And, until lately, many Botanists seemed disposed to adopt this distribution ; viewing the two
orders rather as sections of one, than two distinct families, though, so long ago as 1806, they
had been distinguished. In 1810 Mr. Brown declared them abundantly distinguished by the
position of the ovules and structure of the seed, added to the divisions and aestivation of the
corolla. While fully coinciding with that most learned Botanist, as to the propriety of their
separation, I do not think the reasons assigned quite satisfactory. As regards the position of the
ovules and structure of the seed on which he founds it, they are interblended, unless we are prepared
to advance a step further, and constitute the section Chionanthece a separate order. This
might be done without doing much violence to nature, though I consider it scarcely required ;
and yet, unless we do so, the ovary, ovules, and seed furnish only partially distinguishing marks
between the two families. Brown’s character of the ovary and seed of Oleacece is, “ovarium
simplex, biloculare, loculis dispermis, ovulis pendulis collateralibus. Semina albumine dense
carnoso, c o p io s o o f Jasminece, “ovariunxjoculis 1-spermis; ovulis erectis. Semen albumine
nullo vel parcissimo.” 1 am enabled to setfpgainst these, the fact of the ovules of the section,
Chionanthece, of Oleacece, being ascending 5 r even erect, and the seed exalbuminous, and that,
of many species of the section “ Trifoliate” of Jasminum, having dispermous cells, and the
ovules descending, or at all events amphitropous, that is, attached by the middle. I certainly
have not yet met albumen except very sparingly in any Jasminum, but examples of the others
will be adduced, showing that, as regards the ovary, ovules, and seed, the two orders pass into
each other and are intimately blended. We must look therefore to other parts of their organization
for distinctive marks.
Character of the Order. Flowers hermaphrodite or dioicous. Calyx monophyllous,
persistent, 4-cleft or 4-toothed. Corolla monopetalous, hypogynous, 4-cleft or sometimes 4-
petaled, with the petals united by pairs by the filaments, sometimes wanting, by abortion,
in the female flowers, sub-valvate in aestivation. Stamens 2, attached to the base, alternate
with the lobes or petals; anthers 2-celled, dehiscing longitudinally. Ovary simple, free, without
a hypogynous disk, 2-celled, with 2 collateral, pendulous or amphitropous ovules in each; style
1 or none ; stigma bifid or undivided. Fruit drupaceous, or baccate, or capsular, often 1-seeded
by abortion. Seed usually pendulous, albumen generally copious, dense, fleshy; sometimes
sparing or wanting. Embryo, when albumen is copious, straight, about half the length of the
seed, cotyledons foliaceous ; when wanting cotyledons amygdaloid.—Trees or shrubs with opposite,
simple or unequally pinnate leaves. Racemes or panicles axillary, or terminal, one brac-
teate. Flowers often fragrant, white or lilac coloured.
De Candolle remarks that the species of this order often greatly differ both by habit
ancT character, and that their affinities are best brought to light by grafting. “Thus the Lilac
will graft on the Ash, the Chionanthus on Fontanesia, and I have even succeeded in making the
Persian Lilac live 10 years on the Phillyria latifolia. The Olive will take on the Phillyria and
even on the Ash, but we cannot graft Jasmine on any of the Olive tribe, a circumstance . ■'which
confirms the propriety of separting these two orders.”
Af f in it ie s . As regards the more remote relationships of this order, Botanists seem
pretty well agreed, nearly all referring to Ebenacece, Sapotacece, and Ilicinece, as being more or
less remotely related; the nearer ones, with the exception perhaps of Jasminece, are not so
obvious. In De Candolle’s Prodromus it stands at the head of the series of orders distinguished
by their dicarpillary ovaries, as forming the transition order, in the lineal arrangement, between
the two groups of corolliflorous orders ; those, namely, having more than two carpels, and those,
with rare exceptions, limited to two. Bearing this structure in mind, as well as its extreme
constancy in orders where it does occur, it provides a nearly invariable line of separation
between the Olives and Ebonies, Hollies, &c., otherwise very nearly allied. I say, nearly
invariable, for one genus, referable to Ilicinece, has a 2-celled ovary. The line of distinction on
the other side is not so well marked, I mean that between the Olives and other dicarpillary
orders. The order is itself highly complex, being composed of a number of groups each of
which might almost be elevated to the rank of an order.
