Introduction

Due to its great ecological plasticity, the genus Cebus (capuchin monkeys, macacos-prego, and caiararas) is one of the most widely distributed geographically among neotropical primates (Silva Júnior 2001; Fragaszy et al. 2004). Groves (2001) divides Cebus into eight species (C. capucinus, C. kaapori, C. albifrons, C. olivaceus, C. apella, C. nigritus, C. xanthosternos, C. libidinosus), whereas Silva Júnior (2001) considers 11 species for the genus (the aforementioned eight, plus C. robustus, C. cay, C. macrocephalus). Species of the genus are found in virtually all forested habitats of South and Central America, including humid areas, forests on the Atlantic Coast, dry areas, llanos in Venezuela and the Brazilian Caatinga (Silva Júnior 2001; Madden et al. 2007). Cebus libidinosus occurs in most of northeastern Brazil (Rylands et al. 2000; Silva Júnior 2001), in areas of the Caatinga and in Cerrado biomes (Silva Júnior 2001).

The recent rediscovery of C. flavius in areas of coastal Atlantic Forest in northeastern Brazil (Oliveira and Langguth 2006) suggests that the diversity of this genus, as well as the distribution limits and possible hybridization areas of its species, are not sufficiently known, particularly for the new and rediscovered species (Silva Júnior 2001; Ferreira et al. 2007; Fialho et al. 2007). C. flavius has a yellowish fur compared to C libidinosus (in fact, local people call C. flavius the blond monkey, distinguishing them from the red monkey—C. libidinosus). C. flavius also have a more homogenous coloration throughout the body while C. libidinosus has darker forearms [see Fig. 2 and Silva (in prep) for a thoroughly description of differences between these two species]. Earlier surveys have confirmed the suggestion made by Oliveira and Langguth (2006) that the distribution of C. flavius is restricted to the coastal Atlantic Forest, from south Rio Grande do Norte (RN) to the left margin of the São Francisco River in Alagoas state, with the western limit coinciding with the limits of the Atlantic Forest biome (Ferreira et al. 2007). These surveys recorded small populations of C. flavius inhabiting isolated patches of Atlantic Forest in northeastern Brazil (Ferreira et al. 2007), suggesting that the species is at risk of extinction. The IUCNs recent evaluation (November 2007) confirmed the critically endangered status of the species (unpublished data).

Rio Grande do Norte state, in the northeast of Brazil, exhibits a diversity of environments such as Atlantic Forest and Caatinga biomes, and transition zones. Currently, Caatinga ecosystems occupy 93.6% of RN territory, while only 6.4% are characterized as Atlantic Forest formations (Brasil 2005). Official and scientific reviews of mammal presence in the RN Caatinga are scarce, and nonexistent for Cebus. Reis et al. (2006) report the occurrence of only two mammals (Tolypeutes metacus and Dasyprocta prymnolopha) in RN, while the official list of the Brazilian Environment Ministry (Brasil/MMA 2003) contains five mammals (Alouatta belzebul, Dasyproctus novemcintus, Euprhactus sexcinctus, Tolipeutes tricinctus in coastal areas, and Cabassous sp. in the Caatinga). However, Porpino and Silva (submitted) indicate the presence of 23 non-flying mammals in the RN Caatinga, most of them listed by the IUCN (2007) and the Brazilian Environment Ministry (Brasil/MMA 2003) (see Table 3) as non-threatened.

According to Kinzey (1997), the primate order in RN is represented by three species and three families: Callithrix jacchus (Callithrichidae), Alouatta belzebul (Atelidae), and Cebus libidinosus (Cebidae) (although a question mark indicates the absence of an official record of Cebus in RN). In this paper, we present the first official record of two Cebus species in RN, C. libidinosus and C. flavius, and refine the distribution limits for both species.

Methods

Between January 2006 and October 2007, 12 expeditions (of 5 days each) to different regions of the RN Caatinga were conducted to locate remnant Cebus groups. Regions were chosen based on habitat conservation status indicated on satellite images. We interviewed local residents (small farmers and hunters) using a semi-structured questionnaire, taking names, time of residence in that area, economic activity, and other socio-economic data (see Table 1). During the interviews, we showed them two plates with pictures Cebus (libidinosus, flavius), Alouatta, and Callithrix species so they could indicate which of them had been observed in the surroundings, their areas of occurrence, how long these animals had been in this area, and some other information about the habits of animals (e.g., crop raiding, sleeping areas). As a negative control, plates also included pictures of primates that have never occurred in the area (e.g., C. xanthosternos, Brachyteles arachnoides).

