A dermatologist's guide to ants-in-the-plants
Adapted from: Schmidt RJ (1985) The super-nettles: a dermatologist's guide to ants in the plants. International Journal of Dermatology 24(4): 204-210
A widely distributed, well-defined, but not extensively studied group of higher plants comprises the stinging nettles. Because of their ability to inflict a more or less painful sting on human skin contact, such plants are a dermatological hazard (Evans & Schmidt 1980). Somewhat more dramatic are the "super-nettles" — plants that harbour ants. [The term "super-nettle" is taken from Brown (1960)]. In their natural habitat, such plants are analogous to the stinging nettles. Contact with the plant can elicit a more or less aggressive response from the ant inhabitants which, depending on species, rush out from within the plant to bite and/or sting the intruder. Like the stinging nettles, the super-nettles show a marked variability in their virulence; the skin reaction may vary from a slight irritation to a burning or deeply painful sensation, often accompanied by wheals or even blisters. Any such skin reaction may be classed as a pseudophytodermatitis; it is normally self limiting and, since very little is known about the chemical nature of most ant venoms, treatment if necessary is symptomatic.
Many dramatic accounts exist of late 19th Century and early 20th Century explorers and botanists being bitten and/or stung by super-nettles. There are also a few similar accounts published by contemporary botanists and ecologists studying the plants, yet case reports in the medical literature are conspicuous by their absence. There are several possible reasons, including the fact that the vast majority of the super-nettles occur naturally in tropical rain forests where few people live or venture. Secondly, most of the plants are unknown in cultivation, but a few instances have been recorded where local ants have inhabited such a plant being cultivated away from its normal habitat (Bequaert 1922, Teuscher 1974). With the increasing interest in the study of the ecology of the world's tropical rain forests, it may be supposed that man's encounters with super-nettles are likely to increase in frequency.
The most primitive ants in evolutionary terms are equipped with an extendable, unbarbed stinger enclosed in a sheath (which may be barbed) and connected to a venom gland. Stingers are found in the sub-families Myrmeciinae (found almost exclusively in Australia), Pseudomyrmecinae (compulsorily arboreal ants of Asia, Africa, and America), and Ponerinae (found mainly in tropical regions), as well as in the sub-families Aneuretinae, Cerapachynae, and Dorylinae that are not normally found associated with ant plants. In a further sub-family, the Myrmicinae, stinger reduction has occurred frequently, often independently among species of the same genus; thus only some Crematogaster and Pheidole species can sting, although the stings are usually feeble (Dumpert 1981).
Stinging ants may also have powerful mandibles with which they can bite. Many are known to bite and sting simultaneously, the bite apparently being necessary to gain mechanical advantage whilst the stinger is inserted (Beard 1963). The myrmeciine and pseudomyrmecine ants bear a definite resemblance to wasps (Dumpert 1981) — wasps, bees, and ants are closely related entomologically. The most virulent of the super-nettles house stinging ants of the sub-family Pseudomyrmecinae.
A second, somewhat heterogeneous, group of ants attacks primarily by biting with their more than usually powerful mandibles; these have been described as mandible fighters. Their stings are either poorly developed or non-existent (Dumpert 1981). Very few mandible fighters, with the exception of Pheidole species, are found associated with ant plants.
A third group of ants comprising the sub-families Formicinae and Dolichoderinae are equipped with chemical weapons in place of stings. The formicine ants are equipped with a formic acid spray that may travel a distance of upto 30 cm in some species; the dolichoderine ants secrete a wider variety of more or less aromatic chemicals (Dumpert 1981, Beard 1963). These ants equipped with chemical weapons are also capable of biting. Since the secretions of the dolichoderine ants are essentially harmless to man (Dumpert 1981), these ants are in effect only mandible fighters. The formicine ants, however, are noted for their strategy of biting and at the same time curving their gasters over their heads and spraying the wound with formic acid (Wheeler 1921, Weber 1943, Smith 1973). The formic acid may thus reach the intra- and sub-dermal tissues through the wound made by the mandibles (Beard 1963). The reaction is sharply painful but otherwise of little importance (Smith 1973).
