Water dispersal of seeds in the rainforest
In the often adjacent mangrove ecosystem, there is obviously a dominant trend towards water dispersal of seeds with woody, buoyant fruits. In the rain forest, there is also evidence of this in the design of many fruits. This is not unreasonable to expect, as rain forests grow in, and are the result of, high rainfall areas, and there are many streams. Further, there is often a graduated transition in species composition, and thus overlap, between the mangrove and the rain forest. Fruits such as the large seeded pod of the 'Black bean' Castanospermum australe (below) float well in water, and in fact are even used by children for doll's canoes (Scarth-Johnson 2000)
The fruit of the 'Black Palm', Normanbya normanbya (Cape Tribulation)
As these fruits float down along a rainforest stream, they may yet be transported by an animal. In some rain forests of the world, most significantly those based around the great waterways of South America, some of the over 2000 species of freshwater fish are dispersers of many rain forest fruits (Whitmore 1999). In comparison, Australia has a very poor freshwater fish fauna, with about 300 species (Allen et al 2002). Nevertheless, fish in local rainforest streams may act as dispersers of seed. For example the 'Jungle Perch', Kuhlia rupestris, has been reported to eat fruits, such as figs (Allen et al 2002).
Animal dispersal of seeds in the rainforest
One of the things that even a casual visitor in any rainforest may observe, and that researchers have quantified, is that fleshy fruits are very common. Studies of subtropical and tropical rainforest around the world have shown that these fleshy fruits are produced by over 70% of the plants (Wilson et al 1989). Many are have much higher proportions; some studies in the Neotropics have shown up to 98% of understorey plants have fleshy fruits (Loiselle and Blake 1999). In the rainforests of Queensland, the percentage of plants with fleshy fruits is has been recorded at about 85% (Wilson et al 1989).
Ultimately fleshy fruits have evolved to attract animals to ingest them and thus act as dispersal agents for the seed. And in the end, the seed is not only dispersed away from the parent tree, but it is also provided, by way of the animal feces, a microsite of moisture and extra nutrients (Grice 1996). This latter advantage may be very important in the competitive rainforest, and in some of the forests that grow on relatively nutrient poor soils. To a certain extent, it appears that plants may aim their fruit at particular types of animals. Animal dispersed fruits advertise with various cues such as colours and smells. They then rewards the disperser with nutrients such as water, carbohydrates and fats.
Animals that feed mainly on such fruit as known as 'frugivores'. While there are, for example, no such specialized fruit eating birds in the whole of the northern temperate region, the rain forests of Asia, the Americas and Australasia are rich in many avian frugivore species (Snow 1981). And many of these frugivores are very effective dispersers of rainforest plants. This has been confirmed where rain forest has been established by flying animals in areas that were completely destroyed of vegetation, as happened on the volcanic islands of Krakatua in Indonesia (Whittaker and Jones 1994). As the rain forest is well known for it's biological diversity, there is a correspondingly wide range of dispersal agents. Fish have already been mentioned, and birds and various mammals, such as bats and rodents are well known vectors. But there are other animals. Seed dispersal by reptiles is called 'Saurochory'. Actual dispersal of fruits by reptiles has been recorded in Neotropical rain forests by turtles, both terrestrial and aquatic, that came up our of the river to defecate (Moll and Jansen 1995). In the rainforests of Australia, the eating of the occasional fruit by 'Boyd's Forest Dragon' Hypsilurus boydii has been recorded by Torr (1997).
However, not all frugivores are necessarily effective dispersers. In fact one way of categorizing fruit eating animals is based on this effectiveness as a seed disperser. There are 'seed predators' who destroy and/or digest the seed. There are 'seed thieves' who eat the fruit, but then drop the seed at the site of the providing parent tree, therefore not effectively dispersing the seed. And then there are the 'seed dispersers'.
Seed dispersers themselves can also be then be measured for their effectiveness. This is often measured by how far they disperse the seed; for, generally speaking, the further away from the parent tree the better. A study in Australia by Green (1993) observed how long particular birds stayed feeding in fruiting trees, to measure their effectiveness as seed dispersers, and was also used to establish the birds that are 'good seed disperser' and 'poor seed dispersers'. The same study also identified birds that were better seen as 'fruit thieves' and 'seed predators' (Green 1993).
