American Society of Mammalogists

Food Habits of Certain Colombian Bats

Author(s): Andrew A. Arata, John B. Vaughn and Maurice E. Thomas
Source: Journal of Mammalogy, Vol. 48, No. 4 (Nov., 1967), pp. 653-655
Published by: American Society of Mammalogists
Stable URL: http://www.jstor.org/stable/1377595 .
Accessed: 16/11/2013 18:48

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

American Society of Mammalogists is collaborating with JSTOR to digitize, preserve and extend access to
Journal of Mammalogy.

http://www.jstor.org

This content downloaded from 148.61.13.133 on Sat, 16 Nov 2013 18:48:37 PM

All use subject to JSTOR Terms and Conditions
November 1967 GENERAL NOTES 653

FOOD HABITS OF CERTAIN COLOMBIAN BATS

Generalizations in the literature suggest that most genera of the Phyllostomatidae are
obligate fruit, flower, and nectar feeders. The partially insectivorous genus Micronycteris
and the omnivorous genera Phyllostomus and Vampyrum are considered to be unique in
this family (Allen, 1939; Dunn, 1933; Goodwin and Greenhall, 1961).
In a survey of certain zoonotic infections in Colombian bats, an effort was made to test
this generalization. More than 1600 bats were collected in the Departamento del Valle del
Cauca in July and August 1964. Concurrentwith removal of organs for isolation of patho-
gens and parasites, stomachs of 1000 bats were removed and preserved in 10% formalin
for future study. A majority (80%) of the stomachs were empty. This was because live
animals were preferred for viral isolation attempts, and many bats sacrified in the laboratory
each day were collected the previous night. Further, for valid salmonellae isolation, the
animals were not fed or watered after capture. This report is based on the stomach contents
of 180 animals, representing 4 families, 9 genera, and 13 species.
The principal areas of bat collection were in the vicinity of Call and to the east of
Buenaventura. Cali is situated in the Cauca Valley between the western and central Andean
cordilleras (altitude 1000 m; mean monthly temperatures, 24.50 C - 0.5?; annual rainfall,
1500 mm). The area is heavily cultivated and retains little native forest vegetation. Buena-
ventura is on the Pacific coast (elevation of collecting sites was less than 100 m; mean
monthly temperatures, 30? C - 2?; annual rainfall, more than 7000 mm). The vegetation
is tropical rain forest.
Near Cali, most collections with mist nets were made in plantations of bananas, plan-
tains, citrus fruits, mangos and papayas, cocoa and coffee. Near Buenaventura,some collec-
tions were made in cultivated areas, but most were made in forest clearings or over streams.
Peropteryx,Molossus, Desmodus, and about half of the Carollia were taken in roosts or by
netting at the entrances of roosts. All other bats were taken in mist nets during periods
of their normal activity.
Identification of the well-masticated contents of bat stomachs is especially difficult in
South American forms in the absence of adequate reference collections of fruits, seeds,
and insects. Thus we have not attempted to identify food items beyond general categories
(fruit pulp, seeds, insects, flesh, etc.). Pulpy material was scored as unidentifiable in
Molossus and Peropteryx unless chitinous insect remains were also present. Pulp was
scored as plant in the known fruit eaters, as in virtually all cases it occurred along with
recognizable plant epidermis, seeds, or fibers. Stomachs of nearly all bats contained loose
hair, probably the result of grooming activity. Predation or cannibalism was noted by the
presence of bat flesh with hair and claws in several stomachs.
A summaryof the stomach contents of 180 bats from southwestern Colombia is presented
in Table 1. Certain bats are representedby small numbers, and observationsare inconclusive
for these species. Obligate insectivores such as Peropteryx and Molossus yielded nothing
but insects, providing no new information, but the 120 phyllostomatids and 23 Desmodus
examined yielded somewhat unexpected results.
More than a fifth of the Carollia perspicillata contained insect remains. Many of the
insects were large; their frequency, and quantity in individual stomachs, rule out accidental
ingestion while eating fruit. One of us (Arata) has maintained Carollia in captivity, and
has observed limited flight, including banking and turning in confined quarters (3 X 3 X
2 ft), indicating surprising maneuverability and the ability to take insects in flight. As
Carollia feeds at least twice nightly (Goodwin and Greenhall, 1961), percentages indicat-
ing insectivorous activity may be more significant than is first apparent, and the frequency
of predation on insects by Carollia may be several times higher than revealed in this study.
Flesh, matted hair, and skin also were found in Carollia stomachs. The hair was identi-
fied as that of Carollia. In one instance a pair of claws was found in association with matted
hair and flesh. The amount of hair and its association with remnants of flesh and claws

