Neotrop Entomol (2013) 42:351358

DOI 10.1007/s13744-013-0131-7

ECOLOGY, BEHAVIOR AND BIONOMICS

Mini-Winkler Extractor and Pitfall Trap as Complementary Methods

to Sample Formicidae
FHO SILVA1, JHC DELABIE2,3, GB

DOS

SANTOS4, E MEURER1, MI MARQUES1

Instituto de Biocincias, Univ Federal de Mato Grosso, Cuiab, MT, Brasil

Lab de Mirmecologia, Centro de Pesquisa do Cacau (CEPEC) CEPLAC, Itabuna, BA, Brasil
3
Depto de Cincias Agrrias e Ambientais (DCAA), Univ Estadual de Santa Cruz (UESC), Ilhus, BA, Brasil
4
Fundao Univ Federal do Tocantins, Curso de Cincias Naturais Campus Araguana, Araguana, TO, Brasil
2

Keywords
Ant diversity, ground-dwelling ants inventory,
landscape, sampling techniques
Correspondence
FHO Silva, Programa de Ps-graduao em
Ecologia e Conservao da Biodiversidade,
Instituto de Biocincias, Univ Federal de
Mato Grosso, Av. Fernando Corra da Costa,
n.2367, Boa Esperana, 78060-900, Cuiab,
MT, Brasil; fabio.silva@svc.ifmt.edu.br
Edited by Wesley AC Godoy ESALQ/USP
Received 29 May 2012 and accepted 13
March 2013
Published online 26 April 2013
* Sociedade Entomolgica do Brasil 2013

Abstract
The aim of the present study was to evaluate the use of mini-Winkler
extractor and pitfall traps as appropriate and complementary methods
to sample ant communities in the phytophysiognomy mosaic in the
Pocon Pantanal region, state of Mato Grosso, Brasil. Seven units were
studied for landscape, located within a 25 km2 collection area, formed
by thirty 250-m transects, at 1-km intervals in a 55 km area. Five
collection points were marked in each transect at 50-m intervals, totaling 150 points. A collection was made at each sampling point with miniWinkler extractor and pitfall traps. Using the mini-Winkler extractor,
1,088 individuals were collected distributed in 20 genera and 55 species,
with Solenopsis invicta Buren and Pheidole (gr. biconstricta) sp.1 as the
most frequent ants. Using pitfall traps, 2,726 individuals distributed in
24 genera and 48 species were sampled and Dorymyrmex (gr. pyramicus) sp.1 and Pheidole (gr. biconstricta) sp.1 were the most frequent
species. A significant difference between the methods was observed in
measured species number. The Principal coordinates analysis discriminated two species groups exclusively sampled by the mini-Winkler
extractor and another by the pitfall methods. Therefore, it was concluded that these methods were complementary for ant diversity inventories in the Pocon Pantanal region.

Introduction
Including terrestrial invertebrates in biodiversity and ecosystem management inventories has rapidly increased
since the end of the last century. Inventories of ant species
have been considered an important tool in studies on
tropical biodiversity conservation (Alonso & Agosti 2000).
The abundance and occurrence of ants in almost all
terrestrial ecosystems, especially in the tropics, suggest
that their capture is a relatively easy task and that the
myrmecofauna can be characterized in a given region using
only a single and simple sampling method (see, for example, Souza et al 2012). However, the different research

objectives and the various habitats of ants have marked

the development of a range of methods to capture and
study them (Bestelmeyer et al 2000, Delabie et al 2000a,
Sarmiento 2003).
Methods commonly used for litter ant inventories include fall traps, extraction from the litter, and sampling by
squares and bait (Olson 1991, Bestelmeyer et al 2000).
Fisher (1999) and Alonso & Agosti (2000) considered the
mini-Winkler extractor a technique that could be easily
standardized for myrmecofauna inventories. However, this
method favors small-sized, cryptic species or species with
little mobility that live in habitats where litter has accumulated (Bestelmeyer et al 2000), such as tropical forests.