Fruit eaters that are actually 'seed predators' are not really considered at length here. But as an example in Australian rainforests, the parrots are probably best known, ripping through woody fruits with their sharp beaks to consume seeds. An example in the local rainforest are the 'Maples', Findersia spp. which, despite their hard shell and sharp 'tubercles' can often be seen in pieces on the ground, chewed up by cockatoos. Another good example of the differences between 'seed predators and 'seed dispersers' is actually provided within the one group of birds, the pigeons. Some of the pigeons eat the fruit but their tougher gizzards destroy the seed, while other pigeons have softer gizzards and void the seed unharmed (Snow 1981).
However, one type of seed predator does manage to disperse the seeds over time. These animals are those that may bury many of their seeds for later use. They are known as 'scatterhoarders' and their dispersal works through the idea that for every large amount of seeds they bury then retrieve to eat, there may be a seed they leave in the ground. Here, it may protected and safe from some other seed predators and thus effectively dispersed. Examples in the Australian tropical rain forest include the 'White-tailed Rat' Urmoys caudimaculatus and the 'Musky Rat Kangaroo' Hypsiprymnodon moschatus, although studies have shown that the former animal retrieves, and thus seed mortality, is virtually 100% (Strahan 1998).
The different types of fruits and their vectors in the tropical rain forest
The tropical rain forest not only is rich in the mount of fleshy fruits, but also in the range of different types. The connection between different types of fruits being eaten by different types of animals has probably been known since humans evolved as hunter-gatherers, but the serious study of such dispersal 'syndromes' really only began with quantitative studies in the twentieth century, and especially from 1970's onwards. Since then there has been a growing amount of both literature-search type study (e.g.; Snow 1981), and more field based tests (e.g.: Crome 1975), working on these assumptions and attempting to determine trends.
One of the more obvious trends is the correlation between the size of the fruit and the animal. Studies have shown that the larger the fruit, the less the range of dispersers, and the more specialized those visitors have to be, while smaller fruits have many more visitors, both in terms of number of individuals, and in number of species (Green 1993). However, it also a generally rule that the larger the animal, the wider the possible dispersal. Thus, many of the seeds and their fruits in the rain forest are large and attractive to vertebrate animals. Seed dispersal by vertebrate animals is called 'endozochory' (Wilson et al 1989).
The determination of the general type of vertebrate dispersing agent can be determined by examining the features of the fruit (the size of the fruit, the number of seeds within), as well as the attracting agents (the colour, contrast, and again, also the size of the fruit). Much of the determinations made in the 'classic' literature is done through contrast; for example, direct comparisons are made between 'bird fruits' and '(terrestrial) mammal fruits'. Large fruits with lots of smaller seeds tend to be eaten by mammals that have teeth and chew, but hopefully swallow some of the smaller seeds to be passed out later. Smaller fruits are usually chosen by smaller animals as they need to stay light and lack the guts to break down larger fruits, they also tend to have bigger single seeds that are gulped whole down the toothless beaks of birds. However, it is the colour of fruit that is often the most obvious difference to us visually orientated humans. Fruits with bright and contrasting colours such as red or black, and sometimes yellow and blue tend to attract the more vision-reliant animals such as birds (Wilson et al 1989). (As the same author notes, however, they appear red or black only to our eyes and birds may be picking up different hues, as at least some birds species can detect ultraviolet light.) Mammals on the other hand, often feeding in the lower light levels of night, tend to feed on lighter and duller coloured fruits; browns, greens, orange or yellows (Wilson et al 1989). Mammals generally do not respond to colours as well as birds but tend to be more olfactory. The exceptions are of course the primates, such as ourselves, and monkeys, that have shifted from olfactory senses to visual. These differences between different types of fruits aimed at different types of animals are probably best compared in a large group of related plants, so as to eliminate the complications and bias of differing phylogeny; the figs are a good example to use.
