Animal Behaviour 79 (2010) 819–826
Contents lists available at ScienceDirect
Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav
Elaborate courtship enhances sperm transfer in the Hawaiian swordtail cricket, Laupala cerasina Tagide N. deCarvalho*, Kerry L. Shaw 1 Department of Biology, University of Maryland
a r t i c l e i n f o Article history: Received 19 July 2009 Initial acceptance 10 September 2009 Final acceptance 19 November 2009 Available online 25 January 2010 MS. number: A09-00482R Keywords: cricket ejaculate protection Laupala mating success nuptial gift
Males of many insect species engage in ritualized behaviours during courtship that include the donation of a food gift to their partner. Although there is extensive diversity in nuptial gift form, a gift provided before mating typically serves to facilitate copulation, thereby increasing a male’s mating success. Unlike other insects, the Hawaiian swordtail cricket Laupala shows protracted courtship that includes the serial donation of nuptial gifts prior to mating. Males transfer multiple spermless ‘micro’ spermatophores over several hours before the transfer of a single sperm containing a ‘macro’ spermatophore. By experimental manipulation of male courtship, we tested several hypotheses pertaining to the adaptive significance of protracted courtship involving nuptial gift donation in this system. We found that extensive courtship interactions improve insemination by causing the female reproductive tract to take in more sperm. This suggests that the enhancement of sperm transfer is due to the serial donation of microspermatophores, which would represent a relatively novel function for nuptial gifts provided before mating and highlights the importance of examining cryptic processes of sexual selection. Ó 2009 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
Males of many insect species provide nuptial gifts to their mates in the form of body parts, glandular secretions, regurgitated food, prey items and spermatophores (reviewed in Vahed 1998). If nuptial gifts increase a male’s mating success, sexual selection via female choice could provide an explanation for the evolution of nuptial gift donation. However, sexual selection via sperm competition might also contribute to the evolution of nuptial gifts if such gifts increase a male’s ejaculate transfer. In species where females mate multiply, increases in ejaculate transfer improve male fertilization success either by increasing a male’s numerical representation of sperm or by altering female reproductive behaviour (Parker 1970). The timing of nuptial gift donation relative to the timing of insemination varies considerably between species, and some authors have suggested a relationship between the timing and the adaptive significance of the donation: nuptial gifts provided before sperm transfer are hypothesized to improve male mating success, whereas nuptial gifts provided during insemination are hypothesized to provide ‘ejaculate protection’ by reducing
* Correspondence and present address: T. N. deCarvalho, Department of Embryology, Carnegie Institution for Science, 3520 San Martin Drive, Baltimore, MD 21218, U.S.A. E-mail address:
[email protected] (T.N. deCarvalho). 1 K. L. Shaw is now at the Department of Neurobiology and Behaviour, Cornell University, Ithaca, NY 14853, U.S.A.
female interference with ejaculate transfer and improving sperm competitive ability (Gwynne 1997). The Ensifera, crickets and katydids, provides many valuable systems with which to examine hypotheses about the adaptive significance of nuptial feeding behaviour because of the diversity of gift forms and the ease of experimental manipulation in this group. In the Ensifera, the ejaculate is enclosed within a capsule, the spermatophore ampulla, which remains external to the female after copulation (Loher & Dambach 1989). Sperm flows from the spermatophore ampulla into the female, and the duration of spermatophore attachment significantly influences the amount of sperm a female receives (e.g. in crickets: Sakaluk 1984, 2000; Simmons 1986). Several species of Ensifera provide food items to females before copulation that increase the probability of spermatophore transfer (i.e. mating success). For example, in the sagebrush cricket, Cyphoderris strepitans, females mount males and feed on specialized wing pads while males form the spermatophore (Dodson et al. 1983). Males that lack hindwing material are less likely to have a female remain mounted for spermatophore transfer and consequently, such males suffer lower mating success (Eggert & Sakaluk 1994). In contrast, male crickets that provide food gifts that females eat during insemination (and thus, that are offered during copulation or after spermatophore transfer) benefit by delaying female spermatophore removal. For example, many katydid and one cricket species produce a spermatophylax, a gelatinous substance that
0003-3472/$38.00 Ó 2009 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.anbehav.2009.12.017
820
T.N. deCarvalho, K.L. Shaw / Animal Behaviour 79 (2010) 819–826
coats the spermatophore (Gwynne 1995, 1997; Brown & Gwynne 1997). Females are distracted from prematurely consuming the spermatophore capsule while they eat the spermatophylax, resulting in fuller insemination (e.g. Sakaluk 1984; Wedell & Arak 1989; reviewed in Vahed 1998). In several species of crickets, an increase in sperm number increases a male’s paternity success when females multiply mate (Sakaluk 1986; Simmons 1987; Wedell 1991; Sakaluk & Eggert 1996). Additionally, some cricket species transfer ejaculate substances that manipulate female behaviour (e.g. causing increased rates of oviposition; Murtaugh & Denlinger 1985; Stanley-Samuelson & Loher 1986; Wedell & Arak 1989). These chemicals typically act in a dosage-dependent manner (Stanley-Samuelson et al. 1986); therefore, greater ejaculate transfer may also increase substances other than sperm that benefit males in sperm competition. The Hawaiian swordtail cricket Laupala (Gryllidae: Trigonidiinae) has an elaborate courtship, which includes nuptial feeding in the form of spermless ‘micro’ spermatophores. Courtship consists of repeated elements that include male song, antennation and microspermatophore (micro) transfer to the female, which occurs over the course of several hours in advance of a larger, sperm-containing macrospermatophore (macro) (Shaw & Khine 2004; deCarvalho & Shaw 2005). The transfer of both micro- and macrospermatophores involves genital coupling and insertion of the sperm tube into the female. Following each type of spermatophore transfer, the pair typically antennate for a period of time. After mutual antennation, the pair disengages physical contact and the female removes and consumes the spermatophore. Laupala courtship behaviour differs from that of other Ensiferan and insect taxa because it involves the donation of multiple gifts several hours in