The Bats of Puerto Rico
M.R. Gannon, M. Rogriguez-Duran, A. Kurta, and M.R Willig
The composition of the fauna of Puerto Rico is a dynamic consequence of three biogeographic and evolutionary processesimmigration or colonization, local extinction, and in situ speciation. As a result of these processes, the number of species (species richness) of mammal on Puerto Rico, like that of most islands, is considerably less than the species richness of comparable areas on the mainland at similar latitudes (Willig and Gannon, 1996). Indeed, the mammalian fauna of Puerto Rico comprises only 13 living, native species, all of which are bats. This is far less than the richness of mainland sites of similar size and habitat complexity (mean terrestrial richness of mammals = 83).
Even if one focuses exclusively on bats, Puerto Rico supports far fewer species than do mainland sites (mean richness of bats = 42). The living bats of Puerto Rico include one species of fishing bat (Noctilionidae), three mustached or ghost-faced bats (Mormoopidae), five leaf-nosed bats (Phyllostomidae), two plain-nosed bats (Vespertilionidae), and two free-tailed bats (MolossidaeTable 3). The island does not harbor any species from four New World familiesthe sac-winged bats (Emballonuridae), funnel-eared bats (Natalidae), smoky bats (Furipteridae), and disk-winged bats (Thyropteridae). Nonetheless, these families contain few species and make only small contributions to the richness of mainland sites. The most notable cause of low species richness of bats on Puerto Rico pertains to the leaf-nosed bats. Even though Puerto Rico harbors five species of phyllostomid, that richness is less than one fifth the local richness on the mainland (mean richness of phyllostomids = 27).
Although nonflying mammals are not part of the native fauna today, a number of such animals have been introduced to the main island and its surrounding cays. Native Americans, for example, brought dogs and perhaps guinea pigs to Puerto Rico (Wing, 1989). In addition, Taíno from Hispaniola apparently transported a hutia, about 1-kilogram in size, to Puerto Rico as a source of food in pre-Columbian times; remains of this animal are known only known from kitchen middens, with no reliable evidence of the animal ever having lived freely on Puerto Rico (Miller, 1918; Olson, 1982).
European settlers also imported intentionally a number of domesticated mammalsdogs, cats, goats, pigs, cattle, horses, and donkeys (Lawrence, 1977; Miller, 1929; Wing, 1989; Wing et al., 1968). Some introduced species exist in a feral state, even to this day, either because escaped animals have adapted to local conditions and populations are self-perpetuating or because domesticated individuals that stray or are abandoned continually enter feral groups, thereby rescuing marginally fit populations from extinction. The potential ecological impact of introduced animals is well known and quite detrimental (e.g., overgrazing by goatsBreckon 2000), especially on smaller islands and cays. However, in other situations, such as heavily forested sites on Puerto Rico, feral cats and dogs have an unknown, but possibly appreciable, impact on indigenous animals, i.e., organisms that did not evolve in the presence of any member of the cat or dog families (Garcia et al., 2001; Willig and Gannon, 1996).
Unlike domesticated species, some mammals were introduced inadvertently by humans. The black rat likely arrived in Puerto Rico as a stowaway with Ponce de León in 1508 (Snyder et al., 1987), and it ultimately was joined on the island by the house mouse and Norway rat. These mammals flourished wherever food, water, and shelter were available, and today they are found throughout Puerto Ricoin urban, rural, and natural areas. Worldwide, these commensal rodents are responsible for large economic losses, due to destruction and contamination of human food, and are a potential health hazard (Jackson, 1982). In addition, introduced rodents, particularly rats, may significantly alter community composition and structure of food webs in natural settings, because of their large body size and potentially high densities (e.g., the Luquillo Experimental ForestWillig and Gannon, 1996).
Although domesticated species occasionally escape captivity and rats and mice hitchhike their way to new lands, the Indian mongoose actually was released intentionally by humans into the wild on Puerto Rico, in 1877, to prey on rats living in sugarcane plantations (Horst et al., 2001). Mongooses, however, are not capable of regulating rat populations, and once established, these carnivores generally alter the ecological structure of natural habitats that they invade. For example, six species of vertebrate are considered endangered on Puerto Rico, partly because of interactions with rats and mongooses (Raffaele et al., 1973). Negatively affected species include the Puerto Rican boa, Puerto Rican vine snake, Puerto Rican parrot, Puerto Rican short-eared owl, and possibly the Key West quail-dove and Puerto Rican whip-poor-will.
