Illustrations of Indian Botany, Vol. 2/Cucurbitaceae

by Robert Wight

LXVIIL— CUCURBITACEÆ.

A large, very natural and most important order, principally confined to the tropics and warmer regions of the earth. In this country some species so greatly abound that they are to be met with climbing in almost every hedge and thicket, others are equally frequent spreading over the surface of the ground. Those frequenting the former situations often attain so great a size as completely to cover large trees with their luxuriant foliage. This though a well known order and some of the species very extensively cultivated, is still far from being well understood, its structure and habits being so peculiar, that it is difficult to find any other with which to compare it, and learn, by analogy, its true relations in the vegetable kingdom. While the order thus stands almost alone, in a manner isolated in the system of plants, and its species can

scarcely be confounded with any other, except perhaps Passifloreœ, there is not one where the difficulty of assigning definite limits to either genera or species is more felt.

Deferring to their proper place, all remarks on the genera and organs from which the characters are taken, I may observe here, that the discrimination of the species is most difficult, as no dependence can be placed on the form of the foliage as affording specific characters, almost every variation of form from simple, up to much divided leaves, being found in the same species and even occasionally on the same plant. Nearly all are annuals, with climbing succulent stems furnished with tendrils, supposed to be abortive lateral stipules, or according to modern Botanical doctrines transformed leaves, stipules being considered modified leaves. The flowers are usually unisexual, the male and female generally on the same plant or even springing from the same axil ; more rarely they are on different plants as in the example given, ( Trichosanthus palmata ), usually the flowers are white, red or yellow. The coloured portion of the flower is supposed by some Botanists to be a petaloid calyx, and the apparent calyx, merely certain external appendages, a view not likely to find many proselytes. The stamens in this order are peculiar and present many variations of form. These have recently been employed to supply sectional characters for the distribution and more easy definition of genera. The fruit, like every other part of these singular plants, is quite sui generis and is in consequence designated by its own name, Pepo or peponida, whence Peponiferœ the name given by Bartling and Endlicher to the class. A true pepo is, as I have recently shown, (Madras Journal of Science) a tri-car- pillary fruit but with the carpels inverted : that is with the dorsum of the carpellary leaf in the axis, and the placentiferous margins turned towards the circumference ; instead of, as usual with other fruits, towards the axis. Differences so great, in the construction of this most essential organ, mark this as a peculiar order, the affinities of which have still to be discovered.

These rather extended introductory remarks I have felt to be necessary to facilitate the right understanding of this very difficult and curious order.

" Calyx 5-toothed, sometimes obsolete. Petals 5, distinct or more or less united, sometimes scarcely distinguishable from the calyx, strongly marked with reticulating veins, sometimes fringed. Stamens 5, distinct or triadelphous : anthers 2-celled (or rarely 1 -celled), usually long, and sinuous, rarely ovate. Ovarium adhering to the tube of the calyx, of 2 or 3 carpels. Carpels inverted, that is, having the dorsum in the axis and placentiferous margins in the circumference, hence the fruit 2-3 celled but with 4 or 6 parietal placentas ! ovules solitary or indefinite, imbedded in pulp: style short ; stigmas 2 or 3, 2-lobed, very thick, velvety or fringed. Fruit fleshy, usually a peponida. Seeds usually ovate and compressed, enveloped in a juicy, or dry and membranous, arillus : testa coriaceous, often thick at the margin. Albumen, none. Embryo straight: radicle next the hilum : cotyledons foliaceous, palmatinerved. Stem succulent; climbing by means of tendrils usually lateral, and formed of abortive stipules. Leaves palmatinerved, alternate. Flowers unisexual"

