Copyright Ó 2007 by the Genetics Society of America DOI: 10.1534/genetics.106.064949

Genetic Architecture of Conspecific Sperm Precedence in fasciatus and A. socius

Seth C. Britch,*,1 Emma J. Swartout,* Daniel D. Hampton,† Michael L. Draney,‡ Jiming Chu,§ Jeremy L. Marshall** and Daniel J. Howard* *Department of Biology, New Mexico State University, Las Cruces, New Mexico 88003, †Duke University School of Medicine, Durham, North Carolina 27706, ‡Department of Natural and Applied Sciences, University of Wisconsin, Green Bay, Wisconsin 54311, §Health Occupations Program, Dona Ana Branch Community College, Las Cruces, New Mexico 88003 and **Department of Entomology, Kansas State University, Manhattan, Kansas 66506 Manuscript received August 17, 2006 Accepted for publication April 1, 2007

ABSTRACT The evolution of barriers to gene exchange is centrally important to speciation. We used the crickets and A. socius to investigate the genetic architecture of conspecific sperm precedence (CSP), a postinsemination prezygotic reproductive barrier. With amplified fragment-length polymor- phism (AFLP) markers and controlled crosses we constructed linkage maps and estimated positions of QTL associated with CSP. The majority of QTL have low to moderate effects, although a few QTL exist in A. socius with large effects, and the numbers of QTL are comparable to numbers of genes accounting for species differences in other studies. The QTL are spread across many unlinked markers, yet QTL placed with linked markers are on a small number of linkage groups that could reflect the role of the large Allonemobius sex chromosome in prezygotic isolation. Although many QTL had positive effects on conspecific sperm utilization several QTL also exerted negative effects, which could be explained by intraspecific sexual conflict, sperm competition, or epistasis of introgressed genes on novel backgrounds. One unexpected outcome was that A. socius CSP alleles have a stronger effect than those from A. fasciatus in hybrid females, causing hybrids to behave like A. socius with regard to sperm utilization. Implications of this asymmetry in the Allonemobius hybrid zone are discussed.

PECIATION research is currently one of the most lization of sperm from conspecific males in fertilization S dynamic fields in modern biology. As noted by when both conspecific and heterospecific males have Coyne and Orr (2004), there has been more activity in inseminated a female’’ (Howard 1999, pp. 110–111). this area of scientific endeavor over the course of the The precedence may occur because conspecific sperm past 20 years than in the previous 125 years. One in- outcompete fertilization-competent heterospecific sperm teresting characteristic of this burst of activity is that or because of postinsemination incompatibilities be- most current studies of speciation focus on the evo- tween heterospecific males and females (i.e., non- lution and genetics of reproductive barriers (Coyne competitive gametic isolation). Consequently, there are and Orr 2004). This focus on reproductive barriers is many mechanisms that can underlie such heterospe- easy to understand. The evolution of barriers to gene cific disadvantages. exchange irrevocably separates two lineages and assures Although barriers to fertilization operating at the their future independence. Hence, the evolution of level of sperm and egg have long been recognized as these barriers is of central importance in every spe- important in the reproductive isolation of broadcast- ciation event among sexually reproducing organisms. spawning marine invertebrates (Loeb 1915; Lillie Enough work has now been done on reproductive 1921), the importance of postinsemination barriers to barriers that a number of clear patterns have begun to fertilization in terrestrial did not become ap- emerge; one of the clearest is the important role played parent until the 1990s. It was not until this period of by conspecific sperm precedence in the reproductive time that gamete competition studies were regularly isolation of closely related species (Howard incorporated into investigations of reproductive iso- 1999; Simmons 2001; Coyne and Orr 2004). Conspe- lation. As a result of these analyses, we now recognize cific sperm precedence is defined as ‘‘the favored uti- that conspecific sperm precedence isolates closely re- lated species in groups as divergent as vertebrates and (Howard and Gregory 1993; Gregory and Howard 1994; Wade et al. 1994; Price 1997; Howard 1Corresponding author: USDA-ARS, Center for Medical, Agricultural, and rice rown ady immons Veterinary Entomology, 1600/1700 SW 23rd Dr., Gainesville, FL 32608. 1999; P et al. 1999; B and E 2001; S E-mail: [email protected] 2001; Fricke and Arnqvist 2004).