It is evident, therefore, that several types of trauma may be effected by the super-nettles. At present there is no specific treatment available for the bite and/or sting of any particular species of ant, and thus the identity of the ant is of little consequence. Myrmecologists have however identified many of the ant species inhabiting various ant plants; interested readers are advised to consult the original literature from which this review was prepared.
Because of the close relationship between bees, wasps, and ants, it is rational to treat ant stings in the manner of the far more common (and occasionally fatal) wasp stings. The venoms of ants, wasps, and bees are allergenic (Beard 1963) and may therefore induce anaphylactic reactions (Smith 1973). A small probability exists of a person having or acquiring hypersensitivity to ant stings. In the unlikely event of an ant sting in the oral mucosa, there is a possibility of severe oedema obstructing respiration (Frankland 1963).
There are definite qualitative differences between the local reactions elicited by the stings of different ant species, and the local reactions produced by common wasps and bees. No specific palliative treatment (except for jumping in the nearest river — see Wheeler 1942) has yet been recommended in the literature. In the absence of infection, which has proved fatal following wasp stings (Marshall 1957), local reactions to super-nettle stings may be regarded as self-limiting.
The super-nettles are myrmecophilous plants (Greek myrmex = ant; philos = friend) but are often described as ant plants or myrmecophytes. Myrmecophily in plants can take several forms (after O. Warburg 1892, cited by Bequaert 1922):
One variant of the myrmecodomic type of association is the "ant garden", a structure to be found on trees in tropical forests. An ant garden is an arboreal ant nest that has been built amongst the aerial roots of an epiphyte or group of epiphytes. According to Bequaert (1922), the botanical collector in the tropics soon learns of the partiality of ants to such sites for nesting purposes. The interrelationships between ants and epiphytes are not merely accidental, but a result of mutual adaption (Benzing 1970, Janzen 1974). The species of plants found growing in ant gardens are determined in part by the species of ants involved, and in part by the species of plants to be found in any particular locality. In the tropical rain forests of S.E. Asia, Indomalaysia, and northern Australia, associations of Dischidia (fam. Asclepiadaceae), Hydnophytum and Myrmecodia (fam. Rubiaceae), and Lecanopteris (fam. Polypodiaceae) species are commonly found as ant gardens harbouring ants of the genera Iridomyrmex or Camponotus (Bequaert 1922, Janzen 1974, Docters van Leeuwen 1929, Horwood 1968). Janzen (1974) states that the Iridomyrmex species is relatively unaggressive, but if the colony is disturbed by such a thing as ripping one of the myrmecophytes off its substrate, the ants swarm out over the offending hand and do persistently attack. They cannot sting but have a very potent defence against eyes, nose, and mouth; they release an unidentified chemical that feels roughly like very concentrated acid when applied to mucous membranes. O. Beccari (cited by Bequaert (1922)), on the other hand, noted that the inhabitants of Myrmecodia species act as a formidable army of defence, for no animal dares to meddle with a plant guarded by a host of biting ants ready to assault the imprudent invader in myriads. Horwood (1968) acknowledged that the Camponotus species have a fierce and painful bite.
