- Top of page
In many insects, mating is affected by the day–night cycle, i.e., diurnal rhythm. Although there are many reports that mating and other reproductive behaviors are controlled by daily rhythms in various taxonomic insect species, little attention has been paid to the effect of daily rhythms on male fighting behavior. Here, we investigate whether the frequency and escalation of male–male aggressive interaction exhibit diurnal rhythms under a long-day condition in the bean bug Riptortus pedestris. Despite the fact that male aggressive behaviors were most often observed in the middle of the later half of light periods, no interaction was found between escalation of fighting and the time period. The results, at least, suggest that male aggressive behaviors are influenced by diurnal rhythms like other reproductive behaviors in R. pedestris.
Mating success depends on synchronization of mating activity between the sexes (Thornhill & Alcock 1983; Miyatake et al. 2002; Baužiene et al. 2004). In sexually reproducing organisms, mating plays the most important and fundamental role in producing progeny. In many insects, mating is associated with the day–night cycle evoked by the Earth's rotation about its own axis (e.g. Saunders 1982; Baužiene et al. 2004; Wang & Shi 2004; Rymer et al. 2007). Control of mating by daily rhythm reduces competition between species for a resource, synchronizes mating activity, and facilitates sympatric speciation of sibling species (Saunders 1982). These studies of daily rhythms serve not only to investigate insect mating behavior (Wang & Shi 2004) but also to improve pest control methods, such as sterile insect technique (Matsumoto et al. 2008; Fuchikawa et al. 2011).
Generally, reproductive activity is associated with various components such as courtship manners, courtship song, and sexual signaling for mating opportunity (Thornhill & Alcock 1983). These reproductive behaviors also exhibit daily rhythms like mating. For example, in the melon fly Bactrocera cucurbitae, pulse train intervals of male courtship songs correlate with the length of the circadian locomotor rhythm (Miyatake & Kanmiya 2004). In the Hawaiian cricket Laupala cerasina, mating behavior begins at the end of the light phase, and nuptial gift production is similarly affected by the daily rhythm (deCarvalho et al. 2012).
Like courtship behavior, the outcome of male–male contests is also one of the critical selection pressures affecting male reproductive success (Tachon et al. 1999; Berglund & Rosenqvist 2000; López et al. 2002; Double & Cockburn 2003; Rantala & Kortet 2004; Savage et al. 2005; Thomas & Simmons 2009). Some studies on vertebrate species also found that agonistic behavior occurred during particular time periods (fish, Radilova et al. 1991; mammals, Landau 1975; Russell & Singer 1983; Martensz et al. 1987; humans, Laubichler & Ruby 1986; Manfredini et al. 2001). To our knowledge, however, few studies have investigated the effects of daily rhythms on the male's aggressive behavior in insects or any other invertebrate (but see Dixon & Cade 1986; Farca Luna et al. 2009).
The bean bug, Riptortus pedestris (Fabricius) (Heteroptera: Alydidae), feeds on and damages leguminous crops (e.g. soybeans Glycine max, black-eyed peas Vigna unguiculata, and kidney beans Phaseolus vulgaris; Tomokuni et al. 1993). Riptortus pedestris males have exaggerated hindlegs used in ritualistic male–male competition for a bean plant as a mating territory (Natuhara 1985; Okada et al. 2011, 2012). Riptortus pedestris exhibits daily rhythms in oviposition (Kadosawa 1982; Numata & Matsui 1988), feeding (Kadosawa 1982, 1983), locomotor activity (Kadosawa 1982, 1983), and cuticle deposition (Ikeno et al. 2011). Likewise, in this species, mating shows diurnal variation, and mating most often takes place in the end of the light period (Kadosawa 1983). Thus, if male–male competition is affected by daily rhythm, males should aggressively engage in fighting and the fighting should most drastically escalate at the end of the light phase, when mating frequently occurs. In the present study, we investigated whether the frequency and escalation of male–male competition show diurnal rhythms under a long-day condition in R. pedestris.
A laboratory population was cultured from approximately 50 individuals collected from Fukuyama City, Hiroshima, Japan, in late autumn 2006 (Kimura et al. 2008). Insects were reared on soybean seeds, red clover (Trifolium pratense) seeds, and water containing ascorbic acid (0.05%) (Kamano 1991). Nymphs were reared in plastic cups (95 mm diameter, 40 mm depth) at a density of between 10 and 20 individuals. Each emerging adult was housed in a separate Petri dish (90 mm diameter, 15 mm depth) until the following experiment. The laboratory was maintained at a temperature of 25°C and 60% relative humidity with 16:8 h light:dark conditions (light period: 0700–2300). Bugs were marked individually on the thorax using paint markers (PC-5M, Mitsubishi Pencil Co. Ltd, Tokyo, Japan) 5 days prior to the experiments. We used adult bugs between 20 and 35 days old in this study. A plastic container (64 mm diameter, 25 mm depth) with a filter paper (55 mm diameter) was used as the combat arena. A soy bean was stuck in the center of the filter paper.
Two males, randomly picked from the stock culture, were simultaneously introduced into the combat arena, and their interaction was recorded on digital video camera (GZ-MG880, Victor Co. Ltd, Tokyo, Japan) for 2 h. At most four pairs were observed at once. Aggressive interactions were observed at six time periods: 0900–1100 (n = 21), 1300–1500 (n = 32), 1700–1900 (n = 55), 2100–2300 (n = 28), 0100–0300 (n = 24), and 0500–0700 (n = 24). In the dark period, observations took place under red light. The intensity of male aggression was scored on a scale of 1–6 that describes the fight: 1, one male kicked the other with one hind leg; 2, one male kicked the other with both hind legs; 3, both males kicked each other with one hind leg; 4, both males kicked each other with both hind legs; 5, both males kicked each other furiously with both hind legs and either or both males lifted their abdomens with their backs to the opponent and flapped their wings; and 6, one male raised his hind legs to grasp and squeeze the opponent's body or both males squeezed each other. After the fight, males were frozen at −20°C, and thorax width was measured as an index of body size (see Okada et al. 2011) using a dissecting microscope monitoring system (VM-60, Olympus, Tokyo, Japan).