Although the only native mammals living on Puerto Rico today are bats, there is fossil evidence for the past presence of a number of other species, including bats and nonflying mammals. These extinct animals include one shrew, one sloth, three leaf-nosed bats, and five rodents (Table 5). Cause of each extinction remains poorly understood, but as on other islands of the Antilles, disappearance of a few species may be related to climatic changes at the end of the Pleistocene Epoch (Pregill and Olson, 1981). Many extinctions, however, apparently were induced by humans, caused either by alteration of habitats and exploitation by natives and Europeans or by interactions with rats, cats, dogs, and mongooses (Morgan and Woods 1986; Woods and Eisenberg, 1989).
Beyond the well-documented pattern that number of species on islands are reduced compared with equivalent areas on the mainland, one of the most pervasive biogeographic patterns is a reduction in species richness with reduction in area of an island. Indeed, the equilibrial theory of island biogeography (MacArthur and Wilson, 1967) formalizes the view that number of species on an island represents a balance between rates of loss (primarily extinction) and rates of gain (primarily colonization and in situ speciation). Size of an island affects extinction rates, because area should be proportional to the number of animals that the island can support and to the number of habitats that it contains. All other things being equal, small islands should have small populations, and species with small populations should have a greater likelihood of going extinct, thereby diminishing the richness of a particular island.
Colonists must come from somewhere, and the distance from such source pools (e.g., the mainland) should be inversely proportional to the rate at which individuals, and hence new species, colonize an island. Therefore, all other things being equal, islands that are close to source pools should have more species than do islands that are distant from source pools. In addition, islands close to source pools should have fewer endemic taxa, which usually result from in situ speciation, because immigrants from adjacent populations constantly are arriving, thus preventing the genetic isolation needed for speciation to occur.
The Caribbean region contains a plethora of islands of various sizes and distances from the mainland, providing a rich source of variation in the primary factors thought to mold species richness. At the same time, different islands have different geological histories, elevational profiles, primary source pools (e.g., North America versus Central America or South America), secondary source pools (i.e., nearby islands that might provide colonists), and exposure to anthropogenic and natural disturbances. Such uncontrolled differences among islands lead to considerable complexity, sometimes confounding the ability to detect patterns related to size or distance. Nonetheless, at broad levels of resolution, the way in which bats, birds, amphibians, and reptiles differ in richness among the islands of the Greater Antilles is generally consistent (Fig. 10). Specifically, bat and amphibian faunas are always more depauperate than avian and reptilian faunas, with species richness of bats being the lowest of the four groups on all islands.
The number of species of bat on islands of the Antilles is reasonably consistent with predictions based on size of the island (Griffiths and Klingener, 1988). Indeed, 87% of the variation in bat species richness in the Antilles is explained by variation in island area alone (Fig. 11). The richness of Puerto Rico, however, is slightly below the predicted value for an island of its size in the Antilles, as is that of Hispaniola, whereas richness for Cuba and Jamaica are above predicted values. These differences among islands likely are related to lower sea levels during the Pleistocene Epoch. As a result, Florida and the Grand Bahama Bank were larger, providing North American bats with easier access to Cuba. Similarly, the Nicaraguan Plateau and the Yucatan Peninsula were significantly larger, providing shorter, over-water routes for Central American species colonizing Jamaica, and to a lesser extent, Cuba. Puerto Rico and Hispaniola, in contrast, were relatively isolated from both sources of colonists, even during the Pleistocene Epoch.
