ArticlePDF Available

Forest Fragments with Larger Core Areas Better Sustain Diverse Orchid Bee Faunas (Hymenoptera: Apidae: Euglossina)



Male orchid bees were attracted to chemical baits and collected in nine Atlantic Forest fragments in southeastern Brazil. Fragments differed in size and shape. Three additional sites were also sampled in a nearby large fragment. Three hypothetical core areas of each fragment were measured as the total area minus an area of 50, 100, and 200-m-wide perimeter. Abundance and richness were not correlated with either fragment size or ratio area/perimeter, but were positively correlated with the size of core areas. These results suggest that orchid bee conservation requires the preservation of the fragments with the largest possible core areas. Neither size nor shape alone (area/perimeter ratio) seemed to be good indicators of the value of a given fragment for sustaining diverse and abundant faunas of orchid bees.
July - August 2010
Forest Fragments with Larger Core Areas Better Sustain Diverse Orchid
Bee Faunas (Hymenoptera: Apidae: Euglossina)
Depto de Zoologia, Univ Federal de Minas Gerais, CP 486, 30161-970, Belo Horizonte, MG, Brasil;
Edited by Fernando Noll – UNESP
Neotropical Entomology 39(4):555-561 (2010)
ABSTRACT - Male orchid bees were attracted to chemical baits and collected in nine Atlantic Forest
fragments in southeastern Brazil. Fragments differed in size and shape. Three additional sites were also
sampled in a nearby large fragment. Three hypothetical core areas of each fragment were measured
as the total area minus an area of 50, 100, and 200-m-wide perimeter. Abundance and richness were
not correlated with either fragment size or ratio area/perimeter, but were positively correlated with
the size of core areas. These results suggest that orchid bee conservation requires the preservation of
the fragments with the largest possible core areas. Neither size nor shape alone (area/perimeter ratio)
seemed to be good indicators of the value of a given fragment for sustaining diverse and abundant
faunas of orchid bees.
KEY WORDS: Atlantic Forest, conservation, Euglossine, habitat fragmentation
Deforestation almost always results in fragmentation of
the original forest into isolated patches of tall trees embedded
in a modi ed matrix (Tocher et al 1997). Species richness
and population sizes of forest-dependent animals and plants
usually decline as a result of forest loss and fragmentation
(Franklin & Forman 1987, Collinge 1996, but see Cane
2001). Because there are no methods to determine the
minimum areas of reserves with reference only to ecosystem
properties (see Soulé & Simberloff 1986, Beier 1993),
biologists have been forced to conduct viability analysis for
a few “indicator” or “umbrella” species as an ef cient way
to address the viability of the whole system (Soulé 1987,
Noss 1991). These analyses, however, have focused on large
vertebrates, which require large areas (e. g. Picton 1979,
Freemark & Merriam 1986, Dodd 1990, Laurance 1990,
1994, Beier 1993, Lankester et al 1991, Newmark 1991,
Opdam 1991, Herkert 1994, Brooks et al 1999, Chiarello
1999) but little is known about the effects of fragmentation
on faunas of invertebrates (see Hopkins & Webb 1984, Klein
1989, Daily & Ehrlich 1995).
The few studies involving fragmentation and orchid bees
were carried out in recently fragmented landscape in the
Amazon Basin (Biological Dynamics of Forest Fragments
Project, Powell & Powell 1987, Becker et al 1991) or in
Atlantic Forest areas in which fragmentation took place
over a century ago (Bezerra & Martins 2001, Tonhasca Jr
et al 2002, Souza et al 2005, Nemésio & Silveira 2007,
Aguiar & Gaglianone 2008, Farias et al 2008). These studies
involved few fragments and only related fragment size to
bee diversity.
Although orchid bees are able to y several kilometers
each day in search for food and aromatic compounds (Janzen
1971), there are evidences that some species are unable to
cross open spaces only a few dozen meters wide between
two forest fragments (Powell & Powell 1987, Becker et al
1991). This suggests they are strongly dependent on forest
Male orchid bees are the main (and frequently the only)
pollinators of about 650 species of orchids (Ackerman 1989),
as well as other plant species (reviewed by Dressler 1982).
For this reason, their conservation is a matter of concern.
Most euglossine species are only found in forest habitats
(Roubik & Hanson 2004) which have been widely destroyed
in the Neotropics. To effectively conserve orchid bees in
remnant forests, it is necessary to understand all the effects
of habitat fragmentation on these bees. The main goal of
this paper was to assess the in uence of size and shape of
fragments on their orchid bee community structure.
Material and Methods
Study sites. Data were collected in nine forest fragments
near the urban areas of the Belo Horizonte metropolitan
region (state of Minas Gerais, Brazil), with more than three
million inhabitants. Belo Horizonte (19º58’ - 20º06’S, 43º55’
- 44º04’W, elevation: 800-1,100 m) is at the border of two
major Brazilian biomes, the Atlantic Forest and the “Cerrado”
(Brazilian savanna). The dominant forest in the region is the
semideciduous forest, called “low mountain rain forest” by
Rizzini (1979), at elevations of 300-800 m. The forest canopy
reaches 15-25 m and trunks vary from 40 cm to 60 cm in
556 Nemésio & Silveira - Forest Fragments with Larger Core Areas Better Sustain Diverse Orchid Bee Faunas...
diameter. There are relatively few epiphytes and lianas, but
the understory is well developed. Denser stands of larger
trees grow in the humid ravines. The forest gets sparser and
shorter as the altitude increases, being substituted at the top
of the tallest hills by patches of Cerrado or (above 1,000 m)
by “campos rupestres” (rocky elds). The regional climate
is the AW of Köppen (tropical with rainy summers and a dry
winter with mean annual temperature of 18C). These forest
fragments are surrounded by areas with different degrees of
anthropogenic disturbance. Locations, sizes, elevations and
shape parameters of the nine sampled fragments are given
in Table 1.
Besides the fragments above, three sites were also
sampled at Reserva Particular do Patrimônio Natural da Serra
do Caraça (RSC), a large (>10,000 ha) forest area situated
in the municipality of Catas Altas, ca. 60 km from Belo
Horizonte, with similar vegetation (Vasconcelos 2000). The
rst site (RSC-1; 20º02’37” S - 43º30’17”W) was situated at
850 m above sea level. The second one (RSC-2; 20º04’31”
S - 43º30’37”W) at ca. 1,200 m and the third one (RSC-3;
20º05’44” S - 43º29’44”W) at ca. 1,350 m.
Sampling. Male orchid bees were captured monthly at a
single xed spot in each site, between 10:00h and 16:00h,
during one year, between May 1999 and April 2000. Five
chemicals (benzyl acetate, 1,8-cineole, eugenol, methyl
trans-cinnamate, and vanillin) were used to attract the bees.
