SciELO - Scientific Electronic Library Online

vol.23 número1Daily patterns of activity in free-living Rio Negro tuco-tucos (Ctenomys rionegrensis)New observations of living Echimys saturnus (dark tree rat, Echimyidae) from eastern Ecuador índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




  • No hay articulos citadosCitado por SciELO

Links relacionados


Mastozoología neotropical

versión impresa ISSN 0327-9383versión On-line ISSN 1666-0536

Mastozool. neotrop. vol.23 no.1 Mendoza jun. 2016



Population seasonal variation of Gracilinanus agilis (Mammalia: Didelphidae) in semi-deciduous forest fragments


Wellington Hannibal1,2, Valquiria V. Figueiredo1, and Nicolay L. da Cunha2

1 Laboratório de Ecologia e Biogeografia de Mamíferos, Universidade Estadual de Goiás, Avenida Brasil, nº 435, CEP 75860-000, Quirinópolis, GO, Brazil. [Correspondence: Wellington Hannibal <>]
2 Programa de Pós-Graduação em Ecologia e Conservação, Universidade Federal de Mato Grosso do Sul, Avenida Costa e Silva, s/n, CEP 79070-900, Campo Grande, MS, Brazil.

Recibido 15 octubre 2015.
Aceptado 15 febrero 2016.
Editor asociado: A Farías


Population fluctuations of small tropical mammals are often related to temporal and environmental variations in resource availability. Here, we aim to associate population variation of Gracilinanus agilis with seasonality, and type and position of traps in fragments of semi-deciduous forest in Southern Goiás, Brazil. We conducted a mark-recapture study from August 2013 through August 2014. We set wire-cage and Sherman® traps in four transects, with five stations per transect, during three consecutive nights per month. Each station consisted of 1 wire cage and 1 Sherman® trap, placed alternately on the ground and in the understory (1.5- 2 m height). We registered 66 individuals in 121 captures (7.8% capture success). Gracilinanus agilis was more abundant during the dry season in the understory. Males were more abundant in the dry season and females in the wet season. Data reported here are of great value for understanding population and behavioral patterns of G. agilis in the Brazilian savanna.


Variación poblacional estacional de Gracilinanus agilis (Mammalia: Didelphidae) en fragmentos de bosque semideciduo.

Las fluctuaciones poblacionales de los pequeños mamíferos tropicales suelen estar relacionados con las variaciones temporales y ambientales en la disponibilidad de recursos. Aquí, nuestro objetivo es asociar la variación poblacional de Gracilinanus agilis con la estacionalidad y el tipo y la posición de las trampas en fragmentos de bosque semideciduo en el sur de Goiás, Brasil. Realizamos el muestreo a través del método de marcado y recaptura, desde agosto de 2013 a agosto de 2014. Establecimos cinco estaciones de captura por transecto, durante tres noches consecutivas por mes. Cada estación de trampas consistió en una jaula de alambre y una trampa Sherman®, colocadas alternativamente en el suelo y sotobosque (1.5-2 m de altura). Se registraron 66 individuos en 121 capturas, lo que resulta en un éxito captura de 7.8%. Gracilinanus agilis fue más abundante en la estación seca, en el sotobosque. Los machos fueron más abundantes en la estación seca y las hembras en la estación húmeda. Los datos reportados aquí son de gran valor para la comprensión de los patrones de población y de comportamiento de G. agilis en la sabana brasileña.

Key words: Age class; Dry season; Sex class; Sherman® traps; Understory.

Palabras clave: Clase de edad; Clase de sexo; Estación seca; Sotobosque; Trampas Sherman®.


Population fluctuations of small tropical mammals are often related to temporal and environmental variations in resource availability (Bergallo and Magnusson, 1999; Santos-Filho et al., 2008; Andreazzi et al., 2011; Gentile et al., 2012). Variations in rainfall between seasons and years influence the composition and abundance of available food (Bergallo and Magnusson, 1999; Santos-Filho et al., 2008). Thus, species capable of shaping their diet according to such variations are expected to show minor fluctuations in abundance between seasons, while species with restricted or specialized diets may suffer because of resource limitation, which contributes significantly to the population size variance between seasons (Bergallo and Magnusson, 1999; Santos-Filho et al., 2008).