This content downloaded from 148.61.13.133 on Sat, 16 Nov 2013 18:48:37 PM

All use subject to JSTOR Terms and Conditions
654 JOURNAL OF MAMMALOGY Vol. 48, No. 4

TABLE 1.-Stomach contents of 180 Neotropical bats taken in the Departamento del Valle
del Cauca, Colombia, July and August, 1964.
Number
ex- Matted hair, Not
Species amined Plants Insects claws, flesh Blood identifiable

Phyllostomus hastatus 7 6 (86%) 3 (43%) - - -

Phyllostomus discolor 1 1 (100%) 1 (100%) - -
Glossophaga sorcina 16 15 (94%) 6 (37%) 1 (6%) -
Carollia perspicillata 74 71 (96%) 16 (22%) 6 (9%)
Sturnira lilium 5 5 (100%) -

Vampyrops helleri 3 3 (100%) - - -

Vampyrops dorsalis 4 3 (75%) 1 (25%)
Artibeus jamaicensis 1 1 (100%) - - -
Artibeus lituratus 4 3 (75%) 1 (25%) - -
Artibeus cinereus 5 5 (100%) 1 (20%) - -
Desmodus rotundus 23 - 4 (17%) 4 (17%) 23 (100%)
Molossus major 20 - 15 (75%) - - 5 (25%)
- 5 (29%) - - 12 (71%)
Peropteryx macrotis 17

rule out incidental ingestion during grooming activities. Though isolated claws are not
very diagnostic, those found in Carollia were indistinguishable from those of the same
species. Whether these instances of cannibalism represent direct predation within the
roosts or eating of young or dead animals is not known. The size of the claws did not sug-
gest a young animal, however.
Carollia is generally considered to be an obligate fruit eater. Goodwin and Greenhall
(1961) listed a large variety of fruits (22 species) eaten by Carollia. Most other reports
are general, referring to bananas and other cultivated fruits, or simply stating that Carollia
subsists on "variousfruits." In actuality, little detailed work has been done on phyllostoma-
tid food habits. Most of the work that has been done is observational,which, though impor-
tant, should be balanced with stomach content analysis. Insects taken in flight and less
frequently eaten foods may be overlooked in purely observationalstudies.
As reported previously (Carvalho, 1961), Glossophaga also feeds on insects to some
extent. The large size of the insects encountered suggests that they were taken as food and
not as incidental captures associated with nectar feeding as suggested by Goodwin and
Greenhall (1961). In any case, the small size of an insect found in a bat stomach is not a
valid indicator of accidental ingestion, as Ross (1961) has shown that even obligate insecti-
vores such as Tadarida feed on insects between 5 and 9 mm in total length. A single
individual also contained matted hair still attached to flesh. Carnivorous behavior in
Glossophaga is unreported and unexpected.
Although all Desmodus stomachs contained blood, several contained insect remains and
muscle fibers. The muscle fibers were in chunks (5 mm and more), as though taken in a
single bite and bolted with no mastication. These pieces of flesh were common in the
specimens in which they occurred. Goldman (1920) commented that Desmodus removes
a piece of skin in biting, but the quantity of flesh observed in the stomach contents suggests
that additional pieces of flesh are also ingested, or that a number of initial bites are made.
The insect remains found in Desmodus stomachs consisted of a whole ingested ectoparasite
(not included in Table 1) and well-broken remains of larger insects.
More than a fourth of the 120 phyllostomatids examined contained material other than
plant. This high proportionis not dependent upon the incidence in any one form, as seven
of the 10 species contained insects or flesh. Rather, those that contained only plant material
in their stomachs were represented by too few specimens to warrant any conclusion. These
large frequencies of insect and bat material in the stomach contents of bats that we have
casually considered to be obligate fruit eaters suggest the need for more studies on the

