Review Article

4/18/01

aandapeck@home.com

Ants play a vital role in rainforest ecology. They have developed mutual relationships with many types of trees and other plants, some so important that without one another they would not be able to survive. By developing these relationships, both ant and plant create an optimal environment for each other in which they can avoid predation, provide protection, and cycle nutrients and waste products with each other. Some plants have even developed ways to not only provide a habitat, but they also secrete food products specifically for the ants living with them. In turn, the ants give an amazing amount of protection not only from herbivores, but also from certain types of plant diseases.

There are certain costs to having this type of relationship, however. The plant must give up large amounts of valuable resources to provide food for the ants. In certain cases, the size and defensive efficacy of the ant colony, as well as the individual ants themselves, can be directly related to the investment of resources a particular plant may make. Furthermore, many types of ants have become so dependant on food provided by the plant that they could not survive without it. Some plants are at a much greater risk if they are without their ant defenders because they have evolved to use ants as a defense instead of other types of anti-herbivore resistance. While some types of trees are inhabited by a single obligate ant colony, many types of ant-plant relationships are facultative and some trees and other plants attempt to attract many different types of ants by excreting food types that less competitive foraging ants may be attracted to.

Because the ants are a defense mechanism for their host plants, they have evolved ways that they are induced into certain defensive behaviors depending on the type of cues they are subjected to, such as leaf damage or presence of an herbivore.

It is known that the rainforest is home to thousands of different types of plant and insect life. Many plants have developed ways to recruit ants to form a type of defense mechanism--some have even become dependant on these relationships. Both organisms are able to greatly increase their chances of survival by exploiting the other’s abilities for their own benefit. This paper will explore the different types of relationships that have developed between ant and plant, including the differences between myrmecophytic and myrmecophilic plants, and how these strategies affect the ways in which the plants are protected. It will also discuss the mechanisms used to create these relationships such as size variation, and how this affects the way the ants and plants have developed and how they interact with each other. There are also many different ways in which the ants themselves can respond to cues in order to protect their host tree, depending on the different cues they are subjected to. These can include leaf damage, physical disturbances and type of herbivore encountered. This, as well as the ways in which ants and plants can exchange nutrients will be discussed in this article.

There are two main types of plants who maintain mutualistic relationships with ants: myrmecophytic and myrmecophilic. Myrmecophytes are plants that possess specialized structures that harbor nesting ants. By contrast, myrmecophilic plants attract ants into their vicinity by offering food, and gain protection due to their foraging behavior. (Gaume L, McKey D, Anstett MC 1997; Heil M, Hilpert A, Fiala B, Linsenmair KE 2001). The two types vary in their strategies used and benefit differently as well (Gaume L, McKey D, Anstett MC 1997).

In the relationship between Cecropia trees and Azteca ants, a new colony is formed when a mated Azteca queen burrows through a weakened portion of the tree stem called the prostema. She then enters a hollowing in the stem called the domatia, after which, she begins to lay her eggs. Over time, the prostema grows closed and she and her offspring are sealed inside the domatia. As the colony matures, the workers bore out of the stem and begin to forage and patrol the plant (Sagers CL, Ginger SM, Evans RD 2000). The same is true for founding queens of Aphomomyrex afer (Gaume L, Matile-Ferrero D, McKey D 2000). The queen of A. afer, however, actually seals herself in the domatia by depositing debris around the hole, as well as allowing it to grow closed on its own. It is likely that the queen never leaves that particular domatium for the entire rest of her life (Gaume L, Matile-Ferrero D, McKey D 2000).

Many of the specialized relationships between myrmecophytes and ants in ecosystems are often mediated by homopterans (Gaume L, Matile-Ferrero D, McKey D 2000). Softscales (Coccidae) and mealybugs (Pseudococcidae) are tended by ants inside the specialized hollowed structures of myrmecophytes called domatia. One or many species of homopteran can be found in one host plant. Some ants have been reported to carry their mealybug trophobionts, including some that transfer them to different parts of the plant, or others who carry them on their backs to new nest sights. Other times the founding queen of a new colony will carry them in her mandibles on her flight to establish a new nest (Gaume L, Matile-Ferrero D, McKey D 2000).

