Jump to ContentJump to Main Navigation
The Ornaments of LifeCoevolution and Conservation in the Tropics$

Theodore H. Fleming and W. John Kress

Print publication date: 2013

Print ISBN-13: 9780226253404

Published to Chicago Scholarship Online: January 2014

DOI: 10.7208/chicago/9780226023328.001.0001

Show Summary Details
Page of

PRINTED FROM CHICAGO SCHOLARSHIP ONLINE (www.chicago.universitypressscholarship.com). (c) Copyright University of Chicago Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in CHSO for personal use (for details see http://www.chicago.universitypressscholarship.com/page/privacy-policy). Subscriber: null; date: 17 January 2018

The Future of Vertebrate-Angiosperm Mutualisms

The Future of Vertebrate-Angiosperm Mutualisms

(p.441) 10 The Future of Vertebrate-Angiosperm Mutualisms
The Ornaments of Life

Theodore H. Fleming

W. John Kress

University of Chicago Press

Abstract and Keywords

This chapter reviews the conservation status of plant-visiting vertebrates and the ecosystem services they provide. It describes their major threats and their ecological consequences, and offers an overview of how these threats can be mitigated. It begins with a discussion of three recent studies that illustrate the ecological consequences of disrupted mutualistic interactions between vertebrates and their food plants.

Keywords:   conservation, plant-visiting vertebrates, ecological impact, mutualisms, vertebrates, food plants

As described in the first nine chapters of this book, a great diversity of mammals and birds are dependent on plants for nutrition in the form of nectar and/or fruit. Likewise, many species of plants are dependent on vertebrates as vectors for pollen transfer or dispersal of propagules. In both vertebrates and plants, these mutualisms are spread across a broad phylogenetic spectrum and have evolved repeatedly in different tropical and temperate habitats. Pollination and seed dispersal, often described as “ecosystem services,” are critical species interactions for the long-term functioning of ecological networks and communities. Although perhaps not as dramatic as the ecological consequences associated with the loss of apex predators in many marine and terrestrial ecosystems (Estes et al. 2011), the breakdown of these interactions will undoubtedly have a cascading effect that begins with the mutualists themselves and may eventually encompass a large segment of the entire community (e.g., Jordan 2009; Kearns et al. 1998; Kiers et al. 2010; Pauw and Hawkins 2011; Traill et al. 2010).

Many if not all of the habitats where these mutualisms are found are in the midst of profound alterations as a result of human activities. Whether these activities result from the direct degradation of environments locally or at landscape scales or from indirect modifications due to climate change, the future of these mutualisms, which have evolved over millions of years, is uncertain. The threats to plant-vertebrate mutualisms include habitat fragmentation, invasive species, diseases, and bushmeat hunting. Each of these threats varies in time and space depending on geographic location and habitat type, but taken together these human-based ecological pressures (p.442) have the potential to profoundly affect the mutualistic interdependencies among tropical vertebrates and their botanical partners. As a result of these threats, many plant-visiting birds and mammals are species of considerable conservation concern today.

Our aim in this final chapter is to briefly review the conservation status of plant-visiting vertebrates and the ecosystem services they provide before we describe their major threats and their ecological consequences and provide an overview of how these threats can be mitigated. The literature on the conservation of tropical habitats and their inhabitants is enormous, and we will take a very broad-brush approach to this topic here. Entries into this vast literature can be found in Corlett (2009a), Corlett and Primack (2011), Ghazoul and Shell (2010), Laurance and Peres (2006), and Sodhi et al. (2007), among others. As an introduction to this topic, we will describe three recent studies that illustrate the ecological consequences of disrupted mutualistic interactions between vertebrates and their food plants. These disruptions, of course, also have evolutionary consequences, but ecological consequences generally occur over a much shorter time-span and will be our focus here.

Although it is well-known that populations of various pollinating and seed-dispersing vertebrates have been significantly affected and in some cases have seriously declined as a result of a variety of different threats (see discussion below), documented examples of the direct impact on their dependent plant species are few. It is relatively easy to determine if the absence of one partner in a very specialized mutualism has an immediate effect on the other member with regard to a specific service, for example, pollination or seed dispersal, during a single season. However, documentation of the long-term effects over multiple seasons or multiple generations is a much more difficult and time-intensive task, especially if the mutualisms involve multiple partners in a more generalized interaction system involving long-lived plants, as is the case of many seed dispersal mutualisms. These kinds of studies involving vertebrate mutualists are still uncommon.

One of the best documented cases of the impact of the decline of vertebrate pollinators on their host flowering plant species involves a mutualism that evolved on the North Island of New Zealand (Anderson et al. 2011; also see chap. 4). Rhabdothamnus solandri (Gesneriaceae) is an endemic bird-pollinated member of the forest understory. Its tubular, yellow and pink flowers depend on three species of endemic nectar-feeding birds for pollination: the bellbird (Anthornis melanura) and the tui (Prosthemadera novaeseelandiae—both Meliphagidae—and the stitchbird (Notiomystis cincta; (p.443) Notiomystidae). Native silvereyes (Zosterops lateralis; Zosteropidae) also visit the flowers but usually rob nectar rather than pollinate the flowers. Shortly after the introduction of nonnative mammal predators to the North Island in the late 1870s, two of the three species of pollinators (bellbirds and stitchbirds) were extirpated. Fortunately, all three birds are still present on several small adjacent islands where the plants also occur. By comparing the reproductive success of this shrub on the North Island versus the smaller islands, Anderson et al. were able to demonstrate that fruit set and seed number were reduced by 84% due to the absence of two of the native pollinators. Even more important, despite the persistence of adult shrubs in mainland forests, seedling recruitment was also reduced there by >50%. They concluded that even though one native species of pollinating bird remained, its services were not sufficient to compensate for the extirpation of the other two species. Furthermore, they observed that the tui often preferred to visit nonnative species of plants with more nectar-rich flowers than Rhabdothamus, thereby perhaps increasing the negative impact on reproductive success of this shrub.

This study has at least three important messages. First, partners in a pollinator-plant mutualism can be eliminated due to the introduction of a nonnative predator into the environment. Island mutualisms are particularly vulnerable to this threat (Traveset and Richardson 2006). Second, the extirpation of two of the three native pollinators had a dramatic effect on the pollination success of Rhabdothamus shrubs during a single season as well as a cascading effect on the abundance of seedlings from past reproductive generations. Finally, it showed how an introduced nonnative plant competitor with nectar-rich flowers may have further reduced the service of the sole remaining native pollinating bird in the habitat. The main conclusion of this study was that the decline and eventual extirpation of native bird pollinators in a local population had cascading effects on the reproductive success of their mutualistic plant species. We suspect that this will be a general result from the disruption of many vertebrate-plant mutualisms.

This study has at least three important messages. First, partners in a pollinator-plant mutualism can be eliminated due to the introduction of a nonnative predator into the environment. Island mutualisms are particularly vulnerable to this threat (Traveset and Richardson 2006). Second, the extirpation of two of the three native pollinators had a dramatic effect on the pollination success of Rhabdothamus shrubs during a single season as well as a cascading effect on the abundance of seedlings from past reproductive generations. Finally, it showed how an introduced nonnative plant competitor with nectar-rich flowers may have further reduced the service of the sole remaining native pollinating bird in the habitat. The main conclusion of this study was that the decline and eventual extirpation of native bird pollinators in a local population had cascading effects on the reproductive success of their mutualistic plant species. We suspect that this will be a general result from the disruption of many vertebrate-plant mutualisms.

In a comparable, though more restricted study, Traveset and Riera (2005) demonstrated a similar disruption to the reproductive success of a perennial shrub on the Mediterranean island of Menorca as a result of the decline of its vertebrate seed disperser. On this island the extirpation of the frugivorous lizard Podarcis lilfordi (Lacertidae) over 2,000 years ago by introduced carnivorous mammals has had a significant effect on seed dispersal and seedling recruitment of Daphne rodriguezii (Thymelaeaceae) in coastal (p.444) shrublands, where it is now considered at risk of extinction. As in the case of the extirpation of the pollinators of the New Zealand shrub, the researchers were able to show that on isolated islands near Menorca where the native lizards still persist, seedling recruitment of this plant was significantly greater.

Finally, as described in chapter 4, McConkey and Drake (2006) showed that reduced densities of Pteropus fruit bats on islands of the Tongan archipelago in the South Pacific have resulted in significantly reduced levels of seed dispersal in several species of large-seeded trees. Unfortunately, unlike the above two studies, the consequences of this for seedling recruitment and future plant population growth have not yet been studied but will likely be significant.

These studies are among the few detailed examples of the decline of a mutualism in a natural habitat as the result of the extirpation (or nearextirpation) of one or more vertebrate partners. However, it is not difficult to foresee how such a process could be extrapolated to similar cascades in many other environments with other species of plants, pollinators, and dispersers (see below). Most ecological models of parasites and mutualists suggest that such coextinctions may be the most common type of biodiversity decline and may result in the loss of thousands of species (Dunn et al. 2009; Koh et al. 2004). These particular cases also suggest that specialization in a mutualism may result in a greater risk of extinction of the dependent species (Sekercioglu 2004). More studies are needed, however, before any concrete generalizations can be made about such coextinctions.

The Conservation Status of Plant-Visiting Birds and Mammals

Before we discuss factors that threaten the integrity of mutualisms between tropical vertebrates and their food plants, we need to review the conservation status of families of vertebrate mutualists and the ecosystem services they provide. In this section we address the question of how endangered these animals are. The International Union for Conservation of Nature monitors the population status of the world's plants and animals, and we have used their 2004 summary (Baillie et al. 2004) to examine the status of the families of nectar-feeding and fruit-eating birds and mammals listed in table 1.1. In that table we indicate the percentage of “threatened” species in most families, as reported in appendixes 3d and 3f of Baillie et al. (2004). We summarize those data in figure 10.1. By “threatened,” the International (p.445)

The Future of Vertebrate-Angiosperm Mutualisms

Figure lO.l. Frequency distribution of families of nectar-feeding (A) or fruit-eating (B) birds and mammals in terms of number of “threatened” species according to the International Union for Conservation of Nature Red Book (Baillie et al. 2004).

Union for Conservation of Nature means “threatened with extinction.” Of course, all species on Earth have finite life expectancies and thus face extinction sooner or later. But as a result of the direct and indirect activities of our species, extinction rates of many taxa are currently orders of magnitude greater than background levels. It is this elevated level of threat that we are concerned with here.

As seen in figure 10.1, levels of threat vary both ecologically and taxonomically. Fewer families of nectarivores contain high percentages (i.e., >50%) of threatened species than frugivores, and mammals contain more highly threatened families than birds. Among nectarivores, only Hawaiian honeycreepers (Fringillidae, Drepanidinae) are highly threatened and have suffered the loss of numerous species in historic times (Ziegler 2002). This family or subfamily is an extreme example of the precarious nature of island life, particularly once islands have been colonized by Homo sapiens. One reason why most vertebrate nectarivores do not currently face imminent extinction is that they are small and are not routinely hunted for food by (p.446) humans. Large generalist pteropodid bats that eat both nectar and fruit are an exception to this. They are avidly hunted and eaten in many parts of Southeast Asia and the South Pacific (Mickleburgh et al. 2009). As a result, about 40% of their species are currently classified as “threatened”

Most mammalian frugivores are larger than their avian counterparts and are being severely hunted as bushmeat throughout the tropics (Fa and Brown 2009). As a result, a substantial number of their families contain many threatened species. These include families of medium-to-large primates weighing >2 kg as well as truly large mammals such as elephants and tapirs. Among frugivorous birds, only cassowaries and guans are currently highly endangered and are also important sources of bushmeat. Most species of small frugivorous birds and mammals are not currently being threatened by bushmeat hunting.