advance of mating (i.e. defined as the transfer of the sperm-containing macro). This extremely protracted sequence of nuptial gift donation prior to mating suggests the hypothesis that micro donation functions to improve macro transfer success (i.e. mating success). While micros are not directly used to entice females into mounting because the male and female are physically separated while the female consumes the gift, females may prefer to accept macro transfers from males that produce micros. A plausible alternative, however, is that micro donation benefits males via increasing ejaculate transfer. This could be accomplished by several strategies. First, micro transfer might delay female macro removal. If sperm flows from the macro at a gradual rate as in other crickets, an increase in the duration of macro attachment would result in fuller insemination. Although premating micro donation cannot directly protect the macro, micros might delay macro removal by females through some other means (e.g. gustatory satiation after micro consumption). Second, micros might alter female reproductive activity by delaying remating or stimulating early oviposition, thereby reducing the potential for sperm competition. Micros are spermless, but their donation might represent a strategy to increase the transfer of nonsperm substances. Third, micro transfer might induce greater sperm uptake from the macro into the female reproductive tract. This hypothesis is based on our own preliminary observations of macro sperm drainage patterns and represents an alternative to the ‘ejaculate protection’ hypothesis because it does not involve female macro removal behaviour. To understand the adaptive significance of micro donation in Laupala, we experimentally manipulated courtship using L. cerasina as the focal species. Here we consider micro donation in the broad sense, which includes the elements of courtship involved in their donation such as copulation, transfer and consumption. We compared mating success between two treatment groups: (1) females that received a natural number of micros followed by the sperm-containing macro, and (2) females that received only the
macro. We tested the mating success hypotheses by comparing the number of males that successfully transferred a spermcontaining macrospermatophore between the two groups. Using the same basic experimental design, we also investigated the ejaculate transfer hypothesis by measuring macro attachment duration, remating interval and sperm transfer quantity. METHODS Field Study Prior to laboratory experiments, we observed natural courtship behaviour at Kalopa State Park, Hawaii, U.S.A. Field observations were performed to document the onset of courtship relative to the photoperiod and number of micros transferred. In the laboratory, the number of micros produced by a male depends on the time at which pairs are introduced; late introductions result in fewer micros before mating. Therefore, the field data were used to determine when to introduce pairs, relative to the photoperiod, to achieve a natural number of micros for experimental treatments. Field observations of mating behaviour were conducted every day during 21–29 July 2004 at Kalopa State Park, Hawaii. Morning twilight occurred at 0529 hours, dawn occurred at 0553 hours, sunset occurred at 1901 and evening twilight ended at 1924 hours, according to the United States Naval Observatory. To observe the courtship sequence between a single pair of mating individuals, we marked the location of focal males found under tree bark in Metrosideros polymorpha (ohi’a) and in the recesses of Psidium guajava (guava). To ensure that we observed the onset of mating, we checked each site for female presence shortly after dawn and then every 10–20 min. Once courtship was established at a particular site, we recorded the timing of the following behaviours: onset of courtship (designated by mutual antennation in a face-to-face body position), micro production, and macro production and transfer. Laboratory Experiments Laupala cerasina nymphs were collected on the island of Hawaii in 2004–2006 and studied at the University of Maryland. Crickets were contained in quart-sized (0.094-litre) glass jars or plastic specimen cups lined with moist paper towels and Kimwipe tissues. Crickets were maintained on a diet of Fluker’s Cricket Feed (Fluker Farm, Port Allen, LA, U.S.A.), at 20–22 C on a 12:12 h light:dark cycle. Nymphs were isolated to same-sex containers to remain virgins for mating experiments. All three experiments shared the same basic methods and experimental design. Males and females were paired in plastic petri dishes with water-saturated Kimwipes. Pairs were established approximately 3.5 h after the onset of the photophase (8.5 h before the dark phase), similar to natural courtship observed in field portion of study. Two treatment groups were established for each experiment: (1) ‘micros absent’ females received only the macro; (2) ‘micros present’ females received a series of micros and the macro. Females were randomly assigned to each group. Group 1 was arranged by first pairing a male with a nonexperimental female, which accepted a series of micros. Courtship was allowed to progress until the macro was formed by the male (both types of spermatophores are externally extruded and macros become opaque with sperm approximately 10 min after formation), at which point the nonexperimental female was replaced with an experimental female. This was accomplished by slowly separating the petri dish when the nonexperimental female and male had moved to opposites halves. A second petri dish containing the experimental female was separated and connected to the half containing
T.N. deCarvalho, K.L. Shaw / Animal Behaviour 79 (2010) 819–826
the male. In this way, crickets were not directly handled and minimally disturbed. Experimental females interacted with males for approximately 45 min prior to mating, during which mutual antennation usually occurred. To control for the effect that petri dish disturbance may have had on female behaviour in group 1, the dish containing the mating pair in group 2 was likewise separated and reconnected at the same point in the courtship sequence (i.e. after macro formation, prior to macro transfer). Experiment 1: male mating success and female macro retention time We first tested the hypothesis that micro donation increases mating opportunities. Following the mating treatment described in the general methods, we recorded the number of females that accepted the macro (i.e. mated) for both groups. We also measured macro retention time, which has been known to affect ejaculate transfer in other systems. After macro transfer was complete, we recorded female macro removal behaviour. Females remove spermatophores using their hindleg, and it usually takes multiple attempts to complete sperm tube removal. Therefore, we recorded the time interval (min) between macro