Affinities. These are not by any means clearly understood. Hitherto Cucurbitaceae have been, by the almost universal consent of Botanists, arranged among orders with one-celled ovaria and parietal placentas, a place they can no longer be allowed to occupy if the above explanation of the formation of a pepo be found correct, since in it, we have a peculiarity of structure by which this family is widely separated from every other known order of the vegetable kingdom. Lindley places them in his Epigynous group, among a suit of orders all having a central placenta?, but distinguishing the Cucurbital alliance by the character " Placenta? parietæ" and remarks, in another place they are so closely allied to Passiflore " that they scarcely differ except in their sinuous stamens, unisexual flowers and exalbuminous seed." Endlicher places them between two parietal groups, associating with them another very curious family, Begoniaceœ, the affinities of which, if his view is correct, must up to this time have been totally misunderstood. The very intimate union of the calyx and corolla, which led Jussieu to view the flower as apetalous, and range the order under his class Apetalœ, is a point of structure which cannot be overlooked in determining its affinities ; for, though I think Jussieu's view on this point essentially incorrect, yet, it has had the effect of pointing out a relationship with Euphorbiaceae, in some points striking, but which might otherwise have been overlooked. As an order the Cu- curbitaceae are distinguished from all others by their stamens, hut more especially the anthers, the cells of which, in most of the order, are very long, winding upwards and downwards on the exterior surface of the connectivum. This structure of the anther, combined with the very remarkable one of the ovary and fruit, added to the habit of the plants and lateral tendrils, widely separate this from every other order in the vegetable kingdom. Various attempts have, however, been made to find associates near which to place them in the system of plants. Of the orders thus selected, as relations, some have superior some inferior fruit, some albumenous some exalbumenous seed, but all, except Lindley's group of Epignosœ, among which he places it are objectionable on account of their one-celled ovary : but, as a set off against this advantage he constructs his Cucurbital alliance of plants, having Parietal placentas which properly speaking, excludes Cucurbitaceœ. This group, however, remodeling the alliance will, I apprehend, be found upon the whole the best place, as agreeing with their general character "ovary inferior usually having an epigynous disk" but upon no other account, as in truth they have no relations here. The term Peponiferae under which Meisner groups 9, and Bartling 8 orders is most incorrect as being only applicable to one member of the group.

Geographical Distribution. This family though essentially a tropical one and of more frequent occurrence in India than in any other country, has yet a wide distribution over the world, a few being found even so far north as Europe and one in Britain. At the Cape, we learn from Harvey's " Genera of South African plants" that there are species referable to seven different genera: one species is found in Norfolk island : but generally they are rare in Australia. In equinoctial America and Africa they are of more frequent occurrence, but no where so abundant as in India and her islands, extending eastward to China and Japan. Blume enumerates 46 species found by him in Java alone, which leads one to the inference that the rest of Asia will be found to produce at least two or three times as many more. M. Seringe however only assigns 70 to all Asia including Java, which single fact, shows how little this family is known. Since more than half of the whole number are from that small portion. Let us hope this statement, added to the increasing facilities , which our improved knowledge of their structure confers, in the determination of species, may lead to greater attention being directed to their elucidation, which will unquestionably repay the enquirer with a rich harvest of interesting discoveries.

Properties and Uses. These are varied. Acrimony and a drastic tendency pervade many species, the fruit of some of which afford cathartics of remarkable power, acting, in even small doses, with great energy on nearly the whole line of the alimentary canal.

Generally speaking, however, this intensity of power is of rare occurrence though the property is found more or loss active in every part of the plant ; mildly in the roots of some and the leaves and young shoots of others, but in greatest intensity in the pulp surrounding the seed. The seed themselves do not partake of that property, being in nearly all, mild and oily. There is reason to believe that some at least, if not all the edible sorts, owe their freedom from this property to cultivation, as some of them in the wild state are found to possess it in great intensity. The Lagenarea vulgaris or bottle gourd may be cited as an example, it being recorded that some sailors were poisoned by drinking beer that had been standing in a flask made of one of these gourds: and Dr. Royle mentions a somewhat similar case, where symptoms of cholera were induced by eating the bitter pulp. The fruit of many of the species of Cucumis, the genus to which the Melon and Cucumber belong, are powerfully cathartic, among these C. Hardwickii, Royle and C. psudo colocynthes Royle, may be enumerated as the chief, but even the Cucumber, especially the less highly cultivated varieties of this country, are sometimes known to prove strongly aperient in susceptible constitutions. C. colocynthes (now Citrullus) is necessarily removed from the genus but affords one of the most valuable medicinal agents derived from the order.

The Melon, C. melo, and C. ulilissimus, so far as I have been able to learn are free from it. The fruit of some species of Luffa are violently cathartic such as L. amara and L. Bindual of Roxburgh, while those of L. acutangula (Cucumis acutangulus Ainslie) are a favourite potherb of the natives, and esteemed very wholesome. Some of the species of Bryonia especially B. alba and B. dicica partake of the cathartic properties of the family in great intensity, Curiously enough it is said that the juice of the root is strongly cathartic, and is often employed as such, while the young shoots are so free from the property that, they are used as a pot -herb and said greatly to resemble asparagus in flavour. The purgative properties of the root have been long known and in the opinion of some modern writers have fallen into unmerited neglect, as being fully equal in power, even when dried and powdered, to Jalap and when recent much more so. But of all those yet. mentioned none approach the Elaterium in the concentrated virulence of this quality ; a few grains of the pulp being known occasionally to bring on symptoms of poisoning, and a case is recorded by Dr. Christison where a person after carrying a specimen in his hat was attacked with headache succeeded by colic pains and frequent bilious vomiting and purging.