There can also be competition between different ant colonies in the same tree. In one species of tree called Macaranga, it was found that the deciding factor in how many queens there can be in a certain colony at any one time is the size of the tree itself (Feldhaar H, Fiala B, Hashim R bin, Maschwitz U 2000). In trees with a diameter of less than 10cm there is usually only one Crematogaster queen per colony. However, in those trees with a diameter greater than 10cm, there can be many queens per colony actively laying eggs (Feldhaar H, Fiala B, Hashim R bin, Maschwitz U 2000). By having many available queens at any one time, the ant colony is able to continuously make use of the tree’s resources and also maintain a large patrolling force at all times, even though the ants have a comparatively shorter lifespan than their host trees. The new queens all originate from the original colony, as opposed to new queens from other colonies moving in and competing with established colonies. If however, the queen dies and the colony shrinks too much to effectively patrol the plant, other queens may be able to penetrate and colonize other domatia. If the habitat is very dry, then non-sibling queens may co-operate with each other to produce workers rapidly, which also maximizes plant growth. (Feldhaar H, Fiala B, Hashim R bin, Maschwitz U 2000).

Variation in ant size

Some plant-ants are very large, such as the ferociously stinging Tetraponera aethiops, which protect Barteria fistula against herbivores as large as Colobus monkeys or even elephants (Meunier L, Dalecky A, Berticat C, Gaume L, McKey D 1999). At the other end of the spectrum, there can be minute ants such as the previously mentioned Petalomyrmex phylax whose benefits are not always clear because of its small size and timid behavior (Gaume L, McKey D, Anstett MC 1997).

There can be certain trade-offs between worker size and worker numbers. Worker size would be expected to vary between systems due to the various requirements for protection of different types of plants. Worker size can also vary within a single colony as well. Variation in sizes of ants may have functional significance in terms of division of labor (Meunier L, Dalecky A, Berticat C, Gaume L, McKey D 1999). It has been shown that in a colony of the ant Aphomomyrmex afer, worker size can vary as much as 3 to 12 times that of the smallest worker. This means that there must be an evolutionary benefit to the tree or colony to provide 3 to 12 times the resources to produce the different sizes. In other types of colonies, worker size may vary very little, such as those in the species Petalomyrmex phylax. Meunier L, Dalecky A, Berticat C, Gaume L, and McKey D (1999) proposed three different hypotheses to explain why worker size variability might be different in different plant-ants. First, size range or degree of polymorphism of each plant-associated ant is advantageous in its respective environment and also advantageous to the plant as well. Second, The size range or degree of polymorphism is advantageous for the ant, and evolved after the origin of association with the plant possibly as an adaptation that enhances mutualistic interactions with the plant. Third, worker size range or degree of polymorphism confers no particular advantage with respect to the host plant as environment, and simply reflects ancestral characters maintained by phylogenetic inertia. Due to the large number of different types of mutualistic ant species, any of these’ or even a combination of the three could apply to any one in particular.

Some plants provide extrafloral nectaries (EFNs) to attract ants that feed on the nectar. Others may have aggregations of honeydew secreting homopterans, tended to by ants that harvest the honeydew. Five main species of ants have been reported who benefit by gathering the nectar, which appears to play a key role in their diets. These include Dolichoderus, Crematogaster, Camponotus, Azteca and Cephalotes. (Bluthgen N, Verhaagh M, Goitia W, Jaffe K, Morawetz W, Barthlott W 2000). de la Fuente MAS, and Marquis RJ (1999) experimented with EFNs by studying the effects that ants had on host plants. They excluded ants from certain plants’ and left others as controls. They found that plants deprived of ants had significantly more herbivore damage, and grew at a much slower rate as well. They showed that ants provided plant protection, although the protection varied with time and light environments. Surprisingly, the control plants also demonstrated a significant decrease in pathogen attack of leaves. Although the ants did not decrease herbivore numbers, they did deter them from feeding. In the absence of ants, more herbivores cause more damage, showing that they don’t reduce herbivore densities, but reduce herbivory by deterring herbivore feeding.

In these systems there can be several different ant species on a single type of plant at any given time due to the fact that the ants and plants are not directly dependant on each other. There can be competition over different types of food, though it has been shown that honeydew tends to be a more valuable diet to the ants because it is richer in amino acids than most EFN sources. This usually results in a single ant colony dominating parts of, or even an entire tree. In contrast, there can be as many as 34 different ant species that visit the same EFN. By providing a more nutritious type of food, through a resource that is cared for by the ants, the tree that has homopteran honeydew insights a more defensive behavior from the ants. The EFNs have developed a more generalized approach to attracting ants, which brings many types of ants that may be less aggressive because they are not necessarily directly affected by the plants overall health. (Bluthgen N, Verhaagh M, Goitia W, Jaffe K, Morawetz W, Barthlott W 2000). This shows that by providing ants with food alone, plants can gain a certain amount of protection without becoming totally dependant on the ants, furthermore the ants can gain a valuable source of food from the plants (Bluthgen N, Verhaagh M, Goitia W, Jaffe K, Morawetz W, Barthlott W 2000).