The fluctuation of population density of G. agilis has been associated with the wet season, a period with higher abundance of fruits and arthropods (Mares and Ernest, 1995; Camargo et al., 2014a). However, studies with Marmosops noctivagus, Micoureus demerare and G. agilis in seasonal semi-deciduous for­est showed higher abundance during the dry period, but the highest availability of the pri­mary food of this group-i.e. arthropods-was recorded during the wet season (Santos-Filho et al., 2008).

In addition, the seemingly inconsistent rela­tionship between population density of small mammals and resource availability can also be a result of sampling artifact. Traps with baits have reduced attractiveness during the wet season (Santos-Filho et al., 2008). Furthermore, most Neotropical marsupial genera are recognized as arboreal or scansorial (Vieira and Camargo, 2012). Thus, variation of the type and position of traps could distort estimates of population abundance and community composition (Astúa et al., 2006; Santos-Filho et al., 2006; Cáceres et al., 2011; Hice and Velazco, 2013).

Deviations in the sex ratio also appear to be related to environmental variation between seasons, as males of the marsupial Thylamys macrurus are more abundant in the dry season and females more abundant in the wet season (Cáceres et al., 2007). This pattern also occurs in G. agilis populations, though this can be a result of life history traits and a predisposition of marsupials toward semelparity (Lopes and Leiner, 2015).

Most of the studies within species of Gracilinanus have focused on diet composition (Bocchiglieri et al., 2010), variation of feeding habits related to sex, reproductive condition, and seasonality (Camargo et al., 2014b), as well as variations of niche between seasons (Camargo et al., 2014a). However, fundamental data about population variation of this species with respect to seasonality are scarce.

Gracilinanus agilis is a small marsupial with body mass ranging from 13 to 40 g, head and body length between 81 and 115 mm, length of tail always greater than the body, ranging between 110 and 158 mm (Rossi and Bianconi, 2011). This species is classified as insectivorous-omnivorous, with fruits being an important part of their diet. Although it feeds on abundant insects in the environment, there is a higher preference for termites and bedbugs, and a lesser preference toward ants during the dry season. During the wet season, this species tends to feed according to availability of resources in the environment (Camargo et al., 2014b).

We investigated variation in the structure of populations of G. agilis during dry and wet seasons in two fragments of semi-deciduous seasonal forest in the Brazilian savanna. We address the following questions in our study: (1) is there a difference of abundance of G. agilis between seasons, and type and position of traps? We predicted that G. agilis would be more abundant in the dry season, because of life history traits, including a semelparous reproductive strategy (Lopes and Leiner, 2015). We expected the species to be most commonly captured in Sherman® traps in the understory, because of its small body size and arboreal habits (Voss et al., 2005; Cáceres et al., 2011). (2) Does the abundance of G. agilis vary seasonally and distinctly between sexes and age classes? Because marsupials are known to exhibit seasonal variation in population density as a result of a well-defined breeding season (Cáceres et al., 2007; Gentile et al., 2012), we expected to find higher abundance of young individuals during the wet season, with females being more common than males.


Study area

Samples were taken in two fragments of semi-deciduous forest (F1: 65.11 ha-18° 24' 51" S and 50° 38' 46" W, and F2: 48.15 ha-18° 26' 08" S and 50° 38' 46" W), which were 2792 m from each other, in the municipality of Quirinópolis, central-west of Brazil. The climate is Tropical semi-humid, markedly seasonal, with dry winter (Aw in KÖppen climate classification), with mean annual temperatures of 23 °C (max. 40 °C and min. 14 °C), and mean annual rainfall of around 1600-1900 mm, irregularly distributed in two distinct seasons. The dry season occurs between April and September, and wet season from October to March (Alvares et al., 2014).


Live traps were set along four transects (two per fragment) during three consecutive nights each month, from August 2013 to August 2014. We installed five trap-stations, 15 m apart, in each transect. Each trap-station consisted of a wire-cage and a Sherman® trap, alternating between placements on the ground and in the understory (1.5-2 m height). We used a mixture of banana, bacon and cod liver oil as bait. We checked the traps daily, and all captured individual were individually marked with numbered ear-tags. This way, we were able to get the type of trap (cage or Sherman®) and catching position (ground or understory). We also recorded the sex (male or female) and age (young or adult). Individuals with incomplete dentition (less than four upper molars) and deciduous third superior premolars were clas­sified as young (Rossi et al., 2012). Three specimens were collected and deposited in the collection of small mammals of the Laboratory of Ecology and Biogeography of Mammals of the Universidade Estadual de Goiás, Campus Quirinópolis (Appendix 1). The captures were made with the collection permit SISBIO nº 37519/2013.