Such being the predominating quality of the family, it is well to be cautious in the use of even the best known — many however are in use as pot herbs, among these may be mentioned with just encomiums the red gourd Cucurbita maxima (C. nispeda Ainslie) the flesh of which when boiled somewhat resembles in taste a fine tender carrot. The water melon Cucurbita citrullus so highly esteemed for the cool refreshing juice of its large fruit. The white gourd ( Benincasa cerifera or Cucurbita pepo) which Ainslie informs us (under Cucurbita hispida) is presented at every native marriage feast, being supposed to ensure prosperity to the wedded pair. The vegetable marrow ( Cucumis ovifera) justly esteemed one of our finest culinary vegetables ; and a few others.

All the numerous cultivated varieties of the melon and cucumber are known to be wholesome. Some, if not all the Indian species of Momordica, seem equally safe. The fruit of several species of Trichosanthes especially, those of T. anguina are in daily use, even among Europeans, dressed in curries ; but those of T. palmata are not used and are considered poisonous by the natives. Those of our Cocoinia indica, {Momordica monadelpha, Roxb.) so common in every hedge, is eat by the natives in their curries and when fully ripe, (quite red and pulpy) seem to afford a favourite repast to many birds. Notwithstanding the drawbacks mentioned above, this is certainly a most useful family of plants, owing to the great size of their fruit and the large quantity of nutritious matter which the edible sorts afford, and which, on that account, are largely cultivated in every part of India. Those unfit for food, supply many useful medicines, but even the best known, ought to be used cautiously when not ameliorated in their qualities by cultivation.

Remarks on the Genera and Species. This being a family not yet well understood, the limits of the genera are consequently imperfectly determined, whence, in the opinion of some Botanists, several very unsuitable combinations of species are met with among them. This seems probable enough, but is an error not easily avoided in families so natural, unless we are very careful in the selection of our characters, and attentive not to introduce anything extraneous, by the employment of characters derived from organs apt to vary in their forms, for, while they appear to give greater precision, they actually weaken the definition, or may even render it altogether useless. M. Serenge for example (D.C. Prod.) employs the relative size and shape of the calyx segments as generic characters, parts in themselves liable to vary, even in the same species, both in size and shape, therefore quite unfit to enter into a generic character, and when so employed are liable either to mislead or to constitute very artificial genera, and, what I consider still more objectionable, he constitutes one genus on account of its male flowers being furnished with a large bractea, but excludes from it Trichosanthes palmata, the bracteæ of which are equally conspicuous.

The order is divided into two tribes of very unequal magnitude, the one, Nhandirobeae, containing only two genera and very few species, the other Cucurbileae, to which Meisner assigns 35, but which are reduced by Endlicher to 28 genera.

The first of these tribes Nhandirobeae, has by Endlicher been raised to the rank of an order. Whether in this view, he is correct I am unable to say, as I have, not a female flower wherewith to examine the ovary, on the structure of which, as compared with that of true Cucurbitareae, the decision of the question must mainly depend. If the carpels are similarly inverted in both, which the section of the fruit in the accompanying figure seems to indicate, then, I think it may very well be retained in its present position or at most removed as a sub order : if different, that is, if the placenta are, in the usual acceptation of the term, truly parietal, then it ought undoubtedly to be removed and plated near if not actually united with Passifloreae. Hut in either case, the points of difference, between Nhandirobeae and Cucurbiteae, seem quite sufficient to distinguish them as separate orders, should that be deemed a judicious arrangement, which is not improbable, were it only for the purpose of simplifying the ordinal character.

Its claims to rank as a sub-order of Cucurbitaceae may be at once admitted, supposing the carpellary structure the same, on account of the differences in the flowers and anthers which are very distinct in the two tribes and by Nhandirobeae having axillary not lateral tendrils.

Two genera only are assigned to this tribe or order — Zanonia, (Plate 103), and Fevellea. The first of these, so far as yet known are altogether of Asiatic origin and extends from Malabar eastwards as far as Java, where Blume found 2 species. The other, which until lately, were only known as natives of America, have recently been found in Assam, whence I have, through the kindness of Captain Jenkins, received two species, but unfortunately the male flowers only.

The 35 genera of Cucurbitaceae, Meisner in his table, distributes according to the stamens into two leading sections — A — stamens free — B— stamens variously united among themselves. To the first of these sections he refers only two Indian genera Luffa and Citrullus, others have since been added. To the other, ten are assigned, They are further distributed according to the number and mode of union of the stamens — the petals free or united, the anthers whether connate or distinct, the cells regular or bent and anfractuose, &c.

