UConn Home Banner
HOME COLLECTIONS TEACHING RESEARCH VISITING STAFF

Cultivation of the epiphytic ant-plants Hydnophytum and Myrmecodia

This article was originally published in Cactus and Succulent Journal 72:142-147 (2000).
It is posted here with permission of the author & editor.

INTRODUCTION

When we think of interactions between insects and plants in cultivation, we tend to focus on pollination or the depredations of herbivores. In nature, however, those interactions can be far more complex. Some of the most interesting symbiotic relationships involve ants and specialized plants known as myrmecophytes which possess structures that house ant colonies. In some cases, the plants also provide food for the ants in the form of nectar or protein-rich Beltian bodies In exchange, the myrmecophytes receive nutrients or protection from herbivores. Two genera of myrmecophytes, Hydnophytum and Myrmecodia, are of particular interest to caudiciform enthusiasts.

ANATOMY AND ECOLOGY

Hydnophytum and Myrmecodia are members of the Hydnophytinae, a subtribe of the coffee family, Rubiaceae (Huxley and Jebb, 1991). Most species of both genera are epiphytes Figure 1, though a few grow as terrestrials in nutrient-poor highlands.

Figure 1: Myrmecodia tuberosa in Sarawak.

As seen here, Myrmecodia plants frequently grow with the tuber hanging down. Photograph by Wilson H.P. Chua. All remaining photographs by the author.

In Hydnophytum and Myrmecodia, as well as the related genera Anthorrhiza, Myrmephytum, and Squamellaria, the base of the stem develops into a large tuber. Tissue dies back within the tuber to form a series of galleries connected to entrance holes on the surface. In some cases, tiny pores further ventilate the chambers. The chambers provide excellent nest sites for ants, and the vast majority of Myrmecodia plants examined in the wild are inhabited. Hydnophytum are somewhat less likely to be inhabited, especially at high altitude (Huxley 1978). The primary benefit of the ant inhabitants to the plant is nutritional. The ants do not seem to protect the plants against large predators (such as botanists), and it is unclear whether the plants are protected against herbivorous insects (Janzen 1974). Experiments with radioactive tracers, however, have shown that warts within the chambers are capable of absorbing nutrients from the ants' waste, mostly leftover insect parts (Huxley, 1978; Rickson, 1979). In Myrmecodia, the internal galleries are divided into dry, smooth-walled chambers that the ants inhabit and warty chambers where they dump their garbage. The galleries of Hydnophytum are less specialized, but the warts tend to be clustered at the ends of the chambers (Huxley 1978).

Despite their close relationship, it is easy to distinguish Hydnophytum from Myrmecodia. Myrmecodia tend to have just one or a few thick, unbranched stems with large leaves (exceptions are M. lamii and M. beccarii). The stems of Myrmecodia are frequently covered with clypeoli, unusual shield-like structures surounding each leaf base. The clypeoli may be bordered with spines, and they give the stem a curious armored appearence. It should be noted, however, that in some Myrmecodia species, the clypeoli are obscure or absent. Myrmecodia flowers form in hollows called alveoli which develop along the stem between the clypeoli. Hydnophytum have thinner stems with longer internodes and smaller leaves. They often have multiple stems which can be branched. The stems lack clypeoli and alveoli.

Myrmecodia tubers often have ridges and wart-like bumps; one unimpressed family member told me that my myrmecodias look as though they have a nasty skin condition. The tubers are frequently covered with spines, actually modified roots, that can be simple or branched. In contrast, Hydnophytum tubers generally lack spines, although one species in cultivation does have ridges. Tuber shape in both genera is very variable, and color ranges from dark brown to silvery grey.

The range of Hydnophytum extends from Malaysia and the Philippines, south to the Cape York Peninsula in Queensland, and east as far as Fiji. It includes southern Thailand, Cambodia, and Vietnam. Myrmecodia has a similar but slightly more restricted range, with the highest species diversity in New Guinea (Huxley and Jebb, 1991; Huxley and Jebb, 1993). Unlike many epiphytes, ant-plants are quite rare in lowland rainforest. More typical habitat is open woodland where the plants are exposed to bright light; in myrmecophyte habitat, nutrient availability probably limits growth, while in deep forest, lack of light would reduce the benefit from additional nutrients provided by ants. The plants are found more frequently on branches than trunks (Huxley and Jebb, 1991) and can be very common in disturbed habitat such as plantations (Huxley, 1978). During the time that he studied Hydnophytum formicarum and Myrmecodia tuberosa at Bako National Park in Sarawak, Janzen (1974) noted that rain fell almost daily in the afternoon.

