Epizoochorous Dispersal by Barbs, Hooks, and Spines in a Lowland Moist Forest in Central French Guiana


Scott A. Mori and John L. Brown
piclecpo.jpg - 4328 Bytes

Reprinted with permission from Brittonia 50(2): 165-173. 1998. © 1998. The New York Botanical Garden. Changes from original manuscript: Pavonia schiedeana is now considered P. castaneifolia. Please report comments and corrections to Scott A. Mori (smori@nybg.org).

Abstract | Introduction | Study Site | Methods | Results | Discussion
Figure 1 | Table 1
Acknowledgements | Literature Cited

Mori, S. A. and J. L. Brown (Institute of Systematic Botany, The New York Botanical Garden, Bronx, New York 10458-5126 U.S.A.). Epizoochorous dispersal by barbs, hooks, and spines in a lowland moist forest in central French Guiana. Brittonia 50: 165-173. 1998.--An analysis of the flowering plant flora of a lowland moist forest in central French Guiana reveals 25 species with adaptations for epizoochorous dispersal by barbs, hooks, or spines (stick-tights). This represents 1.5% of the flowering plant flora. Stick-tights are represented among the monocotyledons of the flora of central French Guiana by five species of Poaceae. The Fabaceae and Asteraceae, each with six species with stick-tight dispersal are the richest families of flowering plants in terms of stick-tight dispersal in central French Guiana. This region possesses numerous potential mammal and bird dispersal agents, therefore it is a puzzle why stick-tight dispersal, especially among species of the canopy, is not more frequent.

Key words: dispersal, epizoochory, French Guiana, lowland moist forest, stick-tights.

In lowland tropical forests, mammals and birds are often attracted to and disperse seeds because of the reward provided by edible parts of the fruits or seeds. The greater diversity of tropical plants and animals in comparison to their temperate counterparts may be attributed, at least in part, to the coevolution between plants and animals searching for food (Emmons, 1989; Fleming & Estrada, 1993; Gentry, 1983; Howe & Westley, 1988; Pijl, 1982). Nevertheless, other dispersal systems such as wind (Mori & Brown, 1994) and passive animal dispersal may play important roles in seed dispersal in tropical forests.

The purpose of this paper is to describe the role of stick-tight dispersal, a special kind of epizoochory, among the flowering plants of a lowland moist forest in central French Guiana.

Study Site

This report is a by-product of a floristic study of the forests surrounding the village of Saül (3° 37'N, 53° 12'W) in central French Guiana (Mori et al., 1997).

Annual rainfall for Saül, based on data from 1956 to 1975, averages 2413 mm. There is a distinct dry season from July to November and a less pronounced drier period that sometimes occurs for several weeks in February to March.

Other than a zone of minor disturbance surrounding Saül, the region is covered by undisturbed forest. Most of the study area lies between 200 and 400 meters elevation and, within this band, lowland, non-flooded, moist forest prevails. This forest type has been described by Oldeman (1974) and Mori and Boom (1987) for this part of French Guiana. Considerable differences in the vegetation occur due to variation in soil moisture and fertility, with the most obvious deviation from the predominant type of forest being the Euterpe oleracea dominated swamps in low, wet areas along streams. Occasional large, granitic rocks and domes project above the surrounding forest, and these areas harbor plants not found elsewhere. The largest granitic outcrop within the limits of this guide is found to the NE of Saül on Pic Matécho (Mori et al., 1997).

A shift in species composition appears to occur above 500 m elevation (Granville, 1986), such that cloud forest is formed on the highest peaks. The best example of cloud forest in the area of this flora is that occurring on the 760 m high Mont Galbao.


The dispersal systems of the 133 families, 699 genera, and 1861 species (426 monocots and 1435 dicots) known from a 140,250 hectare area in central French Guiana were examined for adaptations for external dispersal by animals. We defined epizoochory as a type of dispersal in which seeds are carried away from parent plants by attachment to the surface of animals. For the purposes of this paper, we considered only those species with special morphological "adaptations" for attaching diaspores (diaspore = unit of dispersal) to animals. Our concepts follow those of Sorensen (1986), who made a clear distinction between fruits that attach to animals by barbs or hooks and fruits that adhere by viscid exudates. Sorensen calls the former "burrs" and the latter "viscid fruits." We excluded viscid fruits from our analysis and differ from Sorensen by using the term "stick-tight" rather than burr to discribe this type of epizoochorous diaspore. In addition, we included diaspores that attach by spines that stick into the animals surface; that is, in our analysis barbs or hooks do not always have to be present for a species to be considered a stick-tight (e.g., species of Sida). We determined what part of the diaspore possessed the adaptations for epizoochory by examining herbarium collections of the species at The New York Botanical Garden (NY).