Table 1 Summary of interviews
Full size table

When the presence of Cebus was indicated in the interviews we would proceed to conduct active surveys in the areas to observe Cebus or remains of their presence (i.e., tool use sites). On these surveys, we followed hunters’ trails for 3 days, but the size of the area sampled varied according to weather and terrain conditions and whether or not we found a positive indicator of Cebus presence. Based on previous records of tool use by Cebus species in Caatinga environments (Fragaszy et al. 2004; Moura and Lee 2004; Visalberghi et al. 2007), we recorded this evidence to infer the presence of capuchins. We considered a tool use site one or some small stones with marks of use (i.e., hammers) on top of larger stones (i.e., anvils) with cracked seeds scattered around. We did not consider sites without broken seeds to be tool use sites, even if there were hammers and anvils with wear-marks This hammer-anvil-cracked seed set is clearly different from the random, scattered distribution of stones on the environment. We considered monkeys to have used these sites because (1) according to the interviews the local inhabitants do not crack nuts in this way, but do report monkeys cracking nuts this way, and (2) the cracked nuts we found were of the same species (Atallea and Siagrus) cracked by C. libidinosus at other locations (Fragaszy et al. 2004; Visalberghi et al. 2007). We also recorded direct observations of other mammals, and proof of their presence, such as feces, tracks, and carcasses.

Results

Local inhabitants indicated the presence of capuchins in ten out of 17 areas visited (Table 2). In four of these we recorded evidence of their presence (tool use sites), and in two we were able to observe and count the number of animals in the group (see Fig. 1).

Table 2 Surveyed localities and type of evidence obtained for the occurrence of Cebus species
Full size table
Fig. 1
figure 1

Occurrence sites of Cebus in the RN Caatinga. Sites are identified as in Table 2

Full size image

In Martins (Fig. 1, Table 2), we counted a group of 53 Cebus libidinous on October 15, 2006, including adult males and females and several immature individuals. On December 3, 2006, the group was resighted and a juvenile was followed for 10 min. C. jacchus is a sympatric primate species in this area.

The animals were observed ranging on a privately owned hill at approximately 750 m a.s.l. At this altitude, temperature varies between 16 and 25°C, with a mean yearly rainfall of 1,100 mm. Local vegetation is classified as arboreal Caatinga (i.e., trees between 3 and 6 m in height), but large fruit trees are rare, except for two introduced mango trees. The lower slope of the hill is surrounded by maize fields which, according to interviews with the local people, are regularly invaded by capuchins during harvest time. There is an abundance of stones and pebbles of all sizes and weights due to the granitic nature of the region, which hinders access to the area where the monkeys were observed.

In Jucurutu (Fig. 1, Table 2), we observed a group of at least 45 Cebus flavius on August 28, 2006, and September 7, and 16, 2007. The group was composed of adult males, adult females, and several immatures (Fig. 2). Stone tool sites were also recorded (Ferreira et al. in prep.). Interestingly, C. jacchus was not observed at this site in any of the visits or reported in the interviews.

Fig. 2
figure 2

Group of C. flavius at arbustive Caatinga (Jucurutu, RN, Brazil) (photo by A. Roque)

Full size image

These Cebus flavius were observed on a private property of 750 ha (maximum height of 680 m), which has been protected by the owners for at least 30 years, and which was officially registered as a Private Ecological Sanctuary (Reserva Particular do Patrimônio Natural “Stoessel de Brito”) in 1986. The area is located in a difficult-to-access range of hills (Serra do Estreito) and vegetation is classified as arbustive (bush) Caatinga (i.e., trees 1–3 m height and deciduous leaves).

Evidence of the presence of seven additional non-flying mammal species was found in the visited areas. Furthermore, the occurrence of seven other mammals was reported by local people, including four previously unrecorded in the RN Caatinga (Monodelphis domestica, Gracilinanus agilis, Leopardus pardalis, Leopardus wiedii) and the vulnerable (Brasil/MMA 2003) Puma concolor (Table 3).