Silva et al

352

In contrast, fall traps, such as the pitfall trap, can be

used because it is also a method considered suitable for
capturing invertebrates that are more active in the soil,
including Formicidae (Spence & Niemel 1994, Bestelmeyer
et al 2000). Furthermore, this collecting technique is considered as complementary to sampling by the mini-Winkler
extractor method (Delabie et al 2000a). Another factor to
be considered is that this method seems more indicated
for arid regions and open areas with little plant cover and
thin litter (Fisher 1998), the situation found in the study
area of the present research, because the efficiency of ant
collection methods depends essentially on the habitat under study (Romero & Jaff 1989).
The Pantanal is considered one of the largest wetlands in
Latin America, formed by a mosaic of different communities
with abrupt changes in its landscape (Prance & Schaller 1982),
resulting in interactions between edaphic, hydrologic, and
biogeographical factors (Lourival et al 2000).
Thus, the objective of the present study was to analyze the
complementariness between the mini-Winkler extractor and
pitfall trap collection methods, as suitable techniques to assess
the composition of the Formicidae community in the Pocon
Pantanal region, considering the phytophysiognomic variations of this region and the specificities of the methods used.

Material and Methods

Study area
The study was carried out in the Cuiab-Bento GomesParaguaizinho Pantanal, called the Pocon Pantanal, in
the location of Pirizal (161524S to 175432S, 563624
W to 575623W), municipality of Nossa Senhora do Livramento, state of Mato Grosso, Brasil. Sampling was carried
out during November 2005 during the dry season with
little rainfall in the region.
The Pocon Pantanal region is formed by several phytophysiognomies that provide a set of varied habitats that were
classified in the present study according to Veloso et al (1991),
PCBAP (2007), Silva et al (2000), Santos et al (2004) and
Nunes-da-Cunha et al (2007). Five environments were sampled; the Landi and Cambarazal were characterized as forest
areas while natural fields, natural open fields, and pastures
were characterized as open areas.
Landi. This formation is characterized by a low and continuous canopy with individual height varying from 3 to 7.5 m and
is associated with the water courses. The sinsia tree vegetation is dominated by Licania parviflora (Chrysobalanaceae),
Calypthrantes eugenioides (Myrtaceae), Mabea sp.
(Euphorbiaceae), and Calophyllum brasiliense (Crusiaceae)
(Nunes-da-Cunha et al 2010).

Cambarazal. This formation corresponds to a dense homogeneous flooding area, where cambar (Vochysia divergens
(Vochysiaceae), predominates an Amazon species that
varies from 5 to 18 m in height, and that is a colonizer of
flooded natural fields in the Pantanal of PoconMT. Its
local distribution is probably due to the seeds coming from
individual specimens located in nearby riparian forests
(Nunes-da-Cunha et al 2010).
Natural fields. These represent about 30% of the vegetation
in the Pantanal (Silva et al 2000) where the murundu fields
are inserted, made up of flat areas which are flooded in the
rainy season and where there are countless little hills
(murundu), susceptible to flooding when there is extreme
flood. The flat areas and the smaller murundus are covered
by rustic vegetation and the largest areas by woody Cerrado
vegetation. Termite activities and erosion processes seem to
shape the murundus into a rounded or elliptical form, with 1
2 m maximum in height (Nunes-da-Cunha et al 2010).
Natural open fields. They are characterized by the mimoso
pasture grass Axonopus purpusii (Poaceae), a perennial vegetation resistant to temporary submersion. Widely distributed
throughout the Pantanal, they occur in cerrado fields on the
edges of bays (permanent and temporary), and mainly in
areas of seasonal open fields in the sandy Pantanal areas
(Nunes-da-Cunha et al 2010).
Pastures. These regions are basically made up of Brachiaria
humidicola (Poaceae) grass originally from Africa which adapted
well to Brasil, mainly in waterlogged soils. In these areas, there
are also pasture fields which are not periodically renewed,
where there is a mixture of B. humidicola and A. purpusii, and
are known as mixed fields (Nunes-da-Cunha et al 2010).

Field procedures
The ant community was sampled using mini-Winkler extractor
(Fisher 1998, Bestelmeyer et al 2000) and pitfall traps (Adis
2002). Twenty-nine 250-m transects were marked at intervals
of 1 km in an area of 25 km2, following the Rapid Assessment
Protocols and Long-Term Ecological Research, which allows
nontendentious estimates of the distribution, abundance, biomass, and biogeography of the species among sites studied
(Fig 1) (Magnusson et al 2005). Of the transects sampled,
three were marked in the Landi, two in the Cambarazal, four
in the Murundu field, 13 in the natural open field and seven in
the pasture phytophysiognomies. Each transect characterized
a sample point where five collections of 1 m2 litter were made
with the mini-Winkler extractor and then five pitfall traps were
installed. An interval of 50 m was established between each
collection point to guarantee sampling independence because

