The Journal of the Linnean Society.

THE
JOURNAL

OF

THE LINNEAN SOCIETY

BOTANY.

VOL XXI W

LON DON:
SOLD AT THE SOCIETY’S APARTMENTS, BURLINGTON HOUSE,
AND BY
LONGMANS, GREEN, AND CO,
AND
WILLIAMS AND NORGATE.
1887,

Dates of Publication of the several Numbers included in this Volume.

No.141, pp. 1- 98, published December 24, 1885.

4 149, n 00-108, o April14, 1996.
» d » 169-212, — ,, April 30, 1886.
n 14

50145. 213-997; 5 July 23, 1886.

» 140, ,, 297-340, S October 15, 1886.
» 147, ,, 841-401, X October 27, 1886.
» 148, ,, 401-472, 2 March 23, 1887.
» 149, 473-560, a June 30, 1887.

PRINTED BY TAYLOR AND FRANCIS,
RED LION COURT, FLEET STREET.

LIST OF PAPERS.

Í

DBAKZERB, J. G., F.R.S., F.L.S.
: On a Collection of Ferns made in North Borneo by the Bishop

Page

| of Singapore and Sarawak. (Plates XI. & XIL) .......... 222

| Further Contributions to the Flora of Madagascar .......... 44]

| Bar, Jonn, F.R.S., M.R.LA., F.L.S.

: Contributions to the Flora of the Peruvian Andes, with Remarks

j on the History and Origin of the Andean Flora............ 1
Notes on the Botany of Western South America ............ 137

Borus, Harry, F.L.S., and N. E. Brown, A.LS.
Contributions to South-African Botany.—Orchidee. Part II.
With Additional Notes by N. E. Brown, A.L.S. (PlateL).. 65

Boxavra, E., M.D., Brigade-Surgeon, Indian Med. Depart.

On the probable Wild Source of the whole Group of Cultivated
True Limes (Crus acida, Roxb., C. medica var. acida of
Brandis, Hooker, and Alph. de Candolle). (Communicated by
W. T. Thiselton Dyer, C.M.G., F.R.S., F.L.S.) (With 3
woodcuts). eroi ees was 1 orm ERR Ure 213

Brown, N. E., A.L.S., and Harry Borvs, F.L.S.
Contributions to South-African Botany.—Orchidee. Part II.
With Additional Notes by N. E. Brown, A.L.S. (PlateL.) 65

CLARKE, CuanLEs Baron, F.R.S., F.L.S.
Botanical Observations made in a Journey to the Naga Hills
(between Assam and Muneypore), in a Letter addressed to Sir
J. D. Hooker 128

ow 9 «5» v» a » b. eR A e € Ve V P B PP do Pie Oe ee ee

lv
Page
Darwin, Francis, F.R.S., F.L.S.
On the Relation between the “ Bloom " on Leaves and the Dis-
tribütion of the Stone so we cs vase EE 99

DrveEry, Cuares T., F.L.S.
On a new Instance of Apospory in Polystichum angulare, var.
pulcherrimum, Wills. (With 3 woodcuts.)..........+.+-6- 437

GARDNFR, J. STARKIE, F.L.S., F.G.S. P
Remarks on some Fossil Leaves from the Isle of Mull (Scotland) 219

Hzwsrry, W. Borring, A.L.S., Assistant for India in the Herba-
rium, Royal Gardens, Kew.
Report on the Vegetation of Diego Garcia .................. 332

Hens Low, Rev. GEoncz, F.L.S.
A Contribution to the Study of the Relative Effects of Different
Parts of the Solar Spectrum on the Transpiration of Plants.. 81

Iro, Tokxvrano, F.L.S.
Berberidearum Japonie Conspectus. (Communicated by W. T.
Thiselton Dyer, C.M.G., F.R.S., F.L.S.) (Plate XXL) .... 422

LvusB5ock, Sir Joux, Bart., Pres. Linn. Soc., F.R.S., M.P., D.C.L.,
LL.D.

Phytobiological Observations: On the Forms of Seedlings and
the Causes to which they are due. (With 134 woodcuts.) .. 341

Masters, Dr. MaAxwELr T., F.R.S., F.L.S.
Contributions to the History of certain Species of Conifers.

Oates IL- X. and woodcuts.) |... o. u. -oeo eee 169
On the Floral Conformation of the Genus Cypripedium. (Plate
XA. and 10 woodcuts) >- eror oe e ra Ee ML 402

MirTEN, WiLLIAM, A.L.S.
Some new Species of the Genus Metzgeria ...........0..005. 241
The Mosses and Hepatice collected in Central Africa by the late
Right Rev. James Hannington, Bishop of Mombasa, F.L.S.,
F.G.S., &c., with some others, including those gathered by
Mr. H. H. Johnston on Kilimanjaro. (Plates XV.-XIX.).. 298

Page
Ripiey, Henry N., M.A., F.L.S., Assistant, Botanical Department, :
British Museum.
On Dr. Fox's Collection of Orchids from Madagascar, along
with some obtained by the Rev. R. Baron, F.L.S., from the

same island MO M P ou ere ee 116
On the Freshwater Hydrocharidez of Africa and its Islands.

(Elates ALPE & XIV) Sov a a redu e. ee 232
A Monograph of the Genus Ziparte. . oo rk. 244

Trimen, Henry, M.B., F.L.S., Director Roy. Bot. Gardens, Pera-
deniya, Ceylon.
Note on Balan hora Thwaitemt, Piche o.. 6-6 cnc cuss 330

val

EXPLANATION OF THE PLATES.

I. Sovrn-Arrican Omcurps. Floral parts of species of Disperis.
II. ABIES AMABILIS, sterile branch.
III. Bririsu grown cones of A. grandis.
IV. A. woniLis, British grown cone and sections of leaf.
V. Cones of A. nobilis, var. magnifica.
VI. A. RELIGIOSA, cone and foliage.
VII. Fnurr ann Fouce of Cephalotaxus pedunculata, var.
spheralis.
VIII. — PrcEA OmorrKa, cones and foliage.
IX. Cones anv ForiaaE of Pseudolarix Kempferi.
X. Branca and Fruit of P. Kempferi. z
XI. } Bornean Ferns, Nephrodium stenophyllum, and Trichomanes
XII. -Hosei, after J. Œ. Baker.
XIII. | Arrican Hyprocuarwea, viz. Boottia exserta and Blyxa radicans,
XIV. as described by H. N. Ridley.
XV.) i
XYL) CENTRAL-AFRICAN Mosses and Hepaticæ, various species and
XVII. details, accompanying Mr. W. Mitten's paper on Bishop Han-

XVII. | : , UT .
XIX.) nington’s and Mr. Johnston’s collections.

llustrations of Dr. Masters's
paper on History of Conifers.

XX. PzronRiíATE Flower of Selenipedium caudatum, illustrating floral
conformation by Dr. Masters,

XXI. ACHLYS TRIPHYLLA, var. japonica, as defined by Tokutaro Ito.

ERRATA.

Page 37, twelfth line from bottom, Jor "1NCcANUM" read 1NCARUM, the author
desiring the species to bear the name of the Incas.
In Plate XXI. for Achrys read Achlys; the text, p. 435, is correct.

THE JOURNAL

OF

THE LINNEAN SOCIETY.

Contributions to the Flora of the Peruvian Andes, with Remarks
on thé History and Origin of the Andean Flora. By Jonn

Bart, F.R.S., M.R.LA., F.L.S., &c.
[Read 5th November, 1885.]

In the month of April 1882, I was able to make a short excur-
sion in the Peruvian Andes, near to Lima, and to collect a good
many plants of the middle and higher regions. Finding that my
collections include a fair proportion of undescribed species, and
present an occasion for some remarks on the vegetation of the
great range of the Andes, I offer to the Linnean Society an enu-
meration of the species collected, with brief descriptions of those
hitherto undescribed, and of some remarkable varieties, with a few
preliminary observations on the general character and possible
origin of the flora.

Owing to the disorganized state of the country during the war
between Chili and Peru, it was necessary for me to keep within
the portion of the territory occupied by the Chilian forces. This
fortunately included the remarkable line of railway destined to
connect Lima with the interior by the valley of the Rimae, and
the pass leading thence to Oroya, which lies on the eastern side
of the highest range of the Andes. Before the collapse of the
finances and regular administration of the government of Peru,
theline had been completed and opened as far as Chicla, the

LINN. JOURN.—BOTANY, VOL. XXII. B

2 MR. J. BALL ON THE FLORA

highest village in the valley of the Rimac, and to that place I
proceeded in company with my friend Mr. Richard Ward, to
whom I am indebted for kind assistance in collecting plants and
for several plants gathered by him in a subsequent visit, which
have added twelve phanerogams and one fern to the following list.

Of the three days spent at Chicla two were employed in the
neighbouring part of the Rimac valley, while one was destined
to an excursion to the upper region towards the summit of the
western ridge of the Andes, commonly distinguished in Peru as
the Cordillera. Unforeseen difficulties and delays limited the
extent of this excursion. Less than two hours were available for
botanizing, about 600 feet above the wretched hamlet of Casa-
palta, at a height of about 14,300 feet above the sea, and in going
and returning it was necessary to pass by spots which would
have afforded many additional species. It is obvious that a col-
lection made under such circumstances must fail to give an ade-
quate representation of the flora of even the single valley to
which I was confined ; but as I find that it includes much more of
novelty than I could have expected, it seems to supply a slight
contribution to our knowledge, and occasion for discussing some
questions connected with the history and distribution of the
Andean flora.

I must remark that, as my excursion from Lima was limited to
the western slope of the outer range of the Peruvian Andes, the
statements here made must be understood to refer to that region
only. Of the extensive plateau region lying between the great
parallel ranges—the western and highest, known in Peru as the
Cordillera, and the eastern, to which the term Andes is there ex-
clusively confined—and of the eastern slopes towards the Brazilian
frontier, forming what is termed the montaña, 1 have no personal
knowlege. The climatal conditions of the two regions are very
different, those of the montaña in particular being characterized
by abundant, and in some parts by excessive, rainfall.

As is well known, the coast region of Peru is marked by exces-
sive aridity, and is commonly termed rainless. "This is not
strictly accurate, as slight showers of rain recur at intervals,
which may be as long as three or four years, and during the
winter (June to August) dense fogs prevail over the low lands,
which must deposit a considerable amount of moisture, often in
the form of fine rain or Scotch mist, and the country, at least in |
the neighbourhood of Lima, then produces an abundant though |
short-lived vegetation. The ridge of the Cordillera in Central

OF THE PERUVIAN ANDES. 3

Peru is in most parts within less than 100 miles of the sea-coast,
and the flora suffers in consequence from the deficient supply of
moisture, and is poorer both in individuals and in species than
in the plateau region, which is better watered. On this subject I
found, however, reason to modify somewhat the statements of
writers of authority which are summed up in the 20th chapter of
Grisebach's great work on the * Vegetation of the Earth.'

The distance from Lima to Chicla is little over 70 miles ; but as
the former stands ouly 468 feet above the sea, and Chicla at the
level of 12,220 English feet, the ascent surpasses in height and
steepness anything hitherto accomplished by railway engineers.
For a distance of about 24 miles the valley slopes upward very
gently, and to a person travelling, as I did, at the latter end of
the dry season, the appearance of the surface is that of absolute
sterility, except along the banks of the Rimae, or along rare
irrigation channels. The land is not, however, absolutely devoid
of vegetation; some small Amarantacee (Alternanthera, Telan-
thera), the cosmopolitan Portulaca oleracea, Heliotropium parvi-
Jlorum, Boerhavia viscosa, Franseria ambrosioides, and one or two
other unattractive flowering plants, occur here and there, and
prove that the climate here is not so absolutely dry as it is in
southern Peru and northern Chili, where the high range of the
Andes is more distant.

As the valley becomes gradually narrowed between the project-
ing ridges of the mountains, appearances of vegetation rapidly
increase. Cactoid plants (apparently belonging to three species
of Cereus) and the nearly withered stems of a large Bromeliaceous
plant (Puya?) are seen on the rocky slopes, which are furrowed
by channels cut by running water; and about San Bartolome,
nearly 40 miles from Lima and 5000 feet above the sea, a tinge of
green on the declivities becomes perceptible even at this season,
and a slight trickle of water may be noted in some of the lateral
ravines. Flowering plants, especially Composite of the genera
Tessaria, Baccharis, Viguiera, Encelia, Bidens, &c., show them-
selves in the bed of the valley, but the species are for the most
part the same that are seen in irrigated places near the coast.
At Surco, 6655 feet above the sea, are some gardens in which
tropical fruits (Anona Cherimolia, bananas, granadillas, &c.) are
successfully cultivated. Up to this point it would appear that
the change of climate, so far as regards temperature, must be very
slight. The decrease due to increasing elevation is compensated

B2

4 ME. J. BALL ON THE FLORA

by the fact that the valley is to a great extent protected from
the influence of the southern currents of the air and ocean,
which cool, to an abnormal extent, the sea-board of Peru; while
the inereased effect of solar radiation probably makes the heat
by day greater than at Lima. In this part of the valley tree-
vegetation of a stunted aspect becomes more prominent. Besides
Salix Humboltiana, which follows the banks of the Rimae, Schinus
Molle, and an Acacia (?) with thick trunk and stiff spreading
branches, but only 12 or 14 feet in height, are the most con-
spicuous forms.

About 55 miles from Lima, and 7800 feet above the sea, stands
San Juan de Matucana, the chief place in the valley. Judging
from the fact that many of the characteristic species and even
genera of the lower zone disappear above this height, while
numerous types characteristic of the Andean region are first seen
above Matucana, I have fixed on that place as the lower limit, in
the Rimac valley, of what I may call the middle zone of the
western Andes. So far as I know, that region is not marked in
Peru by the predominance of any vegetable group, such as we
find in the Cinchona forests of Columbia. Its natural limits are
fixed by a temperate climate, with a very restricted range of the
thermometer, combined with occasional though scanty rainfall
throughout the year. In this valley its lower limit is marked by
that of Heliotropium peruvianum, which appears to range from
about 8000 to 10,000 feet above the sea; while its upper limit
may be indicated by that of the suffrutescent Calceolarie and of
Lupinus paniculatus.

The most positive conelusion to which I was led by my visit to
Chicla was that, so far as regards this part of Peru, the limit of
alpine vegetation has been placed by writers on the subject far
too low. There can be no serious error as to the heights given in
this paper, which are based on the measurements of the railway
engineers ; and in departing for Chicla, 12,220 feet above the sea,
I fully expected to find myself in the alpine zone of the Andean
flora, whieh Grisebach, on the testimony of Tschudi and other
travellers, has placed for Peru and Bolivia between the level of
11,508 English feet and the line of perpetualsnow. Following
Humboldt, Grisebach places the inferior limit in the Andes of
Ecuador and Columbia at the still lower level of 10,872 feet. In
the valuable notes contributed by M. André to the French edition
of Grisebach's work many facts are given to show that the out-

OF THE PERUVIAN ANDES. 5

lines laid down by Humboldt for the vast region lying within
eight or ten degrees on either side of the equator must be exten-
sively modified; and, as regards the Peruvian Andes, my brief
experience satisfied me that received opinions require no less
considerable correction.

On first sallying forth on the morning after my arrival at
Chicla, I found with surprise that all the conspieuous types of
vegetation belonged to the temperate, and not to what naturalists
designate as the alpine zone. Five species of Calceolaria, Alon-
soa, a beautiful Clematis, a large Lupinus forming dense masses
on the slopes, a large red-flowered Echeveria, several conspicuous
species of Bidens, fruticose species of Solanum, a Nicotiana,
Verbena diffusa, and many others of similar character, all pro-
claimed the prevalence of a temperate climate in which frosts
are rare and of short duration. The impression thus received
was confirmed when I was able to compare the vegetation at this
level with that of the true alpine zone, about 2000 feet higher
than Chicla. Of 46 species hastily gathered at the upper station,
only eight were common to Chicla, and of these, three only could
be reckoned as characteristically alpine, viz.:—Draba siliquosa,
Alchemilla pinnata, and Saxifraga cordillerarum. It is needless
to say that in all countries the limits in altitude of zones of
vegetation are subject to variation from local causes, and it is use-
less to attempt great accuracy. Judging, however, from my own
observation, I should place the lower limit of the alpine zone on
the western side of the Peruvian Andes between 12,500 and
13,000 English feet above the sea, sometimes, perhaps, even above
the latter limit. In ascending the valley above Chicla, at a height
rather exceeding 13,000 feet, we passed a farm-house where corn
of some kind had been sown, and where some trees of Sambucus
peruviana, probably planted, had reached a height of 25 or 30
feet. This, which is the sole representative of arboreal vegeta-
tion in the upper valley, could scarcely thrive in the true alpine
zone.

The explanation of the relatively high extension of the tem-
perate flora in this region is naturally to be sought in climatal
conditions, which are in many respects peculiar. Observations
of the thermometer are, indeed, not forthcoming; but there is
abundant proof that the oscillations of temperature are very
moderate throughout the year, and that, although the nights are
always cool, frost is scarcely known. Between the 21st and 24th

6 MR. J. BALL ON THE FLORA

April—corresponding to October in the northern hemisphere—
I found the mean temperature at Chicla about 52? Fahr., the
extremes observed being 65^7 at noon on the 21st, and 42° soon
after sunrise on the 22nd, before the rays had reached the adjoin-
ing slopes. Rain, according to local information, occurs sparingly
at irregular, but not distant, intervals throughout the year. On
the night of the 21st rain fell in some abundance, and lay in the
form of snow on the higher parts of the range above the level of
about 15,000 feet, but had nearly disappeared two days later. It
would appear that as to the two main factors of climate—tem-
perature and moisture —there is a considerable difference between
the puna, or plateau region between the two great ranges of the
Andes, and places of equal elevation on the western slope.
Poeppig, Tschudi, and other scientific travellers, agree in speaking
of the frequency of storms in the puna region, which, at heights
little exceeding 12,000 feet, are often accompanied by heavy snow,
and they record relatively sharp frosts at places not much higher
than Chicla. Speaking roughly, I am inclined to place the lower
limit of the Alpine zone on the puna at about 12,000 feet. It
may be taken for granted that with a moister climate the alpine
—or proper Andine—flora is more fully developed there than on
the drier western slopes, while a few xerophilous species may be
confined to the latter region.

Speaking of the Equatorial Andes, M. André is disposed to
divide the region above the temperate zone, admitting a subandine
as well as an andine zone, the latter corresponding to the alpine
zone of Grisebach and other writers. I was not satisfied that
the facts justified such a subdivision in the limited field of ob-
servation open to me. Some species are doubtless limited to
the lower part, and some others to the upper part of what I
have termed the temperate zone; but I perceived no indications
of a general change in the constituents, or in the aspect of the
flora, excepting those which I have already specified, and I should
divide the vegetation of the valley that extends from the water-
shed of the western Cordillera to the mouth of the Rimac near
Callao into three zones—a subtropical dry zone, ascending from
the coast nearly to the level of 8000 feet; a temperare zone,
reaching to a height of 12,500 or 13,000 feet; and, finally, an
alpine zone, extending upward to the crest of the Cordillera.
The projecting eminences on the crest must certainly attain the
height of 17,000 feet above sea-level, and some of them may

OF THE PERUVIAN ANDES. 7

probably exceed that limit by nearly 1000 feet; but I doubt
whether snow lies permanently on any of these summits. Com-
paring the rather scanty details given by Poeppig*, who followed
the more direct line from Lima to Cerro de Pasco by Canta and
Culluay, I should infer that the character of the flora of the
western slope is there very similar to that of the upper valley of
the Rimac.

In discussing the regions of vegetation in the Equatorial
Andes, M. André says that there is there no zone characterized by
tbe growth of alpine grasses, and that where these abound they
have occupied ground which, in a natural state, was covered by
shrubs and low bushes which have been destroyed by fire. This
statement obviously refers to a zone considerably lower than that
which I saw above Casapalta, between 13,800 and 14,300 feet
above the sea. Ido not believe that this can have ever supported
a vigorous growth of shrubs; and, as regards the number of in-
dividuals, grasses undoubtedly took the first place, although here,
as everywhere in South America, the Compositæ showed the
greatest variety of species. As happens in all high mountain
regions, they are intermixed with species of the characteristic
alpine flora, but the grasses give the dominant aspect to the
vegetation.

The most singular plant seen by me in the higher zone was a
member of the Cactus family, of which I am unable to find any
description. It forms large, low, cushion-like masses, several
feet, or even several yards, in diameter, completely covered by
long glistening silky white hairs which conceal the slender needle-
like spines with which the whole plant is beset. At a little dis-
tance the plants were taken for patches of snow. Unfortunately
the specimen sent to Kew with a view to cultivation arrived in a
state of partial decomposition.

It has been generally asserted, or taken for granted, by writers
of authority that the flora of the tropical Andes is relatively poor
as regards variety of species. There is some reason to think that,
as compared with the highlands of Mexico, this idea is well
founded ; but I doubt whether it holds good on a comparison with
the floras of other high mountain regions. It must be remem-
bered that the vast region included under the term Andes is still
most imperfectly explored by naturalists. Excepting, indeed,
the Audes of Central Chili, between the 32nd and 37th degrees

* Reise in Chili und Peru, vol. ii.

8 MR. J. BALL ON THE FLORA

of south latitude, only a few points at wide distances asunder
have been visited by botanists, and each of these has supplied
numerous species not yet known to grow elsewhere. Strong
confirmation of this opinion may, I think, be found in the fact
that a hurried visit to a valley situated in the most sterile portion
of the great range should have produced out of a total number
of about 210 indigenous species, 17 undescribed, to which I have
felt it necessary to assign names, besides five or six more, almost
certainly new, which I have preferred to leave unnamed. Of the
17 species, three are identical with plants, hitherto unnamed, ex-
isting in European herbaria, and I feel sure that many more
Andean species remain to be added from this source. A slight
visit to the herbarium of M. Lombardi at Lima satisfied me that
a further considerable amount of undescribed materials must
exist among the collections of that gentleman and of other South
American botanists.

The accompanying list of plants collected in the upper Rimac
valley, between the limits of 7,800 feet and 14,300 feet above the
sea, includes 224 species and six marked varieties, of which two
or three may hereafter be ranked as distinct species. Of the
plants enumerated nine appear to have been certainly introduced
by the agency of man, viz.:—Capsella Bursa-pastoris, Lepidium
virginianum, Erodium cicutarium, Erodium malacoides, Medicago
denticulata, Melilotus indica, Centaurea melitensis, Solanum tube-
rosum, and Paspalum stoloniferum, and I have thought it best to
omit these from the tables given below. To the list of intro-
duced species many botanists would add the following five :—
Cerastium glomeratum, Stellaria media, Galium Aparine, Gnapha-
lium luteo-album, and Poa annua. As regards each of these,
however, I think that the weight of evidence goes to prove that
their wide diffusion throughout the world, and specifically in South
America, is mainly due to agencies other than the interference of
man, and especially to transport by birds.

In the subjoiued table I have sought to exhibit the distribution
of the species collected in the upper Rimac valley, distinguishing
those found only in what I may call the lower temperate zone,
between 7,800 and 11,000 feet above the sea, from those found
only about Chicla, between 12,000 and 12,600. The number in
the former category has been inereased by 13 species, collected
by Mr. Ward above Matucana in the month of May, and not
seen by me.

OF THE PERUYIAN ANDES. 9

Table showing the Distribution of the Indigenous Species belonging
to each Natural Order of Plants collected in the Upper Valley
of the Rimac.

|

About Chicla

ne
ly.
Alpine zone

Natural Order.

exclusive
Common to
lower tempe-
to Chicla,
exclusively.
Common to
Chicla and
alpine zone.
14,000' to

Lower tem-
perate zo
rate zone and |

Total number
of species

Hanunculacese: rak wb 2. oec ccce dese
Crucierm —-. f s2 2 3

Capparidez ...... lC OI NAGSSPp ene poil Carer ae
Polygalaceae c eee 1

Caryophylem -| | sees 6 T 1
Portulacacem:..-| 2... qe 2 1
Malvaces ......... 1
Geraniacer ...... 1
Leguminosz ...... 1
Rosacemi.-sue aee.
Sbwhernccceuducgui dp o fe
Crassulaces® =n: p 5.—— "TD wx
Onagrariem ...... 1

Loasacez ......... 1 1
Passifloracez ...... li
Cucurbitàcem. |] d | see D pee x
Datiscem ......... 1 bcc ceca

se | vevos

s.....

Wdéerss F wesess F | vespeq- | 44 pete

ard uu dr e
Caprifoliacem | c6 eI eee ] p AA:
Rübiacem ........ li qe l [5 1
Maleraneb o] 3. qo hv 2
Compositz......... 6 7 21
Campanulacem J 0 qo pone dee 1
Asclepiades ...... xc Ip
Gentianace® ...... IL [2x
Polemoniacem . | se Do 15$.
Hydrophylimes | =a: [| -=
Boraginez ......... 2
Convolvulacee ... 1
Solanacee ......... 5
Serophularinez ... 1

1
1
2

C to gU to m m E WH Qe m ma oi COTO OTE TOD

.
M
.
M
t2
_
uum

Bignoniacee ......
Verbenacee ......
Labtec
Plantaginess —— | e
Nyctaginee ...... 1 aes
Amarantacem ... Lpa ee s: [525
1
1

Chenopodiacee ...
Phytolaccacer ...
Polyponacem | ee Del eee
PSantalacem. oo os es ae iu Dou
BUrgoacem [D uo poc bec 5 1
GneltaceiB- 222275 | BREN | D ccce
Inde. 45. eel. d 50st d T do 5s
pAmarydüidee — | |. [| tes 1 cuu us
Talmeese............ lc [xol poeiveesd cms
Commelinacez ... 1
Junno C b uus v soie
Graminese ......... 3
Eibces .— 2 1
2

24 103 9 37 215

EET T ] . 9 oas- [| . ewenun. ] . a

ATI, GAO ad ch dS T — Ubxebes* 7 dh Ve

d isis (ak bod je poto sds ts i

ecsooso | /— tt

10 MR. J. BALL ON THE FLORA

As regards the proportion in which the natural families of
plants are represented in the Andean flora, the figures here given
agree pretty well with results derived from other sources. The
number of Composite amounts to nearly a quarter of the whole
number of species, and this proportion seems to prevail through-
out the temperate regions of South America. But the propor-
tion in my list for the higher or alpine region is a full third of
the whole, and, as far as I can judge, this is not far from the truth
as regards the Andes of Peru and Bolivia. Weddell’s ‘Chloris
Andina' is unfortunately incomplete, and it includes plants of the
upper region throughout the entire range from Columbia to Chili,
so that it does not allow of aceurate comparison. In that work
478 species of Composite are enumerated, and in the second
volume, including all the remaining Gamopetale and nearly all
the Polypetale, he reckons about 700 species. The remaining
phanerogams could scarcely amount to 400 species, so that the
proportion of Composite must approach to that above mentioned.

Next to the Composite the grasses play the most important
part in the Andean flora. In my list they amount to very nearly
one eighth of the whole collection, and in the alpine region to
one sixth. I am disposed to estimate the true proportion in this
part of the range at about one tenth.

It is not to be expected that in so small a collection the number
of species belonging to the less predominant groups should cor-
respond nearly to the real proportions, yet on comparison with
other materials I have been surprised to find a considerable
amount of agreement.

The third place in my list, as in Weddell’s work, is taken by
the Scrophularinee, which supply about five per cent. of the
entire number of phanerogams, and the fourth by Solanaceae,
while Crucifere, Caryophyllee, and Leguminose are each repre-
sented by 7 species, or one thirtieth of the whole number, probably
corresponding nearly to the true percentage.

Conspicuous by its absence from my list is the family of Cype-
racec. These are comparatively rare in the higher region of the
Andes, and apparently almost altogether absent from the dry
regions of Peru and Bolivia. .Primulacee, which are such con-
spieuous ornaments to the mountain floras of the old world, are
everywhere rare in America, and their absence is not a matter for
surprise.

The most remarkable feature in the local flora of Chicla is the

OF THE PERUVIAN ANDES. 11

presence of four species of Crassulacee, of which three appear to
be new. The order is very sparingly represented in South America,
and, so far as I know, no species of true Sedum has been previ-
ously recorded.

The figures in the foregoing list show a total number of 178
species collected in the temperate zone, and only 46 in the alpine
zone; but they suffice, in my opinion, to prove that the flora of
the western slope of the Peruvian Andes, and especially that of
the temperate zone, is by no means so poor as has been hitherto
supposed. It will be seen that after excluding the species intro-
duced by man, and those subsequently found by Mr. Ward, I was
able to collect 53 species by jumping from the train during very
short halts between the levels of 7800 and 11,000 feet, and that
136 species were found in the immediate neighbourhood of the
village of Chicla. The flora of the alpine region doubtless gave
me the impression of poverty; but it must be remembered that
the 46 species enumerated were all found within a very restricted
area, presenting little variety of surface or exposition, and I am
not sure that under similar disadvantages other high mountain
regions would give a better harvest.

It is not easy to find published data for comparison. The
results obtained in a district that has been fully examined by
botanists having time for the purpose are, of course, misleading ;
while in short excursions one usually attends mainly to the new
or rarer species, and overlooks the common and widely spread
forms. The only materials I have found that in some degree
admit of comparison with the results of my excursion to Chicla,
are contained in the appendix to Hooker and Ball’s * Marocco and
the Great Atlas. In the valleys of Ait Mesan and Amsmiz we
collected specimens of all, even the commonest species; and in
the accompanying table the distribution of each, between a middle
zone extending from the level of 1200 metres to that of 2000 m.,
and a superior zone reaching from 2000 to about 3500 metres, is
exhibited. These zones correspond fairly well with what I have
here distinguished as the temperate and alpine zones in the
Andes, but the cireumstances were in many ways different. In
Ait Mesan we spent the greater part of six days ; a much larger
extent of ground was traversed, and in great part carefully ex-
amined by two botanists. Our collections there gave a total of
385 species of phanerogams and ferns, of which 283 species were
found in the middle zone, and 154 in the superior zone, 52 species

12 MR. J. BALL ON THE FLORA

being common to both. In the Amsmiz valley the conditions
were more nearly similar, and afford better ground for comparison.
In that valley we collected 225 species, 162 being found in the
middle and 91 in the superior zone, 28 species being common to
both zones. The opportunities for collecting in the middle zone
were less than at Chicla, but much more favourable in the superior,
as we were able to examine the ground rather closely in the
course of an ascent and descent through a vertical height of
more than 4000 feet. My general conclusion is that, as regards
variety of vegetation, the western and most sterile slope of the
Peruvian Andes is at least as rich as the northern slopes of the
Great Atlas, and probably far superior to the southern slopes of
that range *.

In considering the bearing of the facts here brought together
on questions connected with the past history and origin of the

* Since the above lines were written I have had opportunities for comparing
the results obtained in the Andes with those procurable under similar con-
ditions in the Swiss Alps. During three days, between the 12th and 15th
August, I noted all the species seen in flower or fruit above the Baths of
Rosenlaui in the Bernese Oberland, between the levels of 4500 and 5200 feet
above the sea, corresponding as well as I could estimate with the range of my
short excursions from the village of Chicla. On the 2nd of August, I noted in
the same manner the species observed in two separate areas above the summit
of the Furca Pass, at about 8000 feet above the sea, the one slope facing south-
west, the other nearly due north. In both the latter observations, while devoting
about the same time that I was able to give in the Andes above Casapalta, I
covered more ground, with the result of obtaining many additional species.
The reason for omitting from my Alpine lists plants not seen either in flower
or fruit is, that in the Andes I abstained from collecting plants which I should
be unable to determine.

My Rosenlaui list includes 143 phanerogams and 7 ferns against 129 phanero-
gams and 7 ferns collected about Chicla ; and it is worthy of note that in each
list nearly three fourths of the phanerogams belong to the same natural orders.
In the Furca lists the better exposed area afforded me 65 species, and the less
favourably situated slope only 45 species, 30 species being common to both.
Bearing in mind the fact that a larger area was in both cases examined, I came
to the conclusion that, as regards the true alpine zone, the portion of the Andes
seen by me supports a vegetation as varied as that of Central Switzerland ; while
as regards the subalpine region, the Alps display a somewhat more varied vege-
tation. The range of individual species appears to be decidedly greater in the
Alps than in the Andes, as is shown by the fact that out of 80 species seen
above the Furca Pass, growing on crystalline schists near the superior limit of
vegetation, 18 species were found in the valley of Rosenlaui at a level lower
by 2500 or 3000 feet, and growing on calcareous soil.

I have no doubt that on the southern side of the main chain the Alps usually
exhibit a much more diversified vegetation.

OF THE PERUVIAN ANDES. 13

Andean flora, it will be best to exclude altogether the ferns, as
well as the species introduced by man. The 206 remaining
species include representatives of 49 natural orders. Of these,
fiveorders— Passiflorece, Bignoniacee, Nyctaginee, Phytolaccacee,
and Commelinacee, each appearing only by a single species which
does not rise above the level of 10,000 feet, are mainly inhabi-
tants of the tropical zones of the Old and New Worlds. Hydro-
phyllacee, represented by one species, and Polemoniacee, by two
species, are orders whose original home appears to lie in the
north-western part of the American continent, and which are
sparingly represented in South America. Loasacec alone, of
which four species are enumerated in the following list, form a
group which may be regarded as specially Andine. With the
singular exception of the monotypic genus A7ssenia in east tropi-
cal Africa, the other genera of this family, including about 100
species, appear to have all originated in the mountain regions of
Central and Southern America, which I shall here designate as
the Andes, using that term in a broad sense. Setting aside the
somewhat doubtful representative of the small anomalous order
Datiscee, we find that 193 species in the annexed list belong to
40 natural orders, which are all very widely distributed through-
out the world.

Descending from natural orders to the categories of suborder
and tribe, we discover few indications of a special Andean flora.
The genus Balbisia represents the small distinct Andean tribe of
Wendtiee belonging to Geraniacee, and Malesherbia belongs to a
tribe of Passifloracee peculiar to the Central and Southern Andes.
The genus Calceolaria, which by itself constitutes a tribe of
Scrophularinec, is represented at Chicla by five species ; but there
are very strong reasons for referring its origin to the Antarctie
region rather than to the Andes. To many a botanist the most
distinctive element in the Andean flora is afforded by the Muti-
siacec, a tribe of the omnipresent Composite, which is not, indeed,
confined to the South American continent, but which there finds
its chief development, especially in the Chilian Andes. Of 57
known genera*, containing about 420 species, 42 genera and
fully 350 species are confined to South America. The rest are
scattered from South Africa to China, and only one American
genus (Trichocline) is represented elsewhere, 2.e. in Australia.

* Throughout this paper I have adhered closely to the limits of genera as
defined in Bentham and Hooker's * Genera Plantarum,' and in the following
remarks I have applied the same rule in comparing my results with those of
Weddell.

14 MR. J. BALL ON THE FLORA

In attempting to speculate on the possible origin of this remark-
able tribe the first difficulty is to answer the question, whether
we should regard them as all descendants from a common ances-
tor. It is quite true that a complete natural arrangement, if
such were attainable, would make community of descent the cri-
terion for the establishment of genera as well as of larger groups;
but it may well be doubted whether, in cases where a vast number
of forms have been simultaneously developed by continually fresh
divergences from one or a few ancestral types, it is possible to
reconstruct the genealogical tree. In a remarkable paper, pub-
lished in the Journal of this Society, Mr. Bentham explained the
principles by which he was guided in arranging the Composite for
the ‘Genera Plantarum. He must be a bold man who would, in
the present state of our knowledge, undertake to improve upon
Mr. Bentham's classification of the 780 genera and some 10,000
species that are already known of the vast family of Composite;
but I think that in pointing out, as he has done, the complex
affinities of the Mutisiacee with other tribes—Cynaroidee, Inu-
loidee, and Senecionidee—he implied a doubt whether the plants
which must under any logical plan of classification be ranked in
the same tribe, may not have originated through very different
lines of descent. 1t is impossible, for instance, to compare some
species of Trixis with species of Senecio closely resembling them
in habit, and differing only to a slight extent in floral characters,
and to retain the belief that they have no nearer genealogical tie
than that which may have existed between the ancestor of the
Senecioids and that of the group including such remote forms as
Mutisia, Chuquiraga, and Nassauvia. I venture to believe that
under Mutisiaceæe are included plants originating in many very
different lines of descent, but that among these there are some
minor groups distinguished by high relative antiquity. That
which includes Barnadesia, Chuquiraga, and Mutisia apparently
had its origin in the high mountains of tropical South America ;
Nassauvia and a few allied genera may have arisen in the southern
Andes, while Trichocline, Chetanthera, and some of their allies
may perhaps trace their origin to an Antarctic home, whence the
ancestor of Gerbera and a few other Old- World forms may have
reached South Africa.

We have seen that, so far as the materials go, endemie natural
orders and tribes do not appear to play an important part in
the Andean flora. It is when we examine the smaller groups
of genera and species that we perceive better the true character

OF THE PERUVIAN ANDES. 15

of the flora. Of the genera in my list a full half, or 63 out of
122 geuera, are Cosmopolitan, that is to say, are widely spread
throughout the world, and extend to the temperate regions of
the northern hemisphere, including Europe ; and these 63 genera
include 127 species or five eighths of the entire number. A small
number of genera, which I have designated Amphigean, extend
to the Old World, but have a limited distribution and do not reach
Europe. Three genera (Halenia, Castilleja, and Muehlenbergia)
are common to North America and to temperate Asia, and may
for the present purpose be reckoned under the head of American
genera, common to the northern and southern parts of the conti-
nent. Of these, including the three above named, I reckon 19
genera, to which belong 25 species. I next note six genera which
are common to South America and the cooler extratropical regions
of the southern hemisphere, and which I designate as Antarctic
genera. These are Calandrinia, Acena, Oreomyrrhis (or Caldasia),
Calceolaria, Ourisia, and Muehlenbeckia, and are represented in
my collection by 12 species. There remain the genera peculiar
to the Andean region, using that term in a wide sense to include
the highlands of Central America and Mexico, as well as those of
Brazil and Venezuela. Of these, termed Andean genera, I reckon
27, including 32 species. It is thus apparent that, so far as
regards genera, the flora of the Peruvian Andes is less widely
separated from the rest of the world than has been commonly
supposed, and it is only when we come to scrutinize the species
that we find the local character of the vegetation strongly im-
pressed. In faet, very nearly five sixths of the species are
exclusively Andean, and of these, the majority appear to occupy
only a limited area within that region. The subjoined table will
exhibit more clearly the distribution of the species.

Table showing the relative Extension of the Area occupied by
the indigenous Species of the Rimac Valley.

No of No. of No. of No. of

Genera. ox Cosmopoli-, American | Andean

gu imn species.| species. species.
Cosmopolitan ......... 63 9 | 18 100
Aimplipem. ..——— Y. 7 1- ues 5 5
American cic cents 9 qu ce 6 19
Autarede Loser v 6 PL donus It djs 12
Andean ~.e dic rcc e m S. 32
d. Robes. 122 $1 2 168

16 ; MR. J. BALL ON THE FLORA

In attempting to draw any conclusions from the materials
at my disposal, I was naturally confronted with the doubt whether
a collection so small could be regarded as in any degree repre-
sentative of the Andean flora. Unfortunately we possess no
complete work in which the existing materials for a flora of any
part of the tropical Andes have been brought together. The
‘Chloris Andina’ of that excellent botanist the late Mr. Weddell
unfortunately remains incomplete, and for purposes of compari-
son it is difficult to use, because it includes a large number of
species confined to the Chilian Andes, and belonging to genera
either not at all, or very slightly represented in the equatorial
region. There isalso some inconsistency in the limits of elevation
assumed as characteristic of the Andean flora. I have attempted
roughly to compare my results with those obtainable from the
* Chloris Andina,’ and am led to the conclusion that they are more
fairly representative than I could have anticipated.

The only large group as to which Weddell’s work is complete
is that of the gamopetalous dicotyledons, including the Composite.
For the latter the inclusion by Weddell of a large number of
genera and species of Mutisiaceæ which do not extend to the
tropical Andes would make the comparison misleading ; and for
similar reasons I have restricted the comparison of the remaining
natural orders of Gamopetale to the Bicarpellate as defined by
Bentham and Hooker.

Tables showing Comparative Results as to two large Groups of
Plants of materials in the ‘Chloris Andina’ and in the Col-
lection made in the Rimac Valley.

Composite, excluding Mutisiacee.

Chloris Andina. Rimac Valley.

No. of No. of No.of | No. of

genera. species. genera. | species.
Cosmopolitan genera ...... 10 214 9 | 33
Amphigean genera ......... 2 18 2 | 2
American genera ............ 6 43 4 | 7

Antarctic genera ............ 2 3 |

Andean genera ............... 17 88 t 4
dose. s 37 366 19 | 46

OF THE PERUVIAN ANDES. 17

Bicarpellate.

Chloris Andina. Rimae Valley.

No. of No. of No. of No. of

genera. species. genera. species.
Cosmopolitan genera ...... 19 154 10 24
Amphigean genera ......... 1 6 1 L
American genera ............ 8 28 10 10
Antarctic genera ............ 2 46 2 6
Andean genera ............... 18 39 4 4
"Totali... es 48 218 21 45

Allowing for the numerous circumstances leading to discre-
pancies in the results of such a comparison, the figures given
above show a remarkable amount of agreement, and lead to the
inference that (when the special elements of the Chilian flora are
excluded from consideration) the small collection made by me is
fairly representative of the general character of the flora of the
Peruvian Andes. I have therefore thought it worth while to
consider what indications it may afford to elucidate the problems
of the past history of the South-American flora, availing myself
where possible of the fuller materials supplied by the * Chloris
Andina.’

The first fact which must be taken into account is the prepon-
derance in this flora of types that are very widely, almost uni-
versally, diffused throughout the world. It has been already
pointed out that nearly nineteen twentieths of the species (193
out of 206) belong to 40 natural orders that are found in almost
all the mountain-regions of the world. Comparing the flora of
New Zealand, as one of the most distinct and isolated, I find that
only two of the above 40 orders, Capparidee and Asclepiadee, are
wanting in that region. I further note that (omitting Datiscee)
more than 60 per cent. (126 out of 206) of the species in my col-
lection belong to 62 cosmopolitan genera, many of which have a
range nearly as wide as the natural orders to which they belong.
These proportions would doubtless have been somewhat altered
if I had been able to collect more largely in the higher region.
I should have added largely to the number of generic types

LINN. JOURN.—BOTANY, VOL. XXII. c

18 MR. J. BALL ON THE FLORA

peculiar to the Andes, but the proportion of species belonging to
cosmopolitan genera would not probably have been very much
reduced. Ihave tabulated the proportions found in the * Chloris
Andina' for genera and species belonging to the gamopetalous
dicotyledons which here constitute about one half of the entire
phenogamous flora.

Analysis of the Genera and Species of Gamopetalous Dicotyledons
in the Andean Flora.

Number of Number of species

genera. in Andean flora.
Cosmopolitan genera ............ 34 422
Amphigean genera ............... 5 50
Wide-spread American genera... M 76
Antarctic genera ............. e 6 57
Andoan Zenera iei: 62 274
Potalas oiai s: 124 879

It will be recollected that this table includes a large proportion
of the Chilian flora, in which the endemic elements are much more
strongly marked than in the Equatorial Andes. Yet even here,
we find nearly one half of the species belonging to 34 cosmopolitan
genera, a full third being supplied by the 6 genera Valeriana,
Gentiana, Bartsia, Erigeron, Gnaphalium, and Senecio; while
much less than a third of the species are ranked under the 62
endemie genera of the Andean flora.

I may here remark that, so far as the available materials admit
of comparison, the Andean flora appears to be one of the most
distinct existing in the world, if we take the proportions of en-
demic genera and species as criteria. It would appear that about
one half of the genera and fully four fifths of the species are
limited to that region, showing a degree of separateness scarcely
surpassed elsewhere. Allowing for the fact that the endemic
elements of the flora of each large region are more conspicuous
in the mountains than in the low country, it is remarkable that,
comparing the same large group of Gamopetale in the Australian
and Andean floras, we find in the first only 97 endemic genera
out of a total number of 353, whilein the latter one half of the
genera are endemic.

OF THE PERUVIAN ANDES. 19

The faet, which I take to be beyond dispute, that in all the
mountain-regions of the earth a comparatively small number of
the same vegetable types form prominent, if not predominant,
elements in the existing floras, seems to me strong evidence of
their great antiquity, even as geologists use that term. Of many
of them it may be asserted that they possess no special facilities
for transport over wide areas of ocean, and it must be remarked
that, while in the large majority of cases genera thus widely dif-
fused are represented by species differing widely in structure,
we yet find that in cases where the diffusion appears to have
occurred in recent geological times, the species themselves have
remained unaltered, and scarcely exhibit the amount of change
which botanists designate by the term variety. As compared
with the historical period, the separation of the British Islands
from Northern and Central Europe is an event of high antiquity,
though recent in a geological sense. Many species have been
exposed to new conditions since that date; -but not a single new
species, scarcely even a well-marked variety, has been developed
during the interval.

I trust that I shall be forgiven if in spite of the great authority
attaching to the name of Sir Joseph Hooker, who in this matter
has been followed by Mr. Wallace, I prefer to designate as Cos-
mopolitan, rather than Scandinavian, the types of vegetation of
which the only fact certainly known is that they are widely
spread throughout the world. In the present state of our know-
ledge it is, I think, impossible to form any safe conjecture as to
their origin, but there appear to be strong reasons for attributing
to them a high antiquity. As to some of them, there are grounds
for supposing that the colder North Temperate Zone was, if not
their original home, the region from which they have migrated in
recent times; but of these many may, with more reason, be
referred to North America than to Scandinavia, while as to many
others the balance of evidence points to an original home in the
high mountains of lower latitudes.

The constituents of the Andean flora furnished by the genera
which I have called Amphigean do not demand detailed discussion
in this place. Only five or six per cent. of the vegetable popula-
tion belong to this class; they belong chiefly to families that are
common to the Old and New Hemispheres, but which do not
extend far beyond the tropics. Any explanations that can be
offered as to their origin must be, to a large extent, con-

c2

20 MR. J. BALL ON THE FLORA

jectural, and in each case should be considered separately with
reference to the facts of its distribution and its structural affi-
nities.

The class which I have designated American genera, including
those which are widely spread through both portions of the con-
tinent and, with the exception of a few representatives in North-
eastern Asia, do not extend to the Old World, forms a considerable,
though not a preponderant, constituent in the Andean flora.
When we consider that, although subsidence has probably at various
times separated the two portions of the continent, the highlands
of Mexico and Central America have, in all probability, served
during long periods as a bridge over which some portions of the
mountain vegetation may have been transferred from North to
South and vice versá ; we are led rather to feel surprise at the
amount of separateness now existing than at the presence of
many genera and of a few identieal species in the floras of the
Andes and that of the Rocky Mountains. It is true that I have
reckoned as Andean genera and species many that extend north-
ward as far as Mexico, and it may well be that that region, 80
rich in varied forms of vegetation, is the original home of some
that now appear to be more fully developed in the mountain
ranges of Western North America. Among the wide-spread
American types we must note two natural orders whose original
home may, with some confidence, be placed iu the north-western
part of the continent. The Polemoniacee, of which about 140
species belong to that region, are represented in the Andes by
five species of Gilia, one of Collomia, and by the endemic genus
Cantua. They have sent to the Old World two or three species
of Phlox in Northern Asia, and a single emigrant which has
reached Britain, the Jacob's Ladder of old-fashioned gardens
which maintains a struggling existence in several isolated spots
in Europe. The other specially American family is that of the
Hydrophyllacee, of which 12 genera are known in North America,
but which is represented in the Andean chain only by four species
of Phacelia. An opposite phenomenon is presented by the
natural order of Loasacee, which we may, with much probability,
hold to have had its origin in the mountain regions of South
America. The genus Mentzelia, which extends in South Americà
from the plains of Patagonia through the temperate zone of the
Andes, must at some remote period have travelled northward to
Mexico, and thence to temperate North America. It has there |

OF THE PERUVIAN ANDES. 21

developed several distinct species, one of which, at least, has spread
northward into British North America.

Among the American genera that are conspicuous in the
Andean flora one of the most interesting is Baccharis. Among
some 280 described species which are scattered over every part of
South America, from the Straits of Magellan to the equatorial
zone, and which extend on the west side of the continent to
California and Nevada, we find a surprising variety in habit and
in the form and arrangement of the vegetative organs, along with
a remarkable amount of uniformity in the reproductive organs,
which are constantly dicecious. I confess that the arguments
that have led some distinguished botanists to consider the great
family of Composite as of comparatively recent origin appear to
me altogether inconclusive. When I consider the vast variety of
forms which it includes, the degree in which some large groups are
localized in different regions of the earth, while others, such as
Senecio, have representatives in every zone, I shrink from the
conclusion that their origin can be, even in geological language,
at all recent. It is, of course, not inconceivable that plants which
we class together under the name Composite may have come into
existence by different lines of descent through gradual modi-
fication from different ancestral types; but when we consider
the general agreement in the structure and arrangement of the
essential organs, I think that the balance of probability inclines
decidedly towards the belief in a community of origin of all the
various existing forms. Be that as it may, we are, I think,
justified in looking to the mountain regions of South America
as the original home of many large groups, such as the genus
Baccharis, most of the Mutistacee, and many genera of other
tribes.

I could not fail, while travelling in South America, to reflect on
the probable origin of the types of vegetation which are common to
that portion of the continent and to distant spots in the southern
hemisphere, separated by wide expanses of ocean. The connec-
tion between these distant floras is proved by the presence of
closely allied, and even of some identical species, belonging to
genera not known elsewhere in the world, and extends to many
forms that do not appear to be provided with special facilities
for transport to a distance.

I was struck by several considerations which concur in indi-
cating the same probable explanation of these remarkable facts.

22 MR. J. BALL ON THE FLORA

We know that at a recent geological period, and very probably
at many preceding periods, the climate of the northern polar
regions was so far different as to favour the growth of plants now
characteristic of the warm temperate zone. Whatever may have
been the causes of such a change in the distribution of tempera-
ture, it is eontrary to all analogy to assume that in the long record
of the vicissitudes in the condition of our planet similar causes
have not produced effects in some degree similar in the southern
hemisphere. Further than this, our limited knowledge of the
geology of the southern hemisphere suffices to show that most,
if not all, portions have undergone very great oscillations of level,
involving as a necessary consequence alternate extension of the
areas of land and ocean.

The difficulties of antarctic navigation have, unfortunately,
hitherto limited our knowledge of the south polar regions; and,
while many believe that a considerable continental area, wholly
or in part covered with perennial snow or ice, now surrounds the
southern pole, other geographers, with whom I am disposed to
agree, think it more probable that the south polar lands consti-
tute a great archipelago formed by numerous islands, some of
which are probably of great extent. Whatever may be the fact
at the present time, there is a strong probability that, in the
course of past changes of temperature and level, a mild climate
and an extensive area of Antarctic land may once, or very possibly
many times, have occurred simultaneously.

Here, at a period remote even in geological language, we should
look for the original home of those types of vegetation which we
designate as Antarctic. In opening a new region for habitation
to a comparatively few vegetable ancestors, the favourable con-
ditions for the development of new generic types were provided,
and the process would have been hastened by the frequent vicis-
situdes of temperature and moisture to which the climate of such
a region must have been subjected.

I was altogether unconscious that in following this train of
speculation I was merely developing a suggestion contained in
the 12th chapter of the ‘ Origin of Species’ *, which had failed

* The passage, which has only very recently been recalled to my recollection,
runs thus :—* This difficulty almost disappears on the view that New Zealand,
South America, and other southern lands were long ago partially stocked from

a nearly intermediate, though distant point, namely, from the Antarctic islands,

when they were clothed with vegetation, before the commencement of the Glacial
period."

OF THE PERUVIAN ANDES. 23

to impress itself upon my memory when originally presented to
it, and I further find that the same view has since been defended
by Mr. Wallace in his remarkable work on ‘Island Life,’ and
still more recently by Professor Engler. If I have rightly ap-
prehended the brief statement of his views given by Mr. Wallace
in pages 489 and 490 of the work above cited, they do not com-
pletely accord with the conception that seems to me best to
account for ascertained facts ; and I may here specify the points
of difference. Mr. Wallace regards the peopling of the Antarctic
lands with their characteristic generic types as a comparatively
recent event. He supposes that those already existing types
were carried southward from South America, or other circum-
polar countries, to a preglacial Antarctic continental region, and
having there spread widely, were thence conveyed northward to
the widely scattered islands where they still survive. To me it
seems much more probable that the origin of the special generic
types of the Antarctic flora should be referred to the Antarctic
region itself. They belong, without exception, to the great
groups or natural orders which are now almost universally
diffused throughout the world; and the ancestral types from
which they originated were probably carried to that region at a
remote period when the physical conditions of the earth’s surface
were widely different from those now prevailing. Their present
distribution points rather to migration from a higher latitude to
their present homes, than to transport in an opposite direction.
On large continents where no barriers, such as wide deserts or
high mountain-chains, intervene, the diffusion of plants occurs
most easily in directions between east and west, because the
climatal conditions are more nearly uniform. But the reverse
obtains where the transport takes place across wide spaces of
ocean. The two chief agents are ocean-currents, with or without
floating ice, and migratory birds ; and, speaking generally, the
predominant direction in which these travel approaches to the
meridians of longitude.

I may here venture to express the belief that the efficacy of
winds as agents for the dispersal of plants over wide ocean spaces
has been considerably overrated. We know that the spores of
eryptogams may be conveyed for indefinite distances by aérial
eurrents; but the seeds of flowering plants are immeasurably
heavier; and I am not aware of a single faet that authorizes us
to infer that mode of transport over wide intervals of ocean as

24 MR. J. BALL ON THE FLORA

efficient in the present condition of things. No doubt tornadoes
carry many seeds to considerable heights, and if they travelled
far with currents of equal violence, might carry them across
spaces of many hundred miles in width; but, if at all, such a
possibility can scarcely exist in temperate latitudes. If winds
were efficient for the purpose, I feel sure that we should find
evidence of the fact in the much wider oceanic distribution of
species having small seeds or winged appendages.

On the other hand, it appears to me that Mr. Wallace has not
sufficiently dwelt on the predominant influence of migratory
birds in the dispersal of plants, where physical obstacles prevent
their gradual diffusion. If it be true that many facts of distri-
bution may possibly be accounted for by ocean-currents carrying
floating timber or ice-rafts laden with soil, it remains true that
most of these facts are at least equally well explained by the inter-
vention of birds; whilst for numerous other facts transport by
birds affords the only probable explanation. I do not know
whether it has been remarked that, as between regions in the
temperate zone and others nearer to the pole, the chances of the
transfer of species by birds from bigher to lower latitudes is very
much greater than in the opposite direction. As a rule, birds
winter in the milder climate and resort to higher latitudes in
spring or early summer*. Inautumn, when seeds and fruits are
mature, they carry them to a milder region, where, under favour-
able conditions, these may establish themselves. If we admit
the probability of an Antarctic centre of distribution for the types
of vegetation which we are agreed to call Antarctic, but suppose
those types to have originally reached the Antarctic region from
South America or Australia, we find it very diffienlt to conceive
the means by which they could have reached the polar continent ;
while if that continent were their original home there is no diffi-
culty in understanding their subsequent diffusion in a northerly
direction.

It may be reasonably objected that a similar argument might
be urged against the primitive peopling of an Antarctic continent
by the ancestors whence the special Antarctic genera must have
been derived ; and the answer to this objection requires a very

* Professor Giglioli has called my attention to the fact that some birds mi-
grate from low latitudes in summer to higher latitudes in winter. Such cases

are rare and exceptional, but they may serve to explain some anomalous facts
of plant distribution.

OF THE PERUVIAN ANDES. 25

brief discussion of the reasons which go to prove the remoteness
of the period at which the original colonization of the Antarctic
region must have occurred. The undoubted fact of the diffusion
throughout the world of those natural groups of plants which I
have called cosmopolitan leads directly to the conclusion that
their diffusion must have been effected at a period when either
the relative distribution of land and sea must have been widely
different, or when the physical conditions of the earth's surface
were quite other than those now prevailing. Accepting, within
moderately wide limits, the doctrine of the permanence of con-
tinental areas, I cannot resort to the first alternative, and I see
in the second the more probable explanation. The skill and
energy with which the doctrine of the uniformity of physical
conditions of the earth's surface during past geological periods
has been advanced by eminent modern authorities has gone far
towards obtaining general assent; but I cannot believe that it
will maintain its ground against the stringent arguments with
which it has been met by physicists and astronomers. Allowing
a considerable margin for the action of modifying causes, it seems
impossible to doubt that, at a period not very remote in a geolo-
gical sense, the temperature of the earth's surface was sensibly
higher, the difference between polar and equatorial temperature
greater, and in consequence the currents both of the air and
ocean morerapid. Along with this the action of the tides was
far more violent, and the inequalities of the earth's surface more
considerable. The relief of the surface at any given period
represents the sum of the results through previous ages of two
sets of opposing forces—those of upheaval, and those which lower
eminences and fill up depressions. Chief amongst the former is
contraction of the earth's crust through gradual cooling ; and
this must necessarily have been more efficient in the earlier stages
of the earth’s progress, while at a later period the levelling forces
have gradually assumed a supremacy which they obviously possess
at the present time.

Various considerations tend to the conclusion that the dispersal
of the chief cosmopolitan genera of plants may have coincided
with the period of the deposition of the older Secondary rocks,
and at that period physical agencies far transcending those of
our experience prevailed throughout the earth. If the ancestors
of the Antarctic types of vegetation were then established in a
south-polar continental area, and were developed from them by

26 MR. J. BALL ON THE FLORA

gradual modification, I see no difficulty in believing that they may
have maintained themselves through suecessive gradual changes
of physical conditions within the same region, and even that some
may still survive within the Antarctic Circle. This conjecture pre-
sented itself forcibly to my mind when passing some days in mid-
winter at Sandy Point in the Straits of Magellan. Many delicate
plants, impatient of sharp frosts, maintain themselves there in
the same latitude with not distant places, such as South Georgia,
whieh are almost completely covered with permanent snow. The
reason is sufficiently obvious. The accumulation of snow or ice
taking a permanent form occurs only where the supply, by pre-
cipitation or condensation, exceeds the amount annually melted,
and this is limited to special portions of the earth's surface,
usually to high lands, exposed to moist currents from the sea *.
Having considered the constituents of the Andean flora which
have most probably been derived from other regions of the earth,
Imay briefly notice the remaining types which are peculiar to
this part of the earth. From the figures given in the preceding
pages we are led to conclude that the large majority of the species
and about one half of the genera of the Andean flora are not
known elsewhere. Of these endemie species and genera the
majority are allied to the cosmopolitan types to which I have so
often referred, and extend northward to Central America and
Mexico. Others, and especially the genera which exhibit less
affinity to types known elsewhere, are confined to the proper
range of the South-American Andes, and some of them to a small
portion of the range. We find in these facts reason to believe
that, while South America has from a remote period been able to
receive occasional colonists from the north, there has been a very
long period of relative isolation during which a large number of
separate species, and not a few genera, have been differentiated.
As to some groups at least, such as the Loasacee, many Mutisi-
acec, and others where the structure has been profoundly modified,
we must allow a very long interval for changes so considerable.
If it be objected that we have reason to believe the elevation of
the Andean chain to be a comparatively recent geological event,
I would reply, in the first place, that there is no reason to suppose

* This is not the place to discuss the views which Dr. James Croll has main-
tained with singular ability and a wide grasp of the facts and principles of
physical science. I find it impossible to accept his theory of a polar ice-cap and
the important consequences which he derives from it.

OF THE PERUVIAN ANDES. 27

the great movements of alternate elevation and depression to
have been nearly contemporaneous throughout the vast length
of that range, and that it is more probable that some portions
remained as lofty mountain-masses, while others were nearly or
altogether submerged. This remark applies especially to the
highlands of Bolivia and Peru, where the Paleozoic rocks do not
appear to be overlain by recent marine deposits.

Whether this view be valid or the reverse, we must, I think,
look to an adjoining region as the original home of the majority
of the special forms of vegetation that are now limited to South
America, as well as of many others that have spread more or less
widely to other parts of the earth. In a zoological as well as a
botanical sense Brazil is one of the most distinct and separate
regions of the earth. It is, in large part, a granitic region, from
which vast masses of superincumbent strata have been denuded,
and where the granite itself has undergone a great amount of
decay and ablation. We there see the ruins of one of the greatest
mountain-masses of the earth, where a very ancient fauna and
flora were developed, of which portions were able to migrate to a
distance, while others have been modified to adapt themselves to
the gradual changes of the environment. Many vegetable
groups which are but slightly represented in the higher region
of the Andes, but which doubtless originated in South America,
such, for instance, as the Melastomacee, probably had their origin
in the mountains of ancient Brazil.

If in the foregoing pages I have omitted to notice and discuss
the views of Professor Engler, expressed in his important work
‘Versuch einer Entwicklungsgeschichte der Pflanzenwelt,' it is
because I felt that such a discussion would outgrow the proper
limits of the present paper. Agreeing in several of Professor
Engler’s hypothetical views, and entirely dissenting from others, I
could not do justice either to him or to myself without an amount
of detail which would be altogether inappropriate.

28 MR. J. BALL ON THE FLORA

List of Plants collected in the Upper Valley of the Rimac
in the Peruvian Andes.

RANUNCULACEX.

CLEMATIS PERUVIANA, DO., Syst. Nat. i. 141, var. ANDINA, nob.

Differt a typo sepalis longioribus acutioribus, foliis magis dis-
sectis. Chicla!

The description of this species above cited is fuller and more
accurate than that subsequently given in the * Prodromus, but
does not represent the ordinary form. It was drawn up from
specimens in the Paris Museum collected by Dombey in Peru,
and would appear to have much smaller flowers than I find in the
numerous specimens existing in the Kew Herbarium. DeCan-
dolle's words are, * sepala patentissima ovato-oblonga acuta sta-
minibus duplo fere breviora ;" whereas, in what I presume to be
the ordinary form of the species, the sepals are about once and a
half the length of the stamens, and are often somewhat reflexed
when in full lower. The Chicla plant, which apparently inhabits
a higher zone than any of the other specimens I have seen, is
extremely ornamental, the large crowded flowers, of a delicate
sulphur-yellow, often completely concealing the bushes over
which the plant climbs. The leaves and petioles are slightly
pubescent, but the broadly ovate, entire, or slightly toothed bracts,
2 or 4 in number, which form a detached involucre about the
middle of the one-flowered peduncle, are covered with soft villous
hairs. The leaves are much divided, bipinnatipartite, with stalked
three-lobed pinnules.

TuanicrRUM PODOCARPUM, H. B.A. -Chicla! This recalls the
appearance of the rare Pyrenean T. macrocarpum.

RANUNCULUS CHICLENSIS, n. Sp. Chicla, in moist ground!

Bipedalis et ultra, indumento totius plant: e pilis subadpressis
hirtulo; radice e fibris validis subinerassatis; foliis radicalibus et
eaulinis inferioribus pinnatisectis; pinnis alternis trisectis vel
pinnatifidis, superioribus oblongo-cuneatis varie incisis; floribus
majusculis longe peduneulatis; petalis 5-10 angustis oblongo-
cuneatis, apice truncatis vel leviter emarginatis, lineis tenuibus
obseuris notatis; acheniis compressis acute carinatis, in rostrum
longiusculum, sæpe nee semper recurvum, sensim angustatis.

A very distinct species, especially remarkable for the form of
the petals.

OF THE PERUVIAN ANDES. 29

CRUCIFERZ.

SISYMBRIUM MYRIOPHYLLUM, H. B. K. Valley of the Rimac,
ascending from about the level of 9000 feet to Chicla !

SISYMBRIUM TITIOACENSE, Walp. Rel. Meyen. in Nov. Act.
Acad. Leop.-Carol. xix., Suppl. 1, p. 249. Above Casapalta,
about 14,300 feet! "The stems are 1-14 foot long, prostrate or
suberect, branching from below. The length of the siliqua seems
to be variable in this group, and I doubt whether S. athrocarpum,
A. Gray, U. S. Expl. Exped. i. 59, tab. 3, can be maintained as a
distinct species.

SISYMBRIUM HIRSUTUM, Planch. et Triana, DC. (sub 'Turrite).
Chicla! Upper Rimac valley, about 10,000 feet !

DRABA srpLiQUOSA, Hook. fil. Chicla! Above Casapalta!

CAPSELLA Bursa-pastoris, DC. In many places about Chicla!
This is one of the species that seem to owe their wide-spread dif-
fusion in most regions of the earth to the agency of animals, and
at the same time to their capability of adaptation to very different
physical conditions. I have met with it in stations to which the
seeds must apparently have been carried by birds; but for its
original introduction in some remote regions it may be presumed
that the unintentional interference of man is responsible. Such
an instance is to be found in the Straits of Magellan, where 1
found it rather common and still flowering in midwinter about
Sandy Point.

LEPIDIUM vrRGINIANUM, L. Upper valley of the Rimac!

LEPIDIUM AFFINE, Wedd. in Ann. Sc. Nat. ser. 5, i. 284, var.
Chicla !

Lerrpium Humsoxrpti, DC. Chicla!

CREMOLOBUS PINNATIFIDUS, Hook. Ic. Pl. 100,— C. parviflorus,
Wedd. in Am. Sc. Nat. ser. 5,1. 283,=C. aphanopterus, A. Gray,
U. S. Expl. Exped. i. 55. Chicla! The breadth of the wing of
the silicule in this and the allied species is extremely variable,
and without a series of connecting forms the identity of the above-
named species might well be doubted. Some of my specimens
are exactly A. Gray’s C. aphanopterus, while others can scarcely
be distinguished from Hooker’s type specimens of C. pinnatifidus.
I may here point out the necessity for suppressing the name of

30 MR. J. BALL ON THE FLORA

another species in this genus. Oremolobus rhomboideus, Hook.
Ic. Pl. 32, and C. sinuatus, Hook. Ic. Pl. 99, are both founded on
specimens sent by Jameson from the Cuesta de Purruchuco in
Peru, and an examination of the original type shows clearly that
they are forms of the same species. ©. sinuatus is founded on a
depauperate specimen in which the silicules are quite wingless.

CAPPARIDEX.

CLEOME CHILENSIS, DC. Upper Rimac valley, near Matucana,
R. Ward.

POLYGALER.

MONNINA ORTUSIFOLIA, H. B. K., var.? Chicla! This plant
belongs to a group which demands the careful study of all the
available materials. My plant appears to be exactly the same as
two unnamed specimens in Herb. Kew collected by Mathews in
Peru near Chachapoyas, and numbered 1517 and 1518. To the
same species must also, I think, be referred his number 3024,
also from Chachapoyas. The latter is attached to the same sheet
with a specimen labelled Monnina pubescens, H. B. K.?, but
cannot be referred to that species. My plant is evidently nearly
related to M. obtusifolia, H. B. K., and M. crassifolia, H. B. K.;
and I provisionally rank it as a variety of the former.

CARYOPHYLLEA.

LYCHNIS CHILENSIS, Naud. in C. Gay, Fl. Chil. i. p. 256, var.
PERUVIANA, nob. Chicla! This belongs to a group of forms, all
nearly allied to the Arctic L. apetala, L., which extend in
S. America from the equator to the Straits of Magellan. From
a cursory examination of numerous specimens, I am led to con-
clude that they may best be reduced to three species, as follows :—

1. L.thysanodes, Hook. fil., Fl. Antarct. ii. p. 246, — Silene thy-
sanodes, Fenzl., in Endl. Nov. Stirp. Dec. No. 39, — Silene andicola,
Gill., in Hook. Bot. Mise. iii. p. 147. Melandryum cucubalodes,
Fenzl., quoted by Grisebach (Pl. Lorenz. p. 28), probably belongs
also to this species ; but I am unable to 1dentify it with certainty,

This has relatively broad leaves ; the stems usually have many
flowers, which are drooping before fertilization. The calyx is
rather less deeply cleft than in the following species, and the
outer nerves of each of the adnate segments forming the calyx-

OF THE PERUVIAN ANDES. 381

tube are confluent about halfway. It has been found at several
stations in the Andes from Ecuador to Central Chili.

2. L. chilensis, Naud. l. c., =L. terminalis, Naud., in Fl. Chil.
i. p. 255,— L. nubigena, Phil. MSS. This has very narrow stem-
leaves, stems usually simple, one-flowered, calyx divided to about
half its length, the outer nerves of each segment confluent near
the base, the alternate stamens adnate to the base of the petals,
the filaments ciliated, and angular seeds. To this I refer many
Chilian specimens, as well as mine fram Chicla, and No. 989 of
Mandon's * Bolivian Plants," although the latter are in some
respects intermediate. This approaches more nearly to L. ape-
tala, L., than either of its allies.

3. L. magellanica, Lam. Distinguished by its very narrow stiff
leaves, incurved at the edges when the plant is mature, and the
comparatively large flower.

CERASTIUM GLOMERATUM, Thwil. Forma sepalis angustis
acutis. Chicla; perhaps introduced !

It is, I think, time for botanists to free themselves from the
servitude of adhering to a Linnean name when the object to
whieh the name properly belongs is involved in confusion, and
the practice serves only to prolong profitless controversy. Every
one knows what is meant by Cerastium glomeratum, Thuil., but
who is quite sure as to C. vulgatum, L.?

STELLARIA LEPTOPETALA, Benth. Pl. Hartw. p. 163. Chicla!

STELLARIA MEDIA, Vill. Frequent about Chicla! I much
doubt the presence of this species in distant regions being due to
the agency of man.

ARENARIA ALPAMARCE, Á. Gray, Bot. U. S. Expl. Exped. i.
p.114. Above Casapalta! This very much resembles Coloban-
thus subulatus, Hook. fil.

DRYMARIA NITIDA, n. sp. Chicla!

Herba annua, pumila, glaberrima, e basi ramosa, ramis sub-
erectis 3-6-pollicaribus; foliis late elliptico-ovatis, apice acutis
vel brevissime euspidatis, sessilibus, integerrimis, subcarnosis ;
stipulis minimis, setaceis, fugacissimis; cymis terminalibus,
parvis, densis; floribus breviter pedicellatis, virgineis nutanti-
bus; pedicellorum bracteis parvis, ovatis, scariosis; sepalis late
ovatis, obtusiuseulis, coneavis, eveniis, demum purpurascentibus ;
petalorum laminis in unguem tenuem angustatis, profunde fissis ;

32 MR. J. BALL ON THE FLORA

capsulis subglobosis, oligospermis. Semina in exemplaribus
nostris immatura.

I have not seen authentic specimens of D. divaricata, H. B. K.;
and from the description in the ‘ Synopsis Plantarum ZEquinoctia-
lium’ one is led to suppose that it is nearly allied to the common
D. cordata, Willd. But the species is said toinhabit the coast of Peru
near Lima; and specimens in Kew Herbarium ofa plant differing
widely from D. cordata, collected near Lima and at Lurin in the
same region, have been (probably correctly) named D. divaricata,
H.B.K. This is the nearest ally of my Andean plant; but is
more than twice as large, with laxer inflorescence, acute sepals,
and the petals much larger and broader and less deeply divided.

D. glaberrima, Bartl., in Rel. Henk. i. p. 7, differs from
D. nitida, among other particulars, in having petiolate leaves, and
the cymes of flowers on long axillary peduncles. Specimens col-
lected by Mathews “in moist places, Huamantanga,” in Peru,
were, I think, correctly referred by Planchon (in Herb. Kew) to
D. glaberrima, though differing in some respects from Bartling’s
detailed description. The Galapagos plant, referred by Hook. fil.,
in Trans. Linn. Soc. xx., to D. glaberrima, differs more widely
from Bartling's description.

DryMaria LEPTOCLADOS, Hemsl. Diagn. Pl. Nov.i.p.2; Biol. |
Centr.-Amer. i. p. 74, tab. 3, var. peruviana, nob.? Chicla! A
minute plant scarcely 3 inches high, with euspidate leaves ending
in a sharp point, scarious bracts tapering to a long acuminate
point, with a green midrib, and a few subsessile flowers in a con-
gested cyme. It is certainly nearly allied to the Mexican species
above quoted, to which I think must also be referred No. 950,
Mandon’s ‘ Plants of Bolivia.’

PoRTULACACES.
CALANDRINIA CAULESCENS, H. B. K. Chicla!

CALANDRINIA P Chicla? Possibly a dwarf variety of the
last, or an undescribed species. It is less than half the size of
C. caulescens; and I did not notice any intermediate forms; but
I have not detected any further differences.

CALANDRINIA ACAULIS, H. B. K. Above Casapalta !
MALVACER.

MALYASTRUM LIMENSE=Malva limensis, Z. Chicla! Alsoin
the upper valley of the Rimac !

OF THE PERUVIAN ANDES. 33

MALVASTRUM PERUVIANUM, Á. Gray, Bot. U. S. Expl. Exped.
i. p. 146. Frequent in the upper Rimac valley, ascending to
Chicla !

MALYASTRUM CAPITATUM, Griseb. Pl. Lorentz. p. 42,=Malva
eapitata, Cav. Upper valley of the Rimac ascending to Chicla!

Matvastrum ——? [I collected in the upper valley of the
Rimac, at between 9000 and 11,000 feet above the sea, two plants
which, though differing in some respects, must be referred to the
same species. The flowers are in dense glomerules on axillary
or terminal peduncles, with no indication of the secund inflores-
cence so characteristic of M. peruvianum. The leaves are three-
lobed, and the lobes more deeply and sharply toothed than is usual
in that species. As Grisebach has remarked, M. peruvianum is
an eminently variable species; and probably my specimens
should be referred to it. However that may be, I am inclined to
think that a plant in the Kew Herbarium labelled “ Malvastrum,
n. sp., No. 90, Parry and Palmer, * Plants of Central Mexico, ”
is not distinct from my plant.

It is remarkable that although I collected six species of Sida in
the lower valley of the Rimac, none of these were seen to ascend
above the level of 6000 feet.

GERANIACEX.

GERANIUM SESSILIFLORUM, Cav. Chicla! A dwarfform, much
crowded, with short petioles, above Casapalta !

ERODIUM MALACOIDES, Willd. Rimac valley above Matucana,
about 9000 feet! Doubtless introduced.

ERODIUM CICUTARIUM, Willd. Extremely common in the
upper valley of the Rimae and about Chicla, ascending to the
level of at least 12,500 feet !

The wide extension of this species in South America is very
remarkable, especially as it has not shown the same readiness
to establish itself in North America. In Chili, in Argentaria,
and in North Patagonia it has spread throughout the low
country ; and it has been observed that it appears to follow
domestie animals rather than man, not being seen at new stations
in the interior until eattle have been driven there. Its fre-

LINN. JOURN.—BOTANY, VOL. XXII. D

34 MR. J. BALL ON THE FLORA

quency in the Peruvian Andes, or at least in the district visited
by me, is less easy to account for. So far as I know, it has
not been found on the coast, or in the lower valley of the
Rimac. Doubless many articles from the United States, as
well as provisions from the neighbourhood of Lima, were carried
up during the construction of the railway; but this does not
much help us to explain the appearance of the plant. If we
conjecture that seeds may have been carried in hay from Chili
for the use of the Chilian force stationed at Chicla, it is scarcely
conceivable that in a single year the plant could have spread
as it has done, ascending steep rocky slopes inaccessible to
horses.

TROPHOLUM TUBEROSUM, Ruiz d Pav. About the Puente
Infernillo, 11,000 feet! Chicla! One of the ornaments of the
Andean flora; formerly cultivated by the Indians.

BALBISIA VERTICILLATA, Cav. Rimac valley above Matucana ;
flowering in May. Coll. R. Ward.

OXALIS FILIFORMIS, H. B. K. Chicla!

OXALIS PUBESCENS, H. B. K., var. glabra ? I found this Oxalis
near Chicla, and also near the Puente Infernillo, about 11,000
feet above the sea! The latter is a larger plant, with erect
stems branching from the base; but except in being quite
glabrous, I observe no characters by which to distinguish them
from O. pubescens. But the South-American species of Oxalis
are extremely embarrassing to the systematie botanist, very many
species having been described of which specimens are not to be
found in the chief European herbaria. The group with fleshy
stems and leaves, to which O. pubescens belongs, is the more diffi-
cult that in drying the plant falls to pieces, as I found to be the
case both with this and the following species.

OXALIS P Common in the crevices of rocks about Chicla!
l have seen nothing in herbaria that seems to be nearly allied to
this plant, but am unwilling to describe it as new. The rather
large golden flowers are solitary, on short peduncles, which, like
the leaves, spring from the crown of the thick fleshy root (or
rhizome), which extends deep into the crevices of the rocks.

OF THE PERUVIAN ANDES. 35

LEGUMINOSZ.

LUPINUS PANICULATUS, Desr., var. ARGENTEUS. This is the
dominant social species about Chicla, covering the steep slopes in
many places up to a level of about 12,500 feet !

LUPINUS MICROPHYLLUS, Desr., var. ARGYREIUS, 200.? Above
Casapalta, about 14,300 feet! A dwarf plant, with decumbent
branches, 3 or 4 inches in length, the leaves, and especially the
calyx and braets, covered with silky hairs, the petiole twiee or
thrice as long as the seven narrow leaflets. The immature
legume appears to be broader and less pointed than in the true
L. microphyllus. The same plant was collected by Mathews at
Chachapoyas, bearing his number 1570, and may perhaps deserve
to rank as a separate species.

MEDICAGO nENTICULATA, Willd. In the upper Rimac valley,
between 9000 and 10,000 feet; not uncommon, but doubtless
introduced !

Mxzrirorvs INDICA, All.=M. parviflora, Desf. In the valley of
the Rimae, ascending above the level of 9000 feet. Doubtless
introduced!

TRIFOLIUM AMABILE, H. B. K. Chicla! This differs a little
from the ordinary form in having the petioles hairy, and the
peduncles clothed with short adpressed hairs.

TRIFOLIUM AMABILE, H. B. K., var. PENTLANDI,200.? Chicla!

Differt a typo hirsutia totius plantz, ramis brevibus prostratis,
peduneulis axillaribus brevissimis, calycis dentibus brevioribus,
tubo quam in typo longiore.

The same plant was collected by Pentland near the lake of
Titicaca. It may be the same as 7. peruvianum, Vog. Walp.
Rep. i. p. 644, of which I have seen no authentie specimen, or
may rank as a distinct species; but provisionally I prefer to place
it as a variety of T. amabile.

TRIFOLIUM CHICLENSE, n. sp. Found sparingly near Chicla!
Perenne, e collo radicis crass lignose ramos plurimos breves
prostratos edens; foliis apicem versus ramorum confertis, longi-
uscule petiolatis, petiolis sparsim pilosis, foliolis glabris late
obovato-cuneatis argute serrulatis; stipulis lineari-lanceolatis,
acuminatis; capitulis 4—8-floris, subsessilibus, versus basin ramo-
D2

36 MR. J. BALL ON THE FLORA

rum approximatis ; calycis dentibus triangulari-lanceolatis, tubo
brevi longioribus; vexillo brevi calycem parum superante ;
legumine dispermo.

A T. amabili, lorum structura affini, differt imprimis calycis
tubo dimidio breviore et stipulis angustioribus; habitu omnino

diversum, capitulis ad basin ramorum glomeratis, refert potius
T. suffocatum, L.

Darra onopaycuis, DC. Upper valley of the Rimac 8000-
10,000 feet! Scarcely distinct as a species from D. AMutisii
Kunth.

AsTRAGALUS Pickerinen, A. Gray, Bot. U. S. Expl. Exped.
p. 415, var. SERPENS, nob. Above Casapalta !

Rhizomate tenui, ramoso, caudiculos humifusos longe serpentes
sepius subterraneos, stipulis ultra medium coalitis vaginantibus
instruetos, demum sub diu foliiferos, edente; foliis confertis,
cano-villosis, 10-15-jugis; foliolis oblongis, obtusis, truncatis;
pedunculis axillaribus, paucifloris, petiolis vix equilongis ; calycis
tubo longiusculo, quam dentes subequales lineares obtusiusculi
duplo longiore; petalis ochroleucis, vexillo calycem vix duplo
superante ; legumine — ?

A small specimen in Kew Herbarium from Wilkes's Exploring
Expedition shows something of the same mode of growth, but
much less marked than in the Casapalta specimens.

ASTRAGALUS CASAPALTENSIS, n. Sp. Above Casapalta. about
14,300 feet !

Pygmeus, rhizomate lignoso; ramis plurimis brevissimis pro-
stratis, stipulis fere usque apicem coalitis obtectis, deinde foliiferis
et florigeris ; foliis novellis sericeis, mox glabrescentibus, 10-14-
jugis; foliolis late oblongis ovalibusve, obtusis, sæpius truncatis;
floribus axillaribus, solitariis vel geminis, subsessilibus ; calycis
pilosi tubo campanulato, dentibus oblongo-lanceolatis, obtusi-
usculis, :quilongis ; vexillo lete purpureo calyce bis longiore;
ovario piloso, 4-ovulato.

A very distinct species, with smaller flowers than any other of
this group that I have seen. A. Alpamarce, A. Gray, Bot. U.S.
Expl. Exped., has fewer leaflets, and differs in the structure of
the calyx, which has sharp triangular teeth, and in the two-
celled ovary, which bas only one ovule in each cell. A. uniflorus,
DC., which has flowers twice as large, differs more widely from
the Casapalta plant.

OF THE PERUVIAN ANDES. 37

RosaceEz.

ACHNA LAPPACEA, Ruiz 4$ Pav. Chicla!

ACXENA ——? Chicla! A plant doubtless belonging to this
genus, but without flower or fruit.

ALCHEMILLA PINNATA, Ruiz § Pav. Chicla! Casapalta !

SAXIFRAGACES.

SAXIFRAGA CORDILLERARUM, Presl, var. DECIPIENS, nob.
Chicla !

Tenuiter glanduloso-hirtella, foliis caudiculorum latis trifidis,
lobis exterioribus sepius bifidis, nervis lateralibus furcatis, sca-
porum florentium foliis integris bracteiformibus.

SAXIFRAGA CORDILLERARUM, Presl, var. NIVALIS, nob. Above
Casapalta !

Pygmea, foliis omnibus subequaliter trifidis, lobis oblongo-
linearibus, dense glanduloso-pubescentibus.

In spite of considerable differences, I have no hesitation in
regarding these as varieties of the same species, which, as Engler
has shown, is an eminently variable one, not widely differing
from the European S. decipiens, Ehrh., which it represents in the
southern hemisphere.

CRASSULACEZX.

TILLÆA connata, Ruiz d Pav.— T. rufescens, H. B.K. Com-
mon about Chicla! The flowers are tetramerous, the leaves more
or less completely connate.

CoTYLEDON INCANUM, n. sp. Chicla!

Caule erecto, 6-8-pollieari, inferne ramoso; ramis lateralibus
brevibus crebre foliatis, foliis ovali-cylindricis teretibus; floribus
8-12 in racemum brevem dispositis, erectis, superioribus sessili-
bus, inferioribus breviter pedicellatis, pedicellis basi bracteatis ;
sepalis ovali-lanceolatis, viridibus margine pallidis ; corolla sub-
tubulosa; petalis ochroleucis vel pallide rubentibus, lineari-
lanceolatis, acutis, quam sepala sesquilongioribus, usque ad
tertiam vel dimidiam partem coalitis.

A very distinet species, with the leaves and habit of a Sedum
rather than those of the characteristie American group of Eche-
verie, with which, nevertheless, it must be united.

38 MR. J. BALL ON THE FLORA

CoTYLEDON CHICLENSIS, n. sp. Chicla!

Radice crassa, fusiformi, pluries furcata; foliis radicalibus
dense rosulatis, glaberrimis, lineari-lanceolatis, acutis, 6-8 pollices
longis, crasse succulentis; caule erecto, 12—18-pollicari, foliaceo ;
foliis caulinis numerosis, oblongo-lanceolatis, sessilibus, basi calca-
ratis ; floribus omnibus pedicellatis, erectis vel demum nutantibus,
pedicellis inferioribus 6-8-linearibus, superioribus sensim brevio-
ribus, bracteis pedicello subequilongis ; calycis segmentis ovatis,
obscure virentibus ; petalis coccineis, lineari-oblongis, acuminatis,
calycis longitudinem ter superantibus, basin versus usque ad
quartam partem coalitis. Carpellis siccis (emarcidis) sensim
acutatis, subrecurvis, divaricatis.

An ornamental plant, with a general resemblance to several of
the cultivated species of the Echeveria group; but I find it
impossible to identify it with any of those hitherto described.

SEDUM ANDINUM, n. Sp. Chicla!

Planta pygmea, cespitosa, e basi ramosa; ramis numerosis,
tenuibus, 2-3-polliearibus, florigeris erectiuseulis; foliis hemi-
sphericis minutis; cyma subquadriflora, flore terminali sessili,
lateralibus breviter pedicellatis ; sepalis ovato-elliptieis, dimidiam
longitudinem petalorum :zequantibus vel paullo superantibus;
petalis oblongo-lanceolatis, obtusiusculis, atrorubentibus, ima basi
eoalitis. Stamina 10.

Very distinct from any Mexican or other American species
hitherto described; and the first Sedwm, so far as I know, that
has been found in South America.

ONAGRARIEX.

(ExoTHERA MULTICAULIS, Ruiz ( Pav., var. PETIOLARIS, nob.
Chicla !

Foliis radicalibus et caulinis inferioribus in petiolum attenuatis.

CENOTHERA PSYCHROPHILA, n. sp. Chicla!

E basi ramosa ; ramis adscendentibus glabris vel superne pube-
rulis; folis ovato-lanceolatis, in petiolum sensim attenuatis,
margine subintegerrimis, vix aut ne vix obscure denticulatis,
glabris ; calycis tubo incano, sub anthesi supra medium constricto,
laciniis liberis, petalis saturate roseis subequilongis; capsula

OF THE PERUVIAN ANDES. 39

(haud omnino matura) sessili, clavata, striata, suleata. Affinis
Œ. prostrate, Cav., a qua differt foliis subintegerrimis petiolatis
glabris, et calyce fructifero duplo breviori; ab Œ. epilobiifolia
longius distat florum colore, capsula sessili, non tetraptera,
ceeterisque notis.

(ExorHERA ROSEA, Ait. Upper valley of the Rimac, 8000-
10,000 feet! A form with leaves broader and more numerous
than usual.

(ExoruERA ? Above Casapalta, about 14,300 feet. A
mere imperfect fragment of a dwarf species of this genus.

LOASACER.

MENTZELIA HISPIDA, Willd. Rimac valley, above Matucana,
R. Ward.

Loasa XANTHIIFOLIA, Juss.? Chicla! A single dwarf spe-
cimen of a seedling with two small flowers; rather uncertain.

Loasa GRANDIFLORA, Juss. Ann. Mus. Par. v. p. 26, tab. 4;
Lam. Dict. Chicla! All the leaves in my specimens are alter-
nate, the flowers of a deep orange. The long hairs which clothe
the entire plant are extremely stiff, easily penetrating thick
gloves.

BLUMENBACHIA CARDUIFOLIA = Loasa carduifolia, Presl. Com-
mon about Chicla! Puente Infernillo, R. Ward. The slender
but rather stiff stems twine, although in this district the plant
ordinarily grows in positions where it can obtain no support by
so doing.

PASSIFLORACEÆ.

MALESHERBIA THYRSIFLORA, Ruiz § Pav. Above Matucana,
flowering in May, collected by Mr. R. Ward. A curious species,
very unlike the Chilian plants formerly ranked under Malesherbia,
but now united to Gynopleura in Benth. & Hook. f. ‘Gen. Plant.’

CUCURBITACEX.

Srcyos SUBCORYMBOSUS, Cogn. in DC. Monog. Phan. iii. p. 887.
Chiela, on rocky ground, climbing over low bushes!

40 MR. J. BALL ON THE FLORA

DATISCEX.

Damrsca GLOMERATA, Benth. 4 Hook. fil.=Tricerastes glo-
merata, Presl, A. DC. Prod. xv. pars 1, 411. Above Matucana
in the valley of the Rimac, at above 9000 feet! This may possibly
have been introduced during the construction of the railway,
though such an accident is no way probable. It appears to have
its proper home in Mexico, whence it has spread northward to
S. California.

UNMBELLIFERX.

Bow esta PALMATA, Ruiz § Pav.? Chicla! Upper valley
of the Rimac, about 10,000 feet! The group of species to which
this and the following Bowlesie belong appears to be extremely
variable; and I much doubt the constancy of the characters on
whieh several species are founded, nor do I feel confident that
those found by me in Peru are here correctly referred to the
species of the authors quoted.

This is a comparatively large plant, from 12 to 18 inches in
height, with numerous suberect branches; the hairs are stellate;
the lower axillary umbels are shortly pedaneulate, but the ter-
minal umbel is usually on a long peduncle.

Bow testa ACUTANGULA, Benth. Pl. Hartw. p. 186. Chicla!
This has a feeble little branched stem, rarely erect, beset with
long simple hairs, the leaves thin, with very acute angles to the
lobes and incisions, the umbels on iong peduncles, the fruit beset
with stiff stellate hairs.

Bow1esta LOBATA, Ruiz (d Pav., var. CHILENSIS, nob.?
Chicla! This has suberect branches, with stellate hairs, the
leaves three-lobed, with the middle and longer lobe oblong acute,
and the lateral lobes entire or slightly toothed, the umbels on

peduncles equalling the length of the petioles, the hairs of the
fruit glochidiate.

Bow testa LOBATA, Ruiz & Pav., var. uumirusa, nob.? Chicla!
A small depressed plant with numerous short prostrate branches,
the leaves divided into five equal lobes, the umbels subsessile, or
on short peduncles. It is with much doubt that I refer this and
the preceding plant to forms of B. lobata, The typical plant

OF THE PERUVIAN ANDER. 41

differs from both in not having glochidiate hairs on the fruit, and
from the latter, among other points, by the umbels being on
moderately long peduncles. Other differences may be pointed
out between my specimens and B. incana, Ruiz & Pav., and B.
pulchella, Wedd.; but I am inclined to think that both of these
will also be ranked as varieties of B. lobata.

OREOMYRRHIS ANDICOLA, Endl. Gen. p. 787 *. Chicla!

APIUM LEPTOPHYLLUM, F. Muell.= Helosciadium leptophyllum,
DC.=Ammi leptophyllum, Z. Upper valley of the Rimac near
Matucana !

DAUCUS roRILIOIDES, DC. Upper valley of the Rimac, about
10,000 feet !

CAPRIFOLIACER.

SAMBUCUS PERUVIANA, H. B. K. Hereand there in the upper
valley of the Rimac, extending to a height of above 13,000 feet,
but possibly planted at the higher stations! It forms a tree
from 20 to 30 feet in height, and about a foot in diameter.

* [have here blindly followed the authority of Endlicher, sanctioned by that
of the authors of the * Genera Plantarum ; but I must remark that in this case
the first has committed, and the second have sanctioned, an error in nomencla-
ture, having apparently overlooked a note in the fourth volume of De Candolle's
* Prodromus,' p. 229, wherein the facts are clearly stated. "The Andean plant,
which is the representative of a group of forms peculiar to thesouthern hemi-
sphere, was described by Kunth in the fifth volume of the great work of Hum-
boldt and Bonpland, * Nova Genera et Species,’ &c., which was published, in 1821,
under the name Myrrhis andicola. In the same year Lagasca, rightly regarding
the plant as the type of a new genus, published his description of the genus Cal-
dasia, and in a letter to A. P. de Candolle, published by the latter in the col-
lection of his botanical dissertations, named the plant Caldasia andicola. In
preparing his great work on the genera of plants, Endlicher found that, long
before Lagasca, Willdenow had given the name Ca/dasia to a Mexican plant
forming a distinct genus of Polemoniacem. Concluding that Lagasca's genus
Caldasia could not be retained, he gave to it the new name Oreomyrrhis. But
in so doing he overlooked the fact, pointed out by De Candolle, that Willdenow's
genus Caldasia was identical with Bonplandia of Cavanilles, published seven
years earlier; and that in consequence the generic name Ca/dasia had fallen
tothe ground when Lagasca applied it to the plant from the Andes, which should,
according to the established rules of nomenclature, be termed Caldasia andicola,
‘Lag. The priority of Cavanilles’s generic name Bonplandia over Willdenow’s
Caldasia is recognized by Bentham and Hooker in Gen. Plant. ii. p. 824.

42 MR. J. BALL ON THE FLORA

RUBIACER.

MALLOSTOMA THYMIFOLTA, Benth. et Hook. fil= Hedyotis thy-
mifolia, Ruiz & Pav. Upper valley of the Rimac above Matu-
cana, R.Ward. Inthe single specimen collected by Mr. Ward the
segments of the calyx are larger than in the specimens preserved
in Kew Herbarium; but in this group, as in the European genus
Asperula, the length of the calyx-segments, as well as that of the
corolla-tube, varies much in forms of the same species.

RELBUNIUM TUBERCULOSUM=Rubia tuberculosa, Benth. Pl.
Hartw. p. 195,=Galium Benthamianum, Walp. Rep. vi. p. 17.
Above Casapalta, about 14,300 feet! This cannot be separated
from the genus Relbunium. That genus is well characterized by
involuerate flowers and articulate pedicels; but no one of the
species is distinguishable in appearance or habit from some species
of Galium. The present plant might easily be taken for some
Pyrenean or Alpine Galium.

GALIUM Aparine, L. Chicla!

VALERIANEE.

VALERIANA ARMERLEFOLIA, Schlecht. in Lechler, Pl. Peruv.
Above Casapalta! My specimens closely agree with those of
Lechler, named by Sehlechtendahl; but I doubt whether they
are sufficiently different from V. coarctata, Ruiz & Pav., to deserve
separate specific rank.

VALERIANA REMOTA, n. sp. Chicla!

Annua vel biennis, radice incrassata; caule szpius solitario,
simplici, erecto, glaberrimo ; foliis radicalibus lanceolato-spathu-
latis, in petiolum longum sensim attenuatis, subintegerrimis,
preter marginem et nervum dorsalem scabriusculos, glaberrimis,
caulinis paucis ssepius integris, lanceolatis, in petiolum alatum
contractis, interdum lobo brevi angustato utrinque adjecto; in-
florescentia aphylla ex cymis binis oppositis ad quodvis interno-
dium, inferioribus remotis, breviter peduneulatis, laxiusculis,
summis approximatis densis; achenio 2 millim. longo, pappi sex-
radiati setis basin versus membranaceo-alatis, connatis.

A very distinct species, seemingly common about Chicla!
varying in height from 6 inches to 2 feet.

VALERIANA INTERRUPTA, Ruiz 4 Pav. Chicla! <A plant from
2 to 3 feet in height, with much the habit of the European
V. officinalis.

OF THE PERUVIAN ANDES. 43

VALERIANA INTERRUPTA, Ruiz d Pav., var. MINOR, mob.
Chicla! This is very much smaller than the typical species, with
stems from 5 to 6 inches in height ; but I have not observed any
further difference. I did not notice intermediate forms, although
both appear to be rather common about Chicla.

COMPOSITÆ.
EUPATORIUM PERSICIFOLIUM, H. B. K. Chicla!

Evraroriom Barin, Oliv. in Herb. Hook. Ic. Pl. ser. 8, v.
tab. 1462. Chicla! A dwarfform above Casapalta, about 14,300
feet above the sea!

I quote Professor Oliver's description :—'* Fruticosum, ramulis
glanduloso-tomentellis, foliis lineari-ovalibus utrinque attenuatis
obscure crenulatis, margine vernatione plus minus revoluto,
supra rugulosis puberulo-hirtellis, subtus tomentellis areolato-
reticulatis, capitulis multifloris majusculis pedunculatis in cymis
pleiocephalis terminalibus dispositis, bracteis involueri plurise-
riatis, exterioribus ovato-ellipticis acutis striatis parce piloso-
tomentellis, interioribus oblongo- vel lineari-lanceolatis acumi-
natis disco paullo brevioribus, acheniis gracilibus 5-costatis
glabris inferne angustatis, pappo uniseriato setaceo." A shrub
5-8 feet in height, the leaves on very short petioles, the flowering
heads on peduncles that lengthen after fertilization, finally
drooping; belonging to a group of which many forms are found
in the Andes. It is distinguished from Æ. glutinosum, Lam.,
E. salviefolium, H. B. K., E. lavandulefoliun, DC., E. Gaya-
num, Wedd., and E. gynoxioides, Wedd., by the much larger, less
crowded, and drooping capitula; from the first and from several
others by the leaves tapering to the base, and not at all cordate.

EuPATORIUM STrERNBERGIANUM, DC. Common in the upper
valley of the Rimae, growing to a height of several feet! About
Chicla; a reduced form!

EvrATORIUM ——? Puente Infernillo, about 11,000 feet!
Genus uncertain ; too imperfect for determination.

STEVIA ? Chicla! I am unable to identify this plant, of
which I have but an imperfect specimen, with any of the described
species. It is allied to, but certainly different from, S. paniculata,
Lag. The stem is slender, woody, and much branched. The

44. MR. J. BALL ON THE FLORA

leaves are slightly toothed towards the summit, oblong-lanceolate,
the lower opposite, tapering at the base to a very short petiole,
the upper alternate and sessile. The heads of flowers are in à
compact dense corymb.

Conyza SOPHIÆFOLIA, H. B. K. Chicla!

ERIGERON CANADENSIS, L., var.— E. crassicaulis, Sch. Bip.
Chicla! Upper valley of the Rimac! There can, I think, be
no doubt that the form distinguished by Schultz must be referred
as a variety to the cosmopolitan Linnean species.

BACCHARIS GENISTELLOIDES, Pers., var. ALPINA. Above
Casapalta! A dwarf form of a species widely diffused in South
America.

BACCHARIS ALPINA, H. B. K., var. SERPYLLIFOLIA, Wedd. Chl.
And. Above Casapalta!

Baccuaris cHsprtosa, Pers. Above Casapalta! This often
much resembles the last species, but is, I believe, truly distinct.

Baccuaris —— ? Common about Chicla, forming evergreen
bushes from 3 to 6 feet in height. This plant is intermediate in
appearance aud structure between Baccharis obovata, M. B. K.,
and B. Chileo, H. B. K.; but without further observation, I do
not venture to unite those species under a single name. It is
remarkable that, excepting only the moist hot equatorial zone,
the western side of the American continent, from California to
the Straits of Magellan, should almost everywhere produce some
forms of Baccharis belonging to this group differing very slightly
in structure and appearance. Although I have endeavoured to
make myself familiar with their distinctive characters, I doubt
whether if I were to see growing together the bushes which I
have observed in their native homes at Monterey in California, at
Arica in Peru, at Valparaiso in Chili, and at Sandy Point in the
Magellan Straits, I should at once be able to identify them. In
this great and protean genus, so largely developed in South
America, there is room for prolonged study and observation by
the monographer of the future.

ACHYROCLINE SATUREIOIDES, DC. Var. foliis angustissimis
velutinis ? Chicla! This apparently agrees with an unnamed
specimen in Kew Herbarium, collected in New Granada at a
height of about 10,000 feet.

OF THE PERUVIAN ANDES. 45

GNAPHALIUM SPICATUM, Lam., var.— G. sphacelatum, H. B. K.
Chicla !

GNaAPHALIUM Manponu, Sch. Bip. Chicla! Upper valley
of the Rimac! This appears to be a variety of G. oxyphyllum, DC.
Along with many other names given by Schultz Bipontinus, this
appears in a list of Mandon’s “ Plants of Bolivia,” published in
the ‘ Bulletin of the Botanical Society of France,’ vol. xii., without
descriptions. Such a mode of publication is of doubtful validity,
and open to grave objections.

GNAPHALIUM LUTEO-ALBUM, L. Upper valley of the Rimac,
ascending to 10,000 feet !

GNAPHALIUM MELANOSPILEROIDES, Sch. Bip. Chicla!

GNAPHALIUM FRIGIDUM, Wedd. Chl. And. i. p. 146, tab. 24 a.
Above Casapalta!

GNAPHALIUM PIPTOLEPIS, Griseb. Symb. p. L86= Merope pipto-
lepis, Wedd. Chl. And. i. p. 162. Near Chicla!

AMBROSIA ARTEMISLEFOLIA, var. PERUVIANA. Common in the
Rimac valley, ascending above Matucana!

JJEGERIA HIRTA, Less., var. Chicla! Very small and imma-
ture specimens, doubtless referable to this species, which has a
wide range both in area and in altitude.

COREOPSIS FASCICULATA, Willd. Chicia! A form with the
leaves narrower than usual.

Brpens anpicona, H. B.K. Chicla! Some of my specimens
present an unusually luxuriant form of this species, with larger
flowers and stems from 2 to 3 feet in height.

Bipens HumiLIs, H. B. K. Common and very variable about
Chiela, ascending above it to a height of about 14,300 feet!
Some specimens have very narrow leaves, and should perhaps be
distinguished as a variety.

BIDENS MACRANTHUS, Griseb. Pl. Lorentz. p. 188? Upper
valley of the Rimac, about 10,000 feet! My specimens are imma-
ture, but approach most nearly to Grisebach’s plant from the
Andean highlands in the north of Argentaria.

46 MR. J. BALL ON THE FLORA

Brpews PILOSA, L. Common in the Rimac valley, and ascend-
ing to Chicla! It is somewhat remarkable that a species so
common throughout tropical America is able to thrive at a height
exceeding 12,000 feet above the sea.

Bivens cHILENSIS, DC. Upper valley of the Rimac and at
Chicla !

BIDENS BIPINNATA, L. Upper valley of the Rimac and at
Chicla! We can scarcely suppose the three species last enume-
rated to be aboriginal natives of the Peruvian Andes. The
readiness with which the achenes attach themselves accounts for
their wide diffusion by man and by animals.

ScHKUHRIA ISOPAPPA, Benth. Pl. Hartw. p. 205. Near the
Puente Infernillo, about 11,000 feet !

TAGETES MULTIFLORA, H. B. K. Upper valley of the Rimac,
ascending to about 9000 feet !

VILLANOVA OPPOSITIFOLIA, Lag. Chicla! A specimen in
Kew Herbarium is said to have been found near Lima. This is
scarcely probable, unless the plant had been casually carried
down from the Andes by the torrent of the Rimac.

Liasum OVATUM — Paranephelium ovatum, 4. Gr. Probably
a variety of L. (Paranephelium) uniflorum. Above Casapalta!
The large stemless golden-yellow flowering heads are very con-
spicuous.

Senecio Brineesi, Hook. § Arn. Common in the upper
valley of the Rimac! Chicla! I cannot distinguish my specimens
from the ordinary Chilian plant.

SENECIO HYOSERIDIFOLIUs, Wedd. Chl. And.i. p. 108. Chicla!
A very distinct species with large rayless capitula.

Senecio PENTLANDIANUS, DCO., var.—S. adenophylloides, Sch.
Bip. Chicla!

SENECIO GLAREOSUS, Sch. Bip. Chicla!

SENECIO COMOSUS, Sch. Bip. Above Casapalta, about 14,300
feet !

SENECIO GLACIALIS, Wedd. Chl. And. i. p. 118, tab. 18 A.
Above Casapalta !

OF THE PERUVIAN ANDES. 47

SENECIO HUMILLIMUS, Sch. Bip. Var. involucri foliolis latio-
ribus. Above Casapalta !

SENECIO CASAPALTENSIS, n. sp. Above Casapalta!

Pygmeus, subacaulis, rhizomate crasso verticali, fibros validos
laterales emittente; folis omnibus radicalibus rosulatis, ambitu
obverse lanceolatis, profunde lyrato-pinnatifidis, lobis angulato-
subrotundis, supremis confluentibus; capitulo solitario sub-
exscapo, discoideo, majusculo, multifloro, diametro subpollicari ;
involucri foliolis linearibus adpressis, apice acutis.

SENECIO MODESTUS, Wedd. Chl. And. i. p. 105, tab. 18 B.
Above Casapalta! I do not think that S. Mandonianus, Wedd.,
can be maintained as a species distinct from this.

Amid the multitude of stemless plants belonging to Senecio or
nearly allied genera that abound in the upper region, it is very
difficult at once to distinguish the different species, though on
examination many of them display well-marked characters.

WERNERIA STRIGOSISSIMA, Á. Gray, Proc. Amer. Acad. v.
p. 140,2 W. setosa, Wedd. MSS. Above Casapalta!

CENTAUREA MELITENSIS, L. Upper valley of the Rimac,
ascending to about 9000 feet! This plant may probably have
been introduced from Spain at an early period in the history of
her American colonies. It is now widely diffused through most
parts of the former Spanish territory where the climate is not too
moist and hot to permit it.

CHUQUIRAGUA SPINOSA, Don, var. INERMIS. About Casapalta!
Forming erect bushes about 2 feet in height. The lower limit
observed by me was about 13,700 feet above the sea.

Morista vICLEFOLIA, Cav. Chicla!

POLYACHRUS ECHINOPSOIDES, DC. Chicla!

PEREZIA MULTIFLORA, Less. Var. involucri squamis latioribus
apice vix mucronatis. Chicla!

Perezia CILIARIS, Hook. et Arn.=Clarionea ciliaris, Don.
Upper valley of the Rimac above Matucana, R. Ward.

HIERACIUM CHICLENSE, n. sp. Chicla!

Perenne, rhizomate brevi, crasso, verticali; caule elato 2-3-
pedali, ramoso, striato, inferne villoso, superne glabrescente ; foliis

48 MR. J. BALL ON THE FLORA

omnibus plus minusve albo-villosis, radicalibus ante anthesin emar-
cidis, cum caulinis inferioribus in petiolum longum sensim
attenuatis, superioribus paueis sessilibus, oblongo-lanceolatis,
remote repando-dentieulatis, in pagina inferiore nervosis; ramis
adscendentibus, oligocephalis, subcorymbiferis; pedunculis
tomentellis vix glandulosis ; involucri subcylindrici squamis irre-
gulariter imbrieatis, linearibus, acutis, fusco-villosis, intimis
pappo equilongis ; achenio columnari, ecoronato (scil. costis non
prominentibus), pappo niveo.

This is one of the group of Hieracia, of which several are now
known from Central America and the Andes which by their
pure white pappus approach to the genus Crepis. But in America
it is clear that that character cannot be relied upon as of generic
importance, as these plauts are all closely related to species of true
Hieracium. Yn appearance my specimens approach H. tricho-
donta, Sch. Bip. in Linnea, xxxiii. p. 761, from which I think it
impossible to separate H. jubatum, Fr. Vet. Ac. Forh. Stock. 1856,
p. 146, and Epicrisis 147 ; the latter, according to Fries, is the
same as H. frigidum, Wedd. Chl. And. i. p. 225, tab. 428. But
my plant differs in many respects ; e. 7. in being clothed in white
instead of reddish hairs, in the white pappus, and especially in the
form and structure of the involucre, which approaches nearly that
of the cymose group of the genus Pilosella, while H. trichodonta
must be ranked in the subgenus Stenotheca.

HYPOCHÆRIS ELATA, Griseb. Pl. Lorentz. p. 188=Achyro-
phorus elatus, Wedd. Above Chicla in the fissures of rocks!
This species bears at first sight a striking resemblance to the
common 47. radicata.

Hypocr aris SESSILIFLORA, H. B. K. Chicla! This in most
respects nearly approaches Achyrophorus Meyenianus, Walp.,
which must, I think, be united with the older species.

HYPOCHÆRIS P Above Casapalta! This plant, which
exactly agrees with No. 281 of Mandon's * Plants of Bolivia,” is
allied to H. sessiliflora, but can scarcely be united to that species.
The flowering heads are much smaller, and have not nearly so
many florets.

HyPOCILERIS TARAXACOIDES=Oreophila taraxacoides, Meyen

et Walp. Nov. Acta Acad. Leop. xix., Suppl. i. p. 291 = Achyro-
phorus taraxacoides, Wedd. Chlor. And. Above Casapalta !

OF THE PERUVIAN ANDES. 49

CAMPANULACEZ.

WAHLENBERGIA PERUVIANA, A. Gray, Proc. Amer. Acad. v.
p.152. Above Casapalta! A small inconspicuous plant.

A dwarf shrubby plant found at Chicla may possibly belong to
the Vacciniacee; but as I saw neither flower nor fruit, I can
merely express a conjecture.

ASCLEPIADE.

LUGONIA ANDINA, n. sp. On rocky ground about Chicla!

Frutex scandens, ramis tenuibus, volubilibus, 4-5 ped. longis ;
foliis oppositis, breviter petiolatis, ovato- vel oblongo-lanceolatis,
basi eordatis, apice acutis vel acuminatis, superne glabris, in
pagina inferiore tomentellis ; pedunculis 1-2-floris, bractea soli-
taria fuleratis; calycis quinquepartiti segmentis ovato-lanceolatis
puberulis; corolla late campanulata subrotata, sat magna, ealy-
cem ter excedente, diametro 13-2-pollicari, extus puberula, in
lobos quinque rotundatos ultra medium fissa, sordide purpureo-
aurantiaca, tubo atropurpureo; follieulo abortu (an semper?)
solitario, fusiformi, glaberrimo, 3—4 pollices longo.

Two specimens in Kew Herbarium belong clearly to this
hitherto undescribed species, though showing some slight varia-
tions of structure. They were placed by Mr. Bentham in the
same generic cover with Lugonia lysimachioides, Wedd., and
though not named, are referred to in the ‘Genera Plantarum’
(vol. ii. p. 758) as constituting a second species of the genus
Lugonia. Of these specimens, one marked * Peru, Maclean,"
has the leaves rather more deeply cordate, the calyx-segments
longer, and the corolla (apparently) of lighter colour, verging on
orange. The other specimen, a mere fragment from Nuttall's
herbarium, gathered at Charcan (in Peru ?) shows no other dif-
ference from the Chicla plant than being rather more downy on
the surface.

GENTIANACES.

GENTIANA SEDIFOLIA, H. B. K., var. CASAPALTENSIS, nob.
Above Casapalta!

Foliis lineari-lanceolatis basi dilatatis, vaginantibus, muticis ;
scapo brevissimo; corolle lobis intermediis primariis subzqui-
longis, acutis, denticulatis.

LINN. JOURN.—BOTANY, VOL. XXII. E

NMTSECM TII
IVI Aw M7 \ 3

50 MR. J. BALL ON THE FLORA

Specimens of this Gentian, which in appearance much recalls
the G. prostrata of the northern hemisphere, from different loca-
lities in the Andes, show so many varieties of structure, that they
have received many different specific names. Mr. Weddell, who
had ample opportunities for observing the living plants, united
these as varieties of the form first collected by Humboldt and
Bonpland. The Casapalta plant, though evidently belonging to
the same aggregate species, differs so far from the forms enume-
rated by Weddell, that I have found it necessary to name it as &
distinct variety.

GENTIANA VAGINALIS, Griseb. Chicla!

GENTIANA MULTICAULIS, Gill. var.? Upper Rimac valley,
above Matucana, R. Ward. The single specimen is incomplete,
and I do not feel quite confident as to its identity with the
Chilian species.

HALENIA GENTIANOIDES, Wedd. Above Casapalta!

POoLEMONIACEX.
Gitta LACINIATA, Ruiz and Pav. Chicla!

COLLOMIA GRACILIS, Dougl. Chicla! A dwarf form of the
species.

HyDROPHYLLACES.

PHACELIA CIRCINATA, Jacg. Chicla! As is well known, this
species extends throughout the larger part of the American con-
tinent, from Oregon to the Straits of Magellan; throughout that
wide area it is singularly constant, exhibiting no marked varieties.

BonAGINERX.

HELIOTROPIUM PARVIFLORUM, L. Upper valley of the Rimac,
about 9000 feet! This species, whose proper home is in the
lower zone, ascends from Lima along the valley of the Rimac to
some height above Matucana.

HELIOTROPIUM PERUVIANUM, L. Upper valley of the Rimac,
up to about 10,000 feet! This seems to be a native of the middle
zone of the Andes. I did not observe it below the level of 7000
feet, and it becomes rare above that of 9000 feet, although it
appeared to be common between those limits.

OF THE PERUVIAN ANDES. 51

CYNOGLOSSUM REVOLUTUM, Ruiz & Pav. Chicla!

ERnrTRICHIUM LINIFOLIUM, Á. Gray, Proc. Amer. Acad. x. 54
=Anchusa linifolia, Lehm. Asperif. No. 158 (1818)= Anchusa
oppositifolia, H. B. K. Nov. Gen. et Sp. iii. 91, tab. 200— Anti-
phytum linifolium, DC. Prod. x. 122. Chicla! Walpers de-
scribed under the name Anchusa Kunthii (Antiphytum Walpersii,
A. DC. Prod. l. c.) a plant differing from Lehmann's type only
in having leaves “twice or four times as long.” My specimens
vary very much in this respect, and I doubt whether the name
E. Walpersii should be retained even for a variety. Lehmann
founded his description on specimens contained in Willdenow's
herbarium, to which the latter had given, in manuscript, the names
Anchusa linifolia and A. leucantha ; Roemer and Schultes declare
this to have been an error on Lehmann's part, and that instead
of one species with two names, there existed in Willdenow's her-
barium two different species both labelled Anchusa linifolia.

Erirricuium Manpontt, n. sp. Chicla!

Bienne? caule e basi ramoso, ramis adscendentibus, foliaceis,
6-12 poll. longis; foliis cum ramis et calyce strigoso-pilosis,
sessilibus, semiamplexicaulibus, oblongis, apice sensim acutatis ;
floribus subsessilibus in cymam densam congestis, uno alterove
axillari remotiuseulo; calycis segmentis in fructu elongatis,
linearibus ; corolla parva, ochroleuca ; nuculis corrugato-lacunosis.

My specimens, from which the description is taken, agree with
No. 378 of Mandon’s * Plants of Bolivia.’ In appearance the plant
recalls Amsinckia angustifolia, Lehm.

CONVOLVULACER.
DicHoNDRA REPENS, Forst. Chicla!

CUSCUTA CHILENSIS, Ker. Upper Rimac valley, near Matucana.
R. Ward.
SOLANACER.

SotanuM TUBEROSUM, L.? Growing among dense bushes
beside the torrent a few hundred feet above Chicla! I was un-
able to obtain the root of this plant, and I do not feel certain as
to its identity. In spite of the assistance derived from Mr.
Baker’s careful investigation of this group of Solanum, I find it

E2

52 MR. J. BALL ON THE FLORA

very difficult to trace clear lines of demarcation between the
species, even as limited by him. Solanum Maglia in particular
appears to vary very much, and I perceive no constant diagnostic
character by which to recognize it with certainty. Even if the
plant found by me should be a true potato, I do not regard its
occurrence as evidence of S. tuberosum being indigenous in Peru.
The line of communication between the sea-coast and the high-
lands of Peru by the upper valley of the Rimac was doubtless in
frequent use under the rule of the Incas, and we know that the
potato was cultivated by the Peruvians long before the Spanish
conquest. It would, in my opinion, require much more evidence
than we now possess to make it probable that it occurs as a true
native anywhere north of Chili.

SOLANUM ANGUSTIFOLIUM, Ruiz 4 Pav. var. =S. pulverulentum,
Pers. Chicla!

SOLANUM PENTLANDI Dun. Chicla! Upper valley of the
Rimac, 8000 to 10,000 feet. I doubt whether this should be
separated from S. chenopodioides, Lam.

SOLANUM RADICANS, Ruiz d Pav. Upper Rimac valley, from
9000 to 11,000 feet above the sea! S. ruderale, Cl. Gay, Fl.
Chil., seems to me not different from this species.

SoLANUM ———, n. sp.? Upper valley of the Rimac, about
9000 feet !

A Solano rhamnoide, Dun. in DC. Prod. xiii. 160, differt pluri-
bus notis gravioribus :—1. planta tota glabra, nisi nervo medio et
nervis lateralibus primariis (prominentibus, nec vix conspicuis),
in pagina inferiori foliorum fusco-pulverulentis, cum fasciculo
lane tomentose ad insertionem nervorum lateralium ; 2. calycis
segmentis rotundatis, obtusissimis, nec triangularibus, acutis;
3. stigmate clavato, nec obcordato. Planta insignis est ob pedi-
cellos etiam ante anthesim recurvos, pendulos. Segmenta corollina
ochroleuca, trinervia. Bacca magnit. nucis avellanz, fusco-atra.

Amidst the vast erowd of species, many of which I have not
seen, this may possibly have been described, and I think it better
not to coin a new name.

LYCOPERSICUM PERUVIANUM, Mill. Ascends from the coast
along the Rimac valley to a height of 8000 feet!

OF THE PERUVIAN ANDES. 53
HEBECLADUS BIFLORUS, Miers=Atropa bifora, Ruiz ( Pav.
Chicla!

NicANDRA PHYSALOIDES, Gaertn. Upper Rimac valley, be-
tween 8000 and 10,000 feet !

PHYSALIS PERUVIANA, L.=P. edulis, Sm. Upper Rimac valley,
at about 10,000 feet !

NICOTIANA GLUTINOSA, L., var. foliis basi non cordatis, corolla
breviore. About Chicla frequent !

SCROPHULARINEE,
CaLcEOLARIA VIRGATa, Ruiz § Pav. Chicla!
CALCEOLARIA LOBATA, Cav. Chicla!
CALCEOLARIA TENUIS, Benth. Chicla!

CALCEOLARIA ovata, Sm.=C. biflora, Ruiz § Pav., non Lam.
Chicla!

CALCEOLARIA BARTSLEFOLIA, Wedd. Chicla!

ALONSOA ACUTIFOLIA, Ruiz Y Pav. Chicla! A very orna-
mental plant, growing from 2 to 4 feet in height.

Mimvtus araBRATUS, H. B. K. In moist places above Chicla!

OvrIsIa ? Upper valley of the Rimac, near Matucana !
A single and imperfect specimen, evidently allied to O. chame-
drifolia, Benth., but apparently different from that species.

CASTILLEJA FISSIFOLIA, L., fil. var. pumila=C. pumila, Wedd.
Above Casapalta! In his * Chloris Andina,’ vol. ii. pp. 118, 119,
Weddell has united under C. físsifolia no less than eight species
described by various authors, including his own C. pumila, under
which name he had previously described a plant closely agreeing
with my Casapalta specimens. The opinions of such a competent
botanist, who enjoyed such ample opportunities for observation,
are deserving of all respect. It would appear from his remarks
that, in the group of forms here united, the length of the style
and that of the stamens are as subject to variation as are the
length of the calyx and corolla, and the form of the leaves.

Barrsta PATENS, Benth. Above Casapalta!

BARTSIA GRACILIS, Benth. Chicla!

54 MR. J. BALL ON THE FLORA

BIGNONIACER.
Tecoma saMBUCIFOLIA, H. B. K. Upper valley of the Rimac,
near Matucana, R. Ward. Collected in May.

VERBENACE®.

VERBENA LITORALIS, H. B. K. Upper valley of the Rimac!
Chicla! This species, so widely spread in South America, ran-
ging from the coast to a considerable height in the Andes, and
extending from the Northern United States to Chili and Pata-
gonia, is very nearly allied to V. officinalis, and it is often difficult
to distinguish them.

VERBENA HISPIDA, H. B. K. Upper valley of the Rimac,
ascending to 10,000 feet !

VERBENA DIFFUSA, Wedd. Chicla!

LABIATS.

SALVIA RHOMBIFOLIA, Ruiz § Pav. Upper valley of the
Rimac! Chicla!

SALVIA STRICTIFLORA, Hook. Rimac valley, above Matucana,
R. Ward.

SALVIA RUIKSHANKSII, Benth. Rimac valley, above Matucana,
R. Ward.

PLANTAGINEE.

Prantaco Hartweatt, Decne. Chicla! This is nearly allied
to the cosmopolitan P. major, L.

PLANTAGO SERICEA, Ruiz & Pav., var.— P. leucophylla, Decne.
Chicla!

NYCTAGINEX.

COLIGNONIA BIUMBELLATA, n. sp. Upper valley of the Rimae,
near Tamboraque, about 9000 feet above the sea!

Caule erecto, glaberrimo, nodoso; foliis oppositis levissimis,
late deltoideo-ovatis vel rhombeo-orbicularibus, basi truncatis, vix
aut ne vix cordatis; petiolis tenuibus limbi dimidiam vel tertiam
partem longitudine sequantibus; ramis herbaceis, umbellatis,
iterum in umbellam florigeram superne divisis ; umbellz radiis
(vel pedunculis) umbellulam 10-20-floram gerentibus ; staminibus
perigonium subduplo superantibus ; petalis pro genere majusculis ;
fructu nimis immaturo.

Very distinct from the two described species, C. parviflora,
Chois., and C. scandens, Benth.

OF THE PERUVIAN ANDES. 55

AMARANTACEE.

AMARANTUS CHLOROSTACHYS, Willd. Frequent in the valley
of the Rimac, extending up to the level of about 9000 feet!

The Amarantacee, which abound in the lower region of tropical
and extratropical South America, do not seem to ascend to high
levels. "The present species, which is almost universally diffused
throughout the continent, is the only representative of the order
seen by me in the middle region of Peru. A few miles below
Matucana five or six species of this order were seen on the side
of the railway line.

CHENOPODIACER.
CHENOPODIUM AMBROSIOIDES, L. Upper valley of Rimac!
This ascends to about the same level as the last species !

Urrvcus TUBEROSUS, Mog. On rocky ground about Chicla!

PHYTOLACCACER.
PHYTOLACCA OCTANDRA, L. Upper Rimac valley, near Matu-
cana!

PorYGoNaACEZX.

MvuEHLENBECKIA RUPESTRIS, Wedd., var. VULCANICA, nob. =M.
vulcanica, Meisn. in D. C. Prod. Very common about Chicla!
I have no doubt that the two forms described as M. rupestris and
M. vulcanica should be referred to the same species. The former
is distinguished by larger size, more rigid woody branches, larger
leaves, more persistent ochre, and shorter pedicels to the flowers ;
but I found all these marks to be eminently variable in the plant
which abounds at Chicla, and which well agrees with authentic
specimens of M. vulcanica. I doubt whether this can be well
distinguished even as a variety.

The synonymy of these plants has become a little complex,
but JL rupestris was the first correctly referred to the genus
Muehlenbeckia.

M. rupestris, Wedd. in Ann. Sc. Nat. ser. 3, xiii. 256 (1849) =
Polygonum fruticulosum, Walp. Nov. Act. Acad. Leop. xix.
suppl. 1, 407.

M. vulcanica, Meisn. in DC. Prod. xiv. 148 (1856)= Polygonum
voleaniceum, Benth. Pl. Hartw. 81 = Muehlenbeckia volcanica,
Endl. Gen. suppl. 4, pars 2, p. 51.

56 MR. J. BALL ON THE FLORA

SANTALACER.

QuINCHIMALIUM ERICOIDES, A. Brogn. Voy. Cog. tab. 51.
About Chicla! Above Casapalta! This species has been united
by Alph. de Candolle (Prod. xiv. 625) with Q. gracile, A. Brogn.
l e. They appear to me to be quite as distinct as any of the
species of this genus.

URTICACER.

URTICA FLABELLATA, H. B. K. Above Casapalta! To a
European botanist it is somewhat surprising to find a species of
this genus attaining nearly to the extreme limit of phanogamous
vegetation.

PARIETARIA DEBILIS, Forst. Chicla!

GNETACER.

EPHEDIA AMERICANA, H. B. K., var.=E. rupestris, Benth. Pl.
Hartw. 253. Chicla! In the passage cited, Mr. Bentham
pointed out the affinity of this plant with E. americana. With
more ample materials I think he would not have hesitated to
unite them under one specific name.

InrpEEX.
SrsyRINCHIUM JUNCEUM, E. Meyer. Chicla! Casapalta!
SIsYRINCHIUM JUNCEUM, E. Meyer, var. =S. scirpiforme, Poepp.

Fragm. Syn. 2. Chicla! This should perhaps be regarded rather
as a luxuriant state of the typical species than as a true variety.

AÁMARYLLIDEX.

ALSTREMERIA PEREGRINA, L.? Chicla! The solitary speci-
men is very imperfect, and I do not feel confident as to its identity
with the Linnean species.

LirrACEX.

ANTHERICUM ECCREMORHIZON, Ruiz & Pav. Puente Infer-
nillo! A fragment from near Matucana collected by Mr. Ward.

COMMELINACER.

COMMELINA FASCICULATA, Ruiz (f Pav., var.=C. nervosa,
RH. P. Puente Infernillo! Upper valley of the Rimac, near
Matucana, R. Ward.

c
“I

OF THE PERUVIAN ANDES.

JUNCACEE.
LUZULA RACEMOSA, Desv. Chicla!

LUZULA RACEMOSA, Desv., var. HUMILIS = L. humilis, Buchen. in
Abhandl. Nat. Ver. zu Bremen, 1874, 125. Above Casapalta,
about 14,300 feet! This plant, which bears a striking resemblance
to our European Z. spicata, appears to differ only in its smaller
size and more contracted inflorescence from L. racemosa, as the
learned author of the species, to whom we are so much indebted
for our knowledge of the South American Juncacee, appears
finally to hold. See * Kritische Zusammenstellung der bis jetzt
bekannten Juncaceen aus Süd-Amerika, by Franz Buchenau, in
the 6th volume of the Proceedings of the Bremen Natural
History Society.

LUZULA CHILENSIS, Nees et Meyen, var.? Chicla! I found
but a single specimen of this curious plant, which does not closely
agree with any of the described species, although I have been
able to compare it with authentie specimens of nearly all of them.
The stem is about a foot in height, with several broad flat leaves,
villous, especially at the opening of the sheath, and passing
gradually into the bracts at the base of each spike. Of the latter
there are three, few-flowered, lax and interrupted. So far, as
well as in the hexandrous flowers, this agrees well with L. chilensis
of Nees and Meyen; but the segments of the perigone are much
longer and sharper, and dark brown ın colour. The scarious
bracts are very conspicuous, broadly ovate, and tapering to a long
point, twice the length of the flowers, and covered, or at least bor-
dered, with white villose hairs. In examining the South American
specimens of this genus in Kew Herbarium, I have been struck
by the great degree of variation in nearly all the organs, and I
suspect that it may hereafter be necessary to reduce the number
of species. Although they are very unlike in appearance, there
is little difference in structure between Z. chilensis and L. alope-
curus, Desv.

GRAMINEX.

PASPALUM STOLONIFERUM, Bosc, in Trans. Linn. Soc. ii. 83
(1792) — P. racemosum, Jacq. Collect. Suppl. 32 (1796). Upper
valley of the Rimae, at about 9000 feet! This grass is abundant

58 MR. J. BALL ON THE FLORA

about Lima, where it is much valued for fodder, and has doubtless
been carried by accident to the upper part of the valley.

OnYZOPSIS CJSPITOSA, Var. PERUVIANA, 70b.=Nassella cespi-
tosa, Griseb. Pl. Lorentz. 210, var. Common at Chicla and in
the upper valley of the Rimac! This differs from the type in its
larger size, more lax inflorescence, and very short two-eared
ligule. It is in several respects intermediate between the Chilian
species of Nassella and the European Piptathera, which have all
been united under Oryzopsis by Bentham and Hooker. Grisebach
describes the awn as twice or thrice the length of the floret; in
his specimens, as in mine, 1 found it about four times as long.

MUEHLENBERGIA ? Upper valley of the Rimae, about
9000 feet! My specimens, which are only imperfect fragments,
are allied to M. adspersa, Trin., and M. clomena, Beauv., but
differ from both in the extreme shortness of the minute outer
glume.

AGROSTIS NANA, Kunth. Above Casapalta, about 14,300 feet!

POLYPOGON rwrERRUPTUS, H. B. K. Upper valley of the
Rimae, about 9000 feet! My specimens exhibit an unusually
large form of tbis species, which has a wide range in South
America both in area and elevation.

CHATOTROPIS ANDINA, n. Sp. Chicla!

Ceespitosa, multiculmis ; culmis subpedalibus, erectis, foliorum
vaginis levibus striatis vestitis ; foliorum lamina anguste lineari,
plana, preter marginem asperulum glaberrima, panicula breviore;
ligula bifida, oblonga, breviuscula; panicula contracta, densa, sub-
lobata ; floseulis parvis ; glumis exterioribus sterilibus, subæqua-
libus, lanceolato-acuminatis, in carina scabriusculis, cæterum gla-
berrimis, scariosis, szepius purpurascentibus; gluma fertili exteri-
oribus paullo breviore, sub lente pubescente, flosculum involvente,
infra medium arista dorsali ipsam glumam bis superante instructa,
extus ad basin pilis brevibus obliquis munita; rhachilla ultra
florem in setam brevissimam pilosam producta.

It has been extremely difficult to fix the true affinities of the
plant here described, but in all essential respects it agrees with
Chetotropis, of which the only known species appears to be rare
in Chili, but common in the island of Juan Fernandez. This has

OF THE PERUVIAN ANDES. 59

been separated from Agrostis, to which it is in other respects
nearly allied, by having ordinarily, though not always, the rhachilla
prolonged to a short point beyond the base of the solitary floret,
and by the dense crowded panicle of minute florets. In both
respects the Chicla plant agrees, except that the prolongation of
the rhachilla appears to be of constant occurrence; but owing to
the presence of comparatively long dorsal awns (which give the
plant the appearance of a Polypogon), and the longer relative size
of the fertile glume (which is hairy at the base), they are specifi-
cally widely different. The Chicla plant may, in technical charac-
ters, be thought akin to Deyeuxia, from which it chiefly differs in
having the short hairs at the base of the fertile glume set on
obliquely on the outer half, instead of in a complete ring, in the
awn being straight instead of twisted, and in the extreme short-
ness of the prolongation of the rhachilla. But in my opinion its
true affinity is with the group of Andine Agrostis, of which I
have been able to examine A. virescens, H. B. K., and A. foliata,
Hook, fil. These have the panicle nearly as dense as in our plant,
and nearly straight awns. A. foliata, which is the nearest in
appearance, differs in having the outer glumes nearly blunt, not
acuminate, and much longer than the fertile glume. I have
noticed in that species some slight and very short hairs at the
base of the fertile glume, but no trace of a prolongation of the
rhachilla. I venture to think that the true conclusion from the
facts is that the genus Chetotropis should be merged in Agrostis.

DEYEUXIA coARCTATA, H. B. K. Chicla! Either the species
named or a near ally.

DEYEUXIA Manponrana, Wedd. in Bull. Soc. Bot. Fr. xxii.
179, var. Above Casapalta, about 14,300 feet, growing by a
rivulet, with culms two feet in height! This differs from Mandon’s
type specimens in having the culms quite smooth throughout, and
wanting the slight hairiness at the opening of the limb of the
leaf described by Weddell. The leaves of the sterile shoots are
numerous, erect, rigid, tightly rolled into rough wiry cylinders;
the culms have few (two or three) leaves, the sheaths inflated, the
ligule long and acuminate.

DrEYEUXIA , Sp- nov.? Above Casapalta!
Pumila, vix ezspitosa; culmis spithameis, levibus ; foliis gla-
bris, radicalibus brevibus, anguste linearibus, planiusculis, vagiuis

60 MR. J. BALL ON THE FLORA

striatis parum inflatis, limbo brevissimo, canalieutato, margine in-
voluto ; ligula bifida, lata, oblique truncata, culmum amplectente ;
panieula eontraeta, bipollieari; pedicellis pilosiusculis ; glumis
sterilibus subequilongis, seariosis, in carina scabriusculis, gluma
fertili sterilibus dimidio breviore, cartacea, floseulum arcte invol-
vente, pilis basilaribus subduplo longiore, medium versus arista
dorsali torta ipsa gluma duplo longiore munita; palea hyalina,
binervi; rhachilla ultra florem in setam penicillatam glume fertili
equilongam produeta.

B. Weddell in Bull. Soc. Bot. Fr. xxii. p. 176 (sphalmate typo-
graphi, 156) species Andinas generis Deyeuxie 60 enumeravit, et
clavem analyticam adjunxit. Descriptiones fusiores specierum
novarum 37 a Weddelio propositas cel. auctor, morte prematura
abreptus, nunquam publici juris fecit, et plurimarum specimina
in herbaris Anglicis desunt. Etsi planta nostra Casapaltensis
speciei nulli descriptæ convenire videtur, nolui nomen novum
proponere. Caracteres e ligula desumptas, in hoe genere ad
species definiendas utilissimas, in clavi analytica supra citata
auctor pretermisit.

DxscmaMPSIA MarmuEWsIT, n. sp. Above Casapalta!

Ceespitosa, glabra; culmis numerosis, erectis, 3—6-pollicaribus,
superne nudis, basin versus foliorum vaginis striatis subinflatis
vestitis; foliis rigidis, anguste linearibus, subeanaliculatis, muticis,
callo cartilagineo apiculatis; ligula brevi oblongo-ovali; panicule
parve ramis paucifloris; spiculis bifloris; glumis sterilibus sub-
equalibus, ovato-lanceolatis, acutis, in carina scabriuseulis ; gluma
fertili hyalina flosculum involvente; palea oblonga, apice lacera;
rhaehilla glabra ultra flores brevissime produeta.

The same species, with the outer glumes slightly shorter
and less acute, was long ago collected by Mathews in the Peru-
vian Audes. His specimens in Kew Herbarium are numbered
689.

I collected above Casapalta a single specimen of a Deschampsia
differing from the species above described in several respects ; but
all the flowers examined were abortive, and I believe it to repre-
rent merely a diseased condition of the same species.

MeErica ? Chicla! A single incomplete specimen drawn
with some difficulty from amidst a group of thorny bushes. The
flowers much resemble those of M. laxiflora, Cav. ; but the leaves

OF THE PERUVIAN ANDES. 61

are short and glabrous, and the sheaths do not approach the
summit of the culm as they do in that species, which has much
longer leaves, always more or less scabrous.

DissANTHELIUM SUPINUM, Trin. Chicla! A minute species
with culms two inches in height.

ERAGROSTIS TEPHROSANTHOS, Spreng. (sub Poa)? Upper valley
ofthe Rimae! A doubtful species.

The species of the genus Poa have been so much confused, and
the descriptions of many species of which I have not seen
authentie specimens are so imperfect, that, with the exception of
the first, I think it best to designate by numbers the forms which
l eollected in the Peruvian Andes, trusting to be able at some
future time to submit them to the examination of the accom-
plished agrostologist Professor Hackel, who has undertaken the
great task of revising the entire natural order.

Poa annua, L. Chicla! It is an illustration of the remark-
able power of adaptation to different external conditions possessed
by several grasses, that in the course of the same tour I found
this cosmopolitan species flowering in the Andes, in the Straits
of Magellan (in winter), and within the tropies in Brazil.

Poa No.1. Chicla! <A tall grass with culms 3 to 6 feet in
height ; the leaves of the flowering culm striated, nearly smooth
on the surface, but somewhat scabrous at the edge, those of the
barren shoots convolute, the ligule short and rounded at the top;
the panicle is large, widely spreading, the lower branches verti-
cillate, the uppermost solitary, slightly rough.

Poa No.2. Chicla! Culms little more than a foot in height;
the leaves all nearly flat, more scabrous than the last; the panicle
contracted, with short smooth branches of nearly equal length.

Poa No.3. Chicla! Somewhat larger than the last, with the
branches of the panicle scabrous. Probably a variety of the
same species.

Poa No. 4. Chicla! This approaches our No. 1 in the form
of the panicle, but on a much smaller scale, and with the branches
decidedly scabrous. The ligule is oblong and truncate at the top.

62 MR. J. BALL ON THE FLORA

Poa No. 5. Above Casapalta! A small cespitose species,
with roots of numerous strong fibres; the culms 6-8 inches in
height; the leaves filiform, convolute ; the branches of the small
panicle scabrous.

Festuca MYURUS, Ehrh.=F. muralis, Kunth. At Chicla and
in the upper valley of the Rimae, between 9000 and 10,000 feet
above the sea! A doubtful native of South America, though now
widely spread.

FESTUCA SUBULIFORMIS, Benth. Pl. Hartw. 262. Above Casa-
palta !

FESTUCA CASAPALTENSIS, n. sp. Above Casapalta !

Humilis, cæspitosa, culmis adscendentibus spithamæis, angula-
ribus; foliorum vaginis brevibus, subinflatis, limbo anguste lineari,
flexuoso, canaliculato ; paniculæ laxiuscule ramis acute 3—4-angu-
laribus, sub lente scabriusculis, apice incrassatis; glumis inferio-
ribus longe acuminatis, extima uninervi, interiore trinervi, spiculis
3-4-floris, flosculo supremo sterili; gluma fertili tenuiter pube-
scente, quinquenervi, in aristam terminalem scabram ipsa gluma
duplo longiorem desinente; palea bicarinata; ovario glabro;
stylo subnullo; stigmatibus plumosis.

This is a very singular grass, agreeing in all essential points of
structure with Festuca, but not at all resembling any species of
that genus which I have seen. In appearance it comes nearer to
some of the Mediterranean species of Bromus, but the nervation
of the glumes makes it unfitting that genus. Among Festuce
its least distant alliance seems to be with the Mexican F. livida,
Willd. (Helleria of Fournier) and its very near neighbour F.
cacuminis, Hemsl. MSS.

Festuca ? F.Molli, Kunth, proxime affinis. Chicla! 1
drew from amidst dense bushes a siugle imperfect specimen of a
grass 4 or 5 feet high with a large lax few-flowered panicle,
which appears to be nearly allied to Kunth's Festuca mollis, and
may perhaps represent an extreme state of that species. It is
certainly a Festuca, with four or five florets in each spikelet, and
short oval awnless glumes, recalling the appearance of some
species of Bromus of the Serrafaleus group.

OF THE PERUVIAN ANDES. 63

Bromus H.gxKzaNvus, Presl. Chicla, and in the upper valley
of the Rimac! I feel uncertain whether this should be kept
distinct from B. unioloides, H. B. K., which has a very wide
range in South America.

Bromus tents, Presl? Found in the upper valley of the
Rimae and at Chicla, the specimens varying much in size and ap-
pearance! I have not seen authentic specimens of Presl’s plant,
but mine agree fairly well with his description. By a printer's
blunder the species appears as B. tenuis, Presl, in Steudel's Synop-
sis. No. 1370 of Mandon's ‘ Plants of Bolivia’ agrees closely with
my specimens from Peru. It has been placed in Kew Herbarium
in the same cover with B. auleticus, Trin., but cannot, I think, be
united to that species.

BROMUS FRIGIDUS, n. sp. Above Casapalta !

Culmis subsolitariis, tenuibus, 9-15-pollicaribus, erectis, levi-
bus; foliorum vaginis striatulis, dimidiam culmi altitudinem vix
attingentibus, limbo brevi, plano, presertim margine piloso;
ligula lacera; panicule depauperate secunde ramis pubescenti-
bus, spiculam nutantem 3-6-floram gerentibus; glumis vacuis
inzqualibus, superiori quam inferior sesquilongiori; gluma fertili
ovato-lanceolata, acuminata, tenuiter quiquenervia, puberula,
apice bidentata, sub apice aristata, arista dimidiam glume
longitudinem attingente ; palea bicarinata, carinis ciliatis.

This is allied to an undescribed Andean Bromus—No. 1368 of
Mandon’s * Plants of Bolivia’—but cannot, I think, be united with
it. In Mandon's specimens the flowering glume is elliptical,
almost truncate at the summit, the teeth scarcely perceptible, and
the awn decidedly shorter. The nerves of the flowering glume
are salient in his plant, while in ours they are very feeble.

FInicts.

Woopsta OBTUSA, Torr., var.=Cheilanthes crenata, Kunze.
Chicla!

CxsTOPTERIS FRAGILIS, Bernh. Chicla!
ADIANTUM JETHIOPICUM, L., var.— A. chilense, Kauly. Chicla!

CHEILANTHES MATHEWSII, Kunze. Chicla!

64 MR. J. BALL ON THE FLORA OF THE PERUVIAN ANDES.

PELLXA PULCHELLA, Fée. Upper valley of Rimac near Matu-
cana, R. Ward.

ASPLENIUM GiLLrESIANUM, Hook. Abundant about Chicla!
Most of the specimens of this fern have come from the Chilian
Andes, and except mine I have seen none from Peru. Some of
my specimens exactly agree with those first collected by Gillies,
but others are intermediate between Hooker’s species and A.
fragile of Presl. These represent the northern A. viride in South
America, and it is allowable to doubt whether either should be
regarded as a distinct species.

ASPLENIUM TRIPHYLLUM, Presl. About Chicla, very sparingly!
Upper valley of the Rimac, a single frond. R. Ward.

POLYPODIUM ATACAMENSE, Baker, MSS. Chicla! My speci-
mens closely agree with one from the higher part of the Atacama
desert sent to Kew Herbarium by Dr. Philippi, labelled by him
Polypodium squamatum, but certainly not the species so named by
Linneus. Mr. Baker has affixed the manuscript name P. ataca-
mense, and will doubtless describe the species in the next edition
of the ‘Species Filicum.'

NorocHLENA squaMmosa, Fée. Upper valley of the Rimac,
above Matucana. R. Ward.

I cannot close this enumeration without expressing my obliga-
tions to Prof. Oliver and to Mr. Baker for the friendly assistance
which they have repeatedly afforded to me in determining my
Andean specimens, as well as those collected in other parts of
South America.

CONTRIBUTIONS TO SOUTH-AFRICAN BOTANY. 65

Contributions to Soyth-Afrienn Botany.—OnmcmnipEx. Part II.
By Harry Bolus, F.L.S. With Additional Notes by N. E.
Brown, A.L.S.

[Read November 19, 1885.]
(Prats I.)

CALANTHE NATALENSIS, Reichb. f. (Bonplandia, 1856, p. 322).
Folia 5-6, oblongo-laneeolata, nervoso-plicata, basi angustata,
glabra vel minute pubescentia, 25-35 cm. longa, 6-8 cm. lata;
scapus strictus, teres, pubescens, vaginis 2-3 distantibus preditus,
35—45 em. longus; spica laxe 15-18-flora, braeteis ovario parum
brevioribus persistentibus ; flores expansi 3 em. lati; sepala lan-
ceolata acuta, l[:5 em. longa, extus pubescentia; petala lateralia
oblanceolata, equilonga, glabra; labellum trilobatum, porrectum,
subequilongum, lobo medio antice subtruneato, emarginato, sub
apice mucronato, lobis lateralibus euneatis multo brevioribus sub-
obtusis, disco seriebus tribus callorum instructo, basi in calear
filiforme areuatum plerumque ovario longius productum ; ovarium
gracile, arcuatum, pubescens, 2:5 em. longum.— Bot. Mag. t. 6844.
—C. sylvatica, var. natalensis, Reichb. f. in Linnea, xix. p. 374.
(Ex exempll. viv. duobus cultis sub num. 6101 exsiccatis.)

Hab. In sylvis primevis Perie Bush dictis, ditione King
William’s Town, Kaffrarie Britanniez, flor. Dec.-Jan., legit A.
Walsh; tum in hortis botanicis Port Elizabeth et Cape Town
culta.— Bolus 6101. “ Thickest and rockiest part of Perie Bush,"
D Urban No. 108. “Kirkman’s cutting, Maritzburg Road,
1500-2000 ft." Natal, Sanderson 1047; “ Bush Swamp, The
Bluff, head of Natal Bay,” Sanderson 1003; Natal, Buchanan,
In herb. Kewensi, &e.

The flowers are lilac, the labellum showing a tendency towards
a salmon-colour as they advance in age. The species is nearly
allied to C. Masuca, Lindl., Bot. Mag. t. 4541. The habitat is
interesting as being the first time that any species of this genus
has been recorded from the Cape Colony proper.

[The labellum of this species appears to vary considerably in
the size of the lateral lobes ; in the typical form as described by
Reichenbach, they are very small, and the middle part of the
labellum between them and the front lobe narrow ; in Mr. Bolus's
plant they are much larger and about equal the front lobe in

LINN. JOURN.—BOTANY, VOL. XXII. F

66 MR. H. BOLUS'S CONTRIBUTIONS

expanse, and the middle part is broader than in the type; other
speeimens in the Kew Herbarium are intermediate, and one
specimen (Natal, Buchanan) has both forms of lip on the same
spike.—N. E. B.]

SATYRIUM OCHROLEUCUM, Bolus, n. sp. Glabrum, erectum,
subdebile, spithameum vel pedale. Folia duo, humifusa, elliptica
subobtusa vel acuta, multinervia, basi vaginantia, 8-10 cm. longa,
5-9 em. lata, superiora 3 sensim in vaginas ampliatas foliaceas
apice patentes abeuntia; spiea laxe multiflora, 8-20 em. longa,
2:5 em. lata, bracteis lanceolatis reflexis ovarium plus minus exce-
dentibus; sepala lateralia oblonga, subfaleata, obtusa, 6 mm.
longa: sepalum alterum et petala lateralia linearia obtusa, 5 mm.
longa ; labellum galeatum, 7-8 mm. longum, ore anguste elliptico
margine angustissime recurvo, apice producto, erecto vel subre-
flexo, obtuso, minute erosulo, calcaribus filiformibus patento-diva-
rieatis 1-13 cm. longis, ovarium parum superantibus ; rostellum
oblongum antice angustatum ; columns lobus stigmatiferus semi-
orbiculatus ; ovarium costis acutis, cirea 1 cm. longum. (Ex ex-
empll. plur. viv. in Herb. Norm. Austr.-Afr. sub num. 318 distrib.)

Var. B. major. Scapus rectior robustior, flores majores ochro-
rubescentes ; labellum galeatum ore late elliptico vel hippocrepi-
forme.— Orchis bicornis, Jacg. Hort. Schenbr. ii. p. 26, t. 179
(non Linn.).

Hab. Yn clivis orientalibus montis Diaboli prope Cape Town,
alt. 1500 ped., flor. Oct., legit A. Bodkin (Bolus 4982!) ; Tulbagh
Kloof, Oct., Herb. Norm. Austr. Afr. 318! Hex River, Tyson
645 —Var. B cum forma typica Tulbagh Kloof, Bolus 5371! ;
Drége 1257a!

The colour of the flowers is a pale ochre-yellow, and the whole
plant has the scent of Anthoxanthum odoratum. The species is
intermediate between 5S. candidum, Lindl. (with which, as Mr. N.
E. Brown informs me, Lindley himself confused it) and S. folio-
sum, Sw. (not the plant Lindley described as S. foliosum). The
habit of this species is weaker, the spike laxer, the flowers more
spreading, and their size smaller than in S. candidum, besides
some differences in the shape of the perianth. The var. f is
more like S. candidum in the dried state, but is less robust. The
colour of the flowers in S. candidum is always white or a very
pale rose. From 8. foliosum, which our plant more resembles in

habit, it may be readily known by its much shorter spurs, and
the larger lower leaves.

TO SOUTH-AFRICAN BOTANY. 67

SATYRIUM EMARCIDUM, Bolus, n. sp. Glabrum, erectum, 10-
18 em. altum ; folium radicale ovatum vel subrotundum acutum,
basi caulem amplectens, 5-6 em. latum et longum, caulina 3—4
ovata sensim minora, superiora in vaginas reducta ; spica 6-12-
flora, 2-5 em. longa; sepala lateralia subeymbiformia, medio con-
stricta quasi incisa, acuta; sepalum alterum ligulatum aeutum;
petala lateralia ovata acuminata ; labellum galeatum, ovatum, apice
eristato, acuminatum, calearibus filiformibus ovario squilongis,
totis erectis alte connatisque, apicibus cito marcescentibus, sepalo
intermedio deflexo excepto; galea 8 mm. longa, calcaribus 6 mm.
longis, ceteris 6 mm. longis; columna deflexa; rostellum apice
umbonatum, stigmati emarginato squilongum ; antheræ connec-
tivum in acumen liberum subulatum productum ; ovarium ellip-
ticum costatum 6 mm. longum. (Ex exempll. plur. viv. sub num.
4847 distrib.)

Hab. In solo arenoso Fish Hoek juxta oras sinus False Bay
Peninsule Capensis, fl. Sept., Bolus 4847 ; Herb. Norm. Afr.-
Austr. No. 159.

The flowers are a yellowish or dirty white, the tips of the
perianth-segments very soon twisting and withering. In floral
structure it comes next to S. ligulatum, Lindl., and by those who
judge of it in a dried state only might be taken for a variety of
that species. But it seems to me to differ sufficiently by its
broader and shorter leaves, shorter spike, its differently shaped
side-sepals and labellum, shorter stigmatic lobe, &c. The label-
lum in S. ligulatum is lanceolate and has a very unusual sigmoid
flexure; here it is strictly ligulate and simply deflexed. It is to
be noted, too, that while the typical form of S. ligulatum grows
with us (on the Cape Peninsula) only on the mountains at about
2500 feet, and does not flower until December, the present
species is found on the sea-level, and near the coast, and flowers
in September.

SATYRIUM DEBILE, Bolus, n. sp. Glabrum, debile, decumbens,
15-20 em. altum ; caulis gracilis, laxe foliosus, basi vaginis 2-3
membranaceis vestitus; folia 4-5, elliptica vel lanceolata, ad-
scendentia, acuta, subundulata, basi vaginantia, infimum 5-6 em.
longum, 2 em. latum, superiora sensim minora; spica sublaxa
angusta, 8—5'5 em. longa, bractew lanceolate patento-erecte,
inferiores floribus longiores, superiores breviores; flores cum
ovario 5-7 mm. longi; sepala lateralia elliptica; sepalum alterum

F2

68 MR. H. BOLUS'8 CONTRIBUTIONS

oblongum ; petala lateralia subrhomboidea, omnia obtusa basi
connata, patentia, vix 2 mm. longa; labellum galeatum obtusum
(apice non depresso nec deflexo), ore latiore quam altiore, 2-25
mm. longum, basi saccis brevissimis obtusissimis auctum ; columna
brevis fere sessilis, lobus posticus brevis, apice emarginatus, bical-
losus, stigma parvum convexum semiorbieulatum ; rostellum tri-
angulare brevissimum ; anthere loculi dissiti; ovarium late ob-
longum margine utrinque bicostatum, costa antica et postica fere
obsoleta. (Ex exempll. plur. viv. sed floribus plerisque nimis
emarcidis, sub num. 5907 ! distrib.)

Hab. In udis, Kleinpoort montis Winterhoek Tulbaghensis,
alt. circa 3000 ped., Decembri, anno 1884, legit A. Bodkin.

Flowers greenish with dull red markings. Very distinct from
any other species known to me by its slender weak habit, thin
spike, very short sacs of the labellum, and nearly sessile column.
In floral structure it is very similar to S. Lindleyanum, mihi.

Disa TENUICORNIS, Bolus, n. sp. Erecta, glabra, 10-30 em.
alta; caulis foliosus vaginis imbricantibus obtectus ; folia laxa
linearia acuminata, basi laxe vaginantia, adscendentia, inferiora
6-9 cm. longa, interdum medio attenuata, superiora sensim minora,
basin spiez vix attingentia; spica subdensa longitudine e tertia
parte usque ad dimidium altitudinis plante ; bractes ovate, longe
cuspidate, membranaces, reticulato-venose, floribus subæqui-
longe; sepalum alterum posticum incumbens, galeatum, late
obovatum, obtusum, apiculatum, 1:5 em. longum, 1 em. latum, basi
bisacculatum, ealeare recto subfiliformi inter sacculos porrecto,
4-5 mm. longo; sepala lateralia falcato-ovata acuta, basi obliqua
deflexa, 1 cm. longa; petala lateralia minuta, oblonga, lobo faleato
apice aucta, horizontalia, columnz adnata ; labellum subulatum,
supra basin dilatatum, acutum, 1 cm. longum ; rostellum erectum,
breve, integrum, glandulis polliniorum approximatis, utrinque
tuberculatum ; anthera horizontalis ; stigma excavatum ; ovarium
cylindricum, 1:2 em. lougum. (Ex exempll. plur. viv. sub num.
4967 distrib.)

Hab. Yn rimis turfosis inter saxa in planitie inferiori montis
Tabularis juxta rivulum * Hout's Bay Stream ” dictum, alt. 2500
ped., flor. Oct., legit H. Bolus. No. 4967 ! in Herb. Kew ; Mac-
Owan 2590!

The colour of the flowers is mainly white, the galea spotted
with blood-red, and the sepals, labellum, and petals edged with

TO SOUTH-AFRICAN BOTANY. 69

the same colour, with which also the ovaries, bracts, and rhachis
of the spike are more or less tinged ; leaves dark green.— This
is à very distinct species. In general habit it resembles luxuriant
specimens of D. tabularis, Sond., but is readily distinguished
from that by its twice larger flowers and the very peculiar thin,
nearly straight, intrusely set spur, which is different from that of
any other species known to me. I suppose it to be a somewhat
rarely flowering species, for though I have collected on that part
of Table Mountain and at that season for several years past, I
never saw it until Oct. 1884; and it is not to be found amongst
Ecklon and Zeyher’s and Dr. Pappe's Collections in the Cape
Government Herbarium.

Disa XMvLA, Bolus, n. sp. Erecta, valida, glabra, 12—2 pedalis ;
scapus dense foliosus ; folia lanceolata acuminata, basi vaginantia,
margine ssepius undulata, erecto-patentia, inferiora 15-20 cm.
longa, superiora sensim minora in bracteas abeuntia; spica mul-
tiflora, 22-30 em. longa, 2:5-3'5 em. lata, bracteis lanceolatis acu-
minatissimis flores paulo superantibus ; sepala lateralia oblongo-
falcata, acuta, 1:4 cm. longa, 6 mm. lata; galea horizontalis, acuta,
apice emarginulata, 1:7 em. longa, calcare filiformi patente apice
deflexo, 1*6 em. longo; petala lateralia sub galea recondita, falcata,
obtusa vel retusa, apice incurva, basi rotundata, incurva ; labellum
oblongum vel oblongo-lanceolatum, integrum, obtusum, patenti-
deflexum, 1 em. longum, 3:0—3:5 mm. latum ; anthera resupinata ;
rostellum breve; ovarium obtuse triangulare. (Ex exempll.
duobus vivis sub num. 4330 exsiccatis.)

Hab. ln planitiebus arenosis prope Tygerberg, Nov., P. Mac
Owan (Bolus 4330! in herb. Kew); prope Groenekloof, Oct.,
Bolus 4330!; Salt River prope Cape Town, W. H. Harvey 244 !,
in Herb. Kew.

The sepals are usually a dull leaden blue, sometimes nearly
purple, the labellum dirty yellow variously marked with brown
patches or in the form of a large cross, the leaves glaucous green
with bands of red spots on the under surface near the base. The
species is closely allied to D. cornuta, Sw., with which it has been
confused in herbaria, resembling that species in colour, habit, and
general aspect. The differences may be summed up as follows:
the spike of our plant though dense is usually narrower than in
D. cornuta; the bracts are narrower and more acuminate ; the
galea is acute though minutely notched (not obtuse); the spur

70 MR. H. BOLUS's CONTRIBUTIONS

spreading for some distance (not immediately deflexed) ; the side
sepals are narrower and more acute; the labellum exaetly or
nearly oblong (not broadly obovate); colour as above, while in
D. cornuta it is invariably a deep velvety purple.

It is an interesting question whether this is the D. macrantha
of Thunberg (Flor. Cap. ed. 1823, p. 8). It agrees with the
deseription in the acute galea and the more spreading spur; but
Thunberg says the spike of his plant is denser than in D. cornuta,
the flowers slightly larger, and the labellum acute and carinate,
which is not the case with our plant. Then there is the difficulty
that there is no type of D. macrantha existing in Thunberg's
herbarium, and no other plant so far as known which agrees with
his description; while it is certainly desirable that our plant
should be distinguished and published, a course I have thought
best to adopt in spite of the suspicion I cannot avoid entertain-
ing that this may be a form of Thunberg's species.

Lindley (Gen. & Sp. Orch. p. 349) said that D. macrantha was
unknown to him; but the name was subsequently applied by
gardeners to a very different plant, viz. D. crassicornis, Lindl.
(D. megaceras, Hook. f. Bot. Mag. t. 6529), an Eastern Cape
species. Under the latter number Hooker remarks :—“ As to
D. macrantha of Thunberg, it is clearly a species very near to D.
cornuta, if not a variety of that plant; there are numerous speci-
mens thus named in the Kew Herbaria, amongst them one from
the Sands about Salt River, near Cape Town, collected by Dr.
Harvey, who has appended to it a ticket with * D. macrantha of
Thunberg. Differs from D. cornuta merely in its labellum. 1
cosnider it only a variety, yet its habit is different.’ Whatever
the difference of habit is, it is lost iu the drying ; for the speci-
men is in this state undistinguishable from D. cornuta, and, like
it, has flowers not half the size of those of D. megaceras, with
minute included petals.”

I will only add that I have drawn both species from life and
carefully compared them; and though I have only seen two living
examples of the present plant at an interval of five years, I had
no difficulty or hesitation whatever in distinguishing them from
D. cornuta, the latter being extremely common.

Disa Scvrry: Bolus, n.sp. Glabra erecta sequipedalis. Folia
inferiora ligulata acuminata, basi arcte vaginantia, 10-12 cm.
longa, 1:0-1:5 em. lata, superiora sensim minora, ad vaginas ad-

TO SOUTH-AFRICAN BOTANY. Ti

pressas apicibus liberis reducta ; spica cylindracea, sublaxiflora,
8-17 cm. longa, 2:5-3'0 cm. lata, bracteis lanceolatis, acuminatis,
erectis, apicibus reflexis; galea infundibularis erecta, 1:1 em. longa,
calcare filiformi adscendente, apice interdum subincrassato, 3 em.
longo, aucta; sepala lateralia elliptiea subobtusa, infra apicem
valde apiculata, cum apiculo 1:4 em. longa; petala lateralia ob-
longa, erecta, margine postico aurieulato, antice ampliato, emargi-
nato, vix 1 em. longa; labellum anguste rhomboideo-lanceolatum
vel lineari-lanceolatum, basi angustatum, obtusissimum, minutis-
sime crenulatum, circa 1 em. longum, 2-3 mm. latum. (Ex ex-
empll. duobus exsiccatis ut infra.)

Hab. In uliginosis, Menziesberg, prope Stockenstrom, Regione
Orientali Colonize Capensis, flor. Jan., legit W. C. Scully. (No.
5915 Bolus in herb. Kewensi, &e.) Caffraria, Mrs. Barber No.
23! in Herb. Kew.

The habit is straight and lanky with two or three leaves at
base, and half a dozen close or little-spreading sheaths on the
scape. The species comes next to D. Cooperi, Reichb. f., to
which it is, perhaps, too nearly allied. Nevertheless, though it
has sepals of a similar shape and nearly as large, it differs in the
petals and labellum. The former in D. Cooperi are very fleshy
and have no ear-like lobe on the back margin ; the labellum is
still more different in shape, and three times as large superficially
as in this plant. Judging from my specimens the habit is also
different ; tlie present plant being less robust, the leaves fewer
and narrower, and the spikes rather shorter and much less dense.
The sepals appear to be rose-coloured, with darker veins.

As Prof. Reichenbach has not given the colours of D. Cooperi,
I may quote them from a note by Mr. Tyson on his No. 1606 :—
“Galea and spur creamy, with pinkish sides ; petals greenish
yellow; labellum lemon-yellow. All the spurs lie uniformly
pointing upwards along the rhachis. The flowers are strongly
clove-scented.”’

' Disa AFFINIS, N. E. Br. (in Gardeners’ Chron. 1885, vol. xxiv.
p.402). Erecta vel subdecumbens, 1-13 pedalis; folium inferum
ellipticum, subobtusum, multinervium, basi vaginans patens, 10-
19 em. longum, 5-6 cm. latum, superiora (1-2 vel rarius 3) sensim
minora, acuta, mucronata,in bracteas abeuntia; scapus vaginis 4-5
submembranaceis arcte adpressis vestitus; bractex late ovate acu-
minatie, membranacee, retieulato-venoss, ovaria calcariaque laxe

72 MR. H. BOLUS’S CONTRIBUTIONS

amplectentes et iis parum breviores; spica sublaxa 8-12 cm.
longa; sepalum dorsale limbo fornicato OU lem. longo, calcare
filiformi dependente apice emarginato, 2:5 cm. longo, ovario parum
longiore ; sepala lateralia oblonga, subacuta, trinervia, patentia,
5-6 mm. longa; petala lateralia oblique ovata, subfalcata, pleraque
emarginata, trinervia, galea parum breviora; labellum oblongum,
nune obtusissimum, nunc apice angustatum, 8-9 mm. longum;
rostellum brachiis retroversis, stigma magnum prominens, antice
rotundatum, supra convexum, vix superans, et ab eo lacuna quasi
excavata sejunetum. (Ex exempll plur. viv. sub num. 4555
distrib.) —Monadenia rufescens, Lindl.! Gen. & Sp. Orch. p. 356,
excluding synonyms.