How does the threatened status of families of plant-visiting birds and mammals compare with birds and mammals in general? Is nectarivory or frugivory a riskier feeding adaptation than other feeding modes? While this question deserves a rigorous, phylogenetically controlled analysis (e.g., Davies et al. 2008), we will provide a first-order answer by simply comparing the cumulative threat-status distributions of all families of nonmarine birds and mammals with those of families of nectarivores and frugivores. Results are shown in figure 10.2. Families of avian nectarivores and frugivores appear to be no more threatened than other bird families. The cumulative percentage curves of all three groups are nearly identical (fig. 10.2A). There are too few families of mammalian nectarivores (n = 3) for a meaningful comparison, but it is likely that their cumulative curve is similar to the curve for all nonmarine mammals (fig. 10.2B). In contrast, the cumulative curve for frugivores lags behind that for all mammals (i.e., more families of frugivores are threatened than all families of mammals), but the difference is not significant in a Kolmogorov-Smirnov two-sample test (Dmax [0.215] < D005 [0.313]). From these results, we cannot conclude that being frugivorous is riskier in mammals than having another diet. Nonetheless, because frugivorous mammals tend to be large, they often represent a large portion of bushmeat in Central Africa and the Amazon (Fa and Brown 2009). Therefore, in an ecological sense, being a large frugivorous bird or mammal is a risky life style.

Our overall conclusion here is that being nectarivorous or frugivorous does not necessarily predispose birds and mammals to higher risks of extinction than seen in other kinds of nonmarine birds and mammals. But this conclusion is no cause for complacency. As described by Hoffmann et al. 2010, (p.447)

The Future of Vertebrate-Angiosperm Mutualisms

Figure 10.2. Cumulative frequency distributions of families of birds (A) and mammals (B) in terms of number of threatened species by diet.

substantial numbers of birds and mammals are currently classified as threatened, especially in the tropics, and these numbers are steadily increasing. Increased conservation efforts are clearly needed to halt the slow but steady march of many species toward extinction. And, as we have emphasized repeatedly in this book, the loss of vertebrate mutualists will have profound ecological and evolutionary consequences for their ecosystems.

The Effects of Threats on the Mutualistic Partnership

One critical question in the conservation of mutualistic relationships is how widespread, taxonomically, the threats are to one of the partners as a result of the extinction of the other partner. We coupled the APG III phylogenetic tree (fig. 6.3) with the International Union for Conservation of Nature Red List of Threatened Animals in an attempt to understand the phylogenetic distribution of plant taxa across the angiosperms that may suffer decline and even possibly extinction as a result of the current threats to their mutualist (p.448) fruit and seed dispersers. The families of animals that are particularly specialized for fruit dispersal (those in bold in table 1.1) were analyzed with respect to the most important plant orders and families associated with their dispersal services based on table 3.9. Taxonomic families of animal fruit dispersers were sorted into those that have 25%–50% of their species threatened with extinction and those that have >50% (up to 100%) threatened with extinction according to the Red List (table 1.1). Although some of the minor animal disperser families (those not in bold in table 1.1) may also have high extinction risks, their relatively small effect on plant dispersal makes them less important for this analysis. The most important families of animal dispersers, their levels of threat of extinction, and their associated plant orders and families are listed in table 10.1.

We recognize the fact that only some of the species in these orders and families are dispersed by animals threatened with extinction and that the majority is not. Nonetheless, this analysis is intended to emphasize primarily the phylogenetic distribution of plant groups that may be affected as a result of the threat to the seed dispersers of some of their species. The plant orders associated with these major dispersal groups can be categorized according to the level of threat to their dispersers. We therefore assigned each of these plant orders to one of three character states (0 = <25% dispersers threatened, 1 = 25%–50% dispersers threatened, 2 = >50% dispersers threatened) and then mapped these states onto the APG III phylogeny using Mes-quite. The results suggest that four of the five major groupings of flowering plants (basal angiosperms, monocots, asterids and rosids) have at least one or more lineages that contain species that may be affected by the extinction of their dispersal mutualists (fig. 10.3). Only the basal eudicots are not affected, while the rosids (especially Anacardiaceae, Burseraceae, Clusiaceae, Combretaceae, Elaeocarpaceae, Fabaceae, Meliaceae, Moraceae, Myrtaceae, Ulmaceae, Salicaceae, and Sapindaceae) and asterids (especially Cactaceae, Ebenaceae, Rubiaceae, Sapotaceae, and Solanaceae) contain the most families with members who have fruits dispersed by animals under a significant threat of extinction. Several families in the basal angiosperms (Annonaceae, Myristicaceae, and Piperaceae) and monocots (Musaceae) also contain species that are dependent on animal dispersers that are increasingly under threat of extinction.

These results emphasize our earlier observation, discussed in chapter 8, that the dispersal of fruits by specific animal mutualists has evolved repeatedly across the flowering plants. Our results also suggest that as one side of (p.449)

Table 10.1 Proportion of Threatened Species among Major Fruit and Seed Dispersal Agents and Primary Associated Plant Groups

Animal Dispersal Group

Percentage of Species Threatened

Mutualistic Orders (Families) of Plants

25–50% threatened species:

    Old World fruit bats (Pteropodidae)


Ericales (Ebenaceae), Myrtales (Combretaceae, Myrtaceae), Rosales (Moraceae), Sapindales (Anacardiaceae), Zingiberales (Musaceae)

    American leaf-nosed bats (Phyllostomidae)


Caryophyllales (Cactaceae), Malpighiales (Clusiaceae), Piperales (Piperaceae), Rosales (Moraceae), Solanales (Solanaceae)

    Marmosets /capuchins (Cebidae)


A subset of plants eaten by spider monkeys and howlers

    Night monkeys (Aotidae)


A subset of plants eaten by spider monkeys and howlers

    Sakis /titis (Pitheciidae)


Ericales (Lecythidaceae), Fabales (Fabaceae), Myrtales (Myrtaceae), Sapindales (Sapindaceae)

    Spider monkeys /howlers (Atelidae)


Ericales (Sapotaceae), Fabales (Fabaceae), Magnoliales (Annonaceae, Myristicaceae), Malpighiales (Salicaceae), Rosales (Moraceae)

    Old World monkeys (Cercopithecidae)


Ericales (Ebenaceae, Sapotaceae), Fabales (Fabaceae), Magnoliales (Annonaceae), Rosales (Moraceae, Ulmaceae), Sapindales (Meliaceae, Sapindaceae)

    Raccoons (Procyonidae)


Rosales (Moraceae)

    Palm civets (Viverridae)


Magnoliales (Annonaceae)

<50% threatened species:

    Cassowaries (Casuariidae)


Laurales (Lauraceae), Myrtales (Myrtaceae), Oxalidales (Elaeocarpaceae), Rosales (Moraceae)

    Large lemurs (Lemuridae)


Gentianales (Rubiaceae), Malpighiales (Clusia-ceae, Euphorbiaceae), Rosales (Moraceae)

    Gibbons (Hylobatidae)


Ericales (Ebenaceae, Sapotaceae), Magnoliales (Annonaceae), Malpighiales (Euphorbiaceae), Rosales (Moraceae), Sapindales (Meliaceae)

    Great apes (Hominidae)


Ericales (Ebenaceae, Sapotaceae), Fabales (Fabaceae), Magnoliales (Annonaceae), Malpighiales (Euphorbiaceae), Rosales (Mora-ceae), Sapindales (Burseraceae, Meliaceae)

Note. The animal dispersal groups are categorized by the percentage of threatened species within each family according to the IUCN Red List.

the mutualistic partnership is threatened with extinction by habitat degradation and other processes discussed in this chapter, then a similar fate may await their host plant partners distributed across the angiosperms as well.

Ecosystem Services Provided by Plant-visiting Vertebrates, Redux

In recent years it has become fashionable to discuss the conservation of nature in terms of the ecosystem services provided to our species. Daily (1997, 3) defines these services as “the conditions and processes through (p.450)

The Future of Vertebrate-Angiosperm Mutualisms

Figure 10.3. Phylogenetic distribution across the flowering plants of orders containing species that will most likely be affected by the extinction of their mutualist fruit dispersers.

(p.451) which natural ecosystems, and the species that make them up, sustain and fulfill human life. They maintain biodiversity and the production of ecosystem goods [Daily's italics], such as seafood, forage, timber, biomass fuels, natural fiber, and many pharmaceuticals, industrial products, and their precursors…. In addition to the production of goods, ecosystem services are the actual life-supporting functions, such as cleansing, recycling, and renewal, and they confer many intangible aesthetic and cultural benefits as well.” Often these human benefits are given an economic (dollar) value, presumably to indicate why such services are essential to humans. For example, to highlight the ecological importance of bats to mankind, Boyles et al. (2011) estimated that insectivorous bats in North America annually prevent agricultural losses of more than US$3.7 billion as a result of insect suppression. But whether this knowledge will substantially increase efforts to conserve North American bats remains to be seen.

Although we have not stressed their economic benefits in this book, it is clear that the pollination and seed dispersal services provided by plant-visiting birds and mammals have played a very important role in the ecology and evolution of tropical (and subtropical) habitats. We eschew attempting to place a monetary value on these services, however, simply because in our estimation their value is inestimable. How does one really put a dollar value on biodiversity and the ecological services it provides? What is the economic value of an Amazonian hermit hummingbird and the plants it pollinates or a Sumatran orangutan and the plants it disperses? In our opinion, these animals and their habitats must be conserved regardless of their economic (or medical, etc.) contributions to Homo sapiens. For our species to do otherwise would be unconscionable and unethical.

Despite our view of the folly of trying to put a monetary value on wildlife and its ecological services, two recent reviews have attempted to do so for birds and mammals (Kunz et al. 2011; Whelan et al. 2008), and we will summarize some of their results here. No such reviews currently exist for primates (C. Chapman, pers. comm.; but see Astaras et al. [2010] for a review of seed dispersal differences between terrestrial and arboreal African primates and their conservation implications). Whelan et al. (2008) indicated that birds provide four essential services recognized by the United Nations Millennium Ecosystem Assessment: provisioning (i.e., production of fiber, clean water, or food), regulating (i.e., affecting climate, water, and human disease), cultural (e.g., recreation or aesthetics), and supporting (i.e., all other ecosystem services). Pollination and seed dispersal represent supporting (p.452) services and, as this book emphasizes, birds are important pollinators and seed dispersers of plants worldwide, though Whelan et al. indicate that birds are important pollinators of only a few human food crops (i.e., about 5.4% worldwide based on estimates by Nabhan and Buchmann [1997]). In their table 2 they also indicate that birds are important seed dispersers worldwide but focus only on temperate North American plant genera that are bird dispersed. Economically important plants dispersed by birds in North America are mainly trees (e.g., Pinus, Juniperus, Magnolia, Fagus, Quercus, Prunus, etc.). A much greater diversity of plants are dispersed by birds in the tropics (see, e.g., chaps. 4 and 5). These include many trees that are likely to have significant economic value, at both local and international levels (e.g., species of Lauraceae, Meliaceae, Myristicaceae, etc.; Corlett 2009a).

