Abstract 1. Non-consumptive effects of predation can strongly impact reproduction and demography of prey species. Still, the underlying mechanisms that drive non-consumptive effects are not fully understood, and the circumstances under which chronic physiological stress may mediate these effects remain unclear. 2. Benefiting from over 23 years of environmental, physiological and demographic data, we tested the hypothesis that predation risk may impair reproduction of mountain goats through chronic elevation of physiological stress. We conducted path analyses to assess the relationships between predation risk, faecal glucocorticoid metabolites and hair cortisol concentration, and reproduction, while taking into account the potential effects of age class, sex, body mass, season, and within individual variation in glucocorticoid concentration. 3. Predation risk had a direct positive effect on the average annual faecal glucocorticoid concentration in the population, which in turn negatively affected the proportion of reproductive females. The same pattern was observed with hair cortisol concentration, but these results were inconclusive potentially due to methodological challenges in estimating annual average of hair cortisol at the population level. 4. Our study presents one of the first robust evidence that stress-mediated breeding suppression can occur in a wild ungulate following increased predation risk, thereby providing a major insight on the mechanisms underlying non-consumptive effects of predation in wild mammals.

The data set contains four sheets of data collected on marked mountain goats at Caw Ridge, Alberta, Canada between 1994-2016. The first and second sheets include individual data on glucocorticoid concentration (respectively FGM and HCC), age, sex and age-specific mass as well as annual data on reproduction, predation risk, population size, and faecal crude protein peak date. This data was used in all the analyses presented in the main manuscript and to produce Figure 3, Figure 4, Figure 5, Figure S4.1, Figure S5.1, and Figure S5.2. The third and fourth sheets is restricted to females and include individual data on glucocorticoid concentration (respectively FGM and HCC), reproduction, age, and age-specific mass as well as annual data on predation risk, population size, and faecal crude protein peak date. This data was used for the individual level analysis presented in supplementary material S6 and to produce Figure S6.1.