Hab. Yn rimis saxorum locisque umbrosis in montibus Penin-
sule Capensis, 1400-2500 ped., fl. Sept.-Oct. Bolus 4555, Herb.
Norm. Austr.-Afr. 170 ; Clanwilliam, Drakensteenbergen et Ge-
nadendal, Drége 1252 a !, 1252 d!; Zeyher 39251; Craggy Peak,
Zwellendam, Burchell 7321, 7357 ; Pappe 52 in hort. Lindl. §e.

This is one of the most distinct species of this group, being
readily known by its large broad lower leaves, its lax almost
weak habit, and its pale colour both when living and in the dried
state. The leaves are a sap-green, the sepals sulphur-yellow, the
petals bright golden. The column is also well marked, the stigma
being large and bossy with a deep depression between it and
therostellum. The sepals are thin, the petals thicker but scarcely
fleshy as in most of this section. It loves the steep sides of
rocky hills or crags which are turned away from the sun.

Disa PYGM xA, Bolus, n. sp. Erecta, glabra, pusilla, 4°5—-9 cm.
alta; caulis e basi gracili foliosus ; folia infima 1-2 in vaginas
reducta, superiora 3-5 ovata acuta, vel rarius lanceolata acuminata,
basi vaginantia, erecto-patentia, obscure 3-5-nervata, 1:5-2:5 em.
longa; spica densa, 8-24-flora, longitudine fere totius plante;
bractez nunc ovate obtuse floribus equilonge, nunc lanceo-
latee acuminate flores superantes; flores cum ovario 7-9 mm.
longe ; galea oblonga, obtusa, apiculata, calcare tereti, incurvo,
obtuso, sspe subinflato, 2-3 mm. longo, ovario breviore, galea
cum calcare 7 mm. longa; sepala lateralia ovata, obtusissima,
patento-recurva, 4-5 mm. longa; petala lateralia subtriangularia
acuta, angulo postico obtuso longitudine fere galew; labellum
oblongum obtusum, basi angustatum, deflexum, 2:5 mm. longum ;
rostellum stigma breve foboridsbum superans, processubus latera-

"P

TO SOUTH-AFRICAN BOTANY. 73

libus antheram versus omnino reflexis ; caudicule elongate polli-
uiis æquilongæ ; ovarium breve obovatum, pustulis hyalinis obtec-
tum, 3 mm. longum. (Ex exempll. viv. 11 sub num. 4970 distrib.)

Hab. In arenosis montis Muizenberg [Montis Steenberg, in
Herb. Kew.—N. E. B.] Peninsul: Capensis septentrionem versus,
alt. 1500 ped., fl. Nov., legit A. Bodkin. Bolus 4970! in Herb.
Kewensi &c.

The sepals are a dull red, the petals and labellum pale yellow.
Allied to Monadenia micrantha, Lindl., from which it differs by its
smaller size, greater relative length of spike, broader back sepal,
rounded side sepals, wider labellum which is not thickened at the
apex, differently shaped and not fleshy side petals, higher rosteilum,
longer caudicles of the pollinia (which are almost absent in M.
micrantha), rounded stigma, aud, lastly, its shorter ovary covered
with pustules. The spur is variable, but usually shorter, thicker,
and more curved than in M. micrantha. I rely chiefly upon the
differences in the column, for M. micrantha is itself a variable
species in respect of size, bracts, and spur; yet I have never seen
any small specimens of that species which resemble this.

Disa nETICULATA, Bolus, n. sp. Erecta vel subdecumbens,
glabra, 10-30 em. alta; caulis foliosus basi sepius gracilis ; folia
inferiora lineari-lanceolata acuminata, basi vaginantia, erecta, su-
periora sensim minora ovato-lanceolata, basi spice parum longiora;
spica sublaxa, longitudine circiter dimidium plantæ ; bractez late
ovate, acumine longo torto przditze, membranaces, reticulato-
venose, floribus parum longiores, ovarium eum calcare involventes ;
flores eum ovario 1°8 em. longi; galea obovata obtusa 8 mm.
longa, calcare filiformi aeuto, 1 cm. longo, ovario parum breviore ;
sepala lateralia oblonga obtusa, deflexa, obscure trinervia, 6 mm.
longa; petala falcato-ovata, subacuta, carnosa, 5 mm. longa;
labellum linguzforme, obtusum, carnosum, deflexum, 5-6 mm.
longum; rostellum erectum, latius quam longius, utrinque tuber-
eulatum, fissura inter rostellum et stigma tenui; stigma breve
obtusum ; ovarium 1-1:5 cm. longum. (Ex exempll. viv. 8, sub
num. 4988 distrib.)

Hab. In humidis montis Constantisz, in Peninsula Capensi, alt.
2000 ped., flor. Dee. A. Bodkin, Bolus 4988!, in clivis montis
Tabularis, alt. 2800 ped., flor. Dec., Bolus 4897!

The flowers have a yellowish ground colour with dull red on
the edges, back of galea, &e., the labellum and the whole flower

74 MR. H. BOLUS's CONTRIBUTIONS

sometimes darker; leaves a dark green. The species is distin-
guished by its usually slender stem (though stouter specimens
oceur), and especially by its very broad membranous reticulately-
veined braets, low rostellum, and shallow cleft between that and
the stigma. On the Cape Peninsula at least, it is exclusively a
mountain-plant.

Disa LINEATA, Bolus, n. sp. Erecta, glabra, laxe foliosa, 3-9
pollicaris ; folia 5—6, linearia acuta, adscendentia, basi vaginantia,
3-6 cm. longa, superiora sensim in bracteas abeuntia ; spica
subdensiflora (6-20-flora), cylindrica, bracteis ovato-lanceolatis
acuminatis, erecto-patentibus, flore plerisque brevioribus ; flos cum
ovario circa 1:3 em. longus ; sepalum alterum posticum galeatum,
erectum, acutum, margine subpatenti undulatoque, basi obtusissi-
mum (non calearatum nee saccatum), 9 mm. longum; sepala late-
ralia oblonga acuta, patentia, 7 mm. longa ; petala lateralia oblique
oblongo-faleata, reclinata, apice incurva, dentata, basi column: ad-
nata, 5 mm. longa ; labellum e basi lata lanceolatum acuminatum,
integrum vel obsolete crenulatum, apice recurvum, 5 mm. longum ;
anthera reflexa ; rostelli brachia erecta, parallelia, eum glandulis
arcte approximata, basi utrinque tuberculata ; stigma amplum;
ovarium cylindraceum 6-8 mm. longum. (Ex exempll. plur. viv.
sub num. 7001 distrib.)

Hab. In clivis humidis montis Constanti: in Peninsula Capensi,
alt. eire. 2700 ped., fl. Sept. (-Oct.), legit 4. A. Bodkin. Bolus
4966 !, 7001! in Herb. Kew. &e.

The sepals are a tawny yellow, the galea marked by definite
purple lines much in the manner of Satyrium lineatum, Lindl., and
somewhat like the galea of that plant in general appearance. The
petals have dark purple tips, the labellum a large deep red-purple
centre with a yellow tip. The species comes near to D. neglecta,
Sond. ; but Mr. N. E. Brown informs me that the flowers are two
diameters larger than in that species, of which the lip is described
as narrow linear and as long as the sepals, while in our plant it is
shorter and nearly ovate.

Disa BopkiNr, Bolus, n. sp. Ereeta, robusta, glabra, 6-18 em.
alta; caulis foliosus sæpe flexuosus ; folia 6-8, e basi lata linearia
acuminata, basi vaginantia, erecta, gracilia, flexuosa, 6-10 em. longa,
suprema inflorescentiam superantia; bractez late ovate acuminate,
floribus equilonge ; flores 1-6, corymboso-capitate, capitulis majo-
ribus 5 cm. latis, post anthesin sæpe in spicam brevem productis

TO SOUTH-AFRICAN BOTANY. 75

sepalum alterum anticum cucullatum quasi medio deflexum, vix
galeatum, apice angustatum, leviter emarginatum, 1:5 cm. longum ;
sepala lateralia ovato-elliptica, subacuta, concava, erecto-patentia,
1-6 em. longa, 1 em. lata; petala lateralia oblonga subfalcata,
obtusa, carnosa, recurvato-adscendentia, 7 mm. longa; labellum
oblongum, obtusissimum, carnosum, adscendens, 1 em. longum, 5
mm. latum ; rostellum erectum, brachiis parallelis approximatis ;
anthera leviter recurvata; ovarium rectum, 1 em. longum. (Ex
exempll. plur. viv. sub num. 4968 distrib.)

Hab. In humidis montis Tabularis prope Cape Town locis
pluribus, alt. 2200-2800 ped., mense Nov. (anno 1884) legit 4.
Bodkin. H. Bolus No. 4968! in Herb. Kew ; Herb. Norm. Austr.-
Afr. No. 333!

The colour of the sepals is a dull red, the petals and labellum
deep purple in the middle with orange tips. Its nearest congener
is D. Melaleuca, Thunb.; but it greatly differs from that by its
relatively much larger odd sepal, its wide obtuse labellum, and
otherwise, so that there is little probability of confusion even in
the dried state; in a living state their appearance is widely dif-
ferent. The base of the stem is sometimes surrounded with the
fibres of old leaves.

PrERYGODIUM MAGNUM, Reichb. f. in Flora, anno 1867, p. 117.
—This species was very briefly described ex sicco and without
any published locality, or other indication of its origin. The
author had under his hands a single specimen sent to him by
Prof. MacOwan, whieh was returned with this name attached in
the author's hand-writing. It is probably the type specimen,
but certainly an authenticated one. As I have seen this, and
also one other, and have had the advantage of examining it from
flowers freshly preserved in spirit, I venture to append a fuller
description :—

Erectus, validus, glaber, 60—70 em. altus; scapus dense foliosus;
folia lanceolata, obtuse acuta, basi vaginantia, superiora angus-
tiora, acuminata, 15-20 cm. longa, ad basin spice attingentia ;
spica dense multiflora, 20 em. longa, 4 cm. lata; bracteæ subu-
late vel anguste lanceolate, infime flores multo superantes (in
exempl. MacOw. 5-6 em. longs), superiores ovario æquilongæ ;
flores expansi ambitu fere orbiculares 1-2 em. diametro ; sepalum
dorsale oblongo-lanceolatum acutum, 9 mm. longum; lateralia
equilonga, subconfurmia, magis concava, parum obliqua; petala

76 MR. H. BOLUS'S CONTRIBUTIONS

lateralia cuneato-flabellata, marginibus superiori et anteriori fim-
brillatis, multinervia, cirea 9 mm. longa; labellum exacte flabella-
tum, petalis margine anteriori multo minus fimbrillatis, appendice
magno, cucullato, suberecto, media antice fisso, galea Disperidis
cucullate fere instar; columna hippocrepica, glandulis polli-
niorum antice subapproximatis; ovarium gracile, 3:5 cm. longum.
(Ex exempll. duobus exsice. ut supra.)

Hab. In summo monte Kagaberg, 5000 ped., fl. Febr. Mac-
Owan, in herb. Gub. Capensi ; ad ripas rivuli monte Currie, Gri-
qualand East, 5100 ped., Febr., legit Domina P. Brownlee, Tyson
No. 1604 in herb. Bolus.

The habit is entirely that of Pt. inversum, Sw., and very dif-
ferent from that of Pt. venosum, Lindl., to which Prof. Reichen-
bach compares it, probably by a slip of the pen, or typographical
error. The curious appendage of the labellum is different from
anything else in the genus ; the column most resembles that of
Pt. carnosum, Lindl.; but the glands of the pollinia are nearer
together. Tyson's No. 1604 is less robust and shorter than
MacO wan's, but in all essential characters it agrees exactly. To
the last-named gentleman I am indebted for flowers preserved
in spirit.

[The Kew Herbarium contains the following specimens of
this fine species, which, according to Mr. Hutton, “ grows to 5
feet in height, flowers green, striped with light red; January;
soft tubers like Disa, leaves thick and suecculent."— Kagaberg,
3000 ft., Hutton!, Natal, Wood 909!, Cooper 3621!, Mrs. Fan-
nin!; Nottingham, Natal, Buchanan! Some of these specimens

are between 3 and 4 feet high.— N. E. B.]

Disperis oxvGLOossa, Bolus, n. sp. (Pl.I.figs.6-12.) Erecta.
gracilis, glabra, 20-25 cm. alta ; folia 2-3, oblonga acuta, basi vagi-
nantia, 2—3 em. longa, 6 mm. lata, internodiis 4-5 em. longis ; race-
mus laxe 2-3-florus ; bractex ovate acuminate, complicate, ovario
subequilonge ; sepalum dorsale fornicatum, apice longe acumi-
nato, 1°6 em. longum ; sepala lateralia lanceolata acuminata, ar-
cuato-patentia, 1°6 em. longa, infra medium ealeari obtuso 5 mm.
longo aucta ; petala lateralia sub galea agglutinata, eaque subzequi-
longa unguiculata, oblique lanceolata acuminata, margine anteriori
lobo auriculeformi praedita ; labellum basin versus lineari-unguicu-
latum, ima basi column: adnatum, deinde bipartitum, labiis valde
divergentibus, labio antico lanceolato acuminato, semi-complicato,

TO SOUTH-AFRICAN BOTANY. 77

arcuato-porrecto, basi intus minute tuberculato, labio postico bre-
viori, lineari, angustissime cymbiformi, obtuso, utrinque ala late
triangulari hyalina decurrente instructo. Ex exempll. 2 exsicc.
ut infra.

Hab. In lat. australi montis Currie supra cataractam, ditione
Griqualand East, C. B.S., alt. cirea 5500 ped., flor. Febr., legit
W. Tyson No. 1603! in herb. meo. Natal; heights above Kar-
kloof, J. Sanderson No. 1071. Kaffraria; mountains, in rather
damp localities, Mrs. Barber No. 28; Baziya mountain top,
4000 ft. alt., fl. Feb., Baur No. 813.

The species is characterized by its long, narrow, shallow, and
taper-pointed back sepal, and by the shape of its labellum. It
comes near to D. Wealii, Reichb. f.*, and the habit is similar.
It differs from that by its more robust habit, larger and more
taper-pointed flowers, by the anterior lobe of its labelluin being
longer (not shorter) than the posterior lobe, and by the posterior
lobe being winged and differently shaped. In the dried state it
might easily be confused with it.

Disprknis MacowaNr, Bolus,n.sp. (Pl.I.figs. 13-17.) Erecta,
pusilla, 9-15 em. alta; folia duo, patula, ovata acuta, cordato-
amplexicaulia, ciliolata, obscure reticulato-venosa, ex sicco sub-
membranacea, 1:2-1:5 em. longa, 1:0-1:4 em. lata, infimum nune
humistratum, nune caulinum, internodio 2-3 em. longo ; scapus
pergracilis, 1-florus, hispidulus, bractea foliacea, foliis conformi,
6 mm. longa, ovario breviori; sepalum dorsale galeatum, 6 mm.
altum, apice acuto porrecto ; sepala lateralia ambitu lanceolata,

* [In habit and general appearance this species appears to me far more
nearly to resemble D. paludosa, under which name it has been distributed by
Mr. Baur, but is readily distinguishable from that species by its larger flowers,
more acuminate galea, and a very different lip. The flowers are purple with
some green warted spots on the petals. Of this species I have made fresh
drawings of analyses, and substituted for those originally sent by Mr. Bolus.
Figs. 6, 8, 9, 11, & 12 were made from Mr. Bolus'stype specimen (Tyson 1603),
which he kindly sent to Kew for my inspection, and figs. 7 and 10 were drawn
from Baur's specimen No. 813, which exhibits a decided difference in the shape
of the dorsal sepal, although the lateral sepals and lip are the same as in Tyson's
plant. Sanderson's and Mrs, Barber's specimens are also like that of Baur, but
as I do not notice any other difference, I think they can scarcely be regarded in
any other light than as conspecific with Tyson's plant. The dorsal sepals are
shown sideways and pressed flat. The petal, fig. 10, is shown with its claw in its
natural position, whilst in 9 it is pressed flat in a line with the limb, which is
shown pressed flat in both 9 and 10,—N. E. Brows, Kew.]

78 MB. H. BOLUS'S CONTRIBUTIONS

patula, apice deflexa acuta, 4:5 mm. longa, medio calcare obtuso
sub apice extenso nee dependente, 2 mm. longo, aucta; petala
lateralia sub galea agglutiuata, lanceolato-falcata, margine ante-
riori basin versus lobo rotundato predita, 5-6 mm. longa ; labellum
7:5 mm. longum, antheram multo superans, e basi lineare, in lami-
nam rhomboideam expansum, deinde dente lacinula retrorsa apice
acuto recurvo in pagina anteriori, tum denique in processum fili-
formem subearnosum papilloso-hispidulum productum apice tri-
lobum, lobis obtusis ineurvis; rostelli brachea acuta; ovarium
circa 1 em. logum. (Exexempll. 2 exsicc. ut infra.)

Hab. Ad ripas rivuli ad lat. montis Boschberg pone Bester's
Hoek, Somerset East, Colonia Capensi, alt. circa 4500 ped., estate
anni 1880 legit P. MacOwan No. 2626! (in herbb. Kew et Bolus).
[1n rupibus prope Grahamstown et in monte Boschberg, 2000-
4000 ped., flor. Decemb., MacOwan No. 807! in Herb. Kew;
two small specimens 23 inches high.—N. E. Bnows.]

Colour of the flowers * pale bluish " (this being very unusual
in the genus as far as my experience goes), the side petals with a
few greenish spots. The affinities of the species are with D.
Woodii, mihi. The labellum, which is the great mark of distinc-
tion in all the Disperids, is of a similar type, but quite dif-
ferent in detail, besides which all the other parts of the periauth
differ. We have but two plants, all that Mr. MacO wan, who so
diligently explored that fine region, ever found. Nevertheless,
we hope to get it again from some of our younger friends who
are collecting in the mountains of the Eastern districts.

Disveris Woopir, Bolus, n. sp. (Pl. Ï. figs. 18-22.) — Erecta,
pusilla, glabra, 5:0—7:5 cm. alta; folia duo, radicalia, ovata acuta,
cordato-amplexicaulia, undulata, retieulate-venosa, 1*5 cm. longa,
1 em. lata; scapus gracilis, aphyllus, uniflorus ; bractea ovata sub-
obtusa, ovario brevior; sepalum dorsale altigaleatum vel in calcar
extinctoriiforme adscendens productum, basin versus inflatum,
antice acutum, porrectum, 1:0-1:5 em. longum; sepala laterali
deflexa, lanceolata acuminata, 6 mm. longa, saccis brevibus obtusis
aucta; petala lateralia subrhomboidea acuta, sub galea aggluti-
nata; labellum in fundum galeæ extensum, 8-9 mm. longum, €
basi lineare, lacinula cordata, retrorsa, foliacea, acuta, decurva in
pagina anteriori, postice caruncula peruleformi (nectariferi ?)
preeditum, deinde sursum filiforme, apice bifidum, lobis approxi-
matis obtusis. (Ex exempll. 4 exsice. ut infra.)

TO SOUTH-AFRICAN BOTANY. 79

Hab. In colle graminoso, Inanda, Natal, alt. 2000 ped., flor.
Aprili, J. M. Wood No. 127 !* ; Natal, J. Sanderson No. 1010!

The sepals are a pretty rose-pink with darker markings, the
petals with pale green dots in front; the scape is reddish, and
the leaves dark green with pale reticulated veins. The galea is
somewhat variable in length. From Mr. Wood's excellent spe-
cimens I have been enabled to ascertain the form of the labellum
with a fair approach to correctness; and I have further had the
advantage of seeing a finely executed drawiug by Miss A. Clarke
from the liviug plant, which that lady very kindly sent to me.
Altogether this species is a little floral gem.

DisPERIS PURPURATA, Rchb. f., var. PARVIFLORA, Bolus. Gra-
cilior, hirsutior, 8-20 em. alta ; folia laxiora ; flores dimidio fere
minores; galea compressa ore angustiora (anguste elliptica) ;
sepala lateralia saccis pro magnitudinem longioribus; petala
lateralia latiora ; labellum angustius.

Hab. In arenosis, prope Groene Kloof, alt. circa 300 ped., flor.
Oct., Bolus No. 4337 !; prope Stellenbosch, Sanderson No. 937!
(in Herb. Kewensi).

[The flowers are white according to Mr. Sanderson.—N. E.
Bnowsx.]

DiseEmis Tyson, Bolus, n. sp. (Pl. I. figs. 28-33.) Glabra,
erecta, pedalis vel ultra ; scapus rectus, rigidiusculus, laxe folio-
sus; folia 4—5, ovato-lanceolata acuta, basi amplexicaulia, 2:0—
2:5 em. longa, 1 em. lata, internodiis 4-7 cm. longis; spica laxe
12-20-flora, 10-12 cm. longa; bractee ovate acute, complicate,
fere erectz, ovariis subequilonge ; galea cucullata, subconica, ob-
tuse acuta, fundo obtusissimo, 9 mm. longa; sepala lateralia cir-
cumscriptione anguste lanceolata acuminata, concava, medio sacco
brevissimo aucto, 7 mm. longa; petala lateralia falcato-spathulata
acuta, margine anteriori tuberculato, circa 4 mm. longa; labellum
basin versus lineari-unguiculatum, ima basi columns; adnatum,
deinde bipartitum, lobo antico triangulari acuto, margine reflexo,
intus tuberculato, lobo postico longiori, subulato, obtuso, resupi-
uato ; rostelli brachia glandulifera brevia. (Ex exempll. 2 exsice.
sub num. 1079 Tyson ; floris partes ex spec. in spir. vini asservatis.)

Hab. In uliginosis prope Kokstad, Griqualand Orientalis, alt.
5000 ped., Jan., legit W. Tyson No. 1079! in Herb. Kew, &e. ;

* [The specimens of Wood, No. 127 at Kew, are from “ Noodsberg, Natal,
May."—N. E. Bauows.]

80 CONTRIBUTIONS TO SOUTH-AFRICAN BOTANY.

Somerset East, MacOwan 521; Caffraria, in humidis clivis montis
Baziya, alt. 2000 ped., flor. Mart. Apr., Baur 151!

The species is allied to D. Cooperi, Harv. Thes. Cap. ii. t. 172 *;
and must also be allied to D. stenoplectron, Reichb. f. (which the
author says is very near to D. Cooperi, Harv.), but differs from
that by its galea being cucullate, not fornicate, and obtusely acute,
not acuminate; also by the posticous lobe of its labellum being
narrower, not wider, than the anticous lobe. In this species, as
in D. secunda, D. cucullata, the apex of the anterior lobe of the
labellum is closely fitted to and under the apical point of the
galea.

DESCRIPTION OF PLATE I.
South-African Orchids. Floral parts of species of Disperis.
A. Figs. 1-5. Disperis Wealii, Reich. f.

B 6-12. oxyglossa, Bolus, n. sp.
€. , 13-17. —— Macowani, Bolus, n. sp.
D. ,, 18-22. —— Woodii, Bolus, n. sp.

EK „ 2. namaquensis, Bolus.

E: 28-33. Tysoni, Bolus, n. sp.

The figures are numbered consecutively in the Plate, but the following small
letters apply similarly to each species :—

a. Galea, or back sepal, side view.

b. Side sepal, front view.

c. Side petal, side view.

d. Labellum, side view.

e. Ditto, front view. |

Ff. Ditto, front oblique view.

g. Ditto, in front and from above.

| Enlarged 3 diameters.

Enlarged 6 diameters, except in D. ory-
glossa, which are x3 diam., and
D. Tysoni, which are x 4 diam.

In the side petals the median nerve shows the line of attachment to the galea ;
the anterior or free portion is generally marked by coloured spots; in D. ory-
glossa it is to the right ; in D. Wealii to the left.

Nearly all the figures are drawn from dried specimens of actual types. In
the case of D. oxyglossa, figs. 6, 8, 9, 11, and 12 are drawn from Tyson's plant
No. 1603, and figs. 7 and 10 from Baur’s specimen No. 813. In fig. 9 the
claw is straightened out in a line with the limb; and in fig. 10 it is shown more
in its natural position, although the limb is flattened out, not arched, as it
would be when in its natural position.

Figs. 23-27 (D. namaquensis) are drawn from living specimens. Figs. 28-
33 (D. Tysoni) have been drawn from flowers preserved in spirit, viz. specimens
of Tyson's No. 1079, whieh were compared by me, and found to agree with
specimens of MacO wan's No. 521, which had been marked ** D. Cooperi, Harv.,"
by Prof. H. G. Reiebenbach. Fig. 33 is viewed from above, the anterior lobe
being pulled down to show the upper surface.

[* The flowers of D. Cooperi, Harv., are twice as large, and not nearly 5°
numerous, as those of Mr. Bolus's new species D. Tysoni.—N. E. Bnows.]

ON THE TRANSPIRATION OF PLANTS. 8l

Parts of the Solar Spectrun/on the Transpiration of Plants.
By the Rev. George HreNKrow, F.L.S.
[ Read 3rd December, 1885.]

Inrropuction.—The conviction that light is an important factor
among the causes of the phenomena of plant-growth has long
been held; but it is only within the last twenty years that
satisfaetory experiments have been made to ascertain the relative
effects of different rays of light upon transpiration. Sachs, in
his * Physiologie Végétale'*, says :—“ La lumière est un des agents
qui agit le plus efficacement sur la transpiration. Mais on ne
peut pas dire positivement si elle agit par elle-méme, ou par son
union intime avec une élévation de température."

That author repeats this opinion in his * Text-book ' t : —* It is
still doubtful whether light, 2. e. radiation as such, independently
of the elevation of temperature caused by it, influences transpi-
He subjoins the footnote, ** Dehérain's researches f do

A Contribution to the Study wf Relative Effects of Different

ration.”

not decide the question.”
Prof. Daubeny in 1836 carried out some experiments with

coloured lights, but forbore to give any numerical results, as he
met with some apparent anomalies §. He came, however, to the
conclusion that “the processes [the exhalation of moisture from
the leaves, and the absorption of it by the roots] are probably
dependent on the combined action of heat and light, coupled with
those mechanical influences which operate upon dead as well as
upon living organic matter " ||.

Although the glasses used in his experiments were not tested
by the spectroscope, he appears to have come to some conclusions
very nearly the same as those arrived at by the latest observers,
e.g. Wiesner ; yet he thinks that they are exceptional instead of
being the rule, as the latter believes them to be. “ Now, although
the experiment," he adds, * tended to show that the extrication

* P. 250 (1868). t Second English ed. (1882) p. 678.

1 Ann. des Sci. Nat. sér. 5, xii. (1869), p. 1.

$ Sachs remarks as follows on Daubeny's experiments :-—“ Ch. Daubeny, qui
s'est occupé de cette question, ne s'exprime qu'avec une extrême prudence, et
ses observations ne paraissent pas l'avoir conduit à des résultats positifs.”—
Phys. Vég. p. 251.

| Phil. Trans. 1836, i. p. 159.

LINN, JOURN.— BOTANY, VOL. XXII.

82 REV. G. HENSLOW ON THE

of moisture, ceteris paribus, was most abundant in proportion to
the intensity of the light admitted (orange glass in general
causing more moisture to be exhaled than red or green), yet in
some instances blue and purple glasses, and, still more remarkably,
bottles filled with the cupreous solution [ammonio-sulphate of
copper] would cause a more abundant exhalation than orange or
even transparent glass. Here, however, another principle seems
to come into play, namely, the influence of heat radiated from the
surface of the screen ” *.

Dehérain's researches alluded to above were made upon leaves
in a saturated atmosphere, and he came to the following eon-
clusions, amongst others t :—

“1°. L'évaporation de l'eau par les feuilles est déterminée par
la lumière et non par la chaleur.

“9°. Cette évaporation se continue dans une atmosphère
saturée.

*3*, Les jeunes feuilles évaporent plus d'eau que les an-
ciennes.

* 4^, Les rayons lumineux (jaune et rouge) efficaces pour déter-
miner la décomposition de l'acide carbonique, sont aussi ceux qui
provoquent l'évaporation la plus abondante.

“5°. La différence d'action des divers rayons lumineux est
encore sensible quand on s'efforce deles amener à une intensité
lumineuse égale."

The coloured fluids used by M. Dehérain were as follows :—

“1. Dissolution rouge de carmin dans l'ammoniaque.

“2. Dissolution jaune de chromate neutre de potasse.

“3. Dissolution verte de chlorure de cuivre.

* 4, Dissolution bleue de sulphate de cuivre ammoniacal.

“5. Dissolution violette d'iode dans le sulfure de carbone."

The quantity of water * evaporated" in one hour by a leaf of
Barley, reduced to percentages of the whole weight of the leaf,
was as follows :—Red 93:6 p. c., yellow 63:4, green 5:8, blue 633,
violet *05.

The relative intensities of the rays, however, were not estimated
in any way. Nor were the lights tested by the spectroscope—?
most important and, in fact, necessary procedure; as it is pretty
certain that they were not monochromatic f.

* Phil. Trans. 1836, i. p. 160. t Ann. Sci. Nat. sér. 5, xii. (1869) p. 28.
i A violet glass of a peculiar reddish tint in my possession transmits the
whole of the spectrum, and is therefore useless for experimental purposes.

TRANSPIRATION OF PLANTS. 83

Wiesner's * researches led him to draw different conclusions
from those of M. Dehérain, in that it is not the (optically) brightest
part of the spectrum, but those parts which correspond to the
absorption-bands of chlorophyll, which are mainly concerned in
the process. Later observers, including myself, tend to confirm
his views.

A fact of importance, to which he early calls the reader's
attention, is the chief difficulty in carrying out experiments suc-
cessfully, from the ever-varying amount of light (not pure sun-
shine). He illustrated this by the following experiment. He
chose young plants of Maize, having their roots in water, pro-
tected above by a layer of oil to prevent evaporation. He placed
them in one scale of a balance in equilibrium. He then removed
10 milligrammes ; and as soon as the scales were again in equili-
brium he removed 10 more, and so on. Equilibrium was succes-
sively restored after the periods of 6 min. 15 sec.; 7,15; 4,30; 4,20;
7,45 ; and 5,10. The temperature and humidity were constant:
Hence the variations of transpiration could only be accounted for
by the unequal illumination of the sky caused by passing elouds.

He also adds several experiments showing the great differences
which result between the effects of bright sunshine, diffused light,
gas, and in obscurity, and concludes with the remarks :—“ Dans
tous les eas, ces quelques expériences montrent qu'on s'expose à
de graves erreurs quand on étudie l'aetion de la lumière sur la
transpiration sans tenir compte des changements de l'éclairage."

Unfortunately the intensity of light is just the one thing which
at present it is impossible to estimate.

The next point of importance to which Wiesner draws atten-
tion is the part which the ultra-red calorific rays play as a cause of
transpiration. Alluding to M. Dehérain's experiments he says t:—
“ M. Dehérain a exposé une plante dans l'air saturé derrière une
solution d'iode dans le sulfure de carbone qui ne laisse passer
que les rayons calorifiques obscurs, et il en conclut que dans ces
conditions, les rayons obscurs sont pour ainsi dire saus effet sur
la transpiration.

* J'ai fait moi-méme deux séries d'expériences avec de jeunes
Mais et des rameaux d'1f. Seize expériences à la lumière solaire

* “Recherches sur l'influence de la lumière et de la chaleur rayonnante sur
la transpiration des plantes," par M. J. Wiesner, Ann. des Sci. Nat. sér. 6,
iv. (1876), p. 145.

T Ann. Sci. Nat, sér. 6, iv, (1876) p. 159.
a2

S4 REV. G. HENSLOW ON THE

et à la lumière du gaz m'ont conduit à ce résultat: gue les rayons
calorifiques obscurs agissent trós-fortement sur la transpiration, et
que cette influence, relativement à celle des autres rayons du
spectre, est plus grande quand on se sert de la lumière du gaz que
quand on opere à la lumi: re solaire "*.

After detailing his methods of experiment he concludes :—
* TI y a entre la valeur calculée de la transpiration à l’obscurité
et la valeur observée unc differénce de 11 pour 100.

* En passant sur ces erreurs d'expériences, on trouve d'une
manière approximative que pour 100 d'eau transpirée, 79 revien-
nent aux rayons lumineux et ultra-violets, et 27 aux rayons
ealorifiques obscurs.”

It is worth while to recall the fact that Daubeny in 1836 per-
ceived the importance of heat-rays. He attributed the excess of
transpiration under blue, purple, and the cupreous solution, in
great part to the heat radiated from the screen, for “a bottle
filled with water, blaekened with ink to such a degree as to
transmit just as much light so far as could be measured by the
eye, as that filled with the copper solution was found to do,
‘aused an equally considerable amount of water to be evolved by
the plant . . . Now as water, with the addition of a little ink, is
known to absorb the rays proceeding trom all parts of the spec-
trum in an equal ratio, it follows that the effect produced in
either instance must be ascribed to the heat radiated, and not to
any peculiar virtue of the violet extremity in stimulating the
vegetative functions.”

“ Yet,” he concludes, “ the presence of some light seems essen-
tial to the due continuance of the process "t.

* “La flamme du gaz donne bien plus de rayons calorifiques que de rayons
lumineux (Tyndall).”

+ Ann. Sci. Nat. sér. 6, iv. (1876) p. 160. Iam not aware of any other attempts
to ascertain the exact amount of difference between evaporation, due to heat alone,
and to transpiration, due tolight. From some experiments (not yet sufficient for
publication) I have found very marked differences in the amount of water lost
in the same times and conditions, as well as in the rapidity with which desic-
cation takes place under the two processes. Such, at least, was found to obtain
when one half of a living leaf transpired under sunlight; while the other half,
having been suddenly killed by scalding, could only evaporate. I hope to be
able to differentiate these two processes more accurately on a future occasion.
An important paper indirectly bearing upon this subject is one by M. Maquenne,
entitled, ** Recherches sur la détermination des pouvoirs absorbants et diffusifs
des feuilles " (Annales Agronomiques, tom. 6, 1880, p. 321).

A discovery of Maquenne's (which would seem to furnish an important aid

TRANSPIRATION OF PLANTS. 85

In ascertaining the amount of transpiration induced by differ-
ent rays of the spectrum, Wiesner adopted two methods: first,
by placing the plant in certain rays of the spectrum itself; aud,
secondly, under coloured fluids. The results were mutually cor-
roborative.

The following were the results by the first method :—Red gave
136 milligr. per hour; yellow-orange, 122; blue, 146; ultra-
violet, 70; obscurity, 62. He next compared these results with
others from leaves placed as accurately as possible in the parts of
the spectrum corresponding with the absorption-bands of chloro-
phyll. These latter gave the following results :—

Red (with absorption-band No. L) ...... 343 milligr. per hour.
Yellow-orange (between bands II. and LHT.) 32-0 y Es
Green (between IV. and V.) .......... 304 Ee s
Blue (corresponding to band VI.) ...... 387 ii js

From this he concludes :—* Il est done bien évident que ce ne
sont pas les rayons les plus lumineux, les rayons jaunes, qui
favorisent le plus la transpiration; mais que celle faculté est
répartie dans tout le spectre, de telle manière que les rayons les
plus actifs sont précisément ceux qui correspondent aux sept
bandes noires du spectre de la chlorophylle.

“Tl est curieux d'observer la plus forte transpiration daus les
rayons qui correspondent à la bande VI. M. von Wolkoff a
fait voir récemment* que c'est dans cette partie du spectre chlo-
rophyllien que se fait la plus puissante absorption de lumière.

* Les parties du spectre situées entre les bandes d'absorption,
et qui sont toujours plus ou moins obscurcies par le passage à
travers une solution de chlorophylle, ne sont pas sans action sur
la transpiration ; mais cette influence est inféricure à celle des
rayons complètement éteints dans cette solution "t.

That the presence of chlorophyll is intimately connected with
transpiration is clear from the different results obtained by

to distinguish between the two processes evaporation and transpiration) is that
older leaves absorb more heat than the younger ones of the same kind, and
evergreens more than deciduous leaves; yet, according to Dehérain, it is exactly
the reverse which transpire the most. Hence, if this function be identical with
the purely physical process of evaporation, it would be difficult to reconcile the
above fact with it.

* * Die Lichtabsorption in den Chlorophyllésungen :’ Heidelberg. 1876.

f Ann. Sci. Nat. sér. 6, iv. (1876), p. 169.

86 REV. G, HENSLOW ON THE

Wiesner with etiolated and green plants as well as from those
obtained with coloured flowers, to which the reader is referred
for further information*.

My object in pursuing analogous experiments to those of
Wiesner was to try to test his conclusions, using, however,
coloured glasses instead of fluids; and the results I have obtained
all tend to corroborate his.

It is impossible to ascertain absolutely the amount of wafer
transpired which is due to any particular band of colour in the
solar spectrum; for the difficulties are insuperable. To say
nothing of other influences at work to aid in the elimination of
water, such as external and internal variations of temperature,
dryness of the air, and above all, variations in the intensity of
sunlight, it is impossible to get glasses monochromatieally pure
in any colour excepting red. Hence the results cannot be more
than relative or approximately true. Such as they are, how-
ever, I find, like Wiesner, that the largest amount of water
transpired is coincident with those parts of the spectroscope
wherein lie the strongest absorption-bands of chlorophyll.

CHARACTER OF THE COLOURED Grassrs.—bBefore giving the
results of my experiments, it will be desirable to describe the
character of the glasses employed.

Red. This is a pure monochromatic ruby-red, which transmits
no other rays whatever besides red light; and it is that portion
of the spectrum which contains the very strong chlorophyllian
absorption-bands Nos. I. and II. The thickness of the glass is
one tenth of an inch.

Yellow. This glass passes not only the yellow, but all the red
and green rays as well, up to F in the blue; but stops the rest

* Ann. Sci. Nat. sér. 6, iv. (1876), p. 157. For additional information
corroborative of Wiesner the reader may consult ‘Memorie della Reale
Accademia dei Lincei’ (1879-80), and an abstract in ‘Comptes Rendus, xci.
p. 6, Aug. 9th, 1880, the general conclusions of which are as follows :—
“Apart from, and in addition to, other factors which promote evaporation,
the actinic influence of light largely affects transpiration in plants. Plants
transpire more in light than in darkness, and more in proportion to the
intensity of the light. The effect is therefore most marked just after midday.
Only the portion of light absorbed produced this effect ; consequently
highly coloured plants are more affected than others. Plants transpire
least in monochromatic light than in their own colour, and most in the
complementary colour. Thus, a green leaf transpires least in green aud most
in red light, other conditions being the same." (Hort. Journ. 1880.)

TRANSPIRATION OF PLANTS. 87

of the most-refrangible half of the spectrum. As this glass
not only transmits yellow andjgreen which contain the fainter
chlorophylian bands IL, ILL., and IV., but the powerful one
No. l. as well, one would, on à priori grounds, infer that more,
and not less, transpiration would occur with yellow glass than
with red; but such is not usually the case. Moreover, the loss
of water under ordinary clear colourless glass (one twelfth of an
inch in thickness) is sometimes less than under either red or
violet glasses alone; so that from these facts one is led to infer
that the presence of the brightest or yellow rays is an actual
impediment to transpiration, or in some way hinders the action
of the red and violet. The thickness of the yellow glass used is
one twelfth of an inch.

Green. This excludes red and violet rays, but transm'ts
light which includes the position of the chlorophyllian bands III.,
IV., V.,and VI. It is one twelfth of an inch in thickness.

Blue. This has the red end greatly subdued, appearing quite
black in diffused light; butin direct sunlight a broad black baud
is seen in the midst of a small quantity of dull red. It includes
green; buta black band in the position of the chlorophyllian
band No. IV. Hence the glass transmits light which would
include the bands No. V., VI., and VII. "The thickness of the
glass is one tenth of an inch.

Violet. This passes less green, but rather more red, than the
blue glass. The light transmitted would include bands Nos. Ll.,
V. VL, and VH. A dark absorption-band occurs about the
position of No. IV.

Clear. Ordinary transparent colourless glass, of a thickness of
one tenth of an inch.

NATURE OF Experiments.—In commencing my experiments I
at first employed eut shoots and detached leaves of various plants,
as has so generally been done by previous experimenters. lu
every case they were weighed both before and after their expo-
sure for definite periods to daylight transmitted through the
coloured glasses. Different methods were adopted; but the
most satisfactory for short periods, such as are indeed only pos-
sible with detached specimens, was to cut the shoot under distilled
water, inserting it in a small test-tube submerged at the same
time. This secured the cut end from exposure to the air. The
surface of the water in the tube was prevented from evaporating
by a few drops of oil forming a thin layer. The whole is easily

88 REV. G. HENSLOW ON THE

and accurately weighed to the 500th part of a gramme. The tube
with the shoot or leaf thus prepared was enclosed in a box
covered above with the sheet of coloured glass.

Although experimenters have so generally employed cut shoots
and leaves, I soon found, on making my caleulations from data
accumulated from the weights, that although some may appear
quite fresh to the eye for three or even more days after the com-
mencement of the experiment, yet the vitality of the shoot or leaf
had, nevertheless, been becoming enfeebled all the while, and the
amount of water transpired steadily decreased day by day
irrespective of the characters of the coloured glasses used; so
that finally the relative amount of loss in successive days
became untrustworthy.

To ascertain the differences in the amount of water trans-
pired under the influence of different rays of the spectrum, pro-
longed and repeated exposure of the same specimen to the same
kind of light is necessary. The reason for this is obvious; for
there are (as stated above) so many disturbing influences which
may materially affect the results, that unless they be reduced
to a minimum, the effect due to colour alone cannot be even
approximately ascertained.

If, however, the experiments be carried on under conditions
which will reduce the above disturbing elements to a minimum,
then the differences due to the colour of the light transmitted
will be the most powerful agent in the process of transpiration ;
and it is only by taking the mean of many experiments that the
above influences can be virtually eliminated.

In attempting to do this with cut specimens, none will be
found to last long enough unaffected by the lesion to give very
trustworthy results. Hence, although such experiments may
tend to corroborate those obtained by more perfect methods, yet
I do not think it worth while to enumerate more than a brief
series of my own, numerous as they have been, as by themselves
they would uot furnish a sufficiently accurate basis for induction.
They may have their use, however, in showing the somewhat
negative, or at least uncertain, results which are generally only
obtainable from cut specimens.

The next method I adopted was to take small plants with their
roots carefully lifted and freed from soil, and to insert them in
test-tubes as before. The results were even more unsatisfactory
than with cut specimens, for, excepting Radish (the thick root of

TRANSPIRATION OF PLANTS, 89

which presumably acts as a reservoir of water), the others, having
fibrous roots, gave a steadily decreasing loss day by day, showing
no maximum or minimum, except in some cases with clear glass on
the sixth day, on which the loss was greater than in the preceding.

One suspects, therefore, that roots naturally grown in soil
and transferred to water, cannot carry on their function of ab-
sorption in a normal manner beyond a very short period of time.
My experience seems to corroborate that of Sachs*, quoted by
Duchartret :—* Ce physiologiste a reconnu que les racines qui
se sont produites dans la terre ne peuvent végéter ensuite dans
l'eau et que réciproquement celles qui ont pris naissance dans
l'eau ne peuvent remplir leurs fonctions dans la terre."

The plan I finally adopted was to grow small plants in miniature
pots, 2 inches high and nearly 2 inches in diameter. These can
be entirely wrapped up in gutta-percha sheetiug, which is carefully
bound round the stem of the plant with cotton-wool within and
around the stem. — This effectually prevents any evaporation from
the surface of the earth or pot; and allloss of weight is due to
the transpiration from the exposed surface of the plant alonef.

My experiments were made upon Lettuce, Box, Echeveria,
smal] seedling Palms, Ferns, Cactus, and many other kinds of
shrubs and herbs; having selected them with very various de-
grees of density in the epidermis, as well as of different families.
The results would seem to entirely corroborate the conclusion of
Wiesner, that transpiration is mainly effected by the Red, Blue,
aud Violet rays; while the (optically) brightest rays of yellow
and green are generally less able to effect it, even if they do not
hinder it. I emphasize this sentence, as there appear to me to be
grounds for coming to such a conclusion, as will be seen hereafter.

Description or ExrEnIMENTS.— The experiments were all con-
ducted in a room with one window of N. aspect, into which the sun
never entered, except just before setting in midsummer, and then
only at one corner ofthe window. The light was, moreover, par-
tially obscured by foliage of high trees in front of it. The window
was never opened, and the temperature varied but very little, the
maxima ranging from 61? to 66? F. by day, and the minima from

* Bot. Zeit. 1860, p. 113.

f Éléments de Botanique, 2e éd. 1877, p. 289.

1 Even in this case we meet with a difficulty, in that after several days the
absence of air to the roots is liable to cause small and delicate plants to suffer ;
so that other precautions must be taken with them, in not allowing the soil to
be too wet, and in admitting air from time to time.

90 REV. G. HENSLOW ON THE

57° to 61? by night, so that the effects from this cause may be
praetieally put on one side. Similarly the humidity of the air of
the room may be neglected. "They were all earried on between
May and September in the years 1880-85.

The most disturbing element was evidently the variable amount
of light or its intensity ; and this is unfortunately, as remarked,
just the very thing which cannot be estimated*. I have already
quoted Wiesner's experiment, whieh showed how even a passing
cloud may affect the amouut of vapour transpired, so that to
deduce mean results of any value, a very considerable number of
comparative experiments must be made. The following examples
may be taken in illustration. The specimens employed were cut
shoots in test-tubes as described ; and it will be seen that while
in the case of the Euonymus japonicus and Pelargonium the
blue light gives a minimum, yet this result does not agree with
those resulting from more prolonged experimeuts with growing
plants, as will be seen further on, where the mean loss of several
days’ duration is almost invariably greater under blue than
yellow glass; and very rarely did the yellow glass permit of a
greater loss than the red. Again, clear glass usually gives
another maximum, as in the case of the Portugal Laurel and
Fern; but with the Euonymus it gave the same loss per hour
as the green glass, under which it is usually feeble.

1. Experiments during Short Periods with Detached Leaves.
Portugal Laurel.

Duration. | Colour. | Total loss. Loss per | Mean hee |

| | hour. | 3 |

PRENDE | —
| gramme, | gramme. A |

9.53 a.m. to 11.34 A... RB. 995 59 | . 5005. |
11.34 io 44 PNG. Y. -305 -12 600 |
27 PM to. 42 E G ‘335 “14 60:0 |
zd Cw 548 | UR -190 14 610 |
545 to 686 = V. 120 14 600 |
635 to 7.35 Cl. 170 7. | m6 |

* I attended a meeting this spring (1885) at which this subject was dis-
cussed by the leading meteorologists of the day. The general conclusion arriv
at appeared to be that no instrument at present invented can give more than
roughly approximate results of the quantity of direct sunshine recorded ; none
the actual amount of sunlight, and that some instruments in use give very
erroneous results indeed.

t The mean temp. is obtained from the max. and min. of the period.

TRANSPIRATION OF PLANTS. 9]

Asplenium, sp.

| |

| Duration. | Colour. | Total loss. a M | aeons ail) os
| aour. F.

| |

| | gramme. gramme. (s

| 10.15 a.m. to 11.46 a.m...) R. 185 :120 59:5

| 11.46 to Pub | Y :215 088 60:0

| 2.12 r.m. to. 4.37 G. "190 ‘078 60:0

| 4.37 to 5.55 B. 105 092 61:0

| 5.55 to 6.45 z ME "07 084 60:0

| 6.45 to 7.40 sf OL 090 098 59:0

| |

Euonymus japonicus.

| |

Duration. Colour. | Total loss. | r ss per | Mean temp. |

| 10ur. F. |

gramme. | gramme. à |

10:25 sar: to 11.40 al... R. “060 la 3037 59:5 |
11.40 to 2.10 P. M... Ye 045 | 018 60:0
2.10 p.m. to 4.30 G. 035 | “O15 60:0
4.30 to 5.50 B. 010 1 007 61-0
5.50 to 6.38 V. ‘O15 = 018 60:0
6.38 to 7.38 Cl. D15 I 05 59:0

2. Examples of Experiments with Detached Leaves lasting 24 hours

under each glass. Range of temp. 557-57? F.

Aucuba japonica. Adiantum, sp.
Colour. Total loss. Colour. Total loss.
espe 405 HH. i5. 400
Y 40 "285 M. 260
G aa. 245 E vC 15
BI 315 Bi. 12i 265
NI UNDE 325 Ve eee 300
CL oo n 334 OL es 170
Coleus (red-leaved). Pelargonium.
Colour. Total loss. Colour. Total loss.
129540550259 :550 Rees ie 000
NS eee 215 BAUR 265
e. ox. -EZO O T, 235
B5 :180 Boll ees 205
iu UP UD E *335 Vere eek :295
CL s :260 Cl sa. -280

92 REV. d. HENSLOW ON THE

3. Experiments with Rooted Plants in Water for 24 hours under
each glass. Range of temp. 57°-61° F.

Euphorbia. Wallflower.
Colour. Total loss. Colour. Total loss.
Reine 1:47 Be IU 1:91
AV RM E 1:93 Vo a 1°34
G o Ns T25 33 71... ee 1°34
B 5o i$. '82 Bu A 91
WI IE “TD NEUE 15
QI o er ar Qi. vc 81
Grass. | Radish.
Colour. Total loss. Colour. Total loss.

Be iy eee 1:93 | noe IS
Yo UE 1:91 WI UC MALUI TD
G one 1:43 | Gc un LU 1:08
B ela 118 B oa 86
V 1:16 Vi cheers oles “91
(gem 1:36 OL. 5e :99

In these examples, excepting the last one, the loss steadily
decreased from R. to V., showing no maxima, as already stated.
The last one alone shows an increase under violet as well as
clear glass, which may perhaps be accounted for by the suc-
eulent nature of the root, as already suggested.

4. Experiments for Longer Periods.—l selected six young
Lettuce plants on May 15th and May 23rd, 1883, respectively,
from a bed and transferred them to miniature pots. They were
allowed to become well established, and then the pots were com-
pletely invested in gutta-percha sheeting as described. Each
plant was placed under the coloured glasses in succession for 24
hours, with the following results (loss in grammes) :—

Mean Temperature 64°.

No. of | 5 | |
Pot R. X G. | B. | v | Cl.
I 1-499 116 106 | 1:29 La7 | 139
1E 61 07 S6 Oe | 9 | ‘61
HL... 1:05 116 122 | 125 | 216] 99
W 1:43 1:05 tie- | H6 | reo | 172
i Vell 1-09 89 1:02 | 148 | 1:13
YL 3: 1:39 1:62 1-48 1-21 175 1310
Totals... 6:68 63190 | 030 | 0600 | 9473. | 688 |
I ! |

The temperature during the six days oceupied with this experi-
ment varied only eight degrees, from 56° to 64° F.; and to show
how little this affects the results, the temperature for pot 1. under

TRANSPIRATION OF PLANTS. 93

red glass ranged from 59° to 64^; while that for the same plant
under yellow glass was from 59? to 62°; and yet in this instance
the yellow glass gave a rather higher loss than the red. Again,
for No. II., the temperature for red glass varies from 59? to 60°,
that for yellow from 59? to 64°, showing that in both cases the
temperature was lower, when the loss was comparatively greater,
contrary to what would be expected, since it is known that an
elevation of temperature is one of the causes of increasing the
loss of moisture. The differences, therefore, must be put down
to the unascertainable variations in the amount of sunlight.

Hence wil be seen the importance of a protracted series of
experiments; and the results given under the line of * Totals "'
will undoubtedly be a much nearer approximation to the truth.

These give one maximum for the red, and another for blue and
violet, together; while yellow and green together furnish a
decided maximum.

An independent maximum is also supplied by clear glass.
Comparing this with the effects of the yellow glass, which also
transmits red light, it seems that so far from the optically
brightest rays being a chief cause of transpiration, the yellow, aud
those rays on the more-refrangible side of it, where the chloro-
phyll absorption-bands, viz. Nos. II., III., and IV., are feeblest,
must have actually a retarding influence upon the effect of red
and violet lights, which per se are most powerful.

Lettuces (second series).

No. Pe Ye Be et
E. 113 1:05 1:04 115 | 142 | 1:54
IE 81 54 | 102 38 | 108 68
Hr ms dune] | ae 15
I9... 1:89 110 bs. Pot | P0 | 150
Ys 92 | 135 74 43 | i20 | 198
XL 19S | 140 | 129 | im | 241 | 158
Totale.| 790 | 664 | 731 | 675 | 925 | 787

The temperature for the six days ranged from 57° to 66°.

As in the preceding series, one maximum occurs under the red,
another under the violet or most-refrangible end, while yellow is
again a minimum. Green gives a rather higher result than
before. Combining the two series, the mean of the two totals
is as follows :—

R. Xx. G. B. V. CI.

7:29 6:39 6:98 6-70 9°49 7-37

94. REV. G. HENSLOW ON THE

This result clearly shows the preponderating effects of the red
and violet ends of the spectrum ; and they are of course just
where the strongest chlorophyllian absorption-bands occur.

Clear glass gives a decided maximum, though in many cases
the loss is less than under violet alone; hence I would repeat,
and I believe the observation has not been made before, that
yellow light has a decidedly retarding influence upon the amount
of water exhaled by red and violet lights.

Though such is the case with the rapidly transpiring herbaceous
leaves of the Lettuce, it appears to be somewhat different with
Palms and Box, as the following series will show; for out of
ten observations on Palms on seven oceasions, the loss under clear
glass was greater than under violet, and on six occasions greater
than under red. With Lettuces the loss under clear glass was
on as many occasions greater than that under the red glass as
the reverse. In the case of the Box, however, the loss under
red and violet was always greater than under clear glass.
But the preponderance of red over yellow is not so pronounced
as with Lettuces.

Lettuces (third series).
Range of Temperature 53°-58°.

No. R. Y. duo E Cl.
I. 70 62 48 1744 1 207 46
HL. 30 -2 81 | 49 ‘BB 62
A 80 -55 77 | -59 -89 62
wo D mue reci 1:83 70
Yo | 29 43 ‘2 30 -81 59
vuo 144 | 141 | 1:24 | 176 97 | 213
Total...) 4:11 | 328 | 397 | 430 | 612 | 512
Mean..| 68 | 05 | HI | — | 102 -85

The numbers with asterisks are omitted in calculating the
means, as the foliage had become partly flaccid, and they are u2-
doubtedly too high. It was on these two occasions that I first
realized the important difference between evaporation from
dying or dead matter and transpiration from an actively growing
orliving plant. Subsequent experiments have proved that eva-
poration proeeeds much more rapidly when life is enfeebled or
extinct than when the plant is alive. In the latter case this
purely physical process is to some extent kept in check, while
transpiration, so to say, takes its place.

TRANSPIRATION OF PLANTS. 95

Taking the mean of the three series, the results are as
follows :—

R. y- G. S B ME Cl.
5:09 4:46 482 4°70 6:66 5:20

These give decided maxima under red, violet, and clear glass,
with minima under the central portion of the spectrum.
Box.

Six small Box plants, well rooted, were grown in pots as
described, with the following results after twenty-four hours’
exposure in each case.

Range of temperature 57°-65°.

modos ow em NM ROM
DES 91. Oh *& | 9i | I0 | ee
I 137 | 137 | 138 | 193 | 152 | 114
iM. 130 | 88 | 121 | 104 | 148 | 117
a ees 171 | 196 | 263 | 247 | 282 | 143
e 348 | 364 | 372 | 354 | 315 | 240
ao 258 | 291 | 233 | 283 | 267 | 235
Total..| 1119 | lt | 1203 | 1192 | 1264 | 923
Mean..| 1:36 | 1-90 2:00 | 1:98 | 2-10 | 1:54

In this, as in the following case of Palms, the differences are
not so pronounced ; but violet glass still shows a maximum.

Palms.

Three small Palms of undetermined genera and species were
grown and protected from evaporation in the same manner as in
the preceding experiments, and subjected to each of the glasses
for twenty-four hours successively. The losses in weight were as
follows :—

Temperature varied from 60? to 68°.

No. I.
|n at a B. vY a
— I

1:02 1:09 $4 ‘73 95 111

70 97 -90 :96 93 113

1:22 87 '85 '85 99 79

73 88 62 58 71 69

56 63 :53 69 66 81

(Totals.......... | 493 444 374 3:81 494 4:53

96 REV. G. HENSLOW ON THE

No. II
E Tox ad B v Cl
93 | 8° | 110 | 80 79 | 145
174 | 92 46 -88 44 | 105
61 | 50 ‘87 ‘69 72 89
-47 54 -48 34 7] 77
92 | 35 56 | 101 | 63 32
Pub 407 | 324 | 377 | 372 | 359 | 448
No: ILE
| | |
ERR UK Bebo o* ch |
| wl 53 | w 40 | 50 9|
64 | 69 | 47 49 | -70 58 |
47 | -46 | 88 um lg 45 |
34 36 | Al 4l 37 45 |
-54 44 | 32 48 43. 62 |
| Totals....... 259 | 248 | 243 | 315 | 239 | 335 |

Mean of the three totals :—
3:83 3:38 aol 908 3:41 4°12

A result much the same as in the case of the Lettuces. Red,
blue, and violet rays give manifest maxima; yellow and green
minima; while clear yields the highest maximum of all.

These experiments with the Palms show the necessity of
making a prolonged series with the same plants ; for the totals
for No. I. taken alone would lead us to suppose that the yellow
glass was most favourable of all the colours; but variations of sun-
light, coupled with slight variations of temperature, may account
for it. Thus, when the yellow glass gave a loss of ‘97 gramme,
and red only ‘70, the temperature ranged from 63° to 66° with
the former, and only from 61° to 64° with the latter on June 28,
1883. Again, yellow gave a loss of ‘S88 gr. on July 16th, with à
temperature ranging from 57° to 62°; but on the day before
the red glass gave a loss of'75, while the temperature rang
from 59° to 64°; hence, in all probability, it was a duller
day.

These irregularities, however, become eliminated when so large
a series of observations as eighteen for each colour are taken ; for

TRANSPIRATION OF PLANTS. 97

I consider the proof of their elimination to reside in the general
agreement of their results with those of other plants treated in
the same way.
Cactus.
A small specimen of a cylindrieal Cactus, growing in a pot and
protected as described, gave the following results :—

| | RB | ¥ | € | B | V. | OL. | Range of temp.

| | ua bons sone] 08 [do Isis 449 — 59^
OL | 35 | ot | -16 | 05 | 10 409—539?
4s | ar | 10 | 1v 122 ae 549—639

| 13 | 38 () | or L593 | 08 549 —61?

| 32 | 44 | 31 | :03 | -26 | -15 559—649

| i8 le 35 L9 | 12 | 07 549—60?

I |

| Total ...... | ‘84 | -99 | -57 72 | 83 | 61 449— 64°

| Mean Hee | 4S | 46 | ib Poe | 4 p 54^

In this case yellow glass gives a higher loss than red ; and as
this also occurred with Box, and with one of the Palm series,
but with neither of the Lettuce series, one is led to imagine tbat
the character of the epidermis may be an important factor, and
alter the effect of the light. :

Fern.

A small Fern, apparently a species of Asplenium, was experi-
mented with, simultaneously with others. It gave the following
results :—

| R | Y G. 5 1 t | o |
i |
| 78 77 86 72 ue
87 66 -83 60 79 50
76 86 62 65 67 | -75
154 87 Bi | 115 | 108 /| 188
Mota 2.05 | 3-95 296 | 315 | 312 | 326 | 3-41 |
Mean 23... | 99 4 “79 "^8 ‘81 85

This result obviously agrees with the preceding.

Echeveria metallica.

Two plants were treated as the others, and were placed under
each of the coloured glasses for three days. The mean of the
total losses under each colour respectively was as follows :—

LINN. JOURN.—BOTANY, VOL. XXII. H

9S ON THE TRANSPIRATION OF PLANES.

R. Y. G. B. V- CL
“261 :243 204 265 -263 -274

In this case the transpiration, as is well known, is slight, in con-
sequence of the thick cuticle. I have therefore taken it to
three places of decimals ; but here, again, yellow gives a decided
minimum, and clear glass the highest maximum, the other
maxima being but slightly different from each other.

CowcLusioN.— The above experiments, selected from a large
series, seem to me to abundantly prove that Wiesner's results
are correct ; and while recognizing the fact that obscure heat-rays
cause a certain amount of the loss of water by evaporation, that
transpiration per se (theoretically distinct from the purely phy-
sical process of evaporation, which takes place from all moist
surfaces and bodies, dead or alive) is especially, if not solely,
referable to those particular bands of light which are absorbed by
chlorophyll, and that such light, being arrested, is converted into
heat, which then raises the temperature within the tissues and
causes the loss of water. The only additional fact which I have
here advanced, somewhat tentatively, is that yellow light has &
retarding influence upon transpiration, for the reasons given above.
That “life ” has a retarding influence upon evaporation as distinct
from transpiration, I think my experiments (which I hope to
continue hereafter) have distinctly proved.

RELATION BETWEEN ' BLOOM " AND DISTRIBUTION OF 8TOMATA. 99

On the Relation between the * Bloom” on Leaves and the Distri-
bution of the Stomata. By Francis Darwin, F.R.S., F.LS.

[Read 4th February, 1886.]

Sacus* has pointed out that there is a connection between the
distribution of stomata on leaves and their protection from wet by
the wax-like coating commonly known as “ bloom."

He says :—“ It is especially the surfaces of leaves that are
well provided with stomata which seem to be protected against
the adherence of water. The leaves of water-plants such as
the Nympheacex, Polygonum amphibium, Hydrocharis, &c., are
thoroughly wetted on their lower surfaces, which have no
stomata; but water runs off in round drops from the upper
surface, where the stomata occur. The meaning of this fact in
the economy of the plant is evident; the mouths of the stomata
would be closed by prolonged contact with water, and would
thus prevent the rapid ingress and egress of gas." T

In the year 1878 my father was engaged in studying the
bloom on leaves, and it fell to my share to follow up the sug-
gestion of Sachs—that one function of bloom 1s to be found in
the protection of the stomata from wet. The mere fact that
stomata close when the leaf is wetted might lead us to expect
that water interferes with their function, even if we had no
theoretical reasons for believing so. Barthélemy (as quoted in
the Botan. Centralblatt, vol. xix.) has recorded a fact demon-
strating the closure of stomata by water. The leaves of Ne-
lumbium give out bubbles of gas when immersed in water and
exposed to sunshine, but the production of bubbles ceases if the
* bloom ” is removed so that water comes into contact with the
stomata. The conclusion that the closure of the stomata is due
to contact with water must, however, be cautiously received, for
it seems probable, as Garreaut states, that the act of washing

* «Physiologie Végétale ’ (French translation), 1868, p. 178.
+ In Lundstróm's interesting paper, “ Die Anpassungen der Pflanzen an
Regen und Thau” (Nova Acta Reg. Soc. Sci. Upsal. 1884), the author points
out that in Trifolium repens the under surface of the leaflets is wetted by rain,
while the upper bloom-protected surface remains dry. He connects this fact
with the occurrence of stomata on the upper surface,
t Ann. Sci. Nat. 1850.
LINN, JOURN.—BOTANT, VOL. XXII.

100 MR. F. DARWIN ON THE RELATION BETWEEN ‘‘ BLOOM”

off the bloom stops up the stomata. There are certain facts
which go to show that stomata tend to be developed in parts
protected from rain. The well-known fact that in a large num-
ber of leaves the stomata are exclusively on the lower surface,
where they are not so likely to be wetted, is in accordance with
this statement. In vertical leaves, as is well known, the stomata
are equally distributed on the two surfaces*, aud in most such
leaves the surfaces would be equally wetted. An interesting
confirmation of this view may be found in the fact mentioned by
Haberlandt t that lenticels are “fairly equally distributed all
round vertical branches, while on horizontal branches they are
much more numerous on the lower than on the upper side. We
might expect that the function of lenticels would be interfered
with by wetting in the same way as applies to stomata, and it is
therefore of interest to find that (at least in young branches,
according to Haberlandt) they tend to be developed chiefly on
the underside, where they are more or less protected. In
young leaves, which are often more nearly vertical than the
adult leaves of the same speeies, we may perhaps believe that
stomata are protected by not being open at first, but whether
the opening of the stomata corresponds with the assumption of
the horizontal position I cannot say. Von Héhnel has shown
that the eutieular transpiration of young leaves is very large, 80
. that even with closed stomata they may transpire sufficiently.
The idea that the tendency to accumulation of stomata on
the under surface of leaves is an arrangement by which they
are protected against rain, is in accordance with Hohnfeldt’s
observationst. He found that in underground leaves the sto-
mata show no marked tendency to accumulate on the under
surface; indeed the contrary is often the case. This is pre-
cisely what might be expected on the theory that they are deve-
loped on the under surface of aerial leaves as a protective against
rain, &c., since the underground leaves are of course not ex-
posed to such dangers. Hobnfeldt quotes Caruel's remarks on

* In the nearly vertical leaves of Lactuca Scariola I found the stomata on
the morphologically lower side to be the most numerous— exceeding those on
the other side in about the proportion of 140: 100.

t Physiolog. Pflanzenanatomie, p. 317.— Haberlandt gives no explanation of
the fact.

1 As reported in the Botan. Centralblatt, 1880, p. 1161.

AND THE DISTRIBUTION OF THE STOMATA. 101

Passerina hirsuta: the upper surfaces only of the leaves have
stomata, and these surfaces are protected by being pressed closely
to the stem or against other leaves.

With regard to the kind of wetting which is prevented by the
bloom on leaves, it is probable that it is more effective against
rain than against dew, since at least some bloom-protected
leaves are easily covered with a coating of dew in very finely
divided drops. It is curious to observe the effect of immersion
on such a leaf— which may be plunged into water still retaining
its coating of dew, and will be found to be perfectly dry on
being removed from the water—so that the readiest way of
drying such a leaf is to dip it into water. Another fact may
perhaps be mentioned as tending, as far as it goes, to confirm
the view here adyocated—namely, that bloom is frequently ab-
sent from the projecting nervatures, though present on other
parts of leaves—the nervatures being the parts of leaves on
Which stomata are not developed.

Finally, it must be remembered, since bloom certainly has
other functions besides the protection of leaves from wetting,
that it cannot be expected that every species should show a cor-
respondence in the distribution of bloom and stomata. Lund-
strom’s essay, above alluded to, is of interest in this connection
He shows that in some species arrangements exist for the col-
lection and distribution and absorption of rain as it falls on the
leaves; and here the surfaces are so modified, by secretions &c.,
that they are wetted with especial ease. This class of cases
must no doubt have a bearing on the subject of the present
paper. Such cases as those of Trifolium repens and Vaccinium
Vitis-idea, where the lower surface of the leaf is especially wetted
by rain, are of much interest.

The case of Lobelia Erinus also conveys a caution. Here,
according to Lundstróm, there are small scale-leaves at the base
of the flower-stalk which collect the rain on their upper surfaces ;
yet it is on these surfaces that the stomata exclusively occur.
This case is not in reality opposed to my results, since the
stomata function as organs of absorption.

The results here obtained were worked out in the year 1878,
and it was intended to publish them in a work of my father’s to
be devoted to the subject of bloom. I shall on another occasion

give some account of my father’s results, and of some further
12

102 MR. F. DARWIN ON THE RELATION BETWEEN “ BLOOM”

work of my own in the same direction. The present paper, with
the exception of the introductory remarks, is written entirely
from my notes of 1878.

Through the kindness of Sir J. D. Hooker I was enabled to
carry out the work in the Jodrell Laboratory in the Royal
Gardens at Kew. I was thus enabled to obtain a large supply
of fresh material, a necessary condition to the success of my in-
vestigation. It is a pleasure to me to express my sincere thanks
for the opportunity which he so kindly placed at my disposal. I
wish also to express my thanks to Mr. R. Irwin Lynch, now
Curator of the Cambridge Botanic Gardens, but who at that
time held a post at Kew. "Through Mr. Lynch's kind assistance
I was able to obtain good specimens of the many species which
I required. .

I first made a list of species of which the number of stomata
on the leaves have been ce using for this purpose the
works of

Czech ( Botan. Zeitung, 1865).

Kareltschikoff (Bull. de la Soc. des Naturalistes de Moscou, 1866).
Morren (Bull. de l'Académie R. de Belgique, t. xvi.).

Weiss (Pringsheim's Jahrb. vol. iv.).

This list I gave to Mr. Lynch, asking him to supply me with
specimens of species chosen at random from among those in my
list ; I hoped in this way to get some idea of the number of plants
which have bloom-protected leaves. But I was obliged to sup-
plement the list with leaves selected as having bloom, so that no
accurate conclusions of this kind can be drawn from my tables.
In these cases, as well as in many others, I made use of my own
observations on the number of stomata.

Having obtained my specimens, I had merely to note the dis-
tribution of bloom on the surface of the leaves. This is not so
easy as it might seem, since many leaves occur which are per-
fectly dry after immersion, and which, nevertheless, have no true
bloom. In the following tables the leaves noted as having bloom
are such as examination with the naked eye sufficed to classify
among “ bloom” leaves. Bloom is here used to mean a coating
of minute particles of a waxy character which is removable either
by warm water (about 40? C.) or in any case by ether. Leaves
occur which emerge from water dry on the lower surface, but
which have no bloom in the above sense.

AND THE DISTRIBUTION OF THE STOMATA. 103

Between leaves such as these and those which are completely
wetted, a number of gradations occur. We find more or less
perfect developments of a certain character of the surface—
which may be called greasiness. The water coating the surface
shrinks into patches or drops, leaving large areas dry. The leaf,
in fact, behaves towards water like certain metallic or grease-
coated surfaces. The degree of greasiness varies a good deal,
some leaves exhibiting it in great perfection, while others do so
only when the leaf is shaken. The stomata of many of this kind
of leaf must, no doubt, be protected against wet; but as they
graduate into those which are easily wetted, I have placed them
(except those mentioned on p. 108) in the class having no
bloom. I am inclined to think that immersion in water is
not a thoroughly satisfactory test of the leaf's capacity of
resisting natural sources of wet, since some of the “ greasy"
or “metallic ” class of leaf seem to be much more completely
wetted by rain than by a single immersion in water. This is a
further reason for keeping the best protected leaves, those which
have a true bloom, in a separate class. Leaves such as Hydrangea
quercifolia and the Raspberry have their lower surfaces almost
perfectly protected against wetting, but this is due to a fine
coating of hairs. As such cases do not bear on the function of
bloom proper, but only on the general question of the protection
of stomata, they are excluded from my tables.

It must be remembered that bloom is in many cases easily
injured, and that in old leaves it is sometimes washed off by rain
or removed by rubbing against other leaves, &e. It is therefore
possible that some leaves noted as having no bloom may show
this coating under other circumstances ; I can only say that I
have been aware of, and tried to avoid, this error.

In dealing with the results I have been obliged to use a few
simple symbols. The number of the stomata on a leaf is ex-
pressed as a fraction, of which the numerator expresses the
number of stomata on the upper surface and the denominator
the number on the lower. Thus im would mean that the upper
stomata are to the lower as 100 to 150. My figures being taken
from different sources cannot be taken to represent any absolute
number of siomata per square millimetre, but merely the
proportion between the stomata on the upper and lower

surfaces.

104 MR. F. DARWIN ON THE RELATION BETWEEN /' BLOOM ”’

When there are no* stomata on the upper surface the state

. : 0 : :
of things is expressed by &, the letter S being employed in
the denominator to show that the symbol expresses the distribu-
tion of the stomata and not of the bloom, in which latter case the
letter B is employed. According to this plan the floating leaves
of water-plants would be symbolized by 2 :—no stomata on the

under surface. Precisely similar symbols are used to express
the distribution of bloom :—

p Bloom on the upper as well as on the lower surface.
0
g^ Bloom on the lower surface only.

B
97 Bloom on the upper surface only.

Q7 No bloom on either surface.

The leaves examined are divided into four classes according
to the distribution of the bloom, and each class is analysed into
groups according to the distribution of the stomata.

: 0 f
The leaves forming Class > have no bloom on either surface ;

and those in Group 1 have no stomata on the upper surface.

Class Q.—No bloom on either surface.

0
Group l.—Stomata- gs 76 species.

Acer platanoides. Araucaria Bidwellii.
Aconitum Napellus. Asarum europæum.

A. lycoctonum. Asclepias curassavica.
Actæa spicata. Begonia hydrocotylifolia.
Ægopodium Podagraria. B. quinquifolia.

Æsculus Hippocastanum. Betula alba.

Amygdalus communis. Brexia madagascariensis.
A. persica. Broussonettia papyrifera.
Andromeda speciosa. Buxus sempervirens.
Anemone japonica. Camellia japonica.

* In leaves usually described as having no stomata on the upper surface,
there are often a few scattered ones, especially near the veins.

AND THE DISTRIBUTION OF THE STOMATA.

Carpinus betulus.
Celtis occidentalis.
Cerasus Mahaleb.

C. vulgaris.

Citrus Aurantium.
Clematis Viticella.
Cratzegus oxyacantha.
Fagus sylvatica.
Fieus benjamina.

F. cordata.

F. elastica.

Fraxinus excelsior.
Gastonia palmata.
Gentiana cruciata.
Girondia manieata.
Gleditschia triacantha.
Goldfussia glomerata.
Hedera Helix.
Helleborus niger.
Hydrangea hortensis.
H. quercifolia.

Ilex Aquifolium.
Juglans regia.
Kalmia latifolia.
Lilium speciosum.
Maclura aurantiaca.
Mahonia Aquifolium.
Maianthemum bifolium.

i

Mereurialis perennis.
Morus alba.

Negundo fraxinifolia.
Olea europea.
Philadelphus coronarius.
Piper magnolieefolium.
Pittosporum Tobira.
Platanus occidentalis.
Pseonia sp.

Pritzelia zebrina.
Prunus domestica.

P. Laurocerasus.
Pyrus communis.

P. Malus.

Rhodea japonica.

Rhododendron hirsutum.

R. pontieum.

Ribes aureum.

Rubia tinctorum.
Salisburia adiantifolia.
Solanum tuberosuin.
Syringa vulgaris.
Taxus baccata.
Theophrasta Jussieui.
Trollius europzus.
Vinca major.

Vitis vinifera.

105

1 é
Group 2.-—Stomatal formula between 5 and 5; that is to say,

one third, or less than one third of all the stomata are on the

upper surface.

Antirrhinum majus ............ E

Apium Petroselinum............
Asclepias incarnata ............
Aster chinensis ................-.
Atropa Belladonna
Brassica Bapa 3.
Campanula persicifolia .........

G-urücsetolia e. cet

29 species.

Capsella Bursa pastoris

Galium boreale

Geum urbanum

Centaurea phrygia............
Convolvulus arvensis..........-*
Dianthus barbatus......... hiie

D. deltoides <. 2i. A veges?

Gratiola officinalis ...........+- A

vetro an

106 MR. F. DARWIN ON THE RELATION BETWEEN ' BLOOM ”

: 9
Impatiens parviflorum ......... dá
S : 48
Levisticum officinale ............ 9
E 8

Menyanthes trifoliata ......... D
T

Orobus vernus ......... ue E ai
38

9

iPapaver Whwas i .5.0206. 005 << 7
: 18

Phaseolus multiflorus ......... 103
4l

Plumbago Larpentee ............ ioi

: 55

Populus dilatata............... * 370
: 54

E. mgra — Pea ee BIS
m 11-2

E- pyramidalis -n -n 306
. F . 14
Saponaria officinalis ............ Tå
Senecio vulgaris =
garis iiio sins candace 78

; z
Tradescantia subaspera ......... B

Group 3.—Stomatal formula 5 to (excluding = including 3

The stomata on the upper surface form more than a third, but
not more than half, of the whole number of stomata.

species.

Amarantus caudatus ............... ES

Astor DyTOBIOUM ... Loi cere ert
Atraphaxis spinosa
Barbarea vulgaris <c.
Bidons teiparuts oni
Calendula officinalis ...............
Chenopodium Bonus-Henricus ..
Chrysanthemum Leucanthemum
Convolvulus tricolor ...............
Dianthus Seguierii...... d ise eus
Helianthus annuus...
Isatis tinctoria
Lepidium rudoralë- saco
Lychnis Viscaria ..................

Pentstemon barbatus.......... eos

Plantago lanceolata
P; mediano Seer e r
Populus Caroline ...... Rene $e
P. monilifera ....... A S CAE

Rhipsalis crispata
(Sueeulent.)

Eumex Acetosa -ae E e
Scabiosa arvensis
Silene inflata
Statice latifolia ............. resi
Tanacetum vulgare
Valeriana Phu
Viscum album

DII

Yucca aloifolia ........ ET
(Thick and leathery.)

Zea Mays ....... R e

Zinnia elegans

30

AND THE DISTRIBUTION OF THE STOMATA. 107

Group 4.— More stomata on the upper surface than on the

lower: stomatal formula greater than 1- 9 species.

Acacia decipiens .................. yg Vertical.
Achillea Millefolium............... 3
Alisma Plantago ......... B
UU 00
Araucaria: imbricata .— ea a is
Ranunculus Flammula ............ 5 Some leaves have bloom above.
Sedum aro -> e ee 4 Succulent.
So album: e ae ee 7 Suceulent.
: 107
Sempervivum tectorum ............ oa Succulent.
iir 15
Staticelncana esos E

Among the above 143 plants are some which cannot fairly be
compared with the rest, from their different structure. These

e:

* Stomata.

Ana dos 103

ACID dOCIpIeHIB o eea ccisa erinnere echt rores oí

D 10

Ebipsalis crispata -1 es erre ere iea ne T

6

Sed unr acro o a caer e E a A

i 14

S albi ci C ses. ies 7

: 3

Sempervivum tectorum ........... ee eee ee eren A

Du 6

Vacca aloifoliBe- 6e eee osea sui cess pas

Acacia is excluded because of the vertical position of its leaves.
The succulents are obviously not comparable with ordinary leaves
in the matter of stomata distribution. Ina thick fleshy leaf, with
relatively few stomata, the few that do exist may be necessarily
more evenly distributed than in a thin leaf.

If we divide the species into two sets—(A) in which the
stomata on the upper surface are one third or less than one third
of the whole number of stomata; (B) where they are more than
one third of the whole—we have :—

104 out of 137 species, or 76 per cent., in Class A.
34 out of 137 species, or 24 per cent., in Class B.

108 MR. F. DARWIN ON THE RELATION BETWEEN “ BLOOM ”

These results show clearly that, in leaves with no bloom on
either surface, there is a strong tendency towards accumulation
of stomata in the lower surface. This is aiso shown by the fact
that, out of the 137 species, 75 (or 55 per cent.) have no stomata

on the upper surface; that is, in Class o (bloom o^ the group :
are 54 per cent.

0
Class B

Leaves with bloom on the lower surface only.

Here are included a few plants of which the lower surface is of
such a character as to come out very dry, though not actually
covered with bloom; such as Cobæa scandens, Geranium Roberti-
anum, Solanum Dulcamara, &c. 19 species.

Stomata. Remarks.
Apios canadensis -ois s- eeaeee *
E DB oes o e eae ies x
Acer Pseudoplatanus..................... $
ATAOJA uUo, odaie oee $
Ailanthus glandulosus .................. $
Cassia floribunda <. e. neee. H
Qobæa MOIS... iere ct ct H
Daphne Mesereum . ..................... S Young leaves have bloom above,
Geranium Robertianum ............... S
Hut Tarai eese £ :
a ee 8 (This twisted grass-leaf has its
2: o | physically upper side free from
Pharus la DE. asa pcr cee s r1 stomata. (The ^ physically
upper side is morphologically
Polygonatum multiflorum ............ $ ME edens icu
Primula farinosa aoo ae e n i Mealy underneath.
Quercus pedunculata .................. 5

* M. A. de Candolle has been so good as to call my attention to the fact that
Juniper, according to Dutrochet (Mémoire, ii. p. 100), has stomata only on the
upper surfaces of the leaves, and the branches hang down so that the upper
surface of the leaves is downwards.

AND THE DISTRIBUTION OF THE STOMATA. 109

Stomata. Remarks.
Salix babylonica -is e- rie 0

Solanum Düleamatar -e e: ceee —

S : : 0 = B
Sophora japonica e.e cece ress g Young leaves are 5.

B

Tradescantia virginica .................. E

The number of species is small; nevertheless it comes out
fairly clearly that where the under surface is protected against
wet the tendency of the stomata to accumulate on the lower
surface is stronger than in the class where bloom is absent
altogether.

Fifteen leaves, or 83 per cent., have stomata exclusively on the

lower surface ; whereas in Class : only 55 per cent. were in this
condition.

It wil be seen that in the three leaves that bave any,
stomata above, their number does not exceed one fourth of the
whole number of stomata.

Two leaves may be mentioned in which a coating of hairs on
the lower surface serves almost as well as ** bloom " to keep the
leaf dry during immersion in water, and in whieh the stomata are
chiefly or exclusively on the lower surface. These are the Rasp-
berry and Hydrangea quercifolia.

Class B
0
Leaves which have bloom only on the upper surface.

Stomata,
Conyallaria. majas.. -A aeree- EE
Lathyrus pratenBs -oirne =
‘Primula-Auricula — 1 ser ee: $

Ranunculus Flammula.................. = Some leaves have no bloom.

Trifolium repens € — e Bo
te 65
Vicia Cracca ous ee oe Sie 22
x 100
Vi sativa DM M S T 33

Seven species*: of which one has all its stomata on the upper

* It was thought fairer to include only one species of Primula and of Tri-
folium.

110 MR. F. DARWIN ON THE RELATION BETWEEN “ BLOOM”

surface, four have at least twice as many stomata above as below,

: eee 38 (93
while the remaining two leaves have the formula ii (309) and

1 (100):

Class

B

B

Leaves with bloom on both surfaces.

Group 5; or those in which the stomata are exclusively on the

lower surface.

Alstræmeria aurantiaca.

(The stomata are on the phy-
sically lower, but morpho-
logically upper side.)

Aquilegia vulgaris.
Berberis sp.
Bomarea Caldasii.
Chelidonium majus.
Corydalis capnoides,
Cytisus Laburnum.,

They are 16, or 38 per cent.

Euphorbia Cyparissias.

Hypericum calycinum.
(Bad bloom above.)

Linaria vulgaris*.

Oxalis sp.

Poinsettia sp.

Poterium Sanguisorba*.

Rhus sp.

Robinia Pseudacacia.

Rosa sp.

Group $ to =

The next group comprises those in which there are stomata on
the upper surface, but in which they never equal those on the

lower surface in number.
of the whole Class f.

Brassica oleracea .................. Ee
301
is 6
Caladi n lp. ou ci eere ipe aes: ió
: 1:4
Ohloré porfoliata = a o nnna zi
Crambe maritima .— is
; : 0:
Erythrina cristata |. ls
. 5-5
Glaucium bp, =e- i na -
` t 138
Mimosa pudica .........-... eres 302

Lastly, there is the Group :

They are 14 in number, or 33 per cent.

E 82
Pelargonium glaucum ............ vx
1

Ruta graveolens ..................-- $
11

Sonchus oleraceus ................-- 20
: 6
THUS dealbata oe sss T
TEF 95 565
Trifolium pratense ......... 57 OF gi;
: E 47
Tritoma sp: tect Cares ce en ck Pl
Tropzol bout) 2
FOPROLUM 95. 5. (about) 3

to re or those in which the

l

* These species are placed in Group H as the stomata on the lower

surface are so much more numerous than those on the upper.

AND THE DISTRIBUTION OF THE STOMATA. 111

stomata on the upper surface are equal to or more in number
than the lower stomata. They are 12, or 29 per cent.

Artemisia Dracunculus ............ E Lotus corniculatus.................. P.
E 45 3 79
Crinum capense ten i is Lathyrus sylvestris ............... 5s
: : 15 "UE : 118
Desmodium gyrans ...... (about) i Medicago lupulina.................. 100
: T8 e . "
Dianthus spi sco. m = Melilotus officinalis ............... si
Gypsophila perfoliata nie Nicotiana glauca E
ypsop perfoliata e. gs | Nicotiana glauca .................. oa
8 58 pe i
Lmumflyum. — — 0 A Triticum sativum eont; =

The Class E is thus divided into—

2 rani. 6 por cent.,
0 1 i

N to T 33 ”

+ and over... ..<:.. 29 o»

or they may be arranged thus :—

Those in which upper stomata are fewer than lower 71 per cent.
Those in which upper stomata are equal to or more
numerous than the lower stomata..................... 29 ,,

It is well known that succulent plants tend to be covered with
bloom. The following plants are excluded from comparison in
the matter of stomata :—

2
Aloe ArbOreSCODE.-...— (oes E
ý ve 5:5
Beschorneria: Tornellit ........... n. =
3°9

Calanthe crenata. v. 05. sss ix
2:8

Ootyledon bracteosum -nii oi pz

Velumbium speciosum, with stomata only above, and Marsilea,
with more above than below, are also excluded, having the leaves
of a water-plant. It is obvious that if these leaves were included
in the above groups, they would favour my conclusions unfairly.

The three following tables give the results of the analysis of
the above 204 species :—

112 MR. F. DARWIN ON THE RELATION BETWEEN “ BLOOM ”

lst Comparison.

o| 0 b] $
| Bloom—;,. | Bloom — 3. | Bloom= . | Bloom= 3.
er] cei
No stomata above............ | 55 per cent. | 83 per cent. | O per cent. | 38 per cent.
With stomata above......... | 45 IT 5 [100 ,, 02:555

2nd Comparison.

Upper stomata equal to |

or more numerous 5 per cent. | O per cent. | 71 per cent.| 29 per cent.
han LOWER ae. e |
Upper stomata less fre- } É A 2
quent than the lower . 95 2 100, | 29 eee "b
{

3rd Comparison.

|
Stomatal formula 3 or less | 76 per cent.

100 per cent.
Formula greater than 3 ... 24 ,, 0

O per cent. 52 per cent.
s | 49:5 |

39

Remarks on the Comparisons.

Ore. = 0
85 with Class = we see that where, as in 4»

the stomata on the under side of the leaf have the double pro-
tection of position and bloom, the percentage of leaves with
stomata exclusively on the lower surface is decidedly greater—
83 per cent. compared with 55 per cent.

Comparing Clas

Comparing Class : with Class +, we find the number of leaves

having stomata above to be 62 per cent., instead of 45 per cent.
That is, more leaves have some stomata above when the upper
surface is protected by bloom. But it is remarkable that so

many (38 per cent.) of Class = should have all their stomata

below*. This may, perhaps, be explained by the fact that bloom
is in some cases removed by rain, rubbing of other leaves, &c. ;
so that, even where there is bloom above, the lower surface
remains the safest position for the stomata. Nevertheless, we

see that in i there are more species with some stomata on the
upper surface than without any on that surface; whereas in

Classes > and $ the reverse is the case.

* These leaves sem, on the whole, to be thin delicate leaves; so that one
might imagine that stomata on one surface would suffice, and if so, the under
surface would, as suggested, be the best place.

AND THE DISTRIBUTION OF THE STOMATA. 115

If welook at theSecond Comparison, and compare again Class =

with = or ue we see the influence of bloom on the upper surface
determining the distribution of the stomata; for 29 per cent. in
B :

g have the upper stomata equal in number to, or more numerous

than, the lower stomata, as compared with 5 per cent. in m:

or O per cent. in s

To put itin another way. In Class E 45 per cent. have some

stomata on the upper surface; but only one ninth of these have
the number of upper stomata equal to or greater than the number

of lower stomata. In Class E 62 per cent. have some stomata

above, and nearly half of these have the upper stomata equal in
number to, or greater than, the number of the lower ones.

Lookingat the third Comparison, and comparing Class : with

Class $ we see the same accumulation of stomata on the lower

surface, especially when this surface is bloom-protected. Com-

paring Class £ with Class = we see that this tendency dis-

appears when bloom appears on the upper surface. Lastly, in

$ we see accumulation of stomata on the upper surface when

this is the only surface which is bloom-protected.

The number of species in Class 2 is too few to draw any

trustworthy conclusions from. The great tendency to accumu-
lation of stomata on the upper surface is remarkable, especially
when this fact is taken in connection with what we know of

Class 5, where the upper surface is likewise bloom-protected,

but where the tendency to accumulation on the upper surface is
much smaller. It is possible that the distribution of stomata is
determined in this class = by some other circumstances, and that
the development of bloom on the upper surface alone has followed
as protection to the pre-existing stomata; whereas, under other
circumstances, we might imagine bloom to be developed for some
other use, and then the distribution of stomata to be altered

ll4 MR. F. DARWIN ON THE RELATION BETWEEN “ BLOOM”

accordingly. In whatever way we look at the question, the
contrast between Class 3 and Class = in the distribution of
stomata is certainly striking. :

Having been interested by the distribution of stomata in the
Aurieula, I thought it worth while to investigate a few other
species of the genus Primula.

Stomata.
Pod eoruscides © J Or very rare on | Imperfect ** greasiness"
puc ut S | upper surface. on both surfaces.
P. denticulata 3 f Young leaves mealy on both sides; older
< donticülata «ooo I3| ones mealy beneath.
TO pūrparea t. Leere $ “Greasy” above ; good mealiness below.
E ieoa n = Good *' greasiness” on both surfaces.
PXfarnosa 5.35. 00 an Lower side mealy.
P^skkonenss — <50os. sce ees 5 No mealiness on either side.
o.[ Perfect mealiness on the under side ;
P. capitata ao aese & there appear to be traces of meal
on the upper.
DS MOY GUA s sees soos sacs = [us on Dh serine, Dot ee UD
07 | upper surface.
E negr S ines s Imperfect mealiness on upper side only.
P. Auricula T ooe osn s Meal on upper side only.

The meal or bloom on some Primulas ïs so easily washed off or
injured that it is difficult to make certain of the distribution of
the mealiness on plants growing in the open air. Nevertheless,
I think the results are of some little interest:—(i) Where there
is no mealiness, or where the lower surface is so protected, there
is a tendency towards accumulation on the lower surface. When,
as in the species printed in italies,* the meal is exclusively above
or better developed above, we have the stomata exclusively or
chiefly above.

The genus Trifolium appeared to be interesting in the matter of
bloom; I therefore examined the species accessible to me. The
genus (as far as I have examined the species) may be divided

into two groups :—
(i)
(ii)

Those which have bloom on both surfaces.

B
= Those which have bloom only above.

AND THE DISTRIBUTION OF THE STOMATA. 115

B
Group 1. B

Trifolium.
DM nei cone 00.

"T Ancher 0S4 2 Bd

100

77 1

T elegans. i aaa y M
T. Ea pinaster kits ig i
JT medium 2.0. se $ io)
T MINUS eee 21 x S Ix
P. pannonicum — —— = io
E. pratense. o ee 9: iet
T striatum. ees i 300

Two species have all the stomata below. Omitting these, the

- 109
average stomatal formula is 15.

AB
Group ii. 0°

Stomata. Taking the stomata on
DOMO — the lower surface= 100.
S

8
E. alpestre -A i re 0 T 0
T. cmspitosum 0. 35 100
: 100 180
T. fragiferum —.-...—- a 15 100
68 243
T. montanum . e ea 38 100
T 160 163
+ repons.. reer eri 98 100
: 86 344
r E
F: uniflorum «s = 100

One species (alpestre) has all its stomata on the upper surface
—the reverse of what holds good in Group i. = where two species

have all their stomata below. Omitting T. alpestre, the average
227
100°
Thus, when the bloom is on the upper surface only, the average
number of stomata on the upper surface (as compared with the
lower) is twice as great as in those cases where both surfaces are
covered with bloom.

LINN. JOURN.—BOTANY, VOL. XXII.

is

K

116 MR. H. N. RIDLEY ON THE

Expressed in symbols :

Trifolium.
Group = Group 2
Stomata= = Stomata = zd

This confirms my former result, viz. thát when bloom is deve-
loped on the upper surface only, there is a strong tendency to
accumulation of stomata on that surface.

Finally, it may be pointed out that the genus Trifolium confirms
the generalization above made, that when bloom is on both surfaces
a large proportion of the stomata appear on the upper surface ;
although some cases occur in which the stomata occur eaclusively
ou the lower surface.

On Dr. Fox’s Collection of Orcids from Madagascar, along with
some obtained by the hoy/A Baron, F.L.S., from the same
island. By Henry N. Rivtey, M.A., F.L.S., Assistant Bota-
nical Department, British Museum.

[Read 17th December, 1885.]

I nave to thank the authorities at Kew for placing in my hands
for description a collection of Orchids recently made by Dr. Fox
in the neighbourhood of Imerina. There were in the collection
about 50 numbers and almost as many species, the larger number
of which were plants previously obtained by one or other of the
botanists who have recently been exploring almost the identical
locality. There is, however, on the whole, a very large proportion
of novelties, and of these several are peculiarly interesting. The
collection adds three new genera to the flora, viz. Arnottia, à
Mauritian genus of perhaps two species, which it was to have
been expected would occur also in Madagascar ; the very curious
genus Brownleea, hitherto only known from Southern Africa,
where it is represented by 3 or 4 species; aud Holothrix, an
East- and South-African genus. The Madagascar species is à
very remarkable one, differing at first sight from the rest of
the genus in many striking particulars. Another interesting
novelty is Satyrium gigas, probably much the largest species of
the genus known.

ORCHIDS OF MADAGASCAR. HT

It is clear from this collection that much yet remains to be
done in the island, even in the best-worked localities, and indeed
there are still several species known only from Thouars's collec-
tions, that are not represented in the herbaria of this country.

I have incorporated also with the account of these plants some
notes and descriptionsof other new and interesting species recently
received from Mr. Baron in Central Madagascar.

BULBOPHYLLUM APPROXIMATUM, n. sp.—Rhizoma validum breve,
"pseudobulbis conicis monophyllis uncialibus seriatim approximatis;
folia elliptica spathulata obtusa 2 uncias longa, j lata; scapi
nutantes 5-unciales ; racemus laxiuseulus multiflorus ; flores iis
B. Baroni, Ridl., majores ; bractez ovate cucullate quam ovarium
longiores, 3 uncis longs ; sepala ovata lanceolata omnia libera;
petala subsimilia breviora obtusa; labelli lamina tenuis oblonga
* obtusa apice biloba; columns dentibus acutis brevibus.

Central Madagascar, Baron no. 4128.

Allied to B. Baronii, Ridl., but a larger plant. Flowers bigger
and fewer; lip thin, with the edges curled in beneath, and a

bilobed apex.

BULBOPHYLLUM Baroni, ARidl—Ankeramadinka. Flowers
white. Dr. Fox.

BULBOPHYLLUM CONCHIDIOIDES, n. sp.— Planta pusilla cæspi-
tosa, pseudobulbis ovoideis monophyllis 74; uncie altis, haud
remotis; folia elliptica ovata crassiuscula ferme # unciæ longa
1 uncie lata; scapi tenues gracillimi teretes unciam alti biflori ;
bractes ovate quam pedicelli brevissimi longiores; flores parvi
3 unciæ longi, pallidi (sicci); sepalum ovatum lanceolatum ob-
tusum, lateralia lanceolata subacuta, linea mediana obscuriore ;
petala ovata spathulata, $ lateralium zqualia; labellum glabrum,
lamina ovata, marginibus subtus involutis, ungue brevi lato;
columna brevis obliqua, alis latis tenuibus, dentibus brevibus
acutis; anthera ovato-obtusa.

Central Madagascar, Baron no. 4471.

"This species is allied to B. seychellarum, Reichb. f., differing in
its smooth lip &c. The lip is ovate or almost ovate-lanceolate, but
in the natural state the margins are so much curled in on the
underside that it appears to be hastate. The texture is thin, and
it appears to be of a reddish hue, while the rest of the flower seems

to have been pale coloured.
K 2

118 MR. H. N. RIDLEY ON THE

BULBOPHYLLUM (sect. BULBOPHYLLARIA) OPHIUCHUS, n. sp.—
Rhizoma crassum, vaginis pallidis membranaceis tectum ; pseudo-
bulbus angustus elongatus parum incrassatus biuncialis, vaginis 2
membranaceis pallidis amplis tectus. Folia circiter 4, lanceo-
lata obtusa coriacea, apice vix biloba 6 uncias longa $ unciæ lata ;
scapus elatus 16 uncias longus, vaginis 6 dissitis membranaceis
punctulatis laxis 14 uncias longis ; racemus densus erectus multi-
florus 5-uncialis, rhachide carnosula ; flores parvi virides (siccati
rubescentes) ; bracteæ ovate obtuse, floribus vix æquantes ; sepa;
lum postieum lanceolatum angustum, lateralia connata carnosula
ovata; petala linearia spathulata papillosa abrupte recurva ;
labellum breve, lamina ovata hastata obtusa carnosa rugosula,
subtus quasi complicata quam sepala multo brevior, unguis
angustus linearis ; columna brevissima, brachia brevia incurva,
rostellum breve obtusum ; anthera complanata biloculata, loculis
valde discretis.

Ankeramadinka. Flowers green, turning red on fading.—
Dr. Fox; Baron no. 4468.

This is the first recorded Madagascar species of the section
Bulbophyllaria, which is characterized by the lateral sepals being
connate into a somewhat fleshy ovate body. The petals are
narrow and flat, with the ends a little broadened, covered with
small warts, and folded back on themselves. They are apparently
darker than the rest of the flower.

BULBOPHYLLUM GRACILE, Thouars, Orch. Iles Afr. t. 101.

Rhizoma lignosum ; pseudobulbi dissiti oblongi conici 13 uncias
longi, vagina albescente, pseudobulbum et scapi basin tegente ;
folia anguste ligulata spathulata obtusa 3-5 uncias longa,
vix 3 uncis lata; scapus erectus vix pedalis validulus teres,
parte inferiore vaginis paucis laxis dissitis, superiore laxe
racemosa ; bracteze parve quam ovaria breviores ovate lanceo-
late acute ; flores subsessiles ; ovaria brevia conica ; sepala ovata
subacuta, lateralia breviter connata; petala multo breviora lan-
ceolata lorata recta obtusa; labellum breve integrum oblongum
subacutum abrupte convolutum carnosulum, lineis medianis
tribus ; columna alis latis, dentibus minimis obscuris; capsula
ovata globosa 3 uncis longa.

Central Madagascar, Baron no. 4129. Ins. Mascarenenses, Sir
J. Macgregor.

I believe this to be the plant intended by Thouars'sfigure. It
is closely allied to B. ophiuchus, mihi.

ORCHIDS OF MADAGASCAR. 119

BULBOPHYLLUM cORIOPHORUM, n. sp.—Rhizoma $ uncis cras-
sum lignosum, vaginis brunneis membranaceis tectum ; pseudo-
bulbus oblongus conicus 24-uncialis ; folia bina ? lanceolata, apice
biloba 7 uncias longa 14 lata; scapus validulus erectus teres pe-
dalis, vaginis laxis dissitis quarum suprema ampliata cucullata ;
racemus densus conicus, rhachide incrassata earnosa ; flores parvi
copiosi; bractes floribus sequales ovate subobtuse ; sepalum pos-
ticum lanceolatum obtusum quam labellum brevius, lateralia
multo majora connata, apicibus solo liberis, carnosa semiovata,
medio depresso, marginibus exterioribus acute carinatis, carina
denticulata; petala quam sepalum postieum breviora anguste
linearia lanceolata; labellum parvum earnosulum, lamina cordi-
formi; columna brevis, dentibus parvis ; anthera compressa.

Comoro Islands, Humblot no. 337.

This plant is a very singular species, allied, as Professor Rei-
chenbach (‘ Flora,’ 1885, p. 544) intimates, to B. conitum,
Thouars. The flowers in his specimen were deticient, so that he
was unable to describe it. Those in the herbarium of the British
Museum are in very good condition, and show that the plant 1s
distinct from Thouars’s species, which, however, I only know from
his figure. In this plant, as iu the rest of the section, the
lateral sepals are connate and form an oval blunt body with a
cireular depressed space in the centre, in which lie the lip,
petals, and columns. Round the depression the sides of the
sepals are thickened and fleshy, while the outer edges taper
off to a thin irregularly denticulate keel, while only the tips
are free. In Thouars’s figure of B. conitum I see no keel
nor denticulate margin; the dorsal sepal, again, is repre-
sented as being longer than the lip, while in this plant it is
shorter, and the column is figured without teeth, which are cer-
tainly present in this plant, although they are small. Lastly,
B. conitum appears to have but a few flowers sunk in the fleshy
rhachis, whereas B. coriophorum has a very large number. On
the whole, therefore, I conclude that this is not the plant intended
by Thouars under the name of B. conitum. The small flattened
flowers with the denticulate margins bear some resemblance to
some species of Tingis (Hemiptera), hence the specific name.

LIPARIS PUNCTICULATA, n. sp.— Caulis elatus 10-uncialis gra-
cilis basi vix incrassata, vaginis membranaceis puncticulatis striatis
dissitis multis 14 uncias longis tectus; folia 8-4 dissita patula

120 MR. H. N. RIDLEY ON THE

ovata lanceolata vel lanceolata acuta, 3 uncias longa 1 unciam lata
(quo latissima), striata; scapus gracilis angulatus pedalis flexuosus;
bracteze remotze lanceolate acute ; flores olivacei, verosimiliter iis
L. longicaulis equales ; sepalum posticum ligulatum angustum ;
lateralia late lanceolata obtusa 4-nervia falcata; labellum cuneatum
bilobum, ungue angusto longo, lobis rotundatis divaricatis denticu-
latis; columna suberecta apice curvula, alis brevissimis obscuris ;
pedicelli graciles erecti, quam bractez multo longiores.

Central Madagascar, Baron no. 4334.

This is a very remarkable species, allied most closely to L.
longicaulis, Ridl. Like that species it has a tall slender stem
hardly at all dilated into a pseudobulb, but covered with a number
of loose whitish membranous sheaths, which in this species do not
overlap, there being a short but distinct space between each
sheath. The sheaths are minutely dotted. The leaves, usually 4
in number, at the top of the stem, are a little separated from each
other, and seem less thin in texture than those of L. longicaulis.
The scape rises to a considerable height and is usually flexuous ;
numerous bracts are scattered throughout its whole length. The
flowers are chiefly remarkable for the shape of the lip, which
recalls somewhat that of Z. crenulata, Lindl., having a long
narrow base ending in a bilobed lamina, the ends of the lobes of
which are denticulate. The claw lies parallel to the column,
and at the bend there is a kind of rhombiform depression formed
by two low ridges which run towards each other and almost
meet. Down the middle of this depression run two veins slightly
raised and sinuous. This rhombiform depression is, in the dry
plant, of a darker colour than the rest of the lip. The stems of
the plant, which is doubtless terrestrial, grow close together.

Liparis LONGICAULIS, Ridl., was obtained at Ankeramadinka
by Dr. Fox; and L. rrAvEscENs, Lindl., at Ambatovory.

EULOPHIA MACRA, n. sp.—Pseudobulbus oblongus subconicus
biuncialis ; folia 4 angusta linearia ultra pedalia (ineompleta) 1
uneiz lata ; scapus circiter tripedalis, vaginis dissitis magnis ferme
3-uneialibus membranaceis acuminatis ; panicula laxa diffusa,
ramis tenuibus dissitis; bracteæ ramulorum angustissime lan-
ceolatw acuminate pallide, florales minute; flores parvi, iis .E.
luride equales; sepala lanceolata acuta; petala subsimilia paullis-
per latiora, striis 4 obscuris conspicuis; labellum obovatum
cuneatum margine fimbriato, venis medianis parce cristatis ; calcar

ORCHIDS OF MADAGASCAR. 121

breve clavatum saccatum obtusum, ovario multo brevius; columna
brevis; authera subconica, apice obscura.

Central Madagascar, Baron no. 3423.

A straggling plant with long narrow leaves and a lax thin
panicle of small flowers. It seems to be most nearly allied to
the West-African E. lurida, Lindl.

POLYSTACHYA FUSIFORMIS, ZLindl.—Dendrobium fusiforme,
Thouars. Angavokely, near Ambatovory, Dr. Fox, no. 39. To
this species must be referred, I fear, my P. minutiflora. I was
misled by a couple of flowers in the British-Museum Herbarium
received from Du Petit Thouars as those of P. fusiformis, but
which seem rather to belong to P. cultrata, Lindl.

POLYSTACHA ROSEA, Ridl., with flowers both purple and white,
was obtained at Ambatovory by Dr. Fox.

AERANTHUS POLYANTHEMUS, n. sp.—Radices lati crassiusculi ;
caulis biuncialis; folia angusta ligulata obtusa vix biloba striata,
circiter 7 uncias longa 3 uncis lata; racemi graciles e basi caulis
orti penduli 5-unciales, 9-flori; flores laxi mediocres virides-
centes inexpansi ; bractez lanceolate dimidio ovarii brevis sub-
æquales; sepala lanceolata acuminata, lateralia basi angulata ;
petala subsimilia equalia, basi dilatata vix angulata, ?-uncialia ;
labellum basi concavum, marginibus brevissime lobulatis, medio
perforato, lamina lanceolata subacuminata multinervia basi an-
gustata; columna brevissima semiteres, rostellum breve, stigma
semilunare; calcar breve crassiusculum obtusum incurvum.

Ambatovory, Imerina, no. 32. Flowers nearly transparent,
greenish yellow.—Dr. Foz.

This species, a true Aeranthus, differs from A. grandiflorus and
A. Arachnites in its closer raceme of many small green flowers.
The lip is adnate to the base of the column and its base forms a
somewhat deep depression, on each edge of which is a short blunt
lobe. Just before the lip expands into its thin lanceolate pointed
lamina, the basal depression is perforated by the entrance to the
short blunt incurved spur. The blade of the lip is thrown up so
as to be approximately parallel to the petals, and forms an angle
with the pedicel. The lateral sepals have an abrupt angle, almost

a lobe at the base.
Angraecum palmiferum, Thouars, is, I suspect, a near ally of

122 MR. H. N. RIDLEY ON THE

this plant; but I have only seen a single flower of it, in the
British-Museum Herbarium, not in sufficiently good condition to
base much upon.

Both the other species of Acranthus, viz. A. grandiflorus and
A. Arachnites, were also obtained by Dr. Fox at Ambatovory.

fEonts ROSEA, Ridley. Ambatovory, Dr. Fox, no. 26. This
differed from the type in having a white lip. Some time ago I
was shown at Kew a coloured drawing of a similar variety.

MysTACIDIUM VIRIDE, n. sp.—Radices longi tenues; caulis
subnullus; folia brevia oblonga ovata, 4 uncie longa $ lata,
obtusa obscure inequaliter biloba; scapi unciam longi tenues;
flos singulus parvus viridis ; bractea ovata acuta, quam pedicellus
tenuis multo brevior; sepala atque petala anguste linearia re-
curva; labellum ovatum acuminatum cymbiforme concavum;
calcar filiforme attenuatum ferme 13 uncias longum; columna
brevissima; anthera depressa; polliniorum caudieule duo di-
stinete complanate, discus singulus; pedicellus ovario bis
longior.

Ankeramadinka, Dr. Fox.

This little plant, with its short blunt leaves aud solitary flowers,
is remarkable for the lip being continued directly into the spur
without any break or bend at its junction. The long slender
spur, too, looks quite out of proportion to the rest of the flower.
Among Deans Cowan's drawings is a figure of a plant from
“trees, Ankafana,” which seems identical with this species. The
flowers are light green.

Anerecum CarcEOLUS, Thouars, Orch. Iles Afr. p. 78.—
Aeranthus? Calceolus, S. Moore, Flor. Maur. 4 Seych. p.393.
Was obtained at Ambatovory by Dr. Fox. It has not hitherto
been recorded for Madagascar, but occurs both in Mauritius and
Bourbon.

GnAMMATOPHYLLUM RGMPLERIANUM, Reichd. fil. in Gard.
Chron. 1876, p. 240, was omitted from my previous list of
Madagascar Orchids. I have not seen it.

CxNosoRcHIS SPECIOSA, n. sp.—Tubera ignota; folia bina lan-
ceolata lorata augusta, usque ad 10 uncias longa $ uncis lata;
scapus ultra pedalis validulus; vagina singula ampliata acumi-
nata; racemus multiflorus laxus; flores magni, iis C. uniflore
ferme equales; bracteæ lanceolate acuminate circiter # uncle

ORCHIDS OF MADAGASCAR. 123

long; sepala oblonga ovata obtusa lateraliafalcatula; petala multo
angustiora lanceolata acuta; labellum magnum quadrilobum,
unciam longum, lobis oblongis obtusis, apice paullisper dilatatis ;
calear quam ovarium brevius pendulum versus apicem attenua-
tum; anthera dorso carinata; rostellum quadratum latum erec-
tum quam anthere apices multo brevius, antheram ferme
superans; anthers apices porrectze anguste subacute ; stigmata
ferme celata. :

Ambatovory, no. 18, Dr. Fox.

This is allied to C. angustipetala, but differs in the foliage,
lower racemes, &e. There are about twelve flowers in the
race. ‘ White, or sometimes pale pink, with a violet central
spot.”

CYNOSORCHIS GLANDULOSA, n. sp.—Folia bina lanceolata acuta
flaccida, 6 uncias longa vix unciam fata; caulis debilis pedalis
glanduloso-pubescens, vagina singula lamina lanceolata acuminata
pubescens; racemns laxus multiflorus ; bractez lanceolate acumi-
nate, quam flores longiores glanduloso-pubescentes ; flores parvi,
sepala et petala subsimilia lanceolata ; labellum cuneatum bifidum,
calcar eylindricum versus apicem attenuatum ovari glanduloso-
pubescente, multo brevius; antherz apices pro genere breves,
ferme recti; rostellum angustum ; lobi ovati duo ad columns
latera.

Angavokely ; Ambatovory. “ Dark-green, subpilose," Dr. Fox.

This is a near ally to C. lilacina, Ridl.; but is easily distin-
guished by its lanceolate leaves, smaller flowers, bifid lip, and
short points to the anther. The whole plant, except the radical
leaves and perianth, is remarkably glandular-pilose.

CYNOSORCHIS AURANTIACA, n. sp.—Tubera parva; folium sin-
gulum angustum lineare, quam caulis multo brevius ; eaulis gra-
cillima hispida presertim ad basin, vaginis tribus dissitis hispidis ;
racemus laxus; flores 8-9, iis Bicornelle parviflore æquantes,
aurantiaco-flavescentes ; bracteze lanceolate acuminata, 3 ovarii
- seouantes ; sepalum posticum galeatum ovatum acuminatum, late-
ralia angusta lanceolata deflexa; petala lanceolata acuminata ;
labellum trilobum porrectum, lobis lateralibus angustis linearibus
faleatulis, medio longiore cuneato; calcar basi dilatatum hine
attenuatum curvum; antheræ apices porrecti, angusti, latera ad
petalorum bases adnata; ovarium glanduloso-pubescens, quam
calcar longius.

Ankeramadinka. “ Lip flat, lowers orange-yellow," Dr. Foz.

124 ME. H. N. RIDLEY ON THE

CYNOSORCHIS TENELLA, n. sp.— Gracilis; tubera parva; folia 4»
radicalia oblonga ovata acuta 1-13 uncias longa } lata; caulis
tenuis glaber, vaginis duabus acuminatis dissitis, 4-6 uncias
longus; flores 4—5 dissiti parvi; bractez ovate acuminate vagi-
nantes pedicellis equales; sepalum posticum ovatum acutum,
lateralia majora falcata oblonga obtusa patula; petala lanceolata ;
labellum porrectum trilobum, lobis lateralibus curvis subaeutis,
medio oblongo obtuso longiore; ealear breve clavatum, sepalis
lateralibus subzequilongum ; antherz apices breviusculi curvi ob-
tusi; rostellum subsquale ; stigmata ovalia carnosa, haud celata ;
ovarium parce pubescens.

Ambatovory, Dr. Fox. Mohely, Comoro Isles, Boivin.

This little plant has much the appearanee of Zabenaria tener-
rima, Ridl.; but is a true Cynosorchis. The other species obtained
by Dr. Fox were C. grandiflora, Ridl., Ambatovory; C. flexuosa,
Lindl., Ankeramadinka; C. fastigiata, Lindl., Angavokely ; C.
lilacina, Ridl., Angavokely and Ambatovory.

HABENARIA CONOPODES, n. sp.— Caulis 18-uncialis foliatus ;
folia dissita patentia laneeolata acuminata longe vaginantia, basi
amplexicaulia, longissimum 3 uncias longum, 4 uncis latum;
racemus laxus ; flores iis H. arachnoidis æquales ; bractex semi-
unciam longs, lanceolate acuminate, ovario zquales; sepalum
posticum lanceolatum obtusum erectum, lateralia subacuta
deflexa; petala bifida, lacinia postiea erecta linearis, antica an-
gustior decurva; labellum trifidum, basi cuneata brevissima
quam stigmata breviore, lacini: longc subequales, angustissim?
lineares; calcar pendulum ovario subequale clavatum acumina-
tum, apice paullo curvo; apices anthere breves obtusi ; stigmata
longa porrecta obtusa, apice incrassata ; lobi columnares laterales
parvi ovati erecti obtusi. Planta sicea nigrieans.

Ambatovory, Dr. Fox. ;

This plant is allied to H. arachnoides and it has very much
ihe habit of H. lancifolia, Richard.

Hapenarta Foxit, n. sp.— Radices copiose longe, tubera
ovalia; caulis 6-18 uneias longus, validulus foliatus; folia 5-6
ovato-lanceolata acuminata patula, 5 uncias longa 14 lata; race-
mus laxus; flores plures mediocres virides; bractes inferiores
lanceolatze foliacez, superiores breviores angustiores, ovariis multo
breviores ; sepalum posticum erectum ovatum lanceolatum, late-
ralia subsimilia deflexa ; petala bifida, laciniis angustis linearibus

ORCHIDS OF MADAGASCAR. 125

ferme equalibus, postica quam antica paullo latiore et longiore ;
labellum trilobum, lobis angustis linearibus obtusis subsequalibus ;
calcar elongatum clavatum porrectum, 3 uncis longum, quam
ovarium paullo brevius; apices antheræ breves oblongi obtusi;
lobi column: laterales ovati breves; stigmata elongata porrecta
oblonga obtusa, apice paullo dilatata; ovarium angustum, #
unci:z longum.

Shady woods, Ambatovory, Dr. Fox.

HABENARIA ICHNEUMONIFORMIS, n. sp.—Semipedalis; foliaradi-
ealia2-uncialia lanceolata acuta angusta, caulina lanceolata dissita ;
flores pauci 4—5, laxi parvuli ; sepalum posticum ovatum lanceo-
latum, lateralia lanceolata deflexa; petala ligulata erecta subob-
tusa ; labellum profunde trilobum, basi brevissima, laciniis linea-
ribus media quam laterales paullulo longiore, carnosulis ; calcar
elongatum ovarium multo superans filiforme, versus apicem dila-
tatum, hine aeuminatum ; anthera gracilis, apiees elongati curvi ;
rostelli lobus medius longus recurvus; stigmata brevia oblonga
papillosa ad latera calcaris ostii ; lobi columnares exteriores breves
subtriangulares papillosi.

Central Madagascar, Baron, no. 3879.

The following species of Habenaria were also represented in
Dr. Fox's collection :— ZZ. truncata, Lindl, H. Hildebrandtii,
Ridl., and H. cirrhata, at Ambatovory. H. tenerrima, Ridl., at
Ankeramadinka. Amphorchis calcarata, Thouars, was found at
Ambatovory. Bicornella gracilis, Lindl., was common in boggy
places at Ankeramadinka and Ambatovory.

HororHRIX GLABERRIMA, n. sp.—Tubera angusta elongata;
folium singulum lanceolatum aeutum glabrum 5-nervium ; caulis
erectus glaber subpedalis, vaginis longe acuminatis appressis
tectus; racemus multiflorus densus secundus; flores parvi fla-
vescenti-virides copiosi glabri; bractes ovate lanceolate acumi-
nate ; sepala angusta linearia obtusa; petala latiora lanceolata
acuta, basi antica angulata; labellum trilobum basi angusta, lobi
laterales breves falcatuli, medius lanceus acutus; calcar pen-
dulum rectum acuminatum, ovarium superans; columna brevis,
anthera subacuta, loculis discretis ; rostellum breve subteres sub-
erectum ; lobi laterales columns (auricule Benthamii) rotundati
papillosi.

Ambatovory, Imerina. Rare; open elevated downs, Dr. Fox.

126 MR. H. N. RIDLEY ON THE

This plant would almost form the type of a new subsection
of the genus, differing, as it does, from the rest of the genus
in the entire absence of any pubescence, and the erect lanceolate
leaf; the rest of the species, as far as known to me, having one
or two ovate or orbicular radical leaves lying flat upon the ground.
The flowers are rather larger than in most species, densely packed
in a secund raceme. According to a drawing made by Deans”
Cowan, they are greenish-yellow in colour. The lip is three-lobed,
the Jateral lobes shorter and narrower than the lanceate triangular
middle one. There are, as in most species of the genus, two
small oval granular bodies on either side of the column, homo-
logous to the lateral lobes, or auricles, occurring in some of
the Habenarie. It seems likely that they are rudiments of the
two lost stamens.

SaTYRIUM GIGAS, n. sp.—Caulis validus elatus foliatus tripe-
dalis ; folia inferiora lanceolata oblonga plicata magna, 10 uncias
longa, 3 uncias lata, superiora sensim in bracteas attenuata;
racemus compactus multiflorus ; flores pro genere magni; bractex
lanceolate acuminate tandem deflexz, flores superantes ; sepa-
lum postieum anguste lineare, lateralia falcata obtusissima
minutissime ciliata; petala sepalo postico subsimilia sub:equalia :
labellum ovatum obtusum; calcaria longissima filiformia 1j
uncias longa recurva; columna longa teres; anthera apiculata
granulosa ; rostellum subtriangulare latiusculum ; stigmata biloba
ovata elliptiea.

Ambatovory, Imerina, Dr. Fox.

This is certainly the largest Satyrium known to me. The
great size, long curved spurs, and narrow petals make it an easily
distinguishable plant. Other Satyria obtained by Dr. Fox
were :—S. trinerve, Lindl., common in damp places, Andrango-
loaka, Ambatovory, &e.; and S. amenum, Lindl., Ambatovory,
not hitherto recorded from Madagasear.

BROWNLEEA MADAGASCARICA, n. sp.— Tubera oblonga lanata ;
caulis ultra pedalis debilis ; folia 3 dissita lanceolata acuminata
amplexicaulia, 2 uncias longa, unciam lata, tenuia ; racemus laxus
5-7-florus ; bracteze foliacee lanceolate, ovariis subequales vel
(inferiores) longiores ; sepalum postieum petalis (dorsis nec
marginibus) adnatum, apicibus liberis, galeam ovatam formans;
calcar longum ceylindrieum reeurvum, apice obtuso-truncalo ;

ORCHIDS OF MADAGASCAR. 197

sepala lateralia lanceolata subfalcata ; labellum brevissimum, lobo
medio erecto lineari elavato; columna brevissima; anthera in
dorsam reflexa, apicibus breviusculis.

Imerina, Dr. For.

This plant belongs to a small South-African genus not hitherto
recorded from Madagascar. The present species is most closely
allied to B. cerulea, Harv., differing notably in the petals being
adnate to the upper sepal for the best part of their length. The
petals are not attached by the margins as is usual in a galea
formed of the sepals and petals, but by their backs; so that on
the inside of the galea the edges of the petals are seen in the form
of free ridges. The flowers appear to have been blue.

PraATYCORYNE PERVILLEI, Reichb. f. Specimens from Baron,
nos. 4545, 4579, agree very well with plants from Pervillé in
Herb. Kew, and with Professor Reichenbach's description ; but
I cannot find any second lobe to the petals, as mentioned in the
‘Genera Plantarum.’

Disa INCARNATA, Lindl. “ Light red, with yellow spots.
Damp banks, Ambatovory." Dr. Fox.

Disa Bucwenaviana, Avranzlin=Satyrium caleeatum, mihi,
was obtained at Ankeramadinka by Dr. Fox.

ARNOTTIA INERMIS, S. Moore. Andrangoloaka, Dr. Fox. This
seems identical with the Mauritius species. It has not hitherto
been recorded for Madagascar.

Professor Reichenbach, in ‘ Flora,’ 1885, p. 537, refers Mono-
chilus gymnochiloides, Ridl., to the genus Cheirostylis, under the
name of C. gymnochiloides, Reichb. f.

DISPERIS TRIPETALOIDEA, Lindl., has been obtained in Central
Madagascar by Mr. Baron, no. 4288. It has not been hitherto
recorded for the island.

128 MR. C. B. CLARKE’S BOTANICAL OBSERVATIONS

Botanieal Observations made in a Journey to the Naga Hills
(between Assam and Muneypore), in a Letter addressed to
Sir J. D. Hooker, K.C.S.L, F.R.S., by C. B. aevo Esq.,
F.R.S., PLS:

[Read 1st April, 1886.]
Kohima, Naga Hills,

Oct. 30, 1885.
My dear Sir J. D. HOOKER,

I arrived here ten days ago. The station is 4750 feet alt. ;
and “ Jakpho," distant 10 miles, is 9980 feet alt. Having heard
that the vegetation here was totally unlike that of the Khasi
Hills, I thought that there must be a great harvest of new species
to be got here. But Col. Robertson, commanding now at Debro-
gurh, who knows both Kohima and Darjeeling, warned me that
I should find Kohima to be very much Darjeeling over again.
This I find so to be. Nearly all the plants that grow both in
Sikkim and Khasia appear to grow here (the place is exceedingly
rich botanically); besides these there is a large percentage of
Sikkim plants not known in Khasia, and besides these there is
a small percentage of Khasi plants not known in Sikkim, and a
small percentage of new species. Also, the country here above
5000 feet being nearly all jungle (the open grass only small
patches), the Sikkim plants altogether predominate in number of
individuals, and make Kohima resemble Darjeeling, not Shillong.
In marching upon Kohima, for miles together the road was so
like that from Darjeeling to Mungpo that I could not have been
sure that I was not in Sikkim ; and the vegetation from Kohima
to the top of Jakpho closely resembles that from the little Run-
gait to the top of Tonglo. I understand that about 30 miles
south of Kohima I shall enter on open grass and the Khasi flora;
and that the change from the jungle to grass is sudden. It is
also to be recollected that the highest point in the Khasi Hills
is under 6500 feet alt., and that therefore we could hardly expect
Khasi plants between 7000 and 9980 feet alt. here. Still it is
remarkable that we are on one range of hills here with Khasia
not 100 miles off, while Darjeeling is not only 800 miles off, with
the whole Brahmapootra valley between, but that Shillong comes
almost in a line between this and Darjeeling,—yet the whole
charaeter of Kohima is totally unlike Khasia and very much
like Darjeeling.

The country here may be divided much as you divided Sikkim,

IN A JOURNEY TO THE NAGA HILLS. 129

viz. into :—1, the'region of cultivation, alt. 2000-5000 feet ; 2, the
low level jungle below 2000 feet : 3, the upper level jungle, 5000-
8500 feet ; and 4, the small peak of Jakpho, 8500—9900 feet, where
a subalpine flora with Rhododendrons just comes in. The rocks
here are much as in Sikkim but more disintegrated; the upper
levels (above 5000 feet) may be about as steep as the slopes in
Sikkim at 5000—10,000 feet; butthe cultivated region, 2000—5000
feet, are much more open, less steep, valleys, with evidently
richer soil than Sikkim. Major Beresford, who accompanied me
to Jakpho, again ard again exclaimed, * I never saw such culti-
vation in all my life." The land is all terraced and irrigated,
and covered with a heavy crop of a strong growing rice; it looks
like the margin of the plain of North Italy near the lakes.
There is also the common hill-rice as in Sikkim. The Naga
Hills are as Sikkim, but with a vastly ameliorated climate, warmer,
much less rain. The Nagas now are much what the Khasis
were when you were among them; but the population here is
very large instead of as in Khasia and Sikkim. The Nagas eat
dog extensively (dogs are largely imported for the table from
Muneypoor), in this point appoaching the Chinese.

It is perhaps rash for me to make so strong a statement about
the vegetation of Kohima (and the levels above it) after being
here but ten days ; I will therefore offer a little detailed support
of it before going further. I take the Commelinacee, the Indian
species of which I know pretty well specifically. I have seen
here every Sikkim species (except the Pollia with six perfect
stamens), while not one of the species peculiar to Khasia have I
yet collected here. I have only one new species, collected 1200
feet only above sea, an Aneilema of no interest except to a mono-
grapher.

Take Rubus: the Sikkim species abound, the two Darjeeling
silver-leaved Rubi (one with 3-partite, one with 5-partite leaves)
are here (not in Khasia at all) ; and besides these there is another
simple-leaved silver-leaved species of the same section of Rubus,
very handsome (I believe new).

Take Senecio: we have here S. Wallichii and its closely allied
species (I forget specific name), unknown in Khasia; and we
have here I believe nearly every Sikkim Composite that grows
below 9000 feet. We have here, however, also one Khasi species;
viz. the plant 1 called Senecio Rabani, which you have reduced in
the * Flora of British India’ under another well known-species (to
which it is very closely allied). I have also got two new Senecios,

130 MR. C. B. CLARKE’S BOTANICAL OBSERVATIONS

very fine, but one may be Senecio Simonsii, which I have not seen
for twelve years (I have no books nor even a list of genera), so
I cannot say.

Takethe Ferns: I havenot one new species here yet. The middle-
level species are mostly those that grow both in Sikkim and Khasia.
The upper-level species are all Sikkim species aud Sikkim forms
(there is one exception noted below, viz. Lomaria glauca). Thus we
have Lastrea patentissima, Wall., the * red-legged" Tonglo Athy-
rium ; Lastrea apiciflora, var. non-apiciflora, abundant here at 9000
feet, exactly as at Tonglo at the same level.. Indeed, so like are
the Ferns to the Sikkim forms, that I doubt whether Levinge
himself could say that a Kohima bundle did not come from Tonglo.
I have seen no tree-fern here except Asophila glabra in the low
levels. ;

But besides this, I see here numerous plants which I regard as
representative Sikkim genera, and which I do not know from
Khasia: such are Patrinia, Notochete, Dobinca, Kalanchoe. The
“ tasselled" straw-yellow Cymbidium is also here.

Nearly all the Natural Orders tell the same tale as the Com-
melinacee, Rubus, and Senecio; while, on the other hand, many
common Khasi plants are conspicuously absent. I have not seen
one plant here of Impatiens chinensis, so common in Khasia, nor
of any of the other regular Khasi balsams. Nor have I seen here
one plant of any species of Lespedeza, so prominent in Khasia.
Nor have I seen here any one of the marked Khasi species of
the Cheira Sacred Wood, or of the Shillong Hill.

The pines, and in a degree the oaks, are an exception. The
only two pines I have seen here are Khasi species, viz. 1, Pinus
Khasiana, of which I have only seen some poor trees at 5500 feet
scattered in mixed tree-jungle at one spot; and 2, Cephalotaxus
Mannii, Hook. f. MS., which I see scattered at 5000-8000 feet,
and there are battered trees of it about 40 feet high on Jakpho
at 9000 feet alt. :

As to the oaks, they are in such grand force here, both as to
species and individuals, that I have not nearly learnt them yet.
I have collected nearly the whole Sikkim series here (not Quercus
Andersoni); but there are also here several of the peculiarly Khasi
species, such as Quercus Griffithii, A. DC., and the allied sessile-
leaved Shillong species. There are here two oaks which I have
seen before in herbaria but never collected, and I cannot recol-
leet where they come from ; and I have one oak which I do not

IN A JOURNEY TO THE NAGA HILLS. tət

recollect ever seeing before. The great mass of the forests at
5000-8000 feet is oak all round Kohima; the most abundant
tree at the 5000-6000 feet level is Alnus, I think the same species
as in Sikkim. It grows here into a tree with an enormous trunk,
some as big as the largest English oak I ever saw; but big trees
of the Alnus are scarce, as the Nagas pollard it at six feet from
the ground, and cut the innumerable sprouts for firewood.

I have also collected in the 5000-6000 feet level two Diospyros,
largish trees, one with largish smooth fruit, the other with large
hairy fruit, both of them trees that have a tropical, non-Sikkim,
aspect to me.

The flora here is thus very interesting, and exceedingly rich,
nevertheless my collections contain few new species. I had a
vague idea that I could, in a mere passing visit, collect a nearly
complete set of the Kohima species in flower in October; but,
if forno other reason, I cannot do this for want of paper. Trans-
port here is exceedingly difficult; the approach to Kohima is
admitted to be the worst *line"' in India; the sixty-four miles
through the Nambre swamp-forests often take the Government
convoys ten to twenty days, and the road is decorated with
broken carts and the bones of bullocks, and sanded sometimes
for half a mile with Government grain. Kohima is not a place
where you can get paper of any kind. I have therefore in col-
lecting often gathered only a serap of the species where well
known—this will be enough for identification.

In the richer soil and warm climate of Kohima many plants
grow unusually large. One plant here which the English denizens
have not overlooked is the Kohima thistle; it is abundant and
attains fifteen to twenty feet (I am told that in some places it is
twenty-five feet high) ; this is I believe only a large form of the
Sikkim Cnicus (like our European Cn. eriophorus) with yellow
flowers. Similarly I gathered here an enormous Butea; even of
a small leaf of it the ordinary-sized paper cannot hold one leaflet;
at first I thought it quite new, but I now believe it is only alarge
form of Butea minor, Wall. There is another “ Carduoid " here
(I forget the genus, but it is not Cnicus), which attains ten to
fifteen feet, and is quite unlike my recollections of the plant
(this may be sp. nova). The rice is, as I have stated, the Coix,
and all the crops grow exceedingly strong. Naga-land is in fact
à very rich country, and it is not a little place like Sikkim or
Khasia; there must be 200 miles of it, all nearly equally good.

LINN. JOURN.—BOTANY, VOL. XXII. L

132 MR. C. B. CLARKE’S BOTANICAL OBSERVATIONS

I have been up Jakpho; I went to the new summit, which is
ascertained to be fifty feet higher than the old, and to the old
summit also. The ascent is from * Kegwima," alt. 5800 feet,
a village ten miles from Kohima. The ascent would thus appear
4000 feet, but in starting from Kegwima you go one and a half
miles down hill to begin with, so that the whole pull up is about
5000 feet. It is a very easy ascent, no mountaineering at all, and
you can get about anywhere ; you are not confined to the path, nor
is there a very grand path. I went up aftera thirty hours’ rain,
and the lower half was very slippery (all this shaly Kohima
mud) ; the top 1500 feet is excellent going ; small Rhododendron
trees with ferns below. I describe the ascent at some length,
because you may have heard that it is a difficult or dangerous
one. As a matter of fact it is made by many persons who
do not profess to be mountaineers ; and it is reckoned that from
Kegwima it is four hours’ march up. At the same time when Isay
the ascent is easy, I mean as to mountaineering ; you will know
what an ascent of 5000 feet up Tonglo with no well-used path
implies, and Jakpho is still finer exercise (the lower part) from
the slippery ground. Up to 9000 feet the vegetation on Jakpho
is much as on Tonglo; there is one curious difference. You
may recollect that all round Darjeeling from 6000-8000 feet the
ground under the forest trees is covered with a grove of Lomaria
($ Plagiogyria) pycnophylla. On Jakpho, from 7000-9000 feet the
ground under the forest is covered by the closely allied Lomaria
(8 Plagiogyria) glauca ; very densely covered, so as at this season
to exclude nearly everything else. Lomaria glauca we reckon a
rare fern in Khasia; here it is very plentiful and large. The
Rhododendrons come in strong at 8500 feet; I only picked scraps
of Rh. arboreum, Falconeri, and about four other species. They
were not in flower of course, nor was the fruit near ripe ; and I
fancy Dr. Watt (who was on Jakpho in May) got all that is to
be got; so I paid little attention to the Rhododendrons. The
ridge on the very top, connecting the old and new summits, ÍS
clothed with dwarf bamboo (I suppose an Arundinaria) about
six feet high, dense, and dripping wet. Just at the old summit
where there is some open rock, and on one side a drop of 2000
feet at an angle of 80^, I collected several species (besides the
ferns above mentioned by name), viz. :—

Arundinella (a large species).

Juncus (membranaceus, or an adjacent species).

IN A JOURNEY TO THE NAGA HILLS. 133

Juncus, very slender, with small flowers.

Luzula, large; with long flowering-stems and basal tufts of
broad green leaves.

Carex: a large broad-leaved species, perhaps an extreme state
of C. speciosa, Kunth; collected also in Sikkim by Gamble.

Vacciniaceæ * three Tonglo species.

Microglossa: large shrubby, with white ray; but I think only
a form of the common species.

Rubus: half a dozen species, 3 “ silver-leaved " ; moreover R.
moluccanus on the very summit. I mean by R. moluccanus a
species with undivided sub-5-lobed leaves which is abundant at
2000-4000 feet.

Senecio Wallichii, and its allied species ; and the two common
scandent Sikkim species.

Pimpinella (Hookeri, I think), i.e. the very fine cut species
you got on Tonglo.

Ophelia: a 5-sepalous species with reflexed purplish corolla;
a Sikkim species at 9000 feet.

Crassulad: a common Tonglo Sedum.

Pedicularis: a Tonglo species.

Ainsliea: two species.

Actinidia.

Impatiens : the Tonglo small yellow.

Very near the top I saw the common Sikkim white Michelia.
Also, there grew in great beauty a small Campanulad (Codonopsis).
This species grows in Sikkim as a poor little straggling weed
nearly prostrate, with few flowers; in Jakpho it forms small dense
upright bushes about eighteen inches high—a mass of flower—I
think 500 flowers on a single stem.

By working down 100-200 feet from the summit on the north-
east face of Jakpho I came upon the two best plants I got, viz.
too large Senecios of the section Ligularia. One of these is (I
am pretty sure, without a book) the old Ligularia racemosa, DC.,
a plant which I have never seen in Sikkim under 11,000 feet ; and
Jakpho is much warmer than Sikkim at corresponding levels.
The other is a Ligularia, five to six feet high, with very large
leaves, and few florets to the head; it is therefore not a typical
Ligularia. Iam pretty sure that I have seen this plant before
but I fancy not in Sikkim, but in or near Kashmir (this is only a
fancy).

L 2

134 MR. C. B. CLARKE’S BOTANICAL OBSERVATIONS

The levels 5000-7000 feet on Jakpho (and elsewhere here) are
mostly forests of shrubby Strobilanthes six to twelve feet high,
just as in Sikkim. The three or four species now in flower are
Sikkim species.

There is a tree on Tonglo, at about 8000-9000 feet alt., which
has been much noticed by the Europeans here, as they believe it
to be the English holly. Ithought they were wrong, though
the leaf is much like Zex Aquifolium. But I was not able to get
either flower or fruit. The trees I saw were thirty to forty feet -
high, erect, with single cylindric stems about eight inches diam.
The bark is clean, much like holly, and I now think the European.
residents may be right.

There are several laurels on Jakpho, two (at least) of which
are Liteas; but I only saw the fruits of one laurel, which I did
not collect as I could not get the leaves belonging.

The above account of my Jakpho ascent will explain to you that
though the place is very interesting, it has hardly supplied mewith
one species that I know ean be new. It seems to me that I may
get here one or two new species in each large Order, but that I
am not likely to get a new genus.

Up to 5000 feet the Convolvulacee are very prominent. The
snow-creeper (Porana) is abundant about Kohima, and there is
another small one here. At 1000 feet alt. is a splendid species
with flowers fully as large as the Sikkim mauve Porana, but white,
downy without. Of these large flowers there are twenty to thirty
in each spike, and the spikes very numerous. lt is a beautiful
plant; but I have an idea that Kurz got it on the Karen Hills.
From 2000 to 4000 feet alt. white-flowered Ipomzas abound, none
I think new.

The gingers have been numerous here, but are now off flower.
There remains one (a very fine one) like Hedychium coronarium,
but with rich orange flowers (I think anew species). Also I got
down in the Nambre forest, at 500 feet alt., a large Roscoea with
purple flowers (many) in subquadrangular spikes. This also is
a new species so far as I can recollect.

The orchids here are numerous (the European residents say.
very numerous), but not so many perhaps as in Khasia. I have
colleeted several Sikkim species. Of peculiarly Khasi species
Vanda cerulea is plentiful; also the Celogyne with brown
fringed lip. There are two Zuhabenarias still in flower.

In Rubiacez, as in so many other Orders, if you take the Flora of

IN A JOURNEY TO THE NAGA HILLS. 185

British India, and make a list of every species common both to
Sikkim and Khasia, that will give you very nearly what I have
noted here. I have two (or three) species of Spiradiclis, Carle-
mannia, and two of Coffea. Also I have three species of one
marked genus near Myrioneuron (this may be new, but I think
not). There is one Luculia plentiful, I imagine the Khasi (not
the Sikkim) species.

In the Nambre swamp forest I could not get at the trees.
Mucuna, Pueraria, Argyreia abound. There is one gigantic
climber with large racemes of dark red flowers (a Leguminosad)
new to me. Leguminose are not strongly represented near Ko-
hima. I have one that I suppose new, viz. of Dumasia, which
“ replaces " the species common in Sikkim. There is also in the
Nambre forest, and up to 1500 feet alt., the curious small-flowered
Thladiantha, which you got in Khasia; but I could not get one
female plant either in flower or fruit; so the problem remains
where you left it.

It is rather too late in the season to collect Cyrtandracez, but
there are here all the species common to Sikkim and Khasia.
Of ZEschynanthus I have seen three species. In Lysionotus there
is.the common L. ternifolia, and there is also a bushy Lysionotus,
five to six feet, hairy and with very slender peduncles: this I
think is the imperfectly known Lysionotus Griffithii, of which
Griffith collected some fruiting fragments in Upper Assam. In
Chirita there is only in flower Ch. acuminata, R. Br., and Ch.
pumila, D. Don. Of Didymocarpus I have only one in fruit ; it
is not any Khasi species. There are the usual Rhyncoglossum and
Rhyncotechum. Then I have one or two scraps in fruit of a genus
(somewhere near Stauranthera) which Griffith got in Upper
Assam.

In Gentianaces, I have collected three Ophelias, two Exacums,
one Crawfurdia.

I have collected no Primulad; I doubt there being many in
these jungles at any season. Ardisia, with its small allied genera,
is common here with several species; also Mesa, Myrsine, Embelia,
and one very large Choripetalum.

Begonias very plentiful, many Sikkim species; I have seen
none that I suppose new.

Araliads numerous.

Acanthads not numerous (except Strobilanthes); khereis Thun-
bergia coccinea, Wall., common. Also Codonocanthus.

136 BOTANICAL OBSERVATIONS IN THE NAGA HILLS.

Euphorbiacez seem to me fewer than in Sikkim; except Anti-
desma Y have collected hardly anything.

I have collected here Holmskoldia, Cardiopteris, Tacca, and
several other of such characteristie genera.

I have seen no Garciniad ; nor any Corydalis or Dicentra; nor
Crucifer (except one or two weeds); nor Capparid. I recollect
no other Order well represented in Sikkim that is not fairly well
represented here.

The list of plants at the top of Jakpho above noted is only of
a few selected plants; I also saw Polygonatum, two species,
Koenigia, Anaphalis, two species, Epilobium, Pyrus, two species,
Daphne, Viola, &c., &e.; all I believe identical with Tonglo plants.

I do not intend to ascend Jakpho again as I do not think I
shall get much more there. I shall endeavour to visit a patch
of grass at 4000 feet, a few miles hence, also some patch of grass
at 7000-8000 feet. Then I march to Muneypoor town (about
ninety miles) and thence back to Cachar, about 140 miles. I
go to Muneypoor so as to meet the Chief Commissioner, and the
Governor-General’s Agent there,

I enclose a few seeds of an Elettaria (?), or at any rate of one
of those gingers that flower from the root. I do not imagine
they ean be of much value to you at Kew; but they are the only
Seeds that my hand has fallen on, and I send them to show
that I have not forgotten your indent for seeds of any rare or
beautiful plants.

Capt. Plowden, Commandant of the Frontier Police, tells me
that only twenty miles hence, at the Kohima level (5000 feet),
there is a piece of grass with Khasi pines, two days’ march in
length; and apparently the whole Khasi flora there comes in.

No plant I have collected here surprises me so much as a Si
lene which grows in the wet jungle near Kohima, at about 5500
feet. lt is about three feet high, very straggling, hardly any
"leaves; in short I cannot recollect anything near it this side of
Kashmir, but my memory may be defective.

Believe me to be,
Yours very truly,
C. B. CLARKE.

ON THE BOTANY OF WESTERN SOUTH AMERICA. 137

Notes on the Botany of Western South America.
By Joun Batt, F.R.S., F.L.S.

[Read 18th February, 1886.]

Iw the course of a hurried journey round South America, I
visited, in 1882, some points, chiefly on the western coasts of that
continent, of which the vegetation is imperfectly known; and I
have since endeavoured, by comparison of the available materials,
especially those preserved in the herbarium at Kew, to clear up
some doubtful questions relating to the distribution of species.
I have thought it desirable to bring together such of the notes as
I have made on particular places visited by me as appeared to
contribute some slight additions to our existing knowledge, or to
correct errors that have crept into works of authority.

Before entering into details, however, I would premise a few
remarks on the climatal relations of the western sea-board of
South America, which have had such a remarkable influence on the
development of organie, and especially of vegetable life. For an
accurate knowledge of the chief factors of climate—temperature
and rainfall— we are unfortunately at a great disadvantage; for
there are few equally extensive regions of the earth where meteo-
rologieal observations are so scanty and so imperfect as on the
west coast of tropieal South America. So far as I know, none
whatever are available from any point on the coast between
Panama and Lima, a distance of fully 1600 miles; and our know-
ledge of the climate is derived merely from the testimony of
travellers and the few Europeans who have resided in the country.

The Isthmus of Panama (between 9° and 10° N. lat.) has an
annual mean temperature of about 81° F., rarely varying more
than two or three degrees above or below that average. The
rainy season extends for seven months, from May to November,
with an average monthly fall of about 8} inches, while during the
relatively dry season, from December to April, the total rainfall
for five months reaches the same amount, giving the rainfall in
the year about 68 inches. The climate of the coast of Darien is
said to retain the same general characteristics ; but as we travel
southward towards the equator the length of the rainy season
gradually increases, until, in the Gulf of Choco and the adjoining
coast, there is no season of even relative dryness, and heavy rains
occur frequently throughout the year. The reports to this effect

188 MR. J. BALL ON THE BOTANY

were confirmed by my own brief experience on finding at Buena-
ventura, on open ground elear of the forest, an abundant growth
of Cyperacee with a minute Utricularia, while, under a nearly
vertical sun and a clear sky, the soil was everywhere moist.

The maximum of rainfall appears to lie some three or four
degrees north of the equator, for the aspect of the coast at Tu-
maco, on the frontier of Ecuador, indicates a distinct change of
climate: open spaces of grass-covered plain, affording good pas-
ture to horned cattle, could not exist amid the perpetual moisture
of the coast only 100 miles farther to the north. So far as I can
learn, the low eountry of Ecuador between the Andes and the
sea has what may be called the normal equatorial climate *—a
rainy season extending through more than half the year, with an
interval of comparative cessation of rainfall There is reason,
however, to conjecture that some projecting points on the
coast, especially the Capes San Lorenzo and Santa Elena, are
affected by the southerly currents of the air and ocean that main-
tain the exceptional aridity of the coasts of Northern Chili and
Peru, and I shall have occasion to adduce some evidence to that
effect.

The coasts of the Gulf of Guayaquil possess, as might be ex-
pected, an equatorial climate—an average temperature of about
82° Fahr., subject to very slight variation, with considerable
though not excessive moisture. On the southern side of the
gulf, at Tumbez, the frontier port of Peru, the same climate 1s
said to prevail; but on issuing from the gulf and rounding the
headland of Cape Blanco, little over 4° S. lat., an abrupt change
of climate and vegetation is encountered. Under the influence
of the cold Humboldt current which, issuing from the southern
ocean, sets along the west coast of S. America, and of the breeze
which constantly flows in the same direction, the mean tempera-
ture is reduced by more than ten degrees of Fahrenheit, and, the
land being almost always warmer than the sea-breeze, rain falls
very rarely and at long intervals,

The same conditions, with scarce any variation, prevail along
the coast from Cape Blanco to Copiapó, or somewhat south of
that place, over more than 23 degrees of latitude and a coast-
line of about 1700 miles. The maximum of aridity appears to be
reached a little north of the Tropic of Capricorn; but the mean

* In some places near the equator there appear to be two rainy seasons in the

year, with intervening dry seasons; but there are few indications of this phe-
nomenon in equatorial America.

OF WESTERN SOUTH AMERICA. 139

temperature on the coast throughout a great part of the year is
nearly the same—from 68? to 70? Fahr. As we approach the
southern tropic, differences of season become marked; and in
Southern Peru and Northern Chili we encounter differences of
from 10 to 12 degrees of Fahrenheit between the means of
summer and winter. Throughout tbe whole region the mean
temperature increases inland, even at considerable heights on the
slopes of the Andes, as the influence of the southerly currents
becomes less effective *.

Somewhat south of Copiapó the desert region of Western South
America becomes gradually modified; the climate passes slowly
from the state of excessive aridity to a medium condition some-
what resembling that of the drier parts of Southern Europe,
where occasional, and sometimes heavy, rain in winter is succeeded
by along dry summer. Such is the character of the climate of
Central Chili between the parallels of 31? and 36? S. latitude ; and
itis marked by the appearance of a peculiar flora, including a
large number of endemie genera as well as species, for the most
part nearly allied to forms inhabiting the middle and higher zones
ofthe Andes. Even at Coquimbo, in latitude 30^ S., where the
average annual rainfall is said to be only about 13 inch, a good
many of the characteristic types of the flora of Central Chili are
able to maintain themselves. It should be remarked that through-
out the so-called rainless zone of Peru and Chili fogs frequently
rest in winter on the hills near the coast, and that an appreciable
amount of precipitation in the form of dew must be deposited.
To this we must attribute the appearance of a comparatively
abundant vegetation at that season, which has been observed at
several places on the coast, as, for instance, on the hills about
Lima. A recent journey of Professor Friedrich Philippi has
made known the curious fact that on some of the higher ranges
near the coast of Northern Chili, where cloud or fog is seen to
rest during a great part of the year, abundant vegetation exists,
including several plants of the moist region of Southern Chili.
This ceases abruptly at the lower limit to which the cloud habitu-
ally extends.

In Southern Chili a very rapid change of climate occurs, owing
to the influence of relatively warm westerly winds, which appa-

* In his excellent * Handbuch der Klimatologie, p. 127, Dr. Hann has dis-
cussed the general causes which, in tropical and subtropical regions, depress the
coast-temperature on the western side of continents and raise it on the eastern
coasts,

140 MR. J. BALL ON THE BOTANY

rently reach a maximum intensity about the parallel of the island
of Chiloe. The change seemingly commences about lat. 36? S.,
and about a degree further south, at Concepcion and at Lota, is
so marked as to affect materially the aspect of the flora by the
disappearance of most of the characteristic types of Central Chili.
We possess, unfortunately, no trustworthy observations of rainfall
on the coast between Valparaiso, in lat. 33? S., where the average
fall is little over 131 inches, and Valdivia, in lat. 40? S., where it
surpasses the average of 117 inches; while at Ancud in Chiloe,
in lat. 42? S., it is said to reach the yearly mean of 137 inches.

Although the annual mean temperature is low, about 51? F.,
the mildness of the winters allows several plants of subtropical
affinities (Laurinee, Myrtacee, Bambusee, &c.) to flourish in
Chiloe, where at Ancud the mean temperature of July, the coldest
month, is 46° F.

From our limited knowledge of the flora of the Chonos Archi-
pelago, it would appear that most of the representatives of sub-
tropical types disappear about the parallel of 44° S., and that may
be assumed as the northern limit of what naturalists have agreed
to style the Antarctic Flora, and which extends southward with
a surprising degree of uniformity to Cape Horn, over 12 degrees
of latitude. It is true that several of the characteristic herba-
ceous types of the islands of high southern latitudes are represented
in this region, and that there are plausible grounds for referring
to south polar lands as the original home of a great portion of
the flora ; but, on the other hand, it may be urged that the climate
of the entire region, however rude, is essentially temperate, and
that the dominant features of the vegetation—the beech-forests,
the abundance of delicate ferns (Gleichenia, Hymenophyllum, &c.),
and such peculiar types as Drimys, Desfontainea, Mitraria, and
several others—suffice to entitle the flora of this region to rank
as a separate province, for which the name Magellanie may, peT-
haps, be appropriate. We possess but very incomplete data for
discussing the climate save what are afforded by the vegetation,
which is now fairly well known ; but the fact that at Sandy Point,
on the eastern side of the Straits of Magellan, south of the paral-
lel of 53° S., the temperature of the coldest month (very nearly
35° F.) is higher than that of Holland or of the plains of Northern
Italy, fully proves the exceptional character of the climate.

I proceed to give the following brief notes as to some places
on the coast which I was able to visit.

OF WESTERN SOUTH AMERICA. 141

Flora of Buenaventura in Colombia.

Buenaventura, a small port on the Pacific coast, about four
degrees north of the equator, is, I believe, the chief place in the
province of Choco in New Grenada. In the very short time
allowed during the halt of the steamer I collected the following
plants on the skirts of the forest close to the little town :—

SAUVAOGESIA ERECTA, Z. Indigenous in tropical America and
in east and west tropical Africa.

Ryanra crocozEwsis, PZ. et Tri. This species is known only
from Buenaventura.

VIsMIA RUFESCENS, Pers. Syn. ii. 86; H. B. K. Gen. et Sp.
Nov. v. 182. Though long known, this appears to be a very rare
species. I have seen only a single imperfect specimen in Kew
Herbarium. Planchon and Triana find the Buenaventura plant
to agree with the original specimen collected by Humboldt and
Bonpland in Guiana (or Venezuela ?).

CrvsrA ? Collected in fruit. Approaches C. pseudo-
mangle, Pl. et Tri., but is apparently different, perhaps an unde-
scribed species.

DESMODIUM ADSCENDENS, DC. A species widely spread in
South America and in the West-Indian Islands.

ConosTEGIA POLYANDRA, Benth. Bot. Sulph. A native of the
coast-region of Southern Colombia and Northern Ecuador.
Doubtless through error supposed to have been found in the
Andes of Eeuador.

CLIDEMIA DEPENDENS, Don. This is a very variable species, of
which two forms were seen, one in open ground on the skirts of
the forest, the other in more shaded places. It extends through
equatorial South America.

CLIDEMIA HIRTA, Don. This is widely spread in tropical
America.

NEPSERA AQUATICA, Naud. A native of tropical America,
tending to the West-Indian Islands.

Miconia rEDUCENS, Triana. Known only from Buenaventura
and the adjoining district.

142 MR. J. BALL ON THE BOTANY

Miconta syMPLOCOIDEA, Triana. The only published locality
for this species is the lower region of the Andes of Bogota.

Mriconra DECUSSATA, Don. This extends through equatorial
America from Panama to Bahia in Brazil.

SPERMACOCE LINEARIS, H. B. K. A native of Mexico and
Venezuela. If Borreria podocephala, DC., should, as I suspect,
be referred to this species, it has a wide range in tropical
America.

SpPERMACOCE PARVIFLORA, Weyer, This extends from Mexico
and Florida through tropical America to the northern provinces
of Argentaria.

MANETTIA ASPERULA, sp. nov. As this differs considerably
from all the described species, I have found it necessary to name
it, and I here subjoin a brief description :—

Frutex alte scandens, ramis teretibus volubilibus; foliis inte-
gerrimis ovato-lanceolatis, apice acuminatis, breviter petiolatis, in
pagina superiore asperulis; stipulis liberis ovalibus, pro genere
magnis (2-3 lin. longis) ; pedunculis florigeris axillaribus, su-
premis solitariis paucifloris, inferioribus geminis oppositis, 8-10-
floris ; racemo (etiam fructifero) quam folia breviore; calycis
segmentis 4, lineari-oblongis, acutiusculis, dentibus 4 alternis
sæpe (nee semper) interjectis; corolla pro genere parva ochroleuca,
extus strigosa, loborum margine ex pilis rigidiusculis patentibus
ciliato.

I found this climbing over the smaller trees on the margin of
the forest.

ANTHOPTERUS WARDII, Ball in Hook. Ic. Pl. ser. 3, v. No.
1465.

I reproduce the description above referred to :—“ Suffrutex
epiphyticus, glaberrimus, foliis alternis, ovatis, obtusiusculis,
basi rotundatis vel subcordatis, margine integerrimo subrevolutis,
subtus elevatim nervosis; floribus 5-10 in racemo brevi subco-
rymbosis; pedicellis fere pollicaribus, basi bracteatis, cæterum
nudis; calyce cum pedicello pulchre coccineo, sub anthesi obco-
nico, late alato, alis in pedicellum decurrentibus, segmentis liberis
ovato-triangularibus, venosis, tubo subzquilongis ; corolla urceo-
lata, quinquealata, saturate rubra, fauce constricta breviter den-
tata, filamentis connatis."

This beautiful species was procured for me by my friend Mr.
Richard Ward, to whom I have dedicated it. Being, at the same

OF WESTERN SOUTH AMERICA. 143

time with myself, in a different part of the forest, he remarked
the brilliant red flowers on the trunk of a tall tree, which was
readily climbed by the negro who accompanied him. It is very
different from the described species, Anthopterus racemosus, Hook.
le. Pl. iii. No. 243, and A. mucronatus, Beuth. Pl. Hartw., but
approaches near to an undescribed species of which I have seen
specimens in Kew Herbarium. One of these was collected
somewhere in Peru by Maclean ; the other is No. 2716 of Triana's
plants of New Grenada. "These specimens are in bad condition,
but apparently belong to the same species. They differ from my
plant in having shorter pedicels; the calyx-wings narrower,
and seemingly not decurrent; the calyx does not appear to
have the same brilliant red colour, which in my specimens con-
tinued to come off on the paper long after they were dry.

UTRICULARIA

? In damp ground on the margin of the
forest I found growing amongst Cyperacee a minute species of
this genus, with very small flowers supported on stems about two
inches in height. I have not been able to identify it; but as
several very small species have been found in equatorial America,
some of which I have not seen, I do not venture to name my
specimens.

SOLANUM MAMMOSUM, Z., var.? On the bare branches of
small bushes growing near the forest I found the fruits of a So-
lanum, the leaves being quite withered. The specimens, now in
the Museum at Kew, appear to belong to a large-fruited variety
of this species. The fruit is supposed by the natives to be so
poisonous that they think it dangerous to handle it.

EUPHORBIA SERPENS, HM. B. K. Apparently common here.
The typical form of this species extends rather widely through
the warmer parts of America; but a nearly allied form, ranked
by Boissier as a variety, is spread through tropical Asia.

ALCHORNEA ? The species of this genus are not easily
identified in the dry state, and I an unable to name this.

CRINUM ERUBESCENS, Ait. Growing in the deep shade of the
forest. This, and several nearly allied forms, which should,

perhaps, be regarded as varieties, is widely spread throughout
equatorial America.

Tonina FLUVIATILIS, dub]. This has a wide range in tropical
America.

144. MR. J. BALL ON THE BOTANY

CYPERUS LUZULÆ, Rottb. This appears to be confined to the
equatorial zone of South America.

Ruxwcnosronga TENUIS, Willd.=Dichronema gracilis, Kunth.
A species with a wide range in tropical America, extending south-
ward to Montevideo and northward to Mexico and Cuba. It is
not to be confounded with the very similar species, 2h. spermodon
of Grisebach 2 Dichronema setacea, Kunth. The latter has nearly
as wide a range in tropical America, and, according to Bóckeler,
extends also to Australia.

FIMBRISTYLIS POLYMORPHA, Bockl., var.— F. laxa, Vahl. A
cosmopolitan species, with a wide range in the tropics and the
warmer temperate zones.

FUIRENA UMBELLATA, Rottb. Like the last, this species ex-
tends round the world, but appears to be confined to the tropies.

PASPALUM CONJUGATUM, Berg, var.? The spikes are shorter
and the florets decidedly larger than in numerous specimens from
other parts of tropical America. The species is widely spread in
tropieal America, and also, as i$ appears, in tropieal Africa.

PANICUM POLYGONATUM, Schrad. A tropical American species.
MENIscIUM RETICULATUM, Sw. Equatorial America.

LYCOPODIUM cernuum, L. Widely spread through the
warmer regions of the world.

I did not attempt to collect arborescent ferns, of which several
species were seen; and it is needless to say that, under the cir-
cumstances, it was impossible to obtain specimens of the larger
trees forming the forest.

The very slight specimen which the above list affords of the
flora of the Pacific coast-region of equatorial America suggests 4
doubt whether, as regards the warm and moist regions of South
America, the differences between the floras of different portions
have not been too much insisted on by writers on botanical geo-
graphy, and especially by Engler in his valuable and important
work, ‘ Versuch einer Entwicklungsgeschichte der Florenwelt.’
He includes Venezuela and Colombia with tropical and subtropi-
cal Central America and Mexico, under the name * Subandine
Province," as one of the great divisions of the American flora,
making a separate botanical province of North Brazil and Guiana.

OF WESTERN SOUTH AMERICA. 145

It will be noticed that, with a single exception, all the plants in
the above list belong to genera that are widely spread throughout
equatorial America, and that, including two new species and one
doubtful, there are but seven species probably confined to the
Pacific coast. It is extremely difficult to collect materials for
comparison, inasmuch as the excellent, but unfinished, * Prodro-
mus Flore Novo-Granatensis’ of MM. Triana and Planchon
includes only the Natural Orders of Thalamiflore, probably little
more than one tenth of the entire flora. I have analyzed the
contents of the volume, omitting the mountain-species and taking
account only of the genera which are represented in the warm
zone of the Pacific coast, which may be reckoned to extend from
the shore to a height of about 1500 metres, and I find that the
99 genera enumerated may be divided into the following cate-
gories :—

G genera cosmopolitan, 2. e. widely spread throughout the
world.
38 , amphigean, i.e. in the tropics of both hemi-
spheres.
46 , tropical American, extending widely through
the continent.
„ endemic, not extending far from the Pacific
eoast.

9

Total 99

Several of the genera ranked as amphigean are represented in
tropical America and tropical Africa, but not elsewhere in the
world. <A few of the tropical-American genera extend to the
temperate zones in North or South America. The large majority
extend to Brazil; but a few (four or five) do not appear to range
beyond Guiana. It must be remembered, however, that this
implies a range of at least 1600 miles from east to west, while the
great majority extend over an area of at least 3000 miles in
diameter.

The endemic genera are those demanding most attention.

Hampea and Matisia are genera belonging to the tribe Bom-
bacee, which is, in many respects, intermediate between Malvacee
and Sterculiacee. Along with two genera which are common to
the tropics of the Old and New World, this tribe has developed
a number of forms so distinct that they have been ranked as

146 MR. J. BALL ON THE BOTANY

separate genera, each consisting of one or, at most, very few
species. Hampea has one species in Mexico and one in New
Grenada, while Matisia appears not to extend beyond the lower
eastern, northern, and western slopes of the Andes of New Gre-
nada. To the same tribe belong seven other genera (of which
four are monotypic) whose range is confined to limited areas in
tropical America, and an equal number, of which also four are
monotypic, are found in the tropies of the Old World.

Hasseltia, with only two or three species, is a genus of Tiliacee
which appears to be confined to New Grenada; but Iam not sure
that it extends to the Pacific coast. The allied genus Prockia
has a widerange in tropical America.

Pelliciera, a monotypic genus of Ternstrémiacee, is a curious
tree with the habit of the mangroves, and, like that tribe, in-
habits the muddy shores of the Pacifie from Buenaventura
northward to Central America.

Cespedesia and Godoya—the former with four, the latter with
two species—are genera of Ochnacee, whose range is said to be
confined to Northern Peru and New Grenada, but which doubt-
less extend to the lower region of Ecuador. Along with four
other allied genera inhabiting Guiana and Northern Brazil,
they form the tribe Luwembergie, which is strictly confined to
equatorial America.

Pilosperma, Clusiella, and Balboa are monotypic genera, nearly
allied to Clusia, which are confined to the Pacific slopes of the
Andes of New Grenada. The whole tribe of Clusiee is confined
to tropical, and mainly to equatorial, America. Although most of
the species inhabit the warmer zone they bave a very limited
range, and several of them are mountain species inhabiting the
middle zone of the Andes, between 6000 and 9000 feet above the
sea. To this category, indeed, belongs Clusiella, which has not
been found below the level of 6000 feet.

It is apparent from these facts that, if we may judge of the
tropical flora of New Grenada from the portion hitherto pub-
lished, it is not characterized by the presence of many peculiar
types of vegetation. With the single exception of Pelliciera, the
few endemic genera are nearly allied to others inhabiting other
parts of tropical America, and belong to tribes which are else-
where marked by the tendency to variation in portions of their
structure that are held to constitute generic characters. Several
of the generic types regarded as characteristic of the western side

OF WESTERN SOUTH AMERICA. 147

of tropical America, are clearly of Andean origin, some species
of which have been adapted to the climate of the lower region.
To this category belong several genera of Vacciniacee, of which
an example is found in the new species of Anthopterus above
enumerated.

