29

ANT-SUPPORTING PLANTS.

BY JAMES BRITTEN, F.L.S.,

BOTANICAL DEPARTMENT, BRITISH MUSEUM.

[PLATE CXVIII.]

مه

THE

THE relations between the animal and vegetable worlds, and the mutual dependence of one upon the other, offer some of the most interesting problems in natural science. The subject is one which has attracted special attention of late years, and the numerous observations of competent and trustworthy persons have rendered us familiar with many striking illustrations of it. The absolute dependence of certain plants upon particular insects for their fertilisation has been amply demonstrated; and it is now shown that insects may, in certain cases, serve the requirements of plants in another way, by supplying them with food. And this dependence extends more widely than might at first sight appear. Mr. Darwin gives an instance of this when speaking of the fertilisation of the red clover by the visits of the humble-bee. "The number of humble-bees in any district depends in a great degree on the number of fieldmice, which destroy their combs and nests; . . . the number of mice is largely dependent, as every one knows, on the number of cats. . . . . Hence, it is quite credible that the presence of a feline animal in large numbers in a district might determine, through the intervention first of mice and then of bees, the frequency of certain flowers in that district."*

It is not, however, only as fertilisers or as food-suppliers that insects are connected with plants. Botanists are well acquainted with the fact that certain trees and plants offer homes to various species of ants; but this is not very generally known, and, so far as I am aware, no attempt has been made to bring together what has been recorded on the subject in scattered papers, or to collect the references, often merely incidental,

"Origin of Species," ed. i. p. 74.

which have been made to the various ant-tenanted plants. I have for some time been collecting notes upon this subject, and have published a short paper upon one or two points connected with it.* Since then, however, so much additional material has come under my notice, that a brief résumé of the whole matter may be of some interest.

At first sight there might appear to be little remarkable in the residence of insects in trees. Many insects, of course, pass a large portion of their life in this situation; but in the instances to which I am about to refer, it would appear that the whole existence of the ants is passed in these vegetable homes; or rather, that when the ants have once taken up their abode in them, a colony is formed, which is only dispersed by the destruction of the tree. Not only is a dwelling-place found, but in some instances a means of support as well; and all this not only without injury, but even with absolute benefit to the tenanted plant.

These ant-homes are found not only in the hollow trunks of certain trees, where their presence would be less remarkable, but also in tubers, leaves, and thorns. I shall consider each of these separately, commencing with the tubers, the special adaptation of which to the requirements of the ants is truly wonderful.

By far the most striking instance of this is given by Myrmecodia tuberosa, to the very existence of which it would appear essential that its tuber should be tenanted by these insects. Rumpf, in the "Herbarium Amboinense," published in 1750, appears to have been the first to figure and describe this plant, under the name of Nidus germinans formicarum rubrárum, which he terms "mirum prodigium naturæ." He seems to have been uncertain whether the whole was a vegetable, or whether the tuber was an ant's nest from which the plant sprung: he says it is to be regarded as a zoophyte among vegetables! His account does not seem to have attracted any attention until about 1825, when the plant was described by Dr. Jack as a genus of Rubiaceae, under the name it now bears. It presents the form of a large irregular tuber, growing on the branches of old trees; from this spring a few thick fleshy stems, having a small number of smooth, leathery, oblong leaves crowded together at their summits. The small white sessile flowers are situated at the base of the petioles, and are almost concealed by the large persistent stipules. The tuber is tenanted by small and very fierce red ants, which rush out upon the intruder if their dwelling is attacked. The way in

• “Field,” Feb. 7, 1874.

"Trans. Linn. Soc.," xiv. 122.

which these ants take possession of the Myrmecodia, and the intimate relation which exists between the plant and the insect, are thus referred to in Professor Caruel's recent paper upon the genus.* The account is quoted from a manuscript note by Dr. Beccari,† who collected the plant in Borneo :

"I have carefully followed the development of this tuber, having been able to observe the young plants in all stages of growth from the period of germination. The seed is surrounded by a viscid pulp, resembling that of our mistletoe, which readily attaches itself to the branches of the trees upon which it falls. Its dissemination is probably caused by means of the birds which eat the fruit, the undigested seed passing through them and adhering to the branches. The seed soon germinates and unfolds its cotyledons, especially if it has fallen in an opening of a branch where lichens have collected, or if it be placed in mould; the stem developes itself to the length of from three to six millimetres, widening towards the base, acquiring a somewhat conical shape, with the two cotyledons at its apex. In this condition it remains until a particular species of ant burrows a small lateral cavity at the base of the stem; if this does not happen, the stem does not develope itself, and the plant dies. The wound caused by the bite of the ant determines a great development of cellular tissue, in the same way as the sting of the cynips causes the galls on the oak. The tuber now enlarges and the stem developes; the ants soon find sufficient space for forming a colony, and excavate galleries in the interior of the tuber in all directions, thus making for themselves a living habitation—a circumstance which is necessary to the existence of the plant. The plant could not live or even arrive at maturity unless the ants contributed to the formation of the organ which must be the source from which it derives its support, while in all probability the ants could not exist or propagate themselves unless they had discovered this mode of constructing so ingenious an habitation. The fleshy substance of this formicarium is formed of cellular tissue; the channels and galleries with which it is perforated have their entrance near the lower part of the tuber."