They were imbued in cotton waddings hanging from branches
at about 1.5 m above the soil surface and distant from each
other at least 2.0 m. Monthly, the three sites at RSC were
always sampled on the same day, as well as the three sites
at Barreiro (Barreiro large, median, and small) and the two
sites of Catarina (Catarina large and small). This practice was
adopted to avoid bias due to in uence of possible different
climatic conditions if sampling in sites of the same area were
taken in different days. All collected specimens were pinned,
identi ed and deposited at the entomological collection of
the Taxonomic Collections of the Universidade Federal de
Minas Gerais. Taxonomy follows Nemésio (2009).
Data analysis. Sizes and “shapes” of forest fragments were
correlated with the abundance and species richness of their
orchid bee fauna through the Spearman rank correlation
test, considering a 5% signi cance level. Shape of the forest
fragment was calculated as: (i) the area/perimeter ratio and
(ii) the core area size. The core area is that area resulting
from the exclusion of a uniform border of a given width
off the fragment. Three core areas were estimated for each
fragment, excluding borders 50 m, 100 m, and 200 m wide
(respectively, CA
, CA
and CA
) measured from the
forest edge. When, after excluding a given border, a fragment
was split into two or more core areas, only the largest one was
used for analysis. When no area was left after the exclusion
of a border stripe of a give width, we tried two analyses: (i)
this fragment was not considered in the correlation test for
that category; (ii) the area of the fragment was considered
to be zero and it was considered in the correlation test. The
sites in RSC were not used for core area analyses, since they
Municipality Brumadinho
Brumadinho Brumadinho Nova Lima
Elevation above
sea level (m)
1,400 1,100 1,100 1,100 900 900 1,350 1,100 950
Matrix (surroun
ding area)
Cerrado and
Cerrado and
Cerrado and
Cerrado and
Cerrado and
Cerrado and
urban areas
Cerrado and
Total area (ha) 1.0 2.0 45.0 180.0 5.0 119.0 353.8 93.0 100.0
0.01 0.2 32.0 120.0 1.1 68.0
69.0 67.0
0.0 0.0 20.0
59.0 51.0
0.0 0.0 13.0
46.0 35.0
Perimeter (km) 0.4 1.4 7.2 37.0 2.2 28.7 36.7 21.3 14.9
Ratio a/p
2.5 1.4 6.3 4.9 2.3 4.1 9.6 4.4 6.7
Split into two areas of 149.2 ha and 143 ha, respectively;
Split into four areas of 10.5 ha, 10.5 ha, 18.0 ha and 25 ha, respectively;
Split into two areas of 17.5 ha and 19.5 ha, respectively;
Split into three areas of 3.0 ha, 124.6 ha and 123.4 ha, respectively;
Split into four areas of 1.5 ha, 3.0 ha, 7.5 ha and 12.0 ha, respectively;
Split into two areas of 10.0 ha, and 11.0 ha, respectively;
Split into two areas of 91.1 ha and 95.7 ha, respectively.
Table 1 Sampled sites in Belo Horizonte metropolitan region and some important features. Areas: SM = Serra da Moeda;
BS = Área de Proteção Especial (APE) do Barreiro (small fragment); BM = APE do Barreiro (medium-sized fragment); BL =
APE do Barreiro (large fragment); CS = APE do Catarina (small fragment); CL = APE do Catarina (large fragment); FCH =
APE de Fechos; PM = Parque das Mangabeiras; TAB = APE de Taboões. Different fragments in a same area are named after
their relative sizes: s = small; m = median; l = large. Campo = campo rupestre (rocky eld); Cerrado = Brazilian savanna.
CAs are estimates of core area, obtained through subtraction, from the total area of the fragment, of the corresponding area
of 50, 100, and 200 m of edge. Ratio a/p = Ratio area/perimeter.
July - August 2010 Neotropical Entomology 39(4)
are not fragments, but sites situated in the same continuous
large area. For the same reason, they were considered as the
largest fragments when effect of fragment size was analyzed.
To avoid bias, when the nine fragments of the Belo Horizonte
region were analyzed in respect to their core area, two sets
of data were generated: the rst including all nine fragments
and the second excluding the two sites situated at the highest
elevations (FCH and SM, Table 1), leaving only the seven
sites situated at approximately the same elevation. Data and
analyses focusing on elevation were published elsewhere
(Nemésio 2008).
The similarity in faunistic composition among the
twelve sites was estimated by the percent similarity index
of Renkonen, recommended by Wolda (1981) for small
samples. Based on those similarities, the areas were grouped
using UPGMA (Sneath & Sokal 1973). The resulting
similarity matrix was correlated to a matrix of geographic
distance among the sites. Nevertheless, since the elements
are not independent (Fortin & Gurevitch 1993), the Mantel
permutation test was used for these correlations (Douglas
& Endler 1982, Manly 1994, Sokal & Rohlf 1995). For
calculating Z statistics, 1,000 permutations were used, as
recommended by Fortin & Gurevitch (1993).
A total of 2,381 male orchid bees belonging to at least 14
species were collected at the nine areas in Belo Horizonte
and the three sites at RSC (Table 2). Abundance and species
richness were not correlated with fragment size, independent
of the data set employed.
No correlation was found between core area size and
abundance or richness considering the nine fragments.
Nonetheless, when the two sites situated at the highest
elevations (Serra da Moeda and Fechos, both above 1,300
m) were excluded and only the seven fragments situated
approximately at the same altitude (900-1,100 m) were
considered, both abundance (CA
and CA
: r
= 0.90, n
= 5, P < 0.05) and richness (CA
and CA
: r
= 0.98, n =
5, P < 0.05) were correlated with the core area of fragments
for the widest perimeter categories (CA
and CA
result was also achieved when the fragments with CA
= 0 were considered to have area zero and included in the
analysis (for abundance CA
and CA
: r
= 0.69, n = 7,
P < 0.05; for richness CA
and CA
: r
= 0.72, n = 7, P <
0.05). Abundance and species richness were not correlated
with the area/perimeter ratio in any analysis.
The ordination of the sites according to their faunas (Fig
1) shows a great overall similarity among the sites, with the
most distinctive of them (RSC-1) still sharing more than 40%
similarity with the others. The seven sites at approximately
the same altitude in Belo Horizonte region showed more
similarity to each other than to the two other sites at the
highest elevations or to the RSC sites.
When similarity was correlated to geographic distance
through the Mantel test, a signi cant correlation was obtained
(r = -0.494; t = -2.68; n = 12; P = 0.004), i.e., the shorter the
distance, the greater the similarity among sites.
Orchid bee species richness and composition. The species
collected in the present study are essentially the same
collected in a previous work carried out in the same region
(Nemésio & Silveira 2007), although only the fragment
Parque das Mangabeiras was sampled in both studies. The
only species recorded in the present study and not collected
by Nemésio & Silveira (2007) was Eufriesea violacea
(Blanchard). Nonetheless, this species was collected at RSC
and not in the Belo Horizonte region (see Table 2). The
same is true for species composition; details on the currently
known distribution of these species were also presented by
Nemésio & Faria Jr (2004) and by Nemésio (2009). The high
similarity among orchid bee faunas of all sampled sites may
re ect the connections among the fragments and also their
obvious common biogeographic history.
