Yolk testosterone organizes behavior and male plumage coloration in house sparrows (<Emphasis Type="Italic">Passer domesticus</Emphasis>)

Vertebrate eggs contain maternal steroid hormones in their yolks; and in avian species the concentration of these steroids vary within and among clutches. The organizational actions of these variable doses of maternal steroids in the avian egg, analogous to those regulating the development of sexually dimorphic traits, are little explored. In this study, we examined the organizational effects of yolk testosterone in the house sparrow, Passer domesticus, a sexually dichromatic passerine bird in which males are characterized by the presence of a throat patch or badge of black feathers that varies in size among males. We injected a physiological dose of testosterone or vehicle into eggs in the field, removed nestlings, and hand-raised them in the laboratory. At 5 months of age, we investigated treatment effects on plumage and behavior. Males that hatched from testosterone-injected eggs developed a larger badge than males that hatched from vehicle-injected (control) eggs. However, testosterone did not induce the expression of a badge in females. In staged dyadic encounters, both males and females hatched from testosterone-injected eggs were more successful at obtaining and defending a food source than individuals of the same sex hatched from control eggs. The results suggest that variable concentrations of maternal testosterone in the eggs of the house sparrow organize the expression of a plumage trait in males and behavior in both sexes.Communicated by J. Graves     

Sexual conflict in the house sparrow: interference between polygynously mated females versus asymmetric male investment

Summary The reproductive success of female house sparrows mated with polygynous males depended to some extent on the aid received from their mates. Polygynous males fed nestlings at the same rate as monogamous males although polygynous males gave aid almost exclusively to one of their mates (the preferred). As a consequence, the number and quality of young raised by preferred females were similar to those of monogamous females, whereas nonaided females experienced a reduction of clutch size, hatching success, and fledgling quality. Males aided the harem female that hatched her clutch earliest (in 11 out of 12 cases). However, harem females that laid eggs earliest (i.e., the potential preferred) lost eggs and/or nestlings more frequently than harem females that delayed laying. This resulted from competing females' infanticide suited for redirecting males' aid. All polygynously mated females delayed the start of a new clutch and reproduced fewer times per season than monogamous females. Consequently, monogamy seems to be the optimum mating situation for females; whereas polygynous males raised more young per year than monogamous males, interference between harem females reduced their maximum expected success.     

Kin-preferential cooperation,dominance-dependent reproductive skew,and competition for mates in communally nesting female house mice

Little is known about the behavioural mechanisms facilitating kin-preferential communal breeding in wild house mice (Mus domesticus). We evaluated the effect of kinship and male availability on aggression, social structure and reproductive skew in groups of female mice freely interacting and reproducing in semi-natural indoor enclosures. Triplets of either sisters or non-sisters were established in enclosures provided with either one or three littermate males, which were unrelated and unfamiliar to the females. Sisters were more spatially associated and less aggressive than non-sisters, leading to higher incidences of communal breeding and reproduction. This is in agreement with theoretical considerations on kin selection in house mice. Reproductive success was highly skewed in favour of dominant females due to subordinate infertility or complete loss of first litters, which might have been caused by dominant females. In spite of this, subordinates only rarely dispersed from the enclosures, suggesting that perceived dispersal risk generally outweighed relatively reduced reproductive potentials. Aggression levels among females were significantly higher when one male was available, compared to when three males were available. We suggest that this might result from higher female-female competition for mates, due to the risk of missing fertilisation when synchronously oestrous females encounter limited numbers of males in a deme. Our results indicate that, first, communal nursing in house mice might have evolved to make the best out of a bad job rather than to enhance offspring fitness; and, second, that female-female mate-competition might play an important role in shaping female social structure in this polygynous mammal.Electronic Supplementary Material Supplementary material is available in the online version of this article at Communicated by F. Trillmich     

Does the cost of polygyny in house wrens include reduced male assistance in defending offspring?