transfer and both the female’s initial attempt and complete removal of the macro. Experiment 2: female remating and oviposition interval We next tested the hypothesis that micro donation delays female receptivity and remating. Following the mating treatments described in the general methods, females were prevented from removing the macro in order to standardize macro attachment time across females. Immediately after macro transfer, females were coaxed down a plastic funnel into a 0.2 ml microcentrifuge tube, which was narrow enough to restrict leg movement. After 60 min, females were removed from the tube and anaesthetized with C02. The macro was removed with forceps and examined under a dissecting microscope for sperm. To measure the female remating period, females were paired with a different male each day subsequent to the initial mating until they remated. We did not introduce females to another male on the same day out of concern that females may vary in their anaesthesia recovery time. We recorded the interval (in days) between the initial mating and both the acceptance of a subsequent micro and macro. It was important to record both of these behaviours because females may accept one or more micros from a certain male and then terminate courtship, therefore remating (macro acceptance) may not occur until another day. We considered micro acceptance a signal of receptivity and macro acceptance a signal of remating. Females were introduced to males using the same methods as the initial mating and remained paired with males for at least 6 h per day. Females were paired with males every day for at least 16 consecutive days and then at least twice a week. Nonvirgin males were used for remating trials, which had prior mating success (defined by the transfer of a macro) at least 3 days before the remating trial. Work on other Laupala species has shown that well-fed, laboratory-reared males are able to produce a macro 1 day after mating (J. Jadin, personal communication). To determine the onset of oviposition, containers housing individual females were checked every day for eggs, beginning after the initial mating and ending when females remated. Experiment 3: sperm transfer success We tested the hypothesis that micro donation increases sperm uptake by the female reproductive tract. Following the mating treatments described in the general methods, females were observed until they attempted to remove the macro. At this time, females were immediately anaesthetized with CO2. To assess the amount of sperm that drained from the macro into the reproductive tract, macros were quickly removed from the female and
821
photographed on a haemacytometer without a coverslip (see Fig. 1). We then dissected the female to remove the spermatheca, which was placed on a haemacytometer, compressed with a coverslip and photographed (see Fig. 1). The haemacytometer design is such that raised edges hold the coverslip 0.1 mm off the marked grid; therefore, each spermatheca was flattened to the same degree. After photography, we dissected spermathecae and noted whether sperm cells were present. We also measured female femur length as a measure of body size. To evaluate whether micro substances add significant volume to the spermatheca, we added two treatment groups: (1) virgin females and (2) females that received only courtship/micros. The ‘micro only’ group was arranged by pairing crickets using the same protocol as the other experiments, the key alteration being that females were removed from the petri dish as soon as a male produced a macro. Spermathecae from these additional treatment groups were measured using the same protocol for mated females. Digital photographs of macro ampullae and spermathecae were taken with a JVC TK-1280U colour video camera mounted to an Olympus CH2 compound microscope. We performed line drawing and area calculations using image-analysis software (ImageJ 1.39, National Institutes of Health, Bethesda, MD, U.S.A.). To assess the amount of sperm drainage from the macro, we circumscribed the perimeter of the sperm mass and macro ampulla. We subtracted sperm mass area from the total ampulla area to calculate the ‘sperm drainage’ area. To assess the amount of sperm transferred to the female, the edge of the flattened spermatheca was circumscribed to calculate spermathecal area. The spermatheca is elastic and spherical, which allowed us to measure sperm volume indirectly. Unfortunately, we were unable to perform direct measures of sperm number despite extensive efforts to do so. Laupala sperm agglutinate to a degree that precluded individual cell counts and they are not readily dissociated either by standard methods or by several methods that we attempted to develop. Statistical Analyses Analyses were performed using JMP 7.0 for Windows (SAS Institute, Inc., Cary, NC, U.S.A.). All tests of significance were twotailed and results are reported as means SE, unless otherwise noted. Each laboratory experiment consisted of two treatment groups; therefore Student’s t tests and Pearson chi-square test (or Fisher’s exact) analyses were used for comparisons when possible. In the experiment to compare the remating interval between the two groups, many females remated the next day, which resulted in a non-normal, right-skewed distribution. We performed a permutation randomization test involving 10 000 iterations in SAS 9.1.3. This data set also included remating intervals of varying precision. There were two females in each group that mated much later than the rest (23–32 days); therefore, after a certain point these females were assayed for receptivity only twice a week. This resulted in an interval of 2–4 days during which these females regained receptivity, so we used the interval midpoint for our analyses. To test the sperm transfer enhancement hypothesis, we performed two replicate experiments in the spring and autumn of 2006. To determine whether it was appropriate to pool data from both experiments, we performed an ANCOVA to evaluate interactions between our treatment groups and experiments. There was no significant interaction between treatment groups and experiments for any of the dependent variables in our study; therefore, we combined the two data sets for the final analyses. To evaluate the relationship between sperm transfer and time, we performed regression analyses of sperm drainage and spermatheca area on macro attachment duration. We used macro attachment duration
822
T.N. deCarvalho, K.L. Shaw / Animal Behaviour 79 (2010) 819–826
Figure 1. Representative images of macro ampullae and spermathecae measured for sperm transfer experiment. The macro photos represent consecutively larger sperm drainage areas (i.e. smaller amounts of sperm) from left to right. The final macro on the right was devoid of sperm. The spermatheca photos portray consecutively larger spermathecae from left to right. The first spermatheca on the left was devoid of sperm.