Complementary Methods to Sample Formicidae

353

Fig 1 The 25 km2 plot used to

collect ants in the
phytophytophysiognomy
mosaic of the Pocon Pantanal,
MT, Brasil. Source:
Geoprocessing Laboratory of
the Pantanal Study Nucleus
(NEPA) UFMT adapted from
Google Earth. Letters (A, B, C,
D, E, F) and numbers (1, 2, 3, 4,
5) represent trails northsouth
and eastwest, respectively.
Letters and numbers (A1, A2,
A3) indicate the transect
location.

the probability of collecting ants of the same colony in two

successive samples is then remote and, according to Baccaro
et al (2011), subsamples with a minimum distance of more
than 10 m can collect a more diverse number of species.
Each pitfall trap was installed for 48 h at each sampling
point and the litter samples were suspended for 72 h in the
mini-Winkler extractor in a covered environment at environmental temperature (Bestelmeyer et al 2000).
Laboratory procedures
Ants were screened and identified at taxonomic level of
the genus, species, and/or morphospecies level following
Bolton et al (2006), Baroni-Urbani & Andrade (2007), and
LaPolla et al (2010). Vouchers were deposited at the Coleo de Referncia Entomolgica, Laboratrio de Taxonomia
e Ecologia de Artrpodes, Universidade Federal de Mato
Grosso.
Data analyses
Only workers and their absolute frequency were assessed
in the data analysis. According to Romero & Jaff (1989),
the capture measurement units of fauna surveys should be
at least equivalent. Thus the capture frequency, the

measurement used in this study, substituted abundance

in comparisons between the methods to estimate the
probability of finding a determined species in the study
location and reduced the effect of abundance of the social
insects in the species spatial distribution studies. According
to Hughes (1986) and Longino (2000), for social groups, the
number of individuals per se is a weak and erroneous
measurement, because using the mini-Winkler extractor
and pitfall traps in an area where ants with massive recruitment behavior occur, such as the army ants, Pheidole
or Solenopsis species, would give tendentious results that
do not represent the local community.
The species number was compared to the estimated
species number using three species richness estimators,
Jackknife 1, the ChaoJaccard similarity index, and the index
based on the concept of cover of the sample, and species
accumulation and rarefaction curves were constructed using
the EstimateS version 7.5 software (Cowell 2005).
The nonparametric statistical KruskalWallis method
was used to compare the number of species per sample
among the methods in each environment and among the
environments in each method using the STATISTICA version
7 software.
Ants were classified by functional guild according to
Silvestre et al (2003) and Brando et al (2009). The

354

Silva et al

EstimateS software (Cowell 2005) version 7.5 was used to

calculate the ChaoJaccard similarity coefficient (Chao et al
2005) among collection methods.
The principal coordinate analysis (PCoA) was used to
assess whether there was a pattern in the species distribution between the two methods, using capture frequency
data standardized by traps and calculations based on the
Bray Curtis similarity measurement using the Principal Coordinate Analysis software (Anderson 2003).

Results
A total of 3,814 Formicidae individuals were obtained by
the pitfall trap and mini-Winkler extractor methods, distributed in 33 genera and 73 species, with Pheidole
(Westwood), Solenopsis (Westwood), Camponotus (Mayr),
and Hypoponera (Santschi) presented the greatest richness
(see Electronic Supplementary Material; ESM). Pitfall traps
(2,726 individuals) allowed the collection of nearly three
times more ants than the mini-Winkler extractor (1,088
individuals). Twenty genera and 55 species were sampled
with the mini-Winkler extractor as compared to the 24
genera and 48 species collected with the pitfall traps.
Pheidole and Solenopsis commonly represented in both
methods, while Camponotus appeared as one of the richest genus in the pitfall traps and Hypoponera in the miniWrinkler method (see ESM).
Solenopsis invicta (Buren) (16 occurrences), Pheidole
(gr. biconstricta) sp.1 (14 occurrences), Brachymyrmex cf.
heeri (12 occurrences), Nylanderia sp.1 (11 occurrences),
and Solenopsis (gr. globularia) sp.1 (10 occurrences) were
the most frequent using the mini-Winkler (see ESM).
Dorymyrmex (gr. pyramicus) sp.1 (26 occurrences), Pheidole (gr. biconstricta) sp.1 (22 occurrences), S. invicta (19
occurrences), and Pheidole sp.10 (16 occurrences) were the
most frequent using the pitfall traps (see ESM).
Using the mini-Winkler extractor, 442 individuals were
sampled in the forest environment, which were distributed
in 15 genera and 26 species. Solenopsis (4 spp.), Nylanderia
(3), and Hypoponera (3) presented the greatest richness.
Using pitfall traps, 209 individuals, 11 genera, and 21 species
were sampled; Pheidole (4), and Solenopsis (3) were the
groups with the greatest number of species (see ESM).
Using the mini-Winkler, 646 individuals were sampled in
natural open fields, which were distributed in 22 general
and 47 species. Pheidole (6 spp.) and Solenopsis (5) were
the genera with greatest richness. Using pitfall traps, 2,545
individuals were sampled, 26 general and 46 species. Pheidole (7 spp.), Solenopsis (5), and Camponotus (4) were the
groups with the greatest number of species (see ESM).
No significant differences were observed between the
methods regarding the number of species when assessed

