Metaleptea 5 SCIENCE REPORT Sex can be dangerous: me the unique opportunity to study the evolu- but rather that flies were more likely to locate Acoustically-orienting parasitoids on field tion of an acoustic mating display by compar- a male with a greater proportion of long chirp crickets (Orthoptera: Gryllidae) ing populations of the same species. Zuk and in his songs (Zuk et al. 1998). her colleagues have described T. oceanicus Gita Raman Kolluru populations varying in parasitoid prevalence If flies prefer the same song structure vari- from 0% to 31% (Zuk et al. 1993; Rotenberry ables as female crickets, then male crickets Department of Biology et al. 1996). These populations have a corre- may face a compromise between attracting fe- Univcrsity of Calilo~nia sponding variation in calling song structure, males for mating and also attracting flies (e.g., Riverside, CA 92521 suggesting that selection by the parasitoid has Wagner 1996). In this case, cricket song may played a role in song evolution (Zuk et al. either not change much over evolutionary Telephone: (909) 787-3952 FAX: (909) 787- 1993; Rotenberry et al. 1996). time because of stabilizing natural and sexual 4286 email: [email protected] selection pressures, or female cricket choice The long term goal of my research is to de- may be relaxed in the parasitized populations The Orthopterists' Society generously termine how natural selection imposed by the such that directional selection by flies is the awarded me grants in 1995 and 1997 to con- parasitoid fly and sexual selection imposed by predominant force affecting song evolution. duct research on Teleogryllus oceanicus (Or- female crickets interact to shape the evolution Alternatively, if flies differ from female crick- thoptera: Gryllidae) on the Big Island of Ha- of T. oceanicus reproductive biology. My re- ets in their song preferences, then cricket song waii. Here I report results to date from search focuses on the following questions: is expected to evolve away from what flies fieldwork conducted in the past few years. prefer and toward what female crickets prefer. 1) Does O. ochracea influence male cricket Direct tests of fly preference are required to Teleogryllus oceanicus is native to the Pact- reproductive success even before the cricket distinguish between these alternative predic- fic Islands and Australia and has been intro- dies? tions. duced into Hawaii (Kevan 1990; Otte and Al- exander 1983). Like other field crickets, males Ormiine parasitoids sign)ficantly reduce In the summer of 1998 I conducted song produce a conspicuous calling song to attract host lifespan (e.g., Lehmann and Heller 1997), broadcasts to test for song preferences of para- females for mating (Fig. 1). However, in some and my work on T. oceanicus indicates that sitoid flies in Hilo, Hawaii. I broadcasted parts of the cricket's range a singing male parasitization may also have detrimental fit- pairs of songs differing in one variable only, risks also attracting the acoustically orienting ness consequences prior to host death. T. and collected flies attracted to each song. A parasitoid fly Ormia ochracea (Diptera: Ta- oceanicus males transfer sperm to females in total of 342 flies were collected over 27 chinidae: Ormiini), which parasitizes Gryllus discrete spermatophores. In two separate ex- nights. I found that flies sign)ficantly pre- species on the mainland USA (Cede 1975; periments of spermatophore replacement rates ferred songs with many long chirp pulses and Walker 1986) and T. oceanicus in Hawaii in parasitized and unparasitized males con- few short chirps. In the future I intend to di- (Zuk et al. 1993). Ormiines are unique be- ducted in 1997, parasitized males produced rectly test for fly preferences in other parasi- cause they have specialized ears that enable sign)ficantly fewer spermatophores than un- tized populations, to evaluate the generality of them to locate their hosts by their songs (Rob- parasitized males. An examination of mating the Hilo results and to determine whether fly ert et al. 1992; Allen 1995). A gravidO. och- behavior in 1998 supported the spermatophore preferences vary among populations within racea female locates a cricket and larviposits replacement studies and showed that parasi- the Hawaiian islands. The expectation is that on and around it; larvae burrow into the host tized males copulated sign)ficantly less fre- fly preferences will be strongest in Kauai, and develop within the host for 7- 10 days be- quently than unparasitized males. These re- which has the highest parasitoid prevalence, fore emerging to pupate, killing the host with- sults suggest that parasitization reduces male and weakest in Oahu, which has the lowest in one day of emergence (Adamo et al. cricket reproductive success, and confirm that parasitoid prevalence (Rotenberry et al. 1996). 