Summary This study investigated whether reduced male aid in defending offspring potentially reduces the fitness of females choosing already-mated males in the house wren (Troglodytes aedon), a small, territorial songbird. Frozen snakes were placed at 23 nests of monogamously mated males and 12 secondary nests of bigamously mated males. All presentations were made during incubation stages of females attending focal nests. Snakes were placed at nests of secondary females when nests of their primary counterparts contained young 5–9 days old. Males are most attentive to primary nests during this period and should therefore be relatively inattentive to secondary mates and nests. Nevertheless, an equal proportion of monogamous and bigamous males discovered snakes within 15 min, and mean time to discovery, when discovery occurred, did not differ with nest status. Monogamous and bigamous males were also equally likely to attack snakes physically once discovered. Monogamous males appeared no more likely to discover snakes than bigamous males for two main reasons. First, although monogamous males were near focal nests (i.e., < 10 m) more often than bigamous males, monogamous males tended to stay out of view of nests for long periods. In contrast, bigamous males always went immediately to focal nests upon arriving in their vicinity. Second, about one-third of monogamous males in this study spent much of their time during trials at the far edges of their territories advertising for secondary mates. Our experiment suggests that reduced male aid in defending nests against small, diurnal predators probably does not contribute to the cost of polygyny in house wrens. Correspondence to: L.S. Johnson     

Intake and coversion of food in the fish Limanda limanda exposed to different temperatures

In the flat fish Limanda limanda L., feeding rate and conversion efficiency were studied as functions of body weight, sex, temperature and food quality. When offered herring meat at 13 °C (series I), females (live weights 1 to 150 g) consume more food than males; the magnitude of this difference is body weight-dependent. With increasing wieght, both females and males consume less food per unit body weight per day. Variations in daily ration are considerable; the range of deviation from mean feeding rate is about 60% for males and 40% for females. The range of deviation does not vary significantly among females and males of different body weights. At the same temperature level (13 °C; series II), females consume almost the same, or even less, cod meat than males. Among individuals of series I and II, there is a little difference in the feeding rate; however, herring-fed individuals obtain about 2 times more energy than cod-fed individuals. Each gram wet weight of herring meat yields 2001, each gram cod meat 1137, calories. Small individuals completely cease to feed at 3°C; they feed little at 8 °C. Larger females consume maximum amounts at 8 °C. Small individuals consume maximum amounts at higher temperatures. Thus, with increasing body weight (age), the temperature for maximum feeding shifts downwards. Feeding with cod or herring meat results in considerable changes in composition and calorific content of L. Limanda. The magnitude of these changes depends both on temperature and food quality. Food conversion efficiency values of herring-fed individuals are about 1 1/2 times higher than of cod-fed individuals. In series I and II, females are more efficient converters than males. In individuals weighing more than 50 g, conversion efficiency decreases in the order: 8°, 13°, 18° C; in smaller individuals this order is 13°, 18°, 8 °C. Conversion rate is about 2 to 5 times faster in individuals fed herring meat than those receiving cod meat. Conversion rate decreases in the order 13°, 8°, 18 °C in males, and in the order 18°, 13°, 8 °C in females; females of more than 80 g are exceptional in that they reach the maximum at 8 °C. From the data on food intake and food conversion, the biologically useful energy available for metabolism has been calculated for each test individual kept at 13° and 18 °C. At these temperature levels, the weight exponents are about 0.6; the a value or metabolic level for the 18 °C series is about 2 times higher than that at 13 °C. Thus, temperature affects metabolic rate but not the exponential value. The exponential value for the body weight-metabolism relation at 13 °C is for dab fed herring meat 0.9; the a value amounts to about half that for dab fed cod meat. Food quality, unlike temperature, alters not only the exponential value but also metabolic rate.     

Acquisition of breeding coloration depends on badge size in male house sparrows Passer domesticus

Summary The black throat badge of the male house sparrow Passer domesticus, which functions as a status signal of dominance rank, changes in size during late winter because light feather tips gradually wear off. Males change the size of their visible badge earlier if their final badge size is large. The differential timing of the change from winter to breeding coloration in relation to badge size is partially controlled by the behaviour of the individual male. Feather abrasion of the badge is mainly due to preening and dust bathing. During late winter, male house sparrows preen the badge area more than females preen the homologous area, and males with large badges preen more than males with small badges. The seasonal change in size of the visible badge of male house sparrows may reflect the balance of different selection pressures. A large badge size signals dominance status in autumn and winter while predation by visually searching avian predators may constitute the main opposing selection pressure. Sexual selection causes a stronger selection pressure for a large badge in spring and summer. Correspondence to: A.P. Moller     

Female nutritional state affects the rate of male incubation feeding in the pied flycatcher Ficedula hypoleuca

Summary Male pied flycatches Ficedula hypoleuca regularly feed their mates during incubation. By experimentally supplying some females with extra food we studied how the female's nutritional state affected her incubation schedule and the rate at which her mate fed her. Females that received extra food spent more time on the nest and shorter periods away from it, compared with control females. This suggests that nest attentiveness is governed by the amount of energy available to the female. When females reccived extra food, males decreased their rate of incubation feeding. They also did so in response to increasing ambient temperatures, whereas incubation schedules were unaffected. We, therefore, conclude that our results support the female nutrition hypothesis, i.e., that the food provided by the male constitutes a significant nutritional contribution to the incubating female.     