Fig 2 Box plot of the KruskalWallis non parametrical analysis. Comparison of the species number between the pitfall trap and mini-Winkler
methods in forest environment (a) and open fields (b). Comparison of
the species number between environments using pitfall (c) and miniWinkler (d). The horizontal line indicates the average value, the box
corresponds to the percentile (2575%) values, whisker represents the
minimum and maximum values of the whole sampling.

Complementary Methods to Sample Formicidae

355

Table 1 Observed and estimated ant species richness in different phytophysiognomies collected by different methods in the Pocon Pantanal,
MT, Brasil. Number of species observed, number of species after rarefaction (Coleman method), number of species estimated by Jackknife I, and
Chao 2, index based on the sample cover index (ICE). Values in parentheses correspond to the standard error.
Sample methods

Observed and estimated ant species richness

Observed

Pitfall trap
Mini-Winkler
Pitfall Trap+mini-Winkler

48
55
73

Rarefaction

Jackknife 1

Chao 2

ICE

47.7 (0.6)
54.2 (0.9)
72.3 (0.8)

61.5 (4.5)
82.0 (9.2)
101.0 (10.5)

60.5 (7.9)
95.6 (17)
103.2 (12.8)

60.7 (0.01)
99.3 (0.1)
114.1 (0.01)

in the forest environment (KruskalWallis=0.1012; df=1, 10;

P=0.7503) (Fig 2a). However, there was significant difference when the number of species was assessed in the
natural open fields (KruskalWallis=14.0879; df=1, 48;
P=0.0002) (Fig 2b). When the performance of each
method was analyzed in relation to the environment,
no significant difference was observed when the number of species collected by pitfall traps was compared
among the environments (KruskalWallis=9.1879; df=1, 29;
P=0.6582) (Fig 2c). A similar pattern was observed when the
results obtained for the mini-Winkler were assessed (Kruskal
Wallis=3.5656; df=1, 29; P=0.0590) (Fig 2d).
An application of the richness estimators showed that
55 species were sampled by the mini-Winkler extractor
and 48 using the pitfall traps (Table 1). The association
among the methods showed that 73 species were observed, indicating a good complementariness of these
methods for the number of species collected (Table 1).
This complementariness can also be observed when the
species accumulation curve is analyzed. A larger number
of species is collected with a smaller number of samples
when methods are associated compared to the results
obtained using isolated methods (Fig 3). Furthermore, a
0.67 coefficient similarity was observed between methods
using the ChaoJaccard Index.

Fig 3 Accumulation curve of

the number of Formicidae
species collected in different
phytophysiognomies in the
Pocon Pantanal, MT, Brasil,
using the mini-Winkler
extractor and pitfall traps.
Captions: dashed line observed
richness, filled circle miniWinkler, filled square pitfall
trap, filled diamond miniWinkler, plus sign pitfall trap.

A 22.5% variation was observed in the data using the

PCoA, which can be explained by the differences between
the methods used in sampling this community. Two groups
of species were defined, one that predominated in the
sampling with the mini-Winkler extractor and the other
with the pitfall trap (Fig 4).

