Introduction
Bats utilise a wide variety of roost types; the vast majority of species use pre-existing and long-lasting sites, such as caves and hollow logs. A few species use roosts that are actively built by them, such as cavities between roots and foliage, holes in termite mounds and modified leaves called ‘tents’ (Rodríguez-Herrera et al. Reference Rodríguez-Herrera, Medellín and Timm2007). Tents are roosts that protect bats from predators, rain, wind and retain the heat released by bats (Kunz and Lumsden Reference Kunz, Lumsden, Kunz, Lumsden and Fenton2003; Rodríguez-Herrera et al. Reference Rodríguez-Herrera, Víquez-R, Cordero-Schmidt, Sandoval and Rodríguez-Durán2015). Additionally, using these roosts enables bats to ‘camp’ near their feeding sites (Charles-Dominique Reference Charles-Dominique1993; Villalobos-Chaves et al. Reference Villalobos-Chaves, Vargas-Murillo, Rojas, Keeley and Rodríguez-Herrera2016).
Twenty-one species have been recorded in the Neotropics that use this type of roost, some facultatively, for example, Artibeus jamaicensis, which also uses caves and hollow logs, and others that exclusively use tents, such as Ectophylla alba, which mainly uses modified Heliconia leaves. Currently, the species known as neotropical tent-roosting bats are: Artibeus jamaicensis, A. lituratus, Dermanura anderseni, D. cinerea, D. glauca, D. gnoma, D. phaeotis, D. rosenbergi, D. tolteca, D. watsoni, Ectophylla alba, Mesophylla macconnelli, Platyrrhinus incarum, P. recifinus, Uroderma bilobatum, U. convexum, U. magnirostrum, Vampyressa pusilla, V. thyone, Vampyriscus nymphaea and Rhinophylla pumilio (Breviglieri Reference Breviglieri2021; Rodríguez-Herrera et al. Reference Rodríguez-Herrera, Medellín and Timm2007).
Despite its great diversity and size, Brazil has only four publications on tent-roosting bats (see Appel et al. Reference Appel, da Cunha Tavares, de Assis and Bobrowiec2021; Breviglieri Reference Breviglieri2021; Zortéa Reference Zortea1995; Zortéa & Brito et al. Reference Zortéa and De Brito2000), and none of these studies aimed at a systematic, directed search for this type of roost. This work presents the first census of tent-roosting bats conducted in Brazil, specifically in the northern Atlantic Forest. We present new information on bat roosts for Brazil and a new record of a tent-roosting bat.
Methods
The study was conducted at the Murici Ecological Station (MES) (Lat. 9°11’05”–9°16’48” S; Long. 35°45’20”–35°55’12” W) in Alagoas state within the Pernambuco Endemism Center (PEC) (Figure 1). The PEC is an ecoregion of the Atlantic Forest, delimited by the São Francisco River to the south, the Caatinga biome at its western and northern borders, and the Atlantic Ocean along its eastern coast (Feijó et al. Reference Feijó, Beltrão, da Costa-Pinto, da Rocha, de Freitas, Campos, Cordeiro-Estrela, Pereira Filho, França, Alves and Vasconcellos2023). The protected area covers 6116 hectares and has an altitude ranging from 150 to 600 m. It consists of several forest fragments of varying sizes within a submontane dense ombrophilous forest, semideciduous forests and rocky outcrops, mostly located on hilltops within a matrix of sugarcane or pasture (Oliveira et al. Reference Oliveira, Amorim and Lyra-Lemos2020). The climate is type As, according to Köppen’s classification, with a dry season from September to February, followed by a rainy season from March to August. Annual precipitation ranges from 800 to 1800 mm, and temperatures vary between 22 °C and 26 °C (Alvares et al. Reference Alvares, Stape, Sentelhas, Gonçalves and Sparovek2013).
Map of localities with records of bats tent-roosting in Brazil, including the location of Murici Ecological Station (upper right) and occupied tents (stars) found in this study (lower right). References of localities (black points) are: (1) Breviglieri Reference Breviglieri2021. (2) Zortea and Brito Reference Zortéa and De Brito2000; Zortea Reference Zortea1995. (3) Zortea and Brito Reference Zortéa and De Brito2000. (4) Appel et al. Reference Appel, da Cunha Tavares, de Assis and Bobrowiec2021. (5) This study. Detailed information on all localities is found in Supplementary material 1.