Since the publication of this tabular arrangement of the genera, Schrader, a German Botanist, has published a revised distribution of them, founded on a minute examination of a large proportion of the order, establishing his arrangement of the genera according to characters taken principally from the male flowers. The original memoir, published in a German periodical (Linnæa vol. 12), I have not yet seen, but Dr. Arnott obligingly prepared for my use a conspectus of all the genera of that memoir, adding several new ones of his own. This he has permitted me to publish should I think it desirable

Having satisfied myself, by the examination of several of the sections and genera, of the correctness generally of the characters and the facility of applying them in practice, as well as of the greater precision which their adoption confers on our generic characters, I can have no hesitation in subjoining this Conspectus and recommending, to careful examination, the principles on which it proposes to construct the genera of this very obscure and difficult order, as, I think, with some slight modifications it will be found to merit general adoption. Until, however, I have had more extended opportunities of doing so with recent specimens I refrain from saying more in its favour.

The characters of the tribes were not given, which is of little consequence as the tribe Cucurbiteae is, with a single exception, the only one found in India and the whole of the genera of this tribe are given, whether Indian or not, to enable those who may give their attention to this investigation, to determine genera not hitherto introduced into the Indian Catalogue. It may be necessary here to observe, that the fruit, in those genera said to have it baccate, is not truly a bacca, but only a slight modification of the peponida, the placenta being only apparently, not truly parietal, as in the true bacca.

The following explanatory extracts from the letter which accompanied the Conspectus, may not inappropriately be introduced.

" I have lately been revising our East Indian Cucurbitaceae, in consequence of Schrader's paper in the Linnæa, vol. 12. At first I was inclined to consider it worse than useless to subdivide old genera, especially Bryonia, as be has done : but when I came to consider his sectional characters, and that the form and position of the stamens and anthers and stigma and fruit are, really, the only characters hitherto employed for genera by other Botanists ; and that all Schrader has done, is to keep only those species in their old genera that agree with the character and turning out, and making new genera of those that do not, then I felt inclined to go great lengths towards adopting his views. I intend here to give you a Conspectus, or abridged generic characters, which I have drawn up, not only for the Indian, but also for those of the whole world, which I shall not object to your publishing as abridged characters of the genera of the tribe Cucurbiteae of Schrader."

At the conclusion of the conspectus he continues, — " These seem to be all the genera known, that truly belong to Cucurbiteae, they have all unisexual flowers. Gronovia has them bisexual, but is otherwise very closely allied. Allacia cannot be of this order, unless we suppose the description quite erroneous ; and if so, Loureiro may have had before him, in part at least, the Telfairia pedata ; Myrianthus cannot belong to Cucurbitaceae. Thladianthus Runge is imperfectly described as to the stamens, but may perhaps form a 7th tribe.

" I have laboured under great difficulty in making out these characters, partly because the published descriptions were very imperfect, partly because I had not several of the genera, and partly, from the extreme difficulty of examining the anthers after being dried and pressed. I would therefore suggest to you and other Indian botanists, to re-examine all the Indian ones on living plants, and have drawings made, paying particular attention to the representation of the anthers.

" At first you may, as I did, confuse section 7 with section 9, but if you will compare the flower of Citrullus, Momordica or Lagenarea with Cucurbita or Coccinia, you will readily see the difference. In section 7, the connectivum is lobed, and the anther cells are placed along the edge of the lobes— in sections 8 and 9 the connectivum is not itself lobed, but the anther cell is bent," (it winds upwards and downwards along the back of the connectivum).

Warned by the concluding paragraph of the difficulty attending the description of these plants from dried specimens, I took occasion, as opportunity offered, to compare some of the sectional characters with recent specimens, and feel disposed to think the sections too numerous, and not sufficiently distinguished. I have not yet succeeded in comparing the whole, but would suggest the following alterations, which 1 think would improve the arrangement.

Section 6 might with advantage be suppressed, and its only genus referred to section 5. Sections 7 and 9 would be better united, the anthers being the same in both ; transferring, however, Cucurbita to section 8, on account of the anthers, which are similar to those of Trichosanthes, making the insertion of the filaments a matter of secondary consideration, a generic not a sectional distinction.

The difference between sections 7 and 8 would then be — not that in the former the anther is lobed and in the other entire, but that in section 7 the back is traversed by an elevated gyrose ridge, on the top of which the long gyrose anther cell is placed, while in section 8 there is no such elevation, the anther cell being sunk into the substance of the connectivum, not elevated on a ridge with a deep furrow between each bend. To this may be added that the oonnectiva in section 8 is elongated ; hence, from the union of the three, a cylinder results, while in the other their union produces a sort of capitulum.

