Abstract
In our study of bat diversity in the Amazon Basin, we cap-
tured bats in undisturbed continuous forest and in forest 
fragments at the Biological Dynamics of Forest Fragments
Project (BDFFP) near Manaus, Brazil, from January 1996
until July 1999. We recorded 72 species of bats in a sample
of more than 7700 individuals caught during 29,900 mistnet
hours in terra-firme forest. Species accumulation curves and
mathematical estimates of species numbers based on the
number of species captured with standardized methodology
suggest that we sampled about 95% of the entire expected
bat fauna of the area, including aerial insectivorous bats. Our
results are similar to those of other mistnetting inventories of
Amazonian bat assemblages in terms of species composition
and number of species per bat family. Some species consid-
ered widespread in Central Amazonia and expected at our
study site were not recorded. We interpret their absence as
effects of sampling bias and of local ecological conditions.
We know from acoustic monitoring (i.e., identification of
bats by their echolocation calls) that our mistnet data are
incomplete for aerial insectivorous species. We conclude that
the development of comprehensive inventories of key verte-
brate taxa such as bats derived from a combination of several
standardized sampling procedures is essential to develop
meaningful, conservation-oriented plans for land-use and
management of protected areas.
Resumo
Em nosso estudo sobre diversidade de morcegos na Bacia
Amaz?nica, n?s capturamos morcegos em floresta cont?nua
n?o perturbada e em fragmentos florestais no Projeto
Din?mica Biol?gica de Fragmentos Florestais (PDBFF),
pr?ximo ? Manaus, Brazil, de Janeiro de 1996 at? Julho de
1999. N?s registramos 72 esp?cies de morcegos capturando
cerca de 7700 indiv?duos em 29,900 horas de captura em
?reas de terra-firme. As curvas de acumula??o e modelos
matem?ticos baseados no n?mero de esp?cies capturadas
com metodologia estandartizada indicam que n?s regis-
tramos cerca de 95% da fauna esperada para a ?rea, incluindo
as capturas de esp?cies de morcegos a?reos inset?voros.
Nossos resultados s?o equivalentes a outros invent?rios
baseados em redes de captura de morcegos na Amaz?nia, em
termos de composi??o de esp?cies e em n?mero de esp?cies
para cada fam?lia de morcegos. Algumas esp?cies consider-
adas como uniformemente distribu?das na regi?o Amaz?nica
e que deveriam tamb?m ocorrer na nossa ?rea de estudo n?o
foram coletadas. N?s interpretamos a aus?ncia de esp?cies
esperadas como um efeito de limita??es na metodologia e
devido a condi??es ecol?gicas locais. Sabemos por moni-
toramento ac?stico (identifica??o de morcegos atrav?s dos
sinais de ecoloca??o), que nossos dados de redes de captura
est?o incompletos para morcegos inset?voros a?reos. N?s
conclu?mos que o desenvolvimento de invent?rios de grupos
A Biodiversity Assessment of Bats (Chiroptera) in a Tropical
Lowland Rainforest of Central Amazonia, Including
Methodological and Conservation Considerations
Erica M. Sampaio1,6,7, Elisabeth K. V. Kalko2,5,6, Enrico Bernard3,7, Bernal Rodr?guez-Herrera4,7 and Charles O. Handley Jr.6?
1University of T?bingen, Animal Physiology, Auf der Morgenstelle 28, 72076 T?bingen, Germany
2University of Ulm, Experimental Ecology, Ulm, Germany
3Department of Biology, York University, Toronto, Ontario, Canada
4Costa Rica National Museum, Natural History, San Jos?, Costa Rica
5Smithsonian Tropical Research Institute, Balboa, Panam?
6National Museum of Natural History, Washington, D. C., USA
7Biological Dynamics of Forest Fragments Project, INPA, Manaus, AM, Brazil
Studies on Neotropical Fauna and Environment 0165-0521/03/3801-017$16.00
2003, Vol. 38, No. 1, pp. 17?31 ? Swets & Zeitlinger
Received: 3 August 2000
Accepted: 21 August 2002
Correspondence: E. K. V. Kalko, University of Ulm, Experimental Ecology, Albert-Einstein Allee 11, 89069 Ulm, Germany. Fax: +49-731-
50-22683; E-mail: elisabeth.kalko@biologie.uni-ulm.de
18 E.M. Sampaio et al.
chave, como morcegos, com a combina??o de v?rios
m?todos de amostragem estandardizados ? essencial para o
desenvolvimento de planos conservationistas significativos
no uso da terra e no controle de ?reas protegidas.
Keywords: Chiroptera, inventories, mistnets, sampling
methods, conservation, Central Amazonia.
Introduction
Mammalian diversity in the region of the Brazilian Amazon
is very high with about 320 species currently recognized
(Fonseca et al., 1996; Voss & Emmons, 1996; Emmons &
Feer, 1997). Bats (Chiroptera) account for around 40% of this
diversity (see Koopman, 1993; Marinho-Filho & Sazima,
1998). Overall, the region of the Brazilian Amazon, along
with the Guyana region (Brosset et al., 1996; Simmons &
Voss, 1998; Cosson et al., 1999; Lim & Engstrom, 2001),
holds one of the most diverse bat faunas in the Neotropics
(e.g., Mok et al., 1982; dos Reis & Peracchi, 1987; Hutterer
et al., 1995; Voss & Emmons, 1996; Bernard & Fenton,
2002).
Because of the global biodiversity crisis, caused mainly
by anthropogenic habitat destruction and degradation, the
search for answers to the questions: ?Which species rely on
undisturbed forests? and ?Which species readily adapt to dis-
turbed habitats?? is becoming increasingly important. In this
context, detailed studies on the diversity of local species
assemblages are crucial, particularly in forests that are seri-
ously affected by the continuing increase of the human pop-
ulation and consequent urbanization and land conversion
processes (Dale et al., 1994; Ferreira & Laurance, 1996;
Granjon et al., 1996; Whitmore, 1997; Gascon & Lovejoy,
1998; Laurance et al., 1998). Studies on diversity related to
conservation issues are needed to develop and apply efficient
management programs to preserve and maintain the remain-
ing diversity. However, our knowledge of species diversity is
still poor for many taxa. Faunal inventories in the Amazon
Basin continue to uncover new species even in well-studied
groups such as primates (Ferrari & Lopes, 1992; Mittermeier
et al., 1992; Queiroz, 1992; van Roosmalen et al., 1998).
There is increasing evidence that Neotropical bat assem-
blages are strongly affected by habitat alterations. A common
result is the loss of rarer and specialized species and an
increase in abundance of a few generalists (Fenton et al.,
1992; Simmons & Voss, 1998; Cosson et al., 1999).
However, we cannot be sure yet about the overall effects of
habitat alterations on bats because most inventories of bat
faunas are limited to few sites and to mistnetting at ground
level. Adding other methods such as mistnetting in higher
forest strata result in a more complete list of species present
(e. g., Simmons & Voss, 1998; Kalko & Handley, 2001;
Bernard, 2001).
To increase our knowledge about the bats in the Amazon
Basin we applied canopy- and ground-level mistnets to
inventory bat assemblages near Manaus, Brazil, in the 
area of the Biological Dynamics of Forest Fragments 
Project, BDFFP (Bierregaard et al., 1997). Here, we pool 
the data of two projects that were conducted in parallel at 
the BDFFP to give a first overview of local species 
composition. Mistnetting during a period of more than three
years revealed new records for some bat species and 
new information on the distribution patterns of others. To
evaluate the quality of our inventory data, we applied 
the concept of species accumulation curves (Gotelli &
Graves, 1996) and species richness estimators (Colwell &
Coddington, 1996) and discuss differences between number
of recorded and expected species. Detailed analysis of the
individual projects are given in subsequent publications 
(e.g., Bernard, 2001; Berhard & Fenton, 2002).
Materials and methods
Study area
The Biological Dynamics of Forest Fragments Project
(BDFFP; Fig. 1), formerly the Minimum Critical Size of
Ecosystems project, MCSE (Lovejoy & Bierregaard, 1990),
is a cooperative project between the Instituto Nacional 
de Pesquisas da Amaz?nia (INPA) and the Smithsonian 
Institution, Washington, D. C. (USA), represented by the 
Smithsonian Tropical Research Institute (STRI) in Panam?.
The research area is located 80km north of Manaus 
(2?24?S, 59?43?W, 2?25?S, 59?45?W; Lovejoy & Bierregaard,
1990). The predominant habitat in this region of the Amazon
Basin is unflooded, lowland primary forest (terra-firme). The
soils are well-drained, nutrient-poor, yellowish latosols
(Chauvel, 1983). The BDFFP area receives between 2200
and 2600mm of annual precipitation that mostly falls during
the rainy season from January to May. The dry season usually
extends from June to December. Canopy height is between
30 and 37m with emergent trees reaching 50m. The 
dominant trees belong to the families Chrysobalanaceae,
Lecythidaceae, Sapotaceae, Leguminosae, Burseraceae, 
and Bombacaceae (Gentry, 1990; Prance, 1990; Rankin-
De-Merona et al., 1992). The ground level is dominated by
palms (Scariot, 1999) of the genera Astrocaryum, Attalea,
and Bactris. Flowering and fruiting peaks usually occur in
the dry and rainy season, respectively (Rankin-De-Merona 
et al., 1992).
Mistnetting of bats
We set mistnets at ground level and in the canopy following
standardized procedures. Project 1 (Effects of forest frag-
mentation on bat assemblages) spanned three and a half years
(January 1996?June 1999), covering three consecutive rainy
and dry seasons. Project 2 (Vertical stratification of bats) ran
for one year, from August 1996 to August 1997. We avoided
sampling during third-quarter and full moon when light
intensity is usually high and bat activity is mostly low
Bat Species Richness in Amazonia 19
(Crespo et al., 1972; Morrison, 1978). All mistnets were 12
m long and 2.5m high, with four shelves and mesh dimen-
sions of 1.5?3 ? 1.5?3cm. We calculated capture effort for
each sampling area in mistnet hours. One mistnet hour (mnh)
corresponds to one 12-m mistnet open for one hour. Ground
nets were tied to poles, with the first shelf string at ground
level. Canopy nets were set in treefall gaps by using a pulley
system following the technique described by Humphrey et al.
