suunv tj

S ALUTi

^LUbraryqe

OTYOROOTANICA

Digitized by the Internet Archive
in 2016 with funding from
BHL-SIL-FEDLINK

https://archive.org/details/amaryllidaceaeOOherb

■

Flah.il>.

"WJIerbeTt.del .

jjji. Wt&dalL.fc.

AMARYLLIDACEA3 ;

PRECEDED BY

AN ATTEMPT TO ARRANGE

THE

MONOCOTYLEDONOUS ORDERS,

AND FOLLOWED 15Y A

TREATISE ON CROSS-BRED VEGETABLES,

*”■> library

NEW YORK

SUPPLEMENT, bot.am.

tiAklic’N

BY THE

HON. AND REV. WILLIAM HERBERT.

“ Flores hujus generis exiinii sunt ; nescio num secundatn parem habeat; hinc Bellae
donnee dictse plures. Bella donna Virgilii, Amaryllis dicta, nomine transiit in proverbium
de onmi grato.” — Linn, de Ainaryllide, Hurt. Cliff, p. 135.

WITH FORTY-EIGHT PLATES.

LONDON :

JAMES RIDGWAY AND SONS, PICCADILLY.

MDCCCXXX vir.

TO HIS MAJESTY

LEOPOLD,

KING OF THE BELGIANS.

Sir,

My recollection of many interesting conversations
with your Majesty on botanical and horticultural topics
assures me, that the subject of this work, however inadequately
treated, will not fail to he interesting to your Majesty, not-
withstanding the important occupations of your present ex-
alted station. I am induced by a sense of the unvarying
kindness shewn to me by your Majesty, while resident in this
country, to inscribe this volume, with your gracious per-
mission, to your Majesty, and to subscribe myself respectfully,
with every wish for the prosperity of your Majesty and your
kingdom,

Your Majesty’s

Most grateful

And obedient humble Servant,

Spojforth,
December 26, 1836.

WILLIAM HERBERT.

JAN | 3 1926 fa*

ADVERTISEMENT.

LIBRARY

NEW YORK
BOTANIC A!
Garden

~*v

The receipt of a fresh supply of valuable specimens, for
which I am indebted to Sir W. Hooker, after the printing of
the body of this work had been completed, has enabled me to
add amongst the supplemental observations the description of
several fine plants before unknown. The supplemental articles
have been arranged in alphabetical order, as the most conve-
nient for reference, and it will be apparent what place each ad-
ditional plant should occupy in the consecutive arrangement.
The reader is particularly requested to refer from Alstroemeria,
Bomarea, Crinum, Habranthus, Haemanthus, Iiermione, Hip-
peastrum, Oporanthus, and Sternebergia to the Supplement.
From each of the magnificent umbels of additional species
of Bomarea, I could only afford space to represent a single
flower and leaf, which will be sufficient to facilitate their
identification. The publisher having been desirous of offering
some copies to the public with coloured plates, it must be
understood that, where no live specimen has been seen in
this country, and no precise memorandum of the colours has
been given by the collector, the plates could only be made to
represent the existing tints of the dry specimens, which are
in many cases very fallacious. Those who have been accus-
tomed to examine dry specimens, will however be able to
form a better judgment by seeing what the existing colours
are.

It is my intention, as opportunities may occur, to prepare
memoranda for rectifying and supplying the deficiencies of
this work, and any communications relating to it, if left for
me free of expense at the publishers, will be attended to
thankfully. Any dry specimens, seeds, or roots, of newly
introduced Amaryllidaceous plants, especially from the

b

VI

ADVERTISEMENT.

Western hemisphere, will be gratefully received. Crinum
Forbesianum, Amar. grandiflora, Brunsv. striata, minor, ra-
dula, Nerine marginata, Strumaria, Hessea, Imhofia, Carpo-
lyza, and Pane. Canariense amongst the African, Sterneber-
gia, Erinosma (Leucojum) Carpathicum, Lapiedra, Vagaria,
Tapeinanthus, Hermione elegans, serotina, Queltia juncifolia,
pumila, pusilla, Ganymedes cernuus (triandrus), capax, and
reflexus, seed of Q. foetida, odora, montana, Maclaeana, are
more particularly wanting to me amongst the species of the
Old World.

The reference to the figures of Pancratium Cambayense
and longiflorum, pi. 42, is omitted by accident, p.207-8, but
is marked in the index. The Glossary, for the use of un-
learned readers, will be found at p. 417.

I have not thought it adviseable to load this work with
voluminous quotations and references to old works, which the
reader would probably not be desirous of consulting. It will
be understood that, where I have not expressed the contrary,
I mean to assent to the synonyms and references given by the
modern writers who have treated of the several plants de-
scribed, especially Reaumer and Schultes, and the editors of
the Bot. Mag. and Reg.

It has been suggested to me to extend my labours to Iri-
dacese, and perhaps to Liliacese, and ultimately embrace the
whole hexapetaloid portion of agynandrous exspadiceous
plants. I doubt whether I shall have sufficient leisure, and
it remains to be seen whether the reception of this work will
encourage the publisher, who has liberally promoted it, to
any further undertaking; and indeed, whether I can obtain
access to the plants it would be necessary to investigate ; but,
if those who may possess them will assist me in that respect,
I am not indisposed to make preparations for the accomplish-
ment of such a work.

WILLIAM HERBERT.

Spojforth , Dec. 1836.

EXPLANATION OF TERMS.

As these pages are intended for the use and assistance of
the unlearned cultivator, as well as for the edification of the
scientific, I wish to premise a few words concerning some
terms which will be used in them. It has been justly con-
sidered that, in plants which, like the Amaryllideae, have no
calyx, the outer divisions or segments of the flower stand in
lieu of calyx, and the inner of corolla, and such a flower is
called a perianth, the outer or calycine segments sepals, the
inner or corolline segments petals. The number of filaments
being equal to those segments at or below the base of which
they are usually inserted, I propose to call those which belong
to the outer segments the sepaline filaments, and those which
belong to the inner the petaline filaments. I find a great
laxity in the language of botanists in defining flowers which
have a tubular appearance, though perhaps no tube at all,
which they call indiscriminately tubular, tubulosi, so that it
cannot be ascertained from their definitions whether the peri-
anth is really tubular or divided to the base. I propose to
rectify this by the following appropriation of distinct terms;
tubatus, tubed or having a tube, whether long or short ; tubu-
losus, long-tubed ; tubiformis, tube-shaped, or having the sem-
blance of a tube ; tubaeformis, trumpet-shaped : and I hope
that such a necessary distinction may be observed in all fu-
ture definitions. I find also that it is not possible from the
definitions of any botanist to ascertain with certainty whether
a perianth is really divided or not, as they are in the habit of
calling flowers trifid or six-cleft, tripartite or sexpartite, of
which the segments are connected at the base ; I shall call the
perianth sexpartite where the segments are unconnected, deeply

B

2

EXPLANATION OF TERMS.

cleft where there is an annular union, and I consider a mere
annular union of the base to be scarcely a tube. In plants
which are only known to me by the description of others, I
cannot ascertain the fact, in consequence of the laxity of their
expressions. Latin adjectives in osus always signify an excess
of the quality; such words have been greatly misapplied by
botanists, and I erred in naming a plant in the Botanical Re-
gister Pharium fistulosum, which must be called P. fistulatum,
for fistulosum really means either full of pores like a sponge,
or having one enormous pore. I shall use such words strictly,
ex. gr. Spadicosus having a large spadix ; spadiceus or spa-
dicatus having a spadix. Dr. Lindley has unfortunately
made an extensive use of the termination osus, which it is very
necessary to rectify. The termination inus may be conveniently
substituted to answer his purpose ; but the subversion of
latinity by the misuse of a termination which has a decided
meaning is very objectionable. We are forced to create mon-
grel Latin to express what the Romans never thought of ex-
pressing, but we should write correct Latin where we can ; on
which account I protest against the use, too frequent amongst
botanists, of 'tills and eis for those with a genitive following,
as semina illis Amaryllidis conformis instead of semina Ama-
ry^llidi conformis or conformia. By a linear leaf I under-
stand, that the liijes of its margin are parallel. Much con-
fusion has been made by using the word linearis to signify
also indiscriminately a line long, or a line wide, which should
be expressed by lineam latum ; nor does it seem to be uni-
formly established, what portion of an inch is meant by a
line, being the tenth part of an inch according to Dr. John-
son, and the twelfth according to French mensuration. The
unlearned reader is further referred to a glossary at the end
of the volume.

PRELIMINARY TREATISE.

Having been requested to prepare for the press a second
edition of a treatise, which I published in 1821, on the
plants included previously under the genera Amaryllis,
Crinum, Pancratium, and Cyrtanthus, I have felt that it
would be desirable to render it as complete as the further
information I now possess concerning them will enable me,
though it is in some respects still defective, and it was repre-
sented to me that it would be preferable to extend the scheme
of my work to the whole natural order of Amaryllideae ;
but, conformably with the latest practice, 1 have adopted the
superior term Amaryllidaceae, and confined the former name
to a subordinate portion. I have to regret however that,
not having contemplated a further publication, I have
neglected during the last fifteen years opportunities of noting
down points, which I cannot now readily ascertain. In
pursuance of my present undertaking, it was a principal
object to arrange according to their affinities the different
kinds of plants of which the order consists, a task by no
means easy, because the concatenation of vegetables does
not proceed in a straight line, and perhaps the truest arrange-
ment would be in a circle, with lateral lines from some
points in the circumference either falling into it at some
other point, or branching off* to meet some other order. For
this reason, although a consecutive arrangement is necessary
for convenience, it cannot be expected that the concatenation
should be perfect. In proceeding with such a work, the
first point to be considered is what plants are comprised
under the natural order Amaryllidaceae, and what are its
essential characteristics. A great difficulty occurs here at
the outset; for although the system of classing plants by what
modern botanists have called natural orders is entirely the
fashion of the present day, every writer who has treated of
this order refers to Dr. Brown’s Prodromus for the definition,
and on reference to the work of that most distinguished
botanist, on whose accuracy in all points that he has tho-
roughly investigated we may peculiarly rely, we find a defi-

b 2

4

PRELIMINARY TREATISE.

nition founded upon facts incorrectly assumed, in consequence
of an incomplete knowledge of the plants which it is reputed to
comprehend. It is therefore the first necessary part of my
labour to frame a new definition of the natural order, so as
in truth to comprise the plants it contains; and my definition
will be found to include some extensive genera which have
not been usually placed under it. Of such I shall think it
sufficient to point out the characters, without entering into a
full account of the species which are numerous, especially in
Ilypoxis, and not well known to me, but of which the details
may be found, as far as they are ascertained, in the valuable
and comprehensive work of Reaumer and Schultes. Before
I can proceed to the execution of the task imposed upon me,
it becomes necessary to consider what is that sj^stem of
natural orders which has nearly superseded the Linnsean, or,
as modern writers term it, the artificial system. It appears
to me that a more gross misuse of words has rarely invaded
any department of science, for if ever an arrangement was
artificial, it is that now adopted of vegetable orders, the
characters of which depend upon a variety of features taken
ad libitum , and in many cases erroneous in consequence of
insufficient information, and which when defined have been
thrown together in a variable mass of confusion according to
the successive notions of different writers concerning their
affinities to each other, and which cannot possibly be placed
in any natural succession, because they are like octagon,
hexagon, pentagon, and other angular figures, coming in
contact with each other at their various sides, so that some
natural affinity must be torn apart when they are put in
succession. If we refer to Browm’s valuable Prodromus
and to Sweet’s Hortus Britannicus, we find the orders placed
in different succession ; and even if the point of precedence
could be definitively settled, which is not likely to be the
case, it depends on opinion and not on fact ; nor can the
arrangement be used for reference without an index ; for
no person, unless very deeply versed in the science, can
be expected to remember the relative position of between
two and three hundred orders, or to come easily at the
facts on which the arrangement is built. I must pray the
forgiveness of those who are much more deeply versed than
I am in botanical studies, when I venture to say that
something better digested is requisite for a general system of
botany. The Linnsean, with its imperfections, is an arrange-

PRELIMINARY TREATISE.

merit at least as natural (for indeed all arrangements
must be artificial) and it is founded upon facts of easier
access, the succession of classes being for the most part
necessary, and laying no great tax on the memory.

So vague are the given characters of the orders however
amended, and upon so weak a foundation does botany stand
at present, that although perhaps the only point that truly
divides Asphodeleae from Amaryllidese of Brown is the
flower of the latter growing above the germen and of the
former under it, that feature is passed over as undetermined
in Dr. Brown's definition of Asphodeleae, and he rests
mainly the character of that order on the seed-shell being
black crustaceous and brittle, and those of Ilemerocallideae
and Amaryllideae on the seeds being neither black nor
brittle, which distinctions appear to vanish upon further
examination ; for I have at this moment before me
seed of Albuca amongst the Asphodeleae with the shell
softer and less fragile than any of the black-seeded Amaryl-
lideae or at least equally so, and that of Hemerocallis, from
which the second of these orders is named, with the shell
black as jet, and perhaps as fragile as that of any seed that
exists. In the next place on the assumption of the facts
above stated, we find Dr. Brown subsequently constructing
a new order of Hypoxidese ( App . to Flinders ), separated
from Asphodeleae by the flower above the germen, and from
Amaryllideae by the hardness of the seed-shell and the vague
direction of the radicle. A black-shelled seed, almostequally
hard, will be found in Pancratium Illyricum, and others of that
order, and there seems to be rather a graduated difference of
hardness than a diversity of structure. I have not made any
microscopic observations on the interior of the minute seed of
Hypoxis, and any feature of such difficult investigation seems
to me unfit to characterise a natural order of plants, though
very proper to be subjoined as a subsidiary observation. I
apprehend that Dr. Brown must mean that the original
posture of the radicle in the embryo is a little irregular, a
point which does not appear to me of high importance, for
I apprehend that in a hard-shelled seed it must ultimately
issue at the natural passage which is the foramen. The
ultimate direction of the radicle is vague in Crinum and
Hymenocallis, and it pierces the fleshy mass with an irregular
direction, and does not often issue at the natural passage.
Its direction is also in those seeds often vague from the first.

6

PRELIMINARY TREATISE.

Dr. Brown has mentioned elsewhere the strange circum-
stance, that when those seeds are ripe and detached from the
capsule, the embryo is often not discoverable, yet will become
apparent after it has lain awhile on the ground, and that
an artificial direction may be occasionally given to it ;
and in such cases the original direction of the embryo is
vague. For instance an Hymenocallis in my stove had
ripened a considerable number of seeds above an inch long ;
I cut open several and found a cavity within the fleshy
mass, but no appearance of an embryo, and the seeds might
have been supposed to be incapable of vegetating ; the
remainder were laid upon a pot of earth, and after the lapse
of several weeks every one of them sprouted vigorously, the
radicle issuing on the side in contact with the earth. In such
case it must have been drawn towards the earth by some
influence, and, if not bent at its earliest manifestation, the
point of the embryo must have been kept straight at first by
the toughness of the inner coat of the cavity, and have turned
the moment it had issued from it; but its being kept straight
by the toughness of the inner coat of the cavity in which it
lies, till it can find a vent, cannot be a distinguishing feature
of very high importance. I should have supposed that the
embryo, though not discoverable by the magnifier I applied
to it, must have existed originally in a very minute state,
but Dr. Brown asserts in the Linnaean Transactions, xii. 149,
that its formation is subsequent to the separation of the seed
in such cases ; a recondite point, which I have not had a
sufficiently powerful magnifier to investigate. I beg not
in any manner to be considered as imputing blame to a
gentleman whose botanical skill and information exceeds
mine immeasurably, and who has minutely enquired into
many difficult points connected with the science, on account
of any lapses, which are the necessary consequence of an
imperfect knowledge of the profusion of various yet kindred
vegetables, with which the Almighty has adorned our
world ; but when fundamental errors meet me in the
outset at each step, without travelling beyond the im-
mediate object of my labours, I must be allowed at least
to say that the science is yet in its infancy, and that its
most distinguished professors are still feeling their way in
the dark, or at least have not been yet able to emerge into
broad day-light.

My friend, Professor Lindley, whose writings are daily

PRELIMINARY TREATISE.

7

adding to his deserved reputation, conscious of the unsatis-
factory and chaotic state of the present system of botany,
has attempted in a very ingenious and useful tract to arrange
the orders by some intelligible and regular course of divisions.
Valuable as his labours are in this and in all other respects,

I may say without any offence to him, that such are the
radical and insuperable defects of the system, that the
attempt only serves to make darkness visible, and shew the
points wherein it is utterly artificial and repugnant to nature.

I cannot, however, proceed with any observations that have
the least bearing on the botanical labours of that gentleman,
without previously expressing not only the high opinion I
entertain of his talents and acquirements, but the obligations
under which I feel myself to him, not only for the liberality
with which he has entrusted to me portions of his library
and valuable herbarium, but for the invariable urbanity and
kindness with which he has favoured me with his opinion
and instruction on some points, concerning which I found
my general information deficient, and indeed upon every
occasion in which I was desirous of his advice. I may take
this opportunity of returning equal thanks to Sir. W. Hooker
for the liberality with which he has sent tome from Glasgow
the whole of bis valuable specimens of Amaryllidaceous
plants, enabling me to describe and make outlines of a great
number hitherto unknown, and for the alacrity with which he
has replied to any questions I had occasion to put to him.
Mr. Bentham has kindly communicated his specimens of
Narcisseae. To Dr. Brown and Mr. Bennett of the British
Museum and to Professor Don I am indebted for the civility
with which they assisted me to inspect the Banksian her-
barium and those of Mr. Lambert ai>d the Linnsean
Society ; and from Dr. Graham, the distinguished professor
of the Edinburgh University, and Dr. Neill, of Canon-mills
Cottage, near Edinburgh, I have received some interesting
communications. To Mr. Sabine and to the garden of the
Horticultural Society I am indebted for a copious supply of
fresh specimens of Narcissi ; to the Rev. William Ellicombe for
some others ; and I owe thanks to the liberality of Mr.
Murray of the Glasgow botanic garden, and Mr. Loddiges
of Hackney, for the opportunity of examining some plants
which I should not otherwise have seen ; and to Mr. Anderson
of the Chelsea Garden for seeds of Narcissi, and for some
information concerning the several varieties.

8

PRELIMINARY TREATISE.

Dr. Lindley perceived, with his usual clearness of intellect,
that (assuming the natural orders to be correct, or at least
capable of being so reformed as to become unexceptionable)
they must be distributed and classed under a succession of
more satisfactory superior divisions before they could be
made available for a scientific arrangement. Through his
ample information concerning the structure and diversity of
vegetables, I yet hope that he will be able ultimately to put
together a stable and true system of botany ; and it will be
enough for me, who have not sufficient knowledge of facts to
enter into all the details, if I can point out the fundamental
errors of the existing system, and the mode by which accord-
ing to my view a better must be constructed. There seem
to me two radical errors in Professor Lindley’s alliances. I
do not, however, impute them to him, but to the system,
which it was perhaps not easy for him to deal with more
accurately in its present state. The first is, that he attempts
to place the greater part of vegetables in subdivisions equally
distant from the point of universal agreement, which is
artificial in the extreme, and in direct opposition to nature.
At the first, step he gives us five classes, which in truth are
formed of a primary, secondary, and tertiary division
united. He next subdivides the dicotyledonous plants
into four grades, viz. sub-classes, groups, alliances, and
orders; the monocotyledonous, being fewer, into three, viz.
groups, alliances, and orders; the sexless and root-flowering
into orders only. But the Almighty did not thus restrict
himself in the affinities of his creation ; he did not make the
immense variety of vegetables with a limitation that they
should branch out into five or six or any defined number of
points of difference from one original archetypal structure.
There is a point in which all vegetables must agree ; that
which separates them from the other parts of the creation.
Some may detach themselves perhaps from the whole mass
of vegetables by a single peculiar feature; some may separate
themselves at an early period, and others after a long series
of previous subdivisions. There is no law that equalizes
their affinities to each other ; and any system that shall
pretend to subdivide them by a limited number of grades,
must be fundamentally false. My second objection is that
the divisions are inconsistent and contradictory, which is,
perhaps, a necessary consequence of the first wrong step.
For instance, Dr. Lindley’s second group of many-petaled

PRELIMINARY TREATISE.

9

bicotyledonous plants is Epigynosae, that is to say, having the
ovary under the stamens ; but this group of epigynous
plants is not opposed by him to those which are not epigy-
nous, but only includes such of the epigynous plants as he
finds it convenient to place in it, while he puts two whole
alliances of epigynous orders in the first subclass, though
the being epigynous is the distinguishing characteristic of
the second. 1 repeat that I am not finding fault with him,
but with the chaos from which he was struggling to emerge,
if it had been practicable. Let me be clearly understood.
After having separated (as I shall propose to do) corolliform
flowers, from those which are glumaceous or scaly, like
grasses, it may perhaps be found advisable in one of those
classes to place the division epigynous and its converse
hypogynous before, and in the other after, the divisions
hexandrous and triandrous, if such a disposition shall appear
to group the vegetables better according to their general
aspect and affinity ; but having once assumed the epigynous
position of the stamens as the limiting feature of a class,
we cannot place epigynous plants in the other classes which
have been cut off by the absence of that feature ; for if the
class epigynous is limited to those epigynous plants only which
have not such and such other features, those other subordi-
nate features are thereby exalted into a primary station and
the separations are confounded ; the points assumed for
distinction may possibly be correct, but they are not properly
exhibited. This cannot be illustrated better than by examin-
ing Dr. Lindley’s third subclass, viz. the monopetalous
plants, which he thus divides into five groups ; the first,
consists specially, of those whose ovary is composed of many
carpels ; the third of but one perfect carpel ; the fifth of two
carpels ; the second of plants with epigynous stamens, i. e.
inserted above the ovary ; the fourth of those whose flower
has a lip. Here are three features, the number of carpels or
folded leaves in the structure of the ovary, the position of the
stamens, and the lip of the flower, set in opposition, though
not opposite, but consistent with each other. Consequently
we find the whole fourth class composed of ovaries with two
carpels, though that is the characteristic of the fifth ; the
second partly of those with two, and partly of those with
many carpels, which should belong separately to the first
and fifth ; the first and third contain epigynous plants,
which belong properly to the second ; and the fifth is

10

PRELIMINARY TREATISE.

described as consisting of unsymmetrical flowers, yet there
are others unsymmetrical in the third. The view of the
whole is therefore confused, and no true division is made.
But this may be reduced to a lucid order. Supposing the
number of carpels to be of high importance, concerning
which I do not wish to give any opinion, the first division*
would be Aggyrosse or rather Monocarposae, i.e. of one
carpel; 2. Dicarposge, of two carpels; 3. Polycarposae, of
three or more carpels. Each of those divisions may then be
divided into symmetrical and unsymmetrical as to the flower ;
and those divisions again into epigynous and its opposite
hypogynous. There is a like confusion at the outset, where
vegetables are divided improperly into five primary classes.
1. Outgrowing, witli reference to the mode of increasing
their bulk by outward deposits. 2. Naked-seeded, which
are also outgrowing. 3. Ingrowing, or increasing their
bulk inwardly. 4. Root-flowering, which are not stated
whether to be outgrowing or ingrowing. 5. Top-growing.
Of these the two first are dicotyledonous, having two seed-
lobes, the third monocotyledonous, having one, and the
fifth acotyledonous, having none. The four first of these
are sexual, and the last sexless or rather sex not apparent.
The first real division is Sexual and Sex not apparent.
The next division of sexual is Acotyledonous, which takes
his whole fourth class, Monocotyledonous, which takes the
third, and Dicotyledonous, which covers the first and second ;
and dicotyledonous may be then, if it be thought fit, divided
into Naked-seeded and Close-seeded. If Dr. Lindley will
apply his extensive knowledge and correct judgment to
elucidate the system in this manner, he will place it on a
satisfactory and permanent footing, being careful however
not to let any feature rank above those which should pre-
cede it. I believe him (as well as that distinguished con-
tinental botanist Monsieur Decandolle) to have a mind much
above the narrow wish of adhering to any thing because
he has published it, or rejecting any thing because it
is suggested by one like myself, of very inferior botanical

* In the second edition of Dr. Lindley's Natural System of Botany these
divisions are altered to, 1. Polycarpous. 2. Epigynous. 3. Monocarpous. 4.
Nucamentaceous. 5. Dicarpous. The real arrangement should be (supposing
the points assumed to be the most valuable that could be found). 1. One-
carpelled. 2. Two-carpelled ; a. capsular, b. nucamentaceous. 3. Many-
carpelled ; a. hypogynous. b. epigynous.

PRELIMINARY TREATISE.

11

knowledge ; and I would earnestly excite him to perfect
that system upon which he has thrown a strong, but as yet
insufficient, light. I apprehend no ultimate difference of
opinion between me and him, or any other person of clear
understanding and unprejudiced mind, because I build only
on fact, and by that test I wish every thing I advance to be
tried, and should instantly correct any thing I found incon-
sistent with it. It is a main object of this Treatise to reduce
the divisions which rest on opinion, to their proper insig-
nificancy.

The principal merit of the Linnaean division was, that
each separation rested on a single fact of pretty easy access,
although it might remove to too great a distance genera,
associated by other peculiarities. The system now in vogue,
will be found to rest in many respects on features quite as
artificial, which separate kindred genera as objectionally as
its predecessor. I am at a loss to conceive, in what
manner it can possibly be substantiated, that the position of
the stamens adopted by Jussieu is a more natural feature
for classification than their number, by which Linnaeus was
guided in most of his classes. They are evidently facts of
like nature and deserve about equal weight ; but the position
of the stamens, instead of ranking high amongst natural sub-
divisions as assumed by Jussieu, ought to occupy a very
subordinate place. To illustrate this I may state, that there
are plants amongst Amaryllideae, which but for the difference
of having the perianth and stamens superior instead of in-
ferior to the ovary, would be almost identical with others
amongst Asphodeleae; for instance, if the scentless Alliums
of the latter, (an occidental race forming, I believe, a separate
genus, which might be called Pseudoscordum), had the
ovary inferior, it would require nice discrimination to sepa-
rate them from Lapiedra of the former, and their general
aspect would touch very close upon Strumaria, though there
would be points of distinction; yet plants which are separated
from actual identity of genus by little more than that feature
and some difference of seed, which escape the observation
of an uninstructed observer, are not merely removed by Dr.
Lindley’s arrangement into a different alliance, but two
whole groups, ten alliances, and twenty-five natural orders
intervene between them. Such incongruities are found in
every mode by which the orders have been arranged. In
Sweet’s Hortus Britannicus we find the water-lilies close to

12

PRELIMINARY TREATISE.

the berberries: and such is the natural system which is to
supersede the artificial system of Linnaeus. It is founded in
part upon facts of no greater importance, in part on an
assumed coincidence of a variety of points, concerning which
the knowledge of man is still very imperfect, and which from
time to time, as our information becomes more ample, evince
themselves to have been incorrectly assumed. In conse-
quence of the discovery of such lapses, fresh subdivisions
are made, and new orders successively carved out of the
wrecks of the original divisions, to be themselves overturned
in like manner by fresh curtailments : and in the mean time
the orders, between two and three hundred in number, are a
mass of unfathomable confusion.

There is but one mode of proceeding, with a view to
place the divisions on a sound and durable footing, that is,
to found every separation on a single fact, and to work
downwards from the first division, with cautious examination
of the relative importance and consequent priority of the
facts, by which the subordinate divisions are to be limited.
This has not yet been done ; but, whether I live to see it
acomplished or not, I am confident that sooner or later it
must be effected, because it is the only mode of classification
consistent with nature. It has been a subject of very general
complaint amongst those who, without having applied their
minds to botanical study, are interested in the cultivation of
plants, and consequently in the general outlines of botany,
and especially in its nomenclature, that it appears to be based
upon a very vague and changeable foundation; and the
frequent alteration of the names, with which they were
familiar, becomes a source of considerable annoyance. It
must however be recollected, that if the alteration of name
is consequent on the detection of an error in the preexisting
arrangement, the retention of that which had been used
before would be an irrational adherence to that which is false;
and that our unwillingness to accede to alterations which
arise necessarily from a corrected view of the subject, indicates
an indolence of mind that would obstruct the progress of
human knowledge. On the other hand it is no less evident,
that if alterations are made capriciously and not based upon a
correct and tangible foundation, discredit is thrown on the
science by the instability of the views of its professors, and
the students become disgusted on the very threshold of the
building, which they are invited to enter and contemplate.

PRELIMINARY TREATISE.

13

So difficult is it for man to penetrate the mysteries of nature,
that no science can be cultivated successfully, of which the
professors are tenacious of the sentiments they have adopted,
and not willing at all times frankly to reconsider the opinions
they have advanced, as different facts, or different views of
facts already ascertained, are presented to their notice. But
it is scarcely less important to establish some fundamental
principles by which the mode of forming and arranging the
inductions to be drawn from the facts which are ascertained,
may be divested of capricious uncertainty; or the information
that is obtained will become a mass of confusion, the more
palpable, from the futility of the attempts to disentangle it.

The first great division of the vegetable creation is between
those in which sexual propagation is manifest, and those in
which, as in funguses, the mode of increase is a concealed
mystery : they have therefore been called phanerogamous
and cryptogamous, that is, the first having manifest, the
second concealed, wedlock ; which seems to me preferable
to the later terms sexual and sexless, because the latter
word assumes a fact which cannot be substantiated. The
next division of phanerogamous plants is into monocotyle-
donous and dicotyledonous, the latter having two seedling
leaves, the former one, or rather that which stands in lieu
of cotyledons; and perhaps acotyledones without any, con-
sisting of the root-flowering plants like Rafflesia, which are
little understood. In the former the growth is said to be
made by successive additions to the outside, in the second,
to the centre of the plant, whence they are also called
Exogenous and Endogenous ; the former, at least in their
perfect formation, being recognised by a distinct deposit
of bark, wood, and pith. These are the great separations
of the vegetable creation, and it will not appear that any
kindred races are found indiscriminately in either division,
thougli there is a point of approximation between the mono-
cotyledonous and dicotyledonous plants. These lines are
therefore clear and substantial; but we cannot proceed a step
further without danger of error, by placing a less important
before an essential distinction. The Amaryllidaceie, which I
have undertaken to arrange, belong to the monocotyledonous
plants, which, being less numerous, can of course be more
easily classed. Although I have not sufficient knowledge
of all the orders they contain, to flatter myself that I can
make a perfect disposition of them, yet, as it is necessary for

14

PRELIMINARY TREATISE.

me to reform the errors which I find in the existing character
of Amaryllidese, I will try so to trace the peculiarities which
separate them from other vegetables, as to chalk out the
scheme by which that whole subdivision should be arranged.
It is incumbent on me, however, first to enumerate some at
least of the inaccuracies which I observe in Dr. Brown’s
character of Amaryllidese and the cognate orders, for the
purpose of shewing that they cannot stand as at present con-
stituted; their consideration being forced upon me by the
subject of my undertaking. It must however be premised,
that as Dr. Brown did not enumerate all the genera of which
the several orders are now reputed to consist, he is by no
means answerable for the disagreement of some of those
genera with the characters he has defined, though further
enquiry may have shewn that they cannot be excluded
without inconsistency. To commence with Amaryllidese:
1. Perianth in six divisions (sex-divisum) is incorrect, for it
excludes the tube which is very common in the order; and
if not meant to exclude the tube or annular junction of the
segments, it seems incorrectly worded, and at all events
nugatory, since all the cognate orders are equally six-cleft
at the apex. 2. Regular is inaccurate, if I rightly under-
stand the expression, that it is intended to indicate that the
sepals and petals are respectively uniform, witness the genus
Alstroemeria. 3. Stamina inserted at the base of the seg-
ments is inapplicable, for they are frequently inserted in the
tube, agreeing thereby with his character of Hemerocallidese,
and sometimes on the disk of the germen. 4. Anthers
anterior, cannot be assumed as the distinction of one order,
unless the same distinction is carried through the kindred
orders, yet in Melanthaceae we find anthers both anterior and
posterior. It is however an important feature and seems
there improperly confounded. 5. Style one is not correct;
the style is triple or tricomposite, often tripartible. I have
even seen it tripartite; but it is equally triple in all ttie cog-
nate orders. 6. Stigma 3-lobed is not correct; the stigma
in many genera is not 3-lobed, as it is stated to be, and the
angles are even obsolete, nor is it distinguishable from that
of Hemerocallideae, which he terms 3-lobed or simple.
7. Pericarp 3-valved or a berry, is not an accurate fact. In
Crinum it is not valved, yet it is not wliat Dr. Brown himself
calls bacca or berry, but something intermediate between
that and a capsule. 8. I demur to its ever being a berry,

PRELIMINARY TREATISE.

15

if Dr. Lindley’s definition of a berry is correct, that it is “ a
succulent fruit, the seeds of which lose their adhesion when
ripe and lie loose in a pulp;” and as a berry is usually such,
and cannot be two different things, I am satisfied that this
definition is true. That which Dr. Brown calls a berry in
Tamus, has never been properly described. It has a thin
(and usually bright coloured) outer skin; a soft juicy pulp
between it and the inner coat, which is a three-valved
dehiscent capsule, bearing the dissepiments on the valves,
the seeds lying in three distinct cells. This remarkable fruit
shews the futility of separating the berry-like pericarp of
some plants in the cognate orders, from the valved capsule.
9. Seeds with a shell neither black nor crustaceous is entirely
without foundation, and applies to only a small portion of
the order. 10. Embryo straight, with the radicle pointing to
the umbilicus seems to me incorrect. The radicle properly
points to the foramen; but I have seen the embryo in
Crinum at an early period very curved, and its ultimate
direction is quite vague. Neither in Crinum nor in Hyme-
nocallis does it usually issue near the umbilicus. There is
not in truth a single point that I can ascertain to separate
Amaryllidacese from Hemerocallideae, but the germen inferior
to the perianth and stamens in the former, and superior in
the latter. There is no true separation made between Aspho-
deleae and Hemerocallideae. Dr. Brown’s distinction of the
latter by seeds neither black nor crustaceous, and the former
by seeds with a shell black fragile and crustaceous, is cer-
tainly inaccurate, the seed of Hemerocallis itself being
black and fragile, and that of Albuca amongst Asphodeleae
soft and foliaceous. There is not in the given character of
Dioscoreae a single true point to separate it from Amaryllideae
except being dioecious, which is not even a sure generic
distinction, as may be exemplified by Vitis. Tameae are
separated by nothing but a supposed berry (which in fact is
not a berry) instead of a capsule, a variation which occurs
elsewhere both in Amaryllideae and Asphodeleae, and is
therefore no distinctive mark of an order. If any separation
of Tamus and Dioscoreae can be founded on the veins, it is
unnoticed by Dr. Brown. Smilaceae are admitted to differ,
from the portion of Asphodeleae which have berry-like fruit,
in little but the integuments of the seeds, and in the style
being often trifid: if the style is not always trifid, its being
so sometimes is no distinction, and the closely allied genera

l(i

PRELIMINARY TREATISE.

Bulbocodium and Colchicum, distinguished by a tripartite
and a tripartible style, shew that the actual separation of the
three styles, which are usually either conjoined or consoli-
dated, does not furnish a substantial feature to distinguish a
natural order; and in Dr. Brown’s own character of Melan-
thacese we find style trifid or tripartite. Posterior anthers
would have afforded a safe character for that order or rather
suborder, but Dr. Brown has included plants with anterior
anthers, which makes the given character nugatory, there
being no other feature really decisive. With respect to
difference in the integuments of the seeds, the striking
anomaly in that feature between the most closely united
genera of Amaryllideae (as for instance Pancratium and
Hymenocallis, Lencojum and Galanthus) shews that it is of
no certain weight in distinguishing natural orders of plants.
The beaked umbilicus, by which Dr. Brown characterises
Hypoxideae, is very dissimilar in Hypoxis and Curculigo, and
I understand that it is entirely wanting in that portion of
Curculigo which must be detached under the name Moli-
neria.

I cannot too strongly disclaim any intention to depreciate
the valuable labours of a gentleman whose life has been
devoted to the cultivation of a science, which to l^een

only the amusement of some leisure hours, and . . whose

botanical knowledge my own is comparatively light as air.
He is one of the most conspicuous ornaments of the study
he has pursued; but in the arrangement of these orders he
appears to have supposed the features, which he observed in
some genera, to be more prevalent than they prove to be;
and I have no doubt that he is fully aware that there are
imperfections in the characters framed many years ago,
which more intimate acquaintance with the subject matter
makes it necessary to remove, and I am far from flattering
myself that I can perfectly accomplish their reformation.

We should constantly bear in mind that there can be but
one real division, that is identity of kind, or, in other words,
generic distinction; the secret bar by which the Almighty
has made His works incapable of being blended with each
other, and confounded in their propagation. All divisions,
except that of generic identity are artificial, and rest on the
supposed agreement of different individuals in one or more
essential points of structure; and when any such point is
assumed for classification, the question must arise whether

PRELIMINARY TREATISE.

17

it does or does not separate individuals which agree in
another point of greater, or in more than one of equal,
apparent importance. The individuality of kinds in the
sexual creation rests on the possibility of their being propa-
gated by union, and is limited by a secret impediment, pro-
bably depending on the unsearchable adaptation of the
minutest vessels through which that union is to be effected ;
but the object of the limitation being to prevent the union of
things dissimilar, there will be outward appearances to
indicate the existing bar, and the skill of the botanist must
be exerted to distinguish which are the features that are
really symptomatic of the impediment, and which, from not
being so, are unessential. Botany is therefore a science of
conjecture in its fundamental office, the distinction of genera.
We assume from observation and analogy, that certain points
indicate an absolute diversity of kind; horticultural experi-
ments bring the accuracy of those assumptions to the test,
and either confirm or refute them, by proving the possibility
of sexual intermixture to be, or not to be, limited in accord-
ance with those points. The Almighty has allowed the
several genera of vegetables to disport themselves in nume-
rous forms of species and local or accidental varieties, which
are more qr less capable of intermixture according to their
constitute , and diversity, with various degrees of fecundity
and sterility in the united produce. It seems to me utter
waste of words to argue whether vegetables, if of one genus
or identical kind, are species or varieties; if they are found
different in a natural state and maintain their diversity when
removed to different localities, they are usually termed
species; if they return to the more general type when
removed, or if the difference is trifling, they are properly
termed local varieties, in contradistinction from cultivated
and accidental variations. But this distinction is of very
subordinate importance; it is a matter of convenience and
opinion, and not of fact. The discrimination of genera, or
identical kinds, by whatever term that identity shall be sig-
nified, is the true basis of botanical labours.

We are unable to ascertain when or how their subdi-
vision took place into the numerous forms which now adorn
the earth, and have been termed species and varieties.
The various races of mankind were certainly distinguished
at a very early period after the deluge, probably at the very
time of the miraculous dispersion, and separation of tongues ;

c

18

PRELIMINARY TREATISE.

and I apprehend that the distinction of vegetable species is
mainly referable to a period as early, and was consequent on
the dispersion of seeds into different climates by the operation
of the deluge. Looking to the multitude of forms which the
several kinds or genera have assumed, it becomes a matter
of considerable difficulty, and requires much discernment, to
ascertain which are the features that may be relied upon as
indicating an absolute and original diversity. The specific
differences in one kind may embrace a wide range ; the
actual difference between another kind and those most nearly
resembling it may be very small. The feature which is
symptomatic of individuality in one family of kindred plants,
may not be so in another ; such as the absence or presence
of a tube or any particular appendage. In forming generic
characters we are but seeking signs whereby to come at the
knowledge of a fact, namely, the individuality of a thing
described. For the purpose of assisting our view of nature,
we arrange them in groups, to which however no distinct
limits were assigned by the Creator; and, though wre are
trying to find out the ways of nature, our classifications, by
whatever name we may call them, are artificial, and if we
proceed beyond one step at a time, we must be liable to find
ourselves baffled by the reality. When I began, many years
ago, to write concerning vegetables, I had to combat an idea
that the Almighty had created each species of our botanical
catalogues as it now exists, and that plants being able to
breed together was the test of their identity as species. It
has since been ascertained beyond dispute, as I then antici-
pated, that in some genera all the species are capable of easy
intermixture ; and that even some, which botanists had
erroneously placed in different genera,* could produce a fer-
tile cross-breed. It cannot be asserted on the other hand that
such are merely instances of erroneous multiplication of
species by botanists, because they include almo^ every sort
of diversity which, when found among spontaneous vege-
tables, have generally been taken as specific characteristics,
and the facility of intermixture seems to depend less on the
botanical than the constitutional affinity of kindred species ;
so that either the whole existing machinery of species must
be upheld, or they must fall together. The opinion, which
I had at first to combat, is therefore falling to the ground by

* For instance, Crinum and Amaryllis.

PRELIMINARY TREATISE.

19

the irrefragable test of facts, which are daily becoming more
notorious ; and we cannot proceed a step without flounder-
ing into inconsistencies, unless we take the generic characters
to be the definitions of identity of kind, and in strict agree-
ment with that, they must, where it shall appear necessary,
be carefully reformed. Jussieu has stated, in the perspicuous
Latin Introduction to his History of Plants, that the founda-
tion of botanical science rests on the distinction of species,
which he defines to be a perennial succession of like indi-
viduals renewed by continual generations, and adds that
although they may be sometimes a little diversified by
climate, disease, or culture, the seminal produce will return
to the original type. He then considers the union of species
in genera to be a discretionary work for the assistance of the
memory, attempted at first without rule, and latterly con-
structed with more propriety by reference to the fructifica-
tion. Upon the same premises I should have come to the
same conclusion ; but that statement is contradicted by the
complete confusion of vegetable species which may be pro-
duced by reciprocal intermixture ; and horticultural expe-
riments have established, beyond the possibility of refutation,
that botanical species are only a higher and more permanent
class of varieties ; and that even the local and cultivated
varieties to which he alludes, if they do not come iu contact
Avith other varieties, instead of reverting, as he supposed, to
an original type, are renewed with considerable uniformity
in their generations. It is now certain that individuality
does not, as Jussieu had imagined, coincide with the species
of botanists, but with a higher and more comprehensive
grade, which seems to accord with such genera as are truly
defined, and would agree with all if properly reformed ; and
if he had known, what our later experiments have established,
his logical mind would certainly have assented to my pro-
position, which is in truth but the application of his own
doctrine to facts which have since come to light. The labours
of the many distinguished cultivators of the science have been
daily tending to place the genera on a correct footing, but it
will never be perfectly accomplished till the erroneous notion
of the original diversity of all vegetable species is thoroughly
discarded. If the original ' constructors of the system had
known all that is at present ascertained, they would perhaps
have called the botanical genera species, and the species
varieties ; but it does not appear to me advisable at present

c 2

20

PRELIMINARY TREATISE.

to make such an alteration, on account of the extreme incon-
venience of such an extensive change. We are not certain
that there is an exact analogy as to the diversification of
individuals in their generation, between animals and vege-
tables, the latter having branched more extensively into
different grades of variety, some of which can re-unite with
facility, while others seem so constitutionally altered, as to
be almost incapable of now mixing with the other grades ; so
that the arrangement and nomenclature is not objectionable,
if we only bear in mind that the botanic genera define the
individuality of kind. If any botanist should tell me, that
in framing a generic character he was not seeking signs
whereby to come at the knowledge of a fact, namely, the in-
dividuality of the thing described, I should answer, that if
such is not his object, whatever name he may assign to the
character, he has nothing in his science real or tangible;
that it is the vague offspring of variable opinion ; that he is
amusing himself with child’s play, and might as well build
castles with the sand. If, instead of proceeding by single
steps from the first point of general agreement to the last
important point of generic identity, we throw together a num-
ber of kinds, and include jthem under the character of an
order framed to cover their diversities, we are arranging
them in groups to which no such limits were assigned by the
Creator ; and our classifications, by whatever name we may
call them, are artificial ; and we must be liable to find our
labours, on further investigation, defeated by the reality. If
we divide all vegetables, according to their agreement, by
single points of difference, till we reach that which separates
them from their nearest kindred, every plant, when rightly
known, will have its proper place, and the whole system will
be built on truth, and not the offspring of opinion.

To make a perfect arrangement of the monocotyledonous
plants, to which division the plants of which I am about to
treat belong, it would be requisite that I should have a tho-
rough knowledge of all the orders it contains ; and even in
framing the character of Amaryllidaceae, I must be liable to
lapses, from the want of perfect acquaintance with those from
which the}^ are to be separated; and I must apologize for any
such errors, if they should occur, which will be easily recti-
fied by those who are more intimately acquainted with the
other portions of the previous divisions. If any botanist
should, therefore, perceive that my limitation does not effec-

PRELIMINARY TREATISE.

21

tually exclude some race, to which his attention has been
more closely directed, let him not blame me lor the want of
more accurate knowledge, which I do not pretend to have
acquired, but let him substitute the term which his own more
perfect investigation shall have ascertained to be more effec-
tual for their exclusion.

The monocotyledonous plants were divided by Jussieu into
three classes: — 1. Having the stamens inserted above the
germen or future seed-vessel. 2. Inserted round it. 3. In-
serted below it. This is a distinction of easy access ; but to
judge what rank it should hold of itself, we must see whe-
ther all the plants in each of those classes are more akin to
each other, than to any which are not included in it ; and it
is evident that Asphodelese, Hemerocallideae, and Melantha-
cese, are more nearly allied to Amaryllideae in another class,
than to Palms and Rushes in their own. It is therefore bad
on the face of it as a natural division, that is, a limitation
of natural groups ; it is as much an artificial division as the
number of stamens, nor is it more convenient than the
Linnaean arrangements, though it might be combined with
them. Another mode of classification suggested itself from
the structure of seeds, and Dr. Brown founded the charac-
ters of several orders mainly on the colour, hardness, and
separability of their external integument, and the direction
of the embryo. These characters were assumed on a partial
view of the seeds of some of the plants they were intended
to include, or at least of those which they are now held to
comprehend, and the result is, that they are inconsistent
with the fact. Hardness and inseparability, softness and
separability, of the seed-coat occur in plants so closely
allied, that although I consider it a proof of generic differ-
ence, their generic distinction has not yet been universally
admitted by botanists ; as, for instance, Pancratium and
Hymenocallis. I have already noticed the seeming vague-
ness of the radicle in some plants where Dr. Brown has
assumed its direction to be certain, and if correct, it is too
secret and concealed a feature for general application. The
hardness and softness of the integument is so uncertain, that
it is found soft in plants in which, according to Dr. Brown’s
character, it should be crustaceous and fragile. It is there-
fore evident, that the hardness and separability of the seed-
coat, and the direction of the embryo, furnish no satisfactory
characters for the definition of natural groups amongst mo-
nocotyledonous plants. It is certainly possible that the in-

22

PRELIMINARY TREATISE.

teguments of the seed, which are variable in one family,
may assume a more certain character in others. I do not
presume to give an opinion concerning those which I have
not examined, but we should admit with caution in one race
that which has utterly deceived us in another. The form of
the leaves might be expected to furnish some distinguishing
feature, but it is so variable, even in particular genera, that
little reliance can be placed upon it, and it must be admitted
with great circumspection. For instance, Professor Lindley
characterized Hypoxideae as having plicate leaves, whereas,
if the fact were absolutely true with respect to the particular
plants he meant to designate, we find plain and plicate
leaves so intermixed in the nearly related order of Iridese,
and even in the several genera thereof, that it cannot be
taken here as a natural distinguishing character, without the
most striking inconsistency. I wish it were true, that in the
leaves of exogenous or bicotyledonous plants, the veins are
concurrent and form a kind of network, and in endogenous
or monocotyledonous are parallel and cross-barred ; and that
scitamineous plants are distinguishable amongst the latter
by being feather-veined, that is, having oblique veins pro-
ceeding from the midrib. I have before me the leaf of a
Nepal Arum, closely allied to Arum Dracontium, in which the
veins are as concurrent as in the leaf of a lime tree ; and of
Caladium, in which they are confluent, though with less of net-
work ; of a Dioscorea, in which the venation is very like that
of a poplar ; of Costus among the Scitamineae, of which the
leaf is not more feather-veined than of Crinum spectabile,
and other petiolated wide-leaved Amaryllideee ; and on the
other hand, of a lofty Dracaena or Cordyline, which I raised
by seed from Norfolk Island (perhaps #C. stricta) amongst
Asphodeleae, which is completely feather-veined. Dr. Brown
describes parallel veins even as distinguishing an Austra-
lian species of Dioscorea from others of that genus, and some
such variability appears according to the representations of
different species of Peperoma. These peculiarities require
more careful inspection than they have yet received.

Another feature presents itself in the nature of the cap-
sule or fruit, some opening by valves, some indehiscent;
and in the integuments of the seed, some hard, some fragile,
others pulpy and soft. To shew that such peculiarities are

* The Botanical Magazine gives parallel veins to that plant, I believe in-
correctly.

PRELIMINARY TREATISE.

23'

good for nothing as characters of natural groups, though
valid in generic separations, it is sufficient to name Haeman-
thus with a pulpy berry-like fruit, and Buphane with a
valved capsule, yet so closely allied that they have been
heretofore considered one genus, and are not distinguished
by any other striking feature. The shape of the seed-pod
cannot be relied upon with certainty, even as a generic dis-
tinction ; and Linnaeus, when he divided tetradynamia into
siliculosa and siliquosa made a very unsatisfactory separa-
tion. The simple fact, whether the seed is naked or covered,
furnishes, perhaps, a limitation of more value, if any have
it really naked, but it does not help us to subdivide mono-
cotyledonous plants, because they all have it covered, except
where it is exposed by the early disruption of the pericarp.
It is something to have cleared the way of features which,
though they have been used, are evidently not available as
the distinctions of natural groups ; and although these con-
siderations, when duly appreciated, must shew the necessity
of superseding a great part of the present arrangement, they
will help to lead us to something more substantial. I will
suggest, as well as the limited information 1 possess will
enable me, the features that appear to me to separate the
monocotyledonous plants into natural groups, calling to mind
that natural groups are properly such as will strike the un-
scientific eye as having some general resemblance, without
searching for a minute point of agreement. I have to re-
gret, that in order to ascertain any one point concerning any
class of vegetables, it is necessary to examine afresh almost
every plant of which it consists, from the silence or lax ex-
pressions of botanists, and the absolute falsehood of many
of our engravings in material points ; and, with respect to
the venation of leaves in particular plants, our books are
generally silent, and the engravings frequently give an un-
true representation, so that I cannot obtain the information
I desire.

The palms, aroid, and piperaceous plants, form a con-
spicuous division amongst the monocotyledonous plants,
and the spadix seems to me the great point of agreement.
Spadix originally meant the inflorescence of the palm, that
is to say, flowers without the semblance of a corolla, closely
set round a stalk which has an involucre below. There is a
defective spadix, which w'ants the involucre, in piperaceous
plants and bidlrushes ; and there may, perhaps, be a dif-

24

PRELIMINARY TREATISE.

ference between a fibrous or woody, and a succulent, spadix.
The division opposed to Spadiceous, consists of those with
petaloid flowers which are not closely set round a stalk with
an involucre at the base ; they have not the spadix, and may
therefore be called Exspadiceous. In all divisions of vegeta-
bles, individuals will be found that depart a little from the
limits assigned, because the jCreator has not strictly drawn
those limits of classification, but has softened down, as it
were, the edge of difference by an easy transition. The
groups, however, are not the less real on that account, though
some individuals will exceed the boundaries; so we find
Orontium, Tupistra, Aspidistra, and Tacca, forming a sort of
limbo between the spadiceous and scapaceous plants, and,
on the other hand, Pontederia and some others, together
with the Scitamineous plants, advancing from the corolliform
towards the habit of the spadiceous plants. They may
therefore be well placed in an intermediate division as Subspa-
diceous, that is, having a tendency to the form of a true spadix.

The spadiceous plants, I believe, may be divided into,

1. Involucrate or true, including Palms and Aroid plants.

2. Nudae, Naked, including Piperaceae and Typhaceae. The
involucrate into Ligneous, covering the palms, and Succulent,
designating the Arums, &c., but I am not certain of the truth
of the distinction of fibrous and succulent here, and do not
pretend to make a perfect subordinate arrangement. If suc-
culency does not universally distinguish the Aroidae, the
petaloid involucre will furnish one feature of separation, the
mode of fructification another, and probably the veins of the
leaves another. The subspadiceous will divide, I believe,
into 1. Subcorolliform, and 2. Corolliform. The first covers
Orontium and those which have no involucre amongst Oron-
tiaceae, the Acoroid and fluvial plants ; the second will be di-
visible into monoperianthine and biperianthine, the former
covering the aquatic, the latter the scitamineous plants.

The corolliform are divisible into gynandrous, of which
the stamens are consolidated with the style, covering the
Orchidaceous plants and their kindred, and agynandrous, in
which they are not so consolidated. The agynandrous may
be subdivided into tripetaloid having the semblance of three
petals and hexapetaloid of six. The tripetaloid into mono-
gynous having one style, which are Bromeliaceae and Comme-
linaceae, and pleiogynous having more styles, which are Buto-
maceae and Alismaceae. The hexapetaloid into triandrous with

PRELIMINARY TREATISE.

25

three stamens, and hexandrous with six. The triandrous into
epigynous having the germen or ovary underneath, (viz. Irida-
ceae, Burmanniacese and Haemodoraceae limited,) and hypogy-
nous having the germen above, that is a portion of the present
order Haemodoraceae, which I propose to call Wachendorfaceae.
The hexandrous divide in the same manner into epigynous
and hexandrous. That hexandrous epigynous division is the
immutable line by which I limit the character of the order
Amaryllidaceae, and the corresponding hypogynous hexan-
drous division forms the boundary of Liliaceae. Both are
capable of further subdivisions into suborders and sections,
and whether those subdivisions shall be exalted or not into
the station of orders, is a question of discretion, and not of
fact. If it should appear that any invariable feature, which
distinguishes the subdivisions, affords a line of demarcation
more conspicuous than those which I have taken of hexan-
drous or epigynous, it should be inserted in the fittest situa-
tion to characterize and unite the plants it may include. For
instance, if on further acquaintance with the several species,
we should ascertain that any certain venation leaves or mode
of inflorescence distinguishes Dioscoreae, Smilacese, and Rox-
burghia from Bomarea and Asparagus more decidedly than
the features which I have chosen associate them, that feature
must constitute a division prior to either epigynous or hexan-
drous, for the purpose of detaching them ; but the best judg-
ment that I can form, in the want of more precise informa-
tion, is that such features are not to be depended upon for the
separation of high grades, and are more fit to distinguish the
suborders and sections.

The epigynous portion have all one triple style, which in
some genera is tripartite and even occasionally tripartite.
This division includes, with the plants usually called Amaryl-
lidese, the hexandrous. part of Haemodoraceae, Hypoxideae
closely connected with them, Fourcroya and Agave, Dioscoreae,
Tameae and others of which the present position cannot
possibly be maintained. This epigynous division, consti-
tuting the order Amaryllidaceae, separates itself into 1. Ra-
mosae, 2. Caulescentes, 3. Scapaceae, or branching, caule-
scent, and scapaceous, meaning by a scape a succulent stalk
supporting a spathed flower or umbel, and inarticulate below
the spathe. Of the branching, there are but three known
genera; the caulescent are, I believe, all schistandrous, i. e.
having their anthers, which are bilocular or two-celled, open-

26

PRELIMINARY TREATISE.

ing by a vertical slit from top to bottom. They seem divisi-
ble however into Operculous and Nonoperculous. By oper-
culous I mean having the base of the style which is enlarged,
persistent after the decay of the rest, and forming a hollow
top to the ovary, more or less prominent, and becoming part
of the seed-vessel. The operculous may be called Hypoxideae,
and divided into 1. Hypoxidiform, 2. Lanariaeform, covering
the hexandrous part of the present Haetnodoraceae, 3.
Alstroemeriaeform ; or those portions may form three subor-
ders. The nonoperculous may be divided into Dioscoreae
and Agaveae, the latter into Ixiaeform and Agaviform. The
scapaceous are divisible intoSchistandrous, containing the true
Amaryllideae and Narcisseae, and Porandrous of which the an-
thers open only by the upper part, which I call Galantheae.

The hypogynous division consists of Liliaceae, that is
Asphodeleae, Hemerocallideae, Liliaceae or Tulipaceae accord-
ing to the separations of different writers, Smilaceae and Me-
lanthaceae. A portion of Melanthaceae is clearly separable
by posterior anthers, those with anterior anthers being
improperly joined with them. The separation of He-
merocallideae cannot be possibly maintained, the main dis-
tinction of the seminal integument being contrary to the fact,
and the difference of a longer tube quite trivial for the cha-
racter of an order. Smilaceae, according to Dr. Brown’s ad-
mission, are distinguishable from those which have been
called berried amongst Asphodeleae, only by the stjde being
oftener tripartible, which, not being absolute, is no distinction.
This hypogynous division, Liliaceae, maybe properly divided
into suborders as the Amaryllidaceae, and perhaps in this man-
ner, Branching, Caulescent, and Scapaceous. Caulescent
erect, Asphodeleae. Caulescent twining, Smilaceae, inclu-
ding asparagus medeola, &c. Scapaceous ; anthers anterior,
Allieae ; anthers posterior, Melanthieae; the termination aceae
being reserved for the higher division. I can lay no great
stress, except as a generic distinction, on the difference of
seminal integument, when I look to the seeds of Leucojum
and Galanthus, Pancratium and Hymenocallis, nor of cap-
sule valved or valveless, which latter has been erroneously
called a berry, when I look to Haemanthus and Buphane, nor
of erect and twining stem when I look to Bomarea and Al-
stroemeria, all respectively confounded heretofore in genus ;
nor on the dioecious character, when I see one species of vitis
dioecious and another not so ; but they are features which

PRELIMINARY TREATISE.

27

may be available in inferior subdivisions though very little
reliance can be placed on them. The fundamental principles
on which I build may be thus briefly recapitulated. 1. There
is a general point of agreement between all vegetables,
namely, the character which separates them from the other
portions of the creation. 2. As individual vegetables differ
in a variety of points, those points taken singly will separate
the mass, step by step, from the point of general agreement
to that of individuality. 3. The only mode of avoiding con-
fusion is to work downwards from the point of agreement to
that of individuality. 4. Great attention is requisite in de-
ciding which points of separation should have the priority.

5. That point should have the priority which shall be found
not to separate (or, if that be impracticable, to separate the
fewest) individuals connected either by a feature apparently
more important, or by a prevailing weight of other features.

6. Each point must be taken singly, and contrasted with that
which differs from itself, and not with some other feature con-
sistent with it. 7. No point of difficult investigation should
be put foremost, if one of more easy access is found to coincide
with it. 8. The differences between the absence and presence
of a feature being in many cases not absolutely defined, but
intermediate appearances being found to intrude, there are
but two ways of dealing with them, either to place them in
an intermediate division, or in that to which they appear most
nearly allied, stating them to differ through superfluity or
defect; for instance, Commelina is an hexandrous plant, trian-
drous by defect ; Azalea is a decandrous Rhododendron, trian-
drous by defect, and some species of Gethyllis and Vellosia
are pleiandrous or more-stamened by superfluity. 9. Inter-
mediate divisions, if adopted, are further separable by various
modifications. 10. The Creator not having made all vege-
tables to differ from each other by any fixed number of points,
but in the process of separation by distinct features, some
being detached from the whole mass at an early and others at
a later step, any system of arrangement that shall group them
by an equal and fixed number of subdivisions or steps from
the point of universal agreement, must be inconsistent with
nature and fundamentally untrue. Therefore, if it be thought
advisable for the purpose of assisting the memory to adopt
any given number of principal divisions, such as class, alli-
ance, group, order, suborder, and section, it can only be done
with truth by admitting in each of those divisions an unli-
mited number of grades, and the determination of the

28

PRELIMINARY TREATISE.

number of grades at which a new division shall com-
mence will he a matter of discretion and not of fact, and in
truth of very little importance. 11. That the distinction of
groups is of high value in shewing the harmony of the crea-
tion, hut the basis of botanical labours is the discrimination
of individual kinds. 12. That whatever names we may give
to subdivisions, if one subdivision designates individuality,
another cannot designate it also. 13. That the older botanists
thought to designate individuality by the subdivision named
species, and Jussieu said that in the discrimination of species
the fundamental labour of the botanist consisted. 14. That
it has since appeared by the further investigation of species,
and by the experiments of horticulture, that in some very ex-
tensive* genera the species are diversities of one individual,
capable of breeding together and frequently of producing fer-
tile offspring, the fertility seeming to depend more on the con-
stitutional, than the closer botanical, affinity of the parents ;
and that the species so proved to be one have the several
diversities by which botanists have usually distinguished
species from each other. 15. It follows, therefore, that in-
dividuality does not reside in the division named species by
botanists, but in a higher division usually named genera.
16. That the individuality of other species, distinguished by
like features, must fall with those whose individuality has
been disproved. 17. That as individuality is proved to re-
side in the division usually called genera, it must reside there
uniformly. 18. That there are but two ways of rectifying
the great error of older botanists, either to confine the name
genus to the point of individuality, or to change the whole
nomenclature, and call the entire mass of genera (requiring,
as they do, in either case, reformation in many instances),
species, and the mass of species permanent varieties. 19.
That the latter mode of proceeding would produce infinite
disturbance, whereas the former requires nothing but the
clear understanding and declaration of a fact, that indivi-
duality in the botanical divisions resides in the genus. 20.
That there is no essential difference between species and per-

* Crinum, Hippeastruin, Gladiolus, Rosa, Pelargonium, Calceolaria, &c. If
any one shall assert that Hippeastrum and Habranthus are not separate genera,
but two species of one genus, and their respective species only varieties, the as-
sertion is merely a different application of terms, but he who asserts it must con-
cede on the other hand by analogy, that there is in that case only one species of
rose, pelargonium, &c and that their reputed species are only varieties.

PRELIMINARY TREATISE.

29

manent varieties ; and that in some genera even cultivated
varieties preserve themselves distinct in their generations
more easily than natural species in others when approximated.
2 1 . That in some genera intermediate diversities from different
localities so confound the limits of species, that it is waste of
words to argue whether a plant is a species or a permanent
local variety; though it is a matter of convenience to assign
some specific features, and to arrange the subordinate forms
as varieties, but that the distinction of closely allied speci6s
is of very trivial importance ; botanical species being merely
the long established varieties of genus or kind, perhaps con-
sequent on the dispersion of seeds by the deluge and refera-
ble to the unknown period when the different races of man-
kind assumed their peculiar appearance. 22. That the dis-
crimination of an immutable generic feature is the discrimi-
nation of a fact, the limits of species and varieties are the
offspring of human opinion. 23. That a mixed offspring
from the mutual contact of any two plants, if not absolutely
conclusive as to the identity of their origin, gives strong rea-
son to presume it, and to doubt the validity of the features by
which they may have been distinguished. 24. That the
fecundity of such mixed offspring establishes the identity of
kind.

In framing the character of an order or genus, for the
sake of perspicuity and correctness no feature should pro-
perly be included but such as are necessary to distinguish
the order or genus from all others, and from which a depar-
ture would decidedly exclude a plant that should contravene
it ; and all important points that appear to coincide with the
character should be added as observations subsidiary to it ;
nor will the system be perfected till each genus is separated
from its nearest kin by a single primary and immutable
feature, its subordinate peculiarities being subjoined. Per-
fection of the system cannot, however, be effected without a
complete knowledge of the peculiarities of all the indivi-
duals which each order contains, and our ignorance of im-
portant points, even concerning those which are considered
as having been described, presents an obstacle which can
only be surmounted by time and assiduity. I doubt the
correctness of any generic character which rests upon cumu-
lative features, of which no one has sufficient weight by
itself to distinguish and uphold it ; and I think that no
botanist has sufficiently considered what are the features

30

PRELIMINARY TREATISE.

that can be most depended upon in each order as indicating
a diversity of kind. My own thoughts have been earnestly
directed to that inquiry, but I feel my decision obstructed
not only by the want of perfect knowledge of the natural
order which I have attempted to arrange, but of the diversi-
ties of the whole vegetable kingdom. I am inclined to
believe that every true generic character is manifested by,
and therefore should be founded on, some difference of struc-
ture in the male or female constituents of the plant, the male
being the perianth and stamens, the female the germen,
ovary, style, and stigma, and of course the mature fruit.
The difficulty is to ascertain which are substantial differences
and which are only modifications of structure, and how far
the points of structure which are invariable in the indivi-
duals of one race may be variable in another. Concerning
this, which is the basis of botany, the opinion of its pro-
fessors is fluctuating and undecided. Of late years their
attention has been directed with great propriety to the fruit,
but it has not been yet established in what an absolute diversity
of the seed consists. The bent of my own opinion is, that
the nearer to the base of either the male or female consti-
tuent any diversity manifests itself, the greater is its impor-
tance ; and with great diffidence in making any suggestion
on the subject, which I feel that I have not sufficient infor-
mation to substantiate, and which, if it should prove to be
upon the whole correct, will be assuredly subject to modifi-
cations or exceptions, I am nevertheless desirous of drawing
the attention of botanists to its consideration. I look upon
the perianth as superadded to the stamens, and therefore
further from the base than the anthers. Under this view it
would be apparent that the absolute disseverment of the base
of the segments of the perianth (a point so entirely overlooked
in botanical characters, that we cannot ascertain from the
terms used whether the perianth is quite cleft or deeply
cleft) is of more importance in framing a generic character
than any difference in the form or position of its upper part ;
that the absence or presence of a tube is of more importance
than the form of the limb, the stamen being of greater con-
sequence than the perianth, to which the stamen is the inte-
rior and proper base. Under this view it would be apparent
that the stigma is the least important part of the female con-
stituent. The point concerning which I feel the greatest
difficulty is the degree of weight which is to be assigned to a

PRELIMINARY TREATISE.

31

difference in the attachment of the anther, and exactly what
variability of anther is consistent and what is inconsistent
with identity of kind, and I believe it to be the most difficult
point that botanists have to investigate. The insertion of
the filaments appears to be of the highest importance, their
more or less continued adhesion to the tube, when not abso-
lutely inserted, to be of very little weight. Whatever may
be the truth in these respects, it is evident that a distinct view
of the relative importance of the several differences that exist
between vegetables, in ascertaining their generic identity and
in the disposition of natural groups, is the great desideratum
to give stability and consistence to the labours of botanists.

On minute examination of the riches of the several her-
bariums which have been opened and entrusted to me by the
liberality of their possessors, the difficulty of deciding whe-
ther or not to consider individual specimens as local varieties
of one species, is greater than when a more perfect know-
ledge of their peculiarities is obtained from the sight of living
plants. The mode in which I propose to surmount this
difficulty is very simple, and I think its advantage will be
felt, and that it will be universally adopted. I suggest that
in every botanical work varieties should be specified as be-
longing to one of four characters, viz. 1. local; 2. acci-
dental; 3. cultivated ; 4. hybrid. 1. Varietas, i. e. var. loci;
2. var. fortuita ; 3. var. hortensis; 4. var. hybrida. That
the system of marking varieties by letters of the Greek alpha-
bet (which the unlearned cannot read and pronounce, and
when spoken give the very unsatisfactory names alpha, beta,
gamma, &c.) should be abandoned as unmeaning and incon-
venient, and each local permanent variety distinguished by
a Latin name not found in that genus or in any other genus
so nearly allied as to make their union not quite impossible,
and that such name should agree grammatically with the
generic name; and, further, that closely allied species should
be placed in groups headed by their most striking features
of agreement. It will then appear that it makes little diffe-
rence whether they are looked upon as varieties of one or as
several allied species, as their affinity will be equally shown
and the peculiar name would stand the same in either case.
Thus I find a race of Crinum, occupying the coast of South
America, distinguished by a short column, a stoloniferousbulb,
diverging leaves with a rough margin, flowers from 4 to 8,
oftener 6, and cinerascent anthers, which were the features of

32

PRELIMINARY TREATISE.

C. erubescens. It varies in different localities very much,
and I call the varieties, 1. rubrilimbum, &c. 8. octoflorum.
If any person thinks them more distinct than I do, he puts
forward the name rubrilimbum or octoflorum, which at all
times may be used alone for brevity, but the plant equally
stands in its place with its affinities made evident. We have
an American race with an ovate stoloniferous bulb, more
erect leaves, and four flowers (rarely five), of which the first
known was Americanum. This is followed by Commelini,
strictum, and Loddigesi, more distinguished from each
other than the varieties of erubescens, and therefore placed
as species ; but if intermediate plants shall be found to bind
the connexion closer, they will stand with the same names
as varieties 1, 2, 3, &c. following Americanum, the var.
princeps, or first of the family. A very important objection
to the naming permanent local varieties by either figures or
letters, is that it forces their arrangement to be made
according to priority of notice and not according to affinity,
or if those of later introduction are inserted in their proper
places, the distinguishing marks and titles will be ever
varying. I therefore entreat all botanists who may have
occasion to name any new species in the genera I am about
to describe, not to give to it anv name which I shall have
affixed to a variety in the same or a closely allied genus.
The distinction of local varieties by their place of abode, if
surely ascertained, has great advantages, and the objection
that they may, perhaps, be found in more situations is of
no importance. I take this opportunity of cautioning all
botanists against the introduction of any new species of
Crinum, Hippeastrum, & c. without certain assurance of their
being of wild growth in some specific place, because the
hybrid varieties which have been raised by myself and others
are innumerable, and from the multitude of species collected
together now in the gardens at Calcutta, where I sent most
of the hybrid as well as Occidental and African species,
mules have lately originated there by accidental intermix-
ture, and Dr. Carey found that he could no longer depend
on the seed ripened in his garden. The accidental and cul-
tivated are of a different character from the local varieties,
and should have no Latin name, but be designated thus, —
var. (fort, flore albo.) accid. fl. white, var. gard.. fl. double,

var. hybrid by . Very great confusion is produced by

the nurserymen giving a Latin name to every garden seed-

PRELIMINARY TREATISE.

33

ling, and men of science should set their faces decidedly
against the practice, which Mr. De Candolle very inauspi-
ciously sanctioned with respect to hybrid plants. Where
garden varieties are much multiplied florist names ought to
be used, as with hyacinths, tulips, &c. Hybrid plants which
are found of spontaneous growth in the wild abodes of their
parents should rank as species marked Hyb. Sp. or sponta-
neous hybrid ; those of complicated or uncertain intermix-
tures in our gardens should be marked as variety garden
hybrid. It would very much tend to preclude confusion if
ail substantive genitive cases were abandoned to cultivators
for the distinction of their varieties, and the names of all
species and permanent local varieties confined to adjectives.
With this view I venture to alter all the proper names
adopted in this order to an adjective form, writing Calda-
siana for Caldasi ; and I earnestly press the convenience of
this arrangement on the consideration of botanists, by which
it may be understood at once that B. Caldasiana must be a
species, or permanent local variety, and that B. Caldasi
would designate a seminal or hybrid variety ; and as it will
be vain to urge nurserymen not to dignify their productions
with Latin names, I wish to request them to confine them-
selves to genitive cases of proper names, names of romance
or heathen deities, or of substances, as flammse instead of
flammeus, eboris instead of eburneus ; and, if the botanical
editors of popular periodical works will attend to this sug-
gestion, we shall get rid of the overwhelming confusion which
garden productions are creating. At present, in our best
botanical catalogues, every seedling, Camellia Japonica, or
Hippeastrum, is dignified with a Latin adjective name, and
the endless garden intermixtures of Calceolarias are named
like the natives of South America, very much to the disad-
vantage of science. Cultivators will have an ample fund of
names if all genitives are given up to them, and the change
of the few genitives that have been used in the scientific
nomenclature into the form of an adjective will produce no
inconvenience.

The kindred genera, also, of each order should be so
arranged in different groups, having certain common fea-
tures, that, if it should hereafter be found that two genera,
now distinguished, are in fact identical, they would remain
as sections of the leading genus without any disturbance of
the arrangement or nomenclature. For instance, Phycella,

D

34

PRELIMINARY TREATISE.

Habranthus, and Zephyranthes, very dissimilar at their
extreme points, touch very closely at the intermediate points
of approach, hut being properly grouped, if it should here-
after appear (which I do not think it ever will) that a Phy-
cella can breed with a Zephyranthes, they will stand as
three subgenera or sections of the leading genus without
occasion for any further alteration than that of the word
genus, subgenus, or section. Botanists are too apt to con-
sider that they have discharged their part when they have
carefully defined a plant; any accurate man can do this;
but only half the duty of a botanist has been discharged
before he has ascertained to what other plants his specimen
is most nearly related, and by what points it is separated
from them ; and for that information we usually either look
in vain, or find it stated with little consideration and cor-
rectness. I wish to see something more like sound system
and regularity in our proceedings ; and I protest against the
custom of placing plants provisionally in any genus with
which they are known not to accord. It would never have
been done if any clear principles had been established for
the construction of generic characters, and such cannot be
adopted till it is clearly understood what a genus is.

I hope that nothing in these pages will be offensive to
any of the distinguished cultivators of the science of botany,
from whose steps I have found it necessary to depart. It
has been my anxious wish to pursue, with careful delibera-
tion, the path of accurate inquiry, without using any polemic
expression, and without depreciating the writings or setting
at nought the opinion of any individual. If, by misfortune,
an abrupt expression shall have been any where incautiously
admitted, I pray that it may be considered as an oversight,
which, if observed, would have been sedulously rectified.

W. H. 1835.

35

POSTSCRIPT.

The foregoing preliminary treatise was prepared for the
press last year, but was delayed for the purpose of instituting
a particular investigation of the Narcissi and someotherplants
of which specimens could not be then obtained, and it would
have been more satisfactory to myself to have delayed this
work longer. I have, however, now had the advantage of look-
ing into Dr. Lindley’s second edition of a Natural System of
Botany, a work of exceeding great value, which I hope will
reach many editions, and be improved till it shall ultimately
exhibit that distinct view of the vegetable creation which in
the foregoing pages I was desirous of exciting him to pro-
duce. I observe with great pleasure at the commencement
of the volume what is there termed an artificial analysis of
the orders, in which the faulty system of a limited number
of subdivisions is abandoned, and the successive grades of
difference are set forth in the general manner I have sug-
gested. It is termed an artificial analysis, as I conclude,
because its author is conscious that it does not arrange the
orders according to his own view of their affinities. He may
be, however, assured that, if such be the case, it is not because
such an analysis is necessarily repugnant to the most impor-
tant affinities of nature, but because the principal points have
either not been duly set forth, or have not obtained the
precedence they deserved. It is impossible that any man
should, at the first, bring such an arrangement to perfection.
It is much to have begun the work, and to have produced
an analysis which is of infinite value, both from the clear
view which it presents and the facilities it offers for effecting
a more perfect arrangement. What remains is to examine
carefully how the points assumed and arranged therein sepa-
rate orders that ought to be approximated, and to see how
either by arranging those points in a different order of prece-
dence, or by the assumption of other points, the analysis can
be made to exhibit a more natural view of the concatenation

n 2

POSTSCRIPT.

3G

of vegetables. The work will not be perfected till the arrange-
ment made in such an analysis shall be the proper arrange-
ment for the body of the work, to which I am sorry to find,
though with some variation, that my former objections apply.
There are alterations, which make some of the particulars of
my foregoing observations not exactly applicable to the im-
proved edition of this work ; they were, however, by no
means intended as criticisms, but as exemplifications of the
mode of classification to which I objected. But, as the
arrangement is in some respects different, it is incumbent
upon me to examine very carefully what it contains concern-
ing the Amaryllidaceous plants, and, indeed, the Monoeoty-
ledonous plants in general.

I find them divided at the outset into six groups : —
1. Epigynous; 2. gynandrous ; 3. hypogynous; 4. reticu-
lately veined; 5. spadiceous ; 6. glumaceous. In the first
place, epigynous means simply having the ovary below the
stamens ; but the character given as an interpretation of it
includes features of a different description, and the limita-
tion, “ ovary inferior or if superior then the leaves either
scurfy or equitant,” renders the division of Group 1 ovary
inferior, and Group 3 ovary superior, nugatory. If the cir-
cumstance of the leaves being scurfy or equitant is of primary
importance, it ought to be placed before the feature which is
thus made subordinate to it : but the fact is, at least as I
conjecture, that Dr. Lindley does not think it of primary
importance, and, if so, it ought not to supersede and nullify
the division he has made. The second group gynandrous
being a portion of the epigynous group ought not to be con-
trasted with it, but made a subordinate division. The third
group is characterized as having coloured ternary flowers,
which has nothing to do with its title, ovary superior. The
fourth makes a division according to the veins of the leaves,
which l fear is not perfectly correct, and is loaded with the
addition of very different features which are not peculiar to
itself. The fifth and sixth are spadiceous and glumaceous,
but I observe the spadiceous group does not contain all the
palms, the true palms, though spadiceous, being placed in
another group. The division, therefore, thus proposed of
monocotyledonous plants into six groups seems liable to all
the objections I felt to the adoption of the former arrange-
ment. The point which principally concerns the immediate
object of my labours is the order Hsemodoracese, which

POSTSCRIPT.

37

Dr. Lindley proposes to preserve, and I have given the most
unprejudiced consideration to his remarks on that subject.
It is for the purpose of bringing Haemodoraceae and Brome-
liaceae into the group of plants with the ovary inferior, that
the limitation of leaves equitant or scurfy has been devised
to stand in lieu of the inferior ovary in plants which have
the ovary superior, viz. equitant leaves to bring in the Has-
modoraceae, and scurfy to include Bromeliaceae. There is,
however, no connexion between those two qualities of the
leaf which are severally assumed, nor have either of them
any relation to the quality of which they are made to com-
pensate the absence ; nor are they qualities generally be-
longing to the plants with which an association is to be
effected through their means. Dr. Lindley says he is not
aware that scurfy and equitant leaves are found amongst
other plants with superior ovary. Undoubtedly they cannot
be so found, if he takes out all that have them, but we have
no other scurfy-leaved plants, and only one order with equi-
tant leaves, in the group to which he has transferred them.
Scurfy leaves or equitant leaves may be thought sufficient
grounds for detaching plants from those which have them
not, but not for superseding a more important point of dif-
ference. If the leaves standing edge to edge is of more con-
sequence than the position of the ovary, we ought to have a
primary division of leaves equitant and not equitant, but
scurf, I think, could never be taken to characterize a high
division. There seems to be no other feature than equitant
leaves to separate the hexandrous epigynous Haemodoraceae
from Amaryllidaceae, for the equitation of the sepals is almost
obsolete in Alstroemeria, and I doubt its being absolutely
wanting in Haemodoraceae ; the woolly surface is a very
weak feature, and their limb is not in fact smaller when
compared with the tube than it is in some Cyrtauthi and
Cooperia, supposing such a feature to be more important than
I consider it to be. Taking, therefore, leaves equitant to be
the true distinction by which it is proposed to hold together
the discordant order Haemodoraceae, a very important
question arises whether leaves equitant or standing edgeways
with respect to each other, like those of Iris, instead of face
to face, is a feature which ought to have general priority
over the position of the ovary and the number of the
stamens ? The effect would be to sever Iridaceae, which are
defined as having leaves equitant, together with Haemodo-

38

POSTSCRIPT.

race®, from the hexapetaloid group at an early stage ; but
that separation having been made, we must, in order to pre-
serve any consistency, either leave all those whose leaves are
not equitant in one order, or we must divide the equitant
upon the same principles as those with leaves not equitant;
viz. into triandrous and hexandrous, epigynous and hypo-
gynous, the result of which would be that Hsemodoraceae
would be equally broken up in the manner I had proposed
to do, the only difference being that the hexandrous portion
would be placed in closer affinity with Iridaceae. It is, I
think, evident that it would not be desirable to introduce this
feature for the mere purpose of making that discordant asso-
ciation, unless it should appear to work well if applied as a
previous limitation to all other monocotyledonous plants,
which Dr. Lindley does not appear to have considered. But
whatever might be the effect, I believe it is, like other pecu-
liarities of leaves, too variable to be taken as a chief charac-
teristic, for it is untrue with respect to Iridaceae, the only
other order which is said to have leaves equitant. The whole
of the common bulbous Irises or Moraeas have channelled
leaves fronting each other like the Amaryllideae with a
cylindrical base. They can scarcely be called equitant in
Crocus. It seems, like the reticulation of the veins in
Dioscoreae, which is not forthcoming in D. lucida, or like
the plication of leaves, too uncertain to uphold an order
or perhaps even a suborder, though it may be good as a
generic characteristic. It appears to me, therefore, that it
cannot be taken as a previous feature of separation, and that
the hexandrous epigynous portion of Haemodoraceae must
unite with the Amaryllidaceous plants in the first instance.
I have distinguished them in such manner as I find con-
sistent with the truth as far as I can ascertain it ; and
whether that portion so distinguished shall be considered as an
order, or a suborder, or a section, is a matter of discretion
concerning which I feel perfectly indifferent ; but if it be
made an order, other suborders of Amaryllidaceae should be
equally exalted, which I think will not be found convenient;
indeed, I understand the use of the termination acece to have
been adopted for the express purpose of including the sub-
orders, and preventing an unnecessary multiplication of
independent orders. The fact is, that all the bulbous Ama-
ryllidaceae have equitant leaves, though cylindrical at the
base, instead of being compressed. The definition of the

POSTSCRIPT.

39

term by Jussieu is one leaf doubled enclosing another; every
Amaryllidaceous leaf encloses another with its base, and
some vaginate very high, but such base is usually cylindrical.
When it is flattened, as if by compression, the true equitant
form is exhibited; but the difference is the compressed habit,
not the structure. I observe that Dr. Lindley calls leaves
flat in contrast with equitant, which is an unsatisfactory
term, since most of those so called flat are deeply chan-
nelled, and the others often really flat ; but there is no true
contrast, and therefore an applicable expression could not be
found. If all leaves that enclose another with their base are
properly equitant, the distinction of such would be, base
compressed and base cylindrical, lamina edgeways and
lamina confronted. It will even be found that some Iri-
daceous plants, as Homeria, have a compressed base with
confronted leaves, bearing an intermediate form.

Dr. Lindley has associated Xerophyta, Vellosia, and
Barbacenia, with Bromeliaceae. It is impossible that a
person embracing such a very wide field as an arrange-
ment of the whole vegetable system, should have at the
first under his view all the considerations which present
themselves readily to one who is investigating a parti-
cular branch of that system, and I feel quite confident that
my worthy friend on mature consideration will perceive that
this cannot do. Bromeliaceae had two definite and impor-
tant features, ovary superior, and perianth tripetaloid,
that is, having three conspicuous petals, and three short
sepals like a calyx. Dr. Lindley, detaching Vellosia and
Barbacenia from Hypoxideae, and uniting Xerophyta with
them, brings the order Bromeliaceae out of its proper place
to unite with them, and in so doing overturns its character.
Its ovary is superior, and it is brought into the class with
inferior ovary on account of its mealy leaves, but not one of
those three genera to which it is brought has mealy leaves ;
and it cannot be understood how that feature is to stand in
lieu of the other. It thus becomes necessary to engraft on the
character of Bromeliaceae an alternative which is fatal to it
as a distinguishing character (“ calyx usually calycine, some-
times petaline"), and entirely to drop the position of the
ovary, which in other parts of the system is made conspi-
cuous. I cannot too distinctly declare that nothing polemic
is intended by my remarks ; that I have no wish but to con-
sider dispassionately how a satisfactory arrangement can be

40

POSTSCRIPT.

made, and to lend my humble aid to those who are more com-
petent to effect it. After mature deliberation I do not see that
1 can improve the disposition made of those three genera in
my MS. where they form the first suborder, being distinguished
by branching stems. Whether as such they shall form a sepa-
rate order is a discretionary point, of too little importance to
deserve much discussion. I think it will be inconvenient to
detach them so, and I am certain that, if it be done, the cau-
lescent portion must make another separate order, and Ama-
ryllidaceae be confined to the scapaceous plants ; but this
would narrow, instead of enlarging, our views of the creation.

It is gratifying to me to observe, that although my arrange-
ment of the monocotyledonous plants was made upon the
principles which I have laid down, without any reference to
that division made by Dr. Lindley, which he calls nixus or
alliances denoted by the termination ales, when I subsequently
examined my scheme to see how it would affect his alliances,
it appeared to dissever but one, which I think he may perhaps
abandon on further consideration, the union of the bright-
petaled Xyris and its fellows with the grasses. The rushes
seem to me more closely allied to the grasses than the
plants which have some coloured petals ; but the half-glu-
maceous orders stand in need of careful reconsideration. It
is impossible, as I have before said, that any consecutive
arrangement should exhibit a perfect concatenation of
vegetables ; but I have effected one in strict conformity
with the principles I had previously laid down, and
though of course it must be capable of improvement and
rectification, when the plants it contains shall be more per-
fectly understood, it will be found that, proceeding systemati-
cally on those principles, I have disjoined fewer affinities than
has yet been done by any of the existing arrangements, which
are based on no fixed principle. For instance, Bromeliaceae,
which are chiefly epiphytes, find themselves next to the gy-
nandrous epiphytes, where every cultivator would place them;
and the gynandrous epiphytes to the Scitamineous plants,
which have almost a gynandrous appearance ; whereas, I
find Bromeliaceae placed by Dr. Lindley between the Irises
and the aquatic frogbits which intervene between them and
the other epiphytes. The closely allied Amaryllidaceous and
Liliaceous plants, which are all called lilies by the unlearned,
and are only separated by ovary inferior and superior, are
brought in contact by my systematic arrangement, though

POSTSCRIPT.

41

Dr. Lindley separated them by fourteen orders. The palms
and pandanaceous plants, which the unlearned equally call
palms, are brought together, though I find the liliaceous
plants between them, and in that respect I think nature and
my system appear to work well together. The aquatic frog-
bits find themselves by system, as if accidentally, next to the
floating river plants, instead of intruding between Bromelias
and other epiphytes. The breaking up of the order Haemo-
doraceae does of itself arrange the disjointed parts where
their affinities are evident. In the position which I necessarily
give to Gillesiaceae, it becomes the point of contact between
the liliaceous plants and those which advance from the
grasses to meet them, so that it is almost immaterial on which
side the line of division it is placed, whereas it connects no-
thing where it now stands. I find therefore the axioms,
which I have laid down as the basis of the botanical science,
work more satisfactorily in the detail than I could have anti-
cipated before their application.

The circumstance that, working upon different plans, Dr.
Lindley and I should have come so near to the same juxta-
position of certain orders as he has adopted in his alliances,
though to a very different position of the alliances themselves,
seems very confirmatory of the soundness of his views with
respect to subordinate affinities, and on the other hand their
agreement with my systematic arrangement of the whole
must tend to uphold its propriety.

The arrangement which I have made is intended to
enable any person, however little skilled in botany, under-
standing the terms used, which are explained in the glossary,
to ascertain easily to what order any monocotyledonous plant
before him belongs ; but many of the orders are so unsatis-
factory in their construction, so overlaid with alternatives
which obstruct the view of their fundamental diversities,
though the separation may be correct, that I have not the
means of defining them by as satisfactory a feature as I should
wish to do, and I can only take the best that has been found,
or at least asserted, to be invariable.

W. H. 1836.

42

1. Communis inter plantas omnes consentio est, qua ab
aliis rebus discrepant. 2. Singulae quoniam variis qualitati-
bus discordant, qualitates singulae catervas plantarum ab
universis, singulas a catervis distinguunt. 3. Separandae
sunt gradatim universae, dum singulae singulis secernuntur.
4. Quibusnam qualitatibus in ea discriminatione locus prior
tribuendus sit, solicite deliberandum est. 5. Ea qualitas
anteferenda est, quae plantas singulas qualitate aliqua majoris
momenti aut pluribus qualitatibus consentientes haudqua-
quam aut quam minime secernit. G. Qualitas quaeque singu-
latim qualitati a se diversae objicienda est, minime verb qua-
litati, quae cum ipsa consentire possit. 7. Ubi duae qualitates
diversae semper associatae esse videntur, ea ex duabus quae
magis aspicienti est manifesta, potissimum sumenda est.
8. Qualitatis cujuslibet non semper absoluta est praesentia vel
absentia ; quamobrem formae sunt intermediae, quibus aut
locus intermedius praebendus est, aut locus quam maxime
consentaneus superfluitate aut defectu laborantibus. Catervis
plantarum locus intermedius magis idoneus ; singulae cum
consentaneis, superfluitate vel defectu nonobstante, locandae ;
ex. gr. Commelina hexandra est defectu triandra, Azalea
Rhododendron decandrum defectu pentandrum vel lieptan-
drum, Gethyllis et Vellosia hexandrae interdum superfluitate
pleiandrae. 9. Formae intermediae aliis qualitatibus saepe
inter se discrepant. 10. Deus Creator plantis singulis inter
se uno quodam certo qualitatum diversarum numero discre-
pare non dedit, sed aliae priore, aliae posteriore gradu ab
universis in processu separationis secernuntur ; unde quamli-
bet plantarum dispositionem, quae divisiones universarum
certo quodam graduum numero a puncto universae consen-
tionis deductas constituet, naturae ipsi repugnare et men-
dacem esse necesse est. Idcirco si, memoriae subveniendi
gratia, numerus divisionum praecipuarum certus aliquis
statuetur, uti classis, nixus, phalanx, ordo, subordo, et sectio,
falsa erit ista dispositio, nisi in divisione unaquaque grad us
convenientes absque certo numero praescripto constituantur ;

43

parvique divisiones istae praecipuae momenti erunt, utpote ad
libitum locatae, non ex natura rei provenientes. 11. Ad
harmoniam rerum creatarum demonstrandam perutilis est
catervarum ordinatio ; rei verb botanicae praecipuum in
plantarum singularum charactere vera et immutabili dig-
noscendo opus consistit. 12. Quodcunque nomen divisioni
cuique velis assignare, verus singularum plantarum character
in una divisione, non in diversis, existere necesse est. 13.
In divisione, quae vocatur species, character iste verus in
generationum serie immutabilis, teste cl. Jussieu, sedere puta-
batur. 14. Ulteriore specierum investigatione experientiaque
in studiis liortulanis nacta, nuper innotuit, species in generibus
quibusdam singulae plantae formas diversas praebere, utpote
ex mutuo contactu progeniem eamque saepe fecundam, pari-
turas ; et progeniei ista fecunditas idiosyncraseos in parentibus
consentione, magis quam ex affinitate botanica, oriri visa est ;
speciesque istae, quibus liac ratione characterem unum
eundemque inesse probatum est, diversitates eas ipsas exhi-
bent, per quas botanicorum mos est species universas secer-
nere. 15. Speciebus idcirco omnibus supradicto modo dis-
cretis characterem immutabilem non tribuendum esse parili
ratione probatum est. 16. Ergo non in divisione, quae a
botanicis species vocatur, sed in praestantiore ilia quae genus
nominatur, character individuus immutabilis positus est.
17. Si unquam ibi positus est, semper ibi positum esse
necesse est. 18. Modi sunt duo, quibus error iste a botanicis
admissus emendari potest ; vel in generibus solis characterem
individuum collocando, vel genera omnia (reformatis quae
reformari debent) pro speciebus, species pro varietatibus
sumendo. 19. Maxima ex hoc turba, ex illo nulla orietur.
20. Nulla est inter species et varietates generatione conti-
nuatas differentia naturalis, inque generibus quibusdam varie-
tates hortenses perfectius, quam in aliis species, ubi juxta
positae sunt, generatione sincera se continuant. 21. In
quibusdam generibus formae plantarum variae ex diversis
regionibus ortae specierum fines conturbant et commiscent,
quamobrem vana et futilis est contentio, utrum speciei aut
varietatis generatione continuatae (fortuitis varietatibus
exceptis) plantae cuilibet nomen tribuendum est. 22.
Character generis individuus immutabilis rerum factarum
est discriminatio ; specierum ac varietatum fines arbitrio
humano constituuntur. 23. Ubi ex mutuo duarum planta-
rum contactu planta intermedia prognata est, ex una aliqua

44

stirpe primaria parentes deductos fuisse, et characterem qao
discreti fuerint invalidum esse, locus est suspicari. 24. Fe-
cunditate plantse istius intermedi® parentium origo una
eademque esse probatur.

Sociis artis botanic® fautoribus conclusiones hasce breves
consectarias, rebus natura immutabilibus confidens, sibi diffi-
dens, humiliter proponit.

GULIELMUS HERBERT.

Spofforthia, A.D. 1835.

MONQCOTYLEDONES.

Plants having hut one seed-lobe ; found also to increase their
bidk by deposits in the centre, without the distinction of bark,
pith, wood, arid medullary rays.

1. SPADICEOUS. — Having a spadix ; that is, flowers not

petaloid, set round a stalk.

Involucrate. — Having an involucre below.

Ligneous. — Flower-stem woody.

( Palmales . Lindley.)

Palmaceae. — Spadix more loose or branched ; 1-3-seeded.
( Pandales . Lindl.)

Cyclanthaceae. — Flowers spirally arranged ; many-seeded.
Pandanaceae. — Flowers closely set; many-seeded.

§§. Succulent .* — Flower-stem juicy.

(Arales. Lindl.)

Araceae. — Flowers uni-sexual, petals wanting ; excluding
such aroid plants as have not an involucre.

§. Naked. — Wanting the involucre.

( Typ kales . Lindl.)

Typhaceae. — Flowers uni-sexual ; sepals three or more,
sometimes a bundle of hairs ; petals wanting.

(Piperales. Lindl.)

Piperaceae. — Flowers hermaphrodite. If retained amongst
monocotyledones, but removed by Dr. Lindley to dicoty-
ledones. Both these orders are hypogynous.

2. SUBSPADICEOUS. — Having a tendency to the form

of an imperfect spadix, either by an inferior in-
volucre or close-set spike, or flowers neither co-
rolliform nor glumaceous.

§. Subcorolliform. — Having a tendency to a petaloid
appearance in the flowers.

§§. Epigynous. — Having the stamens above the germen.
( Narcissales . Lindl.)

Taccaceae. — An anomalous race, with an Amaryllidaceous
scape, a perianth approaching to that of Tupistra, a tuber
and foliage approaching to Arum.

46

MONOCOTYLEDONES.

Hypogynons. — Having the stamens below the ger-
men.

( Aracece . Lindl.)

Orontiaceae. — Anthers turned inwards ; spathe none ?
Orontium, Tupistra, Aspidistra. This is an anomalous
race of plants. Orontium and Tupistra are decidedly
subspadiceous. Aspidistra and Tupistra are not truly
petaloid, nor yet glumaceous, which is contrary to the
habit of the exspadiceous plants, amongst which they
would appear misplaced. If Aspidistra has really eight
segments and four cells it must stand by itself, but it
looks in the figure as if two of the segments were merely
bifid at the point or 2-lobed. I venture, by analogy, to
doubt the fact of Orontium having posterior anthers
like Araceae.

Acoraceae. — Anthers turned inwards; spathe leaf-like.

( Fluviales. Lindl.)

Juncaginaceae. — Flowers in spikes; seeds erect.

Naiadaceas. — Flowers in spikes; ovules solitary, pen-
dulous.

Pistiaceae. — Flowers two, on the margins of the leaves.

§. Corolliform. — Having the semblance of petals.

Monoperianthine. — Having a simple perianth.

§§§. Tripetaloid. — Having the semblance of three
petals.

(Hy dr ales. Lindl.)

Hydrocharidaceae. — With an inferior involucre.

§§§. Hexapetaloid. — Having the semblance of six
petals.

(Liliales. Lindl.)

Pontederaceae? — Flowers with a close spadix-like
spike, or with an involucre. I am not acquainted
with those said to have an umbel. This order,
which is insufficiently described, belongs, perhaps,
to Liliacese, or requires reformation. I place it
here with doubt, not being thoroughly acquainted
with it.

Biperianthine. — Having a two-fold perianth.

( Amomales . Lindl.)

Zingiberaceae. — Middle stamen fertile.

Marantaceae. — One lateral stamen fertile.

Musaceas. — Stamens irregularly abortive ; flowers spa-
thaceous.

MONOCOTYLEDONES.

47

3. EXSPADICEOUS. — Not having the form of a spadix.

§. Corolliform. — Having the semblance of petals.

Gynandrous. —Having the style consolidated into a
column with the stamens.

C Gynandrales. Lindl.)

Orchidaceae. — Capsule 1-celled, 3-valved; seed-coat loose.
Vanillacese. — Fruit 1-celled, succulent; seed-coat tight.
Apostasiaceae. — Style free from the stamens the prin-
cipal part of its length, therefore less perfectly gynan-
drous ; fruit 3-celled.

§§. Agynandrous. — Not having the style consolidated
with the stamens.

§§§. Tripetaloid. — Having the semblance of three pe-
tals.

§§§§. Monogynous. — Having the semblance of one
style.

( Bromelales . Lindl.)

Bromeliaceae. — Sepals persistent ; leaves rigid.

( Commelales. Lindl.)

Commelinaceas. — Sepals leafy ; leaves not rigid.

§§§§. Pleiogynous. — Having more or distinct styles.
( Alismales. Lindl.)

Butomaceae. — Seeds numerous, attached to the
internal surface of the fruit.

Alismaceae. — Seeds 1-2, attached to the suture, at
a distance from each other.

§§§. Hexapetaloid. — Having the semblance of six pe-
tals.

§§§§. Triandrous. — Having three stamens.

§§§§§. Bpigynous. — Having the stamens above
the germen.

( Ixiales . Lindl.)

Iridaceae. — Anthers opening outwardly.
(Narcissales. Lindl.)

Burmanniaceaj. — Anthers opening trans-
versely.

Haemodoraceae. — Anthers opening inwardly.
Only the triandrous epigynous portion of
Dr. Brown’s order.

48

MONOCOTYLEDON ES.

§§§§§. Hypogynous. — Having the stamens below
the germen.

W achendorfaceae. — Anthers opening inwardly.

§§§§. Hexandrous. — Having six stamens.

§§§§§. Epigynous. — Having the stamens above
the germen.

Amaryllidaceae. — Anthers turned inwards.
Division 1. Ramosae. — Branching.

Suborder 1. Xerophyteae. — Schistandrous, £.e.

anthers opening the whole
length.

Division 2. Caulescentes. — With a stalk.
Suborder 2. Hypoxideae. — Schistandrous ;

operculous, i. e. the base of
the style forming a promi-
nent top to the seed-vessel.

3. Agaveae. — Schistandrous ; not

operculous.

Division 3. Scapaceae. — With a spathaceous
scape.

Suborder 4. Amaryllideae. — Schistandrous ;

petaline filaments excelling.

5. Narcisseae. — Schistandrous; se-

paline filaments excelling.

6. Galantheae. — Porandrous, i. e.

anthers opening partially.

( 7? Taccaceae? — Filaments hooded;

tuber and leaves arum-like.)
§§%§§- Hypogynous. — Having the stamens below
the germen.

(Liliales. Lindl.)

§§§§§§. Anthers turned outwards.

Melanthaceae.

§§§§§§• Anthers turned inwards.

Liliaceoe.

Veins not reticulate.

Suborder 1 . Allieae. — Scape with spa-
thaceous umbel.

■ 2. Asphodeleae. — Spike brac-

teate ; leaves not suc-
culent.

— 3. Aloeae. — Spike bracteate ;

leaves succulent.

4. Tulipeae ; scape without

spathe or bractes.

5. Convallarieae. — Having a

rhizoma or creeping
tuber.

MONOCOTYLEDONES.

49

Suborder 6. Asparageae.— Stalk leaf-bear-
ing-

Veins reticulate? Qu. whether always.

( Retosce . Lindl.)

■ 7. Smilaceae. — If not, they

unite with asparageae.

§§§§. Subhexapetaloid. — Having some of the six

petaloid segments occa-
sionally deficient.

Roxburghiacese. — Fruit 1 -celled. An order not
well established as now con-
stituted; perhaps a section of
Smilaceae with reticulate
veins, but Dr. Lindley gives
an alternative which is fatal
to it, viz. veins reticulate or
alternate.

§. Subglumaceous. — Having a tendency to scaly flowers.

(all hypogynons)

( Liliales . Lindl.)

Hexandrous defective.

Gillesiaceae. — Perianth minute, surrounded with scales ;
seeds attached to the axis.

( Glumosce . Lindl.)

Xyridaceae. — Tripetaloid; anthers turned outwards. Pla-
centae parietal, i. e. seeds not attached to
the centre of the fruit.

( Juncales . Lindl.)

§§. Hexandrous.

Juncaceae. — Not tripetaloid ; anthers turned inwards.

Triandrous.

Philydraceae. — Perianth two-leaved.

§. Glumaceous.

( Glumosce . Lindl.)

Ovules pendulous.

Restiaceae. — Placentae central, i. e. seeds attached to the
centre of the fruit. Anthers unilocular — •
anthers bilocular ; perianth 2-6-parted or
wanting. Requires reformation.
Eriocaulonaceae. — 2 segments anterior ; i posterior or de-
ficient.

Ovules not pendulous.

Desvauxiaceae. — Scape solid filiform ; inflorescence spa-
thaceous; fruit 1 -seeded utricles.

E

50

MONOCOTYLEDONES.

Cyperaceae. — Stems solid and angular ; fruit crustaceous
or bony.

Graminaceae. — Steins fistular ; fruit-coat membranous,
scarce distinguishable from the seed.

The circle is completed by returning to the spadiceous
division, of which the alliance to graminaceous plants is
manifested by the approximation of calamus amongst the
palms to bambusa of the latter.

Notwithstanding its unavoidable imperfections, I believe
the foregoing arrangement will nearly enable any person who
has no previous knowledge of botany, except understanding
the terms used, to discover readily the natural order to which
any monocotyledonous plant before him may belong. If
my knowledge of all the individuals that compose those or-
ders had been perfect, I should have distinguished each of
them from others in the same subsection by one sure feature,
the most obvious and easy to ascertain that could be found ;
and I should have subjoined in italics, as auxiliar, all other
features that seemed to be invariable. But I have to deal
with orders, some of which are inaccurately constituted, and
require to be reformed, and of which important features have
been either overlooked or but partially ascertained ; and I
have to wade through a long and unsatisfactory string of al-
ternatives, from which no certain and invariable distinction
can be elicited. I feel no hesitation in saying that all such
characters require to be remoulded. Alternatives may
severally head the sections of an order, but cannot together
distinguish one order from another. For instance, the words
“ tube short or long , limb flat or erect,” furnish points for
subdivision, but no point of distinction for the whole, ex-
cept the existence of a tube besides the limb, which would be
properly expressed by the words perianth tubed. In the cha-
racter of the order from which those words are taken, I find,
also, embryo orthotropous, heterotropous , or antitropous, from
which it results, either that the order embraces things so
dissimilar as to require reformation, or that the position of
the embryo is immaterial ; but such alternatives, presenting
every possible position of the embryo, can furnish no peculiar
distinction of the thing described, and are worse than super-
fluous, for they so encumber the real marks of separation,
that it becomes almost impossible to discover them.

With respect to the suborders I suggest for Liliaceae, it
must be observed, that it is immaterial whether Melanthaceae

MONOCOTYLEDONES.

51

be erected into a separate order by the previous separation
of anthers anterior and posterior, or considered as a sub-
order distinguished by that feature. It is equally a matter
for future consideration, to ascertain whether all Smilaceae
have confluent veins, and, if it should appear that they have,
whether it would be more consistent to separate them into
an order distinct from the nearly allied Asparageae on that
account, or merely to distinguish them as a suborder. I am
indebted to Dr. Brown for a sight of specimens of Dioscorea
lucida, of which the numerous cross veins are long and beau-
tifully parallel, though here and there one may be observed
bifid at the point, but preserving its direction to the longi-
tudinal rib, and never running into another cross-vein or
retroflex. It is true that the fruit of that plant has not been
seen, and, although from its aspect it can scarcely be doubted
that it is a Dioscorea, it may prove to be distinct from that
genus, of which I have examined fifteen species with con-
fluent veins. But if it were so distinguished, its close affi-
nity to those plants would not the less shew the insufficiency
of the feature to characterize an order. I find, also, the
cross-veins of Rajania quinquefolia to be parallel, though in
an oblique instead of a rectangular direction. I am on the
whole of opinion that the venation of leaves is not suffi-
ciently invariable to furnish a sure character. I believe that
a sound feature may be found in Inflorescence axillary
and Inflorescence not axillary, to detach Dioscoreae, Tamus,
Smilaceae, Asparageae, and Roxburghiaceae, from Amarylli-
deae and Liliaceae, before the subdivision into hypogynous
and epigynous, thus —

§§§§. Hexandrous.

§§§§§. Inflorescence axillary.

§§§§§§• Epigynous.

Dioscoreae. — (With Tamus.)

§§§§§§• Hypogynous.

Asparageae. — Veins parallel ; fruit 3-celled.

Smilaceae. — Veins reticulate? fruit 3-celled.

Roxburghiaceae. — Fruit 1 -celled.

§§§§§. Inflorescence not axillary.

§§§§§§• Epigynous.

Amaryllidaceae.

§§§§§§• Hypogynous.

Liliaceae.

E 2

52

MONOCOTY LEDON ES.

If the fact should prove to be correct, that the inflores-
cence of the plants proposed to be so detached is invariably
axillary, I should certainly prefer that separation ; but that
distinction has never been noticed, and I am fearful lest it
should not be supported by the fact.

I am aware that the foregoing arrangement is liable to
the objection, that the spadix itself, being in its perfect form
composed of more than one feature, does not fall precisely
within my own limitation ; and that strictly^ the flowers being
apetaloid, set round a stalk, and involucrate, are three fea-
tures which should be exhibited separately. Perhaps the
following arrangement may be preferable : —

1. SPADICEOUS. — Meaning thereby simply flowers set round a

stalk. §. Petai.otd. §§. Biperianthine. — Zingiberacece. Maran-
tacece. Musacece. §§. Monoperianthine. — Pontederacece ? §. Sub-

petaloid — Orontiacece. Acoracece. Juncagineee. Apetaloid.

Naked. — Naiadacece. Typhaceee. Piper acece ? §§. Involucrate.

§§§. Ligneous. — Pandanaceee. Cyclanthaceee. Palmacecc. §§§. Suc-
culent.— Araceee.

2. EX SPADICEOUS. — Flowers not set round a stalk. §. Ape-

tai.oid. §§. Not glumaceous. — Pistiacece. §§. Glumaceous. — Gra-
minaceee. Cyperacece. Desvauxiacece. Eriocaulon acece. Restiacecc.
§. Subpetaloid. — Juncacece. P hilydracece . Xyridaceoc. Gillesiacece.
Taccace.ce ? §. Petaloid. §§• Agynandrous. Subhexapetaloid. —
Roxburghiacece? §§§. Hexapetaloid. §§§§. Hexandrous. *§§§§§. Hy-
pogynous — Liliaceee. Epigynous. AMARYLLiDACEiE. §§§§. Tri-

androus. §*$§*§• Epigynous. — Burmanniacece. Iridacece. Heemodo-
racece limited. Hypogynous. — Wachendorf acece. §^§. Tripe-

taloid. §§§§• Pleiogynous. — Butomacece. Alismacece. §§§§. Mono-
gynous. — Hydrockaridacece. Commelinacece. Bromeliacece . §§. Gy-

nandrous. — Apostasiacece . Vannillacece . Orchidacese. — The circle
returns to Zingiberaceae. — It will be observed that Pistiaceae seems to
interfere between Palmaceae and Graminaceae, but the connexion is
lateral between Araceae and Pistiaceae, Palmaceae and Graminaceae.

Araceae. Pistiaceae.

Pandanaceae.

Cyclanthaceae.

Palmaceae. — Graminaceae.

Cyperaceae, &c.

By a singular fatality in forming the character of Heme-
roeallide® as a suborder, Dr. Bindley has also made an over-
sight concerning its seed-coat, which he defines to be pale
and soft, though in truth it is dark glossy black and very
brittle. That of Agapanthus is black also.

53

MONOCOTYLEDON ES.

AMARYLLIDACEiE.

Character. — Expadiceae; corolliforines; agynandrae; liex-
apetaloidese ; epigynae ; hexandrae. [Stylo triplici, antheris
anticisS\ Vel, si mavis; Canlis aut scapus non spadiceus ;
germen inferum, nisi defectu, triloculare ; perianthium co-
rolliforme simplex hexapetaloideum ; stamina sex, rarius
superfluitate numerosa ; styli tres non staminiferi, saepius
consolidati, interdum adnati partibiles, rarissime distincti.

Observations. — Antherae quoad novi, anticae, biloculares.
Stylorum cognatorum bases (ideoque stigmata, ubi stylus
rectus est) lobique operculi in apice germinis costas sepa-
linas respiciunt; val varum sutura costae sepalinae, dissepi-
mentum petalinae, continuatio est ; dissepimentum placentae
medio incumbit, ideoque series binae ovulorum in loculis
singulis diversis placentis adhaerent. Valva septigera inter
costas sepalinas adempta, loculorum duorum pars dimidia
apparet, ovulis, ubi cumulata sunt, divergentibus ; peri-
carpii parte inter costas petalinas adempta, loculus integer
patet, ovulis, ubi cumulata sunt, convergentibus. In Ama-
ryllidacearum omnium seminibus (ni fallor) foramen chalazae
seu vertici oppositum est; embryonis radicula ad foramen
spectat, numquam vere umbilico approximatur, nisi quod
vicinus sit foramiui umbilicus ; umbilicus inter foramen et
verticem situs est ; funiculus umbilicalis foramen attingit ;
quamobrem, ubi foramini contiguus est umbilicus, uti in
Hippeastro, gracilior est funiculus; ubi remotus, uti in Cur-
culigine, latior, spatium scilicet intermedium (quod cum
umbilico ipso veriiis hilum nuucupatur, ab umbilico
mero satis diversion) cooperiens. Ovula seminum tes-
taceorum, quoad novi, ab axe recte fere procedunt (ca-
nali umbilicali ad verticem, quae chalazae regio est, circum-
lato, atque ibi cotyledoni subministrante) ; fecundata obli-
quantur, vertice declinato, foramine surgente. Ovula semi-
num carnosorum vel fune brevi forti crasso alliguntur atque
in crescendo vix obliquantur, vel hilo latiore placentae ipsi
adnata esse videntur.

In Hypoxidiformibus et Galantheis, quoad novi, fila-

54

AMARYLLIDACE.E.

menta subaequalia sunt; in Alstrcemeriaeformibus, et in Ama-
ryllideis, ubi non subsequalia, petalina, in Narcisseis sepalina
semper praestantiora sunt. Folia plerumque sunt paralelo-
venia; sed in late ovalibus (ex. gr. in Crino petiolato, Grifti-
nia, Urceolina) partim semipennivenia sunt, uti etiam in
genere Costo inter scitamineas, venis multis a costa media
oriundis; et Tamus cum Dioscoreis plerisque (sed non om-
nibus) venis bifidis confluentibus distinguitur. Plicatio
foliorum vix, nisi inter Hypoxidiformes, invenitur, lusu
naturae excepto in Crino anomalo ; Galanthi plicati non
vera est plicatio, sed margo, uti in Amaryllideis quibusdam,
magis verb conspicue, retroflexus. Antherae biloculares sunt;
petalinae et sepalinae minuta quadam diversitate affiguntur.
Antherae praestantioris positio a me potissimum designatur.

AMARYLLID ACEiE.

Character. — Not-spadiceous, corolliform, not-gynandrous,
hexapetaloid, epigynous, hexandrous.

Stalk or scape without a spadix (meaning thereby flowers ,
which have not the semblance of petals, set round a stalk, and
when perfect having an involucre below ) ; germen inferior, and,
unless by defect, 3-celled ; perianth simple ( that is, as con-
trasted with the scitamineous plants), having the semblance of
six petals; stamens six, rarely numerous by superfluity; styles
three, not bearing the stamens (as contrasted with the gynan-
drous plants), mostly consolidated, sometimes partibly adnate
to each other, rarely separate.

Observations. — The anthers, I believe, in all the genera
are 2 celled, and face inwards. The bases of the styles, and
consequently the lobes of the stigma, if strait, and of the
opercle on the summit of the germen, are opposite the ribs
of the sepals ; the sutures of the valves of the seed-vessel are
continuations of the line of the sepaline, the dissepiments of
the petaline, ribs. The valve, which bears the dissepiments
between the lines of two sepaline ribs, being taken oft", half of
two cells is exposed, the ovules, when they are heaped,
diverging; the portion of the pericarp between the line of
two petaline ribs being taken off, an entire cell is exposed,
the ovules converging. Unless I am deceived, in the seeds
of all Amaryllidaceous plants the foramen is opposite to the
chalaza or summit ; the radicle of the embryo points to the

AM ARY LLIDACEiE.

55

foramen, but is never truly approached to the umbilicus,
except because the umbilicus is near the foramen ; the um-
bilicus is situate between the foramen and summit of the
seed ; the breadth of the umbilical cord reaches to the fora-
men, wherefore, if the foramen is near to the umbilicus, as
in Hippeastrum, the cord is slender ; if it is remote, as in
Curculigo, wider, that is, occupying the intermediate space
(which, together with the umbilicus, constitutes properly the
hilum), more extensive than the mere umbilicus. The
ovules of the testaceous seeds appear to me to proceed at
right angles from the axis, the umbilical channel making a
circuit by the line of the raphe to the summit, which is the
region of the chalaza, and there furnishing nourishment to
the cotyledon ; and after fertilization the ovule appears to
slope, the summit being lowered and the foramen propor-
tionally elevated. The ovules of the fleshy seeds seem to be
either attached by a short thick cord, or to adhere with a
wider hilum to the placenta.

My own observations would lead me to believe that there
is a very minute communication with the foramen, soon be-
coming obsolete, along the margin of the umbilical cord, but
I have not had the use of a powerful microscope, and I can-
not venture to dissent from the received opinions concerning
the fertilization of ovules, though from their extreme impro-
bability my understanding does not willingly yield assent to
them.

First Division. — BRANCHING.

Suborder 1. Xerophytea:. — Leaves rigid.

1. Xerophyta. Jussieu. — Flowers terminal.

% Flowers not terminal.

2. Vellosia. Vandelli. ( Campderia . Kth. Raddia. Rich.)

— Perianth scarcely tubed ; filaments some-
times numerous, superfluous.

3. Barbacenia. Vandelli. — Perianth tubed.

Second Division.— CAULESCENT.

Suborder 2. HypoxidiLe. — Schistandrous, i. e. anthers open-
ing their whole length. Operculate, i. e. the
point of the germen superior, and forming a base
to the style.

56

AMARY LLIDACEA2.

§. 1. Hypoxidiformes. — Sepals and petals conformable;
leaves plicate.

(Weldenia? Schlechtendal. If really a depau-
perated hexapetaloid epigynous plant, which I
doubt.)

4. Curculigo. Gaertner. — Tube cylindrical upwards;

anthers distant ; seeds with a deep-gulphed
hilum.

5. Molineria. Colla. — Tube short ; anthers fascicu-

late ; seeds round.

6. Hypoxis. Linnceus. (Fabricia. Thunb.) — Perianth

deep-cleft, patent in the sun, base annular.

7. Ccelanthus. Schultes. — Perianth campanulate con-

niving.

§. 2. Lanari.eformes. — Leaves equitant ; sepals and petals
conformable.

J Perianth tubed.

8. Anigozanthus. La Billardiere. ( Schwcegrichenia.

Spr.) — Limb short, more deeply cleft on the
lower side.

9. Lanaria. Alton. ( Argolasla. Jussieu ). — Limbregular.

;{; Perianth deep-cleft.

10. Lophiola. Ker. — Limb regular reflex ; anthers ver-

satile ; style tripartible.

11. Conostylis. R. Brown. — Limb regular, half ex-

panded ; style conical dilated.

12. Phlebocarya. R. Brown. — Anthers subsessile ; style

filiform.

13. Campynema. La Billardiere. — Filaments recurved ;

anthers versatile ; styles separate recurved.

§. 3. Alstrcemeri^eformes. — Sepals and petals not respec-
tively uniform. Petaline filaments excelling.

J Inflorescence umbellate.

14. Choeradodia. Herbert. — Style strumous ; capsule

triangular. If it should prove not operculous it
must be placed next before Doryanthes.

1 5. Alstrcemeria. Linnceus.— Valves of the capsule open-

ing from the bottom. ( Stalk erect.)

16. Collania. Herbert. — Pericarp pulpaceous ; more

than half the ovary superior. ( Stalk rigid, umbel
nodding.)

17. Sphserine. Herbert. — Ovary but little superior; cap-

sule indehiscent. ( Stalk tapering, flexuous.)

AM ARY LLI DACEJi. 57

18. Bomarea. Mirbel. — Capsule dehiscent from the top.

{Stalk twining.')

Suborder 3. Agaves. — Schistandrous, not operculous.

1. Dioscore^formes. — Spikes (I believe always) axil-
lary. Obs. Veins of the leaves generally con-
fluent; plants dioecious , climbing.

19. Tamus. Linnaeus. — Fruit with outer coat valveless,

middle coat pulpaceous, inner coat 3-valved
bearing the dissepiments,

^ Capsule valued; seeds flat, shelly.

20. Testudinaria. Salisbury. — Filaments inserted in the

base of the segments.

21. Dioscorea. Linnaeus. — Filaments inserted in the

disk. N. B. If that distinction does not hold
good, testudinaria must probably fall into Dios-
corea. The statements as to the fact are con-
tradictory.

£ Capsule one-seeded , with a long curved wing.

22. Rajania. Linnaeus.

§. 2. IxIjEformes. — Root bulbous.

23. Bravoa. La Llave. ( Ccetocapnia. Link et Ot.) — In-

florescence spiked ; flowers tubular.

24. Ixiolirion. Herbert. — Flowers deep-cleft ; stamens

perfect.

(Tecophilea? Bertero. — Flowers deep-cleft;
stamens three fertile, three abortive. I believe
this to be a truly triandrous plant belonging to Iri-
daceae .)

§. 3. Agaviformes. — Semibulbous or stemmed.

J Inflorescence spiked or branching.

25. Fourcroya. Ventenat. — Perianth not tubed ; style

strumous.

2G. Agave. Linnaeus. — Perianth tubed ; style filiform.

J Inflorescence umbellate.

27. Doryanthes. Correa. — Perianth tubed.

Third Division.— SCAPACEOUS.

Scape succulent, spathaceous, not articulate below the spatlie.

Suborder 4. Amaryllide^e. — Schistandrous, not operculate.
Petaline filaments, unless equal, excelling.

■f Cavae. — Scape hollow ; seeds compressed with a black
shell, capsule 3- celled ‘3-valved.

58

AMARYLLIDACE./E.

§. 1. Cyrtanthiformes. — Tube wide-mouthed; germen
oval, pedunculated. African.

28. Cyrtanthus. Alton. — Tube curved; filaments strait,

decurrent.

29. Gastronema. Herbert. — Tube curved; filaments

conniving, three bent downwards.

30. Vallota. Herbert. — Tube strait, filaments strait ad-

nate.

§.2. Hippeastriformes. — Tube narrow-mouthed; germen
wider above, constricted in the middle. Occidental.
X Perianth declined , tubeless.

