Introduction
Animal foraging represents a key ecosystem function for plant ecology and evolution, influencing not only seed dispersal but also herbivory and other processes that shape vegetation regeneration. Despite the advances in the understanding of plant–animal interactions in the Neotropics (Galetti et al. Reference Galetti, Donatti, Pires, Guimarães and Jordano2006), for several plant species dispersal agents are still inferred based on fruit or seed morphology (Fleming et al. Reference Fleming, Breitwisch and Whitesides1987; Howe Reference Howe2016; Lososová et al. Reference Lososová, Axmanová, Chytrý, Midolo, Abdulhak, Karger, Renaud, Van Es, Vittoz and Thuiller2023), which rely on the match between functional traits of fruits and seeds, and the dispersal agents (Rojas et al. Reference Rojas, Zampini, Isla and Blendinger2022). To date, only a few studies have reported direct observation of plant–animal interactions in the Cerrado ecoregion (Bizerril et al. Reference Bizerril, Rodrigues and Hass2005; Reis et al. Reference Reis, Dias, Sousa, Chiarello, Sá and Ramos2023), particularly interactions between plants and small rodents, whose foraging activities may encompass both seed dispersal and herbivory, thereby contributing to local species coexistence and flora diversity. Despite their potential ecological importance, small rodents, especially echimyids, remain poorly documented as seed interactors in the Cerrado.
Rodentia is the most diverse terrestrial mammal group in the Neotropics (Mendonça et al. Reference Mendonça, Percequillo, Camargo, Ribeiro, Palma, Oliveira, Câmara and Vieira2018), playing important roles in ecosystem function and plant communities dynamics due to seed predation, consumption, and potential dispersal processes. Echimyidae comprises 15 genera and nearly 70 species (Wilson and Reeder Reference Wilson and Reeder2005) of spiny rats native to the Neotropics, known for their distinctive spiny or bristly fur. These rodents are primarily herbivorous, feeding on fruits, seeds, leaves, and plant bark (Reis et al. Reference Reis, Peracchi, Pedro and Lima2011), with some species also consuming roots and tubers. Thrichomys pachyurus is a relatively large-bodied Neotropical small rodent standards, with body mass of 300–350 g and a head length of 200–250 mm (Reis et al. Reference Reis, Peracchi, Pedro and Lima2011). Species of Thrichomys mainly occur in Cerrado, Caatinga, and Pantanal ecoregions in Brazil (Bonvicino and Oliveira Reference Bonvicino, Oliveira and D’Andrea2008; Reis et al. Reference Reis, Peracchi, Pedro and Lima2011). Among this gender, only one species is recorded in the state of Goiás in Central Brazil: Thrichomys pachyurus Wagner, 1845 (Hannibal et al. Reference Hannibal, Zortéa, Calaça, Carmignotto, Bezerra, Carvalho and Moreira2021).
Even in the Atlantic Forest, the most well-studied Neotropical biome, the role of spiny rats (Echimyidae) in seed dispersal remains poorly understood. While research in other Neotropical regions has identified spiny rats as both seed predators and potential dispersers (Brewer Reference Brewer2001; Forget Reference Forget1991), their ecological role in the Cerrado remains underrepresented. This knowledge gap is primarily due to their nocturnal and elusive behaviour, which limits detectability and poses significant methodological constraints for direct observation, as well as a historical research focus on plant–animal interactions involving larger, diurnal rodents such as agoutis. Consequently, Echimyidae are often omitted from major frugivory datasets (Bello et al. Reference Bello, Galetti, Montan, Pizo, Mariguela, Culot, Bufalo, Labecca, Pedrosa, Constantini, Emer, Silva, da Silva, Ovaskainen and Jordano2017; Fuzessy and Pizo Reference Fuzessy and Pizo2025), reflecting sampling bias rather than a true absence of interactions in nature.