Bats probably colonized Puerto Rico via stepping-stone dispersali.e., movement first from the mainland to a near-shore island, then to a more-distant island, and so on (Griffiths and Klingener, 1988). The ancestors of most species initially left Central or North America and occupied the westernmost of the Greater Antilles, with some eventually flying further east to Hispaniola and Puerto Rico. Although colonization of Puerto Rico could have been accomplished by animals traveling from South America, northward through the Lesser Antilles, biogeographers generally consider this route less likely for two reasons (Koopman, 1989). First, for those species of bat that currently live both on the mainland and on the Antilles, individuals from the Greater Antilles generally are more similar morphologically to those of Central or North America than they are to their relatives in South America. Second, in those genera that are endemic to the West Indies, there is higher species diversity in the Greater than in the Lesser Antilles, suggesting a longer period of residence in the Greater Antilles.
Community Ecology within the Caribbean Basin
Whatever the initial origin, exactly which species successfully colonized and ultimately survived or evolved on each island in the Caribbean was not due totally to chance, because the resulting communities of bats are not random samples of the species available from mainland source pools (Rodríguez-Durán and Kunz, 2001). Instead, biogeographers have identified a core community of bats, the members of which are similar on most islands of the West Indies. This core community consists of one species of Antillean long-tongued bat (genus Monophyllus), one species of Antillean fruit bat (Brachyphylla), the Jamaican fruit bat, greater bulldog bat, Brazilian free-tailed bat, and velvety free-tailed bat. On the larger islands that are part of the Greater Antilles, the core community expands to include three species of ghost-faced or mustached bat (Mormoops and Pteronotus, respectively) and one species of flower bat (Erophylla). Puerto Rico contributes to this general pattern, with 10 of the 13 species of bat on the island representing members of the core community (Table 3).
Why did this particular combination of species evolve on many different islands in the West Indies? One factor that probably played a role in determining the makeup of these island communities was partitioning of food resources (Rodríguez-Durán and Kunz, 2001). Although a random sample of species from the mainland theoretically could include only frugivores or only insectivores, for example, the core community that exists on Caribbean islands typically contains species that specialize on different foods, including nectar/pollen (long-tongued and flower bats), fruit (Antillean and Jamaican fruit bats), fish (greater bulldog bat), and insects (Brazilian free-tailed, velvety free-tailed, ghost-faced, and mustached bats).
Another key factor that helped shape community composition on West Indian islands was type of roosting site utilized by each species (Rodríguez-Durán and Kunz, 2001). Except for the velvety free-tailed bat, all members of the core community roost either predominantly or exclusively in caves. In addition, over 40% of the 56 living species of bat that occur in the West Indies (80% of species on Puerto Rico) use caves. These underground roosts provide excellent protection from wind and rain, especially during tropical storms, such as hurricanes. Over 100 hurricanes have ravaged at least some part of the region over the last 500 years (Colón, 1977), and species that roost in more exposed situations, such as tree hollows or foliage, likely suffer greater direct mortality from these common storms than do cave-dwelling species (e.g., Gannon and Willig, 1994).
In addition, islands usually have fewer food resources than do comparable sites on the mainland (e.g., Janzen, 1973). Consequently, reptiles, birds, and mammals that live on islands often are characterized by having lower rates of energy expenditure (mass-specific basal metabolic rates) than do their mainland relatives. These low metabolic rates presumably are pre-adaptations that favored some groups that colonized islands over others or these low rates of energy expenditure are adaptations evolved by members of endemic taxa in response to the low level of resources available on islands (Faaborg, 1977; Kurta and Ferkin, 1991; McNab, 1994, 2001).
Many cave-dwelling bats of the West Indies, especially endemic species (e.g., flower bats, ghost faced bats, Antillean long-tongued bats), prefer to roost in hot caves, where temperatures often exceed 30°C (Silva-Taboada, 1979). Because these species spend most of their lives in very warm environments, heat loss is minimal, and they typically posses a reduced rate of heat production, i.e., a low basal metabolic rate (Bonaccorso et al., 1992; Rodríguez-Durán, 1995). This physiological trait presumably allows such species to attain higher population densities on islands, making extinction less likely, despite the low availability of resources. Similarly, insectivorous species of bat have lower mass-specific basal rates of metabolism than bats having other dietary preferences (McNab, 1982), and this characteristic partly may explain why insectivorous species are more common in the core community of bats on West Indian islands than species that consume primarily fruit, nectar, or fish.