The evolution of male courtship signals such as the bioluminescent flashes of fireflies may be shaped, at least in part, by female preference for particular characteristics of the male signal. These female preferences for male courtship signals may arise as a result of the benefits of choosing males with particular traits. One possible benefit of mate choice occurs if females can use male courtship signals as an honest indicator of male nutritional contributions at mating, nuptial gifts. This paper reviews female preference for male flash characteristics in Photinus fireflies (Coleoptera: Lampyridae), and the potential for females to use male flash characteristics to predict nuptial gift quality. In Photinus firefly species with single pulse flashes females preferentially respond to flashes of greater intensity and duration. Male Photinus provide a nuptial gift to females at mating in the form of a spermatophore and flash duration serves as a good predictor of spermatophore mass for males collected early in the season. However, Photinus fireflies do not feed as adults, so spermatophore mass decreases with subsequent matings. In response, nutrient-limited females may stop preferentially responding to longer duration flashes, increasing their overall responsiveness later in the mating season as they forage for spermatophores. Therefore, the evolution of male courtship signals in Photinus fireflies is the product not only of female preference for male flash characteristics, but also the costs and benefits of female choice that shape these preferences.
Male courtship signals take a wide variety of forms including the unique bioluminescent courtship flashes of fireflies. Charles Darwin (1871) was the first to propose that these elaborate male traits may evolve through female mate choice. Female mate choice and female preferences for male characteristics that generate these choices have been demonstrated in many species (reviewed by Andersson, 1994; Johnstone, 1995; Andersson and Iwasa, 1996). However, debate continues concerning how female preferences for elaborate male traits have evolved in these species.
Female preferences for male courtship signals may arise through a variety of evolutionary processes (Andersson, 1994). These processes differ in the potential benefit to the female of preferring a particular male phenotype. In a broad range of species, males provide a nutritional contribution at mating (reviewed by Boggs, 1995; Vahed, 1998). These nuptial gifts represent a potential benefit of mate choice if females prefer males whose courtship signals indicate they will provide more substantial nutritional contributions than competing males. The relationship between female preference, male courtship signals and nuptial gifts can influence sexual selection on both male traits and female preference behavior.
Courtship behavior in Photinus fireflies provides a unique opportunity to explore the interaction between female preference, male courtship signals and nuptial gifts. Male flash behavior can be simulated using light emitting diodes (LEDs), and female preference can be measured as the proportion of male flashes to which a female responds with bioluminescent flashes of her own. This paper reviews the flash courtship behavior of Photinus fireflies and discusses the potential for females to preferentially mate with males whose courtship signals indicate that they will provide a large nuptial gift.
Male Photinus fireflies fly and produce bioluminescent courtship flashes of single or multiple pulses repeated at regular time intervals to which female Photinus respond to from perches in the vegetation with bioluminescent flashes of their own after a species-specific time delay (Lloyd, 1966). After observing a female response a male will usually land, and then the male and female will engage in a dialog of male flashes and female responses until the male locates the female. Without these female responses males are unable to locate and mate with the female so differential female responsiveness as a function of male flash characteristics may provide a mechanism for mate choice (Lloyd, 1979; Lewis and Wang, 1991; Branham and Greenfield, 1996; Vencl and Carlson, 1998; Cratsley and Lewis, 2003). On the other hand increasing her overall responsiveness to male flashes represents one way in which a female might compete for males.
Female preference for male flash characteristics
A number of different male flash characteristics have been demonstrated to influence the likelihood of eliciting a female response in Photinus fireflies (Branham and Greenfield, 1996; Vencl and Carlson, 1998; Cratsley and Lewis, 2003). Branham and Greenfield (1996) found that Photinus consimilis females respond more frequently to trains of multiple simulated flash pulses in which the pulse rate is higher than the mean pulse rate observed in P. consimilis males. While this result is consistent with preferences observed in insects with multiple-pulse, acoustic courtship signals (Branham and Greenfield, 1996), most nearctic Photinus firefly species produce single-pulse flashes (Lloyd, 1966). Vencl and Carlson (1998) demonstrated female preference for greater flash intensity in the single-pulse flashing species Photinus pyralis. Lloyd (1966) also demonstrated that flash duration played an important role for females discriminating between conspecific and heterospecific flashes in a number of single-pulse flashing Photinus species.
In order to explore the importance of both flash intensity and flash duration in single-pulse flashing Photinus fireflies, Cratsley and Lewis (2003) tested Photinus ignitus female responsiveness to simulated flashes from LEDs varying in duration, intensity and distance from the female. When presented with computer-generated, simulated flashes representing the range of recorded P. ignitus male flash durations adjusted for temperature, female P. ignitus preferentially responded to the longest duration flashes (Fig. 1). Female P. pyralis show the same pattern of preference for longer duration flashes when presented with simulated flashes across the range of P. pyralis male flash duration (Cratsley, 2000).