as a covariate when it had a significant relationship with dependent variables and had no significant interaction effect between treatment groups. To compare the amount of sperm that a female received between the two groups, we used both macro sperm drainage area and spermatheca area. However, we first performed a regression analysis of spermatheca area on macro sperm drainage area to determine whether spermatheca area indicates the amount of sperm a female has received, and we found a significant relationship between sperm drainage and spermatheca area. Female femur length (i.e. body size) was not correlated with spermatheca size, and therefore, was not included in our analyses.
RESULTS Field Study Twelve mating pairs of L. cerasina were observed in the field, although some variables were not collected for certain pairs. Courtship began at 11:03 1:19 h:min (mean SD, N ¼ 10), which was approximately 5 h after sunrise (8 h before dark). Macro production occurred at 15:50 0:25 h:min (mean SD, N ¼ 8) and macro transfer occurred at 16:48 0:20 h:min (mean SD, N ¼ 8). The mean SD total number of micros produced was 6.6 0.91 (N ¼ 8).
We compared the number of micros produced by field males to males from experiment 1, which were paired 3.5 h after the beginning of the laboratory photophase (8.5 h before dark). Males in experiment 1 produced a mean SD of 6.9 0.98 micros (N ¼ 24). There was no significant difference between the two groups in the number of micros (t test with equal variance: t30 ¼ 0.654, P ¼ 0.52).
Laboratory Experiments Experiment 1: male mating success and female macro retention time We found no significant difference in the number of females that mated between the two groups (Fisher’s exact test: P ¼ 0.35; micros present ¼ 77.8% mated (14/18); micros absent ¼ 93.3% mated (14/15)). Two females from the micros present group overtly rejected macro transfer; rejection was indicated by the male’s failure to transfer the spermatophore after moving into the copulation position. The three other mating failures (across both groups) were due to early termination by the male, indicated by the male’s removal of his macro without attempting to copulate with the female. After copulation, there was no significant difference between the two groups in the amount of time before females attempted to remove the spermatophore (t test with equal variance: t26 ¼ 0.745, P ¼ 0.46; Table 1) or successfully removed the macro (t test with equal variance: t26 ¼ 0.715, P ¼ 0.48; Table 1).
T.N. deCarvalho, K.L. Shaw / Animal Behaviour 79 (2010) 819–826
823
Table 1 Mean SE intervals from macro transfer to initial removal attempt and successful removal of macro, and from initial mating to subsequent acceptance of a micro (i.e. receptivity) and a macro (i.e. remating), and mean SE sperm drainage area of macro ampullae and spermathecae for each group Treatment group
Initial removal attempt, complete macro removal (min)
Micro acceptance, macro acceptance (days)
Sperm drainage area (mm2)
Spermatheca area (mm2)
Micros present Micros absent Micros absent (including only females that received sperm) Micros only Virgins