(1968), a system successfully applied in previous bat studies
(Handley, 1967; Kalko & Handley, 2001). After capture, we
removed bats from the mistnets and placed them temporar-
ily in soft cloth bags. Bats were identified using a dichoto-
mous key for the bats of the Amazon Basin developed by
Handley, Sampaio, and Kalko (unpublished). The taxonomy
used herein follows Koopman (1993), with the exception of
Anoura caudifer, Artibeus (Koopmania) concolor, Eptesicus
chiriquinus, Glyphonycteris daviesi, Glyphonycteris sylve-
stris, Lampronycteris brachyotis, and Trinycteris nicefori
that follow the classification of Simmons & Voss (1998) 
and the subdivision of the genus Tonatia into Lophostoma
and Tonatia proposed by Lee et al. (2002). With the 
exception of rare species or species whose identification was
problematic, all bats were released at the respective capture
sites. Bats collected for further identification and a few 
individuals that died during handling were fixed in 10%
formaldehyde and preserved in 70% alcohol as voucher 
specimens (Handley, 1988). Vouchers (n = 124) were sent 
for identification to the National Museum of Natural 
History (NMNH), Washington, DC (USA) and were sub-
sequently deposited in the INPA collection in Manaus for
further reference (Table 1).
Project 1
In Project 1, we collected bats with mistnets set at ground
level in a variety of habitat types. We sampled bats on 
standardized transects in three replicates each of 1 and 10ha
forest fragments (Colosso, Dimona, and Porto Alegre) and in
three replicates each of 1 and 10ha plots embedded in 
continuous forest (Cabo Frio, Florestal, and km 41) resulting
in a total of 12 study sites (Fig. 1). The transects design 
was adapted to the existing trail system which consisted 
of parallel lanes crossing the fragments or plots in N?S 
direction. Seven 12-m mistnets were set. The capture 
effort for each replica was about 150mnh per mistnet night.
In addition to the transects, we repeatedly collected bats 
with ground mistnets in three different pasture areas, in sec-
ondary growth composed of Vismia sp. and Cecropia sp., and
at three forest edges, penetrating up to 400m into continu-
ous forest. We opened the nets from 18:00 until 24:00 and
patrolled them at 20-minute intervals. The total capture effort
in project 1 amounted to 28,000mnh. In addition to the tran-
sects, we occasionally set nets in the BDFFP area across
streams or near water and in front of potential bat roosts. We
also netted outside the BDFFP near the road ZF2 leading 
to silvicultural research sites of EMBRAPA (Brazilian 
Agricultural Research Corporation) and at the entrances of
caves near the village of Presidente Figueiredo about 100km
Manaus
K
m 41
Rio Solim?
es
Flo
restal
G
avi?o
C
abo Frio
Po
rto Alegre
Di
mona
B
r
 
1
7
4
Z
F
-
0
3
C
olosso
300 1500 3000
Scale Meters
N
Geo
Rio Urubu
60  05? 00? W 60  00? 00? W 59  55? 00? W 59  50? 00? W 59  45? 00? W 59  40? 00? W
O
O
02
  2
0?
 0
0?
 S
   
 
O
02
  2
5?
 0
0?
 S
   
 
O
OOO O O
Forest
Clear cut
Camp
Roads Drainage
River
Stream
L e g e n d
Lake
Highway
Farm Road
Unpaved Road
Biological Dynamics of Forest Fragments Project reserves
Source: INPE Landsat TM 5,4,3 - RGB, 1995. 
Prepared in June 1998 by Venticinque, 
E. M. and  Fernandes, T. L. N. 
Fig. 1. Experimental area of the Biological Dynamics of Forest Fragmentation Project, about 90 km north from Manaus, Amazonian, Brazil.
Map obtained by BDFFP office. Used with permission of the authors.
20 E.M. Sampaio et al.
Table 1. Specimens collected during the field work and deposited at the mammal collection of the Zoological Museum at INPA, Manaus
(Dr. Maria Nazareth Pereira da Silva, curator). The data given are: INPA number, species, locality1, collector2, and date.
INPA 2493, Lasiurus cf. castaneus, ?PDBFF, km 41?, C. O. H., 26.01.96; INPA 2494, Lasiurus cf. castaneus, ?PDBFF, km 41?, C. O. H.,
26.01.96; INPA 2495, Centronycteris maximiliani, ?PDBFF, CO?, E. M. S., 16.02.96; INPA 2496, Phyllostomus latifolius, Pres. Fig., C. O.
H., 19.11.97; INPA 2497, Phyllostomus latifolius, Pres. Fig., C. O. H., 19.11.97; INPA 2498, Lionycteris spurrelli, ?PDBFF, GA?, E. B.,
12.04.97; INPA 2499, Micronycteris microtis, ?PDBFF, GA?, E. B., 07.02.97; INPA 2500, Trinycteris nicefori, ?PDBFF, GA?, E. B.,
13.02.97; INPA 2501, Micronycteris schmidtorum, ?PDBFF, GA?, E. B., 06.08.97; INPA 2502, Centronycteris maximiliani, ?PDBFF, GA?,
E. B., 30.04.97; INPA 2503, Cynomops abrasus, ?PDBFF, km 41?, E. B., 01.02.97; INPA 2504, Peropteryx leucopterus, ?PDBFF, FLO?, E.
M. S., 26.08.97; INPA 2505, Carollia brevicauda, ?PDBFF, FLO?, E. M. S., 13.06.97; INPA 2506, Choeroniscus minor, ?PDBFF, FLO?, E.
M. S., 26.08.97; INPA 2507, Trinycteris nicefori, ?PDBFF, FLO?, E. M. S., 26.08.97; INPA 2508, Myotis riparius, ?PDBFF, CO?, E. M. S.,
18.06.97; INPA 2509, Myotis riparius, ?PDBFF, GA?, E. M. S., 24.01.97; INPA 2510, Eptesicus chiriquinus, ?PDBFF, km 41?, E. M. S.,
04.02.97; INPA 2536, Ametrida centurio, ?PDBFF, GA?, E. B., 09.08.97; INPA 2537, Artibeus gnomus, ?PDBFF, km 41?, E. B., 31.10.96;
INPA 2538, Vampyressa brocki, ?PDBFF, km 41?, E. B., 31.10.96; INPA 2539, Eptesicus chiriquinus, ?PDBFF, km 41?, E. B., 30.01.97;
INPA 2540, Lophostoma schulzi, ?PDBFF, GA?, E. B., 06.08.97; INPA 2541, Centronycteris maximiliani, ?PDBFF, DI?, E. M. S., 23.09.97;
INPA 2542, Cormura brevirostris, ?PDBFF, CO?, E. M. S., 21.11.97; INPA 2543, Saccopteryx leptura, ?PDBFF, CO?, E. M. S., 21.11.97;
INPA 2544, Molossus molossus, ?PDBFF, DI?, E. M. S., 05.10.97; INPA 2545, Glyphonycteris sylvestris, ?PDBFF, FLO?, E. M. S.,
04.11.97; INPA 2546, Lophostoma schulzi, ?PDBFF, km 41?, E. M. S., 26.10.96; INPA 2547, Platyrrhinus helleri, ?PDBFF, PA?, E. M. S.,
03.12.97; INPA 2548, Myotis nigricans, ?PDBFF, PA?, E. M. S., 14.10.97; INPA 2549, Myotis riparius, ?PDBFF, km 41?, E. M. S.,
28.10.96; INPA 2571, Choeroniscus minor, ?PDBFF, DI?, E. M. S., 15.01.98; INPA 2572, Saccopteryx leptura, ?PDBFF, km 41?, E. M. S.,
27.01.98; INPA 2573, Myotis sp., ?PDBFF, CO?, E. M. S., 06.02.98; INPA 2574, Trinycteris nicefori, ?PDBFF, PA?, E. M. S., 09.02.98;
INPA 2575, Centronycteris maximiliani, ?PDBFF, FLO?, E. M. S., 09.03.98; INPA 2576, Molossus molossus, ZF3 km 18, E. M. S.; INPA
2577, Choeroniscus minor, ?PDBFF, FLO?, E. M. S., 08.03.98; INPA 2578, Glyphonycteris daviesi, ?PDBFF, DI?, E. M. S., 18.02.98; INPA
2579, Micronycteris minuta, ?PDBFF, FLO?, E. M. S., 09.03.98; INPA 2622, Carollia perspicillata, ?PDBFF, PA?, E. M. S., 06.03.96; INPA
2623, Carollia perspicillata, ?PDBFF, PA?, E. M. S., 06.03.96; INPA 2624, Pteronotus parnellii, ?PDBFF, km 41?, E. M. S., 20.03.96;
INPA 2625, Glyphonycteris daviesi, ?PDBFF, km 41?, E. M. S., 24.03.96; INPA 2626, Carollia brevicauda, ?PDBFF, PA?, E. M. S.,
12.10.96; INPA 2627, Carollia perspicillata, ?PDBFF, PA?, E. M. S., 12.10.96; INPA 2628, Peropteryx macrotis, Pres. Fig., E. M. S.,
20.11.96; INPA 2629, Carollia perspicillata, ?PDBFF, CO?, E. M. S., 06.12.96; INPA 2630, Carollia perspicillata, ?PDBFF, CO?, E. M. S.,
07.12.96; INPA 2631, Micronycteris minuta, Silv. Treat., E. M. S.; INPA 2632, Ametrida centurio, ?PDBFF, DI?, E. M. S., 02.08.97; INPA
2633, Ametrida centurio, ?PDBFF, DI?, E. M. S., 02.08.97; INPA 2634, Pteronotus parnellii, ?PDBFF, PA?, E. M. S., 05.08.97; INPA 2635,
