Maternal behaviour and the evolution of chemical signalling by offspring in the European earwig (Forficula auricularia)

Parental care provided to current offspring such as food and/or protection increases offspring development and survival which contributes to the parent’s fitness. However parental investment by means of time and energy may also reduce future chance to reproduce and therefore entails parent’s lifetime reproductive success. We expect parents to adjust their parental investment equally among their different offspring within and between broods in order to maximise their fitness. From the point of view of one current offspring, parental care should be rather positively biased towards itself compared to current or future siblings. Thus the different genetic interest over the duration and amount of parental investment is expected to lead to parent-offspring conflict. Resolution of this conflict may be achieved by the evolution of an offspring solicitation signal that regulates parental care as predicted by various theories and models. Although several empirical studies have supported the presence of offspring solicitation signals, mostly in birds, the origin and driving forces for the evolved signalling function have not been clearly demonstrated.
The European earwig, Forficula auricularia, displays facultative maternal care, i.e. offspring can survive in absence of a caring mother but have significantly higher survival when attended by a mother. Thus one can test the on-going selection for an offspring cue to evolve a function of solicitation signal, which is predicted to be condition-dependent and regulating maternal care. Chemical cues are the main means of communication in insects (i.e. pheromones), which also regulate reproductive physiology (i.e. hormones). Therefore I decided for my thesis to explore the possible evolution of chemical signalling by offspring in the context of maternal care in the European earwig.
In a first experiment, I manipulated the nutritional condition (low-food, LF, versus high-food, HF) of earwig first instar nymphs and extracted their cuticular hydrocarbons (CHCs). Caring mothers were presented to these different extracts or a solvent control (C) and their effects on maternal foraging as well as food provisioning to their brood were measured. By gas-chromatography coupled with mass-spectrum analysis, I found that nymphs of different nutritional state produce similar total amount of CHCs but they differ in the relative abundance of specific chemical compounds, particularly long-chain CHCs. Mothers exposed to offspring condition-dependent CHCs adjusted their maternal care behaviours. They foraged and later provisioned significantly more food to their brood when exposed to extract from HF. This first result demonstrated that CHCs of offspring are used as solicitation signals and that mothers may select for an offspring chemical signal of quality.
In a second experiment, I investigated the effects of offspring condition-dependent chemical signals on the maintenance of maternal care among broods and the distribution of maternal food within broods. Mothers were isolated from their 1st instar brood for 3 days and continuously exposed to chemical signals extracted from broods of experimentally manipulated nutritional state (HF, LF). After re-introducing mothers to their brood, a range of maternal behaviours were quantified. I found that earwig mothers groomed their offspring significantly more after exposure to chemical extract from HF brood in comparison with mothers exposed to extract from LF brood, which in turn displayed significantly more aggressive behaviour. Furthermore, I manipulated offspring individual nutritional condition within the brood to evaluate the effect of offspring state on the within-brood food distribution. Within broods, poorly fed individuals received significantly more food than well-fed individuals. These contrasting results of offspring condition-dependent signals observed at the brood and individual levels suggest various selective pressures, such as scramble competition within brood and maternal selection among broods, shaping offspring solicitation signals.
Finally, to test whether offspring chemical signals can per se manipulate the lifetime reproductive success of mothers, I measured long term consequences of exposure to offspring chemical signals on mothers’ residual fecundity. The probability to have a second clutch by females was not affected by offspring chemical signals. However, the predictability for females to lay a second clutch within a certain interval was significantly affected by the condition-dependent chemical signals produced by offspring. The date of laying a second clutch was highly related to date of first clutch laying/hatching and strongly predictable when females were exposed to extract from HF offspring. The importance of timing of the second clutch may be critical for seasonal species like F. auricularia in order to ensure offspring survival. This last result confirms the potential for offspring chemical signals to manipulate maternal future fecundity, yet mothers may actively select for this offspring signal of quality in their best interest in order to optimally adjust their investment between current and future broods.
In conclusion, I showed that earwig first instar nymphs produce CHCs that vary in their relative abundances depending on offspring nutritional state. Earwig mothers adjust their maternal care behaviours (food provisioning, grooming vs. aggressiveness) according to these condition-dependent . Higher food provisioning and more grooming by mothers exposed to HF brood extract suggest maternal selection for offspring chemical cues of quality. Finally, the potential of offspring chemical cues per se to influence future maternal fecundity confirms their evolving function as solicitation pheromone in the context of maternal care.