skip to main content

Title: Variability of weaning age in mountain gorillas ( Gorilla beringei beringei )
Abstract Objectives

Weaning is a key life history milestone for mammals that represents both the end of nutritional investment from the perspective of mothers and the start of complete nutritional independence for the infants. The age at weaning may vary depending on ecological, social, and demographic factors experienced by the mother and infant. Bwindi mountain gorillas live in different environmental conditions and have longer interbirth intervals than their counterparts in the Virunga Volcanoes, yet other life history characteristics of this population remain less well known. We use long‐term data from Bwindi Impenetrable National Park, Uganda to examine factors related to weaning age.

Materials and methods

We analyzed data on infants born in four mountain gorilla groups in Bwindi to quantify their age of weaning (defined as last nipple contact) and to test if the sex of offspring, parity, and dominance rank of mother influences age of weaning. We also compared the age at weaning and time to conception after resumption of mating in Bwindi and Virunga gorillas.


Bwindi gorillas were weaned at an average age of 57.5 months. No difference was found between age of weaning for primiparous and multiparous mothers, nor did maternal dominance rank influence age of weaning, but sons were weaned at a later age than daughters. The majority of Bwindi mothers were still suckling when they resumed mating and mothers generally conceived before they weaned their previous offspring. The age of weaning was significantly later in Bwindi than in Virunga gorillas. After mothers resumed mating, the time to conceiving the next offspring was not significantly longer for Bwindi females than Virungas females (6 vs. 4 months).


Later weaning age for sons than daughters is similar to findings of other studies of great apes. Bwindi mountain gorillas are weaned at approximately the same age as western gorillas and chimpanzees, which is more than a year later than Virunga mountain gorillas. The results of this study suggest that variation in ecological conditions of populations living in close geographic proximity can result in variation in life history patterns, which has implications for understanding the evolution of the unique life history patterns of humans.

more » « less
Award ID(s):
Author(s) / Creator(s):
Publisher / Repository:
Wiley Blackwell (John Wiley & Sons)
Date Published:
Journal Name:
American Journal of Physical Anthropology
Page Range / eLocation ID:
p. 776-784
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract Objectives

    A key feature of human life history evolution is that modern humans wean their infants 2–4 years earlier on average than African apes. However, our understanding of weaning variation in apes remains limited. Here we provide the first such report in chimpanzees by examining weaned age variation using long‐term data from Gombe National Park, Tanzania.

    Material and Methods

    We analyzed 41 years of observational behavioral data from 65 offspring of 29 mothers to examine the relationships between weaned age (defined as cessation of suckling) in wild chimpanzees and maternal age, dominance rank and parity, and offspring sex. We used Cox proportional hazards regression with mixed effects to model time to weaning and to examine potential sources of variation in offspring weaned age.


    We found that male offspring were less likely than female offspring to wean by a given age and that weaned age of males varied more than weaned age of females. In addition, maternal dominance rank interacted with offspring age, such that low‐ranking mothers were less likely to wean offspring early, but this effect decreased with offspring age.


    We found that male offspring and offspring of low‐ranking females were less likely to wean early, but did not find evidence for variable weaning according to maternal age or parity. As more data accumulate, we will be better able to disentangle the effects of maternal dominance rank, age and parity. Such studies will not only provide a richer understanding of living ape life history characteristics, but will also provide an important framework for understanding the evolution of early weaning in humans.

    more » « less
  2. Abstract Objectives

    Availability of fruit is an important factor influencing variation in great ape foraging strategies and activity patterns. This study aims to quantify how frugivory influences activity budgets across age‐sex classes of mountain gorillas in Bwindi Impenetrable National Park, Uganda.

    Materials and methods

    Daily proportions of fruit‐feeding and activity budgets were calculated using 6 years of observational data on four habituated groups. We fitted generalized linear mixed models to test for age‐sex differences in the amount of fruit‐feeding, and to test whether these factors influence the proportion of time spent feeding, resting, and traveling.


    Bwindi mountain gorillas spent on average 15% of feeding time consuming fruit, with monthly variation ranging from 0 to 70%. Greater amounts of fruit‐feeding were associated with more time feeding and traveling, and less time resting. Immatures tended to spend more feeding time on fruit than adults, but less overall time feeding and more time traveling. There were no significant differences in the amount of fruit‐feeding and overall feeding time between adult females and silverback males, despite differences in body size.


