Sexual coercion in Panorpa scorpionflies?—The function of the notal organ reconsidered

Abstract

Conflict between the sexes over the occurrence and timing of copulation is a common feature of sexual reproduction. In Panorpa scorpionflies (Mecoptera, Panorpidae), sexual conflict occurs over mating duration, and it has been suggested that the notal organ—a clamp-like structure on the males’ abdomen—is an adaptation to coerce mating duration, enabling males to prolong mating against female interests. However, as recent studies suggest that female scorpionflies have considerable control over mating decisions, we question this hypothesis and reexamined the function of the notal organ with respect to prolonged mating. We manipulated the notal organs of male Panorpa vulgaris and compared copulation durations achieved by males with functional and nonfunctional notal organs. There were no effects on the overall copulation duration or the period of time a copulation continues after the last nuptial gift has been delivered. We therefore reject the hypothesis that the notal organ is a male adaptation to extend copulation duration against female interests.

Having dismissed the traditional view that reproduction relies on cooperation and harmony between the two sexes, it is now the general opinion that sexual conflict, that is, the conflict between evolutionary interests of individuals of the two sexes (Parker 1979), is a common feature of sexual reproduction. It occurs when a trait that enhances the reproductive success of one sex reduces the fitness of the other sex (Parker 1979). Male and female interests often diverge with respect to mating decisions such as the timing and duration of copulation, as well as to the number of mating partners, parental investment, and/or fertilization decisions. Sexual conflict can therefore constitute a potent force in male–female coevolution (e.g., Stockley 1997; Holland and Rice 1998; Chapman et al. 2003; Arnqvist and Rowe 2005).

The rate at which mating occurs has an especially high potential for sexual conflict, as the optimum will be different for males and females in most species (e.g., Bateman 1948; Parker et al. 1972; Trivers 1972; Holland and Rice 1998). Thus, studies on the implications of sexual conflict have largely focused on conflicts over the occurrence and timing of mating (e.g., Arnqvist and Rowe 1995; Sakaluk et al. 2006; Blanckenhorn et al. 2007; for reviews see Chapman et al. 2003; Parker 2006; Wedell et al. 2006; Gwynne 2008). As Gwynne (2008) points out, sexual conflict over mating and fertilization decisions will also play a major role in the evolution and maintenance of nuptial feeding in insects.

In the gift-giving scorpionfly Panorpa vulgaris (Imhoff and Labram) sexual conflict occurs over the length of copulation. Just as in other Panorpa species, males of P. vulgaris provide females with nutritious saliva secretions during copulation (Sindern 1996; Sauer et al. 1997, 1998; Sauer 2002). Because a male's ability to produce nuptial gifts strongly depends on its body condition, females can rely on this trait as an honest signal for male quality (Fleck and Sauer 1995; Sauer et al. 1998; Kurtz and Sauer 1999; Engels and Sauer 2006). Females adjust copula duration to the number of salivary masses they receive and, because sperm is transferred continuously throughout mating, males that provide numerous nuptial gifts will also transfer large amounts of sperm (Sindern 1996; Sauer et al. 1997, 1998; Sauer 2002). In multiply mated females, paternity allocation follows the “fair raffle principle” (sensuParker 1990), that is, paternity correlates with the relative numerical representation of sperm a male contributes (Sauer et al. 1999). Males that achieve long copulations by offering numerous salivary masses will therefore also fertilize an accordingly large proportion of the female's eggs (Thornhill and Sauer 1991; Sauer et al. 1999). Consequently, it is in the female's interest to terminate copulation once the male ceases to produce any further nuptial gifts. Males, on the other hand, should always aim at extending mating duration in order to increase paternity.

When in copula male scorpionflies secure the anterior edge of one of the female's forewings by means of the so-called “notal organ,” a clamp-like structure is formed from parts of the dorsum of the male's third and fourth abdominal segments (Mickoleit 1971). This notal organ has been suggested to be an adaptation to coerce mating duration, enabling males to prolong mating duration against female interests (Thornhill 1980; Thornhill and Sauer 1991).