Myrmecophytes actually provide shelter for nesting ants in domatia. The benefits to the ants are more dramatic in these systems. Not only do the trees provide food, but they also give the ants a home and a place to raise their young. The benefit to the tree is still protection by the patrolling ants. The trees tend to be more dependent on the ants as a form of defense in this type of relationship, as opposed to myrmecophilic plants (Heil M, Fiala B, Maschwitz U, Linsenmair KE 2001). Heil M, Fiala B, Maschwitz U, and Linsenmair KE (2001) also found that over periods of time, trees deprived of ants suffer quite dramatically in comparison to those with an intact defensive colony. Ant-free Macaranga triloba plants lost an average of nearly 50 % of their total leaf area. They also found that two thirds of M. hosei trees lost up to 80% of new twig growth.

One example of a myrmecophytic plant is the Cecropia tree, which normally houses Azteca ants. The Azteca workers patrol the plant and remove unwanted insects and encroaching vegetation. In addition to providing the ants with a home, the tree secretes food bodies specifically for the ants called Mullerian bodies and Pearl bodies. The Mullarian bodies are rich in lipids, carbohydrates, proteins and amino acids. The Pearl bodies are rich in lipids (Sagers CL, Ginger SM, Evans RD 2000). Workers in the ant colony appear to consume insects for a greater proportion of their diets, and prefer to feed the specialized food bodies to the larva (Sagers CL, Ginger SM, Evans RD 2000).

In order to further understand what costs are involved in maintaining a mutualistic ant colony, Sagers CL, Ginger SM, and Evans RD (2000) experimented to find out what resources are used by Cecropia trees in creating food bodies for their primary occupants, Azteca ants. They found that Cecropia nitrogen isotopic consumption varied between trees that had Azteca ant colonies from trees that did not. The Cecropia trees who had active colonies appeared to be using a nitrogen source different from those that did not. They found that tissue in the trees with colonies had an isotopic signature that is a close match to that of the ants that live in them. This suggests that the trees are assimilating nitrogen from ants or ant deposited debris. It was calculated that an average of 93% of the occupied Cecropia trees’ nitrogen was provided from ant sources.

Benefits to ants in these systems are obvious: nest sights and often food sources. Even a tiny, seemingly defenseless ant called Petalomyrmex phylax can greatly benefit its host tree, Leonardoxa africana. Benefits to plants can vary among these systems and are not always clear. It has been shown that leaves deprived of ants are at 7 to 12 times greater risk to leaf chewing insects than those patrolled by ants (Gaume L, McKey D, Anstett MC 1997).

How do ants know what is and is not an enemy, and how do they solve the problems they face in protecting their host trees? What is the benefit to the tree by using ants as a defense as opposed to some sort of chemical? The general biology of ants makes them well suited for use as an inducible plant defense. First, they have acute sensory mechanisms to detect disturbance and chemical cues. Second, they are often aggressive and have well–developed defense mechanisms. Third, they have recruiting mechanisms that make them deployable and reclaimable and fourth, individuals will give their lives in order to defend the colony or its resources (Agrawal AA, Dubin-Thaler BJ (1999).

As previously mentioned, there is a tiny ant called Petalomyrmex phylax that seems as if it could have no defensive capability at all. When its host plant is brushed up against or disturbed, it rushes inside the domatia of the plant, seemingly to protect itself as opposed to its host plant (Gaume L, McKey D, Anstett MC 1997). It seems as if the ant has no benefit to offer its host at all. However, a closer study found that the ants were small for a good reason. Leonardoxa africana, the host plant, has no known mammalian herbivore enemies. The main enemies to this plant are chewing and plant sucking insects, "young leaf specialists". The tiny P. phylax ant actually plays an important role in maintaining and protecting its host plant. It is just the right size to effectively patrol the leaves in adequate numbers to find and remove these small insects. Due to the fact that the plant has no major mammalian herbivore predators it does not need to be unnecessarily aggressive, and has adapted to perfectly fit its role in ant-plant mutualism (Gaume L, McKey D, Anstett MC 1997).