The effect of intrauterine position on the survival,reproduction and home range size of female house mice (Mus musculus)

Summary In laboratory studies using albino house mice, a female's prior intrauterine position can affect many postnatal physiological, morphological and behavioral characteristics. Females flanked by males in utero (2M females) exhibit more aggressive dominance than females flanked by females (OM females). Thus, wild 2M females may be most successful during peak population densities when their aggressive nature would allow them to displace other females from limited resources. 2M and 0M females and males delivered by cesarean section were individually marked and released as young adults on two occasions onto a highway island (the area enclosed by exit and entrance ramps at an interchange) to determine whether 2 M females have a competitive advantage over 0 M females in the field. Males were included to create realistic population structure; their intrauterine position was not a treatment. Feeding stations afforded individuals an opportunity to exhibit their dominance by maintaining home-ranges at or near the stations. The populations were monitored by periodic live-trapping and reproductive success was determined using field body weights and by post-mortem examination for uterine implantation scars. Survival and capture rates were estimated, using a modified Jolly-Seber markrecapture program, for each of four intervals between trapping occasions over the course of 7 weeks. There were no overall differences in survivorship between 2M and 0M females, neither type of female was caught more frequently at feeding stations and they did not differ in measures of reproductive success. However, 2M females had significantly larger home-range sizes than 0M females and thus space use may be a trait masculinized by prior intrauterine position. Although there are a number of life-history characteristics that differ between 0M and 2M females in the laboratory that we did not test specifically in the field, our findings and other features of wild house mouse biology suggest that prior intrauterine position does not have a strong effect on survival and reproduction in the wild. Offprint requests to: W.J. Zielinski at the present address     

Dominions and desert clickers (Orthoptera:Acrididae): influences of resources and male signaling on female settlement patterns

Summary Previous field work on the grasshopper Ligurotettix coquilletti revealed that females were not evenly distributed among male mating territories, Larrea tridentata (creosote) bushes, but were clustered at particular locations. These sites generally harbored several signaling males simultaneously and also possessed foliage preferred by the insects as a food source, this preference being based on the relative concentrations of various extra-foliar compounds. The clustering of females, therefore, could result from a preference for specific bushes because of the resources (i.e., food) available there and/or an orientation to groups of males per se. Here, we present the results of 3 field experiments in which we controlled the spatial distribution and intensity of male signals using a computer-operated system of loudspeakers and monitored the movement of individually marked females released in the study area. When male song was identical at high and low quality territories (all having single loudspeakers), females still aggregated at the high quality sites, indicating that variability in host plant quality alone may be sufficient to promote a skewed distribution of females. Among high quality territories, females did not discriminate between sites with one versus three loudspeakers (all broadcasting the same signal), but displayed a strong preference for sites (all having single loudspeakers) with a high intensity signal over a low intensity one. Field measurements showed that the songs of grouped males were more intense than those of lone males, implying that the signaling of grouped males may have enhanced the settlement of females at the bushes harboring male groups above and beyond that influenced by territory quality alone. We conclude that female attraction to host plants is influenced primarily by male signaling, whereas their subsequent retention is more dependent on territory quality. An experiment on male settlement failed to show an aggregative tendency, suggesting that male groups form through the passive accumulation of individuals at high quality sites.     

Fifteen-spined stickleback (Spinachia spinachia) females prefer males with more secretional threads in their nests: an honest-condition display by males

In the fifteen-spined stickleback (Spinachia spinachia), the male alone builds a nest and provides care for the eggs until they hatch. The nest is made of filamentous algae held together with shiny secretional threads of a glycoprotein, here called tangspiggin. Tangspiggin is produced by transformed kidney cells. I investigated how tangspiggin production was related to food intake by giving males low or high food rations. Males in the high-food group produced significantly more tangspiggin than those in the low-food group. Since low food rations did not lead to a significant drop in resting metabolic rate (oxygen consumption), tangspiggin production appears to be one of the first energy-demanding processes to be dropped during food deprivation. Among males given equal food rations, females preferred to mate with those with elevated levels of tangspiggin in their nests. Tangspiggin threads may signal safety for the offspring by keeping the eggs within the nest and protecting them from egg predators. Moreover, the threads become less shiny with age, due to particle deposition, and older nests that may house older (more valuable) eggs are thereby well camouflaged. Tangspiggin provides an honest signal of male condition, acting as an extra-bodily ornament, signaling direct benefits to females or possibly also indirect benefits of "good genes" from a male who is a successful forager.     