The conclusion to which I am led from a study, doubtless very
incomplete, of the flora of tropical America, is that the vast
region including the warm and moist parts of South and Central
America should be regarded as a single Botanical Province,
in which the same generie types are represented by species of
which a large proportion are endemie, and confined to compara-
tively small areas. Along with these we find in various parts of
the same region a few forms so distinet as to be ranked as
separate genera, mostly represented by one, or very few, species,
and nearly allied to types of wide distribution.

If along with these elements of the flora we find representatives
of very different types of vegetation in many parts of the same
great region, closer examination shows that, with rare exceptions,
these are more or less modified forms of groups whose origin is
either in the Andes or iu the mountains of Guiana and Brazil.

If this view be correct, it undoubtedly entails the inconvenience
that it is impossible for the phyto-geographer to make the
natural divisions of the flora agree with defined geographical
boundaries. But this is an inconvenience which has become
apparent elsewhere in the world as wellas in South America.

In a broad sense it may be said that the most natural divisions
of the vegetation of the earth are wide areas of low country, over
whieh, with more or less of modifieation, a limited number of
types have extended, with islands of high land, which are the
homes of the special types that form the characteristic features
of the floras of different regions. From these mountain-centres
we ordinarily find some representatives of these special types
that extend through the low country to a greater or lesser
distance from their original home, but as to whose mountain-
origin there is little room for doubt. I am well aware that real
or apparent exceptions to this view may readily be cited. Such,
for instance, are the Vochysiacee of South America, which are now
confined to the warm tropical zone. It is a fair matter for
discussion whether it be more probable that they have originated
in that zone, or whether the existing species are the modified
descendants of ancestors that inhabited the ancient, and now

LINN. JOURN.— BOTANY, VOL. XXII. M

148 MR. J. BALL ON THE BOTANY

worn down, high mountains of Brazil. To some botanists the
Melastomacee may appear to present a more important exception.
Looking at the very numerous modifications which the earlier
types of that family have undergone, and to their wide diffusion
throughout the world, we are, in any case, forced to attribute to
them a very great antiquity. Further, we find so many repre-
sentatives of the chief American types of the Order inhabiting the
subalpine zone of the Andes that, as it appears to me, the
probability of their mountain origin is at least a not unreasonable
hypothesis. An illustration of the view here defended is
afforded by the tribe Thibaudie of Vacciniacee, which, excepting
two Himalayan genera, are exclusively American, and mainly
inhabit the slopes of the Northern Andes. Several representatives
of the group, so distinct as to be ranked as separate genera, have
extended to the tropical zone of equatorial America, and some of
these appear in the islands of the Caribean Sea. Yet few
botanists, I think, would hesitate to regard the Northern Andes
as the home whence these aberrant types have been developed.
It is easily conceivable that in the course of their gradual
migration, while exposed to many vicissitudes of external con-
ditions, new species, and even new generic types would bave
been developed.

I cannot close this slight notice of the flora of Buenaventura
without remarking that it is a spot well deserving the attention
of any naturalist visiting South America. It is easy of access,
is said to have a fairly healthy climate, and would afford a con-
venient ceutre for excursions in a region singularly rich in forms
of life, and which has hitherto been very imperfectly explored.

Flora of Payta in North Peru.

Although there is very little to record, I think it may be
useful to note the species collected during a walk of about twe
hours over the plateau behind the little town of Payta in North
Peru. The vegetation of the hot and dry desert regions of the
arth everywhere exhibits some common features. The existence
of such regions is determined by the extreme scarcity of aqueous
precipitation in the form of rain or dew. The surface being
ordinarily rock, more or less covered with sand, few seeds are
developed even on the rare occasions when sufficient moisture is
present; and such plants as are seeu are scattered at wide
intervals. Annual species, which appear after the rare occasions

OF WESTERN SOUTH AMERICA. 149

when sufficient rain has fallen, are speedily withered, and even
perennials are often unable to survive the long intervals of
drought. On the other hand, the climatal conditions are emi-
nently favourable for the preservation of the vitality of many
seeds which may be carried by the wind over wide spaces. One
result, whieh has been frequently verified by observations in
Northern Africa, is that at any given spot the vegetation is
subject to singular variation in successive seasons. Some species
disappear, while others, not before seen, exhibit themselves, and
after a few seasons again give place to the old residents or
to other new comers. The other result is that over a large area
the flora is mainly composed of relatively few identical species
which spread far and wide. Of the plants inhabiting the vast
zone of desert, extending from Beluchistan to the Atlantic coast
of Marocco, south of the Great Atlas, the large majority are
found at intervals throughout the entire area. It would be of
some interest to ascertain how far these conclusions are verified
in the desert region of western South America, which extends
northward, from about 27? S. latitude in Chili to about 4° S.
lat. in Peru, through a distance of about 1700 English miles.
The conditions are in some respects so different that we should
expect the phenomena to be here less marked than in North
Africa or Australia. In the first place the wind is much less
powerful, rarely exceeding in force a gentle breeze, and it sets
uniformly in the same direction from south to north. Hence
this agency for the dispersal of seeds is less efficient, and while
we should expect to find a good many species carried from Chili
along the sea-board of Peru, this mode of dispersion cannot act
in the opposite direction. Further than this, the desert zone is,
for the most part, a narrow strip running parallel to the chain
of the Andes, and many Andean types of vegetation have been
modified to suit them to the climate of the western sea-board.
From this source we should expect in the coast-flora a less amouut
of uniformity than is usual in desert-floras ; but it seems likely
that the inea which give rise to a great degree of variation in
the vegetation of each particular place will act here much as
they do elsewhere. To test this conclusion it would, be well if
travellers landing at various points on the coast would register
the species observed at each place. Excepting the rare occasions
when numerous plants enjoy a short-lived existence, the number
of species to be observed is so small that the observation would

give little trouble.
M 2

150 MR. J. BALL ON THE BOTANY

Payta lies about five degrees south of the equator, very near
the northern limit of the so-called rainless zone, but within less
than 100 miles of places on the southern side of the Gulf of
Guayaquil which are described as within the region of tropical
rains, and as possessing the characteristic vegetation of the
Equatorial zone. Rain to an appreciable amount had not fallen for
three years before the date of my visit, and the immediate
vicinity of the port appeared to the eye absolutely devoid of
vegetation. Assisted by several fellow-travellers, who were kind
enough to join in the search, I think it probable that I col-
lected every plant then to be found within a radius of more than
a mile from the town on the plateau about 100 feet above the
sea, which in most places approaches close to the shore. My
collection contains only the following 12 species :—

TEPHROSIA CINEREA, Pers. A stunted form of this widely
diffused species is found in many parts of tropical America, and
extends southward to Monte Video and Buenos Ayres. It is
wrongly described as an annual in Persoon’s ‘ Enchiridion’ and in
DeCandolle’s * Prodromus.’ It has a woody rootstock, sometimes
an inch in thickness, and the branches are often woody at the
base. Tephrosia litoralis, Pers., and 7. procumbens, Macf., are
scarcely distinguishable even as varieties, and several other

described species approach very near, both in structure and
appearance.

HorrxaxsEGGIA viscosa, Hook. et Arn. Bot. Misc. iii. p. 209.
I believe that this species is confined to North Peru. It is true
that specimens were distributed by Cuming with the label
* Panama et Colombia occidentalis; " but there is reason to
apprehend that the labels to Cuming's collections were affixed
without any close regard to topographical accuracy *. The genus
extends from the southern border of the region which I propose
naming Argentaria, chiefly along the west side of the continent, to
New Mexico, and thence eastward to Texas, and, like many other
genera mainly American, has two representatives in South Africa.

PROSOPIS LIMENSIS, Benth. in Hook. Journ. Bot. iv. p. 350,
forma spinosa. This appears to be a characteristic plant of the
rainless zone of Peru, but I have not seen specimens from any

* It would seem as if but two labels were used for all Cuming’s plants

collected on the west coast of tropical America, between 18° or 20° S. lat. and
10? N. lat.—either that above quoted or “ Lima et Peruvia septentrionalis."

OF WESTERN SOUTH AMERICA. 151

locality farther south than the neighbourhood of Lima. Bentham
(Trans. Linn. Soc. xxx. p. 378) has expressed a doubt whether
this is more than a local variety of the wide-spread and eminently
variable species Prosopis juliflora, Sw. In the paper quoted,
containing his final views on the classification of this most
difficult group of Mimosee, Bentham has united to P. juliflora
not less than 17 species of Prosopis described by various
authors.

Acacia tortuosa, Willd. This and the nearly allied A. macra-
cantha, H. B. K., have a wide range in tropical America.

ACACIA ? A slender spiny bush without flower or fruit,
so that the species, and even the genus, remain uncertain.

ENCELIA CANESCENS, Cav., var. PARVIFOLIA. Having observed
specimens of Zncelia growing at numerous stations on the
west coast of South America, I have no hesitation in asserting
my belief that they were all forms of the same species, and
that the form growing at Payta, which is Æ. parvifolia, H. B. K.,
as well as Æ. oblongifolia, DC. Prod., should be united to the
original species of Cavanilles, it being difficult to fix their limits
even as varieties. I venture to doubt whether Asa Gray’s
Californian species Æ. farinosa should not also be united to the
older species.

COLDENIA DICHOTOMA, Lehm. A variable species, having a
wide range on the drier portions of the western side of
America, from the Tropic of Capricorn to Mexico. It has been
erroneously described as an annual, as it has a woody rootstock.
We are indebted to Asa Gray for uniting under the Linnean
genus Coldenia, founded on a plant which extends through
Africa and tropical Asia to the Philippines and Northern
Australia, a number of allied American plants which had been
ranked under five different generic names. Several of these are
strikingly alike in appearance, yet exhibit great differences in
the structure of the fruit. In regard to this I may express a
suspicion, founded on the imperfect examination of several
specimens, that in this group carpological differences such as
usually mark generic distinction, may be found within the limits
of the same species.

GALYESIA LIMENSIS, Juss., var. GRANDIFLORA, Benth. in DO.
Prod. x. p. 296. Found flowering in some abundance in the

152 MR. J. BALL ON THE BOTANY

little gullies that descend from the plateau to the sea-beach
close to Payta. The form described by Jussieu, which seems to
differ only in having smaller flowers, was first collected by
Dombey near Lima, and has been since found at several stations
in the dry country, about the foot of the Cordillera in Peru.
The large-flowered variety is known, I believe, only from Payta
and from Manta in Ecuador, whence there is a specimen in Kew
Herbarium collected by Dr. Sinclair. I know very little of the
coast-region of Ecuador north of the Gulf of Guayaquil, but I
was struck by the arid aspect, as seen from the sea, of the portion,
about 100 miles in length, between the headlands Cabo San
Lorenzo and Cabo Santa Elena. It seems probable that this tract
shares to a considerable extent in the influence of the Humboldt
current, and the accompanying cool southern breezes which
maintain the exceptional aridity of the coasts of Peru and
Northern Chili. The presence of Galvesia at Manta, a small
seaport close to the Cabo San Lorenzo, strongly confirms this
conjecture.

I must here be allowed to express a doubt whether Galvesia
should be maintained as a genus distinct from Antirrhinum. So
long as we had to deal with the Old-World forms of the latter
genus, where the mouth of the corolla is closed by the inflexion
of the palate, or lower lip, the Peruvian plant appeared to be so
far different that no one would have thought of ranking it in the
same genus. But the progress of discovery in western North
America has made known to us a complete series of forms
connecting our common Snapdragon with species in which the
mouth of the corolla is completely open. Two species in parti-
cular, Antirrhinum speciosum, A. Gr., and A. junceum, A. GT.,
forming the section Gambelia of Gray and of Bentham, approach
so nearly in structure to our Galvesia, that it seems to me that
the latter must be placed in the same section of the genus, being
in most respects intermediate between them in character and
appearance, while the bright crimson corollas of the three closely
resemble each other. A. speciosum was first described as the
type of a new genus, Gambelia, by Nuttall; and A. junceum was
described by Bentham, in the ‘Botany of the Voyage of the
Sulphur, as Maurandia juncea. In preparing the general
revision of the Scrophularinee, for the tenth volume of
DeCandvlle’s * Prodromus, Bentham left the last-named species
in Maurandia, and in describing Galvesia remarks that it is

OF WESTERN SOUTH AMERICA. 153

distinguished from Antirrhinum by the form of the corolla.
Several years later, with the ample additional materials supplied
by American explorers, Asa Gray, in one of those remarkable
papers whieh he has published in the * Proceedings of the
American Academy of Sciences,’ reduced the scattered fragments
to order, and arranged the North-American species in sections of
the genus Antirrhinum, including in the section Gambelia the
AMaurandia juncea, Benth. In his final revision of the Order for
the * Genera Plantarum, Bentham adopted the views of Asa
Gray as to the limits of Antirrhinum and the groups under
which the American species should be classed; but he maintained
the genus G'alvesia for the Peruvian species. On a scrutiny of
the generic characters assigned to this and to Antirrhinum, I can
find only three points in which they appear to differ. The style
in Galvesia is said to be minutely two-lobed. In several Antir-
rhina the tip of the style is slightly thickened, as in Galvesia,
and the very minute cleft in the latter is all the difference that
can be pointed out. The cells of the anther in Galvesia are said
to be confluent at the summit, while in Antirrhinum they are
separate. Although I have been unable to verify this character,
I place absolute reliance on the unfailing accuracy of Mr. Ben-
tham; but I do not know whether the anthers of the sections
Maurandella and Gambelia of Antirrhinum have been examined,
or whether the assigned differenee has been found to exist.
Finally, the cells of the capsule in Galvesia are said to open by
an irregular pore below the summit. I am not sure whether
Bentham considered the word irregular to indicate any important
difference between Galvesia and the American Antirrhina, which
have the capsule nearly or quite ovoid and the cells opening by a
single pore to each cell. But in any ease the difference is much
less than that existing between the latter and the European
species, in which the capsule is very oblique, one cell opening by
a single pore, the other by two pores with toothed edges. I may
mention that in Galvesia the corolla is more gibbous at the base,
and therefore nearer to the typical Antirrhinum than it isin A.
junceum. I must not omit to notice the fact that, from the figure
and description of a plant brought from Cerros Island and
described by Kellogg in the ‘ Proceedings of the Californian
Academy,’ vol. ii. p. 17, as Saccularia Veatchii, Bentham conjec-
turally added a second Californian species to the genus Galvesia,
with which Saccularia appeared to him to be identical.

154 MR. J. BALL ON THE BOTANY

On comparing Kellogg's figure and description with the original
specimens of Antirrhinum junceum Y was struck by a suspicion
that they refer to the same plant,and this suspicion was much
strengthened on finding, in the * Synoptieal Flora of North Ame-
rica,’ that Asa Gray has received specimens of A. junceum from
Cerros Island, while he makes no mention of Kellogg's Saccularia
Veatchii, a plant which does not appear to have been distributed.
If my suspicion should be verified, it will go far to establish the
conclusion that Galvesia must be united with the Gambelia section
of the genus Antirrhinum.

[Since the above lines were written I have been informed by
Dr. Gray that, in the fortheoming Supplement to the American
Gamopetale described in his * Synoptical Flora,’ he has identified
Kellogg's Saceularia Veatchii with his own Antirrhinum junceum,
thus confirming my conjecture on this score. But he further
informs me that on examining the living plant of Gambelia speciosa,
Nutt., hitherto known only from dried specimens, he has found
that the projecting palate closes the mouth of the corolla, as in the
true Antirrhina; while it would appear,as well from Kellogg’s figure
as from the dried specimens, that the lower lip in A. junceum, A.
Gr., is nearly or quite plane, as in Galvesia. This being the case,
it would appear that, although Galvesia is nearly allied to some
American species of Antirrhinum, it may be retained as a separate
genus including G. limensis and G. juncea, the synonyms for the
latter being Maurandia juncea, Benth. in Bot. Sulph., Antirrhinum
junceum, A. Gr., and Saccularia Veatchii, Kell.]

Lrrrra RePrans, H. B. K. A single plant without flower or
fruit. It is, however, impossible to mistake the deeply plicated
penninerved leaves of this species, which has a wide range in the
drier parts of tropical America. Zapania cuneifolia, Torr. (Lippia
of Steudel and recent authors), a native of New Mexico and Ari-
zona, which has been overlooked by Schauer in the * Prodromus,

appears to approach very near to the older species, but the leaves
are less deeply plicated.

TELANTHERA DENSIFLORA, Mog. in DC. Prod. xiii. 2. 366.
This is said to be a native of North Chili and (on the authority
of Cuming) of Colombia. For the reasons stated above, no con-
fidence should be placed in the latter locality. A specimen in

Kew Herbarium from near Lima should probably be referred to
this species,

OF WESTERN SOUTH AMERICA. 155

TELANTHERA PERUVIANA, Mog. l.c.,var. DECUMBENS. This differs
from the type specimens only in having the numerous short
branches that issue from the woody rootstock prostrate instead of
erect. The species appears to be confined to the arid coast-region
of Peru. Cuming’s specimen in Herb. Kew. is said to come
from Lima, and specimens from Seemann and Spruce in the same
herbarium were both collected in the desert of Piura.

EUPHORBIACEARUM, sp.? Two or three plants without flower or
fruit, with leaves much like those of Buddleia incana, Ruiz & Pav.,
were seen growing on the bare stony plateau. I was told that the
local name is Sapote, that the plant is considered poisonous, and
that cattle, though eagerly devouring almost any green thing, will
not eat the leaves or the fruit.

This meagre collection gives somewhat negative results. Of
the twelve species seen two are indeterminable. Ofthe remaining
ten species four are, so far as we know, confined to North Peru
or its immediate neighbourhood, each having a maximum range
of about ten degrees of latitude. Three other species that are
confined to the Pacific coast have a much wider range, and two of
these, at least, are nearly allied to plants inhabiting South Cali-
fornia, Arizona, or New Mexico. Finally, three species are
plants that range widely through tropical America and extend to
the Atlantic coast.

A botanist who may be fortunate enough to visit Payta soon
after one of the rare recurrences of abundant rain would doubtless
find numerous annuals, of which none were seen by me, and his
collections would probably throw a fresh light on the conditions
of vegetable life in this region.

There is little worth recording respecting the other places
which I was able to visit on the coast of Peru.

The flora of Lima and its immediate vicinity is pretty well
known, as most botanists who have visited Peru have spent some
time at the capital. Aided by the local knowledge of Mr. Nation,
a resident at Lima (who, amidst many difficulties, has pursued the
study of natural history), but at the least favourable season, I was
able, in the course of three short excursions, to collect 101 specics
of flowering plants, chiefly in the gravelly bed of the Rimac or
along the irrigation channels which alone render cultivation pos-
sible in this arid region.

156 MR. J. BALL ON TITE BOTANY

Arica is one of the few spots on the coast where water from
the Andes has made its way across the arid plains, chiefly through
underground channels, and the result is to enable a good many
species to maintain a continuous existence. On the sandy ground
north of the port clumps of shrubby Composite, formed of Pluchea
Chingoyo, DO., Baccharis lanceolata, Ruiz & Pav., Tessaria
ambigua, DC., and Trixis frutescens, P. Br., rise to a height of six
or eight feet. In as hort walk, during the brief halt of the steamer,
I was able to secure 25 species, including among them the fine
Cesalpinia Gilliesti, possibly introduced for the sake of ornament.

At Tocopilla, a place formerly belonging to Peru, but now
annexed to Chili, I, for the first time, found a spot absolutely
devoid of all traces of vegetation. Our stay there was very short,
but I went for some distance along the base of the rocky slopes
that descend very near to the beach, and ascended here and there
where the ground was not too steep. I was unable to detect the
slightest speck of lichen on the rocks, and I was still more im-
pressed by the appearance of the surface, which nowhere showed
any indication of weathering. Even in the depressions the angles
were all perfectly sharp, and there was no sign that water had
ever flowed or trickled over the surface.

For a distance of nine degrees of latitude, or about 620 English
miles, from Arica nearly to Caldera, the appearance of the coast,
wherever we approached sufficiently near to examine it, showed
the same unvaried aspect of absolute sterility ; but I was assured
by competent witnesses that at several places rain has, at long
intervals, been seen to fall in some abundance, and is certain to
be speedily followed by the appearance of numerous flowering
plants. At Taltal, a small place about 25? S. lat., the general
aspect much resembles that of T ocopilla, but on landing and ex-
amining the rocks I was at once struck by slight indications of
water action ; and on ascending a little steep gully on tbe slope
I was able to secure three plants, two of which were in flower or
fruit. One of these is a Oristaria, probably C. Spinole of C. Gay,
or possibly the nearly allied form described as C. foliosa in the
* Florula Atacamensis’ of Philippi. Many described species of this
difficult genus, peculiar to temperate South America, are not to
be found in European herbaria, and it is scarcely safe to express
a decided opinion as to the value of the diagnostic characters
assigned to them. The second plant found at Taltal, and also
seen near Caldera, in each case without flower or fruit, may

OF WESTERN SOUTH AMERICA. 157

perhaps be likewise referable to Cristaria, although very different
in appearance from the first. If so it is probably C. viridi-luteola,
Gay, nearly allied to C. seselifolia, Turez. The third plant, which
I found growing in some abundance at one place on the rough
slopes, is certainly Teucrium nudicaule, Hook. Bot. Misc. ii. 235.
This well-marked species appears to be one of the characteristic
plants of the arid litoral region of Northern Chili, and to the
best of my belief does not extend beyond it; but it is worth while
to point out the errors, very excusable but very misleading, which
have embarrassed in this as in many other instances the study
of the botanical geography of South America. In 1830 Sir
William Hooker received from Mr. Cruickshanks a considerable
collection of plants from the western coast of South America.
The large majority were collected in Peru in the neighbourhood
of Lima, and most of these in the valley of Canta, while a small
proportion were found in Northern Chili, chiefly at a place called
Arqueros, north of Copiapó, and in the adjoining Atacama desert
region. With his aceustomed activity Sir W. Hooker described
the plants of this collection, many of which were then new to
science, in the * Botanical Miscellany,’ and in describing Teucrium
nudicaule mentions the fact that the label had been mislaid, but
that he believed the plant to have come from the valley of Canta
in Peru. The consequence was that in the 12th volume of the
* Prodromus? Bentham gave that valley, with a note of interroga-
tion, as one of the native localities for Teucrium nudicaule.
Since that date no one of the numerous botanists who have
visited Peru has found the species, while several specimens
have been brought from the desert region of Northern Chili,
chiefly from the neighbourhood of Copiapó. In C. Gay's *Flora
Chilena' the plant is said to inhabit Northern Chili, and Arqueros
is specially mentioned as one locality. There can be little or
no doubt that the plant originally sent to Sir W. Hooker came
from that place. Besides the specimens from Northern Chili
there is one in Kew Herbarium labelled ** Concepcion, Bridges."
If it were authentic this would be a fact of botanieal distribution
more anomalous than the supposed Peruvian habitat; for Con-
cepcion, which is nearly ten degrees of latitude south of Copiapó,
has a moist climate, and the general character of the flora is
markedly different. I have found, however, that the same
locality —Concepcion—is affixed to nearly all the Chilian plants
sent by Mr. Bridges to Sir W. Hooker, probably because they

158 MR. J. BALL ON THE BOTANY

were despatched from that place, and no labels indicating localities
were affixed to them. However that may be, it is abundantly
clear that no relianee should be placed on these labels in Kew
Herbarium as indieations of the origin of the specimens to which
they are affixed. Several of the plants so labelled are mountain
species whose home is on the slopes of the Andes; many others
are desert plants from Northern Chili. A few anomalies of dis-
tribution, if they stood apart, might not be altogether ineredible,
but that a single collector should have found a large number of
plants, known to characterize other regions, at one single point
on the coast where no other botanist has found them, may be
safely dismissed from our belief.

It is remarkable that Teucrium nudicaule is not enumerated
amongst the plants of the desert region in Philippi’s ‘ Florula Ata-
camensis.’ Under the number 268 he mentions, however, an un-
certain species of Zewcrium, of which the specimens had been
lost, but which he had collected in two localities, and these most
probably belonged to T. nudicaule, the only species of the genus
hitherto known to grow in that region.

Flora of Caldera in Northern Chili.

A brief halt at Caldera, which I preferred to devote to a walk
over the sandy and rocky flats near the port rather than to
visiting the neighbouring town of Copiapó, enabled me to add a
few interesting plants to my collection. For the first time since
leaving Arica some appearance of vegetable life was visible on
the coast, in the shape of a few bushes of Baccharis confertifolia,
Colla, growing near the landing pier.

The following list includes the few plants collected, chiefly on
the rocky ground about a mile from the port, with four others
for which I am indebted to an English gentleman who casually
gathered them in the same neighbourhood. As Copiapo les at
the southern border of the desert of Atacama, I have added a

reference to the numbers under which most of them appear in
the * Florula Atacamensis.’

CRISTARIA VIRIDI-LUTEOLA, C. Gay? Quite uncertain, being
without flower or fruit, but certainly the plant seen at Taltal.

Faconta aspera, C. Gay (Phil. 64). This appears to be, as
C. Gay admits, no more than a variety of F. chilensis, Hook. &
Arn.; but both are very nearly allied to the F. cretica, L., the

OF WESTERN SOUTH AMERICA. 159

widely spread species of the hot and dry parts of the Mediterra-
nean region. The wide diffusion of nearly allied forms of this
genus in the desert regions of the Old World, of Western North
America, and of Chili, is an interesting fact of distribution; and
it is perhaps still more remarkable that in Chili both forms seem
to be limited to a small zone and do not extend northward across
the Atacama desert or into Peru. Philippi found our species
no farther north than the neighbourhood of Caldera. Neither
of the Chilian forms is more different from Fagonia cretica than
are several described Oriental species which Bentham and Hooker
are disposed to regard as varieties of the Linnean type. One
is tempted to speculate on the possibility of an early introduction
of the European species by the Spaniards and subsequent modi-
fications of the type.

Loasa CHiLENSIS— Huidobria chilensis, C. Gay (Phil. 110).
This appears to be strictly confined to the Atacama region. Al-
though, as Bentham and Hooker have shown, there are no sufli-
cient grounds for separating this generically from Loasa, it is a
very distinct species with a peculiar habit. The flowering
branches are often nearly glabrous, with very few of the glochi-
diate hairs characteristic of the genus.

GYNOPLEURA RUGOsA- Malesherbia rugosa, C. Gay. Three
species, all allied to Dalesherbia humilis, Don, and inhabiting the
arid region of Northern Chili, were described in the ‘ Flora Chi-
lena,’ and to these Philippi added three more found by him in the
Atacama desert. Further observation is required to ascertain the
constancy of the diagnostic characters adduced by these writers.
My specimen closely agrees with those of M. rugosa received
from the Paris Museum, where C. Gay's herbarium is deposited.

CRUIKSHANKSIA HYMENODON, Hook. et Arn.? I cannot distin-
guish this from the original specimen preserved in Kew Her-
barium, but I am unable to say whether the imperfect specimen
given to me has a perennial root. C. tripartita, Phil. Fl. Atac. No.
165, differs chiefly in being an annual. It was found by Philippi,
who does not enumerate C. hymenodon, at several places near
Copiapó; but I have not seen specimens of his plant.

BaccHARIS CONFERTIFOLIA, Colla in Mem. Acad. Tur. xxxviii.
15 (1835) (PhiL 193)=B. Chilquilla, DC. Prod. v. 419 (1836)
=B. margivata, DC. Prod. v. 402 (1836) — B. parviflora, Pers,

160 MR. J. BALL ON THE BOTANY

non Poir. (sec. DC. l.c.). Most of the specimens in herbaria
labelled B. marginata, DC., and B. glutinosa, Pers., should be
referred to this plant, which should retain Colla's older name,
unless they should all be considered as forms of B. glutinosa,
Pers., as to which I am somewhat doubtful. B. confertifolia
has a rather wide range on the western side of S. America from
Central Chili to Peru. A specimen in Kew Herbarium from
Santiago del Estero in Argentaria, and named by Grisebach B.
glutinosa, Pers., is certainly different from B. confertifolia, the
nervation of the leaves being quite dissimilar.

ENCELTA CANESCENS, Cav., var. TOMENTOSA— E. tomentosa,
Walp. (Phil. 218). As I have already remarked, all the specimens
of Encelia from South America seen by me appear to be referable
to E. canescens. The present variety is very near to A. Gray's
E. farinosa from California.

SkvrALANTHUS ACUTUS, Walp. et Meyen (Phil. 228). This
small bush is rather abundant in some spots near Caldera, but its
area seems to be extremely limited, not having been hitherto
found except in this immediate neighbourhood, though too con-
spicuous to be overlooked. The authors of the ‘Genera Plantarum’
have maintained the name SÁytanthus as originally written by
Meyen, though the discoverer of the species subsequently cor-
rected the name so as to conform to its correct etymology.

CoLpENIA LITORALIS, Phil. Fl. Atac. No.948. Philippi found
this species on the sands of the sea-shore near Caldera; I found
it in some abundance on the sandy flats about a mile inland, with
cream-coloured flowers having much the appearance of a Fran-
kenia. lt is certainly distinct from ©. dichotoma, having the
calyx-segments longer and divergent when iu fruit, while in C.
dichotoma they are adpressed and usually close together over the
fruit. The nutlets in the Caldera plant are subspherical and
granulated on the surface, while in C. dichotoma they are smaller,
ovoid, nearly black and polished. The leaves also are different,
being distinctly petiolate in C. litoralis, and in C. dichotoma
sessile but tapering to the base.

CHENOPODIACEARUM, sp.? Too imperfect. Even the genus
uncertain.

SrrPA ToRTUOSA, Desv. (Phil. 392). This species appears to
be rare and confined to Northern Chili.

OF WESTERN SOUTH AMERICA. 161

STIPA CHRYSOPHYLLA, Desv. This is described and figured in
the ‘Flora Chilena’ (tab. 76), but is not included by Philippi
among the species collected in the Atacama desert. The speci-
mens distributed from Cl. Gay's herbarium were found near
Copiapó.

The flora of Coquimbo is sufficiently well known, but it includes
several curious plants which appear to be confined to a narrow
zone of northern Chili. On the hill behind the town I found,
among others :— Cassia coquimbana, Vog., a very ornamental plant
in full flower; Ophriosporus triangularis, Meyen; Haplopappus
hispidulus, DC., and H. parvifolius, A. Gr. ; Gnaphalium ulophyl-
lum, Hook. et Arn. ; Cephalophora litoralis, Phil.; and Cochranea
stenophylla (Hook. et Arn., sub Heliotropio). lalso observed a
dwarf Mammillaria with crimson flowers growing in some abun-
dance, but of which I found but a single specimen in flower. It
appears to be undescribed, but I do not venture to name it.

No place in South America is better known to naturalists than
Valparaiso, and although I visited it at the most unfavourable
season, at the end of autumn (May), and before the first winter
rain had fallen, I was struck by the exceptional abundance and
variety of the vegetation for a region possessing a warm climate
with a mean annual rainfall of only 132 inches.

Flora of Lota in Chili.

My short visit to Lota, a port about 20 miles south of Con-
cepcion, was chiefly occupied by the remarkable garden, or park,
of Madame Cousino, the owner of the mine of excellent lignite
which has raised this to a place of commercial importance. A
temperate climate, with sufficient rainfall (probably about 40
inches a year)*, makes this spot admirably fitted for the cul-

* The observations of rainfall at stations in Chili exhibit results so inconsis-
tent that it is difficult to feel any confidence in many of them. There is reason
to believe that the observations at Santiago are carefully recorded and that a
suitable instrument is there employed, but as regards the other stations equal
reliance cannot be placed in either respect. The chief cause, however, of the
great discrepancies to be found in the published tables is the extreme variability
of the climate, especially of Central and Northern Chili.

In an interesting work by Don B. Vicuña Mackenna, entitled ‘ Ensayo His-
tórico sobre el clima del Chile,’ that author has traced back from the early
records to the present time the recurrence at irregular intervals of exceptional
seasons with excessive rainfall or extreme dryness, and has shown that, even in
years that cannot be called exceptional, the rainfall in successive years frequently
varies in the proportion of 1 to 2 or 3. It is obvious that the results for a

162 MR. J. BALL ON THE BOTANY

tivation of plants of various regions; and, although it is sur-
passed by many others as regards the variety of species under
cultivation, I have seen very few to equal it in interest.

Of indigenous plants very few were in flower, but the beautiful
Lapageria rosea in full bloom, climbing over the shrubs and
smaller trees, was alone sufficient to adorn the scene. The fol-
lowing short list includes chiefly the indigenous plants found on
the low cliffs faeing the shore, but also at least four species in-
troduced from Europe which have apparently become naturalized
of late years.

Linum AQUILINUM, Mol., Hook. et Arn. Bot. Mise. iii. 149—
Linum Maerzi, Benth. Bot. Reg. 1326. This extends along the
coasts of Central and Southern Chili from Valparaiso southward.

comparatively short period which are available for most of the stations men-
tioned below must vary very widely according as years of excessive rainfall or
exceptional dryness happen to be included. As the best available sources I have
compared the measurements given in millimetres by Dr. Julius Hann in his
‘Handbuch der Klimatologie, by M. V. Raulin, in the ‘Zeitschrift’ of the
Austrian Meteorological Society, vol. xi. p. 137, and by D. B. V. Mackenna in
the work above cited, for the places of which the latitude is set down.

Stations. S.latitude. | Hann. | Raulin, | V. Mackenna.
Shs mm mm. mm,
Copiapó Cio aes AN. 22 8 T9 9
ae ees x oe | a 38:5 30
Valparaiso............... SEOIDODOMO 343 350
ue A 33 27 | 360 358-2 419
Constitucion ............ 3 W o D ossi 536
Mala eee 35 26 | 500 504 527
Concepcion ............ gk 49 T oos 2366 1364
Vedi oss 39 49 | 2930 | 9933 2557
Corral 55 di 9952 | 2030 25318 E
Puerto Montt. ......... 4l 30 | 2450 2448-7 233
Andud oa 41 46 | 3400 | 18209 | 2035
Punta Arenas ......... 53 10 | 570 572:8 495

With two exceptions the figures given by Dr. Hann and M. Raulin are
evidently derived from the same materials ; but the sum set down by M. Raulin
for Concepcion is anomalously high, while that for Ancud is little over one
third of that given by another set of observations. If, as appears probable, the
rainfall of 536 millim. at Constitucion is nearly correct, it is not probable that
at Concepcion, only 14° farther south, the average amount is more than twice
as great. Having consulted M. Raulin on the subject, that gentleman has been
good enough to inform me that he places no confidence in the result for Con-
cepcion recorded by him as expressing the annual mean rainfall of that place,
and I incline to believe that even the smaller figure (1364 millim.) is above the
average rate.

OF WESTERN SOUTH AMERICA. 163

Oxatis LAXA, Hook. et Arn. Bot. Beech. Voy. i. 13? This
approaches near to O. albicans, H. B. K. O. laxa extends from
Coquimbo to Chiloe, while O. albicans has a rather wide range in
tropical America.

FRANCOA SONCHIFOLIA, Cav. The five described species of
this genus, which is peculiar to Chili, are very nearly allied; and
Mr. Rolfe (‘ Gardeners’ Chronicle,’ for 26 Aug. 1882) has therein
maintained that the genus consists of a single variable species.
My opportunities for studying the question have been insufficient,
but I may remark that if his conclusion be adopted the name
sonchifolia should be retained in honour of its original discoverer
Feuillée, who (‘ Journal des Observations,’ ii. 742, tab. 81) has
given a very accurate description with a correct figure of the
flowering plant, under the name Llaupanke amplissimo Konchi
folio. It extends through Central and Southern Chili.

EUGENIA APICULATA, DC. Prod. iii. 276?—E. Luma, Berg.
According to C. Gay this species is confined to Southern Chili,
and there is a second (allied) species, not described by him, to
which the Lota plant may perhaps be referred.

GRISELINIA ALATA, sp. nova, This plant has long existed in
herbaria, but has not been described. It was apparently first
collected by Macrae near Concepcion, and subsequently at Lota
by Lechler. I found the male flowers fully opened, and the
female plant with ripe fruit and with flowers just expanding.
It appeared to grow as an epiphyte at the base of some of the
larger trees in the park (Myrtacee or Laurinee), but in my
haste I failed to ascertain the point with sufficient care. I sub-
join the description.

Frutex (ad basin arborum epiphyticus ?) ramosus ; ramis rigidis
quadrangulis, junioribus (presertim pl. feminez) late alatis; foliis
elliptico- vel ovato-lanceolatis acutis, margine integris, vel basin
versus spinoso-denticulatis, coriaceis, nitidis, in pagina inferiore
insigniter reticulato-nervosis, brevissime petiolatis, petiolo cum
ramo articulato; inflorescentia paniculata, pedicellis brevibus
floribus subzquilongis, pubescentibus, cum flore articulatis;
floris masculi filamentis brevibus; antheris lateraliter dehiscen-
tibus; floris feminei apetali calycis dentibus minutis; ovario
uniloculari; bacca monosperma. Species distinctissima, ramis
alatis, forma foliorum, ceterisque notis insignis.

LINN. JOURN.—BOTANY, VOL. XXII. N

164 MR. J. BALL ON THE BOTANY

It is remarkable that Grisebach, who, in his memoir on the
collections made by Philippi and Lechler in Southern Chili, has
described a new species of this genus under the name .Decostea
iodinifolia (* Bemerkungeu', p.34), should not have been acquainted
with the present species, which appears to have been twice dis-
tributed by Lechler (under numbers 1481 and 3269). There
can, I think, be no second opinion as to the propriety of uniting
Decostea of Ruiz and Pavon with Griselinia, as first proposed by
Baillon; but there may remain room for discussion as to the
true place of the genus in the natural arrangement. In the
memoir above cited (pp. 18 et seq.) Grisebach has discussed the
affinities of Decostea at some length, and has shown that jn many
important respects it approaches more nearly to Jlicinee than to
Cornea, amongst which it is placed in the ‘Genera Plantarum.’
It adds one more example of a group of vegetable forms inter-
mediate in structure between families which in the course of
development have diverged from the ancestral type in different
directions.

The geographical distribution of all the known species points
to the probable origin of Griselinia in the lands of high southern
latitudes. With the exception of one species found in the Organ
Mountains in Brazil, the genus is known only in New Zealand
and on the west side of South America.

SCABIOSA MARITIMA, L. Certainly introduced from Europe,
now naturalized.

EUPATORIUM GLECHONOPHYLLUM, Less. This species appears
to be common in Central Chili, not extending far to the north
of Valparaiso, while Lota is near to its southern limit.

Baccuaris EvpATORIOIDES, Hook.et Arn. A native of Southern
Chili whose northern limit is apparently about Lota. There is
a specimen in Kew Herbarium, collected by Dr. Eights who
accompanied a United States surveying expedition, labelled
* La Moche Island, S. Pac. Oen." This name was doubtless taken
from a French chart, and indicates the small island of Mocha,
about 20 miles from the coast, and about a degree S. of Lota.
Dr. Eights collected plants in several other islands on the Chilian

coast, to which on the labels has been added the descriptive phrase,
* South Pacitic Ocean.”

Though not incorrect this is certainly
misleading.

OF WESTERN SOUTH AMERICA. 165

ANTHEMIS CorUrA, L. Introduced from Europe, but thoroughly
naturalized in Chili.

CotTuLa CORONOPIFOLIA, L. A cosmopolitan species, whose
origin in America is uncertain. It is undoubtedly indigenous in
Australia, Tasmania, New Zealand, and in Southern Africa, but
is now thoroughly established in many other parts of the world.
Antarctic islands may have been the original home of this and
several allied species.

SENECIO VULGARIS, L. Apparently of recent introduction in
Chili, but now naturalized. It is not included amongst the 113
species of Senecio described in the ‘ Flora Chilena.’

Muttsta ILICIFOLIA, Cav. Like most species of this singular
genus this has its proper home in Chili. It has been found at
several places in the Andes, and extends southward to the island
of Chiloe.

Hypocueris LECHLERI? My single specimen closely agrees
with a plant found by Lechler in the province of Valdivia, and
named by Schultz Bipontinus Achyrophorus Lechleri. In his
revision of the species of Achyrophorus (‘ Pollichia,’ 1859, p. 66)
Schultz subsequently identified his own A. Lechler? with A. tene-
rifolius, Remy. "That name is objectionable as being too like the
older name A. tenuifolius, DC. ; and if this be retained as a species
it may best be called Hypocheris Lechleri. It appears to me that
the species of this section of the genus Hypocheris have been
unduly multiplied; but I have not had access to sufficient materials
to form any judgment as to the true limits of the numerous
forms that abound in Chili and Peru. I should conjecturally
refer this as a variety to H. apargioides, Hook. et Arn. (sub Se-
riola).

CREPIS VIRENS, L. Doubtless introduced from Europe.
Lechler collected a Crepis somewhere in this part of Chili which
appears to have been named, but not published, by Scbultz Bipon-
tinus as Crepis Lechleri. In Kew Herbarium there are specimens
of Crepis virens from Valdivia ; but no botanist has detected any
indigenous species of Crepis in Chili, and it is allowable to suspect
that Lechler's plant, which I have not seen, is merely a form
of the common European species.

LOBELIA TENERA, H. B. A. This does not appear to have

166 MR. J. BALL ON THE BOTANY

been found elsewhere in Chili, and may possibly be an escape
from the Garden at Lota. It is indigenous in Peru, Ecuador,
and Venezuela.

DzsrowTAINEA SPINOSA, Ruiz §& Pav. This remarkable plant
extends in South America from Fuegia to the Andes of Ecuador
and Colombia in several forms, showing considerable varieties of
foliage, which have been described as distinct species; but these
are connected by intermediate forms. Its true affinities were
fully diseussed by Bentham in the first volume of this Journal;
but it has no close alliance with any other existing genus, and its
origin must be sought at a remote period of vegetable history.

Lorantuus Escusonotzranus, Mart. First found by the
traveller whose name it bears in the neighbourhood of Concep-
cion. It appears to be confined to this part of Chili.

Ruopostacuys BICOLOR, Benth. et Hook. f. Gen. Pl. ii. 662 =
Bromelia bicolor, Ruiz 4 Pav.? A very singular and highly
ornamental plant, which appears to be common on the sea-cliffs
about Lota. The flowers, which are crowded into a nearly flat
sessile head, are surrounded by a large rosette of rigid leaves,
spinosely serrate, with the upper surface bright red (almost scarlet)
along the middle, and green at the edges. The lower leaves are
more than a foot in length, and gradually pass into papyraceous
oblong bracts, fimbriate at the summit, surrounding the head of
flowers. These bracts are of a paler rose hue, and the innermost
almost pure white. The leaves are broader than in specimens
of R. bicolor which I have seen, and my plant does not exactly
agree with the descriptions of the seven or eight published species,
most of which appear to be little known, and several are not to
be found in European herbaria.

LAPAGERIA ROSEA, Ruiz § Pav. This beautiful plant belongs
to a group of genera which Bentham classed as a distinct tribe of
Liliacee, intermediate between Asparagee and Polygonatee. Of
the seven genera four are monotypic, and none contains more
than three species. Three genera are confined to extratropical
South America, one is limited to South Africa, and one to Aus-
tralia. Of those remaining, one is represented by a single species
in Australia, and by another nearly allied in New Caledonia and
the Pacific Islands ; while the last, Luzwriaga, has two species in
Chili, and a third which is common to Magellania and New
Zealand. These facts, which point to the extinction of a large

OF WESTERN SOUTH AMERICA. 167

number of intermediate types, suggest the probability of a very
remote origin of the entire tribe in high latitudes of the Southern
hemisphere.

NoTHOSCORDUM striatum, Kunth, Enum. iv. 459 = Allium stri-
atum, Jacq. This species seems to have a very wide range in
Ameriea, extending from Canada to Chili and Argentaria, sup-
posing that the plant mentioned by Grisebach as Allium euosmon,
Lk. & Ott.,is one of the numerous forms which have been de-
scribed as distinet species by various authors, but are united to
N. striatum by Regel in his excellent monograph of the genus
Allium.

SCIRPUS SETACEUS, L., var. PYGM.EUS = Isolepis pygmæa, Kth.
Bockeler has judiciously united the several forms of this cosmo-
politan species. This variety is mainly spread throughout the
Southern hemisphere. Bóckeler mentions Crete as a locality ;
but if I am not mistaken it has been found in many other stations
in the Northern hemisphere.

GLEICHENTA CRYPTOCARPA, Hook. Thisisa characteristic form
of Southern Chili, and appears to have been found also in the
Falkland Islands.

GLEICHENIA PEDALIS, Kaulf. Known only from S. Chili and
the island of Juan Fernandez.

ALSOPHILA PRUINATA, Kze. This extends from the West
Indies through the moister parts of tropical America to the
Chilian coast, but does not appear to have spread southward on
the east side of the continent.

I traversed the Channels of Western Patagonia and the Straits
of Magellan in June, after the winter season had fairly set in;
and of course I was not able to collect many plants in a satisfac-
tory condition. I was able to land at two places in the Channels—
Eden Harbour and Puerto Bueno. The former appears to be
an excellent station for a botanist, as it offers considerable variety
of ground. Although the superficial crust of spongy soil was
hard-frozen, many delicate plants appeared to be unaffected by
the season. Amongst other interesting species I gathered Pseu-
dopanax racemiferum (Miq., sub Sciadophyllo)= Aralia letevirens,
Fl. Chil, Griselinia ruscifolia= Decostea ruscifolia, Clos in Fl.
Chil., Baecharis magellanica, Pers. (in fruit), Pernettya empetri-

LINN. JOURN.— BOTANY, VOL. XXII. o

168 ON THE BOTANY OF WESTERN SOUTH AMERICA.

folia, Gaud., and P. pumila, Hook. f., Desfontainea spinosa, Ruiz
& Pav. (with flowers only partially withered), Mitraria coccinea,
Cav. (in full fruit), Dacrydium Foncki (Phil., sub Lepidothamnio),
Libocedrus tetragona, Endl., Podocarpus nubigena, Lindl., appa-
rently about the southern limit of this curious species, PAylesia
buxifolia, Lam., Rostkovia grandiflora, Hook. f, aud Elynanthus
laxus, Benth. & Hook. f. Gen. Pl.=Chetospora laxa, Hook. f.—
Chetospora rhyncosporoides, Steud.= Carpha paniculata, Phil. This
was first described by Sir Joseph Hooker in the ‘ Flora Antarctica’
from a specimen collected by Darwin at Cape Tres Montes ; and
a second species, Chetospora antarctica, is described and figured
in the same work from a specimen, also collected by Darwin
above Cape Tres Montes. This latter appears to be a mountain
variety of the common species, which extends from the Cordillera
de Rauco in Chili to the Channels of Patagonia, but has not been
found so far south as Fuegia or the Straits of Magellan. The
only grass collected was Hierochloa antarctica, R. Br., var. = H.
altissima, Steud. With reference to the last-named species 1
may add the remark that, with the single exception of H. tereti-
folia, Steud., all the specimens of Hierochloa that I have seen
from Chili or Magellania appear to me to be referable to H. ant-
arctica of R. Brown, under which should be ranked as synonyms,
or as varieties, H. altissima, Steud., H. redolens, R. S., H. magel-
lanica, Hook. f., H. utriculata, Kunth, and H. arenaria, teud.
The characters drawn from the smoothness or roughness of the
sheaths or leaves, and the presence or absence of hairs on the paleæ
seem to be unreliable, as these vary among plants growing in the
same locality, and even the relative length of the glumes varies
considerably.

The forest at Eden Harbour appears to consist altogether of
Fagus betuloides, Mirb., and the deciduous species (F. antarctica)
was seemingly uncommon in the northern part of the Channels.
It is, however, very difficult to distinguish the closely allied
species F. betuloides aud F. Dombeyi, and they may grow together
at Eden Harbour, as they certainly do at Hale Cove, somewhat
farther north in Messier's Channel.

CONTRIBUTIONS TO THE HISTORY OF CERTAIN CONIFERS. 169

Contributions to the History of certain Species of Conifers.
By Maxwzrr T. Masters, M.D., F.R.S., F.L.S.

[Read 21st January, 1886.]
(PrArzs II.-X.)

Tne following notes are put together with a view of supplying
additional information concerning various species of Conifer
whose history has been imperfectly known or misunderstood.
In compiling them I have availed myself of the resources of
London Museums and Herbaria, of communications from Sir
Joseph Hooker, Prof. Sargent, Mr. Syme, the late Dr. Engelmann,
my lamented friend Andrew Murray, and other botanists. I
have also specially studied during the last few years the mode of
growth as well as very numerous fresh specimens of foliage and
cones observed in or obtained from the Royal Gardens, Kew,
from various nurseries and private establishments in this country,
on the continent, and in the United States. To Mr. Nicholson
and Mr. J. W. Reed I am also under great obligations for
numerous preparations illustrative of leaf-anatomy. The obser-
vations founded on these materials will supply the basis for a
communication on the general morphology of the Order, which I
hope to lay before the Society on a future occasion.

The present notes refer almost exclusively to species of which
I have been enabled to examine living examples in cultivation as
well as dried specimens, and of which the nomenclature has
from various causes become very involved. Only those references
specially important for the present purpose are here cited, and
no attempt has been made to give full bibliographical references
or complete lists of synonyms. These have been, for the most
part, frequently published in standard works and monographs
that are easily accessible, and to which these notes are intended,
so far as they go, to be complementary.

The accompanying illustrations are reproduced from the pages
of the * Gardeners' Chronicle, and have been selected either for
their intrinsic interest or because no other representation of
the species has hitherto been published in a generally available
form. The drawings have been executed, for the most part, by
Mr. Worthington Smith, some from native, others from cultivated
Specimens.

LINN. JOURN.—BOTANY, VOL. XXII. P

170 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

Blunt leaf and section from
sterile branch.

Pointed leaf and section
from fertile branch.

Abies amabilis.—O one dark purple, slightly reduced. Native specimen.

TO THE HISTORY OF CERTAIN CONIFERS. 171

Abies amabilis, Forbes. (Plate II.)

Abies amabilis, Forbes, Pinetum Woburn. p. 195, t. 44 ; Lindley
and Gordon; Carrière; Engelmann in Gard. Chron. December 4,
1880, p. 720, f. 186-141; Veitch; Sargent, Report on the Forests
of North America (1884), p. 213.

Pinus amabilis, Douglas, Comp. Bot. Mag. ii. p. 93 (1825) ; End-
licher, Parlatore, etc.

Picea amabilis, Loudon, Gordon, Newberry, etc.

Pinus grandis, Lambert [not Douglas].

Abies grandis, var. densiflora, Engelmann olim (Engelmann in
litt. 1).

“Valley of the Fraser River, Engelmann and Sargent, and
probably farther north; south along the Cascade Mountains of
Washington Territory and Oregon." Fraser River, Jeffrey 393!
Cascade Mountains, Lyall!

In the year 1880 Dr. Engelmann, accompanied by Prof. Sargent
and Dr. Parry, ascended Silver Mountain near Fort Hope, Fraser
River, and discovered at an altitude of 4000-5000 feet a Fir
which they recognized to be the Pinus amabilis which Douglas
had first made known fifty-five years previously, but the identity
of which had become confused. A few weeks after, Prof. Sargent
ascended the very mountain, just south of the Cascades of the
Colombia, where Douglas originally discovered the plant. Dr.
Engelmann thus describes the tree :—

“Tt is a magnificent tree, at about 4000 feet altitude; the
largest specimen, growing on the banks of a mountain-torrent,
was probably 150 to 200 feet high, with a trunk about 4 feet in
diameter, branching to the ground and forming a perfect cone of
dark green foliage. The bark of the old tree is 13 to 2 inches
thick, furrowed and reddish grey ; that of younger trees, less
than 100 years old, is quite thin and smooth, light grey or almost
white. It certainly is very closely allied to A. grandis, but is
readily distinguished by its very crowded darker green foliage
and its large dark purple cones. The technical characters are
the following :—

* Leaves densely erowded, dark green, channelled, and with-
out stomata above, keeled, and with two white bands below,
slightly notched on the lateral and sterile branches, acute on the
leaders and on the cone-bearing branches (these acute leaves often
with a few stomata on the upper side towards the tip) (fig. 1);
resin-ducts close to the epidermis of the lower side. Cones

P2

172 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

deep purple, 4 and sometimes 5 inches long, 2 to 2} inches in
diameter, slightly tapering to a retuse tip: scales in 2i order,
not much broader than high, their bracts often more than half as
long as the scale, obovate, abruptly acuminate, deep purple
upwards. Wings of seeds obliquely cuneate, as long as wide, or
longer (fig. 2).

Abies amabilis.—Bract, scale, and seed. (From a sketch by Dr. Engelmann.)

“ A. grandis, which is common on the low lands along the rivers
in Oregon, Washington Territory, and the southern parts of
British Columbia, and abnormally ascends, has lighter-coloured,
usually longer, and always less crowded (distichous, with 2-4
ranks of shorter leaves on the upper side of the branchlets), also
without stomata above, more deeply emarginate on the branches,
obtuse or very shortly euspidate on the fertile branchlets, acute
or acuminate only on the leaders; cones 3-4 inches long, 14 inch,
or scarcely more, in diameter, apple-green, and usually less
resinous ; scales much broader than high, their bracts about one
third the length of the scale, obcordate or retuse mucronate, light
green; wings of seeds hatchet-shaped, wider than long.

“ A. amabilis has the purple cones and sharp-pointed leaves
(on fertile branches) of A. subalpina ; but this latter has not such
crowded leaves, much smaller cones, and is especially distinguished
by the parenchymatous resin-ducts. 4. subalpina, common in
the Rocky Mountains, has been found also in Oregon (probably
only east of the Cascades) and northward, but we have not as yet
met with it here.

* We send a photograph of a cone-bearing branch (woodcut,
fig. 1), and slightly smaller than natural, one of a sterile branch
of an old tree (Pl. IT.); on the latter may be seen the crowded
arrangement of the leaves ; on the former the shape of the cone
and the sharp-pointed leaves at its bases. It will thus appear
that Lambert’s figure of the cone-scale and bract of his Abies

TO THE HISTORY OF CERTAIN CONIFERS. 173

grandis, though a little exaggerated, is in the main correct, espe-
cially as to the long abruptly acuminate bract, the figure of
which has given rise to the confusion with the Californian
A. magnifica.” —G. ENGELMANN.

To these remarks of Dr. Engelmann we may add a few notes
concerning the amabilis generally cultivated in this country :—

Habit as of A. Nordmanniana.

Buds small, ovoid, conie ; scales shining.

Young lateral shoots horizontal, flattish, or with a slight upward
convexity, oblong obtuse in outline.

Branches smoky brown or olive-grey, slightly hirtellous or
glabrous, marked with circular leaf-scars ; shoots given off nearly
at right angles.

Leaves very fragrant, on the sterile branches densely crowded,
in many rows all of about the same length (10-12 lines), those on
the leader-shoot appressed, straight, with the white surface outer-

Fig. 3.

Abies amabilis. —Outline of leaf-section magnified, showing the position of the
resin-canals in a fresh specimen. (From a sketch by Dr. Engelmann.)

most; leaves on the lateral branches given off at an angle of 60°-
80°, all turned to one side, the uppermost appressed and parallel
to the long axis of the shoot, which they completely conceal, the
lowermost rather longer and diverging so as to expose the brownish
bark, all linear obtuse, notched at the apex or, in the case of the
smaller more crowded ones at the base of the shoots, pointed,
dark shining green above, furrowed in the centre, and with few
or no stomata ; lower surface with a prominent midrib, and with
a glaucous band with about six rows of stomata on either side
between it and the green recurved margins. On transverse
section the leaves have a somewhat oblong obtuse figure depressed
in the middle above, prominent beneath, The hypoderm or
strengthening cells are continuous throughout, the palissade cells

174 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

well marked, the resin-canals subepidermal and surrounded by
strengthening cells continuous with the lower hypodermal cells ;
the central fibro-vascular bundle is double (fig. 3).

Abies grandis, Lindl. (Plate IIL.)

Abies grandis, Lindley in Penny Cyclopedia, i. p. 80; Engel-
mann in Bot. Californ. ii. p. 148 ; Sargent, Forest Trees of North
America, p. 212; Masters in Gard. Chron. Feb. 5, 1881, p. 179,
f. 33-36 ; Veitch, Manual, p. 97, f. 23, 24, aliorumque.

Pinus grandis, Douglas, Endlicher, McNab, alior.

Picea grandis, Loudon, Murray in Gard. Chron. July 31, 1875,
p. 185, f. 28.

Abies Gordoniana, Carriére, Bertrand.

A. amabilis, Murray, non alior.

Picea lasiocarpa, Balfour, not Hook., fide Barron in Gard.
Chron. vol. v. 1876, p. 78.

* Vancouver’s Island south to Mendocino County, California,
near the coast ; interior valleys of Western Washington Territory
and Oregon south of the Umpqua River, Cascade Mountains below

Fig. 4.

pred

\
\

N

1

/
f

Bract, scale, and seed of Bracts of Abies grandis to show
Abies grandis,

their variability.

4000 feet elevation, through the Blue Mountains of Oregon to
the eastern slope of the Cour d’Aléne Mountains, the Bitter-root
Mountains, Idaho, and the western slopes of the Rocky Mountains
of Northern Montana,” Sargent, l.c. Columbia River and Cascade
Mountains, Lyall! San Juan Island, Lyall!

In cultivation A. grandis varies in the length and colour of the
leaves, some of the forms having received distinct names, such as
Gordoniana &c., but these are hardly worth keeping up.

As seen in gardens grandis forms an elegant tree of elongated
pyramidal form, the branches slender, arranged in irregular ver-
ticils, and diminishing in length gradually towards the top of the

TO THE HISTORY OF CERTAIN CONIFERS. 175

tree, which thus forms an elongated pyramid. The shoots spread
horizontally are elongate oblong, flat, with a tendency to produce
a few lateral buds at irregular intervals. The bark of the young
shoots is olive-brown, glabrous or hirtellous ; the buds small,
cylindric, elongated, ovoid-conical, rather blunt at the apex, not
globose as in concolor. The scales are oblong-obtuse, dark brown,
covered with a thin film of whitish resin. "The leaves are appa-
rently nearly distichous, spreading nearly at a right angle in one
flat horizontal plane. Both the upper and lower leaves of the

Fig. 5.

—Ó—RMÁ—— p EET E
Eo I a ns Mp

(ey Ly )

MCA

Plan of section of leaf, magnitied, of Abies grandis. RC, resin-canals.

horizontal branches are twisted at the base and leave the shoot
exposed on both sides, very dark green above, with few or no
stomata, but with a central groove, white beneath, and with five
or seven rows of stomata on each side of the midrib, linear oblong
obtuse, channelled on the upper surface and notched at the point.
The leaves on the upper surface of the lateral branches are about
half the length of those immediately beneath them.
The greenish-brown cones are figured on Plate III.

Var. Lowrana, Mast.= Picea Lowiana, Gordon, Pinetum
Supplement (1862), p. 75 ; Pinetum, ed. 2 (1875), p. 218.

Abies Lowiana, Alc Nab in Gard. Chron. 1876, v. p. 78.

Pinus Lowiana, McNab, Proc. R. Irish Acad. 2 ser.
vol. ii. p. 680, t. 46, fig. 5, folior. anat.

Picea Parsonsiana, Barron, Catalogue, 1859; Gard. Chron.
1876, v. p. 77 ; Parsons in Gard. Chron. 1876, Jan. 8, p. 45
(extract from ‘American Gardeners’ Monthly’).

Abies Parsonsiana, guorundam.

Picea sew Abies lasiocarpa, hort. var., non Balfour nec
Hook.

Originally introduced from British Columbia by Jeffrey
in 1851 (fide W. R. McNab), and from California by Lobb
in 1855. California, Calaveros, Upper Sacramento, alt.
4000 ft., and Silver Mount Pass, Hooker and Gray!

176 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

In A. grandis, var.
Lowiana, the habit is
generally rigid; the
branches more formally
and closely verticillate
than in A. grandis pro-
per; the branches rela-
tively stouter, shorter,
and diminishing in
length upwards more
abruptly, so that the
general outline is more
bluntly pyramidal. The
young shoots are flat,
olive-green or brownish,
shining. Buds elongate
ovoid, obtuse or pyra-
midal; scales rounded,
brown, more or less re-
sinous. The leaves are
dark green, linear ob-
long, blunt (not notched
at the point), in few rows
(apparently only in two
rows), relatively very
long, generally all nearly
of the same length,
thinly set so as to ex-
pose the shoot, and
spreading horizontally
in a flat plane, some-
times upturned, and
given off nearly at a
right angle (fig. 6).
Stomata exist on both
surfaces. Even in acci-
dental cases, where the
leaves are contracted in
size owing to some check

to growth, they are all Plan of section of leaf, magnified, of
nearly of the same A, grandis, var. Lowiana.

"i grandis.

Abies grandis, var. Lowiana, cultivated
~ specimen, to show foliage.

length, unlike those of the typical

TO THE HISTORY OF CERTAIN CONIFERS. 177

Var. PALLIDA, Mast. ? =A. concolor, Engelmann (partly).

Leaves of unequal length, flat, pale in colour.

A short-leaved form

: m Fig. 7.
raised from Califor- 2
nian seed, and run-
ning into 4. grandis,
var. Lowiana,and A.
concolor.

California, Low 3
[P Lobb], with which
latter it was identified
by Dr. Engelmann (in
herb. Kew); Upper
Sacramento, Hooker
and Gray!

Most of the garden
plants named concolor (A) bract, and (B) scale of cultivated speci-
are probably seedling men of Abies grandis, var. Lowiana.
varieties of the North-
ern Californian type
above mentioned, and
not of the South Cali-
fornian and Colorado
form, though of late

years the true concolor
has been introduced Imperfect seeds of A. grandis, var. Lowiana,

= S | Vf if ea

x {

from New Mexico by Roezl.

Abies concolor, Lindl.

Abies concolor, Lindley et Gordon, in Journ. Hort. Soc. Lond.
(1850) vol. v. p. 210; Engelmann, partly ; Sargent, Report, p. 212,
partly; Murray, in Gard. Chron. 1875, p. 105. Coulter, Botany
Rocky Mts. (1885) p. 430.

Pinus concolor, Engelmann ex Parlatore, in DC. Prod. xvi. ii.
(1868) p. 427; McNab, in Proc. R. Irish Acad. ser. 2, vol. ii.
p. 681 (1876).

Pieea concolor, Gordon, Pinetum, ed. 2, p. 216.

Picea concolor violacea, Roezl, Gard. Chron. vol. xii. (1879)
p. 684, et in herb. prop.

Santa Fé Mountains, New Mexico, Fendler 828!; Roezl!;
Rio de los Animos, South California, Engelmann, G. R. Vasey! ;

178 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

foot hills of Pike’s Peak, Cañon, at Glen Eyrie, Colorado, Engel-
mann!, Hooker and Gray!; mountains of Southern Colorado,

Fig. 8-

Braets of A. concolor, (After A. Murray.)

Brandegee!; Manitou, Hooker and Gray!; Plumas county,
California, Mrs. Austen 133!

|
Fig. 9. |

Abies concolor.— Cone, foliage, leaf, and section of native specimen
collected by Roezl.

TO THE HISTORY OF CERTAIN CONIFERS. 179

If this be not a species, it is an exceedingly well-marked
variety, owing its peculiarities to climatal and geographical
influences, as pointed out by the late Mr. Murray.

Fig. 10.

NEAL
"
\

A. Bract and scale of Abies concolor, from the back ; B, scale with seed.

In A. concolor (true) the young shoots are grey, hoary, yel-
lowish or olive-green. The buds (fig. 11) are rounded or dome-
shaped, pale-coloured, larger than in the chestnut-brown conoid or
pyramidal buds of A. grandis, var. Lowiana, often very prominent ;
scales oblong obtuse. In addition to the terminal buds, others
appear at irregular intervals on the side shoots. The leaves are
in many ranks, thinly set, so that the shoot is exposed, more or
less upturned (fig. 9), all of about equal length, concolorous and
stomatiferous on both surfaces. As compared with the leaves of
A. grandis, var. Lowiana (fig. 6), they are less regularly arranged,
in more numerous planes, they are given off from the branch at a
more acute angle, and are usually paler, narrower, and more
sharply pointed.

The majority of the plants cultivated as A. concolor have the
leaves more nearly resembling those of A. grandis, var. Lowiana,
and many of them may be included under my variety pallida.
The extreme forms may be considered as geographical varieties.
In any case, it is quite certain that great variation exists among

_the seedlings in the same seed-bed. Supposing external circum-
stances to remain unchanged, much of this early variation may,
and probably will, disappear as the plants grow older; but it is
easy to see how some of these forms may be better adapted for
certain conditions than others, and how they might in that case
be perpetuated.

180 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

Fig. 11.

Abies concolor (true). Foliage and buds of cultivated examples.

Great difference of opinion exists, as will be apparent from
what has been already said, as to the identity and specific limi-
tations of the plants variously included under the names grandis,
Gordoniana, lasiocarpa, Lowiana, Parsonsiana, and concolor.
For the purpose of this communication, and with a view to
facilitate the discrimination of specimens, I have considered the
tree found in Colorado, Utah, and South California to represent
the true Abies concolor, which differs in various particulars,
especially in the leaves and buds, from A. grandis, or A. grandis,

TO THE HISTORY OF CERTAIN CONIFERS. 181

var. Lowiana. Engelmann and Sargent, however, from their
examination of the trees in their native forests, combine with
A. concolor the tree found on the Californian Sierras, and
variously known in gardens as Lowiana, Parsonsiana, or lasio-
carpa (not of Hooker). This is usually very different from con-
color proper, although in nursery seed-beds it is true that the
various forms run so much one into the other that, while it is
easy to pick out the extreme forms, the intermediate forms defy
precise differentiation in the young state.

Professor Sargent and Dr. Engelmann, as already remarked,
consider the Colorado and the Californian trees to be specifically
identical. The former author, in his ‘ Forest Trees of North
America’ (1884), p. 213, gives the range of the species, as he
understands it, as follows :—‘‘ Northern slopes of the Siskiyou
Mountains, Oregon, and perhaps further north in the Cascade
Mountains, south along the western slope of the Sierra Nevadas
to the San Bernardino and San Jacinto Mountains, California,
along the high mountains of Northern Arizona to the Mogollon
Mountains, New Mexico, northwards to the Pike region of
Colorado, and in the Wahsatch Mountains of Utah.”

Prof. Sargent also says of it, * Perhaps merely a southern form
of the too nearly allied 4. grandis, Lindley.”

Andrew Murray also considered A. concolor to be only a form
of grandis, and arrived at this ccnclusion after having seen
grandis growing on the Rocky Mountains. “ The characters,” he
writes in Gard. Cbron. April 1875, p. 465, “ which were supposed
to differentiate it [concolor] from grandis were its white colour,
the same on both sides of the leaf, its somewhat falcate form,
and some supposed difference in the braet. The former are
.... . common attributes of P. grandis in Utah, where the
hoary hue of the vegetation of the plains extends also into the
mountains; and as to the more erect altitude and slightly bent
leaf, the variation is very slight, and occurs also in P. grandis.
The [cone and] the bract also in no respect differ from those of
P. grandis -i .. . ki

In Utah, Murray continues, the colour of the foliage * seemed
to depend a good deal upon the nature of the place, whether
near a stream or barren and exposed, the more barren the more
while ..... There were, in fact, very great differences in the
habit of trees of P. grandis growing side by side, especially when
young. Qne would bear its leaves scattered thinly and sparingly

182 DR. MAXWELL T. MASTERS’S CONTRIBUTIONS

and wide apart; another would have its foliage twice as thick ;
one would have long leaves, as in Lowana, another quite near
it would have them only half the length; and even in colour a
great differenee was constantly to be seen—one would be of a
darker green, another yellowish-olive, a third nearly white, and
so on.”

Sargent, writing in the Gard. Chron. January 2, 1886, says :—
* 4. grandis, as it grows in Vancouver's Island, looks certainly
different enough from <A. concolor of Southern Colorado ; but
there are no real characters to distinguish them, except the
length of the leaves and the number and position of the stomata,
not very valuable or constant indications. The cones and the
leaf-strueture of the two species are identieal, their bark does
not differ more than might be expected in individuals of the
same species scattered over such an immense territory and sub-
jected to such different climatic influences. I have traced this
species, or these species, from Vancouver’s Island, inland at the
north to the extreme eastern limits of their distribution in that
direction, the western base of the Rocky Mountains, and Mon-
tana. I have followed them south through Washington, Oregon,
and California to Arizona and Colorado. For garden purposes
it will be wise, perhaps, to keep these species separate, and even
to admit A. Lowiana, which is only the A. concolor of the Cali-
fornian Sierra Nevada; but, looking at the matter broadly, I am
inclined to believe that these different forms—A. grandis at the
north, A. Lowiana or lasiocarpa in California; A. concolor in
Utah, Arizona, New Mexico, and Colorado— distinct enough in
many minor respects, but not to be distinguished by essential
characters, must all be referred to one species of immense geo-
graphical distribution.”

Taking the broad view of species, the opinion of those best
qualified to give an opinion from having seen the species in their
native haunts is the more worthy of acceptance ; for, however
great the differences that are observable in particular instances,
they are not sufficient to invalidate the probability that they may
have sprung, at no very remote period relatively, from a common
source.

The arrangement above proposed may, in some sort, be looked
on as a compromise justifiable on the grounds of practical utility.
As it will be seen, it assumes the specific distinction of A. grandis
and of A. concolor, referring the long-leaved form Lowiana, and

TO THE HISTORY OF CERTAIN CONIFERS. 183

some of the pale-leaved forms to grandis, and keeping separate
the form found in New Mexico, Utah, &c.

Abies subalpina, Englm.

Abies subalpina, Engelmann in Amer. Naturalist, x. p. 554;
Synopsis Amer. Firs, Trans. Acad. St. Louis, 1878, p- 597, et in
Gard. Chron. Feb. 19, 1881, p. 236, f. 43, 44, 45 ; Sargent, Forest
Trees of North America (1884), p. 211; Coulter, Bot. Rocky Mts.
1885, p. 430.

Abies subalpina, Eng.— Pieea bifolia, A. Murr.—Native specimen.
z A, Cone; B, leaf and cross section.

184 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

? Pinus lasiocarpa, Hook. Flor. Bor.-Amer. ii. p. 163 (1840),
partly; McNab in Proc. R. Irish Acad. ser. 2, vol. ii. (1876),
p. 682, t. 46. f. 7.

Abies lasiocarpa, Nuttall ex Sargent, l. c. p.211; Lindley and
Gordon, Journ. Hort. Soc. Lond. (1850), v. p. 210.

Picea amabilis, Gordon, Pinetum, ed. 2, p. 213, partly.

Fig. 13.

A. Bract and seed-scale of Abies subalpina, from the back; B, scale and seed.

Abies bifolia, Murray in Proc. Hort. Soc. Lond. iii. p. 320,
f. 51-56; et in Gard. Chron. 1875, p. 465, f. 96-97 (Picea) ;
McNab, l. c. t. 47. f. 8, folior. anat.

A. grandis, Engelmann, ex Sargent, l. c. (not Lindley).

Pinus amabilis, Parlatore in DC. Prod. xvi. ii. p. 426, partly.

Picea bifolia, Murray in Gard. Chron. 1875, p. 106.

“Valley of the Stakhin River, Alaska, lat. 60° N. (Muir),
through British Columbia and along the Cascade Mountains to
Northern Oregon (Collier); through the Blue Mountains of
Oregon and the ranges of Idaho, Montana, Wyoming, Utah, and
Colorado," Sargent, l. c. p. 211; Cascade Mountains of Oregon,
Moseley!; Columbia River, Lyall!; Galton Range at 7000 ft.
Lyall!; east side of Cascade Mountains, lat. 49? N., not un-
common up to 6000 ft. above the sea, “ Mareilp" in coll. Zyall! ;
Pringle! ; Colorado, Derry!; Forest City, Engelmann, Hooker
and Gray !, Sargent !

The lasiocarpa of Hooker's description is not easily recog-
nizable. The leaves of the type specimen in the Kew Herbarium

TO THE HISTORY OF CERTAIN CONIFERS. 185

are like those of A. amabilis; but their resin-canals (fig. 14) are
parenchymatous, not subepidermal, and correspond very closely
in structure with undoubted swbalpina wild and cultivated (see
woodcut, fig. 17).

Fig. 14.

Engelmann speaks of it as the Western representative of A.
balsamea. But for the doubt attaching to Hooker's name Jasio-
carpa, Nuttall’s name of Abies lasiocarpa should have precedence

Fig. 15.

Leaves of Abies subalpina, Sterile branch.
LINN. JOURN.— BOTANY, VOL. XXII. Q

186 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

on the ground of priority, but under the circumstances it seems
preferable to adopt Engelmann's designation. In the form of
the seed-scale the present species resembles A. magnifica. The
habit of the wild plant is peculiar: the branches, often suffering
periodic arrests of growth, have in consequence a contracted
. necklace-like appearance. The leaves

on the cone-bearing
branches are sharply pointed.

Fig. 16.

Cone-bearing branch of Abus subalpina. Native specimen.

In cultivation there are at present only small examples. The
bark of the young branches is greyish brown, hirtellous. Buds
ovoid, broadly obtuse. Scales oblong, brown, resinous. Leaves
in many rows given off on all sides at an acute angle, linear

TO THE HISTORY OF CERTAIN CONIFERS. 187

oblong, somewhat dilated near the top, subspatulate, pointed, or
shortly notched, flattish and channelled on the upper, rounded on
the lower surface, with a slightly prominent midrib, on each side

Fig. 17.

Ó

(cA

O c

Plan of section of leaf of A. subulpina, magnified ; cultivated.

of which are two glaucous bands of stomata arranged in five
longitudinal series.

ABIES NOBILIS and A. MAGNIFICA.

The confusion that has reigned as to these plants is analogous
to that already mentioned in the case of 4. grandis, &c. It is
not necessary tő go into the history of the matter, about which so
much has already been written E different botanists, and from
various points of view.

It may suffice here to say that, according to the latest views of
the botanists best qualified to give an opinion (the late Dr.
Engelmann and Prof. Sargent), A. nobilis and A. magnifica are
to be considered distinct species. The points of distinction
relied on are principaily the flattish leaves of A. nobilis, grooved
more or less on the uppér surface (see PI. IV., top fig.), as con-
trasted with the four-sided leaves of A. magrifica, in which there
is no groove, but in which, on the contrary, the midrib is promi-
nent on both surfaces (see woodcut, fig. 20). A. nobilis is a
native of Oregon, not extending south beyond North California ;
while A. magnifica is stated not to occur in Oregon, but to extend
from Mount Shasta along the western slopes of the Sierras to
Kern County, lat. 33? N.

Other alleged points of difference will be alluded to later on;
suffice it here to say that they are in my judgment not constant,
and that although for horticultural, and probably for forestal,
purposes, it may be most convenient to keep the two extremes
distinct, yet the numerous variations indicate the probability of a
common origin, and consequently justify the establishment of a
single species or super-species, to which the name nobilis as the
older of the two may be given. The following arrangement is

therefore proposed :—
a2

188 DR. MAXWELL T. MASTERS S CONTRIBUTIONS.

Abies nobilis, Lindl. (Plate IV.)

Abies nobilis, Lindley in Penny Cyclopedia, i. p. 805; Engelmann
in Botany of California, ii. p. 119; Gard. Chron. November 29,
1879, p. 685; Sargent, Forest Trees of North America, p. 214;

Veitch, aliorumque.

Pinus nobilis, Douglas in Comp. Bot. Mag. ii. p. 147 ; Lambert ;
Hooker; Endlicher; Parlatore in DC. Prod. xvi. ii. 419, et alior. ;
McNab in Proc. R. Irish Acad. ser. 2, ii. p. 699, t. 49. f. 29.

Picea nobilis, Loudon, Gordon, Lawson, Pinetum Brit. ii. p. 181,
t. 28, 29.

Pseudotsuga nobilis, Bertrand in Bull. Soc. Bot. France, xviii.
p. 86; et in Ann. Sciences Nat. sér. 5, tom. 20, p. 86.

“ Oregon, Cascade Mountains from the Columbia River south
to the valley of the Upper Rogue River between lat. 42° and
43° N., and along the summits of the Coast Range from the
Columbia to the Nestucca river," Sargent, l. c. ; on the High
Mountains in the Grand Rapids on the Columbia and near the
base of Mount Hood (1825), Douglas in Herb. Kew! ; Cascades of
the Columbia, Dr. Gardiner!; Cascade Mountains, Engelmann
and Sargent!; Oregon, Mary's Peak, Moseley 3992 (forma
strobilis minoribus)!; California, Mt. Shasta, lat. 41° 50’, alt.
6000 ft., Jeffrey!

A. nobilis in cultivation presents the following characteristics :—
Branches verticillate. Young shoots reddish brown, hirtellous.
Buds ovoid-oblong, brown, more or less resinous. Scales oblong,
the lower ones pointed. Lateral shoots spreading horizontally,
oblong obtuse in outline, more or less four-sided (not flat) owing
to the arrangement of the leaves. Leaves densely crowded in
several rows, those on the lower surface twisted at the base so as
to bring them into the horizontal position, and allowing the
reddish-coloured bark to be seen between their bases ; those on
the upper surface more or less curved, assurgent with the
points direeted either upwards or away from the apex of the
branch, 7. e. towards the trunk, entirely concealing the surface
of the shoot, and giving it the appearance of a flat brush,
the leaves representing the bristles. The typical nobilis may
generally be recognized by its flattish, broadly linear leaves
grooved on the upper surface (see Pl. IV., top fig.), either at the
base only, or for the greater part of its length, with whitish bands
of stomata, on the under surface only or on both. The cones are
conoidal, rarely cylindric, and the bracts projecting, acuminate,

TO THE HISTORY OF CERTAIN CONIFERS. 189

the points generally turned downwards. But even on the same
tree the leaves vary, and those on robust leader-shoots and
on the fertile branches often differ from those on the lateral or
younger shoots. Four-sided leaves are not uncommon on the
fertile branches. In the dimensions of the cone and the extent
to which the bracts project there are equally great variations.
The cones are, when young, frequently of a purple colour,
fading into olive-brown. Sir John Walrond informs me that
the cones formerly produced on his trees in cultivation were
purple, but that of late years, although *the cones maintain
their size, they are of a full brown colour, though produced on

the same tree."

Var. GLAUCA.

In cultivation forms of A. nobilis are obtained with broader,
closer, and more glaucous leaves than ordinary, with the
same structure as in ordinary nobilis. In the young state

Fig. 13.

(onc M
l Ses
| Taek

Bract, scale, and seed of Abies nobilis, var. glauca.

the plant is altogether handsomer than the typical nobilis.
A cone of this variety, produced some years since in the
Cranston Nursery at Hereford, was remarkable for the in-
flexion of the points of the bracts (see woodcut, fig. 18).

Var. MAGNIFICA. (Plate V.)

Abies magnifica, Murray, Proc. Hort. Soc. Lond. iii.
(1862), p. 318, f. 42-50; Gard. Chron. 1875, vol. iii. p. 105,
and vol. iv. n. s. p. 134; Engelmann in Gard. Chron. 1879,
p. 655, £.116 ; in Botany of California, ii. p. 119, et in Coulter's
Botanical Gazette, vii. p. 4; Veitch, Manual ; Sargent, Forest
Trees of North America, p. 214.

A. eampylocarpa, Murray in Trans. Bot. Soc. Edinb. vi.

p. 370.

190 DR. MAXWELL T. MASTERS’S CONTRIBUTIONS

Cone of Abies nobilis, var, magnifica. Native specimen.

TO THE HISTORY OF CERTAIN CONIFERS. 191

A. nobilis robusta, Carrière, Traité, ii. p. 269, et hort. quo-
rund.

Pinus amabilis, Parlatore in DC. Prod. xvi. ii. 426,
partly; McNab in Proc. R. Irish Acad. vol. ii. (1876),
p. 700, t. 49. f. 30.

Pseudotsuga magnifica, McNab in Proc. R. Irish Acad.
ser. 2, ii. p. 700, t. 49. f. 30, 30 a (1876).

* California, Mount Shasta, south along the western slope
of the Sierra Nevadas to Kern County,” Sargent, l. c.

Fig. 90.

Leaves and leaf-section of Abies mobilis, var. magnifica. A, A, resin-canals.

To this variety may be referred specimens received from
California, Lobb 441, a,b (1852) !, Lemmon ; Mount Stan-
ford, Silver Pass and Shasta, Hooker and Gray!; Roezl!;
Low!; Gordon in herb. !

Fig. 21.

SN VADA e,
9

Seed-scales of A. nobilis, var. magnifica, A, from the back, showing the bract ;
and B, from the front, showing the seed. Native specimen,

192

DR. MAXWELL T. MASTERS S CONTRIBUTIONS

In the typical magnifica the young shoots are elongate and
brush-like as in 20bilis, olive-brown, hirtellous. The buds are
small, oblong obtuse. Scales oblong, the lower ones pointed,
the leaves many-ranked, upturned, longer, deflexed, assur-
gent, and with their points often directed towards the apex
of the branch away from the trunk, straighter and more
slender than in nobilis, four-sided and pointed, generally not
at all grooved but sometimes slightly so at the base. The
leaves on the older and on the cone-bearing shoots are often
much shorter, stouter, and more abruptly curved than in
faster growing shoots. The cones are blunt, usually less
tapering than in nobilis, and the bracts usually so much
shorter than the scales that they are completely concealed
by them. The scales of magnifica are, moreover, very broad
and deep, transversely oval with a long wedge-like stalk (see
woodcut, fig. 21). In nobilis the scales are usually smaller,
not so deep, more angular in outline, and with a relatively
shorter stalk (woodeut, fig. 18, p. 189, and Plate IV. B).

The nobilis robusta of Carriére, and of some English gardens,
is, I believe, identical with, or a mere form of, magnifica.

It may be well here to quote Carriére’s description of
Abies nobilis robusta, which agrees well with the magnifica
of English gardens ;—

“ ABIES NOBILIS ROBUSTA, Veitch.

Abies amabilis, hort. non Forbes.

Picea amabilis magnifica, hort.

Picea amabilis, hort. angl.

Picea magnifica, hort.

Picea amabilis robusta, hort, aliq.

Abies magnifica, hort. aliq.

" Cette forme, que l'on trouve aujourd'hui dans le com-
merce sous le nom d’ Abies amabilis, présente les caractéres
suivants :—Branches verticillées, rapprochées, trés-étalées.
Ramules et ramilles distiques. Feuilles trés-nombreuses, "
retournant vers le dessus des branches qu'elles cachent à
peu prés complétement, étroites, parfois subtétragones m
rhomboïdales, ordinairement falquées, un peu contournees
et relevées au sommet. Puis, lorsque les arbres vieillissent,
ces feuilles deviennent plus larges en se régularisant, et
les arbres, alors, ont beaucoup plus de ressemblance avec

TO THE HISTORY OF CERTAIN CONIFERS. 193

VA. nobilis. Du reste ces plantes paraissent trés-variables,
de sorte que, dans un semis, on pourrait facilement en faire
plusieurs variétés trés-distinctes, mais qui, peut-étre, per-
draient peu à peu leurs caractères exceptionnels. Néan-
moins lA. nobilis robusta, beaucoup plus facile à élever que
PA. nobilis, est bien préférable à celui-ci; aussi tend-il à le
remplacer."

The magnifica of English gardens agrees with this descrip-
tion, and is noticeable for its robust habit and formal whorls.
The leaves on the leading shoot are appressed spirally around
the stem, and often stouter than those on the lateral
branches.

A. magnifica and A. nobilis robusta of gardens (for with-
out cones they cannot be separated) differ in habit from
A. nobilis, being much more formal and rigid in the mode of
erowth; the branches are more nearly of equal length, and
the leaves on the sterile branches longer, straighter, and
more slender than in nobilis. Sir John Walrond writes of
one of his trees that “ for several years the lower branches
feathered to the ground and grew in regular formal whorls,
so that the tree was perfect and symmetrical throughout.
For some years, however, the lower branches have died off,
now leaving one third of the trunk bare, although there is
no want of healthy vigour above. These trees have never
shown sigus of coning, though all those of A. nobilis have
done so freely.” A. magnifica starts into growth in this
country much later than A. nobilis, and is therefore less
liable to injury from frost. If what is above said be correct,
it would seem that both forms extend northwards into
Oregon, but that the typical nobilis is replaced in the south
by the magnifica form. Assuredly the character of the four-
sided and ungrooved, as contrasted with the flat and grooved
leaves (Pl. IV., top fig.), cannot be entirely depended on any
more than the size, proportion, and form of the scales and
bracts. The habit in gardens is certainly different. Inter-
mediate forms, however, occur, one of the most remarkable
of which is that figured in Pl. V., and received from
Mr. Heale, the obliging manager of the Cranston Nursery
Company. In this specimen the habit and the foliage are,
for the most part, those of magnifica, but the leaves on the
sterile branches are slightly grooved as in the true nobilis,

194 DR. MAXWELL T. MASTERS's CONTRIBUTIONS

while those on the cone-bearing branches are four-sided,
1} inch long, upturned, and with the midrib prominent on
both surfaces. Tbe olive-brown cones are in form more
nearly like those of magnifica, as also are the scales, but the
long golden-brown bracts are more like those of A. nobilis.
Instead of being deflexed, however, they project at right
angles as shown in the illustration (Pl. V.). In one of
the cones the bracts, as they dried, became partially bent
downward, but this was not the ease with the other two
cones. Little stress, however, can be placed on the direction
of the bracts, nor even on their relative length. Alexander
Braun (Sitz. Bot. Vereins Brandenburg, June 26, 1874,
describes a cone of another species wherein the bracts and
scales of the upper half of the cone were reflexed, while those
in the lower part were in their usual position. Professor
Sargent writing in the Gard. Chron., January 2, 1886, of
the illustration above referred to (Pl. V.), considers it to
represent probably a form of magnifica with exserted bracts,
and Mr. Syme also, in litt. (Feb. 1886), considers the foliage
to be more decidedly like that of A. magnifica than that of
A. nobilis. The cone, too, is more like that of magnifica except
for the exsert bracts. Sargent also adopts Engelmann’s latest
views published in Coulter’s ‘ Botanical Gazette,’ vil. p. 4,
as follows :—* Abies nobilis (Douglas), Lindiey, is peculiar to
the higher mountains of Oregon, and has not yet been found
in California, nor, as far as I can see, in Washington Terri-
tory. Its greenish leaves crowded on the branchlets, and its
Jarge purple cones, with long exsert recurved bracts, well '
characterize it. The tree on Mount Shasta which has gone
by this name (also in the ‘Flora of California’) is distin-
guished by its quadrangular leaves keeled on the upper side ;
its large cones considerably resemble those of nobilis and
have often, not always, exsert and recurved bracts. It is a
form of A. magnifica, Murray, the common Red Fir of the
Californian Sierras, which has bracts normally enclosed.”

Abies religiosa, Schlect. (Plate VI.)
Abies religiosa, Schlectendal in Linnea, v. p. 77; Lindley ;
Spach; Carrière; Mast. in Gard. Chron. Jan. 10, 1885, p. 56.
Picea religiosa, Loudon,' Arboretum, iv. p. 2349; Gordon ;
Murray in Gard. Chron. April 29, 1876.

TO THE HISTORY OF CERTAIN CONIFERS. 195

Pinus religiosa, Humb. Bonpl. et Kunth, Nov. Gen. et Spec. ii.
p.59; Parlatore in DC. Prodr. xvi. ii. p. 420 (1868); Seemahn,
Botany of * Herald.’

Pinus hirtella, Humb. Bonpl. et Kunth, l. c.

Abies hirtella, Lindley ; Carrière.

Picea hirtella, Loudon.

In montibus Mexici et Guatemala, alt. 4000 ped., Humboldt,
Hartweg!, Roezl l, Linden 831, Bourgeau 1148 !, Christy!, Parry
et Palmer 847 !, Seemann!; Guatemala, Skinner !

This beautiful Fir varies in the amount of pubescence of the
stem, in the colour of the leaves and cones, and also in degree of
hardihood.

The leaves start from all sides of the branches, but are so
twisted as to form about four rows, one on each side of the
branch, from which the leaves come off, at an angle of about 40°,
and two others disposed along the centre of the horizontal
branches, the individual leaves being given off at an angle of 10°,
the lateral leaves thus spreading more or less horizontally away
from the branch, the uppermost ones appressed to it, and more
or less parallel in direction to it. The largest leaves measure
about 13 inch. In form they are linear, pointed, curved, deeply
grooved on the deep green upper surface along the midrib, silvery-
grey on the under surface between the prominent midrib and the
strongly recurved margins. Ona cross section the deep central
groove on the upper surface is very conspicuous, as also are the
revolute margins. The two resin-canals are close to the epidermis
on the lower surface, and the strengthening cells or hypoderm
are either continuous, or in other specimens they do not form a
continuous layer but are broken up into detached masses. In
all the native Mexican specimens examined by me the hypoderm
was continuous.

The buds at the end of the lateral shoots are subglobose,
covered with whitish or pale violet obtuse scales.

The cones are erect, 5 by 23 inches, cylindric, oblong, slightly
narrowed towards the top, rich dark violet, or in some cases paler,
covered with whitish resin, with prominent bracts ending in an
acuminate point and markedly reflected. (See Pl. VI.)

My knowledge of the cultivated plant is derived from speci-
mens grown in the garden of the Hon. and Rev. J. T. Boscawen,
Lamorran, Cornwall, and of Mr. Smith Barry, Fota Island, Cork.
In the latter garden there are two varieties, differing as above

196

DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

of A. hirtella.

mentioned in colour and in hardihood, which is significant in
vfew of ihe question as to the specifie distinctness or otherwise

= : ; PY,

SO

á J A y
7 Pi
Vi t 7 A !
A ae A A
f A er — E: | /
A I — = a j ^
d = ee : T | Wl /
£ ~ PL, y
N / f
/ 7
f "7
VY ,

NS S Y M /
S 1 Y H ij

ge <á

Foliage of Abies Fortunez,

TO THE HISTORY OF CERTAIN CONIFERS. 197

Abies Fortunei, 4. Murr.

Abies Fortunei, 4. Murray, Pines and Firs of Japan (1863),
p. 49; Hance in Journal of Botany, vol. xx. (1882) p. 32; Gor- .
don, Pinetum, ed. 2, p. 27; Masters in Journ. Linn. Soc. vol. xviii.
p. 522, et in Gard. Chron. March 15, 1884 [April 3, 1866], c. ic.
xylog. hic repet.

Abies Jezoensis, Lindl. in Paxtons Flower Garden, May 1850,
et in Gard. Chron. 1850, p. 311; Flore des Serres, vol. vii. 223,
vol. ix. p. 7; not of Siebold and Zuccarini.

Picea Fortunei, Murray, Proc. Hort. Soc. 1862, p. 421.

Keteleeria Fortunei, Carriére, Revue Horticole, 1868, p. 132,
e ic. col.

Pseudotsuga J ezoensis (nomen tantum), Bertrand in Ann. Sc.
Nat. sér. 5, Bot. t. xx. p. 87. à

Pinus Fortunei, Parlatore in DC. Prod. xvi. ii. p. 430 (1868).

In China austro-orien- Fig. 23.
tali ad templum prope
Foo Chow Foo, Fortune,
n. 50!; Alfred Hance
(1873) ; et W. Hancock
(1881), fide Hance supra
citat. Frequens in mon-
tibus ad septentrionem,
Foo Chow, Maries!

The history of this tree
is now well known, and
the confusion with 4. Je-
zoensis has been well
cleared up by the late
Mr. Andrew Murray.

Fortune speaks of it as
a magnificent tree with
the habit of a Cedar of
Lebanon, with beautiful
erect purple cones thickly
grouped, like rows of
soldiers, on the branches.
Fortune saw but the one
tree in the vicinity of a

temple, where it h ; x p
pie, as been Abies Fortunei. Portion of the bark to show its
seen by other travellers, ~eork-like character.

Cultivated specimen.

198 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

In this country the
iree has not thriven;
but in the Nursery of
Signor Rovelli, at Pal-
lanza, is a magnificent
specimen of oblong or
pyramidal form, with
rather dense verticils
of horizontally spread-
ing branches and with
thick spongy cracked
bark like that of the
Cork -Oak (woodcut,
fig. 23). The smaller
branches — are grcy,
marked with the circular
leaf-scars of a Silver
Fir. The young shoots
are glabrous, orange-
red. The buds are
ovoid oblong, with ob-
long, rather acute

brownish scales, mem- Abies Fortunei,—Cone erect. Cultivated
branous at the edges. - specimen.
The lowermost scales

; SN NINN
are acuminate, and tran- AN n a SANAN
sitionalin form between |
the leaves and the bud-
scales. The leaves are
in many ranks, spread-
ing more or less on all
sides. On section, they
show a layer of hypo-
derm beneath the epi.
dermis, palissade cells,
transfusion-tissue, and
a central fibro-vascular
bundle surrounded by

a well-marked sheath. : ed:
The resin-canals are t Abies Fortunei. —A, seed ; B, seed-scale and wil
3 pe EO SWO c, side of seed; p, side of seed-scale, showing
in number, one on each bract ; E, bract and seed-scale.

TO THE HISTORY OF CERTAIN CONIFERS. 199

side above the lower Fig. 25.
epidermis and near to
the edge of the leaf.
The cones (fig. 24),
which are stalked, vary
considerably in size and
less so in form. The
bracts in the Italian
specimen are shorter
than the rounded scales;
and usually concealed ;
.but here and there the
persistent acuminate tip
of the bract belonging
toa given scale projects
in the interspace be-
tween the upper part of
the two scales beneath
it; and the seeds are
provided with a hatchet-
shaped wing. Mr. Ben-
tham was quite correct
when, in the ‘ Genera
Plantarum,' iii. p. 442,
referring to Carriére’s
genus Keteleeria, he
said :— Keteleeria ....
est verisimiliter Abietis
species, strobili squamis
diu persistentibus."
Hance (7. c.) concurs in
this opinion ; and as the
leaf-insertion is that of
the Silver Firs, and the
cones* ereet, not, as ; V¢
once supposed, pendent, dE

there is no ground for @

including it among the |
Spruces (Picea), still
less for considering it
the representative of

Cone-bearing branch of Adies Fortunei, reduced
one half; cultivated specimen.

Fig. 26.— A'hrotazis laxifolia, cultivated specimen, with cones.

CONTRIBUTIONS TO THE HISTORY OF CERTAIN CONIFERS. 201

a distinet genus intermediate between the Spruces and the
Silvers.
Athrotaxis laxifolia, Hook.

Athrotaxis laxifolia, Hook. Ic. Pl. t. 573; Lond. Journ. Bot. iv.
p. 449; Fl. Tasman. p.354; Parlatore in DC. Prod. xvi. pars ii.
p. 434; Masters in Gard. Chron. Nov. 7, 1885, p. 584.

Of this species, originally described by Sir Joseph, then
Dr. Hooker, Parlatore writes, l. c. :— Species in plantis vivis
melius illustranda." In truth, the figure heretofore published
(destitute of cones) and the descriptions given were not sufficient
to enable the botanist to decide whether A. laxifolia was entitled
to specifie rank, or whether it was not better to consider it a mere
form of A. cupressoides, or perhaps an intermediate between that
species and A. selaginoides. The evidence now available would
rather lead to the notion that although all three may be deriva-
tives from one form, they are now sufficiently differentiated to be
kept distinct. A. lawifolia, as the name implies, has the leaves
not so closely appressed in the other two species, and different
also in shape. The cones are subglobose, the scales spirally
arranged, each with a long stalk-like portion, dilated above
into a thickened club-shaped extremity, from the dorsum of
which proceeds an ovate acute, up-turned, originally leafy pro-
cess. The seeds are roundish, provided with a deep wing on
either side.

The specimen figured (woodcut, fig. 26) was obtained from the
collection of John Rashleigh, Esq., Menabilly, Cornwall.

The species of Athrotaxis may be compared with the figures of
Walchia imbricata, Schimper* (Permian), and especially with
Echinostrobus Sternbergii from the lower Cretaceous beds.

Cephalotaxus pedunculata, Siebold § Zucc.
Cephalotaxus pedunculata, Siebold and Zuccarini, Flor. Japon.
ii. t. 183; Endlicher, Carriére ; Parlatore in DC. Prod. xvi. ii.
p. 003; Mast. in Gard. Chron. Jan. 26, 1884, p. 113.
In Japonia, Siebold ; China, Fortune T
Of this species, only the male plant is described by authors ;
but if I am correct in my determination, some of the female
plants called in gardens C. Fortunei really belong here. Seeds

* Renault, ‘Cours de Botanique Fossile,’ 1885, tab. 8, Walchia; tab. 12.

f. 14, Echinostrobus.
LINN. JOURN.— BOTANY, VOL. XXII. R

202 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

have been produced on some of these plants, and which are placed
on deflexed stalks about half an inch long, rather shorter than
the ovoid fawn-coloured fleshy seeds. Fortune, on his second
visit to China, ascertained that this plant was quite distinct from
C. Fortunei; and, on his authority, C. pedunculata is cited as
a Chinese plant, though there are no specimens from that country

Seed-bearing shoot of Cephalotaxus pedunculata. Drupe-like seeds, olive-
coloured. Cultivated.

in the herbarium (see Fortune in Gard. Chron. Nov. 28, 1863,
p. 1134). It is supposed that the flowers of a female plant
of this species which bore seed in Messrs. Paul's Nursery at
Cheshunt might have been fertilized by the pollen of the Yew;
but there is unfortunately no record of the seeds having been
sown.

In the garden of the Rev. J. Goring, of Steyning, Sussex, is ?

TO THE HISTORY OF CERTAIN CONIFERS. 203

bush which seems to differ from C. pedunculata principally in the
globular berry-like seeds, unlike anything yet described in the
genus. The shoots, moreover, have a brownish chestnut colour.
As the foliage is so like that of C. pedunculata, I have preferred
to consider it as a variety of that species. It is very probable
that this is the female plant above referred to which Fortune
met with on his second journey (Gard. Chron. 1863, p. 1134),
and which appeared to him to be * new to science." Be this as
it may, Mr. Goring obtained both trees—pedunculata true, and
the present one (pedunculata, var. spheralis)-—from a nursery at
Bagshot in 1865 or 1866, under the name of C. Fortunei; but
certainly neither of them seems entitled to that appellation. It
should be stated that the globular form is much less apparent in
the dried specimens.

The new variety, of which an illustration is appended, may
be thus characterized :—

Var. SPHHRALIS, nov. var. ramis pallide castaneis; novellis
viridibus glabris; foliis 14-2 pollic. lineari-falcatis sensim acumi-
natis, planis vel leviter arcuatis; gemmarum terminalium squamis
oblongis acutis, haud acuminatis, amentis masculis . . . . pedun-
culis seminiferis versus basin ramulorum aggregatis 4 poll.
long. seminibus drupaceis sphericis parum brevioribus. Ex hort.

Goring. (Plate VII.)

Picea Omorika, Panic. (Plate VIII.)

Picea Omorika, Pantic, C. Bolle, Monatsh. des Vereines zur
Beforderung des Gartenbaues, 1877; Reichenbach in Botanische
Zeitung, 1877 ; Gard. Chron. April 14, 1877, p. 470, et May 19,
1877, p. 620, March 8, 1884, p. 308; Ascherson, Sitzb. der
Gesells. Naturfreunde zu Berlin, 1881; Boissier, Flora Orien-
talis, v. p. 701, sub P. orientali obiter.

* Arbor excelsa, coma anguste pyramidali; ramis brevibus sub-
verticillatis, superioribus erectis, mediis horizontaliter patentibus
ae inferioribus pendulis cum apicibus arcuatim adscendentibus ;
ramulis hirsutis; foliis solitariis, rectis aut incurvis planiusculis,
nervo utrinque prominulo, obsolete tetragonis apice acuminatis
acutis aut obtusis cum apieulo cartilagineo, superiore pagina
eximie glaucis; strobilis sat parvis oblongis in ramis varie longi-
tudinis erectis, horizontaliter patentibus aut pendulis; squamis a
basi cuneata subrotundis, dorso sub apice striatis margine eroso-
denticulatis; bracteis obovato-cuneatis apice denticulatis, sua

squama multo brevioribus; uuculis parvis obovatis ala obovata-
x 2

204 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

cuneata subobliqua margine subintegra triplo brevioribus."—
J. Pancic.

This tree is still so little known that it may be convenient to
repeat Dr. Pan'ic's diagnosis, and lay before the reader an illus-
tration from a specimen collected. by Dr. Pančic in the mountains
of Servia (Plate VIIL). It also occurs in Bosnia and Monte-
negro. For the opportunity of doing this and of examining the
specimen, I am indebted to Messrs. Froebel of Zurich and to
Mr. George Nicholson.

The shoots have the “pegged” character common to the Spruces,
due to the projections at the base of the leaves, and are moreover
covered with short stiff hair. The buds (woodcut, fig. 28) are
ovoid conic, dull brown in colour, and invested by ovate-lanceolate
scales. The leaves are
flattish, straight or late-
rally curved, between half
and three quarters of an
inch in length, linear,.
oblong obtuse, convex,
and shining green on the
dorsal surface, slightly
flattened near the point ;
the ventral surface is
marked by a prominent
midrib, on eitker side of
which are lines of sto-
mata. There is no indi- 5
cation of any twisting of &
the leaf; hence the sto-
mata are really on the ;
upper surface ofthe blades, Pi» Omorika.—Plan of section of leaf, much
as in some J unipers and magnified: bud X3; seed x2; leaves, front
other plants. A trans- and back, x2.
verse section, when examined under the mieroscope, shows a layer
of hypoderm-cells beneath the epidermis, and investing the ground
tissue of the leaf, which consists of spheroidal cells with undu-
lating boundary-walls and with a very slight tendency to form
palissade-cells on the lower side, that furthest removed from the
stomata. Close beneath the hypoderm on the lower surface are
two resin-canals, one on each side midway between the centre
and the margins; each is surrounded by strengthening-cells.

Fig. 98.

TO THE HISTORY OF CERTAIN CONIFERS. 205

The central bundle of fibro-vascular tissue is surrounded by a
well-marked bundle-sheath, and has the xylem above and the
phloem beneath in the usual manner.

Picea orientalis, which has been compared with it, has shorter
leaves, more acutely pointed and more four-sided in section.
The cells also which surround the central bundle-sheath are in
P. orientalis of an oblong form, and radiate from the central
bundle on all sides. Their walls are undulated. Hence, then,
the superficial appearance, no less than the internal structure of
the leaves, is quite distinct in the two species; a comparative
examination of which, therefore, confirms Dr. Panécie's view, that
he had to do with a previously undeseribed species, and not with
a form of P. orientalis. Moreover, the cones of P. orientalis are
longer and the scales broader than in P. Omorika.

ul

Picea orientalis. —Leaves and plan of leaf-section.

-

Pinus Peuke, Griseb.

Pinus Peuke, Grisebach, Spicil. Flor. Rumel. i. p. 239;
Christ, Europ. Abiet. p. 9; Endlicher, Conif. Synops. p. 144;
J. D. Hooker in Journ. Linn. Soc. 1865, p. 146, et in Gard.
Chron. Feb. 24, 1883, p. 244; Boissier, Flora Orientalis, vol. v.
(1884) p. 698; Engelmann in Trans. Acad. St. Louis, iv. 1880,
p. 170, Revision, p. 18.

Pinus excelsa, Parlatore in DC. Prod. xvi. ii. p. 405, ex parte,
haud Wallich.

P. Cembra, var. fruticosa, Grisebach, Reise in Rumelien, p. 189,
teste Parlatore, l. c.

“ In [Macedonia] Seardo australi in monte Peristeri prope
Bitoliam, alt. 2400-5000 ped.," Grisebach, Heldreich (ex Par-
latore). In monte Perimdagh supra Nevrekop Janka, in monte
Rilo Thracie, Pančicii, teste Boissier, l. c.

For my knowledge of this Pine I am entirely indebted to Sir

Y. n
) .
7 ^ 7 Y 7 A ZEA BX ZE ^
4M EN g Yy Gym LZ Hy z—— "
np ffpyy
m E,
p, Z2

n E
// ;

] ij | = Sa.
~ a \\ )
NW | W y | NN ' Wig
Tj WA M

E
d
o
Š
À
g
S,
À

CONTRIBUTIONS TO THE HISTORY OF CERTAIN CONIFERS. 207

Joseph Hooker; and I can add but little to what he has pub-
lished on the subject; but as the tree is still comparatively little
known, it may be well to repeat the illustration already published
in the ‘ Gardeners’ Chronicle,’ and add a few details taken from
the cultivated tree.

The chief interest attaching to the tree lies in its close resem-
blance to the Himalayan Pinus excelsa, a resemblance so close
that at first it was considered merely as a geographical variety of
that species. On further acquaintance, however, the differences
between the two trees become more obvious; and, although the
resemblances are sufficiently great to justify the hypothesis that
the two may have had a common origin, the divergences warrant
the keeping of the two forms distinct for practical purposes.

Boissier thus sums up the differences :— Affinis P. excelse a
qua tamen differt foliis brevioribus in ramulo toto persistentibus
(nec prope apicem ramulorum confertis); strobilis multo brevius
peduneulatis abbreviatis 3-43 pollices nec 6-7 pollices, longis ;
squamis superne sulcatis ; seminis ala longiore."

The trees at Kew, from which tbe illustration was taken, were
derived from seed collected in Grisebach’s original locality by
Professor Orphanides, and sown at Kew in 1864.

The bark of the trunk is smooth smoky brown, that of the
branches olive-coloured. The herbaceous shoots are glaucous
green, destitute of leaves near the base. The buds, like those of
P. excelsa, are elongate-ovoid, conic, covered by numerous lan-
ceolate acuminate brown scales somewhat reflexed at the tips,
The leaves are similar to those of P. excelsa, but shorter. Their

Fig. 31.

Pinus Peuke, —Portion of leaf, magnified. Plan of leaf-section, magnified.
Imperfect seed, real size. Cultivated specimen.

208 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

anatomical conformation is essentially the same (woodcut, fig. 31).
There is a layer of hypoderm beneath the epidermis, which en-
circles the homogeneous parenchyma of the leaf. The bundle-
sheath is well marked. The resin-canals are placed one on each
side at the dorsal side of the leaf near the epidermis, and sur-
rounded by strengthening-cells. Engelmann attributes peripheral
ducts to this species ; but I find them just within the epidermis.
The historian of the genus Pinus also says there is little hypo-
derm, and none round the ducts. It is possible that there may
be some variation in these particulars, or Engelmann may not
really have examined the true P. Peuke. That there is some
confusion is shown by the fact that in his tabular statement he
includes Peuke among species with few strengthening-cells and
none round the ducts; while in the more detailed notes he
mentions the hypoderm as forming a continuous layer. The
male catkins are in clusters of a yellow colour, each oblong-
obtuse (anther-crest roundish, pale violet). The stalk of the
cone and the cone itself, as shown in the illustration (fig. 30,
p. 206), are much shorter than in P. excelsa.

Endlicher, after alluding to other points of distinction from
P. excelsa, goes on to describe the venation of the wing of the
seed as parallel, while in P. ewcelsa it is “tortuose venoso-
reticulata."

Pseudolarix Kempferi, Gordon. (Plates IX. & X.)

Pseudolarix Kempferi, Gordon, Pinetum, p. 292, 1858 ; Henk.
et Hochst.; Carriére.

Larix conifera, Kempfer, ex Parlatore.

Abies Kempferi, Lindley in Penny Cyclopedia, p. 34, et in
Gard. Chron. 1854, p. 255, f. 455 ; A. Murray, Pines and Firs of
Japan, 1863, p. 100, c. ic. xylog.

Larix Kempferi, Carrière in Flore des Serres, xi. p. 97.

Pinus Kempferi, Lambert, Pinus, ed. 3, p. vii, ex Parlatore in
DC. Prod. xvi. ii. p. 413 (1868).

In montibus Chine orientali-borealis provincie Che Kiang ad
3000 ped. circ. elev. Fortune!

The so-called Golden Larch was in the first instance botani-
cally described by Lindley in the ‘ Gardeners’ Chronicle,’ 1854,
pp. 255 and 455, with a figure of cone (woodcut, fig. 32), under
the name of Abies Kempferi, under the impression that it was
one of the plants mentioned by Kempfer in Japan. This, how-

TO THE HISTORY OF CERTAIN CONIFERS. 209

ever, appears to be a mistake. The tree is not Japanese so far as
known, but is a native of the central, northern, and eastern pro-
vinces of China. It was introduced to this country by Fortune,
who called it a Larch (L«riz), although the scales of the cone,
instead of being persistent, “are so deciduous, that it is scarcely
possible to hold them together."

Fig. 32.

Cone of Pseudolarix Kempferi, real size. Native.

Fortune could not assent to its being called an Abies unless
Cedars and Larches are also referred to the same genus. “It
is," he says, *apparently a plant exactly intermediate between
the Cedar and the Larch; it has deciduous cone-scales, like the
Cedar, and deciduous leaves like the Larch, and a habit some-
what of the one and somewhat of the other" (* Gardeners’ Chro-
nicle, 1855, p. 643). Gordon, recognizing its distinctness from
Abies, Cedar, or Larch, called it in his ‘ Pinetum’ Pseudolarix
Kempferi, thus putting it into a new genus, to which objection
was raised at the time on the score that the proposed name was
an “ill phrase—a vile phrase” !

A. Murray preferred to keep to Lindley's nomenclature, and
LINN. JOURN.—BOTANY, VOL. XXII. s

210 DR. MAXWELL T. MASTERS'S CONTRIBUTIONS

was of opinion that it was nearer to the Silver Firs than to the
Larches. Carrière called it Larix Kempferi. Lambert and
Parlatore grouped it under Pinus.

It is not necessary to cite other authors, who have only copied
what their predecessors have said; but it may be stated that
Mr. Bentham, in the ‘Genera Plantarum,’ ii. p. 442 (1880),
after pointing out its peculiarities, concluded that the tree
must either be affiliated to the Cedars or considered as a distinct
genus; the male flowers, however, were unknown to him.

This deficiency I have the satisfaction, thanks to the courtesy
of Signor Rovelli, of removing. One of the most remarkable
trees in the nursery of Signor Rovelli, at Pallanza, Italy, is a
tree of this species, which in 1884 put forth male catkins. These
male catkins settle the point. It is clearly not a Pinus (except
in the very widest and practically most inconvenient sense). lt
is not an Abies (Silver Fir), nor a Picea (Spruce), nor a Cedar,
nor a Larch; but is the representative, as Gordon said, of a
distinct genus. In spite of “his ill-phrase," we consider our-
selves, therefore, bound to adopt his designation. The male
catkins are different from those of any Conifer known to us. To
some extent they remind us of those of the Ginkgo, which are
stalked, grow in tufts in the same manner, but which are mixed
with the leaves. It is needless to say there is little or no
other resemblance to the Ginkgo. The catkins resemble those
of Cunninghamia sinensis; but are pendulous, not erect, and are
arranged in umbels at the apex of short spurs invested by brown
membranous scales (Plate X.). Each catkin is stalked, the stalk
being recurved, so that the eatkin itself is pendulous, oblong,
about three quarters of an inch long, apple-green in colour.
The lower anthers are subglobose, almost sessile, the upper ones
are prolonged into a long appendage, sometimes lanceolate, some-
times, as shown in the drawing, 3-lobed at the apex. The pollen-
grains, drawn by Mr. W. G. Smith, have each two bladders, as
in the Pines &c.

The tassel-like catkins, not fully developed in our specimen,
must clearly be very ornamental, and are quite different from
the solitary sessile catkin of the Larch. The adult foliage and
cones from Signor Rovelli’s plant are shown on Plate IX.

Fortune first met with it as a dwarfed pot-plant, telling us that
when at a height of some 18 inches the Chinese contrive to make
it assume the guise of an aged Cedar of Lebanon. Ultimately

TO THE HISTORY OF CERTAIN CONIFERS. 211

he found the tree growing in the vicinity of a Buddhist monas-
tery near Chekiang. The trunk of one specimen had a girth of
5 feet and a height of 120 or 130 feet. At Quanting, 20 miles
away from the first locality, Fortune came upon a sloping hill
* eovered with the beautiful objeet of our search." The trees
he notes all appeared to have been planted. “One tree in par-
ticular seemed the Queen of the Forest, the trunk measuring
8 feet in circumference and 130 feet in height, with the lower
branches nearly touching the ground." The circumstance that
the trees appear to have been planted is of importance with re-
ference to the origin of the tree. Two explanations have been
offered of the fact that curious trees, such, for instance, as the
Abies Fortunei, are found in the vicinity of temples and monas-
teries in Japan, China, Sikkim, Tibet, and other Buddhist
countries. One explanation is that the trees have been brought
from elsewhere by the Buddhists—but if so, whence? The
other is the view taken by Dr. Hance (whose authority on such a
point is very great), that the trees form the remnants of the indi-
genous forest preserved around the temples, from feelings of
veneration, but destroyed for timber or by fire in other parts of
the country. -

In the * Journal of Botany, 1875, p. 138, as has been kindly
pointed out to me by Mr. F. B. Forbes, Dr. Hance alludes to
the “Gold Pine" (King-sung or King-ts-ien-sung), found by
the Rev. G. E. Moule in the hills west of Ningpo. This Dr.
Hance refers to Larix davurica, but, as Mr. Forbes suggests, it
is probable that Pseudolarix Kempferi is really the tree in-
tended. Mr. Moule says that the tree is not so spiral in figure
as the European Larch, but more like a Cedar, except that the
branches do not stratify.

Bertrand (Ann. Se. Nat. t. xx. 5 sér. p. 90) includes this plant
under Larix, of whose leaf-structure he gives a figure and de-
scription. He distinguishes it from the other species of Larix
(except L. europea) by the absence of stomata on the upper
surface, while, according to him, there is hypoderm and four
rows of stomata on either side of the midrib in the European
species, as contrasted with the absence of hypoderm and the
presence of nine rows of stomata in the Chinese species. The
only leaf I have yet examined of this species presented a dif-
ferent structure from that indicated by Bertrand, one in many
points unlike that of the Larch. The leaf-section, is transversely

219 CONTRIBUTIONS TO THE HISTORY OF CERTAIN CONIFERS.

oblong, the midrib prominent on both surfaces, especially below ;
the epidermal cells are papilliform, and project like those of the
petals of a flower. There is no hypoderm; but beneath the
upper epidermis are two rows of cuboidal cells, or imperfect
palissade-cells full of chlorophyll. The centre of the leaf is occu-
pied by transfusion-tissue, and the single central fibro-vascular
bundle is surrounded by a ring of very large colourless cells
constituting the bundle-sheath. The resin-canals are three in
number—one on the upper surface under the projecting midrib,
a most unusual position, and two others on the under surface of
the leaf, one on each side above the epidermis, and near to the
edge. All three are surrounded by strengthening-cells.

DESCRIPTION OF THE PLATES.

Prats II.
Abies amabilis, Forbes. Sterile branch, nat. size.

Pare III.
Abies grandis, Lindi. British-grown cones, nat. size.

Puate IV.
Abies nobilis, Lindl. British-grown cone, nat. size, kindly lent by Messrs.
Veitch. a. Plan of section of leaf, showing grooved upper surface;
AA. Resin-canals. B. Seeds, scales, and bracts of cultivated specimen.

PrarE V.
Abies nobilis, var. magnifica, hort. Cones of home-grown specimen.

Puate VI.
Abies religiosa, Schlecht. Cone and foliage. a. Bract, two views. B. Scale.
c. Leaf and cross section.

Puare VII.

Cephalotaxus pedunculata, var. spheralis. Fruit and foliage. A. Longitudina
section of fruit.

Puare VIII.
Picea Omorika, Pantic. Cones and foliage.

Puare IX.
Pseudolarix Kempferi, Gord. Cones and foliage, nat. size, a. Leaf and cross
section. B. Scale and bract.

PLATE X.
Pseudolarix Kempferi,Gord. Branch with fruit. a. Male catkins. B, C, D, E.
Sectional outline of stamens. F& a. Pollen.

DR. E. BONAVIA ON CULTIVATED TRUE LIMES. 218

On the probable Wild Souree of the whole Group of Cultivated
True Limes (Citrus acida, Roxb., C. medica, var. acida of
Brandis, Hooker, and Aiph. de Cando!lle). By E. BoxYvia,
M.D., Brigade-Surgeon, Indian Med. Department. (Com-
municated by W. T. Tuisgyro Drže, F.R.S., OMG.)

[Read 4th February, 1886.]

Dn. Bnasprs, in his ‘ Indian Forest Flora, considers the small
sour Lime as a variety directly descended from the Citron,
Citrus medica, Linn. Sir J. D. Hooker, in his * Flora of British
India, follows Dr. Brandis; and Alph. de Candolle, in his
‘Origin of Cultivated Plants,’ follows them both.

It always struck me that the characters of the Lime-tree were
very different from those of the Lemon and Citron trees; and 1
could not therefore persuade myself to accept the relationship
given by Dr. Brandis as a finally settled point.

Some time ago I took up the study of the cultivated forms of
Citrus of India with a commercial object, that is, with the view
of ascertaining how this useful genus might be turned to some
better account in India and Ceylon than it has hitherto been.
I collected specimens from all parts, and observed leaves and
fruit; and in lately travelling from the North-West Provinces of
India to Ceylon by the Western coast, I had opportunities of
seeing Limes and their leaves. I have always been struck with the
persistence of the winged petiole in the Lime (C. acida, Roxb.),
while the Lemon and Citron rarely had it. One cannot help being
struck also by the aspect of the Lime-tree, so different from the
Lemon and Citron trees.

In Ceylon, one day, I asked Dr. Trimen to show me what he
had in the way of Citrus-trees in the Royal Botanical Garden
of Peradeniya. While going over the ground, his foreman
placed in my hands a Citrus-fruit, which I had never seen before,
and which he said he had obtained from an adjoining Cocoa
estate. This Citrus in Ceylon is called Lima, or Kàdalu dehi
(Leech Lime). It flashed across my mind that 1 held in my
hand the wild form of the ancestor of all the cultivated true
Limes. Dr. Trimen and myself then visited the tree. The
immense winged petiole of the “ Lima” leaf rather staggered me.
The large and small spines and the blade of the leaf were like

LINN. JOURN.—BOTANY, VOL. XXII. T

214 DR. E. BONAVIA ON THE WILD SOURCE

those of the cultivated form of Lime. I collected some small!
leaves of this interesting plant and compared them with the
well-developed leaves of an adjoining cultivated Lime-tree, and
found that the undeveloped petiole of the “ Lima” corresponded
to the fully-developed petiole of the Lime. The lime-leaves are
thinner, and so is the lime-skin. Thin leaf and thin skin
often go together. The blossoms of this Lima were small, like
those of the Lime, with either four or five petals, tinged purple
(at the Hakgala Gardens, Newera Eliya, I gathered off a Lime-tree
blossoms also tinged purple); some of the Zima flowers were
male, others hermaphrodite. I afterwards learnt that this Lima
is never eaten ; that its fruit is used boiled, as an insecticide, for
washing the head ; that it is used by estate coolies, and natives,
smeared over their feet, as an antidote against land leeches; that
it is often cultivated near villages for these purposes, and is also
sometimes found in jungles. This Lima appeared to me to have
the characters of a wild species ; its fruit is warty, thick-skinned,
and the rind has an aromatie, resinous scent, and a bitter acrid
flavour; the pulp is sour, and of an unpleasant flavour. A riper
specimen had an aroma sui generis in its greenish pulp, not
unlike that of the Indian Lime. The subjoined woodcuts give
outline drawings of this Lima and its leaves.

Drawings accompanying the MS. give an idea of its warty ex-
terior, one figure distinctly showing a mamilla. The girth of the
two specimens is given. All the drawings are natural size *. The
section is of A, and shows the thickness of the skin, and the
numerous seeds. The woodcuts here given, figs. 1-8, show the
leaves of this Lima of Ceylon. They are dark green, thick and
leathery, with their edges slightly turned backwards. Fig. 1 is

* The illustrations referred to were :—Lima of Ceylon, or Küdalu-Déhi (Leech
Lime); Limo tuberosus martinicus, Rumph. Flor, Amboy. vol. ii. tab. 26.
fig. 1, or Citrus Hystrix, DC., Kurz. Flor. Br. Burma.

Fig. A. Exterior very warty; the warts being studded with pinholes or
openings of the oil-cells. Pale yellow when ripe. The rind has an aromatic
resinous scent, and a bitter acrid flavour. Essential oil of a gamboge-yellow,
and stains paper permanently yellow.

Fig. B. Trans. sect. of A. Thick-skinned ; oil-cells numerous, and crowded
together ; pulp greenish; and juice-vesicles very small. ‘The pulp is sour, and
of a very unpleasant flavour. Seeds greenish when cut. Many-seeded.

Fig. C. A third, riper specimen of this Lima had greenish pulp and very
sour, with a peculiar aroma, not unlike that of the cultivated Lime. The
rind had a very unpleasant acrid taste and resinous scent,

OF CULTIVATED TRUE LIMES. 21?

Occasional form of Lima-leaf.

Fig. 3.

A. Well-developed typical form of
Lima-leaf of Ceylon; thick, leathery,
dark-green edges, slightly turned back-
wards. B. Spine.

A common form of the Indian
cultivated Lime.

of a fully developed leaf, probably after the rains ; fig. 2, a small

and differently shaped, probably a spring leaf. I have shown

these, in order to compare their petioles with that of a fully-
T2

216 DR. E. BONAVIA ON THE WILD SOURCE

developed cultivated Lime-leaf, fig. 3. The petioles are then
identical in form. The spines of this wild form and the adjoining
cultivated form are identical. The form of the blade of leaf 1
corresponds to that of fig. 3; a very common form in all the
eultivated Limes I have seen is rather rounded and broad in
proportion to its length. In other respects this Zima has all the
characters of a Citrus.

By the courtesy of Dr. Trimen, I had the opportunity of looking
over Rumphius's ‘ Herbarium Amboinense. I found that this
Ceylon Zima had all the characters of his * Limo tuberosus mar-
tinicus, vol. ii tab. 26. fig. 1. Making allowances for Rum-
phius’s artist’s inaccuracies, I consider these two identical. The
flowers are pictured with 4 petals, and he says “it is used for
washing the head.”

Vol. ii. tab. 27 of Rumphius gives Limo agrestis (in Dutch,
Wilde Pap-Lemoen). This is evidently a near relative to Limo
tuberosus, Mart. It has retained the broad winged petiole, and
has lost some of its warts. It is pictured with 4 petals; is of a
yellow colour, and its pulp sour and acrid.

Another variety closely connected with Limo tuberosus is that
shown in Rumphius, vol. ii. tab. 28. It is also pictured with 4
petals and has broad winged petioles. I do not know the name
given to this by Rumphius, as the copy of the Peradeniya Library
has pp. 105 and 106 missing. [Limo Ferus, Lemon Swangi,
Wilde Lemoen-Boom, Latine Limo Ferus; juxta Malaicense
Limoen Swangi, ob irregularem et deformem ejus figuram:—
Rumph. Herb. Amb..ii. 106.]

In Rumphius, vol. ii. tab. 26. fig. 3, is given a smooth form of his
Limo tuberosus, with its identical leaf and 4-petalled flowers, Why
Rumphius calls it “ ramum Limonis feri" I do not know. Of this
ovoid smooth form there are many cultivated Limes in India.
Some trees produce nothing but this ovoid smooth form.

Rumphius's Limonellus aurarius, vol. ii. tab. 30, pictured with 4,
and also 5 petals, is, he says, the smallest of all Limes, not used
for food, but by goldsmiths for cleaning gold. Observe the
large winged petiole he gives to this small form. Iu India
there is a small smooth form, with the smaller petiole of the
Indian varieties, which is also used by goldsmiths for a similar
purpose. The leaves of this Limonellus aurarius are pictured
entire. Probably this is an oversight ; I never saw a Citrus-leaf

OF CULTIVATED TRUE LIMES. 211

of any sort entire, as a typical leaf. Sometimes the small spring
leaves appear entire on a superficial inspection.

In Rumphius, vol. ii. tab. 29. fig. B, we have for the first time
the Lime, pictured with 4 petals and a small winged petiole, like
all the varieties we have in India, like the cultivated Limes of
Ceylon, and the Lime of Montserrat, shown in tab. 6745 of the
* Botanical Magazine.’

Nota bene.—Loureiro gives Citrus limonum, ‘ Flor. Cochin
Chin., as a variety of Citrus medica, and refers the reader to
Rumph. Flor. Amboyn. vol. ii. tab. 29. Unfortunately this
plate gives two different things—Fig. A, which is a Lemon, with
unwinged petiole, and probably a descendant of C. medica; and
Fig. B, a true Lime, like the Indian cultivated varieties, and
pictured with 4 petals, and which is probably a descendant of Limo
tuberosus of Rumphius. (Can this plate have given rise to the
notion that C. acida of Roxb., the Lime, is closely connected with
C. medica, the Lemon? Rumphius says that on rare occasions
this form is produced (op. cit. ii. tab. 29. fiz. A) (though probably
not on the same tree) Its spines and leaves are those of
C. medica, Rumph., vol. ii. tab. 25, and, curiously enough, he also
gives a smooth form in this latter plate.)

After examining Rumphius’s plates and descriptions, I turned
to Kurz's ‘Flora of Br. Burma,’ and under Citrus Hystrix, DC.
(Shouk-pote), I found the same Limo tuberosus of Rumph. and
Lima of Ceylon in a wild state. If I were writing a description
of the Lima of Ceylon I would copy Kurz’s of C. Hystriz—It is
“armed with longer or shorter straight axillary spines, all parts
glabrous ; leaves oval or ovate, 15-2, rarely 3 in. long, petiole 1-13,
often 2-3 in. long, leaf-like, expanded, and often larger than the
blade itself, obversely cordate or obovate-oblong; at base con-
tracted in a simple petiole ; flowers 4 or 5-merous, small, white [in
the Ceylon Lima I found them tinged purple—E. B. |, on very short
glabrous pedicels....; ovary obovate, terminated by a very short
thiek style; berries obovate, or irregularly globose, the size of a
citroon or larger [a vague size; I have seen Citrons from 3 to
10 inches long, E. B.], very uneven and warty, almost juiceless,
with a thiek yellow skin. Not unfrequent in tropical forests of the
Martaban hills. . ..; also in Upper Tenasserim. Often cultivated
in native gardens."— Kurz, Flor. Brit. Burma, 195-7.

Dr. Trimen also thinks that the Lima of Ceylon and the
C. Hystrix of Kurz are the same thing.

218 DR. E. BONAVIA ON CULTIVATED TRUE LIMES.

l have not had an opportunity of seeing a larger and riper
specimen ; but in my mind I have little doubt that the Lima of
Ceylon and Rumphius’s Limo tuberosus are direct descendants of
Kurz’s wild Citrus Hystrix, and that from these semi-cultivated
descendants all our varieties of Limes have come. It is not
improbable that with the Dutch and their Malay servants came
to Ceylon the Lima as a hair-wash as well as the smooth Lime.
Afterwards, the Cinghalese, finding the Lima a good autidote
against land leeches, gave it the additional name of Küdalu dèhi
(déhi is the name of their cultivated edible Lime, and Kudalu
means leech).

This semi-wild Lima I have never seen or heard of in India,
but I think its derivatives are the cultivated forms of Lime,
such as the small goldsmiths’ Lime, Limonellus aurarius and
Limo Ferus of Rumphius, and other similar common forms of
Lime. In any place in India, if you find no other Citrus, you
are sure to find the small thin-skinned Lime, the Kaghzi nimboo
(Kaghzi means “of paper thickness” with reference to the skin),
and it is always raised from seed.

It is not unlikely that the true Limes of India were also
introduced by the Dutch. The * Jambiri” of India, which is a
Sanskrit name, is I think a totally different thing, and owes its
origin to another source, although now the two names “Jambiri”
and “Kaghzi nimboo" are often mixed up.

Conclusion.—In my opinion the wild Citrus Hystrix, DC., is
the ancestor of Limo tuberosus, Rumph., and Lima of Ceylon,
and all their derivatives— Limo agrestis, Limo Ferus, Limonellus
aurarius, and those figured in tab. 28 & 29, all of Rumphius;
and also the ancestor, more distantly, of all our other culti-
vated true Limes of India, Ceylon, and other parts. In my
opinion, the reason why the Lime has so persistently a winged
petiole, is the immense winged petiole of its progenitor, Citrus
Hystrix, it being too prominent a feature to be obliterated by
cultivation. If this be not conceded, at all events it will be
allowed that the true Lime, the C. acida of Roxb., is a nearer
relative of C. Hystrix than of C. medica.

MR. J. STARKIE GARDNER ON SOME FOSSIL LEAVES. 219

Remarks on some Fossil Leaves from the Isle of Mull (Scotland).
By J. STARKIE disp. F.L.S., F.G.S.

[Read 21st January, 1886.]

Tue series of plants is not very extensive, but many of the speci-
mens are in remarkable preservation, and being in a hard matrix
are quite beyond the risk of injury.

The flora comprises but one Fern, and that is indistinguishable
from the living Onoclea sensibilis of Western America and Eastern
Asia. It is represented by both fertile and barren fronds, and
though very limited in range, is extraordinarily abundant in in-
dividuals. With itis an Equisetum ; but the most careful search
failed to bring any other Cryptogamous plants to light.

The Conifere are better represented. A Ginkgo is indi-
stinguishable from the existing species, and is more abundant and
luxuriant in this locality than in any other. In those remote
times it seems to have formed a circumpolar belt such as Firs and
Larches do at the present day, and must have been well-nigh
universal in northern regions. Geological changes, such as
the intrusion of sea and desert, changes of climate, and the
competition of stronger forms, have been peculiarly fatal to it.
The type of Podocarpus represented at Mull has a wider dis-
tribution at the present day. The Taxus was formerly supposed
to be a Sequoia, but it does not agree in any of its minuter
characters with the surviving Redwood, the foliage of which also
breaks up in quite a different way. There are, moreover, no
Sequoia-cones associated with it; but it does, on the other hand,
agree in every respect with an existing Yew of Japan. The
determination of the Cryptomeria rests on its identity with a
species found in the somewhat newer beds of Antrim, with
which Cryptomeria-cones are abundantly associated. All these
I have described in the volume of the Paleontographical Society
which is on the eve of publication.

Monocotyledons are only represented by one sword-shaped
leaf with rectangular netted venation.

I have not yet made any attempt to determine the Dicotyledons,
except by comparing them with other fossils.

Three very well known Arctic Tertiary species are the most
abundant, and these have been named Platanus, Corylus, and
Grewia. For the present I can only regard them as of geo-

220 MR. J. STARKIE GARDNER ON SOME FOSSIL

logieal, and not botanical value. "With regard to the supposed
Platanus, it is quite unlike the real Platanus in form, texture,
and venation, and perhaps more like some Cretaceous leaves
known as Credneria and Protophyllum. There is no reason to
doubt, however, that true Platanus was already in existence at
this period, as it has been found in our lower Eocenes in England,
and in still older rocks in America. A curious burr-like fruit,
perhaps belonging to one or other of these species, is not
uncommon. There are about twenty species of Dicotyledons in
all, but I believe that further search will greatly increase the
number.

The differences between this flora and those of the County
Antrim, which are included in the same Basaltic flows, though
on a higher level, are very remarkable. There must have been
a considerable lapse of time between them, but the difference is
too great to be altogether accounted for in this way. Forests
disappear at times with great rapidity, and the arrival of new and
vigorous species will sometimes exterminate the indigenous trees.
There is the well-known instance in Denmark of the Beech
killing off the Birch, the Oak, and the Fir. A change in the flora
does not therefore necessarily imply a fresh geological period.

There is another and perhaps more ingenious way of account-
ing for the change. The enormous flows of basalt must have
been continually substituting a Laterite formation with a rich
and ferruginous soil for the previous gneiss and limestone, and
this we may assume would lead to a considerable change in the
character of the forests *. A flora may be produced in a sheltered
spot of quite different aspect from the average flora of a locality.
Thus no one wandering over the Isle of Mull would suppose
that m Bute Island, in almost the same latitude, there are

* As an example of this I may quote from Mr. Kurz, in his preliminary
Report on the Forest and other Vegetation of Pegu. ‘No other formation
except metamorphic and volcanic ones, can boast of such a variety of species, in
spite of its apparent sterility, as Laterite. It is this rock that affects vegetation
so much that the great difference between the floras of Malacca, Borneo,
Sumatra, &e., on the one hand, and that of Java, on the other side, is produced.
It is also this formation which allows so many Australian genera, like Mela-
leuca, Backea, Tristania, Leucopogon, &e., to spread so far to the north-
west, some of which, like Tristania, spread as far north as the Ava frontier.
If all the Laterite plants were to be erased from a list of the plants of Pegu
proper, the flora would be rendered very uninteresting indeed.”

LEAVES FROM THE ISLE OF MULL. 2»

luxuriant woods in which the Arbutus flourishes and ripens its
fruit.

The Mull flora is interesting as supplying fresh confirmation
to the view, first propounded by Asa Gray, that formerly the
entire northern temperate regions possessed a very uniform flora.
It was not previously demonstrated that the Greenland and
North-American Tertiary flora extended to Europe in Eocene
times. Its final disappearance from Europe was doubtless
due to the causes enumerated by Asa Gray during a Glacial
period; but there must also have been numberless minor migra-
tions and changes that will be found to have given local colour-
ing to fossil floras, as these are successively brought to light.
The flora will acquire a far higher interest when the plants
have been studied. The difficulties in determining them are
well known, but I trust that care and perseverance will prove
them not to be insurmountable. Work in our own country
wil at best but afford us glimpses of the past. The data
in regard to the former inhabitants of the land are incom-
plete and must ever remain so, but we must not undervalue them
on that account. If we really wish to investigate the origin of
the existing vegetation of the north temperate world, we shall
have to examine the enormous series of plant-deposits in Spitz-
bergen, whieh attain a thickness of some 3000 feet, and were
apparently being deposited continuously throughout that critical
period when existing dicotyledonous genera were taking shape.
If a small vessel could be despatched on an expedition to Spitz-
bergen and Greenland, the results of a single summer's collecting
would, I believe, far surpass our highest anticipations.

I desire to express in' conclusion my great acknowledgments
to the Duke of Argyll, on whose property the “Leaf Beds" are
situated, for assistance rendered, both personally and through
his agents; nor must it be forgotten that we might still have
been in ignorance of their existence had he not brought them

into notice.

222 MR. J. 6. BAKER ON A COLLECTION

On a Collection of Ferns made in North Bornoó by the Bishop
of Singapore and Sarawak. By J. G. Barer, F.R.S., F.L.S.

[Read 4th March, 1886.]
(Prarzs XI. & XII.)

Dr. G. F. Hose, Bishop of Singapore and Sarawak, has just sent
home a beautifully dried and carefully annotated series of Ferns
from the north of Borneo, collected principally in the neighbour-
hood of Sarawak. The following is a complete enumeration of
the species, with deseriptions of the novelties. Besides scattered
notices, three papers have been published on the Ferns of the
island, viz. :—1. Cesati's memoir in vol. vii. of the * Atti della R. Ac-
cademia delle Scienze Fisiche e Matematiche di Napoli,’ founded
mainly on the material collected by Dr. Beccari; 2, my own
report in vol. viii. p. 37 (1870) of the ‘Journal of Botany,’ on
the collection of Mr. F. W. Burbidge ; and 3, Burck’s paper in
vol. iv. of the ‘Annals of the Botanic Garden of Buitenzorg,’ p. 88
(1884). Of the numbers used, those given for every species are
Dr. Hose’s collecting-numbers ; and those given for the novelties
within brackets indicate the position they occupy in the series
followed in our ‘Synopsis Filicum.’
2. GLEICHENIA FLAGELLARIS, Spreng.

l. G. vestita, Blume. Matang, 3000 feet. Seems conspe-
cific with G. hirta, Blume.

4. CxaTHEA Bnuxowis, Wall. Caudex 2 feet long.

5. HEMITELIA ALTERNANS, Hook. As Dr. Hose points out,
and as his specimens clearly show, this has a rudimentary invo-
lucre. I believe now that Cyathea Lobbiana, Hook., is the same
species.

6. ALSOPHILA LATEBROSA, Hook. Caudex 8-10 feet long.
Stipe unarmed.

7. A. BURBIDGEI, Baker.

9. A. RAMISPINA, Hook. Matang, 3000 feet. Only gathered
before by Lobb long ago.

10. A. conraminans, Wall. Caudex 12 feet or more long.
Stipe muricated.

14. HyuENOPHYLLUM DILATATUM, Sw. Matang, 2000 feet.
15. H. Ners, Hook.

OF FERNS FROM NORTH BORNEO. 223

22. H. JAVANICUM, Spreng.

17. TRICHOMANES SAXIFRAGOIDES, Presl. Cannot, I think, be
separated specifically from T. parvulum.

16. (61*) T. Hoser, Baker, n. sp.; rhizomate filiformi late
repente, stipite brevi nudo haud alato, frondibus parvis ovato-
rhomboideis tripinnatis, pinnis inferioribus maximis deltoideis
basi postice cuneato-truncatis, segmentis ultimis linearibus uni-
nerviis integris ascendentibus, soris terminalibus solitariis,
involucro infundibulari angusto alato labiis 2 parvis semiorbicu-
laribus przdito, receptaculo breviter exserto. (Plate XII.)

A small finely-cut species allied to the Philippine 7' brevipes,
Baker, and T. Smithii, Hook., and Polynesian T. tenue, Brock.
Fronds about 2 inches long, 3 inch stipe included, erect, spaced
out upon the long filiform rhizome. Final one-nerved entire
segments about 3 of a line broad, not more than j-i in. long.
Involuere d in. long, narrowly winged on both sides nearly or

quite to the top of the tube.

25. T. MAXIMUM, Blume.
23. T. FÆNICULACEUM, Bory. Matang.

12. (18*) Dicxsonta (§ PATANIA) AMPIA, Baker, n. sp. ; rhi-
zomate repente paleaceo, frondibus amplis deltoideis glabris tri-
pinnatis, pinnis oblongo-lanceolatis, pinnulis lanceolatis multi-
jugis acuminatis ad costam vel alam angustam pinnatipartitis,
segmentis ultimis adnatis oblongis parallelis contiguis, venis
copiose pinnatis venulis 6—8-jugis erecto-patentibus, soris parvis
globosis marginalibus incurvatis, involucri valva exteriori majori
glabra semiorbiculari.

A very distinct species, allied to the Tropical-American D.
adiantoides, H. B.K. Rhizome epigeous, creeping, as thick as a
man’s finger, clothed with dark palee. Fronds 4 feet or more
long, moderately firm in texture. Rhachis brown, unarmed,
pubescent only down the channelled face. Lower pinnæ the
largest, 2-3 feet long, nearly a foot bread. Tertiary segments
Lin. broad. Sori placed on the edge of the tertiary segments,
principally near the base. Veins 7-8 on a side in the tertiary
segments, the lower forked, the upper simple.

11. (23*) D. (§ PATANIA) GoMPHOPHYLLA, Baker, n. sp.;

frondibus amplis glabris deltoideis decompositis, rachibus in-
ermibus, pinnis infimis maximis deltoideis inzquilateralibus,

224. MR. J. d. BAKER ON A COLLECTION

pinnulis oblongo-lanceolatis, segmentis tertiariis multijugis ob-
longis, quaternariis multijugis subsessilibus ineequaliter cuneatis
parvis obtusis profunde pinnatifidis, venis subflabellatis, soris

parvis globosis incurvatis, involucri valva exteriori majori semi-
orbiculari.

Matang, Sarawak, 2000 feet.

A very large decompound species with the habit of D. moluc-
cana, Blume, and D. scandens, Blume, but the rhachises unarmed.
Primary rhachis half an inch thick at the base of the lamina.
Lower pinn: 14-2 feet long, a foot broad. Quaternary segments
not more than $ inch long, parallel with the rhachis on the upper
side at the base, cut away on the lower side. Sori small, placed at
the base of the sinuses of the quaternary segments.

28. DAVALLIA PARALLELA, Wall.
29. D. ALPINA, Blume.

35. D. PINNATA, Cav., var. LUZONICA, Hook. Matang, Sarawak,
3000 feet.

32. D. sorida, Smith.
39. LINDSAYA CULTRATA, Sw.
40. L. CULTRATA, var. LoBBraANA, Hook.

41. L. scaNpENs, Hook. Known previously only as a simply

pinnate iype; but Dr. Hose's specimens are bipinnate, with a
habit like L. trapeziformis.

46. L. tanuernosa, Wall.

47. L. DIVERGENS, Wall.

Əl. ADIANTUM CaPILLUs-vENERIS, L. Kudat.
53. Preris LONGIFOLIA, L.

51. P. enstrormis, Burm.

56. P. qUADRIAURITA, Retz.

e P QUADRIAURITA, var. DIGITATA, Baker.
99. P. incisa, Thunb., var. AURITA, Blume.

61. BLECHNUM ORIENTALE, Linn. The type and a copiously
bipinnate form, gathered also by Burbidge.

62. B. Frntaysonranum, Wall,

64. ASPLENIUM MACROPHYLLUM, Sw. A very dwarf slightly
compound form,

66. A. HIRTUM, Kaulf.

OF FERNS FROM NORTH BORNEO. 225

67. AsPLENIUM NITIDUM, Sw. “Stipe and rhachis always
black in Borneo."

68. A. sUBAQUATILE, Cesati. Discovered by Beccari, and also
collected by Burbidge.

69. A. PORPHYRORACHIS, Baker. Sarawak and Perak.

72. (205*) A. ($ DIPLAZIUM) JEQUIBASALE, Baker, n. Sp. 5
caudice erecto, stipitibus cespitosis basi paleaceis, frondibus
oblongis glabris simpliciter pinnatis, pinnis 9-11 adnatis lanceo-
latis acuminatis crenulatis ad basin integram sequilateraliter
attenuatis, venis perspicuis erecto-patentibus simplicibus vel fur-
catis, soris medialibus parallelis, involucro angusto glabro per-
sistente.

River-banks at Sarawak.

Allied to A. pallidum, Blume, and A. porrectum, Wall. Tex-
ture rather thin. Both surfaces bright green, and free from
hairs and palew. Stipe 3-4 in. long, with a-few black lanceolate
spreading palee near the base. Lamina 6-8 in. long, 4-5 in.
broad. Pinns alternate, distant, 4 inch broad at the middle.
Sori just falling short of the midrib and margin.

76. A. POLYPODIOIDES, Mett.
74,75,77,78. Forms of the very variable ASPLENIUM MAXIMUM,
Don.

79, 80. A. CORDIFOLIUM, Mett. Type and a form connecting
it with var. integrifolium.

81, 82, 88. A. LINEOLATUM, Mett.

84. DIDYMOCHLÆNA POLYCARPA, Baker.

85. ASPIDIUM SEMICORDATUM, Sw.

87, 88. A. ARISTATUM, Sw.

89. A. PLATANIFOLIUM, Mett.

90. NEPHRODIUM IMMERSUM, Hook.

91. N. caLcARATUM, Hook., var. FALCILOBUM, Hook.

94. N. cRASSIFOLIUM, Hook.

95. (131*) N. ($ LASTREA) SARAWAKENSE, Baker, n. sp. ; cau-
dice erecto, stipitibus ewspitosis semipedalibus ad apicem paleis
setosis squarrosis preditis, frondibus submembranaceis ovato-

lanceolatis subpedalibus tripinnatis dorso ad costas parce palea-
ceis, pinnis infimis maximis inequilateraliter deltoideis postice

226 MR. J. G. BAKER ON A COLLECTION

produetis breviter petiolatis, pinnulis multijugis oblongo-lanceo-
latis infimis solum pinnatis, segmentis tertiariis adnatis contiguis
integris, venis pinnatis, venulis paucijugis erecto-patentibus
infimis furcatis, soris parvis medialibus, indusio parvo deciduo.
Sarawak.
Allied to the South-Indian N. scabrosum and pulvinuliferum.
Stipes and main rhachis clothed throughout with small squarrose
rigid dark-brown pales. Frond thin in texture, green on both
sides, obscurely pubescent on the rhachises and main veins, about
a foot long, 33-4 inches broad at the base. Only the lowest
pinne opposite and deltoid, the others alternate, sessile, and
oblong-lanceolate, with divisions of pinnules not reaching down
far. Largest tertiary segments } inch long, 1 inch broad, with
about 4 veins on each side.

86. (131*) Nepnroprum ($LasrREA) ACICULATUM, Baker,
n. 8p. ; stipitibus elongatis ubique eum rhachibus paleis brunneis
subulatis patulis vel squarrosis dense vestitis, frondibus magnis
oblongo-deltoideis firmulis glabris tripinnatis, pinnis infimis
maximis inequilateraliter deltoideis postice productis, pinnulis
multijugis lanceolatis profunde pinnatifidis vel basi pinnatis,
segmentis tertiariis oblongis obtusis integris contiguis, venis
pinnatis, venulis 6-8-jugis erecto-patentibus immersis obscuris,
soris parvis subcostularibus, indusio parvo membranaceo deciduo.

Matang, Sarawak, 2000 feet.

Allied to the Indian N. scabrosum and intermedium. Stipes
1-1} foot long, densely beset throughout, like the rhachises, with
squarrose or spreading subulate brown palee. Lamina mode-
rately firm in texture, about 2 feet long by a foot broad. Lower
pinne much the largest, and the pinnules of the lower side much
the longest. Ultimate segments t inch broad, reaching 3-1 inch
long. Sori many to a final segment, placed near the midrib on the
base of the obscure veinlets.

97. N. sETIGERUM, Zaker.

98. (139*) N. (§ Lasrrra) MULTISETUM, Baker, n. sp. ; stipi-
tibus ezxspitosis elongatis pallidis cum rhachibus ubique paleis
subulatis rigidulis castaneis patulis vel Squarrosis dense vestitis,
frondibus magnis deltoideis decompositis dorso setis hyalinis
sparsis przditis, pinnis oblongo-lanceolatis pedalibus vel sesqui-
pedalibus, pinnulis multijugis lanceolatis copiose pinnatis, seg-

OF FERNS FHOM NORTH BORNEO. 227

mentis tertiariis adnatis lineari-oblongis profunde pinnatifidis,
segmentis ultimis parvis deltoideis, soris minutis medialibus,
indusio membranaceo deciduo.

Matang, Sarawak, 2000 feet.

Habit of Nephrodium setigerum and Polypodium ornatum, Wall.,
but the stipes and rhachises everywhere densely beset with
spreading or squarrose bristly pales. Frond 4 feet long. Pinnæ
patent, the lower 13 foot long, 4-6 inches broad. Pinnules
patent, 4-2 inch broad, the largest with 12-15 adnate distinct
deeply pinnatifid segments on a side below the pinnatifid apex.
Tertiary segments yy inch broad. Veins forked in the largest -
final segments. Sori minute, not more than 1-2 in a final
segment.

96. (141*) NEPHRODIUM ($ Lastrea) MEGAPHYLLUM, Baker,
n. sp. ; caudice decumbente, stipitibus elongatis basi paleis cas-
taneis dense vestitis, frondibus magnis deltoideis glabris tripin-
natis, pinnis infimis maximis inequilateraliter oblongo-lanceolatis
postice productis, pinnulis oblongo-lanceolatis acutis basi pin-
natis, segmentis tertiariis oblongis obtusis infimis leviter pinnati-
fidis, venis in segmentis quaternariis pinnatis venulis paucijugis
simplicibus erecto-patentibus, soris parvis medialibus, indusio
membranaeco deciduo.

Sarawak.

A huge species, most resembling in habit the glabrous forms
of the Tropical-American N. villosum. Lower pinne 13-2 feet
long, 7-8 inches broad. Largest tertiary segments 3—} inch
long by half as broad, crenate or shallowly pinnatifid. Veins
pinnate opposite the quaternary lobes, with not more than two
veinlets ona side. Sori 10-12 to a pinnule, one to cacb final lobe.

105, 120, 176. N. Levzeanum, Hook.

100. N. procurrens, Baker.

104. N. rRUNCATUM, Presl.

103. N. FErox, Moore.

107. N. (§ SAGENIA) rERNATUM, Baker. The type and a

5-lobed form.

110, (219*) N. (S SAGENIA) STENOPHYLLUM, Baker, n. sp. ; stipi-
tibus gracilibus cespitosis nudis castaneis, frondibus parvis
deltoideis glabris bis tripartitis, segmentis linearibus integris
aeuminatis, venis in areolas hexagonas copiose anastomosantibus,

228 MR. J. G. BAKER ON A COLLECTION

soris in segmentis subbiserialibus, indusio magno glabro per-
sistente. (Plate XI.)

River-banks, Sarawak.

A remarkable plant, like nothing else that is known except
N. nudum, Baker in Journ. Bot. 1879, p. 41, one of Burbidge’s
discoveries. Stipes 4—5 inches long. Frond 4-5 inches long
and broad, consisting of three subequal thirds, each made up of
three linear segments, which are each 2-3 inches long, j inch
broad at the middle, and taper gradually to the base and acumi-
nate apex. Veins all faint and indistinct. Soriina single irre-
gular row between the midrib and edge of the segments.

111. NEPHRODIUM ACUTA, Presl.

113. OLEANDRA NERIIFOLIA, Cav.
112. O. NERIIFOLIA, var. PHYLLARTHRON, Kunze

119. POLYPODIUM prrFORME, Blume.

115. P. MINIMUM, Baker. Matang, Sarawak. One of Bur-
bidge’s discoveries.

114. P. FLABELLINERVIUM, Baker. Lingga mountain, Sarawak,
4000 feet. A very distinct species, gathered before only by
Dr. Beccari.

116. P. ALTERNIDENS, Cesati.

117. P. nEPANDULUM, Mett. Matang, Sarawak, 2000 feet.
124. P. NUMMULARIFOLIUM, Mett.

126. P. sonipENs, Hook. Matang, 3000 feet.

125. (811*) P. ($ PuywATODES) SARAWAKENSE, Baker, n. Sp. ;
rhizomate gracili longe repente ramoso calvato, frondibus parvis
simplicibus subcoriaceis glabris oblongo-lanceolatis acuminatis
basi cuneatis, venis omnibus subtilibus immersis occultis in
areolas hexagonas anastomosantibus, soris magnis sparsis glo-
bosis leviter immersis.

Matang, Sarawak. Gathered bu by Lobb, but never
described.

Allied to P. superficiale, Blume, and P. normale var. sumatra-
num, Baker. Rhizome, in Dr. Hose's specimen, nearly a foot
long, less than a line in diameter, bearing 4 distant erect fronds,
on stipes 12-3 inches long. Frond 4-6 inches long, 11-12 inch
broad at the middle, narrowed gradually to the base and acumi-

OF FERNS FROM NORTH BORNEO. 229

nate apex. Sori ij inch diam., scattered irregularly all over the
under surface of the frond except the apex.

127. (331*) POLYPODIUM (§ PLEURIDIUM) CAMPYLONEUROIDES,
- Baker, n.sp.; rbizomate gracili, epigeo late repente paleis parvis
lanceolatis adpressis vestito, stipitibus brevibus erectis, frondibus
simplicibus membranaceis glabris parvis oblanceolatis obtusis e
medio ad basin sensim attenuatis, venis primariis subtilibus
erecto-patentibus parallelis, intermediis transversalibus areolas
subquadratas formantibus, soris 8—4 inter costam et marginem
dispositis inter venas primarias uniseriatis.

Matang, Sarawak, 2000 feet.

Allied to P. Melleri and the New-Caledonian P. selligueoides.
Rhizome cylindrical, dark green, about 4L; inch diameter. Stipes
erect, naked, 3-14 inch long. Frond 4-6 inches long, à-i inch
broad, narrowed gradually from the middle to the base. Main
veins fine and indistinct, the cross ones almost as regular as in
the American Campyloneurons.

130. P. RUPESTRE, Blume. I do not think this and P. trique-
trum can safely be kept up as distinct species.

129. (334*) P. (§ PLEURIDIUM) LEUCOPHORUM, Baker, n. sp.;
rhizomate gracili glauco, epigzo longe repente paleis filiformibus
brunneis subpatulis deciduis instructo, stipitibus brevissimis
erectis, frondibus simplicibus rigide coriaceis oblongo-lanceolatis
acuminatis facie glabris lucidulis viridibus punctis albis cretaceis
minutis preditis, margine recurvato crebre cretaceo-punctato,
dorso obscure viridibus impunctatis tenuiter pilosis, venis pri-
mariis subtilibus rectis erecto-patentibus parallelis, venulis inter-
mediis irregulariter anastomosantibus, soris minutis superficialibus
inter venas primarias irregulariter biseriatis.

Matang, Sarawak.

A very distinct and well-marked species, remarkable for the
white cretaceous dots scattered over the upper surface and form-
ing a close row on the incurved margin of the frond. Khizome
creeping to a length of a foot or more, the shining filiform pales
4 inch long. Stipes not more than 2-7 inch long. Frond 6-9
inches long, 1-13 inch broad at the middle.

. 131. P. IRIOIDES, Lam.

132. P. MUSÆFOLIUM, Blume.
LINN. JOURN.—BOTANY, VOL. XXII. U

230 MR. J. G. BAKER ON A COLLECTION

135. Potypopium INCURVATUM, Blume. Matang, 3000 feet.

137. P. DILATATUM, Wall., var. GRANDIDENTATUM, Cesati.
Differs from the type by its repand pinne; and may be a distinct
species.

144. GYMNOGRAMME QUINATA, Hook.

142. G. LOBBIANA, Hook.

141. G. BORNEENSIS, Hook.

140. G. BORNEENSIS, var. MAJOR, Baker.

148. MzNisCIUM TRIPHYLLUM, Sw.

160. (4*) M. Host, Baker, n. sp.; rhizomate breviter re-
pente, stipitibus contiguis gracilibus stramineis cum rhachibus
obscure pubescentibus sterilibus brevibus fertilibus elongatis,
frondibus oblongo-lanceolatis glabris simpliciter pinnatis, pinnis
lateralibus 7-8-jugis lanceolatis sessilibus basi auriculatis, venulis
2-3-jugis valde ascendentibus, soris demum confluentibus.

River-banks, Sarawak.

Closely allied tothe Ceylonese M. Thwaitesii. Stipe of sterile
fronds 2-6 inches long, of the fertile much longer. Fronds
about i foot long, 21-3 inches broad. Pinnæ i-i inch broad,
entire or obscurely crenulate, auricled on one or both sides at
the truncate base. Pinne of the fertile frond at most about an
inch long, + inch broad.

158. VITTARIA SULCATA, Kuhn.
150. ANTROPHYUM PARVULUM, Blume.
157. DrymoeLossum PILOSELLOIDES, Presl.

158. (71*) AcnosTICHUM (§ POLYBOTRYA) STENOSEMIOIDES,
Baker, n. sp.; stipitibus elongatis castaneis deorsum paleis
paucis lanceolatis præditis, frondibus sterilibus oblongo-lanceo-
aatis bipinnatifidis submembranaceis glabris margine paleis paucis
hyalinis deciduis instructis, pinnis sessilibus lanceolatis multi-
jugis ad costam late alatam pinnatifidis, segmentis secundaris
oblongis obtusis, venis liberis in segmentis secundariis pinnatis,
venulis 6-8-jugis inferioribus furcatis superioribus simplicibus,
frondibus fertilibus consimilibus stipitibus longioribus pinnis
laxioribus minoribus.

Matang, Sarawak, 2000 feet.

A very distinct species, most like A. repandum, var. Quoyanum
in habit, but the veins entirely free. Stipe of the barren frond a

OF FERNS FROM NORTH BORNEO, 231

foot long, of the fertile nearly 2 feet. Barren frond 1-14 foot
long, 7-8 inches broad. Lower pinne 444 inches long, 1-1}
inch broad. Secondary segments 4— inch broad.

159. (108*) AcrosticHum ($ CHRYsoDIUM) MoDEsTUM, Baker,
n. sp.; rhizomate gracili late repente paleis parvis lanceolatis
apice predito, stipitibus gracillimis elongatis, frondibus sterilibus
simplicibus linearibus integris acuminatis glabris, venis omnibus
in areolis hexagonis anastomosantibus, frondibus fertilibus con-
similibus sed minoribus.

River-banks, Sarawak.

A small inconspicuous plant, most like the Ceylonese 4. Wallii,
Baker. Rhizome trailing to a length of half a foot, the fronds
spaced out upon it. Stipe of sterile frond 2-3 inches long
Lamina 4-6 inches long, 1-3 inch broad at the middle, tapering
gradually to the base and acuminate apex. Fertile frond 2-3
nches long, 4 inch broad.

162. (110*) A. (§ CHRYSODIUM) ANTROPHYOIDES, Baker, n. sp. ;
rhizomate breviter repente, frondibus sterilibus subcoriaceis
glabris sessilibus oblanceolatis acutis basi rotundatis, venis im-
mersis occultis omnibus in areolis parvis anastomosantibus, fron-
dibus fertilibus linearibus rigidis dimidio superiori soriferis deor-
sum ad alam angustam costularem nudam attenuatis.

Matang, Sarawak.

A very distinct species, with sterile fronds most resembling
those of Antrophyum subsessile. Barren fronds 6-8 inches long,
3-1 inch broad ; fertile 8-9 inches long, 3-3 inch broad.

161. A. QUERCIFOLIUM, Retz.

163. A. BicusPE, Hook., type and var. integrifolium. Sarawak,
mountains at 4000 feet. “Only a few of the barren fronds

forked.”
164. A. AUREUM, L.
165. A. sPICATUM, L.
171. LYGODIUM SCANDENS, Sw.
172. AwGIOPTERIS EVECTA, Hoffm.
173. OpHIoGLossUM RETICULATUM, L.
175. HELMINTHOSTACHYS ZEYLANICA, Hook.

v2

282 > MR. H. N. RIDLEY ON THE

DESCRIPTION OF THE PLATES.

Puare XI.
Nephrodium (§ Sagenia) stenophyllum, Baker, n. sp.— Fig. 1. Frond. Fig. 2.
Enlarged portion, showing sori.

Puar XII.

Trichomanes Hosei, Baker, n. sp.—Fig. 1. Frond. Fig. 2. Enlargement of
thece.

On the Freshwater Hydrocharidew of Africa and its Islands.
By H. N. Rri»1vf, M.A., F.L.S.

[Read 1st April, 1886.]
(Puarss XIII. & XIV.)

Or the eleven genera of Hydrocharidex inhabiting fresh water,
seven occur in Africa and the adjacent archipelago. The section
Hydrillez is represented by the genus Hydrilla and Lagarosiphon.
The former genus consists of a single polymorphic species, very
widely distributed and common throughout Tropical Asia, and
occurring also in Australasia; it is found in Africa, but is
apparently not so abundant. Lagarosiphon is a genus very cha-
racteristic of Africa, all the species known being natives of that
region or of Madagascar, with the single exception of L. Ros-
burghii, an Indian plant of very different habit, which has, how-
ever, beeu found in Socotra, but there it is possibly an intro-
duction. The species are all very closely allied, and not easy
to study from dry specimens, as the flowers are exceedingly deli-
cate and the male flowers are comparatively rarely met with. One
of the chief distinctions between Hydrilla and Lagarosiphon 18
that the leaves in the former are in whorls, in the latter scattered ;
in some species of Lagarosiphon, however, there is a great ten-
dency for the leaves to become opposite, especially towards the
apex of the stems. This is notably the case in L. rubella, mihi.
In the ‘Genera Plantarum’ it is stated that there are no
staminodia in the female flower; this is usually the case, but
they do occur in at least one species in the form of slender short
filaments.

The Vallisneri»e consist of three small genera, two of which
occur within the area. One, Vallisneria, comprises à single
species widely spread over the whole world. Bly«a contains the
common Indian plant, B. Roxburghii, Rich., which is found also

F
FRESHWATER HYDROCHARIDEÆ OF AFRICA. 233

in Madagascar, and an undescribed species, B. radicans, mihi,
from Angola. The Stratioteæ are well represented by a number
of species of Ottelia and Boottia.

Only two of these genera are common to both the Old and New
World, viz. Vallisneria and Ottelia, and of the latter but a single
species occurs in the New World, viz. O. brasiliensis in Brazil.
None of the genera are endemic to Africa, but all occur more or
less widely throughout Tropical Asia. Five of the African species
are natives also of Asia, viz. Hydrilla verticillata, Casp., Blyxa
Roxburghii, Rich., Lagarosiphon Roxburghii, Benth., Ottelia alis-
moides, Pers., Boottia cordata, Lindl. ; and it is very interesting
to note that all these species are, as far as at present known,
confined to Eastern Africa, the islands of Madagascar and Mau-
ritius. From this it may be conjectured that the Hydrocharideæ
pushed their way into Africa from the East, perhaps, for the
most part, after the separation of the continents of Africa and
America was complete. At the same time it is possible that some
of these plants may have been introduced into the country in rice-
cultivation.

HyDRILLER.

LAGAROSIPHON MUSCOIDES, Harvey, Hook. Journ. Bot. iv.
p. 230, t. 22.