N¼14, 25.64.0, 36.54.8 N¼14, 30.24.6, 41.55.0
N¼29, 4.81.6, 5.61.7 N¼29, 2.91.3, 3.41.3
4
4
N¼23, 0.3650.010 N¼22, 0.2540.022 N¼16, 0.2890.025
N¼23, 0.8220.042 N¼20, 0.7370.057 N¼15, 0.8040.063
4 4
4 4
4 4
N¼12, 0.3380.051 N¼17, 0.3670.017
4 Measurements not applicable.
Experiment 3: sperm transfer success We compared spermatophore removal behaviour between the two groups and found no significant difference in the time until the first attempt to remove the macro (micros present females: 33.9 2.2 min, N ¼ 23; micros absent females: 37.6 3.0 min, N ¼ 25; t ¼ 1.06, P ¼ 0.29). Furthermore, there was no relationship between the duration of macro attachment and the amount of sperm drained from the spermatophore for either group (linear regression: micros present: r2 ¼ 0.002, N ¼ 23, P ¼ 0.83; micros absent: r2 ¼ 0.069, N ¼ 22, P ¼ 0.23). There was a significant relationship between duration of macro attachment and spermatheca size within the micros present group (r2 ¼ 0.226, N ¼ 23, P ¼ 0.021) but not within the micros absent group (r2 ¼ 0.107, N ¼ 20, P ¼ 0.16). When individual females that did not receive sperm were removed from the analyses, there was still no significant relationship between macro attachment duration and spermatheca size for the micros absent group (r2 ¼ 0.049, N ¼ 15, P ¼ 0.42). To assess the effect of micro donation on sperm transfer into the female, we compared the amount of sperm that drained from the macro ampulla between the groups. Micros present females had significantly larger sperm drainage areas in the macro ampulla than in the micros absent group (t test with equal variance: t43 ¼ 4.62, P < 0.0001; Table 1). We also compared the number of females that failed to receive any sperm between the two groups. Sperm transfer failure was identified by a full macro in combination with an empty spermatheca. All females in the micros present group received sperm (0/24 failures). However, 31.8% (7/22) of females in the micros absent group failed to take up sperm, constituting significantly more sperm transfer failures than the micros present group (Fisher’s exact test: P ¼ 0.0038). In addition to the large proportion of females that failed to take up sperm, females that received a nonzero amount of sperm in the micros absent group had a smaller sperm drainage area, on average, compared to females in the micro present group (t test with equal variance: t37 ¼ 3.18, P ¼ 0.003; Table 1).
To determine whether spermatheca size is affected by the amount of sperm drained from the ampulla, we assessed the relationship between the sperm drainage area in the ampulla and spermatheca area. Spermatheca area increased significantly with sperm drainage area (linear regression: r2 ¼ 0.244, N ¼ 37, P ¼ 0.0019; Fig. 2). To assess the effect that micro donation had on sperm uptake by the female reproductive tract, we compared the spermatheca size between the groups. Females had significantly larger spermatheca areas in the micros present group than in the micros absent group (ANCOVA: F2,40 ¼ 4.27, P ¼ 0.021; treatment group: P ¼ 0.038; macro attachment duration as a covariate: P ¼ 0.013; Table 1). However, this relationship did not hold when females that did not receive sperm in the micros absent group were removed from the analysis (ANCOVA: F2,35 ¼ 2.43, P ¼ 0.10; group: P ¼ 0.27; macro attachment duration: P ¼ 0.035; Table 1), or when macro attachment duration was not used as a covariate (t test with equal variance: t41 ¼ 1.226, P ¼ 0.23). To evaluate whether micro substances add a significant amount of material to the female sperm storage organ, we compared spermatheca size between the virgin and micro only groups. There was no significant difference in spermatheca area between virgins and females that received only micros (t test with equal variance: t27 ¼ 1.05, P ¼ 0.30; Table 1). DISCUSSION During natural courtship, males of the Hawaiian swordtail cricket L. cerasina provide several specialized spermless microspermatophores (micros) well in advance of mating (i.e. the transfer of a sperm-containing macrospermatophore; Shaw & Khine 2004). We conducted experiments to determine whether the donation of micros improves ejaculate transfer or mating success. We consider the effects of micro donation in the broad sense,
1.4
Spermatheca area (mm2)
Experiment 2: female remating and oviposition interval Micro donation did not significantly affect female remating behaviour. Most females remated (95.2%; 60/63). A large proportion of females in each group remated on the subsequent day, but there was no significant difference between the two groups (chisquare test: c259 ¼ 2.61, P ¼ 0.11; micros present ¼ 48.4% females mated (15/31) and micros absent ¼ 69.0% females mated (20/29)). There was no significant difference between groups in the number of days to accept a micro (randomization test: P ¼ 0.35; Table 1) or the macro (randomization test: P ¼ 0.32; Table 1) in remating trials. There was also no significant difference between the groups in the number of females that laid eggs before remating (chi-square test: c257 ¼ 1.68, P ¼ 0.19; micros present: 27.6% oviposited (8/29); micros absent: 13.8% oviposited (4/29)) or in the number of days to oviposition (t test with equal variance: t ¼ 1.141, P ¼ 0.26; micros present: 10.63 2.16, N ¼ 23; micros absent: 14.25 9.71, N ¼ 18).
1.2 1 0.8 0.6 0.4 0.2 0
0.1 0.2 0.3 0.4 Macro sperm drainage area (mm2)
0.5
Figure 2. Relation between macro sperm drainage area and spermatheca area.
