.
lentiscifolia, 256.
leptobotrys, 255.
rhoifolia, 254.
Richii, 257.
subcinerea, 258.

Cuphea Balsamona, 606.
flava, 606.
ingrata, 606.
ligustrina, 606.

Cussonia Lessonii, 719.
thyrsiflora, 714.

Cyclanthera Matthewsii, 647.

Cytisus scoparius, 396.

tener, 396.

Cynometra grandiflora, 470.

INDEX, 755

Cynometra faleata, 472.
ineequifolia, 473.

D.

Dalbergia cassioides, 457.
multijuga, 457.
myriantha, 458.
torta, 458.
variabilis, 458.

Dalea cylindrica, 401.

Mutisii, 401.
Onobrychis, 401.

Dalibarda geoides, 503.

Darwinia fascicularis, 571.
laxifolia, 572.
taxifolia, 572.

Daucus australis, 711.

brachiatus, 711.
Carota, 711.
Montevitlensis, 711.
pusillus, 711.
toriloides, 711.

Daviesia acicularis, 382.

corymbosa, 382.
ulicina, 582.
Davilla Brasiliana, 15.
rugosa, 15,

Decarphe, 593.

Decaspermum fruticosumy 547.

Decumaria, 663. id

Dendrolobium australe, 431.

umbellatum, 431.

Dendropthoe Forsterianus, 737.

Derris uliginosa, 457.

Descurainia, 60.

Descurea, 60.

Desmanthus depressus, 475.

| Desmodium ancistrocarpum, 434.

axillare, 434.
australe, 481.
barbatum, 482.
brachypodum, 434.
capitatum, 433.
Chamissonis, 433.
Gangeticum, 433.
Gunnianum, 439.

756

Desmodium heterocarpum, 432.
incanum, 434.
leptopus, 436.
Limense, 433.
nervosum, 432.
obovatum, 434.
Peruvianum, 484.
polycarpum, 432.
purpureum, 432.
reptans, 434.
Scorpiurus, 435.
Sandwicensis, 433.
tortuosum, 435.
triflorum, 432.
Deutzia, 663.
Dianthus prolifer, 109.
Dicerma pulchellum, 432.
Dichoglottis australis, 118.
Dichroanthus mutabilis, 46.
Diclidocarpus Richii, 200.
Didiscus albiflorus, 694.

Didymochiton Gaudichaudianum, 241.

Richii, 239.

Dillenia speciosa, 20.

volubtlis, 20.

Dilleniaceze, 15.

Dillwynia ericeefolia, 383.
glaberrima, 384.
parviflora, 384.
pedun@alaris, 383.
tenuifolia, 384.

Dimereza, 254.

Dimorphanthus edulis, 715.

elatus, 715.

Dioclea violacea, 439.

Diosma succulenta, 328.

Diosmec, 327.

Diplochita, 5938.

Discostigma caudatum, 217.

Vitiense, 216.

Disemma Herbertiana, 638.

Disoxylon Samoense, 241.

Dissochzeta Cumingii, 600.

Ditoca muscosa, 180.

Dodonxa Brasiliensis, 260.
Burmanniana, 260.
calycina, 262.

INDEX.

Dodonzea cuneata, 262.
erlocarpa, 261.
Jamaicensis, 260.
neriifolia, 262.
spathulata, 261.
triquetra, 262.
viscosa, 260.

Dolichos altissimus, 439.
falcatus, 453.
galeatus, 441.
luteolus, 452.
luteus, 452.

Donatia fascicularis, 688.

Magellanica, 688.

Draba Brackenridgei, 53.

cryptantha, 51.
imbricatifolia, 58.
Macleani, 52.
Pickeringii, 54.

Dracontomelon sylvestre, 274.

Draytonia rubicunda, 207.

Drosera Anglica, 100.

Cunninghamii, 98.
dichotoma, 99.
intermedia, 98.
longifolia, 100.
micrantha, 100.
pedata, 99.
peltata, 99.
petiolaris, 100.
pusilla, 100.
pygmea, 100.
spathulata, 100.
uniflora, 98.

Droseracez, 98.

Drymaria apetata, 126.
cordata, 122.
divaricata, 124.
fasciculata, 125.
glaberrima, 123.
grandiflora, 126.
macrantha, 126.

rotundifolia, 123.
viscidula, 124.

Drymis Chilensis, 24.

punctata, 24,
Winteri, 24.

Dryptopetalum, 607.

Duvaua dentata, 370.
dependens, 370.

Dysemone lobata, 631.

K.

Edwardsia Chilensis, 460.
chrysophylla, 459.
grandiflora, 458.
microphylla, 458.

Eleeocarpus bifidus, 205.

cassinoides, 204.
eyaneus, 202.
Griffthii, 203.
laurifolius, 203.
petiolatus, 203.
pyriformis, 205.
reticulatus, 202.
Elsneria crateegifolia, 696.
Entada Adenanthera, 478.
Gigalobium, 473.
monostachya, 473.
Pursetha, 473.
scandens, 473.

Entelea arborescens, 195.

Entoganum levigatum, 350.

Eperua, 469.

Epilobium alsinoides, 622.
atriplicifolium, 622.
Billardierianum, 623.
Bonplandianum, 621.
cinereum, 628.
glabellum, 622.
hirtigerum, 623.
incanum, 628.
junceum, 623.
micranthum, 623.
microphyllum, 622.
nerterotdes, 622.
nivale, 621.

nummulariefolium, 622.

pallidiflorum, 623.
parviflorum, 621.
pedicellare, 621.
pedunculare, 622.
pubens, 622.

INDEX.

190

757

Epilobium thymifolium, 622.
tetragonum, 621.
virgatum, 628.

Epiphylium, 661.

Eriocalia major, 704.

minor, T05.

Hriodendron trischistandrum, 182.

Kriostemon buxifolium, 330.

cuspidatum, 330.
salicifolium, 330.
verrucosum, 330.
Erodium cicutarium, 317.
malachoides, 317.

Eryngium coronatum, 706.
ebracteatum, 706.
paniculatum, 706.

Erysimum pusillum, 57.

Erythrina Indica, 444.
macrophylla, 444,
monosperma, 444.
ovalifolia, 445.

Erythroxylacez, 270.

Erythroxylon deciduum, 270.

nanum, 270.
nitidum, 270.
pulchrum, 270. ~

Escalloniee, 663.

Escallonia affinis, 664.
arguta, 665.
Berteriana, 665,
illinita, 664.
leucantha, 665.
Meriensiana, 664.
pulverulenta, 665
resiniflua, 665.
revoluta, 664.
rubra, 664.
serrata, 663.

Eucalyptus acervula, 553.
botryoides, 554.
marginata, 553.
miniata, 554.
multiflora, 554.
pallens, 554.
paniculata, 554.
platypodos, 554,
radiata, 553.

758

Eucalyptus rostrata, 553.
subulata, 553.

Eucladus suffruticosus, 187.

