There are 16 families, 78 genera and approximately 300 living species of marsupials worldwide, with a great diversity of food guilds and habitats [1]. The order Didelphimorphia groups 19 genera and 92 species [2,3], of which 12 have been recorded as vulnerable or at critical risk of disappearance, due to habitat fragmentation and other forms of human disturbance [3]. In the Americas, their distribution ranges from southern Canada to northern Patagonia in Argentina, including some Caribbean islands [4], being present in most Neotropical ecosystems between 3,000 and 4,200 m.a.s.l. [5] and even in vegetation relics of anthropized environments [6]. In the ecosystems where they live, they perform important ecological functions as seed dispersers, consumers of invertebrates and small vertebrates, and as prey for large and small carnivores [2]. 

In Mexico, the order Didelphimorphia is represented by two subfamilies: Caluromyinae, with one medium-sized species (Caluromys derbianus), and Didelphinae, with five medium-sized species (Chironectes minimus, Didelphis marsupialis, Didelphis virginiana, Metachirus nudicaudatus and Philander opossum). Additionally, there are three small-size species: Tlacuatzin canescens, Marmosa mayensis and Marmosa mexicana [7]. Known as the Mexican opossum mouse, Marmosa mexicana is a small marsupial found along the Gulf of Mexico, from southern Tamaulipas to Yucatan, and on the southern. 

Pacific slope, from Oaxaca to western Panama. It is a solitary, semi-arboreal, and mainly nocturnal mammal [8,9]. In the wild, it feeds mainly on insects and fruit, although its diet may also include small rodents, birds, lizards and eggs. The opossum is prey to different carnivores such as foxes, owls and snakes [8,10]. Its habitat includes tropical deciduous forests, humid tropical forests, cloud forests, and secondary vegetation tracts up to 1,800 m.a.s.l. 

M. mexicana appears in biological inventories of small mammals in different plant formations [11,12,13,14], as a host of certain parasites [15], and in taxonomic reviews [16]. In all of these studies, the species is present in low abundance (n = 2), which can be attributed to the capture techniques employed, mostly consisting of placing Sherman traps placed on the ground [17]. However, aside from ground level activity, this species is known to have semi-arboreal habits, and has been reported to carry out foraging activities on trees 1 to 1.5 m high [18,19]. 

Another determining factor in the capture of individuals of the Marmosa genus, who are known to consume a variety of foods in the wild, is the type of bait used [9]. Among the most popular baits are oats and fruit [20,13], banana [21], papaya, and sweet corn [22]; but meat, ground raisins, and broken corn have also been used [23]. However, studies on the success of different baits and trapping techniques in the capture of M. mexicana individuals remain scarce [11,12]. The aim of this study was to comparatively evaluate bait preferences and trap placement (ground vs. tree) in a peri-urban park in Xalapa, Veracruz.

Study Site

The study was conducted in the Ecological Park El Haya, located in the outskirts of Xalapa, Veracruz, Mexico, at a latitude of 19 ° 31’09’’ N and a longitude of 96 ° 56’33’’ W, and an altitude of 1,418 m.a.s.l. (Figure 1). The park extends over 13 has. The climate in the area is temperate (fm), with an average annual temperature of 18° C and an average annual rainfall of 1,500 mm. The dominant vegetation type is secondary vegetation with coffee plantations, and some typical species of the montane cloud forest. The arboreal stratum includes Eriobotrya japonica, Palicourea padifolia, Coffea arabica, Leucaena sp., and Citrus sp; and in the undergrowth Piper sp. and Odontonema callistachyum are found. Soil types are mostly phaeozem, regosol, andosol and luvisol.

Bait Types

On each sampling day, 100 grams of different bait types were prepared [22,23], namely apple-blueberry (50 g of chopped apple and 50 g of blueberries), papaya-sweet corn (50 g of chopped papaya and 50 g of canned sweet corn kernels), banana-cricket meat (50 g of banana, 40 g of ground meat and 10 g of crushed crickets), oats-peanut butter (50 g of oat flakes and 50 g of peanut butter). Latex gloves were used during preparation to avoid impregnation of foreign odors. 