Kunz et al. (2011) review the ecosystem services of bats, focusing on tropical as well as temperate systems. In table 10.2 we indicate the major groups of plants that have economic or especially important ecological value that are either pollinated or dispersed by bats, and this impressive array of plants has considerable significance as sources of food and timber locally or globally. Kunz et al. (2011, table 4) provide details about how these plants are used. However, putting a dollar value on these plants, particularly in terms of the relative contribution of bats compared with other pollinators or dispersers, is extremely difficult, if not impossible, for at least two reasons. First, as discussed in chapter 4, we lack detailed knowledge of the relative contributions of bats to pollination and/or seed dispersal for most of these plants. Second, many of these species are now in cultivation and do not need the services of animal pollinators or seed dispersers for their propagation. Commercially grown species of Agave and Musa are two good examples. The wild progenitors of commercial crops still need animal services for their reproduction and genetic diversity, but their “domestic” relatives do not.

It is even more difficult to assign a dollar value to the pollination and seed dispersal services of bats and other vertebrates for plants that play especially important roles in the regeneration and maintenance of plant community structure (e.g., table 10.2B). In our view, these services are priceless, and conservation of the species responsible for ensuring the reproduction of ecologically important plants in habitats throughout the world is critical. That being said, the fate of many of these species of birds and mammals hangs in the balance today. We now turn our attention to the threats faced by these animals and what steps need to be taken locally and globally to ensure their survival. (p.453)

Table 10.2. Examples of Economically and Ecologically Important Tropical and Subtropical Plants Serviced by Bats

Plant Family and Subfamily



A. Economically important plants:


Anacardium occidentale, Mangifera indica, Spondias spp.



Annona spp.



Anthurium and Philodendron spp.



Acrocomia, Astrocaryum, Bactris, Euterpe, Phoenix, Prestoea, Roystonea, Sabal, and Socratea spp.



Agave spp., subgenus Agave



Cordia dodecandra



Many genera in subfamily Cactoideae, tribe Pachycereeae

Pollinated, Dispersed


Carica papaya




Pollinated, Dispersed


Cecropia peltata



Chrysobalanus icaco



Clusia, Symphonia, and Vismia spp.



Terminalia catappa



Carludovica palmata



Diospyros digyna, D. kaki


    Fabaceae, Faboideae

Andira inermis, Dipteryx odorata


    Fabaceae, Mimosoideae

Inga vera, Parkia speciosa

Pollinated, Dispersed


Lecythis pisonis



Malpighia glabra


    Malvaceae, Bombacoideae

Ceiba spp.


    Malvaceae, Helicteroideae

Durio and Ochroma spp.


    Malvaceae, Sterculioideae

Guazuma ulmifolia



Brosimum alicastrum; Artocarpus and Ficus spp.



Muntingia calabura



Musa spp.

Pollinated, Dispersed


Anomomis umbellulifera and Psidium guajava; Syzgium spp.



Passiflora spp.



Piper aduncum



Coccoloba uvifera



Eriobotrya japonica



Coffea arabica



Casimiroa edulis



Flacourtia indica



Meliococcus bijugatus, Sapindus saponaria



Chrysophyllum cainito, Mimusops elengi; Manilkara and Pouteria spp.



Trema micrantha



Vitus vinifera


B. Ecologically important plants:


Agave spp.



Many New and Old World genera


    Cactaceae, Cactoideae

Many columnar cacti in several tribes of this subfamily

Pollinated, Dispersed


Cecropia spp.



Vismia spp.


    Malvaceae, Bombacoideae

Adansonia, Bombax, Ceiba, Pachira, Pseudobombax, etc. spp.



Ficus spp.



Piper spp.



Solanum spp.



Trema micrantha


Source. Kunz et al. (2011), which should be consulted for further details.

(p.454) Threats to Vertebrate Pollination and Seed Dispersal Mutualisms

Many different types of changes in a particular environment may have deleterious effects on one or the other partner in a plant-vertebrate mutualism. These environmental alterations may be natural, such as severe geologic upheavals, including volcanic eruptions and extraterrestrial impacts, or localized immediate catastrophes, including weather-related storms (e.g., Rathcke 2000). However, the most significant immediate threats to the majority of species are caused by human activities. These threats include habitat modifications and fragmentation, introduced and invasive species, pathogens and diseases, bushmeat hunting, and commercial wildlife trade. Climate change, primarily in response to increased carbon dioxide in the atmosphere, is a separate category, which many conservationists now believe may be the biggest threat to biodiversity and hence plant-vertebrate mutualisms (Dawson et al. 2011).

Environmental Degradation

The primary threat to tropical plant-vertebrate mutualisms is the degradation and conversion of natural habitats to human-dominated landscapes. Whether this conversion is due to small-or large-scale agroforestry, agriculture, logging, mining, wildfires, or simply human population expansion, the result is often the same: disruption of the interaction between and/or population decline and even extinction of one or more mutualistic partners (Keitt 2009). For example, in a survey of a wet forest site in Thailand, about one-third of the plant species were assessed to be vulnerable to extinction because they are dispersed mostly by large frugivores, which were deemed intolerant of any human impact on the environment (Kitamura et al. 2005). These authors suggested that when plant extinctions take place, these forests may become dominated by plant species dispersed by abiotic vectors or species with small-seeded fruits (also see chap. 4). A study in the Neo-tropics showed that plant species with bird- and monkey-dispersed fruits were more common in intact forest habitats, whereas species with passive fur-dispersed seeds were more common in deforested habitats (Mayfield et al. 2006). In their review of recent studies of avian responses to forest conversion, Tscharntke et al. (2008) found two trends. First, species richness of large frugivores and insectivores (especially terrestrial and understory species) declines in agroforests. And second, nectarivores, small-medium insectivores (p.455) (especially migrants and canopy species), omnivores, and sometimes granivores and small frugivores do better or thrive in agroforestry. In general, agroforest bird communities are richer in species of nectar- and fruit-eating birds than intact forests or nontree agricultural habitats, and agricultural habitats have higher proportions of granivores and lower proportions of insectivores, frugivores, and nectarivores than agroforests or intact forests. Similar results have been reported for bird communities in four regions of Costa Rica (Karp et al. 2011). Our overall conclusion here is that different trophic groups of vertebrate mutualists (chap. 9) respond differently to human-modified, formerly forested habitats. Small species of birds and mammals appear to be more ecologically flexible than large species.

The transition from primary to disturbed habitats may not always be abrupt or immediate, and hence the transition in the status of a mutualism may take place in stepwise fashion. With respect to one specific activity that is an increasing threat to tropical wet forests, namely, wildfires caused by human activities, it has been shown that specific families of trees that produce fleshy fruits become less abundant than expected in once- and twice-burned habitats in central Amazonia, suggesting that tree mortality can be nonran-dom in terms of fruit dispersal type (Barlow and Peres 2006). Moreover, populations of large frugivores declined significantly in response to single fires, and most primary forest dispersal specialists were extirpated from twice-burned forest habitats. Such studies suggest that habitat effects on the disruption of plant-vertebrate mutualisms will certainly vary with the specific partners involved and the degree of specialization of the interactions.

An as yet unanswered question is how rapidly mutualisms present in tropical forests will recover after these disturbances and degradation. Although sufficient field and modeling data are lacking to make accurate predictions, one study on vertebrate dispersal agents in the Atlantic coastal forests of Brazil estimated that once a habitat is significantly altered, it may require 100–300 years to regain the percentage of animal-dispersed species (80% of the total plant species) found in mature predisturbance forests (Liebsch et al. 2008).

The conversion of primary habitats to agricultural lands is one of the principle sources of environmental degradation in both the temperate zone and the tropics. Most studies on the effects of widespread agriculture on pollinators have been focused on insects, especially bees, and have demonstrated both positive and negative responses (Burkle and Alarcón 2011; Potts et al. 2010). This mixed response is because bees are pollinators of (p.456) both native plant species as well as many agricultural species. In some cases, pollinator decline is due to the absence of native species, while in others, pollinators increase in abundance because of an increase in total floral resources. Although data are limited on vertebrate pollinators and dispersers in this context, it has been shown that bird diversity and abundance decline in tropical agricultural environments, such as coffee plantations, unless these habitats are supplemented with native fruiting plants as shade trees, living fences, and windbreaks or unless forest remnants are preserved close to the plantations (Luck et al. 2003).

This same pattern has also been shown for the abundance of plant-visiting bats in the agricultural landscape. In a comparison of secondary forests, riparian forests, forest fallows, live fences, pastures with high tree cover, and pastures with low tree cover in Nicaragua, Medina et al. (2007) showed that riparian forests had the highest mean bat species density and abundance whereas the lowest values were found in pastures with low tree cover. Results of this study suggest that agricultural landscapes must retain a heterogeneous assemblage of tree species in order to maintain a diverse bat assemblage. This study contrasts with a similar investigation in lowland Amazonia where frugivores and nectarivores were abundant in areas that had been converted to agriculture (Willig et al. 2007). However, these authors cautioned that their results may have been scale dependent and that if habitat conversion to agricultural use continues to fragment the landscape, then source populations of bats may decline severely.

With respect to the effects of logging and secondary growth on volant vertebrates, several studies in the Amazon region have shown that the abundance of nectarivorous and frugivorous phyllostomid bats (Glossophaginae, Lonchophyllinae, Carolliinae, and Stenodermatinae) increased in logged sites where the canopy was more open and the understory was denser and in less-disturbed areas of Cecropia-dominated regrowth (e.g., Peters et al. 2006). In some cases many phyllostomids disappeared from habitats that experienced constant disturbance, whereas frugivorous stenodermatids favored heavily altered areas (Bobrowiec and Gribel 2010.). In tropical forests of southeastern Mexico, Castro-Luna et al. (2007) found that common species of frugivorous bats were very abundant in young successional stages, but rare species inhabited only primary forest and seldom foraged in secondary growth, even when it was close to primary vegetation. These results suggest that a mosaic of habitats modified by human activities, in combination with original vegetation, may be necessary to conserve the full spectrum of (p.457) frugivorous and nectarivorous phyllostomid bats in an area (Castro-Luna et al. 2007).

Similarly, the abundance and diversity of nectarivorous and frugivorous birds in tree plantations established after major habitat degradation is in part dependent on the floristic composition of the regenerating vegetation. Understory frugivorous birds in Colombia appear to function at a larger spatial scale than the patchiness created by plantations, whereas nectarivorous birds respond to small-scale patchiness within secondary growth (Durán and Kattan 2005). Furthermore, patch size as well as tree species composition influence the number and duration of bird visits in forest restoration plots in southern Costa Rica (Fink et al. 2009). Investigations of the tolerance of frugivorous birds to habitat disturbance in a Costa Rican cloud forest indicate that large-bodied species are moderately tolerant of intermediate habitat disturbance but are intolerant of severe disturbance, whereas tolerance in medium- and small-bodied species is often greater in these same habitats (Gomes et al. 2008). As also seen in bats, the conservation of frugivorous birds will thus require a variety of forest types representing different levels of disturbance.

