Volume 6
Issue 2

To avoid hungry bats, katydids in some parts of the world have had to adopt new tunes to survive—

Insects are an important source of food for bats, providing them with a diet rich in protein, fats, and other nutrients. Historically, insects have also played a significant role in the evolution of bats and their behavior, ranging from the development of teeth designed to crush hard insect bodies, to timing migration or the birth of young to coincide with seasonal insect `blooms.'

Similarly, bats have influenced the development of behavioral and physical adaptations in many insect species, all designed to avoid being eaten by bats. The specialized ears of many species of moths, green lacewings, crickets, and mantids 1-4 are a good example. With an acute sensitivity to the high-pitched echolocation calls of hunting bats, these insects can veer away from approaching bats and thus lessen their chances of being eaten. In the presence of bat ultrasound, some moths take their defenses one step further, and `talk back to bats' with their own ultrasound to advertise the fact that they are poisonous and should not be eaten 5.

Until recently, anti-bat adaptations were unknown in katydids, another major group of insects. Large, slow-moving, grasshopper-like insects, katydids are closely related to crickets and roaches. In many parts of the world, they are a dominant component of the insect fauna and are well known for their spectacular abilities to mimic leaves or blend into their surroundings, protecting them against predators that rely on vision.

Most species of katydids are active at night and, in the absence of light, they rely on calling to attract a mate. Male katydids, like frogs and songbirds, attract females by singing `songs' that are unique to their species. In katydids, these songs are produced with specialized structures on the main or first set of wings. Singing males advertise not only who and where they are to nearby females, but also inadvertently broadcast this information to all other animals within hearing range. Since katydids are plentiful, they are an important source of food for a variety of nocturnal and diurnal animals, including a small and not very well known group of bats called foliage gleaners (literally, ‘leaf cleaners').

Bat foraging styles
Although they may not know them by name, most people are familiar with the kind of bats known as `aerial insectivores'— the common little bats that can be seen on warm summer evenings almost anywhere in the world chasing small insects over lakes and around street lights. For the most part, aerial insectivores are worldwide in distribution, are swift and agile fliers, have small ears, and use echolocation to locate and track mosquito-to-June-bug-sized insects that are eaten in flight.

Much less familiar are the foliage gleaners. Most of them occur only in tropical areas of South America, Central America, Africa, and Australasia. Usually found in dense forest, they spend a considerable amount of their time perched on branches or vines on the lookout for large insects, such as katydids, moths, roaches and beetles. Instead of capturing their prey on the wing, gleaners feed on larger stationary insects that are plucked from the ground or foliage.

Foliage gleaning bats and aerial insectivores do not have much in common 6. In addition to how they hunt, they differ in their flying styles and in the size of their feeding territories. Aerial insectivores forage swiftly over fairly long distances, a feat made possible by their relatively long and narrow wings. In contrast, foliage gleaning bats have broad wings, giving them a highly maneuverable flying style that allows them to hover as they pursue prey, but that restricts their ability to fly swiftly over long distances or for prolonged periods of time. As a result, foliage gleaners are loyal to a particular feeding site, foraging in the same small areas and usually from the same feeding perches, night after night 7. Because the insect prey of these bats are large (in some cases almost as large as the bats themselves), a catch cannot be consumed in flight like the smaller prey of the aerial insectivores. Instead, an insect is brought back to a secluded feeding roost where it is slowly eaten after the wings, legs, and other inedible parts are carefully culled. Because only a few feeding roosts are used, parts from discarded prey accumulate below the roost in large numbers.

Compared to other insect-eating bats, most foliage gleaners also have unusual echolocation calls. Their calls are shorter, fainter, and broader in spectrum than those of the aerial insectivores, and it appears that their echolocation does not allow them to distinguish prey from the surroundings 8. Perhaps as a result of this, these bats also have developed very long and highly mobile ears. Like `hearing horns,' such ears facilitate the location of insects by their calls, rather than by echolocation.

