Dr Primrose Moss – Freelance Veterinary Writer

Paquet and Smiseth aim to elucidate whether prenatal maternal effects on offspring phenotype influence paternal behaviour, thereby enabling females to manipulate males by biasing sexual conflict over parental care. Since Trivers’ seminal work¹, research into sexual conflict over parental care has proliferated. Little research, however, exists into the possible contribution of maternal effects. Previous theoretical and empirical work on sexual conflict over parental care has focused on the sexually symmetrical behavioural mechanisms of negotiation, matching and sealed bid responses for the resolution of such conflict^2,3. Existing research into maternal effects on this conflict targets specific candidate mechanisms, such as maternal androgen deposition^3,4, but fails to provide conclusive evidence for the role of maternal effects in sexual conflict. Paquet and Smiseth³ discuss possible candidate mechanisms for such effects, including egg size, coloration and components such as juvenile hormone and ecdysone (in invertebrates), but emphasise the importance of initially conducting less mechanism-focused studies to find evidence of male manipulation by prenatal maternal effects as a result of ecological conditions. This paper follows on from the authors’ previous review in an innovative attempt to provide empirical evidence by modifying a specific, ecologically relevant, prenatal condition: male presence or absence. 

In this paper, prenatal environment was experimentally manipulated in the burying beetle, Nicrophorus vespilloides, either by allowing males to remain with females after mating, or by removing them. Cross-fostering was then implemented, crucially distinguishing prenatal and postnatal maternal effects. The investigation did not utilise specific hypotheses, instead exploring potential effects on parental care and food consumption by recording a range of variables including parental care, parent and offspring weight changes and lifespan. This produced several potentially significant results. The study’s conclusion is undermined, however, by a lack of evidence of long-term fitness benefits.

Methods

The study is generally clearly presented and carefully designed. Nicrophorus vespilloides is an ideal model species, with exceptionally well characterised behaviour and ecology. N. vespilloides exhibit facultative biparental care. This trait has been the subject of numerous studies, due to their rapid generation time and the ease of rearing them in laboratory conditions that enable natural behaviours. Benefits of their use include well established protocols to aid experimental design and their use of temporal cues for kin recognition⁵, facilitating cross-fostering. Further benefits include the ecological relevance of experimental treatments, since uniparental female care is observed in 39% of broods⁷, as well as the extension of research into prenatal maternal effects and sexual conflict into a new taxon, as previous research has targeted birds.

The experimental design controls for a wide range of relevant variables and prevents experimenter bias through blinding. Statistical controls are used to enable analysis of different factors. Considering the relatively small sample size additional factors, for example the sex of the mouse, could advantageously have been controlled; only limited research exists regarding potential impact of vertebrate hormones on insect development. The sample size is sufficient for assumptions of normality according to the Central Limit Theorem, although increasing sample size would reduce standard error and minimise the effect of any uncontrolled random variables.

The behavioural assessments were effective in assessing direct parental care. They were rigorous, being conducted over the entirety of the period of direct care⁸ using an instantaneous sampling method. This ensured a thorough assessment of direct care behaviour, in accordance with established protocol. These assessments were limited in that they failed to consider other, potentially relevant behaviours. They recorded only direct parental care, defined as regurgitation of food to the larvae, manipulation of carrion and regurgitation of food into the crater. Male parental care in laboratory conditions is typically redundant, however, as they focus on guarding the brood from competition and predation rather than providing direct care^7,9. Additional behaviours, including time spent with access to the carcass for feeding or offspring begging behaviour, could therefore have been recorded. 

Statistical analyses

The statistical analyses in the paper are somewhat perfunctorily described and raise concerns as to their reliability. The final and minimal models referred to in the paper and supplementary information are never clearly defined, whilst the insets in Figures 1, 2A and 2B are based on predicted values from these ‘final models’; the relevance of this is unclear. In addition, the use of certain statistical tests is not explained. Table S1 lists F values corresponding to F-tests, yet in Table S2 Z values seemingly corresponding to t-tests are recorded. The original data from the study are not available, but the statistical analysis would nonetheless be impossible to replicate due to this lack of explanation. 