The classification system of Cronquist (1981) for the flowering plants was followed in this analysis.


Distribution of Stick-tight Dispersal Among the Flowering Plants

In the flora of central French Guiana there are 27 species of flowering plants with stick-tight dispersal (Fig. 1, Table I). This represents 1.5% of the flowering plant flora. Nine families in six of the 11 flowering plant subsclasses display stick-tight dispersal in central French Guiana. The Fabaceae (45 species known for the flora) and Asteraceae (22 species known for the flora), each with six species, possess the greatest number of stick-tight dispersed species. Only one family of monocotyledon (Poaceae with five species) has evolved stick-tight dispersed fruits in the flora of central French Guiana.


In the flora of central French Guiana the diaspore of epizoochorous plants may be 1) the infructescence or part of the infructescence (11%), 2) the fruit plus some accessory (37%), 3) a segment of the fruit (30%), or 4) the fruit itself (22%) (Table I). In species of Pharus (Fig. 1), hooked hairs on the lemma become attached to animals and entire infructescences or parts of infructescences are carried away. In several species the fruit is surrounded by a structure with hooked hairs. For example, the fruit of Priva lappulacea (Fig. 1) is enclosed by an inflated calyx that possesses hooked hairs and the fruits of Rolandra fruticosa are enclosed in a bract (chaff) with a pointed awn that sticks into the fur, feathers, or clothing of passing animals. In Cyathula prostrata (Fig. 1), hooked hairs are found on the bracts and perianth parts of sterile flowers surrounding fertile flowers -- when in fruit, the hooks become attached to passing animals which then carry away the sterile flowers as well as the fruits. In Petiveria alliacea the fruits possess downward pointing spines at the apex, and in the epizoochorous species of Asteraceae (except Rolandra fruitcosa) the fruits are are adapted for epizoochory by specialized barbed pappuses. In other species the entire fruit or part of the fruit is dispersed. For instance, the species of Mimosaceae possess a specialized fruit, a craspedium, in which the suture of the fruit is armed with spines or hooks. After the fruit becomes attached to the dispersal agent, segments containing single seeds break free and eventually fall to the ground. Desmodium develops another specialized fruit, the loment (Fig. 1). This kind of legume, armed with fine hooks, is constricted into individual one-seeded segments that may or may not break apart during the course of dispersal. Special one-seeded fruit segments called mericarps are the diaspores of Pavonia schiedeana and of the species of Sida in our flora. In many of the epizoochorous species dispersed by barbs, hooks, and spines, the dispersal unit contains a single seed, either from the onset or by the eventual break-up of multi-seeded fruits.

Habit and Habitat

All stick-tight dispersed species in the flora of central French Guiana are terrestrial herbs, subshrubs, or shrubs (Table I, with two exceptions: Mimosa guilandica is a liana with hooks inserted around the suture of the fruit, and Trymatococcus amazonicus is an understory tree with hooked hairs on the fruit surface. All of the stick-tight species in the flora of central French Guiana, except T. amazonicus, grow in open, weedy habitats, and many of them have very wide geographic distributions.


As pointed out by Mori and Brown (1994), ongoing study of extant collections as well as the collection of taxa new to the flora will modify estimates of the number of taxa found in central French Guiana for many years to come. Even in relatively well documented floras, such as that of California, as many as five to 10 new taxa were added annually as recently as 1987 (Shevock & Taylor, 1987). Nevertheless, the conclusions reached in this paper will not be significantly affected by new discoveries because both non stick-tights and stick-tights will be added to the flora with further botanical exploration. It should be noted, however, that epizoochorously dispersed species certainly make up a slightly lower percentage of the flora than indicated in this paper because they mostly occur in easily collected habitats and are therefore probably more adequately sampled at the present time than are non-stick-tights.