The correlation between geographic distance and
similarity of faunas revealed by the Mantel test and clearly
seen in Fig 1 should be pointed out. Besides, it is noticeable
that fragments situated approximately at the same elevation
grouped together (FCH and RSC-2; SM and RSC-3 and the
seven fragments situated at approximately the same elevation
in Belo Horizonte – [BS + (CL + CS)] + [(BL + BM) + TAB
+ PM]). Interestingly, BS grouped rst with the two sites of
Catarina reserve (CL + CS) instead of grouping with BL and
BM, its neighbor sites. This is due to the strong in uence of
Eulaema nigrita Lepeletier (see Table 2), a species regarded
as typical of open and/or disturbed areas (see Tonhasca Jr
et al 2002 and Nemésio & Silveira 2006b; for alternative
hypothesis, see Bezerra & Martins 2001 and Nemésio &
Silveira 2006a). The four largest fragments of the Belo
Horizonte area situated at similar elevations (BM, BL, TAB,
and PM) grouped together (sites where the dominance of El.
nigrita is weaker than in BS, CL and CS) (see Fig 1).
Fragment size. Nemésio & Silveira (2007) found a positive
correlation between fragment size and abundance (but not
to species richness, though it was suggested that fragment
size could influence species richness). Nevertheless, the
data presented here do not corroborate such correlation. The
difference between those results may be due, primarily, to the
areas sampled in each study. All four fragments sampled by
Nemésio & Silveira (2007) are at similar elevations (850-1,100
m), whereas in this study fragments between 900 m and 1,400
m were sampled. Moreover, the four fragments of the rst study
were quite distant from each other (2.3-12.8 km) while, in the
present study, different degrees of connection were selected.
Barreiro-small (2 ha) presented a high abundance and
species richness, most probably because it is between two
larger fragments (45-180 ha) and distant only a few tens of
meters from both of them. Thus, many of the bees collected
there may have been attracted from the larger fragments
nearby. This, surely, contributed for the reduction of the
correlations, since high values of abundance and richness
were attributed to one of the smallest areas. It also should
be noted that the orchid bee faunas of neighbor fragments
tended to be the most similar to each other, when distance
between fragments was smaller than 100 m (Barreiro large,
558 Nemésio & Silveira - Forest Fragments with Larger Core Areas Better Sustain Diverse Orchid Bee Faunas...
Table 2 Number (N) and relative abundance (%) of male orchid bees according to species collected in the twelve sampling sites: Barreiro-small (BS), Barreiro-
median (BM), Barreiro-large (BL), Catarina-small (CS), Catarina-large (CL), Fechos (FCH), Serra da Moeda (SM), Taboões (TAB), Parque das Mangabeiras (PM),
and the three sites at RSC (RSC-1, 2, and 3).
Species N % N % N % N % N % N % N % N % N % N % N % N %
Eufriesea violacea
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 2 0 0 0 0 3
Ef. nigrohirta
1 0.3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
Euglossa aratingae
0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0.2 0 0 0 0 0 0 0 0 1
Eg. fimbriata
Rebêlo & Moure
18 5 25 8 22 7 4 4 4 2 1 4 0 0 30 7 83 21 7 4 0 0 5 26 199
Eg. imperialis
5 1 3 1 6 2 0 0 0 0 0 0 0 0 19 4 4 1 0 0 0 0 0 0 37
Eg. leucotricha
Rebêlo & Moure
2 0.6 1 0.3 2 0.6 0 0 0 0 0 0 0 0 1 0.2 3 1 0
0 2 5 3 15 14
Eg. melanotricha
54 16 36 12 35 11 18 20 41 19 2 8 0 0 100 23 46 11 4 2 1 2 2 11 339
Eg. pleosticta
0 0 3 1 3 1 0 0 0 0 0 0 0 0 3 1 1 0.5 0 0 0 0 0 0 10
Eg. securigera
10 3 13 4 6 2 5 6 6 3 2 8 0 0 14 3 17 4 13 7 1 2 1 5 88
Eg. stellfeldi
24 7 32 10 44 14 0 0 2 1 0 0 0 0 27 6 56 14 0 0 8 18 2 11 195
Eg. truncata
Rebêlo & Moure
16 5 44 14 36
11 3 3 16 7 8 34 0 0 51 12 42 10 102 59 9 21 2 11 329
Eulaema marcii
1 0.3 0 0 1 0.3 0 0 0 0 0 0 0 0 1 0.2 1 0.5 8 5 2 4 0 0 14
El. nigrita
211 62 155 50 170 52 61 67 145 68 11 46 1 100 186 43 146 36 36 21 21 48 4 21 1,147
Exaerete smaragdina
0 0 1 0.3 0 0 0 0 0 0 0 0 0 0 0 0 3 1 0 0 0 0 0 0 4
Total 342 313 325 91 214 24 01 433 402 173 44 19 2,381
July - August 2010 Neotropical Entomology 39(4)
median, and small; Catarina large and small; distances
between all other fragments were larger than 1,000 m).
This also suggests that some migration between close-by
fragments does occur. Moreover, fragment sizes and shapes
were more homogeneous among the forest patches studied
by Nemésio & Silveira (2007), probably promoting similar
environmental conditions among fragments.
Fragment shape. It is known that two patches of the same
size but with different amount of edge may have different
population dynamics (Fahrig & Merriam 1994) and it is
common sense that, among fragments of the same size, the
one with shape most closely approaching a circle should be
preferred for conservation purposes, since it would best reduce
the edge effects (e. g. Diamond 1975, Begon et al 2006).
Although Game (1980) argued that “in certain circumstances
the optimal shape may be other than circular”, she recognizes
that, if extinction rate is highly dependent on shape, then
the optimal shape is circular (Game 1980:631). In relatively
isolated fragments or sets of fragments, as in the present study,
immersed in an urban matrix, the main challenge is to reduce
the extinction rate and not to increase the immigration rate.
The area/perimeter ratio is suggested as a practical way
to assess the “shape quality” of fragments. The higher its
value, the more similar to a circle a fragment will be; the
lower the value, the higher the edge effects will be. However,
this ratio is of limited use when fragments of different sizes
are compared, since the area/perimeter ratio of two areas of
same shape but with different sizes are not equivalent, with
the larger area also presenting a larger area/perimeter ratio.
Given that it is not shape itself that counts for organisms
depending on deep-forest environments, but the actual
area that is isolated from edge effects in the fragment, the
use of core areas should be preferred as a tool to evaluate
fragment quality for conservation. Moreover, the study of
the correlation between core areas and population abundance
and species richness is a practical tool for estimating the
absolute distance below which edge effects are important
for different kinds of organisms. Thus, our data, combined
with those presented in a previous work (Nemésio & Silveira
2006b), suggest that at 50 m from the edge, the orchid bee
community is still heavily affected by edge effects. Data on
the orchid bee fauna of a large fragment of Atlantic Forest
(36,000 ha) showed that the orchid bee faunas at 400 m and
500 m from the forest edge are more similar to those at 2,000
m and 4,000 m from the edge than to that at 50 m (Nemésio
& Silveira 2006b).
