Primates are distinguished from other groups of animals by having large brains, enhanced manipulative abilities, and more complex cognitive skills (Garber, 2004). These traits allow nonhuman primates to perform complex behaviors such as tool use, a behavior previously considered to be limited to humans (Panger, 1998). Most instances of manipulating objects as tools have been recorded in apes (Beck, 1975; Goodall, 1964). There are some records for Old World monkeys, including baboons (Papio) and macaques (Macaca) (Van Lawick-Goodall et al., 1973; Tomasello and Call, 1997; Westergaard, 1992); in New World monkeys, tool use has been observed in capuchin monkeys (Cebus) (Beck, 1972, 1975; Chevalier-Skolnikoff, 1989; Fragaszy et al., 2004; Ottoni and Mannu, 2001; Phillips, 1998; Struhsaker, 1977; Vauclair and Anderson, 1994; Visalberghi, 1990; Westergaard, 1988).

Beck (1975) defined tool use as “the manipulation of an unattached environmental object, the tool (not part of the user's body), to alter more efficiently the form or position of a separate object, when the user holds or carries the tool in toto during or just prior to use and is responsible for the critical connection between tool and incentive” (p.414). Urbani and Garber (2002), however, warned that several reports of tool use cited in the scientific literature are better classified as “proto tool-use or object manipulation.” True tool use involves the detachment and manipulation of both the object of change and the agent of change (the tool), whereas in proto tool-use, only the object of change is detached and manipulated (Panger, 1998; Parker and Gibson, 1977).

Here we report a case of substrate manipulation by a brown howler monkey (Alouatta guariba clamitans Cabrera, 1940). It was recorded by F. Koch during a study of the ecology and behavior of a group of brown howlers at the Morro da Extrema (30°12′s, 51°04′W), Porto Alegre, Rio Grande do Sul, Brazil. On 16 October 2002, around 15:00 h, the sky became overcast and the wind picked up, signaling an approaching rainstorm. The study group began moving away. At 15:20 h, the group came to a gap in the canopy of about 2 m. All successfully leaped across the gap to the next tree except an infant (in the process of becoming independent from its mother). The branches were blowing about vigorously because of the high winds, and the infant stopped and vocalized (cried) while holding onto the end of the branch. The group members did not return to help the infant, which made no attempt to jump but continued vocalizing loudly, until eventually its mother went back to rescue it. In order to help her infant, the mother manipulated a nearby branch (without detaching it) of the tree she was in until it was positioned close to the infant. The infant immediately used this branch as a bridge to traverse the gap. Once safely across, it quickly climbed onto its mother's back. Given the configuration of the arboreal canopy, the only way for group members to cross the gap was by leaping from one tree to the other. A similar situation involving the same mother-infant pair was observed on a second occasion when there was a strong wind but a clear sky.

This note reports the observation of a complex behavior performed by a howler monkey to solve a problem commonly faced by arboreal primates. This is the first record of the manipulation of an object to help an infant howler monkey travel across a gap in the canopy. Previous reports indicate that adult howler monkeys may use their bodies to form a “bridge” in order to help immatures cross such gaps. According to the definition proposed by Beck (1975), this behavior cannot be considered as true tool use because the animal (mother) did not detach the branch used as a bridge from the tree. This use of the substrate as an object, however, can be classified as proto-tool-use or object substrate manipulation (sensu Parker and Gibson, 1977).