Mating calls
It has long been suspected that foliage gleaning bats use the mating calls of their prey to locate them for food, and this has recently been confirmed in the New World tropics. Perhaps the most dramatic example is the Frog-eating bat (Trachops cirrhosus), a carnivorous foliage gleaner that uses the mating calls of small frogs to locate them9. The Round-eared bat (Tonatia silvicola) has also been shown to use mating calls of katydids as feeding cues. Round-eared bats and Frog-eating bats are very closely related and share habitat in the lowland forests of Panama. Studies of these two species demonstrated the importance of mating calls in locating prey and also showed that different species of foliage gleaners can avoid competition for limited food resources by responding to the calls of different prey species 10. Not surprisingly, as an anti-predator defense, frogs that occur in the same areas as Frog-eating bats were discovered to produce calls that made it difficult for the bats to locate the source. Would this adaptation prove to be true also of katydids?

The study begins
With these discoveries as background, I initiated a two-year study at the Smithsonian Tropical Research Institute, on Barro Colorado Island in Panama to examine the interactions between insect-eating foliage gleaning bats and the katydids on which they feed 11. The Institute's facilities include large screened flight cages— ideal structures in which to study the behavior of bats and their prey. In addition, the tropical moist forest of Barro Colorado Island supports a diverse katydid fauna of over 70 species and is home to at least 12 species of foliage gleaning bats. The bats all belong to the family, Phyllostomidae, a large group known as New World spear-nosed or American leaf-nosed bats. The family includes about 140 species, many of which feed on fruit, pollen, or nectar12. The gleaners are in their own sub-family, the Phyllostominae, and are characterized by very long ears and a diverse diet. About half of them are strictly insectivorous, while the rest also include small vertebrates or fruit in their diets.

Learning what gleaners eat
To determine the specific diet of insect-eating foliage gleaners, their feeding roosts had to be located. They usually can be found in hollow trees, under large trunks that have fallen to the forest floor forming small protected alcoves, or even in abandoned buildings. A roost is identified by the presence of bats (up to about a dozen) or by an accumulation of insect remains, provided that scavenging ants haven't removed them.

During this study, no such roosts were found for Round-eared bats. However, 27 Little big-eared bat (Micronycteris megalotis) roosts containing the remains of over 12,000 insects, and 11 roosts of Hairy big-eared bats (M. hirsuta) with over 10,000 insect remains, were located. The insects were painstakingly reconstructed to determine exactly what the bats had eaten. About 40% of the diet of Hairy big-eared bats consisted of katydids belonging to only five of the 70 or so species present on Barro Colorado. Why these particular species were taken, and not others, is still unknown. In contrast, only 6% of the diet of the Little big-eared bat consisted of katydids. Instead, these bats included a large proportion of dragonflies in their diets, which is unusual since these insects are active strictly during the day. At night dragonflies hang quietly from vegetation, usually from the undersides of leaves. How the bats find them is not known.

With the exception of katydids and dragonflies, the remainder of the diets of these species was remarkably similar. Both fed on roaches and scarab beetles that were identical except for size. The prey of the larger species, the Hairy big-eared bat, were about twice the size of that taken by the smaller Little big-eared bat. How the bats distinguish size in their respective non-calling insect prey, apparently without the help of echolocation, remains to be determined. ¥When katydids sing£

In the field, male katydids, serving as bait to attract bats, were placed in small screen cages near mist nets set up to harmlessly capture the bats. The katydids usually began to sing 10 or 15 minutes after being placed next to a net, which was placed in a densely forested area known to contain foliage gleaners. Response to the insect calls was often dramatic, sometimes within a matter of minutes. On some evenings, however, no bats were caught–presumably because the nets were not placed within a bat's feeding territory.

Four species of foliage gleaners responded to the calls of the singing katydids–the small Little big-eared bat, the medium-sized Hairy big-eared bat, and the larger Round-eared bat.á One Frog-eating bat was also attracted to the calls. Hungry bats responded by attempting to land on the screen cages containing the calling insects. Other cages baited with female katydids, which do not sing, did not attract bats, demonstrating that the bats were attracted to insect song, not merely to large insects.

Changing their tune

There was no doubt that the mating calls of katydids attracted bats hunting for a meal, but the next step was to learn the influence this had on the calling behavior of the katydids. I recorded and compared the mating calls of katydid species occurring in habitats with and without foliage gleaners— either in densely forested areas, which provide foliage in which bats can perch and roost, or clearings and sparsely forested areas, where gleaning bats are less likely to be found.