The data presented here suggest that the orchid bee
faunas of the fragments with the largest core areas at least
100 m far from the closest edge are richer and more abundant
than those occurring in areas with limited core areas. Thus,
large but narrowly linear reserves will not be effective in
conserving orchid bees. Our data do not allow us to estimate
the minimum area for effective conservation of euglossine
species dependent on deep-forest environments. However,
considering the fact that orchid bees are believed to y a
few to several thousands meters in search of the resources
they need (Janzen 1971), it can be expected that those forest
species demanding well preserved environments will need
reserves of several hundreds to a few thousands hectares
for long-term conservation. However, small fragments are
important to conserve less restrictive species.
This study suggests that the best areas to be preserved in
the Atlantic Forest domain, as far as orchid bee conservation
is concerned, are those still holding well preserved core areas
at least 100 m far from the closest forest edge. However,
complementary studies involving larger number of fragments
are necessary to de ne which the minimum size of such
core areas would be for each species. The employment of
core areas, thus, seems to be a useful tool for conservation
policy, since areas to be preserved can be objectively selected.
Since different organisms will respond differently to speci c
distances to the forest edge, combinations of minimum edge
distances for core areas estimated for several taxa should be
employed in selecting the best areas to be preserved.
We are indebt to James Cane, who kindly read the rst
Fig 1 Clustering of the twelve sampled sites in the metropolitan region of Belo Horizonte and at the Reserva Particular do
Patrimônio Natural da Serra do Caraça (RSC) according to the similarity of their orchid bee faunas. BS = Barreiro-small, BM =
Barreiro-median, BL = Barreiro-large, CS = Catarina-small, CL = Catarina-large, FCH = Fechos, SM = Serra da Moeda, TAB =
Taboões, PM = Parque das Mangabeiras, and the three sites at RSC (RSC-1, 2, and 3).
Percent similarity
560 Nemésio & Silveira - Forest Fragments with Larger Core Areas Better Sustain Diverse Orchid Bee Faunas...
version of this manuscript and made valuable comments on
it. Dr. Fernando Consoli, as the Chief Editor of Neotropical
Entomology, and the anonymous referees also made
important criticisms to previous versions of this manuscript.
We also thank A. G. Damasceno and R. Scholte, who helped
with eldwork. We thank the Prefeitura Municipal de Belo
Horizonte for allowing us to sample (through the Park’s
administration) the Parque Municipal das Mangabeiras; the
direction of the Reserva Particular do Patrimônio Natural da
Serra do Caraça for allowing us to sample that area and the
Companhia de Saneamento de Minas Gerais, COPASA-MG,
for allowing us to sample their reserves at Fechos, Taboões,
Barreiro and Catarina. This study was partly supported by the
Ministério do Meio Ambiente and CNPq, through PROBIO
(process #380678/1999-3), which also granted a fellowship
to the rst author. This paper is part of the rst authors
M.Sc. Dissertation.
Ackerman J D (1989) Geographic and seasonal variation in
fragrance choices and preferences of male euglossine bees.
Biotropica 21: 340-347.
Aguiar W M, Gaglianone M C (2008) Comunidade de abelhas
Euglossina (Hymenoptera: Apidae) em remanescentes de mata
estacional semidecidual sobre tabuleiro no estado do Rio de
Janeiro. Neotrop Entomol 37: 118-125.
Becker P, Moure J S, Peralta F J A (1991) More about euglossine
bees in Amazonian Forest fragments. Biotropica 23: 586-591.
Begon M, Townsend C R, Harper J L (2006) Ecology: Individuals,
populations and communities. 4th ed, Oxford, Blackwell
Scienti c Publications, 738p.
Beier P (1993) Determining minimum habitat areas and habitat
corridors for cougars. Cons Biol 7: 94-108.
Bezerra C P, Martins C F (2001) Diversidade de Euglossinae
(Hymenoptera: Apidae) em dois fragmentos de Mata Atlântica
localizados na região urbana de João Pessoa, Paraíba, Brasil.
Rev Bras Zool 18: 823-835.
Brooks T, Pimm S L, Oyugi J O (1999) Time lag between
deforestation and bird extinction in tropical forest fragments.
Cons Biol 13: 1140-1150.
Cane J H (2001) Habitat fragmentation and native bees: a premature
verdict? Cons Ecol 5: 3. [online] URL: http://www.consecol.
Chiarello A G (1999) Effects of fragmentation of the Atlantic
forest on mammal communities in south-eastern Brazil. Biol
Cons 89: 71-82.
Collinge S K (1996) Ecological consequences of habitat
fragmentation: implications for landscape architecture and
planning. Land Urban Plan 36: 59-77.
Daily G C, Ehrlich P R (1995) Preservation of biodiversity in small
rainforest patches: rapid evaluations using butter y trapping.
Biodiv Cons 4: 35-55.
Diamond J M (1975) The island dilemma: lessons of modern
biogeographic studies for the design of natural reserves. Biol
Cons 7: 129-146.
Dodd Jr C K (1990) Effects of habitat fragmentation on a stream-
dwelling species, the attened musk turtle Sternoterus depressus.
Biol Cons 54: 33-45.
Douglas M E, Endler J A (1982) Quantitative matrix comparisons
in ecological and evolutionary investigations. J Theor Biol 99:
Dressler R L (1982) Biology of the orchid bees (Euglossini). An
Rev Ecol Syst 13: 373-394.
Fahrig L, Merriam G (1994) Conservation of fragmented
populations. Cons Biol 8: 50-59.
Farias R C A P, Madeira-da-Silva M C, Pereira-Peixoto M H,
Martins C F (2008) Composição e sazonalidade de espécies de
Euglossina (Hymenoptera: Apidae) em mata e duna na Área
de Proteção Ambiental do Rio Mamanguape, Rio Tinto, PB.
Neotrop Entomol 37: 253-258.
Fortin M J, Gurevitch J (1993) Mantel tests: spatial structure in
eld experiments, p.342-359. In Scheiner S M, Gurevitch J
(eds) Design and analysis of ecological experiments. New York,
Chapman & Hall, 530p.
Franklin J, Forman R (1987) Creating landscape patterns by forest
cutting: ecological consequences and principles. Land Ecol 1:
Freemark K E, Merriam G (1986) Importance of area and habitat
heterogeneity to bird assemblages in temperate forest fragments.
Biol Cons 36: 115-141.
Game M (1980 Best shape for nature reserves. Nature 287: 630-
Herkert J R (1994) The effects of habitat fragmentation on
Midwestern grassland communities. Ecol Appl 4: 461-471.
Hopkins P J, Webb N R (1984) The composition of the beetle
and spider faunas on fragmented heathlands. J Appl Ecol 21:
Janzen D H (1971) Euglossini bees as long-distance pollinators of
tropical plants. Science 171: 203-205.