    This study confirms that gorillas are frugivorous, and only the Virunga mountain gorilla population can be characterized as highly folivorous. Along with other frugivorous great apes, Bwindi mountain gorillas alter their activity patterns in response to varying amounts of fruit in their diet. A better understanding of how variable ecological conditions can drive diversity even within a subspecies has important implications for understanding relationships between ecology, body size, and foraging strategies in great apes.

    more » « less
  3. Abstract Objectives

    Several theories have been proposed to explain the impact of ecological conditions on differences in life history variables within and between species. Here we compare female life history parameters of one western lowland gorilla population(Gorilla gorilla gorilla) and two mountain gorilla populations(Gorilla beringei beringei).

    Materials and Methods

    We compared the age of natal dispersal, age of first birth, interbirth interval, and birth rates using long‐term demographic datasets from Mbeli Bai (western gorillas), Bwindi Impenetrable National Park and the Virunga Massif (mountain gorillas).


    The Mbeli western gorillas had the latest age at first birth, longest interbirth interval, and slowest surviving birth rate compared to the Virunga mountain gorillas. Bwindi mountain gorillas were intermediate in their life history patterns.


    These patterns are consistent with differences in feeding ecology across sites. However, it is not possible to determine the evolutionary mechanisms responsible for these differences, whether a consequence of genetic adaptation to fluctuating food supplies (“ecological risk aversion hypothesis”) or phenotypic plasticity in response to the abundance of food (“energy balance hypothesis”). Our results do not seem consistent with the extrinsic mortality risks at each site, but current conditions for mountain gorillas are unlikely to match their evolutionary history. Not all traits fell along the expected fast‐slow continuum, which illustrates that they can vary independently from each other (“modularity model”). Thus, the life history traits of each gorilla population may reflect a complex interplay of multiple ecological influences that are operating through both genetic adaptations and phenotypic plasticity.

    more » « less
  4. Abstract Research Highlights

    Having more maternal kin (mother and siblings) is associated with spending more time near others across developmental stages in both male and female capuchins.

    Having more offspring as a subadult or adult female is additionally associated with spending more time near others.

    A mother's average sociality (time near others) is predictive of how social her daughters (but not sons) become as juveniles and subadults (a between‐mother effect).

    Additional variation within sibling sets in this same maternal phenotype is not predictive of how social they become later relative to each other (no within‐mother effect).

    more » « less
  5. Abstract

    Lactation length and weaning age provide important information about maternal investment, which can reflect the health and nutritional status of the mother, as well as broader reproductive strategies in mammals. Calcium‐normalized strontium (Sr) and barium (Ba) concentrations in the growth layers of mammalian teeth differ for nursing animals and those consuming non‐milk foods, thus can be used to estimate age‐at‐weaning. To date, this approach has been used only for terrestrial animals, and almost exclusively for primates.

    The goal of this study was to determine whether Sr and Ba concentrations in the cementum of Pacific walrusOdobenus rosmarus divergensteeth can be used to estimate weaning age. Teeth from 107 walruses were analysed using laser ablation inductively coupled plasma mass spectrometry, and calcium‐normalized88Sr and137Ba concentrations were quantified.

    For most walruses, both Sr and Ba concentrations exhibited rapid changes in early life. Ba concentrations matched closely with expected patterns in the published literature, rapidly declining from high to low concentrations (typically from ~10 ppm to ~5 ppm). In contrast, Sr exhibited a pattern opposite to that presented in studies of terrestrial mammals, appearing nearly identical to Ba (typically declining from ~400 ppm to ~200 ppm). To explain these findings, we present conceptual models of the factors generating weaning signals in Sr and Ba for terrestrial mammals, as well as a new, hypothetical model for walruses. Both a visual and mathematical approach to weaning age estimation indicated a median weaning age of walruses at the end of the second year of life (in the second dark layer of the tooth cementum), with many walruses estimated to have weaned in their third year of life, and a smaller group weaning in their fourth or fifth year. This is later than expected, given a published estimate of walrus weaning at 18–24 months.

    These results do not conclusively support the use of tooth Sr and Ba for estimating weaning age in walruses, and further research is warranted to better understand the drivers of the observed patterns of Ba and Sr accumulation in walrus teeth.

    more » « less