Co-ordination of the sexual signalling system and the genetic basis of differentiation between populations in the brown planthopper, Nilaparvata lugens

Substrate-transmitted sexual signals and female responses to synthetic male signals were analysed for three populations of brown planthopper, Nilaparvata lugens (Homoptera: Delphacidae). Mean female preferences closely match mean male signal pulse repetition frequencies (PRF) in each population, despite marked differences among populations in PRF of both male and female signals. Female response windows (the range of male signal PRFs that elicit responses) were similar in all populations and wide relative to male signal variation. The Wright-Castle-Lande method was used to estimate the number of loci affecting differences between populations in male and female sexual signals and female preferences. Estimates obtained using the Lande formulae were compared with estimates from an iterative model-fitting approach. No sex linkage was observed. The characters conformed to the expectations of additive inheritance, with directional dominance in two cases. Estimates for the number of loci involved in signal characters were low, ranging from 1.4 to 5.3. After corrections for bias in the estimation method, estimates range from 2 to 16 but are still substantially lower than the recombination index of about 25. Estimates of the number of loci influencing female preference were unreliable. The two estimation methods gave compatible results but the iterative estimation method made better use of the data and gave clearer guidance on the support for the estimates. The results are discussed in the light of potential causes of divergence in sexual signalling systems.