824
T.N. deCarvalho, K.L. Shaw / Animal Behaviour 79 (2010) 819–826
because multiple behavioural elements are involved in their donation during protracted courtship. Other male crickets that donate nuptial gifts typically do so while in copula, a behaviour that has been shown to increase ejaculate transfer (Gwynne 1997; Vahed 1998), which in turn improves sperm competitive ability. Our observations suggest that L. cerasina males are likely to experience sperm competition as almost all females remated in our first experiment, most within 1–2 days after the first mating. In addition, we found that females usually remated before oviposition. Thus, males that increase the quantity of sperm transferred to the female sperm storage organ could potentially improve their fertilization success when their sperm is combined with another male’s prior to oviposition. Results from our third experiment clearly show that protracted courtship increased the amount of ejaculate transferred to the female from the sperm-filled macrospermatophore. Although we could not count sperm directly, we compared the amount of ejaculate that drained from the macro between the treatment groups. We observed that females that received micros had macros that were more fully drained of sperm, on average, than females that did not receive micros. Females that received micros prior to mating also had larger spermathecae. Together, a positive correlation between macro sperm drainage area and spermatheca area (Fig. 2) supports the conclusion that males that engaged in protracted courtship achieved greater ejaculate transfer with their mating partners. By dissecting spermathecae, we verified that the presence of sperm cells contributed to variation in spermatheca size. Furthermore, because we found no significant difference in spermatheca size between virgin females and females that received only micros, we can rule out that micros, by themselves, contributed significant volume to the spermatheca. Therefore, we conclude that courtship had a positive effect on the deposition of sperm from the macro into the female spermatheca. In contrast to the potential for micro donation to increase ejaculate transfer, we found that micros were not only unnecessary for females to mate, but that their donation did not increase mating success. This somewhat surprising result may indicate that females are indifferent to micros, or that females are unable to tell whether a male is transferring a micro or a macro. In other taxa that provide food gifts prior to mating, females often reject males that fail to provide a gift, such as sagebrush cricket males that show reduced mating success when females are not able to feed on specialized wing structures prior to spermatophore transfer (Eggert & Sakaluk 1994). Furthermore, females of other insect taxa also reject males that offer gifts of poor quality, as in the hanging fly Bittacus apicalis, in which females discriminate against males that offer relatively small or unpalatable prey as nuptial gifts (Thornhill 1976). Because Laupala females mated with males regardless of whether nuptial gifts were provided, this suggests that females do not exert premating choice based on variation in micro characteristics such as number or transfer rate. While our results show that micro donation enhanced sperm transfer, apparently it does not alter female behaviour via mechanisms observed in other crickets. First, micro donation did not delay female macrospermatophore removal as nuptial gifts have been shown to do in other Ensiferan taxa (Gwynne 1997). In two independent experiments, we found no significant difference between females that received micros and those that did not in the latency to the female’s first attempt to remove the macro or to completely remove the macro. Thus, the significant difference in the average amount of sperm transferred between the two groups cannot be explained by different macro attachment times. Second, remating and oviposition behaviour did not differ between treatment groups in our first experiment. Micro donation did not significantly lengthen the female latency to remating, suggesting that micros
lack receptivity-inhibiting substances, as has been found in other insect ejaculates (reviewed in Gillott 2003). Even without receptivity-inhibiting substances, females might be expected to remate sooner if they receive a smaller ejaculate. Although we did not measure sperm transfer in our second experiment, females probably received more sperm in the micros present treatment because of the enhanced effects of micro transfer. Even so, we found no significant difference in the remating period between treatments. Micro transfer also did not increase the occurrence of oviposition before remating. In this study, most females remated before laying eggs; however, there was no significant difference between the two treatment groups in the number of females that did lay eggs before remating or in the onset of oviposition. This further suggests that micro donation does not benefit males by manipulating female reproductive behaviours. Our ability to observe a difference between treatment groups may have been constrained by sample size, and similarly, the variation between groups in the amount of ejaculate transferred may have obscured a robust pattern. In Laupala, nuptial gift donation could enhance sperm transport in a number of ways. In many insects, sperm are transported to the storage organ by properties of the female reproductive tract, which includes peristalsis, chemotaxis and fluid reabsorption (reviewed in Bloch Qazi et al. 1998). In crickets, sperm is evacuated from the spermatophore via the pressure bodies located in a posterior compartment of the ampulla (Khalifa 1949). How sperm are transported through the spermathecal duct has not been well studied, however there is indirect evidence that peristaltic contractions of the reproductive tract are involved in some species (Eberhard 1996). In Laupala, the female reproductive tract may be responding to chemical or physical properties of nuptial gifts or the process of their transfer. For example, gifts may contain bioactive molecules that stimulate a peristaltic response. Mechanical stimulation of the female reproductive tract could be another way that nuptial gift donation directly affects the reproductive tract. Alternatively, the perception of nuptial gift properties, such as taste or number, may elicit an indirect response of the female reproductive tract. Furthermore, some other act of the courtship process, such as antennation, may be indirectly stimulating female-mediated sperm transport. Finally, micro donation may not act to induce femalemediated transport but rather result in the transfer of substances that aid in sperm motility. Sperm transfer enhancement by nuptial gift donation could either represent a new form of male manipulation or female choice. In recent work by Kullmann & Sauer (2009) on the Caucasion scorpionfly, Panorpa similis, gift-giving males were found to benefit by increased sperm transfer compared to those that did not provide gifts, independent of copulation duration. These results contrast with those for other scorpionflies, in which the linear relationship between copulation duration and sperm transfer results in copulation duration determining male insemination success (e.g. Engqvist & Sauer 2003). Females appear to exert control over copulation duration in other Panorpa species (e.g. Sauer et al. 1998); however, it is unknown whether P. similis females have an internal form of control over sperm transfer rate or whether males differ in ejaculate properties. In crickets, investigation of female control over insemination has focused largely on spermatophore removal behaviour, which is the functional equivalent of copulation duration in insects with internal fertilization. Most studies of ejaculate transfer in crickets show a linear relationship between sperm transfer and time (Sakaluk 1984, 2000; Simmons 1986). Thus, females can directly control the quantity of ejaculate transfer by timing the removal of the spermatophore. However, based on Simmons’s (1986) G. bimaculatus work, in which sperm transfer rate varied among individuals, Eberhard (1996) put forward the idea that female crickets may also have a form of internal control.