Genetics 176: 1209–1222 ( June 2007) 1210 S. C. Britch et al.

An group in which conspecific sperm pre- tion we developed AFLP markers (see below) unique to each cedence (CSP) has been particularly well studied is the species by tracing AFLP fragments through two generations ground genus Allonemobius. Detailed studies (F1 and F2 backcross) that were absent in individuals of the enedix oward species to which the backcrossing was done (Figure 2). over the course of many years (B and H The QTL trait of interest, CSP, was measured by documenting 1991; Howard and Gregory 1993; Howard et al. 1993; the frequency with which males of the species of interest (A. Gregory and Howard 1994; Doherty and Howard fasciatus when the introgressed genes in the F2-backcross 1996; Gregory et al. 1998; Howard et al. 1998a,b; female were from A. fasciatus and A. socius when the intro- Britch et al. 2001) have demonstrated that the strong, gressed genes were from A. socius) produced offspring when an F2-backcross female was mated once each to an A. fasciatus but incomplete reproductive isolation between the male and to an A. socius male (Figure 3). Prior studies of sperm closely related species Allonemobius fasciatus and A. socius precedence among parental types demonstrated that order of is due to a single type of reproductive barrier—CSP. The matings has no significant effect on sperm utilization patterns simplicity of the system, a single barrier to gene exchange (Howard and Gregory 1993; Gregory and Howard 1994); isolating two closely related species, is extremely rare thus, for the sake of simplicity F2-backcross females in both experiments were mated first to A. fasciatus males and second among species pairs that have been thoroughly investi- to A. socius males. All matings were closely observed to ensure gated (Coyne and Orr 2004). In this case, should we that only a single spermatophore was transferred to the female achieve an understanding of the genetic control of CSP, by each male. A period of 24–48 hr was imposed between first we will have achieved an understanding of the genetic and second matings. changes that have given rise to new species. Females that successfully mated to both males were placed in individual cages and left to oviposit in both soil and cotton Here we report in detail the results of QTL studies of media for 2 weeks, after which they were frozen at 80°. Soil CSP in A. fasciatus and A. socius, preliminary results of medium is provided specifically for oviposition, but females which were published earlier (Howard et al. 2002). Al- also oviposit in approximately equal frequency (S. C. Britch, though several studies have looked at other isolating unpublished data) in water-soaked cotton provided for dietary mechanisms through QTL analysis (Bradshaw et al. moisture. Oviposition media were gradually cooled and ex- posed to an artificial overwintering period of 3 months in a 4° 1998; MacDonald and Goldstein 1999; Fishman et al. ao urnberger constant-temperature room. After overwintering, oviposition 2002; T et al. 2003; N et al. 2003), this study media were gradually warmed to room temperature and is among the first to document QTL for conspecific nymphs were left to emerge in individual family cages. When sperm precedence (see also Civetta et al. 2002). The nymphs reached second or third instar they were frozen en QTL approach allows us to estimate the number and masse at 80° to await paternity analysis, which was accom- plished with allozyme phenotyping. The resulting data were location of genetic factors responsible for a difference used to determine the pattern of sperm utilization by each F2- between two species in a trait, as well as the magnitude of backcross female, in particular the frequency with which each the effect of each QTL. male fathered offspring. AFLP typing of parents, males, F1 females, and F2-backcross females was done post hoc, since many females were expected MATERIALS AND METHODS to either not complete a second mating or not mate at all; similarly, many males will not mate in the laboratory. Following Although preliminary QTL analyses and a linkage map for restriction ligations of cricket genomic DNA, extracted using A. fasciatus using amplified fragment-length polymorphisms QIAGEN (Valencia, CA) DNEasy kits (no. 69504), we used the (AFLPs) (Vos et al. 1995) were previously described (Howard ABI Regular Genome mapping kit (no. 4303050) to do pre- et al. 2002), since the publication of those results, the Howard selective and selective PCR amplification of samples, which Lab has switched from an ABI 377 to an ABI 3100 automated were then run on the ABI 3100 sequencer. A prior survey ( J. L. sequencer. Given the ease and speed of analysis of the ABI Marshall, unpublished data) identified five combinations of 3100 and some discrepancies in fragment sizes, we reanalyzed selective amplification primers that consistently produced all individuals in the A. fasciatus mapping population, and $70 AFLP fragments per individual. We code these combina- we analyzed individuals from an A. socius mapping popula- tions here as B5, B6, G3, G7, and H5, but primer names as well tion. In both cases, we used five primer combinations to create as all molecular protocols are available from the correspond- AFLP linkage maps and performed QTL analyses to document ing author. Unique fragments were traced using the rules linked or unlinked single markers or groups of markers outlined in Figure 2 and tabulated by F2-backcross female in a eer strongly associated with CSP in F2-backcross females. spreadsheet format suitable for Map Manager Qtxb19 (M The QTL experiments for both species were similar in et al. 2002). Linkage groups were created in Map Manager at protocol and are described here in brief. Field-caught crickets the P ¼ 0.001 level using the Kosambi map function, which from three focal populations in the East Coast (EC) transect assumes intermediate interference. At P ¼ 0.001 the threshold (Figure 1), EC 49 and EC 60A (A. fasciatus, north of mixed of at least eight linkage groups was found in both data sets, the populations), and EC 65 (A. socius, south of mixed popula- known haploid chromosome number in Allonemobius (Lim tions), were brought to the lab and screened for species 1971). A chi-square test showed that several AFLP markers identity using allozymes (Howard 1983, 1986; Howard and from both species showed patterns of segregation distortion, Furth 1986). Screening was necessary due to the presence in so we activated the feature accounting for segregation distor- both populations of introgressed individuals and cryptic con- tion in Map Manager. Once linkage groups were established geners in low frequencies. Following the breeding design we used the Ripple command, which refines the order of shown in Figure 2 we hybridized field-caught A. fasciatus males markers on all linkage groups by testing local permutations of eer and A. socius females and backcrossed F1 females with males the order (M et al. 2002). As a result of the breeding design from first-generation lab-reared parental lines to produce F2- (Figure 2) F2-backcross females are heterozygous for all backcross females. For linkage mapping and QTL identifica- candidate markers, and so, despite dominance of AFLP Conspecific Sperm Precedence 1211