Male katydids singing from clearings sound like miniature chain saws. Their calls are loud and raspy and are produced continually for tens of minutes at a time over several hours or more in a night. Even in the wild, these insects are easy for a human observer to locate in a matter of minutes. In contrast, katydids that inhabit heavily forested areas, and thus are potential prey for foliage gleaning bats, have calls that make them much more difficult to find. Resembling high-pitched whistles, they sing once or twice per minute for only a fraction of a second each, making them virtually impossible for a human observer to locate. To learn whether bats, like humans, experience difficulty in finding insects that only call sporadically, bats were brought into flight cages for study.

Single Round-eared bats were placed in a cage with a katydid— either Scopiorinus fragilis, a species that sings frequently (at least one call per second) or Acanthodis curvidens, a species that sings infrequently (about one call per minute). The time required for a bat to localize singing insects, and the ease with which they could do this, was used as a measure of the protection afforded by each call type.

The difference in response time to the two types of calls was dramatic. Bats located katydids that produce frequent calls in about 26 seconds, with an immediate and direct flight from their perch to the singing insect. In contrast, bats required nearly 34 minutes to locate the katydids that called less often. The short calls appear to be advantageous to the katydid because they do not allow sufficient time to pin-point its precise location. Instead of a direct flight to the sound source, the bat made short aimless flights, lasting only as long as the insect's call. The few seconds that the insect sang were not long enough for the bat to successfully locate its target. Moreover, during the intervals when the katydid was silent, the bat flew about the cage as if in search of the insect it knew was there but could not find. On several occasions during such flights, the bat landed within a few inches of a silent katydid, but could not locate it. Since continuous flight is energetically inefficient for the broad-winged foliage gleaning bats, it is presumed that if the experiment were repeated in the wild, the bat would have left the area to forage elsewhere.

The calls of about 40 species of katydids were recorded on Barro Colorado, and all forest species examined, with one exception, produced very short calls. The exception appears to prove the rule that short calls are advantageous in the presence of bats that respond to insect calls. Ischnomela pulchripennis is a large forest katydid whose long, noisy calls sound remarkably like those of katydids found in clearings. This katydid has only been found on a species of plant in the pineapple family (Aechmea magdalanae) that grows to be about six feet tall and is covered with inch-long, sharp spines. Since foliage gleaning bats do not forage near these plants with spines that could injure their delicate wings, the katydids in them can, therefore, sing freely without the threat of being eaten.

The cost of silence
Although calling less frequently is a good adaptive strategy for katydids that co-habit areas with large numbers of foliage gleaning bats, it defeats the principal purpose of calling–to acquire a mate. To compensate, the katydids studied were discovered to supplement their short calls with signals that could be perceived by other katydids, but not by bats. The katydid signals its presence with tremulations, body vibrations which are produced when the insect shakes rapidly in an up and down motion, appearing to do push ups. As it sits on leaves, plant stems, or other bits of vegetation, a tremulating katydid sets up a series of vibrations sometimes strong enough to visibly shake the plants. The vibrations travel through the vegetation to another katydid and are perceived with specialized sensory structures on the legs. Like audible calls, the way in which they tremulate is unique to each species.

Males tremulate to attract females, and females respond to males of their species with tremulations of their own. In the laboratory, where this has been observed, katydids on a common surface, but unable to see one another, will communicate in this manner for hours at a time, separated by distances of up to 10 feet.

Katydids worldwide are considered to be among the most acoustically active of insects. Therefore, the finding that some species purposely reduced their use of sound was unexpected. Until these studies, tremulation was not known as a major form of communication in katydids; it appears to be a direct response to bat predation. However, tremulations also appear to be more energetically costly to produce.

Prior to these studies, little was known about the feeding ecology of New World foliage gleaning bats. Even less was known about the effects of bat predation on the development of anti-predator behavior in the insects on which they feed, and the results dramatically demonstrate the degree to which predator and prey can affect one another. To understand the complexity of such relationships, it will be important to conduct long-term studies on animals in their natural environments and in the presence of the other organisms that affect them.


