Pollinating fig wasps' simple solutions to complex sex ratio problems: a review

Local mate competition (LMC) favours female biased clutch sex ratios because it reduces competition between brothers and provides extra mating opportunities for sons. Fig wasps seem to fit LMC model assumptions and lay female-biased sex ratios as predicted. These female biased sex ratios increase fitness greatly. In line with predictions, their sex ratios become less female-biased as the number of mothers laying in the same fig increases. However, this variation results in comparatively small fitness benefits compared to just biased ratios and data suggest substantial mismatches with LMC theory. The mismatches are due to several factors. (1) Multiple foundresses typically lay too many daughters. (2) Single foundress sex ratios are explained by sequential oviposition and ladies-last models. (3) Mortality that typically exceeds 10% may decouple the link between primary sex ratios, the focus of model predictions, and secondary sex ratios of adult wasps that are counted by researchers. (4) Model assumptions are frequently violated: (a) clutch sizes are unequal, (b) oviposition may not be simultaneous (c) cryptic/multiple wasp species inhabit the same host, (d) foundress numbers are systematically undercounted, (e) inbreeding coefficient calculations are inaccurate, and (f) male wasps sometimes disperse. These data and calculations suggest that alternative explanations must be considered seriously. Substantial data show that wasps typically lay most of their male eggs first followed by mostly female eggs require a new approach. These "slope" strategies result in more accurate sex ratios that are automatically adjusted to foundress number, own and relative clutch sizes and to sequential clutches. This effect will alter sex ratios in all species once the egg capacity of a fig is crossed or when interference reduces clutch sizes. In addition to this passive response, the females of about half the studied species have a conditional response that reduces female bias under higher foundress numbers by laying more sons. Therefore, wasps seem to use a very simple strategy that increases their fitness. Natural selection could have optimized parameters of the slope strategy and possibly the existence of the slope strategy itself. Variation in the slope strategy that is the result of natural selection is adaptive. Research should therefore focus on quantifying variables of this slope strategy. Currently, it is unclear how much of the variation is adaptive as opposed to being coincidental by-products.

Consequences of population structure for sex allocation and sexual conflict

Both sex allocation and sexual conflict can be modulated by spatial structure. However, how the interplay between the type of dispersal and the scale of competition simultaneously affects these traits in sub-divided populations is rarely considered. We investigated sex allocation and sexual conflict evolution in meta-populations of the spider mite Tetranychus urticae evolving under budding (pairing females from the same patch) or random (pairing females from different patches) dispersal and either local (fixed sampling from each subpopulation) or global (sampling as a function of subpopulation productivity) competition. Females evolving under budding dispersal produced less female-biased offspring sex ratios than those from the random dispersal selection regimes, contradicting theoretical predictions. In contrast, the scale of competition did not strongly affect sex allocation. Offspring sex ratio and female fecundity were unaffected by the number of mates, but female fecundity was highest when their mates evolved under budding dispersal, suggesting these males inflict less harm than those evolving under random dispersal. This work highlights that population structure can impact the evolution of sex allocation and sexual conflict. Moreover, selection on either trait may reciprocally affect the evolution of the other, for example via effects on fecundity.

The efficacy of natural selection in producing optimal sex ratio adjustments in a fig wasp species

Ever since Darwin's discovery of natural selection, we expect traits to evolve to increase organisms' fitness. As a result, we can use optimization models to make a priori predictions of phenotypic variation, even when selection is frequency-dependent. A notable example is the prediction of female-biased sex ratios resulting from local mate competition (LMC) and inbreeding. LMC models incorporate the effects of LMC and inbreeding. Fig wasp sex ratio adjustments fit LMC predictions well. However, the appropriateness of LMC models to fig wasps has been questioned, and the role that a coincidental by-product plays in creating the apparent fit has been clearly illustrated. Here, we show that the sex ratio adjustments of a fig wasp are the result of a dual mechanism. It consists of a standard facultative LMC response favoured by natural selection, as well as a mechanism that may be the result of selection, but that could also be a coincidental by-product. If it is a by-product, the fitness increase is coincidental and natural selection's role was limited to fine-tuning it for higher fitness returns. We further document a case of an apparent fitness-reducing sex ratio adjustment. We conclude that the use of the adaptationist approach demands that our understanding of traits must be remodelled continually to rectify spurious assumptions.