Furthermore, some of the figures quoted in the supplementary information do not seem to correspond to the graphs provided in the paper. Table S1, for example, lists the estimate for the intercept (mean) of male weight change as 0.134, whilst Figure 2B shows the largest male weight change as approximately half this value. Such discrepancies impugn the validity of the statistical analyses.

Results & Conclusions

Larvae from eggs laid in the presence of a male were 3.4% lighter at hatching, which the authors suggest is due to reduced female investment in expectation of male assistance, redirecting costs to the postnatal period when they are shared.

This is supported by the literature; females in cooperatively breeding species lay smaller eggs with increasing numbers of helpers³. Alternatively, mothers may increase egg size in stressful environments. Males’ main parental role is to guard the brood from competition and predation, so male absence may act as a cue for increased risk of these threats. Maternal size is, however, positively correlated with egg size¹⁰, and the paper does not clarify whether the effect of maternal size has been statistically controlled. If it has not then random maternal weight differences between the treatment groups could easily account for the difference in brood weights. Regardless, males gain more weight when caring for lighter larvae; whilst this provides evidence of maternal effects in response to male presence, it is not suggested to constitute female manipulation of males.

Despite their lower weight at hatching, broods laid in the presence of a male gained more weight and were heavier at dispersal. Maternal effects on postnatal brood weight are therefore stronger than initial weight differences. Surprisingly, the increased weight gain of broods laid in the presence of a male was not the result of increased parental care. Both parents exhibited similar levels of parental care across the treatment conditions, with the exception of paternal care at 25 hours after hatching which was found to be decreased in broods laid with a male present. This finding is, however, based on only 14 males that provided care out of 60. The authors did not analyse care at 49 hours as only 6 males provided care. Arguably, however, care at 25 hours still provides too small a set of data points for valid analysis. Further assessment of male behaviour, including whether the majority of males never provided care or were simply resting at the time of the survey, may have provided a better insight into differences in male behaviour between the treatments.

The correlation of male weight change with treatment and brood weight at dispersal is one of the most noteworthy findings. While both male and female weight change are negatively correlated with brood weight at dispersal, only male weight change sufficiently explains the higher weight at dispersal of broods laid in the presence of a male. When male weight change was fitted to a linear model of brood weight it eliminated the effects of treatment and brood weight at hatching (Table S3). Maternal effects are thus argued to cause increased larval weight gain at the expense of male weight gain. The authors claim that the fitness gain to the offspring of increased larval weight and fitness cost to the male as a result of these maternal effects constitutes female manipulation of the male, as outlined in their paper’s introduction. The authors fail, however, to establish a causal link between male weight change and brood weight at dispersal. 

Paquet and Smiseth conclude that reduced weight gain in males is due to female suppression of male food consumption. Although credible, this is based on male weight change being used as a proxy for food consumption, with no evidence that this is accurate. Other factors such as activity level may affect weight change. Recording additional behavioural measures such as male presence on the carcass might have provided further evidence for this claim. Alternatively, future studies could measure food consumption directly using tracers in the carcass.

An interesting result was that male weight change was positively correlated with care at 25 hours after hatching. This finding is limited by the small number of data points, but is supported by previous work showing that parents who spend more time feeding offspring had greater access to food⁶. Previous studies have shown that male (but not female) provisioning is significantly influenced by social and non-social factors¹¹; the correlation of care with weight change provides evidence that changes to offspring phenotype may affect male weight change via altered male behaviour.

Finally, the paper investigated the influence of prenatal maternal effects on long-term fitness consequences. Notably, the only significant result was that offspring laid in the absence of a male had a longer lifespan after eclosion. The authors hypothesise that this may be due to selective disappearance of larvae before eclosion, but further investigation of pre and post eclosion lifespans would be required to confirm this. Regardless, the only significant effect was the reverse of that expected due to manipulation. This casts doubt on the study’s conclusions which require a fitness benefit to be observed.