Table 3 List of non-primate mammals recorded in the RN Caatinga during the present study and type of evidence obtained for their occurrence
Full size table

Discussion

Up to the present time, Cebus libidinosus has been considered to be the only Cebus species occurring in Rio Grande do Norte, given its presence in the Caatingas of the neighboring states of Paraíba and Ceará (Silva Júnior 2001). However, museums contain no Cebus specimens collected in RN, nor are there up-to-date reports of capuchins in this state. Therefore, the present record of C. flavius not only broadens the diversity of known primates in RN but also records the first primate species at critical risk of extinction in the state.

The Stoessel de Brito ecological sanctuary is now the northernmost and westernmost area of occurrence of Cebus flavius. The previous northern limit of C. flavius was a forest fragment in the Associação dos Plantadores de Cana-de-Açúcar—Asplan (6°31′12.7″; 35°8′29.32″W), municipality of Camaratuba, in the coastal forest of Paraíba, and the former westernmost limit was Passo de Camaragibe (9°14′S; 35°30′W), in the coastal forest of Alagoas (Oliveira and Langguth 2006), areas far from Caatinga formations.

All the regions where Cebus (or evidence of their existence) were sought or were indicated to occur belong to a series of mountains which, before human encroachment, were covered with arboreal caatinga and Atlantic Forest remnants in interior areas after the Pleistocene retraction of the Amazonian forest. More recently, vegetation composition has been changing to more xerophytic arbustive Caatinga formation and some near-desert areas after 200 years of human influence (Oliveira et al. 2005).

Currently, these areas have similar characteristics: they are rocky vertical hills difficult to access by both humans and other large mammals (i.e., cattle and sheep), but near maize fields, from which capuchins (Cebus libidinosus but not C. flavius) are said to steal food frequently. This indicates capuchins may have been forced into these regions due to increased human occupation of nearby areas and therefore may be adopting alternative habits to ensure survival, including tool use behavior. The area where C. libidinosus was found still contains patches of arboreal caatinga while that occupied by C. flavius is exclusively arbustive caatinga. The size of the observed groups is also larger than that described for other areas occupied by Cebus libidinosus and C. flavius and larger than that of subgenus Sapajus groups in general (Ferreira 2003 for review of size and social dynamics of Cebus).

Considering this is the first record of C. flavius in the Caatinga, one hypothesis is that the presence of the monkeys in the area could be the result of the release of exogenous animals in this area. However, interviews with local people and with environmental agencies indicate there were no official releases in these areas. Additionally, elderly local inhabitants reported that these capuchins have been present for at least 40 years.

Hitherto, it was thought that Cebus flavius followed a parapatry pattern in relation to Cebus libidinosus similar to that observed in other northeastern Brazilian primates. Titi monkeys (Callicebus coimbrai and Callicebus melanochir) have distributions restricted to the Atlantic Forest, whereas Callicebus barbarabrownae is typical of the Caatinga, distribution being parapatric in both of these areas (Jerusalinsky et al. 2006). The present record of C. flavius refutes what was proposed by Oliveira and Langguth (2006) that this species was restricted to the Atlantic Forest and indicates that this species can also occupy environments in the Caatinga biome.

There are a number of populations of another critically endangered capuchin, Cebus xanthosternos, in the Atlantic Forest of eastern Bahia and Sergipe states, and several populations are also known in Caatinga-like environments in western Bahia (Kierulff et al. 2005). However, contact with other Cebus species is limited by rivers, unlike what is seen for C. flavius and C. libidinosus. This highlights the importance of investigating the factors that are limiting the distribution of Cebus flavius and C. libidinosus, since they are neither restricted to a particular biome nor limited by any apparent physical barriers.

This C. flavius population may be a rare example of a process of geographic distribution retraction, similar to that proposed for Alouatta belzebul. The restricted distributions of C. flavius and A. belzebul are associated to an overall retraction of the Atlantic Forest in northeastern Brazil, due to natural and anthropogenic events (Oliveira et al. 2005). The present records demonstrate that C. flavius distribution is less restricted than previously thought, and broadens our knowledge of the diversity of habitats that the species is able to occupy. Remarkably, these data reveal the co-relation between Caatinga (savannah like) environments and tool use behavior by another Cebus species. This finding indicates the importance of further studies of the species in the Caatinga, and suggests that this biome be included in management scenarios for species conservation as well as for more detailed studies regarding tool use capabilities in Cebus.