CULTIVATION

Hydnophytum and Myrmecodia are fairly large genera with 50 and 26 species respectively (Huxley and Jebb, 1991; Huxley and Jebb, 1993), but only a few are likely to be seen by amateur growers. In the United States, Hydnophytum plants commonly available from commercial nurseries fall into two categories which may be two variable species or two groups of closely related species. The first group consists of plants are usually labeled Hydnophytum formicarum (see Figure 2).

Figure 2: Hydnophytum species mounted on bark).


This is one of the plants commonly sold as Hydnophytum formicarum. The plant is mounted on cork bark with some osmunda fiber around the roots. The tuber is 11 cm (4.3 inches) long.

These plants have somewhat fleshy leaves with obscure veins. The tuber is usually globular or pear-shaped, and color can be grey or light brown. Large entrance holes frequently form on any part of the tuber. In my experience, these plants can be tricky to grow well and are quick to shed their leaves if they are allowed to dry out or become too cold. The second group are larger plants sold either as Hydnophytum species or Hydnophytum formicarum. They have narrower leaves with more prominent veins. The tuber is typically hemispherical, and entrance holes seem to be restricted to the flat underside. Often, the tuber has ridges which may be barely visible or very prominent (see Figure 3 and Figure 4).

Figure 3: Hydnophytum species with ridged tuber.

The leaves have more prominent veins than those of the plant in Figure 2. This is a three year-old seedling growing in an 8 inch bulb pan. The plant is 27 cm (10.6 inches) tall with a tuber diameter of 12.5 cm (4.9 inches).

Figure 4: Hydnophytum seedling with prominently ridged tuber

Compared to the plant is Figure 3, this seedling has narrower leaves and a more prominently ridged tuber. It remains to be seen whether these characteristics are retained at maturity. The tuber is 6.5 cm (2.5 inches) in diameter.

These plants seem to be easier to grow than the first group and are less likely to shed leaves when underwatered. Other species may be growing in botanic gardens, but they are not widely available to amateur growers. Unfortunately, I have not been able to positively identify any of my Hydnophytum plants, due to the difficulty of finding literature on the genus.

The genus Myrmecodia has been recently revised by C.R. Huxley and M.H.P. Jebb (1993), so cultivated myrmecodias are easier to identify. The majority of myrmecodias offered for sale are forms of the widespread and very variable species Myrmecodia tuberosa (see Figure 1 and Figure 5).

Figure 5: Myrmecodia tuberosa seedlings.

Left, M. tuberosa from Bako National Park, Sarawak, Malaysia. This plant may correspond to Huxley and Jebb's (1993) M. tuberosa "bracteata." Right, The variety sold as Myrmecodia solomonensis (Huxley and Jebb's M. tuberosa "salomonensis"). Both seedlings are one year old and about 9 cm (3.5 inches) tall.

M. echinata is a synonym that is sometimes still seen on plant labels. This species has brown or greyish tubers, stems that sometimes lack clypeoli, and spines that can be simple or branched. Numerous entrance holes are often visible in arcs on the upper surface of the tuber. The plant distributed by the Huntington Botanic Gardens plant sales as Myrmecodia sp. #51501 is M. tuberosa from Sarawak, Malaysia. The original material was collected in Bako National Park and first grown at the Marie Selby Botanic Gardens (Donna Atwood, personal communication). Myrmecodia solomonensis, a plant sometimes offered for sale, is considered a variety of M. tuberosa by Huxley and Jebb, and a Myrmecodia sp. from New Britain is probably M. tuberosa "dahlii." At least some of the plants sold as "M. platyrea" (sic) are also M. tuberosa.

Other plants in my collection include a young M. beccarii and a plant from New Guinea which might be M. platytyrea (see Figure 6).

Figure 6: Myrmecodia species from Papua New Guinea.

This mature plant seems to fit the description of M. platytyrea (Huxley and Jebb, 1993). Clypeoli and alveoli are visible on the stem. The plant is growing in osmunda fiber in a 4 inch clay pot. The tuber is 13 cm (5.1 inches) long.