Habitat Fragmentation

The fragmentation of undisturbed habitats into a mosaic of pristine and degraded areas is an increasingly common precursor to widespread environmental destruction. Over the last several decades the role of fragmentation in biodiversity loss, including effects on mutualisms, has been widely investigated across the tropics (e.g., Laurance and Bieregaard 1997). With respect to pollination and dispersal mutualisms, it is overwhelmingly clear that plant reproduction is highly susceptible to habitat fragmentation and that fragmentation reduces plant reproductive success (Aguilar et al. 2006). Variation in both the life history traits of plants as well as habitat type has varying impacts on the effects of fragmentation. For example, trees with different-sized seeds usually respond differently; small-seeded plants are more resilient to forest fragmentation than large-seeded species (Cramer et al. 2007). In a study of fragmented habitats in Spain, the number of species with the ability for long-distance dispersal increased in more isolated patches in the highlands whereas the number of species with short-distance dispersal increased in isolated patches in the lowlands (Aparicio et al. 2008). In some cases the effects of fragmentation on pollination and dispersal are plant-species specific and are contingent on (p.458) the animal biota involved (Guariguata et al. 2002). For example, the fallen seeds of Dipteryx panamensis (Fabaceae) were dispersed more quickly by small scatter-hoarding rodents in forest fragments than in intact forest in and around La Selva, Costa Rica; fallen seeds of Carapa quianensis (Meliaceae), in contrast, were dispersed more quickly in intact forest than in forest fragments.

Edge effects and the type of vegetation matrix surrounding fragments also play a significant role in determining the types of mutualisms that survive fragmentation. In a study in the Atlantic forest of northeastern Brazil, secondary habitats and edges of fragments contained an assemblage of trees exhibiting reduced diversity of pollination systems, a higher abundance of reproductive traits associated with pollination by generalist diurnal vectors, and an increased abundance of hermaphroditic trees. From this, Lopes et al. (2009) concluded that narrow forest corridors and small fragments will become increasingly dominated by edge-affected habitats that will no longer contain the full complement of tree life history diversity and its attendant mutualists. Finally, climate change itself presents a potentially severe threat to biodiversity and mutualisms in fragmented habitats (see more below on climate change). As rates of environmental change accelerate due to increased C02 in the atmosphere, many species will be required to disperse rapidly through fragmented landscapes and large-scale corridors in order to keep pace with the changing climate. In many cases, standard dispersal mechanisms may not be able to respond rapidly enough to these changes (Pearson and Dawson 2005).

Birds may be particularly susceptible to habitat fragmentation because forest is the primary habitat for the majority of the avifauna with restricted geographic ranges—species that are especially prone to extinction (Oostra et al. 2008). Many of these bird species are important pollinators and seed dispersers in tropical habitats. For example, in a fragmented landscape in southern Costa Rica, nonforest hummingbirds occurred less frequently in fragments than in intact habitat, but fragments still supported a mixture of forest-interior and canopy-dwelling hummingbird species along with a diverse group of hummingbird-pollinated plants (Borgella et al. 2001). In contrast, in a species-rich scrub forest in Western Australia, the honeyeater pollinator community showed no significant response to fragment size (Yates et al. 2007a). Outcrossing rates of one bird-pollinated shrub were not significantly correlated with plant population or fragment size. However, mating in small populations occurred between much smaller numbers of (p.459) plants, which could affect population fitness in subsequent generations. Further observations on these same populations showed a strong positive correlation between the number of seeds produced per fruit and population and fragment size (Yates et al. 2007b), indicating that fragment size does matter in terms of plant reproductive success. Borgella et al. (2001) reached the same conclusion in their study.

With regard to seed dispersal, numbers of species of frugivorous birds (and some primates) in forest fragments in montane Tanzania declined with decreasing fragment size. In addition, recruitment of seedlings and juveniles of 31 animal-dispersed tree species was more than three times greater in continuous forest and large forest fragments than in small fragments (Cordeiro and Howe 2001). Interestingly, recruitment of wind- and gravity-dispersed trees of the forest interior was unaffected by fragmentation in this region. A similar study in Los Tuxtlas, Mexico, demonstrated that habitat disturbance influences avian visitation patterns to fruiting trees; this may in turn affect recruitment patterns in some tree species, although the results were not consistent among the tree species studied (Graham et al. 2002). This result led the authors to suggest that it may be difficult to generalize about the effects of forest fragmentation on assemblages of frugivorous birds. For large fruit-eating birds such as keel-billed toucans, guans, and some hornbills, fragmentation may not have large effects on their seed dispersal abilities because they appear to be able to move through degraded habitats and along corridors of vegetation connecting fragments (Graham 2001; Pizo 2004; Raman and Mudappa 2003). Nonetheless, for many birds, habitat reduction and forest edges have substantial effects on fruit selection, which will ultimately affect plant fitness in forest fragments (e.g., Galetti et al. 2003).

Bat pollinators and dispersers appear to respond quite differently than their avian counterparts to habitat fragmentation. It has been shown in Veracruz, Mexico, that a combination of continuous forest, forest fragments, and agricultural habitats helps to conserve a diverse assemblage of bat species in the local landscape (Estrada and Coates-Estrada 2002). Similarly, in the Atlantic forests of Paraguay, species richness of bats was highest in partly deforested landscapes, whereas evenness was greatest in forested habitat; the highest diversity of bats occurred in landscapes comprising moderately fragmented forest habitat (Gorresen and Willig 2004). Nonetheless, despite the abundance and diversity of bats in disturbed habits, the effects of fragmentation on pollination and seed dispersal services by bats are still unclear. In Australian tropical rain forests, the common blossom bat (Syconycteris (p.460) australis) is a significant pollinator and transports large quantities of pollen among flowers in forest fragments, but the quality (i.e., the effect of geographic genetic distance) of this pollen is not known (Law and Lean 1999). In Mexico, pollination and the reproductive success of the bombacaceous tree Ceiba grandiflora in dry forest habitats was negatively affected by habitat disruption. The effective pollinators of this species (Glossophaga soricina and Musonycteris harrisoni) visited flowers of trees in disturbed habitats significantly less often than trees in undisturbed habitats (Quesada et al. 2003). However, in a broader study of trees in the same plant family in Mexico and Costa Rica, the effects of forest fragmentation on bat pollinators, plant reproductive success, and mating patterns varied depending on the particular plant species, which suggests that the effects of forest fragmentation may differentially affect flowering patterns, bat foraging behavior, and plant self-incompatibility systems in these trees (Quesada et al. 2004).

In studies of bats as fruit and seed dispersers, Klingbeil and Willig (2009) demonstrated a guild-specific and scale-dependent response of bats to fragmented Amazonian rain forests. Abundance and richness of bats, including frugivorous species, were higher in moderately fragmented forest than in continuous forest. In contrast, large frugivores accounted for a higher proportion of total captures in continuous forest than in forest fragments, whereas small frugivores showed the opposite pattern in undisturbed forest and fragmented agricultural land in Peten, Guatemala (Schulze et al. 2000). These authors concluded that the relative abundances of large frugivores, which feed on large fruits of mature forest trees, and small frugivores, which feed on small-fruited plants occurring in early succession, are an indicator of forest disturbance. In Veracruz, Mexico, higher abundance and diversity of frugivorous bats were recorded for riparian sites than for isolated fruiting trees in pastures in a fragmented tropical landscape; the abundance of these bats decreased with distance from the nearest forest fragment (Galindo-Gonzalez and Sosa 2003). Although these few studies on bats are not necessarily consistent in the effects of fragmentation on bat diversity and abundance, they do suggest that bats may have a less severe response to habitat degradation than birds. In a comparison of bat- and bird-generated seed rain at isolated fruiting trees in pastures in fragmented tropical rain forest in Veracruz, Galindo-Gonzalez et al. (2000) found that seed diversity was similar between day and night seed captures and that the contribution of birds and bats to seed rain differed among seasons. Nonetheless, in this region both birds and bats are important seed dispersers of both pioneer and (p.461) primary species in pastures; their dispersal activities help to connect forest fragments and maintain plant diversity in fragmented tropical forests. This result is likely to have broad generality throughout the tropics.

In contrast to bats, other small mammals appear to be significantly affected by habitat disturbance and fragmentation. In a study in the Atlantic forests of Brazil, total abundance and alpha diversity of small mammal populations were lower in small and medium-sized fragments than in large fragments and continuous forest (Pardini et al. 2005). Moreover, isolated fragments showed lower diversity and abundance compared to connected fragments, which highlights the importance of corridors for buffering the effects of habitat fragmentation on small mammals in tropical landscapes. For mammal-dispersed woody plants in tropical dry forests, the number of species declined with decreasing forest cover in a number of reserves in Central America (Gillespie. 1999 Mammal-dispersed plants were rarest in the smallest fragments, perhaps as a result of the loss of dispersal vectors or because of other life history characteristics of the plants. For two Amazonian rodents (Myoprocta acouchy and Dasyprocta leporina), which are the most important dispersers of several large-seeded tree species, the larger species (D. leporina) was initially less affected by forest fragmentation than the smaller one, and continued fragmentation of Amazonian forests will most likely have strong negative consequences for the smaller species (Jorge 2008). In a comparison in a temperate fragmented forest of South America, it was found that the seeds of the mistletoe Tristerix corymbosus (Loranthaceae) are dispersed solely by the endemic marsupial Dromiciops gliroides (Rídriguez-Cabal et al. 2007). Fragmentation negatively affected marsupial abundance, fruit removal, seed dispersal, and seedling recruitment, and local extirpation of D. gliroides resulted in the complete disruption of mistletoe seed dispersal. Thus, the effects of forest fragmentation on this dispersal mutualism have clear demographic consequences for the survival of mistletoe populations.

Forest fragmentation can also have a profound effect on the behavior and ecology of primates, including their seed dispersal services. For example, forest fragmentation in the Orinoco Basin of Colombia has led to the local extinction of certain ateline monkeys (e.g., species of Lagothrix and Ateles) that originally inhabited the lowlands at the base of the Andes. Their absence has had negative effects on local plant populations because atelines play important roles as seed dispersers in these forests, especially for large-seeded plants (Stevenson and Aldana 2008). Ateline extinctions have (p.462) apparently resulted in reduced species diversity in local plant communities in this region. While many other studies on the effects of habitat fragmentation on primates exist (e.g., reviewed in Arroyo-Rodriguez and Dias 2010; Harcourt and Doherty 2005; Isabirye-Basuta and Lwanga 2008; Onderdonk and Chapman 2000), none of these have explicitly focused on the effect of fragmentation on seed dispersal by primates. Nonetheless, because we know that forest fragments generally contain a reduced subset of the local or regional primate fauna, smaller group sizes per species, a lower biomass of large tree species favored as food sources by primates, and primates that sometimes move among and out of patches to find food, we can make the following predictions. (1) Large-seeded, primate-dispersed trees that occur in forest patches characterized by the long-term absence of primates will experience severely limited seed dispersal (compared with the same species in intact forests). (2) Both dispersal and recruitment limitation (see chap. 4) will be greater in patches that do contain primates compared with intact forest. (3) Long-distance seed dispersal mediated by primates will still occur for trees located in forest fragments, but it will be much less frequent than in intact forest. (4) Seeds of forest trees deposited in agricultural fields or other nonforest habitats by primates will have lower probabilities of producing new recruits than those dispersed in intact forest. Overall, we expect that seed dispersal mutualisms between primates and their food plants will be disrupted by habitat fragmentation, as is likely to be the case in many other tropical vertebrate-plant mutualisms.