Figure 1.—The inset shows a map of North America with the approximate ranges of A. fasciatus and A. socius. The dark pat- terned area is the estimated extent of the zone of overlap and hybridization between the species. The main map shows the eastern United States, the transects through the A. fasciatus–A. socius hybrid zone, and the populations used in this study. The Illinois (IL) transect runs through Illinois, the Mountain (MTN) transect spans West Virginia, Virginia, and North Carolina, and the East Coast (EC) transect runs from New York along the eastern seaboard to North Carolina. A. fasciatus populations are represented by solid circles, A. socius populations by open circles, and mixed populations by half-open/half-solid circles. Differential adaptation to temperature underlies the mosaic pattern, causing the zone to widen across the Appalachian Mountains where A. fasciatus and mixed populations at cooler high elevations may occur south of warmer low-elevation A. socius populations. The three pop- ulations from which individuals were collected for this study are in New Jersey and include two A. fasciatus populations, EC 49 and EC 60A, and one A. socius population, EC 65. fragments, data were treated as ‘‘codominant backcross’’ in group (we also used quick test LRS values to filter results of Map Manager’s linkage evaluation. single-marker regression analysis). As a general guideline for On the basis of trials using several standard transformations interpreting LRS scores, the underlying P-value for the LRS in Mapmaker/QTL 1.1b (Lincoln et al. 1993) we determined must be ,0.0001 at one point on a linkage group for a ge- that square-root transformation of CSP trait data for both nomewide P-value of 0.05, and the equivalent LRS at one point species was the closest approximation to the normal distribu- on a linkage group is 15 for a backcross pedigree design tion. Raw trait data were then imported into Map Manager and (Meer et al. 2002). For readers accustomed to the LOD score, square-root transformed. Before QTL analyses we had Map LOD ¼ LRS/4.6; thus the LRS of 15 is roughly equivalent to Manager hide redundant loci, i.e., markers mapping to a loca- the critical LOD of 3.0 often cited in QTL studies. However, tion already occupied (refer to results and Figures 4 and 5). the quick test adjusts the LRS score on the basis of the ex- We performed single-marker regression analysis for QTL in perimental data. Map Manager under the additive regression model at the least Interval mapping in Map Manager produces a graph of the restrictive significance level of P ¼ 0.05. At P ¼ 0.05 we were LRS for the trait along each linkage group as well as the ad- able to group significant single-marker associations manually ditive regression coefficient, which is positive if the presence of instead of running several separate analyses at the five more the marker tends to increase the trait value and negative if it restrictive levels of significance available in the program. tends to decrease it. The graph is marked with the three LRS Before interval QTL mapping we used the quick test to thresholds calculated in the quick test. A table accompanies establish critical likelihood-ratio statistics (LRS). LRS are em- each graph with values at 1-cM intervals for the LRS and re- pirical significance level estimates across all linkage groups, gression coefficient, as well as the percentage of the total trait built by randomly permuting markers and phenotypes (Hartl variance explained by a QTL at that location. We activated the and Clark 1997) for the transformed trait values. This feature bootstrap test as a further means of localizing and assessing generated three values, suggestive, significant, and highly significant trait–marker associations on each linkage group. significant, with which we gauged the significance of the LRS Bootstrap LRS values are displayed as a histogram on the produced by interval mapping of the trait on each linkage graph, the width of which shows the confidence interval in 1212 S. C. Britch et al.

