cases adduced among Primulacese, where a somewhat similar structure occurs, that no great
weight can be attached to that circumstance. Habit and geographical distribution must, I
apprehend, be looked upon, in this case, as the really essential distinguishing feature. The
section, or according to Alph. D. C. the order, Theophrastece has alternate sterile stamens
and extrose anthers as in Sapotacece, showing a transition from the one into the other and
indicating a direct affinity between the twb orders, which, however, are kept distinct by the
difference of the ovary, which is several-celled in Sapotaceae, and also by the very different
structure of the seed. Whether or not Theophrastece ought to be separated from Myriseneacece
is a question which scarcely comes within the scope of this work, there being no Indian
species of the former, but, were I called upon to give an opinion, I think it probable it would
be adverse to their separation partly on the ground of analogy furnished by Sapotacese, between
the genera of which it appears to me there are as wide discrepancies as between Theophrastacece
and Myriseneacece. But on the other hand many families, are less clearly defined and therefore
it is my impression that the verdict of A. De G. will ultimataly be confirmed as it rests
on some strong points.
G eographical D istribution. In 1833 when Alph. De Candolle’s paper on this family
was read in the Linnean Society, he was only acquainted with 180 species excluding Theophrasta,
since then the number has been nearly doubled. In his three tabular statement of the geographical
distribution of the order he assigns only 9 to “Ceylon and the Indian Peninsula,”
I cannot say how many I have, indigenous within these limits, but I am nearly certain the
number is little under twenty.
I extract the following notice on this subject from my Spicilegium Neilgherriense—“This
family is widely but unequally distributed, apparently preferring those countries enjoying a
rather high but equable temperature. They most abound in the Islands of the Indian Archipelago,
next to which ranks Bengal, Burmah, and the Tenasseram coast. The Indian Peninsula
and Ceylon, are placed low in the scale, whether owing to their possessing few or to these
floras being less known, I do not know, but I know that I have m my own collection nearly
twice as many as De Candolle assigned to both countries in 1833 when his very excellent paper
was read to the Linnean Society.”
Dr. Hooker in his “Botany of the Antarctic voyage,” gives an admirable account of the
geography of this family, which, though rather long, 1 am tempted to quote in full as a model
of perfection.
“Myrsineaceae are for the most part inhabitants of climates whose temperature is equable
and they particularly abound in insular localities, as the Islands of the Indian Ocean, Mauritius,
Bourbon and Madagascar. Their utmost Northern limit in the Old World seems to be the
Azores, lat. 39° N., Madeira, lat. 32°, and Tenriffe; but in no part of the adjacent continent
of Africa do they cross the Northern tropic; in Europe they are entirely wanting, and in Asia
extend only to Japan in North lat. 40°. The order is very rare in N. America, and especially
to the northward of Mexico, only one species inhabiting the United States, the M . Jloridana,
A. De C., and that is confined to the southern state, whose name it bears, lat. 30° N. In the
southern hemisphere they nowhere (except in New Zealand), are found to the southward of
the 36th parallel, and there in S. Brazil only. In Africa they reach the 33d, and in Australia
the 34th. Their extension into the 53d degree in the South Pacific Ocean, is hence a remarkable
circumstance, and probably in some measure to be accounted for by the uniform temperature
which the New Zealand Islands possess; further, they there bear a larger proportion
to the other dicotyledonous vegetation than they do in any other part of the globe. I have
alluded to the Suttonia divaricata having a considerable range in latitude, a circumstance not
without parallel in the order to which it belongs. Of this, Myrsine africana is an extreme
instance, that plant being found both at the Cape of Good Hope, in Abyssinia, and in the
Azores. The species of the Natural Order are, however, as M. A. De Candolle well remarks
( Linn. Trans, vol. xvii. p. 99), very confined as regards their geographical limits, Melastomacece
and Myrtacece, being two of the very few groups containing about the same or a greater
number of species which are more so.”—Jos. Hooker, Bot. o f Antarctic Voyage, p. 52.
ILLUSTRATIONS OF INDIAN BOTANY. 139
In addition to the above, it only remains for me to remark that generally they exhibit a
decided prediliction for sub-alpine forests, which may in some degree account for their non-
appearance on the African continent; the interior mountains, having been as yet imperfectly
explored and the same fact will, I think, account for so few having as yet been found in Ceylon
and the Southern Provinces of India. Linnaeus only knew one Indian species of the order,
his Samara Iceta, De Candolle only 9 in 1833, and now my Herbarium has about 20 species,
figures of 10 of which I have published.
P roperties and U ses. On this head little is known, the fruit of some of the species
possess considerable pungency and it is said, that those of Embelia ribes, which resemble black
pepper are occasionally used to adulterate that article. The fruit of a species of Samara are
called in Tamil ‘Devil’s pepper’ indicating its possession of . pepper-like properties.
Dr. Royle mentions that the fruit of Embelia robusta and Myrsine bifaria, are esteemed
gently purgative by the inhabitants of that part of the country where they are indigenous.
Upon the whole it appears, so far as present information extends, that they are deficient in active
properties.
R emarks on Genera and Species. The order is divided into three very distinct groups
by A. De Candolle, namely, Mcesece having the ovary inferior and many-seeded fruit—Embeliece
ovary superior : corolla polypetalous—and Ardisiece ovary superior : corolla monopetalous. The
first of these contains one genus, the second two, Embelia and Samara (Choripetalum, D. C.),
and the third 18, six of which have Indian representatives. Of these six, two only, Myrsine and
Ardisia, so far as I am aware have been found in the southern provinces. These are distinguished
at first sight by their inflorescence, in the former fascicled along the branches, in the
latter more or less distinctly panicled or thyrsoid.
The genus Samara (Lin.), has long been an enigma. Dr. Arnott lately cleared up the
doubts which hung over it by an examination of the original specimen. The genus was originally
established on a specimen now actually existing in the Linnean herbarium, and was well
defined, but the author, at the same time, quoted as a synonym, a figure which had no relationship
with the plant before him. The plate being well known, but not the specimen, it in course
of time came to be quoted as the authority for the genus and the original specimen of the plate
being subsequently examined was found not to correspond with the generic character. This in
place of suggesting the suspicion that Linnaeus had merely committed an error in his synonym,
was held as a proof that no such genus as he defined existed, and on that supposition A. De C.
constructed his genus Choripelalum which, as shown by a comparison of the two generic
characters, is identical with Samara, hence the latter, being the older name, must be retained
to the exclusion of the other.
EXPLANATION OF PLATE 145.
1. Ardisia polycephala (Wall.), flowering branch, nat.
size.
2. Flower-bud ready to expand.
3. Expanded flower.
4. Corolla detached and split open, showing the stamens
opposite the lobes of the corolla.
5. Anthers back and front views.
6. Calyx and ovary.
7. Ovary detached.
8. Ovary cut vertically showing the free central
placenta.
9 . -- - cut transversely.
10. A mature fruit.
11. The same cut transversely showing the solitary
seed and remains of numerous aborted ovules; the
embryo lying transversely across a copious homy albumen.
12. Embryo detached.
T