Phenotypic variation and covariation indicate high evolvability of acoustic communication in crickets

Abstract

Studying the genetic architecture of sexual traits provides insight into the rate and direction at which traits can respond to selection. Traits associated with few loci and limited genetic and phenotypic constraints tend to evolve at high rates typically observed for secondary sexual characters. Here, we examined the genetic architecture of song traits and female song preferences in the field crickets Gryllus rubens and Gryllus texensis. Song and preference data were collected from both species and interspecific F1 and F2 hybrids. We first analysed phenotypic variation to examine interspecific differentiation and trait distributions in parental and hybrid generations. Then, the relative contribution of additive and additive-dominance variation was estimated. Finally, phenotypic variance-covariance (P) matrices were estimated to evaluate the multivariate phenotype available for selection. Song traits and preferences had unimodal trait distributions, and hybrid offspring were intermediate with respect to the parents. We uncovered additive and dominance variation in song traits and preferences. For two song traits, we found evidence for X-linked inheritance. On the one hand, the observed genetic architecture does not suggest rapid divergence, although sex linkage may have allowed for somewhat higher evolutionary rates. On the other hand, P matrices revealed that multivariate variation in song traits aligned with major dimensions in song preferences, suggesting a strong selection response. We also found strong covariance between the main traits that are sexually selected and traits that are not directly selected by females, providing an explanation for the striking multivariate divergence in male calling songs despite limited divergence in female preferences.

Demography and selection shape transcriptomic divergence in field crickets

“…The theoretical support for speciation with gene flow driven by divergence in secondary sexual characters is very thin at best (van Doorn et al 2004;Weissing et al 2011;Servedio 2015). Here, we provide exciting and rare evidence for speciation with primary gene flow while combined evidence from phenotypic (Gray and Cade 2000), quantitative genetic (Blankers et al 2015b(Blankers et al , 2017, and genomic analyses (this study) highlight a role for selection on (acoustic) mating behavior in driving reproductive isolation. A compelling alternative interpretation of the findings here is that the peripatric origin of G. rubens has allowed for an initial phase of reduced gene flow; during this phase mating signals and preferences may have diverged sufficiently (aided by a founder effect following a population bottleneck) to maintain reproductive isolation during a subsequent phase of range expansion culminating into the contemporary, widespread, and largely overlapping species' distributions.…”

Field Crickets Compensate for Unattractive Static Long-Distance Call Components by Increasing Dynamic Signalling Effort

“…Two males had low pulse rates for G . texensis (<56 pulses per second) suggesting they might be misidentified ( sensu [18], Fig 1a). However, both males called at carrier frequencies above 5.5 kHz, with trill durations about 400 ms, and trill rates above 1.2 trills/second, suggesting they are G .…”

“…However, both males called at carrier frequencies above 5.5 kHz, with trill durations about 400 ms, and trill rates above 1.2 trills/second, suggesting they are G . texensis ( sensu [18], Fig 1c and 1d, respectively). Further, given our study site was located about 100km west of the most western end of the range of G .…”

“…In general, signal composition is determined by static traits, while measures of signalling effort are more dynamic [18]. This holds true in Texas field crickets, as carrier frequency, sound pressure level (amplitude), and pulse components (duration, pause, period, rate, and duty cycle) all have low coefficients of variation (CV<18; Table 1; sensu [11]), whereas trilling components (duration, pause, period, rate, and duty cycle) and calling effort all have high coefficients of variation CV>25; Table 1; sensu [11]).…”

Multivariate female preference tests reveal latent perceptual biases

“…Could males evolve to exploit the female bias and evolve long trills? This would require sufficient phenotypic and genetic variation, often observed in crickets [36,37], and available acoustic niche space. Both G. firmus and G#13 are sympatric with other Gryllus that produce long trills, which in theory could limit the sound space available to them.…”