Study of the effect of habitat fragmentation on the structure and stability of mutualistic networks is in its infancy (reviewed by Gonzalez et al. 2011). As discussed above, we expect fragments to contain a subset of animals and plants that co-occur in intact habitats, but the extent to which this threatens the integrity of mutualistic networks is still an open question. Indeed, even delimiting the spatial scope of these networks in a fragmented landscape can be problematic. High mobility on the part of many pollinators and seed dispersers (chaps. 7 and 8) will help to mitigate to some extent the effects of fragmentation on plant reproductive success in habitat patches. Nonetheless, the persistence of many species, especially ecological specialists among both animals and plants, and the stability of mutualistic networks in the long run are likely to decrease with increasing habitat fragmentation. There can be no doubt that fragmentation will accelerate the loss of species and their interactions. The question then becomes, how many species and which interactions?

(p.463) Introduced and Invasive Species

When they are introduced into new habitats, birds and mammals can act as competitors with or, more frequently, as predators on native bird and mammal nectarivores and frugivores. As described in the example from New Zealand at the beginning of this chapter, an introduced plant species can also disrupt vertebrate-plant mutualisms by competing with native plant species for vertebrate services. How far-ranging these alterations may be and how resilient the interacting species are to such disruptions remains to be determined (Traveset and Richardson 2006). Here we discuss what is known about the effects of introduced species, primarily animals, on vertebrate pollinator and seed dispersal mutualisms. Most of these studies occur on islands where introductions can have dramatic effects. Less is known about the effect of animal introductions in mainland ecosystems.

One of the best documented cases of an alien animal species that has had a wide-ranging affect on the native flora and fauna is the brown tree snake (Boiga irregularis) on the Pacific island of Guam (Mortensen et al. 2008). In a study of two bird-pollinated tree species native to Guam—Bruguiera gymnorrhiza (a mangrove tree in the Rhizophoraceae) and Erythrina variegata (a forest tree in the Fabaceae)—flower visitation and seed set were both significantly higher on Saipan, where the tree snake is absent, than on Guam, where the invader has caused severe losses of native vertebrates. The authors concluded that the bird-pollinated tree species were highly dependent on avian visitors for reproduction and that the decimation of flower-visiting birds by this snake has severely disrupted mutualistic interactions, as we described for two other island systems above. Each of these examples demonstrates the cascading effects of introduced predators on both partners of a vertebrate-plant mutualism.

In a more general and widespread fashion, invasive species that become integrated into networks of interacting mutualists may significantly alter the web structure that has evolved within a community. Aizen et al. (2008) analyzed the extent of mutual dependencies between interacting species (primarily insects and plants) in forests of the southern Andes as well as on several oceanic islands with respect to the pervasiveness of different alien species. They found that weaker, more asymmetrical mutualisms were present in highly invaded webs whereas mutualisms tended to persist better in less-invaded webs. The presence of aliens did not alter overall network connectivity, but connections were transferred from generalist native species to “super-generalist” alien species after invasion. They concluded that the (p.464) introduction of alien species may leave native species open to new ecological and evolutionary dynamics with respect to their mutualistic partners.

In contrast to the study conducted in the Andes, a survey of plant-pollinator networks on the tropical island of Mauritius showed that newly introduced plant invaders had a relatively low impact on visitation rates to native plant species by both native invertebrate and vertebrate pollinators, suggesting that these invaders offered little direct competition for pollinators with native plant species (Kaiser-Bunbury et al. 2009). Conversely, the introduction of honey bees on the island of New Caledonia, which contains an evolu-tionarily unique flora, appears to have had a significant effect, including a change in patterns of gene flow, on a number of native plant species that were originally pollinated by native short-tongued bees (Kato and Kawakita 2004). Finally, nonpollinating introduced vertebrate species may also affect local plant-pollinator interactions. In one case, there is strong evidence that the introduction of domestic cattle has significantly modified the structure of local mutualistic networks. A comparison of plant-pollinator interaction networks in native forest sites with and without domestic cattle in montane Argentina demonstrated that the effect of cattle on the network structure was due primarily to the modification of only a few specific but common interactions, which resulted in a significant disruption of critical components of the overall network (Vazquez and Simberloff 2003). Although these studies were not focused solely on vertebrate pollinators, they nonetheless indicate that introduced exotic species have the potential to significantly disrupt plant-pollinator mutualisms.

Before we discuss the impact of invasive species with respect to specific partners of the mutualistic interactions, it is important to note that vertebrates, especially fruit and seed dispersers, have played a major role in the propagation of invasive plant species in new environments. This issue is becoming an area of particular importance in weed and environmental management (Buckley et al. 2006). Moreover, as fragmentation of landscapes becomes increasing common in both temperate and tropical habitats, many invasive plants and their native dispersers readily use these disturbed environments and fragment edges to increase the dispersal of nonnative species. Where invasive plants are an important part of the diet of native frugivores, a conflict will inevitably arise between the control of the invasive plants and the conservation of populations of these frugivores, especially in cases in which other environmental threats have already reduced populations of native fruit-producing species. For example, in eastern Australia the (p.465) Bitou bush, Chrysanthemoides monilifera (Asteraceae), is an invasive weed of coastal habitats that produces fleshy bird- and mammal-dispersed fruits. This species, along with other nonnative plant invaders, has substantially altered the temporal pattern of fruit availability in the coastal vegetation of this region. Characteristics of its fruits as well as its seasonal pattern of fruit production are particularly attractive to native frugivores and have contributed to its successful spread as an invasive (Gosper 2004).

Mutualisms between plants and birds, both as flower pollinators and as seed dispersers, have received more attention regarding the effects of invasive species than plants and mammals. In an Australian study of bird pollination that focused on the use of exotic and native nectar resources by native species of birds, it was found that native plant genera such as Banksia and Grevillea produced significantly higher volumes of nectar than such exotic genera as Camellia and Hibiscus (French et al. 2005. Banksias also produced a significantly higher sugar reward per floral unit than the other three genera. The three most common local meliphagid nectarivores—the red wattlebird (Anthochaera carunculata), the little wattlebird (Anthochaera chrysoptera), and the noisy miner (Manorina melanocephala)—all preferred flowers of Banksia and Grevillea and spent significantly more time in flowers of Banksia than in those of any other genus. Therefore, contrary to expectations, the native genera of plants were not only a more valuable food source than the exotic genera, but they were also the preferred flowers for these nectarivorous birds (French et al. 2005).

In contrast, invasive alien plants and insects may also impede natural regeneration of native plant species by altering plant-animal interactions such as pollination. In Mauritius, the pollination ecology of a rare endemic cauliflorous tree, Syzygium mamillatum, was compared in a restored forest (all alien plant species removed) and an adjacent unrestored area (degraded by alien plants; Kaiser et al. 2008). Flowers of this tree were visited only by generalist bird species. Results indicated that fruit set and the number of seeds per fruit were lower in the restored forest than in the unrestored forest as a result of lower bird visitation rates in the restored area. In addition to differences in bird visitation rates, the difference in reproductive performance of S. mamillatum between the two localities was also caused by differences in the attack rates of insect herbivores on flower buds, indicating that multiple interspecific interactions may have compounding effects in habitats altered by invasive species.

Introduced insects also have direct effects on the reproductive ecology (p.466) of bird-pollinated native plant species. The effects of introduced honey bees on the nectar-feeding activity of two endemic nectarivorous birds—the grey white-eye, Zosterops borbonicus mauritianus, and the olive white-eye, Z. chloronothos—at two endemic flowering trees, Sideroxylon cinereum and S. puberulum (Sapotaceae), was studied on Mauritius by Hansen et al. (2002). Results indicated that the introduced bees interfered with the interactions of endemic bird and plant species by reducing nectar levels and forcing birds to forage elsewhere. The authors suggested that native plant-pollinator mutualisms in island ecosystems may be especially vulnerable to disruption by introduced honey bees.

With regard to frugivorous birds and invasive species, most studies have concentrated on the effects of frugivores on the dispersal of introduced plant species. It is commonly assumed that exotic fruiting trees in degraded areas are attractive to frugivorous birds and may become centers of regeneration for invasive species. Supporting this is a study of the frugivore assemblage and seed rain/seedling establishment of exotic guava trees (Psidium guajava) in farmland adjacent to native forests in Kenya (Berens et al. 2008). Results indicated that 40 species of frugivorous birds visited guava trees, that 100% and 82% of the seed and seedling species found under guava crowns, respectively, were animal dispersed, and that the majority of these species were late-successional native forest species. Furthermore, the abundance of frugivorous shrubland birds, animal-dispersed seeds, and late-successional seeds at or under guava trees increased, rather than decreased, with increasing distance from primary forest. Therefore, even though guavas are an exotic species, these trees may have a positive effect on forest regeneration and may prove valuable for management plans concerning forest restoration in this area.

In contrast, Cordeiro et al. (2004) demonstrated that native bird disperses also had a significant effect on the range expansion of an introduced tree species. Their study examined whether several generalist avian frugivores facilitated the invasion of the exotic early successional tree Maesopsis eminii (Rhamnaceae) in the East Usambara Mountains, Tanzania. Hornbills were shown to disperse more than 26 times more seeds of this species than monkeys and more than three times as many seeds as turacos per visit and were thus considered the most important disperser of this tree. They concluded that the extensive invasion of M. eminii in the East Usambara Mountains was enhanced in both speed and spatial scale by the silvery-cheeked horn-bill, an extremely effective dispersal agent of this introduced tree.

The Hawaiian Islands have lost nearly all their native seed dispersers but (p.467) have gained many frugivorous birds and fleshy-fruited plants through introductions. In this fragile ecosystem, introduced birds may not only aid invasions of exotic plants but may also be the sole dispersers of native plants. In a study including both native- and exotic-dominated forests, Foster and Robinson (2007) showed that introduced species of birds were the primary dispersers of native seeds into exotic-dominated forests, which may have enabled six native understory plant species to become reestablished. Introduced birds also dispersed seeds of two exotic plants into native forest habitats, but dispersal was localized and establishment of these exotics was minimal. Without suitable native dispersers, most common understory plants in Hawaiian rainforests now depend on introduced birds for dispersal, and these introduced species may actually facilitate perpetuation, and perhaps in some cases restoration, of native forests. However, the authors emphasized that restoration of native forests in Hawaii through seed dispersal by introduced birds depends on the existence of native forests to provide a source of local plant seeds. In contrast, on Mauritius the introduced red-whiskered bulbul (Pycnonotus jocose, Pycnonotidae), which is an effective disperser of many fleshy-fruited species, frequently moves from invaded and degraded habitats into less-disturbed native forests, thus potentially acting as a mediator of continued plant invasion into these areas. This exotic dispersal agent has been a major factor in the continued reinvasion of exotic plants such as Clidemia hirta (Melastomataceae) and Ligustrum robustum (Oleaceae) into secondary and restored conservation management areas. The initial invasion of native forests by exotic plants on Mauritius may not have happened as rapidly without efficient avian seed dispersers such as the red-whiskered bulbul (Linnebjerg et al. 2009).