Ridley (1910) reported that in the region of Singapore, ants of the genus Dolichoderus nest among the aerial roots of the pigeon orchid (Dendrobium crumenatum Sw., fam. Orchidaceae) and also under the epiphytic fern Platycerium biforme Blume (syn. Platycerium coronarium Desv.). These ants are small but bite viciously; the bite, though irritating. is not severe. A viciously stinging ant of the genus Odontomachus nests under another common Malayan epiphytic fern Asplenium nitidus L., otherwise known as the bird's nest fern. The pain produced by the sting of this ant was described by Ridley (1910) as long-lasting. Ferns of the genus Platycerium growing on trees in Queensland provide nesting sites for ants of the genus Myrmecia whose stings are dreaded by the natives (Bequaert 1922). Ferns of the genus Lecanopteris (formerly Polypodium) growing in Malaya are associated with ants of the genera Technomyrmex and Crematogaster. Such ant-epiphyte systems were described by Yapp (1902) who noted that attempting to remove Polypodium carnosum Christ (syn. Lecanopteris carnosa Blume) from a tree resulted in great numbers of its tiny inhabitants sallying forth and attacking the intruder with great pugnacity. In contrast, Jermy and Walker (1975) noted that the ants inhabiting Lecanopteris species in Celebes tend to flee rather than attack when disturbed. It is evident that the virulence of the Indomalaysian ant epiphytes as super-nettles is very much dependent on the species of ant involved.
The ant gardens of tropical America are rather more varied. Horich (1977) has described the tropical American orchid ant gardens in some detail. He noted that Epidendrum dolabrilobum Ames & Schweinf. (fam. Orchidaceae), a Costa Rican species, is always associated with ants as are many Catasetum, Gongora, and Stanhopea species. J. Rodway (cited by Bequaert 1922) mentioned that in British Guiana, many epiphytic orchids, especially of the genera Coryanthes, Gongora, and Oncidium shelter large communities of ants in the oval masses of their fibrous roots. He noted that collectors find it very difficult to dislodge the plants without being severely bitten. Teuscher (1974) states that the ants that infest species of Coryanthes are quite formidable because they bite furiously and painfully. Coryanthes speciosa Hook., the most common species, is especially known for this and collectors are always wary of this orchid. One needs only to lean against a tree that harbours a plant of Coryanthes speciosa and the ants come storming out ready for war (Teuscher 1974). Horich (1977) claims that all Coryanthes species are actually associated with somewhat more complex structures that he calls "beehive nests". Not only is there a standing army of one or more species of ant nesting amongst the roots of these orchids, but also an air force of excitable wasps, bees, or hornets. If one of these seemingly harmless root balls is disturbed, hundreds of insects pour out; it becomes an untouchable heap of ants surrounded by a menacing cloud of hornets (or whatever species happens to be cohabiting with the ants) that attacks the intruder in swarms. According to Horich (1977), attempting to dislodge a beehive nest orchid from a tree is an unforgetable experience. Ants seem to occupy such beehive nests for many years but the wild bees, wasps, or hornets may only be temporary guests since they are not always present. Beehive nests have also been found in Epidendrum imatophyllum Lindley and Epidendrum schomburgkii Lindley (Horich 1977).
Several tropical American epiphytic orchids are myrmecodomic in the true sense, having hollow pseudobulbs that provide nesting sites for ants. Such plants include species of Caularthron (syn. Diacrium) and Schomburgkia (Horich 1977, Jones 1966). The ants that nest in these orchids belong to the genera Azteca, Crematogaster, Monomorium, and Neoponera (Bequaert 1922, Wheeler 1942).
Another common feature of the epiphytic flora of tropical America are the Tillandsia species of the family Bromeliaceae. The plants form bulb-like rosettes, the innermost cavities of which contain water. The outer cavities harbour ants (Wheeler 1942). Only one species of ant seems to occupy a single rosette, but several species from many genera of ants have been found nesting in Tillandsia species. Benzing (1970) notes that the ants often exhibit aggressive behaviour when the plants are disturbed, and he states that he was bitten on several occasions whilst studying Tillandsia species.