Carollia perspicillata, ?PDBFF, DI?, E. M. S., 07.10.97; INPA 2636, Sturnira tildae, ?PDBFF, PA?, E. M. S., 09.10.97; INPA 2637, Carollia
perspicillata, ?PDBFF, CO?, E. M. S., 22.11.97; INPA 2638, Carollia brevicauda, ?PDBFF, DI?, E. M. S., 14.01.98; INPA 2639, Carollia
perspicillata, ?PDBFF, CO?, E. M. S., 06.02.98; INPA 2640, Mimon crenulatum, ?PDBFF, DI?, E. M. S., 18.02.98; INPA 2641,
Lonchophylla thomasi, ?PDBFF, FLO?, E. M. S., 31.05.97; INPA 2642, Carollia perspicillata, ?PDBFF, CO?, E. M. S., 18.06.97; INPA
2643, Thyroptera tricolor, ?PDBFF, DI?, E. M. S., 24.09.97; INPA 2644, Carollia perspicillata, ?PDBFF, PA?, E. M. S., 02.12.97; INPA
2645, Carollia brevicauda, ?PDBFF, km 41?, E. M. S., 12.04.98; INPA 2646, Artibeus jamaicensis, ?PDBFF, km 41?, E. M. S., 13.04.98;
INPA 2647, Carollia perspicillata, ?PDBFF, PA?, E. M. S., 14.04.98; INPA 2648, Saccopteryx cf. canescens, ?PDBFF, DI?, E. M. S.,
16.04.98; INPA 2649, Uroderma cf. magnirostrum, ?PDBFF, DI?, E. M. S., 16.04.98; INPA 2650, Cormura brevirostris, ?PDBFF, km 41?,
E. M. S., 12.04.98; INPA 2651, Glyphonycteris sylvestris, ?PDBFF, CO?, E. M. S., 05.05.98; INPA 2652, Tonatia saurophila, ?PDBFF, km
41?, E. M. S., 24.05.98; INPA 2653, Centronycteris maximiliani, ?PDBFF, km 41?, E. M. S., 25.05.98; INPA 2654, Artibeus jamaicensis,
?PDBFF, km 41?, E. B., 31.08.96; INPA 2655, Cormura brevirostris, ?PDBFF, km 41?, E. B., 03.09.96; INPA 2656, Koopmania concolor,
?PDBFF, km 41?, E. B., 05.09.96; INPA 2657, Myotis sp., ?PDBFF, GA?, E. B., 04.10.96; INPA 2658, Molossops greenhalli, ?PDBFF, GA?,
E. B., 01.09.96; INPA 2659, Eptesicus andinus, ?PDBFF, km 41?, E. B., 03.09.96; INPA 2688, Artibeus cinereus, ?PDBFF, DI?, B. R.-H,
31.08.98; INPA 2691, Carollia perspicillata, ?PDBFF, FLO?, B. R.-H, 23.04.98; INPA 2692, Platyrrhinus helleri, ?PDBFF, DI?, B. R.-H,
04.08.98; INPA 2693, Cormura brevirostris, ?PDBFF, DI?, B. R.-H, 05.08.98; INPA 2694, Rhynchonycteris naso, Silv. Treat., E. M. S.,
29.08.98; INPA 2695, Phylloderma stenops, ?PDBFF, DI?, B. R.-H, 14.09.98; INPA 2696, Choeroniscus cf. minor, ?PDBFF, FLO?, B. R.-
H, 15.09.98; INPA 2697, Lonchophylla thomasi, JAU, B. R.-H, 06.09.98; INPA 2698, Myotis riparius, JAU, B. R.-H, 07.09.98; INPA 2699,
Cormura brevirostris, JAU, B. R.-H, 07.09.98; INPA 2700, Trachops cirrhosus, JAU, B. R.-H, 07.09.98; INPA 2701, Vampyrum spectrum,
?PDBFF, CF?, B. R.-H, 24.09.98; INPA 2702, Trachops cirrhosus, ?PDBFF, CF?, B. R.-H, 25.09.98; INPA 2703, Thyroptera discifera,
?PDBFF, CF?, B. R.-H, 26.09.98; INPA 2704, Thyroptera discifera, ?PDBFF, CF?, B. R.-H, 26.09.98; INPA 2705, Myotis sp., ?PDBFF, CF?,
B. R.-H, 26.09.98; INPA 2706, Phylloderma stenops, ?PDBFF, CF?, B. R.-H, 26.09.98; INPA 2707, Ametrida centurio, ?PDBFF, PA?, B. R.-
H, 13.10.98; INPA 2708, Vampyressa bidens, ?PDBFF, PA?, B. R.-H, 13.10.98; INPA 2709, Phylloderma stenops, ?PDBFF, CO?, B. R.-H,
16.10.98; INPA 2710, Artibeus cinereus, ?PDBFF, CO?, B. R.-H, 16.10.98; INPA 2711, Carollia brevicauda, ?PDBFF, CO?, B. R.-H,
16.10.98; INPA 2712, Lophostoma brasiliense, ?PDBFF, CO?, B. R.-H, 17.10.98; INPA 2723, Ametrida centurio, ?PDBFF, DI?, B. R.-H,
04.11.98; INPA 2724, Trachops cirrhosus, ?PDBFF, CF?, B. R.-H, 17.11.98; INPA 2725, Trinycteris nicefori, ?PDBFF, CF?, B. R.-H,
18.11.98; INPA 2726, Sturnira lilium, ?PDBFF, PA?, B. R.-H, 15.12.98; INPA 2727, Platyrrhinus helleri, ?PDBFF, PA?, B. R.-H, 15.12.98;
INPA 2728, Desmodus rotundus, ?PDBFF, DI?, B. R.-H, 26.11.98; INPA 2729, Artibeus jamaicensis, ?PDBFF, DI?, B. R.-H, 26.11.98;
INPA 2839, Platyrrhinus helleri, ?PDBFF, CO?, E. B., 17.01.99; INPA 2840, Saccopteryx canescens, ?PDBFF, DI?, E. B., 20.01.99; INPA
2841, Molossus molossus, ?PDBFF, DI?, E. B., 21.01.99; INPA 2842, Molossus molossus, ?PDBFF, DI?, E. B., 21.01.99; INPA 2843,
Artibeus anderseni, ?PDBFF, CO?, E. B., 02.02.99; INPA 2844, Carollia perspicillata, ?PDBFF, GA?, E. B., 03.02.99; INPA 2845,
Glyphonycteris daviesi, ?PDBFF, GA?, E. B., 03.02.99; INPA 2846, Lonchophylla thomasi, ?PDBFF, CF?, E. B., 20.03.99; INPA 2847,
Rhinophylla pumilio, ?PDBFF, CO?, E. B., 09.04.99; INPA 2848, Uroderma bilobatum, ?PDBFF, CO?, E. B., 11.04.99; INPA 2849,
Platyrrhinus cf. helleri, ?PDBFF, CO?, E. B., 11.04.99; INPA 2850, Myotis nigricans, ?PDBFF, PA?, E. B., 19.05.99; INPA 2851, Carollia
cf. castanea, ?PDBFF, CF?, E. B., 05.06.99; INPA 2852, Carollia perspicillata, ?PDBFF, CF?, E. B., 05.06.99; INPA 2853, Artibeus
obscurus, ?PDBFF, CF?, E. B., 05.06.99.
1 Reserves: CF = Cabo Frio reserves, CO = Colosso reserves, DI = Dimona reserves, FLO = Florestal reserve, GA = Gaviao reserve, PA =
Porto Alegre reserves, Pres. Fig = Presidente Figueiredo, Silv. Treat = Silvicultural Treatment.
2 Collector(s): B. R.-H = Bernal Rodriguez C. O. H. = Charles O. Handley, Jr., E. B. = Enrico Bernard, and E. M. S. = Erica Sampaio. Cf. =
to confirm, Myotis sp. = M. riparius o. nigricans.
Bat Species Richness in Amazonia 21
north of the BDFFP. Capture effort at the additional sites
varied from 30 to 100mnh per night. The additional species
that we collected outside the BDFFP areas are listed but they
are not included in the evaluation of the completeness of our
samples.
Project 2
Ground and high nets (height 17 to 30m) were set inside 17
small tree-fall gaps in undisturbed forest (Gavi?o, km 41;
Fig.1). Additionally, ground-level mistnets were placed in the
forest near gaps. Sampling took place for 12 hours per night.
The mistnets were opened at 18:00 and checked at intervals
of 20 to 45 minutes. Each gap was sampled five times at 50
day intervals during the study period (August 1996?August
1997). Because the number of mistnets used in each gap dif-
fered, total capture effort per gap varied between 114?322
mnh. Total capture effort of project 2 encompassed 3398.5
mnh (Bernard, 2001; Bernard & Fenton, 2002).
Species accumulation curve
To qualitatively assess the completeness of our data, we
pooled the data from both projects and plotted the number 
of newly recorded species against the number of individuals
captured (Fig. 2a). We randomized the data of the species
accumulation curves to minimize the effect of order in which
the nights and hence the species captured accumulated on 
the x-axis (see Colwell & Coddington, 1996). For the number
of sampled individuals (x-axis) we excluded all subsequent
recaptures at a given capture site. We applied parametric and
non-parametric estimators to our data (Chao, 1984; Bunge &
Fitzpatrick, 1993; Colwell & Coddington, 1996) to quantify
sampling completeness and to estimate the minimum capture
effort necessary to achieve a satisfactory result of local
species richness (>90% of the expected species).
Guild classification
We classified all bats captured in this study into guilds fol-
lowing Kalko (1997) and Schnitzler and Kalko (1998). Each
guild consists of species foraging in similar ways in similar
habitats for similar foods (Root, 1967).
Complementarity
In order to estimate the number of species common to both
projects and at both levels (ground and canopy) we calcu-
lated the index of complementarity U/Stotal as described in
Colwell and Coddington (1996):
Where U is the number of unshared species or
Uniques = S1 + S2 - 2 * Shared species
and Stotal = S1 + S2 - Shared species
(S1 = number of species in project 1; S2 = number of species
in project 2).