Cape of Good Hope, Harvey 1187; Zeyher 1732!; Drege
2276 b c!

Var. MAJOR, n. var. Planta validula, habitu Hydrille verticil-
late. Folia crassa atro-viridia opaca, 2 unciam longa, ;1;—} lata,
costa obseura; dentes breves obtusiores basi dilatati circiter 70 in
utraque margine; margines pallide folii angustissime ; spatha
floris feminei ferme 3 unciam longa, dentibus paucis obtusis brevi-
bus ; spatha mascula $ unciam longa.

South Africa, Undezine River, Victoria, Cooper 17! Transvaal,
Hoggeveld, Bronkerspruit, no. 6559! and Trigardsfontein, 6678!
Dr. Rehmann. Herb. Brit. Mus.

This variety, though looking very distinct, seems to be only a
remarkably large and coarse form. It has the habit and texture
of Hydrilla. The leaves are so thick as to be almost opaque
under the microscope, instead of thin and transparent as in the
well-known form, so that the midrib is quite obscured. The teeth
also are shorter and blunter, slightly thickened at the base. The
spathes, although quite similar in form, are much larger.

284 MR. H. N. RIDLEY ON THE

L. conporanvs, Caspary in Pringsh. Jahrb. i. p. 503.—Arasch
Cool, Nubia, Kotschy no. 170! Herb. Brit. Mus.

L. Nyassa, n.sp. Planta tenuis bryoides ; caules tenuissimi ;
folia lineari-lanceolata acuta lete viridia, 1 em. longa, 3 mm. lata;
dentes in utraque margine 26-33, longiusculi subappressi; flores
feminei minutissimi; petala lanceolata obtusa; styli graciles ad
dimidium longitudinis in stigmatibus fissi ; ovarium subelongatum
conicum ; tubus perianthii unciam longus capillaceus; spatha
lanceolata edentula subintegra.

Lake Nyassa, Dr. Laws! Herb. Brit. Mus.

I have seen only female flowers of this species. It is a smaller
and more delicate plant than any of the other species, and the
lanceolate petals and deeply cleft style, with the simple entire
spathe, sufficiently distinguish it.

L. SrEUDNERI, Caspary in Schweinfurth’s Beiträge Flora
4Ethiop. p. 200, t. 4.— Wadela, East Abyssinia, Stewdner.

L. RUBELLA, n. sp. Caules 3-5 pedes longi, purpurascentes
graciles ; folia remotiuscula subopposita, linearia acuminata, 7 mm.
longa, 3 mm. lata, herbaceo-viridia ; dentes circiter 30 in utraque
margine, breves; internodi nonnunquam ] cm. longi: flores
feminei minuti rubelli, bractea pallida ad basin ovata obtusa
quam fructus brevior, dentibus 10 in utraque margine dissitis,
spatha integra edentula ; tubus perianthii perlongus, 3 cm., gracil-
limus; petala ovata rotundata obtusissima; styli crassiusculi,
breviter bifidi, trichomata stigmatica brevia; ovarium elongatum
ellipticum conicum; capsula 5 mm. longa; semina 3 elongata
fusiformia, 3 mm. longa: flores masculi, spatha brevis obconica
quadrata $ unciam longa; stamina plura.

Angola, Huilla, frequens in flumine Humpatensi, socialis cum
Utriculariis et Potamogetonis speciebus, Welwitsch no. 6472!
Herb. Brit. Mus.

This is a lax species, chiefly remarkable for the more distant
leaves and the purplish flowers. It, to some extent, resembles
L. Steudneri, Casp., but has fewer teeth to the leaves, and the
petals are broader and more obtuse.

L. SCHWEINFURTHII, Caspary, Bot. Zeit. 1870, p. 88.— Central
Africa, in Djur land, Schweinfurth 2452! ; Bongo land, Schwein-
furth 2158! Herb. Kew.

FRESHWATER HYDROCHARIDEZX OF AFRICA. 235

L. pENSUS, n. sp. Planta densa, elongata, habitus Nazadis in-
dice. Folia copiosa et dense congesta, atro-viridia, anguste
linearia acuta, $ unciam longa, $ mm. lata; dentes 45-50 in
utraque margine, longiusculi subbappressi, versus apicem folii
copiosiores; flores minuti; petala et sepala ovata obtusa;
stigmata subclavata erassiuscula breviter fissa; staminum rudi-
menta tria tenuia; ovarium obconicum; perianthii tubus gracilis
capillaceus, 4 unciam longus vel ultra; bractea lanceolata longe
acuminata abrupte obtusa, profunde fissa edentata; capsula
elliptica 4 mm. longa.

Madagascar, Imerina, Hildebrandt 3804! Herb. Brit. Mus.

A dense, compact, dark green plant, with very numerous leaves.
The female flowers are all that I have seen. The stigmata much
resemble those of L. Steudneri, Casp. The rudimentary anthers
are fairly large and conspicuous, and the perianth-lobes are ovate
and obtuse. The bract subtending the ovary is very deeply cleft,
the laciniæ being rather narrow and acuminate and longer than
the ovary.

L. MADAGASCARIENSIS, Caspary in Rel. Rutenb. ii. p. 252,
t.18.—Madagascar. Prope Marovay : flowers white ; May 1880—
Hildebrandt no. 3422! Herb. Brit. Mus., Kew.

Hildebrandt’s plant, which I take to be the species described
by Caspary, is a laxer, shorter-leaved plant than the preceding.
The leaves in the upper part of the stems are often no more than
5 mm. long and 4 broad, somewhat recurved, bright green when
moistened, and with a very distinct midrib. The leaf-buds
and the adjacent leaves are frequently purplish. The spathe of
the female flower did not appear to me to extend so far beyond
the ovary as is shown in Caspary’s figure, and the lacinie in this
plant were blunter. The fruit was not seen by Caspary. It is
oval elliptical in outline, 3 em. in length. The seeds are cylin-
drical, with an elliptical outline, beaked at each end, longitudinally
striate and dark brown; they are 1 mm. in length, and there
are 4 or 5 in the capsule.

The plant has more the appearance of a Hydrilla than L. densus,
on account of the shortness of the leaves, especially in the upper
parts, where, too, they are almost opposite. The length of the
internodes is often as much as 5 mm., which also adds to the
resemblance.

Another species was obtained in Madagascar by Hildebrandt,
Imerina, no. 3523, which is quite distinct from either of the other

236 MR. H, N. RIDLEY ON THE

Madagascar species, and more resembles L. muscoides, Harv. It
is remarkable for its very spinulose leaves, the teeth of which are
singularly long and numerous, and the stem is distinctly angled.
Unfortunately I can find neither fruit nor flowers upon the spe-
cimens, so cannot properly describe it.

L. Roxsvnanrr, Benth. in Benth. § Hook. f. Gen. Pl. iii. p. 451.
—Socotra, Boivin! in herb. Kew. A letter with this specimen,
from Dr. Ascherson, suggests that the plant may have been intro-
duced into this region with rice from India.

HYDRILLA vERTICILLATA, Rich. Mém. Inst. Par. 1811, p. 961,
t. 2.—Mauritius, Robillard! Herb. Brit. Mus. White Nile,
Grant! Herb. Kew.

VALLISNEBIA.

VALLISNERIA SPIRALIS, Linn. Sp. Pl. ed. i. 1015.—Deep water,
Sierra Leone, Don! ; Lagos, Barter !; White N ile, Schweinfurth! ;
Bahr-el-Arab, Gazelle River, Schweinfurth!; Lake Nyassa, Dr.
Laws!

Apparently common in Africa, except in the south. V. ethio-
pica, Fenzl, based on a plant collected by Kotschy in Senaar,
Nubia, no. 284, seems to be merely a remarkably stunted form
with dentate leaves, somewhat analogous to the Australian plant
called V. nana by Robert Brown. There is a very similar plant
in the British Museum, labelled “ Ind. Or., Soc. Un. Fratr.,” which
has also dentate edges to the leaves.

Bryxa Roxsvnamrr, Rich. Mém. Inst. Par. 1811, p. 63.—Mada-
gascar, Imerina, no. 3805! prope Marovay, no. 3443! Hilde-
brandt. Herb. Brit. Mus.

The Madagascar plant does not seem to differ in any material
points from the Indian form. It does not seem to have been met
with yet on the mainland of Africa.

Bryxa RADICANS,n. sp. (Plate XIV.) Caulis gracilis prostratus,
radicans, radicibus paucis longis ; folia graminea angusta linearia
obtusa ad bipedalia, 1 unciam lata, basi dilatata in vagina membra-
nacea, plura singula dissita in caule; seapus gracilis 9-uncialis.
Flores masculi: spatha biuncialis angustasubcylindrica, basi paullo
dilatata; tubus quam spatha haud longior; sepala tenuia ligulata
obtusa pallide virescentia, $ unciam longa ; petalal ongiora angus-
tiora $ unciam, alba; stamina sex, gracilia, linearia acuminata ;
pistillodia tria brevia, basi dilatata. Flores feminei: spatha sub-

FRESHWATER HYDROCHARIDEX OF AFRICA. 997

similis sed tubus perianthii longe exsertus, 21 uncias supra
spatham ; sepala viridia chartacea ligulata obtusa, $; unciam longa;
petala alba, angustissima linearia filiformia, $ unciam longa; styli
validuli complanati tres, petalis subequalibus; fructus carnosus
longe cylindraceus.

Huilla. Frequens et gregaria in rivulo Lopollo imprimis ad
Catarractam, rarius florens. Fructus immaturi eum floribus initio
Maii 1860, JVelwitsch. Herb. Brit. Mus.

This species is easily distinguished from B. Roxburghii by its
creeping stem, which throws up distant solitary leaves sheathing
at the base. At intervals the leaves appear to become crowded
and the flowers are then produced. The female scape is usually
shorter than the male. The spathes of each are similar and re-
semble in form those of the other species ; but in the female flower
the perianth has a long tube projecting for two inches beyond the
spathe, while the perianth-tube in the male is no longer than the
spathe. The sepals in the male flower are thin and transparent,
while in the female they are thicker and opaque. The petals in
both are thin and white, and in the female exceedingly narrow
and filiform. The stamens are similar to those of the other species,
only narrower and acuminate. The pollen-grains are globular
and covered with short processes.

STRATIOTEX.
OrTELIA.

O. ALISMOIDES, Pers. Syn. Pl. i. p. 400.—Nubia, Egypt, Sir
Gardner Wilkinson!, Delisle!; Uah-el-Behariah, Ascherson
( Gisterr. Bot. Zeit. 1876) ; El Aguri, Lybian Desert, Ascherson! ;
Damietta, Bromfield!; Cordofan, Kotschy !

This is the most widely spread species in the genus, occurring
also in India, Borneo ( Grabowsky), Timor (R. Brown), Philippines
(Cuming), Java (Horsfield), China, and Japan.

O. VESICULATA, n. sp. Acaulis; folia lanceolata longe petiolata,
lamina 3-4 uncias longa ł unciam lata, petioli 3-unciales ; scapi
3-15-unciales, compressiusculi,breviter bialati, paullo infraspatham
sensim incrassati, compresse tetragoni angulis lateralibus acute
alatis; spatha vesieulatim inflata compresse lato-ovoidea extus
purpureo-vittata, ore bidentato demum lacero, uneiam longa, 3
unciam lata (in sicca); sepala lanceolata subacuta $-uncialia;
petala lutea 4 unciam longa; stamina 6; styli 7 (fide Welw.);
ovarium incomplete 9-loculare.

238 MR. H. N. RIDLEY ON THE

Huilla. In rivulis lente fluentibus et stagnis limpidis prope
Lopollo et Mumpulla. Flores in Octobri et Decembri, etiam
Aprili Folia primum (in vivo) rigidula fragilia pellucida lete
viridia, Welwitsch no. 6497! Herb. Brit. Mus.

O. ULVÆFOLIA, Buchen.—Madagasear, Imerina, Hildebrandt,
3803 ! Hilsenberg! Lyall 149! Baron 900! Parker ; “ Tan-Tang-
hin-ranoo."

O. rnANTAGINEA, Welw. MSS. Caulis brevis crassus descendens,
radices multas emittens; folia copiosa ovata petiolata crispa, 4
uncias longa (petiolus 14 unciam), 24 lata; scapi 10-unciales flac-
cidi; spatha saccata, uncialis, alis brevissimis, dentibus circiter 10,
brevibus obtusis ; floris pedicellus e spatha non productus; sepala
lanceolata obtusa 3 unciam longa quam petala multo breviora, viri-
dia, marginibus scariosis ; petala 2 unciam longa, flava vel lutea.

Huilla, gregaria in stagnis juxta ripas rivorum prope Catumba.
Radicis fibri sub epiderme eneo-nitentes. No. 6469, Welwitsch!
Herb. Brit. Mus.

This plant has a distinct, though short, conical descending
stem, by which, together with its ovate crisped leaves and nu-
merous teeth to the spathe, it is easily distinguished.

O. LANCIFOLIA, Richard, Tentamen Flor. Abyss. ii. p. 280, t. 95.
—Ambaca in lacu ad sinistram fluvii Caringa. Flores et fructus
Junio. No. 6466, Welwitsch! Herb. Brit. Mus.

Var. FLUITANS, n. var. Folia elongata angustiora longissime
attenuata in petiolis, pedalia 14 unciam lata.

Pungo Andongo. Frequens cum Nympheis in stagnis editio-
rum rupium Presidii Pedrasongue ad jugum inferius. Flores
medio Februario, fructus Aprili. No. 6468, Welwitsch! Herb.
Brit. Mus.

I have little doubt as to this being the plant intended by
Richard ; and if so, it is a true Ottelia and not a Boottia at all, as
suggested in the ‘Genera Plantarum.’ It exactly resembles the
figure quoted, but I have not seen Abyssinian specimens. The
plants referred by Mr. Baker (Trans. Linn. Soc. xxix. p. 151) to
this species, which were collected by S peke and Grant and Schwein-
furth, seem to me to be aspecies of Boottia allied to B. abyssinica,
mihi. The flowers in Dr. Welwitsch’s specimens were deep yellow.
I suspect that the white-flowered plants obtained by Schweinfurth
at Monbuttu, no. 3938, and one of those collected by Speke and
Grant, are new species of this genus.

FRESHWATER HYDROCHARIDE® OY AFRICA. 239

BoorrT1.

B. cRAsSIFOLIA, n. sp. Rhizoma crassum 3 unciam longum,
oblique descendens radices multas emittens, squainis ovatis dense
tectum ; folia ligulata lorata obtusa, basi in petiolum attenuata,
carnoso-subcoriacea rigida, obscure virentia, 5—7-nervia, 19 uncias
longa, 2-3 uncias lata, marginibus crispulis; scapi 15-unciales ;
spathasubcylindrica tubulosa, 7-canaliculata, basi paullo ampliata,
subcompressa, utrinque alata sex-costata, dentibus brevibus obtu-
sissimis; sepala viridia, 2 unciam longa, 4 lata, lanceolata obtusa ;
petala flava.

Angola. In stagnis profundis et rivulis Mumpulla sparse, et
Lopollo. Flores et fructus Aprili, fructus etiam Octobri. Wel-
witsch, no. 6470! Herb. Brit. Mus.

The stout rhizome of this plant is covered with close-set
broad and blunt sheathing-leaves, and descends somewhat
obliquely into the mud; the leaves are very long and stiff, resem-
bling to some extent the fronds of a Laminaria. The species is
somewhat closely allied to B. scabra, Benth., but has not the
peculiar scabridity of that species.

B. scasra, Benth. et Hook. f. Gen. Pl. ii. p. 454.—Ottelia
scabra, Baker, Trans. Linn. Soc. xxix. 151.—Mogren el bahr,
Gazelle River, Schweinfurth 1121! Speke & Grant! Herb. Kew.

B. ABYSSINICA, n. sp. Rhizoma crassum, submersum, unciale,
radices copiosas longas emittens; folia plura tenuia lanceolata
petiolata obtusa elongata, 15 uncias longa, 13 lata; scapi crassi
debiles 9-uneiales ; spatha oblonga elliptica, apice vix attenuata,
2 uncias longa, circiter 3 lata; flores in spatha duo; sepala
lanceolata obtusa coriacea, 3 unciam longa, j lata; petala flavida
visa haud multo sepala superantia.

Abyssinia. Zana prope Carrata, Dec. 1863. Schimper, iter re-
centius, no. 1452! Herb. Brit. Mus.

The long, narrow, thin leaves, not unlike those of Ottelia ulve-
Jolia, and the straight elliptical spathe not tapering to the apex, but
almost of the same diameter throughout its length, sufficiently
distinguish this plant. The rhizome, too, is not covered with
large scale-leaves, as in B. crassifolia, but seems to resemble that
of O. plantaginea, and is apparently deeply seated in the mud in
its native haunts. The flowers are hardly open in the only
specimen known to me, and the neck of the perianth barely
projects out of the mouth of the spathe.

240 ON THE FRESHWATER HYDROCHARIDE OF AFRICA.

B. corpata, Zdl.—Madagascar, Forbes no.1! Herb. Brit.
Mus.

B. ExsERTA, n. sp. (Plate XIII.) Rhizoma subnullum ; folia
coriacea ovata elliptica obtusa 5-nervia longe petiolata, lamina 23-
3-5 uncias longa, 11-2 uncias lata, petiolus ad 9 uncias longus ;
scapus 10-uncialis vel brevior; spath sangustz 1-2 uncias longa,
subeylindriez vix dilatate apice trifide, latere uno profundiore
fissæ ; flores g circiter 6 in spatha, tubus longe exsertus ; sepala
tenuia quam petala multo breviora; petala flava versus apices albe-
scentia, ovata obtusa ; stamina flava; flos 9 singulus in spatha 2
uncias longa, tubus exsertus; sepala angusta coriacea viridia, acu-
minata, 2 unciam longa, obtusa.

Eastern Africa. River Shire, a few miles above Shamo, Aug.
1859; R. Zambesi, Shupanga, Jan. 5,1860, Sir John Kirk in
Herb. Kew! Madagascar, no. 2, Forbes! Herb. Brit. Mus.

This is a small species with hard leaves something like those
of Potamogeton natans, L., and rather small yellow flowers,
the tubes of which project for some distance beyond the spathe.
Both the specimens collected by Sir John Kirk and the
drawing made by him (for the use of which I have to express my
thanks to the authorities at Kew, and especially to Mr. Baker)
were in male flower; but in the British-Museum herbarium are
two specimens collected by Forbes in Madagasear, which have
both male and female flowers. The male flowers, besides being
more numerous in the spathe than the female, are larger, with a
more delicate tube further exserted from the mouth of the spathe,
and the sepals are quite thin and shorter and blunter. Two flowers
apparently are open at the same time. The petals are yellow,
deepest in the centre and passing into white at the edges. ‘The
female flower has a shorter and thicker tube, and larger thicker
sepals, very narrowly lanceolate acuminate but blunt. Ihave not
seen the petals.

DESCRIPTION OF THE PLATES.
Puare XIII.
Boottia exserta, Ridley, n. sp.
Fig. l. The entire male plant, nat. size, from an original drawing by Sir John
Kirk, F.L.S.
2, Spathe of male plant, laid open.
3, Spathe of female flower, laid open ; petals fallen away.
4, A stamen.
5. Rudimentary pistil of male flower.

MR. W. MITTEN ON NEW SPECIES OF METZGERIA. 241

PrarE XIV. :
Blyxa radicans, Ridley, n. sp.
. Female plant, of nat. size.
. Female flower, enlarged.
- Male spathe, enlarged.
Male flower, laid open and enlarged.
. À stamen, enlarged.
. Pollen-grains, enlarged.

Fig.

Ow oo bo

Some new Species of the Genus Metzgeria.
By W. MrrrÁs, A.L.S.

[Read 6th May, 1886.]

SrwcE the publication of the monograph of this genus by Dr.
Lindberg in 1877, when he distinctly separated ten species, and
for the time exhausted the subject, eight species have been added
by Dr. Spruce in his * Hepatic Amazonice et Andinz,’ seven of
these being from the regions he had visited and one from New
Zealand. All these species are, excepting the difference in rami-
fication and the degree of their pubescence, very uniform in their
appearance, and coincide very nearly with the common M, furcata,
which might thus well be considered the typical and representa-
tive species. The discovery, however, of a few stems recently
sent from New Zealand by Mr. Reader, has revealed a form which
could hardly have been anticipated, and thus leads to the conclu-
sion that other modifieations will in time be discovered, which
must change the idea of the genus, and may alter its place in
arrangements.

METZGERIA sACCATA, sp. n.; frons dichotome divisus, supra
convexulus, l»vis marginibus ubique lobis alternatim dispositis
saccatis patenti-ascendentibus compressulis ad ventrem frondis
incurvis.

Hab. New Zealand, on bark, with a Frullania and several
mosses, a very few stems only; Tasmania, forests between the
Gordon and Franklin rivers; M‘Quarie Harbour, 1846, Dr. Mil-
ligan, a fragment.

The specimens of this species are as yet too scanty to give an
idea of what may be its usual state, but they are enough to show
its complete departure, in the presence of the saccate lobules, from

242 MR. W. MITTEN ON NEW SPECIES OF METZGERIA.

allpreviously known. To the unassisted eye they scarce offer
more than small stems of M. furcata, with which they agree in
substance and colour; and itis only when their apparently sinuose
outline is detected that any difference is obvious. When, however,
the fronds are examined with sufficient power on the under or
ventral side, it is seen that they have really marginal inflated
lobules not very different from those usual in the genus Frullania.
This appearance is in part produced by the incurvation of the
inflated sac, so that its opening, which is towards the costa,

Fig. 1.

M. saccata.—A, natural size. B, much enlarged, showing the under or ventral
side. C, a sac bent back in plane with the frond, D, transverse section of
frond, to show the many-celled costa.

MR. W. MITTEN ON NEW SPECIES OF METZGERIA. 248

appears much less than it really is when the sac or lobule is bent
back in plane with the frond. All these appendages were empty.
The substance and areolation of the fronds offer nothing different
from that usual in M. furcata; the costa is, however, of a dif-
ferent structure, the cells being more numerous. As to the cilia,
it is believed several were seen on the first fragment examined,
but none have since been met with. Young male inflorescence,
of the form usual in the genus, is on the New-Zealand spe-
cimen.

METZGERIA SCOBINA, sp. n.; frons dichotome divisus, supra
convexulus, dorso cellulis omnibus elevationibus conicis asper ;
marginibus ciliis binatis ad latus ventrale anguste incurvis, costa
cellulis majoribus 2 dispositis.

Hab. Borneo: Sarawak, on Matang Mountain, Mr. Everett.

Fragments only of this have been found on mosses from trees.
To the unassisted eye it is altogether like JM. hamata, but on
inspeetion with the microscope, even in the dry state, the promi-
nent cells showing on the incurved margins at once prove to be
quite different, and on section all the cells are seen to be pro-
tuberaut.

Ne SSS
FEES

M. scobina.—Transverse section (enlarged), to show the pointed cells.

METZGERIA NITIDA, sp. n. ; frons dichotomus, costa e cellulis
2 dispositis; margine ciliis singulis binatisve paucis, cellulis
magnis pellucidis levibus iis M. furcate quadruplo majoribus.

Hab. Australia, Apollo Bay, Sir F. von Mueller; New Zea-
land, Rev. W. Colenso, on a specimen of Homalia pulchella,
a 279.

This appears almost exactly similar to M. furcata, but its
fronds are composed of cells so much longer and translucent
that they have a shining look.

244. MR. H. N. RIDLEY’S MONOGRAPH

A Monograph of the Genus Liparis.
By H. N. Roni MA TLS.

[Read 3rd June, 1886.]

ArTHOUGH the literature treating of Orchids is very extensive,
and is still very largely on the increase, the number of Mono-
graphs of genera is remarkably small, and, indeed, it seems tbat
there is no Order of equal size and importance now so meagrely
provided for in this respect. Species innumerable have been
deseribed as they were discovered or introduced into cultivation,
often several species of different genera together, and that with-
out correlation or comparison with other species. The result of
this is that the species of many genera are inextricably confused,
and it is by no means easy, even with a great deal of labour, to
collate all the described species of any large genus. It not un-
frequently happens also that plants are introduced, perhaps
almost accidentally, into cultivation, and shortly after dying out
are entirely lost, no specimens having been preserved, and only a
description, often, especially among the older writers, of an in-
sufficient character, by which the species can be identified. I
have therefore thought it may be useful to those interested in
this group of plants to collect what information I could glean
concerning the genus Liparis into a single paper. No list has
been published of the genus since the publication of the ‘ Genera
and Species of the Orchidaceous Plants,’ by Lindley in 1840, in
which thirty-three species are described, while the number of
species at present known is above 100. The collections to which
I have had access are those of the British Museum, of Kew,
including the Lindley herbarium, and the Wallichian herbarium
of the Linnean Society ; and I have here to thank Messrs. Fran-
chet, H. Bolus, F. Moore, and Baron von Mueller, for placing in
my hands specimens of rare and critical species, and Mr. A. Fryer,
of Chatteris, for notes and living specimens of our native species.

The genus was originally founded by L. C. Richard (Mém.
Mus. Par. iv. p. 52) for the species Zoeselii and liliifolia, formerly
included under Malaxis. Later Reichenbach (‘Conspectus,’ p. 69)
changed the name to Sturmia, on the ground that the name
Liparis was already in use for a genus of Moths. This objection
is no longer considered valid, and the older name is commonly
retained. Liparis, as it stands by itself, is a perfectly natural
genus, most nearly related to Microstylis, but distinguished,

OF THE GENUS LIPARIS. 245

usually easily, therefrom by its longer and more slender column
not embraced by the lip. To it the authors of the ‘ Genera
Plantarum’ have added Ephippianthus. This seems, however,
to be distinct both in the habit and form ofthelip. It is a mono-
typic genus found in the island of Sachalin, and is referred to
the neighbourhood of Malawis by Prof. Reichenbach. Gastro-
glottis, Blume, Bijdr. p. 397, is an unknown plant which, as the
authors of the * Genera Plantarum’ suggest, may be a species of
this genus. Platystylis was made by Blume as a section of
Malaxis, and included two species, P. decurrens and P. atro-
purpurea. Lindley raised the section to the rank of a genus, but
Bentham and Hooker (Gen. Plant. vol. iii. p. 496) refer it to Li-
paris on the strength of a plant collected by Lobb in Java, which
has been identified by Prof. Reichenbach as P. atropurpurea. I
have not seen the type of this species; but the description, so far
as it goes, tallies very fairly with Lobb's plant, which is certainly
a Liparis, and a very different plant from the P. decurrens, and
of which there is a specimen in the herbarium of the Natural-
History Museum labelled by Blume Apatales purpurascens, a
name I cannot find published anywhere. P. decurrens is a very
peculiar plant, differing from all other species of the genus m
possessing a short thick column with four wings, two on each side.
A specimen in the above-mentioned herbarium, exactly similar
to Blume's figure (Bijdr. t. 54) is labelied by Blume himself
Platystylis alata, a name which is, as far as I can find out, un-
published. This plant, standing alone, would certainly con-
stitute a distinct genus, but L. aurita, Ridl., and L. resupi-
nata, mihi, form connecting links between the two genera. The
former of these two species possesses the rudiments of the
second pair of column-wings in the form of two small ears, one
on either side of the column, in addition to the ordinary wings.
The flowers, too, in both are resupinate, one of the chief cha-
racters in the genus Platystylis. I therefore think it will be
preferable to regard the genus as a section only of Liparis.

The species of this genus are usually smali plants with incon-
spicuous flowers, occurring terrestrially in damp spots in bogs or
woods, or among rocks, or epiphytieally upon trees. The truly epi-
phytic species belong to the section Conrrrorrz, with distinct stout
pseudobulbs and thick or leathery leaves. These are confined
to the tropics of the Old World. Many of the soft-leaved species,
especially those with creeping stems, grow upon moss or rotten

LINN. JOURN.— BOTANY, VOL. XXII. x

246 MR. H. N. RIDLEY’S MONOGRAPH

tree-trunks in the damp forests. It is commonly stated that our
native species, L. Loeselii, Rich., is epiphytic upon Sphagnum ;
but Mr. A. Fryer, who has had opportunities of observing the
plant in the fens of Cambridgeshire, and who has kindly supplied
me with living plants, informs me that he has found it sometimes,
though rarely, growing directly upon clay. He suggests, in this
case, that the moss may have decayed away, and that the plant
manages to do without it. This may be the case, but it is not
very difficult to grow the plant in ordinary soil or leaf-mould,
provided it be rich enough for it ; and it cannot be said to be more
epiphytic upon the moss than Droseras or Pinguiculas, which are
so often seen in similar situations.

Distribution —The Liparides are one of the most widely-spread
genera of the Order, only being eclipsed by the much larger
ubiquitous genus Habenaria. Probably it owes much of this ex-
tensive distribution to its being equally adapted for temperate
and tropical regions. It occurs inall portions of the globe, with
the exception of the extreme north and south, and of Arabia,
Persia, and New Zealand. Z. Loeselii, Rich., is at the same
time the most northern species and the most widely distributed,
occurring in North America as far north as Canada, and in
Europe to Sweden; while the most southern species is pro-
bably Z. reflexa, R. Br., from Australia. L. angustifolia, Lindl., is
perhaps the most widely distributed of the tropical species, being
found in Madagascar, Ceylon, and Malaya. As a rule, however,
as is the case with most Orchids, the species are very local. The
largest number of species occurs in tropical Asia, which includes
nearly all the section CORIIFOLIE, besides a considerable number
of Mollifoliz.

The Stem.—In most species of the genus the rhizome is very
short, the pseudobulbs being thrown up close together & In
many of the epiphytie species the old pseudobulbs remain for
some time after they have become effete, as in Z. luteola, Lindl.,
where they are closely set on the rhizome, the terminal one only
being floriferous and foliiferous. In Z. Loeselii, and species of
similar habit, as a rule but two pseudobulbs are met with together,
one of the present, the other of the previous year. In many
species it is even unusual to find the pseudobulb of the previous
year remaining, and the rhizome also has almost entirely dis-
appeared. In Z. nepalensis, Lindl., the rhizome is extensively

* See Pfitzer, Grundzüge vergl. Morphol. Orch. pp. 149-152; Orépin, Notes
aur plantes rares Belgiques, fasc. v. p. 103.

OF THE GENUS LIPARIS. 247

developed, stout, woody, and persistent, the pseudobulbs being
often as much as 2 inches apart. In Z. repens the rhizome creeps
for a considerable distance; it is not woody, but slender and
herbaceous, throwing up numerous small pseudobulbs, each of
which bears a single leaf, from the axils of some of which flower-
spikes spring. From this form of rhizome we pass easily into
the creeping section of the genus (Ramose). These are South-
American plants, growing on damp moss or on wet trunks of
trees. The stem is covered with small petiolate leaves, without
any trace of a pseudobulb ; towards the extremity of the creeping
stem one or more scapes are thrown up singly from the axils of
the leaves. The most striking species of this section is L. brachy-
staliz, Reichb. f.

L. Welwitschit, Reichb. f., is another quite aberrant species, its
habit closely resembling that of Goodyera, having a creeping stem
sending up tufts of leaves, which are sometimes barren, and some-
times bear erect scapes.

The pseudobulb is naturally best developed in the epiphytic
species, and is sometimes of considerable size. It is, however,
somewhat variable even in the same species; thus in typical
L. spathulata it is conical, about 2 inches in length; but I have
seen plants, differing in no other respect from the typical plant,
with short, almost oval pseudobulbs, and again in the true
L. longipes, Lindl., of which L. spathulata is but a variety, the
pseudobulb is drawn out sometimes to a length of nearly 6 inches.
It is to be remembered also that, after flowering, the pseudobulb
not unfrequently lengthens and alters its form somewhat. In
many terrestrial species besides those mentioned above, there is no
true pseudobulb, the stem being elongate and narrow, covered
with leaf-sheaths, often loose, white, and membranous (L. punc-
ticulata, Ridl., L. atropurpurea, Lindl., etc.).

Leaves.—The leaves vary to some extent, and afford good
characters for breaking up the genus. In the greater number of
the truly epiphytic species they are hard and thick in texture,
and lanceolate or lanceolate spathulate in outline. They have no
distinct petiole, and break off abruptly when dead. These are the
characteristics of the section CoRIIFOLLE, which is quite confined
to the tropics of the Old World, excluding Africa. The remaining
species have thin leaves, with an outline varying from lanceolate
to ovate, or rarely orbicular, often petiolate, and spreading. Many
have crisped or waved margins to the leaves, and not unfrequently

x2

248 MR. H. N. RIDLEY'S MONOGRAPH

they are coloured or tinted with purple (L. Saundersiana, &c.).
As a rule, there are two or three leaves to each pseudobulb, more
rarely one or more than three.

The Scape rises from the apex of the pseudobulb. It is usually
in the Conirronrz angled and winged, sometimes flattened. The
base is, as a rule, nude except for a few scattered empty bracts,
the lowest of which is not unfrequently, especially in the Co-
RIIFOLIA, foliaceous. The inflorescence is a lax raceme, often
few-flowered; but in the Z.-longipes group the flowers are
numerous and densely crowded. The most striking form is that of
the section Dist1cH», where the raceme is shortened and flexuous,
and the bracts are arranged in a distichous manner, like those of
a Dendrocolla.

The bracts are lanceolate and green, more rarely ovate-lanceo-
late, usually shorter than the pedicellate ovary, and sometimes,
but seldom, reflexed after flowering.

The Flower.—ln the majority of species the flowers are rather
small and few in number. The largest are those of L. grandi-

lora, Ridl., the smallest those of L. longipes, Lindl., and its allies,
in which they are not only minute but very numerous. ‘The
texture is usually thin, especially in the Drsrrcx x ; but in L. elata
and many allied species the flowers, and more particularly the lip,
are somewhat fleshy. The colour is most commonly yellowish
green, but it varies from whitish to yellow (L. flavescens, &c.) and
purple (L. Sauadersiana). The sepals are usually narrow and lan-
ceolate, the two lateral ones being frequently falcate and placed
under the lip. In this case they are usually broader than the
upper one, and are sometimes connate to a greater or less degree.
This is not infrequent as an accident, and occurs normally in
L. disepala, Reichb. f. When the upper sepal is dissimilar to the
others, it usually resembles the petals, which are invariably linear
and narrow. The petals and sepals are constantly spreading
or reflexed ; but in L. viridiflora, Lindl., the flower seems never
to open entirely. The lip is most commonly entire, more rarely
distinctly 2- or 3-lobed. The base is usually narrowed, with the
edges raised and often thickened, so as to form a channel down
the centre. The claw stands parallel to the face of the column,
the lamina being bent abruptly downwards. In this case there
are usually two (rarely a single) conical calli at the base of the
claw, formed by the thickening of the lateral veins of the lip;
and all stages of the development of these protuberances are to

OF THE GENUS LIPARIS. 249

be met with, from the simple thickening and sometimes colouring
of the veins, as in L. elliptica, L. neuroglossa, &c., to the complete
distinct calli (L. elata, &e.). In L. aurita, Ridl., and its allies
the veins at the base of the lamina are pushed a little to the
side and thickened, so as to form a horseshoe-shaped callus with
a slight depression in the centre. In many species the lip has
no erect claw, but the whole of the lip lies parallel, or almost so,
to the column; and in Z. elliptica, Reichb. f., the sides of the lip
are so curved up as to enclose and conceal the column.

The column is usually long and slender and more or less curved,
with the base and apex dilated. The sides of the stigma are
flanked by two small membranous wings more or less triangular
in outline. They are rarely absent (L. W'elwitschii, Reichb. f.),
and in Z. decurrens (Platystylis, Blume) there are an extra pair
outside the normal pair, rudiments of whieh may be seen in
L. aurita, Ridl., a closely allied plant, in the form of a pair of
minute lobes. The species of this section are also remarkable
for their short, thick, straight column. The column of Z. dis-
ticha, Lindl., is similar in shape, though the plant is very dif-
ferent in other respects; but the two other species of the same
group (Distiche), L. Cuming?, Ridl., and L. compressa, Lindl.,
have the slender arcuate column of the rest of the genus.

The anther-cap is usually ovate and acute, sometimes, as in
L. ornithorrhynchos, Ridl., distinctly beaked. It is more or
less distinctly bilocular, and contains two pairs of ovoid pollinia
usually adhering together in pairs by their apices. The stigma
is usually rather small, elliptical or oval, rarely (LZ. Welwitschit)
deeply sunk.

The capsule is usually pyriform globose in the small epiphytic
species of the genus, such as L. cespitosa, Lindl., and L. longi-
pes, Lindl. In the terrestrial species it is commonly elongate
conical (L. Loeselii and L. elata). When ripe it has usually distinct
ribs, which are rarely sinuous (L. Forbesii, Ridl.). As the fruit
ripens the pedicels elongate a little, and in some of the epiphytic
species not unfrequently become recurved. This is especially
noticeable in L. flaccida, Reichb. f. In L. Loeselit, Rich., and
species of similar habit, the fruit remains stiffly erect, and as it
were adpressed to the rhachis. It is possible that this may bear
some relation to the seed-distribution, for whereas in the epiphytic
species it would be advantageous to the plant to spread its cap-
sules, that the wind might catch them more readily and drift

250 MR. H. N. RIDLEY'S MONOGRAPH

ihe seeds upon other and lower branches of the trees on which
the plants grow, in the case of the marsh-loving species, were the
seed-pods to be recurved, there would be considerable risk of the
seeds falling too close to the parent plant.

Fertilization.—1l cannot find that, with one exception, any
observations have been published upon the fertilization of any
species ofthe genus. Mrs. Barber (Journ. Linn. Soc., Bot. x. 1869,
p. 455) suggests that small Hymenoptera or Diptera might be the
impregnators of L. Bowkeri, Harv., which species she cultivated in
her garden in South Africa, but found them not fertilized there.
She remarks that she never found any honey in the blossoms ;
neither have I been able to detect any in any of the species
which I have been able to examine in a living state. The
flowers do not appear as a rule to have any scent; but L. longipes
is an exception, it has an unpleasant odour. Notwithstanding
this, they seem to be in some way attractive to insects, as I have
seen ants busily running about upon the lip of L. Saundersiana,
Reichb. f., as if there was something there worth seeking.

The insect attractions to the plant are ofthree kinds. In plants
of the L. longipes type there is a smooth polished band of a
darker green than the rest of the lip running for the whole
length of the labellum, very similar to that which occurs in the
lip of Listera ovata, R. Br. The lip in these species is so curved
that its highest part is immediately under the anther, so that an
insect alighting upon the apex of the lip and creeping up would
strike with its head the anther and receive the pollinia. In some
of the larger-flowered plants, in which this would appear to be
the means of fertilization, there is an ochreous or orange patch
in the centre of the lip (L. reflexa, R. Br., L. laciniata, Ridl.).
In another series of plants the veins at the base of the lip are
thickened, and sometimes (L. neuroglossa, Reichb. f.) brightly
coloured. These doubtless are attractive to insects; and once ina
flower of L. elliptica, Reichb. f., from Venezuela, I found the base
of the lip had been gnawed by some animal, and there were a
pair of pollinia lying in the stigma. This, however, may have
been accidentally done in pressing the flower.

A considerable number of species, again, have these veins raised
at the extreme base of the lip into two conical calli. In this case
the lip usually, as stated above, has a distinct channelled claw
and a broad deflexed lamina ; sometimes there is only one callus,
which is in some cases notched at the apex, from which it appears

OF THE GENUS LIPARIS. 251

probable that it was formed by the amalgamation of the two calli.
L. disticha, Lindl, is peculiar in having a patch of irregular
warty calli at the base of the lip.

In many of the calliferous species the insect is directed to the
callus by the recurved edges of the claw ; butin Z. Rheedii, Lindl.,
there are two guiding ridges, apparently formed by the pinching-
up of the lamina which run from the base of the lip, flanking the
calli, till they are gradually merged in the lamina.

In the two last series the point of attraction lies at the base of
the lip close to the column, while the pollinia are situate in the
first case over the middle ofthe flat lamina, and in the second over
the highest point of the lip, that is, where the lamina joins the
unguis and forms an angle with it; so that if the insect receives
the pollinia as it leaves the flower, as is usually the case, they
would in all probability be attached to its back. In the first
series, however (the LZ. longipes section), and in the case of L. au-
rita, Ridl., the point of attraction is not at the base of the lip but
at its highest point, 7. e. in the former at the top of the curve
formed by the lip, and in the latter at the junction of lamina
and unguis. The latter has been alluded to under the general
description of the lip; but I will proceed to give further details.
The veins which arise from the base of the lip run straight up
the claw till they reach the lamina, where they become sigmoid,
each forming a semicircle in their course, the ends of the semi-
circles being so closely approximated as almost to meet, especi-
ally at the upper end. This forms a horseshoe-shaped ridge with
a distinet depression in the centre at the highest point of the
lamina. The anther overhangs this ridge, so that in this plant, as
well as in the Z. longipes section, the head of the fertilizer is
naturally exactly under it, and in all probability the pollinia
would be attached to its head.

Divisions.—The genus is best divided up into ConrrrForrz, the
hard-leaved epiphytic section, and Morrrrornrz with soft leaves.
The names, which originated with Professor Reichenbach, are
preferable to Lindley's Liparis vera and Cestichis, the latter a
name invented by Thouars for some of the hard-leaved Mascarene
species.

The Morrrronrx have frequently no pseudobulbs, but a
straight, longer or shorter stem covered with sheathing-leaves,
often whitish and membranous. The leaves are thin and plicate,
often more than two in number, not rarely petiolate and oval, or

252 MR. H. N. RIDLEY'8 MONOGRAPH

ovate-lanceolate. They do not drop off abruptly, but wither on
the stem. Their distribution is that of the genus.

The ConirroLrz, including Thouars’s Distichis, have distinct
pseudobulbs, oval to conical in shape, the leaves stiff and
leathery, usually two in number, and lanceolate, breaking off
abruptly when dead from the top of the pseudobulb, and leaving
a scar, never withering on the pseudobulb. They are usually
epiphytic, and are confined to the tropics of the Old World.

It is true that these sections pass into each other, as Bentham
and Hooker (Gen. Plant. iii. p. 495) state, but they are certainly
the most satisfactory and natural.

Liraris, L. C. Richard,

Mém. Mus. Par. iv. p. 52; Benth. & Hook. f. Gen. Pl. iii. p. 495.
Sturmia, Reichenbach, Conspectus, p. 69; Iconogr. Bot. t. 956.
Alipsa, Hoffmannsegg, Linnea, xvi.

Malaxis, auctorum veterum.

Empusa, Lindley, Bot. Reg. sub 825.
Empusaria, Reichb. Consp. p. 69.
Platystylis, Lindley, Orch. Pl. p. 18.

Herb perennes, epiphytice vel terrestres, caulibus primariis
plus minusve repentibus, ramis sæpe pseudobulbosis, brevibus et
crassis, vel longioribus angustioribus, foliis vaginantibus sspe
albescentibus membranaceis tectis, arcte approximatis vel rarius
dissitis. Folia evoluta, 1-6, membranacea, ovata petiolata vel lan-
ceolata, aut carnosa vel coriacea, lanceolata spathulata, in apicem
pseudobulbi subrosulata vel raro dissita in rhizomate. Scapi
singuli ex apice pseudobulbi orientes, teretes, vel angulati et alati.
Racemus sepius laxus pauciflorus aut aliquando densus multi-
florus. — Bractes sepissime ovariis breviores lanceolate, raro
bifariam distiche. Flores mediocres vel minores, haud raro
minimi ochrei flavidi vel virides, rarius purpurascentes. Sepala et
petala omnia libera reflexa (perpaucis exceptis) angusta linearia,
vel lanceolata, petala semper linearia, Sepissime quam sepala
angustiora, nunquam latiora. Labellum unguiculatum, ungue
sepius ad columnam parallelo, lamina latiora integra vel bifida,
rarius subtrifida vel irregulariter laciniata, callis nullis vel duobus
conicis rarius singulo ad basin. Columna apoda saepissime
gracilis arcuata, alis brevibus duabus subtriangulis ad latera
foveæ, rarissime (in una specie) alis quatuor. Anthera ovata

OF THE GENUS LIPARIS. 253

biloculata, loculis vix omnino segregatis. Pollinia 4, cerea,
ovoidea, s: pe acuminata, apicibus per paria conjunctis; stigma
hemispherica rarissime profunda. Capsula pyriformis, globosa,
sepe pendula, vel elongata conica, erecta, costis spissime
distinctis.

$ Morrrronrz.—Folia membranacea sæpius ovata subpetiolata.
Petiolus longe vaginans, haud articulatus. Pseudobulbus
raro distinctus.
Caulis hornotinus, annotino appressus ; rhizoma brevissimum.
Elate.—Folia plura erecta lanceolata vel oblonga angusta
acuta ; flores parvi carnosuli ; lamina labelli deflexa ; callis
duobus (rarius unico) ad basin unguis.
Sepala lateralia libera lanceolata,
petalis latiora.
Species Americane et Africane.
Labellum haud crenulatum.
Sepala lateraliafaleata.... 1. L. elata.
Sepala lateralia recta .... 3. L. bicornis.
Labellum crenulatum.

Flores virides (Africana)... 4. L. guineensis.
Flores albi (Americana) .. 5. L. Kappleri.
Species Áustralasica...........- 2. L. habenarina,

Species Asiatice.
Calli conspicui.
Caulis vaginis foliaceis tectus
hine validulus.
Labellum costis elevatis
duabus in lamina...... 6. L. Rheedii.
Labellum ecostatum emar-

ginatum.
Flores flavescentes vel

WiIGU eo a cs 7. L. paradoza.
Flores purpurei ........ 8. L. nervosa.

Caulis tenuis vaginis membranaceis,
pseudobulbus parvus globosus.
Folia patula.
Folia plura, bractez post anthe-
sin haud deflexa.
Labellum obtusum ...... 9. L. bituberculata.
Labellum apiculatum .... 10. L. odontostoma.

254

Folia bina evoluta, bractex post
anthesin deflexa.
Folia 4 unciam lata .... 11.
Folia 3 uncias lata s EB
Calli inconspieui vel subnulli,

folia longissima acuta
acuminata.
Flores iis L. elate squales,
Buil dicii occus 13.
Flores minimi, apice scapi
SS PTCA Lia es 14.
Sepala lateralia libera linearia, pe-
talig subamulis ..- 0. 2v. st 39.
Sepala lateralia connata.
Lamina labelli linguiformis
mteora CILE LL DOR 15.
Lamina labelli cuneata retusa. 16.
Elliptice.—F olium evolutum singulum.
Folium ellipticum lanceolatum erec-
tum.
Labellum ovatum obtusum ...... 17.
Labellum oblongum apiculatum .. 26.
Labellum subtrilobum bicallosum . 18.
Folium ovatum cucullatum repandum.
Labellum latum apiculatum.
Planta 8-uncialis ............ 19.
Planta 4-uncialis ............ 20.
Labellum oblongum obtusum .... 95
Folia bina ovata distincte petiolata,
subdissita.
Labellum dentatum obovatum .. 21.
Labellum integerrimum vel rarius
obscure crenulatum.
Labellum flavidum .......... 22.
Labellum purpureum ........ 23.
Folia bina non vel brevissime petiolata
congesta.
Flores mediocres; labellum eeallo-
sum.
Sepala lateralia angusta récta.
Bractex pedicellis equales .... 24.

L.

MR. H. N. RIDLEY' S MONOGRAPH

olivacea.

L. platyphylla.

L.

L.

L.

MN

Serruginea.
acutissima.

longa.

. abyssinica.
. disepala.

. elliptica.
. glossula.

campestris.

. elypeolum.

cuspidata.

. rupestris.

. auriculata.

. hawaiensis.
. viridipurpurea.

. arnoglossophylla.

OF THE GENUS LIPARIS.

Bractes pedicellis multo minores.

Folia ultra dimidium longitu-
dinis liia i. e eon

Folia vix dimidio longitudinis

lata.

Labellum basi purpureo-ma-

culatum

Labellum concolor, oblon-

gum acutum..........

Sepala lateralia falcata..........

Flores mediocres virescentes ; label-

lum callosum.
Caulis brevis

foliaceis.
Petala sepalis subzequilonga.
Labellum obtusum

"^. |: n | t5: n9

uncialis; vaginis

30.

oe ee ee n n

Labellum longe cuspidatum . 29.
51.

DLL

Petala sepalis longiora
Folia ultra duo, ovata vel ovato-lanceolata.
Rhizoma brevissimum.
Sepala cum petalis connata......
Sepala et petala libera.

Flores majores; caulis breviusculus.. 32.

Caulis elatus; vaginis albescentibus

tectus.
Labellum integrum ............ 34
Labellum emarginatum: mar-

gines crenulate ecallosum . . 35.
margines crenulate bicallosum.. 38.
margines integrae 39.

Flores parve virescentes vel flavescentes.
Sepala angusta, petalis similia.
Racemus densus, multiflorus .
Racemus laxus.
Folia oblongo-lanceolata acuta.

DELLI

. 40.

Labellum obovatum oblongum . 41.
Labellum pentangulum retusum. 43.

Folia lanceolata, basi longe acumi-

nate flaccida.. ors ss ie xs 50
Folia cuneata lanceolata.
Labellum serrulatum .......... 52.

27.

255

25. L. rupestris.

L. Beddomei.

. L. liliifolia.
. L. ornithorrhynchos.

L. Bowkeri.
L. Krameri.
L. longipetala.

. L. connata.

L. ochracea.

mes longicaulis.

L. puncticulata.
L. polycardia.
L. longa.

L. capensis.

L. Loeselii.
L. campylostaliz.

. L. Gerrardi.

L. tricallosa.

256 MR. H. N. RIDLEY'S MONOGRAPH

Folia ovata vel subcordata.
Labellum integrum.
Flores omnino virides ...... 42, L. japonica.
Labellum emarginatum serrula-
tüm i E eee 45. L. Wendlandii.
Sepala lateralia, petalis latiora falcata.
Planta pusilla, 3-uncialis ........ 46. L. Guingange.
Planta ultra, 3-uncialis.
Labellum integrum.
Pedicelli strictim erecta .... 47. L. lutea.
Pedicelli patentes.
Sepala lateralia ovata; flores

BA = eL XUL vis 48. L. xanthina.
Sepala lateralia lanceolata ;
flores olivacel ........ 49. L. parva.

Labellum bilobum ligulatum .. 37. L. Eggersi.
Flores parvi purpurascentes.

Labellum glabrum ecallosum .... 44. L. Sauwndersiana.
Labellum glabrum callosum .... 53. L. brachyglottis.
Labellum hispidum ........... . 54. L. barbata.
Pusillæ vix triunciales.
Labelum bilobum ............ 55. L. biloba.
Labellum integrum cordatum
acutum -> a r Eus in 56. L. Jovis- Pluvii.
Caulescentes.
Labellum oblongum .......... 57. L. Walkerie.
Labellum obovatum serratum .. 58. L. montana.
Flores majores.
Labelli margines antice inflexe
pectinat® =- S Svo 59. L. pectinata.
Labellum haud pectinatum ob-
longum ovatum .......... 60. L. atropurpurea.
Labellum integrum orbieulare.. 61. L. Wightiana.

Rhizoma diu persistens longa; caules
secundarii dissiti.
Pseudobulbi distincti.
Folia plura; flores virescentes . 62. L. nepalensis.
Folia singula; flores purpurei . 63. L. repens.
Pseudobulbi nulli.
Labellum dentatum ........ 65. L. purpurascens.
Labellum ligulatum bilobum . 64. L. stenoglossa.

OF THE GENUS LIPARIS. 257

Labellum integrum.
Flores minimi aggregati .. 67. L. Welwitschii.

Flores parvi dissiti ...... 66. L. neuroglossa.
Flores majores; labellum
g unciam longum ...... 96. L. flavescens.

Ramosæ.—Caulis primarius longe repens foliis ovatis petiolatis
solitariis omnino tectus; racemi in axillis superioribus.
Flores magni ; labellum 3 unciam

longum atropurpureum ...... 68. L. crispifolia.
Flores mediocres.
Labellum album obtusum .... 70. L. ramosa.
Labellum album cuneatum fla-
Be ue S S 71. L. Wageneri.
Labellum viridecuspidata...... 69. L. brachystaliz.

§ Conrrrorrz.— Folia coriacea vel subcoriacea rigida cum pseudo-
bulbo articulata.
Racemus laxus, flores haud copiosi,
` raro resupinati. Columna gracilis
bialata.
Plantæ majores circiter pedales.
Labellum obovatum bilobum.
Labellum ? unciam longum...... 72. L. grandiflora.
Labellum vix $ unciam longum.
Scapus teres.
Braetes pedicellis multo breviores. 73. L. crenulata.
Bracteæ pedicellis zquales seu

longiores o. ee ois 74. L. latifolia.
Scapus anceps.
Folium enrülum —.: 75. L. Forbesi.
Folia bina ce. os 76. L. chloroxantha.
Labellum margine lacerato ........ 77. L. lacerata.
Labellum oblongum ellipticum den-
Scuba -e 78. L. Beccarii.
Labellum subquadratum crenulatum. 79. L. diodon.

Labellum integrum.
Folia singula rarius bina multi-
nervia.
Racemus laxus.

Columna crassiuseula ...... 80. L. plicata.

258 MR. H. N. RIDLEY’S MONOGRAPH

Columna gracilis cycnicollis.

Labellum integrum ...... 81. L. bootanensis.
Labellum crenulatum . .. 84. L. celogynoides.
Racemus densiusculus ........ 82. L. Griffithu.

Folia bina patula majora.
Labellum planum ellipticum
Potuindalum.-....: s.c. 85. L. plantaginea.
Folia plura erecta.
Labellum unguiculatum, ma-
cula ochrea -= 86. L. reflexa.
Labellum trifidum .... esie 83. L. Mannii.
Plantæ minores, humiles, flores parvi.
Flores aperti.
Pseudobulbus 4 unciam crassus

tumidus.
Labellum subbilobum crenu-
latam dq oes 87. L. pachypus.
Labellum divergenti-bilobum,
denticūlatum ....--.--- 88. L. grossa.
Labellum integrum ........ 90. L. Hookeri.

Pseudobulbus angustior.
Labellum lineari-panduratum. 89. L. bistriata.
Labellum obovatum.

Bracteæ floribus breviores.

Labellum bieallosum ...... 92. L. luteola.
Labellum ecallosum ........ 93. L. revoluta.
Bractee flores superantes .... 94. L. cespitosa.
Flores maperti 3: os oo o. 91. L. viridiflora.

Platystylis—Columna crassa brevis lata; flores parvi resupi-
nati.
Ale columnares evolute due.
Flores virides; labellumintegrum 95. L. resupinata.

Flores aurantiaci ; labellum
apice frilobum. .......... 96. L. aurita.
Ale columnares evolute 4 ...... 97. L. decurrens.
Distiche.—Racemus brevis congestus, bractes bifarie.
Folium 13 unciam latum ........ 98. L. compressa.
Folium 4 unciam latum.
Columna arcuata gracilis...... 99. L. Cumingit.

Columna brevis crassa recta ..100. L. disticha.

OF THE GENUS LIPARIS. 259

Densiflore.—F lores minimi copiosi densi.
Labellum integrum.
Folia lanceolata spathulata.
Labellum e basi columns oriens.. 101. L. longipes.
Labeli anguli pone columnam

brevissime produeti.......... 102. L. pusilla.
Folia lineari-lanceolata angusta. 108. L. clavigera.
Labellum retusum crenulatum .. 104. L. vestita.

Labellum trilobum.
Lobus medius triangulus integer. 105. L. stachyurus.

Lobi omnes dentieulati ...... 106. L. Stricklandiniana.
Lobus medius ovatus lanceo-
latus o oa Se Lt 107. L. triloba.
Lobus medius obtuse tridentatus 108. L. dendrochilum.
Lobus medius bilobus ........ 109. L. parviflora.
Labellum apice bilobum ........ 110. L. flaccida.

$Morrrronrnr;.

Herbs tenues, sspe terrestres, foliis herbaceis viridibus
purpurascentibus ovatis vel lanceolatis, in apicibus caulium

secundariorum sspe vix basi incrassatorum marcescentibus nec
articulatis.

1. L. ELATA, Lindl. Bot. Reg. t. 1175; Orch. Pl. p. 27.

Caulis basi pseudobulbosus, vaginis viridibus tectus, uncialis ;
folia lanceolata acuminata erecta 4-6 uncias longa, unciam lata ;
scapus pedalis vel altior, angulatus et alatus, basi nudus; race-
mus laxus multiflorus; flores parvi, fusco-viridis vel purpuras-
centes; bractez lanceolate 2 lineas longs; sepalum posticum
angustum lanceolatum obtusum, lateralia latiora subfalcata ;
petala multo angustiora linearia obtusa; labellum carnosulum
unguiculatum obovatum emarginatum, sepius macula pustulata
obscuriore ad basin; calli duo conici in basin; columna sub-
erecta apice curva subclavata, alis brevibus latis obtusis; pedi-
celli erecti cum ovariis } unciam longis.

West Indies: Jamaica, Morris!; St. Vincent, Anderson!;
Dominica, Imray! Costa Rica, Wendland. Ecuador, Banos
near the river Verde, Spruce no. 5382! Venezuela, Fendler
no. 2141! Caripe, Moritz no. 618! Merida, Linden! no. 6771!
Brazil, Burchell 2277! Organ Mountains, Gardner, 675, 676!
Miers no. 2447! Minas Geraes, Widgren!

260 MR. H. N. RIDLEY’S MONOGRAPH

Var. rNUNDATA, Rodriguez, Gen. § Sp. Orch. p. 86.

Minor, hydrophila, foliis plicatis eum marginibus incurvis,
floribus flavis, labello oblongo subquadrato, apice retuso.— Brazil.

Var. LATIFOLIA. Caulis magis dilatatus ad basin; folia plura,
late ovata, 6-9 uncias longa, 2-31 uncias lata, obtusa vel acuta;
scapus elatus angulatus et breviter alatus; flores fusco-virides
vel flavi, labello fusco; sepalum posticum lanceolatum ligulatum
obtusum, lateralia lanceolata latiora subacuta haud falcata.

Cuba, Wright no. 1495! Paraguay, Valenzuela, Balansa
no. 4542!

Var. RUFINA. Flores rufi, obseuri ; labellum vix emarginatum.

Africa: Lagos, Barter! (“flowers deep red") ; Sierra Leone,
Morson !

2. L. HABENARINA, F. Muell. Fragm. Fl. Austral.iv.p. 131; Benth.
Flor. Austral. vi. p. 273.—Sturmia habenarina, F. Muell. l. c.

Pseudobulbus oblongus, foliis vaginantibus tectus more L.
paradoxe, biuncialis ; folia evoluta bina, lanceolata erecta acuta,
5 uncias longa, 3 unciam lata ; scapus crassiusculus vix angulatus
9-uncialis; flores parvuli laxi; bracteæ lanceolate anguste ;
sepala lateralia faleata obtusa carnosula, posticum petalaque
linearia obtusa membranacea ; labellum carnosulum breve, sepalis
lateralibus vix equans, lamina oblongo-cuneata obtusa deflexa;
apice paullo retuso cum cuspide mediana minuta; calli parvi
conici ad basin unguis ; columna crassiuscula erecta paullo curva,
apice dilatata, alis parvis subacutis; anthera ovata acuta, loculis
distinctis.

Australia : Queensland, Rockingham Bay, Dallachy!

3. L. Bicornts, Ridl. Journ. Linn. Soc., Bot. vol. xxi. p. 458.

Differt racemo densiusculo, labello oblongo-cuneato emarginato,
sepalis ovato-lanceolatis haud falcatis, sepalo postico lanceolato.

Madagascar, Hildebrandt! Baron!

4. L. eviNEENsrS, Lindl. Bot. Reg. t. 1671.

Pseudobulbus “ovatus parvus, avellane nucis magnitudine " ;
folia lanceolata acuminata plicata, 51—63-uncialia vix unciam lata ;
scapus angulatus, 11-uncialis ; racemus longus laxiusculus ; flores
virides 20-30; bractez lanceolate acuminate, i-unciales, quam
ovarium breviores ; sepalum posticum angustum lineare oblongum
obtusum, lateralia ovata oblonga faleatula obtusa, labello æqui-
longa; petala linearia quam sepalum postieum angustiora quam
sepala longiora; labellum cuneato-obovatum bilobum, margine

OF THE GENUS LIPARIS. 261

crenulato, calli cornuti breves; columna crassiuscula, apice curva,
alis rotundatis brevibus latis ; anthera ovata obtusa.
Western Tropical Africa, Sierra Leone, Wilford! Whitfield

5. L. KAPPLERI, Reichb. fil. in. Walp. Ann. vi. p. 218.—Sturmia
Kappleri, Reichb. fil. in Bonplandia, ii. p. 22.

Folia oblonga acuta, basi cuneata membranacea ; scapus elon-
gatus in sicco sulcatus apice pauciflorus ; bractez lineares, sub
anthesin reflex» ; sepala oblonga obtusa subzqualia; petala
linearia obtusa; labellum cuneatum antice retusum obscure bis
emarginatum crenulatum ; columna incurva alata ; flores albi.
Differt L. guineensi, Lindl., caule alato et forma labelli.

Surinam, Para, Kappler (non vidi; forsan varietas L. elate).

6. L. Raren, Lindl. Orch. Pl. p. 27; Mig. Fl. Ind. Bat. iii.
p. 621.—Malaxis Rheedii, Blume, Bijdr. p. 389, t. 54.

Caulis subcylindricus vix pseudobulbosus, foliis vaginantibus
pluribus tectus ; folia evoluta tria erecta lanceolata acuminata
acuta, subpetiolata, 3-6 uncias longa, 13 lata; scapus elatus
crassiusculus angulatus 9-uncialis usque pedalis; racemus multi-
florus ; flores carnosuli virescenti-flavi ; bractew lanceolate acu-
ininatz foliacew, fuiu } unciam longs. ovariis equantes vel infe-
riores paullo superantes; sepalum posticum anguste ligulatum
obtusum, 4 lineas longum ; lateralia late oblonga elliptica falca-
tula; petala anguste linearia ligulata; labellum obovatum rotun-
datum obtusissimum marginibus minute crispulis, costis duabus
in lamina ad callos conicos euntibus ; columna crassiuscula basi
et apice dilatata apice arcuata, alis brevibus vix conspicuis;
anthera depresso-conica.

Java: Mt. Seribu, Blume ; Tjikoya, Zollinger 146, Horsfield !

7. L. PARADOXA, Reichd. fil. in Walp. Annals, vi. p. 218.—
Empusa paradoxa, Lindl. Bot. Reg. t. 825, Orch. Pl. p. 17; Miq.
Prolus. Fl. Jap. p. 135.—Empusaria , Reichb. f. Conspectus,
p. 69.—Malaxis lancifolia, Smith in Rees’s Cyclop.

Caulis sepius vix pseudobulbosus brevis; folia lanceolata
acuminata suberecta, 5 uncias longa, 14 lata; scapus angulatus et
ad basin alatus, 6-uncialis ; racemus laxus terminalis multiflorus ;
bractez lanceolate acuminate erectz dimidio ovarii xquales, 4
unciam longs; flores iis ZL. elate æquales viridi-flavi; sepala
oblonga obtusa lateralia semiovata falcata basi obscure paullo
connata ; petala linearia angusta; labellum obovatum retusum

LINN. JOURN.—BOTANY, VOL. XXII. Y

262 MR. H. N. RIDLEY'S MONOGRAPH

cuspide minuto, marginibus ad apicem minute denticulatis; calli
conici; columna gracilis arcuata, alis brevibus ovatis.

India: Nepal, Kumaon, Blinkworth, Wallich herb. no. 1937
pars! Canara and Mysore, Law. Malabar, Hooker & Thom-
son, Liparis, ii.! Guliwal, Falconer. Moulmein, Parish! Kha-
siya hills, Hooker & Thomson!

Siam and Cambodia, Angkor Than,Iles de Phuquoc, and Bayoc,
Godefroy-Leboeuf! Java, Modjiterti, Zollinger, Iter. ii. no. 8766.

L. odorata, Lindl. Orch. Pl. p. 26; Reichb. fil. in Walp. Ann.
vi. p. 217.—Malaxis odorata, Willd. no. 6.—Katton pounam
Marawara, Rheede, Hort. Malabar, xii. p. 55, t. 28, mihi videtur
hzc species esse, nam figura satis bene congruit, forma coloreque,
est etiam species vulgaris Malabarica. Thwaites (Enum. Pl. Ceyl.
p. 295) credidit L. odorata esse L. elata, Lindi. (i. e. L. bituber-
culata, Lindl.), L. odorata, Benth., Flora Hongkong, p. 352, sine
dubio L. nervosa est. Itaque, specie omnino dubia, nomen etsi
antiquius melius delendum.

Foliola duo sunt in herbario Wallichiano sub numero 1937 et
nomine Empuse paradoxe, in collectionibus societatis Linnzane ;
harum unum (1937 a) continet specimen singulum L. olivacee
cum tribus E. paradoxae, alterum (1937 b), specimen L. paradoxae,

detioratum, cum tribus- £, bituverculate.

8. L. nervosa, Lindl, Gen. Ẹ§ Sp. Orch. Pl. p. 26.— Reichb. fil.
in Walp. Ann. vi. p. 217; Benth. Fl. Hongkong. vol. vi. p. 852;
So Mokou Soussetz, xviii. p. 67, fig. C; Franchet § Savatier,
Enum. Pl. Jap. ii. p. 21.—Ophrys nervosa, Thunb. Fl. Japon.
p. 27.—Epidendrum nervosum, Thunb. Ic. Pl. Jap. dec. i. t. 10.
—Sturmia nervosa, Reichb. fil. in Bonplandia, iii. p. 250.

Habitus L. paradoxe. Caulis brevis vix pseudobulbosus ; folia
ovata vel lanceolata acuminata, 2-3 rarius 4 tenuia; scapus
3-6-uncialis ; racemus laxus, flores iis L. paradoxe equales rubri
vel purpurei; sepala lateralia oblonga lanceolata falcata, posticum
angustius, ellipticum lanceolatum, petala linearia, omnia pallide
purpurea vel flavo-rubra; labellum obovatum o¥tusum emargi-
natum, marginibus integris purpureum ; calli parvi; columna
curva, alis distinctis subacutis.

China: Hong Kong, Victoria Peak, Champion! Wilford, Hance.
Japan; Nippon, Thunberg; Niko, Savatier ; Nagasaki, Mazimo-
wicz !

Formosa, Oldham! | Loochoo Islands, Wright no. 342!

OF THE GENUS LIPARIS. 263

Hee planta (presertim sicca) L. paradoxam simulat, sed differt
semper ut videtur colore. Hane ex India non vidi nec illam ex
regione Japonica-sinica.

9. L. BITUBERCULATA, Lindl. Bot. Reg. p. 882.—Cymbidium ?
bituberculatum, Hooker, Exot. Flor. t. 116.—Sturmia bituber-
culata, Reichb. fil. in Bonplandia, ii. p. 22.

Caulis distinctus triuncialis, vaginis laxis albescentibus (in
sicco) tectus, basi vix dilatatus, radices crassi lanati; folia plura
lanceolata acuminata patula, 5 uncias longa, 14 lata; scapus 8-12-
uncialis angulatus ; racemus laxus multiflorus ; bracteæ lanceo-
late; pedicelli 3-unciales, graciles; flores purpurei; sepala linearia
ligulata, lateralia faleatula ; petala multo angustiora; labellum
oblongum cuneatum paullo retuso, marginibus integris; calli
conici; columna erecta, apice curva, alis brevibus triangulis sub-
acutis.

India: Nepaul, Wallich ‘no. 1937, 2 pars! Kumaon, Blink-
worth, sub nomine L. paradoxe, Lindl! Sikkim, J. D. Hooker!

Ceylon, Thwaites no. 3180, 3375.

Var. ? FORMOSANA.—L. formosana, Reichb. f. in Gard. Chron.
1880, p. 394.

Pseudobulbi elongati ferme cylindrici, folia oblonga acuta,
crispa; labellum sagittatum oblongum acutum.

Formosa, introd., Maries.

10. L. ornowrosrOowa, Reichb. fil. in Linnea, xli. p. 97.

Pseudobulbus subteretiusculus, apice plurifolio ; folia circiter 4
euneato-oblonga acuta membranacea ; scapus exsertus pluriflorus
racemosus; sepalum posticum lanceo-trinerve, lateralia oblongo-
ligulata obliqua 4-nervia curva; petala linearia uninervia ; labelli
unguis basi utrinque obtusangulus, lamina subquadrata antice
emarginata cum apieulo ungui squilonga, calli conici gemini
ante basin unguis; columna basi et apice ungulata (? angulata).
Planta bipedalis; flores verosimiliter olivacei.

Sikkim, 3000-5000 feet, Hooker.

11. L. OLIVACEA, Lindl. Wall. Cat. no. 1942; Orch. Pl. p. 27.
Caulis uncialis angustus, basi pseudobulbo parvo ovali dis-
tincto; folia bina lanceolata acuminata, 5-nervia, 3 uncias longa,
4 lata (quo latissima); scapus 6-uncialis angulatus, basi nudus;
racemus multiflorus, flores iis L. paradoxe equales, olivacei ;
bractee lanceolate acuminate, post anthesin deflexe ; sepala an-

gusta ligulata obtusa lateralia faleatula; petala anguste linearia ;
x2

264 MR. H. N. RIDLEY'S MONOGRAPH

labellum euneatum obovatum emarginatum cuspide minuto, mar-
ginibus integris; calli breves conici; columna gracilis arcuata, alis
obscuris.

Nepal, Wallich no. 1942!

L. odorata, Lindl., valde affinis, sed bracteis 3 unciam longis,
sub anthesin deflexis, floribusque majoribus et colore diverso di-
stinctus. In herbario typico Wallichiano est etiam specimen
singulum cum Z. paradoxa commixta sub numero 1937 a.

12. L. PLATYPHYLLA, n. sp.—Caulis l-2-uncialis, angustus ;
folia bina ovata vel ovata lanceolata acuta, patula, 7-nervia, 4
uncias longa, 3 lata; scapus erectus debilis basi nudus pedalis;
flores plures laxe racemosi ; bractew lanceolate acuminate, primum
erecte comoss demum reflexe pedicellis equalibus; sepala lan-
ceolata ligulata, lateralia labello supposita; petala angustiora ;
labellum obovatum rotundatum paullo retusum marginibus crenu-
latis, unguis brevis; calli conici duo ; columna crassiuscula brevis,
alis obtusis longis angustis.

Anamallays, 3500 ped. alt., Beddome! Herb. Brit. Mus.

Preecedenti affinis sed foliis latioribus, labello crenulato diversa.

13. L. FERRUGINEA, Lindl. in Gard. Chron. 1848, p. 55.

Planta sesquipedalis vel bipedalis. Pseudobulbus ovalis brevis,
caulis rectus elongatus; folia 3—4 angusta lanceolata acuminata
stricta 6-15 uncias longa, 4 unciam lata; scapus elongatus vix
angulatus exalatus, validulus, superne racemosus ; racemus laxus
multiflorus, flores iis Z. bituberculate equales ; bractex lanceolate
minime demum deflexe ; sepalum posticum angustum lanceo-
latum, lateralia latiora labello supposita, vix faleatula ; petala an-
guste linearia; labellum euneatum rotundatum retusum cuspide
minuto, ferrugineum vel roseum etubereulosum ; columna curvula,
alis obtusis.

Borneo, Labuan, Lobb! Malacca, Griffith !

Cambodia, Mont de Pursat, Godefroy-Leboeuf !

L. paradox et olivacea affinis, sed ecallosa.

14. L. AcUTISSIMA, Reichb. fil. Otia Hamb. i. p. 37.

Caulis basi breviter pseudobulbosus, subglobosus, vaginis tectus ;
folia 4—5 angusta linearia lanceolata acuminata stricta, 3-6 uncias
longa, j unciam lata; seapus gracillimus subteres, 14-pedalis, pro
maxima parte omnino nudus ; racemus multiflorus laxus 21
uncias longus ; flores minimi “flayo-brunnei ?; pedicelli 2; unciam
longa, bractex lanceolatz minute ; sepalum posticum lanceolatum

OF THE GENUS LIPARIS. 265

obtusum, lateralia latiora ovata lanceolata faleatula; petala an-
gusta linearia filiformia; labellum cuneatum oblongum bilobum,
breve carnosulum, callis obscurissimis.

Cambodia, Angkow Thom and Bayoc, island of Phuquoc, Gode-
Jroy-Leboeuf!

15. L. ABYSSINICA, Richard, Tentamen Flore Abyssinice, vol. ii.
p. 281.—Sturmia abyssinica, Reichb. fil. in Walp. Ann. iii. p. 527.

Pseudobulbus squamis laxis obtectus; caulis 3—4-uncialis ;
folia szepius bina rarius tria, ovalia acuta vel ovali-oblonga ; racemus
pauciflorus ; flores purpurascentes; bracteæ ovales lanceolate
acuta, ovario equales; sepalum posticum lanceolatum obtusius-
culum trinervium, lateralia pendula ultra mediam partem coalita,
semiovalia obtusa; petala longiora linearia; labelli basis erecta
semicanaliculata, lamina linguiformis obtusa crassa integra supra
basin carunculata.

Abyssinia prope Adoua, Quartin Dillon (non vidi).

16. L. DISEPALA, Reichd. fil. in Linnea, xli. p. 97.

Ultra pedalis, vultus onmino Z. elate. Folia angustiora tria ob-
longa ligulata acuminata bene nervosa ; scapus longus exsertus laxe
racemosus; bractez triangulo-lineares parve, ovariis pedicellatis
longe breviores; sepalum posticum ligulatum obtuse acutum,
lateralia bene latiora connata bifida; petala linearia deflexa ;
labellum cuneatum oblongum retusum cum apiculo in medio,
callo parvo lamelliformi subquadrato obscuro in medio apiculato ;
columna apice ampliata.

New Caledonia, Vieillard, Deplanche.

Omnes precedentes, inter se plerumque valde affines, sectio-
nem Elate formant, distinct labello unguiculato, bicalloso, et
habitu.

17. L. ELLIPTICA, Heichb. fil. in Walp. Ann. vi. p. 218; Grise-
bach, Flor. Brit. West Indies, p. 612.—L. Lindeniana et Galeot-
tiana, Hemsley in Gard. Chron. 1879, i. p. 559.—L. jamaicensis,
Lindl. ez Griseb. Cat. Pl. Cub. p. 261.—Cymbidium vexilliferum,
Llave § Lex. Nov. Veg. Descr. fasc. ii. p. 11.— Malaxis Galeot-
tiana et Lindeniana, Rich. Ann. Sc. Nat. sér. 8, iii. p. 18.—Sturmia
elliptica, Reichb. fil. in Linnea, xxii. p. 833 ; Bonplandia, il. p. 22.

Pseudobulbus ovatus semiuncialis, folia vaginante tectus ; folium
singulum erectum oblongum lanceolatum acutum, 3j-4 uncias
longum, 4 unciam latum ; scapus anceps 5-uncialis ; racemus com-
pactus 12-30-florus; bractes foliacee lanceolate acuminate

266 MR. H. N. RIDLEY S MONOGRAPH

semiunciales ovariis equales; flores mediocres purpurascentes,
* luridi " (Llave), * atropurpurei " (Galeotti); sepalum postieum
lineare obtusum, lateralia lanceolata obtuso labello supposita ;
petala angustissime linearia, sepala lateralia paullo superantes ;
labellum ovatum obtusum marginibus undulatis, lateribus versus
apicem emarginatis, ecallosum, vena mediana elevata, versus basin
ramos duos laterales emittens ; columna arcuata gracilis, alis parvis
brevibus obtusis; anthera parva ovata truncata apice emarginata.

West Indies, Jamaica, Macfadyen !

Mexico, Vera Cruz, Galeotti ; Mirador, Linden 199!

Guatemala, Bernoulli 290 !, Skinner!

Colombia, Moritz 889; Banos in Pastis, Spruce! Venezuela,
Lagunitans, Moritz 1973! Caripe, Moritz 625! Cumana, Funcke
675! Caraccas, Wagener. Santa Martha, Purdie!

British Guiana, Watu Ticaba, Schomburgk !

Species in regione Americana-australi equatorea, ut videtur
vulgaris, variabilis, lorum dimensionibus varians.

18. L. CAMPESTEIS, Barb. Rod. Gen. & Sp. Nov. Orch. i. p. 36.

Pseudobulbi minimi obovales, bracteis vestiti monophylli ;
folium oblongum acutum erectum semiplicatum; spica multi-
flora; caulis trigonus duplo major foliis ; bractes oblonge acute
ovario duplo major foliis; sepala linearia oblonga obtusa, late-
ralia largiora; petala linearia; labellum oblongum subtrilobum,
apice acutum cum linea in centro elevato desinente in callum
bidentatum ; columna semiarcuata, apice bidentata; flores flavo-
virescentes.

Brazil, Caldas et St. Paul, Rodriquez, Regnell.

19. L: CLYPEOLUM, Lindl. Gen. & Sp. Orch. Pl. p.29; Nadeaud,
Flor. Tahit. p. 86.—Epidendrum clypeolum, Forster, Prodr.
no. 823.—Cymbidium clypeolum, Willdenow, no. 22.

Radices copiose lanatæ; pseudobulbus biuncialis vaginis albis
membranaceis laxis tectus; folium singulum orbiculatum basi
cucullatum patens, 4 uncias longum et æquilatum ; scapus gra-
cilis teres 6-uncialis parte majore inferiore nudo; racemus laxus ;
flores mediocres circiter 8; bractes minim: ; sepala et petala
anguste linearia patentia; labellum magnum tenue orbiculatum
integrum ; columna parum arcuata gracilis.

Society lsles, Bidwill! Tahiti, Forster! Moseley! Samoa,
Powell! Arue, Nadeaud.

20. L. cusPrDATA, n. sp.—Pusilla; caulis vix pseudobulbosus,
vaginis 2-3 tectus ; folium singulum tenue late ovatum, 2 uncias

OF THE GENUS LIPARIS. 267

longum, 14 latum; scapus inferne nudus gracilis. angulatus
3-unciale; bractez brevissime ; flores 6-8 mediocres, pedicellis
longiuseulis gracilibus suberectis ; sepala anguste linearia, petala
subsimilia angustiora; labellum tenue patens cordatum cuspida-
tum pallidum, linea media obscuriora 1 unciam longum ; columna
gracilis curvula, alis parvis.

Society Islands, Tahiti, Collie and Lay! Herb. Brit. Mus.

Planta habitu precedentis, sed minor labello cuspidato. In
itinere Beecheyano lecta, sed adhue przetermissa.

2]. L. Avn1CcULATA, Mig. Prolusio Fl. Jap. p. 135 (1866), non
Reichb. fil.

Pseudobulbus ovatus, vaginis laxis tectus; folia bina ovata
rotundata petiolata; petiolus basi paullo dilatatus, 14 uncias;
lamina 2 uncias longa, 11 lata, 8-nervia ; scapus erectus 6-uncialis,
angulatus nec alatus; flores circiter 10 erecti iis L. Loeselii majores
in racemo terminali laxo ; bractex lanceolate acute quam pedicelli
dimidio breviores, } unciam longs; sepala ligulata lanceolata
obtusa recurva, petala linearia; labellum cuneatum obovatum
minute dentatum, 2 unciam longum linea media (in sicco) discolori
macula obscura ad basin paullo incrassata; columna arcuata
graeilis, alis parvis quadratis.

Japan: Nippon, Tschonoski !

22. L. nawaIENSIS, Mann, Proc. Amer. Acad. vi. p. 207;
Wawra, in Flora, 1875, p. 246.

Pseudobulbus parvus; folia bina ovalia vel oblonga obtusa
membranacea plicata petiolata, lamina 21—4 uncias longa, 1-23-
uncialis lata, petiolus circiter uncialis ; scapus angulatus, folia bis
superans; racemus pluriflorus ; sepala lanceolata lateralia mar-
gine revoluta, petala filiformia; labellum obovatum integerrimum
vel crenatum flavidum 3 lineas longum; capsula obovata clavata

eostata.
Sandwich Isles, Mann & Brigham, Wawra, Hillebrandt!

23. L. VIRIDI-PURPUREA, Griseb. Cat. Pl. Cub. Wright, p. 261.
Caulis vix pseudobulbosus uncialis; folia bina ovata elliptica
acuta petiolata, 3—4 uncias longa, 11-2 lata subdissita ; scapus
3-6-uncialis angulatus ; racemus pauciflorus laxiusculus ; bracteæ
j-unciales, lanceolate acuminate ; sepala linearia ligulata rufes-
centi-viridia, petala angustiora; labellum cordatum ovatum sub-

acutum, “ purpureum."
Cuba, Wright 1691 !

268 MR. H. N. RIDLEY S MONOGRAPH

24. L. ARNOGLOSSOPHYLLA, Reichb. fil. in Bonplandia, 1856,
p. 217.—Ophrys rotundifolia, Ruiz, MSS.

Folium inferius 3-4-unciale oblongum acutiusculum horizon-
tale prope 3 pollices latum, superius dimidium minus; bractew
lineari-lanceolate, pedicellis subequales ; scapus 3J-uncialis
alatus ; racemus 2}-pollicaris ; flores quam iis Z. liliifolie paullo
minores minus tenues; petala linearia, nec subulata; labellum
oblongum acutiusculum nec euneatum obovatum apiculatum.

Central America, New Spain, Ruiz; “exstat specimen unum
diphyllum," non vidi.

25. L. RUPESTRIS, Griffth, Notule, ii. p. 276.—L. rostrata,
Reichb. fil. in Linnea, xli. p. 44.

Pseudobulbus 1}-uncialis parum dilatatus laxe vaginatus ;
folium 1 rarius 2 tenue late ovatum acutum repandum (^ glaucum,”
fide Griffith) 4 uncias longum, 3 uncias latum, 18-nervium ; scapus
brevis gracilis nutans, 4—6-uncialis, angulatus alatus, racemus laxi-
usculus pluriflorus; bractee minimæ ovate lanceolate; flores
mediocres virides, pedicellis longis erectis tenuibus ; sepala linearia
reflexa ; petala angustiora subsimilia ; labellum porrectum oblon-
gum obtusum, apice minute emarginatum cuspide minute; columna
cycnicollis arcuata, apice incrassata, alis obtusis; capsula semi-
uncialis elliptica.

Assam, Khasia, Nunklow, Griffith! Hooker & Thomson!
N.W. India, Royle! Thomson!

Var. PURPURASCENS, n. var.—Labellum pallide obscure purpu-
reum ; folia bina subpetiolata.

Lachen, Sikkim, J. D. Hooker!

26. L. arossuLa, Reichb. fil. in Linnea, xxi. p. 43.

Pseudobulbus ovoideus, caulis pauci-vaginatus monophyllus;
folium euneatum oblongum obtuse acutum seu obtusum ; scapum
angulatum, foliis longior seu :equalis; racemus pluriflorus; brac-
teis lanceis ovaria pedicellata infima dimidio squantibus; sepala
lineari-lanceolata, petala linearia; labellum basi columnam am-
plexum, utrinque angulo retrorso cuneata, basi brevissima ob-
longa obtusa apiculata ecallosa; columna basi et apice dilatata ;
flores iis Z. elliptice simulantes sed L. liliifolie equales.

Ind. or.

Hanc plantam nescio, an L. rupestris, Griff., diversa ?

27. L. BeppoMEr, n. sp.—Caulis uncialis vix bulbosus; folia
bina rarius terna ovata lanceolata acuminata, 1 unciam longa, 3
unciam lata; scapus gracilis biuncialis angulatus ; racemus laxus

OF THE GENUS LIPARIS. 269

pauciflorus; bractee lanceolate acuminate uninervie 1 linea
longi; ; flores parvi, pedicellis gracilibus suberectis 3 lineas longis ;
sepala angusta ligulata linearia; petala angustissime linearia
acuminata; labellum planum tenue oblongum rotundatum ob-
tusum viride macula purpurea in medio; columna gracillima
areuata, alis minimis; anthera ovata obtusa.

Southern India, Shembanganoor, Pulney mountains, 5000 feet

alt., Beddome! in Herb. Brit. Mus.

28. L. LILITFOLTIA, Richard, Mém. Mus. Par. iv. p. 52 ; Sprengel,
Syst. Veget. ii. p. 740 ; Lindley, Orch. Pl. p. 28; et Bot. Heg.
t. 882; Torrey, Fl. New York, ii. p. 271; Chapman, Flor. U. 8.
p. 494; So Mokou Soussetz, xviii. p. 67, fig. A; Franchet & Sa-
vatier, Flor. Jap. ii. p. 21; Asa Gray, Bot. Jap. p. 409.—Ophrys
tiliifolia, Linn. Sp. Pl. 946; Redouté, Liliac. t. 437.—O. trifolia,
Walt. Fl. Carol. p. 221.—Malaxis liliifolia, Swartz. Act. Holm.
1800, p. 235 ; Elliot, Bot. S. Carol. § Georg. ii. p. 503 ; Nuttall,
Gen. Amer. Pl. ii. p. 196; Andrews, Rep. t. 65; Pursh, Flor. ii.
p. 992, t. 2004; Bot. Cab. t. 198 ; Ehret, Phil. Trans. lii. t. iv. p. 81.

Pseudobulbus ovoideus, vaginis ampliatis membranaceis ; folia
bina ovata obtusa, 11—5 uncias longa, 21 uncias lata; scapus 31-
8-uncialis, infra nudus; racemus laxus 12-80-florus; bractes
parv: ovate acute quam pedicelli multo breviores ; sepala linearia
obtusa, petala linearia angustissima viridia; labellum cordatum
oblongum integrum breviter mueronatum ecallosum purpuras-
centi-viride; columna gracilis arcuata medio angustata, alis
brevibus; ovaria et pedicelli graciles patentes purpurei.

N. America, Philadelphia, Nuttall! Buffalo, N. Y., Clinton!
Cincinnati, J. Clark! Virginia, Clayton! Missouri, Letterman!

Japan, Nippon and Yeso, Hakodate, Wright!

29. L. Kramert, Franchet 4 Savatier, Flora Jap. ii. pp. 22,
509; So Mokou Soussetz, xviii. p. 68, fig. D.

Pseudobulbus parvus, iis Z. Loeseli? similis ; folia bina ovata
vel ovata lanceolata spathulata subacuta crispula, 21 uncias longa,
ł lata; scapus gracilis alatus 4—S-uncialis (rarius multo minor
biuncialis, foliis l-uncialibus); flores pauci ad 9 flavescenti-
virides laxi; bractez ovate lanceolate minutissime ; sepala an-
gusta lineari-lanceolata acuminata, lateralia } unciam longa,
labellum superantes; petala angustiora linearia acutissima; la-
bellum geniculatum basi angustatum, callo magno subcylindrico
apice eroso ad basin, ad geniculationem carinis duabus in lamina
labelli attenuatis ; lamina ovata longe cuspidata, margine minute

270 MR. H. N. RIDLEY'S MONOGRAPH

dentato ; columna erassiuscula parum curva, alis brevibus erectis
cornutis; alabastra aeuminata.

Japan, Nippon, 7Schonoski! Kramer.

30. L. Bowxznr, Harvey, Thes. Cap. ii. p. 7, t. 109.

Caulis breviusculus 1—}-uncialis; folia sspius bina evoluta
ovata acuta tenuia haud petiolata, 4 uncias longa, 14 uncias lata ;
seapus debilis angulatus et brevissime alatus 4 uncias longus;
bractez 2-unciales ovate acuminate virides; flores ochracei ad
15 laxe racemosi, pedicellis longiusculis ; sepalum posticum an-
guste lanceolatum obtusum, lateralia elliptica lanceolata latiora
et falcata obtusa, venis 4 prominulis, labello subzequalia ; label-
lum unguiculatum ungue brevi, lamina ovalis rotundata obtusa
obscure crispula, nervis prominulis prsesertim medianis; calli
duo perparvi subconici; columna gracilis arcuata basi dilatata,
alis brevibus obtusis; anthera ovata subacuta.

S. Africa: Caffraria, Kreili's country, Harvey! Grahamstown,
Atherstone! Natal, Gerrard! Kai Kloof, Saunders!

31. L. ORNITHORRHYNCHOS, Ridl. in Journ. Linn. Soc., Bot.
xxi. p. 460, differt labello ovato-cordato, venis haud prominulis ;
callis nullis, anthera acuminata, floribus virescentes.

S. Betsileo, Madagascar, Hildebrandt! Deans Cowan!

32. L. OCHBACEA, Ridl. in Journ. Linn. Soc., Bot. xxi. p. 461,
differt a precedente ungui labelli longiori, lamina abrupta deflexa
oblonga obtusa macula ochrea in mediis (eo L. reflexe similis),
foliis pluribus.

Ankafana, Madagascar, Deans Cowan !

83. L. connata, Ridl. in Journ. Linn. Soc., Bot. xxi. p. 462.
Insignis sepalis connatis, reliquas perianthii includentibus, venis
elevatis et conspicuis, labello erenulato, callis duobus, forsan est
monstrum. Affinis Z. Bowkeri, sed preter alia, foliis pluribus

diversa.
Madagascar.

34. L. rosG1caAULIS, Ridl. in Journ. Linn. Soc., Bot. xxi. p. 461.
Differt caule elatiore tenui, vaginis albescentibus membranaceis
pluribus tecto; scapo elato gracili; labello oblongo crenulato.

Madagascar, Ankafana, Hildebrandt! Deans Cowan!

35. L. PUNCTICULATA, Ridl. in Journ. Linn. Koc., Bot. xxii.
p. 119, habet caulem 10-uncialem altum tenuem, vaginis albescen-
tibus membranaceis, dissitis, puneticulatis; labello bilobo, lobis
denticulatis divaricatis.

Madagascar, Baron !

OF THE GENUS LIPARIS. 271

36. L. rnAvESCENS, Lindl. Bot. Reg. sub t. 882; et Orch. Pl.
p.29; Spreng. Syst. Veg. ii. p. 741; Ach. Richard, Orch. Maur.
p. 51; S. Moore, Flor. Maur. & Seych. p. 842; Boj. Hort. Maur.
p. 320.—Malaxis flavescens, Thouars, Orch. Iles Afr. t. 25.

Rhizoma diu repens; caulis biuncialis haud incrassatus, vaginis
albescentibus ampliatis tectus; folia bina ovata lanceolata bre-
viter petiolata, 2 uncias longa, 1 unciam lata; scapus exalatus
4-uncialis pro majore parte floriferus ; bractes lanceolate acute
foliaceze ; pedicelli erecti; racemus pluriflorus laxus; flores iis
L. ornithorrhynchi minores; sepala lateralia faleata lanceolata,
labello supposita, posticum ligulatum ; petala linearia; labellum
vix ultra 3 unciam longa, quam sepala multo brevius, subcordata
rotundata lata apiculata, venis conspicuis; callus unicus lingui-
formis; columna breviuscula curvula.

Madagascar, Lyall! Fox! Mauritius, Bojer. Seychelles,
Horne! Bourbon, Thouars.

37. L. Eaaerst, Reichb. fil. in Ber. Bot. Gesellsch. iii. p. 278.

Affinis L. flavescenti. Usque spithamea ; folia gemina seu terna
cuneato-oblonga acuminata medium pedunculum non attin-
gentia ; pedunculus præsertim inter inflorescentiam anceps, sub
inflorescentia pauci ac distanter vaginatus purpureus usque 5-
florus ; bracteæ triangulæ pedicellos nune subæquantes deflexe ;
sepalum posticum ligulatum obtuse acutiusculum lateralia latiora
breviora ovata acuta subcurva; petala filitormia apice latiora
obtusa spathulata deflexa ; labellum ligulatum antice dilatatum,
bilobum basi bicorniculatum ; columna apice bidentata, mem-
brana cingente postiea humili emarginata. Dimensiones L. Loe-
selii fortis.

West Indies, Baron Eggers.

88. L. POLYCARDIA, Reich. fil. in Flora, 1885, p. 543.

Ultra pedalis; caulis basi vix tumidus, vaginis amplis acutis
membranaceis vestitus ; folia genuina oblonga acuta membranacea
24 uncias longa, #-1 unciam lata; scapus elongatus laxiflorus ;
bracteæ in basi probabiliter fatus majores cordato-lanceolate,
Superiores vere subæquales minores, ovarii pedicellato dimidio
ferme squantes; sepalum posticum lineare, ligulatum, lateralia
faleato-ligulata; petala filiformia: labellum cuneatum flabellatum
dilatatum antice emarginatum, marginibus anticis crenulatis ;
callus depressus foreipatus in basi (sc. lamina antrorsum bicus-
pidata); columna arcuata apice et basi incrassata.

Regio Africana austro-orientalis, sine loco speciali aut collec-

272 MR. H. N. RIDLEY 8 MONOGRAPH

tore. Non vidi, probabiliter est LZ. puncticulata affine, sed folia
pauciora, ete.

39. L. roxaa, Reichd. fil. in Linnea, xli. p. 97.

Caulis elatus subteres, basi paullo incrassato vaginatus ; folia
genuina subopposita oblonga acuminata; scapus racemosus ex-
sertus inferne subnudus squama una hic inde, bractezx semilancez
subuninervie (nervo sc. unico tantum evoluto), ovariis pedicellatis
subequales ; sepala ligulata obtusa lateralia et petala linearia sub-
falcata ; labelli pars basilaris angusta utrinque semicordata paullo
dilatata in partem anticam latissimam obtusangulo subquadratum,
antice medio bilobum ; callus depresso-bidentatus in basi; columna
arcuata apicem versus dilatata.

Hane plantam, eujus locum natalem non dixit auctor, nescio.

40. L. capensis, Lindl. in Ann. Nat. Hist. ii. 1840, p. 314.—
Sturmia capensis, Sonder in Linnea, xix. p. 71.

Semipedalis; pseudobulbus parvus cum caule 1—14-uncialis,
vaginis albis membranaceis laxis tectus; folia bina, late ovata
obtusa vel acutiuscula patula, 1-12 uncias longa, 2-1 unciam lata ;
scapus erectus crassiusculus angulatus, 1-3-uncialis, multiflorus ;
flores parvi congesti; bractez lanceolatz acuminate, pedicellis
squales, j-3-unciales; sepalum posticum angustum lineare ob-
longum, lateralia oblonga obtusa falcatula ; petala angusta linearia
quam lateralia longiora ; labellum quam sepala brevius trilobum,
lobis lateralibus perparvis obtusis, lobus medius oblongus apice
rotundato ; marginibus carnosulis ; columna erecta angusta, apice
curvo, alis brevibus obtusis; capsula oblonga elliptica, 1-uncialis.

South Africa: prope Rondebosch, Bolus no. 4598! Cape
Town, Bolus 4552! Zwellendam, Bowie! Zeyher 3887!

41. L. LoxsErtr, Richard, Mém.Mus. Par. iv.p.52; Spreng. Syst.
Veget. ii. p. 740; Lindl. Gen. & Sp. Orch. Pl. p. 28; Ilustr.
Orch. Pl. Genera, t. i.; Bot. Reg. t. 882; Syme, Eng. Bot. ix.
p. 133, t. 1488.— L. Correana, Spreng. Syst. iii. p. 740.—Ophrys
Loeselii, Linn. Sp. Pl. 947; Eng. Bot. no. 47.—O. liliifolia,
Huds. Fl. Angl. p. 389.—O. paludosa, Muell. in Flora Danica,
t. 877.—Malaxis Loeselii, Swartz, Act. Holm. 1800, p. 235;
Smith, Eng. Flora, iv. p. 48; Willdenow, no. 8.—M. Correana,
Barton, Prodr. Flor. Philad. p. 86; Nuttall, ii. p. 196.—M.
longifolia, Barton, Flor. N. Am. iii. p. 12, t. 75.—Sturmia Loeselii,
Reichb. fil. Ic. Fl. Germ. t. eccexcii.; Te. Bot. emlvi.; Koch, Syn.
Fl. Germ. & Helv. ed. 2, p. 808.

Pseudobulbus ovatus, vaginis albescentibus membranaceis

OF THE GENUS LIPARIS. 278

tectus; folia bina oblonga, lanceolata obtusa vel subacuta flaccida
1-3 uncias longa, 3-1 lata; scapus folia superans acute triqueter,
3-8 rarissime 13-uncialis (specimen Canadense) ; racemus laxus,
flores 3-20 parvi erecti flavescenti-virides; bractez parve lan-
ceolate acute, pedicellis zquales; sepala angusta lineari-lan-
ceolata obtusa 3 lineas longa; petala linearia; labellum obo-
vatum oblongum integrum vel obscure trilobum apiculatum mar-
ginibus plus minus crenulatis ; columna brevis crassiuscula ferme
recta, alis brevibus; anthera ovata obtusa, stigma quadratum;
capsula oblonga elliptica erecta, costis prominulis,ferme 3-uncialis.

Europe, central; Sweden, Ostrogotha, Blytt! Holmgren! Got-
land, Nyman. Norway, Borgeby, Hallbeck. Denmark. Eastern
England: Norfolk, Suffolk, Cambridge, Hants, Kent? Belgium.
Holland. Sehleswig-Holstein, Hansen 281! Ins. Mohn. Germany :
Mecklenburg, Wiistrei!; Hamburg; Bavaria, Schultz, Herb.
Norm. ii. 147, 147 bis; Leipzig, Awerswald! France: Dunkirk,
Woods!; Dessine prés Lyon, Bonnamour in Billot herb. 3238!
Tyrol, Bozen, Grabmayr! Hungary, Epenet, Haslinsky! Switzer-
land, Vevey, Leresche!; Geneva,Meisner! Transylvania. Russia:
Lithuania, Besser!

Eastern N. America: New York, Buffalo, Clinton!; Hastings
County, Lake Ontario, Macoun!; Washington, Falcon Valley,
Suksdorf!; Philadelphia, Barton!; Massachusetts, Cambridge,
Nuttall!; Virginia, Clayton!

42. L. sapontca, Maximowicz, Bull. Ac. Imp. St. Pétersb. xii.
p. 544.—Microstylis japonica, Mig. Prolus. Fl. Jap. p. 135;
Franch. & Sav. Enum. Pl. Jap. ii. p. 21.

Caulis pseudobulbosus sed paullo dilatatus, 13-uncialis ; rhi-
zoma validum et suberosum ; folia bina ovata obtusa, 2-3 uncias
longa, 1i lata ; scapus sexuncialis conspicue alatus ; flores 7-8 laxe
racemosi erecti virescentes ; bractez minute ovate obtuse ; pedi-
celli breves, cum ovariis, 4 unciam longi; sepala lanceolata ligulata
obtusa, petala linearia; labellum obovatum cuneatum ecallosum
breviter cuspidatum marginibus minute crenulatis; columna ferme
cyenicollis gracilis, basi dilatata, alis brevibus latis rotundatis.

Japan: Hakodate, Maximowicz! Mandshuria austr. Radde,
Maximowiez.

43. L. CcAMPYLOSTALIX, Reichb. fil. in Linnea, xli p. 43.

Pseudobulbus tunicatus ; caulis paucivaginatus diphyllus ; folia
euneata oblonga acuta; scapus angulatus folia bene superans race-
mosus; bractex triangule minute ; flores iis L, Loeselit equales

274 MR. H. N. RIDLEY'S MONOGRAPH

sepala linearia lancea; petala linearia; labellum basi columnam
amplexum utrinque angulum retrorsum obtusangulum pentan-
gulum antice retusum ecallosum ; columna bene curva antice
juxta foveam ampliata.

India oriens, locus specialis ignotus.

44. L. SauNDERSIANA, Reichb. fil.in Gard. Chron. 1872,p. 1008.

Pusilla ; caulis brevis pseudobulbosus; pseudobulbus globosus 3-
uncialis; folia bina ovalia cordata acuta quorum unum majus,
unciam longa, $ unciam lata; scapus 2—3-uncialis, purpurascens
breviter alatus debilis; flores parve 8-9 in racemo laxo ; bractew
lanceolate virides, 3 unciam longs, pedicellis purpureis haud
multo breviores; sepala angusta ligulata viridia, petala angustiora
filiformia purpurea; labellum ellipticum oblongum obtusum,
vel minute apiculatum, 4 unciam longum, purpureum ; columna ~
gracilis cycnicollis, basi dilatata; anthera purpurea.

Jamaica, Morris! v. s. et v. cult. in hort. Kew.

45. L. Wzwprannrr, Reichb. fil. Orchid. Wendland. p. 98.

Humilis; caulis 2-1-uncialis, vagin: pseudobulbi externe
suberose ; folium cuneatum late ovatum acutum; racemus cir-
citer 13-florus subsecundus nune quaquaversus pluriflorus;
flores virides; sepala ligulata obtuse acuta; petala linearia; la-
bellum basi utrinque minutissime retrorsum auriculatum, late
ligulatum, antice dilatatum emarginatum, toto limbo minute
serrulata; lineoli callosi terni breves a basin disci; columna
humillima.

Costa Rica, JWendland.

46. L. Gurneanea, Reichb. fil. in Flora, 1867, p. 103.

Pusilla, triuncialis; pseudobulbus minutus globularis; caulis
brevissimus; folia bina lanceolata mucronata patula, 4 unciam
longa, 2 lineas lata ; scapus gracilis angulatus; racemus pauci-
florus : bractez lanceolate acuminate vel ovate, 13 lineam longe ;
flores parvi ex albido pallide flavescentes ; pedicelli erecti 4 lineam
longi; sepalum postieum angustum lineare }-unciale, lateralia
ovata faleatula, labello zquales; petala linearia angusta quam
ovarium longiora ; labellum ovatum apiculatum ; columna gracilis
erecta, apice incurva; ovarium }-unciale oblongum ellipticum.

West Africa: Huilla, Morro de Monino, Welwitsch no. 723!

47. L. vores, Ridl.in Journ. Linn. Soc., Bot. xxi. p. 458, differt
a precedente foliis angustis, pedicellis erectis, bracteis lanceolatis,
2-3 in scapi parte inferiore dissitis; sepalis ovatis falcatis; est
etiam planta omnino major. Madagascar!

OF THE GENUS LIPARIS. 275

48. L. XANTHINA, n. sp.—Pseudobulbus ovalis conicus, vaginis
tectus, superioribus foliaceis ; folia bina lanceolata ovata subob-
tusa; scapus gracilis, bracteis dissitis lanceolatis; racemus multi-
florus laxus; flores circiter 13, parvi, flavi; bracteæ inferiores
pedicelli dimidio æquales, superiores breviores; pedicelli graciles,
subpatuli; sepalum posticum angustum lineare, lateralia ovata
oblonga obtusa; petala angustiora linearia; labellum oblongum
obtusum ; columna erassiuscula parum curva.

Madagascar: Ankafana in arboribus, Deans Cowan!

49. L. parva, Ridl. in Journ. Linn. Soc., Bot. xxi. p. 462.—A L.
xanthina, Ridl., differt floribus olivaceis nee flavis, sepalis laterali-
bus lanceolatis nec ovatis. Planta pusilla Madagascarica.

90. L. GERRARDI, Reichb. fil. in Flora, 1867, p. 118.

Caulis paullo dilatatus uncialis ; folia 3-4 angusta lanceolata
ligulata acuta, basi longe angustata flaccida, 3 uncias longa, 3
unciam lata ; scapus erectus gracilis subteres, exalatus, 6-uncialis ;
racemus laxus pauciflorus; bractew lanceolate acuminate virides,
4 unciam longs; pedicelli longi tenues; sepala ligulata oblonga
subacuta; petala filiformia angusta; labellum obovatum acutum,
callis minutis duobus ad basin.

Natal, Gerrard!

91. L. LONGIPETALA, Ridl. in Journ. Linn. Koc., Bot. xxi. p. 459.

Habitu Z. Gerrardi habet, sed pseudobulbus est distinctus,
sepala lateralia late semicireularia, labellum oblongum obtusum
breve, petala longiora.

Madagascar.

92. L. TRICALLOSA, Reichb. fil. in Gard. Chron. n. s. xiii. p. 684.

Caulis vix pseudobulbosus, breviuseulus; folia 3-5, late lan-
ceolata plicata petiolata; scapus 9-12 uncias longus angulatus ;
racemus laxus multiflorus; sepala lanceolata ligulata obtusa;
petala filiformia; labellum ovatum breviter unguiculatum, lamina
obtusa tenuis marginibus denticulatis, venis rufescentibus, un-
guis brevissima erecta lateribus incrassatis atro-rubris; callus
singulus mamilliformis obtusus albus; columna suberecta, apice
incrassata curva, alis magnis obtusis latis, basi kermesina; pedi-
cellus gracilis longus; bracte minim: lanceolate; capsula erecta
elliptica.

Borneo; Buut Island, Sooloo, Burbidge!

53. L. BRACHYGLOTTIS, Reichb. fil., Trimen,Syst. Cat. Ceyl. p.87.

Planta pusilla, 3-6-uncialis ; caulis biuncialis, basi dilatatus ;
folia 1-2 ovata acuta marginibus crispis, 2 uncias longa, 1 lata ;

276 MR. H. N. RIDLEY S MONOGRAPH

scapus brevis gracilis; flores parvi, laxe racemosi; bracteæ

ovate lanceolate, pedicelli breves graciles; sepalum posticum

lanceolatum viride, lateralia ovata labello supposita, basibus ad-

natis, quam labellum breviora; petala linearia; labellum oblongum

obtusum emarginatum, marginibus paullo incrassatis ; callo unico

lato brevi emarginato ad basin ; columna brevis crassiuscula erecta.
Ceylon, Wattakeele Hill, Thwaites no. 4002!

54. L. sARBATA, Lindl. Orch. Pl. p. 27.

Planta pusilla, 2-4-uncialis ; caulis vix pseudobulbosus vaginis
laxis tectus; folia bina ovata flaccida patentia, 2 uncias longa,
1 lata; seapus brevis angulatus; racemus laxus pauciflorus;
bractee breves aeutz, pedicelli sublongi; flores parvi virides;
sepala lanceolata obtusa recurva; petala anguste linearia ob-
tusa; labellum oblongum cuneatum, basi in medio depresso, vel
potius suleato, hispidum; columna erecta, parum curvula, alis
rectis obtusis, apice dilatata.

Ceylon, Macrae! (specimen singulum in herbario Lindleyano
nec cl. Thwaitesio, unquam reperta).

55. L. BILOBA, Wight, Icones, vi. p. 1633; Reichb. fil. in Walp.
Ann. vi. p. 218.

Pusilla; caulis 2-uncialis basi in pseudobulbo (1-uncialis) dila-
tata ; folia bina vel terna ovata acuta undulata petiolata, unciam
longa, 3-3 unciam lata; scapus nutans, 1j unciam longus, superne
laxe racemosus ; flores parvi purpurei pauci; bractee 1 unciam
longer, lanceolate; sepala lanceolata ovata acuta, lateralia vix
faleata; petala anguste linearia obtusa; labellum ungue brevi,
lamina cuneata biloba, laciniis lanceolatis obtusis ; calli duo;
anthera ovata; columna arcuata, alis parvis obtusis. Planta
omnino purpurascens.

Southern India, Ootacamund, Wight!

56. L. Jovis-Pruvir, Parish § Reichb. f. in Trans. Linn. Soc.
Xxx. p. 155.

Pseudobulbus obconieus, vaginis pallidis demum fibris reductis
tectus, 1-3 uncias longus; folia 3-4 ovata lanceolata acumi-
nata, albo-tessellata, 3 uncias longa, 13 lata; petiolus aliquando
uncialis ; scapus 7-uncialis exalatus, basi nudus ; racemus multi-
florus; flores parvi purpurascentes; bracteew anguste lineares
lanceolate tandem deflexe 3 unciam longs» ; sepala linearia, petala
angustiora filiformia; labellum ad columnam ungue adnatum,
lamina cordata acuta ; columna breviuscula, alis quadratis.

Moulmein, Twakabin, Parish !

OF THE GENUS LIPARIS. 277

97. L. WarnkERLE, Graham in Bot. Mag. t. 3770; Thwaites,
Enum. Pl. Ceyl. p. 295.

Caulis vix pseudobulbosus 23-uncialis, vaginis purpurascenti-
bus tectus; folia terna ovata, 2 uncias longa, 14 lata, petioli
$-unciales ; Scapus gracilis erectus, 33-uncialis, bracteis vacuis 1-2
dissitis, superne laxe racemosus ; flores parvi plures ; sepala ob-
longa acuta purpurea, lateralia vix obliqua, petalis multo latiora ;
petala linearia angusta purpurea; labellum oblongum, suborbi-
culare integrum crenulatum, deflexum viridi-purpureum; calli
duo conici; columna breviuscula crassiuscula, paullulo curva,
alis parvis subacutis.

S. India, Nilgherries, Wight no. 2946! Ceylon, Thwaites
no. 2736!

98. L. MONTANA, Lindl. Orch. Pl. p. 29; Mig. Flor. Ind. Bat.
iii. p. 621.—Malaxis montana, Blume, Bijdr. p. 388.

Caulis biuncialis vix pseudobulbosus, vaginis pallidis membra-
naceis tectus; folia quaterna ovata longe acuminata acuta petio-
lata marginibus crispis, 3 uncias longa, 11 lata, petiolis uncialibus ;
scapus gracilis angulatus 7-uncialis; racemus laxus terminalis ;
flores parvi; bractes anguste lanceolate acuminate 2 lineas
longe quam pedicelli multo breviores ; sepala lanceolata ligulata
obtusa lateralia subobliqua; petala linearia quam sepala lon-
giora; labellum late obovatum, marginibus dentatis serratis,
lateribus ad basin revolutis; callus ad basin obtusus; columna
breviuscula crassiuscula parum curva basi dilatata, alis latis
obtusis hyalinis.

Java, Pangerango mt., Zollinger no. 387 ; Lobb 194! Blume.

59. L. PECTINATA, n. sp.—-Caulis vix pseudobulbosus, vaginis
tectus 2-uncialis; folia pauca ovata acuminata, 34 uncias longa,
13 lata ; scapus subteres exalatus 4-uncialis (vix tamen evolutus) ;
racemus laxus; flores parvuli; sepala lanceolata subobtusa,
similia; petala ligulata acuminata; labellum panduratum mar-
ginibus apicis inflexis pectinatis ; calli duo conici in basin labelli;
columna suberecta parum curva, alis latis obtusis crenulatis.

Philippines, Cuming! Specimen singulum in herbario Lind-
leyano, Kew.

60. L. ATROPURPUREA, Lindl. Orch. Plants, p. 28; Thwaites,
Enum. Plant. Ceyl. p. 295.—L. olivacea, Wight, Icones, t. 903
(haud Lindley).—L. Walkerie, Wight, Ic. t. 905 (haud Graham).
— L. livida, Lindl. in Royle Ill. Herb. Himal. p. 364.

LINN. JOURN.—BOTANY, VOL. XXII. Z

278 MR. H. N. RIDLEY'S MONOGRAPH

Caulis 5-uncialis basi paullo dilatato, vaginis laxis tectus; folia
terna vel quaterna sepius remotiuscula ovata petiolata 9-nervia
crispa, 8-6 uncias longa, 2} lata, petiolo unciali; scapus teres
9-uncialis, bracteis vacuis lanceolatis pluribus munitus; floresin
racemo congesto terminali pauci mediocres; bractes 14 lineas
longs pedicelli dimidio equales ; sepala purpurea anguste oblonga
lanceolata, lateralia vix faleatula, obtusa, quam labellum longiora ;
petala purpurea linearia angustiora; labellum porrectum oblon-
gum ovatum obtusum, marginibus crenulatis; calli breves duo
subconici obtusi ; columna crassiuscula parum curva, alis brevi-
bus obtusis ; anthera ovata obtusa, apice papillis viridibus ornato.

Southern India: Pulney Mountains, Wight; Travancore,
Beddome! Anamallay Mountains, Beddome! Ceylon, Macrae!
Thwaites no. 3181! Gardner no. 846! Maxwell, &c.

Northern India: Khasiya, Lobb! Mussooree, Royle! (v. ic. in
Hb. Kew.).

L. livida, Lindl., differt ut videtur solummodo in labello magis
rotundato et floribus magis congestis.

61. L. Wienttana, Thwaites, Enum. Pl. Ceyl. p. 295.—L.
atropurpurea, Wight, Icones, t. 904, non Lindl.

Caulis brevis vix uncialis, vix pseudobulbosus, vaginis albis
membranaceis laxis tectus; folia pauea 2-3 ovata lanceolata,
margine crispa basi cordata, 1-23 uncias longa, 3-1 lata; petioli
unciales ; scapus erectus angulatus; bracteis paucis parvis dissitis
3-12-uncialis ; flores pauci in apice scapi congesti mediocres, iis
L. nepalensis ferme equantes tenues, viridi-purpurascentes, pedi-
cellis 3-uncialibus ; bractez lanceolate acuminate lineales ; sepa'a
oblonga acuta; petala linearia marginibus revolutis ; labellum
orbiculare vel suborbiculare planum, integrum vel minute retusum,

2 unciam longum, callo lunato ad basin; columna exalata gracilis
arcuata.

Ceylon, Thwaites no 3179!

Southern India ; Courtallum ?
Wight.

62. L. wEPALENSIS, Lindl. Bot. Reg. t. 882; Orch. Pl. p. 28;
Sprengel, Syst. Veg. ii. p. 741.—Malaxis cordifolia, Smith in
Rees’s Cyclop.—Acianthus petiolatus, Don, Prodr. Nep. p. 28.

Rhizoma longe repens, crassiusculum, caulibus sepe longe
remotis vix pseudobulbosis, vaginis 1-2 et petiolorum basibus
dilatatus tectus; folia bina ovata cordata acuminata flaccida,
lamina 34 uncias longa, 25 lata, petiolus biuncialis ; scapus 4-6-
uncialis angulatus; bracteæ plures anguste lanceolate acuminate,

OF THE GENUS LIPARIS. 279

superne comosse, 3 lineas long; ; racemus laxus ; flores mediocres
virescenti-albidi, pedicellis 5 lineas longis; sepala angustata lan-
ceolata linearia acuminata; petala angustiora; labellum por-
rectum oblongum ovatum longe cuspidatum ecallosum ; columna
gracilis arcuata, basi paullo dilatata, alis parvis subacutis.

Nepaul, Sheopore, Wallich no. 1945! Wight no. 109! Dar-
jeeling, E. Bengal, Griffith! Kumaon, Atkinson.

63. L. REPENS, mihi.—Malaxis atropurpurea, Blume, Bijdr.
p. 390.—Platystylis atropurpurea, Lindl. Orch. Pl. p. 18.—
Apatales purpurascens, Blume, MSS. in herb. Brit. Mus.

Rhizoma diu repens, pseudobulbi parvi plures dissiti ovato-
oblongi, semiunciales; folia singula rarius bina, ovata acuta
petiolata, lamina 14 uncias longa, 2 unciam lata, petiolus uncialis ;
scapus folia superans, 4-uncialis, angulatus; bracteæ plures lan-
ceolate acuminate carinate superne comose j-unciales; flores
mediocres pauci in apice scapi congesti atro-purpurei, pedicellis
gracilibus ; sepala lanceolata; petala anguste linearia; labellum
oblongum cuspidatum 2 lineas longum ; columna gracilis arcuata,

alis brevibus.
Java, Lobb no. 193! Mt. Prahu, Horsfield! Mt. Salak, Blume!

64. L. srENoGLOssaA, Parish d Reichb. fil. in Trans. Linn.
Soc. xxx. p. 154.

Caulis vix pseudobulbosus; folia 3-4 lanceolata acuta petio-
lata dissita, 11-3 uncias longa, 1 unciam lata, petiolus 1} uncias
longus ; scapus 4-uncialis, basi nudo, bractea hine inde excepta ;
racemus laxus; flores plures dissiti parvuli purpurascentes ;

bractez lanceolate acuminate 3-unciales; pedicelli }-unciales ;

sepala lanceolata ligulata lateralia, labellum superantia faleatula ;
petala filiformia ; labellum ligulatum bilobum ; columna gracilis
arcuata, alis rotundatis brevibus.

Moulmein, Parish !

Habitus L. purpurascentis.

65. L. purpurascens, Lindl. Bot. Reg. sub t. 882; Orch. Pl.
p- 29; A. Richard, Orch. Maur. p. 583 ; Bojer, Hort. Maur. p. 320.
= Matri purpurascens, Thwaites, Orch. Iles Afr. tt. 26, 27.

Caulis 1-3-uncialis haud pseudobulbosus purpurascens ; folia
bina rarius trina, vix petiolata dissita ovata cordata subacuta
crispa, 2 uncias longa, 1 lata; scapus brevis uncialis vel 1}-
unciales purpurascens angulatus; racemus multiflorus subden-

us; bracteæ lanceolate pedicellis subzequales, } unciam longs ;
z 2

280 MR. H. N. RIDLEY'S MONOGRAPH

flores purpurei parvi; sepala lorata ligulata, lateralia labello

supposita; petala linearia; labellum unguieulatum, ungue brevi

suberecto, lamina cordata quadrata dentata, emarginata cum

apiculo; columna breviuscula subclavata curvula apice incrassata,

alis grandiusculis ; capsula 4 unciam longa conica clavata erecta.
Mauritius, Bouton! Blackburn!

66. L. NEUROGLOSSA, Reichd. fil. Xenia Orchidacea, iii. p. 26.

Rhizoma diu repens, eo L. nepalensis similis sed gracili-
bus, caules erectos, basi vaginis albescentibus membranaceis
ampliatis tectos emittens ; folia singula, cordata orbieularia eris-
pula petiolata; petiolus 14-unciales, scapi basin amplectens,
lamina 14 uncias longa, 14 lata, quo latissima ; scapus debilis
biuneialis angulatus; racemus laxus pauciflorus; flores parvi 4
unciam longi, tenues ; bractes ovate acute acuminate vel lanceo-
late foliaces, quam pedicelli dimidio breviores; sepala linearia
ligulata, petala angustiora; labellum panduratum apice lato
cuspide mediano brevi, venis tribus elevatis purpureis conspicuis ;
columna gracillima arcuata (in sicco aurantiaca), alis nullis ; cap-
sula elliptica oblonga erecta.

Bolivia: inter Laripata et Cani; Queliguaya prope Sorata,
Mandon no. 1136!

67. L. WeLwitscutl, Reichd. fil. in Flora, 1865, p. 184.

Rhizoma repens, caules erectos haud pseudobulbosos 3 uncias
altos emittens; folia 3-4 dissita ovata obtusiuscula petiolata,
lamina 1-1} uncias longa, 14-2 lata; petioli semiunciales canali-
culati vaginantes, vaginis ampliatis membranaceis ; scapus 2-4-
uncialis angulatus purpureus ; bractes cordate ovate membra-
naceæ longe acuminate pedicellis equilonge ; racemus 11-uncialis
densiflorus; flores plures minimi; sepala ovata lanceolata late-
ralia multo latiora ovata oblonga, labello zqualia ; petala linearia
sepalis subzequalia, omnia obtusa atro-purpurea ; labellum breve
cordatum obtusum carnosulum, atro-purpureum margine albido,
ecallosum ; columna parum curva semiteres, crassiuscula exalata,
stigma profundum; anthera obtusa; capsula elliptica circiter
j-uncialis. Habitus ferme Goodyere.

Angola: Golungo Alto, Welwitsch no. 660 !

Ramose.

68. L. CRISPIFOLIA, Reichb. fil. in Linnea, xli. p. 43.
Caulis repens, folia dissita ovata subacuta, crispa petiolata,

OF THE GENUS LIPARIS. 281

lamina 2 uncias longa, l} lata, petiolus 2-uncialis; scapus
erectus brevis, triuncialis, racemus pluriflorus ; flores magni
speciosi; braetez lanceolate acuminate, $ pedicelli æquales ;
sepala et petala anguste linearia acuminata reflexa semiuncialia
verosimiliter viridia, petala angustiora; labellum $ unciam
longum, 2 latum, rhomboidale acuminatum, atro-purpureum, venis
multis obscurioribus, ecallosum ; columna gracilis cycnicollis, alis
brevibus obtusis.

Peru, Jameson!

69. L. BRACHYSTALIX, Reichb. fil. in Linnea, xli. p. 43.

Caulescens, caulis teres undique foliatus 5-6-uncialis; folia
copiosa ovata petiolata subdissita, petiolis vaginantibus brevi-
bus, laminis 9 lineas longis, 5 latis, marginibus crispis; racemi
multiflori laxi laterales ex axilla foliorum superiorum orientes 43-
uncialis, rhachide subtereti ; bracte:x lanceolate foliacee crispe,
ovariis subzquales; sepala et petala linearia ligulata reflexa an-
gusta “ brunnea,” petala angustiora; labellum oblongo-cordatum
retusum cuspidatum, 4 lineas longum, viride, linea mediana ob-
scuriore ; eolumna brevissima crassiuscula erecta quam unguis
labelli brevior, alis nullis.

Ecuador: in valle Lloensi ad arborum truncos 9000 ped. alt.
No. 449, Jameson!

70. L. naMosa, Poeppig d Endlicher, Nov. Gen. et Sp. ii.
p ut ux

Rhizoma diu repens, crassiusculum bipedale, ramosum ; folia
remota, secunda (internodiis uncialibus) erecta, petiolata, pe-
tiolus 4 unciam longa, lamina 1} uncias longa 1 unciam lata, quo
latissima, ovata cordata subacuta; scapus subterminalis, 6-
uncialis, ferme omnino racemosus ; flores “albi,” plures; sepala
ligulata, 4 unciam longa; petala angustiora linearia subsqualia,
omnia reflexa viridia (ex Poeppig); labellum porrectum, 1 unciam
longum, ellipticum obeuneatum apiculatum ; columna brevissima,
arcuata, subgracilis, 1 labello :equalis.

Venezuela: Tovar, Fendler no. 1422! Tolina prope Oyuear,
Goudot! Peru: Pampayaco, Poeppig.

Valde affinis Z. brachystalice, Reichb. f., sed omnino maior.

71. L. Wacenert, Reichb. fil. in Bonplandia, iii. p. 78.

Folia violacea plantaginea geminata subpetiolata, lamina ovalis
acuta basi cuneata rotundata, petiolus lamina æqualis vel brevior ;
racemus spithameus basi nudus; flores iis Z. /Jatifolie squales,

282 MR. H. N. RIDLEY’S MONOGRAPH

albi, sepala et petala linearia; labellum cuneatum flabellatum
margine denticulato ; bractez quam ovario ter breviores.
Colombia: San Pedro, Ocana, Wagener (non vidi).

Species hujus sectionis haud satis note.

L. anata, Scheidweiler, in Otto § Dietr. Allgem. Gartenz.
1842, p. 293; Lindl. Bot. Reg. 1843, Misc. no. 12 (non Richard).

Folia 6, oblongo-ovata acuta petiolata, plicata, quam spica
erecta multiflora breviora; scapus erectus squamosus 6-uncialis
purpureus, alis 6 membranaceis; sepala et petala linearia; labellum
mucronatum revolutum atro-purpureum ; columna viridis; flores
Y unciam longer, purpurea, labello kermesino.

Mexieo, culta ad Lacken a rege Belgiorum.

L. rRADEsCANTIFOLIA, Lindl. Orch. Pl. p. 29.—Malaxis trades-
eantifolia, Blume, Bijdr. p. 389.

Bulbi compressiusculi vaginati monophylli; folio ovata dis-
colora membranacea; bractez: minute ovate; labelli limbus
ovatus integerrimus ; sepala subeequalia patentia ; labellum sepalis
paullo latius basi callosum ; anthere loculi subdisjuncti.

Java, in umbrosis Mt. Salak, October, Blume.

Anne Microstylis commelynifolia, Reichb. f. ?

$ CoRIIFOLIZ.

Herbe gerontoge:s tropice plerumque epiphytice rarius rupi-
cole, pseudobulbis incrassatis, foliis paucis lanceolatis spathulatis,
marcescentibus caducis, quasi pseudobulbis articulatis.

72. L. GRANDIFLORA, Ridl. Journ. Bot. xxii. 1884, p. 333.

Pseudobulbus adhue deest,....; folium singulum magnum,
8 uncias longum, 2} latum, ovatum lanceolatum costatum ; scapus
crassiusculus teres pedalis, pro majore parte laxe racemosus ; flores
magni remoti ; pedicellis gracilibus uncialibus ; bractew lanceolate
acute vix > pedicellorum zquales ; petala et sepala ligulata lanceo-
lata obtusa angusta, viridescentia, petala angustiora ; labellum cu-
neatum obovatum trilobum, crenulatum, marginibus pubescenti-
bus ł unciam longum, et quo latissimum 4 unciam latum, ochrea
linea mediana obscuriore; columna curva gracilis basi paullo
dilatata, alis brevibus rotundatis latis; anthera obscure viridis
depressa obtusa.

Borneo, Mindai-Pramassan, Grabowsky ! Everett!

Flores in genere maximi.

OF THE GENUS LIPARIS. 283

73. L. cnEzNULATA, Lindl. Gen. & Sp. Orch. Pl. p. 80; Mig.
Fl. Ind. Bat. ii. p. 623.—Malaxis crenulata, Blume, Bijdr.
p. 393.

Pseudobulbus 2—3-uncialis eylindrico-conicus ; folium singulum
vel 2, lanceolatum acutum, 12-costatum, 8-10 uncias longum,11-22
latum; scapus teres, ferme omnino racemosus 7-8 uncias longus;
flores mediocres * obscure aurantiaci"' dissiti; pedicelli unciales,
patuli; bractez lanceolate foliacee 4 unciam longs ; sepala lan-
ceolata obtusa ; petala subsimilia angustiora; labellum 5 lineas
longum et latum, cordatum obovatum bilobum, lobis crenulatis ;
columna arcuata gracilis, alis brevibus obtusis.

Java, Lobb 307! Gede, Zollinger no. 3122! Junghuhn no. 295 ;
Blume.

Var. SUMATRANA.—Bractee lanceolate triangulares acuminate
breviores; pedicelli breviores, callus parvus ad basin labelli.

Sumatra, Forbes 3102! in Herb. Brit. Mus.

74. L. LATIFOLIA, Lindl. Gen. 4 Sp. Orch. Pl. p. 30; Reichb.
Jil. in Gard. Chron. 1885, April 25, p. 532; Miq. Fl. Ind. Bat.
iii. p. 623.

Pseudobulbus pyriformis 14-34 uncias longus, radicibus gra-
cilibus ; folium singulum lanceolatum acutum, 6-8 uncias longum,
$-1 unciam latum, erectum spe scapum superans; scapus multi-
florus 6—7-uncialis; racemus laxiuseulus ; “ flores aurantiace ; "
bractez lanceolate acute pallide pedicellos superantes superne
comosæ, 4-unciales ; sepala elliptica ovata lanceolata obtusa;
petala linearia obtusa; labellum euneatum bilobum, marginibus
dentatis; columna gracilis areuata, alis obtusis; anthera ovata
rotundata obtusa ; capsula ovata vix 1 unciam longa pyriformis.

Java: Tankabar, 410 F, Salak, 3117, Zollinger! 304, Junghuhn!
1079. S.E. Java, H. O. Forbes! Lobb 170!

75. L. ForBeEstt, n. sp.— Pseudobulbus brevis conicus 1-uncialis,
vaginis membranaceis laxis tectus, superioribus foliaceis; folium
singulum exacte lanceolatum acutum petiolatum semipedale,
ferme unciam latum ; scapus gracilis angulatus, anguste sed con-
Spieue alatus, S-uncialis ; racemus laxus circiter 12-florus ;
bracteæ breves lanceolate acuminate 1 lineam longs»; pedicelli
patentes 6 lineas longi; flores quam iis latifoliz: minores ; sepala
lanceolata ligulata obtusa, lateralia falcatula ; petala linearia
angustissima ; labellum late obovatum, oblongum emarginatum,
callis duobis parvis hemisphericis ad basin; columna brevis

284 MR. H. N. RIDLEY'S MONOGRAPH

erassiuseula parum curva, alis linearibus ligulatis obtusis ; costs
ovarii sinuate ; capsule pyriformes j-unciales nutantes.
S.E. Java, 1153a H. O. Forbes! Herb. Brit. Mus.

76. L. CHLOROXANTHA, Hance in Journ. Bot. 1883, p. 231.

Planta 8-pollicaris, radicibus dense pilosis ; pseudobulbi ovoidei
teretes 9 lineas longi, vaginis 4 tecti; folia bina lineari-lanceolata
acuminati 6 polliees longa, 9 lineas lata, 5-nervia, nervo medio
subtus prominulo ; scapus valde compressus anceps ; bractez bifariz
lineari-subulate acutissime, inferiores ad 8 poll. (sic, potius lin. ?)
longs, florales 3 lineales; racemus 15—20-florus, 4 poll. longus ;
flores 4 lin. longis adhue clausi virides, expansi luteoli ; sepala
lateralia oblonga obtusa recurva, posticum filiforme patens;
petala filiformia patentia; labellum late depresso-orbiculatum
arcte revolutum, basi bitubereulatum, margine superiore denticu-
lato ipso apice ex emarginatura dense obtuso terminato; columna
margine edentato.

Hongkong, Mount Parker, et terra firma, Ford: non vidi.

77. L. LACERATA, n. sp.— Pseudobulbi aggregati oblongi conici ;
folia bina, ligulata spathulata acuta 6 uncias longa, 1 unciam lata ;
scapus longior crassiusculus laxe racemosus ; flores mediocres
dissiti; braetez breves lanceolate; sepala et petala ochroleuca,
lanceolata ligulata; petala angustiora; labellum ungue brevis,
lamina oblonga quadrata deflexa apice laciniata, laciniis circiter
10 aeutis, ochroleuca, macula mediana aurantiaca; columna
gracilis curvula, alis parvis.

Borneo: Lawas River, Burbidge (v. icon.) in Herb. Brit. Mus.

Planta hee habitum Z. reflexe habet, sed foliis binis labello
apice laciniato facile distinguenda.

78. L. Beccarit, n. sp.—Rhizoma diu repens; pseudobulbi
valde (3 uncias) dissiti, baculiformes recti vix incrassati, bi-
unciales ; folia dissita 3, lanceolata obtusa basi angustata cos-
tata, 7 uncias longa, $ unciam lata; scapus subteres 9-uncialis, basi
bracteis paucis dissitis exceptis nudus; racemus laxus circiter
l9-florus; flores mediocres, pedicellis gracilibus j-uncialibus ;
bractez lanceoiatæ lineares, unciam longs; sepala lanceolata
subobtusa 2 unciam longa, lateralia labello supposita, latiora sub-
falcata ; petala linearia filiformia; labellum ungue erecto, lamina
oblonga elliptica, biloba, lobis rotundatis quadratis, minute denti-
eulatis; eolumna gracilis arcuata, alis brevibus.

Sumatra: Mount Singalan, Beccari no. 380! in Herb. Kew.

OF THE GENUS LIPARIS. 285

79. L. DIODON, Reichb. fil. in Linnea, xli. p. 43.

Folium longe spathulatum carnosum membranaceum oblongum
obtusum ; scapus angulatus longe nudus, apice racemosus ;
bractez minute ; sepala linearia; petala filiformia; labellum basi
cuneatum utrinque angulatum dein dilatatum subquadratum
antice crenulatum cum apiculo; columna gracilis arcuata basi
bidentata ; anthera mutica.

India, Dehra Doon, cult., Wilson Saunders.

80. L. pricata, Franchet, Enum. Pl. Jap. vol. ii. pp. 22, 509;
So Mokou Soussetz, vol. xviii. p. 22.

Pseudobulbus oblongus conicus ferme uncialis, vaginis mem-
branaceis 4 laxis tectus, superiore acuminata; folium singulum
lanceolatum acutum, 5 uncias longum, 1 latum, costis tribus con-
spieuis; scapus erectus 6-uncialis vel paullo ultra, complanatus
conspicue alatus, superne racemosus; bractez lanceolate acumi-
natæ, pedicellis ferme æquantes ; flores mediocres dissiti erecti
vel patentes; sepala lanceolata obtusa vix faleatula; petala
linearia angusta; labellum unguiculatum ecallosum, lamina cu-
neata obovata, paullisper retusa, nonnunquam cuspide mediano
minuto; columna crassiuscula erecta, alis brevibus latis sub-
quadratis.

Japan, cult.in hortis Yedoensibus, Franchet! Hongkong, 521
Fortune ? herb. Lindl.!

81. L. pooTANENSE, Griffith, Ic. Pl. celxxxvii.; Notule, iii.
p. 278; Itinerary Notes, ii. p. 98.

Rhizoma sæpe diu persistens, pseudobulbi approximati, vaginis
viridibus tecti, 1} uncias longi; folia singula rarius bina lanceo-
lata acuminata plicata, sepius basi longe angustata, 6 uncias
longa, 2 lata; scapus gracilis alatus, 9-uncialis ; racemus laxus;
flores remoti, mediocres, pallide flavo-virides et ochrei ; bracteæ
lineari-lanceolat acuminate, inferiores 3, superiores vix j unciam
longe ; pedicelli 4-unciales patentes; sepala lanceolata ligulata
angusta; petala filiformia; labellum cuneatum retusiusculum,
basi unguiculatum ; columna gracilis cyenicollis, alis triangulis.

Bootan, Durunga, Griffith 1460! Churra, Pomrang, Hooker
& Thomson, Liparis, 18 !

82. L. Grrrrrrutt, n. sp.—Pseudobulbus sub anthesin vix dis-
tinetus, foliis vaginantibus 4 tectus ; folia evoluta 1-2 spathulata
lanceolata ad basin angustata plicata, 6 uncias longa, 1j lata;

286 MR. H. N. RIDLEY'8 MONOGRAPH

scapus brevior validulus, late alatus ; racemus brevis compactus ;
flores parvi plures; bractex lanceolate acuminate foliacee, pedi-
cellis equales; sepala lanceolata, lateralia labello supposita vix
falcata; petala linearia; labellum integrum ovatum cuneatum
crenulatum ; columna gracilis arcuata.

Bootan ? Griffith 5069! in Herb. Kew.

Habitus Z. plicate, sed flores multo minores congesti, pedicellis
brevibus.

83. L. Mannu, Zeichb. fil. in Flora, 1872, p. 275.

Dense esspitosa; pseudobulbus plurivaginatus vagina summa
elongata acuminata, ipse cylindraceus vix pollicaris monophyllus ;
folium chartaceum lineari-lanceolato-acuminatum; pedunculus
anceps apice longius ac minus racemoso ; bractew setacez ovaria
pedicellata infima superantes, summa longe non æquantes ; sepala
linearia ligulata obtusa acuta; petala angustiora; labelli trifidi
lacini: laterales semifalcate, isthmo brevissimo angusto, lacinia
antica transverso ovata antice crenulata ; anthera acuta; columna
utrinque superne ampliata.

Assam, Obres, Mann: non vidi.

84. L. c&Loaxvorpxs, F. Muell. Fragm. Fl. Austral. ii. p. 71;
Benth. Fl. Austral. vi. p. 273.

Pseudobulbi conici ovoidei approximati parvi, semiunciales,
vaginis pallidis tecti; folia erecta 1-2, circiter 3-uncialia, raro
pedalia angusta lanceolata spathulata acuminata et versus basin
angustata, j unciam lata vel ultra, multinervia ; scapus pedalis vel
brevior anceps gracilis, aliquando subflexuosus ; bractes virides
lanceolate acute pedicellis tenuibus squales; racemus laxus,
pauciflorus ; flores parvi tenues ; sepala lanceolata acuta; petala
linearia; labellum sepalis æquale membranaceum porrectum secu-
riforme obtusum crenulatum flavescens; columna semiteres in-
curva; capsula clavata ellipsoidea.

Australia: Clarence River, Beckler !

85. L. PLANTAGINEA, Lindl. Orch. Pl. p. 29.—L. selligera,
Reichb. fil. in Linnea, xli. p. 42.—1L. orbicularis, Lodd., ex
Hemsley in Gard. Chron. n. s. xvi. (1881) p. 592.

Pseudobulbus oblongus cylindrieus 2 uncias longus, vaginis
lanceolatis acutis foliaceis tectus; folia evoluta bina, lanceolata
plicata acuta cuspidata, iis L. plicate similia, 4-6 uncias longa,
1-13 lata; scapus angulatus exalatus 8 uncias longus; racemus
laxus, flores dissiti plures suberecti magni virides ; bractes lan-

OF THE GENUS LIPARIS. 287

ceolatæ acuminate latiuscule, 3—-}-unciales pedicellis æquales :
sepala lanceolata ligulata, lateralia labello supposita; petala
linearia patula; labellum latum integrum planum viride cuneatum
rotundatum, marginibus obscure crenulatis, basi bicallosum, $
unciam longa et lata; columna suberecta longiuscula gracilis, alis
latis brevibus paullisper decurvis ; cost: ovarii sinuose.

India, locum specialem nemo dicit: v. sc. in hort. Kew. 1863, et
icon. a Catheartio et Royleo pictam in Herb. Kew.; sine dubio
hee est planta sub nomine Cymbidi plantaginei, Wallich, in
Icones Mus. Angl. Ind. no. 633 depicta, etsi specimina in herbario
Wallichiano desunt.

86. L. nkrLEXa, Lindl. Bot. Reg. t. 882; Heichb. fil. Beitr.
Syst. Pflanzenk. p. 46; Benth. Fl. Austral. vi. p. 272.—
L. foliosa, Lindl. Bot. Reg. t. 882 ; Lodd. Bot. Cab. t. 1097 ; Hooker,
Bot. Mag. t. 2709; A. Rich. Orch. Maur. p. 52; Lindl. Orch.
Pl. p. 29.—L. decursiva, Reichb. fil. in Gard. Chron. 1884,
July 12, p. 38.—Cymbidium reflexum, R. Br. Prod. p. 331.

Rupicola; pseudobulbus uncialis latus ovatus, aliquando plures
seriatim aggregati; folia vaginantia 3-4, evoluta bina terna
oblonga lanceolata acuta submucronulata erecta, 5 uncias longa,
1 unciam lata; scapus 4—6-uncialis alatus ; racemus laxus 16-20
florus ; bracteæ lanceolatz:e acuminatz i-unciales, ovarii dimidio
zquantes; flores mediocres ; sepala lanceolata obtusa angusta;
petala lineari-lanceolata subacuta, omnia recurva viridia ; labellum
ungue angustato, lamina oblonga obtusa deflexa, apice ssepius
paullo emarginata, viride vel flavidulum, macula ochrea in medio ;
columna crassiuscula, apice curva, alis parvis; anthera ovata;
capsula elliptica 5 lineas longa, pedicellus 4-uncialis.

Australia: Port Jackson, A. Brown, 5506!; North Rocks,
Port Jackson, Caley!

Speeimen optimum vivum, quod aecepi a cl. Moore, Hort.
Glasnevin, 1885, sub nomine Z. decursiva, Ind. or.! est omnino

L. reflexa, R. Br.

Var. B. cuNEILABRIS.— L. cuneilabris, F. Muell. Fragm. FI.
Austral. iv. p. 164; Benth. Fl. Austral. vi. p. 273.—Sturmia
angustilabris, F. Muell. l.c.

Flores paullo minores; folia acutiora et magis acuminata;
bracteæ breviores; alabastra longiora et angustiora; calli duo
conici ad basin labelli.

Queensland : Rockingham Bay, Dallachy ! v. v. cult. Hort. Kew.

288 MR. H. N. RIDLEY’S MONOGRAPH

87. L.pacuypus, Parish & Reichb. fil. in Trans. Linn. Soc. Xxx.
p. 155.

Pseudobulbi crassi tumidi unciam alti circiter 3 in diametro,
pyriformes; folia singula lanceolata acuta, 2 uncias longa, 3
unciam lata; scapus angulatus, paullisper alatus vix 3-uncialis,
ferme omnino laxiuscule racemosus ; bracteæ lanceolate 3 unciam
longs, quam pedicelli paullo breviores; flores parvuli virides ;
sepala anguste ligulata; petala filiformia; labellum cuneatum,
subbilobum crenulatum basi bieallosum ; columna gracilis sub-
erecta, alis obscuris.

Moulmein, Parish!

88. L. erossa, Reichb. fil. in Gard. Chron. 1883, vol. i. p. 27.

Pseudobulbus brevis erassus pyriformis diphyllus ; folia ampla
lato ligulata obtusa apicibus inequalibus ; racemus sursum pluri-
florus, rhachide depressa ima basi vagina nunc acuta foliacea com-
pressa, sursum vaginis acuminatis scariosis pluribus, flores brun-
nescentes flavi; bractee lineari-euspidatz, ovariis pedicellatis
vix equales ; sepala ligulata acuta, petala linearia ; labellum ima
basi minute utrinque obtuse auriculata ligulata medio abrupte
constricto antrorsum divergenti bilobo, lobis lateralibus qued-
ratis extrorsis margine externo denticulatis; calli obscuri sub-
evanidi gemini in basi; columna albo-virescens.

Burma, introd., H. Low.

Affinis L. pachypode, sed multo major, foliis latissimis obtusis
et labello diverso.

89. L. Bisrrtara, Parish & Reichb. fil. in Trans. Linn. Soc.
xxx. p. 155.

Pseudobulbi elongati subcylindrici 2 uncias longi; radices
copios: graciles: folia bina patentia elliptica oblonga obtusa
2 uncias longa, 1 lata; scapus gracilis angulata exalata, 3 uncias
longa; racemus laxiuseulus multiflorus; bractes lanceolate pal-
lid: pedicellis quales; flores parvuli virides lineis duabus brun-
neis ad basin labelli; sepala ligulata angusta; petala linearia
filiformia ; labellum lineari-panduratum apice retuso denticulato ;
columna cyenicollis, alis parvis.

Moulmein, Parish !

90. L. Hooxert, n. sp.—Pseudobulbi oblongi ovati crassi
$unciam longi; folia bina spathulata lanceolata acuta, multinervia,
3 uncias longa, 1 unciam lata; scapus debilis (aliquando crassius-
cula) complanatus alatus, 43 uncias longus; flores plures, resu-

OF THE GENUS LIPARIS. 289

pinati iis L. bistriate majores ; bracteæ lanceolatz acute, 1 unciam
longer, pedicellis subæquales ; sepala lanceolata ligulata; petala
linearia angustiora; labellum unguiculatum, lamina deflexa ob-
longa subquadrata obtusissima; columna gracilis cyenicollis sub-
clavata basi dilatata.

Khasiya, Mamloo (Liparis 17), J. D. Hooker! in Herb. Kew.

Affinis L. bistriate, sed pseudobulbo diverso, labello integro
oblongo.

91. L. VIRIDIFLORA, Lindl. Orch. Pl. p. 31; Thwaites, Enum.
Pl. Ceyl. p. 295.—L. elliptica, Wight, Icones, v.1735.—L. Wigbtii,
Reichb. fil. in Walp. Ann. vi. p. 218.—Malaxis viridiflora, Blume,
Bijdr. p. 392, t. 54.

Pseudobulbi ovato-globosi unciales aggregati; folia bina ellip-
tica spathulata, 2 uncias longa ł unciam lata; scapus angulatus,
6-8 uncias longus ; racemus subdensus subsecundus, flores parvi
inaperti 40-20; bractes lanceolate acuminate, floribus sub-
zquales ; sepalum posticum ovatum lanceolatum, lateralia lan-
ceolata erecta; petala subzqualia linearia obtusa; labellum
ovatum cucullatum acutum marginibus crispulis ; columna sub-
erecta, alis parvis obtusis.

Nilgherries, Wight. Upper Assam! sine collectore, in Herb.
Kew. Ceylon, Macrae! Hautane, Champion, Thwaites no. 3178.

Java, Blume, Lobb!

Planta hsc habitum Z. pachypodis habet, sed est insignis ob
flores inapertos verosimiliter cleistogamos.

92. L. LuTEOLA, Lindl. in Wall. Cat. no. 1944; Orch. Pl.
p. 32; Griffith, Notule, in. p. 277.

Cæspitosa ; pseudobulbi parvi subglobosi, seriatim approximati ;
folia bina lineari-lanceolata acuta mucronulata angusta 2 uncias
longa vix 3 unciam lata ; scapus alatus, 4-uncialis ; racemus pauci-
florus laxus sæpe flexuosus ; bractez lanceolate acuminate ; flores
parvi dissiti ochracei-flavi; sepalum posticum lanceolatum ob-
tusum lateralia latiora falcatula, labello supposita ; petala linearia
deflexa; labellum oblongum obovatum obtusum, callis conicis
duabus ad basin ; columna gracilis arcuata, basi et apice dilatatis ;
capsula pyriformis globosa, 4 unciam longa, pedicello æqualis.

India: E. Bengal, Churra Pungee, Nullah, Griffith!; Mt.
Pundooah, Wallich ! no. 1944; Assam, Griffith!; Khasiya, Lobb ,

93. L. n&voLvTA, Hooker, Bot. of Beechey's Voyage, p. 70, t. 16.
Pseudobulbi semiunciales, oblongi conici, seriatim approai-

290 MR. H. N. RIDLEY'S MONOGRAPH

mati, vaginis laxis tecti; folia singula ensiformia lanceolata
spathulata obtusa, 23 uncias longa, 3 lata; scapus 4-uncialis inferne
late alatus superne ferme teres, bracteis vacuis dissitis paucis ;
racemus laxus multiflorus ; flores parvi; bractez lanceolate longe
acuminate ovariis ferme :equantes ; sepala lanceolata acuta;
petala linearia angustiora ; labellum ovatum integrum subaeutum,
lamina deflexa; columna crassiuscula curvula exalata; anthera
ovata; capsula ovato-globosa, cum pedicello 4 unciam longo.
Society Islands, Tahiti, Collie !

94. L. cxsPrTOsA, Lindl. Bot. Reg. sub t. 882; Gen. d Sp.
Orch. Pl. p. 32; Richard, Orch. Maur. p. 53; Boj. Hort. Maur.
p. 321; S. Moore, Flor. Maur. § Seych. p. 342.—L. angustifolia,
Lindl. Gen. & Sp. Orch. Pl. i. p. 81; Thwaites, Enum. Pl. Ceyl.
p. 296.—Malaxis cespitosa, Thouars, Orch. Iles d Afr. t. 90.—M.
angustifolia, Blume, Bijdr. p. 393; Mig. Fl. Ind. Bat. iii. p. 622.

Cespitosa ; pseudobulbi 1 unciam longi, seriatim approximati,
vaginis pluribus laxis ; folia 1-3, scapis subequalia, 13-33 uncias
longa, j-;3; lata, lanceolata subspathulata acuta mucronata;
scapus compressus alatus, 2—3-uncialis; racemus laxiusculus
pluriflorus ; flores virescentes minuti; bractes lineares subulate
flores superantes; sepala ovata lanceolata reflexa; petala an-
guste linearia reflexa; labellum oblongum obtusum, margine
retuso dente minimo, nervi duo ad basin incrassati; columna
arcuata gracilis, alis parvis obtusis; capsula parva ovata.

Mauritius, Bojer! Bouton. Bourbon, Bernier! Madagascar,
Deans Cowan!

Ceylon, Thwaites no. 2351! Gardner 847!

Java, H. O. Forbes no. 914! Mt. Salak and Gede, Blume.

Sumatra, H. O. Forbes no. 2574 6, pars!

Platystylis. — Pusille, cespitose; pseudobulbi parvi angusti;
folia subcoriacea ; scapus teres vel valde compressus alatus
flexuosus ; flores parvi resupinati ; labellum oblongum ungui-
culatum, callis nullis, venis ad basin lamins incrassatis ;
columna recta, brevis, lata, alis 2 vel 4.

95. L. RESUPINATA, n. sp.—Pseudobulbi parvi 4 unciam longi
angusta, conico-cylindrica; folia dissita lanceolata acuminata
versus basin angustata, 3 uncias longa, } lata; scapus 44-uncialis,
gracilis, ferme omnino laxe racemosus, subteres ; bractez angustx
lanceolat# acuminate 2 lineas long»; flores parvi plures, flavo-
virides, pedicellis ferme 4 lineas longs; sepalum posticum lan-

OF THE GENUS LIPARIS. 291

ceolatum obtusum marginibus minute papillosis, lateralia lauceo-
lata obtuso-deflexa, marginibus convolutis; petala angusta linearia
deflexa; labellum ungue erecto, lamina oblonga deflexa, obtusa
quam sepala lateralia multo brevius; columna subrecta brevius-
cula parum curva, alis duabus grandiuseulis.

India, Darjeeling, Khasiya, Mumbree, Griffith! Lobb 122!
Sikkim, Hooker! Neilgherries, Wight!

96. L. AURITA, Ridl. in Append. to Forbes’s Naturalist's Wan-
derings in Eastern Archip. p. 518.

Pseudobulbi parvi ovati approximati vix $ unciam longi; folia
linearia lanceolata subacuta basi angustata; scapus gracilis 44-
uncialis teres, ferme omnino laxe racemosus; bractee ovate
acute; flores parvi plures, aurantiaci; sepala obtusa lanceolata,
petala linearia; labellum oblongum cuneatum apice trilobo, lobis
Subeequalibus brevibus angustis obtusis, lateralibus convolutis,
venis duabus ad basin incrassatis costam hippocrepiformem for-
mans; columna crassiuscula recta, alis brevibus latis obtusis,
duabus multo minoribus extus; anthera ovata conico-depressa ;
capsula i-uncialis subglobosa, pedicello ferme } unciam longo.

Timor, Bibicucu, no. 3714, H. O. Forbes!

97. L. DECURRENS, Reichb. fil.—Platystylis decurrens, Lindl.
Orch. Pl. p. 18.—P. alata, Blume, MS. in herb. Brit. Mus.—
Malaxis decurrens, Blume, Bijdr. p. 390, t. 54.

Pseudobulbi cespitosi, tenues vix incrassati, ferme cylindriei ;
folia bina lanceolata spathulata acuta versus basin angustata vix
coriacea unciam longa, 3 lata; scapi vix 6-unciales, graciles com-
pressi, flexuosi late alati; flores pauci dissiti, parvi resupinati ;
bractew lanceolate acuminate; sepala ovata obtusa, posticum
erectum, marginibus omnium minutissime ciliata vel papillosa ;
petala filiformia ciliata; labelli ungue erecto, lamina oblonga
cuspidata abrupte deflexa, lateribus ad basin laminsz elevatis
tortisque; columna crassa, alis 4 carnosulis obtusis.

Java, Mt. Salak, Blume! Zollinger!

Khasiya, Lobb!

Distiche. Racemi breves pauciflori, bractez distichw
majuscule; flores tenues mediocres.

98. L. compressa, Lindl. Gen. & Sp. Orch. Pl. p. 32; Miq.
Flor. Ind. Bat. iii. p. 621.—Malaxis compressa, Blume, Bijdr.
p- 390, t. 54.

292 MR. H. N. RIDLEY'S MONOGRAPH

Pseudobulbus 2-23 uncias longus, compressus; folium singu-
lum, ensiforme acutum, carinatum, coriaceum 9 uncias longum,
1} latum; scapus compressus semipedalis gracilis alatus; race-
mus nutans 2-uncialis; braeteze remotiuscule; flores rubri “au-
rantiaci" ; sepalalanceolata unciam longa ; petala angustissima
linearia ; labellum obovatum suborbiculare, 4 lineas longum, et
quo latissimum equilatum, papillis duabus ad basin; columna
breviuscula crassiuscula arcuata, basi et apice incrassata.

Java, Lobb no. 168!; Buitenzorg, Bantam, Tjanjor, Blume.
Salak, Zollinger 680. Sumatra, Mt. Singalan, Beccari!

99. L. CuuiNarr, n. sp.—Pseudobulbus adhue abest; folium
lanceolatum obtusum, apice mucronatum, 5 uncias longum, 2latum ;
scapus 3-42 uncias longus, gracilis compressus alatus; bracteæ
approximate lanceolate carinatex ; flores tenuissimi, verosimiliter
virescentes ; sepala angusta lmearia lanceolata, 3 lineas longa;
petala angustissima linearia lanceolata paullo longiora; labellum
obovatum obtusum margine minutissime crenulatum, papillis
duabus brevibus obtusis in basin; columna gracilis arcuata, alis
brevibus latis obtusis quasi quadratis ; anthera ovata lata, apice
obtuso; capsula elliptica 1-uncialis; pedicellus 4 unciam longus,
bractez zequalis.

Malacea, Cuming no. 2141! in Herb. Brit. Mus.

100. L. pisrICHa, Lindl. Bot. Heg. sub t. 882; A. Rich.
Orch. Maur. p. 54; Mig. Flor. Ind. Bat. iii. p. 623 ; Bojer,
Hort. Maur. p. 321; Decne. Ann. N. H. Mus. iii. p. 366.—
L. gregaria, Lindl. Gen. & Sp. Orch. Pl. p. 33 ; Moore in Bak..
Fl. Maur. p. 842; Thwaites, Enum. Pl. Ceyl. p. 296.—Malaxis
disticha, Thouars, Orch. Iles Afr. t. 89.—Stelis micrantha, Sieb.
Herb. Maur, ii. no. 168.—L. mucronata, Lindley, Orch. Pl. p. 32.
—Malaxis mucronata, Bl. Bijdr. p. 391.

Rhizoma diu repens, pseudobulbi pyriformes remoti ; folia sin-
gula, 8 uncias longa vix $ unciam lata, lanceolata angusta acuta ;
scapus 6-10-uncialis tenuis anceps alatus ; racemus 1-2-uncialis ;
flores tenues aurantiacl; bractee ovate acuminate carinatse
ovarii dimidio s: quales; sepala lanceolata acuta; petala linearia
subeequilonga; labellum obovatum emargiuatum, cuspide medio
brevi, marginibus ad basin inflexis; callus verrucosus; columna
brevis antice complanata lata recta marginata paullisper curva,
alis distinctis nullis.

Mauritius, Sieber! Wallich! Macgregor Bourbon, Thouars.

OF THE GENUS LIPARIS. 293

Ceylon, Gardner, Macrae! Thwaites no. 2359! Moon! 467,
Mrs. Walker.

Moulmein, Parish !

Philippines, Cuming no. 2099!

Timor, fide Decaisne.

Densiflore. Pseudobulbi conico-cylindrici vel baculiformes ;
vaginis paucis membranaceis ad basin tecti; folia suboppo-
sita, lanceolata vel spathulata, sæpius bina; racemus densi-
florus, flores minimi virescentes, petala et sepala subsimilia,
patentia vel recurva; labellum nec unguiculatum nec
callosum.

101. L. rowaiPEs, Lindl. in Wallick’ s Plant. Asiat. Rar. vol.
i. p. 81, t. 35; Gen. § Sp. Orch. Pl. p. 40; Wight, Icon. no.
906 ; Reichb. fil. in Bonplandia, 1857, p. 57 ; Thwaites, Enum.
Pl. Ceyl. p. 295; Benth. Fl. Hongkong. p. 332.—L. pendula,
Lindley, Bot. Reg. 1838, Misc. p. 180; Regel, Ind. Sem. Hort.
Petrop. 1868, p. 79.—L. elegans, Lindl. in Wall. Cat. no. 1943
(in parte); Gen. § Sp. Orch. Pl. p. 30.—L. nesophila, Reichd.
fil. Otia Hamb. i. p. 50; Xenia Orch. iii. p. 31.—L. Boottii,
Kegel, Ind. Sem. Hort. Petrop. 1864, p. 14?

Pseudobulbusclavatus vel ferme cylindricus, 3—52 unciaslongus,
vaginis paucis pallidis tectus, radicibus crassis lanatis; folia
bina erecta vel recurva oblonga lanceolata acuta crassa, 6-8
uncias longa, 2-1} lata; scapus erectus vel subnutans, sepius
inferne alatus; flores dense aggregati rhachin exalatum ferme
omnino tegentes; bracteæ lanceolate acuminate membranacex,
1 lineam longs; sepala lanceolata acuta; petala linearia, pallide
virides (marcescentia albida); labellum oblongum ovatum re-
curvum integrum viride (marcescens aurantiacum) ; columna sub-
erecta apice parum curva, crassiuscula, alis obscuris ; anthera
ovata; eapsula obovata pyriformis vix unciam longa, pedicello
zequalis.

india: Khasiya, 1000-3000 ft., Sikkim, 3000-5000 ft., Hooker g
Thomson!; Nepal, E. Bengal, Wallich!; Churra, Khasiya,
Griffith! ; Courtallum, Wight !; Tinnevelly, Beddome! ; Bootan,
Griffith no. 107 ! ; Ceylon, Macrae, Thwaites no. 3177! ;

China: Hongkong, Mt. Victoria, Lamont! Harland, Champion
335.

Java, Zollinger no. 777.

Samoa, Whitmee! Fiji Islands, Horne! Seemann!
LINN. JOURN.— BOTANY, VOL, XXII. 2A

294 MR. H. N. RIDLEY'S MONOGRAPH

Var. SPATHULATA.—L. spathulata, Lindl. Bot. Reg. 1842,
p. 81, Mise. no. 189.

Pseudobulbis ovatus oblongus brevis, biuncialis; folia lanceo-
lata spathulata; seapus brevior anceps; labellum magis ovatum
subobtusum.

India, Falconer!; Khasiya, Sikkim, Hooker & Thomson!;
Moulmein and Tavoy, Parish; Lakat Mts., Wallich !; Sadiya
Mts., Assam, Griffith !

Planta variabilis sepe culta. Z. elegans, Lindl. in Wallich
no. 1943, Penang (Gen. Porter), constat hzc species florifera et
fructifera cum alia specie fructibus multo majoribus ; flores nullos
inveni, labello retuso ut dixit et pinxit Lindley in herbario suo,
forsan floribus ullius speciei intermixtis.

102. L. PUsILLA, mihi.—L. auriculata, Reichb. fil. in Flora,
1872, p. 77, non Miq.

Subesspitosa; pseudobulbus parvus ovatus conicus, j unciam
longus, vaginis albescentibus tectus ; folia iis precedentis similia
sed minora singula spathulata lanceolata ferme 3 uncias longa,
i lata; seapus alatus 4-uncialis; racemus subdensus; flores
minuti copiosi ; bractezx lanceolate acuminate, 1-2 pedicellorum
zquales; sepala et petala anguste lanceolata ligulata reflexa;
petala angustiora; labellum oblongum panduratum obtusum
angulis basalibus pone columnam paullo auriculatis; columna
paullo arcuata, alis brevibus; anthera ovata obtusa.

India: Khasiya, Hooker & Thomson! Anamallays, Beddome!

103. L. crAvIGERA, n. sp.—Pseudobulbi approximati elongati
baculiformes 4 uncias longi, basi incrassato, vaginis laxis tecto,
superne nudi; folia bina angusta lineari-lanceolata obtusa, 4—9
uncias longa, 1 unciam lata, divaricata; racemus gracilis elongatus,
circiter 4-uncialis, vagina laxa ad basin supra folia vix semi-
uncialis ; flores remotiusculi minutissimi, iis L. longipedis simu-
lantes; bractee anguste lanceolate setacee; sepala ovata
oblonga obtusa; petala anguste linearia; labellum ecallosum
ovatum oblongum obtusum integrum; columna arcuata gracilis,
basi dilatata, alis brevibus quasi quadratis ; anthera ovata sub-
obtusa.

Java, Zollinger no. 865! in Herb. Brit. Mus.
104. L. vEsTITA, Heichb. fil. in Flora, 1872, p. 274.

Pseudobulbi appressi subteretes, baeuliformes, juniores vaginis
amplis vestiti, diphylli; folia cuneato-oblonga ligulata acuta;

OF THE GENUS LIPARIS. 295

scapus teretiusculus inferne nudus superne dense racemosus ;
bracter lanceolate ovariis subequales ; sepala ligulata, obtusa ;
petala filiformia; labellum oblongum retusum, nune utrinque
angulatum, antice crenulatum, ecallosum ; columna basi et apice
ampliata.

Assam, Obres, Mann.

105. L. sracnyurus, Reichb. fil. in Flora, 1872, p. 274.

Pseudobulbi ligulati (ancipites?) diphylli; folia cuneataoblonga
ligulata acuminata ; pedunculus subæqualis densiflorus ; bracteæ
setaceæ ovariis pedicellatis subæquales; sepala ligulata obtusa;
petala linearia; labellum trifidum, laciniis lateralibus basi semi-
cordatis; antice obtusangulis, lacinia media triangula ; columna
apice dilatata ; anthera acuta.

Assam, Mann.

106. L. STRICKLANDINIANA, Reichb. fil. in Gard. Chron. 1880,
p. 232.

Pseudobulbi ovoidei conici 1-8-phylli, vulgo 2-phylli ;
folia lineari-ligulata acuta; peduneulus compressus utrinque
alatus hine ensiformis ; racemus pluriflorus densiusculus, bractew
triangule ovariis pedicellatis non æquales; sepala ligulata ;
petala linearia apice retusa ; labellum a cuneata basi transverso
dilatatum trilobum, lobis lateralibus rotundatis, lobo medio
triangulo brevi, omnibus lobes antice minutissime denticulatis ;
columna curvata antice vix conspicue dilatata; calli minuti in
basin labelli ?

Locus ignotus, introd., Mann.