Eugenia affinis, 537.
Amicorum, 524.
apiculata, 537.
arenaria, 517.
australis, 514.
Benthamii, 520.
Brasiliensis, 517.
Brackenridgei, 521.
Bridgesti, 536.
brevifolia, 531.
Chekan, 587.
clusizefolia, 528.
confertiflora, 523.
correcefolia, 542.
corynocarpa, 526.
Cruckshanksti, 540.
Cumingii, 536.
effusa, 524.
elliptica, 518.
Jerruginea, 542.
Gayana, 530.
Gilliesti, 537.
gracilipes, 513.
inophylloides, 521.

leptospermoides, 536.

Maire, 518.
Malaccensis, 510.
maritima, 536.
Michelit, 516.

multiflora, 5386, 539.

myrtifolia, 514.
nemoralis, 516.
neurocalyx, 512.
obtusa, 541.
ovata, 5386.
paniculata, 525.
planipes, 536.
quadrangulata, 511.
rariflora, 514.
Richii, 510.
rotundifolia, 517.
rubescens, 525.
Sandwicensis, 519.
Salzmanni, 517.

INDEX.

Kugenia Savaiensis, 530.
Selkirkii, 536.
Smithii, 518, 526.
stenophylla, 540.
tenella, 517.
Temu, 539.
Tutuilensis, 529.
Ugni, 544.
uniflora, 516.
zanthoxyloides, 517.
Zeylanica, 520.
Kuodia drupacea, 332.
hortensis, 332.
longifolia, 332.
Kurya Pickeringii, 211.
Richii, 212.
Sandwicensis, 209.
Vitiensis, 210.
Kuryodes, 211.
Kuryomyrtus diffusa, 571.
Eustylis, 710.
Kvia dulcis, 374.

F.

Fabricia myrtifolia, 568.
Fagara Euodia, 332.
Fagelia bituminosa, 453.
Fagonia aspera, 326.
Fendlera, 663.
Firmiana diversifolia, 185.
Flacourtia inermis, 75.
Flacourtiaceze, 73.
Flemingia strobilifera, 454.
Feeniculum vulgare, 709.
Fragaria Chilensis, 500.
vesca, 500.
Fragosa cladorhiza, 702.
multifida, T02.
spinosa, 703.
Fuchsia affinis, 624.
coccinea, 624.
excorticata, 624.
integrifolia, 623.
lycioides, 624.
rosea, 624.

Fumaria agraria, 45.
capreolata, 45.
media, 45,
officinalis, 45.

Fumariacee, 45.

G.

Galactia hispidula, 438.
Galega littoralis, 407.
Gamochleena, 173.
Garcinia Mangostana, 267.
Gastonia Oahuensis, 726.
Gaya subtriloba, 166.
Geissois ternata, 679.
Gelpkea, 509.
Gemella trifoliata, 249.
Genista Cumingii, 396.
Gayana, 397.
Maderensis, 396.
umbellata, 397.
virgata, 396.
Geraniaceex, 308.
Geranium arboreum, 315.
Berterianum, 309.
crassipes, 309.
cuneatum, 312.
diffusum, 308.
dissectum, 308.

microphyllum, 308.

multiflorum, 311.

multipartitum, 310.

ovatifolium, 314.

Patagonicum, 308.

ptlosum, 308.
potentilloides, 308.
sessiliflorum, 309.
retrorsum, 308.
Geum Chilense, 501.
Chiloense, 501.
intermedium, 501.
involucratum, 501.
Magellanicum, 501.
Quillyon, 501.
parviflorum, 501.
ranunculoides, 501.

INDEX. 759

Geum strictum, 501.
urbanum, 501.
Glossocentrum, 598.
Glossospermum, 198.
Glycyrhiza astragalina, 406.
Godetia tenuifolia, 620.
Gomphia semula, 360.
Hookert, 361.
tteoides, 360.
linearis, 359.
oliveeformis, 361.
parviflora, 360.
stipulacea, 360.
Gompholobium barbigerum, 380.
glabratum, 381.
glaucescens, 381.
grandiflorum, 380.
latifolium, 381.
pinnatum, 381.
virgatum, 381.
uncinatum, 381.
Goniocarpus citriodorus, 626.
depressus, 626.
micranthus, 626.
serpyllifolia, 626.
tenellus, 626.
tetragyna, 625.
teucroides, 625.
Gonocarpus micranthus, 626.
Gossypium album, 179.
Barbadense, 179.
croceum, 179.
herbaceum, 179.
nigrum, 179.
religiosum, 179.
Gouania cordifolia, 285.
denticulata, 282.
integrifolia, 284.
orbicularis, 284.
Richii, 282.
smilacina, 285.
vitifolia, 283.
Gourliea decorticans, 460.
Graffenrieda jucunda, 594.
Grewia Amicorum, 198.
glabra, 199.
leevigata, 199.

760

Grewia Mallococca, 197.
persiczefolia, 198.
prunifolia, 199.
Richardiana, 198.
tiliefolia, 198.

Griselinia lucida, 734.

littoralis, 734.

Grossulaceze, 662.

Guaiacum dubium, 256.

Guarea purgans, 2438.
tuberculata, 243.

Guatteria pallida, 27.

psilopus, 26.

Guilandina Bondue, 461.

Guioa lentiscifolia, 256.

Gunnera Chilensis, 629.

Falklandica, 631.
lobata, 631.
Magellanica, 631.
petaloidea, 629.
scabra, 629.

Guttiferee, 214.

Gymnophyton, 745.

Gynandropsis affinis, 65.

pentaphylla, 65.

Gynopleuru tubulosa, 635.
Gynotroches reticulata, 607.
Gypsophila australis, 112.

H.

Heemocharis semiserrata, 218.

Hallia cordata, 388.
imbricata, 888.
virgata, 388.

Halorages, 624.

Haloragis alata, 625.

Cercodea, 625.
depressa, 626.
elata, 625.
elata, 626.
filiformis, 628.
micrantha, 626.
pinnatifida, 627.
serra, 625.
tenella, 626.

INDEX.

Haloragis Tetragonia, 625.
tetragyna, 625.
teucroides, 625.

Haplopetalon Richii, 608.

Hardenbergia ovata, 438.

Harmogia densifolia, 571.

Harpeetropis speciosus, 446.

Hartighsea Billardieri, 239.

Forstert, 239, 240.
Fraseriana, 239.
spectabilis, 238.

Hartigia, 593.

Hecastophyllum Brownei, 454.

Hedera crassifolia, 718.

Gaudichaudi, 719.
Lessoni, 719.
platyphylla, 720.

Hedysarum australe, 431.

barbatum, 482.
capitatum, 433.
diphyllum, 423, 424.
heterocarpum, 432.
lagopodioides, 430.
lagopoides, 430.
umbellatum, 481.

Heisteria Raddiana, 305.

Heisteria pungens, 106.

Helicteres ovata, 184.

Helosciadium laciniatum, 708.

lateriflorum, 708.
leptophyllum, 707.
ranunculifolium, 708.

Hemigyrosa Perrottetii, 251.

Henriettea Brasiliensis, 585.
succosa, 585.

Heretiera Fomes, 184.
lanceolata, 184.
littoralis, 184.
minor, 184.

Hermas villosa, 713.