In addition, scorpionfly mating systems are often referred to as example for a “resource defence polygyny” (sensuEmlen and Oring 1977), claiming that male scorpionflies monopolize females by monopolizing food resources critical to female reproduction and are thus in control over the occurrence of mating (Thornhill 1979, 1981; cited, e.g., in Andersson 1994; Krebs and Davies 1996). This view, together with the postulate that the notal organ is an adaptation to coerce mating duration, leads to the conclusion that in scorpionflies it is largely the males that are in control over mating decisions.

Over the last few years, however, a lot of evidence has come up suggesting that female Panorpa scorpionflies are much more in control over mating than previously thought. A recent study showed that the mating systems of at least two Panorpa species (P. vulgaris and Panorpa alpina) evidently cannot be characterized as resource defence polygynous (Missoweit and Sauer 2007). Further studies on P. vulgaris (e.g., Sauer et al. 1998; Kurtz and Sauer 1999; Sauer 2002; Engels and Sauer 2006) as well as studies on Panorpa cognata (Engqvist and Sauer 2003; Engqvist 2007a, 2007b; Engqvist et al. 2007) and Panorpa germanica (Gerhards 1999; Kock et al. 2006; Kock and Sauer, 2008) point to mating systems that are predominantly based on female choice of males signaling their quality via the ability to produce nuptial gifts. For P. germanica, there is evidence that, in addition to mate choice based on saliva secretion, females choose for male quality already precopulatory based on male pheromones and are hence the ones to decide if mating takes place (Kock et al. 2007; Kock and Sauer 2008). P. vulgaris females are able to counteract sperm transfer, most likely by means of the so-called “compression muscle” located at the spermatheca (Vermeulen 2004; Vermeulen et al. 2008). From these studies, together with the observation that copulations are always terminated by females kicking their hind legs (personal observation), we conclude that at least in P. vulgaris, P. germanica, and P. cognata, it is rather the females than the males who are in control over mating and fertilizing decisions.

In the light of these new insights, we question the findings by Thornhill and Sauer (1991) that males are able to force females into longer copulations than it is in their interests and therefore reexamined the function of the notal organ in P. vulgaris. In this experiment, we manipulated the notal organs of males to determine whether a functional notal organ led to an increase in copulation duration. Due to the reasons outlined above, we do not expect to find any effect of the notal organ on copulation duration.

MATERIALS AND METHODS

Breeding protocol

Individuals of P. vulgaris used in this study were reared from adults collected in the field near Freiburg i. Br., Germany. F1-offspring obtained from adults collected in spring 2006 represented our first experimental generation and F1-offspring of adults collected in summer 2006 a second generation. Field collected adults were held in pairs in plastic boxes (10 × 10 × 7 cm) containing moist tissue paper, ad libitum food, and a small Petri dish filled with moist peat for oviposition. F1-larvae were reared at an 18:6 h light:dark cycle at 18 °C on moist tissue paper with ad libitum food at a maximum density of 20 larvae per Petri dish (12 cm diameter). Third-instar larvae were transferred into soil-filled, open-bottomed plastic cylinders (40 cm diameter) placed outdoors in the ground, where they finished their development. After hatching, imagoes were kept separately in plastic cylinders (diameter 3.5 cm, height 8 cm) containing moist tissue paper as water supply. Throughout their lives all animals were kept on an 18:6 h light:dark cycle in climate chambers at 18 °C. As Panorpa scorpionflies are opportunistic scavengers that mainly feed on dead arthropods (Sauer 1970; Kaltenbach 1978; Byers and Thornhill 1983), imagoes were regularly provided with segments of last-instar mealworms T. molitor (see below).