107. L. rRILOBA, n. sp.— Pseudobulbus deest ; folium singulum
(an semper?) lanceolatum acutum recurvum scapum superans,
6 uncias longum, 2 latum ; scapus omnino racemosus triuncialis ;
flores copiosi iis Z. pusille squales; bractesz lanceolate longs
acuminatee dimidio ovarii longiores; sepala lanceolata subacuta,
lateralia recta; petala linearia ligulata angustiora; labellum
trilobum, lobis lateralibus erectis obtusis, medio longiore porrecto
ovato-lanceolato obtuso; columna ferme horizontalis arcuata ;
alis brevibus obtusis.

Philippines, Cuming! in herb. Lindl.

Affinis L. pusille, sed major, et labello trilobo distincta.

108. L. DENDROCHILUM, Aeichb. fil. in Flora, 1872, p. 278. ;
Pseudobulbi oblongi apiee attenuati approximati diphylli ;

296 MR. H. N. RIDLEY'S MCNOGRAPH

folia angusta cuneata oblonga lanceolata acuta ;* bractes lanceo-
late acuminate inferiores floribus equantes, superiores breviores ;
sepala oblonga acuta, petala linearia acuta; labellum ‘columna
bene longius ecallosum oblongum versus apicem obtuse trilobum,
lobus medius angustior, nunc obtuse tridentatus ; columna apice
ampliata ; flores straminei labello aurantiaeo.

Cult. in horto Dayano.

109. L. PARVIFLORA, Lindl. Orch. Pl. p. 31.—Malaxis parvi-
flora, Blume, Bijdr. p. 292. — Elasmium parviflorum, Blume,
MSS. in Herb. Brit. Mus.—L. condylobulbona, Reichb. fil. in
-Hamb. Gartenz. 1862, p. 34.

Pseudobulbi elongati cylindrici 3-unciales ; folia bina lanceolata
obtusa multinervia, 10 uncias longa, 3 unciam lata ; scapus longis-
simus gracilis, 19-uncialis subangulatus, majore parte racemosus ;
flores minuti copiosissimi; bractee lanceolatæ acuminate $3
pedunculi :quales, 14 lineam longa; sepalum postieum lanceo-
latum ligulatum obtusum lateralia subsimilia pendula subfaleata ;
petala anguste linearia: labellum trilobum, lobis lateralibus erectis
obtusis, medio bilobo angusto lobis acutis forcipatis brevibus
ciliatis; columna arcuata gracilis; anthera hemispherica sub-
acuta. EE

Java, Forbes! Blume! Moulmein, Parish!

Descriptio pessima Blumei Malazidis parviflore-cum Elasmio
parvifloro, Blume! haud male convenit: cl. Lindleyus plantam non
vidit.

110. L. Fraccrpa, Reichb. fil. in Linnea, xli. p. 44.

Pseudobulbus cylindricus, 3-34 uncias longus; folia 1-2 an-
gusta lanceolata acuta costata, 7-9 uncias longa, 1j lata; scapus
gracilis pedalis; racemus subdensus elongatus; flores minuti
copiosissimi; bracteæ lanceolate acuminate membranacee 1
lineam longs; sepala ovata lanceolata obtusa, petala linearia
angustiora; labellum angustum oblongum ellipticum bilobum,
marginibus minute ciliatis, lamina abrupte deflexa; columna
gracilis curva, alis brevibus obtusis; anthera ovata obtusa.

Java, Lobb 171! Mt. Harrang, Horsfield !

Forsan preeedenti nimis affinis, sed validior, et lobi parvi
laterales labelli hujus absunt.

Species dubiz hujus sectionis.
L. PALLIDA, Lindl. Orch. Pl. p. 31.—Malaxis pallida, Blume,
Bijdr. p. 391.

OF THE GENUS LIPARIS. 297

* Foliis in bulbo oblongo compressiusculo vaginato solitariis
lanceolatis ; bracteis lineari-lanceolatis remotis labelli limbo
rotundato Duet crenulato ; flores olivacei." Java, Mt. Salak.

L. AFFINIS, Lindl. Orch. Pl. p. 31.—Malaxis affinis, Blume,
Bijdr. p. 892, t. 54.
“ Foliis in bulbo ovato oblongo solitariis, lanceolata linearibus
inferne canaliculatis, pedunculo nutante, foliorum longitudine ;
bracteis subulatis, labelli limbo rotundato." Java, Bantam.

L. u1x1Ma, Lindl. Orch. Pl. p. 32.—Malaxis minima, Blume,
Bijdr. p. 391.

“ Foliis in bulbis (seriatis ovatis compressiusculis) lineari-
lanceolatis, pedunculo folium zquante compresso, floribus laxe
Spicatis distichis; capsulis triquetris.” Affinis L. cespitose.
Java: Mt. Salak and Pantjar.

Excluded species.
. alata, A. Rich. Orch. Nilgh. p. 13, videtur Microstylis.
. Bernaysii, F. Muell. Fragm. Fl Austral. xi. p. 21, est
Microstylis Bernaysii, ejusd.
- bidentata, Griffith, Notul. iii. p. 277, figs. celxxxviii., planta
omnino dubia sed certe non Liparis.
. bracteosa, Lindl. in Wall. Catalogue, no. 7406, est Platy-
clinis sp.
. densiflora, A. Rich. Orch. Nilgh. p. 14, est Microstylis.
. intermedia, A. Rich. Orch. Nilgh. p. 18, videtur Microstylis.
. priochilus, Lodd. Bot. Cab. 1751, est Microstylis versicolor.
. prionotes, Lindl. MSS. in Herb., est Sarcochilus sp.
serreformis, Lindl. Orch. Pl. p. 53; Wall. Cat. no. 1946, est

Sarcochilus, sp.

. labiata, Sprengel, Syst. iii. p. 741 — Epidendrum labiatum, Sw.
. monophylla, Sprengel, l. c. — Epidendrum ? monophyllum, Hook.

PUB DEINER D N NS

Nomina nuda.

. cylindrostachya, Lindl. ex Hemsley, in Gard. Chron. n. s. xvi.
(1881) p. 592. Brasilia.

. anceps, Lodd. Orchids cult. at Hackney, p. 4. India.

. elegantissima, Veitch, Catalogue of Garden Plants for 1881.

. obcordata, * Vahl," Atkinson, Plants of N.W. India, p. 318.

. tristis, Lodd. Orchids cult. at Hackney, p. 4. Ceylon.

. “ zeylanica,” Lodd. Orchids cult. at Hackney, Pn. 79, p. 4.
Ceylon.

LINN. JOURN.—BOTANY, VOL. XXII. 2n

PME M

298 MR. W. MITTEN ON MOSSES AND

The Mosses aud Hepatice collected in Central Africa by the late
Right Rev. James Hannington, Bishop of Mombasa, F.LS.,
F.G.S., &e., with some others, including those gathered
by Mr,/H. H. Johnston on Kilimanjaro. By WiLL1AM
Mirrex, A.L.S.

[Read 3rd June, 1886.]

(Puates XV.-XIX.)

DvnixG his first journey from the coast opposite Zanzibar to the
Lakes, the Bishop, then known as the Rev. James Hannington,
collected many very small specimens, chiefly on the Usagara
Mountains and in the forest of Ugogo; many of these were
entirely barren, and have not yet been identified with any de-
scribed African species; and they seem to indicate a region not
abounding in Mosses or Hepatice. In the Usagara the conditions
must have been more favourable, for he sent thence by letter
many Ferns ; but at the time he despatched these his health had
been so broken down with the hardships and privations of the
journey through a country so deficient in food and water, that he
had quite given up all hope of being able to return to the coast.
This, however, he eventually did; and during the voyage home,
and more rapidly after, regained his usual health.

Induced to return as Bishop of Eastern Equatorial Africa,
with residence at Mombasa, the importance of finding a better
rout to the Missionary station in Uganda must have seemed to
him, who had experienced the miseries of the older road, a matter
of urgency. Being a man possessed of great courage and the
faculty of winning others to do as he wished, he soon made an
attempt to ascend Kilimanjaro. On his return from this expe-
dition he wrote from Mombasa (May 13, 1885) :—* I only got up
to about 8800 feet; I wanted to get to the snow-line, and made a
desperate struggle, but the ascent is so gradual, that it takes
such a dreadfully long time. I started from about 5000 feet,
and walked from morning till 4 p.m. with scarce any stoppages,
and then never got out of the dense forest-tangle. The same
Mosses seem to me to crop up here as in Usagara; my plants all
got spoilt with the intense rain, the Mosses running a very narrow
shave, and many are discoloured. The rainfall has been very ter-

HEPATICZE COLLECTED IN CENTRAL AFRICA. 299

rible, 6 inches in four days ; so that I am expecting an unhealthy
time shortly. This new road is perfectly healthy; I think the
dangers have been greatly exaggerated. The road being healthy,
one can endure much vexation and delay ; travelling has been as
though I was in England or Wales."

On another slip of paper he goes on :—** I found three or four
brambles on Kilimanjaro, also Sibthorpia and Arthrolobium. I
found also another Hydnora; and have determined in my own
mind beyond any question that there are two species—growth,
habit, size, substance, colour, shape entirely different ; what more
can you want to make a species?” This refers to doubts expressed
when he brought home from his first journey a painting of a species
he had met with.

In the interval between the Bishop's two expeditions a box
was sent from the Usagara by Mr. Last, containing sundry objects
of natural history, a variety of living plants and ferns packed in
lichens, chiefly Parmelia perforata, amongst which were a few
additional Mosses. The living plants were Dorstenia indica; a
tetraquetrous Euphorbia? with mottled green and white stems
and oblanceolate foliage; also Asplenium trifidum. With these
were dead Orchids and Ferns; some of the latter being planted,
gave rise to Streptocarpus caulescens and S. Kirkii, as well as
Impatiens Sultani ; these were supposed to have come from an
altitude of about 4000 feet.

All the species in the following enumeration, unless otherwise
stated, were collected by the Bishop himself.

Arcuipium, Brid.

A. AFRICANUM, sp.n. (PI. XV. figs. 1-4) Monoicum; folia
caulina patentia lanceolata integerrima, nervo continuo, cellulis
oblongis, perichetialia subduplo longiora latioraque, apice sub-
ulato acuminata, thecam globosam obtegentia; sporis obtuse an-
gulatis rotundatis (17) ; flore masculo gemmiformi.

-Hab. Usagara Mountains. :

Similar to the European A. alternifolium, and differing in
habit and also in the oblong not abbreviated cells of its leaves
from the African 4. Ecklonii ; the subulate part of the perichzetial
leaves is about one third of their whole length; the same portion
of the leaves in A. Ecklonii occupies two thirds.

It appears probable that the species of this genus will become
2522

300 MR. W. MITTEN ON MOSSESJAND

numerous; for besides the two African species above named,
another was collected by Dr. Rehmann near Cape Town :—

A. REHMANNI, sp. n. (Pl. XV. figs. 5-8.) Monoieum, pul-
vinatum cespitosum ; folia patentia deltoideo-ovata, nervo per-
cursa, e cellulis breviter oblongis areolata, perichetialia e basi
orbieulari convoluta in subulam angustam producta patula;
theca globosa; sporis levibus (12); flore masculo gemmiformi
stipitato.

Stature that of A. alternifolium ; stems short, with many short
branches.

All these African species agree generally with the oldest known
European species, and very similar ones are found in India and
Australia; but a great departure is presented by a species sent
from Monte Video by M. Gibert, No. 1249.

A. GisEnRTIsp.n. Monoicum ; caules erecti in cespites com-
pactos aggregati graciles humiles; folia subjulacea imbricata
erectiuscula subdeltoideo-ovata brevissime acuminata integer-
rima, nervo angusto percursa, cellulis minutis rotundis ubique
areolata, perichetialia late ovata brevi apiculata convolutacea ;
theca globosa tenera; sporis magnis sublevibus; flore masculo
gemmiformi in apicibus ramorum. Caulis circiter centimetrum
altus cum foliis 0:25 mm. latus. Folia juniora subglauco-viridia,
seniora pallide fusca.

This, in its julaceous stems, is analogous to those species of
Pleuridiwm like the Australian P. tenuissimum.

Trematopon, Rich.
T. LIGULATUS, Rehmann, 22.
Hab. Usagara Mountains.

Dicranum, Hedw.
D. ($ HEMICAMPYLUS) DrCHOTOMUM, Brid.
Hab. Kilimanjaro.

Originally described from Bourbon specimens : itis found also
in Madagascar.

D. ($ EvbicRANUM) JonwsTONI, sp.n. Caulis pallide fusco to-
mentosus; folia laxa patula apicalia faleata secunda linearia
sensim loriformi-angustata, nervo dorso bicarinato infra summum
apicem latiusculum desinente marginibusque e medio ad apicem

HEPATICZE COLLECTED IN CENTRAL AFRICA. 301

breviter denticulatis, cellulis angustis inferioribus parietibus con-
tinuis alaribus externis obseuris internis hyalinis.

Hab. Kilimanjaro, 10,000 feet. “A fragment (a single stem)
of a large piece of moss; the rest accidentally lost.” —H. H.
Johnston, No. 52.

Somewhat similar to a slender state of D. majus; unlike any
state of D. Scoparium, from the narrower lower portions of the
leaf.

LepropontTium, Hampe.

L. rapicosum, Mitt—Didymodon, Mitt. Journ. Linn. Soc.
vii. p. 149.

Hab. Kilimanjaro, Bishop Hannington. Usagara Mountains,
Mr. Last.

Rather smaller in all its parts than the specimens from the
Cameroon Mountains; and it seems but little different from the
Indian L. squarrosum.

The Usagara specimens are the most slender and elongated ;
but in none is the form of the leaves so much narrowed up-
wards as in L. stellatum.

L. rristicuus.—Zygodon, C. Müll. Bot. Zeit. 1855, p. 764.—
Z. strictissimus, Rehm. Seems identical with Z. Priessianus,
Hampe, Linn. 1860, p. 6832 Didymodon papillatus, Hook. f. et
Wils., Leskia rubricaulis, Taylor. If there proves no distinc-
tion in the fruit, the species ranges from S. Africa to Abyssinia,
Australia, and New Zealand.

CAMPYLOPUS, Brid.

C. PERPUSILLUS, sp. n. (Pl. XV. figs. 9-12.) Gracilis laxe
cespitosus; folia erecto-patentia elongate oblongo-lanceolata,
apicem versus abrupte acuminata caviuscula, nervo excurrente
apice pauce dentato, cellulis superioribus ovali-rotundis inferi-
oribus oblongis rectangulatis, perichetialia vix diversa; theca
minuta ovalis in pedunculo flexo pendula sepulta, operculo
conico acuminato, calyptra brevis basi fimbriata, peristomio den-
tibus teneris incompletis.

Hab. Ugogo.
In appearance this small species is more like the young states

of Anisothecium varium than of any species of Campylopus. The
plants are buried in sandy earth to near their apices; and the
fruit is quite invisible until sought for.

302 MR. W. MITTEN ON MOSSES AND

Fragmentary specimens of several other species of this genus
were collected during the first journey, but all too small in
quantity to determine.

OncopHorts, Brid.

O. ($ RuABDOWEISIA) CYATHICARPUS (Zygodon, Mont.).

Hab. Kilimanjaro, Mr. H. H. Johnston, No. 12, with fruit.
Under 8000 feet without fruit, Bishop Hannington.

This Moss appears to inhabit the whole of Africa where there
are suitable mountain-regions.

Scuistomitrium, Dozy et Molk.

S. ACUTIFOLIUM, sp. n. Cæspitosum, humile densifolium ;
folia patentia recta secundave a basi ovali concava, lamina utraque
e cellulis tribus lata inde marginibus incurvis subulato canalicu-
lato, apice excavata mucrone parvo angusto terminata lævia inte-
gerrima, perichætialia caulinis longiora subulata obscuriora.

Hab. Usagara Mountains, Bishop Hannington and Mr. Last.
Natal, Mrs. Saunders. Madagascar, Dr. Meller.

Caulis 1 cm. altus, cum foliis vix 3 mm. latus. Folia superiora
glauco-viridia, inferiora albescentia, 2 mm. longa.

Similar to S. cucullatum, Thw. et Mitt., but with more subulate
contracted leaves.

All the species yet known to belong to this genus agree very
closely in the form of their leaves and in their having a mucro-
nate point. S. apiculatum and S. robustum have been figured
with fructification, which is absent in S. cucullatum and in S. acu-
tifolium, although in the last perichetia at the apices of some of
the stems are present. The genus, so far as known, seems re-
stricted to the Indian Archipelago, Ceylon, and Tropical Africa,
and, beside those mentioned, must include another gathered on
Kina Balu by Mr. Low :—

S. Lowi, sp.n. Folia secunda falcatave dense inserta oblongo-
lanceolata concava inferne limbo e seriebus cellularum oblongarum
4-marginata inde marginibus incurvis canalieulata, apice abrupte
excavato, apieulo parvo terminata levia perichetialia vix diversa ;
theca in peduneulo gracili aurantiaco levi cylindracea, calyptra
basi fimbriis divergentibus.

This is a little less than S. robustum, and thus much larger
than S. apiculatum or S. acutifolium.

HEPATICH COLLECTED IN CENTRAL AFRICA. 308

GarMM1A, Ehrh.
G. LEUCOPH £A, Grev.
Hab. Kilimanjaro, covering stones at 12,000 feet, H. H.
Johnston.
SYRRHOPODON, Schwegr.

S. (SOnTHOTHECA) ASPER, Sp.n. Cespitosus ; caulis curvatus ;
folia sordide viridia dein pallescentia, siccitate secunda curvata,
humida patentia, a basi oblonga latiora hyalina, margine superne
dentato, subulato angustata longe linearia, apice acuta, margine
limbo calloso densissime papilloso duplicato dentato, nervo dorso
papilloso asperrimo, cellulis densis obscuris areolata.

Hab. Usagara Mountains and Kilimanjaro, Bishop Han-
nington.

Similar to S. circinatus, but with narrower leaves, the base
entirely occupied with hyaline cells.

Caulis 2-4 cm. altus ; folia 5 mm. longa.

Fragments of a species of Eusyrrhopodon, like S. Gaudichaudi,
were also amongst the Mosses from Kilimanjaro.

THYRIDIUM, Mitt.

T. AFRICANUM, sp. n. Caulis decumbens ; folia humida erecto-
patentia sicca appressa apicibus ineurvis oblongo-lanceolata
acuta supra basin parum latiora, nervo percursa, margine flexuoso
anguste hyalino limbata remote brevissime serrulata, cellulis
minutis rotundis obscuriusculis basalibus mediis oblongis hya-
linis ad tertiam partem, folii longitudinis juxta nervum protractis
areolata,

Hab. Usagara Mountains, Mr. Last.

Caulis subuncialis. Folia 4 mm. longa, 1 mm. lata, sordide
viridia, juniora pallidiora sub glauco-viridia.

A very distinct species of which the specimens are too incom-
plete ; it appears to be a little more robust than 7. fasciculatum.

CALYMPERES, Sw.

C. LrGULARE, sp. n. Humile; folia patentia ligulata inferne
haud dilatata, apice plus minusve acuta, nervo longiore producto
mucronata, margine integerrima limbo nullo, cellulis hyalinis
spatium breve subquadratum occupantibus reliquis minutis rotun-
dis viridibus; fructus C. Afzelii.

304 MR. W. MITTEN ON MOSSES AND

Hab. Usagara Mountains, Bishop Hannington.

Statura adspectuque omnino C. Afzeliz; folia longitudine
latitudinem quinquies superant.

In its leaves having no limb this species is distinct from the
other African species, and comes nearer to the wide-spread
C. Dozyanum of the Indian Archipelago and Pacific Islands.

SxsrEa1UM, Bruch et Schimp.

S. UNGUICULATUM, sp. n. Caulis brevis simplex ramosusve ;
folia in comam disposita patentia sicca incurvata crispata elongate
lanceolata, perichetialia paululum latiora omnia canaliculata
marginibus superioribus involutis, nervo latiusculo in mucronem
unguieularem excurrente, cellulis minutis rotundis inferioribus
oblongis; theca parva immersa ovali-globosa, calyptra stylo
elongato.

Hab. Qua Chiropa.

Caules in cæspitibus laxis aggregati 1:50 mm. alti. Folia lon-
giora 2 mm. longa luteo-viridia, nervo pallidiore; theca pallide
aurantiaca, 0:25 mm. longa.

Smaller than the European S. crispum, but corresponding

generally with it; and nearly similar species seem to be found in
India and S. America.

Hyopuina, C. Müll.

H. PLrcATA, sp.n. (Pl. XV. figs. 13-16.) Monoica, gregarie
ceespitosa humilis ; caulis inferne foliis parvis sparsis apice rosu-
latim foliatus; folia patula obovata spathulata obtusissima inte-
gerrima, nervo ad apicem percurrente levi, cellulis superioribus
parvis rotundatis basalibus majoribus quadratis pallidioribus,
perichetialia ovali-oblonga obtusa; theca in pedunculo gracili
pallido erecta elongate ovali-cylindracea acute octangulata, oper-
culo conieo acuminato recto, peristomio brevi dentibus subulatis
solidis rubris levibus infra os insertis 16 annulo composito;
flos masculus foemineo similis.

Hab. Usagara Mountains, on the ground.

In its manner of growth and foliage this resembles many
species of Mosses referred to Hyophila, such as H. javanica and
H. involuta; but the plicate capsule is peculiar, and the peristome
is rather that of a Weisia than of an incomplete Tortula,

HEPATICH COLLECTED IN CENTRAL AFRICA. 805

ANGCTANGIUM, Hedw.

A. PUSILLUM, sp. n. (Pl XV. figs. 17-19.) Humile; folia
patentia incurva sicca contorta ligulata obtuse acuta, nervo
valido pallido dorso levi sub summo apice desinente carinata,
margine subintegerrima, cellulis densis obscuris basalibus paucis
mediis suboblongis pellucidis; theca in pedunculo gracillimo
pallido ovata.

Hab. Kilimanjaro, Bishop Hannington.

Folia juniora eruginoso viridia, seniora fuscescentia in verti-
cillis interruptis disposita.

A small species, which agrees very nearly with A. Mariei,
Besch., from Nossi-bé ; but the apices of the leaves are wider
and their nerve is not papillose. Among the small specimens
gathered there is no indication of the stems being tomentose

with rootlets.
Macromirrium, Brid.

M. (§ Macrocoma) Dnzxazr, Zornsch. Linn. xv. p. 181.

Hab. Usagara, Bishop Hannington and Mr. Last.

The difficulty in extricating the species of this group without
examination of the original specimens has led to some confusion,
and the following appears to have been mistaken :—M. CONFUSUM,
Sp.n. Dioieum; M. Lycopodioidi simillimum sed robustior, foliis
e cellulis distinctis areolatis.—JM. Lycopodioides, C. Müll., Rehm.
1655, 166; M. tenue, var. brachypus, ejusd. 162; M. cespitans,
C. Müll. (planta mascula), Rehm. 159.

Gathered on Table Mountain by most collectors, and in Natal
by Capt. Rooper.

Always a little larger than the true M. Lycopodioides gathered
by Dr. Burchell and also by Dr. Rehmann (M. Dawsoniomitrium,

C. Müll., Rehm. 160).

Zxaopox, Hook. et Tayl.

Z. (§ SrENoMITRIUM) EROSUM, sp.n. Habitus staturaque Z.

trichomitrii, sed foliis angustioribus apice apiculo parvo pellu-
cidiore acuto marginibus minutissime crenulatis, cellulis basi
pro spatio quadrato elongatis pellucidis reliquis minutis parietibus
latis pellucidioribus.
Hab. Kilimanjaro, a few barren stems, Bishop Hannington.
With the same appearance as Z. trichomitrius, Hook. et

806 MR. W. MITTEN ON MOSSES AND

Wils. (Z. runcinatus, C. Müll), Z. Rehmanni, C. Müll, and
Z. rugifolius, ejusd., all gathered in S. Africa by Dr. Rehmann ;
and agreeing very nearly with the S.- American Z. pentastichus,
Z. quitensis, and Z. crenulatus, Musc. Aust.-Amer. The Ana-
calypta stellatifolia, Hampe, probably belongs to the same group.
The fruit is seldom in a state to afford a correct idea of the
structure of the internal peristome; but it is impossible to avoid
the considerations arising from an examination of the S.-American
species without recalling the remarks upon them by Wilson,
Bry. Brit. 264, in his notes on Paludella; and the more natural
arrangement may be to bring together Zygodon, Paludella, and
Aulacomnion, at least so far as relates to the bog species, if
indeed these two latter genera are truly more than steps higher
in peristomial development in a natural genus containing such
great diversity in that appendage as is observable in undoubted
Zygodons.

Bryum, Linn.

B. (§ BRacHYMENIUM) PULCHRUM, Hook. Bot. Miscell. 1880,
t. 38.

Hab. Usagara Mountains, forming large patches and in fruit,
Bishop Hannington.

B. (BRACHYMENIUM) CAPITULATUM, sp.n. Monoicum; caulis
humilis inferne radicellis fuscis foliisque sparsis parvis obtectus
superne subrosulatim foliatus innovationibus conformibus infra
perichetium oriundis flore parvo masculo terminatis; folia obovato-
spathulata, nervo in setam rectam exeurrente apiculata, margine
limbo e duplici seriebus cellularum versus apicem indistincto
pauei serrulato inferne recurvo, cellulis reliquis breviter ovalibus
areolata, perichetialia lanceolata angustiora ; theca erecta inclina-
taque oblongo-obovata clavzeformis, operculo satis parvo breviter
conico peristomii dentibus a basi sensim angustatis interni mem-
brana usque ad medium dentium producta.

Hab. Kilimanjaro, Bishop Hannington ; he gathered it also in
the Usagara Mountains. Cameroon Mountains, Mann ; Central
Madagascar, G. W. Parker.

Entirely similar to B. (§ Brachymenium) nipalense in habit but
rather smaller in stature. It differs from B. flexifolium, from
Abyssinia, in its more strongly nerved and firmer leaves as well as
in the more regularly erect capsules. The form of the operculum

HEPATICZE COLLECTED IN CENTRAL AFRICA. 307

seems to vary with the age of the capsule; from conie it becomes
in age hemispherical.

B. ($ EUBRYUM) nosEUM, Schreb.

Hab. Kilimanjaro, with perfected fruit, Bishop Hannington;
Usagara Mountains, barren, Mr. Last.

B. integrifolium, Rehm., from Natal, appears to be not different,
the foliage being dentate, as usual in European specimens of
B. roseum. It was gathered also in Sierra Leone by Barter and
in Madagascar by Dr. Meller; and it is most probable that the
Indian B. giganteum, Hook., and the S.-American B. grandi-
folium, Tayl., are but states induced by some different conditions
of one widely spread species.

B. Cowwznsonrr, Schw.

Hab. Kilimanjaro, barren.

From this B. umbraculum, Burchell, differs scarcely at all.
In some state the species appears spread over the Southern
hemisphere.

B. ALPINUM, Linn.

Hab. Usagara Mountains, barren, Bishop Hannington.

Of the usual colour and appearance; this handsome species is
found in the Cameroon Mountains and in S. Africa.

B. ARGENTEUM, Linn., var. ARGYROTRICHUM.—BD. ($ Argyro-
bryum) argyrotrichum, C. Müller, Linn. 1875, p. 385.

Hab, Usagara Mountains, Bishop Hannington.

At first sight this differs greatly from the common states of
the species as seen in Europe, but the differences seem com-
pletely connected by specimens from the Nilgiri Mountains, and
in one from Abyssinia both states of the foliage are observable on
the same specimen.

B. coronatum, Schw.
Hab. Usagara Mountains.

Mnivum, Linn.

M. rostratum, Schrad.
Hab. Kilimanjaro, without fruit, Bishop Hannington.
Inflorescence as usual and not in any way differing from

European states.

908 MR. W. MITTEN ON MOSSES AND

Ruizocontum, Brid.

R. ($ PyRRHOBRYUM) SPINIFORME, Linn.
Hab. Kilimanjaro, Bishop Hannington.

HYPOPTERYGIUM, Brid.
H. LARICINUM, Brid.

Hab. Usagara Mountains ; also gathered by Mr. Last.

RuacorinuM, Brid.

R. AFRICANUM, Mitt. in Journ. Linn. Soc. vii. p. 162.

Hab. Msilala, and on Kilimanjaro, Bishop Hannington.

Specimens to all appearance the same come from Mauritius
and Madagascar; but as none have been seen in fruit except
those from the Cameroons it cannot be decided. The Abyssinian
R. longe-aristatum, C. Müll., does not appear to differ; R.
capense, C. Müll, Rehmann 267, seems to be R. mucronatum,
Beauv., it is found in Natal. Besides these there have been
described R. angusti-stipulaceum, C. Müll., from Comoro; R.
spelunce, ejusd., from Kalunga, Bongo; R. pectinatum, ejusd.,
from Dar Fertit, with the preceding collected by Dr. Schweinfurth;
there is also the R. mauritianum, O. Müll., R. angustifolium, and
R. prelongum, Schimp., enumerated by Bescherelle from Bourbon,

Mauritius, and Comoro. A well-marked species was sent by
Dr. Ayres from Mauritius :—

R. AYRESII, sp. n. Caulis pinnatim ramosus sed sæpius

longe procerus parce tomentosus; folia inter se remota pectinatim
distichacea ovali acuminata, nervo obscuriore in aristam dimidiam
folii longitudinis æqualem excurrente lævi, margine apicem
versus subintegerrima, in ramis denticulata, cellulis omnibus
mollibus ovalibus pellucidis areolatis; folia stipuliformia ovata,
nervo in subulam equilongam lævem excurrente.

Hab. The Pouce Mountain.

Dr. Ayres himself noticed that this differed from the other
species he had gathered (R. africanum) in being without the
tomentum. In its areolation it is quite distinct, but it may be
really one of the species before mentioned as already described,
in which the form of the cells is not stated.

The description of the areolation of R. prelongum given by
M. Bescherelle in his ‘ Florule de la Réunion,’ &c., corresponds
very nearly with that found in R. africanum.

HEPATICH COLLECTED IN CENTRAL AFRICA. 309

Hookznia, Sm.

H. ($ CarnrcosrA) verstcotor, Mitt. — Lepidopilum, Mitt.
Journ. Linn. Soc. vii. p. 159.

Hab. Kilimanjaro, Bishop Hannington.

This closely agrees with the S.-American H. Langsdorfii and
the Indian H. Utacamundiana as well as the Javan H. macropus.
Great difficulty arises in the comparison of specimens of the
species of this group from the uncertain manner of their revival
by maceration in water after having been dried ; in some the cells
refuse to expand at all, and thus the areolation may appear
completely different from what it was in the living moss.

The Hookeria (§ Lepidopilum) angolensis, Duby, Mém. de la
Soc. de Phys. de Genéve, tom. xxi. (1872), seems to be the same
species, which differs from H. Pappeana, Hampe, Ic. 11, in its
undulated and shining leaves when dry and its more narrow
areolation. Fragments which appear to belong to H. Pappeana
were amongst Mosses collected in the Usagara.

LzPipoPiLUM, Brid.

L. HaANNINGTONI, sp. n. Monoicum ; caulis subsimplex ; folia
distichacea media intermediaque ovato-oblonga lateralia patentia
longiora omnia subulato acuminata, margine superne serrulata,
limbo nullo, cellulis elongate oblongis nervis binis medio evanidis,
periehzetialia parva ovato-lanceolata ; theca in pedunclo brevi fere
levi ovalis, peristomii dentibus lamina externa rubra supra
totam latitudinem lamins interne hyaline protracta, processus
&quilongis in membrana brevi exserta insidentibus; calyptra
parva sublzvis.

Hab. Usagara Mountains, Bishop Hannington.

Caulis unciam altus cum foliis explanatis 8 mm. latus. Folia
subnitida, stramineo-viridia plus minus rubescentia.

L. Lastu, sp. n. Monoieum ; caulis subsimplex ; folia disti-
chacea compressa media intermediaque ovalia lateralia longiora
oblonga omnia apice subacuminata acuta marginibus versus
apicem serrulatis, limbo nullo, cellulis oblongo-elongatis, nervis
brevibus, perichetialia parva late ovata acuminata; theca in
pedunculo brevi setulis densis obtecto ovalis erecta papuloso
aspera opereulo acuminato, peristomii dentes elongati, lamina
externa rubra angusto j tantum latitudinis lamine interne tegens,

810 ME. W. MITTEN ON MOSSES AND

processus perist. intern. equilongis in membrana brevi impositis
minute punctulatis ; calyptra ramentis brevibus inspersa.

Hab. Usagara Mountains., on small branches, Mr. Last.

Caulis unciam altus cum folis 8-4 mm. latus. Folia stra-
mineo-viridia vix nitentia. Pedunculus 3-4 mm. longus. Theca
1mm.

Closely resembling L. Hanningtoni, but very different in its
fruit.

Darron, Hook. et Tayl.

D. ratura, Mitt. in Journ. Linn. Soc. vii. p. 162.

Hab. Usagara Mountains, Mr. Last; Kilimanjaro, Bishop
Hannington.

The Kilimanjaro specimens are a little more robust than those
from the Cameroons and Usagara; but too little has been seen
to be certain that they belong to one species.

Hepwieta, Ehrh.

H. (§ BRAUNIA) secunpa, Hook. Musc. Exot. t. 99.

Hab. Kilimanjaro, perfectly fruited, Bishop Hannington.

In size, colour, ramification, and foliage the specimens agree so
closely with Mexican, that it is impossible to avoid the conclusion
that they belong to the same species; and the question at once
arises are the Braunia sciuroides of S. Europe, the B. indica, so
luxuriant in the Nilgiri Mountains, and the Abyssinian B. Schim-
peri really different, or are they not rather most probably slight
variations of one wide-spread species ?

H. ($ Evnepwrera) crrzraTA, Ehrh.

Hab. Kilimanjaro, Bishop Hannington. In no way different
from European specimens.

CzPuza, Brid.

C. ($ EUCRYPHEA)LAXIFOLIA, sp. n. Monoica; rami elongati,
ramulis paucis remotis elongatis vix pinnatim ramosi; folia pa-
tentia sicca appressa laxa ovato-lanceolata, apice latiuscula sub-
mucronata breviterve apiculata ibique subserrulata, margine
inferiore recurva, nervo infra apicem evanido carinata e cellulis
| parvis rotundatis areolata, perichetialia lateralia patentia anguste
cylindraeea; foliis internis thecam ovali-cylindraceam totam

HEPATIC COLLECTED IN CENTRAL AFRICA. 311

vaginantibus inde abrupte in mucronem aristiformen patentem
productis.

Hab. Usagara Mountains, Mr. Last.

Another closely allied species is C. DENTATA, sp. n. Rami
stricti curvative, ramulis sparsis pinnatim ramosi interdum
subsimpliei; folia imbricata ovata latiuscule acuminata, apice
dentata, margine anguste recurva, nervo sub apice desinente,
perichetialia lateralia parva ; foliis internis erectis spathulatis,
apice paucidentatis, nervo inferne evanescente in mucronem
setaceum erecto patentem excurrente ; theca immersa ovali-cylin-
dracea, operculo brevi conico acuminato, peristomio interno
externum breviore.

Hab. Umgoe Mountains, Natal, Mr. Plant ; ; gathered also by
Mr. Keit, No. 3.

Both these African species agree generally with the Mexican
C. patens, Hsch., but are more slender ; they have no near resem-
blance to the Indian C. spherocarpa. In the dried state the
two species are not very dissimilar; but on wetting them a great
difference is observable in the loose spreading foliage of C. laxi-

folia.

C. ($ AcRrocRYPHÆA) WELWITSCHII, Duby, Mém. de la Soc. de
Phys. de Geneve, tom. xxi. p. 426, t. 2. f. 1.

Hab. Usagara Mountains, and on Kilimanjaro, Bishop Han-
nington. It is found also in Madagascar.

The specimens yet seen of this species, if indeed they all be-
long to one, are in too small quantity to afford suflicient means
of comparison; similar forms are found in India (C. concavifolia
and C. incurvifolia), and in 8. America several others. All agree
in their mode of fructification, and differ chiefly in their leaves ; and
it may reasonably be doubted if in reality they are all distinct,
or states of one or two species of which C. julacea, Hornsch., from
Brazil, was the earliest known. It is not an easy matter to find
in some of the specimens what may truly be a cauline leaf on
stems which are merely fertile branches.

Prionopon, C. Müll.

P. Renmanni, sp. n. Folia subcompressa patula inferne
ovali-lanceolata anguste ligulato attenuata acuta (swpissime dif-
fracta), nervo sub apice desinente, margine late implana, superne

812 MR. W. MITTEN ON MOSSES AND

dentibus brevibus divergentibus armata, inferne ut plurimum
bisuleata e cellulis minutis areolata punctulis asperula.

Hab. Kilimanjaro, Bishop Hannington. Transvaal, Dr. Reh-
mann, Also in Mauritius ?

Very nearly resembling P. densus, C. Müll. (Hypnum, Sw.),
from Jamaica, Mexico, and Tropical America—regions in which
that species is now known to be accompanied by many others,
but until the discovery of the African species none was known
to have been found elsewhere.

PrEROBRYUM, Hornsch.

P. FLAGELLIFERUM, sp.n. (Pl XVI. figs. 1-3.) Rami in-
ferne breviter stipitati flexuosi, ramulis variis pinnati flagellis
minute foliatis gracilibus ; folia patentia ovato-lanceolata, nervo
infra apicem planum desinente caviuscula integerrima, cellulis
elongatis, perichetialia anguste oblongo-lanceolata sensim angus-
tata supra medium nervata, theca ovato-oblonga operculo convexo
brevi acuminato, duplo longiora, integerrima.

Hab. Usagara Mountains, Mr. Last.

P. HasxiNGTONI, sp.n. (Pl. XV. figs. 20-22.) Rami a caule
primario rhizomeformi ascendentes subplumeformes, curvuli ; ra-
mulis plurimis patentibus distichis fere ad basin usque pinnatis;
folia patentia laxe imbricata ovato-lanceolata parum acuminata
concava sieca subplicata, nervo infra apicem planum desinente
integerrima e cellulis elongatis areolata, perichetialia oblongo-
lanceolata subulato attenuata ad 3 usque nervata, thecam oblon-
gam 3 superantes.

Hab. Usagara Mountains,

Both these species are nearly allied to the South-American
P. trichomanoides, Spruce, P. filicinum, Sw., and P. angustifolium,
C. Müll. ; and although very nearly resembling each other, when
once the differences are apprehended are readily separable.

P. Hanningtoni has much more numerous and shorter branch-
letsanda denser foliage, besides the differences in the perichetium.

CALYPTOTHECIUM, Mitt.
C. AFRICANUM (Trachyloma, Rehm.).
Hab. Kilimanjaro, Bishop Hannington. Usagara Mountains,
Mr. Last, gathered also by Burchell. All the specimens
barren.

HEPATICE COLLECTED IN CENTRAL AFRICA. 313

So far as known, the greatest number of the species of this
group of Neckera-like Mosses is found in the Indian Archi-
pelago, Pacific Islands, and Australia; and should some of
those as yet known only in a barren state prove to produce
an exserted capsule, they would approach very closely to Trachy-
loma. The Trachyloma stipitatum, Mitt. in Journ. Linn. Soc. vii.
p. 156, is more correctly a species of Hypnodendron, very nearly
allied to T. arcuatum and the other species of the same group
as H. Archeri. It has the same tree-like mode of growth and the
same areolation, the ends of the cells having a callosity which
gives the leaves a dotted appearance, but which does not show as
prominences when the leaf is folded over. It is the only species
which has a nerve not continued to the point of its leaves.

Ervopium, Brid.

E. Hanninertont, sp. n. (Pl. XVI. figs. 4-7.) Monoicum;
caulis procumbens, radicellis fuscis repens; ramis fertilibus as-
cendentibus paucis ramosus; folia caulina patentia laxe sursum
imbricata ovalia obtusa caviuscula e cellulis parvis rotundis viri-
dibus in dorso folii papilliformi protuberantibus obscuriusculis
limitibus pellucidioribus areolata, apice in pilum sequilongum
levem albidum e cellulis angustis elongatis hyalinis compositum
subito producta; folia in ramis fertilibus appresse imbricata in-
feriora ovata e cellulis rotundatis ovalibusque pellucidis limitibus
angustis, perichetialia interna majora sensim in subulam hyali-
num angustata, cellulis folii heteromorphis papillis grossioribus
asperrima; theca immersa oblonga, operculo convexo apiculato
annulo lato, sporis levibus, calyptra brevis campanulata basi
multifida plicata apice subaspera; flos masculus gemmiformis
parvus.

Hab. Lake Nyanza, on Adansonia digitata (Bishop Hannington).

Caulis procumbens uncialis cum foliis 1 mm. latus; ramis fer-
tilibus 2-3 mm. longis. Folia caulina 0°5 mm. lata cum pilo
1:50 mm. longa. :

E. Schweinfurthii, C. Müll., Linn. 1875, p. 412, is described yis
cordate leaves, *acumine a basi latiusculo breviter attenuato,
and must therefore be quite different, although in some particulars
relating to the fruit it seems to agree pretty nearly ; it is also
stated to be diccious. Both species are allied most nearly to the

S.-American E. coronatum ; but still more closely agree with a

LINN. JOURN.— BOTANY, VOL. XXII. 2c

3814 MR. W. MITTEN ON MOSSES AND

remarkable species collected during the visit of the * Challenger’
to Japan, to be more fully described elsewhere.

ERPODIUM JAPONICUM, sp. n. Folia ovata obtusa acuminatave
in pilum breviorem pellucidum producta; theca emergens, peri-
stomio e dentibus subulatis 16, calyptra fere ad basin thece ovali
oblong descendens 14-16-plieata.

In this the whole appearance is that of E. Hanningtoni, the
point of the leaf being shorter; it is less conspicuous, and instead
of covering little more than the operculum, the calyptra includes
nearly the whole of the capsule ; and the presence of a completely
developed simple peristome is a new character in the genus.

Meteorium, Brid.

M. (§ EUMETEORIUM) IMBRICATUM, Beauv. Schw. t. 165.

Hab. Usagara Mountains and Kilimanjaro, without fruit,
Bishop Hannington.

Very variable in colour from pale green to orange-brown, also
in the form of the leaves of the stems, less so in those of the
branches ; their arrangement in series is chiefly evident on the

secondary ramuli. The species appears spread across Tropical
Africa, but not seen from the south.

M. (§ PAPILLARIA) AFRICANUM, C. Müll.
Hab. Usagara Mountains, Mr. Last; there and also on Kili-
manjaro, Bishop Hannington.

A wide-spread African moss, of which the fruit is wanting on
all our specimens.

TnAcHYPUs, Schw.
T. sERRULATUS, Beauv.
Hab. Kilimanjaro, Bishop Hannington.
Originally described from Bourbon, it is now known from
Madagascar and also from the Cameroon Mountains. It appears

to be always more compressed and less rigid than the nearly
allied Indian T. erispatulus.

Lertopon, Brid.
L. Smitu, Mohr.
Hab. Kilimanjaro, Bishop Hannington.

Previously known
from S. Africa.

HEPATICH COLLECTED IN CENTRAL AFRICA. 315

Porotricuum, Brid.

P. PENNÆFORME, C. Müll.

Hab. Kilimanjaro, Bishop Hannington. Found also in S.
Africa (Thamnium complanatum, Schimp.); also in Madagascar
(P. madagassum, Kiaer.), and on the Cameroon Mountains, G.
Thomson. tis very similar to P. longirostrum.

P. prerops, Rehm.
Hab. Kilimanjaro, Bishop Hannington. Natal, Dr. Rehmann.

P. vsaGARUM, sp.n. Folia rami primarii rotundo-ovata, apice
abrupta, apieulo parvo, nervo valido infra apicem abrupte desi-
nente, ramulina rotundo-oblonga apiculo brevi, omnia margine
apicem versus serrulata e cellulis densis ovalibus areolata.

Hab. Usagara Mountains.

Closely resembling the P. elegans from Samoa, in size, colour,
ramification, and appearance, but distinct in the form of its
foliage. It is one of a number of small species which differ more
in appearance than in any other particular from those which
agree in aspect with P. longirostrum.

These constitute the section denominated by C. Müller ** Pinna-
tella,” and he has described two species collected by Dr. Schwein-
furth in Monbuttu, and another from Comoro, P. Geheebii (Linn.
1876, p. 272), the description of the leaves of which applies nearly
to P. usagarum ; and this last may be but a more grown-out state
Which has not subsimple stems sparingly pinnate as stated in
the description. Other species are found in Madagascar and
Mauritius; all have a general resemblance to P. elegans, and agree
in the dull, not glossy, foliage and tree-like stems. From the
above the group of species designated by C. Müller Anastrephi-
dium differ in their stiff and more glossy foliage, and a great
tendency of the branches to run out into slender attenuated
flagelliform shoots. To this group belong the P. ramulosum, Mitt.
in Journ. Linn. Soc. vii. p. 160, from the Cameroons, from which it
seems difficult to distinguish P. pennaforme var. brachyphyllum,
Rehm. 334, from Inanda, Natal, gathered also in S. Africa by
Cooper, and P. flagellaceum, Mitt. in Trans. Linn. Soc. xxiii. t. 5,
from whieh the Homalia elongata, Duby, Mém. de la Soc. de
Phys. de Genéve, tom. xxi. t. 2. f. 3, may not be different.

2c2

816 MR. W. MITTEN ON MOSSES AND

PLAGIOTHECIUM, Bruch et Schimp.

P. NITIDIFOLIUM, Mitt.

Hab. Kilimanjaro, a few fragments only.

Quite different from P. membranosulum, C. Miill., Rehm. 389,
and from another species gathered by Dr. Rehmann on Table
Mountain.

STEREOPHYLLUM, Mitt.
S. nrvvraRE, Mitt. in Trans. Linn. Soc. xxiii. t. 5. f. 4.
Hab. Msilala.

The Hypnum andongense, Welw. et Duby, seems to be the
same species.

S. NITENS, Mitt. 7. c.—Homalia linguefolia, Welw. et Duby,
Mém. de la Soc. de Phys. de Genève, tom. xxi. p. 431.
Hab. Usagara Mountains.

So far as can be judged from the description, Duby’s species
appear to belong as above.

Fapronta, Raddi.

F. SCHIMPERIANA, Hampe.

Hab. Usagara Mountains, and on Kilimanjaro, Bishop Han-
nington.

In the specimens collected some of the leaves are slightly
serrulate and some almost entire; the length of the leaf-point
seems also variable, from one third to half the length of the leaf
itself.

Microruamnivum, Mitt. Muse. Aust.- Amer.

M. PSEUDOREPTANS, Jaeger, Adumb. ii. p. 494.—Hypnum,
C. Müll. Bot. Zeit. 1856, p. 439.

Hab. Kilimanjaro, Bishop Hannington.

M. fruticellum from the Cameroons is more slender, and may

be only a smaller form ; but M. frondosum from the same place is
quite different in its short wide leaves.

Prerogonium, Sw.
P. GRACILE, Sw.

Hab. Kilimanjaro, Bishop Hannington. Collected also in
Abyssinia, and by Dr. Rehmann in S. Africa.

HEPATICEH COLLECTED IN CENTRAL AFRICA. 817.

Hypnum, Dill.

H. (§ HELICODONTIUM) suscompressum, C. Müll.— Leskea,
Bruch et Schimp.

Hab. Kilimanjaro, Bishop Hannington.

The few stems gathered agree very nearly with the original,
collected by Schimper in Abyssinia (Un. Itin. 4816). The
Species very closely agrees with the S.-American H. conostomum
CEriodon), Montagne, and the H. Sprucei and H. salicicola.

H. (§ HELICODONTIUM) USAGARUM, sp. n. Monoicum, pusillum,
repens ; folia patentia ovato-lanceolata, margine superiore subser-
rulata, nervo longe infra medium obsoleto, cellulis superioribus
parvis oblongis, inferioribus ad angulos densioribus quadratis,
perichetialia apicibus subulatis; theca in pedunculo levi sub-
erecta ovalis, opereulo rostro subulato, peristomio dentibus
angustis pallide rubris, interno processus ciliiformibus sequilongis
rufidulis.

Hab. Usagara Mountains, on decayed wood, Bishop Han-
nington.

Caulis 5-6 mm. longus eum foliis 0°75 mm. latus ; folia pallide
viridia subnitida, 0°50 mm. longa; pedunculus 5 mm. longus,
ruber; theca matura viridis, 0°70 mm. longa. Operculi rostrum

paullo brevius.

H. (§ RIGODIUM) roxanroN, Schw.

Hab. Kilimanjaro, Bishop Hannington.

The specimens are without fruit; but they present no differ-
ences from states of the species as found on the western side of
S. America. The African specimens agree very nearly in size and
colour with those collected by Lechler (629), which were named
by Schimper Rigodium Lechleri, and described by C. Müller
in Bot. Zeit. 1858, p. 172, as Hypnum arborescens ; these are
rather a small state ; other Chilian examples, and especially those
from Chiloe, are much more grown-out, but in foliage and other
particulars present no differences. The species is remarkable
for its rigid stems, ramification, and foliage, which resemble those
of the larger Thuidia, without, however, having any papillæ.

H. (§ BRACHYTHECIUM) IMPLICATUM, Hornsch.
Hab. Kilimanjaro, Bishop Hannington.
Gathered in S. Africa by most collectors.

318 MR, W. MITTEN ON MOSSES AND

Brachythecium Dicksoni, Rehm. 383b, and B. Knysne, C. Müll.,
Rehm. 387, seem to be almost the same species. B. afro-
salebrosum, C. Müll. Rehm. 385, is more robust. B. velle-
rum, Mitt, from the Cameroons, is larger and very similar to
B. salebrosum ; its capsules are erect.

Awomopon, Hook. et Tayl.

A. DEVOLUTUS, Mitt.

Hab. Kilimanjaro, Bishop Hannington.

A few stems only, which seem to agree with the Indian species,
and are, with it, probably not distinct from the American
A. Toccoe, Sull.

Lrskza, Hedw.

L. cLavrraAMEA.—Pseudoleskea, C. Müll,

Hab. Kilimanjaro, Bishop Hannington.

Imperfect specimens, which agree better with Dr. Rehmann’s
so named than with the Pseudoleskea plagiostoma, Schimp., from
Abyssinia.

THUIDIUM, Schimp.

T. LxvIPES, sp. n. Monoicum ; caulis pinnatim divisus phyl-
lidiis vestitus, ramis simplicibus, ramulis paucis nudis; folia
caulina hastato cordata sicca incurva, nervo infra apicem evanes-
cente, ramea compressa patula ovali-oblonga oblique acuta
asymmetrica lateralia semitorta planiuscula, marginibus minute
crenulatis, cellulis rotundis minutis obscuris areolata, nervo
infra apicem evanescente pallidiore, perichetialia elongata api-
cibus subulatis inferne breviter fimbriata; theca in pedunculo
levi horizontalis ovalis.

Hab. Kilimanjaro, Bishop Hannington.

Habit and size that of T. versicolor, Hornsch.; but different in
the form of its branch-leaves. A considerable number of nearly
allied species have already been described from Africa, many of
them having, like T. gratum, a rough fruit-stalk.

T. TAMARISCINUM, Hedw.
Hab Kilimanjaro, Mr. H. H. Johnston and Bishop Han-
nington.
Corresponding in all respects with European specimens. It
` has been collected in S. Africa by Cooper.

HEPATICIE COLLECTED IN CENTRAL AFRICA. 319

FissrpENs, Hedw.

F. cELLULOSUS, sp. n. Pusillus flaecidus decumbens ; folia
alterna oblongo-lanceolata aeuta ubique limbo e duplici serie
cellularum anguste marginata, nervo angusto obscuriore infra
apicem evanido; cellulis magnis oblongo-hexagonis pellucidis
areolata.

Hab. Usagara Mountains, on the earth, Bishop Hannington.

Caulis 3-6 mm. altus pallidus tenuis. Folia patentia 1:50 mm.
longa.

These very incomplete specimens indicate the presence of
another species of that small group which contains F. hyalinus,
F. dealbatus, and the Conomitrium (§ Polypodiopsis) Metzgeria,
C. Müll., with a very lax pellucid areolation and an incomplete
or obsolete nerve.

PLAGIOCHILA, Nees et Mont.

P. SARMENTOSA, Lehm.

Hab. Kilimanjaro, Bishop Hannington.

The specimens are larger than those figured by Lindenberg,
and in this they agree with some collected on Table Mountain
by Mr. Eaton. Other specimens still more elongate have been
collected at Fernando Po, Mann, on the Niger, Barter; and
closely similar ones were sent out of M. Bojer's herbarium at
Mauritius by Dr. Ayres. All these seem to be really states of
one species, which varies in colour and in stature, but very little
in the outline of its leaves, which is similar to that of the European

P. asplenioides.

P. nrcuorowa, Web.
Hab. Kilimanjaro, gathered by Dr. Meller on Mount Chirad-

zura, Livingstone's Zambesi Expedition. It is also from the
Cameroons and Niger.

P. TEREBRANS, Nees.
Hab. Usagara Mountains and Kilimanjaro,

P. sqvAMULOSA, Mitt. in Journ. Linn. Soc. vii. p. 165.
Hab. Kilimanjaro.

P. stnvosa, sp. n. (Pl. XVI. figs. 8-10.) Rami ramulis sub-
pinnatim in flabellum dispositi phyllidiis pluribus parvis lanceo-

320 MR. W. MITTEN ON MOSSES AND

latis ventralibus obsiti; folia conferte imbricata fere ubique
squalia deltoidea, apice truncata rotundata pauci breviter den-
tata integerrimave, margine ventrali usque ad basin rotundatum
pluries sinuato undulata integerrima, dorsali integerrima basi
flexura quasi rotundato dilatato caulem cum basis folii oppositi
omnino tegente, involucralia conformia apice margine ventrali
labiisque perianthii obovati subtruncati undulati spinoso-dentata.

Hab. Kilimanjaro, Bishop Hannington. Usagara Mountains,
Last. Madagascar, Baron and Pool. Mauritius, Thomas.

PLAGIOCHILA JAVANICA, Nees et Mont.
Hab. Kilimanjaro. Gathered in S. Africa by Cooper, and on
Table Mountain by Mr. Eaton.

P. Lastir, sp. n. (Pl. XVII. figs. 1-3.) Rami ramulis
paucis irregularibus subpinnati; folia patula divaricatave ovato-
oblonga, apice satis angusta dentibus variis, margine dorsali
recurva decurrente, ventrali e medio ad apicem paucidentata,
basin versus rotundata plus minusve dentata, e cellulis rotundatis
areolata involucralia latiora, margine ventrali spinoso-dentata,
perianthium obovatum, ala dorsali apice dentata, labiis subtruncatis
spinoso-dentatis ; spice mascule apicibus foliorum rotundatis
divaricato-recurvis rotundatis integerrimis.

Hab. Usagara Mountains, Last.

This somewhat resembles small P. cristata, Sw., and P. Necke-
roidea, Mitt. in Trans. Linn. Soc. xxiii. t. 6; but it differs in the
more entire leaves of a different form.

P. BARTERI, sp. n. (Pl. XVI. figs. 11, 12.) Rami dichotomi
furcatimve divisi; folia imbricata divergentia ovato-ligulata,
apice rotundata margineque ventrali breviter dentata basin versus
dentibus longioribus magis approximatis, dorsali recurva in caulem
longum descendente basi denticulis circiter tribus eiliata.

-Hab. Sierra Leone, Barter.

P. pickaNna, sp. n. (Pl. XVII. figs. 4-6.) Rami furcati ;
folia imbricata a basi latiora patente sensim angustata subligulata
eurvata falcata divergentia, apice in erura duo subulata sinu lato
divisa, margine dorsali integerrima, ventrali dente brevi versus
medium imposito integerrimove involucralia, margine ventrali
inferiore spinoso-dentato ; perianthium obovatum, dorso ala con-
tinua integerrima, labiis rotundatis b:eviter ciliato-dentatis.

HEPATICZE COLLECTED IN CENTRAL AFRICA. 321

Hab. Madagascar, Andrangoalaka, G. W. Parker. Central
region, Baron.

P. ABYSSINICA, sp. n. (Pl. XVII. figs. 7-9.) Rami ramulis
paucis pinnatim dorsi latere ventrali ramentis crispulis vestiti ;
folia imbricata patenti-divergentia deltoideo ovata, apice lata
rotundata obtusissima, margineque ventrali breviter denticulata
basin versus integerrima sinuato-undulata, dorsali integerrima
recurva quasi abrupta caulem tegente, involucralia conformia
vix majora dentieulis parum majoribus, perianthium obovatum,
ala dorsali angusta, apice paucidentieulato, labiis rotundatis
dentibus brevibus.

Hab. Abyssinia, Schimper.

Letoscypuus, Mitt.

L. inruscatus. (Pl. XVII. figs. 10-14.)—L. repens, Mitt.
in Journ. Linn. Soc. vii. p. 165.

Hab. Kilimanjaro.

The habit of this species is nearly that of Lophocolea bidentata ;
aud in all the specimens the colour is brown. This is in most
respects very similar to a species collected in Java by Motley :—

L. Morrzvi; foliis suborbiculatis integris apiceve sinu latius-
culo breviter bidentata, amphigastria caule vix latiora quadriden-
tata, perichetialia breviter dentata, perianthium oblongum, labiis
rotundatis brevidentatis.

Hab. In monte Megamendong, alt. 4000-6000 ped., ad lignum
putridum. Excepting the rounder leaves, there seems to be little
difference.

JUNGERMANNIA, Linn.

J. MINUTA, Dicks.

Hab. Kilimanjaro, amongst Oncophorus cyathicarpus, Mr. H.
H. Johnston.

Cuanponantuus, Mitt. in Hook. Handb. N. Zeal. Fl. ii. 753.

C. HIRTELLUS.— Jungermannia hirtella, Web.—J. fimbriata,
Hook. Muse. Exot. t. 79.

Hab. Kilimanjaro; also à single stem amongst moss sent from
the Usagara by Mr. Last.

The leaf in the original specimen from Richard is four-lobed ;
the fourth being small, was overlocked in the figure.

322 MR. W. MITTEN ON MOSSES AND

Besides the above, five species referable to this group are
known :— C. javanicus, from Java ; C. indicus and C. rufus (Jung.
tripartita, Tayl.?), C. squarrosus, from New Zealand; and
C. fimbriatus, from Tahiti.

LEPIDOZIA, Dumort.

L. cuenEsstiNA, Lindenb.—Jungermannia, Sw.

Hab. Kilimanjaro.

Tall and much-branched specimens, some with perianths; they
seem to agree very nearly with Swartz's original specimen in the
form and substance of the leaf, although the stems and branches
are longer.

BAZZANIA, Gray.

B. convexa.—Jungermannia convexa, Thunb.
Hab. Usagara Mountains and Kilimanjaro.

B. pumita, sp. n. (Pl XVIII. figs. 1, 2.) Humilis, dense
cespitosa; caulis ramis dichotomis flabellatim dispositis ; folia
opposita patenti-divergentia imbricata late ovata, margine
dorsali arcuato, ventrali subrecto, apice truncato dentibus plus
minusve distinctis subtridentata, cellulis carnosulis rotundis ovali-
busque subpellucidis, amphigastria quadrato-rotundata, apice
erosa, basi cum basibus foliorum utroque anguste confluente
coalita.

Hab. Kilimanjaro.

Much less than B. decrescens, so common in Mauritius, and
with its areolation of smaller cells.

B. DEFLEXA.—Herpetium deflexa, Wees.
Hab. Kilimanjaro.

Isotacuis, Mitt.

I. AUBERTI.— Jungermannia Auberti, Schw.—J. serrulata,
Hook. Muse. Exot. t. 88.
Hab. Kilimanjaro. A fragment only, Bishop Hannington.

Herserra, Gray.
H. picrana, Tayl.
Hab. Kilimanjaro.
The specimens agree very nearly with Indian, both in ramifica-
tion and foliage, but are barren.

HEPATICZE COLLECTED IN CENTRAL AFRICA. 323

Rapvra, Dumort.

R. Boryana, Nees.
Hab. Kilimanjaro.

R. appressa, Mitt.
Hab. Kilimanjaro.

R. MASCARENA, Steph.
Hab. Kilimanjaro.
Fragments of some other species were also gathered.

Ponzura, Dill.

P. capensts.—Madotheca capensis, Géttsche.

Hab. Usagara Mountains, Bishop Hannington.

Very different in appearance from the S.-Afriean specimens,
being deep green and luxuriant.

PunaAGMiICOMA, Dumort.

P. FLOREA, sp. n. (Pl. XVIII. figs. 3, 4) Monoica; rami
dichotomi flabelliformes ; folia divergentia arcuata oblique acuta,
margine ventrali sinuata incurva, lobulo obsoleto, amphigastria
caulem triplo latiora orbiculata ; folia involucralia parum minora
subconformia, lobulo parvo liguleformi, amphigastrium oblon-
gum, apice recurvum sinu parvo bidentatum lateribus dentibus
paucis nullisve armatis; perianthia sepe plura oblonga obtusa,
dorso levia ventre apicem versus unicarinata, spice mascule
siepe prelonge.

Hab. Niger, Barter.

Brown, somewhat flaccid in stratified tufts.

P. EMERGENS, Mitt.
Hab. Usagara Mountains and Kilimanjaro; associated with
Leptodon Smithii.

P. PAPPEANA, Nees.

Hab. Kilimanjaro.

The specimens, mostly barren, appear to belong to this
S.-African species.

324 MR. W. MITTEN ON MOSSES AND

LEJEUNEA, Lib.

L. (§ ODONTOLEJEUNEA) TORTUOSA, Lehm. et Lindenb.—-Phrag-
micoma Palisoti, Tayl. Flores monoici; perianthium ventre lævi
uni- bicarinatumve.

Hab. Usagara Mountains ; gathered also by Mr. Last.

L. ($ OponroteseuNEA) Hannineront, sp. n. (Pl. XVIII.
figs. 5-8.) Dioica ? ; rami fertiles assurgentes foliis patenti-diver-
gentibus, inferioribus subovalibus obtusis integerrimis, superiori-
bus ovato-oblongis oblique obtusis, margine breviter paucidentatis
integerrimisve ventrali incurvis; basi lobulo parvo pliceformi,
amphigastriis fere orbiculatis parum longioribus quam latis
paucidentatis ; foliis involucralibus ovatis acutis amphigastrioque
ovali magis dentatis, lobulo fere obsoleto, perianthio obcordato
haud exserto carina ventrali marginibusque dentatis; rami
masculi foliis patentibus ovato-lanceolatis, lobulo elongato, mar-
ginibus amphigastrioque integerrimis.

Hab. Usagara Mountains, on leaves.

L. ($ LoPHoLEJEUNEA) ATRA, sp.n. (Pl. XVIII. figs. 9-12.)
Monoica, gracilis nigrescens humilis; folia late ovata obtusius-
cule acuta, lobulo fere ad medium marginis ventrali producto
ovato-oblonga, apice rectangulari, amphigastria caulem triplo
latiora orbieularia involucralia majora perianthium obcordatum,
limbo dentato alatum, ventre carinulis apice paucidenticulatis
tegens.

Hab. Kilimanjaro.

Similar in habit and appearance to Z. subfusca; but differing
from most of the allied species in the outline of its leaves.

L. ($ LoruorLEJEUNEA) siNUVATA, sp. n. (Pl. XVIII. figs.
13-16.) Repens, ramosa; folia ovato-rotundata lobulo ad
medium marginis ventralis ascendente ibique sinuata, lobulum
subobeordatum ostendente; amphigastria orbiculata caule quad-
ruplo latiora, involucralia majora conformia, lobulo oblongo, peri-
anthio obovato ventre convexo lateribus ala integerrima limbato.

Hab. Mauritius, creeping on bark, Dr. Ayres.

L. ($ PRIONOLEJEUNEA) SERRULA, sp. n. (Pl. XIX. figs. 1-3.)
Monoica ; folia ovata acuta, cellulis marginalibus angulis promi-
nulis acutis erenulata grossiuscule areolata, lobulo subovato

HEPATICE COLLECTED IN CENTRAL AFRICA. 325

involuto, amphigastria caule parum latiora; perianthium obcor-
datum, marginibus superioribus denticulatis.
Hab. Niger River, creeping on Plagiochila neckeroidea (Barter).

LEJEUNEA (§ PRIONOLEJEUNEA) DEPLANATA, sp. n. (PI. XIX.
figs. 4-6.) Monoica; folia lata ovata obtusa subacutave, margine
minute crenulata, cellulis minutis areolata, lobulo obsoleto, amphi-
gastria caule vix duplo latiora; perianthio obcordato, mar-
ginibus superioribus crenulatis.

Hab. Island of Bourbon, on Buzzania decrescens, ex herb.
Montagne.

L. ($ LEeProcorEa) apua#siva, sp. n. (Pl. XIX. figs. 7-9.)
Monoica, ad folia viva repens planissima affixa pinnatim ramosa ;
folia rotundato-ovata imbricata, margine cellulis tenuissimis an-
gustis a reliquis diversiformibus pallidis limbata, lobulo in laci-
niam lanceolato-ligulatam ovalemve cauli subparallelam producto
in ramis fertilibus subquadrato saccato varieformi; folia involu-
cralia majora obovata lobulis subovatis; perianthium (juvenile ?)
immersum late obovatum planum ventre linea carinzformi nota-
tum; antheridia in saccis foliorum infra involucralium inclusa.

Hab. Usagara Mountains.

Very pale brown, adhering to the leaves on which it grows as
if by the suctorial leaf-margins ; in this particular it agrees with
many other species of this group. The perianths are all alike
and of the same small size, and are probably immature. The
form of the lobule is unusual ; but the same elongation is seen in
the Indian Z. producta.

L. (8 Correa) prerratis, sp. n. (Pl. XIX. figs. 10, 11.)
Repens, pinnatim ramosa; folia patenti-divergentia subovato-
oblonga, margine ventrali involuto apice in cucullum oblongum
teretem obtusiusculum tertiam partem foli longitudinis occu-
pantem producta amphigastria profunde, bicrura.

Hab. Usagara Mountains, on leaves.

The fragmentary specimens show that this is a species in size
and habit approaching the L. superba of the Pacific Islands.

L. (8 CERATANTHE) CORNUTA ?, Lindenb.

Hab. Usagara Mountains.

The specimens, mere fragments, clearly indicate the presence of
a species of this group and, like some others from the western
coast, are too incomplete to be certain of their identity.

326 MR. W. MITTEN ON MOSSES AND

Leseunea (§ EULEJEUNEA) FLAVA, Sw.
Hab. Usagara Mountains.

L. (8 EvLEJEUNEA) acuta, Mitt.
Hab. Usagara Mountains.

FnurLLANIA, Raddi.

F. Eckrouir, Spreng.

Hab. Kilimanjaro and Usagara Mountains.

All the S.-African specimens which appear to belong to this
species are not in a condition to show if the inflorescence is
uniformly monoicous, as is the case in those from Central Africa
and in F. emergens from the Cameroons. Similar species are
found in India as well as in S. America, and it is most probable
that one wide-spread species inhabits all these regions.

F. squARROSA, Nees.
Hab. Kilimanjaro.

F. usagara, sp.n. (Pl. XIX. figs. 12-14) Caulis ramis
paucis remotis, subpinnatim ramosus ; folia patentia patulave late
ovata apiee rotundata, lobulo magno semigloboso evolutove;
amphigastria rotunda caule triplo latiora sinu acuto fissa,
dentibus acutis ad latera externa paucidentatis; folia involu-
eralia lobulo acuto dentato cum amphigastrio bipartito, laciniis
dentatis coalita; perianthium emergens obovatum triquetrum
iuferne phyllidiis parvis et supra in carinis pluriseriatis asperu-
lum ; spice masculz ovales.

-Hab. Forest of Ugogo and Usagara Mountains, on bark.

The general appearance of this species is like that of F. squar-
rosa, but it is in all its parts much smaller. The small lacini
on the two lateral and ventral carina of the perianth appear to
be arranged in several rows.

F. rrinervis, Lehm. et Lindenb.
Hab. Kilimanjaro and Usagara Mountains.

F. BRUNNEA, Spreng.
Hab. Kilimanjaro. <A fragment only.

F. corpata, Mitt. in Journ. Linn. Soc. vii. p. 168.

Hab. Kilimanjaro.
The inflorescence is in this monoicous.

HEPATICZE COLLECTED IN CENTRAL AFRICA. 327

FRULLANIA ANGULATA, Mitt. in Journ. Linn. Soc. vii. p. 169.
Hab. Usagara Mountains.

F. REPLICATA, Nees.
Hab. Gold Coast, Capt. Burton and Com. Cameron.

MzrzaERIA, Raddi.
M. rurcata, Linn.
Hab. Usagara Mountains.

M. myriavopa, Lindberg.
Hab. Usagara Mountains.

CyaATHODIUM, Kunze, Gottsche, Lindenb. et Nees, Syn.
Hep. p. 577.
Synhymenium, Griffith, Notul. p. 844.
Riecia, Dickson, fasc. 4.

C. AFRICANUM, $p. n. Sporis levibus ; frondes tenerrima pal-
lide viridia sicca chartacea laxe areolata dentes capsule octo
e seriebus sex cellularum fuscarum conditi; spore fusce sub-
hemisph:erieze depresse, margine rotundate.

Hab. Usagara Mountains, Bishop Hannington.

So far as can be seen in the small quantity obtained, this
species is entirely of the same habit and structure as the Indian
C. aureonitens, Griff. Icon. 69. D. ii. It differs in its smooth
spores and probably in the shorter teeth-like divisions of the
apex of the capsule; but both species agree in having the in-
volucra situate in the axil of the marginal divisions of the edge
of the fronds, as represented by Griffith, and is also shown in the
figure given by Dickson of his Riccia spuria, fasc. 4, t. ii. fig. 16.
The plants appear to grow in their earlier stages radiating from
a centre, but soon become confused from the lobes of the same
or adjacent individuals growing over them ; and as the fructifi-
cation is concealed from view above, they might be readily passed
over as the prothallus of some fern. The areolation is of so loose
and tender a substance that it revives after desiccation very irre-
gularly or not at all.

Griffith had the opportunity of examining the Indian species in
a living state, and his original drawing has no doubt been so
modified by the native artist that the mouth of the involucre is
represented as a globular body opening by fissures of which two

|

i
t

328 MR. W. MITTEN ON MOSSES AND ;

are lateral, and these should represent the orifice of the pouch-
like involuere, the middle lobe having to all appearance been
added by the artist in place of the apex of the capsule, the other
figures (2, 3, 4, 6, 7) being true to nature.

TanarOoNIA, Micheli.

T. HYPOPHYLLA, Linn.

Hab. Kilimanjaro and Usagara Mountains.

There seems to be no difference between these specimens and
European, the spores are alike in all.

Fimprrarra, Nees.

F. Boryana ?, Mont.
Hab. Kilimanjaro.
Specimens with very young fruit may belong to this.

DESCRIPTION OF THE PLATES.
Prate XV.

Figs. 1-4. Archidium africanum, sp. n. 1, Natural size; 2, portion of stem
with perichetium and male flower; 3, stem-leaf; 4, perichetial leaf,
enlarged.

Figs. 5-8. Archidium Rehmanni,sp.n. 5, Natural size; 6, entire rim ; 7, stem-
leaves ; 8, perichzetial leaves.

Figs. 9-12. Campylopus perpusillus, sp. n. 9, Natural size; 10, stem-leaves ;
11, perichextium with capsule and calyptra; 12, portion of peristome,
enlarged.

Figs. 18-16. Hyophila plicata, sp. n. 18, Natural size; 14, leaves; 15, capsule
with (a) calyptra and (5) portiors of peristome; 16, male flower, all
enlarged.

Figs. 17-19. Anæctangium pusillum, sp. n. 17, Natural size; 18, leaves; 19,
capsule, enlarged.

Figs. 20-22. Pterobryum Hanningtoni, sp. n. £0, Natural size; 21, stem-
leaves; 22, perichzetium, with (a) leaf detached.

Puare XVI.

Figs. 1-3. Pterobryum Jlagelliferum, sp. n. 1, Natural size; 2, stem-leaf ; 3,
perichetium, enlarged.

Figs. 4-7. Erpodium Hanningtoni, sp. n. 4, Natural size; 5, portion of stem
with fruit-bearing branch ; 6, stem-leaves and (a, b) sections; 7, cap-
sule and calyptra, enlarged. :

Figs. 8-10. Plagiochila sinuosa, sp. n. 8, Natural size; 9, leaves ; 10, perianth
and involucral leaf, enlarged.

Figs. 11, 12. Plagiochila Barteri, sp. n. 11, Natural size; 12, leaves, enlarged.

HEPATICH COLLECTED IN CENTRAL AFRICA. 329

Puate XVII.

Figs. 1-3. Plagiochila Lastii, sp. n. 1, Naturalsize; 2, leaves ; 3, perianth with
involucral leaf.

Figs. 4-6. Plagiochila dicrana, sp. n. 4, Natural size; 5, leaves; 6, perianth
and involucral leaf, enlarged.

Figs. 7-9. Plagiochila abyssinica, sp.n. 7, Natural size; 8, portions of stem
with leaves ; 9, perianth and involucral leaf, enlarged.

Figs. 10-14. Leioscyphus infuscatus, nob. 10, Natural size ; 11, portion of stem
with leaves ; 12, amphigastrium ; 13, involucral leaves and perianth ;
14, perianth seen from the side, enlarged.

Puate XVIII.

Figs. 1,2. Bazzania pumila, sp. n. 1, Natural size; 2, portion of stem with

leaves and amphigastrium.
igs. 3, 4. Phragmicoma florea, sp. n. 8, Stem, natural size; 4, portion of same
with perianth and male spike, enlarged.

Figs. 5-8. Lejeunea ($ Odontolejeunea) Hanningtoni, sp.n. 5, Natural size; 6,
portion of stem with leaves; 7, fertile branch and (a) section of
perianth; 8, portion of male spike, enlarged.

Figs. 9-12. Lejeunea (§ Lopholejewnca) atra, sp. n. 9, Natural size; 10, portion of
stem with leaves and amphigastria ; 11, perianth and involucral leaves ;
12, transverse section of perianth.

Figs. 13-16. Lejeunea (§ Lopholejeunea) sinuata, sp. n. 13, Natural size; 14,
portion of stem with leaves and amphigastrium ; 15, perianth with
involucral leaves; 16, transverse section of perianth.

Puate XIX.

Figs. 1-3, Lejeunea ($ Prionolejeunea) serrula, sp. n. 1, Natural size ; 2, portion
of stem with leaves and amphigastria; 3, perianth with involucral
leaves and (a) transverse section, enlarged.

Figs. 4-6. Lejeunea ($ Prionolejeunea) deplanata, sp. n. 4, Natural size ; 5, por-
tion of stem with leaves and amphigastria ; 6, involucral leaves and
perianth with (2) its transverse section. :

Figs. 7-9. Lejeunea ($ Leptocolea) adhesiva, sp. n. 7, Natural size; 8, portion
of stem with leaves and (a) cells more enlarged ; 9, fertile branch. ;

Figs. 10-11. Lejeunea (8 Colura) digitalis, sp. n. 10, Natural size; 11, portion
of stem with leaves and amphigastria, enlarged.

Figs. 12-14. Frullania usagara,n.sp.? 12, Natural size; 13, portion of stem
with leaves and amphigastria ; 14, perianth and involucre, enlarged.

LINN. JOURN.—BOTANY, VOL. XXII. 2D

330 MR. H. TRIMEN ON BALANOPHORA THWAITESII.

Note on Balanophora,Thwaitesii, Eichl.
By Henry Tery, M.B., F.LS.

[Read 3rd June, 1886.]

At the commencement of the present year my friend Mr. Fawcett,
of the British Museum, then engaged on the paper upon certain
Balanophoree which he has since communicated to this Society,
wrote to me in Ceylon asking me if I could give him any infor-
mation about a plant mentioned by Eichler in his monograph of
the Order in the * Prodromus, and named by him Balanophora
Thwaitesii, but not described. In reply I informed Mr. Fawcett
what material existed at Peradeniya; and this I have brought
home with me, and now exhibit to the Society in illustration
of this note, which can be regarded only as a small addendum to
Mr. Fawcett's paper.

The name B. Thwaitesii was given (“pro tem. ") in 1873 * to a
supposed new species from Ceylon, represented only by a coloured
drawing of a male plant, sent by Thwaites to the Kew herbarium f.
Eichler’s MS. note attached to this drawing is as follows:—
“ Hee nova species mihi videtur nomine Balanophora Thwaitesii
in DC. Prod. commemorata nee tamen descripta." In the ‘ Pro-
dromus ' he gives the few characters visible in the drawing; but
nothing further has been published concerning the plant.

The original specimens from which the drawing was made are
preserved in the Peradeniya herbarium, and are now exhibited.
They were collected in * Dec. 1854" at Raxawa, a place at an
elevation of some 4000 feet. The late Dr. Thwaitesi did not
distinguish them from B. indica, and placed them under the
same C.P. number 200. No female specimens were collected.

In January 1881 one of the Garden coolies brought me fine
living specimens, both male and female, of this plant, which he
sald he had colleeted at Nuwara Eliya, a locality at over 6000
feet elevation. From an examination of this fresh material $ I
then arrived at the conclusion that Eichler was correct in con-
sidering the plant distinet as a species from B. indica; and in

* DC. Prod. xvii. p. 146.

+ Being a facsimile of one in the Peradeniya Collection.

1 Thwaites, Enum. Fl. Zeyl. p. 293 (1861).

$ I regret that, not knowing at the time the rarity of the plant, I preserved
dried specimens only, and put none into spirit.

MR. H. TRIMEN ON BALANOPHORA THWAITESII. 331

my recently published List of Ceylon Plants * I so entered it.
It has not been again met with in Ceylon, and must be rare,
whilst B. indica is very frequent in the mountain forests t.

The following description of B. Thwaitesii was made from
the living plants above mentioned :—Rhizome coral-like, much
branched or lobed, studded at intervals with very large stellate
pustules; heads apparently dicecious, bursting irregularly through
the rind of the rhizome, ultimately rather long-stalked. Male
plant entirely bright lemon-yellow, the young bracts orange-
coloured, peduncle variable in length, completely concealed by
large fleshy, elongated, oblong-obtuse decurrent leaf-scales, which
pass into flower-braets above; flower-bracts always conspicuous,
being as long as the buds in the upper part and as the pedicels
ot fully opened flowers below, thick and fleshy, strongly ribbed
above and channelled beneath so that the ends are erenate in
outline; flowers large, not very densely crowded; perianth-

segments over 2 inch long, reflexed; anther-head j inch, cells

pon Female plant rather taller, the head 3 inches by
21, ovold-clavate, and exserted beyond the leaf-scales, brownish
yellow; flowers minute, the ovaries shortly stalked, ovoid or
pyriform, nearly always inserted on the receptacle, and surround-
ing the bases of the stalked oblong-ovate truncate brownish
spadicels ; styles very long, tapering, somewhat exceeding the
spadicels.

The whole plant is full of a waxy juice and has a strong mousy
smell. Its large size and bright yellow colour give it at first
sight a very different look from B. indica, but yet it is difficult to
formulate definite characters. Those which appear to be the
most valuable are :—the large size of the stellate pustules on the
rhizome, the prominent, longer, crenated flower-bracts of the male
heads, the ovoid or pear-shaped exserted female flower-heads, and
the short-stalked female flowers (ovaries) inserted on the re-
ceptacle and very rarely attached to the bases of the spadicels.
I had hoped that this last character would have proved an abso-
lute one, the rest being merely comparative ; but a further
examination and comparison along with Mr. Fawcett has shown me
that occasionally the ovaries have their insertion on the spadicel-

stalks, as in B. indica.

* Trimen, Syst. Cat. Ceylon Plants, p. 77 (1885). ik
t The specimens in spirit of B. indica exhibited were collected by Prof. Bower

of Glasgow, during his visit to Ceylon in the early part of this year.

332 MR. W. BOTTING HEMSLEY ON THE

B. Thwaitesii should be searched for in the S.-Indian moun-
tains. In looking over the sheets of B. indica in the British
Museum, a female plant collected by Wight in the Nilgiris came
under notice, which may probably be the same. The flowers,
however, are very much destroyed, which renders the disposition
of the ovaries somewhat uncertain.

In the Kew Herbarium is a male plant from Ceylon (Gardner,
no. 839) which is unlike the usual condition of B. indica; but it
is too young to enable one to say positively that itis B. 'Awaitesi?.
Possibly this may be one of the specimens referred with doubt
to B. elongata by Dr. (now Sir) Joseph Hooker in his well-known
paper in our Transactions *.

Whether B. Thwaitesii, Eichl., is entitled to rank as a species
may remain a matter of opinion until further investigation. I
am inclined now rather to regard it as a well-marked variety of
B. indica. No doubt it will again be met with in Ceylon, and
this note may assist its identification in Peninsular India or

Malaya. "

f

Report on the/ Vegetation of Diego Garcia. By W. Borrine
Hemstey, A.L.S., Assistant for India in the Herbarium,
Royal Gardens, Kew.

[Read 17th June, 1886.]

Dırco Garcia is the name of an island and three contiguous
islets forming part of the Chagos Archipelago, in the Indian
Ocean, and lying in about the same latitude and eighteen degrees
east of the Seychelles, and in the same longitude as the Laccadive
and Maldive Islands. They are a dependency of the Mauritius,
yet until recently nothing definite was known of their vegetation.
Though smaller in extent than the largest island of the Seychelles,
it was thought possible that they might also possess an endemic
element in their flora, especially as the only plant recorded from
Diego Garcia was a Fern, Asplenium equabile, Baker, which is
not known to occur elsewhere.

* Trans. Linn. Soc. xxii. p. 45 (1858).

VEGETATION OF DIEGO GARCIA. 333

Bory de St. Vincent (‘Voyage dans les quatre Principales
Iles des Mers d'Afrique; 1804, iii. p. 247), refers to the
islands in these words :—“ On arme à l'Ile de France diverse
embarcations, qui ne font d'autre commerce que d'aller a
Praslin à Diégo Garcias et sur d'autres ecueils chercher des
cocos pour en faire de l'huile." In 1885, in connection with
the expedition of Mr. Bourne, to be referred to hereafter, Prof.
Moseley, F.R.S., kindly drew the attention of the Director
of Kew to the following extract from Findlay's ‘Directory
for the Navigation of the Indian Ocean, 2nd ed. London,
1870, p. 459:—" On nearly every part of these islands there
grow at intervals great clumps of gigantic trees, the Bois
Mapan or Rose Tree, which attains an enormous size and height,
even to 200 feet. Their growth and decay is most rapid, and
their fallen and decayed trunks form a large portion of the vege-
table world of the Archipelago.” This statement increased our
desire that the islands should be explored botanically. In 1883
an opportunity occurred. H.M.S. ‘Moorhen’ was to call at
Diego Garcia on her homeward voyage from China; and at the
instigation of Mr. Wykeham Perry (Secretary to the Commodore
in Chief, China Station), Mr. A. Hume (Assistant Paymaster,
R.N.) made a small collection of plants there and forwarded
them to Kew. This collection, although regarded as very im-
perfect by Mr. Hume, dispelled to a great extent the idea of the
existence of endemie forms, as it consisted entirely of the com-
monest of tropical littoral plants. Nevertheless it did not con-
clusively prove that the flora was wholly of comparatively recent
derivation, and as Diego Garcia had by this time been made a
coaling-station of the ‘ Orient’ line of steamers from London to
Australia, an opportunity for a more complete exploration was
not far distant. In 1885 Mr. Gilbert C. Bourne, F.L.S., was
furnished with funds by the Government Grant Committee of
the Royal Society for investigating the marine invertebrate fauna,
and he at the same time undertook to botanize the islands. This
he appears to have done thoroughly; but before proceeding to
discuss his collection, I will here give some particulars concerning
the islands, communicated by Mr. Bourne.

* Diego Garcia," he writes, ^is an Atoll fourteen miles long
by six miles and a half broad, entirely of coral formation, and
nowhere rising more than ten feet above high tide, excepting
where, in a few places, sand heaps have been piled up by the

334 MR. W. BOTTING HEMSLEY ON THE

wind. The ring of land encircling the lagoon is nearly continuous,
and averages one third of a mile in width. At the north end
the strip of land is discontinuous; here there is an entrance into
the harbour five miles wide, broken up by three small islets into
four channels. The whole of the main island is densely covered
with vegetation, Scevola Koenigii and Tournefortia argentea
reaching down to the high-tide level on either shore, and forming,
especially the Scevola, a thick impenetrable mass along both
shores. Within this mass of bushes come cocoa-nut palms,
which have been artificially plauted only in a few places; over
the larger part of the island they have grown wild. The under-
growth beneath the palms is often very thick, consisting generally
of Guettarda speciosa. There are no cocoa-nut trees on the three
islands in the mouth of the harbour; but the whole of the main
island is densely covered with them. I have frequently seen
cocoa-nuts germinating on the shores in places where the young
plant was lapped by the waves at the highest spring-tides ; these
young plants seemed to be perfectly healthy. The shores are
not, I should think, favourable for the germination of many
kinds of seeds that may be fortuitously brought to them, being
composed almost entirely of coral blocks or coral-sand. There is
a little cultivation in the island at the three principal habita-
tions of Point Marianne on the west side of the island, and East
Point and Minny Minny on the east side. In the Gardens
bananas, sweet potatoes, bitter oranges, citrons, and a few other
tropical fruits are grown. Maize is cultivated at Minny Minny.
Ihave forgotten to mention that the banyan tree occurs on the
island. There are several large specimens, but it does not appear
in my collection, for it neither flowered nor fruited during my
stay. The only noticeable feature about it is that, owing no
doubt to the poverty of the soil, its leaves are much smaller than
usual."

In connection with the vegetation of oceanic islands generally
I published a list of the plants collected by Mr. Hume in the
Botany of the * Challenger' Expedition, part 3, p. 115, omitting,
however, to add the sign of doubt after the name Achyranthes
velutina, Hook.et Arn. This list and the collections sent by Mr.
Bourne the latter part of last year and the beginning of this,
prove that the flora of Diego Garcia is altogether derivative, and
apparently of comparatively recent derivation. The fern, Asple-

VEGETATION OF DIEGO GARCIA. 335

nium cquabile, Mr. Bourne sought in vain ; consequently some
doubts arise concerning the home of the e specimen sent
by Lady Barkly ; but if it exist in Diego Garcia, it must be ex-
ceedingly scarce. As to the gigantic treesof Findlay's ‘Directory’
Mr. Bourne writes:—“ The inhabitants knew nothing of any
such trees, and pointed out Pisonia as being the largest tree on
the island. That same Directory makes several remarkable and
untrue statements about the island, among which must be classed
the gigantic trees."

Without accepting the enormous dimensions recorded in Find-
lay, I think there is a probability that the statement as to the
former existence of large trees was based upon facts; and “ Bois
Mapan ” is perhaps a misprint for Bois Mapou, the name of a
district in Mauritius, and a name applied to various trees in dif-
ferent islands of the Indian Ocean. At the present day, according
to Mr. Bourne, Pisonia inermis, the largest tree in the island,
bears the name of Bois Mapou in Diego Garcia; and on the
authority of Dr. Balfour (‘Botany of Rodriguez, p. 64), the
endemic Pisonia viscosa is known by the same name in Rodri-
guez.

Going farther back, it appears from Bojer (‘Hortus Mauri-
tianus,’ p. 265), that this name was applied to Pisonia macro-
phylla in the island of Galega, and to Cissus Mappia in Mauritius
itself, where it has given the name to a distriet; while in Mail-
lard’s ‘Notes sur l'Ile de la Réunion,’ 2™ éd. i. p. 145, it is
stated that Monimia rotundifolia is the Mapou of Bourbon
Island. What the meaning of the word “ Mapou” may be is
not evident ; but as it is employed to designate such very differ-
ent trees as those named, it may well have included another in
former times.

It is very probable, however, that the statement in Findlay’s
Directory is a jumble of fact and fiction, and all that we can
accept of it is that there were formerly clumps of large trees all
over the island. Mr. Thiselton Dyer suggests that the big tree
of earlier times was the * Gayae" (Afzelia bijuga), basing his
opinion on a passage in an unpublished report on the timber
trees of the Seychelles by Mr. C. Button, Crown Conservator of
Forests, Seychelles. He says :—‘ This tree, already very scarce,
acquires gigantic proportions, more especially in the sandy flats,
the largest trees having been found in the Chagos group, Solomon,

336 MR. W. BOTTING HEMSLEY ON THE

Six Iles, &e. ; their diameter attaining six or seven feet, and their
height fifty or sixty feet from the base to the branches. Its
timber has been so much sought after that in many places it has
been completely destroyed without pity, and its place taken by
cocoa-nut trees. Gayac is in great demand for ship-building;
nearly all the ships built at Seychelles having been principally
constructed of this wood.”

The Gayac is a member of the Leguminose found in the Sey-
chelles, Madagasear, the Malay peninsula and archipelago, and
in Western Polynesia, and is everywhere valued for its timber.
When Mr. Bourne was at Diego Garcia there was only one clump
of four or five trees remaining; and he was informed that it did
not increase in consequence of the rats eating the seeds as soon
as they fall.

Cordia subcordata is another tree inhabiting Diego Garcia
which attains a large size, but only when it forms forests in humid
situations. On the authority of Forster (Seemann, ‘ Flora Vi-
tiensis, p. 169), it occasionally reaches a height of eighty feet
and upwards; and when Darwin visited the Keeling Islands, in
the eastern part of the Indian Ocean, there were considerable
forests of it in those islands. Still this could scarcely have been
the * Bois Mapan” of Diego Garcia, because it has a very hard
durable wood. Hibiscus tiliaceus, a common oceanic tree, not
now found in Diego Garcia, grows rapidly and has a soft perish-
able wood. But what the large trees of Diego Garcia were
must remain conjectural, unless there are any records in the
Paris Museum, for it seems that the French surveyed and
explored the islands in 1744, though no results have been
published.

Below is a tabular view of the probably indigenous plants of
Diego Garcia with their general distribution.

YEGETATION OF DIEGO GARCIA. 337

A List of the Vascular Plants found in Diego Garcia, which pro-
bably reached the Islands independently of the agency of Man*.

d ; »
S -J gott.
z e a a
PORTULACACEÆ.
1. Portulaca quadrifida, Linn............ en 9 11
MALvACEX.
2. Sida diffusa, H. B.K. ......... eem Lod ee ae
TILIACE.
3. Triumfetta procumbens, For&f. ............ ee S * *
SIMARUBEZ.
4. Suriana maritima, Linn. ........ ens x ate, Seder pees © ve
LEGUMINOSE.
5. Afzelia bijuga, 4. Gray... mnn Sejt a
Mad.
COMBRETACEÆ®.
6. Terminalia Catappa, Linn. ....... nmn M i*a
MxYRTACEX.
7. Barringtonia speciosa, Linn. f. ...... en M |I ea
RuB1IACEX.
8. Morinda citrifolia, Lin. ......... enn * * | # | owe | x
9. Guettarda speciosa, Linn. .....e en NI, TI. D
COMPOSITÆ.
10. Vernonia cinerea, Less.........e.eseosserersereeeree. x i-l|sl*
ll. Ageratum conyzoides, Linn. ...... m ee | Re e
GoopENOVIE.
12. Scævola Koenigii, Vahl ......sseceerrereener seers sS oT.T-*
APOCYNACER.
13. Ochrosia borbonica, Gel. ........... 2i * | *
BonaAGINEE.
14. Cordia subcordata, Lam.......... mns kriki
15. Tournefortia argentea, Linn. f. ..... nn 2& L4]. 13
CoNVOLVULACEZ.
16. Ipomoea grandiflora, Lam. ...«««eem eis te ae
t: biloba, Forsk. TT * xisi]
SCROPHULARINEÆ.
18. Herpestis Monniera, HBE es 56 x |a xx
19. Striga hirsuta, JDenthe E ctia * tivit

* In the African column S. denotes Seychelles, M. Mauritius, and Mad.
Madagascar; and the plants to which they refer have not been found in Conti-
nental Africa.

LINN, JOURN.—BOTANY, VOL. XXII. 2E

388 MR. W. BOTTING HEMSLEY ON THE

CEN
: Spe
B b E £ B
E al ey cH
a lalala]
VERBENACES.
20. Lippia nodilora, Mean c.s.. Vics as eie tho Nasa a TTITIL
21. Promna serratifolia; Denm. eee x * XS
NYCTAGINE.
22. Boerhuavin diffusa; Linn..-......-.. o nes * wc
23-Pisonia Inermis, LOr. <... r- o. ao dedere
AMARANTACES,
24. Achyranthes aspera, Linn, ..................eeeese * * oe *
LavunINEE.
25. Hernandia peltata, Meissn. .......................- * Xx NN
26. oyipera; LANN. 3 eee oe * *
21- Gassytheefibformis Linn. sts: ec nos * * x [x *
EuPHonRBIACEE.
28. Euphorbia pilulifera, Zénn............ 414.11113- * * x X x
29. Phyllahtliud Niruri, LANN. -en eoe en * e
30. Acalypha indica, ath. -oa ne wow Ba]
CaASsUARINES.
31. Casuarina equisetifolia, Forsz...................... E x pas ow
CyYPERACE.
32: Cyperus hoularis, Lanm. oi e * x pw ox
33. Kyllinga monocephala, Pott. ..................... * x bx a es
34. Fimbristylis glomerata, Nees | ..................... * eth te | cal
GRAMINES.
35. Panicum sanguinale, Linn, ........................ pet ahd um n
36. Stenotaphrum complanatum, Schrank. ......... * Li Ds
Lycopopracea.
31; Estlotum-triguebrum: Nn.. 2 x * xiaj at
Fitices.
38. Asplenium squabile, Zaer........................
39 longissimum, Blume ios s acces, M. | «
40; — — Nidas: D ee 7 VOX * x [C ex
4] — — falcatum: Lom — ee V 1* E es) ote |
42. Nephrodium unitum, R. Br, ..................... »lcákk]elqerp.
49. Pteris marginata, Bory ue vvv sve v e xls. * x |a

Of the thirty-six flowering plants in the foregoing table, fifteen
are trees or shrubs; and fourteen of the total occur in Africa
(or the Mascarene Islands), Asia, Polynesia, Australia, and
America; twenty-five in four out of fiveof these geographical areas;
and thirty-four of them in three of the areas.

Mr. Bourne writes that he believes they fully represent the

YEGETATION OF DIEGO GARCIA. 339

indigenous flora of Diego Garcia, except the “ Lily of the Island,”
which he was unable to preserve, from the effects of damp after
drying a specimen. This * Lily" is probably one of the common
maritime species of Crinum, though Mr. Bourne was of opinion
that it was peculiar to the islands *.

From what is known of the distribution of the plants here
treated as indigenous, they may have all reached the islands by
natural means, such as oceanic currents and birds; yet it is also
possible that some of them may have been intentionally or unin-
tentionally introduced by man. Mr. Bourne states that he was
careful not to collect evidently introduced plants; but the fol-
lowing collected by him can only be classed as such :— Gossypium
barbadense, Linn., Triphasia trifoliata, DO., Passiflora suberosa,
Linn., Asclepias curassavica, Linn., Capsicum frutescens, Linn., and
Rivina levis, Linn. ; hence they were not included in the table.

It would unduly lengthen this sketch to enter into details re-
specting the diffusion of each species of the native plants; yet
I may be permitted to add that full particulars on this point for
nearly all of them are given in the Botany of the ‘ Challenger’
Expedition. Not one species is restricted even to the islands of
the Indian Ocean; thus Ochrosia borbonica, which, more than any
other, is characteristic of the region, is also found in the tidal
forests of Ceylon, the Andaman Islands, and Singapore. It has,
however, been proved that the seeds of most of them are un-
injured by long immersion in the sea that they are often cast
ashore on remote islands in a germinating condition, and that
some of them at least are chiefly littoral in their wide distribution
from these causes, but they flourish equally well beyond the
immediate influence of the sea.

Mr. Bourne collected a few cellular Cryptogams, including
some new forms, descriptions of which are appended.

MUSCI.

Catymperes ($ HYOPHILINA) Garcia, Mitt., n. sp.
“ Cæspitosum, humile; folia supra basin erectam pallidam
obovatam, margine superiore crenulatam, intus cellulis pellucidis

* Since writing the foregoing, I have seen a young plant of a Crinum raised
at Kew from seed sent thither by Sir John Kirk, who obtained it from Diego
Garcia. Until the plant flowers it will be impossible to determine the species,
but Mr. Baker regards it as being nearly allied to Crinum Hildebrandtit, Vatke
(Bot. Mag. t. 6709), a native of the mountains of Johanna Island.

340 ON THE VEGETATION OF DIEGO GARCTA.

areolatam patentia oblongo-ligulata, apice obtusiuscule mucro-
nata sepius autem angustata anomala, cellulis minutissimis haud
opacis areolata, nervo crassiusculo dorso minute scabro percursa,
margine incurva integerrima, limbo ubique carente.

* Island of Diego Garcia, Mr. Œ. C. Bourne.

* More slender than the best known species of this group, which
agree generally with the African C. Afzelii, aud differing in
the entire absence of the marginal or intramarginal limb. In
some species from the islands of the Pacific the limb becomes
obsolete, but in these the leaf-outline is usually wider."
W. MITTEN.

ALGE.
CAULERPA MAMILLOSA, Mont.

* CLADOPHORA DEMISSA, J. Ag. MSS., ad nullam descriptam
speciem hanc referre ausus sum.

“ De hac mihi adnotavi:

“ Ad Confervam Echinus habitu paulisper accedit, at structura
abunde diversa, fere Valoniam referens (insunt inter fila Confervæ
fragmenta Corallin: et Microdictyi).—J. G. AGARDH.”

* DasYA INDICA, J. Ag. MSS., est species nondum descripta.

* De hae specie mihi adnotavi: Nonnullis EE ad
Dasyam stuposam J. Ag. (Sp. Alg. vol. ii. pars 8, p. 1197) —
D. crassipes, Harv. Alg. Ceylon, no. 7 (non Harv. Austral);
aliis characteribus ad Dasyam Cardicanam accedit. Frons est
inferne corticata, ramulis fasciculato-fastigiatis sursum porrectis,
subsecundatis, inferiore parte articulata at corticata, superiore
monosiphonia.—J. G. AGARDH.”

HALIMEDA Opuntia, Lamour.
SARGASSUM, sp. (too imperfect for determination).

TURBINARIA VULGARIS, var. DECURRENS, J. Ág.

Funar.

POLYPORUS SANGUINEUS, Fries. An almost cosmopolitan
species.

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 341

Phytobiologieal Observations; On the Forms. of ‘Seedlings and
the Causes to which they are due. By Sir Jonny LUBBOCK,
Bart., Pres. Linnean Society, F.R.S.; M.P., D.C.L., LL.D.

[Read as a Presidential Address, 24th May, 1886.]
o Forms or LEAVES AND OF COTYLEDONS.
Introduction: Forms of Leaves.

I nave elsewhere * called attention to the forms of Leaves, and
discussed the causes to which we might ascribe the endless differ-
ences which they assume. Vertical leaves, for instance, are gene-
rally long and narrow, horizontal ones have a tendency towards
width, which brings the centre of gravity nearer to the point of
support. Wide leaves, again, are sometimes heart-shaped, some-
times lobed. The former shape is obviously that which would
arise if a leaf is gradually widened at the base; and I have pointed
out that in many species with lobed leaves—for instance Passi-
flora, Cephalandra, Hibiscus, &c.—the first, or few first, leaves are
entire and more or less cordate. The cordate form, then, appears
to be the early, the palmate a later form. But what advantage
does the palmate form present ?

Now in cordate leaves with veins following the curvature of
the leaf, as, for instance, in Zamus (fig. 1, p. 842), the vascular
bundles pursue necessarily a curved course ; while in palmate
leaves, as in Acer (fig. 2), the veins are straight; and it is
clearly an advantage that the main channels which convey the
nutritive fluid should hold a direct course. In such cases the
leaf naturally assumes the lobed form with a vein running to
the point of each lobe. There has indeed been some question
whether the path of the sap lies mainly in the cell-walls or in
the cell-cavities ; but the evidence seems to point strongly to the

latter view t. The tracheids of, say, the Yew *are at least
times as long as they are broad, so that in

acheid the water-
3? +
T.

seventy or eighty
travelling transversely the length of a single tr
current has to traverse seventy cell-walls instead of one
then, to the question why some plants should have

In reply,
s have palmate

cordate leaves with curved veins, while other

* Journal Royal Institution, 1885; ‘Contemporary Review, 1885; and

‘Nature Series— Flowers, Fruits, and Leaves.’
+ See, for instance, Darwin and Phillips, “ On the Transpiration Stream in

Cut Branches,” Proc. Cambridge Phil. Soe. vol. v. 1 Loc. cit. p. 364.
LINN. JOURN.— BOTANY, VOL XXII. 2F

342 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

leaves with straight ones, I suggested that the first retains the
old form of leaf, while the latter has assumed one which presents
certain advantages.

Leaf of Tamus, to show the Leaf of Sycamore, to show the
curved course of the veins. straight* course of the veins.

I also pointed out that in many cases the form of the leaf was
determined by the distance between the buds taken in connection
with the strength of the stem, and illustrated this by numerous
examples. For instance, the Spanish Chestnut has a stronger
stem than the Beech, but the distances between the buds are not
much greater; consequently the leaves may be much larger, but
cannot be much broader; hence they assume the dagger-like form
characteristic of that species. Again, in Conifers, the species with
short leaves generally retain their leaves for several years, those
. with longer ones for a shorter period; the length of the leaf seeming
to be a function of the strength of the shoot, taken in relation to
the longevity of the leaf. The vast majority, indeed, of Conifers
are evergreens, though some few, such as the Larch (Larix), are
deciduous. This species flourishes especially in Siberia and the
higher alpine slopes of Switzerland, while elsewhere the Spruce
gains the upper hand. These two species are deadly enemies.
Generally the Spruce has the advantage, as the young Spruce will
grow under Larch, while the shade of the Spruce is fatal to most
other plants. But in Siberia and in the upper Alpine slopes of
Switzerland the short summer is perhaps too short for the Spruce,
while giving sufficient time for the Larch.

It has been suggested by Mr. Grant Allen that the tendency
to division in subaqueous leaves is dueto the comparative paucity
of carbonic acid in water as compared with air. I have ventured
to suggest a different explanation. Of course it is important to

SIR J. LUBBOCK —PHYTOBIOLOGICAL OBSERVATIONS. 343

expose as large a surface as may be to the action of the water.
We know that the gills of fish consist of a number of thin plates,
which, while in water, float apart, but have not sufficient con-
sistence to support even their own weight, much less any ex-
ternal force, and consequently collapse in air. The same thing
happens with thin, finely-cut leaves. In still water they afford
the greatest possible extent of surface with the least expenditure
of effort in the formation of skeleton. This is, I believe, the ex-
_ planation of the prevalence of this form in subaqueous leaves, as,
for instance, in Myriophyllum, Hottonia, Utricularia, Ranunculus,
and many others.

Again, in still air, the conditions, except so far as they are
modified by the weight, would approximate to those of water ;
but the more the plant is exposed to wind the more would it
require strengthening. Hence, perhaps, the fact that herbs so
much oftener have finely-cut leaves than is the case with trees.
In the Umbellifers, for instance, almost all the species have the
leaves much divided, more, I need hardly say, than is the case
with trees. Shrubs and trees are eharacterized by more or less
entire leaves, such as those of the Laurel, Beech, Hornbeam,
Lime; or by similarly-shaped leaflets, as in the Ash, Horse-
chestnut, and Walnut.

There are, however, many groups of plants which, while
habitually herbaceous, contain some shrubby species, or vice versd.
Let us take some groups of this description, in which the herba-
ceous species have their leaves much cut up, and see what is the
character of the foliage in the shrubby species.

The vast majority of Umbellifers, as I have just observed, are
herbaceous and with leaves much divided, the common Carrot
being a typical example. One European species, however (Bu-
pleurum fruticosum), is a shrub attaining a height of more than
6 feet, and has the leaves coriaceous and oblong-lanceolate.

Most of our common Groundsels (Senecio), again, are low herbs
with much-cut leaves. Some species of Senecio, however, are
shrubby, and their leaves assume a totally different character—
Senecio laurifolius and S. populifolius having, as their specific
names denote, leaves respectively resembling the Laurel and
Poplar. In the genus Oxalis, again, there is a shrubby species—
Q. Laureola, with leaves like those of a Laurel.

I would venture, then, to suggest that this throws light on the
reason why herbaceous plants so often have their leaves cut up.

2r2

944 SIR J. LUBBOCK— PHYTOBIOLOGICAL OBSERVATIONS.

On Seedlings.

The consideration of leaves naturally led to that of Cotyledons.

I had already paid some attention to the form and structure of
Seeds, on which I had the honour of addressing the Royal Insti-
tution in a lecture which appeared in its * Journal,’ and which I
have since republished in the small volume already referred to.

I have since made a further study of Seeds and Seedlings, with
special reference to the characteristics presented by the cotyledons
and the first leaves. In this I have been greatly aided by the
kindness with which the immense resources of Kew have been
placed at my disposal, and for which my grateful thanks are due
to Sir Joseph Hooker, Mr. Thiselton Dyer, Mr. Watson, and
indeed the whole staff. I have also been permitted to make use
of the Laboratory for which Science is indebted to the liberality
of the late Mr. Jodrell. Mr. Carruthers and the other officers
of the British Museum have kindly aided me in various ways;
Mr. Lynch has sent me some seeds from the Botanie Garden at
Cambridge ; and Mr. Hanbury has been good enough to supply
me with several seeds from his rich garden'on the Riviera.

No one who has ever looked at seedlings can fail to have been
struck by the contrast which the cotyledons afford, not only to
the final leaves, but even to those by which they are immediately
followed.

I propose to take certain plants (especially, as far as possible,
the commonest and most familiar), and see what light can be
thrown on the varied forms which their seedlin gs present. Look,
for instance at the familiar Mustard and Cress; the first (fig. 3)
has kidney-shaped cotyledons, one of them rather larger than the
other; while the Cress, Lepidium sativum (fig. 4), on the other
hand, has the cotyledons divided into three lobes. The Pink
(Dianthus) has (fig. 22) broad cotyledons, the Chickweed and
Cerastium (fig. 23) narrow ones ; those of the Beech (fig. 13) are
fan-shaped in outline; those of the Sycamore shaped almost like
a knife; those of the Eschscholtzia (fig. 40) divided like a hay-
fork ; those of the Bean or Acorn thick and fleshy.

Mustard and Cress were the delight and wonder of our child-
hood; but at that time it never occurred, to me at least, to ask
why they were formed as they are, and why they differed so much.
So they grew, and beyond that it did not occur to me, nor I think
to most, that it was possible to inquire.

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 345

I propose, however, to-day to suggest reasons which may account
for some, at any rate, of these differences.

In the Memoirs to which I have already referred, I have dis-
cussed the causes which regulate the forms of seeds, and will not
now therefore enter into them. I may just, however, observe
that the shape of the cotyledons seems to have little, if any,
influence on the form of the seed.

Fig. 3. Fig. 4.

fugare ae

Seedling of Mustard (Brassica Seedling of Cress (Lepi-
nigra), X 3. dium sativum), X 3.

Forms of Ovules.

It may be convenient that I should shortly define a few tech-
nical terms which I shall have to use frequently. The fruit con-
sists of the ovary and whatever other parts persist when the seed
is ripe. The seed is the ovule fertilized by the pollen. The festa
is the outer shell or coat of the seed. The micropyle is a small
opening through which the pollen-tube obtains access to the
interior of the ovule. The funicle is the stalk by which the seed
is attached. The Aum is the scar left on the seed at the point
where it was attached to the funicle. The chalaza is the spot to
which the nutritive vessels reach before ramifying in the seed.
In some seeds the chalaza and hilum coincide; when this is not

346 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

the case they are connected by the raphe. The embryo or young
plant consists (1) of the radicle or base of the embryo, which
gives origin to the root, (2) of one or two cotyledons or future
seed-leaves, and (3) the plumule, or bud of the future stem between
the cotyledons. It is embedded in perisperm, a mealy, oily, fleshy,
or horn-like substance, which serves to nourish the young plant,
but is in many cases entirely absorbed by the embryo by the time
the seed is ripe.

If we take one of the simplest seeds in avery young state, say
that of a Buckwheat or Nettle, we shall find that it is a small,
upright, more or less pear-shaped body attached at the base,
and with a small opening at the free end, which is called the
* mieropyle" or little doorway. Seeds thus formed are called
“orthotropous” or straight. But this, though the simplest, is
the least common form of seed. More often, as in the Bean,
Chickweed, and many others, the ovule during its growth is
curved on itself, so that the apex and the “micropyle” are
brought down very near to the base. Such ovules are known as
* eampylotropous " or curved.

In a third category the seed itself is straight, but it stands as
it were at right angles to the base of attachment. These are
called “ half anatropous”’ or half reversed.

Lastly, in a very large number of seeds—in fact by far the largest
—the seed is quite inverted on its base; the funiculus, or stalk
growing with the seed, forming a sort of ridge or “ raphe ” as it
is technically called, which is very conspicuous in the ovule, but
gradually becomes less and less apparent, and which completely
reverses the seed. The “ chalaza” or place of attachment of the
seed itself, and from which the nutritive vessels ramify, is thus
carried right away from the “hilum” or point of attachment
to the opposite end, and the whole seed is reversed, so that the
true base is removed from, and the true apex brought close to,
the point of union with the ovary. Such seeds are called “ ana-
tropous” or reversed.

Now this seems a very curious and roundabout arrangement.
It is described in all works on general Botany, but those which I
have seen do not give any explanation of its object or purpose.
will for the moment omit any consideration of “ campylotropous
and ** balf anatropous " seeds, and confine myself to orthotropous
an4 anatropous forms.

In orthotropous species, as for instance in the Buckwheat, the

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 347

ovule is straight, upright, and attached by its base. At the free end
is the “ mieropyle," and immediately below it, embedded in the
seed, is a small group of special cells—the rudiment or embryo of
the future plant. Now when the pollen falls on the stigma it
soon begins to grow. It pushes out a little tube, which rapidly
elongates, passes into the cavity of the ovary, enters the micro-
pyle or little door, left open, and conveniently placed for it, and
vivifies the rudiment of the future plant.

But ovaries thus constituted are, as I have already mentioned,
quite exceptional. In a great many plants the seed, instead of
being upright and rising at the base of the ovary, is, on the con-
trary, attached to the summit and pendent; and in still more the
ovary contains a large number of seeds. Now in a pendent
“orthotropous”’ ovule, the micropyle would be turned away
from the pollen-tube; but the object and effect of the seed
being reversed or *anatropous " is to bring the micropyle back
into a convenient position. Soalso, when there are many ovules,
the result of the anatropous arrangement is again to bring the
micropyle into a suitable position.

The structure and arrangement of the ovule have been the
subject of various important memoirs, which, however, have been
almost exclusively descriptive. Dalmer* indeed quotes Schleiden,
who, speaking of Berberis, pointed out that among the normal
anatropous ovules here and there one occurs which is ortho-
tropous, and that these never develop into seeds (** Die Befruch-
tung der Ovule scheint mir aber in ganz bestimmter Beziehung
zu dem vorgeschriebenen Lauf der Pollenschliuche zu stehen,
deun stets beobachtete ich, dass diese regelwidrige atropen
Eichen unbefruchtet blieben"). The pollen-tubes, however,
would naturally follow the regular course; and we could not,
I think, deduce a general conclusion from such rare and
abnormal cases. Dalmer himself does not seem to bave done
80; for, after referring to the different forms of seeds, he ob-
serves that “sometimes the form of the ovule appears to be
adapted so as to facilitate the entrance of the pollen-tube ”
(“ scheint zuweilen die Gestalt des Ovulums dem Eintritt des Pol-
lenschlauches angepasst zu sein"). Even here, then, the expla-
nation I venture to suggest seems referred to rather as an excep-
tional occurrence than as a general explanation of this remarkable

* * Ueber die Leitung der Pollenschlàuche bei den Angiospermen," Jenaisch.
Zeits. 1880, p. 530.

318 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

arrangement. It must be admitted that there are some cases in
whieh the anatropy of the seed appears at first sight rather dis-
advantageous than otherwise. Most of these can, I believe, be
explained, while in some it is possible that the plants retain, even
perhaps to their present disadvantage, au arrangement inherited
from an ancestral condition in which it was beneficial. I hope,
however, to enter into this more fully on a future occasion.

When we cut open a seed, we find within ita more or less dif-
ferentiated embryo ; sometimes, as in the Ash (Fraxinus, fig. 48),
or Castor Oil (Ricinus, fig. 54), &e., a lovelj little miniature plant,
generally white, but in some species green, distinctly formed,
but still embedded in food-material or ** perisperm." Such plants
occupy an intermediate place between those in which, as for
instance in the Larkspur (Delphinium, fig. 130), the embryo is
very small, and those, on the other hand, where, as in Hippophaé
(fig. 55), the perisperm is reduced almost to a film, or, as in the
Bean, where the embryo occupies the whole seed, and the
nourishment intended for the young plant is stored up in the
cotyledons themselves.

In a Walnut, for instance, the two halves of the seed are the
two cotyledons, and attached to them may be found the little
plantlet with a delicate white root, and a little stalk bearing five
or six minute rudiments of leaves, often just tipped with green.

I wil now pass from seeds to seedlings. No one who has
ever looked at young plants ean fail to have been struck by
the eontrast they afford to the older specimens belonging to the
same species. This arises partly from differenees in the leaves,
partly from the contrast which the cotyledons, or seed-leaves;
afford, not only to the final leaves, but even to those by which
they are immediately followed.

This contrast between the cotyledons and true leaves is 50
great that one might almost be pardoned for asking whether they
can be brought into actual correlation, or whether the cotyledons,
with the portion of stem belonging to them, might occupy some
such relation to the rest of the plant as the prothallium does to
the fronds of the Fern.

This, however, is not the direction in which we must look for
the true explanation of the differences.

So far, indeed, are the cotyledons from agreeing with the forms of
the leaves, that the difficulty is to find any which have been clearly
influenced by them. One species of Ipomea CE Pescapræ) has both

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 349

cotyledons and leaves, as the name denotes, somewhat like the foot
of a goat; but the leaves vary considerably, and it is probable that
the resemblance may be accidental. A clear case is, however,
afforded by the Onagrariex, where in @nothera and some allied
genera the form of the mature cotyledons is evidently related to the
leaves, as I hope to show in a subsequent communication. Even
here, however, the resemblance is confined to a basal portion of
the cotyledon which makes its appearance subsequent to germi-
nation, aud no trace of it is shown in the eotyledons themselves
when they first appear.

The forms of the cotyledons in many species have been the
subject of special memoirs by Tittmann, Irmisch, Wichura,
Winkler, Tscherning, and other botanists; but they have not
given any reasons for the various forms assumed.

Klebs, indeed, in his interesting Memoir on Germination *,
refers to this diversity of form, and expressly says that these dif-
ferences are an enigma (“ Sind gewiss diese Verschiedenheiten in
den Blattformen hinsiehtlieh ihrer biologischen Bedeutung für die
Pflanze ein Rüthsel"). He observes, however, that on the whole the
forms of cotyledons are much simpler than those of leaves, and he
suggests that while in some cases, perhaps like the first leaves, they
retain the form which characterized the species in bygone ages,
we may rather, as a more generally applicable explanation, apply
to them the suggestion of Goebel with reference to stipules, and
regard them as simplified by arrest t (* Ihre einfache Form einer
ähnlichen Hemmung verdanken, welche vielleicht überwunden
wird, wenn es gelingt, die Kotyledonen lüngere Zeit als Haupt-
assiniilationsorgane bei jungen Pflanzen zu erbalten”). Even,
however, if this suggestion were the real explanation of the
comparative simplicity, it would throw no light on the differences
between the cotyledons of different species.

Though cotyledons do not present nearly such extensive va-
riations as leaves, still they do differ considerably from one
another.

Forms of Cotyledons.

Some are narrow, in illustration of which I may mention Feni-
culum (fig. 5), Coreopsis (fig. 6), Ceratocephalus (fig. 7), and
Ferula (in the hollow stalk, or ferule, of which Prometheus
E “ Beiträge zur Morphologie und Biologie der Keimung," Untersuch. Botan.
Inst. v. Tübingen, 1884.

t Loc. cit. p. 613,

350 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

brought down fire from heaven), &c., where the ultimate leaves
are much divided; Platanus (fig. 8) and Acer (fig. 9), where the
ultimate leaves are palmate; and Chenopodium (fig. 10), where
they are more or less triangular.

Fig. 5. Fig. 6.

Fig. 5. Seedling of Feniculum vulgare. Half nat. size.

Fig. 6. Seedling of Coreopsis filifolia. Half nat. size.

Fig. 7. Seedling of Ceratocephalus falcatus. Nat. size. The numerals indicate
the successive leaves.

Some cotyledons are broad, in illustration of which I give
figures of Ricinus (fig. I Impatiens (fig. 12), Beech (Fagus)
(fig. 13), Brassica (fig. 3), Hippophaé (fig. 14), Rivina (fig. 15),
Ruellia * (fig. 16), Rhus typhina (fig. 17), and Flax (Linum)
(fig. 19). We find some specles with narrow cotyledons and
broad leaves, as Menispermum (fig. 18) and Olea cuspidata (fig. 20);

* Tn Asystasia coromandeliana there is an interesting peculiarity. The first
pair of leaves of each branch, or at any rate of the lower branches, approximate
to the form of the cotyledons,

SIR J, LUBBOCK —PHYTOBIOLOGICAL OBSERVATIONS. 351

Seedling of Platanus. Seedling of Acer Pseudo-Platanus,
Nat. size? Half nat. size.
Fig. 10. Fig. 11.

Seedling of Chenopodium Bonus- Seedling of Ricinus sanguineus.
Henricus. Nat. size. One fourth nat. size.

while in others the cotyledons are broad and the leaves narrow, as
in Flax (Linum monogynum) (fig. 19), Hakea (fig. 21), and the Pink
(Dianthus) (fig. 22), &e. :

In many cases we find instances of broad and narrow cotyledons
in the same family, as in Chickweed (fig. 23) and Pink (fig. 22),
belonging to the Caryophyllacee ; sometimes even in the same
genus, as Galium saccharatum (fig. 24) and Galium Aparine
(figs. 113 & 114). 3

In some cases the two cotyledons are unequal, as in the

352 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

Fig. 13.

— XL

ELE m

Seedling of Zmpafiens Balsamina. Seedling of Beech (Fagus sylvatica).
Half nat. size. Half nat. size.

Fig. 14.

Seedling of Hippophaé rhamnoides, Seedling of Rivina levis.
Half nat. size. Nat. size.

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 353

Fig. 16.

Seedling of Ruellia flava. Seedling of Rhus typhina.
Half nat. size. Half nat.'size.
Fig. 19.
Fig. 18.

|

Seedling of Menispermum canadense. Seedling of Flax (Linum mono-
Half nat. size. gynum), Nat. size.

354 SIR J. LUBBOCK—PHYTOBIOLOGICATL OBSERVATIONS.

Seedling of Olea cuspidata. Seedling of Hakea acicularis.
Two thirds'nat. size. Half uat. size.

Fig. 22. Seedling of Pink (Dianthus Caryophyllus). Nat. size.
Fig. 23. Seedling of Cerastium arvense. Half nat. size.
Fig. 24. Seedling of Galium saccharatum. Nat. size.

~

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS 355

Mustard (fig. 3), Cabbage, Radish, Cereus (fig. 79), Pachira*, &c.:

in some the two halves of each cotyledon are unequal, as in the
Geraniums (fig.

25), or they are otherwise unsymmetrical, as in
the d. Laburnum (fig. 26), Clitoria (fig. 27), Schinus (fig. 28).

Fig. 26.

Seedling of Geranium sanguineum. Seedling of Laburnum (Cytisus
Nat. size. vulgare). ' Nat. size.

Seedling of Clitoria ternatea. Seedling of Schinus terebinthifolius.
Half nat. size. Two thirds nat. size.

* As described by Mr. Lynch in our Journal, vol. xvii. p. 147.

356 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

Seedling of Delphinium elatum.
Two thirds nat. size.

Seedling of Delphinium Staphisagria. Seedling of Delphinium nudicaule.
Half nat. size. Half nat. size.

SIR J. LUBBOCK— PHYTOBIOLOGICAL OBSERVATIONS. 357
Sometimes they are sessile, as Acer (fig. 9), Hakea (fig. 21),
Laburn um (fig. 26), &e. ; sometimes they are supported on petioles,
as in Microloma (fig. 29), which again are occasionally connate.
These differences sometimes occur in very closely allied species ;
for instance in Delphinium Staphisagria (fig. 31) the cotyledons are
sessile, while in D. elatum (fig. 30) they are petioled, and in
D. nudicaule (fig. 32) the petioles are connate.

Generally the cotyledons are entire, but sometimes crenate, as
in Cordia (fig. 33), or lobed, as in Pelargonium (fig. 34) and

Seedling of Pelargonium australe.
Half nat. size.

Seedling of Cordia subcordata.
Half nat. size.

Malva (fig. 85). Often they are emarginate, as in Impatiens
(fig. 12), Mustard (fig. 3) and Cabbage, {pomea (fig. 106), Con-
volvulus, Galium (fig. 114), Eucalyptus (fig. 38), Pentapetes (fig.
39), &e. Nay, they are sometimes even bifid, as in .EscAscholtzia
(fig. 40) and Zpom«a dasysperma (fig. 111) ; trifid, as in the Cress
(Lepidium, fig. 4), or in four long lobes, as in Pterocarya (fig. 41).
Sometimes auricled at the base, as in Poferium (fig. 42) and
Cuphea. Sometimes they are large; sometimes small. Generally
they are leaf-like; but sometimes, as in Rhus Thunbergii (fig. 44),
Sapindus (fig. 43), and among English plants in the Oak, Nut,
LINN. JOURN.—BOTANY, VOL. XXII. 2G

358 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

Peas, Bean, Mercurialis perennis*, Melittis melissophyllum t,
Nympha, Nuphar, Rhamnus Frangula, Trientalis, Daphne, &c.,
they are thick and fleshy.

Seedling of Malva moschata.

Seedling of Spathodea campanulata.
Nat. size.

Nat. size.

Ga
ERN

Seedling of Catalpa Kempferi. Seedling of Eucalyptus, SP-
Half nat. size. Half nat. size.
* Winkler, * Ueber die Keimpflanze des Mercurialis perennis," Flora, 1880.
t Irmisch, ‘‘ Zur Naturg. v. Melittis,” Bot. Zeit. 1858, p. 283.

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 859

Fig. 40.

Seedling of Pentapetes phenicia. Seedling of Eschscholtzia californica.

Nat. size. Nat. Size.
Fig 42.

x d Bee

TEX

1

Seedling of Píegrocarya caucasica. Seedling of Poterium Sanguisorba,
Nat. size. Twice nat. size.

262

3860 SIR J. LUBBOCK—PIIYTOBIOLOGICAL OBSERVATIONS.

Narrow Cotyledons.

Let us now begin with such species as have narrow cotyledons,
and see if we can throw any light on this characteristic. The

Fig 43.

Seedling of Sapindus inequalis. Seedling of Rhus Thunbergit.
Half nat. size. Half hat. size.

problem is simple enough in such cases as Platanus, where we have
on the one hand narrow cotyledons (fig. 8) and on the other hand
a long narrow seed (fig. 45, p) fully occupied by a straight
embryo. Again, in the Ash (Fraxinus, fig. 46), Ursinia (fig. 47),
&c., the cotyledons lie parallel to the longer axis of the seed, which
is narrow and elongated. Such cases, however, are comparatively
few; and there are a large number of species in which the seeds
are broad and even orbicular, while yet the cotyledons are narrow,
as for instance in Chenopodium (fig. 48) and Menispermum (fig. 49).

In these it will generally be found thaf the cotyledons lie
transversely to the seed. In Menispermum (fig. 18) the fruit

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 3531

Fig. 45.

Fig. 45. A. Fruit of Platanus, longitudinal section, X 6. w, woody part.
B. Transverse’section, X 12.
Fig. 46. A. Samara of Fraxinus excelsior, nat. size, with on

the seed pulled out. pa, placental axis; «6,
f, funicle. B. Longitudinal section of seed.

e half removed and
abortive ovule ;

Except where otherwise stated, c stands for cotyledon; oc, outer cotyledon ;
te, inner cotyledon ; ch, chalaza; e, embryo ; f, funiculus; 2, hilum; A, micro-

pyle; p, perisperm ; r, radicle; /, testa.

Fig. 47. Achene of Ursinia speciosa. Longitudinal section, X 2). Trans-

verse section; x 10.

Fig. 48. Chenopodium Bonus-Henricus, vertical and transverse sections of seed,

362 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

(figs. 49 and 50) is laterally compressed and horseshoe-shaped, with
a crest along the edge; the seed (fig. 51) conforms to the shape
of the fruit, and the embryo is curved and linear, the cotyledons
being applied to one another face to face, and at right angles to
the plane of the seed, so that the edges of the cotyledons touch
the walls of the seed at each side.

Fig. 49. Fig. 50.

Menispermum canadense.
Fig. 49. Germinating seedling, x 2. Fig. 50. Vertical section of seed, X 4.
Fig. 51. Transverse section of seed, x 2.

Fig. 52.

Acer Pseudo-Platanus.
Fruit, nat. size. A, B. Embryo, showing two modes of arrangement of
cotyledons.

The Sycamore (Acer Pseudo-Platanus, fig. 9) has also narrow
cotyledons; but the arrangement is very different. The fruit
(fig. 52) is winged, the seed somewhat obovoid and aperispermic—
that is to say, the embryo, instead of lying embedded in food-
material, occupies the whole cavity of the seed. Now if we.
wished to pack a leaf into a cavity of this form, it would be found
convenient to choose one of a long strap-like shape, and then roll
it up into a sort of ball. This is, I believe, the reason why this
form of cotyledon is most suitable in the case of the Sycamore.
The mode of folding, however, as shown in fig. 52, A and B, is not

SIR J. LUBBOCK—PHY TOBIOLOGICAL OBSERVATIONS. 363

always the same. I shall suggest a reason for the difference

further on.
Broad Cotyledons.

I now pass to species with broad cotyledons. The Acorn, Nut,
Bean (Phaseolus, fig. 53), and Pea aff rd familiar cases, in which

Fig. 54.

- p

Fig. 53. Phaseolus multiflorus. Section of seed parallel and vertical to coty-
ledons, three fourths nat. size. a, @, auricles; ar, arillode.

Fig. 54. Ricinus sanguineus. A. Longitudinal section of seed, nat. size. B. Trans-
Y verse section of seed, nat. size:

Fig. 55. Hippophaé rhamnoides. A. Longitudinal section of seed, x 4. D.
*'l'ransverse section of seed. dd, depressed line.
the two broad, fleshy, thickened cotyledons conform to and occupy
the whole seed. In the Castor-oil plant (Ricinus, fig. 11) the
seed (fig. 54) is ovoid, oblong, somewhat compressed dorso-
ventrally, and beautifully mottled, while the projecting knob at the
hilum gives it very much the appearance of a beetle or large tick.
The perisperm is abundant, fleshy, white, and surrounds the
embryo. The embryo is straight, flat, large, central, and white ;
the cotyledons broad, obtuse oblong, and approximately following
the general outline of the seed. In Hippophaé (fig. 14) we have
a somewhat similar case; but the cotyledons are fleshy and occupy
almost the whole of the seed (fig. 55). In Euonymus, again,
the seed is obovoid, slightly compressed laterally. The peri-
sperm is abundant, fleshy, firm, and white, entirely surrounding
the embryo. The embryo is straight, flat, central, pale green,
extending very nearly from one end of the seed to the other.
The arrangement is very similar in the Apple. In the Flax the

361 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

seed is ovate, obliquely pointed, glabrous, plano-convex, laterally
much compressed, placed edgeways on the placenta, and the coty-
ledons lie parallel to the flattened axis of the seed. In other
cases the seeds are still more flattened, as in Ailanthus, Passiflora,
Cobea, Stephanotis, &e. É

The Composite generally have narrow cotyledons. In MMos-
charia, however, they are somewhat broader, the seed (fig. 56)
being obovoid, with the cotyledons lying the broad way.

A Fig. 57.

Fig. 56. Moscharia pinnatifida. A. Longitudinal section of seed, x 8. B.
“Transverse section of seed, x 8. a a, auricles.

Fig. 57. Ruellia longifolia. A. Longitudinal section of seed, x 10. B. Trans-
“Verse section of seed, x 10. FH, fringe of hairs; LAw, larger,
and SAw, smaller auricle.

In many cases, seeds of the same shape produce cotyledons
of very different form.

Compare together, for instance, Rwellia (fig. 16 seedling, and
fig. 57 seed) and Cerastium (fig. 23 seedling, and fig. 58 seed).
Both have compressed, nearly orbicular seeds, but in Ruellia
the cotyledons are broad, in the Chickweed they are narrow. If,
however, we make sections of the seeds, the cause of this difference
becomes obvious, because in one (fig. 57) the cotyledons lie parallel,
in the other (fig. 58) transverse, to the seed.

The form of the cotyledons often differs greatly even in the

same family.

The Caryophyllacee, the family to which the Pink belongs,
afford an ‘interesting illustration. The cotyledons are placed
with their backs to the placenta, and in most species are narrow,
as 1n Cerastium (fig. 23). In some of them, however, such
as the Pink itself (Dianthus, fig. 22) and Tunica, they are

wide.

SIR J. LUBBOCK — PH YTOBIOLOGICAL OBSERVATIONS. 365

Now in most genera, as in the Chickw eed, Stellaria, Spergu-
laria, Cerastium, &e., the seeds are laterally ‘compressed ; the
cotyledons consequently lie transversely to the seed, and their
width therefore is limited by the thickness of the seed, as in
fig. 8. The case is, however, somewhat complicated by the fact
that the seed and embryo are both curved.

Fig. 58.

Cerastium arvense, X 195.

L4
On the other hand, in the Pink (fig. 59) the seeds are not
lateraly but dorsally compressed, attached to the columnar
placenta by the middle of the interior face, so that the cotyledons
are straight, parallel to the seeds, and have in consequence plenty
of room to widen out.

Bigs 50.

a

Dianthus Caryophyllus, X 15.
In figs. 58 & 59 the letters a, b, and c, d indicate the directions of the sections.

In Solanum the fruit is globose, scarlet when ripe, gla-
brous, indehiscent, and many-seeded. The seeds are kidney-
shaped, much compressed laterally, and placed with their narrow

366 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

edge to the placenta, surrounded by a paler margin, glabrous
and white; the hilum is small, and on the middle of the ventral
edge. The embryo is, in the mature seed, much curved, em-
bedded in, but lying near the outer edge of, the perisperm. The
radicle occupies the lower and narrow part of the seed. The
cotyledons are linear, not broader than the radicle, curved, with
their tips close to the hilum, and their back to the placental axis,
and at right angles to the plane of the seed, the whole width of
which aecordingly they oceupy, so that they cannot grow any
wider. On the other hand, while the fruit of Cestrum is not very
unlike that of Solanum, the seeds are very different in shape, being
peltate, more or less obovate, with the broad end towards the
apex of the seed, so that the cotyledons have room to widen
themselves.

Sometimes we meet with species having both narrow and broad
cotyledons, even in the same genus. For instance, Coreopsis fili-
Jolia (fig. 6) has narrow, Coreopsis auriculata broad, cotyledons.

Fig. 62.

Coreopsis filifolia.

*
Fig. 60. Achene, x 14. Fig.61. Longitudinal section, x 14.
Fig. 62. Transverse section, x 14. Fig. 63. Embryo, x 14.

If, however, we examine the seeds we find that those of C. fili-
folia are narrow or subcylindrieal (fig. 61), while those of C.
auriculata (fig. 64) are broad obovate; and as in both cases the

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 367

embryo fills the seed, this difference sufficiently accounts for the
dissimilarity in the cotyledons.

aunt

Coreopsis auriculata.

Fig. 64. Achene, x 14. Fig. 65. Longitudinal section, x 14.
Fig. 66. Transverse section, x14. Fig. 67. Embryo, x 14.

The genus Galium is an interesting case. Here also we find
some species with narrow, some with broad, cotyledons ; but the
contrast seems to be due to a very different cause. G, Aparine
(fig. 114) has broad, G. saccharatum (fig. 24) narrow, cotyle-
dons. In G. saccharatum the fruit (figs. 68-70) is deeply lobed,
two-celled, iwo-seeded, indehiscent, and densely covered with
tubercles. The seed is globose, deeply hollowed on the ventral

368 SIR J. LUBBOCK—-PHYTOBIOLOGICATL OBSERVATIONS.

surface. The embryo (fig. 68, 7 c) is curved, following the concavity
of the hollow, with the larger part of the perisperm lying towards
the periphery. The cotyledons are linear and obtuse. Fig. 69
shows that, so far as the form of the seed is concerned, there is
no reason why the cotyledons should not be much broader than
they are. The explanation may perhaps be found in the struc-
ture of the pericarp (fig. 68, p.c), which is thick, tough, and corky.

Fig. 68. Fig. 71.

Galium saccharatum.

Galium Aparine.

Fig. 68. Longitudinal section of seed, x 8. Fig. 69. Transverse section of
seed, X 8. Fig. 70. Germinating seedling, X 4.

Fig. 71. Longitudinal section of seed, x 8. Fig. 72. Transverse section of
seed, x 8. Fig. 73. Germinating seedling, x 4.

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 369

It is very impervious to water, and may be advantageous to the
embryo by resisting the attacks of drought and of insects, and
perhaps if the seed be swallowed by a bird, by protecting it
from being digested. On the other hand it does not split open,
and is too tough to be torn by the embryo. The cotyledons,
therefore, if they had widened, as they might otherwise have
done, would have found it impossible to emerge from the seed.
They evade the difficulty, however, by remaining narrow. On
the other hand, in G. Aparine the pericarp is much thinner
(figs. 71-73), and the embryo is able to tear it open (fig. 73).
In this ease, therefore, the cotyledons can safely widen without
endangering their exit from the seed.

The thick corky covering of G. saccharatuin is doubtless much
more impervious to water than the comparatively thin testa of
G. Aparine. The latter species is a native of our own isles,
while G. saccharatum inhabits Algiers, the hotter parts of
France, &c. May not, then, perhaps the thick corky envelope
be adapted to enable it to withstand the heat and drought ?

In all these species the cotyledons are flat or nearly so, but a
large number are enabled to widen themselves by being more or
less folded. One form of this is afforded by the Radish (Rapha-
nus) and Brassica (fig. 8). Fig. 74 shows a seed of Raphanus
sativus, and, as shown in figs. 75-78, the latter of which repre-
sents a young seedling, the cotyledons are applied to one another
face to face, and then folded along the middle.

Unequal Cotyledons.

I now turn to those species in which the two cotyledons are
unequal in size.

Several of these cases have been discussed by Darwin*, who
attributed the inequality to the fact of *a store of nutriment
being laid up in some other part, as in the hypocotyl, or one of
the two cotyledons, or one of the secondary radicles.” I differ
with the greatest hesitation from so high an authority; but do
not see the connection between the store of food being partly
laid up in some other part of the plant and the inequality
between the cotyledons. Why should. it affect one more than
the other? I venture to suggest that the difference is rather due
to the position of the embryo in the seed, which in some cases
favours one cotyledon more than the other. For instance, in

* ‘Movements of Plants,’ p. 94,

370 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

many cases the cotyledons are what is called “ incumbent,” that
is to say the radicle is folded upon one of the cotyledons, and in
such species the outer cotyledon is often rather larger than the
other, as for instance in Hesperis matronalis.

Fig. 74. Fig. 75. Fig. 78.

Raphanus sativus.
Fig. 74. Outline of seed, x 4: m, micropyle; A, hilum. Fig. 75. Embryo
extracted from seed, x 4. Fig.76. Embryo, x 4; vertical section.
Fig. 77. Embryo, seen from the side, x 4: o.c, outer cotyledon ;
i.c, inner cotyledon; r, radicle; 7, testa. Fig. 78. Germinating
seedling, x 2, showing the cotyledons still folded.

In the Mustard (fig. 3), Cabbage, Radish (Raphanus), and some
other Crucifers the difference is more marked, and is due to à
different cause. The cotyledons, as just mentioned, are applied to
one another face to face (figs. 75—77), and then doubled longitu-
dinally one inside the other. The outer one, therefore, having
more space, becomes larger and the petiole is longer. In Cereus
(fig. 79) the embryo is much curved, and the cotyledons being
thick and fleshy, the inner one is naturally smaller than the outer.

In Abronia umbellata (fig. S0 B) the embryo is large, much
curved or doubled on itself, and lies outside the perisperm, the
two edges of the cotyledon reaching almost to the radicle. The
second cotyledon (ic) is minute, generally not more than one
seventh the length of the outer one, though in one seed it attained
nearly half the length of the other. In another species of the

SIR J. LUBBOCK —PH Y TOBIOLOGICAL OBSERVATIONS. 371

same genus, A. arenaria, the smaller cotyledon is reduced to a
mere knob (fig. 80 A, ic).

Fig. 79. Fig. 80.
; ic

B

Fig. 79. Cereus JVapoleonis. Section of seed. oc, outer coat of testa; zc, mner
coat of testa.

Fig. 80. A. Embryo of Abronia arenaria, x 6: ic, smaller cotyledon. B. Em-
bryo of A. umbellata, x 6: ic, right and inner cotyledon; le, left

cotyledon.
Cases in which one of the cotyledons is rudimentary also occur
in the genera Ranunculus*, Carumt, Cyclamen, &c.
In Petiveria octandra (fig. 81) the cotyledons are very interest-
1ng. The one is about 3 centim. long, 14 broad, oblong, tapering

Fig. 83. Fig. 84.

Petiveria octandra.
Fig. 81. Seedling, half nat. size. Fig. 82. Embryo partly unfolded, x 6.
Fig. 83. Outer and shorter cotyledon. Fig. 84. Inner and longer cotyledon.

* See Irmisch, ‘ Beit. zur vergl. Morphol. der Pflanzen’ (Halle, 1854).
t Hegelmaier, Vergl. Untersuchungen, 1875.

372 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

at both ends, and entire. The other is shorter and broader,
2 centim. long by 13 broad, subeordate, with a large terminal
lobe, and one, more or less pronounced, on each side. At first
I thought this curious want of symmetry must be accidental.
It is, however, normal, and is explaiued by the peculiar form of
and arrangement of the embryo (figs. 82-84) in the seed. The
fruit is an achene of peculiar form; it is oblong linear, sub-
cuneate, and laterally compressed, bifid at the apex, and crowned
with 2-6 unequal, closely reflexed, acute bristles, more or less
hairy, one-celled, one-seeded, and indehiscent. The seed is oblong
linear, and tapers somewhat, conforming to the interior of the
fruit. The embryo is large; the longer cotyledon is doubled on
itself; the broader one is also turned over at the end, and wraps
round the narrower one*, not reaching, however, to the narrow
end of the seed: the terminal lobe of the shorter and broader
cotyledon is the part which is folded over, and the lateral lobes,
which are much smaller in the embryo, are also due to the fold,
as shown in figs. 82-84.

Unsymmetrical Cotyledons.

In other cases, as in the Geraniums (fig. 25), Schinus (fig. 28) (the
false Pepper), Clitoria (fig. 27), Laburnum (fig. 26), Lupines, &e.,
there is inequality, not between the two cotyledons, but between
the two halves of each cotyledon. In the Geraniums this is due
to the manner in which the cotyledons are folded. In Cabbage
and Mustard we have seen that one cotyledon is folded inside the
other; in the Geranium they are convolute (fig. 85), one half of

Fig. 85.

Section through embryo of Geranium, showing the mode of folding
of the cotyledons.

each being folded inside one half of the other—the two inner
halves being the smaller, the two outer the larger ones. ;
In the Laburnum (fig. 26) and in Clitoria (fig. 27), where

* The narrow cotyledon is sometimes absent.

?
SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 373

the arrangement is very similar, the inequality in the two sides
of the cotyledon is due to the inequality between the two sides
of the seed (fig. 86).

Seed of Laburnum vulgare, X 6. OC, outer coat ; IC, inner coat.

In Heritiera macrophylla the cotyledons fill the seed, which
conforms to the shape of the carpels, and the fact that these are
somewhat unequal-sided renders the seed and consequently the
cotyledons so likewise.

In the Lupines the seeds are obliquely oblong, compressed.
laterally, and without perisperm, the embryo being large, fleshy,
yellowish, and occupying the whole seed. It is doubled on itself,
and the cotyledons are folded along the radicle, which nearly equals
them in length, with the smaller halves turned towards the radicle,
and in such a manner that they and the radicle together occupy
one half of the seed, and about equal the larger halves of the
cotyledons, which fill the other.

In Zriphasia the inequality is due, partly at any rate,to a
different cause. The seeds are oval, somewhat flattened, especially
on the ventral aspect. The embryo is large, and occupies the
whole seed. The cotyledons are very unequal in size, and the
smaller one is more or less enclosed in the larger. But, in addition
to this, there are often, indeed generally, two and sometimes
three embryos in each seed ; these differ in size, and the smaller
ones often intrude more or less on one of the cotyledons belong-
ing to the larger ones.

Crenate Cotyledons.

The vast majority of plants have the edges of the cotyledons
entire. There are some few, however, in which they are more or
less crenate, as for instance in Cordia subcordata (fig. 33).

In this species the embryo occupies the whole of the ovoid-
conical seed, There is no perisperm, and the cotyledons, in order

LINN. JOURN.— BOTANY, VOL. XXII. 2H

974 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

to occupy the whole space, are longitudinally folded (fig. 87),
thus giving rise'to the crenations on the margin.

Fig. 87.

VE i

Embryo of Cordia, subcordata, X 2.

Accumbent and Incumbent Cotyledons.

There are two ways in which the radicle may be bent over the
cotyledon.

Sometimes it is turned up over the back of one of the coty-
ledons, as in figs. 88 and 89, and is said to be incumbent; while
sometimes it is turned along the edge of the cotyledons, as in
figs. 90 and 91, which are then termed accumbent. The divisions

Figs. 88 and 89. Sections of seed of He: peris matronalis, x 10.
v, hollow in testa.

Figs. 90 and 91. Section of seed of Cheiranthus Cheiri, X 10.

of the Cruciferz are based to some extent on this character, some
groups being accumbent and some incumbent. I puzzled for
some time over the reasons which could account for this differ-

SIR J. LUBDOCK—PHYTOBIOLOGICAL OBSERVATIONS. 375
ence, though the explanation which I would suggest is very simple
when once stated. I have not, however, been as yet able to
examine a sufficient number of cases to justify me in doing
more than throwing it out as a suggestion.

Seen in section the two forms would be as shown in figs. 88 and
90. Now if from the form of the pods, or for any other reason, it
is an advantage that the seed should be compressed, as in fig. 90,
then, the thickness of the cotyledons remaining the same, it is
better that the radicle should be aeeumbent; while, on the other
hand, in a thicker or globular seed, as in fig. 88, the incumbent
arrangement is most convenient. In fact we find that in groups,
such as the Arabideæ, where the seeds are as a rule compressed,
the radicle is almost always accumbent; while in incumbent
groups, such as the Sisymbreæ, they are, on the contrary, more or
less turgid. As an actual e xample of an incumbent form I give
Hesperis matronalis (figs. 88 & 89), and of an accumbent, Chei-

ranthus Cheiri (figs. 90 & 91).

Petioles.

The cotyledons are sometimes sessile, as in Acer (fig. 9), Hip-
pophaé (fig. 14), Hakea (fig. 21), Clitorea (fig. 27), sometimes
supported on petioles, which in many cases, as in Microloma
(fig. 29), attain a considerable length. :

Occasionally we meet with both sessile and petioled cotyledons
even within the limits of the same genus. For instance, Delphi-
fig. 31) bas the cotyledons sessile, while

nium Staphisagria (
In Bryonia laciniosa

those of D. elatum (fig. 30) have a p »tiole.
(fig. 92) the cotyledons are nearly sessile, while those of B. dioica
(fig. 93) have petioles.

There is nothing, so far as I kn
seed to account for this difference. It is observable, however,

that while the cotyledons of Bryonia laciniosa (fig. 92) and those
of D. Staphisagria (fig. 31) are raised by the caulicle somewhat
above the level of the ground, those of Bryonia dioica (fig. 93)
and of D. elatum (fig. 30) are attached close to the ground. In
fact the cotyledons are carried up in both cases, but in B. laci-
niosa and D. Staphisagria by the caulicle or stalk of the plant,
in B: dioica and in D. elatum, on the other hand, by their own

petioles.
In short we may say
sessile when they are raised by t

ow,in the strueture of the

that the cotyledons are, as a general rule,
he growth of the caulicle, while
2H 2

376 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

they are petiolate when they take their origin close to the
ground. There are no doubt some exceptions; for instance,

Fig. 92,

Fig. 92. Seedling of Bryonia laciniosa. Half nat. size.
Fig. 93. Seedling of B, dioica. Nat. size.

in some species of Hedysarum the cotyledons are radical and
yet sessile. I have, however, often in Algeria seen seedlings of
this group in hot exposed situations, where they * held the field"
alone, and being sure of ample heat and light, did not require to
be raised above the surface.

The opposite exception is perhaps more common. e. when
the cotyledons, though raised, are still petiolate. Here, however,
the cotyledons are probably petiolate for the same reason as
the leaves—viz., when the foliage is large, leaf-stalks are an
advantage in carrying the lower leaves out of the shadow of
those immediately above them.

In another species of Delphinium (D, nudicaule, fig. 32) the
cotyledons are raised well above the surface of the ground on astem
consisting of their own two stalks or petioles, which are connate,
though readily separable from one another. Connate cotyledons

SIR J. LUBBOCK—PHYTOBIOLOGICAT, OBSERVATIONS. 874

also occur in Phlomis tuberosa, Smyrnium perfoliatum, Polygonum
Bistorta *, &c. Gray t observes that the economy of this arrange-
ment is not apparent. Assuming, however, that the elevation of
the cotyledons is an advantage, perhaps, as I have suggested,
from carrying them above the surrounding herbage, the combi-
nation of the two petioles—reversing the old fable of the buudle
of sticks—would, with the same amount of material, give a cen-
siderable addition of strength.

In other cases, however, the existence of petioles apparently
has reference to the arrangement of the embryo in the seed.

In the Geraniums, for instance, as has been already mentioned,
the cotyledons are folded on themselves, one half of each lying
within the other. Fig. 94 represents an embryo partialy un-
folded, and it will be seen that in the position assumed by the
cotyledons the petioles are necessarily as long as half their breadth.
In Eucalyptus Globulus, again (fig. 112), the mode of folding of
the cotyledons would be impossible but for the petioles.

Fig. 94,

Germinating seedling of Geranium bohemicum, X 6.

* Winkler, * Ueber die Keimblütter der deutschen Dicotylen," Brandenburg
Bot. Ver. 1874. 1 ‘Structural Botany,’ p. 21.

378 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

Lastly, in cases where the cotyledons do not leave the seed,
the petioles leave room for the free growth of the plumule, as in
Sapindus (fig. 43).

Lobed Cotyledons.

The great majority of cotyledons are entire, but some are more
orless lobed. For instance, those of the Mallow (fig. 35) are
broadly ovate, minutely emarginate, cordate at the base, and 3-
lobed or angled towards the apex, with three veins, each running
into one of the lobes.

Those of Lavatera and Althea are similar. The embryo is
green, curved, and occupies a great part of the seed. The cotyle-
dons are applied face to face, then, as growth continues, the
tip becomes eurved, and depressed into a median longitudiual
furrow, the fold of the one lying in that of the other.

The embryo is a small body of the form shown in fig. 95 : the
horn or process 7 is the radicle; the rest is the cotyledon, of
which the free end f is folded on itself and turned downwards.
In this way the embryo occupies the seed, leaving a small space
between the cotyledons and also between f and » which is occupied
by perisperm. Perhaps it may make the arrangement clearer
to take a piece of note-paper, cut it into the form of an egg

Fig. 95. Fig. 96. Fig. 97.

Fig. 95. Embryo of Mallow. Enlarged.
Figs. 96 & 97. Piece of paper prepared to show the mode and effect of the
folding of the embryo.

(fig. 96), turn the broad portion ab downwards, so that the
parts a and b have their under faces turned to one another, and
then press down the line ef, and bring the points c and d
together, so that c ef and d e f have their upper surfaces together,
and the apex f pointing downwards. We shall then have an
object shaped as in fig. 97, with a sharp point at c, which would
not conform to the rounded shape of the seed. If, now, to
make it do so, we cut off the point c along the dotted line
and then unfold the paper, we shall find that it has the form of
the cotyledon of Malva (fig. 35) with a bay or notch on each side.
In Erodium the arrangement is somewhat similar, and it seems

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 879

clear that the lobes are due to the manner in which the embryo
is folded.

In Gianothera and some allied species the cotyledons also
present a terminal lobe. This, however, is not due to folding.
The terminal lobe is the original cotyledon, and the basal portion
is altogether subsequent growth, whieh moreover, to some extent,
assumes the character of the true leaf. I shall refer to this in-
teresting group in more detail on a subsequent occasion.

The ease of Petiveria octandra (fig. 81) has been already

described. à
Emarginate Cotyledons.

In a great many species the cotyledons are emarginate, or even
more or less deeply bifid. No explanation of this has, so far as I
know, yet been offered. It is, in fact, as I shall hope to show, by
no means always due to the same cause,

One of the simplest cases is that of the Oak, where the thick
fleshy embryo occupies the whole of the seed. The chalaza is
situated at the centre of one end, at the extremity of the cotyle-
dons, and the walls of the seed being at that point somewhat
thickened, the cotyledons are slightly pressed in. The same ex-
planation applies to various other species, as for instance to the
Impatiens (figs. 12 and 98), Poterium (figs. 42 and 99), Cuphea
(fig. 100), and Nettle (Urtica, fig. 101).

" Fig. 98.

ch

| aN

Fig. 98. Longitudinal and transverse sections of seed of Impatiens parviflora,
x10. Pl, Plumule.

Fig. 99. Longitudinal section of Poterium Sanguisorba, X 9. ct, calyx-tube ;
a, achene ; re, receptacle.

3S0 SIR J. LUBBOCK-—PHYTOBIOLOGICAL OBSERVATIONS.

In Helianthus Cucumis the seed itself is slightly notched at the
point where it artieulates with the receptacle; and the cotyle-
dons, which, with the rest of the embryo, eventually oceupy the
whole interior of the seed, conform to this notch.

In such cases as the Mustard (fig. 3), Cabbage, and Radish, the
emargination is due to a totally different cause. The seed (fig. 74)
is oblong, thick, and slightly narrower at one end than the other.
There is no perisperm, so that the embryo oceupies the whole

Fig. 100. Longitudinal and transverse sections of Cuphea silenoides, X 10.
oc, outer coat of testa ; ic, inner coat of testa.

Fig. 10i. Achene of Nettle ( Urtica dioica), x 30. J, testa.
seed and as this is somewhat deep, the cotyledons, in order to
occupy the whole space, are folded and arranged one over the
other, like two sheets of note-paper, as shown in figs. 75-78, the
radicle being folded along the edge. Fig. 77 represents the
seed a little opened out; and fig. 76 a section showing the
radicle and the outer and inner cotyledons. To this folding the
emargination is due. If a piece of paper be taken, folded on
itself, cut into the form shown in fig. 74, with the fold along the
edge from m to h, and then unfolded, the reason for the form of
the cotyledon becomes clear at once. Zilla myagroides affords a
similar case. :

But it may be said that in the Wallflower the seed has a simi'ar
outline, and yet (fig. 102) the cotyledons are not emarginate.
The reason of this is that in the Wallflower, Cheiranthus (figs.
90, 91), the seed is more compressed than in the Mustard and
Radish (figs. 75, 76), and consequently the cotyledons are nof
folded; so that the whole, not the half, of each cotyledon
corresponds to the form of the seed.

SIR J. LUBBOCK —PHYTOBIOLOGICAL OBSERVATIONS. 381

In the Bignoniacesz, again, a large number of species have
emarginate cotyledons ; and this would appear also, as for instance
in Pithecoctenium Aubletii, to be due to the chalaza, though in a
different manner. The seeds themselves are transversely oblong,
much compressed dorsally, surrounded on all sides except the
base by an extremely thin, transparent, membranous wing, which
is traversed by nerves radiating from the central part of the seed,

- Fig. 102,

Seedling of Cheiranthus Cheiri. Two thirds nat. size.

and is uneven at the margin. The raphe is ventral, extending
from the hilum to the centre of the embryo. The chalaza is
attached to the embryo-sac a little above the middle of the embryo.
The radicle is very small, distant from, but pointing to, the hilum.
The embryo is straight and flat; the cotyledons grow until
they come to the point of attachment of the chalaza, when they
extend forwards on each side, forming two lobes.

. In Oroxylum indicum the general structure of the seed is very
similar, but the growth of the two lobes of the cotyledons is even
more Juxuriant, so that they actually overlap. A structure more
or less similar occurs in other genera of this family.

The emargination is very much deeper in other groups, and
due to other conditions, for instance in the Convolvulacee. In
Convolvulus Soldanella (fig. 103) the embryo, which is eventually
very large, lies at first straight in the seed embedded in a clear
jelly-like perisperm, and rests on a solid, ovate, grooved, white
ridge (figs. 104 and 105, a), which rises from close to the micropyle.
This tongue-like ridge grows with the embryo. At the opposite
end of the seed the raphe and chalaza form a somewhat pro-

|

382 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

minent ridge (5) projecting into the perisperm. The cotyledons
in this stage are plano-convex, applied face to face, orbicular,
entire, green, with distinct petioles, 5-nerved, with two lateral,
subopposite branches from the midrib some distance below
the apex. The plumule and radicle are small. The cotyledons

Fig. 103. Fig. 104. Fig. 105.

Convolvulus Soldanella. «~

Fig. 103. Embryo, x 2 Fig. 104. Section of seed after removal of the dorsal
surface, embryo, and. perisperm, x2. Fig. 105. Side view of ditto, X 2.

gradually increase in size and grow over the process in which
the radicle (a) lies, extending to the apex of the seed, doubling
over and abutting against the ridge formed by the raphe and
chalaza (b), and thus becoming more and more emarginate at the
apex. The notch is therefore due to their continuing to grow at
the sides after their apex has reached this ridge.

In Zpomea purpurea (figs. 106-108), where the seed is con-

Fig. 106. Fig. 107. Fig. 108.

Ipomea purpurea.

Fig. 106. Embryo, x 2. Fig. 107. Section of seed after removal of dorsal
surface, embryo, and perisperm, x 2. Fig. 108. Side view of ditto, X 2.

structed generally on the same model as in Convolvulus Soldanella,
the ridge formed by the raphe and chalaza (figs. 107 & 108, 2), is
more prominent, and consequently the notch of the cotyledon is
deeper. Lastly, in Zgom«a dasysperma (figs. 109-111) the pro-
jecting ridge of the chalaza is still more developed and grows
nearly to the process which supports the radicle; the cotyledons

,

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 883

are thus precluded from growing in length, and in consequence
send out two long wings, so that they are divided almost to the
base (fig. 111). :

In Shorea, again, the division of the cotyledons appears to be
due to an internal process of the seed. I have not, however, had
an opportunity of examining a specimen,

Fig. 109. Fig. 110. Fig. 111.

Ipomea dasysperina.

Fig. 109. Embryo, x 2. Fig. 110. Section of seed after removal of dorsal
surface, embryo, and perisperm, x 2. Fig. 111. Side view of ditto, x 2.

In Zucalyptus (fig. 38) we have a different case. The embryo
is (with the exception of the petioles) straight or nearly so,
fleshy, white, oecupying the whole of the seed, and conforming
to it in general outline; the cotyledons are deflexed and convolute
round the radicle, which the lobes equal in length, while half of

Fig. 112.

Embryo of Eucalyptus Globulus, one cotyledon being cut away. p, petiole ;
p', cut’end of petiole; i.l, inner lobe of cotyledon. x4.

one cotyledon lies over half the other; one half of each conse-
quently lying against the testa. The radicle is stout, fleshy,
truncate at the end where it lies against the testa, but otherwise
entirely enclosed by the folded cotyledons. The true length of
the cotyledon is determined by the distance between the end
of the petiole and the opposite pole of the seed. The side of the
cotyledons, however, being folded back, that part which lies beyond
the petioles is enabled by folding round the radicle to widen, and
this consequently gives the cotyledons their more or less pro-
nounced hourglass shape.

384 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

Moreover, in speaking of emarginate cotyledons, we must dis-
tinguish between two very opposite cases, of which I may take
Galium Aparine and GZnothera Lindleyana as illustrations. In
the former the cotyledons commence with an entire end (fig. 113),
and subsequently, not as a rule till they have left the seed, become

Fig. 113.

Fig. 113. Young seedling of Galium Aparine, X 2.
Fig. 114. Ditto, a few days older, 269,

emarginate (fig. 114); in @nothera Lindleyana, on the contrary,
the cotyledons are at first emarginate, but gradually cease to
be so. The embryo gradually appropriates all the perisperm,
but the supply being largest at the wider end of the seed, this is
the last part to be absorbed. In neither of these cases does the
emargination appear to be directly due to the structure of the
seed, nor to be in itself of any advantage to the plant. It seems
rather to depend on the conditions of growth. In G. 4parine
the cotyledon terminates in a peculiar gland *, which would

* This gland has already been mentioned by Gravis in his work on Urtica
dioica, p. 139. He observes that about a dozen small stomata occur on it,
while they are entirely absent on the rest of the upper surface. He regards
it as a water-gland, “un organ destiné à remédier à l'excès de tension dans
l'appareil aquifére.”

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 385

appear in this and other cases, after the emergence of the cotyle-
don, to stop its growth at that point, and thus to produce the
emargination.

In Senecio, again, the majority of species have entire cotyle-
dons. ' In some, however, as in S. erucefolius, they are emarginate.
Even here, however, they are at first entire (fig. 115), and the
emargination does not make its appearance until after germi-
nation, when the cotyledons gradually become much widened
(fig. 116). In fact, Senecio squalidus, S. viscosus, S. vulgaris,

Fig. 115.

Fig. 115. Young seedling of Senecio erucefolius.
Fig. 116. Ditto, a few days older.

&e. have the cotyledons narrow and entire; while in S. eruce-
folius and S. Cruentus, where they grow more in width than
in length, they become emarginate. Among other cases where
the cotyledons are at first entire, but after germination become
emarginate, may be mentioned some species of Lithospermum.

Bryonia laciniosa (fig. 92) also has the cotyledons emarginate,
while in Bryonia dioica (fig. 93) they are entire. They are, how-
ever, originally entire in both cases, and the emargination in
B. laciniosa is due to the fact that in that species the cotyledons
grow much more than in B. dioica. There is no great difference
in size between the seeds, those of B. laciniosa being perhaps
one tenth larger. On the other hand, the cotyledons of B.
laciniosa attain a length three times greater than those of B.
dioica, as shown in the figures (figs. 92 & 93). In the genus
Lacsonia, again, the cotyledons are entire in T. Van- Volxemi
and T. Leschenaultii, and emarginate in 7' ignea. Here also,
however, they are at first entire, and only become emarginate
after leaving the seed.

Divided Cotyledons.
The genus Pterocarya has very curious cotyledons (fig. 41),
due to a cause entirely different from any of those we have
considered hitherto.

386 SIR J. LUBBOCK— PI YTOBIOLOGICAL OBSERVATIONS.

They are bipartite, each primary division narrowing to a
cuneate base, and being again deeply divided, so as to make in
all four ultimate, linear, oblong, obtuse, entire segments. In
this case the endocarp is thickened, bony, and divided at the base
into four cells (fig. 117,0 c cc) by the thickening and consequent
intrusion of the dorsal and ventral walls.

The seed (fig. 118) is conieal above, deeply 4-lobed below
(fig. 118, 7777), one lobe (7) passing into each of these cells (c).
The embryo again follows suit, and each cotyledon sends a lobe
into each of the four cells, thus assuming the peculiar form
characteristic of the species.

Fig. 117. Fig. 118.

Pterocarya caucasica.

Fig. 117. Transverse section of nut, x 6.
Fig. 118. Transverse section of seed, x 6.

In Eschscholtzia (fig. 40) the cotyledons are deeply bifid, re-
sembling a hay-fork with two long prongs. In this case we find
no such peculiarity of the fruit or seed to account for the pecu-
liarity. My first idea was that such cases might possibly be due
to some difference in the perisperm, as occurs in certain Umbellifers,
Delphinium, &c., and which might have permitted growth more
readily in certain directions than in others. Thin sections, how-
ever, showed no such differences. Moreover, Schizopetalon
Walkeri (one of the Cruciferz) (figs. 120-123) has the cotyledons
as deeply divided as in Eschscholtzia ; and as there is no perisperm,
but the embryo occupies the whole seed, within which the long
lobes of the cotyledons wind about more or less irregularly, the
division cannot be due to differences in the perisperm. ‘There are,
moreover, other species, such as the Sycamore (fig. 52) and Hop,
where the cotyledons are also narrow, winding, and occupy the

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 387

whole seed, but are not divided. We must therefore seek some
other explanation, and I will suggest the following.