igure F 2.—Design of pedigreed matings between A. fasciatus and A. socius for the breeding of F2-backcross females for use in both QTL experiments. Not shown are matings using field-caught crickets to produce parental lines of each species in both experi- ments. Parental lines are extended through two generations to provide males for mating trials with F2-backcross females (Figure 3). In both experiments marker M signifies an AFLP fragment present in the focal species of the parental generation that is never present in the opposite species. F2-backcross females are scored as ‘‘present’’ (Mm) or ‘‘absent’’ (mm) for a suite of AFLP frag- ments unique to the focal species. Owing to the pedigree design the AFLP fragment is heterozygous in all instances of present in the F2 and can be treated as a codominant marker in QTL analyses. The relationship of marker M to the CSP trait is thus attribut- able to the effect of the focal species’ genome. centimorgans for the QTL on the linkage group. There is no (17 F1 families) for the A. socius map. Additionally, 13 F2- accompanying table for bootstrap values. For each linkage backcross males to A. fasciatus and 20 to A. socius were group we ran two blocks of composite-interval mapping included in the mapping data to increase the resolution analyses that control for effects of linked and unlinked markers and reduce ambiguity of adjacent QTL. The first of AFLP linkage maps. Lineages are based on a unique block controlled for genomewide background QTL with LRS parental hybrid mating (Figure 2). Over 1000 AFLP $ 15 that had been identified with single-marker regression. fragments across five primer combinations were scored The second block controlled for effects of adjacent QTL on for the two A. fasciatus parents of which 116 were not the same linkage group. In the second block we limited found in A. socius individuals used to produce F back- analyses to linkage groups containing more than two markers 2 (not including redundant markers) and controlled for each crosses in the A. fasciatus experiment. Of .2500 frag- marker, in turn, as a background effect. We updated the ments scored for the five A. socius parents, 275 were location of a QTL only if controlling for a background QTL not found in A. fasciatus individuals used to produce F2 increased the actual (not bootstrapped) LRS at a marker or backcrosses in the A. socius experiment. Except for rare intermarker space on a linkage group; however, bootstrap instances, all AFLP primer combinations yielded $70 values higher than significant were taken into account in borderline cases. fragments in all individuals. Linkage groups created at the P ¼ 0.001 linkage criterion for A. fasciatus are shown in Figure 4 and those for A. socius in Figure 5. We RESULTS mapped 65 A. fasciatus markers to 13 linkage groups We produced 100 F2-backcross females from two pa- (four redundant markers), spanning 833.9 contiguous rental lineages (five F1 families) for the A. fasciatus map cM with an average distance between markers of 17.7 cM and 111 F2-backcross females from five parental lineages (range: 4.4–45.9 cM); and we mapped 39 A. socius