Fecundity studies on Acartia tonsa (Copepoda: Calanoida) in standardized culture

Acartia tonsa Dana longevity and egg production data were studied over a 4-year period under standardized culture conditions. Egg-laying patterns and daily rates were evaluated as bioassay tools for measuring the nutritional value of various copepod foods, as well as assessing other aspects of environmental quality. Altogether, 337 females were observed in various aspects of this work. Experiments were performed with newly matured females in loosely covered crystallizing dishes. Fresh water and a standard concentration of algae were supplied daily, and a running daily tally was kept of the eggs produced. The effects of male copepods on egg production were analyzed. Males were either removed after the first eggs were observed or they remained with the female throughout her life span. Daily egg-laying rates of this latter group of females appeared to be more stable, but abnormal embryo development occurred if the males were not removed after fertilization. Likewise, the results of females which were not mated were studied. Life spans were shorter and egg production rates were lower than those of contemporary mated females. Unmated females were subjectively observed to be less active than their mated counterparts. The effects of continuous and temporary starvation on egg-laying were also observed. Under both normal and starved conditions, males appeared to have shorter life spans than either mated or unmated females. During starvation there were no real differences between the life spans or egg productions of mated or unmated females. Non-nutritional particulate matter was evaluated as a food source. It was ingested by the females, but did nothing to prolong the egg-laying period. Finally, the three algal components of the standard ration were evaluated. Females produced equally well with a mixed diet or with Thalassiosira pseudonana (100 m³) or Chroomonas salina (140 m³), the two larger species of this diet, but when Isochrysis galbana (40 m³) was fed, females reacted as if in a starved situation.     

Sexual behavior is related to badge size in the house sparrow Passer domesticus

Summary Badge size, which functions as a signal of dominance status in male house sparrows Passer domesticus, was significantly related to their sexual behavior. Males with large badges participated in communal displays (multi-male chases directed towards single females) more often than males with small badges, irrespective of whether the female involved was the male's mate or not. Experimentally released females were more often chased if they were fertile than if they were nonfertile. Estradiol-implanted females were chased more often than control females without a hormone implant, and males with large badges chased estradiol-implanted females more often than did males with small badges. Both forced extra-pair copulations during communal displays and unforced extra-pair copulations were more often achieved by males with large than small badges. Male house sparrows with large badges also copulated with their mates at a higher rate than did males with small badges. A higher certainty of paternity therefore is hypothetized to accrue to male house sparrows with large badges.     

The conflict between feeding and territorial defence in the great tit

Summary A field experiment investigated whether feeding and territorial defence competed for time in the activity budget of territorial male great tits during the spring. Feeding tables were placed on the territories of five males. Five other territorial males also fed at these tables, while ten additional males had no access to any feeding tables. A standardised intrusion using playback of territorial song and a stuffed mount of a male great tit was conducted on each of the twenty territories, and the response of the resident male measured. All of the males with access to a feeding table responded more vigorously to the intrusion than the males who received no extra food. These results are not attributable to the fact that males defended their territories more vigorously because of the addition of feeding tables, since males that gained extra food outside the boundaries of their own territories also defended more vigorously than birds with no access to a feeding table. The results support the conclusion that the provisioned males defended more vigorously because they could afford to take more time out from feeding.     

Sex-related asymmetry in competitive ability of sexually monomorphic barn swallow nestlings

In diverse taxa, offspring solicit parental care using complex displays, which may evolve as reliable signals of condition or as mechanisms to manipulate parental investment. Differential sex allocation may therefore result from adaptive parental decisions or sex-related variation in competitive ability or because of sex-related asymmetries in kin selection. Under normal food provisioning, female barn swallow (Hirundo rustica) nestlings begged more loudly but did not receive more food than male nestlings. After food deprivation, begging call loudness of males but not females increased. Begging loudness positively predicted the number of feedings received by the nestlings, and males gained more mass than females after food deprivation. Male nestlings are more severely affected by chronic food reduction and may therefore accrue a larger benefit compared to females by increasing their food intake under short-term conditions of food scarcity. These results suggest that either females do not increase begging intensity to favour male broodmates which are more vulnerable to prolonged food stress, or that males prevail in scramble competition despite being similar in size to females.     