In most of the species which I have examined, when the coty-
ledons remain in the seed they do not leave the ground. In
some cases, however, as in Anona (fig. 119), the caulicle is long,

Fig. 119.

stout, and rises in the form of a loop
during germination, while the cotyle-
dons, which ‘at first are very small,
gradually increase almost to the length
and breadth of the seed, throw them-
selves into undulations, and it being
perhaps on this account impossible to
withdraw themselves from the seed,
are eventually torn from the axis. In
Bignonia insignis, again, the cotyle-
dons, though flat and leaflike, are un-
able to emerge from, or at any rate do
not emerge from, the seed. This may
possibly give us a clue to such cases as
Eschscholtzia and Schizopetalon, which,
Iwould venture to suggest, may have
reference to the manner in which the
cotyledons free themselves from the
seed. If this is delayed, the young
plant suffers eonsiderably, and indeed
often perishes. That the process is
not, however, so simple as might be
imagined, may be seen from the inter-
esting case afforded by the Cucurbi-
tacez, where,in Mr. Darwin’s words*,
"the seed-coats are ruptured by a
curious contrivance, described by M.
Flahaut. A keel or peg is developed
on one side of the summit of the radicle
or base of the hypocotyl, and this holds
down the lower half of the seed-coats

Germinating seedling of
Anona.

(the radicle being fixed into the ground), whilst the continued
growth of the arched hypocotyl forces upwards the upper half,

* * Movements of Plants, p. 102. In Welwitschia Bower has pointed out that
a corresponding process serves to absorb the perisperm, acting in fact as a
feeder to the young plant (Quart. Journ. Mier. Sci. vol. xxi.).

388 SIR J. LUBBOCK—PHYTOBIOLOGICAL ORSERVATIONS

and tears asunder the seed-coats at one end, and the cotyledons
are thus easily withdrawn."

May not the narrowness of the cotyledons in Eschscholtzia and
their deep fission be due to a similar cause ? The seed is slightly
pyriform, and the radicle emerges from the narrower end. It
bursts through the soil in an arch, and instead of leaving the seed
behind as usual, the cotyledons carry it up with them. Then the
two arms of the cotyledons separate, thus widening the orifice
and drawing themselves out.

This suggestion seems to be confirmed by the evidence of
Schizopetalon (figs. 120-123), one of the other few cases where the

Fig. 123.
Fig. 120.

Figs. 120-123. Stages in the growth of seedling of
Schizopetalon Walkeri, x 24.
cotyledons are bifid. Here, also, the radicle emerges through a
comparatively small orifice, and the seed, from which the coty-
ledons seem to have some difficulty in freeing themselves, is carried
up by the embryo, while eventually the lobes of the cotyledons
draw ihemselves out one by one. In Opuntia basilaris, again,
which differs from O. Labourtiana in having narrow cotyledons,
the seed is similarly carried up, and the cotyledons free them-

SIR J. LUBBOCK— PHYTOBIOLOGICAL OBSERVATIONS. 889

selves by divergence. In this species, also, it is interesting that
one or both cotyledons are often bifid. Is it possible that the
multiplicity of the cotyledons in Conifers (fig. 124) can be due
to the same cause ? à !

In Ephedra there is a special membrane, which remains
attached to the root, and thus prevents the testa from being
carried up on the tips of the cotyledons.

The common Cress (Lepidium sativum) (fig. 4), to which I
have already referred, is a very interesting case, for while in the
other species, at any rate in the other English species, of the
genus the cotyledons are entire, in Lepidium sativum, on the
contrary, each possesses two long, narrow lateral lobes.

Fig. 125 represents a section through the seed of L. gramini-
Jolium, which may be taken as representing the ordinary arrange-
ment in the genus. The seeds, conforming to the shape of the

Fig. 126.

ne 125. ters EEE feme d Rae ien. x5;

Fig. 126. Section of seed of = sativum, X 15.
capsule, are somewhat triangular, with the radicle in the narrow
end. The embryo occupies the whole of the seed, there being no
perisperm. In Z. sativum (fig. 126) the seed is of the same form,
but nearly twice as large, and if, therefore, the cotyledons were to
occupy the whole additional space, they would become extremely
thick. In perispermie seeds this would present no difficulty, as the
additional space would be simply filled by perisperm. In Lepi-
dium, however, this device cannot be resorted to; the two lobes,
however, just fill up the vacancy.

LINN. JOURN.—BOTANY, VOL. XXII. 21

390 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

Auricled Cotyledons.

Some cotyledons are markedly auricled. As illustrations I
give Poterium (fig. 42) and Hakea (fig. 21). This form is, I am
disposed to sugzest, a provision to fill up vacant space in the seed.
In the seed of Hippophaë (fig. 55) the form of the cotyledon
leaves at each side of the base of the cotyledon two spaces (d),
which are occupied by perisperm. In Cuphea (fig. 100), Ruellia
(fig. 57), and Poterium (fig. 99), on the other hand, theré is no
perisperm, and it is consequently an advantage that the cotyledon
should develop auricles in order to fill up the space.

If this is the explanation of the auricles, we should expect to
find them developed principally in families where the perisperm
is deficient. Now in the species I have examined, auricled coty-
ledons occur in 35 genera, belonging to 22 families, of whieh 13
bave no perisperm, while in 6 of the 9 others it is reduced almost
to a film.

The argument in the case of Cuphea is further strengthened by
the peculiar conformation of the radicle (fig. 100), which is
three-lobed, the reason being, I would suggest, that the radicle

cooperates with the cotyledons in the endeavour to fill up the
vacant spaces.

Fig. 127.

Three stages in the growth of the seedling of Cuphea silenoides.

In support of this view I also would observe that the auricles
seem to be of little use to the young plant. For instance, the
embryo of Cuphea, while in the seed (fig. 100), has very large
auricles, which in the seedling (figs. 127—129) soon disappear. In
Ruellia, again (conf. figs. 57 and 16), we have a similar case.

Size of Seeds.
As regards the size of seeds, if we could imagine a state of

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 391

things in which every seed grew and attained maturity, then to
keep up the number of any given species existing at any time, it
would be sufficient if each plant produced but one or two seeds
during its whole life. There is, however, an enormous destruction
of seeds. The great majority are eaten by animals, or fail to secure
a suitable site for germination; of those which do germinate, again,
many are crowded out by their fellows; and Darwin observed that
out of 357 seedlings which came up in a space of 3 feet by 2,
no less than 295 were destroyed by slugs and insects. The greater
the chance against any given seed reaching a suitable locality and
attaining maturity, the larger number of seeds must the plant
produce in order to maintain its numbers, and, as a general rule,
the smaller will the individual seeds be. On the contrary, the
greater the chance that each seed enjoys of arriving at maturity,
the smaller the number of seeds that is necessary, and in such
cases it is an advantage that the seeds should be large.

Hence parasitic plants generally produce a large number of
very small seeds, though there are exceptions due to other consi-
derations, as, for instance, in the Mistletoe (I believe, indeed, all
the Loranthaces), where the seeds are carried by birds.

An interesting illustration is afforded by certain species which
produee two kinds of pods, as, for instance, Cardamine cheno-
podifolia of Brazil. Besides ordinary pods, which resemble those
of any other Cardamine, and contain several seeds, this plant pro-
duces a second sort of pod underground. Now in the ordinary
pods the number of seeds increases, of course, the chance that some
one will find a suitable place. On the other hand, the subterranean
pods are sown, as it were, by the plant itself. In this case, if
there were a number of seeds they would only get in one another's
way, and hence, perhaps, the fact that the subterranean pods
only produce one or two seeds. In most species the seeds vary
somewhat in size; but in such cases it would not be conclusive
to contrast the produce of large seeds with that of smaller ones,
because it might fairly be said that the former were better
nourished, and inherently, perhaps, more vigorous. In Carda-
mine chenopodifolia of Brazil, however, the seeds from the under-
ground pods are larger than the others, and Grisebach found that
they produced more vigorous seedlings *.

There are, on the contrary, other considerations which may

* “Der Dimorphismus der Fortpfl. v. Cardamine chenopodifolia,' Göttinger

Nachrichtungen, 1878, p. 334.
212

392 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

make it an advantage that the number of seeds produced by a
flower should be reduced, as, for instance, in the case of the
Composite, where the agglomeration of a number of flowers into
a single head, as in the Daisy, and their consequent diminution
in size, renders it an advantage that each floret should produce
but one seed.

The species with large seeds may, as already mentioned, be
divided into two groups—first, those in which the embryo is
surrounded by perisperm, and, secondly, those in which it occupies
the whole seed.

In the former the arrangement of the embryo presents no spe-
cial difficulties, as the perisperm simply fills up all vacaut spaces.
In the latter, on the contrary, Nature has to exercise much
ingenuity, and adopts various devices to fill up the whole space.

One plan is to arrange the cotyledons face to face, and then
roll them up in a ball. This is adopted, amongst other cases, in
the Sycamore (fig. 52), and hence the long strap-like form of the
cotyledons.

Another is to arrange the cotyledons face to face, and then double
them up, as in the Cabbage, Mustard, Radish (figs. 75-77), &c.

In a third the cotyledons are convolute edgeways, as in Caly-
canthus.

In Lepidium sativum the cotyledons are trifid (figs. 4 and 126) ;
in Cordia they are thrown into plaits(fig. 87). In others we have
still more complex folds, as in the Beech.

In such cases as the Lupine the cotyledons become so fleshy and
thiekened, that they almost lose the appearance of leaves; in
this instance they are set free by the splitting of the testa.
When, however, the testa does not readily split, and where in
large seeds there is no perisperm, the difficulty of unfolding the
cotyledons and extricating them from the seed becomes greater,
and we arrive at cases where Nature seems to have abandoned
the attempt, and, as in the Oak and Horse Chestnut, the coty-
ledons never quit the seed. Thus, ' among the Juglandes,
Pterocarya has leaf-like cotyledons, while those of the Walnut
never quit the shell. Every one, however, must have observed
the elaborate folds into which the two cotyledons are thrown—
folds which seem to have no significance or importance now, and

which carry us back to a time when the Walnut, like the Ptero-
carya, had foliaceous cotyledons.

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 393

If these suggestions be correct, we should expect that species
with non-emerging cotyledons would generally have large seeds
and be aperispermie. This certainly appears to be the rule;
among the species with reference to which I have notes, there
are 37 genera in which the cotyledons are subterranean or remain
in the seed. The seeds themselves are notably large, and all but
three are aperispermic. Occasionally we meet with subterranean
and foliaceous cotyledons in the same genus, as, for instance,
in Rhus, Rhamnus, Mercurialis *, Phaseolus, &c.

Phaseolus vulgaris presents us with an intermediate stage, the
cotyledons being aerial and green, but fleshy and by no means
true leaves. In Melittis melissophyllum, again, according to
Irmisch f, the fleshy éotyledons generally remain in the seed, and
are held together by the testa; but they sometimes burst the
shell, and stand out from one another. Like true subterranean
cotyledons, they have no stomata.

When the cotyledons are large, thick, and fleshy they often
contain sufficient nourishment to render the plants for some time
independent of any fresh supply. In such cases the seediings
sometimes push up for awhile without any fully-developed leaves,
the first few being reduced to very small size or almost obsolete,
as in Rhus (fig. 44), Sapindus (fig. 43), l'ropeolum, &c.

LJ

Size of Embryo.

As already mentioned, there are many cases, in fact many
whole orders, in which the ripe seed is entirely occupied by the
embryo; in other cases, again, as in Delphinium (fig. 130), the
embryo is very small, and examples of every intermediate stage
might be given.

Where it is an advantage to the plant that germination should
be rapid, this of course can be more readily secured if the embryo
is large. In fact, we find that species with large embryos, such,
for instance as Cabbage, Pea, &c., germinate much more rapidly
than those, such as Umbellifers, RHanunculacez, &e., in which the
embryos are small. ;

On the other hand, in other cases, time is less important, and

* Winkler, * Flora," 1880, p. 339.
t “Zur Naturgeschichte von Melittis melissophyllum,” Bot. Zeit. 1858, p. 233.

994 SIR J. LUBBOCK— PHYTOBIOLOGICAL OBSERVATIONS.

here other considerations come into play. The protection of the
embryo is mainly effected by the outer coverings; but the peri-

Fig. 130.

Long and trausverse sections of seed of Delphinium Staphisagria,
X 12: m

t

sperm itself contributes also *, and hence a small embryo is less
liable to injury.

Size of Cotyledons.

It is hardly necessary to say that the size of the plant does
not determine that of the cotyledons. Winkler has pointed
out that the largest of our Nettles has the smallest cotyledons T.
It is, on the other hand, natural that large seeds, as a general rule,
produce large cotyledons. This is, however, by no means a com-
plete explanation. There are many cases in which the cotyledons
grow considerably after quitting the seed. In the wonderful
genus Welwitschia it was, at one time, supposed that the two
great leaves were persistent cotyledons. This view is now aban-
doned. In many of the Monocotyledons, however, the coty-
ledons acquire a considerable length. I have already had occasion
to allude to cases among the Dicotyledons in which the cotyledon
continues to grow for some time after quitting the seed.

* See Marloth, “Ueber Mech, Schutzmittel der Samen gegen schädliche
Einflüsse von Aussen,” Engler's Bot. Jahrb. 1883.

T “Ueber die Keimblátter der deutschen Dicotylen,” Verh. bot. Ver. Bran-
denburg, 1874, p. 11.

SIR J. LUBBOCK —PHYTOBIOLOGICAL OBSERVATIONS. 395
Streptocarpus Rexii affords, perhaps, the most remarkable case.
The cotyledons are at first small, rotund, and shortly petiolate.
Soon, however, one of them begins to grow, and becomes an
at first ovate, then oblong, and finally pandariform, obtuse,
entire, persistent leaf, attaiuing a length of more than 18 inches.
The other cotyledon retains the original form, and soon drops off.
In some specimens, however, both cotyledons retain their original
form, and it is the first true leaf which develops.

In the cultivated form of the Mango, the cotyledons are di-
vided into more or less irregular lobes ; and in a specimen kindly
given me by Mr. Ridley, one of the lobes developed into an inde-
pendent plant.

In Qnothera and some allied species, as already mentioned,
the. full-grown cotyledons have a terminal lobe, which is, in
reality, the original cotyledon, while the larger, basal part is due
to subsequent growth. I hope to describe these more fully in a
future memoir.

Concluding Remarks.

The conditions under which the seedlings are grown naturally
exercise some influence on the form of the leaves. For instance,
in Mimulus luteus, if the seedlings have sufficient room the first
leaves are shortly stalked and deltoid, while the primary nodes
are but little developed; on the contrary, if they are more crowded,
the internodes and petioles are longer and the leaves are oval.

In the Primrose we find an arrangement which almost seems
as if it were intended to give the seedling some power of loco-
motion. The caulicle is sometimes horizontal, and throws out
strong adventitious roots, the upper part, however, becoming
vertical as usual.

As a general rule, the first buds produced by the seedlings are
in the axil of the leaves, or more rarely of the cotyledons. In
some species of Linaria, however, the caulicle itself throws out
one or more buds which develop into branches. The advantage
of this may be that, if the main shoot is cropped or broken even
down to the root, the plant is capable of throwing up another
stalk.

We thus find an almost inexhaustible series of beautiful adap-
tations to purpose. On the other hand, there are not wanting cases
in which it would seem that the adaptation is not complete, or that

396 SIR J. LUBBOCK— PHYTOBIOLOGICAL OBSERVATIONS.

a certain change, which has been of superior importance, has
involved a minor evil.

The Oak, Beech, Nut, &e. afford us a very interesting series of
species. Fig: 132 represents a diagram of a nut with the parts
somewhat separated from one another, so as to show the relations
more clearly. The micropyle (m) is at the apex of the seed. The
ovule, however, is not straight and orthotropous, which would be,
or at any rate seem to be, the simplest arrangement. Quite the
contrary, we find a long placental axis (pl) which extends to the
apex of the nut, and from which starts a raphe (7), which returns
about halfway back again to the place where the true attachment
or chalaza (ch) is situated. I have in vain endeavoured to discover
or imagine any circumstances which would render this complex
arrangement specially adapted to present conditions. It would
seem as if it would be simpler and give Nature less trouble if
the ovule sat directly with its base on the stalk, thus doing
away with both the placental axis (pl) and the raphe (7).

This view is strengthened by the fact that such an arrange-
ment has been in fact nearly attained by the Oak. The ovule in
that genus is theoretically anatropous, but the placental axis and
the raphe are both greatly shortened (fig. 131), so that the
distance which the nourishment has to traverse is much less,
though the actual place of attachment remains the same. The
Oak in fact seems to have appreciated the difficulties of the
situation, and to have in great measure neutralized them. Is it
fanciful to imagine that some ages hence the Oak may be prac-
tically orthotropous ?

But why should these species be anatropous if it is an
advantage to be orthotropous? On this question some light is
thrown by the fact that while one seed only comes to maturity,
the ovary contains originally several cells each with one or two
ovules, though none of the others come to anything. They can,
however, easily be seen, either at the apex of the seed, as in the
Nut (fig. 132, o), Beech (Fagus), &c., or, as in the Oak (fig. 131, 9),
near the base. Their presence appears to indicate that these
species are descended from ancestors the fruit of which was com-
posed of several cells, each with more than one seed—a state of
things therefore very unlike the present, and when the anatropous
condition would be an advantage. If this view be correct, the
structure of the fruit in the Nut, Beech, &c. becomes peculiarly

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 397

interesting, because it represents a case in which the present
arrangements are not those in all respects most convenient to the
plant, and render it probable that the same explanation may
apply to other cases of difficulty.

Fig. 131. Seed of Oak (Quercus pedunculata). Nat. size. o, abortive ovule.
Fig. 132. Seed of Nut (Corylus avellana). Nat. size. o, abortive ovule.

We meet indeed with a great many cases in which a larger
or smaller number of the ovules, often all but one, fail to
become developed. In some of them we may perhaps see a pro-
vision to increase the chances of fertilization. There are others,
however, in which this explanation will not hold good. In
Ptelea, for instance, the ovary consists of 2-3 cells, each con-
taining 2 ovules, one inserted rather above the other, and the
inferior ovule constantly fails to develop. In this and other
similar eases we can bardly doubt that the existence of this
second, and now functionless, ovule carries us back to a time
when the ancestors of our present Pelea habitually produced
two seeds.

In Paliurus the fruit normally contains three, but sometimes
only two loculi, each with a single seed. Again, in Myagrum
the ovary is spuriously 3-celled, but the two outer cells pro-
duce no seed. In Hemanthus there are three cells, each con-
taining a seed, only one of which, however, comes to maturity.
In Convallaria there are three cells, each with two ovules, but
only one of the six is generally developed. In Phillyrea there are
two ovules, only one of which is developed. In Canarium there are
two, only one of which is developed; and we meet with similar
cases in Gyrinopsis, Jasmine, ZEsculus, Cordia, and many others.

From the point of view as to the origin of these differences
the variations in seedlings offer great interest. For example,
out of 135 seedlings of Lepidium sativum (which, as already men-
tioned, differs from thé rest of the genus in having tripartite

LINN. JOURN.—BOTANY, VOL. XXII. 2K

398 SIR J. LUBBOCK —PHYTOBIOLOGICAL OBSERVATIONS.

cotyledons), no less than 25, or as much as 187 per cent., differed
from the type, aud had this character more or less imperfectly
developed. Under cultivation the seedlings of Primula sinensis
often have one of the cotyledons deeply bifid; in one lot of
seedlings this was the case with, for instance, over 20 per cent.
of the plants.

In Poterium Sanguisorba the calyx-tube generally contains one,
but sometimes two or even three achenes. In Ranunculus, occa-
sionally, the petioles of the cotyledons are connate.

In @nothera the cotyledons are either straight, or with one or
both involute. Irmisch states that the cotyledons of Clematis
recta, which are usually aerial, sometimes remain below ground ;
while the reverse is the case in Melittis, and, according to Winkler,
in Dentaria and Mercurialis, for the cotyledons, which are
generally fleshy and subterranean, sometimes rise above ground
and form small green leaves.

Many species occasionally have one or both cotyledons divided*.

In Rheum the cotyledons are generally parallel, but in some
cases one of them is placed more or less obliquely with reference
to the other.

In Fagopyrum the position and arrangement of the cotyledons
in the seed vary greatly. The cotyledons are turned about
after striking into an angle of the seed, and then, following the
testa, assume very various positions. In Carum carvi the coty-
ledons sometimes have their backs to the axis of the fruit, and
sometimes their edges, while at others they are oblique.

In Cheiranthus pygmeus the cotyledons are said to be sometimes
accumbent, sometimes incumbent, and straight or convolute, even
in the same pod.

It would be easy to multiply such cases, but I will only
mention one more, in which I will venture to suggest a reason
for the variation. In Acer (fig. 133, 1), the embryo origi-
nates in a short tubular prolongation of the micropyle, and
is at first straight with an extremely short turbinate radicle,
and ovate obtuse, closely adpressed cotyledons. As grow th
continues the embryo extends itself along the lower side of the
seed, and curves with it, becoming gradually lanceolate, or oblong
lanceolate (fig. 133, 2). When the cotyledons have reached the
upper, narrow end of the seed, the curvature of the wall turns

* See Goebel, Grundziige Syst. Spec. Pflanzen Morphologie, p. 505.

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 399

them down again on themselves (fig. 133, ,). This growth is con-
tinued until the tips reach the radiele again, and the ultimate
arrangement of the embryo differs aecording to whether they
then curve inwards or outwards. This, again, seems to depend on
the exact direction of the growth of the cotyledons. If they
strike (fig. 133, 5) against the process which encloses the radicle,

Fig. 133.

Acer Pseudo- Platanus.
. s: St .
| Sections of seed in seven successive stages, showing growth of embryo. x 3.

then their general direction naturally carries them outwards,
until the wall of the seed again turns them upwards, so that
they become plicate ; if, on the contrary, the tips of the coty-
ledons pass just within the mieropylar process and touch the
radicle, then they are compelled to grow in the opposite direction,
and they become spirally coiled. Inthe specimens examined the
latter arrangement was exceptional.

Forms of the First Leaves.

Time does not permit me to enter into any particulars with
reference to the first leaves and the transition between them aud
those of the final form. I will only observe that the first leaves
are generally simple, or at any rate simpler than those which
follow.

In species with trifoliate leaves, the first leaf is generally
simple, as in the Clovers. When the mature leaves are pinnate,
the first ones are generally trifoliate; and when the final leaves
are bipinnate, the first ones are generally pinnate. I have already
observed that species with lobed or palmate leaves often commence
with one or more which are entire and heart-shaped.

400 SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS.

In most cases, therefore, the first leaves are simpler than those
which follow. In species from very dry localities, however, the
reverse is often the case. For instance, in Lasiopetalum ferru-
gineum (fig. 134), from Adelaide, the first few leaves are spatu-
late and more or less lobed, the final ones linear. Again, in
Dodonea viscosa, also from Adelaide, the first leaves are lobed,
the subsequent ones simple.

Fig. 134.

Seedling of Lasiopetalum Jerrugineum, Half nat. size.

Relation of the Seedling to the Seed.

To return for one moment to the case of seedlings. The question
of course arises whether the embryo conforms to the seed, or
whether the shape of the seed is determined with reference
to the form of the cotyledons. The seeds, however, evidently
depend on the habits, conditions, &c. of the parent plant. I
have elsewhere dealt with the structure of the seed, and must
content myself here with the simple statement that we have no
reason to suppose that it is influenced by the form of the embryo.
On the other hand, it seems equally clear that the form of the
embryo, and especially of the cotyledons, is essentially influenced
by that of the seed.

The Tea (Thea), for instance, presents us with a very interest-
ing case, in which the cotyledons vary greatly in shape, following
that of the seed, and depending on the number of ovules which
develop; they are contained in a wooden capsule, and are
variously compressed. In Citrus also the cotyledons are unequal

SIR J. LUBBOCK—PHYTOBIOLOGICAL OBSERVATIONS. 401

and irregular, several embryos being contained in each seed, and
sometimes squeezed together in the utmost confusion.

In several other cases, among those which have been already
referred to, we can hardly doubt, I think, that the form of the
cotyledon has been affected by the seed and not vice versá.

Let me in conclusion take one other illustration. The coty-
ledons of the Sycamore (fig. 9) are long, narrow, and strap-like ;
those of the Booch (fig. 13) are short, very broad, and fan-like.
Both species gte aperispermic, the embryo occupying the whole
interior of the seed.

Now in the Sycamore the seed is more or less an oblate
spheroid, and the long ribbon-like cotyledons, being rolled up
into a ball, fit it closely, the inner cotyledon being often
somewhat shorter than the other. On the other hand, the nuts
of the Beech are more or less triangular: an arrangement like that
of the Sycamore would therefore be utterly unsuitable, as it
would necessarily leave great gaps. The cotyledons, however,
are folded up like a fan, but with more complication, and in such
a manner that they fit beautifully into the triangular nut.

Can we, however, carry the argument one stage further? Why
should the seed of the Sycamore be globular, and that of the
Beech triangular? Is it clear that the cotyledons are consti-
tuted so as to suit the seed? May it not be that it is the seed
which is adapted to the cotyledons ? In answer to this we must
examine the fruit, and we shall find that in both cases the cavity
of the fruit is approximately spherical. That of the Sycamore,
however, is comparatively small, say } inch in diameter, and
contains one seed, which exactly conforms to the cavity in which
it lies. In the Beech, on the contrary, the involucre is at least
twice the diameter, and contains from two to four fruits, which
Consequently, in order to occupy the space, are compelled (to
give a familiar illustration, like the segments of an orange) to
take a more or less triangular form.

Thus, then, in these cases, starting with the form of the fruit,
we see that it governs that of the seed, and that of the seed
again determines that of the cotyledons. But though the coty-
ledons often follow the form of the seed, this is not invariably
the case: other factors must also be taken into consideration ;
but when this is done, we can, I venture to think, throw much

light on the varied forms which seedlings assume.

LINN. JOURN.— BOTANY, VOL. XXII. 2n

402 DR. M. T. MASTERS ON THE FLORAL

On the Floral Conformation of the Genus Cypripedium.
By Maxwet T. Masrvas, M.D., F.R.S., F.LS., &c.

[Read 18th November, 1886.]
(Prate XX. and ten Woodcuts.)

GENERAL CONFORMATION OF ORCHID FLOWERS.

The usual conformation of an Orchid flower *, the relative dis-
position of its parts, and the general course of their development
are so well known that for the present purpose it is only neces-
sary to advert to the relatively excessive development of the
median petal or lip, and to allude to the construction of the
* column " from three, or more frequently from six (potential)
stamens arranged in two rows, an outer and an inner, and con-
crescent with three styles.

According to the plan followed by Darwin t, and now gene-
rally adopted, the outer stamens are, for-convenience’ sake, de-
noted as Ai, A2, As, the three inner ones being marked by
corresponding smaller letters a1, a2, as. Similarly, the carpels of
the gynæcium may be marked Gi, Ge, Gs. S may stand for the
sepals, P for the petals, Ps or L for the median petal or lip, and X
for the staminodei. In the great majority of Orchids the outer
median stamen A1, placed opposite the median petal, is the only
one that attains its full development. Any or all of the other
five may, under exceptional circumstances, be present, either in
a perfect anther-bearing state, or more commonly in the guise of
petals. Thus, in Odontoglossum crispum Y have met with all six
stamens fertile and free $. Arundina pentandra is described by
Reichenbach as having five stamens ; and Fritz Müller mentions

* See Eichler, ‘ Blüthen Diagramme’ (1875), i. p. 179 e£ seg.; Gérard, “ Sur
l'homologie et le diagramme des Orchidées,” Ann. Sc. Nat. Bot. 6° sér. t. Vil
p.213, 1878; Van Tieghem, ‘Traité de Botanique' (1884), p. 1387, and the
same author, ‘Sur l'Anatomie comparée de la Fleur’ (1871), p. 140; Pfitzers
‘Morphologische Studien über die Orchideenblüthe' (1886) has been issued
since this paper was written.

t ‘ Fertilization of Orchids,’ ed. 2 (1877), p. 236.

1 It must be borne in mind that throughout this paper the flower is supposed
to be in the adult stage subsequent to any torsion, and that the numbers A1 &e.
do not apply to the order or period of development of the several parts.

§ Masters, ‘ Pflanzen Teratologie, ed. Dammer (1886), p. 439, figs. 213, 214.
In this edition numerous instances are cited of polyandry in Orchids in addition
to those mentioned in the original edition (1869), p. 382. See also Masters,
* Orchids single and double,” in ‘ Gardeners’ Chronicle’ (1885), May 5, p. 597-

CONFORMATION OF THE GENUS CYPRIPEDIUM. 403

an Epidendrum one form of which is constantly triandrous (Bot.
Zeit. 1872, no. 10, p. 152). By some observers it is considered
that traces of the undeveloped stamens of the outer, and perhaps
also of the inner, row are to be seen in the shape of lateral out-
growths from the column, or, as in Cypripedium, in the form -of
lateral lobes of the staminode ; but this opinion, as also the view
that one or more stamens are incorporated with the lip, I believe,
for reasons hereafter given, to be erroneous *.

In Zygostates cornuta (according to Pfitzer, J. c. p. 88), in ad-
dition to the ordinary median stamen Au, there are present As,
As, and as, in the form of sterile filaments. Lindley, however
(‘ Vegetable Kingdom, ed. 3 (1853), p. 183^), referred the
staminodes of this plant entirely to the inner series. In The-
lychiton, Endlicher, Iconog. t. 29, there is a central stigma
surrounded by a lobed cup bearing on one of its lobes an
anther. Crügert describes and figures an Isochilus in which
the flower is normally triandrous, but which often bears five
anthers with a filament proceeding from the front of the column
just beneath the stigmatic cavity. Van

Tieghem f, relying exclusively on the distri- Bract.
bution of the vascular bundles, gives an ac- Sı
count of the arrangement of the parts, in some Pi Hn
measure at variance with the generally received A,
opinions, but he concludes thus :—“ l'androcée al a2

des Orchidées est done constituée idéalement Gi

par six étamines superposées, en deux verticilles Ga Gs
ternaires aux divisions du perianthe." a3

The ternary nature of the gynecium is gene- da 4s
rally admitted, though it is marked by abortion Ps
and conerescence. In a flower of Pleione bir- 82 85
mannica recently examined all three stigmas Axis.
were present. The distribution of the nerves Diagrammatie

illustration of the

in the column showed that the large m up of arrangements of
the column, which might readily be mistaken the parts of an
for lateral stamens A2, As, were really subdivi- Orchid flower.

sions of A1.
The appended scheme, showing the relative position of the

* See Robert Brown, “Observations on Apostasia," in Wallich, Plante Asiat.
Rarior. i, p. 74, and Miscellaneous Botanical Works, ed. Bennett (1866), i. 499.
+ Journ. Linn. Soc., Bot. viii. p. 135.

1 ‘Anatomie Comparée de la Fleur,’ p. 142.
212

404 DR. M. T. MASTERS ON THE FLORAL

parts in the generality of Orchids, may be useful for comparison.
Letters in italies indicate parts that are generally absent, but
which are potentially, and sometimes actually, present.

CONFORMATION OF CYPRIPEDIUM.

In an ordinary Cypripedium flower, as seen in the adult con-
dition, there are three sepals, the median one free, the two lateral
ones not only concrescent, so as to form in appearance a single
sepal, but removed from their typical lateral position into à
median one opposite to the median sepal. The result is that the
flower appears to have a two-parted calyx, the significance of
which wil become apparent when the details to be hereafter
given are considered. The corolla, or inner row of the perianth,
consists of two lateral petals, and of one median labellum or lip,
usually bag-shaped and differing greatly in appearance from the
lateral petals.

The andrecium is composed of one median stamen dilated into
abroad shield-like staminode, and of two lateral fertile stamens
within the preceding. The gynecium consists of an inferior
ovary, distinctly 3-celled in Selenipedium, but 1-celled with three

parietal placentas in other Cypripedia. One Bract.
of the three carpels is median, in the same line $i

as the median staminode, the other two carpels Pi Po
being lateral and superposed to the lateral x
sepals. The styles are concrescent with the xi az
stamens into a column, the apparently solitary Gi
stigma being developed in the median line, Go G3
superposed to the staminode. a3

Thus the largest sepal Si, the largest petal A A
Ps, the staminode x, and the stigma G are all

: 5 Ps
in the median plane of the flower. This arrange- S S3
ment may be expressed by the accompanying ee
plan. Axis.

The points specially worthy of note are the Diagrammatic

lip, the andreecium, and the gynecium. In the illustration st e
development of the flower, acecording to Crüger*, iiam the
irregularity first shows itself in the median flower in CP
petal which develops itself into the lip. pedium. |
As to the andreecium, the generally accepted view is that which
R. Brown originally propounded (but which he subsequently
modified). According to this view the large staminode is the

* Journ. Linn. Soc. Bot. vol. viii. p. 134, t. 9.

CONFORMATION OF THE GENUS CYPRIPEDIUM. 405
median barren stamen of the outer row, Ai. The two lateral
fertile stamens, a1, a2, belong to the inner row. All the other
stamens are absent. It is possible, especially in cases of Peloria
(see post.), that the lateral stamens of the outer row, Az, As, may
be sometimes represented in the lateral lobes of the staminode,
although, from the distribution of the vessels, I do not think this
is usually the case. This staminode, moreover, is often more or
less three-angled, three-lobed, or two-lobed with a central knob
or short projecting filament. Link, as cited by Eichler *, was of
opinion that the fertile anthers originated from the division of
one median stamen, the staminode being looked on as an enor-
mously developed connective. But although the staminode often
shows signs of bi-partition, the whole weight of evidence, whether
from morphology, teratology, or anatomy, is against Link’s view.
The teratological evidence will be alluded to later on, and that
furnished by the anatomy of the column may be deferred till the
nature of the style is considered.

The large stigma of Cypripedium occupies, as previously

stated, a median position, and this position Bract
gives rise to some perplexity on account of its G i
appearing to be a single organ. G G
The readiest mode of explaining the existence 3 à
iagrammatic

of a single stigmatic lobe would evidently be to illustration of the
assume that the two lateral carpels, at least their position of the
: : AE stigmas in Cypri-
stylar portions, were abortive. An examination pedium, which,
of the arrangement of the vascular bundles however, are con-
: crescent, not 180-

of the peduncle and of the column supplies Jatea.

another and more correct explanation.

DISTRIBUTION OF THE VASCULAR BUNDLES.

Tracing the nerves from the stalk of the flower upwards to the
ovary, and thence into the column, the arrangement is seen to be
the following :—In the peduncle there are six vascular bundles in
one ring. Following these upwards, three are found to corre-
spond to the dorsal sutures of the carpels and three to the
ventral or placenta-bearing margins. Although at first in one
ring, at the level of the emergence of the perianth-segments,
they form two rings one within the other, and their arrange-
ment may be diagrammatically represented as follows, Ni, Ne, N3
occupying the same positions as those occupied by the sepals, outer

* * Blüthen Diagramme,’ p. 181.

406 DR. M. T. MASTERS ON THE FLORAL

stamens, and carpels, while nı, n2, n3 are in a line with the three

petals and with the three inner stamens. N

In fact, the median nerve Ni, after leaving dí m
the ovarian wall, divides radially into two n3
divisions, one of which passes into the median N; Ns
sepal Si, where it soon breaks up laterally S
into numerous subdivisions, while the other nerves at the base
passes into the staminode A x . of the flower.

The nerves N2, Ns pass into the corresponding lateral sepals
without dividing radially, so that the cords corresponding to the
stamens A2, As are, in this case, entirely wanting. The inter-
mediate nerves ni, n2 each divides radially, one division passing
into a lateral petal, the other into a fertile stamen. The median
nerve of this series, na, passes into the lip, Ps, where it speedily
trifurcates by lateral subdivision, but does not divide radially ;
for, as the inner median stamen a3 is not developed, there is no
vascular cord required for it*. The three cords for the gynecial
row are numerically as complete above as they are below the
perianth, the primary nerves Ni, N2, Ns supplying the three
carpels Gi, Ga, Gs.

A. difficulty, as before remarked, arises as to the large and
apparently single stigmatie lobe which is placed in the median
line. It might readily be assumed that the two lateral styles
were in this case absent, while the median one, Gi, was alone
present. But this inference does not accord with the position
of the placentas, two of which are placed laterally and one in the
median line superposed to the lip. The placentas of course alter-
nate with the styles, the styles in fact being prolongations of the
dorsal sutures, while the placentas correspond to the united
margins of three carpels. An examination of the vascular
bundles in the style above the emergence of the stamens shows
the three bundles Ni, N2, Ns very distinctly alternating with the
3-lobed mass of conducting tissue continuous below with the
placentas. Into the stigmatic lobe itself, which is a triangular
cushion-shaped body with the apex of the triangle in the middle
line, it has appeared to me that the median bundle, Ni, does not
always enter, though it does so in some cases. The other two
bundles, N2, Ns, corresponding to G2, Gs, however, are always

* Darwin's view that the nerves corresponding to A2, À3 are in any degree
incorporated with the lip is, I believe, an error; the nerves figured by him at
the side of the lip result merely from the lateral branching of the bundle n3.

CONFORMATION OF THE GENUS CYPRIPEDIUM. 407

present. The mass of the stigmatic lobe is made up of large
thick-walled polygonal cells, many with nuclei and nucleoli;
through this mass of parenchyma the vascular cords pass, the
tracheæ being nearest to the lower or anterior edge. The lower
or anterior part of the stigmatic lobe consists of several layers of
cells of a very different character from those constituting the bulk
of thelobe. They are elongated or club-shaped, in several layers,
disposed more or Jess horizontally, thin-walled and nucleated, and
bounded on the lower surface by a layer of papular cells.

This layer of stigmatic tissue, though to outward appearance
unbroken, is seen, on inspection of microscopic sections, to pass
inward and to separate the lobe into two divisions, correspond-
ing to G2 and Gs. Perhaps in some species G1 may be cut off in
a similar way, but this has not been the case in any of the flowers
examined by me. The stigmatic lobe, therefore, so far from being
single, as it appears to be at first sight, is certainly two-fold and,
perhaps, in some species three-fold, the two or the three styles
at their uppermost ends being united in the median line like
the two lateral sepals. The column of Cypripedium, therefore,
is made up of three stamens and of three styles. Of the three
stamens the upper median belongs to the outer row and is deve-
loped as the staminode; the other two lateral ones are fertile and
belong to the inner row. Of the three ovaries and three styles all
remain, but of the three stigmas the upper or median one becomes
abortive, while the two lateral ones are joined into one compound
lobe *.

TERATOLOGY OF CYPRIPEDIUM.

Adverting now to the teratological appearances presented by
this genus and its allies, I propose to confine attention to those
malformations which more directly elucidate the plan of Orchid-
structure, and to omit reference to those numerous cases of pure
distortion or perversion which have no direct morphological signi-
ficance from the present point of view.

So regarded, these malformations may conveniently be grouped
under the heads of defect, of excess, or of perversion of the natural

* The description above given of the course of the vascular bundles of the
column &c. agrees in the main with that given by Van Tieghem, ‘ Anatomie
Comparée de la Fleur,’ p. 144, tab. xii. figs. 209-219, and with that propounded
by Gérard, l. c. p. 236. Van Tieghem, however, says nothing about the arrange-
ment of the cords in the stigmatic lobe, nor is it alluded to by any other author,

80 far as I am aware.

408 DR. M. T. MASTERS ON THE FLORAL

process of development. Someof these changes may thus be con-
sidered as progressive, others as retrogressive, and not infre-
quently as reversions to a prior and simpler condition bei

OLIGOMERY.

I. Reduction of Parts.—A. dimerous condition is one of the
commonest deviations from the usual typical
structure of Orchids in general, This arrange- P
ment may be thus represented, the two sepals S A 8
being lateral, the solitary fertile stamen A and G
the stigma G being in the median line. This L
dimerous condition has some indirect analogy , Diagrammatic
: : : illustration of di-
with the natural arrangement of parts in Cypri- merous Orchid.
pedium, where in the adult flower there are two 7
sepals only (one of these, however, resulting from the concres-
cence of two), and in the presence in the median line of two
much-modified parts, the lip and the staminode,

the inner staminal whorl being represented by Bract.

a pair of organs placed at right angles to these, S

and, as it were, decussating with them. P P
In abnormal flowers of Cypripedium Sedeni x

the dimerous condition is associated with the al a2

presence of a single median fertile stamen in G

the position usually occupied by the stami- L

node x. In these cases the median sepal is S S

generally absent, whilst the two lateral ones erence 7

are either united or partially separated, and Axis.

more or less dislocated or turned to the lower Diagrammatic

part of the flower. No lateral petals are illustration of ar-

rangement of parts

formed, but a petal is placed above in the
median line, in the position usually occupied
by the median sepal, and in form and colour like one of the
lateral petals. Opposite to this petal is placed a second one,
lip-like in character, or sometimes flat T.

in Cypripedium.

* A very large number of specimens have been examined by myself during
the last few years, and upon this examination these remarks are, for the most
part, based. I avail myself, however, of this opportunity of thanking Mr.
Hansen, of St. Albap's, for the privilege of inspecting a large number of draw-
ings executed by him in the rich Orchid nursery of Mr. Sander.

t See Le Marchant Moore, in ‘Journal of Botany’ (1879), vol. xviii. p- =
Masters, ‘ Vegetable Teratology,’ p. 401; German edition, ed. Dammer (1886),
p. 460 ; Magnus, Sitzungsbericht d. Gesellsch, Naturforsch. Freunde zu Berlin,

CONFORMATION OF THE GENUS CYPRIPEDIUM. 409

In a monandrous flower of C. barbatum sent me by Mr. Bull
there were two lateral sepals turned downwards, an upper median
Fig. 1. Fig. 2.

Fig. 1. Shows flower of C. barbatum described in the text. The column is
shown as seen from the side.

Fig. 2. Diagram showing the arrangement of the parts in fig. 1.

petal P, and, opposite to it, a lip L. The column had one fertile

median stamen, Ai, and in front of it one stigma, SM, the other

parts being absent. In Mr. Hansen's series of P
drawings [ have seen illustrations of a similar Pt
malformation and dislocation in C. Sedeni x, G
C. barbatum, C. venustum, and C. Lawrenceanum. L
The relative position of the parts in such s s
flowers may be thus expressed :— ;

The lateral sepals are usually conjoined and, Diagrammatic

as it were, dragged out of place and made to illustration of figs.
assume a median position more or less opposite uix
the upper median petal ; but Mr. Hansen's drawings show various

(1878) June 18 and (1885) July 21; also Verhandl. d. Bot. Vereins d. Prov.
Brandenburg (1882), xxiv. p. 111, and xxv. p. x, where other references to the
literature of the subject are given. Dimerous flowers associated with the pre-
sence of one median fertile stamen, Ar, are frequently met with in this hybrid

Cypripedium.

410 DR. M. T. MASTERS ON THE FLORAL

intermediate stages between perfect union and perfect dialysis of
the sepals. "Where the upper median sepal is present, it some-
times happens that the two lateral sepals are greatly reduced in
size, so that various stages occur between the ordinary condition
and one of dimery and decussation of parts. In all cases where
the median stamen, A1, is fertile, the large shield-like staminode,
x, which forms so characteristic a feature in ordinary Cypripe-
dium flowers, is absent. In one flower of C. Lawrenceanum in
which the perianth was dimerous, the andrce-

cium was trimerous, the stamens present being P
one of the outer row, A1, and two of the inner S A $
row, al, a2, thus :— G
It will be observed that one stigma alone was ao a
developed in the median line, but whether that L
one represented Gi or a combination of G2 and , Diagrammatic
illustration of ar-

Gs, as under ordinary circumstances, I was not rangement in a
able to ascertain. monstrous flower
: of C. Lawrence-
Monandrous and diandrous flowers have been anum.
already alluded to; so that I may now pass on

to flowers in which the number of stamens was greater.

PLEIOMERY.

IL Increased Number of Parts—This is illustrated by the
occasional presence of an increased number of lips. This aug-
mentation may be due to an actual increase in the number of
petals, to a division of one petal into two, or, more frequently,
either to the development of the lateral petals in the form of lips,
or to the appearance of one or more of the stamens in the guise
of labella. An instance of increased number of petals in Cypri-
pedium superbiens was lately communicated to me by Mr. James.
In this flower the sepals, andreecium, and the gynecium were
all normal; but the corolla was represented by four petals, the
two lower ones in the form of lips (see figs. 3, 4). Possibly these
were not really two labella, but were constituted of one dividing
into two at a very early stage; but this could not be determined
in the adult flower.

Plurality of the stamens occurs in different degrees. Hitherto
I have not met with any absolute increase in number beyond the
typical six; but I have now met with instances of the presence
in Cypripedium of from one to six stamens in different flowers,
either in a fertile or in a petaloid condition, and upon some of
these I shall now comment.

CONFORMATION OF THE GENUS CYPRIPEDIUM. 411

Fig. 3.

Two-lipped flower of Cypripedium superbiens.

Triandrous Flowers.—Magnus*
describes a triandrous flower in
Cypripedium barbatum ; and I have
seen a flower of the same species
shown me by Mr. Lowne also with
three stamens, viz. A1, a1, a2. In
two flowers of C. Lawrenceanum
(fig. 5), received from Mr. Stevens,
the perianth was tetramerous and
dicyclic, while the andrecium was
trimerous, the outer median stamen,
A1, being fertile, and the two lateral
stamens of the inner series, a1, a2,
represented by shield-like stami-
nodes, thus exactly reversing the

Fig. 4, B
=

Cypripedium superbiens,
Plan of flower (fg. 3).—8,

sepals; P, petals; COL, column
or staminode; at, 22, inner fer-
tile stamens.

* Sitzungsber. d, Gesellsch. Nat. Freund. Berlin, June 18, 1878.

412 DR. M. T. MASTERS ON THE FLORAL

ordinary staminal arrangement in this genus. The stigma, Gi,
was superposed to the median fertile stamen. In a specimen of
C. Spicerianum sent to me by Mr. Fraser, the three sepals were

free, and the an-
droecium was repre-
sented by the three
stamens of theinner
row al, a2, a3 all
fertile. In C. Seden?
x I have also seen
all three stamens
present in the case
of flowers the sub-
ject of regular pe-
loria to be hereafter
mentioned.
Tetrandrous

Flowers——In Uro-
pedium there are at
least four stamens
present, viz. Al, asa
staminode, and a1,
a2, a3 fertile. The
staminode in this
case may possibly
represent A 2 and A3
as wellas A1. Asa
Gray describes* a
tetrandrous flower
of Cypripedium can-
didum ; and I have
lately examined a
flower of C. Law-
renceanum in which

Flower of Cypripedium Lawrenceanum, with 4-parted
perianth and three stamens, one outer, Ar, fertile,
two lateral, a1, a2, barren. A, flower; B, column,
from the front; C, column, from the back; D,
column, from the side, one of the staminodes, 42,
removed.

there were four stamens present, viz. A1 as a staminode, a1, a2
fertile, and a3 developed as a lip, which was, as it were, slipped
within the cavity of the true lip, so that there were two lips one
within the other (figs. 6 and 7).
In a Cypripedium sent me by Mr. Haywood there were also
* Seemann’s Journ. of Botany, iv. (1866) p. 318.

CONFORMATION OF THE GENUS CYPRIPEDIUM. 413

four stamens—A1, a1, and a2 developed as flat petals, while a3
was present in the form of asmall lip. In this case not one of the
stamens was fertile; but all were petalodic, as was also one of
the styles Gi, thus affording an instance of a double-flowered

Cypripedium * (fig. 8, p. 414).

Fig. 6.

PIA

Fig. 7.
=

v
RU
rà

s Ax OL
ayy Pr :

x 7r |

2» TD LLL

Cypripedium Lawrenceanum.

Plan of arrangement of parts.

La

a3 E ó

Ww lt,

Cypripedium Lawrenceanum.—Flower in section, with
some of the parts removed to show one lip, 25, en-
closed within the normal one L. S, sepals, the two
lower ones concrescent; PP, petals cut across at
their base; Ai staminode ; S, stigma.

* Masters in Gard. Chron. May 9, 1885, p. 597, fig. 112.

414 DR. M. T. MASTERS ON THE FLORAL

Fig. 8.

A double-flowered Cypripedium.

Sr, S2, sepals; P, P, petals; L, lip; Ar, outer petal-like stamen; a1, 42, 43»
three inner stamens, of which two are petal-like and one lip-shaped ; G1, a
petal-like style.

Hexandrous Flowers.—Last in the series may be mentioned a
flower of Cypripedium Bedeni x in which all six stamens were
present, two fertile, A1 and as, opposite one another, while
all the others were represented by saccate lips (irregular
peloria).

CONFORMATION OF THE GENUS CYPRIPEDIUM. 415

Tendency to Increase the Inner Row of Stamens.

It is perhaps worthy of note that in the numerous cases I have
examined, wherein the number of stamens in Orchids generally
has been increased, the increase has been relatively more frequent
in the inner staminal cycle than in the outer; thus, in my
experience, it has been more common to find the inner series
a1, a2, and even aa, developed either in a fertile or in a petalodic
condition than it has been to meet with corresponding changes
in the outer series A1, A2, A3. The frequent abortion of the
outer row may be indeed induced by the necessities of space,
or by the rhythmic alternation of activity and quiescence, which
is so marked a feature in plant-life. In the nearly allied groups
Burmanniaces and Apostasiaces the inner row of stamens is the
one most usually developed, as it is also in Restiacez and Xyri-
dacez ; while in Iridacez, on the contrary, it is the outer episepa-
lous row which is usually present, and the inner row is absent*.

PARTITION.

Under this heading may be included those cases in which an
organ, primitively simple, becomes, in course of development,
more or less divided by a so-called process of fission. They
require to be carefully discriminated from instances of dialysis,
or the separation of parts usually concrescent or inseparate. An
illustration of this partition was afforded by a flower of Cypripe-
dium Spicerianum forwarded by Mr. Hansen. In this flower the
dorsal or median sepal, usually entire, had two distinct midribs
and a shallow notch at the apex; the lower united sepals, the
petals, the lip, the inner row of stamens a1, a2, and the stigma, G,
were all of the usual form and appearance. The staminode, how-
ever, was divided to the base into two polygonal, shield-shaped
lobes occupying a lateral, not a median, position. At first sight,
therefore, it seemed as if the usual staminode representing A1
were absent, while the other two of the outer row As, As were
present in the form of petaloid shields. In fact, the appearance
was exactly the same as in certain flowers mentioned in this
communication in which the two side stamens of the inner row,
al, a2, were represented by shield-like staminodes.

* For a case in which the inner stamina] series in Jris was developed, see
Heinricher, Sitzungsb. der k, Akad. Wissensch. Wien, November 3, 1881, p. 509,

separate paper, p. 51, tab. v.

416 DR. M. T. MASTERS ON THE FLORAL

That they did not, however, in this case, belong to the inner
row was manifest, because the inner row was present in its
normal state and position. Moreover, their insertion was dis-
tinctly external to that of the fertile stamens, while the back of the
column presented a well-marked central ridge concealing a single
central vascular bundle, which, near the apex, divided into three
divisions—one central, minute, and two lateral, passing into the
petal-like shields. The minute central portion evidently repre-
sents the central branch of the filament which in ordinary Orchids
bears the anther, while the two lateral lobes are side-outgrowths
from the same filament, thus resembling the trifid stamen of
Allium, in which the central division bears the anther. Again,
the indieations of fission in the median sepal correspond to the
similar division of the staminode superposed to it. Indications of
this bipartition, or rather tripartition, of the posterior or median
staminode are common in many species of Cypripedium, which
present a central knob or boss and two lateral lobes.

On making sections through the peduncle, above the insertion
of the bract, a ring of six vascular bundles was seen. At the
base of the ovary the posterior bundle bifurcated; and this twin
bundle was traced throughout the whole length of the prominent
ridge forming the dorsal suture of G1, and in to the median sepal
Si, where the bundles formed the double midrib. A similar
section across the column at its base also showed six bundles,
five single and one bifurcated; this bifureated bundle passed
into Ai, its two divisions passing into the shield-like divisions
of the staminode. Thus the partition began at the

base of the flower in the posterior median line, and Sı
continued in it all the way up; so that whereas, Al
under ordinary circumstances, the arrangement in G1

the median line would be— L

in the partieular flower now under consideration the arrangement
was— SS

the gynxcium being only partially implicated, and A A

the lip not at all. G G

L
DISPLACEMENT.

III. Change in Position of Parts.—This has already been
incidentally alluded to in connection with the reduced number
of parts; but the importance of the subject necessitates some
further mention. In the dimerous forms of Cypripedium the
upper or median sepalis usually absent, the lateral ones are

CONFORMATION OF THE GENUS CYPRIPEDIUM. 417

present, either separate, when each occupies a truly lateral
position, or united from concrescence, when the conjoined pair
occupy a median, inferior, or anterior position. In these flowers
the side petals are absent, the petalline whorl being repre-
sented by a single median petal opposite to the lip, as shown in
Mr. Le Marchant Moore’s monandrous Cypripedium, and in
those figured in this communication, and as commonly happens
when the flowers of Orchids are dimerous. Professor Charles
Morren* describes and figures a flower of Cypripedium insigne
in which, by torsion of the axis, the inner sepal had become
lateral and, by a further twist, one of the lower sepals was
made to occupy the position of one of the lateral petals. Of the
two side petals, one was completely absent, while the other, by
a continuance of the twisting process, was placed in a median
position opposite to the lip. All the stamens were absent except
one of the inner row, apparently what we should now designate
as ai. Morren cites this flower as an instance of what he terms
“spiranthy.” The flower may probably have been susceptible
of other and simpler interpretations than that offered by the
Belgian Professor; but it is unnecessary to enter into the
question here, where the only object is to cite the flower as an
instance of displacement of parts and of a monandrous condition.

In a flower of C. Lawrenceanum given me by Mr. O’Brien
the flower had a peculiar oblique lopsided appearance by reason
of the lateral petals being united with the upper sepal; and
while the sepal was, as it were, dragged downwards, the petal
Was correspondingly raised; the consequence being that neither
part occupied its natural relative position, 90°, but one was about
midway between the two, 7. e. 45^. The other lateral petal was

normal in form and in position.

PELORIA.
Regularity of Form and Arrangement of
Parts, or Peloria.

Peloria in Cypripedium, as in other plants, is either:—a, Re~
gular, when: the tendency to regularity is brought about by an
increase in the number of regular portions, a tendency which
seems to be a reversion to a simpler and probably primordial
condition, as is the case perhaps in Uropedium ; or, B, Irregular,
when symmetry is restored by the development in an irregular

* Bull. Acad. Roy. Belgique, t. xviii. p. 196, tab. ; Lodelia (1851), p. 55.
LINN, JOURN.— BO ANY, VOL. XXII. 2M

IV. Tendency towards

418 DR. M. T. MASTERS ON THE FLORAL

form of parts which are usually regular. Irregular Peloria,
when not associated, as it sometimes is, with compensatory defect
in other whorls, may be regarded as an instance of enhanced
complexity and of potential progressive development. The
opposite view, that the most highly complex is the form whence
the less differentiated case of regular peloria is derived by a pro-
cess of degeneration, finds, so far as I know, no warranty, either
in organogeny or in anatomy.

Regular Peloria.—
Prof. Asa Gray, as
before mentioned, has
described a ease of
regular peloria in
Cypripedium candi-
dum associated with
a reduced number of
parts, forming thus a
regular symmetrical
and complete, but di-
merous, Orchideous
flower—the first ver-
ticil of stamens bar-
ren, the second an-
theriferous, and alter-
nating with the car-
pels. Here we have,
as Gray remarks,
“clear, and perhaps
the first, direct de-
monstration that the
Orchideous type of
flower has two sta-
minal verticils, as
Brown always in-
sisted."* Mr. Tautz
has lately furnished d j
me with a specimen of f

Cypripedium (or Sele- Peloriate flower of Selenipedium Sedeni. Column, from
Jp P ( à dis beneath and from the Em anther, lip L, and stigmas
nipedium) Sedeni x, G,G,G* All magnified.

* Gray in Seemann’s Journal of Botany, vol. iv. (1866) p. 378, reprinted
from ‘American Journal of Science,’ xlii. July 1866. See also Magnus, ‘ Sitz-

CONFORMATION OF THE GENUS CYPRIPEDIUM. 419

in whieh there were three free sepals, two lateral petals, and one
flat lip, like the other petals in shape but broader. The column
had the usual staminode A x Fig. 10.

and two fertile stamens, a1 :
and a2, with, in addition, a
third stamen a3, projecting
in front of the stigma. The
stigma itself was three-lobed,
the posterior lobe G1 being
of the usual large size, the
two lateral lobes smaller (see
figs. 9, 10). The formula for
this very interesting flower
is thus 8S 3P x 8a 8G.
The usual zygomorphic state
was replaced by an actino-
morphic condition, as is met with in some Dendrobes, Paxtonia,
Thelymitra, Phalenopsis, Telipogon, &c. Endlicher's Thelychiton
(Endl. Gen. 1616) may also be mentioned as a pelorioid form
with a regular perianth and a cup-like andreecium, one lobe
of which bears an anther.

In a specimen of C. caudatum sent me by Mr. Bull (see
Plate XX.) the three sepals were free, the lateral petals as usual
prolonged into the long tails, which constitute so remarkable a
feature of the species, while the median petal, instead of having
the form of a long sac, something like the finger of a glove, was
partially flattened out and partially contorted— presenting, in
fact, an intermediate condition between that of the ordinary lip
and that of one of the ordinary lateral petals. The median
stamen of the outer row, A1, was present as a staminode ; and the
ear-like process on either side of it possibly represented the two
lateral stamens of that series, Az, As, though, from the analogy
of other cases, I am more disposed to regard them as sterile
branches of Ai, and to consider that A2, A3 were absent. The
three inner stamens, ai, a2, a3, were all present and all fertile.
The ovary was, as customary in this plant, 3-celled. So that
there was in this flower a tendency towards regular peloria, and

Plan of peloriate flower of Seleni-
pedium (see fig. 9). ~

ungsber. d. Botan. Vereins d. Prov. Brandenburg, xxi. January 31, June 27,
and November 28 (1879); also his * Kurze Notiz über dimere Zygomorphie
Orchideenblüthen und über ein monstroses Cypripedium," in the Abhandlung.
d. Bot. Vereins Prov. Brandenburg, xxi. p. 97.

2x2

420 DR. M. T. MASTERS ON THE FLORAL

a reversion to the (assumed) primitive type, as manifested by the
dialysis of the sepals (which,it may be incidentally mentioned,
occurs normally in the Canadian C. arietinum), partial regularity
of the lip, and the presence of the third or median stamen (a3)
of the inner row. The usual absence of this stamen seems to be
correlated with the great development of the labellum. The
stigma Gi (marked S~ in the drawing) was superposed to as,
either owing to the non-development of G» and Gs, or, judging
from other cases previously mentioned, it might have been really
compounded of G2 and Gs. Unfortunately the arrangement of
the vessels in the stigma of this flower was not examined.

In the flower in question the labellum was less highly organized
than usual, and the stamen a3 eonsequently had more space
wherein to develop, and thus the defect of the one was, to use
Sir Thomas Browne’s expression, “pierced out by the excess of
the other.” The close resemblance to Üropedium * in Mr. Bull's
flower was manifest in the dialysis of the sepals, the tendency in
the lip to become regular, and in the presence of all the stamens
of the innercycle. In the Bulletin of the Amsterdam Botanical
Congress, 1865, p. 62, Prof. Reichenbach describes a flower of
Selenipedium caudatum in which the lip was undulated and
twisted, bearing at its point a little thickening, as at the orifice
of the bag-shaped lip. “If,” says M. Reichenbach, “ this flower
had shown an anther in front of the labellum, it would have
proved the identity of the two types (Selenipedium and Urope-
dium); but the flower, in spite of the petaloid nature of the
labellum, had the column of Selenipedium, and on that account
offers additional argument in favour of the distinctness of the
two genera." In the specimen above described the third stamen
was present in front of the lip, and so far it satisfied Prof.
Reichenbach’s postulate. For a figure of a very similar mal-
formation in S. caudatwm see also Reichenbach fil. in Nov. Acad.
Nat. Cur. xxxv. (1870), n. 2, e. tab. In Reichenbach's plant the
three sepals were not separate.

Irregular Peloria.—Cases of partial irregular peloria in Cypri-
pedium ave not very uncommon, as, for instance, flowers in which
the lateral petals (or one of them) assume the saccate form of
the lip. This may occur without any increase in the number of
parts, or, on the contrary, some of the usually undeveloped

* Uropedium, see Brongniart, in Ann. Sc. Nat. sér. 3, xiii. p. 113; Reichen-
bach fil., ‘Xenia’ (1858), t. i. tab. 15; Bot. Zeit. 1876, p. 41.

CONFORMATION OF THE GENUS CYPRIPEDIUM. 421

stamens may be present in the form of lips, as in a flower of
C. Sedeni x , previously mentioned.

The change from the ordinary zygomorphie or irregular form
to a state of actinomorphy or regularity is thus shown, not only
in the equalized form and size of the parts of the flower, but also .
in their regularity of arrangement, either in a binary and decus-
sate or in a ternary radiating manner. Whether the peloria
be regular, or whether it be irregular, the result is thus seen to be
the same—a zygomorphic flower becomes actinomorphic.

COMBINATION OF TENDENCIES.

The changes just referred to may occur separately, but it is
more common to find them associated with others: thus an in-
creased number of stamens is associated with regular peloria,
and sometimes with independence of the sepals, as in Mr. Bull’s
specimen, or it may be that the increase in the number of stamens
is connected with their development in the form of lips, as in
irregular peloria. In irregular zygomorphic flowers, or in those
which show a tendency to become so, it often happens that if one
part, say the median sepal in Cypripedium, is larger than the lateral
petals, the part opposite to it in the succeeding whorl is also
larger than its fellows: thus in Cypripedium the large median or
dorsal sepal is on one side of the flower, the largest petal (the lip)
on the other; the large shield-like staminode is on one side, the
large stigma on the other. This is of course a frequent tendency
in flowers, as in Eremophila, where one large sepal is placed on
the opposite side of the flower to the largest lobe of the corolla.

CHANGES RESULTING FROM HYBRIDIZATION.

The Cypripedium Sedeni x , which presezts so many monstrous
conditions, is a hybrid raised in Messrs. Veitch’s nursery, between
C. longifolium and C. Schlimii, Rehb. f., in Gard. Chron. (1873),
p. 1431. By its 3-celled ovary it falls into the section Seleni-
pedium, which some prefer to consider a distinct genus, not
only for structural, but also for geographical reasons, as it is the
tropical American representative of the tropical Asiatic group.
It is noteworthy that, according to Mr. Seden, the raiser, Cypri-
pedium Sedeni x is produced whichever species be used as the
seed-parent. Whether C. longifolium be fertilized by the pollen
of C. Schlimii, or the reverse, the result is the same*. The

x Reichenbach, in Gard. Ohron. /.c.; Veitch Orchid Conference Report,
Journ, R. Hort. Soc. Lond. vii. (1886) p. 30.

422 BERBERIDEARUM JAPONI® CONSPECTUS.

petals of the hybrid are in form like those of C. longifolium,
while their colour is that of C. ScAlimii.

The changes which occur in this and other hybrid plants and
their derivatives may be ranged under three categories :—

1. Those in which there is an intermixture of parental cha-
racteristics, in varying degrees, but without any material change
in the floral symmetry. The changes, in fact, occur in those cha-
racters which are more or less directly of an “adaptive” nature.

2. Those in which there is in the offspring more or less dis-
sociation of the hybrid or mixed characters of the parents, 80
that the plant reverts more or less completely to one or other of
its immediate parents, as pointed out by Naudin.

3. Those in which the change is decidedly teratological, and
more or less affecting those * congenital" characters which con-
stitute the symmetry of the flower. These may be: 1, reversions
to an ancestral form, in which the structure was simpler, less
differentiated, the parts possibly more numerous, and isolated
rather than inseparate ; or, 2, they may possibly be indications of
the direction in which progressive modifications may take place
in the future.

EXPLANATION OF PLATE XX.

Peloriate flower of Selenipedium caudatum, real size (see p. 419). A, section of
ovary, magn. 4 diam. ; B, side views of column, enlarged 3 times. A1 A2 A5;
staminode; Az, A3 are probably not separate stamens of the outer row as
the lettering would indicate, but merely subdivisions of Ar; 41, 42, 43,
fertile stamens of the inner row; SM, stigma. In the plan the signification
of the letters is the same.

Berberidearum Japonie Conspectus, auctore ToKUTARO Iro.
(Communicated by W. T. TurseLron Dyer, C.M.G., F.R.S.
F.L.S.)

[Read 4th November, 1886.]
(PrarE XXI.)

Noras que de Japonicis Berberidaceis sequentur auctor compo-
suit eo tempore quo in Herbario Kewensi Aprili presentis anni
suum 'Florula Fudsisanensis ’ parabat, cujus partem hee con-
stituit. Quum tamen rem diligenter cogitavisset in conclusionem
pervenit si tales notas intra fines flore localis inseruisset hane de
toto esse permixturum sed amicis eximiis cll. Prof. Dan. Oliver
et W. T. Thiselton Dyer benigne consulentibus auctori per-

BERBERIDEARUM JAPONIE CONSPECTUS. 423

suasum est ut notas seorsum poneret. Quamquam auctor Ber-
beridacearum omnes species in J aponia cognitas includere
statuerat notas in monographiam tam elaboratam componere non
In ejus consilium venit. Maximi autem momenti observationes
de hae familia in iis invenientur. Diligens erat preparatio, de-
fectum tamen de multis rebus, genere prsesertim Epimedio, auctor
timet.
Srauntonia, DC.

Duz species ante cognite, alia Japonica, alia Chinensis, inter
se conjungi a me existimantur; una ergo tantum species huie
generi est relicta.

Sr. HEXAPHYLLA, Decne. Ann. Sc. Nat. xii. 1839, 105 (2°
série) ; Sieb. et Zuce. Fl. Jap. i. 140, tab. 76 ; Fam. Nat. n. 359;
Mig. Prol. 197 ; Franch. et Sav. Enum. i. 21, n. 92; Engler, Bot.
Jahrb. vi. 1885, 581.—Rajania hexaphylla, Thunb. Fl. Jap. (1784)
149.— Ya-mokkwa, Tokiwa Akebi, Shimada Yonan, Kwa-wi, Arb.
ii. fol. 2—Mubé, Iwasaki Tsunemasa, Honzō Dsufu, vol. xxx.
fol. 10 recto; Ito Keiske, Nihon Sanbutsushi, pt. prov. Omi, ii.
fol. 14-15, cum icone. Id., Koishikawa Shokubutsuyen Somoku
Dsusetsu, seu Icon. et Descript. plant. in hort. Koishikawa, Arb.
ii. 8, pl. iii.

Hab. in Japonie insula Kiusiu: prope Nagaski (Oldham!
1862, n. 536 in part, flor., 1863 n. 32 flor., et n. 33 fructif.). In
principali insula: mult. provinciis Omi, Yamato, Ise (monte
Asakuma-yama), Kii (Kumano) et alibi (Jto Keiske, loce. citt.) ;
circa Yokosuka (Savatier, n. 46). In archipel. Riukiu, ins.
Oshima (Déderlein ex Engler).

Corea: ad portum Hamilton (C. Wilford! 1859, n. 338 flor.).

Japonice: Tokiwa Akebi, i. e. Akebia persistens.

B. cutnensts, DO. (sp. pr.), Syst. i. 513 (1818); Prodr. i. 96;
Decne. loc. cit. 107 ; Bentham, Fl. Hongk. 14.

Hab. in China (Fortune! 1845, n. 166 flor): Hongkong
(Wright! for., Champion! flor. et frf., Wilford! 1858), mon-
tibus Gough et Victoria (Wilford! Feb. 1885, n. 338 flor.).

Distinctio inter St. hexaphyllam, Decne., et St. chinensis, DC.,
est valde obscura. Decaisne, Annales des Sciences Naturelles,
2° série, t. xii. 1839, 105, differentiam inter duas species mon-
stravit: illam “foliolis lanceolatis mueronato-setaceis," hane
*ovali-oblongis integris." Sed et Chinensis et Japonice plantze
folia sunt valde variabilia, atque specimen St. hezaphylle, quod

424 BERBERIDEARUM JAPONIJE CONSPECTUS.

ab Oldham in Nagasaki lectum (n. 32) in herb. Kew. vidi, ovali-
oblongis, emarginatis, sed specimen Hongkongense (n. 43),
St. chinensis foliolis 3-7 lanceolatis, acuminatis, vel oblongo-
ovatis, emarginatis, vel juvenalibus mucronato-setaceis mihi vide-
tur. Variabilitas foliorum a Benthamio in Flora Hongkongensi,
14, jam demonstrata est, qui dixit folia magnitudine valde varia-
bilia esse, superiora plerumque foliolis circiter 5 obovatis, aut
ovali-oblongis, circiter 2 poll.long., cum petiolulis 3 poll, sed
inferiora sepe foliolis 6 vel 7 oblongo-acuminatis.

Itaque duas species vix differre et in unam conjungendas mihi
videtur. Sive St. chinensis varietas St. hexaphylle seu forma
tantum sit investigandum est. Quamquam in presentia eam
colloco varietatem, forma tantum forsitan est.

Cl. Maximowicz qui nuper scripsit * accurata est diligentia
tua de Stawntonia” mihi notas ineditas benigne transcripsit,
quas infra dedi :—

St. chinensis.—St. chinensis et St. hevaphylle de differentia
suum dubius. Differentem nominat Decaisne eam foliorum et
mucronis formam quam unam et eandem comperi. Benthami
autem descriptio cum altera specie sque ac altera congruit,
exceptis purpureo-viridibus floribus, sed Wright (in schedula
plante Hongkongensis) albos flores nominat. Denique distinctio
sequens, a Candolleo data superest que probabiliter seque mala:

Foliola utrinque elevato-reticulata .. St. hexaphylla.
Foliola superne lavia.............. St. chinensis.

Cll. F. B. Forbes et Hemsley mihi roganti benigne responde-
runt, foliorum reticulationem esse in eadem planta eujus folia
venas immersas habeant. Id quod przterea meam existima-
tionem de unitate St. hexaphylle et St. chinensis confirmat.

AKEBIA, DC.

1. A. QUINATA, Decaisne, loc. cit. 107 ; Sieb. et Zucc. Fl. Jap.
i. 143, tab. 77; Fam. Nat. n. 355; Mig. Prol. 197; Franch. et
Sav. Enum. i. 21, n. 88.— Rajania quinata, Thunb. Fl. Jap. 148.—
Akebi, Iwasaki Tsunemasa, Honzó Dsufu, vol. xxx. fol, 8 verso.—
Tsuneno Akebi, Jt0 Keiske, Sómoku Dsusetsu, Arb., vol. ii. 1,
pi. 1. \
Hab. in Japonia (Buerger ! sub nom. Rayana quinata, Thunb.,
comm, ex herb. Lugd.-Bat. in herb. Kew. flor.); Kiusiu, circa

BERBERIDEARUM JAPONIE CONSPECTUS. 425

Nagasaki (Thunberg, Oldham! April. 1862, n. 536, flor. et frf.).
In principali insula: ad Yokohamam (Maximowicz ex Franchet
et Savatier ; Oldham! Juli. 1881, n. 217 fructif., Dickins! 8.);
Kioto (coll. * Challenger’ Expedit.! April. et Mai. 1875, flor. in
herb. Kew.); in montibus Nikko et Fudsi-no-yama (Maries! flor.);
cirea Yokosuka (Savatier, n. 48).

Korea: Chemulpo (JF. Charles! n. 14 et 46 flor.) ; ad portum
Hamilton (Oldham! n. 34, Mar. 1863, cum floribus foliisque
juven.).

China (Fortune! n. A 31, 1845, flor.) : Ningpo (Everard); ad
Nin-kan-jou (W. Hancock! 22 April. 1877); Kiu-Kiang (Dre.
Shearer! 1873, flor.). Herb. J. Gay! in schedula sub “ Jardin
des plantes, école, le 28 Avril, 1858, Originaire de Chine,” in
herb. Kew. vidi.

Japonice: Akebi.

2. A. LOBATA, Decaisne, loc. cit. ; Sieb. et Zucc. Fl. Jap. i. 145,
tab. 78; Fam. Nat. n. 356; Mig. Prol. 197; Franch. et Sav.
Enum. i. 21, n. 89; Engler, Bot. Jahrb. vi. 1885, 58.—Mitsuba
Akebi, Iwasaki Tsunemasa, Honzó Dsufu, vol. xxx. fol. 9 verso;
Ito Keiske, Simoku Dsusetsu, Arb. ii. 2, pl. 2.

Hab. in Japonie (Buerger! in sched. sub Rajania trifolia,
Sieb. comm. ex herb. Lugd.-Bat. in herb. Kew.) ins. Kiusiu:
circa Nagasaki (Buerger, Textor); in fruticetis montis Kawara-
yama (Buerger). In principali insula: in monte Hakone (Sie-
bold ipse); circa Kobe (Moseley! ‘Challenger’ Expedit. April.
1875, flor.) ; frequens circa Yokosuka (Savatier, n. 49) ; Simoda
(Small! in herb. Columbia Coll. New York); Yokohama (Dickins!
1864, flor.); Kioto et Maizuru (Döderlein ex Engler). Insula
Yezo, prope Hakodate (Maximowicz, Dre. Albrecht! 1861, flor.).

China: Kiu-Kiang (Dre. Shearer! 1873, specim. stcril.).

Japonice: Mitsuba Akebi, i. e. Akebia trifolia.

Variat formis foliorum :—

a. CLEMATIFOLIOLA.—A. clematifolia, Sieb. et Zucc. Fl. Jap.
i 146; Fam. Nat. n. 357; Mig. Prol. 197; Franch. et Sav.
Enum. i. 21, n. 90.

Hab. in insula Yezo, inde vivam accepit Siebold.

Hoe specimen typicum non vidi sed potest fieri ut sit aut
formam A. lobate aut fortasse tantum forma trifoliolata A. quinate.

f. QUERCIFOLIOLA.—A. quercifolia, Sieb. et. Zuce. Fl. Jap. i.

426 BERBERIDEARUM JAPONI® CONSPECTUS.

146; Fam. Nat. n. 358; Mig. Prol. 197 ; Franch. et Sav. Enum.
i. 21, n. 91.

Hab. in monte Kimpo-zan (1/0 Keiske); Yokohama (Oldham !
Mai. 1863, n. 985, flor.).

BERBERIS, Le

Conspectus specierum.
Racemi 1-4-flori, foliis oblongo-ovatis
v. cuneato-ellipticis integerrimis .. B. Thunbergii, DC.
Racemi multiflori, 2.
2. Foliis integerrimis .......... B. Tschonoskiana, Regel.
Foliis serratis v. setoso-ciliatis, 3.
3. Foliis elliptieis v. cuneato-ellip- :
ticis, spinuloso-serratis ...... B. vulgaris, L., var. japo-
nica, Regel.
Foliis oblongo-lanceolatis v. ob-
longo-ellipticis, dense setoso-
ciliatis, brevipetiolatis ...... B. Sieboldi, Miq.

1. B. TuvwsEnon, DO. Syst. ii. 9; Prodr. i. 106; Regel,
Descript. plant. nov. Turkest. fasc. i. 1873, 21.—B. cretica, Thunb.
Fl. Jap. 146.—B. sinensis, C. Koch, in Ann. Mus. Bot. Lugd.-
Bat. i. 252; Mig. Prol.i.; Franch. et Sav. Enum. i. 22, n. 98,
ii. 272 (non Desf.).—Megi, Its Keiske, Nihon Sanbutsushi, pt.
prov. Yamashiro, ii. fol. 15.—Ko Hebinoborazu, Itō Keiske,
Somoku Dsusetsu, Arb. ii. 6, pl. iv. (fig. var. a typica est).

a. TYPICA, Regel, loc. cit. 20; Gartenflora, 1872, 238.—Foliis
rotundato-obtusis, supra viridibus, subtus albido-glaucis.

Hab. in Japonie (herb. Lugd.-Bat.! in herb. Kew. steril., C.
Wright!, Dickins! Oct. 1881, fructif., hort. Kew. April. 1882,
specimen typicum Bot. Mag. t. 6646! flor.); principali insula :
Yokohama (Mavimowicz! iter secundum, 1862, fructif.) ; Yoko-
suka (Savatier, n. 50) ; Kintoki (J. Bisset! Mai. 1876, n. 19 flor.);
ad pedem m. Fudsi-no-yama (ipse).

B. Maxrwowiroezrr, Regel, loc. cit.; Franch. et Sav. Enum. n.
272.—B. Maximowiczii, Regel, Gartenflora, 1872, 238.— Foliis
aeutis v. rarius obtusis, utrinque viridibus. :

Hab. in principali insula: Yokohama (Maximowicz! iter se-

BERBERIDEARUM JAPONIJE CONSPECTUS. 427

eundum, 1862, flor., Dickins! n. 904, flor. et fructif.); in mon-
tibus Nikko et Fudsi-no-yama (Maries! flor.).
Japoniee: Ko Hebinoborazu, i. e. Berberis minor (It6 Keiske).

2. B. Tscmowoskrawa, Regel! loc. cit. 21; Franch. et Sav.
Enum. ii. 272, n. 2510.

Hab. in principali insula: alpe Nikko legit Tschonoski! 1864
(Maximowez, iter secundum, ex herb. horti bot. Petrop. in herb.
Kew.).

Japonice: Miyama Hebinoborazu, i. e. Berberis montana
(mihi).

Planta in Kiu-Kiang, China, inventa a Maries! n. 206 flor. in
herb. Kew. posita est, hine speciei forma foliorum et aliis cha-
racteres proxime congruunt: Japonica et Sinica inter se dif-
ferunt, quod in hac ramuli sulcati, angulati, spinulosi simpliciter
quod in illa ramuli tereti, striati, spinis nullis. Sed Regel
(J. c. 21) dicit * ramis teretibus, spinis nullis a B. Thunbergii et
B. sinensi facile dignoscitur," que sunt illius note ut hanc
speciem a duabus illis secernat. Id quod fortasse eam esse
nom distinctam probat cui novum nomen B. Mariesii m. dari

cet.

3. B. VULGARIS, L., var. JAPONICA, Regel, loc. cit. 15; Franch.
et Sav. Enum. ii. 275.—B. vulgaris, Thunb. Fl. Jap. 146; Mig.
Prol. 364; Franch. et Sav. Enum. i. 22, n. 95.—Hebi-noborazu,
Ito Keiske, Nihon Sanbutsushi, pt. prov. Yamashiro, i. fol. 18.—
Tsuneno Hebinoborazu, [td Keiske, Somoku Dsusetsu, Arb. ii. 4,
BL. v. A.

Hab. in principalis insule regionibus montanis et humidis
Provv. Yamashiro, Omi, Mino, Shinano, et alibi crescentes (teste
Ito Keiske, Z. c.) ; prov. Nambu legit Tschonoski! (JMazimowicz,
iter secundum, 1865, flor.) ; in alpe Nikko (Savatier, n. 2090). In
insula Yezo (Small! in herb. Columbia Coll. New York).

Japonice: Tsuneno Hebinoborazu, i.e. Berberis vulgaris (Ito

Keiske).
4. B. SrgBorpr, Mig. Prol. 1; Hegel, loc. cit. 18; Franch. et
Sav. Enum. i. 22, n. 94.—Hirohano Hebinoborazu, It Keiske,

Somoku Dsusetsu, Arb. ii. 5, pl. v. B.

Hab. in Japonia (C. Wright! flor.) loco non commemorato, a
bot. indigen. lecta (ex Miquel, l.c.) ; in monte centrali Japoniæ
alt. 27,000 ped. (Maries!); Fudsi-no-yama (4tó Keiske, l. c.,

428 BERBERIDEARUM JAPONIJE CONSPECTUS.

ipse); alpe Nikko (It6 Keiske, l. c., Bisset! n. 1289, flor. in
herb. Kew.) In regionibus montanis insule Yezo (teste Ito
Keiske, l. c.).

Japonice: Hirohano Hebinoborazu, i. e. Berberis latifolia (Ito
Keiske).

Planta Bissetii sine spinis esse videtur, cetere partes ex toto
B. Sieboldi, Miq., congruunt. An planta diversa ?

Species exclusa.

B. JAPONICA, R. Brown, App. Congo, 22; Mig. Prol.2 ; Franch.
et Sav. Enum. i. 29, n. 96; Schlechtd. Bot. Zeit. 1854, 606 ; Sieb.
et Zucc. Fam. Nat. n. 290; C. Koch, loc. cit.—Ilex japonica,
Thunb. Fl. Jap. 79; Icon. Pl. Jap. Decas. 4, fol. 2.—Mahonia
japonica, DC. Syst. ii. 22; Prodr. i. 109.—O. Hebinoborazu, £76
Keiske, Somoku Dsusetsu, Arb. ii. 7, pl. vi, non distincta e
B. nepalensi, Spreng., videtur.

Ex sententia Ito Keiskei (7. c.) hee planta in hortis Japonen-
sium tantum culta, sponte crescentem non invenit. Tamen à
monte Hakusan prov. Kaga originaliter venisse maxime dubium
est. Quam speciem ex ea enumeratione quamquam nomen speci-
fieum B. japonicam gerit excludere necesse est.

Japonia ( Herb. Lugd.-Bat.! folia una in herb. Kew.). Kiusiu,
in hortis Nagasaki (Oldham ! 1862, n. 686 frf.).

Berberis Bealei, Fortune, in Gard. Ohron. 1850, 212; Hook.
Bot, Mag. t. 4852, e China, videtur non differre ab hac specie*.

CAULOPHYLLUM, Micha.

C. THALICTROIDES, Michx. Fl. Amer. i. 205, fol. 21; Mig. Prol.
2; Franch. et Sav. Enum. i. 23, n. 97.—Rui-yo Botan, Somoku
Dsusetsu, vol. vii. fol. 26; Itd Keiske, Nihon Sanbutsushi, pt.
prov. Omi, vol. ii. fol. 3 et 4 (eum iconibus).

Hab. in Japonie (C. Wright! frf.) principali insula: in monte
Ibuki, prov. Omi ; Ono, prov. Shinano; Hagiwara, Suwara et
Kurokawa, Kiso; in montibus Komono et Sakamoto, prov. Ise
(teste 7/0 Keiske, l.c.) ; in monte Nikko, prov. Shimotsuke
(Bisset! Mai. 1877, n. 981, flor.) ; in monte centrale Japoniæ
alt. 27,000 ped. (Maries! flor.. In insula Yezo (Maries!

* Determinationem nominis B. Bealei, cl. Prof. Babingtonio debeo propterea
quod ejus originem a cl. Bakero inquisivit.

BERBERIDEARUM JAPONIJX CONSPECTUS. 429

fructif.) ; circa Hakodate (Dre. Albrecht! 1862, fructif., C. Wright!
in herb. Columbia Coll. New York).

In insula Sachalin: Dui (Glehn! Juni 1861, flor. et fructif.).

Amur (Maximowicz! sub nom. Caulophyllum robustum, Maxim.,
flor. et fructif.).

America boreali (Richardson! fl. et frf., Grevith! flor., Dau-
beny! frf): Lexington, Kentucky (Dre. Robert Peter! April.
1834, flor.) ; Rochester (Domina Thakeray! flor. in herb. Cantab.);
Cunnington (Torrey! Mai. 1833, flor.) ; Charloth, Vermont (C. G.
Pringle! Mai. 1877, fl. et frf.); West Chester, Townsend (Oleg-
horn! frf.); in silvis penitus Kentucky (C. W. Short! fl. April.,
Mai.). Brit. bor. Amer. (Dre. Hooker! Franklin Expedit., flor., in
herb. Cantab.) ; Canada (Domina Dalhousie! flor.).

NaNnINA, Thunb.

N. pomesrica, Thunb. nov. gen. diss. i. 14; Fl. Jap. 147;
Banks, Ic. Kempf. t. 18, 14; DC. Syst. i1. 22; Prodr. i. 109;
C. Koch, loc. cit. 253; Mig. Prol. 2; Bot. Mag. t. 1109; Franch.
et Sav. Enum. i. 23, n. 98.—Nandsjokf, vulgo Natten vel Nandin
Tsikku, Kempf. Amen. 776.—Nanten, Honz6 Dsufu, vol. Ixxxviii.
fol. 16-28 ; It Keiske, Somoku Dsusetsu, Arb. ii. 9, pl. viii. A et B.

Hab. in Japonie (Buerger! flor. comm. ex herb. Lugd.-Bat.
in herb. Kew.) principali insula: circa Yokosuka (Savatier, n. 53,
sub ubi videtur frequens sed verisimiliter introducta, e Franchet
et Savatier); Yokohama (Mazximowicz); Simoda (Wright! in
herb. Columbia Coll. New York). In insula Kiusiu; prope
Nagasaki, prov. Hizen (Oldham! Mai. 1862, n. 334, sub in hor-
tis). In archipel. Riukiu; ins. Oshima (Oldham! Oct. 1861,
n. 334, fructif.). In insula Tsushima freto Corea (C. W. Wilford!
Mai. 1859, flor. n. 876, sub * cult. in hortis °).

China: prope Ningpo (R. Oldham! Juni 1861, n. 72, flor.)! Kiu-
Kiang (Dre. Shearer! 1878, flor.); fl. Pekin (Dre. Bretschneider!
n. 51 bis, fructi£.) ; Shanghai (Dre. A. C. Maingay ! 1861 et 1862,
n. 299, frf.) ; portus Chusan (Cantor! frf.).

India orientalis (ex herb. Griffith! sub ‘culta, flor.) quod
speeimen cum parvis foliis

430 BERBERIDEARUM JAPONIH CONSPECTUS:

EPIMEDIUM, L.
Clavis specierum.
Petala calcarata (Epimedium).... E. macranthum, Morr. et
Petala non calcarata (Aceranthus), [Decne.
2. Flores racemosi, folia conjugata,
foliola reniformia denticulis
sepius obsoletis .......... E. diphyllum, Lodd.
Flores paniculati, folia ternata,
foliola sagittata setoso-ciliata H. sinense, Sieb.

1. E. MACRANTHUM, Morr. et Decne. in Ann. Sc. Nat. 2° série,
ii. 252, tab. 13; Lindl. Bot. Reg. t. 1906; Paston, Bot. Mag. v.
151, cum icone; C. Koch, in Ann. Mus. Lugd.-Bat. i. 253;
Mig. Prol. 2; Franch. et Sav. Enum. i. 23, n. 99; Baker, in
Gard. Chron. xiii. 1880, 683; Moore, in Journ. Linn. Soc. xviii.
1880, 377.—Ikariso, Somoku Dsusetsu, ii. fol. 46, n. 45; Honzo
Dsufu, vol. vi. fol. 11, recto fig. sinistra.

a. TYPICUM.

Hab. in Japonie (Siebold teste C. Koch) principali insula:
prov. Omi, circa lacum Biwa (‘Challenger’ Expedit.! April. et Mai.
1875, flor. in herb. Kew.); in monte Hakone, prov. Sagami
(Savatier, n. 52, teste Franchet et Savatier). In insula Yezo,
circa Hakodate (Dre. Albrecht | 1861-63, flor., Maximowicz! 1861,
flor.).

China boreali: prov. Shing King (John Ross! April. 28, 1876,
n. 174, flor. in herb. Kew.) ; Chiu Kiang et Kiu Kiang (Maries !
flor); Changdien-Quandien (ex Moore).

B. vioLaceuM, Morr. et Dene. (sp. pr.), loc. cit. 354, tab. 12;
Bot. Mag. n. s. xiii. 1840, t. 83731; Bot. Reg. xxvi. 1840, t. 43;
Paxton, Bot. Mag. v. 1837, t. 123; C. Koch, loc. cit. 253; Miq.
Prol. 2; Franch. et Sav. Enum. i. 28, n. 100; Honzð Dsufu,
vol. vi. fol. 10 verso (?) et 11 verso.

Hab. in Japonie (Siebold, teste C. Koch) principali insula:
prov. Shinano legit Tschonoski! flor. (Maaimowicz, iter secundum,
1864, ex herb. horti bot. Petrop. in herb. Kew.).

y.? HUMILE, T. Itō. Planta humilis multo minor quam cetera,
foliis triternatis, rigidis, glabris, floribus albidis.—ZZonzó Dsufu,
vol. vi. fol. 11, fig. dextra (bona). :

Herb. J. Gay! sub “ E. macranthum, Morr. et Decne. Jardin
des plantes " in herb. Kew.

Ab E. pumilo (spece. ex hort. Edward Leeds in herb. Kew.! et

BERBERIDEARUM JAPONIJ CONSPECTUS. 431

Baker, in Gard. Chron., Joc. cit. 683) mihi differe videtur, que
est tantum planta juvenalis E. macranthi, var. (3. violacei. For-
tasse tantum parva forma a. typici.

E. MvsscuraNvuM, Morr. et Decne. loc. cit. 853; Graham in
Bot. Mag. t. 8745 ; Baker! loc. cit. 683; C. Koch, loc. cit. 258 ;
Mig. Prol. 2.—Japonia (It Keiske, fide C. Koch, Ll. c.). Herb.
ex horti bot. Petrop. flor.! in herb. Kew. Spece. sice. in hort.
Kew. culta in herb. Kew. fl. albis.

Ab E. macrantho distincta est bene hzc planta fere, hujus
calcaribus petalis brevioribus, longioribus illius.

Cl. Maximowicz mihi notam sequentem scripsit :—

E. Musschianum.—Hujus duas formas habeo ; aliam albis flori-
bus (veterem) quam in Europa sola coli scio, aliam violaceam a
me in hortis Nagasaki lectam. Ambe hybride videntur; vio-
lacea inter E. macranthum fl. violaceum et E. diphyllum foliis
integris obtusis, calcaribus brevissimis; alba origine et formis
foliorum incerta, magnitudine florum, et longitudine calcarium
valde variabilis.

E. concinnum, Vatke, in Regel, Gartenflora, vol. xxi. 1872,
165, tab. 726; Baker! in Gard. Chron. loc. cit. 683.

Hee planta, quam descriptione ieonibusque in ‘ Gartenflora '
(l. e.) et typico Bakeriano specimine (in Gard. Chron. l. c.) in herb.
Kew. tantum scio, proxima AE. Musschiano videtur. Preterea
immaturum Bakerianum specimen videtur et ejus specimini Æ.
Musschiani est simillimum, sed caulis ad summitatem habet 2 folia
opposita, trifoliata, pedunculum pauciflorum, sepalos pallidos,
petalos purpureos vel rubro-purpureos.

E. RUBRUM, Morr. in Belg. Hort. 1854, 33, tab. 6. figg. 1-6;
Regel, Gartenflora, 1857, 21, et 1862, 311, tab. 373. figg. 4-7;
Index sem. hort. Petrop. 1856, 33; Baker! loc. cit. 620.—E. al-
pinum, L., var. rubrum, Hook. Bot. Mag. 1867, t. 5671; Morren,
in Belg. Hort. vol. xviii. 1868, 199, tab. xiv. n. 2.—E. alpinum,
Thunb. Fl. Jap. 62.—E. macranthum, Morr. et Decne. var. f.
Thunbergianum, Mig. Prol. 2 (in part) ; Franch. et Sav. Enum.
i. 23.—Spec. ex hort. bot. Petrop. flor. in herb. Kew. vidi.

Hee planta florum forma similior Æ. alpino, L., quam E. ma-
crantho, Morr. et Decne., esse mihi videtur, propterea quod illius
speciei varietati in Bot. Mag. t. 5671 jam relata est. Quam
igitur propius E. alpino locandam esse non incredibile videtur.
Neque tamen hujus plante sub distincto capite constituendo»
necessitas scientia nostra imperfecta in presentia declarari potest.

432 BERBERIDEARUM JAPONIE CONSPECTUS.

Nam eam eultam tantum scimus sed sua sponte crescentem non
invenimus. Ulteriore sane studio opus est. Opinio, qua hee
planta hybridus inter E. alpinum et E. macranthum existimatur,
a Morreno proposita et a Bakero secuta, manifeste non est acci-
pienda, si de eo quod Thunbergii specimen typicum est in herb.
Mertensii, sub nom. Æ. alpinum, eadem eum cl. Maximowicz cre-
damus qui mihi minute scripsit sic :—

E. rubrum.—Hee planta est E. alpinum, Thunb. Fl. Jap. 62,
ut cerno specimine in Japonia a Thunberg ipso lecto et in herb.
Mertensii locato quod in hortis botanicis Petropolis est. Mirum
est Upsali herb. Thunbergii plantam deesse. Mertensii speci-
men est florescens ; authenticum etiam de toto, propterea quod a
Thunbergio ipso denominatum est atque glutinatum est in iisdem
Batavis papyris ac cetera ejus collectionis speeimina Upsali.
Sed dimidia tantum est charta, reliquo tortuose abscisso tamquam
si alterum specimen in eadem charta esset. Valde autem inere-
dibile Thunbergii state Yedo esse hybridum inter Æ. alpinum
et E. macranthum, quem Bakerus E. rubrum esse dicit! Quam
opinionem Bakerus a Morreno (Belgique horticole, 1854, p. 33,
eum tabula) accepit, sed Morrenus Æ. rubrum ex aliorum sen-
tentia esse introductum a Sieboldio, aliorum esse hybridum inter
E. macranthum et E. alpinum dixit et quesitionem non decretam
linquit. Manifestum est validius esse testimonium pro J aponica
origione : evanuit opinio que eain plantam hybridain nominat.

2. E. DIPHYLLUM, Lodd. Bot. Cab. xix. 1832, n. 1858; Graham,
in Bot. Mag. ix. 1835, t. 3448; Baker, loc. cit. 683.—E. japoni-
cum, Sieb. herb. in Mig. Prol. 3.—Aceranthus diphyllus, Morr.
et Decne. loc. cit. 349, tab. 14; C. Koch, loc. cit. 253; Mig. Prol.
2; Franch. et Sav. Enum. i. 24, n. 108.— Baikwa Ikari, S0mo-
ku Dsusetsu, ii. fol. 47, n. 46; Honzó Dsufu, vol. vi. fol. 12
recto.

Hab. in Japonie (Itō Keiske, Siebold, teste Miquel ; herb.
Lugd.-Bat.! in herb. Kew. flor. insula Kiusiu: prov. Higo, 1m
urbe Kumamoto et in monte Kipon-san (Maximowicz ! iter se-
cundum, 1863, ex herb. horti bot. Petrop. in herb. Kew., nec
non teste Honzo Dsufu, l.c.). Herb. J. Gay! flor. in herb. Kew.

Japonice: Baikwa lkariso, i.e. Epimedium pruniflorum (P.
Mume, Sieb. et Zuce.).

3. E. SINENSE, Siebold herb. in Mig. Prol. 2; Hance, Journ.
Bot. 1882, 2.—E. Ikariso, Regel, Index sem. hort. Petrop. 1868,
89; Franch. et Sav. Enum. i. 24, n. 101.—E. sagittatum, Mazim.

BERBERIDEARUM JAPONIJE CONSPECTUS. 433

Mél. Biol. 1876, ix. 713; Baker, loc. cit. 683.—Aceranthus
sagittatus, Sieb. et Zucc. Fl. Jap., Fam. Nat. n. 296; C. Koch,
loc. cit. 253; Miq. Prol. 2; Franch. et Sav. Enum. i. 24, n. 102.—
A. triphyllus e£ macrophyllus, C. Koch, loc. cit. ; Sómoku Dsusetsu,
ii. fol. 45, n. 44; Honzó Dsufu, vol. vi. fol. 10 recto.

Hab. iu Japonia (herb. Lugd.-Bat.! in herb. Kew. pl. steril.).
In Japonensium hortos cultos e China introducta, etiam in prov.
Koshiu sponte crescere dicitur (teste Somoku Dsusetsu, Z. c.).

China: Kiu Kiang (Dre. Shearer ! 1873) ; juxta urbem I-chang,
prov. Hu-peh, legit 7. Watters (ex Hance).

Japonice: Hozakino Ikariso, i. e. Epimedium paniculatum,

E specimine Japonico, loco spec. non indicato (flor.!), quod
ante oculos mihi est, sequentes characteres notavi:—

Foliis setoso-ciliatis, subtus nervis prominentibus radicalibus
ternatis; caule erecto simplici, foliis oppositis cum folio juvenali
medio inter vetera exorto, foliis ternatis longo petiolatis, petiolis
glabris ad segmenta fuscis, leviter striato-angulatis v. canalicu-
latis; foliolis obovatis v. ellipticis, acuminatis v. acutis, basi
foliorum inequaliter sagittatis v. sagittato-hastatis, apicibus lobium
acutis v. acuminatis rarius submucronato-obtusis ; racemo com-
posito gracili multiflori glabro leviter striato, petiolis gracilissimis
1—3-floris ad basin bracteatis, bracteis linearibus ; sepalis ovato-
ellipticis, obtusis; ovario subinequaliter elliptico, stylo lon-
giore.

Diversum videtur a proxime affini E. pubescente, Maxim., quod
non vidi, sed cognovi descriptione iconibusque cl. Maximowiczii,
foliis hastatis subtus non crebre pilosis sed glabris, flore majore,
sepalis non acuminatis sed obtusis, et pedicellis glabris non
glanduloso-pilosis.

E cl. Bakeri sententia (Gard. Chron. xiii. 1880, 683), Æ.
Ikariso, Regel, Index seminum hort. Petrop. 1868, 89, est
hybridus inter E. alpinum et E. Musschianum, sed e Regeli
descriptione, foliola e basi sagittato-cordata v. inequaliter sagittata
oblongo-elliptica, a qua mihi, ut cl. Maximowiez in Mélanges
Biologiques, vol. xi. 713, demonstravit, synonymus Æ. sinensi
manifeste videtur.

E. longifolium, Decaisne, in Revue Horticole, 1854, 285,
mihi ignotum.

DırAYLLEIA, Miche.

D. Gmaxr, Fr. Schmidt, Fl. Sachal. 109; Mig. Prol. 364;
Franch. et Sav. Enum. i. 25, n. 104.—D. cymosa, A. Gray, Bot.

LINN. JOURN.—-BOTANY, VOL. XXII. 2N

434 BERBERIDEARUM JAPONIJE CONSPECTUS.

Jap. 380 (non Michz.).—Sanka Yo, Honzd Dsufu, vol. xxiii. fol.
5 verso; Somoku Dsusetsu, vii. fol. 25; Ito Keiske, Nihon Sanbut-
sushi, pt. prov. Omi, ii. fol. 22, cum icone.

Hab. in Japonie (C. Wright! frf.) principali insula: monte
Togakushi (7:0 Yudsuru!); Tengu-lwa in monte Ibuki, prov.
Omi; in pagis Hagiwara-Mura et Naka-Mura, Kiso; prov.
Etsigo; in monte Hakusan, prov. Kaga; in montanis Nikko,
prov. Shimodsuke (teste [16 Keiske, l.c.) In insula Yezo: ad
promontorium Soya (Small); prope Hakodate (Maximowicz).

Tum in Sachalin (Angustinowicz! flor.).

Japonice: San-ka-yo.

Ab affini D. cymosa, Michx., fructibus majoribus polyspermis,
seminibus pyriformibus rectis, cyma pubescente facile distin-
guitur.

D. cymosa, Michx., hab. in America boreali: in locis humidis
sylvarum region. med. mont. Smoky, Tennessee (Agel! flor.
Aprili et fructif. Aug. 1884, spp. 2); in montibus altioribus
Virginie et Caroline septentrionalis (A. Gray et J. Carey !
Julio 1841, spp. 3, frf.).

Poporuytium, L.

1. P. savontcum, T. It in Maxim. Mél. Biol. t. xii. 1886, 417.

Hab. in Japonie principali insula: monte Togakushi (/¢0
Fudsuru !).

Japonice: Togakushi So, 7. e. Planta e Togakushi (mihi).

Hee planta, eujus folium cum florum figuris ad Maximo wiez ,
notissimum Academicum Petropolis, specimen misi, ab illo in
diagnose plantarum asiaticarum nuper descripta est. Hee
tamen est manifeste nova species, cui cetere PodopAylli species
de toto dissimiles sunt ternatis foliis. Mihi autem a speciminum
paucitate positio generis adhuc dubia videtur. Fructus tamen

nunquam vidi, sed cl. Tanaka vidisse baccas nigro-esruleas memi-
nisse mihi dixit.

Species non satis nota.
2. P. PELTATUM, L. ex Maxim. loc. cit. 418; Honzoó Dsufu,
vol. xxiii. fol. 7 recto ?
“ Nippon in alpibus " (Tanaka, japonice Sankayo).
“Summitas speciminis florentis missa dubia est propterea quod
stamina tantum 6. Cetera: foliorum indoles, floris inserto,

BERBERIDEARUM JAPONIJE CONSPECTUS. 435

stamina, stigma fungosum— omnino identica ; pubes tamen paginz
folii inferioris ad basin nervorum parcissima.” (Maximowicz.)

Cl. Maximowicz adumbrationem hujus plantz, cui sunt duo
folia, sua maru ex unico specimine in Japonia lecto recte de-
scriptam, ad me benigne misit. Que adumbratio mihi non diversa
a P. peltato videtur. Eandem sententiam cl. Maximowiez in
epistola ad me scripta extulit, sed dubitat utrum specimen culta
planta Americana, necne sit. Nec P. peltatum usquam in Japonia
inventum esse audivi, nec de eo quicquam dicere possum. Atta-
men plantam in Honzo Dsufu (Joe. cit.), delineatam pro affinitatem
Diphylleie, huic esse proxime similem videri animadverto, qua, si
esse eandem sumamus, indigena Japoniæ existimanda est, quam-
quam auctor Honzo Dsufu loeum speciale non memorat.

Acurxs, DC.

A. TRIPHYLLA, DC. Syst. ii. 35; Prodr. 1. 112; G. Don, Syst.
i. 1831, 120; Hook. Fl. Bor.-Am. i. 1833, 30, t. 129; Torr. et
Gray, Fl. N. Am. i. 53; Torr. Bot. U. S. Explor. Exped. 219 ;
Newberry, Pac. R. Rep. vi. 66; Cooper, id. xii. 56; Boland. Cat.
4; Gray, Proc. Am. Acad. viii. 376; Watson, Bibl. Index N.
Am. Bot. 83; Brewer et Watson, Bot. Calif. i. 16.—Leontice
triphylla, Smith, Rees’s Cycl. n. 5; Schult. Syst. vii. 21; Dietr.
Syn. ii. 1073.

&. TYPICA.
Hab. in America boreali occ. (Douglas! 1829 flor., Scoulter!

flor.) : California!; Oregon (W. Lobb! n. 313 flor., Elihu Hall!
1871, n. 21, frf.); fl. Lower Frazer, 49° lat. bor. (Dre. Lyell!
1859 in schedula sub ‘in silvis, fin. April. flor); New Georgia
(Menzies! herb. Brownianum in herb. Kew. flor); Columbia
(Dre. Gaiden! April. 1834, n. 32, in schedula sub ‘ shallow wood,’
flor. et frf.); monte Tacoma, in territ. Washington (J. Bryce!
comm. A. Gray in herb. Kew.); in insula Vancouver (Dre. Lyell!
1858 flor., Dre. C. B. Wood! 1862, frf.).

f. sapontca, T. It0.—A. japonica, Mazxim.! Mél. Biol. t. vi.
960.—Planta minor quam in typica; foliola minora, terminalis
apicis leviter triloba cetera integerrima.

Hab. in Japonie principali insula: in silvis subalpinis prov.
Nambu legit Tschonoski! 1855 subdeflor. ex herb. Maximowicz,

iter seeundum.
Japonice: Nambu So, i. e. planta e Nambu (mihi).

436 BERBERIDEARUM JAPONIE CONSPECTUS.

A. triphylla foliis trifoliatis valde variabilia est, aut parvis aut
magnis et bene expansis, aut sinuato-dentatis, aut remote lobatis,
aut spicis, aut densis aut interruptis. Planta ex insula Vancouver,
a Dre. C. B. Wood lecta, illi Japoniez parvitate est similis ; simi-
liores etiam e Nova Georgia et California lect» sunt, ita ut,
quod ad Californie plantas fere nequeamus discernere quo
Americane et Japonice inter se distincta sint.

Cl. Maximowiez, postquam descriptionem superiorem vidit,
mihi epistolam sequentem scripsit :—

Achlys japonica : Quam eam descripsi, specimina tantum, quibus
eam conferrem, ex insula Vancouver a Dre. Lyell lecta habui.
Neque alia adhue recepta esse video. Itaque nihil quicquam
novum de planta dicere possum. Fortasse te recta dixisse potest.
Fructus tamen, quoad scio, adhuc incognitus est et forsitan dis-
tinctiones exhibet.

Cl. Prof. Oliver, qui meam opinionem de bae planta valde
credibilem esse dixit, non modo auxilium benignum de hujus
plantz determinatione sed etiam multa alia officia debeo.

Postscriptum.—Postquam meum “ Berberidacezm Japonies "
paravi, monographia “Sur les espèces du genre Epimedium ”
cl. Franchet auctore in ‘Bulletin de la Société Botanique de
France,’ tome viii. n. 2, Mai 1886, edita est, que mihi persuasit,
ut observationes paucas addam. Quantum ad Epimedium viola-
ceum, Morr. et Decne., attinet, cl. Franchet in unam et eandem
conclusionem mecum pervenit. Namque ille quoque hane plantam
in varietatem tantum F. macranthi, Morr. et Decne., esse redu-
cendam statuit. Quamquam antea plantam sub nomine E.
macranthum, Morr. et Decne., var. fl. violaceum, T. Ito, reducere
proposueram, libenter cl. Franchet auctoritati jure prioritatis
jubente concedam.

E. rubrum, Morr., cl. Franchet (loc. cit. 107), sententiam ab
Hooker (Bot. Mag. 1867) jam propositam habet, qui plantam
esse varietatem E. alpini, L., viz. E. alpinum, L., var. rubrum,
Hook., existimavit. Nihil tamen de hae planta dicere potero;
donee evidentiam de origine occasionis habuerimus.

Cl. Maximowiez ad me nuper scripsit quod specimen (sterile)
a Miquel dictum Thunbergii Z. alpinum, quod Miquel E. ma-
cranthum, M orr. et Decne., var. B. Thunbergianum, Miq. (Prol. 2),
adduxit, re vera esse E. rubrum, Morr. Ceterum Miquel flores

ON APOSPORY IN POLYSTICHUM ANGULARE. 437

et Thunbergii E. alpini et E. macranthi unos et eosdem descripsit,
quamobrem Miquelianum specimen (non Thunbergianum) Æ. al-
pini (ut nominatur) esse re vera E. macranthi, Morr. et Decne.,
est sumendum.

EXPLICATIO TABUL/E XXI.
Achlys triphylla, DC., var. japonica, T. Ito.

Fig. 1. Specimen subdeflorens, mag. nat.
2. Spice pars cum carpellis immaturis, mag. auct.
3. Flos subdefloratus cum 3 staminibus persistentibus, mag. auct.
4. Stamen, mag. auct.

On a new Instance of Apospory in Polystichum ı gigulare, var.
pulcherrimum, Wills. By Cuartes T. Drvery, F.L.5.

[Read 16th December, 1886. ]

In December 1884, and previously, I brought before the notice
of this Society the results of certain investigations into a mode
of reproduction in Athyrium Filix-femina, var. clarissima,
hitherto unobserved, and to which Prof. F. O. Bower gave the
name of apospory, the spore being eliminated from the life-cycle
of the fern, the prothalli springing direct from the stalks of
aborted sporangia. This discovery in the variety of Athyrium in
question was followed by Mr. G. B. Wollaston's discovery of a
kindred phenomenon in an abnormal form of Polystichum angu-
lare, var. pulcherrimum, Padley, in which plant, however, it is
developed in an altogether different manner, the prothalli being
produced in this case by simple extension and dilatation of the
growing tips of the pinnules, so that not only the spore, but the
entire sorus is excluded from the normal cycle of existence.

The particular specimen of Polystichum angulare in which this
phenomenon was remarked was found many years previously,
growing wild in North Devon, since which time it had constantly
produced its characteristic expansions of the pinnule-tips, which,
however, were regarded at first as a minute form of that cresting
common to so many abnormal forms of Ferns, and consequently
their true character as protballi proper remained for a long time
unsuspected.

One of the peculiar characteristics of this Fern is a marked
elongation of all the lower divisions ofthe pinna, 7. e. piunules, the
upper ones retaining nearly or quite their normal outline ; and

438 MR. €. T. DRUERY ON A NEW INSTANCE

as it was mainly upon these attenuated divisions that the pro-
thalli were formed, the points running out into threads at the
ends of which the prothalli appeared, it struck me that as these
attenuated pinne were not confined to this individual Polysti-
chum, others presenting the same peculiarity might develop the
rest of the phenomena under suitable treatment. I consequently
applied to Colonel Jones, of Clifton, for whose aid throughout
these investigations I cannot too strongly express my thanks, and
was duly furnished with a young plant of the form known as
P. angulare var. pulcherrimum, Wills, found by the late Dr. Wills
in Dorsetshire some years ago. This form, which, it will be
noted, was found some seventy or eighty miles distant from
Mr. Padley’s station, moreover differs markedly in general cha-
racter, so that no doubt can reasonably be entertained of its
distinct origin ; the characteristic prothalloid extensions had also
never been remarked upon it.

On receipt of the plant early in the present year (1886), I at
once placed it in a close Todea frame, with the result that as soon
as the fronds attained the length of some 6 inches or so, I was
delighted to find the tips of the pinnules running out and dilating
into prothalli in profusion, until the pinne were absolutely fringed
with them, precisely in the same manner as in Padley’s form.
Later in the season, however, the close treatment disagreed
with the plant, and on removal from the frame to 4 cool
fernery, the prothalli dried up and perished; fortunately, how-
ever, not before Prof. Bower was able to inspect the plant and
obtain material for further investigation. It seems highly pro-
bable from the rapid disappearance of the prothalli, in this case
under nearly normal conditions, that they would not be deve-
loped under ordinary circumstances to an extent likely to attract
observation.

So far the phenomena observed had been precisely similar
to those noticed in Padley's form ; but upon further close exami-
nation of the fronds I was struck by the existence of several
hydraform bodies, clearly visible to the naked eye, which
appeared to be attached to the upper surface of the pin
nules well within the margin, and in this respect, as well as
in their shape, entirely different from the prothalliform point-
extensions.

On examining several of these under the microscope, I found
their distinct character emphasized by the fact that they were 1n

OF APOSPORY IN POLYSTICHUM ANGULARE. 439

every case produced at the ends of excurrent veinlets pro-
truding from the surface of the pinnules, well within the margin,
and thickening at the distance of about -4 of an inch into a pear-
shaped body, from which radiated in all directions numerous root-
like hairs. The brown and shrivelled condition of these bodies
led me to the belief that it was essential to their full develop-
ment, whatever that might be, that they should at an early stage
find some suitable nidus whence the hairs, assuming them to be
root-hairs, might obtain the needful elements for growth. After
a long search I was fortunate enough to find one of these bodies
in a young state, the pear-shaped head only being formed, and
the hairs just beginning to develop. I at once removed the
plant into a large shallow pan and pegged down the frond so that
the excrescence was brought into very close proximity to the soil.

Fig. 1. Hydraform excrescence, nat. size, in Polystichum angulare, var. pul-
cherrimum, Wills.—Fig. 2. The same, magnified.—Fig. 3. Further development,
showing radiating filaments.—Fig. 4. When assuming prothalloid form, nat. size.
—Fig. 5. The same enlarged, showing incipient sinus (a) and developed roots.

This was on September 5th, when it presented the appearance
shown in figs. 1 & 2. On September 10th I observed that the rovt-
hairs had much developed, radiating stiffly in all directions with-
out apy apparent bias towards the soil (fig. 3). On October 2nd,
after a temporary absence, I found that some water had been
spilt over the object under examination, partly burying it in soil;
on removing which I found that it had sent several long threads
from the lower end into the soil, and that the upper ones had
either aborted or been broken off, as nothing but short projections
remained. At this date the shape had materially altered in the
prothallus direction, having flattened out laterally in the same
plane as the soil, and formed an incipient but unmistakable

410 ON APOSPORY IN POLYSTICHUM ANGULARE.

sinus (figs.4 & 5). Since that date growth has continued, though
slowly, with the result that itis now an undoubted prothallus,
though much thicker in substance than those produced by exten-
sion of the pinnule-tips.

From these observations it will be seen that the formation of
the prothallus in this case is preceded by a very different series
of phenomena from those already recorded. In the one case the
prothalli are simple extensions of the cellular substance of the tips
of the pinnules, commencing at points quite beyond the venation,
and, so far as I have observed, produce no root-hairs unless
brought into contact with the soil. In the other case, however,
the prothallus is a direct outgrowth of the tip of a veinlet pro-
truded through the upper surface of the pinnule, and at once
produces root-hairs in abundance long before it assumes any
other characteristic of a prothallus; and, finally, the resulting
prothallus is much thicker in substance. Whether archegonia
or antheridia are present upon the solitary specimen I have had
under examination, I cannot say, as the disturbance necessary
for microscopic scrutiny would probably have been more destruc-
tive than profitable.

These excrescences, it will be observed, bear several striking
resemblances to the pear-shaped pseudobulbils first observed in
Athyrium Filiz-fomina, var. clarissima. Their size, shape, and the
production at an early stage of radiating root-hairs are peculiarly
akin, as also the dilatation of the free end of the body into a
prothallus; on the other hand, they are differentiated by their
isolation and independent position with regard to the sorus.

I will conclude my remarks by pointing out that in all three cases
of apospory now observed there appears correlated with the pheno-
menon a marked elongation of the ultimate divisions of the frond.
Athyrium Filiz-femina, var. clarissima, is very remarkable for
this character; and on the two species of Polystichum the lower
pinne are so characterized, while the upper pinne, which nearly
approach the normal form, are comparatively barren of prothalli.
I mention this as it may form a clue to the discovery of the phe-
nomenon in other Ferns which, like the one under special notice,
may only require close treatment to yield similar results. Iam
therefore sanguine that other workers, in not only British, but
exotie Ferns, may thus light upon new instances of the pheno-
menon in question, and so increase the opportunities of com-
pletely studying it.

MR. J. G. BAKER ON THE FLORA OF MADAGASCAR. 441

Further Contributions tothe Flora of Madagascar.
By J. G. Baxtr, F.R.S., F.L.S.

[Read 18th November, 1886.]

Styce I last reported to the Linnean Society on the Flora of
Madagascar, in the Session of 1884-85, two large boxes of plants
have been received from our indefatigable correspondent, the
Rev. R. Baron, F.L.S. One of these, owing to the unsettled
state of political affairs in the island, did not reach us for more
than a year after the letter announcing it came to hand, so that
we had quite given it up for lost. These two boxes carry up
his collecting-numbers to nearly 5000. The present paper con-
tains descriptions of the principal new genera and species in-
cluded in these boxes of which the specimens were sufficiently
complete. There appear to be seven new genera—one in Meni-
Sspermaees, one in Geraniacew, two in Melastomaces, one in
Rubiaceæ, and two in Composite. The species are distributed
through the primary divisions as follows, viz. :—Thalamiflore
51, Calyciflore 48, Gamopetalz 93, Incomplete 31, Monocotyle-
dons 27, Vascular Cryptogamia 5.

As before, the great bulk of the new species belong to the
large, well-known, widely-spread tropical genera, such as Garcinia,
Hibiscus, Begonia, Vernonia, Vitis, Ficus, Piper, and Cyperus.
Of characteristically Cape types we have Pelargonium, Stoebe,
Belmontia, and Cineraria added to the flora of Madagascar, a
second species of Crassula, several flesh y-leaved Kleinioid Senecios,
and a curious dwarf Aloe, allied to the Cape Aloe aristata of
Haworth. Of familiar European genera we have Celtis and
Deyeuxia added and new species of Nasturtium and Ajuga. Of
the endemic Madagascar genera we get new species of Asteropeia,
Rhodolena, Dichetanthera, Veprecella, Gravesia, Kitchingia,
Dicoryphe, Oncostemum, Mascarenhaisia, and Dypsis. There is
nothing materially fresh affecting the relations of Madagascar
to Tropical Africa, Mauritius, and Bourbon. The feeble affinity
of the Madagascar to the Indian and Malayan flora is strength-
ened by the discovery of the genus Cyclea and of new species of
Alysia, Didymocarpus, and Strobilanthes. Of types of certain
or possible economic interest we have species of Dalbergia,
Macaranga, and Strychnos, two species of Commiphora ( Balsamo-
dendron) and four of Garcinia. A curious Euphorbiaceous

LINN. JOURN.—BOTANY, VOL. XXII. 20

442 MR. J. G. BAKER ON THE

plant seems to belong to the American genus Pedilanthus, but |
the material is incomplete.

For convenience of reference I give a list of the recent papers
whieh have appeared in English periodicals, in whieh new or im-
perfectly known plants from Madagascar have been described.

Genera. Species.
1876. Ferns collected by Mr. and Mrs. William Pool.
Baker, Journ. Linn. Soc. xv. p. 411 ............... € 26
1877. Ferns collected by Miss Helen Gilpin. Baker,
Journ. Linn S00. xvi. Pp DUE eis e e: 21
1879. Flowering Plants collected by L. Kitching. Baker,
Journ. Linn. Soc. xviii. p. 264 ........ IAR acd 2 33

1880. Ferns collected by L. Kitching. Baker, Journ.
Bot 1880: aston cent este) de. 13
1880-85. Endemic Genera figured in Icones Plantarum.
Micronychia, Oliver, t. 1337; Epallage, DC.
t. 1394; Rhaphispermum, Beuth. t. 1402; Car-
diochlamys, Oliv. t. 1403; | Bembicia, Oliv. :
t. 1404 ; Xerochlamys, Baker, t. 1413; new ... 3 3
1882. New Plants collected by Messrs. Baron and Parker.
Baker, Journ. Bot. pp. 17, 45, 67, 109, 137, 189,
218, 243, 266 ; reprint paged differently 2 108
New Plants collected by Messrs. Baron and Parker.
Baker (with synopsis of Cyperus by C. B. Clarke),

Journ. Linn. Soc. xx. pp. 87-304 .................. 5 392
1883. New Monocotyledons collected by Deans Cowan.
Ridley, Journ. Linn. Soc. xx. p. 829 ............ 1 10

1884. On Zyalocalyr, a new genus of Turneracee.

Rolfe, Journ. Linn. Soc. xxi. p. 256, t. 10
New Piants collected by Baron. Baker, Journ.

Linh. Soo: xui. pp. 317, 407 o aeaa 8 190
Cyperacez nove. Ridley, Journ. Bot. p. 13......
Ferns collected by Humblot. Baker, Journ. Bot.

EMS o SESS s 15

1885. Complete list of all the known Orchids of Mada-
gascar, with descriptions of new and little-known

species. Ridley, Journ. Linn. Soc. xxi. p. 456. 74

New Orchids collected by Fox and Baron. Ridley,
Journ. Jann: Soc xxi p. H6 ues voee E 19
1858. The present paper iiie Suites 7 255
29 1163

THALAMIFLORÆ.
POPOWIA MICRANTHA, n. sp.

Arborea, glabra, ramulis gracilibus purpurascentibus copiose lenticellatis,
foliis oblongis petiolatis, floribus axillaribus solitariis longe pedicellatis,
sepalis parvis orbicularibus, petalis 6, exterioribus orbicularibus, interio-

ribus oblongis, staminibus carpellis zequilongis, carpellis 8-9 ovoideis
brunneo-sericeis.

FLORA OF MADAGASCAR. 413

A tree, glabrous in all its parts except the carpels, with
copious, very slender branchlets. Leaves alternate, distinctly
petioled, 12-2 in. long, moderately firm in texture, green on the
upper surface, glaucous beneath, with fine inconspicuous veins.
Flowers copious, solitary, axillary, on very slender spreading
pedicels, sometimes an inch long. Calyx persistent, din. in diam.
Petals deciduous, outer 3 in. long and broad, inner about half as
broad. Carpels very small in the flowering stage, densely silky.
—Baron 4773! Popowia is united to Unona by Baillon. There
are several Madagascar species already known.

CYCLEA MADAGASCARIENSIS, n. sp.

Sarmentosa, glaberrima, caulibus gracilibus angulatis, foliis ovatis
acuminatis subcoriaceis nitidis venulis ad faciem inferiorem omnibus ex-
sculptis, petiolo supra basin insertis, floribus masculis in paniculas parcas
axillares dispositis, pedicellis flore longioribus, sepalis 6 oblongis basi
solum coalitis, petalis in corollam globosam coalitis ore crenato, columna
staminea apice peltata corolla zequilonga.

A shrub, with slender angled twining stems. Leaves glossy,
firm in texture, 2-3 in. long, green on both sides, with the petiole
inserted a short distance above the truncate or slightly emargi-
nate base, with 5 veins radiating from its insertion upwards and
2 horizontally, all the veins and veinlets raised on the under
surface. Flowers in narrow peduncled axillary panicles, as long
as or shorter than the leaves; pedicels articulated at the apex,
1-1 line long. Bud globose, blackish, not more than half a line
in diameter. Corolla shorter than the calyx.—Baron 3766!
Adds this well-known Indian and Malayan genus to the Mada-

gascar flora.
GamopoDa, genus novum Menispermacearum.

Dioieum. Flores masculi. Calyx polyphyllus, sepalis 6 oblanceolatis
obtusis imbricatis extus dense pilosis. Petala 6 oblonga glabra concava,
calyce 2-3plo breviora. Stamina 3-4, filamentis deorsum connatis apice
liberis, antheris terminalibus adnatis tetragonis bilocularibus. Flores
feminei ignoti.

G. nENSIFLORA, Baker.

Species sola.

An erect shrub, with slender terete woody branchlets, clothed
with dense short persistent brown pubescence. Leaves alter-
nate, shortly petioled, ovate-lanceolate, entire, acute, 3-5 in.

long, 1-2 in. broad below the middle, rigidly coriaceous, green
202

444 MR. J. G. BAKER ON THE

and glabrous above, clothed with deciduous brown hairs on a
glaucous groundwork beneath, all the main veins and anastomos-
ing veinlets raised. Flowers many, arranged in dense sessile or
shortly peduncled congested cymes in the axils of the leaves.
Calyx +; in. long, densely clothed with brown-drab hairs on the
outside. Petals brownish, at most jin. long. Stamens about
as long as the petals, the square anthers entirely adnate to the
tip of the filaments.— Baron 2927! Allied to Triclisia, Benth.,
from which it differs by its adnate anthers and flower-wrapper
divided out into a distinct calyx and corolla.

Nasturtium MILLEFOLIUM, n. sp.

Annuum, caulibus dense czspitosis brevibus eiespitosis humifusis, foliis
petiolatis oblongis decompositis segmentis minutis crassis oblongis, flori-
bus axillaribus solitariis breviter pedicellatis, sepalis parvis oblongis, petalis
obovatis unguiculatis albis calyce duplo longioribus, staminibus calyci
sequilongis, fructu oblongo-lanceolato stylo brevi apiculato.

An annual, with a fusiform root, with many short stems a few
inches long radiating from its crown. Leaves 1-3 ir. long, de-
eompound like those of Sisymbrium Sophia, with very small
ultimate segments. Flowers solitary from the axils of the leaves
all down the stem ; pedicels short, erecto-patent. Calyx jin.
long. Petals 2 in. long. Pod 1-1 in. long, tipped with a short
persistent style.— Baron 4428 ! Antananarivo, Hildebrandt 4056,
ex parte.

APHLOIA MINIMA, n. sp.

Fruticosa, ramosissima, glabra, foliis parvis oblongis breviter petiolatis
supra basin argute serratis, floribus solitariis terminalibus breviter pedi-
cellatis, sepalis orbicularibus, staminibus demum calyci zequilongis, ovario
ampullzformi stigmate peltato.

A very much branched small shrub, glabrous in all its parts.
Leaves alternate, 4—4 in. long, moderately firm in texture, green
and glabrous on both surfaces, obtuse, cuneate and entire at the
base. Flowers solitary, terminal; pedicels i-lin.long. Budjin.
in diam. ; sepals 5, orbieular, the two outer thicker and darker in
colour than the three inner. Stamens very numerous. Ovary
glabrous; stigma large, capitate.— Baron 4514!

PITTOSPORUM PACHYPHYLLUM, n. sp.

Fruticosum, glabrum, ramulis validis,
vel verticillatis oblanceolatis Obtusis cras
basin angustatis, umbellis paucifloris

foliis breviter petiolatis sparsis
sissimis rigidissimis e medio ad
sessilibus terminalibus, pedicellis

FLORA OF MADAGASCAR. 44.5

fructui zequilongis, gynophoro nullo, valvis 2 percrassis rigidissimis divari-
catis, ovulis ad placentam singulam 8-10, seminibus globosis castaneis.
An erect shrub, glabrous in all its parts. Leaves 2-3 in. long,
#-1 in. broad above the middle, very thick and rigid in texture,
green above, drab beneath, the veins fine and immersed ; petiole
4-3 in. long. Umbels sessile at the end of the branchlets,
bracteated by full-sized leaves. Flowers not seen. Fruit-pedicels
3-2 in. long. Valves of the fruit orbicular, under 3 in. long, very
thick and rigid, brown on the outside.— Baron 1174! 3950!

PrrTOSPORUM VERNICOSUM, n. sp.

Fruticosum, ramosissimum, glabrum, foliis irregulariter confertis brevi-
ter petiolatis oblongis vel lanceolatis rigidulis utrinque viridibus facie
nitidis, umbellis multifloris terminalibus, pedicellis fructu longioribus,
fructu globoso glabro brunneo magnitudine pisi valvis 2 patulis, gyno-
phoro brevi, seminibus 2 castaneis nitidis.

An erect shrub, glabrous in all its parts, with irregularly
crowded leaves. Leaves 14-2 in. long, about 4 in. broad at the
middle, acuminate, glossy green above, paler green beneath, with
fine immersed veining. Flowers as many as 12, arranged in
sessile terminal umbels, with many full-sized leaves at and near
the base; fruit-pedicels i-j in. long. Fruit globose, brown,
coriaceous, lin. long and broad, tipped with a short slender
style, the rigid valves finally spreading horizontally. Seeds 2,
subglobose, mahogany-brown, arising one each from the base of
the two placentas. Flowers not seen.— Baron 4942 !

POLYGALA LEPTOCAULIS, n. sp.

Annua, caule erecto gracillimo pubescente virgato simplici vel furcato,
folis subsessilibus glabris linearibus vel lanceolatis, racemo terminali
elongato laxissimo, pedicellis brevibus inferioribus cernuis, sepalis pallidis
viridi venosis, exterioribus parvis oblongis, interioribus magnis obovatis
venis sursum anastomosantibus, petalis czeruleis calyce paulo longioribus,
carina apice parce cristata, fructu oblongo calyci «quilongo.

An annual, with slender erect finely pubescent stems 3-12 ft.
long. Leaves extending all the way up the stem to the base of
the raceme, laxly disposed, erecto-patent, l-nerved, 1-14 in.
long. Raceme 2-5 in. long; pedicels much shorter than the
calyx. Outer sepals 4j; in. long; inner above å in. long, with
the 3 green veins connected upwards with cross-veinlets. Upper
petals 4 in. long; keel rather longer, with a small crest at the
tip of the midrib. Stamens as long as the keel-petal. Capsule
oblong, compressed, $-4 in. long, emarginate at the apex.— Baron
4548! 4590! 4598! Allied to P. hyssopifolia, Bojer.

446 MR. J. G. BAKER ON 1HE

GARCINIA CERNUA, n. 8p-

Arborea, glabra, foliis brevissime petiolatis subcoriaceis obovato-oblongis
cuspidatis basi deltoideis venis subtilibus inconspicuis, floribus foemineis
ad foliorum axillas brevissime pedicellatis cernuis, sepalis 4 orbicularibus
parvis, ovario ampullzformi, stigmate tenui plano peltato lobis 8 orbicu-
laribus patulis.

A tree, glabrous in all its parts. Leaves 4-5 in. long, about 2
in. broad, moderately firm in texture, with 20-25 indistinct veins
on each side of the midrib. Flowers solitary, drooping. Calyx
$ in. long; sepals 4, decussate, suborbicular. Petals and stamens
not seen. Ovary narrowed gradually into a long neck. Stigma
} in. in diam., thin, flat, with 8 orbicular spreading lobes.—
Baron 2653!

GARCINIA ORTHOCLADA, n. sp.

Arborea, glabra, foliis brevissime petiolatis rigide coriaceis oblongis
obtusis basi deltoideis venis faciei superioris exsculptis faciei inferioris
immersis, floribus foemineis axillaribus et terminalibus breviter pedicellatis,
ovario globoso, stigmate crasso peltato 4-lobato lobis recurvatis.

A tree, glabrous in all its parts, with long straight main
branchlets. Leaves very thick and rigid in texture, 2-3 in. long,
1-13 in. broad, with all the main veins raised on the upper sur-
face in the dried specimens, but not so conspicuous beneath.
Flowers produced from the side and tip of the young leafy
branchlets. Calyx, sepals, and petals not seen. Ovary 4 in. in
diam. after the calyx and corolla fall; stigma 1 in. in diam.,
with 4 deflexed lobes.— Baron 3633 !

GARCINIA CAULIFLORA, n. Sp.

Arborea, glabra, foliis magnis rigide coriaceis obovato-oblongis breviter
petiolatis venulis subtilibus, floribus feemineis ad ramos crassos vetustos
breviter pedicellatis, sepalis oblongis acutis coriaceis, petalis oblongis,
ovario globoso ampullzformi, stigmate magno peltato 4-lobato lobis
deflexis.

A tree, glabrous in all its parts. Leaves rigid in texture, 6-7
in. long, 2-3 in. broad above the middle, deltoid at the apex,
narrowed gradually from the middle to the base, the numerous
veins and veinlets but little raised. Flowers produced from
thick leafless branches ; pedicels 3-1 in. long. Sepals and petals
reflexing, each 4, oblong, 1-8 in. long, imbricated. Ovary globose,
j in. in diam. before the petals and sepals fall, narrowed suddenly
to a short neck ; stigma thin, iin. in diam., with 4 orbicular
deflexing lobes.— Baron 1786!

FLORA OF MADAGASCAR. 447

GARCINIA POLYPHLEBIA, n. 8p.

Arborea, glabra, foliis breviter petiolatis oblongis obtusis basi deltoideis
rigide coriaceis venis primariis multijugis venulisque pulchre exsculptis,
floribus feemineis terminalibus pedicellatis, ovario subgloboso, stigmate
crasso umbilicato lobis 4 rotundatis recurvatis.

A tree, glabrous in all its parts. Leaves 2-3 in. long, 1-1} in.
broad, very rigid in texture, bright green, deltoid at the base,
both the nearly spreading main veins and the intermediate vein-
lets raised on the under surface. Pedicels erecto-patent, + in.
long. Calyx, stamens, and petals not seen. Ovary } in. long and
broad after the petals and sepals fall; stigma 4 in. diam., thick,
brown, umbilicate iu the midale, with 4 recurved semiorbicular

lobes.— Baron 3064 ! 3101!

RuopornxNwa BakrnrANA, Baill. in Bull. Linn. Soc. Par. (1886)
pp. 566, 571, — R. altivola, Baker in Journ. Linn. Soc. xx. p. 95.

Dr. Baillon tells me that the plant which I deseribed in detail
(Baron 1980 ! 2173 !) is distinct specifically from the original type
of Thouars, which has lately beeu regathered by Humblot (24).
Mr. Baron says that R. Bakeriana is a tall erect tree, with an
edible fruit, and that its native name is Fotona. R. altivola, Thou.,

is a climber.

HIBISCUS XIPHOCUSPIS, n. 8p.

Herbaceus, caulibus elongatis erectis sursum parce pilosis, foliis petio-
latis glabris margine ciliatis inferioribus serratis ovatis simplicibus vel
trilobatis, superioribus lanceolatis integris, floribus inferioribus solitariis
axillaribus superioribus corymbosis longe pedunculatis, bracteolis pluribus
linearibus hispidis, calycis segmentis deltoideis acuminatis, petalis viola-
ceis glabris, capsule valvis ovatis acuminatis.

A herb, with very slender elongated stems. Leaves distant,
alternate; lower broad ovate, entire, or palmately 3-lobed, 14-3
in. long ; upper lanceolate, entire, shortly petioled, rounded at
the base. Flowers in a corymb at the end of the stem and
solitary from the axils of the upper leaves; peduncles slender,
3-14 in. long. Calyx 3 in. long; tube campanulate. Epicalyx
of about 10 adpressed hispid linear bracteoles. Petals oblong,
i-i in.long. Valves of the hairy capsule rather longer than the
calyx.—Baron 4533 ! 4581! 4675! 4679! 4797!

HipiscUs CYTISIFOLIUS, n. sp.
Annuus, caulibus gracilibus elongatis scabris, foliis parvis petiolatis
simplicibus vel digitatim trifoliolatis foliolis linearibus vel lineari-oblongis,

448 MR. J. G. BAKER ON THE

floribus solitariis axillaribus, bracteis pluribus liberis linearibus dense
setoso-ciliatis, calycis tubo campanulato segmentis lanceolato-deltoideis
tubo zquilongis, petalis luteis calyce duplo longioribus, genitalibus in-
elusis, fructu membranaceo globoso piloso magnitudine pisi.

An annual, with slender green stems a foot or more long.
Leaves sometimes all simple, sometimes all trifoliolate ; leaflets
1-3 in. long, usually linear or linear-oblong, rarely ovate, mode-
rately firm in texture, green and scabrous on both surfaces.
Pedicel ascending, about an inch long. Calyx, capsule, and
braeteoles all about 4 inch long. Petals glabrous, obovate un-
guiculate. Capsule splitting to the base into 5 oblong-lanceolate
valves 4 in. broad.— Baron 708 ! 3942! 4594!

HisiscUS OBLATUS, n. sp.

Fructicosus, ramulis lignosis, folis longe petiolatis integris orbiculari-
oblatis subcoriaceis facie viridibus seabris dorso dense pubescentibus,
floribus magnis axillaribus breviter pedicellatis solitariis, bracteolis 5 in
eupulam magnam rigidam dentibus deltoideis coalitis, calycis tomentos!
segmentis parvis ovatis, petalis paunosis obovato-unguiculatis, genitalibus
longe exsertis.

A shrub, with terete woody branchlets, thinly clothed upwards
with grey lepidote tomentum. Leaves 6-8 in. broad, 4-6 in.
long, eordate at the base, broadly rounded at the apex, entire or
nearly so, greenish and scabrous above, persistently matted be-
neath with soft whitish drab tomentum. Epicalyx drab, above
3 in. long. Calyx under an inch long, matted with whitish-drab
pubescence ; segments } in. long and broad. Petals pale brown,
li in. long, densely pubescent. Column of stamens exserted
nearly an inch beyond the petals.— Baron 3353! A very distinct
large striking fruticose species, allied to H. Ellisii.

HIBISCUS NUMMULARIFOLIUS, n. sp-

Herbaceus, perennis, caulibus gracilibus apice stellato-pilosis, foliis
petiolatis orbicularibus serratis facie calvatis subtus parce hispidis, floribus
paucis axillaribus et terminalibus, bracteolis pluribus linearibus hispidis,
calycis segmentis lanceolato-deltoideis, petalis violaceis glabris calyce
duplo longioribus.

A much-branched perennial herb, with short slender stems.
Leaves orbicular, subcoriaceous, under an inch long and broad ,
petioles $—j in. long; stipules lanceolate, persistent. Flowers
mainly solitary at the end of the branches. Calyx 4 in. long;

segments twice as long as the tube. Epicalyx of 10-12 linear

FLORA OF MADAGASCAR. 449

adpressed pectinate bracteoles. Petals 4-4 in. long. Capsule
not seen.— Baron 4827! Nearly allied to H. xiphocuspis, but
quite different in the leaves.

DomMBEYA ACERIFOLIA, n. sp.

D. ramulis validis, stipulis orbicularibus pilosis persistentibus, foliis mag-
uis longe petiolatis cordato-orbicularibus breviter palmatifidis denticulatis
rugulosis utrinque pilosis, umbellis simplicibus multifloris axillaribus
longe pedunculatis, pedicellis elongatis dense patentim pilosis, calycis
dense hirsuti segmentis lanceolatis, petalis latis scariosis magnitudine
mediocribus, staminodiis oblanceolatis, staminibus fertilibus 10 tubo
urceolato.

A shrub, with stout woody branchlets. Leaves sometimes
nearly a foot long and broad, with many unequal deltoid lobes,
a basal sinus 1-14 in. long, rugose like the leaf of a Salvia, thinly
bristly above, lacunose and more densely pilose beneath ; petiole
6-8 in. long; stipules persistent, coriaceous, cuspidate, ? in. in
diam. Peduncles 5-6 in. long, densely clothed with soft brown
spreading hairs. Umbel 4 in. in diam. ; pedicels capillary, 13 in.
long. Calyx 4 in. long, densely pilose. Petals $ in. long,
reddish, scariose. Staminodia 7 in. long. Anthers 4l in. long.
Style reaching nearly to the tip of the petals.— Baron 3446! A
very handsome species, near D. platanifolia, Bojer.

DomMBEYA MEGAPHYLLA, n. sp.

D. ramulis lignosis sursum tenuiter lepidotis, foliis longe petiolatis integris
cordato-orbicularibus subcoriaceis facie viridibus glabris dorso tenuiter
albido-lepidotis, floribus parvis in cymas densas axillares dichotomiter
furcatas pedunculatas aggregatis, bracteis lanceolatis, calyce lepidoto
segmentis oblongo-lanceolatis, petalis cuneatis scariosis, staminibus petalis
brevioribus fertilibus 10 tubo brevi cupulari.

Branchlets stout, only thinly lepidote. Leaves 3-1 ft. long
and nearly as broad, entire, deeply cordate at the base, with
strongly-raised veins beneath connected by cross veinlets ; petiole
2—4 in. long. Flowers in dichotomous axillary cymes much
shorter than the leaves. Bracts ~; in. long. Calyx 4 in. long,
rough with persistent drab pubescence. Petals scariose, twice
as long as the calyx. Staminodia linear, reaching to the tip of
the petals. Stamens connate only at the very base.— Baron
3443! Allied to D. spectabilis, Bojer (D. Humblotii, Baillon).
D. lucida, Baillon in Bull. Linn. Soc. Par. (1885) p. 496, is the
same species as D. floribunda, Baker, founded on Baron's 2373.

450 MR. J. G. BAKER ON THE

DoMBEYA INSIGNIS, n. sp.

. D. ramulis stellato-pilosis, foliis longe petiolatis cordato-ovatis acutis
inciso-crenatis subcoriaceis facie viridibus calvatis dorso dense pubes-
centibus, floribus maximis solitariis cernuis axillaribus, bracteis persis-
tentibus laceratis, calycis dense pilosi segmentis lanceolatis, petalis latis
scariosis cuneatis, staminodiis longissimis, staminibus fertilibus 30 et ultra
tubo brevi cupulari.

A shrub, with slender branches thinly tomentose upwards.
Leaves 3-4 in. long, dotted over when young with stellate tufts of
hairs above, densely matted beneath with pale brown pubescence ;
petiole 13-2 in. long, densely pubescent. Flowers solitary in
the axils of the leaves; peduncle about an inch long. Bracts 2
in. long. Calyx 1-11 in. long, densely roughly stellato-pilose.
Petals nearly 2 in. long, 3-1 in. broad. Staminodia an inch
long. Anthers linear, 4 in. long, as long as the free part of the
filament. Style reaching the tip of the petals——Baron 3388!
Nearly allied to D. macrantha, Baker (Baron 710! Hildebrandt
3895!). Belongs to the section T'rochetiantha of Baillon.

DomBEYA BIUMBELLATA, n. sp.

D. ramulis dense brunneo-pilosis, foliis longe petiolatis cordato-orbicu-
laribus obscure palmatim trilobatis utrinque pilosis, umbellis furcatis
axillaribus longe pedunculatis, pedicellis elongatis dense patentim pilosis,
calycis dense pilosi segmentis ovato-lanceolatis, petalis cuneatis scariosis
calyce sesquilongioribus, staminibus petalis brevioribus urceolo brevissimo,
fertilibus 10.

A shrub, with branchlets, peduncles, and pedicels densely
clothed with short soft brown pubescence. Leaves 6-8 in. broad,
subcoriaceous, persistently pilose on both sides ; petiole about as
long as the lamina. Stipules deciduous. Umbels forked, about
2 in. in diam.; pedicels 3-2 in. long. Calyx à in. long, densely
pilose. Petals much imbricated, truncate, 4 in. long. Anthers
and free filaments each yz in. long. Ovary densely pilose ; style
in. long.— Baron 3752! Allied to D. viburnefolia, Bojer.

Domseya (§ HILSENBERGIA) Baroni, n. sp.

D. ramulis pilosis, stipulis deciduis, foliis longe petiolatis cordato-orbicu-
laribus obscure palmatis utrinque persistenter pilosis, pedunculo elongato
cernuo, floribus in cymam densam dichotomam dispositis, calycis dense
pilosi tubo campanulato segmentis lanceolatis, petalis cuneatis rubellis
scariosis calyce zquilongis, filamentis in tubum cylindricum calyce duplo
longiorem coalitis, staminibus fertilibus circiter 15, staminodiis staminibus
sequilongis, stylo exserto.

A shrub with pilose branches. Leaves 6-8 in. broad, deeply
cordate, obscurely palmately lobed in the upper half, clothed with

FLORA OF MADAGASCAR. 451

short persistent pubescence on both sides; petiole 8-4 in. long.
Peduncle 6-8 in. long. Cyme dense, 3-4 in. in diam. Calyx
densely pilose ; segments 3 in. long. Staminal tube more than
twice as long as the calyx and corolla. Anthers Lin. long; free
apex of filament as long as the anther. Style exserted j in.
beyond the stamens.— Baron 3432! Closely allied to D. canna-
bina, Bot. Mag. t. 3619.

Me tocura ($ REIDLEIA) BETSILIENSIS, n. sp.

M. ramulis gracilibus apice obscure pilosis, foliis petiolatis ovato-lanceo-
latis serratis glabris, stipulis lanceolatis persistentibus, floribus axillaribus
solitariis vel ad ramorum apice paucis corymbosis pedicellis brevibus,
bracteolis 3 linearibus, calycis tubo campanulato dentibus parvis deltoideis,
petalis lilacinis, filamentis in tubum cylindrieum connatis apice liberis,
fructu globoso haud inflato, stylis basi connatis.

A shrub with long slender terete brownish branches. Leaves
1-2 in. long, moderately firm in texture, green and glabrous on
both surfaces, acute, rounded at the base, sharply serrated ; petiole
1-1 in. long. Stipules 4 in. long. Flowers not more than 2-3
at the end of the branches, and one each from the axils of the
upper leaves; pedicels hairy, at most } in. long ; bracteoles j in.
long. Calyx jl in. long and broad; teeth much shorter than
the tube. Petals l in. long. Stamens longer than the calyx.
Capsule 4 in. diam., setose.— Baron 271! 4678! 4757! Nearly
allied to the widely-spread M. corchorifolia, Linn.

ByTvrNERIA MELLERI, n. sp.

Arborea, ramulis rectis lignosis lepidoto-rugosis, foliis petiolatis cordato-
orbicularibus apice profunde emarginatis lobis rotundatis rigide coriaceis
facie viridibus glabris dorso dense lepidotis venis venulisque exsculptis,
eymis axillaribus paucifloris, fructu magno globoso lepidoto echinato,
valvis 5 lignosis e basi ruptis.

A tree 20 feet high, with stout woody branchlets. Leaves
2-3 in. long and broad, very thick, the apical sinus 2-3 in. deep,
the erecto-patent main veins of the under surface much raised,
4 radiating from the apex of the petiole, which is 1-12 in. long.
Flowers not seen. Capsule 13 in. long and broad, with 5 woody
valves, armed with spreading prickles j in. long, rough outside
with scales, smooth inside, each half valve } in. broad.—Anevoca,
by streams and moist places in woods, Dr. Meller! Allied to
B. biloba, Baill. Adans. x. 178.

ByYTTNERIA BAUHINIOIDES, n. sp.
B. ramulis gracillimis apice tenuiter lepidotis, foliis petiolatis ovatis glabris

452 MR. J. G. BAKER ON THE

utrinque viridibus basi truncatis ex apice ad medium bifidis lobis deltoideis
acutis, cymis axillaribus multifloris, bracteis deciduis, pedicellis fore
longioribus, sepalis ovatis nigris lepidotis, petalis apice appendicibus
ovatis preditis, urceolo stamineo calyce breviori.

A shrub, probably sarmentose, with slender blackish branches
lepidote towards the tip only. Leaves about 3 in. long, moderately
firm in texture, quite glabrous, cleft down to the middle ora
little beyond; petiole slender, 1-14 in. long. Cymes about as
long as the petiole; pedicels 4 in. long, slender, lepidote ; bracts
brown, lanceolate, deciduous. Calyx i in. long, blackish and
thinly lepidote; segments connate at the base. Petals brown-
purple, the brown tip exserted beyond the calyx. Staminal
urceolus campanulate, dark purple. Fruit not seen.—Baron
4627! Allied to B. heterophylla, Hook. Bot. Mise. i. 291, t. 61.

CoRCHORUS HAMATUS, n. sp.

Annuus, caulibus elongatis erectis gracilibus glabris, foliis breviter
petiolatis membranaceis glabris lanceolatis vel oblongo-lanceolatis acutis
argute serratis basi bi-aristatis, floribus paucis axillaribus brevissime pedi-
cellatis, sepalis oblongis membranaceis cuspidatis, petalis orbiculari-
unguiculatis luteis, antheris linearibus filamentis brevibus, fructu duro
cylindrico curvato rugoso rostro cylindrico integro.

An annual, with elongated slender glabrous stems. Leaves
2-3 in. long, 3-3 in. broad, thin in texture, sharply serrated,
with a large awn from each side at the base ; petiole very short ;
stipules lanceolate acuminate, persistent. Flowers 3—4 from the
axils of the leaves. Sepals membranous, } in. long, with a large
cusp at the tip. Petals bright yellow, 4—4 in. long. Stamens
about 30; anthers linear, Ł in. long, twice as long as the filaments.
Capsule sickle-shaped, abruptly curved, 1 in. long, 4 in. in diam.,
narrowed gradually into a cylindrical beak 1 in. long, obscurely
lobed at the tip only.— Baron 4712! Mojanga, Hildebrandt 3409 !

ELEOCARPUS DALECHAMPIOIDES, n. sp.

Arborea, ramulis pilosis, foliis petiolatis alternis vel oppositis obovato-
oblongis obtusis subintegris rigide coriaceis subtus pilosis, pedicellis
cernuis, sepalis subcoriaceis tomentosis lanceolato-deltoideis reflexis, petalis
latis cuneatis rubellis pubescentibus margine laciniatis, antheris apiculatis
dense pilosis.

A tree, with moderately stout pilose branchlets. Leaves 4-5
in. long, 13-25 in. broad, very obtuse, entire or obscurely denti-
culate, rounded at the base, green and glabrous on the upper
surface when mature, softly pilose, especially on the main veins

FLORA OF MADAGASCAR. 453

beneath, with 6-8 pairs of much-raised main veins on each side ;
petiole 3-4 in. long. Pedicels cernuous, 1-1} in. long. Calyx
coriaceous, 3 in. long; sepals connate above the base. Petals
an inch long, nearly an inch broad, dark red, much imbricated,
pubescent on the outside mainly on the flabellate ribs. Stamens
very numerous, half as long as the petals; anthers densely
pilose, narrowed into an apiculus j in. long above the cells.—
Baron 3822! Allied to FE. rhodanthus and quercifolius.

PsoROSPERMUM EMARGINATUM, n. Sp.

Fruticosum, ramulis glabris ancipitibus, foliis distincte petiolatis oblongis
obtusis vel emarginatis basi deltoideis utrinque viridibus glabris, umbellis
multifloris pedunculatis terminalibus, pedicellis elongatis glabris, sepalis
oblongis glabris valde nigro-lineatis, petalis calyce sesquilongioribus facie
dense pilosis, staminibus pentadelphis antheris circiter 35.

A shrub, glabrous in all its parts except the face of the petals,
the branchlets distinctly flattened towards the tip. Leaves 2-3
in. long, moderately firm in texture, green and glabrous on both
sides ; lower surface paler; main veins distinct, distant ; petiole
4-1 in. long. Flowers 15-30 in an umbel; peduncle 1-11 in. long;
pedicels 3-3 in. long. Calyx } in. long; sepals glabrous, with
copious black lines and dots outside. Petals j in. long, very
hairy on the face. Stamens as long as the calyx. Free styles
as long as the ovary.— Baron 4397 !

PsoROSPERMUM POPULIFOLIUM, N. 8p.

Fruticosum, ramosissimum, ramulis apice ferrugineo-pilosis, foliis dis-
tincte petiolatis oblongis acutis obscure crenulatis adultis utrinque glabris
viridibus, umbellis terminalibus multifloris, pedicellis pilosis flore longiori-
bus, sepalis oblongis pilosis, petalis oblanceolatis calyce duplo longioribus
facie pilosis, staminibus pentadelphis antheris circiter 40.

A much-branched erect shrub, the slender branchlets pilose
towards the tip only. Adult leaves 2-3 in. long, acute, generally
narrowed to the base, moderately firm in texture, green on both
sides, the main veins distant, arcuate ; young leaves matted with
brown pubescence on both sides. Umbels 1-2 from the axils of
the top leaves; peduncle and pedicels brown-pilose, the latter
i-i in. long. Calyx py in. long. Petals 4 in. long, lineolate and
hairy on the face. Stamens and pistil both as long as the petals.—
Baron 4611! 4617! 4810! 4812! Allied to P. Fanerana, Baker,
in Journ. Bot. 1882, p. 19.

454 MR. J. G. BAKER ON THE

PELARGONIUM ($ PELARGIUM) MADAGASCARIENSE, N. 8p.

Perenne, caulibus erectis elongatis pilosis, foliis petiolatis membranaceis
cordato-orbicularibus parce pilosis breviter lobatis inciso-crenatis, umbellis
3-4-floris, bracteis ovato-lanceolatis, pedicellis flori zquilongis, calycis
calcare brevi segmentis ovato-lanceolatis, petalis parvis anguste cuneatis
pallide lilacinis superioribus majoribus deorsum venosis, carpellorum rostro
cylindrico pubescente.

A perennial, with a fusiform root and tufted leafy erect
pubescent stems above a foot long. Basal leaves with a petiole
3—4 in. long, and a blade 3 in. broad, and basal sinus an inch
deep; upper growing gradually smaller; lobes 3-5, shallow.
Umbels produced from the axils of the leaves all down the stem ;
pedicels under 4 in. long ; bracts 5-6, membranous, pilose. Calyx
l in. long, exclusive of the short spur. Petals 3 in. long.
Beak of the carpels 3 in. long, with a short style and 5 revolute
stigmas.— Baron 3422! This is the first Pelargonium that has
been found in Madagascar. It is allied to the Cape P. alchemil-
loides and the Abyssinian P. glechomoides.

Impatiens HILDEBRANDTII, Baill. in Bull. Linn. Soc. Par.
(1886) p. 595.

Herbacea, perennis, glabra, caulibus erectis simplicibus gracilibus, foliis
oblongis vel oblongo-lanceolatis acutis integris vel parce dentatis breviter
petiolatis, floribus corymbosis longe pedicellatis, sepalis lateralibus parvis
lanceolatis, postico magno ovato-naviculari calcare gracillimo unciali,
petalis rubellis supremo orbiculari inferioribus cuneatis bifidis sepalo
postico longioribus, ovario clavato glabro.

A herbaceous perennial, glabrous in all its parts, with slender
simple erect stems a foot or more long. Leaves 10-12 to a stem,
distant, membranous, 1-3 in. long, narrowed from the middle to
the apex and base, often furnished with 1-2 bristles at the apex
of the petiole. Flowers many to a stem from the top and axils
of the upper leaves; pedicels very slender, 1—2 in. long. Blade
of the spurred sepal å in. long. Petals reddish, the two lower
1-$& in. long. Capsule glabrous, narrowed from the middle to
the apex and base.— Baron 3704! 3935! Also Hildebrandt
3828, from shady woods in the province of lmerina. Nearly
allied to J. firmula, Baker. Hildebrandt's 3382, also named
I. Hildebrandti by Dr. Baillon, is I. emirnensis, Baker.

TRIMORPHOPETALUM, gen. nov. Balsaminearum.
Sepala 3, 2 lateralia parva lanceolata viridia; posticum maximum
galeatum haud calcaratum petaloideum venosum. Petala 3, anticum
externum concavum sepalo postico simile; lateralia orbiculato-cuneata

FLORA OF MADAGASCAR. 455

unguiculata, unguibus basi unilateraliter auriculatis. Stamina 5, filamentis
cohzrentibus circiter ovarium conniventibus, antheris apicalibus. Ovarium
obliquum 5-carpellatum, ovulis multis axillibus superpositis, stigmate
sessili capitato. Fructus maturus ignotus.

T. DORSTENIOIDES, Baker.

Species sola.

Root-stock slender, creeping, sending out simple, erect, rather
fleshy fragile stems 6-9 in. long, glanduloso-viscose, especially
towards the top. Leaves 5-6 to a stem, all arising from its
upper half, ascending, distinctly petioled, oblong, acute, mem-
branous, glabrous, 14-2 in. long, deltoid and entire at the base,
shallowly broadly toothed, with little glandular cusps in the
sinuses. Flowers solitary, lateral, pedicellate, with two small
bracteoles outside the two lateral sepals. Green membranous
lateral sepals 4; in. long. Upper galeate sepal 4 in. long, veined
with brown on a drab-whitish groundwork, like an Aristolochia.
Lateral petals with a blade 4 in. long and broad, and a claw $ in.
long, auricled on the outer side. Ovary and stamens j in. long
before the petals fall.— Baron 4476! A very distinct new
generic type nearest Zmpatiens.

ERYTHROXYLUM SPARSIFLORUM, n. sp.

Ramosissimum, glabrum, foliis parvis breviter petiolatis oblongis obtusis
subcoriaceis utrinque viridibus basi cuneatis, stipulis ovatis, floribus 1-2-
nis axillaribus pedicellis brevibus, sepatis ovatis acutis basi coalitis, petalis
oblongis calyce 2-3plo longioribus, staminibus petalis subzequilongis.

A much-branched shrub, glabrous in all its parts; branchlets
slightly compressed. Leaves rigid in texture, 1-13 in. long,
narrowed from the middle to the base and point, the main veins
raised beneath, anastomosing by cross veinlets close to the
margin; petiole very short; stipules and bracts ovate, acute.
Flowers 1-2nate; pedicels 3-4 in. long. Calyx 3 line long.
Petals4l; in. long. Stamens nearly equalling the petals; anthers
oblong, minute; filaments joined at the base in a cup.— Baron
4619!

ERYTHROXYLUM AMPULLACEUM, n. Sp.

Glabrum, ramulis compressis, stipulis ovatis, foliis breviter petiolatis
oblongis obtusis subcoriaceis basi deltoideis, floribus l-2nis axillaribus,
pedicellis elongatis, sepalis oblongo-lanceolatis, filamentis elongatis basi
coalitis, fructu ampullzformi.

A shrub, glabrous in all its parts, with flattened branchlets.

456 MR. J. G. BAKER ON THE

Leaves alternate, laxly disposed, moderately firm in texture,
2-23 in. long, green and glabrous on both surfaces, with about
12 pairs of raised veins beneath anastomosing by cross veinlets
some distance from the edge; petiole very short; stipules 4 in.
long. Pedicels 3 in. long. Sepals oblong-lanceolate, reflexing,
persistent, 4l, in. long. Free filaments jl in. long. Immature
fruit ampulleform, 4-3 in. long.— Baron 3688!

TRISTELLATETA EMARGINATA, n. Sp.

Sarmentosa, ramulis glabris, foliis petiolatis oblongis subcoriaceis
emarginatis basi cuneatis utrinque viridibus glabris, racemis paucifloris
brevibus, pedicellis elongatis, sepalis oblongo-lanceolatis, ovario conico
piloso, stylo brevi, samaræ alis circiter 18 rigidis cuneatis patulis basi
pilosis.

A climber, with terete glabrous branchlets. Leaves 2-3 in.
long, moderately firm in texture, green and quite glabrous on
both surfaces, obtuse, conspicuously emarginate, deltoid at the
base. Racemes 4-6 in. long, short peduncle included ; pedicels
1-14 in. long, lower sometimes branched. Calyx 4 in. long.
Petals and stamens not seen. Wings of the fruit i in. long,
à in. broad at the tip, narrowed gradually to the base.— Baron
3078 !

TRISTELLATEIA STENOPTERA, n. sp.

Sarmentosa, foliis oppositis longe petiolatis subcoriaceis ovatis facie
glabris dorso pubescentibus, racemis laxis longissimis, pedicellis arcuatis
flore longioribus, sepalis oblongis crispatis dense pilosis, petalis oblongis
unguiculatis, antheris lanceolatis filamento longioribus, ovario piloso
globoso, stylo elongato, samarze alis circiter 15 angustis pubescentibus.

A climber, with slender terete branchlets. Leaves 3-4 in.
long, moderately firm in texture, slightly cordate, deltoid at the
apex, when mature green and glabrous above, thinly coated with
pale pubescence beneath, with about 6 pairs of arcuate main
veins. Racemes axillary and terminal, the end ones 8-9 in. long ;
lower pedicels 4-1 in. long. Calyx 1 in. long. Petals bright
yellow, 3 in. long. Wings of the fruit 2 in. long, 4!; in. broad at
the tip, sometimes emarginate.— Baron 4754!

EvoniA DISCOLOR, n. sp.

Arborea, ramulis apice pubescentibus, foliis simplicibus lanceolatis vix
petiolatis obscure pubescentibus facie viridibus dorso pallidis, floribus
tetrameris in paniculam densam terminalem ramis deltoideis dispositis,
pedicellis brevibus pubescentibus, calycis segmentis orbicularibus, petalis

FLORA OF MADAGASCAR. 457

oblongis crebre nigro-punctatis, staminibus exsertis, pistillo in flore
masculo rudimentario.

A tree, with branchlets pubescent and flattened at the tip.
Leaves subcoriaceous, subobtuse, sometimes half a foot long, an
inch broad at the middle, narrowed to the base, greenish white
beneath, with raised main veins and copious minute immersed
dots. Panicles on pubescent peduncles from the axils of the
large uppermost leaves. Calyx 4!; in. in diam., brown, pubescent.
Corolla 4; in. long. Female flower and fruit not seen.—Baron

3754!

EvopDtA FLORIBUNDA, n. sp.

Arborea, ramosissima, glabra, foliis longe petiolatis trifoliolatis, foliolis
oblongis cuspidatis basi attenuatis, floribus tetrameris in paniculas densas
terminales corymbosas dispositis, pedicellis pubescentibus flori zequilongis,
calycis pilosi segmentis orbicularibus, petalis oblongis pallidis dense
glanduloso-punctatis, staminibus exsertis, pistillo in flore masculo rudi-

mentario.
A much-branched erect tree, with slender terete brown

glabrous branchlets. Petiole 1-2 in. long; leaflets 2—4 in. long,
moderately firm in texture, green and glabrous on both surfaces,
dotted with copious minute immersed black glands, cuneately
narrowed at the base into a short winged petiolule. Flowers in
copious ample dense terminal panicles, about as long as the
upper leaves; branches erecto-patent ; final branchlets densely
corymbose. Calyx 3 line in diam., brown, densely pubescent.
Petals 4 in. long. Filaments finally as long as the petals ; anthers
oblong, versatile.— Baron 2221!

OcHNA MACRANTHA, n. Sp.
Glabra, ramulis gracilibus, foliis breviter petiolatis oblongo-lanceolatis
subobtusis denticulatis, cymis axillaribus paucifloris, pedicellis elongatis,
sepalis oblongis magnis inzqualibus, petalis magnis luteis orbiculari-
unguiculatis, staminibus permultis antheris linearibus, carpellis maturis

turbinatis.
A tree, glabrous in all its parts, with slender terete branchlets.