As in the Indomalaysian ant gardens, the constituent epiphytes in those found in tropical American rain forests are apparently selected by the ant species involved. Ants of the genus Azteca produce ant gardens comprising Philodendron myrmecophilum Engl. (fam. Araceae), Nidularium myrmecophilum Ule (fam. Bromeliaceae), Ficus myrmecophila Warb. (fam. Moraceae), Markea formicarum Dammer (fam. Solanaceae), Ectozoma ulei Dammer (fam. Solanaceae), Codonanthe formicarum Fritsch (fam. Gesneriaceae), and other unidentified members of the family Gesneriaceae (Wheeler 1921). Somewhat larger ant gardens comprise the following species: Philodendron myrmecophilum, Anthurium scopolodendrinum Kunth var. poiteauanum Engl. (syn. Anthurium gracile Lindley, fam. Araceae), Streptocalyx angustifolius Mez (fam. Bromeliaceae), Peperomia nematostachya Link (fam. Peperomiaceae), Aechmea spicata Martius (fam. Bromeliaceae), Codonanthe uleana Fritsch (fam. Gesneriaceae), and Phyllocactus phyllanthus Link (fam. Cactaceae). With the exception of the Anthurium and Phyllocactus species, the above ant epiphytes are so intimately associated with ants that they are not found in the Amazon Basin in any other situation (Bequaert 1922). These large ant gardens are almost always inhabited by two species of ants; a Crematogaster species occupies galleries in the outer parts of the nests, whilst a larger and more aggressive species of Camponotus occupies the innermost galleries. If such an ant garden is slightly disturbed, the Crematogaster species rush out in great numbers to attack the intruder (Wheeler 1921). The Crematogaster, though small, are so numerous, active and vicious as to be annoying. They possess a sting that is too small to pierce human skin in most exposed places unless given plenty of time, but a bite that may be felt (Weber 1943). If the nest is rudely shaken or broken up, the formidable battalions of the Camponotus rush to the surface (Wheeler 1921). These ants are much larger and have strong mandibles but no sting. Like other formicine ants, they bite and at the same time curve forward the tip of their abdomen and spray the wound with formic acid so that a person may form the hasty impression of being stung (Weber 1943).
The myrmecoxenic potato-ferns of the tropical American rain forests have quite recently been described (Gómez 1974, Wagner 1972). These ferns (Solanopteris and Microgramma species, fam. Polypodiaceae) grow epiphytically in association with other ferns, aroids, bromeliads, and orchids, and are thus constituents of ant gardens. They possess hollowed tuber-like modified stems that are frequently inhabited by ants of the genera Azteca and Camponotus, less commonly with Pheidole and Solenopsis species. Gómez (1974) noted that if an ant inhabited specimen was disturbed, Azteca ants rushed out in large numbers and fiercely defended their nest and tubers. If the disturbance was increased, the Camponotus would join the defences. Pheidole and Solenopsis would only swarm out if attacked or disturbed directly. Allen (cited by Wagner 1972) gave a more vivid account. He noted that the small, but exceedingly bellicose species of ant inhabiting the hollow tubers would make itself known on the slightest disturbance of the plant. The ants race at top speed over one's protesting anatomy until fortune favours them with a particularly succulent spot to clamp on to, which they do with a vengeance.
As well as the plants that grow on ant gardens, a large number of other species are known to harbour ants within hollow branches, fistulose swellings, hollow thorns, and other such myrmecodomatia. Members of over 40 plant families are known to be myrmecophilous, but with the exception of those described below, the majority have not been described as being hazardous to handle. This may be because of their rarity or inaccessibility in their forest habitats, or otherwise because they house relatively unaggressive or even timid ants.
There is only one noteworthy ant plant in this family, and that is Epitaberna myrmoecia Schumann which occurs in tropical West African rain forests. The branches have caulinary swellings inhabited by a particularly vicious ant of the genus Pachysima (see also Barteria spp., fam. Passifloraceae). G. Tessmann (cited by Bequaert 1922) described how the Pangwe tribe of Southern Cameroon and Spanish Guinea used this plant in their religious ceremonies. During the initiation of the Sso-cult of the Yaunde, the candidates are obliged to pass through several tests, one of which involves crawling through low huts in which nests of stinging ants and branches of Epitaberna myrmoecia are hung or placed. Pods of Mucuna pruriens DC. (fam. Leguminosae) covered with dangerously itching hairs are also thrown inside. The candidates are, of course, fearfully stung.