Discussion
The efficiency of the complementariness between the two
collection methods applied in the present study for the
observed and estimated richness among the ant community was previously tested in an ant diversity study on the
island of Madagascar (Fisher et al 2000) and were essential
to reinforce the importance of complementariness among
these methods. Studies that proved that a number of
species were collected by the mini-Winkler extractor than
by the pitfall trap in the Brazilian Atlantic Rainforest when
efficient sampling of ant communities is desired (Delabie et
al 2000b, Orsolon-Souza et al 2011).
Our data for the Pocon Pantanal region corroborated
findings by Delabie et al (2000b) because the mini-Winkler
is a method that samples a greater number of Formicidae
species, especially small sized and the cryptobiotic ones,

356

Silva et al

Fig 4 Indirect ordination axis

of the ant community sample
in the phytophysiognomical
mosaic of the Pocon Pantanal,
Mato Grosso, Brasil, using the
mini-Winkler extractor and
pitfall trap methods. Captions:
filled diamond mini-Winkler,
plus sign pitfall trap.

such as Hypoponera spp., Strumigenys denticulata (Mayr),

Strumigenys subdentata (Mayr), and Oxyepoecus vezenyii
(Forel). This data demonstrates the need of its use when
collecting in the soils in the Mato Grosso Pantanal. However, ants with large and wide dispersion throughout the
environment can be subsampled by this method. This
limitation can, however, be compensated by the use of
the pitfall trap because this method depends on the daily
foraging activity of the ants (Adis 2002).
According to Bestelmeyer et al (2000), the pitfall trap
also has limitations when sampling slow-moving species
and those with cryptobiotic behavior. Thus, it was observed that the limitations of one method can be mitigated
by the attributes of another.
The results obtained by the ChaoJaccard similarity index showed that the two methods used collected partially
similar groups of ants. The similarity observed was justified
by the generalist characteristics, dominance, and wide
distribution of some species found in the area sampled.
For example, Brachymyrmex cf. heeri had 12 records with
the mini-Winkler extractor and 13 with the pitfall traps (see
ESM). This occurrence was justified by the underground
opportunistic habit of the ant that constructs its nests in
various locations and forages over extensive areas (Fernndez
2003, Silvestre et al 2003).
In addition to this species, S. invicta, Solenopsis
(gr. globularia) sp.1, Nylanderia sp.1, Pheidole (gr. biconstricta) sp.1, and Camponotus abstinens (Forel), considered
generalists and dominant in the soil and vegetation), were
also sampled by the two methods (Fernndez 2003,
Silvestre et al 2003).
However, the PCoA and the similarity index reflected
the difference in species composition between the methods used. For example, Hypoponera (5 spp.) was sampled
only with the mini-Winkler extractor while Dorymyrmex (2),

and Pogonomyrmex naegelii (Emery) were captured very

frequently by the pitfall traps (see ESM). Hypoponera are
common inhabitants of decaying wood, trunks, and hollow
branches fallen on the ground and especially of the litter
(Lattke 2003) which favors its capture by the mini-Winkler
extractor.
In addition, Cyphomyrmex sp.3, Cyphomyrmex sp.4,
Gnamptogenys striatula (Mayr), S. denticulata Mayr, S.
subdentata Mayr, Rogeria blanda (Smith), Trachymyrmex
fuscus (Emery), and Trachymyrmex iheringi (Emery) were
sampled exclusively with the mini-Winkler extractor (see
ESM). These taxa are frequently cryptic, litter inhabitants
(Fernndez 2003, Silvestre et al 2003), and this method
was especially efficient in sampling habitats with greater
litter accumulation.
The cryptobiotic O. vezenyii (Forel) was exclusively sampled by the mini-Winkler extractor (see ESM), and the
frequency crytobiotic species have been collected in fauna
surveys has increased because of the development of litter
extraction techniques (Albuquerque & Brando 2004,
Castilho et al 2007).
Species collected mainly by pitfall traps, such as Dorymyrmex sp.1, Dorymyrmex (gr. Pyramicus) sp.1, and P.
naegelii, build their nests in the soil in arid and semiarid
regions, preferring open habitats with little plant cover
(Cuezzo 2003, Fernndez 2003). Thus the higher capture
frequencies of these taxa with pitfall traps corroborated
data by Fisher (1999), who considered this method ideal for
arid regions and open areas with little plant coverage.
In addition to these ants, Acromyrmex fracticornis
(Forel), Acromyrmex subterraneus (Forel), and Atta laevigata (Smith) were sampled only by pitfall traps. These
species are common inhabitants of open Cerrado (central
Brasil savanna) areas with little litter (Silvestre et al 2003),
similar to the field areas we sampled.

