Ethology Divergence in Female Duetting Signals in the Enchenopa binotata Species Complex of Treehoppers (Hemiptera: Membracidae) Rafael L. Rodrı´guez & Reginald B. Cocroft Biological Sciences, University of Missouri–Columbia, Columbia, MO, USA Correspondence Abstract Rafael L. Rodrı´guez, 223 Tucker Hall, Biological Sciences, University of Missouri– Sexual communication often involves signal exchanges between the Columbia, Columbia, MO 65211–7400, USA. sexes, or duetting, in which mate choice is expressed through response E-mail: [email protected] signals. With both sexes acting as signalers and receivers, variation in the signals of males and females may be important for mate choice, Received: May 18, 2006 reproductive isolation, and divergence. In the Enchenopa binotata species Initial acceptance: June 5, 2006 complex – a case study of sympatric speciation in which vibrational Final acceptance: June 26, 2006 (S. A. Foster) duetting may have an important role – male signals are species-specific, doi: 10.1111/j.1439-0310.2006.01285.x females choose among males on the basis of signal traits that reflect spe- cies and individual differences, and female preferences have exerted divergent selection on male signals. Here, we describe variation in female signals in the E. binotata species complex. We report substantial species differences in the spectral and temporal features of female sig- nals, and in their timing relative to male signals. These differences were similar in range to differences in male signals in the E. binotata complex. We consider processes that might contribute to divergence in female sig- nals, and suggest that signal evolution in the E. binotata complex may be influenced by mate choice in both sexes. patterns of reproductive isolation and sexual selec- Introduction tion that may lead to population divergence. Sexual communication often takes the form of duet- Female signals are often simpler and shorter than ting – signal interchanges between males and male signals, with the timing relative to male signals females – across a variety of taxonomic groups and being the feature that most varies between species signaling modalities (Claridge 1985; Henry 1994; (Lloyd 1966; Claridge 1985; Bailey & Hammond Greenfield 2002; Bailey 2003; Hall 2004; Virant-Dob- 2003; Bailey 2003). However, female signals can also erlet & Cokl 2004; Cocroft & Rodrı´guez 2005). resemble male signals in their spectral and temporal In the mating systems of many duetting species, features, and may likewise be species- or population- female signals express mate choice (Claridge 1985; specific (Henry 1994; Wells & Henry 1998; Virant- Henry 1994; Bailey 2003; Virant-Doberlet & Cokl Doberlet & Cokl 2004). Variation in signal length 2004; Cocroft & Rodrı´guez 2005). In these cases, and specificity may reflect the balance of natural males must signal attractively, and also detect and and sexual selection stemming from mate choice and react appropriately to female responses; females, on signaling costs (Greenfield 2002; Bailey 2003): very the other hand, must choose among males, and short female response signals may evolve when the effectively express their choice with their signals. risk of detection by predators is high, while longer Because each sex is a signaler and a receiver, vari- responses may be favored when male mate choice ation in the signals of males and females may influ- has a relatively stronger role. Comparing the length ence their reproductive success and contribute to and specificity of the signals of males and females Ethology ª 2006 The Authors Journal compilation ª 2006 Blackwell Verlag, Berlin 1 Divergence in Female Duetting Signals in Enchenopa Treehoppers R.L. Rodrı´guez & R.B. Cocroft can thus generate hypotheses about the selection the stronger female preferences (Rodrı´guez et al. regimes under which they evolve, and about their 2006), indicating that female preferences have been contribution to variation in reproductive success and important in male signal divergence, and suggesting divergence. that sexual communication may be important in Here, we describe variation in female signals and the diversification of this clade. compare it with variation in male signals in a clade Here, we turn to the other side of sexual communi- of plant-feeding insects where duetting sexual com- cation in the E. binotata complex, and assess variation munication may be important in speciation. The in female signals. We studied three aspects of female Enchenopa binotata complex of treehoppers consists signals: frequency, length, and timing relative to male of 11 or more species that specialize on different signals. Male signal frequency and length are import- host plants across eastern North America, and it is ant for female choice in the E. binotata complex (Rod- a case study of sympatric speciation through shifts rı´guez et al. 2004, 2006), and differences in female to novel host plants (Wood & Guttman 1983; Lin & signals may in turn influence male-mate choice. Wood 2002; Cocroft et al. 2007). Host shifts pro- Since duetting is an interactive process, the relative mote reproductive isolation between populations on timing of male and female signals can influence pair ancestral and novel hosts, through a combination formation through their effects on detection by the of changes in life history timing and high host receiver (Bailey & Hammond 2003; Bailey 2003). We fidelity; host shifts also result in divergent ecological focused on four species whose male signals span selection among host plants (reviewed in Wood much of the range of variation in the complex. 