For the most part, identifying a plant as having stick-tight dispersal offers little chance for error, because barbs, hooks, or spines are easy to recognize. The presence of diaspores on clothing after a day in the field also indicates that a particular species is epizoochorous. On the other hand, several species of Asteraceae (Acmella spp. and Ageratum conyzoides) have hairs along the margins of their fruits and appear in the literature as epizoochorous (Ridley, 1930). However, the fruits of these species in our flora have no special adaptations for adherence to animals. Moreover, they do not easily become attached to clothing, or presumably fur or feather, when a human or other kind of animal brushes against them. Similarly, species of Begonia, Orchidaceae, Cyperaceae, and some others have small seeds that may adhere to passing animals, especially when wet, and species of Loranthaceae (showy mistletoes) and Viscaceae (Christmas mistletoes), which have viscid substances surrounding the seeds, are dispersed epizoochorously. However, none of these nor similar taxa were included in our analysis, because they do not have the barbs, hooks, or spines that facilitate epizoochorous dispersal.

It has been demonstrated that plants such as Sarcopterium spinosum (L.) Spach (Rosaceae) of Israel possesses smooth, rounded seeds without any special adaptations yet they are carried in the fur of animals (Shmida & Ellner, 1983). However, the only way to document this kind of epizoochorous dispersal would be to examine all possible animal dispersal agents and comb their fur, feathers, or clothing for diaspores.

Hooked, barbed, or spined diaspores have been shown to be efficiently dispersed epizoochorously. In a study of hares (Lepus capensis) of the dry thornbush of east Africa, it was demonstrated that they carried diaspores of at least 17 species of plants in their fur. Moreover, because half of the captured hares carried seeds, because the hares groomed themselves at least twice a day, and because the hares reached high population densities, it was concluded that epizoochorous dispersal by hares plays a significant role in the distribution and maintenance of at least some plant populations (Agnew & Flux, 1970). Epizoochory on man and his livestock (e.g., sheep and goats) can account for a significant amount of diaspore movement, as has been shown in the Mediterranean chapararral (Shmida & Ellner, 1983). Carlquist and Pauly (1985) have experimentally demonstrated the adherence capacity of selected species of southern Californian epizoochorously dispersed plants.

Epizoochorous dispersal is considered to be an excellent means of long-distance dispersal and has long been recognized as playing an important role in getting plants established on oceanic islands (Carlquist, 1980; Gentry, 1983; Renner, 1992; Willson et al., 1990).

Although epizoochory via stick-tights is an effective means of dispersal for some plants, this mode of dispersal is not employed by large numbers of species in any of the vegetation types heretofore studied. Our data from central French Guiana supports this assumption, as only 1.5% of the flowering plant species in the flora are dispersed by stick-tights. Hoffman and Armesto (1994) note that very few species of plants in the Mediterranean ecosystems of Chile, California, and Australia possess epizoochorous dispersal systems; and Willson et al. (1990) suggest that external dispersal by stick-tights and by adhesion by sticky substances is generally uncommon (<15% of the species in the numerous temperate floras they studied).

Stick-tight dispersal among the monocotyledons in central French Guiana is restricted to the Poaceae, a family well-known from other areas for its epizoochorous dispersal (Silberbauer-Gottsberger, 1984).

The lack of a significant number of flowering plants dispersed by stick-tights in the flora of central French Guiana can not be attributed to the absence of possible dispersal agents. The fur-bearing animal fauna of the area comprises nine species of opossum, three species of anteater, two species of sloth, numerous species of bat, at least seven species of monkey, the bush dog, a raccoon, the coati, the kinkajou, the grison, the tayra, an otter, the ocelot, the margay cat, the oncilla, the jaguarandi, the puma, the jaguar, two species of peccary, three species of deer, several species of squirreles, many species of mouse and rat, two species of porcupine, the paca, the red rumped agouti, and the green acouchy (Emmons, 1990). The bird fauna is even richer in numbers of species (Dick et al., 1984).

Especially striking is the near absence of stick-tights among canopy plants, i.e., epiphytes, hemiepiphytes, lianas, and trees (Table I). Stick-tight dispersal is probably not more prevalent in the canopy plants because arboreal animals such as monkeys, kinkajous, squirrels, and some rats are so dexterous that they readily remove diaspores attached to them. Sorensen (1986) has also suggested that the relatively large seeds of trees and the lack of attractants for animals in stick-tight diaspores may also limit epizoochory in trees. In the case of epiphytes, the same kinds of barbs, hooks, and spines serving to attach diaspores to animals are not well adapted for securing the diaspores onto the woody substrates on which epiphytic plants grow.