To examine whether the frequency of aggressive interaction differed among the six time periods, we used logistic regression with the dependent variable (1, fight; 0, no fight) and time period as predictor variables. In addition, since the occurrence of fighting was affected by differences in body size of the two opponents, we added the difference of body size and interaction between time period and body size difference to the predictor variables. Ordered logistic regression was used to examine the effect of time period on the escalation of male aggressive behavior. In this analysis, replications where no aggressive interaction was observed were excluded from the data set. Male aggression scores of each replicate were assigned to the dependent variable as an ordinal variable. We applied time period, body size difference, and their interaction to predictor variables. In each analysis, non-significant interaction terms were removed from the full model (Grafen & Hails 2002). When a significant effect was detected in time period, we performed a post-hoc test and corrected the significance level using the step-up false discovery rate (FDR) (Benjamini & Hochberg 1995; Garcia 2004) to control for type I errors caused by conducting multiple tests. All analyses were performed using JMP version 9.0.2 for Windows (SAS Institute, Cary, NC, USA).
Logistic regression detected a significant effect of time period on the occurrence of fighting, but there was no significant effect of body size difference (logistic regression: time period, χ2 = 29.997, df = 5, P < 0.001; body size difference, χ2 = 0.104, df = 1, P = 0.746, Fig. 1A). Post-hoc logistic regression showed that the frequency of fighting was significantly higher during the time period 1700–1900 (Fig. 1A). However, there was no significant effect of time period and body size difference on escalation of aggressive behavior (logistic regression: time period: χ2 = 8.258, df = 5, P = 0.143; body size difference: χ2 = 0.504, df = 1, P = 0.477, Fig. 1B).
Figure 1. (a) Frequency of fighting and (b) mean score of fighting escalation in each time period. Treatments carried out in light and dark periods are indicated by white and black bars, respectively. The error bar shows the standard error. The same letters on each bar indicate that there is no significant difference on the basis of the logistic regression by the step-up false discovery rate (FDR) method (Benjamini & Hochberg 1995; Garcia 2004).
Download figure to PowerPoint
In R. pedestris, although escalation of male–male competition did not significantly differ among the six time periods, individuals engaged in fighting most frequently in the 1700–1900 period (Fig. 1A). Hence, it seems that male aggressive behavior exhibits a diurnal rhythm along with other behaviors such as locomotor activity, mating (Kadosawa 1982, 1983), and oviposition (Numata & Matsui 1988). The present results suggest that the aggressiveness of male R. pedestris is affected by diurnal rhythms. Kadosawa (1982, 1983) reported that although females of R. pedestris frequently fed and mated from late in the light period to early in next light period, mating often began during the last of the light period. Aggressive interactions between males were frequently observed in the middle of the last half of the light period, which was a little before mating time. This time period accords with the peak of a female's locomotor activity (Kadosawa 1983). Therefore, R. pedestris males can engage in a contest over territory while females search for host plants to feed on and then mate with a female that begins feeding in his territory.
Enquist and Leimar (1987) predicted that fighting escalates in proportion to the value of the resource being contested, and many empirical studies support this theoretical prediction (e.g., Simmons 1986; Tachon et al. 1999; Bridge et al. 2000; Okada & Miyatake 2004; Brown et al. 2007; Buena & Walker 2008; Small et al. 2009). For example, in the sap beetle Librodor japonicas, escalated fighting was observed only when a female was present in the fighting arena (Okada & Miyatake 2004). Males of the house cricket Acheta domesticus increase aggression in the presence of female pheromones (Buena & Walker 2008), and males of the orb-weaving spider Metellina mengei become more aggressive when they defend a larger and more fecund female (Bridge et al. 2000). In the stalk-eyed fly Teleopsis dalmanni, the greater the value of a resource, the more likely males are to engage in aggression and escalate aggressive interactions (Small et al. 2009). Consistent with these studies, our results revealed that R. pedestris males exhibited greater aggressiveness in the time period when female availability might be highest because females visit male territory to feed more frequently than in other time periods. In first half of the light period, females do not often feed and show less mating activity (Kadosawa 1982, 1983). Furthermore, most females mate from the last part of the light period to the first part of the next light period (Kadosawa 1983), so in these time periods, female availability is relatively low, and thus, males show little motivation to fight. Hence, the daily rhythm of male aggressive behavior may be brought about by the diurnal change of female availability.
Again, body size difference did not affect escalation of male fighting in the present study (Fig. 1B). Generally, a resource holding potential such as body size also has an effect on the occurrence and escalation of a male's aggressiveness (Parker 1974; Maynard Smith & Parker 1976). Many theoretical models suggest that when the body size difference of contestants is large, the fight outcome is easily predicted, and therefore settled sooner (e.g. Parker 1974; Maynard Smith & Parker 1976; Parker & Rubenstein 1981; Maynard Smith 1982). In R. pedestris, however, it is interesting to note that male body size difference affects neither the frequency nor escalation of fighting. Another hypothesis is needed to reveal the mechanism of fighting escalation in this species.