31. Sprekelia. Heister. — Connexion annular ; filaments

fasciculate declined.

X Perianth declined ; tube abbreviated and faucial
membrane deficieiit on the lower side.

32. Hippeastrum. Herbert. — Limb with fourfold dis-

parity.

^ Perianth declined ; tube not abbreviated.

33. Phycella. Lindley. — #Perianth convolute into the

form of a tube; anthers incumbent; faucial
membrane, when manifest, equal.

34. Habranthus. Herbert. — Perianth not convolute ;

anthers incumbent, versatile, attached by the
middle ; faucial membrane annular ; leaves
hiemal.

A. The upper sepaline filament prolonged ; two or
more-flowered.

B. Upper sepaline and lower petaline filaments
abbreviated ; 1-flowered. — Or a distinct genus,
Zephyrites ? Herbert. — To be further investigated.

X Perianth suberect.

35. Zephyranthes. Herbert. — Filaments of alternate

length, inserted at the base of the segments, pa-
tent, a little conniving ; anthers linear, attached
below the middle, suberect ; pollen oval ; flowers
expand in the sun ; lobes of the stigma patent.
Argyropsis ? Herbert. — Anthers erect, after inver-
sion subulate and spirally twisted at the point ;
lobes ofthestigma erect; pollen irregularly shaped.
X Perianth erect.

* Eustepliia (Cavanilles) must be expunged ; no such plant will be found.
It seems to be a species of Phycella, of which the description is unintelligible,
and the accompanying figure repugnant to all probability.

AM ARYLLIDACEiE.

59

36. Cooperia. Herbert. ( Scepti'anthes. Graham.) — Tube

erect, cylindrical, wider at the mouth, limb stel-
late in its prime ; anthers inserted at one-third
from the base, fasciculate; stamens inserted, with
little alternate variation, at the mouth of the tube.
Flower (Qu. whether always) nocturnal.

37. Pyrolirion. Herbert. — Tube cylindrical ; limb cam-

panulate, with reflex points ; filaments erect,
patent.

38. Haylockia. Herbert. — Filaments conniving, sepa-

line inserted in the base of the limb, petaline
higher. ( Germen subterraneous.)
f Solidae.- — Scape solid.

^ One-flowered.

§. 3. Oporanthiformes. — Perianth without cup ; — seeds
testaceous ?

39. Gethyllis. Linnaeus. — Tube long, cylindrical, conso-

lidated with the lower part of the style ; filaments
inserted at the mouth of the tube, sometimes
superabundant. ( Germen subterraneous.)

40. Sternebergia. Kitaibel. — Tube cylindrical; style

free; capsule 3-valved ; seeds black. ( Germen
subterraneous.)

41. Oporanthus. Herbert. — Tube short, somewhat fun-

nel-shaped ; filaments inserted equally within
its mouth.

£ Many flowered. .

42. Lapiedra. Lagasca. — Perianth stellate, deep cleft;

seeds smooth, angular. Scape supposed solid.

§. 4. Pancratiformes. — Cup staminiferous.

J Seeds black shelly.

43. Tapeinanthus. Herbert. — Perianth deep cleft ; cup

short; scape supposed solid.

44. Chlidanthus. Herbert. — Only the rudiments of a

cup ; membranous wings to the filaments, ad-
hering to the perianth ; tube long, cylindrical,
widened at the mouth ; limb semipatent.

45. Clinanthus. Herbert.— Membranous wings to the

filaments, free, contiguous (if not united at the
base?) tube long funnel-shaped, limb short.

46. Urceolina. Reichenbach. (Urceolaria. Herb. App.

Collania. Schultes.) — Tube short, cylindrical,
pendulous; limb ventricosely campanulate; sup-

GO

AMARYLLIDACE.-E.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

posed by me to have some rudiments of a cup,
from its evident affinity to Leperiza. If it lias
not, it must follow Lapiedra.

Leperiza. Herbert. — Tube strait subcylindrical pen-
dulous, cup short.

Carpodetes. Herbert. — Germen erect, tube curved
slender cylindrical, limb subcampanulate ; cup
short.

Coburghia. Street. — Germen rather drooping ; tube
long curved cylindrical, enlarged upwards ;
limb regular half patent ; cup tubular.

Stenomesson. Herbert. ( Chrysiphiala. Ker.) — Tube
nearly strait, constricted in the middle, ventri-
cose upwards ; limb short regular ; cup short.

Eucrosia. Ker. — Tube declined ; cup declined, ab-
breviated on the upper side, shovel-shaped ; limb
compressed, curved upwards.

Elisena. Herbert. ( Liriope Herb. App.) — Tube cy-
lindrical ; limb reflex ; two segments sloped
downwards; cup sloped.

Pancratium. Linnaeus.— Tube cylindrical strait, cup
conspicuous, limb patent, anthers short sub-
erect.

A. Subsessile; seeds compressed, black, foliaceous;
leaves persistent.

B. Pedunculated ; seeds roundish, harder shelled.
Genus ? Halmyra. Salisbury.

C. Sessile ; leaves deciduous, fruit unknown.
Asiatic. Genus ? Tiaranthus. Herbert, if the
seeds prove to be fleshy, as stated by Rheede.

;{; Seeds fleshy, green ; tube cylindrical.

Hymenocallis. Salisbury. — Limb reflex; tube strait;
filaments erect, a little conniving ; anthers long,
attached at a point below the middle, versatile,
pendulous ; seeds large oblong.

Choretis. Herbert. — Tube nearly strait ; limb
reflex; filaments erect conniving ; anthers long,
attached by a callosity above the middle, pendu-
lous from the first ; seeds shorter, oblong.

Ismene. Salisbury. — Tube curved ; filaments con-
niving, deflex ; anthers long, attached below the
middle ; seeds round.

(Callithauma ? Herbert. — Limb patent; cup about
erpiul ; filaments exceeding.)

AMARYLLIDACEjE.

61

57. Calostemma. R. Brown. — Limb strait, half patent

suberect ; cup sometimes split ; filaments erect ;
anthers short, erect ; germen usually 2-seeded,
the dissepiments obsolete.

58. Vagaria. Herbert. — Cup imperfect, 6-cleft to the

base ; limb substellate ; seed unknown.

59. Eurycles. Salisbury. ( Proiphys.Herb.App .) — Limb

strait; filaments erect; anthers erect; dissepi-
ments imperfect ; ovules protruding the embryo.
4. Amaryllidiformes. — Scape solid; seeds fleshy;
stamens not connected by a cup.

60. Grifflnia. Ker. — Tube declined cylindrical, limb

reflex, lower segments divaricate, lowest pro-
jected.

61. Lycoris. Herbert. — Tube declined, wider at the

mouth, curved upwards ; limb curved upwards.

62. Olivia. Lindley. ( Imatophyllum. Hooker.) — Germen

pendulous ; perianth curved on the upper side
only, with four-fold disparity of limb; fruit
valveless, with pulpaceous middle coat.

63. Hsemanthus. Linnaeus. — Tube strait, perianth re-

gular; fruit valveless, with pulpaceous middle
coat.

64. Buphane. Herbert. — Tube strait, perianth regu-

lar ; capsule 3-valved, dry ; ovules attached by
a cord.

65. Ammocharis. Herbert. — Tube strait, perianth re-

gular ; capsule not pulpaceous ; ovules adnate
to the edge of the placenta; anthers short.

66. Crinum. Linnaeus. — Germen thicker in the middle ;

tube slender, cylindrical ; fruit soft, valveless ;
dissepiments becoming obsolete ; anthers long.

67. Amaryllis. Linnaeus. — Tube narrow-funnel-shaped ;

petaline filaments inserted at the base of the
segments, sepaline lower ; capsule obovate.

68. Brunsvigia. Heister. — Tube short or none ; perianth

and filaments recurved ; capsule 3-cornered tur-
binate.

69. Nerine. Herbert. — Limb reflex, base annular ; fila-

ments equally adnate, with a gibbous monadel-
phous base ; style filiform ; anthers incumbent.

70. Strumaria. Jacquin. — Base of limb annular; fila-

ments connected at the base, all or alternately

62 AMARYLLIDACE^E.

often adnate to the lower part of the style ; style
strumous ; anthers incumbent.

71. Hessea. Herbert. — Tube short, limb regular ; fila-

ments equal, subulate, reflex ; anthers short,
erect; style filiform.

72. Imhofia. Herbert. — Perianth deep cleft, patent ;

style thick or strumous ; filaments inserted at the
disk.

73. Carpolyza. Salisbury. (Hessea. Bergius.) — Tube

short, funnel-shaped ; limb half-patent ; fila-
ments adnate ; style thick, sulcate.

Suborder 5. Narcissi. — Scapaceous, schistandrous, not
operculate ; sepaline filaments excelling ; cup
including them.

74. Corbularia. Salisbury. — Tube and cup funnel-

shaped ; style and filaments declined, recurved ;
sepaline inserted at the base, petaline adnate
near the base of the tube ; anthers short, at-
tached at the middle, incumbent, versatile. (Cup
longer than the tube)

75. Ajax. Salisbury. — Style strait, subulate ; filaments

strait, adnate to the lower part of the tube ; an-
thers linear, strait, attached below the middle,
after inversion inveloping the top of the fila-
ment. (Cup longer than the tube)

76. Ganymedes. Salisbury. — Style strait, slender ; fila-

ments adnate to the upper part of the tube more
unequally than in Queltia, and the sepaline
more prolonged ; limb decidedly reflex. (Cup
not exceeding the tube)

77. Queltia. Salisbury. — Style strait, more or less at-

tenuated upwards ; filaments adnate unequally
to the upper part of the tube ; anthers linear,
affixed below the middle, recurved, margins
meeting at top above the filament, but not en-
veloping it. (Cup not equal to the tube)

78. Narcissus. Linnceus. — Style strait, slender ; fila-

ments straight, free at the point only, adnate un-
equally at and near the mouth of the tube ; an-
thers attached below the middle, short, suberect,
point recurved, margins not meeting behind.
(Cup near four times or more shorter than the
tube)

AMARYLLI DACE/E.

63

79. Hermione. Salisbury. — Style strait, slender; fila-

ments conniving free at the point only, adnate
unequally at and near the mouth of the tube ;
anthers acute oval, attached at the middle, in-
cumbent. (Cup shorter than the tube)

Suborder 6. Galantheae. — Scapaceous, porandrous (i. e. an-
thers opening partially), not operculate.

§ 1. Scape solid ; seeds white or greenish.

80. Galanthus. Linnceus. — Sepals and petals dissimilar ;

anthers apiculate, opening at the top ; style fili-
form, tapering.

81. Acis. Salisbury. — Sepals and petals a little dissimi-

lar ; anthers not apiculate, opening at the side
near the top ; style filiform, attenuated upwards;
stigma a little divided.

82. Erinosma. Herbert. — Sepals and petals similar ;

anthers not apiculate, opening at the top ; style
clavate ; stigma triangularly acuminate.

§ 2. Scape hollowr ; seeds black with a loose shining shell.

83. Leucojum. Linnceus. — Segments nearly equal ; an-

thers dehiscent laterally from the terminal ori-
fice, but not to the base ; style clavate ; stigma
acuminate.

Suborder 7 ? Taccaceje ? — With aroid tuber and leaves ;

filaments hooded at top, longer than the anthers.
Referred by me to the subspadiceous plants, believing
the perianth not to be truly petaloid ; but I have never seen
its flower. It has an evident affinity to Orontiaceae.

Tacca. Linnaeus. — Perianth regular persistent ; fila-
ments inserted at the base of the segments, di-
lated below ; anthers adnate within the hood ;
ovary 1 -celled, with three parietal polyspermous
placentae.

Ataccia? Prezl.

Divisio Prima. — RAMOSiE.

Subordo 1. Xerophyteas. — Schistandrae.

1. Xeropliyta. — Perianthium tubatum limbo regulari
persistente ; filamenta basi laciniarum inserta ;
antherae longae lineares subsessiles ; stigma ob-

64

AMARYLLIDACE^E.

long'um dilatatum ; capsula scabra trilocularis
polysperma. (Genitalia recta?) Planta suffruti-
cosa ramulis alternis vaginis foliorum deciduorum
imbricantibus scabra , folds uti in pinu rigidis acu-
tis, in apice ramulorum confertis , jioribus termina-
libus subsolitariis. Ex insula Madagascar. Spe-
cimen Commerson.

2. Vellosia.— Per. campanul. vix tubat. (stam. inter-

dum 12-15-18-24, phalangibus 3-6 connata) fi-
lam. brevia, anth. erect® ; (pollen globosum ?)
stylus rectus tripartibilis ; capsula trilocularis
trivalvis valvis in vertice horizontalibus vel im-
perfectis ; sem. quadrata cuneata testa simplici
chartacea. Fide cl. Martii, milii ignota. Plant®
Brazilian®, Yucc®formes.

3. Barbacenia. — Germen erectum ; perianth, conti-

nuum tubatum limbo reflexo; filamenta bifida an-
tlieris adnatis longiora ; stylus acuminatus stig-
matibus longior ; sem. parva numerosa, compres-
sione irregularia, testa badia albo marginata.

Divisio Secunda.— CAULESCENTES.

Subordo 1. HypoxiDEiE. — Schistandr® ; operculat®.

§ 1. Hypoxidiformes. — Sepala petalis conformia ; folia
plicata.

Weldenia ? si verb epigyna hexapetaloidea de-
pauperata, quod vix credo.

Perianthium tubulosum erectum, limbo pa-
tente tripartito ; filamenta tubi fauci inserta,
alterna breviora ; anther® erect®, fundo inter
loculos affix® ; stylus filiformis erectus ; stigma
trigono-capitatum. Fide cl. Karwinsky, mihi ig-
nota. Superumne germen an inferum, sepalane an
petala abortiva sint, nescio.

4. Curculigo. — Germen erectum, fere subterraneum,
bracteatum ; tubus stylo accretus, superne cylin-
dricus ; limbus regularis patens ; filamenta bre-
via ori tubi inserta ; anther® erect® distantes ;
semina nigra pendentia, funiculo crasso albo si-
num grandem intra umbilicum et foramen re-
plente.

AMARYLLIDACEiE.

65

5. Molineria. — Folia longe petiolata; caulis apice cur-

vatus capitulatus multiflorus ; germen nutans
bracteatum ; tubus stylo accretus brevis, limbus
regularis patens ; filamenta brevia ori tubi in-
serta ; antherae erecte fasciculate (monadel-
phe ?) semina nigra rugosa rotunda. Nomen a
Collce hort. Rip. mutuatus sum. Species seminibus
sine processu laterali procul dubio distinguendce sunt.

6. Ilypoxis. — Germen erectum; perianthium profunde

fissum basi annulari, sub sole patens, persistens ;
filameuta brevia subulata d-isco inserta ; antherae
sagittate erecte intra lobos affixe ; stylus brevis
uscpie ad operculum trisulcus tripartibilis ; stigma
breve erectum fimbriatum ; capsula trivaivis
valvis ab operculo transverse disruptis ; superne
deliiscentibus, infra conjunctis ; semina parva
nigra foramine apiculato, umbilico minutissime
rostra to.

Genera duo sub Hypoxide latere suspicor, alte-
rum bidbo foliisque annuls, alterum persistentibus.

7. Ccelanthus. — Scapus folio complicato vaginante fere

totus inclusus pauciflorus ; germen sessile, lim-
bus (sexfidus ?) campanulatus connivens ; fila-
menta basi laciniarum inserta Genus obscurum
fide Schlechtendal in Reliq. Willd. specie unica ,
patrui ignotd.

<§. 2. LANARieFORMES. — Sepala petalis conformia ; folia
equitantia.

8. Anigozanthus. — Perianthium tubulosum tubo cur-

vato limbo inferne profundius fisso laciniis sub-
equalibus inferne distantibus, filamenta basi
dilatata flexa brevia ore tubi inserta ; anthere
erecte adnate; stylus curvatus deciduus; stigma
clavatum ; capsula 3-locularis operculo dehiscente;
semina numerosa obovata nigra.

9. Lanaria. — Perianthium tubatum persistens, limbo

regulari ; filamenta basi laciniarum inserta ; an-
there versatiles ; capsula trilocularis loculis 2-3-
spermis.

10. Lophiola. — Perianthium limbo profunde fisso regu-
lari reflexo; filamenta erecto-patentia ; anthere
oblonge reclinate ; stylus usque ad operculum
tripartibilis, stigma simplex; operculum pre-

F

OG

AMARYLLIDACEiE.

grande, capsula ovato-py rami data, trilocularis
loculis biseriatim polyspermis.

11. Conostylis. — Periantliium limbo profunde fisso, re-

gulari, semipatente, persistente ; filamenta brevia
erecta ; antherae erect® ; stylus conico-dilatatus
cavus tripartibilis persistens ; stigma breve ; cap-
sula operculo dehiscente ; semina numerosa.

12. Phlebocarva. — Periantbium profunde fissum persis-

tens; filamenta basi laciniarum inserta ; anther®
tetragon® subsessiles ; stylus filiformis ; stigma
simplex ; ovarium defectu uniloculare trisper-
mum ; capsula corticata monosperma.

13. Campynema. — Per. profunde sex-fid um persistens ;

filamenta laciniarum basi inserta recurva; an-
ther® versatiles ; styli (nisi potius stigmata sint)
distincti recurvi ; capsula trilocularis tripartibilis
apice introrsum dehiscens ; semina numerosa
complanata testa spongiosa.

§. 3. Alstrcemeriasformes. — Sepala petalis nonconformia;
filamenta petaliua pr®stantiora.

14. Cboeradodia. — Sepala et petala valde disparia ; sty-

lus strumosus ; capsula trigona. Planta radice
fibrosa, folds radicaVibus majoribus, caulinis mino-
ribus, caide alto umbellato multifioro ; fores albucce
habitu , fide cl. Molince.

15. Alstrcemeria. — Germen ovulis suberectis sexcosta-

tum (superne 12-angulare 12-costatum) perian-
tbium sexpartitum laciniis quater disparibus,
petalis duobus erectis imo porrecto ; filamenta ubi
matura recurvata, glandulis in disco deciduis in-
serta, laciniarum basi vix adnata tardius distincta,
petalina basi acute ovali sepalina ultra-semicir-
culari discum signantia ; anther® basi affix® ;
stylus recurvus ; capsula oblongo-rotunda acute
operculata sexcostata valvis crustaceis septigeris
dissilientibus axe ab ima parte trifariarn disrupto,
costarum dimidio inseparabiliter pedunculo ad-
b®rente ; semina subrotunda testa tuberculata
difficulter separabili, liilo l®vi, cbalaza circulari,
endopleura ab albumine corneo inseparabili.

Plantai occidentals caule ( quoad novi) erecto
folioso vel squamato peduncidis \-plurifioris brac-
teatis, radice tuberosd pcdmatd.

AMARYLLIDACEjE.

67

16. Collania. — Caulis rigidus erectus apice curvatus ;

folia rigida? germen pendulum turbinatum, oper-
culo ad basin sty li tardius maxime amplificato ;
periantliium sexpartitum sepalis petalisque dis-
paribus sub-tubiforme inter se paribus; filamenta
et stylus recti ; antherse basi affix® ; pericar-
pium parte majore operculosum (molle? pulpa-
ceum ? edule ?)

17. Sph®'rine. — Caulis rectus superne attenuatus ; folia

subrigida; (pedunculi uniflori?) periantliium
sexpartitum breve sepalis petalisque disparibus
inter se paribus; (filamenta recta?) anther® basi
affix® ; (stylus rectus ?) capsula oblongo-rotunda
indehiscens. (Semina pulpa pauca?) Plant®
occidentales (caule solitario?)

18. Bomarea. — Caulis volubilis ; germen trigone turbi-

natum ; periantliium sexpartitum sepalis peta-
lisque disparibus inter se vix disparibus ; fila-
menta glandulis in disco deciduis inserta laci-
niarum basi vix aut nequaquam adnata; anther®
basi affix® ; capsula evalvis coriacea trigone
turbinata sexsulcata operculo obtuso, serius cor-
rugata dissepimentis integris operculo late deliis-
cente ; semina obovate subrotunda pericarpio
maturo longum adh®rentia exopleura molli
(semperne rubra aurantiaca vel aurea?) sarco-
pleura pulpacea, endopleura subfusca album ini
corneo adh®rente, chalaza interna annulari in
endopleura conspicua, extus vix discernenda.

Plantca occidentales radice phis minus tuberose t ;
stylo nescio an semper tripartibili.

Subordo 3. Agaves. — Schistandr® non operculos®.

§. 1. Dioscoreasformes. — Spica axillaris. Vena; folio rum
accuratius inspiciendce sunt. Plantoe scandentes,
dioicoe.

19. Tamus. — Periantliium profunde fissum, regulare,

patens, in feminis supra germen coarct.atum.
Pericarpium exopleura evalvi, sarcopleura pul-
pacea, endopleura valvis septigeris; semina albu-
mine corneo, fere ut Alstrcemeria.

20. Testudinaria. — Periantliium profunde fissum, regu-

lare, semipatens ; filamenta basi laciniarum
f 2

68

AMARYLLIDACE®.

inserta ; pericarp, membranac. valvis septigeris.
Semina complanata testa obscura. Plant an Afri-
can ce meridionales radice quadrate angulosa.

21. Dioscorea. — Perianthium sex- partition (laciniis toro

inodioliformi in disco insertis teste Salisbur.
Parad. Pond. 75) filamenta toro inserta alterna
interiora, teste Salisb. laciniarum basi, teste
Brown; (utri credendum est? D. bulbiferd ab
utroque designatd ) stylus in feminis profunde tri-
fid as ; ovarium 3-loculare loculis 2-3-spermis ;
capsula loculis 2 quandoque abortientibus.

Si reverd species sunt Dioscorea in quibus fila-
menta laciniis perianthii inseruntur, ad Testudina-
riarn transfer endee sunt, filamentis in Dioscorea toro
inodioliformi insertis. Si in omnibus filamenta laci-
niis, uti cl. Brown prcedicat, inseruntur, Testudi-
naria radice quadrate angulosa vix a Dioscorea ,
nescio qua alia ralione , secernitur.

Observ. — Vellcm characterem Tamiformibus et
Dioscoreceformibus ex foliorum venis suniere, cum in
Tamo communi et Dioscoreis plerisque prcecipuce
bifidee sint, confiuentes, et reticulata ; contra verb
in Diosc. lucidd simplices.

22. Rajania. — Perianthium semipatens profunde fissum,

supra germen in feminis coarctatum ; capsula
alata monosperma, loculis ovulisque caeteris abor-
tientibus. Capsula monosperma ala magna cur-
vatula. Semen subrotundum. Perianthium basi
indivisum esse intelligo. Pessimus est botanicorum
mos perianthium profunde fissum sexfidum aut tri-
fidum dicer e ; meliora sperare fas est.

§. 3. I xi.® formes. — Radix bulbosa ( pericarpium non pulp a-
ceum; semina albumine corneo, testa dura?)

23. Bravoa, v. s. — Caulis spicatus bracteatus ; germen

oblongum subtrigonum ; perianthium cylindri-
cum ore tubi ampliato, limbo laciniis brevissimis
subconniventibus, alterne minoribus, persistens ;
filamenta fundo tubi inserta, filiformia ; antherae
versatiles ; stylus incrassatus ; stigma triquetrum ;
capsula oblonga trilocularis trivalvis ; semina
duplici serie numerosa reniformia.

24. Ixiolirion, v. s.— Caulis pedunculis axillaribus vel

terminalibus bracteatis 1-2-floris; germen ob-

AM ARYLLIDACEjE.

69

longum erectum ; perianthium profunde fissum
(sexfidum non credo) regulare semipatens laciniis
alternis aequalibus ; filamenta recta basi lacinia-
rum inserta, alterna sequalia; antherae versatiles;
stylus rectus ; capsula oblonga striata trilocularis
trivalvis ; semina numerosa ovali-oblonga.

Tecopbilea? v. s. Radix parva bulbosa pyri-
formis indusiis filamentose membranaceis, folia
1-pauca, caule 1-paucifloro, germen oblongum,
perianthium profunde fissum (regulare ? semi-
patens ?) filamenta 3 fertilia, 3 abortientia, ste-
rilia. Teste cl. Bertero. Nescio an filamenta un-
<piam perfecta. Credo equidem potius plantam
esse ex Irideis vere triandram processibus tribus
interstamineis. Filamenta non vidi, neque
innotuit utra sint abortiva, petalina an sepalina.

C 3. Agaviformes. — Semibulbosoe vel caudicales, seminibus
complanatis testa riujra vel subfused, (filamentis
cequalibus ? )

25. Fourcroya. — Bulbus imperfectus aut caudex longae-

vus ; germen oblongum infra attenuatum cer-
nuum ; perianthium sexfidum regulare subpatu-
lum laciniis alternis aequalibus ; filamenta brevia
subulata basi obovate incrassata disco inserta
conniventia ; antherae versatiles ; stylus rectus
cavus triqueter infra strumosus stigma trigonum
fimbriatum ; capsula oblongo-ovalis 3-valvis
3-locularis, semina nigra numerosa. An pror-
sus sexfidum perianthium sit, inspicias.

26. Agave. — Bulbus imperfectus vel caudex longaevus;

perianthium tubatum limbo non patulo ; fila-
menta tubo adnata filiformia longa ; antherae
longae versatiles ; stylus triquetro - filiformis ;
capsula utrinque attenuata 3-valvis 3-locularis ;
semina nigra numerosa.

1. Germen cylindraceo-oblongum, pedunculatum hori-

zontale ; limbus infundibuliformis ; filamenta
fauce tubi inserta ; stigma capitatum vix loba-
tum. Genus Chloropsis? Spec, lurida. Bot. Mag.

2. Germen ovale sessile subhorizontale; limbus revolu-

tus ; filamenta basi laciniarum inserta ; stigma
inconspicuum. Genus Littaea ? Spec, gemini-
flora.

70

AMARYLLIDACEiE.

<§. 3. Germen oblongo-ovale perianthium erectum vel as-
surgens limbo filamentis adpresso; stigma tri-
lobum lobis emarginatis. Genus Agave? Spec.
Yirginica. In speciebus diversis flares et fructus
accuratius inspiciendi sunt.

27. Doryanthes. — Bulbus imperfectus, longaevus; cau-

lis capitulatus bracteatus ; perianthium deciduum
tubatum ( fide cl. Schultes et fly. Bot. Mag. tacen-
tibus cl. Sims et Brown ) infund ibuliforme laciniis
subaequalibus reflexis ; filamenta subulata basi
laciniis (ettubo?) adnata; anther® long® erectae
basi ex t inc tori i form i affix®; stylus trisulcus;
stigma trigonum; capsula turbinate ovata 3-locu-
laris 3-valvis ; semina subfusca foliaceo-rugosa.

Divisio Tertia.

Scapace®; i. e. scapo carnoso succulento spathaceo, infra

spatham non articulato.

Subordo 3. A m a rylli n e^e . — Schistandrce, non operculosce.

t Cav® ; filamentis petalinis pr®stantioribus. Scapus
cavus ; semina compressa testa nigra ; capsula
trilocularis trivalvis.

§. Cyrtanthiformes. — Germen ovale, pedunculatum,
perianthium tubulosum tubo ampliato laciniis
alterne ®qualibus. Plantce Africans.

28. Cyrtanthus. — Germen pendulum aut declinatum,

tubus curvatus, filamenta recta decurrentia tubi
parte superiore inserta, sepalina vix profundius,
anther® tertia parte superiore affix® suberect® ;
stylus incurvatus. A. foliis persistentibus.
B. foliis deciduis.

29. Gastronema. — Germen declinatum, tubus infra cur-

vatus gracilis cylindraceus, superne amplus cam-
panulatus, lacini® breves reflex®, filamenta de-
currentia conniventia, superiora tria longiora
incurvata, petalina summo sepalina medio fere
tubo inserta, anther® breves, stylus declinatus
petalo imo adpressus. Folds deciduis.

30. Vallota. — Germen erectum, tubus rectus amplus,

limbus infundibuliformis, filamenta conniventia
latere tantum tubo (petalina summo tubo vel
laciniis, sepalina profundius) adnata ; anther®

AMARYLLIDACE/E.

71

tertia parte inferiore affixse suberectae ; stylus
declinatus. Foliis persistentibus.

§. 2. Hippeastriformes. — Germen trigonum superne la-
tius medio constrictum ; tubus arctus. Plantce
occidentalcs.

1 . Sprekelia. — Per. pedunculatum declinatum tubo
subnullo laciniis superis reflexis inferis devexis
basi convolutis; filamenta pariter membrana con-
nectente ad basin perianthii inserta, fasciculata,
resurgenter (uti stylus) declinata, e petalinis sum-
mo e sepalinis imo caeteris longiore ; antherae
a tertia parte superiore pendulae. ( Stigma tri-
fidum)

32. Hippeastrum. — Per. pedunculatum declinatum tubo

infra oblique abbreviato, sepalo summo latiore,
petalo imo angustiore ; membrana faucialis, ubi
manifesta, non annularis, (infra scil. imperfecta,)
filamenta resurgenter declinata tubo gradatim
inserta, e petalinis imo breviore profundius, e
sepalinis summo longiore altiiis inserto ; an-
therae a tertia parte superiore pendulae; stylus
resurgenter declinatus. (Stigma trifidum aut
trigonum)

33. Phycella. — Per. tubiforme, tubo brevi laciniis con-

volutis apice reflexo ; membrana faucialis, ubi
manifesta, aequalis ; filamenta alterne aequalia
summo tubo pariter inserta decurrentia apice re-
curvata ; antherae breves incumbentes ; stylus
devexe porrectus, (stigma dilatatum apice fim-
briato)

Fustephia. Delenda, utpote falso delineata
diagnosi pessima. Nunquam tabs invenietur.
Proculdubio Phycellae species mendosissime de-
picta et descripta.

34. Habranthus. — Per. declinatum breviter tubatum

subcampanulatum laciniis alterne subaequalibus ;
membrana faucialis annularis ; filamenta summo
tubo pariter inserta resurgenter declinata fascicu-
lata ; antherae medio affixae incumbentes ; stylus
resurgenter declinatus. (Stigma trifidum ; cap-
sula turbinata.) A. Subbi-pluriflorae filamento
sepalino summo longiore. B. Uniflorre fila-
mento sepalino summo petalino imo sociis bre-

72

AMARYLLIDACEjE.

viore. — Si constanter, genusne diversum, Ze-
phyrites? mihi, ulterius inquirendum est.

35. Zephyranthes. (Scapus uniflorus ?) Per. suberec-

tum plus minus tubatum laciniis sub sole paten-
tioribus alterne subaequalibus ; filamenta pariter
basi laciniarum inserta conniventia alterne lon-
giora ; anther® lineares infra medium affix®
suberect®; pollen ovale l®ve ; stylus aversus.
A. Polline ovali l®vi, antheris linearibus stig-
mate t.rifido patulo. B. Pollen difforme scabrius-
culum, anther® erectiores post inversionern subu-
lat® apice recurvis, stylus aversus, stigma lobis
erectis ; (folia crassa cylindrica persistentia)
species Candida. Misceri, ut videtur, nequeunt.
Genusne diversum, Argyropsis? mihi, ulterius
inquirendum est.

36. Cooperia. — Germen erectum ; per. tubulosum tubo

cylindrico erecto (fauce nuda?) limbo (sub
nocte ?) patente laciniis alternis ®qualibus ; fila-
menta sub®qualia apice libera, a fauce tubi
pariter decurrentia ; anther® erect® non versa-
tiles tertia parte inferiore affix®, fasciculat® ; sty-
lus erectus ; pollen difforme. (Stigma obtuse
trilobum ; spatlia univalvis; scapus uniflorus;
folia linearia tortilia)

37. Pyrolirion. — Per. erectum tubulosum tubo cylin-

drico limbo ventricose campanulato laciniis apice
patentibus alterne ®qualibus ; [filamenta (longa)
fauce tubi inserta ®qualiter patentia? anther® a
parte superiore pendul® ?] stylus suberectus.
(Stigma trifidum ; germen oblongum, sessile ;
scapus uniflorus)

38. Haylockia. — Per. tubulosum, tubo cylindrico, limbo

semipatente laciniis alterne ®qualibus ; fila-
menta conniventia (brevia) laciniis, sepalina
profundius, inserta; anther® medio affix®, in-
cumbentes ; stylus erectus. (Stigma trifidum ;
germen subterraneum ; capsula turbinato- ro-
tunda ; semina dorso rotundato ; folia linearia)
f Solid®, scapo solido.

V 3. Oporanthiformes. — Scapus solidus; periantliium
non coronatum. Semina testacea?

39. Gethyllis. — Per. tubulosum tubo recto cylindrico

AMARYLLID ACE.E.

73

(saepius partim subterraneo) limbo subaequali stel-
lato ; filamenta subulata erecta tubo prope faucem
inserta a senis usque ad numerosa superflua;
antherae sagittatae erectae superne (plerumque ?)
spiraliter flexae ; stylus filiformis erectus inferne
cum tubo consolidatus ; capsula pulpa eduli tri-
locularis; semina parva rotunda. Plantce Afiri-
cance.

40. Sternebergia. — Per.tubulosum tubo erecto cylindrico
(saepe partim subterraneo) limbo semi-patenti ;
capsula trivalvis ; semina nigra nitida punctata,
(strophiolata ?). Africance, Europece. filamenta
filiformia, basi dilatata, erecta, alterna longiora ;
stylus superne crassior ; stigma bilobum.

4 1. Oporanthus. — Per. tubatum tubo erecto brevi subin-
fundibuliformi, limbo semipatente laciniis alternis
aequalibus ; filamenta tubo pari ter prope faucem
inserta, (scapus uniflorus ; spatha univalvis ;
germen sessile ovale complanatum rectum ; ovula
subrotunda compressione angularia magaritacea)
semina testa nigra ? Europece.

42. Lapiedra. — (Ex conjectura mea scapo solido, seminibus

testaceis ; si carnosis, post Carpolyzam locanda
erit .) Scapus pluriflorus ; spatha 2-3-phylla ;

per. stellatim patens regulare ; filamenta erecta ;
antherae sagittato-hastatae (incumbentes ? fideLa-
yascce) semina parva angularia fide Clusii. Si
porandra esse invenietur , post Leucojum statuenda
erit.

4. Pancratiformes. — Corona staminifera.

A. Semina testa nigra.

43. Tapeinanthus. — Perianthium profunde fissum, co-

rona brevis, filamenta patentia, antherae breves
incumbentes, stylus erectus ; (clavatus ?) stigma
obtusum. Planta bidbo pyriformi, folds gracilli-
mis, scapo prcecoce spatha monophylla. Tacente
cl. Cavanillesio, dubia est scapi soliditas.

44. Chlidanthus. — Per. tubulosum tubo cylindrico tri-

gono recto superne ampliato, limbo regulari se-
mipatente; filamenta apice incurvo libera, mem-
brana tenui dentibus sex alterne inaequalibus
tubo et petalorum parti inferiori adnata inserta,
antherae prope basin affixae ; stylus rectus : ovula

74

AM ARY LLI DACE.®.

multa opaca complanata biseriatim cumulata vix
imbricantia. Stigma trifidum, patens.

45. Clinanthus. — Per. tubulosum tubo infundibuliformi

(declinato?) limbo regulari brevi vix semipa-
ten te ; filamenta brevia subaequalia acuminate
alata (basi coliaerentia corona perfecta ? vel certe
contigua) antberae basi aftixae erectae. (Germen
subrotundum trisulcum ; folia linearia)

46. Urceolina. — Germen pendulum ovatum trisulcum;

tubus rectus subcylindricus ; limbus ventricose
campanulatus ; stigma dilatatum ; semina plura
nigra. (Folia lata petiolata.) Observ. Coronam
plus minus perfectam inesse ex scapo solido seminibus
nigris et mira Leperizoe ajfinitate conjicio, cui tamen
nomen Crinum a cl. Ruiz datum repugnat. Si nulla
inerunt corona vestigia, post Oporanthum ponenda
erit ; si corona fuerit, a Leperizd vix diversa inve-
nietur ; unde si genera consolidanda sint, nomen
Urceolina anteferendum est.

47. Leperiza. — Germen pendulum ovatum trisulcum ;

tubus rectus subcylindricus, limbus ventricose
campanulatus ; corona brevis ; stigma dilatatum ;
semina plura oblongo-subrotunda. (Folia lata
petiolata, bulbus ovatus)

Bulbum tunicis imperfectis squamafiormibus et
filamenta sinuosc fiexa esse ex fig. Flor. Per. non
ex Ruiz. text, apparet. Figuram, ut scape in Flor.
Per. mendacem esseputo.

48. Carpodetes. — Germen erectum pedunculatum ; tu-

bus curvatus gracilis cylindricus, limbus sub-
campanulatus, corona brevis, capsula obovata tri-
sulca medio constricta ; semina magna. (Folia
linearia, bulbus oblongus)

49. Coburghia. — Germen ovale trisulcum peduncu-

latum subcernuum ; per. tubulosum tubo curvato
cylindrico superne ampliato, limbo brevi regu-
lari semi-patente ; filamenta aequalia sub-recta
vix conniventia, corona tubata, antberae rectae,
stylus recurvatus, stigma obtusum ; ovula biseri-
atim confusa. (Folia linearia ; bulbus ovatus)

50. Stenomesson. — Tubus sub-rectus medio constrictus

superne ventricosus, limbus brevis regularis,
corona brevis, filamenta recta, antberae incum-

AMARYLLIDACEAS.

75

bentes ; capsula late ovata trisulca trivalvis basi
valde ampliata. (Folia lanceolata margine com-
presso ; bulbus subrotundus)

51. Eucrosia. — Tubus declinatus, antice abbreviatus ;

limbus sursum curvatus compressus ; corona de-
clinata concavo-rutelliformis basi cylindrica erec-
tiore ; filamenta infra dilatata complanata ; an-
ther® a tertia parte superiore pendul® polline
minore ; stigma dilatatum complanatum ; cap-
sula ovata trisulca. (Folia lata petiolata ; bulbus
late ovatus)

52. Elisena. — Per. tubatum tubo cylindrico, limbo re-

flexo laciniis duabus depressis ; corona declinata
(ima parte staminifera summa filamenta labro
comprehendens ?) genitalia declinata recurvata;
anther® breves incumbentes.

53. Pancratium.— Per. tubulosum tubo cylindrico la-

ciniis patentibus ; corona conspicua ; filamenta
rigidula conniventia ; anther® breves sub-erect®.
(Stigma simplex vel trigonam; flores albi, extus
subvirescentes)

A. Subsessiles, foliis persistentibus ; semina testacea
complanata dorso rotundato margine foliaceo hilo
albo. B. Peduculat® foliis deciduis ; scapus su-
perne cavus ; semina oblongo-rotunda testa du-
riore, raphe albicante. Genus Halmyra. Salis-
bury. C. Sessiles ; foliis deciduis. Seminibus
nondum aspectis dubito an genus sint a c®teris
decernendum, cum cl. Rlieede in H. Mai. semina
Catulli pol® albicantia carne albicante aquea
scripserit ; si vere, genus mihi Tiaranthus.

B. Semina carnosa, integumento molli inseparabili.

54. Hymenocallis. -Per. tubulosum, tubo recto angu-

late cylindrico, laciniis longis linearibus angustis
patulis ; filamenta gracilia distantia subconni-
ventia ; anther® long® lineares versatiles a
tertia parte inferiore pendul® ; pollen reticula-
tum scabrum ; germen sessile vel breviter pedun-
culatum angulatum ovulis suberectis ; semina
viridia oblonga capsulam immaturam s®pe rum-
pentia. (Folia persistentia ; bulbus et folia simul
ex semine oriuntur)

55. Choretis. — ' Tubus cylindricus subrectus, laciniis et

76

AMARYLLIDACE.E.

corona patulis ; filamentis distantibus sub-erecte
conniventibus ; antherae callositate quadam in
parte superiore affixae pendulae; semina oblongo-
rotunda (glaucescentia ?)

56. Ismene. — Tubus curvatulus cylindricus, laciniis an-

gustis linearibus patulis; filamenta conniventia
(tria superiora in coronam plerumque deflexe
curvata) antherae modicae ; semina viridia ro-
tunda. ‘ (Bulbus ovatus ; folia aestiva decidua ;
bulbils semine maturo mox ortus non ante aesta-
tem sequentem folia protrudit)

Callithauma? — Non satis notum. Bulbus cylin-
draceo-oblongus ; limbus patens; corona lacinias
aequans ; stamina exerta. Spec. P. viridiflorum.
Flor. Per. Ex bulbo cylindraceo-oblongo pal-
mari, scapo orgyali, nectario enormi, flore sma-
ragdino, generibus notis non convenire fretus
sum. Vera generis indicia latent.

57. Calostemma. — Tubus brevis costatus, laciniis sub-

aequalibus semipatentibus ; corona saepe fissa ;
filamenta brevia erecta ; antherae brevia erecta ;
stylus gracilis sensim attenuatus ; stigma minu-
tum ; germen costatum, defectu saepe biloculare ;
capsula dissepimentis obsoletis unilocularis 1-2-
sperma; semina viridia rotunda, latere uno com-
pressione complanato.

58. Vagaria. (Vix Calostemmatis sectio secunda.) — Tubus

subcylindricus angulatus ; limbus substellatus
laciniis lanceolatis; filamenta dentate alata coro-
nae sexties fissae speciem referentia ; stylus subu-
latus superne tenuissimus curvatulus ; stigma
parvulum. (Germen oblongum non costatum ;
folia linearia dorso convexo ; stigma trifidum ?)
Spec. Pane, parvifiorum, patria ignotd. Redout e.

59. Eurycles. — Tubus subcylindricus ; limbus brevis

laciniis lanceolatis ; corona saepius imperfecta
sexties fissa, filamentis basi alatis; antherae erectae
basi affixae ; semina rotundata bulbo ante matu-
ritatem (vel in ovulis) protruso. (Germen disse-
pimentis imperfectis ; folia aestiva decidua, petio-
lata, latissima)

§. 4. Amaryllidiformes. — Scapus solidus ; semina car-
nosa.

AMARYLLIDACE/E.

77

60. Griffinia. — Tubus declinatus cylindricus ; laciniae

reflexae, inferiores divaricatae, ima porrecta ; fila-
menta cum laciniis pariter in tubum coalescentia,
decurrentia, divaricata, recurvata; antherae incum-
bentes versatiles ; stylus recurvatus ante matu-
ritatem devexus; ovula parallelo-erecta, obovata,
latere exteriore erosa ; semina obovata nitida
albescentia chalaza fusca. (Folia lata, petiolata;
stigma simplex ; pedunculi *in seminando cur-
vati)

61. Lycoris. — Tubus obsolete trigonus ampliatus, sur-

sum ad faucem curvatus ; laciniae sursum cur-
vatae ; filamenta cum laciniis pariter in tubum
coalescentia, recurvata, alterna aequalia ; stylus
recurvatus; stigma acutum circumfimbriatum ;
ovula subrotunda. (Folia linearia ; capsula et
semina non visa ; capsula certe trivalvis)

62. Clivia. — Germen pendulum ; per. superne incurva-

tion, infra rectius ; tubo cylindrico ampliato ;
laciniis quater disparibus ; genitalia recta, fila-
menta alterne aequalia, tubo pariter decurrentia ;
pericarpium evalve exulcum coloratum sarco-
pleura molli pulpacea (spatlia plurivalvis ; folia
lorata rigida persistentia ; semina obovata hya-
lina ; bulbus imperfectus fibris carnosis).

63. Haemanthus. — Per. regulare tubo recto ; filamenta

tubo summo inserta, recta ; antherae breves, sub-
erectae ; stylus rectus attenuatus ; pericarpium
evalve exulcum [loculis monospermis dissepi-
mentis obsoletis, saepe coloratum] sarcopleura
molli pulpacea. (Stigma lobis tribus vel obso-
letis ; spatlia 3-multivalvis ; semina hyalina)

64. Buphane. — Tubus cylindricus; limbus regularis;

filamenta extra tubum inserta recta distantia;
stylus rectus ; ovula distincta; capsula turbinata,
trilocularis, trivalvis, trisulca. (Stigma simplex;
semina plura; viridia? spatha bifolia ^

65. Ammocharis. — Tubus trigone cylindricus ore am-

pliato (sepalorum marginibus non imbricantibus)
limbus regularis ; filamenta pariter fere ad laci-
niarum basin tubo adnata declinata recurvata ;
antherae breves medio fere affixae ; stylus decli-
natus recurvatus ; ovula biseriatim latere pla-

78

AM ARYL LID AC E/E.

cent® intermedi® adnata (semina, qu® vidi, dif-
formia, pollinis particul® Crini polline dimidio
minores ; stigma simplex ; folia basi non tubu-
losa.)

66. Crinum. — Germen medio crassius; tubus cylin-

dricus germine gracilior; filamenta vix extra
tubum pariter inserta recurvata, anther® incum-
bentes ; pericarpium molle evalve exulcum, dis-
sepimentis obsoletis; semina diftormia. (Folia
basi tubulosa)

67. Amaryllis. — Germen trigone obovatum ; tubus an-

guste infundibuliformis, laeini® reflex® ; fila-
menta petalina petalorum basi sepalina profundius
tubo adnata ; capsula trilocularis trivalvis obo-
vata (pr®cociter s®pius disrupta) semina sub-
globosa. (Folia hiemalia, sub-linearia, arcuata,
scapus autumnalis, multiflorus, pedunculis diva-
ricatis)

68. Brunsvigia. — Germen trigonum; tubus brevis aut

subnullus, perianthium et genitalia recurvata;
filamenta tubo annulari adnata ; capsula trigone
turbinata trilocularis trivalvis ; semina subro-
tunda. (Folia hiemalia, lata, recumbentia ;
scapus autumnalis multiflorus pedunculis diva-
ricatis ; spatha lata)

69. Nerine. — Tubus subnullus annularis; laeini® re-

flex® ; filamenta basi gibbosa monadelpha laci-
niarum basi adnata. (Folia lorato-linearia ;
scapus autumnalis; pollen minus; stigma de-
mum trifidum)

70. Strumaria. — Tubus subnullus annularis, filamenta

basi inter se connata, stylo plerumque vel omnia
vel alterna adnata; anther® incumbentes; stylus
strumosus angularis sulcatus; semina pauca sub-
globosa. (Umbella multiflora pedunculata; spatha
bivalvis ; folia lorato-linearia ; stigma trifidum.)

7 1 . Hessea. — Tubus brevis, limbus regularis semipatens ;

filamenta ®qualia subulata, erecta dein reflexa,
fauce tubi inserta ; anther® breves erect® basi
affix® ; stylus filiformis. (Folia lorato-linearia ;
spatha bifida ; umbella multiflora pedunculata ;
germen parvulum, mox pr®cociter inflatum
turbinatum)

amaryl.lidace.iE.

79

72. Imhofia. — Perianthium patens ; filamenta ad basin

perianthii (germini ipsi?) inserta subaequalia ;
stylus crassus vel strumosus, trisulcus; antlierae
breves erectae basi affixae ; semina viridia angu-
laria dorso rotundato. (Folia filiforniia ; scapus
autumnalis)

73. Carpolyza. — Perianthium tubo brevi infundibuli-

formi laciniis semipatentibus ; filamenta toti tubo
adnata; antlierae oblongae, erectae, basi affixae ;
stylus crassus sulcatus superne tenuior ; capsula
trilocularis trivalvis ; semina viridia dorso rotun-
dato. (Stigma trifidum reflexum fimbriatum ;
folia filiformia)

Subordo 5. Narcisse.e. — Scapaceae; schistandrae ; non oper-
culatae ; filamentis sepalinis praestantioribus co-
rona inclusis.

Observ.- — Scapus plus minus superne fistulatus
infra medulla repletus ; pedunculi solidi ; semina
testa nigra; coronce , ubi triloba cst, fissure preci-
puce costis sepalinis, loborum crence petalinis, oppo-
site; sepala, ubi disparia ( semper , ni fallor') petalis
latiora ; ovula 2-4 seriebus sepius imperfectis con-
flicts ; stylus scepe tripartibilis , casu aliquando tri-
partite.

74. Corbularia. — Stylus ut filamenta declinatus recur-

vatus ; sepalina basi tubi inserta, petalina prope
basin adnata; antlierae medio affixae, breves, ver-
satiles. (Tubus et corona infundibuliformes.
Corona tubo longior)

75. Ajax. — Stylus rectus subulatus ; filamenta recta tubi

parti inferiori adnata; antherae infra medium
affixae longae, lineares, marginibus retroflexis fila-
mentum summum cooperientes. (Corona tubo
longior)

76. Ganymedes. — Stylus rectus gracilis ; filamenta tubi

parti superiori inaequaliter adnata; (majore quam
in Queltia disparitate, sepalina magis producta)
limbus absolute retroflexus. (Corona tubo vix
aut non longior)

77. Queltia. — Stylus rectus superne attenuatus ; fila-

menta tubi parti superiori inaequaliter adnata,
recta ; antlierae lineares recurvatae marginibus

80

AIMAR YLLIDACEiE.

sunimis retro concurrentibus filamentum sum-
mum non cooperientes. (Corona tubo brevior)

78. Narcissus.— Stylus rectus gracilis, filamenta recta

apice tantum libera, in®qualiter tubo ad et prope
faucem adnata ; antherse infra medium afRxae
breves suberect® apice recurvat® marginibus
non concurrentibus. (Corona tubo quater fere
brevior)

79. Hermione. — Stylus rectus gracilis, filamenta apice

tantum libera conniventia, in®qualiter tubo ad
et prope faucem adnata ; anther® medio affix®
acute ovales incumbentes. (Corona tubo bre-
vior)

Subordo G. Galanthe®. — Porandr® ; non operculat® ; sca-
pace®. (Genuine pendulo, filamentis ®qualibus ;
polline minutissimo)

<§. 1. Scapus solidus ; semina subalbida.

80. Galanthus. — Perianthium sexpartitum sepalis conca-

vis, petalis brevioribus emarginatis; filamenta
libera germinis toro inserta subulata; anther®
erect® basi affix®, poris terminalibus, reflexe
apiculat®. (Spatha monophylla 1 -flora uno latere
fissa, altero translucens)

81. Acis. — Perianthium (sexpartitum?) laciniis alterne

paribus ; filamenta toro germinis inserta ; an-
ther® convergenter erect® basi affix®, poris su-
perne lateralibus; stylus superne attenuatus;
stigma parvulum fimbriatum. (Folia filiformia)

82. Erinosma. — Perianthium sexpartitum ; lacini® con-

similes ; filamenta brevia libera erecta toro ger-
minis inserta; anther® erect® basi affix® poris
rotundis terminalibus ; stylus crasse clavatus ;
stigma triangulare acute attenuatum; pericarpium
turbinatum evalve ; semina albescentia. (Folia
lorato-linearia ; spatha monophylla, uno latere
superne fissa, altero translucens)

§. 2. Scapus cavus : semina testa nigra.

83. Leucojum. — Scapus marginato-anceps ; spatha mo-

nophylla, profunde fissa marcescens ;p erianthium
sexpartitum, laciniis sub®qualibus(apicem versus
angustatis) filamenta erecta toro germinis inserta;

AMARY LLI DACE/E.

81

antherae erectae, basi affixae poris terminal, et
latere dehiscentes; stylus tenuis clavatus; stigma
triangulare acute attenuatum ; capsula trigona
trilocularis trivalvis.

Subordo 7. Tacceje? — Bulbo et foliis Aroideis. Filamentis
superne cucullatis antheras superantibus.

A me ad subspadiceas relatus ob perianthium vix, ni
fallor, petaloideum, affinitate non dubia ad Orontaceas
spectans. Flores nunquam aspexi.

Tacca. — Per. regulare, persistens ; filamenta basi
laciniarum inserta infra dilatata ; antherae cu-
cullae intus adnatae ; ovarium 1-loculare placentis
3 parigtalibus polyspermum.

Ataccia? — Prezl.

First Division. — BRANCHED.

Suborder 1. Xerophyte^:. — Scliistandrous, i.e. anthers open-
ing laterally and completely.

1. Xerophyta. — Perianth tubed, limb regular persis-

tent ; filaments inserted at the base of the segments ;
anthers long, linear, subsessile ; stigma oblong, di-
lated; capsule rough, 3-celled; seeds numerous.
(Style and stamens straight?)

A suffruticose plant, with small alternate branches, harshly
clothed by the dry imbricating sheaths of the old leaves ;
leaves rigid, acute, like those of pinus, crowded at the ends
of the branches; flowers terminal, generally solitary. Native
of Madagascar, only known by Commerson’s dry specimen.

2. Vellosia. — Perianth eampanulate, scarcely tubed ;

stamens sometimes 12, 15, 18, or 24, connected in
3 or 6 groups ; filaments short ; anthers erect ; (pollen
globular?) style straight tripartible; capsule 3-celled,
3-valved ; valves at the top, horizontal or imperfect,
(probably meaning a transverse fissure below the
opercle as in Hypoxis) seeds square, cuneate, with a
single shell.

1. Squamata. — Pohl Braz. ic. v. i. t. 9. flowers blue. —
2. Ramosa. Martius. Braz. t. 6. blue. — 3. Aloaefolia.
ib. t. 7. blue. — 4. Asperula, ib. t. 8. blue. — 5.Plicata,
ib. t. 9. white.

o

82

AMARYLLI DACEiE.

Tliis genus is unknown to me, but amply describe^ by Dr.
Martins. They are natives of Brazil, having a habit between
Barbacenia and Fourcroya, and give a peculiar appearance
to the country where they abound, the largest acquiring some
resemblance to the form of a Yucca, with less rigid leaves, and
flowers which are either white or blue, borne on a single stalk.
Yellosia has the same remarkable peculiarity as Gethyllis,
producing an occasional superfluity of stamens in groups.

3. Barbacenia — Germen erect, perianth continuous tubed;
filaments bifid longer than the anthers ; style acumi-
nate longer than the stigma ; seeds small, numerous,
angular by contact with a bay shell margined with
white. PI. 1. f. 29. (Limb reflex. Inhabit Brazil.)

1. Purpurea. — Bot. Mag. 54. 2777. — Flowers purple. —
2. Tricolor. Mart. Braz. t. 10. orange ; branching.
— 3. Tomentosa. ib. t. 11. orange; branching. —
4. Longiflora. ib. t. 12. yellow. — 5. Bicolor, ib. t. 13.
yellow, green without ; branching. — 6. Rubro-vi-
rens. ib. t. 14. f. 1. dull red.— 7. E'xscapa. ib. t.
14. f. 2. yellow; scarcely branching.

These plants have great resemblance to Yellosia, though
of very inferior stature. I believe that no species but pur-
purea has been introduced into this country. It was raised
by me from seeds concealed in the small quantity of vegetable
mould and moss adhering to some epiphytes which had been
collected in the neighbourhood of Rio Janeiro, which it is
always advisable to save, and spread upon a pot full of
earth, for it will generally be found that some seeds are
lurking in it, and probably better preserved than if they had
been gathered intentionally.

It is immaterial, as far as concerns the Amaryllidaceous
plants, whether the branching, the caulescent, and the scapa-
ceous divisions be considered as three separate orders, or as
portions of one; but they should not be separated without
carefully examining how far a similar limitation would be
desirable in other orders, and whether it is not more conve-
nient, for the purpose of giving a clear and easy view of the
whole vegetable creation, to reduce the number of orders by
converting some of those now defined into suborders, than
to multiply them by further dismemberment ; I entertain no
doubt upon the subject, and I consider that Dr. Lind ley’s

AMARYLLIDACEA2.

S3

alliance Narcissales should be dropped, and the whole hex-
androus hexapetaloid portion of the monocotyledonous plants
included under the alliance Liliales, the triandrous under
Ixiales. But the determination of this, which is a discre-
tionary point of convenience, should be postponed, till a final
and satisfactory arrangement of the entire system shall have
been effected, when it should be made with a view to the
consistency of the whole.

Second Division. — CAULESCENT.

Suborder 1. LIypoxide^e. — Schistandrous. Operculous, i. e.

the base of the style forming a prominent portion
of the ovary.

§. 1. Hypoxidiformes. — Sepals and petals conformable;
leaves plicate. There is a great disposition to
hairiness of the exterior of the germen and flower
in this section.

Weldenia ? — If truly an epigynous hexapetaloid
flower depauperated, or in an imperfect state,
which I scarcely believe ; it rests on the report
of Karwinsky, who has omitted to state whether the
germen is superior or inferior. Perhaps it may be
allied to Commelina.

Perianth long-tubed ; tube erect ; limb patent,
tripartite; (whether the sepals or the petals are abor-
tive, is not stated) filaments inserted at the mouth
of the tube, the alternate shorter ; anthers erect,
affixed at the base between the cells; style filiform,
erect ; stigma triangularly capitate.

1. Candida. — Schult. fil. in Bot. Zeit. 1829. p. 1. 1. 1.

4. Curculigo. — Germen erect, bracteate, nearly subterra-
neous ; tube adhering to the style, cylindrical
upwards ; limb regular, patent ; filaments short,
inserted in the mouth of the tube ; anthers distant,
erect; seeds black, pendant by a thick white chord
filling a large indenture between the umbilicus and
foramen; umbilicus conspicuously beaked, proceeding
from the side. (Leaves broadest near the base ?)

1. Orchioides. Bot. Mag. 27. 1076. Roxb. Cor. 1. 13. —
2. Plicata. Jacq. H. Sch. 1. 88. — var. glabra.
Bot. Reg. 4. 345. Forbesia plicata. Ecklon topog.
g 2

84

AMARYLLIDACE7E.

— 3. Brevifolia. Dryand. Hort. Kew. ed. 2. — 4.

Ensifolia. Brown Prod. 290. Australia. — 5. Stans.

Schultes. — 6. Latifolia. Bot. Mag. 46. 2034. Bot.

Reg. 9. 754. Flowers villous externally.

The ovules of latifolia are said to be round ; if the seeds
should prove not to have the extraordinary beaked umbilicus
of the genus, it can scarcely be a Curculigo; but it may have
been overlooked in the ovules or not yet developed. We
have no account of its seed. Gaertner describes the fruit of
C. orchioides to have as many cells as there maybe seeds, one
above another ; but such are perhaps rather membranous
envelopements. I have not perceived them, nor have I ever
found in the capsule an upper seed, as stated by him, without
the prominent umbilicus.

5. Molineria. — Stalk curved at the top, capitulate, many-

flowered ; germen nodding, bracteate ; tube adhering
to the style, short ; limb regular, patent ; filaments
short, inserted at the mouth of the tube ; anthers
fasciculate, erect (said to be monadelphous or united),
style straight ; stigma dilated ; seeds red, wrinkled,
round ; outer integument brittle ; inner brown, crus-
taceous. I have taken the name f rom Colla Hort. Rip.
who has applied it to a plant allied to Curculigo, with a
character resting on points of no value, except that on
inspection of the ovules he found no appendage ; meaning ,
I suppose, no rostrate umbilicus. I may, therefore,
assume that his plant was one of the species with round
seeds, which can scarcely belong to Curculigo. ( Leaves
long, petiolated. These plants have the outside of the
germen and perianth hairy.')

1. Recurvata. Bot. Reg. 9. 770. — 2. Sumatrana. Lodd.
Bot. Cab. 443. — 3. Capitulata. Leucojum capitula-
tum. R. et Schultes. Lour. C. Ch. 246. Leaves wide,
petiolated, plicate ; petiole a foot long ; head of
flowers large, roundish ; scape short, radical, reflex;
spathes, or rather bractes, 1 -flowered; segments
adhering together at the base, thickened at the
point, golden within, brown and villous without;
stamens six, short, equal ; stigma simple. It seems
strange that a plant so described should have been
mistaken for a Leucojum.

6. Hypoxis. — Germen erect ; perianth deeply divided with

AMARYLLIDACEiE.

85

an annular base; expanding in the sun, persistent;
filaments short, subulate, inserted in the disk; anthers
sagittate, erect, affixed between the cells ; style short,
three-furrowed, tripartible down to the opercle ; stigma
short, erect, fimbriated ; capsule 3-celled, 3-valved,
the valves breaking transversely from the opercle,
parting laterally upwards, united below ; seeds small,
black, the foramen a little pointed, the umbilicus
lateral, minutely beaked.

Flowers yellow. American. 1 . Erecta. — 2. Caroliniensis.
— 3. Juncea. — 4. Mexicana. — 5. Pusilla. — 6. Hu-
milis. — 7. Elongata. — 8. Breviscapa. — 9. Decum-
bens. — 10. Scorzonersefolia. — 11. Gracilis. — Afri-
can. 12. Sobolifera. — 13. Villosa. — 14. Obliqua.
— 15. Obtusa. — 16. Stellipilis. Bot. Reg. 663. — 17.
Tomentosa. — 18. Angustifolia.- — 19. Graminea. —
20. Serrata. — 21. Linearis. — 22. Tabularis. — 23.
Juncea. Ecklon. — 24. Luzulaefolia. — 25. Veratri-
folia. — 26. Ovata. — Australian. 27. Pratensis. —
28. Hygrometrica. — 29. Marginata. — 30. Glabella.
— Asiatic. 31. Minor. — 32. Aurea. — 33. Minuta.
Flowers yellow , white, or blue ; Cape of Good Hope.
34. Affinis. — 35. Pumila. — 36. Alba.= — 37. Dubia.
— 38. Aquatica. — 39. Stellata. — 40. Tridentata. —
41. Caulescens. — 42. Elata.

F)oubtful Species.

43. Subtrigyna. Willd. — 44. Scabra. Lodd. Cat. —
Flowers blue. 45. Geniculata. — 46. Acuminata. —
47. Laxa. Ecklon. — Flowers yelloiv. Flavescens,
Longifolia, Aurea, Tenuifdlia, Flavipetala, Fili-
folia. Ecklon. — Flowers white. Minor. Spatha-
cea. — Of Ecklon’s species little is known to me
besides the names. The species require more
careful investigation than they have yet received.
It is difficult to believe that stellata with an annual
tuberous root should be of the same genus as those
with persistent root and leaves. I have never seen
the fruit of stellata.

Ccelanttius. — Scape few-flowered, nearly enclosed in
a sheathing leaf; germen sessile; limb deeply cut
(probably not absolutely six-cleft), campanulate, con-
niving ; filaments inserted at the base of the seg-

86

AMARYLLIDACE.E.

ments ; anthers versatile. — An obscure genus resting
on the report of Schlechtendal, Reliq. Willd. — One
only species; native country unknown, perhaps the
Cape of Good Hope.

1. Amplificatus. — Herbar. Willd. Bulb of the size of a

small nut ; scape 3 inches ; sheathing leaf lanceo-
late, acuminate, undulate, plicate, with the point
bent downwards ; flowers sessile, with long acumi-
nate slender bractes.

2. LanarijEformes (or a suborder Lanarieae). — Leaves

equitant ? Flower more or less downy.

8. Anigozanthus. — Perianth downy ; tube long, curved ;

limb more deeply cleft underneath ; segments nearly
equal, standing apart underneath ; filaments dilated
at the base, short, bent, inserted at the mouth of the
tube ; anthers erect, adhering longitudinally ; style
curved, deciduous ; stigma clavate ; capsule 3-celled,
dehiscent by the opercle ; seeds numerous, obovate,
black.

1. Rufa. — Labill. voy. 1. 411 — 22. Nov. Holl. 2. 119.

Stalk downy, anthers not bearded.

2. Flavida. — Red. lil. 176. Bot. Mag. 29. 1151. grandi-

flora. Par. Lond. 97. Stalk smooth, anthers with
a reflex slender point. Natives of Australia.

9. Lanaria. — Perianth tubed, persistent; limb regular; fila-

ments inserted at the base of the segments ; anthers
versatile; capsule 3-celled; cells 2-3-seeded.

1. Plumosa. Willdenow. — Argolasia. Jussieu. — Native
of the Cape. Perianth downy without.

Obs. — At this point the epigynous plants make a close
approximation to the feathery-stamened Asphodeleae through
Narthecium, which would stand between Lanaria and Lo-
pliiola, if its ovary were not absolutely superior.

10. Lophiola. — Perianth deeply cleft, regular, reflex ; fila-
ments erectly patent; anthers oblong, recumbent;
style tripartible to the opercle; stigma simple; opercle
prominent ; very little of the germen inferior ; cap-
sule ovately pyramidal, 3-celled ; seeds many, in two
rows.

1. Aurea. Bot. Mag. 39. 1596. Conostylis Americana.
Pursh FI. Am. Sept. 1. 224. Disk downy.

AMARYLLIDACEiE.

87

11. Conostylis. — Perianth downy, deeply cleft, regular,

semipatent, persistent ; filaments short, erect ; an-
thers erect ; style conically dilated, hollow, tripar-
tible, persistent; stigma short; capsule dehiscent in
the opercle ; seeds numerous.

1. Aculeata. Bot. Mao;. 57. 2989. — 2. Serrulata. Brown
Prod. 300. — 3. Setigera, ib. — 4. Breviscapa, ib.
301. — The perianth is downy within in the two last
species.

12. Phlebocarya.- — Perianth deeply cleft, persistent; fila-

ments inserted at the base of the segments ; anthers
quadrangular, subsessile ; style filiform ; stigma sim-
ple ; ovary 1 -celled by defect; 3-seeded ; capsule
1 -seeded.

1. Ciliata. Brown. Prod. 301. Native of Australia.

13. Campynema. — Perianth regular, deeply cleft, persistent,

alternate segments equal ; filaments recurved ; an-
thers versatile ; styles distinct, subulate, recurved ;
capsule 3-celled, tripartible, dehiscent inwardly at the
points ; seeds numerous, flat, with a spongy shell.

1. Linearis. Labill. Nov. Holl. 1. 93. t. 121. Root fas-
ciculate, stalk a foot or more, leaves linear gra-
mineous smooth, those on the stem half embracing
it; flowers 1-4, terminal; stigmas simple (unless
the stigmas are long and mistaken for styles). Cape
Van Dieman. Seeds glossy rufous.

§. 3. ALSTR(EMERiiEFORMES (or a suborder Alstrcemeriese).
Sepals differing in form from the petals. Petaline
filaments exceeding. (In this section there is a
disposition to hairiness of the filaments, almost
obsolete.)

14. Chceradodia. — Root fibrous; leaves both radical, and

borne on the stalk ; stalk erect ; umbel many-
flowered ; sepals and petals very unequal ; style
strumous; capsule triangular. — This genus is not
sufficiently known ; if it should prove not to be
operculous, it would stand next before Fourcroya.

1. Chilensis. — Strumaria Chilensis - Molina, Saggio s.
Chil. Sepals and petals very unequal, the one white,
the other white tipped with red. Stalk erect, 5 or
G feet high, bearing 3 or 4 small alternate leaves,

88

AM A RYLLIDACEiE.

which embrace it ; radical leaves many, 2 feet long,
smooth, acute, of a fine green. Flowers said to re-
semble Albuca. Called in Chili Thekel. A cold
infusion of its leaves is purgative and diuretic.

15. Alstrcemeria. — Germen six-ribbed, above 6- or
12-angled, with a 3-lobed point; ovules suberect;
perianth with fourfold disparity divided to the base ;
petals 2 upper erect, the lowest stretched forward ;
filaments successively curved upwards, inserted in
deciduous glands on the germen, scarcely adhering
at the back to the segments, ultimately distinct ; the
petaline glands marking the disk with an acute oval
scar, the sepaline with a scar exceeding a semicircle
in breadth ; anthers attached at their base ; style
ultimately recurved ; capsule roundish, more or less
oblong, acutely operculate, 6-ribbed, with crusta-
ceous valves bearing the dissepiments, springing
asunder by the axis splitting from the base into
three parts, the lower half of the ribs remaining
inseparable from the peduncle ; seeds roundish, with
a tubercled shell not easily separated ; hilum smooth;
chalaza circular; inner coat inseparable from the
horny albumen.

Occidental plants ; as far as I knoic with an erect stalk,
leafy or scaly, peduncles one or more flowered, usually hrac-
teate.

1. Peduncles 1 -flowered; flowers a little incurved, tube-
formed, narrow ; upper sepal longer. Brazilian
plants.

1. Psittacina. — Bot. Mag. 57. 3033. Bot. Reg. 18. 1540.

Sweet Brit. fl. g. 2. 15. Barren stem subrosulace-
ous. Flowers red, tipped with green spotted.

2. Isabellana. — Spec. M. Isabelle. Herb. Hooker. Coptis

19. S. Brazil. From the province of Rio Grande
do Sul. Stem a foot ; leaves remote (four), sessile,
suberect, lance-linear, less upwards on the stalk,
rather stiff ; involucral bractes very small ; pe-
duncles 1 1 , simple, 1 ^ inch long ; perianth in-
curved, tube-formed, 1^ inch long (yellow tipped
with green ?) ; sepals wider, subspatulate, sub-
acute ; petals narrowly spatulate, acuminate ; bar-

JSWedtiell.se.

AMARYLLIDACEAE. 89

ren stalk not known. It seems to be allied to
Monticola. PI. 6. f. 4 — 6.

3? Monticola. — Martins ap. Schultes. Stalk 2 feet,
smooth ; leaves lance-linear, remote ; sessile, glau-
cous underneath ; peduncles 6, simple ; sepals
(red ?) ; spatulate with a point obsoletely crenu-
late ; petals lanceolate acute, the upper a little
longer with purple lines, 19 or 20 lines long. Bar-
ren stem not known. Found on Serra dos Lagos,
in the province of Bahia.

4? Plantaginea. — Martius ap. Schultes. Leaves nar-
row, lanceolate, smooth, erect, resupinate (lance-
ovate on the rosulaceous barren stem) ; involucral
bractes many, longer than the peduncles ; sepals
obovate, spatulate, obtuse ; petals obovate, oblong,
apiculate ; colour uncertain ; appeared not to be
spotted. On high hills of Minas Geraes.

2. Peduncles one-flowered; upper petals erect; lower one
abbreviated. Chiefly Brazilian plants.

5. Caryophyllaea. — Jacquin. H. Sch. 4. 465. A. ligtu.
Bot. Mag. 4. 125. Barren stalk rosulaceous; fer-
tile scaled; involucral bractes leaf-like; upper petals
spatulate, with a long slender point, white, some-
times lined and speckled with red and the point
red ; lowest petal almost obsolete ; sepals crimson
upwards, tipped with white, the upper much
longest. Brazil. This plant is generally known to
cultivators by the erroneous name of ligtu, given
through great inattention by Mr. Curtis, for it has
no affinity to the ligtu of Feuillet, which he states
to be a Chilian plant. In R. and Sch. the name
Curtisiana is substituted ; but I can see no reason
for that, since the error had been long ago set right
by Jacquin, who named it Caryophyllaea on ac-
count of its fine fragrance, and there can be no
doubt at all as to the identity of the plants. It is
easily cultivated in the stove, requiring absolute
drought in the autumn and early winter, and will
send up flower stems as soon as it is started in the
very early spring, if placed on a hot flue and
abundantlv watered. When it has done flower-
ing, it may be removed into a cooler situation. Its

90

AMARYLLIDACEi'E.

time of flowering will depend upon the time of
ceasing and recommencing to water it. The bar-
ren stalks follow the flower stems. It varies in
our stoves with brighter coloured flowers, and is
probably capable of much improvement from seed.

Var. 2. Catharinensis. — PI. 2. fig. 6. Mrs. Graham, Rio ;
and also Tweedie, St. Catharine’s, Brazil, Herb.
Hooker. Involucral bractes very slender ; petals
spotted upwards.

6. Macraeana. — Spec. Valparaiso. Macrae. Feb. 1825,

Herb. Lindl. Leaves crowded at the base, sub-
erect, lance-oval, 3 inches long, besides the petiole
lj, smaller upwards on the stalk; involucral
bractes three, narrow, 1^ inch long; peduncles
erect, above an inch long ; flowers suberect, lowest
petal abbreviated, upper longest and narrow ; la-
teral sepals broadest. This Chilian plant has very
much the foliage of Caryophyllaea, but bears its
flowers on the leafly stalk ; it has a like abbrevia-
tion of the lower petal, but appears to differ much
in the form of the segments. It will prove a green-
house or hardy plant, whenever it may be intro-
duced.

7. Inodora. — PI. 2. fig. 1. Tweedie Herb. Hooker. In

crevices of rocks near Solta, S. Brazil. Stalk
slender, leaves about half an inch apart, oval, 1^
inch long besides the petiole f ; involucral bractes
leaf-like with short petioles; peduncle |- , upper
petals narrow, spatulate, with an acute point,
spotted, 1|- long, lowest about f ; sepals more
obtuse. (Colour seemingly red, yellow, and

green ?) Mr. Tweedie writes, that on its first
shooting, this plant has much the appearance of
Caryophyllaea, but as it advances, becomes very
different ; that its tubers are less palmate, and its
flowers scentless.

8. Longistaminea. — Martins ap. Schultes. Stalk

smooth, 20 inches high, scaled below ; leaves
lance-linear, acute, resupinate ; involucral bractes
obsolete ; peduncles 5 ; sepals oblique ovate, petals
less oblong, lanceolate, with a slender point, the
lowest least (colour roseate ? not spotted).

AMARYLLIDACEiE.

91

3. Peduncles 1 and \ -2 flowered ; leaves short, linear ; peri-

anth small, I’efiex.

9. Linifolia. — Humb. Kuntli 3. 282. Stalks not a foot

high, purple, densely leaved, scaled below ; leaves
scattered, rather obtuse and coriaceous, 5-nerved,
8-9 lines long ; petioles short, thick, hairy ; sepals
nearly equal, red ; outside downy ; petals slen-
derer, just longer, yellow spotted with red, the
lowest petal protruded ; segments reflex ; style
downy. On the cold parts of the Paramo de Al-
maguer, between the river Marmato and the con-
fines of Puntaurcu.

10. Revoluta. — PI. 7. fig. 9. Flor. Peruv. ubsque
icone. Poeppig fragm. syn. p. 7. Stalk erect,
12-13 inches, scaled below in the specimen ; leaves
scattered, erect, linear, scarce an inch long; pe-
rianth purple, reflex ; petals less than the sepals,
the two upper yellow, spotted with purple below
the middle; peduncles 1-2 flowered? Poeppig
says leaves glaucous, peduncles 6-9, 2-flowered,
leaves very fleshy, colour of the flower varying
between rose-coloured and purple ; segments re-
flex, not revolute. In stony Alpine fields by the
torrent Rucue in S. Chili, flowering in December.

4. Peduncles \ -3-flower ed ; leaves resupinate.

a. leaves downy.

11. Foliosa. — Martius ap. Schultes. Stalk 1-2 feet;
leaves lance-linear, underneath glaucous with thick
down; peduncles 5-9, 1-2 flowered, sepals spatu-
late, scarlet with a white tip; petals red -lead
colour, with a scarlet tip and purple lines.

Var. 2. Leaves narrower. Var. 3. Stalk shorter, leaves
crowded on the middle of the stalk.

b. leaves smooth : barren stalk not rosulaceous.

12. Peregrina. — Pelegrina. Feuillet Obs. Bot. Mag. 4.
139. Peduncles 1-flowered. Too well known to
need particular description here. Its leaves are
resupinate and very fleshy ; its flower remarkable
by the high shoulders of its sepals, which are ob-
liquely cross-nerved. This plant will live out of
doors against the front of a green-house, having
some saw-dust heaped upon it in winter. It re-
quires an airy situation, if sheltered under glass.

92

AMARYLLIDACEAS.

Var. 2. Peduncles 2-3-flowered. There is much sport-
ing in the different specimens, not worth particu-
larizing.

Var. 3. Albescens. — PI. 3. fig. 4. Cmnming 568. Val-
paraiso. Herb. Hooker. Peduncles 1 -flowered ;
the shoulders of the sepals white, the middle green;
the petals marked with purple: stalk robust, 15-16
inches.

Var. 4. Squammata. — A. M. 317. Valparaiso, Sept.
Steep cliffs near the sea. Herb. Hooker. Sharp
scales instead of leaves.

Var. 5. Flore albo ; hortensis. A beautiful white and
green variety raised from seed in England : white
and red seedlings are both raised from its seed :
the white variety is tenderer than the red.

13. Ligtu. — Feuillet. Obs. 710. Described and figured
by Feuillet, who brought the first plant of this genus
before the public, and stated by him to grow near
Conception in Chili. The name was by a great and
unaccountable error in the Bot. Mag. applied to a
tropical plant, in no manner resembling it, which
has been generally cultivated under a wrong name,
being properly A. Caryophyllsea of Jacquin. Feuil-
let’s plant has the leaves 2-3 inches long, 5-16ths
wide, nearly sessile, peduncles bracteate 3-flowered,
flowers large, of a fine purplish red, streaked lon-
gitudinally on the two upper petals with white ;
sepals not auriculate or shouldered like those of
peregrina. Jacquin says that he had once in cul-
tivation a plant which nearly answered the descrip-
tion. It appears by the specimens under Ruiz’s
name lineatiflora, to admit variations of the same
nature and extent as those which occur in pere-
grina ; leaves broad and narrow ; peduncles one or
more flowered ; but it is readily distinguished from
that plant by the different form of the sepals.

Var. 2. lineatiflora. — Flor. Peruv. 3. 60. 289. Peduncles
2-3-flowered ; leaves ovate acute, 1 inch and J
wide ; the purple of the flowers deep. Ruiz’s spe-
cimen, Herb. Lambert, has very broad acute leaves
and bractes, and the sepals not auriculate. I find
a one-flowered variety in Dr. Lindley’s Herba-
rium, with leaves little broader than those of pere-

AM ARYLLID ACE^E.

93

grina, and varieties with 2-flowered peduncles, both
broad and narrow leaved.

Var. 3. Cumming 568, near Valparaiso. Herb. Lindl.
Leaves broad; peduncles 1 -flowered, short, thick.

Var. 4. Macrae. Conception. Herb. Lindl. Leaves nar-
row; peduncles 1 -flowered, short.

14. Pulchra. — Bot. Mag. 50. 2421. Tricolor. Hook. Ex.
fl. 63. Flos martini. Bot. Reg. 9. 731. Leaves sub-
glaucous, acute, resupinate ; peduncles 2-3 flower-
ed ; perianth white, tipped with red and green ;
upper petals blotched and streaked with purple,
having a bright transverse band of yellow. Petals
spatidate, shortly acuminate ; style and filaments
white ; anthers yellow ; pollen straw - coloured.
Near Valparaiso ; Cumming. In shady places on
the gravelly hills near Concon in N. Chili; flower-
ing in October; Poeppig. Fragm. Syn. [Pe-
duncles erect, often 9 inches long, five to the fork.]
Var. hort. Bicolor. Loddiges Bot. Cab. 1497.
Perianth white with green tips ; upper petals
broader and shorter than in the former (from the
seed of which it was raised) with a deep yellow
blotch streaked with green. Another two-coloured
seedling, without any purple, was raised by Mr.
Loddiges from the seed of bicolor, which had the
upper petals intermediate in length and breadth
between those of bicolor and the original pulchra.
The parent of bicolor stood near the white variety
of peregrina, but there seem no features in it that
could justify a belief that it was obtained by a cross
with that plant. It has the foliage and habits of
pulchra, and seems to be merely a beautiful semi-
nal variety. It is very fertile.

A. pulchra is very distinct from peregrina in root and
habit, as well as in flower and foliage. Its tubers endure a
very long period of rest, and bear our winters in a dry situa-
tion without any covering, but it is very difficult to preserve
the young shoots from the bite of slugs in the spring on their
fisrt appearance. Its leaves are very thin, those of pere-
grina fleshy and shining; its sepals are not auriculate. Its
root is disposed to run like couch-grass, and it bears tubers
on the lower part of the stalk, which is not the case with

94

AMARYLLIDACEvE.

peregrina. The name pulchra has the priority over tricolor
and flos martini. The substitution of the latter name cannot
be admitted ; the Chilian name, flower of St. Martin, does
not even seem to be confined to this species.

5. Peduncles \-Jlowered ; leaves spatulate, not resupi-
pinate.

15. Spathulata. — PI. 1. f. 49. Presl. Rel. Hank. 2. 122.
Specim. Andes of Chili and Mendoza, with three
flowers. Gillies. Herb. Hooker. The leaves of the
specimen are spatulate with a small point, 1 of
an inch long, not resupinate, \ an inch wide ;
peduncles f ; perianth 1|- ; sepals not auriculate.
Colour pale ? or red evanescent in the specimen ?
not spotted.

Var. 2. Bridgesiana. — PI. 1. f. 50. Specim. Los ojos
de agua. Bridges, 38. Herb. Hooker. Only the
barren stems. Leaves above two inches long, f
to wide, rather less obtuse than the latter.

Var. 3. Curbrana. — PI. 1. f. 51. Specim. ex Curbran.
Herb. Hooker. Stalk short and more rosulaceous ;
leaves 1|- long, % wide, very obtuse.

G. Peduncles \-2-jlowered ; leaves subspatulate ; not
resupinate.

16. Neilliana. — Neillii. Bot. Mag. 58. 3105. Specim.
Herb. Hooker. Leaves rather more spatulate than
those of peregrina, and not resupinate. The flowers
blush pink, the petals marked with deeper colour,
and the two upper with a yellow blotch. I am in-
formed by Dr. Neill, after whom it was named,
that it is a very shy plant, and has yielded no seed
or increase from the root, since it has been in his
possession.

7. Leaves narrow , not resupinate.

A. Peduncles 1-flowered.

17. Tenuifolia. — PL 1. f. 47. Gouan, No. 1. Herb.
Hooker. Stalk not a foot high ; leaves very slen-
der, 1^ inch long; peduncles two, 2 inches long,
bracteate near the middle ; perianth above 1^ long,
(intensely red ? with paler tips ?) segments acumi-
nate, very slender at the base. The bractes on the
peduncles perhaps indicate that this plant has
sometimes 2-flowered peduncles.

AMARYLLIDACEiE.

95

18. Versicolor. — Flor. Peruv. 3. 59. absque icone.
Stalk six inches, 2-3-flowered ; leaves lance-
linear, sessile; perianth yellow with purple marks;
sepals nearly equal ; petals narrower, the lowest
widest.

19. Pallida. — Bot. Mag. 58. 3040. Specim. Herb.
Hooker. Leaves sessile, above two inches long,
narrow ; sepals broad, rounded, with a small point ;
perianth denticulate, pale rose-colour ; two upper
petals yellow in the middle and marked with red,
lower down with orange.

B. Peduncles 2-3-flowered.

20. Preslana. — PI. 1. f. 43. Albiflora. Presl. Rel. Hank.
2. 121. Poeppig Frag. syn. Leaves above two
inches long, 3- 16ths wide, acute ; peduncles four,
2-flowered ; perianth denticulate, deep rose-colour,
with thick brown tips ; upper petals pale yellow,
marked with purple. Poeppig states that the co-
lour is so evanescent that it appears to be white in
the dry specimen by which Presl was misled to
give the plant an erroneous name. The rosy colour
seems to fade in most of the dry specimens of this
genus. The name albiflora cannot be preserved
without absurdity. This plant seems chiefly sepa-
rable from pallida by its 2-3-flowered peduncles,
and is perhaps to be considered as a variety thereof.
It grows in S. Chili in rich meadows near
Antuco.

21. Hookeriana. — A. rosea. Hook. Ex. fl. 181. non Ruiz.
A very hardy plant, preserving its foliage through-
out the winter out of doors against the front of a
stove, and never going to rest ; flowering in suc-
cession through the whole summer, and ripening
its seed. Its germen is strongly marked with
purple and very round. See PL 1. f. 1, 2, 4, 5,
and 17.

Stalk 5-12 inches high ; leaves glaucous, sessile,
a little tortuous, not resupinate, 3 inches long or
less, 3s wide ; peduncles erect, bracteate and forked
near the base, 2-4-flowered ; secondary peduncles
erect, contiguous ; sepals obovate apiculate ; petals

96

AMARYLLIDACE.E.

obovatc acuminate ; the lower portion of the upper
petals whitish and line-speckled, a faint yellow
mark above the speckles; the rest of the perianth
pale rosy ; points greenish ; perianth about IV inch
long or less; capsule round, purple-ribbed; point
much shorter than in Pulchra.

22. Cummingiana. — Raised from Chilian seed collected
by Mr. Cumming. I am by no means clear that
it should not be considered a variety of Angusti-
folia. — Specim. Versicolor Valparaiso, Bridges, 35.
and specim. “ Common on mountains near Val-
paraiso, Nov. and Dec.” Herb. Hooker, seem to
be this plant.

Stalk 12-15 inches high ; leaves subglaucous,
sessile, tortuous, not resupinate, 4 inches long or
more, 3-16ths wide ; peduncles 3-5, diverging,
bracteate and forked 2-2| inches above the base,
3-6-flowered ; secondary peduncles diverging; seg-
ments acuminate, a little tortuous, petals line-
speckled their whole length ; the inferior part of
the two upper petals yellowish, the rest of the
perianth pale brownish or yellowish red (very vari-
able in different seedlings), perianth If inch long;
sepals wide, obovate ; upper petals 5-16ths wide,
below marked at the point with green, the whole
below the green dashed with short red streaks ; the
other four segments with a green blotch at the
point extending down the rib on the outside ; low-
est petal acuminate ; pollen pale dull purple ; style
and stigma pale. Sterile stems weak, with linear
glaucous leaves. The colour, and the acuteness
of the petals varies in different individuals. I
have seen three out of five peduncles on one stalk
six-flowered. This plant is impatient of confined
air, and thrives well out of doors with a little pro-
tection. A. Angustifolia v. Selloensis seems to
come near this plant in general appearance.

23. Angustifolia. — Leaves long, linear, sessile, acute ;
perianth a little more than an inch long, colour
evanescent, probably purplish rose ; peduncles 3-
flowTered.

"W. Herbert, del

AMARYLLIDACEAt.

97

Var. 1. Conferta. — PI. 3. f. 1. Specim. Gumming, 354.
Herb. Lindl. Leaves crowded, 1|-2| inches long;
peduncles an inch long.

Var. 2. Intermedia. — Cumming, Herb. Lamb. Cum-
ming, 354. Herb. Hook. Leaves 4-5 inches long,
wide ; peduncles 2 ^ long.

Var. 3. Solliana. — PI. 4. 1’. 2. Solly. Chili. Herb. Lindl.
Leaves 4 inches long; peduncles long, diverging.

Var. 4. Acuminata. — PI. 4. f. 3. Specim. Chili. Bridges,
marked “Versicolor Martius.” Herb. Hook. Leaves
5-7 inches long, not ^ wide; peduncles two inches,
robust; perianth acuminate, purplish rose colour ;
upper petals streaked.

The colour of these four plants is evanescent in the dry
specimens, except that in var. 4. it can be seen to have been
purplish rose, and the upper petals streaked. .It is a vigorous
plant, not at all agreeing with versicolor of Ruiz. Var. 1.
and 3. appear so different, that if they had not been approached
to each other by the specimens of intermedia, which come
nearest to Solliana, but are numbered by the collector as if
identical with Conferta, I should have supposed them very
distinct.

24. Recumbens.- PI. 3. f. 1. Specim. Cumming, 384.
Herb. Lindl.- Lambert — Hooker. Sterile, stem

wanting. Fertile stem recumbent, curved, scaled
below, the scales acute, becoming more crowded
and leaf-like upwards; umbel 10 or more flowered,
very crowded; involucral bractes crowded, leaf-like,
shorter than the peduncles; perianth near 1^ inch
long, acuminate, bright purple ; upper petals
crossed with bright yellow streaked with purple. —
Some of the specimens of this plant are consider-
ably larger than that which I have sketched, and
preserve a brilliant colouring of purple and yellow.

§. 8. Peduncles 2- flowered. Stalks tall. {Perianth
usually vermilion or golden .)

A. Leaves smooth.

26. Quillotensis. — PI. 2. f. 2. Specim. near Quillote.
Bridges, 36. Herb. Hooker. — Herb. Lindl. — Fertile
stem above a foot high, scaly below ; scales
becoming longer, more linear and leaf-like upwards,
slender, not an inch long ; peduncles about ten, an
H

98

AM ARYLLIDACEiE.

inch long, 2-flowered, bracteate; flowers 1^ or near
If of an inch long, slender, golden.

Var. 2. Foliolosa. — Specim. Buenos Ayres; ex hort.
Tweedie. Herb. Hooker, with a more leafy stalk.

26. Aurea. — Bot. Mag. 61. 3350. Aurantiaca. Sweet.
Br. fl. g. Stalk 2-4 feet high, leaves glabrous,
resupinate, 4| inches long or less, f wide (persistent
after the ripening of the seed till the approach of
winter;) peduncles about five, 2-3-flowered, about
four inches long, with leaf-like bractes ; perianth
orange, two upper sepals lanceolate, streaked with
red ; capsule oblong with a blunt point ; seed
round pale chesnut colour. Native of the island of
Cliiloe, and raised by seed imported from thence by
Mr. Lowe, Nurseryman, at Clapton. Specimens
gathered by Cumming (No. 562), on the banks of
the river of Valdivia are less robust, but decidedly
identical. This beautiful plant is very hardy, and
thrives under a south wall, where it forms a great
tuft, and flowers and ripens seed abundantly. PI.
1. f. 3, 6, 7, 8, 9, 10, 1 1, 12, 13, and 14.

Var. 2. Valparadisiaca. — Specim. Cumming, 293. Val-
paraiso. Herb. Hooker. Herb. Lindl. Leaves an
inch wide, peduncles nine, 3-flowered; perianth
larger and more brilliantly coloured. This plant
has been raised from seed from Valparaiso by Mrs.
Bridgman of C. Weston near Thetford, but has not
yet flowered. By the dry specimens it appears to
Ije the most splendid plant of the genus.

A young plant of Aurea, planted out in front of a green-
house two years ago, has now about 50 flower-stems which
are four feet high, more or less, each with from three to eight
diverging peduncles, two or three-flowered, usually three-
flowered, but in no instance more ; the peduncles are from
4 to 4g inches high, forked at 2 or 2| inches from their base.
This splendid species ripens its seed abundantly, and increases
so fast by the root, and is so hardy, that I believe it will soon
become the ornament of our cottage gardens. By planting
it in a shady and cool situation a later succession of bloom
will be obtained. Its principal enemies are slugs, which
must be destroyed when the shoots first appear, and the
caterpillars of the lambda moth in July and August, which

AM AR YLLIDACE/E.

99

devour botli the leaf and flower. It conceals itself under-
ground in the day-time, hut may he easily killed at night
when it feeds. When the hlack aphis attacks the beans, it
molests this plant.

B. Leaves ciliated.

27. Hsemantha. — Flor. Peruv. 3. 60. Province of Here
in Chili. A variable species, distinguished from
Aurea by ciliated leaves of a more glaucous colour,
a more acutely pointed capsule, and, as far as I
have seen, by darker coloured seeds, not regularly
round, but deformed with a shorter and conspicu-
ously white hilum.

Var. 1. Simsiana — PI. 1. f. 15. 16. A. pulehella. Bot.
Mag. 49. 2353. Bot. Reg. 12. 1008. Hook. ex. fl.
64. Sweet Br. fl. g. 267. Upper sepal and lower
petal much abbreviated ; upper petals very narrow,
much elongated, acute, yellow (except at the point),
with red streaks ; sepals wider; general colour bril-
liant red ; peduncles erect, forked 5g or 6i inches
from the base, 4-5-flowered ; secondary peduncles
short. There appear to me to be two varieties cul-
tivated in this country besides pilosa, one with
longer flowers than the other. That which I pos-
sess growing out of doors, in front of the green-
house, has the stem not exceeding 13 inches, the
peduncles 5-flowered, 8 or 9 inches long, much
more erect than those of aurea, the bractes smaller.
Sweet refers the name Hsemantha of Ruiz specially
to the var. pilosa, excluding Simsiana, in which he
was mistaken, because Ruiz describes a white va-
riety, and Poeppig found Simsiana to vary with
white flowers, but not the smaller flowered plant,
which has the leaves more strongly ciliated.

Var. 2. Albida. — Perianth white with red lines on the
upper petals. Mentioned by Ruiz, Flor. Peruv.
Poeppig states (Frag. syn. p. 6.) this beautiful plant
to be so variable, that he has found specimens
growing together promiscuously, in the fields and
meadows near Antuco, in South Chili, in December,
vermilion, red-lead, orange, yellow, lemon-colour,
and white, but always with the dark-red marks on
the upper petals. He adds, that in his herbarium

h 2

100

AMARYLLIDACEiE.

lie finds one with smooth leaves, which is probably
aurea. 1 have found no approximation in the
foliage of seedling plants of aurea and Simsiana, of
which the parent plants were growing at a small
distance from each other. The flower-stem of Hse-
mantha perishes almost before the seed is ripe, and
the plant goes to rest for about six months ; aurea
preserves its leaves till cut by severe frost, and is
not much disposed to rest entirely.

Var. 3. Pilosa. — Bot. Reg. 17. 1410. Ciliaris ? Poeppig
Fr. syn. A shorter plant? with the ciliation of the
leaves more conspicuous, and the sepals serrate in-
stead of denticulate. The colour is much too pale
in the plate in the Bot. Reg. which would have pre-
vented Poeppig from recognizing it in his Ciliaris,
which I believe to be the same plant, and in that
case, whether it be considered as a variety or a sepa-
rate species, the name pilosa has the priority.

Ciliaris? Poeppig Fr. syn. absque icone. Stalk
2-3 feet ; leaves lance-oval, about 4 inches long,
half an inch wide, ciliated with bristles, petiole 1J
inch; upper leaves narrower; peduncles 9-12, 2-
flowered, 3 inches long; flowers less than in hae-
mantha, always vermilion. A very rare plant; on
the mountains near Concon in North Chili, flower-
ing in Oct. It seems to differ chiefly in a harsher
ciliation, and smaller flowers, which scarcely dis-
tinguish it from a species said to be so variable. The
longest petals of pilosa, which I believe to be this
plant, are 1 inch | long, those of the longest var. of
Simsiana, 2 inches and \ ] the shorter comes nearer
to pilosa.

§ 9. Stalk scarcely above ground , one-flowered.

28. Pigmaea. — PI. 8. fig. 4 — 13. Specim. Matthews,
865. near Pasco, Peru. Dec. Herb. Hooker. Root
tuberous, white, palmated ; leaves just above
ground, lance-linear, about an inch long, glaucous;
one-flowered, perianth bright yellow ; petals
marked below with dark brown ; stigma deeply 3-
cleft. This curious little plant is very unlike an
Alstrcemeria in many respects. In the specimen the
germen is covered by the leaves, and the segments

W. ! Dscberb .del .

AMARYLLIDACEiE.

101

appear very similar in form and regular ; but
Matthews has transmitted some outlines, by which
the germen and ovules appear to conform exactly
with Alstrcemeria, the sepals to be spatulate, and the
petals lance-oval. He represents the style and
filaments straight, which would not agree with
Alstrcemeria, but the specimen has a curved style.
Fig. 7 to 13 are Matthews’s outlines, which are
evidently too large. Fig. 4-6 are from the dry
specimen.

Doubtful Species.

29. Pulchella. — PI. 4. fig. 1. Spec. Herb. Banks, ab
arm. J. Banks in Brazilia lect. This specimen is
unfortunately very much mutilated, especially as to
the flowers. It is labelled in the hand-writing of
Dryander, and was certainly supposed to be the
plant intended by Linnaeus, Suppl. 206. Linnaeus
described his pulchella not from a specimen, but
from a drawing, and no reliance can be placed upon
such a foundation. The specimen in the Banks,
herb, has the lower part of the stalk naked or only
scaled, the leaves smooth, petiolated, petiole leaf
1^, in all about 2 inches long, lance-oval, acute, 7-
24ths of an inch wide in the middle, equally attenu-
ated; peduncles 1 -flowered, 1-2 inches, perianth about
1|-; the sepals seem lance-ovate, red with yellow,
upper petals seem obtuse, spatulate, pencilled with
red, green upwards. No such plant appears to
have been met with by any later collector, and
there is no memorandum as to the part of Brazil
in which it was found. It is certainly distinct
from all the plants I have described, unless it be a
variety of Psittacina, which I do not believe. It is
strange that Dr. Sims should not only have con-
founded Haemantha with the pulchella of Linnaeus,
but assumed as nearly certain that it was also the
Ligtu of Feuillet, the one being vermilion, the
other purple striped with white.

Brasiliensis (Sello) is much too insufficiently
described to be ever identified. It is supposed to
be a variety of Monticola.

I have thus to the best of my ability arranged the various
Alstroemerias of which we have any knowledge. That know-

102

AMARYLLIDACEA2.

ledge is in many respects imperfect, and the variability of
the species renders their arrangement very difficult. It is
possible that the whole seventh section may upon more
intimate acquaintance be considered as local varieties of angus-
tifolia, a point of little importance, as they are placed together
under one head, and would retain the same names as subor-
dinate, if that were the case. I do not, however, believe that
it will be so found, for neither did the many seedlings which
were raised of Cummingiana, nor the garden seedlings of
Hookeri, vary in the aspect of the stalk and foliage. Being
chiefly natives of Alpine situations, these beautiful plants
require free air, and (with the exception of Caryophyllaea
amongst those we possess) very little protection, except from
severe frost. A. Hookeri planted in front of one of my stoves
formed a large patch, the foliage resisting all frost in that
situation, and flowered throughout the summer ; but the two
last dry summers have greatly reduced it. They are very
thirsty plants in the season of their growth, and should be
abundantly watered in dry weather at that time. I appre-
hend that the singular circumstance mentioned by Poepipg,
of his finding A. haemantha growing promiscuously of every
shade of vermilion, orange, yellow, sulphur, and white, arose
from the vicinity of the two plants mentioned by Ruiz, the
vermilion and the white variety having bred together, and
their intermediate produce again with either parent. I do
not believe that we shall be able to produce the same result
even by garden cultivation and sowing the seed of the
vermilion plant, without first obtaining a white variety to
cross with the vermilion. The known variability of the
genus, however, the white peregrina and the beautiful two-
coloured variety of pulchra which have been raised in
England, afford a great encouragement to cultivators. It is
very remarkable that the stigma of Alstroemeria does not
come to perfection till after the decay of its anthers. The
stamens advance successively like those of Nerine undulata,
and like them nod before they rise, the petaline filaments
taking the lead, but the two upper not simultaneously with
the lower. It results from this that the stigma must either
be fertilized by the pollen of another flower, or that its own
scattered pollen must be efficient after it seems to be dried
up and lost ; in either case there is a greater probability of
the intrusion of the pollen of another individual, than when
the stigma and anthers are mature at the same time.

AM A RYLLIDACEA2.

103

The variation in the form and colour of the flower of A.
pnlchra, and the two coloured varieties, as well as of the seed-
lings of A. Cummingiana from imported seed, should render
botanists very cautious not to multiply species freely on the
appearance of such diversities in natural specimens from
different localities, and makes it very difficult to fix on the
true distinguishing features. Little attention is to be paid to
the length of the style in Alstrcemeria ; its maturity is very
tardy ; it is very short at first, grows out slowly, and at last
the stigma, which had appeared to be simple, expands and
becomes trifid and patent. The anthers discharge their
pollen long before the maturity of the style. The late
developement of the stigma should make the genus very liable
to spontaneous intermixtures of the species, but render it
difficult to obtain artificial crosses. I failed in getting seed
at all from an attempt to fertilize the red peregrina by the
white, which must have arisen from having neglected the
proper moment for fertilizing the style or for selecting the
pollen.

A. Psittacina, as well as hremantha and aurantiaca,
flowers well in the open ground, if covered witli straw or a
thick coat of leaves in the winter. The soil should be light,
and the tubers set pretty deep ; and any heading that would
throw the wet off’ in the winter will be found advantageous.
It is absolutely necessary to pick the slugs oflP the border,
which will otherwise devour every shoot at its first appearance
above ground : and it will be found advantageous to cover
the bed in the spring with dry sawdust which the slugs do
not like to crawl over, and it will keep moisture in the
ground. A top covering of peat is also disagreeable to slugs,
which I find very troublesome in biting the flower-stalks of
Gladioli on sandy loam, but they rarely do so on a border of
black earth.

16. Collania. — Stem rigid, erect, curved at the summit;
leaves rigid (not reversed ?) umbel pendulous ; ger-
men turbinate, the operculous base of the style be-
coming enormously enlarged, and forming the main
part of the fruit ; perianth rather tubeformed, and
not at all patent. (Pericarp probably soft and pul-
paceous, since the fruit in at least one species is sweet
and eatable.) Filaments and style straight ?

1. Involucrosa.— PI. 9. Specim. Matthews, Peru, 863.

104

AMARYLLIDACE.E.

Alst. grandiflora. Herb. Hooker. Stalks 3-4 feet
high, numerous, round, smooth ; leaves 4 inches
long, more or less, £ wide, crowded, attenuated at
each end, ensiform or falcate, blueish green, lighter
beneath, strongly nerved ; involucral bractes much
wider, darker green, more densely nerved, nume-
rous (sometimes lb); perianth pendulous, pale
yellow tinged with green, petals not spotted ; fila-
ments longer than the perianth, style still longer,
stigma trifid; flowers 1 to many; root thick
fibrous (growing at the foot of a shrub, or on the
side of a rocky precipice). Found near Culluay ?
and St. Mateo in November. Specimens, Matthews,
863. Herb. Lambert, and Herb. Lindl. In the
former, the involucral bractes are much narrower.

2. Dulcis. — PI. 7. fig. 1 — 9. Alstr. dulcis. Hooker B.

Misc. Alst. uniflora. Matthews MS. Stalk 5-14
inches, smooth; leaves suberect, |-1^ inch long, a
line wide, margins reflex, glaucous green, glossy ;
peduncles one or two, 1-2 flowered ; perianth near
sepals light red, tint between crimson and scar-
let, tipped with green ; petals yellower with more
green, and a border of yellow speckled. Flowers
generally solitary, sometimes four. This plant is
called Campanillas coloradas, and the fruit is
sweet and agreeable to the taste, and much sought
by children, “ the seeds being enveloped in a red-
dish gelatinous substance by which I understand
that it has a pulpaceous pericarp, as the case is
with Tamus and Hsemanthus. It grows beneath
rocks at Huayllay, near Pasco, at an elevation of
from 12 to 14,000 feet. It appears (see fig. 2.) to
have a creeping rootstalk with* small pyriform
tubers appended.

3. Glaucescens. — PI. 10. fig. 1, 2. Specim. Jamieson.

Herb. Hooker. Paramo of Cayambe. Alst. Glau-
cescens. II. and B. Kunth, 3. 282. Stalk 1-2 feet
high ; leaves numerous, erect, lanceolate, acumi-
nate, sessile, rigid, glaucous, downy underneath,
1^-2 inches long or less, pressed close to the stalk,
the upper widest; involucral leaves widest; pe-
duncles 4-6, half an inch long, smooth, with shorter

AM ARY LLI DACEA2.

105

lanceolate bractes at tlieir base ; perianth smooth,
not half an inch long ; sepals oblong, obtuse, scar-
let, paler witbin; petals slenderer, yellow with rusty
spots ; filaments shorter. “ This plant is found
also on Chimborazo and on the ridge of Assuay.”
Jamieson MS. Found by Humboldt on the slopes
ofPachincha, between Palmacucha and the spring
of Cantuna.

4. A ndimarcana. — PI. 8. fig. 1 . Specim. Matthews, Peru,

1164. Herb. Hook. Stalk smooth; leaves nume-
rous, 4 inches long, scarcely § wide ; margin re-
flex, attenuated at both ends, acute (probably
glaucous), with thick white down underneath ; two
broader involucral leaves; peduncles three, 1-2
flowered ; common peduncle strong, bearing a
lanceolate bracte 1^ long, secondary slender; pe-
rianth 1-| long, sepals oblong, seemingly red tipped
with green, about § wide ; sepals spatulate, yellower
tipped with green ; anthers just exceeding the limb,
style a little longer ; stigma very little cleft. Found
in July on the lofty mountains of Andimarca.

5. Puberula. — PI. 11. fig. 1. Specim. Matthews, Peru,

1165. Herb. Hooker. Stalk downy; leaves nume-
rous, suberect, paler underneath ; downy on both
sides, margins reflex, 3-4 long or more, about 3-16ths
wide, acute; involucral leaves wider ; peduncles 5,
slender, 2-3 flowered, with a lanceolate bracte at
the fork ; perianth an inch long, sepals oblong, ^
wide, seemingly red with a green tip, yellow
within ; petals yellower with more green at the
end ; style and filaments shorter than the perianth.

Var. 2. Acicularis. — PI. 2. fig. 2. Spec. Matthews, Peru,
1165. No. 2. Herb. Hooker. Leaves 2^-3 inches
long, scarcely a line wide ; peduncles 2, 2-flowered.

Both were found in July on the high mountains of Andi-
marca.

It is much to be lamented that seeds of the Collanias, as
well as numerous species of splendid Bomareas which are
herein described, have not been gathered by collectors, and
remitted to Europe ; most of them would live out of doors in
England with a little covering in winter of sawdust, leaves,
or ashes, and probably in the south of Europe would require

106

AMARYLLIDACEyE.

no care, except perhaps watering- in dry summers. We pos-
sess only, as far as I know, the tender edulis, the Chilian
salsilla, and the Mexican hirtella and acutifolia, unless a
Chilian variety of ovata figured in the Bot. Mag. be in this
country. B. hirtella will live well at the foot of a south wall,
sawdust being heaped over the roots in winter. With me it
flourishes against the front of a greenhouse, and twines up a
pole to the height of about eight feet. Acutifolia has lived
through a winter in the same situation, but it does not go
to rest like hirtella for six months or more, and therefore is
much more liable to injury, and requires more protection,
but it is impatient of heat under glass in the summer. Its
root is less tuberous and substantial, but it will succeed in
an open border if set deeper and covered with leaves to pro-
tect it from the frost. There is a rage at present for the
introduction of Orchideous plants, of which the greater part
cannot compare with the Amaryllidaceous in beauty, and
require a degree of heat, and a damp atmosphere, which is
very unpleasant to the cultivator, and no pains are taken to
introduce the rich variety of splendid plants in this natural
order, which the western continent possesses in its temperate
and even cold regions. I wish that I could excite some of
our wealthy cultivators to turn their thoughts to the acqui-
sition of the plants here described, and their numerous
kindred, which are profusely scattered over the slopes of the
Andes, and waste their beauties in the woods, as yet un-
noticed by any European traveller.

17. Sphjerine. — Stalk straight, attenuated upwards; pe-
duncles simple ; sepals equal ; petals equal ; style
and filaments straight (leaves subrigid ; perianth
under an inch) ; capsule indehiscent.

A perfect diagnosis of the genus cannot he obtained; the
perianth and germen appear to conform with the smallest
flowered Bomareas, hut on the authority of Ruiz all the spe-
cies with a short straight stem have an indehiscent capsule, with
a little pulp , and , if so, they cannot he Bomareas ; and, when-
ever they can be inspected, further differences will he found.
The specimen with fruit, probably immature, in Mr. Lambert’s
herbarium (see PI. 12. fig. 4.) has the fruit spherical, quite
unlike that of Bomarea. Imperfect as my knowledge of this
genus is, I cannot set down as Bomareas plants which are as-
serted to have an indehiscent capsule. I believe that a material

AM ARY LLIDACEiE.

107

difference will be found at the base of the style, and that the
(jennen is much less operculate ; but that is conjectural.

1. Distychophylla. — PI. 12. f. 3.4. Specim. Ruiz. Herb.

Lambert. Alstr. distychifolia. Flor. Peruv. 3. 60.
sed non t. 287. fig. pessima ex Bomarese tomen-
tosse var. Pangoensi, vel. B. setacea, conflata. —
Stalk solitary, filiform, 2 feet high ; leaves 2^ to
above 3 inches long, about § wide, lance-oval, with
five hard nerves and six pair of weaker interme-
diate nerves, alternate, two usually opposite at the
summit; petiole about ^ long; the lower leaves
less ; involucral bracte one or more, subulate, small;
peduncles 3-6, \ an inch to 1^ long; perianth \ an
inch long, purplish red ; petals a little wider, blunt,
with a small point ; capsule ovate, obtusely trian-
gular, furrows neai’ly obsolete, not dehiscent;
seeds roundish, yellow, with little pulp. In the
woods of the Peruvian Andes, in the high cold
mountains of Munna. The name distychifolia
being a compound of Greek and Latin, is neces-
sarily corrected. Ruiz’s specimens of distycho-
phylla, in Mr. Lambert’s herbarium, were con-
founded by him with specimens of Bomarea setacea,
and the figure in the Flora Peruviana, which does
not agree with the description, has been taken from
a wrong specimen, and does not represent the plant
of which it bears the name.

2. Secundifolia. — PI. 12. f. 1.2. 5. Spec. Ruiz, Herb.

Lambert. Alst. secundifolia, Flor. Per. 3. 60. t. 290.
Stalk solitary, filiform, two feet high, glabrous,
pendulous ; leaves alternate, 1^ palm long, 1 inch
wide (in the specimen there are two opposite at
the top, which seems usual in this genus, near 3^
long, wide, much less ovate) ; the inferior
smaller, underneath downy; peduncles 4-6, 1^
inch long; involucral bractes small, subulate;
perianth f long an inch, Ruiz in the text,
wrong) ; sepals downy without, purplish red, obo-
vate, a little widest; petals cuneiform, yellow witli
black spots and a green tip ; gennen and stigma
doivny; capsule hexangularly oblong, with the
alternate angles fainter, not dehiscent, yellow,

108

AMARYLLIDACE2E.

wrinkled, rough ; seeds roundish, obovate, red,
(nestling in pulp ?) In the woods near Munna, as
well as the last.

Ruiz’s descriptions appear to me generally accurate, but
the figures in his work the very reverse : I think, however,
that he was probably wrong in saying that the seeds nestle
in pulp, which is the true definition of a berry, as that of a
currant, &c. ; but I have found no berry in this order.
Whether the seed, like Bomarea, or the pericarp, like
Tamus and Haemanthus, has a pulpaceous coat, remains
doubtful.

3. Coccinea. — PI. 16. fig. 1. Alst. Coccinea. Flor. Peruv.

3. 62. t. 291. Stalk 3 feet high, with ovate scales
below the middle, flexible upwards ; leaves alter-
nate, villous underneath, smaller near the summit,
2 opposite at top ; bractes none ; peduncles villous,
two inches long; perianth half an inch long; sepals
oblong, scarlet with a green tip, petals cuneiform,
greenish yellow with purple spots ; capsule downy,
oval, not dehiscent ; (seeds nestling in pulp ?) In
the woods of Huassahuassi, Peru. The outlines
I have given of this, and the capsule of secundi-
folia, are taken from the Flor. Peruv. I have seen
no specimen of Coccinea.

4. Nervosa. — PL 13. Spec, alterum. Matthews, 1661.

Herb. Hooker. Leaves alternate, 4-41 inches long,
ovate, lanceolate, acutely attenuated, f wide ; 29
principal nerves equal, with a like number of alter-
nate inferior ones ; two upper leaves opposite, and
two opposite bracte-like leaves much smaller, half
an inch above ; interior bractes filiform; peduncles
19-20, 1 \ inch long, downy; germ downy; perianth
half an inch long (seemingly red, yellow, and
green).

5. Brevis.— PI. 18. fig. 1. Spec. Matthews, Peru, 1660.

Herb. Hooker. Stalk 14 inches, a little spiral near
the top ; the lower half scaled, the lower leaves
less ; leaves lance-oval, attenuated at the base,
strongly nerved underneath, three inches or more
• long, half an inch wide, alternate, two upper oppo-

site, and two smaller bracte-like leaves* above oppo-
site ; interior bractes small; peduncles three, f of

AMARYLLIDACE^E.

109

an inch, a little downy, as well as the germen ; peri-
anth half an inch (seems orange and green), sepals
horned.

I have no knowledge of the fruit of either of the two last
species, being only acquainted with the specimens of which
1 have given the outline ; but I venture to predict that, if the
statement of Ruiz concerning the capsule of the three former,
which I have no reason to doubt, be correct, the fruit of these
two will be found indehiscent also. Concerning the first I
could entertain no doubt; the spiral twist near the top of the
stem of brevis at first gave me the idea of a Bomarea, but,
after I had long studied it, I felt assured that it was not. It
will be observed that all five agree not only in having the
lower part of the stem scaled and the first leaves smaller, but
in having the rest alternate, except the two upper, which are
opposite, a feature that does not occur in Bomarea ; that all
have short simple peduncles, and secundifolia, which has the
perianth f of an inch long, is the only one in which it exceeds
half an inch. As far as we know, they are said to have soli-
tary stems. The two new species having smooth leaves
should properly follow distychopliylla, but are placed at the
end, on account of the want of certain knowledge of their fruit.

18. Bomarea. — Stalk twining; germen triangularly turbi-
nate ; segments of the perianth inserted in deciduous
glands on the germen ; sepals nearly equal, petals
nearly equal unlike the sepals ; capsule valveless,
coriaceous, triangularly turbinate, six furrowed, with
an obtuse opercle, when ripe wrinkled; the dissepi-
ments entire, the opercle widely dehiscent; seeds ad-
hering long after maturity, roundish, tapering at one
end, outer coat soft (perhaps always red or yellow) ;
middle coat pulpy, inner coat brownish, adhering to
the horny albumen; umbilicus a little pointed ; raphe
inconspicuous, chalaza annular on the inner coat,
scarcely visible on the outer, opposite the foramen ;
umbilical cord entering the raphe nearly half way
between them ; (root more or less tuberous ?)

The style of Bomarea, as far as I know, is tripartible; but
I have no confidence in the invariability of that feature,
which is irregular amongst the Narcissece. Salsilla has a
decided flexion of the filaments, which, if it exists in Hirtella
and Acutifolia, must be more equivocal ; but we are ac-

110

AM ARYLLIDACE.E.

quainted with so few of the species in a fresh state, that I
cannot properly investigate this feature. In the Alstrce-
merieae, the filaments seem to change their altitude so much
at different periods, that it is difficult to assign a posture to
them. The capsule of Salsilla agrees with that of Hirtella
and acutifolia. I have not seen its ripe seed.

§. 1. Peduncles two-flowered or more.

§§. Perianth nearly equal.

A. Leaves smooth , flowers about half an inch long , crowded.

1. Salsilla. — Feuillet Obs. 2. 713. t. 6. var. 1. obtusa.
Alst. oculata. Bot. Mag. 61. 3344. Lodd. Bot. Cab.
t. 1851.

Var. 2. Prsecipua. — Plate 16. fig. 4. foliis majoribus.
Herb. Hooker. Cumming, 345. Valparaiso.

Var. 3. Subfalcata. — PI. 16. fig. 5. Herb. Hooker. Cruik-
shank, 33. Chili, Herb. Hooker. Cumming, 345.
Herb. Lindl.

Flowers purple with a dark eye-like spot on the lower
part of the two upper petals, and a paler one on the lowest.
It has a strong flexion of the filaments. Salsilla was first
figured and described by Feuillet together with Alst. ligtu and
peregrina, all three stated to be Chilian plants. By a strange
mistake, when Bomarea edulis and Alst. caryophyllaea, both
tender tropical plants, were brought from the East coast and
West Indies, they were confounded with Salsilla and Alst.
ligtu, and have usurped their names in our stoves and in
modern botanical works. It so happens that, numerous as
the genus Bomarea is in the higher latitudes, not a single
species, except Salsilla, and Ovata var., has been discovered
in Chili, unless the specimens in Professor Lindley’s herba-
rium from Conception, gathered by Macrae in 1825, with
the flowers not blown, and varying in leaf from lanceolate-
acute to lance-ovate, be B. glomerata, which is a Peruvian
species, and approaches to Salsilla in leaf. It is quite cer-
tain that the edulis of Tussac, which has usurped the name,
is quite distinct from Feuillet's plant, and the original name
must be restored to the Chilian twiner, which has been since
figured as oculata. The name oculata is therefore to be alto-
gether expunged. B. salsilla is a hardy greenhouse plant,
of which the seedlings vary very much in brightness of colour
and in leaf. I have not found Salsida amongst the Peruvian
specimens.

AMARYLLIDACEjE.

Ill

B. Leaves smooth, flowers middle sized , umbel loose.

2. Edulis. — Tussac FL des Antill. Alst. Salsilla. Bot. v
Mag. 39. 1613. Bot. Rep. 649. Vandesia. Salisb.

Var. 2. grandis. • Specim. Herb. Hooker. Brazil, Mr.
Booz. Peduncles long, three or four flowered. Sal-
si lloeides, Martius ap. Schultes. In hedges near
Rio Janeiro. This is Booz’s plant, certainly not dis-
tinguishable as a species. The name Salsilloeides,
being founded on the original error, cannot stand.

Var. 3. Maranensis. — Specimen Herb. Lindl. Maran-
hao, Hesketh. Taja de bibro. Leaves 3f inches
long, not •g- wide, acute, peduncles above two
inches long, simple, with a diminutive bracte half
an inch from the base.

Linnaeus probably confounded this plant with salsilla,
in consequence of having received some dry specimen at a
time when salsilla was considered to be the only twining
Alstrcemeria, and without sufficient examination he sup-
posed his specimen erroneously to be Feuillet’s plant. The
tubers are said by Tussac to be eaten in St. Domingo like
those of the species of sunflower called Jerusalem artichoke.

It is a stove plant. If Maranensis should prove to have always
simple peduncles, with such narrow leaves, it must be re-
moved from its place as a variety of edulis to the second sec-
tion. I am inclined to separate it.

Poeppig says he found edulis (salsilla B. M. 1613) fl. Oct.
Dec. in coppices, S. Chili, 7000 f. alt. Strange, if the W.
Indian plant be found in such a situation. It was probably
ovata var. B. M. 2848.

3. Caraccensis. — PI. 16. f. 2. Specim. Fanning, Herb.

Lambert, from Caraccas. Leaves smooth, lanceo-
late, acuminate, six inches long ; peduncles
about six inches long, two-flowered, branching
near the end with a small bracte ; involucral
bractes lanceolate, about 2i inches long ; perianth
about an inch long, sepals apparently red, petals
green spotted. It must be closely allied to edulis,
and may perhaps be more properly called a variety
of it.

C. Leaves smooth, flowers large, umbel crowded.

4. Formosissima. — PI. 14. f. 4. Ruiz, Specim. Herb.

Lambert. Als. formosissima, Flor. Per. 3. 64. t. 296.

• t, /°- /oq -/ 1

p uU'*. [ t.

112

A MARY LLTDACEjE.

This magnificent plant is said to produce an umbel
with from 36 to 80 flowers, of a purplish red and
yellow, the petals being richly spotted. The stalk
grows 12 feet high, and the leaves are an inch wide
and very acute. It grows in the woods of the Pe-
ruvian Andes near Munna.

D. Leaves smooth , flowers large , umbel loose.

5. Pauciflora. — H. and B. Kunth, 3. 285. Leaves scat-

tered, acuminate, two inches long or more, one
wide ; petioles three lines, peduncles four, 4-5
inches long, 2-3-flowered ; bractes lanceolate, not
an inch; sepals two inches (purple?) petals
orange ? Near Santa Fe de Bogota on the slopes
of Mount St. Quadalupe.