Complementary Methods to Sample Formicidae

The association of techniques used in the present study

was more efficacious in sampling in regions such as the
Pocon Pantanal because the mini-Winkler extractor favored species collection in habitats with accumulated litter
while the pitfall trap favored sampling in locations with
little plant cover and thin litter. Therefore, pitfall traps
should be used in association with mini-Winkler extractors
also because pitfall traps are efficient in sampling in open
field areas, unlike the extractor that requires the accumulation of litter for a good performance in the sampling of
soil ants. Thus, combined the use of these sampling methods contemplates the characteristic of the environment on
the spatial scale assessed and the bioecological peculiarities of ants in the region studied.
Other studies, such as in Costa Rica, have reported that
pitfall traps captured generalist species with a larger body
size, which occurs less frequently when methods with litter
extraction are used (Olson 1991). This pattern was also
observed in the present study, for example, for the Ponerinae Odontomachus bauri (Emery), Odontomachus brunneus (Patton), and Pachycondyla harpax (Fabricius).
Because of their body sizes (1 cm or more) and the diameter of the pitfall trap, these ants were exclusively collected by this technique.
However, if the study is to focus only on open areas
such as natural fields and introduced grass pastures, the
pitfall trap associated to bait and hand collection is an
efficient method for a satisfactory sampling, as long as
traps remains in the field for a period longer than 48 h.
However, if sampling is carried out for a short period of
time, the mini-Winkler extractor is the suggested method
to be associated with pitfall traps for the efficient collection
of the highest number of species and functional groups.
Assessment of the complementariness of the two methods in forested and open areas using the concept of functional groups showed that the results between the
methods differed in the composition of the genera collected. For example, predator and generalist ants were more
frequently collected by pitfall traps in open areas. Thus, the
mini-Winkler extractor favors smaller and less active species among the functional groups while the pitfall trap
favors those that are larger and mobile (see ESM).
It was observed, therefore, that the results of the present research are important for knowledge on the soil and
little ant community in the Pantanal biome, since this was
the first study to use the association of the pitfall trap and
mini-Winkler extractor collection methods for this region.
Acknowledgments This paper is dedicated to the memory of the late
Professor Dr. Joachim Adis for his dedication to the study of arthropods in
the Pantanal of Mato Grosso who passed his knowledge on to the
teaching staff, employees, and students of the Laboratrio de Taxonomia
e Ecologia de Artrpodes, Universidade Federal do Mato Grosso. The
authors also thank Professor Dr. Carlos Roberto Ferreira Brando,

357
Dr. Rodrigo Feitosa, and Dr. Rogrio Rosa da Silva for their help in the
identification of the specimens sampled here. The authors also thank the
students and the technician of the Francisco de Assis Rondon of the
Laboratrio de Taxonomia e Ecologia de Artrpodes, Universidade Federal
do Mato Grosso for their help in the field collections. We finally thank the
Programa de Ps-graduao em Ecologia e Conservao da Biodiversidade
do Instituto de Biocincias da UFMT, Centro de Pesquisas do Pantanal
(CPP), Ncleo de Pesquisas do Pantanal (NEPA), and the Programa de
Pesquisas em Biodiversidade (PPBio) for logistic help. Thanks are due to
the CNPq for the fellowships and research grants to the first two authors.
Electronic supplementary material The online version of this article
(doi:10.1007/s13744-013-0131-7) contains supplementary material, which
is available to authorized users.