In Fraooinece (Ash tribe), for example, we have polygamous, apetalous, and also polypetalous
flowers, with samaroid fruit and albuminous seed. In Syringece (Lilacs), we have monopetalous,
tubular flowers, capsular fruit, and albuminous seed. In Oleinece (true Olives), we have both
tubular and deeply-parted flowers, baccate or drupaceous fruit, and abundantly albuminous seed.
And lastly, we have in Chionanthece the flowers and habit of Oleinece combined with ascending
or amphitropal, not pendulous, ovules and exalbuminous seed. These might all perhaps be raised
to the rank of orders, especially the last, on nearly the same ostensible grounds that led to the
elevation of Jasminece, viz. exalbuminous seed and ascending ovules, but still it seems to me
improper to raise them higher than sections, since all have tetramerous flowers and possess the
property of intercommunication by grafting, which the Jasmines do not. But while thus complex
in itself we find no other that can be confounded with it. Even the Jasmines, which some
Botanists seem disposed to unite, and nearly all place next each other, is by others considered but
remotely allied, and, if such indeed be the case, we may then look upon both as isolated orders
having affinities but no positively near relationships. Lindley observes, “To me I confess
that the unsymmetrical flowers of Jasmines offer a great difficulty in the way of placing them
in even the same Alliance as Olives, the more especially because that peculiarity is connected
with a decidedly nucamentaceous fruit. The two stamens usually present in Oliveworts may
be taken to show that the flowers of the order are really %/ which is confirmed by Tesserandra,
which has four stamens; the 2 stamens of Jasmine worts are probably connected with a
quinary type. The true affinity seems to be with Night-shades, as is indicated by the dicarpillary
fruit, regular symmetrical monopetalous corolla, axile placenta and undivided fruit of both
orders.” On these grounds he places the Oleacece in his Solanal alliance, next Solanacece; and
Jasminece in his Echial, next Boraginece and Labiatce, observing that “their unsymmetrical
flowers and deeply lobed fruit suggest” that affinity, and seem to point distinctly to those monopetalous
orders in which the number of stamens is different from that of the divisions of the
corolla, such as Labiates and Verbenas. Different persons draw different conclusions from the
same premises and, in common, I believe, with most other Botanists, I confess I cannot see much
relationship between Oleacece and Solanacece, but think that I can trace a good deal of affinity
between them and Apocynacece and Loganiacece, and so much between them and Jasminece,
that I cannot tell where the one ends and the other begins, though in their extreme forms so
abundantly distinct. Between Notelcea and Jasminum officinale, the distance is extreme—the
former having simple leaves, sub-polypetalous flowers, ovaries with 2 pendulous ovules in each
cell, a single sub-drupaceous fruit and copiously albuminous seed; in every one of which particulars
the reverse is the case in the other ; which is still further removed by the form and
number of its calyx lobes, and the aestivation of the lobes of the corolla. But if we compare
Notelcea with Jasminum auriculatum we find the distance considerably diminished by the
approximating forms of the calyx, and the ovaries in both having 2 ovules in each cell; the
seed are still different, the one being albuminous the other exalbuminous. Passing from Oleacece
to Chionanthece, we find the distance still further reduced, for now the seed of both families
are exalbuminous, the ovules of Chionanthece lose their distinctly pendulous character, while
those of some Jasmines lose their distinctly ascending one, and become amphitropous. The
habit still, however, marks their difference, Jasmines being generally twining and Olives
erect, the Jasmines besides have generally a well-developed style, while that of Olives is much
reduced or obsolete.—Advancing one step further, we come to Chondrospermum which unites
the two families. In this genus, the corolla is either 4- or 5-cleft with valvate aestivation, and
the style is reduced as in Olives, but the cells of the ovary has only one ascending ovule, and
the habit is twining as in Jasmines. The seed is unknown. Here then we have the flower of
Oleinece with the ovary and habit of Jasminece: to which of the two does it belong ? And
lastly, in J . rigidum we have a 4-lobed calyx. The transition between the two orders being
thus progressive, I cannot quite coincide with Brown in saying of Jasminece, “Ordo ab Oleinis
abunde diversus situ ovulorum structura seminis nee non divisione et aestivatione corollas.”
And still less can I agree with Lindley in viewing Olives as more nearly akin to night-shades
than to Jasmines or Jasmines to Echiums than to Olives. The want of symmetry observable
in the flowers of Jasmines is not, it appears to me, akin to that of Labiates, but to that of
Potaleads, and the rest of the structure more in conformity. On these grounds I view Olives