We visited the area twice to conduct exhaustive, purposeful surveys for tents, once during the dry season (23–25 October 2024) and again during the rainy season (8–9 April 2025). We sampled the same trails within the protected area through direct observations at each visit, complemented by binoculars and photographs. When a tent was found, we documented its height, distance from the trail, the tent’s status (old or new leaf), tent construction completeness (complete or incomplete), identified the plant species and determined the tent’s architecture based on Rodríguez-Herrera et al. (Reference Rodríguez-Herrera, Medellín and Timm2007).
If bats were found inside tents, we photographed and tried to capture them using a hand net for species identification. Captured bats were collected and deposited in the Zoological Collection of the Federal University of Paraíba. The specimens collected were fixed in 10% formalin, preserved in 70% ethanol, and subsequently, the skulls were extracted. The survey and collection of specimens were allowed by the Brazilian environmental agency (SISBIO licence number 77804-6 MMA, IBAMA and ICMBIO).
Results
In total, we covered an area of 3,340 m2 twice—once during the dry season and once during the rainy season. We found 45 tents during the two surveys. Twenty-one were found in the dry season and 24 in the rainy season, for an average of 3 tents per hectare. Only four tents were occupied by bats, all found in the rainy season. In the dry season, only 2 of 21 tents were occupied (9.5%), but on both occasions, bats flew away, preventing us from identifying the species. During the rainy season, the tent occupancy rate is 16.6% (4 of 24 tents). Only one plant individual had tents in both seasons. A Philodendron ornatum with four tents was observed during the dry season, whereas in the rainy season, the same plant had fewer tents, all of which were old and/or dead. All were unoccupied.
Three occupied tents housed Artibeus obscurus. The three tents were constructed in the hemi-epiphyte Philodendron ornatum (Araceae) (Figure 2a, b). All bats were found roosting alone, and all three modified leaves were umbrella-shaped (Rodriguez et al. Reference Rodríguez-Herrera, Medellín and Timm2007). The plants ranged in height from 4 to 5 m, and the modified leaves measured 70–80 cm in length and 30–40 cm in width. The tents feature semicircular cuts at the base of the first 4–5 veins, causing the leaf base to collapse and form the tent (Figure 2c, d). One male individual was collected (UFPB 12775).
(a) Record of Artibeus obscurus. (b) Tent-roosting in Philodendron ornatum. (c) The modified leaves presented an umbrella architecture with semicircular cuts in the base of the first 4–5 veins. (d) The modified leaves are 70–80 cm long and 30–40 cm wide at 4–5 m height.
One tent harboured two Vampyressa pusilla on the hemi-epiphyte Philodendron fragantissimum (Figure 3a, b). The tent was 3.5 m high, and the single modified leaf was 58.5 cm long and 43.5 cm wide. The tent had an inverted boat architecture, with bite cuts parallel to the central vein on both sides of the central portion of the leaf (Figure 3c, d). One male individual was collected (UFPB 12774).
(a) Record of Vampyressa pusilla. (b) tent-roosting in Philodendron fragantissimum. (c) The modified leaf presented an inverted boat architecture by presenting bite cuts parallel to the central vein and on both sides of it at the central portion of the leaf. (d) the modified leaves are 70–80 cm long and 30–40 cm wide at 3.5 m from the ground.
Discussion
We present the first record of plant species from the Araceae family used as tents in this region, as well as the first record of umbrella architecture in Brazil. We also present the first records of tent-roosts for the northern Atlantic Forest in Brazil. Few bat tent surveys have been conducted worldwide. Our results show that the tent density in Murici is three tents per hectare, a value similar to other studies in the Tirimbina Biological Reserve located in the Caribbean rainforest of Costa Rica (4 tents per hectare) (Villalobos-Chaves et al. Reference Villalobos-Chaves, Vargas-Murillo, Rojas, Keeley and Rodríguez-Herrera2016) and those in Petén, Guatemala (3.5 tents per hectare) (Arévalo et al. Reference Arévalo, Trujillo, Sánchez-Calderón, Medellín and Rodríguez-Herrera2025).