Figs and their different animal dispersal agents
Figs, Ficus spp., are of course one of the best known and widespread of all plant genera. In fact, Figs are considered by some authors to be the most diverse genus of woody plants when looking at habit, growth forms and life forms (Kalko et al 1996). And because the genus is ubiquitous in all the world's rain forests, it is often studied (example, all of the Journal of Biogeography number 23). It is not just the distribution that attracts interest, however, for the genus has some important and fascinating ecology. Perhaps best known is it's highly specific and reliant relationship with 'fig wasps' (family Agaonidae); the plant relies on the wasps for pollination, while the wasps rely on receptacle to reproduce (Compton et al 1996). Because of this relationship, it is thought figs have to provide these fruit receptacles more or less continously, and thus many figs fruit asynchronously. That is, trees fruit out of sync with other individuals of the same species (Kalko e al 1996). This also means that figs are being produced all year round, and thus can provide a regular food source for many animals in the rainforest, and this is especially important when food is scarce at certain times of the year. And thus figs are often considered 'keystone' species. They have traditionally been thought of as providing fruits mainly for generalist fruits eaters, and over 50 species of birds in Australia have been recorded feeding on fig fruits (Green 1993). However more recently it has been acknowledged that some figs may target specific dispersers and that there are some animals that may almost entirely feed on figs. It is no accident that some Australian birds have the 'fig' prefix in their common name, such as the 'fig parrot' and the 'figbird', Specotheres viridis (Slater et al 2001); the latter bird is thought to be almost an obligate fig eater (Green 1993).
A study by Kalko et al (1996) in the neotropics on figs highlighted the difference of fruits within this same genus that were aimed at different types of animals; 'bat fruits' and 'bird fruits'. 'Bat figs' were found to be greenish when ripe, while 'bird figs' were brighter, usually red. There were differences in the pattern of fruiting, and where on the tree they fruited; 'bat figs' were produced all at the same time, while 'bird figs' were asynchronous. There were even differences in the amount of protein and carbohydrates, with 'bird figs' being cheaper in what they offered in terms of nutrition. Even the form of the fruiting tree itself may determine the dispersal vector, with the unusual habits of 'ramiflory' and 'cauliflory' (above) possibly evolving for easier access for bats (Whitmore 1999).
Bats and seed dispersal
The traditional division of 'bird' vs 'mammal' fruits becomes more complicated when bats are examined, as they have the senses of mammals, but the flight and mobility (and thus inherent problems and limitations) of birds.
Fruits that are attractive for flying fox have certain attributes that appeal to their senses and their morphology. They are often attracted to fruits that are duller or lighter colored (Richards 1987), juicy and sometimes strong smelling. The fruits themselves are varied in size, and can often be quite large, so large in fact, that most birds can not disperse them (Whittaker and Jones 1994). However, some of the fruits that are successfully transported by bats have many small seeds; as large seed will not pass through the smaller oesaphagal openings of flying fox, and thus larger seeds will be dropped nearer the tree (Whitaker and Jones 1994). Compared to the oily fruits for obligate frugivorous birds, bat fruits tend to be much more 'sugary' (Whitaker and Jones 1994). And so generally, flying fox find many of the same fruits that humans find attractive. In fact, many of the fruits that are dispersed in the wild by these bats have been cultivated by humans, such as bananas and mangos. Unfortunately, this also means orchards of these fruits become the target of these animals (Tidemann and Nelson 1987).
Flying-fox have been suggested by many researchers and their studies to be very effective seed dispersers. This is despite the fact that are first glance they would seem inefficient, for flying fox do not necessarily eat a whole fruit. They are often observed to chew on the fruit and will suck out the juice to spit out the pulp and thus often the seeds. However, many of the fruits favored by such animals (and other mammals such as ourselves) have many smaller seeds that are difficult to separate from the surrounding sloppy pulp. A study of flying foxes in the Philippines showed that figs that that had passed through the animals were almost twice as likely to germinate than those that hadn't (Ultzurrum and Heideman 1991); and so those that were ingested had a chance of being transported away as the animal flew elsewhere, while those that were spat out stayed near the tree. (Although, interestingly, the reasons suggested by the authors were not simple physiological effects, but more complex ecological causes, see paper). In a study of figs in the Philippines a third of the thirty species of figs were eaten by the local flying fox (Kalko et al 1996). Another study on islands in the Pacific showed a simple but telling correlation; where there were less flying-fox, there was less seed dispersal (Fuentes 2000). In Australia there is a species, the 'Spectacled Flying-fox' Pteropus conspicillatus, which is considered primarily a rainforest endemic (Strahan 1998), and it has been shown to disperse viable seeds for over 26 plants (Hall and Richards 2000), with 23 of these being in the tropical rainforest (Richards 1987).