1995a). Although larvae deposited near a the fly is an important selective agent for male may parasitize female crickets attracted crickets. This research is ongoing and I plan Neuroethological studies suggest that fe- to that male, females have a relatively low to address whether the fecundity of females male crickets generally also prefer the long parasitoid prevalence compared to males (Zuk mated to parasitized males is lower than that chirp (Pollack and Hoy 1981), so that female et al. 1993, Adamo et al. 1995b). of females mated to unparasitized males. crickets and flies may converge in their song preferences (e.g., Wagner 1996). However, if Figure 1. Sonogram of a typical T. oceanicus 2) Do female flies and female crickets prefer a female approaches a male whose song has calling song. the same aspects of male song? attracted flies, she risks also becoming parasi- tized. Therefore, it is possible that females are Because the same calling song produced by Acoustically-orienting parasitoids are general- less choosy in populations with high parasi- male crickets to attract females is used by flies ly more likely to find males with high pulse toid prevalence. For example, female crickets to locate hosts (Cede 1975), Hawaii males numbers (e.g., long chirp, Fig. 1) in their respond differently to male songs depending face a trade-off between producing elaborate songs (Wagner 1996; Lehmann and Heller on the perceived risk of predation (Hedrick songs to attract females and minimizing sing- 1998). Zuk et al. (1998) confirmed this by and Dill 1993; Csada and Neudorf 1995). I ing to avoid attracting the fly. Previous studies showing that parasitized T. oceanicus males will therefore also conduct laboratory experi- of the effects of parasitization on cricket song within a population had more long chirp and ments to directly test the song preferences of have been confined to comparisons of species less short chirp components in their songs females from populations varying in parasi- that differ in parasitization; such interspecific than unparasitized males. These differences toid prevalence, to test the hypothesis that fe- comparisons may have confounding effects among individuals were apparent even imme- males from heavily parasitized populations that have nothing to do with the parasitoid diately after infestation, suggesting that differ- are less choosy. (Rotenberry et al. 1996). T. oceanicus offers ences in calling were not due to parasitization, 6 Metaleptea SCIENCE REPORT 3) Does cricket calling activity correlate In 1997 I examined the individual calling pat- predation risk on mate choice in female Ache- positively with female cricket attraction and terns of 39 male crickets in Hilo and found a ta domesticus crickets. Ecol. Entomol. 20: negatively with parasitoid prevalence? pronounced dusk peak but no dawn peak in 393-395. calling activity. Data from fly captures Calling in T. oceanicus and other field showed that the calling activity peak coincides Doolan, JM and GS Pollack. 1985. Phonotac- crickets is energetically expensive (e.g., Ho- with the peak in fly searching activity, so that tic specificity of the cricket Teleogryllus back and Wagner 1997). Therefore, it is possi- males do not appear to have shifted calling to oceanicus: intensity-dependent selectivity for ble that females prefer males who are able to avoid flies. I also intend to examine calling temporal parameters of the stimulus. J. Comp. sustain high levels of calling during a night patterns in Oahu and Kauai, to determine Physiol. A. 157: 223-233. (Walker 1983; Crnokrak and Roff 1998), or whether a shift in the timing of calling has oc- that males that call a lot are simply more like- curred in those populations. I am especially French, BW and WH Cade. 1987. The timing ly to be calling when a female passes (Zuk interested in the heavily infested Kauai popu- of calling, movement, and mating in the field and Simmons 1997). The female may then lation, where selection by the fly has poten- crickets Gryllus veletis, G. pennsylvanicus, evaluate male quality based on song structure tially resulted in crickets calling more at dawn and G. integer. Behav. Ecol. Sociobiol. 21: elements such as pulse rate, frequency, and in- than at dusk. 157-162. tensity (Pollack and Hoy 1981; Doolan and Pollack 1985). In either case, because calling This research will increase our understand- Hedrick, AV and LM Dill. 1993. Mate choice also attracts flies, calling activity is expected ing of how cricket behavior evolves and, more by female crickets is influenced by predation to be negatively correlated with parasitoid generally, of how conflicting selection pres- risk. Anim. Behav. 46: 193- 196. prevalence (Cede l 991). sures interact to produce phenotypes over evo- lutionary time (e.g., Andersson 1994: 234). Hoback, WW and WE Wagner, Jr. 1997. The In the summer of 1997 I determined the energetic cost of calling in the variable field calling activity of 39 caged males for two to REFERENCES CITED cricket, Gryllus lineaticeps. Physiol. Entomol. eight consecutive nights in Hilo, Hawaii. Field 22: 286-290.