Klein B C (1989) Effects of forest fragmentation on dung and
carrion beetle communities in central Amazonia. Ecology 70:
Lankester K, van Apeldoorn R, Meelis E, Verboom J (1991)
Management perspectives for populations of the Eurasian
badger (Meles meles) in a fragmented landscape. J Appl Ecol
28: 561-573.
Laurance W F (1990) Comparative responses of ve arboreal
marsupials to tropical forest fragmentation. J Mamm 71: 641-
Laurance W F (1994) Rainforest fragmentation and the structure
of small mammal communities in tropical Queensland. Biol
Cons 69: 23-32.
Manly B F (1994) Multivariate statistical methods: a primer. Second
July - August 2010 Neotropical Entomology 39(4)
edition. London, Chapman & Hall, 464p.
Nemésio A (2008) Orchid bee community (Hymenoptera, Apidae) at
an altitudinal gradient in a large forest fragment in southeastern
Brazil. Rev Bras Zoocienc 10: 249-256.
Nemésio A (2009) Orchid bees (Hymenoptera: Apidae) of the
Brazilian Atlantic Forest. Zootaxa 2041: 1-242.
Nemésio A, Faria Jr L R R (2004) First assessment of orchid bee
fauna (Hymenoptera: Apidae: Apini: Euglossina) of Parque
Estadual do Rio Preto, a cerrado area in southeastern Brazil.
Lundiana 5: 113-117.
Nemésio A, Silveira F A (2006a) Deriving ecological relationships
between host and parasitic species – an example with orchid
bees. J Biogeog 33: 91-97.
Nemésio A, Silveira F A (2006b) Edge effects on the orchid bee
fauna (Hymenoptera: Apidae: Apini: Euglossina) at a large
remnant of Atlantic Rain Forest in southeastern Brazil. Neotrop
Entomol 35: 313-323.
Nemésio A, Silveira F A (2007) Orchid bee fauna (Hymenoptera:
Apidae: Euglossina) of Atlantic Forest fragments inside an urban
area in southeastern Brazil. Neotrop Entomol 36: 186-191.
Newmark W D (1991) Tropical forest fragmentation and the
local extinction of understory birds in the eastern Usambara
Mountains, Tanzania. Cons Biol 5: 67-78.
Noss R F (1991) From endangered species to biodiversity, p.227-
246. In Kohm, K A (ed) Balancing on the brink of extinction.
Washington, Island Press, 398p.
Opdam P (1991) Metapopulation theory and habitat fragmentation:
a review of holarctic breeding bird studies. Land Ecol 5: 93-
Picton H D (1979) The application of insular biogeographic theory
to the conservation of large mammals in the northern Rocky
Moutains. Biol Cons 15: 73-79.
Powell A H, Powell G V N (1987) Population dynamics of male
euglossine bees in Amazonian forest fragments. Biotropica 19:
Rizzini C T (1979) Tratado de togeogra a do Brasil. Vol 2,
Aspectos sociológicos e orísticos. São Paulo, HUCITEC &
USP, 473p.
Roubik D W, Hanson P E (2004) Orchid Bees of Tropical America:
biology and eld guide. San Jose, INBIO, 370p.
Sneath P H A, Sokal R R (1973) Numerical taxonomy. The
principles and practice of numerical classi cation. San Francisco,
W H Freeman, 573p.
Sokal R R, Rohlf F J (1995) Biometry. 3rd ed, New York, W H
Freeman & Company, 887p.
Soulé M E (1987) Introduction, p.1-10. In Soulé, M E (ed) Viable
populations for conservation. New York, Cambridge University
Press, 390p.
Soulé M E, Simberloff D (1986) What do genetics and ecology tell
us about the design of nature reserves? Biol Cons 35: 19-40.
Souza A K P, Hernández M I M, Martins C F (2005) Riqueza,
abundância e diversidade de Euglossina (Hymenoptera: Apidade)
em três áreas da Reserva Biológica Guaribas, Paraíba, Brasil.
Rev Bras Zool 22: 320-325.
Tocher M D, Gascon C, Zimmerman B L (1997) Fragmentation
effects on a Central Amazonian frog community: a ten-year
study, p.124-137. In Lawrence W F, Bierregaard Jr R O
(eds) Tropical Forest remnants – ecology, management, and
conservation of fragmented communities. Chicago, Chicago
University Press, 312p.
Tonhasca Jr A, Blackmer J L, Albuquerque G S (2002) Abundance
and diversity of euglossine bees in the fragmented landscape of
the Brazilian Atlantic Forest. Biotropica 34: 416-422.
Vasconcelos M F (2000) Reserva do Caraça: história, vegetação e
fauna. Aves 1: 3-7.
Wolda H (1981) Similarity indices, sample sizes and diversity.
Oecologia 50: 296-302.
Received 27/XI/08. Accepted 02/IV/09.
... Then, the negative effects of coffee cover on Euglossini communities are linked to forest cover loss, especially in the focal forest patches. This indirectly corroborates other observations showing lower euglossine richness and abundance in smaller forest patches (Brosi 2009;Nemésio and Silveira 2010). Forest cover (%) also played an important role in maintaining species richness and abundance within modified landscapes of Brazilian Cerrado and Atlantic Forest (Sousa et al. 2022;Carneiro et al. 2022). ...
... Our study thus reinforces the idea that the maintenance of euglossine diversity in human dominated Cerrado landscapes requires the maintenance of patches of natural vegetation in these landscapes (Sousa et al. 2022). In this sense, where large tracts of natural vegetation still exist, efforts should be directed at creating new protected areas, especially considering that relatively large areas are needed for maintaining viable populations of some species (Nemésio and Silveira 2010). On the other hand, in rural areas, efforts should be made to reinforce the compliance on existing legislation. ...
Full-text available
Natural landscapes are facing rapid transformation in many parts of the world, but the consequences of such changes for insects are poorly known. We evaluated how the conversion of the savannas and forests from Brazil´s Cerrado into lands devoted to intensive agriculture, livestock, or plantation forestry affects euglossine bees. We determined the effects of land-use change on the species richness of euglossines, and functional traits related to the body size and wing fluctuating asymmetry (FA) of a habitat generalist species. Few species were recorded in the anthropogenic land uses and most presented lower abundances in the converted than in the native habitats. The exception was Eulaema nigrita, whose average abundance in cattle pastures and in soy fields did not differ from that observed in the native habitats. However, El. nigrita males collected in pastures or in soy fields were significantly smaller than those from savannas and forests, whereas those from eucalypt tree plantations were smaller than those from savanna. Furthermore, El. nigrita males from the tree plantations had higher levels of FA in wing shape than those from native forests. Implications for conservation Our results showed that land-use change in the Cerrado biodiversity hotspot causes the impoverishment and homogenization of the Euglossini fauna. Furthermore, we found evidence that the converted habitats present more stressful conditions for the larval development of some species. This indicates that the conservation of euglossine populations in human- modified Cerrado landscapes depends largely on the presence of protected areas, including those within rural private land as required by Brazil´s Native Vegetation Protection Law.