E. Leaves pubescent , flowers about an inch , umbel loose.

6. Bracteata. — PI. 3. f. 3. Specim. Ruiz, Herb. Lambert.

Alst. bracteata, Flor. Per. 3. 61. t. 291. The
figure in the FI. Per. is very incorrect and greatly
exaggerated in size, on which account I have given
an exact outline of a portion of Ruiz’s own speci-
men. Involucral bractes five, lanceolate ; leaves
lanceolate with a spiral point ; peduncles five,
long, 2-3-flowered, bracteate ; sepals purple tip-
ped with green, petals spatulate, green speckled
with purple, yellow at the base. In the woods of
Huassahuassi, Peru.

7. Acutifolia. — Bot. Mag. 58. 3050. Leaves slightly

haired on the nerves beneath, lanceolate, acumi-
nate, half an inch wide or more; peduncles 1-2-
flowered ; sepals red, petals orange. Mexico.

8. Obovata. — PI. 14. fig. 2. Spec. Col. Hall, 7. Herb.

Hooker. “ Road of Angus, Peru, at the height of
4000 feet; flowers crimson.” Perianth 1| inch,
sepals crimson; petals spotted, some tipped with
green ; involucral bractes lanceolate, acuminate,
3^ inches long, near an inch wide, lesser 2| inches
wide ; peduncles 8 inches long, 4-flowered, brac-
teate at the forks. The leaves and rest of the
umbel are wanting.

9. Hirtella. — Sweet, Brit. FI. g. 228. H. and B. Kunth,

3. 284. Gloriosa, Deppe et Schiede ap. Schlecht

W. Herbert .del .

AMARYLLIDACEiE.

113

Linn. 6.51. Spec. Herb. Lamb. Ovata, B. M. 2848 ?
Stem about 7 feet high ; leaves 4 or 5 inches long,
lanceolate, acuminate, with a harsh pubescence
underneath ; peduncles 5-7, 1-3-flowered, 4-6
inches long ; sepals red ; petals green spotted with
red ; stamens, germen, and base of the petals,
pubescent. Between Mexico and Tingello on the
road to Tolucca. This fine plant thrives well in
England at the foot of a south wall, or against the
front of a greenhouse, with the precaution of pour-
ing a small heap of sawdust or coal-ashes over it in
the winter to keep the frost from the root. It lies
at rest through the winter, and late in the spring
the shoots rise like asparagus. It ripens its seed,
unless the season is very unfavourable. — Chili?

10. Latifolia. — Flor. Per. 3. 64. t. 295. Leaves a palm
wide, a palm and half long; peduncles 7-14, a
palm long, bracteate ; perianth above an inch long;
sepals red tipped with green, petals green speckled
and streaked with purple. In Peru on the hills
near Atiquipa called Lomas.

11. Cordifolia. — Flor. Per. 3. 64. t. 290. Stalk 2-3
yards high ; leaves 4 inches long, nearly 3 wide,
petiole three lines, peduncles six ; involucral leaves
ovate, lanceolate, acuminate; perianth an inch long,
yellowish red ; germen a little downy. In the
woods of the Andes of Peru by the road ofPozuzo.

12. Grandifolia. — H. and B. Kunth, 3. 285. Differs
from cordifolia in a ^narrower leaf, longer petiole,
shorter peduncles and leaves, not alternate. The
colour of the flower is not perfectly ascertained, and,
as it is the native of a hotter climate, it is probably
distinct.

13. Ovata. — Var. 1. Cavanillesiana. Cavanilles ic. p. 54.
76. Leaves alternate, with the edge revolute ; pe-
duncles about five, two-flowered, bracteate at the
fork ; sepals bright red tipped with green ; petals
green speckled with black. In Peru.

Var. 2. Tatiana. — PI. 15. fig. 3. Specim. Mexico. Herb.
Hooker. Cervantes Mexico. Herb. Lambert. Leaves
3| inches long, li broad, ovate, acutely acuminate ;

i

114

AMARYLLIDACE^E.

petiole | long ; involucral bractes four, conformable,
smaller ; peduncles five, 3 inches long, bracteate at
the fork and below the secondary flower, therefore
probably occasionally three-flowered; perianth 1^-
long, seemingly yellowish red and green ; petals
scarcely ± wide. — Ovata ? Bot. Mag. see Hirtella.
Var. 3. Nobilis. — PI. 15. f. 2. Specim. Culb. Herb.
Hooker. Leaves 4| long, wide, ovate, acumi-
nate, bractes four, conformable, smaller ; peduncles
5^ long, two-flowered, bracteate at the fork and on
the secondary peduncle; perianth 1^ long, petals
above wide ; sepals seemingly reddish yellow,
petals yellowish green ; approaches to cordifolia.

14. Macrocarpa. — Flor. Peruv. 3. 63. t. 294. Pillao in
Peru. Spec. Matthews, Herb. Hooker, 483. At the
roots of bushes. Huamantanga, Peru. April. Spec.
Matth. Herb. Lindl. Leaves a palm long, an inch
wide (in the specimens about three inches long,
^ wide, lanceolate, acuminate); peduncles 6-18,
(in the spec, four, 2-3-flowered) ; bractes small;
perianth an inch (in the spec. 3-4ths); yellowish red.
The leaves are more crowded and much less ovate
than in the three foregoing ; the sepals seem much
redder, the petals more spatulate, speckled at the
edge.

15. Hirsuta. — H. and B. Kuntli, 3. 285. Leaves sparse,
acuminate, paler underneath and hirsute, 4 inches
long and more, 1J wide ; petioles half inch ; peri-
anth crimson, sepals obtuse, 7 lines ; petals longer,
not an inch. On the woody mountains of New
Granada, between Fusagasuga and Icononzo.

16. Cornuta. — PI. 17. f. 4. Specim. Matthews, 1161.
Mount Parahuanca, Peru, July; Herb. Hooker.
Upper leaves subsessile, 5| inches long, 1| wide;
a harsh down on the nerves underneath ; a slender
setaceous point; involucral bractes conformable,
shorter; peduncles five, 9-10 inches long, brac-
teate at the fork 6-7 inches from the base, 3-4
flowered ; bracte an inch or more, narrow ; peri-
anth an inch long, (apparently dark red ?) sepals
ending with a horn \ of an inch long, petals with
a bristle.

AMARYLLIDACEA2.

115

§§. 2. Petals much longer than the sepals.

F. Leaves smooth , flowers about an inch long, umbel loose.

17. Dispar. — PI. 17. fig. 1. Specim. Matthews, 1G58.
Herb. Hooker. Leaves lance-ovate, acuminate,
strongly nerved, 8 inches long, 2 wide, petiole \ ;
peduncles 5, near 5 inches long, 2-3-flowered,
bractes slender, ^ long ; perianth an inch, petals
one-third longer than sepals ; sepals obtuse-oval,
seemingly (orange ?) tipped with green ; petals
less highly coloured, spatulate ; involucral bractes
almost obsolete.

§. 2. Peduncles one-flowered.

§§. 1. Perianth nearly equal.

G. Leaves smooth, flowers about \ an inch, umbel crowded.

18. Glomerata. — PI. 1 5. fig. 1. Spec. Peru, Matthews,
1662. (Do. 1661.) Herb. Hooker. Leaves about

inches long, ^ wide, closely nerved, lanceolate,
acute ; petiole about apart ; outer involucral
bractes small, lanceolate, inner setaceous ; pedun-
cles about half an inch, perianth half an inch, ap-
parently red, petals edged with orange. This plant
in the dry specimens has much the aspect of B.
salsilla, but may be readily distinguished by the
thickness of its leaves, its short simple peduncles,
and the different colour of its flowers. Matthews,
1662. Herb. Lindl. is a variety with the leaf wide,
4^ long, with a very long slender point.

19. Torta. — H. and B. Kuntli, 3. 283. Leaves scat-
tered, acuminate, 1^ inch long, margin revolute;
peduncles numerous, half an inch ; sepals obtuse,
red, half an inch ; petals scarcely shorter, green
spotted with black, yellowish at the base. In the
cold parts of Peru, near Coxamarca and Parama de
Yanaguanga.

20. Cumbrensis. — PI. 18. fig. 2. Spec. Jamieson. Herb.
Hooker, found “ between Cumbre and Juna at 9000
feet elevation ; flowers light purple, tipped with
green.” Leaves numerous, lanceolate, 2 inches
and less, | wide ; involucral bractes shorter ; pe-
duncles slender, about an inch, flowers about ten ;
perianth f long; sepals obovate, seemingly reddish

i 2

116

AMARYLLIDACE^E.

with a g'reen tip, petals spatulate, wider (green,
except at the base ?).

II. Leaves smooth, flowers about an inch long, umbel crowded.

21. Fanningiana. — PI. 16. fig. 3. Spec. Fanning, Ca-
raccas. Herb. Lambert. Leaves about three inches
long, and 1 1-1 6ths wide, growing smaller upwards;
involucral bractes very numerous, reflex, lanceo-
late, 1| inch long; peduncles very crowded; pe-
rianth about inch long. It seems allied to
Bredemeyeranu, which is densely pubescent.

22. Fimbriata. — Flor. Peruv. 3. 63. t. 293. fol. vix pub.
Leaves lanceolate, alternate, very acute, near three
inches long and one wide ; petiole 3 lines ; flowers
25-50 or more; bractes as many, lanceolate, reflex,
inner less; peduncles an inch; perianth near an
inch, yellow variegated with orange ; petals tipped
with green, fimbriated on the margin.

Var. 2. Paltarumensis. — Spec. Matthews, Paltarumi, Peru,
867. Herb. Hooker. Stem climbing, 18-20 feet ;
leaves closely nerved, 2 \ to 3^ long, ^wide; perianth
a little above an inch, probably redder than Ruiz’s
fimbriata ; petals spotted with deeper colour. It has
the reflex lanceolate bractes, and peduncles a little
more than an inch. The name Paltarumensis will
equally distinguish it, whether it prove to differ
more than I apprehend from Fimbriata or not. It
flowers in Peru in December.

23. Floribunda. — H. and B. Kunth, 3. 283. Leaves
scattered, acuminate, two inches and a half long,
half an inch wide ; petiole short ; peduncles one
inch long; sepals pubescent without, tubercled at
the point, an inch long (red ?) petals a little longer,
(yellow ?) spotted with red ; germen and style
downy. On the temperate Andes near la Palmilla.

24. Anceps. — Flor. Per. 3.61. Stalk slender, leaves
alternate, lanceolate, spirally tipped. ; petiole short ;
peduncles about 17 ; perianth equal, intensely pur-
ple, within dark yellow with blackish purple marks,
in the woods of Iluassahuassi on the Andes of Peru.

I. Leaves smooth, flowers about an inch long, few.

25. Cornigera. — PI. 17. fig. 2. Specim. Matthews,

AMARYLLIDACEvE.

117

1659. Herb. Hooker. Stalk slender, leaves about
2\ long, from \ to f wide, lance-ovate, acute, pe-
tioles about ^ ; peduncles 1-4, naked, about 1^
long ; perianth not an inch long, (seemingly yel-
lowish ?) sepals horned ; involucral bractes, one
larger, the rest small and acute. This plant is
very unlike any other species. It is horned like
cornuta, but very dissimilar in other respects. Its
native spot is not mentioned, but it is a Peruvian
plant. Var. Matthewsiana, PI. 10. fig. 4. Mat-
thews, 1659. Herb. Lindl. seems a different plant.
It has larger leaves, four large involucral bractes,
and four flowers, and, if horned, I overlooked the
horns.

26. Ilalliana. — PI. 10. fig. 3. Specim. Col. Hall. 20.
Patacocha, Herb. Hooker. Leaves about 4|- inches
long, above half an inch wide, spirally acuminate ;
bractes narrow, near an inch long; peduncles about
six, ljlong; perianth an inch, seemingly yellowish
red; sepals oval ; petals wide, spatulate, an inch
long ; stalk, peduncles and germen pubescent.

K. Leaves smooth, flowers large , crowded.

27. Superba. — PI. 6. fig. 1. Specim. Matthews, Peru,
1663. Herb. Hooker. Stalk smooth ; leaves two
inches or more, ^ wide, closely nerved ; bractes
smaller, peduncles about 12, 1^ inch long, perianth
If long, seemingly red and orange; it does not
seem to have spots or green tips ; sepals obtuse,
petals spatulate, both about 5-16ths wide. A beau-
tiful plant.

L. Leaves pubescent, flowers about half an inch, crowded.

28. Setacea. — Flor. Peruv. 3. 62. t. 292. Spec. Ruiz,
Herb. Lambert. Stalk 3-4 yards ; leaves lanceo-
late, 1^ inch long, ^ wide ; bractes small, lanceo-
late, acute ; peduncles 18-35, 1-2 inches long, with
a setaceous bracte below the middle, perianth
scarcely half an inch, sepals purplish red, petals
yellow. In the woods near Pillao on the Peruvian
Andes.

29. Tomentosa. — Flor. Peruv. 3. 62. t. 292. Stalk 4
feet, leaves 1-3 inches, 1 inch wide, closely nerved

1 18

AMARYLLIDACEJE.

underneath ; bractes small ; peduncles numerous,
bracteate at the base of the flower ; perianth small,
sepals scarlet, petals yellowish. In the woods at
Munna and Pillao of the Peruvian Andes.

Yrar. 2. Ebracteata. — Specim. Peru, Matthews, 1666.
Herb. Hooker. With peduncles non-bracteate,
the leaves three-quarters of an inch wide.

Var. 3. Pangoensis. — Foliis angustioribus, umbella pau-
ciflora. Spec. Peru, Matthews, 1162. (from the
woods of Pangoa, July) Herb. Hooker. Leaves
about 2^ inches long, scarcely § wide, more atte-
nuated and acute ; peduncles without bractes there-
on, 1^ long, about six ; perianth f long.

32, Denticulata. — Flor. Peruv. 3. 62. t. 293. Stalk
smooth, leaves 1-3 inches long, 1 inch wide, densely
pubescent underneath; petioles 2 lines, repand den-
ticulate ; bractes small, peduncles numerous, pe-
rianth small (about |), reddish and yellow, seg-
ments equal ; petals cuneiform. On the loftiest
cold mountains near Patasaria in the Peruvian
Andes. The figure in the Flor. Peruv. gives the
margin of the leaves denticulate, which is probably
an error. It seems closely allied to tomentosa, but
the petals are probably a little less spatulate.

M. Leaves pubescent, flowers about an inch, and crowded.

31. Bredemeyerana. — De Schlechtend. fil. ap. R. et
Schultes. Stalk thickly pubescent ; leaves lan-
ceolate, glaucous underneath ; petioles short ;
peduncles numerous, pubescent, an inch long ;
involucral bractes 1 inch, reflex; perianth an inch,
sepals a little shorter. Near Caraccas. See 21
Fanningiana, a native of the same country, glabrous
with longer flowers.

32. Rosea. — Flor. Peruv. 3. 91. Stalk 5 feet high,
smooth ; leaves alternate, ovate-lanceolate, acute ;
petioles short ; involucral bractes many, linear ;
peduncles about 18 with a subulate bracte at the
middle; sepals rose-coloured with a green tip,
petals greener with black spots towards the end.
In the woods of Huassahuassi, Peru.

33. Purpurea. — Flor. Peruv. 3. 63. t. 294. Stalk 18 feet

AM ARYLLlDACEvE.

119

high, pubescent upwards; leaves lanceolate, a palm
long, near an inch wide; bractes narrow ; peduncles
*20-60, 1| inch, sometimes bracteate in the middle;
perianth an inch long, purple, sepals cuneiform,
petals spatulate.

N. Leaves pubescent, flowers about an inch, few.

34. Simplex. — PL 15. fig. 5. Spec. Matthews, 786.
Herb. Hooker, misnamed by Mattli. tomentosa.
Leaves 2^ inches long, ovate, above an inch wide ;
involucral leaves very small ; peduncles five ; a
small bracte thereon towards the middle ; perianth
an inch long. This plant is much more allied to
ovata of Cavanilles than to tomentosa, but it differs
in having one-flowered peduncles, though the
bractes indicate a disposition to form a secondary
one. Found near Purruchuco in Peru.

O. Leaves pubescent, flowers large, peduncles long.

35. Crinita. — PL 15. fig. 4. Spec. Matthews, Peru,

1664. Herb. Hooker. Stalk downy ; leaves
harshly nerved, about 4 inches long, f wide,
lanceolate, acute; petioles near i; involucral
bractes an inch, like hairs or bristles ; peduncles
ten, 6 inches long ; perianth 2 inches long, seem-
ingly red and orange.

P. Leaves pubescent, flowers large, crowded.

36. Crocea. — Flor. Peruv. 3. 61. Stalk 18 feet high,
smooth ; leaves linear, lanceolate, sessile, very
acute, a palm and a half long, 4 lines wide ; invo-
lucral bractes many, broader, but thrice shorter ;
peduncles arcuate, downy; perianth 1^ inch, equal,
saffron-coloured. In the woods at Chumpulla in
the Peruvian province of Tarma.

§§. 2. Perianth very unequal.

Q. Leaves smooth, flowers large, crowded.

37. Caldasiana. — Alstr. Caldasi. H. and B. Kuntli 3.
283. Stalk smooth, leaves ovate, lanceolate, rather
rigid, acuminate, 3-3 inches or longer, an inch
wide or nearly ; peduncles l^inch, sepals tubercled
at the point, scarlet, 9 lines long, petals much more
than an inch, orange spotted with red; germen and

120

AMARYLLIBACEiE.

style downy. In temperate situations near Alan-
gasi, Piso, and Chillo of the province of Quito.

R. Leaves downy , flowers large, crowded.

38. Pardina. — PI. 14. fig. 1. Spec. Col. Hall. 19. Herb.
Hooker. Leaves thickly pubescent underneath,
lance-oval, spirally acuminate, 2\ inches long or
more, 1 or more wide on the upper part of the stem
(inferior leaves unknown) ; peduncles about 20, f
long ; involucral bractes, two outer leaf-like, interior
filiform; sepals If long, petals richly spotted near
2i long. Found at Patacocha on the western
declivity of the Andes, at an elevation of 6000 feet.
A superb species.

39. Patacocensis. — PI. 14. fig. 3. Spec. Col. Hall.
Herb. Hook. Stalk a little downy ; leaves lanceo-
late, lj inches long, ^ wide, frequent near the
summit; involucral bractes broader, few ; peduncles
slender, 30 or more, 2 inches long or more, naked ;
germen small, downy; sepals an inch and 3-16ths,
narrow, obtuse, about 3-16ths wide; petals about an
inch and 9-lGths long, half an inch wide upwards,
spatulate ; longest filaments nearly as long as the
petals. Perianth seemingly reddish yellow? sepals
rather redder? a green spot at the pointof the petals?
Another magnificent plant also from Patacocha at
the same elevation.

S. Leaves downy, flowers large, flowers few.

40. Lutea. — PI. 5. fig. 3. Spec. Col. Hall, 14, No. 4.
Herb. Hooker. Stalk downy ; leaves upwards 2-
long, not \ wide, lanceolate, very little pubescence ;
involucral bractes near an inch long, narrow,
downy ; a bracte above half an inch long, downy
near the middle; germen downy; perianth (“bright
yellow,” Col. Hall.) segments with darker longitu-
dinal stripe ; sepals narrower about 1 4 inch long,
petals near two inches, above half an inch wide
upwmrds. A beautiful species found by the road
to Mindo on the western declivity of the Andes at
an elevation of 9000 feet.

The greater part of these beautiful plants are natives of
elevated situations and dislike a high temperature. They
will be found to thrive best out of doors in this country in

\&cta/nias sstm

AMAR YLLIDACE/E.

121

summer time, and will endure the winter if planted pretty
deep in light soil and covered over with leaves in the cold
season, especially if any sloped heading be laid on to throw
off the wet. Even Acutifolia, which in the greenhouse
keeps its leaves through the winter, will succeed with that
treatment.

The genus Bomarea was named by Mirbel, who divided
it from Alstroemeria, founding the separation on edulis,
erroneously named salsilla, ovata, and multiflora, a species
which cannot be identified and must be altogether erased,
having been described only from a drawing. Poiret (suppl.
encyl. meth. p. bot. t. 1. p. 677) states that Mirbel's charac-
ter of Bomarea was perianth and filaments straight, capsule
rounded and flattened from top to bottom; of Altroemeria sepals
reflex ? (renver sees) filaments curved, capsule lengthened, which
appeared to Poiret insufficient. Mirbel was on the right
scent, but he had not laid hold of the substantial differences.
The filaments of B. salsilla are much bent. The elongation
of the capsule is a weak feature, and varies in the different
species of Alstroemeria.

Suborder 3. Agaves. — Schistandrous, not operculate.

§. 1. Dioscoreasformes. — Inflorescence axillary, spicate.

Plants tuberous, climbing, dioecious, with the
veins of the leaves usually, but not always, con-
fluent. (Petiole articulate ?)

Dr. Lindley relies on the articulation of the footstalk of
the leaf in Dioscoreae, and a small portion of Smilaceae, in-
cluding only Smilax and Ripogonum, of which I believe
but one species is known, for the character of an order, and
even of a group of orders ; but I doubt there being such a
decisive difference between Dioscorea and Bomarea ; and
if it were established, I do not see how such weight can be
attributed to that feature, while articulations of the seed-pod
are not considered of sufficient importance to effect a like
separation. I believe that the axillary inflorescence will be
found a constant feature in the plants I have characterized
as Dioscoreaeformes. Dioscoreae may *be taken as a third
suborder, and Agaveae confined to Ixiaeformes and Agavi-
formes, if it be thought preferable. It is immaterial, except
with a view to consistency, when the other orders shall have
been rectified and arranged.

122

AM ARYLLIDACEAS.

19. Tamus. — Perianth deeply cleft, patent, regular ; con-

tracted above the germen in the female ; pericarp
with the outer coat valveless, the flesh-coat pulpy,
the inner coat with valves bearing the dissepiments ;
seeds with horny albumen, (having some resemblance
to those of Bomarea.) Plants with great tuberous
roots ; veins of the leaves confluent.

1. Communis. Engl. Bot. 91. — -2. Cretica. Willd.

20. Testudinaria. — Perianth deeply cleft; limb regular,

half-patent, reflex ; filaments erect, subulate, enlarged
at the base, inserted in the base of the segments ;
anthers roundish, erect, adnate. Natives of Africa ;
root with square angular prominences. The male
inflorescence is spiked, the female flowers said to be
usually solitary. The veins of the leaves are con-
fluent.

1. Elephantipes. — Bot. Reg. 1 1.921. — Tamus. Bot. Mag.

23. 1347.

2. Montana. — Burch, trav. Leaves glaucous.

21. Dioscorea. — Perianth deeply cleft {Hook. Bot. Mag.

55. 2825.) in six segments, inserted in a bed shaped
like the nave of a wheel on the disk or upper surface
of the germen {Salisb. Par. Lond. 75.) ; filaments
erect, equal ; inserted {according to Salisbury , and the
engraving Bot. Mag. 2825.) in the bed, {according to
Brown , Prod.) into the base of the segments ; {Qu.
which is correct ? the same species , D. bulbifera , being
described by both) ; anthers roundish ; style (or perhaps
stigma) in the female deeply trifid ; ovary 3-celled ;
cells 2-3-seeded ; capsule compressed, with two of
the cells sometimes abortive ; seeds with a winged or
foliaceous margin.

Obs. If in truth there be species of Dioscorea differing
from the rest in having the filaments inserted into the base of the
segments, they should be transferred to Testudinaria, Dioscorea
being distinguished by filaments inserted in the nave-shaped bed
or disk ; if, as Dr. Brown declares, the filaments in all are
inserted into the base of the segments, Testudinaria can be
scarcely separated by its angular root, and I can see nothing
else to distinguish it. Sir W. Hooker s engraving of Cinnarno-
mifolia {Bot. Mag.) represents the perianth , as he also describes

AMAItYLLIDAClliE.

123

it, to be only deeply cleft, and falsifies the generic character
which calls it Sexpartitum, and it represents the filaments, as
Salisbury described them, inserted in the bed,. I have had no
opportunity of ascertaining from fresh specimens to whom the
error is attributable.

1. Pentaphylla. Willd. Rheede. Mai. 7. 35. — 2. Tri-
phylla. Willd. Jacq. ic. t. 627. — 3. Quinqueloba.
Willd. Ksempf. ic. t. 15. — 4. Braziliensis. Willd. —
5. Heterophylla. Hort. Beng. — 6. Aculeata. Willd.
Rheed. Mai. 7. t. 37. — 7. Rubella. Hort. Beng. —
8. Purpurea. Hort. Beng. — 9. Angustifolia. Spreng.
— 10. Nummularia. Willd. Rumph. Amb. 5. t. 162.
— 1 1. Alata. Willd. Rheed. Mai. 7. t. 38. — 12. Bul-
bifera. Par. Lond. 17. — 13. Crispata. Hort. Beng. —
14. Altissima. Spreng. Plum. ic. 117. f. 2.- — 15. Co-
riacea. Spreng. — 16. Angelina. Hort. Beng. — 17.
Pulchella. ib. — 18. Atropurpurea. ib. — 19. Glo-
bosa. ib. — 20. Fasciculata. ib. — 21. Glabra, ib. —
22. Sativa. Willd. Rheed. Mai. 8. 51. — 23. Piperi-
folia. Willd. Plum. ic. 117. f. 1. — 24. Nepalensis.
Sweet H. Brit. — 25. Cinnamomifolia. Bot. Mag. 55.
2825. — 26. Quaternata. Pursh. Walt. fl. c. 246. —
27. Villosa. Jacq. ic. 626. Quinata. Walt. Panicu-
lata. Mich. fl. A. 2. 239. — 28. Oppositifolia. Petiv.
gaz. t. 31. f. 6. — 29. Lucida ? Brown. Prod.

The tubers of Dioscorea are eaten and called yams. The
flower is very insignificant. Jacquin’s figure of D. villosa
makes the filaments bifid, each point bearing a detached cell
of the anther which is bipartite. I cannot believe such an
extraordinary feature, if correctly given, to be only a diffe-
rence of species. A. Dioscorea, with large cordate leaves, at
the Chelsea Garden, has 13 strong longitudinal veins, and
the cross veins confluent, a little oblique, branching in vari-
ous directions, some retroflex. Fifteen species, of which I
have examined dry specimens, have the cross veins confluent,
but those of lucida are most correctly parallel, very rarely
bifid at the point, but in such case continuing their direction
to the next rib or the margin, and never retroflex or running
towards another parallel vein. Its fruit not having been
seen, it is possible that it may belong to a separate, but
closely allied, genus.

22. Rajania. — Perianth deeply cleft, half-patent ; contracted

124

AMARYLLIDACE,®.

above the germen in the female ; capsule with a large
incurved wing on one side ; seed roundish, with a
prominence. See PI. 43. f. 53.

Obs. — I understand the perianth to be undivided at the
base, the only feature mentioned, except its capsule and
seeds, by which it can be distinguished from Dioscorea.

1. Hastata. Willd. Plum. Am. 84. t. 98. — 2. Cordata.
Willd. Plum. ic. 155. f. 1. — 3. Quinquefolia.Willd.
Plum. ic. 155. f. 2.

A leaf from the Chelsea garden, said by its able conductor
to be certainly that of a Rajania (perhaps hastata), has three
strong veins tending to the point, two weaker near the mar-
gin, tile cross veins oblique and confluent, but none decidedly
retroflex. Quinquefolia has three principal veins, and the
lesser intermediate veins are represented oblique and feather-
like, but not confluent. It must be observed that even the
species of Dioscoreseform plants, which agree in not having
the veins parallel, disagree amongst themselves as to their
actual direction. When I alluded (above, p. 51) to the possi-
bility of detaching Smilaceae by an axillary inflorescence, 1
should have modified the term so as to include the inflores-
cence of Ruscus, which is borne on the margin of the leaf
instead of the axill. That circumstance does not however
interfere with the separation, being a still further removal
from the other division, but it renders it more difficult to
characterize the separation by a single word.

§. 2. IxiiEFORMEs.' — Root bulbous ; pericarp not pulpace-
ous, (seeds with a horny albumen and hard
shell ?)

23. Bravoa. — Stalk spiked, bracteate ; germen oblong,
faintly triangular ; perianth persistent ; cylindrical,
with the mouth of the tube widened, the segments of
the limb very short, alternate segments smaller ; fila-
ments inserted at the base of the tube, filiform ; an-
thers versatile ; style thickened ; stigma triangular ;
capsule oblong, 3-celled, 3-valved ; seeds numerous,
kidney-shaped, in two rows.

1. Geminiflora. — PL 12. f. 8, 9. Spec. Herb. Lindl. De
la Llave et Lexarsa, Nov. veg. desc. f. 1. 1824.
Caetocapnia geminiflora (et Zetocapnia) Link et
Otto. PI. rar. h. r. Berl. p. 35. ic. 18. 1828.

AMARYLLIDACEvE.

125

Peduncles axillary in pairs ; perianth about an inch
long, red ; style just longer, stamens just shorter than
the perianth ; root said to be solid, that is, not a tuni-
cated bulb, but a pear-shaped tuber, with inferior
fasciculate small tubers; leaves keeled, spotted at the
base. Native of Mexico.

24. Ixiolirion. — Stalk bracteate with peduncles axillary or
terminal, 1-2 flowered; germen oblong, erect; peri-
anth deeply cleft, (I believe not divided) regular,
half-patent ; alternate segments equal ; filaments
straight, inserted at the base of the segments, alter-
nately equal; anthers versatile; style straight; capsule
oblong, striped, 3-celled, 3-valved ; seeds numerous,
oval-oblong.

1. Montanum. — PI. 20. fig. 3. Spec, ex Aleppo. Herb.

Lamb. — Am. montana. Red. lil. 241. Labill. Syr.

dec. 2. p. 5. t. 1.

2. Tataricum. — PI. 19. Spec, ex Altai. Herb. Lamb. —

Am. Tatarica. Pall. it. 3. ap. 85. t. D. f. 1.

Var. Scythica. — PI. 20. f. 1. Specim. Cyananthes. Pallas.

MS. Herb. Lindl. ex Scythia.

Montanum and Tataricum, as far as can be judged from
the dry specimens, approach very closely to each other in
the flower, which is probably blue in the whole genus ; but
Montanum, the Syrian plant, is distinguished by the remark-
able feature of having a partly spiked inflorescence, contrary
to the usual habit of the order, while the flowers of all the
specimens 1 have seen of Tataricum are terminal. Redoute’s
figure represents Montanum with flowers from the axills of
the bractes, though not so decidedly as the specimen I have
sketched.

Tecopiiilea? — Root small, bulbous, pear-shaped, with
thready membranaceous integuments ; leaves 1 or
few; stem 1 -few-flowered (perhaps never more than
2) germen oblong ; perianth deeply cleft (regular?
half-patent ?) filaments, according to Bertero, three
fertile, three abortive barren. I have not been able
to see the filaments, and cannot learn whether the
sepaline or petaline are abortive, nor whether they
are ever perfect. I believe it rather to be a true
triandrous plant with three processes between the
stamens, and referable to Irideae.

126

AMARYLEIDACETE.

1. Violaeflora. — PI. 24. f. 16, 17. Spec, ex Chili. Gum-
ming et Bridges. Herb. Hooker. — Bertero lett.
Bullet, de sci. nat.

§. 3. Agaviformes. — Semibulbous or stemmed; seeds flat,
with a black or brown shell.

25. Fourcroya.— Bulb imperfect, or long-lived stem ; ger-
men oblong, attenuated below, drooping ; perianth in
six segments, regular, nearly patent, alternate seg-
ments equal ; filaments (short) subulate, obovately
thickened at the base, conniving ; anthers versatile ;
style straight, hollow, triangular, enlarged below ;
stigma triangular, fringed ; capsule oblong, oval,
3-valved, 3-celled ; seeds black, numerous. [/ have
had no opportunity of ascertaining with my own eyes
whether the perianth is absolutely divided, but rely on
the report of Zuccarini .] See PI. 34. fig. 20-25. Cer-
tainly distinct from Agave.

1 . Longaeva. — ( Beyond all comparison the most magnificent

plant in the order .) Zucc. IN ov. Act. Phys. Med. Ac.
Cses. Leop. Car. v. 16. p. 2. 666. ic. 48. Stem
forty feet high, leaves in the form of Dracaena or
Charlwoodia, less rigid and erect than Y ucca ; inflo-
rescence terminal, 30 feet high, pyramidal, lower
branches thereof 12 to 15 feet long ; flowers white,
innumerable, much like those of gigantea. Sup-
posed to be 100 years, in all probability at least
50, coming to maturity. Native of the summit of
Mount Tanga in Mexican Oaxaca, 10,000 feet
above the sea, on slopes amongst oaks and arbutus,
where the water is frozen for a long continuance in
winter. Flowers in May and ripens fruit the fol-
lowing winter. There can be little doubt that this
marvellous plant would flourish in this country.
Karwinski has introduced it into Europe, and it
may be seen alive at Mr. Loddiges at Hackney.

2. Gigantea. — Bot. Mag. 48. 2250. — Dec. PI. grass.

126. Agave foetida. Willd. — 3. Tuberosa. Comm.
Hort. 2. f. 19. — 4. Cubensis. Jacq. Am. pic. t.
260. f. 25. — 5. Rigida. Haworth. Syn. — 6. Austra-
lis. ib. — 7. Madagascariensis. Horneman. — 8. Can-
tala. Haworth Suppl. China.

Mr. Decandolle rejected the genus Fourcroya because he

AM ARYLLIDACE/E.

127

had seen the filaments shorter and longer in different flowers
of Agave vivlpara; which is a most inconclusive reason.
Length of stamens is a very variable feature, insufficient, I
should think, in most cases to support a generic character,
but that of Fourcroya does not depend on it. I have inserted
filaments short in a parenthesis, as a feature probably per-
vading the whole genus, though not necessarily. A plant
would be equally a Fourcroya, however long its filaments
might be, with the formation described.

26. Agave. — Bulb imperfect, or long-lived stem ; perianth
tubed, limb not patent ; filaments adnate to the tube,
filiform (long) ; anthers long, versatile ; style trian-
gularly filiform ; capsule attenuated at both ends,
3-valved, 3-celled ; seeds black, numerous.

§. 1. Germen cylindrically oblong, pedunculated, hori-
zontal ; limb funnel-shaped ; filaments inserted
at the mouth of the tube ; stigma capitate,
scarcely lobed. Genus Chloropsis? Milii. Spec.
Lurida. PI. 33. f. 16-19.

§. 2. Germen oval, sessile, nearly horizontal ; limb revo-
lute; filaments inserted in the base of the seg-
ments ; stigma inconspicuous. Genus Littgea ?
Haworth. Spec. Geminiflora.

§. 3. Germen oblong-oval ; perianth erect or assurgent,
limb pressed against the filaments ; stigma
3-lobed, lobes emarginate. Genus Agave. Spec.
Virginica.

This genus requires to be carefully re-examined. I can-
not allot the species to the three sections, but have stated the
species which are their types, having had no opportunity of
inspecting them.

The flowers of five species named by Zuccarini are not
described. They are at present only known by the leaf.

Stemless; leaves (jlaucous. — 1. Americana. Andr. Rep.
438. — 2. Milleriana. Flaworth Syn. — 3. Mexicana. Haw.
Sup. — 4. Fkiccida. Haw. Syn. — 5. Brachystachys. Red. lil.
485. — 6. Spicata. Cavan. Bibl. reg. Matr. ic. ined. — 7.
Yuccaefolia. Red. lil. 328.329. — 8. Virginica. Bot,. Mag. 29.
1157. — 9. Potatoria. potatorum. Zuccarini. 674. Scape 5-6
feet, growing the third or fourth year. — 10. Heteracantha.
ib. 675. — 11. Macracantha. ib. 676. — 12. Pugioniformis.

128

AMARYLLIDACEiE.

ib. 676.- — 13. Karwinskiana. ib. 677. Caulescent. — 14.

Lurida. Bot. Mag. 37. 1522. — 15. Angustifolia. Ilaw. Syn.
Leaves green. — 16. Keratto. Haw. Syn. — 17. Vivipara. Comm,
prael. t. 15. — 18. Polyacantha. Haw. Syn. — 19. Gemini-
flora. Bot. Reg. 14. 1145. — 20. Striata. Zucc. 678. nearly
allied to Geminiflora.

27. Doryanthes. — Bulb imperfect, long-lived ; stalk capitu-

late, bracteate; perianth deciduous, tubed, (on the
faith of Schultes and the fig. in the Bot. Mag., Brown
and Sims being both silent as to the tube) funnel-
shaped ; segments nearly equal, reflex ; filaments
subulate, adnate to the base of the segments and tube ;
anthers long, erect, affixed at the base, which is
shaped like an extinguisher ; style 3-furrowed ;
stigma triangular; capsule turbinately ovate, 3-celled,
3-valved ; seeds with a brown wrinkled foliaceous
shell.

1. Excelsa. Bot. Mag. 41. 1685,

The bulbs of this plant, though imperfect, may be brought
dry from Australia to Europe without injury. They seem im-
patient of much water. Flower tubed. Corraea Linn. Trans.

Third Division., — SCAPACEOUS.

Scape succulent , spathaceous , not articulate below the spathe.
Suborder 4. Amaryllideje. Schistandrous, not operculate.

Petaline filaments, unless equal, excelling.

j' Cavae. — Scape hollow ; seeds compressed with a black
shell, capsule 3-celled, 3-valved.

§. 1. Cyrtanthiformes. — Tube wide-mouthed; germen
oval, pedunculated. South African plants.

28. Cyrtanthus. — Germen pendulous or declined; tube

curved, narrow funnel-shaped, often a little ventri-
cose ; segments short ; filaments straight, decurrent,
inserted in the upper portion of the tube, the sepa-
line scarcely lower than the petaline; anthers attached
at the upper third part, suberect ; style curved down-
wards.

A. With persistent leaves.

1. Obliquus. — Bot. Mag. 28. 1133. Flowers ten or
more, pendulous, large, orange, yellow, and green ;
leaves an inch wide, subglaucous.

I

AMARYLLIDACE.E.

129

2. Carneus. — -Dot. Reg. 17. 1462. Flowers eight or

more, pendulous, subventricose, 3 inches long, nar-
rower than those of obliquus, pink, paler at the
base ; leaves blunter than those of obliquus ; fila-
ments inserted higher.

B. With deciduous leaves. To he kept dry in winter.

3. Pallidus. — Bot. Mag. 51. 2471. Flowers about five,

pendulous, subventricose, dull pink, paler upwards;
leaves dark green, about \ wide, acute, attenuated
at both ends. — Never having possessed this plant
I am not certain that it is deciduous.

4. Collinus. — Bot. Reg. 2. 162. Flowers about nine,

pendulous, subventricose, poppy-scarlet ; leaves 7
or 8 inches long, glaucous, above ^ wide, subacute,
attenuated below, purplish at the base.

5. Spiralis. — Bot. Reg. 2. 167. Flowers about seven,

pendulous, subventricose, orange-red, yellow below;
style shorter than the tube ; leaves spiral, glaucous,
about ~ wide.

6. Ventricosus, Willdenow. Angustifolius, Jacq. H.

Sell. 1. p. 40. t. 76. Flowers declined, orange-red,
subventricose, wider than in spiralis ; style much
longer than the tube ; leaves not spiral, green.

7. Striatus. — Bot. Mag. 52. 2534. Flowers 3 or 4,

pendulous, 2^ inches long, narrow funnel-shaped,
red striped with yellow ; filaments shorter than the
style, longer than the perianth ; leaves a foot long,
half an inch wide, subacute, speckled with red be-
low.

8. Angustifolius. — Bot. Mag. 8. 271. Flowers 4 or 5,

pendulous on one side, narrow funnel-shaped,
orange-red; style just longer than the limb and
filaments ; leaves above | wide, subobtuse, red
below.

9. Odoms. — Bot. Reg. 6. 503. Flowers about 4, crim-

son, fragrant, declined, narrow funnel-shaped ; fila-
ments short ; style half an inch longer ; leaves
linear, 1-1 2th wide.

10. Lutescens. — Albo-hiteus. Burchell. Herb. 7144.

Monella ochroleucra. Herb. App. lapsu. Closely
allied to odorus, but in all the specimens perianth

K

130

AMARYLLIDACEvE.

yellowish white. Anthers almost sessile, three just
within the tube, three 3-8ths of an inch below it.
Native specimens 2-3-flowered. Found at the
mountain station between Zwellendam and George
by Dr. Burchell. 1 regret that by an accidental
omission I have not mentioned Dr. Burchell’s name
amongst the gentlemen to whom I am indebted.
To him I owe much for the inspection of his herba-
rium and drawings, as well as the kind commu-
nication of some of the bulbs he imported.

Cyrtanthus obliquus appeared to differ so much from the
rest in the structure of the flower, that I was formerly in-
duced to think it generically distinct, and adopted Mr. Salis-
bury’s MS. name Monella for the other species then known.
Soon after, I saw reason to doubt the propriety of the sepa-
ration, and to believe that this genus admits greater varia-
tion of structure than most others in the same order ; and
that, as the filaments are decurrent, it is not essential whe-
ther the union with the tube is continued a little further or
not in different species. In Vallota purpurea, of which the
filaments are only adnate, there is a difference sometimes as
to the point of adherence in different flowers of the same
umbel. The subsequent appearance of C. Carneus with
persistent leaves scarcely distinguishable from those of obli-
quus, and flowers more like pallidus, confirmed me in consi-
dering the plants with deciduous leaves to have been impro-
perly detached. The diversities consist in the tube being a
little ventricose in some species, the angular ribs and chan-
nels which mark that of obliquus, the more or less prolonged
adhesion of the filaments to its sides, and the curved or more
erect habit of the peduncle.

They are altogether plants of difficult culture, the bulbs
being more disposed to dwindle and rot, than to increase in
bulk. Mr. Griffin was, I think, more successful than most
others in the cultivation of C. obliquus, of which he had
m my strong bulbs on a shelf very near the glass in his stove,
where the heat was never great. A common greenhouse is
usually too damp for it in winter, and the air of a hot stove
too confined. A light soil which is not retentive of water
will be found to suit the whole genus ; and I think that the
use of peat will always be dangerous to them. Those with
persistent leaves should be cautiously watered in winter, the
deciduous species not at all. C. Carneus is one of the most

AMARYLLIDACEiE.

131

difficult to manage ; twice I lost it, notwithstanding the
greatest care, and have at last succeeded in establishing one
with better hopes by giving it water very seldom the first
year, and rather more after it had formed a strong leaf,
keeping it as much as possible in a draft of air in the green-
house. It is planted in a mixture of white sand with a little
light loam, with an open under drain.

With respect to the other species there is some peculiarity
in the soil congenial to them which it is very difficult to
analyze. When I lived at Mitcham in Surrey, C. angusti-
folius was a weed with me, ripening seed freely, and the
seedlings quickly came to a flowering age and were vigor-
ous, being potted in the soil of Mitcham common, which
was a light brown earth with a little admixture of dead furze
leaves on a gravelly substratum. Since I have lived in
Yorkshire I have been able to find no soil that suited it, and
although many changes were tried, the plants dwindled and
all perished : nor have I found any species of Cyrtanthus
succeed well in the soils to which I have access here. Mr.
Rollisson had equal success with C. angustifolius at his
nursery at Tooting near Mitcham. C. lutescens has, I be-
lieve, never been in Europe, but Dr. Burchell has many
specimens of it. It has very narrow leaves, and comes very
near to C. odorus except in its colour, which is invariably a
yellowish white. Ventricosus, figured by Jacquin under the
name angustifolius, is only known to us by his plate and
description. It was probably one of Masson’s plants from
the East coast, and is allied to Collinus. Mr. Ker conceived
that Jacquin had by mistake represented a scape of spiralis
with the foliage of angustifolius ; but it is evident that his
plant has not the inflorescence of spiralis.

The recollection that Hippeastrum equestre, single and
double, which will not exist in the light soils to which I have
access in Yorkshire, throve exceedingly with me at Mitcham
in Surrey, in the same soil that peculiarly suited Cyrtanthus
angustifolius, and that all the Cyrtanthiform bulbs are
disposed to rot in light earth at Spoff'orth, persuades me that
wherever their cultivation is found difficult, a soil that is
more disposed to set firm, and not fall to pieces when turned
out of the pot, should be substituted, with good drainage
and cautious watering. The difficulty is to find a light soil
which has a little tenacity. There is a yellow earth of that
nature in which I have observed Erica cinerea thrive with

k 2

132

AMARYLLIDACEiE.

much greater vigour than in any black soil, in the neighbour-
hood of the New Forest, which would perhaps suit the
Cyrtanthi. In a soil of that nature all Mr. Woodford’s
Eric® were cultivated at Rickmansworth. The earth of
Mitcham common was so congenial to the Ixias that in it
I have had 72 flowers from one bulb of Ixia longiflora, and
nearly as many from one of Sparaxis grandiflora, whereas
the confluent soils of this neighbourhood, though favourable
to the hardier Gladioli, destroy the Ixias and Babianas and
are not favourable to Sparaxis.

29. Gastronema. — Germen declined ; tube below slender,
curved above, wide-campanulate ; limb short, reflex;
filaments decurrent, conniving; three upper longer,
incurved ; the petaline inserted at the top, the
sepaline near the middle, of the tube ; anthers short ;
style declined, pressed against the lowest petal.

1. Clavatum. — Bot. Mag. 49. 2291. Cyrtanthus uni-
dorus. Bot. Reg. 2. 168. Amaryllis clavata.
L’Heritier. Sert. Angl. Amaryllis pumilio. Hort.
Kew. Am. tubiflora. Specim. Herb. Soc. Linn,
et Herb. Banks, et MS. Bibl. Mus. Brit. — Flowers
1 or 2, white striped with red ; leaves slender,
attenuated below, dark green.

This very pretty little bulb, which is nearly akin to Cyr-
tanthus, but presents strong points of difference, appears to
have been first described by L’Heritier under the name A.
clavata. The name clavatum has therefore priority over
pumilio and uniflorum, which last is incorrect as it often
bears a two-dowered scape, of which there are three or four
specimens in Dr. Burchell’s herbarium, and he assures me
that lie found it with one or two dowers promiscuously in
Africa. Mr. Ker having conceived that Amaryllis pumilio
Hort. Kew. was a different plant and of a different genus,
and that there existed a specimen of it in the Banksian
herbarium, I have carefully inspected the Banks, herb, and
MSS. with the kind assistance of Mr. Bennet, and the result
is decisive that A. pumilio Hort. Kew. is G. clavatum, and
the specimen marked A. pumilio in the herbarium Oporan-
thus luteus. In the herbarium of the Linnaean Society there
is a specimen of G. clavatum marked Amaryllis tubidora,
with a reference to a MS. in the Banks, libr. In the Banks,
herb, is a like specimen marked A. tubidora by Dr. Solander,

AMARYLLIDACE^E.

133

which is erased by a line of the pen, and Amaryllis clavata
added by Dryander. It is memorandumed as growing in
the fields beyond Camtours rivier. In the MS. catalogue
pumilio is also erased and clavata substituted by Dryander.
A. pumilio was the name given to this plant in the Hort.
Kew. when it had been ascertained that it was not the tubi-
flora of L’Heritier ; and when Dryander afterwards per-
ceived its identity with A. clavata of L’Heritier, the right
name was substituted. Whether the description in the Hort.
Kew. be correct or not, these entries are decisive as to their
identity, and Southern Africa has not produced auy other
one-flowered plant of this suborder, except gethyllis. In
plate 21. fig. 2. I have given an exact outline of the specimen
in the Banks, herb, marked Amaryllis pumilio Hort. Kew.
It is a garden specimen from Kew, which seems to have been
shrivelled before it was laid in, and is much damaged ; with
it there is a fragment of a leaf reversed, with top and bottom
written by Dryander. There is no entry concerning this
specimen in any of the MS. books. Fig. 3. I have placed
beside it an exact outline of a flower of Oporanthus luteus
better preserved, and it will be apparent to the most unprac-
tised eye that they are specimens of the same species.
Whether the erroneous description of A. pumilio in the
Hort. Kew. be from the pen of the younger Linnaeus, as cited
by Mr. Ker, or (as I have been told) from that of Dr.
Solander, it is needless to inquire ; it will be evident to any
person who will examine and compare it with G. clavatum
and the specimen marked A. pumilio in the Banks. Herb,
of which I have given an outline, that the description is an
amalgamation of the two, probably in consequence of an
attempt to reconcile the description first made from G.
clavatum with the garden specimen of Oporanthus, erro-
neously laid in under name of A. pumilio, and so confounded
with it. At all events Amaryllis pumilio must be expunged
as a non-entity.

G. clavatum requires to be kept dry in the winter. Dr.
Burchell’s bulbs flowered well in his garden border, having-
been set there in the spring, but they all perished afterwards.
30. Vallota. — Germen erect; tube straight, wide; limb
funnel-shaped ; filaments conniving, adhering by one
side only to the tube (the petaline to the summit of
the tube or even to the petals, the sepaline lower ;)

134

AMARYLLIDACEjE.

anthers affixed at one-third from the bottom, or rather
nearer to the middle, suberect; style declined.

1. Purpurea. — Amaryllis purpurea. Bot. Mag. 35.
1430. Am. elata. Jacq. H. Sch. 1. 32. t. 62.
Am. Speciosa. L'Herit. S. Aug. 12.

Var. 2. Minor. — Am. purpurea. Bot. Reg. 7. 552.
Flowers smaller and paler.

This beautiful plant is so closely allied to Cyrtanthus,
that I have even entertained doubts of its being distinct, and
should wish to see it ascertained by further experiments
whether it is incapable of mingling with that genus. Those
who confound it with Hippeastrum are quite in error. So
ill are the affinities of plants in this order understood, that I
observe some Botanists persist in uniting it with Amaryllis,
from which it is widely removed. Dr. Burchell told me that
it was the only bulb of the order that he had found growing
in boggy peat in Africa. It delights so much in wet, that it
will thrive even in water. Its decayed coats are, however,
so very retentive of moisture, and impenetrable to the air,
that the plant does not flourish in the stove or greenhouse,
unless they are pulled oft’ if the bulb be above ground : and,
although it is very thirsty, the bulbs are often lost by
rottenness, and are very apt to decay on the voyage from the
Cape. I think in a pot it succeeds better in loam or loam
and peat mixed than when planted entirely in peat. It
must be kept always growing, and shows no dislike to the
stove in winter. All Cyrtanthiform pollen is small.

§. 2. Hippeastriformes. — Germen triangular, wider above,
constricted in the middle. Tube narrow-mouthed.
Capsule 3-celled, 3-furrowed, 3-valved. Occidental
plants.

31. Sprekelia.- — Jacobean lily. Perianth declined, tube
scarcely any, upper segments reflex, lower sloped
downwards, convolute at the base ; filaments inserted
equally with a connecting membrane at the base of
the perianth, fasciculate, declined and recurved as
the style ; the upper sepaline and lower petaline
shorter than the others. Anthers pendulous, affixed
one-third from the top.

1. Formosissima. — Amaryllis formosissima. Bot. Mag.
2. 47.

AMARYLLIDACE^E.

135

I cannot find that we had any knowledge of the precise
natural habitation of this splendid flower ; but Mr. Bateman
has just communicated to me bulbs collected by Mr. Skinner
in Guatimala, which prove to be this plant.

The bulbs are perfectly hardy and appear to like a low
temperature, but they will not flower willingly unless they
have a season of drought. They succeed well against the
wall of a stove in the open ground, flowering in the spring
and sometimes again in the autumn, if the summer has been
very dry. They rarely blossom if watered through the
winter in a greenhouse, but if kept dry and warm for a few
months, they will flower as soon as they are watered in the
spring. I have twice known this plant produce a two-
flowered scape accidentally, but it is usually one-flowered.
No instance has come to my knowledge of its bearing seed in
this country. Its ovules are longer than those of Hippeastrum,
and the particles of its pollen longer, blunter, and often
bent. See plate 34. fig. 32. The pollen is very abundant
and perfect, but its semination seems to depend upon some
very nice adaptation of temperature and moisture. Num-
berless unsuccessful attempts have been made to cross it with
other genera, especially Hippeastrum and Zephyranthes. It
likes a fertile soil, and the bulbs may be advantageously
planted out in spring, and taken up and dried on the
approach of winter.

32. Hippeastrum. — Perianth declined, tubed ; tube abbre-
viated underneath ; faucial membrane, when mani-
fested, defective on the lower side (not annular as in
Habranthus), the upper sepal wider, the lower petal
narrower; filaments declined, recurved, inserted in
the tube with gradations, the upper sepaline longer
and inserted higher, the lower petaline shorter and
inserted lower; style declined, recurved ; stigma
trifid or triangular.

Calyptratce. — Tube screened.

1. Aulicum. — Am. aulica. Bot. Reg. 6. 444. Flowers
two, crimson, with a green fleshy screen at the
mouth of the tube; sepals narrower than the petals ;
lowest petal outstretched, embracing the filaments.

Var. 2. Platypetalum. — Am. aulica var. Bot. Reg. 12.

13G

AM ARYLLlDACEiE.

1038. Limb wider, fleshy screen less conspicuous,
green star larger.

Var. 3. Glaucophyllum. — Am. aulica var. Bot. Mag. 57.
2083. Sepals nearly as wide as the petals, green
star longer, screen nearly obsolete.

2. Calyptratum. — Am. calyptrata. Bot. Reg. 2. 164.

Flowers twro, green, with faint red tesserated
marks; sepals narrower than the petals; screen
conspicuous.

3. Psittacmum. — Am. psittacina. Bot. Reg. 3. 199.

Flowers two, green, beautifully streaked and
margined with bright red, sepals widest ; screen
conspicuous; upper petal depressed.

Laves vel barbatce. — Tube smooth or bearded.

4. Solandriflorum. — Lindley Coll. Bot. t. 11. var. chlo-

roleucum. Flower green and yellowish white,
eight inches long, tube smooth; style nearly as long
as the perianth ; filaments a little shorter.

Var. Striatum. — Bot. Mag. 52. 2573. Flower from
nine to ten inches long, faintly striped without
with red ; tube from 3 to 4^ inches long.

Var. Rubrituba. — lb. absque icone. Flower greenish
white, tube purplish red.

Var. Conspicuum. — Var. (3 ■ vittata. Bot. Reg. 11. 876.
Flower ten inches and a half long, faintly striped
outside wfith red, tube purplish red.

5. Ambiguum. — Per. tubuloso, segmentis intus rubro-

striatis fauce leviter barbata, stylo filamenta super-
ante, perianthium subaequante.

Var. 1. Longiflorum, — Bot. Mag. ined. 62. 3542. peri-
anthio 8 unciali. Ex Lima Peruviae allatum.

Var. 2. Tweedianum. — PI. 21. f. 3. perianthio sub-6-
unciali. Specim. Herb. Hooker. Ex Brazilia
meridionali ? Tw^eedie.

The first variety with the flower eight inches long, white
with red stripes within and slightly bearded, was sent from
Lima to the Glasgow Botanic Garden, where several of the
bulbs flowered ; and singularly resembled a cross-bred plant

Tloofo. 13.

AM ARYLLIDACE/E.

137

at Spofforth, raised from a complicated intermixture of
Regium, vittatum, and solandriflorum. The second, only
known by Tweedie’s specimen, has a shorter flower, but in
other respects very similar. Flower five inches and f long,
white, with two deep purple stripes, with a white stripe
between, on each segment ; segments acuminate ; tube only
one inch and long ; style nearly equal to the perianth, an
inch longer than the stamens ; stigma trifid.

6. Breviflorum. — PI. 21. f. 4. Bot. Mag. ined. 62.

3549. Specim. Herb. Hooker (ex Braz. meri-
dionali?) Tweedie. Peduncles about 2^ inches,
tube \ of an inch long, limb two inches, white,
striped outside with red, and stained with yellow ;
style \ an inch longer than the stamens, | shorter
than the limb ; stigma trifid, reflex ; lateral sepals
about wide, lateral petals nearly equal to them.
Sent by Tweedie to the Glasgow Botanic Garden,
where it has flowered.

7. Vittatum. — Am. Vittata. Bot. Mag. 4. 129. Flowers

4 to 9, white doubly striped with dull red.

Var. 2. Latifolium. — v. major. Lindl. Coll. Bot. 12.
Larger, with broader and more glossy leaves ;
perianth five and a half inches long, broad dull red
stripe with green star; style five inches long, fila-
ments a little shorter.

Var. 3. Harrisonianum. — Bot. Reg. 12. 988. Tube
greenish, limb obtuse, doubly striped with red
within ; said to grow near Lima.

8. Reticulatum. — Am. reticulata. Bot. Mag. 18. 657.

Flower purplish red, beautifully reticulated with a
deeper colour ; star white. Style nearly as long as
the limb ; pollen very pale ; stigma triangular,
obtuse ; leaves very deep green, attenuated below :
inside of the ripe capsule red ; seeds fewer and less
compressed than usual in the genus.

Var. 2. Striatifolium. — Amaryllis. Bot. Mag. 47. 2113.
Bot. Reg. 5. 352. Flower larger, paler; leaf with
a strong broad longitudinal white line ; capsule
and seeds as in the prototype. Some imported
individuals have the flower small and darker as in

138

AMARYLLIDACEiE.

the green-leaved plant. Am. principis de Salm
Dyk Schultes is evidently this plant.

9. Barbatum. — PI. 21. fig. 1. Crinum barbatum. Linn.

MS. cum spec, ex Surinam in Herb. Linn. Soc.
flore albo, Stella viridi, tubo viridi barbato. Ama-
ryllis. dubia. Linn. Amcen. 8. 254. with reference to
the figure of H. equestre, misnamed Belladonna,
Merian Surinam, t. 22. Tube f of an inch, slender,
green, bearded; limb about 2%, white with a green
star; upper petals 1 and 3-16ths wide, lower 1 and
1- 16th ; lower sepals 13-16ths wide; leaf sub-obtuse,
attenuated at both ends, a little undulated, margin
whitish and recurved; greatest width 1 and 3-lGths.
Flowers three or more. Grows near Surinam.

10. Equestre. — 1. minus. Am. equestris. Bot. Mag. 9.
305. Merian Surinam, t. 22. Flower bright orange,
with a green star; tube cernuous, fimbriated at
the mouth ; upper segments very reflex, lower
protruded; stigma with short round lobes. Surinam,
Trinidad, West Indies.

Var. 2. Majus. — Amaryllis. Bot. Reg. 3. 234. With
much larger flowers, and longer more erect leaves.
Demerara.

Var. 3. Semi-plenum. — Pulcherrima hortulanorum.
Leaves like the prototype; flowers semidouble, found
by Fraser in Cuba near the Havannah : imported
also from Bahama.

11. Stylosum. — Bot. Mag. 49. 2278. ic. nimis obscur.
Am. Maranensis. Bot. Reg. 9. 719. Flowers of a
dull coppery flesh colour, with a short cernuous
greenish tube, fimbriated ; filaments more than half
an inch, style more than an inch longer than the
perianth ; leaves much more glossy than those of
equestre. Grows near Maranham.

I never saw a specimen with the filaments so much
prolonged as in the fig. in the Bot. Reg. and I doubt its
correctness in that point. The name Stylosum has priority
over Maranense, which Schultes has wrongfully preferred.

12. Miniatum. — Flor. Peruv. 3. 57. without a figure.
Flowers three or four, cernuous; tube three lines
long, with six fimbriated white scales at the mouth ;

AMARYLLIDACEiE.

139

sepals obtuse, wider than the petals, upper one reflex ;
lowest petal narrow ; style and filaments equal to the
perianth ; stigma triangularly three-lobed. Peru.
It seems to form a link between Equestre and
Regium. A. Chilensis (Ruiz) is not distinguishable
from this.

13. Regium. — Amaryllis reginae. Bot. Mag. 13. 453.
Flowers scarlet with a green star, cernuous, much
more funnel-shaped than those of equestre ; seg-
ments more equal ; upper sepal not so reflex ;
tube much thicker and shorter ; scape shorter.
Named from having flowered first in England on
the queen’s birthday, in 1728. Said to be from
Mexico ; seemingly native of a hotter climate ; per-
haps from the neighbourhood of Vera Cruz. A
bulb of Skinner’s from Guatimala looks like it.

14. Glaucescens. — Martius apud Schultes. (Without a
figure.) Differs from Bulbuliferum, var. fulgidum, in
having glaucous leaves, and the upper sepal non-
revolute ; from var. acuminatum in having the seg-
ments less acuminatum, and scarcely undulated ; from
Bulbuliferum generally, in having the mouth of the
tube scaly or bearded ; and as it is a native of the banks
of Rio des Contes in the hot province of Bahia, I
doubt its being one of the bulbuliferous family.
It is known only by Dr. Martius’s description. It
is probably a variety of the Maranliam stylosum.
It lias the tube short and thick.

15. Bulbulosum. — This species inhabits the middle
latitudes of Brazil, where it branches into a vast
number of local varieties, some of which have been
described as separate species ; but they all agree in
producing blind offsets round the bulb (some of
which have been known to lie dormant for years
refusing to vegetate), in having orange-coloured
flowers of various tints, with the upper sepal
more or less recurved, and a cernuous tube of which
the mouth is either smooth, or only shewing a very
slight disposition to become bearded, and the leaves
arcuate.

Var. 1. Subbarbatum. — Bot. Mag. 51. 2475. Flowers
highly coloured, sepals much wider, lower petal

140

AMARYLLIDACE7E.

very narrow, tube red with a little vestige of beard ;
leaves 2 to 2^ inches wide.

Var. 2. Unguiculatum. — Martins ap. Schultes. With-
out a figure. Stated to approach closely to sub-
barbatum, differing in a less cernuous tube with a
thin membrane between the filaments; the upper
sepal to be less reflex than in fulgidum. The leaves
agree with those of subbarbatum. It grows in the
glens of Mount Corcovado, and on the hills near
St. Sebastian’s in Brazil.

Var. 3. Rutilum. — Am. rutila. Bot. Reg. 1.28. Flowers
brilliant, tending to scarlet, smaller, tube about
half an inch long, with a minute knob at the
mouth ; leaves under an inch wide.

Var. 4. Fulgidum. — Am. fulgida. Bot. Reg. 3. 22G.
Flowers larger, paler, tube about an inch long,
throat quite smooth.

Var. 5. Pallidum. — Like the former, but the flower
smaller and much paler. Imported some years
ago by Mr. Tate.

Var. 6. Simsianum. — Am. miniata. Bot. Mag. 45. 1943.
Flower red lead colour, tube short, smooth at the
mouth ; leaves growing near the ground, arcuate.
Certainly not the A. miniata of Ruiz and Pavon ;
but so closely allied to Fulgidum, as to leave no
doubt in my mind that it came from Brazil ori-
ginally, though said vaguely to have been from the
West Indies.

Var. 7. Equestriforme. — Flower very similar in struc-
ture to equestre minus, but the mouth of the tube
smooth, the leaves and bulb like Fulgidum. Sent
growing in the midst of some epiphytes unperceived
by my collector from the neighbourhood of Rio.

Var. 8. Acuminatum. — Am. acuminata. Bot. Reg. 7.
534. and 14. 1188. Flowers pale, undulate, acu-
minate, tube smooth ; leaves with a glaucous
bloom.

Var. 9. Pulverulentum. — Bot. Mag. 49. 2273. Flowers
darker, less acuminate ; leaves with a glaucous
bloom. Both the varieties acuminatum and pul-
verulentum grow naturally near Rio, where my

TlaisJL,

v ;.JTerT)afLc[el.

AMARY LLIDACEiE.

141

collector found them growing in black vegetable
earth, with a scape above three feet high. I can-
not distinguish the var. longipedunculata, 1 188,
from acuminata, 534, Bot. Reg. In the latter the
shortness of the peduncles is an error of the artist.

I think there must be a mistake in the account that
longipedunculata was found in a wild state near
San Pablo Quatro-Venados in the State of Mexican
Oaxaca. It seems very improbable that this
variable genus should manifest itself exactly in the
same form in two spots 40 degrees asunder, in the
North and South hemispheres. It is, however,
possible. I believe the bulbs of pulverulentum
from Buenos Ayres to be garden productions.

Var. 10. Ignescens. — Scape and leaves more robust
than fulgidpm, dark green, flowers smaller, more
undulated and crowded, fiery orange, approaching
more to Crocatum in form.

Var. 1 1. Crocatum. — Amaryllis. Bot. Reg. 1. 38. Scape
and leaves robust, flowers smaller and more undu-
lated than fulgidum, upper sepal less reflex. The
flowers of this plant are very subject to the distemper
called the curl.

II. Spathaceum, hybridum, Bot. Mag. 49. 2315. is
Rutilo-Johnsoni, a cross-bred seedling of the same race as
that figured under the name Splendens, Herb. App. — Am.
Braziliensis, Red. Lil. 8. 469. is the English mule Johnsoni
or Regio-vittatum. All other named varieties belonging to
this genus are cross-bred intermixtures, which are almost
interminable ; for all the species appear to breed freely
together ; and their produce is fertile and easily crossed
again. The system of giving a Latin specific name to every
cross-bred seedling, which prevails amongst cultivators, and
has been unfortunately sanctioned to a certain degree by M.
De Candolle, and completely adopted by Sweet in his Hort.
Brit.., where he has overwhelmed the natural species of this
genus by the association of one or two hundred cross-bred
seedlings, bearing names of the like description with those
of the original species, is very objectionable; and it is high
time that the writers of all botanical works should set their
faces decidedly against it. I will enumerate the principal
crosses which have been effected in this genus, retaining

142

AMARYLLIDACE.E.

Sweet’s appellation, where he has used the genitive of a
personal name, but rejecting the Latin adjectives which
ought to be reserved for the natural species, and I have
given such names in their place as will exemplify the
nomenclature I propose for such demi-species. It will be
recollected that I have recommended the change of the few
genitives of proper names which occur amongst the species
into an adjective termination, for the purpose of confining the
genitives to the plants of mixed origin. I consider the two
converse productions obtained by reversing the sexes of the
parents to be included under one name, covering all the
crosses raised from different varieties of the same two species;
for instance that the mixture between Vittatum and any
variety of Bulbulosum should bear the same name, and its
diversities only rank as subordinate varieties. Even wfith
that limitation the artificial demi-species will be very nume-
rous, but florist’s names, such as Juno, Ceres, Camillus,
Napoleon, Alompra, Sigismunda, See., must be given to the
subordinary varieties, and those deserve no place in a
botanical arrangement, belonging property to the catalogues
of cultivators. I think it will be much best to limit the
names of cross-bred demi-species strictly to genitives of
proper names or of the place where they wrere raised. In
the descriptive double name, that which is placed first
indicates the male, but the converse is intended to be
included under it.

Hybrid or mixed Crosses. — 1. Johnsoni, or Regio- vit-
tatum; raised by Johnson, wrho had a small garden in
Lancashire, in 1810; by me at Mitcham, in 1811, and at
Highclere later. Am. Braziliensis, Red. Lil. 8. 469. Am.
Carnarvoni. De Candolle, improperly, since the name
Johnsoni had been previously published in the Hort. Trans,
and was in general use. Johnson’s cross was probably
accidental, for he thought it a mule between Vittatum and
Sprekelia formosissima, and it passed for such during several
years. — 2. Seymouri, or Aulico-vittatum, named from my
present gardener. — 3. Allmanni , or Calyptrato- vittatum;
raised at Colvill’s where I saw it in flower; named from the
Professor of Botany at Dublin. — 4. Andersoni , or Bulbuloso-
vittatum ; named from the conductor of the Chelsea garden ;
very variable, according to the variety of bulbulosum from
which it is raised. — 5. Hookeri, or Goweni-vittatum ; very
beautiful ; named from the Botanical Professor at Glasgow.

AM ARYLLIDACEiE.

143

—6. Grahnmi , or Johnsoni-vittatum ; approximating to
vittatum, but with larger and more brilliant flowers and
tenderer constitution ; named from the Botanical Professor
at Edinburgh. — 7. Digrveedi, or Striati folio- vittatum ; named
from the foreman in the Highclere garden. — 8. Griffini, or
Psittacino-Johnsoni ; very beautiful and variable in its
colour, but always distinguishable from Psittacinum by the
upper segment not being depressed ; the only mule raised by
Mr. Griffin. — 9. Sweetii, or Reticulato-Johnsoni, including
the crosses from Striatifolium ; raised first by Sweet at
Colvill’s Nursery. I have a seedling from one of the Striati-
folio-mules, which so closely resembles in foliage and flower
the original reticulatum that it might be mistaken for it. —
10. Broohesi , or Bulbuloso-Jolmsoni. The cross by the
variety Crocatum was first raised at Brookes’s nursery; that
by rutilum at Highclere; by fulgidum at Spofforth. — 11.
Benthami, or Styloso-Johnsoni ; named from Mr. Bentham,
whose botanical labours are well known; raised at Spofforth;
flowers of a gloomy but variable colour. — 12. Baubenii, or
Griffini-Johnsoni ; a very beautiful cross raised at Spofforth ;
named from the Botanical Professor at Oxford.— -13. Carnar-
voni, or Solan drifloro-Johnsoni; a magnificent cross raised at
Highclere and at Spofforth ; named from my lamented
brother in lieu of regio-vittatum to which Mr. De Candolle
had applied the name inadvertently. — 14. Altaclaree , or
Psittacino-Griffini ; named from Highclere, where it was
raised. — 15. Haylocki, or Solandrifloro-bulbulosum ; raised
at Spofforth, and named by Sweet from my deceased gar-
dener.— 16. Herberti, or Solandrifloro-stylosum ; raised at
Spofforth, and named by Sweet. I have one of this cross
with a pale orange flower, the largest I have seen in the genus,
though that of stylosum is small. — 17. Harrisoni, or Reti-
culato-stylosum ; raised by R. Harrison, Esq. — 18. Barken ,
or Bulbuloso-reticulatum ; named from the foreman at Spof-
forth.— 19. Hoodii, or Equestri-regium ; raised at Colvill’s ;
unknown to me ; named by Sweet. — 20. Henslowi, or regio-
bulbulosum ; raised at Spofforth ; very brilliant ; named
from the Botanical Professor at Cambridge. — 21. Batemanni,
or Equestri-bulbulosum ; raised at Colvill’s, named from J.
Bateman, Esq. — 22. Goweni, or Reticulato-bulbulosum ;
raised at Highclere from pulverulentum ; very beautiful;
named from R. J. Gowen, Esq. — 23. Munroi, or Psittacino-
equestre ; raised at Colvill’s ; unknown to me ; named after

144

AMARYLLIDACEjE.

the gardener of the Horticultural Society. — 24. Baconi, or
Psittacino-regium, ib. named from my deceased friend Ant.
Bacon, Esq., a zealous cultivator of plants of this order. — 25.
Colvillii , or Reticulato-regium ; raised at Colvill’s. — 26.
Cartoni, or Aulico-Sweetii ; named from Carton the gardener
at Highclere, when it was raised there. — 27. Lindseyi, or
Aulico-reticulatum ; named from the present gardener at
Highclere where it was raised. — 28. Lamberti, or Cartoni-
Grahami ; raised at Spofforth ; named from A. B. Lambert,
Esq. — 29. Bonnii, or Hookeri-Haylocki, a complicated cross,
raised at Spofforth, in most of which the stripe of striatifolium
has descended from the first cross; named from the Botanical
Professor of King’s College. — 30. Spufforthice, or Aulico-
Carnarvoni ; named from Spofforth; not yet blown. — 31.
Lindleyi, or Griffini-Carnarvoni ; raised at Spofforth ; named
from the Botanical Professor of the London University.

Many years ago, when, in a letter published in the Hort.
Soc. trails. I first distinguished this genus from the plants
with which it had been confounded, I retained for it the name
Amaryllis, and proposed that of Coburghia for Belladonna and
Blanda. I was not then aware that Linnaeus had given the
name Amaryllis to Belladonna, with a playful reason as-
signed ; but as soon as I learnt it, I felt, besides the general
law of priority, that the jeu d' esprit of a distinguished man
ought not to be superseded, and that no continental botanist
would submit to the change. I therefore restored the name
Amaryllis to Belladonna, and gave that of Hippeastrum or
Equestrian star to this genus, following up the idea of Lin-
naeus when he named one of the original species equestre.
Mr. Sweet has improperly given the name Amaryllis to these
bulbs, and made Belladonna a generic name, to which he
subjoined a new specific one. This was doubly wrong, for
with his view' he ought to have adopted the proposed name
Coburghia, which has been since applied to another genus.
The first institution of the genus Amaryllis was by Linnaeus
in Hort. Cliffort. p. 135, published in 1737. The name was
given expressly to supersede Tournefort’s Lilio-narcissus,
which he rejected as a compound word. It so happens that
the few species enumerated there by Linnaeus are of different
genera, as Sprekelia, Zephyranthes, Nerine and Oporanthus ;
and it was meant to comprise every thing called Lilio-narcis-
sus by Tournefort : but he says that he gives the title in allu-
sion to the name Belladonna, by which several species wrere

AMARYLLIDACE^E.

145

known, because Amaryllis was the bella donna of Virgil, and
her name was become proverbial for loveliness ; and he adds
a further conceit, that some of the bulbs were said to be
bitter, amarellas. Amaryllis belladonna is not one of the few
species defined in that article, because, though he knew of its
existence, he had it not to enumerate from the Clifford gar-
den. Mr. Sweet was perhaps misled by knowing that equcstre,
which is one of the plants described, was called belladonna
by Merian ; but Merian only called it another belladonna,
with reference to the plant of the Italian gardens, thinking
erroneously that it was of the same genus. Barrelius had
previously, in the year 1714, described the pink and white
belladonna, as cultivated by that name in the gardens of
Italy, and to the plant of Barrelius both Merian and Linnaeus
alluded. It was the exquisite blending of pink and white in
that flower, as in the female complexion, that suggested the
common name in Italy, and to those lovely tints Linnaeus
referred, when he assigned to it the name of a beautiful
woman. To suppose he could have alluded to a bright
orange flower would be perfectly absurd. It is, therefore,
quite indisputable that Belladonna is the type of the Linnaean
genus Amaryllis, and it would be an idle insult to the memory
of Linnaeus to remove it without any cause. It is equally
clear, that this genus Hippeastrum is quite distinct from
Amaryllis, and, as belonging to a different section of the
order, it has no proximate affinity to it. It is divided from
Amaryllis Belladonna by the following features. 1. Hollow
scape. 2. Black shelly seeds. 3. The seeds not bursting
the capsule prematurely. 4. The oblique mouth of the tube
abbreviated on the under side by the deeper incision of the
perianth. 5. The fourfold instead of alternate diversity of
the segments. 6. The fourfold instead of alternate insertion,
(7.) the fourfold instead of alternate length, of the filaments.
8. The nectareous beard or screen in several species. 9. The
germen sloped from the peduncle. 10. The tube sloped from
the germen. 11. The constriction of the germen in the
middle. 12. The capsule widest instead of narrowest at its
base. 13. The flower accompanying or following instead of
preceding the leaves. 14. The growth of the leaves being
vernal instead of autumnal. 15. The plants Occidental in-
stead of African.

An idea was adopted by Mr. Ker, that Amaryllidem some-
times bore one sort of seed and sometimes another. In 1821

L

14 G

AM AltYLLIDACE.E.

I stated that after many years attention to their cultivation,
I had never seen the slightest variation, and that it would not
seem more improbable to me that a fowl should sometimes
lay a duck’s egg, or gooseberry bushes produce peaches,
which I fear was too lightly expressed, and may have given
unintentional offence. After fifteen years additional expe-
rience and constant attention to this natural order, I can re-
assert that I have seen no such variability, and that the
notion that Crinum Capense sometimes bore flat black shelly
seeds was certainly founded on erroneous information. The
seed of Hippeastrum has an outer black foliaceous separable
integument, which after the decay of the seed will lie under
ground for years without perishing. It has an inner brown
coat adhering more closely to the almond-like albumen in
which the embryo lies with the point directed to the orifice.
Seeds of Crinum and Amaryllis, which Mr. Salisbury
used to call albuminous, very improperly, since they seem
to contain no real albumen, have the outer coat soft and in-
separable from a fleshy mass which is the middle or fleshy
coat (in Hippeastrum quite dried up and obsolete), and the
inner coat is not separable from it ; within lies the embryo,
usually (I believe always) without any albuminous bed. I
have opened hundreds of Crinum and Hymenocallis seeds
without discovering any, but I have not had the same fre-
quent opportunities of cutting through those of Amaryllis.
There is no instance in the order of such fleshy seeds being
produced by a plant with a hollow scape. I was deceived,
when I published the appendix, by a little diversity in the
seed of Hippeastrum reticulatum and the variety Striatifolium,
to believe that they constituted a separate genus. The con-
trary was soon proved by their producing fertile crosses with
the other species, and 1 have them now blended with Auli-
cum, regio-vittatum, pulverulentum, See. and breeding again
freely with any species or variety. On re-examination of the
seed it appeared that there was no substantial difference, but
that the seeds of reticulatum are few, and therefore rounder
and not flattened by compression ; the inside of its capsule is
remarkable by its orange colour, but this disappears in the
first cross, as well as the roundness of the seeds, while the
white stripe of striatifolium endures through many successive
alliances. My attention having been kindly called to the
difference between the seed of Iris foetida and other species,
with a view to the soundness of the principle on which I build,

AMARYLLI DACEiE.

147

tliat plants of tlie same genus cannot have seeds of different,
structure, I have compared them attentively with those of
spuria and ochroleuca, and I find no real difference of struc-
ture, though much of appearance. Those of Iris fcetida are
plump and of a bright orange ; those of the two others arc
rather shrivelled and dark brown. The outer coat of the two
latter is separable, thin, and pellucid ; the flesh coat pulpy,
and gradually drying up, in consequence of which the outer
coat becomes shrivelled ; the inner or third coat separable
from the albumen in the interior. The outer coat of the seed
of foetida is equally separable but orange (being exactly the
same difference as between the pericarp of Hsemanthus hya-
locarpus and coccineus), and the second or fleshy coat is
more full, consistent, and durable ; but I perceive no differ-
ence of structure. It is a case somewhat analogous to that
of H. reticulatum. Another section of the genus Iris has the
outer and inner coats thicker and a cavity between them, the
pulpy coat being quite obsolete. It is not requisite that I
should give any opinion here as to the generic identity of
these plants, nor have I sufficiently investigated it. It is cer-
tainly possible, where there is a pulpy coat that dries up
sooner in one species than in another, that it may be entirely
absorbed at an earlier period in a third, thereby thickening
the other coats and leaving a cavity between them, without
any original difference of structure : but wherever we find
any difference of seed corresponding invariably with a differ-
ence in the structure of the flower, I think we may with to-
lerable confidence pronounce the plants to be of distinct
genera, and incapable of breeding together.

There is some difference as to the cultivation of the various
species of Hippeastrum, in consequence of the several lati-
tudes, altitudes, and situations in which they are found.
Capricious watering is their bane ; they should be watered
pretty freely while they are making leaves, more sparingly
after they are grown, and not at all when at rest. Aulicum
I have found very difficult to manage ; I have had but two
or three roots of it, and have not been satisfied with their
treatment. Calyptratum flourished well with me in light,
soil on the hothouse flue, growing all the year round, till I
was told by a gentleman that they had been found to succeed
better in the greenhouse, and having transferred them, ac-
cording to his advice, I lost all my bulbs of that species.
Psittacinum and the beautiful mules between it and Reofio-

O

148

AMARYLLIDACEyE.

vittatum are hardy greenhouse plants, requiring absolute rest
in the winter, and flower freely in the spring ; they grow
weak in the stove, and will not flower without rest. Solan-
driflorum and Stylosum are tender stove plants, but should
rest in winter time. Vittatum is a greenhouse plant requir-
ing rest in winter, and may be brought into the stove in spring
to flower it. In Surrey it lived well, flowered yearly, and
sometimes ripened seed, in the open ground near the south
front of my house, a small heap of ashes being thrown over
it in the winter. The mules between Psittacinum and vitta-
tum would perhaps bear as much exposure, if the wet could
be kept from them in winter. Reticulatum and striatifolium
are tender stove plants, and I believe the former is nearly
lost, and its habitation has never been exactly ascertained.
Of latter years the striped-leaved variety has been frequently
sent from Brazil, but the original plant has not been met with.
The mules between striatifolium, and different varieties of
bulbulosum, as well as regium and regio-vittatum, have a
hardier constitution, and many of them come so near to the
reticulated parents, that they will be preferred in cultivation
for ornament. Equestrc is a plant of singular constitution,
and frequently lost in the stoves : though a native of the
hottest regions of the west, it will not live if watered con-
stantly in the stove. It requires absolute rest in winter in a
moderately cool but not damp situation ; it will flower early
in the summer, and after flowering should be placed in the
greenhouse, or in the open air, where it will grow better than
in the stove. Regium requires less care, the stove, and rest
in winter. The whole family of bulbulosum, except croca-
tum, are easily managed. By giving them two periods of
rest, in winter and again at midsummer, they, as well as the
mules regio-vittatum and rutilo-regio-vittatum, may often
be made to flower in the spring and autumn. I have found
great advantage with bulbs that were to stand on a hot flue in
placing under them a shallow tray made of tin or zinc, and
nearly filled with sand. In pursuance of this system of en-
couraging their growth by moist warm sand underneath, a
gentleman to whom I had given several tender bulbs in-
formed me that he had constructed a pit for them with a
chamber into which was introduced a slender steam-pipe,
perforated with small holes ; and the chamber was covered
with hurdles, over which he placed a layer of brush-wood,
and on that a body of sand in which the pots were plunged.

AMARYLLIDACEjE.

14.9

The steam worked its way through into the sand, and kept
up a moist warmth which was very congenial to the tender
bulbs during their season of growth ; and I do not conceive
that any better mode of cultivation could be adopted. A bed
of the various splendid Hippeastra successfully cultivated in
a low narrow house, would exceed most vegetable displays in
beauty. Some of the varieties of H. bulbulosum, if not all,
may be found in South America, growing in black vegetable
earth. My collector found pulverulentum in such soil with
the scape three feet high and the leaves as long : and I dis-
covered equestriforme growing unperceived in a mass of pa-
rasitic plants, Cereus and Pitcairnia, which had been torn off
a stem or the face of a rock. I have however lost so many
bulbs by the use of peat at various times, that I am generally
fearful of using it. There is so much variety in the vegeta-
ble black earth of different places, that it should be tried
cautiously. I have been told that H. calyptratum has been
found growing on the branches of trees, and that it has been
necessary to shoot off the limb by repeated discharges of a
gun in order to get the bulb : and I have seen it grown in a
pot of moss. The principal causes of the sickly state of Hip-
peastra in cultivation are too light a soil, want of water when
the leaves are pushing, and too much water after. I have
observed them grow with unusual vigour in a split or broken
pot in consequence of a better drainage. The finest bulbs I
ever saw were three self-sown seedlings from a cross-bred
plant, which established themselves in the pot where Con-
volvulus Gangeticus was growing. They killed the Convol-
vulus and at last broke the pot, and have not been so vigor-
ous, since, in consequence of its falling to pieces, it became
necessary to shift them. It is evident that good drainage is
essential to their health. With earth that sets firm, that
object may be effected better by a single crock placed care-
fully so as to cover only part of the hole, than by many, of
which the lowest covers the aperture, and the remainder be-
come choked by the earth settling amongst them. I have
had seedlings of crosses with vittatum which sent up two
stems of blossoms from a pot scarcely twice the size of the
bulb. A self-sown seedling established itself in one of my
stoves, and is growing freely on a stump of wood, into a
cavity of which a little peat had been thrown to encourage
the growth of a Pleurothallis ; and I do not doubt the bulbs
being often found on old trees amongst the ferns and other

150

AMARYLLID AC£iE.

parasites ; blit I consider a well-drained rich alluvial soil to
be most fit for bringing them to perfection. They appear to
have gone rather out of favour lately with cultivators, pro-
bably from failures through mismanagement, for certainly
they can be surpassed by few flowers in beauty ; and most
of them may be cultivated in a warm greenhouse, if they are
kept quite dry in the winter ; but it should be always re-
membered that very tender bulbs, which are to be kept dry
in a greenhouse, will rot if above ground from the dampness
of the atmosphere, though they would have been uninjured
if closely covered by dry earth.

It is now pretty well understood that although cuttings
of Camellia Japonica strike root readily in sand, a light and
confluent soil is fatal to the growth of the plant, causing the
young leaf to turn yellow, become spotted, and fall oft",
especially if exposed to the sun ; a well -drained stronger
soil being essential to their health. Most sorts of Hippe-
astrum seem to me liable to suffer from the same cause,
which is apparently too rapid evaporation of the moisture
which they require : the more frequent watering, which be-
comes necessary to the developement of their leaves, occa-
sions the decay of the fibres. Due attention to this point
will make the cultivation of the bulbs of this genus easy to
those who have been unsuccessful in their treatment, always
bearing in mind that, the stronger the soil used, the more
perfect should be the drainage. Strong loam and a cool
situation, with complete rest in winter, suits H. vittatum,
and I believe that Aulicum will succeed best with the same
treatment. I consider that Ilippeastrum generally does not
thrive well in soil which is powdery when dry, and does
not set.

If half the outer shell of the seed of Hippeastrum is care-
fully taken off" with a penknife, the duct from the umbilicus
to the opposite extremity, which is the region of the chalaza,
will be seen passing circuitously at some distance from the
embiyo ; and another duct from the foramen to the radicle of
the embryo, the two apertures being closely approximated.
The main nourishment from the umbilical cord appears to be
conveyed to the summit of the cotyledon.

I have no reason to believe that this occidental genus has
been found in a wild state north of the tropic of Cancer. Its
natural limits seem to coincide nearly with the tropics in the
western hemisphere, exceeding them a little in the southern

AMARYLLIDACEiE.

151

latitudes. The discovery by Drummond of the Chilian
Allium Cowani and Bonarian Habranthus Andersonianus in
Texas, renders it less improbable that Hipp. acuminatum
should be a native of Mexican Oaxaca as well as of Brazil.

33. Piiycella. — Perianth with segments regular, convo-
lute ; filaments alternately equal, inserted alike at
the mouth of the tube, decurrent, straight, recurved
at the point; anthers short; incumbent, versatile;
style sloping downwards. (Observ. Bulb ovate ,
black; leaves linear, lorate; umbel unequally pedun-
culated; flower declined ; stigma simple, dilated, fim-
briated at the end; fiaucial membrane not defective
on the lower side, as in Hippeastrum ; properly seti-
form or subulate, but its diversities are not sufficiently
ascertained.)

1. Ignea. — Leaves tipped with red; peduncles long;
perianth yellow below, scarlet upwards ; two awl-
shaped processes at the foot of each sepaline fila-
ment ; style longer than the filaments, filaments
than the perianth ; perianth about 2i inches long.

Var. 1. Corusca. — Lindl. Trans. Hort. Soc. 7. 1. 74.
Am. ignea, Bot. Reg. 10. 809. Style much longer
than the filaments, which are white. According to
Professor Lindley, the processes are one- third the
length of the filaments ; but I cannot help suspect-
ing an error in his memorandum, for I have met
with no process of such dimensions in the genus.
Found in sandy soil by Poeppig near Concon in
N. Chili.

Var. 2. Glauca. — Bot. Mag. 53. 2687. Leaves glau-
cous ; style scarcely longer than the filaments ;
filaments red ; processes very minute.

2. Cyrtanthoeides. — Bot. Mag. 50. 2399. Leaves green;
peduncles very short ; perianth greenish yellow at
the base, the rest red ; petaline filaments quarter
of an inch longer than the others, quarter shorter
than the style ; all red. The inside of the flower
was not examined by Dr. Sims, whose specimen
was evidently diseased. Professor Poeppig recog-
nised this plant, as distinct from ignea, on the sandy
maritime hills near Concon in N. Chili, but he
gives no particulars ; and he did not favour me

152

AMARYLLIDACE.E.

with any answer to the inquiries I took the liberty
of making about the Amaryllidaceous plants he
collected in Chili, concerning which the notices
he printed are not sufficiently explicit : but it is
rather my duty to apologise for having intruded on
the occupations of Professor Poeppig, than to com-
plain of his silence.

3. Magnifica. — Perianthii tubo semunciali, ilimbo 2f ;

stylo filam. et perianth, semunc. longiore (appendi-
cibus, ni fallor, inconspicuis).

Var. 1. PI. 24. f. 16. Specim. ex Limache et Colmo,
Bridges. Herb. Hooker, foliis 2-2i ped. i unc. vcl
ultra lat. subacutis, scapo pedali ; spatha 2-unc.
pedunc. unc. germ, oblongo, tubo rubro, limbo
rubro colore aureo inferne admixto.

Var. 2. PI. 24. f. 12. 13. et 14. Specim. ex Valparaiso,
Gumming, 456. Herb. Hooker, absque foliis. Sca-
po 9-unc. pedunc. 2^, tubo rubro, limbo duabus
parti bus inferioribus aureis, superiore rubra. Much
the most splendid species of Pliycella yet known.
The leaves of Var. 1. grow to the length of two feet
and a half ; the tube half an inch long, red ; and
the limb red upwards, red and golden mixed be-
low. Var. 2. has longer peduncles on a shorter
scape, tube red, but the lower part of the limb pure
golden. In both the flower is three inches and a
quarter long. I cannot ascertain that they have
any faucial appendages, but I have only seen dry
specimens. This genus, which shows much inter-
nal diversity with a great general similarity of
aspect, seems to be mainly distinguishable into
species by the modifications of the faucial mem-
brane, and without an opportunity of inspecting a
fresh flower, it is difficult to come at the knowledge
of the facts. Magnifica has the style half an inch
longer than the limb, the filaments not exceed-
ing it.

4. Graciliflora. — PI. 25. fig. 4. Specim. ex Mendoza.

Gillies. Herb. Lambert. Specim. ex valle inter
Viliam Vicentio et apieem montis Paramillo. Gil-
lies. Herb. Hooker. Folia 1 9-unc. § lat. utrinque
attenuata, obtusa; scapo 3-9 unc. pedunc. 6-7, in-

AM ARY LLIDACEAS.

153

eequalibus, 1-l^unc. perianth l£ unc. filam. peta-
linis perianth, longitudine ; stylo exuperante ; pro-
cessibus sex parvis ad basin filamentorum. Found
by Dr. Gillies in November, deeply rooted in hard
stony ground, between Villo Vicentio and the top
of the Paramillo ; said to grow also in Chili under
the name Mancayo, but the identity of the plants
is not ascertained. Professor Lindlcy’s note on
Dr. Gillies dry specimen from Mendoza is, “ pro-
cessibus tribus squamiformibus fimbriatis but I
have the benefit of Dr. Gillies’ description made
from the fresh plant of his specimen in Sir W.
Hooker’s herbarium, stating that it has six faucial
appendages, and it is certainly the same plant.
Cells about 25-seeded.

5. Attenuata. — Perianthio rubro attenuate infundibuli-
formi (appendicibus faucialibus inconspicuis ? )
stylo perianthium et stamina superante.

Var. 1. Obtusifolia. — PI. 25. f. 2. Specim. 33. Chili,
Bridges. Herb. Lambert. Leaves obtuse, a foot
long, above a quarter of an inch wide ; perianth
If long.

Var. 2. Latifolia. — PI. 25. f. 2. Specim. ex Valparaiso,
Herb. Lindl. Leaves half an inch wide, perianth
If long. Grows in deep sandy loam.

Var. 3. Macraeana. — Spec. Herb. Lindl. Prope Co-
quimbo. M‘Rae. Leaves narrow acute ; perianth
1 i inch long, spathe about an inch ; umbel
many-flowered, segments of the limb obtuse, vivid
crimson. (Appendicibus faucialibus inconspicuis?
Lindl. MS.)

This species has red funnel-shaped flowers (with incon-
spicuous faucial appendages, according to a memorandum by
Dr. Lindley concerning Var. 3, by which 1 understand
scarcely discernible in the specimen) ; style longer than the
limb, filaments shorter. The leaves of Var. 3. are acute and
narrow ; of Var. 1 . wider and obtuse ; of Var. 2. half an inch
wide. The two last mentioned have flowers larger than
Macraeana, which perhaps may differ in the internal struc-
ture. I have no means of knowing whether they have any
faucial appendages, but I cannot by the dry specimens dis-
tinguish them as species from Macraeana, which, according

154

AMARYLLIDACEiE.

to Dr. Lindley’s memorandum, has the faucial appendages
inconspicuous. I apprehend that means only not discernible
in the dry specimen, where, however, they would be likely
to elude the search.

G. Herbertiana, — Bot. Reg. 16. 1341. Leaves narrow,
obtuse ; limb slender, very little attenuated, pale
red, equal to the filaments, shorter than the style ;
faucial membrane, according to Professor Lindley,
annular, fimbriated. From Cumbre, a pass be-
tween Valparaiso and St. Jago, four or five inches
below the surface. The soil adhering to the im-
ported bulbs is said to have been black, but I
should advise caution in the use of peat for any
Phycella in a pot.

7. Brevituba. — Perianthum If unciale, basi annulari,

vix tubatum, infundibuliforme, rubrum, intus
luteo-striatum ; stylus perianthium et filamenta
superans ; filamenta pallida apice rubra, petalina
limbum subsequantia ; processus fauciales sex, k
unciales, tenuiter subulati ; folia semunciam vel
ultra lata. Perianth 1 inch and f, red, yellow-
streaked within, tube scarcely more than annular,
style exceeding the limb and filaments ; filaments
pale tipped with red, the petaline nearly equalling
the limb. Introduced into this country by Mr.
Knight, of the King’s Road, Chelsea, from Chili.
It is remarkable from the tube being nearly obsolete.

8. Bicolor. — Amar. bicolor. Flor. Peruv. 2. 57. Leaves

ensiform, umbel 6-12-flowered ; peduncles brac-
teate ; perianth sub-campanulate, 1^ inch long ;
tube short ; segments straight, a little patent at
the end, lanceolate, red with green tips ; filaments
straight, shorter than the perianth ; two-horned
scales at the base of the segments. According to
Molina (Saggio, p. 129) the scape is flattened and
two-edged, two feet high, flowers four, red outside,
yellow often marked with red inside ; segments
rounded ; leaves many, channelled, obtuse. Bico-
lor is unquestionably a Phycella from Peru. Mo-
lina’s plant is from Chili ; and as he does not
mention the green colour, it is very questionable
whether his plant was the Bicolor of Ruiz, espe-

AMARYLLIDACEiE.

155

daily as he limits the number of flowers to four
with such a vigorous scape.

9. Chloracra. — Haemanthus dubius, Humb. et B.
Kunth. 1. 281. Leaves oblong, lanceolate, sub-
acute, smooth, above an inch wide, near five inches
long ; scape round, spatlie many-leaved (I appre-
hend including bractes); umbel 7 or more flower-
ed ; peduncles 6-9 lines ; perianth red with green
tips, pedunculous, 1£ inch long, tube-formed ; tube
short ; segments conniving ; style straight, longer
than the filaments ; stigma obtuse. Grows by the
river Guallabamba, near Quito, in temperate
mountainous situations. This plant, allied to bi-
color, is remarkable by an unusual breadth of leaf ;
which is a departure from the habits of the genus,
perhaps the consequence of its inhabiting a richer
and more humid soil.

It will be observed that the genus Phycella has, as far as
we know, red flowers ; that in the sandy and lower parts of
Chili it seems disposed to have yellow at the lower part of
the flower, in the higher regions of Peru green tips. I
entertain no doubt of bicolor and chloracra belonging to this
genus ; the latter has probably large bractes, which gave
the notion of a many-valved spathe like Haemanthus, but I
scarcely think it will be found to have many outer valves to
the spatlie. There are frequently in scapaceous plants inner
valves or larger outer bractes, of which it is difficult to say
in which light they should be considered. I should how-
ever by no means think a redundancy of valves to the spathe
a sufficient cause for removing it from the genus.

I find a decided difference of habit between Phycella,
Habranthus, and Zephyranthes. Some years ago I planted
three species of Phycella out of doors in front of a green-
house, throwing a small heap of sawdust over them in
winter. In that situation one of them flowered early in the
summer, and they go to rest in the hot dry season. They
are tempted by mild weather to push their leaf in the winter,
in which case they suffer injury from severe frosts that may
ensue, though they will endure a good deal ; and their habit
is to flower after the leaf has acquired its growth before they
go to rest. The Phycellas have been found difficult to cul-
tivate, because they have been often set in peat, though they

156

AM A RYLLIDACEjE.

grow naturally in a sandy or strong soil on a dry rocky
substratum, and proper rest has not been allowed them. They
should be planted in light soil well drained, and be left dry
from the moment their leaves show a disposition to wither,
till the bulbs on examination show a disposition to push out
fresh fibres at their base. The old fibres in this genus seem
always to perish before the plant vegetates again ; it cannot
therefore be injurious, and may be advantageous, to take the
bulbs out of the ground when the leaves perish, and set them
again when they are disposed to move. They will be best
preserved while at rest in dry sand. Concerning the differ-
ence between Pliycella and Habranthus, see the two first
species of Habranthus, and the observations at the end of
that genus. I consider that Phycellas should begin to grow
in February and go to rest in August. If the leaf endures
later than August, they should have six months rest before
they are watered again. A sunny aspect at the foot of a
south wall appears to suit them.

Eustephia. Cavanilles.

This genus must be erased, as not existing in truth ; the
generic character being obscurely worded, and supported
by a figure of the plant which I do not hesitate to pronounce
fundamentally erroneous, both as to the extraordinary fila-
ments there represented, and the pits in the tube under their
point of insertion. I have no doubt that the plant intended
was a Phycella, with appendages at the mouth of the tube,
which have been incorrectly transferred to the mouth of the
limb and fastened on the extremity of the filaments ; and the
tube of the flower, injudiciously called in the text a pit, has
been multiplied into six pits under the several filaments,
where it is almost impossible that such cavities should be
placed. If such had existed, they would have been between
the filaments. I apprehend that the author must have for-
gotten the precise intention of his own notes on the plant,
and framed the figure from erroneous recollection drawn
from his notes, and not from the plant itself, and I am con-
fident that the representation given will never be verified.

34. Habranthus. — Flower declined; perianth short-tubed,
subcampanulate, not convolute, more or less patent ;
alternate segments nearly equal, filaments (properly
of four lengths) inserted alike at the mouth of the
tube, declined, recurved, semi-fasciculate ; faucial

AM ARYLLID ACEA2.

157

membrane (when manifested) annular ; anthers af-
fixed at the middle, incumbent, versatile ; style de-
clined, recurved. ( Leaves linear lorate , not exceed-
ing half an inch wide; umbel precocious, unequalhj
pedunculated; stigma ii-lobed ; capsule turbinate ; fau-
cial membrane , as far as we know , never wanting. ) See
the observations at the end of the article Habrantlius.

§. 1. Pluriflorre. — ( Filamento , quoad novi , supero inter
sepalina longiore, infimo inter petalina breviore.
Upper sepaline filament prolonged, lower petaline
abbreviated.')

1. Phycelloides. — Bot. Mag. 17. 1417. Large black
round bulb ; leaves glaucous, obtuse, |- wide ;
umbel 6-flowered, peduncles long, unequal; peri-
anth yellow below, scarlet above, tube limb 2|,
style longer than the perianth and filaments ;
tipped with red ; filaments inserted in an annular
membrane.

In the character of the filaments in the Bot. Reg., the
words extends and intends are transposed by a lapse. I
have placed this species first ; not as the type of the genus,
but because it approximates to Phycella so much as to have
raised a question, whether the genus Phycella could stand as
distinct from Habrantlius. The genus Phycella did not
originate with me, and I entertain doubts whether it is in
fact a distinct genus, or only a section of Habrantlius with
convolute perianth, filaments of alternate length and droop-
ing style ; neither do I feel satisfied of its identity with
Habrantlius, so as to presume to reduce it, and in fact I
believe it to be distinct. The approximation of the extreme
points in two nearly allied genera, is not without precedent.
H. phycelloides agrees with all the Habranthi with more
than one flower that I have had an opportunity of inspect-
ing, in the elongation of the upper sepaline, and abbrevia-
tion of the lower petaline filament, in the recurved style, and
the perianth not convolute, and in truth its resemblance is
rather in colour to Phycella ignea than in structure, which
is conformable to that of Habrantlius angustus. There is
some difficulty in defining the genera intermediate between
Hippeastrum and Haylockia; for although there can scarcely
be two less similar than Phycella and Cooperia, the edges of
the intermediate genera come so nearly in contact, as to

158

AMARYLLIDACE/E.

make it very difficult to be certain of the limiting features.
I have never however seen a dry specimen of Phycella or of
Habranthus, which I could hesitate to distinguish from each
other at the first sight, or a bulb of Phycella which I could
fail to recognise as such ; but there is an intuitive perception
of difference, especially in the look of bulbs, which it is not
easy to define. The bulbs of Phycella, as far as I have seen,
are never as round as those of Habranthus, nor do they ever
elongate themselves beyond their regular structure. The
elongation of the bulb of Habranthus is not the natural form,
but its mode of drawing itself to a greater depth under
ground. I wish I could excite cultivators to try if they can
produce a cross-breed between Phycella and Habranthus;
but I do not think they will succeed.

2. Speciosus. — PI. 23. f. 2. Specim. Herb. Lindl. ex
Conception, Chili. M‘Rae. Foliis sub-sesquipeda-
libus (16 unc.), angustis, longe attenuatis, 3-16-1-
latis; scapo subpedali, spatha 1^ unc. umbella tri-
flora; pedunculis 1| unc. germ. 1 unc. tenui ; tubo
brevi, gracili ; perianth 2J unc. campanulato, satu-
rate rubro; stylo i, filamentis f, unc. perianthio
brevioribus ; stigmate obtuso. In Sir W. Hooker's
herbarium there is a flower of this species ticketed
Am. Chilensis. Bridges. It is not the plant re-
cognised as Chilensis by Poeppig, which has the
stigma deeply trifid, while this has it undivided.
The Chilensis of Flor. Peruv. is clearly an orange
bearded Hippeastrurn allied to Regium and Mini-
atum, a variety which can never, from want of par-
ticulars, be identified. That of L’Heritier, exclud-
ing his quotation of a Pyrolirion from Feuillet, is
an Habranthus, concerning which he has said so
little, that it cannot be ascertained from his de-
scription what plant he meant. He calls it purple,
yet quotes an orange flower as identical. I shall
therefore consider Dr. Poeppig’s plant, which does
not contradict any thing stated by L’Heritier, to
be the plant that must bear the name. H. specio-
sus seems by the specimens to be of a deep crim-
son. It has a stigma obtuse like that of Hipp.
equestre, a variation which, as it occurs in that
genus, cannot surprise us in this. It brings Ha-

AMARYLLID ACE.E.

159

branthus a little nearer to Phycella, by removing
the difference of a trifid stigma ; but in both the
specimens of this plant the style is recurved, and
the division of the stigma in this order is very
variable.

3. Pratensis. — Poeppig. Fragm. Syn. p. 5. Leaves

linear, glaucous, with the back rounded ; scape a
foot or more, 2-3-flowered ; perianth scarlet, yellow
at the base ; tube scarcely any ; limb two inches,
campanulate, with unequal segments, the upper
widest ; faucial scales six, serrate ; filaments de-
clined. Flowers in November in the meadows of
Antuco in S. Chili. This plant seems to me allied
to speciosus, except in leaf, which appears to have
a different character. Dr. Poeppig calls it “ ge-
nus Placea, Miers,” alluding, I suppose, to the
faucial scale, but they would be insufficient to dis-
tinguish it except as a species. Placea (Miers) is
a mere name, without any definition or description
of the plant for which it was intended.

4. Kermesinus. — Am. kermesina. Bot. Reg. 19. 1638.

Leaves 3-16ths wide, obtuse, bright green, slightly
glaucous (underneath ?) peduncles unequal, about
1£ inch, perianth about long, deep crimson,
ribbed at the base with yellow, suberect; filaments,
“ two much shorter than the rest” imply that the
upper sepaline is prolonged, the longest rather
more than half the length of the perianth ; style
one-third longer than the filaments ; stigma trifid,
recurved. Brazil ; I doubt not, S. Brazil.

5. Nemoralis. — Specimina ex Rio Grande in sylvis

prope flumen (Yazeguay?) Tweedie. Herb. Hooker,
absque foliis. Scapo 4-unc. trifloro, spatha acu-
minata 2-unc. pedunculis inseq. subbiunc. perian-
thio kermesino biunciali, anguste campanulato, vix
stylo semunciam longiore ; stylo fil. vix superante,
stigm. trifido. This flower, called bright scarlet by
Tweedie, but seeming bright crimson, was found in
the woods by some river which cannot be deciphered
(the Yazeguay, I believe), near Rio Grande. It
comes very near to kermesinus, and is probably a
variety of it, but the specimens agree in a 4-inch

ICO

AMARYLLIDACEiE.

scape, longer spathe, and filaments greatly longer

than those of kermesinus. The leaves are wanting.

©

0. Intermedius. — Am. intermedia. Bot. Reg. 14. 1148.
Ex Brazilia. Leaves bright green, obtuse, ~ wide;
scape green (six inches?) peduncles f to If, green;
perianth about 2 inches, patent, red, greenish
yellow below ; style and filaments as in Nemoralis.
A much more expanded and differently coloured
flower.

7. Spathaceus. — Spatha univalvi, latere fisso, floribus

magnis purpureis. Spathe one-valved, slit on one
side ; flowers large purple ; leaves green.

Var. 1. Leaves broader, flowers larger.

Var. 2. Angustus. — Bot. Mag. 53. 2639. Leaves green,
narrow, spathe one-valved, slit on one side to the
base, tube green, limb, style, and filaments purplish
red ; segments narrow. In all points narrower
than spathaceus. V. 1. Membrane annular.

8. Blfidus. — Bot. Mag. 52. 2597. Spathe 2-valved,

2^ inches long, green ; scape green ; flowers four,
2 inches long, purplish red, darker and lined with
green below. Membrane annular.

Var. 2. Litoralis. — Specim. ex Monte Video. Tweedie,
Herb. Hooker. Spatha bifida 5-flora, perian. satur.
purpureo ; absque foliis. Spathe bifid 5-flowered,
perianth deep purple; without leaves. Plentiful
within tide-mark at Monte Video. Tweedie writes
that the leaves are like those of Nerine venusta,
commonly called Guernsey lily, but they are pro-
bably not so wide. This may, perhaps, be the
plant formerly named by me lorifolius, from Mal-
donado, which has never flowered here, and, as I
cannot ascertain whether it is distinct from all
Tweedie’s specimens, it becomes necessary to drop
the name, at least for the present. The bulbs, so
named, increase fast by offsets, and elongate them-
selves very much, appearing to strive to get deep
under ground, and, I believe, they will not flower
while the bulb is elongated. I am now trying
various modes of culture to bring it to a flowering-
state. Litoralis growing within tide-mark may
require salt.

FiaJte. /9.

Y7 Herbert . del .

AMARYLLIDACE^E.

161

9. Pulcher. — PI. 26. f. 1. Specim. Tweedic. Herb.

Hooker, absque foliis. Scapo 3| unc. spatlia 24
unc. bifida apice non diviso, pedunc. quinque, f-
3| unc. germine crasso, perian. 1-| (albo ? eru-
bescente ?) stylum \ unc. superante ; stylo fila-
mentis longiore, stigmate trifido. Flowers five,
apparently white or blush. Upper sepaline fila-
ment decidedly prolonged, and lower petaline
abbreviated. The inequality of the peduncles is
very conspicuous. From Buenos Ayres.

10. Pedunculosus. — PI. 26. f. 3. Specim. Tweedie.
Herb. Hooker, absque foliis. Scapo unciali bifloro ;
spatha tubulosa apice bifido ; pedunc. 3-unc. ger-
mine longo gracili, perianthio angusto 2-unc. vel
ultra, laciniis acutis, stylo unc. fere longioribus
(pallide rubris ?) filamentis stylo brevioribus ; stig-
mate trifido reflexo. From Buenos Ayres. A
remarkable plant, the scape being only an inch,
and the two peduncles each three inches long, the
peduncles remarkably slender and the spatlie tubu-
lar. If it should appear on inspection of these six
Bonarian plants that the annular membrane at the
base of the filaments is precisely similar, they
might, perhaps, be united as local varieties of one
Habranthus Platensis ; but without more intimate
acquaintance with them, the union would not be
advisable. I have some Habranthi from the Plata
which have not yet flowered with me, and vary a
little in leaf, but certainly belonging to the same
family. These species all require to be plentifully
watered in a greenhouse or frame in the winter,
and their leaves ought to grow out to the length of
near two feet. As soon as the leaves decay, they
should be left quite dry, and if placed in a hot
situation it will promote their flowering.

11. Hesperius. — Foliis glaucis, perianthio patente re-
flexo, annulo fauciali fimbriato, filamentorum qua-
terna discrepantia obsoletiore.

Var. 1. Advena. — Amar. advena. Bot. Reg. 10. 849.
Bot. Mag. 28. 1125. Leaves channelled, rather
glaucous, | inch wide ; 2-6-flowered ; flowers red
upwards, yellowish green below ; peduncles 2 inches

M

162

AMARYLLIDACEiE.

and under. A delicate plant, apt to rot ; probably
a native of dry situations. It should certainly not
be planted in peat.

Var. 2. Pallidus. — Lodd. Bot. Cab. 1760. Am. advena,
var. citrina. Bot. Reg. 10. 849. Specim. septem-
florum. Cumming, 395. Valparaiso. Herb. Hooker.
Spec. 4-florum. Am. advena, Cumming. Valpa-
raiso. 183. Herb. Lindl. Leaves less channelled,
more glaucous and recumbent than those of Ad-
vena, i wide, obtuse ; flowers 2-7, pale yellow,
sometimes tinged with red ; varying in different
individuals between sulphur, white, and red ; pe-
duncles shorter than in Advena ; umbel more
numerous than Chilensis.

Var. 3. Miniatus. — PI. 26. f. 2. Sweet Brit. fl. g. ser. 2.
213. Specim. ex Valparaiso, Cumming, 355. Herb.
Hooker, with the scape 85 inches, spathe 3 inches
acuminate, peduncles J- to 2^, perianth patent, If
long, segments acute, style trifid, half an inch
shorter, but a little longer than the filaments ;
upper sepaline filament little elongated, but not
equal, as stated by Sweet ; leaves wanting, said by
Sweet to be glaucous. Faucial membrane not re-
corded, but it will probably be found to agree with
A dvena and Pallidus, under which expectation I ven-
ture to include it under a name implying that they
inhabit the extreme West. I do not believe the seed
of Advena would ever produce Pallidus, and it
would be incorrect to call the latter a variety of the
former; but where the distinction, however perma-
nent, is not great, it will be found very convenient
to unite those which agree in most essential points
under a superior name, which produces no change
in the nomenclature, but gives a clear view of the
affinities of the local variations of the genus.

12. Bagnoldianus. — Bot. Reg. 17. 1396. Specim. Cum-
ming ex Coquimbo. Herb. Hooker. Bulb large,
black; leaves obtuse, glaucous, f wide ; scape green,
6-flowered ; peduncles 3 inches and under, tube a,
limb 1 and 1 5- 16ths long, yellow, spotted within with
pale red, especially on the three upper segments ;
upper sepaline filament much prolonged, and lower

AMARYLLIDACEiE.

163

petaline shortened; faucial membrane annular fim-
briated. In the Bot. Reg. the words internis and
externis, relating to the filaments, are transposed
by a lapse.

Var. 2. Gillesianus. — PI. 23. f. 1. Specim. ex Melo-
coton. Dr. Gillies. Herb. Hooker. “ Perianth yel-
lowish with a tinge of red, deeply divided, segments
ovate-lanceolate with a point, inch long, at the
base attenuated and greenish ; filaments yellowish,
fimbriated membrane at their insertion ; anthers
green before inversion ; bulb size of a pigeon’s
egg, black. Found in gravelly soil near Meloco-
ton.” Gillies. The peduncles do not exceed an inch,
the spathe two inches ; leaves four, a foot long, only
wide ; flowers much paler than in Bagnoldianus,
sulphureous ; but it is certainly to be called a
variety of that plant. It will be observed that both
these plants have the glaucous leaf and fimbriated
membrane of the Hesperian family, from which
they are separated by the more decided prolonga-
tion of the upper and abbreviation of the lower
filament, the appearance of the bulb, and harder
foliage. They are natives also of the same coast as
Hesperius. I lost Bagnoldianus by planting it in
a peaty compost, not being then aware that it grew
in dry gravel, and I understand that it has rotted
in most, collections. A most singular lusus occurs
in a specimen of that plant in Sir W. Hooker’s
herbarium, of which one of the peduncles is com-
pound, bearing two flowers and a bracte, like one
of the valves of the spathe, at the fork.

§. 2. Subbiflorae. — One or two flowered.

13. Roseus. — Sweet. Brit. fl. g. ser. 2. 107. Leaves
glaucous ; flowers two, rose-coloured, green at the
base, expanding wide ; filaments conspicuously of
four lengths. Faucial membrane not recorded.
From the island of Chiloe.

14. Chilensis. — Am. Chilensis. L'Heritier S. A. p. 11.
(excluso Feuillet cit.) non Ruiz Flor. Per. — Poeppig
Frag. syn. 5. — Am. lutea. Specim. Ruiz. Herb.
Lamb. Zephyranthes lutea. Herb. App. — Am.
ochroleuca? Ker. Bauer tab. pict. Bibl. Banks,

m 2

1G4

AMARYLLIDACE^.

filamentis (vix recte?) conniventibus. Leaves
linear ; spatlie 1 or 2-leaved, according as it has
one or two flowers, pedunculated. h’Herit. Leaves
vernal, scarcely to be gathered at the same time as
the scape ; flowers very pale sulphur or vermilion,
the colour being very variable ; stigma deeply
trifid. Poeppig. L’Heritier’s description is quite
insufficient for the identification of his plant, and
his quotations are erroneous. It is strange that he
should have called his flower purple, and yet quoted
Feuillet’s orange Pyrolirion as the same. Ruiz’s
Chilensis is a bearded Hippeastrum, with leaves
attenuated at both ends, a variety of Miniatum or
Regium. Dr. Poeppig has brought home speci-
mens, and if his plant be not Habranthus pallidus,
it must keep the name Chilensis. I believe the
Am. lutea of Ruiz and Pavon's herbarium to be
L’Heritier’s plant. I have a small bulb, sent from
Chili under the name Chilensis, with two leaves
scarcely a line wide, acute, shining green, which
is probably this plant. According to Poeppig the
flower of Chilensis is autumnal, rising before the
leaves, which in my plant are hiemal ; he calls
them vernal, that is they endure till the hot season.
Chilensis was found by him in sandy fields near
La Vega de Conception in S. Chili. Molina, p. 129,
calls it coccinea, flowers bright red, seldom two ;
anthers yellow, genitals red, spathe two-leaved ;
scarce exceeding a foot ; leaves linear, about equal
to it. I suspect that Mr. Ker’s Amaryllis Ochro-
leuca, taken from a drawing by Bauer in the
Banksian Library, must be one of the varieties of
this plant. The drawing represents the flowers
with obtuse deeply divided segments, of a deep
greenish yellow with reddish veins, erect, very
little expanded, conniving filaments, and anthers
short, incumbent, and attached by the middle.
Mr. Ker’s engraving ( Journ. Sc. and Arts ) is very
accurate. It is decidedly not a Zephyranthes,
from the structure and attachment of the anthers,
nor has any two-flowered Zephyranthes been seen.
I think the drawing must have been made before
the full expansion of the flower, and before either

TlcSe.ZO.

AMARYLLIDACEiE.

165

the perianth or stamens had assumed their proper
attitude. No memorandum exists concerning the
plant, which flowered at Kew at a period when
Chilian bulbs were certainly not likely to find their
way there. If it be one of Masson’s bulbs from
the west coast of Southern Africa, it must belong
to a genus entirely unknown to us; for if the draw-
ing exhibits a correct representation of the inflo-
rescence in its perfect state, it agrees with no genus
yet described. It wants the tube that should con-
nect it through Vallota with the Cyrtanthiform
race, and has no affinity to any other African
bulb.

15. Gracilifolius. — Bot. Mag. 51. 2464. From Mal-
donado. Bulb small, leaves very slender, a foot
and half long, nearly cylindrical, with a deep
channel on the upper surface ; scapes successive,
seven or eight inches high ; spathe green, tubular,
slit at the point ; flowers 1-2, 1 inch and long,
tube green with a thick green annular membrane,
limb pale purple, closing at night, and expanding
widest in the sun, stigma trifid. Flowers freely in
September after three or four months’ rest. Gra-
cilifolius and Andersonianus expand in the sun,
but with the flower declined and the lower petal
protruded, not patent like Zephyrantlies.

Var. 2. Boothianus. — Tab. pict.'et descr. W. B. Booth.
Spatha pedunculo et genuine erubescentibus, ger-
mine et pedunculo magis elongatis, filamentorum
quaterna discrepantia obsoletiore, stigmate majore.
In specimine scapus 1-florus. Ex Maldonado.

This plant, which has flowered at SirC. Lemon’s, is only
distinguishable from Gracilifolius, which grows in the same
vicinity, by a rather longer gcrmen and peduncle, a red
tinge on the germen, peduncle, and spathe, and the diversity
of the filaments being less marked, but they are not exactly
of two lengths in the specimen. The flower seems also a
little more declined, and rosier. The specimen was 1 -flow-
ered ; whether it will be found to be always so I know not ;
but I think it not improbable, as the plant is evidently, from
the near equality of its filaments, the intermediate point
of transition from the two-flowered to the one-flowered

166

AMARYLLIDACEjE.

section, which has exactly the opposite discrepancy of fila-
ments.

16. Sylvaticus. — Martius ap. Schultes. Leaves shorter
than the spathe, 1 line and ^ wide; scape green,
4\ inches; spathe 15 lines, one-third divided;
peduncle 14-15 lines; germen ovate, 1£ line;
perianth inclined, 2| inches ; tube half an inch,
green, with a thin faucial membrane; limb funnel-
shaped, green below, above purplish scarlet ; upper
sepaline filament prolonged, lower petaline abbre-
viated. In the woods called Catingas, of the pro-
vince of Bahia. There are lofty Sierras in Bahia,
where I suppose this plant is found ; for the genus
Habranthus, so far as we know, only exists in tem-
perate situations.

§.3. Uni florae. — One-flowered.

Filamento sepalino supcriore abbreviato. Upper sepa-
line filament shorter.

17. Versicolor. — Bot. Mag. 51. 2485. Leaves ^ inch
wide, above a foot long, acute ; scape, spathe, and
perianth at first rosy, fading before expansion ;
perianth two inches long, white tinged with pink,
tipped with red and lined below with red, on each
side of a green midrib; stigma large, white, trifid,
patent ; faucial membrane fimbriated. If my dis-
section of the flower in the Bot. Mag. is correct, it
disagrees with Robustus and Andersonianus in the
prolongation of the lower sepaline filament. I have
had no opportunity of re-examining it since the
introduction of those species, my gardener having
destroyed the bulbs of Versicolor by watering them
during the season of rest.