Results
The bat fauna of the BDFFP
Pooling both ground and canopy netting data sets, we docu-
mented six of the nine families of bats expected to occur in
the area of Central Amazonia (Koopman, 1993; Voss &
Emmons, 1996; Emmons & Feer, 1997; Marinho-Filho &
Sazima, 1998). We did not sample members of three fami-
lies (Noctilionidae, Natalidae, and Furipteridae) within the
BDFFP (but see results on species caught outside the
BDFFP). In project 1 (forest fragmentation) more than 6700
bats were caught in about 26,500mnh (Table 2). We obtained
61 species with confirmed identifications representing 39
genera and six families. The pooled capture rate per sample
area averaged 0.16bats/mnh. The most common species in
the forest fragments was Carollia perspicillata (52% of all
captures), followed by Rhinophylla pumilio (9%), and Art-
ibeus obscurus (7.6%). This pattern differed for the plots in
continuous forest. There, Rhinophylla pumilio was the most
 
 
number of individuals 
nu
m
be
r o
f s
pe
ci
es
 
 2000 400 600 800
0
10
20
30
40
50
60
70
80
number of individuals 
(a)
(b)
nu
m
be
r o
f s
pe
ci
es
 
2000 400 600 800
0
10
20
30
40
50
60
70
80
 
Fig. 2. Species accumulation curves (randomized ? 100) of the
bats sampled with mistnets at the BDFFP project near Manaus,
Brazil. (a) Data of both projects pooled; (b) data presented sepa-
rately for canopy- (C) and ground-level (G) mistnets.
22 E.M. Sampaio et al.
common species with 10% of total captures followed by
Pteronotus parnellii (5%). We achieved a recapture rate of
about 8?10% for most areas with the exception of the plots
in continuous forest (Cabo Frio, Florestal) with lower recap-
ture rates (2.5% and 4.9%) (Table 2).
Project 2 (vertical stratification) accumulated a total 
of 936 individuals in about 3400mnh representing 52 
species in 32 genera and six families (Table 2). Of those, 
37 species were recorded in the canopy nets. As in the
ground-nets set in the forest fragments, three frugivorous
species made up more than half (57.9%) of all captures. Car-
ollia perspicillata had the highest capture rate followed by
Artibeus (Koopmania) concolor, and Rhinophylla pumilio.
Pooling the data from project 1 and 2 revealed that five
species (Micronycteris schmidtorum, Chiroderma villosum,
Vampyressa brocki, Diaemus youngi, and Molossops
(Cynomops) abrasus were exclusively captured in canopy
nets (Table 3). Each of these species was represented only by
one individual.
Overall, nets in the canopy showed a higher average
capture rate per sampling effort (0.36bats/mnh) than the
ground-level mistnets in project 2 (0.26bats/mnh) and in
project 1 (0.16bats/mnh). Recapture rates in the canopy were
much lower than in the groundnets of project 1 and 2, ranging
from 0.7 to 1.05bats/mnh (Table 2).
Species recorded outside the BDFFP area
At the silvicultural research site (ZF2), our brief netting 
sessions yielded two Furipterus horrens (Furipteridae) and
one Rhynchonycteris naso (Emballonuridae). In Presidente
Figueiredo, we caught Peropteryx kappleri and P. macrotis
(Emballonuridae) and the rare Phyllostomus latifolius (Phyl-
lostominae). The two specimens of P. latifolius are clearly
distinct in forearm length and cranial measurements from the
similar-looking, syntopic P. elongatus.
Species accumulation curve
We sampled a total of 67 species with standardized ground
and canopy mistnets. This corresponds to more than 90% of
the number of species (63?78) predicted by most of the
selected species-richness estimators (Table 4). We therefore
consider our sample as a satisfactory approximation of the
total number of species. Based on our species accumulation
curve (Fig. 2a), we conclude that about 2000 captures of 
individuals are needed to reach the 75% mark, and that more
than 6700 captures of individuals are required to reach about
95% of the expected number of species. Adding the five
species to our list that we captured outside the BDFFP area
brings the total number of species up to 72.
Table 2. Summary of both projects including ground- and canopy-level mistnetting. Mnh = mistnet hour (one 12-m mistnet open for one
hour).
Nights Families Genera Species
Localities (N) Mnh No. of ind. No. of recaptures (N) (N) (N)
Project 1 (fragmentation)
Reserve km 41 39 4,011 507 41 (8%) 5 24 43
Florestal 28 3,876 991 49 (4.9%) 5 25 38
Cabo Frio 18 2,583 473 12 (2.5%) 5 20 32
Dimona 31 3,613 572 41 (7.1%) 5 20 31
Porto Alegre 39 3,696 1013 100 (9.8%) 4 23 37
Colosso 35 3,822 1037 91 (8.7%) 4 24 44
Forest edges 22 3,511 516 60 (11.6%) 5 20 28
Secondary growth 38 1,419 1671 133 (8%) 5 24 44
Total 250 26,531 6780 527 (8.2%) 6 34 61
Project 2 (vertical stratification)
Reserve km 41 34 1,738.50 271 2 (0.7%) 5 22 33
Gavi?o 30 1,680.50 665 7 (1.05%) 6 32 48
Total 64 3,419.00 936 9 (0.9%) 6 32 52
Ground vs. canopy mistnets
Ground mistnets 294 28,875 7333 6 38 62
Canopy mistnets 20 1,075 383 5 24 36
Pooled data 314 29,950 7716 6 40 67
Bat Species Richness in Amazonia 23
Table 3. Species list of bats captured with ground and canopy mistnets between January 1996 and July 1999 in the area of the BDFFP near
Manaus, Brazil. See Table 4 for description of feeding guilds.
Project 1 Project 2
Ground Ground Canopy New Conservation
mistnets mistnets mistnets occurrence status Guild
Emballonuridae
Centronycteris maximiliani x ? x * (-) II
Cormura brevirostris x x x (-) II
Peropteryx leucoptera x ? ? (-) II
Saccopteryx bilineata x x x (-) I
Saccopteryx canescens x ? x (-) II
Saccopteryx leptura x ? x (-) II
Mormoopidae
Pteronotus gymnonotus x ? ? (*) (-) II
Pteronotus parnellii x x ? (-) III
Phyllostomidae-Phyllostominae
Chrotopterus auritus x x ? (?) V
Glyphonycteris daviesi x ? ? * (?) IV
Glyphonycteris sylvestris x x ? (?) IV
Lampronycteris brachyotis x ? ? (-) IV
Lophostoma brasiliense x x ? (-) IV
Lophostoma carrikeri x ? ? (?) IV
Lophostoma schulzi x x ? (?) IV
Lophostoma silvicola x x x (-) IV
Micronycteris hirsuta x x ? (-) IV
Micronycteris megalotis x x ? (-) IV
Micronycteris microtis x (-) IV
Micronycteris minuta x ? ? (?) IV
Micronycteris schmidtorum ? ? x (-) IV
Mimon crenulatum x x x (-) IV
Phylloderma stenops x x x (?) VIIIb
Phyllostomus discolor x ? x (-) X
Phyllostomus elongatus x x ? (-) X
Phyllostomus hastatus x ? x (-) X
Tonatia saurophila x x x (-) IV
Trachops cirrhosus x x ? (-) V
Trinycteris nicefori x x ? (-) IV
Vampyrum spectrum x ? ? (+) V
Glossophaginae
Anoura caudifer x ? ? (-) IX
Choeroniscus minor x x x (-) IX
Glossophaga soricina x ? x (-) IX
Lonchophyllinae
Lionycteris spurrelli x x (?) IX
Lonchopylla thomasi x x ? (?) IX
Carollinae
Carollia brevicauda x x ? (-) VIIIb
Carollia perspicillata x x x (-) VIIIb
Rhinophylla pumilio x x x (-) VIIIb
Stenodermatinae
Ametrida centurio x x x (-) VIIIb
Artibeus cinereus x x x (-) VIIIa
Artibeus (Koopmania) concolor x x x (-) VIIIa
Artibeus gnomus x x x (-) VIIIa
Artibeus jamaicensis x x x (-) VIIIa
Artibeus lituratus x x x (-) VIIIa
24 E.M. Sampaio et al.
Complementarity of ground and canopy
mistnetting
The estimated completeness of the bat fauna was lower for
the canopy mistnets (between 73 and 85%) than for the
ground-level mistnets, which actually exceeded the respective
estimates (>100%; Table 4). The complementarity of species
captured between both projects was 0.29. Project 1 yielded 17
unique species in contrast to five species that were taken only
in the canopy nets. Complementarity between both netting
placements (ground vs. canopy) was 0.51 with 31 species that
were exclusively captured in ground-level mistnets (Table 3).
Because the collecting effort in project 1 (forest frag-
mentation) was much higher than in project 2 (vertical 
Table 3. Continued
Project 1 Project 2
Ground Ground Canopy New Conservation
mistnets mistnets mistnets occurrence status Guild
Artibeus obscurus x x x (-) VIIIa
Chiroderma trinitatum x x x (-) VIIIa
Chiroderma villosum ? ? x (-) VIIIa
Ectophylla macconnelli x x x (-) VIIIb
Platyrrhinus helleri x ? ? (-) VIIIa
Sturnira lilium x x x (-) VIIIb
Sturnira tildae x x x (-) VIIIb
Uroderma bilobatum x ? x (-) VIIIb
Vampyressa bidens x ? x (-) VIIIa
Vampyressa brocki x
Desmodontinae
Desmodus rotundus x x ? (-) VI
Diaemus youngi ? ? x (-) VI
Thyropteridae
Thyroptera discifera x ? ? * (-) II
Thyroptera tricolor x x ? (-) II
Vespertilionidae
Eptesicus chiriquinus x x x (-) II
Lasiurus castaneus (x) ? ? * (-) II
Myotis albescens x ? ? (-) II
Myotis nigricans x ? ? (-) II
Myotis riparius x x ? (-) II
Molossidae
Molossops (Cynomops) abrasus ? ? x (-) I
Molossops (Cynomops) greenhalli x ? x (-) I
Molossus bondae x ? ? (-) I
Molossus molossus x ? ? (-) I
37 36
Total number of species 67
Project 1 61
Project 2 52
Ground mistnetting 62
Canopy mistnetting 36
Complementarity indices
Project 1 ? Project 2 0.29
Ground ? canopy mistnetting 0.51
* = Species that have not been previously reported in the literature for the Manaus area; x = captured and identified species; (x) = prelimi-
nary identification; * new occurrence for the area with voucher specimen. All voucher specimens are deposited and listed at INPA, Manaus
(see Table 1). (*) = New occurrence without voucher specimen. Conservation status (-) = stable; (?) = potentially vulnerable; (+) = vulner-
able (Aguiar & Taddei, 1995; Wilson, 1996). Complementarity indices follow Colwell and Coddington (1996).