The introduction to the paper states that ‘to demonstrate female manipulation, it is crucial to document fitness benefits to females and/or offspring and fitness costs to males.’ In light of the reduced lifespan of offspring laid with a male present, this criterion is apparently unfulfilled. The authors focus on larval weight at dispersal as an indicator of increased fitness but this does not necessarily provide any inherent benefit. For this to contribute to fitness, it would have to influence the offspring’s reproductive success. This may, perhaps, be fulfilled by the significant influence of brood weight at dispersal on offspring adult size, as shown in Table S4. Body size determines the outcome of intraspecific competition over carcasses¹², so increased size would be of significant benefit. Whether this increase in adult size can be attributed to the effect of treatment on brood weight at dispersal is, however, unclear and the omission of offspring adult size from the main paper suggests it might not be. Direct measurement of offspring fecundity may have provided a more convincing measure of fitness. For female manipulation to be evolutionarily stable, the fitness benefit to offspring must be sufficient to offset the cost of reduced weight incurred by males, resulting in incentivisation rather than deception. If prenatal maternal effects due to male presence do not have any positive effect beyond increased brood weight at dispersal, then the paper has failed to provide sufficient evidence of the benefits of female manipulation.

This paper has produced various interesting results which suggest that the focus of further research should be on male food consumption rather than direct paternal care. The suggestion that females lay lighter eggs when expecting male assistance in facultative biparental species is also an area deserving of further research. The study would have been of greater use, however, had it investigated further variables such as male presence on the carcass in order to better inform future research. The use of larval weight at dispersal as a measure of fitness is unsatisfactory: a lack of evidence for other fitness benefits, in addition to unclear statistical analyses, casts a measure of doubt on the validity of the paper’s conclusions. Nevertheless, the study took an original approach to studying prenatal maternal effects in sexual conflict and provides a useful basis for further research.

References

1          Trivers, RL. Parental investment and sexual selection. In: Campbell B, editor. Sexual Selection and the Descent of Man 1871-1971. Chicago, Ill.: Aldine; 1972, pp. 139– 79.

2          Lessells CM. Sexual conflict. In: Royle NJ, Smiseth PT, Kölliker M, editors. The Evolution of Parental Care. Oxford: OUP Oxford; 2014.

3          Paquet M, Smiseth P. Maternal effects as a mechanism for manipulating male care and resolving sexual conflict over care. Behavioral Ecology. 2015;27(3):685-694. 

4          Moreno-Rueda G. Yolk androgen deposition as a female tactic to manipulate paternal contribution. Behavioral Ecology. 2007;18(2):496-498. 

5          Müller J, Eggert A. Time-dependent shifts between infanticidal and parental behavior in female burying beetles a mechanism of indirect mother-offspring recognition. Behavioral Ecology and Sociobiology. 1990;27(1). 

6          Lock J, Smiseth PT, Moore A. Selection, Inheritance, and the Evolution of Parent‐Offspring Interactions. The American Naturalist. 2004;164(1):13-24. 

7          Scott M. THE ECOLOGY AND BEHAVIOR OF BURYING BEETLES. Annual Review of Entomology. 1998;43(1):595-618.

8          Smiseth PT, Darwell C, Moore A. Partial begging: an empirical model for the early evolution of offspring signalling. Proceedings of the Royal Society B: Biological Sciences. 2003;270(1526):1773-1777. 

9          Muller J, Eggert A, Sakaluk S. Carcass maintenance and biparental brood care in burying beetles: are males redundant? Ecological Entomology. 1998;23(2):195-200. 

10        Steiger S. Bigger mothers are better mothers: disentangling size-related prenatal and postnatal maternal effects. Proceedings of the Royal Society B: Biological Sciences. 2013;280(1766):20131225. 

11        Smiseth PT. Behavioral dynamics between caring males and females in a beetle with facultative biparental care. Behavioral Ecology. 2004;15(4):621-628. 

12        Otronen M. The effect of body size on the outcome of fights in burying beetles (Nicrophorus). Ann. Zool. Fenn. 1988;25:191-201.