Herniaria illecebroides, 127.

Hesperomeles cuneata, 507.

Heteroloma, 433, 484.

Heteropterys chrysophylla, 266.

laurifolia, 266.
megaptera, 267.
metallochroa, 266.

Heteropterys nitida, 266.
sericea, 266.
venosa, 266.
Hibbertia camphorosma, 21.
canescens, 21,
dentata, 20.
diffusa, 20.
fasciculata, 21.
linearis, 20.
obtusifolia, 20.
volubilis, 20.
Mibiscus Abelmoschus, 172.
Adthiopicus, 173.
Arnottianus, 176.
bacciferus, 179.
Boryanus, 177.
Brackenridgei, 175.
bifurcatus, 174.
diversifolius, 174.
Jiculneus, 174.
hastatus, 178.
heterophyllus, 174.
physaloides, 173.
populneus, 179.

Pseudo-Abelmoschus, 172.

Rosa-Sinensis, 178.
scaber, 174.
Sturtii, 173.
tiliaceus, 178.
tricuspis, 178.
Youngianus, 174.
Hippocratea micrantha, 286.
Hippocrateaceze, 286.
Hiptage Javanica, 267.
myrtifolia, 267.
Hireea Gaudichaudiana, 270.
glabra, 269.
Hirtella oblongifolia, 492.
Hoffmanseggia viscosa, 462.
Hoheria angustifolia, 180.
Lyallii, 181.
populnea, 180.
Hosackia Macrazi, 400.
Hovea lanceolata, 386.
longifolia, 386.
Huberia ovalifolia, 596.
resinosa, 096.

INDEX.

Hugueninia, 60.

Hydrangiee, 664.

Hydrocotyle Asiatica, 693.

batrachioides, 698.
Bonariensis, 692.
elongata, 693.
glechomoides, 686.
gummifera, 701.
hirta, 6938.
interrupta, 692.
laxiflora, 693.
leucocephala, 693.
microphylla, 694.
modesta, 693.
peduncularis, 694.
plebeia, 693.
quinqueloba, 692.
ranunculoides, 693.
vulgaris, 692.

Hypericacez, 219.

Hypericum Brasiliense, 220.
Guianense, 219.
glandulosum, 219.
gramineum, 220.
grandifolium, 219.
humifusum, 220.
involutum, 220.
Japonicum, 220.
perforatum, 220.
pusillum, 220.

Hypopterygium, 371, 372.

Icacineze, 299.

Tlex anomala, 296.
Canariensis, 294.
Mate, 294.
Paraguariensis, 294.
Paraguayensis, 294.
Vitiensis, 295.

Illecebreze, 121.

Indigofera adesmizefolia, 405.

Anil, 403.
australis, 405.
coriacea, 406.

761

762

Indigofera filiformis, 406.
glutinosa, 405.
linearis, 405.

Obrajillensis, 404.

tephrosioides, 404.
tinctoria, 403.
Truxillensis, 403.
viscosa, 405.
Inga affinis, 487.
Cumingiana, 487.
dulcis, 487.
Feuillei, 486.
semialata, 486.

Velloziana, 487.

Intsia Amboinensis, 467.

Involucraria, 645.

Tonidium longifolium, 88.
monopetalum, 87.
orchidifiorum, 88.
setigerum, 746.

Isatis tinctoria, 64.

Isodendrion laurifolium, 96.

longifolium, 95.
pyrifolium, 93.

J.

Jacksonia scoparia, 382.
Jambosa australis, 514.
domestica, 510.
Malaccensis, 510.
myrtifolia, 514.
pterocaulis, 512.
Jamesia, 663.
Jossinia cotinifolia, 515.
Jucunda Lhotzkyana, 594.
Martiana, 594.
Juliania Huaucui, 371.
Jungia imbricata, 571.
Jussizea erecta, 618.
longifolia, 619.
octofila, 619.
octonervia, 619.
Peruviana, 618.
pilosa, 618.
repens, 618.

INDEX.

K.

Kageneckia crategoides, 493.
lanceolata, 493.
oblonga, 493.
Kalanchoe Brasiliensis, 690.
Kallstreemia, 325.
Kampitzia albens, 556.
Karivia Samoensis, 642.
Kennedya coccinea, 437.
monophylla, 438.
ovata, 438.
rubicunda, 437.
Kibessiec, 576.
Kleinhovia Hospita, 187.
Koniga maritima, 51.
Kotschya, 429.
Kuhlia Brasiliensis, 74, 746.
parviflora, 73.
Kunzea corifolia, 565.
Schaueri, 566.

L.

Lablab vulgaris, 453.
Laguncularia purpurea, 615.
racemosa, 615.
Lamanonia ternata, 681.
Laplacea semiserrata, 213.
Lardizabala biternuta, 41.
triternata, 41.
Lardizabalez, 41.
Laretia acaulis, 704.
Larrea divaricata, 326.
Larrea viscosa, 462.
Lasiopetalum ferrugineum, 187.
Lasiandra adenostemon, 603.
argentea, 603.
Gaudichaudiana, 603.
gracilis, 604.
Kunthiana, 604.
phalacrostemon, 603.
spoliata, 608.
Lathyrus epetiolaris, 421.
pubescens, 422.
sericeus, 422.

Lathyrus sessilifolius, 421.
tomentosus, 422.
Laurus caustica, 370.
Lawrencia, 180.
Leandra angustifolia, 585.
villosa, 585.
Leea Manillensis, 274.
sambucina, 274.
Staphylea, 274.
Leguminosz, 378.
Legnotidex, 607.

Leiospermum parviflorum, 670, 678.

Lepidium Bonariense, 62.
bipinnatifidum, 62.
didymum, 63.
graminifolium, 63.
Owahiense, 63.
piscidium, 63.
ruderale, 62.

Leptocyamus elongatus, 438.

microphyllus, 438.

Leptosperm um arachnoideum, 570.

collinum, 558.
ericoides, 569.
flavescens, 569.
juniperinum, 570.
lanigerum, 569.
myrtifolium, 569.
perforatum, 558.
scandens, 557.
Scoparium, 569.

Lessertia monticola, 411.

Leucena Forstert, 477.

Leucolena pannosa, 696.

Lichtensteinia lacera, 709.

pyrethrifolia, 709.

Ligusticum Pansil, 709.

Lilenia dentata, 75.

Limacia yelutina, 40.

Limonia minuta, 235.

Linacez, 319.

Lindackeria, 72.

Linum Africanum, 320.

aquilinum, 319.
Macreei, 319.
marginale, 320.
monogynum, 319.

INDEX. 763

Linum oligophyllum, 319.
polygaloides, 319.
Lithreea caustica, 370.
venenosa, 370.
Loasa Aldunata, 634.
atriplicifolia, 633.
contorta, 633.
coronata, 684.
grandiflora, 633.
heterophylla, 684.
incana, 633.
' pinnatifida, 634.

Loasaceze, 633.

Lobelia bryoides, 117.

Lobularia maritima, 51.

Loranthacesxe, 735.