18. Robustus. — Sweet Brit. FI. g. 2. 14. Lodd. B.
Cab. 1761. Am. tubispatha. L’Her. S. Ang. 9.
Am. Berteri. Spreng. S. V. 2. 49. Leaves sub-
glaucous, channelled, scape robust, spathe undi-
vided two-thirds of its length, a little shorter than
the peduncle ; flower 3| inches long, much de-
clined, purplish pink fading to white; tube scarcely
any ; faucial membrane greenish, fimbriated ?
sepals much wider than the petals, but four grada-

AMARYLLIDACEiE.

167

tions very distinct ; upper sepaline and lower peta-
line filaments abridged ; stigma trifid. This plant
increases fast by seed and offsets. The bulbs are
killed by frost, if left through the winter in the
open ground in England without protection. They
will grow freely if kept dry in the winter, and
planted out in the spring ; but they should pre-
serve their leaves through the winter, and be kept
dry and hot in the spring or early summer to pro-
mote their flowering, which under that treatment
is abundant. Mr. Ker was too hasty in identifying
a Jamaica plant with Commerson’s Am. tubispatha
from Buenos Ayres. I think there can be little
doubt that this Habranthus, which equally agrees
with L’Heritier’s imperfect description and is a
native of Buenos Ayres, must be the true plant, and
strictly I ought to restore the name ; but the rule
concerning priority of names was established for
public convenience to avoid confusion, and, as the
name tubispatha has now been generally applied to
a tropical Zephyranthes, it would be inconvenient
to make a change, and it is quite sufficient to point
out the fact.

19. Pumilus. — Lodd. Bot. Cab. 1771. Leaves narrow,
scape short, flower patent, cernuous, rose-coloured ;
faucial membrane and filaments not observed.
From Chili. I have never seen Sweet’s 2-flowered
Hab. roseus, and am not sure of its diversity from
this plant, of which I have not seen the flower.

20. Maculatus. — L’Herit. S. A. 10. Scape speckled
with red lines ; spathe 2-valved linear, flower pe-
dunculated, filaments and style declined ; perianth
campanulate. From a specimen brought from
Chili by Dombey. Colour not known. The de-
scription of this plant is much too vague to admit
of its ever being identified with any certainty.
The name should be attached to the first Chilian
Habranthus with a speckled scape that shall be
discovered.

21. Andersonianus. — Bot. Reg. 16. 1345. Lodd. B.
Cab. 1677. — Sweet Br. fl. g. ser. 2. Leaves narrow,
linear, acute, green or subglaucous ; spathe tubular.

168

AMARYLLIPACEiE.

divided upwards, peduncle about II inch long or
more, perianth golden or copper-coloured, streaked
outwardly and marked at the base within with
reddish brown ; faucial membrane annular. Lower
petaline and upper sepaline filaments abbreviated.
Peduncle elongated in seeding. From Buenos
Ayres. Varies in the depth of colour of the flower,
and the breadth and glaucous hue of the leaf, in
different localities. The variation of leaf seems
permanent in the seedlings.

V. 1. Aurea, golden. V. 2. Cuprea, coppery. V. 3. Ob-
scura, dark, especially in the bud. V. 4. Brevi-
limba, short-flowered, with broader leaves. V. 5.
Parvula, small. PI. 26. f. 4. V. 6. Texana, Texan,
with roundish obtuse segments.

This bright-coloured Habranthus was imported from
Monte Video in 1829, and I noted the first four varieties
amongst the bulbs which flowered in 1830, having been par-
celled as different by the collector. The fifth is known to me
only by a Bonarian specimen from Tweedie, Herb. Hooker.
The sixth was sent by Drummond from Texas, where the
Chilian Allium Cowani has also manifested itself in the
Northern Hemisphere. I doubt whether it is distinguish-
able from the first variety, but I have not seen them in flower
together. Both Andersonianus and Robustus are disposed to
flower earlier than Gracilifolius and Versicolor.

22 ? Andicola. — Am. Poeppig Fragm. Syn. 5. Diar. 3.
833. Leaves linear, glaucous, smooth ; scape 6-7
inches, glaucous, one-flowered ; spatlie approached
to the flower, two-valved, reaching to the middle
of the perianth ; perianth erect, near two inches
long, subbilabiate, segments nearly equal, splendid
violet colour ; tube 3-4 lines long, throat smooth ;
stamens very short, deflex. Dr. Poeppig does not
state whether there is a peduncle shorter than the
spatlie or none, but he refers the plant to Ha-
branthi. The perianth said to have a tendency to
be bilabiate and the filaments to be deflex point to
that genus ; but a sessile or subsessile germen,
erect perianth, and very short stamens, are at vari-
ance with that genus, and I cannot but doubt whe-
ther it will not be found to be a Zephyranthes

. • eszr>//fM/ms /’}/■ //y /■//)/

Place. 2/

AMARYLLIDACE^E.

169

allied to Carinata. The insertion of the filaments
into the anthers should be examined; and it should
be observed whether the spathe is absolutely divided
to the base. This beautiful little plant flowers in
January, in cold gravelly situations, on the Andes
of Antuco in S. Chili, especially on Pico del Pilque,
and, if we could obtain it, would probably thrive
in our climate.

It may be observed that no Habranthus has been yet
found with the leaf more than -|ths of an inch wide ; nor
any Phycella much exceeding half an inch, except Chlo-
racra (the Haemanthus dubius of Humboldt), which may
possibly prove to be a genus by itself; and that no Hippeas-
trum has been discovered with the leaf near so narrow.
This is not a very fit circumstance to be set down in a generic
character, but it is an important confirmatory observation,
and valuable as a guide in the discrimination of unknown
plants of this family, especially in an order of which many
genera as Brunsvigia, Eurycles, Stenomesson, and others, \
may be recognized by the leaf. Concerning some of the
Habranthi my information is defective, but fourteen are
ascertained to have the faucial annular membrane, and the
absence of it has not been observed in any one. In Hippeas-
trum it is never annular, but confined to the upper region ; \
in Zephyranthes, when the membrane is at all manifested, 1
it is nearly obsolete ; in Phycella its exhibition is properly
setiform, and the departures from that form require to be
carefully examined, which I have had no opportunity of
doing. Dr. Lindley noted down the membrane in Phycella
Herbertiana to be annular, but it may, perhaps, not have
been precisely the same structure that I call annular in Ha-
branthus. The flower of Habranthus rises after the dry
season of rest, and is followed by the leaves, which endure
through the winter; Phycella flowers after the complete pro-
duction of the leaves, and rests after flowering ; Zephy-
ranthes rests in the winter and flowers with or after the rise
of the leaves, the flower expanding in the sun ; Argyropsis
has the leaves perennial and the flower autumnal ; Cooperia
ippears to flower from the earliest spring till the autumn,
the flower being at its prime in the first night of its expan-
sion. The Habranthi in general are pretty hardy, but as
their leaf should be in perfection in the winter, it must be
liable to injury from frost if not protected in some manner;

170

AMARYLLJDACEiE.

they require, in order to prepare their blossom, a hot period
of rest, which would be often wanting to them if exposed to
our climate. When cultivated in a border, they should be
covered with a glass frame, to keep them hot and dry in
May, June, and July, and any covering of mats or straw
that will prevent injury from severe frost may be sufficient
in winter; or they may be taken up when the leaves decay,
without breaking the fibres, kept in sand, and reset three
months after. As most of these bulbs are found in dry
gravelly situations, they must require the border to be well
drained, which should be done by a layer six inches deep of
stones covered with an inverted sod, or at least with heath,
furze, or straw. The same system may be pursued with
advantage in deep pots for all plants that are liable to suffer
from wet, as Habranthus Bagnoldianus, and Hesperius,
placing a thin inverted sod, or some other covering, over
the crocks or stones, to prevent the drainage from becom-
ing choked, and with that precaution stronger soil may
be used than would suit otherwise, and less water will be
necessary.

35. Zephyranthes. — Leaves linear; scape one-flowered ;
spathe one-valved; germen sessile or pedunculated,
erect ; tube short, funnel-shaped ; perianth sub-
erect ; faucial membrane inconspicuous, not annu-
lar, manifested (if at all) by six very minute points
above the insertion of the filaments, which is at the
base of the segments just without the tube, the
sepaline inserted a very little lower, and shorter;
anthers suberect, versatile, attached below the
middle, sloped. (Stigma usually trifid, patent ; cap-
sule ovate, deeply 3-furrowed ; seeds less numerous
than in Hippeastrum ; style generally declined, the
opposite filament being averted.) Leaves narrow,
(estival, produced in the spring ; flower simultaneous
or later.

A. Pedunculated.

1. Mesochloa. — Bot. Reg. 16. 1361. Leaves green;
spathe looped ; perianth half-green, above white,,
stained with red outside, segments acute, tube very
short, smooth within, style white, an inch shorter
than the limb, very little longer than the filaments. I
have placed this first, not as the type of the genus,

AMARY LLIDACEJE.

171

but because it approaches nearest to Habranthus,
the flower being produced later in the summer,
and not expanding so freely in the sun as the rest
of the genus ; and if my representation of its
anthers in the Bot. Reg is quite correct, there is a
difference which requires to be re-examined. It
varies with a spathe bifid at the point and flower
yellower, and with a shorter peduncle and flower
not marked with red. From Buenos Ayres. It
flowers and seeds freely. Some seedling bulbs
having been left out in a bed covered with leaves
in winter, though but just under-ground, survived,
and sprouted in the first week of May. I do not
doubt its succeeding in a bed of white sand covered
with leaves in winter.

2. Depauperata. — Poeppig Frag. Syn. 4. Diar. 3.

795. Leaves vernal, very narrow, linear; pedun-
culated, spathe far from the flower, tubular, bifid ;
limb erect, campanulate, regular, very acute, pale
sulphur with purple external streaks ; throat
naked ; filaments straight, nearly equal. In sandy
stony fields of the Andes of South Chili. It seems
very nearly allied to Mesochloa. Dr. Poeppig
has not stated in what respect it is depauperated.

3. Atamasco. — Am. Atamasco. Bot. Mag. 7. 239. Lodd.

B. C. 1899. Catesby Car. 3. 12. Leaves green,
tube half an inch, limb large, white, bright red
outside in the bud, white after expansion. Abun-
dant in pastures of Virginia and Carolina. To be
kept dry in the greenhouse in winter ; if planted
out it should be set in white sand, to prevent the
roots rotting while at rest. Sprengel made a great
mistake in uniting the sessile Verecunda with this
species as a variety.

Var. 2. Minor. — With smaller flowers, more obtuse.
Introduced by Fraser from Carolina; not materially
different.

4. Tubispatha. — Bot. Mag. 38. 1586. Not L’Herit. S.

Angl. whose plant is probably Habranthus robustus,
possibly Z. mesochloa. Leaves green, sometimes
rather glaucous ; tube very short, rounded within
by the bending of the base of the filaments ;

172

AMARYLLIDACEjE.

perianth wliite, green below. Native of the Blue
mountains in Jamaica. The name .Tubispatha
having been now generally applied to this species,
though certainly not L fieri tier’s Bonarian plant, it
is not advisable to disturb it. This is properly a
stove plant, requiring a sandy soil, rest in the
winter, and much water in the summer. Ripens
seed freely.

Var. Hybrida. — Spofforthiae. SpofForthiana. Bot. Reg.
21. 1746. A very pretty pink mule from Tubi-
spatha impregnated by Carinata.

5. Nervosa. — H. and B. Kunth, 1. 285. Leaves linear,

smooth, 10-11 inches long, scape 6-8; peduncle
scarce longer than the spathe; perianth oue inch
and a half long, white, green below ; filaments
alternately equal; style longer. Six small processes
near the insertion of the filaments, if I rightly
understand Mr. Kunth. The addition that the
seeds are bulbiform (Kunth), if thereby green
fleshy seeds are meant, I have no hesitation in
pronouncing to be quite erroneous, nor were the
travellers likely to have seen both flower and seed.
America produces no genus of that section.

6. Gracilis. — PI. 29. f. 1. Specimina Ruiz ex Port.

de S. Maria. Herb. Lamb, absque foliis. Scapo
biunciali filiformi, pedunc. f unc. spatha § unc.
perianthio (albo ?) semunciam excedente, stylo
isomctro filamentis longiore suberecto, stigmate
minuto lobis brevibus. The specimens of this
singular and diminutive plant are without leaves ;
the form of the flower is funnel-shaped with the
points a little reflex. The stigma is less cleft than
usual in this genus. Scape filiform two inches,
peduncle spathe perianth half an inch (or a
little more), seemingly white; style equal, longer
than the filaments.

7. Minima. — PI. 24. f. 3. Specim. Tweedie ; from

Buenos Ayres. Herb. Hooker. Bulbo parvulo
collo semunciali, foliis filiformibus, scapis pluribus
2-3-uncialibus, spatha f tubata apice divisa, pe-
dunculis i uncialibus genuine 1, tubo £ , limbo
albo, sepalis extus rubescentibus, stylo bre-

AMARYLLIDACEiE.

173

vlore. Leaves filiform, scape 2-3 inches, pedun-
cles germen i, tube limb from 3 to 5-16ths,
white, sepals red outside, style shorter than the
limb, spathe slit at the point. This plant is still
more minute than the former, and not funnel-
shaped to the base in the dry specimen, but more
tubular.

8. Carinata. — Bot. Mag. 52. 2594. — Zeph. grandiflora.

Bot. Reg. 11. 992. quoad florem, non quoad folia
et sem. Leaves 5-16ths wide, channelled, keeled,
green, red at the base ; spathe longer than the
peduncle, slit partly on one side, tube green, f
long, limb rose-coloured, 2 inches long ; anthers
long, style white. From Mexico. Dr. Lindley
has withdrawn the name Grandiflora, because the
plate and description of Grandiflora in the Register
appertains to this species as to the flower, and to
Lindleyana as to the leaf and seed, through a
mistake of the gardener. A most beautiful species
flowering abundantly in the greenhouse (if kept
quite dry in the winter) in light sandy loam. I
have never seen it make any advance towards the
formation of seed, though tried in various aspects
and temperatures, but a mule as above stated has
been obtained from its pollen.

9. Rosea. — Bot. Reg. 10. 821. Bot. Mag. 52. 2537.

Leaves bright green, narrow, recumbent; flower
much smaller than Carinata, segments rather
obtuse, rose-coloured. Native of the mountains in
Cuba; a greenhouse plant, but its flowering is
promoted by heat in the summer.

Var. 2. Bifolia. — Lamark Enc. 1. 122. Poirret Enc.
Sup. 1. 316. Leaves acute, usually two, one a foot,
one four inches long (probably so only at the mo-
ment of flowering) ; scape a foot, as thick as a pen ;
limb an inch long; filaments shorter; stigma tri-
fid. In the woods of St. Domingo and Cayenne.
Bulb a little longer than a nut. They cite Plunder
M. S. t. 3. f. 137. Plum. cat. 7. and Aubl. H. Guin.
304. No. 3. There appears nothing to distinguish
it from Rosea but the alleged habit of the leaves,
and a more robust scape. A native of Cuba would

174

AMARYLLIDACEjE.

probably be found in St. Domingo. The leaves of
Rosea are more numerous, so that bifolia would not
do for the chief name.

10. Commersoniana. — PI. 29. f. 3. Specim. Commer-
son. ex Monte Video. Herb. Linn. Soc. Am. Ata-
masco minor. Red. Lil. 454. Foliis angustis line-
aribus, scapo debili, spatha pedunculo longiore,
perianthio circiter pollicari roseo. Leaves narrow,
scape slender, spathe longer than the peduncle,
perianth about an inch long, rose-coloured. The
native country of Redoute’s plant is not known ; it
agrees with Commerson’s specimen from Monte
Video, and is sufficiently distinct from Atamasco,
which is a native of North America. I understand
that the plant is lost at Paris ; it has never been in
this country.

1 1 . Lindleyana.' — PI. 35. f. 5. Grandiflora, Bot. Reg.
11. 902. quoad folia et semina, non quoad florem.
Am. minuta? H. and B. Nov. gen. 1.278. Kunth
Syn. 1. 285. Bulbo parvulo, foliis linearibus, pe-
dunculo semunciali, germine |, tubo viridi J unc.
limbo pallide roseo, unc. aculeis sub lente sex
minutissimis ad basin petalorum. This plant, a
native of Mexico, is distinguished by its smaller
size, pedunculated germen, and pale rose-coloured
flowers, from Verecunda, which it otherwise closely
resembles. It is the plant of which the leaves are
represented and the seed described under the name
Grandiflora, in consequence of the flower of Cari-
nata having appeared before the leaves on its first
importation, and the leaves added to the drawing
at a later period from a bulb of this species by a
mistake of the gardener. My plants of Lindleyana
are from an offset taken by Dr. Lindley’s hand
from the plant of which he described the leaf and
seed. If a pot of these bulbs, after remaining dry
in the greenhouse all winter, be placed on a warm
flue and watered in May, the flower buds will ap-
pear in a few hours. I believe this to be Hum-
boldt’s minuta, but it cannot be identified with
certainty, and the smaller species since discovered
make the name not desirable.

AMARYLLIDACEA3.

175

B. Sessile.

13. Sessilis. -Foliis gracilibus semicylinclricis viridibus,
ortu tortilibus ; germine sessili ; tubo ultra semun-
ciali, aculeis binis minutissimis ad petali cujusque
basin ; limbo albo, sepalis extus plus minus rube-
scentibus ; scapo in seminando deflexo ; stylo valde
deflexo.

Var. 1. Verecunda. — Bot. Mag. 52. 2593. Germen
above tube green limb 1^- long, white, tinged
with blush outside ; sepaline filaments shorter,
petaline just longer, than the style.

Var. 2. Striata. — Bot. Mag. 52. 2593. Sepals streaked
outside with red ; style longer than any of the
filaments. This plant does not appear to maintain
its distinguishing features with certainty when
raised by seed.

Var. 3. Ackermannia. — From Guatimala. Leaves rather
broader ; sepals red externally before expansion ;
perianth internally pure white ; style much longer
than in either of the other varieties. It does not
flower near so freely as them, and is more disposed
to make offsets. These plants require no trouble, but
to keep them dry in winter ; they may be crowded
in a small pot, stowed away any where dry, and
set out of doors in May or June, when they will
flower immediately ; or they may be kept in dry
sand, and planted out in May in a border of sandy
soil. Their flowers expand quite flat when the
sun shines, and are produced abundantly, and
every flower is followed by a seed-pod. The seed-
lings flower at an early age.

14. Grahamiana. — PI. 29. f. 2. Specim. Graham 358.
B. Herb. Lindl. Mexico, absque foliis. Scapo tri-
unciali, spatha Itc dimidio tubulosa, basi inflata
superne bifida (vel fenestrata ?) germine sessili l
unc. perianth 1-|, tubo gracillimo ^ unc. limbo unc.
laciniis ovalibus ultra latis (in spec, sicco pallide
purpureis infra lutescentibus) filamentis ^ unc.
liberis, stylo suberecto tubum superante, stig-
mate lobis obtusis, antheris limbo semunciam bre-
vioribus. This plant is nearly allied to sessilis.
The flower is more purpurascent, the tube rather

176

AMARYLLIDACEiE.

longer, and the stigma less trifid. The leaves are
wanting to the specimen.

Obs. The genus Zephyranthes, requiring rest in winter,
may be kept dry at that season, and planted out in the full
sun in very sandy soil in the spring.

§. 2. Antheris subulato-tortilibus, erectis apice reflexo ;
polline difformi ; stigmate obtusiore, lobis sub-erec-
tis. Genus diversum ? Argyropsis.

1. Candida. — Bot. Mag. 53. 2607. Am. Candida. Bot.
Reg. 9. 724.

Var. 1. Flore majore. — Specim. Herb. Hooker. Mat-
thews, 434. ex hortis veteribus per Lim® vallems.

Var. 2. Flore minore. — PI. 24. f. 2. Specim. Tweedie
ex ripa fluminis La Plata , ita ab argenteo plant®
hujusce floridissim® aspectu nuncupati. Herb.
Hooker.

Var. 3. Rubro extus suffusa. — From Buenos Ayres.
Reddish outside.

Var. 4. Fortuita. — Quadrilocularis. Ex Bonaria. Peri-
anthii segmentis et staminibus octo, stigmatis lobis
et capsul® loculis quatuor ; scapo pedali ; flore
majusculo.

This plant, conspicuous by its fleshy, semicylindrical and
rush-like leaves, which resist the severest frost of our usual
winters, has ripened its seeds with me after snow had lain
upon them for three weeks. I have seen the quicksilver
fifteen degrees below the freezing point (Fahren.) without
its losing more than the ends of its leaves. I have not been
able to ascertain that it is indigenous in the west of South
America, though abundant in old gardens in the valley of
Lima. There is no difference in the hardiness of the consti-
tution of the bulbs from Lima and those from Buenos Ayres,
where the banks of the Plata are so covered with it that it is
understood that the river was called La Plata, meaning
silver, on account of the profusion of its white blossom on the
shore. I have had seventy flowers expanded at once on a
small patch of the plant at Spofforth. It is strange that this
plant, which thrives in the hot valley of Lima, should have
stood out of doors here nine or ten years unprotected, with-
out ever losing its leaves entirely. Perhaps the strong cur-
rent of air which must accompany the rush of such a great

2i /’!. /h i<///i n

v <•/////.

'^oe. Hat/nud

''/Herbert,, del.

AMARYLLIDACEJE.

177

body of water as the Plata, and the evaporation from it, occa-
sions a degree of cold on its immediate banks which the latitude
would not otherwise admit. There is, however, a mystery in
the constitution of plants, which does not always depend upon
their native climate. I raised two species of Gesneria from ac-
cidental seeds lodging amongst the roots of the same plant of
Pitcairnia, plucked off a rock in Brazil, one of which objected
to the heat of a stove, and the other could not live through
the winter without it. Candida flowers abundantly here
under a warm south wall, upon the first autumnal rains after
a season of drought, which does not however at all affect its
foliage; and, if they cease, its blossoming will be suspended,
unless it be watered, and recommence later. I have an ex-
traordinary variety (var. 4.) from Buenos Ayres, which
flowers year after year with eight segments to the flower,
eight stamens, four lobes to the style, and four cells to the
capsule. It has a more robust scape a foot high, and very
conspicuous large flowers, expanding full three inches. I
have seen accidental flowers of Gladioli with a supernume-
rary stamen, sepal, and petal, and even with two, and a
regular fertile four-celled capsule on Camellia Japonica, but
this is the only instance I have observed of a seedling with
such a variation permanent. My belief is that this plant
belongs to a genus intermediate between Zephyranthes and
Cooperia, though I will not disturb it till the further species
of those genera can be thoroughly examined. It differs
from Zephyranthes, and agrees with Cooperia in having
deformed pollen, a point which I am persuaded must indi-
cate generic diversity, though we may be sure that the di-
versity, if positive, would not be confined to such a secret
feature. It differs from Zephyranthes in the stamen, and
from the known Cooperias in the absence of a cylindrical
tube and in its expansion under sunshine ; from both in the
substance of the leaf and their permanency. The erect lobes
of its stigma are peculiar to itself ; and it has as yet resisted
my repeated attempts to cross it with Zephyranthes. I sug-
gest the name Argyropsis, if its separation should be esta-
blished on further investigation, in allusion to its silvering
the banks of its native river, and giving name to the Argen-
tine republic. It comes very near, however, to Z. sessilis,
which forms a link between it and the pedunculated species,
and I can only separate it as a section of that genus.

I learnt from my lamented friend, Dr. Carey, that the

N

178

AMARYLLIDACEyE.

species of Zephyrantlies which I had sent to him had mul-
tiplied prodigiously in the East Indies, as well as the Hippe-
astra, the former being a substitute for our crocuses, the latter
for our tulips, in a tropical garden ; and he was earnest in his
application for every species of Zephyranthes.

There is a little mystery in the expansion of Candida. I
have seen it open quite flat in warm sunshine, but I have,
nevertheless, in very cold gloomy weather with a north
wind, seen its flowers standing at 3-4ths expansion at night,
and a few days after, when the wind was south-west and
warmer the flowers were not near so open even in the day ;
as if its expansion depended on the dryness of the at-
mosphere.

36. CooPERfA. — Germen erect; tube erect, long, slender,
cylindrical, widened at the mouth ; limb in its prime
stellate; filaments inserted at the mouth, nearly equal,
erect ; anthers erect, affixed at one-third from the
base, not versatile, fasciculate ; style erect. (Stigma
three-lobed, fimbriated, viscous ; lobes furrowed, ob-
tuse ; leaves linear, tortuous ; scape one-flowered ;
spathe one-valved.)

1. Drummondiana. — PI. 24. f. 2. 5 — 11. PI. 41. f. 16.
Drummondi, Bot. Reg. 22. 1835. ^Tar. 2. — Cldo-
rosolen. — PI. 24. f. 1. Bot. Mag. 63. >2482.

Leaves 1-1 2th to l-8th wide, 12-18 inches long,
tortuous, green, tending to glaucous; scape 4-13
inches ; spathe about 1J, slit or looped at the end ;
tube inches, greenish, often fading red ; limb
1|- long, white, acquiring often six broad red
stripes on the back in fading ; sepals tipped on
the back with green : style sometimes shorter than
the tube, sometimes exceeding the stamens ; fila-
ments free, about £ of an inch ; flower expanding
in the evening, sometimes beginning to close a
little in the morning, sometimes lasting four days
before it withers.

This plant is so variable, that three bulbs sent by Drum-
mond separately, and perhaps from different localities, flow-
ered at Spofforth, one with the style shorter than the tube,
one longer, but shorter than the stamens, and the third
longer than the stamens ; the difference of stature and colour
was also considerable, but the first of the three bulbs having

jfMARYLLIDACE.E.

179

produced, in the space of six months, five successive scapes,
has itself exhibited successively all the diversities which were
at first supposed to distinguish the three bulbs, and it is vain
to separate them. The flower always expands in the even-
ing, and is not usually seen in its perfect state after the first
night, the perianth becoming less stellate and the margins of
the segments curled, but it endures for three or four days in
that state. The bulb figured in the Bot. Reg. having pro-
duced a scape in August, when the weather was very fine
and hot, was placed out of doors in the full sun. It ex-
panded as usual in the evening, and in its prime the limb
was pure white ; the next morning it had become less stel-
late, and it lasted so four days, at the end of which, as it
began to wither, the tube became red, and a deep broad red
stripe appeared on the back of each segment, widest on the
sepals. The nocturnal flowering of this plant is an anomaly
in the order, and the more remarkable, because its nearest
kin, Zephyranthes, requires a powerful sun to make it ex-
pand. The flower is fragrant, smelling somewhat like a
primrose. I have the bulbs in a sandy compost, which
clearly suits them. They do not seem impatient of either
heat or cold, but like a free supply of water; as soon as the
seed on one scape is ripe, another seems ready to rise. The
flower has no irritability, and does not seem affected by the
presence or absence of sun after it has once expanded. A
single flower is not very conspicuous, but in a tuft they
would be very ornamental. Severe frosts occur in Texas, and
it may perhaps prove hardy; but its habit appears to be to
flower successively from the earliest spring till September,
the leaves growing principally late in the autumn, and in
the winter if protected.

2. Pedunculata. — PI. 42. f. 3, 4, 5. Seeptranthes Drum-
mondi. Graham. Edin. phil. Jour. 40. 413. 1836.
Zephyranthes Drummondi. Don. Sweet, Brit. fl. g.
ser. 2. 328. Nocte patens, fragrans ; bulbo de-
presso, foliis glaucis subsesquipedalibus i unc. latis
subobtusis, scapo superne attenuato unifloro, spatha
If, apice semunc. bifida, pedunculo f-1 unciali,
perianthio 2f-2j unciali albo, limbo 1a vel ultra,
antheras unciam superante, tubo 1-1 a unc. antheris
k uncialibus, petalinis sepalina vix quarta parte
superantibus, stylo f unciali erecto, stigmate sub-
erecto inc.luso. Bulb black, flattened at top, about
n 2

180

AMARYLLIDACE^.

If diameter; leaves glaucous, \ of an inch broad,
near ^ an inch close to the bulb, scape 8 inches
high, \ diameter ; peduncle an inch ; spathe bifid
at the end, at first covering the germen, afterwards
disclosing it on one side ; flower 2 inches and f or
more, tube about an inch, limb white, tinged with
pale blush in the bud, white when blown, striped
outside with red in withering, an inch longer than
the anthers ; anthers ^ of an inch long, the peta-
line scarcely exceeding the sepaline in position by
a quarter of their length ; style f of an inch long,
or shorter, erect; stigma suberect, included in the
tube ; capsule three-furrowed, f-ths long, | wide,
not widened at bottom.

This is another of Drummond’s bulbs, which has flowered
at Dr. Neill’s, in the Edinburgh and Glasgow Gardens, and at
Spoflorth. It was not observed during the night in Scotland,
nor seen to expand perfectly. It did not flower at Spoflorth
till the 20th of September, the weather being then very cold
and cloudy. The scape and spathe were pale reddish, the
bud tinged with a faint blush. A little before sunset it
expanded a little, the limb having grown pure white ; in
that posture it remained till the same hour the next evening,
when it opened a little wider, and on the third evening it
made a fresh effort and reached a state of about half expan-
sion ; the next morning the sepals acquired a red tinge on
the outside, and the flower began to shrivel. It had the
same primrose-like fragrance as Drummondiana. The peta-
line filaments were prolonged a little, adhering to the base
of the petals ; the style in my specimen was only half the
length of the tube. The tube was quite as erect and cylin-
drical as in Drummondiana, the figure in Sweet’s Br. fl. g.
being very incorrect in that respect. I was at a loss to
understand the capricious non-expansion of these nocturnal
flowers, but I am convinced that it arises from the manner
in which they have been treated. Increase of temperature
prevents their expansion, and probably would obstruct the
opening of any night-blowing flower. The requisite for
producing the flower is a certain mean temperature, but a
gradual decrease of temperature, such as usually takes place
at sunset, is necessary for its expansion. Therefore if the
plant be in a stove or warm greenhouse, and the weather
cold and cloudy, there is no decrease of temperature at even,

AMARYLLiDACEiE.

181

but perhaps an increase on shutting the lights and making
up the fire ; and so circumstanced the flower of the Coope-
rias obstinately refuses to open. If placed in the open air
the day before the flower is to blow, it feels the natural
evening refrigeration and expands like a star, and having
once attained that posture it preserves it till it withers, that
is about three days. The first' day my C. pedunculata tried
to open it had been very cold and gloomy, and where it
stood there could have been very little decrease of tempe-
rature at night ; on the two following days the weather
became gradually warmer, so that it felt a little more dimi-
nution of warmth each evening, and consequently opened a
little more, but the change was never great enough where it
stood to cause a perfect expansion. The natural effect
arising from decrease of temperature may be counteracted in
like manner by the artificial treatment of animals. The
African Whidah bird acquires fine plumage and a prodigious
tail at its vernal moult ; at the approach of winter it puts on
brown feathers and a short tail ; but if during the summer
it is kept in a cold, sunless, and airy situation, and just
before its autumnal moult is brought into a close and heated
room, it will acquire a renewal of its summer dress ; and
under such treatment it will never put on its winter plumage.
I can entertain no doubt as to the generic identity of this
plant and Drummondiana.

Professor Don looked upon it as a Zephyranthes, not
having seen the live specimen, and relying on the engraving
in Sweet’s Br. fl. g. from a drawing by Mr. M‘Nab; but the
engraving is very incorrect in the form and posture of the
tube. The outline I have given is from a drawing also by
Mr. M‘Nab, which was communicated by Dr. Graham to Sir
W. Hooker, and kindly forwarded by him to me, together
with the dry specimen of the flower received from Dr. Graham.
In that drawing, which is correct, and of which the outline
had been sent to the engraver before the bulb flowered at
Spofforth, the flower has exactly the form and posture of
Cooperia, and the only difference of structure is a very
trifling increase in the prolongation of the petaline stamens,
which are, however, only 1-1 6th of an inch longer than the
sepaline. If Mr. Don had seen the live specimen or Mr.
M‘Nab’s drawing, he would have entertained a different
opinion concerning this plant. Dr. Graham named it Scep-
tranthes, considering that its less expanded limb, its shorter

182

AMAUYLLIDACEiE.

tube, and pedunculated germen, distinguished it from
Cooperia. The expansion of C. Drummondiana depends as
much on the decrease of temperature; and I had one flower
of it in the winter that never opened at all, though it ripened
its seed. The comparative shortness of the tube can be only
looked upon as a specific feature ; witness the great diffe-
rence of its length in Hipp. breviflorum and solandriflorum,
Hymen. Carihhea and pedalis ; the pedunculated and sessile
germen occur together in the most nearly allied genus
Zephyranthes, as well as in Ismene and in Hymenocallis ;
and I can discover no feature of sufficient weight to separate
it generically from Drummondiana. I am indebted to Dr.
Neill for seeds ripened in his collection, and an excellent
sketch of the bulb with the scape and capsule, which is not
enlarged at bottom, as usual in Zephyranthes. The seeds
are like those of Drummondiana, flat and black. They vege-
tate readily. The leaves of Drummondiana have a glaucous
tendency, which perhaps is common to the whole genus.

Species dubia, jiore nondum viso.— Doubtful species.

3 ? Mexicana. — Bulbo depresso, foliis glaucis, canali-
culatis, sesquipedalibus, tortuosis, reciwnbentibus,
uncial, latis subobtusis. Bulb flattened at
the top ; leaves glaucous, channelled, a foot and a
half long, tortuous, recumbent, 3-16ths of an inch
wide, rather obtuse. Imported by Mr. Loddiges
from Mexico. I have not yet succeeded in obtain-
ing a flower from it, nor has it blossomed at
Hackney, where Mr. Loddiges has many bulbs of
it in a thriving state, but I think it will prove to
be a white Cooperia. It has the same flattened
summit to the bulb as C. pedunculata. I suspect
the cause of its not flowering to have been the
cessation of watering at the approach of winter,
and consequent delay of the growth of its leaves
till the spring. It is now making great growth of
leaves in September, and I doubt not that by en-
couraging them in the winter it will be induced to
flower in the spring or summer.

37. Haylockia. — Leaves linear; scape concealed, one-
flowered ; spathe one-valved, half-concealed, .di-
vided upwards ; germen concealed in the bulb ;
tube cylindrical, enlarged at the mouth, limb funnel-

AMARYLLIDACEjE.

183

shaped below, semipatent upwards ; filaments of
alternate length, conniving, inserted in the seg-
ments of the limb, the sepaline at their base, the
petaline higher ; anthers incumbent, versatile, at-
tached in the middle; style erect; stigma deeply trifid,
erect, with recurved points, concealed in the tube ;
capsule protruded on a short peduncle, round, 3-fur-
rowed, 3-valved ; seeds with a rounded back.

1. Pusilla. — Bot. Reg. 16. 1371. — Sternebergia Ame-
ricana. Hoffmansegg. verz. pfl. p 197. cum fig.
Berl. 1824. — Var. 2. Rubella. — Hoffmansegg, ib.
Variety with pink flowers. — Bulb small, leaves
hiemal, very narrow ; flower autumnal after a sea-
son of drought, tube an inch, limb an inch and
white tending to straw-coloured, stained with
purple without ; style white. Hoffmansegg states
that the plant is abundant at Maldonado, and that
it is found red as well as straw-coloured. The
flower has some affinity to the European Sterne-
bergia, but it belongs to a different section of the
order. The seed of Haylockia approaches to that
of Zephyranthes, with which it is closely con-
nected.

38. Pyrolirion. — Leaves attenuated at both ends ; scape
one-flowered ; germen sessile ; tube cylindrical,
erect ; limb campanulate with reflex points ; fila-
ments equal or alternately equal, suberect, patent ;
anthers versatile ; style erect or reclining ? stigma
trifid.

1. Aureum. — PI. 29. fig. 4. Spec. 400. Herb. Lindl.
PI. 23. fig. 3. Spec. Dombey Herb. Soc. Linn.
A. aurea. Flor. Per. 3. 56. 286. A. tubiflora, L’Her.
S. A. 10. Lil. Narc. Feuill. Obs. 3. 29. 26. A. Pe-
ruviana. Ker. Lamark. Bulb beset with blind off-
sets, scape a foot high, flower 4-4f inches long,
golden, mouth of the tube scaled ; style scarcely
exceeding the filaments. Dombey’s specimen,
which is L’Heritier’s plant from Lima, has the
flower more than four inches long. The peduncles
in the plate of Flor. Per. are false, and not men-
tioned in the description. It flowers in January
and February in the cornfields of Peru at Lurin,

184

AMARYLLIDACE^E.

Surco, Magdalena, Lurizanclio, and Chorillo, near
Lima ; called Hamanco de Antibo.

2. Flammeum. — Flor. Per. 3. 56. 286. Lilio-narc. mon.

coccin. Feuill. Obs. 3. 29. 21. Flower shorter,
limb only two inches, fiery orange, faucial scales
short, truncate, and crenate. In the cornfields of
Cliancay, near Huaura, and in the fields and on
the hills of Conception in Chili.

3. Flavum.— Spec. Ruiz. Herb. Lamb. — Pyrol. aureum,

fauce laevi. Bot. Reg. 20. 1724. Scape short ;
flower 3 inches or less, pale golden ; tube smooth ;
filaments much shorter than the style ; stigma
deeply trifid.

There is a specimen from Peru with the scape shorter
than the flower, which is decidedly flavum ; two other spe-
cimens of aureum are erroneously marked flavum. These
three plants are very nearly allied, but as Ruiz distinctly
describes the faucial scales of aureum and flammeum, the
plant in which Dr. Lindley found none, must be the third
species which Ruiz named, but did not describe. It also
agrees with it and differs from aureum in the short scape,
long style, and paler colour ; and disagrees with aureum in
not having the circle of blind offsets, described by Ruiz,
which I find in an imported bulb of Pyrolirion, which has
not flowered. The figures in the Flor. Per. are quite un-
worthy of faith, but the descriptions of Ruiz appear to me
singularly correct. The three species are probably dis-
tinct ; aureum with larger golden flower, style little exceed-
ing the filaments, and faucial scales longer; flammeum with
smaller flowers, fiery orange, and crenate scales ; and flam-
meum with paler flowers, longer style, and short scape.
The filaments of flavum in Mr. Lambert’s specimen are as
nearly equal as possible.

Species dubia. — Doubtful species.

4 ? A lbicans. — Lil. Narc. monanthus flore albicante,
tubo prselongo. Feuillet Obs. 3. 29. f. 20. Scapo
tereti, virente, 8-unc. tubo 2-unc. 1^ lin. lato ;
limbo circiter 2-unc. laciniis acutis apice reflexo ;
foliis 7-8-unc. 1^ lineas latis, viridibus, acutis,
canaliculatis. In valle Ylo intra montes Peruviae,
situ aridissimo. This plant, from Ylo in Peru, is

AMARYLLIDACEiE.

185

only known to us by Feuillet’s description and bad
plate, with a whitish tubular flower, and limb
reflex at the point ; it agrees with no genus but
Pyrolirion. There is no spathe in the figure. It
is impossible to be quite certain that this plant may
not be a Cooperia ; but both the leaves and reflex
points of the limb seem to correspond with those
of Pyrolirion.

I have not possessed this genus long enough to be sure of
its habits. It flowers before the full growth of the leaves in
S. America in January and February, that is at and after
midsummer, therefore the leafing of the plant is autumnal.
As it had had leaves all the last summer at Spofforth, I left
it dry through the winter and started it in the spring;
but, as it appears disposed to push more leaves this year in
September, I apprehend that its proper season of rest will be '
the spring and fore-part of the summer.

§. 3. Oporanthiformes. — Scape solid ; filaments free.
[Seeds testaceous?]

39. Gethyllis.— Bulb ovate, coats often imperfect ; leaves
linear ; scape and germen concealed in the bulb ; tube
cylindrical, long, adhering to the lower part of the
style; limb regular, patent; filaments short, straight,
diverging, sometimes by superfluity doubled or
trebled, or multiplied numerously, inserted at the
mouth of the tube ; anthers erect, attached at the
base ; style erect ; stigma simple or triangular ; cap-
sule with the scape extended, diaphanous, pulpy ;
seeds small, round. S. African plants.

1. Spiralis. — Bot. Mag. 27. 1088. with minute and

excellent particulars of the genus and species by
Mr. Ker. Papiria. Thunberg. Act. Phys. 1. 2. 111.
Leaves narrow, smooth, a little spiral ; tube white,
near 3 inches long, limb 1^, white, purplish with-
out; filaments six.

2. Ciliaris. — Jacq. Sclioen. 1. 79. Leaves narrow, cili-

ated ; tube cylindrical, widened near the base; limb
white, patent ; filaments six, with three anthers on
each, stigma obtuse. Flowers before the leaves.

3. A fra. — Bot. Reg. 12. 1016. — Leaves narrow, smooth,

spotted ; tube purple, 2 inches long, thicker than

186

AMARYLLIDACE^E.

in spiralis ; limb white within, purplish without ;
anthers large, 10-12; style longer than filaments ;
stigma triangular, almost 3-lobed. Flowers sweet.

4. Villosa. — PI. 25. fig. 7. Specim. Masson. Herb.

Banks. Papiria villosa. Thunberg. act. 1. 2. 111.
Bulb ovate, small, with a cylindrical neck ; leaves
very villous and narrow ; tube slender, near 3 inches
long ; limb white, about filaments and anthers
six, 3 shorter.

5. Verticillata. — PI. 25. fig. 6. Specim. Masson Herb.

Banks, absque flore. Vide Brown prod. 290. Bulbo
parvulo ovato, foliis quinque filiformibus ; fructu
ovali unciali ; scapo unciali. Bulb small, oblong,
ovate ; leaves five, like threads, 4 inches long ;
fruit an inch long on a scape an inch long. I know
nothing of its flower.

6. Undulata. — PI. 25. fig. 5. Specim. Masson Herb.

Banks, absque flore. Bulbo oblongo ovato, lj unc.
lato, collo foliis sub terra vaginantibus producto
5-unciali ; foliis 20, 7-uncialibus, erectis, undula-
tissimis, setose ciliatis. This singular plant, with
leaves undulated extremely and in a most remark-
able manner, and ciliated with strong bristles, is
only known by Masson’s specimen, with no account
of the flower, which, perhaps, he never saw. It is
the largest known species.

7? Lanceolata. — Thunb. act. 1.2. 111. cum fig. Bulb
small, ovate ; leaves smooth, of the length of the
flower. On the hills near Buffalo river and else-
where. Mr. Ker refers this plant to Curculigo. I
have not been able to obtain a sight of Thunberg’s
figure.

8 ? Rosea. — Ecklon topogr. verz. 4. Flowers rose-co-
loured ; of smaller stature than the rest. At the
foot of Bavian’s-berg, near Gnadenthal. I know
nothing further of this plant.

40. Sternebergia. — Bulb ovate ; leaves linear, following
the flower ; germen subterraneous ; scape autumnal,
one-flowered ; tube erect, cylindrical ; limb semi-
patent; filaments filiform, dilated at the base, con-
niving, alternately longer; anthers short, oblong,

AMARYLLlDACEiE.

187

versatile ; style thicker upwards ; stigma 3-lobed ;
seeds black, shining, dotted, with a thick white
spungy cord.

1. Colchiciflora. — Kit. et Wald. 2. 157. M. v. Bieber-

stein, 1. 261. Flor. Graec. Narcissus autumn. Clus.

2. 164. Specim. Herb. Lindl. Bulb small, leaves
erect, keeled, tortuous, blunt, about a line wide ;
flower autumnal, yellow7, sweet; tube long, limb
shorter. Native of Hungary and Tauris. Seed
ripe in June ; capsule spherical. Called in Thrace
Ciden Sair.

Var. 2. Dalmatica. — Humilior flore minore lsete flavo,
laciniis angustioribus tubum subaequantibus, stylo
stamina excedente. In Dalmatia prope Czerno.
Reichenb. FI. Germ. p. 87. Variety of lower
growth, with a smaller bright yellow flower, seg-
ments narrow, nearly as long as the tube; the style
longer than the stamens. Found by Reichenbach
near Czerno in Dalmatia.

2. Clusiana. — Ker. Narcissus Persicus. Clus. 2. 163.

Leaves lorate, tortuous, glaucous, erect ; flower au-
tumnal, pale yellow, stinking; sepals wider. Sent
to Clusius from Constantinople.

3. Citrina. — Flor. Graec. 4. 311. Limb about an inch

long or more, pale yellow, segments narrow ; tube
above an inch, style 1$, a little exceeding the longer
stamens; stigma trifid, lobes recurved; leaves erect,
a little tortuous. Flowers late in the autumn on
Mount Olon in the Morea.

4 ? iEtnensis. — Rafinesque Schmatz Caratteri. p. 84.
t. 18. f. 2. Am. iEtn. Filaments said to be equal,
probably inaccurately ; otherwise scarcely to be
distinguished from Citrina. Leaves linear, acute,
spiral ; spathe acute, subulate, of the length of the
germen ; scape shorter than the flower ; perianth
(probably meaning limb) an inch long, erect, cam-
panulate ; segments oblong, obtuse, pale yellow ;
filaments filiform, anthers very small, round; stigma
3-lobed. Amongst the fern in the woods of Mount
iEtna, near Nicolosi. The tube is not mentioned.

Caucasica. — Willd. is a misquotation from Marsh, von

188

AMARYLLIDACEjE.

Bieberstein, referable to Merendera Caucasica. Sterne-
bergia Colchiciflora has been introduced into this country
lately from Hungary by the Hon. W. Fox Strangways, with
whom it has flowered.

41. Oporanthus. — Bulb roundish; leaves hiemal ; scape

autumnal, one- flowered ; spathe tubular, divided at
the point ; germen erect ; tube short, erect, tending
to funnel-shaped ; limb regular ; filaments inserted in
the tube below the limb, decurrent, conniving ; an-
thers versatile ; stigma trifid ; ovules roundish, irre-
gularly angular by compression. Seeds nowhere
described.

1. Luteus. — Am. lutea. Bot. Mag. 5. 200. Redoute lil.

148. Narcissus autumnalis major. Clus. 2. 164.
Mountain pastures in Spain, Italy, and Greece.

Var. flore pleno. Hill’s Eden.

It is very strange that no writer has described the seed
of this plant, which is much cultivated, nor have I ever seen
it. Hill speaks of sowing the seed in beds, as if he had
readily obtained it, and asserts that the seedlings vary much
in the shade of yellow, and he gives a figure of a double
variety which is probably lost. That the seed is roundish
and black I entertain no doubt. The only variation I have
seen is a narrower and a broader leaf. This plant is hardy,
but the bulbs often rot with me in the open ground if the
summer is wet. Probably they should be taken up and dried
when the leaf decays.

2. Exiguus. — Amaryllis. Schousboc Morocco, 1. 160.

Scape an inch high; leaves 1-3, short; tube short,
limb campanulate, yellow ; segments equal, obtuse,
with a deeper yellow midrib ; filaments nearly as
long as the limb; style filiform. Tangiers.

42. Lapiedra. — Perianth regular, expanded, star-shaped ;

filaments straight, erect; anthers arrow-headed (in-
cumbent?); seeds small, angular. PI. 42. f. 10. 11.

1. Placiana. — Sparganium Pla^ae. Clus. PI. rar. 164.
Lapiedra Martinezii. Lagasca Nov. sp. et gen.
p. 14. Crinum Martinezii, ib. Barrelius, ic. 993.
A.D. 1714. Flowers about eight, white; style and
stigma white, spathe 2-3 leaved; leaves (two only?)
longitudinally striped with white, linear, obtuse.

AMARYLLIDACEiE.

189

It is very remarkable that this plant should have been
described above 200 years ago by Clusius, with particulars
concerning it which as yet we only know from his report,
and with a precise indication of the spot in which it still
grows spontaneously, and that no botanist, as far as I can
perceive, has since noticed his account of the plant under
the name Sparganium Pla^ae, by which it is indicated on
the margin of the page in which it is described. He states
that it was pointed out to him in its native locality, on the
stony heights above Valentia, by Dr. Pla^a, a physician of
that town ; and in that very spot it is still pointed out by
Lagasca, as growing amongst the clefts of rocks, without any
reference to Clusius. It is difficult to understand, even in
the vague state of botany at that period, how Dr. Pla^a
should have looked upon it to be a Sparganium, which is
an ancient name for a genus of a very different family, but
there is no reason for rejecting the specific name which
Clusius gave to it in commemoration of its first discoverer.
It seems to have been rather overlooked than intentionally
laid aside, and as the later name has had little currency I
have thought it undoubtedly proper to restore the original
one. According to Clusius it usually has two leaves, in
form like those of Oporanthus luteus, but marked with a
longitudinal white stripe. He states the seeds, which are
not noticed by Lagasca, to be small and angular. I con-
jecture the scape to be solid, and the seeds testaceous. It is
strange that Spanish plants of such easy access, and whose
locality is so well known, as Lapiedra and Tapeinanthus,
should never have been brought into cultivation, nor speci-
mens even introduced into any herbarium. Lapiedra appears
to be one of the points by which Amaryllidese approach the
hypogynous Allium and Ornithogalum. If the seeds, con-
trary to my expectation, should prove to be fleshy, which I
think very improbable, the genus would properly follow
Carpolyza. According to Lagasca it grows also near the
church of San Fuen, near Algesiras, and near Malaga, and
it might certainly be easily obtained. The anthers are
asserted to be arrow-headed and incumbent ; with the form
of an arrow-head I should have expected them rather to be
erect, like those of Hypoxis. The plant having been called
a Crinum, I assume it to be schistandrous ; it may, how-
ever, prove to be porandrous, it which case it would
stand amongst Galantheae, probably next to Leucojum.

190

AMARYLLIDACE.E.

§ 4. Pancratiformes. — Cup staminiferous.

£ Seeds black, shelly.

43. Tapeinanthus. — Perianth with ovate-oblong segments,

short cup, filaments long and diverging, anthers short
and incumbent, style erect with an obtuse stigma ;
scape rising before the leaves. — N. B. Spathe is said
to be one-leaved ; bulb small and pear-shaped.

1. Humilis. — Pancratium humile. Cavanilles icones.
This curious little Spanish plant is only known by
the representation and description given by Cava-
nilles. It has a short scape rising in the autumn
before the leaves, bearing two yellow flowers ; the
leaves are two, and extremely slender. The style
is probably club-shaped. This plant, which is a
native of the territory of Seville, appears to furnish
a link between Oporanthus and Pancratium, one
of those lateral connections which falsify continu-
ous arrangements. The tube of the flower is per-
haps little more than an annular connexion of the
limb at the base of the inconspicuous cup, but it is
not particularly described. Its separation from
all the other genera of this section is evident ; the
staminiferous cup distinguishes it from all the other
sections. It is much to be wished that some
person, who has the means, would interest himself
to procure the bulbs from Seville.

44. Chlidanthus. — Tube erect, cylindrical, triangular,

widened at the mouth ; limb nearly equal, semi-
patent ; filaments very short, curved, inserted in the
points of the alternately unequal teeth of a thin
membrane adhering completely to the tube and base
of the petals, but partible ; anthers attached near the
base ; style erect. (Germen erect, oblong, trian-
gular ; stigma trifid, patent ; leaves linear-lorate,
sheathing at the base ; umbel few-flowered.)

1. Fragrans. — PI. 27. f. 2. Herb. App. 46. Bot. Reg.
8. 640. Lindl. Coll. 34. Leaves about a quarter of
an inch wide, glaucous, erect ; flowers yellow, fra-
grant ; germen subsessile ; tube about two inches,
limb 1^ or more ; style longer than filaments,
shorter than limb ; stigma widely trifid ; ovules

AM ARY LLIDACEJE.

191

22 or more in each cell, opaque, flattened, heaped
in two rows, scarcely lapping; [scape very flat, with
winged furrowed margins]. Fruit not seen.

This plant was named and first described by me in the
Appendix ; and shortly after was represented and described
by Professor Lindiey, who discovered a slight membranous
connexion of the base of the filaments, and irregular half-
abortive dentate wings to the shorter or sepaline filaments,
which I did not perceive in the specimen on which I founded
the genus ; but it had travelled 260 miles, from Highclere,
and was a little withered, and the stigma had, as it appeared
afterwards, been disfigured so as to appear obtuse. I have
since re-examined it, expecting from its solid scape, that,
being an occidental plant, it must appertain to the genera
with a membranous cup. I find a fine six-toothed mem-
brane, in which the filaments are inserted, adhering to the
tube and lower part of the petals, the petaline teeth being
prolonged. This membrane while the flower is fresh is
partible, separating from the perianth like acuminate wings
to the decurrent filaments, by taking hold of the point of
the filament and pulling it. Those points are very short
and ultimately curved. Here, therefore, we have the con-
necting link by which the solid-scaped shell-seeded Ama-
ryllidese without cup, like Oporanthus, come in contact
with the Pancratiform section ; the membrane in Cldidan-
thus being an imperfect or rather incipient manifestation
of a cup. It brings Chlidanthus nearer to Clinanthus, in
w'hich, both from the appearance of the specimens and from
the name Pancratium given to it by Ruiz, I judge the
membranous teeth to be free from the perianth, and, unless
I have been deceived in a difficult examination, equal.
I may take this opportunity of stating, that, although the
plates in the Flor. Peruv. are disgracefully inaccurate,
wherever I had doubted the text of Ruiz, 1 have found
subsequent reason to bear testimony to his great accuracy.
He would not have named any plant Pancratium which had
not some appearance of a cup. Chlidanthus increases so
rapidly by offsets and splitting of the main bulb that it is
difficult to keep bulbs of a size to flow'er. It grows very
vigorously, and flowers in June or July, in a border of
which the soil is much warmed by contact with the wall
of a stove. In the open garden it flourishes, if taken up

192

AMARYLLIDACEjE.

and kept dry in winter, but the bulbs seem liable to canker
in peat. It has very much the constitution of the tender
Narcisseae, and likes a fertile loam; but, except where the
ground is heated by a flue, the bulbs, if not killed by frost,
are injured by moisture. They should be taken up at the
approach of winter without destroying the fibres, and placed
in a pot of sufficient size to contain them, dry sandy soil
being poured in to cover them ; they may then be set in
any dry warm situation till April, when, however dry they
may be, they will begin to sprout. All offsets should then
be taken off, and they may be set either in pots or in a
sunny border. No native specimens of this plant occur
in any herbarium with which I am acquainted. Mr.
Brookes’s bulbs were said to have come from Chili, but I
have some reason for doubting the fact. Bulbs of Chlid.
fragrans were sent to my brother from Buenos Ayres about
the same time, I believe at the same time precisely ; but it
may have been an inhabitant of gardens there, like Ismene
Calatliina, which accompanied it. Mr. Ker and Dr. Lindley
were quite mistaken in identifying this sessile plant with
Pancratium luteum of Ruiz, the Clinanthus luteus of my
appendix, a pedunculated plant, concerning which see the
next genus.

45. Clinanthus. — Tube long, funnel-shaped ; limb short,
continuous ; filaments acuminately winged (connected
by the wings ?) decurrent ; anthers short, broad at the
base, erect, attached at the base. [Scape peduncu-
lated ; spathe 2-valved ; germen round ; leaves linear
lorate, sheathing at the base.] Native of Peru.

]. Luteus. — PI. 27. f. 1. Herb. App. 40. Specimina
Pancr. luteum. Ruiz. Herb. Lamb. Leaves 6 or 7
inches long (at the time of flowering, perhaps
longer when full grown) ; j|-| wide, subacute ;
scape 2-flowered ; peduncles -§-1? long; perianth
yellow, near two inches long, tube li, limb near
f, segments rather obtuse. The specimens seem
to be all two-flowered, two of them being perfect,
the others broken.

It seems strange that, after I had described this plant
and Chlid. fragrans, this with long peduncles and small
flowers should have been mistaken for a half-blown specimen
of Chlid. fragrans, which has the flowers nearly twice as

F^.Z7.

AWARYLLIDACEJE.

U)3

long, with a limb four times as large, and the germen sub-
sessile ; indeed it is described in the Collectanea as actually
sessile. With a little moi’e consideration, before I was
charged with having made two genera of one species, it
might have been perceived that it was as impossible for a
flower, which has a long peduncle before it attains its full
size, to have none, and become sessile when full blown, as
for a man to become shorter in the maturity of life; but the
most accurate of men ai*e liable to such occasional oversights.
The comparison of two plants in the same state, whether dry
or fresh, is more satisfactory than that of a fresh plant with
a dry specimen, in which the parts may collapse unequally :
I have therefore given an exact outline of a dry specimen of
Chlid. fragrans in the same plate with the outlines of Cli-
nanthus luteus, where their difference is very manifest. My
description of C. luteus in the App. was made from the
upper of the two specimens of that plant therein represented,
in which the flower has a most decided slope from the ger-
men, instead of being erect as in Chlidanthus. The second
specimen, laid in since that time, makes it a little doubtful
what the posture of the living flower may be, but the dis-
section gives an interior different from Chlidanthus. The
round pedunculated germen gives reason to expect also a
difference in the fruit. I think the flowers cannot be natu-
rally erect, but that some of them, in the newer specimens,
have been forced into an erect posture in pressing them
under paper. If it should hereafter appear, that in conse-
quence of any variability of an imperfect cup, they can be
generically united, I should wish the name Chlidanthus to
be preferred. I believe, however, that the dentate cup of
Clinanthus will be found to be perfect, and the genus
separate. It is evident that they are two very distinct
species.

64. Urceolina. — Bulb roundish ; leaves petiolated, broad
oval, sestival ; scape festival, germen ovate, three-fur-
rowed; peduncles curved, flowers pendulous; tube
straight, slender, cylindrical, enlarged at the mouth ;
limb ventricose ; (filaments a little diverging?) an-
• thers incumbent, style straight, stigma obtuse. Seeds
numerous, small (black ?)

I. Pendula. — Crinum urceolatum. Flor. Per. Urceo-
laria. Herb. App. Urceolina. Reichenbaclu Colla-

o

194

AMARYLLIDACE^E.

nia. R. et Schultes. Leaves petiolated, 3 palms
long and half a palm wide ; scape 1 foot, flowers
5, two inches long, lower half of the limb yellow,
upper green with white edges ; filaments and style
longer than the limb. Grows in the woods of the
Andes at Pozuzo, and in the district of Pampa-
marca in Peru ; and flowers from June to November.
The size of the flower is probably grossly exagge-
rated in the plate in the Flor. Peruv. If it is not,
the plant I have named Fulva is very different
from it.

2. Fulva. — PI. 26. fig. 5. Specim. Matthews, 868. C.
urceolatum. Herb. Hooker. Petiole 4 inches, leaf
9 inches by 4 ; scape 7J, spatlie 2 inches, with five
valves ; peduncles of various length, flowers eight,
germen ovate, tube slender, above half an inch
long, limb near an inch (tawny, with pale margins,
and tipped with green ?) filaments and style longer
than the limb. Found at Parcahuanca in Peru in
December.

This may perhaps be only a variety of the former, but its
flowers are smaller and more numerous, and its colour seem-
ingly different : Ruiz states five to be the number of flowers
in the first species. This is distinguishable from Leperiza
(Ruiz’s Pancratium latifolium) by no outward feature besides
the supposed absence of a cup, but a longer and slenderer
tube and broader leaf; and I consider the two genera to be
so closely allied, that I think some attempt to form a mem-
branous cup must be found in this genus when better known.
Under that persuasion I place it in this section. If there is
no exhibition of the membrane, it must be removed to the
second section, after Lapiedra. The plants which I possess,
probably the second species, produce from two to three leaves,
which suffer very much if exposed to a hot sun, from which
they require to be screened. They must be kept quite dr}r
in winter in the greenhouse, and in very hot weather I found
it advisable to place them out of doors behind a north wall.
They are accustomed to the shade of woods. The bulbs in-
crease by offsets, of which the leaf pushes up at some dis-
tance from the parent. The leaf is much like that of Eucro-
sia and Griffinia, but with a longer and slenderer footstalk ;
the habit precisely that of Eucrosia. It seems to dislike

AMARYLLIDACE.E.

195

strong heat and sunshine so much, that it will perhaps suc-
ceed best in the open ground, taking up the bulbs when the
leaves decay. It seems to thrive better in a fertile loam than
in light soil. I expect, from the increased size of my prin-
cipal bulb, to see it flower next spring.

47. Leperiza. — Bulb roundish ; (coats imbricating ?) leaves

petiolated, wide oval, aestival; scape aestival ; pe-
duncles curved ; germen ovate, three-furrowed, pen-
dulous ; tube short ; filaments decurrent in the cup ;
anthers incumbent ; style straight ; stigma obtuse ;
capsule ovate, three-furrowed ; seeds many, small,
oblong, roundish (black ?)

1. Latifolia. — Pancratium latifolium. Flor. Per. 3. 54.
284. Leaves about four, narrower than in Urceo-
lina ; scape 8-9 inches ; spathe 5-6 valved, or with
spathe-like bractes ; flowers about five, li inch
long, reddish yellow tipped with green ; cup
toothed between the filaments; filaments a little,
style much, exceeding the limb. Grows in the
shady and damp woods of the Peruvian Andes
near Vitoc, on the hills and lands of Tarma.

The extraordinary bulb, represented in the plate of the
Flor. Per. with opposite coats like the scales of a lilium, and
the sinuosity of the filaments, are not warranted by Ruiz’s
description, and are probably false, like many other things
in those engravings. If that be the case, it will be separated
from Urceolina by the cup and tube only, and, if Urceolina
should prove to agree with it in those respects, Leperiza will
merge in that genus. Its habit seems, by Ruiz’s account, to
be precisely similar.

48. Carpodetes. — Bulb oblong ; leaves vaginating, attenu-

ated upwards ; spathe one-leaved, peduncles erect ;
germen oblong, constricted in the middle ; tube cy-
lindrical, slender, curved (enlarged upwards?) limb
regular (cup short?) filaments straight, alternately
equal ; style straight ; stigma obtuse ; capsule obo-
vate, three-furrowed, constricted in the middle ; seeds
large. In steep broken ground near Obragillo, in
the province of Canta in Peru ; called Chihuanhuaita
by the natives.

1. Recurvata. — Pancratium recurvatum. Flor. Per. 3.
54. 285. Bulb purplish with black spots and a long
o 2

A MARY LLTDACEiE.

196

neck; leaves half ail inch wide, about 10 long-,
rather obtuse ; scape rising in the centre shorter ;
spathe very large, purple, slit on one side, near
4 inches long; peduncles 3, unequal, lf-4J long;
colour of the flower between purple and yellow ;
style scarcely longer than the filaments.

This plant is evidently of a genus quite distinct from any
other that is known, and differs widely in bulb and foliage,
as well as in flower, fruit, and habit, from Stenomesson and
Leperiza, with which Mr. Ker wished to unite it. It is
much more nearly allied to Coburghia.

49. Coburghia. — Bulb ovate ; germen ovate, 3-furrowed ;
tube bent, cylindrical ; subventricosely enlarged ;
limb shorter than the tube, equal, half open, cernu-
ous ; filaments nearly equal, a little conniving, con-
nected by a tubular cup ; anthers erect; style a little
recurved; stigma obtuse, triangular ; capsule erect;
oblong, triangular, 3-lobed, 3-celled, 3-valved ; seeds
black.

1. Incarnata. — Sweet Br. fl. g. s. 2. 17. Pane, incarn.

Kunth. nov. gen. 1. 223. Leaves thick, glaucous,
obtuse ; peduncles short ; flowers under 5 inches,
crimson, with a green spot on each segment ; cup
campanulate, 12-toothed ; style equal to the fila-
ments, shorter than the limb. From Quito, on
the banks of the river Machangara. The bulb is
larger, the leaves larger, and less obtuse than in
Trichroma.

2. Trichroma. —Pane, trichromum. Nov. veg. d. De

la Llave et Lexarza. fasc. 1. Leaves thick, glau-
cous, obtuse ; scape a little longer, 5-flowered ;
flowers near 3 inches ; tube 2 inches, enlarged,
furrowed, light red ; segments ovate, suberect, with
an oblong ovate spot, white within, green without,
lined with green ; sepals hooked, petals acute ; cup
very short, with green-tipped teeth ; filaments
erect, much longer ; anthers linear, erect, after
inversion versatile. Cultivated in pots with great
care at Mexico, where it flowers at various seasons.
Native habitation not recorded.

3. Variegata. — Pane, varieg. Flor. Peruv. 3. 55. Leaves

thick, glaucous ; spathe with 2 outer valves, and

c ///)'/ ■( ch'njSas&s/rn

Plate. 28.

'’WHerber't.daL

AMAKYLLIDACEiE.

197

5 spathe-like bractes ; peduncles short, curved ;
Howers a span long-, cernuous ; tube yellow and
rose-coloured ; the limb yellow, margined with
rose colour, with a spot on each segment, green
without and white within; sepals wider, longer;
cup tubular, much shorter than the limb, with six
little forked reflex green teeth ; filaments shorter
than the style, which equals the limb. In gardens
at Lima. Precise habitation not known.

These three species are not very easily distinguished from
each other in bulb and foliage; they increase rapidly by
offsets, and are very unwilling to flower with us. The first
is larger, with broader leaves.

4. Fulva. — Bot. Reg. 18. 1497. Bot. Mag. 60. 3221.
Leaves sub-glaucous, thinner than the three former;
peduncles short; tube 3^ inches, fulvous ; limb 1^,
fulvous, tipped with green ; cup half an inch; fila-
ments nearly equal to the limb, style longer. The
precise habitation of this beautiful plant is not
known, the bulbs having been purchased at Liver-
pool by J. Wilroore, Esq. of Oldfield, near Bir-
mingham ; by whose gardener it was obligingly
communicated to mine, who mislaid his direction.
It was figured in the Register before I had ascer-
tained to whom I was indebted for it, and indeed I
never learned till 1 read it in the B. Magazine.

I have ascertained that all those bulbs succeed well in a
strong rich alluvial soil, and probably they will like old rotten
manure ; they thrive well in the open ground in summer,
but must be taken up and kept dry, or nearly so, in winter.
The bulbs are not delicate, but will not endure our winter,
except near the front wall of a stove, or with some protec-
tion to keep them dry. I consider the application of heat,
after the full growth of the leaves, to be the most likely way
to promote their flowering, but the first growth of the leaves
should be made in a cool and airv situation, or they will be
weak.

5? Discolor. — Feuillet,Obs.2. p.29. Lilio-narcissus poly-
anthus. cum icone. Bulb 2 inches long, rather less
wide, chesnut coloured; scape 2 feet high, a little
flattened and two-edged ; leaves 6 or 7, surrounding
the scape, 9 inches long and \ an inch wide at the

198

AMARYLLIDACE.E.

time of flowering, fine green, deeply channelled,
smooth, obtuse ; umbel 4-5-flowered ; perianth red
on the outside, variegated with red and yellow
within ; tube about one inch long, one line wide
at the bottom, enlarged upwards ; limb with
rounded lobes. Found on mountains of Chili,
and in 17° 39' south lat. which must be in La
Paz of Peru. It is possible that this plant may
be a Clinanthus, but I have very little doubt in
referring it to Coburghia, and 1 apprehend that
it must be allied to Fulva.

50. Stenomesson. — Bulb roundish, neck very narrow; leaves
at first compressed at the margin ; umbel 2-6 or more
flowered, pedunculated ; tube constricted in the mid-
dle, wider upwards, a little curved ; limb short ; fila-
ments straight, connected by a membrane ; anthers
short, incumbent ; style straight, before maturity
sloping ; stigma dilated ; capsule broad-ovate, 3-
furrowed, 3-lobed, 3-celled ; seeds black, obliquely
oblong.

1. Flavum. — PI. 28. f. 1. Bot. Mag. 52. 2641. Pancrat.

flavum. Flor. Per. 3. 54. 284. Chrysiphiala flava.
Bot. Reg. 10. 778. Perianth golden; teeth of the
cup straight, irregular ; filaments a little, style
much, longer than the limb. On sandy hills in
Peru by Lurin and Pachacama.

2. Curvidentatum. — PL 28. f. 2. Bot. Mag. 52. 2640.

Sphaerotele Peruviana. Prezl Rel. Flank, p. 119.
t. 16. Specim. Matthews, 399. ex collibus Aman-
caes prope Limam Herb. Hooker. Herb. Lindl.
Perianth golden, slender ; teeth of the cup long,
bifid, recurved ; style and filaments equal, longer
than the limb ; flowers 2-7. The Sphaerotele of
Prezl is unquestionably this plant.

3. Pauciflorum. — PI. 28. f. 3. Chrysiphiala pauciflora.

Hooker. Ex. Flor. t. 132. Perianth golden, thicker,
shorter than in Curvidentatum ; tube and tips
green, teeth of the crown short, bifid, recurved.

4. Aurantiacum. — Humb. Kunth, 1.280. Leaves lance-

linear, 5 flowered or under; perianth orange, 14-15
lines long ; peduncles 1-1^ inch; intervals of the
cup repand ; filaments shorter, style longer, than

AM ARY LLIDACEjE. 199

the limb. Found near Chillo in the province of
Quito.

5. Breviflorum. — PI. 28. f. 7. lb. Specim. Herb. Lindl.

Herb. Hooker. Matthews, 657. Spec. nov. ex Tar-
ma Peru vise mense Jul. floribus 7-8 roseis, foliis
nondum natis. Scapo 10 unc. spatha 2-unc. pedunc.
inaequal. f-1^ unc. perianthio 1^ unc. roseo, stylo
vix limbum superante. Seven or eight rose-coloured
flowers lj inch long.

6. Coccineum. — PI. 28. f. 5. Panel*, coccin. Flor. Per.

3. 54. 2853. Ex Tarma Peruviae ; floribus coccineis,
perianthio biunc. graciliore; spatha graciliore ; fila-
mentis limbum superante, stylo longiore. Cocci-
neum has a slenderer spathe, fewer flowers, scarlet,
slenderer and longer than Breviflorum. In the dry
specimens it looks almost like a Phycella.

I am not without suspicion that Ruiz has made a great
error concerning his P. coccineum, and that the figure in the
Flor. Per. is a representation of his specimen labelled P.
rubrum, and his specimen of coccineum a Phycella, allied to
graciliflora. It does not seem likely that the artist through
inaccuracy should have so varied from the truth, as to give
precisely the usual shape of the flower of Stenomesson, of
which he had no knowledge, when the specimen departs from
it : wherefore I think he must have copied rubrum, of which
the plate is a fair representation. If my suspicion should
prove correct, rubrum will be to be erased. The confusion
will probably have in some degree extended to the text.

7. Rubrum. — PI. 28. f. 6. Specim. Ruiz. Herb. Lam-

bert. Floribus quinque, pedunculis fere aequal.
unc. perianthio rectiore, rubro, latiore, 1^ unc. fila-
mentis limbum superantibus, stylo longiore. This
plant has in the specimen (perhaps not in fact)
the flowers straighter and more erect than usual,
the flowers large and red.

8 ? Croceum. — PI. 28. f. 4. Specim. Dombey Herb.
Soc. Linn. Pane, croceum. Redoute lil. 187. Si
tides iconi adhibenda est, a flavo foliis praecipue et
stylo breviore distinguenda ; pessime depicta, et
flavi forsan var. certe non coccineum neque auran-
tiacum. Dombey s specimen is certainly the same
as Redoute’s plant, and it looks different from flavum.

200

AMAKYLLIDACEi'E.

The leaf is described as channelled, and having the
margins recurved like flavuin, but the engraving-
gives leaves quite foreign to the genus, and perhaps
taken by mistake from some other plant, as the
plant had no leaves at the time of dowering. Mr.
Ker was misled to refer this plant to Coccineum,
which differs in form as well as colour, by the
faulty engraving in the Flor. Per. The difference
will be seen at once on comparing the outline of
Dombey s specimen with that of Ruiz’s coccineum.
See PI. 28. f. 4 and 5.

This genus was named by me from the singular slender-
ness of the middle part of the tube, which is wider below and
bell-shaped above. Mr. Ker afterwards altered the name to
Chrysiphiala, likening it to an hour-glass, from the same
feature of constriction in the middle of the tube, and he pro-
posed to add it to the genus Leperiza and Carpodetes. Whe-
ther those two genera ought or ought not to merge in it, the
name Stenomesson having the priority must be retained, nor
could there be any reason for changing it for one founded on
the same feature. The proposed addition to the genus is
equally objectionable, for the affinity of Carpodetes, as far as
we know, is to Coburghia, and Leperiza seems to approach
nearer to Urceolina, but neither of them can possibly belong
to this genus. I have given the outline of the capsule of
Stenomesson croceum from Dombey’s specimen, which agrees
with Ruiz’s representations. It is very broad at bottom and
tapering to a point : that of Carpodetes very broad at top and
constricted in the middle. Stenomesson likes a sandy soil,
shade, and plenty of moisture in summer, complete rest in
winter. It flowers before the leaves rise. The reader must
be cautioned, that Mr. Ker’s descriptions of plants from the
Flor. Per. in the Journal of Science and Arts, cannot be per-
fectly relied upon, not from any error of his, but because
they have been made indiscriminately, not from the text of
Ruiz only, which may generally be trusted, but from the
plates also, in which the grossest inaccuracies have since been
detected.

51. Eucrosia. — Bulb round ; leaves wide, petiolated ; scape
tapering ; umbel 4- (more?) flowered, pedunculated ;
germen erect ; ovules oblong, heaped in two rows
alternating, attached at the inner angle of the cell ;

Flah.19

3.^. (/rmmwTMrruanaS.

terbgrfc.del.

AMARYLLIDACEX.

201

tube oblique, abbreviated underneath ; limb com-
pressed, recurved ; segments alternately alike ; cup
declined, abbreviated and rostrate above, shovel-formed
and prolonged below ; filaments long, recurved ; an-
thers attached at one-third from the top, pendulous ;
style at first sloped down, afterwards recurved ; stigma
obtuse, dilated, downy. Fruit not seen, but capsule
evidently ovate, 3-furrowed.

1. Bicolor. — Bot. Reg. 207. Bot. Mag. 51. 2490. I
have given an ample and careful account of this
plant in the Bot. Mag. to which the reader is re-
ferred. The figure in the Reg. is far from correct ;
and on comparison of the figure in the Mag. with
the drawing from which it was made, it appears that
the colourer has not made it near bright enough ;
the colour is nearer that of vermilion, the stripe in
the buds dark green unmixed with yellow, the cup
pale vermilion. The yellow colour of the flowers
in the Magazine is quite incorrect. I know not on
what authority Mr. Sweet stated it to be a native of
Cape Horn ; had it come from so cold a situation,
Mr. Lee would scarcely have lost all his bulbs by
leaving them in a cold frame in winter. I only
know that it came from South America. 1 have
usually kept it in the stove, or a very dry part of
the greenhouse without water in the winter ; in
summer it is thirsty and requires shade, having
much the same habit as Urceolina. It thrives in a
pretty strong alluvial soil.

52. Elisena. — Bulb roundish ; leaves linear, lorate, erect ;
scape few-flowered ; germen oblong ; tube short, cylin-
drical ; limb reflex ; segments linear, (4 recurved, 2
declined ?) cup declined, cylindrical, with a repand
recurved margin ; filaments and style filiform, declined,
recurved ; alternate filaments equal ; anthers incum-
bent. Fruit unknown. Scape presumed to be solid.

1. Ringens. — Pancratium ringens. Flor. Peruv. 283.
Liriope ringens. Herb. App. scape 5-flowered ;
peduncles very short ; tube short, green ; tube
greenish, f of an inch long ; filaments about equal
to the limb, style longer. In gardens in Peru,
Precise habitation not known.

202

AMARYLLIDACE^E.

This is a singular plant, if any reliance can be placed on
the engraving. The declined cup marks an affinity to Eu-
crosia, the only other genus in which it is found.

53. Pancratium. — Bulb ovate or roundish ; perianth with a
cylindrical tube and patent limb ; filaments stiff and
conniving; anthers short, suberectly incumbent. A.
flowers subsessile, with persistent leaves. B. pedun-
culated, with deciduous leaves. C. sessile or subses-
sile, with deciduous leaves. N.B. The colour of the
flowers in this genus is invariably white with a tinge
of green on the outside, except in Illyricum which
has no green.

A. Species subsessiles, foliis persistentibus.

1. Maritimum. — Bot. Reg. 5. 161. Red. lil. 8. Flor.

Graec. 4. t. 309. Semina mihi, t. 34. f. 21, 22. an-
thera, 42. f. 7. Limb with segments attached to
the cup, at about one-third from their base ; cup
campanulate with two teeth between every pair of
stamens ; filaments free about one-third of their
length, decurrent in the cup ; flowers fugacious.

2. Carolinianum. — Bot. Reg. 11. 927. foliis perperani

viridibus. P. maritimum, Pursh. Anthera, PI.
42. f. 8.

B. Pedunculate, foliis deciduis.

3. Canariense.— Bot. Reg. 2. 174.

4. Illyricum. — Bot. Mag. 19. 718. Limbo rigidulo, co-

rona brevi patenter infundibuliformi, interdum fissa,
antheris erecto-incumbentibus infra medium affixis,
stylo superne attenuato, seminibus rotundato-oblon-
gis, raphe albicante. Genus Halmyra ?

C. Sessiles aut subsessiles, foliis deciduis. Genus Tiaranthus ?

5. Malabathricum. — Rheede, Hort. Mai. 11. 79. t. 40.

Catulli-pola, floribus 7-8, (seminibus carnosis?)

6. Verecundum. — Bot. Reg. 5. 413. Hort. Kew. Biflo-

rum et Triflorum, Roxburgh.

7. Zeylanicum. — Bot. Mag. 52. 2538. Bot. Reg. 6. 479.

8. Cambayense. — PI. 42. f. 1. Specim. ex Cambay.

Hove. Herb. Banks. O — 1.

9. Longiflorum. — PI. 42. f. 2. Specim. Roxburgh. Herb.

Banks.

10. Maximum. — Forsksel. Flor. iEgypt. Arab. 72.

AM ARYLLIDACEiE.

203

The name Pancratium indicates that the bulb was sup-
posed by the Greeks to be of universal efficacy in medicinal
application. The Pancratium of Clusius appears to be Scilla
maritima, and it is very probable that Pliny intended to de-
signate the same plant ; this however is certain that the name
belongs to a maritime European bulb ; and amongst the
plants to which it has been attached by later botanists, P.
maritimum, the seaside European species, must be considered
as the type of the genus. It is evident that the American
plants which have been confounded with it, cannot be per-
mitted to usurp the name. Whether or not the three sec-
tions, into which I have divided the genus, are several genera,
I cannot decide without further knowledge of the plants, and
some better assurance than the assertion of Rheede, who is
not always correct, concerning the fruit of the Asiatic species.
I have found the scape of maritimum and Illyricum to be
solid, that of Illyricum being a little hollow near the summit.
The scape of the other species has not been examined, but I
have seen no plant of this order with a staminiferous cup and
hollow scape. The scape of Illyricum in this respect ap-
proximates to the Narcissese, in which the fistular cavity is
more or less filled up in different species.

1. Maritimum has acute glaucous leaves of a hard tex-
ture, and sheathing at their base. The representation of the
plant in the Bot. Reg. is very bad, and the cup quite out of
drawing. Even Clusius’s rough figure under the name of
Hemerocallis Valentina is better. The plant is a native of
the sea sand of the Mediterranean. I have two bulbs that
were dug up by my lamented brother the Earl of Carnarvon
on the coast of Ischia, and he told me that the sand where it
grew was so hot when the sun shone powerfully, that he
could not bear to keep his hand upon it. This circumstance
marks the difficulty of cultivating it, as it enjoys an air at
some times cool and temperate, a burning sand to the bulb
at others, and perhaps the fibres may reach below the high
water mark. I believe the best way of treating it is that re-
commended long ago by Hill ; to set it out doors late in the
spring, and at the latter part of the summer to bring it into
the stove to flower, removing it afterwards into the green-
house. When placed on a hot flue it requires frequent
watering, and the leaves are soon scorched if it is neglected.

2. Carolinianum. — I am rather inclined to consider this
plant a variety of Maritimum, that is to say the distinguishing

204

AMARYLLIDACEA2.

features, as far as I know, are not very considerable, and
their permanency may be doubted ; but it is not absolutely
identical with the Mediterranean plant as Pursli supposed.
I received from Dr. Carey bulbs which had b£en long culti-
vated at Calcutta under the name of Maritimum, and flow-
ered there freely, concerning Avhich he wished my opinion,
as he found the European Maritimum, which he had received
from me, different in foliage, and very impatient of the East
Indian climate, where it did not flower and was with difficulty
preserved. His bulbs were undoubtedly the Maritimum of
Pursh. With me they have flowered once, and increase by
offsets rapidly. The principal difference is that the leaves
are much less acute, and wider (I have measured a vigorous
leaf seven-eighths of an inch wide), and they do not vaginate
near so high as the European plant, indeed but little above
ground. Their texture is also smoother and softer to the
touch, and the glaucous hue not quite so intense, and more
easily washed oft’ by heavy rains. It is however quite a
mistake to suppose that the leaves are not glaucous. The
figure in the Bot. Reg. 927. has the leaf very incorrect in
that respect. I knowr that the plant from which it was made,
belonging to my lamented brother, Mr. George Herbert, had
glaucous foliage, and if the leaf appeared to the artist of the
colour represented in the engraving, it must have been quite
discoloured by being packed wet. The figure of the Caroli-
nian Pancratium, in Catesby’s work, is unquestionably a very
bad representation of Hymenocallis rotata, with bright green
leaves, and it is stated by him to grow in a bog, as H. rotata
does, near Pallachacula an Indian town in Georgia, which is
quite contrary to the habit of the genus Pancratium, of which
the fibres rot if too much watered. I have not had an oppor-
tunity of comparing the living flowers of Maritimum and
Carolinianum, but I can see little to distinguish them ; and
in dry specimens I observe some variation amongst indivi-
duals. I have given a representation of the seed of P. mari-
timum, which was ripened several years ago and sent to me
by the kindness of the Rev. Mr. Hutton. It exactly accords
with the figure of the seed given by Clusius, which is very
correct. The seeds imbricate, lying in two horizontal rows ,
they have a thin brittle black shell, a broad back, a wedge-
like edge, and a margin a little inclined to be foliaceous. —
These two species do not lose their leaves in the winter, which
is an important distinction. The two bulbs from Ischia have

Pla,te>JO.

AMARYLLIDACEiE.

205

been several years at Spoffortli without producing an offset,
but the broad-leaved or Carolina plants increase rapidly. I
cannot perceive the difference of anther mentioned by Mr.
Ker, and indeed such a difference, if strongly marked, would
be scarcely consistent with generic identity. The anthers of
both are what I call suberect, that is neither incumbent nor
quite erect. In dried specimens the posture becomes much
disarranged. The attachment is nearer the middle than in
Hymenocallis.

3. Canariense. — One bulb of this species was brought to
Europe by Dr. Schmidt, and found its way into the collection
of Mr. Griffin, where it flowered and ripened seeds, which,
as Mr. Griffin told me, had some tendency to a foliaceous
margin. I know not into whose hands the bulb passed after
Mr. Griffin’s death, nor have I heard of any other having
been imported. It lost its leaves in the winter like Illyricum,
with which it agrees in being pedunculated.

4. Illyricum. — This bulb, which is often sold erroneously
as imported from Holland under the name of Maritimum, is
perfectly hardy, flowering and ripening seed freely in our
gardens : but the seedlings are of slow progress, and do not
flower till they are many years old. It likes a rich and even
manured soil. Its seeds are round-oblong, covered with a
pretty hard black shell, and have an elevated white raphe,
while those of Maritimum are more foliaceous, wedge-shaped
with a round back, and the hilum a white speck without any
elevated raphe. I communicated the seeds of both to a dis-
tinguished botanist, without mentioning to what plants they
belonged, and asked whether he should suppose they were of
one genus or not, and the answer was he should rather sup-
pose not. If the seed of Canariense should be found to agree
with that of Illyricum, considering that both are peduncu-
lated and deciduous, I should not doubt the correctness of
Mr. Salisbury’s separation of Illyricum, under the name
Halmyra Illyrica. The flower of Illyricum is distinguishable
from all the rest in having the inside of the cup and the base
of the limb yellow, the tube yellowish, the cup wide funnel-
shaped, almost patent, and so short that it is little more than
a connection of winged bases to the filaments, and 1 have
found it occasionally split to the base between the filaments
in one or two places, which makes it approximate a little to
Vagaria parviflora, concerning which our knowledge is im-
perfect. On the whole, in our present ignorance of the fruit

206

AMARYLLIDACEiE.

of the Asiatic section, and insufficient acquaintance with the
fruit and stamens of Canariense, I wish to postpone deciding
whether the seminal difference between Illyricum and Mari-
timum is absolute or not, especially as I have no opportunity
of re-examining the style and stamens of Maritimum. I am
inclined however to think that it is, and that there are suffi-
cient symptomatic features to indicate the diversity.

5. Malabathricum. — This plant is only known from
Rheede’s Hort. Malab. where it is called by the native name
Catulli-pola. Mr. Ker quotes it as synonymous to Verecun-
dum, which he calls 3-4-flowered. I have never known or
heard that the flowers of that species exceed three, and it
was only known to Dr. Roxburgh as bearing either two or
three, to which he particularly limited it by his name tri-
florum ; but Rheede described his plant as having from seven
to eight. Rheede says that its fructification is similar to that
of a Crinum, and he gives a very correct representation of
a Crinum seed ; and adds that the seed of Catulli-pola is
whitish, with a whitish watery flesh, and a watery taste. If
this account be true, and without some grounds I cannot
reject such a positive assertion, though he is often incorrect,
the Asiatic species must be of a different genus, which I
should call, as proposed many years ago, Tiaranthus, follow-
ing the idea of Mr. Salisbury when he changed the name
Zeylanicum without cause to Tiarreflorum. 1 have vainly
attempted to obtain seed of any Asiatic species. Dr. Wallich
should try to procure this Catulli-pola from the Malabar
coast. It is stated by Rheede to grow in sandy places, but
he does not say on the sea-side. Whether it should be called
a polyanthous variety of Verecundum, or a distinct species,
I cannot judge without seeing the plant, and knowing how it
agrees in other respects. Very little reliance can be placed
on the figures of Rheede. He did not always consider the
difference between a peduncle and a tube.

6. Verecundum is abundant in the meadows near the
river in the vicinity of Calcutta, but Dr. Carey could never
send me a seed of it, and answered that he had never seen
one. Whether the plant only perfects its seeds in a cooler
situation, from whence they may be brought down by the
floods, or my worthy friend did not inquire beyond the limits
of his own garden, I know not, but, though I wrote several
times on the subject, I could obtain no account of the seed
of any oriental species. This plant is very difficult to culti-

amaryllidacea*.

207

vate here, though Dr. Carey said it was free and abundant
at Calcutta. In the hothouse its leaves flag whenever the
sun shines upon them, and the bulbs are apt to rot during
the season of rest when kept dry in the stove. I have lost
the species, and I imagine that it required an alluvial and
stronger soil than I gave it, and that the bulbs should have
been removed into a cooler place in the winter. They come
over very sound, when sent from Calcutta. The scape is
2-3-flowered ; the peduncles not half an inch long ; tube 2
inches long according to Solander Hort. Kew., 3^ in my
plants from Calcutta figured in the Bot. Reg., limb in both
shorter than the tube ; cup with six interstamineous teeth ;
leaves bright green, acute, f of an inch wide.

7. Zeylanicum. — Scape one-flowered ; germen sessile ;
tube about 1|- inch ; limb longer than the tube, the lower
part adhering to the cup ; filaments exceeding the teeth of
the cup by about the length of the cup and teeth ; leaves
bright green, narrow-lanceolate, acute, narrower than those
of Verecundum. Unless the flower varies more than I ima-
gine it does, the cup and stamens are too long in the figure
in the Bot. Reg. It is less difficult to cultivate than Vere-
cundum, but very liable to perish here, and scarce in the
gardens of Calcutta.

8. Cambayense. — Specimina Herb. Banks. O — 1. ex
montibus Cambay Ind. Orient, prope Guzzerat, in planitie
arenosa. A. P. Hove. Bulbus subrotundus collo cylindrico,
folia linearia obtusa ^ unciae latae, scapus ultra 4-uncialis,
flos unicus sessilis, tubus gracilis 2f unc. limbus 2^ unc.
stylus || unc. limbo brevior. There are many specimens of
this plant in the Banks, herb, gathered by Hove on a sandy
plain on the Cambay hills in the East Indies, near Guzze-
rat. The specimens vary very little, and they are strongly
distinguished from any species that has been described by
their narrow linear obtuse leaves. The tube is much
slenderer than that of Longiflorum. The entry in the
Banks, herb, in Dryander’s writing, is as follows: —
“ Pane, longiflorum fl. sessili solitario nectario duas tertias
partes limbi excedente, laciniis limbi patentibus, limbo
tubo longissimo breviore. Linn. fil. — P. maximum Forsk. ?
— I believe this to be the same as P. Zeylanicum. If the
different specimens from Koenig and Hove, together with
Hove’s drawing, are compared with the figure in Com-
melin, it will be found to vary very much as to the length

208

am-aryllidace;e.

of the tube, and the nectary is 12-dentate in Hove's fig.
Feb. 90.” — Hove’s drawing, No. 64, is marked “ Gool
Sobool, sandy plains, Cambay.” Bad as Hove’s drawing is,
the tube therein is 2f of an inch long, and that of the spe-
cimen of Zeylanicum in the Banks, herb, only 1|. Hove’s
drawings are so bad, that he had probably never handled
a pencil before ; and in this drawing, where the leaves
are a great deal too broad and pointed, he has even
omitted the style. It is strange that Dryander should have
formed an opinion on such a performance, when he had
several good specimens before him. The plant has no re-
semblance to Forskael’s P. maximum, to which he inju-
diciously refers it, as well as to longiflorum ; and the varia-
tion of the tube of Zeylanicum from 1|- to 1| in length, to
which he alludes, is of no importance. In Zeylanicum the
limb is usually from two to three times the length of the
tube, in all the specimens of Cambayense it is shorter than
the tube, but the leaves of the two plants are quite different,
the margins in this being parallel, differing in that respect
from all the other Asiatic species.

9. Longiflorum. — Specimina Herb. Banks. Leaves acute,
lanceolate, half an inch wide, scape very short (about an
inch and a quarter), one-flowered, germen sessile, tube
inches long (six according to Roxburgh), limb about three,
style nearly equalling the limb. Native of Molucca. It has
long been lost at Calcutta ; the bulbs have never, I believe,
been brought to Europe. It is very different from Cam-
bayense.

10. Maximum. — This plant was described by Forskael,
who saw only one specimen, which had been gathered by
his companion, Dr. Niebuhr, near Taaes, in Arabia, a town
about fourteen German miles west by north of Mockha.
Forskael did not see its leaves. It was a solitary flower,
and he describes severally both the tube and the cup to
be a foot long, the limb to be longer than the stamens and
patent ; but, he adds, not reflex as in Zeylanicum. It is
however evident that he must have considered the cup to be
a part and prolongation of the tube, and only meant that,
taken together, the tube and cup measured a foot in length;
for if the cup alone had been a foot long, the stamens, being
longer than the cup, and the limb, as he states, exceeding
them, the expansion of the flower would have measured at
least a yard in width, which is quite incredible. He speaks

F!ako.2>l

*

AMARYLLIDACE7E.

209

of the nectary as enormous, considering the tube to be part
thereof ; but he does not say that the expansion of the limb
was very great. The main part of the length must there-
fore be occupied by the tube, perhaps nine inches long and
the cup three, which would very well entitle the flower to
the name Maximum.

A. Pancratium, by name iEgyptiacum, is mentioned in
an old catalogue of the plants at Mr. Loddiges’ nursery. I
learn from him that the plant so designated was brought by
a gentleman from Egypt, and appeared to him to be a Pan-
cratium, but perished without having flowered, and no me-
morandum was preserved concerning it. It cannot there-
fore be ascertained what it was, and the name must be
expunged.

54. Hymenocallis. — Bulb roundish ; tube cylindrical, slen-
derer than the germen ; perianth equal, patent, flac-
cid ; crown more or less patent (funnel-shaped or
rotate) ; filaments equal ; anthers long, pendulous,
attached at one-third of their length from the base,
tipping after expansion base-upwards ; pollen rough ;
style long, flaccid, declined; stigma minutely fimbri-
ated, roundish ; ovules oblong, 2-8 in a cell, erect;
seeds large, oblong, green. Occidental plants with
leaves usually persistent, the scape flattened 2- angular,
curved downward in seeding , flowers 1?-21, white, very
fragrant, pollen orange or deep yellow ; germen sessile,
except II. speciosa.

A. Petiolata. Leaves petiolate.

1. Speciosa. — Breviter pedunculata.

Var. 1. Longe petiolata. — P. speciosum. Cup dentate.
Bot. Mag. 35. 1453. Red. lil. 412. P. amcenum.
Lodd. B. C. 286.

Var. 2. Humilis, petiolo brevi. — Petiole short, leaves
more recumbent.

Var. 3. Augustifolia, vix petiolata. — Leaves narrower,
scarcely petiolated.

Var. 4. Fragrans, tubo breviore. — Red. lil. 413. Tube
4 inches long, limb 4|, cup not dentate, but in-
dented between the filaments.

This beautiful species is distinguishable by the short
peduncle from all others that are known to us ; and, agree-

p

210

AMAR YLLIDACEiE.

ing’ with Amoena and Guianensis in having petiolated broad
leaves, it may be always recognized by their greater thick-
ness and consistence, by a more robust tube, and much more
conspicuous flowers. It differs from all the other species in
sending up its young leaves not successively, but at periods
(perhaps twice a year) simultaneously, the flower-stem just
preceding them. P. fragrans, Redoute, is certainly a va-
riety of Speciosa with the foliage of var. 2, but a shorter
tube ; it is more robust, and has a longer tube than Amoena.
This species is the most beautiful, and is cultivated without
any particular care in the stove. The petiolated sorts are
all natives ol very hot countries, and it may be that the
habit becomes obsolete in a more temperate climate ; for I
find in the plants of this species and of Amoena raised from
seed produced in my stove, a disposition to depart from the
decidedly petiolated habit of the parent, and produce leaves
of a more lorate form.

2. Guianensis. — Folia petiolata, mollia, successiva ;
tubus limbo longior. Leaves softer than Speciosa,
successive ; tube 5-8 inches long, longer than the
limb.

Var. 1. Prmceps. — Bot. Reg. 4. 265. Tube 6 inches
or more; limb 4-5.

Var. 2. Undulata. — Kunth. 1. 222. Tube 5 inches,
cup toothed in the intervals.

Var. 3. Tubiflora. — Salisbury. Hort. Soc. Tr. 1. 341.
Tube 8 inches, cup not toothed.

This species occurs in a variety of forms along the coast
of Honduras and Guiana, always distinguishable by broad
petiolated foliage of a thinner texture than speciosa, much
slenderer and less conspicuous flowers, with a tube reaching
from 5 to 8 inches in length. The undulation of the limb,
from which var. 2. has been named, is of little importance in
this genus ; I have seen it vary in the same umbel. The
dentation of the cup is not less variable. Var. 1. has long
petioles, var. 3. much shorter, but endless shades of variation
will be found between them. They are plants of very deli-
cate constitution, and if planted in light loam in a dry stove
they soon perish. I believe them to be natives of alluvial
soil and a half swampy situation. A plant of var. 3. which
I have lately placed in water, seems not to be offended by it,
but perhaps will not thrive there permanently.

AMARYLLIDACEiE.

211

3. Aincena. — Folia petiolata, mollia, successiva ; tubus
limbo brevior. Leaves softer than speciosa, suc-
cessive, tube shorter than the limb.

Var. 1. Princeps. — P. amoenum. Bot. Mag. 36. 1467.
P. fragrans. Andr. Bot. Rep. 556. Lodd. Bot. Cab.
834. Tube 2 inches, limb 3, filaments exceeding
the tube 2^, style 3j, cup f .

Var. 2. Ovata. — P. ovatum. Bot. Reg. 1. 43. Broader
leaves.

Var. 3. Lorata, hortensis ; foliis lane. lor. suberectis
longioribus.

This species is mainly distinguishable from Guianensis
by the shortness of its tube, which does not equal the limb,
while that of Guianensis is preposterously long. It is tender,
but not so difficult to manage as Guianensis. I had from
Mr. Griffin the very ovata figured in the Register, but
there was very little to distinguish it from Amoena, but rather
broader and less erect foliage, and the seedlings raised from
the two plants could not be distinguished. Var. 3. was
raised at Spoff'orth from seed of var. 1. and has long, erect,
lorate leaves. The circumstance is very singular. The
plant had been deprived of its pollen for an experiment, and
touched with pollen of another and hardy genus ; but some
particle of its own pollen must have escaped, as frequently
occurs, in the operation, no other Hymenocallis having been
in flower at the time. Seeds were produced, very imperfect
in their appearance, small, and discoloured. Four seedlings
were raised, all exactly alike, and having the appearance of
a different species from the parent. I entertained doubts at
first whether a bigeneric mule had been obtained, but the
plants had not inherited any hardiness from the supposed
male parent, and when they flowered the inflorescence dif-
fered in nothing from that of Amoena. Does this remark-
able occurrence exemplify the manner in which permanent
local varieties are produced by peculiarities of temperature,
soil, or situation, or by an insufficient quantity of the pollen?
I am confident that no other species was in flower at the
time ; but, supposing me to be mistaken in that point, if the
seedlings had been produced by the pollen of H. expansa, to
which they approximate in foliage, the form of the flowers
must have been affected as well as that of the leaf, which is

p 2

212

AMARYLLIDACEiE.

not the case. These seedlings are much disposed to increase
by offsets.

4. Ovalifolia. — Lodd. B. C. 510. This is a very small

and delicate species, quite distinct from the variety
of Amcenum, to which the name ovatum had been
previously given. It has small oval recumbent
leaves. It has been lost at Spofforth and Hackney;
I know not whether it is preserved at Liverpool,
where it was first introduced by Mr. Shepherd. I
never saw it in flower, but do not doubt the cor-
rectness of the figure cited. The flowers are very
slender.

B. Loratce ; kumifusce vel arcuatce ; Meridionales. Leaves
lorate, spread on the ground or arcuate ; S. hemisphere.

5. Caribgea. — Foliis late loratis non arcuatis ; tubo

crasso subtriunciali laciniis breviore.

Var. 1. Declinata, vel Princeps. — P. Caribgeum. Linn.
Sp. pi. ed. 2. 1. 418. Bot. Mag. 21. 826. P. decli-
natum. Jacq. Am. 51. t. 102. Hort. Vind. 3. 11.
t. 10. Red. Lil. 414. Lodd. B. C. 558. Spec. Herb.
Banks. Linn. fil. Germen and tube from 2fths to
3|-th long; style longer than the filaments, shorter
than or equal to the limb, which is from 4 to 4\
long; leaves broad, very acute, of a very lively
green.

Var. 2. Cinerascens ; foliis obtusioribus, pallide cine-
reo-viridibus, tubo vix 2-unc. limbo 5^-5§ canali-
culato, petalis latioribus, basi undulatis, corona
lj unc., filamentis 3§, stylo 4f, tubum superante.
Cinerascens is nowhere figured or described ; it is
closely akin to the preceding, but has the limb and
style much more prolonged ; leaves 19-20, 20 inches
long, 2f to 3 inches wide, more obtuse, and of a
pale, dull, cinereous green.

Var. 3. Patens. — P. patens. Red. lil. 380. et nota ad
414. Leaves H-3 feet long, 2 inches wide, ger-
men and tube 3 inches long, limb near 4.

I cannot find that there exists in any collection or her-
barium such a plant as Patens of Redoute distinct from his
Declinatum, which is H. Caribaea, var. 1. or princeps. It
seems to be distinguished from that plant as a variety by

I

AMARYLLIDACEA2.

213

narrower longer leaves, less proportional length of filaments,
less curve of the limb and undulation of the base of the seg-
ments, the cup not toothed, the filaments a little shorter than
the style. I cannot discover the plant anywhere, and its
description rests entirely on the confused account in Redoute.
It must be considered a narrow-leaved variety of Caribaea.
The name Declinata lapses, being posterior to Caribaea, or it
may remain to designate var. 1 . from the others. I suppose
it was given in allusion to the declension of the scape in
seeding, which is common to the whole genus.

6. Expansa. — Bot. Mag. 44. 1941. Foliis Caribaea lon-

gioribus angustioribus erectioribus, scapo altiore,
tubo circiter 4-unciali ; Caribaeae, si mavis, varietas.
The figure of this plant is not a very good one, and
there is no other. It is a large vigorous species,
allied to the cinerascent variety of Caribaea, with
leaves longer, narrower, and more erect, but less
arched than those of pedalis and Caymanensis.
Leaves from 2to2 f.6, 2inches to 2f wide, channelled,
attenuated below. In a flourishing state it has
produced as many as 2 1 flowers on an umbel ; the
flowers vary very much in different seasons, accord-
ing to the temperature, viz. 1816, tube 4^; 1817,
Aug. 4|- ; Nov. 4, 1820, 4^ ; 1836, 3^ ; limb, 1816,
5f ; 1817, Aug. 5^; Nov. 6, 1836, 5 ; cup, 1817,
Aug. lj; Nov. 1^, 1836. Style exceeding the

filaments 2 inches, about equal to the limb. The
original bulb was purchased at Mr. Evans’s sale,
and I distributed many seedlings from it. I have
since had from Lima varieties closely allied to it,
but differing in foliage, and they will perhaps be
multiplied from different localities, till the specific
distinction between it and Caribaea will be found
to vanish ; and in that case it must range as a va-
riety of Caribaea.

7. Tenuiflora. — (P. litorale ? Kunth, 1. 222.?) Foliis

recumbentibus, loratis, 27-uncialibus, 2| latis ; um-
bella 11 -flora, tubo tenui prope 5 unciali, limbo
tenui 4^, corona f unc. This plant was purchased
by me at Mr. Evans’s sale, 1814 ; its locality is not
known. Its leaves are recumbent, above two feet
long, 2^ inches wide at the utmost, flowers 1 1, tube
very slender, 4! long, limb slender 4^, cup |ths

214

AMARYLLIDACEiE.

wide and long. Rather a delicate plant in the stove.

1 suspect that it will be found on the coast between
Carthagena and Portobello.

8. Angusta.— P.angustum. Bot. Reg. 3. 221. This plant

has broader and more arcuate foliage than Tenui-
flora, to which it approximates in the slenderness
of its flowers, but the tube of its flower is under
two inches. It requires a high temperature, and
was knowm in our nurseries by the name of P. lito-
rale, before it was figured under the name Angusta.
Mr. Kennedy of the Hammersmith nursery, where
it was first introduced, asserted that it was the South
American sea-side plant. I entertain very little
doubt that either this plant or Tenuiflora is the
true litoralis that grows on the island Tierra Bomba,
near Carthagena, and on the. coast from thence to
Portobello ; and that Jacquin, by some confusion
of labels, has affixed the name erroneously to a
Mexican plant. See 12. Adnata, var. 1.

9. Pedalis. — P. pedale. Lodd. B. C. 809. Bot. Reg. 19.

1641. fig. non laudanda, tubo nimis gracili, nisi
varietas sit minus speciosa et mihi ignota. A large
species with green arcuate leaves, about 2 feet long,

2 inches wide, acute, attenuated below ; robust
flowers a foot long, limb about 5, tube 7 inches.
Introduced by Mr. Shepherd of the Liverpool gar-
den. Locality not ascertained.

10. Caymanensis. — P. patens. Lindley Hort. Soc. Tr. 6.
87. non Redoute. Folia saturate viridia, nitentia,
tripedalia, 2^ unc. lata, canaliculata, arcuata, acuta;
tubus limbo longior, Caribsese tubo gracilior. Leaves
dark glossy green, about 3 feet long, 2^ inches
wide, channelled, arcuate, acute. This plant does
not agree with Patens of Redoute, which has the
tube shorter than the limb. It is a native of the
island called Grand Cayman, and has foliage more
like that of pedalis than of any other species, the
tube longer than the limb, and not so thick as in
the Caribsean family, which have all the tube
shorter than the limb.

C. Loratce, suberectce ; Septentrionales. Leaves lorate,
suberect; N. Hemisphere.

AMARYLLIDACEjE.

215

11. Crassifolia. — Ex portu St. Mary, Floridae Orien-
talis. lat. circit. 29. Foliis crassis, suberectis,
loratis, obtusis, canaliculatis, subtripedalibus, 2-unc.
latis, viridibus, perianthii corona fere uti in Cari-
baea, tubo longiore ; bulbo e majoribus. Three
bulbs of this species were brought from Florida by
Fraser, about twenty years ago ; two of them were
purchased by myself, the other, as Fraser informed
me, was bespoken by Lord Mountnorris ; whether
it is living, 1 know not. One of mine was lost by
neglect a few years ago ; the other flowered the
first summer, but never since, nor has it yielded
any increase. The inflorescence was sent to Dr.
Sims, who, by some mistake, mislaid it, and I had
made no memorandum concerning it. Its long
erect thick leaves, scarcely attenuated, are very
remarkable. It probably requires a cooler tempe-
rature than I have ventured to give it.

12. Adnata. — Laciniarum basi coronae adnata.

Var. 1. Princeps, vel Litoralis? P. litorale Jacq. Am.
99. t. 179. Hort. Vind. 3. 41. t. 75. Salisbury
Linn. Soc. tr. 2. 74. t. 13. P. distychum Bot.
Mag. 44. 1879. quoad figuram et Leei plantain.
Tubo laciniis longiore. Olim P. Mexicanum hor-
tulanorum.

Var. 2. Driandrina. — P. litorale /3. Bot. Mag. 21. 825.
P. Dryandri Ker. J. Sc. and A. Tubo 4-unciali
laciniis parum longiore.

Var. 3. Distycha. — P. distychum Bot. Mag. 44. 1879.
quoad Herberti plantam, non quoad figuram. Tubo
et laciniis 4^-uncialibus, foliis parum latioribus
nervosioribus.

Var. 4. Acutifolia.— Bot. Mag. 53. 2621. P. Mexicanum
Bot. Reg. 11. 940. P. acutifolium. Sweet H. Brit.
Tubo 3^ unciali laciniis unciam breviore, foliis
angustioribus minus erectis, stylo perianthium
subsequante, filamentis unciam longiore, corona
1% unciali margine denticulato. Patria Mexico.

Var. 5. Staplesiana. — Tubo 3J unciali laciniis unciam
longiore, stylo perianthium aequante filamentis
paullulurn longiore, corona f unc. dentibus inagnis

216

AMARYLLIDACEjE.

staminiferis ; bulbo quam maxime prolifero, foliis
erectis nitentibus bipedalibus, vix unciam latis,
canaliculato-costatis, inferne attenuatis.

These are all distinguished from the rest of the genus by
the adherence of the lower part of the limb to the cup ; they
are much hardier than the rest of the genus and are decidedly
aquatic or swamp plants. Jacquin has stated positively
that var. 1. grows on the island of Tierra Bomba near Car-
thagena, lat. 11. The identity of his litoralis, as well as that
of Mr. Salisbury, with the variety 1. of this species is un-
questionable, but I consider it impossible that his statement
should be correct that it grows on the sea-level so near the
line, or that it can be the plant found by Humboldt between
that place and Portobello in the sea-sand. I purchased the
plant above 20 years ago at the Hammersmith nursery under
the name P. Mexicanum, and Mr. Kennedy asserted it to
be from Mexico, and Mr. Loddiges has lately imported many
bulbs which appear to be similar, direct from Mexico. It
is so hardy that an offset which I set against the front wall
of the stove about 14 years ago, grew vigorously there, and
although the snow lay upon it some weeks the first winter,
its leaves were not killed quite to the ground, and it grew
into a tuft with many offsets, and flowers most summers. I
lately submerged a pot of seedlings of this variety in a cool
cistern in the stove, and the plants immediately began to
grow rapidly, and young white fibres formed themselves
abundantly on the surface of the earth, and some of them are
now floating in the water. Amoena, speciosa, and angusta,
planted against the same wall of the stove close beside it, not
only perished in the winter, but could not thrive in the sum-
mer. I think therefore that I may venture to say that
Jacquin must have been deceived by some confusion of labels,
and that this plant, which he has called litoralis as growing
on the sea-shore in an ardent situation, cannot be the plant
that grows there ; and that if Angusta and Tenuiflora be not
the real inhabitants of that coast, the true litoralis has not
yet been introduced into Europe, and is not the plant he has
described and represented. If it can grow naturally in such
a burning sand, and yet thrive with increased vigour in
water in a cool situation, and flourish under snow, it must
have a constitution different from that of all other vege-
tables. I have had this plant with 16 flowers on a scape.
Var. 2. and 3. are only distinguished by a little difference of

lO . 7<S. Ua -ia/uata

Tla,U. 3 i

AMARYLLIDACE^E.

217

tube, which in both is shorter than in the first. Var. 4. has
the leaves narrower and less erect than the others. It was
sent to Sir Alex. Johnstone direct from Mexico, and is such
a decided aquatic, that I have a seedling plant of it now in
flower in a cistern, where it has been submerged from a few
weeks after it sprouted from the seed some years ago. It is
a hardy greenhouse plant, but a little warmth promotes its
flowering. Var. 5. was sent to Mr. Tate of Sloane Street,
Chelsea, by Mr. Staples, from Mexico. It is remarkable
from the multitude of offsets it produces, which rather im-
pede its flowering. It is the hardiest of all the varieties,
and having been placed about half a yard from the wall of
the stove in front of var. 1. it grows with such luxuriance as
almost to overpower the bulbs behind it, forming a very
thick and increasing tuft of leaves above two feet high.
Some bulbs of it, which were placed in 1835 in a border in
the middle of the garden, and covered with a few leaves in
the winter, have survived, but seem to want frequent water-
ing ; a pot the bulbs set on a step in the pond grows more
vigorously. It has flowered and ripened seeds out of doors
in front of the stove. I think Mr. Ker was right in stating
(Bot. Mag. 27. 1082.y that the P. Mexicanum of Linnaeus
was a weak specimen of var. 2. of this species, and the name
Mexicana, if at all preserved, must belong to some variety
of this species. I believe them all to grow in Mexico, and
as yet I know of no other species from thence. Litoralis
could not be taken as the chief name of the species without
absurdity and untruth, and it remains to be ascertained
whether it does really belong even to the first variety.

13. Rotata. — Foliis suberectis subspathulatis 9-20 unci-
alibus ^-1§ latis, nitidis, viridibus, scapo 2-4 floro,
corona magna patula.

Var. 1. Quadriflora. — P. rotatum. Bot. Mag. 21. 827.
Lodd. B. C. 19. P. Carolinianum Catesby Carol,
fig. pessiina.

Var. 2. Disciformis ; biflora, omni parte minor, corona
minus rotata. P. Mexicanum. Linn. Ker. J. Sc.
and A. cum ic. P. rotatum a. Bot. Mag. 27. 1082.
P. disciforme. Red. Lil. 155.

These two plants, the second of which Dryander erro-
neously supposed to have come from Mexico, where we have
no reason to believe it is known, are natives of bogs in the

218

AM ARYLLIDACEiE.

United States ; the latter in Florida, the former extending
northwards into Virginia. It is quite evident that the figure
of Dryander’s P. Mexicanum, published by Mr. Ker, is a
bad representation of this plant. It differs in nothing ma-
terial except the attenuated point of the leaf ; and when I
look to the incorrect form given to the style and stamens, I
can only regard that as another error in the representation.
They differ from the rest of the genus in some particulars.
Though in a stove or warm greenhouse they will preserve
their leaves through the winter, they are more disposed to
go to rest at that season, and I believe it is best to leave them
dry for a time. The cup is more expanded and green at its
base, the seed-vessel is stretched by the growth of the seeds,
and does not burst prematurely like the rest, of which the
growing seeds become exposed. The young rising leaves of
both in the stove or greenhouse perish if a drop of water
lodges among them, which makes the difficulty of cultivating
them. It has seemed to me lately that the same injury does
not take place when the pot stands in a pot of water as when
the treatment of the bulb is drier. The flower of var. 1.
approximates in form and colour to that of Choretis glauca:
it survived a year or two out of doors in front of the stove,
but perished without having flowered there. There is a
little difference in the form of the cup of var. 2. but they
agree in constitution, and every other respect, except size
and number of flowers.

Species hybrida. — Spofforthise, or Distycha-rotata. Raised
eighteen years ago at Spofforth from rotata var. 1. by ad-
nata var. 3. but has never perfected a flower. The scape
comes up always diseased, the flowers dead and discoloured,
or in a perishing state. I had despaired of ever seeing it
flower, and had suspected that it arose from the barrenness
of the mule ; but it has occurred to me that it perhaps re-
quires to stand in water at the time of flowering, and the
pot is now set in a cistern where the roots are thriving, and
I hope that it may be enabled to bring its flowers to per-
fection.

14. Quitoensis. — PI. 22. f. 4. Specim, Ruiz ex Quito.
Herb. Lambert. Folia 13-uncialia utrinque atte-
nuata lata, scapus 7-uncialis uniflorus, germen
sessile § unc. tubus 4|, laciniae 3f, corona biunci-
alis (2| lata?), filamenta circiter unc. coronam

AM ARY LLI DACE.®.

219

superantia, stylo breviora, stigma crassum. This
very remarkable plant is only known by Ruiz's
specimen in Mr. Lambert’s Herbarium. It is the
only one-flowered Hymenocallis known, and in
other respects it approximates to the Pancratium
of the Eastern Continent ; so that, from the mere
examination of the dry specimen, I do not feel
quite confident of its genus. It is barely possible,
though not probable, that the Asiatic form of
Pancratium may manifest itself on the heights of
Quito. The two-flowered Hym. rotata is the species
most nearly allied to it, though very inferior in the
size of its cup. I much lament the impossibility
of obtaining any knowledge of its fruit, and I wish
to call the attention of collectors to this plant, and
especially to its seeds. I am at a loss to imagine
how it could have been supposed in Germany that
this plant was allied to Urceolina pendula : it is
very remote from it indeed. — Leaves 13 inches,
attenuated at both ends, -|ths wide; scape 7 inches,
one-flowered ; germen sessile, f long ; tube 4f,
limb 3f , cup 2 inches long, perhaps 2^ wide ; fila-
ments about -|ths longer than the cup, shorter than
the style; stigma thick.

Species dubia.

15? Paludosa. — Specim. folii imperf. Tweedie 701.
Herb. Hooker. Ex paludibus profundis prope flu-
men Yazuguay, Rio Grande, Braz. merid. foliis
If unc. latis infra attenuatis (2h pedalibus?) scapo
2-4 pedali, floribus odoratis 15-20. Hymenocallis,
vel Crinum ? vel generis nondum noti. Prince
Maximilian mentions a white Amaryllis on the
banks of the Rio Grande, which is probably this
plant, which has a scape from 2 to 4 feet high, and
15 or 20 sweet flowers. No Crinum is known to
inhabit America with a scape bearing so many
flowers.

Choretis. — Filaments attached to the upper part of
the anther in a prominent callosity ; anthers recurved
at top and bottom. — Seeds short, oblong, i. e. inter-
mediate in form between those of Hymenocallis and
Ismene, shorter than the former, but not round, as in

220

AMARYLLIDACEiE.

the latter. Perianth of Hymenocallis, habit of Ismene.
Scape round (Qu. whether always?) which is two-
edged in Hymenocallis. Natives of Mexico and
Texas.

1. Glauca. PI. 35. f. 1. seed PL 41. f. 32. Bulbus
indusiis nigris ; folia erecta, glauca, plus minus
obtusa, 2^ unc. lata, sesquipedalia ; scapus teres,
14 uncialis, 3-(4?) florus ; spatha valvis angustis,
erectis, sejunctis, biuncialibus ; germen sessile,
loculis dispermis ; tubus viridis, sexuncialis vel
ultra; limbus circiter 3|, albus, patulus, sepalis
viridi-costatis; corona ultra-uncialis, patula, rotata,
alba, macula magna in medio viridi, dentibus senis
magnis reflexis, margine laciniato ; filamenta pa-
tulo-conniventia, subulata, 1^- uncialia ; antherae
apice et basi retroflexis ; semina glauca, nitida.
This beautiful plant has erect glaucous leaves, 2^
inches wide, obtuse (but not equally so in all indi-
viduals), about a foot and a half long ; the scape
round, with angles nearly obsolete, about 14 inches
high ; spatlie two inches, valves erect, narrow,
disjoined, and not covering the umbel ; flowers
three (or four ?) sessile ; tube above six inches,
green ; limb about 3f, white, ribbed with green ;
cup above an inch long, patent, rotate, with large
interstaminear teeth, and a jagged reflex margin;
filaments 1|- long, anthers recurved at both ends ;
cells two-seeded ; seeds large, glaucous, smooth. It
is a greenhouse plant, liking a very sandy soil, and
perfect rest in the autumn and winter, approaching
in habit to Ismene, but in the perianth to Hymeno-
callis rotata. Before I had seen the flower of this
plant I entertained no doubt that it would prove
to be an Ismene, and when the flower appeared, I
had much difficulty in satisfying myself whether
it should be considered generically distinct from
Hymenocallis or not. I have not been able to
learn that such a diversity of anther exists in any
well-constituted genus, and particularly the sepa-
ration of Habenaria from Orchis seems to be very
analogous. I have since been fully satisfied of the
diversity of this plant by the vegetation of the
seeds, which, though sown in May (the plant

FLcbfo.ZL.

AMARYLLIDACEjE.

221

having been forced, to satisfy my impatience), have
produced strong dormant bulbs under ground at
the bottom of the pot, in the manner of Ismene,
which will not vegetate till next year. Hymeno-
callis, like most other bulbs, forms leaf and bulb at
the same time. The genus is perhaps confined to
the Mexican dependencies, including Texas. The
seeds were obtained by cutting off the rind of the
germen and laying open the cells, and, though two
of the ovules had a slice cut off in the operation,
they did not suffer from the loss, and their growth
seemed facilitated by the removal of the rind, which
they must otherwise have burst. The scape of this
plant was bent downwards every night, and be-
came more erect again in the day time while the
flowers lasted.