Figure 3.—Mating trial design for measuring the CSP trait. F2-backcross fe- males (Figure 2) are used in mating trials with males from parental lines of both species. Females are mated first to a male of A. fasciatus and must be mated to an A. socius male 24–48 hr later. Matings are observed di- rectly to ensure only one successful mating per male and only a single spermatophore transferred to the female. The CSP trait is simply the frequency of offspring sired by the con- specific male, assessed using allozyme phenotyping of the female, both males, and a random sample of not more than 30 offspring. Conspecific Sperm Precedence 1213

Figure 4.—Linkage groups of A. fasciatus. Linkage groups were calculated using Map Manager (Meer et al. 2002) at P ¼ 0.001 using the Kosam- bi map function and account- ing for segregation distortion. The linkage criterion of P ¼ 0.001 yielded the number of linkage groups closest to eight, the known haploid number of chromosomes in Allonemobius (Lim 1971). Markers are prefixed by the code for the primer combina- tion responsible for them (B5, B6, G3, G7, or H5), fol- lowed by the size of the AFLP fragment in base pairs. Mark- ers redundant for QTL detec- tion (i.e., markers mapping to a location already occu- pied) are included in both maps. We used MapChart 2.1 (Voorrips 2002) to draw maps of linkage groups.

markers to 10 linkage groups (8 redundant markers, in- F2-backcross females used to generate the A. fasciatus cluding 4 on 3 linkage groups made exclusively of mark- map, 83 had sufficient trait data for QTL analysis. ers mapping to the same location) spanning 361.1 cM The corresponding number for A. socius was 95 of 111 with an average distance of 17.2 cM (range: 4.8–41.4 cM). F2-backcross females. The trait data for A. fasciatus, the The remaining 51 markers in A. fasciatus and 236 mark- percentage of offspring sired by the A. fasciatus male, ers in A. socius were unlinked. The smaller number of were bimodally distributed with an average of 0.39 markers mapping to fewer linkage groups and the large (interquartile range, IQR 0.00–0.68). Trait data for A. number of unlinked loci in A. socius were probably due socius, the percentage of offspring sired by the A. socius to the large number of small families, sharing few mark- male, were also bimodal, with an average of 0.58 (IQR ers, used in that experiment. The Ripple command 0.15–1.00). The results of single-marker regression are affected only linkage group 2 of A. fasciatus—it reversed given for A. fasciatus in Table 1 and for A. socius in Table the order of G7.144.14 and G3.272.24 and gave rise to 2. Although several additional markers in both species some changes in map distances on the linkage group. were found at the P ¼ 0.05 level in single-marker However, interval QTL mapping (see below) on both regression we filtered results based on the quick test versions of A. fasciatus linkage group 2 did not produce experimentwide LRS levels, requiring a minimum LRS greatly different LRS graphs or bootstrap histograms; of 5.6 for A. socius markers and 6.8 for A. fasciatus therefore we retained the original marker order. markers. Quick test critical LRS values were comparable Although all 211 F2-backcross females mated with between species (Tables 3 and 4). conspecific and heterospecific males, not all produced Most significant QTL were associated with unlinked offspring, and in some cases offspring paternity was markers and significant QTL associated with linked not discernible due to degraded allozymes. Of the 100 markers were grouped onto a small number of linkage 1214 S. C. Britch et al.

Figure 5.—Linkage groups of A. socius. Linkage groups were cal- culated and drawn with the crite- ria described in Figure 4.