Bryonia Garcini, doubtfully referred to Bryonia, is a new species of Pylogyne : Bryonia leiosperma, I find, from the examination of dried specimens, is a second species of Mukia, with which it agrees well in habit.

Notwithstanding these differences of opinion, it is not my intention to alter the conspectus, but print it simply as it reached me, the few additions I have to make, being included within brackets — thus [ ]. Before proceeding further it may be well to explain what is meant by the term tri-adelphous, as applied to this family, which is of such frequent occurrence in the following characters. The normal structure of Cucurbitaceae is to have five stamens, in place of which we usually find only three ; but when these are carefully examined it appears that two of them are twice the size of the third, and are actually made up of two united : each set is then called an adelphi or brotherhood, and the three together tri-adelphous. This structure is readily seen in the Pekunkie ( Cucumis acutangulus, Ainslie) where the anthers do not cohere. In those where they do cohere it is not so clearly seen, as they then require to be separated artificially before it can be made out.

In some genera the anthers are described as being one or two-celled; these characters require to be used with caution, as being generally of very difficult application in practice. Theoretically every anther is two-celled, and herein examining a number of instances with particular care, under a high magnifier, I have found most of them actually two-celled, though on less careful examination they appeared only one celled. If such is the case when examined with fresh specimens, how much more liable to error must we be when working with dried ones.

CONSPECTUS OR ABRIDGED CHARACTERS OF THE GENERA OF THE TRIBE CUCDRBITEÆ* OF SCHRADER.

§ 1. — Filaments 5, inserted on the throat of the corolla ; anthers distinct or 3-adelphous, anticous, straight ; fruit,——— baccate, Jew seeded.

1. CONTANDRA (Schrad.) Corolla 5-partite : connectiva conniving, oblong, conical : fruit beaked. South Africa.

2. Cyrtonema (Schrad.) Limb of corolla 5-partite : filaments 5, incurved, connectiva incrassated 3-adel- phous ; anthers fixed laterally under the apex. South Africa.

§ 2. — Filaments di or tri-adelphous, inserted on the tube of the corolla ; anthers lateral, straight, 2-3-adelphous.

3. SICYDIUM (Schlecht.) Corolla 5-petaled, petals undivided : filaments 3-adelphous, dilated and incurved at the apex ; anthers without a beak. Mexico.

4. BRYONOPSIS (Arn.) Corolla 5-partite, lobes obovate, entire, undulated : filaments 3-adelphous, inserted on the throat, straight ; anthers pointless : stigma fringed : berry few seeded. East Indies — Courtallum.

5. ACHMANDRA (Arn.) Lobes of the corolla undivided : filaments tri-adelphous very short ; anthers anticous, inserted along the margins of the connectivum, linear oblong ; connectivum prolonged into a short beak beyond the anther : fruit baccate (always ?) beaked. East Indies. Bryonia epigœa, rostrata, deltoidea, and an undescribed species from Malabar.

OBS. — Perhaps this and the two last genera might be joined to Melothria.

6. MELOTHRIA (Linn.) Lobes of the corolla undivided, denticulated: filaments 3-adelphous; connectivum pointless : fruit baccate, not beaked. America.

OBS. — Schrader notices an East Indian one, but that may perhaps be an Achmandra.

7. CERATOSANTHES (Schrad.) Lobes of the corolla linear bifid : filaments 3-adelphous. America.

8. ANGURIA (Linn.) Lobes of the corolla entire : stamens diadelphous : fruit somewhat 4 -angled. America.

§ 3. — Filaments S-adelphms inserted on the top of the tube, anthers all cohering by means of their connectiva, and applied at the back along the margins of the connectiva, sigmoid. (?) two-celled.

9. SCHIZOSTIGMA (Arn.) Style simple, stigma peltate, fleshy, cleft into 10-12 radiating linear lobes. Andes of Mendosa. (Cucurbita asperata. Gill.)

S 4._ Filaments distinct or tri-adelphous. inserted on the throat of the corolla, anthers 6 or 3-adelphous, gyrose, anticous.

10. SPHENANTHE (Schrad.) Mexico.

§ 5. — Filaments tri-adelphous, inserted at the base of the corolla, anthers lateral, straight, 3-adelphous.

11. PILOGYNE (Schrad ) Calyx campanulate; segments of the corolla patent, much longer than the calyx : anthers one-celled : styLe entire; stigma 1, pileate : fruit baccate, few seeded, obtuse. South Africa. [ East Indies — Bryonia Garcini ?]