Loranthus Americanus, 736.
aphyllus, 737.
buxifolius, 736.
Cactorum, 737.
citricola, 736.
celastroides, 740.
Cunninghamii, 741.
dichroos, 735.
Eschscholtzianus, 736.
eucalyptoides, 741.
Forsterianus, 737.
heterophyllus, 736.
insularum, 738.
Jacquini, 736.
marginatus, 736.
maytenifolius, 739.
Philippensis, 741.
pendulus, 741.
stelis, 737.
tetrandrus, 736.
tetrapetalus, 737.

Lotus angustissimus, 399.

arboreus, 411.
australis, 400.
Candolleanus, 400.
corniculatus, 400.
divaricatus, 399.
Fluminensis, 487.
Fraseri, 400.
Macrzxi, 401.
maritimus, 450.

764

Lotus uliginosus, 400.
Ludwigia jussizoides, 619.
Luffa insularum, 644.
Luma Cheken, 536.
correvefolia, 542..
Cruckschanksii, 540.
ferruginea, 542.
obtusa, 541.
stenophylla, 540.
Temu, 539.
Lumnitzera coccinea, 615.
purpurea, 615. -
Lupinus alopecuroides, 391.
Cruckshanksii, 393.
Lindleyanus, 391.
microcarpus, 391.
nubigenus, 391.
paniculatus, 392.
Pickeringii, 392.
prostratus, 392.
Termis, 393.
tomentosus, 392.
Lychnis thysanodes, 114.
Chilensis, 115.
Lythrum Hyssopifolia, 606.
maritimum, 606.
Pemphis, 605.
Salicaria, 605.

M.

Macherium acutifolium, 455.

angustifolium, 454.
incorruptibile, 455.

muticum, 499.

oblongifolium, 455.

Macrolobium bijugum, 467.

Macroptilium, 450.

Mamillaria flavescens, 659.
prolifera, 659.

Magnoliaces, 24.

Maiten, 290.

Malacmea fluminensis, 263.

Malesherbia thyrsiflora, 635.

Malesherbiaceze, 635.

Mallea Rothii, 244.

INDEX.

Mallococca crenata, 197.

Malochia, 441.

Malpighia Armeniaca, 263.

Martiana, 263.

Malpighiacese, 263.

Malwa acaulis, 150.
Americana, 148.
Belloa, 150.
Coromandeliana, 148.
Domingensis, 148.
humilis, 150.
leprosa, 157.

Trimensis, 146.
Lindheimeriana, 148.
obtusiloba, 171.
ovata, 147.
Peruviana, 146.
plumosa, 147.
scabra, 147.
scoparia, 147.
subhastata, 148.
sulphurea, 150, 157.
Timorensis, 147.
tricuspidata, 148.

Malvaceze, 146.

Malvastrum acaule, 150.
anthemidifolium, 152.
aretioides, 153.
carpinifolium, 148.
Cavanillesii, 156.
coccineum, 150.
compactum, 152.
humile, 150.
Macleani, 152.
parnassizefolium, 150.
pedicularizefolium, 152.
Peruvianum, 146.
Phyllanthos, 152.
Pichinchense, 152.
pinnatum, 154,
plumosum, 147.
Purdizei, 150.
Richii, 149.
scabrum, 147.
scoparium, 147.
spicatum, 147.
stenopetalum, 155.

Malvastrum tricuspidatum, 148.

ulophyllum, 150.
Mangifera Indica, 364.
Mangium celsum, 614.
Margyricarpus alatus, 495.
setosus, 495.
Marlierea excorticata, 535.
suaveolens, 534.
Matthiola Maderensis, 46.
Mauria multiflora, 371.
Maytenus Brasiliensis, 290.
Chilensis, 290.
obtusifolius, 290.
Mayna Brasiliensis, 72.
Mazeutoxeron reflecum, 329.
Medicago denticulata, 397.
lupulina, 397.
sativa, 397.
Medinilla heterophylla, 598.
rhodochlzena, 600.
Melaleuca eriocephala, 566.
erubescens, 567.
ferrea, 566.
florida, 557.
Fraseri, 567.
genistifolia, 567.
linariifolia, 567.
lucida, 557.
myrtifolia, 568.
nodosa, 567.
perforata, 558.

INDEX.

Melastomex, 576.
Melhania Leprieurii, 194.
Meliaceze, 236.
Melicope barbigera, 351.
cinerea, 350.
elliptica, 353.
grandifolia, 354.
simplex, 345.
spathulata, 352.
ternata, 346.
Melicytus macrophyllus, 97.
ramiflorus, 97.
Melilotus parviflora, 398.
Melistaurum distichum, 81.
Melocactus, 661.

Melochia aristata, 193.

corchorifolia, 191.
concatenata, 191.
odorata, 192.
Vitiensis, 193.
Melothria Fluminensis, 641.
Samoensis, 641.
Memecyleze, 573.
Menecylon Calderense, 574.
Vitiense, 573.
Menispermacee, 36.
Menispermum Japonicum, 36.
peltatum, 36.
Mentzelia aspera, 633.
scabra, 633.
Mesembryanthemez, 145.

352.

765

‘squarrosa, 568.
styphelioides, 566.
thymifolia, 567.
villosa, 558.
Melandrium apetalum, 115.
Magellanicum, 115.
thysanodes, 114.
Melastoma adpressum, 602.
cereacea, 581.
longifolium, 602.
Malabathrica, 601.
polyanthum, 602.
staminea, 594,
Taitense, 601.
Vitiense, 601.
Melastomacez, 573.

Mesembryanthemum australe, 145.
glaucum, 145.
nodiflorum, 145.
tenuifolium, 145.
Metrosideros Amboinensis, 467.
buxtfolia, 558.
capitata, 566.
collina, 558.
corifolia, 565.
diffusa, 558.
florida, 557.
floribunda, 518.
glomerulifolia, 556.
hypericifolia, 558.
lucida, 557.
lutea, 560.
192

766

Metrosideros Macropus, 564.
perforata, 558.
polymorpha, 562.
robusta, 557.
rugosa, 561.
scandens, 558.
spectabilis, 558.
tomentosa, 557.
umbellata, 557.
villosa, 558.

Miconia affinis, 595,

attenuata, 595.
brunnea, 594.
cheenopleuroides, 596.
cinerascens, 594,
depauperata, 595.
divaricata, 594.
fasciculata, 595.
latecrenata, 596.
paludosa, 595.
prasina, 595.
Sellowiana, 595.
staminea, 594.
seplaria, 595.

Microjambosa, 509, 520, 528, 530.

Microlotus, 400.
Micromelum glabrescens, 235.
pubescens, 235.

Micropetalon lanuginosum, 115.

Micropteryx, 446.

Milletia Luzonensis, 456.

Milnea edulis, 237,

Mimosa albida, 476.
balsamica, 426.
ephedroides, 477.
elliptica, 476.
Gilliesii, 477.
glandulosa, 477.
juliflora, 474.
Mangium, 482.
pudica, 476.
scandens, 4738.
sepiaria, 476.
semilata, 486.
Velloziana, 476.
viva, 476.

Mimosez, 473.

INDEX.