T.N. deCarvalho, K.L. Shaw / Animal Behaviour 79 (2010) 819–826
Our observations from Laupala suggest that sperm flow from the macro may not occur at a constant rate. In some instances, sperm appeared to drain rapidly from the ampulla into the female and then stop abruptly, suggesting that females may actively transport a certain amount of sperm. The idea that micros may activate female-mediated transport is supported by our observation that all females in the micros present group received some sperm, while approximately one-third of the micros absent females received no sperm. Internal control over ejaculate transfer would represent another mechanism of cryptic female choice in crickets that remains to be explored. The field component of our study allowed us to interpret our observations in the context of natural populations. We found that most courtships began in late morning, approximately 9 h before dark. We set up our laboratory experiments 8.5 h before the dark phase, resulting in no significant difference in mean micro number between field and experimental matings. In the field, we most likely observed the higher end of the natural range of micro number because of our method of scanning focal males in the early morning until a female arrived. However, the field data do not represent the upper limit of micro production. Males are able to produce more micros in the laboratory, and micro number is a function of the time of day that males begin courtship (T. N. deCarvalho, D. J. Fergus & K. L. Shaw, unpublished data). In the laboratory, males and females were artificially paired earlier than onset of pairing in the field; therefore, the larger quantity of micros produced in the laboratory was probably due to timing rather than to better male condition. In this study, we did not explore the effects of natural variation in micro number/courtship duration on sperm transfer, and we probably observed an extreme difference in sperm transfer response between the two groups. Therefore, the relationship between the time of pair formation, micro number and insemination success will be an important avenue to explore in future studies. In addition to sperm transfer enhancement, the restricted pattern of macro formation and transfer suggests that serial micro transfer may also serve as a form of mate guarding (Shaw & Khine 2004). While precopulatory guarding has not been well studied in crickets, recent work has shown that field cricket, Gryllus bimaculatus, males engage females with mate-guarding behaviours when they do not have a pre-made spermatophore and need time to prepare one (Parker & Vahed, in press). In Laupala, males may benefit by occupying females until an optimal time for sperm transfer, which may correlate with a narrow window of female receptivity (e.g. crustaceans; Jormalainen 1998) or fertilization, if oviposition occurs at night or early the next day (e.g. butterflies; Svard & Wiklund 1988). In L. cerasina, females do not appear to be restricted in their receptivity (T. N. deCarvalho, D. J. Fergus & K. L. Shaw, unpublished data); therefore, it is more likely that males are guarding females from other males prior to oviposition, although oviposition rhythmicity is yet unknown for this species. In conclusion, the possibility that greater sperm uptake by the female reproductive tract is due to micro donation would be a relatively novel function for nuptial gifts in insects. While it appears that serial micro donation served to enhance insemination rather than to ‘protect’ the ejaculate, the timing of micro donation did not predict the adaptive significance of these nuptial gifts because they were presented well in advance of mating. Micro donation appears to be a sperm competition adaptation that acts to increase the numerical representation of sperm from a given male. It remains to be investigated which aspect of micro donation (e.g. gustatory, chemical or tactile) affects sperm uptake and whether the effect of micro donation on sperm transfer is male or female mediated. Also, other courtship variables, such as antennation or familiarity, should be tested independently to determine their
825
potential contribution to the observed phenomenon. Furthermore, the elaborate nuptial gift system of Laupala may represent a new form of male manipulation of female reproduction or cryptic female choice.
Acknowledgments We are grateful to Brian Coyle for assistance in cricket collection and comments on the manuscript. We also thank members of the Shaw lab and T.N.D.’s dissertation committee, Charles Mitter, Kaci Thompson, Barbara Thorne and Jerry Wilkinson for their intellectual contribution to this work. This work was financially supported by an Orthopterist’s Society Grant and an Animal Behavior Society Student Research Grant to T.N.D. and by a National Science Foundation award 344789 to K.L.S.