Data Analysis

Abundance and capture success were used to determine population patterns of G. agilis. The number of captured individuals was our proxy for abundance. Capture success is the number of captures divided by the sampling effort multiplied by 100. To verify differences in the total abundance of G. agilis between the periods of the year (dry and wet seasons), type (cage and Sherman® traps), and position (ground and understory) of traps, we used chi-square test with Yates correction. To test the dependence of the sex and age classes as a function of seasonality, we used Fisher's exact test. We used the Holm-Bonferroni method to correct the P-values for multiple non-independent comparisons.


From August 2013 to August 2014, 66 individuals of G. agilis were captured, including 37 males and 29 females, during 1560 trap-nights (capture success of 7.8%). Gracilinanus agilis was more commonly captured in the dry season, in the understory and using Sherman® traps (Table 1). Males were more abundant in the dry season and females in the wet season (P = 0.03, Fig. 1). However, the abundance of young and adults was not related to seasonality (P = 0.72, Fig. 1).

Table 1 Capture success of Gracilinanus agilis in different seasons, and relative to the type and position of traps in fragments of seasonal forest in southern Goiás.

Fig. 1. Comparisons of abundance (number of captured individuals) of Gracilinanus agilis between sex and age class in seasonal forest fragments in southern Goiás.


General patterns of the population

Gracilinanus agilis was more abundant during the dry season and captured more often in Sherman® traps set in the understory, confirming our initial predictions. This result is concordant with other studies done in areas of the Brazilian Cerrado (Lopes and Leiner, 2015) and Pantanal wetland (Aragona and Marinho-Filho, 2009; Andreazzi et al., 2011). Mating of G. agilis begins in July, and around August-September all females are pregnant or lactating, which significantly increases the number of individuals during the dry season, i.e. between June and October (Lopes and Leiner, 2015). However, in gallery forests of Cerrado in central Brazil, peaks of abundance of G. agilis occurred during the wet season and population declined during the dry season (Mares and Ernest, 1995). Beyond seasonality, local factors such as competition with mor­phologically similar species (e.g. Cryptonanus agricolai), habitat heterogeneity (gallery forest, woodland savannah, semi-deciduous forest) and resource availability may play important roles in the variation of abundance of G. agilis. In woodland savannah fragments, G. agilis and C. agricolai tended to occupy the ground and the understory, respectively (Hannibal and Cáceres, 2010). This difference may be explained by competition (Vieira and Camargo, 2012). The peak of individuals of G. agilis in the dry season occurred in woodland savannah and semi-deciduous forest (Andreazzi et al., 2011; Lopes and Leiner, 2015), while the peak in wet season occurred in the gallery forest (Mares and Ernest, 1995). Gracilinanus agilis feeds heavily on Isoptera and Hemiptera dur­ing the dry season, showing that these insects are relevant for its diet, especially in cool-dry season (Camargo et al., 2014b). Therefore, the availability and composition of resources are important to determine the variation in abundance of G. agilis.

We had a higher number of captures in Sherman® traps set in the understory than in cage traps in any other position within the fragments. Many other studies have reported similar results (Santos-Filho et al., 2006; Aragona and Marino-Filho, 2009; Andreazzi et al., 2011). Gracilinanus agilis is a species with arboreal habits (Paglia et al., 2012) and commonly uses the arboreal or semi-arboreal forest layer (Voss et al., 2005). Cage traps appear to be more efficient in the capturing of larger species like Didelphis albiventris and Thrychomys pachyurus (Andreazzi et al., 2011; Cáceres et al., 2011).

Sex and age class composition

We found deviations in the sex ratio of captured individuals between seasons, with males being more commonly captured during the dry season and females more frequently in the wet season. This result may be associated with the life history traits, reproductive patterns and predisposition toward semelparity in G. agilis (Aragona and Marinho-Filho, 2009; Lopes and Leiner, 2015). At the end of the dry season, as is typical of didelphids, there is an increase in activity of males searching for females for breeding (Cáceres et al., 2007; Aragona and Marinho-Filho, 2009; Andreazzi et al., 2011).