Mirbelia reticulata, 385.
Misandra Magellanica, 631.
Misodendron brachystachium, 745.
oblongifolium, 745,
punctulatum, 745.
Mischocarpus, 252.
Mniarum biflorum, 129.
pedunculatum, 130.
Modiola reptans, 171.
Mollia tenuifolia, 122.
Mollugo bellidifolia, 130.
Berteriana, 180.
dichotoma, 130.
nudicaulis, 130.
Schrankii, 130.
stricta, 130.
triphylla, 130.
verticillata, 130.
Momordica Charantia, 646.
Senegalensis, 646,
Monnina angustifolia, 107.
linearifolia, 107.
macrostachya, 107.
Monocera Griffithii, 202.
petiolata, 202.
paniculata, 202.
Monoxora rubescens, 547.
spectabilis, 546.
Montia fontana, 144.
Montinia acris, 621.
Moschoxylum elegans, 242.
Richardianum, 242.
Moulinsia cupanioides, 250.
rubiginosa, 250.
Mucuna altissima, 448, 446.
gigantea, 442.
platyphylla, 443.
urens, 4438.
Mulinum cuneatum, 7038.
Echinus, 703.
proliferum, 703.
spinosum, 703.
ulicinum, 703.
Muraltia diffusa, 106.
Heisteria, 106.
mixta, 106.
Murrya exotica, 235.

Myrcia Browniana, 533.
Kriopus, 533.
elliptica, 532.
ellipticifolia, 532.
exsucca, 531.
ferruginea, 532.
prunifolia, 532.
Pseudo-Mini, 532.
pubescens, 531.
ramulosa, 532.
rostrata, 534.
sepiaria, 534.
spectabilis, 533.
spheerocarpa, 532.
subsericea, 533.

Myriophyllum proserpinacoides, 628.

verrucosum, 629.

Myristica castanezefolia, 32.

cinerea, 35.
hypargyrea, 33.
inutilis, 34.
macrophylla, 83.
Myristicacese, 32.
Myroxylon orbiculatum, 78.
suaveolens, 76.

Myrrhis andicola, 712.

Myrtaceze, 508.

Myrtus bullata, 543.

Candollet, 545.

Cheken, 537.

communis, 543.

Coquimbensis, 536.

Fernandesiana, 586.

Gudilla, 540.

Luma, 537, 539.

Moline, 537.

multiflora, 536.

nummularia, 544.

pedunculata, 544.

Raran, 541.

rufa, 542.

smilactfolia, 546.

tenuifolia, 545.

tomentosa, 546.

trinervia, 547.

Ugni, 544. )
Myzodendron brachystachium, 745.

INDEX. 767

Myzodendron oblongifolium, 745.
punctulatum, 745.

N.

Nasturtium officinale, 46.
palustre, 47.
sylvestre, 47.
terrestre, 47.
Nelitris fruticosa, 547.
Jambosella, 547.
Vitiensis, 548.
Nephelium pinnatum, 259.
Nephroica Ferrandiana, 39.
Neptunia gracilis, 475.

'. pubescens, 475.
Neurocarpum cajanifolium, 437.
Neurophyllodes, 310.
Nicolsonia barbata, 482.

capitata, 438.
Cayennensis, 432.
reptans, 452.
venustula, 432.
Nissolia incorruptible, 455.
Noisettia longifolia, 88.
orchidiflora, 88.
pyrifolia, 88.

O.

Ochna Jabotapita, 360.
stipulata, 360.
Ochnacex, 359.
(Anothera acaulis, 620.
albicans, 620.
grandiflora, 620.
mollissima, 619.
multicaulis, 620.
nocturna, 620.
odorata, 620.
prostrata, 620.
stricta, 620.
tenuifolia, 620.
Olacacese, 299.
Olacinez, 304.

768

Olax imbricata, 305.

Onagracex, 618.

Oncocarpus Vitiensis, 365.

Ononis capillaris, 395.

Opuntia, 661.

Oreomyrrhis andicola, 712.

lasiopetala, 712.

Ormocarpum sennoides, 422.

Ornitrophe glabra, 250.

Osmorhiza Berteri, 713.

Chilensis, 713.

Osteomeles anthyllidifolia, 507.

Otolepis nigrescens, 259.

Otophora Blancoi, 259.

Outea bijuga, 467.

Oxalis Barrelieri, 321.
bipunctata, 321.
cataracte, 324.
ciliifera, 520.
corniculata, 320.
crassifolia, 320.
elegans, 322.
exilis, 8320.
filicaulis, 325.
floribunda, 321.
fruticosa, 321.
geminata, 322.
hedysaroides, 321.
hirta, 324,
lactea, 324.
lasiopetala, 321.
lineata, 322.
lobata, 322.
Magellanica, 324.
Martiana, 321.
microphylla, 320.
nubigena, 323.
parvula, 323.
Pickeringii, 322
polyantha, 322.
purpurea, 325.
pygmea, 322.
reptans, 320.
urbica, 321.
versicolor, 325.

Oxalidaceze, 320.

Oxymeris ciliata, 592.

INDEX,

Oxymeris pulchra, 591.
Oxylobium hamulosum, 379.
Pultenee, 378.

P.

Palava moschata, 146.
Palavia moschata, 146.
Panax arboreum, 717.
Gaudichaudi, 719.
fruticosum, 716.
Lessoni, 719.
platyphyllum, 720.
sambucifolium, 717.
Samoense, 717..,
simplex, 718.
ovatum, 719.
Panke, 629.
Pangiaces, 72.
Papaveracexe, 44.
Papayacez, 640.
Papilionaceze, 378.
Paratropia multijuga, 722.
Samoensis, 722.
Paritium tiliaceum, 178.
tricuspis, 178.
Parinarium insularum, 488.
laurinum, 490.
Margarata, 489.

Sumatranum, 489.
Griffithianum, 491.

Jackianum, 491.
Parivoa, 469..
Paronychia Andina, 128.
Bonariensis, 127.
Brasiliana, 127.
Chilensis, 127.
communis, 127.
illecebroides, 127.
ramosissima, 129.
Passiflora albida, 637.
bilobata, 638.
elliptica, 638.
footida, 638.
littoralis, 637.
hibiscifolia, 638.
peduncularis, 639.

Passiflora piligera, 638.
Herbertiana, 688.
tetrandra, 637.
trifoliata, 639.

Passifloracese, 637.

Pastinaca(?), 714.

Paullinia falcata, 248.
discolor, 248.
Guarumima, 247.
multiflora, 248.
rubiginosa, 248.

wienmanniefolia, 248.

Barbadensis, 249,
Payonia communis, 171.
hastata, 171.
sepium, 171.
Pectophytum, 697, 702.
Peixotoa glabra, 265.

Pelargonium Acugnaticum, 817.

anceps, 317.
australe, 317.

clandestinum, 317.

cucullatum, 317.

myrrhifolium, 318.

Pelea auriculefolia, 3438.
clusizfolia, 340.
lucida, 348.
oblongifolia, 343.
rotundifolia, 344.
Sandwicensis, 345.
volcanica, 346.

Pemphis acidula, 605.