12. ZEHNIRIA (Endl. ?) Lobes of the corolla quite entire : style trifid ; stigmas 3, flabelliform, quite entire : fruit baccate, few seeded, blunt. South Africa, East Indies and ? Norfolk Island. — Bryonia Mysorenses. B. Hookeriana.

Perbaps this is only a subgenus of Pilogyne.

13. KARIIA (Arn.) Calyx urccolate, corolla scarcely exserted, lobes minute, quite entire ; anthers 2-cell-ed; style entire; stigma pileate, 3-fid : fruit a peponida, many seeded, blunt or with a short thick beak. East Indies, Bryonia umbellata. B. amplexicaulis.

14. RHYNCHOCARPA (Schrad.) Lobes of the corolla denticulate, ciliated ; style trifid ; stigmas 3, jagged and toothed : fruit with a long slender beak. Guinea.

§ 6. — Filaments tri-adelphous, inserted at the base of the corolla, anthers all cohering, posticous, linear, straight.

15. MUKIA (Arn.) Style entire, stigmas 3, more or less cohering, erect. Anthers distinct, one-celled, lateral ; [connectivum prolonged, forming a projecting point: a globose abortive ovary in the bottom of the calyx.] East Indies — Bryonia scabrella. [B. leiosperma.]

§ 7. — Filaments 5, or tri-adelphous, inserted at the base of the corolla ; connectiva toothed or lobed, anthers applied at the back along the margins of the connectivum, and therefore flexuose, gyrose, or anfractuose.

16. BRYONIA (Linn.) Corolla 5-cleft : anthers triadelphous, one-celled : style trifid ; stigmas subreniform or bifid : fruit ovoid or globose, baccate ; few seeded. Europe and East Indies — B. laciniosa, India. B. alba. B. dioica, Europe. Perhaps also B. Garcini and leiosperma, but of these 1 have no male flowers by me to examine [B. Garcini. Stamens tri-adelphous one-celled : cells linear, marginal, not sigmoid, style one, stigma dilated, peltate : ovary few seeded ; pepo inverse reniform, 2 seeded. This is probably a new genus, but if not seems to belong to Pilogyne rather than any other here. It cannot possibly belong to Bryonia. ]

In the European plants, the type of the genus, there are two ovules in each of the 3 cells of the ovary — nearly all Blame's species belong to other genera.

17. CITRULLUS. (Schrad.) Corolla persistent, 5-parted, subrotate : anthers tri-adelphous, bilocular : style 3-fid; stigma obcordate, convex: fruit a fleshy or dry and fibrous, many-seeded peponida. Africa, East Indies. Cucurbita citrullus and Cucumis colorynthis.

18. ECBALIAM (Rich.) Corolla 5 cleft : anthers triadelphous: ovules in two rows in each cell; stigmas three, two horned : fruit an elastically and irregularly bursting peponida. Europe — Momordica elaterium.

19. MOMORDICA (Linn.) Petals 5, adnate to the base of the calyx, deciduous : anthers all cohering: ovules

• This tribe contains all the East Indian genera except Zanonica — Schrader refers Erythropapum of Blume to Cucurbiteae but that genus is very closely allied to Mackaya (Arnott in Jardine's Mag. of Zool. and Bot. vol. 2), and does not belong to the order.

in a single* row in each cell ; stigma two lobed ; fruit a capsular, elastically bursting, 3 valved peponida. E. Indies, South Africa and America.

[* This though practically correct is not theoretically so, the carpellary structure being the same here as in others; each margin has iis placenta and ovules: and though at any one section only one row appears, we do not find the ovules always attached to the same line of placenta on slicing the ovary successively from end to end but sometimes on the one, sometimes on the other side of the cell ; such at least I find it in Momordica Charantia.]

To this genus seems to belong, Muricia, Loureiro and Neurospermum, Kaf.

20. LUFFA (Cav.) Petals 5, inserted in the base of the calyx, deciduous : anthers all distinct or di-triadelphous : style 3 fid ; stigma reniform or bipartite : fruit a peponida, at length dry and internally fibrous, usually opening by a terminal lid, rarely indehiscent. East Indies and Arabia. There are 3 sections of this genus.

1st. Stamens distinct, LUffa pentandra, acutangula, and Kleinii.

2d. Stamens 3-adelphous. L. amara, Roxb. and nearly all the species of Turia, Forsk.

3d. Stamens di-adelphous — L. tuberosa, Roxb.

21. BENINCASA (Sav.) Corolla (yellow), five -parted, patent : anthers 3-adelphous : style undivided, very short; stigma large, thick, irregularly lobed and plaited : peponida, fleshy indehiscent. Asia.