... Then, the negative effects of coffee cover on Euglossini communities are linked to forest cover loss, especially in the focal forest patches. This indirectly corroborates other observations showing lower euglossine richness and abundance in smaller forest patches (Brosi 2009;Nemésio and Silveira 2010). Forest cover (%) also played an important role in maintaining species richness and abundance within modified landscapes of Brazilian Cerrado and Atlantic Forest (Sousa et al. 2022;Carneiro et al. 2022). ...
Orchid bees (Euglossini) are pollinators sensitive to landscape pressures related to agricultural land use, such as coffee farming. Coffee crops occupy a large land area in Brazil, and understanding the effects of coffee farming on bee communities is essential to pollinator conservation in modified landscapes. Here, we evaluated the Euglossini communities in forest patches surrounded by coffee crops in the Atlantic Forest. We hypothesized the negative effects of coffee cover (%) on euglossine richness and abundance. The euglossine males were sampled at a sampling point within forest patches of 15 landscapes in southeastern Brazil. A total of 1890 euglossine males in four genera and 14 species were sampled. Eulaema nigrita Lepeletier, 1841 was the dominant species (55.1%), followed by Euglossa cordata (Linnaeus, 1758) (25.5%). We found a new record for Euglossa liopoda Dressler, 1982, increasing the species’ known range in the Atlantic Forest. The results showed that the euglossine richness and species abundance decreased in forest patches surrounded by a high coffee cover (%). These negative effects of coffee cover on the Euglossini communities are related to forest cover substitution by monocultures with low or no floral attractiveness for these bees. This study highlights that forest patches in agricultural landscapes sustain high levels of euglossine richness. Thus, we indicate the conservation importance of these Atlantic Forest patches for bee species requirements.
... • Number of Patches in the landscape of each land use and cover class and respective area; • Largest Patch Index, the percentage of the landscape comprised by the largest patch; • Core Area, the interior area of patches after a userspecified edge buffer is eliminated. This research used buffers of 50 and 100 m, following Nemésio and Silveira (2010), also in a study in the highly fragmented Atlantic rainforest; • Core Area Index, which quantifies the core area for the entire class or landscape as a percentage of total class (in a scale of 0 to 1). ...
Habitat loss and fragmentation are ecological impacts of development projects known to have severe effects on mammals’ populations. Here, we study the role of connectivity assessment in Environmental Impact Assessment, focusing on its contribution to determining appropriate mitigation, in particular, biodiversity offsetting. For that purpose, we analysed the dispersal of large mammals in a region potentially influenced by a mine located in the Atlantic rainforest in Brazil and anticipate the long-term outcomes of restoration offsets by modelling a 30-year post-offset scenario. Results show that: (i) offsets enhance ecological connectivity and are used as an alternative habitat for the animals dispersed due to habitat loss and fragmentation; (ii) the location of restoration areas should be carefully planned to maximize their contribution to increase ecological connectivity and enhance habitat quality. The reviewed case suggests that best practices to incorporate connectivity analysis in environmental impact assessment include: (i) monitoring biodiversity to quantify losses and gains; (ii) modelling offset areas alternatives for both protection and restoration; and (iii) conducting rigorous analysis of project alternatives, aiming at impact avoidance.
... We also emphasize that the areas of average to high priority accompany the areas of greater natural vulnerability. In general, these areas correspond punctually to small (>10 ha) and medium (10-50 ha) fragments, which emerge due to their greater vulnerability to the edge effect and size reduction (dos Fahrig, 2003;Mendonça et al., 2015;Montealegre-Talero et al., 2021;Nemésio and Silveira, 2010). In fact, these areas require prioritization in forest conservation projects, where AFS are highly recommended. ...
Full-text available
The ecological functions restoration in agricultural areas is a major challenge on a landscape scale. In the specific case of active restoration through Agroforestry Systems (AFS), the absence of a specific direction hinders ecological restoration processes, especially in regions that prefer intensive agriculture. Thus, this study aims to develop a Spatial Indicator of Priority Areas to guide Agroforestry Systems implementation in agricultural landscapes. A spatial multicriteria decision analysis (MDCA) was carried out based on environmental factors: soil, geology and slope (which determine the natural vulnerability of the land) and anthropogenic factors: land use and land cover, forest fragments, potential land use capacity and legal protected areas in rural properties (which reflects human pressure and land use suitability). Subsequently, four priority levels were classified for agroforestry interventions: (1) Low priority; (2) Average priority; (3) High priority; (4) Extreme priority. A final map was made to identify priority areas for landscape recovery in 9 cities located at the mouth of the Mogi Guaçu River Hydrographic Basin, State of São Paulo, Brazil. Considering the natural vulnerability of the land and the multifunctional aspects of the landscape, the scenarios projection allowed a consensus for forest conservation and agricultural suitability perspectives. A final combination of the explored aspects culminated in the spatial indicator, which model foresees 22,300 ha available for urgent actions for restoration, reforestation and sustainable exploitation through agroforestry systems. We emphasize the challenges in reconciling the socioeconomic and ecological functions in the agroecosystem, however, the metric provides a more inclusive and assertive management strategy for natural resources and advances towards the goal of reforestation and implementation of payment for environmental services (PES) schemes.
... Since agricultural cover can affect larger bees at the landscape scale and they are more likely to reach the canopy, therefore this may result in a lower proportion of bees flying to the canopy near agricultural areas as compared to canopies in undisturbed habitats. Moreover, other factors that may influence richness and abundance must be considered when comparing understory and canopy communities such as bee and plant phenologies, ambient temperature, precipitation, water vapor pressure, wind speed, forest fragmentation, among others (Kammerer et al. 2021;Margatto et al. 2019;Nemésio and Silveira 2010;Perillo et al. 2021Perillo et al. , 2017. Overall, higher resolution studies that evaluate these factors and potential interactions with elevation and latitude are needed to appraise whether changes in orchid bee communities could result in some cases in differences between forest strata. ...
Full-text available
Vertical stratification of animal assemblages has been observed in several arthropod taxa. However, for some groups there is conflicting evidence, such as for the neotropical orchid bees. Here, we used a meta-analysis to evaluate trends in abundance and diversity of euglossine bees across strata. We found no evidence of stratification in terms of diversity, but abundance was higher in the understory. Our results highlight the effect of sampling technique as a source of heterogeneity. For instance, studies that used McPhail traps reported higher diversity, higher similarity between strata, and were more accurate for quantifying abundance, as compared to improvised traps. Studies that used higher number of bait types sampled higher diversity in general. The higher abundance in the understory tended to decrease with the number of bait types used. Variables like elevation and latitude of the study influenced bees’ distribution. We also discuss other influential variables, ecological implications, and recommend areas for future research.
... However, some studies have shown that small patches support more species abundance and richness than large patches [1,47,68,122]. Some studies have also found that the size of forest patches does not affect bee populations [36,65,79,98,110], and others claim that the capacity of small patches to support pollinators is the same as that of large ones [78,118,130]. The shape of forest patches also affects the presence of bees. ...