Capture of Individuals

For the capture of individuals, six linear transects 200 m long were established, along which 40 Sherman® traps of 5.08 x 6.35 x 16.51 cm, and 7.62 x 8.89 x 22.86 cm were placed, separated 5 m from each other. Samplings were carried out between January and July 2017, leaving the traps on site for 18 days each month. The baits described above were randomly distributed among the traps (10 g per trap), so as to minimize the effect of learned behavior [24]. Each type of bait was present in ten traps. Ground traps alternated with tree traps at 4 m above the ground, placed on trees using brackets 20 cm long [25]. Traps were placed at 5:00 pm and checked at 7:00 am the following day. Standard morphometric measurements of the captured individuals were recorded, such as sex and reproductive status; and in order to record recaptures, these details were marked on their right hind limb with an indelible-ink Pelikan Scribtol®. Finally, individuals were released in the capture sites, and their position was geo-referenced. Traps were washed with neutral soap to eliminate odors and facilitate future catches. 

Data Analysis

Capture Success

Success was measured using the following formula: catches/nights-trap*100, which considers the total of catches and the total of sampling efforts (number of nights per trap), multiplied by 100 [26].

Figure 1: Study site in the city of Xalapa, Veracruz, Mexico corresponding to the El Haya ecological par

Effects of bait type and trap placement 

Essential to the evaluation of bait type and trap placement was to determine whether the number of individuals had a normal distribution, so a Kolmogorov & Smirnoff Test was run. If normal distribution was not the case, the data were transformed to ranges [27]. The effect of trap placement and bait type on the number of captured individuals was evaluated through a variance factorial analysis with second level interactions, where the response variable was the number of captured individuals transformed to ranges, and the effect variables the site of placement and the type of bait. To identify differences in the possible interactions between the levels of each factor, a Tukey Post-hoc Test with a confidence level of 95% was conducted, P <0.05 [27].

Capture Success

A total of 26 individuals were captured during 5,040 trap nights, with a capture success of 1.48%, plus 49 recaptures, making for a total of 75 Marmosa mexicana records. 

Effect of Bait Type and Trap Placement

Only the site of trap placement had a significant effect (gl = 1, F = 7.43, P = 0.0067) in the capture success of M. mexicana. Traps placed on trees at 4 m above the ground captured more individuals than those placed on the ground (182.95 ± 10.56 vs. 154.04 ± 10.56). Of the 75 individuals caught, 53 were in tree traps, and 22 on ground traps (Figure 2). 

Figure 2: Effect of trap placement site. Significant differences were found in the number of individuals captured by traps placed at different sites (transformed to ranges). Different letters indicate significant differences p<0.05 (mean ± SE). 

Although this study in a peri-urban park obtained more catches of M. mexicana than previous studies in Mexico had obtained [11,14], capture success was comparatively lower (1.48% with 5,040 trap nights, 75 captures, 49 of them recaptures). For example, Rodríguez-Santiago [28] reported a success rate of 2.61% with 880 trap nights in a remnant of montane cloud forest (MCF) on the Xalapa- 
 

Naolinco road. This suggests that results may be influenced by the characteristics of the study site (i.e. natural habitat vs. peri-urban park). However, more studies are necessary to sustain this hypothesis. The Naolinco site referred to was not fragmented, and had connectivity to other vegetation fragments with varying structural complexity, whereas in the case of our study, the site was a small and fenced vegetation fragment limited by roads, which may have acted as a barrier to the movement of animals [29]. Moreover, the fact that the site is used as a recreational area may also have had an effect on the number of catches.