With the record of Artibeus obscurus, the total number of neotropical tent-roosting bats increases to 23 (Rodríguez-Herrera et al. Reference Rodríguez-Herrera, Medellín and Timm2007). Moreover, A. obscurus is the third-largest species to use leaf tents as roosts (Forearm: 55.4–65.0 mm), behind A. lituratus (FA: 65–78 mm) and A. jamaicensis (52.0–67.4 mm). Like A. jamaicensis and A. lituratus, A. obscurus does not exclusively roost in tents; cavities in standing logs, rock crevices, caves, human structures and foliage are also known roosts for these species (Haynes and Lee Reference Haynes and Lee2004). Despite its widespread distribution in Colombia, Guyana, Ecuador, Peru, Venezuela, Bolivia and Brazil (Haynes and Lee Reference Haynes and Lee2004), we believe that it has not been found in tents due to a lack of tent searches, as there is generally very little research on tents in South America. However, if bats of this species have been found or collected in tents, that information did not make it into a publication, since details of fieldwork are rarely reported, including ecological details such as the presence of bats in tents.
In fact, these are the first records of tents in the northern Atlantic Forest of Brazil. Although we have no direct observation, the size of the cuts on the leaf suggests that A. obscurus was the species that modified the leaf into a tent. Moreover, in all three of the tents that we found A. obscurus, all presented the same biting pattern (umbrella). This is the first record of the umbrella architecture in a tent for Brazil, as previous records included only boat, boat/apical, bifid and conical architectures (Supplementary Material).
Vampyressa thyone and V. pusilla are known to roost in tents (Rodríguez-Herrera et al. Reference Rodríguez-Herrera, Medellín and Timm2007). For V. pusilla, reports on tents have included leaves of Heliconia (Heliconiaceae), Philodendron (Araceae), Simira (Rubiaceae) and possibly Piper (Piperaceae) plants with boat and boat/apical architecture but solely reported in Espírito Santo, Brazil, within the Atlantic Forest (Zórtea and de Brito Reference Zortéa and De Brito2000). Vampyressa pusilla is distributed in the Atlantic Forest of Brazil, Paraguay and Argentina (Lewis and Wilson Reference Lewis and Wilson1987). The limited information available on the roosting ecology of this species suggests that it is an obligatory tent species, like the phylogenetically close genera Ectophylla and Mesophylla (Tavares et al. Reference Tavares, Gardner, Mcdonough, Maldonado, Gutiérrez, Velazco and Garbino2022).
We found very few occupied tents in both seasons (6 of 45), although we encountered unoccupied tents with leaves in good, healthy condition. The dynamics of tent occupation remain poorly understood, as bats use tents not only as day roosts. Bats can build tents for various purposes, which can vary in space and time, and unoccupied healthy tents can serve as alternative roosts after escaping predators (Brooke Reference Brooke1990; Timm and Clauson Reference Timm and Clauson1990), to have temporal close access to seasonal food (Charles-Dominique Reference Charles-Dominique1993), and as exclusive night roosts for feeding (Melo et al. Reference Melo, Rodríguez-Herrera, Chazdon, Medellín and Ceballos2009; Rodríguez-Herrera et al. Reference Rodríguez-Herrera, Medellín and Timm2007). However, the potential for human error during detection must be considered, as locating tent roosts is difficult amid a ‘sea of leaves’ in rainforests, even with surveyors’ extensive experience.
The Atlantic Forest has historically experienced significant habitat loss due to deforestation and fragmentation, which are more pronounced in the PEC than in other parts of the biome (Filho et al. Reference Filho, França, Alves and Vasconcellos2023). Thus, the MES is a critical sanctuary for tent-roosting bats and other wildlife in northeastern Brazil, particularly within the PEC, as the biome faces escalating threats. More active searches and censuses are needed in the PEC and throughout the Atlantic Forest to identify which plants and architectures are used by these bats, as well as their role in plant dispersal by collecting seeds under the tent-roost. This data will provide a better understanding of the ecology of tent-roosting bats in the highly threatened Atlantic Forest.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0266467426100546.
Acknowledgments
The National Geographic Society for providing financial support for these expeditions as part of the Meridian project NGS-102401R-23: ’Exploring bat architecture: Their homes, neighbourhoods, and environmental services they provide us’, awarded to B. Rodriguez Herrera. We thank environmental analyst and head of the conservation unit, Marco Antônio Freitas, for the entry authorisation and logistical support, and to the environmental brigade member Cícero Junior for his assistance and field guidance.
Funding statement
This work was supported by the project NGS-102401R-23: ‘Exploring bat architecture: Their homes, neighborhoods, and environmental services they provide us’ granted to BRH.
Competing interests
None.
Ethical statement
Bat capture and handling protocols followed the guidelines of the American Society of Mammalogists for the use of wild mammals in research (Sikes 2016).