References Bloch Qazi, M. C., Aprille, J. R. & Lewis, S. M. 1998. Female role in sperm storage in the red flour beetle, Tribolium castaneum. Comparative Biochemistry and Physiology Part A Molecular & Integrative Physiology, 120, 641–647. Brown, W. D. & Gwynne, D. T. 1997. Evolution of mating in crickets, katydids and wetas (Ensifera). In: The Bionomics of Grasshoppers, Katydids and Their Kin (Ed. by S. K. Gangwere & M. C. Muralirangan), pp. 281–314. Wallingford: CAB International. deCarvalho, T. N. & Shaw, K. L. 2005. Nuptial feeding of spermless spermatophores in the Hawaiian swordtail cricket, Laupala pacifica (Gryllidae: Trigonidiinae). Naturwissenschaften, 92, 483–487. Dodson, G. N., Morris, G. K. & Gwynne, D. T. 1983. Mating behavior of the primitive orthopteran genus Cyphoderris (Haglidae). In: Orthopteran Mating Systems: Sexual Competition in a Diverse Group of Insects (Ed. by D. T. Gwynne & G. K. Morris), pp. 305–318. Boulder, Colorado: Westview Press. Eberhard, W. G. 1996. Female Control: Sexual Selection by Cryptic Female Choice. Princeton, New Jersey: Princeton University Press. Eggert, A. K. & Sakaluk, S. K. 1994. Sexual cannibalism and its relation to male mating success in sagebrush crickets, Cyphoderris strepitans (Haglidae, Orthoptera). Animal Behaviour, 47, 1171–1177. Engqvist, L. & Sauer, K. P. 2003. Determinants of sperm transfer in the scorpionfly Panorpa cognata: male variation, female condition and copulation duration. Journal of Evolutionary Biology, 16, 1196–1204. Gillott, C. 2003. Male accessory gland secretions: modulators of female reproductive physiology and behavior. Annual Review of Entomology, 48, 163–184. Gwynne, D. T. 1995. A phylogeny of Ensifera (Orthoptera): a hypothesis supporting multiple origins of acoustical signaling, complex spermatophores and maternal care in crickets, katydids and weta. Journal of Orthoptera Research, 4, 203–218. Gwynne, D. T. 1997. The evolution of edible ‘sperm sacs’ and other forms of courtship feeding in crickets, katydids and their kin (Orthoptera: Ensifera). In: The Evolution of Mating Systems in Insects and Arachnids (Ed. by J. A. Choe & B. J. Crespi), pp. 110–139. Cambridge: Cambridge University Press. Jormalainen, V. 1998. Precopulatory mate guarding in crustaceans: male competitive strategy and intersexual conflict. Quarterly Review of Biology, 73, 275–304. Khalifa, A. 1949. The Mechanism of insemination and the mode of action of the spermatophore in Gryllus domesticus. Quarterly Journal of Microscopical Science, 90, 281–292. Kullmann, H. & Sauer, K. P. 2009. Mating tactic dependent sperm transfer rates in Panorpa similis (Mecoptera; Panorpidae): a case of female control? Ecological Entomology, 34, 153–157. Loher, W. & Dambach, M. 1989. Reproductive behavior. In: Cricket Behavior and Neurobiology (Ed. by F. Huber, T. E. Moore & W. Loher), pp. 43–82. Ithaca, New York: Cornell University Press. Murtaugh, M. P. & Denlinger, D. L. 1985. Physiological regulation of long-term oviposition in the house cricket, Acheta domesticus. Journal of Insect Physiology, 31, 611–617. Parker, D.J. & Vahed, K. in press. The intensity of pre- and post-copulatory mate guarding in relation to spermatophore transfer in the cricket Gryllus bimaculatus. Journal of Ethology. doi:10.1007/s10164-009-0176-6. Parker, G. A. 1970. Sperm competition and its evolutionary consequences in the insects. Biological Reviews, 45, 525–567. Sakaluk, S. K. 1984. Male crickets feed females to ensure complete sperm transfer. Science, 223, 609–610. Sakaluk, S. K. 1986. Sperm competition and the evolution of nuptial feedingbehavior in the cricket, Gryllodes supplicans (Walker). Evolution, 40, 584–593. Sakaluk, S. K. 2000. Sensory exploitation as an evolutionary origin to nuptial food gifts in insects. Proceedings of the Royal Society B, 267, 339–343.
826
T.N. deCarvalho, K.L. Shaw / Animal Behaviour 79 (2010) 819–826
Sakaluk, S. K. & Eggert, A. K. 1996. Female control of sperm transfer and intraspecific variation in sperm precedence: antecedents to the evolution of a courtship food gift. Evolution, 50, 694–703. Sauer, K. P., Lubjuhn, T., Sindern, J., Kullmann, H., Kurtz, J., Epplen, C. & Epplen, J. T. 1998. Mating system and sexual selection in the scorpionfly Panorpa vulgaris (Mecoptera: Panorpidae). Naturwissenschaften, 85, 219–228. Shaw, K. L. & Khine, A. H. 2004. Courtship behavior of the Hawaiian cricket Laupala cerasina: males provide spermless spermatophores as nuptial gifts. Ethology, 110, 81–95. Simmons, L. W. 1986. Female choice in the field cricket Gryllus bimaculatus (De Geer). Animal Behaviour, 34, 1463–1470. Simmons, L. W.1987. Sperm competition as a mechanism of female choice in the field cricket, Gryllus bimaculatus. Behavioral Ecology and Sociobiology, 21, 197–202. Stanley-Samuelson, D. W. & Loher, W. 1986. Prostaglandins in insect reproduction. Annals of the Entomological Society of America, 79, 841–853.
Stanley-Samuelson, D. W., Peloquin, J. J. & Loher, W. 1986. Egg-laying in response to prostaglandin injections in the Australian field cricket, Teleogryllus commodus. Physiological Entomology, 11, 213–219. Svard, L. & Wiklund, C. 1988. Prolonged mating in the monarch butterfly Danaus plexippus and nightfall as a cue for sperm transfer. Oikos, 51, 351–354. Thornhill, R. 1976. Sexual selection and nuptial feeding behavior in Bittacus apicalis (Insecta: Mecoptera). American Naturalist, 110, 529–548. Vahed, K. 1998. The function of nuptial feeding in insects: review of empirical studies. Biological Reviews of the Cambridge Philosophical Society, 73, 43–78. Wedell, N. 1991. Sperm competition selects for nuptial feeding in a bushcricket. Evolution, 45, 1975–1978. Wedell, N. & Arak, A. 1989. The wartbiter spermatophore and its effect on female reproductive output (Orthoptera, Tettigoniidae, Decticus verrucivorus). Behavioral Ecology and Sociobiology, 24, 117–125.