Full-text available
Background The growing human population and the need for more food in the world have reduced forests and turned them into agricultural land. Many agricultural products are dependent on pollinating bees, so it is possible to increase crop production by increasing the population of bees in agricultural landscapes and preventing further deforestation. In agricultural landscapes, bees use forest patches as nesting habitats and, therefore, are highly dependent on these patches. Therefore, by creating new forest patches within agricultural fields, we can increase the pollination rate, and thus the crop production. In this regard, understanding the role of forest patches and their effects on bee populations is a key step in successfully implementing the patch creation strategy. To determine the effects of forest patches on bees and pollination services, we reviewed 93 articles examining the effects of forest patches on bees in agricultural landscapes. We divided these effects into three categories based on the sampling method: (1) distance-dependent, (2) amount, and (3) structural effects. Methods We searched for published studies related to the effects of the forest patches on bees in agricultural landscapes using the ISI Web of Science. We conducted our search from May 1991 to May 2021 using the following search string keywords: forest fragment, forest patch, forest fragmentation, pollination, and bee. Results Approximately, 79% of studies showed that by increasing the distance (up to 2 km) from forest patches, regardless of the type of species, the type of agricultural product around the patches, the size and number of patches, the bees’ diversity and abundance decrease. Approximately, 76% of the studies showed that the presence of forest cover within a radius of 2 km from the target sites has a positive effect on bee populations. Our data also show that larger forest patches maintain a larger population of bees than smaller ones. Conclusion It was not clear what percentage of a landscape should be covered by forest or how much habitat was sufficient to maintain a viable population of bees. Therefore, we suggest future studies to find the thresholds of forest amounts below which the bee population is rapidly declining.
... In contrast, well-preserved natural habitats may enhance the diversity of pollinator assemblages (Menz et al., 2011;Senapathi et al., 2015). Studies identified bee species richness to be affected by landscape composition (Andersson et al., 2013), resource availability (Peters and Carroll, 2012) and the accessibility of natural habitat such as forests (Brosi et al., 2008;Klein, 2009;Nemésio and Silveira, 2010;Ricketts, 2004). However, tropical forests are among the most severely human-modified biomes and require urgent protection (Green et al., 2020;Kremen et al., 2007;Rosenzweig, 1995), especially as tropical species may be more susceptible to habitat loss than temperate species (Melo et al., 2018). ...
Animal pollinators are globally threatened by anthropogenic land use change and agricultural intensification. The yield of many food crops is therefore negatively impacted because they benefit from biotic pollination. This is especially the case in the tropics. For instance, fruit set of Coffea arabica has been shown to increase by 10–30% in plantations with a high richness of bee species, possibly influenced by the availability of surrounding forest habitat. Here, we performed a global literature review to (1) assess how much animal pollination enhances coffee fruit set, and to (2) examine the importance of the amount of forest cover, distance to nearby forest and forest canopy density for bee species richness and coffee fruit set. Using a systematic literature review, we identified eleven case studies with a total of 182 samples where fruit set of C. arabica was assessed. We subsequently gathered forest data for all study sites from satellite imagery. We modelled the effects of open (all forest with a canopy density of ≥25%), closed (≥50%) and dense (≥75%) forests on pollinator richness and fruit set of coffee. Overall, we found that animal pollination increases coffee fruit set by ~18% on average. In only one of the case studies, regression results indicate a positive effect of dense forest on coffee fruit set, which increased with higher forest cover and shorter distance to the forest. Against expectations, forest cover and distance to open forest were not related to bee species richness and fruit set. In summary, we provide strong empirical support for the notion that animal pollinators increase coffee fruit set. Forest proximity had little overall influence on bee richness and coffee fruit set, except when farms were surrounded by dense tropical forests, potentially because these may provide high-quality habitats for bees pollinating coffee. We, therefore, advocate that more research is done to understand the biodiversity value of dense forest for pollinators, notably assessing the mechanisms underlying the importance of forest for pollinators and their pollination services.
Understanding the effects of island shape and dispersal limitation on species assemblages is important for the conservation of bryophytes in fragmented landscapes. In order to explore the comparative contributions of dispersal limitation vs environmental filtering, island shape vs other island attributes to bryophyte species assemblages in fragmented landscapes, we investigated bryophyte flora and environmental variables (island shape, area, elevation, and isolation degree) of 71 islands in the Zhoushan Archipelago in the East China Sea. We used redundancy analysis, canonical correspondence analysis and generalized linear mixed models to explore comparative contributions of dispersal limitation vs environmental filtering, shape vs other explanatory variables to bryophyte species composition and species richness. Dispersal limitation independently accounted for 15% of the total variation in species composition, which was higher than environmental filtering (6.6%). When island shape was the only constraining explanatory variable, it explained 22.5% of the total species richness variation. Variance partitioning showed that island shape per se explained 3.8% of the total variation in species richness and 18.7% of the variation was confounded by island shape and other variables. Island shape combined with area and elevation was a good predictor of bryophyte species richness, independently explaining 40% of the total SR variation. Among twenty tested bryophyte categories, eighteen categories had positive relationships of their species richness with island shape irregularity, their species richness linearly increasing with island shape irregularity to varying extents. Island shape irregularity exerted a stronger effect on the species richness of acrocarpous mosses than that of pleurocarpous mosses. Therefore, dispersal limitation exerted significant effects on bryophyte species assemblages, and island shape imposed a significant and taxon-specific effect on bryophyte species richness. Our findings implied that a forest reserve with a high irregular shape is favourable for bryophyte conservation.
Habitat loss and fragmentation in the human-modified landscape affect bee assemblages by reducing the available natural resources. The life history features and adaptations of bees to environmental conditions affect how they perceive and interact with landscape structure. Therefore, it is crucial to quantify the magnitude of the effects of the landscape structure on this group at multiscale. We address the relative contribution of landscape structure in explaining the abundance of orchid bees, richness, and diversity in landscapes of Brazilian Cerrado. We mapped the surrounding landscape around 18 focal patches and sampled male orchid bees attracted by chemical baits. We used VIF and generalized linear models (GLM) to evaluate the effects of landscape structure (composition and configuration) at different spatial scales. We sampled 658 euglossine bees from five genera and 12 species. The savanna cover, forest cover, and fragmentation per se (i.e., the number of patches) were the best predictors of the richness of euglossine bees. The abundance of bees was explained by the savanna cover. The increase in natural habitat area and the reduction in landscape fragmentation favoring the assemblages of orchid bees are supported by our results. Savanna cover also modulates the abundance of euglossine bees, confirming that the habitat amount is essential for maintaining the populations of these bees. Our results reinforce the importance of habitat loss and fragmentation effects on the richness, abundance, and composition of euglossine bees, particularly in the naturally heterogeneous landscapes. bee biodiversity / habitat loss / landscape diversity / multiscale effects / pollinators
Full-text available
(1) In the modern agricultural landscape habitat patches for badgers are distributed like fragments in large suboptimal areas. In this paper the badger population in the Netherlands is considered as a metapopulation. A stochastic model is presented for a single badger clan. Subsequently, the model is extended to simulate a metapopulation consisting of badger clans that are connected through dispersal, while taking account of the present biological information. (2) The models are used to address the perspectives of management measures for badgers in the Netherlands. The results show that the viability of a single badger clan is especially sensitive to changes in adult mortality. (3) For a metapopulation of clans a similar result is obtained. In addition, the number of territories and the rate of dispersal between the clans are major factors that influence the viability of the metapopulation. (4) Management measures aimed at improving conditions for badgers should be directed at reducing the adult mortality and restoring or maintaining a large number of potential territories for clans while providing ample opportunity for dispersal between the clans.