Bat Species Richness in Amazonia 25
stratification), both on a nightly basis and as a whole, direct
comparison of species composition and patterns of relative
abundance between ground level and canopy strata is diffi-
cult. Nevertheless, we observed different trends in the pref-
erence of either canopy or ground level between and within
guilds. Gleaning insectivorous bats, both in number of
species and individuals, as well as Pteronotus parnellii
(highly cluttered space aerial insectivore), were mainly
caught in ground-level mistnets (Table 3). Of the two
members of the blood-feeding guild, Desmodus rotundus was
sampled exclusively at ground level and Diaemus youngi
only in the canopy.
Discussion
Species accumulation curves
Although the use of species accumulation curves in inven-
tory studies of bats was initiated only recently, we consider
it an essential tool to assess the quality of inventory data (see
Kalko et al., 1996a; Longino & Colwell, 1997; Moreno &
Halffter, 2000). The shape of the curves indicates how well
a local bat fauna has been sampled. In case the curve is
nearing an asymptote, most species that can be adequately
sampled with a certain method have been recorded. Overall,
our samples confirm the high numbers of New World 
leaf-nosed bats (Phyllostomidae) and Pteronotus parnellii
(Mormoopidae) that are dominant components of bats in
Neotropical lowland forests. Both are well-sampled with
mistnets (e.g., Fenton et al., 1992; Kalko, 1997, 1998).
Because our curve closely approached the plateau of the
expected number of species as determined by species-rich-
ness estimators (Colwell & Coddington, 1996), we conclude
that we captured most of the local bat fauna (Fig. 2a).
Based on pooled data from various local inventories in the
Amazon Basin, the expected total number of species for the
Amazon Basin is about 117 on a regional scale (Marinho-
Filho & Sazima, 1998). Up to now none of the mistnetting
bat inventories reported for this area (e.g., Mok et al., 1982;
dos Reis & Peracchi, 1987; Hutterer et al., 1995; Voss &
Table 4. Results of selected species richness estimators for the expected number of species sampled during both projects (N = 7716 indi-
viduals) at the BDFFP in ground (N = 62 species) and canopy (N = 36 species) mistnets. SD = standard deviation; ACE = Incidence Cover-
age Estimator; Jack1 = Jacknife 1 estimator; MMMean = Michaelis?Menten Mean estimator.
Pooled data Ground Canopy
Estimators* all mistnets SD % complete mistnets SD % complete mistnets SD % complete
ACE 72.7 0.4 93 56.3 0.2 110 45.1 2.5 78
Chao1 70 4 97 56.0 0.3 110 41 4.2 85
Jack1 79 4.5 86 60 2.4 103 48.2 2.9 73
Bootstrap 72 94 58.4 106 41.8 83
MMMean 63.2 >100 55.4 111 42 83
*Estimates: http://viceroy.eeb.uconn.edu/estimates, R. K. Colwell (1999).
Table 5. Guild structure of the bat fauna sampled in the BDFFP area based on ground and canopy mistnetting.
Guilds Expected Observed New occurrences
I Uncluttered space aerial insectivores 23 5
II Background cluttered space aerial insectivores* 24 13 4
III Highly cluttered space aerial insectivores 1 1
IV Highly cluttered space gleaning insectivores 17 15 4
V Highly cluttered space gleaning carnivores 3 3
VI Highly cluttered space gleaning piscivores 1 0
VII Highly cluttered space gleaning sanguivores 3 2
VIIIa Highly cluttered space gleaning canopy frugivores 20 12 1
VIIIb Highly cluttered space gleaning shrub frugivores 10 9
IX Highly cluttered space gleaning nectarivores 11 5
X Highly cluttered space gleaning omnivores 4 3 1
Total 117 67 10
The classification follows the bat guilds proposed by Kalko (1997) and Schnitzler and Kalko (1998). Natalus stramineus is included in guild
II because of its foraging behaviour close to vegetation where it captures insects in flight (Kalko, pers. obs.). The expected number of species
is based on a list for bats proposed by Marinho-Filho and Sazima (1998) for the Brazilian Amazon region. The new occurrences are included
in the total number of observed species.
26 E.M. Sampaio et al.
Emmons, 1996; Marinho-Filho & Sazima, 1998), including
our results, have yet reached or have come close to this
number of species. Although the results of the studies are
based on well-sampled communities, some species are still
underrepresented or lacking, particularly aerial insectivores.
This bias is inherent to the exclusive use of mistnets (Voss
& Emmons, 1996; Kalko, 1998; Simmons & Voss, 1998)
because aerial insectivorous bats forage mainly in spaces that
are difficult or impossible to sample with mistnets. Addi-
tional investigations are needed to reduce this bias such as
the identification of aerial insectivores by their echoloca-
tion calls (e.g., Kalko, 1998; Kuenzi & Morrison, 1998;
O?Farrell & Gannon, 1999; O?Farrell & Miller, 1999).
Comparison of canopy- and ground-level mistnets
The species accumulation curves of both projects give a first
impression of the complementarity between the placement of
the mistnets at ground and at canopy level (Longino &
Colwell, 1997; Kalko & Handley, 2001; Fig. 2b). Because
both projects were located in the same area, comparison of
project 1 and 2 showed rather low overall species comple-
mentarity (<30%) that is high similarity of species composi-
tion whereas the comparison between ground and canopy
mistnetting revealed higher complementarity (>50%), in
other words less similarity. One reason for the lower simi-
larity between canopy and ground mistnets is unequal
capture effort. Capture effort was almost 30 times higher for
the ground-level nets (projects 1 and 2 combined) than for
the canopy nets (28,000 vs. 1000mnh). This leads to a higher
probability to capture more species at ground-level than in
the canopy.
Another reason is the differential use of space. Our studies
confirm trends in vertical stratification found in Neotropical
bats (Handley, 1967; Simmons & Voss, 1998; Kalko &
Handley, 2001; Lim & Engstrom, 2001). However, although
some bats do show differential use of forest strata, vertical
stratification may not be as strict in bats as it is in other
species such as birds (see Kalko & Handley, 2001; Bernard,
2001). For instance, most forest bats are not limited to a par-
ticular forest level even though some species might forage
more frequently in particular strata. Kalko & Handley (2001)
argue that the classification of bats into ?ground? or ?canopy?
bats is still premature due to low sample size and insufficient
information on the diet, behavior, and sensorial flexibility of
many species.
In our study, only a few species were captured exclusively
in canopy nets, as, for example, one individual each of
Diaemus youngi and one of the ?mostly fig-eating? bat,
Vampyressa brocki. Although D. youngi seems to be rarely
collected with other vampire species (Hutterer et al., 1995),
the absence or low capture rates of this species might to some
degree be due to a capture bias because of fewer canopy nets
compared with the ground nets. Diaemus youngi is a bird-
blood specialists and probably feeds mainly on birds roost-
ing in the canopy (Uieda, 1992). It was also captured
exclusively in canopy nets in a study at Bel?m (Kalko &
Handley, 2001). Frugivorous bats that are known to feed
mainly on canopy trees such as Artibeus concolor (Kalko &
Handley, 2001) were well represented in our canopy mist-
nets. Comparing both mistnetting levels, we observed more
species of aerial insectivorous bats in our canopy samples
which is the stratum frequently used by these bats as forag-
ing areas. Overall, the combination of canopy- and ground-
level mistnets produced a higher number of species per
mistnet hour than either placement alone (Fig. 2a, b).
Comparisons with other inventories
Overall we documented 72 species of 43 genera from 117
species predicted for the greater Amazon region (Marinho-
Filho & Sazima, 1998; Table 3). We recorded 10 species new
for the area, but already known from other locations in the
Amazon Basin (dos Reis & Schubart, 1979; Mok et al., 1982;
dos Reis & Guillaumet, 1983; dos Reis, 1984; dos Reis 
& Peracchi, 1987; Emmons & Feer, 1997; Marinho-Filho 
& Sazima, 1998). These species include Centronycteris 
maximiliani, Glyphonycteris daviesi, Micronycteris hirsuta,
Micronycteris microtis, Phyllostomus latifolius, Trinycteris
nicefori, Vampyressa brocki, Lasiurus cf. castaneus, Thy-
roptera discifera, and Pteronotus gymnonotus. The results
compare well with other studies in the Amazon region and
in the adjacent Guyanas, all of which document an impres-
sively high species richness of bats. Currently, the highest
numbers of species recorded locally in those areas encom-
pass 86 species at Iwokrama Forest in Central Guyana (Lim
& Engstrom, 2001), 78 species at Paracou, French Guyana
(Simmons & Voss, 1998) and 72 species at Alt?r do Ch?o,
Central Amazonia, Brazil (Bernard & Fenton, 2002).
None of the inventories, however, is close to completeness.
This applies also for our study as revealed by a comparison
of the number of recorded species with the number of species
based on information on distribution patterns in the current
literature (e.g., dos Reis & Schubart, 1979; Mok et al., 1982;
dos Reis & Guillaumet, 1983; dos Reis, 1984; dos Reis &
Peracchi, 1987;  Emmons & Feer, 1997; Marinho-Filho &
Sazima, 1998; Bernard & Fenton, 2002). Interestingly, the
proportional composition of species in each family is 
similar to earlier inventories in the Amazon Basin and the
Guyanas. Probably, the expected number of species is too
high because it includes all species recorded in previous
inventories, which were scattered over a huge area. Further-
more, some of the species may not occur in the area of the
BDFFP because they are patchily distributed in the Amazon
Basin or do not find the appropriate ecological require-
ments there such as rocky outcrops for cave-roosting bats.