It has been shown that in the association between Azteca species of ants and Cecropia trees, damage to leaves can triple the number of individual ants on the leaf (Agrawal AA, Dubin-Thaler BJ 1999). The Azteca ants respond to the damage by alarming nest-mates. Ants present on the leaf at the time of damage immediately become excited and begin to swarm around the damaged sites to investigate the cause. What causes this type of swarming behavior? It has been found that the ants begin to swarm in response to volatile plant cues from the damaged leaf. Ants on the leaf at the time of damage actually leave the leaf and walk down to the main stem. These ants are sometimes seen walking with a curled gaster, possibly indicating that the ant is releasing an alarm pheromone (Agrawal AA, Dubin-Thaler BJ 1999).

In more facultative ant-plant relationships, plants may attract ants to wounded sites by producing substances that attract the ants, which then eat the herbivore that is attacking the host plant. This would give the ants the additional benefit of food as well as maintaining the health of the plant as a whole (Agrawal AA 1998). Leaf damage is probably not necessary to induce ant recruitment, but the magnitude of response to herbivores is much lower than that to leaf damage. This suggests that leaf damage is a much more potent inducing stimulus of ant activity than presence of an herbivore. Signals associated with leaf damage are likely to be more easily detected by ants then those produced by herbivores (Agrawal AA 1998).

It has been shown that ants and plants can be invaluable resources for each other. The ants can greatly reduce damage caused by herbivores, and plants can provide food and even homes for the ants that defend them. Many types of ingenious relationships have developed between many types of ants and plants. These relationships can be obligate or facultative, specific or nonspecific, and they can involve different types of ants, and even other insects such as homopterans. Ants can respond to different types of cues such as volatile leaf compounds in order to better defend their host plant. Although much is now known about these interactions, more discoveries inspire more questions every day. There is still much research needed to further our understanding of these incredible relationships between ant and plant.

Agrawal AA (1998) Leaf damage and associated cues induce aggressive ant recruitment in a neotropical ant-plant. Ecology 79: 2100-2112

Agrawal AA, Dubin-Thaler BJ (1999) Induced responses to herbivory in the Neotropical ant-plant association between Azteca ants and Cecropia trees: response of ants to potential inducing cues. Behav Ecol Sociobiol 45: 47-54

Bluthgen N, Verhaagh M, Goitia W, Jaffe K, Morawetz W, Barthlott W (2000) How plants shape the ant community in the Amazonian rainforest canopy: the key role of extrafloral nectaries and homopteran honeydew. Oecologia 125: 229-240

de la Fuente MAS, Marquis RJ (1999) The role of ant-tended extrafloral nectaries in the protection and benefit of a Neotropical rainforest tree. Oecologia 118: 192-202

Feldhaar H, Fiala B, Hashim R bin, Maschwitz U (2000) Maintaining an ant-plant symbiosis: secondary polygyny in the Macaranga triloba-Crematogaster sp. association. Naturwissenschaften 87: 408-411

Gaume L, McKey D, Anstett MC (1997) Benefits conferred by "timid" ants: active anti-herbivore protection of the rainforest tree Leonardoxa africana by the minute ant Petalomyrmex phylax. Oecologia 112:209-216

Gaume L, Matile-Ferrero D, McKey D (2000) Colony foundation and acquisition of trophobionts by Aphomomyrmex afer (Formicinae): co-dispersal of queens and phoretic mealybugs in an ant-plant-homopteran mutualism? Insectes sociaux. 47: 84-91

Heil M, Hilpert A, Fiala B, Linsenmair KE (2001) Nutrient availability and indirect (biotic) defense in a Malaysian ant-plant. Oecologia 126: 404-408

Heil M, Fiala B, Maschwitz U, Linsenmair KE (2001) On benefits of indirect defense: short- and long-term studies of antiherbivore protection via mutualistic ants. Oecologia 126: 395-403

Meunier L, Dalecky A, Berticat C, Gaume L, McKey D (1999) Worker size variation and the evolution of an ant-plant mutualism: Comparative morphometrics of workers of two closely related plant-ants, Petalomyrmex phylax and Aphomomyrmex afer (Formicinae). Insectes Sociaux. 46: 171-178

Sagers CL, Ginger SM, Evans RD (2000) Carbon and nitrogen isotopes trace nutrient exchange in an ant-plant mutualism. Oecologia 123: 582-586