Pentacena polycnemoides, 129.

ramosissima, 129.

Pentaloba, 92.

Pereskia, 661.

Perrottetia Sandwicensis, 291, 746.

Quinduensis, 292.
Petrocarya Sumatrana, 489.
excelsa, 491.
Petroselinum peregrinum, 707.
prostratum, TOT.
Peucedanum capillaceum, 710.
Phaca canescens, 413.
ochroleuca, 418.
Richii, 414.
Phaseolus amenus, 449.

INDEX.

193

Phaseolus Cumingii, 449.
maritimus, 450.
Mungo, 449.
rostratus, 449.
Schottii, 449.
Truxillensis, 449.
vestitus, 450.

Philadelphus, 663.

Philotheca australis, 329.

Reichenbachiana, 329.

Phoradendron chrysocladon, 743.

pepericarpum, 742.

Schottii, 742.
Phylacium bracteosum, 423.
Phylica stipularis, 285.
bicolor, 285.
Phyllota aspera, 383.

Baueri, 383.
Phyllodium pulchellum, 431.
Pigea filiformis, 87.

monopetala, 87.
Piliostigma acidum, 470.
Pilocarpus spicata, 331.

Pimela Luzonica, 374.

Piptadenia communis, 473.

macrocarpa, 474.
Pitavia, 340.
Pithecolobium Clypearia, 485.
dulce, 485.

Pittosporacese, 221.

Pittosporum acacioides, 222.
arborescens, 223.
Brackenridgei, 2265.
confertiflorum, 232.
cornifolium, 223.
crassifolium, 223.
eugenioides, 223.
glabrum, 229.
microcarpum, 223.
Pickeringii, 227.
pimelioides, 223.
Richii, 224.
rhytidocarpum, 228.
tenuifolium, 222.
terminalioides, 231.
tobiroides, 226.
umbellatum, 223.

769

770

Pittosporum undulatum, 222.
Pitumba, 79.
Plagianthez, 180.
Plagianthus betulinus, 182.
divaricatus, 181.
sidoides, 181.
urticinus, 182.
Platylobium formosum, 386.
parviflorum, 386.
Pleiochiton crassifolia, 583.
Plerandra Pickeringii, 729.
Pleroma adenostemon, 603.
argenteum, 603.
Fontanesianum, 603.

Gaudichaudianum, 6038.

gracile, 604.
Kunthianum, 604.
spoliatum, 603.
Pleurandra acicularis, 23.
Camphorosma, 22.
parviflora, 23.
scabra, 22.
stricta, 23.
Pleuropetalon Samoense, 299.
Pleuropetalum, 302.
Ploiarium elegans, 218.
Podalyria biflora, 378.
subbiflora, 378.
Podocarpium, 436.
Podolobium obovatum, 379.
trilobatum, 379.
scandens, 380.
Pecilodermis populnea, 185.
Poinciana, 461.
Polanisia icosandra, 68.
viscosa, 68.
Polycarpzea tenuifolia, 122.
glabrifolia, 122.
Polycarpon Magellanicum, 688.
Polycarpon tetraphyllum, 122.
Polygala cestrifolia, 104.
cyparissias, 104.
erloptera, 105.
Jrutescens, 105.
Garcini, 105.
Heisteria, 106.
lancifolia, 104.

INDEX.

Polygala paniculata, 104.
stenophylla, 104.
Polygalaceze, 103.
Polystemon pentaphyllus, 681.
Pomaderris discolor, 281.
ericeefolia, 281.
lanigera, 281.
multiflora, 281.
ztzyphoides, 278.
Pometia pinnata, 259.
Pomum-draconis sylvestre, 374.
Pongamia glabra, 455.
religiosa, 457.
triphylla, 457.
uliginosa, 457.
Portulaca hirsutissima, 140.
lutea, 189.
oleracea, 139.
pilosa, 140.
quadrifida, 140.
sclerocarpa, 141.
villosa, 140.
Portulacacez, 139.
Potentilla procumbens, 500.
Tormentilla, 500.
Poupartia dulcis, 374.
Poza Ranunculus, 698.
Priestleya sericea, 387.
Prinos, 295.
Prockia crucis, 73.
completa, 73.
Prockiaria, 73.
Prosopis affints, 474.
dulcis, 474.
juliflora, 474.
strombulifera, 475.
globosa, 477.
Pseudarthia capitata, 483.
Psidium cerasoides, 549.
Decaspermum, 547.
Guajava, 550.
polycarpon, 550.
pomiferum, 550.
pyriferum, 550.
Psittacanthus dichroos, 735.
Jacquint, 736.
Psoralea Americana, 401.

Psoralea aphylla, 402.
bituminosa, 402.
dentata, 401.
decumbens, 402.
glandulosa, 403.
lasiostachys, 403.
patens, 402.
- pinnata, 402.
tenax, 402.
Psychrophila appendiculata, 18.
andicola, 12.
dionezefolia, 14.
sagittata, 12.
Pterocarpus frutescens, 458.

Pterospermum diversifolium, 194.

Pukateria littoralis, 734.
Pultencea aspera, 383.
biloba, 385.
daphnioides, 384.
elliptica, 385.
euchila, 384.
linophylla, 384.
polifolia, 385.
retusa, 385.
stipularis, 385.
villosa, 385.
Putterlickia pyracantha, 287.
Pyrranthus littoreus, 615.

Q.

Quillaia Moline, 493.
saponaria, 493.
Smegmadermos, 493.

Quintinia serrata, 666.

R.

Rafnia axillaris, 387.

Ranunculus acaulis, 7.
Aucklandicus, 8.
biternatus, 6.
Bonplandianus, 6.
Chilensis, 6.
exiguus, 6.
Hawaiensis, 10.
hirtus, 7.

INDEX.

Ranunculus inundatus, 9.
lappaceus, 9.
Mauiensis, 11.
plebeius, 8.
preemorsus, 5.
rivularis, 7.
Ranunculaces, 1.
Raphanus Raphanistrum, 64.
Reissekia cordifolia, 285.
Retanilla Ephedra, 276.
trinervia, 276.
Reynoldsia pleiosperma, 724.
Sandwicensis, 723.
Rhadinocarpus, 423.
Rhamnacex, 274.
Rhamnuus zizyphoides, 278.
incana, 278.
Rhexia Fontanesii, 603.
gracilis, 604.
Langsdorfiana, 603.
Leuzeana, 596.
nympheifolia, 596.
Rhizophora Candelaria, 613.
Mangle, 613.
mucronata, 613.
Rhizophoraceze, 607.
Rhizophoree, 613.
Rhodamnia cinerea, 546.
trinervia, 547.
Rhodomyrtus tomentosa, 546.
Rhus atrum, 367.
caustica, 370.
lucida, 370.
Sandwicensis, 369.
simarubeefolia, 367.
Taitensis, 368.
Rhynchanthera dichotoma, 604.
pentanthera, 604.

Schrankiana, 604.

Rhynchosia minima, 454.
Rhytidandra Vitiensis, 303.
Ribes albiflorum, 662.
glandulosum, 662.
Magellanicum, 662.
punctatum, 662.
Richella monosperma, 28.
Riedleia concatenata, 191.