Cratsley and Lewis (2003) also tested P. ignitus female responsiveness with simulated flashes from LEDs modified to represent fireflies with different size lanterns. Flash intensity observed by the females varied both with artificial lantern size and distance from lantern to female. Females responded preferentially to flashes from larger or closer lanterns, but did not distinguish between very close small lanterns and distant large lanterns (Fig. 2). While Vencl and Carlson (1998) manipulated flash intensity produced by the LED rather than lantern size or distance, their results are consistent with a general preference for greater intensity flashes in P. pyralis. Therefore, in at least two single-pulsing Photinus species, P. ignitus and P. pyralis, females respond preferentially to higher intensity and longer duration flashes.
Flash signals as indicators of male quality
Sensory bias in females for greater male flash intensity and duration could explain the pattern of female preference observed in P. ignitus and P. pyralis. However, females exhibit decreased flash responsiveness when presented with male flashes outside the range of conspecific male flash duration (Lloyd, 1966; Cratsley and Lewis, 2003). Female preference for greater flash intensity and duration could also have evolved if these flash characteristics serve as honest indicators of male quality. Evidence that females cannot distinguish between greater intensity as a function of lantern size versus distance suggests that intensity may be a relatively poor indicator of male quality (Cratsley and Lewis, 2003). Flash duration on the other hand could provide females with an opportunity to assess the quality of a male nuptial gift prior to mating.
Male Photinus fireflies provide a nuptial gift to females at mating in the form of a spermatophore that the female digests internally, using the nutrients to provision her eggs (van der Reijden et al., 1997; Rooney and Lewis, 1999; Lewis et al., 2004). The nutrients from male spermatophores are particularly important for female fecundity (Rooney and Lewis, 2002) because Photinus fireflies do not feed as adults (Williams, 1917; Lloyd, 1997). If male flash duration predicts spermatophore quality, females may prefer flashes of greater duration in order to obtain a large nuptial gift.
While flash intensity seems less likely to serve as an honest indicator of male quality because of the confounding effect of male distance, there is still the potential for flash intensity to covary with spermatophore mass. Male lantern size varies directly with body size in both P. pyralis (Vencl and Carlson, 1998) and P. ignitus (Cratsley, 2000), and male spermatophore mass and protein content also vary with body size in P. ignitus with heavier males producing larger spermatophores (Cratsley et al., 2003). However, the relationship between body mass and spermatophore mass is strongest for the first spermatophore collected from a male and decreases with the second spermatophore (Fig. 3). P. ignitus male spermatophore mass declines with successive matings (Cratsley et al., 2003), likely due to the limited nutrient reserves of these non-feeding adults. Females could potentially benefit from mating with larger males, but only early in the mating season when males have not mated multiply. There is some evidence for differential mating success as a function of body size in field populations of P. pyralis and P. ignitus (Vencl and Carlson, 1998; Cratsley, 2000), but it is unclear what role male flash characteristics play in determining these mating outcomes.
In order to determine the potential for females to use male flash duration and flash intensity as indicators of spermatophore mass, Cratsley and Lewis (2003) measured male flash duration, lantern width, lantern area, body mass and spermatophore mass in P. ignitus males collected early in the season when they were unlikely to have mated. In early season males flash duration was the best predictor of spermatophore mass, while neither lantern width nor lantern area contributed to explaining the variation in spermatophore mass (Table 1). Although more research needs to be conducted on the relationship between male morphology and flash intensity, these results suggest that female preference for flash duration in particular could have evolved to allow females to choose mates who will provide the largest spermatophore contributions.
While many aspects of the bioluminescent courtship behavior of fireflies have been studied in detail over the last 50 years, we are just beginning to understand the evolutionary significance of female preferences for male flash characteristics. These preferences may have shaped the evolution of flash characteristics such as pulse rate (Branham and Greenfield, 1996) intensity (Vencl and Carslon, 1998) and flash duration (Cratsley and Lewis, 2003) in Photinus fireflies. The preference functions themselves may be determined at least in part by interspecific differences in the potential for flash characteristics to predict nuptial gift quality and the relative benefits of choosing males who produce larger nuptial gifts. Even within species there is evidence that females may only exhibit preference for male flash characteristics early in the season when flash characteristics are good predictors of spermatophore mass (Cratsley and Lewis, 2003). Later in the season in a number of Photinus species the sex ratio becomes female-biased and females show overall increased responsiveness to male flashes (Lewis and Wang, 1991; Cratsley, 2000), suggesting that females may be foraging for mates and the accompanying spermatophores rather than choosing mates as a function of spermatophore size. Future research on female preferences for male flash characteristics and the potential benefits of those preferences should help to explain the wonderful diversity of firefly bioluminescent courtship behavior observed in the field.
From the Symposium Flash Communication: Fireflies at Fifty presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 4–8 January 2003, at Toronto, Canada.
I would like to thank Sara Lewis and Jennifer Rooney for discussions that shaped many of the ideas presented above, one anonymous reviewer for comments on the abstract, and NSF and SICB for supporting this symposium. This research was supported by Tufts University, Fitchburg State College and NSF (IBN 9816432).