Right after mating, there is a drastic reduction in the number of adult males, which can result in a complete disappearance of these individuals between November and December-wet season (Lopes and Leiner, 2015). Further, females tend to increase their activity in the wet season to find enough resources for breastfeeding as well as parental care and weaning of the litter (Cáceres et al., 2007; Lopes and Leiner, 2015). There is an association between age class and season, as the abundance of adults is clearly higher in the dry season. Our results contrast with those found elsewhere, which report that juveniles are predominantly captured in the wet season (Mares and Ernest, 1995; Lopes and Leiner, 2015). One possible explanation is trap bias favoring adult trappings. Another conjecture may be the increased abundance associated with low reproductive activity, which could indicate immigration (Andreazzi et al., 2011).


This study shows strong evidence of popula­tion change over seasons, especially sex ratio which was inversely abundant in each season. Although our study was performed in a small geographic area, data reported here are of great value for understanding population and behavioral patterns of G. agilis in the Brazilian Savannah. Intensive studies evaluating the use of ground, understory and canopy levels by G. agilis are, to the best of our knowledge, non-existent. Further studies should test 1) the effect of seasonality on the use of vertical strata by G. agilis, and 2) the variation of population structure on different vertical strata. Such approaches may be of great importance for better understanding patterns of vertical strata habitat use and its variation in population structure.


We thank members of the Laboratory of Ecology and Biogeography of Mammals for their help in fieldwork, and J. L. Holladay for the English revision. We are also indebted to the Curso de Ciências Biológicas, UEG, Câmpus Quirinópolis for logistical support, and to the Programa de Iniciação Científica e Tecnológica (PIC&T), UEG, for sponsoring V.V. Figueiredo.


1. ALVARES CA, JL STAPE, PC SENTELHAS, JLM GONÇALVES, and G SPAROVEK. 2013. KÖppen's climate classification map for Brazil. Meteorologische Zeitschrift 22(6):711-728.         [ Links ]

2. ANDREAZZI CS, V RADEMAKER, R GENTILE, HM HERRERA, AM JANSEN, and PS D'ANDREA. 2011. Population ecology of small rodents and marsupials in a semi-deciduous tropical forest of the southeast Pantanal, Brazil. Zoologia 28(6):762-770.         [ Links ]

3. ARAGONA M and J MARINHO-FILHO. 2009. História natural e biologia reprodutiva de marsupiais no Pantanal Mato Grosso, Brasil. Zoologia 26(2):220-230.         [ Links ]

4. ASTÚA D, RT MOURA, CEV GRELLE, and MT FONSECA. 2006. Influence of baits, trap type, and position for small mammal capture in a Brazilian lowland Atlantic Forest. Boletim do Museu de Biologia Mello Leitão. 19:31-44.         [ Links ]

5. BERGALLO HG and WE MAGNUSSON. 1999. Effects of climate and food availability on four rodent species in southeastern Brazil. Journal of Mammalogy. 80(2):472-486.         [ Links ]

6. BOCCHIGLIERI A, AF MENDONÇA, and JB CAMPOS. 2010. Diet composition of Gracilinanus agilis (Didelphimorphia, Didelphidae) in dry woodland areas of Cerrado in central Brazil. Mammalia 74:225-227.         [ Links ]

7. CÁCERES NC, RP NAPOLI, and W HANNIBAL. 2011. Differential trapping success for small mammals using pitfall and standard cage traps in a woodland savannah region of southwestern Brazil. Mammalia 75:45-52.         [ Links ]

8. CÁCERES NC, RP NAPOLI, WH LOPES, J CASELLA, and GS GAZETA. 2007. Natural history of the marsupial Thylamys macrurus (Mammalia, Didelphidae) in fragments of savannah in southwestern Brazil. Journal of Natural History 41:1979-1988.         [ Links ]

9. CAMARGO NF, JF RIBEIRO, AJA CAMARGO, and EM VIEIRA. 2014a. Intra- and inter-individual variation show distinct trends as drivers of seasonal changes in the resource use of a neotropical marsupial. Biological Journal of the Linnean Society 111:737-747.         [ Links ]

10. CAMARGO NF, JF RIBEIRO, AJA CAMARGO, and EM VIEIRA. 2014b. Diet of the gracile mouse opossum Gracilinanus agilis (Didelphimorphia: Didelphidae) in a neotropical savanna: Intraspecific variation and resource selection. Acta Theriologica 59:183-191.         [ Links ]