References
Adis J (2002) Recommended sampling techniques. In: Adis J (ed)
Amazonian Arachnida and Myriapoda. PENSOFT, Sofia, Moscow, pp
555576
Alonso LE, Agosti D (2000) Biodiversity studies, monitoring and ants:
an overview. In: Agosti D, Majer JD, Alonso LE, Schultz TR (eds)
Ants: standard methods for measuring and monitoring biodiversity.
Smithsonian Institution Press, Washington, pp 18
Albuquerque NL, Brando CRF (2004) A revision of the Neotropical
Solenopsidini ant genus Oxyepoecus Santschi, 1926 (Hymenoptera:
Formicidae: Myrmicinae). 1. The Vezenyii species-group. Pap
Avulsos Zool 44(4):5580
Anderson MJ (2003) PCO: a FORTRAN computer program for principal
coordinate analysis. Department of Statistics, University of
Auckland, New Zealand
Baccaro FB, Ketelhut SM, Morais JW (2011) Efeitos da distncia entre iscas
nas estimativas de abundncia e riqueza de formigas em uma floresta
de terra-firme na Amaznia Central. Acta Amazon 41(1):115122
Baroni-Urbani C, Andrade ML (2007) The ant tribe Dacetini: limits and
constituent genera, with descriptions of new species
(Hymenoptera, Formicidae). Annali del Museo Civico di Storia
Naturale "Giacomo Doria" 95:1191
Bestelmeyer BT, Agosti D, Leeanne E, Alonso T, Brando CRF, Brown
WL, Delabie JHC, Silvestre R (2000) Field techniques for the study
of ground dwelling ants: an overview, description, and evaluation.
In: Agosti D, Majer JD, Alonso LE, Schultz TR (eds) Ants: standard
methods for measuring and monitoring biodiversity. Smithsonian
Institution Press, Washington and London, pp 122144
Brando CRF, Silva RR, Delabie JHC (2009) Formigas (Hymenoptera).
In: Panizzi AR, Parra JRP (eds) Bioecologia e Nutrio de Insetos.
Base para o Manejo Integrado de Pragas. Braslia, Embrapa
Informao Tecnolgica, pp 323369
Bolton B, Alpert G, Ward PS, Naskrecki P (2006) Boltons catalogue of ants
of the world 17582005. Harvard University Press, CD-ROM, Cambridge
Castilho ACC, Delabie JHC, Marques MI, Adis J (2007) Novos registros
de Creightonidris scambognatha Brown (Hymenoptera, Formicidae),
formiga criptobitica endmica do Brasil. Neotrop Entomol 36(1):150
152
Chao A, Chazdon RL, Colwell RK, Shen TJ (2005) A new statistical
approach for assessing similarity of species composition with incidence and abundance data. Ecol Lett 8(2):148159
Cowell RK (2005) EstimateS: statistical estimation of species richness
and shared species from sample. Version 7.5
Cuezzo F (2003) Subfamilia Dolichoderinae. In: Fernndez F (ed)
Introduccin a Las Hormigas de la Regin Neotropical. Instituto
de Investigacin de Recursos Biolgicos Alexander von Humboldt,
Bogot, Colombia. XXVI+, pp 291297

358
Delabie JHC, Fisher BL, Majer JD, Wright IW (2000a) Sampling effort and
choice of methods. In: Agosti D, Majer JD, Alonso LE, Schultz TR (eds)
Ants: standard methods for measuring and monitoring biodiversity.
Smithsonian Institution Press, Washington and London, pp 145154
Delabie JHC, Agosti D, Nascimento IC (2000b) Litter ant communities of
the Brazilian Atlantic rain forest region. In: Agosti D, Majer JD, Alonso
LE, Schultz TR (eds) Sampling ground-dwelling ants: case studies from
the worlds rain forests. Bulletin no. 18. Curtin University School of
Environmental Biology, Perth, Australia, pp 117
Fernndez F (2003) Subfamilia Myrmicinae. In: Fernndez F (ed)
Introduccin a Las Hormigas de la Regin Neotropical. Instituto
de Investigacin de Recursos Biolgicos Alexander von Humboldt,
Bogot, Colombia. XXVI+ pp 307330
Fisher BL (1998) Ant diversity patterns along an elevation gradient in
the Rserve Spciale dAnjanaharibe-Sub and on the western
Masoala Peninsula, Madagascar. Fieldiana Zool 90:3967
Fisher BL (1999) Improving inventory efficiency: a case study of leaflitter ant diversity in Madagascar. Ecol Appl 9(2):714731
Fisher BL, Malsch AKF, Gadagkar R, Delabie JHC, Vasconcelos HL,
Majer JD (2000) Applying the ALL Protocol. In: Agosti D, Majer
JD, Alonso LE, Schultz TR (eds) Ants: standard methods for measuring and monitoring biodiversity. Smithsonian Institution Press,
Washington and London, pp 207214
Hughes RG (1986) Theories and models of species abundances. Am
Nat 128:879899
LaPolla JS, Brady SG, Shattuck SO (2010) Phylogeny and taxonomy of
the Prenolepis genus-group of ants (Hymenoptera: Formicidae).
Syst Entomol 35:118131
Lattke JE (2003) Subfamilia Ponerinae. In: Fernndez, F (ed)
Introduccin a Las Hormigas de la Regin Neotropical. Instituto
de Investigacin de Recursos Biolgicos Alexander von Humboldt,
Bogot, Colombia. XXVI+, pp 261276
Longino J (2000) What to do with the data. In: Agosti D, Majer JD,
Alonso LE, Schultz TR (eds) Ants: standard methods for measuring
and monitoring biodiversity. Smithsonian Institution Press,
Washington and London, pp 186203
Lourival R, Harris M, Montambault JR (2000) Introduo ao Pantanal,
MS, Brasil. In: Willink PW, Chernoff B, Alonso LE, Montambault JR,
Lourival R (eds) Rap Bulletin of Biological Assessment. A Biological
Assessment of the Aquatic Ecosystems of the Pantanal, Mato Grosso
do Sul, Brazil. Conservation International. Washington, pp 146151
Magnusson WE, Lima AP, Luizo R, Luizo F, Costa FRC, Castilho CV,
Kinupp VF (2005) RAPELD: uma modificao do mtodo de Gentry
para inventrios de biodiversidade em stios para pesquisa
ecolgica de longa durao. Biota Neotropica 5(2):16
Nunes-da-Cunha C, JUNK WJ, Leito-Filho H (2007) Woody vegetation
in the Pantanal of Mato Grosso, Brazil: a preliminary tipology.
Amazoniana 19(3/4):159184