Age determination and growth of the bonnethead shark Sphyrna tiburo: a comparison of two populations

From July 1982 to December 1986, a study of the age and growth of the bonnethead shark Sphyrna tiburo was conducted in Tampa and Florida Bays, Florida, USA. Tetracycline-injected sharks held in captivity and, to a lesser extent, tagged, released and recaptured, were utilized for validating the annual nature of the rings (herein defined as the narrow, translucent regions) appearing on vertebral centra. The technique was validated for all age groups (0 to 6+yr) included in the study. Marginal increment analysis likewise suggested annual ring formation. The rings formed during the winter, when water temperatures were lowest. Using the vertebral ring aging-technique, von Bertalanffy growth curves for males and females from both tampa and Florida Bays were constructed. Growth of sharks born and held in captivity demonstrated that the male and female growth curves diverge after 1 yr and that mean sizes at age are statistically distinct after 2 yr. In both populations, females grew to larger sizes tha males and apparently are longer-lived. Reproductively mature females from Tampa Bay were significantly larger than those from Florida Bay.     

Seasonal observations on the biometry and development in Copepodite Stage V of Calanus finmarchicus from the Firth of Clyde

Relationships between linear dimensions of casts produced at the final moult in the laboratory and definitive sex confirm that Calanus finmarchicus (Gunn.) is sexually dimorphic in Copepodite Stage V. Two forms of Stage V copepodite were present at all times of the year in the Firth of Clyde, Scotland. Except in early spring, when size was similar, these were consistently distinguished by the prosome length. The small form moulted to adult females and the large form to adults of either sex. Relative to the metasome, the length of the cephalosome was greater in the large than small form of Stage V copepodite. This, to an increased extent, distinguished males from all females in the adult. Differences between females of the large and those of the small form were similar in the Stage V copepodite and adult. Feminization in the large form entailed an arrest in the differentiation of secondary male sex-characters. Most morphological females moulting from both forms of Stage V copepodite possessed an apparently normal ovary, with eggs extending into oviducal diverticula. In the case of the large form, however, eggs were occasionally absent from diverticula. The gonad in two specimens resembled an immature testis. In males, a normal testis was always present. Differentiation into the adult male entailed the internal suppression of feminizing factors. On average, within collections, females of the large form moulted later than the small form and definitive males. A similarly consistent difference between females of the small form and males was absent. Moulting times in the large form of Stage V copepodite depended on the presence or not of feminization. Differentiation into the adult male required the environmental stimulation of development. Comparisons with moulting patterns in the sea suggest a mechanism that regulates the recruitment of males to the adult.     

Abnormal gonadal morphology in intersex,<Emphasis Type="Italic"> Echinogammarus marinus</Emphasis> (Amphipoda): a possible cause of reduced fecundity?

Recent reports have demonstrated a cost associated with intersexuality in Amphipoda, including reduced fecundity and fertility. In this study, the gross morphology of the gonads in normal and intersex Echinogammarus marinus (Amphipoda) were compared to determine whether resource allocation to gonadal tissue accounted for this reduced fitness. Evidence for the presence of the male sex-determining hormone, androgenic gland hormone (AGH), was compared between sexual phenotypes using MALDI mass spectrometry. Two distinct intersex phenotypes (male intersex and female intersex) were found, with variation in gonadal structure corresponding with external phenotype. Examination of male intersexes revealed normal testicular development (testes, seminal vesicles and vas deferens), but also revealed the formation of an oviduct. Ovaries of intersex females showed normal ovarian development, but were reduced in length by approximately 20% due to the presence of vas deferens. The number of vas deferens in intersex females was equal to the number (one or two) of genital papillae. We hypothesise that the reduced ovarian length observed in intersex females is a likely cause of the reduced brood size previously reported in intersex females of this species. Variation in the sexual phenotype corresponded both to development of the androgenic gland and to expression of a peptide fragment corresponding to the A chain of androgenic gland hormone (AGH). Androgenic glands and a putative AGH peptide were present in males. However, in both normal and intersex females, the androgenic glands were only present in a rudiment form and the peptide was not detected. Intersex males were found to possess abnormal glands that appeared hypertrophied. However, AGH peptides were not detected, supporting the suggestion that the intersex phenotype is manifested via perturbations of AGH.An erratum to this article can be found at Communicated by J.P. Thorpe, Port Erin     

Experimental studies on water strider mating dynamics: spatial variation in density and sex ratio