Experimental setup

In order to reexamine the hypotheses that the notal organ in P. vulgaris is an adaptation to prolong mating against female interests, we performed a series of experimental tests, each involving the manipulation of the notal organ. In the test groups, we made the notal organs nonfunctional by covering them with a layer of the adhesive Protemp II (3 M ESPE Protemp II). For control groups, we pseudomanipulated the males by applying Protemp II lateral to the notal organ, leaving it operative as a clamp.

As Thornhill and Sauer (1991) found the effect of the notal organ to be dependent on body condition and the males’ ability to produce salivary masses, we assigned our test individuals arbitrarily to two different feeding regimes: in the first generation, half of the males as well as half of the females received one segment of a last larval instar of T. molitor every second day, whereas the other half was provided with one segment only every fourth day. However, as we found that animals receiving food only every fourth day still remained in rather good condition, we adjusted the feeding regime in the second generation: Half the animals received one segment of a last larval instar of T. molitor every 3 days, the other half only once a week.

Matings were arranged by placing 1 male and 1 female together in a plastic box (10 cm × 10 cm × 6 cm). We recorded when copulations began, when salivary masses were transferred, when copulation ended, and the duration of the time interval from delivery of the last salivary mass to the end of copulation (=last Δt). In the first experiment, a total number of 218 copulations took place, of which 124 involved males with nonfunctional notal organs. The 203 copulations of the second generation included 99 performed by males with nonfunctional notal organs.

Statistical analyses

All statistical analyses were carried out using the SPSS 12.0 software (SPSS Inc., Chicago, IL). Quoted significance values are for two-tailed tests, and the level of significance was set to P ≤ 0.05.

RESULTS

Effect of the notal organ on copulation duration and the last Δt

To test if an operative notal organ extends the length of copulation, we compared mean copulation durations achieved by males with functional notal organs versus males with nonfunctional notal organs. Because of significant deviations from normal distribution, data were transformed (square root transformation) before being subjected to parametric tests. There were no significant differences in mean copulation durations in any of the two test generations (Table 1).

Table 1

Mean copulation duration and mean last Δt achieved by males with functional notal organs versus males with nonfunctional notal organs

	State of notal organ	N	Copulation duration (mean ± SD) (min)		t-tests (copulation duration)	Last Δt (mean ± SD) (min)		t-tests (last Δt)
Generation	Functional	94	194.24	+144.64	t = 1.730	66.64	+66.33	t = 1.147
				−104.65			−43.64
1	Nonfunctional	124	162.74	+165.04	P = 0.085	58.51	+59.33	P = 0.253
				−107.85			−38.77
Generation	Functional	104	155.58	+176.04	t = 0.299	66.72	+70.77	t = −1.101
				−110.20			−45.47
2	Nonfunctional	99	150.05	+143.29	P = 0.765	75.72	+69.34	P = 0.272
				−95.71			−40.00

As outlined in the Introduction, male P. vulgaris provide females with nutritious salivary masses during copulation (e.g., Sauer et al. 1998), and it is in the female's interest to terminate copulation once the male ceases to offer further gifts. The male, however, will aim at extending mating duration. Consequently, there is a strong conflict between the two sexes over the period of time the copulation will continue after the last nuptial gift has been transferred. This period from the last salivary mass to the end of copulation is called the last Δt. To test if the notal organ could be used to extend this last Δt, rather than to extend the overall mating duration, we compared the length of the last Δt achieved by males with functional notal organs versus males with nonfunctional notal organs. There were no significant differences in any of the two test generations (Table 1).

Effects of male nutritional status and salivary secretion on copulation duration

For P. vulgaris, it has repeatedly been shown that the length of copulation is positively related to the number of salivary masses a male provides and a male's ability to produce nuptial gifts is dependent on body condition (e.g., Sindern 1996; Sauer et al. 1997, 1998; Engels and Sauer 2006, 2007, 2008).