Several species of Cordia have spindle-shaped myrmecodomatia at the tips of their branches. Wheeler and Bequaert (1929) described a species of Neoponera that forms small colonies in the myrmecodomatia of Cordia hispidissima DC. They describe the ant as being rather timid and note that it does not try to defend itself when disturbed; they do however state that it is a very active ant which stings rather severely when handled. Wheeler (1942) described colonies of an Allomerus species on Cordia nodosa Lam. Each colony comprises thousands of workers and normally occupies all or most of the caulinary swellings of the bush. All of the cavities in the swellings are connected with one another and with the forest floor by a peculiar system of galleries or arcades (measuring 5—10 mm in diameter) constructed on the surface of the plant by the worker ants. The galleries consist of minute particles of earth built up around, and supported by, the long red hairs that cover the plant. These ants thus live on the Cordia but forage amongst the dead leaves on the forest floor for their food. Wheeler (1942) also states that when one handles the plant for some time, the worker ants swarm over one's clothes and body and for some time keep on stinging; their stings are so feeble that they produce merely an unpleasant itching, and that only on parts with very thin epidermis. A third species, Cordia alliodora Cham., may be found inhabited by either an Azteca species or a Pseudomyrmex species. Wheeler (1942) encountered principally Azteca inhabited specimens, and noted that these ants were small, timid, and lethargic, and would bite only infrequently. However, Wheeler (1942) also acknowledged that other botanists have described the presence of very vicious and fierce ants on this species — they evidently encountered Cordia alliodora inhabited by Pseudomyrmex species.
Myrmecophytic Cordia species are found in the tropical American rain forests (Wheeler 1942).
Several genera in this family are known to contain myrmecophytes. These include Alchornea, Endospermum, Mabea, Macaranga, and Sapium. Furthermore, the ants inhabiting these species belong to genera that are known to contain aggressive biting or stinging species. Rickson (1980) studied the hollow-stemmed Macaranga species of S.E. Asia, which he noted were inhabited by ants of the genus Crematogaster. He found that these ants bite as readily as do the ants of Cecropia (fam. Urticaceae) and Acacia (fam. Leguminosae) species when the plants are handled.
Probably the best studied group of myrmecophytes comprises the "swollen thorn acacias" (Acacia species) of tropical America. They are unusual among myrmecophytes since they occur in hot, dry regions and not in tropical rain forests. There are also a number of East African acacias that harbour ants. These species also grow in hot and dry regions, but differ from the American species in that the ants that occupy their thorns are not obligate tenants as are the ants in the tropical American species. The American acacias are almost always inhabited by aggressively stinging ants of the genus Pseudomyrmex, whilst the African species house ants of the genus Crematogaster.
There is unanimous agreement that the Pseudomyrmex acacia ants of tropical America are very aggressive and sting fiercely (Brown 1960). In 1874, T. Belt (cited by Wheeler 1942) noted that if one of the swollen thorn acacias be touched, or a branch shaken, the little ants swarm out of the hollow thorns and attack the aggressor with jaws and sting. They sting severely, raising a little white lump that does not disappear in less than 24 hours. P.C. Standley (cited by Record and Hess 1943) reported that the ants protect the plant efficiently against animals which might molest them, and sally forth full of fight when the branches are jarred. They bite severely and painfully and are perhaps the most ferocious of all Central American ants. (It would appear that Standley, as did many other authors, failed to recognise that the ants inflict a sting whilst biting). The stinger of the Pseudomyrmex is not barbed but, when attacking mammals, the barbed stinger sheath is inserted to hold the ant in place (Janzen 1966). The ant acacias are thus typical super-nettles. Their virulence does however vary with the weather — in cool conditions or in heavy rain the ants are almost inactive (Janzen 1967). The Crematogaster acacia ants of East Africa are also very aggressive, but rely on their bite as a weapon of attack (Brown 1960).