With the capture of approximately 3000 individuals we
reached about 90% of the expected species richness. This
compares well to the Paracou Project in French Guyana
(Simmons & Voss, 1998) where over 1500 bats bats were
sampled with ground-level mistnets in about 2500mnh
(24,957 net-meter-hours (nmh)/12m mistnets) resulting in a
Bat Species Richness in Amazonia 27
total of 78 species with 65 species sampled in ground-level
and 39 species in canopy nets. Lim and Engstrom (2001)
caught 2117 bats representing 73 species in Guyana with a
capture effort of about 1660mnh (495,136m2/h; one 12-m
net approximates 300m2). Bernard & Fenton (2002) docu-
mented 72 species of bats with a capture effort of 6116mnh
over 102 nights with more than 3900 individuals at a highly
diverse savanna-forest site in Brazil (Alt?r do Ch?o, Par?).
In Mexico, Moreno and Halffter (2000) netted 20 nights 
(3853nmh, or about 320mnh) in different habitat types to
obtain about 90% of the expected fauna with ground-level
mistnets. One of the long-term bat projects on Barro Col-
orado Island (BCI) in Panam? (Kalko et al., 1996a) spanned
more than 10 years and included more than 48,000 cap-
tures/recaptures. An average of 30 net nights was needed to
compile 30 species. After additional years of sampling
including other methods (canopy netting, acoustic monitor-
ing, roost search) the number of species recorded for BCI has
recently reached 72 (E. Kalko, pers. comm.), which is similar
to the species richness in our study when we include our 
captures of areas adjacent to the BDFFP.
In all studies, a large proportion of the bat fauna has
already been recorded during the first few samples. Most of
the less common species accumulated gradually in the course
of the study. After capturing about 70% of the total number
of species for a given locality, the increase in species accu-
mulation usually slows down drastically. Even though in
most Neotropical bat inventories a relatively large number of
species (between 70 to 80% of the total number of species
ultimately recorded in the respective projects) are captured
within 30 nights or in about 1000mnh or with a capture effort
of about 1000 individuals, an intense and committed capture
effort is necessary to go beyond the more common species
and to also include rare species.
Unsampled species
In our study, we caught only nine out of the 14 expected
emballonurids, six out of the nine expected vespertilionids
and four out of the 16 expected molossids. The most remark-
able captures of aerial insectivores were the rare Centronyc-
teris maximiliani and Lasiurus cf. castaneus, resulting in
substantial range extensions for both species (e. g., Masson
& Cosson, 1992; Simmons & Handley, 1998). Based on our
extensive echolocation call recordings of aerial insectivorous
bats in the BDFFP area and subsequent species identification
we expect an additional 10?15 species bringing the total
number of bats in the area to 82?87 species. Details about
the results of the echolocation call analysis will be published
elsewhere.
In our mistnetting study at the BDFFP, we missed several
families and species which are considered to be widespread
in the Central Amazon Basin (dos Reis & Schubart, 1979;
Mok et al., 1982; dos Reis & Perracchi, 1987; Koopman,
1993; Emmons & Feer, 1997), in particular Noctilionidae,
and Natalidae. The absence of noctilionids and Macrophyl-
lum macrophyllum (Phyllostominae) in our mistnet data may
be correlated with the almost complete absence of large
rivers and other large water bodies in our study area. Noc-
tilio leporinus is well known for its fish-eating habits and N.
albiventris mainly for its insect-eating habits (e.g., Hooper
& Brown, 1968; Hood & Pitocchelli, 1983). Both species of
Noctilio forage predominantly over water surfaces (e.g.,
Hood & Jones, 1984; Schnitzler et al., 1994; Kalko et al.,
1998). We recorded both species frequently and in large
numbers over the Amazon as well as the Branco and Madeira
river. Macrophyllum macrophyllum is the only phyllostomid
known to forage regularly over water. This is reflected in 
its diet that includes, among other foods, water striders
(Hemiptera, Gerridae; Gardner, 1977). Adding those species
to our list in addition to the species identified by echoloca-
tion brings the total number of bats to 85?90 species.
Some bat species prefer or require caves as roost sites.
Hence, the lack of caves in the BDFFP area may explain the
absence of Natalus stramineus (Natalidae). The absence of
moormopids such as P. personatus which is also known to
roost predominantly in caves (Kalko et al., 1996a; Voss &
Emmons, 1996), might also be in part due to sampling bias.
Pteronotus personatus forages mostly at forest edges and in
gaps (Table 4), and is usually not well sampled by ground
mistnets. However, we recorded the characteristic echoloca-
tion calls of this species frequently at the BDFFP. Inter-
estingly, its congener, Pteronotus parnellii is the only
mormoopid that is well sampled with mistnets (Kalko et al.,
1996a; Bernard et al., 2001; Bernard & Fenton, 2002). This
is probably mainly linked to its foraging habits because it 
frequently forages along man-made trails inside the forest, 
as documented by ultrasound monitoring. At Presidente
Figueiredo, where we netted bats at cave entrances, the
capture effort was too low (two nights) to permit compre-
hensive conclusions about the bat fauna roosting in those
caves. The species we caught included Pteronotus parnellii,
Phyllostomus latifolius, and Carollia perspicillata.
Although frugivorous bats are well represented in our
samples (Table 4), our inventory did not record some species
of the genera Platyrrhinus, and Vampyressa, and the nec-
tarivorous bats Lichonycteris degener and Scleronycteris ega
that are expected to occur in the area (dos Reis & Peracchi,
1987; Kalko et al., 1996a; Voss & Emmons, 1996; Moreno
& Halffter, 2000). The distribution and abundance of frugiv-
orous and nectarivorous bats is tightly associated with the
temporal and spatial availability of certain fruits and flowers
(Fleming, 1986a; Marinho-Filho, 1991). In particular,
species such as figs (Moraceae, Ficus spp.) have keystone
functions for frugivorous bats and other frugivores in part
because of their year-long fruiting within populations and the
asynchronous fruiting of individuals (Handley et al., 1991;
Kalko et al., 1996b; Korine et al., 2000). In contrast to 
the BDFFP area, frugivorous bats of the genera Artibeus,
Platyrrhinus, and Vampyressa are common on BCI, Panam?,
and other localities where figs are abundant (Handley et al.,
1991). This discrepancy may be mainly due to the nutrient-
28 E.M. Sampaio et al.
poor latosols of Manaus region (Chauvel, 1983) that produce
a distinctive local flora where poor-soil specialist trees pre-
dominate. Many bat-dispersed fruits, such as Piper and Ficus
require richer soils (Gentry, 1990). The low abundance of fig
trees in the Manaus area is likely to be one of the main
reasons for our low capture rates or absence of canopy fru-
givores considered to be mainly fig-eaters (e.g. Artibeus
jamaicensis, A. lituratus, Vampyrodes carracioli, Vampyressa
pusilla, Platyrrhinus brachycephalus, and P. infuscus; see
Handley et al., 1991; Galetti & Morellato, 1994; Kalko et al.,
1996b; Bernhard & Fenton, 2002).
The most common species we encountered in our mist-
netting samples were ground story frugivores equivalent to
the ?highly-cluttered-space ground-level frugivorous bats?
described by Kalko (1997) and Schnitzler and Kalko (1998).
For instance, Carollia perspicillata was the most abundant
species in this guild in the BDFFP area and contributed to
more than half (52%) of all captures. In contrast to C. per-
spicillata and C. brevicauda, which are both widespread,
common, and often occur sympatrically in the Amazon
Basin, C. castanea, was not represented in our samples. To
date, the nearest capture of C. castanea has been in Acre, Rio
Branco, in the western Brazilian Amazon (Uieda, 1980). In
Panam?, Carollia castanea and C. perspicillata overlap in
habitat use and certain dietary resources (Thies et al., 1998).
Both species eat Piper fruits but C. castanea is clearly a
Piper specialist and may be thus strongly limited by the avail-
ability of these plants. Carollia perspicillata and C. brevi-
cauda have broader food preferences and fecal analysis
indicates that in the BDFFP both species feed almost exclu-
sively on fruits of Vismia sp., a common secondary shrub
(Sampaio, pers. obs.). The dominance of Carollia in the
BDFFP area conforms well to similar results of bat assem-
blages in Alt?r do Ch?o, Central Brazil (Bernhard & Fenton,
2001) and in French Guyana (Simmons & Voss, 1998).
We also did not find Rhinophylla fischerae, which has
been recorded from southern Venezuela (Gardner, 1988), and
was described from the upper Amazon (Ucayali) of Per?
(Carter, 1966), and is expected to occur elsewhere in the
Amazon Basin (Marinho-Filho & Sazima, 1998). It is rela-
tively common in Alt?r do Ch?o, along the Tapaj?s river
(Bernard et al., 2001) but appears to be rare in most bat
surveys conducted thus far. Probably, essential ecological
requirements of this species are not met in our study area.
Ecological limitations may also play a role in the rareness of
other phyllostomids in our samples where we captured less
than five individuals in 13 species.
Diversity and conservation of Amazonian lowland bats
Our Central Amazonian bat inventory reveals a high number
of coexisting species. If we add to our current data base about
15 species of aerial insectivores which we missed in our mist-
netting inventory but whose presence has been confirmed by
ultrasound recordings (Sampaio & Kalko, unpublished), and
four bat species that are considered to be widespread in 
Amazonia but that were also absent in our samples, leads to
a total of 91 species. This result compares favorably with the
high bat diversity documented for other Neotropical locali-
ties (Handley, 1967; McNab, 1971; Mok et al., 1982; dos
Reis, 1984; Fleming, 1986b; Willig, 1986; dos Reis & Per-
acchi, 1987; Brosset & Charles-Dominique, 1990; Findley,
1993; Voss & Emmons, 1996; Kalko & Handley, 2001;
Bernard & Fenton, 2002). However, none of the inventories
is close to completeness and in most cases we do not fully
understand why certain species are absent. We recognize the
urgency in acquiring a better understanding of the biology
and demographics of Amazon bat assemblages. This knowl-
edge is necessary for developing strategies for conserving
habitats and their associated fauna. For instance, we captured
nine species in our study that are considered as ?potentially
endangered? in their range and one species (Vampyrum spec-
trum) that is already included in the official Brazilian list of
threatened bat species (Aguiar & Taddei, 1995; Wilson,
1996). The reason for the ?rarity? of V. spectrum may be
related to its trophic position as a top predator.
Aguiar and Taddei (1995) used rarity of species and the
degree of endemism of species to define their vulnerability
to habitat changes such as fragmentation and isolation. 