The lack of data constrains the understanding of ecological interaction networks in Cerrado and the impacts of defaunation in plant community dynamics (Dirzo et al. Reference Dirzo, Young, Galetti, Ceballos, Isaac and Collen2014). This understanding is particularly relevant in the Cerrado ecoregion, where more than 50% of the original area is already modified, mainly by intensive agricultural expansion (Alencar et al. Reference Alencar, Shimbo, Lenti, Balzani Marques, Zimbres, Rosa, Arruda, Castro, Ribeiro, Varela, Alencar, Piontekowski, Ribeiro, Bustamante, Sano and Barroso2020), resulting in elevated extinction risk of several species (Colli et al. Reference Colli, Vieira and Dianese2020). Habitat loss leads to decline of large-bodied vertebrates and disrupts their key role in seed dispersal, particularly of large-seed plants (Dirzo et al. Reference Dirzo, Young, Galetti, Ceballos, Isaac and Collen2014). As these large herbivores decline, smaller animals such as rodents, often more tolerant to habitat modifications, contribute to compensatory ecological functions, including partial replacement of seed-related processes (Diaz and Ernest Reference Diaz and Ernest2022).
Rodents are commonly recognized as seed predators; however, they can potentially disperse seeds because they can gnaw fruits and seeds without damaging them, and they can also carry fruits and seeds to burrows or caves which can never be recovered and can thus germinate (Gomez et al. Reference Gómez, Schupp and Jordano2019). Although small rodents may disperse seeds over relatively short distances, they often deposit them in sites that are highly favourable to germination, thus enhancing the quality of seed dispersal. Nevertheless, their role in seed dispersal is still poorly acknowledged, particularly in tropical ecosystems where direct observation remains a challenge (Brewer Reference Brewer2001; Forget Reference Forget1991; Feer and Forget Reference Feer and P‐M2002). Although some studies in the Atlantic Forest have documented the importance of spiny rats in seed fate (Adler and Kestell Reference Adler and Kestell1998; Lessa and Costa Reference Lessa and Costa2009), these contributions are frequently omitted from general reviews of Neotropical frugivory and seed dispersal. This bias is reinforced by methodological constraints, as alternative approaches such as stomach content and faecal analyses are time-consuming and often limited in their ability to accurately identify plant taxa. In this context, field-based observations can provide valuable insights into plant–animal interactions and the behaviour of frugivorous species, particularly where direct observation is difficult.
Pouteria ramiflora (Mart.) Radlk. is a small zoochore tree widely distributed in the Cerrado, Amazon, Caatinga, and Atlantic Forest ecoregions, in both savannas and forests (Alves-Araujo et al. Reference Alves-Araújo, Swenson and Alves2014). Its fruits are green drupes with 60 mm maximum length, foraged by medium to large-bodied mammals such as monkeys (Sapajus cay, capuchin monkey), carnivores (Lycalopex vetulus, hoary fox), ungulates (Dicotyles tajacu, collared peccary, and Tayassu pecari, white-lipped peccary), Mazama sp. (deer), Tapirus terrestris (lowland-tapir), and rodents (Cuniculus paca, paca, and Dasyprocta sp., agouti) (Alves-Araujo et al. Reference Alves-Araújo, Swenson and Alves2014; Dalponte and Lima Reference Dalponte and Lima1999; Goebel et al. Reference Goebel, Vitorino, Frota and Santos‐Filho2023). Because P. ramiflora produces relatively large fruits (20–60 mm; Silva et al. Reference Silva, Silva, Junqueira and Andrade2001), it is commonly assumed that its dispersal is restricted to medium- and large-bodied vertebrates. However, this assumption may overlook the potential role of small mammals, which can contribute to fruit handling, hoarding, or stomatochory (McConkey et al. Reference McConkey, Sushma and Sengupta2024).
In the NeoFrugivory database (Fuzessy and Pizo Reference Fuzessy and Pizo2025), a comprehensive compilation of 10,175 frugivory interactions across the Neotropics based on 419 studies spanning 1967–2023, the rodent species reported interacting with seeds of P. ramiflora include C. paca, Dasyprocta ssp., Myoprocta spp., and the small rodents Proechimys oconnelli and Proechimys semispinosus. Specifically, in Cerrado, only Proechimys longicaudatus has been recorded consuming fruits of P. ramiflora (Goebel et al. Reference Goebel, Vitorino, Frota and Santos‐Filho2023).
Herein, we contribute to fill the knowledge gap about seed foraging in Cerrado ecoregion, reporting the interaction between the small rodent T. pachyurus and the tree P. ramiflora in a savanna remnant in the Brazilian Cerrado ecoregion for the first time. We observed frugivory, fruit handling and moving suggesting a potential role of this small rodent in seed dispersal processes.