Invasive insects may also have an impact on vertebrate-plant dispersal mutualisms. On Christmas Island in the Indian Ocean, Davis et al. (2010) have suggested that the invasion and formation of high-density supercolonies by the yellow crazy ant, Anoplolepis gracilipes, may severely disrupt frugivory and seed dispersal by endemic birds. This invasive ant, whose high densities are sustained through a mutualism with introduced scale insects, rapidly reduces fruit handling times by endemic island birds and may therefore reduce seed dispersal by these frugivores. Although additional studies need to be done on the effects of this introduced insect, this study complements the other investigations discussed above by showing that any disruption of mutualistic partners in a community network may have cascading effects on other members of the network.

(p.468) Although investigations are limited on the effects of invasive species on mammal-plant interactions, it is expected that introduced plants as well as introduced vertebrates will significantly alter these mutualisms. At least one study has shown that the introduced brushtail possum (Trichosurus vulpecula; Phalangeridae) into New Zealand has contributed to the dispersal of seeds of 17% of the total species in a successional forest landscape (Dungan et al. 2002). As in the case of many introduced frugivorous birds, these possums have not only increased the spread of invasive weeds, but their seed dispersal behavior has also provided conservation benefits by accelerating forest regeneration in native vegetation. Because of the decrease in numbers of large-gaped native birds as fruit disperser over the last century, exotic possums may now be the only dispersal agent for large-seeded native species in many New Zealand habitats.

Pathogens and Diseases

The introduction of new parasites and pathogens to populations of vertebrates often accompanies human activities such as habitat degradation and the release of nonnative species. Primates living in forest fragments, for example, sometimes incur higher parasite loads because of contact with humans (or their feces) and their domestic pets (Arroyo-Rodríguez and Dias 2010; Gillespie and Chapman 2008; Mbora and McPeek 2009). Conversely, humans can also be infected with new pathogens as a result of increased contact with wildlife (e.g., Wolfe et al. 2005). But, except in the case of Hawaiian honeycreepers (Fringillidae, Drepanidinae), the impact that pathogens actually have on vertebrate-plant mutualisms is poorly known.

Hawaiian honeycreepers are a classic example of the adaptive radiation of a clade of birds on an isolated oceanic archipelago. In a period of about 4 Ma, these birds evolved into a substantial array of feeding niches, including nectarivory, occupied by about 20 genera and 50 species. Seventeen species and subspecies are now extinct, another 14 species are endangered, and only three species still have robust populations (Atkinson and LaPointe 2009; Spiegel et al. 2006). The nectar feeders of tribe Drepanidini included six genera and eight species (of which about five are now extinct) that evolved close ecological relationships with a number of native flowers, especially those in the Campanulaceae (Lobelioideae). Although causes of extinction or population loss include habitat degradation and predation by introduced mammals such as rats, dogs, and mongooses, avian malaria and birdpox, which were introduced into the archipelago in the late nineteenth or early (p.469) twentieth centuries with the release of exotic birds, are thought to have played a major role in these extinctions. Populations of birds living below an elevation of about 1,700 m have been especially hard-hit by malaria. As a result of these avian extinctions plus habitat degradation and herbivory by feral goats, about 30 species of lobeliads have gone extinct in the past 100 years (Cox and Elmqvist 2000; Smith et al. 1995). In some places, remaining plant population densities are too low to support viable populations of their avian pollinators—an example of an “extinction vortex.” Smith et al. (1995) reported that as a result of low lobelioid flower densities as well as the extinction of the 'o'o (Moho nobilis, Meliphagidae), which was behaviorally dominant to the i'iwi at ohia flowers, the nectar-feeding i'iwi (Vestiaria coc-cinea) has switched to feeding on flowers of ohia trees (Metrosideros poly-morpha; Myrtaceae) with significant evolutionary results. Whereas its bill was originally strongly curved for feeding in the curved, tubular flowers of lobelioids (Speith 1966), current populations of V. coccinea have slightly shorter, straighter bills, presumably as a result of selection for feeding on the corolla-less flowers of M. polymorpha.

This example has several general messages. First, it illustrates the concept of cascading ecological consequences (i.e., an extinction vortex) associated with the extinction of mutualist partners. The extinction of nectar-feeding drepanidids presumably led to reduced reproductive success of their (coevolved) food plants that led to their near or actual extinction and a significant change in the composition of the understories of Hawaiian forests. Second, it indicates how the extinction of one pollinator (the 'o'o) can create a new ecological opportunity for another pollinator (the i'iwi) that helped to prevent its extinction. Third, it emphasizes the importance of adaptability in the face of ecological change. If V. coccinea had not been able to successfully extract nectar from ohia flowers, it might have also gone extinct. Adaptive flexibility is a hallmark of this clade of birds (Lovette et al. 2002). Finally, this example shows us the precarious nature of strong ecological and evolutionary associations. Mutualisms such as pollination and frugivory are always vulnerable to disruption once one or more key partners are lost.

Bushmeat Hunting

As indicated above, bushmeat hunting is a major threat to large-bodied frugivorous birds and mammals. The extraction of protein from forests via hunting is widespread in the tropics and is often unsustainable. It is more intense in Africa and Asia than in South America because of higher human (p.470) population densities. In the forest belt of Central Africa, for example, Fa and Brown (2009) reported that about 4 million tons of dressed bushmeat are extracted annually and that 55% of all forest mammals are being hunted there. In contrast, Peres and Palacios (2007) reported that about 0.16 million tons of wild game are eaten annually in the Brazilian Amazon, where about 28% of all forest mammals are hunted. In regions where hunting has been long-standing, tropical forests are often devoid of large terrestrial herbivore/frugivores such as elephants, tragulids, cervids, and scatter-hoarding rodents as well as medium-to-large species of terrestrial and arboreal primates. Large fruit bats and birds such as guans, curassows, and hornbills are also hunted in many tropical areas (Kinnaird and O'Brien 2007; Mickleburgh et al. 2009; Trail 2007).

Since these birds and mammals are often the sole dispersers of large-seeded plants (i.e., seed length >15 mm), the question becomes, how badly has bushmeat hunting disrupted seed dispersal mutualisms? We addressed this question in some detail in chapter 4 and will only give a few additional examples here. Brodie et al. (2009) studied the effects of variation in the density of three species of mammalian frugivores (lar gibbons, muntjac, and sambar deer) on seed removal and seedling density of the canopy tree Choerospondias axillaris (Anacardiaceae) in four national parks in Thailand. Three of these parks were affected by hunting, but the fourth was effectively protected. They found that the proportion of undispersed seeds decreased and the seed density in potential germination sites, especially light gaps, increased with increasing density of the three frugivores across these parks. They used population matrix simulation models (see chap. 4) to determine the effect of reduced seed dispersal on future population growth of this tree. Their results indicated that reduced dispersal slightly reduced population growth rates, but compared with the effects of juvenile and adult survival rates, the elasticity of decreased seed dispersal (i.e., the relative effect of this factor on a population's λ. compared with other life history factors; chap. 4) had little effect on long-term population growth. As we discussed in chapter 4, owing to their low adult mortality rates, the extinction of tropical trees will occur long after their mammalian or avian dispersers have disappeared. The long time lag between loss of dispersal agents and the extinction of vertebrate-dispersed tree species creates an “extinction debt.” All plants whose key dispersers or pollinators have been extirpated will incur this debt and are living on borrowed time. They will repay this debt on their death.

Additional recent studies of the effects of hunting on rates of seed dispersal (p.471) in tropical trees include those of Holbrook and Loiselle (2009), Nunez-Iturri et al. (2008), and Vanthomme et al. (2010). Holbrook and Loiselle (2009) studied rates of visitation and seed removal by frugivores in the tree Virola flexuosa (Myristicaceae) at one hunted and one unhunted site in lowland Ecuador. At the site where monkeys and toucans, which are the most important dispersers of this tree, were hunted, the proportion of seeds removed and frugivore visitation rates were reduced significantly. The remaining frugivorous birds at the hunted site included barbets, cotingids, and thrushes—species that do not swallow large Virola seeds but instead drop them under the canopies of fruiting trees. Hunting clearly reduces effective dispersal in this system. Nuñez-Iturri et al. (2008) compared the species richness and density of seedlings and juveniles of trees dispersed by medium-to-large primates in transects away from fruiting trees at three hunted and three unhunted sites in lowland Peru. In forests in which primates have been depleted, species richness of seedlings of trees dispersed by monkeys was reduced by 46% and the frequency of seedlings of abiotically dispersed species increased by 284% compared with unhunted forests. These researchers concluded that hunting results in severe recruitment limitation that will eventually (on a time scale of decades to centuries) lead to reduced densities of primate-dispersed trees and less food for primates. Hunting thus appears to lead to an extinction vortex, in which both frugivores and their food plants maybe doomed to extinction. Finally, Vanthomme et al. (2010) censused diurnal frugivorous birds and mammals and compared seed sizes and the diversity and density of vertebrate-dispersed seedlings in transects around five species of primate-dispersed trees at one hunted and one unhunted forest site in the Central African Republic. As expected, animal densities were generally lower (including some extirpations), and mean seed lengths and diversity and density of seedlings of large-seeded species were lower at the hunted sites.

In summary, current evidence indicates that hunting has a significant negative effect on the seed dispersal services provided by frugivorous birds and mammals. As discussed in chapter 4, the long-term effects of hunting are less easy to predict but likely include decreased densities of vertebrates and their food plants. In the case of long-lived forest trees, these effects will occur over long periods of time so that the immediate botanical effects of extirpating frugivorous birds and mammals will not be obvious. Given this time lag (and its extinction debt), skeptics could say that it is too early to say that all is doomed for certain fruit-frugivore mutualisms because the (p.472) jury is still out. We disagree with this point of view and contend that we cannot wait until the extinction debt has been repaid to reach unassailable conclusions. By the time sufficient evidence has accumulated to show that the high levels of bushmeat hunting of today will have effects on forest or other habitat dynamics that last for centuries (e,g., Nuñez-Iturri et al. 2008), it will be far too late to save substantial numbers of vertebrate frugivores and their food plants. We know that bushmeat hunting is unsustainable in much of the tropics today (Corlett 2007a; Fa and Brown 2009). If it continues to be unregulated, there will be dire consequences for both plants and animals involved in the frugivory mutualism. Bushmeat hunting throughout the tropics must be regulated and/or its effects mitigated via alternative forms of protein.

Climate Change

During the last 2 decades of the twentieth century, concern with environmental deterioration and the loss of biodiversity soared as habitats were being degraded at enormously increased rates due to the activities of human populations. As the twenty-first century dawned, a more all-encompassing environmental threat emerged with the realization that steadily rising global temperatures, primarily as a result of increased levels of atmospheric C02 from the burning of fossil fuels and tropical forests, would have a profound effect not only on biodiversity, but on human civilization as well. The future effects of climate change on the functioning of ecosystems have now eclipsed all other environmental concerns so that biodiversity loss and the threats to ecosystem performance are predominantly a function of increasing global temperature. To assess any current threat to biodiversity, we must consider the vulnerability and sensitivity of particular species or set of interacting species to climate change, the level of their exposure to environmental change, and the adaptive capacity of populations to cope with these changes (Dawson et al. 2011; Kiers et al. 2010).