The existing records of M. mexicana have come mostly from biological inventories, with low capture success. For example, a sampling effort in the Montes Azules Biosphere Reserve in the state of Chiapas with 27,972 trap nights from June 1990 to May 1991, reported a low capture success of 0.046% (equivalent to 13 individuals). Equally low values have been reported in the south of Veracruz [11] and the Sumidero Canyon National Park in Chiapas [30]. In the former case, with 250 trap nights in the months of June, January and March of 1997-1998, only two catches were obtained in disturbed vegetation fragments with agricultural crops, pastures, and urban and industrial waste. In the latter case, the results were obtained through interviews. In all these cases, low catches (< 13 individuals) may be directly related to inadequate trapping methods or baits, or both [31].

In our study, the type of bait had no effect on the capture of M. mexicana, which leads us to conclude that all of the food items described above may prove efficient in future studies. Similar results have been reported in the capture of rodents and marsupials at both the Amazon Reserve in southeastern Peru, and the Rio Doce State Park, in Brazil [32]. In Mexico, opossum mice (Tlacuatzin canescens, Marmosa mayensis and M. mexicana) are omnivores, so they can feed on birds’ eggs as well as on insects, fruit from various types of vegetation, and even the nectar of the cactus Stenocereus queretaroensis [33]. Therefore, it is important to consider different food items in the capture protocols [34].

As for traps, those placed at 4 m above the ground were the most effective in capturing M. mexicana individuals, which suggests a preference for vertical vegetation space. Several studies within the Neotropical realm have reported the placement of aerial traps in their capture protocol, as a way to evaluate the vertical use of vegetation by medium and small mammals [25,35]. Some of these studies focus on the capture of marsupials such as Micoureus paraguayanus and M. demerarae [36]. Our results agree with those reported by Díaz [19] in Iquitos-Nauta Peru, where the use of aerial traps placed on trees at 6 to 19 m above the ground made the capture of marsupials (such as Marmosa regina, Caluromys lanatus and Marmosops noctivagus) more efficient. Similarly, in the state of Colima, Mexico, tree traps successfully captured individuals of Tlacuatzin canescens, a nocturnal and arboreal species that spends 76.6% of the time on trees [33]. Marmosa mexicana is considered a species of semi-arboreal habits that builds nests on tree trunks and avails of abandoned bird nests [2]. However, despite the existing evidence that the placement of traps at different heights in the vertical stratum can favor the capture of different species of this group of mammals [37], tree traps continue to be rarely used in the monitoring of small arboreal mammals, such as rodents and marsupials [25]. 

Many small marsupial species, including M. mexicana, live in secondary vegetation tracts such as urban parks, which ought to be included as habitats worthy of biological research. Any protocol seeking better results in the capture of M. mexicana ought to consider the types of baits used in this study, as well as the placement of traps on trees at 4 m above the ground.

Acknowledgment

We thank the Environment Coordination of the Municipality of Xalapa, Veracruz, for their permission to conduct this research project, and the National Council of Science and Technology (Conacyt) for the scholarship granted to students of members of the National System of Researchers. Finally, we thank all the people who helped during fieldwork.

1. Dickman, R.C. “Marsupial of the World: An Introduction.” Walker’s Marsupials of the World, edited by R.N. Nowak, Johns Hopkins University Press, 2005, pp. 1–68.

2. Ceballos, G., and G. Oliva. Los Mamíferos Silvestres de México. Comisión Nacional para el Conocimiento y Uso de la Biodiversidad, Fondo de Cultura Económica, 2005.

3. IUCN (International Union for Conservation of Nature). “Didelphidae.” The IUCN Red List of Threatened Species, 2015, https://www.iucnredlist.org/.

4. Gardner, A.L. Mammals of South America. Volume 1. Marsupials, Xenarthrans, Shrews and Bats. University of Chicago Press, 2007.