Carlquist (1974), in his search for an explanation of the relatively high number of epizoochorously dispersed species on some islands, pointed out that "these are dry islands, however, and the dispersal of barbed and bristly fruits may represent successful dispersal of dry-forest elements, as opposed to the fleshy fruit types found in wet forests." In a personal communication to us he adds that, "dry forests and dry scrub show greater seasonality than wet forests in flower and fruit production. Thus, birds cannot depend on a year-round supply of fruits in dry habitats and therefore frugivorous birds tend to be scarce in dry forests and dry scrub but relatively common in wet forests." Therefore, the wetter the forest type, the greater the importance of frugivory and the lesser the importance of epizoochory by barbs, hooks, or spines -- a conclusion that is supported by the relatively low numbers of epizoochorously dispersed species in central French Guiana.

Ridley (1930) was one of the first to suggest that epizoochory decreases from open plains and savanna habitats to more closed forest habitats. For example, Gottsberger and Silberbauer-Gottsberger (1983) have demonstrated that the more open cerrado of central Brazil possesses 15% epizoochorous species, in contrast to the more closed cerradão, where only 1% of the species are of this dispersal type. An exception to this trend, however, is the discovery by Schultka and Hilger (1983) of a high percentage of epizoochorous plants in the ground layer of an evergreen mountain rain forest in north Kenya.

An aspect of epizoochorous dispersal supported by our data is the very strong correlation between the understory habitat and stick-tight dispersal. All of the species with stick-tights known from central French Guiana are understory plants, except, as we mentioned before, the liana Mimosa guilandica and the understory tree Trymatococcus amazonicus. Previous studies have emphasized that epizoochory decreases as plant stature and longevity increase (Ridley, 1930; Schultka & Hilger, 1984; Shmida & Ellner, 1983; Willson et al., 1990).

Species of stick-tights are often found along trails or in open, weedy habitats (Willson et al., 1990), usually have very wide distributions, and are frequently alien (Milton et al., 1990). In central French Guiana, stick-tights are frequently represented by widespread weeds, and some -- e.g., Bidens pilosa, Cyathula prostrata, Petiveria alliacea, and Synedrella nodiflora -- are found in the Paleotropics as well as in the Neotropics. Because of the non-native origin of some of the species dispersed epizoochorously by barbs, hooks, and spines in central French Guiana, the actual percentage of native species with this dispersal type is probably lower than the 1.5% calculated for the present-day flora.

Epizoochorous diaspores attach to the surface of animals by 1) sticky substances exuded by bracts, fruits, or seeds; 2) increased adherence caused by wetting; 3) becoming entangled in fur or feathers because of structures adapted principally for wind dispersal; 4) lodging in fur or feathers because of small size; 5) appearing in mud which in turn becomes attached to the feet of animals, especially waterfowl, and 6) attaching to animals by barbs, hooks, or spines. All of these mechanisms, although not yet documented, are probably found among the flowering plants of central French Guiana. However, there is one mechanism completely missing from this flora: that of solitary seeds with barbs, hooks, or spines.


We thank Bobbi Angell for preparing the illustration, Michel Hoff for developing the database upon which this study is partially based, John D. Mitchell for his numerous suggestions and leads into the literature, and Sherwin Carlquist, Carol Gracie, Emmet Judziewicz, John Mitchell, Michael Nee, and Susanne Renner for reviewing the manuscript. We are grateful to the Beneficia Foundation, The Eppley Foundation for Research, Inc., The Fund for Neotropical Plant Research of The New York Botanical Garden, the Andrew W. Mellon Foundation, and the National Science Foundation (BSR-9024530) for providing financial support for various aspects of this study.

Literature Cited

Agnew, A. D. Q. & J. E. C. Flux. 1970. Plant dispersal by hares (Lepus capensis L.) in Kenya. Ecology 51: 735-737.

Carlquist, S. 1974. Island biology. Columbia University Press, New York.

_______. 1980. Hawaii: a natural history. National Tropical Botanical Garden, Lawai, Hawaii.

_______ & Q. Pauly. Experimental studies on epizoochorous dispersal in Californian plants. Aliso 11: 167-177.

Cronquist, A. 1981. An integrated system of classification of flowering plants. Columbia University Press, New York.