Co-foundress confinement elicits kinship effects in a naturally sub-social parasitoid

Kinship among interacting individuals is often associated with sociality and also with sex ratio effects. Parasitoids in the bethylid genus Goniozus are sub-social, with single foundress females exhibiting post-ovipositional maternal care via short-term aggressive host and brood defence against conspecific females. Due to local mate competition (LMC) and broods normally being produced by a single foundress, sex ratios are female-biased. Contests between adult females are, however, not normally fatal, and aggression is reduced when competing females are kin, raising the possibility of multi-foundress reproduction on some hosts. Here, we screen for further life-history effects of kinship by varying the numbers and relatedness of foundresses confined together with a host resource and also by varying the size of host. We confined groups of 1-8 Goniozus nephantidis females together with a host for 5+ days. Multi-foundress groups were either all siblings or all nonsiblings. Our chief expectations included that competition for resources would be more intense among larger foundress groups but diminished by both larger host size and closer foundress relatedness, affecting both foundress mortality and reproductive output. From classical LMC theory, we expected that offspring group sex ratios would be less female-biased when there were more foundresses, and from extended LMC theory, we expected that sex ratios would be more female-biased when foundresses were close kin. We found that confinement led to the death of some females (11% overall) but only when host resources were most limiting. Mortality of foundresses was less common when foundresses were siblings. Developmental mortality among offspring was considerably higher in multi-foundress clutches but was unaffected by foundress relatedness. Groups of sibling foundresses collectively produced similar numbers of offspring to nonsibling groups. There was little advantage for individual females to reproduce in multi-foundress groups: single foundresses suppressed even the largest hosts presented and had the highest per capita production of adult offspring. Despite single foundress reproduction being the norm, G. nephantidis females in multi-foundress groups appear to attune sex allocation according to both foundress number and foundress relatedness: broods produced by sibling foundresses had sex ratios similar to broods produced by single foundresses (ca. 11% males), whereas the sex ratios of broods produced by nonsibling females were approximately 20% higher and broadly increased with foundress number. We conclude that relatedness and host size may combine to reduce selection against communal reproduction on hosts and that, unlike other studied parasitoids, G. nephantidis sex ratios conform to predictions of both classical and extended LMC theories.

Stability in fig tree–fig wasp mutualisms: how to be a cooperative fig wasp

Fig tree–fig wasp mutualisms are diverse and underpin much biodiversity. The wasps (Agaonidae) are the sole pollinators of the trees (Ficus). Figs are enclosed inflorescences, each of which contains many small flowers. Female wasps (foundresses) enter receptive figs to spread pollen and to lay their eggs individually into fig flowers. As they oviposit, wasps also inject chemicals that transform individual flower ovaries into galls that will feed and house wasp offspring. For fig tree–fig wasp mutualisms to persist, the trees must set seed; therefore, the wasps have both to pollinate and to fail to gall all flower ovaries. However, wasps that avoid pollination costs and/or gall all flowers are predicted to outcompete more cooperative conspecifics, resulting in destabilisation of the mutualism. Here, I review the literature on why wasps pollinate by focusing on how trees reduce investment to unpollinated figs, resulting in ‘sanctions’ to wasps that fail to pollinate via reduced production of offspring. I also review the mechanisms that prevent wasps from galling all flowers, mainly those in monoecious Ficus, that also result in wasps predominantly galling longer flowers whilst leaving shorter flowers to become seeds. I make suggestions for future work and conclude by reaffirming why multiple processes promote stability in fig tree–fig wasp mutualisms.

Egg load is a cue for offspring sex ratio adjustment in a fig-pollinating wasp with male-eggs-first sex allocation

Fig-pollinating wasps (Agaonidae) only reproduce within fig tree inflorescences (figs). Agaonid offspring sex ratios are usually female-biased and often concur with local mate competition theory (LMC). LMC predicts less female-bias when several foundresses reproduce in a fig due to reduced relatedness among intra-sexually competing male offspring. Clutch size, the offspring produced by each foundress, is a strong predictor of agaonid sex ratios and correlates negatively with foundress number. However, clutch size variation can result from several processes including egg load (eggs within a foundress), competition among foundresses and oviposition site limitation, each of which can be used as a sex allocation cue. We introduced into individual Ficus racemosa figs single Ceratosolen fusciceps foundresses and allowed each to oviposit from zero to five hours thus variably reducing their eggs-loads and then introduced each wasp individually into a second fig. Offspring sex ratio (proportion males) in second figs correlated negatively with clutch size, with males produced even in very small clutches. Ceratosolen fusciceps lay mainly male eggs first and then female eggs. Our results demonstrate that foundresses do not generally lay or attempt to lay a 'fixed' number of males, but do 'reset to zero' their sex allocation strategy on entering a second fig. With decreasing clutch size, gall failure increased, probably due to reduced pollen. We conclude that C. fusciceps foundresses can use their own egg loads as a cue to facultatively adjust their offspring sex ratios and that foundresses may also produce more 'insurance' males when they can predict increasing rates of offspring mortality.