In the study by Thornhill and Sauer (1991), test animals were either fed or starved before copulation and an extended length of copulation was found in all their tests, except one test in which starved females were mated with well-fed males who produced numerous salivary masses. In order to account for the effect of the nutritional status of males and their ability to produce salivary masses, we performed Analysis of Covariance (ANCOVA) analyses taking copulation duration as the dependent variable, nutritional status as well as the state of the notal organ as fixed factors and the number of salivary masses as a covariate. In the first test generation, only the number of salivary masses had a significant effect on copulation duration (number of salivary masses: F_1,213 = 280.58, P < 0.001; male nutritional status: F_1,213 = 0.65, P = 0.421; state of notal organ: F_1,213 = 2.98, P = 0.086; nutritional status*notal organ: F_1,213 = 0.25, P = 0.617). The results for the second generation were slightly different. Again, the number of salivary masses had a significant effect on copulation duration (F_1,198 = 343.97, P < 0.001). Neither the nutritional status of males (F_1,198 = 0.24, P = 0.623) nor the state of the notal organ (F_1,198 = 0.23, P = 0.632) had a significant effect, but the interaction of both did (F_1,198 = 5.26, P = 0.023). Looking at males of high and low nutritional status separately, we found that within the well-nourished males, those with operative notal organs achieved an average copulation duration of 161.10 min (standard deviation, SD = +181.31, −113.75 min, N = 51), whereas those with nonfunctional notal organs had an average copulation duration of 140.05 min (SD = +141.06, −92.41 min, N = 49; t-test: t_{well-nourished} = 0.80, P = 0.429). Within the poorly nourished males, those with functional notal organs had an average copulation duration of 150.36 min (SD = +172.72, −107.46 min, N = 53), whereas those with nonfunctional notal organs achieved a mean copulation duration of 160.17 min (SD = +145.90, −99.07 min, N = 50; t-test: t_{poorly nourished} = −0.38, P = 0.707). Thus, well-nourished males achieved longer copulations when the notal organ was operative, which was not the case for poorly nourished males. Yet, these differences are not significant.

ANCOVA analyses were also used to compare the length of the last Δt achieved by males with functional versus males with nonfunctional notal organs, taking effects of male nutritional status and salivary production into account. In both generations, only the number of salivary masses had a significant effect on the last Δt (Table 2).

Table 2

Effect of saliva production, male nutritional status, and state of notal organ on the last Δt

	Dependent variable: last Δt (min)	ANCOVA
Generation 1	Number of salivary masses	F1,213 = 7.08	P = 0.008**
	Male nutritional status	F1,213 = 1.38	P = 0.242
	State of notal organ	F1,213 = 1.03	P = 0.312
	Interaction nutritional status*notal organ	F1,213 = 2.82	P = 0.095
Generation 2	Number of salivary masses	F1,198 = 21.85	P < 0.001***
	Male nutritional status	F1,198 = 0.19	P = 0.667
	State of notal organ	F1,198 = 1.92	P = 0.168
	Interaction nutritional status*notal organ	F1,198 = 3.30	P = 0.071

**

P < 0.01

***

P < 0.001.

DISCUSSION

Sexual conflict over mating decisions is a common phenomenon reported for many different species. This conflict often concerns the frequency as well as the duration of copulation. Along with this, males of some insect species have evolved morphological traits used in sexual coercion. Male water striders possess an abdominal grasping structure that they use to initiate copulation as well as to influence the duration of copulation (e.g., Arnqvist 1989; Arnqvist and Rowe 1995, 2002; Ortigosa and Rowe 2003). Similar traits that are also used in the context of sexual coercion are the clasping structures on the front legs of male dung flies (Allen and Simmons 1996) and the gin trap of male sagebrush crickets (Sakaluk et al. 1995). The notal organ of male Panorpa scorpionflies belongs to this same class of morphological structures that could potentially be used in coercive mating behavior. Thornhill and Sauer (1991) suggested that the notal organ of male scorpionflies is an adaptation to prolong mating against female interests. In contrast to the results obtained by Thornhill and Sauer (1991), in the present study, there were no differences in the mean copulation duration in any of our two test generations. In both experiments, copulation duration was strongly dependent on the number of salivary masses a male produced, but there were no effects of the notal organ or male nutritional status. In the second generation, however, we found a significant effect of the interaction of male nutritional status and the state of the notal organ. Nonsignificantly, well-nourished males apparently achieved longer copulations when the notal organ was operative, whereas it was the other way round for the poorly nourished males. Although these results appear somewhat obscure, they still do not support the hypothesis that males can extend the copulation duration against female interests: Males that are in poor body condition cannot produce large amounts of saliva and should therefore be the ones that “have to force” females into copulations. Yet, in our experiment, the poorly nourished males actually had shorter copulations when the notal organ was operative, although the difference was not significant.