Methods
We set 92 camera traps (model HC900a) in 23 landscapes of the LAND LTER (Figure 1A) project (Agricultural Landscape Dynamics and Impacts on Biodiversity Long-Term Ecological Project). In each sampling site, four cameras were installed at 30 cm above the ground, and all tree species close to the camera were identified, in a 10 m radius buffer, as part of our project protocol. The survey was conducted during the rainy season, between January and February 2023, coinciding with the known fruiting peak of several Cerrado woody plant species. The camera traps remained in the field for 45 days, operating 24 hours/day in each landscape. We programmed the equipment to take three pictures followed by recording a video for 10 s, setting the sensor sensitivity to a low level to reduce false takes. This study was conducted within the broader framework of the LAND LTER monitoring programme; however, the specific observation reported here represents an opportunistic record obtained at a single camera trap station and not observed in other areas of the project.
The LAND LTER project (Agricultural Landscape Dynamics and Impacts on Biodiversity) locality in Brazilian Cerrado (A), and the sampling site in a savanna remnant where the interaction between Thrichomys pachyurus and Pouteria ramiflora was recorded (B).
Results
The interaction between T. pachyurus and P. ramiflora was recorded in one of the landscapes (Figure 1B) in Silvânia district, State of Goiás, Brazil (−16,453231; −48,343031, DATUM WGS84). We monitored an area around a single adult P. ramiflora tree (Figure 2A) and recorded fruit removal and transport by T. pachyurus. The tree had unripe (Figure 2B) and later ripe fruits (Figure 2C) during this period. Thrichomys pachyurus individuals were recorded on 10 different days between January and February, at night or twilight, totalling 19 records (Supplementary Data).
The tree and the small rodentia species interactions. (A) The adult tree of Pouteria ramiflora; (B) unripen fruits; (C) ripen fruits showing the edible mesocarp, the unique seed, and the green fruit colour; (D) an individual of T. pachyurus. foraging on a P. ramiflora fruit; (E) a drawing of the interaction recorded, made by Márcio J. Pereira.
Nine foraging or searching events were observed beneath the tree canopy, each lasting between one and three minutes, during which individuals moved and sniffed the ground (Supplementary Data). On three occasions, fruit transport events were recorded, lasting two to four minutes each, in which rodents searched for, collected, and carried fruits away from the site (Supplementary Material videos: DSCF0076, DSCF0224, DSCF0088, and DSCF0536). In addition, two direct fruit consumption events were documented, lasting two to three minutes each, where individuals were observed feeding directly on fruits (DSCF0220 and DSCF0592). The observed records totalled approximately 20 minutes of direct interaction of Thrichomys pachyurus with the fruits of Pouteria ramiflora. However, it is likely that the actual interaction period was longer, since the recorded duration was limited by the configuration of the camera-trapping equipment, which restricts continuous capture time per event. Since only one tree was observed and seed fate was not tracked, additional studies are required to confirm effective dispersal.
Discussion
An individual P. ramiflora tree can set 10 to 400 fruits in a single fruiting season (Silva et al. Reference Silva, Silva, Junqueira and Andrade2001). A large number of fruits ripening during the rainy season, from December until February, may be an important factor to attract small rodents to forage its fruits, a hypothesis we still need to test in the near future. Due to its relatively large size, P. ramiflora’s seeds are not expected to be swallowed by small rodents such as T. pachyurus. However, small rodents can gnaw the fruits, but it can also handle and move away from the plant, potentially dropping it on burrows or caves leading to a potential high-quality dispersal event (Silva et al. Reference Silva, da Rosa Silveira, Aumond, da Silveira and Cademartori2017). This behaviour may also be interpreted as stomatochory, that is, the removal of fruits by small rodents. Such interactions have been described as a strategy of seed dispersal (Silva et al. Reference Silva, da Rosa Silveira, Aumond, da Silveira and Cademartori2017; Vander-Wall Reference Vander-Wall2001), although our data do not allow us to confirm the final fate of the seeds. Therefore, we consider that the observed event indicates only the potential for seed dispersal, which should be further investigated in future.