22. LAGENARIA (Ser.) Corolla (white) five petaled : anthers 3-adelphous : style almost none ; stigmas 3, thick and 2-lobed : peponida fleshy and indehiscent. India, South Africa.

§ 8. — Filaments, 3-adelphous, inserted on the tube of the corolla ; connectiva entire, anthers 3 or mon-adelphous, posticous, linear, bent upwards and downwards : calyx long, tubular.

23. TRICHOSANTHES (Linn.) Segments of the corolla lacerated and fringed : anthers 3-adelphous ? or all united : style trifid ; stigmas oblong, subulate : fruit a peponida, many seeded. E. Indies.

I reunite Involucrarea to this as a mere section depending on the bracteas, the character taken from the anther not holding good, at least T Cucumerina has frequently the anthers all united, and I suspect also T. anguina ; perhaps they only become tri-adelphous after fecundating.

[In all the species I have had an opportunity of carefully examining, the anthers are monadelphous or united. The style is not trifid, nor properly speaking the stigmas subulate, as they cohere nearly to the apex by their central face, though the stigmatic surface extends for some distance outwardly, and presents a somewhat subulate outline.

In T. anguina they are never tri-adelphous, the anthers cohere to the last as represented in the accompanying figure. This last species with T. globosa, and trifoliata, Blume, and Involucrarea, Serange (T. Wallichana) form a very characteristic section, perhaps a sub-genus, distinguished by their curiously bracteated male flowers]. R. W.

24. GYMNOPETALUM (Arn.) Calyx constricted at the mouth ; corolla (yellow) 5-parted ; segments quite entire : anthers all closely cohering : fruit baccate, ovate, beaked, few-seeded : seeds large, roundish, with a blunt margin E. Indies. There are two species —

1. G. Ceylanicum (Arn.) Leaves deeply 5-lobed; perianth glabrous : Bryonia tubiflora W. and A.

2. G. Wightii (Arn.) Leaves 3-5-angle-lobed ; perianth hairy. Courtallum.

§ 9. — Filaments usually tri-adelphovs, inserted at the base of the perianth ; connectiva entire, unless when produced into appendages beyond the anthers : anthers linear, posticous, bent upwards and downwards (calyx campanulate or rarely infundibuliform.)

25. CUCUMIS (Linn.) Corolla 5-parted : anthers tri-adelphous, or all of them slightly cohering, with appendages at the apex ! Peponida fleshy, indehiscent, or rarely irregularly dehiscent, polyspermous : seeds ovate, compressed, sharp edged. Asia, Africa and America.

26. CUCURBITA (Linn.) Corolla campanulate, 5-cleft : filaments tri-adelphous at the base or quite monadelphous ; anthers all cohering, without appendages : peponida fleshy, indehiscent, polyspermous : seed with a slightly thickened edge. Asia and America.

27. ELATERIUM (Lin.) Petals scarcely united at the base : filaments monadelphous, anthers all cOhering : style thick ; stigma capitate : fruit a coriaceous, one-celled, few seeded capsule, bursting elastically by two or three valves. America.

28. SCHIZOCARPUM (Schlch.) Corolla infundibuliform, quite entire : filaments 3-adelphous : anthers all cohering; peponida many seeded, bursting by several valves that cohere by their apex. Mexico.

29. COCCINEA (W. and A.) Corolla campanulate, segments acuminated : filaments monadelphous, anthers tri-adelphous, conniving, without appendages : peponida somewhat baccate, many-seeded. [Usually of an oblong oval shape and bright red when ripe.] East Indies.

§ 10. — Filaments monadelphous, connate into a column, which is capitate at the apex, and then bearing the gyrose posticous anthers.

30. CEPHALANDRA (Schrad.) South Africa.

EXPLANATION OF PLATE 103.

ZANONIA INDICA.

1. Flowering branch (male plant) natural size.

2. Male flower, the petals removed, showing the 3-lobed calyx and insertion of the anthers.

3. An expanded flower showing the petals and stamens.

4. Corolla and calyx detached, the stamens more highly magnified, to show the form of the anther.

5. A fruit cut transversely.

6. A seed.

For this figure and dissection the draftsman is wholely responsible. It was executed while I was in England, and 1 have had no opportunity of comparing it with living plants to ascertain its accuracy. I suspect however there is an error in making it a monœcious plant.

EXPLANATION OF PLATE 104.

TRICHOSANTHES PALMATA.

1. Flowering branch of the male plant, natural size.

2. A flower split open to show the insertion of the stamens.

3. Stamens detached.

4-5. The same split open, showing; the long cells of the anthers, winding upwards and downwards on the back of the connectivum.