In addition to the overall length of copulation, we also took a closer look at the time interval copulations continued after the delivery of the last nuptial gift (last Δt). During this phase of copulation, sexual conflict should be most intense, as it is in the female's interest to terminate copulation once the male ceases to produce any further nuptial gifts, whereas males should always aim at extending mating duration in order to increase paternity. Comparing the length of the time interval copulations continued after the delivery of the last nuptial gift of males with functional versus males with nonfunctional notal organs we found no significant differences in any of the two test generations. ANCOVA analyses revealed that in both generations only the number of salivary masses had a significant effect on how long copulation continued after the last gift had been delivered. Apparently, females do not only adjust the overall copulation duration to the male's quality, but also the time interval a male is allowed to copulate per salivary mass delivered. In neither of the two generations we could find effects of the notal organ or male nutritional status on the time interval after the delivery of the last nuptial gift.

Overall, our results do not support the hypothesis that the notal organ of male scorpionflies is an adaptation to coerce mating duration. Other observations also contradict its evolution as a means to force females into copulation. Furthermore, the notal organ is used in all copulations irrespective of whether the male provides any nuptial gift (Byers and Thornhill 1983; Thornhill and Sauer 1991). The question arises as to what function the notal organ might have, if it has not evolved as a male adaptation to sexual conflict over mating duration.

One possible function could be to influence female mate choice. During struggles, females could measure male body condition and through this carry out a form of mate assessment influencing their mating decisions. However, this explanation seems to be very unlikely, because salivary masses appear to be a more useful measure to a female to assess a male's condition. Females adjust copulation duration to the number of salivary masses a male provides and a male's ability to produce nuptial gifts is dependent on body condition (e.g., Sindern 1996; Sauer et al. 1997, 1998; Engels and Sauer 2006, 2007, 2008). Moreover, if there are any struggles before or during a mating, the female will always be able to avoid/terminate mating. Therefore, the notal organ must have evolved in a different context.

When in copula Panorpa scorpionflies adopt a characteristic V-shaped mating position. The male's wings are positioned on top of the female's wings and the anterior edge of one of the female's forewings is held in the clamp-like notal organ throughout mating. To initiate copulation, the male comes up to the female's side and lashes out his abdomen at hers, meanwhile aiming to grab one of the female's forewings with the notal organ. Possibly, the notal organ's function is to ease initiation of copulation and it might further have its role in stabilizing the characteristic position throughout mating. Being securely attached to each other during mating could help to resist disturbances and/or the transfer of salivary masses might be facilitated.

Although it remains to be tested if the notal organ's function is to initiate and stabilize copulation, we can conclude from this study, that males cannot force females to stay in copulation. Thus, we reject the hypothesis that the notal organ of male scorpionflies is an adaptation to coerce mating duration, advancing the view that female Panorpa scorpionflies have considerable control over mating decisions.

FUNDING

This project was financially supported by the Deutsche Forschungsgemeinschaft [SA 259/7–1 to 4].

We thank Julia Leven for help in breeding P. vulgaris in the laboratory. We would like to thank Leif Engqvist for statistical advice and 2 anonymous referees for their helpful comments on earlier versions of this manuscript.