1993). As with many insects that live on plants (Claridge 1985; Henry 1994; Virant-Doberlet & Cokl Methods 2004; Cocroft & Rodrı´guez 2005), the members of the E. binotata complex communicate with plant- Experiments were performed during April to August borne vibrational signals (Hunt 1994; Sattman & 2003–2005. We elicited female signals with playbacks Cocroft 2003). Mate-searching males produce adver- of recorded and simulated male signals (see below). tisement signals, and receptive females alternate To ensure that females were sexually receptive and their own signals with those of the male (Fig. 1), responsive, we tested virgin females 4–6 wk after eliciting localized searching by the male. Male sig- their adult molt, at the peak of their sexual receptiv- nals vary among species in the E. binotata complex, ity. We reared females from nymphs collected in the and females discriminate among males by selec- field, or from eggs laid by females collected in the tively responding on the basis of species and indi- field. Collecting sites were in Boone County, Mis- vidual signal differences (Rodrı´guez et al. 2004). souri, near the University of Missouri–Columbia Male signals closely match the preferred values of campus. We reared nymphs on their host plants in an outdoor facility at the University of Missouri– Columbia. We tested the E. binotata species that occur on Celastrus scandens, Cercis canadensis, Ptelea trifoliata, 1600 and Viburnum rufidulum plants. [We also used these 200 ms experiments to describe female signal preferences (Rodrı´guez et al. 2006).] Species in this complex await description, so we refer to them by the names Frequency (Hz) Frequency 200 of their host plants (e.g. E. binotata ‘Celastrus’). 1 0 Stimulus Design Relative –1 amplitude Whine pulses To describe female signal frequency and length, we elicited female responses with playbacks of recorded Male signal Female signal male signals having characteristics close to the mean values of each species in their Missouri populations Fig. 1: Spectrogram (top) and waveform (bottom) of a portion of a (using one recording for each species), obtained from duet of Enchenopa binotata ‘Celastrus’, showing one male signal and a library of signals (R.B. Cocroft, unpub- one female response. The male signal has two components: the whine E. binotata (a tone with harmonics that sweeps downward in frequency) and a lished data). Recordings were made using laser vibr- series of pulses. The female signal consists of a single, long compo- ometry (see below), by placing individual males on nent having a set of harmonically related frequencies the stem of a potted plant, within a few cm of a Ethology ª 2006 The Authors 2 Journal compilation ª 2006 Blackwell Verlag, Berlin R.L. Rodrı´guez & R.B. Cocroft Divergence in Female Duetting Signals in Enchenopa Treehoppers 6 signals have a constant frequency. Preliminary trials E. binotata showed these stimuli to be about as effective in 3 ‘Cercis’ eliciting female responses as playback of natural 0 signals (unpublished data). The whine or pulse components of male signals 6 may influence female response timing. The pulse E. binotata component probably does not have an influence: 3 ‘Celastrus’ stimuli with many pulses are often overlapped by 0 female responses, and stimuli with and without pulses result in similar or identical female response timings 6 (unpublished data). Thus, females trigger their E. binotata 3 ‘Ptelea’ responses by the whine component, and we focused Number of responses on whether the trigger is the beginning or end of the 0 whine. We compared response timing between stim- 6 uli with 400 and 800 ms – long whines (Fig. 1). These E. binotata values span the range of variation among the popula- 3 ‘Viburnum’ tions where our females were drawn. Other stimulus 0 features were set to the population means. We used stimuli with whine and pulses, varying only the 300 900 100 300 500 700 500 700 900 100 whine, because such stimuli elicit female responses 1100 1100 Latency (ms) Delay (ms) more effectively than stimuli having only the whine component (unpublished data). We measured the time from the beginning and end of the whine to the Latency beginning of the female response (latency and delay; Fig. 2, Table 1). If females trigger their responses by the beginning of male signals, latency should not be affected by whine length; if the trigger is the end of Delay male signals, delay should not be affected. Once we determined the feature of the whine that influences Fig.