2. Galvestonensis. — PI. 41. f. 34. 35. Specim. Herb.
Hooker, (absque foliis) Drummond, 412. ex Gal-
veston bay in prov. Texas. N. lat. 29-30. (f, 36.
represents the outline of the upper part of one of
the leaves of a bulb sent by Drummond from
Texas labelled genus like Crinum, supposed by
me to be this plant.) Scapus 9-11 uncialis, spatha
circiter biunciali valvis sejunctis, umbella 4-flora,
germine sessili, tubo vix 2± unciali, limbo 2^-
3-unc. stylo vix limbum superante, filamentis un-
ciam brevioribus, corona unciali. Flowers four,
scape 9-11 inches, spathe 2 inches long, valves dis-
joined, germen sessile, tube scarce 2j inches, limb
24-3 inches, style scarce exceeding it, filaments an
inch shorter. There are no leaves to the specimen.
On the first sight of Drummond’s leafless bulb
labelled “ Genus like Crinum,” without leaf, I
pronounced it to be an Ismene, if of any known
genus, and probably allied to the plant since
named Chor. glauca, but the integuments are
much paler, and not black as in that bulb. It pro-
duced last spring eight suberect glaucous leaves
not vaginating, having been forced early, and it
has gone to rest before the end of July as well as
Chor. glauca, and I entertain little or no doubt
that it is the plant which I have named Choretis
Galvestonensis from the specimen inSirW. Hooker’s

222

AMARYLLIDACE/E.

herbarium, which is not accompanied by the leaf.
It has certainly no immediate affinity to Crinum.

I am not sure whether there is a third species
amongst Drummond’s bulbs, or only a duplicate
with a different label. The bulbs are sulky and do
not sprout readily before they are well established.
The stamen in Drummond’s specimen is exactly
similar to that of glauca.

56. Ismene. — Filaments deflexedly conniving, ovules erect ;
seeds fleshy, green, round ; bulb ovate ; leaves deci-
duous, linear-lanceolate ; tube curved, cylindrical ;
cup large ; stigma obtuse ; anther attached scarcely
below the middle.

1. Amancaes. — Pancratium. Bot. Mag. 30. 1224. Leaves

acute, sheathing cylindrically ; germen sessile ;
tube green, furrowed, equal to the limb ; limb
yellow ; cup nearly as long, yellow with six green
stripes, and bifid jagged intervals.

Var. 1. Bot. Mag. 30. 1224. Limb semipatent, cup
cleft into six lobes at the insertion of the filaments ;
smell unpleasant. Peru.

Var. 2. Limb patent; cup not cleft into lobes, but in-
cluding the filaments and projecting a quarter of
an inch beyond their insertion ; scent more agree-
able. Peru.

2. Calathina. Bot. Mag. 53. 2685. Pancratium. Bot.

Reg. 3. 215. P. calathiforme Red. lil. 353. Leaves
blunter than Amancaes, sheathing cylindrically ;
germen sessile ; tube 4i inches, green ; limb four
inches, white ; cup three inches, white without,
green striped within, with large bifid jagged lobes.

3. Pedunculata. Tab. 35. f. 2. Bulbo valde prolifero,

foliis laete viridibus ; Ismenes Amancaes foliis acu-
tioribus minus vaginantibus, terrae magis adpressis ;
scapo foliis longiore, ancipite ; genuine peduncu-
late ; tubo viridi sesquiunciali ; limbo extus vire-
scente, intus albo, biunc. corona alba subvirescente,
sesquiunciali, viridi-striata ; stylo limbum subae-
quante; odore vix grato. Species frigoris et liumi-
ditatis patientior. Peru. Leaves more acute, and
not sheathing so high as those of Amancaes ; scape

AMARYLLIDACEiE.

223

about two feet, pedunculated ; tube green, 1 j inch ;
limb greenish without, white within, two inches;
cup greenish and white, with deep green lines,
U inch long, with bifid jagged lobes ; style equal
to the limb.

4. Nutans. — Pancratium. Bot. Mag. 38. 1561. Leaves
erect, rather spatulate, scarcely vaginating ;
flowers three or more ; germen sessile ; tube green,
shorter than the limb ; limb near 2f long, white,
equal to the style, f longer than the cup ; cup
white, , dentate and jagged. I believe there is a
mistake in the account of this plant having been
received from Brazil ; at least, bulbs brought by
Fraser from E. Florida, which were soon lost by
planting them in peat and watering them when at
rest, appeared to be this plant. I doubt whether
Nutans is preserved in any European collection to
this day. I have never seen its flower.

Hybridse. 1. Spoffbrthise. Calathina-amancaes. Bot. Reg.
20. 1665. Flore sulphureo.

Absolute rest in winter is essential to this genus, which
delights in very light sandy soil ; its cultivation is easy when
those two requisites are observed. Amancaes seems to
thrive best in pure white sand, at least in the vicinity of
the bulb. I have flowered it in the open ground by putting
a pot full of white sand with the bulbs into the border. Ca-
lathina is less particular as to soil, and pedunculata is hardier
than either, vegetates in a lower temperature, and flags
sooner in hot weather. They should be planted in a border
of light compost in April, and the bulbs must be taken up
when the leaf is cut by frost in November or sooner, without
breaking off the thick fleshy fibres which will endure through
the winter after the bulbs are taken up. They must be put
in a box or large pot, and covered with dry sand or earth,
and kept quite dry till the following April or May. If
Amancaes be set in the stove at the beginning of May, and
watered, it will flower immediately, and should be removed
into a greenhouse as soon as the first bud is ready to expand.
The sulphur-coloured mule may be forced as easily. It is a
beautiful plant, and has produced flowers in which the ex-
pansion of the cup was 3g- inches, and of the limb 5^. Its
ovules, three in a cell, are bold, and its pollen seems fertile.

224

AMARYLLIDACEiE.

I have obtained this summer strong seedling bulbs from the
mule crossed again by the pollen of Amancaes, from which I
expect an improved Amancaes with a hardier constitution.
The mule seems nearly as vigorous as Calathina. I have
also one seedling from a flower of Amancaes which had been
deprived of its anthers and touched with the pollen of the
mule. The seed of Ismene is large and round, and vegetates
immediately in a remarkable manner, forming a bulb as big
as itself (sometimes much bigger) far under ground without
pushing any leaf. As soon as the seed rots, the young bulb
must be left without water till the next spring. A person
unaware of the peculiarity of this genus and Choretis, when
he found the seed rotten, would be likely to throw away the
earth without suspecting the formation of the bulb near the
bottom of the pot. If the seedlings of Amancaes are grown
in loam, I believe they will be twenty years before they
attain size to flower ; in pure white sand or a very sandy
compost, I think they may flower the third. I have a mule
seedling from Amancaes, from seed of last year, which is
now near two feet high with five leaves. The seedling bulbs
raised this year from the mule are larger than the natural
Amancaes from seed that was sown at the same time.

The ovules of Amancaes are sometimes four in a cell,
two being abreast, and a second row behind them a little
higher, and sometimes three in a file, the base of one behind
the head of another, and probably six ovules to the cell may
be a perfect complement (two abreast in three tiers), but I
have never seen more than four. By removing the sepaline
ribs from the germen when the flower withers, the ovules
will be discovered, and their advance to maturity may be
observed. The name Calathina, being of Greek derivation,
should not be accented on the i. Ismene Calathina is said
to be a native of Brazil, but I am not satisfied of the cor-
rectness of that statement. It has been brought to England
from Buenos Ayres, where it is much cultivated in gardens,
but its natural locality in S. America has not been ascer-
tained, and it has not found its way into the herbariums of
collectors in that quarter, except from the gardens of Buenos
Ayres. Mr. Skinner found it wild in Guatemala, and
brought it to this country. I think his plant is less fragrant,
and differs a little in the length of the tube and the point of
the leaf from that which we obtained from Buenos Ayres,
but the difference, if at all permanent, is insignificant.

AMARYLLIDACE7E.

225

Callithauma? — Bulb oblong cylindrical ; leaves ensiform;
limb patent; crown equal to, filaments exceeding,
the limb. Not sufficiently known to enable a perfect
generic character to be made, but seemingly distinct
from all other known plants.

1. Viridiflorum. — Pancratium viridiflorum. Flor. Per.
3. 55. Bulb a palm and half long; leaves long,
flat, suberect, diverging ; scape the height of a
man, or 6 feet ; spathe with a white deciduous
bracte to each flower ; flowers 4-5, large, beau- -
tiful, entirely emerald green ; cup enormous ; seg-
ments of the limb acuminate. Plentiful in the
woods of Huassahuassi, and on the headlands and
stony places of Palca in Peru. Ruiz’s specimens
of this marvellous plant were lost by shipwreck. It
is barely possible that it may be an Ismene, but
very improbable.

2? Spathulatum.— Bulb large, round; leaves thick,
coriaceous, dark green, petiolated, spatulate, de-
ciduous ; scape said to be a yard high ; flowers
large, green. Imported by Richard Harrison, Esq.
having been dug from a mountain said to be 500
miles east of Truxillo, but the distance is probably
exaggerated. It is probably of a separate genus,
intermediate between this plant and Leperiza, with
which it agrees in the form of the leaf. It has
dwindled in many collections, but thrives pretty
well with me in loam. It has not flowered in Eu-
rope, nor has any specimen of the flower been re-
ceived. It seems to like heat.

57. Calostemma. — Bulb ovate ; leaves linear lorate ; umbel
many-flowered, pedunculated ; germen by defect 1-
celled ; ovules 2-3 ; tube cylindrical ; limb funnel-
shaped ; crown often irregularly slit ; filaments short,
erect; anthers small, erect, attached near the base;
style attenuated; stigma simple, small; seeds 1-2,
green, fleshy, flattened on one side by contact. Na-
tives of Australia.

1. Purpureum. — Bot. Reg. 5. 422. Leaves about 15
inches, scape 16-20 flowered ; peduncles about an
inch long, bracteate ; spathe 3-valved ; tube
limb half an inch, purple, midribs continued down
Q

226

AMARYLLIDACE^E.

the tube and germen; 12 triangular teeth to the
cup between the filaments ; cup often slit in one
or more places.

2. Liiteiun. — Bot. Reg. 5. 421. Leaves narrower,

flowers yellow with green ribs, and six purple
spots at the base of the cup; cup with teeth as in
Purpureum.

3. Album. — Brown, prod. 298. Flowers white; teeth

of the cup linear, emarginate.

58. Vagaria. — Leaves linear-lorate ; umbel many-flowered,
pedunculated ; germen roundly triangular, ob-
long ; cup imperfect, filaments dentately winged
on each side, inserted at the base of the limb ; limb
patent ; filaments short, alternate ; anthers short
(incumbent?) style tapering, a little bent; stigma
small, slightly trifid.

1. Parviflora. — Pancratium parviflorum. Red. lil. 8.
471. Desfontaine’s Supp. Cat. Hort. Par. It is
not known from whence this plant was introduced
into the Parisian garden, where it produced its
flower stem in the autumn from the centre of the
old leaves, which appear to vaginate more than
those of Calostemma, or, when the leaves had pre-
viously decayed, before the autumnal growth of
young foliage. Having seen the cup of Calo-
stemma irregularly slit to the base, though that
feature was not observed by Dr. Brown, I felt some
doubt whether it might be regularly slit in another
species, and in complete ignorance of the ovarium
and fruit, as well as the native country, of this
plant, there seemed to be a possibility that it
might be an Australian Calostemma, differing in
that respect, in a wide-expanded limb, a style a
little curved, and minute splitting of the stigma.
It is however very improbable that an unknown
Australian plant should have made its appearance,
at the period when this bulb was introduced, in the
Parisian garden, and there only, without any re-
cord concerning its introduction ; and it is much
more probable that it will be found to be a native
of Spain or Egypt, or some other country border-
ing on the Mediterranean ; in which case it is so

AMAKYLLIDACEJE.

227

unlikely that an Australian genus should manifest
itself there, that we may with safety assume that
its fruit will approach to that of Pancratium rather
than Calostemma : it is certainly not a Pancratium.
The flowers are white within, and mainly green on
the outside, approximating to those of Ornithoga-
lum nutans in a kindfed order, and I suspect in
fructification to P. Illyricum. It has some affinity
to Eurycles.

59. Eurycles. — Bulb ovate; leaves very wide, petiolated ;
scape scarcely preceding the leaves ; spathe 3-valved;
umbel many-flowered ; germen erect, cells nearly
obsolete, 2-(rarely 3-)seeded ; tube cylindrical, angles
obsolete; limb regular, alternate; segments equal;
cup more frequently imperfect ; filaments diverging,
inserted at the base of the limb, more or less winged,
and often scarcely connected at the base by a mem-
brane ; anthers short, affixed at the base ; style erect,
cylindrical ; stigma small, fimbriated ; one ovule often
erect, the other pendulous; seed with two thick fleshy
integuments, the bulb usually protruding from the
seed in the immature capsule ; capsule obovate, 3-
ribbed, indehiscent; dissepiments decayed.

1. Amboinensis. — Pancratium Bot. Mag. 1419. Cri-
num nervosum L’Herit. S. A. S. Pancr. nervi-
folium Parad. Lond. 84. Eurycles Salisb. Hort.
Soc. Trans. Proiphys. Herb. App. Cepa sylvestris,
Rumph. Amboin. 6. 160. t. 70. f. 1. Narcissus.
Comm. H. Am. Rudb. Elys.

This plant, a native of Amboyna and the Philippines, is
subject to great variation of the crown. It was complete,
though deeply cleft, in the specimens figured in the Bot.
Mag. and Par. Lond. but was almost obsolete, the filaments
being scarcely connected by a thin membrane at the base, in
the plant cultivated by me ; the filaments, as Mr. Sweet
asserts under Eurycles nuda, Hort. Sub. Lond. were not
even winged in another variety that flowered at Colvill’s,
which cannot, however, properly be separated as a species
on that account solely, when we look to the intermediate
variation. It is a stove plant, requiring rest in winter, and
liable to rot if wet gets into the neck of the root. Ovules
obovate, dehiscent at the top, with the embryo protruded.

Q 2

228

AMARYLLIDACEiE.

2. Australasica. — Bot. Reg. 9. 715. Eurycles alata.
Sweet H. S. L. Calostemma Cuninghami. Hortu-
lanorum. A smaller species from Australia, of
which the cup is split to the base, approaching in
that respect to the second var. of Amboinensis. It
differs however in a much narrower leaf and a
hardier constitution, and will live in a greenhouse.
I have not seen its flower.

Mr. Salisbury’s name Eurycles, being very objectionable,
was altered with his consent to Proiphys. Eurycles was
derived by him, as he informed me, from the obsolete nomi-
native of the Greek word kladi, which however would be
hi as , not kies , and means a branch, but never a leaf ; and the
breadth of the leaf on which he founded the generic name
was unsatisfactory for the fundamental feature. I omitted
to state in the App. that the alteration was made with Mr.
Salisbury’s approbation ; and, as Eurycles has been since
adopted in several works, and has the priority, it seems now-
advisable to retain it. My friend, Sir W. Hooker, has sug-
gested another derivation for the wrord, which had not oc-
curred to its framer. Seeds of Crinum and Hymenocallis
will often germinate in an over-ripe capsule, as grain sprouts
in the ear, and Clivia before the capsule appears quite ripe ;
but I know of no other instance of the embryo protruded
from the ovule at the point furthest from its attachment, and
the formation of a bulb proceeding in the immature fruit,
wdiich is conspicuous in this genus. It seems however not
to be an invariable occurrence, for I received two fresh seeds
of Australasica which had not germinated. This genus
comes very near to Calostemma, and inhabits the same
vicinity.

5. Amaryllidiformes. — Scape solid, seeds fleshy.

60. Griff'inia. — Leaves w7ide,petiolated; umbel many-flowered;
germen pedunculated; tube cylindrical, declined; seg-
ments reflex, the low7er divaricate, the lowest stretched
forward ; filaments decurrent in the tube, recurved ;
anthers incumbent. Seeds large, roundish, white
witli a small black chalaza.

1. Hyacmthina. — Bot. Reg. 2. 163. This plant grows
in woods on the hills behind Rio Janeiro, eight
inches deep in strong loam, the scape and leaves
rising to the height of two feet. Cultivated in the

?latt 36.

AMARYLLIDACEtE.

229

sunny gardens of Rio it only acquires half the
size. Potted in our stoves it rots in strong- loam,
and seems to succeed best in light peat and sand.
It cannot thrive in its native soil in a pot, without
very good drainage and cautious watering.

2. Intermedia.-— Bot. Reg. 12. 990. A plant of inter-

mediate size, introduced by Mr. Harrison. I have
never seen it,

3. Parviflora. — Bot. Reg. 6. 511. A much smaller

species, which I never saw but in Mr. Griffin’s
collection.

This Brazilian genus is allied in leaf to the South
American Eucrosia and Urceolina, but in flower it ap-
proaches very near to the Asiatic genus Lycoris. The
flowers in all the species are purplish blue and white.

61. Lycoris. — Leaves linear, hiemal; umbel many-flowered;
germen declined ; perianth with tube declined, obso-
letely triangular, wider at the mouth, and curved
upwards; limb curved upwards ; filaments long, re-
curved, inserted equally at the mouth of the tube;
anthers oblong, incumbent ; style long, recurved,
circumfimbriated near the summit ; stigma simple.

1. Aurea. — Am. aurea. Bot. Mag. 409. Bot. Reg. 8.

611. Jacq. H. Sch. 1. 38. t. 73. Flowers golden.

I have a variety of this plant from the Chinese
islands, which usually flowers with two supernume-
rary segments to the flower.

2. Radiata. — Am. radiata. Bot. Rep. 95. Bot. Reg. 7.

596. Specim. ex Japonia. Herb. Lambert. Flowers
crimson.

Lycoris ought properly to stand beside Nerine; Bruns-
vigia is more nearly allied to Nerine, and Griffinia and
Lycoris cannot intervene in a consecutive arrangement
without interrupting the affinities of the African plants.
The leaves of both species are glaucous. They are natives
of the extreme East, Aurea of China and the Chinese
islands, radiata of Japan. Asia yields its tribute of the
genus Crinum, which is remarkable for extending round
the whole belt of the world, within and near the tropics ;
but, with that exception, it produces few plants in this

230

AM ARY LLIDACE^E.

natural order. Lycoris flowers in the autumn, after a period
of rest ; radiata but rarely in our stoves. It is hardier than
aurea, and will live in a green-house. Radiata cultivated
at Calcutta, flowers after the hot season, but sometimes fails
to yield any blossom for two or three seasons. Aurea likes
a light sandy soil.

62. Olivia. — Spathe many-valved ; germen pendulous ; pe-
rianth curved on the upper surface, straighter beneath;
tube cylindrical, widened; limb with fourfold diver-
sity ; filaments straight, alternately equal, equally
decurrent in the tube ; style straight ; fruit valve-
less, with a coloured outer coat, and fleshy middle
coat, seeds round-oblong, angular by contact, vege-
tating often in the fruit.

1. Nobilis. — Bot. Reg. 14. 1182. Imatophyllum Ay-
toni. Bot. Mag. 55. 2856. Cells 3-seeded; seeds
pearl-coloured ; flowers numerous, scarlet, tipped
with yellow and green ; leaves with a harsh
margin.

This beautiful plant was first discovered by Dr. Burchell,
in whose herbarium, soon after his return from Africa, I saw
a fine specimen, which not having been carefully examined,
had been mistaken for an Agapanthus, to which its root and
leaves have a striking affinity. I soon after became pos-
sessed of a plant of this species, brought over by an officer
who had been employed on the Caffre frontier, and I recog-
nized it to be the plant I had seen in Dr. Burchell’s herba-
rium, but concluded it to be an Agapanthus. Not long
after, I obtained for Mr. Tate, from the kindness of Dr.
Burchell, a precise account of the spot where he had seen
this plant; the result of which was a large importation of the
roots ; but after they had vegetated, Mr. Tate mistook them
for the common Agapanthus, and was about to dispose of
them as such, when I saw them accidentally, and imme-
diately recognized them. One of the plants flowered for the
first time in this country in the collection of the Duchess of
Northumberland, after whom it was named. By a singular
accident it appeared on the same day in the Bot. Reg. and
Mag., being named in the latter work Imatophyllum Aytoni,
but the name Clivia nobilis has been generally preferred.
Sir W. Hooker was mistaken in supposing Mr. Bowie to

AMARYLLIDACEjE.

231

have been the first discoverer of this plant. Indeed, roots of
it were in my hands before he had seen it in Africa. It is a
plant of perfectly easy culture, requiring no particular care
but to shelter it from frost, and it flowers freely in the green-
house if placed near a front light, and ripens its seeds ; but
the seedlings are of very slow growth. The reason of its not
having flowered in my collection earlier was that from the
slowness of its growth 1 had been induced to put it in the stove,
hoping that it might grow there more freely; but the heat
increased its sulkiness. Its berries in a cool greenhouse are
sometimes a year ripening ; they will hang on a long time
after, and on opening them the seeds will often be found to
have vegetated within the pericarp. The fruit as figured in
the Bot. Mag. is shrivelled and unlike its usual appearance.
The ripe pericarp is properly smooth, without furrows, and
scarlet, with a juicy pulp between the outer and inner coat
like that of Haemanthus and Tamus ; but I have now before
me a fruit of Clivia, nearly ripe, in which the seeds have
split the pericarp, which is outwardly of a rough brownish
green, and instead of juicy pulp has a tough greenish flesh,
and will probably never become red and juicy; shewing how
easy is the transition from a coloured and juicy to a dry
capsule.

This plant has been likened to Cyrtanthus, and a ques-
tion entertained whether it is distinct from that genus, and
Sir W. Hooker says that it has the habit of a Cyrtanthus,
but to me it appears to have scarcely any point of resem-
blance, except the scarlet colour and pendulous posture of
some of that genus. Professor Lindley observed its real
affinity to Hsemanthus, which it approaches so nearly in its
most important features as to have made me hesitate to pro-
nounce that it was not an Haemanthus, and capable of breed-
ing with species of that genus. Those who likened it to
Cyrtanthus did not consider the distinctions of scape, seed,
and bulb, nor that Cyrtanthus has the flower with a long
wide-mouthed tube and a very short open limb ; whereas
Clivia has a short close tube and a long limb almost closed
at the mouth. The perianth of Clivia has been inconside-
rately called tubular, which is not the case, though the seg-
ments are closed so as to give a tubular appearance. The
most skilful botanists have often failed to consider suffi-
ciently which features are compatible and which are incom-
patible with identity of genus ; but the affinities of plants

232

AM ARYLLIDACEiE.

cannot be determined without such consideration. Many of
the points set forth in generic definitions are, in themselves,
unimportant, except as being symptomatic of some more
essential difference. The distinctions of a hollow or a solid
scape, and of seminal structure, are insurmountable barriers,
and I believe it to be as impossible for Clivia to breed with
a Cyrtanthus or Phycella, to which it has been also likened,
as with an oak tree ; but when I compare it with Haeman-
thus I find such a conformity, that to distinguish it I must
rely on a drooping peduncle, an inequality and obliquity of
the perianth, the number of ovules, the permanency of its
leaves, and the length of time the seeds are ripening ; and I
can only rely upon such distinctions by judging from ana-
logy that they probably indicate a natural diversity, and I
still look to experimental attempts to cross it with Haeman-
thus, in order to establish whether they are symptomatic or
unimportant features ; but when I consider the graduated
diversity of form in the genus Hsemanthus, that carneus has
a disposition to drooping peduncles, that Puniceus has some
obliquity of perianth, that multiflorus and puniceus disagree
marvellously from the rest in root and leaf, yet depart from
each other and agree respectively with different species in
the flower, I cannot look upon the distinction of Clivia with
decided confidence, though I believe it to be well founded.

63. Htemanthus.— Spathe 3-many-valved ; perianth nearly
regular, tube straight ; filaments straight, inserted in
the top of the tube ; anthers short, suberect ; lobes of
stigma three or obsolete ; pericarp valveless, furrow-
less, often coloured, middle coat pulpy ; cells with
one ovule, dissepiments obsolete.

§. 1. Bidbo ovato, folds cylindraceo-vaginantibus, undulatis.
Bulb ovate, leaves cylindrically sheathing, undulated.

A. Limb patent.

1. Multiflorus. — Bot. Mag. 24. 961. et 45. 1995. Lodd.

B. C. 1948. Spathe 3-leaved, drooping, and
withering. Sierra Leone.

2. Abyssinicus. — Specim. Salt. Herb. Lambert, ex

Abyssinia. Spatlia plurivalvi 1| - unc. purpurea
reflexa, floribus numerosissimis, pedunc. 1^-unc.
floribus H. multifioro dimidio vel ultra minoribus.
Many-valved purple reflex spathe, flowers very

Fla .■

AMARYLL1DACEA3.

233

crowded, scarcely half the size of those of mul-
tiflorus.

3. Delagoensis. — Specim. Forbes 101. Herb. Lindl.

Spatha 3 (vel 4?) valvi, l^-unc. bractea una
^ unc. ^ unc. lat. cseteris gracilibus ; floribus nu-
merosissimis, pedunc. unc. vel ultra, germ, gracili,
tubo cylindrico vix \ unc. perianthio rubro gracili
patente filarnentis breviore. From Delogoa Bay.

These two plants seem to form a link between multiflorus
and puniceus, and shew that the genus takes a wide range
in Africa. It is impossible from the dry specimen to be
certain whether the limb is patent as in multiflora or not.

B. Limb erect.

4. Puniceus. — Bot. Mag. 32. 1315. Lodd. B. C. 912.

Spathe many-valved, green, suberect, persistent ;
flowers erect, closed, pale yellowish red.

Var. fortuita. Subalba. Variety with flowers nearly
white, imported by Mr. Lee from the Cape.

§. 2. Foliis erectis, undulatis; bulbo cylindraceo-ovato. Leaves
erect, undulated ; bulb cylindrically ovate.

5. Undulatus. — PI. 30. fig. 1. Masson. Herb. Banks.

Bulbo unc. lat. foliis 6-unc. i unc. lat. undulatis-
simis, erectis ; scapo 2- unc. spatha coccinea, sub-
unc. filarnentis alternis stylo longioribus. This
curious little plant has lain long unnoticed in the
Banks. Herb, having been one of Masson’s collec-
tion. The inflorescence is almost destroyed, but
one flower and part of three valves of a scarlet
spathe remain. The bulb seems to approach to
that of the first section, and is not imbricated.

§. 3. Foliis non vaginantibus , non undulatis ; bulbo com -
presso imbricato. Leaves not sheathing, not undu-
lated ; bulb compressed, imbricated.

A. Limb patent.

6. Amaryllidioides. — Jacq. H. Sc. 4. p. 5. t. 408. Two

leaves with a red tip, acute, erect, an inch wide,
smooth, not spotted ; scape 5 inches, pinkish ;
spathe lanceolate, red with white edges; ped. green,
an inch long ; germen red ; flowers semipatent ;

234

AM ARYLLIDACEAS.

limb scarcely half an inch long, rose-coloured;
stigma minute, trifid.

7. Carneus. — Bot. Reg. 6. 509. Bot. Mag. 61. 3373.

H. Roseus. Link enum. 1. p. 309. Leaves broad,
hairy all over ; spathe withering, not coloured ;
flowers semipatent, rose-coloured. This species is
subject to variation. The bulb figured in the Mag.
has longer peduncles and more acute leaves ; my
bulbs have the leaves obtuse.

8. Strigosus. — PI. 30. fig. 2. Specim. Masson. Herb.

Banks. Foliis binis recumbentibus, 2-unc. lj| lat.
scapo 6-7 unc. spatha reflexa | unc. 3 (4 ?) valvi,
lane. lin. acuta, floribus 15-16, ped. 4 unc. perianth,
pallide ? roseo, cum germ. fil. vix brevioribus,
stylo vix longiore ; stigmate minute trifido. An-
other curious plant of Masson’s which has been
overlooked ; allied to carneus.

9. Lanceaefolius. — Jacq. H. Sch. 1. 31. t. 60. Leaves

recumbent, ovate, acute, 6 inches by 2, thickly re-
curvedly ciliate ; flowers 7, like pumilio, but
smaller.

10. Pumilio. — Jacq. H. Sch. 1. 32. t. 61. Leaves sub-
erect, f wide, spotted at the base ; spathe large,
coloured ; flowers 4-5, white, widely patent.

11. Brevifolius. — PI. 30. fig. 3. Specim. Masson. Herb.
Banks. Foliis tribus uncialibus, § latis ; scapo un-
ciali; spatha reflexa, 4-valvi, gracili, acuta; floribus
10, una cum germ. § unc. albis. This very curious
diminutive plant, with three broad leaves an inch
long, scape an inch high, pendulous spathe and
ten white flowers, is another plant of Masson’s.
Both this and strigosus have MS. names affixed to
the specimens, which were certainly given with an
intention of publishing them in the Hort. Kew.
Undulatus is named by me.

B. Limb erect , close.

12. Humilis. — Jacq. H. Sch. 4. 6. t. 411. Two leaves
six inches by II, narrow at the base, strongly ci-
liated backward, suberect ; scape 2-3 inches, pale
green; spathe conspicuous, purplish pink; peduncle
| ; flowers white, just shorter than the filaments.

AMARYLLIDACEjE.

235

13. Virescens. — Bulbo viridi, foliis pubescentibus vel
ciliatis, spatha non colorata, perianthio albescente
filamentis breviore. Bulb green, leaves pubescent
or ciliated, spathe not coloured, perianth whitish,
shorter than the filaments.

Var. I. Pubescens. — Bot. Reg. 5. 382. Lodd. B. C. 702.
Leaves broad, hairy, varying in obtuseness ; spathe
many-leaved, green ; perianth white, shorter than
the stamina and style. In the Banks. Herb, pu-
bescens has the leaf acute, albiflos obtuse. My
plant of pubescens has obtuse leaves.

Var. 2. Intermedius. — Bot. Mag. 31. 1239. Leaves
smooth, margin villously ciliated, spathe white
with green veins ; style more prolonged.

Var. 3. Albiflos. — Bot. Reg. 12. 984. Jaccp H. Sch. 1.
31. t. 39. Lodd. B. C. 602. Leaves smooth,
margin ciliated ; spathe whitish ; style not pro-
longed. I suspect that these natural varieties will
be found permanent by seed, and that they are
distinct in their localities, though properly to be
united under one head ; but pubescens is not a
right name to comprehend the three, though very
fit to designate the one which is downy. Their
green bulb distinguishes them from all the species
I know.

14. Quadrivalvis. — PI. 31. f. 4. Jacq. H. Sch. 1. 30. t.
58. Bot. Mag. 37. 1523. Leaves narrow, acute,
spotted (not always) on the base behind, hairy on
the surface ; scape spotted ; spathe large, bright
red ; perianth red, very close, tipped with white ;
stigma trifid.

15. Rotundifolius. — PI. 31. f. 8. Bot. Mag. 1618. Leaves
pressed to the ground, large, round, with a rough
margin ; scape and spathe deep red, 4-valved ;
perianth pale red with white tips, shorter than
stamina and style.

Var. 2. Multivalvis.— Lambertianus. Martins ap. Schultes.
Callosus. Burchell Cat. et tab. pict. Differs in
nothing but more valves to the spathe, and a trifling
variation of the leaf.

16. Sanguineus. — PI. 31. f. 5. Jacq. 11. Sch. 4. 4. t.

23G

AMARYLLIDACEjE.

407. Leaves large, unspotted, when young tipped
with red ; scape deep red ; spathe-valves acute,
narrow, veined with deep red ; flowers crowded,
but not compressed ; perianth red, with white tips
and base, shorter than filaments ; stigma trifid.

17. Hyalocarpus. — Jacq. H. Sell. 4. 5. t. 409. PI. 31.
fig. 9. Leaves broad, not spotted ; scape spotted ;
spathe longer than the flowers, bright red ; ped.
and germ, each half an inch, pale; limb red, tipped
with white ; stigma trifid ; fruit purplish white.
Seed one (represented blackish?)

18. Moschatus. — PI. 31. f. 7. Jacq. H. Sch. 4. 6. t. 410.
Leaves 1 i foot long, 4 i inches wide, at first
thinly pubescent, crossed-barred with red below,
and deep green upwards ; spathe light scarlet ;
flowers crowded, compressed ; ped. and germ, pale
green, flower red, tipped with white, close ; fila-
ments half an inch longer.

The last five species agree with the next in having the
spathe red and the perianth tipped with a white glandular
callosity ; the two first are widely distinguished by the foli-
age ; the latter have the leaves more nearly allied, but they
appear to be distinct.

19. Coccineus. — Foliis rubro apiculat.is magnis linguae-
formibus, umbella multivalvi colorata floribus sub-
longiore, limbo rubro apice albo calloso filamentis
breviore, scapo maculato.

Var. 1. Grandivalvis. — H. Coccineus. Hort. Kew. Bot.
Mag. 27. 1075. Spathe with large deep-red
roundish valves ; leaves not barred.

Var. 2. Coarctatus. — PI. 31. f. 6. Jacq. H. Sch. 1.
30. t. 57. Bot. Reg. 3. 181. Spathe close, orange-
red or salmon-coloured ; perianth dull red, pale at
bottom ; leaves smaller, not barred.

These agree in foliage, not barred, but tipped, when
young, with red, spathe as long as the flowers, limb shorter
than the filaments, and tipped with a white callosity.

Var. 3. Carinatus. — Hort. Kew. This plant seems only
distinguishable from Coccineus by the leaves being
much narrower and more channelled. It has not
flowered with me yet, but the particulars recorded

AMARYLLIDACEiE.

237

are insufficient to separate it from Coccineus.
Leaves much narrower, more channelled, and much
longer than those of coarctatus, not barred.

I have four other species or varieties which have not yet
flowered, but seem allied to Coccineus. 1. Leaf like coarc-
tatus, with two or three red spots. 2. Leaf closely barred
with narrow greenish-red bars towards the base; root given to
produce offsets. 3. Leaf barred more rarely with a redder
colour. 4. Leaves narrower than grandivalvis, with broad
reddish bars. All smooth and tipped with red, which if the
flower agree, as I expect, with Coccineus, will make 7 varie-
ties, unless No. 1. should prove to be coarctatus. I have that
plant with the spathe not more compressed than grandivalvis.
I do not believe that in the native specimens of the latter
the spathe is only 4-valved, but many-valved in the same
when cultivated, as stated in Hort. Kew.

20. Incarnatus. — PI. 31. f. 1 . Burchell 4556. Tab. pict.

1818. Foliis recumbentibus 7-unc. obtusis obovatis,
infra attenuatis, scapo rubro, spatha multivalvi flo-
ribus breviore pallide rubro, ped. unc. tubo et
limbo rubro, stylo limbo longiore filamentis bre-
viore, stigmate minuto. I believe I possess this
plant, which is allied to tigrinus, but it has not
flowered. Leaves not barred, scape red, flower
entirely red, style longer than the limb, shorter
than the filaments, stigma small ; spathe pale red ;
valves short and narrow.

21. Tigrinus. — PI. 31. f. 3. Jacq. H. Sell. 1. 29. t. 56.
Bot. Mag. 41. 1705. Leaves ciliated, obovate, very
little attenuated; scape green, speckled; spathe
red, valves short, narrow ; germ, and ped. green ;
tube white ; limb red ; style and filaments equal,
longer.

22. Crassipes. — PI. 31. f. 10. Jacq. H. Sell. 4. 7. t. 412.
Leaves closely ciliated, arcuate, 6 inches long,
much spotted and barred with red, 1^ wide; scape
spotted ; spathe broad, scarlet, nearly as long as
the flower, which is pale red, shorter than the fila-
ments, stigma trifid. Allied to quadrivalvis, but
has not the callous tip or hairy surface.

23. Zebrinus. — Apud me nondum floridus, foliis 8-11-
unc. hevibus, obtusis, superne 1^ latis, infra at-

238

AM ARYLLIDACEjE.

tcnuat. | unc. lat. clorso ad dimid. longitud. den-
sissime, superficie rarius, clathrato-macul. Species
distincta, nisi crassipedis varietas. Leaves 8-1 1
inches, obtuse, 1^ wide, below only half an inch ;
closely barred half their length below, less densely
on the upper surface, smooth.

24. Concolor. — PI. 31. f. 2. Burchell Herb. B. 276.
Foliis lorato-lanceolatis laevibus subacutis 8 unc.
1|^ lat. scapo 6 unc. spatha 1^ rubra circ. 8 valvi
Horibus vix brev. perianthio stylo filamentis conco-
loribus, filamentis limbo longioribus stylo brevio-
ribus, stigmate trifido. A very desirable small
species, being as well as crassipes and quadrivalvis
equal to the larger in the inflorescence.

I have lately received from Mr. Loddiges a small Hae-
manthus, imported by him from the Cape under the name
Coccineus, which has a smooth leaf not an inch wide, and
dotted at the base with red, which is probably allied to Con-
color. Haemanthus dubius H. and B. Kunth. must be a
Phycella. Haemanthus vaginatus Thunb. Flor. Cap. from
its six linear leaves, and two-valved spathe, does not agree
with Haemanthus, and I entertain no doubt of its belonging
to my genus Hessea intermediate between Nerine and Stru-
maria. It is referred with doubts by Schultes to Bruns-
vigia, with which it does not agree.

It will be observed that in this genus there is an extra-
ordinary diversity between the first and second section in
the bulb, leaf, and habit, such as occurs in no other genus ;
but I cannot find any diversity in the flower and fruit, and a
like variation of species with close and with patent flowers
occurs in both divisions. The difference is so great as to
raise a doubt whether they form one genus. In the first
section the leaves have a long fistulous or hollow cylindrical
foot-stalk, which acts as a sheath : in that respect, perhaps,
the difference is not much greater than in Amaryllis, of
which Blanda vaginates above six inches ; but the conse-
quence of this vagination in Haemanthus is a different form
of bulb. That of Haemanthus has been called imperfect, but
I see no imperfection in it. All such bulbs consist of coats,
which are the permanent base of the leaves after the upper
part has perished. In the species which have only two very
broad leaves without a footstalk, the sheathing base is necessa-

AMARYLLIDACEyE.

239

rily much wider one way, which makes the shape of the bulb
compressed, and not round like those which are formed by
leaves that sheath cylindrically ; consequently the bulbs of
the first section are not compressed, but cylindrically ovate.
In all bulbs the outer coats must be stretched by the internal
growth of new leaves ; when the leaves sheath cylindrically,
the neck or orifice being small, the old coats in stretching can-
not slip downwards; but, in the second section of Haemanthus
from the breadth of the leaves, which have no cylindrical
base, the mouth of the bulb is wide one way ; and, as fresh
leaves arise, the outer coats, being stretched by the internal
growth, drop, and consequently become shorter each year
till they decay, which gives an outward appearance of imbri-
cation ; but the only structural difference is the leaves not
sheathing cylindrically at the mouth of the bulb. The most
startling difference is that those of the first section flower
with the leaves in vigour, instead of before their appearance.
The very singular plant I discovered in the Banks, herb,
though in a very imperfect state, with undulated foliage,
seems to form a link between the two sections, and it is to be
regretted that we cannot obtain a perfect knowledge of it.

G4. Buphane. — Scape precocious, 100-230 flowered, pedun-
culated ; spathe 2-valved, tube cylindrical, limb
expanded, filaments inserted without the tube, erect,
diverging; stigma a single point; capsule turbinate,
3-celled, 3-furrowed, 3-valved, few seeded.

1. Distycha. — Patterson trav. t. 1. Bulbo subrotundo

maximo indusiis nigris. Bulbs of this species were
imported by Mr. Tate, a few years ago, as large as
a man's head.

2. Toxicaria. — Brunsv. tox. Bot. Reg. 7. 567. loculis

dispermis.

Var. 2. Obtusifolia. — Burchell. herb, et tab. pict. This
species varies much in foliage, some having the
leaves quite acute, others obtuse, some very little,
some much, undulated. Some have the bulb long
and cylindrical, others shorter and thicker. It
does not appear from Dr. Burchell's drawing that
there is any corresponding difference in the inflo-
rescence. The coats of the bulb are pale brown.
It is called the poison bulb, and said to be fatal to
cattle, from which the generic name is derived.

240

AMARYLLIDACEjE.

3. Ciliaris. — Brunsv. Ciliaris. Bot. Reg. 14. 1153.

Amaryllis. Linn. Hsemanthus. Willtl. This plant
lias, 1 believe, never flowered in England, except
in the collection of J. H. Slater, Esq. of Newick
Park, who appears to have been particularly suc-
cessful in the treatment of this genus and Bruns-
vigia. The bulbs of this species rot if not kept
perfectly dry in summer, and bleed if exposed to
much heat at that time. The only bulb of toxi-
caria I have seen in flower was cultivated above
ground, but so treated they are apt to contract
damp from the atmosphere and rot. I believe they
should all be grown in light loam, placing white
sand in contact with the bulb. Capricious water-
ing is the bane of all such bulbs. They form their
leaves naturally in the cool and rainy season, and
rest during the period of heat, flowering with
the first rains that follow it. It is evident, there-
fore, that the temperature may be advantageously
increased after the leaves are full grown, but heat
at their first pushing will make them grow weak
and is unnatural to them ; they must all have a
free supply of water while the leaf is growing, and
it must not be allowed to wither prematurely. It
has been found, as well as several Mesembryan-
themums, growing in clay at the Cape, but it is
doubtful whether the evaporation in this country is
sufficiently rapid to enable it to thrive in a strong
soil here.

4. Guttata. — Linn. Syst. veg. ed. 13. p. 265. Leaf Herb.

Banks, marked Haemanthus Guttatus, folio 10-unc.
unciam lato, infra attenuato, margine spinis semi-
uncialibus. PI. 22. fig. 1. I have given the repre-
sentation of this leaf from a specimen named by Lin-
naeus Amaryllis guttata. After his death, from
his notes, it was changed to ciliaris. It may be
that he had ascertained by other specimens that it
was only a variety of ciliaris ; but, although ciliaris
varies a little in the form and ciliation of its leaves,
I cannot think this narrow attenuated leaf, with
such long marginal spines, can belong to the same
species. If it prove to be only a variety differing
in leaf, it is yet so remarkable as to deserve a sub-

A M ARY LLIDACEiE.

241

ordinate name. That given it by Linnaeus implies
a spotting, as if by drops of water sprinkled on it,
which does not belong to Ciliaris.

The leaves in this genus are annual; the peduncles trian-
gular, crowdedly suberect, afterwards diverging to form a
spherical head. The perianth is smaller, the stigma more
acute, than in Ammocharis. It is placed under Brunsvigia
in the Bot. Reg. referring to the character of that genus at
p. 192, which is framed with a long string of alternatives,
including three differences of seed, and cannot possibly be
maintained. The inflorescence of this genus approximates
closely to a portion of Hgemanthus, especially H. carneus,
but it departs widely from that genus in its fruit, which
approaches nearer to Brunsvigia.

65. Ammocharis. — Leaves vernal, not sheathing; tube cy-
lindrical, enlarged, sepals not imbricating thereon ;
filaments adnate at the base of the limb almost equally;
anthers short, affixed at the middle ; pollen minute ;
stigma obtuse, one or two lobes obsolete ; capsule
turbinate, 3-valved, 3-celled, 3-furrowed.

1. Falcata. — Amar. falc. Bot. Mag. 35. 1443.

2. Coranica. — Am. Coran. Bot. Reg. 2. 139.

Var. 2. Pallida. — Bot. Reg. 15. 1219.

Ammocharis is an intermediate gradation between Bu-
phane and Crinum, differing from the latter in anthers,
pollen, capsule, filaments inserted just within, instead of
without, the tube, which is wider at the mouth, a shorter
proportionate limb, and leaves not sheathing at the base.
The particles of its pollen are much smaller. From Buphane
it is distinguished by leaves that sprout again identically,
like those of Crinum, after a period of rest, the wider mouth
of the tube, the insertion of the filaments within the tube,
and the difference of the ovules which are more numerous.
Those of Ammocharis and Crinum adhere inseparably by
one whole side to an intermediate body or placenta ; those
of Buphane are separately attached, closing together from
two threads, instead of proceeding from a solid mass, which
is a decisive generic feature. The habits are quite different
from those of Amaryllis, Brunsvigia, and Nerine, which
flower in autumn before the appearance of their leaves.
Ammocharis produces its foliage in the spring, some time
before the blossom appears, and lies at rest in the winter.

R

242

AM A RYLLIDACEiE.

The seed approaches to that of Crinum. Professor Lindley
(Bot. Reg. 1219.) formerly expressed a doubt of the distinc-
tion between Ammocharis and Nerine. It will be found,
however, on examination that those two genera are very far
removed from each other in natural affinity. The genus
Nerine is distinguished by characters that cannot be misap-
prehended, the gibbous monadelphous union of the filaments
with the limb which has only an annular connexion. The
nearest affinity of Ammocharis is unquestionably to Buphane;
that of Nerine to Brunsvigia and Strumaria. The ovules of
Nerine do not adhere like those of Ammocharis to an inter-
mediate placenta.

Bulbs of Ammocharis require complete rest in winter,
when they must not be watered. They are exceedingly
thirsty in summer, and if planted in light earth and left fora
short time in the sun without water, the leaves will die back.
It is essential, therefore, to supply them constantly with
moisture. To make them flower the pot should be plunged,
in the summer, in a hotbed after the leaves have grown to
some length. In that situation it should be shaded from
scorching sunshine, and it should not be kept longer than
necessary in the hotbed. A rich and not very light soil is
suitable to it in cultivation, and I have found it even succeed
pretty well in peat ; but I have found the bulbs at all times
liable to unexpected rottenness, from incautious watering.
It does not seem possible to distinguish falcata fromCoranica
by the bulb or leaf, yet there is sufficient diversity between
their inflorescence to have induced Mr. Ker to place them in
different, genera, though he was certainly mistaken in so
doing. I have found the cells in falcata 6-8-seeded, in Cora-
nica 15-16-seeded. The germen of Coranica has the same
angles as falcata, but they are faintly marked, and on the
peduncle they become almost obsolete. The difference be-
tween them is perhaps sufficient to separate them as species,
though it can be only seen in the blossoming.

66. Crinum. — Germen thickest in the middle; tube cylin-
drical, slender ; filaments inserted just outside the
tube, more or less recurved ; anthers incumbent, ver-
satile ; stigma 3-cornered or trifid ; capsule soft, de-
formed, without valves or furrows ; dissepiments
obsolete. Seeds very irregular in form, size, and
number. (Leaves properly tubular at the base.)

AMARYLLIDACE^E.

243

1. Patentes. — Flowers patent.

A. Inclinatce. — Buds inclined.

1. Bulbo longe columnar i, folds dinar icantibus, umbelld nu-
merosd pedunculatd. Asiatics, et Australasicce. — Long
columnar root-stem , leaves spreading , flowers numerous,
pedunculated. Asiatic and Australian.

1. Asiaticum. — Loculis monospermis. Cells with one
ovule. Native of Asia. Perianth white ; end of
the style and filaments red.

Var. 1. Toxicarium. — Roxb. Hort. Beng. Asiaticum.
Bot. Mag. 27. 1073. Leaves acute, tube under
3 inches, limb rather longer; style and peduncle
short. A tender plant, and one of the least orna-
mental. A great bulb of this plant surrounded
with offsets was brought from Florida by Fraser,
who assured me that he had dug it up on the sea-
side. It must have been washed on shore from
some ship that was bringing it from the east.

Subvar. Minor. — Ex Sumatra. A small variety sent
from Sumatra by Sir Stamford Raffles. Leaves
acute.

Var. 2. Bracteatum. — Subvar. 1. Brevifdlium. Bot.
Reg. 4. 179. Jacq. H. Sch. 4. 495. Brevifolium.
Hort. Beng. This plant is a native of Mauritius,
very like Toxicarium in flower, but smaller, and
with very blunt leaves. It is not more hracteate
than other species, but the name, Bracteatum, has
unfortunately the priority over one more appro-
priate.

Subvar. 2. Angustius. — An equally small plant, with
much narrower leaves, but not less obtuse. Its
natural abode is not known, but it had been in the
stoves of this country, long before the introduction
of the former. It is probably the Sundeep Island
variety, which is of easier cultivation at Calcutta
than the former ; and so is Angustius here. Both
have cells with one ovule, and are closely allied to
Toxicarium. I have seen the three ovules of this
plant fertilized, one taking the lead and growing
into the other cells, turning the umbilical cords
which were between them to the outside, and by
r 2

244

AM ARYLLIDACEvE.

the pressure becoming’ inseparably united with the
two weaker ovules, forming one large pear-shaped
mass, with three irregular furrows and a sharp-
pointed trifid base.

Va r. 3. Anomalum. — Herb. App. Bot. Mag. 47. 2121.
p. 5. Plicatum. Bot. Mag. 56. 2908. This plant
is a native of China. It is very remarkable, be-
cause it differs from the whole genus in having the
leaves split on opposite sides alternately to the base
of the bulb, so that the bulb is imperfect in all its
coats. It approximates in the inflorescence to
Toxicarium and Bracteatum. The leaves have a
strange disease, being, as far as I have seen, always
more or less plicate, the plication being sometimes
quite monstrous. This will of course be more
conspicuous in some seedlings than in others, and
it is less observable in the growth of some seasons
than of others in the same individual. The tube is
usually 1^ inch long, limb 2|, filaments about an
inch shorter; style ^-li inch longer than the tube.
The leaves are acute. The name Anomalum has
the priority.

Var. 4. Declinatum. — Bot. Mag. 48. 2231. This plant
has also cells with one ovule, and is closely allied
to Toxicarium, but remarkable for the singular
manner in which the peduncles detach themselves
successively from the cluster and become horizontal,
and for the bud nodding, which contravenes the
habit of the section, and shews that the division of
the inclined and the nodding buds is not decisive.
It is a large plant, native of Silhet, with the tube
and limb longer than Toxicarium, the outside of
the sepals mottled with red.

2. Sinicum. — Roxburgh. Bot. Mag. 47. 2121. p. 7.
Foliis 3i pedalibus saturate viridibus 5 unc. latis
margine undulatissimo, scapo tripedali, germine
unciali breviter pedunculato, tubo 4-4! unc. limbo
4 unc. albo, filamentis stylo longioribus limbo bre-
vioribus. Native of China, from whence it was
received by Dr. Roxburgh in 1809.

Subvar. (fortuita) Variegata; foliis pulcherrime varie-
gatis. This variety, with beautifully variegated

AMAIIYLLIDACE.E.

245

leaves was sent to me by Dr. Carey, but I have lost
it. He informed me that a large caterpillar is apt
to eat its way into the column of this and other
species, and, by the admission of wet into the heart,
the plants become variegated through disease, and
often die: while in that state I believe they are apt
to produce variegated seedlings. The plant I had
was very big, grew vigorously with the larger por-
tion of the leaves yellowish white, and did not seem
diseased. It was however very sensitive to any wet
lodging amongst the leaves, and was killed by it.
I have not seen Sinicum in flower, and have no cer-
tainty of the number of its ovules, but I have a
large drawing of it by an Indian artist, which was
sent me by Dr. Carey.

3. Procerum. — Bot. Mag. 53. 2684. According to

Dr. Carey, this is the tallest and largest species of
Crinum, but it does not manifest its superior size
with me, and does not seem to enjoy the confined
air of the stove. It is a native of the neighbour-
hood of Rangoon, and I entertain no doubt of its
being one of the parents of C. amabile. This
plant increases by the splitting of the columnar
bulb into two equal parts, like the Australian
family. Its limb is tinged on the outside with
red, and the flowers are conspicuous. I neglected
to examine its ovules, though indeed it was in
flower here this season ; but if Sinicum and Pro-
cerum should prove to have one-ovuled cells, the
above six might be considered as local varieties of
Asiaticum.

4. Amabile. — Bot. Mag. 53. 2684. Superbum. Hort.

Beng. Hybridum spontaneum. Cultivated for its
beauty in Sumatra, from whence many bulbs of
it were sent by Sir S. Raffles, and where it is
said to be found in the interior. Neither Dr.
Carey nor Sir S. Raffles ever knew it produce
seed. Here it is not only barren, but its pollen
is dry, its anthers shrivelled, and it fails to im-
pregnate any other species. It is decidedly a
spontaneous cross, probably between Procerum
and Zeylanicum.

246

AMAUYLLIDACEiE.

5. Augustum. — Bot. Mag. 50. 2397. Bot. Reg. 8. 679.

Hybndum spontaneum. Received at Calcutta from
Mauritius, where it is said to be indigenous. Bar-
ren like Amabile, and apparently a cross be-
tween Bracteatum and Zeylanicum. It is of
smaller stature than Amabile, with more obtuse
leaves. It is much to be wished that the natural
production and reproduction of these two remark-
able plants were investigated. We have no cer-
tainty whether they are frequently produced, or
whether the cultivated plants are all offsets from
one respective original.

6. Australe. — Foliis laevibus sparsis, apice cucullato,

scapo multifloro, loculis (quad novi) triovularibus,
perianthio albo ; stylo saepissime filamentis bre-
viore.

Var. 1. Pedunculatum. — Bot. Reg. 1. 52. Brown,
Prod. 1. 297. Taitense, Red. Lil. 408. Australe.
Donn Cat. Foliis obtusis, germine crasso, tubo
4 unciali, stylo tubum unciam superante. Leaves
very blunt ; germen thick, tube 4 inches, style an
inch longer; column greenish white.

Var. 2. Rubricaule. — Columna et scapo purpurascenti-
bus, germine crasso, tubo viridi 2| unc. limbo sub-
aequali (vix triunciali), laciniis albis usque ad |-
latis, stylo et filamentis superne purpureis, stylo
tubum unc. superante, filamentis unc. breviore.
Column and scape purple, tube 2| long, limb
scarce 3 inches— style an inch longer than the
tube, an inch shorter than the filaments,
v Var. 3. Exaltatum. — Bot. Mag. 47. 2121. p. 5. Co-
lumna non colorata 15 unciali, foliis 4| pedalibus,
3-3i unc. latis, scapo viridi 33 nncias coluinnam
superante, floribus (apud me A. D. 1819, 40.) odo-
ratis, pedunc. uncial, germine gracili (media parte
hand multum crassiore) tubo viridi 4 unciali,
limbo 2g-2f unc. laciniis albis sub-|- latis canalicu-
latis, stylo superne purpureo tubum unciam su-
perante, filamentis apice purpureis semunc. stylo
longioribus. Leaves narrower, more acute than
pedunculatum ; germen slenderer than in any
known Crinum, unless perhaps Flaecidum ; tube

AMARYLLIDACEiE.

247

4 inches, limb 2|, or less. Purchased by me at
Mr. Evans’s sale above twenty years ago. Local
habitation not ascertained, but evidently belonging
to the Australian family, which extends into the
islands of the Pacific. Amongst the loftiest of the
genus.

Var. 4. Canaliculatum. — Roxburgh. Bot. Mag. 47.
2121. p. 5. Columna 9-10-unc. pallida vix pur-
purascente, foliis 3s pedalibus arcuatioribus, scapo
16-26 unc. pedunc. uncial, vel ultra, germine
crasso i-unc. tubo 21-21 unc. limbo 3i-'3b unc.
stylo nunc brevissimo, nunc Is tubum superante,
saepissime filamentis f unc. breviore, filamentis 1|
limbo brevioribus. The shortest of the family
with leaves acute and more arched, with a shorter
tube and longer limb 1 purchased two bulbs of
it at Mr. Evans’s sale, and have freely distributed
seedlings from them, but have never known it since
imported. Precise local habitation not ascertained.

It appears evident that these four natives of Australia and
the extreme south-east, although they preserve their peculi-
arities when raised from seed, may be conveniently united
under one general title, preserving the subordinate names to
distinguish them. I saw a few years ago a variety in flower
at Colvill’s nursery, which seemed to exceed them all in
general stature, and in the size of the flowers, but I did not
make any memorandum concerning it, and I cannot say with
certainty whether it differed materially, but I believe not. It
was there called, from its large stature, a white Amabile, but
I could not learn from whence it had been received. I con-
sidered it to have belonged to the Australian family, and to
have perhaps found its way there from Kew. If it be forth-
coming, and has triovular cells like the rest, it may be called
Australe, Maximum, var. 5. The illiberal system esta-
blished at Kew Gardens by Sir Joseph Banks, whereby the
rare plants collected there were hoarded with the most nig-
gard jealousy, and kept as much as possible out of the sight
of any inquirer, led in the first instance to a feeling of satis-
faction, whenever it was known that the garden had been
plundered, and some of its hidden treasures brought into
circulation ; and the indifference with which such thefts
were regarded, if they were not actual! v winked at, by

248

AIM All YLLI DACE-iE.

cultivators, led to such great laxity of conduct, that, until
the practice was stopped by a prosecution, every private
collection became exposed to like depredations ; and the
falsehoods, that were told to cover the theft, occasioned a
great deal of confusion concerning the native habitation of
plants introduced at that period. It was the narrow-minded
doctrine of Sir J. Banks, that he could only render the king’s
collection superior to others by monopolizing its contents;
and by so doing he rendered it hateful and contemptible:
whereas, if he had freely given and freely received, and
made its contents easily accessible to those who were inter-
ested in them, it would have been a pleasure and a pride to
the nation. It is now near twenty years since I have visited
that odious and useless establishment. Formerly I went
there often, but always in vain, for if I inquired for any rare
plants, which 1 had reason to believe were in the collection,
excepting those which, from their size, could not be con-
cealed, my conductor always denied any knowledge of them ;
and, if I asked whether I could speak to a person better ac-
quainted with the plants, I was told that I could obtain no
further information. The multitude of rare plants that have
flourished and perished there unobserved, I believe to be very
great. I owe no thanks to that establishment, but for the
mere permission to walk straight forward through the houses.
I must do Mr. Salisbury the justice to say, that be repeatedly
remonstrated with Sir Joseph Banks in vain on the subject.

The size of C. pedunculatum and exaltatum is limited, by
the column splitting into two equal parts when it has ac-
quired its full growth. That is their mode of increase both
here and in cultivation at Calcutta, nor have I ever known
them produce a subsidiary offset. I have had Canaliculatum
above twenty years, but I have not found it increase in any
manner except by seed, which they all bear pretty freely.

2. Bulbo b reciter coluvmari ; foliis suberectis ; uvibelld
numerosd subsessili. Asiatics. — Bulb short-columnar;
leaves suberect ; Jloivers numerous , subscssile. Asiatic.

7. Rigidum. — Foliis rigidulis, floribus albis.

Var. 1. Sumatranum. — Bot. Reg. 13. 1049. A very
remarkable plant with rigid leaves, that give it the
appearance of a Yucca. It grows to a great size,
though the colump is short, arid, it is not delicate,
but it yields no increase, and is unwilling to flower,

Route H

AMARYLLIDACEiE.

249

seeming to require a great heat to make it blossom,
or perhaps more water than it has been thought
prudent to give it. Seeds very large.

Var. 2. Macrocarpon. — Carey. Bot. Mag. 48. 2231.
p. 2. Obtained by Dr. Carey from the neighbour-
hood of Rangoon, but lost soon afterwards. Known
to me only by Dr. Carey’s letters. Foliis longis
rigidis angustis, perianthii laciniis albis latioribus
quam in C. toxicario, seminibus maximis 2 uncias
latis. Leaves long, rigid, narrow ; segments of the
perianth white, wider than those of toxicarium ;
seeds two inches wide — The description makes its
close affinity to Sumatranum evident.

B. Nutantes. Buds nodding.

1. Bidbo breviter columnari, umbelld pauciori- flora sessili.
Bulb shortly columnar ; flowers fewer, sessile.

8. Mauritianum. — Loddiges Bot. Cab. G50. Foliis

tripedalibus glabris acutis suberectis apice pendulo,
scapo breviore, floribus 4 (vel ultra?) albis rubro
apiculatis, stylo staminibus superne rubris longiore.
This plant was imported many years ago by Mr.
Loddiges from Mauritius, of which it was stated to
be a native. It approaches nearer to Brachyan-
drum than to any other, but is inferior in beauty.

9. Brachyandrum. — Bot. Mag. 47. 2121. p. 8. — 48.

2231. p. 2. Columna 5-unciali albicante, 3 unc.
diam. foliis 20, suberectis, 3-5-pedalibus, vix 2£
latis, glabris, saturate viridibus, obtusis uti folia
C. defixi ; scapo viridi, ultra-bipedali ; floribus 11
( apud me; 20-30. Carey), 7 simul nutantibus, csete-
ris paullo postea ; genuine subsessili, tubo 2f unc.
viridi, laciniis albis, 3J unc. § latis ; filamentis vix
unciam tubo longioribus, recurvatis, rubro apicu-
latis ; stylo unciam longiore, superne rubro, stig-
mate parvulo albo ; loculis dispermis. This re-
markable Crinum with two-seeded cells was raised
by Dr. Carey from seeds which he received from
Australia, certainly within the tropics, and sent to
me by him from Calcutta. It flowered at Spofforth
in 1821 and two or three seasons successivelv,
never producing more than 1 1 flowers, but died
without yielding any increase, and I fear it has

250

AMARYLLIDACE^E.

been lost at Calcutta also. It has no near affinity
to C. Australe.

2. Bulbo ovato breviter columnari, foliis humi divaricantibus,
undulatis. Africaner Occident ales. Bulb with a short
columnar neck, leaves spreading on the ground, undu-
lated. W. Africa.

10. Purpurascens. — Ex insula Fernando Po. Afric. occ.

Bulbo parvulo ovate cylindrico, foliis ultra 20
undulatissimis subscabris obscure viridibus 28 unc.
vix unc. latis porrectis apice acute producto arcua-
tim depresso, umbella 2-9 -flora, scapo purpura-
scente, spatha biuncial i purpurea marcescente,
germ, subsessili, perianthio ante expansionem nu-
tante saturate purpureo, dein extus pallide purpu-
reo intus albo, tubo unciali, limbo 2^-2|,

laciniis £ latis, filamentis superne purp. stylo pa-
rum longioribus, antheris subfuscis polline aureo,
stylo purp. tubum 2^ unc. superante, stigmate
parvo purpureo. The flowers of this singular little
African species are 8 or 9 in number, deep purple
before expansion, but the colour fades as they open,
and they are white within. It is of easy culture
and increases fast by offsets, but it is not stolonife-
rous. The tube is usually above twice as long as
the limb.

3. Bulbo breviter columnari stolonifero , foliis sub-arcuatis,
floribus sub-octo. Occidentales. Column short, stoloni-
ferous , leaves rather arched, flowers sessile, not exceed-
ing eight. Occidental plants.

11. Cruentum. — Bot. Reg. 2. 171. Two bulbs of this
fine plant (which is allied to erubescens, and
scarcely distinguishable as a species by its more
glossy and erect leaves of a darker green, and the
rich purple limb of its flowers, which never exceed
six) were purchased by me at Mr. Evans’s sale in
1814, and from them all the plants of the species
now in cultivation in Europe are descended. It
is certainly occidental, and native of a very hot
climate, but its exact abode is not known. 1 find
however in Dr. Hooker’s herbarium a five-flowered
Mexican Crinum with the tube 8 inches long and
the limb seeming to have been purple, which

AM AllYLLlDACEiE.

251

appears to agree with Cruentum except as to the
length of the style, which is a variable feature. It
is not accompanied by the leaf, but was gathered
in the month of August at San Bartolo in Mexican
Oaxaca by Mr. Andrieux (No. 74.), and I think
it must be a variety (probably a hardier one) of
either Cruentum or Loddigesianum. I have a
seedling from Cruentum, of which the flower tends
a little more to white.

12. Erubescens. — Folia arcuata margine scabro, spatio
inter nervos medios apiculato, perianthio tubuloso,
antheris cinereis, polline aureo. Leaves arched,
margin rough, with a point prolonged from the
space between the two mid veins, tube long, an-
thers dusky, pollen golden.

Var. 1. Rubrilimbum. — Bot. Mag. 30. 1232. foliis rec-
tioribus 2| unc. latis, limbo extus rubescente.
Leaves straighter, limb red outside.

Var. 2. Majus. — Foliis altioribus 22-3-pedalibus, flori-
bus 6-7 majoribus, limbo albo. Leaves larger,
flowers 6-7 larger, white. Cells with five ovules.

Var. 3. Minus. — Foliis latis humilioribus, floribus 6,
limbo albo. Smaller variety with six white flow-
ers ; leaves broader and less erect than var. 1. This
was the variety chiefly known in our stoves
twenty years ago.

Var. 4. Corantinum. — Foliis rectioribus, superficie la-
cunose subradulosa, floribus 4 vel ultra limbo lato
albo, revoluto ; columna vix purpurascente. Co-
lumn greenish; leaves straighter and stifler than
2 and 3, much pitted on the surface ; flowers 4
or more ; limb broad, white, and revolute. Intro-
duced by Lee and Kennedy from the banks of the
Corantyne river in Dutch Guiana, I believe in
Berbice.

Var. 5. Braziliense. — Reliq. Willd. Minimum, foliis hu-
milibus, floribus 4 vel ultra albis revolutis. In
inuudatis prope Rio Janeiro. A small variety with
about four revolute white flowers, found by my
collector, as described by Willdenow, growing in
water (he called it a pond) near Rio. Certainly

252

AMARYLLIDACE^E.

the mother of C. submersum, which was found in
company with it. It was rather impatient of pack-
age and cultivation, and the bulbs I had are all
lost.

Var. 6. Caraccense. — Foliis humilibus 17 uncialibus, 2 g-
latis, margine scaberrimo, floribus albis revolutis ;
praecedenti valde affinis. Prope Caraccam. A
small variety from Caraccas closely allied to var. 5,
but of a stouter constitution ; not half so big as
var. 3.

Var. 7. Octoflorum. — Bulbo ambitu 14 unciali, foliis
2^ pedalibus, floribus 8 subsessilibus, genuine un-
ciali crasso ovali viridi, tubo viridi 4-5J unciali,
filamentis apice rubris limbo 2 uncias brevioribus,
stylo subaequali vel longitudinis incertae. Ex Vene-
zuela. A fine variety lately imported from the
Spanish Main by W. Gordon, Esq. of Haffield,
affording the only instance yet known of an occi-
dental Crinum of the patent narrow- limbed section
producing eight flowers. In all the varieties the
flower is more or less revolute at first, at last flaccid
and pendulous.

The stoloniferous Crinum erubescens, bearing usually six
flowers, but often four in the lesser varieties, and eight in
Mr. Gordon’s plant, occupies nearly the whole tropical east-
ern coast of South America. It has a remarkable point to the
leaf, which is easily recognised, and before inversion the
anthers are of a blackish ash colour.

13. Lindleyanum. — Ex Surinam et Maranham. Eru-
bescens v. viridifolia. Herb. App. Bulbo non pur-
purascente ovate columnari, foliis lucidis sesqui-
pedalibus vel ultra, unciam latis vel ultra, margine
vix sub lente scabriusculo, lineari-lanceolatis arcua-
tim porrectis (tubo 6-unciali viridi, limbo revoluto,
denique pendulo ; — extus purpurascente ?) I can-
not distinguish Dr. Lindley’s revolutum, Hort.
Soc. Tr. vi. p. 284, from erubescens, by any thing
in his description except greater flatness of leaf.
The name revolutum, having been long pre-occu-
pied by another species, cannot remain with it. I
believe the plant to be the same 'as those I possess
from Surinam and Maranham, which I formerly

AM ARYLLIDACEvE.

253

designated as the glabrous green-leaved variety
of erubescens. I think it may be separated as hav-
ing, besides a smooth shining leaf, a more ovate
bulb, which is rather disposed to rest from vegeta-
tion in winter, being in habit intermediate between
erubescens and Americanum. I propose to call it
Lindleyanum ; whether as a variety of erubescens,
or as distinct, may be determined by further ex-
amination. It does not flower freely, probably re-
quiring great heat; and, although I have possessed
it many years, I have not sufficiently attended to
its inflorescence. I describe the leaves from my own
plants; the flower from Dr. Lindley’s report, under
a doubt whether his plant be this, or a variety of
erubescens closely akin to Braziliense.

14. (Jndulatum. — Hooker Ex. Flor. 200. This plant I
have never seen, and cannot learn where it is to be
found. I did consider it to be one of the varieties
of erubescens, but, not having seen its bulb and
foliage, I can pronounce no decided opinion.. Its
flowers are four, white, with the tube 8 inches
long.

4. Bulbo ovato vel spheerieo stolonifero , folds suberectis,
umbella sessili 4 -fiord. Bulb ovate or sphcerical, stolo-
niferous ; leaves suberect ; flowers four , sessile. Occi-
dental plants.

15. Loddigesianum. — Folia non undulata, suberecta,
2i pedalia, oblique apiculata, saturate viridia, mar-
gine scabriusculo ; scapus 2-uncialis; spatha tubata
(an ita semper?) tubus viridis, 8-uncialis, erectus,
superne curvatus ; limbus 3-31 uncialis, apice sa-
turate purpureo, extus purpura saturate striatus ;
filamenta cum stylo fere sequali saturate purpurea
basi albicantia, 4 declinata, 2 aversa. Ex Mexico.
Tube 8 inches, limb tipped with dark purple and
striped with the same on the outside. Imported
from Mexico by Mr. Loddiges.

16. Strictum. — Bot. Mag. 53. 2635. This plant is
intermediate between Loddigesianum and Ameri-
canum, and I make no doubt of its being a Mexi-
can plant, from whence Mr. Tate, who sent it to
me as it arrived (by some mistake labelled as a

254

AM ARYLLID ACE,®.

Neottia from Ceylon), imported many plants at the
same period. It has no oriental affinities. The
name has been altered by R. and Schultes, on ac-
count of C. Strictum occurring in Hornemann’s
catalogue described as having leaves longer than
the scape, whereas my strictum had leaves shorter.
This is a very variable feature, even in the same
bulb at different times, and unfit to make a specific
separation ; and I find nothing in Hornemann’s
brief description to mark that his plant may not be
the same ; but at all events I cannot admit Horne-
mann’s plant at all, for the quarter from whence it
was derived is not stated, and it is very likely one
of my own hybrid productions, which were freely
distributed, and the description is insufficient for
a possibility of its being ever recognised. It is a
public nuisance to load our books with vague and
uncertain species. Rejiciendum est C. strictum
Hornem. habitatione incerta, diagnosi imperfecta;
forsan supradicti varietas, forsan C. Defixum aut
Commelinianum vel hybridum hortense.

17. Americanum. — Bot. Mag. 26. 1034. A well known
plant with four white flowers exquisitely fragrant.
This section might be considered as varieties of
one Crinum Americanum, giving the subordinate
name odoratissimum to this particular plant; and
Cruentum might be properly joined w ith Erubes-
cens, under the name C. occidentale. I raised a
lovely mule between this plant and the narrow-
leaved Bracteatum.

18. Commelinianum. — Jacq.H.Sch. 2. 202. This plant
has never been brought to Europe, since its first
introduction into the garden at Schcenbrun, and its
precise locality is not ascertained. It wras brought
to this country by iny brother, the late Earl of
Carnarvon, in his carriage, from the garden of the
Grand Duke of Tuscany, being the produce of
Jacquin's plant. It increases freely with me by
offsets and seeds, preserving its distinct characters,
and sometimes bearing only two flowers.

5. Bulbo elongato stolonifero foliis erectis umbella 6-flora,
Asiatics. Bulb elongated, stoloniferous, leaves erect,
flowers six. Asiatic.

AM ARYLLIDACEiE.

255

19. Defixum. — Bot. Mag. 48. 2208. Asiaticum. Rox-
burgh. Belluta pola tali. Rheede. Mai. p. 75. This
little Crinum grows in the wet ditehes near Cal-
cutta, and its bulb, sinking deep in the mud, is
elongated like a ruler; and Dr. Carey, who sent
it to me, could not recognise its ovate green bulb in
the engraving, which is the effect of cultivation
above ground. It will flower immersed in water,
but does not require it, and increases freely by
runners. The seed is longer than usual in the
genus.

20. Ensifolium. — Bot. Mag. 49. 2301. Cells with two
ovules. A native of Pegu, closely allied to Defixum,
with more acute leaves and flowers redder on the
outside, and of a much more delicate constitution.
Dr. Carey had lost it at Calcutta, and I sent back
to him, a year before his death, one of two bulbs I
possessed. This and the one preceding it might
be united under a superior name, C. Scrobense, or
ditch Crinum.

21. Pusillum. — PI. 32. fig. 3. Specim ex insulis Nico-
bar. Herb. Linn. Soc. et Herb. Banks. Bulbo co-
lumnari cylindrico 4 unc. vix. § diam. foliis acutis
9 unc. * latis. pedunculis 1J unc. genuine gracili
| unc. tubo et limbo triuncialibus, stylo et fila-
mentis subaequalibus, plus unc. limbo brevioribus.
1 conjecture that this singular little plant must be
a stoloniferous ditch Crinum like Defixum, and that
its little columnar bulb would contract its length if
cultivated above ground : it differs from the two
last in having long peduncles.

6. Bulbo sphcerico vel ovato, foliis semirecumbentibus. Bulb
sphcerical or ovate ; leaves half -recumbent.

22. Amoenum. — Roxburgh. M. S. cum tab. pict. Ker J.
Sc. 6t A. bulb round, without elongated neck ;
leaves 1-2 feet, 1^ inch wide, a little channelled;
margin slightly scabrous ; flowers white, 4-8, ses-
sile, tube 3-4 inches, limb about equal. From Silhet.
Found also in Nepal and the Birman empire. Cells
with 5 ovules.

Var. 2. — Caudiceum. — From Ceylon. Bulb round with
an elongated neck ; leaves gradually attenuated ;

256

AMAIIYLLIDACEAE.

flowers 10, subsessile ; tube 3|, of which 2^ are
persistent; limb 3 inches; style j an inch ; filaments
an inch shorter than the limb ; cells with 5 ovules.

These plants increase by the bulb splitting in halves.
They are very thirsty while growing in heat, and I think
flower most freely when they are left dry in the winter. Dr.
Carey sent me a bulb he had received from Rangoon, which
he had named Verecundum, stating that, though it approached
very near to Amoenum in foliage, it was distinct in its inflores-
cence. I have not seen its flower, and am not satisfied of its
diversity, or sufficiently acquainted with its peculiarities, to
think it right to give it a separate place as a species.

Var. 3. Verecundum. — Amcenum, var. 2. Herb. app.
Bulbo sphaerico foliis humifusis saturate viridibus,
lacunosis, obtusioribus, margine laevi.

23. Humile. — Bot. Mag. 53. 2636. Leaves short, acute,
pitted ; flowers pedunculated. A very small spe-
cies from the extreme East ; brought to England in
an Indiaman by Capt. Craigie, but it was not ascer-
tained where it was taken in. Not known at Cal-
cutta.

24. Pratense. — Bulbo ovato vel sphasrico, foliis loratis
flaccide recumbentibus, umbella sessili multiflora.
In pratis siccis prope flumina proveniens, inunda-
tionis annuae tempestate florens umbella natante.

Var. 1. Longifolium. — Roxburgh. Ker J. Sc. and A.
Leaves 2-3 feet long, 2 inches wide ; flowers sub-
sessile 8-12, white; tube 4 inches, limb a little
shorter ; filaments and style nearly equal with the
limb; anthers before inversion brown. Bengal.
Floruit Altaclarae in Britannia A. D. 1822, floribus
10 ante expansionem nutantibus, spatha majore non
marcescente, germine subsessili, in medio umbellae
semunciam pedunculato, tubo viridi graciliore q.
in lorifolio, 4 unciali, laciniis 3 unc. vix semunc.
latis acutioribus, stylo filamentis et limbo breviori-
bus superne rubris f unc. longiore, debili, recum-
bente, rubro, stigmate minus quam in lorifolio
fimbriato, antheris ante inversionem fuscis, scapo
rigidiore erecto, germine longiore crassiore, um-
bella diflfusiore.

_PLaie.l3.

AMARYLLIDACE-E.

257

Var. 2. Lorifolium. — Roxburgh. Ker. J. Sc. and A.
Leaves 2-5 feet long, scarcely distinguishable from
longifolium, the margin of both a little scabrous.
Flowers sometimes 20, less pedunculated than in
longifolium (which is contrary to Dr. Roxburgh’s
observation). Pegu. Floruit. A.D. 1822. floribus
8, spatha parvula marcescente, pedunculis brevi-
bus, tubo viridi 3 unciali, laciniis 2 b unc. longis,
| unc. latis, filamentis albis lacinias eequantibus,
antheris fuscis, stylo superne rubro, stigmate con-
spicue fimbriato trifido, bracteis fugacioribus, scapo
recumbente umbella erecta conferta, genuine vix
pedunculo crassiore.

The following particulars of the inflorescence of longifo-
lium and lorifolium will establish their diversity, though
they can scarcely be distinguished from each other when not
in flower: —

Lorifolium.

Scape reclining quite horizontally.

Umbel turning up at right angles to
the scape or nearly so, compact, 8-flow-
ered, buds nearly erect.

Peduncles extremely short, except in
the middle of the umbel ; becoming
afterwards more elongated than in lon-
gifolium.

Germen scarcely thicker than the
peduncle.

Bractes small, very slender, thread-
like, fading when the umbel expands.

Spathe small, soon withering.

Tube green, 3 inches long.

Limb 2 J inches long ; segments J
inch broad, blunt at the point, white.

Filaments equal to the limb, pure
white, conniving.

Anthers brown ; stigma trifid, con-
spicuously fimbriated.

Style equalling the filaments, erect,
upper half red.

Longifolium.

Scape more rigid, upright, less com-
pressed.

Umbel more diffuse, 10-flowered,
buds nodding.

Peduncles extremely short, some be-
coming afterwards half an inch long,
other germs remaining nearly sessile.

Germen longer and thicker than in
lorifolium.

Bractes slender, green, more per-
sistent.

Spathe larger, firmer, more durable.

Tube green, slenderer, 4 inches
long;

Limb 3 inches long ; segments not
quite half an inch wide, more actite,
white.

Filaments shorter than the limb,
slender, straighter, erect; upper half
red.

Anthers brown ; stigma not conspi-
cuously fimbriated.

Style f ths of an inch longer than
the filaments, slender, recumbent, all
red.

Var. 3. E'legans. — Bot. Mag. 52. 2592. from Rangoon.
Closely allied to the two former, with flowers less

s

258

AMARYLLIDACEjE.

disposed to be pedunculated, and a more elongated
bulb ; scape recumbent. In longifolium the style
is red, the filaments tipped with red, in the two
others the filaments white and the style tipped with
red. The tube of elegans is 2i inches, and shorter
than the limb which is 31 ; in longifolium the tube
4 and the limb only 3, in lorifolium the tube 3 and
the limb only 21. From Rangoon. It increases
by the bulb splitting in halves.

Var. 4. Venustum. — Carey. Herb. App. From Silhet.
Leaves more obtuse than elegans; flowers about 30,
entirely white, about the size and form of those of
longifolium.

Var. 5. Canalifolium. — Carey. Herb. App. Leaves an
inch wide, more channelled than the three former,
rather tending to glaucous ; flowers about 10, pale
red on the outside.

The two former grow in meadows near the great rivers
of India, on sound turf, liable to inundation, and in the flood
season their leaves are elongated, and they flower with the
umbel floating like a water-lily. It is evident from the
sloping scape of elegans that its habits are similar, and I
have no doubt that the two others have the same. They are
difficult to cultivate, because they are impatient of continued
moisture, and it is difficult to accommodate them with the
luxury of a dry sod and a seasonable inundation. They are
so evidently akin to each other that I thought they might be
conveniently brought together under the superior name of
Meadow Crinum. They will not long exist in light earth,
and should be planted in rich and pretty strong alluvial
soil.

25. Erythrophyllum.- — Carey. Bot. Mag. 47. 2121.
p. 7. Ex Rangoon, foliis saturate sanguineis.
Dr. Carey lost this remarkable plant without hav-
ing seen its flower, and vainly attempted to obtain
it and Macrocarpon again. There is no doubt of
its being a distinct species, but I cannot tell what
are its affinities. Leaves as red as those of a red
cabbage.

26. Gracile. — Meyer Rel. Hank. p. 120. Ex insula
Luzon. Known only by Meyer’s description from
a Manilla specimen. Leaves lanceolate, 1^ foot

AMARYLLIDACEiE.

259

long, 1^ inch wide, attenuated at both ends ; scape
a foot; spathe obtuse, 1^ inch; flowers 6-7 ; tube
more than twice as long as the limb ; filaments ^rd
shorter than the limb, rather declined, recurved.
Perhaps allied to the following Australian group.

27. Australasicum. — Bulbo spheerico, foliis crassis rigi-
dulis, filamentis basi gibbosa, germine sessili, sub-
sessili, vel breviter pedunculate).

Var. 1. Arenarium. — Bot. Mag. 49. 2355. Leaves with
the margin slightly scabrous, peduncle long,
tube near 4 inches, pale green, limb white 2|; fila-
ments and style purple upwards ; style longest,
shorter than the limb. Cells with six ovules.

Var. 2. Blandum. — Arenarium ft. Bot. Mag. 49. 2535.
Leaves larger, peduncles longer, limb broader, blush-
coloured without, filaments white, style red up-
wards, much longer, stigma much smaller than in
arenarium.

These were found growing in Water Island, lat. 14° 3'
south long. 150° east of Greenwich, 10 inches below the sur-
face of a barren sandy soil ; thermometer in the shade, sea-
breeze blowing, 94° Fahr. ; heat of the sand 130° near the
surface.

Var. 3. Confertum. — Bot. Mag. 51. 2522. Leaf nar-
rower than the former, with a smooth margin,
flowers sessile, cells with 5-8 ovules, tube erect,
3 inches long, shorter, red, white at the base. Sent
to England under the name Angustifolium.

Var. 4. Angustifolium. — Brown Prodromus. 1. 297.
Foliis margine scabro. I am not aware from Dr.
Brown's short description that it differs from con-
fertum in anything but a scabrous margin to the
leaf, but on comparison of the leaf of confertum
with Dr. Brown’s specimen that difference was
very apparent. I have seen no bulb of the true
angustifolium.

28. Venosum. — Brown Prodromus. 1. 297. Germen
subsessile, tube twice as long as the limb, segments
strongly veined elliptic-lanceolate, filaments half
the length of the limb, anthers equal to the fila-
ments ; leaves unknown. This plant being insuf-

s 2

260

A MARYLLID ACE^E.

ficiently described, I cannot determine whether it
is allied to the foregoing group, or to 9. Brachyan-
drum.

§. 2. Semipatentia. — Half-patent.

A. Foliis scepius undulatis, umbelld l -'20-flord, laciniis latis,

germ. sess. vel subsess. — Leaves usually undulated; umbel

1 -20 -flowered, segments broad ; germen sessile or subsessile.

29. Broussonetianum. — Bot. Mag. 47. 2121.

Var. 2. Pluriflorum. — Parad. Londin. 52. C. Yuccse-
florum.

Var. 3. Yuccseides. — Foliis non undulatis. Spec, in
Herb. Soe. Linn.

The first of these produces usually a solitary flower, with
a fine purple stripe on each segment, the second two or more;
the third variety, which is more delicate, has the leaves not
undulated, the flowers two or more. They are all natives of
Sierra Leone, and not materially distinguishable. It is ad-
visable to leave them dry in the winter. Probably from
some constitutional difference I failed in one or two attempts
to cross this species with Capense, which shews no disincli-
nation to breed with any Crinum of Asia or America.

30. Distychum. — Am. Ornata. a. Bot. Mag. 31. 1253.
This little plant, a native also of Sierra Leone,
differs from all the rest of the genus in having
leaves not multifarious, but alternating to the right
and left; and, contrary to the prevailing habit of
this section, they are not undulated. It produces
usually a solitary flower, striped with red, and is
allied to Broussonetianum.

31. Petiolatum. — Foliis latis petiolatis, costosis, nervis vix
parallelis ex costa oriundis, floribus albis, umbelld sub -
8 -fiord.

Var. 1. Spectabile. — A. ornata. ft. Bot. Mag. 923. C.
Giganteum And. Rep. 169. This fine species with
broad and large conspicuous white flowers, is a
native of the tropical west coast of Africa. The
name, giganteum, has the priority, but it would be
disgraceful to science to retain the name gigantic
for a species of inferior size. Such names, as this
and Thunbergia fragrans for a scentless plant,
must be altered, being much more objectionable

Kah.lL.

AM AR YLLI DACEjE.

2G1

than those rejected by Linnaeus as too complex ;
hut it is only where a specific name announces an
untruth that I would countenance its rejection.
The name spectabile occurs in Lodd. Cat. and else-
where. I understand that it grows in ditches in
an island, called, I believe, St. Thomas, on the
west coast of Africa.

Var. 2. E'xtorre. — Foliis sub-angustioribus, nitentiori-
bus, erectioribus, undulatissimis. Leaves rather
narrower, more shining, erect, and undulated.
Found by my collector growing near the sea in the
neighbourhood of Rio Janeiro. It is a more thirsty
plant than the African, and requires the earth to
be always moist. I cannot tell whether it be a
variety from the African plant introduced into
Brazilian gardens and naturalized in the vicinity,
or an ancient native of the west.

Var. 3. Virgineum. — Martius ap. Schultes fil. Tube
shorter than the limb, lowest petal not divaricate.
Brazil, on the road near Mandiocca. This plant is
distinguished by a very variable feature in the short-
ness of the tube from the African plant, and having-
been found near the road-side, the same doubt may
arise whether it was growing either in the site of
an old garden, or where it had been thrown by
accident, or whether it is really indigenous in the
neighbouring country. It will be a circumstance
rather singular, if this native of tropical Africa,
distinguished by petiolation and a consequent dis-
turbance of the veining of the leaves, and which
seems less willing to breed with C. Capense than
the Crinums of other quarters, should manifest
itself naturally with these slight variations in Brazil
also, and there only. I have one young seedling,
which was raised after many failures from Capense
by Spectabile, but it is not yet arrived Jat an age to
flower. It is very singular that the Sierra Leone
kinds should be unwilling to cross, while all the
others seem ready to intermingle. The petiolated
Crinum has usually six flowers, in weak specimens
four, in stronger, if I recollect right, eight, per-
haps more.

262

AMARYLLIDACEjE.

32. Scabrum. — Bot. Mag. 47. 2180. — Bulb large, leaves
very long and glossy, of a lively green, with a sca-
brous edge ; flowers striped with bright red and
fragrant ; very beautiful. In or by the side of
upland woods near Rio Janeiro.

33. Submersum. — Bot. Mag. 51. 2468. Hybridum Spon-
taneum. A single bulb was found by my collector
in company with Erubescens (var.) Braziliense in
a natural pond or spot still flooded after a long
course of dry weather near Rio Janeiro. Certainly
a spontaneous mule from that plant impregnated
by scabrum. I have found it absolutely barren,
with very dry pollen ; its flowers are larger and
brighter than represented from its first flowering
before it was well established. After fourteen years’
cultivation it has just yielded one offset. It differs
very little from the mules raised from erubescens,
var. 3. but is rather inferior to them, as having pro-
ceeded from a less conspicuous variety.

34. Concinnum. — Martius ap. Schultes fil. Scape
green, 12-flowered; germen subsessile, tube scarcely
curved, 15-17 lines long; limb 2-2^ inches long,
segments 4-5 lines wide, sepals rather obovate,
petals acute, filaments declined (except the upper),
half as long as the limb ; style nearly equal to the
limb, stigma obtuse. Flowers in the dry specimen
violet-purple. Fresh colour and leaves unknown.
In the mountain fields of Minas Geraes, Brazil.
It is much to be wished that this plant could be
obtained from Brazil. This imperfect notice is the
only one of any occidental Crinum with an umbel
exceeding eight flowers.

35. Ornatum. — Bulbo sphaerico, caudice gracili, foliis
undulatis, umbella sub-viginti-flora, perianthio la-
ciniis latis extus striatis, loculis polyspermis. Bulb
sphaerical, column slender, leaves undulated, umbel
20-flowered or under, segments of the perianth
wide, striped without ; cells with many ovules.
Asiatic.

Var. 1. Zeylanicum. — Am.Ornata. Bot. Mag. 29. 1171.
Am. Latifolia. And. Bot. Rep. 478. Bulb and
column very deep red ; leaves very undulated, the

AMARYLLIDACE-iE.

263

young ones tipped with red, scape purple, 2|-3 feet
high, flowers 12-20, deep purple, declining on the
edges to white.

Var. 2. Herbertianum. — Wallich. p. 39. t. 145. Bulb
and column not red; leaves 14-2 feet long, 4 inches
wide, attenuatedly acuminate, much undulated,
with a secondary undulation on the margin, dark
green, very glaucous underneath (not so in the
plate), scape purple on one side, flowers 12-16 fra-
grant, deep purple edged with white. Flowers in
the rainy seasons in ditches near Rangoon in com-
pany with Zeylanicum.

Var. 3. Latifolium. — Bulbo et caudice uti in Zeylanico,
minus saturate rubro ; foliis undulatis margine un~
dulatiore, ortu rubro non apiculatis, scapo viridi
breviore, floribus Zeylanico pallidioribus. It may
be distinguished from Zeylanicum by the secon-
dary undulation of the margins of the leaves, and
the absence of the red tip to the young leaves, as
also by its green scape and paler flowers. The
plant figured in the Bot. Reg. is not Latifolium.
It is very scarce in the Calcutta gardens ; Dr.
Carey informed me a few years ago that he had
but two bulbs of it, and although it ripened seed,
they rarely vegetated with him, and when they
did, from the number of species collected together
in his garden, he was no longer sure of obtaining
the true plant. Bulbs of Zeylanicum, and others
not genuine, have been repeatedly sent to England
from the Botanic Garden under the name of Lati-
folium. Dr. Roxburgh omits in his description to
mention the style which is not prolonged as in the
plant figured in the Bot. Reg., but he says the leaf
tapers to a blunt point, which does not agree either
with that plant or with his own Latifolium, which
I have had in cultivation a great many years, and
still preserve. The plant compared with it under
the name Lineatum, by Dr. Roxburgh, as quoted
there, is C. Zeylanicum. Native of Bengal ; the
true plant will, perhaps, not be found now in the
gardens of Calcutta. Floruit. Spofforthioe, A.D.
1828. Scapo 16.| unc. basi purpurasc. germ, ob-
longo subscss. f unc. pallide viridi-purpurasc. flo-

264

AMARYLLIDACEiE.

ribus 6, tubo pallido, 3-unc. curv. limbo 4-unc.
unguiculate acuminatissimo albo striis latis purpu-
reis ; stylo superne rubro, limbum non aequante,
filamentis longiore, polline pallido.

Var. 4. Longistylum. — Bulbo minore, caudice breviore
vix rubro, foliis humilioribus, minus arcuatis, apice
obtuso; perianthio albo, roseo-striato, tubo 3|,
limbo 3f unc. undulatissimo ; stylo rubro filamen-
tis limbo 1 ri, unc. longiore. Native of Bengal ;
differs from Moluccanum in having the style longer,
and the tube shorter than the limb, which is more
undulated, and the leaves obtuse.

Subvar. acutior. — Latifolium. Bot. Reg. 15. 1297. This
plant is a sub-variety of the bulb named by me
longistylum in Sweet’s Hort. Brit. ; distinguished
by little but the acute point of the leaf, whereas
the plant which I have cultivated a great many
years as longistylum has a blunter point than any
of the cognate bulbs. The aspect of these two
bulbs is quite different from that of Roxburgh’s
latifolium, which comes near to Zeylanicum in
general appearance ; these approach to Speciosum
and Careyanum.

Var. 5. Speciosum. — Bot. Reg. 48. 2217. Leaves less
undulated than those of Zeylanicum, and narrower ;
flowers 15-20, striped with pink, almost white
in the cold season. Flowers freely. Native of
Bengal.

Var. 6. Moluccanum. — Bot. Mag. 49. Tulipa Java-
nica Rumph. Amb. A smaller plant, leaves much
undulated, scape 1J-2 feet high, green, 5-6 flow-
ered. Increases so much by offsets that it rarely
flowers in the stove, and scarcely at all with us,
unless placed in a hotbed.

Var. 7. Careyanum. — Bot. Mag. 51. 2466. This beau-
tiful plant was brought to light by Dr. Carey, late
of Serampore, and I had the pleasure of naming it
after one of the best, the most amiable, gifted, and
indefatigable of men ; whose virtues and talents
adorned his country, and whose labours have pro-
moted the glory of the Almighty. I never saw
that excellent man, but fifteen years' correspondence

AMARYLLIDACEjE.

265

had accustomed me to look upon him as a dearly
valued friend. His life was devoted to the diffu-
sion of the Gospel ; horticulture, natural history,
and botany, afforded the brief recreation he allowed
himself from his daily toils. His favourite plants
were the Amaryllidaceous family, and to him we
are indebted for our knowledge of many of them.
He was born in 1761, at Hackleton, in Leicester-
shire, and embarked for India in 1793. In 1800
he was settled at Serampore, and he closed the
labours of his useful life in 1834, beloved by all
who knew him, honoured by all whom his name
has reached, having translated and superintended
the publication of the gospel in forty oriental
languages, which he had the perseverance to ac-
quire for that purpose. Born in the humblest cir-
cumstances, often uncertain of his daily bread, at
first a journeyman shoemaker, then a village school-
master, he had before his departure from England
taught himself to read the Bible in Greek, Latin,
Hebrew, French, Italian, and Dutch, and had be-
come conspicuous by his eloquent preaching, and
his ardent desire to bring about the mission to
India which originated in his powerful mind.
When he arrived there lie found it necessary to
offer his services by a hand-bill to make or repair
shoes, and after he had risen to the head of a flou-
rishing establishment, and occupied the chair of
three professorships, he was not ashamed to nail up
the original hand-bill against the wall of his study,
but took pleasure in considering from what a hum-
ble grade he had been lifted up to a metre useful
and distinguished station by the grace of God and
his own virtuous perseverance.

Var. 8. Insigne. — Am. insignis. Bot. Reg. 7. 579. A
very fine plant, introduced to this country from
Ceylon. It is of larger stature than any of this
family, with more glossy and less undulated leaves
than the rest. Dr. Carey had another kindred
bulb, which he called C. ornatum, with pink-
striped flowers, from Mauritius. He sent me a
bulb of it, but it perished during the voyage. I
have no particulars by which I can designate it.

266

AMARYLLlDACEiE.

( Species dubia. Lanceanum, mild Sweet. Hort. Brit, ex
Surinam. Nondum floridum, et forsitan C. Brous-
sonctiani ex Africa ibi delati varietas. Formosum,
Herb. Append, idem est cum Arenario blando var.
falso ex Brazilid allatum fuisse dictum .) C. Os-
beckii Desfontaines Cat. Hort. Par. habitatione
ignota non inter species recipiendum est : hybridum
verisimiliter hortense ex ineis ; forsan Scabro-Ca-
pense ? Rejiciendum est parili ratione Multi-
florum Desf.

B. Folds non undulatis, umbelld pedunculatd. — Leaves not un-
dulated; umbel pedunculated.

1 Perianthio patentiore ; Australes. — Flowers more patent ;

Australian.

36. Brevillmbum. — Bot. Mag. 47. 2121. p. 7. “ Foliis
rigidissimis attenuate acutis; limbo latobrevi primu-
lae speciem praebente, albo; filamentis rubris. Planta
magnitudine Crini erubescentis,” Carey MS. This
singular plant, with rigid and acutely attenuated
leaves, short broad white limb, like that of a pri-
mula, and red filaments, was raised by Dr. Carey
from seed received from Australia, probably from
the same unknown quarter as Brachyandrum. This
insufficient description does not satisfy me where
to place it; the description of the flower comes
nearest to that of Flaccidum, but the leaves would
suit rather with Sumatranum and Macrocarpon.
He did not state whether it was pedunculated or
not, but called it about equal to Erubescens in size.
It was unfortunately lost by him soon after its
flowering.

37. Flaccidum. — Bot. Mag. 47. 2133. Am. Austral-
asica Bot. Reg. 5. 426. This plant has seeds
smoother than those of any other Crinum I have
seen, and inclining to pearl colour, but I have
crossed it with both Asiaticum and Australe. It
succeeds in a cool part of the stove, and is also
capable of living in the open ground, where I have
had it some years in front of the stove, but it has
not flowered there. In a greenhouse it should be
left dry in winter.

AMARYLLIDACEjE.

267

2. Flore minus patente; loculis poly spermis. Africans meridio-
nales. — Flower less patent ; cells with many ovules. South
Africa.

38. Forbesianum. — Trans. Hort. Soc. 6. p. 83. Tabula
picta ap. Hort. Soc. Londini. Bulbo magno spliae-
rico, foliis loratis canaliculat.is glaucis ciliatis, um-
bella 30-40-flora, floribus lsetissime purpureis.

Var. (fortuita ) punicea. Ib. floribus la3te purpureo-
kermesinis magis revolutis.

This superb species, with from 30 to 40 purplish red
flowers, was found by Forbes on the banks of the Delagoa
river, on the south-east coast of Africa. The bulbs were very
large, but of difficult culture, and I fear that they may have
been all lost by rottenness. I believe the exposure of the
bulbs to our damp atmosphere has been the cause of their
perishing. I raised three seedlings from C. Careyanum im-
pregnated by it, but they were also very delicate. I lost the
one I reserved for myself, and I believe the others are dead
also. It is greatly to be regretted that no cross should have
been raised from C. Capense by this beautiful species, and if
any person in this country still possesses it, I hope he will
not miss an opportunity of attempting it.

39. Revolutum.- — Am. revoluta. A. Bot. Reg. 8. 623.
Bulbo ovato foliis angustis, lorato-linearibus, acute
attenuatis, humifusis, sestivis, autumno deperiun-
tibus, floribus roseo-striatis. In dubio est, an lmjus
varietas sit Am. revoluta. Bot. Mag. 915. absque
foliis ? This plant is quite distinct from Am. va-
riabilis. Jacq. the revoluta /3. Bot. Reg. 8. 615. A
light loam suits it best. It is a very ticklish plant, re-
quiring absolute drought, in a cool situation, during
six winter months. The bulb must not be kept
above ground, but the neck just emerging. It will
flower in the summer in the greenhouse, and ap-
proaches nearer to the tropical Crinums than any
other native of the Cape. It will bleed to death if
forced in the winter, and rot if watered at that sea-
son. It is difficult to say what Mr. Hibbert’s bulb
was (Bot. Mag. 915. without leaves) which perished
after flowering. It was probably a variety of this
plant, but if meant for the identical plant it is very
incorrect. The tube of this plant is variable, and
not always shorter than the limb; in Mr. Burchell’s

268

AM AUYLLIDACEjE.

native specimens and drawing it is longer, and the
dried inflorescence might be mistaken for Capense,
but the bulb and leaf agree with Bauer’s drawing
of the plant of the Hort. Kew.

40. Variabile. — Am. Variabilis. Jacq. H. Sell. 4. 426.
Am. revoluta, /3. Bot. Reg. 8. 615. C. Crassifo-
lium. Herb. Append. I have restored Jacquin’s
appropriate name, variabilis, which I ought to
have adopted at first. Foliis lsete viridibus erecto-
arcuatis crassis germine gracili, floribus ortu albis
rubro pallide extus notatis, dein saturate rubris.
Species frigoris patiens, bulbulis quam maxime
prolifera. This is the hardiest known species ;
out of doors it preserves its leaves in winter longer
than Capense, and it shoots earlier in the spring.
Both this plant and revolutum have a slender
germen, but the idea conceived from Jacquin’s
plate by some persons of its cells being monosper-
mous was erroneous. It leaves are deep green.
The flowers turn to a rich purplish red, so that
flowers of two colours are always on the same
umbel, as represented by Jacquin. The figure in
the Register was taken at too early a period for a
just representation, before the expansion of the
flowers. Mr. Ker considers Am. revoluta (Bot. Mag.
1178.) to be this plant. The plant offered at Mr.
Woodford’s sale, which Mr. Wykes, his gardener,
asserted to be the one from which the figure was
made, was not distinguishable from a common
glaucous-leaved C. Capense ; and I observe in
Mr. Ker’s description, he says of the leaves rather
glaucous, which is not the case with any bulb I
ever saw of variabile, which has the green very
bright ; and if the plant had been variabile, the
two decaying flowers in the figure would have
been intensely red. The figure, therefore, if in-
tended for C. variabile, is quite incorrect ; but
it agrees better with some plants that I have of C.
Capense. Mr. Ker draws a peremptory distinc-
tion, that in variabile the tube is shorter than the
limb, and in Capense longer; but in his own figure
of Capense (Am. longifolia Bot. Mag. 18. 661.) it
is shorter. Usually, however, the tube is longer than

AMAR VLLIDACEiE.

269

the limb in Capense, but it is a very variable plant,
and in some seedling varieties its flowers change
to red, as in variabile, which I consider to be much
more akin to it than to revolutum. It is therefore
best to discard all consideration of that plate, and
of Mr. Ker’s description accompanying it, as far as
it disagrees with his amended description in the
Bot. Reg. 8. 615. I grieve that he should ever
have imagined on this subject, that I could have
intended the least incivility to him, from whom,
in whatever intercourse I had with him on botanical
matters, I had met with the utmost urbanity.
When I published that Mr. Griffin disclaimed all
knowledge of Mr. Woodford’s bulb, which the
editor of the Bot. Reg. had stated to be in his col-
lection, it never entered into my mind that I could
be supposed to accuse him of wilful inaccuracy,
and I was cautious not to deny the fact, but only
Mr. Griffin’s knowledge of it. On further expla-
nation it appeared that the bulb never had been
in Mr. Griffin’s hands, nor seen by him ; but Mr.
Ker had misunderstood a memorandum written by
Mr. Griffin, which was certainly so worded through
inattention as to render the misconception almost
unavoidable. I am glad to have this opportunity
of stating, that although I have found some of Mr.
Ker's views to be erroneous, (and where is the man
whose views are free from error?) I place great
reliance on the general accuracy of his botanical
labours.

41. Capense. Var. 1. princeps. — Am. Capensis. Miller
G. diet. Am. longifolia. Bot. Mag. 8. 631. The
name longifolium had been pre-occupied by an
Asiatic species, before this plant was recognized as
a Crinum. Miller’s name, Capense, is therefore
adopted.

Var. 2. Riparium. — Bot. Mag. 53. 2688. Bot. Reg. 7.
546. A very fine dark purple variety with fewer
ovules; introduced by Ur. Burchell from the banks
of the Black River. Permanent by seed, if kept
by itself; not with certainty if near the other
varieties.

270

AMAUYLLIDACEJE.

Var. 3. (fortuita ) flore albo. A very handsome white
variety, introduced from the Dutch gardens, pro-
bably raised in Holland. Not permanent by seed,
if near other varieties ; but nearly so if kept apart
while in flower.

The seedlings of this species, which is easily recognized
by its long glaucous leaves, vary so much in the length of
the peduncle, tube, and limb, and in the shade of colour,
which is usually a mixture of dull red and white, that it is
not advisable to designate any further accidental varieties.
It is a very hardy species, endures the winter, and flowers in
profuse succession during five or six months in a bed covered
with leaves in the winter, and with me it ripens seed by the
bushel. It delights in wet, and will flower in a pond, but
its fibres are rather disposed to rot in the water of a cold
pond in the winter. In a warmer situation it may remain
always in water. I do not know that its fibres would rot if
it was growing in the soil under the pond. It might be
advantageously planted by the edge of any ornamental piece
of water, and would form a beautiful clothing for a small
island, where it would afford thick covert for waterfowl.
Nursery gardeners might easily rear it from seed to sell by
the hundred. A covering of leaves is not necessary to it,
and its own dead foliage would give it a good deal of pro-
tection. I have had the neck of a bulb, which was left in a
pot standing in a small pond, clasped tight by ice two
inches thick for a fortnight, without receiving any injury
from it.

42. Campanulatum. — Bot. Mag. 47. 2121. p. 7. Aqua-
ticum. Bot. Mag. 49. 2352. Specim. Herb. Bur-
chell. A most remarkable plant ; very like vari-
able in foliage, but its leaves are less vigorous,
more fleshy and tortuous. Flowers rosy red, with
a green waxy membrane at the mouth of the tube,
which appears in no other species, and is so foreign
to the genus, that I was at first almost tempted to
believe it was generically distinct, and to detach it
under the name Crinopsis ; but, although its fruit
is unknown, I expect that it will be found to con-
form with the other species. Dr. Burchell imagined
that the spot on which he found it was always
under water. He must, however, have been de-

AMARYLLlDACEvE.

271

ceived in that respect, for it is very prone to rot, if
incautiously watered. I lost three plants in con-
sequence of having placed them out of doors in
front of a stove, where they cannot endure the cold
dampness of the winter. I lost another last year
by leaving it out in the rain in September, and the
finest bulb of the kind at Highclere was killed by
one profuse watering in a lowT temperature in Octo-
ber. It is a plant of difficult culture on this
account, and though it may bear much wet in the
hot season, it is certainly not an aquatic plant. 1
have therefore restored the name which was first
published; as the second, which was substituted in
compliment to Dr. Burchell’s MS. label, is falla-
cious, and may lead to its loss in cultivation.

43. Longiflorum. Hybridum vix spontaneum ; ex erube-
scente et Capensi. Am. longifolia y. Bot. Reg. 4.
303. loculis 9-1 1-spermis. ex Jamaica. Varietas
alia est ex Demerara, ubi C. Capense in scrobibus
viget. — Hybridum hortense, Erubescente-Capense
Bot. Mag. 49. 2336. vix diversum est, loculis
14-spermis. The plant figured in the Bot. Register
is a mule between erubescens and Capense from
Jamaica ; another mule differing a little was
brought from Demerara, where C. Capense, origi-
nally planted in gardens, is naturalized in the
ditches: and they are not materially different from
the artificial mules raised from the same inter-
mixture in this country. The little difference that
exists arises perhaps from their having proceeded
from different varieties of erubescens.

44. Paxtoni. —Hybridum fortuitum ; ex India orientali.
Bulbo 20 unc. circu. caudice 8 unc. foliis scabro-
erubescentis foliis affinibus nitentioribus, scapo
bipedali purpurascente, umbella 20-30-flora incli-
nata; germine sessili, tubo viridi purpurascente
5 unciali, limbo 3^ pallidissime erubescente saturate
rosco-striato, expansione 6 unciali, stylo puniceo
filamentis parum longiore, limbo \ unc. breviore,
antheris tortilibus corrugatis, polline subsicco,
loculis ovulis 7-8 indistinctis. The flowers of this
fine plant, which was sent by Lord William Ben-
tinck, from the East Indies to the Duke of Devon-

272

AMARYLLIDACE^.

shire, come very near to those of Scabro-Capense,
having the tube rather longer and the expansion
greater, the colour nearly similar ; but they are
much more numerous. 1 take it to be cross be-
tween Scabrum and some other species in the Cal-
cutta garden. The leaves are very like those of
Scabro-erubescens, but more glossy. Its dry pol-
len and shrivelled anthers shew that it is not a
genuine species. I have named it after the Duke
of Devonshire's very skilful and intelligent gar-
dener at Chatsworth.

Species dubice. — Doubtful species.

45. Algoense. — Mihi ap. Sweet Hort. Brit. Nondum
floridum ex Africa meridionali, prope Algoa bay,
foliis C. revoluto affine, angustioribus, magis cana-
liculatis. Forsan revoluti varietas.

46 ? Album. — Am. alba, foliis lineari-lanceolatis, flore
albo, spatha 10-flora, floribus declinatis. Kurmge.
Arab. Soraf.— -Forsksel. FI. iEg. Ar. 209. I can
scarcely doubt that this plant, with ten white flowers,
must be a Crinum, and from its abode in Arabia,
it is probably distinct from all that have been
described.

Caffrum. — Mihi ap. Sweet Hort. Brit, nondum floridum
C. campanulato simile et vix diversum est. — Caf-
frum et Formosum Herb. App. delenda sunt.
Formosum est Arenarium v. blandum.

Hybrid or mixed crosses. — 1. Gowcni, or Zeylanico-Ca-
pense. Hort. Soc. Trans, ic. vol. 3. p. 187. First raised from
Capense impregnated by Zeylanicum by J. R. Gowen, Esq.
at Highclere. I received, a few years ago, a bulb of the
same cross accidentally produced from seed of Capense in
in Dr. Carey’s garden in the East Indies. — 2. Mitchamice,
or Australi-Capense. First raised by me from pollen of
pedunculatum at Mitcham in Surrey ; a sub-variety very
different from pollen of Canaliculatum, and another from
pollen of Exaltatum, afterwards at Spofforth. These crosses
are fertile, and ripen some seed every year out of doors in
front of the stove. Some of the seedlings are like the cross-
bred parents, some, having been produced by the pollen from
a large bed of Capense close by, revert very near to the type
of Capense. When there are no plants of Capense in flower

AMARYLLIDACEjE.

273

they come true to the mule type. The cross from peduncu-
latum is of very large stature, and follows the habit of the
male parent in the splitting of its huge columnar bulb, but
produces offsets. They produce a succession of scapes till
winter, and are very handsome. They are perfectly hardy ;
but the climate of Yorkshire is seldom hot enough to make
them flower in the middle of the garden.

3. Herberti, or Scabro-Capense ; named by Sweet Hort.
Brit. A plant of great beauty, bearing 1 1 flowers on a
scape 3 feet high, peduncles short, with the germen little
more than an inch, tube 4 inches, limb 3^, blush with deep
red stripes, expansion about 5 inches ; style as in scabrum,
longer than the limb, red towards the end; filaments, an-
thers, and pollen white ; stigma crimson, small ; ovules per-
fect, 22-24 in a cell ; leaves 5-6-feet long, above 4 inches wide.
It requires a warmer situation where the wall is a little heated
to make it flower finely, which it does several times each
year in such a situation. This cross had continued barren
14 or 15 years, till the summer of 1834, when it produced,
in the border out of doors, one smallish seed, which vege-
tated ; but the leaves of the young plant were colourless and
it died in a few months. In 1835 one larger seed was pro-
duced which was sown in white sand and grew freely. It
is now above a year old, shewing no signs of approximation
to Capense, the leaves being of a bright green, with no glau-
cous tinge, and the plant must be either the genuine offspring
of the mule, or the seed may have been fertilized by Pedun-
culato-Capense which grew close beside it.

4. Puseyce , or Specioso-Capense ; named after my niece
Lady Emily Pusey. 5. Wallichii, or Careyano-Capense ;
named after Dr. Wallich. 6. Seymouri, or Revoluto-Ca-
pense. 7. Roxburghi , or Defixo-Capense. This cross has
produced seed. 8. Ebordci, or Asiatico-Capense. 9. Alta-
clarce, or Erubescente-Capense ; Bot. Mag. 49. 2336. This
cross, raised at Highclere and at Spofforth, from two different
varieties, does not flower freely, producing a superabundance
of offsets, on which it wastes its vigour. The converse Ca-
pensi-erubescens is a handsomer variety. 10. Shepherdi, or
Cruento-Capense ; named after the late conductor of the
Liverpool garden. It has been standing a great many years
in the border in front of a stove beside Scabro-Capense,
but has never flowered, probably requiring more heat. 11.
Baconi , or Zeylanico-erubescens. Raised in the stove of the

T

274

AMAKYLLIDACEjE.

late A. Bacon, Esq. 12. Stapletoniee, or Zeylanico-pedun-
culatum. Raised in the stove of the late Earl of Carnarvon,
and named after his daughter Lady Harriet Stapleton. This
plant strongly supports the opinion I had long entertained
concerning the origin of C. amabile, having altogether the
semblance of an inferior variety of that plant. 13. Louisce,
or Specioso-defixum ; named after my eldest daughter. An
exceedingly pretty and remarkable plant, of which I raised
but one bulb many years ago, which has not yielded any in-
crease by offsets, and is sterile. The leaves are long, lori-
form, and flaccid, unlike the habit of its female parent, and
the flowers are striped with purplish pink. 14. Brownii , or
Americano-bracteatum ; named after Dr. R. Brown. A very
ornamental cross. 15. Letitiee, or Erubescente-bracteatum ;
named after my wife. A remarkable plant, of larger
growth than either parent, with a red column, and an umbel
of about 10 large white flowers. It flowers very frequently,
and is very ornamental, increasing by offsets. 16. Digweedi,
or Scabro-Americanum ; named by Ssveet after Digweed,
who raised it in the stove at Highclere. It does not flower
freely ; in foliage it resembles scabrum very much. 17. Ceci-
lies, or Exaltato-scabrum ; named after my younger daughter.
One plant thereof was raised at Spofforth many years ago,
with a tall red column, the leaves more flaccid than Amabile.
The flowers are striped with red. It has yielded no increase.
18. Cooperi , or Specioso-longifolium ; raised by Cooper at
Wentworth. 19. Parkeri, or Americano-erubescens; raised
at Spofforth. 20. Murrayi, or Forbesiano -Careyanum ;
raised at Spofforth ; named after the conductor of the Glas-
gow garden. 21. DecandoYd , or Flaccido-canaliculatum ;
named after Mons. Decandolle ; a very remarkable cross
with pendulous flaccid leaves. They have not yet flowered.
22. Haylocki, or Flaccido-bracteatum. One plant of this
curious cross was raised many years ago at Spofforth and
has grown very slowly. It has not yet flowered, but may be
expected to do so soon. It has an oval bulb with a long slen-
der column arising from it, and very long pendulous flaccid
leaves. 23? Clarouis, or Spectabili-Capense. One seedling
which has not flowered, raised at Spofforth, and named from
the Hundred of Claro,but the cross, from its aspect, is doubtful.

It will be observed, that with the exception of the imper-
fect notice derived from Dr. Martins of his C. concinnum
with 12 flowers, and the nine-flowered inflorescence of pur-

AMARYLLIDACEAS.

275

purascens from a West African island, no occidental or West
African Crinum has been found with more than 8 flowers,
and we may probably assign 12 as the maximum. The many-
flowered species belong to East Africa, Asia, and Australia.

Sjovanni pola tali (Rheede H. Mai. p. 77.) agrees with no
known species. I suspect that it is a most inaccurate repre-
sentation of Amcenum, but the figure is more like the African
spectabile.

The pedunculated species in this genus are for the most
part disposed only to incline the bud, the sessile and sub-
sessile to nod.

67. Amaryllis. — Leaves liiemal, arcuate ; scape autumnal,
before the leaves ; umbel many-flowered, pedun-
culated, divaricate ; germen triangularly obovate ;
tube narrow funnel-shaped ; petaline filaments ad-
hering to the petals, sepaline to the mouth of the
tube ; anthers incumbent, attached in the middle ;
capsule obovate, disposed to burst prematurely.

1. Belladonna. — Bot. Mag. 9. 733. The type of the
Linnaean genus Amaryllis. Hort. Cliff'. See article
Hippeastrum.

Var. 2. Pallida. — Red. Lil. 479. Bot. Reg. 9. 714.

Var. 3. Latifolia. — Foliis latioribus. Leaves very broad.

These bulbs are natives of the Cape of Good Hope, and
are naturalized in Madeira, having been probably dissemi-
nated from gardens. They are exceedingly hardy, but two
things are necessary to their flowering, a strong growth of the
leaves, and absolute rest from Midsummer till their period of
flowering in September. If the leaves sprout early, and are
so much damaged by severe frosts that a vigorous growth
does not ensue in spring, or if they are prevented by drought
in the spring from developing themselves, or are preserved
green by a wet summer after the period of rest, or if they
have not moisture in September to promote the blossoming,
it will fail, and in the latter case the abortive flower-buds will
be thrown out of the ground when the leaves sprout after-
wards. I have tried cutting the leaves oft’ in a wet summer,
but that did not give the rest from moisture and the heat
during the period of rest, which is requisite to the formation
of the flowers, nor did it allow the return of sap to the bulb
at the moment of desiccation It is quite clear thatMr. Knight
was in error in imagining; that the flower of such bulbs is

t 2

276

AMARYLLIDACEiE.

formed during the growth of the leaves ; though a strong
previous growth of the leaf seems necessary to its formation
by the return of the sap from the leaf at the time of their
decay ; the seasonable drying up of the leaf seems to be the
requisite circumstance, if followed by a sufficiently high tem-
perature. In a pot it is very easy to regulate the points
above stated ; in the open ground it can only be done by
putting a glass covering or awning over the bulbs, so as to
keep off the rain in a wet summer. At Spofforth the bulbs
of A. belladonna are placed at two feet distance from the
front wall of a stove and greenhouse ; and there, in a fa-
vourable season, near 200 stems rise in September in
thick patches. I have seen 30 stems in close contact. A
rich and even strong and manured soil is agreeable to them,
and a south wall promotes the drying of the roots in summer.
They very rarely flower with mein the middle of the garden.
The proper treatment of all greenhouse bulbs that flower
before the appearance of the leaf in autumn, is to promote a
vigorous growth of leaf in as airy a situation as may be found
suitable to their respective constitutions, to keep them hot
and dry while at rest (a glass frame in the full sun will
answer the purpose), and to water them freely when the
time for vegetation arrives. The flower-bud is probably
formed at the moment when the leaf perishes, but without
sufficient heat during the time of apparent rest, the bulb
will not supply it with the juices necessary to bring it to
perfection, which in many cases enable it to sprout without
any aliment to the bulb. I possess a plant belonging to the
confused mass of Ornithogalum, which produces a strong
spike of flowers a yard high, and ripens its seed if kept with-
out water; but a single copious watering is fatal to the scape
and brings forward the leaves, which should not appear till
later. The seeds of Belladonna are pearl colored, purple on
the exposed side, and the seedlings are many years coming
to maturity. The plant long cultivated in Europe is beauti-
fully pencilled with red ; the variety brought of later years
from the Cape is paler, and turns to a deeper red as it fades ;
the leaves of the broad variety shew more diversity than
the flower. Some imported bulbs have a much taller and
more robust scape, and their flowers in fading assume a
deeper red.

With respect to the doubts expressed by Mr. Ker, Bot.
Mag. 733. & 1450. concerning this and the following plant,

AMARYLLIDACE.E.

277

I have to state that the pale-coloured variety, imported
from the Cape, has seeds precisely similar to those of the
old variety ; that I have received both, and other shades of
variation from the Cape ; that the notion of a pale vernal
Belladonna, originating with Miller, is quite erroneous, and
that, whenever a vernal scape has been produced, it has been
the delayed autumnal scape, which is usually thrown up
abortive in the spring, preserved by some accident of the
season through the winter ; that the pale-flowered variety of
Belladonna, which has been often confounded with blanda,
B. M. 1450, is quite distinct from it ; that the bulb imported
by the Hort. Soc. from Madeira, and supposed at first by
Mr. Sabine to be vernal, flowers with me in the autumn, and
has no peculiarity. The bulbs of Sir J. Banks and Van
Royen were the pale belladonna, and not blanda. The
leaves of blanda are much more erect, strongly ribbed and
nerved, and vaginate more; its constitution is much ten-
derer. Mr. Sweet was wrong in altering the Linnaean name
of this genus, and shewed complete ignorance of its affinities
in placing it at a distance from Brunsvigia, between Habran-
thus and Hippeastrum.

2. Blanda. — Bot. Mag. 35. 1450. This beautiful plant
was found by Niven, who collected for Mr. Hib-
bert, and I believe has never since been met with
by any collector. I purchased one of the bulbs
when Mr. Hibbert disposed of his collection, and
Mr. Griffin had another. Mr. Knight, of the Kind's
Road, Chelsea, who had the rest, killed them by
planting them in the open ground, which they will
not endure in this country, and 1 believe there are
no bulbs of it in Europe but the produce of those
two. I lost two by planting them in front of the
stove ; one died the first winter, the other only
lingered till the second. The leaves of this and
the following species, when cut by frost or drought
at the points, will not continue to grow like those
of Belladonna. It requires an airy situation in the
greenhouse in winter, drought and dry heat in
summer, and will then flower magnificently in Sep-
tember. Whatever may have been the growth of
its leaves, it will not flower if it is left in a cold
situation while dry.