All of the other threats to vertebrate-plant mutualisms as described in the first part of this chapter are either exacerbated or exceeded by the enormity of climate change effects. Within this century, these effects have been predicted to include an increase in mean global air temperature of 2–3°C, a rise in mean sea level of at least 0.5 m (and probably much more), significant shifts in the locations of habitats, particularly at high elevations and high latitudes, and a substantial increase in the extinction rates of plants and animals (but see He and Hubbell [2011] for a cautionary view about this).

(p.473) Perhaps one of the most important things we have understood as global change has transformed our concern for biodiversity conservation is that historical change as reflected in the fossil record has much to tell us about the magnitude of these problems, the similarity of changes that the earth has gone through in the past, and what may happen in the future (e.g., Benton 2010; Erwin 2009; Ezard et al. 2011; Finarelli and Badgley 2010; Willis et al. 2010). It is quite difficult, if not impossible, to determine how global climate change will affect vertebrate-plant mutualisms discussed in this book (see Kiers et al. [2010] for an overview of this problem). We will therefore provide a brief analysis of past changes in vertebrate and plant interactions that may serve as a guide to and provide some insights into how climate change might affect these mutualisms in the future.

From a historical perspective, climate change has been the norm, not the exception, for most of the earth's history. This can be clearly seen in a plot of estimated global air temperatures throughout the Cenozoic Era (fig. 6.1). The earth was much warmer during the Paleocene-Eocene Thermal Maximum than at any time since, and global air temperatures have declined steadily since the Mid-Miocene Climatic Optimum. As we describe in chapter 6, Cenozoic temperature changes, along with associated changes in precipitation, have caused the distributions of humid tropical forests to expand and shrink, which, in turn, has had a significant effect on the diversification and distribution of tropical plants and animals.

We used data on the geological ages of families of nectar- and fruit-eating birds and mammals (table 1.1) plus the temperature data found in figure 6.1 to address the question of what effect, if any, climate change has had on the diversification rates of these animals during the Cenozoic. Although diversification rates are usually calculated as the difference between rates of speciation and extinction at the species level (e.g., Mittlebach et al. 2007; Ricklefs 2010), we will use this term here to mean time of family origin. Thus, our primary question can be restated as: What effect has climate change (i.e., change in global air temperature) had on timing of the origins of families of mutualistic birds and mammals and their food plants? Have these families evolved primarily under conditions of elevated or reduced temperatures relative to climatic conditions before or after their origins? Or have these origins occurred independently of climatic conditions? Note that climate change is only one of several possible abiotic factors that are known to affect rates of biological diversification (e.g., Benton 2010; Finarelli and Badgley 2010). Tectonic events, including plate movements, orogeny, and volcanism, (p.474) are also known to be important abiotic drivers, as we discussed in chapter 9. Thus, we would ultimately like to know the importance of climate change relative to that of various tectonic events as an abiotic driver of biological diversification.

To address our main question, we tallied the number of origins of families of nectar-or fruit-eating birds and mammals in 10 Ma time blocks throughout the Cenozoic. Results (fig. 10.4, birds and mammals combined) indicate that there are two peaks in the distribution of nectarivore origins (at 50–59 Ma and 30–39 Ma) and a single peak for frugivore origins (at 20–29 Ma). Next, we divided the Cenozoic Era temperature curve (fig. 6.1) into segments based on whether temperatures were higher, lower, or stable relative to temperature changes before and after particular time blocks (see col. 2 of table 10.3). We then compared the observed and expected number of origins of families of nectarivores and frugivores (birds and mammals

The Future of Vertebrate-Angiosperm Mutualisms

Figure 10.4. Estimate of global temperature during the Cenozoic (A) and distributions of time of origin of families of vertebrate nectarivores and frugivores (birds and mammals combined) in 10 Ma time blocks (B); based on data in table 1.1. Data for (A) come from Blois and Hadly (2009). Geological ages (at top): Pl = Pliocene; M = Miocene; O = Oligocene; E = Eocene; P = Paleocene; K = Cretaceous.


Table 10.3. The Times of Origin of Families of Avian and Mammalian Mutualists as a Function of Relative Temperature Changes during the Cenozoic Era

Number of Family Origins



Relative Temperature Condition

Time Periods* (Ma)



Total (Expected Origins)



Total (Expected Origins)


59–45; 26–15 (25)



5 (5.04)



30 (22.3)


45–34; 15–0 (26)



7 (5.28)



14 (23.3)


34–26 (8)



0 (1.68)



9 (7.42)

Note. See fig. 6.1 and text for further discussion.

(*) Total Ma given in parentheses following time periods.

combined) occurring in each of the three segments by chi-square (table 10.3). Results indicated that nectarivore origins were independent of direction of temperature change (x2 = 2.24, df = 2, P >> 0.05) but that more frugivore origins occurred during periods of elevated temperatures than expected and fewer origins occurred during periods of reduced temperature = 6.71, df = 2,0.05 > P > 0.025). From this simple analysis, we tentatively conclude that (1) the origins of families of nectarivores appear to be independent of Cenozoic temperature changes (although the small number of nectarivore families limits our ability to detect a significant trend) and (2) the origins of families of frugivores are not independent of Cenozoic temperature changes; these origins were more common during periods of elevated temperatures than expected. It is important to note that the peak in the origin of families of vertebrate frugivores does not coincide with highest temperatures in the Cenozoic (i.e., with the Paleocene-Eocene Thermal Maximum). Instead, this peak is associated with a period of increasing temperature in the Late Oligocene and Early Miocene, which suggests that relative temperatures were more important than absolute temperatures in the diversification of modern families of fruit-eating birds and mammals (as well as mammals in general; Stadler 2011).

Rates of biological diversification are driven by both abiotic and biotic factors. In addition to competition and predation, biotic factors for vertebrate mutualists include new food resources resulting from the evolution of new families of food plants. However, as we discussed in chapter 6, most of the major families of angiosperms that provide nectar or fruit for birds and mammals evolved much earlier (e.g., in the Late Cretaceous) than their modern mutualists. Thus, there is not a close temporal association between the origins of families of vertebrate-pollinated or -dispersed plant families and families of their modern vertebrate mutualists. This does not preclude (p.476) the possibility, however, that the origins of families of mutualistic birds and mammals involved coradiations with their food plants (chap. 5). That is, although origins of certain angiosperm families and their modern vertebrate mutualists were not coeval, perhaps major radiations of these plants did coincide with the origins of their mutualists. Some support for this idea comes from the evolutionary history of the cactus family, Cactaceae, and its phyllostomid bat pollinators. According to the phylogenetic analysis of Arakaki et al. (2011), this New World family first evolved in the Late Eocene but didn't undergo a major radiation of its crown lineages, including tribes of columnar cacti that are both pollinated and dispersed by bats, until the mid-Miocene in association with cooler and drier climatic conditions worldwide (fig. 10.5). The radiation of columnar cactus lineages also coincides with the first appearance of nectar-feeding phyllostomid bats (Datzman et al. 2010; fig. 10.5). If this cactus-bat example is general, then it supports the hypothesis that climate change can lead to an evolutionary response by plants, which in turn can favor an evolutionary response by vertebrate mutualists. We clearly need more detailed and dated plant and animal phylogenies in order to

The Future of Vertebrate-Angiosperm Mutualisms

Figure 10.5. Evolutionary history of the Cactaceae and other families of arid-adapted plants with the time of origin of flower-visiting phyllostomid bats indicated by the red star. Based on Arakaki et al. (2011), with bat data from Datzmann et al. (2010).

(p.477) determine the generality of this hypothesis. Data presented by Givnish et al. (unpubl. ms.) suggest that such a scenario also holds for the diversification of epiphytic bromeliads and their hummingbird pollinators during the mid-Miocene Andean uplift. Even if this hypothesis is generally true, however, it may not be applicable to the current period of climate change because of the speed with which this change is occurring. Coradiations likely occur on a time scale of thousands, if not millions, of years and far too slowly to respond quickly to changes that we are currently witnessing.

In conclusion, there can be no doubt that climate change has played an important role (in both a positive and negative sense) in the diversification of the earth's biota, and the evolution of vertebrate mutualists is no exception to this. The appearance of many modern families of vertebrate frugivores appears to have occurred during periods of elevated temperature in the latter half of the Cenozoic, perhaps in response to climate-induced expansions of tropical forests. Once they evolved, these vertebrates could have facilitated the radiation of their food plants via their seed dispersal behavior, including long-distance dispersal (e.g., Muellner et al. 2008; Viseshakul et al. 2011), but much more research is needed before these links are firmly established.

Whether the current global warming trend will have a net positive effect on frugivores and their food plants is difficult to say. This is because climate change is only one of many anthropogenically mediated global changes that are affecting biodiversity today. We suspect that evolutionary trends that held in the past have become void with the rise of modern Homo sapiens. As the tropical ecologist Ariel Lugo has said (in plenary addresses to the Association for Tropical Biology and Conservation in 2003 and 2006), we have now entered a new evolutionary era—the Anthropocene (also see Economist 2011). In this era, the past is not necessarily a mirror to the future of life on Earth. Like it or not, humans have become a major factor in the evolution of life on our planet.

Conservation of Vertebrate Pollination and Dispersal Mutualisms

What do vertebrate mutualists and their food plants need to avoid (premature) extinction as a result of human activities? Intact and extensive habitats is the easy answer. In the case of migratory mutualists, “extensive” can mean an extremely large area that often encompasses multiple political (p.478) units. Many nectar- or fruit-eating birds fly thousands of kilometers annually between their wintering and breeding grounds, and a few phyllostomid and pteropodid bats fly nearly this far (Epstein et al. 2009; Fleming and Eby 2003; Greenberg and Marra 2005). However, most of the land encompassing migratory routes currently lacks conservation protection (e.g., Epstein et al. 2009; Fleming 2004b; Richter and Cumming 2008). Without serious conservation efforts, this easily identifiable goal will become more and more difficult to achieve with each passing year. In a world of increasingly fragmented habitats, shrinking wildlands, increasing temperatures, and ever-increasing human pressure, many of our ornaments of life will disappear within a few human generations unless we are proactive about their conservation. Given that these mutualisms occur in networks of interacting species embedded in larger sets of species in communities or habitats, their conservation clearly demands habitat or ecosystem conservation rather than simply species-based conservation. And it almost always involves in situ, rather than ex situ, conservation.

It is generally agreed that effective measures for conserving individual species or habitats and ecosystems involve a multi-pronged approach that includes environmental education, legislation, enforcement, and the involvement of local communities. Many international and local conservation agencies are involved in each of these activities, and we will not review this general topic here. Instead, we will focus on two topics that have special relevance to the conservation of nectar- and fruit-eating birds and mammals: pollinator and disperser protection campaigns and ecotourism.