The contribution of small rodents in seed dispersal is still poorly understood in the Neotropics. For instance, traces of fruit mesocarp and endocarp of Cabralea canjerana (Meliaceae) were found in faeces of Trichomys aperoides in Atlantic Forest, southeast Brazil, but seeds were not found, raising the question whether the rodent is only a seed predator or may also disperse the seeds (Lessa and Costa Reference Lessa and Costa2009).
Because the seeds of P. ramiflora are large, measuring between 20 and 30 mm in length (Silva et al. Reference Silva, Silva, Junqueira and Andrade2001), and we observed the rodent handling and transporting the fruits, it is plausible that the species also contributes to seed dispersal. Echimyidae species such as Proechmys semispinosus may partially assume the ecological role of seed dispersers in Neotropical forests, particularly in environments where larger frugivores are absent. They have been reported dispersing species such as Scheelea sinensis (Arecaceae) and Calophyllum longifolium (Calophyllaceae) in the Atlantic Forest (Adler and Kestell Reference Adler and Kestell1998) as well as Clidemia urceolata and Miconia holosericea (Melastomataceae) in Cerrado (Lessa and Costa Reference Lessa and Costa2009). In addition, Myrtaceae species in the Atlantic Forest have larger seeds (>3 cm) such as Eugenia cambucarana, E. multicostata, and E. neoverrucosa and are mainly dispersed by rodents such as Dasyprocta sp. (Dasyproctidae) and Proechimys spp. (Echimyidae) (Gressler et al. Reference Gressler, Pizo and Morellato2006).
Additionally, T. pachyurus may also exploit another potential food resource, such as larvae found within P. ramiflora fruits. We noted fly larvae infesting P. ramiflora’ fruits, which can also be a resource for the small rodent, a subject to be investigated further. In the LAND LTER landscapes, Anastrepha spp. (Diptera: Tephritidae) uses fruit from Pouteria gardneriana and P. ramiflora as oviposition sites (Veloso et al. Reference Veloso, Ferreira, Fernandes, Daza and Zucchi1996). This highlights an important aspect of the rodent’s diet that needs further investigation, especially considering the role of these larvae in the rodent’s foraging behaviour.
Our foraging report highlights the importance of small rodents as potential seed dispersers in Neotropical ecoregions and for P. ramiflora and brings new perspectives to future studies regarding the roles of small rodents in seed dispersal. We can point out some questions to be addressed in the near future, such as (i) the effect of fruit yield on frugivore attractiveness; (ii) the effect of T. pachyurus foraging behaviour on seed germination; (iii) seed fate after handling and moving behaviour; (iv) the role of flies in seed predation and mortality; (v) and the role of flies in rodent diet. In conclusion, our study highlights the interaction between the rodent T. pachyurus and the plant P. ramiflora as an example of the potential role of small rodents in seed dispersal in the Cerrado, providing new perspectives for future investigations into these ecological processes in Neotropical ecosystems. Future research on small rodents and their interactions with plant communities is essential for fully comprehending the vegetation dynamics and functioning of Neotropical ecosystems. More broadly, natural history observations and detailed field studies remain indispensable for revealing overlooked interactions and generating the empirical foundation upon which ecological theories and models of community dynamics are built.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S026646742610073X.
Data availability statement
The data that support the results of this study are available as supplementary material.
Acknowledgements
This work was supported by grants to the research network LAND PELD supported by FAPEG (project no. 202010267000404, 202510267001635) and CNPq (project no. 445419/2024-5). ALR was funded by FAPEG (project no. 202110267000877), CNPq fellowship program (project no. 317734/2021-0), and FAPESP (project no. 2024/20380-7). RGC has continuously received productive grants from CNPq, which we gratefully acknowledge.
Author contributions
RGC and ALR conceived the study and developed the sampling design. RGC and ALR funded the work. ALR and MJP performed the fieldwork. MJP and RGC wrote the manuscript with inputs from ALR. All authors approved the final version of the manuscript.
Competing interests
The authors declare that there have been no competing interest or involvements that might raise the question of bias in the work reported or in the conclusions, implications, or opinions stated.
Animal welfare statement
All procedures performed with wild animals in this study followed guidelines of the ethics of animal care and welfare, namely the Guidelines of the American Society of Mammalogists for the use of wild mammals in research.