11. GENTILE R, BR TEIXEIRA, and HG BERGALLO. 2012. Dinâmica populacional de marsupiais brasileiros. Pp. 309-324, in: Os marsupiais do Brasil: biologia, ecologia e conservação (NC Cáceres, org.). Editora UFMS. Campo Grande.         [ Links ]

12. HICE CL and PM VELAZCO. 2013. Relative effectiveness of several bait and trap types for assessing terrestrial small mammal communities in Neotropical rainforest. Museum of Texas Tech University 313:1-16.         [ Links ]

13. LOPES GP and NO LEINER. 2015. Semelparity in a population of Gracilinanus agilis (Didelphimorphia: Didelphidae) inhabiting the Brazilian Cerrado. Mammalian Biology 80:1-6.         [ Links ]

14. HANNIBAL W and NC CÁCERES. 2010. Use of vertical space by small mammals in gallery forest and woodland savannah in south-western Brazil. Mammalia 74:247-255.         [ Links ]

15. MARES MA and KA ERNEST. 1995. Population and community ecology of small mammals in a gallery forest of central Brazil. Journal of Mammalogy 76(3):750-768.         [ Links ]

16. PAGLIA AP, GAB FONSECA, AB RYLANDS, G HERRMANN, LMS AGUIAR, AG CHIARELLO, YLR LEITE, LP COSTA, S SICILIANO, MCM KIERULFF, SL MENDES, VC TAVARES, RA MITTERMEIER, and JL PATTON. 2012. Annotated Checklist of Brazilian Mammals. 2ª Ed. Occasional Papers in Conservation Biology, nº 6. Conservation International, Arlington, VA.         [ Links ]

17. ROSSI RV and GV BIANCONI. 2011. Ordem Didelphimorphia. Pp. 31-69, in: Os mamíferos do Brasil (NR Reis, AL Peracchi, WA Pedro, and IP Lima, eds). 2ª Ed. Londrina: Nelio R. dos Reis.         [ Links ]

18. ROSSI RV, MV BRANDÃO, AP CARMIGNOTTO, CL MIRANDA, and JJ CHEREM. 2012. Diversidade morfológica e taxonômica de marsupiais didelfídeos, com ênfase nas espécies brasileiras. Pp. 23-72, in: Os marsupiais do Brasil: biologia, ecologia e conservação (NC Cáceres, org.). Editora UFMS. Campo Grande.         [ Links ]

19. SANTOS-FILHO M, DJ SILVA, and TM SANAIOTTI. 2006. Efficiency of four trap types in sampling small mammals in forest fragments, Mato Grosso, Brazil. Mastozoología Neotropical 13:217-225.         [ Links ]

20. SANTOS-FILHO M, DJ SILVA, and TM SANAIOTTI. 2008. Variação sazonal na riqueza e na abundancia de pequenos mamíferos, na estrutura da floresta e na disponibilidade de artrópodes em fragmentos florestais no Mato Grosso, Brasil. Biota neotropical 8:115-121.         [ Links ]

21. VIEIRA EM and NF CAMARGO. 2012. Uso do espaço vertical por marsupiais brasileiros. Pp. 345-362, in: Os marsupiais do Brasil: biologia, ecologia e conservação (NC Cáceres, org.). Editora UFMS. Campo Grande.         [ Links ]

22. VOSS RR, DP LUNDE, and SA JANSA. 2005. On the contents of Gracilinanus Gardner and Creighton, 1989, with the description of a previously unrecognized clade of small didelphid marsupials. American Museum of Natural History 3482:1-34.         [ Links ]

Specimens of Gracilinanus agilis collected in a semi-deciduous seasonal fragment, southern Goiás, Brazil. Measurements: W = weight in g and length in mm-HB = head-body, T = tail, HF = hind foot, E = ear.
27.VIII.13, [18°25'44"S; 50°40'32"W], male, W = 45 g, HB = 110.5, T = 133.5, HF = 14.3, E = 20.7, captured in cage trap, on the understory, Hannibal et al. cols. (ZUEG-007). 27.VIII.13, [18°24'58"S; 50°39'22"W], female, W = 29 g, HB = 105.9, T = 126.5, HF = 11.6, E = 20.9, captured in Sherman® trap, on the understory, Hannibal et al. cols. (ZUEG-008). 29.VIII.13, [18°25'33"S; 50°40'23"W], female, W = 28 g, HB = 98.1, T = 108.5, HF = 12.1, E = 21.8, captured in cage trap, on the ground, Hannibal et al. cols. (ZUEG-012).


Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons