Silja Parri

Viability Costs of Condition-Dependent Sexual Male Display in a Drumming Wolf Spider

According to the conditional handicap models females use male ornaments as honest signals of male viability. The assumptions for honest signalling are that the traits are costly and that they reflect male phenotypic condition, and hence optimal trait size is largest in the most viable males. However, experimental evidence for the costs of signalling are scarce. In this study we experimentally tested whether acoustic signalling, drumming, in a wolf spider Hygrolycosa rubrofasciata is a condition dependent, costly trait, and thus offers an honest signal of quality to females. Males of this species court females by drumming dry leaves with their abdomen. Females prefer to mate with males of high drumming rate, but body mass of males does not affect female choice. We manipulated phenotypic condition of males by keeping them in high, intermediate and low food levels. Males in a high food level treatment maintained their drumming rate at a high level, while males with intermediate and low food levels exhibited a reduction in drumming rates. Thus, phenotypic condition of the males affects their sexual signalling. We induced another set of males to increase their drumming activity by presenting females in proximity. These males suffered higher mortality and lost significantly more weight than other males, confirming that drumming is costly. However, within the increased treatment group males that drummed most actively survived better than less active males. Thus, males vary in their ability to bear the costs of drumming, which suggests that drumming is an honest signal of male quality (= conditional handicap) for females.

ENERGETIC COSTS OF SIZE AND SEXUAL SIGNALLING IN A WOLF SPIDER

A prerequisite for honest handicaps is that there are significant condition–dependent costs in the expression of sexual traits. In the wolf spider Hygrolycosa rubrofasciata (Ohlert), sexual signalling (drumming) is costly in terms of increased mortality. Here we investigated whether this mortality may be caused by increased energy expenditure. During sexual signalling, metabolic rate was 22 times higher than at rest and four times higher than when males were actively moving. Metabolic rate per unit mass was positively related to absolute body mass during sexual signalling but not during other activities. This positive relationship is novel to any studies of metabolic rates. Indeed, it seems that the largest males can drum only 12 times per minute before reaching the maximum sustainable metabolic rate, whereas the smallest males may drum up to 39 times per minute. However, there is no relationship between body mass and drumming rate, indicating that larger males are able to compensate for the higher cost of drumming. There was a quadratic relationship between relative abdomen mass and overall body mass, which may provide a partial explanation for the increased energy expenditure of largest males while drumming. Altogether, our results indicate that sexual signalling is highly energetically demanding, which may be the main reason for the honesty of signalling in this species. In addition, the energetic costs are surprisingly strongly size dependent, which may compensate any disadvantage of small male size.

SEXUAL SELECTION IN A WOLF SPIDER: MALE DRUMMING ACTIVITY, BODY SIZE, AND VIABILITY

Females are often believed to actively choose highly ornamented males (males with extravagant morphological signals or intense sexual display), and ornaments should be honest signals of male viability. However, this belief is relying only on some pieces of empirical evidence from birds. Our study reports active female choice on sexual display that indicates male viability in spiders. We established trials in which we studied female choice in relation to male courtship drumming activity and body size. Females chose the most actively drumming males as mating partners, but the body size of the males did not seem to be selected. Male drumming activity turned out to be a good predictor of male viability, whereas male viability was independent of male body mass. Our results suggest that by actively choosing mates according to male drumming performance, but independently of male body mass, females are preferring viable males as mates. Because Hygrolycosa rubrofasciata males do not provide obvious direct benefits to their offspring, females may gain some indirect benefits; offspring may have higher chance of survival, or the offspring may inherit the attractiveness of their father.

Mate choice for offspring performance: major benefits or minor costs?

There is considerable disagreement over whether or not gaining viability benefits to offspring could be substantial enough to overcome the costs of female choosiness. A recent review suggested that the ‘lek paradox’ might be resolved by large indirect benefits as indicated by highly heritable ornamental traits. We selected males of a wolf spider Hygrolycosa rubrofasciata in relation to their sexual signalling rate (audible drumming). The estimated correlated response in offspring viability was rather small (0.12 s.d.). However, it may be large enough if the costs of being choosy are small. In fact, females mate with better–than–average males just by responding passively to a random drumming signal, and the active choice by females seemed to increase this benefit only slightly. In many mating systems, females obtain better–than–average males as a consequence of intense male–male competition or because of the extraordinary variance in male signalling. The costs of any additional choice may be so minor that female choice for honestly signalling males may evolve even with minute benefits in offspring viability. This may be the general solution to the lek paradox, as most studies report no apparent fitness benefits. Publication bias favouring statistically significant results may have led to an overemphasis on the few studies with large effects.

Female choice for male drumming in the wolf spiderHygrolycosa rubrofasciata

Mate preferences in invertebrates have usually been studied with simultaneous choice exper- iments alone, which allows eVective detection of any preferences but does not tell much about the strength of inter-sexual selection. Under natural conditions females frequently have to rely on sequential choice, and choosy females may incur opportunity and direct costs such as loss of time when they reject a male. Female preference in the wolf spider Hygrolycosa rubrofasciata for two components of male courtship signalling, rate and volume, was investigated. Both of these characteristics were tested with a sequential choice set-up and the eVect of volume also with a simultaneous choice method. Females responded more quickly to male signals with a higher rate and volume. This suggests that females use a threshold level when responding to male courtship signals and that they are prepared to suVer some costs of waiting for an opportunity to choose between males.

Acoustic signalling in a wolf spider: can signal characteristics predict male quality?

While there has been considerable interest in female choice for male sexual signals, there have been few studies of the underlying information that different aspects of the signal calls convey. Such studies, however, are essential to understand the significance of signals as honest handicaps, arbitrary Fisherian traits and/or in species recognition. We studied the somewhat exceptional system of audible drumming in the wolf spider Hygrolycosa rubrofasciata. We estimated the repeatabilities of signal components, the levels of between-male variance, the symmetry of the signal, the correlations between different aspects of drumming and their correlations with body weight. While in other taxa the frequency of audible signals may convey honest information of male size, in this species signal frequency was not related to male size and had a low repeatability. The pulse rate within each drum was highly repeatable but had a relatively small between-male coefficient of variation. In previous studies on this species, these traits were not important for male mating success. Among the traits directionally preferred by females, signal volume had considerable repeatability. Signal length was repeatable with high variability between males. In one population, signal length and volume were positively correlated with the rate at which males produced the drumming signals, a trait essential for male mating success. Thus, while signal length may reliably indicate male quality, other signal characteristics such as peak frequency and symmetry were not repeatable or were static and not related to any other male traits. Copyright 2000 The Association for the Study of Animal Behaviour.

Sexual signalling and viability in a wolf spider (Hygrolycosa rubrofasciata): measurements under laboratory and field conditions

Abstract This study examined the crucial prediction of the conditional-handicap theory, the relationship between male sexual trait size and male viability, in the wolf spider Hygrolycosa rubrofasciata. In this species, males court females by drumming dry leaves with their abdomen, and males with the highest drumming rate enjoy highest mating success. We determined male drumming rate, body mass, and mobility, which reflects mate-searching activity, in relation to male survival. Because it is often difficult to know how results obtained from laboratory studies reflect the natural world, particularly when the measured variable is survival, we repeated our study in both laboratory and field conditions. Males drumming at the highest rate survived better than males drumming at a lower rate in both laboratory and field conditions. These results are in accordance with the predictions of conditional-handicap models of sexual selection. Survival was independent of male body mass and there was no significant correlation between male drumming activity and body mass. However, large males moved further than smaller males, and males moving longer distances lost less mass than those moving shorter distances. These results suggest that, moving, and consequently mate-searching activity, may be a condition-dependent trait and that there may be a advantage for large males in mate searching.

Honesty of agonistic signalling and effects of size and motivation asymmetry in contests

Abstract Game theoretical models predict that the main function of fighting behaviour is to assess the relative fighting ability of opponents. The sequential assessment game has often been used to investigate contests, while honest signalling theory has received much less attention. With the wolf spider Hygrolycosa rubrofasciata we investigated whether male agonistic signalling can reveal honest information about fighting ability, and how size and motivation asymmetries affect male fighting behaviour. We also determined whether male–male competition affects the courtship behaviour of the males. We found that agonistic drumming activity is an honest indicator of male fighting ability, and that relative size asymmetry and motivation to fight both contribute to the fighting ability. We also found that male–male competition decreases the courtship drumming rate of subdominant males, suggesting that male–male competition limits the opportunities for female choice. We conclude that sequential assessment and honest signalling may both be used in settling contests, and that they may be used simultaneously.

Restrictive mating by females on black grouse leks

In bird species with pair bonds, extra‐pair matings could allow females to choose genetically superior males. This is not needed in lekking species because female choice is not constrained by pairing opportunities. However, polyandry has been reported in most lekking species studied so far. Using 12 microsatellite loci, we determined the paternity of 135 broods of black grouse sampled between 2001 and 2005 (970 hatchlings and 811 adult birds genotyped). The paternity assignments were combined to lek observations to investigate the mating behaviour of black grouse females. About 10% of the matings seemed to take place with males displaying solitarily. Forty per cent of the copulations between males displaying on the studied leks and radio‐tagged females were not recorded. This was due to difficulties in identifying the females and because our observations did not cover all the possible time for matings. However, females of the undetected copulations had chosen males that were already known to be successful on the leks. There was a strong consistency between the observations and true paternity, even when the copulation was disturbed by a neighbouring male. Multiple mating and multiple paternities were rare. We can now confidently ascertain that most females mate only once with one male for the whole clutch. This mating behaviour requires that a single insemination is sufficient to fertilize a clutch and that females can determine whether the sperm has been successfully transferred. Grouse Tetraoninae with many lekking species may be the only bird taxon that has evolved these traits.

Microhabitat selection and audible sexual signalling in the wolf spider Hygrolycosa rubrofasciata (Araneae, Lycosidae)

Abstract We studied the microhabitat selection and male sexual signalling behaviour in the wolf spider Hygrolycosa rubrofasciata (Ohlert). Males strike dry leaves with their abdomen, producing an audible sexual drumming signal, and females use this signal to choose mating partners. In the field we followed male drumming rate and microhabitat selection using both the mark–recapture method and direct observations. In the laboratory we conducted an experiment on male microhabitat and drumming substrate selection. We found that in the field males were not distributed randomly among the habitat; fewer males were found in areas that had high sedge cover, low elevation, and low dry leaf cover. In the laboratory experiment, males spent more time on dry leaf substrate. Drumming rate in the field was positively correlated with dry leaf cover and in the laboratory males clearly preferred dry leaves as drumming substrate. Temperature was positively correlated with male drumming rate, and with male and female mobility. We conclude that in H. rubrofasciata male distribution and sexual signalling rate, and thus mating success, are greatly affected by environmental factors. Therefore, males may be sexually selected to make effective use of their signalling habitat by active microhabitat choice.