5. Tyndale-Biscoe, H. Life of Marsupials. CSIRO Publishing, 2005.

6. De Almeida, A.J., et al. “Use of Space by Neotropical Marsupial Didelphis albiventris (Didelphimorphia: Didelphidae) in an Urban Forest Fragment.” Revista Brasileira de Zoologia, vol. 25, no. 2, 2008, pp. 214–219.

7. Ramírez-Pulido, J., et al. “List of Recent Land Mammals of Mexico.” Museum of Texas Tech University Special Publications, vol. 63, no. 1, 2014, pp. 1–69.

8. Alonso-Mejía, A., and R.A. Medellín. “Marmosa mexicana.” Mammalian Species, vol. 421, no. 1, 1992, pp. 1–4.

9. Medellín, R.A. “Marmosa mexicana Merriam, 1897.” Mamíferos Silvestres de México, edited by G. Ceballos and G. Oliva, Comisión Nacional para el Conocimiento y Uso de la Biodiversidad, Fondo de Cultura Económica, 2005, pp. 98–99.

10. Medellín, R.A., and M. Equihua. “Mammal Species Richness and Habitat Use in Rainforest and Abandoned Agricultural Fields in Chiapas, Mexico.” Journal of Applied Ecology, vol. 35, no. 1, 1999, pp. 13–23.

11. Sánchez-Hernández, C., et al. “Mamíferos en Cuatro Áreas con Diferente Grado de Alteración en el Sureste de México.” Acta Zoológica Mexicana, vol. 84, no. 1, 2001, pp. 35–48.

12. Pacheco, J., et al. “Diversidad, Historia Natural y Conservación de los Mamíferos de San Bito de Coto Brus, Costa Rica.” Revista de Biología Tropical, vol. 54, no. 1, 2006, pp. 219–240.

13. Ibarra-Portillo, R. “Registro de Marmosa mexicana en Cinquera, Departamento de Cabañas, El Salvador.” Ocelotlán, vol. 6, no. 1, 2009, pp. 3–4.

14. Palacios-Romo, T.M., et al. “Inventario de Mamíferos en Sistemas Cafetaleros de Sombra Asociados a la Cuenca del Río Copalita, Oaxaca, México.” Therya, vol. 3, no. 3, 2012, pp. 303–310.

15. Guzmán-Cornejo, C., et al. “Metazoarios Parásitos de Tlacuatzin canescens y Marmosa mexicana (Mammalia: Didelphimorphia) de México.” Revista Mexicana de Biodiversidad, vol. 83, no. 2, 2012, pp. 557–561.

16. Rossi, R.V., et al. “A Revision of the Didelphid Marsupial Genus Marmosa. Part 1, The Species in Tate’s ‘mexicana’ and ‘mitis’ Sections and Other Closely Related Forms.” Bulletin of the American Museum of Natural History, 2010.

17. Santos-Filho, M., et al. “Edge Effects and Landscape Matrix Use by a Small Mammal Community in Fragments of Semideciduous Submontane Forest in Mato Grosso.” Brazilian Journal of Biology, vol. 68, no. 4, 2008, pp. 703–710.

18. Prevedello, J.A., et al. “Vertical Use of Space by the Marsupial Micoureus paraguayanus (Didelphimorphia, Didelphidae) in the Atlantic Forest of Brazil.” Acta Theriologica, vol. 54, no. 1, 2009, pp. 259–266.

19. Díaz, M.M. “Marsupiales (Didelphimorphia: Didelphidae) de Iquitos y sus Alrededores (Loreto, Perú).” Therya, vol. 5, no. 1, 2014, pp. 111–151.

20. Villa, R. “Mamíferos del Soconusco Chiapas.” Anales del Instituto de Biología Universidad Nacional Autónoma de México, Serie Zoología, vol. 19, no. 1, 1948, pp. 485–528.

21. Kelly, M.K., and T. Caro. “Low Density of Small Mammals at Las Cuevas, Belize.” Mammalian Biology, vol. 68, no. 6, 2003, pp. 372–386.