Many other plants in the family Leguminosae are known to be ant plants, but their reputations are rather more local. The name of the South American genus Tachigalia was derived from the word tachi meaning stinging ants or ant nest, a word employed by the Indians of northern South America. These low growing trees bear petiolar sacs that provide nesting sites for ants of the genus Pseudomyrmex (Wheeler 1942).
Of the four species of Barteria in Africa, Barteria dewevrei De Wild. & T. Durand and Barteria fistulosa Masters are known to be myrmecophytes (Bequaert 1922, Janzen 1972). Barteria fistulosa is a small tree that grows in the tropical rain forests of West Africa. A fully grown specimen may be occupied by 1000—4000 ants of the genus Pachysima. The ants nest in the hollow branches into which they gain access through specially gnawed holes. These ants are not particularly sure-footed and frequently fall from the branches. There is thus a slow rain of these ants from the crown of the tree, and this increases somewhat if the colony is disturbed (Janzen 1972). The natives of the forest are familiar with the plant and aware of its being inhabited by very aggressive ants. They usually avoid the unpleasant task of cutting this tree when clearing ground (Bequaert 1922).
The ants are an extremely effective protective agent against animals, and especially against man. On venturing under the tree, the intruder cannot avoid the slow rain of ants falling from the branches. Once they land on the intruder, the ants do not immediately bite or sting, but walk about until a bare patch of skin is found. They then attack by grabbing on with their mandibles and proceed to bite and sting; they are very difficult to remove. The pain from a Pachysima sting is not felt for several seconds after the stinger has been inserted. A deep throbbing pain then develops, which lasts for 1—2 days during which time the muscle is sore and stiff (Janzen 1972). A similar story was told by H. Kohl (cited by Bequaert 1922), who noted that the ants are extremely pugnacious and always ready for a fight as they are equipped with excellent weapons, their stings and mandibles. If a Barteria tree is roughly handled, innumerable hosts of the ants rush out of all openings and woe to him who approaches them closely. The pain of the stings spreads instantly over the whole of the affected limb and continues for a long time, and on the following morning returns with full intensity during one's ablutions. Kohl also recorded that he was told by one of his black servants that it was customary in his part of the country to punish unfaithful wives by tying them to plants inhabited by the ants. According to Janzen (1972), hundreds of Pseudomyrmex stings from Central American ant acacias (see Leguminosae) can be tolerated if there is a compelling reason to invade the tree, but 1—5 Pachysima stings render the recipient very reluctant to return to the tree.
The same ant species has been found inhabiting the hollow lateral branches of the less common Barteria dewevrei as well as the hollow swellings of the unrelated Epitaberna myrmoecia Schumann (see Apocynaceae) (Bequaert 1922).