Summary We used field surveys, field experiments and experiments in artificial pools to study the effects of variation in sex ratio and density on mating dynamics of a stream water strider, Aquarius remigis. Our field survey documented the existence of hot spots, sites of higher than average total gerrid density, a male-biased sex ratio, and higher than average female mating activity. Female gerrids frequently changed sites, usually moving upstream, perhaps to spread their eggs among many sites. Male gerrids showed two movement strategies: some males frequently changed sites, while other males were stationary at hot spots. Surprisingly, smaller males tended to be stationary at hot spots. A field manipulation of the availability of refuges for females to avoid harassment by males supported the notion that males prefer hot spots because they are sites where a scarcity of refuge for females makes it relatively easy for males to intercept females. Experiments in plastic pools compared sites with 20 males: 5 females (simulating hot spots) to pools with 5 males: 5 females. The rate of male harassment of females was higher in 20:5 pools as compared to 5:5 pools. In response to increased male harassment, females reduced their activity on the water and increased their time spent out of the water and thus unable to forage. Males showed a large male mating advantage (LMMA) in 5:5 pools, but, surprisingly, not in 20: 5 pools. This pattern can explain the field observation that small males prefer hot spots. A behavioral mechanism that can explain the LMMA is as follows. Mating occurs when males overcome female resistance. Larger males have a mating advantage over smaller males if females resist heavily. Increased harassment (e.g., in 20:5 pools as compared to 5:5 pools) might result in reduced female resistance to males and thus a reduced LMMA. Females also showed some non-random mating by size that might reflect an interplay between male preference for large females and the avoidance of males by large females. Correspondence to: A. Sih     

Re-copulation and post-copulatory mate guarding increase immediate female reproductive output in the dragonfly Nannophya pygmaea Rambur

After copulation, male Nannophya pygmaea dragonflies mate guard by hovering over ovipositing females and repelling conspecific males. Copulation is not always a prerequisite for oviposition in the females of this species because females can store the sperm received during previous visits/copulations. An oviposition episode consists of several bouts of oviposition separated by periods of perching. We conducted two types of male-removal experiments to examine the effects of mating and post-copulatory mate guarding on the oviposition behaviour of females. In the first experiment, we removed all males from the habitat to eliminate the effect of re-copulation, mate-guarding and harassment by males. In the second experiment, we removed males immediately after copulation to eliminate the effects of guarding and other post-copulatory male-female interactions. We compared these experimental data with data obtained under natural conditions. The dipping rate in an oviposition bout was not influenced by copulation or guarding. However, guarded females made more dips per episode than did solitary females. The proportion of time actually spent ovipositing (total bout duration/oviposition episode duration) of guarded females was higher than that of solitary females. Solitary females often oviposited in more than one territorial site, while guarded females usually oviposited within a single territorial site during an oviposition episode. Because males tend to hold territories at sites where egg survival is high, guarded females (and the male guardian) benefit from guarding in terms of egg hatchability. The possible benefits for solitary females are discussed.     

Selective males and ardent females in pipefishes

Summary In the pipefishes Syngnathus typhle and Nerophis ophidion, males have been shown to limit female reproductive rate, and females to compete for access to males. Hence, these species fit the criteria for sex-role reversal. Males brood the eggs and provide the offspring with nutrients, oxygen and an osmoregulated environment. Moreover, in S. typhle both sexes prefer a larger mate when given a choice. Sexual selection theory predicts that males should be more choosy than females, and that was experimentally demonstrated in this study. We predicted that S. typhle males should be less eager to copulate than S. typhle females with an unattractive (i.e. small) mate. We measured eagerness as the time from the start of the experiment until copulation occurred. Males with unattractive partners took significantly longer to copulate than females with unattractive partners. Moreover, females invariably initiated the courtship dance, and resumed it quicker after copulation than did the males, again suggesting reproductive hesitation in males. Neither male nor female size per se was correlated with time until copulation. In N. ophidion, where we have previously shown that males prefer larger to smaller females, we found that females did not select males with regard to size. Our results are consistent both with earlier findings (males limit female reproduction and females compete for males) and with operational sex ratios in nature: in seven annual field samples in June, the numbers of S. typhle females with ripe eggs always significantly exceeded numbers of receptive males. Hence, the potential cost of being choosy in terms of lost matings is much higher in females than in males. In conclusion, S. typhle females were somewhat choosy, but less so than males, whereas N. ophidion females were not choosy at all. Correspondence to: A. Berglund