6. The staminal column cut transversely, showing the anthers one-celled.

EXPLANATION OF PLATE 105.

TRICHOSANTHES PALMATA.

1. Flowering branch of the female plant.

2. A flower split open showing 4 sterile anthers. This number is not uniform.

3. Ovary, style and stigma.

4. Ovary cut vertically, through some oversight the more important transverse section is not given.

5. A mature fruit, natural site.

6. The same cut transversely, showing that the placenta at this period contract adhesions to the parietes, which is not the case at first.

7. A seed with its pulp.

8. The pulp removed.

9. The same cut transversely.

10. A cotyledon with its minute radicle.

11. The detached radicle and plumula. I fear this last figure is not quite correct.

EXPLANATION OF PLATE 105* OR 121.

OBS. — This plate was prepared long after the others which will account for the error in the numbering. The object of it is to illustrate my remarks on some of the sections of Dr. Arnott's Conspectus.

CUCURBITA MAXIMA.

1. Female flower the corolla removed, showing the ovary, style, stigmas and disk.

2. Male flower corolla removed, showing the calyx and stamens, anthers like those of Trichosanthes.

3. Detached stamens.

4. Female disk, the style removed to show the sterile gland -like anthers.

5. Ovary cut transversely, showing the inverted carpels with the aid of colouring a little more distinctly than in nature.

6. The ovary cut vertically.

COCCINEA INDICA.

1. Female flower.

2. Male flower, corolla and calyx removed to show the stamens which correspond in form with the accompanying figures of section 7, but not with Cucurbita.

3. Style and stigmas.

4. Ovary cut vertically.

5. Cut transversely. The carpillary lines though apparently single, are actually double and do not cohere in the very early stages of the ovary.

6. A full grown fruit reduced in size.

7. The same cut transversely, showing the changes it undergoes in its progress towards maturity.

8. A seed.

9. The same slightly magnified.

10. (By mistake l6) a cotyledon showing the radicle at the base.

11 . A seed, testa removed showing the naked cotyledons.

MOMORDICA CHARANTIA.

1. Male flower and a detached petal, showing the stamens in situ.

2. Detached stamens.

3. Stamens separated showing the connective, not more lobed than in Coccinia.

4. Apex of the ovary, with calyx, styles and stigmas.

5. Ovary cut transversely.

6. A seed.

7. A cotyledon and radicle.

8. A seed cut transversely.

LUFFA PENTANDRA.

1. Male flower corolla removed, anthers undulating on the margin, but not lobed.

2. Female flower similarly dissected.

3. Ovary cut transversely.

4. A portion of it cut vertically.

5. A seed.

6. Cut transversely.

MUKIA SCABRILLA.

1. Male flower.

2. The same opened showing the anthers from within, not cohering, and abortive ovary.

3. Anthers different views.

4. Female flower.

5. The same, corolla removed.

6. A mature fruit.

7. The same cut transversely, 2-celled with 2 seed in each.

8. A seed somewhat smaller than nature.

9. The same magnified, rough on the surface.

10. Cut transversely.

11. Divided longitudinally.

12. Cotyledons, testa removed.

CUCUMIS TRIGONUS.

1. Male flower, split open.

2. Stamens detached, one of them separated and seen from within.

3. Female flower dissected, showing the styles and stigmas

4. A young fruit.

5. The same cut transversely.

6. A seed divided vertically.

7. Cut transversely.

LAGENAREA ULGARIS

1. Stamens as seen when the corolla is simply removed.

2. The same, the anthers separated to show their lobed form.

3. A detached stamen showing it lobed but not much more so than in Coccinea.

4. Ovary, style and stigmas.

5. The same cut transversely.

ZEHNERIA HOOKERIANA

1. Male flower, stamens free

2. The same split open, showing the insertions of the stamens and large abortive ovary.

3. (By mistake 8) detached stamens.

4. Female flower and ovary.

5. The same split open to show the styles and stigmas and abortive stamens

OBS. — This figure was prepared from dried specimens, but appears sufficiently perfect to show that § 5 and 6, might without inconvenience, perhaps with advantage be united so far as the stamens are concerned for the anthers of Mukia do not cohere though connivent and less distinctly free than those of Zehneria.

The plate as a whole amply I think establishes the position with which I started, that the sections are too numerous.

NHANDIROBEAE. CUCURBITACEÆ.

ZANONIA INDICA (LINN.)

CUCURBITEÆ. CUCURBITACEÆ.

TRICHOSANTHES PALMATA? (ROXB.)

CUCURBITEÆ. CUCURBITACEÆ.

TRICHOSANTHES PALMATA? (ROXB.)

CUCURBITACEÆ.