Detecting non-binomial sex allocation when developmental mortality operates.

Area of Science:

• Evolutionary ecology
• Behavioral ecology
• Population genetics

Background:

• Optimal sex allocation theory explains non-random offspring sex ratios under specific conditions like population subdivision.
• Detecting non-binomial sex allocation is hindered by developmental mortality, where offspring sex is unknown until maturity.

Purpose of the Study:

• To address limitations in current methods for detecting non-binomial sex allocation when developmental mortality occurs.
• To develop and present a novel Bayesian model selection approach for improved detection of non-random sex ratios.

Main Methods:

• Developed a new procedure incorporating explicit models of sex allocation and mortality.
• Utilized Bayesian model selection and calculated Bayes factors to compare binomial, double binomial, and multiplicative binomial distributions.
• Employed a simulation study and analyzed empirical data from parasitoid wasps.

Main Results:

• Current methods have limited ability to detect non-binomial sex allocation in the presence of developmental mortality.
• The new Bayesian approach significantly increases the power to detect non-binomial sex allocation, especially with common mortality.
• The approach offers improved descriptive capability and quantifies evidence for different hypotheses.

Conclusions:

• The novel Bayesian method provides a more powerful and accurate tool for analyzing sex allocation patterns.
• This approach is crucial for understanding evolutionary ecology, particularly in species with significant developmental mortality.
• The R package facilitates the application of these advanced statistical methods to real-world data.