Pollinator and Disperser Protection Campaigns

As a result of the United Nations Convention on Biological Diversity starting in 1992, a number of national, regional, and international efforts have been launched to increase awareness of the threats to pollinators and to protect them from further decline. The International Pollinators Initiative was established in 2000 at the fifth conference of the parties to the Convention on Biological Diversity as the International Initiative for the Conservation and Sustainable Use of Pollinators. This initiative to address the worldwide decline in pollinator diversity was followed by a plan of action, which was prepared by the Food and Agriculture Organization of the United Nations and the Convention on Biological Diversity secretariat and adopted in 2002 at sixth conference of the parties. The aim of the International Pollinators (p.479) Initiative is to promote coordinated worldwide action to monitor pollinator decline and identify its causes and its impact on pollination services; address the lack of taxonomic information on pollinators; assess the economic value of pollination and the economic impact of the decline of pollination services; and promote the conservation, restoration, and sustainable use of pollinator diversity in agriculture and other ecosystems. So far a number of reports have been prepared by the International Pollinators Initiative to further these aims (see www.internationalpollinatorsinitiative.org).

In addition to this international effort by the United Nations, the North American Pollinator Protection Campaign was established in 1999 as a private-public collaborative body to promote and protect pollinators in North America. The campaign is coordinated by the Pollinator Partnership, a 501(c)3 nonprofit organization that works with partners worldwide to further their goals. At present the campaign is made-up of more than 120 organizations that include scientists, researchers, conservationists, government officials, and volunteers who have organized major programs to protect pollinators, to raise awareness of pollinator-related issues by the public and governmental bodies, and to benefit the health of all species of pollinators, especially those most threatened by all of the factors described in the present chapter.

The North American Pollinator Protection Campaign's primary mission is “to encourage the health of resident and migratory pollinating animals in North America” by raising public awareness about the importance of pollinators for agriculture, ecosystem health, and food supplies; encouraging collaborative, working partnerships with federal, state, and local government entities and strengthening the network of associated organizations working on behalf of pollinators; promoting the conservation, protection and restoration of pollinator habitats; and documenting scientific, economic, and policy research on pollinators as well as developing and maintaining an international database of pollinator information. Similar efforts have now been instituted around the world, including the African Pollinator Initiative, the Brazilian Pollinator Initiative, the Colombian Pollinator Initiative, the European Pollinator Initiative, the French Pollinator Initiative, and the Oceania Pollinator Initiative.

Despite the diversity and breadth of these efforts to protect pollinators and their natural habitats, it is difficult to assess how effective these initiatives have been in preventing or at least slowing the decline of populations of pollinators. In addition, most of these efforts are aimed at insect pollinators, (p.480) especially bees, because of their importance in agricultural systems. In fact, very little activity has been aimed at halting or slowing the loss of vertebrate pollinators. We know of no similar international programs that address the decline of vertebrate dispersal agents, though many agencies exist for the general protection of birds (e.g., US National Audubon Society, Royal Society for the Protection of Birds, etc.) and primates (e.g., Conservation International, Wildlife Conservation Society, World Wildlife Fund, etc.). Clearly, there is much room for further efforts specifically aimed at conserving the world's vertebrate pollinators and seed dispersers.

One agency that has taken considerable interest in the conservation status of vertebrate pollinators and seed dispersers is Bat Conservation International (www.batcon.org). Founded in 1982 by Dr. Merlin Tuttle and based in Austin, Texas, Bat Conservation International has a membership of about 14,000 people worldwide and has initiated or encouraged education and conservation efforts in many tropical countries around the world. Specifically, it has supported student research projects, many of which deal with phyllostomid and pteropodid bats, in 55 countries; its Global Grassroots Bat Conservation Fund supports local conservationists and researchers in countries such as Ghana, Kenya, Myanmar, and Vietnam; and its Wings across the Americas program focuses on conservation of two species of nectar-feeding bats in Mexico and the southwestern United States as well as conducting workshops on bat conservation throughout Latin America. In addition, it currently supports a special program on bats of the Philippines, which focuses on conservation, education, and research (Locke 2006; Waldien 2008). One of the major goals of this initiative is to set national priorities and to begin proactive conservation of this bat fauna, including at least 24 species of plant-visiting pteropodids (Heaney and Roberts 2009; D. Waldien, pers. comm.).

Ecotourism—Biodiversity Savior or Threat?

Ecotourism, or nature tourism as coined by Corlett (2009a), has become big business around the world in recent decades. For example, when one of us (Fleming) began working in Costa Rica in 1970, that country's main sources of income were coffee, bananas, and cattle. Within a decade or so, however, ecotourism had become one of its main income sources (currently second only to silicon chips), and this country continues to be a major destination for ecotourists, mostly from the United States. While staying in small-scale ecolodges in Costa Rica, guests can explore tropical forests (p.481) where, depending on location, they can view manakins, macaws, toucans and araçaris, tanagers, quetzals and other trogons, three-wattled bellbirds and other cotingids, many species of hummingbirds, and several species of primates—myriads of the ornaments of life. Ecolodges are especially popular with bird watchers, and many well-run lodges cater specifically to this segment of the tourist industry. The world's avian ornaments surely are an important draw for these travelers.

In theory, ecotourism should be a win-win situation for a country: it offers visitors intact habitats full of flora and fauna while permitting countries to cash in on their natural resources without destroying them. Countries that attract ecotourists can parlay resource conservation into an important source of income. When it involves locally owned and operated small-scale lodges and local transportation systems, this income remains in the local economy. From a conservation standpoint, therefore, ecotourism can be a far better use of land and its natural resources than alternatives such as logging, farming, ranching, and mining.

In practice, however, ecotourism has its costs as well as its benefits. These costs include overuse of and damage to habitats caused by too many visitors; inadequate protection or misuse of the environment owing to too few park guards, poorly trained guides, and so on; an emphasis on profit over environmental protection by some tour companies; and heavy reliance on foreign capital and donors who do not necessarily respect a country's environmental ethics or laws (Dasenbrock 2002; Krüger 2005). As a final concern, ecotourism may not be a sustainable source of income in the long term because it relies significantly on a strong world economy (to generate ecotourists) and stable governments and safe conditions in destination countries (to attract ecotourists).

Ecotourism, of course, is only one of many options for conserving tropical biodiversity (Corlett 2009a; Ghazoul and Shell 2010). As Krüger (2005) indicates in his meta-analysis of the sustainability and conservation value of ecotourism, ecotourism is a major revenue producer in only a few countries (e.g., Costa Rica and Kenya), and of the world's 25 biodiversity hotspots identified by Myers et al. (2000), only six countries encompassing these localities are among the top 10 countries in terms of frequency of ecotourism case studies. Clearly, many areas of major conservation concern are not attracting ecotourists. Based on Krüger's (2005) analysis, major factors predicting whether countries are likely to have thriving ecotourism industries include proximity to major markets and the presence of species of charismatic (p.482) birds and mammals (many of which are ornaments of life) or internationally recognized charismatic vertebrates (e.g., gorillas in Rwanda).

The Future of Vertebrate Pollination and Dispersal Mutualisms

We live in a world that continues to teem with many ornaments of life, especially in tropical ecosystems. One can take any number of well-run ecotours or walks through pristine as well as secondary forests to see beautiful nectar-or fruit-eating birds and mammals living with their plant hosts in apparently healthy habitats throughout the tropics (but which are decidedly depauperate in terms of large seed-dispersing birds and mammals; see Corlett 2007a). But what does the future hold for these systems? How many more generations of humans will be able to enjoy and appreciate these ecological interactions and these natural history phenomena? We suspect not many.

Our pessimism stems from two lifetimes of research by the authors in tropical ecosystems around the world. We have watched as forests have been butchered and reduced to low-productivity agricultural systems or, even worse, ecological deserts devoid of the life forms we expect to see in wet, ever-warm habitats. These observations are indeed discouraging and we do not need to dwell further on the threats to the ornaments of life that we have elaborately described in this volume. A paleontological colleague has stressed to us that if we are concerned with the future of the earth and the lineages of plants and animals that have evolved here, then we need not worry. The historical record testifies that our planet and its life forms will exist long after this current environmental onslaught has abated. However, he also has warned us that if we are worried about the future of our own species and the species we now find on Earth, then we should be extremely concerned (W. DiMichele, pers. comm.). And we are. As Davies et al. (2008), 11562) state in their analysis of extinction risk in mammals: “Diversity will almost certainly rebound after the current extinction event; however, it may be composed of species descended from a different … subset of lineages from those that dominate now, and humans will likely not be included among them.” If nothing else, we hope that this book on the interactions between vertebrates and the plants they pollinate and disperse will help to inform our scientists, our citizens, and our policy makers that we have a responsibility to make sure that the ornaments of life will continue to prosper (p.483) and flourish long into the future both for their own value and for the benefit and enjoyment of all.


In this book we have covered a wide range of topics dealing with the past, present, and future of interactions between nectar-feeding and fruit-eating birds and mammals and their food plants. In the 30–40 plus years that we have been studying these systems, we have seen tremendous advances in our understanding of many facets of these interactions. In the late 1960s and early 1970s, pollination and seed dispersal mutualisms were still being studied as natural history subjects without strong theoretical, evolutionary, or experimental underpinnings. One prominent theoretical ecologist was even questioning the importance of mutualisms in the dynamics of communities (May 1973). Most of our knowledge about tropical vertebrate mutualisms came from a few sites in Latin America, especially in Panama and Costa Rica.

As this book attests, all of this has changed in the last few decades. We now know (though this wasn't really news to field naturalists) that mutualisms play an extremely important role in the history of life; we have solid theoretical underpinnings for the study of pollination and dispersal mutualisms, which are now being studied experimentally everywhere; the geographic scope of mutualism studies now encompasses all tropical regions; and, owing to the surge of phylogenetic studies with the advent of molecular genetics and sophisticated computer-based analytical techniques, we can now study these mutualisms in an increasingly robust evolutionary framework. Although our basic understanding of the ecology and evolution of vertebrate-plant mutualisms has increased many-fold in recent decades, there is still much to be learned about them at all levels of biological organization, from genes through biogeographic realms, as we have repeatedly emphasized throughout this book. What we don't know about these mutualisms will keep “curious naturalists” busy for the foreseeable future. But, as we all know, we are in a race against time in this task because the habitats and organisms that we love to study are steadily disappearing as humans proclaim their dominion over land and sea in ever-increasing fashion. So the ultimate question concerning the fate of these mutualisms becomes: What will it take for our species to realize that biodiversity conservation in (p.484) general, and conservation of vertebrate-plant mutualisms specifically, is one of the great environmental issues of our time? In writing about conserving biodiversity in tropical East Asia, Richard Corlett (Corlett 2009a, 209) said: “There is a small but growing minority of well-educated, mostly urban, ecologists and conservationists, both amateur and professional [in this region]. In between these groups, however, there is a high degree of ignorance about environmental problems in general and biodiversity conservation in particular, encompassing almost everyone from rural smallholders to urban civil servants and politicians. The extent to which this ignorance contributes to behavior that has harmful impacts on biodiversity is hard to judge, but combating ignorance is a crucial first step towards building the broad public support needed to sustain biodiversity conservation in the region” The same can be said for much of the world. It is our hope that this book will make an important contribution to combating ignorance and misunderstanding about the importance of a relatively small but nonetheless critical set of ecological and evolutionary interactions between plant-visiting vertebrates and their food plants. We cant imagine living in a world devoid of these ornaments of life.