Leaves moderately firm in texture, 3-4 in. long, an inch broad
at the middle, rather glossy above, green on both sides, finely
veined, ciliato-denticulate; petiole not more than 3-4 in. long.
Cymes axillary, 3-4-flowered ; pedicels spreading, 4-2 in. long.
Sepals i-i in. long, green, glabrous, firm in texture, Petals
bright yellow, an inch long, with a blade 4 in. broad, cuneate at
the base, and a distinct claw. Stamens half as long as the petals.
LINN. JOURN.— BOTANY, VOL. XXII. 2P

458 MR. J. G. BAKER ON THE

Fruit-carpels much shorter than the persistent calyx.— Baron
3927!4811!4818! Native name Lanary. Allied to O. ciliata,
Lam.

APODYTES EMIRNENSIS, n. sp.

Arborea, ramulis glabris, foliis petiolatis oblongis acutis basi deltoideis
dorso obscure pubescentibus, floribus in panieulas multifloras axillares
peduneulatas dispositis, pedunculis pedicellisque pilosis, calyce parvo
dentibus deltoideis, petalis oblanceolatis obtusis, staminibus petalis zequi-
longis antheris oblongis, ovario ovoideo stylo brevi.

A tree, with slender woody naked branchlets. Leaves 3-4 in.
long, 18-21 lines broad, quite entire, moderately firm in texture,
finely veined, only slightly pubescent on the midrib beneath.
Panicles 2-3 in. in diam., 3-4-pinnate; branches pubescent;
ultimate pedicels much shorter than the flowers; bracts minute,
ovate. Flowers pentamerous. Calyx 4 linein diam., pubescent.
Bud oblong, glabrous, yy in. long. Anthers as long as the
filaments, oblong, reaching to the tip of the petals Baron 4791!
4865! Nearly allied to the Mauritian 4. mauritiana, Planch.

TURRZA RHOMBIFOLIA, n. sp.

Glabra, foliis parvis rhomboideis breviter petiolatis, floribus solitariis
tetrameris terminalibus breviter pedunculatis, ealycis parvi tubo campa-
nulato segmentis ovatis acutis, petalis luteis oblongo-lanceolatis longe

_ unguiculatis, tubo stamineo eylindrico petalis zquilongo, staminodiis
parvis lanceolatis profunde bifidis, antheris dense pilosis connectivo pro-
ducto apiculatis, stigmate exserto capitato.

A tree or shrub, with slender glabrous branchlets. Leaves
moderately firm in texture, not more than i-i in. long, cuneate
at the base, green and glabrous on both surfaces. Pedicels
slender, glabrous, 4 in. long. Calyx lin.long. Petals an inch
long, $-i in. broad. Filaments united in a cylindrical column
an inch long; teeth 8, laciniated at the tip; anthers 4. in. long,
very hairy. Style reaching to the tip of the anthers; stigma
eapitate.— Baron 4569 !

TURRXA VENULOSA, n. sp.

T. ramulis pilosis, foliis oblongis acutis rigide coriaceis facie nitidis
viridibus dorso obscure pubescentibus venis exsculptis, floribus solitariis
tetrameris breviter pedicellatis, calycis pilosi segmentis lanceolatis, petalis
oblanceolatis unguiculatis, tubo stamineo cylindrico petalis sequilongo,

dentibus quadratis bifidis, antheris 8 glabris connectivo producto apiculatis,
stylo longe exserto.

FLORA OF MADAGASCAR. 459

A tree or shrub, with long slender pilose branchlets. Leaves
1-2 in. long, shortly petioled, rigid in texture, bright green
above, dull green beneath, with the main veins much raised and
connected by arches within the margin. Pedicels 1 in. long,
Calyx 7 in. long; teeth shorter than the tube. Petals 14 in.
long. Filaments united in a cylindrical tube as long as the
petals, with 8 minute quadrate bifid teeth. Anthers 4^; in. long.
Style exserted half an inch beyond the stamens.— Baron 4199!

CoMMIPHORA (BALSAMODENDRON) FRAXINIFOLIA, N. Sp.

Arborea, glabra, ramulis validis inermibus, foliis imparipinnatis longe
petiolatis utrinque viridibus glabris, foliolis 5-7 ovatis vel oblongis sessili-
bus, panieulis axillaribus laxis foliis brevioribus, ramis inferioribus paucis
brevibus corymbosis, pedicellis gracilibus flore interdum longioribus,
calycis segmentis ovatis, petalis angustis calyce paulo longioribus, ovario
ampulleformi glabro stigmate sessili coronato.

A tree, with stout glabrous brown sulcate branchlets. Leaves
8-9 in. long, including the 2-3 in. petiole; leaflets 3-4 in. long,
shallowly crenate, moderately firm in texture, green and glabrous
on both surfaces. Panicles crowded at the end of the branchlets,
contemporary with the young leaves, 2-3 in. long, very lax, the

lower branches sparingly compound; pedicels slender, 1-j in.

long. Calyx glabrous, j in. long; tube short, campanulate ;
segments acute, firm in texture. Petals oblanceolate, 4 in. long.
Ovary glabrous, narrowed very gradually to the sessile stigma.—

Baron 44.04!

COMMIPHORA (BALSAMODENDRON) LAXIFLORA, n. sp.

Arborea, ramulis validis junioribus pilosis, foliis longe petiolatis impari-
pinnatis utrinque viridibus pilosis, foliolis 5-7 oblongis cuspidatis serratis,
racemis elongatis laxissimis breviter pedunculatis, pedicellis crassis brevibus
infimis furcatis, calycis segmentis ovatis rigide coriaceis glabris.

A tree, with stout branchlets, at first pilose. Leaves 8-9 in.
long, including the 3-inch petiole; leaflets 2-4 in. long, the end
one largest, minutely serrated, rounded at the base, moderately
firm in texture, green and obscurely pilose above, paler and more
hairy beneath, with 8-10 pairs of fine ascending raised main
veins. Racemes axillary, as long as the leaves; rhachis slender,
glabrous ; flowers few, distant, solitary except the very lowest ;
pedicels 45-3 in. long. Sepals 4, ovate, decussate, ~y in. long.
Petals and fruit not seen—Baron 4837! Closely allied to
C. grandifolia, Engler.

2P2

460 MR. J. G. BAKER ON THE

GYMNOSPORIA CUNEIFOLIA, n. sp.

Glabra, ramulis virgatis angulosis spinosis, foliis parvis obovatis obtusis
fasciculatis brevissime petiolatis, floribus in cymis axillaribus paucifloris
dispositis, pedicellis brevibus, calycis segmentis ovatis, fructu valvis tribus
latis patentibus emarginatis, seminibus ovoideis castaneis.

A shrub or tree, with long slender angled branchlets, bearing
tufts of leaves in the axils of pungent straight slender spines 4-7
in. long. Leaves firm in texture, very obtuse, pale green and
glabrous on both surfaces, usually about half an inch long.
Cymes usually one from each of the fascicles of leaves, which they
slightly overlap, few-flowered ; pedicels finally 4 in. long. Petals
and stamens not seen. Fruit-valves rigidly coriaceous, 4 in.
long and broad. Seeds ovoid, bright brown, 2 to each cell.—
Baron 4201! Closely allied to G. berberidacea, Baker, which has
just the same Berberis-like habit.

ELÆODENDRON GYMNOSPOROIDES, n. sp.

Fruticosa, glabra, ramulis gracilibus, foliis parvis oblongis obtusis
obscure crenulatis, floribus in cymis 2-3-floris dispositis, pedicellis strictis
flore longioribus, calycis tubo brevissimo dentibus ovatis, petalis orbiculari-
bus, staminibus brevibus, disco pentagono, ovario obconico stigmate
sessili.

An erect shrub or small tree, with very slender branchlets.
Leaves alternate, shortly petioled, 3-1 in. long, deltoid or rather
rounded at the base, finely veined. Cymes axillary or terminal;
pedicels 5-2 in. long. Bud globose. Calyx 4l in. in diam.

me greenish yellow, à in. in diam. Fruit not seen.—Baron
735!

Virts ($ AMPELOCISsUs) SPHJEROPHYLLA, n. sp.

Sarmentosa, glabra, cirrhifera, folis membranaceis longe petiolatis

digitatis, foliolis 5 longe petiolulatis magnis orbicularibus denticulatis,
floribus pentameris in paniculas bipinnatas ramulis racemosis dispositis,

pedieellis brevissimis, calyce patelleformi piloso subintegro, petalis
oblongis, staminibus brevibus.

A large climber, with rather Stout angled woody stems. Leaves
digitately 5-foliolate ; petiole 2 in. long; upper leaflets 3-4 in.
long and broad, much imbricated, rounded or shallowly cordate
at the base, with distinct petiolules 3-1 in. long ; two outer
leaves smaller, cuneate at the base. Panicles thyrsoid, peduncled,

lateral; branches pubescent. Buds globose. Flowers very
small. Fruit not seen.— Baron 4841 !

FLORA OF MADAGASCAR. 461

Vrris (§ AMPELOCISSUS) RHODOTRICHA, n. sp.

Sarmentosa, cirrhifera, ramulis gracilibus lignosis sursum ferrugineo
pubescentibus, foliis simplicibus cordato-ovatis denticulatis facie viridibus
glabris dorso ferrugineo-pubescentibus, panicule ramis dense corymbosis
pedunculis pedicellisque dense ferrugineo-pilosis, floribus pentameris,
calyce obscure dentato, petalis parvis glabris sordide purpureis, ovario

glabro.
A climbing shrub, with slender woody branchlets, clothed

upwards, like the panicle and underside of the leaves, with dense
bright ferruginous pubescence. Leaves 3—4 in. long and broad,
moderately firm in texture, ferruginous on the raised veins
beneath, with a broad open basal sinus. Panicles thrysoid, with
distant short erecto-patent densely corymbose maia branches;
pedicels finally longer than the flowers. Bud globose, under a
line in diameter. Calyx patelleform, densely ferrugineo-
pubescent. Petals not more than a line loug. Ovary glabrous ;
style as long as the ovary Baron 4749! 4838! Nossi-bé,
Hildebrandt 2876!

Viris ($ Cissus) VOANONALA, n. sp.

Sarmentosa, copiose cirrhifera, ramulis gracillimis brevissime pubescen-
tibus, foliis bipinnatis membranaceis glabris, foliolis serratis obovatis vel
oblongis, floribus tetrameris corymboso-paniculatis, pedunculis pedicellis-
que pubescentibus, calyce patelleformi vix dentato, petalis parvis ovatis
viridibus, fructu immaturo turbinato glabro.

A climber, with very slender stems and copious tendrils.
Fully-developed leaves 6-8 in. long, including the long petiole;
blade simply pinnate in the upper half, bipinnate at the base;
leaflets 12-20 to a leaf, about an inch long, bright green, deltoid
or rounded at the base. Panicles lax, axillary or terminal, 1-3
in. broad ; pedicels very short. Flower-calyx jlin. in diam. Bud
globose. Ovary globose; style very short; stigma capitate.—
Baron 4561! Dr. Parker! Native name Voanonala. Allied to
V. microdiptera, Baker. In a fully-developed leaf there are
5 leaves above a branch, 2 branches with 3 pairs of leaves and
2 with 5 pairs.

LEEA CUSPIDIFERA, n. Sp.

Fruticosa, ramis pubescentibus, foliis longe petiolatis rhomboideis
tripinnatis, foliolis ovatis vel oblongis serratis conspicue cuspidatis facie
viridibus glabris dorso pallide viridibus ad venas hispidis, floribus copiose
corymboso-paniculatis, pedicellis calyci zequilongis, calycis tubo campanu-
lato dentibus ovatis, petalis rubellis, tubo stamineo ad medium quinquefido
dentibus quadratis erectis emarginatis.

462 MR. J. G. BAKER ON THE

A. climbing shrub, with pubescent woody branchlets. Petiole
5 in. long; blade 6-8 inches long and broad; leaflets #-1} in.
long, moderately firm in texture, bright green above. Panicle
dense, 2 in. in diam. ; branches densely pubescent ; bracts minute,
lanceolate. Calyx pubescent, 4; in. in diam. ; teeth shorter than
the tube. Petals reflexing, i in. long. Corona rather shorter
than the petals, deeply 5-cleft.— Baron 4785 !

TINA VELUTINA, n. Sp.

Arborea, ramulis velutinis, foliis confertis subcoriaceis facie parce dorso
dense brunneo-pilosis, foliolis 11-13 oblongis acutis serratis venulis faciei
inferioris exsculptis, floribus paniculatis, pedicellis brevibus, bracteis mi-
nutis deltoideis, calycis segmentis parvis obtusis, petalis calyce duplo lon-
gioribus, fructu turbinato glabro.

A tree, with the leaves and panicles crowded towards the end
of the velvety branchlets. Leaves half a foot long including the
short petiole; upper leaflets 2-3 in. long, acute or obtuse, with a
cusp; texture slightly coriaceous ; main veins of the under sur-
face numerous, prominently raised. Panicles about as long as
the leaves; pedicels finally +in. long. Calyx 4 in. in diam. ; seg-
ments very short, obtuse. Fruit turbinate, glabrous.— Baron

1457! 1522! 3968! Differs from T. dasycarpa by its acute
serrated leaflets.

CALYCIFLOR X.

BynsocanPus BARONTI, n. sp.

B. ramis crebre lenticillatis, ramulis dense pubescentibus, foliolis 20-24
petiolulatis oblongis obtusis glabris, floribus cum foliis immaturis produc-
tis, cymis densis paucifloris, pedicellis pubescentibus prope apicem arti-
culatis, sepalis ovatis obtusis pilosis, petalis oblanceolatis calyce 3-4plo
longioribus, staminibus longioribus petalis paulo brevioribus, carpellis
pilosis, stylo brevi, stigmate capitato.

A shrub or tree, with densely pubescent final branchlets.
Fully developed leaves 4-5 in. long, ineluding the short petiole ;
leaflets thin, about an inch long. Flowers contemporary with the
young leaves. Calyx very hairy, 74 in. in diam. ; segments pilose,
much imbrieated. Petals } in. long. Filaments long; anthers
small, globose. Carpels hairy, about as long as the calyx. Fruit

not seen.— Baron 4992! The genus Byrsocarpus is united by
Dr. Baillon with Rourea.

CROTALARIA LUTEO-RUBELLA, n. sp.

Fruticosa, giabra, ramulis gracilibus, foliis longe petiolatis trifoliolatis

FLORA OF MADAGASCAR. 463

foliolis magnis oblongis acutis, racemis terminalibus laxis multifloris, pe-
dicellis elongatis, bracteis minutis caducis, calyce glabro tubo campanulato
dentibus acutis tubo equilongis, petalis glabris calyce duplo longioribus,
vexillo rubello carina alisque luteis, legumine oblongo-clavato rigidulo 2-3-

spermo longe stipitato.
A much-branched shrub, with slender terete virgate branchlets.

Petiole 11-2 in. long; leaflets membranous, sometimes 3-4 in.
long, green and glabrous on both sides. Racemes 4-5 in. long;
lower pedicels 1 in. long. Calyx broadly campanulate, j in. in
diam.; teeth lanceolate from a broad base. Corolla 3 in. long ;
keel broad, with a conspicuous erect beak. Pod glabrous, rigid
in texture. Pod 1 in. long, 4 in. in diam., with a slender
gynophore 3 in. long.— Baron 4708! 4856! Hildebrandt 3441!
Near C. striata, DC., and C. Pervillei, Baill.

CROTALARIA MACROPODA, n. Sp.

Fruticosa, ramulis gracilibus obscure strigillosis, stipulis minutis caducis,
foliis trifoliolatis facie viridibus glabris dorso tenuiter pilosis, foliolis ob-
longis acutis vel obtusis, racemis laxis terminalibus paucifloris, pedicellis
calyei eequilongis, bracteis minutis ovatis caducis, calyce dentibus lanceo-
latis tubo duplo longioribus, petalis rubellis glabris calyce duplo longiori-
bus, ovario oblongo 4-ovulato.

A much-branched erect shrub, with slender terete branchlets.
Petiole 1-14 in. long; leaflets 1-2 in. long, moderately firm in
texture, green on both surfaces. Racemes copious, lax, 1-2 in.
long; pedicels finally 1 in. long. Calyx jin. long, slightly silky ;
tube very short; teeth lanceolate acute. Corolla 3 in. long;
standard 4 in. broad; keel broad, rostrate, up-curved. Mature
pod not seen.—Baron 3789! 3862! Near C. striata, DC.

INDIGOFERA DESMODIOIDES, n. sp.

Fruticosa, ramulis apice strigillosis, stipulis linearibus, foliis impari-
pinnatis breviter petiolatis rigidulis siccitate nigrescentibus utrinque albo-
strigillosis, foliolis 25-27 lanceolatis petiolulatis mucronatis, racemis
densis subsessilibus axillaribus multifloris, pedicellis brevibus, calyce parvo
late campanulato dentibus parvis lanceolatis, petalis glabris calyce multo
longioribus, legumine curvato conspicue toruloso 4-6-spermo.

A shrub, with long slender woody branchlets. Leaves 2-3 in.
long; leaflets about 3 in. long, 4 line broad, firm in texture,
turning blackish when dried, like those of I. Anil and I. tinctoria.
Racemes copiously produced in the axils of the leaves ; pedicels
about as long as the calyx, finally cernuous. Calyx 4 line in diam. ;
tube campanulate, strigillose. Corolla } in. long. Pod sickle-

464 MR. J. G. BAKER ON THE

shaped, under an inch long, conspicuously torulose, naked when
mature, black.— Baron 889! 3965! Group of I. tinctoria.

INDIGOFERA ORMOCARPOIDES, n. sp.

Fruticosa, ramulis gracilibus obscure strigillosis, stipulis lanceolatis
minutis, folis membranaceis petiolatis utrinque obscure albo-hispidulis,
foliolis 5-7 oblongis petiolulatis stipellatis, racemis axillaribus, floriferis
densis brevibus subsessilibus, fructiferis elongatis, pedicellis calyci zequi-
longis, bracteis lanceolatis caducis, calyce parvo piloso dentibus deltoideis,
petalis pilosis sordide luteis, legumine gracili elongato toruloso 5-6-
spermo.

A shrub, with slender terete branchlets. Leaves 3-4 in. long,
including the $ in. petiole; leaflets opposite, 1-2 in. long, mu-
cronate, thin in texture, thinly elothed with adpressed white
bristles on both surfaces. Flower-racemes about an inch long,
nearly sessile, much shorter than the leaves. Calyx oblique,
broadly campanulate, 3 lin. in diam.; teeth minute. Corolla lin.
long. Pod 2 in. long whilst still immature, glabrous, conspi-
cuously torulose.— Baron 4607! 4734! 4822! Group of
I. tinctoria.

MUNDULEA LAXIFLORA, n. sp.

Fruticosa, ramulis gracilibus velutinis, stipulis lanceolatis persistentibus,
foliis imparipinnatis subcoriaceis facie tenuiter dorso dense sericeis, foliolis
7-9 obovatis vel oblongis petiolulatis, racemis laxis, pedicellis calyci zequi-
longis, bracteis caducis, calyce sericeo tubo campanulato dentibus acumi-
natis, petalisrubellis sericeis, legumine velutino.

A shrub, with slender terete woody velvety branchlets. Leaves
3--4 in. long, including the i-l in. petiole; upper leaflets 1-13
in. long; side ones opposite; upper subacute ; lower shorter,
obovate, obtuse. Racemes 2-4 in. long; lower pedicels 4 in.
long. Calyx silky, broadly campanulate, 4 in. long; teeth
deltoid-cuspidate; upper as long as the tube, lower longer.
Corolla 3 in. long; standard silky on the outside. Pod linear,
not seen fully mature.— Baron 3861!

LEPTODESMIA CONGESTA, Benth.

The Madagascar Desmodium lespedezioides, Benth. in Plant.
Junghuhn. p. 223, of which we have specimens from Ellis, Bojer,
and Parker, seems to be conspecific with the Indian Leptodesmia
congesta, Benth. MSS. ; Baker in Hook. fil. Fl. Brit. Ind. ii. p. 152.
Hallia ? Bojeriana, Baill. in Bull. Linn. Soc. Par. p. 432, is a

FLORA OF MADAGASCAR. 465

very distinct second species of Leptodesmia, endemic in Mada-
gascar. We have specimens from Bojer and Justiee Blackburn.
It has a 1-jointed 1-seeded suborbicular compressed legume } in.
long, tipped with the short curved style.

CLITORIA ZANZIBARENSIS, Vatke in @Œster. Bot. Zeitschrift,
1878, p. 261.—Baron’s 4586 and 4615 quite agree with Hilde-
brandt’s 1189 from Zanzibar, on which this isfounded. It seems
to me only a variety of C. heterophylla, Lam., receding from the
type by its lanceolate or oblong-lanceolate leaflets 1-14 in. long,
and larger flowers.

Mucuna (§ SrrzoLoBIUM) AXILLARIS, n. sp.

Herbacea, volubilis, ramulis tenuibus pilosis, foliis longe petiolatis mem-
branaceis facie tenuiter dorso dense adpresse hispidulis, floribus paucis
axillaribus fasciculatis brevissime pedicellatis, calycis pilosi tubo campanu-
lato dentibus ovatis acutis tubo :equilongis, vexillo petalis reliquis
duplo breviori, legumine cylindrico exalato dense persistenter hispidulo-
velutino.

A herbaceous climber, with very slender densely pilose stems.
Stipules minute, lanceolate ; petiole 12-2 in. long; leaflets 2-3
in. long; end one oblong; stipelle minute. Flowers in nearly
sessile fascicles in the axils of the leaves. Calyx j in. long, very
silky ; tubes broadly campanulate ; teeth deltoid or lanceolate.
Corolla above an inch long ; petalsall obtuse. Pod curved, 14 in.
long, 3 in. indiam., clothed with spreading grey bristles — Baron
4877! Native name dgy. Allied to M. pruriens, DC.

RHYNCHOSIA TRICHOCEPHALA, N. Sp..

Volubilis, ramulis dense pilosis, foliis longe petiolatis trifoliolatis sub-
coriaceis facie parce dorso dense pilosis, foliolo terminali cordato-ovato,
racemis densis axillaribus et terminalibus, bracteis magnis ovatis dense
pilosis, pedicellis calyei zquilongis, calycis pilosi tubo campanulato den-
tibus lanceolatis, petalis conspicue venosis calyce duplo longioribus, ovario
dense piloso.

A climber, with slender terete woody densely pilose branchlets.
Stipules scariose, lanceolate ; petiole 1-14 in. long; leaflets 2-3
in. long, acute, shallowly cordate, green and slightly hairy above,
persistently pubescent beneath. Racemes copious, few-flowered,
dense; bracts ovate acuminate, j in. long. Calyx pilose, } in.
long; teeth all acute. Corolla 3 in. long; petals conspicuously
veined with reddish brown on a yellowish ground. Stamens and
style abruptly up-curved.— Baron 3393! Near R. viscosa, DC.

466 MR. J. G. BAKER ON THE

DALBERGIA SCORPIOIDES, n. sp.

D. ramulis gracilibus glabris, foliis petiolatis imparipinnatis firmis gla-
bris, foliolis 9-11 oblongis alternis petiolulatis, floribus copiose paniculatis,
paniculz ramis scorpioideis, pedicellis subnullis, bracteis late ovatis per-
sistentibus, calyce tubo infundibulari dentibus tubo vix sequilongis, petalis
calyce duplo longioribus, staminibus isadelphis, ovario piloso.

An erect shrub or tree, with glabrous branchlets and leaves.
Leaves 3-6 in. long, including the 1-14 in. petiole; leaflets 1-11
in. long, the upper larger and oblong, the lower shorter and sub-
orbicular ; petiolules y% in. long. Flowers in copious terminal
panicles with scorpioid pubescent branchlets. Cal yx d; in. long,
with a pair of small persistent ovate obtuse bracts clasping its
base; teeth obtuse. Petals obtuse, twice as long as the calyx.
Stamens in two bundles of 5 each, nearly as long as the petals.—
Baron 4583 !

Darserera Poort, n. sp.

D. ramulis validis glabris, foliis longe petiolatis firmis glabris lucidis,
foliolis 5-7 alternis longe petiolulatis oblongis vel ovatis, floribus dense
corymboso-paniculatis, panicule ramulis pilosis, pedicellis brevibus, calyce
glabro tubo campanulato glabro dentibus difformibus, petalis calyce duplo
longioribus, staminibus monadelphis.

A tree, glabrous in all its parts. Leaves 4-6 in. long, inclu-
ding the 1-1} in. petiole; end leaflet 2-3 in. long; lower shorter,
ovate, acute; petiolules 1—1 in. long. Panicles terminal; lower
branches from the axils of large leaves. Calyx 4l; in. long ; lower
teeth as long as the tube ; upper semiorbicular, much shorter.
Corolla $ in. long; standard orbicular. Stamens 10, monadel-
phous. Ovary hairy along the upper suture. Fruit not seen.—
Baron 4393! Mr. Pool brought us in 1876 four different kinds
of Voamboana timber, from one of which, he says, handsume fur-
niture is manufactured. The present species is allied to the
Indian D. latifolia, Roxb.

LONCHOCARPUS PAULLINIOIDES, n Sp.

Sarmentosus, ramulis gracilibus teretibus glabris, folis longe petio-
latis trifoliolatis glabris, foliolis oblongis acutis, racemis elongatis laxis,
floribus fasciculatis, pedicellis brevibus, bracteolis ovatis obtusis persis-
tentibus, calyce sericeo tubo campanulato dentibus brevibus obtusis, pe-
talis latis rubellis calyce 3-4plo longioribus, legumine lineari villoso.

A scandent shrub, with slender terete woody branchlets.
Petiole 1-23 in. long; leaflets shortly stalked, the end one the
largest, 2-4 in. long, moderately firm in texture, green and gla-

FLORA OF MADAGASCAR. 467

brous on both surfaces. Flowers in lax racemes and also in
fascicles from the axils of theleaves; pedicels unequal, sometimes
4-} in. long. Calyx } in. long, with a pair of ovate persistent
bracteoles at the base; teeth obtuse, much shorter than the tube.
Petals silky, 3-2? in. long. Pod straight, linear, densely deltoid,
with short persistent brown pubescence.—Baron 4748! 4828!
4807 ! 4863! 4853!

DicHROSTACHYS UNIJUGA, n. sp.

Fruticosa, ramulis gracilibus lignosis, stipulis ovatis scariosis striatis,
foliis breviter petiolatis, pinnis unijugis, foliolis 10—25-jugis parvis confertis
rigidulis inzquilateraliter lineari-oblongis obtusis, floribus in capitulis
oblongis axillaribus breviter pedunculatis aggregatis, superioribus densi-
oribus staminibus antheriferis, calyce sessili obconico segmentis ovatis,
petalis oblongo-lanceolatis.

A shrub, with long slender woody glabrous branchlets. Pinne
1-1 in. long, erecto-patent ; rhachis slender, pubescent ; leaflets
deciduous, + in. long, green and glabrous on both surfaces, pro-
duced at the base on the lower side. Heads 3 in. in diam., on short
peduncles from the axils of the leaves. Corolla 3 line long, three
times as long as the calyx. Staminodes reddish, protruded 3-3
in. beyond the tips of the lower flowers. Pod not seen.— Baron
4602! 4758! 4765!

PrPTADENIA LEPTOCLADA, n. sp.

Fruticosa, inermis, glabra, ramulis gracillimis, foliis bipinnatis, petiolo
glandula magna ad apicem przdito, pinnis 12-18-jugis, foliolis multijugis
rigidulis parvis oblongo-lanceolatis eontertis, floribus pentameris in capi-
tulis oblongis axillaribus breviter pedunculatis aggregatis, calyce basi
bracteato tubo infundibulari dentibus deltoideis, petalis lanceolatis luteis,
staminibus longe exsertis.

A shrub, glabrous in all its parts, with long slender branchlets.
Leaves 2-3 in. long, including the 4 in. petiole, which bears a
large conical brown gland on one side at the top; pinne 1 in.
long; leaflets -4 in. long, crowded, firm in texture, bright green.
Heads solitary from the axils of the upper leaves, on short erecto-
patent peduncles, fin. in diam. Petals 4; in. long. Stamens
twice as long as the petals. Legume not seen.— Baron 4730!

MIMOSA MYRIOCEPHALA, n. sp.

Fruticosa, ramulis validis angulatis glabris cum foliorum rhachibus aculeis
copiosis uncinatis parvis stramineis armatis, foliis bipinnatis, pinnis 10-12-
jugis, foliolis multijugis inzquilateraliter oblongis rigidulis obscure ciliatis

468 MR. J. G. BAKER ON THE

basi postice productis, floribus tetrameris in capitula parva globosa pe-
dunculatis copiose paniculatis dispositis, calyce glabro tubo infundibulari
dentibus parvis deltoideis, petalis oblongis, legumine tenui curvato glabro
inermi 6-8-spermo.

A. shrub, with stout strongly angled branchlets, armed, like
the leaf-rhachides, with copious small toothed prickles. Leaves
sometimes a foot long; punt 2-3 in. long; leaflets firm in
texture, obtuse, imbricated, 1—3 in. long. Panicle a foot or more
long, the lower reae produced from the axils of large leaves ;
heade globose, i in. in diam. Stamens yellow, much exserted.
Legume 3 in. long, 1 in. broad.— Baron 990 ! 2096 ! 3410! 4700.
Near M. latispinosa.

ACACIA XIPHOCLADA, D. sp.

Arborea, ramosissima, glabra, ramulis angulatis, foliis propriis bipinnatis,
pinnis 5-6.jugis, foliolis multijugis parvis oblongis confertis, phyllodiis
rigidulis lanceolatis subsessilibus, floribus in capitula parva axillaria
globosa breviter pedunculatis aggregatis, calyce obconico vix dentato,
corolla segmentis luteis oblongo-lanceolatis, staminibus breviter exsertis,
legumine ignoto.

A much-branched tree, glabrous in all its parts. Proper
leaves apparently only produced when it is young, 3 in. long,
including the short petiole, with pinnæ under an inch long, with
very numerous small crowded leaflets. Phyllodia rigidly coria-
ceous, flat, simple, 2-4 in. long, 4—4 in. broad at the middle, ver-
tically ribbed, narrowed gradually to the base and apex. Heads
globose, } in. in diam., produced in the axils of the phyllodes on
short erecto-patent US Flowers 20-30 to a head. Calyx
lin.long. Pod not seen.— Baron 1695! 4384! Closely allied
to A. heterophylla, Willd., of Mauritius.

ALBIZZIA TRICHOPETALA, n. sp.

Fruticosa, ramulis pubescentibus, petiolo medio glandulifero, pinnis 5-6-
jugis, foliolis 10-20-jugis parvis oblongo-rhomboideis confertis facie glabris
viridibus dorso albidis pilosis, floribus in capitula glabra axillaria pedun-
culatis aggregatis dispositis, pedicellis pilosis flori zequilongis, calyce piloso
segmentis parvis deltoideis, corollz pilosæ dentibus ovatis, tubo staminali
petalis breviori, ovario lineari glabro.

An erect shrub or tree, with slender pilose branchlets. Pe-
tiole 1-14 in. long, with a single large gland at the middle;
blade 3-4 in. long; upper pinne 2 in. long ; lower shorter, with
fewer leaflets ; leaflets} in. long. Heads 11 in. in diam., produced
singly in the axils of the upper leaves on slender erecto-patent

FLORA OF MADAGASCAR, 469

peduncles 2 in. long. Calyx + in. long. Corolla 4 in. long,
densely pilose. Stamens protruded $ in. beyond the corolla.
Legume not seen.—Baron 4644! Near the Cape 4. Forbesii,
Benth.

PARINARIUM EMIRNENSE, n. sp.

P. ramulis gracilibus apice pubescentibus, foliis breviter petiolatis obovato-
oblongis obtusis subcoriaceis lucidis glabris, floribus laxe corymboso-pani-
culatis, ramulis pubescentibus, pedicellis brevibus, sepalis ovatis tomen-
tosis, petalis oblongis unguiculatis, ovario globoso obliquo tomentosc»
stylis 2 glabris ovario æquilongis.

An erect shrub or tree, with slender branchlets. Petiole +4
in. long; blade 2-3 in. long, 1-1} in. broad, green and glabrous
on both surfaces, with 5-6 pairs of arcuate raised main veins
beneath. Panicles both axillary and terminal; branchlets and
pedicels grey-tomentose. Calyx j in. long. Petals dark purple.
Ovary globose, densely pubescent, seated obliquely on the base of
the calyx.— Baron 3672!

WEINMANNIA LEPTOSTACHYA, n. Sp.

Fruticosa, ramulis gracilibus pubescentibus, foliis distincte petiolatis
imparipinnatis rigidulis glabris venis exsculptis, foliolis 7-9 oblongis ob-
tusis inciso-crenatis, floribus parvis pentameris spicatis, calycis tubo cam-
panulato segmentis ovatis, petalis oblongis calyce duplo longioribus, sta-
minibus inclusis, ovario hispido.

A shrub, with slender terete copiously lenticellate purplish
branchlets, pubescent towards the tip. Leaves 2-3 in. long in-
cluding the 1—1-in. petiole; rhachis winged between the leaflets,
which are 2-1 in. long. Flowers arranged in copious slender
spikes 2-3 in. long, with a pubescent rhachis. Calyx slightly
hairy, 4 line long. Petals obtuse, twice as long as the calyx.
Stamens shorter than the petals. Fruit not seen.— Baron
4434!

CRASSULA FRAGILIS, n. sp.

Annua, fragilis, pubescens, caulibus gracillimis dichotomiter furcatis,
foliis sessilibus oblongo-spathulatis acutis, floribus pentameris solitariis
pedicellis filiformibus elongatis, calyce campanulato piloso tubo brevissimo
segmentis lanceolatis, petalis rubris oblongis calyci paulo superantibus,
genitalibus calyci zequilongis.

A fragile annual, with erect dichotomously forked pubescent
stems not more than an inch long. Leaves crowded, opposite,
fleshy, } in. Jong, narrowed from above the middle to the base.
Flowers solitary, on terminal hairy pedicels j in. long. Calyx

470 MR. J. G. BAKER ON THE

Tz in. long. Petals reddish, oblong, alittle longer than the calyx.
Anthers small, red, globose. Carpels as long as the petals ; style
short.— Baron 3348! Belongs to the section Filipedes, as defined
in ‘ Flora Capensis.’

KALANCHOE BREVICAULIS, n. sp.

Perennis, glabra, caulibus brevibus flexuosis, foliis parvis confertis ses-
silibus carnosis obovato-cuneatis obtusis dimidio superiorl serratis, pe-
dunculo stricto brevi, floribus multis in paniculam terminalem modice
densam dispositis, pedicellis flore interdum eequilongis, calyce haud laxo
tubo brevi dentibus lanceolato-deltoideis acutis. corolle rubella tubo
infundibulari segmentis oblongis acutis tubo zquilongis, stylis brevibus.

A fleshy erect perennial, with tortuose branching stem under
half a foot long. Leaves crowded towards the top of the stems,
ascending, decussate, 3-1 in. long, cuneate and entire in the
lower half. Peduucle slender, naked, erect, 2-3 in. long. Pa-
nicle 1-2 in. in diam., dichotomous; branches and pedicels slender,
glabrous; lower 4 in. long. Flowers tetramerous, reddish.
Calyx i in. long. Corolla + in. long. Stamens reaching halfway
up the corolla-segments. Ovaries narrowed gradually into the
Short styles.— Baron 3542! Near K. pumila, Baker in Journ.
Linn. Soc. xx. p. 139, which is the same as K. multiceps, Baill. in
Bull. Linn. Soc. Par. 489, and Hildebrandt’s 3576.

KALANCHOE BRACHYCALYX, n. sp. :

Perennis (foliis ignotis), pedunculo elongato, floribus permultis in pani-
eulam densam corymbosam dispositis, panicule ramis pubescentibus,
pedicellis brevibus, calyce stricto tubo pubescente campanulato, dentibus
latis brevissimis, corolla rubre pubescentis tubo oblongo-cylindrico seg-
mentis parvis oblongis, stylis brevibus.

A perennial, with elon gated erect flower-stems, ending in dense
corymbose panicles, with flowers, pedicels, and branches all
covered with short drab pubescence. Pedicels at most as long
as the calyx. Calyx j, in. long and broad. Corolla 2 in. long;
tube j; in. in diam. ; segments oblong, acute, } as long as the
tube. Stamens and style protruded from the throat of the corolla-

tube. Ovaries glabrous, nearly as long as the corolla-tube.—
Baron 4319!

KALANCHOE PUBESCENS, n. sp. ;

Perennis, caule valido elongato pubescente, foliis petiolatis orbiculari-
bus crenatis carnosis, pedunculo elongato, floribus multis in paniculam
corymbosam dispositis, paniculæ ramis pubescentibus, pedicellis flori mter
dum zquilongis, calyce pubescente baud inflato tubo campanulato dentibus

FLORA OF MADAGASCAR. 471

deltoideis, corolle tubo infundibulari segmentis parvis oblongis, stylo
ovario zquilongo.

A robust erect perennial, with pubescent erect flowering-
stems 2 feet long. Leaves in distant pairs, with a petiole as
long as the blade, which is 2 in. long and broad. Panicle half
a foot broad, with erecto-patent main branches ; pedicels some-
times } in. long. Calyx 4 in. long and broad ; segments as long
asthe tube. Corolla 3 in. long. Stamens reaching out of the
corolla-tube.— Baron 3574!

KALANCHOE GOMPHOPHYLLA, n. sp.

Perennis, glabra, caulibus brevibus, foliis obovato-cuneatis sessilibus
subintegris, pedunculo elongato, floribus parvis permultis in paniculam
thyrsoideam ramis corymbosis dispositis, pedicellis flore brevioribus,
calyce campanulato tubo brevissimo segmentis deltoideis, coroll lutex
tubo ovoideo segmentis parvis ovatis, stylis brevibus.

An erect perennial, with flowering-stems about a foot long.
Lower leaves an inch long, arranged in decussate pairs, very
obtuse, cuneate in the lower half. Panicle of several dense
corymbs ; main branches erecto-patent, the lower subtended at
the base by small leaves. Calyx jin.long. Corolla iin. long,
the tube ys in. in diam.; segments 4—4 as long as the tube. Sta-
mens and style just exserted from the corolla-tube.— Baron 3543 !
Nearly allied to K. Hildebrandtii, Baill.

KALANCHOE INTEGRIFOLIA, n. sp.

| Perennis, glabra, caule brevi, foliis crassis sessilibus oblongo-cuneatis
obtusis integris, pedunculo stricto nudo elongato, foribus pro genere
phrvis in panieulam densam cymosam dispositis, ramis pubescentibus,
p edicellis brevibus, calyce stricto parvo tubo brevissimo segmentis ovatis,
cprolle segmentis oblongis tubo ovoideo zquilongis, stylis brevibus.
A perennial, with stiffly erect flowering-stems under a foot
ong. Leaves crowded, glabrous, very thick, 12 in. long, 3-3? in.
broad, narrowed gradually to the sessile base from above the
middle. Peduncle naked, stiffly erect, 6-7 in. long. Panicle
dichotomously branched, very dense, 11—2 in. in diam.; pedicels
35-3 in. long. Calyx jin. in diam., +4 in. long. Corolla dull
reddish, 3 in. long. Stamens reaching halfway up the perianth-
segments. Style as long as the ovary.— Baron 4377!

KALANCHOE (KirCHINGIA) SULPHUREA, n. sp.
Perennis, glabra, foliis caulinis sessilibus oblongis obtusis carnosis, flori
bus paucis laxe cymosis, pedicellis elongatis cernuis, calyce laxo tubo

472 MR. J. G. BAKER ON THE

campanulato dentibus deltoideis tubo æquilongis, corolla lutez tubo
cylindrico, segmentis latis brevibus, stylis elongatis.

A perennial, glabrous in allits parts. Leaves thick, fleshy,
sessile, oblong, entire. Cymes lax, few E pedicels very
slender, cernuous, j-j in. long. Calyx iin. long. Corolla an
inch long, with a cy minds icu tube 4 in. in diam., and 4 short seg-
mentsas broad as long. Stamens reaching halfway up the corolla-
segments. Ovaries cylindrical, narrowed gradually in a style 2 in.
long.— Baron 4180!

KALANCHOE (KITCHINGIA) STREPTANTHA, n. sp.

Perennis, robusta, glabra, foliis magnis sessilibus carnosis oblongo-
spathulatis acutis, pedunculo nudo elongato, floribus multis in paniculam
ramis corymbosis dispositis, pedicellis brevibus, calyce laxo magno tubo
oblongo, segmentis deltoideis, coroll luteæ tubo infundibulari medio con-
stricto, segmentis ovatis, stylis elongatis.

An erect perennial, with a leafy stem half a foot long. Leaves
in contiguous sessile decussate pairs, 5-6 in. long, 14-2 in. broad
at the middle, entire, acute, thick and fleshy in texture, narrov ed
gradually from the middle to a broad base. Peduncle naked,
stiffly erect, a foot long. Panicle 4-5 in. broad; pedicels 1-2 in.
long. Calyx $-i in. long; teeth 4 as long as the tube. Corolla
pale yellow, 13 in. long; expanded limb 4 in. in diam. Stamens
reaching halfway up the perianth-segments. Style above 3 im.
long.— Baron 4874!

KALANCHOE (KITCHINGIA) LAXIFLORA, n. sp.

Perennis, glabra, foliis oblongis obtusis carnosis crenatis, caule elongató,
floribus in paniculam laxain corymbosam dispositis, paniculz ramis prit-
mariis elongatis ascendentibus, pedicellis flexuosis elongatis cernuis, calyd e
laxo membranaceo rubeilo tubo campanulato dentibus deltoideis, coroll:
lutez tubo medio constricto, segmentis parvis ovatis, stylis elongatis.

A perennial, glabrous in all its parts, with flowering-stems
2-3 feet long. Petiole slender, an inch long; blade 12-2 in.
long, conspicuously crenate. Panicle a foot long, with 3 main
branches, each several times dichotomously forked ; pedicels very
slender, 3-3? in. long. Calyx j in. long. Corolla pale yellow,
twice as long as the calyx, ampulleform at the base, and the tube
dilated again above the middle; segments as broad as long.
Stamens as long as the corolla-tube. Styles 1 in. long.—Baron
4306 !

KALANCHOE (KITCHINGIA) SUBPELTATA, n. sp.
Perennis, glabra, caule elongato, foliis dissitis patulis longe petiolatis

FLORA OF MADAGASCAR. 473

ovatis obtusis crenatis petiolo supra basin insertis, pedunculo elongato,
floribus paucis in paniculam corymbosam dispositis, pedicellis elongatis
flexuosis, calyce laxo late campanulato dentibus late deltoideis cuspidatis,
corolla rubra tubo ampulleformi medio constricto segmentis brevibus
ovatis, stylis elongatis.

A glabrous fleshy perennial with elongated flowering-stems.
Leaves opposite, in distant spreading pairs, which grow gradually
smaller upwards; petiole patent, dilated at the base, lower
above an inch long; blade 13-2 in. long, truncate or slightly
cordate at the base, crenate all round the margin. Panicle 4-5 in.
in diam.; main branches arcuate; flowers in fascicles, subtended
by minute bracts ; pedicels 3 in. long. Calyx broadly campanu-
late, 4 in. in diam. Corolla bright red, above an inch long.
Stamens and styles reaching to the throat of the corolla-tube.
— Baron 3429! Near Kitchingia miniata, Baker.

DicorYPHE RETUSA, n. sp.

D. ramulis gracilibus sursum parce lepidotis, stipulis parvis deciduis,
foliis breviter petiolatis oblongis rigide coriaceis glabris obtusis vel retusis,
floribus 2-4 in capitula terminalia breviter pedunculatis aggregatis, brac-
teis minutis, calycis tubo infundibulari dense lepidoto, segmentis ovatis
tubo triplo brevioribus, petalis purpurascentibus patulis, fructu scabro
infra medium bifido.

A much-branched erect shrub or small tree. Petiole 3—4 in.;
blade 13-3 in. long, 3-17 in. broad, firm in texture, green and
glabrous on both surfaces, the main veins of the underside fine
and indistinct ; stipules oblanceolate, } in. long, similar in tex-
ture to the leaves. Peduncle very short. Flower-calyx 3 in.
long, plicate in the lower third of the tube, clothed with brown
lepidote indumentum. Petals dark purple, obtuse, protruded
lin. beyond the calyx. Anthers exserted a little from the calyx.
Fruit 4 in. long, with a short pedicel, and protruded à in. beyond
the enveloping base of the ealyx.— Baron 4408!

DicorRYPHE GUATTERLEFOLIA, n. sp.

D. ramulis gracilibus sursum parce lepidotis, stipulis parvis Janceolatis,
foliis breviter petiolatis ovatis vel oblongis obtusis rigide coriaceis utrinque
viridibus glabris, floribus 3-4 in capitula terminalia breviter peduncu-
latis aggregatis, bracteis parvis linearibus, calycis tubo oblongo dense
lepidoto segmentis ovatis tubo 3-4plo brevioribus, petalis oblongis, ovario
tomentoso.

An erect shrub or small tree, with slender branchlets, thinly
lepidote towards the tip. Stipules } in. long; petiole 3-3 in. ;

LINN. JOURN.— BOTANY, VOL. XXII. 2Q

474 MR. J. G. BAKER ON THE

blade 13-2 in. long, rounded at the base, subcoriaceous, with 7-8
pairs of slightly raised erecto-patent main veins joined by arches
a space from the edge. Flowers tetramerous, in shortly-pe-
duncled terminal clusters. Calyx i in. long, the tube plicated in
the lower third. Petals produced } in. beyond the calyx. Ovary
protruded a little from the persistent plicate base of the calyx.
Bracts persistent, 4—4 in. long.— Baron 4940!

DICORYPHE LAURIFOLIA, n. sp.

D. ramulis gracilibus glabris, stipulis parvis caducis, foliis breviter petio-
latis oblongis rigide coriaceis glabris obtusis vel emarginatis, floribus cir-
citer 3 in capitula terminalia breviter peduneulatis dispositis, fructu bifido
valvis apice protrusis.

An erect much-branched shrub or tree, with slender branchlets.
Petiole 75-4 in. long ; blade 2-3 in. long, 13-12 in. broad at the
middle, deltoid at the base, green and glabrous on both sides, the
main veins fine and inconspicuous beneath. Flowers about 3 in
peduncled terminal heads. Calyx and corolla not seen. Fruits
ovoid, 2 in. long, smooth, with two divergent valves protruded }

in. bevond the persistent base of the calyx, shortly pedicellate.—
Baron 4941 !

CALOPYXIS MALIFOLIA, n. sp.

C. ramulis gracilibus apice solum ferrugineo-pubescentibus, foliis petio-
latis orbicularibus vel oblongis acutis utrinque glabris, racemis densis bre-
vibus terminalibus, ovario breviter stipitato piloso, calycis limbo cuneato-
infundibulari glabro dentibus 5 deltoideis, staminibus breviter exsertis.

An erect shrub, with slender woody branchlets, pubescent only
towards the tip. Leaves 134-2 in. long, moderately firm in tex-
ture, green and glabrous on both surfaces; petiole 1-3 in. long.
Racemes dense, 1-13 in. long; rhachis pubescent. Ovary pilose,
gin. long. Calyx glabrous, 3 in. long ; teeth spreading or reflexing.
Stamens exserted a little beyond the calyx-teeth; anthers small,
ovoid, apiculate — Baron 4709! 4846! 4847!

Eveenta (§ Jossry1a) OLIGANTHA, n. sp.

E. ramulis gracilibus obscure pubescentibus, foliis petiolatis oblongis
acutis basi deltoideis subcoriaceis utrinque viridibus venis obscuris, flori-
bus paucis axillaribus, pedicellis flore longioribus, calycis tubo turbinato,
segmentis 4 oblongis obtusis inzqualibus, petalis patulis obovato-cuneatis,
staminibus petalis brevioribus.

A much-branched erect shrub or tree. Leaves 2-3 in. long,
about an inch broad at the middle, narrowed to both ends, firm

FLORA OF MADAGASCAR. 475

in texture, green and glabrous on both surfaces; petiole 4-7 in.

long. Flowers produced only from the axils of the upper leaves ;
pedicels 4-4 in. long. Calyx-limb jy in. long. Expanded
corolla jin. in diam. Stamens as long as the calyx-limb.— Zaron
3743 !.

EUGENIA ($ SYZYGIUM) AGGREGATA, n. sp.

Ramosissima, glabra, foliis petiolatis oblongis obtuse cuspidatis rigide
eoriaceis utrinque viridibus venulis omnibus crebris exsculptis, floribus
parvis dense eorymboso-paniculatis, pedicellis brevissimis, calycis tubo
obconico glabro nitido dentibus latis brevibus, petalis cohzrentibus, sta-
minibus permultis calyce longioribus.

A much-branched erect tree, glabrous in all its parts. Pe-
tiole 1-1 in. long; blade 11-2 in. long, narrowed to an obtuse
point, euneate at the base, thick and rigid, with all the veinlets
on both sides raised. Flowers very numerous, arranged in deuse
terminal eorymbose panicles. Calyx 4 in.in diam. Corolla-bud
semiorbicular. Stamens 4 in. long.— Baron 1115! 4321! 4917!
4939! Near E. tanalensis, Baker.

RuoposEPALA, genus novum Melastomacearum tribus
Osbeckie:e.

Calyx setosus, tubo campanulato, segmentis 5 ovatis membranaceis
saturate rubris deciduis, dentibus setiferis persistentibus alternantibus.
Petala 5, orbiculari-cuneata, purpurea. Stamina 10, æqualia, antheris
leviter undulatis apice poro unico dehiscentibus connectivo haud producto
basi antice bicalearato. Ovarium globosum, immersum, ovulis in loculo
permultis ; stylus elongatus subulatus sursum curvatus apice stigmatosus.
Capsula membranacea in calycis tubo perfecte immersa, dimidio superiori
libera. Semina cochleata minutissima copiosa.

Allied to Antherotoma and Dionychia.

RuoposEPALA PAUCIFLORA, Baker. Species sola.

Stems slender, herbaceous, spreading, sparsely bristly, un-
branched below the inflorescence. Leaves few, opposite, shortly
petioled, oblong, obtuse, distinctly triplinerved, $-1 in. long,
thinly bristly and green on both surfaces, moderately firm in
texture. Flowers few, sometimes one only, arranged in lax ter-
minal cymes, the lower branches from the axils of the upper
leaves. Calyx-tube } in. long and broad, thinly beset with
spreading whitish bristies, subtended at the base with a pair of
ovate obtuse dark purple deciduous bracteoles ; segments ovate,

bright red, pj; in. long. Petals j-Y in. broad. Stamens as long
2Q2

476 MR. J. G. BAKER ON THE

as the petals; anthers 4 in. long, bright yellow, with a pair of
small curved spurs from the base of the connective ; filament
nearly as long as the anther. Style finally 4 in. long.— Baron
4909 !

AMPHOROCALYX, genus novum Melastomacearum tribus
Oxysporee.

Calyx glaber, urceolatus, tubo verticaliter multistriatus, ore dentibus 4
brevissimis subdeltoideis instructo. Petala 4, oblongo-unguiculata. Sta-
mina 8, zqualia, antheris rectis apice l-porosis, connectivo basi breviter
producto postice bicalearato, filamentis brevibus deorsum applanatis.
Ovarium 4-loculare, in tubo profunde immersum, ovulis permultis ; stylus
elongatus, filiformis, sursum eurvatus, apice stigmatosus. Fructus inferus
indehiscens 4-locularis, seminibus permultis cuneatis.

Allied to Veprecella and Rousseauxia.

AMPHOROCALYX MULTIFLORUS, Baker. Species sola.

An erect shrub, with glabrous woody rather stout quadran-
gular stems. Leaves opposite, cordate-ovate, acute, 3-5 in. long,
2-3 in. broad, moderately firm in texture, distinctly triplinerved,
with a few bristles on the surface above and on the main ribs
below. Flowers in lax terminal panicles 4—5 in. long and broad,
with cymose branches, the lower subtended at the base by re-
duced stalked leaves; pedicels short. Calyx-tube lin. long and
broad, constricted to a distinct neck 4 in. in diam. ; vertical ribs
about a dozen; teeth very short. Petals deciduous, probably
yellow, 4 in. long. Anthers yellow, 4 in. long, with a couple of
spreading yellow spurs 4—4} line long from the back just below
the base ; filament shorter than the anther. Capsule completely
immersed in the calyx-tube below the neck. Seeds very numerous,
very minute.— Baron 4889 !

DICHETANTHERA CRASSINODIS, n. sp.

Arborea, ramulis validis glabris nodis incrassatis, foliis hysteranthiis
oblongis membranaceis hispidulis, floribus tetrameris in paniculas laxas
deltoidea dispositis, pedicellis elongatis erecto-patentibus supra medium
articulatis, ovario globoso glabro, calycis segmentis deltoideis ciliatis,
petalis purpurascentibus orbiculatis unguiculatis, staminibus 4 rubellis
connectivo infra antheram longe producto, 4 luteis connective breviter
producto.

A tree, with the flowers in lax deltoid panicles 3—4 in. long and
broad, at the end of the stout woody leafless branches ; pedicels
erecto-patent ; lower 1-3 in. long; bracts lanceolate or ovate-

FLORA OF MADAGASCAR. 477

lanceolate. Flower-calyx $ in. long and broad; calyx-segments
under a line long. Petals 1 in. broad, lin. long. Longer sta-
mens as long as the petals; anthers g in. long; connective pro-
duced 4 in. below the anther, bearing 2 yellow posticous spurs ;
smaller anthers with the spurs a short distance below the anther.
Style as long as the petals, narrowed to the base. Leaves pro-
duced after the flowers, oblong, obtuse, cuneate at the base, dis-
tinetly triplinerved, hispid, especially on the under surface.—
Baron 4825! 4925! Mr. Baron has again gathered D. oblongi-
Jolia, Baker= Baron 3838 !

VEPRECELLA BIFORMIS, n. sp.

Glabra, ramulis sterilibus longe sarmentosis, floriferis patulis, foliis bre-
viter petiolatis oblongis obtusis triplinerviis crenatis, paniculz laxe ramis
inferioribus cymosis paucifloris, pedicellis elongatis, calycis tubo globoso
glabro limbo angusto subintegro, petalis latis lilacinis, fructu capsulari 5-
valvi calyci zquilongo.

Long sterile branches climbing up trees perpendieularly or
along the ground from one tree to another, bearing copious
opposite oblong obtuse shortly petioled glabrous triplinerved
leaves often tinged with purple. Leaves of the flowering stems
longer than those of the creeping sterile shoots, 2-3 in. long.

I

Panicle lax, terminal, 5 ft. long; upper branches simple ; lower

cymose; pedicels 4-2 in. long. Fruit-ealyx } in. diam., strongly
ribbed; limb very narrow. Petals suborbicular, à in. in diam.
Capsule of 5 truncate valves not protruded beyond the calyx-
limb. Seeds minute, cochleate.— Baron 1054! 3992! The

sterile stems sent before by Dr. Miller, Dr. Parker, and
Mr. Pool.

GRAVESIA PORPHYROVALVIS, n. sp.

Herbacea, caule gracili decumbente hispidulo apice 1—3-floro, foliis
paucis longe petiolatis oblongis obtusis triplinerviis utrinque parce hispi-
dulis, calycis tubo globoso hispidulo, limbo brevi dentibus deltoideis, fructu
5-valvi apice rubro-purpureo e calyce protruso.

Stems slender, trailing, about a foot long. Leaves about
8 to a stem; petioles ascending, densely hispid, 14-2 in. long ;
blade membranous, 1-2 in. long, slightly cordate at the base,
green on both surfaces, obscurely crenulate, hispid mainly on the
veins beneath and towards the margin. Pedicels 1-2 in. long,
densely hispid. Calyx } in. in diam.; tube with 10 prominent
ribs; limb very narrow. Petal and stamens not seen. Capsule with

478 MR. J. G. BAKER ON THE

9 rigid bright red-purple valves protruding a little beyond the
ealy x-limb.— Baron 4469!

MEDINILLA DIVARICATA, n. sp.

Fruticosa, glabra, ramulis gracilibus, foliis subsessilibus oblongo-lanceo-
latis acutis subcarnosis triplinerviis basi sagittatis, floribus in cymas copiosas
axillares paucifloras dispositis, pedicellis elongatis gracilibus, calycis tubo
globoso glabro, limbo angusto truncato, petalis orbicularibus rubellis,
antheris antice bicalcaratis, postice unicalcaratis, fructu globoso carnoso
magnitudine pisi.

A shrub, glabrous in all its parts, with slender terete woody
branchlets. Leaves 2-3 in. long, rather fleshy, obscurely petioled,
shortly auricled at the base, triplinerved, with the lateral veins
springing from above the base of the midrib. Cymes pe-
duncled, not more than half as long as the leaves; pedicels erecto-
patent, 3-2 in. long. Fruit-calyx 4 in. in diam. Petals 4, 4 in.
long. Stamens 8, as long as the petals.— Baron 3658! Near
M. leptophylla, Baker.

MEDINILLA LINEARIFOLIA, n. sp.

Fruticosa, glabra, ramulis gracillimis, foliis subsessilibus linearibus uni-
nerviis subcarnosis basi sagittato-cordatis, floribus in cymas sessiles axil-
lares paucifloras dispositis, pedicellis fructui æquilongis, fructu globoso
carnoso magnitudine pisi.

A shrub, with very slender terete branchlets with swollen nodes.
Leaves opposite, ascending, 3—4 in. long, 4 in. broad, with only a
single central midrib and two minute auricles at the base. Cymes
sessile in the axils of the leaves ; pedicels 1 in. long, bracteolate

atthe middle. Petals and stamens not seen. Fruit 4 in. in diam.
— Baron 3844 !

RorALA CORDIFOLIA, n. sp.

R. caulibus czespitosis simplicibus glabris, foliis ternis sessilibus cordato-
orbicularibus superioribus approximatis, floribus tetrameris solitariis axil-
laribus sessilibus, ealycis tubo eampanulato demum scarioso dentibus bre-
vibus latis cuspidatis, petalis parvis oblongis unguieulatis, ovario globoso,
stylo elongato.

Stems simple, glabrous, cespitose, 6-8 in. long. Leaves in
whorls of three, $ in. long; lower internodes 3 in. long; upper
leaves imbrieated. Flowers solitary in the axils of the leaves all
down the stem. Calyx ;4 in. long and broad, S-ribbed. Petals
inserted at the throat of the calyx-tube between the teeth, oy in.
long. Stamens inserted low down in the calyx-tube; anthers

FLORA OF MADAGASCAR. 479

small, oblong. Ovary ovoid; style filiform, ;', in. long; stigma
capitate. Mature fruit not seen.— Baron 4313!

ASTEROPEIA SPH EROCARPA, n. 8p.

Fruticosa, ramulis glabris, foliis breviter petiolatis oblanceolato-oblongis
e medio ad basin sensim angustatis rigide coriaceis, floribus in paniculas
densas terminales aggregatis, pedicellis brevissimis glabris, sepalis oblan-
ceolatis obtusis rigide coriaceis, staminibus basi coalitis, fructu globoso
magnitudine pisi apice trilineato.

A shrub or small tree, glabrous in all its parts. Leaves alter-
nate, not crowded, thick and rigid, 13-2 in. long, 1-3 in. broad
above the middle, narrowed gradually from the middle to a short
petiole. Flowers in a dense panicle 13 in. in diam. at the end of
the branchlets ; pedicels short, glabrous. Sepals brown, rigid,
vertically ribbed, ji in. long. Petals deciduous (not seen).
Stamens 10, longer than the sepals, deltoid at the base and
shortly connate. Fruit brown, coriaceous, glabrous, 3-celled,
4 in. in diam., with 3 grooves, along which it splits, radiating from
its apex.—Baron 3401! Closely allied to 4. densiflora, Baker
in Journ. Bot. 1882, p. 19. Dr. Baillon refers the genus to

Ternstromiaces, and unites with it my Rhodoclada.

MODECCA HEDER.EFOLIA, n. sp.

Sarmentosa, cirrhifera, ramulis gracilibus lignosis teretibus, foliis longe
petiolatis membranaceis profunde trilobatis segmentis acutis, flore foemineo
solitario longe pedunculato, calycis tubo brevi campanulato segmentis 5
oblongis reflexis, petalis oblanceolatis obtusis calyce brevioribus, fructu
magno oblongo-fusiformi longe stipitato.

A large copiously cirrhiferous woody climber, glabrous in all
its parts. Cirrhi 3 ft. long, woody. Petiole 2 in. long; blade
5-5 in. long and broad, deeply 3-lobed ; central segment oblong,
narrowed to the base. Peduncle 4-5 in. long, with a cirrhus
below the solitary flower. Calyx 3 in. long. Fruit 5-6 in. long,
2 in. in diam., narrowed gradually into a gynophore an inch long.
— Baron 3875!

BEGONIA ($ QUADRILOBARIA) FRAGILIS, n. sp.

Acaulis, petiolis brevibus glabris, foliis cordato-orbicularibus membra-
naceis leviter repandis denticulatis, pedunculo stricto pedali, floribus in
eymas paucifloras corymbosas monoieas dispositis, bracteis obovato-spathu-
latis, perianthio masculo segmentis 2 cordato-orbicularibus, filamentis
brevibus liberis, ovario obiongo-triquetro alis tribus angustis inzqui-

latis.

480 MR. J. Gd. BAKER ON THE

A fragile annual, with a slender rootstock and very thin leaves.
Petiole 3-12 in. long; leaves usually two; blade 4-6 in. broad.
Peduncle 1-1 ft. long, stiffly erect, slender, fragile. Cyme
dichotomously forked, dense or rather lax; male flowers much
more numerous than female. Perianth-segments reddish, j in.
long and broad. Anthers obtusely apiculate, 4 in. long, three
times as long as the free filaments. Ovary 3 in. long; wings at
most 1 in. broad, confluent at the base of the ovary.— Baron 4677 !
Near B. nossibea, A. DC.

BEGoNra Baront, n. sp.

Perennis, glabra. caule elongato robusto ramoso, foliis magnis membra-
naceis inzquilateraliter cordato-orbicularibus serratis, cymis laxissimis
magnis dichotomiter ramosis, fructu trialato ala unica reliquis 3-4plo
latiori. :

Stem smooth, straight, woody, 1-1 in. in diam. Leaves 6-8 1n.
long and nearly as broad, green, membranous, rounded on one
side at the base, deeply cordate on the other. Cymes in the
fruiting stage 8-9 in. in diam. Flowers not seen. Capsule 3-
celled, oblong, 2 in. long; wings thin but firm in texture, flexible,
horizontally ribbed; one wing oblong, obtuse, 1-1} in. broad,
the two others 3-1 in. broad.— Baron 4353! Allied probably to
B. Lyallá, A. DC.

Panax ($ SPHEROPANAX) GOMPHOPHYLLA, D. sp-

Arborea, glabra, foliis magnis bipinnatis foliolis obovato-cuneatis cus-
pidatis subcoriaceis integris, paniculæ rhachi valida bipedali, ramis simpli-
citer umbellatis, pedunculis brevibus arcuatis, pedicellis brevissimis apice
articulatis, ovario ovoideo pentamero, stylis 5 subulatis patulis.

A tree, glabrous in all its parts. Leaves with a stout rhachis
2 feet long, with several distant pinne ; leaflets 2-3 in. long,
1-11 in. broad at the middle, firm in texture, green and glabrous
on both surfaces, with the main veins beneath fine and hardly at
all raised. Panicle with a stout straight woody axis 2 ft. long;
peduncles slender, 1-11 in. long; flowers a dozeu or more in an
umbel; pedicels 3 in. long. Ovary pentagonal, glabrous, j in.

long, crowned by the crenate calyx-limb. Styles à line long.—
Baron 3793!

MoNoPETALZX.

ScHISMATOCLADA TRICHOLARYNX, n. gp.

Arborea, glabra, stipulis deltoideis, foliis oblanceolato-oblongis acutis
brevissime petiolatis, floribus in cymas multifloras terminales dispositis,

FLORA OF MADAGASCAR. 481

calycis tubo campanulato, segmentis oblanceolatis magnis foliaceis, corollæ
tubo cylindrico apice dilatato intus dense piloso, segmentis parvis ovatis,
fructu clavato pene ad basin fisso.

A much-branched tree, glabrous in all its parts. Leaves 2-3
in. long, 3-# in. broad above the middle, moderately firm in
texture, green and glabrous on both surfaces, with fine slightly
raised veins. Cymes 1-1} in. broad. Calyx } in. long; teeth
3-4 times as long as the tube. Corolla}in. long; segments 1 in.
Capsule $ in. long.— Baron 3632! Near S. psychotrioides,
Baker.

Dawnats LYALLII, n. sp.

D. ramulis pubescentibus, stipulis deltoideis, foliis breviter petiolatis obo-
vatis subcoriaceis glabris, floribus in paniculas sparsas axillares folio multo
breviores dispositis, pedicellis brevibus, bracteis parvis deltoideis, calycis
tubo campanulato segmentis deltoideis, corollz tubo brevi infundibulari
segmentis oblongis tubo duplo brevioribus, fructu globoso glabro pæne ad
basin fisso.

An ereet shrub, with rather stout terete branchlets. Leaves 2-3
in. long, 13-14 in. broad, firm in texture, with 6-8 pairs of raised
arcuate main veins; petiole {-}in. Panicles erecto-patent, 1-14
in. long. Calyx 4 in. long. Corolla j in. long; throat of the
tube very hairy. Anthers halfaslong as the segments. Capsule
black, 4 in. long and broad.— Baron 1479! 3894! 3895! 3981!
Also Lyall 122! Hildebrandt 3004! Closely allied to the Mau-
ritian D. fragrans, Comm.

DANAIS NUMMULARIFOLIA, n. sp.

Fruticosa, ramulis pubescentibus, stipulis parvis deciduis, foliis breviter
petiolatis parvis orbieularibus subcoriaceis utrinque viridibus glabris, flo-
ribus in paniculas sparsas terminales ramis corymbosis dispositis, pedi-
cellis brevissimis, bracteis parvis deltoideis, calycis tubo campanulato
dentibus parvis deltoideis, corolle tubo cylindrico segmentis oblongis,
staminibus segmentis zequilongis.

A much-branched shrub, with slender terete, densely shortly
pubescent branchlets. Leaves about 3 in. long and broad, firm
in texture, green and glabrous on both surfaces. Cymes lax,
terminal, and from the axils of the upper leaves. Calyx jin.
long; teeth shorter than the tube. Corolla-tube 1 in. long;
segments 3-3 as long as the tube. Filaments filiform, inserted
at the throat of the corolla-tube, as long as the lanceolate
anthers. Fruit not seen.— Baron 3657! 3811! Near D. pu-
bescens, Baker.

482 MR. J. G. BAKER ON THE

PENTAS HIRTIFLORA, n. sp.

Fruticosa, ramulis dense pilosis, stipulis fimbriatis, foliis sessilibus ovato-
lanceolatis acutis facie tenuiter dorso dense pubescentibus, floribus in
eymas densas terminales aggregatis, pedicellis brevissimis, bracteis linea-
ribus, calycis tubo brevi eampanulato, segmentis 5 linearibus elongatis,
coroliæ tubo extus piloso ore dilatato intus hirsuto, segmentis brevibus
ovatis, staminibus pilosis, stylo antheris eminente.

Stems long, shrubby, densely pilose. Leaves thin in texture,
dark green above, pale green beneath, 2-3 in. long, about an inch
broad, with numerous parallel arcuate main veins. Cymes 13-2
in. broad; branches and pedicels very hairy. Calyx very hairy ;
segments equal, 1 in. long. Corolla-segments } in. long. Anthers
inserted below the throat of the corolla-tube; filaments very
short. Fruit not seen.— Baron 4732!

DIRICHLETIA INVOLUCRATA, n. sp.

Fruticosa, ramulis pilosis, floribus petiolatis ovatis acutis utrinque
pubescentibus, floribus in capitula globosa terminalia bracteis magnis
ovatis aggregatis, calyce piloso segmentis inzqualibus ovatis acuminatis
foliaceis, corolle tubo eylindrico piloso segmentis parvis oblongis, stami-
nibus limbo zquilongis.

A shrub, with densely pubescent slender terete branchlets.
Leaves 4-5 in. long, 2-21 in. broad, rounded at the base, thin in
texture, green on both surfaces; petiole under 4 in. long. Pe-
duncles pubescent, 1-14 in. long ; heads dense, about 4 in. in diam.;
bracts pilose; outer Zin. long. Calyx-segments 4, ovate-acumi-
nate, unequal, not becoming bright red and scariose, as in the
other two, the largest 4 in. long. Corolla-tube very hairy, 2 in.
long; segments j'j in. long. Stamens as long as the corolla-
lobes. Fruit not seen.— Baron 3979! Dirichletia is united by
Dr. Baillon with CarpAalea of Jussieu, founded on a Madagascar
plant.

DIRICHLETIA TERNIFOLIA, n. sp.

Arborea, stipulis tricuspidatis, foliis sessilibus ternatis oblongo-lanceo-
latis acutis utrinque glabris, floribus tetrameris in eymas densas termi-
nales aggregatis, calycis segmentis rigidulis ovatis inzequalibus saturate
rubris, corolla tubo cylindrico glabro ore intus dense piloso, segmentis
parvis ovatis, staminibus ex tnbo subexsertis.

A tree, with woody glabrous branchlets. Leaves all in threes,
2-3 in. long, 1-13 in. broad at the middle, narrowed gradually to
the base and point, moderately firm in texture, green and glabrous
on both surfaces. Cymes dense, 2-3 in. indiam. Calyx-limb of 4

FLORA OF MADAGASCAR. 488

bright dark red segments of firm texture, the largest 4 in. long,
3 in. broad. Corolla-tube cylindrical, above } in. long, densely

pilose at the throat.— Baron 4280!

DIRICHLETIA TRICHOPHLEBIA, n. sp.

Arborea, ramulis pubescentibus, stipulis deltoideis pilosis, foliis oppo-
sitis sessilibus oblongis acutis basi attenuatis facie glabris dorso ad venas
dense pilosis, floribus tetrameris in cymas densas terminales dispositis,
calycis segmentis oblongis rubris valde inzequalibus, corolla tubo cylindrico
piloso, segmentis parvis ovatis.

A tree, with slender pubescent branchlets. Leaves 14-2 in.
long, 2-14 in. broad at the middle, narrowed to both ends, green
and glabrous above when mature, dull green beneath, very hairy
on the raised veins. Flowers aggregated in dense terminal
cymes 2-3 in. broad; branches pubescent ; pedicels short. Calyx
with 4 bright red oblong very unequal segments, connate at the
base, the largest 3—3 in. long. Corolla-tube 2 in. long, dilated
at the apex, the throat inside very hairy. Stamens not exserted
from the corolla-tube. Style exserted beyond the corolla-seg-
ments.—Baron 110 ! 3999 ! 4512! 4554! Collected also by Bojer,
Wykeham Perry, and Hildebrandt 3059; and may perhaps be
conspecific with Carphalea Pervilleana, Baill. in Bull. Linn. Soc.
Par. p. 188.

GARDENIA SUCCOSA, n. sp.

Glabra, copiose gummifera, ramulis gracilibus, stipulis ciliatis, foliis
breviter petiolatis oblongis obtusis basi deltoideis venis primariis 7-8-
jugis erecto-patentibus parallelis, floribus solitariis axillaribus breviter
pedicellatis, calycis tubo brevi segmentis magnis oblanceolatis foliaceis,
corolhe segmentis orbicularibus tubo infundibulari 2-3plo brevioribus,
fructu globoso coriaceo glabro.

A much-branched erect shrub 8-12 feet high, glabrous in all
its parts. Leaves 14-2 in. long, rigid in texture, lucent, quite
glabrous, with 7-8 pairs of distinct parallel raised veins. Flowers
solitary, on short ascending pedicels from the axils of the leaves,
pentamerous. Calyx-tube j in. long; segments twice as long as
the tube. Corolla 4 in. long. Fruit rigid in texture, brown,
glabrous, 4 in. long and broad, containing a dense mass of flattened
seeds aggregated in a pulpy mass in each of the two cells —
Baron 4620! 4710! Also Hildebrandt 3465!

PLECTRONIA MICRANTHA, D. sp.
Inermis, glabra, stipulis lanceolatis deciduis, foliis petiolatis oblongis

481 MR. J. G. BAKER ON THE

obtuse cuspidatis subcoriaceis venulis gracilibus immersis, floribus penta-
meris in cymas copiosas multifloras axillares dispositis, pedicellis flore
longioribus, calyce late campanulato dentibus parvis deltoideis, corolla
tubo campanulato ore intus piloso, segmentis ovatis tubo æquilongis,
antheris parvis oblongis filamentis brevissimis.

A much-branched erect tree, glabrous in allits parts. Leaves
2-3 in. long, 1-11 in. broad at the middle, firm in texture, glossy
above, deltoid at the base. Flowers in umbels in the axils of the
leaves all down the branch; pedicels 4 in. long. Calyx-tube
llin.in diam. Corolla-tube scarcely 4; in. long. Stamens 4 as
long as the corolla-lobes. Fruit not seen. — Baron 3836! Near
P. buxifolia, Baker.

IxoRA PACHYPHYLLA, n. sp.

Glabra, stipulis deltoideis persistentibus, foliis breviter petiolatis lanceo-
latis crassis rigide coriaceis venulis gracilibus vix exsculptis, floribus parvis
pentameris in panieulas densas corymbosas terminales aggregatis, pedi-
cellis brevissimis, bracteis minutis deltoideis, calycis segmentis deltoideis
tubo campanulato brevioribus, corolle segmentis oblongis tubo infundibu-
lari duplo brevioribus, fructu ovoideo monospermo.

A tree, with straight slender woody branchlets, glabrous in all
its parts. Leaves 3-5 in. long, 2-1 in. broad above the middle,
narrowed gradually to the base, firm and thick in texture, green
and glabrous on both sides, with fine inconspicuous venation.
Cymes dense, terminal, level-topped, 14-2 in. broad. Calyx 7; in.
Ls Corolla $ in. long. Fruit brown, 3-1 in. in diam.— Baron
3772!

Psycuorria ($ GRUMILEA) REDUCTA, n. Sp.

Arborea, glabra, stipulis ovatis obtusis, foliis breviter petiolatis obovato-
cuneatis cuspidatis subcoriaceis venis primariis parallelis exsculptis, floribus
in paniculas parvas corymbosas terminales dispositis, pedicellis brevis-
simis, calycis tubo obconico limbo angusto dentibus brevibus latis, corollz
tubo infundibulari ore intus glabro segmentis ovatis, fzuctu ovoideo nigro
magnitudine pisi, albumine corneo ruminato.

A much-branched erect tree, glabrous in all its parts. Stipules
ovate, obtuse, large, deciduous. Leaves 14-2 in. long, thick and
rigid in texture, green and glabrous on both surfaces, with 6-8
pairs of raised parallel ribs. Calyx-tube } in. long. Corolla-
tube 3 in. long, $ in. in diam at the throat; segments 4 as long as
the tube. Stamens just exserted from the corolla-tube. Style
exserted from the corolla-tube, with two short spreading flattened
branches. Drupe black, 1 in. diam.— Baron 1029! 1778! 1784!

FLORA OF MADAGASCAR. 485

3760! 3997! 4016! 4018! 4399! Near P. mesentericarpa,
Baker.

PSYCHOTRIA ($ GRUMILEA) RETIPHLEBIA, n. sp.

P. ramulis tetragonis sursum pubescentibus, stipulis ovatis, foliis vix pe-
tiolatis subcoriaceis glabris oblanceolatis obtusis vel acutis venulis faciei
inferioris anastomosantibus, floribus in paniculas terminales ramulis capitatis
dispositis, calycis tubo obconico piloso dentibus deltoideis, corolla tubo
infundibulari, segmentis ovatis, fructu globoso magnitudine pisi, albumine
corneo ruminato.

A much-branched shrub or small tree. Leaves 11-21 in. long,
4-3 in. broad above the middle, firm in texture, green and glabrous
on both surfaces, with the veins beneath anastomosing in distinct
areole. Panicles small, level-topped, terminal, with the flowers
in dense clusters. Calyx 4 in. long. Corolla i in. long.
Drupe 4 in. in diam.— Baron 4285! 4908!

Psycuorrra ($ GRUMILEA) PARKERI, n. sp.

Arborea, glabra, stipulis deltoideis, foliis petiolatis obovato-oblongis
acutis subcoriaceis utrinque viridibus glabris venis exsculptis, floribus in
cymas terminales peduneulatas dispositis, ovario obconico, calycis limbo
patelleeformi vix dentato, corolle tubo infundibulari, segmentis brevibus
ovatis, fructu globoso magnitudine pisi, albumine corneo ruminato.

A tree 30-40 feet, glabrous in allits parts. Leaves 4-6 in.
long, 2-21 in. broad above the middle, firm in texture, green and
glabrous on both surfaces, with numerous parallel arcuate dis-
tinctly raised main veins. Cymes level-topped, terminal, dis-
tinctly peduncled, pedicels none. Calyx 4 in. long. Corolla
in. long; expanded limb jin. in diam. Drupe 4 in. in diam.—
Baron 1585! 3527! Also gathered in the forest of Andranga-
loaka by Dr. Parker.

GomPHocaLyx, genus novum Rubiacearum tribus
Spermacoce:e.

Flores hermaphroditi. Calycis tubus obconieus, vittis validis; limbi
dentes 7-8 deltoidei acuti patuli rigiduli. Corolla hypocrateriformis,
tubo cylindrico sursum infundibulari, segmentis 5 oblongis. Stamina 5
ad tubi faucem inserta, filamentis filiformibus brevibus, antheris oblongis.
Ovarium 2-locuiare, ovulis in loculo solitariis; stylus filiformis, pro-
funde bifidus, ramis falcatis. Fructus osseus indehiscens, limbo patulo
coronatus.

Allied to Hydrophylax and Ernodea.

486 MR. J. G. BAKER ON THE

GoMPHOCALYX HERNIARIOIDES, Baker. Species sola.

A much-branched diffuse annual, with the habit of a large Her-
niaria, glabrous in all its parts. Stems slender, 1-1 ft. long;
branches erecto-patent. Leaves opposite, subsessile, oblong-
lanceolate, acute, 1-2 in. long, moderately firm in texture, 1-
nerved, green and glabrous on both surfaces; stipules short,
membranous, connate, persistent, slightly toothed. Flowers
axillary, 1-3nate, subsessile. Flower-calyx obconie, 4; in. long,
with 7-8 teeth, each decurrent as a strong rib to the base of the
tube. Corolla-tube Lin. long ; segments 1-3 as long as the tube.
Stamens shorter than the corolla-segments. Style reaching to
the throat of the corolla-tube, branched from halfway down.
Capsule 4'; in. diam., crowned by the rigid spreading calyx-seg-
ments.— Baron 4585! 4782!

LECONTEA FARINOSA, n. sp.

Volubilis, ramulis gracilibus subtiliter pubescentibus, stipulis deltoideis,
folis longe petiolatis subcoriaceis cordato-ovatis acutis facie viridibus
tenuiter pubescentibus dorso dense persistenter albo-incanis, cymis axil-
laribus longe pedunculatis foliis magnis bracteatis, bracteolis linearibus,
calyce dense pubescente, tubo oblongo segmentis elongatis linearibus,
fructu ovoideo compresso lucido pubescente.

A woody climber, with slender terete stems. Leaves 2-3 in.
long, shallowly cordate at the base, cuspidate, densely clothed
beneath with thin white tomentum ; petiole an inch long. Cymes
overtopping the leaves, bracteated by a pair of reduced petioled
leaves. Calyx-segments } in. long. Corolla not seen. Fruit
şin. long, crowned by the persistent calyx-segments.—Baron
4932!

VERNONIA STENOCLINOIDES, n. Sp.

Fruticosa, ramulis albido-incanis, foliis breviter petiolatis coriaceis
oblongis obtusis integris facie viridibus dorso albo-incanis, capitulis tri-
floris dense corymboso-paniculatis, involucro infundibulari bracteis paucis
pallide brumneis adpressis scariosis, exterioribus gradatim brevioribus,
achenio glabro, pappo albido tlexuoso.

Branchlets very slender, woody, coated with white tomentum.
Leaves 1-14 in. long, firm in texture, cuneate at the base, bright
green above, white beneath, with fine immersed veins. Panicle
3-4 in. long and broad; branches densely corymbose, the lower
springing from the axils of large leaves. Involucre j in. long ;
bracts few, drab; outer ovate; inner lanceolate. Corolla and
pappus +5 in. long. — Baron 3737!

FLORA OF MADAGASCAR. 487

VERNONIA RHODOPAPPA, n. sp.

Fruticosa, ramulis albido-incanis, foliis breviter petiolatis oblongis
acutis integris coriaceis facie viridibus glabris dorso albo-incanis, capitulis
multifloris dense corymboso-paniculatis, involucro campanulato bracteis
imbricatis obtusis exterioribus dense pilosis, achenio glabro, pappo
rubello.

Branchlets short, spreading, coated with thin persistent white
tomentum. Leaves 12-2 in. long, $-1 in. broad, rounded at the
base, firm in texture, the veins beneath fine and inconspicuous.
Heads in congested corymbose panicles at the end of the short
branchlets. Involucre 4 in. long and broad; bracts firm in tex-
ture, drab, adpressed. Corolla and bright red pappus 4 in. long.
Achene only seen immature.—Baron 3608! Also Hildebrandt

3621!

VERNONIA BETONICEFOLIA, n. sp.

Herbacea, caule simplici flexuoso dense glanduloso-pubescente, foliis
alternis longe petiolatis cordato-ovatis profunde crenatis utrinque viridibus
pubescentibus, capitulis multifloris corymboso-paniculatis, paniculæ ramis
dense glanduloso- pubescentibus, involucro campanulato glanduloso pauci-
seriali bracteis rigidulis exterioribus ovatis interioribus lanceolatis, achenio
glabro, pappi setis albis flexuosis.

An erect herb, with green terete densely glandular-pubescent
stems. Lower leaves 3—4 in. long, 1j in. broad, narrowed gra-
dually to the point, deeply irregularly inciso-crenate, membra-
nous; petiole above an inch long; upper growing gradually
smaller. Heads in a lax corymbose panicle, with glandular-
pubescent branches and pedicels. Involucre } in. in diam. ; outer
bracts small, ovate; inner lanceolate, } in. long. Flowers pale
purple, a little exserted from the involucre. Pappus 4 in. long.
— Baron 4733! 4779!

Vernonia (§ STROBOCALYX) CAPREJEFOLIA, n. sp.

Arborea, ramulis dense pubescentibus, foliis alternis petiolatis ovatis
acutis integris subcoriaceis facie scabris dorso pubescentibus venulis ex-
sculptis, capitulis 5-floris dense corymboso-paniculatis, involucro campa-
nulato piloso pauciseriali deciduo, bracteis exterioribus ovatis intimis
oblongis, achenio glabro, pappi setis albidis flexuosis.

Branches woody, terete, densely pubescent. Leaves 2-3 in.
long, 13-1} in. broad, firm in texture, acute, broadly rounded at
the base; petiole pubescent, 4-3 in. long. Heads aggregated in
very dense terminal corymbs 2-3 in. in diam. Involucre j in.
in diam., very deciduous ; outer bracts small, ovate ; inner oblong.

488 MR. J. G. BAKER ON THE

Achene 4-angled, many-ribbed, drab, glabrous, å in. long. Pappus
l in. long.— Baron 3577 !

VERNONIA ($ STROBOCALYX) GRISEA, n. SP.

Arborea, ramulis pubescentibus, foliis alternis petiolatis oblongis acutis
dentatis subglabris utrinque viridibus, capitulis 7-8-floris copiose corym-
boso-paniculatis, involucro campanulato piloso bracteis pauciseriatis ovatis
vel oblongis obtusis, achenio glabro, pappo albido setis exterioribus pluri-
bus brevibus linearibus.

Branchlets woody, terete, grey, pubescent upwards. Leaves
3-4 in. long, 1-12 in. broad at the middle, very acute, thin in
texture, penninerved, pubescent on the midrib beneath. Panicle
4-6 in. broad; branches pubescent, the lower springing from the
axils of large leaves. Involucre } in. in diam. ; bracts few, deci-
duous, brownish black ; outer small, ovate ; inner oblong. Corolla
not seen fully developed. Pappus flexuose, 1 in. long.— Baron
3437 !

Vernonta (§ SrROBOCALYX) EXSERTA, n. sp.

Arborea, ramulis apice tenuiter adpresse pubescentibus, foliis alternis
breviter petiolatis lanceolatis serratis utrinque viridibus glabris, capitulis
5-6-floris copiose corymboso-paniculatis, involucro campanulato bracteis
pauciseriatis adpressis deciduis ovatis vel oblongis, floribus ex involucro
conspicue exsertis, achenio glabro, pappo flexuoso albido setis exterioribus
multis brevibus linearibus.

Branchlets woody, terete, obscurely pnbescent upwards.
Leaves 3—4 in. long, under an inch broad at the middle, narrowed
gradually to the acute apex and short petiole, obscurely serrulate,
thin in texture. Panicle terminal, 4-6 in. broad, the lower
branches springing from the axils of large leaves. Involucre j in.
in diam.; bracts greenish brown, rigid, slightly pubescent ; outer
small, ovate; inner oblong. Corolla reddish, 4 in. long. Flowers
twice as long as the involucre.— Baron 4364 !

CONYZA SERRATIFOLIA, n. sp.

Suffruticosa, caule glabro, foliis sessilibus lanceolatis. membranaceis
serratis glabris, capitulis parvis copiose corymboso-paniculatis, involucro
campanulato bracteis pauciseriatis lanceolatis glabris, floribus filiformibus
pluribus, tubulosis paucis, achenio glabro compresso facie uninervato,
pappo flexuoso albido.

An undershrub, glabrous in all its parts, with long virgate
terete stems. Leaves 3—4 in. long, 2-3 in. broad, narrowed gra-
dually to both ends, green and glabrous on both sides. Panicle

FLORA OF MADAGASCAR. 489

4-5 in. long. Involucre 4 in. in diam.; bracts moderately
firm in texture, green in the middle, pale at the edge. Achene
3 in. long. Pappus j in. long.— Baron 4960! Closely allied, as
is also C. amplexicaulis, to the Brazilian C. triplinervia, Less.

Conyza AMPLEXICAULIS, n. Sp.

Suffruticosa, caule virgato obscure piloso, foliis lanceolatis argute
serratis amplexicaulibus utrinque viridibus parce pilosis, capitulis parvis
multifloris copiose corymboso-paniculatis, involucro campanulato bracteis
pauciseriatis subzequalibus lanceolatis glabris, floribus filiformibus multis,
tubulosis paucis, achenio glabro, pappo rubello.

An undershrub, with terete green stems. Leaves 3-4 in.
long, 3-2 in. broad at the middle, narrowed gradually to both
ends, thin in texture, hairy mainly on the midrib beneath.
Panicle 3-4 in. broad, with branches from the axils of the upper
leaves. Involucre j in. in diam. ; bracts glabrous, moderately firm
in texture, green in the middle, pale on the edge. Achene not
seen fully mature. Pappus setose, $ in. long.—Baron 4036!

Conyza ELLIsH, n. sp.

Fruticosa, ramulis gracilibus sursum obscure pubescentibus, foliis
sessilibus oblanceolatis obtusis serratis glabris, capitulis parvis multi-
floris corymboso-paniculatis, involucro campanulato bracteis pauciseriatis
subzqualibus oblongo-lanceolatis, acheniis eompressis facie uninerviis,
pappo flexuoso albido.

A shrub with slender brown woody terete branches. Leaves
1-11 in. long, 3-2 in. broad, narrowed gradually from above the
middle to the base, moderately firm in texture, green and glabrous
on both surfaces, penninerved, with erecto-patent main veins.
Heads in small dense terminal corymbs. Involucre } in. in diam. ;
bracts green in the middle, with pale edges. Achene glabrous,
llinelong. Pappus j in. long.— Baron 4676! Gathered also

by Ellis and Lyall.

PsrADIA CUSPIDIFERA, D. Sp.

Fruticosa, volubilis, ramulis dense breviter pilosis, folis breviter
petiolatis oblongis acutis serratis penninerviis dense pilosis, capitulis dense
corymbosis, involucro campanulato piloso bracteis lanceolatis acutis
adpressis viridi carinatis, achenio glabro cylindrico, pappo flexuoso
pallide rubello.

A shrubby climber, with slender terete densely pilose branch-
lets. Leaves 1-2 in. long, j-1 in. broad at the middle, cuneate

LINN. JOURN.— BOTANY, VOL. XXII. 28

490 MR. J. G. BAKER ON THE

at the base, not at all rigid nor viscose, green and densely pilose
on both surfaces, conspicuously serrated with a little cusp to
each tooth. Capitula in dense corymbs either from the tips of
the branchlets or axils of the leaves. Involucre 4 in. long and
broad; bracts pauciserial, adpressed, pale green, with a dark
green keel. Corolla and pappus both 4'; in. long.— Baron 4356!
Also Hildebrandt 3538! from damp places in the north of
Betsileo-land.

PsrADIA STENOPHYLLA, n. sp.

Fruticosa, ramosissima, glutinosa, glabra, ramulis gracilibus sulcatis,
foliis subsessilibus linearibus serratis uninerviis, capitulis parvis multi-
floris dense corymboso-paniculatis, involucro campanulato bracteis pauci-
seriatis oblongis rigidulis adpressis, pappo albido flexuoso.

A shrub, with slender woody ascending branchlets. Leaves
ascending, rigid in texture, 11-2 in. long, 4-} in. broad at the
middle, narrowed gradually to both ends, minutely serrated.
Heads in dense corymbs at the ends of all the branchlets. Iu-
volucre 4 in. long and broad; bracts all greenish and obtuse,
the outer small. Achene glabrous. Pappus jl in. long.—
Baron 3549!

PsrADIA MODESTA, n. sp.

Fruticosa, ramulis apice breviter pubescentibus, foliis subsessilibus
oblanceolatis acutis serratis subglabris, capitulis parvis multifloris corym-
boso-paniculatis, involucro campanulato bracteis pauciseriatis rigidulis
adpressis exterioribus ovatis interioribus lanceolatis, achenio cylindrico
glabro, pappi setis albis flexuosis.

A shrub, with erect terete brown calvate main branches and
copious slender ascending branchlets, hairy towards the tip.
Leaves erecto-patent, 1-14 in. long, 4 in. broad, moderately firm
in texture, acute, penninerved, minutely serrated, not viscose.
Capitula arranged in small terminal panicles with corymbose
branches. Involucre glabrous, brownish, } in. long and broad.
Achene cylindrical, glabrous. Corolla i in. long. Style-arms

8
linear, exserted from the corolla.— Baron 4775!

GNAPHALIUM DIFFUSUM, n. sp.

Herbaceum, perenne, albo-lanuginosum, caulibus brevibus diffusis, foliis
parvis sessilibus obovatis, capitulis paucis multifloris ad ramorum apices
glomeratis, involucro campanulato bracteis pauciseriatis oblongis scariosis,

FLORA OF MADAGASCAR. 491

floribus filiformibus pluribus, tubulosis hermaphroditis paucis, pappo albo
flexuoso.

A. perennial, with densely cespitose stems, many short and
sterile, the flowering ones branched, spreading, 2-3 in. long.
Leaves alternate, obovate-cuneate, 4 in. long, persistently matted
with white tomentum on both sides. Heads } in. in diam., densely
tomentose at the base; bracts oblong, Bu brown. Achenes
subeylindrieal, brown, glabrous. Corolla 4—4 in. long. Pappus
as long as the corolla, the bristles falling singly.— Baron 3602 !

HELICHRYSUM ARANEOSUM, n. sp.

Fruticosum, ramulis gracilibus albo-incanis, foliis petiolatis oblongis
obtusis basi cuneatis facie viridibus obscure pubescentibus dorso tenuiter
albido-incanis, capitulis 6-8-floris in panieulam terminalem ramis corym-
bosis dispositis, pedicellis brevibus albo-lanosis, involucro infundibulari
bracteis multiseriatis albo-rubellis obtusis squarrosis deorsum viridi
carinatis, pappo albo setoso fragili.

A shrub, with slender terete woody branchlets, clothed with
white tomentum. Leaves 2-3 in. long, 1-14 in. broad below the
middle, moderately firm in texture, with very distinct ascending
main veins. Panicle 4-5 in. in diam.; branches leafy, clothed
with thick white tomentum. Involucre j in. long, } in. in
diam., lanuginose in the Jower half; bracts in many rows, outer
gradually shorter. Corolla $ in. eae Pappus as long as the
eorolla.— Baron 4894! Nearly allied to H. farinosum.

HELICHRYSUM FARINOSUM, D. sp.

Fruticosum, ramulis robustis dense albo-lanuginosis, foliis breviter
petiolatis. subcoriaceis ovatis facie tenuiter dorso dense albo-incanis,
capitulis 6-8-floris dense corymboso-paniculatis, panicule ramis albo-
lanuginosis, pedicellis subnullis, involuero obiongo bracteis multiseriatis
oblongis obtusis albo-rubellis squarrosis deorsum viridi carinatis exteri-
oribus dorso lanuginosis, pappo albo setoso.

An erect shrub, with terete tomentose leafy branchlets i-i in.
in diam. Leaves 13-2 in. long, 1-1 in. broad, densely lanuginose
below, obtuse, rounded at the base, with very ascending main
veins. Corymbs dense, ege 2-3 in. in diam. Involucre
nearly 4 in. long, } in. in diam.; outer bracts gradually shorter,
the inner with large Errem M obtuse spreading tips. Co-
rollas 4 in. long. Pappus as long as the corolla.— Baron 4926!

HELICHRYSUM AMPLEXICAULE, n. Sp.
Herbaceum, perenne, ramis gracilibus albo-incanis, foliis caulinis crebris

lanceolatis amplexicaulibus facie viridibus subclavatis dorso persistenter
9n 9

492 MR. J. G. BAKER ON THE

albo-incanis, capitulis parvis 8-10-floris dense corymboso-paniculatis,
pedicellis brevibus lanuginosis, involucro oblongo bracteis multiseriatis
oblongis obtusis luteis adpressis glabris dorso convexis, pappo albo
flexuoso.

A herbaceous perennial, with slender terete stems, clothed with
white tomentum. Lower stem-leaves about an inch long, 4 in.
broad, crowded, dilated at the base; upper shorter. Corymbs
1-3 in. broad, with ascending tomentose branches; heads very
numerous. Involucre } in. long, under j in. in diam. Flowers
yellow, à in. long. Pappus as long as the corolla.—Baron
4964! Near H. triplinerve, DC.

HELICHRYSUM PLATYCEPHALUM, n. sp.

Herbaceum, perenne, caulibus erectis dense cæspitosis, sterilibus
brevibus crebre foliatis, foliis linearibus uninerviis utrinque persistenter
albo-incanis, floriferis elongatis 1-3-cephalis foliis dissitis reductis sessili-
bus ascendentibus, involucro magno late campanulato bracteis ovatis
acutis multiseriatis scariosis glabris subæquilongis adpressis, pappo albo
flexuoso.

An erect densely cæspitose perennial, with slender erect
sparsely leafy 1-3-headed flowering-stems under a foot long aud
numerous short sterile branches. Leaves linear, persistently
matted with white on both sides, those of the sterile stems

3-2 in. long, of the flower-stems smaller and suberect. In-
voluere 3 in. in diam., not more than 4 in. long; bracts glossy,
bright yellow, scariose and glabrous down to the base. Flowers
100 or more to a head. Receptacle convex, 4 in. in diam. Flowers
$in.long. Pappus as long as the corolla.— Baron 3485!

APHELEXIS FLEXUOSA, n. sp.

Fruticosa, ramosissima, ramulis gracilibus divaricatis albo-lanuginosis,
foliis lanceolatis rigidulis adpressis dorso acute carinatis, capitulis 10-12-
floris ad ramulorum apices paucis glomeratis, involucro obconico bracteis
scariosis longe unguiculatis apicibus minutis oblongis brunneolis, pappi
setis albidis flexuosis.

A much-branched shrub, with terete woody main stems and
more or less ascending slender branchlets coated with white
tomentum. Leaves adpressed to the branchlets, l in. long.
Heads in very numerous clusters at the tips of all the branchlets.
Involucre $ in. long; bracts glossy, drab-brown, with very small
spreading tips. Corolla 3 in. long. Achene brown, glabrous.
Pappus as long as the corolla— Baron 3510! This and the two
next are allied to A. selaginifolia, DC.

FLORA OF MADAGASCAR. 493

APHELEXIS STENOCLADA, n. sp.

Fruticosa, ramosissima, ramulis confertis gracilibus erectis albo-lanu-
ginosis, foliis lanceolatis rigidulis adpressis dorso acute carinatis, capitulis
6-8-floris paucis ad ramulorum apices dense corymbosis, involucro infun-
dibulari bracteis longe unguiculatis apicibus patulis oblongis sulphureis,
pappi setis albidis flexuosis.

A shrub, with slender densely-crowded ascending ultimate
branchlets, which with the adpressed leaves do not exceed half a
line in diameter. Leaves 4 in. long, imbedded in the white
tomentum that clothes the branchlets. ^ Capitula in corymbs
at the ends of all the branchlets. Involucre J in. long, com-
posed of 15 to 16 scariose bracts with minute spreading tips.
Achene glabrous. Corolla j in. long. — Pappus-bristles falling

singly.— Baron 3464!

APHELEXIS SULPHUREA, n. sp.

Fruticosa, ramosissima, ramulis confertis gracilibus erectis albo-lanu-
ginosis, foliis lanceolatis rigidulis adpressis dorso convexis, capitulis 8-10-
floris paucis ad ramulorum apices corymbosis, involucro infundibulari
bracteis longe unguiculatis apicibus patulis ovatis sulphureis, pappi setis
albidis flexuosis.

A shrub, with woody main stems and crowded branchlets,
which with the adpressed leaves are half a line in diameter.
Leaves 4 in. long, imbedded in the white tomentum of the branch-
lets. Capitula in corymbs at the ends of all the branchlets. In-
voluere % in. long; bracts 10-12, sulphur-yellow, with ovate
spreading tips half a line long. Achene glabrous. Corolla and
pappus both 3 in. long.— Baron 3459!

ASTEPHANOCARPA, genus novum Compositarum
tribus Inuloide:. !

Capitula 2-3-flora homogama discoidea. —Involuerum cylindricum
deorsum lanuginosum, bracteis pauciseriatis oblongo-lanceolatis scariosis
sursum albis glabris. Receptaculum nudum. Corolle tubulose luteze
involucro breviores, apice breviter quinquefidz, segmentis ovato-lanceo-
latis. Styli rami elongati faleati apice truncati. Antherz basi sagittate,
auriculis caudatis. Achenia cylindrica glabra, apice calva.

Habit of Stenocline, from which it differs by being entirely
destitute of pappus.

ASTEPHANOCARPA ARBUTIFOLIA, Baker. Species sola.
A shrub, glabrous in all its parts, with slender long stiff terete
ultimate branchlets, closely leafy to the top. Leaves rigid,

494 MR. J. G. BAKER ON THE

sessile, ascending, oblanceolate, 3—1 in. long, obscurely crenulate,
brown when dried, with the copious reticulated veinlets raised
on both surfaces. Heads very numerous, arranged in corymbose
terminal clusters, bracteated by reduced ovate coriaceous leaves.
Involucre nearly } in. long, the scariose bracts brownish and
woolly in the lower half, pure white and glabrous in the upper
half. Corolla 4 in. long. Achene much shorter than the corolla.
— Baron 3488!

STOEBE CRYPTOPHYLLA, D. sp.

Fruticosa, ramosissima, ramulis confertis gracilibus erectis albo-tomen-
tosis, folis minutis lanceolatis rigidis ad ramum adpressis, capitulis
spicatis erecto-patentibus, involucro cylindrico multiseriali bracteis exteri-
oribus oblongis rigidis interioribus lanceolatis corneis, pappo albido
plumoso flori xequilongo.

A much-branched erect heath-like shrub ; branchlets, including
the adpressed leaves, under a line in diameter. Leaves nearly
black, 4; in. long, quite adpressed to the white tomentose
branchlets. Spikes 4-1 in. long. Involucre } in. long; outer
rows of bracts similar to the leaves in borne but smaller;
inner horny, pale brown, glabrous. Flower solitary, as long as
the involucre. Corolla and pappus d in. long.— Baron 3349!
Adds this characteristically Cape genus to the Madagascar flora.
About 30 species are known at the Cape and 1 in Bourbon.

STOEBE BIOTOIDES, n. sp.

Fruticosa, ramosissima, ramulis confertis gracilibus erectis albo-tomen-
tosis, foliis minutis rigidulis lanceolatis dorso convexis ad ramum adpressis,

capitulis spicatis erecto- -patentibus, involucro cylindrico multiseriali bracteis
exterioribus parvis oblongis rigidis interioribus corneis ovatis vel lanceolatis;
pappo albido plumoso flori zequilongo.

An ericoid shrub, with woody main stems } in. in diam., and
very numerous crowded ascending branchlets a line in diameter,
including the adpressed leaves. ee rigid, brown-black, yy in.
long. Spikes about 2 in. long. Involucre + in. long, with a
br outer bracts similar in texture to the ees, the horny inner
ones much longer, drab-brown. Achene castaneous, glabrous,
is in. long. Pappus and corolla 1 à in. long.— Baron 3504!

EPALLAGE DISSITIFOLIA, n. sp.

E. ramulis dense breviter pubescentibus, foliis paucis remotis petiolatis
ovatis serratis, capitulis paucis longe pedunculatis, involucro late cam-
panulato bracteis herbaceis oblongis acutis adpressis pubescentibus, ligulis

FLORA OF MADAGASCAR. 495

multis brevibus, disco paleis lanceolatis rigidis, achenio glabro, pappo
paleaceo.

A perennial herb, with slender terete pubescent ascending
branchlets. Leaves small, distant, crenate, pubescent. Heads
depresso-globose, 3 in. in diam. JInvolucre 4 in. long, densely
hairy ; bracts broad, greenish, foliaceous. Pales of the disk rigid,
lanceolate, } in. long. Achene subcylindrical, glabrous, p in.

long; pappus coroniform, paleaceous.—Baron 4743!

TEMNOLEPIS, genus novum Compositarum tribus Helianthoidez.

Capitula multiflora homogama, floribus omnibus tubulosis hermapbro-
ditis. Involucrum campanulatum, bracteis pauciseriatis oblongis acutis
rigidulis adpressis. Receptaculum planum, paleis rigidulis laceratis flori-
bus equilongis preeditum. Corolla tubulose, apice breviter quinquefidze.
Anthere basi haud caudate. Styli rami breves complanati. Achznia
glabra cylindrica quadrangula, pappo obscuro coroniformi minute dentato.

Allied to Epallage, from which it differs by its homogamous
flowers, large lacerated bracts of the receptacle, and obscure
minutely-toothed pappus.

TEMNOLEPIS SCROPHULARIHFOLIA, Baker.

An erect much-branched annual, with slightly pubescent
slender stems, 1-2 ft. long. Leaves few, alternate ; petiole 2-1 in.
long; blade cordate-deltoid, obtuse, coarsely irregularly crenate,
1-11 in. long, thin in texture, green on both sides, obscurely
hispid. Heads few, laxly corymbose at the end of the numerous
branches; peduncles long, erect, glandular-pubescent, generally
naked. Involucre campanulate, 4 in. in diam. ; pales few, rigid,
oblong, subequal, pubescent, obtuse or cuspidate. Flowers each
subtended by a tough whitish bract, lacerated at the apex, as long
as the involucre and flower. Corolla yellow, subcylindrical, py
in. long, glandular downwards, constricted above the base. Styles
not exserted. Achene as long as the corolla, clavate, greenish,
glabrous, 4-angled, the pappus only an obscurely minutely toothed
ring.— Baron 4303 !

GYNURA SONCHIFOLTA, n. sp.

G. caule brevi dense piloso, foliis membranaceis lyrato-pinnatis, pedunculo
elongato glanduloso-piloso, capitulis parvis copiose corymboso-paniculatis,
panicule ramis dense pilosis, involucro oblongo bracteis linearibus, flori-
bus involucro equilongis, achenio glabro cylindrico, pappo albo fragili.

A perennial herb, with short densely pilose erect simple stems.
Petiole 1-2 in. long; blade consisting ofa large ovate terminal

496 MR. J. G. BAKER ON THE

segment 2-3 in. long and broad, and a single pair of small lateral
segments. Peduncle slender, fragile, a foot long, with a short
branch from the middle, and a lax end panicle with corymbose
branches densely clothed with soft short crisped hairs. Involucre
3 in. long, pilose like the pedicels; bracts about 8, with pale
edges. Achene din. long, with about 10 ribs. Pappus copious,
soft, deciduous, 1 in. long.— Baron 3426 !

CixERARIA ANAMPOZA, Baker,=Senecio Anampoza, Baker in
Journ. Linn, Soc. xx. 191.

Sent now (Baron 4254!) with mature achenes. They are
flattened and winged, and therefore it must be transferred to the
genus Cineraria as defined in‘ Genera Plantarum,’ to the vicinity
of the Cape C. geraniifolia, DC. Adds another characteristic Cape
genus to the Madagascar flora.

SENECIO ACETOSEFOLIUS, n. sp.

Sarmentosus, glaber, foliis petiolatis subcarnosis inferioribus hastatis
superioribus ovatis, capitulis parvis radiatis paucifloris dense corymboso-
paniculatis, involucro infundibulari bracteis 5 lanceolatis, ligulis luteis,
floribus discoideis involuero longioribus, achenio piloso, pappo albo
flexuoso.

A climber, with slender terete glabrous stems. Leaves distant,
alternate, bright green, glabrous, lower 2 in. long and broad, with
spreading deltoid auricles and a petiole as long as the blade.
Panicles terminal on the branches, 3-4 in. broad, made up of
several corymbose branches ; pedicels as long as or shorter than
the heads, copiously bracteate. Involucre 4 in. long. Flowers
about 6 in a head, 3 ligulate and 3 discoid. Pappus and disk-
flowers 4 in. long.— Baron 4357! 4360! Belongs to the group

Scandentes as defined in ‘Flora Capensis, near S. tamoides of
Natal.

SENECIO ($ KrkrNOIDEI) OYCLOCLADUS, n. sp.

Fruticosus, carnosus, glaber, caulibus flexuosis dichotomiter ramosis,
foliis parvis confertis oblanceolatis sessilibus uninerviis, pedunculo elongato
nudo oligocephalo, capitulis radiatis multifloris, involucro campanulato
bracteis 10-12 lanceolatis, ligulis luteis involucro zequilongis, floribus dis-
coideis involucro superantibus, achenio glabro, pappo albo fragili.

Stems woody, flexuose, two or three times dichotomausly forked.
Young leaves erowded towards the ends of the branches, very
fleshy, 1-12 in. long, $ in. broad. Peduncles slender, stiffly erect,
3-headed ; pedicels 3-13 in.long, minutely bracteate. Involucre

FLORA OF MADAGASCAR. 497

l in. long and broad. Ligules about 8, 4 in. long. Pappus and
disk-flowers 4 in. long. Achene not seen fully developed.— Baron
4272!

SENECIO (§ KLEINOIDET) MELASTOMJEFOLIUS, n. sp.

Fruticosus, carnosus, glaber, ramis cylindricis, foliis sessilibus carnosis
oblongis acutis integris confertis, pedunculo gracili oligocephalo, capitulis
radiatis multifloris, involucro campanulato bracteis 10-12 lanceolatis,
ligulis luteis, pappo albo flexuoso.

Stems short, thick, fleshy, curved, forked, bare except at the
tip. Leaves crowded at the tips of the branches, ascending, 1—
1i in. long, 1-4 in. broad, distinctly triplinerved from the base
upwards. Peduncles slender, naked, 2-3 in. long, 3-4-headed.
Involucre i in. in diam. Ligules 8-10, linear. Disk-flowers not
exserted from the involucre. Pappus 4 in. long.— Baron 3569!

SENECIO (KLEINOIDEI) VERNICOSUS, n. sp.

Fruticosus, carnosus, glaber, foliis confertis sessilibus oblanceolatis
obtusis integris ascendentibus trimervatis, capitulis radiatis multifloris
corymboso-panieulatis, involuero campauulato bracteis 9-10 lanceolatis,
ligulis parvis luteis, floribus discoideis, involucro duplo longioribus, achenio
hispidulo, pappo albido fragili.

Leafy stems woody, 4—4 in. in diam. Leaves 3-4 in. long, $-3
in. broad, shining as if varnished, narrowed gradually from the
middle to the base, triplinerved down to the base. Peduncles
naked, 3-4 in. long, including the corymb ; ascending minutely
bracteated peduncles longer than the heads. Involucre l in.
long and broad. Ligules small, yellow. Achene cylindrical, à

in. long. Pappus soft, white, à in. long.— Baron 3563 !

SENECIO (§ KLEINOIDEI) CICATRICOSUS, n. sp.

Fruticosus, carnosus, glaber, ramis furcatis, foliorum delapsorum basibus
rugosis, foliis confertis sessilibus ascendentibus lineari-subulatis, pedunculo
nudo brevi oligocephalo, capitulis multifloris radiatis, involucro campanu-
lato bracteis circiter 8 basi connatis, ligulis patulis luteis, achenio glabro,
pappo albo fragili.

Stems woody, cylindrical, 2 ft. long, simple or once or twice
dichotomously forked. Young leaves crowded towards the tips of
the branches, very fleshy, acute, 1-14 in. long. Peduncles slender,
stiffly erect, 1-4 in. long, 1-7-headed. Involucre 4 in. in diam.,
glabrous. Ligules about 8, 4 in. long. Disk-flowers rather
longer than the involucre. Achene brown, cylindrical, gin. long.
Pappus and disk-corollas l in. long.— Baron 3564! Near SV.
canaliculatus, Bojer.

498 MR. J. G. BAKER ON THE

SENECIO (§ KLEINOIDEL) MONOCEPHALUS, n. 8p.

Fruticosus, carnosus, ramis crassis foliorum delapsorum basibus rugosis,
foliis confertis sessilibus ascendentibus lineari-subulatis carnosis glabris,
pedunculo brevissimo monocephalo, capitulis radiatis multifloris, involucro
campanulato piloso, bracteis lanceolatis, ligulis parvis luteis, floribus dis-
coideis involucro zquilongis, achenio glabro, pappi setis albidis flexuosis.

A fleshy erect undershrub a foot high, with stout branchlets
rough with the raised bases of the old leaves. Young leaves
crowded towards the ends of the stems, 1 in. long, 4; in. in diam.,
acute. Peduncle not longer than the leaves. Involucre cam-
panulate, 4 in. long and broad; bracts about 12 and a few small
ones at the base. Ligules spreading, 4 in. long. Achene only
seen immature. Pappus 4 in. long.— Baron 3961 !

Senecio (§ KLENTA) HILDEBRANDTII, n. sp.

Fruticosus, carnosus, glaber, ramis crassis deorsum foliorum delapsorum
basibus rugosis, foliis confertis ascendentibus subulatis carnosis, capitulis
parvis discoideis copiose corymboso-paniculatis, pedunculis bracteatis, invo-
lucro infundibulari bracteis 5 lanceolatis, floribus involucro sesquilongiori-
bus, achenio glabro, pappo albo fragili.

Stems 1-13 ft. long, simple or dichotomously forked, leafy in
the upper half. Leaves crowded, ascending, very fleshy, 2-3 in.
long, under a line in diameter. Panicles dense, level-topped,
2—4 in. in diam. Involucre } in. long; pedicels with lanceolate
bracts at the middle and base. Flowers all discoid, 5 in a head.
Achene only seen immature. Pappus and corolla 4 in. long.—
Baron 3562! 4218! Also Hildebrandt 3628! From Andranga-
loaka, on sunny hill-sides.

GERBERA EMIRNENSIS, n. sp.

Perennis, rhizomate dense piloso, foliis petiolatis oblanceolato-oblongis
dentatis subcoriaceis utrinque glabris, pedunculo gracili stricto elongato
pubescente, involuero campanulato bracteis paucis lanceolatis pubescen-
tibus, ligulis exsertis, floribus discoideis involucro sequilongis, achenio
cylindrico piloso, pappo albido setoso.

A perennial, with a thick short erect pilose rootstock, and a
rosette of radical leaves. Leaves 12-2 in. long, 4-2 in. broad,
narrowed gradually to the base, moderately firm in texture, dark
green; petiole pilose, 3-1 in. long. Peduncle slender, erect,
densely pubescent, about 4 in. long. Involucre 4 in. long and
broad; outer bracts gradually shorter. Receptacle naked,

FLORA OF MADAGASCAR. 499

slightly convex, j in. in diam. Flowers yellow. Achenes only

seen immature. Pappus } in. long; bristles white, flexuose.—
Baron 3457 !

PHILIPPIA CRYPTOCLADA, n. Sp.

Ramosissima, glabra, ramulis gracillimis erectis, foliis minutis rigidis
oblongis ternis ad ramulum adpressis, floribus 1—3nis terminalibus, calycis
segmentis ovatis, corolle tubo campanulato segmentis brevibus latis,
staminibus inclusis connatis, stylo incluso stigmate peltato.

An erect shrub, with slender crowded erect branchlets, hidden
by the adpressed leaves, not more than half a line in diameter,
leaves included. Leaves green, oblong, rigid, glabrous, 4, in.
long, quite adpressed to the stems, the upper whorls imbricated.
Flowers few and inconspicuous. Calyx similar to the leaves in
texture. Corolla 3 lin. in diam. Stamens and style not pro-
truded from the corolla.— Baron 3499! Differs from all the
following by its erect adpressed leaves.

PHILIPPIA CAPITATA, n. sp.

Ramosissima, ramulis pallidis dense hispidis, foliis quatetnis lanceo-
latis rigidulis ascendentibus parce glanduloso-hispidulis marginibus late
revolutis, floribus dense capitatis, calycis segmentis ovatis, corolla cam-
panulatze segmentis orbicularibus, staminibus liberis inclusis, stylo longe
exserto, stigmate magno peltato.

A much-branched shrub, with slender pale densely hispid
branchlets. Leaves + in. long, pale green, obtuse, more or less
ascending. Flowers in dense clusters at the ends of the branch-
lets; pedicels as long as the corolla. Calyx half as long as the
corolla. Corolla red, ;; in. in diam. Pistil twice as long as the
corolla.— Baron 3481 !

PHILIPPIA HISPIDA, n. 8p.

Ramosissima, ramulis pallidis dense hispidis, foliis parvis oblongis
rigidulis patulis quaternis dense glanduloso-hispidis marginibus late
revolutis, floribus dense capitatis, calycis segmentis ovatis obtusis, coroll:
campanulate segmentis latis obtusis, staminibus breviter exsertis antheris
apice liberis deorsum leviter coalitis, stylo breviter exserto stigmate haud
peltato.

A much-branched shrub, with drab branchlets densely clothed
with very unequal stiff spreading hairs. Leaves ;'5 in. long,
densely clothed with stiff spreading glandular bristles. Flowers
in dense clusters at the ends of the branchlets; pedicels as long

500 MR. J. G. BAKER ON THE

as the corolla. Calyx densely pilose, half as long as the corolla.
Corolla 4; in. long and broad. Anthers 8, short, slightly ex-
serted. Style just overtopping the anthers.— Baro 3335!

PHILIPPIA TRICHOCLADA, n. sp.

Ramosissima, ramulis pallidis erectis pubescentibus, foliis quaternis
oblongis rigidulis erecto-patentibus parce glanduloso-hispidis marginibus
late revolutis, floribus sparsis l-2nis, pedicellis flori zequilongis, calycis
segmentis ovatis obtusis, corolle tubo campanulato segmentis latis
brevibus, staminibus inclusis connatis, stylo exserto stigmate magno
peltato.

A much-branched shrub, with crowded erect densely pubescent
branchlets. Lower leaves 3-1 in., upper 4, in. long. Pedicels
densely pubescent, 4. in. long. Calyx nearly as long as the
corolla; segments similar to the leaves in texture, ciliated with
bristles. Corolla red, 75 in. long and broad. Pistil twice as
long as the corolla.— Baron 3480 !

PHILIPPIA MINUTIFOLIA, n. sp.

P. ramulis gracillimis erectis pubescentibus, foliis minutis oblongis
rigidulis mucronatis ascendentibus quaternis parce glanduloso-hispidulis
marginibus late revolutis, floribus paucis l-3nis terminalibus, calycis
segmentis ovatis, corollze campanulate segmentis latis brevibus, stamini-
bus liberis inclusis, stylo longe exserto stigmate magno peltato.

A much-branched shrub, with very slender erect branchlets
with a loose white epidermis. Leaves crowded, erecto-patent, at
most a line long, with a few gland-tipped bristles. Flowers
terminal on the branchlets; pedicels very short, densely gland-
ular. Corolla bright red, campanulate, glandular, $ line long
and broad. Calyx glandular, half as long as the corolla. Sta-

mens 8, as long as the corolla. Pistil twice as long as the
corolla.— Baron 4458 !

ARDISIA LEPTOCLADA, n. Sp.

Fruticosa, ramulis gracillimis pubescentibus, foliis breviter petiolatis
ovatis acutis submembranaceis przter costam faciei inferioris glabris,
cymis axillaribus sessilibus 1-3-floris, pedicellis elongatis, calycis seg-
mentis suborbicularibus, corollae segmentis oblongis, antheris liberis
oblongis, fructu globoso glabro.

A much-branched erect shrub with very slender flexuose
branchlets. Leaves 1-12 in. long, rounded at the base, thin
in texture, green on both sides, pubescent only on the midrib
beneath; petiole always very short. Pedicels slender, curved,

FLORA OF MADAGASCAR. 501

3-1 in. long. Flowers pentamorous. Calyx campanulate, y in.
in diam.; segments much imbricated. Corolla 4L in. long; tube
very short. Stamens nearly as long as the corolla.— Baron
3674!

ARDISIA DISSITIFLORA, n. sp.

Fruticosa, ramulis gracilibus pilosis, foliis breviter petiolatis oblongis
acutis subcoriaceis dorso ad costam solum pilosis, cymis axillaribus
pedunculatis circiter trifloris pedicellis elongatis, floribus tetrameris, calycis
segmentis ovatis, acutis, corolla segmentis ovatis, staminibus liberis,
antheris lanceolatis, filamentis brevissimis.

A shrub, probably a climber, with slender terete pilose branch-
lets. Leaves about an inch long, firm in texture, green and
glabrous on both surfaces except the midrib beneath, copiously
dotted with black. Cymes copious, axillary; peduncle and
pedicels each 3—# in. long, slightly pilose. Calyx y; in. in diam.,
tube very short. Corolla å in. in diam. Stamens 4; anthers
lanceolate, much longer than the filaments.— Baron 4511!
Habit of Oncostemum vacciniifolium, but stamens quite free.

OxNcosTEMUM ? POLYTRICHUM, n. sp.

Fruticosum, dense molliter fusco-pilosum, foliis magnis petiolatis
oblongis subcoriaceis, floribus pentameris dense copiose racemosis, pedi-
cellis elongatis, calycis segmentis ovatis, corolle segmentis ovatis acutis
tubo campanulato longioribus, filamentis brevibus connatis, antheris
liberis lanceolatis.

Branches woody, terete, densely clothed with short brown
pubescence. Leaves 3-4 in. long, 13-14 in. broad at the
middle, narrowed gradually to both ends, firm in texture, green
and clothed with short brown pubescence on both surfaces;
petiole 4 in. long. Flowers in dense peduncled panicled ter-
minal racemes ; pedicels 3—4 in. long, densely pilose. Expanded
calyx å in. in diam. Corolla glabrous, j in. in diam. Stamens
shorter than the corolla; anthers lanceolate ; filaments united in
a campanulate cup.— Baron 2219! 3878! Combines the habit
and anthers of Ardisia with the connate filaments of Onco-
stemum.

ONCOSTEMUM MICROSPHERUM, n. sp.

Fruticosum, erectum, ramosissimum, ramulis dense pilosis, foliis breviter
petiolatis orbicularibus parvis glabris, floribus pentameris l-3nis axil-
laribus, pedunculo brevi, pedicellis elongatis, calycis segmentis ovatis
obtusis, corolla tubo brevi segmentis orbicularibus, coronz campanulate
filamentis deltoideis basi solum connatis.

502 MR. J. G. BAKER ON THE

A much-branched small shrub, with very slender branchlets,
densely clothed with stiff brown hairs. Leaves j-j in. long
and broad, moderately firm in texture, green and glabrous on
both surfaces, closely dotted with black. Peduncle very short,
pilose ; pedicels erect, glabrous, 4 in. long. Calyx spreading hori-
zontally, 4l, in. in diam. Corolla 5 in. in diam.; bud globose ;
segments much imbricated. Stamens included.— Baro» 4462!

ONCOSTEMUM VACCINIIFOLIUM, n. sp.

Fruticosum, erectum, ramulis gracilibus pilosis, foliis brevissime petio-
latis oblanceolato-oblongis acutis subcoriaceis przter costam glabris,
floribus tetrameris, cymis axillaribus szpissime bifloris, pedunculis pedi-
celisque elongatis, calycis segmentis ovatis obtusis, corolla segmentis
ovatis, antheris coalitis, filamentis brevissimis.

A much-branched erect shrub, with erecto-patent or spreading
slender pubescent branchlets. Leaves 1-11 in. long, 3-} in.
broad, firm in texture, green on both sides, hairy only on the
midrib beneath. Cymes copiously produced from the axils of the
leaves; peduncle about 4 in. long; pedicels 4—4 in. long, curved,
glandular-pubescent. Expanded calyx flat, in.in diam. Corolla
4 in. in diam. Stamens inserted low down in the corolla; fila-
ments very short.— Baron 3858! 3773 !

ONCOSTEMUM FLEXUOSUM, n. Sp.

Fruticosum, sarmentosum, ramulis flexuosis sursum obscure glanduloso-
pubescentibus, foliis breviter petiolatis oblongis acutis subcoriaceis glabris,
floribus pentameris, cymis axillaribus paucifloris, calycis segmentis orbi-
cularibus, corollee segmentis ovatis obtusis, filamentis perfecte connatis,
fructu globoso magnitudine pisi

A climber, with slender terete branchlets. Leaves 1-2 in. long,
narrowed to the base, moderately firm in texture, green and
glabrous on both sides, petiole 1-1 in. long. Cymes 3-5-flowered
from the axils of the leaves; peduncle }~#in. long; pedicels
very slender, 3-4 times as long as the flowers. Corolla j in.
in diam. Anthers placed inside a ring a little shorter than the
corolla.— Baron 4134 !

ONCOSTEMUM BOTRYOIDES, n. sp.

Fruticosum, glabrum, foliis magnis petiolatis oblanceolato-oolongis ob-
tusis coriaceis, floribus pentameris laxe copiose racemosis, pedicellis arcuatis
flore longioribus, calycis segmentis orbicularibus ciliatis late imbricatis,
corollz segmentis orbicularibus, staminibus connatis.

Branches moderately robust, terete. Leaves 3-4 in. long,

FLORA OF MADAGASCAR. 503

1-13 in. broad, narrowed gradually from the middle to base, firm
in texture, green and glabrous on both surfaces; petiole 4 in.
long. Flowers in copious lax axillary peduncled racemes 3-4 in.
long; pedicels 4 in. long. Expanded calyx Jin. in diam. Corolla
% in. in diam.; tube very short. Filaments connate, in a cam-
panulate cup shorter than the corolla, inside which the stamens
are placed.—Baron 1096! 3714!

PacHYPODIUM DENSIFLORUM, n. Sp.

P. tubero firmo polycephalo, rosulis dense aculeatis, foliis oblongis dorso
pilosis, pedunculo elongato, cymis densis paucifloris, calycis segmentis
ovatis lanosis, corolle tubo cylindrico piloso calyce longiore, segmentis
obovatis, folliculis cylindricis elongatis.

Tuber with several cylindrical branches, densely armed with
spreading subulate pungent prickles lin. long. Leaves only
seen immature, more or less coated with short white pubescence
beneath. Peduncle reaching nearly a foot in length. Flowers
6-8 in a dense cyme. Calyx-segments 1 in. long. Corolla-tube
nearly twice as long as the calyx; segments obovate, 4 in. long.
Follicles pubescent, 4 in. long. Seeds cylindrical, vin. long,
crowned by a corona of silky hairs.— Baron 4246 !

PACHYPODIUM BREVICAULE, n. sp.

P. tubero firmo polycephalo, rosulis copiose aculeatis, foliis hysteranthiis
oblongis, pedunculo brevi pubescente interdum nullo, cymis densis pau-
cifloris, calycis segmentis lanceolatis dense pilosis, corolla lutez tubo
cylindrico pubescente segmentis obovatis tubo brevioribus.

Tuber subglobose, -1 in. in diam. Prickles ina dense cluster,
pale, subulate, erect, i in. long. Leaves at the flowering-time
very immature, oblong, densely pilose. Pedicels very short.
Calyx densely pilose, } in. long; tube very short. Corolla-tube
Zin. long, densely pubescent on the outside ; segments 3 in. long.
Stamens not reaching above the middle of the corolla-tube.—
Baron 4412! Also Hildebrandt 3586!

ALYXIA LUCIDA, n. sp.

Sarmentosa, fruticosa, glabra, foliis 2-4nis sessilibus lanceolatis vel
oblongo-lanceolatis acuminatis lucidis subcoriaceis, cymis laxis paucifloris
terminalibus, calycis segmentis minutis ovatis, corolle tubo elongato
cylindrico, segmentis angustis tubo 4-5plo brevioribus, fructu pedicellato
splendide rubro profunde toruloso.

A climber, glabrous in all its parts, with very slender green

501 MR. J. G. BAKER ON THE

branchlets. Leaves 12-2 in. long, 4-3 in. broad at the middle,
moderately firm in texture, green and glabrous on both surfaces,
glossy above. Cymes few-flowered, shortly peduncled ; pedicels
erect, about j in. long. Calyx #5 in. long. Corolla-tube 3 in.
long; expanded limb 1 in. in diam., very contorted before expan-
sion; segments oblong. Fruit of 1-5 oblong or globose joints
à-iin. long, green at first, finally bright scarlet.— Baron 4551!
4663 ! 4888!

MASCARENHAISIA GERRARDIANA, n. Sp. :

Sarmentosa, glabra, ramulis gracilibus, foliis oppositis breviter petiolatis
oblongis acutis rigide coriaceis, floribus splendide rubris in cymas pauci-
floras terminales dispositis, calycis tubo brevi segmentis ovatis, coroll:
tubo cylindrico segmentis oblongis, antheris lanceolatis infra tubi medium
insertis.

A climber, glabrous in all its parts, with slender purplish-
black branchlets. Leaves erecto-patent, 12-2 in. long, 2-1 in.
broad, narrowed to both ends, firm in texture, green and glabrous
on both surfaces, with the veins and veinlets beneath raised.
Cymes 1-6-flowered, shortly peduncled; bracts small, ovate.
Calyx 4; in. long. Corolla-tube iin. long, constricted in the
lower third; expanded limb above an inch in diameter. Anthers
iin.long. Fruit not seen.— Baron 2510! 4652 ! Also Gerrard
52! B. Kitching! Nearly allied to M. Curnowiana, Hemsley ;
Bot. Mag. t. 6612.

MASCARENHAISIA MACROSIPHON, n. sp.

Glabra, ramulis gracilibus pallidis, foliis petiolatis oblongo-lanceolatis
subcoriaceis venulis faciei inferioris immersis, floribus solitariis axillaribus
breviter pedunculatis, calycis segmentis parvis ovatis, corollz tubo elongato
cylindrico infra medium constricto, segmentis oblique obovatis, antheris
in tubo inclusis infra medium insertis.

A shrub, with slender branchlets, glabrous in all its parts.
Leaves opposite, erecto-patent, 2-3 in. long, 4—3 in. broad at the
middle, narrowed gradually to both ends, moderately firm in
texture, green and glabrous on both surfaces, the veins fine and
indistinct; petiole % in. long. Peduncle 4 in. long, erect. Calyx
py in. long. Corolla apparently white; tube 1i in. long, con-
spicuously constricted below the insertion of the stamens; limb
above an inch in diameter. Fruit not seen.— Baron 3840!

Strycunos BARONI, n. sp. ;
Glabra, ramulis gracilibus copiose lenticellatis, foliis petiolatis oblongis

FLORA OF MADAGASCAR. 505

acutis triplinerviis, cymis axillaribus paucifloris, bracteis ovatis, pedicellis
brevibus, calyce tubo brevi segmentis ovatis obtusis ciliatis, corolla tubo
brevi cylindrico intus dense piloso, segmentis oblongis tubo equilongis,
antheris oblongis ad tubi faucem subsessilibus.

An erect tree, with glabrous branchlets. Leaves opposite,
2-3 in. long, 1-14 in. broad at the middle, narrowed gradually to
both ends, moderately firm in texture, green and glabrous on both
surfaces, triplinerved from a little above the base. Cymes shortly
peduneled, much shorter than the leaves; pedicels à—1 in. long.
Calyx +}; in. long ; segments imbricated, densely ciliated. Corolla
% in. long. Anthers shorter than the perianth-segments. Style
reaching to the top of the anthers. Fruit not seen.—Baron
4648 !

BuDDLEIA SPH ZROCALYX, n. sp.

B. ramis robustis dense pubescentibus, foliis magnis oblongis acutis
crenatis basi late connatis, racemis axillaribus breviter pedunculatis, calyce
globoso ventricoso pubescente dentibus parvis, corolla tubo cylindrico
calyci æquilongo, limbi segmentis parvis orbicularibus, fructu nigro
glabro calyce rumpente.

Stems stout, erect, woody, densely pubescent. Leaves 4—6 in,
long, 11-2 in. broad, moderately firm in texture, finely crenate,
pubescent on both surfaces, connate at the base for half an inch
on each side ofthe stem. Racemes dense, 11-2 in. long; peduncle
short, densely pubescent; pedicels 4 in. long; bracts lanceolate.
Calyx bladder-like, densely pubescent, 4 in. long and broad ; teeth
minute, lanceolate-deltoid. Corolla hypocrateriform ; limb å in.
in diam.; segments 4, orbicular. Fruit lin. in diam.— Baron
4401! A very curious and distinct species.

NuxiA SPHEZROCEPHALA, Baker.

More mature specimens in the present parcel (Baron 3738 !)
of the plant which I described in Journ. Linn. Soc., Bot. vol. xxi.
p. 425, as Buddleia spherocephala show it to have divaricate
anther-lobes, and a corolla falling off by circumscissile dehisceuce,
so that it must be transferred to the genus Nusia, to the neigh-
bourhood of N. capitata, N. terminalioides, and N. pachyphylla.

NvxXIA TERMINALIOIDES, n. sp.

N. ramulis tenuiter pubescentibus, foliis oppositis petiolatis oblanceolato-
oblongis subcoriaceis subacutis serratis dorso obscure pubescentibus, flori-
bus in capitula globosa laxe paniculata aggregatis, calyce coriaceo tubo
companulato, dentibus deltoideis, coroll tubo cylindrico segmentis oh-
longis, genitalibus exsertis.

LINN. JOURN.— BOTANY, VOL. XXII. 28

506 MR. J. G. BAKER ON THE

An erect shrub, with slender shortly pubescent branchlets.
Leaves 2-3 in. long, 1-17 in. broad above the middle, narrowed
gradually from the middle to the base, firm in texture, green on
both surfaces, with 6-8 raised ascending main veins. Heads j in.
in diam., arranged in a terminal panicle about 3 in. long and broad,
the lower branches bracteated at the base by large leaves. Calyx
dark brown, coriaceous, pubescent, 1 in. long; teeth 4—4 as long
as the tube. Corolla-tube jin. long; segments } as long as the
tube.— Baron 3629!

NUXIA PACHYPHYLLA, n. 8p.

N. ramulis pubescentibus, foliis oppositis vel ternis breviter petiolatis
oblanceolato-oblongis obtusis integris subcoriaceis venis primariis 6-8-
jugis exsculptis, floribus in capitula globosa panieulata aggregatis, calyce
piloso coriaceo tubo infundibulari dentibus ovatis, corollze tubo cylindrico
segmentis oblongis obtusis, genitalibus exsertis.

A shrub or small tree, with woody pubescent branchlets.
Leaves opposite or ternate, ascending, 2—4 in. long, 1-2 in.
broad above the middle, narrowed gradually from the middle to
the base, strongly ribbed beneath, at first pubescent, finally gla-
brous. Heads 3-# in. in diam. exclusive of the style and stamens,
arranged in lax deltoid terminal panicles 3—4 in. long. Calyx
brown-black, very coriaceous, 4 in. long. Corolla-lobes ¿4 in. long.
Stamens } in. longer than the calyx; anthers broadly reniform.
Style overtopping the anthers.— Baron 1326! 3389! 4056! 4407!

ANTHOCLEISTA AMPLEXICAULIS, n. Sp.

Glabra, foliis magnis obovatis amplexicaulibus, floribus copiose corym-
boso-paniculatis, bracteis late connatis, calycis segmentis orbicularibus
late imbricatis, corolle tubo cylindrico calyci zquilongo segmentis 9-10
oblongis, antheris segmentis paulo brevioribus.

Branchlets woody, robust. Leaves a foot long, 5 in. broad,
obtuse, narrowed gradually from the middle to the amplexicaul
base, subcoriaceous, green and glabrous on both surfaces, with
8-9 pairs of distant erecto-patent parallel main veins. Panicle
dense, terminal, half a foot broad; lower bracts witha long, upper
with a very short point. Calyx lin. in diam.; inner segments

in. broad. Corolla-segments 3 in. long. Anthers lanceolate, 4 in.

long.—.Baron 3795! Allied to A. Vogelii, Planch., in Hook. Ic.
t. 3795. Native name Landemy.

ANTHOCLEISTA RHIZOPHOROIDES, n. Sp.
Glabra, foliis petiolatis obovato-oblongis obtusis rigide coriaceis, cymis

FLORA OF MADAGASCAR. 507

paucifloris, bracteis parvis, calycis segmentis orbicularibus valde imbricatis,
coroll: tubo calyci zequilongo, segmentis 9-10 oblongis, fructu ovoideo.

Branchlets woody, terete, marked with the rings left by the
fallen leaves. Leaves opposite, erecto-patent ; petiole 4 in. long
blade 3—4 in. long, 1-13 in. broad above the middle, cuneate at the
base, green and glabrous on both surfaces, the side-veins fine and
immersed. Cymes lax, terminal. Calyx } in. long; inner seg-
ments jin. broad. Corolla an inch long; segments as long as the
tube. Fruit 4 in. in diam.—Baron 1952! 3814!

BELMONTIA EMIRNENSIS, n. sp.

Glabra, foliis sessilibus oblongis acutis, floribus copiose corymboso-pani-
culatis, calyce late quinquealato segmentis acuminatis, corolla segmentis
ovatis tubo urceolato longioribus, staminibus ex corolle tubo longe ex-
sertis, filamentis brevissimis.

A glabrous perennial herb, with simple erect stems 3 ft. long.
Leaves opposite, decussate, triplinerved, the central ones about
2in.long, the upper shorter and remote. Flowers numerous,
bright yellow, arranged in a terminal panicle with erect corym-
bose branches; bracts ovate-lanceolate, foliaceous ; pedicels 4 in.
long. Calyx campanulate, } in. in diam., all the lobes acuminate
and broadly winged. Corolla j-$ in. long. Anthers l in. long.
Ovary ovoid; style 4 in. long, reaching nearly to the tip of the
corolla-segments.—.Baron 4187! Adds another characteristic
Cape genus to the Madagascar flora.

Tpoma@a (8 StROPHIPOMG@A) RUBRO-VIRIDIS, n. sp.

Herbacea, volubilis, glabra, caulibus gracillimis, foliis remotis sessilibus
linearibus integris uninerviis, floribus magnis axillaribus solitariis breviter
pedunculatis, calycis segmentis exterioribus oblongis, corolla infundibulari
intus rubella extus viridula, fructu parvo globoso.

An herbaceous climber, glabrous in all its parts, with very
slender stems. Leaves 2-3 in. long, 3;—1 in. broad, narrowed to
an indistinct petiole, moderately firm in texture, green, glabrous.
Peduncle j-j in. long, erect, thickened upwards. Calyx 1 in.
long ; three inner segments longer, narrower, and more acute than
the two outer. Corolla 13-1? in. long, about j in. in diam. at the
throat. Capsule 3 in. in diam., tipped with the persistent style.
— Baron 4728! 4832! Near the Australian J. graminea, R. Br.

Ipom@a (§ ORTHIPOM@A) SYRINGJEFOLIA, n. sp.

Fruticosa, erecta, ramulis sursum dense pubescentibus, foliis longe
petiolatis ovatis integris acutis utrinque breviter pubescentibus, floribus
282

508 MR. J. G. BAKER ON THE

copiose cymoso-paniculatis, pedunculis pedicellisque dense pubescentibus,
ealycis segmentis oblongis, corolla obconica calyce 5-6plo longiora extus
hispidula, segmentis parvis deltoideis, genitalibus corolla duplo bre-
vioribus.

An erect shrub, with slender terete branchlets, clothed, like
the peduncles, leaves, and calyx, with short adpressed whitish
pubescence. Leaves simple, 2—4 in. long, acute, rounded at
the base, quite entire, moderately firm in texture, green and
shortly pilose on both surfaces. Flowers copious, forming a ter-
minal panicle, with ascending level-topped branches, bracteated
at the base by large leaves. Calyx } in. long, the two outer seg-
ments hiding the rest. Corolla an inch long, probably reddish.
Style lin. long ; stigmas globose.— Baron 46741

BREWERIA TILIZFOLIA, n. sp.

Volubilis, fruticosa, ramulis gracilibus pubescentibus, foliis petiolatis
integris cordato-ovatis cuspidatis utrinque pubescentibus, floribus axillari-
bus dense racemosis, pedicellis brevissimis, calyce piloso tubo campanulato
dentibus deltoideis, corolla obconica glabra, segmentis obtusis brevissimis,
stylo staminibus eminente stigmatibus globosis.

A climber, with slender terete densely pubescent branchlets.
Leaves distant, alternate; petiole about an inch long; blade 23 in.
long, thin, green, and shortly pubescent on both surfaces. Racemes
copious, nearly sessile, 1-14 in. long, sometimes compound at the
base. Calyx 4l in. in diam., densely pilose. Corolla } in. long
and broad, probably white. Stamens unequal, less than half as
long as the corolla; anthers small, oblong.— Baron 4580! 4817!

DrpYMOCARPUS PUSILLUS, n. sp.

Annuus, fragilis, caule producto simplici piloso, foliis petiolatis oblongis
obtusis inciso-crenatis facie glabris dorso parce pilosis basi cuneatis, pe-
dunculo terminali piloso 1-2-floro, calycis segmentis lineari-oblongis,
corollz tubo oblongo segmentis brevibus.

An erect fragile annual, about 2 inches high. Leaves 3 in.
long, membranous, green above, pale green beneath, with four
pairs of arcuate main veins ; petiole }in. long. Calyx 4 in. long,
cleft down nearly to the base. Corolla reddish, twice as long as

the calyx.— Baron 3440!
THUNBERGIA CHRYSOCHLAMYS, n. sp.

Volubilis, fruticosa, ramis glabris, foliis petiolatis oblongis acutis in-
tegris coriaceis glabris, floribus pluribus axillaribus, pedicellis elongatis

FLORA OF MADAGASCAR. 509

luteo-incanis, bracteis coriaceis connatis persistenter aureo-incanis, calyce
truncato brevissimo, coroll: tubo piloso lobis brevibus, stigmate exserto.

A shrubby climber, with woody terete stems. Leaves 21-3 in-
long, 1-12 in. broad at the middle, acute, firm in texture, green
and glabrous on both surfaces ; petiole 4 in. long. Flowers a dozen
or more in sessile cymes in the axils of the leaves; pedicels 4—4
in. long, clothed with persistent golden-yellow tomentum like the
bracts. Bracts oblong, 3 in. long, coriaceous, connate nearly to
the top; lobes half-orbicular. Calyx densely pilose. Corolla
1j in. long, clothed with bristly hairs; tube infundibuliform ;
throat l in. in diam.; lobes short, rounded. Stamens reaching to
the throat of the corolla-tube. Capsule not seen.— Baron 4915 !
Numbers 4762 and 4770 are either the same species or a near
ally, but are incomplete. To be compared with Radlkofer’s
genus Pseudocalya.

MIMULOPSIS AFFINIS, n. sp.

Herbacea, caulo erecto gracillimo sursum glanduloso-pubescente, foliis
petiolatis lanceolatis parce pilosis, floribus paucis luteis laxe racemosis,
pedicellis elongatis glanduloso-pubescentibus bracteatis, calycis segmentis
linearibus elongatis inzequalibus, coroll tubo obconico, segmentis orbi-
cularibus, staminibus in tubo inclusis.

Stems herbaceous, very slender. Leaves 1-2 in. long, thin,
green and slightly pilose on both sides; petiole 1-lin. Flowers
few, arranged in a lax terminal raceme, the lowest branch some-
times forked; peduncles and pedicels densely softly glandular-
pubescent; the latter with a pair of small opposite lanceolate
bracts above the middle. Calyx j-iin. long, glandular-pubes-
cent; tube very short. Corolla 3-3 in. long; segments 5, sub-
equal, spreading, 1-1 in. broad. Stamens inserted low down in
the tube ; anthers of the longer pair with a spur at the base of
the larger cell— Baron 3435! 4050! 4294! Very near M. lan-
ceolata, Baker.

STROBILANTHES HISPIDULA, n. sp.

Fruticosa, ramulis gracilibus pilosis, foliis longe petiolatis integris acutis
ovatis vel oblongis, floribus axillaribus solitariis, bracteis foliaceis oblan-
ceolatis, calyce piloso tubo infundibulari dentibus lanceolato-acuminatis
tubo zquilongis, corolla tubo infundibulari segmentis 5 oblongis sub-
zqualibus, staminibus didynamis antheris omnibus perfectis ovatis sub-
zequalibus.

510 MR. J. G. BAKER ON THE

A much-branched shrub, with slender densely hispidulous
branchlets. Leaves 14-2 in. long, rather thin, green and ob-
scurely pilose on both surfaces, deltoid or rounded at the base ;
petiole 1-13 in. long. Flowers solitary in the axils of the close
reduced upper leaves ; bracts oblanceolate, obtuse, generally larger
than the calyx. Calyx i in. long, densely pilose. Corolla 2 in.
long; tube much dilated in the upper, cylindrical in the lower
half; segments 1 in. long, 1 in. broad. Longer stamens reaching
to the throat of the perianth-tube; anthers 4, small, ovoid.

Ovary ampulleform; style reaching up to the anthers.— Baron
4543 !

BanLERIA Krremrwoir, n. sp.

Fruticosa, inermis, foliis parvis ovatis integris hispidis, floribus dense
spicatis sordide purpureis, calycis segmentis exterioribus magnis cordato-
ovatis spinoso-marginatis demum scariosis, interioribus parvis lanceolatis,
Corolle tubo deorsum cylindrico sursum infundibulari, limbi segmentis 5
orbicularibus, antheris anticis cum stylo ex tubo breviter exsertis, posticis
brevissimis.

Stems slender, shrubby, terete, green, glabrous. Leaves at the
flowering-time very small. Spikes dense, oblong. Calyx of 4
segments; two outer nearly an inch long, 4 in. broad, hispid,
scariose, persistent; two inner hidden by the outer, small, lan-
ceolate-aeuminate. Corolla-tube 1} in. long, 4 in. in diam. at the
throat; segments spreading, 4-2 inch broad; expanded limb 13
in.in diam. Longer stamens just exserted from the throat of the
corolla-tube; anthers oblong, i in. long ; shorter stamens inserted
with the others near the base of the tube, not reaching above
its middle; anthers small, globose. Betsileo-land.—Baron 257!
4048! Ibara country, Kitching! A near ally of B. cristata, L.

BARLERIA PHILLYREXFOLIA, n. sp.

Fruticosa, inermis, glabra, ramulis gracillimis, foliis petiolatis ovatis vel

oblongis acutis integris, floribus solitariis axillaribus, bracteis linearibus,
calycis segmentis hispidis, exterioribus magnis lanceolatis, interioribus
parvis rigidulis, corolla tubo infundibulari segmentis obovato-cuneatis,
staminibus anticis ex tuho longe protrusis, posticis parvis staminodio
zequilongis.

An erect shrub, with slender terete woody branchlets. Leaves
13-2 in. long, membranous, green and obscurely pilose on both
surfaces, deltoid at the base ; petiole 4—4 in. long. Flowers solitary
in the axils of many of the leaves ; pedicel very short; bracts

FLORA OF MADAGASCAR. 511

linear, faleate, 1 in. long. Outer calyx-segments leafy, lanceo-
late, } in. long ; inner shorter, narrower, firmer, Corolla-tube
l; in. long, 3 in. in diam. at the throat; segments subequal, 1 in.
long, 3 in. broad. Stamens inserted low down in the tube;
larger anthers oblong, $ in. long; shorter stamens } in. long,
with ovate anthers and a small staminode between them.— Baron

4555 !

HyPOESTES PHYLLOSTACHYA, n. sp.

Herbacea, ramulis gracilibus pubescentibus, foliis longe petiolatis ovatis
integris membranaceis glabris, spicis elongatis floribus singulis folio
sepissime reducto bracteatis, involucro unifloro bracteis exterioribus
lanceolatis obtusis, interioribus foliaceis, calyce parvo, corolle rubelle
tubo involucro paulo longiore, limbo parvo.

An erect herbaceous perennial, with slender pubescent stems.
Leaves 11-2 in. long, deltoid at the base, acute or obtuse, thin,
green and glabrous on both surfaces; petiole 1-13 in. long. Spikes
long, lax, simple or forked at the base. Involucre cylindrical,
pubescent, 3-3 in. long. Calyx much shorter than the involucre.
Corolla-tube 4 in. long; lower lip 4 in. long, cuneate, shortly 3-
lobed. Stamens reaching to the tip of the corolla-limb.— Baron
4907! Also Hildebrandt 3444! Near H. lasiostegia, Nees.

HYPOESTES MICROPHYLLA, n. sp.

Fruticosa, ramulis dense pilosis, foliis parvis brevissime petiolatis lan-
ceolatis integris pilosis rigidulis, capitulis dense spicato-paniculatis, involu-
cro unifloro bracteis lanceolatis rigidulis acuminatis, calyce involucro
breviore, corollze tubo pubescente involucro duplo longiore, limbo involucro
sequilongo labio inferiore trilobato.

An erect shrub, with slender virgate densely pubescent
branchlets. Leaves 2-1 in. long, subcoriaceous, entire, green and
clothed with adpressed hairs on both surfaces. Flowers in ter-
minal panicles, with dense ascending spicate branches. Involucre
1 in. long; inner segments shorter. Corolla-tube 1 in. long; limb
half as long as the tube; lower lip cuneate, 1 in. broad. Stamens
lin. longer than the tube. Capsule lanceolate, acute, glabrous.
— Baron 4382! Near H. saxicola, Nees.

HYPOESTES CONGESTIFLORA, n. sp.

Fruticosa, ramulis gracilibus pubescentibus, foliis linearibus subsessili-
bus rigidulis uninerviis glabris, spicis axillaribus secundis dense congestis,
involucro unifloro piloso bracteis elongatis exterioribus foliaceis interiori-

512 MR, J. G. BAKER ON THE

bus acuminatis, calyce involucro duplo breviore, coroll tubo involucro
zequilongo, limbo triplo breviore.

An erect shrub, with slender virgate densely pubescent
branchlets. Leaves 11-9 in. long, } in. broad, firm in texture,
green and glabrous on both surfaces. Spikes nearly sessile, many-
flowered, densely congested, short, secund. Involuere densely
pubescent, + in. long; bracts all lanceolate ; inner much shorter
and more acuminate than outer. Corolla-tube 1 in. long;
lower lip 4 in. long. Stamens reaching nearly to the tip of the

ligulate upper lip of the corolla.— Baron 3391 !

HYPOESTES OBTUSIFOLIA, n. Sp.

Fruticosa, ramulis gracilibus obscure pubescentibus, foliis petiolatis ovatis
integris obtusis membranaceis glabris, floribus paucis terminalibus confer-
tis, involucro brevi unifloro, calyce involucro subzquilongo, corollz tubo
cylindrico involucro 3-4plo longiore, limbo tubo triplo breviore, labio
infimo cuneato trilobato.

An erect shrub, with slender green branchlets, thinly clothed
with adpressed pubescence. Leaves 1-2 in. long, 4—1 in. broad,
thin, green and. glabrous on both surfaces; petiole 1-4 in. long.
Flowers few, in terminal congested spikes, bracteated by reduced
leaves. Involucre and calyx 4 in. long. Corolla-tube about an
inch long, glanduloso-pubescent ; lower lip 1 in. long, 4 in. broad.
Stamens a little exserted from the corolla-tube.— Baron 4890!

Hx»ozsrEs ACUMINATA, n. sp.

Fruticosa, glabra, ramulis gracilibus tetragonis, foliis petiolatis oblongo-
lanceolatis acuminatis subintegris, floribus laxe spicato-paniculatis, invo-
lucro unifloro bracteis lanceolatis acutis, calyce involucro breviore, corollæ
tubo infundibulari involucro triplo longiore, limbo brevi.

An undershrub, glabrous in all its parts, with slender square
branchlets. Leaves 2-3 in. long, 3—? in. broad, narrowed to the
base, thin in texture, green and glabrous on both surfaces, penni-
nerved, with 8-9 pairs of arcuate erecto-patent main veins ; petiole
i-i in. long. Spikes lax, ascending, peduncled, few-flowered.
Involucre 1—1 in. long; inner and outer bracts similar in length
and shape. Corolla-tube 1-1j in. long, white or reddish; lips

under 7 in. long. Anthers just exserted from the corolla-tube.
— Baron 3407! 3513!

HYPOESTES SESSILIFOLIA, n. sp.

Fruticosa, glabra, ramulis gracilibus tetragonis, foliis sessilibus oblongis

FLORA OF MADAGASCAR. 513

subintegris, floribus dense unilateraliter spicato-paniculatis, involucro
unifloro bracteis rigidulis lanceolatis acutis, calyce involucro paulo breviore,
coroll: tubo involucro 2plo longiore, limbo involucro æquilongô.

An undershrub, glabrous in all its parts, with slender square
green branchlets. Leaves 1-2 in. long, 1-3 in. broad at the
middle, obtuse, narrowed to the base, obscurely undulated on the
margin, green and glabrous on both surfaces, moderately firm in
texture, penninerved, with 6-8 pairs of arcuate-ascending main
veins. Spikes short, dense, secund, stiffly erecto-patent. Invo-
lucre $ in. long, all the four bracts similar in size and shape.
Corolla-tube j-$ in. long; lips j in. long, the lower slightly
3-lobed. Anthers just exserted from the throat of the corolla-
tube.—Baron 4284! Both this and H. acuminata are nearly
allied to H. jasminoides, Baker.

HYPOESTES CHLOROCLADA, n. 8p.

Fruticosa, ramulis gracilibus viridibus glabris, foliis petiolatis oblongis
acutis integris acuminatis, floribus albidis vel rubellis spicato-paniculatis,
involucro unifloro glanduloso-pubescente bracteis lanceolatis obtusis,
calyce involucro breviore, corolla tubo involucro 2-3plo longiore, limbo
tubo triplo breviore.

A shrub, with slender green glabrous branchlets. Leaves 2-3
in. long, 1-1} in. broad at the middle, narrowed gradually to both
ends, bright green, glabrous. Flowers in a dense terminal panicle
with spicate branches. Involucre pubescent, jin. long. Corolla-
tube 3 in. long; limb j in. long; lower lip cuneate, faintly
3-lobed. Anthers reaching halfway up the limb.— Baron 1224!
4045! 4893! Near H. comorensis and H. jasminoides.

CLERODENDRON ($ CYCLONEMA) MIRABILE, n. sp.

Erectum, fruticosum, glabrum, foliis oppositis sessilibus oblanceolato-
oblongis acuminatis serratis, floribus parce corymboso-paniculatis, calyce
tubo infundibulari dentibus magnis lanceolatis, corolla tubo cylindrico
longissimo, limbo parvo obliquo hispidulo.

An erect shrub, with slender terete stems. Leaves 3-4 in.
long, 1-1} in. broad at the middle, moderately firm in texture,
green and glabrous on both surfaces, sharply incised in the
middle half. Panicle sparse, terminal, erect ; pedicels very short.
Calyx ? in. long; teeth half as long as the tube. Corolla-tube
half a foot long; limb seen in bud only, globose, with the stamens
curled up so as to form a circle and a half, with oblong anthers.—
Baron 4755! Near C. macrosiphon, Hook. f. Bot. Mag. t. 6695.

514 MR. J. G. BAKER ON THE

AJUGA OOCEPHALA, D. sp.

Perennis, caulibus decumbentibus simplicibus pilosis, foliis longe petio-
latis ovatis obtusis profunde irregulariter crenatis utrinque parce hispidulis,
floribus dense capitatis, pedicellis brevissimis, calyce tubo infundibulari
dentibus lanceolato-deltoideis tubo brevioribus, corolla tubo calyce paulo
longiore, labio inferiore lobis lateralibus parvis oblongis centrale latissimo
emarginato.

A perennial herb, with simple decumbent stems under a foot
long. Leaves thin, 2-3 in. long, broadly cuneate at the base, green
and thinly hairy on both surfaces; petiole 1-1 in. long. Flowers
in a dense terminal head, the lower whorls only bracteated by
reduced leaves. Calyx } in. long, hispid principally in the upper
half. Corolla blue; tube 1 in. long; lower lip 4 in. broad.
Stamens exserted 3 in. beyond the throat of the corolla-tube.—
Baron 3430! 3520!

INCOMPLETA.

Cetosta (§ LAGREZIA) MICRANTHA, n. sp.

Annua, caulibus elongatis gracillimis obscure pilosis, foliis longe petio-
latis ovatis vel oblongis obtusis basi cuneatis utrinque tenuiter pilosis,
floribus in panieulam ramis primariis elongatis ramulis brevibus cyinosis
dispositis, perianthio segmentis oblongis albis utrieulo brevioribus, utriculo
lentieularo monospermo, stylo brevi, stigmatibus falcatis.

An annual, with very slender erect distantly leafy stems a foot
long. Leaves 1-2 in. long, membranous, entire, green and thinly
pilose on both surfaces ; petiole as long as the blade. Panicles
terminal, 1-3 in. long; branches pilose, ascending; cymes dense,
few-flowered. Utricle green, globose, 4 line long. Perianth-
segments oblong, Scariose, white, tinged with green.—Baron

4229! The three species previously known are characterized
in ‘ Genera Plantarum,’ vol. iii. p. 25.

Prrer (8 Coccosryon) EMIRNENSE, n. sp.

Glabrum, ramulis gracilibus, foliis longe petiolatis cordato-orbicularibus
cuspidatis membranaceis e basi /-nerviis, spicis axillaribus densissimis
longe pedunculatis, bracteis orbicularibus peltatis, ovario globoso sessili,
stylo brevi, stigmatibus 2 globosis capitatis.

Herbaceous, branched, with very slender elongated stems.
Leaves 3-4 in. long, 2-3 in. broad, shallowly cordate, green and
glabrous on both surfaces, with 7 main nerves, only the central
three reaching beyond the middle; petiole 1-1 in. long. Spikes
in the fruiting-stage Ł in. in diam. Peltate bracts minute. Style

FLORA OF MADAGASCAR. 515

nearly as long as the ovary, tipped with two spreading globose
stigmas.— Baron 3677! Allied to P. capense, Linn. fil.

CRYPTOCARYA PAUCIFLORA, n. Sp.

Ramosissima, ramulis gracillimis sursum ferrugineo-pubescentibus,
foliis oblongis obtusis rigide coriaceis glabris facie viridibus dorso glauco
tinctis, cymis paucifloris terminalibus et lateralibus, pedunculis pedicellis-
que dense ferrugineis, perianthio ferrugineo tubo obconico segmentis
ovatis, antheris inclusis.

Branchlets crowded, ascending, slender, ferruginous only
towards the tip. Leaves 1j-2 in. long, obtuse, deltoid at the
base; veins fine, immersed; petiole j in. long Cymes short, 8—4-
flowered. Perianth j in. long; segments longer than the suleate
tube, both densely ferruginous. Stamens inserted at the throat
of the tube, rather shorter than the segments. Ovary immersed
in the perianth-tube. Fruit not seen.— Baron 4470!

OcorEA ($ MESPILODAPHNE) TRICHANTHA, n. Sp.

Ramosissima, ramulis dense pubescentibus, foliis erecto-patentibus
breviter petiolatis oblongis obtusis rigide coriaceis basi cuneatis facie
tenuiter dorso dense pubescentibus venis primariis valde ascendentibus,
eymis axillaribus paucifloris, bracteis pilosis lanceolatis, perianthio dense
piloso segmentis ovatis, calycis tubo accrescente rigido, fructu ex tubo
breviter exserto.

Branchlets crowded, slender, erecto-patent, very pubescent.
Leaves 14-2 in. long, dull green on both surfaces, narrowed
gradually from the middle to the base, with distant very ascend-
ing main veins. Cymes few-flowered. Cupule black, glossy,
} in. long and broad. Fruit drab, j in. long.— Baron 4373!

VISCUM GLOMERATUM, N. sp.

Aphyllum, ramosissimum, ramis primariis nodis compressis ad basin
angustatis, ramulorum brevibus tetragonis, nodis cupulatis, floribus ad
nodos glomeratis subsessilibus, ovario globoso rugoso, perianthii segmentis
parvis ovatis.

Whole plant 3-4 in. long; lower internode of main stem 3-1 in.
in diam., much eompressed, much narrowed to the base; branch-
lets crowded, ascending, made up of several internodes 3-} in. long,
with cupular nodes enclosing a dense mass of minute flowers, of
which there is also a cluster at the tip of each branchlet.— Baron
3757! Ona Eugenia, Allied to V. japonicum, Thunb., and V.
capense, L. fil.

516 MR. J. G. BAKER ON THE

‘VISCUM RHIPSALOIDES, n. sp.

Aphyllum, fruticosum, glabrum, ramis primariis teretibus lignosis,
junioribus compressis erecto-patentibus viridibus subtiliter multisuleatis
basi articulatis, nodis superioribus bracteis parvis oblongis carnosis oppo-
sitis przeditis, floribus masculis axillaribus sessilibus segmentis ovatis tubo
campanulato zequilongis.

A shrub, many times dichotomously, finally distantly pinnately
branched ; branchlets erecto-patent, 1—L in. in diam. ; upper inter-
nodes 13-2 in. long. Upper nodes with two small spreading
obtuse fleshy bracts a line long. Male flower p> in. in diam.
Female flowers not seen.— Baron 4892! Resembles in habit the

Indian and Cape FV. dichotomum, D. Don.

THESIUM CYSTOSEIROIDES, n. sp.

Fruticosum, ramosissimum, siccitate brunneum, habitu Cystoseire,
ramulis gracillimis glabris, foliis subulatis mucronatis, floribus solitariis
axillaribus breviter pedicellatis foliis circiter 3 reductis bracteatis, ovario
globoso pubescente perianthii segmentis conniventibus coronato.

A much-branched shrub, with glabrous stems and leaves, all
turning brown when dried; main stems terete, jy in. in diam.
Leaves scattered, ascending, 1-1 in. long, tipped with a pale horny
mucro. Fruit 3 in. in diam., vertically ribbed, crowned with the
green connivent deltoid perianth-segments, which are much
shorter than the fruit— Baron 3490! Habit of the Cape T.
ericefolium, A. DC.

PEDILANTHUS ? LYCIOIDES, n. sp.

Fruticosus, glaber, ramis lignosis elongatis aculeis pungentibus rectis
patulis armatis, ramulis brevissimis pulvinatis minute aculeatis, folus
sessilibus linearibus obtusis rigide coriaceis, involucris oblongis rubellis
profunde bilabiatis ad ramorum apices glomeratis pedunculis pedicellisque
dichotomis brevissimis, bracteis bracteolisque parvis ovatis, flore foemineo
unico stylis elongatis summo apice furcatis, floribus masculis pluribus.

Stem woody, cylindrieal, a foot or more long, without branch-
ing, j in. in diam., producing irregularly small prickly pulvini
representing branchlets and armed with straight spreading pun-
gent prickles i-i in. long. Leaves sessile, deciduous, rigid,
linear, obtuse, 1-1} in. long, ji; in. broad, distinctly costate,
obscurely penninerved. Heads about 6 in a cluster at the ends
of the main branches; peduncles dichotomously forked; each
head clasped by a small ovate membranous bracteole. Involucre

FLORA OF MADAGASCAR. 517

4-i in. long, composed of two unequal oblong lobes reaching
down nearly to the base. Styles reaching in the flowering-stage
to the tip of the involucre. Male flowers much shorter.— Baron
4772! 4792! This very curious plant merits further investiga-
tion. The genus Pedilanthus is exclusively American.

EUPHORBIA ($ ARTHROTHAMNUS) ALCICORNIS, n. Sp.