278

AMARYLLIDACE7E.

3. Josephiniana. — Josephin®. Redoute Lil. 3. 370. 372.

Var. 2. Flore vix striato. Bruns. Josepliinae. Bot. Mag.
52. 2578.

Var. 3. Griffiniana. — Bruns. Josephinse minor. Bot.

Reg. 192. 193.

It will be remembered that in my Appendix I stated the
great affinity of this plant to Amaryllis blanda, and the
difficulty I found in dealing with the genus Brunsvigia. At
that time I did not possess this species, excepting a small
seedling from the Malmaison bulb, which, though above
twenty years old, is not yet bigger than a fowl’s egg. I have
since obtained four seedlings from Josephini., impregnated
by A. blanda, and failed at the same time to obtain any by
impregnating it by B. multiflora. This is decisive as to
Josephiniana being an Amaryllis, and leaves a probability
that the Brunsvigia of Heister may yet be upheld. Blanda
has a strong midrib to the leaves, which sheath above
ground ; Josephini. no midrib or sheathing : the seedlings
sheath and have a midrib, though much less conspicuous
than that of blanda. They are now ten or eleven years old,
and have reached a larger size than a natural Josephiniana
raised from the same scape, and they shew some diversity of
foliage amongst themselves. I believe that the very large
imported bulbs of this plant and B. multiflora may be
100 years old. As they sprout but once a year, nothing
can be done to accelerate their growth, beyond keeping them
in a healthy and vigorous state : for which purpose the bulb
must be kept under ground, with the neck perhaps above
ground ; but of that I entertain great doubts. If the whole
bulb is exposed, it imbibes moisture from the atmosphere in
the season of rest, which becomes fatal to it, and I have found
the seedling’s of which the neck has never been raised above
ground in the safest state. A rich light loam and abundance
of water in winter, perfect rest and dry heat in summer, are
necessary. There is a good deal of variety in the foliage of
imported bulbs, some having it more obtuse and undu-
lated than others. The lesser variety, which I have named
Griffiniana, because minor is applied to a kindred Bruns-
vigia, is a distinct local plant of very inferior stature, and
I have never seen any intermediate variations.

4. Grandiflora. — Bruns. Grandiflora. Bot. Reg. 16.

1335. This fine plant evidently belongs, as well

AMAR YLLIDACEJE.

279

as Josephiniana, to the genus Amaryllis, and I
expect that my cross between the two last species
will much resemble it. I lost my bulb of grandi-
flora very soon, and I fear that all that were im-
ported at the same time may have been also lost in
consequence of the dangerous practice of keeping
the bulb above ground.

Var. 2. Banksiana. — PI. 32. fig. 2. Spec. Herb. Banks,
ex hort. Kew. Leaf not half an inch wide (7-16ths);
peduncle near 6 inches, curved at the end ; peri-
anth 2f long, segments about wide, about half
an inch longer than the filaments and style ; tube
scarcely any. There is a single flower of this plant
with a portion of a leaf from the Kew Garden, in
the Banksian herbarium. It was probably one of
Masson’s plants, and seems distinct from any
known species, unless it be grandiflora, to which
it is clearly allied; but if the true leaf is attached
to it (which may always be doubted with respect
to plants that produce the leaf and the flower at
different seasons), it is not above one-third the
width, if so much, Never having seen the cap-
sule of grandiflora, nor its flower so as to ascertain
the point of adhesion of its filaments, I judge by its
affinity in bulb and foliage to Josephiniana, and by
the germen of which I have given a representation,
that it is an Amaryllis.

Amaryllis pudica, Ker. is certainly an abortive specimen
of A. Belladonna, of which all the buds had perished except
one; a circumstance not unusual with bulbs that flower on
their first importation, before they have made fibres. When
the leaves grew, it became of course confounded with other
plants of A. Belladonna, and therefore could not be identi-
fied afterwards. In consequence of Mr. Ker’s statement that
a specimen existed in the Banks, herb, conformable to A.
pudica, except in having a sessile instead of a pedunculated
germen, a distinction however which would have been in-
surmountable, I have most carefully examined that herba-
rium, and the result of the inquiry is, that the statement is
erroneous. There are one-flowered specimens of Crinum
Distichum, and one of Crinum yuccaeides (Broussoneti.)
from Kew Gardens, and marked Sierra Leone. It is scarcely

AMARYLLIDACEjE.

280

possible that such plants should have been confounded with
the supposed A. pudica. There is a single flower with its
very long peduncle cut off from the umbel of a magnificent
Amaryllis grandiflora, var. Banksiana, above described, and
beside it the fragment of a leaf. Is it possible that, in a
hasty consideration of the specimen, the absence of spathe
should have been overlooked, and the peduncle regarded as
the scape of a one-flowered plant ? Persons of the greatest
general accuracy are liable to such oversights. There has
certainly been an error of some kind, because the herbarium
furnishes nothing else that can be compared with the sup-
posed A. pudica, unless it be the one-flowered specimen of
Oporanthus luteus, which is labelled A. pumilio ; but Mr.
Ker refers to that as pumilio, and it entirely disagrees with
the description of A. pudica, in not having the lower seg-
ment pushed down by the superincumbent stamens. Con-
cerning that specimen see Gastronema. The species is
stated to rest on the authority of the late Mr. Salisbury, but
I have too often ascertained Mr. Salisbury’s excessive in-
accuracy concerning facts of this kind, to have the least de-
pendance upon it. The species should be finally erased as
founded on a mistake.

Amaryllis alba. Kurmfe. Arab. Soraf. Forskgel. Flor. JEg.
Arab. 209. is probably a Crinum.

68. Brunsvigia.— Leaves broad, recumbent, hiemal ; scape
autumnal, precocious ; spathe broad ; germen angu-
lar ; tube very short (little more than annular) ; pe-
rianth, style, and filaments, recurved ; filaments ad-
nate to the annular tube only ; capsule triangularly
turbinate.

1. Multiflora. — Heister Mon. Bot. Mag. 39. 1619;
scapo viridi.

Var. 2. Rubricaulis. Jacq. H. Sch. 1. 39; scapo rubro.

The statement in the Bot. Mag. that this plant has black
shelly seeds is quite erroneous. All Brunsvigias have green
seeds, perhaps sometimes coloured on the side exposed to
the sun. I have bulbs of rubricaulis raised from imported
seed, now fifteen years old and thriving, but yet no bigger
than a plover’s egg. Imported bulbs have white sand
amongst their coats ; and probably they grow in the sand,
which preserves their delicate bulbs from rotting, and the
vigorous fibres may perhaps penetrate into some stronger

AMARYLLIDACE^E.

281

subsoil. They are very thirsty while in leaf, and it does
not answer to pot them in sand. The same treatment that
is recommended for A. Josephiniana will suit these plants. I
do not know whether the green and red stemmed varieties
are permanent by seed, but I find variation of leaf amongst
the seedlings.

2. Striata. — Jac. H. Schoen. 1. 70. A smaller plant

than multiflora, with the nerves of the leaves
stronger.

Var. rosea; floribus roseis. Spec. Herb. Bank. Spec.
Herb. Lamb, falso B. marginatae specimen esse
putatum.

3. Radulosa. — PI. 22. fig. 2. absque flore, adhuc ig-

noto. Burchell Cat. 2703*. Spec. Burch. Herb,
(una cum bulbis Angliam allatis) foliis margine
(albo?) cartilagineo, obtuse ovalibus, 8 uncialibus,
3^ latis, superficie asperrime et minute radulosa.
Found by Dr. Burchell in the country of the
Bushmen near the Gariep, putting out its leaves
earlier than B. multiflora, but the inflorescence is
unknown. Of three or four bulbs which he
brought to Europe, one was given by him to
Prince Leopold, and may perhaps be still pre-
served at Claremont ; the others perished at Ful-
ham.

4. Minor. — Bot. Reg. 11. 954.

Var. 2. floribus roseis. PI. 32. fig. 1. Burchell herb.
4044. Langekloof. Bulb small and oval, leaves
an inch wide, in form like those of Radula, but
smooth or nearly so on the surface. Dr. Burchell
found also the first variety, with flowers whitish
within, mixed with the rose-coloured.

5. Radula. — Jacq. Hort. Schoen. 1. 68. This plant,

with a very small bulb, anti rasp-like leaves, is
only known by the figure and description in Jac-
quin’s work. It was probably one of Masson’s
bulbs from the western coast of the Cape territory.
Its flower approximates to that of Nerine a little,
but I do not doubt its being a true Brunsvigia.

6? Albiflora. — Ecklon topog. Verz. p. 7. is unknown to
me, and I cannot judge what it may be.

282

AMARYLLIDACE.'E.

I believe all Brunsvigias have a cartilagineous margin
and broad recumbent leaves ; but if Brunsvigia is, as I be-
lieve it to be, distinct from Amaryllis, the distinguishing
features are the triangular capsule, which is almost winged
and diaphanous, and the petaline filaments not adhering to
the petals. The genus was founded by Heister on multiflora,
without any statement of the points by which it was to be
separated from Amaryllis, but merely a description of that
specific plant ; it was adopted by others who distinguished
it, not from the Linnaean Amaryllis, but from the mass of
discordant plants accumulated under that name, or rather
from an imperfect view of some of them ; so that the cha-
racter given became unavoidably more objectionable than
the silence of Heister. I mentioned in the Appendix fifteen
years ago that Josephiniana disagreed with multiflora in the
insertion of its filaments, and in that point, as well as its tube
and germen, I saw that it agreed nearly with blanda, but I
was deterred by the curvature of its corolla from removing it
from the genus where it had been placed. The fact, that
Blanda has bred with it, is decisive that the other points
mentioned furnish the true generic features, and the di-
versity of perianth is such as occurs also in Nerine. Joseph,
not having bred with Multiflora at the same time, goes some
way to uphold the diversity of the genus Brunsvigia : but my
information is not sufficient to pronounce decidedly on its
validity, though I believe in it. I know the capsule of Mul-
tiflora, Striata, and minor, which all disagree with Ama-
ryllis. I do not know the fruit of A. Grandiflora, B. radu-
losa and radula, nor have I had any opportunity of ascer-
taining the insertion of their filaments, or those of Striata
with exactness. There is, however, no fact ascertained
which can throw a doubt on the genus Brunsvigia, except a
general similarity of aspect in the umbels of Josephiniana
and Multiflora, which awakens my suspicions, and I wish all
those who have flowering Brunsvigias would try whether
they can obtain a cross between them and Belladonna. In
bulbs of this order I have often observed an imperfect bulb-
coat acting like an internal bracte to enclose the incipient
inflorescence. In Crinum Cruentum it is prolonged into a
narrow lorate leaf, which indicates long beforehand the for-
mation of blossom in the bulb. In Am. Josephiniana there
are many half-bulb-coats, opposite and alternate.

AMA RYLLIDACE.E.

283

69. Nerine. — Tube none, except an annular connexion;
limb reflex ; filaments with a gibbous monadelphous
base ; stigma ultimately trifid.

Sectio 1. Regulares. — Perianth regular.

Gerviine rectiore , revoluto-patente ;filamentis et stylo fasciculatis,
rectioribus. Germen straighter, revolutely patent; filaments
and style fasciculate, straighter.

Centripetce.

With centripetal inflorescence .

1. Marginata. Jacq. Hort. Schoen. 1 . t.

65. Am. marginata.

2. Curvifolia. Bot. Mag. 725.

3. Coriksca. Bot. Mag. 1089.

Centrifuges.

With centrifugal inflorescence.

4. Ventista. Bot Mag. 1090.

Var. 2. Sarniensis. Bot. Mag. 294.
Var. 3. Rosea. Bot. Mag. 2124.
Var. 4. Minor.

Sectio 2. Distortse — Perianth distorted.

Genuine deflexo ; perianthio filamentis et stylo sursum curvatis.
— Germen deflex ; perianth, filaments, and style, curved
upwards.

Centripetce.

5. Flexuosa. Bot. Reg. 172.

6. Pulchella. Bot. Mag 2407.

7. Lbcida. pi. 26. f. 3. Bot. Mag.

2124. p. 2. A.laticoma. B.M. 497.

Centrifuges.

8. Humilis. Bot. Mag. 726.

9. Undulata. Bot. Mag. 369.

Hybridce. — Hybrid crosses.

Centripetce. Centrifuges.

1 . Mitchamiee, or Curvifolia-undulata. 4. Spofforthiae, or Venusta-undulata.

PI. 43. — Versicolor. Herb. Appen-
dix. Sterile.

2. Haylocki, or Curvifolia-pulchella.

Fertile.

3. Pulchella-undulata.

Inflorescence not yet ascertained.

5. Pulchella-humilis. 6. Humili-undulata. 7. Curvifolia-

venusta.

The genus Nerine, though defined by a feature that
cannot be mistaken, the gibbous union of the filaments at
the point of their junction with the base of the segments of
the perianth which has no tube, except the annular junction
of the segments occupying about an eightli of an inch, con-
tains more extraordinary variations than any genus of the
same order. The first section consists of those whose germen
is continued nearly straight from the peduncle, the perianth
regular, expanded and revolute, and the filaments fascicu-

284

AMARYLLIDACEiE.

late. In the 2nd section the germen is sloped from the
peduncle, the perianth distorted and curved upwards, and
the filaments at first curved downwards and afterwards rising
and curved upwards. Such a difference might have led to
the supposition that the two sections were generically dis-
tinct, but experience has shewn that they breed together
freely. But there is yet a more singular circumstance
attending this genus. In all the other genera allied to it
(indeed, as far as I have observed, in this whole suborder)
the inflorescence is centripetal (that is, the outer buds of the
umbel expand first and the central last), unless perhaps the
very crowded umbel of Buphane ciliaris is divided into
compartments, by which the law is modified. In the genus
Nerine, however, some species have a centrifugal and some a
centripetal inflorescence, and this remarkable peculiarity does
not coincide with the two sections, but divides each of them
almost equally. The hybrid Nerine Mitchamise, which was
figured in my Appendix under the name Versicolor, was pro-
duced by a centrifugal species, Undulata, fertilized by a
centripetal species of the other section, Curvifolia, therefore
doubly removed from each other. Its inflorescence follows
the centripetal habit of the male parent. It is a shy flowerer,
and seems quite sterile. Some raised three years ago from
Pulchella by Curviflora, which, though in different sections,
have the same course of inflorescence, show the type of the
male much more strongly in the foliage, and grow more
freely. The flowers are very similar to those of Mitchamiae,
or Curvifolia-undulata, having the same pale rose-colour,
changing to a greyish purple, but they are produced in a more
vigorous state, appearing to be quite fertile, and I expect to
obtain seed both from the mule, and from Curvifolia ferti-
lized by it. I attribute the difference in the apparent ferti-
lity to the parents having the same habit of inflorescence,
both being centripetal. I was formerly of opinion that the
Am. marginata of Jacquin must form a genus by itself,
intermediate between Brunsvigia and Nerine, having been
misled to that decision by a specimen professing to be that
plant in Mr. Lambert’s herbarium ; but I have since ob-
served that the specimen was not in any manner ascertained
to be Marginata. but was named conjecturally with a quaere,
and on comparison with Masson’s specimen of Brunsvigia
Striata in the Banksian herbarium which has the leaf with it,
and is certainly correct, I am satisfied the specimen supposed
to be Marginata is in truth B. striata. We have therefore

AMARYLLIDACEiE.

285

no knowledge of Marginata excepting from the figure and
description by Jacquin. In his plate it differs from all other
species of Nerine in having broader leaves, with a red mar-
gin and a short blunt spathe, in which respects it approxi-
mates more to Brunsvioia. The inflorescence differs in no
other respect from that of Curvifolia, except being less un-
dulated, and I have lately observed the undulation of Curvi-
folia very much diminished under some circumstances of
temperature, and in particular seedlings. Jacquin mentions
that marginata has a fetid smell, and, having in the autumn
of 1835, placed Curvifolia one sunny day in the stove to
observe the effect of heat on the posture of its filaments, 1
was surprised to find that it had a fetid smell, something
like that which the most fragrant hyacinths acquire when
they are beginning to decay. That circumstance is decisive
as to the near affinity of the two plants, and (the confusion
occasioned by the supposed specimen in Mr. Lambert’s her-
barium being removed) I have no hesitation in saying that
it is a Nerine, with broader leaves than any of the other
species, and differing further in having a red margin. I
have, however, plants of Curvifolia with the leaf nearly as
broad. This remarkable plant must have been one of Mas-
son’s collection ; he followed the Western coast from the
Cape, and no other traveller has pursued the same track, which
is the cause that several of the plants which he introduced
have not been since met with. Marginata agrees with Cur-
vifolia in having the new leaves sprouting during the time
of flowering ; those of Brunsvigia appear after. Curvifolia
flowers freely ; Corusca spawns too abundantly ; Venusta
minor is a beautiful miniature of Venusta, and flowers pretty
freely. I have had it about twenty-five years, but it has
never been figured : the colour deep red. Flexuosa, pul-
chella, humilis, and undulata are of easy cultivation. Un-
dulata flowers strongest out of doors against the front wall of
a greenhouse. The flowering of all is autumnal, in some
pushing a little before the leaves, in some as Flexuosa and
Undulata after. My first mention of Lucida, in the Bot.
Mag. represented it to be a Nerine. I was afterwards misled
by an insufficient examination of a dry specimen to think it
an anomalous Brunsvigia. My first impression was right,
and I have since ascertained that the inflorescence and fruit
exactly agree in all points with Nerine. The only pecu-
liarity it has is, that after a period of rest its old leaves are

286

AMAIIYLLIDACE^E.

more disposed to push again, but I find other species of
Nerine capable of elongating their leaves after the ends have
been injured, which a Brunsvigia cannot do.

To promote the flowering of Nerines a vigorous growth
of leaf must be encouraged in the autumn ; the requisites are
warmth enough to excite them, and air enough to prevent
their growing weak. During the winter they must not be
allowed to lose their foliage either by frost or drought. About
May they should be allowed to rest by ceasing to water them.
After three months’ rest they may be watered again to pro-
mote their growth, at the very beginning of September.
Lucida, which grows near the snowy mountains in Africa,
and is probably there cut by frost in the winter, seems more
disposed to grow whenever it can, pushing out afresh the leaves
of which the tips have been damaged, whether by drought
or frost. I believe its native situation is refreshed in the hot
season by frequent thunder-storms, so that it has not the
decided term of rest, which the dry season forces upon its
congeners in the lower country. I find its leaves very willing
to grow in winter with a little assistance of artificial heat.
The other species suffer more from any injury to the point
of their growing leaves. Dr. Burchell found N. lucida
growing in company with Uncaria procumbens between
Gattikamma or Wittewater and (Aakaap) Riet fonteen, and
coming into flower on February 15. The seed was ripe ten
days after. Its flowering in its native place is therefore
autumnal. At that time Buphane toxicaria was in leaf, and
its flowers long passed. Gattikamma is 22 miles from Kloar-
water, which is in Lat. 28°. 50'. 56". s. Long. 24°. 3 . e.

Hybrida. 1. — Michamiae, curvifolia-undulata, tab. 43. —
Bulbo viridi-purpurascente; foliisf unciae latis, sub glaucis,
obtuse acuminatis ; scapo viridi, 18-unciali, nr lato, basi
aliquando purpurascente; umbella circiter 16-flora, bracteata;
pedunculis uncialibus ; germine loculis circiter 6-spermis,
ovulis plerumque emarcidis ; corollae laciniis semipaten-
tibus undulatissimis apice reflexo, 1| uncialibus, ^ unciae
latis, roseo-rubentibus, dein coeruleo-purpurascentibus costa
media vel stria rubente ; stylo et filamentis roseo-rubris
inferne purpurascentibus, stylo filamentis longiore, corolla
breviore, denuo assurgenter curvato ; stigmate albo.

About four and twenty seedling bulbs of this mule
were raised in 1815, at Spofforth, from Nerine Undulata, im-
pregnated by Curvifolia. No difference has been observed

AM ARY LLIDACEjE.

287

amongst the seedlings, excepting that some have the leaves
a little more glaucous than others, and some have the
scape green and others a little purpurascent. The bulbs
are of a greenish purple, the leaves rather glaucous, f of an
inch wide ; the scape a foot and a half high, 3-16ths of an
inch wide; peduncles about an inch long, flowers about 16;
the laciniae of the corolla £ of an inch wide, lj long, very
undulate, half expanded, with the tips bent back, rose coloured,
changing after a few days to blueish purple, except the mid-
dle rib or stripe, which continues red. The cells, which in
Undulata are 2-seeded, and in Curvifolia about 8-seeded, in
the mule are generally 6-seeded, but the ovules are discoloured
and probably imperfect. In Undulata the germen is always
bent downwards by the premature growth of the seeds in
the upper cell, by which the corolla and filaments become dis-
torted; in the mule the germen continues straight as in Curvi-
folia. In Undulata the filaments are very liable to irregularity
and generally curved downwards, those of the mule approach
nearer in straightness and regularity to those of Curvifolia :
the style becomes at last bent upwards, but not so much bent,
nor is it at first so short, as in Undulata, of which the style is
by degrees considerably prolonged. In some flowers of the
mule the style is entirely wanting. The style, or one or
more of the filaments of the Amaryllideae, are occasionally
abbreviated or prolonged by accident. In Nerine Undulata
they are particularly irregular, in consequence of the distor-
tion of the flower, but in its most perfect state the corre-
spondence of the alternate filaments is evident. Seedlings
of Undulata flower when about three years old; the strongest
of the mules blew at four years : but seedlings of Curvifolia
seem not to flower till they are eight or nine years old. The
figure represents a full-sized seed of each of the parents,
the largest being that of Curvifolia.

70. Strum aria. — Umbel many-flowered, pedunculated ;
spathe 2 valved ; perianth regular ; tube, none except
an annular connexion; filaments connected at the base,
the alternate filaments, generally, more or less adnate
to the style in proportion to its thickness ; anthers in-
cumbent; style strumous (that is, enlarged below), an-
gular, furrowed, stigma trifid ; seeds roundish, few.

1. Augustifolia. — PI. 29. f. 14. Jacq. Coll. Supp. 48. ic.
rar. 2. c. 259, Leaves ^ wide, subacute ; perianth

288

AMARYLLIDACE-iE.

white, lined with red ; style with three bifid pro-
cesses, a little longer than the filaments, of which 3
are free, 3 connate with the style.

2. Truncata. — PI. 39. f. 11. Jacq. ic. 2. t. 357. Specim.

Herb. Banks. Leaves obtuse, \ an inch wide ;
spathe lined with red ; flowers white, red at the
base and in the bud, filaments shortly connate, 3
adhering to the style.

3. Rubella. — PI. 39. f. 12. Jacq. ic. 2. 358. Leaves j

wide, flowers red, filaments cylindrically connected
half their length, 3 connected with the style.

4. Linguaefolia.— PI. 29. f. 10. Jacq. C. S. 45. ic. 2.

356. Leaves obtuse, ^ an inch wide ; flowers
white, lined with green ; filaments connected at
bottom with the perianth ; 3 connected with the
style.

5. Undulata. — PI. 29. f. 13. Jacq. C. S. 50. ic. 2. 360.

Leaves \ wide, subacute ; flowers undulated, white
tipped with red ; filaments shortly connected at the
base, all free from the style.

This genus is distinguished from all the foregoing by the
strumous style. It is not very material whether Strumaria or
£lessea be placed next to Nerine, but it seems advisable to
keep those with an enlarged style together. Properly Stru-
maria and Hessea should stand side by side, the former con-
nected with the subsequent genera by its strumous style; the
latter by its short erect anthers. There is in the Banksian
herbarium, a drawing by Bauer, of a Strumaria without name,
which I believe to be an undescribed species or variety; the leaf
agrees nearly with that of Linguaefolia ; the umbel is repre-
sented rather drooping to one side ; the spathe short ; the pe-
rianth not undulated, white with a very faint blush of pink ;
the dissection having been made before the maturity of the
sepaline anthers, the style, which is green, is not full grown,
and the stigma not developed ; the dissection does not shew
the base, nor whether the filaments, which are crowded round
it, adhere or not. It was probably a representation of one of
Masson’s bulbs which flowered at Kew. From appearances, I
should think it a variety of S. linguaefolia, with a pink blush
instead of a green line outside the sepals. The point of a
young leaf and Bauer’s dissection are given, pi. 29. f. 15.

AMARYLLIDACEiE.

289

A

71. Hessea. — Spathe bifid; limb regular, tube short; fila-
ments equal, subulate, erect at first, afterwards reflex,
inserted at the mouth of the tube ; anthers short,
erect, inserted at the base, after expansion circular ;
style filiform ; stigma trifid, patent, fimbriated ; ger-
men at first very small, swelling large prematurely,
turbinate.

1. Stellaris. — Am. stellaris. Jacq. H. Sch. 1. 37. t. 71.

There are five specimens in Mr. Lambert’s Herb.; one
in Dr. Lindley’s. The plant has never, I believe,
been introduced into this country. The sepals and
petals seem to differ a little in colour.

2. Breviflora. — PI. 29. fig. 6. Specim. Herb. Banks.

leaf PI. 43. f. 4 ; pedunc. unc. erectis, germine
parvulo, floribus 14, tubo brevi, limbo (albo?)
semipatente, staminibus vix longioribus, stylo
breviore. This little plant, which has been long
overlooked in the Banks. Herb., and is probably one
of Masson’s, comes near to Thunberg’s Hsemantlms
vaginatus, but differs in having the style shorter.
Carpolyza spiralis vaginates, but is not his H. va-
ginatus.

Species vix dubia.

3? Vaginata. — Haemanthus vaginatus. Thunb. Flor.
Cap. p. 297. Leaves 4-6, linear lanceolate, sheathed
at the base ; segments of the limb lanceolate, acute,
shorter than the spathe ; scape solitary, smooth,
round, a foot long or more, thick as a goose-quill;
umbel about 20 flowered ; spathe 2-valved, ovate,
lanceolate, concave, smooth ; peduncles filiform,
erect, smooth, 1 inch long ; segments equal, acute,
white ; filaments as long, style longer ; stigma
trifid.

This genus, as well as I can ascertain, is distinguished from
Nerine and Brunsvigia, by its anthers erect and attached at the
base, and the near equality of the stamens; from the former
by their equal insertion, and from the latter, by the absence
of their gibbous union. It is difficult to understand on what
principle it could have been united with Strumaria. The name
Hessea was given by Bergius to Carpolyza spiralis ; I know
not with what view R. and Schultes can have taken the name

u

290

AMARYLLIDACEjE.

Hessea in preference, seeing that Mr. Salisbury had very
many years before published the character of his genus Car-
polyza, with a figure and dissections of the plant, in the
Parad. Londin. The name Hessea was given, as I understand,
in compliment to Mr. Hess, a most respectable Missionary at
the Cape, and I am glad to be able to transfer the name,
which cannot stand in the place of Carpolyza, to a closely
allied genus.

72. Imhofia. — Spathe 2-valved ; limb patent ; style erect,
strumous or thicker below, furrowed ; stigma simple
or just trifid ; filaments equal, enlarged at the base,
separate, inserted in the disk, patent. (Quaere, whe-
ther sometimes adhering to the base of the style.) An-
thers short, attached at the base, erect, after expan-
sion circular. Seeds (like those of Nerine) green, with
one angle and a rounded back.

1. Filifolia. — PI. 29. f. 8. Strumaria filifolia. Jacq. ic.

r. 2-14. Bot. Reg. 4. 440. Leucojum strumosum,
Jacq. ic. t. 361. in icone, non in textu. — Hort. Kew.
Leaves almost filiform, channelled; flowers 5-11,
segments acute, not undulated, white, lined without
with red ; style with a great 3-furrowed protrusion
below the middle ; stigma with 3 short lobes ; an-
thers purple. I have neglected to examine this
plant minutely, when in flower ; it is said to have a
nectareous membrane adhering to the base of the
style. It ripens seed freely, and is easily cultivated,
if not watered in the summer. Grows near the Cape
and Nyland, in sandy places.

2. Crispa. — Strumaria Crispa. Bot. Mag. 33. 1363.

Amaryllis Crispa. Jacq. H. Sell. 1. 37. t. 72. Am.
Cinnamomea. L’Herit. S. A. 16. 27. Leaves fili-
form ; flowers 7 or more, undulated, white and rose-
coloured; anthers purple; filaments subulate; style
thick and 3-furrowed.

3. Burchelliana. — PI. 29. fig. 5. Specim. Burch. Herb.

4954 et 4967. Scapo 4\ unc. spatha f, pedunc.
1^-1^, florib. 7-11, albis, lin. dorsal, (viridi ?) peri-
antliio vix 4, subundulato, staminibus longiore ;
filamentis et stylo basi incrassatis ; antheris albis ;
stigmate simpl. aut vix trifido. Circa Langkloof.

AMARYLLIDACEiE.

291

This little plant, found by Dr. Burchell near Langkloof,
without leaves in March, approaches to filifolia, but differs
in a little undulation of the flowers, lined, as it appears, on
the back with green, anthers white, a more simple stigma, and
a style (as it seems in the dry specimen), broad at the very
base. It may possibly be only a local variety of filifolia.

Var. 2. Virescens. Specim. Burch. Herb. 2683. Major,
floribus 15, albis, virescentibus ; stigmate vix trif.
lobis obsoletis, absque foliis. Dr. Burchell had con-
sidered this, when he gathered it, to be a distinct
species ; I could discover nothing in the dry speci-
mens to distinguish it, except superior size and a
greener hue : but when in flower in Dr. Burchell 's
garden, it was shewn to Mr. Ker, who, as I learn
from Dr. Burchell, said it was Gemmata. If he
meant that it was the very plant figured in the B.
Mag., the flower of gemmata must be grossly ex-
aggerated in the plate, being twice as large as in
the specimen of virescens.

4. Gemmata. — Strumaria. Bot. Mag. 39. 1620. Leaves
wide, ciliated, subacute ; peduncles 3 inches ;
perianth 9-16ths, very much undulated, white,
without pale red, with a green mid-rib ; style
greatly enlarged at the base, so as to occupy the
disk ; filaments less enlarged than in the other spe-
cies, adhering to the base of the style (Mr. Ker
says, inserted in it; there seems to be no real dif-
ference of structure, but that the style intrudes on
the base of the filaments) ; anthers white ; stigma
at first simple, afterwards trifid.

Only known by the figure and description in the Bot.
Mag. I am persuaded that Dr. Burchell’s virescens is not
the same plant. Its flowers are not near so large, nor undu-
lated ; and, according to Mr. Burchell’s notes, they have no
rose-colour. Mr. Ker mentions a limpid drop between the
filaments and stvle. The name Imhofia liavino- been set at
liberty by the union of marginata with Nerine, I have been
glad to fulfil the object of its framer, by applying it to a
nearly allied genus. It was originally intended for Nerine
venusta, but was not defined by any character, nor ever
adopted : and its substitution for Nerine, which has been pretty
generally adopted, would have occasioned useless confusion,

u 2

292

AMARYLLIDACEiE.

and have been inconvenient to the public. I had not observed
the name thus suggested loosely by Heister for venusta,
when the genus Nerine was established, or I should have pre-
ferred it.

73. Carpolyza. — Umbel pedunculated; spathe 2-valved ;
limb semi-patent, regular ; tube short, funnel-shaped ;
filaments adnate to the whole tube, the alternate a
little shorter; anthers oblong, attached at the base,
not versatile; style thick, furrowed, triangular, slen-
derer upwards; stigma trifid, recurved, fimbriated;
germen obovate, 3-furrowed ; capsule 3-celled, 3-
valved ; seeds green, rounded on the back.

1. Spiralis. — Specim. Herb. Bank. PI. 29. f. 9. — Dis-
sections from Salisbury, Paradisus London. 63. —
Crinum tenellum. Jacq. Coll. S. 43. ic. r. 2. t. 363.
Crinum spirale. Bot. Rep. 92. — Amaryllis Spiralis.
L'Her. S. A. p. 10. — Strumaria Spiralis. Bot. Mag.
33. 1834. — Haemanthus spiralis. Linn. fil. in Hort.
Kew. — Thunb. Prod. 58.

A very neat little plant, with filiform spiral leaves, and
the lower part of the scape singularly spiral ; spathe an inch
long, acute, flowers few, sometimes but one, about f- of an
inch, white; sepals red on the outside, tipped with green;
alternate anthers mature first, peduncles about li inch.
Found by Masson near Cape Town.

When each successive writer refers a plant to a different
genus, as in this case, it may with great probability be sur-
mised that it belongs to none of them. Mr. Salisbury de-
fined the genus and named it, giving excellent dissections, in
the year 1807 ; and nothing but the confusion in which the
whole order was involved, could have prevented the earlier
adoption of the name he gave to it. The genera Imhofia and
Carpolyza bear the same relation to Strumaria, as Hessea does
to Brunsvigia and Nerine, distinguished by anthers erect and
not versatile, a most decisive feature. Carpolyza differs
from Imhofia in the tube, the shajoe of the anther, the form
of the style and stigma.

Suborder 5. Narcisse^e. — Perianth tubed ; cup including
the filaments ; sepaline filaments prolonged, sepals
(when unequal, and I believe always), wider than
the petals ; scape hollow upwards, more or less
filled up below; peduncles solid, unequal. The

AMARYLLIDACEvE.

293

principal fissures of the crown when it is three-
lobed opposite the sepaline ribs, the indenture of
each lobe opposite the petaline ; style mostly
tri partible, sometimes by accident (in semidouble
or degenerated double flowers) tripartite ; ovules
in from 2 to 4 rows which are mostly imperfect
and confused; seeds with a black shell.

Mr. Haworth was bold to say that “ it was better for his
future reputation that many years had elapsed without his
genera and species of this section or suborder having been
adopted by others, for that adopted they will be, and then
must be proved his having seen just so far before his com-
petitors that “ his characters are founded on the solid laws
of botanical proportions,” and that “ he knows their vali-
dity.” I wish to give Mr. Haworth full credit for his labours,
though, indeed, I cannot say that his descriptions always
tally with the heads under which he places them ; but after
a careful inspection of the second edition of his Monograph,
without any wish except to ascertain the truth, I must take
leave to say that it is utterly impossible that the scientific
public should ever adopt his generic characters, because they
are founded on trivial features, in some cases unfit even to
support a specific distinction ; and although the names, which
Mr. Salisbury first published without any definitions will be
properly retained, they must be attached to generic charac-
ters very different from those given by Mr. Haworth. The
erroneous features on which he has built are the relative pro-
portion of the tube, limb, and cup, the form of the margin of
the cup, the colour of the flower, the length of the filaments
and style, the smoothness of the scape, the colour and shape
of the leaves, not one of which is fit to be made the founda-
tion of a generic character. It sometimes occurs that through-
out a genus there maybe a very striking disproportion between
the tube and limb, which is so manifest that it is tempting,
and pretty safe, to insert it in amongst the features, though it
would be more properly subjoined as an observation ; the
proportion of the cup to the tube may be more important,
and a peculiar colour may be found to coincide with the
limits of some genera, but neither can support a generic
character which is not founded on some more decisive dis-
tinction. Mr. Haworth is not, however, consistent even in
the use of the features he adopts; for he makes the style in-
cluded a generic feature of Qileus, yet style always included

294

AMARYLLTDACE/E.

characterises one variety, and style as long or longer than the
cup another variety, of even the same species of Tros, and I
find his specific distinctions equally inconsistent. I need
have no hesitation in pronouncing that such characters can-
not be adopted. The exact proportion borne by either the
tube or the cup to the limb is by the analogy of other genera
in this order to be considered as nothing but a specific fea-
ture ; the relative length of the filaments and style to the
limb is scarcely to be relied on as such ; the form of the
margin of the cup is utterly insignificant ; and the genera
Calostemma and Eurycles shew that even its absolute inci-
sion is not to be depended upon, unless coupled with some
decisive feature to which it may be subsidiary ; the form of
the tube is variable in many genera of plants, which may be
well exemplified by Petunia nyctanigeniflora, in which it is
nearly cylindrical, and phoenicea, in which it is ventricosely
campanulate, yet the hybrid or mixed produce of the former
plant, impregnated by the second, ripens seed with me more
freely than either of the parents. Vallota and Cyrtanthus
shew that a longer or shorter adhesion of filaments, which
are adnate and not actually inserted, is a variation that
occurs even in the same umbel in this order ; a little greater
or less comparative length of the style occurs sometimes in
the same umbel of the Narcissiform plants, and the length of
the filaments is exceedingly variable in many species of
Amaryllideee.

An equal or unequal adherence of the sepaline and peta-
line filaments, a decided difference in the form and attach-
ment of the anthers, a decided upward flexion compared with
a decided rectitude or connivance of the stamina, may, as far
as my observation extends, be generally relied upon as signs
of a generic diversity. Applying these principles I cannot
hesitate in pronouncing Corbularia, the hoop petticoat, and
Ajax, the daffodil, to be quite distinct from Narcissus, and
the rest of the Narcissean fraternity. I find a further dif-
ference of structure in the stamens distinguishing Queltia,
Narcissus, and Hermione, which however approximate more
nearly to each other.

The next point of importance is to inspect the fruit, and
see whether any diversity thereof corresponds with the diver-
sity of the stamens. I find the capsule of Ajax turbinate
and not erect, sloping from the peduncle if the scape con-
tinues erect, straight if it falls to the ground, preserving an

AMARYLLIDACEA2.

295

horizontal position ; 1 find the like position in the vernal
species of Narcissus, and a sloping posture also in Queltia.
In Corbularia and Hermione I find the fruit erect from the
time it begins to swell. There is a difference of stamen in
the jonquil and an erect capsule, and the autumnal Narcisseae
have also an erect capsule, and I am doubtful whether, on
close inspection of the latter, they may be found to disagree
with Hermione, and perhaps form a different genus with the
jonquil. I find the seed of Ajax roundish, wrinkled with an
elevated wrinkled raphe and chalaza; of Ganymedes ap-
proaching to that of Ajax ; of Corbularia irregularly flattened,
wrinkled, and somewhat foliaceous ; of Hermione shining,
smooth, properly compressed with a rounded back, irregu-
larly angular by contact, the margins of the angles sulcate
on the side, and the raphe sulcate with elevated margins ;
that of Narcissus roundish, sometimes angular by contact,
minutely lined or wrinkled with an elevated raphe and
chalaza. I have not however had sufficient opportunities
of tracing the diversities of the seed, and I have seen so
much outward difference of appearance between the seeds of
two Hermiones, that I cannot describe the seed of Hermione
with any confidence. I have not been able to obtain that of
any Queltia, except the doubtful species Jonquilla, which
agrees with the autumnal Narcisseae in having an erect cap-
sule. I have admitted the genus Ganymedes with very great
hesitation, having been disposed to look upon it as a section
of Queltia. It forms, however, a very marked group, and I
find no intermediate plants to connect them. Mr. Salisbury
had a remarkable intuitive perception of differences which
has generally proved to be correct, though he was unable to
give any satisfactory account of the existing marks of dis-
tinction. I am indebted to him for nothing in this natural
order, but naked names of genera, excepting Acis and Car-
polyza which he defined in the Parad. Lond. He brought
his MS. to me one morning, and read to me some names and
observations of little importance from it, but did not let me
look into it, and I believe it contained no finished generic
characters. He could give me no account of the features on
which he built his separations, except that he had founded
Hymenocallis on two-seeded cells, which was quite incon-
sistent with the fact, the range of the genus extending from
2 to 8-seeded cells. The names of the Narcissean genera and
Eurycles were published by him in the Hortic. Trans, with-

2 96

AM ARYLLIDACEiE.

out any explanation of his views, and I doubt his having
ever put them on paper, and 1 apprehend that he had for-
gotten the points which had led him to a just conclusion while
the plants were under examination. The affinities of the
vegetable creation were the avowed object of his MS., and he
informed me that he expected his future botanical fame to
rest upon that foundation ; but I believe he left nothing suf-
ficiently finished to be fit for publication.

Miller states that the Narcisseee usually flower the fifth
year from seed, but finer the second and third season than
the first. He says that they should not be removed above
once in three years, because they flower weaker after being
moved, but that jonquils, if left standing, grow long in the
bulb, and do not flower well ; and he recommends removing
them every year. Light loam, with a little dung, is recom-
mended by him, if very light to be strengthened. The bulbs
should be set by the beginning of September, and I believe
the less time they are kept out of ground, the less they will
suffer from removal. There is, however, another cause which
affects them when removed, that the wet earth lies too close
upon the bulb, and when in a tuft, the wet drains through
the interstices between the bulbs. Haworth recommends
loam, an elevated border, and a covering of leaves in winter,
which is certainly right for most of them ; but some are na-
tives of rocky and stony situations, and some appear to like
sandy soil. N. majalis is stated in Eng. bot. to have natu-
ralized itself on a heath in Kent. There can be no doubt
that they will be less liable to rot if the border be raised, and
that good drainage and a winter coat of leaves, is essential
for the tenderer sorts, and some of them will not be safe in
the border without further protection.

74. Corbularia. — Hoop-petticoat. Style and filaments de-
clined, recurved ; anthers short, attached at the
middle, incumbent, versatile. Filaments successively
matured ; the sepaline adhering to the tube near the
base, the petaline inserted at the base ; tube funnel-
shaped, cup large funnel-shaped (longer than the
tube) ; limb little conspicuous, segments slender.

1. Cantabrica. — Haworth. FI. omnino candido. Clus.
166. f. 2. Flower white, cup crenate, style in-
cluded. Totus albus. Swert. Flor. t. 28. fl. omn.
albo. Morris, s. 4. t. 23. fl. white or whitish, brims

AM ARYLLIDACEjE.

297

a little uneven. Park. par. p. 106. no figure. Albi-
cans. Haw. Mon. with an erroneous reference to
Park. par. 106. f. 4. style erroneously stated to be
included and cup quite entire. This white spe-
cies, of which the limb is probably a little tinged
with greenish yellow on the outside, was formerly
well known in this country and Holland, but seems
to have been long lost. It is native of Biscay and
the Pyrenees, There is no reason to suppose that
two white varieties were ever known. If there be
an old engraving with a prolonged style, no depen-
dance can be placed on it.

2. Bulbocodium. — Linn, who cites “ Juncif. 2. fl. flavo
leviter ettenuiter per oras fimbriato. Clus..l66.f. 4.”
Clusius’s plant has the style shorter than the cup,
which is described as slightly fringed on the mar-
gin. It grows between Seville and Lisbon, and
according to Brotero, near Lisbon and Coimbra.
Mr. Haworth’s assertion that this is Bulbocodium
of Linnaeus’s herbarium is cpiite erroneous. His
specimen has the limb §ths of an inch shorter, the
style |4hs longer than the cup, the leaves much
longer than the scape. It is certainly tenuifolia,
but we must look to what Linnaeus published, and
that cannot be reversed in consequence of his hav-
ing confounded another species with it afterwards
in his herbarium, or considered it as a variety not
worth distinguishing. He did not found his cha-
racters of Narcissi on the specimens in his herba-
rium, but in some cases, as in this, on the faith of
other writers, and he did not distinguish two Cor-
bularias. The quotation attaches the name bulbo-
codium to Clusius’s fig. 4.

Var. 1. Lobulata. Haw. — Clus. 166. f. 1. Park. par.
107. f. 7. margin waved and indented; flowers in
April.

Var. 2. Serotina. — Sweet Br. f. g. ser. 2. 164. N. Bul-
bocodium. Bot. Mag. 3. 88. Park. par. 107. f. 8.
Turgida. Salisbury. Flowers in May. Said to
grow on the Pyrenees, and to be abundant near
Tarbes. Cup undulately-lobed. Haw. Parkinson
states that this plant flowered a month later than

298

AMARYLLIDACEjE.

the other three, which he describes. This is the
hardiest kind of Corbularia.

3. Conspicua. Haw. — Var. 1. princeps. Park. par. 107.

f. 6. Sweet Br. fl. g. ser. 2. 326. Taller and larger
than Bulbocodium. Cup plicate, margin repand,
style longer, leaves erect. April.

Var. 2. Minor. — Both varieties are cultivated at the
Chelsea gardens, and are conspicuous by the erect
leaves and long scape. I have not seen the flower.

Var. 3? Gigas? — Theatr. flor. t. 21. Whether the fig.
referred to represents v. princeps or a larger va-
riety, otherwise unknown, I cannot pronounce. I
find reason to place little faith on the correctness
of the engravings of that period.

4. Tenuifolia. — Var. 1. Princeps. Sweet Br. f. g. t. 114.

Bulbocodium. Specim. Linn. herb, sed non Linn.
Syst. Leaves very narrow, cup 6-lobed, limb fths
shorter, style fths longer, than the cup. I had
hundreds of bulbs of this plant in my garden for
many years, but one inauspicious season extermi-
nated them.

Var. 2. Minor. — Smaller in all its parts. Both varie-
ties are cultivated in pots in the Chelsea garden.
In Mr. Bentham’s herbarium I find a singularly
minute variety of this plant stated to grow in Italy
generally, the style exceeding the cup.

5? Obesa. — Salisbury Hort. Soc. Tr. 1. 350. Cup en-
tire inflated, (style exceeding the cup, Haw.) leaves
recumbent on the ground. I have not been able to
discover this plant, which Haworth states to flower
in March. Salisbury had it from Tangiers, and says
that it grows also in Spain and Portugal, and is a
tender plant. He quotes Clus. Pseud, n. June. 1.
p. 165, oris non sinuosis aut fimbriatis, and says,
edge of the cup even. Fl. April.

Aurea, taken up on the authority of a bad figure in Mo-
rison, I have no hesitation in rejecting. It is quite vain to
trust to minute differences in such engravings. There is a
very large specimen figured Theatr. flor. t. 21. which Mr.
Haworth names Gigas. I cannot rely upon the engraving,
and refer it to Conspicua.

AMARYLLIDACEiE.

299

75. Ajax. — Daffodil. Style thick, straight, subulate, three-
furrowed, tripartible; filaments straight, clustering
round the style, adhering to the lower part of the
tube, growing into the back of the anther below
the middle ; anthers erect, straight (the point rarely
a little curved), linear, margins of the lobes bent back,
meeting behind, and enveloping the filaments. Tube
funnel-shaped, crown nearly cylindrical, longer than
the tube, equal to or longer than the limb, which
is conspicuous ; capsule sloping ; seed roundish,
wrinkled, with a prominent wrinkled raphe and
chalaza.

1. Minor. — Var. 1. Princeps. N. minor. Linn. Cunei-
folius. Haw. Scape internally filled up near to
the top ; filaments adnate close to the base un-
equally, see pi. 38. f. 32. ; style a quarter of an inch
shorter than the cup ; limb semipatent, pale yel-
low ; segments not imbricating ; tube half an inch
long ; cup 6-lobed, crenate at the edge ; leaves
glaucous. Native of the Pyrenees. Linnaeus (s. 2.
ed.) describes minor with flowers nodding, three
times smaller than Pseud. Nar. ; scape scarcely
striated ; margin of the cup 6-lobed, undulate,
crisp ; segments of the limb distinct at the base.
Mr. Haworth, therefore, in separating the varieties,
affixed the name minor to a wrong one, with the
segments imbricating.

Var. 2. Humilior. — Minor. Bot. Mag. 1. 6. Minimus,
absque ratione. Haworth. Leaves closer to the
ground, but before expansion touching it; segments
imbricating.

Var. 3. Cuneifolius. — Salisbury. PI. 39. f. 24. Spec.
Herb. Banks. Segments of the limb not imbri-
cating.

Only one of the above is known to me, except the speci-
mens in the Banks. Herb, to which the names are affixed by
Mr. Salisbury ; but Mr. Haworth states that he was ac-
quainted with the three, which he injudiciously distinguished
as species, and two other varieties, one with leaves larger
and more erect than minor, the other narrower. They are
very hardy, and flower in March.

Var. 5. Erectior. — Minor. Haw. This Mr. Haworth

300

AMARYLLIDACEjE.

fancied to be the N. minor of Linnseus's herbarium.
Mr. Haworth adds two other varieties; one with
leaves still more erect, another with leaves nar-
rower. It is quite immaterial what variety Lin-
naeus may have placed in his herbarium as identical
with minor. The name belongs to the plant he
first distinctly described, whatever confusion he
may have afterwards made in this as in many other
cases, either not observing the difference, or not
thinking it worthy of notice.

Var. 5 ? Nanus. Haw. — Only known to me by his de-
scription ; probably a variety of minor, since Ha-
worth first called it minor ft. conspicuus. N. Rev.
p. 112. Sulphureous, semipatent, limb half as long
again as the cup, which is yellow, lobed, and cre-
nate. He notices a narrow-leaved variety with
limb quite patent, and a double variety. I possess
a bulb of Haworth's nanus, but have not yet seen
its flower.

Var. 6. Pumilus. — PI. 39. f. 23. Salisbury. Spec. Van
Royen. Herb. Banks, from the Leyden Garden. A
very diminutive variety of minor.

2. Pseudonarcissus. — Cup the length of the limb. Linn.
Scape two-edged, striated, much more hollow than
that of minor ; style ^ to about § of an inch shorter
than the cup ; filaments adnate close to the base of
the tube ; tube about f long, sometimes § ; sepals
varying in width from |- to -|ths ; cup sometimes
with six lobes cleft almost regularly, sometimes
with none ; more reflex, when cleft; about If long;
limb paler yellow than the cup. — Filaments adnate
near the base of the tube, with a marked inequality.
See PI. 38. f. 40.

Var. 1. Ryticarpus. — Haw. capsulis rugosis. That
which grows wild in the parish of Spofforth, in
Yorkshire, and the neighbourhood, has the germen
very wrinkled as soon as the flower begins to fade,
and is probably the variety thus named.

Var. 2? Festalis. — PI. 43. f. 3. Salisbury. The speci-
men in the Banks, herb, to which Mr. Salisbury
has affixed his name festalis is a plant of small size,
seemingly intermediate between minor and Pseu-

AMARYLLIDACEvE.

301

donarcissus, and I am inclined to think it may be
Mr. Haworth’s Nanus. The variety Eng-1. Bot. t. 17.

I suppose, has smoother capsules than that which
is found in Yorkshire, hut I have not had an oppor-
tunity of comparing them. Haworth enumerates
three shades of colour, besides two double varieties,
and his A. serratus and praecox are two more. A
reference to plate 40. f. 5, 6, 7, 8. which represent
four varieties that grew in the same wild sod of
about 16 inches diameter, taken up at random in a
pasture at Spofforth before the flowers were blown,
and exhibit great diversity of proportions, espe-
cially the limb longer and shorter than the tube,
and some difference of colour, are sufficient to shew
the futility of the features by which the species of
Haworth are distinguished. Fig. 5. and 6. are
from flowers which had been pressed under paper
before the outline was made, and the apparent
width of the tube must not be compared with 7 and
8, which had not been pressed, but 5 may be com-
pared with 6, and 7 with 8, and perfect reliance
may be placed on the exact correctness of the out-
line. All had the same wrinkling of the immature
capsule. I entertain doubts whether any plant of
this genus was originally indigenous in Great Bri-
tain. They seem to me to have remained in the
soil in the site of old cottage gardens long de-
stroyed, and to have spread wherever a root may
have happened in the course of centuries to esta-
blish itself, but I cannot pretend to assert that such
is the case.

Var. 3. Nobilis. — Haw. Pseud. Red. Lil. 158. Ander-
soni. Sabine MS. Limb more patent ; scent un-
pleasant; tube about f of an inch, cup long,
irregularly lobed, plaited, crenulate, a little re-
curved ; segments equal to the cup, rather acute,
tortuous ; sepals fths wide, petals not \ ; style
1 1-1 6ths shorter than the cup; peduncle 5-16ths
long ; germen ^ ; leaves half an inch wide, a little
glaucous.

Var. 4. Telamonius. — Haw. This agrees with the other
varieties in having the limb paler than the cup, the

302

amaryllidacea:.

tube tinged with green, the peduncle from i to |
an inch long, as well as in the length of the style.
Mr. Haworth states that in rich soil this flower
attains an unusual length, but I have not seen it in
such a state.

Mr. Haworth notices three seminal varieties of Pseudo-
narcissus, differing in having the limb paler in different de-
grees, and he distinguished another as a species under the
name Serratus, with three variations, the cup being specified
as serrate, though he also calls the cup of Pseudonarcissus
serrate ; and the only difference that can be extracted from
his characters of Pseudonarcissus and Serratus are that the
limb of the former is equal to the cup, and the latter shorter,
though with his usual inconsistency he has a variety of Ser-
ratus limb nearly equal. The uncertainty of that feature
has been stated above.

I am informed by a gentleman, who has paid much atten-
tion to the wild bulbs of Italy, that Pseudonarcissus in the
vicinity of Naples is oftener found in a wild state with
double than with single flowers. Such plants, I think, must
be vestiges of ancient horticulture. I am also told that
there are many variations of this plant peculiar to different
localities in Italy.

3. Bicolor. — Linn. Sys. ed. 2. Filamenta prope tubi
basin adnata, insequalitate fere obsoleta ; stylo vix
semunciam corona breviore. Tube about half an
inch long, or 1-1 6th more, very pale; petals very
little narrower than the sepals; cup yellow, imper-
fectly lobed ; style about half an inch shorter than
the cup ; filaments adnate near the base of the
tube, with the inequality almost obsolete. See
PI. 38. f. 36. 38. 39.

Linnaeus’s specimen of N. bicolor has the tube only ^
long, and probably belongs to the next closely allied species,
but his comparison of the plant with Ps. nar. (ed. 2.) points
to this species with the longer tube. There is no reason for
supposing that he founded the species on the specimen in his
herbarium, when that is found to disagree with his published
description, and it would be inconvenient to reverse the
names at present in use.

Var. 1. Lorifolius. — Haw. N. bicolor. Bot. Mag. 29.
1187. Flower sw eet, but not so fragrant as brevi-

AMARYLLIDACEiE.

303

flos. Limb about fth of an inch shorter than the
cup ; sepals fths wide; petals scarcely 1 - 16th nar-
rower; cup 1J long, lobed irregularly, plaited,
crenate ; style scarcely half an inch shorter ; leaves
9-16ths wide.

Var. 2. Breviflos. — Haw. Limb a little tortuous, just
shorter than the cup; sepals fths wide, petals fths;
cup If long, almost regularly lobed ; style half an
inch shorter; leaves 5-16ths wide. Delicately
fragrant.

Var. 3. Anceps.- — Haw. Scape more flattened ; scent
powerful, but less agreeable, like the sour-creamy
smell of gladiolus hirsutus ; sepals an inch wide,
petals fths ; cup near 1^ long ; tube fths.

4. Tubaeflorus. — Filamenta tubi ipsi basi adnata inaequa-

litate fere obsoleta, stylo ultra semunciam corona
breviore ; tubo vix semunciali ; limbo pallido laci-
niis latis.

Var. 1. Princeps. — Tubaeflorus. Salisbury. PI. 38. f. 34.
This and the following variety differ from all the
varieties of bicolor in having the filaments parting
from the tube still nearer to its base, as may be seen
by the outlines I have given, but they are very closely
akin to bicolor, and Linnaeus seems by his speci-
men not to have distinguished them. Tube fths;
limb broad, patent, very pale sulphur, If long;
cup equal, yellow, margin reflex, irregularly lobed
and serrate.

Var. 2. Crenulatus. — Sabine MS. PI. 38. f. 35. Dis-
tinguishable from tubaeflorus chiefly by the cup
being less cleft and recurved at the margin, which
gives it the appearance of being longer ; but I have
seen so much variation in the crown of tubaeflorus
in bulbs that were offsets from the same stock, that
I doubt its being a variety permanent by seed.

5. Moschatus. — Style shorter than the cup and

longer than the stamens ; tube inch long, flow-
ers whitish. Linn. ed. 2. says, “cup cylindrical;
margin obsolete, repand, not toothed or crisp;” but
it is quite needless to inquire which of the white-

304

AMARYLLIDACEA3.

flowered varieties he had in view, for it is prepos-
terous to distinguish them as separate species.

Var. 1. Candidissimus. — Red. lil. 3. 188. N. Moscha-
tus 8. Bot. Mag. 32. 1300. Segments tortuous,
sulphurescent, turning white, equal to the cup ;
flower less ; tube half an inch long ; style a quarter
of an inch shorter than the cup.

Var. 2. Tortuosus. — Haw. N. Moschatus Bot. Mag. 24.
924. Limb shorter than cup ; not so white, larger;
tube above § inch long ; style \ inch shorter than
the cup.

Var. 3. Cernuus. — Haw. Sweet. Br. f. g. s. 2. v. 2. 101.
Sulphurescent creamy white.

Var. 4. A'lbicans. — Haw. Park. par. p. 100. t. 101. f. 1.
Segments of the limb ovate-lanceolate, flatter,
whitish ; cup crenate, sulphurescent. This variety
Mr. Haworth says that he saw in 1831 at Young’s
Nursery at Epsom.

There is a double-flowering variety of Ajax Moschatus in
cultivation, of which the root is communicated to me by
Mr. Penny, but I have not seen the leaf or flower. Haworth
refers it to Parkinson’s lesser variety. It is represented in
Sweet’s Br. fl. g. ser. 2. v. 2. 101.

5. Luteus. mihi. — Perianthium luteum concolor; ped.
et tubus | unc. filamenta prope basin tubi inaeq.
adnata ; stylus corona circ. unc. brevior. PI. 38.
f. 33. Perianth all yellow ; ped. and tube f of an
inch long ; filaments adnate, unequally near the
base of the tube ; style about an inch shorter than
the cup.

Var. 1. Maximus.— Haw. The finest variety.

Var. 2. Major. — Bot. Mag. 2. 51. Found on hills near
Limoges.

Var. 3. Propinquus. — N. major /3. Bot. Mag. 32. 1301.
f. sup. ; cup straighter than the two former. I have
a variety with the cup ^ of an inch longer and
more expanded, which is probably Spurius of
Haworth, but not worth distinguishing.

Var. 4. Obvallaris.— N. major, y. Bot. Mag. 32. 1301.
f. ima. N. Sibthorpi. The lower of the two figures
in the plate, Bot. Mag.

AMARYLLIDACEyE.

305

Mr. Haworth has, besides an Ajax Lobularis growing
near Tenby in Pembrokeshire ; limb exactly twice the length
of the tube, cup three lines longer, six-lobed, which he con-
siders the stock of the common double deep-yellow daffodil;
Rugilobus, limb a third part longer than the tube, cup 6-
lobed, wrinkled ; and Cambrieus, differing from the last
merely in the lobes being crenate and but little wrinkled,
and he makes them three distinct species ! They have been
noticed by no other person, and belong, as far as I can judge,
to Pseudonarcissus, having the limb darker than usual.

6? Abscissus. — Haworth. Park. Par. 107. f. 1. Pseudo-
narcissus tubo quasi abscisso. A daffodil with the
cup as if its margin had been clipped. This plant
and the following have the limb paler yellow than
the cup, and are probably varieties of Pseudo-nar-
cissus. Parkinson received this from the Pyrenees.

7? Hexangularis. — Haw. Park. Par. 101. f. 5. Cup
hexangular, dipt. These, together with albus, and
two other daffodils, differing from minor in having
the cup, as if dipt (but only known by the plates
in Rudb. Elys. 68. f. 4. and 72. f. 15), have been
most injudiciously erected by Mr. Haworth into a
genus Oileus, on account of no other peculiarity
than the entire and straight margin of the cup.
They have been long lost ; and the description of
them is too imperfect to enable their proper place
to be ascertained with certainty. The two former
are probably referable to Pseudo-narcissus, the
two latter to minor. Haworth has an Ajax albus
from Rudb. Elys. 70. f. 7. said to have the perianth
white, the cup not fringed, and the style exceed-
ing ; the plant about equal in size to minor. On
reference to Rudbeck, I find a figure with some-
thing indistinct like a curved style. In the text
he does not call it white, as Haworth says, but
white with a large yellow cup, and lie mentions
three varieties, one with the cup fringed, and one
not fringed. No reliance can be placed on such
a figure and description. It comes nearer to Ajax
Sabinianus than to any other plant.

N. Cyclamineus (Rudb. Theat. flor. 20. called there N.
Hispanicus minor, luteus, amplo calyce, foliis reflexis) is

x

AMARYLLIDACEiE.

30 G

another absurdity which will never be found to exist. It is
probably an execrable representation of Ganymedes capax
with the margin of the cup incorrectly given ; and, looking
at the rest of Rudbeck’s figures, I have no hesitation in re-
jecting it as a non-entity. There is no account of the quarter
from which it was obtained.

9. Sabinianus. — PI. 38. f. 41. Bot. Reg. 9. 762. Limb
white, 1|- inch long; cup yellow, tube f ; style
lj long; petaline filaments adnate, ~ of an inch
from the base of the tube, sepaline near
This is a very distinct and remarkable species, which was
brought to notice by a specimen and bulb, sent (as I am in-
formed) to Mr. Sabine from the Botanic Garden at Oxford,
where it may have been growing unnoticed for the last two
centuries, as very little attention seems to have been paid to
its old inhabitants. Mr. Sabine, who has paid great atten-
tion to the Narcisseae, having appeared to think that it was
not a natural species, but a cross-bred garden production, I
have given particular consideration to that point, and I think
it must be a natural species. The reasons for suspecting a
spurious origin are, its departure from the other known spe-
cies of Ajax in a higher adhesion of the filaments to the
tube, a very little tendency to a curvature of the anthers,
and a shorter proportional cup, and in all those features it
draws nearer than any other species of Ajax to the general
character of the genus Queltia. Two considerations present
themselves: in the first place, whether the plant offers any
indications in itself of a spurious and imperfect structure,
such as might be expected from a bigeneric mule, or a plant
cross-bred between two sections of a genus, differing in struc-
ture, if they be considered as forming one genus ; secondly,
whether we know any species of Ajax and Queltia, from
whose accidental intermixture in cultivation, this particular
form of plant could seem likely to have been produced. I
find no indications of imperfection ; the ovules are plump
and well defined, and have the usual form, with the raphe
and chalaza usually observable in other species of Ajax ; nor
did I observe in the fresh specimen any apparent sterility of
the anther. I have known no intermixture of any two vege-
tables, so well distinguishable from each other as Ajax and
Queltia ; and, if their union can possibly take place, I should
conceive the produce must be sterile and imperfect. If that

AM ARYLLID ACEAL.

307

consideration be passed over, it is evident, that as it has a
yellow cup and a whiter limb than any species of Ajax, it
must (if cross-bred) be derived from a two-coloured Ajax
and a Queltia with a white limb ; and the only such Queltias
that we have are Macleaana, and Montana with its varie-
ties ; but there is decisive proof that it cannot be so bred,
because its style is as long as that of any two-coloured Ajax,
and the style of the two white-tinted Queltias is from half to
^ths of an inch shorter ; the style of the cross-bred plant
would have been therefore abbreviated, instead of being as
long and proportionally longer. The style of Q. incompara-
bilis is also quarter of an inch shorter, and the same objec-
tion applies to it besides, that the palest variety of that plant
has not the limb so white as A. Sabinianus, and could not
therefore by its influence have effaced the yellow tint of the
parent Ajax ; besides which, the limb of the whitish incom-
parabilis is lfths of an inch long, and that of the two-
coloured A. bicolor and tubaeflorus in their varieties being-
longer than that of Sabinianus, the cross-bred plant would
not be likely to have it abbreviated so as to be -|ths of an
inch shorter than that of either parent. It does not approxi-
mate in any manner to the reflex species forming the genus
Ganymedes, nor to N. poeticus, nor to the Hermiones; and
any cross, except with the Queltias above named, would have
produced a still more different result, I think therefore that
I must discard the suggestion of its hybrid origin, notwith-
standing my respect for the quarter from whence it has been
derived. I hope that I shall be able to obtain a bulb of it,
being very desirous of more intimate acquaintance with
the plant by cultivating it. Haworth’s genus Oileus,
without a single feature applicable to a generic character,
contains the varieties of Ajax, which are distinguished only
by the margin of their cup being even and not in the
least reflex, and whose existence depends on the faith of
old books. His Diomedes unites Ajax Sabinianus with
Queltia Macleaana, the two plants through which the genera
approximate, but which are much more essentially different
from each other than from the other species of their re-
spective genera.

76. Ganymedes. — Style straight, slender ; filaments adher-
ing to the upper part of the tube with greater diversity
than Queltia ; sepaline stamens more prolonged ;
limb decidedly reflex ; (cup equalling or shorter than
x 2

308

AM ARYLLIDACEjE.

the limb ; tube slender, drooping ; capsule erect)
seed oblong with elevated raphe and chalaza. Plants
with a small bulb and leaves, scape slender, and con-
stitution delicate; 2-7-flowered. The anthers re-
quire further examination ; those of a dry specimen
of G. striatulus now before me are not like those of
a Queltia ; I have had no opportunity of examining a
fresh anther of Ganymedes, since I have fully inves-
tigated the structure of the other genera.

1. Capax. — N. Calathinus. Red. Lil. 3. 177. Perianth

pale yellow; tube and cup each about J-ths, limb
about -|ths of an inch long : style shorter than cup.
Native of the isles of Glenan dep. of Finisterre.

2. Reflexus. -Brotero Flor. Lus. — N. Calathinus Red.

Lil. 410. Perianth sulphureous white ; tube and
cup each about limb ^ inch long ; style shorter
than the cup. On the mountains of Gerez, near
Amaranta and elsewhere in Portugal, and in the
isles of Glenan, The Portuguese plant is called
sulphureous white, that of Glenan is figured quite
white by Redoute.

3. Pulchellus. — PI. 39. f. 16. — N. triandrus, v. luteus.

Bot. Mag. 31. 1262. Sweet Br. f. g. Ser. 2. t. 99.
Tube limb about yellow; cup about half an
inch, very pale.

Var. 2. — Parvicorona, misapplied by Flaworth to stria-
tulus with which it does not agree ; the main dif-
ference of striatulus being the prolonged style.

4. Cernuus. —Illus cernuus. Haw. N. triandrus. Bot. Mag.

2. 48. Tube and cup about ■§, limb l long, whitish.
Style shorter than the cup according to the Bot. Mag.
though Haworth in his character of Illus says longer.

N. triandrus of Linnaeus agrees with this, except that it
is called absolutely white, but there may exist a whiter va-
riety than that figured in the Bot. Mag. The name which
he gave by a great error, having supposed it to bear only
three stamens, because the petaline were hid within the tube,
cannot with any propriety be preserved.

5. Concolor. — Sweet. Br. f. g. Ser. 2. 113. Perianth

pale yellow; tube about § long; cup not ^ ; limb
f ; style exceeding the cup.

AMAKYLLIDACE.E.

309

6. Striatulus. — Haw. According to Haworth, style

exceeding the cup ; segments yellow, tortuous, not
overlapping at the base, cup sulphureous white,
faintly striped, one-third the length of the limb,
margin very repand. This little plant is cultivated
at the Chelsea Garden, and Mr. Anderson assures
me that it is very distinct. It bears three or more
flowers.

7. Nutans. — N. trllobus. Bot. Mag. 24. 945. not trilo-

bus of Linnaeus. Cup deeper yellow than the
limb, style much longer.

N. totus albus reflexus, Morison v. 4. t. 9. f. 13. taken
from Swert Florileg. is a very bad representation of a white
reflex 3-flowered plant; with the style not visible. It is too
bad a representation to deserve much credit ; but it is cer-
tainly not juncif. alb. Park. par. 93. f. 1. which is identified
with it by Haworth, for that plant has not a reflex limb, and
is probably N. dubius of Redoute, an Hermione.

N. coronatus Rudb. El. 75. f. 6. is a 2-flowered plant,
with the limb not reflex ; the figure was not taken from a
living plant, but had reference to N. coronatus Lobel, i. e.
a one-flowered white Ganymedes, with the style salient, of
which Rudbeck’s is an unfaithful copy. The other plant,
which has been called coronatus, or Coornei, from an execra-
ble figure in Morison and Rudb. El. is copied by them, and
others, with like infidelity from Lobel. advers. and called
white, though Lobel says, white with a large yellow cup.
Lobel states that he never saw the flower, but gives the repre-
sentation from a drawing, by Mr. C. Coorne, sent to him in
England, from the Continent, in 1604. The drawing has
been evidently made from a withered flower, of which the
limb was quite shrivelled. It was most probably Ganymedes
striatulus, as the style was longer than the cup. Little de-
pendance is to be placed on the reported colour of an old
drawing from a withered specimen, made by an unskilful
person. The result of a careful examination of the old en-
gravings of Narcisseae, is that the various authors copied from
each other with great inaccuracy, and not from nature, and
that no faith is to be given to any remarkable appearance in
their works which shall not be verified by specimens. N.
coronatus and Coornei must be expunged.

A root called N. capax flore pleno is communicated to

310

AMARYLLIDACEA2.

me by Mr. Penny. It is stated by Haworth to be the
double of Redoute’s calathinus, Lil. 177. whether correctly
or not, I have yet had no opportunity of judging.

Haworth's genus Assaracus is a specific description of the
yellow and the white reflex Narcissi, Ganymedes calathinus of
Redoute, Lil. 410 and 417. His Illus is a specific description
of Ganymedes cernuus, the triandrus of Linnceus, separated
as species, the latter being described as having the limb a
little longer, and the scape more flattened.

77. Queltia. Mock-narcissus. — Style straight, more or less
attenuated ; filaments adhering alternately to the
upper part of the tube, straight, affixed to the back
of the anthers below the middle ; anthers linear,
erect, recurved, with the margins of the cells meeting
behind, above the attachment, but not enveloping
the filaments, (tube sub-cylindrical, more or less
widened, crown shorter than the limb) shorter than
the tube, or nearly equal.

1. Macleaana. — PI. 39. f. 1. N. Macleai. Bot. Mag. 52.

2588. Diomedes minor. Haw. Park. Par. 71. f. 7?
Tube 9-16ths, cup half an inch long, yellow ; limb
f white ; style |-tli shorter than the cup, \ longer
than the anthers. This species perfectly agrees
in material points of structure with the genus
Queltia ; it approaches nearer to Ajax than the
rest, in the greater proportional length, and per-
haps a little less expansion of the cup, points which
are not of absolute importance, and do not affect
the generic character. Its union with A. sabini-
anus, in a separate genus by Haworth, is evidently
erroneous. They form the approximating points
of the genera to which they belong respectively,
but they do not agree with each other in structure.
This plant is a perfect Queltia. I think Parkinson's
plant, above quoted, may be the same, and I
recognize an appearance very like to it in the
Theatr. flor. with wider segments, which seems to
have escaped Mr. Haworth.

2. Montana. — PI. 39. f. 4. Bot. Reg. 2. 133. N. mon-

tanus. Park. Par. 71. f. 6. Tros poculiformis. Haw.
Perianth white; tube 11- 16ths, cup limb |-th of
an inch long ; style just exceeding the anthers,

AMARYLLIDACEjE.

311

about |-th shorter than the cup. Mr. Haworth had
a variety one-third smaller.

Var. 2. Galanthifolia. — Tros. Haw . said to differ in
having the limb more expanded and tortuous, and
the cup more plaited ; the style sometimes exceed-
ing it. There is some apparent difference in the
bulbs, but I have not seen the flower of the latter,
which seems to be of more difficult culture.

3. Fce'tida. — Leaves rather glaucous ; tube § to one
inch long, shorter than the limb, stout, wider up-
wards and strongly six-ribbed ; limb patent, subtor-
tuous, 1^ to 1^ long; cup plaited, six-lobed, about
half an inch long, | to 1 inch wide. (Flower stinking
in all that I have seen, and petals more tortuous
than the sepals).

Var. 1. Incomparabilis. — Bot. Mag. 4. 121. Cup a little
darker yellow than the limb.

Var. 2. Aurantia. — PI. 39. f. 5. before expansion. N.
Gouani. Red. Lil. 158. Cup edged with a deeper
colour, tending more or less to orange. Of this I
have two variations, and Redoute’s plant differs a
little; a specimen from Mr. Sabine was not exactly
similar to any of the three. The double variety
is called Butter and eggs.

Var. 3. Semipartita. — PI. 39. f. 6. A marked variety,
with a paler tint of yellow, the cup more patent,
regularly and deeply lobed.

Var. 4. Concolor. Haw. — Homochroos. Schultes. N.omn.
max. fl. et cal. flavo. Park. Par. 68. 2. This
plant is only known by the account of Parkinson,
who says it differs from incomparabilis in nothing
but having the cup of the same colour as the limb,
and he adds that it is sometimes 2-flowered. It
has not been forthcoming in our days ; the double
variety called sulphur-crown may possibly have
sprung from it ; but I understand that a little ap-
pearance of darker colour in the middle is observ-
able in the doubling of that flower.

Var. 5. Grisea. — PI. 39. f. 7. N. max. griseus cal. flav.
Park. Par. 69. 3. Q. alba. Haw. Q. nivea? Haw.
Parkinson says, “ glistening whitish grey, cup yel-

312

AMARYLLIDACEiE.

low ; peradventure but a difference of seed from
the former.” The limb of this variety is not really
white, but has a dull yellowish tinge, and is inac-
curately termed white by Haworth. I cannot learn
that any person has seen a white variety, though
he had another which he called snow-white. The
orange phoenix is the double of this variety. The
double narcisseae are apt to degenerate in some
seasons, or by accident, to semidouble and single,
and I have seen the Butter and eggs produce a per-
fect single flower of Aurantia. I have seen the
orange phoenix, when becoming semidouble, with
the three styles quite distinct. They were parted
by the process of doubling the flower, and when it
became accidentally single, they were not re-
united. Mr. Sabine informs me that he has twelve
varieties of foetida, but only one single one with a
whitish limb. The variations which I have not
enumerated are of little importance. I think the
convenience of limiting the name incomparabilis to
that variety to which it was affixed in the magazine,
and calling the species by a more applicable gene-
ric name as stinking, will be universally admitted.
Incomparabilis does not appear to be the stock
from which they have all sprung ; the orange rim,
which it wants, seems rather the prevailing feature
amongst them.

4. Orientalis. — PI. 39. f. 2. tube |ths or more; limb

-|th or one inch long ; margins reflex or tortuous ;
cup near I an inch long, about I wide, irregularly
lobed and 3-cleft. N. orientalis /3- Bot. Mag. 24.
948. Schisanthes. Haw.

Var. 2. PI. 39. f. 2, 3. — Cup more patent and deeply
gashed ; leaf keeled, half an inch wide. f. 2 was
made before the expansion of the flower, which was
perhaps not quite fall grown. A specimen from
Mr. Sabine differed a little from both varieties.
The second I had ; the first was from the Hort. Soc.
garden. The species was first united with incom-
parabilis in the Bot. Mag. and afterwards with an
Hermione. It is, however, a perfect Queltia.

5. Odora. — Leaves deep green ; perianth uniform bright

AMARYLLIDACE.E.

313

yellow ; tube slender, about f of an inch long;
limb about an inch long ; cup about half an inch
long, i or | wide; style ^ of an inch shorter than
the cup. Ovules usually in three rows, full, but
irregular.

Var. 1. Campernelliana.— Philogyne Campernelli. Haw.
The finest variety ; scape often 4-flowered ; flowers
equal to, if not exceeding, the var. called N. cala-
thinus, Bot. Mag. and rather more brilliant in
colour.

Var. 2. Calathina. — PI. 39. f. 8. N. calathinus. Bot.
Mag. 24. 934. Linn. S .ed. 2. et MS. in marg. ed.
1. quoad auctor. cit. non quoad descriptionem.
N. odorus Linn. herb.

Linnaeus’s specimen of N. odorus is from the Upsal gar-
den, a two-flowered specimen of a plant of this species, with
a six-lobed cup, the tube and limb together measuring barely
one inch and |-, the limb exceeding the cup This is ^
shorter than calathina of the Bot. Mag. and comes nearer to
v. laeta ; but as I have adopted Linnaeus’s name odora for
the whole species, it is needless to discuss the difficult point,
which variety lies dried in his herbarium. He never had a
specimen of N. calathina, but named it, in MS. on the mar-
gin of the first edition, and published it in the second with a
reference to Clusius, No. 1. juncif. 9. and Rudb. El. 2. p 60.
f. 5. Rudbeck’s figure is a very unfaithful copy of that of
Clusius with a reference to it ; therefore the original plant
of Clusius is the thing meant, and that corresponds very
closely with N. calathinus Bot. Mag. Rudbeck exaggerated
the size of the cup in his engraving, on the faith of which
Linnaeus described it as having the cup and limb nearly
equal, but no such specimen has ever been forthcoming:
and, as Linnaeus had no specimen, but meant to describe the
plant of Clusius, we must look to Clusius alone, rejecting
the garbled figure of Rudbeck. Our calathinus seems a
little more indented than the flower of Clusius, but there is
no other difference. M. Decandolle, in Red. lil., in conse-
quence of Linnaeus’s statement that the cup and limb were
equal, applied the name to a plant with a reflex limb; but
the reference to Clusius’s figure, and even Rudbeck’s, as
well as Linnaeus’s reference to Tazetta, shews that it was not
a reflex flower. His odorus was probably a variety a little

314

AMARYLLIDACE J2.

different from any of those we have now in cultivation. No-
thing can better shew the unsettled state of Linnaeus’s mind
concerning the Narcissi, than his specimen of N. tenuior first
marked by him trilobus, which name he afterwards erased
and substituted Tazetta !

Var. 3. Calathina minor. — PI. 39. f. 9. Flower con-
siderably smaller, but similar in its proportions.

Var. 4. Rugulosa. Haw. — PI. 39. f. 11. Leaf rounded,
and not keeled on the back, strongly nerved, more
concave than heminalis ; segments wider and more
patent ; cup more distinctly lobed. Flowers three
or less. Tube f, cup about i, limb nearly one
inch ; sepals ll-16ths, petals 9-16ths wide.

Var. 5. Interjecta. Haw. — PI. 39. f. 10. Leaf like rugu-
losa ; lobes of the cup more curled ; tube f , cup
about limb about an inch long ; sepals petals
half an inch wider.

Var. 6. Heminalis. Haw. — PI. 39. f. 12. Leaf ^ of an
inch wide, keeled, strongly nerved on the back,
fiat with upstanding margins in front ; scape
roundly two-edged; tube about 11-1 6ths, slender
at the bottom, much enlarged upwards ; limb near
an inch long, cup \ or more, plaited, irregularly
and not deeply six-lobed; perianth of a deeper
yellow than any other variety; very fragrant.

Subvar. Heminalis, minor. — Similar in all parts, but
smaller; leaf not 3-16ths wide.

Var. 7. Triloba. — PL 39. f. 13. Leaf as in rugulosa;
tube f , cup or more, slightly three-lobed ; se-
pals 9-16ths, petals 5-16ths wide.

Subvar. Laeta. Curtisii. Haw. — PI. 39. f. 14. Rather
smaller than the last, and less conspicuous, and
only distinguishable as an inferior variety.

No reasonable person, after comparing the exact repre-
sentations of these flowers in plate 39 of this work, will differ
from me in considering them to be varieties of one species.

6. Juncifolia. — PI. 43. f. 1. Requien. Spec. herb. Ben-
tliam. I find three varieties of this little plant in
Mr. Bentliam’s herbarium. Leaves very narrow,
1-3 flowered, flower bright yellow, limb about

AMARYLLIDACE.-E. 3J5

twice the length of the cup, of which the margin
is undulated and indistinctly lobed.

Var. 1. From Restinctieres near Montpelier, in dry
stony places. Scape about 7 inches high ; leaf
scarce 1 - 16th wide. The Castilian rupicola of
Schultes is probably this plant.

Var. 2. Scape only four inches. Pont du Gord.

Var. 3. Tallest; leaves about 13 inches long and wider.
Capouladoux near Montpelier.

7. Pusilla. — PI. 43. f. 2. Specim. Masson. Herb. Banks.

ex Hispania inter Ayamonte et Huelba. Foliis
filiforinibus subbiuncialibus scapo gracillimo 2^
unciali bifloro, spatha unciali pedunculis longiore,
tubo subcylindrico semunciali luteo inferne vire-
scente, limbo ^ unc. luteo (corona lutea erosa? vel
sex lobata?)

This most singular diminutive species was found by Mas-
son between A3^amonte and Huelba in Spain, and I cannot
find that it has been noticed by any other person, unless the
obscure plant which has been named jonquillioides, from a
mutilated one-flowered specimen in the herbarium of Willde-
now, from Portugal, be referable to this, but there is nothing
in the imperfect description of it, but the filiform leaf to iden-
tify them. My outline gives the exact dimensions of the plant.

8. Jonquilla. — PI. 39. f. 15. before expansion. Bot.

Mag. 1. 15. Leaves narrow and rounded, chan-
nelled on the surface. Tube 1|-, slender, limb
about |ths of an inch long, cup |-th or more.
Flower all yellow, very fragrant.

This plant differs from the foregoing species of Queltia in
having shorter anthers, nor am 1 satisfied that their struc-
ture is precisely similar. Its seed differs a little from any
seed of Narcissese that I have seen. Not having been able
to obtain the ripe fruit of the Queltias, I cannot come at pre-
sent to a decisive opinion whether the jonquill is a species of
Queltia or a genus by itself. Its capsule is erect and not
sloped as those of Queltia appear to be, yet it does by no
means agree with Hermione, to which Sweet referred it.

It flowers later than any Queltia, and has much affinity
to the autumnal species of Narcisseae, which I have no op-
portunity of examining in a fresh state. In all of them the

316

AMARYLLIDACEjE.

capsule is erect, and the tube very slender, from 6 to 8 times
as long as the cup. I cannot ascertain whether their anthers
conform with those of the jonquill.

Var. 1. Major, var. 2. Media, var. 3. Minor, are culti-
vated at the Chelsea garden. I believe the difference to be
very little. I could observe none in the stature of the plants
after the flower was faded. Haworth adds a fourth and
smaller variety, which he calls parvicorona, but I find no
trace of it. Haworth's definitions of his genera Schisanthes,
Jonquilla, and Chloraster, are merely inadequate descriptions
of three species. His genus Tros is a specific description of
Queltia montana, with galanthifolia, which is perhaps a
variety of montana; his Schisanthes of Queltia orientalis, his
Philogyne of odora and its varieties.

9 ? Pumila. — PI. 41. f. 26. N. pumilus, Red. lil. 7. 409.
Leaves very slender, scape slender short, flower all
white, cup about ^ the length of the limb, crenate.
Native situation not ascertained. Only known by
the figure and description in Redoute. I believe
it to be a true Queltia.

78. Narcissus. — Style straight, slender; filaments straight,
free at the point only, the sepaline adnate to the
mouth of the tube, the petalinejust below; anthers
short, with the summit recurved, the margins not
meeting behind. Tube slender, cylindrical, widened
at the mouth ; cup short, spread. Capsule declined ?
and seeds, as far as I have seen, roundish-oblong,
unless angular by contact, wrinkled, with a slender
raphe and prominent chalaza.

1. Gracilis. — Leaves green, narrow ; spathe 1-3-flowered,
slender ; perianth pale yellow ; style equal to the
tube, seeds less round than those of poeticus.

Var. 1. Princeps. — PI. 41. f. 4. Bot. Reg. 816. Sweet
Br. fl. g. ser. 2. v. 2. Tube 1^, limb near one inch
long, cup darker, crenulate, 3-16ths long.

Var. 2. Planicorona. — PI. 41. f. 3. Leaves narrower ;
tube jj, limb about \ an inch ; cup about 3-16ths
long, yellow, margin even.

Var. 3. Tenuior. — PI. 41. f. 3. Bot. Mag. 11. 379.
Tube limb 11-1 6ths, pale yellow, fading nearly
white ; cup yellow, 3-16ths, margin waved.

AM ARYLLIDACEiE.

317

2. Biflorus. — Leaves a little glaucous; 1-4-flowered ;

limb creamy white, cup yellow; seeds not seen by
me.

Var. 1. Sterilis. — Biflorus. Bot. Mag. 6. 197. 2-flowered,
without pollen or ovides in our gardens, perhaps
from having been raised by offsets three centuries dr
more, without renovation by seed. I see no reason
to think it a hybrid production, for it does not ex-
hibit appearances intermediate between those of any
two narcissean genera, or even species.

Yar. 2. Triflorus. — 2-4-flowered, from the south of
France ; ovules perfect, flower rather smaller and
the white clearer.

Var. 3. Dianthus. — Haworth. Unknown to me, but de-
scribed as 2-flowered, with a very plicate and darker
yellow or orange cup.

3. Poeticus. — Linnaeus. Leaves glaucous ; flowers soli-

tary, rarely, if ever, 2 ; limb pure white, unless a
small yellow mark at the base of the segments ;
cup yellow, margined with red or deep orange.
Seeds rounder than those of Gracilis.

Early flowering. — April.

Var. 1. Grandiflorus. — Sabine MS. Poetarum. Haworth,
absque causa nomen Linnaeanum ( poeticus ) repu-
dians. Stamen and anther, pi. 38. f. 3. magnified.
The largest and most conspicuous, but there is no
reason for thinking that this variety was particu-
larly alluded to by the ancient poets.

Var. 2. Angustifolius. — PI. 41. f. 1. before expansion.
— Bot. Mag. 5. 193. Leaves narrow ; flowers smal-
ler than v. 1 .

Var. 3? Ornatus. — The flat-crowned saffron rim. Ha-
worth. A doubtful variety, which I have never
seen. It is described by Haworth, as an early va-
riety, similar in appearance with the late flowering
patellaris ; he cites as synonymus to his Ornatus, tri -
podalis, MS Salisbury, tripedalis, Schultes. 1
am not acquainted with this earlier variety.

Var. 4. Spathulatus. — The lesser saffron rim. Haworth.

318

AMARYLUDACE^E.

Smaller and earlier than v. 8, but in other respects
similar ; margin of the cup plicate.

Var. 5? Albus. — The slightly saffron rimmed. Haworth.
Flowers in the middle of April. I have no know-
ledge of the last three varieties, and I cannot find
them in cultivation at present.

Late flowering . — May.

Var. 6. — Majalis. — PI. 40. f. 2. Engl. Bot. 4. 275.
Anther PI. 38. f. 14.

Var. 7. — Recurvus. — PL 40. f. 1. Sweet Br. fl. g. ser. 2.
Leaves with the points always drooping; limb
waved, reflex.

Var. 8. Patellaris. — PL 40. f. 3. Leaves broad, sepals
more reflex than the petals, whence either this or
the earlier ornatus was called tripodalis, corrupted
into tripedalis, implying that the three reflex seg-
ments represent the feet of a tripod.

Var. 9. Stellaris. — Haworth. Park. Parad. t. 76. 4.
Sweet Br. fl. g. ser. 2. 132. — Wild specimen, dep.
d’Aveyron. Herb. Bentham. Segments of the limb
narrow at the base and distinct. Cultivated in the
Chelsea garden, and by the Hort. Soc.

Diminutive. — May.

Var. 10. Verbanensis. — PL 37. f. 2. A very dimi-
nutive plant, agreeing with Poeticus, but smal-
ler in all its parts, with very narow linear leaves,
and reflex limb, tinged with yellow at the base.
Brought to Bolton Percy in Yorkshire a few
years ago, by Mrs. Robert Markham, who found
it growing in a pasture, about a mile from
Baveno, near Lago Maggiore, on the side of
the road to Milan. It is remarkable that this plant
does not seem to increase by offsets in its native
situation, the bulbs being found single and scattered
about the pasture, instead of growing in tufts, like
most others. It flowers about the middle of May
both there and in England. Parkinson has two
small varieties, one with a purple rim, from France,
which Haworth calls purpuro-cinctus ; another
from the Pyrenees, with an orange rim, -which he

AMARYLLIDACEiE.

319

culls croceo-cinctus. They must be nearly allied to
this plant ; and perhaps they might be separated
without impropriety from poeticus, under the name
exiguus.

The seeds of Gracilis, v. princeps and tenuior, are less
round than those which 1 have seen of Poeticus, namely, re-
curvus and stellaris, and, as I can judge by comparison of
the ovules, grandiflorus, angustifolius, majalis, and patellaris.

I did not inspect the ovules of Verbanensis, and although it
has ripened seeds at Bolton Percy, they had been mislaid, and
I have seen none. Judging from the insufficient view I have
been able to obtain of the seeds of some of the Narcisseae,
and the ovules of others, I believe it will be found that each
genus embraces a certain range of outward appearances in
the seed, and that the differences of the seed will establish
clearer limits to the species. There is a like diversity of seed
in Gladiolus, by which l can perceive in the produce of cross-
bred plants of that genus, even the most remote affinity to
G. tristis. To ascertain exactly what diversity in the outward
appearance of seeds is compatible with generic identity is
a point which requires to be deeply investigated, and we must
look to hybridizing experiments, as the test which must estab-
lish the accuracy of any surmises on that subject.

Haworth’s genus Helene is a specific description of the
pale yellow Narcissus gracilis, with its varieties tenuior and
planicorona, with which he absurdly unites the small varieties
of Narcissus poeticus, such as Verbanensis, which may possibly
be separable as a species, but which it is preposterous to refer
with gracilis to another genus, because their leaves are nar-
row ; he further adds to them, pumilus of Redoute, which I
have never seen, but which, in the absence of more accurate
knowledge, seems to me, knowing nothing of the structure
of its stamens, referable to Queltia.

79. Hf.rmione. — Style straight, slender ; filaments conniving,
with a short curved point, alternately inserted ; the
sepaline at the mouth of the tube, decurrent, scarcely
partible from it, attached to the middle of the anthers;
anthers after inversion acute-oval, incumbent, versa-
tile. (Capsule erect ; tube slender, cylindrical, en-
larged at the mouth ; cup shorter than the tube or
limb.) Observ. I had found seeds of Hermione to be
smooth, shining, compressed, with rounded back, an-

320

AM ARYLLIDACEjE.

gular by contact, with prominent margins edged by a
furrow ; raphe sulcate, with prominent margins ;
chalaza inconspicuous ; but I have seen a departure
from that outward form in seed of another Hermione,
shewing that those of the whole genus and suborder
require careful examination to ascertain the range of
differences that is admitted, by which perhaps the
several species ma3r be limited. It should be carefully
observed whether such variations, as shall be found to
occur, may in any manner correspond with greater or
less repletion of the fistular cavity of the scape, which
is remarkable in different species or varietie's of Ajax
and Hermione.

1. Bifrons. — Leaves deep glossy green; perianth yel-
low ; cup more or less 6-lobed, shallow, patent ;
style just exceeding.

Var. 1. Princeps. — N. bifrons. Bot. Mag. 29. 1186.
With me the tube fths, the limb near fths, cup
3-16ths. Mr. Ker imagined that the primuline
variety imported from Holland, degenerated into the
form of this variety, after some years of neglected
culture. This appears to be a mistake, for the fact
was positively denied by Mr. Salisbury after many
years experience. The segments of the limb are
three times as long as the cup, not lapping.

Var. 2. Primulina. — Bot. Mag. 32. 1299. 4-flowered
or under ; cup 6-lobed, wider; limb three or four
times as long as the cup. Haworth. Tube 6-8ths,
limb near 5-8ths, cup almost flat, 6-lobed ; style
1-1 6th of an inch exceeding the anthers.

Var. 3. Compressa. — 2-6-flowered, cup more erect,
crenate, rather cleft into three lobes. Haworth. 1
apprehend N. tripartitus, Hormem. to be this plant.
In the specimens of this variety received from Mr.
Sabine, and from Mr. Ellacombe, I find the tube
-|ths, the limb broad, above fths long, the cup flat,
6-lobed, the style just exceeding the anthers. I
believe Haworth’s description of the cup as erect,
to be an error.

Var. 4 ? Biscrenata. — Haworth. Unknown to me, but
associated with these varieties by Haworth, who
states it to exceed in stature with 6-9 flowers, the

AMARYLLIDACE^E.

321

cup almost 12-lobed, the six lobes being severally
indented ; the limb broad ovate, more than twice
as long as the cup. I can have no certainty from
Mr. Haworth’s insufficient details, whether the plant
he meant to describe belongs to this species. He
does not notice the style, and it might have been
a variety of H. brevistyla allied to v. 10. represented
pi. 41. f. 10 ; but I have just received biscrenata
from Mr. Penn}% nurseryman, at Godalmin, and
am satisfied by the aspect of the bulb and leaves that
it is a variety of bifrons.

2. Tazetta. — Stigma between the upper anthers ; cup
yellow or orange ; limb white, pale lemon, or yel-
low. Observ. The tube is usually from ^ an inch ,
to -|ths long, and the limb seems to have precisely
the same latitude of variation ; the tube is oftener
longer than the limb, sometimes equal, more rarely
shorter. If it were ascertained that the varieties in
which the limb equals, and in which it exceeds the
tube were found so wild invariably in certain locali-
ties, and that the proportions did not vary in cul-
tivated seedlings, it might be advisable to separate
them on that account, and increase the number of
specific names ; but I can find no account of those
varieties in a wild state, and we have no report of
the variabilities of Hermione in cultivation. It is
utterly impossible to identify Mr. Haworth’s named
varieties, for he has not noticed the style, and his
definitions do not enable me to class them. I do not
pretend to detail the multitude of varieties which
are to be found in cidtivation, but I have given the
outline and particulars of all those to which I had
access last year, which will be quite sufficient to
shew in what manner the genus varies, and enable
any person to arrange the further varieties under
the proper heads, till their seeds and their seminal
variability can be more fully investigated.

Tube equal to the limb.

Var. 1. Cupularis. — Limb yellow, cup orange. Salisb.
H. Tr. 1. 361. N. Tazetta. Bot. Mag. 23. 925.
Red. Lil. 1. 17. Tenore 1. 148. N. flavus. Lagasca
N. G. et s. 13. Soleil d’or hortulanorum. With

322

AMARYLLIDACEjE.

me it lias the tube and limb about fths, the cup
about fth long-, orange. Linnaeus, not having seen
the variability of the length of the cup and limb,
defined Tazetta as having the limb three times the
length of the cup. It is true with respect to this
plant, but that proportion does not appear to afford
any safe distinction. Native of Portugal, Spain,
Italy, Greece, Malta, and the coast of Africa. Te-
nore specifies the mountains near Miseno and
Capri.

Var. 2. PI. 41. f. 15. flower a little shorter.

Var. 3. PI. 41. f. 20. Limb pale lemon, cup deep yellow,
about | long, tube and limb half an inch each.

Var. 4. PI. 41. f. 18. Limb white, cup pale yellow,
tube and limb 9-16ths of an inch each. N. Ta-
zetta from the garden of the Hort. Soc.

Var. 5. PI. 41. f. 21. Very pale sulphur, yellow at the
base, cup bright yellow, 5-16ths, limb and tube
fths of an inch long. This comes near the Greek
mountain plant Flor. Graec. t. 388.

Tube shorter than the limb.

Var. G. Longilimba. — PI. 41. f. 22. Very shewy ; limb
bright yellow, broad, fths long ; tube fths, cup
near fths, long, crenate.

Tube longer than the limb.

Var. 7. PI. 41. f. 17. Limb white, shorter than the
tube, cup bright yellow ; limb about fths long.

Var. 8. PI. 41. f. 19. Czar Monarque? hortulanorum.
Limb white, yellow at the base of the segments,
near fths long, tube fths, cup subcylindrical, fths
long.

Var. 9. PI. 41. f. 16. Imported from China; tube about
fths long, limb white shorter, cup pale yellow.

Var. 10. Sexlobata. PI. 41. f. 23. N. orientalisy? Bot.
Mag. 24. 946. Segments broad roundish, pale
lemon-colour fading to white; cup patent, 6-lobed;
tube fths long. A very beautiful and hardy
variety.

Var. 11. Lacticolor. — Haworth. N. Tazetta Flor. Graec.

AMARYLLTDACEiE.

323

358. Limb sulphur-coloured, cup orange, tube
above -|ths long.

3. Papyracea. — Bot. Mag. 24. 947. N. unicolor. Tenore.

FI. N. 1. 26. Perianth pure white, cup short,
stigma between the upper filaments. Bulb blacker ;
flowers 8-15. (Cup five times shorter than the
limb?) Grows near Mount Vesuvius. Only dis-
tinguishable from Tazetta by the white cup, darker-
coated bulb, and peculiar scent.

4. Dubia. — Red. Lil. 8. 428. Leaves narrow; perianth

all white ; style shorter than all the stamens ; bulb
small. Flowers 2-3 when wild, 6 when culti-
vated.

Var. 1. PI. 43. f. 6. Two-flowered, 7 inches high; leaf
little more than 1 - 16th wide. Avignon. Herb.
Bentham.

Var. 2. PI. 43. f. 7. Three-flowered, tube slenderer ;
leaf similar to var. 1. Restinctieres, near Mont-
pelier. Herb. Bentham. Flowers in March.

Var. 3. PI. 43. f. 8. Tube shorter and thicker, leaf
3-16ths wide, 8 inches long. Near Montpelier,
Herb. Bentham.

5. Corcyrensis. — PI. 37. f. 2. Leaves half an inch

wide, glaucous ; scape 1 -flowered (Qu. whether
evermore?) 6-7 inches; spathe 1-valved, looped;
limb fths of an inch long ; cup yellow, scarce 4th,
trifid ; limb near fths, reflex, segments narrow,
acute, pale sulphur-colour ; peduncle above an
inch ; style shorter than the stamens. I have had
this remarkable plant, which was found growing-
wild in Corfu, near thirty years. It is hardy, the
points of the leaves being, however, injured by the
winter, and it flowers at the beginning of April, or
sooner in forward seasons. I have never known it
produce more than one flower on a scape, though I
cannot say that in a more congenial situation it
might not do so ; but on the other hand I have
never seen any vernal Hermione produce a solitary
flower. Its small stature, one-flowered habit or at
least propensity, slender reflex limb, and deeply
trifid cup, seem to distinguish it as a species from
v 2

324

AMARYLLIDACE.E.

Brevistyla, with which both it and Dubia agree as
to the style. It is not allied to Queltia orientalis,
notwithstanding the trifid cup of both plants.

6. Brevistyla. — Style shorter than the stamens, about
half the length of the tube, or a little more ; limb
white, pale lemon, or yellow ; cup yellow or
orange.

Tube equal to the limb.

Var. 1. PI. 41. f. 9. Limb pale yellow, cup deeper ; limb
much less than var. 5 ; tube and limb about
-§ths each. Flowers early.

Var. 2. Large yellow, pale yellow cup ; tube and limb
about fths each.

Var. 3. PI. 41. f. 12. Limb creamy white, cup yellow;
tube and limb about fths.

Var. 4. Crenulata. — Haworth. Bazelman minor of the
gardeners. Tube and limb each -|ths of an inch
long ; limb white ; cup wide, yellow, edged with
orange.

Tube shorter than the limb.

Var. 5. PI. 41. f. 7. Limb pale yellow, cup deeper,
limb large, near fths long, tube not fths.

Var. 6. Citrina major. — PI. 41. f. 8. Grand citronier.
Sweet Br. f. g. ser. 2. 128. Limb white, broad,
about an inch long ; tube thick fths ; cup pale
yellow ; flowers numerous.

Var. 7. Trewiana. — N. orientalis a. Bot. Mag. Bazel-
man major of the gardeners. Flowers large, limb
white, cup wide crenulated.

Var. 8. Flexiflora. — Haworth. Tube fths long ; limb
white, near an inch long ; cup 5-16ths wide or
more. This and the foregoing are coupled with
crenulata by Haworth as Grandiflorse, and he
fancied them a separate genus ; but in fact the
flower of Citrina major is larger.

Tube longer than the limb.

Var. 9. PI. 41. f. 6. Limb creamy white with yellow
marks at the base of the limb ; cup deep yellow ;
tube fths, limb half an inch, ovate, acute.

AMARYLLIDACE^E.

325

Var. 10. PI. 41. f. 10. Fine yellow, cup deeper; tube
fths, limb not fths.

Var. 11. PI. 41. f. 11. Limb yellowish white, about
half an inch ; cup orange ; tube fths.

Var. 12. PI. 41. f. 13. Limb yellow, about fths; tube
fths ; cup deeper yellow. Jonquil-scented.

Var. 13. PI. 41. f. 14. Yellow, cup deeper yellow ;
cup much larger than var. 12. Jonquil-scented.

Var. 14. PI. 41. f. 5. Limb white, fths; cup lemon-
coloured ; tube fths.

6. Italica. — Style nearly equalling the cup; cup yellow;
limb white or pale yellow.

Var. 1. Princeps. — Bot. Mag. 30. 1188. About 10-flow-
ered ; limb very pale sulphureous, four times as
long as the cup; cup irregularly cleft, pale yellow.

Var. 2. Prsecox. — Tenore, Flor. N. 1. 27. Pale sulphur,
cup citron, 6-cleft.

Var. 3. PI. 41. f. 24. Limb white, cup 3-cleft, yellow;
tube near fths long ; flower approaching to Papy-
racea in scent. A hardy, late-flowering variety.

Var. 4. Brevis. — PI. 41. f. 25. Limb white, cup wide,
yellow irregularly cleft ; tube half an inch long.

§. 2. Autumnales. — Flowering in the autumn.

Not having had an opportunity of inspecting the stamens
and fruit, I cannot be sure that this section may not form a
genus intermediate between Hermione and Narcissus. The
proportions in this section approach more to those of Narcissus.
Serotina has the capsule erect.

1. Elegans. — PI. 41. f. 27. See Hermione elegans. Sup-
plemental Observations. Tube near fths long; limb
acute fths or a little more. N. serotinus ; fig.
7-flora. Desf. Flor. Atl. 1. 283. t. 82. This African
plant is confounded by Desfontaines with the 1-2-
flowered serotina of the south of Europe. I have
given the outline of a flower copied from the umbel
represented by him, and also his outline of serotina,
PI. 41. f. 30. which last I have verified by compa-
rison with dry specimens to be perfectly correct.
It will be seen that the difference is very consider-
able in the form of the flower, that of elegans being

326

AMARYLLlDACEiE.

much longer and more acute ; and as serotina
occupies a very wide range in Spain, Sardinia,
Naples, and Africa, with a one- (rarely two-) flow-
ered scape, without varying its form of flower, 1
think the many-flowered plant of Desfontaines re-
quires to be distinguished. Both are remarkable
for the erect posture of the flowers. According to
Zerapha (Flor. Melit. thes.) this many-flowered
plant, with lanceolate segments, and a very short
entire crenulate cup, grows in the island of Malta,
where it is called Rangis mewahhar. N. oblite-
ratus Willd. Reliq. Schultes is probably this plant.

2. Serotina. — PI. 41. f. 39. and 36. Tube near half
an inch, limb broader than in elegans, about 7-16ths
long. The former represents the tube, cup, and
sepal, from G.Bentham, Esq.’s Sardinian specimens;
tube more or less enlarged in different specimens;
the latter Desfontaines’s outline of the African spe-
cimens, verified by comparison with the Sardinian.
Scape 1-2-flowered; tube 9-16‘ths, limb 7-16'ths,
white, cup 1 - 1 6 1 h , yellow; flow'er erect; flowering
usually before the appearance of the leaves. Found
in Spain, near Badajos, in Sardinia, Naples, and
North Africa. A miserable engraving of this plant
by Clusius was evidently copied and grossly exag-
gerated by Parkinson, who does not on this occa-
sion deserve the name of faithful, given to him by
Flaworth, having set leaves or bractes on the scape.
Miller described the stalk of serotina to be knotted;
Tenore has given a like description, and authors
have been deceived into a belief that the plant has
an articulate scape, quite repugnant to the cha-
racter of the suborder. Desfontaines takes no notice
of this supposed phenomenon, and represents the
scape like that of other Amaryllidean plants. On
examination of many specimens, it appears that
the knots exist in a very small proportion of them,
quite irregular in their number and position, not
being articulations but swellings of the scape, and I
apprehend they must have been occasioned by the
deposit of theeggand larva of some very small insect.
F. 30. represents three such knots on the lower part
of the scape of one of the Sardinian specimens.

AMARYLLIDACE.i£.

327

The coarse figure given by Clusius was intended to
represent a specimen with such knots ; but they
are very ill represented, and the excrescences mag-
nified by successive copyists from a had original,
have assumed a most extraordinary appearance in
the figure given by Parkinson, which has no resem-
blance to a Narcissean plant.

I have accented the name serotina (late-flowering) on the
o, in conformity with the authority of Hilary of Arles, who
has the i short. I should have considered that the analogy
of the words vespertin us and matutinus, also expressing time,
would have overborne his authority ; if Prudent! us,* who
has much weight with me, had not used another adjective of
time, diutinus, with the i short, and perhaps Plautus also.
It is desirable to have a clear understanding as to the accen-
tuation of the many botanical adjectives ending in inus.
Greek adjectives in bios have the i generally short, though
it is used both long and short in oporinos , therefore all ad-
jectives derived from a Greek word, as calathinus, calycinus,
crocinus, faginus, murinus (from muros, Gr.), must bear the
accent on the antepenult, or syllable preceding the i which
is short. The few adjectives in inus derived from the names
of vegetables, of which the accentuation has been ascertained
by their use in verse, as faginus, crocinus, are, to the best of

The deviations of Prudentius from the quantity of such Greek words as
eidola, whicli he wrote idola, did not arise from ignorance or false pronunciation,
but because the Greek word had the o long, but the accent nevertheless on the
first syllable, which was contrary to the law of Latin pronunciation ; and, as it
was necessary, in order to reconcile it, to depart either from the accent or the
quantity of the original word, he sacrificed the latter as the least deviation from
the original sound. There had been a gradually increasing preference for accen-
tual sounds, which (as I pointed out many years ago in an article on the harmony
of language in the Edinb. Rev. v. 6) at first limited the temporal metres of the
Romans, and at last superseded them, by the establishment of the accentual
cadences which regulate the verse of modern Europe. In the days of Virgil the
quantity of the syllables was deemed more important than the accent, and we
learn from a note by Pierius (Georgic. i. 59.) that when Virgil introduced the
Greek word E'piros, his meaning, in preserving the Greek termination os, was to
preserve the Greek accentuation on the first syllable, though the i in the second
was long ; whereas if he had written the word with Latin orthography Epirus,
the rule of Latin pronunciation would have thrown the accent on the i. Such a
departure from the Roman habits of speech could not be generally adopted, and
we find in the words introduced from the Greek at that period, the accent was
changed rather than the quantity ; in a later age the quantity was sacrificed to
preserve the accentuation.

328

AMAUYLLIDACEiE.

my recollection, Greek words, the original Latin practice
having been to form adjectives from the names of the animals
in in us, and of vegetables in eus. The vegetable adjectives
in inus were of late introduction, and built upon the Greek
model, even when derived from Latin names ; they seem to
have been always reputed short, and the contracted adjec-
tives, abiegnus, acernus, quernus, seem to imply the brevity
of abietinus, acerinus, and quercinus, which were not used.
Pliny seems to have attached to the termination inus a pecu-
liar meaning, as virga myrtea or laurea, a myrtle or laurel
twig ; oleum myrtinum, laurinum, myrrhinum, oil extracted
from myrtle, laurel, or myrrh ; folium laurinum, a leaf like
that of laurel. Adjectives in inus, from the names of animals,
are invariably long and accented on the i; as equinus,canmus,
cervinus, viperinus, aquilinus, murinus, from mus. I think it
may be laid down as a general law that all adjectives in inus,
of Greek extraction, or derived from the names of vegetables
(such being framed on the Greek model), have the i short
and unaccented ; all derived from the names of animals, and
generally all other adjectives in inus, the i long and accented,
unless there be special authority to the contrary ; and I call
to mind no other exceptions than serotinus and diutinus,
though it is very possible that others may escape my recol-
lection.

3. Obsoleta. — Haw. N. autumnalis medio obsoletus.

Park. Par. p. 90. t. 89. f. 4. Two flowers declined,
white; cup very small ; yellow, with dun-coloured
edge ; leaves two, narrow ; scape rising between
them. Received by Parkinson from Spain, and
not since noticed. N. Broussoneti (Lagasca),
colour unknown, is perhaps this plant. PI. 41. f. 28.
represents the tube, cup, and sepal, of two-flowered
specimens from Tangiers (herbar. Bentham. and
Lindl.), the flowers of which have the posture of
those of Viridiflora ; the cup minute, the tube a
little exceeding half an inch ; segments of the limb
obovate, equal to the limb ; but the colour of the
flowers cannot be ascertained. If they are not
viridiflora, they are probably obsoleta. The obovate
apiculate sepal seems to distinguish them from the
others.

4. Viridiflora. — 1- (3?) flora. Bot. Mag. 41. 1687.

AMARYLLIDACEiE.

329

Flower green, jonquil-scented, cup very small,
6-lobed ; scape produced before the leaves ; tube
according to the figure fths, limb nearly as long,
leaves rush-like ; bulb small. The specimen
there represented is one-flowered. (Specim. ex
Tangiers? PI. 41. f. 28. Herb. Benth. et Lindl.
2-flor.) This plant grows on the neutral ground
between Gibraltar and St. Roque, and on the coast
of Barbary. I believe the two-flowered specimens
from Tangiers to be viridiflora. Parkinson’s green-
flowering plant, of which the cup according to the
figure was not lobed, had a three-flowered scape.

Var. 2? Integra? — Park. Par. 94. 11. t. 93. f. 6. 3-flow-
ered, cup entire. The existence of this variety
with an entire margin to the cup, depends on the
accuracy of Parkinson’s figure ; and, as he does
not state the margin to be entire, I consider the
figure to deserve very little credit.

This species forms the genus Chloraster of Haworth ; his
generic character is a very incomplete description of the
species, and contains no semblance of a generic distinction.
I have lost the plant many years, and have had no oppor-
tunity of examining the most important features of this and
the other autumnal species.

Suborder 6. — Galanthe^e. Porandrous ; i. e. not having the
anthers slit and inverted, but opening partially ; in
this instance the pollen is discharged through two
small round holes at the summit. Not operculous ;
bedded, i. e. with a glandular spongy covering to
the germen, in which the filaments are inserted.
Pollen not half the size of the smallest pollen of
Amaryllideoe ; peduncle curved.

This suborder is confined to the portion of the old Con-
tinent north of the tropics.

§. 1. Scape solid ; seeds whitish.

80. Galanthus. — Bulb ovate ; leaves linear lorate ; scape
1 -flowered ; spathe tubular below ; above, slit on
one side, transparent on the other ; germen oblong-
ovate, pendulous ; segments separate ; sepals con-
cave, expanding in the sun ; petals much shorter,
obovate, emarginate, 2-lobed, having the semblance of
large nectareous scales ; style filiform, tapering to a

330

AM ARYLLIDACEA2.

point ; filaments short, equal, inserted in the opercle,
free ; anthers erect, affixed at the base, apiculate ;
orifices terminal, round ; capsule valveless ; seeds
whitish; pollen very minute.

1. Nivalis. — Engl. hot. 1. 19. Sepals white; petals

with 5 green lines down the inside of each lobe,
and an undulated green mark towards the top out-
side.

Var. horte?isis, flore semipleno. The doubleness of this
flower consists in a multiplication of the scale-like
petals, on some of which an anther more or less
perfect is usually borne ; which is very usual also
in Camellia Japonica. Dr. Lindley considers this
analagous to the production of supernumerary
anthers in Gethyllis, which occurs also in Vellosia.
It appears to be the necessary result of the imper-
fect conversion of the stamens into petals, the flower
being in fact only semi-double. The like occurs in
the semi-double Hippeastrum equestre. In Ge-
thyllis and Vellosia the stamens are multiplied
without any multiplication of the segments.

2. Plicatus. — Bot. Reg. 7. 545. Bot. Mag. 47. 2162.

Marsh v. Bieberstein. Fl.Cauc. Sup. 225. Leucojum
bulb, praec. Byz. Clus. Hist. 1. 169. Bulb larger ;
leaves larger, with their margins folded back ; scape
more robust ; flower smaller, green, more intense;
petals set more in than the sepals, and having more
the appearance of a scale. The sepals are a con-
tinuation of the outer coat of the germen, but
in plicatum the petals are set on its inner coat, and
their base does not range with the sepals. If
the petals of Nivalis had agreed with it in that
respect, I should have thought that the true petals
were entirely deficient. It is a native of Russia
and Asia.

81. Erinosma. — Bulb ovate; leaves linear lorate ; scape
1-2-flowered, pedunculated; germen triangular, obo-
vate, pendulous ; spathe tubular below, above slit
on one side, transparent on the other ; germen obo-
vate, segments of the perianth separate, similar ; fila-
ments short, equal, free, erect, inserted in the opercle ;
anthers erect, affixed at the base, not apiculate ;

AMARYLLIDACE^E.

331

orifices terminal, and style club-shaped (i. e. thick
at top, attenuated downwards) ; stigma triangular,
slender, tapering to a point ; capsule turbinate, seeds
yellowish white.

1. Vernum. — Leucojum vernum. Bot. Mag. 2. 46.

Spathe longer than the peduncle, perianth | long,
segments concave, oval, with a blunt point ; a
yellowish green spot on the outside of each, just
below the point ; style white below, green upwards ;
the tapering point white. Flower fragrant; in
February.

This is no Leucojum, for that genus has shining testaceous
black seeds like those of Hemerocallis, and a style attenuated
at both ends like the roller used by pastry-cooks. It differs
from Galanthus in having the petals like the sepals, and the
style club-shaped and thick, instead of slender and tapering
upwards ; it differs from Acis in the complete separation of
the segments of the perianth, which are united in Acis, in
the style attenuated downwards instead of upwards, the acute
stigma, and the round orifices of the anthers, which if we may
trust the usual accuracy of Parad. Loud, are more longitudi-
nal in Acis (see Par. Lond. fig. 21.) and the leaves not being
filiform. Erinosma should be planted in peat. In very wet
summers it may be advisable to take up the bulbs.

2. Carpathicum. — Leucojum vernum; var. /3. Bot. Mag.

45.1993. Spathe 2-flowered ; spots on the perianth
yellow.

I have not been able to obtain a sight of this plant. I see
no reason for confounding it with the one-flowered, green-
spotted vernum. I believe that vernum does not ever pro-
duce a two-flowered scape, and that the L. prsecox majus of
Clusius is Carpathicum.

82. Acis. — Perianth just united at the base ; segments nearly
similar ; filaments inserted in the disk ; free, short,
straight; anthers erect, affixed at the base ; dehiscent
on the outer side near the orifice ; style filiform, a little
attenuated upwards ; stigma a little divided ; seeds
fleshy, angular. (Bulb ovate ; leaves filiform ; scape
1 -few-flowered ; spathe 2-valved ; germen pendulous.)

1. Trichophylla. — Leucojum trichop hyllum. Bot. Reg.
7.544. One-flowered, perianth oval, white ; flowers
vernal.

332

AM ARYLLIDACEiE.

2. Grandiflora. — PI. 30. fig. 4. Specim. Herb. Linn.

Leucojum Redoute lil. 217. 2-4-flowered, perianth
oval, white.

3. Rosea. — Sweet Br. fl. g. 297. One-flowered, perianth

oval, pink ; leaves glaucescent.

4. Autumnalis. — Parad. Lond. 21. Leucojum aut. Bot.

Mag. 960. 2-3-flowered, perianth white, red at the
base; segments 3-cleft at the end ; leaves green.

The first is said to produce its blossom in the spring, the
others in autumn before the leaves push. They are all natives
of the countries bordering on the Mediterranean. I have
had no opportunity of inspecting these myself, never having
seen any but autumnalis in flower, and that many years ago.
I have it in the greenhouse, where it requires to be left quite
dry in summer. It grows abundantly near the river in the
neighbourhood of Badajos. Rosea and autumnalis might
have been united as varieties of one species under the name
hiemalis, but they will probably maintain themselves dis-
tinct.

<§. 2. Scape hollow ; seeds shelly, black, shining.

83. Leucojum.— Perianth six-cleft, segments separate, nearly
equal (narrowed near the point), style slender, cla-
vate ; stigma acuminate ; filaments inserted in the
disk ; anthers erect, dehiscent on the outer side, from
the terminal orifice, but not to the base ; capsule soft,
triangular, turbinate, 3-valved, semi-dehiscent (bulb
ovate ; leaves vaginating, erect, linear, lorate ; scape
few -many -flowered ; spathe 1-leaved, deeply slit,
withering; germen obovate.) Seeds nearly round,
outer coat loose, foliaceous, glossy black ; inner coat
brown, soft, separable ; albumen horny, with a large
circular chalaza not visible till the coats are removed ;
embryo curved, reaching the foramen.

1. iEstivum. — Bot. Mag. 30. 1210. Leaves green ; scape

flattened, 2-edged ; peduncles about an inch long
or less ; perianth white, with a faint green spot on
each segment near the end ; seeds as big as a pea.
The valves of the capsule gape near the top, but
adhere at the point, and do not open to the base.

2. Pulchellum. — Parad. Lond. 74. Leaves deep green,

appearing in Nov. ; scape 3-7-flowered, peduncles

AMARYLLIDACE2E.

333

long, green, not crenulated ; periantli white, with
a green spot near the top of each segment ; filaments
white ; style white, with a green spot below the
stigma ; ovules 13-20 in each cell. The figure of
the style and stigma of this plant is doubtless impro-
perly represented, for it disagrees with the descrip-
tion in the text, as well as with the conformation of
Leucojum aestivum. I have never seen the plant.

3. Hernandezianum. — Cambassedes. Leaves linear, 3-4-
lines wide, obtuse, 1-1^ foot long ; scape a little ex-
ceeding the leaves; spathe one-leaved, peduncle
1-1^ inch, filiform ; perianth about £ an inch (4-5-
lines) long ; style shorter than the limb. Differs
from aestivum in having flowers about half the size,
and a more oblong germen. Found by Dr. Her-
nandez, in Majorca, on the mountains near Lluch.

Leucojum capitulatum of R. and Schultes from Loureiro
C. Ch. p. 246. is unquestionably a Curculigo or Molineria.
They have been misled by the alleged thickness of the points
of the limb, without considering the plicate leaves and hairy
exterior of the perianth. Leucojum is remarkably distin-
guishable from Erinosma by the lateral slit of the anther,
which is superadded to the terminal pore, sometimes running-
near to the base, though the anther is not at all inverted, as
in the schistandrous genera ; the stigma is much like that of
Erinosma, but the style slenderer and less clavate.

Suborder 7 ? Tacce^:.