Dick, J. A., W. B. McGillivray & D. J. Brooks. 1984. A list of birds and their weights from Saül, French Guiana. The Wilson Bulletin 96: 347-514.

Emmons, L. 1989. Tropical rain forests. Why do they have so many species, and how we may lose this biodiversity without cutting a single tree. Orion 8: 8-14.

_______. 1990. Neotropical rainforest mammals. A field guide. The University of Chicago Press, Chicago.

Fleming, T. H. & A. Estrada (editors). 1993. Frugivory and seed dispersal: ecological and evolutionary aspects. Kluwer Academic Publishers, Dordrecht.

Gentry, A. H. 1983. Dispersal ecology and diversity in neotropical forest communities. Sonderbd. Naturwiss. Ver. Hamburg 7: 303-314.

Gottsberger, G. & I. Silberbauer-Gottsberger. 1983. Dispersal and distribution in the cerrado vegetation of Brazil. Sonderbd. Naturwiss. Ver. Hamburg 7: 315-352.

Granville, J.-J. 1986. Flore et vegetation. Saga, Cayenne, French Guiana.

Hoffman, A. J. & J. J. Armesto. 1994. 12. Modes of seed dispersal in the Mediterranean regions in Chile, California, and Australia. Ecological Studies 108: 289-310. Springer-Verlag, Berlin.

Howe, H. F. & L. C. Westley. 1988. Ecological relationships of plants and animals. Oxford University Press, New York.

Milton, S. J., W. R. Siegfried & W. R. J. Dean. 1990. The distribution of epizoochoric plant species: a clue to the prehistoric use of arid Karoo rangelands by large herbivores. J. Biogeogr. 17: 25-34.

Mori, S. A. & B. M. Boom. 1987. The forest. In S. A. Mori & collaborators, The Lecythidaceae of a lowland neotropical forest: La Fumée Mountain, French Guiana. Mem. New York Bot. Gard. 44: 9-29.

_______, S. A. & J. L. Brown. 1994. Report on wind dispersal in a lowland moist forest in central French Guiana. Brittonia 46: 105-125.

_______, G. Cremers, C. Gracie, J.-J. de Granville, M. Hoff, & J. D. Mitchell. 1997. Guide to the vascular plants of central French Guiana. Part 1. Pteridophytes, Gymnosperms, and Monocotyledons. Mem. New York Bot. Gard. 76(1): 1-422.

Oldeman, R. A. A. 1974. L'architecture de la forêt guyanaise. Mém. ORSTOM 73: 1-204.

Pijl, L. van der. 1982. Principles of dispersal in higher plants. Ed. 3. Springer-Verlag, Berlin.

Renner, S. S. 1992. I. Seed dispersal. Progress in Botany 53: 435-453.

Ridley, H. N. 1930. The dispersal of plants throughout the world. L. Reeve, Ashford, Kent.

Schultka, W. & H. H. Hilger. 1983. Epizoochore Verbreitung in der Krautschicht beweideter Bergregenwälder des Mt. Kulal (Nordkenia). Beitr. Biol. Pflanzen 58: 333-356.

Shevock, J. & D. W. Taylor. 1987. Plant exploration in California, the frontier is still here. Pages 91-98 in T. S. Elias, editor. Conservation and management of rare and endangered plants. California Native Plant Society, Sacramento.

Shmida, A. & S. Ellner. 1983. Seed dispersal on pastoral grazers in open Mediterranean chaparral, Israel. Israel J. Bot. 32: 147-159.

Silberbauer-Gottsberger, I. 1984. Fruit dispersal and trypanocarpy in Brazilian cerrado grasses. Pl. Syst. Evol. 147: 1-27.

Sorensen, A. E. 1986. Seed dispersal by adhesion. Ann. Rev. Ecol. Syst. 17: 443-463.

Willson, M. F., B. L. Rice & M. Westoby. 1990. Seed dispersal spectra: a comparison of temperate plant communities. J. Veg. Sci. 1: 547-562.

This article was formatted for the WWW by Henry Burke and Scott V. Heald.

Go to    Figure 1 | Table 1
Back to Dispersal Biology Page
Back to Fungal and Plant Diversity of Central French Guiana Home Page