When is a species considered to be rare? Are species truly
rare or simply difficult to be captured or recorded? Answers
to these questions will help us to determine which 
species become rare due to habitat changes that limit or
prevent the natural maintenance of their populations.
Methodological biases in our sampling techniques may result
in the characterization of a species as either ?rare? or
?common? when the opposite may be true. For example,
species that are ?rare? or ?absent? in mistnet surveys, (e.g.,
Diclidurus spp. or Peropteryx spp.) are frequently recorded
in ultrasound monitoring.
Furthermore, designating bat species as endangered or
threatened requires a thorough understanding of their natural
history, especially of the underlying ecological factors deter-
mining population density and size (Brussard, 1991). This
includes for instance lack of adaptability to habitat changes,
low dispersal capacities, and dietary spezializations, forag-
ing habits, and roosting behavior. Furthermore, it is arguable
that rare species are always more prone to extinction than
more abundant and widespread species (Terborgh & Winter,
1980; McIntyire, 1992).
Distribution and occurrence data of many taxa, in par-
ticular bats, are still meager and are often difficult to 
interpret and evaluate due to the lack of standardized
methodology. Consequently, data of species richness and rel-
ative abundance in inventories for many species are still
insufficient to be taken as indicators of rarity and vulnera-
bility to extinction. An adequate understanding of the role of
bats as indicator species and as providers of critical ecosys-
tem services (e.g., pollinators, seed dispersal agents, control
of insect populations), requires considerably more study of
the community organization and ecology of bats across a
range of different sites and habits than is available to date.
Bat Species Richness in Amazonia 29
Acknowledgements
We thank J.R.M de Oliveira (Ribamar), J. Tenasol (Z?), F. M.
Bezerra (Flecha), S.S. de Souza (Seba) and A.M. dos Reis
(Leo) for assistance with field work and the BDFF Project
for facilities as well as financial support. We are most grate-
ful to H.-U. Schnitzler and I. Kaipf (Univ. of T?bingen,
Germany) for support and technical advice and to F. Fahr
(Univ. Ulm, Germany), D. Menne (Univ. T?bingen,
Germany), R. Stallard (U.S. Geological Survey, Boulder), M.
Tschapka (Univ. Ulm, Germany), R. Voss (AMNH, USA), A.
Zillikens (Univ. T?bingen, Germany) and an anonymous
referee for critical feedback. We thank Al Gardner (NMNH,
Washington, D.C.) for additional identifications of bats. E.
Sampaio was supported by a Ph.D. scholarship (CAPES) and
by the World Wildlife Fund (WWF) and E. Bernhard by 
the Conselho Nacional de Desenvolvimento Cinet?fico e 
Tecnol?gico (CNPq-Masters Program) and Bat Conservation
International. Our work was also funded by trust funds of the
Smithsonian Institution and grants from the BDFFP and the
Deutsche Forschungsgemeinschaft to E. Kalko. This publi-
cation is dedicated to Dr. Charles Handley, Jr., who deceased
in June 2000 before the manuscript was completed. This is
publication number 339 in the BDFFP technical series.
References
Aguiar LMS, Taddei VA (1995): Workshop sobre a conserva??o
dos morcegos Brazileiros. Chiroptera Neotrop 12: 24?30.
Bernard E (2001): Vertical stratification of bat communities in
primary forest of Central Amazon, Brazil. J Trop Ecol 17:
115?126.
Bernard E, Albernaz ALKM, Magnusson WE (2001): Bat
species composition in three localities in the Amazon
Basin. Stud Neotrop Fauna Environm 36: 177?184.
Bernard E, Fenton MB (2002): Species diversity of bats 
(Mammalia: Chiroptera) in forest fragments, primary
forests, and savannas in Central Amazonia, Brazil. Can J
Zool 80: 1124?1140.
Bierregaard RO Jr, Laurance WF, Sites JW Jr, Lynam AJ,
Didham RK, Andersen M, Gascon C, Tocher MD, Smith
AP, Viana VM, Lovejoy TE, Sieving KE, Kramer EA,
Restrepo C, Moritz C (1997): Key priorities for the study
of fragmented tropical ecosystems. In: Laurance WF, 
Bierregaard RO Jr, eds., Tropical Forest Remnants:
Ecology, Management, and Conservation of Fragmented
Communities. Chicago and London, The University of
Chicago Press, pp. 515?527.
Brosset A, Charles-Dominique P (1990): The bats from French
Guyana: A taxonomic, faunistic and ecological approach.
Mammalia 544: 509?560.
Brosset A, Charles-Dominique P, Cockle A, Cosson JF, Masson
D (1996): Bat communities and deforestation in French
Guyana. Can J Zool 74: 1975?1982.
Brussard PF (1991): The role of ecology in biological conser-
vation. Ecol Appl 1: 6?12.
Bunge J, Fitzpatrick M (1993). Estimating the number of
species: A review. J Am Stat Assoc 88: 364?373.
Carter DC (1966): A new species of Rhinophylla (Mammalia:
Chiroptera; Phyllostomatidae) from South America. Proc
Biol Soc Wash 79: 235?238.
Chao A (1984): Non-parametric estimation of the number of
classes in a population. Scand J Stat 11: 265?270.
Chauvel A (1983): Os latossolos amarelos ?licos, argilosos
dentro dos ecossistemas das bacias experimentais do INPA
e da regi?o vizinha. Acta Amaz?nica 12: 47?60.
Colwell RK, Coddington JA (1996): Estimating terrestrial 
biodiversity through extrapolation. In: Hawksworth DL,
ed., Biodiversity: Measurement and Estimation. London,
Chapman & Hall, pp. 101?118.
Cosson JF, Pons JM, Masson D (1999): Effects of forest frag-
mentation on frugivorous and nectarivorous bats in French
Guyana. J Trop Ecol 154: 515?534.
Crespo RF, Linhart SB, Burns RJ, Mitchell GC (1972): Forag-
ing behavior of the common vampire bat related to moon-
light. J Mammal 532: 366?368.
Dale VH, O?Neill RV, Southworth E, Pedlowski M (1994): Mod-
eling the effects of land management in the Brazilian set-
tlement of Rond?nia. Conserv Biol 8: 196?206.
Emmons LH, Feer F (1997): Neotropical Rainforest Mammals:
A Field Guide. Chicago, University of Chicago Press, xvi
+ 307 pp.
Fenton MB, Acharya L, Audet D, Hickey MB, Merriman C,
Obrist MK, Syme DM, Adkins B (1992): Phyllostomid bats
(Chiroptera: Phyllostomidae) as indicators of habitat dis-
ruption in the Neotropics. Biotropica 243: 440?446.
Ferrari SF, Lopes MA (1992) A new species of marmoset, genus
Callithrix Erxleben, 1777 (Callitrichidae, Primates) from
western Brazilian Amazonia. Goeldi Zool 12: 1?13.
Ferreira LV, Laurance WL (1996): Effects of forest fragmenta-
tion on the mortality and damage of selected trees in
Central Amazonia. Conserv Biol 113: 797?801.
Findley JS (1993): Bats: A Community Perspective. Cambridge,
Cambridge University Press, xi + 167 pp.
Fleming TH (1986a): Opportunism versus specialization: the
evolution of feeding strategies in frugivorous bats. In:
Estrada A, Fleming TH, eds. Frugivores and Seed Disper-
sal. Dordrecht, Dr W Junk, pp. 105?118.
Fleming TH (1986b): The structure of Neotropical bat commu-
nities: A preliminary analysis. Rev Chil Hist Nat 59:
135?150.
Fonseca GAB, Herrmann G, Leite YLR, Mittermeier RA,
Rylands AB, Patton JL (1996): Lista anotada dos
mam?feros do Brazil. Occas Pap Conserv Biol 4: 1?38.
Galetti M, Morellato LPC (1994): Diet of the fruit-eating bat
Artibeus lituratus in a forest fragment in Brazil. Mammalia
58: 61?665.
Gardner AL (1977): Feeding habits. In: Baker RJ, Jones JK Jr,
Carter DC, eds., Biology of Bats of the New World Family
Phyllostomatidae, Part II, Special Publs Mus Texas Tech
Univ. 13. Lubbock, pp. 293?350.
Gardner AL (1988): The mammals of the Parque Nacional Ser-
rania de la Neblina. In: Brewer-Carias C, ed., Resultados
30 E.M. Sampaio et al.
de la Expedition 1983?1987, Caracas, Editorial Sucre, pp.
695?765.
Gascon C, Lovejoy TE (1998): Ecological impacts of forest 
fragmentation in Central Amazon. Zool-Anal Complex Sy
101: 273?280.
Gentry AHT (1990): Composition and dynamics of the Cocha
Cashu ?mature? floodplain forest. In: Gentry AH, ed., Four
Neotropical Rainforests. New Haven, Yale University Press,
pp. 542?565.
Gotelli NJ, Graves GR (1996): Null Models in Ecology.
Washington and London, Smithsonian Institution Press, 
pp. 1?368.
Granjon L, Cosson JF, Judas J, Ringuet S (1996): Influence of
tropical rainforest fragmentation on mammal communities
in French Guyana: Short term effects. Acta Oecol 176:
673?684.
Handley CO Jr (1967): Bats of the canopy of an Amazonian
forest. Atas do simposio da Biota Amaz?nica: 211?215.
Handley CO Jr (1988): Specimen preparation. In: Kunz TH, ed.,
Ecological and Behavioral Methods for the Study of Bats.
Washington, DC, Smithsonian Institution Press, pp.
437?457.
Handley CO Jr, Wilson DE, Gardner AL (1991): Demography
and natural history of the common fruit bat, Artibeus
jamaicensis, on Barro Colorado Island, Panam?. Smith
Contr Zool, Washington, DC, Smithsonian Institution
Press, pp. 1?511.
Hood CS, Pitocchelli J (1983): Noctilio albiventris. Mammal Sp
197: 1?5.
Hood CS, Jones JK Jr (1984): Noctilio leporinus. Mammal Sp
216: 1?7.
Hooper ET, Brown JH (1968): Foraging and breeding in two
sympatric species of Neotropical bats, genus Noctilio. J
Mammal 49: 310?312.