Concerning these anomalous plants, which form the order
Taccaceae, and in my opinion do not properly belong to Ama-
ryllidaceae, I have little to add to what has been already stated.
The order contains only the very few known species of Tacca,
one of which has been detached by Prezl, under the name
Ataccia. I raised many years ago Tacca pinnatifida and
integrifolia from seed, and their habit was precisely that of an
Arum, throwing out their fibres from the upper surface of
the tuber, which lay dormant in winter. I have never seen
their inflorescence, but notwithstanding its approximation to
that of hexapetaloid plants, I believe they will be found to
have been pi-operly placed (p. 45) as a separate (subspadiceous
subcorolliform) order. I do not recollect that any of the
plants I have included under Amaryllidaceae, produce fibres

334

.AMARYLLIDACEiE.

on the upper surface of a tuber ; I cannot bring to mind
whether the Dioscoreae do so or not, and I have not access to
any of that genus at present, but I observe from the engrav-
ing that Testudinaria does not ; and, if Dioscorea does, that
circumstance will furnish a stronger feature of generic dis-
tinction than any which has been yet noticed ; but we must
look far beyond this natural order to judge of its validity in
limiting the affinity of vegetables, and we should find it dis-
tinguishing Caladium from Arum, which shews that it cannot
be available as a high grade of separation.

ON CROSSES AND HYBRID INTERMIXTURES

IN VEGETABLES.

The first experiments, with a view to ascertain the possi-
bility of producing hybrid vegetables, appears to have been
made in Germany, by Kolreuter, who published reports of
his proceedings in the Acts of the Petersburgh Academy be-
tween 50 and 60 years ago. Lycium, digitalis, nicotiana,
datura,- and lobelia, were the chief plants with which he
worked successfully, and as I have found nothing in his re-
ports to the best of my recollection opposed to my own general
observations, it is unnecessary to state more concerning his
mules than the fact, that he was the father of such experiments.
They do not seem to have been at all followed up by others,
or to have attracted the attention of cultivators or botanists
as they ought to have done; and nothing else material on
the subject has fallen under my notice of earlier date than
Mr. Knight’s report of his crosses of fruit-trees, and my own
of ornamental flowers, in the Transactions of the Horticultural
Society of London. Those papers attracted the public notice,
and appear to have excited many persons both in this country
and abroad to similar experiments.

In the year 1819, having for some years previous paid
attention to the production of hybrid vegetables, but ignorant
of the experiments of Kolreuter, I was induced, rather
against my own inclination, to address some detailed obser-
vations on the subject to the Horticultural Society, which
were published in the transactions of that body. It was, I
say, against my inclination, because I was fully aware, that
a much longer course of experiments was necessary, in order
to obtain any results sufficiently certain to give stability to my
views. It is, however, satisfactory to find at the present day,
after the attention of botanists and cultivators has been fully
called to the subject during the space of many years, and a
multitude of experiments carried on by a variety of persons,
that, although our knowledge of its mysteries is still very

336

ON CROSSES AND

limited, my general views have been fully verified, and my
anticipations confirmed in a manner which I was scarcely san-
guine enough to have expected. Soon after the publication
of that communication to the Society, I was accosted by
more than one botanist in the words, “ I do not thank you
for your mules,” and other expressions of like import, under
an impression that the intermixture of species which had
been commenced, and was earnestly recommended to culti-
vators, would confuse the labours of botanists, and force them
to work their way through a wilderness of uncertainty ;
whereas it was evident to myself, that it would on the con-
trary afford a test whereby the accuracy of their distinctions
might be more satisfactorily investigated, many of the errors
of their system eradicated, and its details established upon a
more solid foundation, and less upon the judgment or caprice
of individuals. The alarm, which some botanists had taken
inconsiderately, appears to have subsided, and admissions
have been already made by some of the most distinguished,
which, if the consequences that flow from them are considered
without prejudice, must lead to much more extensive avowals,
and a final assent to the principle of my statements concerning
specific and generic distinctions. A number of attempts had
been made by the President of the Horticultural Society to
produce new varieties of fruit,* by impregnating the flowers
with the pollen of other individuals, and the success of his
proceedings was communicated to the public, both by his let-
ters to the Society, and by the more substantial production
of the fruits he had raised ; but it must be evident, that less
could be expected in the raising of new fruit-bearing plants
by intermixture, because the hybridising process is to a cer-
tain degree inimical to fertility in the offspring; and that
the flower-garden was more likely to be adorned, than the
kitchen-garden replenished, by the intermixture of species.
The President adopted in his writings a principle or dogma,
which seemed to be then much relied upon by botanists, that
the production of a fertile cross was proof direct that the two
parents were of the same species, and he assumed as a conse-
quence, that a sterile offspring was nearly conclusive evidence
that they were of different species ; and this dictnm was ad-
vanced without suggesting any alteration in the definition of

* There is a paper in the Philos. Transact, concerning the production of
apples, by crossing the pollen, by Benj. Cooke.

HYBRID INTERMIXTURES.

337

the term species, but leaving it to imply what it had before
universally signified in the language of botanists. Having,
in fact, the same fundamental opinion, that the production of
a fertile intermixture, designated the common origin of the
parents, I held also, what experience has since in a great
measure confirmed, that the production of any intermixture
amongst vegetables, whether fertile or not, gave reason to
suspect that the parents were descended from one common
stock, and shewed that they were referable to one genus ;
but that there was no substantial and natural difference be-
tween what botanists had called species, and what they had
termed varieties ; the distinction being merely in degree, and
not absolute ; so that, without first reforming the terms used
in botany, and ascertaining more precisely what was meant
by a species, those who argued on the subject were fighting
the air ; and I suggested, as my view, that the birth of an in-
termixture afforded presumptive evidence that the parents
were of one genus, meaning thereby kind or descent, and im-
plying such an affinity as to enable them to breed together,
and to induce a probability that they had diverged from one
original created type. The real point in discussion at that
period was, whether there did exist a positive and invariable
line of fertility or sterility in all mixed vegetable produc-
tions, founded upon an original identity or diversity in the
parental stocks ; and whether it was possible for two plants,
which were considered according to the general system of
botanists to be distinct species, to produce a fertile cross,
without proving an error of the subdivision in that particular
case. Further experiments have shewn, that the sterility or
fertility of the offspring does not depend upon original di-
versity of stock ; and that, if two species are to be united in
a scientific arrangement on account of a fertile issue, the
botanist must give up his specific distinctions generally, and
entrench himself within the genera. It has been objected
that if any plants, now different, had descended from one
original type, we might expect to find new forms and com-
binations daily arising round us by the process of nature, as
well as by artificial agency ; whereas the catalogue of Eu-
ropean vegetables does not appear to be increased by the
production of new plants in a wild* state ; but it is most

* Ranunculus, Anemone, Hypericum, Scleranthus, Drosera, Potentilla, Geum,
Medicago, Galium, Centaurea, Stachys, Rhinanthus, Digitalis, Verbascum, Gen-

Z

338

ON CROSSES AND

probable, that if the Almighty created the original types ca-
pable of permanent variations under different circumstances,
perhaps of soil or climate, those variations were worked at a
very early period, on the first diffusion of seeds into every
different portion of the world, especially by the operation of
the flood, and may have in part resulted from the changes of
climate which accompanied it and shortened the life of man.
We must recollect, that although the different races of dogs,
which all freely interbreed, are universally admitted to have
come from one type, though now outwardly more unlike to
each other than numberless distinct species of other animals,
we know not what the similitude of that type was ; we have
no record concerning the original wild dog, nor whether there
existed immediately before or after the deluge any dogs in an
undomesticated state ; nor have we any knowledge of the
time or place when any one of the several races, as grey-
hound, terrier, spaniel, bull-dog, &c. took its birth; nor is
there a single known instance of two parent dogs of the same
race, giving birth to individuals of a new race, or materially
dissimilar to themselves, except where they are mongrels, and
one of the ancestral types reappears more strongly than the
other. Neither have we any information concerning the ori-
gin of the different races of mankind, which are as different
in appearance as the species of vegetables ; we have not seen
any new race arise within the period of historical certainty ;
and whatever we do know concerning them, refers the time
of their branching out from the common stock to very re-
mote antiquity, at a period antecedent to or coeval with the
dispersion of mankind over the globe. If it had been other-
wise, the various races would have been blended, instead of
occupying different localities. It is probable that the various
races of dogs owe their origin to a very early period ; to the
days, when the effects arising from change of situation, were
first experienced by the several created members of the ani-
mal and vegetable kingdom : and it is no more essential to
believe that individuals of every one of the present species of
fox, or antelope, or finch (many of which are more like to
each other than the greyhound is to the terrier, though they
do not intermingle), entered with their present respective as-

tiana, Mentha, Quercus, Salix, and Narcissus, are however a long list of Genera
enumerated hy Schiede, 1825, and Lasch Linn. 1S29, as having pioduced spon-
taneous hybrids, to which Crinum may be added.

HYBRID I NTEB MIXTURES.

339

pects into the ark, than that all the calceolarias on the moun-
tains of Chili, or all the mezembryanthemums on the wastes
of Southern Africa, exhibited their present peculiarities in
the days of the patriarch. It was perhaps part of the wise
scheme of Providence, for the purpose of peopling the world
with the immense diversity of forms that occupy it, to give
each created race a disposition to branch into diversities,
acquiring constitutional peculiarities, which should keep
them more or less separated ; and the same phasnomenon is
observable in the languages of man, which are infinitely
numerous ; yet there is no reason to believe that many lan-
guages were given to man on the confusion of tongues ; on
the contrary, the cloven tongues that gave back the power
of universal speech, imply that they were few; but from
these have branched out innumerable languages, which can-
not be reunited, and no person can show when or how any
one of them arose, though wq may trace the mingling of one
with another in the later years of the world. One thing
seems pretty certain, amongst the mysteries in which this
subject is enveloped, that the differences worked, whether
in plants or animals, in a state of domesticity, do not effect
so great a constitutional separation inducing an indisposition
to reunite and produce a prolific offspring, as the changes
which have been wrought by nature in the wilderness.

I have said in the preliminary observations on Amarylli-
dacea?, that a perfect analogy between animals and vegetables
in their generations is not apparent; but I do not mean to
assert, that, if this subject can ever be thoroughly bottomed,
it may not be found to exist. A reformation of Zoology is
in progress ; for example, in Ornithology, the Linnajan ge-
nus Motacilla was after a time confined to the wagtails, a
large group being detached as Silvise ; but later observers
found that group to consist of sevei'al families, and have
since correctly distinguished at least the robins, the red-
starts, the nightingales, the hedge warblers, the fruit-eating
warblers, the sedge warblers, the chats, the troglodyte
wrens, and the greenish wrens, as separate genera with their
respective diversities ; and within those generic limits I
suspect, that the power of crossing ma)^ be confined, and their
several species, however now immutably distinct, may have
respectively branched out from one stock since the period
of the deluge. I have lately had under my observation a
dog, whose father was a fox in an innyard at Ripon, and it

z 2

340

ON CROSSES AND

has singularly the manners as well as the voice of a fox, but
it is the parent of many families of puppies: and I feel satis-
fied that the fox and the dog are of one origin, and suspect
the wolf and jackall to be of the same; nor could I ever con-
template the black line down the back of a dun pony with-
out entertaining a suspicion that the horse, unknown in a
wild state except where it has escaped from domesticity, may
be a magnificent improvement of the wild ass in the very
earliest age of the world : bearing in mind, that both in the
animal and vegetable creation, the diversities arising from
inscrutable causes in the wild races of the forest, are of a more
unalterable character than those which spring up under the
care and cultivation of man. With respect to animals in
their wild state, their union with their own species seems to
be mainly guided by voice and smell, and in domesticity
that instinctive preference is evidently much weaker, and
the will to keep themselves distinct is therefore lessened.
The various species of greenish wrens are so similar ,in plu-
mage, that it requires nice examination to distinguish them,
yet they have different notes, manners, and habits of build-
ing their nests, even when in the same locality ; but we have
no certainty that if their predilection for the voice and smell
of their own race was weakened, they would not be capable
of producing a fertile cross; and we draw our conclusions
from a few instances of domestic mules between species
which happen to be widely removed from each other, as the
pheasant and fowl, the goldfinch and canary bird ; wdiereas
we should apply to this subject, concerning which our know-
ledge is very limited, the consideration of the fact ascer-
tained concerning vegetables, which have no will to interfere
with our experiments, that some crosses are sterile and some
quite fertile, without any apparent reason, except the greater
or less approximation of constitution in the parents : and that
the cross-bred plant, which has seemed for a long course of
years to be absolutely sterile, becomes under some circum-
stances productive.

In accordance with the principle above stated, when it
was shown that the botanic subdivisions of Rhododendron,
Azalea, and Rhodora, comprehended plants which were
capable of intermixing, I asserted that the botanist must
reconsider and alter his subdivisions, and acknowledge that,
notwithstanding their peculiarities, they constituted but one
genus or kind. Conformably with this suggestion, Mr.

HYBRID INTERMIXTURES.

341

Sweet, in the second edition of his Hortus Britannicus, has
since wisely added to Rhododendron the genera Rhodora
and Azalea, with the exception of Azalea procumbens,
though in his subdivision of the pelargoniums he has not
kept in view sufficiently that certain and unalterable guide.
The true meaning of species, not as the word used to be ex-
plained by botanists, but as it is in fact used in all botanical
arrangements, appears to be, the subdivision of the genera
or kinds into branches, which naturally maintain themselves
distinct even when approximated, though they maybe more
or less capable of artificial or accidental intermixture; while
a local variety will reproduce itself when isolated under
particular circumstances of soil and climate ; and a semi-
nal variety will not with equal certainty reproduce itself
in the same form anvwhere, being more ready to inter-
mingle with others of like origin. In fact, there is no
real or natural line of difference between species and per-
manent or descendible variety, as the terms have been ap-
plied by all botanists ; nor do there exist any features on
which reliance can be placed to pronounce whether two
plants are distinguishable as species or varieties. Any per-
son, who attends to the subject, will perceive that no bota-
nist has laid down any precise rules by which that point of
inquiry can be solved, and that the most variable, contra-
dictory, and unsubstantial features have been taken by dif-
ferent persons, and by the same person on different occa-
sions, to uphold the distinctions they proposed to establish ;
the truth being that such distinctions are quite arbitrary,
and that, if two plants are found capable of interbreeding,
when approached by the hand of man, they are as much one
as if they were made to intermix more readily and frequently
by the mere agency of the wind, or assiduity of insects ; and
are not separable with more truth by any positive difference,
than the varieties which cannot be prevented from crossing
with each other when in the same vicinity. It remained to
be ascertained whether there did exist a real, natural, and
indefeasible difference between plants which could produce
a fertile and those which could produce only a sterile off-
spring by blending their races. It was my opinion that
fertility depended much upon circumstances of climate, soil,
and situation, and that there did not exist any decided line
of absolute sterility in hybrid vegetables, though from rea-

342

ON CROSSES AND

sons, which I did not pretend to be able to develope, but
undoubtedly depending upon certain affinities either of
structure or constitution, there was a greater disposition to
fertility in some than in others. Subsequent experiments
have confirmed this view to such a degree as to make it
almost certain that the fertility of the hybrid or mixed off-
spring depends more upon the constitutional than the closer
botanical affinities of the parents. The most striking and
unanswerable proof of this fact was afforded by the genus
Crinum, which is spread round the whole belt of the globe,
within the tropics and within a certain distance from them,
under a great variety of circumstances affecting the consti-
tution of individuals, which nevertheless readily intermix,
when brought together by human agency. The plant called
Crinum Capense (formerly Amaryllis longif'olia), impreg-
nated by either Crinum Zeylanicum or scabrum, both at
that time also called Amaryllis, produced offspring, which
during sixteen years proved sterile, probably because, not-
withstanding their botanical affinity, the first is an extra-
tropical aquatic plant, and the two latter tropical plants
which affect drier habitations and readily rot, at least in
this climate, in a wet situation. The same C. Capense, im-
pregnated by Crinum pedunculatum, canaliculatum, or
defixum, produces a fertile cross, though they are so dis-
similar as to have been placed in different genera, and the
author was formerly reproached by botanists as having com-
mitted an absurdity when he insisted upon uniting them.
The reason of the fertility of their joint produce seems to be,
that they are all aquatic or swamp plants ; and it may be
further observed that the crosses with the two former, the
plants being all extra-tropical, are much more fertile than
that between C. Capense and defixum, because the latter is
a tropical plant. The mules between Scabrum and Capense
having continued so many years with every appearance of ab-
solute sterility, without any change of situation or treatment,
at last produced one good seed in 1834 and another in 1835.
These facts were of such an overbearing nature, that it be-
came impossible for those, who had charged the author with
absurdity for uniting the parents under the genus Crinum
(to which even certain other plants were then asserted to be
more nearly allied than the species at that time called Ama-
ryllis), to contend any longer that they, producing a fertile

HYBRID INTERMIXTURES.

343

offspring, were of different genera, and they will probably
be never again disunited in any botanical work ; but the
facts furnish much ground for the serious consideration of
men of science. It happens (as if expressly designed to
overthrow the theory, that the identity of species is proved
by fertility or sterility in the mixed issue), that, while C.
Capense, Zeylanicum, and scabrum, are very similar in
their general appearance, and yield an offspring which has
been found quite sterile except in the case of the two seeds
above mentioned, C. Capense and pedunculatum are as
unlike as perhaps any two species of any known genus; and
if it were asserted that C. Capense and pedunculatum are one
species, and C. Capense and scabrum two species, the as-
sertion would appear, to any person looking at the plants,
too preposterous to require a serious answer.

In further confirmation of the fact that the sterility de-
pends on constitutional discrepancy, or difference of what
medical men call idiosyncrasy, may be adduced the curious
plant figured in the Botanical Magazine under the name of
Crinum submersum, which was found by my collector in a
pond or flooded spot not far from Rio Janeiro, in company
with a small variety of C. erubescens, and appeared to be
exactly intermediate between that aquatic plant and C. sca-
brum, which grows on high ground amongst the woods. It
is absolutely sterile, the anthers being always shrivelled and
the pollen dry, and it is not materially different from the
mules raised in our stoves between C. scabrum and a larger
variety of C. erubescens, the latter being of course a finer
mule, but with exactly the same barrenness of the anthers.
C. submersum is certainly a natural cross, in consequence of
the pollen of C. scabrum having been brought to the lake
by some humming-bird or insect which touched the stigma
of the aquatic species. The same sterility has been found
in C. amabile and C. angustum, which are undoubtedly
mules accidentally produced between dry-land and swamp-
species, the former probably between 0. Zeylanicum and
procerum, the latter between C. Zeylanicum and bracteatum ;
as also C. longiflorum (Amaryllis longiflora of the Botanical
Register), wdiicli is an accidental cross between C. Capense
and erubescens, one variety of it having been produced at
Demerara, the other in Jamaica. The fact being established
with respect to one genus, that the species which have most
botanical affinity and general likeness, if they delight in a

344

ON CROSSES AND

different state of soil or of atmosphere, produce a barren
cross, while the most dissimilar, if they possess the same
constitutional predilections, give birth to a fertile plant, can-
not remain as an isolated circumstance, but must be consi-
dered by every unprejudiced and philosophical mind with
reference to the whole vegetable creation. I have lately
heard it admitted in conversation by an eminent botanist,
that he had almost arrived at the conviction that there was
but one rose, meaning that there seemed to be no natural
impediment to the fertile intercourse of the great variety of
plants which constitute the known species of that extensive
genus. Let it be observed, if the fact is so, the reason is
apparent enough ; that, although some roses will endure a
little more cold than others, there is a sameness of constitu-
tion throughout the genus, which affects a dry soil and a
temperate atmosphere. The genus Calceolaria embraces
plants very dissimilar to the eye of the botanist, as well as of
the unlearned observer, of which some are absolutely stem-
less, and bear only leaves and flower- stalks, while others are
shrubby, and acquire a strong woody stem some feet in
height ; yet there appears to be no limit whatsoever to their
intermixture, and their produce may be crossed again inde-
finitely. Are we, then, to come to the result that there is
but one Calceolaria, oversetting not only the nicer distinc-
tions of botanical science, but the difference between herb and
shrub? The African Gladioli, excepting those which, like
the European, present their flowers in front of the stalk, have
been intermixed by me without any difficulty occurring, and
the crosses of the most dissimilar have proved abundantly
fertile, and four or five sorts have been blended in successive
generations. Some of the complicated crosses have pro-
duced seed less freely, and one treble cross (Hirsuto-Cardi-
nali-blandus) has as yet produced none that has vegetated,
probably because the last male, G. hirsutus, is of a constitu-
tion much less suited to our climate than the other two. Are
we then to come to the result, that these dissimilar species
are all one natural Gladiolus ? There is no outward sign of
barrenness in G. hirsuto-Cardinali-blandus, which will pro-
bably bear seed under favourable circumstances ; that there
is no insurmountable natural impediment may be proved
thus ; the offspring of G. versicolor by hirsutus, of blandus
by versicolor, and of Cardinali-blandus by tristis, have all
borne seed, shewing that G. hirsutus is not of a separate

HYBRID INTERMIXTURES.

345

race, and that the triple cross is not an impediment. I have
crosses raised by me between the yellow Linaria genistifolia
and the purple purpurea, and also between Penstemon an-
gustifolium and pulchellum, both perfectly fertile and sowing
themselves about the garden, and, from my having given
them many years ago to more than one nurseryman, become
common. It is scarcely possible to assert that these very
unlike plants are respectively one, and at the same time to
distinguish them from the rest of their own genera, espe-
cially the former. That whole portion of Amaryllideae which
constitutes the genus Hippeastrum, and was confounded by
botanists with a portion of the genus Crinum, not only inter-
breed freely, but produce offspring invariably fertile, be-
cause they are all of like constitution, and impatient of exces-
sive moisture, though some will bear more cold than others.
Amongst the Pelargoniums a similar convertibility has been
found to exist within certain limits, which, if duly observed,
will be sure guides to ascertain the genera, into which they
ought to be subdivided, and by which the botanist, who is
desirous that his labours should not be overturned hereafter,
must be in a great measure ruled in classing them.
Amongst the Cacti or Cerei the prickly angular specio-
sissimus, the flexible flagelliformis or whip-plant, and the
flat unarmed phyllanthocides, are nearly the most dis-
similar, yet they have produced mixed offspring, which
readily bears eatable fruit of intermediate appearance,
colour, and flavour. The fruit of the speciosissimus is
large, green, and well -flavoured, round oblong; that of
phyllanthocides small, purple, and very inferior ; the mule
from the former has purple fruit of a medium size and
taste*. The cross from the former by flagelliformis is now
ripening here a short angular fruit, quite unlike that of the
mother plant. The fertility of these crosses, and readiness
to vary the appearance and taste of the fruit, though derived
from such very dissimilar parents, is one of the most striking-
results of our experiments. I have had no opportunity of
attempting to cross them with the plants called echino-cacti,
but I do not see a single point in the generic character given
of those plants which can uphold it, and I believe them to
be of one genus with Cereus, and capable of intermixing ;
but I have had no opportunity of examining the flower of
any of the plants called Echinocactus myself. Amongst
melons I have had the Cucumis osmocarpus from Mexico,

346

ON CROSSES AND

bearing a small egg-shaped white fruit and a small flower
and leaf, very different from the Cucumis melo, fertilized
accidentally by its pollen, thus occasionally producing fruit
of twice the natural size with red flesh. Lobelia speciosa is
a cross between L. siphylitica and fulgens, yet it reproduces
itself abundantly.

The more these facts are considered, and the more they
are multiplied, as they will be by the daily experiments of
cultivators in other genera, the more strongly will my ori-
ginal suggestions impress themselves upon every botanist,
who will look on the subject without prejudice, that the
genera of plants are the real natural divisions; that no plants
which interbreed can belong to separate genera ; that any
arrangement, which shall have parted such plants, must be
revised ; that any discrimination between species and per-
manent varieties of plants is artificial, capricious, and insig-
nificant ; that the question which is perpetually agitated,
whether such a wild plant is a new species or a variety of a
known species, is waste of intellect on a point which is capa-
ble of no precise definition, and that the only thing to be
decided by the botanist in such cases is whether the plant is
other than an accidental seedling, and whether there are
features of sufficient dissimilarity to warrant a belief that
they will be reproduced, and to make the plant deserve on
that account to be distinguished by name amongst its fel-
lows. The effect, therefore, of the system of crossing, as
pursued by the cultivator, instead of confusing the labours
of the botanist, will be to force him to study the truth, and
take care that his arrangement and subdivisions are conform-
able to the secret laws of nature ; and will onlv confound him
when his views shall appear to have been superficial and in-
accurate ; while on the other hand it will furnish him an
irrefragable confirmation when they are based upon reality.
To the cultivators of ornamental plants the facility of raising
hybrid varieties affords an endless source of interest and
amusement. He sees in the several species of each genus
that he possesses the materials with which he must work, and
he considers in what manner he can blend them to the best
advantage, looking to the several gifts in which each excels,
whether of hardiness to endure our seasons, of brilliancy in
its colours, of delicacy in its markings, of fragrance, or sta-
ture, or profusion of blossom, and he may anticipate with
tolerable accuracy the probable aspect of the intermediate

HYBRID INTERMIXTURES.

347

plant which he is permitted to create ; for that term may be
figuratively applied to the introduction into the world of a
natural form which has probably never before existed in it.
In constitution the mixed offspring appears to partake of the
habits of both parents ; that is to say, it will be less hardy
than the one of its parents which bears the greatest exposure,
and not so delicate as the other ; but if one of the parents is
quite hardy and the other not quite able to support our win-
ters, the probability is that the offspring will support them,
though it may suffer from a very unusual depression of the
thermometer or excess of moisture, which would not destroy
its hardier parent. Such is the case with the beautiful mule
Rhododendron Altaclarae, of which the mother was a cross
between Ponticum and Catawbiense, and the father the
Nepal scarlet arboreum. We now possess a further cross by
the impregnation of Altaclarae by arboreum, which will pro-
bably come so near the father in its colour, that if, as ex-
pected, it should be able to endure our winters, we shall have
nearly attained the result, which would be otherwise most
likely impracticable, of acclimating the magnificent Nepal
plant ; for it does not appear that in reality any plant be-
comes acclimated under our observation, except by crossing
with a hardier variety, or by the accidental alteration of con-
stitution in some particular seedling ; nor that any period of
time does in fact work an alteration in the constitution of an
individual plant, so as to make it endure a climate which it
was originally unable to bear ; and, although we are told
that laurels were at first kept in hothouses in this country,
it was not that they were less capable of supporting our sea-
sons than at present, but that the cultivators had not made
full trial of their powers of endurance. The notion of Mr.
Sweet that the roots produced by cuttings are hardier than
those of seedling plants is probably fanciful, if he meant
permanently so, which alone would be of importance. They
may be tougher at the first period of propagation, while the
seedling is in its infancy, but that, if not permanent, could
have no effect in acclimating a plant. In truth it is not the
root that is tougher, but the nucleus or base of the cutting
from which the roots issue, and in which the life resides,
which is tougher than in a young seedling at the first. All
his other experiments only tended to shew that some half-
hardy plants would live through an English winter in very
dry and sheltered situations, or during two or three years,

348

ON CROSSES AND

till a more inclement season cut them off, but not that by
any process of his they had become hardier ; the word accli-
mating seems, therefore, to have been misapplied in his
paper in the Transactions of the Horticultural Society. For
the purpose of obtaining a large or a brilliant corolla, it will
be probably found in the long run best to use the pollen of
the species which excels in those points, because the corolla,
in truth, belongs to the male portion of the flower, the an-
thers being usually either borne upon it, or in some manner
connected with it by a membrane ; but upon the whole an
intermediate appearance may be generally expected, but
with a great disposition to sport, especially in the seminal
produce of the fertile crosses, as in plants which are apt to
break into cultivated varieties.

Before I proceed to consider the various cross-bred pro-
ductions of late introduction, which at present embellish our
collections, I will enter into a short detail of the reports
which I have seen of experiments on this subject, made on
the Continent, together with my view of the opinions which
have been advanced in them, and in a little work by Professor
Rennie, the matter of which is chiefly extracted from the
writings of Mons. De Candolle. Kolreuter’s experiments*
are detailed in the transactions of the Petersburg Acad, in
1777, and the five or six following years. I do not find any
further reports of experiments made in Germany, previous to
that of Gaertner, concerning the observations he had made
in 1825, subsequent to the publication of those of the Presi-
dent of the Horticultural Society of London, and of myself, in
its Transactions. He gave an account of the number of im-
pregnations he attempted to effect, the particular subjects of
his several experiments, and the failure or success of each.
I cannot learn that he has since published any report of the
germination of the seeds which he had obtained by those
experiments, and no later statement was known to Mons. l)e

* Relating to Lychnis and cucubalus, N. C. ac. Petr. t. 20. p. 431 — 448.
Hybrid digitales Act. Ac. Petr. 1777. Do. Journ. de physique t. 21. p. 285 —
299. Other hybrid digitales, Act. Ac. Petr. 1778. Continuation of experi-
ments on hybrid foxgloves, J. de phys. t. 21. 209 — 306. Hybrid lobelia, Act
Ac. Petr. 1777. — J. de phys. t. 23. 100 — 105. Hybrid lycia, Act. Ac. Petr.
1778. Verbasca, 1781. Daturse, 1781. Malvaceous plants, 1782. Flaxes,
Nov. Act. A. P. t. 1. 339 — 346. Pinks, ib. t. 3. 177 — 284. There may
perhaps be some other reports by Kolreuter, of which I may have neglected to
make a memorandum. I believe one concerning Nicotiana.

HYBRID INTERMIXTURES.

349

Candolle, in 1832. I have no hesitation in saying that this
report, which seems to have been accepted as proof of what
Gaertner had done, is utterly fallacious. He has entirely
overlooked the difficulty, and, in many cases, the impracti-
cability with the utmost care of excluding the natural pollen ;
the insufficiency of a bag to shut it out, and the probability
of its having been admitted even before the bag was placed
over the flower. I have learned by endless disappointments
to know, that no attempt to obtain a cross-bred plant can be
looked upon as successful, till the seedlings raised shall have
advanced in growth sufficiently to exhibit the type of both
parents united in themselves ; and I consider Gaertner’s
report of the cross-bred seeds he has obtained , to be nothing
but a mere enumeration of the crosses he has tried to obtain ;
and I believe some of his supposed intermixtures to be im-
possible. The fact is, that in this country, where the passion
for horticulture is great, and the attempts to produce hybrid
intermixtures have been very extensive during the last
fifteen years, not one truly bigeneric mule has been seen ;
and, although I by no means presume to assert that such a
production is impossible, experience shews it to be impro-
bable ; and those, who fancy they have obtained one, must
forgive my wishing to see it forthcoming, and to examine
whether it is certainly of such descent as they suppose.
Gaertner details his mode of proceeding, which is pretty
similar to my own ; but he does not seem aware, that, in
spite of all possible precautions, the pollen will often escape
unobserved, and will penetrate the coverings that may be
used. He asserts that the moist juice of the pollen combines
with that of the stigma, to fecundate the germen, a question-
able point, that need not here be considered. The super-
abundant viscous juice on the stigma of Rhododendron
appears to me to obstruct the fecundation, which I think
takes place more readily when it subsides. Gaertner could
not decide whether the fecundation is slow as Kolreuter
imagined, or rapid as Hedwig asserted ; but in microsco
pical observations the particles of pollen seemed not to be
emptied in less than an hour and a half ; and he found that,
when the fecundation was as he thought complete, the par-
ticles afterwards superadded did not change form or colour ;
but that in hybridizing applications a greater quantity of
pollen seemed requisite, in proportion to the distance of
affinity, and that it was repeatedly consumed ; and he

350

ON CROSSES AND

fancied that its successive applications in such cases made
the seeds more numerous and perfect, which is very probable;
but he says that only in kinds very closely allied did he
obtain the full complement of seeds, as for instance in the
genus Datura, of which Metel and Lsevis mix freety. He
found the life of the stigma more prolonged, when it was
not fertilized by its own pollen ; which might be expected,
because the complete saturation of the stigma had not taken
place, after which it no longer receives the influence of the
pollen. He states that in natural fecundations the change
of the stigma took place sometimes in 85 or 100 minutes,
usually in a few hours, at most in 24 ; but neither he nor any
other person seems to have thought of ascertaining whether
the influence of the pollen really fertilizes the germen within
that period, or merely saturates the stigma; and, although I
have not pursued a course of experiments to make that fact
sure, I have some reason to believe that the truth is not yet
ascertained. If the fertilization was complete, and the office
of the stigma defunct, it might be cut off without any detri-
ment; unless necessary to the mere nourishment of the ovules,
whether fertilized or not, which does not seem probable ; but
I have repeatedly cut it ofl' a few days after I had applied
pollen to Rhododendron, and the result has been that no seed
has been formed. The whole of my observations has led me
to think, that at any period before the decay of the stigma
the access of the natural pollen may supersede the influence
of the foreign that may have been previously applied, if not
from a closely allied species or variety ; but that on the other
hand no foreign pollen can act upon the germen after the
stigma has been fertilized naturally. The incomplete satu-
ration of the stio-ma in the first case enables the natural

O

pollen to gain access ; but, if the absorption of the pollen
first applied causes immediate fecundation, it cannot be ex-
plained how the subsequent access of the natural dust should
supersede it ; and it has seemed to me that the natural pollen
could supersede that of an Azalea on the stigma of an ever-
green Rhododendron even after the flower had fallen off.
This point, which I have not sufficiently investigated, might
be elucidated by cutting off the stigma with portions of the
style of various lengths at different periods after the applica-
tion of pollen, and seeing in what manner the operation in-
terferes with the fructification of the plant. Gaertner thinks
it doubtful whether the corolla is essential to the fertilization

HYBRID INTERMIXTURES.

351

of the stigma ; my observation is, that its early destruction
is very prejudicial to the growth of the germen and stigma,
but that after their developement it is not usually essential.
He observes that the corolla perishes more quickly and com-
pletely after fecundation, and is more persistent and some-
times withers instead of falling off, in cases of hybrid im-
pregnation, as it does where impregnation is prevented ; but
he seems not to have distinguished the cases of successful
and abortive hybrid impregnation. I observe that he admits
that the viscous juice remains on the stigma of Datura and
Nicotiana two or three days after fecundation, which does
not exactly agree with his theory, and seems to mark that
the stigma has not become quite inoperative. Some days
elapse before any other signs of fecundation appear after the
fall of the corolla, such as the enlargement of the peduncle, or
strengthening of its articulation, and that period seems to
him longer in hybrid impregnation, and the interval longer
before the seeds are vivified. Both he and Kolreuter observe
instances, such as I have found, of false hybrid fecundation,
producing an enlargement of the germen, or even seeds with
an imperfect embryo or without any. They remarked, that
they did not usually obtain the full complement of seeds
from a hybrid impregnation, unless the affinity was very
close. My own observation is that this circumstance de-
pends rather on the similarity of constitution, and is by no
means universal, for I had a pod from Crinum Capense
fertilized by revolution, in which every ovule produced a
seedling plant, which I never saw to occur in a case of its
natural fecundation. He cites from Kolreuter that Datura
metel and laevis have each about 600 seeds in a capsule ; he
found that a capsule from one of them fertilized by the other
contained 640, and in another case 284 ; but that Datura
laevis by Nicotiana rustica produced only 108 seeds, which
were however apparently perfect and provided with an em-
bryo ; but I utterly repudiate the probability of that impreg-
nation, of which he has not published the ultimate result.
In Gaertncr’s list, I find Convolvulus sepium by Ipomoea
purpurea (the Convolvulus major of nurserymen) 8 experi-
ments failed ; the converse 10 failed. Iporn. purpurea by
Convolvulus tricolor 6 failed ; 1 successful, which I greatly
doubt. Datura laevis by Metel, 4 failed and 4 succeeded ;
by Hyoscyamus all failed ; by Nicotiana macrophylla 3 failed ;
1 succeeded ; by Nicot. rustica 1 failed, 1 succeeded. Datura

352

ON CROSSES AND

metel by laevis all succeeded ; by Hyoscyamus failed ; by
Nicot. macoph. failed. Glaucium by Papaver failed. I
make no doubt that when the seeds vegetated, the supposed
crosses of improbable origin manifested themselves to be na-
tural seedlings of the mother plant, or produced by the in-
trusion of some kindred pollen. Kolreuter raised mules
(Act. Ac. Pet. 1780) between Lobelia siphylitica and Cardi-
nalis both ways. He found them fertile by the pollen of
either parent, and their pollen fertilized the parents, but he
obtained no seed from the mule by its own pollen. Lobelia
speciosa, or more properly Lowii, Bot. Reg. 17. 1455, was
found in a border where siphylitica and fulgensgrew; it was
a mule from siphylitica, which seeds freely. That mule,
intermediate and purple-flowered like those of Kolreuter,
seeded abundantly with me standing in a border between the
two parents, but the seedlings with one or two exceptions, did
not approximate to either, but reproduced the mule with some
variability of colour. Dr. Wiegman, in a tract published in
the German language, has given an account of some interest-
ing experiments. By sowing Allium porrum and Cepa in
one bed, and tying the flower-stems together, he obtained
plants intermediate between the leek and onion, which were
fertile. By tying together Vicia faba hortensis (the garden
bean) and Vicia sativa (the common vetch), he obtained cross-
bred seed; the seedlings from the bean had flowers more
purple, smaller pods and .seeds, which when sown again,
yielded plants that appeared to him not distinguishable from
what he calls the known red-seeded variety. Those from the
vetch shewed also a difference of blossom. In 1823 he sowed
Pisum sativum (the field pea) and Vicia sativa (the common
vetch) together ; the seedlings showed a departure from the
natural colour, and yielded grey seeds. From the twining
Phaseolus vulgaris albus, and Phaseolus nanus which does
not twine, he obtained crosses ; some seedlings of the latter
twining, and of the former bent and crooked, but not twining.
From Vicia sativa (the common vetch) and ervum lens (the
lentil) he also obtained a fertile cross. If these facts are
correct, it is clear that the closely allied genera Faba, pisum,
vicia, and ervum cannot be upheld as distinct ; but, although
it is a very common practice in England to sow peas and
tares mixed with beans, I have questioned many intelligent
farmers on the subject, and not one had ever heard of any
adulteration in the seed in consequence of the mixed culti-

HYBRID INTERMIXTURES.

353

vation, which, according to Dr. Wiegman’s statement, ought
to be of constant occurrence in such cases. On the other
hand, I have seen cultivated in Yorkshire a plant having the
growth of a vigorous field pea(Pisum), which produces seeds
that no man would hesitate to call beans, and which when
boiled have, I understand, more the flavour of beans than
of peas ; and the plant, though very fertile, has every ap-
pearance of being a mixed production between the two. The
most extraordinary mule, however, that is asserted to have
been produced on the Continent, is a cross between the cab-
bage and horse-radish, which Monsieur Sageret reports that
he has obtained, and that it has produced seed-pods, some of
which resemble the short pod orsilicula of the Cochlearia or
horse-radish, and some the long pod or siliqua of the Brassica
or cabbage. Strange it is, that asserting such a result, he
appears quite unaware of its importance, and does not state
whether those singular and various pods contained seeds or
proved abortive. He does not even state whether the plants
so obtained were annual, like the cabbage, or perennial, like
the horse-radish, nor does he describe them. I must there-
fore, without any offence to him, be allowed to consider the
actual generation of such a mule questionable, till the mule
plants are produced before the public, so that their conforma-
tion may be examined to see whether it is agreeable to their
supposed origin, or at least their peculiarities accurately de-
tailed, and the impossibility of a mistake rendered manifest. I
have, however, always considered the separation of siliquosae
and siliculosae to be very unsatisfactory, and have entertained
great doubts of the established distinctions amongst Cruciferae.
In consequence of M. Sageret’s statement, I tried in 1835 to
impregnate a plant of Brassica with the horse-radish, and
with the pollen of two or three other genera of Cruciferse ;
but I did not obtain a single seed from at least fifty flowers,
on which the experiments were tried, all other flowers being
cut off from the plant. I beg to be understood as not deny-
ing M. Sageret’s assertion, but requiring better proof of the
accuracy of a fact so important to science, in which he may
be mistaken, and more detailed particulars, and especially
the production of the plants ; and I invite M. Sageret to com-
municate one of them to the Horticultural Society of London,
that opportunities may be afforded of examining it carefully.

Experiments have also been made on the Continent
to establish within what limits the cucurbitaceous plants

2 A

354

ON CROSSES AND

(melons, See.), can be intermingled; but, the names used not
being of general currency, I cannot state accurately the re-
sult. Dr. Wiegman extended bis experiments to varieties of
oats. The accuracy of bis observations and those of Sageret
ought to be thoroughly investigated, and the results produced
before the public ; and a more useful office, connected with
its pursuits, could not be undertaken by the Horticultural So-
ciety of London, than to pursue those enquiries and extend
them to other vegetables. An observation made by Gaertner
and Wiegman (Berlin, 1828) as well as by Mr. Knight, that
the offspring of hybrids revert to the maternal and not to the
paternal type, is certainly erroneous, and Wiegman admits
that tobacco (Nicotiana) and oats may be made by crossing
again either to revert to that of the mother, or advance to that
of the father. The offspring of the mule Passiflora ccerulea-
racemosa, both in Mr. Milne’s garden and in mine, have noto-
riously approximated to the type of the father, and lost alto-
gether the red colour of the original mother. It is certainly
not correct as a general law, though some have stated it, that
the number of seeds in one pericarp is smaller in hybrid,
than in cases of natural, impregnation ; it is true in some
cases, and the reverse occurs in others. With respect to the
conditions stated by Professor Rennie, as necessary to ensure
success in crossing vegetables, it must be observed that the
first, namely, that the blossoms should be nearly in the same
state of advancement, is not accurate; for in some kinds, as
for instance, Calceolaria, that which is to bear the seed should
be much less advanced than that from which the dust is taken ;
and in others, as Pelargonium and Alstrcemeria, it should be
much more advanced. In truth, the moment should be
seized, when the stigma in the flower which is to bear the
seed, and the pollen in the other, is in perfection. The se-
cond condition stated, that the anthers should be cut out early
in the morning, is equally liable to objection, and cannot be
applicable to all flowers, some of which blow in the morning,
and others in the afternoon or evening. The necessary con-
dition is, that the anthers be removed from the flower that is
to produce the seed before the dust can escape from them ;
for which purpose in many cases, as for instance in Crocus,
Erica tetralix, and others, the flower must be opened with great
difficulty at a very early stage. The plant must be then
placed in a situation where no natural dust can reach it,
brought either by the wind or by insects ; and the pollen

HYBRID INTERMIXTURES.

355

from another flower, which is in perfection and not beginning
to wither, must be applied to the stigma as soon as it is quite
developed and mature, or rather sooner. The success of
such experiments is always most probable when the plant,
which is to be fertilized, has been forced, and no natural
pollen can be brought to it accidentally from other plants,
and, by forcing one of the intended parents, those, which
flower at different seasons, may be made to intermix. But
it should be always remembered that, except in cases where
the anthers are very accessible, and not mature till after the
expansion of the flower, it is almost impossible to be quite
certain that no particle shall escape from them in the opera-
tion. It is incorrectly stated that we cannot cross plants
which do not ripen seeds with us, for their dust may be used
to fertilize one that will ripen its seed ; for instance, the
pollen of Zephyranthes carinata, which I have never known
to bear seed in England, lias fertilized Z. tubispatha. It is
also very possible, if the fruit of one species is apt to perish
immaturely from the unsuitableness of the climate, and the
germen of another is not usually fertilized with us, in conse-
quence of an imperfect formation of its pollen, that it may
be effectually fecundated by the pollen of the other species,
though neither would have borne seed separately. The
deficiency of pollen is of frequent occurrence in the American
Azaleas from the fault of our climate ; but the p lien of
Sprekelia and of Z. carinata is abundant, and their sterility
does not arise from its defect, but from the temperature or
exposure in which they are placed not being exactly adapted
to the growth of their fruit.

The first hybrid amongst our liliaceous plants that ap-
peared in our gardens was the mule between Hippeastrum
vittatum and regium, which was circulated under the name
of Amaryllis Johnsoni, having been raised by a nurseryman
named Johnson. It was, perhaps, an accidental production,
for it was offered to the public with an incorrect statement,
that it had been raised by impregnating H. vittatum with
the pollen of Sprekelia formosissima. He might, however,
have made various trials, and have been deceived as to which
of them had been successful. That statement has been since
disproved by the failure of every attempt to fecundate any
species of Hippeastrum by the pollen of Sprekelia, of which
the separate generic character is thereby confirmed, and also
by the facility with which plants exactly similar have been

2 a 2

356

ON CROSSES AND

raised between H. vittatum and regium. The next hybrid
of that order, that flowered amongst us, was the Crinum
Goweni, which was raised from seed of C. Capense, impreg-
nated with the pollen of C. Zeylanicum in the greenhouse of
the Earl of Carnarvon, at Highclere, in 1813, by R. J.
Gowen, Esq., and blossomed in my possession at Spofforth ;
and soon after the mules between C. Capense and Canalicu-
latum, which had been first raised by me at Mitcham about
the same time, came into flower with other crosses at Spof-
forth. All the hybrid Crinums raised between Capense and
tropical species, which are now very numerous, are hardy
enough to stand out of doors against the front wall of a
stove, where, if a mat is thrown over them in sharp frosts,
they preserve much of their leaves through the winter, and
from May to November continue throwing up a succession
of flower-stems in great perfection. C. scabro-Capense bears
the most beautiful flower ; C. pedunculato-Capense is of the
largest stature. The only other hybrids of much note in
our gardens at that period were, to the best of my recollec-
tion, as follows : — The Rhododendron Azaleoides, obtained
by the accidental impregnation of an Azalea by Rhododen-
dron Ponticum, in the nursery of Mr. Thompson at Mile-
End ; the Rhododendron glaucum hybridum figured in the
Botanical Register, and Azalea enneandra figured in the
Botanical Magazine, which had both been raised by me at
Mitcham and removed to Spofforth. Since that time we
have had the Rhododendron fragrans of Mr. Chandler, and a
very great number of similar crosses from American white
Azaleas by Rhododendron Ponticum at Highclere. I am
not aware at what period the beautiful mule pink which is
common in our gardens made its first appearance, nor
through whom, or in what manner it was obtained; but it
was probably the produce of an accidental intermixture of a
florist’s pink with a crimson sweet-william. Mr. Sweet
gives no date to the hybrid pinks. Several most beautiful
mule Gladioli and Ericae, which had been raised at Mitcham
between the years 1808 and 1814, and removed from thence
to Spofforth, had also flowered there, but had not been made
known to the public till the year 1819, when an engraving
of Crinum Goweni was published in the Horticultural Trans-
actions, and a figure of two or three crosses of Gladioli
appeared soon after in the same work. Those who raised
pelargoniums from seed had found amongst the produce of

HYBRID INTERMIXTURES.

357

certain species a great disposition to intermix and sport,
which was occasioned by the accidental transmission of the
pollen from one plant to another by the bees, which occurs
perpetually in that genus, because many of its flowers are
occasionally without anthers, or lose them before the stigma
comes to maturity, which causes them to be fertilized by
another flower; and in the year 1812 (taking the date from
Sweet’s Hortus Britannicus) the beautiful cross between Pe-
largonium Citronodorum and fulgidum was obtained from
seed, and afterwards produced under the name ignescens ;
and, being fertile, it has become the parent of an innumer-
able variety of the most beautiful plants that adorn our green-
houses. P. ardens had been raised two years before between
fulgidum and lobatum, and had first pointed out to cultiva-
tors that it was possible, through the pollen of P. fulgidum,
to introduce its brilliant tint of scarlet under a variety of
modifications, in union with the superior qualities of other
species in which it was deficient ; but a long course of expe-
riments has shewn the impracticability of blending the plants
allied to zonale (which are properly detached by Mr. Sweet
under the name Ciconia) with the true Pelargoniums, which
are however certainly of one genus with the bulbous rooted
sorts that are found to interbreed with them, and have been
improperly detached. Such plants as fulgidum and echi-
natum, which have a stem of a semi-tuberous nature and
capable of enduring a long period of drought, form a curious
link between the tuberous and fibrous-rooted species. The
practicability of obtaining a cross between the hardy Passi-
flora coerulea and its more splendid but tenderer congeners
had been suggested in my communication to the Horticul-
tural Society ; and not long after Mr. Milne verified the
suggestion by the production of three fine varieties by seed
from the scarlet racemosa fertilized by coerulea. These
mules, though not absolutely sterile, are indisposed to fruit,
but seedlings were obtained from them by Mr. Milne, which
are approximated more in colour to the male parent coerulea,
and laboured under a suspicion on that account of having
been the fruit of a second cross by coerulea, which was flower-
ing in the immediate vicinity. Some time after a solitary
fruit was borne by one of Mr. Milne’s plants in the conser-
vatory at Spofforth, and although there certainly was a plant
of coerulea in another greenhouse in the garden, at a consi-
derable distance from the plant, there was no probability of

358

ON CROSSES AND

its pollen having reached the conservatory, though it cer-
tainly was possible that it might have done so. Twelve
seedlings were raised from the fruit, which was small, shri-
velled, and quite deficient in juice, and those which have
flowered not only approximated in colour to ccerulea, having
no tinge of the red of its female progenitor, but were inferior
to ccerulea in the beauty of the flower, and tenderer than
that plant ; neither of which circumstances were likely to
have occurred, if they had been derived from a second cross
with ccerulea. In the same manner I have found that the
seedlings from the crosses, between the scarlet G. cardinalis
and the white or purplish G. blandus, are always disposed
to degenerate from the colour of the more brilliant parent
and approximate themselves to G. blandus, whether the
scarlet cardinalis was the male or the female ancestor. It
appears probable that this seeming disposition in fertile
crosses to produce seedlings approaching to the least splendid
of their parents, may arise from the effects of our climate
upon them, which is more congenial to the duller coloured
than to the brighter species ; in which case it would follow,
that if the crosses were planted in the native soil and atmo-
sphere of their more splendid parent, the same degeneration
of colour would not take place. This is, however, a conjec-
ture which I have no opportunity of verifying. I was led in
some measure to form it, by having once observed the flowers
of the hardy Nymphaea alba of a pale rose-colour, after a
fortnight of unusual and intense heat in July, which ap-
peared to point out why the genus Nymphaea, which is white
in our latitudes, is found blue nearer the tropics, and red
under their influence. This suggestion does not, however,
account satisfactorily for the mule offspring, being inferior to
the mules themselves generated in a similar situation ; but I
have observed the seedlings from Hippeastrum Johnsoni or
Regio-vittatum by its own pollen to have often a corolla both
smaller and less brilliant than the mule plant itself, and this
deterioration of the descendants may perhaps be in part
attributable to the fertility of the mule being less vigorous
and perfect than that of the original parents, when there
exists some constitutional difference between them, which is
the case in these three instances, Passiflora ccerulea being
hardier than racemosa, Hippeastrum vittatum than regium,
and Gladiolus cardinalis much more thirsty than blandus.

I have already spoken of hybrid cactaceous plants of the

HYBRID INTERMIXTURES.

359

genus Cereus. Grandiflorus is also said to have been
crossed with speciosissimus at Colvill’s, and Ackermannius
has bred with both phyllanthocides and speciosissimus at
Spofforth, and I have been told that some of them have been
also crossed with the very dissimilar truncatus. There is,
therefore, every reason to suppose that the whole genus
Cereus will intermingle, and the fertility of the existing
crosses seems to open an unlimited field to the expectations
of the cultivator. I entertain such doubt of the truth of the
separation of Echinocactus that I would urge cultivators to
try whether it will cross with Cereus. In no genus, how-
ever, are more valuable results to be obtained than in that
magnificent ornament of our shrubberies, the Rhododendron,
including the subordinate family of Azaleas, which together
with Rhodora form part of the same genus. I had enter-
tained an idea that the dwarf Alpine species would be found
distinct from the rest of the genus, but I am satisfied the
suspicion was unfounded, and I believe all the species to be
capable of intermixing, though I have failed as yet in blend-
ing Az. Indica with any but its own immediate kindred.
There is a strong plant at Spofforth from Rhodora Cana-
densis by Azalea Pontica (of which cross a great number
were raised, but being very delicate when young most of
them perished, as well as another cross from Rhodora by
Rhododendron Ponticum), and small plants were raised from
Rhodora by Azalea triumphans and Rhod. Ponticum. That
by Az. Pontica will flower next spring for the first time.
Four evergreen seedlings obtained from the seed of Rhod.
Ponticum, which I had fertilized at Spofforth with pollen of
Az. Pontica, have flowered at Highclere. Two produced
yellow fragant flowers nearly of the colour of Az. Pontica,
one had flowers of a paler yellow or lemon colour, and the
fourth of an intermediate chesnut. I have raised many
weak plants from the seed of Rhododendron by yellow and
orange Azaleas, but I have found extreme difficulty in rear-
ing them, and have lost them at an early age. I had the
same bad success in trying to rear to maturity a pot full of
mules between the white Australian Nicotian a suaveolens
and the red Virginian Tabacum. The American Azaleas
have intermixed with the Nepal Rhododendron arboreum at
Spofforth, and under the care of Mr. Smith of Norbiton, at
that time the gardener of the Earl of Liverpool, who also
succeeded in obtaining seedlings from Rhododendron Dauri-

360

ON CROSSES AND

cum sempervirens by the scarlet arboreum. The latter,
which were curious little plants, are probably all dead, in
consequence of his injudicious perseverance in exposing them
at an early age in the open ground. I vainly endeavoured
to rescue the last survivor from his hard treatment. It is
remarkable that the difference of constitution between the
Rhododendrons and the American Azaleas seems to render
the mules more impatient of wet than either of the parents,
which is manifested by a sickly variegation of the leaf, ren-
dering it often difficult to rear them, and indicating the
want of a more sandy and drier soil. This may be the con-
sequence of crossing a deciduous with an evergreen species.
It is probably on account of that discrepancy that no seed
has been yet obtained from any Rhododendro-Azalea, though
the crosses of evergreen Rhododendra are sufficiently fertile,
and I have raised seedlings from Azalea Pontica-viscosa v.
alba, and Calendulacea-viscosa v. rubescens, though neither
are disposed to seed freely. The intermixture of the white
Rhododendron maximum, which is not an accidental variety,
but a widely-spread and permanent kind on the mountains
of Jersey in America, with Ponticum, has afforded a beautiful
white cross, which reproduces itself in perfection by seed,
and from that, or the American white itself, with Azalea
Pontica or the yellow Sinensis, or the still more splendid
orange varieties of calendulacea, we may expect to obtain
various Rhododendrons with more decidedly yellow flowers,
hardier constitution, and larger stature than R. Chrysan-
thum. Seedlings from the white American Rhododendron,
and also from a cross between Maximum and Ponticum, im-
pregnated by me with pollen from Az. calendulacea v. chry-
solectra, were in cultivation at Highclere, but they have
been mislaid and perhaps lost. The fragrance of the Azaleas
had been communicated to the Rhododendron, both by the
Mitcham crosses and that of Mr. Chandler. The Indian
Azaleas are probably capable of intermixing with the rest of
the genus Rhododendron, and the beautiful lilac cross obtained
by Mr. Smith abundantly between phoenicea and the white
or ledifolia is very fertile, and has produced a great number
of vigorous seedlings at Spoft’orth. Rhododendron Ponticum
and Catawbiense have produced a cross which far excels the
natural sorts in the size and complication of the umbels of
flowers, and is amazingly florid, and the further cross Alta-
clarse between that and the scarlet arboreum is of a colour

HYBRID INTERMIXTURES.

361

beautiful in the extreme, and quite hardy enough to bear

our winters, though more impatient of wet than the Pontic

and American plants, more fragile, and from its inheriting

the early habits of arboreum, very obnoxious to spring frosts.

A profusion of seedlings, now of large size and flowering,

were reared at Hio-hclere from the American blush-coloured
© «

arborescent Rhododendron, probably the maximum v. pur-
pureum altissimum of Pursh, which is more like to Ponti-
cum than to maximum, and requires a specific name (I
suggest Arborescens), impregnated by the scarlet arboreum
of Nepal, and this cross will probably be of great stature and
magnificence. The plants of that American species or local
variety have broad oval leaves. I have another permanent
variety of American Rhododendron raised from seed gathered
by Fraser from a tree in Pennsylvania, which he stated to
have been the largest he ever sawr, and capable of being
sawed into large planks. It has the leaf narrower than Pon-
ticum, and unlike any of the three kinds which are ranged
under the name maximum. I should include it under the
name arborescens. The white Nepal arboreum, with a fer-
ruginous underside to the leaf, and the beautiful but still rare
campanulatum, are hardier than the red arboreum ; and Dr.
Wallich saw in one situation the red growing at a much
higher elevation than it usually occupies, from which hardier
variety he has given us hopes of obtaining seed ; and from
these sources, as well as from the bristly Rhododendron bar-
batum, when it shall flower with us, and the beautiful Rho-
dodendron venustum of Silhet, which we hope soon to pos-
sess, our means of increasing the varieties of this desirable
family will be multiplied. It is to be hoped that the seed-
lings which I have raised from the white Rh. maximum by
arboreum, will not move so early in the spring and will suit
our variable climate better. An intermixture between the
white arboreum and the yellow or orange Azaleas will yield
a plant of great beauty. The cross between Arboreum and
Caucasicum has flowered and been duly appreciated. The
mule Altaclarae has been crossed again with a large red
Azalea at Highclere, and Azalea Sinensis has yielded a most
beautiful intermixture with the same red Azalea. The finest
flowered cross I have seen is one that I possess between Ar-
boreum and Catawbiense, and having forced this plant more
than one season I have obtained seed from it, no other Rho-
dodendron having been in flower at that time. The result is

362

ON CROSSES AND

important, namely, that it requires no label to distinguish
the offspring, which are as uniform and unlike all others in
foliage as if they were a separate species; and so in fact they
are, and, if planted by themselves in a congenial situation
and climate, would be the parents of a distinct race. This
cross I call Haylocki. The great complaint against the
mules raised by the impregnation of the Nepal arboreum, is
that they all partake of its irritability, and move so early
that they are very frequently damaged by spring frosts.
Finding them all as irritable as the male parent, I have lately
raised from it crosses by R. maximum and others, hoping that
such may inherit the tardier habits of the hardy male parent,
which would greatly increase their value. They are as yet but
two years old, and their constitution has not yet been suffi-
ciently proved. In the lovely genus Rosa, I believe, little has
been done except by the hand of accident, and the necessary
consequences of cultivation and the approximation of species
in gardens. Much remains to be accomplished, but our
climate is not very favourable to the seeding of the more
delicate sorts, and the continental cultivators do not as yet
take pains to obtain the results that might be expected from
a judicious combination of the species, when rosa lutea,
sulphurea, and bicolor shall be brought into union with the
fragrant, the double, and the ever-blowing individuals of
other species. The first decided original cross that we know
was brought by Fraser from America, where it had been
raised between the musk cluster and the ever-blowing Chi-
nese, probably by accident; and, having been sold to Mr.
Noisette, it has been made to bear his name, and, being
more fertile in France than in this country, it has become
the parent of an extensive family of beautiful varieties.
From this plant Mr. Smith raised by impregnation with the
yellowish Indian rose a variety of some merit, but not a good
flowerer under general circumstances ; and Rosa ruga is
understood to have been raised in Italy from the Ayrshire
rose by the pollen of the Chinese odorata, but the fact is not
authenticated, and, if I am rightly informed, the great variety
of cultivated roses, is owing rather to accidental than arti-
ficial intermixture. It is particularly desirable that those,
who reside in quarters congenial to the seeding of roses,
should exert themselves to intermix the qualities of the most
estimable species. Many have been lately introduced of
which I know not the origin. The honeysuckles also offer

HYBRID INTERMIXTURES.

363

an easy opportunity of improvement, by intermixing the
fragrant and more vigorous with the yellow and the scarlet.
Fruit has been grown at Spofforth from the common garden
honeysuckle by Fraser’s scarlet, but it was plundered, when
on the point of ripening, by robins. I have plants which I
have raised from an early pale honeysuckle crossed with
hirsutum, and with flavum. The French have favoured us
with some desirable magnolias from M. Yulan, fertilized by
obovata and gracilis, but the admixture of the Chinese
species with the magnificent grandiflora, and with the very
hardy tripetala is probably still in expectation. One of the
most interesting genera, on which the process of intermix-
ture has been successfully attempted, is that of Calceolaria,
because it embraces plants of a decidedly shrubby and ten-
der habit, and others which are completely stemless, and
capable of retiring to rest under ground in the temperature
of a British winter, and colours very dissimilar, the yellow
and the brownish purple ; and because most of the numerous
species which have been imported appear to intermix with
the greatest readiness, producing an endless variety of
forms. The natural effect of crossing a yellow with a pur-
ple flower should be to produce various shades passing from
the intermediate coppery tinge to the two extremities of
purple and yellow, and such is the case in the mixtures
between arachnoeides and the different varieties of integri-
folia; but the cultivators of this genus were surprised by the
breaking of the intermixture of the purple arachnoeides
with Corymbosa, which has some purple specks on the
corolla, so as to produce yellow flowers, broadly blotched
with dark and even blackish purple ; but the subsequent dis-
covery of a Chilian biennial species which has not yet been
figured, and which I call C. discolor, blotched with a red-
dish purple in a manner somewhat similar, shewed that such
an arrangement of colour was a natural variation of the
genus, which the cultivator might therefore have expected,
if all the natural species thereof had been previously brought
to his knowledge. C. integrifolia in all its varieties, includ-
ing the closely allied viscosa, is a woody shrub, attaining, if
protected, the height of several feet (I have had viscosa ten
feet high), but incapable of resisting many degrees of frost,
while C. plantaginea is absolutely stemless, and so hardy,
that although it loses its leaves in the open border, and dis-
appears in the winter out of doors, yet even in the north of

364

ON CROSSES AND

England it pushes again in the spring, and is only liable to
be lost by drought in summer, or too great a superabundance
of wet in the winter season. The application of the pollen of
the latter to the shrubby integrifolia at once reduces the
stature of the offspring from that of a shrub to a low semi-
herbaceous plant, not absolutely stemless, yet capable of
retiring into winter quarters like C. plantaginea, and not ex-
ceeding a few inches in elevation. C. Herbertiana, though
shrubby, has more affinity to the herbaceous species, being
rather intermediate between integrifolia and corymbosa in
its general appearance, and the effect of the application of its
dust to C. plantaginea is to afford an offspring more abso-
lutely herbaceous, and of which the leaves are partly radical
and partly borne on recumbent sprouts. The same is the
case with the cross between C. plantaginea and arachnoeides,
which, though it pushes out a number of herbaceous branches,
that die back in the winter out of doors, is perfectly hardy
and spreads under ground, so as to form a large clump.
The cross from C. plantaginea by rugosa (figured in the
Botanical Register first under the name ascendens which is
to be struck out, and afterwards a second time under the
right name rugosa) grows but a few inches high, and is
marvellously florid. It is further remarkable, that although
the natural species in this genus have such diversity of habits,
the crosses, as far as has been seen, are all fertile and able to
intermix, ad infinitum, though they will not bear seed as
readily as some of the natural sorts. Unfortunately C.
corymbosa which has given us a cross with the most beautiful
broken colours by intermixture with arachnoeides, called C.
Youngi, from the nurseryman who first flowered it, is one of
the most delicate species that have been introduced. C.
plantaginea is covered with minute specks underneath, and
the cross between it and arachnoeides is inclined to continue
speckled, and not to receive the ornamental blotch, but to
change the whole tint of the corolla. The cross of planta-
ginea with the annual crenatiflora is a hardy biennial one,
but it has perished with me after flowering. C. floribunda,
which endures a Yorkshire winter, may afford the means of
elevating a conspicuous branching scape from an herbaceous
stemless plant, and produce some very desirable crosses
with species that have a more ornamental flower. One very
singular monstrosity has shewn itself, though not perma-
nently, yet frequently amongst the mules from C. planta-

HYBRID INTERMIXTURES.

365

ginea ; the flower has assumed a form totally different from
its natural shape, being like a bag or purse two inches long,
widest in the middle, and gradually tapering almost to a
point at the two extremities. Sometimes one or two such
are on a stalk amongst the natural flowers, and sometimes
nearly a whole head has consisted of them. This may
authorize an expectation of very curious garden varieties
being hereafter produced in this( genus. The whole genus
agrees in constitution, liking a clear air, and a very moist
soil. The hybrid Gladioli, of which a large portion are
sufficiently hardy, flower about the same time as the roses,
and contribute quite as much in general effect to the em-
bellishment of the garden by their fine colours and profusion
of blossom. They succeed very well in the natural soil of
the garden atSpofforth, which is a good yellowish light loam,
suitable for barley, and also in the artificial borders of peat
and sand, where, however, in a dry summer they stand more
in need of water. These hardy crosses are between G. Car-
dinalis, bland us, carneus, inflatus, angustus, and tristis, and
they vary with every shade of colour from white to scarlet,
rose, coppery, and blackish purple, and some are exquisitely
speckled in consequence of the cross with tristis. They
succeed best when grown into a thick tuft, in which state the
profusion of blossom is admirable, the cluster of bulbs and
the old skins of decayed bulbs permitting the wet to drain
away, and preventing the earth from lying too close and
heavy on the bulbs in autumn and winter. Clusters have
now stood undisturbed at Spoffbrth above twenty years, with
the precaution of covering them with leaves from November
to March or April. There is danger in disturbing and
parting them, for numbers will rot if re-set separately ; and,
if they must be divided, it is best to do so in April, or, if it
be done in the autumn, the roots taken up should be potted,
and turned out again in the spring. The beautiful crosses
with hirsutus, recurvus, and versicolor are more delicate
plants, and do not succeed well in the border. I have not
succeeded in obtaining any cross, on the correctness of which
I can depend, by admixture with Gladiolus psittacinus
(Nathalensis), and I do not believe that it will breed with
any of the above. Like all the European species, it pre-
sents its flowers in front of the stem, which is erect ; and
repeated experiments have shewn that every flower of G.
tristis which was touched with the pollen of G. Byzantinus

366

ON CROSSES AND

only, failed of making’ seed, while every flower to which the
natural pollen had access produced it, and I consider the
union of Byzantinus with any of the species above enume-
rated, except Psittacinus, to be impracticable. I consider
Alatus to belong to the same family as Psittacinus, and also
a beautiful species imported, likewise, as I understood, from
the neighbourhood of the Nathal river, which I propose to
call G. oppositiflorus. It is now sold by the Dutch nursery-
men under the name floribundus, which has been long pre-
occupied. Its flowers, twenty-four or more in number, pre-
sent themselves alternately from the two sides of a robust
erect stalk. Blandus, Cardinalis, &c. have the flowers
rising upwards from the back of a bent stalk, and seem to
constitute a family distinct from those which present them-
selves in front. G. oppositiflorus has the flower much undu-
lated, white, dashed with pinkish purple. The genus
Gladiolus ought to be divided into at least two sections or
sub-genera. I consider a sub-genus to be such a portion of
any genus as will not intermingle with the rest, and has
some distinctive appearance, but insufficient to induce a
belief of their original diversity. Seedling Gladioli will
flower often the first autumn ; the best treatment is to sow
the seed in pots, and give them shelter till the seedlings are
pretty strong, and then turn out the ball unbroken into the
border, where they will produce a crowded nosegay of flowers
of various shades of colour.

It is not, however, by crossing different species or local
varieties of plants only, that the cultivator may add to the
beauty of his collection. Much may be done undoubtedly
by crossing judiciously the finest seminal varieties of such
plants as have been already improved in our gardens, and
are disposed to break into a multiplicity of forms or colours.

It is to be observed, that in some cases the seminal va-
rieties of plants preserve themselves almost as distinct in their
generations as if they were separate species : for instance, the
cultivated double holyoaks, of which at least the orange, the
yellow, the white, the black, the red, and the pink, may be
raised with certainty by seed from plants of the several co-
lours, although planted near together in the garden ; and it
is probable that if gardeners were to take the trouble of
crossing them with the pollen of plants of a different colour,
a greater multiplicity of hues would be procured. In car-
nations also the seedlings have a great disposition to follow

HYBRID INTERMIXTURES.

367

the colour of the parent plant. I have had greater success
than any other person in raising from seed double camellias
of various tints and appearance, and some of the best have
been produced either from single flowers or plants raised
from single ones, impregnated by the pollen of double
flowers, preferring, where it can be got, the pollen that is
borne on a petal. The new seedlings that flowered with me
in one spring for the first time were nine full double ; three
semidouble, of which one was very fine, and only three
single ; but such an unusual result is not to be obtained
without particular attention to the mode of treating the
mother plant while in flower and seeding : the method
which I have adopted being to keep it in confined air, with a
superabundance of water, even to the detriment of its health,
and to prevent it from making young shoots, in a great mea-
sure, if not entirely, by which means an exuberant degree of
nutriment is forced to the seed-vessel. The reason that the
seedlings raised by some nurserymen are so very inferior is,
that their plants are in the most luxuriant growth ; and it
cannot be expected that seed gathered from individuals
growing with freedom and vigour, should not be more dis-
posed to reproduce the natural form of the plant, than to
yield the fine cultivated varieties, which are to be obtained
from them when almost diseased by repletion. The finest
double varieties of Camellia Japonica which I have so raised
are as follows : —

From the single white by the pollen of the Pompone, 1. var.
Spofforthise, or Spofforth striped , very large and very
double, white, with a few pink stripes, and occasionally one
or two anthers. — 2. v. Maculosa, or Calypso , do. — 3. v. Hay-
locki, or Haylock's white; pure white, rarely a few anthers.
— 4. v. Eburnea, or Ebur ; very vigorous, pure white; some-
what waratah shaped. — 5. v. Nivosa, or Nitor ; double white,
variable in form.— 6. v. Fortuita, or Fortuna ; very like var. 1.
— 7. v. Lactescens, or Luna ; double white. From seedlings
impregnated by the Pompone, which had been raised from
the common single red by the striped. — 8. v. Pumila, or
Circe ; regularly formed double white dwarf myrtle-leaved.
— 9. v. Ulantha, or Hylas; white striped with pink; flowers
in four uniform compartments. — 10. v. Lysantha, or Eysi-
machus ; flowers if possible more regular than the buff or
old double white, red with a watery white line and margin
to each petal. A very erect plant of rapid growth with

368

ON CROSSES AND

flowers of first-rate merit. — 11. v. Yictrix, or Victoria ; own
sister to Lysimachus, equally regular, of the colour of a
full-blown cabbage rose, paler near the edges. From the
Chinese semidouble by Pompone. — 12. v. Picta, or Alcmene;
very regular in general ; with a pink stripe usually on each
petal, the white changing after some days to blush, some-
times less regular, with one or two anthers ; very beautiful.
From the Pompone. — 13. v. Spofforthiana rosea, or Idiina ;
superior to the Peony-flowered in form and colour ; the
flower has always some anthers like its parent. From the
waratah by the striped. — 14. v. Foliolosa, or Amalthea ;
flower-shaped like the rose-scented peonia edulis, v. rosea,
red, with about 350 petals. — 15. v. Conferta, or Odin; fine
double red, not regular. I have never seen any anthers in
either this or the preceding. — 16. v. Porrecta, or Bellona ;
fine crimson ; branches horizontal or weeping. From waratch
by Pompone. — 17. v. Modesta, or Hebe; flower nearly regu-
lar, of a delicate purplish pink. From waratah by 13. Iduna.
— 18. v. Roslgena, or Penelope; double red. From a seed-
ling from single red by striped, fecundated again by striped.
— 19. v. Molesta, or Nemesis ; very double red, but a delicate
plant. — 20. v. Venosa, or Venus ; flower regular, but not
sufficiently full, red veined with white. This has produced
but one flower yet, and I am not sure of its permanent supe-
riority. Many others of much merit I have not thought
worthy of being named ; and amongst them is one full-
double red, raised immediately from the common small-
flowered single red. I have a great multitude of seedlings
which have not flowered yet, from which I anticipate much
beauty and variety. I scarcely entertain a doubt that the
double pink Camellia Sesanqua (Maliformis of Lindley) is a
cross-bred plant between C. Japonica and Sesanqua; and,
from its seeming sterility, I cannot but suspect that C. reti-
culata is not a genuine species, but a cross, perhaps obtained
from some species still unknown to us.

Mr. Chandler obtained some very fine varieties from the
waratah, impregnated by the striped, one season, but those
which he has raised since have not proved good. It is,
therefore, probable that there was some difference in the
treatment of the plant or plants which bore seed for him
that season, though accidental and unnoticed by him. His
finest productions are eximia, somewhat like imbricata, Bi-
roni, one-while called concinna, a very remarkable flower,

HYBRID INTERMIXTURES. 369

regular and oddly flattened, but very beautiful ; Woodsi, a
large rose-coloured flower, quite double, but cup-shaped and
hollow in the centre, requiring a little warmth to flower it in
perfection ; Chandleri, striped, sometimes very fine, but not
always equally so. His elegans, rosa sinensis, and florida,
are handsome also ; corallina and althaeiflora sometimes, but
often producing poor semi-double flowers. His anemoneflora
alba comes very near in flower to its parent the Pompone,
with a much less hardy constitution. Mr. Gray produced
three cross-bred seedlings, of which Press’s eclipse is the
best, and Colvill’s nursery two speckled seedlings of con-
siderable merit, though very irregular, and too muddy in
the colour. I have seen no other seedling Camellia that de-
serves to be preserved, but I have been told that Mr. Gray
has since raised a good red one. His former plants were
said to have been crosses between the single white and
Chinese semi-double. These observations may perhaps tend
to the raising still finer varieties, when the mode of obtain-
ing them is rightly understood. I have no difficulty in ob-
taining seed from any given flower of the Pompone or Mid-
dlemist’s Camellia, by putting it in a house rather warmer,
arid with less admission of air, than suits greenhouse plants
in general; impregnating the stigma, and taking off the
corolla before it begins to decay, and cutting away the petals
that adhere to the germen or young seed-vessel, that the air
may have free admission to it ; without which precaution it
will perish in most cases from damp. The striped sorts have
usually more white in their flowers when they flower early
in the spring, and it seems th&t the seed ripened earliest in
the year is the most apt to yield white or pied seedlings.
There is a strange mutability in the flowering of Camellias,
of which the Pompone, which has been called on that account
variabilis, furnishes a striking instance. It has four dis-
tinguishable kinds of flower, the pure white and the red-
eyed, which appear promiscuously, the brindled pink,
and the rose-coloured, which may be kept separate with
tolerable certainty by grafting from the branch that bears
them, the rose-coloured form being the Peony-flowered of
the nurserymen. There is a branch on my oldest plant of
the peony-flowered, which has reverted to the pure white
colour, an occurrence less common than the departure from
it. Carnations, which have run to red, very seldom re-
vert to the white-stripe. I have been informed that the

2 B

370

ON CROSSES AND

Chinese do not reckon seedling Camellias confirmed in their
habit, till they have flowered six or seven seasons without
becoming less double. I have not found any of mine, thus
raised several years ago, degenerate from their first appear-
ance. Of the Chinese, the double white, the buff, the
fringed white, and, as far as we know, the red variety, called
imbricata, are the only sorts that never bear anthers. Hav-
ing cultivated the myrtle-leaved above twenty-five years, I
never saw that variety bear an anther in my collection, ex-
cept one season, when all the flowers on even^ plant of the
kind had them, and they were found in two or three late
flowers last year ; but the seedlings reared from its pollen, of
which great expectations were entertained, proved to be the
worst 1 had ever raised, and it seemed that whatever peculi-
arity of the season inclined the flowers to deviate from their
usual double form, and approach nearer to the fertile single-
flowered original, disposed also the pollen to generate single
seedlings. I have seen the myrtle-leaved with anthers at
Mr. Knight’s nursery, though the circumstance has been so
rare in my own collection ; perhaps it may be connected with
the more or less luxuriant growth of the plant.

It is to be lamented that more experiments have not been
tried to improve the races of agricultural vegetables by
crossing. I impregnated in 1834 with great care the Swe-
dish turnip (ruta-baga) with pollen of the white, and another
branch thereof with that of the red-rooted turnip, which pro-
duces perhaps a greater tonage than the white, bearing both
frosts and unfavourable summers better, and thriving in soils
where the white does not succeed. The seed was sown im-
mediately, and the plants of both crosses, though late, formed
pretty good roots. The leaves differed in appearance from
those of the Swedes, and did not, like them, retain the rain-
water on their surface. In the following spring they were
set for seed in two different situations where no extraneous
pollen might have access. The flowers of the greater part
were of the bright yellow of the two male parents ; a smaller
portion in each lot produced straw-colour blossom, like
that of the Swede : but not one shewed the least disposition
to an intermediate tint ; and it seemed as if those two colours
were incapable of blending, or modifying each other. I have
a crop from their seed this year, but the season has been par-
ticularly unfavourable for all turnips ; the fly destroyed the
first sown, and the plants being again too backward, 1 do not

HYBRID INTERMIXTURES.

371

think their value will be fully ascertained from the present
crop.

There seems no reason to doubt that better varieties of
wheat, oats, and barley may yet be obtained by combining
the hardiness of one, with the productiveness of another
sort, and the finer skin or greater weight of a third. I
am inclined to think that I have derived advantage from
impregnating the flower, from which I wished to obtain seed,
with pollen from another individual of the same variety, or
at least from another flower, rather than with its own ; and
as races of animals are known to degenerate, if they are perpe-
tuated by the union of near kindred, it seems not unlikely
that vigour may be given also to any race of vegetables by
introducing across, though of the same kind, and especially
from an individual grown in a different soil or aspect. To
illustrate this, I will state a circumstance which occurred
last summer in my stove. Nine very fine crosses of Hippe-
astrum were flowering there at the same time ; one a natural
seedling from Johnsoni or Regio-vittatum, two Johnsoni-pul-
verulentum, one Johnsoni-vittatum, one Psittacino-Johnsoni,
one from Psittacino-Johnsoni crossed again by Vittato-John-
soni, one from Johnsoni by Solandriflorum, and two from Vit-
tato-Johnsoni by the same. Being desirous of blending again
these plants which were all cross-bred, different flowers were
touched with pollen from their several neighbours and tick-
eted, and other flowers were touched with their own pollen.
Almost every flower that was touched with pollen from ano-
ther cross produced seed abundantly, and those which were
touched with their own either failed entirely or formed slowly
a pod of inferior size with fewer seeds, the cross impregna-
tion decidedly taking the lead. It appears to me that this
circumstance may be analogous to the introduction of a male
from another flock or herd, which has been found advanta-
geous to the breed of domestic animals ; and I would advise
gardeners to try the effect of setting flowers with the dust
from another individual in preference to their own, with a
view to obtain an improved breed.

It is only from the superior efficacy of the pollen of another
plant, that we can account for the circumstance of some hy-
brid plants, which breed freely with plants of either parental
stock and fecundate them, not producing seed readily when
left to themselves; for if their pollen is able to fertilize, and
their ovary to be fertilized, there can be no positive sterility

2 b 2

372

ON CROSSES AND

in the plant, though there may be a want of sufficient energy
under certain, or perhaps under ordinary, circumstances.
Many centuries of experimental cultivation must elapse be-
fore the subject can be, if ever, fully understood ; and I can-
not suppose that my present view of it will not require to be
modified by the results of future investigation. For instance,
there seems little prospect of being able to answer why the
hybridizing process is so easy in some genera and so difficult
in others, if equally facile of access, unless it shall be found
to arise from greater or less constitutional conformity.

The genus Calceolaria affords greater facility than most
others, because its stigma is nearly obsolete before the pollen
of the flower is ready, and, in the earliest stage of the bud, it
is easy to lift up the corolla, and take out the anthers, which
are then comparatively large and exposed, and the stigma
may be fertilized at that early period, when it is defended by
the covering of the corolla from any accidental intrusion.
Amongst the Amaryllideae there is for the most part much
facility of performing the operation in the several genera,
the anthers not being reversed to display the pollen till a
little while after the expansion of the flower ; yet in the
genus Hippeastrum there is a complete readiness of all the
species to intermix when crossed artificially, and in the
genus Crinum nearly so, while in Zephyranthes it is ex-
tremely difficult to obtain hybrid seed, and repeated disap-
pointments occur from the escape of some particles of the
natural pollen in taking out the anthers. In the genus
Crinum one unintelligible impediment appeared for a long
time to exist. C. Capense, which bred freely with every
other species, refused to be fertilized by the tropical Cape-
coast kinds, Broussonetianum, and petiolatum or spectabile.
A seedling has, however, at last been obtained at Spofforth
from C. Capense by the latter, which I believe to be correct,
and it can scarcely be doubted that the difficulty arises from
some constitutional peculiarities in those plants. In general
hybrid plants have been found to be excessively florid, but
sometimes the contrary has been the case, and there appears
to be some impediment to the perfection of their blossom.
The mule between Hymenocallis disticha and rotata, which
was raised many years ago, whether it be in the stove or in
the open air, where it grew against the front wall of the
stove, throws up, after its proper time of flowering, an abor-
tive scape, on which the buds are dead and discoloured, as

HYBRID INTERMIXTURES.

373

Amaryllis Belladonna does at the time its leaves push in the
spring, when the previous autumn has been unfavourable to
its flowering. In the course of above fourteen years since it
was raised, it has only once attempted to expand its flowers,
and that in a very unsatisfactory manner in the stove ; but
I have lately had reason to suspect that more wet is neces-
sary to it than to either of its parents, and perhaps absolute
immersion at the time of flowering. I may take this oppor-
tunity of stating that the plant which I fancied many years
ago to have been obtained from C. Capense by Hymeno-
callis (then Pancratium) disticha, proved, as it advanced, to
have been by C. Canaliculatum. There had been an error
in the memorandum made concerning it, or the flower had
been touched by the pollen of both plants. A very interest-
ing Crinum, of which only one plant was raised, from C. de-
fixum by speciosum for several years put forth abortive
scapes, but it has flowered well the two or three last seasons,
though it has yielded no increase in any manner ; neither
has the fine plant which was raised from Scabrum by Cana-
liculatum. One of the handsomest white sorts, from C. bre-
vifolium by the larger variety of erubescens, having a strong
red root-stem, somewhat like Amabile, has afforded many
offsets but no seed. The only genus in which I had observed
barrenness of the offspring appearing to arise from the bota-
nical difference of parents, whose constitution seemed very
similar, was Nerine, of which the crosses between the divi-
sion with regular and that with distorted filaments had borne
no seed ; but in that case the discrepancy was so important
that it might have been almost supposed to afford a generic
distinction, and Mr. Salisbury had named the Distortae
Loxanthus. In the article Nerine, p. 283, I have given an
account of a mule from the distorted N. pulchella by the
regular curvifolia, of which the flowers are exceedingly
similar to those of the cross between undulata and curvi-
folia, plate 45, but more healthful and free. The last-men-
tioned cross, as far as I have seen, is quite sterile, the parents
having differed not only in regularity of perianth, but in the
mode of flowering ; for the inflorescence of undulata is cen-
trifugal— that of pulchella, as well as curvifolia, centripetal ;
from which conformity I anticipated the more probable fer-
tility of the mule. That conjecture has been verified, since
the former pages were sent to the press, by the production of
healthy seed from the mule curvifolia-pulchella, and an abun-

37 i

ON CROSSES AND

dant crop from two plants of curvifolia by the pollen of the
mule, no other Nerine having been permitted to develope its
anthers on the premises. Here, then, is a feature which had
been overlooked, which seems, nevertheless, to have a pow-
erful influence over the fertility of the offspring. The seed-
ing of this mule is fatal to Salisbury’s genus Loxanthus, if
any doubt could have remained after the production of the
former intermixture. In the tubular African heaths the
pollen remains confined, unless the anthers are touched by
something inserted, as the point of a pin or the proboscis of
an insect, when they spring asunder and discharge it. This
genus, therefore, affords greater facility of intermixing, and
it is probable that some of the native species, which are said
to be quite local, have been produced by accidental intermix-
ture of two other kinds. There is a natural species of Goodia,
quite permanent by seed, which I had many years ago
named intermedia, but which appears in Sweet’s Hortus
Britannicus under the name subpubescens, which is so
exactly intermediate between lotifolia and pubescens in all
points, that it can scarcely be doubted that it might be pro-
duced by crossing those two species. Amongst other crosses
of Ericae, I obtained at Mitcbam many plants from two very
dissimilar, namely, from Jasminiflora by vestita coccinea,
which had the foliage slender and near an inch long. The
late Mr. Salisbury had conceived that those two species,
being; distinguished by a shorter and a longer and more
pointed pod, were referable to two distinct genera to which
he had accordingly assigned names, and he told me that I
should fail in any attempt to cross them ; which was answered
by shewing him the seedlings then several inches high.
They were all lost on, or soon after, removing to Spofforth
before they had flowered, though one of them was above a
foot high. The disposition to sterility which lias been stated
to exist especially in the offspring of parents of different con-
stitutions, offers a great impediment to the unlimited use of
crossing in the fruit-garden, but it is certain that great ad-
vantage may be derived by the cultivator, who will strive to
bring together the various good qualities of the sorts between
which no such obstacle exists ; and the complete fertility of
the fruit-bearing Cerei makes it very uncertain where such
obstacles will be found to interfere, before the experiment is
made. 1 have already mentioned that Crinum scabro-capense,
though the pollen of different species was applied to it had

HYBRID INTERMIXTURES.

375

continued about sixteen years perfectly sterile. In 1834 a
plant of it which had been growing- the greater part of that
time out of doors in front of the stove, produced one small
seed. It vegetated, but the leaf was from the first of a yel-
lowish white, and the plant did not live many weeks. In
1835 it produced another and larger seed, the early part of
the summer having been very hot both those seasons. This
seed was sown in white sand to try to save it from perishing
like the former, and a thriving young plant has been obtained
from it. Whether they are the produce of its own pollen, or
that of Pedunculato-capense, which grew beside it, cannot
yet be judged with certainty ; but the seedling now growing
vigorously, has deep green leaves, and does not shew any
approximation to the glaucous hue* of C. Capense, of which
a large bed was not far off; and that hue would probably
have been very apparent, if it had been so crossed again. I
had often attempted to fertilize C. Capense by the pollen of
this beautiful mule unsuccessfully, but the circumstance of
the two seeds it has borne shews that it is possible to obtain
such a second cross, which would be a great acquisition, as
it would certainly yield a plant of hardier constitution, and
able to bloom in our open gardens, with much greater beauty
of flower than Capense itself.

I have not found as yet the results which might perhaps
have been expected, and which Mr. Knight seems to have
obtained, from carefully blending the pollen of more than
one species before its application. I attempted to fecundate
calceolaria plantaginea with the pollen of twelve species, most
industriously mixed together, but very few seeds were ripened,
and the produce differed very little from those which had
been procured by the pollen of one of the twelve species.
Further experiments are necessary to establish how far the
influence of different males can act simultaneously, by admix-
ture of the dust. I have obtained mule seed and natural seed
from the same capsule, but they were probably formed in

* Since these pages were prepared for the press, the supposed seedling from
C. Capense by Spectabile, which had grown very slowly, having been immersed
in water in the stove has pushed vigorously, and my present opinion is that it will
prove to be a natural Capense produced by the escape of some particle of maternal
pollen, notwithstanding all the precautions which had been taken. This seed-
ling had been glaucous from an early age. The refusal of the W. African spe-
cies, Spectabile and Broussonetianum, to breed with Capense is therefore not yet
overcome.

376

ON CROSSES AND

different cells. Experiments should be made to ascertain
whether, in cases of partial and imperfect fecundation, the
pollen of another species, and even of nearly allied genus
which could not alone fertilize the ovary, can act in conjunc-
tion with a single grain, or at least with an insufficient quan-
tity of the natural dust to effect the fertilization, and occasion
the seed to produce a variety, not actually hybrid, but in
some degree departing from the natural form. See, above,
the account of Hymenocallis amoena, var. lorata, p. 211. It
is certain, by the result of many experiments, made at Spof-
forth, that the pollen of a nearly allied genus, which cannot
effect fhe production of seed that will vegetate, will often
cause some of the ovules to swell to a large, and occa-
sionally to a preposterous, size, and become seed-like masses
without an embryo, and the same circumstance has been
observed in Germany ; and, as it can act so far, I do not see
the impossibility of its influencing the character of the pro-
duce, where the access of natural pollen is insufficient; and
it seems to me questionable whether some of the singular
varieties which occur among vegetables may not have been
so produced.

A very singular occurrence in the history of cross-bred
plants took place last year in the garden of my brother
(Hon. Algernon Herbert), at Ickleton, in Cambridgeshire,
which deserves the attention of naturalists. In 1834 he
purchased a plant, grafted from a hybrid Cytisus, known to
have been raised in France between C. laburnum and pur-
pureus, of which the leaves are as large as those of la-
burnum, though a little different in form, the flowers of a
dingy and rather coppery purple in long racemes. The
plant purchased consisted of a strong laburnum stock about
8 inches, and a grafted scion about 16, high. Its growth
that year was vigorous. In 1835, from a strong branch,
which was not in existence when the plant was purchased,
proceeded a small shoot a foot and half long, covered
with small leaves of the exact size and very nearly of the
form of those of the little C. purpureus, while the rest of the
tree, having reached the height of 8 feet, had the usual
large foliage, approaching in appearance to that of the la-
burnum. This extraordinary branch, which has in a great
measure, though not exactly, assimilated itself to the struc-
ture and habits of that one of the two parents to which the
mule had originally the least resemblance, has this year, like

HYBRID INTERMIXTURES.

377

that parent, produced small flowers, four at each joint, from
the axills of the leaves, in the same manner as C. purpureus,
and of a colour more purple than the pendulous racemes on
the other branches, which had about 16 flowers on a spike.
This fact is scarcely less wonderful than if a mule, between
a mare and an ass, were at three years old to acquire an
ass’s tail. It was positively stated to me, 30 years ago, by a
nurseryman, that Spong’s rose was not a seedling, but an
accidental sucker from the rose de Meaux, which had as-
sumed a different character of leaf and flower, and main-
tained its diversity : a phenomenon which, if true, appears
to be in some degree analogous. I have also seen the flower
of the yellow Austrian rosa lutea borne upon a branch of the
two-coloured var. bicolor, improperly called a Persian spe-
cies, purpurea, by Sweet ; but certainly a garden cross
from lutea. Jacquin had observed, also, that this two-
coloured plant sometimes bore flowers entirely yellow.

I am not informed which was the female parent of the
hybrid cytisus, but I entertain no doubt that it was the la-
burnum, because the foliage approximates to it, and the
flower follows rather the colour of purpureus ; as the mule
Rhododendrons by Azalea Pontica have the evergreen leaf of
the former, and are more disposed to follow the yellow colour
of the latter. The natural leaves of the hybrid cytisus are
about four times as long, and four times as broad, that is,
sixteen times as large as those of the curious branch on which
the leaves are as crowded as on C. purpureus ; the general
foliage of the tree, though altered from the exact shape of the
laburnum, being little, if at all, reduced in size. I have
been told, but cannot verify the fact, that a like circnmstance
has occurred in France to a plant of the same mule. Grafts
or layers from the anomalous branch will in all probability
preserve their acquired character, and be so propagated as a
distinct plant.

It was apparent to me that no botanist had been able to
distinguish Nicotiana, Salpiglossis, and Petunia, except by
features which I knew to be unsupported by the fact, though
Salpiglossis, in my humble opinion, has been erroneously
placed in a different order and alliance from the others,
plants with five stamina being considered Solaneee, and
those in which the fifth is wanting Scrophularineae. I had
ascertained the utter invalidity of that feature, having seen
flowers of the same Salpiglossis with only four stamens, with

378

ON CROSSES AND

five fertile stamens, and with four fertile stamens and one
abortive ; and, as in Nicotiana the fifth stamen is always of
a different power and inserted differently from the other four,
I was inclined to believe them to be one genus, and I made
a great many attempts to cross them, but in vain. On further
examination of those plants I find a diversity of the seeds,
which are very small, and must be examined with a mag-
nifier. Those of Salpiglossis are angular ; see pi. 43. f. 50.
In other respects, although its aspect is easily recognized, it
is not so easily separated by any decided distinction from
Nicotiana; and it has been incorrectly characterized by one
of the stamina barren, since the fifth is sometimes entirely
wanting, and sometimes fertile. The character of Nicotiana
in the Bot. Mag. is quite untrue ; founded on one species,
and inapplicable to others. That genus affords an instance
of the unimportance of the more or less continued adhesion
of decurrent filaments, the adhesion varying greatly in the
different species. See the seeds of Nicotiana, pi. 43. f. 51
and 52. They are not angular, but more oblong and reni-
form than those of Petunia, which are nearly round, and
pitted all over. See Pet. linearis, pi. 43. f. 48. I never
doubted the diversity of Nierenbergia, but it stands yet un-
defined. Dr. Lindley characterised it by the reflex uneate
lobes of the stigma, but the stigma will prove often a trea-
cherous feature in botanical characters ; and Professor Don,
having found in N. aristata a stigma of which the lobes are
very little more reflex than in Petunia, at once assumed that
they were one genus. They are, I doubt not, substantially
distinct, but the generic character of Nierenbergia must be,
Tube slender cylindrical , limb wide-funnel-shaped, stamina ad-
hesive to the style and stiyma. I have never seen more than
one seed perfected in a pod of Nierenbergia, the rest of the
ovules proving abortive, and the capsule so small that it is
often overlooked when ripe ; the seed of filifolia is oval with
the back rounded. See pi. 43. f. 47. I see nothing in
N. aristata to reconcile it to Petunia. The lovely N. caly-
cina has a much greater affinity to Petunia from its trailing
inflorescence and general aspect ; but, if Petunia can be
shewn to be one with Nierenbergia, there will remain no-
thing to separate them from Nicotiana. I am, however,
now almost satisfied of the diversity of Petunia and Nicotiana,
wishing, however, cultivators to persevere awhile in the at-
tempt to cross them ; for Petunia and Nicotiana have much

HYBRID INTERMIXTURES.

379

closer affinity than the others, and their distinction is more
questionable. I am, however, by no means satisfied that, if
they are distinct, Petunia linearis (Salpiglossis integrifolia,
Bot. Mag. Nierenbergia, Sweet's B. f. g.) may not belong
to a fifth genus. It is certainly neither a Salpiglossis nor a
Nierenbergia, and its seeds conform with Petunia, but it has
a different aspect, and I cannot cross it with the other sorts
of Petunia. It will belong at least to a separate section of
Petunia with linear leaves. It is very remarkable that,
although there is a great difference in the form of the flower,
especially of the tube, of P. nyctanigenseflora and phoenicea,
the mules between them are not only fertile, but I have found
them seed much more freely with me than either parent.
The mules I had raised from the former by the latter, having
been forced early in the spring of 1835, set their seed before
any other Petunia was in flower on the premises, and must
therefore have been fructified by its own pollen. The white
impregnated by the dust of the mule, and the mule by it,
produce a great deal of sporting, but from a pod of the above-
mentioned mule to which no pollen but its own had access,

I had a large batch of seedlings in which there was no varia-
bility or difference from itself ; and it is evident that the
mule planted by itself, in a congenial climate, would repro-
duce itself as a species ; at least as much deserving to be so
considered, as the various Calceolarias of different districts
in South America.

I have little to add to this treatise, but my regret that it is
necessarily so imperfect ; and, from the nature of the subject
and the additions made to it since it was first written, of
somewhat too desultory a character. I hope, however, that
it may have the effect of removing some erroneous impres-
sions, and contribute its humble mite towards the elucidation
of truth ; and that, by giving the public a clearer view of
what has been effected, it may enable those, who are dis-
posed to pursue experiments on this subject, to conduct them
with greater advantage. I have by no means enumerated
all the genera in which crosses have been lately introduced,
as for instance, Potentilla and Anagallis, in the last of which
I have seen a remarkable result in the production of a red-
dish purple flower, by the union of the orange with the
bright blue. I have an hybrid from the little Hibiscus
ficulrieus by manihot, which, with leaves that preserve the
form of those of ficulneus, has the vigour and stature of ma-

380

ON CROSSES AND HYBRID INTERMIXTURES.

nihot with its terminal spike, but witli small axillary flower-
ing branches also. I bad likewise a cross from H. palustris
by speciosus, but the plants were so delicate that all died
before they had made a fourth leaf. I apprehend that
several genera are comprehended under the name Hibiscus,
which shews a great diversity of fruit, and an interesting
course of experiments might be conducted to ascertain
whether any cross can be obtained between those which differ
in that respect, and whether they are all convertible within
certain demarcations. Cultivators are too apt to believe they
have obtained the cross they have been desirous of pro-
ducing, when they have really a natural seedling variety.
I can have no doubt in saying that the plant figured under
the name Azalea Rawsoni, (Paxton, p. 123.) which Mr.
Rawson’s gardener fancied to be a cross between Az. Indica
and Rh. Dauricum, is not allied to the latter plant, but a
genuine Az. Indica, perhaps from a cross between two va-
rieties of that plant. In speaking of the varieties of Camellia,
I should have noticed Ford's handsome variety in Paxton’s
work, but I never saw it, and am ignorant of its origin.

AMARYLLIDACEiE.

CHARACTERES BREVIORES.

Div. 1. Ramose. Subordo 1. Xerophytece ; folia rigida.

1. Xerbphyta ; flores terminales, per. tubatum limbo regu-

lari, antherae lineares subsessiles.

2. Vellosia; flor. non term. per. campanulas vix tubatum,

fil. brevia interdum catervatim superflua, antherae
erectae.

3. Barbactnia ; flor. non term. per. tubatum limbo reflexo,

fil. bifida antlieris adnatis longiora.

Div. 2. Caulescentes. Subordo 2. Hypoxidece; schistandrae,
operculatae. ( Weldenia ? vide p. 64.)

4. CurcuUgo ; folia plicata, tubus superne cylindr. fil. dis-

tantia, semina hilo profunde eroso.

5. Molineria ; fol. plic. tubus brevis, antherae fasciculatae,

semina subrotunda.

6. Hypoxis ; fol. plic. per. profunde sexfidum basi annulare,

sub sole patens.

7. Ccelanthus ; fol. plic. per. campanulatum connivens.

8. Anigozanthus ; folia equitantia, per. tubatum limbo brevi

infra profundius fisso.

9. Landria; fol. equit. per. tubatum limbo regulari.

10. Lophiolci ; fol. equit. per. profunde fissum limbo regul.

reflexo, antherae versatiles, stylus tripartibilis.

11. Conostylis ; fol. equit. per. prof. fiss. limbo regul. semi-

expanso, stylus conice dilatatus.

12 Phlebocdrya ; fol. equit. per. prof. fiss. antherae subses-
siles, stylus filiformis.

13. Campynana ; fol. equit. per. prof. fiss. filam. recurva,

styli partiti recurvi.

14. Chceradodia ; caulis umbellatus, sepala petalis diversa,

stylus strumosus, capsula trigona.

15. Alstrcemeria ; caul. umb. rectus, sep. petalis div. per.

quater dispar, stylus filiformis recurvus, caps, non an-

382

AMARYLLIDACEjE.

gulata operculo rigido attenuate) valvis inferne dissili-
entibus, semina subrotunda tuberculata.

16. Collania ; caul. umb. nutans, sep. petalis div. caps, non

angulata, operculo grandi pulpaceo.

17. Sphce'rine ; caul. nmb. subrectus, sep. petalis div. peri-

carpiurn indehiscens operculo obsoletiore.

18. Bomdrea; caul. umb. tortus, sep. petalis div. caps, ob-

tuse trigona superne debiscens.

Subordo 3. Agavece ; schistandrae, non operculatae.

§ 1. Bioscoreaformes ; caul, tortus, inflor. axillaris, venae fol.
saepissime confluentes.

19. Tamus; pericarpium extus evalve, integum. med. pul-

paceo, intimo trivalvi septigero, sem. subrotunda testa
dura.

20. Testudinaria ; caps, tri valvis, semina complanata. fil . basi

lacin. inserta ?

21. Bioscorea ; fil. toro inserta? caps. triv. sem, complanata.

22. Rajania ; caps, monosperma, ala grandi incurva.

§ 2. Ixiceformes ; radix bulbosa.

23. Bravoa ; inflor. spicata, per. tubatum limbo brevi.

24. Ixiolirion ; per. profunde fissum, regulare, stamina per-

fecta.

( Tecophilea ? per. profunde fissum, stamina, tria fertilia,
tria abortiva. Iridaceae ?)

§ 3. Agavece ; bulbus imperfectus vel caudex elongatus.

25. Fourcroya ; caul, spicatus vel ramosus, per. non tubatum,

filam. brevia incrassata, stylus strumosus.

26. Agave; caul. spic. vel ramosus, per. tubatum, genit.

filiformia.

27. Doryanthes ; caul, capitulatus, per. tubatum, antherae

extinctoriiformes.

Div. 3. Scapace^e; scapus succulentus infra spatham non
articulatus.

Subordo 4. Amaryllidece ; sebistandrae, non operculatae.

§ 1. Cyrtanthiform.es ; scapus cavus, germen ovale, caps, tri-
valvis, sem. testa nigra, tubi faux ampla. Plantae
Africanae.

28. Cyrtanthus ; tubus curvatus, filam. recta tubo decur-

rentia.

29. Gastronema; tubus curvatus, filam. conniventia, tria

deflexa.

30. Vallota ; tubus rectus, filam. recta tubo adnata. Sem.
basi longe-alata vacua, chal. obliqua, album, et embr.
verticem (saepius transverse) juxtaposit.

CHARACTERES BREVIORES.

383

§2. Hippeastr formes ; scapus cavus, germ, superne latius
medio constrictum, tubi faux arcta, caps. triv. sem. testa
nigra. Plantse occidentales.

31. Sprekelia ; per. prof. fiss. declinatum ; fil. declin. fascU

culata. 1-2 flora.

Pluriflorse.

32. Hippeastrnm ; per. declin. tubus infra abbreviatus, mem-

brana faucialis ubi manifesta non annularis, limbus
subpatens laciniis quater disparibus.

33. P hy cello, ; tubus sequalis, limbus convolutus, membrana

faucialis sequalis.

34. Habrantlius ; per. declin. tubus sequalis, limbus non con-

volutus, membrana faucialis annularis, fil. recurva, an-
therse medio affixse incumbentes.

Uniflorse.

35. Zephyranlhes ; tubus suberectus, limbus regularis, fila-

menta distantia conniventia, anth. infra med. affixse
suberecte incumb. Stylus declinatus.

36. Cooperia ; tubus cylindr. erectus, limbus regularis (nisi

defectu), stellatus, stylus erectus, fil. brevia fauci in-
serta, anth. tertia parte infer, affixse recurvatim erectae
fasciculatse.

37. Pyrolirion ; tubus cylindr. erectus, limbus campanulate

reflexus, fil. recta patentia, stylus subdeclinatus.

38. Haylockia ; tub. cylindr. erectus, limbus semipatens, fil.

conniventia brevia laciniis inserta, sepalina profundius,
anth. medio affixse versatiles, stylus erectus, gerrnen
subterraneum.

§3. Oporanth for tries ; scapus solidus (semina testacea ?)
Uniflorse.

39. Gethyllis ; tubus longus cylindr. stylo inf. consolidatus,

fil. tubi fauci inserta, interdum catervatim superflua,
gerrnen subterraneum.

40. Sternebergia ; tubus cylindr. stylus liber, gerrnen subter-

raneum.

41. Oporanthus ; tubus brevis subinfundibuliformis, filam.

aequaliter tubo inserta.

Scap. pluriflorus, (solidus?)

42. Lapiedra ; per. profunde fissum, semina angularia nitida.
$ 4 . P oner at formes ; corona staminifera. — 1 . semina testa nigra.

43. Tapeinanthus ; per. prof, fissum, corona brevis.

44. Chlidanthus ; tubus cylindr. limb, semipatens, fil. alis

membranaceis perianthio adnatis. [semina ignota.]

384

A MARY LLTDACE-iE.

45. Clinanthus ; tubus cylindr. limb, semipatens, fil. basi

alata libera [inter se connexa? semina ignota].

46. JJrceolina ; per. pendulum tubo cylindr. recto, limb, ven-

tricose campanulato. [fil basi connexa ?]

47. Leper'iza ; per. pendulum tubo brevi subcylindr. recto,

limbo anguste campanulato, corona brevis.

48. Ccirpodetes ; per. tubo curvato cylindr. limbo subcampa-

nulato, corona brevis.

49. Coburghia ; per. tubo longo curvato cylindr. superne am-

pliato, limbo brevi regulari semipatente, corona cylin-
drica.

50. Stenomesson ; per. tubo suberecto, media parte constricto,

superne ventricoso, limbo brevi regulari, corona brevis.

51. Eucrosia ; per. tubo declinato, limbo compresso recurvo,

corona declinata rutelliformis.

52. Elisena ; per. tubo cylindr. recto, limbo patulo reflexo,

corona cylindr. declinata margine repando.

53. Pancratium ; per. tubo cylindr. limbo patente, corona

conspicua, anther® breves suberecte incumbentes.

2. Semina carnosa viridia.

54. Hymenocallis ; per. tubo recto cylindr. limbo flaccide re-

flexo, fil. subconniventia, anther® long® versat. infra
med. affix® apice inverso pendul®, semina oblonga.

55. Chorctis ; per. tubo subrecto cylindr. limbo reflexulo, fil.

erecta conniventia, anth. long® versat. callositate
quadam supra medium affix® pendul®, semina subro-
tunda.

56. Ismene ; per. tubo curvato, fil. conniventia deflexa, anth.

long® inf. med. affi pendul®, semina rotunda.

— Callithauma ? bulbus cylindr. oblong, per. limbo patente,
corona pr®grandis, stamina exserta.

57. Calostemma ; per. tubo cylindr. limbo semipat. recto,

corona interdum fissa, fil. recta, anth. breves erect®,
germ, dissep. obsoletis s®piss. dispermum.

58. Vagdria ; per. tubo cylindr. limbo substellato, corona im-

perfecta sexfida. [semina ignota ; testa nigra ? genusne
post Pancratium locandum ?]

59. Eurycles ; per. tubo cylindr. limbo semipatente, fil. recta,

anth. erect®, germ, dissep. imperfectis, loculis dispermis.

§ 5. Amaryllidiformes; scapus solidus, semina carnosa,
fil. libera.

60. Grijfinia; per. tubo declin. cylindr. limbo reflexo, laci-

niis inferis divaricatis, ima porrecta.

CHARACTERES BREVIORES. 385

61. Lycoris ; per. tubo declin. ampliato recurvo, limbo surs.

curvato.

62. Clivia; per. pendulum, lacin. quater disparibus, limbi

superficie super, incurva, infer, recta, peric. evalvi, in-
teg. medio pulpaceo.

63. Hoemanthus ; per. tubo recto limbo regulari, peric. evalvi,

integ. medio pulpaceo.

64. Biiphane ; per. tubo recto limbo regul. caps, sicca trival.

turbinate trigona, ovula funiculo affixa.

65. Ammocharis; per. tubo recto limbo regul. caps, oblonga

non pulpacea, ovula placentae adnata.

66. Crinum ; per. tubo subrecto cylindr. limbo subregul. fil.

longa flaccida, anth. long, versat. germ. med. crassius,
peric. dissep. obsoletis uniloculare, sem. difformia.

67. Amaryllis ; per. tubo anguste infundib. fil. petalina

petalorum basi, sepalina profundius inserta, caps, ob-
longa.

68. Brunsvigia ; per. vix tubatum sursum curvatum genit.

recurva, caps, turbin. trigona.

69. Nerine ; per. basi annulari limbo reflexo, fil. pariter basi

monadelpha adnata, stylus filiformis, anth. incumb.
versatiles.

70. Strumaria ; per. basi annul, limbo r eg. subpat. fil. basi

plus minus stylo strumoso adnata.

71. Hessea ; per. tubatum limbo reg. subpat. fil. subul.

aequal. reflexa, stylus filiformis, anth. breves erectse.

72. Imhofia ; per. prof, fissum patens, stylus strumos. fil.

disco inserta.

73. Carpolyza ; per. tubo brevi infundib. limbo semipat. fil.

tubo adnata, stylus crassus sulcatus.

Subordo 5. Narcissta ; scapaceae, schistandrae, non opercu-
latae, corona filam. includente.

74. Corbularia ; fil. declin. recurva, anth. breves med. aff.

incumb. versat. corona tubo longior.

75. Ajax ; fil. recta, anth. longae lineares subrectae infr.

med. affix, filamenta marginibus retro-concurrentibus
amplexae, corona tubo longior.

76. Queltia ; limbus subpatens, fil. recta, anth. longae li-

neares reflexae infr. med. aff. filamenta non amplexae,
marginibus superne retro-concurrentibus, corona tubo
brevior.

77. Ganymedes ; limbus retroflexus, fil. recta, anth. acute

oblongo-ovales med. circiter aff. erectae reflexulae, mar-

2 c

386

AMARYLLIDACEA3.

ginibus vix apicem versus retro-concurrentibus, corona
tubo vix aut non longior.

78. Narcissus ; fil. apice libera, anth. subsessiles subovales

infr. med. aft’, erectae reflexae marginibus retroflexis
non concurrentibus, corona tubo cpiater fere brevior.

79. Hermione ; fil. apice libera, anth. subsessiles acute ovales

medio aff. incumb. versatiles marginibus non retroflexis.
Subordo 6. Galanthece ; scapaceae, porandrae, non operculatae.
Vide p. 80.

EXPLANATION OF PLATES.

PLATE I.

Fig. 1. Alstrcemeria Hookeriana, germen with sepaline rib taken
off. 2. Do. with petaline rib taken off. 3. Alstr. aurea, valve of
capsule. 4. Alst. Hookeriana summit of germen, after the fall of
the perianth, magnified. 5. Do. germen side view. 6. Alst. aurea,
germen. 7. Do. magnified ; a sepaline scars, b petaline. 8. Alst.
aurea, seed magnified. 9. natural size. 10. capsule. 11. pedun-
cle and ribs adhering. 12. valve and ripe seeds. 13. portion of
the axis on which the seeds were borne. 14. same with the seeds.
15. Alstr. Simsiana, seeds. 16. Do. capsule. 17. petaline filament
of Als. Hookeriana. 18. Bomarea hirtella, ripe capsule. 19. Do.
more open. 20. Bomarea acutifolia, germen, and tripartible style
separated. 21. 22. Bom. hirtella, germen with sepaline rib taken
off. 23. dry seeds of Bom. hirtella with the pulpy coat taken off.
24. blunt end magnified. 25. two fresh seeds with pulpy coat on„
26. one of the same seeds, the umbilical chord being pulled from
the raphe to shew its real or ultimate attachment. 27. same seed
with the pulpy coat taken off. 28. The end of do. opposite the
point of attachment, being the internal chalaza. 29. Barbacenia
purpurea, seeds magnified. 30. Curculigo orchioides, seeds, one of
them showing the attachment. 31. same magnified: 32. stamen

of Hypoxis sobolifera. 33. Do. germen, shewing one cell. 34. Do.
style and opercle. 35. Do. germen magnified, a valve and dissepi-
ment being removed, the stigma tripartible. 36. another species,
germen and style. 37. H. sobolifera, germen and tripartible style.
38. transverse section. 39. other species, germen and style magni-
fied, valve and dissepiment removed. 40. style and opercle. 41.
The two species, ripe capsules. 42. Five seeds of H. sobolifera,
nat. size and magnified. 43. Tamus, one of the internal valves of
the capsule. 44. seeds. 45. Conostylis aculeata, capsule magnified
from Bot. Mag. vol. 57. 2989. 46. Do. Germen and style magni-

fied. 47. Alstrcemeria tenuifolia. Herb. Hooker. Gouan. colour
very dark, apparently deep coppery red with pale edges. 48. Alstr.
Preslana. Albiflora Prezl. from Prezl. 49. Alstr. spathulata from
Prezl. Reliq. Hank. t. 122. 50. Leaf of Alstr. spathulata var.

Herb. Hooker, los ojos de agua. Bridges. 51. Alstr. spathulata
Herb. Hooker. Curbran. 52. Vallota purpurea, seed.

PLATE II.

1. Alstroemeria inodora. 2. upper sepal. 3. lateral sepal. 4.
lower petal. 5. lateral petal. 6. Alstroemeria caryophyllsea var.

2 c 2

388

EXPLANATION OF PLATES.

2. Mrs. Graham. Rio. 7. upper sepal. 8. lateral petal. 9. Alstr.
Quillotensis, Herb. Hooker. 10. sepal. Herb. Lindl. 11. petal.
12. stamens ib.

PLATE III.

1. Alstr. angustifolia, var. conferta. 2. Alstr. recumbens. Do.
petal, sepal, and filaments. Herb. Lindl. 3. Bomarea bracteata.
Ruiz, specimen. Herb. Lamb. 4. Alstr. peregrina, var. albescens.
Herb. Hooker. Gumming 568 Valparaiso; petal, and leaf.

PLATE IV.

1. Alstr. pulchella. Herb. Banks. Brazil. Spec, gathered by Sir
J . Banks, name in Dryander’s writing. 2. Alstr. angustifolia, var.
Solliana. Herb. Lindl. 3. Do. var. acuminata Herb. Hooker, ver-
sicolor, Martius. 4. Bomarea macrocarpa Matthews 483. Herb.
Hooker.

PLATE V.

1. Alstr. subrosulacea. Macrae, Valparaiso. Herb. Lindl. fertile
stem. 2. sterile stem. 3. Bomarea Lutea. Herb. Hooker. 4.
petal. 5. sepal.

PLATE VI.

1. Bomarea superba. Herb. Hook. 2. sepal. 3. petal. 4.
Alstr. Isabellana. Herb. Hook. 5. lateral sepal. 6. lateral petal.

PLATE VII.

1. Collania dulcis, immature fruit and style. Herb. Hook.
Matthews 864. 2. four-flowered specimen and tuber. 3. Collania

dulcis Herb. Lindl. 4. 5. 6. Do. Cruikshanks. Herb. Lindl. 7.
outline with fruit by Matthews. 8. Collania dulcis. Herb. Hooker.
9. Alstr. revoluta. Ruiz specim. Herb. Lamb.

PLATE VIII.

1. Collania Andimarcana Herb. Hooker. 2. sepal (orange?)

3. petal paler tipped with green. 4. Alstr. pygmaea. Herb. Hooker.
5. stamen. 6. style. 7 — 13. Matthews’s outlines, seemingly too
large. 8. Do. stamen. 9. germen shewing the ovules. 10. entire
plant. 11. flower. 12. petal. 13. sepal.

PLATE IX.

Collania involucrosa. Herb. Hooker. 1. sepal. 2. petal. 3.
germen in an advanced state, and style. 3. 4,. germen and style,
Matthews’ outline. 5. stamen. 6. stamen immature. 7. point of
sepal, Matthews’ seemingly too large. 8. point of petal do. 9.
Coll, involucrosa. Herb. Lindl.

PLATE X.

1. Collania glaucescens. Herb. Hooker. Paramo of Cayambe.
2. with immature fruit. 3. Bomarea Halliana. Patacocha. Herb.
Hooker. 4. Bomarea cornigera, var. Matth. 1659. Herb. Lindl.

4. one-flowered peduncles.

EXPLANATION OF PLATES.

389

PLATE XI.

1. Collania puberula. Herb. Hooker. 2. var. acicularis.
PLATE XII.

1. Sphserine secundifolia, Ruiz specim. Herb. Lamb. 2. fruit
from Flor. Peruv. 3. Sphserine distychophylla. Ruiz specim. Herb.
Lamb. 4. with immature fruit, ib. 5. Bravoa geminiflora. Herb.
Lindl. 6. leaf. ib.

PLATE XIII.

1. Sphserine, nervosa. Matth. 1661. Second specim. Herb.
Hooker. 2. Bomarea ovata, leaf.

PLATE XIV.

1. Bomarea pardina. 2. Bom. obovata; petal and sepal. 3.
Bom. Patacocensis ; petal and sepal. Herb. Hooker. 4. Bom. for-
mosissima. Ruiz specim. Herb- Lamb.

PLATE XV.

1. Bom. glomerata. Herb. Hooker. 2. Bom. ovata var. nobilis
and leaf. Herb. Hooker. 3. var. Tatiana. Mexico, ib. 4. Bom.
crinita, and leaf. Herb. Hooker. 5. Bom. simplex, and leaf. Herb.
Hooker.

PLATE XVI.

1. Sphserine coccinea from Flor. Peruv. 2. Bom. Caraccensis,
and leaf. Caraccas, Fanning, Herb. Lamb. 3. Bom. Fanningiana
and leaf. Herb. Lamb. Caraccas. 4. Bom. salsilla var. princeps and
leaf. Valparaiso, Cumming. Herb. Hooker. 5. Bom. salsilla and
leaf. Cruikshank 33. Chili. Herb. Hooker.

PLATE XVII.

1. Bomarea dispar. Herb. Hooker. 2. Bom. cornigera Matth. 1659.
Herb. Hooker. 3. Do. broader leaves. 4. Bom. cornuta. Matth. 1161.
misnamed anceps. Herb. Hooker. 5. Bom. edulis. var. Maranensis.
Lindl.

PLATE XVIII.

1. Sphserine brevis. Herb. Hooker. Do. sepal horned, side views.
2. Bom. Cumbrensis. Jamieson. Herb. Hooker. 3. Bomarea fim-
briata var. 4. Bom. tomentosa.

PLATE XIX.

Ixiolirion Tataricum. Altai. Spec. Herb. Lamb.

PLATE XX.

L Ixiolirion Tataricum var. Scythica. Specim. Pallas. Herb.
Lindl. 2. Ixiolirion montanum. Aleppo. Herb. Banks.

PLATE XXI.

1. Hippeastrum barbatum flower, and point of leaf. Herb. Linn.
2. Oporanthus luteus flower and fragment of leaf reversed ; a

390

EXPLANATION OF PLATES.

damaged specimen from Kew garden, labelled erroneously Amaryllis
pumilio. Herb. Banks. 3. Oporanthus luteus from a better pre-
served specimen, marked Am. lutea ; to be compared therewith.
4. Hippeastrum breviflorum spec. Tweedie. Herb. Hooker.

PLATE XXII.

1. BuphanQ guttata leaf, Herb. Linn. 2. Brunsvigia Burchel-
liana, leaf. Herb. Burchell. see B. radulosa p. 281, to be corrected
Burchelliana, for I believe that 1 have the plant, and the green leaf
is not so rasped as it appears when dry. 3. Hippeastrum ambiguum,
(var.) Tweedianum. Herb. Hooker.

PLATE XXIII.

1. Habranthus Bagnoklianus (var.) Gillesianus, umbel and part of
a leaf. Herb. Hooker. 2. Habranthus speciosus umbel, and three
portions of a leaf. Herb. Lindl. 3. Pyrolirion aureum. Dombey’s
spec. Herb. Soc. Linn.

PLATE XXIV.

1. Cooperia Drummondiana var. chlorosolen stamen. 2. Cooperia
Drummondiana. 3. Zephyranthes minima. Herb. Hooker. 4. Ha-
branthus Andersonianus, var. Texana, leaf, inflorescence and stamen.
Herb. Hooker. 5. Cooperia Drummondiana inflorescence and young
leaves. 6. petal. 7. sepal. 8. stigma. 9. mouth of the tube open
and stamens. 10. seed. 11. pollen magnified. 12. Phycella mag-
nifica var. 2. 13. point of the leaf. 14. broadest part thereof.