The genus Myrmecodia was formerly regarded as exclusively Malayan; but it is represented in Java by another species (or perhaps a form of M. tuberosa); and fine Australian specimens which perhaps belong to a new and undescribed species, have been recently received at Kew from Mr. Hill, of Brisbane. They greatly resemble a wasp's nest in external appearance, being of

• "Nuovo Giornale Botanico Italiano,” iv. 170—176 (1872).

This author has since described a new species from Celebes under the name of M. selebica in "Nuovo Giornale Botanico Italiano," vi. 195 (1874). It is intermediate between Myrmecodi and Hydnophytum.

a slaty-gray colour; and the galleries with which they are intersected in all directions are lined by the ants with a thin papery material.

The allied genus Hydnophytum, which differs from Myrmecodia in having a smooth tuber and small deciduous stipules, was also figured and described by Rumpf (as Nidus germinans formicarum nigrarum), and named and redescribed by Jack at the same time as Myrmecodia. In the structure of the tuber and its mode of use by the ants, it seems very similar to Myrmecodia, only that the best known species, H. formicarum, is inhabited not by red but by black ants. Three or four species are described, natives of Tropical Australia, the Fiji Islands, and the Indian Archipelago.

Passing now to those trees, the trunks and branches of which are used as habitations by ants, the most remarkable are those belonging to the South American genus Triplaris, all or nearly all the species of which are described by Meisner in Decandolle's Prodromus as "arbores ramulis fistulosis formicis hospitium præbentibus." Aublet, in his "Histoire des Plantes de la Guiane Françoise," published in 1775, figures and describes a species (doubtfully referred by Meisner to T. surinamensis, Cham.) under the name of T. americana, of which he says: "Ants are found in abundance in the interior of the trunk, branches, and twigs of this tree, in such a manner that when one strikes or cuts it, one is soon quite covered and acutely tormented by them, an accident which I myself have experienced. The only means by which one can get rid of them is to plunge oneself in water." Triplaris is a genus of Polygonacea, forming with Ruprechtia and the monotypic genus Podopterus the sub-tribe Triplaridea; it is characterised by the remarkable shuttlecock-like appearance of the fruit, an appearance caused by the three outer lobes of the perianth growing out into erect membranous wings, and reminding one very forcibly of the fruits of Dipterocarps. The most detailed account of the ant which infests these trees is that given by Weddell,* who says:

"The trunk, the branches, and even the smallest branchlets of the species of this genus are hollow, and serve as habitation to a peculiar species of ant, which gives off when excited a somewhat agreeable scent, resembling that of the Cicindelea. If one happens to touch the trunk of a Triplaris accidentally, especially if it is shaken, the ants rush out by hundreds from the interior of the tree through the small canals by which the medullary canal communicates with the exterior, and if escape is not made as quickly as possible one is covered with these

* "Ann. Sc. Nat. (Bot.)," 3rd series, xiii. 262-267.

dangerous guests, the bite of which is much more painful in proportion than the stings of any other insect with which I am acquainted. It is a singular thing that, at whatever stage of their existence one examines the Triplaris in the forests, one is always certain to encounter these ants. It is still more curious that in Ruprechtia, which some authors unite with Triplaris, they are never met with. I do not think that this

insect has ever been observed in other conditions than those which I have noted; its linear form is especially adapted to its mode of life. I have had occasion to examine it and indeed to suffer from its attacks in many parts of Brazil, in Bolivia, and in Peru, and it has everywhere appeared to me identical. Many travellers have already recorded a portion of the facts in question, and have referred the ant of the Triplaris to Latreille's genus Myrmica, but I am not aware that it has received a specific name; that of triplarina may be applied to it. It is usually of a clear brown; its length is six or seven millimetres, and its breadth one millimetre; the abdomen is cylindrical and a little attenuated towards its lower extremity, which is hairy." The ants swarm especially in two or three species, notably in T. nolitangere (so named by Weddell on this account), which is called Formigueira by the Brazilians, T. Bonplandiana, and T. Schomburgkiana. This last species is described at length by Dr. Schomburghk, but his description adds nothing of importance to what has been already cited; he remarks that the different tribes of Indians in Guiana call it by names which signify "the ant-tree."

*

The well-known Trumpet-tree (Cecropia peltata) is also antinhabited, a fact to which Mr. Belt directs attention in his interesting "Naturalist in Nicaragua." As in Triplaris, the trunk is hollow, and provided within with partitions answering to the position of the leaves on the outside, and it is in the spaces between these partitions that the ants congregate. They "gain access by making a hole from the outside, and then burrow through the partitions, thus getting the run of the whole stem. They do not obtain their food directly from the tree, but keep brown scale-insects (Coccida) in the cells, which suck the juices from the tree, and secrete a honey-like fluid that exudes from a pore on the back, and is lapped up by the ants. In one cell eggs will be found, in another grubs, and in a third pupa, all lying loosely. In another cell, by itself, a queen ant will be found, surrounded by walls made of a brown waxy-looking substance, along with about a dozen coccidæ, to supply her with food. . . . If the tree be shaken, the ants rush out in myriads, and search about for the molester. . I have cut into some

.

• "Annals and Magazine of Natural History," i. 264.