Current concerns about climate change has led to intensive research attempting to understand how climate driven stressors affect the performance of organisms, in particular on offspring of many invertebrate and fish. Although stressors are likely to act on several stages of the life cycle, little is known about their action across life phases, for instance how multiple stressors experienced simultaneously in the maternal environment can modulate the responses to the same stressors operating in the offspring environment. Here, we study how performance of offspring of a marine invertebrate (shore crab Carcinus maenas) in response to two stressors (temperature and salinity) changes if the same stressors are experienced at the time of embryogenesis in cohorts of mothers breeding eggs at different seasons. On average, offspring responses were antagonistic: high temperature mitigated the negative effects of low salinity on survival. However, the magnitude of the response was modulated by the temperature and salinity conditions experienced by egg-carrying mothers. Performance also varied among cohorts, suggesting a role for genetic variation, and/or maternal conditions prior to fertilisation. We speculate that similar maternal effects may occur in other brooding organisms, as a consequence of anthropogenic modification of the environment.

Berried females collected in two seasons (SEASON) were kept in 4 treatments (2 EMBRYONIC SALINITY x 2 EMBRYONIC TEMPERATURE). After embryonic development was completed and larvae hatched, they were distributed in 6 treatments (2 LARVAL SALINITY x 3 LARVAL TEMPERATURE). Survival and developmental time  of the first stage was determined.

Survival data (proportion) were first adjusted using the equation p’= [p(n-1)/n+0.5]/n, (n=10 individuals) and then analysed after logistic (= logit) transformation³⁹, following Griffen et al. 2016. For survival, we applied a five-way factorial model containing embryonic salinity (E_S), embryonic temperature (E_T), larval salinity (L_S), larval temperature (L_T) and season (S).

The duration of development was analysed using the data corresponding to the larvae reared in seawater because we had high mortality rates at a lower salinity of 20 PSU (see Results). The starting model was reduced to a four-way factorial model (the factor “larval salinity”, L_S, was dropped), keeping female (F) as a random factor.