15. var. 1. Herb. Hooker. 16. Tecophilea violaeflora. 17. The
same two-flowered. 18. Zephyranthes (or Argyropsis) Candida;
banks of the Plata. Herb. Hooker. 19. Cooperia Drummondiana,
capsule.

PLATE XXV.

1. Phycella attenuata (var.) Macraeana ; imperfect umbel and
point of leaf. 2. var. obtusifolia ; flower and point of leaf. 3. var.
latifolia ; flower, point of leaf and broadest part. 4. Phycella gra-
ciliflora ; umbel and point of leaf. 5. Gethyllis undulata, leaf. 6.
Gethyllis verticillata, fruit, bulb and leaves. 7. Gethyllis villosa ;
entire plant.

PLATE XXVI.

1. Habranthus pulcher ; umbel, stamens and style. 2. Habran-
thus miniatus, one flower. 3. Habranthus pedunculosus. 4. Ha-
branthus Andersonianus var. parvula. 5. Urceolina fulva ; umbel
and leaf.

PLATE XXVII.

1 . Clinanthus luteus, two specimens, and inside of the tube with
the stamens. 2. Chlidanthus fragrans.

PLATE XXVIII.

1. Stenomesson flavum ; flower, stamens, and seed. 2. Steno-
messon curvidentatum ; flower and stamens. 3. Stenomesson

EXPLANATION OF PLATES.

391

pauciflorum ; bud, flower, and style, from Hooker, ex. fl. 4. Ste-
nomesson croceum ; flower and capsule. Dombey’s spec. 5. Steno-
messon coccineum ; Ruiz. spec. 6. Stenomesson rubrum ; Ruiz,
specim. 7. Stenomesson breviflorum. Herb. Hooker. 8. The same.
Herb. Lindl.

PLATE XXIX.

1. Zephyranthes gracilis. 2. Zepli. Grahamiana. 3. Zephy-
ranthes Commersoniana. 4. Pyrolirion aureum. 5. Imhofia Bur-
chelliana. 6. Hessea breviflora. 7. Hessea stellaris dissection.
8. Imhofia filifolia, 9. Carpolyza spiralis. 10. Strumaria linguse-
folia. 11. truncata. 12. rubella. 13. undulata. 14. angustifolia.

PLATE XXX.

1. Hsemanthus undulatus. 2. strigosus. 3. brevifolius. 4. Acis
grandiflora. Herb. Banks.

PLATE XXXI.

1. Haemanthus incarnatus. 2. concolor. 3. tigrinus. 4. qua-
drivalvis. 5. sanguineus. 6. coarctatus. 7. moschatus. 8. rotun-
difolius. 9. hyalocarpus. 10. crassipes, 3-10 from Jacquin.

PLATE XXXII.

1. Brunsvigia minor. 2. Amaryllis grandiflora, var. Banksiana.

3. Crinum pusillum.

PLATE XXXIII.

1. Pancratium Illyricum germen. 2. the same after the decay
of the flower. 3. flower and anther and magnified stigma.

4. Hymenocallis speciosa, stigma magnified, and anther. 5. Eury-
cles Amboinensis, stigma and anther magnified. 6. stamen and
segment. 7. section of the flower and germen. 8. Ammocharis
Coranica. 9. stigma magnified. 10. Vagaria parviflora from
Redoute. 11. Gastronema, sections of the flower. 12. Cyrtanthus
obliquus. 13. glaucus. 14. lutescens. 15. Vallota purpurea. 16.
Agave lurida, valve of capsule. 17. stigma magnified. 18. germen
and base of the style. - 19. seed. 20. Fourcroya longseva capsule.
21. style. 22. seed. 23. germen and style. 24. stamen magnified.
25. germen and stamens, 16-25. from Zucarelli. 26. Anther of
Hymenocallis amoena.

PLATE XXXIV.

1. Seed of Zephyranthes Lindleyana. 2. Z. sessilis, var. stri-
atula. 3. Z. mesochloa. 4. Z. sessilis, v. Ackermannia. 5. ripe
capsule and seed of Z. atamasco. 6. germen magnified, ovule greatly
magnified, and stamen of Z. tubispatha. 7. seed of Hippeastrum
Regio-vittatum crossed again by Aulico-striatifolium, natural size.
8. Habranthus spathaceus. 2. Habranthus bifidus. 10. Galan-
thus nivalis, petal. 11. sepal. 12. stamen. 13. style. 14. Eri-
nosma vernum sepal. 15. petal. 16. stamen. 17. style. 18. Leu-
cojum Eestivum, stamen and style. 19. Acis autumnalis, stamen.

392

EXPLANATION OF PLATES.

20. style. Parad. Lond. 21. — 21. Pancratium maritimum, seed
sprouting. 22. seeds of the same. 23. Habranthus Andersonianus,
capsule. 24. Zepli. mesochloa. 25. Pancratium Illyricum, ripe
capsule and seeds. 26. germen. 27. The same at a later period.
28. Hippeastrum pulverulentum, section of the flower shewing the
abbreviation of the tube below, and graduated insertion of the fila-
ments. 29. ovule magnified. 30. germen magnified. 31. style
and anther. 32. cell of Sprekelia formosissima, and ovule magnified.
33. Seed of Ismene calathina and amancaes. 34. Inside of the
flower of Calostemma purpureum. 35. germen and style. 36. inside
of the germen with two ovules. 37. capsule, two views. 38. seed.
39. Eurycles amboinensis, seed. 40. The same. 41. ovule. 42. inside
of the germen a little later. 43. seed dissected. 44. capsule.
45. bulb taken out of the ripe capsule. 46. Hyinenocallis augusta,
germen. 47. Hyinenocallis speciosa, ovule cut open to shew the
embryo magnified. 48. germen. 49. Hyinenocallis litoralis, shewing
two cells. 50. Hymen. Caribea seed. 51. H. litoralis, seed cut
open to shew the embryo.

PLATE XXXV.

1. Choretis glauca. 2. Ismene pedunculata. 3. Ismene amancaes
stamen and germen. 4. Hyinenocallis speciosa, stamen. 5. Zepliy-
ranthes Lindleyana.

PLATE XXXVI.

1. Brunsvigia multiflora, capsule and seed sprouting. 2. Ama-
ryllis Josephiniana, capsule and seed. 3. Nerine lucida, flower,
section of flower, stamen, stigma, and unripe seed. 4. Nerine cur-
vifolia, from a dried flower for comparison of the structure. 5. Lycoris
aurea, tube, and magnified stigma. 6. Clivia nobilis, dissections.
7. Bupliane toxicaria. 8. ciliaris, flower and half ripe capsule.
9. germen and tube of Amaryllis Josephiniana. 10. magnified
stigma of Am. blanda. 11. Zephyranthes tubispatha, stigma.
12. Bomarea acutifolia, capsule and seeds ; a marks the foramen
and umbilicus contiguous to it, b the chalaza. 13. seeds of the
same magnified ; the lower represents the seed with the pulp-coat
taken off, the cord and raphe detached, but remaining in their posi-
tion ; on one side the umbilicus more highly magnified ; the foramen
on the protuberance, the umbilicus at its foot. 14. Zephyranthes
carinata, stamen. 15. Z. Lindleyana. 16. Cooperia Drummondiana,
stamen and stigma. 17. Habranthus, stamen.

PLATE XXXVII.

I. Hermione Corcyrensis. 2. Narcissus Verbanensis.

PLATE XXXVIII.

I. Ajax Pseudonarcissus, stamen before and after maturity.
2. Queltia incomparabilis. 3. Narcissus poeticus grandiflorus fila-
ment and anther. 4. gracilis. 5. patellaris. 6. tenuior. 7. an-
gqstifolius. 8. Hermione grand citronier stamen. 9. compressa.

EXPLANATION OF PLATES.

393

10. bifrons. 11. "primulina. 12. Tazetta before maturity, with
section of the tube. 13. Corbularia bulbocodium. 14. N. majalis.
15. Ajax minor ovules. 16. Sulphureus. 17. Telamonius. 18. tubae-
florus. 19. lorifolius. 20. Sabini. 21. Narcissus angustifolius.
22. tenuior. 23. Queltia incomparabilis greatly magnified. 24. Jon-
quilla. 25. triloba. 26. Calathina. 27. more advanced. 28. ori-
entalis. 29. Hermione, grand citronier. 30. before impregnation.
31. Corbularia bulbocodium. 32. Ajax minor. 33. propinquus.
34. tubaeflorus. 35. crenulatus. 36. bicolor anceps. 37. Pseudo-
narcissus. 38. Bicolor lorifolius. 39. Bic. breviflos. 40. Pseu-
donarcissus Andersonianus. 41. Sabini. 42. N. angustifolius.
43. Queltia incomparabilis unripe. 44. calathina. 45. Corbularia
B. 46. Ajax moschatus. 47. Hermione Tazetta, ripe and unripe.
48. bifrons unripe. 49. Ajax propinquus unripe. 50. minor.
51. pseudonarcissus. 52. Hermione ripe seed. 53. Ajax moschatus.
54. Corbularia conspicua. 55. Narc. recurvus. 56. Ganymedes
pulchellus.

PLATE XXXIX.

1. Queltia Macleaana. 2. orientalis var. 3. orientalis. 4.
montana. 5. foetida v. aurantia. 6. var. semipartita. 7. var.

grisea. 8. odora v. calathina B.M. 9. Calathina minor. 10. in-
terjecta. 11. rugulosa. 12. heminalis. 13. triloba. 14. laeta
(Curtisiana Haw.) 15. Jonquilla. 16. Ganymedes pulchellus. 17.
stamen before maturity twice magnified. 18. Q. jonquilla do. 19.
incomparabilis. 20. Ajax pseudonarcissus. 21. Queltia juncifolia,
Herb. Bentham. 21. Qu. pusilla, Herb. Banks. 23. Ajax pumilus,
Salisbury, Herb. Banks. 24. Cuneifolius, Salisb. Herb. Banks.

PLATE XL.

1. Narcissus recurvus. 2. Majalis. 3. patellaris. 4. Ajax
pseudo-narcissus.

PLATE XLI.

1. Narcissus angustifolius. 2. tenuior. 3. planicorona. 4.
gracilis. 5 — 14. Hermione brevistyla varieties. 15 — 23. Hermione
Tazetta varieties. 24 — 25. Herm. Italica. 26. Queltia pumila from
Eedoute. 27. Hermione elegans from Desfontaines Flor. Atl. from
a many-flowered umbel. 28. Tangiers spec, tube, cup, and sepal,
Herb. Benth. perhaps obsoleta. 29. Herm. Serotina, tube, cup, and
sepal. Sardinia Herb. Benth. 30. Knots on a scape thereof. 31.
Outline from Desfontaines verified by a Sardinian specimen. 32.
Choretis glauca seed. 33. Hilum thereof. 34. Choretis Galvesto-
nensis. Herb. Hooker. 35. anther thereof. 36. part of a leaf
of Drummond’s bulb from Texas, supposed to be Chor. Galvesto-
nensis.

PLATE XLII.

1. Pancratium Cambayense. Specim. Hove. Herb. Banks. 2.
longiflorum. Herb. Banks. 3. Cooperia pedunculata. Mr. M’Nab’s
outline. 4. Dissection of C. pedunculata, twice the natural length and

394

EXPLANATION OF PLATES.

breadth. 5. Dr. Graham’s dry specimen, anthers and style seen
through. 6. Tangiers two-flowered autumnal Hermione. Herb.
Benth. perhaps obsoleta. 7. Pane, maritimum anther, Herb. Banks.

8. Pane. Carolinianum, ib. 9. Hymenocallis pedalis and expansa
anther. 10. perianth and stamens of Lapiedra Placiana from Barre-
lius. 11. leaf thereof from the same.

PLATE XLIII.

1. Queltia juncifolia. Herb. Bentham. 2. Queltia pusilla.
Masson. Herb. Banks. 3. Ajax festalis. Salisbury. Herb. Banks.
4. Ajax pumilus. Salisb. Herb. Banks. 5. Ajax cuneifolius.
Salisb. Herb. Banks. 6. Hermione dubia, 2 flowers, Herb. Benth.
7. Do. 3 flowers, var. Herb. Benth. 8. Do. var. Herb. Benth.

9. Sternebergia citrina, from Flora Grseca.

PLATE XLIV.

1. Crinum Capense (var.) riparium, germen and ovules. 2. C.
Capense, germen, stigma, and fimbriae thereon. 3. C. scabrum.
4. C. longiflorum, Bot. Beg. 303. 5. C. bracteatum (v.) angustius.

6. amoenum. 7. ensifolium. 8. spectabile, transverse section. 9.
cruentum. 10. Canaliculatum ovules from one cell. 11. flaccidum.
12. erubescens, capsule burst. 13. Broussonetianum ovules. 14.
spectabile. 15. seed thereof. 16. cruentum seed. 17. speciosum:
18. defixum. 19. brevifolium (v.) angustius. 20. Capense. 21.
exaltatum. 22. Ammocharis Coranica ovules. 23. Crinum varia-
ble. 24. Paxtoni. 25. Lycoris aurea, cell and pollen. 26. Val-
lota pollen. 27. Z. Candida. 28. tubispatha. 29 Clivia. 30.
Hgemanthus puniceus. 31. Eurycles Amboinensis. 32. Hymeno-
callis angusta. 33. Pane. Illyricum. 34. Ammoch. Coranica. 35.
Crinum pollen (in water). 36. C. Capense, pollen. 37. canalicu-
latum. 38. scabrum. 39. Hippeastrum equestre. 40. in water.
41. Sprekelia. 42. Leucojum aestivum and Galanthus. 43. Ajax
minor. 44. Hymenocallis speciosa. 45. Choretis glauca. 46.
Galanthus seed. 47. Nierenbergia filifolia seed. 48. Petunia
linearis seed magnified. 49. Pet. nyctanigenseflora do. 50. Salpi-
glossis. 51. Nicotiana Langsdorfiana. 52. Nic. Persica variously
magnified. 53. Raj ania Herb. Banks, fruit nat. size. 54. Rajania
ovata. Herb. Banks.

PLATE XLV.

1. Nerine Mitchamiae, hybrida. N. versicolor Herb, appendix. 2.
N. undulata, seed. 3. curvifolia, seed. 4. Hessea breviflora, leaf.
5. Strumaria Baueriana flower magnified ; Bauer’s sketch. Banks,
library. 6. Amaryllis ochroleuca, Ker. Part of Bauer’s sketch.
Banks, library.

PLATE XLVI.

1. Bomarea lyncina, one flower from a crowded umbel. 2.
aurantiaca, do. 3. Hookeriana, do. 4. densiflora, do. 5. pur-
purea var. amoena, do. 6. Collania dulcis var. parvifolia. 7. Hae-

EXPLANATION OF PLATES. 395

manthus Hookerianus. For the descriptions see Supplemental Obser-
vations.

PLATE XLVII.

1. Habranthus punctatus. 2. Sternebergia Dalmatica. 3.
Oporantbus Fischerianus. 4. Stenomesson coccineum. 5. Coburghia
incarnata var. 6. Coburghia splendens. 7. Hippeastrum mini-
atum. For the descriptions see Supplemental Observations.

PLATE XLVIII.

1. Hermione sequilimba. Rangis Mewahhar of the island of
Malta. Narc. serotinus Zer. FI. Melit. 2. peduncle, germen, section
of the tube, the style, and three of the stamens. 3. cup. 4. stel-
late limb, the tube and cup being cut off. 5. anther. 6. Alstroe-
meria versicolor, specim. Herb. Hooker. Reynolds. Antuco, S. Chili.
7. Hermione papyracea sepal. 8. petal. 9. germen, section of tube,
the style, and three of the stamens. 10. cup. See Suppl. Observ.

SUPPLEMENTAL OBSERVATIONS.

By a lapse in transcribing the copy, which was not ob-
served in correcting the press, in p. 27, 1. 28, I have stated
Azalea to be triandrous by defect, which should have been
printed pentandrous by defect ; and, in p. 39, line 2, by a like
lapse, every Amary llidaceous leaf has been printed, which
should have been every Amaryllidean leaf; that is, every leaf
of the fourth suborder.

Allium Cowani. — The statement (151) that Allium
Cowanium had been found by Drummond in Texas was
founded upon erroneous information from Glasgow, and proves
to be incorrect.

Alluvial Soil. — The soil which I recommend, under
that name for the cultivation of Hippeastrum, Vallota, Ne-
rine, &c., and even Cyrtanthus, is such as naturally inclines
to produce crowfoot when it is in tillage.

Alstrcemeria Cummingiana. — This plant, having been
in flower the whole summer in a border in front of a green-
house, which is protected by a mat in frosty weather, has still
a stem of flowers in November, and another advancing, after
a heavy fall of snow, and some days subsequent frost. This,
and Hookeriana, which are nearly allied, seem to be the
hardiest sorts yet introduced. The capsule of A. pulchra has
a long and stout operculous point ; that of Cummingiana
conforms more nearly with Hookeriana.

Alstrcemeria Isabellana. — Very fine specimens of this
plant have been just received by Sir W. Hooker, from Mr.
Tweedie, who says it is one of the most beautiful flowers of
Rio Grande ; the colour of the flowers orange tipped with
green. His specimens have fuller and closer heads of
flowers, which have preserved their colour, and some of the
leaves are 5 inches long, and ^ an inch wide. They have a
pale or yellowish cartilaginous margin. Folia suberecta,
non resupinata, 2-5 unc. |-2 unc- ^a^a’ mai'gine cartila-
gineo, umb. sub- 16-flora, perianth, aurantiacum apice viridi.
Found amongst the mountain marshes and rough pastures of
Rio Grande. Very abundant near Portalegro.

JZaie. 16.

■

SUPPLEMENTAL OBSERVATIONS.

397

Alstrcemeria Psittacina. — The seed-pod of this species
has only six angles at the summit of the germen and cap-
sule, the subsidiary six being obsolete ; and its seeds are
much less tuberculated. than the other species. These pecu-
liarities probably extend to the section, with a prolonged and
incurved upper sepal.

Alstrcemeria Pygm^a grows only on the highest parts
of the Cordillera, not more than two inches high, in rich
black soil.

I have said inadvertently (p. 103) that the soil for Alstroe-
meria should be light. No general rule, as to earth, can be
applicable to a genus of which some species grow in the clefts
of rocks, and others in rich meadows. I find aurantiaca suc-
ceed well in alluvial soil, and I have no doubt that Hseman-
tha, and many others, will thrive better in the earth of a
rich meadow than in a light compost, if attention is paid to
the drainage. I even suspect that the diseased appearance
which Alst. haemantha (pulchella of the nurserymen) often
presents, is owing to its being planted in a light, peaty com-
post. I have not found it so diseased in loam.

Alstrcemeria, 20. Subrosulacea. — PI. 5. f. 1-2. Herb.
Lindl. Chili. Flower stem with few leaves, the two upper
opposite ; involucral leaves and peduncles three ; peduncles
bracteate, 3-flowered ; flowers 1 \ inch long ; colour evan-
escent ; barren stem subrosulaceous ; leaves resupinate,
about If long, lance-oval 3-16ths wide. Seems to approach
most nearly to Preslana. This description has been omitted
by accident in the body of the work ; it should stand before
Preslana, which should be numbered 21, and the subsequent
numbers altered accordingly.

Alstrcemeria Versicolor. — PI. 48. f. 6. Specim. Herb.
Hooker. Reynolds, near Antuco, S. Chili. This plant ap-
pears to answer exactly the description given by Ruiz of
Versicolor, without any figure or specimen preserved, and
I entertain no doubt of its identity.

Amaryllis. — As if Lamark’s character of Amaryllis, in
which the features assigned are successively contravened by
contradictory alternatives, were not sufficiently unintelligi-
ble, I find that Trattinich (Auswahl von Gartenpflanzen) has
added Ixiolirion, Hsemanthus, Brunsvigia, Gastronema, Stru-
maria, and Sternebergia, to the mass, and subdivided it into
such fanciful sections, that Ixiolirion Tataricum figures with

398

AMARYLLIDACEA3.

some of the Nerines in one, and Ixilirion Montanum, which
is barely distinguishable as a species from Tataricum, with
Hippeastrum regium, in another section. If the genera and
species were thrown into a hat, and drawn out by lot, I should
scarcely anticipate a more unsatisfactory result.

Bomarea. § 1. Pedunc. bi-plurifori. § § Per. suhcequale.

A. folia glabra, fl. circ. semunc. conferti.

1. Salsilla. — v. 2. praecipua. — v. 3. subfalcata.

B. fol. glabra, fl. med. magnit. laxi.

2. edulis. — v. grandis. — v. Maranensis. 3. Caraccensis.

C. fol. glabra, fl. magni, conferti. 4. formosissima.

D. fol. glabra, fl. magni, laxi. 5. pauciflora.

E. fol. pubesc. fl. circ. unciales, conferti. Leaves downy,
flowers near an inch, umbel crowded.

6. Hookeriana. — PI. 46. f. 3. Specim. Herb. Hooker.
Matthews ex prov. Chacapozas Peruviae. Folia circ. 6f unc.
If lata, subtus nervosa, pedunc. 2-2f unc. numerosissima
( 1 00?) minute superne bracteati, bracteae involucrales semunc.
lineares aut filiformes, per. f unc. lac. f unc. latae, petala infra
magis attenuata, (saturate aurantiaca ? sepala rubra ?)
Leaves about 6f inches, If wide, strongly nerved under-
neath, peduncles 2-2f long, secondary peduncles minutely
bracteate, involucral bractes f an inch linear, inner ones fili-
form, perianth segments f wide upwards, petals more at-
tenuated below, seemingly deep orange ; sepals red ?

F. fol. pub. fl. circ. unc. laxi. 7. bracteata. 8. acutifolia.
9. obovata. 10. hirtella. 11. latifolia. 12. cordifolia. 13. gran-
difolia. 14. ovata, v. 2. Tatiana, v. 3. nobilis. 15. macrocarpa.
16. hirsuta. 17. cornuta.

§ § Petala sepalis longiora.

G. fol. glabra, fl. circiter unciales laxi. 18. dispar.

H. fol. glabra, fl. magni, conferti.

19. Lyncina. — PI. 46. f. 2. Spec. herb. Hooker. Mat-
thews ex Zambrabamba Peruviae. Fol. ovata ac. 2-3 unc.
1-lf lata, bract, iuvol. externae unciales, f- unc. latae, internae
parvulae, ped. unc. sepala angusta If unc. longa, petala spat.
If unc. long, (aurantiaca ?) saturate punctata. Leaves ovate,
acute, 2-3 inches long, If wide, involucral bractes 1 inch
long, f- wide, inner small, peduncles an inch long, sepals
narrow, If inch long, petals spatulate, If- long, dark spotted,
seemingly orange. The flower seems allied to formosissima.

I. fol. glabra, fl. circiter unciales, laxi. 20. dispar.

§ 2. Peduncidi uni fori. §§ 1. Per. suboe quale.

SUPPLEMENTAL OBSERVATIONS.

399

K. fol. glabra, fl. circ. semunc. conferti. 21. glomerata.
22. torta. 23. Cumbrensis.

L. fol. glabra, fl. circ. unc. conferti. 24. Fanningiana.
25. fimbriata. 26. floribunda, 27. anceps.

M. fol. glabra, fl. circ. unc. pauci. 28. cornigera. 29.
Halliana.

N. fol. glabra, fl. magni conferti. 30. Superba.

O. fol. pubesc. fl. circ. semunc. conferti. 31. setacea. 32.
tomentosa. 33. denticulata.

P. fol. pubesc. fl. circ. unc. conferti. 34. Bredemeyerana.

35. Rosea.

36. Densiflora. — PI. 46. f. 4. Specim. Herb. Hooker,
Matthews ex Chacapozos Peruviae. Caulis glaber, fol. ovata
acum. subt. pubesc. circ. 3f unc. If lata, ped. numerosissimi,
circ. If unc. ebract. bracteae involuc. exter. ultra-unciales,
semunc. latae vel ultra, intern, filiformes, peri. unc. (ni
fallor, saturate rubrum ?) sepala ang. obov. pet. spatlmlata.
Stalk smooth, leaves ovate, acuminate, downy underneath,
about 3f inches long, If wide; peduncles very numerous,
about If inch long, bracteless ; outer involucral bractes
above an inch long, \ an inch wide, or more ; inner ones
filiform, perianth long (seemingly deep red ?) ; sepals
narrow obovate, petals spatulate.

37. Purpurea. — Var. 2. Amcena. PI. 46. f. 5. Specim.
Herb. Hooker, Matthews ex Chacapozos Peruviae. Fol. 3f
unc. lane, acuta semunc. lata, pedunculi biunciales, parte
sup. minute bract, per. unc. sepalis f- unc. latis (lilacinis ?)
petalis spathulatis vix f unc. latis. Leaves 3f inches long,
lanceolate acute ; f an inch wide ; peduncles 2 inches long,
with a small bracte on the upper part ; perianth ^ths long ;
sepals f- wide ; petals spatulate, scarce f wide (sepals lilac ?).
Var. 3. Guancana. ex Guancas Peruviae. Herb. Hooker.
Another variety from Guancas, with larger leaves, shorter
peduncles, flowers and leaves apparently darker, which
comes nearer to the plant of the Flor. Peruv.

Q. fol. pub. fl. circ. unc. pauci. 38. simplex.

R. fol. pub. fl. magni, pedunculi longi. 39. crinita.

S. fol. pub. fl. magni , conferti. 40. crocea.

41. Aurantiaca. — PI. 46. f. 1. Spec. Herb. Hooker,
Matth. Caulis dense pubesc. superne f unc. diam. petioli
crassi, fol. circ. 5-nnc. If-lf lata, lane, ovata tenuiter acum.
subtus pallida enervia dense tomentosa, bract.-invol. externae
circ. unciales vix semunc. latae, internae tenues, ped. nume-

400

AMARYLLIDACEiE.

rosi 2\ unc. vel breviores, germ. parv. per. lj unc. Sepala
anguste obovata, petala spathulata. Ex mont. alt. bum.
Panahuanea. — Specim. alt. ex Huanacabra, flor. et fol. col. ut
videtur saturatiore. Stalk densely pubescent , 4 0f an inch
diam. upwards ; petioles thick ; leaves about five inches
long, 1 ^ to 1^ wide, lance-ovate, slenderly acuminate, under-
neath pale, nerveless, very thickly pubescent ; outer involu-
cral bractes about an inch long, ^ wide; inner slender f pe-
duncles numerous, 2^ inches, or less ; bracteless, germen
small, perianth 1 4 inch long ; sepals narrow-obovate, petals
spatulate. General colour orange, since it was named auran-
tia by Matthews. High damp mountains ; Panahuanea.
Var. ? with flowers and leaves seemingly darker, from Hua-
nacabra.

§ § 2. per. valde incequali.

T. fol. glabra , fl. magni , conferti. 42. Caldasiana.

U. fol. pub. fl. magni , conferti , 43. pardina. 44. Pata-

cocensis.

V. fol. pub. fl. magni, pauci. 45. lutea.

Coburghia incarnata, var. PI. 47. f. 5. Spec, ex Pe-

ruvia, Matthews Herb. Hooker perianthio circ. triunciali tubo
saturate coccineo limbo viridi. This splendid specimen of
a Coburghia from Peru must be a variety of incarnata, with
the flower shorter, the tube of a more orange red, and the
limb more completely green.

Coburghia splendens. PI. 47. f. 5. Spec. ex. Peruvia,
Matthews Herb. Hooker. (An incarnatae varietas ?) Peri-
anthium coccineum apice virescens, laciniis longioribus acu-
tioribus ; folia carent. It is impossible to decide, without
more perfect knowledge of these plants, whether this splen-
did flower should be considered as var. splendens of incar-
nata, or as distinct. It differs both in colour and the form of
the segments of the limb. The leaves are wanting.

Coburghia fulva. — There is a specimen of Coburghia
fulva from Lima amongst Matthews’s last specimens.

Collania dulcis, var. parvifolia. PI. 46. f. 6. Specim.
Herb Hooker, Matthews. In mont. alt. hum. Portachuela,
Peruviae. This plant is evidently a small-leaved variety of
dulcis, having, perhaps, the petals yellower.

Crinum Haylocki, or Flaccido-bracteatum, hybridum,
p.274. Since the foregoing pages were printed, this very
remarkable plant has flowered for the first time, Novem-
ber, 1836. It merits a particular description, as it now

SUPPLEMENTAL OBSERVATIONS.

401

stands. — Bulb ovate, yellowish, 2| inches diameter; neck of
the bulb prolonged into a cylinder 5 inches long ; leaves
eleven, bright green, lorate, attenuated to an acute point,
with a white smooth margin, the longest 4 feet 4 inches, 16
inches thereof erect, the rest pendulous, 1^ to 1^ wide;
scape green, two-edged, 15 inches high ; spathe withering;
bractes whitish, filiform, near 2f inches long ; flowers nine,
successive; scent strong, somewhat like that of C. Amabile;
peduncles f ths, germen very slender, \ inch long ; tube
green, 2-2^ long ; limb white, 3 inches long, expansion 5^ ;
sepals with green points ; filaments and style purple up-
wards; filaments near 1^ shorter than the limb, fths longer
than the style ; anthers a little shrivelled ; stigma minute,
white; cells with two or three ovules. This plant presents a
striking instance of the filaments, at first approached to each
other and recurved, becoming divergent by the subsequent
great expansion of the flower, and explains the different pos-
ture in the patent and less-expanding species.

Cuinum Herbertianum. — I have quoted from Dr. Wal-
lich that this plant and Zeylanicum flower, in the rainy
season, in ditches, but I suspect some mistake ; for Dr.
Carey mentioned Zeylanicum, latifolium, and speciosum, as
flowering at various seasons, (speciosum almost colourless in
the cold season,) and did not intimate that they had any
aquatic predilections in Bengal, while he particularly pointed
out defixum as the inmate of the wet ditches, and longifolium
of the flooded meadows. I apprehend the plants seen by Dr.
Wallich must have been on the ditch-bank, or at least in a
situation subject only to occasional short submersion. They
require to be quite dry in the winter, but a short inundation
would not be injurious to them in the first vigour of their
growth. It is evident, that in cultivation these bulbs should
be abundantly watered for a short time while their leaves are
growing.

Crinum Lindleyanum. — A specimen of this plant from
Demerara, in Sir W. Hooker’s herbarium, has the leaves acute,
fths wide, 18 inches long, exactly agreeing with some of the
leaves of my plant ; scape a foot, flowers eight, tube 4 inches,
limb about 2^, style equalling the limb, exceeding the fila-
ments near an inch.

Crinum Loddigesianum. — This plant has flowered late
this winter, with a six-flowered scape, the bulb having grown
to twice the bulk it had last year. Its nearest affinity is evi-

2 D

402

AMAKYLLIDACE^E.

dently to C. cruentum, and it should be removed from the
place assigned to it, and stand No. 12, between cruentum
and 13. erubescens. Its spathe this year was cleft, not tubed,
as before.

Crinum Seed. — See the representations of the seeds and
germen of several species, pi. 43. The seed of the tropical
species of Crinum will often lie for a very long time without
vegetating. It may be made to grow immediately, by cut-
ting carefully off a portion of the fleshy mass, so as to ex-
pose the point of the embryo, after which the seed should be
set edgeways in a small pot of earth, just covering the radi-
cle. The operation requires, however, a cautious hand ; for
if the point is cut by the knife, the vitality of the seed is
destroyed, and its direction is uncertain, though more likely
to tend towards the hilum or scar than elsewhere. Small
bits of the flesh should therefore be pulled off with the point
of a knife, rather than cut, till the embryo is discovered.

Eucrosia Bicolor.— The habitation of this plant is de-
termined by a specimen in Sir W. Hooker’s herbar. Jamieson,
1836, on the descent towards Jaguachi ; elevation 1000 feet.

Flued Borders. — The vigour with which mules of the
genus Crinum, and many other plants, grow out of doors
against the front wall of a stove, persuades me that a great
variety of plants might with a little care be cultivated better
in the open ground than under glass, if the border in which
they are to grow were flued underground, and a tarpauling,
or any waterproof covering, placed over them at the times
when it might be requisite to exclude either rain or cold.
The covering might hang on the two sides of a strong longi-
tudinal pole like the two slopes of a roof, and be made to roll
up either with or without a spring. There are many plants
which seem to enjoy a cool atmosphere, but will not flower
or thrive vigorously without the stimulus of heated earth at
the root. Having chosen a situation where a furnace and
boiler could be placed under ground, I would carry the
smoke-flue as far as its heat would extend on one side, and
hot-water or steam-pipes in a different direction, as might be
found convenient, enclosed in a stone or brick flue, to as
great a length as its influence might reach. In such a border
I believe the genus Hedychium, and many others, would
flower perfectly with the assistance of fire in the summer,
requiring nothing in winter but a covering to throw off the
wet ; and the heat might be turned into other pipes for the

. • r

Plate. 17.

SUPPLEMENTAL OBSERVATIONS.

403

advantage of plants which might require the warmth in winter
rather than in the summer. If in front of a wall, a move-
able verandah, which might be either ornamental or made of
thatched hurdles or hurdle-gates, would throw off the wet,
which is the principal cause of injury in winter, for many
shrubs will endure the access of severe frost to the head, if
all wet can be effectually excluded from the base of the stem
and from the root by any sloped heading. Under such a
verandah, with occasional heat to the flue, during the early
summer, and perhaps in severe frost, Amaryllis, Brunsvigia,
Buphane, Nerine, Hsemanthus, and all the allied genera of
African bulbs, as well as the South American, would certainly
succeed better than with any other treatment. I believe that
not only those, but even some of the tropical Crinums, would
succeed better so than in a stove, and probably many shrubs,
which might not be expected to live there. The advantage
of a verandah or pent covering, however rude, on the north
side of a wall, for the protection of half-hardy plants, such as
Camellia Japonica, Asiatic species of Rhododendron, &c. is
not sufficiently known. It is the excitement occasioned by
the access of the sun that makes such plants liable to injury,
and a south aspect, whether in summer or winter, is preju-
dicial to them. I believe that the covering of a pent roof
in a northern aspect, without any flue, is more congenial to
those plants than a greenhouse, with caution to prevent any
heavy rain or snow from being driven upon them by a strong
north wind, which is easily done by hanging mats along in
such an emergency.

Ganymedes Capax, flore pleno, p. 310. From the aspect
of the bulb and young leaf of the plant sent to me under that
name, I am satisfied that it is not a double Ganymedes, but
a small Ajax allied to minor. I shall probably not see the
flower before the publication of these pages.

Habranthus punctatus. PI. 47. f. 1. Specim. Rey-
nolds 54. ex Chili merid. Herb. Hooker. Folia carent ; sca-
pus 2-i unc. biflorus, spatha 2-2|, ped. f-2| unc. per. If tubo
brevi unc. ?) limbo pulcherrime punctato, (color roseus ?
marginem versus pallidior? punctis quam plurimis, (saturate
purpureis ?) The leaves are wanting. Scape 2^ inches, two-
flowered, spathe from 2 to 2^ inches, peduncles from f to 2^
inches, perianth If, tube short, seemingly only ± long, limb
minutely and beautifully speckled with deep red or purple ;
general colour probably rose or purple, paler near the margins.

2 d 2

404

AM AR YI.I.T D ACEiE.

I find a quaere (pratensis ? Poeppig) attached to the specimen,
but that plant has a scarlet flower without any speckling, and
cannot be the same.

Htemanthus Hookerianus. PL 46. f. 7. Spec. C. B.
Sp. Herb. Hooker, absque foliis. Scapus 8| unc. gracilis,
circ. 17-florus, spatha quadrivalvis 1^ unc. valv. lanceol. ker-
mesinis, ped. semunc. vel ultra, tubus non coloratus, limbus

et genitalia kermesini, lacinire lineares ultra-semunc. stylo et
filamentis breviores. Scape 8^ inches, slender, about 17-

flowered, spathe with 4 lanceolate crimson valves, about \ of
an inch wide, li long, peduncles ^-f of an inch, tube green?
or pale, limb, filaments, and style crimson, segments very
slender, shorter than the style which equals the longest fila-
ments. We have no knowledge of the leaves of this remark-
able Haemanthus, which preserves the most vivid crimson
colour in the dry specimen.

Haemanthus. Roseus Link enum. 1. p. 309 is carneus.
Speciosus, Colla Hort. Rip. p. 63. Hydrophilus, Thunb. Mus.
Ups. Longifolius, H. Nymph. Obliquus, Don H. C. Sinuatus
Thunb. Mus. Ups. are obscure and doubtful species which can-
not be identified. Orbicularis is a synonym for rotundifolius.

Hermione aEQUIlimba. — PI. 48. f. 1.5. Narcissus sero-
tinus, Rangis Mewahhar. Zer. FI. Melit. Bulbus 2i unc.
diam. vel minor, folia 3-4 glauca carinata ultra semunc. lata
obtusa, vaginis ad basim tribus albis viridi-lineatis cylindricis
H unc. diamet. scapus circ. pedalis subanceps glaucus
striatus apice subtortuoso 3-7 floras, germen horizon tale \
unc. tubus viridis -|, limbus stellatus subalbus laciniis ^ unc.
latis & line, longis, petalis rotundato-oblongis, sepalis acuti-
oribus, stylus tubum paullulum exsuperans antheris sepalinis
brevior, corona triloba lutea j unc. longa, \ lata subventricosa.
After the account of Hermione elegans, Haworth , or Narcis-
sus serotinus, Desfontaines Flor. Atlant. with a 7-flowered
scape, had been sent to the press, I received bulbs of Zerapha’s
plant, which grows in a sequestered spot on the shore of the
island of Malta, in a crumbling reddish-yellow soil, in com-
pany (as it seems) with Qrnithogalum Arabicum, of which a
small bulb was sent with them accidentally. They were dug-
up for me by the kindness of a friend late in September, and
reached Spofl’orth the first week in November, when they
were immediately potted. The largest of these wild bulbs
measured inches circumference. They had evidently been
in a dormant state when taken up, w ith the remains of the
dead fibres of the last season. Zerapha has certainly made

Flaw. £8.

SUPPLEMENTAL OBSERVATIONS.

405

an error in referring this African plant to the little 1-flowered
(rarely 2-flowered) N. serotinus of Clusius, which grows in
Spain, Naples, and Sardinia ; and that not even as a variety,
but as identical. The latter plant is very small, with slender
leaves, and is said to produce its scape often before their ap-
pearance. It is generally one-flowered, though I have seen
a two-flowered Sardinian specimen ; but there is no record
of its bearing a greater number of flowers. I have never
seen a fresh specimen, but in the dry specimens the cup is
not 1-1 6th of an inch long and about ^th wide, and it is
expressly described by Linnaeus as 6-cleft, nectario brevissimo
sexpartito. The Maltese plant has the cup full ^ long, and
a wide, with only three indentures opposite the sepaline
midribs ; and as the length of the tube and limb is not very
different in the two plants, the proportion of the cup to them is
very dissimilar. I cannot ascertain the length of the style
in serotina ; in asquilimba it just exceeds the tube. The
leaves of sequilimba rise before the flower, from 3 to 4 in
number, glaucous, keeled, above ^ an inch wide, about as
tall as the scape at the time of flowering. Clusius represents
the solitary flower of serotina erect, and it is nearly so in most
of the specimens I have seen. Those of eequilimba are hori-
zontal, with a stellate limb, which is remarkable from the
nearly equal width of the petals and sepals. The sepals of
serotina are very decidedly wider than the petals, and its
leaves but the 16th of an inch wide. In all others that I
have seen the sepals are widest. — Scape glaucous, lined, a
little tortuous; germen horizontal; tube green, -|ths long;
horizontal ; limb whitish, with a green line on the back of each
segment, stellate ; cup ^th long, 3 wide, a little ventricose,
more or less three-lobed ; style exceeding the tube, shorter than
the sepaline filaments; expansion of the limb 1 inch and
l-16th. The leaves are enclosed at the base by three cylindri-
cal sheaths, white, lined with green, of which the outer is
about ^ths of an inch diameter, the inner about an inch long
above ground. This plant flowers in Malta at the end of Oc-
tober, yet is not truly an autumnal flower, but one of the
earliest species in a situation and climate where there is no
winter. One of the bulbs taken up in September, and
placed in the stove at Spoft’orth the first week of November,
opened its first flower on the 19th of December, the others in
the same pot rather later. The strongest bulb, left in a green-
house with a warm flue and free access of air, will not flower
before January ; and it is clear that, if the roots had been set

406

AMARYLLIDACEjE.

in the open ground with sufficient protection from frost, they
would not have flowered before March or April. H. Tazetta
which does not bloom here before April or May, in its native
Maltese pastures produces its flowers in January. I make no
doubt that Corcyrensis planted on the hot shore of Malta,
would flower in October, and I think it belongs to the autum-
nal section. Their autumnal flowering is not a distinct habit,
as it is in some genera. Last year var. 9 of Tazetta from China,
produced three flower-stems in Nov. and the same patch of
bulbs produced three or four more in the spring, frost having
intervened and damaged the first umbels. Till I had seen
aequilimba, I concluded that it must be identical with Des-
fontaines’s 7-flowered African N. serotinus, (Herm. elegans),
but it proves to be quite different, that plant having slender
leaves not ^th of an inch wide, like serotina, and a diminu-
tive cup. The flowers of aequilimba are sweet. This plant
has been as yet discovered in no other spot ; its early flowering
would make it an acquisition for forcing, but perhaps it will
not flower a second season freely in our cloudy climate. It
might answer to the Maltese to cultivate it with manure
in their sunny island for exportation. H. papyracea flowers
in its native situation in Nov. and Dec. and var. 12 of Tazetta
(see below) in Oct.

Hermione bifrons. Tube green, limb and cup always
yellow, style just exceeding ; leaves glossy green. It comes
near to tazetta, but has a very different aspect. Found wild
near Montpelier.

Hermione brevistyla. Tube green, limb white or yel-
low, cup orange or yellow however pale, style much shorter
than the tube, leaves subglaucous. I find this plant in no
herbarium, and its native habitation is not clearly recorded.
Our gardens are filled with fine varieties of it, which are very
hardy. It seems not to be a native of the neighbourhood of
the western portion of the Mediterranean, from whence the
collected specimens are derived, and was probably introduced
into our gardens from Constantinople, which in the days of
Clusius and Parkinson was in the habit of supplying us with
bulbs. These short-styled plants can scarcely be supposed
to have been raised by cultivation from the little H, dubia,
and are evidently not garden varieties of any of the other
species. Some of them were known 200 years ago, and it
seems most likely that they have been derived from the nor-
thern part of the Turkish empire, beyond the walk of our
collectors. Its native situation being unknown, it has con-

SUPPLEMENTAL OBSERVATIONS.

407

sequently been overlooked by botanists, but it assuredly has
a wild locality somewhere.

Hermione elegans. According to Desfontaines its style
is as long as the tube. It inhabits the neighbourhood of Al-
giers.— See above, Hermione sequilimba.

Hermione Italica. Tube green. — v. 2. praecox. Tube f,
limb cup about f, in a specimen from Pr. Tenore.

V. 4. brevis. I find a specimen from Pr. Tenore, under
the name narcissus neglectus, exactly agreeing with this
variety. It is, therefore, not a garden production.

V. 5. aurea.spec.Requien Herb. Hooker. N. aureus, spec,
nov. Tube near f, cup f , limb f ; style a little shorter than
the cup ; umbel about 10-flowered ; leaves half an inch wide.

V. 6. chrysantha. N. chrysanthus, Redoute. Spec.
Herb. Hooker. Tube about |-, limb scarcely, cup about f,
style equalling the cup. Both golden, and closely allied.

V. 7. subalbida. N. subalbidus. Lois. Tube near f,
limb white; cup very pale yellow, f ; style equalling the
cup. It comes near to v. 3.

The double Roman narcissi are cultivated varieties of H.
Italica, with the scape very fistulous. I have two distinct.

Hermione papyracea. — Tube white, style exceeding it.

V. 1. unicolor. Tenore. 1. 26. Capri and Vesuvius.
Tube exceeding §, cup3-16ths, limb 1 l-16ths, sepals 7- 16‘ths
wide, leaves -§ wide, style shorter than the upper anthers.
Subvar. culta. Paper-white, germen f-, tube exceeding f ,
limb cup 3-16ths, style shorter than the upper anthers,
sepals f- wide, leaves glaucous, channelled, keeled, near fths
of an inch wide. — PI. 48. f. 7 — 10. I had not had a late
opportunity of examining the flower of this plant before
the genus went to the press, but having forced it into
flower, I find it distinguished from Tazetta by the small-
ness of its cup, which is not larger than that of aequilimba,
viz. 3-16ths of an inch long and \ wide, crenulate. It is
distinguishable from the rest of the genus, unless perhaps
dubia, which I have not seen in a fresh state, by the tube
being scarcely green. The segments of the limb are more
lanceolate than usual. They are more than four times the
length of the cup, and exceed the tube, which is fths of
an inch long. Folia semunc. glauca, canaliculata, cari-
nata, obtusa, scapus anceps, germen || unc. tubus f albo-
virescens, limbus stellatus laciniis lanceolatis unc. sepalis f-
unc. latis petalis albis, corona alba unc. crenulata, f unc.
lata, stylus tubum exsuperans, antheris sepalinis brevior.

408

AMARYLLIDACEA2.

Var. 2. Nivea. Loiseleur. Spec. Herb. Hooker. Tangiers.
Germen 5-16ths, tube slenderer than unicolor fths, cup
3-16ths, limb 13-1 Gtlis, sepals 3-16ths wide, style equalling
the cup, leaves 5-16ths wide, or less, umbel 13-flowered.

The variety cultivated in Italy has the leaf and style of
the native, but approaches more to the African variety in the
proportions of the flower. The peduncles in all are shorter
than the spathe.

Hermione Serotina. — I have ascertained by two speci-
mens, one from stony places at the foot of the hills near Ar-
gostoli, one from the hills near Cagliari, that the posture of
this flower is not really erect, as represented in all the en-
gravings we have of it, in consequence of their having been
made from specimens ill dried. The flower will become erect
when the seed sets. It does not always precede the leaves ;
in one Neapolitan specimen I find a leaf as long as the scape,
but scarcely a line wide. There is in Sir W. Hooker's herba-
rium a one-flowerecl specimen from Tangiers, marked sero-
tinus, which is evidently the same species or variety as the
two-flowered specimen represented PI. 42. f. 6. but I cannot
ascertain what its colour has been. The bulb is twice as large
as those of serotina. The union of aequilimba with serotina
in the FI. Melit. is quite erroneous. 1 lament that I cannot
ascertain the precise structure of the stamens of serotina and
viridiflora, which may be found to conform with the jonquill.

Hermione Tazetta. — Tube green, limb yellow or white,
cup orange or yellow, however pale ; stigma between the
upper anthers ; leaves glaucous.

V. 12. Specimens Herb. Hooker. Lady Dalhousie, near
Syrie. Oct. 10. both 4-flowered, peduncles long, tube cup

limb about 11-1 6ths, seemingly very pale yellow.

V. 13. Spec. Herb. Hooker; in moist places. Algiers. Ja-
nuary. Umbel 8-flowered, peduncles short, tube 9-lGths, limb
7-1 Gtlis, light yellow; cup 3-1 Gtlis, darker; leavesabout l wide.

V. 14. Spec. Herb. Hooker; in upland pastures near Ca-
gliari, Sardinia. Umbel 5-flowered, ped. short, tube limb ^
inch, seeming pale yellow (or white ?) cup i, leaves about
^ths wide.

I was surprised at finding the bulbs sent to me of the
native Maltese Tazetta, said to be cupularis or soleil d’or, not
so wide as a common walnut, though rather longer. The
succession of the species would stand more properly 3. bre-
vistyla. 4. Italica. 5. papyracea. 6. dubia. 7. Corcyrensis.
8 Aequilimba, 9. elegans. 10. serotina. 11. obsoleta. 12. viri-
diflora.

SUPPLEMENTAL OBSERVATIONS.

409

I have stated the cup of H. Tazetta to be yellow or orange,
meaning to include pale lemon under that definition, and to
exclude white, of which I know no instance. The var. 8. is
perhaps the Grand monarque. If I had certain information
concerning each of the Narcissean varieties, whether it is a
garden production or a wild local variety, 1 should give or
retain a subordinate Latin name to each of the latter descrip-
tion, and consign those of the former to the catalogue of
florist’s names.

Hippeastrum Acuminatum, y- foliis variegatis, R. et
Schultes v. 7. 2. p. 1 720 is doubtless a cross-bred plant, having
one quarter of the type of reticulatum by descent. I have
raised similar plants, but do not believe any striped acumi-
natum has been found wild, or raised in cultivation without
such intermixture.

Hippeastrum Barbatum. — This plant was called Ama-
ryllis dubia by Linnaeus (Amcen.), not intending dubia to be
its specific name, but because he was in doubt about the
plant. It is a remarkable instance of the very vague ideas
then entertained of generic characteristics, that, after having
referred the plant to Merian’s Hippeastrum equestre, from
which it is principally distinguished by its white instead of
orange limb, he should at last have called it in his herbarium
a Crinum, for no other reason than its agreeing with those
he knew in colour. The plant has been noticed by no other
writer, and has never been brought to Europe.

Hippeastrum Griffini. Amaryllis psittacina hybrida
(Bot. Mag. 61. 3528) is the same plant.

Hippeastrum Miniatum. — PI. 47. f. 7. Spec. herb.
Hooker. Matthews, Quebrada of Panahuanca. Peru. Folia
22-unc. acuta, superne unc. lata, inferne valde attenuata,
scapus biflorus7-10 unc. spatlia If, ped. lj-lf, per. 3|-4 unc.
cernuum, stylus limbum sequans, staminibus longior,
stigma obtusum triquetrum. I cannot doubt these Peruvian
specimens of a deep red Hippeastrum being the Am. miniata
of Ruiz. He says filaments, as well as style, equal to the
perianth, which is very improbable. The lateral petals are
about 1|- wide, the lowest narrow.

Hymenocallis adnata, v. princeps. — Jacquin’s plant,
said (as I suspect erroneously) to have come from Tierra
Bomba has the tube eight inches long, the cup eroded, one-
third the length of the limb, filaments not exceeding the
limb; leaves 3-4 feet long, an inch and half wide. The
seedlings of this plant, which I mentioned p. 216 as thriving

410

AMARYLLIDACE-E.

after they had been plunged in a cold cistern of water, in the
coolest part of the stove, are still (in the month of December,
1836) thriving rapidly in that situation, where it appears
that they might remain permanently, like v. acutifolia. A
pot of bulbs of v. Staplesiana which had been set under
water in a pond at SpofForth last summer has had the points
of the leaves above water cut by the frost, but the lower part
and the bulbs are uninjured. Four bulbs of v. acutifolia
were turned out into a border of peaty compost at the begin-
ning of May. They were taken up in November, without
disturbing the ball, to be kept dry throughout the winter,
and were found to have quadrupled their bulk, though the
season was so unfavourable that the oat harvest was unfinished
at the beginning of December in the neighbourhood. I am
satisfied that the reason why the mule from rotata by adnata
v. princeps has never yet brought to perfection its flowers,
which have invariably perished in the bud, is that it requires
at that period immersion in water. The mule has usually
produced about six abortive buds. I shall be disappointed
if I do not see its inflorescence completed next season by
treating it as a swamp plant at the season of flowering. It
is very probable that Hym. crassifolia may, as well as some
oriental species of Crinum, require a like increase of wet to
promote their blossom.

Kew Garden. — When I spoke (p. 247) of the evil con-
sequences which flowed from the bad system pursued there,
tending to loosen the ties of morality, and to create a feeling
of satisfaction when it was known that cuttings had been
stolen from the large plants hoarded there, by which the
public were enriched without any perceivable loss to the col-
lection, I should perhaps have stated for fear of misconstruc-
tion, that, far from justifying, I much lamented such a feel-
ing ; and I observe that I have incautiously admitted a
stronger expression concerning the unpopularity of the prin-
ciple on which that garden has been conducted, than I should
wish to have used.

Lapiedra. — The generic name of this plant, which grows
in stony situations, appears to be singularly cross-bred be-
tween the Latin and Spanish words for a stone, lapis and
piedra. The name sounds well, and we must overlook its
awkward parentage, but such an amalgamation should not be
drawn into a precedent.

Liriope. — The name Liriope, which I gave fifteen years
ago to a South American genus, appearing to have been long

SUPPLEMENTAL OBSERVATIONS.

411

before applied by Loureiro to the genus Ophiopogon, I have
changed it to Elisena, an ancient name of romance : and, as
the latter genus is only known to us by the description and
engraving of Ruiz, the name Liriope may be abandoned
without much inconvenience to the public. The rule, that
the name first published shall be adopted, was made for the
convenience of the public, to prevent confusion ; but great
inconvenience results from dragging to light names that
have been unnoticed during a long course of years, and
making them supersede those which have been long in vogue
and recognized in systematic works. On that account I
declined altering the name of Z. tubispatha, though certainly
not the original A. tubispatha of LHeritier. The genus
Ophiopogon, well named by Mr. Ker in allusion to the
Chinese appellation, has been acknowledged universally near
thirty years, and I deprecate on principle the substitution at
this late period of one which had been overlooked so long, as
a nuisance to the public : and, unless Loureiro, whose work
I have not at hand to refer to, has defined the genus as well
as named it, and defined it by features that could have been
with certainty recognized, the substitution of his old name
would be irregular as well as inconvenient.

Markers. — Zinc is the only fit ingredient for labels whe-
ther to be used in the open ground or in pots. A sheet of
zinc is easily cut by the gardener with strong scissors into
labels of whatever size he may want. If the zinc is greasy,
the labels should be steeped for a minute or two in diluted
nitric acid. The following receipt for making ink for writing
on the zinc was communicated to me by a gentleman who
was in the habit of using it, and I have found it indelible.
Take Verdigris in powder 3i. Sal Ammoniac ditto 3i. Lamp-
black 3ft. Water 3x- Mix carefully in a mortar. Keep the
ink in a bottle well corked. It must be well shaken before the
pen is dipped in it.

Mule Gladioli. — It should have been mentioned, in
speaking of the cross-bred Gladioli (p. 365), that I obtained
two years ago several seed-pods from G. cardinalis by the
pollen of various choice mules Cardinali-blandus, and Car-
dinali-Tristi-blandus. I had expected the seedlings, having
two-thirds of the type of cardinalis, to lean towards the scarlet
colour and approximate to that plant ; but to my surprise the
approximation (notwithstanding their greater affinity to car-
dinalis) was entirely to the cross-bred mule parent and not
one scarlet flower appeared amongst them. It seems that

412

AMARYLLIDACEiE.

to obtain the scarlet varieties, the first mule must be im-
pregnated by a second cross with the pollen of cardinalis.
The causes are perhaps two ; the disposition of the perianth
to follow preferably the type of the male, and of our climate
to produce the less-brilliantly coloured varieties of plants
which are derived from warmer latitudes.

NarcissejE, p. 79. Observ. add thereto capsula parte
superiore dehiscens ; and p. 293. 1. 7. add the upper portion
of the capsule gaping.

Corbularia bulbocodium and tenuifolia are both found on
the heights called Chambre d’Amour, near Bayonne. Two
varieties of Ajax luteus grow on Mount Hazza, western
Pyrenees.

Natural Orders. — I have perhaps not sufficiently ex-
plained, that, having taken the most decisive feature I could
find to distinguish from each other the monocotyledonous
orders, which I could not pretend to reform where I might
think some reformation necessary, I by no means intended
to assent to the importance attributed to each of those several
features ; for instance, I suspect that on future consideration
Dr. Lindley may find it advisable to make Yanillese a sub-
order or section of Orchidacese, as they seem distinguished
chiefly by a feature which would separate Hippeastrum and
Amaryllis, placing them in distinct orders, if admitted to be
of such importance. It will be understood, that by the in-
sertion within brackets of the names of Dr. Lindley 's alli-
ances in the scheme of monocotyledonous plants (p. 45), I do
not intend to make them a part of that scheme, but merely
to point out where they intervene and how far they coincide,
though in a different consecutive arrangement.

Nerine Haylocki, p. 284. —Seeds by the pollen of this
hybrid plant, produced upon its own scape, and on that of
curvifolia, which had been deprived of its anthers, no other
Nerine having been in flower at the time on the premises,
have vegetated, and amongst the latter, one has produced
the young leaf crimson, an appearance which I have never
seen in the genus, and which will perhaps not be permanent
in the adult plant ; but such a remarkable seminal variation
brings curvifolia in closer affinity with marginata, which is
distinguished by a red margin to the leaf.

Oporanthus Fischerianus. PI. 47. f. 3. — Specim, Dr.
Fischer. Herb. Hooker. “ In collibus graminosis circa cas-
tellum Agh-oglan provinciae Karabogh.” Bulbus ovatus
collo producto sexunc. semunc. lato (subterraneo ?) folia

SUPPLEMENTAL OBSERVATIONS.

413

subpedal, ~ unc. lata obtusa, scapus 9-uncialis (magna parte
subterraneus ?) spatha univalvisj unc. peri, vix lj| (pal-
lide?) luteus, stylus uncialis stam. longior, stigma vix loba-
tum. It is immaterial whether this plant be called O. luteus
Fisclierianus var. or Fischerianus simply. It is very different
from luteus, which has the perianth two inches long, while in
this it is scarcely If, and the colour is seemingly different.
The scape is above nine inches, but, as the bulb has a neck
six inches long and half an inch wide, probably the greater
part of it was under ground.

Queltia. — § 1. Macrantherce. With long anthers. 1. Ma-
cleana. 2. montana. 3. foetida. 4. orientalis. 5. odora.
§ Bracheiantherce . With shorter anthers. 6. jonquilla. 7. jun-
c.ifolia. 8 ? pusilla. 9 ? pumila. The anther of juncifolia
agrees with that of the jonquill. I am confident that pusilla
will conform with it, and I think it most probable that pu-
mila will also. Jonquillioides and rupicola will certainly, if
there be any such distinct plants. I see it suggested by a
collector that Queltia aurantia, which is found wild near
Montpelier, may be perhaps a mule between Narcissus
poeticus and Ajax pseudo-narcissus, growing in the same
pastures, and that narcissus biflorus may have originated in
like manner between poeticus and Tazetta, meaning it seems
the yellow bifrons, which grows in the same vicinity. I
cannot admit the probability of the suggestions.

Stamen.- — It is worthy of observation, that if the bud of a
long-tubed flower, like Hymenocallis, be examined at an
early period, while it is still enclosed in the spathe, the
anther will be found to have acquired its full growth, the
limb being just large enough to contain it, but the tube not
developed, and its rudiments only in existence. The pro-
longation of the tube and peduncle, and the further growth
of the limb, depending upon the nourishment afterwards
afforded to the plant, must be liable to obstruction and
variability. This consideration seems to uphold the view
which I have taken (p. 30), that the anthers are of superior
botanical importance to the perianth. The same observations
tend to shew why the proportion of the limb in Narcissese is
less important than that of the cup.

Sternebergia Colchiciflora v. Dalmatica. — Plate 47.
f. 2. Specim. Dr. Fischer. Herb. Hooker. Or rather Sterne-
bergia Dalmatica, for its aspect and proportions are very
different from that of Colchiciflora, and it is very inferior in
size. Bulb about §ths of an inch diameter, ovate, brown,

414

AMARYLLIDACE.E.

tube exceeding the neck of the bulb §-§ths, limb slenderer,
pale yellow, |-|ths long, style shorter than the limb, a
little exceeding the longest stamens, spathe bifid, filiform.

Stenomesson Coccineum. — PL 47. f. 4. Herb. Hooker.
Matthews 786. All doubts about this plant are set at rest by
fresh specimens received by Sir W. Hooker from Chacopoyas
in Peru. Scapes 4-flowered, perianth scarlet, 1^ inch or a
little more, style | of an inch longer, capsule formed like
that of St. flavum. They agree with Ruiz’s specimen, but
not with the figure in Flor. Peruv. which is incorrect.

Tapeinanthus humilis. — Cav. ic. 129. t. 207. f. 2. This
plant flowers in October. Scape filiform, 3 inches ; cup small,
dentate, or 12-cleft, filaments equal, nearly as long as the
limb ; direction of the tube continuous from the germen and
peduncles ; peduncles unequal ; leaves slender, awl-shaped.

Yallota purpurea. — I have lately had an opportunity of
examining many ripe specimens of the fruit of this plant. The
capsules, which are oblong-oval, begin to gape near the base,
where the valves are attenuated. The seeds, which are 7-16ths
of an inch long, have the lower 5-16ths absolutely empty,
being merely a foliaceous wing attached at the base, where
it is narrow, by a very slender umbilical chord. The chalaza
points obliquely to one side, the other side of the seed having
a curved back. The albumen or kernel lies in the upper
2-16ths transversely, the embryo pointing from the chalaza
to the opposite side of the seed. The seeds which are in the
lower part of the capsule, where it is narrow, cannot develope
themselves into this form, and are narrow and straighter, the
chalaza more vertical, and the kernel and embryo not trans-
verse, but pointing towards the foramen at the base. This is
important to shew how insignificant is the position of the
embryo with relation to the hilum in some cases.

Vellosia abietina. — Martius. fine blue. Omitted by
accident in p. 81.

415

POSTSCRIPT.

Bomarea salsilla. — I mentioned (p. 110) that this plant differed from
the other Bomareas I had seen, in a very decided flexion of the filaments,
but that its half-ripe capsule seemed conformable with theirs. I have
now its ripe seeds before me, and they differ materially from those of hir-
tella and acutifolia. They have no soft pulp, but a tight coat, not easily
separable, but minutely wrinkled, with a very prominent loose wrinkled
raphe and chalaza. We do not possess a sufficient number of Bomareas to
be able to trace out these diversities.

Hymenocallis expansa : var. stricta. Foliis erectioribus superne la-
tioribus, floribus erectioribus laciniis angustis. Leaves above the middle
2§ths of an inch wide, more erect, but equally acute; flowers 11, more
upright; tube 4 inches, limb \ of an inch shorter, just exceeding the style,
li longer than the filaments, segments only ^ths wide, cup fths long, con-
stricted at the base, margin denticulate.

Hymenocallis Caribea, v. prineeps, has the anthers deeper coloured
than v. cinerascens.

The delay which has taken place in the publication of this volume, the
plates not having been ready, has enabled me to clear up some obscure
points concerning the Narcisse^e. The stature of A. nanus, Ham. is more
erect than that of minor, but it has, like v. prineeps and humilior, the
tube \ inch long, the cup Tilths, and the limb shorter than the cup, the style
half an inch shorter. It may be called v. altior ; nanus will not do.
Festalis, Salisb. pi. 43. f. 3. differs only in having a narrower leaf. I have
never had pumilus and cuneifolius ; they may perhaps form a separate
diminutive species. I find unrecorded differences which I had not antici-
pated in the sulphurescent species, and, with me, the tube of tortuosus,
unless there be two varieties of it, scarcely loses the yellow from the end
of the cup. The superior breadth of the leaf of Sabinianus is an additional
proof that it is a genuine species. I arrange the genus thus: 1. Pseudo-
narcissus. Closely allied to minor Corona, lutea limbum pallidiorem
cequans , stylo semunciam longior ; Jilamenta. ( uti in 2 et 3) prope basin
incequaliter adnata. Cup yellow, equal to the paler limb, £ inch longer
than the style ; filaments (as also in 2 and 3) adnate unequally near the
base. 1. Ryticarpus ; 2. nobilis ; 3. Telamonius. Festalis, pi. 43. f. 3.
belongs to minor, on account of the excess of the cup. The other varieties
I have not examined. — 2. Minor. Corona lutea limbum pallidiorem su-
perans, stylo semunc. longior. Cup yellow, exceeding the paler limb,
| inch longer than the style. 1. prineeps ; 2. humilior; minimus, Ham.
Leaves a little shorter, cup deeper, and limb shorter, but the difference
scarcely worth remarking. 3. altior ; nanus , Haw. scape 8 inches ;
leaves less tortuous, fths wide; 4. ? cuneifolius ; o. ? pumilus. — 3. Lu-
teus. Corona lutea limbum concolorem fere parum superans, stylo § unc.
vel ultra longior. Cup yellow, a little exceeding the limb, yellow also,
fths of an inch or more longer than the style. 1. maximus ; 2. major;
3. propinquus ; 4. obvallaris. I have not examined the last. — 3. Tub^e-
florus. Corona lutea limb, pallidiss. cequans , stylo J unc. longior ; Jil.
basi ipsi adnata. Cup yellow, equal to the very pale limb, gth of an inch
longer than the style. 1. prineeps; 2. crenulatus. — 5. Bicolor. Cor.
lutea limbum pallidiss. cequans , stylo f unc. longior ; Jil. prope basin
cequaliter J'ere adnata. Cup yellow, equal to the very pale limb, \ longer
than the style ; filaments adnate near the base almost equally. 1. lori-
folius; 2. breviflos ; 3. an ceps. — 6. Tortuosus. Cor . citrine, colore serius
obsolescente limbum subalbescentem superans, antheris g, stylo | longior ;
Jil. i-A a basi adnata. Cup lemon, fading late and imperfectly to sulphu-

416

POSTSCRIPT.

reous white, longer than the limb, jjths longer than the style, Iths than the
anthers ; filaments adnate, from £ to ^ths of an inch from the base, sepals
Jths, petals jjths wide, tortuous. N. moschatus a Bot. Mag. — 7. Albicans.
Cor. citrina mox subalbescens limburn subalbescentem superans, stylo
unc. anther is 1£, longior ; Jilam. \-rs basi adnata. Cup lemon, turning
whitish soon, longer than the limb, exceeding the style an inch, the
anthers ; filaments adnate as in Tortuosus; sepals fifths, petals 11 wide,
tortuous. — 8. Cernuus. Corona cernua citrina mox subalbescens limburn
subalbescentem superans stylo unciam fere longior ; fil. circ. i unc. a basi
adnata. Cup hanging down, lemon, turning soon to whitish, exceeding
the style near an inch, the anthers near lj; filaments adnate at about gth
from the base, sepals igths, petals 9 wide, tortuous. The filaments of
tortuosus are slenderer, of cernuus thicker, though shorter, than those of
albicans ; the anthers of cernuus a little longer and yellower. I suspect
tortuosus of being a garden cross between albicans and bicolor. — 9. Mos-
chatus. Cor. sulphurea mox albescens limburn album cequans, stylo
5 longior. Cup sulphur, turning to white, longer than the white limb,
exceeding the style 5 of an inch. N. Moschatus S. Bot. Mag. 32. 1300. —
10. Sabinianus. Cor. lutea limburn album superans , stylum cequans;
fil. i-i a basi inceq. adnata. Cup yellow, longer than the white limb,
equalling the style ; filaments adnate unequally | and |ths of an inch from
the base. — Ganymedes. Its anthers are attached nearer the middle than
those of Queltia, and their margins do not come in contact behind.
Striatulus is shorter, stouter, with larger cup and limb than pulchellus,
and its petaline stamens more prolonged ; style long and salient. — Her-
mione. A specimen of wild Hermione, just received from the Hon. Wm.
Fox Strangways, very common near Genoa, exhibits the natural form of
H. brevistyla, which I had found in no herbarium. Flowers 3-4, fragrant ;
tube a little exceeding half an inch, green; limb 3-8ths, white; sepals
much wider than petals, cup deep orange, not 3-16ths long ; style 3-16ths
shorter than the tube ; leaves narrow. Bifrons is the only Hermione
with leaves not tending to glaucous ; no Ajax has green, no Corbularia
glaucous, leaves. — H. Dubia, incorrectly represented in Redoute, has the
style as long as papyracea, of which it is clearly a variety, with shorter
and blunter segments. By a recent examination of a live specimen, I am
satisfied that N. pumilus of Redoute is a weak one-flowered specimen
of H. dubia, var. 1. The flower tallies exactly, and the wild specimens
are often 2-flowered. In cultivation the flowers become more numerous,
and the leaves broader, than in any of the three wild varieties. Herm.
dubia must therefore be transferred to papyracea, as v. 3. dubia, or more
properly Gallica, and its wild variations marked subvar. My opinion
of N. pumilus of Redoute is confirmed by information since received from
Paris, that no trace of it exists there. — One var. of double Hermione is

brevistyla. -Queltia pumila must be expunged as a nonentity. — A fine

2-flowered variety of Queltia odora grows amongst the chesnut trees near
the church of N. Senhora del Monte in Madeira. Var. 8. Isometra ;
germine, tubo, et corona isometris, fere semuncialibus, limbo § unc. corona
stylum -jj superante leviter sexloba. Germen, tube, and cup equal, each
near | inch, limb f , long ; cup exceeding the style l-16th, slightly six-
lobed. The fetid Queltia cannot be, as some have suggested, a cross
between plants of agreeable scent.

The curious mule Rhododendron (see p. 359) from the plant called
Rhodora Canadensis by Az. Pontica is now in flower. The flowers are
of a very pale yellow, with from seven to nine stamens. It will be iigured
in the Bot. Reg.

March 25, 1837.

POSTSCRIPT.

416*

I have to-day received a plant of Pancratium parviflorum of Redoute,
with which I am favoured from the Jardin des Plantes at Paris. The
instant I saw its leaves, I was convinced of its being Lapiedra Placiana.
The leaf exactly agrees with the figure of Barrelius and the description of
Clusius that it resembles that of Oporanthus luteus, with the difference of
having a longitudinal whitish stripe. It is strange that Barrelius and
Lagasca should have entirely overlooked the tendency to a Pancratiform
structure in the stamens, and Redoute have represented the leaf so imper-
fectly, that no person had, or could have, suspected the identity of the
plants. I am confident that the bulb sent to me from Paris must be
Lapiedra. The name Lapiedra Placiana must therefore be removed from
the Oporanthiform section, and substituted for Yagaria parviflora, which
must be expunged. The account of the leaf, seed, and habitation, under
the head Lapiedra, will be correct ; but that of the flower must be taken
from the article Yagaria and the figure of Redoute.

April 2, 1837.

■

»

GLOSSARY.

Acuminate. Having a prolonged point, gradually attenuated.

Adnate. Adhering to laterally.

Albumen. A substance which in its perfect and solid state is ground into flour.
Anthers. The portion of the male part of a flower which contains the pollen or
particles of dust that, by contact with the stigma, fertilize the ovary.

Antitr opous. A seed is so called when the embryo in it is inverted.

Apetaloid. Having no appearance of petals.

Apiculate. Having a prolonged slender point.

Assurgent. Having the ends rising upwards from a horizontal posture.
Auriculate. Having projections like ears.

Articulate. Jointed.

Axillary. Growing out of the axil or angle between the base of a leaf, or its
petiole, and the stalk or branch on which it grows.

Ax/s. The central column in a fruit which supports the style, and connects it
with the footstalk.

Bilabiate. Having the form of two distinct lips, the base of the segments being
united in two separate parcels.

Bracteate. Furnished with bractes.

Bracte. An appendage to the inflorescence, sometimes leaf-like, sometimes fili-
form, from the axil of which a flower-bud usually proceeds. There is no
perfect distinction between bractes and the valves of an involucre or spathe.
A bracte is properly with respect to a flower, that which a stipule is with re-
spect to a leaf.

Calycine. Belonging to, or having the office or semblance of, a calyx.

Calyx. The outer leaf-like envelope of a flower.

Campanulute. Bell-shaped.

Capitulate. Having the flowers in a crowded head.

Caulescent. Having a leaf-bearing or articulate stalk.

Centrifugal inflorescence. Beginning to flower in the centre of the umbel.
Centripetal inflorescence. Beginning to flower at the outside of the umbel.
Chalaza. A marked point in a seed, indicating the position of the cotyledon or
cotyledons, remote from the foramen through which the radicle usually issues.
Ciliated. Edged with hairs like the eyelids.

Clavate. Club-shaped.

Conniving. Closing inwards.

Coroll. The inner envelope of a flower, usually more ornamental than the calyx.
Crenate. Notched.

Cremilate. Slightly notched.

Cumulate. Heaped one on the other.

Cuneate. Wedge-shaped.

2 E

418

GLOSSARY.

Decurrent. Applied to the filaments, when their course down the tube is visible
below the point of their insertion.

Dehiscent. Gaping.

Dentate. Having prominences like teeth.

Denticulate. Having prominences like small teeth.

Depauperated. Having some material part wanting, or defective.

Disk. The superior surface of the germen round the base of the style. The
word has elsewhere had a more extensive and vague application, including the
cup or nectary.

Dissepiments. The lateral partitions of the cells of the germen and fruit.

Emarginate. Having the margin uneven, like the edge of a wound just cleansed
out or dressed ; an expression taken from surgery.

Embryo. The organized body within the seed, consisting of the cotyledons or
undeveloped leaves, the ascending plumule, the descending radicle, and the
neck which connects the cotyledon or cotyledons with the radicle.

Ensiform. Sword-shaped.

Epigynous. Having the position of the stamens above the future seed-vessels,
and consequently thegermen below the perianth.

Epiphytes. Plants that grow upon other plants.

Falcate. Sickle-shaped.

Filaments. That part of the stamen which supports the anther, so called because
its form is most frequently thread-like.

Filiform. Having the form of a thread.

Fimbriated. Terminating in fringe-like processes.

Foramen. A minute aperture in the seed, through which the embryo is fertilized
and the radicle usually issues.

Faucial. Situated on the throat or mouth of the tube of the perianth.

Germen. The future fruit before the decay of the perianth.

Glabrous. Perfectly smooth.

Glumaceovs. Having scales instead of calyx or corolla.

Gynundrous. Having the style consolidated with the stamens.

Heierotropous. Having the position of the embryo in the seed oblique.

Hexandrous. Having six stamens.

Hexapetaloid. Having the semblance of six petals.

Hilum. Tbe scar on a seed where it had been attached to the umbilical chord,
including the umbilicus and the space between it and the foramen, if not the
foramen itself.

Hypogynous. Having the position of the stamens below the ovary, consequently
the germen within the perianth.

Imbricating . Overlapping like tiles on a roof.

Inarticulate. Not jointed.

Incurved. Bent forwards, or down.

Involucre. An envelope to an inflorescence, not on the summit of an umbellate
inarticulate scape.

Involucrate. Having an involucre.

Lanceolate. Shaped like a lance-head, that is, slenderer than ovate.

Limb. The part of a perianth, or coroll, which is not tubular.

Linear. Having the margins parallel and not distant.

Lorate. Shaped like a thong, that is prolonged with parallel margins.

Monadelphous Having the base of the stamens united.

Obovatc. Egg-shaped reversed, the base narrowest.

Opercle. The persistent base of a style, forming a prominent point to an ovary
in an epigynous flower.

GLOSSARY.

419

Operculate. Having such a prominent point to the ovary.

Oval. Egg-shaped, attenuated equally at both ends.

Ovate. Egg-shaped, the base broadest.

Ovule. The future seed before it is fertilized, and in an early state of immaturity.

Parietal. Forming a wall or side ; or belonging or adhering to a wall or side.

Patent . Expanded wide.

Peduncle. The foot-stalk which connects a flower with the scape, stalk, or
branch ; simple, when it connects one ; compound, when it is branched, and
each branch thereof connects one.

Pentandrous. Having five stamens.

Perianth. The foliaceous part A a flower which has no calyx, consisting of
outer and inner segments, called sepals and petals, with or without a tube.

Pericarp. That which contains the seed, and which was the germen or ovary
before the decay of the perianth.

Persistent. Continuing after the usual period of decay.

Petaline. That which has relation to a petal ; petaline filament, that which is
inserted in or below a petal.

Petaloid. Having the appearance of petals.

Petals. The segments of a coroll ; the inner segments of a perianth.

Petiole. The footstalk which connects a leaf with a stalk, branch, or root.

Plicate. Plaited, having folds.

Pollen. The small dust-like particles, which issue from the anther, and by con-
tact with the stigma effect the fertilization of the ovules.

Porandrous. Discharging the pollen through a pore or partial aperture in the
anther.

Recurved. Bent back, or upwards.

Repand. Having the margin bent down backwards.

Resupinate. Properly lying on its back, or thrown back : but used by botanists
in a contrary sense to signify reversed, when, by the twisting of the foot-
stalk, the back of a leaf is turned upwards, or the lowest part of a flower is
uppermost.

Rosulaceous. Having the leaves crowded round the head of a stalk, in the like-
ness of a rose, diminishing in length towards the centre.

Rosule. A rosulaceous or rose-like head of leaves, having the form of an umbel.

Sagittate. Having the form of an arrow-head.

Scape. A succulent inarticulate flower- stem.

Scapaceous. Having a scape.

Secundifolius. Secundus (second) is used by botanists to express the bending
of all the leaves on a stalk one way. That, which is second, follows ; and I
suppose the leaves are considered by their similar inclination to follow or
second each other.

Segment. One of the divisions into which the limb of a flower is cleft.

Sepals. The segments of a calyx ; the outer segments of the limb of a perianth
standing in lieu of a calyx.

Sepaline. That which has relation to a sepal. Sepaline filament, that which is
inserted in or below a sepal.

Serrate. Edged like a saw.

Sessile. Seated on the scape, stalk, or branch, without the support of a footstalk.

Siliqua. A long capsule with two valves, and two parietal placentae, to which
the ovules are attached. Silicula is precisely the same thing shorter and
broader.

Spadiceous. Having a spadix.

Spadix. Originally the inflorescence of a palm, which has apetaloid flowers close

420

GLOSSARY.

set round a stalk, with an involucre below: applied to any like form of in-
florescence, even when the involucre is wanting.

Spu< he. An envelope consisting of one or more leaves, usually called valves, en-
closing the buds of a scapaceous umbel before they break through.

Spat u late. Shaped like a spatula or spoon, wider near the end and attenuated
below.

Stamen. The filament and anther.

Stigma. The summit of the style, often fringed, through which the influence of
the pollen is received and conveyed to the ovary.

Strophiolated. Having strophioles.

Strophioles. A girdle of w rinkles or tubercles around the hilum or chalaza of a
seed. The umbilical chord has sometimes been called a strophiole, but this
creates confusion.

Strumous. Preposterously enlarged in part ; strictly having a wen.

Style. The column proceeding from the summit of the ovary which supports
the stigma.

Sub. In compound prefixed to an adjective signifies a tendency to the quality it
represents, as subglaucous, of a colour tending to glaucous ; subacute, not
acute, but nearly so.

Subulate. Awl-shaped.

Sujfruticose. Having a tendency to the form of a shrub.

Triandrous. Having three stamens.

Tripartible. Separable into three.

Tripetaloid. Having the appearance of three petals. All adjectives formed
from a substantive, with the addition of oid, signify having the appear-
ance thereof.

Tubed. Having a tube.

Tube-shaped. Having the appearance of a tube, though cleft.

Tubular. Consisting of a tube.

Turbinate. In the form of a top, greatly attenuated at the base.

Valves. The portions into wdrich a dehiscent pericarp or a spathe splits.

Ventricose. Bellying out, enlarged towards the middle.

Versatile. Easily moveable on the point of its attachment.

Villous. Fleecy, woolly.

Umbel. A head of flowers of which the footstalks properly diverge from nearly
the same plane, but in crowded umbels the centre is often much protruded.

Umbilicus. The point in a seed of attachment to the chord by which it is con-
nected with the ovary, and through which it is nourished.

INDEX.

Page

Acis (PI. 34. f. 19—21) . 62, 82, 231

grandiflora (PI. 30. f. 4) . 332

Acorace® . . 46

Ajax . . 62, 79, 299

abscissus? . . 305

auceps (PI. 38. f. 36) . 302

bicolor . . 302

breviflos (PI 38. f. 39) . 303

Cambricus 1 . 305

candidissimus, cernuus 304

crenulatus (PI. 38. f. 35) 303

cuneifolius (PI. 39. f. 24. PI. 43. f. 5) 299

■ cyclamineus . . 305

erectior . . 299

festalis (PI. 43. f. 3) . 300

hexangularis ? . 305

humilior . . 299

lobularis ? . . 30.1

lorifolius (PI. 38. f. 38) . 302

luteus . . 304

major — maximus . 304

minor (PI. 38. i. 32) . 299

moscbatus . . 303

nanus . . 300, 415

nobilis (PI. 38. f. 40) . 301

obvallaris . . 304

ovules (PI. 38. f. 15 — 20) 304

propinquus (PI. 38. f. 33) 304

pseudonarcissus (PI. 38. 1 — PI. 39.

20—40. 5—8) . 300

pumilus (PI. 39. f. 23. PI. 43. f. 4) 300

rugilobus ? . 305

ryticarpus . . 300

Sabinianus (PI. 38. f. 41) 306

serratus? . 302

Sibthorpi (obvallaris) . 304

Telamonius . .301

tortuosus . . 304

tubaeflorus (PI. 38. f. 34) 302

Agave® (PI. 33. f. 16—19) 57, 69, 127

Agave® . 57, 67, 121

Agaviformes . 57, 69, 127

Alismkce® . . 47

Alie® . . .48

Allium Cowanium . 151,396

Alluvia! soil . . 396

Almyra, 3ee Halmyra . 202

Aloe® . . 48

Alstrocmdria . 56, 66, 88

acuminata (PI. 4. f. 3) . 97

61bida . 99

albijidra, see Preslana . 95

angustifolia . . 96

aurantiaca, see aurea . 98

Page

Alstroemeria aiirea (PI. 1. f. 3. 6 — 14) 28

Brusiliensis . 191

Bridgesiana (PI. 1. f. 50) . 94

caryophyll®a . 89

Catharinensis (P). 2 f. 6—8) 90

ciliaris? . 190

coccineu, see Spli®rine . 108

conferta (PI. 3. f. 1; • 97

Cummingiana . 96, 396

Curbrana (PI. 1. f. 51) . 94

dislichophylla, see Sph®rine 107

fios Martini, see pulchra . 93

• folioldsa . 91

gtaucescens, see Collania . 104

h®mantha . . 99

Hookeriana (PI. 1. f. 1. 2.4. 5. 17 ) 95

inodora ( PI. 2. f. 1 — 5) . 90

intermedia . . 97

Isabellana (PI. 6. f. 4 — 6) 88, 396

ligtu . . 92

ligtu, see caryophyll®a 89

lineatiflora . . 92

linifolia . . 91

longistaminea . 90

Macraeana . . 90

monticola . . 89

Neilliana . • 94

pallida . . 95

— — peregrina(Pl. 3. f. 4) . 91

plantagiuea . 89

Preslana (PI. 1. f. 43) . 95

psittacina . 88, 397

pulchella (PI. 4. f. 1) . 101

pulchella, see Simsiaua . 99

pygmma (PI. 8. f. 4 — 13) 100, 397

pulchra . . 93

Quillotensis (P. 2. f. 9 — 12) 98

recumbens (PI. 3. f. 2) . 97

revoluta (PI. 7. f. 9) . 91

secundifotia, see Sphmrine 107

Simsiana . 99

Solliana (PI. 4. f. 2) . 97

spathulata (PI. 1. f. 49. 51) 94

subrosulacea (PI. 5. f. 1 — 2) 397

tenuif'olia (PI. 1. f. 47) . 94

• tricolor, see pulchra . 93

twining, see Bomarea . 109

Valparadisiaca . 98

versicolor (PI. 48. f. 6) 94, 397

Alstroemeriseformes . 56. 66. 88

Amaryllidu.ee® . 53, 54

Araaryllid6® . 57, 70, 128

Amaryllidiformes . 61,76,228

Amarvllis(P1.36 .f.2. & 9, 10) 61,78,275,397

422

INDEX.

Page

Amaryllis Banksiana (PI. 32. f. 2),

grandiflora v. . 279

belladonna . 275

blanda . . 277

grandiflora . . 278

Josephiniana . 278

Spofforthiae, hybrids . 278

acuminata, aulica, calyptrata,

equestris, ffllgida, Maranensis
stylosum, psittacina, reginae,
reticulata, solandriflora, striatifo-

lia, vittata, see Hippeastrum 135

jEtnensis 1 see Sternebergia 187

advena, andicola, intermedia, ker-

mesina, maculata, pratensis, see
Habrautlius . . 156

ulba ? Forskael . 281

atamasco, bifolia, depauperata,

nervosa, tubispatha, see Zephy-
rantbes . . 170

aurea, FI. Per. flammea, see Py-

rolirion . 1 83

aurea, L’Her. radiata, see Ly-

coris . . * 229

Austi alasica, see Crinum flacci-

dum . 266

Bei terii, see Habrantbus robustus 166

■ — — bicolor, see Phycella . 154

Broussoneti, latifolia, longiflora,

insignis, ornata, revoluta, varia-
bilis, Zeylanica, see Crinum . 260

chilensis, see Habrantbus 163

chroroleuca (PI. 45. f. 6), see Ha-
brantbus Chilensis . 163

cinnamdmea, crispa, filifolia, see

Imhofia . 290

citrina, Clusiana, colchiciflora, see

Sternebergia . 187

clavata, see Gastronema . 132

Coranica, falcata, see Ammocbaris 241

corusca, curvifolia, flexuosa, humi-

lis, marginata, Sarniensis, ve-
nusta, see Nerine . 283

dubia, Linn. Am., see Hippeas-
trum barbatum . 138

elata, purpurea, see Vallota 133

formosissima, see Sprekelia 134

Fothcrgillii, see Nerine curvifolia 283

g'Sanlea> jagus> see Crinum petio-

latum . 260

bvacintbina, see GrifEnia 228

laticoma, see Nerine lucida 283

longifnlia, see Crinum Capense 269

lutea, exigua, see Oporanthus 188

minuta, see Zepbyrantbes Lind-

leyana . 17 4

montana, Tatarica, see Ixiolirion 125

orientalis, see Brunsvigia multi-
flora . 280

Peruviana, see Pyrolirion aureum 183

pudica . 279

pumilio, see Gastronema clavatum 132

radula, striata, see Brunsvigia 281

spiralis, see Carpolyza . 292

stellaris, see Hessea . 289

tubiflora, see Pyiobrion aui eum 183

Page

Ammocharis (PI. 33. f. 8, 9) 61,77, 241

Anigozanthus . 56,65,86

Apostasiacete . 47

Aracese . . 45

Argolasia . 56, 86

Argyropsis ? . . 176

Asparagem . 48

Asphodele® . . 48

Assaracus, see Ganymedes . 310

Ataccia . 63, 81, 333

Barbacenia (PI. 1. f. 29) . 55, 64, 82

Bomarea(Pl. 36. f. 12, 13) 57, 67, 109, 398

acutifolia (PI. 1. f. 20) . 112

amoena (PI. 46. f. 5), purpurea r. 399

anceps . . 116

aurantiaca (PI. 46. f. 1) . 399

bracteata (PI. 3. f. 3) . 112

Bredemeyerana . 118

Caldasiana . 119

Caraccensis (PI. 16. f. 2) 111

Cavanillesiana . 113

cordifolia . . 113

cornigera (PI. 17. f. 2, 3) 116

cornuta (PI. 17. f. 4) . 114

crinita (PI. 15. f. 4) . 119

crocea . 119

Cumbrensis (PL 18. f. 2) 115

densiflora (PI. 46. f. 4) . 399

denticulata . 118

dispar (PI. 19. f. 1) . 115

ebracteata (PI. 18. f. 4), tomentosa 1 18

edulis . . Ill

Fanningiana (PI. 16. f. 3) 116

fimbriata (PI. 18. f. 3) . 116

floribunda . 116

formosissima (PI. 14. f. 4) 116

gloraerata (PL 15. f. 1) . 115

grandifolia . 11? .

grandis . . 112

Guancana, pupurea v. . 399

Halliana (PI. 10. f. 3) . 117

hirsuta . 114

hirtella (PI. 1. 18, 19. 21—28) 112

Hookeriana (PI. 46. f. 3) . 398

latifolia . 113

lutea (PI. 5. f. 3) . 120

lyncina (PL 46. f. 2) . 398

macrocarpa (PL 4. f. 4) 114

Maranensis (PI. 17. f. 5) 111

ndbilis (PI. 15. f. 2) . 114

obovata (PI. 14, f. 2) . 112

ovata . . 113

Paltarumensis (PL 18. f. 3) 1 16

Pangoensis, Tomentosa . rl8

pardina(Pl. 14. f. 1 ) . 120

Patacocensis (PI. 14. f. 3) 120

pauciflora . 112

praecipua (PI. 14. f. 4) . 110

purpurea

118, 399

rosea .

118

salsilla (PI. 16. f. 4, 5)

110, 415

salsilloeides

111

setacea

117

simplex (PI. 15. f. 5)

119

subfalcala (PI. 16. f. 5)

1 10

superba (PI. 6. f. 1)

117

INDEX

423

Bom&rea Tatiana (PI. 15. f. 3)

tomentosa

torta

Branching

Page
113
115
115
55, 81

Bravda (PI. 12. f. 5, 6) . 57, 68, 124

Bromehaceae . . 47

Brunsvigia (PI. 36. f. 1) . 61,78,280

Burchelliana (PI, 22. f. 2), radulosa 281

minor (PI. 32. f. 1 ) . 281

multifl6ra . 280

radula . . 281

radulosa (PI. 22. f. 2) Burchelliana 281

striata . . 281

albiflora? Ecklon . 281

Buphane (PI. 36. f. 7, 8) 61, 77, 239

ciliaris . . 240

guttata (PI. 22. f. 1 ) . 240

disticha . . 239

toxicaria . . . 239

Burmanniaceaj . 47

Butomaces . . 47

Calceolaria . . 344

ascendens . 364

discolor . . 363

Callithauma? . . 60, 76, 225

viridiflorum . . 225

spathulatum ? . . 225

Calostemma (PI. 34. f. 34 — 38) 61, 76, 225
Campderiu, see Vellosia . 55

Campynema . . 56, 66, 87

Carey, Dr. . . 264

Carpodetes . . 60, 74, 195

Carpolyza (PI. 29. f. 9) . 62. 79. 292

Characteres breviores . . 831

Chlidanthus (PI. 27. f. 2) . 59, 73, 190

Chloraster, see Hermione viridis
Chloropsis . . 89, 127

Choeradodia . 56, 66, 87

Choretis . . 60, 75, 219

Galvestonensis (PI. 41 . f. 34, 35) 221

glauca (PI. 35. f. 1. PI. 41. f. 32) 220

Chrysiphiula . . 60, 200

Cliuanthus . 59, 74, 192

luteus (PI. 27. f. 1) . • 192

Clivia (PI. 36. f. 6) . 61, 77, 230

Coburghia . 60, 74, 196

incamata, var. (PI. 47. f. 5) 400

splendens (PI. 47. f.[6) . 400

Coeianthus . 56, 65, 85

Coetocapnia, see Bravoa . 57,68, 124

Collania . 56, 67*, 103

acicularis (PI. 11. f. 2) . 105

Andimarctina (PI. 8. f. 1 — 3) . 105

dulcis (PI. 7. f. 1— 8) 104,400

glaucescens (PI. 10. f. 1, 2) 104

involucrosa ( PI. 10) . 103

puberula (PI. 11. f. 1) 105

Collania, see Urceolina . 193

Coir.melinacese . 47

Conostylis (PI. 1. f. 45, 46) 56, 66, 87

Convallarieae . . 48

Cooperia ( PI. 36. f. 16) . 59,72, 17 8

Drummondiana (PI. 24. f. 2.

5—11) . 178

chlorsolen (PI. 24. f. 1) . 178

pedunculata . .179

Corbularia
Crinum

Algoense

Americanum

amcenum

angustius (bracteatum)

angustifolium

andmalum

Page

62, 79, 296, 412
61, 78,242, 400
272

254

255
243
259

o) lA

aquaticum, see campanulatum 270

arenarium . 259

Asiaticum . . 243

augustum . 246

Australe . . 246

Australasicum . 259

blandum (arenarium) . 259

brachyandrum . . 249

bracteatum . . 243

Braziliense . 251

• brevifolium, see bracteatum 243

brevilimbuin . . 266

Broussonetianum . 260

Cuffrum, see campanulatum . 272

campanulatum . . 270

canaliculatum . . 247

canalifolium . 258

Capense . . 269

Caraccense . 252

Carevanum . . 264

caudiceum (amcenum) . 255

Commelinianum . 254

concinnum . . 264

confertum . . 259

crassijolium, see variabile 268

cruentum . . 250

declinatum . 244

defixum . 255

disticbum . . 260

elegans . . 257

ensifolium . 255

erubescens . . 251

erythropbyllum . 258

exaltatum ^ . . 246

extorre . . 260

fhiccidum . . 266

Forbesianum . . 267

giganteum, see spectabile . 260

gracile . 258

Herbertianum . 263

Hybrid . 272

bumile . . 256

insigne

Lanceanum

latifolium

Lindleyanum

Loddigesianum

longijlorum .

longifolium

longistylum

lorifolium

marcocarpon

majus (erubescens)

Mauritianum

minus (erubescens)

Moluccanum

multijlorum

octoflorum

265

266

263

252, 401

253, 401
271

256

264

257
249
251

. 249

251
264
266

252

424

INDEX

Page

Crinara ornatum . . 262

Osbeckii . . . 266

Paxtoni . . 271

pedunculatum . . 246

petiolatum . . . 260

plicatum, see anomalum . 244

pratense . . 256

procerum . . 245

purpurascens . . 250

pusillum (PI. 32. f. 3) . 255

revolutum . . . 267

rigidum . . 248

riparium . . 269

rubricaule . . 246

lubrilimbum . 251

scabrum . . 262

seed (PI. 44. f. 15— 21) . 402

sinicum . . 244

speciosum . . 264

spectabde . . 260

strictum . . 253

submersum . . 262

Sumatanum . 248

tenellum, see Carpolyza . 292

toxicarium . 243

variabile . . 268

venosum . . 259

venustum . . 258

verecuadum . 256

virgmeum

undulatum

urceolatum, see urceolina

vuccaeides

Zeylanicum

CucuHgo ( PI. 1. f. 30. 3] )
Cvclanthacese
Cyrtanthi formes
Cyrtanthus (PI. 33. f. 12 — 14)
Desvauxiaceaa
Diomedes, see Ajax
Dioscorea
Dioscoreaeformes
Doryantbes
Elisena

Erinosma (PI. 34. f. 14 — 17)63,80,330,
Eriocaulonace®

Errors corrected

Eucrosia . 60, 75, 200,

Eurycles (PI. 34. f. 39 — 45. PI. 33

56, 64

58, 70,
58, 70,

57, 68,
57, 68,
57, 70,
60, 75,

261
253
193
260
262
, 83

45

128

128

48
307
122
121
128
201
404

49
429
402

5-7)

Fustephia

Explanation of terms
Fabricia, see Hypoxis
Flued borders

Fourcroya (PI. 33. f. 20 —25)
Gaertner, experiments of
Galantheae

61, 76,
58, 71,

227
156
1
56
402

57, 69, 126
348

63, 80, 329

Galanthus (PI. 34. f. 10 — 13. PI. 44.

f. 46) . 63, 80,

Ganymedes (PI. 39. f. 16, 17) 62,79,

capax ? fl.pleno

Gastronema . 58, 70,

Gethyllis . 59, 72,

A fra — ciliaris

lanceolata — rosea

329

307

403

132

185

185

186

Page

Gethyllis spiralis . 185

verticillata (PI. 25. f. 6) . 186

villosa (PI. 25. f. 7) . 186

undulata (PI. 25. f. 5) . 186

Gillesiaceae . 48

Gladiolus oppositiflorus . 366

mules . . 365,411

Graminacete . 48

Griffinia . . 61, 77,228

Habranthus (PI. 34. f. 8, 9. PI. 36.

f. 17) . 58, 71, 156, 403

advena . . 161

Andersonianus (PI. 24. f. 4.

PI. 34. f. 23) . 167

Andicola . .168

angustus . 160

Bagnoldianus . 162

bifidus . 160

Boothianus . 165

Chilensis . .163

Gillesianus (PI. 23. f. 1) 163

gracilifolius . .165

Hesperius . . 161

intermedius . 160

kermesinus . 159

litoralis . . 160

■ maculatus . . 167

miniatus (PI. 26. f. 2) . 162

nemoraJis . . 159

pallidus . . 162

parvula (v.) Andersonianus

(PI. 26. f. 4) . 168

pedunculosus (PI. 26. f. 3) 161

phycelloides . 157

pratensis . 159

pulcber (PI. 26. f. 1) . 161

pumilus . . 167

punctatus (PI. 47. f. 1) . 403

robustus . . 166

roseus . .163

spatbaceus . 160

speciosus (PI. 23. f. 2) . 158

sylvaticus . 166

versicolor . . 166

Haemanthus . 61, 77, 232, 404

Abjssinicus . 232

albiflos . 235

amaryllidioides . 233

brevifolius (PI. 30. f. 3) . 235

• carinatus . 236

carneus . . 234

coarctatus (PI. 31. f. 6) . 236

coccineus . 236

concolor (PI. 31. f. 2) . 238

crassipes (PI. 31. f. 10) . 237

Delagoensis . 2 33

doubtful species . 404

dubius, see Pbycella cbloracra 155

grandivalvis . 236

Hookerianus (PI. 46. f. 7) 404

hyalocarpus (PI. 31. f. 9) 236

humilis . . 234

incarnatus (PI. 31. f. 7) 237

intermedius . 235

lancetefolius . . 234

INDEX.

425

Page

Hsemanthus moschatus (PI. 31. f. 7) 236

multiflorus . 232

multivalvis . 235

orbicularis, see rotundifolius

pubescens . 235

pumilio . 234

punkeus . . 233

quadrivalvis (PI. 31. f. 4) 235

roseus, see carneus

rotuudifolius ( PI. 31 . f. 8) . 235

sanguineus (PI. 31. f. 5) 235

strigosus (PI. 30. f. 2) 234

tigrinus (PI. 31. f. 3) . 237

virescens . 235

undulatus (PI. 30. f. 1) 233

vaginatus, see Hessea . 289

Zebrinus . 237

Haemodorace® . 47

Halmyra? . 202

Haylockia . 59, 72, 182

Helene, see Narcissus . 319

Hermione . 63,80,319,404,415

ajquilimba (PI. 48. f. 5) . 404

Bazelraan minor (v.) brevistyla 324

bifrons (PI. 38. f. 10) 320, 406

biscrenata (v.) bifrons 320

brevistyla (PI. 41. f. 5 — 15) 324,406

citrina (v.) brevistyla . 324

compressa (v.) bifrons 320

Coreyrensis (PI. 37. f. 1) . 323

cupularis (v.) Tazetta 321

dubia (PI. 43. f. 6 — 8) . 323,415

elegans (PI. 41. f. 27) 325,407

flexiflora (v.) brevistyla 324

integra ? . . 329

Italica (PI. 41. f. 24, 25) 325, 407

lacticolor (v.) Tazetta . 322

obsoleta . . 328

• orientalis y. (v.) Tazetta 322

papyracea (PI. 48. f. 7 — 10) 323, 407

praecox (v.) Italica . 325

primuliua (v.) bifrons 320

serotina (PI. 41. f. 29 — 31) 326, 408

Tangiers (PI. 41. 28. PI. 42. 6) 328

tazetta (PI. 41. f. 15— 23) 321,408

Trewiana (v.) brevistyla . 324

viridiflora . . 328

Hessea . 62, 78, 289

breviflora (PI. 29. f. 6. PI. 43.

f. 4) . . 289

stellaris (PI. 29. f. 7 ) . 289

vaginata . . 289

Hessea, see Carpolyza 62, 79, 292

Hippeastrum (PI. 34. 7. & 28 — 31)

58, 71, 135, 409

acuminatum (v.) bulbulosum 140

acuminatum y Scbultes (hybrid) 409

ambiguum . 136

aulicum . . 135

barbatum (PI. 21. f. 1) 138, 409

breviflorum (PI. 21 . f. 4) . 137

bulbulosum . 139

calyptratum . 136

conspicuum (v.) solandriflorum 136

crocatum (v.) bulbulosum . 141

Page

Hippeastrum equestre . 138

equestriforme (v.) bulbulosum 140

glaucescens . 139

glaucophyllum (v.) aulicum 136

Harrisonianum (v.) vittatum 135

Hybrid . . 142,371

ignescens (v.) bulbulosum . Ill

latifolium (v.) vittatum . 137

longiflorum (v.) ambiguum 136

miniatum (PI. 47. f. 7) 139, 409

pallidum (v.) bulbulosum 140

platypetalum (v.) aulicum 135

psittaciuum . 136

pulverulentum (v.) bulbulosum 140

regium . . 139

regin®, see regium . 139

reticulatum . 137

rubrituba (v.) solandriflorum 136

riitilum (v.) bulbulosum . 140

Simsianum (v.) bulbulosum, mi-
niatum . . 140

Solandriflorum (choroleucum v.) 136

Spathuceum, see Hybrid Brookesi 143

Splendens, see Hybrid Brookesi 143

Striatifolium (v.) reticulatum 137

Striatum (v ) solandriflorum 136

Stylosum . . 138

subbarbatum (v.) bulbulosum 137

Tweedianum (v.) ambiguum,

(PI. 21. f. 3) . 136

unguiculatum (v.) bulbulosum 140

vittatum . . 137

Hybrid. Amaryllis , 278

Spofforthi®. blanda- Joseph. . 278

Hybrid. Crinum . 272

Altaclar®, erub. Cap. . 273

Baconi, Zeyl. erub. . 273

Brownii, Amer. bract. 274

Cecili®, Aust. scabr. . 274

Claronis? spect. Cap.? 274

Cooperii spec. long. . 274

Decandolii, flaccid. Aust. 274

Digweedii, scabr. Amer. . 274

Eboraci, As. Cap. . 273

Goweni, Zeyl. Cap. . 272

Haylockii, flacc. bract. 274, 400

Herbertii, scabr. Cap. . 273

Letiti®, erub. bract. . 274

Louisae, spec, defix. . 274

Mitcbami®, Aust. Cap. 272

Murrayi, Forbes. Carey. . 274

Parkerii, Am. erub. . 274

— — Pusev®, spec. Cap. . 273

Roxburghii, Defix. Cap. 273

Seymourii, revol. Cap. . 273

Shepherdii, cruent. Cap. 273

Stapletoni®, Zeyl. ped. . 274

Wallicbii, Carey. Cap. 273

Hybrid. Hippeastrum . 142

Allraannii, (3) calyptrato-vittatum 142

Altaclar®, ( 14) Psittacino-Griflini 143

Andersonii,(4) Bulbuloso-vittatum 142

Baconi, (24) Psittacino-regium 144

Batemanni, (21 ) Equestri-bulbu-

losum . .143

426

INDEX.

283

Page

Hybrid. Benthami, (11) Styloso-Jobn-

soni . . 143

Brookesi, ( 10) Bulbuloso-Johnsoni 143

Carnarvoni, (13) Solandrifloro-

Johnsoni . 143

Cartoni, (26) Aulico-Sweetii 144

Colvillii, (25) Reticulato-regium 144

Daubenii(12) Griffini-Johnsoni 143

Digweedii,(7)Striatifolio-vittatum 143

Donnii, (29) Hookeri-Haylocki 144

Goweni, (22) Reticulato-bulbu-

losum . . 144

Grahami, (6) Johnsoni-vittatum 143

Griffini, (8) Psittacino-Jobnsoni 143

Harrisoni, (17) Reticulato-stylosum 143

Haylocki, (15) Solandriflorc-bul-

bulosum . 143

Henslowi, (20) Regio-bulbulosum 143

Herberti, (16) Solandrifloro-sty-

losura . . 143

Hoodii,(19) Equestri-regium 143

Hookeri, (5) Goweni-vittatum 142

Johnsoni, (1) Regio-vittatum 142

Lamberti, (28) Cartoni-Graharai 144

Lindsevi (27) Aulico-reticulatum 144

Lindleyi (31) Griffini-Carnarvoni 144

Munroei (23) Psittacino-equestre 143

Parkeri(18) Bulbuloso-reticulatum 143

Seymouri (2) Aulico-vittatum 142

Spofforthire (30) Aulico-Carnarvoni 144

Sweetii (9) Reticulato-Johnsoni 143

Hybrid Nerine .

Cecilise.curvif.undul. — Claronis,

pulchella-undul. — Haylocki,
curvifolia-pulchella — Mitcba-
mire,curvifoli5-pulchella — Par-
kerii, pulch. hum. — Seymourii,
humili-undul. — Spoffortbiae,
venusta-undulata — vesicolor,
see Mitcbami*

Hybridizing process
Hydrocbarideae

Hymenocallis (PI. 33. f. 26. PI. 34. f.
46—51. PI. 35. f. 4)

acutifolia (v.) adnata

adnata

amoena

angusta .

angustifolia (v.) speciosa

Caribea

Caymanensis

cinerascens (v.) Caribea

crassifolia

declinata (v.) Caribea

disciformis (v.) rotata

distirha (v.) adnata

disticba-rotata (hybrid)

Dryandriua (v.) adnata

expansa

fragrans (v.) speciosa

Guianensis

humilis (v.) speciosa

Hybrid

litoralis (v.) adnata

lorata (v.) amcena

283

354

46

60, 75, 209
215
215,409
211

214

209
212, 415

210
212

215
212, 415

217
215

218
215
415

209

210
209
218
215
211

21(

Page

Hymenocallis ovalifolia . 212

ovata (v.) amoena . 211

paludosa 1 . 218

patens (v.) Caribea . 212

pedalis . . 211

princeps (v.) amoena . 211

princeps (v.) Guianensis 210

quadriflora (v.) rotata . 217

Quitoensis (PI. 22. f. 4) . 218

rotata . . 217

speciosa . . 209

Staplesiana (v.) adnata 215

tenuiflora . . 213

tubiflora (v.) Guianensis 210

undulata (v.) Guianensis 210

Hypoxidese . . 56, 64, 83

Hypoxidiformes . 56, 64, 83

Ilypoxis (PI. 1. f. 32 — 42) 56, 65, 84

lllus, see Ganymedes . 310

Jmntophyllum, see Clivia . 61

Imhofia . . 62, 79, 290

Burcbelliana (PI. 29. f. 5) 290

Crispa . . 290

filifolia ( PI. 29. f. 8) . 290

gemmata — virescens . 291

inus, adjectives in . . 327

Iridacea: . . 47

Ismene (PI. 3.5. f. 3) . 60, 76, 222

amancaes var. 1 and 2 . 222

Calatbina . . 222

pedunculata (PI. 35. f. 1) 222

nutaus . . 223

Juncacete . . 48

Juncaginacete . . 46

Ixiaeformes . 57, 68, 124

lxiolirion . 57, 63, 125

montanum (PI. 20. f. 2) 125

Tataricum (PI. 19) . 125

v. Scytbica (PI. 20. f. 1) 125

Kolreuter, experiments of 335, 348, 352
Kew gardens . . 410

Lanaria . . 56, 65, 86

Lanariseformes . 56, 65, 86

Lapiedra . . 59, 73,188,410

Leperiza . . 60, 74, 195

Leucojum (PI. 34. f. 18) 63, 80, 232

autumnale, see Acis . 232

capitulatum, see Molineria 84

vernum, see Erinosma . 230

Liliacete . . 48

Liriope . . 410

Littsa . 69, 127

Lophfola . 56, 65, 86

Lycdris . . 61, 77, 229

Marantacese . . 46

Markers . . 411

Melanthaceae . . 48

Molineria . 56, 65, 84

MonocotyUdones . 48, 52

Musace* . 46

Naiadaceae . . 46

Narcisseae . 62,79,292,385, 412

Narcissus (PI. 38. f. 3 — 7) 68, 80, 316

albus ? v. (poeticus) . 318

biflorus . . 317

427

INDEX.

Narcissus croceo-cinctus (v.) poeticus 319

gracilis (PI. 41. f. 4) . 316

grandiflorus (v.) poeticus 31?

majalis fPl. 40. f. 2) . 318

omatus 1 (v.) poeticus . 317

patellaris ( PI. 40. f. 3) 318

planicorona (PI. 41. f. 3) 316

• poeticus . .317

purpuro-cinctus (v.) poeticus 318

recurvus (v.) poeticus ( PI. 40. f. 1) 318

stellaris (v.) poeticus . 318

tenuior (v.) gracilis . 316

triflorus (v.) biflorus . 317

tripodalisl (v.) poeticus 317

spathulatus ? (v.) poeticus 317

Verbanus (v. ) poeticus (PI. 37. f.l) 318

Natural orders . . 412

Nerine . . 61,283

corusca . . 283

curvifolia (PI. 36. f. 4., seed PI.

45. f. 3) . . 283

flexuosa . . 283

humilis . . 283

lucida (P1‘ 36. f. 3) . 283

raarginata . . 283

minor (v.) venusta . 283

— — pulcbella . . 283

rosea (v.) venusta . 283

Sarniensis, see venusta 283

vedusta . . 283

versicolor, see Hybrid Nerine. 283

undulata (seed PI. 45. f. 2 283

Mitcbamia.*, bybrida ( PI. 45. f. 1) 283

Spoffortbite, Venusta-undulata 283

Haylocki, Curvifoli5-pulcbella283, 412

Nicotiana (seed PI. 44. f. 51. 52) 377

Nierenbergia (seed PI. 44. f. 47)j 378

Oileus, see Ajax . 305

Oporanthiformes . 59, 72, 185

Oporanthus (PI. 21. f. 2 — 3) 59, 73, 188

Fischerianus (PI. 47. f. 3) 412

Orcbidace® . . 47

Orontiacea; . . 46

Palmticeas . . 45

Pancratiformes . 59, 73, 190

Pancratium (PI. 34. f. 25—2?) 60, 75, 202

amancaes, calatbiforme, calathi-

num, nutaus, see Ismene.

amcenum, angustum, Caribeum,

declinatum, disciforme, disti-
cbum, expansum, fragrans,
Guianense, litorale, IVIexica-
num, ovalitolium, ovatum, pa-
tens, pedale, rotatum, specio-
sum, tubiflorum, undulatum,

see Hymenocallis . 209

Cambayense (PI. 42. f. 1) 207

Canariense . 205

Carolinianum . 203

coccineum, croceum, flavum.see

Stenomesson . 193

humile, see Tapeinantbus 190

Illyricum (PI. 33. f. I — 3) 206

incarnatum, variegatum, see

Coburghia . . 196

latifolium, see Leperiza . 195

Page

Pancratium longiflorum (PI. 42. f. 2) 208

Malabathricum . 206

Maritimum(Pl.<34. f. 21— 2— 42.7) 203

maximum . 208

parviflorum, see Vagaria 226

recurvatum, see Carpodetes 195

ringens, see Elisena . 201

viridiflorum, see Callithauma 225

verecundum . 206

zeylanicum . . 207

Pandanticeae . . 45

Petunia (seed PI. 44. f. 48 . 49) 378

Philogyne, see Queltia odora . 313

Pbilydraceae . . 48

Phlebocarya . 56, 66, 87

Pbycella . 58, 71, 151

attenuata (PI. 25. f. 1) . 155

bicolor . . 154

brevituba . . 154

cbloracra . . 155

corusca ( v.) ignea . 151

glauca (v.) ignea . 151

gracilitiora (PI. 25. f. 4) 152

Herbertiana . 154

ignea . . 151

latifolia (v.) attenuata (PI. 25. f. 3) 153

Macraeana (v.) attenuata (PI. 24.

f.l) . . 153

magnifies v. 1. (PI. 24. f. 15) 152

v. 2. (PI. 24. f. 12—14) 152

obtusifolia (v.) attenuata (PI.

25. f. 2) . . 153

Piperacete . . 45

Pistiaceae . . 46

Pontederacete . . 46

Pruiphys. see Eurycles . 227

Pseudoscordum ? . 11

Pyrolirion . 59, 72, 183

aureum (PI. 29. f. 4. PI. 23. f. 3) 183

albicans? flammeum ; flavum 184

Queltia (PI. 38. f. 2) 62, 79, 310, 413

aurantia (v.) feetida (PI. 39. f. 5) 311

calathina (v.) odora (PI. 39 f. 8) 3l3

Campernelliana (v.) odora 313

concolor (v.) feetida . 31 1

Curtisiana, see lsta (v.) odora 114

feetida . . » 311

galanthifolia (v. ?) montana ? 31 1

Gnuani, see aurantia . 312

grisea (v.) foetida (PI. 39. f. 7) 312

beminalis (v.) odora (PI. 39. f. 12) 314

incomparabilis (v.) foetida 31 1

interjecta (v.) odora . 314

jonquilla (PI. 39. f. 15 — 18) 314

juncifolia (P1.43. f. 1. PI. 39. f. 21) 315

laeta (PI. 39. f. 14) Curtisiana Haw. 314

Maclaeana (PI. 39. f. 1) 310

montana (PI. 39. f. 4) . 310

odora (PI. 39. f. 8 — 14) 313

orientalis (PI. 39. f. 3) . 312

var. (PI. 39. f. 2) S12

pumila (PI. 41. f. 26) . 316

pusilla (PI. 39. f. 22. PI. 43. f. 2) 315

rugulosa (v.) odora (PI. 39. f. 11) 314

semipartita (v.) foetida . 311

triloba (v.) odora (PI. 39. f. 13) 314

428

INDEX.

Page

R addia, see Vellosia . 55

Rajania (fruit PI. 43. f. 53 — 54) r>7 , 68, 1 0S
Restiacea: . . 48

Roxburgkiacea* . . 48

Sageret, Mons. . . 353

Salpiglossis (seed PI. 44. f. 51) 377

Scapace* . 57, 70, 128

Sceptranthes, see Cooperia 59, 72, 178

Schisauthes, see Queltia orientalis 312

Smilacete . . 48

Solanea; . . 378

Sphterine . 56, 67, 106

brevis (PI. 18. f. 1) . 108

coccinea (PI. 16. f. 1 ) . 108

distichophylla (PI. 12. f. 3 — 4) 107

Nervosa ( PI. 13) . 108

Secundifolia (PI. 12. f. 1 — 2) 107

Sprekelia (PI. 34. f. 32) 58, 71, 134

Stamen . . . 413

Stenomesson . 60, 74, 198

aurantiacum . 198

breviflorum (PI. 28. f. 7 — 7 b) 199

coccineum (PI. 28. f. 5. PI. 47. 4) 198,413

croceum (PI. 28. f. 4) . 199

curvidentatum (PI. 28. f. 2) 198

flavum (PI. 28. f. 1) . 193

pauciflorum (PI. 28. f. 3) 198

rubrum (PI. 28. f. 6) . 199

Sternebergia . 59, 73, 186

Americana, see Haylockia 183

^Etnensis . . 187

Caucasica Citrina (PI. 43. f. 9)

Clusiana. Colchiciflora 187,413

Dalmatica (PI. 47. f. 2) 413

Strumaria . 61,78, 617

angustifolia (PI. 29. f. 14) 287

Baueriana (PI. 29. 15. PI. 45. f. 5) 288

linguasfolia, rubella, truncata, un-

dulata (PI. 29)
Subgenus
Tacca
Taccaceae
Taccete

Tamus (PI. 1. f. 43)
Tapeinanthus

288
366

63, 81, 333
45

63, 81, 333
57, 67, 122
59, 73, 190, 414

Page

Tecophilea (PI. 24. f. 16 — 17) 57, 69, 125

Testudinaria . 57, 67, 122

Tiaranlhus ? . 60, 75, 202

Tros, see Queltia montana . 316

Tulipem . . 48

Turnip . . 370

Typhaceae . . 45

Vagaria (PI. 33. f. 10) . 61, 76, 226

Vallota (P. 1.52. P.33. 15)58, 70, 133, 414
Vellosia . 55, 64, 81

abietina . . 414

alosfolia — asperula — plicata —

raraosa — squamata . 81

Urceolaria, see Urceolina . 193

Urceolfna . 59, 74, 193

ptindula fulva (PI. 26. f. 5) 193

Wacbendorfacese . .48

YVeldenia . 56, 64, 83

Wiegman, Dr., experiments . 352

Xerophyta . 55, 63, 81

Xerophytdae . 55, 63, 81

Xyridaceae . . 48

Zephyranthes (PI. 34. 1 — 6.

PI. 36. f. 11, 14, 15) 58,72,170

Ackermanniana (v.) sessilis 175

atamasco . . 171

bifolia (v.) rosea . 173

Candida? (PI. 24. f. 18) . 176

cariuata . . 173

Commersoniana (PI. 29. f. 3) 174

depauperata . 171

gracilis (PI. 29. f. 1) . 172

Grahamiana (PI. 29. f. 2) 175

grandiflora, see carinata 173

hybrid Spofforthi* . 172

Lindleyana (PI. 35. f. 5) 164

Mesockloa (PI. 34. f. 24) 170

minima (PI. 24. f. 3) . 172

rosea . . 172

sessilis . . 175

striata (v.) sessilis . 175

tubispatha . 171

verecunda (v.) sessilis . 175

Zingeberaceae . . 46

Norman and Skeen, Printers, Maiden Lane, Covent Garden.

ERRATA.

P. 25, ). 7, for and hexandrous, read and by-
pogynous.

— 27, 1. 28, for trian-, read pentan-.

— 39, ). 2, for Amaryllidaceous, read Ama-

ryllidean.

— 52, I. 16, for Juncagines read Juncagina-

ceae.

— 52, 1. 31, Orcbidaceae should be in Italics.

— 54, 1. 2, for subsequalia, read subiequalia.

— 59, place t one-flowered, after seeds tes-

taceous.

— 61, for § 4 read § 5.

— 67, for operculosse read operculata;.

— 71, char, of Habranthus, for fasciculata

read seraifasciculata.

— for § 4. read § 5.

— Recumbens, for fig. 1. read fig. 2.

— 119, 1. 37, for 3 — 3. read 2 — 3.

— 122, 1. 18, for 23 read 33.

— 163, last line, for ochroleuca read cbloro-

leuca.

P. 182, 1. 8, for Caribbea read Caribea.

— 207, 1.26, after Hove, insert PI. 42. f. 1.

— 208, 1. 21, after Banks, insert PI. 42. f. 2.

— 212 ,for Caribasa read Caribea.

— 215 and 239, for Distvchum read Dis-

tichum.

— 267, last line, for Mr. read Dr.

— 281, 1. 13 , for Radulosa read Burchel-

liana.

— 299, 1. 29, for but before read bud before.

— 300, 1. 15, for cup read tube.

— 300, I. 30, for about 1| long read about

1* long. '

— 306, 1. 1, in character of Ganymedes, for

apicem versus vix read non.

— 308, 1. 31, for £ read 1^.

— 310, last line, for 133 read 123.

— 318, for Verbanensis read Verbanus.

— 335, 1. 2, for appears read appear.

— 382, 1. 25, for Agavese read Agaviforraes.

— 399, last line, for 5 read 4.

Shortly will be published ,

A T T I L A,

BY

THE HON. AND REV. W. HERBERT

CONTAINING

1. Attila, or the Triumph of Christianity : a poem.

2. Attila and his Prf.decessors : AN historical treatise.

?T.~~rk Bo'anlc'al Garden Library

QL 55 ,A1 H39
Hei;ber,( William/Amary||idaceae:

3 5185 00075 0123

in

SvA\. 1

%) Wm{{

1/

of '■ f

^§1/1] .

jfyii

1