Silva et al
Nunes-da-Cunha C, Rebellato L, Costa CP (2010) Vegetao e flora:
experincia pantaneira no sistema de grade. In: Fernandes IM,
Signor CA, Penha J. (eds) Biodiversidade no Pantanal de Pocon.
Cuiab: Centro de Pesquisa do Pantanal, pp 3758
Olson MD (1991) A comparison of the efficacy of litter sifting and
pitfall traps for sampling leaf litter ants (Hymenoptera:
Formicidae) in a tropical wet forest, Costa Rica. Biotropica
23(2):166172
Orsolon-Souza G, Esbrard CEL, Mayh-Nunes AJ, Vargas AB, VeigaFerreira S, Folly-Ramos E (2011) Comparison between Winklers
extractor and pitfall traps to estimate leaf litter ants richness
(Formicidae) at a rainforest site in southeast Brazil. Braz J Biol
71(4):873880
PCBAP (2007) Plano de Conservao da Bacia do Alto Paraguai.
Metodologia do Plano de Conservao da Bacia do Alto Paraguai.
Projeto Pantanal, Programa Nacional do Meio Ambiente. Braslia:
MMA/SEMAM/PNMA, v.3, 370 p
Prance GT, Schaller GB (1982) Preliminary study of some vegetation
types of the Pantanal, Mato Grosso, Brazil. Brittonia 34:228251
Romero H, Jaff K (1989) A comparision of methods for sampling ants (Hymenoptera, Formicidae) in savanas. Biotropica 21:
348352
Santos SA, Comastri-Filho JA, Cardoso EL (2004) Princpios de agroecologia no manejo das pastagens nativas do Pantanal. Embrapa
Pantanal, Corumb, 35 p
Sarmiento CE (2003) Metodologas de captura y estudio de las hormigas. In: Fernndez, F (ed.). In: Fernndez F (ed) Introduccin a
Las Hormigas de la Regin Neotropical. Instituto de Investigacin
de Recursos Biolgicos Alexander von Humboldt, Bogot,
Colombia. XXVI+, pp 201210
Silva MP, Mauro R, Mouro G, Coutinho M (2000) Distribuio e
quantificao de classes de vegetao do Pantanal atravs de
levantamento areo. Rev Bras Bot 23:143152
Silvestre R, Brando CRF, Silva RR (2003) Grupos funcionales de
hormigas: El caso de los gremios del Cerrado, Brasil. In:
Fernndez, F (ed) Introduccin a Las Hormigas de la Regin
Neotropical. Instituto de Investigacin de Recursos Biolgicos
Alexander von Humboldt, Bogot, Colombia. XXVI+, pp 113143
Souza JLP, Baccaro FB, Landeiro VL, Franklin E, Magnusson WE (2012)
Trade-offs between complementarity and redundancy in the use of
different sampling techniques for ground-dwelling ant assemblages. Appl Soil Ecol 56:6373
Spence JR, Niemel J (1994) Sampling carabid assemblages with
pitfall traps: the madness and the method. Can Entomol
126:881894
Veloso HP, Rangel-Filho AL, Lima JCA (1991) Classificao da
vegetao brasileira, adaptada a um sistema universal. IBGE,
Rio de Janeiro