The most dramatic of the super-nettles must surely be the Triplaris species of Central and South America. The earliest accounts of man's encounters with super-nettles relate to these plants. In 1653, P. Bernabé Cobo (cited by Wheeler 1942) related that the tree (a Triplaris species) is hollow throughout the trunk to the slenderest twigs, and full of large yellow ants, so virulent that their sting is apt to bring on a fever and is always exceedingly painful. If a single leaf be touched, so many of the ants swarm forth from all parts of the tree as to excite wonder, and they assail the person who touches the tree and, if he does not withdraw in time, martyr him with their stings. In 1775, J.B. Aublet (cited by Wheeler 1942), writing about Triplaris americana L., noted that when one fells the tree, one is at once completely covered and cruelly tormented by the ants. The only way to get rid of them is to throw oneself into the water (by which the trees invariably seem to grow). J. Von Spix and C. Von Martius also encountered Triplaris americana and wrote in 1831 (cited by Wheeler 1942) that on breaking off a branch, a scrambling mass of viciously biting [actually biting and stinging] enemies pours down and leaves many blisters on the skin. Perhaps the most vivid account of the consequences of meddling with these plants was presented by R. Schomburgk in 1848 (cited by Wheeler 1942). He noted that whilst trying to break off one of the boughs, thousands of yellowish brown ants rushed out of the small openings in the internodes, completely covered him and in the greatest fury seized his skin with their jaws and, vomiting a white fluid, buried their terrible stings in his muscles. Thousands more poured out from openings in the stump and also rained down from the tree into his boat. The whole crew including a few tame apes which freed themselves with wild leaps from their tethers, promptly jumped into the river, for this was the only way in which they could escape the savage onslaughts of the ants. Schomburgk confessed that a secret horror crept over him thereafter whenever he passed one of the trees. He also recorded that the swelling, inflammation, and pain of the excruciating stings may persist for several days. E. Ule in 1907 (cited by Wheeler 1942) described the bite [actually the bite and sting] of the ants inhabiting Triplaris schomburgkiana Benth. and Triplaris surinamensis Cham. & Schltr. as being very painful, burning as if one had come into contact with red hot iron, and sometimes producing blisters on the affected parts of the body. Several other reports concerning the ants in Triplaris species were published in the 19th and early 20th Centuries (cited by Wheeler 1942).
The ants in Triplaris species have been identified as species of Pseudomyrmex (Wheeler 1942). One particular species that was found by E. Ule (cited by Wheeler 1942) in Triplaris schomburgkiana characteristically clears the ground of vegetation around the tree, a phenomenon that may also be observed around Acacia species (fam. Leguminosae), Tachigalia species (fam. Leguminosae), Barteria species (fam. Passifloraceae), Macaranga species (fam. Euphorbiaceae), and Cecropia species (fam. Urticaceae), although different species of ant are involved in these cases (Bequaert 1922, Wheeler 1942, Ridley 1910, Janzen 1966, Janzen 1969).
During his studies on Triplaris americana, Wheeler (1942) encountered specimens harbouring various Pseudomyrmex species, some more aggressive than others. In some areas, specimens could be handled and examined with impunity. These specimens were found to be inhabited by small inoffensive (to humans) species of Azteca.
Several other genera in this family, including Coccoloba (syn. Campderia), Symmeria, and Ruprechtia, all of which are found in tropical Central and South America, are known to harbour ants in their medullary cavities (Wheeler 1942), but these do not have the reputation of the Triplaris species as super-nettles.
The genus Cecropia from South America contains several myrmecophytes. The ants, most commonly of the genus Azteca, reside within the hollow stems into which they gain access by gnawing entrance holes through the walls of the stems. In 1874, T. Belt (cited by Wheeler 1942) noted that if the tree be shaken, the ants rush out in myriads and search about for the molester. Bailey (1922) also found that when a myrmecophytic Cecropia species is shaken or touched, the angry and aggressive Azteca ants rush out from their nest and swarm over the whole plant. Janzen (1969) reported that the ants displayed aggression against humans cutting the Cecropia species, and K. Fiebrig (cited by Bequaert 1922) noted that the Azteca ants of Cecropia adenopus Miq. go only short distances from their exit holes unless disturbed, when they become very aggressive.
The fact that the Azteca ants cannot sting, but rely on their bite (Janzen 1972) renders the myrmecophytic Cecropia species rather less noxious than many other super-nettles.
Two closely related genera, Clerodendrum and Vitex, are known to contain myrmecophytes. One species, Vitex staudtii Gürke, which is a creeper that grows in tropical West Africa, has been studied by Bequaert (1922). He noted that when the stems of this plant are slightly touched or otherwise disturbed, large numbers of slender, reddish ants rush out of the hollow stalks ready to attack. The ants are invariably a species of Viticicola. They are exceedingly vicious and alert. Their sting is extremely painful and sometimes produces vesicles on the skin. It is certain that they constitute a very efficient body-guard of their host.