Abstract
Measuring the relative contributions of milk and non-milk foods in the diets of primate infants is difficult from observations. Stable carbon (δ(13)C) and nitrogen (δ(15)N) isotopes in hair can be used to physiologically track infant feeding through development, but few wild studies have done so, likely due to the difficulty in collecting hair non-invasively. We assessed infant feeding at different ages in wild chimpanzees (Pan troglodytes) at Ngogo, Uganda using δ(13)C and δ(15)N of keratin in 164 naturally shed hairs from 29 infants (61 hairs), 6 juveniles (7 hairs), 28 mothers (67 hairs) and 14 adult males (29 hairs). Hairs were collected when they stuck to feces during defecation or from the ground after chimpanzees groomed or rested. We could not distinguish between the hairs of infants and mothers using strand length and diameter. Infants 1-2 years old were most enriched in (13)C and (15)N and showed means of 1.1‰ in δ(13)C and 2.1‰ in δ(15)N above their mothers. Infants at 2 years had hair δ(13)C values like those of their mothers, which suggests infants began relying more heavily on plants around this age. While mother-infant δ(13)C and δ(15)N differences generally decreased with offspring age, as is expected when infants rely increasingly more on independent foraging through development, milk seemed to remain an important dietary component for infants older than 2.5 years, as evidenced by continuing elevated δ(15)N. We showed that stable carbon and nitrogen isotopes in naturally shed hairs can feasibly detect trophic level differences between chimpanzee infants and mothers. Since it can mitigate some of the limitations associated with behavioral and fecal stable isotope data, the use of hair stable isotopes is a useful, non-invasive tool for assessing infant feeding development in wild primates.