Fruticosa, erecta, aphylla, multoties dichotomiter ramosa, ramulis strictis
erecto-patentibus basi solum articulatis, capitulis ad ramorum apices
glomeratis, bracteis parvis ovatis, involucro parvo campanulato, appendi-
cibus 7-8 erectis cuneatis.

An erect leafless shrub many times dichotomously forked;
main stems terete, 4 in. in diam. ; penultimate branchlets 2-3 in.
long; ultimate shorter, subterete, 4 in. in diam. Heads clustered
at the tips of the branchlets; bracts ovate, minute. Involucre
& in. in diam., crowned by 7-8 erect small cuneate emarginate
fleshy green appendages.— Baron 4875!

EUPHORBIA ($ GONIOSTEMA) ORTHOCLADA, n. sp.

Fruticosa, ramis carnosis cylindricis elongatis, folis sparsis alternis
sessilibus oblanceolatis obtusis subcarnosis, cymis terminalibus oligocepha-
lis, involucro campanulato bracteis 2 foliaceis orbiculari-cuneatis stipato
appendicibus 5 orbicularibus integris patulis przedito, ovario seminibusque
levibus glabris.

A shrub, with long cylindrical fleshy unarmed branchlets.
Leaves placed all down the branches, erecto-patent, fleshy, 14-14
in. long, $ in. broad. Heads arranged in dense cymes at the ends
of the branches, each subtended by a pair of foliaceous bracts
iin. long. Involucre j in. in diam. Fruit-carpels 1 in. long.
Seeds large, oblong-quadrate.— Baron 3555 !

EUPHORBIA (8 GONIOSTEMA) BAKERIANA, Baill. in Bull. Linn.
Soc. Par. p. 623.

Arborea, ramosissima, ramulis gracilibus teretibus lignosis, foliis ad
ramorum apices subconfertis petiolatis oblongis cuspidatis subcoriaceis,
cymis axillaribus oligocephalis, bracteis parvis foliaceis ovato-lanceolatis,
involucro parvo campanulato appendicibus 5 patulis integris orbicularibus.

A much-branched tree, glabrous in all its parts, with slender
terete branchlets. Leaves 1-14 in. long, conspicuously cuspidate,
firm in texture, green and glabrous on both surfaces, with few
indistinct erecto-patent main veins. Cymes axillary, shortly

518 MR. J. G. BAKER ON THE

peduneled, composed of 2-3 heads, each subtended by a pair of
foliaceous bracts as long as the involucre. Involucre 4l in. in
diam. Fruit and seeds not seen.— Baron 3763! 3769! 3842!
Allied to E. adenopoda, Baill., and E. erythroxyloides, Baker.

EUPHORBIA ($ GowrosrEwA) pacuysantua, Baill. in Bull.
Linn. Soc. Par. p. 623. ;

Arborea, ramosissima, glabra, ramulis lignosis, folis ad ramorum apices
confertis breviter petiolatis oblongis acutis membranaceo-carnosis, capi-
tulis terminalibus sessilibus, involucro campanulato appendicibus 5 patulis
lunatis integris stipitatis.

A much-branched tree, glabrous in all its parts, with much
stouter branches and branchlets than in E. Bakeriana. Leaves
2-3 in. long, 2-1 in. broad, narrowed to the base and apex, green
and glabrous on both surfaces, with the indistinct distant erecto-
patent main viens anastomosing in intramarginal arches. Heads
solitary at the tips of the branchlets. Involuere, including the
spreading dark-green fleshy appendages, } in. in diam. Fruit and
seeds not seen.— Baron 4437! Allied to E. Bakeriana.

Evrnorsra (§ GoxrosrEwA) Mancrnetta, Baill. in Bull.
Linn. Soc. Par. p. 623.

Arborea, ramosissima, glabra, ramulis gracilibus lignosis teretibus, foliis
petiolatis oblongis acuminatis subcoriaceis venis faciei inferioris perspicuis,
erecto-patentibus inter marginem regulariter in arcubus anastomosantibus,
cymis terminalibus oligocephalis, pedicellis brevissimis, bracteis parvis
ovatis, involucro campanulato appendicibus 5 patulis integris transversa-
liter oblongis.

A much-branched tree, with slender terete woody branchlets,
rough with the raised scars of the fallen leaves. Leaves 2-4 in.
long, 1-1} in. broad at the middle, very acuminate, deltoid at the
base, moderately firm in texture, with fine parallel main yon
anastomosing at the tip in regular arches. Cymes sessile, termi-
nal. Involucre 1 in. in diam., clasped by two small ovate bracts.
Immature fruit globose, 4 in. in diam. Styles falcate, forked
at the tip.— Baron 4442! Allied to Z. adenopoda, Baill.

ANTIDESMA BRACHYSCYPHA, n. sp.

Ramosissima, ramulis gracilibus pubescentibus, foliis breviter petiolatis
oblongis vel oblongo-lanceolatis utrinque viridibus glabris in axillis folio-
rum haud scrobiculatis, racemis masculis laxifloris rhachide pubescente, pe-
dicellis brevissimis, bracteis lanceolatis, calycis tubo brevi, segmentis ovatis,
ovario rudimentario globoso, staminibus 2-3 conspicue exsertis.

FLORA OF MADAGASCAR. 519

Much-branched, with crowded branchlets, glabrous below the
tip. Leaves 11-2 in. long, 4-2 in. broad at the middle, obtusely
aeuminate, narrowed to the base, moderately firm in texture,
green and glabrous on both surfaces. Male racemes about an
inch long; rhachis slender, pubescent. Calyx 4l in. in diam.
Stamens generally 3, twice as long as the calyx.— Baron 4447 !

ANTIDESMA ALNIFOLIA, n. sp.

A. ramulis gracilibus sursum pilosis, foliis breviter petiolatis oblongis
obtusis utrinque viridibus preter costam glabris in axillis venarum haud
scrobiculatis, floribus masculis copiose laxe spicatis, calyce late campanu-
lato segmentis ovatis, ovario rudimentario globoso, stylo brevi staminibus
seepissime 3 conspicue exsertis.

A tree with slender branchlets pubescent towards the tip.
Leaves 2-3 in. long, 1-13 in. broad, rounded at the base, entire,
green and glabrous on both surfaces except the midrib, the
distant arcuate slender main veins anastomosing by intramarginal
arches. Spikes copious, axillary, shortly peduncled, 2-3 in. long.
Calyx 4 in. in diam., greenish, lobed about halfway down. Sta-
mens 4; in. long.— Baron 4666 !

ANTIDESMA ARBUTIFOLIA, n. Sp.

A. ramulis ferrugineo-pubescentibus, foliis breviter petiolatis oblongis
obtusis subcoriaceis utrinque viridibus glabris in foliorum axillis haud
scrobiculatis, floribus foemineis breviter racemosis rhachide pedicellisque
dense pilosis, bracteis ovatis, calyce late campanulato dense piloso, stylis
3 incurvatis ovario :zequilongis.

A tree with straight woody branchlets, pubescent only towards
the tip. Leaves 2-4 in. long, 1-12 in. broad, rounded to the base,
moderately firm in texture, green and glabrous on both surfaces,
with fine erecto-patent parallel main veins anastomosing in intra-
marginal arches. Female racemes axillary, peduncled, an inch
long; pedicels as long as the flowers. Calyx y} in. in diam., twice
as broad as long; segments ovate. Ovary globose, pilose.—
Baron 4062!

CROTON VERNICOSUS, n. sp.

Fruticosus, ramulis sulcatis dense ferrugineo-lepidotis, folis breviter
petiolatis lanceolatis integris coriaceis facie viridibus vernicosis dorso dense
persistenter albido-lepidotis, spicis terminalibus laxifloris, floribus masculis
pluribus, calyce dense lepidoto, tubo brevi, segmentis ovatis, petalis parvis
oblongis ciliatis, staminibus 9-10, flore foemineo solitario, ovario dense
lepidoto, seminibus brevibus nitidis,

520 MR. J. G. BAKER ON THE

A much-branched shrub, with slender sulcate branchlets matted
with dense ferruginous lepidote scales. Leaves 3-4 in. long, 3-2
in. broad, acute, narrowed to the base, bright green and glossy
above, white beneath mixed with ferruginous spots. Spikes lẹ-
3 in. long, with a stiff very lepidote axis. Male calyx yy in.
diam., lepidote like the branches and axis. Fruit 4 in. long.
Seeds drab, glossy.— Baron 4935! Allied to C. argyrodaphne,
Baill.

CEPHALOCROTON CORDIFOLIUS, n. Sp.

Fruticosus, ramulis gracilibus sursum stellato-pilosis, folis longe pe-
tiolatis cordato-ovatis integris acuminatis utrinque stellato-pilosis, stipulis
lineari-subulatis persistentibus, floribus foemineis solitariis calycis segmen-
tis magnis deltoideis laciniis elongatis pectinatis, floribus masculis parvis
in capitulum globosum dispositis pedicellis dense pilosis, calyce piloso
tubo brevi segmentis 3-4 obtusis ovatis vel oblongis, staminibus isomeris,
ovario rudimentario piloso stylo brevi, stigmate polycephalo.

A shrub, with slender terete woody branchlets. Leaves 3-4 in.
long, 2-3 in. broad, very acuminate, slightly pilose above, densely
beneath ; petiole 1-14 in. long. Female flower shortly peduncled ;
segments 5, an inch long, ovate, bipinnate. Male flowers many
in a head 3 in. in diam. ; pedicels longer than the flowers; calyx
Ty in. in diam. Stamens as long as the perianth ; anthers globose ;
filaments glabrous.— Baron 4668 ! 4774!

Macanaxaa (8 EUMACARANGA) RACEMOSA, n. Sp.

M. ramulis elongatis lignosis nigrescentibus sursum pubescentibus, foliis
breviter petiolatis erecto-patentibus oblongo-lanceolatis subcoriaceis facie
parce dorso copiose lepidotis venis 10-1 2-jugis parallelis erecto-patentibus,
floribus foemineis in racemos copiosos axillares breviter pedunculatos dis-
positis, pedicellis erecto-patentibus, calycis segmentis ovatis, fructu globoso
glutinoso-lepidoto monospermo.

Branchlets stiff, straight, a foot long, glabrous below the tip.
Leaves 2-3 in. long, 1-1 in. broad, obtuse, cuneate at the base,
firm in texture, green above, drab beneath, with the raised
parallel main veins distinct to the edge. Female flowers in
copious erecto-patent shortly peduncled axillary racemes 1-12 in.
long; pedicel as long as the fruit, subtended by a lanceolate
bract. Fruit 4 in. in diam., tipped by the persistent subulate style.
Seed black, smooth, filling up the whole fruit.—_Baron 3654!

FLORA OF MADAGASCAR. 521

MACARANGA FERRUGINEA, D. SD. ,

Arborea, ramulis crassis resiniferis sursum ferrugineo-pubescentibus,
folis alternis petiolo supra basin inserto late ovatis repando-dentatis
subcoriaceis, facie tenuiter dorso dense pubescentibus, floribus masculis in
paniculas deltoideas axillares dispositis in axillis bractearum glomeratis,
bracteis magnis ovatis laciniatis pubescentibus, calyce parvo piloso segmentis
ovatis, staminibus paucis.

A tree, with robust woody branchlets, clothed towards the tip
with ferruginous pubescence. Petiole 2-3 in. long, pubescent ;
blade 3-4 in. long and broad, obscurely cordate, with the petiole
inserted i-i in. above the base, repand, with small cusps at the
tip of the lobes, moderately firm in texture, matted above when
young, but only obscurely pubescent when mature, drab and
densely pubescent beneath, with 5-6 pairs of erecto-patent veins.
Female flowers and fruit not seen. Male flowers in axillary
panicles about as long as the petiole, several clustered in the
axil of an ovate pubescent laciniated bract lin. long. Buds
small, globose, densely pilose. Stamens about 3.—Baron 4395!
The stems contain an abundant supply of resin, the nature of
which needs investigation on fuller material.

CELTIS GOMPHOPHYLLA, n. Sp.

Arborea, glabra, ramulis gracillimis lenticellatis, foliis petiolatis oblongis
vel ovatis integris acutis deorsum longe inzquilateraliter cuneatis, floribus
1-2nis axillaribus, pedicellis erectis vel cernuis fruetu longioribus, ovario
ovoideo glabro, stylis brevibus falcatis.

A tree, glabrous in all its parts, with very slender lenticellate
branchlets. Leaves 2-34 in. long, 11-14 in. broad at the middle,
moderately firm in texture, green and glabrous on both surfaces,
generally oblique in the lower half. Flowers solitary or in pairs
in the axils of some of the lower leaves of the branch; pedicels
ascending, slender, 4 in. long. Dried fruit ovoid, 3-4 in. in
diam.; style 4 line long.— Baron 3697! Adds the genus to the

Madagascar flora.

Ficus COCCULIFOLIA, n. sp.

F. ramulis gracilibus pubescentibus, stipulis parvis pubescentibus, foliis
cordato-ovatis obtusis utrinque viridibus pubescentibus venis primariis
4-5-jugis erecto-patentibus, receptaculis parvis pilosis e ramis haud foliatis
ortis fasciculatis basi bracteis ovatis pubescentibus imbricatis, pedicellis
ascendentibus.

An erect tree, with slender pubescent branchlets. Petiole 4 in.
long; blade 3-4 in. long, 2-23 in. broad at the middle, obtuse,

LINN. JOURN.— BOTANY, VOL. XXII. 2T

522 MR. J. G. BAKER ON THE

shallowly cordate at the base, thin in texture for the genus,
green and finely pubescent on both surfaces; main veins distant
and forked before they reach the edge. Receptacles produced
copiously on short leafless branchlets, only seen in a young state,

very pubescent, as are the bracts and ascending pedicels.—Baron
4660 ! 4965 !

Ficus (S UROSTIGMA) PHANEROPHLEBIA, n. 8p.

F. ramulis foliiferis gracilibus teretibus scabris, foliis breviter petiolatis
oblongis cuspidatis subcoriaceis utrinque viridibus facie scabris dorso sub-
tiliter pilosis venulis omnibus pallidis exsculptis, receptaculis parvis glo-
bosis glabris scaberulis breviter pedunculatis, pedunculo minute bracteato.

Leaf-bearing branches slender, very scabrous. Petiole also
rugose, 1-2 in. long; blade 3-5 in. long, 2-21 in. broad, deltoid
or rounded at the base, moderately firm in texture, pale green
beneath, with about 6 pairs of arcuate main veins, all the small
veinlets distinctly raised. Receptacles lin. in diam., produced
from the leafy branches ; peduncle 4 in. long, minutely bracteated

above the middle.— Baron 4474! Allied to F. claoxyloides,
Baker.

Ficus PACHYCLADA, n. sp.

F. ramulis foliiferis validis scaberulis, foliis longe petiolatis oppositis vel
alternis cordato-orbicularibus rigide coriaceis utrinque viridibus scaberri-
mis dorso hispidulis venulis omnibus exsculptis, receptaculis globosis mag-
nitudine mediocribus dense pubescentibus pedunculatis e ramis nudis ortis
solitariis basi haud bracteatis. .

Leaf-bearing branch 1—3 in. in diam.; petiole erecto-patent, 2 in.
long; blade 3-4 in. long and broad, conspicuously cordate, very
rough on the upper surface, pale green beneath, with all the
hispid veinlets raised. Leafless fertile branches more slender.
Peduncle pubescent, erecto-patent, minutely bracteate. Re-
ceptacle globose, densely pubescent, 4 in. in diam.— Baron 4496!
Nearly allied to the Asiatic F. hispida, L. fil.

Ficus ($ CovELLIA) PULVINIFERA, n. sp.

F. ramulis gracilibus apice pubescentibus, stipulis parvis lanceolatis, foliis
petiolatis cordato-ovatis acutis utrinque viridibus glabris venis primariis
/-8-jugis arcuatis, receptaculis magnis globosis e ramis foliatis ortis soli-
tariis basi bracteis tribus parvis ovatis suffultis, pedunculo piloso ascendente
receptaculo zequilongo.

A tree, with rugose woody branchlets, pilose when young.
Petiole 4 in. long; blade resembling that of Populus basalmifera,

FLORA OF MADAGASCAR. 523

3-4 in. long, shortly cordate at the base, bright green above,
smooth and quite glabrous on both surfaces. Receptacle gla-
brous, -2 in. in diam. Peduncle erecto-patent, pubescent, 4 in.

long. —Baron 3406!

Ficus ($ COVELLIA) SAKALAVARUM, n. Sp.

F. ramulis gracilibus pilosis, stipulis parvis lanceolatis, foliis breviter
petiolatis cordato-ovatis integris obtusis subcoriaceis utrinque viridibus
facie scabris dorso pubescentibus venis primariis paucijugis erecto-patenti-
bus, receptaculis magnis duris globosis glabris breviter pedunculatis basi
bracteis parvis ovatis suffultis.

Flowering branches slender, sulcate, densely pilose. Petiole
4-4 in. long; blade 2-3 in. long, 13-2 in. broad at the middle,
shortly cordate at the base, bright green, glabrous and scabrous
above, paler green and pubescent beneath, with 5-6 pairs of
parallel erecto-patent main veins. Receptacles produced from
stouter leafless branches, an inch in diameter, subtended at the
base by three small ovate bracts; peduncle about j in. long.—
Baron 4885! Common in the Sakalava country. Native name
Adabo or Adabolahy.

Ficus (§ CoVELLIA) ALBIDULA, n. sp.

F. ramulis gracilibus pilosis, stipulis parvis lanceolatis, foliis distincte
petiolatis anguste oblongis obtusis subcoriaceis glabris facie viridibus dorso
albidulis venis primariis 6-7-jugis arcuatis, receptaculis magnis solitariis
breviter pedunculatis e ramis foliatis ortis basi bracteis 3 parvis ovatis
suffultis.

A tree, with slender branchlets, hairy only at the tip. Stipules
2 in. long, equalling the petiole; blade 2-3 in. long, 3-1 in. broad
at the middle, obtuse, rather rounded at the base. Receptacles
$ in. in diam., produced from the leafy branchlets; peduncle
very short.— Baron 8525!

Ficus ($ CovELLIA) BOTRYOIDES, n. sp.

F. ramulis gracilibus scabris, stipulis lanceolatis, foliis longe petiolatis
oblongo-lanceolatis acuminatis subcoriaceis utrinque viridibus glabris,
venis primariis arcuatis /-8-jugis, receptaculis magnis globosis e ramis
haud foliatis ortis dense czspitosis, basi bracteis tribus ovatis suffultis,
pedunculis elongatis glabris.

A tree, with slender terete branchlets. Petiole 1-1} in. long;
blade 4—6 in. long, 14-24 in. broad at the middle, very acute,
deltoid at the base, moderately firm in texture, green and glabrous
on both surfaces. Receptacles 3-2 in. in diam., produced in

272

524 MR. J. G. BAKER ON THE

dense clusters; peduncles sometimes an inch long; bract small,
ovate.— Baron 3803!

Ficus (§ COVELLIA) TRICHOCLADA, n. sp.

F.ramulis gracilibus pubescentibus, stipulis lanceolatis pilosis, foliis longe
petiolatis oblongo-lanceolatis integris subcoriaceis utrinque viridibus gla-
bris venis primariis arcuatis 8-10-jugis, receptaculis magnis globosis pubes-
centibus solitariis e ramis foliatis ortis, basi bracteis 3 parvis ovatis suf-
fultis, pedunculis brevibus dense pilosis.

A much-branched erect shrub, with slender sulcate branchlets.
Petiole 3 in. long; blade 3-4 in. long, 1-11 in. broad, subacute,
rounded at the base, moderately firm in texture. Receptacles
3-7 in. in diam., produced singly from the slender branchlets just
below the leaves; peduncle 1-3 in. long — Baron 3547 !

PILEA MACROPODA, n. sp.

Dioica, herbacea, glabra, caule simplici erecto, foliis oppositis zequalibus
longe petiolatis oblongis acuminatis erebre subtiliter serrulatis membrana-
ceis, floribus foemineis in foliorum axillis dense glomeratis, perianthii seg-
mentis ovatis obtusis inzqualibus, achenio ovoideo compresso duplo
brevioribus.

Stems herbaceous, erect, simple, moderately stout, a foot long.
Leaves 3-4 in. long, 1i-1j in. broad, deltoid or rather rounded
at the base, membranous, triplinerved, finely sharply serrated,
especially in the upper half ; petiole 1-1} in. long. Flowers in
sessile clusters, in the axils of the central leaves. Achene ovoid,

glossy, pale drab, half a line long. Male flowers not seen.—
Baron 4450!

ELATOSTEMA HEXADONTUM, n. sp.

Dioicum, glabrum, caulibus cæspitosis simplicibus debilibus apice solum
foliatis, stipulis magnis lanceolatis persistentibus, foliis breviter petiolatis
oblanceolatis supra medium conspicue dentatis, floribus masculis in folio-
rum axillis glomeratis, bracteis magnis ovatis foliaceis, perianthio membra-
naceo globoso segmentis 4 ineequalibus linearibus, antheris 4 subsessilibus.

Stems slender, decumbent and then ascending, 3-12 in. long.
Leaves 4-6 on a side, alternate, 3-1 in. long, 4-4 in. broad, dark
green, membranous, slightly inzquilateral, with about six large
teeth in the upper half or third, the lower half entire, with a vein
just within the margin connecting the main veins. Flowers in a
sessile cluster in the axils of the leaves. Bracts 4—4 in. long, the
outer sterile. Perianth a small membranous bag, with 4 unequal
erect segments.— Baron 505 ! 4059 ! 4475!

Or

FLORA OF MADAGASCAR. 52

MOoNOCOTYLEDONS.

DYPSIS POLYSTACHYA, n. sp.

D. foliis copiose pinnatis, rhachi compresso scabro, segmento terminali
parvo bifido, pinnis multijugis lanceolatis integris acuminatis, spadicibus
copiose pinnatis, spicis elongatis gracillimis rhachi glabro, floribus foemi-
neis petalis ovatis striatis, sepalis orbicularibus minutis, fructu oblongo
parvo monospermo.

Stem and sheaths not seen. Leaf 1-2 ft. long; rhachis flat on
the face, clothed with drab furfuraceous ramentum; pinn®
arranged irregularly, the lower 9-10 in. long, } in. broad at the
middle. Spathe not seen. Inflorescence nearly a foot long;
spikes very slender, 8-9 in. long. Female flower 4; in. long ;
bract broad ovate; petals striated; sepals rigid, not striated.
Dried berry 3 in. long; pericarp very thin.—Baron 1287!

Nearly allied to D. pinnatifrons, Mart.

Dypsis HETEROPHYLLA, n. Sp.

D. foliis superioribus vel omnibus simplicibus cuneatis profunde bifidis,
inferioribus interdum simpliciter pinnatis pinnis lanceolatis integris acumi-
natis, vaginis imbricatis glabris, spathis parvis pedunculo arcte vaginantibus,
spadicibus bifidis vel trifurcatis rhachi glabro sulcato, petalis ovatis striatis,
sepalis orbicularibus imbricatis petalis triplo brevioribus.

Stem 1 in. in diam. Leaves many, spaced out, erecto-patent ;
sheaths about 3 in. long; blade 6-9 in. long, consisting sometimes
entirely of one deeply bifid end-segment, sometimes of a small bifid
end-segment and many erecto-patent lanceolate acuminate pinne.
Spathe cylindrical, about 2in. long. Spikes 2-3 in. long ; bracts
and bracteoles minute, orbicular. Petals strongly ribbed, 4; in.
long; sepals very rigid, glossy, not striated. Fruit not seen.—

- Baron 486! Also Lyall 323! D. forficifolia, Noronha; Mart.
Hist. Palm. iii. 180, 312, tab. 143. fig. 2, is similar to this in
habit, but all the leaves are simple and deeply bifid and the
inflorescence is copiously compound. Mr. Baron has also
gathered this (1490) and D. pinnatifrons, Mart. Hist. Palm. iii.
180, 312, tab. 158. fig. 1.

DYPSIS RHODOTRICHA, n. Sp.

D. foliis copiose pinnatis, segmento terminali parvo bifido, pinnis multi-
jugis lanceolatis integris acuminatis, vaginis ferrugineo-pilosis, spadicibus
simplicibus rhachi glabro, petalis ovatis striatis, sepalis orbicularibus striatis
petalis triplo brevioribus, fructu oblongo semine solitario.

Stem 4 in. in diam. Leaves many, spaced out, erecto-patent ;

526 MR. J. G. BAKER ON THE

sheaths about 3 in. long, imbricated, densely clothed with fine
red-brown hairs; blade 9-12 in. long; pinne erecto-patent,
lanceolate, 2-3 in. long, 4 in. broad at the middle, narrowed
gradually to the acuminate point. Peduncle axillary, 4-5 in.
long. Spadix short, simple; rhachis slender, angled, glabrous.
Petals 7; in. long, ovate, strongly ribbed. Sepals orbicular,
imbricated, also ribbed. Fruit oblique, oblong, à in. long, with a
thin pink pericarp and a single seed with white horny albumen.
—Forests between Tamatave and Antananarivo, in wet clayey
soil, Dr. Meller!, who also colleeted another species with copiously
pinnate leaves, with short truncate leaflets, inciso-serrated at the
oblique apex. This latter, he notes, is “used for thatching.”
The specimen is without flower or fruit.

Dyrsis CONCINNA, n. sp.

D. foliis copiose pinnatis, segmento terminali parvo bifido, pinnis multi-
jugis parvis lanceolatis acutis haud laceratis, vaginis glabris, spathis
angustis, spadicibus bifidis vel trifurcatis, petalis ovatis striatis, sepalis
parvis suborbieularibus, fructu oblongo semine solitario.

Stem j-3 in. indiam. Leaves many, spaced out, erecto-patent ;
sheaths glabrous, imbricated, 3 in. long; blade 6-8 in. long ;
rhachis slender, 4-angled, minutely furfuraceous; side-leaflets
very numerous, 2—4 in. long, 3-3 in. broad, narrowed gradually
to along point. Peduncles axillary, erecto-patent, 8—4 in. long,
clasped tightly by the spathe ; spikes 3-4in. long; rhachis angled,
naked; bracts and bracteoles minute, orbieular. Petals j^; in.
long; sepals not striated, coriaceous, glossy, 3 as long as the
petals. Fruit oblong, rather curved, 3 in. long; pericarp thin.
Seeds conspicuously spirally ribbed; albumen horny, white.—

Baron 1286! 3413! Also received from Mr. William Pool in the
year 1876.

Dypsis Curristt, n. sp.

D. foliis copiose pinnatis, segmento terminali parvo bifido, pinnis multi-
jugis irregulariter dispositis lanceolatis integris longe acuminatis, spathis
cylindrieis, spadice simplici elongato, rhachi dense nigro-furfuraceo, sepalis
orbicularibus striatis, petalis oblongis obtusis calyce triplo longioribus.

Leaf 17 ft. long; rhachis flat on the face, scabrous ; pinne
erecto-patent ; leaves 6-7 in. long, an inch broad at the middle,
narrowed to the adnate base and a long acuminate point. Peduncle
above half a foot long, tightly elasped by the coriaceous spathe.
Spathe simple, 8-9 in. long ; slender axis densely clothed with

FLORA OF MADAGASCAR. 527

black furfuraceous ramentum. Male flower 4', in. long. Petals
oblong, obtuse, valvate, strongly ribbed. Sepals imbricated,
glossy, coriaceous. Fruit not seen.—Curtisll6! Received from
Messrs. Veitch in 1881. This resembles D. pinnatifrons, Mart.,
in the shape of the leaflets, but the inflorescence is unbranched.

PHLOGA POLYSTACHYA, Noronha, in Thouars, Prod. Phytol.
Madag. p. 2 (name only) Dypsis noditera, Mart. Hist. Palm. iii.
312.

Leaves 2-3 ft. long, simply pinnate; rhachis compressed;
end-segment formed of 2-4 confluent side- pinnz», deeply bifid at
the apex; upper and lower side-pinne single; central crowded
into irregular whorls; pinnæ lanceolate, 6-8 in. long, an inch
broad at the middle, narrowed gradually to the base and deflexed
very acuminate apex. Spathe narrow, coriaceous, 4-5 in. long,
tightly clasping the stout peduncle. Spadix copiously bipinnate ;
final branchlets very slender, curved, 6-8 in. long; flowers laxly
disposed, usually solitary ; bract broad, ovate; bracteoles minute,
orbicular. Male flower yy in. long ; petals ovate, valvate, strongly
ribbed; sepals suborbicular, imbricated, } as long as the petals;
stamens 6, half as long as the petals. Female flower with shorter
petals. Fruit oblong, } in. long, brown in the dried specimens,
green when young in the fresh state, red when mature; pericarp
thin, fleshy ; seed single; albumen horny.— Baron 508! Also
Meller! Parker! and received from Hort. Linden as * Areca
madagascariensis.” Native name “ Ouvun," teste Meller; given
as * Ou Van" by Noronha, loc. cit., and “ Anivona” by Dr.
Parker.

PANDANUS MONTANUS, Bory,=Sussea conoidea, Gaudich. Atlas
Bonite, tab. 24.

Gathered in South Betsileo-land by the Rev. Mr. Rowlands,
from whose specimen the following notes are made :—Leaves
4-5 tt. long, rigidly coriaceous, 15-2 in. broad at the base, tapering
gradually to the point, closely margined with lanceolate erecto-
patent small prickles. Peduncle woody, 2 in. thick. Female
head oblong-triquetrous, 2 ft. long, 3in.in diam. Drupe clavate,
3 in. long. Structure as drawn by Gaudichaud.— Baron 4931!

PANDANUS DYCKIOIDES, N. sp.
P. caule lignoso ramoso, foliis dense rosulatis linearibus acuminatis rigi-
dulis aculeis corneis magnis lanceolatis vel deltoideis marginalibus armatis,

528 MR. J. G. BAKER ON THE

capitulis sessilibus centralibus, masculo oblongo bracteis magnis ovatis
intermixto, foemineo solitario subgloboso floribus circiter 20, ovario ampul-
leformi uniloculari, stigmate peltato.

Woody main stem 4 in., tuft-bearing branch lin. in diam.

~

Leaves 9-15 in. long, 4—4} in. broad above the dilated base, firm
in texture, closely ribbed ; spines id in. long, stramineous above
the base. Male flowers almost hidden by the large ovate bracts.
Female head about 3 in. in diam., subtended and overtopped by
3-4 oblong-lanceolate bracts with closely minutely serrated
edges. Ovary under 4 in. long.— Baron 491! 3815! Allied to

P. pygmeus, Thouars, Bot. Mag. tab. 4736.

CRINUM ($ PLATYASTER) MODESTUM, n. sp.

C. bulbo collo oblongo, foliis 6-8 linearibus glabris subpedalibus, pedun-
culo foliis subzequilongo, umbellis 5-6-floris, spathe valvis 2 lanceolatis
reflexis, ovario sessili, perianthii tubo cylindrico, segmentis lanceolatis
patulis tubo brevioribus.

Neck of the bulb in the dried plant 2-3 in. long, 1 in. in diam.
Leaves a foot long, 4 in. broad low down, tapering gradually to
the point, closely finely ribbed, scabrous on the margin. Peduncle
a foot long, 1 in. in diam. Spathe-valves 14 in. long. Flowers all
quite sessile; perianth-tube very slender, 2 in. long ; segments
1i in. long, 1 in. broad at the middle, 10-12-nerved. Filaments
distinctly shorter than the segments.— Baron 48 ! 4223! Allied
to the East-Indian C. humile, Herbert, Bot. Mag. t. 2636.

DIOSCOREA CRYPTANTHA, n. sp.

Volubilis, caule gracillimo pubescente, foliis membranaceis pubescenti-
bus pinnatim trifoliolatis, foliolis obovatis vel oblongis aristato-cuspidatis,
spicis masculis 2-4nis densis brevibus, braeteis magnis ovatis cuspidatis
membranaceis, perianthio oblongo tubo brevi segmentis oblongo-lanceo-
latis, pedicello dense piloso.

Aun herbaceous climber, with very slender obscurely pubescent
stems. Petiole an inch long ; leaflets 1-2 in. long, membranous,
faintly pilose, green above, pale beneath ; end-leaflets obovate,
shortly stalked; side-leaflets oblique oblong, nearly sessile.
Spikes 1-1} in. long, shortly peduncled, 2-4nate, from leafy or
leafless nodes. Bracts ovate, brown, membranous, 4 in. long,
quite hiding the flowers. Male perianth +, in. long, twice as
long as the very hairy pedicel. Anthers ovoid, nearly sessile on
the base of the segments. Female flowers and fruit not seen.—
-Baron 4658! A near ally of the Cape D. crinita, Hook. fil. in
Bot. Mag. t. 6804.

FLORA OF MADAGASCAR. 529

ALoE (§ EUALOE) HAWORTHIOIDES, n. sp.

Acaulis, foliis circiter 30 lanceolatis parvis acuminatis dense rosulatis
aculeis patulis albis deltoideis cuspidatis copiosis przditis, pedunculo
elongato, floribus parvis dense spicatis erecto-patentibus pallide rubellis
viridi vittatis, bracteis ovatis cuspidatis, perianthio tubo subnullo, seg-
mentis oblongo-lanceolatis, genitalibus demum breviter exsertis.

Rosette under 2 in. in diam. Leaves 12 in. long, 1-3 in. broad
at the base, thin in texture for the genus, dark green, tapering
gradually to a long point, furnished on the edge especially
upwards with copious white horny spreading prickles, which are
found also on the back in a less-developed condition. Peduncle
4 ft. long, with a few empty bracts in the upper half. Spike
2 in. long, consisting of about 20 erecto-patent flowers; bracts
4-4 in. long, pale, with greenish ribs. Perianth 4—4} in. long;
segments pale pink, with a distinct green midrib. Stamens just
exserted.— Baron 3424! Allied to the Cape A. aristata,

Haworth.

ANTHERICUM ($ DILANTHES) DIANELLEFOLIUM, n. sp.

A. foliis vetustis basalibus in fibris copiosis dissolutis, productis elongatis
linearibus glabris venis crebris exsculptis, pedunculo nudo subtereti elon-
gato, racemo simplici, nodis inferioribus 2-3-floris, bracteis superioribus
parvis ovatis, pedicellis brevibus medio articulatis, perianthio albo segmentis
oblongis dorso laxe 5-nervatis, antheris magnis lanceolatis filamentis bre-
vibus, stylo elongato scabro.

Root-fibres many, cylindrical. Leaves rigidly erect, 11—2 ft.
long, 1-3 in. broad, with a distinct midrib and 10-12 crowded ribs
on each side of it; the margin scabrous. Peduncle stiffly erect,
about as long as the leaves. Raceme 2-3 in. long; lower bracts
long-pointed, from a broad base ; upper ovate-cuspidate, 1;—1 in.
long; pedicel shorter than the bract. Perianth 4 in. long.
Anther 4 in. long, twice as long as the filament. Style finally
exserted.— Baron 659! 4121! Nearly allied to the Cape A. tri-

florum, Ait.

CHLOROPHYTUM CHLORANTHUM, n. sp.

C. foliis dense rosulatis subpetiolatis lanceolatis membranaceis glabris,
pedunculo gracili nudo foliis breviore, racemis paucis laxifloris, nodis infe-
rioribus 2-3-floris, bracteis parvis acuminatis, pedicellis flore longioribus
medio articulatis, perianthio parvo viridulo, segmentis angustis, antheris
parvis, fructu latiore quam longo.

Root-fibres slender. Leaves 10-12 in a rosette, 8-9 in. long,

1 in. broad at the middle, narrowed gradually to the point and

530 MR. J. G. BAKER ON THE

obscure channelled petiole; ribs about 12; margin transparent.
Peduncle very slender, naked, 3 ft. long. Racemes about three,
2-3 in. long; pedicels 1-1 in. long; bracts };—¢ in. long, acu-
minate from a deltoid base. Perianth 4 in. long. Capsule of 3
acute few-seeded lobes each 1 in. long and broad.—.Baron 3825!

CHSIA SUBULATA, D. sp.

C. foliis basalibus pluribus subulatis glabris, pedunculo gracili foliis bre-
viore, racemis laxissimis paucifloris simplicibus, nodis omnibus unifloris,
bracteis minutis lanceolatis, pedicellis elongatis apice articulatis, floribus
ceruleis segmentis lanceolatis, genitalibus inclusis.

Leaves 10-12 to a rosette, 4 ft. long, 4 line broad low down,
distinctly ribbed, firm in texture. Peduncle very slender, 7-8 in.
long. Raceme about 4 in. long, 5-6-flowered ; lower bracts jin.
long; lower pedicels 4 in. long, deflexed after flowering. Perianth
4 in. long, strongly twisted after flowering; segments 4'; in. long.
Stamens 3 as long as the perianth.— Baron 3524! Nearly allied
to the Cape C. Eckloniana, Roem. & Sch.

ANEILEMA (§ LAMPRODITHYROS) TENERA, n. Sp.

A. caulibus elongatis glabris, foliis oblongis acutis subpetiolatis mem-
branaceis glabris, cymis paucifloris scorpioideis, axillaribus simplicibus
vel furcatis pedunculo elongato deflexo, terminalibus paucis paniculatis,
bracteis ultimis parvis ovatis, floribus parvis pedicellatis, sepalis oblongis
viridibus membranaceis, petalis parvis.

Stems slender,erect. Leaves 5-6 in.long, 11-2 in. broad at
the middle, narrowed to an acute point and a short narrowly
winged petiole. Side-eymes with a deflexed peduncle 1i-2 in.
long. End-panicle of about 4 lax cymes; main bracts oblong or
lanceolate ; pedicel as long as the flower. Sepals 4 in. long,
oblong. Anthers oblong, white ; cells parallel. Fruit not seen.
— Baron 49! 4118!

CorrorRyPE BARONI, n. sp.

C. caulibus glabris elongatis, foliis oblongis acutis subpetiolatis glabris,
vaginis laxis truncatis ore dense ciliatis demum ad laterem ruptis, cymis
multifloris, sepalis lanceolatis pilosis, petalis cæruleis deorsum in tubum
cylindrieum coalitis, filamentis dense barbatis, antheris oblongis loculis
parallelis.

Stems ascending, above a foot long. Sheaths above 4 in. long.
Leaves 3-4 in. long, 1-14 in. broad at the middle, narrowed gra-
dually to the blade. Clusters of flowers produced from the nodes
all the way down the stem. Calyx 4 in. long. Corolla-tube as

FLORA OF MADAGASCAR. 531

long as the calyx; segments cuneate, blue, jin. long. Style
reaching to the tip of the perianth-segments. Fruit not seen.—
Baron 3877!

CLADIUM FIMBRISTYLOIDES, n. sp.

C. foliis basalibus multis elongatis distichis linearibus rigidulis, caulibus
elongatis subteretibus foliis multis reductis przditis, spiculis parvis
oblongis pedicellatis in umbellas laxas simplices vel compositas peduncu-
latas ad foliorum axillas dispositis, glumis rigidulis castaneis, 4 exterioribus
parvis ovatis, interioribus fertilibus majoribus oblongis.

Basal leaves numerous, distichous, erect, firm in texture, 14 ft.
long, jin. broad. Stems about 3 ft. long, with one or more simple
or compound umbels from the axils of the distant reduced leaves
of the upper two-thirds. Spikelets + in. long, composed of 4 small
sterile ovate glumes and two longer oblong ones. Style and
stamens protruded a little beyond the tip of the glumes.—Baron
4193! 4517 is probably a form of the same species from a drier
soil. The inflorescence is much shorter (À ft. instead of 2 ft.
long), its bract-leaves much smaller, and the spikelets are longer
(2 in. long), with 6-7 sterile ovate glumes instead of four.

CYPERUS ($ PYCREUS) MONOCEPHALUS, n. sp.

C. foliis basalibus et caulinis pluribus productis linearibus, caulibus
elongatis gracilibus angulatis, spiculis pluribus lanceolatis in capitulum
globosum sessilem dispositis, bracteis 3—4 linearibus elongatis, glumis 10-
12-jugis ovatis albidis valde imbricatis, nucis obovoideis lenticularibus,
stigmatibus binis.

Stems densely cæspitose, erect, slender, 1-1} ft. long, with
several short linear leaves produced from the base and lower
quarter. Longer bracts 3-4 in. long. Spikelets 20 or more,
arranged in a single sessile head, each 3 in. long. Glumes jin.
long, ovate, whitish, with 3 ribs on the back. Stigmas protruded
beyond the glume.— Baron 3640! Allied to C. globosus, All.

CYPERUS ($ EUCYPERUS) SUBZQUALIS, n. sp.

C. foliis productis multis, caulibus gracilibus teretibus elongatis, umbellze
radiis multis omnibus simplicibus brevibus, bracteis parvis haud foliaceis,
glumis 4-6-jugis ovatis obtusis brunneis valde imbricatis, nucis obovoideo-
triquetris, stigmatibus tribus.

Proper leaves absent. Stems slender, stiffly erect, 3-4 feet
long. Umbel of 30-50 crowded rays, all about an inch long;
bracts small. Spikelets few or many in a head, 3-} in. long.
Glumes under 4; in. long, very convex on the back, bright brown

092 MR. J. G. BAKER ON THE

with a narrow green edge. Stigmas protruded a little beyond the
glume.— Baron 3639! Allied to C. equalis, Vahl.

CYPERUS ($ EvcyPERUS) PLATYCAULIS, n. sp.

C. foliis omnibus rudimentariis scariosis vaginantibus, caulibus dense
cespitosis elongatis applanatis, umbella radiis pluribus brevibus simplici-
bus vel parum compositis, bracteis parvis scariosis, spiculis parvis lanceo-
latis, glumis 6-8-jugis oblongis obtusis castaneis valde imbricatis, nucis
obovoideo-triquetris, stigmatibus tribus.

Stems erect, densely tufted, 3—4 feet long, quite compressed,
I in. broad, sheathed in the lower foot by rudimentary brown
leaves. Rays 20 or more, unequal, simple or slightly compressed,
at most 1-13 in. long. Spikelets few in each cluster, at most
3-1 in. long. Glumes 7; in. long, dark bright brown. Stigmas
protruding a little beyond the glume.—Baron 4456! Allied to
C. equalis, Vahl.

CYPERUS ($ EUCYPERUS) DEBILISSIMUS, n. sp.

C. foliis omnibus rudimentariis scariosis vaginantibus, caulibus longis-
simis debilibus angulatis, umbellz radiis 4-6 simplicibus bracteis paucis pro-
ductis linearibus, spiculis 6-8 lanceolatis, glumis 6-8-jugis ovatis brunneis
valde imbricatis, nucis obovoideo-triquetris, stigmatibus tribus.

Stems very slender, densely cxspitose, 3-4 feet long, with
about three brown sheaths, the bighest reaching up a foot above
its base. Umbels of spikelets all simple, varying from sessile to
peduncled, with stalks an inch long. Spikelets 4 in. long.
Glumes brown, much imbricated, 7; in. long. Stigmas pro-
truded beyond the glumes.— Baron 3374! Allied to C. equalis,
Vahl, and C. Haspan, L.

CYPERUS ($ EUCYPERUS) CUSPIDATUS, n. sp.

C. foliis 3-4 productis subbasalibus linearibus, caulibus gracilibus tri-
quetris, umbelle radiis 8-10 valde inzequalibus simplicibus vel parum
compositis, bracteis 3-4 linearibus foliaceis inzqualibus, spiculis paucis
digitatis lanceolatis, glumis 4-5-jugis ovatis cuspidatis castaneis valde
imbricatis, nucis obovoideo-triquetris, stigmatibus tribus.

Leaves 3-4, produced from the base of the stem, linear, ascend-
ing, 5-6 in. long. Stem very slender, erect, about a foot long.
Umbel very irregular, the longest branches about an inch long,
some simple, some slightly compound. Spikelets at most 4 in.
long. Glumes dark brown, 2 line long, narrowed suddenly to a

distinct cusp.— Baron 3854! Allied to C. dichrostachys, Hochst.
(Hildebrandt, 4016).

FLORA OF MADAGASCAR. 533

ARISTIDA (8 Cu ETARIA) MULTICAULIS, n. sp.

A. caulibus elongatis erectis strictis teretibus ramosis, foliis superpositis
lineari-subulatis glabris rigidulis, spiculis permultis purpureis in panicu-
lum amplam coarctatam ramis diu ascendentibus dispositis, glumis lanceo-
latis acuminatis arcte convolutis, florali in aristis tribus continuis longis
scabris producto.

Stems stiffly erect, 3 feet long exclusive of the panicle. Stem-
leaves 3-4, sheaths 3-4 in. long; blade 3-1 foot long, stiffly
erecto-patent. End-panicle a foot long; those of the side-
branches smaller; branches short, slender ; lower many in a
whorl. Spikelets, including the awns, about an inch long ; cen-
tral awn twice as long as the glumes, all the three slender and
erect.— Baron 4279! 4298!

RoTBOELLIA GRACILLIMA, n. sp.

Dense cæspitosa, foliis basalibus subulatis gracillimis strictis erectis,
caule simplici folis 2-3 reductis przdito, spiculis in spicam elongatam
dispositis, rhachi gracili glabro articulato sulcato, gluma sterili exteriore
oblongo-lanceolata rigidula purpurea internodio zquilonga dorso rotundata
levia, glumis reliquis brevioribus membranaceis.

Nearly allied to R. cespitosa, with which it quite agrees in
general habit; but the leaves are even more slender, the stems
longer and more leafy, the unexpanded spikes very slender and
coloured purple, and the outer sterile glume rather longer and
not ribbed on the back.— Baron 4457 !

RoTBOELLIA CHSPITOSA, n. Sp.

Dense czespitosa, foliis subbasalibus elongatis subulatis gracilibus strictis
erectis, caule simplici folio reducto ad spice basin przdito, spiculis in
spicam elongatam dispositis, rhachi gracili glabro articulato sulcato, gluma
sterili exteriore oblongo-lanceolata erecto-patenti internodio equilongo,
reliquis brevioribus membranaceis.

Leaves and stems densely clustered, stiffly erect, the former
1-1} feet long, pale green, glabrous, not more than 1 line in diam.
Flowering-stems, spike included, shorter than the leaves, with
a leaf clasping the base of the spike, which is 5-6 in. long. In-
ternodes and outer glumes 4 in. long, the latter rigid in texture,
pale green, erecto-patent when the flower expands, furnished
with 3 lax ribs on the back.— Baron 4091! Allied to the North-
American R. campestris, Nuttall.

DEYEUXIA EMIRNENSIS, n. sp.
D. caulibus gracilibus erectis, foliis eaulinis paucis remotis linearibus gla-

534 ME. J. G. BAKER ON THE

bris, spiculis parvis pilosis pedicellatis in paniculam laxam elongatam ramis
flexuosis gracillimis ascendentibus dispositis, glumis omnibus oblongo-
lanceolatis zequilongis, florifera hyalina membranacea breviter aristata.

Stems rather above a foot long, panicle included. Stem-
leaves 3-4, the central sheaths 2-3 in. long; blade linear, gla-
brous, spreading or ascending, under half a foot long. Panicle
half a foot long, the ascending branches very slender and flexuous.
Spikelets 3 in. long, with a tuft of white hairs at the base inside;
sterile glumes oblong-lanceolate, white with a green keel; flower-
ing-glumes entirely hyaline, shortly awned.— Baron 4197 ! Adds
this temperate genus to the Madagascar flora.

DIPLACHNE ARISTATA, n. sp.

D. caulibus elongatis teretibus erectis, foliis remotis linearibus firmis
glabris utrinque conspicue nervatis, spiculis 5-6-floris dense pilosis in
panieulam amplam densam ramis ascendentibus gracillimis dispositis, glu-
mis sterilibus parvis scariosis oblongo-lanceolatis, glumis floriferis lanceo-
latis apice bifidis inter dentes aristatis.

Stems moderately stout, terete. Stem-leaves spreading, 6-8 in.
long, 4 in. broad, firm and thick in texture, flat, pale green on
the face, darker green on the back, with numerous distinctly
raised ribs on both sides. Panicle nearly a foot long, all the
branches very slender and ascending, the lower remote, the upper
crowded. Spikelets 1 in. long, villose, with dense soft white
hairs; sterile glumes subequal, erecto-patent, brown, scariose,
1; in. long; flowers laxly alternate on the axis, erecto-patent;
flowering-glume 2 in. long, including the short ascending or
squarrose awn.— Baron 4945! Allied to the Cape D. andropo-
gonoides.

VASCULAR CRYPTOGAMIA.

CYATHEA RIGIDULA, n. sp.

C. frondibus amplis rigide subcoriaceis viridibus glabris, pinnis oblongo-
lanceolatis rhachidibus inermibus facie pilosis dorso parce paleaceis, pin-
nulis lanceolatis ad costam pinnatis, segmentis tertiariis oblongis contiguis
inciso-crenatis margine recurvatis, venulis furcatis 6-8-jugis, soris densis
costularibus, indusio parvo campanulato ore truneato.

A large tree-fern, with the habit of C. arborea, rigid in texture,
green on both surfaces. Rhachides of the pinne and pinnules
minutely paleaceous. Pinne above a foot long, 31-4 in. broad ;
pinnules very numerous, spreading, 4 in. broad, cut down to the

midrib into crowded ascending segments jl in. broad. Veins

FLORA OF MADAGASCAR. 535

very distinct. Sori filling all the segments except the edges and
tips, often 8-10 to each. Indusium not more than half the size
of the globose sorus, persistent, quite entire and truncate at the
throat.— Baron 3845 !

CYATHEA LEPTOCHLAMYS, n. sp.

C. frondibus amplis bipinnatis utrinque viridibus facie glabris dorso parce
piloso-paleaceis, pinnis oblongo-lanceolatis rhachibus inermibus facie dense
pilosis dorso subcalvatis, pinnulis lanceolatis sessilibus sursum crenatis
deorsum ad medium pinnatifidis segmentis contiguis oblongis, venis parce
pinnatis, venulis 3-4-jugis ascendentibus simplicibus, soris ad pinnularum
costam utrinque laxe uniseriatis, indusio membranaceo fragili.

A large tree-fern, allied to C. canaliculata, Willd. Fronds
moderately firm in texture, green on both surfaces, paleaceo-
pilose on the main ribs beneath. Pinn: above a foot long, 23 in.
broad. Pinnules very numerous, alternate, sessile, but not adnate,
4 in. broad, the largest cut halfway down to the rhachis into con-
tiguous ascending oblong lobes js in. broad. Veinlets not more
than 3-4 on a side opposite each lobe, distinet, simple, very
ascending. Sorus one near the base of each lobe, a short dis-
tance from the midrib of the pinnule. Indusium membranous,
persistent, ;, in. in diam., breaking up irregularly ; receptacle
large, barrel-shaped, glabrous.—.Baron 3665 !

POLYPODIUM ($ PuEaorrEnIs) BARONI, n. sp.

P. frondibus amplis bipinnatis utrinque viridibus glabris, rhachidibus
furfuraceis minute paleaceis, pinnis oblongo-lanceolatis, pinnulis lanceolatis
brevissime petiolatis ad costam pinnatis, segmentis tertiariis lineari-
oblongis erenulatis adnatis, venis pinnatis venulis 5-6-jugis simplicibus
obscuris, soris submarginalibus demum confluentibus.

General habit like that of the phegopteroid forms of Nephro-
dium villosum. Rhachides thinly coated with ferruginous tomen-
tum and furnished with a few small linear brown pales. Longest
pinne in the specimen 9-10 inches long. Longest pinnules 2 in.
long, i-$ in. broad. Lower tertiary segments j in. broad, as-
cending, adnate at the base, with a distinct space between each ;
veinlets distant and indistinct. Sori placed at the tip of the
veins, superficial.— Baron 3920!

LYCOPODIUM XIPHOPHYLLUM, n. 8p.

L.caulibus simplicibus erectis angulatis, foliis laxe dispositis lanceolatis
firmulis nitide viridibus integris patulis vel subsquarrosis basi decurrentibus,
sporangiis ad foliorum centralium axillas solitariis sessilibus.

536 ON THE FLORA OF MADAGASCAR,

Allied to Z. serratum and L. Hamiltonii. Stem 4 foot long,
stiffly erect from a decumbent base, simple in the specimen seen,
brownish, strongly angled, not hidden by the leaves. Leaves
3 in. long, +}; in. broad, acute, decurrent at the base, flat, bright
green, moderately firm in texture, with a very distinct midrib.
Sporangia placed in the axils of entirely unaltered leaves, strami-
neous, reniform, a line broad.— Baron 4512!

SELAGINELLA ECHINATA, n. sp.

S. caulibus brevibus decumbentibus copiose bipinnatis, ramis sut ulieque
laxe dispositis erecto-patentibus, foliis densissimis distichis biformibus
rigidnlis pallide viridibus dense ciliatis apice acuminatis stramineis sub-
pungentibus, planz inferioris ovato-lanceolatis patulis, plane superioris
angustioribus ascendentibus squarrosis, spicis ignotis.

A very distinct species, intermediate in habit between S. ru-
pestris and S. denticulata. Stems 2-4 in. long, with rootlets
extending to the tip; branches of all grades alternate, distant,
erecto-patent; ultimate branchlets very short; leafy branches
à in. in diam. Leaves firm in texture, grey-green, densely cili-
ated. Leaves of the lower plane very acuminate from an ovate
base, 4; in. long, with a distinct nearly central midrib. Leaves of
the upper plane not much shorter, but narrower, ascending, and
recurved. Fructification none in the specimens, but no doubt that
of § Stachygynandrum.— Baron 4226 !

ON THE FLORA OF MADAGASCAR. 537

Posrscrirt.—Since the preceding pages were printed, the
following description of a new species of Chara has been received
from Messrs. H. and J. Groves.

NrrELLA Baront, H. & J. Groves, sp. nov.

Habitus proxime N. pseudo-flabellate. Caulis *4 millim. cras-
sus. Ramuli verticillorum fertilium circa 8, similes, 3—4 furcati ;
radiis divisionis prime plerumque 7, secunde 5-6, tertie et
quartz: 4. Segmenta primaria ramulorum totium longitudinem
dimidiam zquantes. Segmenta ultima bicellularia, cellula in-
feriora *05—06 miliim. crassa, superiora ‘1 millim. longa, :03
millim. crassa. Fructus solitarii, in divisione secunda aut tertia
siti, '53 millim. longa, *4 millim. crassa, circa 9-striati, oospora
'34 millim. longa :23 millim. crassa, liris inconspicuis, coronula
brevi. Antheridia iu divisione secunda.

Differs from N. pseudo-flabellata by the fruit and antheridia
being separate, neither being produced at the first forking, and
by the inconspicuous ridges of the oospore.— Baron 3420.

LINN. JOURN. BOTANY, VOL XXII. 2v

INDEX.

Abies amabilis, 170, 171, 172, 173, 174,
185, 192, 212; balsamea, 185; bi-
folia, 184 ; campylocarpa, 189 ; cana-
densis, 108; concolor, 177, 178, 179,
180; Fortunei, 196, 197, 198, 199;
Gordoniana, 174, 180; grandis, 171,
172, 174, 175, 176, 180, 182, 184, 187,
212, var. densiflora, 171, var. Low-
iana,176,177,179,180, var. pallida,177,
179; hirtella, 195, 196 ; Keempferi,
208 ; Jezoensis, 197 ; lasiocarpa, 175,
180, 181, 182, 184, 185; Lowiana,
175, 180, 181, 182; magnifica, 173,
186, 187, 192, 193, 194; nobilis, 187,
188, 192, 193, 194, 212, var. glauca,
189, var. magnifica, 189, 190, 191 ;
robusta, 191, 192, 193; Nordman-
niana, 173: Parsonsiana, 175, 180,
181; pectinata, 108; religiosa, 194,
212; subalpina, 172, 183, 185, 186,
187.

Abronia arenaria (cotyledon), 371; um-
bellata (embryo), 370, 371.

Acacia decipiens, 106, 107 ; macracan-
tha, 151; sp.? (Peru), 4 151 ; tor-
tuosa, 151 ; xiphoclada, 468.

Aceena, 15; lappacea, 37 ; sp. ?, 37.

Acalypha indica, 338.

Acanthads (Naga Hills), 135.

Accumbent and incumbent cotyledons,
374.

Acer, 357; platanoides, 104; Pseudo-
Platanus, 108, fruit and embryo,
362, seedling of, 351, 362, sect. of
seed, 399, sessile cotyledon, 375,
straight veins of, 341.

Aceranthus diphyllus, 432.

Achillea Millefolium, 107.

Achlys (Japan), 435; triphylla, 4395,
496, 437, var. typica, 435, var.
japonica, 435, 436, 437.

Achyranthes aspera, 338 ; velutina, 334.

Achyrocline satureioides, 44.

Achyrophorus elatus, = H ypochseris ela-
ta, 48; Lechleri, 165; Meyenianus,
48 ; taraxacoides.=Hypocheris ta-
raxacoides, 48 ; tenerifolius, 165 ; ten-
uifolius, 165.

Acianthus petiolatus, 278.

Aconitum Napellus, 104; lycoctonum,
104.

Acrocrypheea (African), 311.

Acrostichum antrophyoides, 231; au-
reum, 231; bicuspe, var. integri-
folium, 231; modestum, 231; quer-
cifolium, 231; repandum, var. Quoy-
anum, 230; spicatum, 231; stenose-
mioides, 230 ; Wallii, 231.

Actsea spicata, 104.

Actinidia (Jakpho), 133.

Adabo or Adabolahy, native name for
Ficus (Covellia) sakalavarum, 523.
Adiantum sethiopieum, var. =A. chi-

lense, 63; Capillus-veneris, 224.

Ægopodium Podgraria, 104.

Konia rosea, 122.

JEranthus Arachnites, 121, 122; Cal-
ceolus, 122; grandiflorus, 121, 122;
polyanthemus, 121.

ZEschynanthus, 135.

oe Hippocastanum, 104, (ovule)
397.

Afzelia bijuga, 335, 337.

Ageratum conyzoides, 337.

Agrostis foliata, 59; nana, 58; vires-
cens, 59.

Ailanthus glandulosus, 108, (seed)
364.

Ainsliæa (Jakpho), 133.

Ajuga, 441; oocephala, 514.

Akebia (Japan), 424; lobata, 425, var.
a. clematifoliola, var. 8. quercifolia,
425 ; quinata, 424.

Albizzia trichopetala, 468.

Alchemilla pinnata, 5, 37.

Alchornea, sp. ?, 143.

Algæ (Diego Garcia), 340.

Alisma Plantago, 107.

Allium euosmon, 167 ; striatum, 167.

Alnus, 131.

Aloe arborescens, 111 ; aristata, 441.

Aloe (Eualoe) Haworthioides, 529.

Alonsoa acutifolia, 53; (Chicla) 5.

Alsophila Burbidgei, 222 ; contaminans,
222; latebrosa, 222; pruinata, 167 ;
ramispina, 222.

540

Alstroemeria aurantiaca, 110 ; peregrina,
56.

Alternanthera (Peru), 3.

Althzea (lobed cotyledon), 378.

Alyxia, 441; lucida, 503.

Amarantacee (Diego Garcia), 338 ;
(Peru), 9, 55.

Amarantus caudatus, 106; chlorosta-
ehys, 55.

Amaryllideæ (Peru), 9, 56.

Ambrosia artemisisfolia, var. peruviana,
45

America, S.W., Botany of (Ball), 137.
Ampelocissus, sect. of Vitis, 460.
Amphorchis calcarata, 125.
Amphoroealyx, 476 ; multiflorus, 476.
Amsinckia angustifolia, 51.
Amygdalus communis, 104; persica,
104.
Anacalypta stellatifolia, 306.
Analysis Gen. & Sp. Gamopetalous Di-
cotyledons in Andean Flora, 18.
Anaphalis (Jakpho), 136.
Anchusa Kunthii, 51 ; leucantha, 51;
linifolia, 51; oppositifolia, 51.
Andean flora, 18.
André, M., opinions regarding flora of
Equatorial Andes, 6.
Andromeda speciosa, 104.
Aneilema (Kohima), 129;
dithyros) tenera, 530.
Anemone japoniea, 104.
Angiopteris evecta, 231.
puru Caleeolus, 122; palmiferum,
21.
Anisothecium varium, 301.
Anivona, 527 (see Ouvun).
ane a Mariei, 305; pusillum,
305.
Anomodon devolutus, 318 ; Toccoee, 318.
Anona Cherimolia, 3 ; (germinating
seedling), 387.
Anthemis Cotula, 165.
Anthericum (Dilanthes) dianellefolium,
929; eccremorhizon, 56.
Anthocleista amplexicaulis, 506; rhizo-
phoroides, 506.
Anthopterus (N. Granada),
"pi 143; racemosus, 143; Wardii,
2.
Antidesma, 136; alnifolia, 519; arbu-
tifolia, 519; brachyscypha, 518.
Antiphytum linifolium, 91; Walpersii,
51.

(Lampro-

Antirrhinum junceum, 152, 153, 154;
majus, 108; speciosum, 152. :

Antrophyum parvulum, 230,

Apatales purpurascens, 245, 279,

Aphelexis flexuosa, 492; stenoclada
493; sulphurea, 493. :

147 ; mucro- :

INDEX.

Aphloia minima, 444.

Apium leptophyllum = Helosciadium
leptophyllum, — Ammi leptophyllum,
41; Petroselinum, 105.

Apocynacer (Diego Garcia), 337.

Apodytes emirnensis, 458.

A pospory, distinctions of kinds, 440.

Apospory in Polystichum angulare, var.
pulcherrimum (Druery), 437.

Apostasiacez, develop. inner row sta-
mens, 415.

Aquilegia vulgaris, 110.

Aralia letevirens, 167.

Araliacez (Naga Hills), 135.

Araucaria Bidwellii, 104; imbricata, 107.

Arauja serifera, 108. :

Archidium africanum, 299 ; alternifo-
lium, 299, 300; Ecklonii, 299.

Ardisia, 135; dissitiflora, 501; lepto-
clada, 500.

Areca madagascariensis, 527.

Arenaria Alpamarce, 31.

Argyreia (Nambre forest), 135.

Argyrobryum (African), 307.

Arica, plants around, 156.

Aristida (Chetaria) multicaulis, 533.

Arnottia, 116; inermis, 127.

Artemisia dracunculus, 111.

Arthrolobium, from Kilimanjaro, 299.

Arundina pentandra, 402.

Arundinella, 132.

Asarum europæum, 104.

Asclepiadeæ (Peru), 9, 17, 49.

Asclepias Curassavica, 104, 339; incar-
nata, 105.

Asophila glabra, 130. m

Aspidium aristatum, 225 ; platanifolium,
225; semicordatum, 225.

Asplenium equibasale, 225, 332, 335,
338 ; (8 Diplazium) sequibasale, 225 ;
cordifolium, 225, var. integrifolium,
225 ; falcatum, 338 ; fragile, 64 ; Gil-
liesianum, 64; hirtum, 224; longis-
simum, 338; lineolatum, 225 ; macro-
phyllum, 224 ; maximum, 225; nidus,
338; nitidum, 225 ; Laeti wed.

olypodioides, 225; porp yroraens,
bag E porrectum, 225 ; trifidum
(African), 299; triphyllum, 64; vr
ride, 64.

Astephanocarpa arbutifolia, 493.

Aster chinensis, 105 ; pyrenseus, 106.

Asteropeia, 441; sphserocarpa, 479. 3

Astragalus Alpamarce, 36; casap
tensis, 36 ; Deis 36, var. ser-

ens, 36; uniflorus, 36. dirae

p enes cupressoides, 201 ; laxifolia,
200, 201 ; selaginoides, 201. ie

Athyrium, 130; Filix-fcemina, var. €
rissima, 437, 440.

INDEX.

Atraphaxis spinosa, 106.

Atropa Belladonna, 105 ; biflora— Hebe-
cladus biflorus, 53.

Aulacomnion, 306.

Auricled cotyledons, 390.

Baccharis alpina, 44, var. serpyllifolia,
44; (Andes) 21; cespitosa, 44;
Chilco, 44; Chilquilla, 159 ; conferti-
folia, 130, 158, 159; eupatrioides,
164; genistelloides, 44, var. alpina,
44; glutinosa, 160; lanceolata, 156 ;
magellanica, 167 ; marginata, 159,
160; obovata, 44; parviflora, 159;
(Peru), 3 ; sp., 44.

Baker, J. G., Collection of Ferns from
N. Borneo, 220.

, Flora of Madagascar, 441.

Balanophora elongata, 332 ; indica, 330,
331, 332 ; Thwaitesii (Trimen on),
330, 331, 332.

Balbisia (Peru), 13; verticillata, 34.

Balboa (Andes), 146.

Ball, J., Flora of Caldera, N. Chili, 158.

, Flora of Payta, N. Peru, 148.

———, Flora of Peruvian Andes, 1.

—, Flora of Rimac Valley (Peru), 9,
28

, Notes on Botany of Western South
America, 137.

Balsaminese (Madagascar), 454.

Balsamodendron, 441; fraxinifolia, 459;
laxiflora, 459.

Bambusez (Chiloe), 140.

Barbarea vulgaris, 106.

Barleria Kitchingii, 510; phillyresefolia,
510.

Barnadesia, 14.

Baron, Rev. R., Madagascar Orchids,
116.

Barringtonia speciosa, 337.

Bartsia (Andes), 18; gracilis, 53; patens,
53.

Bazzania convexa, 322 ; decrescens, 322;
deflexa, 322; pumila, 322.

Bean, embryo of, 348.

Beech (involucre), 401.

Begonia, 441; Baroni, 480; (Quadri-
lobaria) fragilis, 479; (Naga Hills),
135 ; hydrocotylifolia, 104; quinqui-
folia, 104.

Belbunium tuberculosum=Rubia tu-
berculosa, 42.

Belmontia, 441 ; emirnensis, 507.

Berberis Bealei, 428; latifolia, 428;
(Japan), 426; Mariesii, 427; mon-
tana, 421; (ovule), 347; Sieboldi,
426, 427, 428; sinensis, 427; sp.?,
110; Thunbergii, 426, 427, var. f.

541

Maximowiczii, 426, var. a. typica,
426; Tschonoskiana, 426, 427; vul-
garis, var. japonica, 426, 427.

Beschorneria Tornellu, 111.

Betula alba, 104.

Bicornella parviflora, 123; gracilis,
125.

Bidens andicola, 45 ; bipinnata, 46 ;
(Chicla), sp.,5; chilensis, 46; hu-
milis, 45; macranthus, 45; (Peru),
3; pilosa, 46; tripartita, 106.

Bignonia insignis (cotyledon), 387.

Bignoniaceee (cotyledon), 381; (Peru),
9, 13, 54.

Bishop of Singapore and Sarawak,
Ferns of N. Borneo coll. by, 222.

Blechnum Finlaysonianum, 224; orien-
tale, 224.

Bloom on Leaves and Distribution of
Stomata (Darwin), 99.

Bloom, symbols used in, 104.

, both surfaces leaf, 110.

, either surface leaf, 104.

—— ., lower surface leaf only, 108.

———, upper surface leaf only, 109.

Blumenbachia carduifolia — Loasa car-
duifolia, 39.

Blyxa radicans, 236, 241; Roxburghii,
232, 233, 236, 237.

Boeckea, (ftnote) 220.

Boerhaavia diffusa, 338 ; viscosa, 3.

Bois Mapan, on Rose-tree, 333, 336.

Bolus, H., Contributions to S.-African
Botany.— Orchidee, Part II. With
additional notes by N. E. Brown,
65.

Bomarea Caldasii, 110.

Bombaces, S.-Amer., 145.

Bonavia, Dr. E., Wild Source of Cul-
tivated Limes, 213.

Bonplandia = Caldasia, (ftnote) 41.

Boottia abyssinica, 238, 239; cordata,
233, 240; crassifolia, 239; exserta,
240; scabra, 239.

Doragines (Diego Garcia), 337 ; (Peru),
9, 50.

Borneo, Ferns from (Baker), 222.

Borreria podocephala, 142.

Botany of W. S. America (Ball), 137.

Bourne, G. C., coll. plants from Diego
Garcia, 333.

Bowlesia acutangula, 40; incana, 41;
lobata, 40, var. chilensis, 40, var. hu-
mifusa, 40; palmata, 40; pulchella,
41.

Brachymenium (African), 306.

Brachythecium afrosalebrosum, 318;
Dicksoni, 318; Knysns, 318 ; impli-
catum, 317; salebrosum, 318 ; velle-
rum, 318.

542

Brassica (cotyledon), 369; nigra, 345 ;
oleracea, 110; Rapa, 105.

Braunia sciuroides, 310; indica, 310;
Schimperi, 310.

Breweria tilizfolia, 508.

Brexia madagascariensis, 104.

Broad cotyledons, 363.

Bromelia bicolor, 166.

Bromus auleticus, 63 ; frigidus, 63 ;
Henkeanus, 63; lenis, 63; tenuis,
63; unioloides, 63.

Broussonettia papyrifera, 104.

Brown, N. E., Orchidesw of S. Africa,
(see Bolus), 65.

Brownleea, 116; cerulea, 127; mada-
gascarica, 126.

Bryonia laciniosa (cotyledon), 375, 376,
385 ;` dioica (seedling), 375, 376,
385.

Bryum alpinum, 307; argenteum, var.
argyrotrichum, 307 ; capitulatum,
306 ; Commersonii, 307 ; coronatum,
307; flexifolium, 306; giganteum,
307; grandifolium, 307; integrifo-
lium, 307; nipalense, 306; pul-
chrum, 306; roseum, 307; umbra-
culum, 307.

Buckwheat, seed of, 346.

Buddleia incana, 155; spheerocalyx,
505 ; spheerocephala, 505.

Bulbophyllum approximatum, 117 ; Ba-
-ronii, 117; conitum, 119; coriopho-
rum, 119; gracile, 118; (§ Bulbo-
phyllaria) ophiuchus, 118; seychel-
larum, 117.

Bupleurum fruticosum, coriaceous
leaves, 343 ; populifolius, 343.

Burmanniaceæ, develop. inner low sta-
mens, 415.

Butea minor, 131.

Buxus sempervirens, 104.

Byrsocarpus Baroni, 462.

Byttneria Bauhinoides, 451; biloba,
45l; heterophylla, 45l ; Melleri, 451.

Cacti of Peruvian Andes, 3.

Cesalpinia Gilliesii, 156.

Cæsia subulata, 530.

Caladium, sp., 110.

Calandrinia, 15 ; acuulis, 32; caulescens,
32; Sp. P (dwarf var. of caulescens),

Calanthe crenata, 111 ; Masuca, 65 ; na-
talensis, 65 ; sylvatica, var. natalensis,
Calceolaria, 15; bartsiæfolia, 53; bi-
flora, 53; lobata, 53; ovata, 53;
tenuis, 5, 53; virgata, 53.
Calcevlarise (Chicla), 5 ; (Peru), 4.

s

INDEX.

Caldasia, 15; (=Bonplandia), ftnote,
41; andicola, 41.

Caldera, Chili, flora of, 158.

Calendula officinalis, 106.

Callicosta (African), 309.

Calopyxis malifolia, 474. ,

Calycanthus, 392.

Calyciflore (Madagascar), 462.

Calymperes Afzelii, 303, 304, 340;
Dozyanum, 304 ; ($ Hyophilina) Gar-
cim, 339; ligulare, 303.

Calyptothecium africanum, 312.

Camellia japonica, 104. .

Campanula persicifolia, 105; urticæ-
folia, 105.

Campanulaces (Peru), 9, 49.

Campylopus perpusillus, 301.

Canarium (ovule), 397.

Cantua (Andes), 20.

Capparideæ (Rimac), 9, 17, 30.

Caprifoliacezg (Rimac), 9, 41. -

Capsella Bursa-pastoris, 8, 29, 105;
(Rimac), 5.

Capsicum frutescens, 339.

Cardamine chenopodifolia, 391.

Cardiopteris, 136.

Carex speciosa, 133.

Carlemannia (Naga Hills), 135. _

Carpha paniculata, 168; Pervilleana,
483.

Carphalea, united with Dirichletia, 482.

Carpinus betulus, 105.

Carum carvi (cotyledon), 371, 398. —

Caryophyllaceæ (cotyledons), 364; (Rt-
mac, Peru), 9, 10, 30. .

Cassia coquimbana, 161; floribunda,
108.

Cassytha filiformis, 338. .

Castilleja fissifolia, var. pumilla, — pu-
mila, 53.

Castor Oil, seed of, 348.

Casuarina equisetifolia, 338.

Casuarines (Diego Garcia), 338.

Catulpa Kempferi (seedling), 358.

Caulerpa mamillosa, 340.

Caulophyllum (Japan), 428 ; robustum,
422 ; thalictroides, 428.

Celosia (Lagrezia) micrantha, 814.
Celtis, 441; gomphophylla, 521; 0¢
cidentalis, 105. .
Centaurea melitensis, 8, 47 ; phrygia,

105.
Cephalandra, leaves entire, 341.
Cephalocroton cordifolius, 520.
Cephalophora litoralis, . :
Cephalotaxus Fortunei, 201, 202 ; oe
nii, 130; poiunoniate, 201, 202, 20%
var. spheeralis, 203, 212. z
Cerastium arvense (seedling) 954, 066 f
glomeratum, 8, 31; vulgatum, 94

INDEX.

Cerastium, narrow cotyledons, 844.
Cerasus Mahaleb, 105; vulgaris, 105.
Ceratanthe, 325.

Ceratocephalus faleatus, (seedling) 350.

Cereus Napoleonis, (seed) 371; (Peru)
3, 355.

Cespedesia (S. Amer.), 146.

Chzelotropis andina, 58.

Cheetanthera, 14.

Cheetospora antarctica, 168 ; laxa, 168 ;
rhyncosporoides, 168.

Chagos archipelago, coll. plants by G.
C. Bourne, 333; Diego Garcia, an
atoll 333; flora of, 332 ; visited by,
333.

Chandonanthus fimbriatus, 322; hir-
tellus, 321 ; indicus, 322 ; javanicus,
322; rufus, 322; squarrosus, 322.

Chara, new sp. Madagascar, 537.

Cheilanthes crenata, — Woodsia obtusa,
63; Matthewsii, 63.

Cheiranthus Cheiri (ageumbent cotyle-
don), 374, 375, 381; pygmaeus (coty-
ledon), 398.

Cheirostylis gymnochiloides, 127.

Chelidonium majus, 110.

Chenopodiacez (Andes, Peru), 9, 55.

Chenopodiacearum sp., 160.

Chenopodium ambrosioides, 55 ; Bonus-
Henricus, 106, 351, (seed) 361;
seedling of, 351; cotyledon of, 350.

Chili, climate and veget. of, 139.

———, flora of Lota, 161; rainfall in,
(ftnote) 161, 162.

Chirita acuminata, 135 ; pumila, 135.

Chlora perfoliata, 110.

Chlorophytum chloranthum, 529.

Chonos archipelago, climate and flora of,
140. .

Choripetalum, 135.

Chrysanthemum Leucanthemum, 106.

Chuquiragua, 14; spinosa, var. iner-
mis, 47.

Cineraria, 441; Anampoza, —Senecio
Anampoza, 496.

Cissus Mappia, 335.

, Sect. of Vitis, 461.

Citrus acida, 213, 217,218; acida, per-
sistence winged petiole of. 213; Au-
rantium, 105, (cotyledon) 400; Hys-
trix, 214, 217, 218; limonum, 217;
medica, 217, 218, var. acida, 213.

Cladium fimbristyloides, 531.

Cladophora demissa, 340,

Clarionea ciliaris — Perezia ciliaris, 47.

Clarke, C. B., Botanical Observations
in Journey to Naga Hills, 128.

Clematis (Chicla), 5; Peruviana, 28;
recta, (cotyledon) 398; viticella, 105.

543

Cleome chilensis, 30.

Clerodendron (Cyclonema) mirabile,
513.

Clidemia dependens, 141; hirta, 141.

Climatal relations to flora of W.S. Ame-
rica (Ball), 137.

Clitoria, 372, 375; ternatea (seedlings),
355; zanzibarensis, 466 ; var. of he-
terophylla, 465.

Clusia pseudomangle, 141; sp.?, 141.

Clusieæ (Equator. Amer.), 146.

Clusiella ( Andes), 146.

Cnicus eriophorus, 131.

Cobza scandens, 108, (seed) 364.

Cochranea stenophylla, 161.

Codonocanthus, 135.

Codonopsis (Jakpho), 133.

Ceelogyne, 134.

Coffea (Naga Hills), 135.

Coldenia dichotoma, 151, 160; litoralis,
160.

Coleotrype Baroni, 530. .

Colignonia biumbellata, 54; parviflora,
54; scandens, 54.

Collomia (Andes), 20; gracilis, 50. —

Colombia, flora of Buenaventura in,
141.

. Colura, 325, 329.

Combretaces (Diego Garcia), 337.

Commelina fasciculata, var. =C. ner-
vosa, 56; (Kohima) 130, (Naga Hills)
130, (Peru) 9, 13,56. ————

Commiphora, 441; fraxinifolia, 459;
laxiflora, 459. .

Comparisons plants, Weddell's ‘ Chloris
Andina' and Rimae Valley (Ball),
16, 17.

Composite (Andes, Peru), 3, 9, 10, 13,
14, 16, 43, 392; (cotyledons) 364;
(Diego Garcia) 337 ; (not recent)
21

Concepcion, a doubtful plant locality,
157. . .

Conclusions regarding seedlings (Lub-
bock), 395. Lui

Conformation of flower of Cypripedium,
402. . i

Conifers, history of certain species of
(Masters), 169. —

Conomitrum Metzgeria, 819.

Conostegia polyandra . .

Contributions to Study of Relative
Effects different parts of Solar Spec-
trum on Transpiration of Plants,
Rev. G. Henslow on, 81.

—to Flora of Peruvian Andes, J.
Ball on, 1. ; .

——— to History of certain Species of
Conifers (Masters), 169.

544

Convallaria majalis, 109, (ovules) 397. |

Convolvulacee (Andes, Peru), 9, 51;
(cotyledons) 381 ; (Diego Garcia) 337,
(Jakpho) 134.

Convolvulus, 357; arvensis, 105 ; Sol-
danella (embryo), 381, 382; tricolor,
106.

Conyza amplexicaulis, 489 ;
folia, 44.

Coquimbo, plants of, 161.

Corchorus hamatus, 452.

Cordia, 392, (ovule) 397; subcordata,
(erenate cotyledon) 336, 337,373, 374 ;
(seedling) 357.

Coreopsis auriculata, 366, 367; fasci-
culata, 45 ; filifolia, (achene and em-
bryo) 366, (seedling) 350.

Coriifoliæ, sect. gen. Liparis, 257.

, diagnosis, 257, 282.

, key diag. spec. belonging to, 257-
259.

Corydalis capnoides, 110.

Corylus Avellana (seed of nut), 397;
foss. Mull, 219.

Cotula coronopifolia, 165.

Cotyledon bracteosum, 111; Chiclensis,
37; Incarum, 37.

Cotyledons and Leaves, forms of (Lub-
bock), 341, 349, 363; shape of no
influence in form of seed, 345.

Cotyledons, 350, 363, 369, 372, 373,
378, 379, 385, 390; accumbent and
incumbent, 374; auricled, 390;
broad, 363; crenate, 373; divided,
385; emarginate, 379; lobed, 378;
petioled, 375 ; size of, 394; unequal,
369; unsymmetrical, 372.

Crambe maritima, 110.

Crassula, 441; fragilis, 469.

Crassulacee (Andes, Peru), 9, 37.

Crategus oxyacantha, 105.

Crawfurdia, 135.

Credneria, foss., 220.

Cremolobus aphanopterus, 29 ; parvi-
florus, 29; pinnatifidus —C. parvi-
florus=C. aphanopterus, 29; rhom-
boideus, 30 ; sinuatus, 30.

Crepis, 48; Lechleri, 165 ; virens, 165.

Cress, seedling of, 345.

Crinum capense, 111 ; erubescens, 143;
Hildebrandtii, 339; (Platygaster)
modestum, 528.

Cristaria foliosa, 156 ; seselifolia, 157;
Spinole, 156; viridi-luteola, 157,
158.

Crotalaria luteo-rubella, 462; macro-
poda, 463.

Croton vernicosus, 519.

Crucifere (Rimac, Peru), 9, 10, 29.

sophis-

INDEX.

Cruickshanks, coll. of plants from W.
coast of S. America, 157.

Cruikshanksia hymenodon, 159; tri-
partita, 159.

Cryphza concavifolia, 311; dentata,
311; incurvifolia, 311 ; julacea, 311 ;

laxifolia, 310, 311; patens, 311;
spherocarpa, 311; Welwitschii,
311.

Cryptocarya pauciflora, 515.

Cryptomeria, foss. Mull, 219.

Cucurbitacese (Andes, Peru), 9, 39,
387.

Cultivated True Limes, wild source of,
Dr. E. Bonavia on, 213.

Cunninghamia sinensis, 210.

Cuphea, 357, 379; silenoides, (sect.
seed) 380, (stages growth seedling)
390.

Cuseuta chilensis, 51.

Cyathea Brunonis, 222; leptochlamys,
535; Lobbiana, 222 ; rigidula, 534.

Cyathodium africanum, 327; aureo-
nitens, 327.

Cyclamen (cotyledon), 371.

Cyclea, 41 ; madagascariensis, 443.

Cymbidium bitubereulatum, 263; cly-
peolum, 266; plantagineum, 287;
(Kohima), 130; reflexum, 287 ;
vexilliferum, 265.

Cynaroidez, 14.

Cynoglossum revolutum, 51.

Cynosorchis angustipetala, 123 ; auran-
tiaca, 123; fastigiata, 154; flexuosa,
124; glandulosa, 123; grandiflora,
124; lilacina, 123, 124; speciosa,
122; tenella, 124 ; uniflora, 122.

Cyperacese (Colombia), 138; (Diego
Garcia) 338; (Peru), 10. —

Cyperus cuspidatus, 532 ; debilissimus,
532; ligularis, 338; Luzule, 144;
(Pycreus) monocephalus, 531 ; platy-
caulis, 532 ; (Eucyperus) subsequalis,
531.

Cypripedium, 403, 404, 405, 407, 408,
410, 416; arietinum, 420; barbatum,
409, 411; candidum, 412, 418; ca-
datum, 419, 420; insigne, 41 A
Laurenceanum, 409, 410, 411, 412,
413, 417 ; Sedeni, 408, 409, 412, 414,
418; Spicerianum, 415; superbiens,
410, 411, 412; venustum, 409.

—, floral conformation of, 402 ; tera-
tology of, 407.

Cyrtandracex (Naga Hills), 135.

Cystopteris fragilis, 63.

Cytisus Laburnum, 110; vulgare
ling), 355.

(seed-

INDEX. 545

Dacrydium Foncki, 168.

Dalbergia, 441; Poolii, 466; scorpioides,
466.

Dalea Mutisii, 36; onobrychis, 36.

Daltonia patula, 310.

Danais Lyalli, 481; nummularifolia,
481.

Daphne, 358; (Jakpho), 136; Meze-
reum, 108.

Darwin, F., Relation between “ Bloom”
on Leaves and Distribution of Sto-
mata, 99.

Dasya cardicana, 340; crassipes, 340 ;
indica, 340 ; stirposa, 340.

Datisca glomerata='Tricerastes glome-
rata, 40.

Datiscee (Peru), 9, 13.

Daubeny’s exper. heat-rays, 84; light
on plants, 81.

Daucus torilioides, 41.

Davallia alpina, 224; parallela, 224;
pinnata, var. luzonica, 224; solida,
224.

Decostea iodinifolia, 164;
167.

Dehérain's researches on leaves, 82.

Delphinium, 348, 386, 393.

(seedlings) elatum, 356, 357, 375;
nudicaule, 356, 357, (cotyledon)
376; Staphisagria, 356, 357, 375,
(sect. seed) 394.

Dentaria (cotyledon), 398.

Deschampsia Mathewsii, 60.

Desfontainea (Chonos Archipel.), 140;
spinosa, 166, 168.

Desmodium adscendens, 141 ; gyrans,
111; lespedegioides, 464.

Deyeuxia, 441; coarctata, 59; emir-
nensis, 533; Mandoniana, var. 59;
Sp. n., 59.

Dianthus barbatus, 105; broad cotyle-
dons, 344, 351; caryophyllus (seed-
ling), 354, 364, (seed) 365; del-
toides, 105; Seguierii, 106; Sp.,
111.

Dichetanthera, 411, crassinodis, 476;
oblongifolia, 477.

Dichondra repens, 51.

Dichronema setacea, 144.

Dichrostachys unijuga, 467.

Dicksonia adiantoides, 223;
223; gomphophylla, 223;
tania) moluccana, 224;
224.

Dicoryphe, 441; Gualteriwfolia, 473 ;
laurifolia, 474; retusa, 473.
icotyledons, foss. Mull, 219, 220.
icranum dichotomum, 300; John-
stoni, 300; majus, 301 ; Scoparium,
301.

LINN. JOURN.—BOTANY, VOL

ruscifolia,

ampla,
(§ Pa-
scandens,

. XXII.

Didymocarpus,
508.

Didymochlena polycarpa, 225.

Didymodon papillatus, 301.

Diego Garcia, vascular plants of, 337;
vegetation (Hemsley), 332.

Dioscorea cryptantha, 528.

Diospyros, 131.

Diphylleia cymosa, 433, 434; Grayi,
433 ; (Japan), 433.

Diplachne aristata, 534.

Dirichletia involucrata, 482; ternifolia,
482; trichophlebia, 483; united with
Carphalea, 482, 483.

Disa emula, 69; affinis, 71; Bodkini,
74; Buchenaviana, 127; Cooperi,
71; cornuta, 69, 70; crassicornis,
70; inearnata, 127; lineata, 74;
macrantha, 70; Melaleuca, 75; me-
gaceras, 70; neglecta, 74; pygmea,
72; reticulata, 73; Scullyi, 70;
tabularis, 69; tenuicornis, 68.

Disperis Cooperi, 80; cucullata, 80 ;
Macowani, 77, 80; namaquensis, 80 ;
oxyglossa, 76, 80; paludosa, (ftnote)
77; purpurata, var. parviflora, 79;
secunda, 80; stenoplectron, 80; Ty-
soni, 79, 80; tripetaloidea, 127;
Wealii, 77,80 ; Woodii, 78, 80.

Dissanthelium supinum, 61. —

Distribution Indigenous species Nat.
Ord. Plants coll. by J. Ball in Up.
Valley Rimac, Peru, 9, 15.

Divided Cotyledons, 385.

Dobinza (Kohima), 130.

Dodonea viscosa, 400. .

Dombeya acerifolia, 419 ; Baroni, 450 ;
biumbellata, 450; floribunda, 449 ;
Humboldtii, 449; lucida, 449; in-
signis, 450; megaphylla, 449.

Dorstenia indica (African), 249.

Draba siliquosa, 5, 29.

Drimys (Chonos Archipel.), 140.

Druery, T., on Apospory in Polystichum
angulare, var. pulcherrimum, Wills.,
437. 2.

Drymaria cordata, 32; divaricata, 32 ;
glaberrima, 32; leptoclados, 32, var.
peruviana, 32 ; nitida, 81, 32.

Drymoglossum piloselloides, 230.

Dumasia, 135. s.

Dypsis, 441; concinna, 526; Curtisii,
526; forficifolia, 525 ; heterophylla,
525; nodifera— Phloga polystachya,
527; pinnatifrons, 525, 527 ; poly-
stachya, 525, 527; rhodotricha,
525.

Echeveria (Chicla), 5.
Echinostrobus Sternbergii, 201.
2x

135, 441; pusillus,

546

Eden Harbour (Patagonia), plants from,
167.

Elsocarpus dalechampioides, 452.

Elxodendron gyinnosporoides, 460.

Elasmium parviflorum, 296.

Elatostema hexadontum, 524.

Elettaria (seeds), 136.

Elynanthus laxus, 168.

Emarginate Cotyledons, 379.

Embelia, 135.

Embryo, size of, 393.

Encelia canescens, 160 ; canescens, var.
parvilolia, 151, var. tomentosa, 160 ;
farinosa, 151.160; oblongifolia, 151;
parvifolia, 151 ; (Peru) 3 ; tomentosa,
160.

Engler, views of plant-distrib. in S.
America, 144.

Epallage dissitifolia, 494.

Ephedia americana — E. rupestris, 56.

Ephedra (root-membrane), 389.

Ephippianthus, 245.

Epicrisis, 41.

Epidendrum clypeolum, 266 ; labiatum,
297 ; monophyllum, 297; nervosum,
262; triandrous, 403.

Epilobium (Jakpho), 136.

Epimedium alpinum, 431, 432, 433,
436, 437, var. rubrum, 4363 concin-
num,431; diphyllum, 430, 431, 432,
var. humile, 430; (Japan), 430; ja-
ponicum, 432; longifolium, 433;
macranthum, 430, 431, 432, 436, 437;
var. B. Thunbergianum, 431, 436,
var. a. typicum, 430, var. f. viola-
ceum, 430, 431, 436; musschianum,
431, 433 ; paniculatum, 433; pruni-
florum, 432 ; pumilum, 430 ; rubrum,
431, 432, 436 ; sagittatum, 432, 433 ;
sinense, 430, 432, 433.

Equisetum (fossil), 219.

Eragrostis tephrosanthos, 61.

Erigeron (Andes), 18; canadensis, var.
— E. crassicaulis, 44.

Eritrichium linifolium — Anchusa lini-
folia— A. oppositifolia — Antiphytum
linifolium, 51.

Erodium, 378 ; Cicutarium, 8, 33; ma-
lacoides, 8, 33.

Erpodium coronatum, 313; Han-
ningtoni, 313, 314; japonicum, 314 ;
Schweinfurthii, 313.

Erythrina cristata, 110.

Erythroxylum ampullaceum, 455 ; spar-
siflorum, 455.

Eschscholtzia, 386, 387, 398; (forked
cotyledons), 344, 357; californica
(seedling), 359.

Eubryum (African), 307.

Eucalyptus, 357, 358; globulus (coty-

INDEX.

ledon), 377, 383; sp. (seedling),
358.

Eucryphea (African), 311.

Eudicranum (African), 300.

Eugenia (Syzygium) aggregata, 475;
apiculata, 163 ; Luma, 163; oligantha
(Jossinia) 474.

Fuhabenaria, 134.

Euhedwigia (African), 310.

Eulejeunea, 326.

Euloppia lurida, 121; macra, 120.

Eumeteorium (African), 314.

Euonymus (seed), 363.

Eupatorium Ballii, 43; Gayanum, 43 ;
glechonophyllum, 164; glutinosum,
43; gynoxioides, 43; lavandule-
folium, 43; persicifolium, 43; sal-
viefolium, 43; sp., 43 ; Sternber-
gianum, 43.

Euphorbia (Africa), 299 ; adenopoda,
518; (Anthrothamnus) aleicornis,
517; (Goniostema) Bakeriana, 517,
518; Cyparissias, 110; erythroxy-
loides, 518 ; euonymoides, 518 ; Man-
cinella, 18; orthoclada, 517 ; pachy-
santha, 518; pilulifera, 338 ; serpens,
143. :

Euphorbiacese (Diego Garcia), 338;
(Naga Hills), 136; sp., 155.

Evodia floribunda, 457,

Exacums, 135.

Experiments in transpiration of plants
(Henslow), 87, 89.

Fabronia Schimperiana, 316. : =
Fagonia aspera, 158; chilensis, 158;
cretica, 158, 159. T.
Fagopyrum (cotyledon), 398.
Fagus antarctica, 168; betuloides, 168;
Dombeyi, 168; (nut) 396 ; sylvatica,
105, (seedling) 352. ;
Ferns from N. Borneo (Baker), ot
of Diego Garcia, 338; of Naga Hills,
130. E
Ferula (cotyledon), 349. :
Festuca cacuminis, 62; easapaltens®,
62; livida, 62; mollis, 62 ; gii
=F. muralis, 62; sp.?,62; su
formis, 62. y e
Ficus, 441; albidula, 523; been
105; botryoides, 523; oooi py
52l; cordata, 105; elastica, i de
pachyclada, 522 ; (Urostigma) Pi e i
rophlebia, 522 ; (Covellia) p 1

fera, 522; sakalavarum, 523; tr
choclada, 524. i
Filices (Andes, Peru), 63; (Diego

Garcia), 338 ; (Peru), 9, 63.
Filipedes, sect. of Crassula, 470.

INDEX. 547

Fimbriaria Boryana, 328.

Fimbristylis glomerata, 338;
morpha, 144, var. =F. laxa, 144.

First forms of leaves, 399.

Fissidens cellulosus, 319;
319; hyalinus, 319.

Flax (seed), 363.

Flora of Buenaventura in Colombia,
141; of Caldera. N. Chili, 158; of
Lota (Chili), 161; of Madagascar
(Baker), 441; of Naga Hills, 128,
131; of Payta, N. Peru, 148; of Pe-
ruvian Andes (Ball), 1.

Floral conformation of Cypripedium,
Dr. Masters on, 482.

Feeniculum vulgare (seedling), 350.

Forms of first leaves, 399 ; of leaves and
cotyledons ( Lubbock), 341 ; of ovules
(Lubbock), 345; of seedlings, and
causes to which they are due, Sir J.
Lubbock on, 341.

Fossil leaves from the Tsle of Mull
(Scotland) (Gardiner), 219.

Fotona, native name of Rhodolena
Bakeriana, 447.

Fox, Dr., Madagascar Orchids, 116.

Francoa sonchifolia, 163.

Franseria ambrosioides, 3.

Fraxinus excelsior, 105 ; (seed) 361.

Freshwater Hydrocharidee of Africa
(Ridley), 232.

Frullania angulata, 327; brunnea, 326;
cordata, 326; Ecklonii, 326 ; emer-
gens, 326; furcata, 241, 242, 243;
hamata, 243; replicata, 327; squar-
rosa, 326; trinervis, 326; usagara,
326, 329.

Fuirena umbellata, 144.

Fungi (Diego Garcia), 340.

oly-

dealbatus,

Galium, 357; Aparine, 8, 351, 384;
(cotyledon), 367, 368, 369; (seed-
ling), 384; boreale, 105; sacchara-
tum, 351, 354, 367, 368, 369.

Galvesia, 152; juncea, 154; limensis,
154, var. grandiflora, 151.

Gambelia, 152; speciosa, 154.

Gamopetale (Peru), 10, 16, 18.

Gamopoda densiflora, 443.

Garcinia, 441; cauliflora, 446; cernua,
446; orthoclada, 446; polyphlebia,
446.

Gardenia succosa, 483.

Gardner, J. Starkie, Fossil Leaves from
Isle of Mull (Scotland), 219.

Gastonia palmata, 105.

Gastroglottis, 245.

Gayac, a member of Leguminose, of
Seychelles, Diego Garcia, &c., 336.

Gentiana (Andes), 18.

cruciata, 105 ; multicaulis, var.,
50; prostrata, 50; sedifolia, var.
casapaltensis, 49; vaginalis, 50.

Geraniacese (Madagascar), 441; (Naga
Hills), 135; (Peru), 9, 13, 49; (Ri-
mac), 33.

Geranium (cotyledon unsymmetrical),
372; bohemicum (germinating seed-
ling), 377; Robertianum, 108; san-
guineum (seedling), 355; sessili-
florum, 33.

Gerbera, 14; emirnensis, 498.

Germination, opinions thereon, 349.

Geum urbanum, 105.

Gilia (Andes), 20; laciniata, 50.

Ginkgo, foss. Mull, 219.

Girondia manicata, 105.

Glaucium, sp., 110.

Gleditschia triacantha, 105.

Gleichenia (Chonos Archipel), 140;

cryptocarpa, 167; flagellaris, 222;
hirta, 222; pedalis, 167; vestita,

222.

Gnaphalium (Andes), 18; diffusum,
490; frigidum, 45; luteo-album, 8,
45; Mandonii, 45; melanosphe-
roides, 45; oxyphyllum, 45; pipto-
lepis (=Merope piptolepis), 45; spi-
catum, var. (=G. sphacelatuin), 45 ;
ulophyllum, 161.

Gnetacez (Peru), 9, 55, 56.

Godoya (S. Amer.), 146.

Goldfussia glomerata, 105.

Gomphocalyx, 485; herniarioides, 486.

Goodenoviex (Diego Garcia), 337.

Gossypium barbadense, 339.

Graminez (Diego Garcia) 338 (Peru),
35 Bye

Grammatophyllum Remplerianum,
122.

Gratiola officinalis, 105.

Gravesia, 441 ; porphyrovalvia, 477.

Grewia (foss. Mull), 219.

Grimmia leucophea, 303. qon

Griselinia, 164; alata, 163; ruscifolia,
167.

Groves, H. & J., descrip. of new Chara
from Madagascar, 537.

Guayaquil, Gulf of, climate and vegeta-
tion of, 138.

Guettarda speciosa, 334, 337.

Gymnogramme borneensis, 230; bor-
neensis, var. major, 230; lobbiana,
230 ; quinata, 230.

Gymnosporia cuneifolia, 460.

Gynopleura rugosa, 159.

Gynura sonchifolia, 495.

Gypsophila perfoliata, 111.

548
Gyrinopsis (ovule), 397.

Habenaria arachnoidis, 124 ; conopodes,
124; cirrhata, 125; Foxii, 124;
Hildebrandtii, 125; ichneumonifor-
mis, 125; tenerrima, 125; truncata,
125.

—, numbers compared with Liparis,
246.

Haberlandt, distrib. & function of len-
ticels on branches, 100.

Hemanthus (seed), 397.

Hakea, 357, 375, 390; acicularis (seed-
ling), 354.

Halenia, 151; gentianoides, 50.

Halimeda Opuntia, 340.

Hallia Bojeriana, 464.

Hampea, 145, 146.

Hannington, Rev. J., coll. Mosses and
Hepatice from Central Africa, 298.
Haplopappus hispidulus, 161; parvi-

folius, 161.

Hasseltia (N. Grenada), 146.

Hebecladus — biflora— Atropa
53.

Hedera Helix, 105.

Hedwigia ciliata, 310 ; Secunda, 310.

Hedychium coronarium, 134.

Hedysarum, 376.

Helleborus niger, 105.

Helianthoides, tribe of Composite,
495.

Helianthus annuus, 106 ; Cucumis (seed),
380.

Helichrysum amplexicaule, 491; ara-
neosum, 491; farinosum, 491 ; platy-
cephalum, 492 ; triplinerve, 492.

Helicodontium (African), 317.

Heliotropium parviflorum, 3,50; peru-
vianum, 4.

Helminthostachys zeylanica, 231.

Hemicampylus (African), 330.

Hemitelia alternans, 222.

Hemsley, W. Botting, Report on Vege-
tation of Diego Garcia, 332.

Henslow, Rev. G., Solar Spectrum on
Transpiration of Plants, 81; conclu-
sions drawn, 98; in Adiantum, Sp.,
91; in Asplenium, 91; in Aucuba
japonica, 91; in Box, 95; in Cactus,
96; in Coleus, 91; in Echeveria
metallica, 96; in Euonymus japo-
nicus, 91; in Euphorbia, 92 Eun
Fern, 96 ; in Grass, 92 ; in Lettuce,
92, 93, 94; in Radish, 92 ; in Palms,
95; in Portugal Laurel, 90 ; in Pelar-
gonium, 91; in Wallflower, 92.

Hepaticee & Mosses from Central Africa
(Mitten), 298.

Herberta dicrana, 322.

biflora,

INDEX.

Heritiera macrophylla, 373.

Hernandia peltata, 338 ; ovigera, 338.

Herpestis Monniera, 337.

Herpetium deflexa, 322.

Hesperis matronalis (cotyledon), 370;
(sect. seed) 374, 375.

Hexandrous Flowers, 414.

Hibiscus, 441; cytisifolius, 447 ; leaves
entire, 341; nummularifolius, 448 ;
oblatus, 448; tiliaceus, 336; xipho-
cuspis, 447.

Hieracium chiclense, 47 ; frigidum, 48;
jubatum, 48; trichodonta, 48. E

Hierochloa arenaria, 168; altissima,
168 ; antarctica, 168 ; antarctica, var.,
168; magellanica, 168; redolens,
168; teretifolia, 168; utriculata,
168.

Hilsenbergia Baroni, 450.

Hippophaé, 375, 390; (embryo) 349;
rhamnoides (seedling), 352, 363.

Hoffmanseggia viscosa, 150.

Hohnfeldt’s observations stomata on
leaves, 100.

Holmskoldia, 136.

Holothrix glaberrima, 116, 125. :

Homalia elongata, 315; lingueefolia,
316; pulchella, sp. of Metzgeria on,
243.

Hooker, Sir J. D., Clarke’s letter to, on
Botany of Naga Hills, 128. :

Hookeria angolensis, 309 ; Langsdorfii,
309; macropus, 309; Pappeana,
309; Utacamundiana, 309; versi-
color, 309.

Hose, Dr., coll. Ferns from N. Borneo,
222.

Hottonia, subaqueous leaves, 343.

Huidobria chilensis, 159.

Humulus Lupulus, 108.

Hydnora (Kilimanjaro), 299. -folia

Hydrangea hortensis, 105; quercitosa,
103, 105, 109.

Hydrilla rubella, 232; verticillata, 233,
236.

Hydrillee (African), 233. —

iere Nass of Africa (Ridley), 232.

Hydrocharis, 99. 9. 13

Hydrophyllacee (Andes, Peru), 9, 19;
20, 50. BE

Hymenophyllum (Chonos Archipe agi ids
140; dilatatum, 222; javanicum, 228;
Neesii, 222. a

Hyophila involuta, 304; javanıca, ;
plicata, dn bien.

Hypericum calycinum, 110. :

Ho ded Archeri, 313; arcus
tum, 313.

Hypnum andongense, 316 ; MU
317 ; conostomum, 317 ; implica

INDEX.

317; salicicola, 317; Sprucei, 317;
subcompressum, 317 ; toxarion, 317;
usagarum, 317.

Hypocherris apargioides, 165 ; elata, 48;
Lechleri, 165 ; radicata, 48; sessili-
flora, 48; sp.?, 48; taraxacoides,—
Oreophila taraxacoides— Achyropho-
rus taraxacoides, 48.

Hypoestes acuminata, 512, 513 ; chloro-
clada, 513; congestiflora, 511; mi-
crophylla, 511; obtusifolia, 512;
phyllostachya, 511 ; sessifolia, 512.

Hypopterygium laricinum, 308.

aries 105, 134; japonica,

23.

Imerina (Madagascar), Orchids of, 116.

Impatiens, 357; balsamina (seedling),
352; chinensis, 130; (Jakpho), 133;
emirnensis, 454; Hildebrandtii, 454;
parviflorum, 106; parviflora (sect.
seed), 379; sultani, 299.

Incomplete (Madagascar), 514.
Indigofera desmodioides, 463; ormo-
carpoides, 464 ; tinctoria, 463, 464.
Inuloidee, 14; (tribe Composite),

493.
Ipomæa (Strophipomeea) biloba, 337;
dasysperma, 357, 382, 383 ; grandi-

flora, 337; purpurea (embryo),
382; Pes-capree, cotyledons of, 348 ;
rubro-viridis, 507; (Orthipomoea)

syringzefolia, 507.
p develop. outer row stamens,
Tridez (Peru), 9, 56.
Isatis tinctoria, 106.
Isochilus, triandrous & five anthers, 403.
Isolepis pygmza, 167.
Isotachis Auberti, 322.
Ixora pachyphylla, 484.

Jægeria hirta, var., 45.

Jakpho mountain, Botany of, 129, 132;
as compared with Tonglo veget., 132.

Jasmine (ovule), 397.

Johnston, H. H., Mosses and Hepatice
from Kilimanjaro, 298.

Juglans regia, 105.

Juncaces (Peru), 9, 57.

Juncus, 132, 133.

Jungermannia (African), Auberti, 322;
convexa, 322; cupressina, 322; fim-
briata, 321; hirtella, 321; minuta,
321; serrulata, 322; tripartita, 322.

Kalanchoe brachycalyx, 470; brevi-
caulis, 470; gomphophylla, 471;
integrifolia, 471; (Kohima), 130 ;
laxiflora, 472 ; multiceps, 470; pube-

919

scens, 470; pumila, 470 ; subpeltata,
472; (Kitchingia) sulphurea, 471;
streptantha, 472.

Kalmia latifolia, 105.

Keteleeria, 199 ; Fortunei, 197.

Kissenia (Peru), 13.

Kitchingia, 441, 471, 472; laxifolia,
472; minuta, 473; streptantha, 472;
sulphurea, 471; subpeltata, 472.

Kleinia Hildebrandtii, 498.

Koenigia (Jakpho), 136.

Kohima, vegetation of, 129.

Küdalu déhi, or Leech Lime of Ceylon,
213.

Kurz, Limes of Burma, 217.

Kyllinga monocephala, 338.

Labiate (Peru), 9, 54.

Laburnum, 357, (seedling) 355; vul-
gare (seed), 373.

Lactuca Scariola, 100.

Lagarosiphon cordofanus, 234; densus,
235; madagascariensis, 235; mus-
coides, 233, var. major, 233; Nyas-
se, 234; Roxburghii, 232, 233, 236;
rubella, 234; Schweinfurthii, 234;
Steudneri, 234, 235.

Lanary, native name of Ochna macran-
tha, 458.

Landemy, native name of Anthocleista
amplexicaulis, 506.

Lapageria rosea, 162, 166.

Larix conifera, 208; davurica, 211;
europea, 211; Kempferi, 208, 210.

Larkspur, seed of, 348. i

Lasiopetalum ferrugineum (seedling),
400.

Lastrea apiciflora, var. non-apiciflora,
130; patentissima, 130.

Lathyrus pratensis, 109;
111.

Laurels (Jakpho), 134. :

Laurineæ, (Chiloe) 140, 338; (Diego
Garcia), 338.

Lavatera, 378.

Leaves, first forms of, 399; form deter-
mined by distance between buds,
342; forms of, and cotyledons (Lub-
bock), 341, 399; Grant Allen’s views,
division subsequeous leaves from pau-
city carbonic acid in water, 343;
shape of, 341; venation, 341.

with bloom both surfaces (Dar-
win), 110; with bloom, lower surface
only, 108 ; with bloom, upper surface
only, 109; with no bloom on either
surface, 104.

Lecontea farinosa, 486.

Leea cuspidifera, 461.

Leech Lime, 213.

sylvestris,

550

Leguminose (Diego Garcia), 337; Ko-
hima), 135 ; (Peru), 9, 10, 35.

Leioscyphus infuscatus, 321; Motleyi,
321; repens, 321.

Lejeunea acuta, 326; adhesiva, 325,
329 ; atra, 324, 329; cornuta, 325;
deplanata, 325, 329; digitalis, 325,

329; Hanningtoni, 324, 329; flava, |

326; producta, 325; serrula, 324,
329; sinuata, 324, 329; subfusca,
324; superba, 325; tortuosa, 324.

Lemon Swangi, 216.

Leontice triphylla, 435.

Lepidium affine, 29; graminifolium,
389; ruderale, 106, 357; sativum,
345, (divided cotyledons) 344,
(sect. seed) 389, 392, 397 ; virginia-
num, 8, 29.

Lepidopilum Hanningtoni, 309, 310;
Lastii, 309.

Lepidothamnion, 168.

Lepidozia cupressina, 322.

Leptocolea, 325.

Leptodesmia congesta, 464.

Leptodon Smith, 314, 323.

Leptodontium radicosum, 301; squar-
rosum, 301; stellatum, 301; tri-
stichus, 301; Leschenaultii, 385;
ignea, 385.

Leskea claviramea, 318; rubicaulis, 301.

Lespedeza, 130.

Leucopogon, (ftnote) 220.

Levisticum officinale, 106.

Liabum ovatum — ( Paranephelium) uni-
florum, 46.

Libocedrus tetragona, 168.

Light important in plant-growth, 81.

Ligularia racemosa, 133.

Liliacese (Peru), 9, 56.

Lilium speciosum, 105.

Lima, 213, 214, 216, 218 ; leaf, Ceylon,
typical form of, 215; leaf, occasional
form of, 215.

Lime, form of Indian, cultivated, 215.

Limes, Dr. Brandis on descent of, 213 ;
wild source of cultivated sp. (Bona-
via), 213.

Limo agrestris, 216, 218; Ferus, 216,
218 ; tuberosus, 216, 217, 218; tube-
rosus martinicus, 214, 216.

Limoen Swangi, 216,

Limonellus aurarius, 216, 218.

Linaria (caulicle), 395; vulgaris, 110.

Lindeniana, 265; lilifolia, 269.

Lindsaya cultrata, 224; cultrata, var.
Lobbiana, 224; divergens, 224; la-
nuginosa, 224; scandens, 224; tra-
peziformis, 224.

Lindstróm on bloom-protected surface
of Trifolium repens, (ftnote) 99.

INDEX.

Linum aquilinum, 162; flavum, 111;
Macreei, 162 ; monogynum (seedling),
351, 353.

Liparis, anther-cap, 249; bracts, 248;
capsule, 249 ; column, 249; diag.
charac. gen., 252; distribution ol,
246; divis. into sections, 251; ex-
cluded species, 297; fertilization,
250; flower of, 248; gen. founded
by Richard, 244; habit of species,
245; insect attractions, 250; key
diag. spec. belonging to, 253-257 ;
leaves, 247; literature concerning,
244; nomina nuda, 297; relations of,
245 ; scape, 248 ; species dubie, 296;
stem, rhizome, and pseudobulb, 246,
247; stigma, 249; the Mollifolise,
251, 253; the Coriifolie, 252, 257.

, Monograph of Genus (Ridley),

244.

abyssinica, 254, 266; acutissima,

254, 264; affinis, 297; alata, 282,

297 ; anceps, 297 ; angustifolia, 246,

290; arnoglossophylla, 254, 268 ;

atropurpurea, 247, 256, 277 ; auricu-

lata, 254, 267, 294; aurita, 245, 249,

258, 291; barbata, 256, 276; Bec-

carii, 257, 284; Beddomei, 255, 268;

Bernaysii, 297; bicornis, 253, 260;

bidentata, 297; biloba, 256, 276;

bistriata, 258, 288, 289 ; bituber-

culata, 253, 262, 263, 264; bituber-
culata, var. formosana, 263; boota-

nensis, 258, 285; Boottii, 293;

Bowkeri, 250, 255, 270; brachy-

glottis, 256, 275 ; brachystalix,

247, 257, 281; bracteosa, 297;

campestris, 254, 266 ; campylostalix,

255, 273; capensis, 255, 272; cS"

spitosa, 249, 258, 290, 297; chlo-

roxantha, 257, 284; clavigera, 259,

294; clypeolum, 254, 266; cology-

noides, 258, 286 ; compressa, 249, 298,

291; connata, 255, 270; correana,

272 ; crenulata, 120, 257, 283 ; erispt-

folia, 257, 280; Cumingii, 249, 298,

292 ; cuneilabris, 287; cuspidata, 294,

266 ; cylindrostachya, 297; decurrens,

249, 258, 291 ; decursiva, 287 ; den-

drochilum, 259, 295 ; densiflora, 291;

diodon, 257, 285; disepala, 248, 254,

265; disticha, 249, 250, 251, 258,

292; Eggersi 256, 271; elata, 248,

249, 253, 259, 261, 262, 265; elata,

var. rufina, 260 ; elata, var. inundata,

260; elata, var. latifolia, 260; ele-

gans, 293, 294; elegantissima, a

elliptica, 249, 250, 254, 268, pied ,

289 ; ferruginea, 254, 264; emm

249, 259, 296 ; flavescens, 120, 245,

INDEX.

257, 271; foliosa, 287; Forbesii,
249, 257, 283; formosana, 263;
galeotiana, 265 ; Gerrardi, 255, 275 ;
glossula, 254, 268; grandiflora, 248,
257, 282; gregaria, 292; Griffithii, 258,
285 ; grossa, 258, 288; guineensis, 253,
260,261; Guingangee, 256, 274; habe-
naria, 260; habenarina, 253 ; hawai-
ensis, 254, 267 ; Hookeri, 258, 288;
intermedia, 297; jamaicensis, 265 ;
japonica, 256, 273; Jovis-Pluvii, 256,
276; Kappleri, 253, 261; Krameri,
255," 269; labiata, 297; lacerata,
257, 284; laciniata, 250; latifolia,
257, 281, 283; liliifolia, 244, 255,
268, 269; Lindeniana, 265; livida,
277, 278; Loeselii, 246, 249, 255,
267, 269, 271, 272, 273; longa, 254,
255; longicaulis, 120, 255, 270;
longipes, 247, 248, 249, 250, 251,
259, 293, 294; longipetala, 255, 275 ;
lutea, 256, 274 ; luteola, 246,258, 289;
Mannii, 258, 286; minima, 297;
monophylla, 297 ; montana, 256, 277 ;
mucronata, 292; nepalensis, 246,
256, 278, 280; mervosa, 253, 262;
neuroglossa, 249, 250, 257, 280;
obcordata, 297 ; ochracea, 255, 270;
odontostoma, 253, 263 ; odorata, 262,
264; olivacea, 254, 262, 263, 264,
277 ; orbicularis, 286 ; ornithorrhyn-
chos, 249, 255, 270, 271 ; pallida, 296;
paradoxa, 253, 260, 261, 262, 263,
264; parva, 256, 275; pachypoda,
288, 289 ; pachypus, 258, 288; par-
viflora, 259, 296; pectinata, 256,
277; pendula, 293 ; plantaginea, 258,
286; platyphylla, 254, 264; pli-
cata, 257, 285, 286; polycardia,
255, 271; priochilus,297 ; prionotes,
297; puncticulata, 119, 247, 255,
270, 272 ; purpurascens, 256, 279;
pusilla, 259, 294, 295; ramosa,
257, 281; reflexa, 246, 250, 258,
270, 281, 284, 287; reflexa,
var. $. cuneilabris, 287; repens,
247, 256, 279; resupinata, 245,
258, 290; revoluta, 258, 289; Rhee-
di, 251, 253, 261; rostrata, 268;
rupestris, 254, 255, 268; rupestris,
var. purpurasceus, 268; Saundersi-
ana, 248, 250, 256, 274; selligera,
286; serreformis, 297; spathu-
lata, 247, 294; stachyurus, 259,
295; stenoglossa, 256, 279; Strick-
landiniana, 259, 295; Tradescantifolia,
282; tricallosa, 255, 275; triloba,
259, 295; tristis, 297; vera, 251;
vestita, 259, 294; viridiflora, 248,
258, 289 ; viridi-purpurea, 254, 267 ;

551

Wageneri, 257, 281; Walkeris, 256,
277; Welwitschii, 247, 249, 257,
281; Wendlandii, 256, 274; Wighti-
ana, 256, 278; Wightii, 289; xan-
thina, 256, 275 ; zeylanica, 297.

Lippia nodiflora, 338; reptans, 154.

List of Plants coll. in Upper Valley,
Rimac, Peruvian Andes (J. Ball), 23.

List of Vascular Plants, Diego Garcia
(Hemsley), 337.

Listera ovata, 250.

Liteas (Jakpho), 134.

Lithospermum (cotyledon), 385.

Loasa carduifolia, = Blumenbachia car-
duifolia, 39; Chilensis, 159; grandi-
flora, 39; xanthiifolia, 39

Loasaceæ, 9, 13, (Andes) 20, 26,
(Peru) 39.

Lobed cotyledons, 378.

Lobelia Erinus, 101.

Lomaria tenera, 165; glauca, 130, 132 ;
pycnophylla, 132.

Lonchocarpus Paullinioides, 466.

Lophocolea bidentata, 321.

Lopholejeunea, 324.

Loranthaces (seeds), 391.

Loranthus Eschscholzianus, 166.

Lota (Chili), flora of, 161.

Lubbock, Sir J., Phytobiological Ob-
servations, 341.

Luculia, 135. t

Lugonia andina, 49; lysimachioides,
49.

Lupinus microphyllus, var. argyreas, 35;
paniculatus, 4; paniculatus, var. ar-
genteus, 35.

Luxembergis (Equator. Amer.), 146.

Luzula Alopecurus, 57; Chilensis, var.,
57; =L. humilis, 57.

(Jakpho), 133; racemosa, 57,
var. humilis, 57 ; spicata, 57.

Luzuriaga, 166. eae

Lychnis apetala, 30, 31; chilensis (=
L. terminalis =L. nubigena) 31;
chilensis, 30, 31, var. Peruviana,
30; magellanica, 31; thysanodes
(—Silene thysanodes—S. andicola),
30; Viscaria, 106.

Lycopersicum Peruvianum, 52.

Lycopodiacese (Diego Garcia), 338.

Lycopodium cernuum, 144; scandens,
231; xiphophyllum, 539 o

Lysionotus Griffithii, 135; ternifolia,
135.

Macaranga, 441; ferruginea, 521; (Eu-
macaranga) racemosa, 520.

Maelura aurantiaca, 105.

Macrocoma (African), 305.

Macromitrium cespitans, 305; confu-

552

sum, 305; Dawsoniomitrium, 305;
Dregei, 305; Lycopodioides, 305;
tenue, var brachypus, 305.

Madagascar, Flora of (Baker), 441;
Orchids of (Ridley), 116.

Mesa, 135.

Magellanie province, flora of (Ball),
140.

Mahonia Aquifolium, 105 ; japonica,
428.

Maianthemum bifolium, 105.

Malaxis affinis, 297; atropurpurea,
279; cæspitosa, 290; compressa,
291; cordifolia, 278; Correana, 272 ;
crenulata, 283; disticha, 292; fla-
vescens, 271 ; Galeottiana, 265 ; lanci-
folia, 261; Loeselii, 244, 274 ; longi-
folia, 272; minima, 297; montana,
277; mucronata, 292; odorata, 262 ;
pallida, 296; parviflora, 296; pur-
purascens, 279 ; Rheedii, 261;
Tradescantifolia, 282 ; viridiflora,
289.

Malesherbia humilis, 159 ; (Peru) 13;
rugosa, 159 ; thyrsiflora, 39.

Mallostoma thymifolia = Hedyotis
thymifolia, 42.

Mallow (embryo of), 378.

Malva, 378; moschata (seedling), 357,
358.

Malvaces, (Diego Garcia) 337; (Peru)
9; (Rimae, Peru) 32.

Malvastrum capitatum, 33; limense
(=Malva limense), 32 ; peruvianum,
33 ; sp. ?, 33.

Mammillaria, 161 ; Mandonii, 51; Wal-
persii, 51.

Manettia asperula, 142.

Maquenne, on evaporation and transpi-
ration, (ftnote) 84.

Marsilea, 111.

Mascarenhaisia, 441 ; Gerrardiana, 504 ;
macrosiphon, 504.

Masters, Dr. M. T., Contributions to
Lad of Species of Conifers,
169.

——, Floral Conformation of Genus
Cypripedium, 402.
Matisia (S. Amer.), 145, 446.
Maurandella, 153.
Maurandia juncea, 152, 153, 154.
Medicago denticulata, 8, 35 ; lupulina,
11l.
eh divaricata, 478; linearifolia,
8.
Melaleuca, (ftnote) 220.
Melastomacese, 475, 476, (Andes) 27;
FA antiquity of, 148 ; (Madagascar)
l.

Melica laxiflora, 60 ; sp.?, 60.

INDEX.

Melilotus indica, 8, (= parviflora) 35;
officinalis, 111.

Melittis melissophyllum,
353, (cotyledon) 393.
Melochia betsiliensis, 451.
Meniscium reticulatum, 144; Hosei,
230; Thwaitesii, 230; triphyllum,

230.

Menispermaces (Madagascar), 141.

Menispermum canadense (seedling),
353, 362. :

Mentzelia hispida, 39; (S. America),
20.

Menyanthes trifoliata, 106.

Mercurialis (cotyledon), 393, 398;
perennis, 105, 358. :
Meteorium afrieanum, 314; imbrica-

tum, 314.

Metzgeria, n. sp. of (Mitten), 241 ; my-
riapoda, 327; nitida, 243; saccata,
241, 242; scobina, 243.

Michelia (Jakpho), 133.

Miconia decussata, 142 ; reducens, 141 ;
symplocoidea, 142.

Microglossa (Jakpho), 133.

Microloma (seedling), 357, 375.

Microstylis, 244 ; Bernaysii, 297 ; Com-
melynifolia, 282 ; japoniea, 273; ver-
sicolor, 297.

Microthamnium frondosum, 316; fru-
ticellum, 316; pseudoreptans, 316.
Mimosa myriocephala, 467; pudica,

110.

Mimoses: (Peru), 151.

Mimulus glabratus, 53; luteus (seed-
ling), 395.

Mimulopsis affinis, 509.

Mistletoe seed, 391.

Mitraria, Chonos Archipel., 140 ; coc-
cinea, 168.

Mitten, W., Mosses and Hepaticæ from
Central Africa, 298.

New Species of Genus Metz-
geria, 241.

Mnium rostratum, 307.

Modica hederefolia, 479. : :

Mollifoliz, sect. gen. Liparis, diagnosis
of, 253, 259.

Monadenia micrantha, 73; rufescens,

(seedling)

Monnina crassifolia, 30; obtusifolia,
30; pubescens, 30; rotundifolia,
335.

Monochilus gymnochiloides, 127.
Monocotyledons, foss. Mull, 219; (Ma-
dagasear), 525. k
Nom genus Liparis (H. N. Rid-

ley), 244.
Monopetale (of Madagascar), 480.
Morinda citrifolia, 337.

INDEX.

Morus alba, 105.

Moscharia pinnatifida (seed), 364.

Mosses (Diego Garcia), 339.

and Hepatice from Central Africa
(Mitten), 298.

Mueuna (Nambre Forest), 135; (8 Stizo-
lobium) axillaris, 465.

Muehlenbeckia rupestris, var. vulcanica,
55=M. vuleana, 55; vulcanica, 55 ;
rupestris, 55.

Muehlenbergia, 15 ; adspersa, 58; clo-
mena, 58; sp.?,58.

Mull, fossil leaves fróm (J. S. Gardner),
219.

Mundulea laxiflora, 464.

Musci (Diego Garcia), 339.

Mustard, seedling of, 345.

Mutisia, 14; lilicifolia, 165 ; vicieefolia,47.

Mutisiaces, affinities, 14, 16; (S. Ame-
rica) 21, 26.

, diff. lines descent of, Bentham and
Ball’s views, 14.

Myagrum (ovary), 397.

Myrioneuron, 135.

Myriophyllum, subaqueous leaves, 343.

Myrrhis andicola, 41.

Myrsine, 135.

pope (Chiloe), 140; (Diego Garcia),
337.

Mystacidium viride, 122.

Naga Hills, Botany of (Clarke), 128.
Nandina (Japan), 428; domestica, 429.
Nassauvia, 14.

Nassella ceespitosa = Oryzopsis cespi-
tosa, 58.

Nasturtium, 144, 441.

Negundo fraxinifolia, 105.

Nelumbium, gives out bubbles of air,
99; speciosum, 111.

Nephrodium aciculatum, 226; acuta,
228; calcaratum, var. falcilobum,
225; crassifolium, 225; ferox, 227 ;
immersum, 225; intermedium, 226;
Leuzeanum, 227 ; megaphyllum, 227 ;
multisetum, 226; nudum, 228 ; pro-
currens, 227; pulvinuliferum, 226;
sarawakense, 225; scabrosum, 226;
setigerum, 226, 227; stenophyllum,
227 ; ternatum, 227 ; truncatum, 227;
villosum, 227 ; unitum, 338.

Nepsera aquatica, 141.

Nettle, seed of, 346.

New instance of Apospory in Poly-
stichum angulare (Druery), 438.

Nicandra physaloides, 53.

Nicotiana (Chicla), 5 ;
glutinosa, 53.

Nitella Baroni, 537; pseudoflabeVata,
537.

glauca, 111;

553

Notes on Botany of Western South
America, by John Ball, 137.

Nothoscordum striatum, 167.

Notocheete (Kohima), 130.

Notochleena squamosa, 64.

Nuphar, 358.

Nuxiaspheerocephala, 505 ; pachyphylla,
505, 506; terminalioides, 505.

Nyctaginez (Andes, Peru), 9, 13, 54;
(Diego Garcia), 338.

Nymphea, 29, 358.

Oaks of Naga Hills, 130.

Observations, Phytobiological (Lub-
bock), 341.

Ochna macrantha, 457.

Ochnacee (S. Amer.), 146.

Ochrosia borbonica, 337, 339.

Ocotea (Mespilodaphne) trichantha,
515.

Odontoglossum crispum, 403.

Odontolejeunea, 324. d

CEnothera, 378, 395, 397; epilobiifolia,
39; (cotyledons) 349; Lindleyana,
384; multicaulis, var. petiolaris, 38;
prostrata, 39; psychrophila, 38;
rosea, 39.

Olea cuspidata (seedling), 354 ; europea,
105.

Oleandra neriifolia, 228 ; neriifolia, var.
phyllarthron, 228.

Oligomery (Orchids), 408.

Onagrariez, cotyledons of, 349; (Peru),
9,38.

Oncophorus cyathicarpus, 302.

Oncostemum, 441; botryoides, 502;
flexuosum, 502; microspherum, 501;
polytrichum, 501 ; vacciniifolium, 501,
502.

Onoclea sensibilis, 219.

Ophelia, 135 ; (Jakpho), 133.

Ophioglossum reticulatum, 231.

Ophriosporus triangularis, 161. »

Ophrys liliifolia, 289, 272 ; Loeselii,
272; nervosa, 262; paludosa, 272;
rotundifolia, 268 ; trifolia, 269.

Opuntia (cotyledon) basilaris, 389 ; La-
bourtiana, 388. f

Orchid flower, changes resulting from
hybridization, 421; Combination of
tendencies, 421 ; Conformation flower
Cypripedium, 402, 404 ; Displace-
ment, 416; Change in position of
parts, 416: Gen. Plan of parts, 402—
in Odontoglossum crispum, 402; in
Arundina pentandra, 402; in Zygo-
states cornuta, 403 ; in Thelycluton,
403; in Pleione birmannica, 403 ;
in Selenipedium, 404 ;—Distrib. of
vascular bundles, 405; Hexandrous

LINN. JOURN.— BOTANY, VOL. XXII. 2y

554

flowers, 414; Hybrid plants, changes
in, 422; Increase inner row of sta-
mens, 445: Oligomery, 408; reduc-
tion parts 408, in Cypripedium
Sedeni, 408, in C. barbatum, 409,
in C. venustum, 409, in C. Lawren-
ceanum, 409 ; Partition, 415 : Peloria,
or regularity in arrang. parts, 417;
regular peloria, 418; irregular pe-
loria, 420: Pleiomery, 410, in-
creased number of parts, 410, in Cypri-
pedium superbiens, 411; Teratology
of Cypripedium, 407; Tetrandrous
flowers, 412; "Triandrous flowers,
411.

Orchides of S. Africa (Bolus), 65.

Orchids from Madagascar (H. N. Rid-
ley), 116; (Jakpho), 134.

Orchis bicornis, 66.

Oreomyrrhis, 15, 41; andicola, 41.

Oreophila taraxacoides = Hypocheris
taraxacoides, 48.

Orobus vernus, 106.

Oroxylum indicum (seed), 381.

Orthotheca (African), 303.

Oryzopsis ceespitosa, var. peruviana, =
Nassella ceespitosa, 58.

Osbeckies, tribe of Melastomacesz (Ma-
dagascar), 475.

Ottelia alismoides, 233, 237; brasili-
ensis, 233; lancifolia, 238; lanci-
folia, var. fluitans, 238; plantaginea,
238, 239; scabra, 239; ulvefolia,
238, 239; vesiculata, 237.

Ourisia chameedrifolia, 53 ; sp. ?, 53.

Ouvun, or Ou Van, or Anivona, native
name of Phloga polystachya, 527.

Ovules, forms of (Lubbock), 345.

Oxalis albicans, 163; filiformis, 34;
laureola, 343; laxa, 163 ; pubescens,
34, var. glabra, 34 ; sp. ?, 34, 110.

Oxysporeee, tribe Melastomacese ( Mada-
gascar), 176.

Pachira (seedling), 355.

Pachypodium brevicaule, 503; densi-
florum, 503.

Preonia, sp., 105.

Paliurus, fruit, 397.

Paludella, 306.

oe isthmus, climate and veget.,
137.

peg (Spheropanax) gomphophylla,

Pandanus dyckioides, 527; montanus,

: 527 (=Sussea conoidea, 527).
anicum polygonatum, 144; sangui-
nale, 338. de

Papaver Rhoas, 106.

Papillaria (African), 314.

INDEX.

Paranephelium ovatum =Liabum ova-
tum, 46 ; uniflorum, 46.

Parietaria debilis, 56.

Parinarium emirnense, 469.

Parmelia perforata ( African), 299.

Partition in Orchids, 415.

Paspalum conjugatum, 144; stoloni-
ferum, 8; stoloniferum = P. racemo-
sum, 57.

Passerina hirsuta, 101.

Passiflora, lobed leaves, 341 ; seed, 364 ;
suberosa, 339.

Passifloraceee (Andes, Peru) 9, 13,
39.

Patagonia, plants from, 167.

Patrinia (Kohima), 130. :

Paxtonia, aetinomorphie condition of,
419.

Payta, N. Peru, flora of, 148; climate
and physical geography of, 149.

Pedicularis (Jakpho), 133.

Pedilanthus, 442; lycioides, 516.

Pelargium madagascariense, 454. —

Pelargonium, 441; australe, seedling,
357; glaucum, 110; madagascariense,
454.

Pellza pulchella, 64.

Pelliciera (S. Amer.), 146.

Peloria (Orchids), 417. i

Pentapetes, 357; phoenicea (seedling),
359.

Pentas hirtiflora, 482.

Pentstemon barbatus, 106.

Perezia ciliaris — Clarionea ciliaris, 47 ;
multiflora, var., 47. à
Pernettya empetrifolia, 167; pumila,

168.

Petioles, 375.

Petiveria octandra (cotyledon), 371,
379.

Phacelia (Andes) circinata, 50. e

Phalenopsis, actinomorphie condition
of, 419.

Pharus latifolius, 108.

Phaseolus multiflorus (seed), 106, 363 ;
vulgaris (cotyledon), 393.

Philadelphus coronarius, 105.

Philippia capitata, 499; cryptoclada,
499; hispida, 499; minutifolia, 500 ;
trichoclada, 500.

Phillyrea (ovule), 397. : :

Phloga polystachya= Dypsis nodifera,
527

Phlomis tuberosa (connate cotyledon),
377.

Phlox (Andes), 20.

Phragmicoma emergens, 323; florea,
323; Palisoti, 324; Pappeana, 323.

Phylesia buxifolia, 168.

Phyllanthus Niruri, 338.

INDEX,

Physalis Peruviana — FP. edulis, 53.

Phytobiological Observations (Sir J.
Lubbock), 341.

Phytolacca octandra, 55.

Phytolaecacee (Andes, Peru), 9, 12,
55.

Picea amabilis, 171, 184, 192, var. mag-
nifica, 192, var. robusta, 192; bifolia,
183; concolor, 177, var. violacea, 177 ;
Fortunei, 197; grandis, 174, 181;
hirtella, 195; lasiocarpa, 174, 175;
Lowiana, 175; magnifica, 192; no-
bilis, 188; Omorika, 203, 204, 205,
212; orientalis, 203, 205; Parsons-
iana, 175 ; religiosa, 194.

Pilea macropoda, 524.

Pilosella, 48.

Pilosperma (Andes), 146.

Pimpinella (Jakpho), 133.

Pines of Naga Hills, 130.

Pinus amabilis, 171, 184, 191, 192; Cem-
bra, var. fruticosa, 205; concolor,
177; excelsa, 205, 207, 208; For-
tunei, 197 ; grandis, 171, 174, 181;
hirtella, 195; Kempferi, 208; Kha-
siana, 130 ; lasiocarpa, 184; Lowiana,
175; nobilis, 188; Peuke, 205, 206,
207, 208; religiosa, 195; rigida
(seedling), 389.

Piper, 441; (Coccobryon) emirnense,
514; magnolizfolium, 105.

Piptadenia leptoclada, 467.

Pisonia inermis, 335, 338 ; macrophylla,
335 ; viscosa, 335.

Pithecoctenium Aubletii
381.

Pittosporum pachyphyllum, 444; To-
bira, 105; vernicosum, 445.

Plagiochila abyssinica, 321 ; asplenioides,
319; Barteri, 320; cristata, 320;
dichotoma, 319 ; dicrana, 320; java-
nica, 320; Lastii, 320; Neckeroidea,
320 ; sarmentosa, 319; sinuosa, 319;
squamulosa, 319 ; tenebrans, 319.

Pen glauca, 132; pyenophylla,

(cotyledon),

Plagiothecium membranosulum, 316;
nitidifolium, 316.

Plantagines (Andes, Peru), 9, 54.

Plantago Hartwegii, 54; lanceolata,
106; media, 106; sericea, var.,— P.
leucophylla, 54.

Plants, Cruickshanks on, W. coast of S.
America, 157.
indigenous

(Hemsley), 337.
——— of Rimac Valley (Ball), 28.
———, Swiss Alps and Peruvian Andes
compared (Ball), 12.
Platanus, cotyledon of, 350, 351 ; fruit

to Diego Garcia

555

sect., 361; foss. Mull, 219, 220 ; oc-
cidentalis, 105.

Platycoryne Pervillei, 127.

Platystylis, 249, 258; alata, 245, 291 ;
atropurpurea, 245, 279; decurrens,
245, 291.

Plectronia micrantha, 483.

Pleiomery (Orchids), 410.

Pleuridium tenuissimum, 300.

Pluchea Chingoyo, 156.

Plumbago Larpente, 106.

Poa annua, 61,and Poa, vars. (Nos. 1, 2,
8, 4, 5), 61, 62.

Podocarpus, foss. Mull, 219 ; nubigena,
168.

Podophyllum (Japan), 434 ; japonicum,
434; peltatum, 434, 435.

Poinsettia, sp., 110.

Polemoniaces (Andes, Peru), 9, 13, 20,
50.

Pollia (Kohima), 129.

Polyachrus echinopsoides, 47.

Polygala leptocaulis, 445.

Polygales (Rimac, Peru), 9, 30.

Polygonacese (Andes, Peru), 9, 55. :

Polygonatum (Jakpho), 136; multi-
florum, 108. í

Polygonum amphibium, 99; bistorta
(connate cotyledon), 377; fruticulo-
sum, 55; volcanicum = Muehlen-
beckia volcanica, 55.

Polypetalæ (Peru), 10.

Polypodium atacamense, 64; alterni-
dens, 228; (Phegopteris) Baroni,
535; campyloneuroides, 229; dif-
forme, 228; dilatatum, var. grandi-
dentatum, 230; flabellinervium, 228;
ineurvatum, 230; irioides, 229 ; leu-
cophorum, 229 ; Melleri, 229 ; mini-
mum, 228; musefolium, 229; nor-
male, var. sumatranum, 228; num-
mularifolium, 228; ornatum, 227;
repandulum, 228; rupestre, 229 ;
sarawakense, 228; selligueoides, 229;
soridens, 228; squamatum, 64; su-
perficiale, 228 ; triquetrum, 229.

Polypogon interruptus, 58.

Polyporus sanguineus, 340. : :

Polystachya cultrata, 121; fusiformis,
121; minutiflora, 121; rosea, 121.

Polystichum angulare, var. pulcherri-
mum, apospory in (Druery), 437,
438. ne

Poponia micrantha, c

Fomio Carolinæ, 106; dilatata, 106;
monilifera, 106; nigra, 106, pyrami-
dalis, 106.

Porana a. fig

Porella capensis, .

Borini elegans, 315 ; flagellaceum,

556

315; Geheebii, 315; longirostrum,
315 ; madagassum, 315 ; pennseforme,
315; penneforme, var. brachyphyl-
lum, 315; ramulosum, 315; usaga-
rum, 315.

Portulaca oleracea, 3; quadrifida, 337.

Portulacacee (Diego Garcia), 337;
(Rimac, Peru), 32.

Poterium, 357, 390 ; Sanguisorba, 110;
(seedling) 359, 398, (sect. seed)
379.

Premna serratifolia, 338.

Primula Aurieula, 109, 114; capitata,
114; cortusoides, 114; denticulata,
114; farinosa, 108, 114 ; integrifolia,
114; luteola, 114; marginata, 114;
purpurea, 114; sikkimensis, 114;
sinensis (seedling), 398; stomata,
leaves diff. species of (Darwin), 114.

Primulaces (absent in Andes), 10.

Prionodon densus, 312; Rehmanni,
311.

Prionolejeunea, 324.

Pritzelia zebrina, 105.

Prockia (S. Amer.), 146.

Prosopis juliflora, 151; limensis, 150.

Protophyllum (foss.), 220.

Prunus domestica, 105; Laurocerasus,
105.

Pseudolarix Keempferi, 208, 209, 211,
212.

Pseudoleskea plagiostoma, 318.

Pseudopanax racemiferum, 167.

Pseudotsuga Jezoensis, 197 ; magnifica,
191; nobilis, 188.

Psiadia cuspidifera, 489 ; modestus, 490;
stenophylla, 490.

Psilotum triquetrum, 338.

Psorospermum emarginatum,
pulifolium, 453.

Psychotria (Grumilea) mesentericarpa,
485; Parkeri, 485; reducta, 484;
retiphlebia, 485.

Ptelea (ovary), 397.

Pteris ensiformis, 224 ; incisa, var. au-
rita, 224; longifolia, 224. ; marginata,
338; quadriaurita, 224, var. digitata,
224.

Pterobryum angustifolium, 312; filici-
num, 312; flagelliferum, 312; Han-
ningtoni, 312 ; trichomanoides, 312.

Pterocarya, 357, 392 ; caucasica, (seed-
ling) 359, (sect. nut) 386.

Pterogonium gracile, 316.

Pterygodium carnosum, 7 6; inversum,
76; magnum, 75; venosum, 76.

Pueraria (Nambre forest), 135.

Puerto Bueno (Patagonia), plants from,
167.

Puya ? (Peru), 3.

453 ; po-

INDEX.

Pyrrhobryum spiniforme, 308. _
Pyrus (Jakpho), 136; communis, 105 ;
Malus, 105.

Quercus Andersoni, 130; Griffithii,
130; pedunculata (seed), 397. — —
Quinchimalium ericoides, 56; gracile,

56.

Radula appressa, 323; Boryana, 323;
mascarena, 323.

Rajania quinata, 424. :

Ranunculacez, embryos, 393; (Rimac,
Peru), 9, 28.

Ranunculus chiclensis, 28 ; (cotyledon),
371; Flammula, 107, 109; petiole,
398 ; subaqueous leaves, 348.

Raphanus Sativus (cotyledon), 369,
370. ;

Rayana quinata, 424; Rehmanni, 300;
Giberti, 300.

Reidleia betsiliensis, 451.

Relation of the seedling to the seed,
400. ;
Report on vegetation of Diego Garcia,

Hemsley on, 332.

Restiaceæ, develop. inner row stamens,
415.

Rhabdoweisia (African), 302.

Rhacopilum africanum, 308 ; angusti-
folium, 308; angusti-stipulaceum,
308; Ayresii, 308; capense, 308;
longe-aristatum, 308; mauritianum,
308; mucronatum, 308; pectinatum,
308; prelongum, 308; Spelunce,
308.

Rhamnus (cotyledon), 393; Frangula
(seedling), 358.

Rheum (cotyledon), 398.

Rhipsalis crispata, 106, 107.

Rhizogonium spiniforme, 308.

Rhodea japonica, 105. :

Rhodoclada, united with Asteropeia,
479.

Rhododendron arboreum, 132; Fal-
coneri, 132; hirsutum, 105; ponti-
cum, 105. :

Rhodolena, 441 ; Bakeriana, 447 ; alti-
vola, 447. ;

Rhodosepala, 475 ; pauciflora, 479.

Rhodostachys bicolor, 166. SS

Rhus, 393 ; sp., 110; Thabo E
ledon), 357, 360; typhina (seedling),
353.

Rhynchospora tenuis, 144; spermodon,
=Dichronema setacea, 144.

Rhynchosia trichocephala, 465.

Rhyncoglossum, 135.

Rhyncotechum, 135.

Ribes aureum, 105.

INDEX. 557

Riccia spuria, 327.

Ricinus sanguineus (seedling), 348, 351,
363.

Ridley, H. N., Freshwater Hydrochari-
deze of Africa and its Islands, 232.

; Monograph of Genus Liparis,

4

;, On Dr. Fox's coll. of Orchids

from Madagascar, 116.

, On Rev. R. Baron's coll.of Orchids
from Madagascar, 116.

Rigodium Lechleri, 317.

Rivina levis, 339; (seedling), 352.

Robinia Pseudacacia, 110.

Rosa, sp., 110.

Rosacez (Peru), 9, 37.

Roscoea (Nambre forest), 134.

Rostkovia grandiflora, 168.

Rotala cordifolia, 478.

Rotboellia gracillima, 533; cæspitosa,
533.

Rourea, united with Byrsocarpus, 462.

Rubia tinctorum, 105.

Rubiacew (Andes, Peru), 42; (Diego
Garcia), 337; (Jakpho), 134; (Ri-
mac), 9, 42.

, tribe Spermacocee (Madagascar),
485.

Rubus (Kohima), 129, 130; moluc-
canus, 133.

Ruellia, 390; flava (seedling), 353;
longifolia (seed), 364.

Rumex Acetosa, 106.

Rumphius's plates of Limo, 216.

Ruta graveolens, 110.

Ryania chocoensis, 141.

Saccularia Veatchii, 153, 154.

Sachs on stomata on leaves of water-
plants, 99.

Salisburia adiantifolia, 105.

Salix Humboldtiana, 4.

Salvia Cruikshanksii, 54; rhombifolia,
54; strictiflora, 54.

Sambucus peruviana, 5, 41.

Santalace; (Andes, Peru), 9, 55, 56.

Sapindus (seedling), 357, 378, 393; in-
sequalis (seedling), 360.

Saponaria officinalis, 106.

Sargassum, 340.

Satyrium amenum, 126; calceatum,
127; candidum, 66; debile, 67;
emarcidum, 67 ; foliosum, 66 ; gigas,
116, 126; ligulatum, 67; ochroleu-
cum, 66; ochroleucum, var. 8. major,
66; trinerve, 126.

Sauvagesia erecta, 141.

Saxifraga cordillerarum, 5; cordillera-
rum, var. decipiens, 37; cordillera-
ram, var. nivalis, 37 ; decipiens, 37.

Saxifragese (Peru), 9, 37.

Scabiosa arvensis, 106 ; maritima, 164.

Sceevola Koenigii, 334, 337.

Schinus, 372 ; molle, 4, terebinthifolius
(seedling), 355.

Schismatoclada tricholarynx, 480.
Schistomitrium acutifolium, 302; api-
culatum, 302; cucullatum, 302

Lowii, 302; robustum, 302.
Schizopetalon Walkeri (cotyledon), 386,
387, 388.
Schkuhria isopappa, 46.
Scirpus setaceus, var. pygmeus, 167.
Scrophularinee (Andes, Peru), 9, 10,
13, 53; (Diego Garcia), 337.
Sedum, 37; acre, 107; album, 107;
andinum, 38; (Jakpho), 133. :
Seed, relation of seedling to, 400; size
of, 390.

Seedling, of Cress, 345; of Mustard,
345.

Seedlings, Forms and Causes (Lubbock),
341, 344; relation to seed, 400.

Selaginella echinata, 536. à

Selenipedium, 3-celled gyns;cium, 404 ;
caudatum, 420 ; Sedeni, 418.

Sempervivum tectorum, 107.

Senecio (Andes), 18, 441; compared
with Trixis, 14; (Naga Hills), 129.

, 441; acetosefolius, 496 ; ade-

nophylloides, 46 ; Anampoza= Cine-
raria Anampoza, 496 ; Bridgesii, 46;
casapaltensis, 47 ; cicatricosus, 497 ;
comosus, 46; cruentus, 385; cyclo-
cladus, 406; erucæfolius (seedling),
385; glacialis, 46; glareosus, 46;
(Kleinia) Hildebrandtii, 498 ; humil-
limus, var., 47 ; hyoseridifolius, 46 ;
laurifolius, 343; Mandonianus, 47 ;
melastomsefolius, 497 ; modestus, 47;
monocephalus, 498 ; Pentlandianus,
var, 46; Rabani (shrubby leaves),
129; Simonsii, 130; squalidus, 385 ;
tamoides, 496; vernicosus, 497 ;
viscosus, 385; vulgaris, 106, 168,
385; Wallichii, 129, 133.

Senecionidese, 14.

Sequoia, foss. Mull, 219. ;

Sibthorpia, from Kilimanjaro, 299.

Sicyos eae Sean 39.

Sida diffusa, 337.

Silene (Kohima), 136 ; inflata, 106.

Simarubesw (Diego Garcia), 337. —

Sisymbrium athrocarpum, 29; hirsu-
tum, 29; myriophyllum, 29; titica-
cense, 29. i

Sisyrinchium junceum, 56; Junceum,
var. =S. scirpiforme, 46.

Size of cotyledons, 394; of embryo,
393 ; of seeds, 390.

558

Skytalanthus acutus, 160.

Smyrnium perfoliatum (connate cotyle-
dons), 377.

Solanacee (Andes, Peru), 9, 10, 51.

Solanum angustifolium, var., 52; che-
nopodioides, 52; Dulcamara, 108,
109; (fruit & seed), 366, 366 ; Maglia,
52; mammosum, var, 143 ; Pent-
landi, 52; pulverulentum, 52; sp. ?
(Rimae Valley), 52; radicans, 52;
rhamnoides, 52 ; ruderale, 52; tube-
rosum, 8, 51, 52,105.

Solar Spectrum and Transpiration of
Plants (Henslow), 81.

Sonchus oleraceus, 110.

Sophora japonica, 109.

South-African Botany, Contributions
to (Bolus), 65.

South-American Botany and Botanical
provinces (Ball), 147.

Spathodea campanulata (seedling), 358.

Spergularia (seed), 365.

Spermacoce linearis, 142; parviflora,
142.

Spiradiclis (Naga Hills), 135.

Statice incana, 107 ; latifolia, 106.

Stauntonia, (3. chinensis, 423, 424;
hexaphylla, 423, 424.

Stauranthera, 135.

Stelis micrantha, 292.

Stellaria (seed), 365; leptopetala, 31;
media, 8, 31.

Stenomitrum (African), 305.

Stenotaphrum complanatum, 338.

Stenotheca, 48.

Stephanotis (seed), 364.

Stereophyllum nitens,
316.

Stevia panieulata, 43; sp.? (Chicla),
43

316; rivulare,

Stipa chrysophylla, 161; tortuosa, 160.

oe biotoides, 494; cryptophyla,
494.

Stomata in relation to bloom on leaves,
99.

eer of Magellan, Botany of (Ball),
167.

Stratiotes (African), 233, 237.

Streptocarpus caulescens, 299; Kirkii,
299; Rexii (cotyledon), 395.

Striga hirsuta, 337.

Strobilanthes hispidula, 509 ; (J akpho)
134, 135; (Madagascar) 441.

Strychnos, 441 ; Baroni, 504.

Sturmia, 244, 255; abyssinica, 265 ;
angustilabris, 287; bituberculata,
263; capensis, 272; elliptica, 265 ;
Darren, 261; Loeselii, 272 ; nervosa,
262.

Suriana maritima, 337.

INDEX.

Sycamore, cotyledons, 362; seed, 401;
straight veins, leaf of, 342.

Syringa vulgaris, 105.

Syrrhopodon asper, 303;
303 ; Gaudichaudi, 303.

Systegium crispum, 304 ; unguiculatum,
304.

circinatus,

Tacca, 136.

Tacsonia Van-Volxemi (cotyledon),
385.

Tagetes multiflora, 46.

Taltal, paucity of veget., 156, 341.

Tamus, curved veins, leaf of, 341,
342.

Tanacetum vulgare, 106.

Targionia hypophylla, 328.

Taxus baccata, 105 ; fossil, 219.

Tecoma sambucifolia, 54.

Telanthera densiflora, 154 ; (Peru), 3;
peruviana, var. decumbens, 155.

Telipogon, actinomorphie condition of,
419.

Temnolepis scrophularizfolia, 495.

Tephrosia cinerea, 150; litoralis, 150 ;
procumbens, 150.

Teratology of Cypripedium, 407.

Terminalia Catappa, 337.

Ternstromiaces, 146.

Tessaria ambigua, 156 ; (Peru), 3.

Tetrandrous flowers, 412.

Teucrium nudicaule, 157, 158.

'Thalamiflorz (Madagascar), 442.

Thalia dealbata, 110.

Thalictrum macrocarpum, 28; podo-
carpum, 28.

Thamnium complanatum, 315.

Thea (cotyledon), 400.

Thelychiton, pelorioid form of, 419;
stigma of, 403. a

Thelymitra, actinomorphic condition ot,
419.

Theophrasta Jussieui, 105.

Thesium cystoseiroides, 516.

Thibaudiæ (N. Andes), 148.

Thladiantha, 135. : ;

Thuidium gratum, 318 ; lævipes, 318 ;
tamariscinum, 318; versicolor, 318.

Thunbergia chrysochlamys, 508; Coc-
cinea, 135. à

Thyridium africanum, 303 ; fascicula-
tum, 303. :

Tiliaceæ, 146 ; (Diego Garcia), 337.

Tillea connata=T. rufescens, 37.

Tina velutina, 462. 5

Tocophila, paucity of veget., 156.

Tonina fluviatilis, 143.

Tortula, 304.

Tournefortia argentea, 334, 337.

Trachyloma stipitatum, 313.

INDEX.

Trachypus crispatulus, 314; serrulatus,
314.

Tradescantia subaspera, 101; virginica,
109.

Transpiration of Plants and Solar Spec-
trum, Rev. G. Henslow on, 81.

Trematodon ligulatus, 300.

Triana and Planchon, summary plant-
distrib. N. Granada, 145.

Trichocline (Peru), 13, 14.

Trichomanes brevipes, 223; fcenicula-
ceum, 223; Hosei, 223; maximum,
223; parvulum, 223; saxifragoides,
223 ; Smithii, 223 ; tenue, 223.

Trientalis, 358.

Trifolium, 35, 114; alpestre, 115 ; ama-
bile, var. pentlandi, 35; Aucheri,
115 ; ceespitosum, 115 ; chiclense, 35 ;
elegans, 115; fragiferum, 115; Lupin-
aster, 115; medium, 115; minus,
115 ; montanum, 115 ; pannonicum,
115; peruvianum, 35 ; pratense, 110,
115 ; repens, 109, 115 ; repens, bloom
protected surface of, (ftnote) 99;
striatum, 115; uniflorum, 115.

, stomata leaves diff. species of
(Darwin), 115.

Trimen, H., Note on Balanophora
Thwaitesii, Eichl., 330.

Trimorphopetalum, 454 ; dorstenioides,
455.

Triphasia trifoliata, 339, 373 ; (unequal
cotyledons), 373.

Tristania, (ftnote) 220.

Tristellateia discolor, 456; emarginata,
456 ; stenoptera, 456.

Triticum sativum, 111.

Tritonia, sp., 110.

Triumfetta procumbens, 337.

Trixis, compared with Senecio, 14; fru-
tescens, 156.

Trocheliantha (§ of Dombeya), 450.

Trollius europzeus, 105.

Tropsolum, 110; (seedling), 393 ; tube-
rosum, 34.

Tunica (cotyledon), 364.

Turbinaria vulgaris, var. decurrens,
340.

Turrea rhombifolia, 458 ;
458.

venulosa,

Ullucus tuberosus, 55; flabellata, 56.

Umbelliferee (Andes, Peru), 40; (Ri-
mac), 9,40; (embryos), 393.

Unequal cotyledons, 369.

Unsymmetrical cotyledons, 372.

Uropedium, 412, 417, 420.

Ursinia speciosa (achene), 361.

Urtica, 379; dioica (achene), 380, 384.

Urticacez (Andes, Peru), 9, 55, 56.

559

Utricularia, sp. ? (Colombia), 138, 143.
, subaqueous leaves, 343.

Vacciniacee (Jakpho), 133; (New Gre-
nada), 147, 148.

Vaccinium Vitis-idea, 101.

Valeriana armeriefolia, 42; coarctata,
42; interrupta, 42, var. minor, 43 ;
Phu, 106; remota, 42.

Valerianee (Peru), 8, 9, 42.

Vallisneria, 232, 233; sthiopica, 236 ;
nana, 236; spiralis, 236.

Vallisneriz (African), 232, 233, 236.

Valonia (Diego Garcia), 340.

Valparaiso, veget. of, 161.

Vanda ceerulea, 134.

VaseularCryptogamia(Madagascar),534.

Vascular plants, Diego Garcia, indi-
genous (Hemsley), 337.

Veprecella, 441; biformis, 447.

Verbena (Chicla), 5 ; diffusa, 5, 54 ; his-
pida, 54; litoralis, 54; officinalis,
54

Verbenacez (Diego Garcia),338; (Peru),
9, 54.

Vernonia, 441; betonicsfolia, 487 ;
(Strobocalyx) capresefolia, 487; ci-
nerea, 337 ; (Strobocalyx) exserta,
488 ; (Strobocalyx) grisea, 488 ; rho-
dopappa, 487 ; stenoclinoides, 486.

Vicia Cracca, 109 ; sativa, 109.

Viguiera (Peru), 3.

Villanova oppositifolia, 46.

Vinca major, 105.

Viola (Jakpho), 136.

Viscum album, 106 ; glomeratum, 515 ;
rhipsaloides, 516.

Vismia rufescens, 141.

Vitis ($ Ampelocissus), 441; rhodo-
tricha, 461; sphsrophylla, 460; vi-
nifera, 105; (S Cissus) Voanonala,
461.

Vittaria sulcata, 230. :

Voamboana timber (Madagascar), furni-
ture made from, 466. =

Voanonala, native name of Vitis Voa-
nonala, 461.

Vochysiacee (S. Amer.), 147.

Wahlenbergia peruviana, 49.

Walchia imbricata, 201.

Walnut (seed), 348.

Weinmannia leptostachya, 469.

Welwitschia (perisperm), 387, 394.

Wendtieæ (Peru), 13.

Werneria strigosissima, 47,= W. setosa,
47.

Western South America, Botany of
(Ball), 137.

560

Wiesner's conclusions on light and
plant-transpiration, 83-85.

Wild source of cultivated Limes (Dr.
Bonavia), 213.

Woodsia obtusa, var., — Cheilanthus
crenata, 63.

Xyridaces, develop. inner row stamens
of, 415.

Yucca aloifolia, 106, 107.

INDEX.

Zapania cuneifolia, 154.

Zea Mays, 106.

Zilla myagroides (cotyledon), 380.

Zinnia elegans, 106.

Zygodon (§ Stenomitrium) erosum, 305 ;
crenulatus, 306; pentastichus, 306;
Priessianus, 301; Rehmanni, 306;
rugifolius, 306; runcinatus, 306;
strictissimus, 301; trichomitrius,
305 ; quitensis, 306.

Zygostates cornuta, 403.

END OF THE TWENTY-SECOND VOLUME.

PRINTED BY TAYLOR AND FRANCIS, RED LION COURT, FLEET STREET.

HBolus . : j : LINN Soc. Journ. Bor VoL. XXILPr.1. 7

JNFitch hth . . Fitch imp.
S.AFRICAN ORCHIDS SPECIES OF DISPERIS.

A.1-5 D.Weali. B.612 D.oxyglossa C1317 D Macowani.

D.18-22 DWoodit. E 2327 D namaouensis.F 2833 D Iysoni.

Dr. M. Masters, Linn. Soc. Journ. Bor. VOL. XXIL Pr, 3

"Y 1

$^ HR

$ " In f,
r

^

ABIES AMA BILIS. j fj STERILE BRANCH.

Dr. M. Masters. Lryy. Soc. Journ. Bor. Vor. xxii. Pr. 3.

ABIES GRANDIS.

M

cmm
de aad

NOS
wes

vits
ARTT

umm me

Linn. Soc. Journ. Bor. Vor. xxr. Pr. 4.

t
( ue e Mut pum

eae

Section of leaf, magnified, showing groove on
upper surface ; A A, resin canals.

B

Hy
NN | /
Ce
\ ~, N Í
N by i
N e NOS
N \ |

Y

Seed, scales, and bracts of cultivated specimens.

ABIES NOBILIS.

i : i 5
LixN. Soc. Journ. Bor. Vor. xxrr. Pr. 5.
Dr. M. Masters,

UT) il : |
- Oe ili D E:
& 2, = Gs My (i : d ! >
Gere LBA
M
T A es
à) VS
=

ZF

i pr ai
Nen
^ jn D eu
ie e com fpa enr
mb No: INANE | ALD T c
n » all SS d S ; Zi 77 1 M
qme » S S P í m j tale Hf; |

ip

f M

"mU. ud. A xtti
TN DN @ZZ~ fe an "Ul | hi 1

= "
ai

: lij
"Vertit. di

ABIES NOBILIS, VAR. MAGNIFICA, HORT.

Dr. M. Masters. Liny. Soc. Journ. Bor. Vor. xxir. Pr. 6.

ES

ABIES RELIGIOSA, MEXICAN SILVER FIR: ho M

Dr. M. Masters.

Liny. Soc. Journ. Por. Vor. xxix. Pt. 7.

CEPHALOTAXUS PESUNCULATA, VAR. SPHJAERALI3. , i

Lryy. Soc. Journ. Bor. Vor. xxrr. Pr. 8.
Dr. M. Masters,

Linn. Soc. Journ. Bor. Vor. xxit. Pr. 9.

Dr. M. Masters,

Linn. Soc. Journ. Bor. Vor. xxi. Pr. 10.

REFERENCES :—A, Male catkins, section; B, C, D, E, Stamens (B, bottom, E, top, C, D, intermediate positions) ; F, a, Pollen.

J.U-Baker.^ fuos Linn.Soc.Journ.Bor.Vou. XXII Pri.

LNF&ch lih . Fitch ix

NEPHRODIUM (SAGENIA) STENOPHYLLUM, Baker

J.&Baker. Don coe oun BoT Von. XXII Pr 19.

Rg.1.

.J.N Esch | Fitch imp
oe TRICHOMANES HOSEL, Baker. d

HNRic'ay . LINN Soc. Journ. Bor. VoL XXII. Pt 13.

Linn. Soc. Journ. Bor. Vor. XAN Pr 4

HNRidley .

BLYXA RADICAN S, Ridley. Fitch imp.

JNFach lih.

Linn. Soc. Journ. Bor. Vor. XXII. P115.

W. Mitten.

N64
intera wap

HANNINGTONT.

ww adu NNI.
OPHILA PLICATA

T9
13-16. HY

-22.PTEROBRYUM

Ls

SILLUM. 20

l-4 ARCHIDIUM AFRIC^'""
9-12. CAMPYLOPUS PERPUSILLU

-

ANCECTANGIUM PU

17-19.

E Carter sc.

W. Mitten.

Liny. Soc. Journ. Bor. Voi. XXII E116.

T5 eee

us Fa

"m prevea xat d

Mintern imp
l3 PTEROBRYUM PLAGELLIFERUM 4-7 EREODIUM HANNINGTONI.
8-10 PLAGIOCHILA SINUOSA 1- 12 "E BARTERI.

E Carter sc.

MN "nu E
N
bl

E Carter $c.

1-3. PLAGIO

Lun Sóc. Journ Bor Vor. XML PIIR

A. 79. EABYSSINICA
`: 10-14 LEIOCYPHUS INFUSCATUS.

CHILA LASTIL 4-6 EDICRAN
Mintern imp

Linn Soc Journ Bor Vor XXII P118.
as" |

3

ECarter se Mintern imp

1-2 BAZZANIA PUMILA. 3-4. PHRAGMICOMA FLOREA.
5-8 LEJEUNEA HANNINGTONI. 9-12. L. ATRA. 13 16. L.SINUATA.

W.Mitten. Linn Soc. Journ. Bot. Vor XXII. P119.

E Carter sc.

1-à LEJEUNEA SERRULA 4-6 L.DEPLANATA. 7-9 LADH/ESIVA
10-11. 1, DIGITALIS 12-14 FRULLANIA USAGARA.

Dr. M. T. Masters.

Lryn. Soc. Journ. Bor. Vor. xxrr. Pr. xx.

1 "—-— . TN NBN NY
7 E [xd
\ m a aa \
Wd £ ar à \ om })
í SOA dy \ NI i y
a3
PL
AI
F, P
fe NES.
} at ; ` a2
D)
>
a3
S S

man ee”

SELENIPEDIUM CAUDATUM
INTERMEDIATE BETWEEN THE ORDINARY FORM AND UROPEDIUM.

(See page 419.)

‘Wataro Ito del.ad nat . Lr. Soc. Journ. Bor. Vou. XXIL. Pr. Z1.
d 3
ey f
LE i
ek aed EU po
à e KH T M4
SNNT !/7/ Pt
SA ON i yy "a