Humphrey PS, Bridge D, Lovejoy TE (1968). A technique for
mistnetting in the forest canopy. Bird-banding 39: 43?
50.
Hutterer R, Verhaagh M, Diller J, Podlouck R (1995): An inven-
tory of mammals observed at Panguana Biological Station,
Amazonian Peru. Ecotropica 1: 3?20.
Kalko EKV (1997): Diversity in tropical bats. In: Ulrich H, ed.,
Tropical Biodiversity and Systematics. Proceedings of the
International Symposium on Biodiversity and Systematics
in Tropical Ecosystems, Bonn, 1994. Bonn, Zoologisches
Forschungsinstitut und Museum Alexander Koenig, pp.
13?43.
Kalko EKV (1998): Organization and diversity of tropical bat
communities through space and time. Zoology 101:
281?297.
Kalko EKV, Handley CO Jr (2001): Neotropical bats in the
canopy: Diversity, community strucuture, and implications
for conservation strategies. Plant Ecol 153: 319?333.
Kalko EKV, Handley CO Jr, Handley D (1996a): Organization,
diversity, and long-term dynamics of a Neotropical bat
community. In: Cody M, Smallwood S, eds., Long-term
Studies of Vertebrate Communities. Los Angeles, Academic
Press: pp. 503?551.
Kalko EKV, Herre EA, Handley CO Jr (1996b): Relation of fig
fruit characteristics to fruit-eating bats in the new and old
world tropics. J Biogeogr 23: 565?576.
Kalko EKV, Schnitzler H-U, Kaipf I, Grinnell AD (1998):
Echolocation and foraging behavior of the lesser bulldog
bat, Noctilio albiventris: Preadaptations for piscivorous?
Behav Ecol Sociobiol 42: 305?319.
Koopman KF (1993): Order Chiroptera. In: Wilson DE, Reeder
DM, eds, Mammals Species of the World, a Taxonomic 
and Geographic Reference, 2nd ed. Washington, DC, 
Smithsonian Institution Press, pp. 137?241.
Korine C, Kalko EKV, Herre EA (2000): Fruit characteristics
and factors affecting fruit removal in a Panamanian com-
munity of strangler figs. Oecologia 123: 560?568.
Kuenzi AJ, Morrison ML (1998): Detection of bats by mistnets
and ultrasonic sensors. Wild Soc Bull 26: 307?311.
Laurance WF, Ferreira LV, Rankin-De-Merona JM, Laurance S,
Hutchings RW, Lovejoy TE (1998): Effects of forest frag-
mentation on recruitment patterns in Amazonian tree com-
munities. Conserv Biol 122: 460?464.
Lee TE Jr, Hoofer SR, Van Den Busche RA (2002): Molecular
phylogenetics and taxonomic revision of the genus Tonatia
(Chiroptera: Phyllostomidae). J Mammal 83: 49?57.
Lim BK, Engstrom MD (2001): Species diversity of bats 
(Mammalia: Chiroptera) in Iwokrama Forest, Guyana, and
the Guyanan subregion: Implications for conservation.
Biodiv Conserv 10: 613?657.
Longino JT, Colwell RK (1997): Biodiversity assessment using
structured inventory: Capturing the ant fauna of a tropical
rain forest. Ecol Appl 74: 1263?1277.
Lovejoy TE, Bierregaard RO Jr (1990): Central Amazonian
forests and the minimum critical size of ecosystem Project.
In: Gentry AH, ed., Four Neotropical Rainforests. New
Haven, Yale University Press, pp. 60?71.
Marinho-Filho JS (1991): The coexistence of two frugivorous
bat species and the phenology of their food plants in Brazil.
J Trop Ecol 7: 59?67.
Marinho-Filho JS, Sazima I (1998): Brazilian bats and conser-
vation biology: A first survey. In: Kunz THP, Racey PA,
eds., Bat Biology and Conservation. Washington, DC and
London, Smithsonian Institution Press, pp. 342?353.
Masson D, Cosson JF (1992): Cyttarops alecto (Emballonuri-
dae) et Lasiurus castaneus (Vespertilionidae), deux chi-
ropteres nouveaux pour la Guyane fran?aise. Mammalia
56: 475?478.
McIntyre S (1992): Risks associated with the setting of conser-
vation priorities from rare plant species list. Biol Conserv
60: 31?37.
McNab BK (1971): The structure of tropical bat faunas. Ecology
522: 352?358.
Mittermeier RA, Schwarz M, Ayres JM (1992): A new species
of marmoset, genus Callithrix Erxleben, 1777 (Callitrichi-
dae, Primates) from the Rio Mau?s region, state of Ama-
zonas, Central Brazilian Amaz?nia. Goeldi Zool 14: 1?17.
Mok WY, Wilson DE, Racey LA, Luiz?o RCC (1982): Lista atu-
alizada de quir?pteros da Amaz?nia Brazileira. Acta
Amaz?nica 124: 817?823.
Bat Species Richness in Amazonia 31
Moreno CE, Halffter G (2000): Assessing the efficiency of bio-
diversity inventories using species accumulations curves
for a bat fauna. J Appl Ecol 37: 149?158.
Morrison DW (1978): Lunar phobia in a Neotropical fruit bat
Artibeus jamaicensis. Anim Behav 263: 852?855.
O?Farrell MJ, Gannon WL (1999): A comparison of acoustic
versus capture techniques for the inventory of bats. J
Mammal 80: 24?30.
O?Farrell MJ, Miller BW (1999): A new examination of echolo-
cation calls of some Neotropical bats (Emballonuridae and
Mormoopidae). J Mammal 78: 954?963.
Prance GT (1990): The floristic composition of forests of
Central Amazonian Brazil. In: Gentry AH, ed., Four
Neotropical Rainforests. New Haven, Yale University Press,
pp. 112?141.
Queiroz HL (1992): A new species of capuchin monkey, genus
Cebus Erxleben, 1777 (Cebidae: Primates) from eastern
Brazilian Amazonia. Goeldi Zool 15: 1?13.
Rankin-De-Merona JM, Prance GT, Hutchings RW, Silva MF,
Rodrigues WA, Uehling ME (1992): Preliminary results of
a large-scale tree inventory of lowland rain forest in the
Central Amazon. Acta Amaz?nica 22: 493?534.
dos Reis NR (1984): Estrutura da comunidade de morcegos da
regi?o de Manaus, Amazonas. Rev Bras Biol 443: 247?254.
dos Reis NR, Guillaumet J-L (1983): Les Chauves-souris 
frugivoures de la region de Manaus et leur r?le dans la 
dissemination des especes vegetales. Rev Ecol-Terre Vie 
38: 1983.
dos Reis NR, Peracchi AL (1987): Quir?pteros da regi?o de
Manaus, Amazonas, Brazil (Mammalia, Chiroptera). Bol
Mus Emilio Goeldi Ser Zool 32: 161?182.
dos Reis NR, Schubart HOR (1979): Notas preliminares sobre
os morcegos do Parque Nacional da Amazonia m?dio
Tapajos. Acta Amaz?nica 93: 507?515.
Root RB (1967): The niche exploitation pattern of the blue-gray
gnatcatcher. Ecol Monogr 37: 317?350.
van Roosmalen MGM, van Roosmalen T, Mittermeier RA, de
Fonseca GAB (1998): A new and distinctive species of 
marmoset (Callitrichidae; Primates), from the lower Rio
Aripuan?, State of Amazonas, Central Brazilian Amazonia.
Goeldi Zool 22: 1?27.
Scariot A (1999): Forest fragmentation: Effects on palm diver-
sity in Central Amazonia. J Ecol 87: 66?76.
Schnitzler HU, Kalko EKV (1998): How echolocating bats
search and find food. In: Kunz TA, Racey PA, eds., Bat
Biology and Conservation. Washington, DC and London,
Smithsonian Institution Press, pp. 183?204.
Schnitzler HU, Kalko EKV, Kaipf I, Grinnell AD (1994): Fishing
and echolocation behavior of the greater bulldog bat, 
Noctilio leporinus, in the field. Behav Ecol Sociobiol 35:
327?345.
Simmons NB, Handley CO Jr (1998): A revision of Centronyc-
teris Gray (Chiroptera: Emballonuridae) with notes on
natural history. Am Mus Nov 3239: 1?28.
Simmons NB, Voss RS (1998): The mammals of Paracou,
French Guyana: A Neotropical lowland rainforest fauna.
Part 1. Bats. Bull Am Mus Nat Hist 237: 1?219.
Terborgh J, Winter B (1980): Somes causes of extinction. In:
Soul? ME, Wilcox BA, eds., Conservation Biology: an
Evolutionary-Ecological Perspective. Sunderland, MA,
Sinauer Associates, pp. 119?133.
Thies W, Kalko EKV, Schnitzler HU (1998): The roles of echolo-
cation and olfaction in two Neotropical fruit-eating bats,
Carollia perspicillata and C. castanea, feeding on Piper.
Behav Ecol Sociobiol 42: 397?402.
Uieda W (1980): Occorr?ncia de Carollia castanea na 
Amaz?nia Brazileira (Chiroptera: Phyllostomidae). Acta
Amaz?nica 104: 936?938.
Uieda W (1992): Feeding activity period and types of prey of
the vampire bats (Phyllostomidae) in southeastern Brazil.
Rev Bras Biol 52: 563?573
Voss RS, Emmons LH (1996): Mammalian diversity in Neotrop-
ical lowland rainforests: a preliminary assessment. Bull Am
Mus Nat Hist 230: 1?115.
Whitmore TC (1997): Tropical forest disturbance, disappear-
ance and species loss. In: Laurance WF, Bierregaard RO Jr,
eds., Tropical Forest Remnants. Ecology, Management, and
Conservation of Fragmented Communities. Chicago and
London, University of Chicago Press, pp. 3?13.
Willig MR (1986): Bat community structure in South America:
A tenacious chimera. Rev Chil Hist Nat 59: 151?168.
Wilson DE (1996): Neotropical bats: A checklist with conser-
vation status. In: Gibson AC, ed., Neotropical Biodiversity
and Conservation. Los Angeles, California, University of
California. pp. 167?177.