Of course the most obvious advantage to dispersal by flying fox is the fact they can, as the name suggests, fly. They are noted to often roost a long way from their feeding areas (Kalko et al 1996). This can be observed locally; the largest colony of the 'Spectacled Flying-fox' Pteropus conspicillatus in the region is located right in the middle of urban Cairns (Whybird and Clague 2001), and can be seen leaving their camp and flying off every evening. Some species have been recorded traveling up to 40 kilometers from their roosting to their feeding areas (Kalko et al 1996).
'Flying Fox' no doubt fill a very important ecological niche in Australasian rainforests. In an international context, our rainforests are often thought to be unusual for the comparative lower diversity of larger vertebrate dispersal agents (Kroon and Westcott 2001). This is primarily due to a lack of frugivorous mammals in the region (Wilson et al 1989). We are missing many of the mammals found in other rain forests of the world, such as the different rodents (porcupines, agouti, paca), primates (various Old and New World monkeys, lemurs) and even marsupials (four-eyed and mouse opossums in the neotropics), (Crome 1990, Emmons and Feer 1990). Causal observers are not aware of most of these 'missing' animals, except, invariably, for the primates. However, 'flying fox' in Australia are in many ways our closest ecological equivalent to these arboreal primates of other rain forests. This is perhaps even more interesting in light of the arguments that have risen about the evolutionary history of flying fox in the last few decades, for their was some evidence suggesting the Megachiropteran bats may be more closely related to primates than the Microchiropteran bats (e.g: Pettigrew 1988).
General study surveys of the fruits of the rain forest and their presumed dispersal vectors have shown that fruits suitable for flying vertebrates are more common than fruits suitable for non-flying vertebrates (Wilson et al 1989). Being fliers and thus highly mobile, bats are likely to be one of the most successful groups of animals and most effective seed dispersers of the rain forest. However, there is one huge and diverse group of flying vertebrates that are no doubt the most effective, (and best known), seed dispersers of them all. For while the Australasian rainforests may lack a high diversity and abundance of frugivorous mammals, they are often noted for their specialized frugivorous birds
Birds as seed dispersers in the tropical rain forest
Generally, birds are probably the most commonly attracted dispersal agent in the rain forests of the world. One study estimated at least 68% of tree species produced bird dispersed fruit (Wilson et al 1989). And here in the rainforests of Australia's wet tropics bioregion, over 40 species of birds have been recording regularly including fruit in their diet (Jones and Crome 1990). Thus trees and their fruit and the physiology of birds have probably co-evolved to some extent. Depending on the species of bird, seeds can take anywhere from between 10 minutes and 2 hours to pass through the digestive system (Greeen 1993). This is much quicker compared to other non-flying vertebrates, as birds need to lighten the load for flight, but as they are active and mobile, so the seed can potentially be taken away from the parent plant. And in fact, some studies have observed that birds generally don't stay for too long in the fruiting tree, and fly to another before voiding the seed and thus dispersing the seed effectively (Green 1993). Nevertheless, some plants produce fruits that have a laxative effect (Murray et al 1994).
Fruit-eating birds can be further divided into those specialist fruit eaters, or 'obligate frugivores', and the more generalist bird that includes fruit as well as other food in it's diet. The obligate frugivores that rely on fruit and little else for their diet also tend to feed on fruit that is more nutritious and rich in fats and proteins, such as those in the family Lauraceae (Crome 1975). The less specialized fruit eaters feed on the less nutritious fruit that are high in sugars and water content (Snow 1981). The more generalist feeders may visit more trees and therefore have the potential to perhaps be capable of wider dispersal (Green 1993). However, the obligate frugivores tend to have gentler gizzards, and pass the seeds less altered and often thus more viable; this is probably also a way of avoiding the poisonous compounds that are present in many seeds (Stocker and Irvine 1983).
In Australian rainforests, the most diverse and abundant group of avian frugivores are the various pigeons. There are 7 species of obligate frugivorous pigeons, (with 2 being endemic to the country) and several more species that feed opportunistically (Crome 1975). Other avian fruit eaters are the bowerbirds and catbirds, the birds-of-paradise, the old world Orioles, the metallic starling, several of the honeyeaters (Green 1993), and the biggest native animal of the ecosystem, the 'Southern Cassowary', Casuarius casuarius.