22. Enders, R. “Banana Stowaways Again.” Science, vol. 71, no. 1, 1930, pp. 438–439.

23. Woodman, N., et al. “Comparison of Traps and Baits for Censusing Small Mammals in Neotropical Lowlands.” Journal of Mammalogy, vol. 77, no. 1, 1996, pp. 274–281.

24. Mares, M., and M. Rosenzweig. “Granivory in North and South American Deserts: Rodents, Birds and Ants.” Ecology, vol. 59, no. 2, 1978, pp. 235–241.

25. Lambert, T.D., et al. “Variation in Small Mammal’s Species Richness by Trap Height and Trap Type in Southeastern Amazonia.” Journal of Mammalogy, vol. 86, no. 5, 2005, pp. 982–990.

26. Balazote-Oliver, A., et al. “Population Dynamics of Dromiciops gliroides (Microbiotheriidae) in an Austral Temperate Forest.” Journal of Mammalogy, vol. 98, no. 4, 2017, pp. 1179–1184.

27. Zar, J.H. Biostatistical Analysis. 4th ed., Prentice-Hall, 1999.

28. Rodríguez-Santiago, N.V. Temporada Reproductiva de Marmosa mexicana (Merriam, 1987) en un Fragmento de Bosque Mesófilo de Montaña en el Municipio de Naolinco, Ver. Undergraduate dissertation, Universidad Veracruzana, 2012.

29. Capitanachi, C., and S. Amante. Las Áreas Verdes Urbanas de Xalapa, Veracruz. Instituto de Ecología, A.C., Universidad Veracruzana, 1995.

30. Arroyo-Chacón, E., et al. “Riqueza, Abundancia y Diversidad de Mamíferos Silvestres entre Hábitats en el Parque Nacional Cañón del Sumidero, Chiapas, México.” Therya, vol. 4, no. 3, 2013, pp. 647–776.

31. Hice, C.L., and P.M. Velazco. “Relative Effectiveness of Several Bait and Trap Types for Assessing Terrestrial Small Mammal Communities in Neotropical Rainforest.” Occasional Papers Museum of Texas Tech University, no. 316, 2013, pp. 1–16.

32. Astúa, D., et al. “Influence of Baits, Trap Type and Position for Small Mammal Capture in a Brazilian Lowland Atlantic Forest.” Boletim do Museu de Biologia Mello Leitão, vol. 19, no. 1, 2006, pp. 31–44.

33. Kennedy, M.L., et al. “Demographic Features, Distribution and Habitat Selection of the Gray Mouse Opossum (Tlacuatzin canescens) in Colima, Mexico.” Acta Theriologica, vol. 58, no. 1, 2013, pp. 285–298.

34. Patric, E.F. “Bait Preference of Small Mammals.” Journal of Mammalogy, vol. 51, no. 1, 1970, pp. 179–182.

35. Grelle, C.E.V. “Forest Structure and Vertical Stratification of Small Mammals in a Secondary Atlantic Forest, Southeastern Brazil.” Studies on Neotropical Fauna and Environment, vol. 38, no. 2, 2003, pp. 81–85.

36. Pires, A.D.S., and F.A.D.S. Fernández. “Use of Space by the Marsupial Micoureus demerarae in Small Atlantic Forest Fragments in South-Eastern Brazil.” Journal of Tropical Ecology, vol. 15, no. 3, 1999, pp. 279–290.

37. Malcom, J.R. “Comparative Abundances of Neotropical Small Mammals by Trap Height.” Journal of Mammalogy, vol. 72, no. 1, 1991, pp. 188–192.

38. Pereyra-Díaz, D., et al. “Influencia de La Niña y El Niño sobre la Canícula en la Región de Xalapa, Veracruz, México.” Ecosistemas y Recursos Agropecuarios, vol. 16, no. 32, 2000, pp. 71–79

39.