M. beccarii is a lowland species from northern Australia which often grows on mangroves. It is characterized by its small, succulent leaves and its tendency to produce multiple stems. The putative M. platytyrea has large, circular clypeoli and a silvery grey tuber covered with warts. Another plant of unknown origin has very similar clypeoli but the tuber is brownish and globular.

The related genera Anthorrhiza, Myrmephytum, and Squamellaria, do not seem to be widely grown. I am aware of a single Anthorrhiza specimen on display at the University of Oxford Botanic Gardens in England, but no others. I would definitely be interested to hear reports of other plants in cultivation.

The myrmecodias common in cultivation are lowland species found at altitudes below 1000 meters (Huxley and Jebb, 1993). The hydnophytums are also likely to be lowlanders, though this is uncertain in the absence of collection data or definitive identification. The habitat preferences of ant-plants summarized above would suggest cultivation in warm temperatures with high humidity and bright light, though some protection from mid-day sun may be warranted. In North Carolina, I grow my plants outside in summer where they experience high humidity and direct sun after about 2:00 pm. Daytime temperatures are commonly 90-95 ûF (32Ð35 ûC), and nighttime temperatures drop to the 60's (15Ð20 ûC). I would not risk prolonged exposure to temperatures much below 60 ûF (15 ûC). In winter, the plants grow under fluorescent lights with my orchid collection or on a sunny windowsill.

Many growers successfully use a mix of peat and perlite, but I prefer a combination of long-fiber sphagnum moss, chopped fir bark, and perlite which is very open but retains moisture. The plants are watered when the mix is just barely moist, because sphagnum is difficult to re-wet if it is allowed to become bone dry. Either plastic or clay pots are suitable depending on one's watering habits. If humidity is high, plants can be mounted on cork bark or treefern plaques for a more natural appearence (see Figure 2). Mounted plants will, of course, require more frequent watering. The roots of Hydnophytum and Myrmecodia are very brittle, so great care must be taken when repotting. Luckily, however, the plants seem capable of producing new roots from any part of the tuber that is in contact with a moist substrate.

Jolivet (1996) reports that cultivated plants without ants are stunted, but this does not seem to be the case if plants are fertilized. Any balanced fertilizer with trace elements should be suitable. I use Dyna-Gro 7:9:5 at a rate of 1/2 tsp per gallon every second watering, and I throughly flush the pots with water on days when I do not fertilize. Sphagnum moss frequently sprouts from spores in my potting mix, indicating that this procedure prevents buildup of fertilizer salts. Moss growth is especially luxurient in flats of seedlings which are never allowed to dry out.

PESTS AND DISEASES

Ant-plants are susceptible to a number of common pests such as scale and mealybugs. In general, the thin-leaved myrmecodias seem to be more seriously affected than hydnophytums. In dry weather, mites can be a real problem. Check the undersides of leaves for tell-tale webs and silvery scarring. Since my plants are grown as houseplants at least part of the year, I tend not to use toxic systemic pesticides. Washing the foliage with tepid water keeps spider mites under control, and spraying with insecticidal soap or pyrethrin kills scale. Squashing the pests is also an effective and ecologically sound method of controlling them in a small collection; there is no danger of scale evolving resistance to your thumbnail.

In addition to occasional infestations of mites, my myrmecodias seem especially prone to developing ugly brown bumps on the leaves. The bumps begin as whitish swellings on a newly developing leaf and later form a hard corky skin. These defects occur on plants from a variety of sources, and some photographs of wild plants show evidence of the same problem. I suspect that this is not a disease, but rather that it is edema due to low light and excess water. The defects are most severe during cool, dull winter weather and much reduced or completely absent in summer when the plants receive more sunlight. In addition to increasing light intensity, reduction of watering may also help, though underwatering will result in defoliation is some species. Interestingly, my hydnophytums seem immune to this problem, and in Myrmecodia beccarii, the defects are generally restricted to the petiole. Perhaps the thicker leaves of these species are more resistant to the swelling.

