Population and group-specific behavioral differences have been taken as evidence for animal cultures, a notion that remains controversial. Skeptics argue that ecological or genetic factors, rather than social learning, provide a more parsimonious explanation. Work with captive chimpanzees has addressed this criticism by showing that experimentally created traditions can be transmitted through social learning. Recent fieldwork further suggests that ecological and genetic factors are insufficient to explain the behavioral differences seen between communities, but the data are only observational. Here, we present the results of a field experiment that compared the performance of chimpanzees (P. t. schweinfurthii) from two Ugandan communities, Kanyawara and Sonso, on an identical task in the physical domain-extracting honey from holes drilled into horizontal logs. Kanyawara chimpanzees, who occasionally use sticks to acquire honey, spontaneously manufactured sticks to extract the experimentally provided honey. In contrast, Sonso chimpanzees, who possess a considerable leaf technology but no food-related stick use, relied on their fingers, but some also produced leaf sponges to access the honey. Our results indicate that, when genetic and environmental factors are controlled, wild chimpanzees rely on their cultural knowledge to solve a novel task.