The Cassowary as a seed disperser
The cassowary is probably the single most important animal seed disperser of the local rainforest. Studies have shown that the animal is an obligate frugivore, with the bird depending on fruits to survive (Stockerand Irvine 1993). So far, over 200 rainforest plants have been recorded to be eaten by cassowaries (Kroon and Westcott 2001). The relationship may be equally dependent in the other direction. One study recorded 78 species found in Cassowary dung, with at least some of the seeds of 70 of those species remaining viable (Stocker and Irvine 1983). As with many fruit specialists, many of the seeds come out virtually untouched and many of these are far too large for any other extant animal, except for the cassowary, to disperse (Stocker and Irvine 1983). And even though the bird itself is hard to see, their scat can be quite common and conspicuous, and can contain up to 1 kilogram of seed (Stocker and Irvine 1983). One scat contained 13 seeds of about of about 6 centimetres in diameter (Stocker and Irvine 1983).
Being large animals, Cassowaries also have the potential to be effective dispersers over some distance. Radio-telemetry showed that cassowaries, especially in the uplands, had huge territories, sometimes of several kilometres, and may cover that distance in a day (Kroon and Westcott 2001). Further, seeds may take over 10 hours to pass through the digestive system (Stocker and Irvine 1983), thus cassowary dispersed seeds have the potential to be dispersed quite widely.
Wind-dispersal of seeds in the tropical rainforests
As can be seen from the previous pages, there has been a lot of effort put into the interesting field of seed dispersal by animals in the rain forests of both Australia and the rest of the world; too many to include in this essay. This amount of research and information also reflects the dominance, abundance and ecological importance of such relationships. However, there is also other important dispersal methods at work in the rain forest.
While it is common to come across fleshy fruits in the rainforest, it is also surprisingly common to see hard, often large, woody fruits ('follicles'). This is also perhaps an artifact of the fact that the hard woody fruit tends to last longer on the forest floor, and are not a food source as are the fleshy fruits, however, the diversity of woody fruits in the rainforest is still impressive (eg; Cooper and Cooper 1994).
Wind dispersed seeds tend to be small, light and numerous. They are usually contained in a protective woody follicle that protects the seeds from predation. These woody fruits open to release the many small seeds, many of which may be winged in some way to assist in dispersal, such as the jungle vine seeds to the right.
However, one of the more obvious features to human visitors in the rainforest is the lack of wind. Thus, wind dispersal would not seem to be a successful strategy. That is, unless the main wind dispersed seeds were only found on the edge of the rainforest as are 'pioneer species' (e.g.; Commersonia, see Figure), or sticking out above the rainforest as 'emergent species' as in the next family.
While it is generally agreed that the rain forest is botanically very species diverse, there are some cases where particular groups of plants may dominate the forest. In the forests of south-east Asia, for example, one such family dominates the rain forest, both in height and in species; the Dipterocarpaceae. These huge timber trees have winged, wind-dispersed seeds (Whitmore 1999). The rainforests of Australia, although with a very different evolutionary origin and history, have their parallel in the family Proteacae. While there are many smaller species, sometimes with fleshy fruits, there are many Proteaceae that are rainforest emergents also with winged wind-dispersed seeds, such as the 'Silky Oaks' of various genera.
Resembling the seeds of the Dipterocarps even more are the seeds of the Argydendron spp of tropical and subtropical east-coast Australian rainforests. Like many of the Dipterocarps, the large seed is conspicuously winged, and when thrown often twirls around before reaching the ground. As rainforest emergents with wind dispersed seeds, it's a simple equation; studies showed that generally, the taller tree, the further the seed fell away from the parent, or the longer the 'seed shadow' (Nadolny 1999).
The thin papery seeds themselves are much harder to find, as they are usually small and delicate, and are eaten by seed predators. It is quite common to find the woody fruits on the ground before they have opened, and chewed open. In the Australian tropical rainforest, the Sulfur-crested Cockatoos (right), often try to get at such seeds, such as Flindersia, before they are ready to be released.
To avoid being eaten, many of these wind-dispersed plants tend to invest all the resources into one big explosive seeding even, and this peaking is known as 'mast seeding', which occurs in Australian tropical rainforest with species such as the Red Tulip Oaks (Nadolny 1999).