Cold wet weather, injury during shipping, or other accidents may result in rotting of the tuber. This can disfigure the plant but need not be fatal if caught early. The rotten portion should be scraped off to expose healthy tissue and dusted with something to dry out the wound. I have used powdered cinnamon, which among orchid growers is reputed to have anti-bacterial properties. If necessary, the plant can be repotted so that the injured portion of the tuber is not in contact with wet potting mix. Small seedlings will occasional lose their growing stem to rot if they are too wet during shipping. As long as the tuber remains firm, these seedlings should be potted up, as they often sprout new stems and may produce interesting multi-branched specimens.

PROPAGATION

According to Huxley and Jebb (1991), cuttings from Hydnophytum stems will root and grow but do not produce a tuber. Luckily, Hydnophytum and Myrmecodia are very easy to grow from seed. The small white flowers of Myrmecodia tuberosa and the Hydnophytum species in cultivation are self-pollinating. When ripe, the fruit of Hydnophytum and Myrmecodia are bright orange or pink and easily detached from the stem. In Hydnophytum, unripe fruit is visible on the stem, but in Myrmecodia, the fruit does not protrude from the alveolus until it is ripe. Depending on the species, each fruit contains between one and seven seeds. Janzen (1974) reports that dry seed of Hydnophytum formicarum remains viable for at least four months, but I have not had much luck storing dry seed and recommend that it be sown immediately after harvesting. The seed should be removed from the fleshy fruit and sown on a moist surface; sphagnum moss is excellent for this purpose. In a pinch, the seed can be sprinkled around the base of the mother plant. Do not bury the seed. Fresh seed germinates within a week, and the initial swelling of the tuber is visible almost immediately. Seedlings grow rapidly under fluorescent lights if kept constantly moist. Even very young seedlings can be shipped successfully if wrapped in damp (but not wet!) paper towels and sealed in a plastic bag. For shipping seed, it seems best to send intact fruit rather than dry seed which may fail to germinate.

CONCLUSION

Hydnophytum and Myrmecodia are definitely worthy of a place in any collection of caudiciform plants, and it is a shame that only a handful of species from such large, diverse genera are currently available. I can only echo Kimnach (1996) in hoping that more species will be introduced into cultivation. It would be wonderful to grow the giant Myrmecodia lamii (Kimnach 1996), the densly spiny M. augustifolia and M. tuberosa "pulvinata" (Huxley and Jebb, 1993), or Hydnophytum guppyanum with its bizarre tuber (Huxley and Jebb, 1991). Until that time, the commonly available Hydnophytum and Myrmecodia tuberosa varieties will reward growers with a variety of shapes that can be elegant, unusual, and sometimes breathtaking in their ugliness.

ACKNOWLEDGMENTS

I thank Tim Devine for generosity with seeds and seedlings, Wilson H.P. Chua for photographs of Myrmecodia in habitat, and Barry Sullender for interesting discussion and helpful pointers to the literature.

REFERENCES
  1. Huxley, C.R. 1978. The antÐplants Myrmecodia and Hydnophytum (Rubiaceae), and the relationships between their morphology, ant occupants, physiology and ecology. New Phytol. 80:231-268.
  2. Huxley, C.R., and Jebb, M.H.P. 1991. The tuberous epiphytes of the Rubiaceae 1: A new subtribe Ð the Hydnophytinae. Blumea 36:1-20
  3. Huxley, C.R., and Jebb, M.H.P. 1993. The tuberous epiphytes of the Rubiaceae 5: A Revision of Myrmecodia. Blumea 37:271-334
  4. Janzen, D.H. 1974. Epiphytic myrmecophytes in Sarawak: Mutualism through feeding of plants by ants. Biotropica 6:237-259.
  5. Jolivet, P. 1996. Ants and Plants: an Example of Coevolution. Backhuys, Leiden. Kimnach, M. 1996. Myrmecodia lamii, a giant ant-plant of Irian Jaya. Cactus and Succulent Journal 68:134-141.
  6. Rickson, F.R. 1979. Absorption of animal tissue breakdown products into a plant stem-the feeding of a plant by ants. American Journal of Botany 66:87-90.


Info on other myrmecophytes can be found on Nick's website
{Page regenerated on Thu, 15 May 2008 06:39:19 -0400}
UConn       The Web       People
WEATHER          MSDS & LABELS          STATISTICS          DISCLAIMER © Ecology & Evolutionary Biology Greenhouses
75 North Eagleville Road, Unit 3043
Storrs, CT 06269-3043
860-486-4052
email: eeb dot greenhouse at uconn dot edu