Anthropocene enthusiasts assert the dominant human place in nature, conservation policy, and the biosphere, with plans for managing, even reengineering the planet. This worldview deemphasizes the goal of protecting nature for its own sake in favor of protecting and revising the environment for benefits to humans. But this preserves endangered species only so far as they promise benefits for humans. Environmental ethicists raise concerns that this effort to go beyond nature fails to recognize both the Earth's creative genesis and the myriads of species lines, a failure which diminishes the richness of life on the Earth, and simultaneously the fullness of human life.
Few studies directly address the consequences of habitat fragmentation for communities of pollinating insects, particularly for the key pollinator group, bees (Hymenoptera: Apiformes). Bees typically live in habitats where nesting substrates and bloom are patchily distributed and spatially dissociated. Bee studies have all defined habitat fragments as remnant patches of floral hosts or forests, overlooking the nesting needs of bees. Several authors conclude that habitat fragmentation is broadly deleterious, but their own data show that some native species proliferate in sampled fragments. Other studies report greater densities and comparable diversities of native bees at flowers in some fragment size classes relative to undisrupted habitats, but find dramatic shifts in species composition. Insightful studies of habitat fragmentation and bees will consider fragmentation, alteration, and loss of nesting habitats, not just patches of forage plants, as well as the permeability of the surrounding matrix to interpatch movement. Inasmuch as the floral associations and nesting habits of bees are often attributes of species or subgenera, ecological interpretations hinge on authoritative identifications. Study designs must accommodate statistical problems associated with bee community samples, especially non-normal data and frequent zero values. The spatial scale of fragmentation must be appreciated: bees of medium body size can regularly fly 1-2 km from nest site to forage patch. Overall, evidence for prolonged persistence of substantial diversity and abundances of native bee communities in habitat fragments of modest size promises practical solutions for maintaining bee populations. Provided that reserve selection, design, and management can address the foraging and nesting needs of bees, networks of even small reserves may hold hope for sustaining considerable pollinator diversity and the ecological services pollinators provide.
A detailed synopsis of all the orchid-bee species known to occur in the Atlantic Forest Domain, eastern Brazil, is provided, including synonymy, complete type data, diagnoses, relevant data on biology and geographic distribution (with detailed localities of known occurrence of each species), colorful illustrations of onomatophores ("name-bearing type specimens"), and a list with the main references dealing with each species. Fifty-four species are recognized to occur in the Atlantic Forest Domain. Identification keys are presented for each genus and their species occurring in the Atlantic Forest. Euglossa carinilabris Dressier, 1982, Euglossa cyanaspis Moure, 1968, Eulaema (Eulaema) niveofasciata (Friese, 1899) and Exaerete lepeletieri Oliveira & Nemesio, 2003, considered junior synonyms of other species by different authors, are reinstated as valid species. A full discussion on the status of the four orchid-bee species described by Linnaeus is presented, as well as colorful illustrations of the four onomatophores. The two existing onomatophores of orchid bee species described by Fabricius are also illustrated and his Apis cingulata has been shown to be the species recently described as Eulaema (Apeulaema) pseudocingulata Oliveira, 2006, which, thus, becomes a junior synonym (syn. n.). Euglossa aratingae sp. n., Euglossa Carolina sp. n., Euglossa nanomelanotricha sp. n., Euglossa roderici sp. n., Euglossa roubiki sp. n., Eulaema (Eulaema) atleticana sp. n., and Eulaema (Apeulaema) marcii sp. n. are described as new species. Neotypes are designated for Eufriesea violacea (Blanchard, 1840) and Exaerete frontalis (Guérin- Méneville, 1844). Some corrections concerning the repository institutions of some onomatophores of orchid bees were also made: Eufriesea auriceps (Friese, 1899) holotype has been listed as belonging to the US National Museum (Washington) or to the American Museum of Natural History (New York) but, in fact, it belongs to the Zoologisches Museum der Humboldt Universität (Berlin); the lectotype of Eufriesea aeneiventris (Mocsáry, 1896) has been listed as belonging to the Istituto e Museo di Zoologia, Universita di Torino (Turin), but it actually belongs to the Hungarian Museum of Natural History (Budapest). Publication dates of both Exaerete frontalis Guérin-Méneville and Exaerete smaragdina Guérin- Méneville have been listed as 1845 but, in fact, the actual date is 1844. Based on the known geographic distribution and abundance of each species in orchid-bee inventories, IUCN criteria were applied and three species are recommended to be included in future lists of threatened species in one of the IUCN categories of risk: Eufriesea brasilianorum (Friese, 1899) and Euglossa cognata Moure, 1970 are suggested to be listed as "vulnerable", and Euglossa cyanocholora Moure, 1996 is suggested to be listed as "endangered". A fully annotated check list of all known orchid bee species is also presented as an Appendix.
At sites near Manaus, Amazonas, Brazil we measured the impact of forest fragmentation on populations of male euglossine bees. The visitation rates of 15 species of male euglossine bees to three chemical attractants were determined for continuous forest; isolated 1, 10, and 100 ha forest fragments; and a cleared area. For most species, visitation rate declined with fragment size, even though openings separating fragments and continuous forest tended to be narrow (as little as 100 m). Pre- and post-isolation comparisons for individual fragments exhibited the same trend. Bee species attracted to scents in the clearing differed from those of the forest indicating there was little potential for forest pollination by these clearing species. These results are discussed with respect to the impact of forest fragmentation on plant species richness and the design of natural and commercial forest reserves.
Male euglossine bees were captured monthly for one year in modified McPhail traps baited with cineole, methyl salicylate, and skatole. The traps were operated 90 km north of Manaus, Brazil in continuous terra firme forest and forest fragments of 1, 10, and 100 ha. Of the 16 species of euglossine bees captured, Euglossa chalybeata and Eg. stilbonota accounted for 85 percent. The traps had the advantage of operating unattended, but their capture efficiency was low Bee abundance and species richness were significantly correlated and peaked in the wet season when flowering also peaked. Contrary to a previous study in the same area, bee abundance was greater in 10- and 100-ha fragments than in continuous forest. Fragments of 1 ha had the smallest number of individuals and species. It appears that male euglossine abundance may vary considerably over short distances, making it difficult to characterize a forest by sampling a single site.