771

772

Riedleia corchorifolia, 191.
odorata, 192.
Rochea odoratissima, 690.
Rollinia longifolia, 26.
parviflora, 26.
Rosaceze, 493.
Rourea heterophylla, 875.
Rubus alcecefolius, 5038.
australis, 502.
Bergii, 503.
cissoides, 502.
Kglanteria, 502.
fruticosus, 502.
geoides, 503.
Hamitltonianus, 503.
Hawaiensis, 504.
heterophyllus, 5038.
Macrei, 505.
macropodus, 502.
Molluccanus, 503.
reflecus, 503.
roszefolius, 501.
rugosus, 503.
schmidelioides, 502.
tiliaceus, 503.
urticeefolius, 503.
Rulingia pannosa, 188.
Ruta bracteosa, 327.
Rutaceze, 327.

S.

Sagotia triflora, 432.

Sagina crassifolia, 181.
subulata, 182.

Salacia macrophylla, 286.
undulata, 286.

Salomonia Cantoniensis, 105.

Samyda polyandra, 81.
Selloi, 79.

Samydaceze, 79.

Sanicula Sandwicensis, 705.

Sapindacese, 247,

Sapindus baccatus, 259.
cinereus, 258.

Jraxinifolius, 250.

INDEX.

Sapindus rubiginosus, 250.
Saponaria, 252.
Vitiensis, 251.

Sarothamus scoparius, 396.

Saurauja, 208.

Sauvagesia erecta, 97.

Saxifraga bicuspidata, 687.
Bonplandit, 687.
exarata, 687.
Magellanica, 687.
Peruviana, 687.
stellata, 687.

Saxifragacess, 663.

Saxifragec, 664.

Scandix glochidiata, 711.

Schefilera digitata, 715.

Schidiomyrtus crenulata, 571.

diosmeefolia, 570.

Schiedea diffusa, 138, 138.
Hookeri, 133.
ligustrina, 133.
Menziesii, 133.
Nuttallii, 133, 137.
spergulina, 133, 135.

Schinus dentatus, 370.

terebinthifolius, 371.
Molle, 371.
Schizeilema, 698.

Schizophragma, 664.

Schmidelia Cochinchinensis, 249.
glabra, 249.
leevis, 249.
obovata, 249.
racemosa, 249.
serrata, 249.
Schnella macrostachya, 470.
splendens, 470.
Sciodaphyllum, 7238.
Sclerantheze, 129.
Sclerostylis atalantoides, 233.
Scopolia lucida, 734.
Scutula, 575.
Scytalis anomala, 452.
Securidaca ovalifolia, 108.
Sedum nudum, 691.
Peruvianum, 691.
Selenolobium, 458.

Selinum acaule, 704.
proliferum, 708.
spinosum, 708.

Senebiera didyma, 63.

pectinata, 63.

pinnatifida, 63.
Serianthes myriadenia, 485.

Vitiensis, 485.

Seringia platyphylla, 187.

Serjania communis, 247.
cuspidata, 247.

Sesbania tomentosa, 409.

Sesuvium Portulacastrum, 142.

parviflorum, 142.
Sicyocarya, 649.
Sicyopsis, 649.

Sicyos angulatus, 648.
australis, 648.
cucumerinus, 652.
Fretensis, 648.
levis, 648.
macrophyllus, 651.
montanus, 649.
pachycarpus, 650.

Sida acaulis, 156.

acuta, 159.
Africana, 160.
alba, 158.
althxifolia, 160.
anthemidifolia, 152.
atrosanguinea, 160.
Borussica, 152.
bracteoluta, 159.
Canariensis, 158.
canescens, 158.
carpinifolia, 159.
carpinoides, 148.
compacta, 152.
cordifolia, 160.
corrugata, 169.
depressa, 147.
Diellii, 162.
Dombeyana, 160.
fallax, 161.
fibulifera, 165.
floribunda, 159.
glauca, 167.

INDEX.

Sida herbacea, 160.

Hondensis, 158.
humilis, 160.
incana, 168.
Javensis, 160.
linifolia, 157.
maura, 162.
Meyeniana, 164.
microphylla, 158.
multiflora, 160.
paniculata, 159.

parnassizfolia, 150.
pedicularizxfolia, 152.

Philippica, 158.
Phyllanthos, 152.
Pichichensis, 152.
pinnata, 154.
planicaulis, 159.
pulchella, 181.
pygmeza, 150.
retusa, 158.
rhombifolia, 158.
rhomboidea, 158.

rotundifolia, 160, 161, 163.

salicifolia, 158.
Sertum, 163.
spireifolia, 159.
spinosa, 157.
sulphurea, 157.
Tasmanica, 181.
ulmifolia, 164.

venusta, 160.

veroniceefolia, 160.
Sieversia albiflora, 501.

Silene Anglica, 109.

Antirrhina, 112.

cerastoides, 109.
Gallica, 109.
inflata, 112.
lanceolata, 111.

quinquevulnera, 112.
struthioloides, 109.
thysanodes, 114.

Silenez, 109.
Simarubaceex, 355.

Sinapidendron frutescens, 64.

Sinapis frutescens, 64.

194

773

174

Sisymbrium amplexicaule, 61.

__ athrocarpum, 59.
canescens, 59.
gerantifolium, 48.
glaciale, 49.
leptocarpum, 58.
myriophyllum, 59.

pimpinellifolium, 59.

Titicacense, 59.
spathuleefolium, 60.
Sizygium, 509, 518.
Sizygium paniculatum, 525.
zeylanicum, 520.
Skinnera excorticata, 624.
Smithia spicata, 429.
Solea monopetala, 87.
Sonneratia acida, 550.
Sophora macrocarpa, 460.
microphylla, 458.
tetraptera, 458.
tomentosa, 460.
Spadostylis Sieberi, 384.
Spartium junceum, 395.
scopartum, 396.
virgatum, 396.
Spatholobus, 445.
Spathularia longifolia, 88.
Spennera paludosa, 601.
Spergula arvensis, 121.
Spergularia ramosa, 121.
rubra, 121.

Spergulastrum lanuginosum, 115.

Spheeralcea obtusiloba, 171.
Spirzeanthemum Samoense, 667.
Vitiense, 669.
Spondias Cytherea, 374.
dulcis, 374.
Stackhousia linearifolia, 807.
muricata; 306.
spathulata, 306.
Stackhousiacez, 306.
Staphidium biserratum, 598.
pauciflorum, 593.
Stellaria ceespitosa, 119.
cuspidata, 118.
debilis, 118.
elongata, 115.

INDEX,

Stellaria lanuginosa, 115.
leptopetala, 119.
pungens, 119.
squarrosa, 119.

Stephania australis, 38.
Forsteri, 36.
Gaudichaudii, 37.
glaucescens, 37.

Sterculia Ceramica, 184.

Sterculiaceze, 182.

Stereoxylon pulverulentum, 665.

revolutum, 664.
rubrum, 664.
Stigmaphyllon ciliatum, 265.
Gaudichaudinum, 265. |
tomentosum, 265.
vitifolium, 264.