This content downloaded from 148.61.13.133 on Sat, 16 Nov 2013 18:48:37 PM

All use subject to JSTOR Terms and Conditions
November 1967 GENERAL NOTES 655

habits of these animals. Of particular importance would be the study of seasonal and
geographic trends in food habits.
As most phylogenetic interpretationsof chiropteran evolution suggest that ancestral bats
(and more specifically, phyllostomatids) were insectivorous, it is possible that dietic re-
quirements from early habits are still operative in forms that have otherwise adapted to
a different mode of life.
In addition to the purely biological interest in the food habits of these animals, the
increased interest in bats, particularly tropical forms, in zoonotic epidemiology, requires
more detailed information on their food habits for a more comprehensive knowledge of
their parasitic and microbial associates.
Various aspects of this study were supported in part by grants from the National Science
Foundation (GB 2458) and the National Institutes of Health (TW00143), I.C.M.R.T.
Award.
LITERATURE CITED

ALLEN, G. M. 1939. Bats. Harvard Univ. Press, Cambridge, x + 368 pp.

CARVALHO, C. T. 1961. Sobre os habitos alimentares de phillostomideos (Mammalia,
Chiroptera). Rev. Biol. Trop., 9: 53-60.
DUNN, L. 1933. Observations on the carnivorous habits of the spear-nosed bat, Phyllo-
stomus hastatus panamensis Allen, in Panama. J. Mamm., 14: 188-199.
GOLDMAN, E. A. 1920. Mammals of Panama. Smiths. Misc. Coll., 69 (5): 1-309.
GOODWIN, G. G., AND A. M. GREENHALL. 1961. A review of the bats of Trinidad and
Tobago. Bull. Amer. Mus. Nat. Hist., 112: 191-301.
Ross, A. 1961. Notes on food habits of bats. J. Mamm., 42: 66-71.
ANDREW A. ARATA, JOHN B. VAUGHN, AND MAURICE E. THOMAS, Department of Biology,
and Division of Epidemiology, Department of Tropical Medicine and Public Hygiene, Tulane
University, New Orleans, Louisiana 70118, and Universidad del Valle, Cali, Colombia.
Accepted 3 April 1967.

HYPOTHERMIAOBSERVED IN WATER-CHILLED DEER

An adult male white-tailed deer (Odocoileus virginianus) was found at 10:00 AM on
26 November 1966 in Fish Lake on the Cedar Creek Natural History Area of the University
of Minnesota, after having fallen through the ice which partially covered the lake. When
brought to shore, its hair was thoroughly soaked and the animal appeared to be completely
exhausted. The deer did not move for a time, but "running"movements began as it lay
prostrate on the ground. In experimental hypothermia, Simpson and Herring (J. Physiol.,
32: 305-311, 1905) noted a rhythmicalmovement of the limbs of cats with a body tempera-
ture of 24? C.
After an hour of rubbing its body and bending its legs to stimulate circulation, the deer
showed little sign of recovering. It was then hauled to the heated garage at the Cedar
Creek Laboratory. Water temperature, measured the following day, was between 0.4 and
2.0? C.
In the garage, a Yellow Springs No. 406 tubular thermistor probe was inserted in the
rectum to a depth of 5 inches, and the temperature was read on the Yellow Springs Tele-
thermometerModel 46 TUC at 30-minute intervals. The initial rectal temperatureobserved
was 26.4? C, followed by a steady rise to 39.5?, after which it declined to 38.0?. The
behavior of the animal as it warmed up was observed continuously until the temperature
stabilized, 12 hours after being brought to the garage. Shivering occurred sporadically
throughout the warming period and a gradual increase in alertness and activity was ob-
served (Fig. 1). The deer did not stand up for more than a few seconds, and showed no
overt signs of fear.

This content downloaded from 148.61.13.133 on Sat, 16 Nov 2013 18:48:37 PM

All use subject to JSTOR Terms and Conditions