Multivariate Phenotypic Evolution: Divergent Acoustic Signals and Sexual Selection in Gryllus Field Crickets
- 336 Downloads
Predicting the response to selection is at the core of evolutionary biology. Presently, thorough understanding of the effects of selection on the multivariate phenotype is lacking, in particular for behavioral traits. Here, we compared multivariate acoustic mating signals among seven field cricket species contrasting two selection regimes: (1) species producing songs with long trains of pulses for which preference functions for acoustic energy (chirp duty cycle) are linear and likely exert strong directional selection (‘trillers’); (2) species producing songs consisting of short chirps and for which preference functions for chirp duty cycle are concave and directional selection is likely weak or absent (‘chirpers’). We compared the phenotypic variance–covariance matrix (P) among species and uncovered two main patterns: First, surprisingly, pulse rate and chirp rate were positively correlated in six of seven species thus suggesting phenotypic coupling of timescales. Second, chirp rate and chirp duty cycle also covaried, but the direction of covariation differed between chirpers (positive) and trillers (negative). Multi-population Bayesian methods for matrix comparisons, Krzanowski’s subspace comparison and tensor analysis, revealed significant variation in P unrelated to phylogenetic distance, but strongly contrasting chirpers and trillers. We also found differences in the predicted selection response between chirpers and trillers. We thus report that variation in P is higher between than within selection regimes. Although effects from drift and shared ancestry cannot be fully excluded, these findings highlight a role for sexual selection in shaping patterns of phenotypic covariation that can ultimately affect the evolutionary trajectory of a multivariate mating signal.
KeywordsP matrix Sexual selection Acoustic communication Gryllus Bayesian
The manuscript strongly benefitted from comments by Emma Berdan, Jonas Finck, and Michael Reichert and peer review by Derek A. Roff, Katherine Willmore, and four anonymous reviewers. The performed experiments comply with the “Principles of animal care”, publication No. 86-23, revised 1985 of the National Institute of Health, and also with the current laws of Germany. The authors declare no conflict of interest. Data will be deposited in the Dryad Digital Repository. This study is part of the GENART project funded by the Leibniz Association (SAW-2012-MfN-3).
Compliance with Ethical Standards
Conflict of interest
The authors declare no conflict of interest.
- Flury, B. (1988). Common principal components and related multivariate models. New York: Wiley.Google Scholar
- Gerhardt, H. C., & Huber, F. (2002). Acoustic communication in insects and anurans. Chicago: The University of Chicago Press.Google Scholar
- Gray, D. A., Gabel, E., Blankers, T., & Hennig, R. M. (2016a). Multivariate female preference tests reveal latent perceptual biases. Proc R Soc B (in review)Google Scholar
- Hadfield, J. D. (2012). MCMCglmm course notes. http://cran.r-project.org/web/packages/MCMCglmm/vignettes/CourseNotes.pdf.
- Lynch, M., & Walsh, B. (1998). Genetics and analysis of quantitative Traits. Sunderland, MA: Sinauer.Google Scholar
- R Development Core Team, R. (2015). R: A language and environment for statistical computing. In R. D. C. Team (Ed.), R foundation for statistical computing. R Foundation for Statistical Computing.Google Scholar
- Walker, T. J. (2015). Crickets. In Singing insects of North America. http://entnemdept.ifas.ufl.edu/walker/Buzz/crickets.htm.