Stilbocarpa polaris, 714.

Streptodesmia canescens, 427.

Strombocarpa cinerascens, 475.

odorata, 475.
pubescens, 475.
reptans, 475.
strombulifera, 475.
Strongylocalyx, 509.
Strongylodon macrobotrys, 448.
ruber, 445.

Strophostyles, 449.

Struthanthus Eschscholtzianus, 736.

Stylosanthes Guianensis, 424.

viscosa, 424.

Suriana maritima, 307.

Surianiaceze, 307.

Swainsona coronilleefolia, 410.
galegifolia, 410.
microphylla, 410.
monticola, £11.

Swartzia elegans, 467.
grandiflora, 467.

Syllysium, 509.

Syncarpia laurifolia, 556.

I.

Tacsonia gracilens, 639.
peduncularis, 639.

Tacsonia trifoliata, 639.
Talinum patens, 142.
Teesdalia Lepidium, 55.
Teline Maderensis, 396.
Tephrosia Baueri, 408.
bracteolata, 407.
Capensis, 408.
hirta, 407.
piscatoria, 407.
Terminalia Catappa, 615.
glabrata, 616.
Moluccana, 616.
Richii, 616.
Ternstreemia Brasiliensis, 209.
Ternstroemiaceze, 206.
Tetraglochin strictum, 495.
Tetragonia expansa, 144.
Tetrameris Martiana, 601.
Tetrapathzxa australis, 637.
Tetraphyle muscosa, 689.

Tetraplasandra Hawaiensis, 728.
Tetrapterys Guilleminiana, 270.

mogorizfolia, 269.
punctulata, 269.
rotundifolia, 269.
Tetratheca affinis, 102.
ericifolia, 102.
juncea, 101.
setigera, 102.
thymifolia, 102.
Thespesia populnea, 179.
Thlaspi Bonariense, 62.
multifidum, 62.

Thomasia quercifolia, 187.

Thryallis rotundifolia, 264.

Tiliaceze, 195. -

Tillzea connata, 688.
moschata, 689.
rubescens, 689.
verticillaris, 638.
muscosa, 688.

Tontelea undulata, 286.

Torilis nodosa, 712.

Tovaria pendula, 65.

Trachymene ericoides, 695.

lanceolata, 694.
linearis, 695.

INDEX.

Trachymene myrtifolia, 694.
ovalis, 694.
subvelutina, 695.
tenuis, 695.
Tremandra, 102.
Tremandracez, 101.
Trevoa trinervia, 276.
Tribulus cistoides, 325.
Trichilia alliacea, 240.
monophylla, 222.
spectabilis, 238.
Trichilia elegans, 242.
Richardiana, 242.

Trichocephalus stipularis, 285.
Trichosanthes quinquangulata, 645.

Trifolium filiforme, 398.
fragiferum, 398.
Matthewsii, 398.
obcordatum, 399.
Peruvianum, 398.
polymorphum, 399.

Trigonia crotonoides, 108.

nivea, 108.

Trigoniaceze, 108.

Triphasia monophylla, 234.
trifoliata, 234.

Triptolemea montana, 458.

myriantha, 458.

Tristania albens, 556.
laurina, 568.
neriifolia, 568.

Tristerix aphyllus, 737.
tetrandrus, 736.

Triumfetta annua, 197.

eriocarpa, 196.
Fabreana, 197.
Indica, 197.
obscura, 196.
procumbens, 197.
rhomboidea, 196.
trilocularis, 197.
Tropxolacex, 318.
Tropeolum majus, 318.
tricolor, 318.
tricolorum, 318.
tuberosum, 318.
Tupeia antarctica, 742,

775

776

Tupeia Cunninghamii, 742.

pubigera, 742.
Turnera cuneiformis, 636.
salicifolia, 636.
ulmifolia, 636.
Turneraceze, 636.

U.
Ugni, 544.
Ulex Europzus, 395.
Umbelliferze, 692.

Umbilicus pendulinus, 691.

Unona leptopetala, 30.
ligularis, 30.
odorata, 30.

Uraria lagopoides, 430.
linearis, 480.
picta, 430.

Urena lobata, 169.
morifolia, 170.
monopetala, 169.

scabriuscula, 169.

Uvaria amygdalina, 31.

odorata, 30.

V.

Vavea Amicorum, 244.

Ventilago Vitiensis, 274.

Vepris simplicifolia, 336.

Vicia Andicola, 420.

bidentata, 419.
depauperata, 419.
grandiflora, 420.
Matthewsii, 419.
Menziesii, 420.
nigricans, 419.
Selloi, 419.

Vigna anomala, 452.
lutea, 452.
Oahuensis, 450.
Sandwicensis, 451.
Sinensis, 452.
villosa, 450.

Villaresia mucronata, 297.

Viminaria denudata, 382.

INDEX.

Viola Chamissonana, 86.
Commersoniana, 83.
hederacea, 84.
Kauensis, 85.
Magellanica, 84.
longifolia, 88.
tracheliifolia, 86.
tridentata, 83.
orchidiflora, 88.
Siebert, 84.
Sieberiana, 84.
spathulata, 84.
summa, 88.

Violacezx, 83.

Virgilia Capensis, 460.

Viscum antarcticum, 742.

articulatum, 754.
compressum, 754.
moniliforme, 754.
opuntioides, 754.
platycaulon, 754.
Schottii, 742.

Visenia, 193.

Vismia G'utanensis, 219.

Hilarii, 219.

Vitaceze, 271.

Viviania marifolia, 307.

Vivianaceee, 307.

Vochysia emarginata, 617.

tucanorum, 617.

Vochysiaceze, 617.

Vouapa, 469.

W.

Waltheria Americana, 189.
Indica, 190.
Lophanthus, 191.
ovata, 190.
pyrolzefolia, 190.
reticulata, 190.

Walpersia capitata, 285.
stipularis, 285."

Webbia floribunda, 219.

Weinmannia affinis, 674.

Arnoldia, 676.
betulina, 671.

Weinmannia Blume, 676.

fuschioides, 671.

parviflora, 6738.
pinnata, 677.
Richii, 675.
roseefolia, 671.

Samoensis, 677.
spirseoides, 677.

sylvicola, 671.
Wintera aromatica, 24.
~ Winterana aromatica, 24.
Wormia apetala, 18.
Wrenciala, 180.

X.

Xanthosia hirsuta, 695.
montana, 696.
pilosa, 695.
tridentata, 696.

Ximenia Americana, 304.

May 20, 1854.

INDEX.

| Ximenia elliptica, 305.
Xylocarpus Granatum, 243.
obovatus, 243.
AXylophylla longifolia, 718.
Xylosma orbiculatum, 78.
suaveolens, 76.

Zanthoxylee, 327.
Zanthoxylum, 340.

Kauaense, 354.

varians, 382.
Zichya, 437.
Zieria lanceolata, 331.
pauciflora, 331.
Zizyphus Jujuba, 275.
Zornia angustifolia, 423.
reticulata, 424.
Zygophyllaceze, 325.

195

~I

“I

iT