1. Roeder, K.D. 1967. Nerve Cells and Insect Behavior. Harvard University Press, Cambridge, Massachusetts. 188 pp.

2. Miller, L.A., and J. Olesen. 1979. Avoidance behavior in green lacewings. I. Behavior of free flying green lacewings to hunting bats and ultrasound. J. Comp. Physiol. 131:113-120.

3. Moiseff, A., G. Pollack, and R.R. Hoy. 1978. Steering responses of flying crickets to sound and ultrasound: male attraction and predator avoidance. Proc. Nat. Acad. Sci. U.S.A. 75:4052-4056.

4. Yager, D.D., and R.R. Hoy. 1986. The cyclopean ear: a new sense for the praying mantis. Science 231:727-729.

5. Fullard, J.H. 1977. Phenology of sound-producing arctiid moths and the activity of insectivorous bats. Nature 267:42-43.

6. Hill, J.E., and J.D. Smith. 1984. Bats, a natural history. British Museum (Natural History). London.

7. Vaughan, T.A. 1976. Nocturnal behavior of the African false vampire bat (Cardioderma cor). J. Mammal. 57:227-248.

8. Belwood, J.J. 1988. Foraging behavior, prey selection, and echolocation in phyllostomine bats (Phyllostomidae). Pages __ – __ in Animal Sonar: Processes and Performance, P.E. Nachtigall, ed. Plenum Press, NY, in press.

9. Tuttle, M.D., and M.J. Ryan. 1981. Bat predation and the evolution of frog vocalizations in the neotropics. Science 214:677-678.

10. Tuttle, M.D., M.J. Ryan, and J.J. Belwood. 1985. Acoustical resource partitioning by two species of phyllostomid bats (Trachops cirrhosus and Tonatia silvicola). Anim. Behav. 33:1369-1371.

11. Belwood, J.J., and G.K. Morris. 1987. Bat predation and its influence on calling behavior in neotropical katydids. Science 238:64-67.

12. Gardner, A.L. 1977. Feeding habits. Pages 293-350 in Biology of Bats of the New World Family Phyllostomatidae, Part II, R.J. Baker, J.K. Jones, Jr., and D.C. Carter, eds. Spec. Pub. Mus. Texas Tech. Univ. 364 pp.

Jacqueline J. Belwood recently completed a Ph.D. in Entomology (the study of insects) at the University of Florida. She has studied the feeding ecology of insect-eating bats in Ontario, Canada and Hawaii, and has examined bat-katydid interactions and katydid biology in Panama and most recently in Peru with the help of 52 Earthwatch volunteers. She is currently working on the feeding ecology of mustached bats in Jamaica and will return to Panama later this year as a Smithsonian Fellow to resume her post-doctoral studies on bats and katydids.

Opposite page: Unlike the more familiar kinds of bats that catch and eat their prey in flight, foliage gleaners pluck large stationery insects directly from their perches. The Frog-eating bat, a carnivorous foliage gleaner, sometimes supplements its diet of frogs with large insects, such as katydids.

Above: Katydids are well known for their ability to disappear into their surroundings, an adaptation to avoid being eaten. Visually oriented predators would have a difficult time distinguishing this katydid of the Mimetica species from a leaf.

Left: Katydids are an important source of food for foliage gleaning bats such as this Round-eared bat, but because the insect is so large, the bat must take its prey to a feeding roost.

Right: This Round-eared bat has selected an abandoned termite nest for its feeding roost. It will eat the katydid only after it has carefully culled the inedible parts.

Below: the discarded parts of its prey accumulate in great numbers below the feeding roost, allowing researchers to learn more about the bat's diet.

A hungry Frog-eating bat demonstrates dramatically that visual clues to find its prey are unimportant. This one is responding to the recorded mating calls of a mud puddle from. These bats identify and locate frogs by their calls and have had a major impact on the evolution of vocal communication in certain frogs. Male frogs must call to attract mates, but face the dilemma of how to do so without calling a Frog-eating bat instead.

Katydids in densely forested areas have had to learn how to attract mates without advertising their location to foliage gleaning bats. Some species, such as this Melanonotus bradleyi, have developed a system of silent communication through body vibrations called tremulations.