groups (Tables 1 and 2). As discussed in Howard et al. QTL (composite-interval mapping; Tables 3 and 4). In (2002) the relatively small size of the mapping pop- cases where a QTL is located with composite-interval ulations is expected to reduce the number of signifi- mapping, the marker used for background control is cant QTL, but the QTL detected will be of large effect noted. Every linkage group in both A. fasciatus and A. (Tanksley 1993; Bradshaw and Stettler 1995; socius analyzed with simple interval mapping displayed Bradshaw et al. 1995, 1998). The results of these ana- at least one location with a significant to highly sig- lyses reflect this observation. The percentages of trait nificant bootstrap value, yet rarely were these values variance explained by single-marker regression QTL associated with an LRS value above suggestive. We re- ranged from 9 to 17% in A. fasciatus (Table 1) and from port only QTL associated with LRS values that were at 6 to 29% in A. socius (Table 2). Although most markers least suggestive (Tables 3 and 4) and took high boot- in both species had a positive effect on conspecific strap values into account only to resolve two borderline sperm utilization, 4 of the 16 A. fasciatus markers and 4 cases, B5.190.43 and B6.069.43 in A. fasciatus (Table 3). of the 10 A. socius markers had a negative effect, mean- Generally, linked markers strongly associated with ing that the presence of the marker reduced the per- QTL in simple interval mapping also showed high LRS centage of offspring fathered by the conspecific male. values in single-marker analyses (Tables 1–4). One ex- Positive and negative effects are determined by the ception occurred in A. socius,wheretheintervalbetween average percentage of (conspecific) offspring column B5.304.15 and G3.324.66 on linkage group 1 had a strong in Tables 1 and 2. In the A. fasciatus mapping popula- positive effect on the trait variance (although supported tion, positive effects are assigned to markers associated only by a suggestive LRS; Table 4), yet the individual with an increase in fertilization by A. fasciatus males. In markers were not picked up by single-marker regression the A. socius mapping population, positive effects are analysis. assigned to markers associated with an increase in fer- In the first block of composite-interval mapping, tilization by A. socius males. In cases where the average single-marker regression had identified only one mark- percentages and/or IQRs are not notably different or er in A. fasciatus, B6.069.43 in linkage group 3 (Table 1), ambiguous (e.g., B6.315.40 in A. fasciatus, Table 1; with an LRS $ 15 appropriate for use as a background G7.255.32 or G7.260.15 in A. socius, Table 2), and to QTL. We detected no effect of this marker on any link- confirm all other cases, we use the sign of the additive age group, including linkage group 3. Single-marker regression coefficient (not shown). For B6.096.42 (Ta- regression revealed no markers with an LRS $ 15 in ble 1), since the sample size was so small for the number the A. socius data (Table 2). In the second block of of individuals possessing the fragment, we used the sign composite-interval mapping, in which we controlled for of the additive regression coefficient (negative). each marker across each linkage group, two markers in Tables of results for the interval mapping analyses for A. fasciatus were found to be associated with the trait. A. fasciatus and A. socius are divided into markers (or Only one of these, B6.069.43, had already been identi- marker intervals) associated with QTL with no control fied with single-marker regression (Table 3), and the for background QTL (interval mapping) and those other, B5.190.45, had a borderline suggestive LRS. associated with QTL while controlling for background Additionally, composite-interval mapping had the effect Conspecific Sperm Precedence 1215

TABLE 1

Single-marker QTL regression analysis results for A. fasciatus, calculated in Map Manager (MEER et al. 2002)

Average % A. fasciatus AFLP marker State offspring (IQR) n Effect (1/) LRSa % P Linkage group G3.136.69 Present 0.36 (0.02–0.84) 11 14.7 16 0.0001 Unlinked Absent 0.54 (0.23–1.00) 41 B6.240.97 Present 0.51 (0.33–0.68) 3 1 14.2 16 0.0002 Unlinked Absent 0.11 (0.00–0.06) 16 G7.238.46 Present 0.31 (0.05–0.48) 14 13.6 15 0.0002 Unlinked Absent 0.61 (0.32–1.00) 24 B6.088.93 Present 0.32 (0.00–0.62) 13 1 12.6 14 0.0004 Unlinked Absent 0.13 (0.00–0.17) 18 B6.225.37 Present 0.34 (0.14–0.53) 4 1 12.5 14 0.0004 Unlinked Absent 0.19 (0.00–0.25) 27 B6.098.51 Present 0.26 (0.03–0.43) 10 1 10.9 12 0.0010 Unlinked Absent 0.18 (0.00–0.20) 21 B6.158.92 Present 0.33 (0.00–0.65) 8 1 10.8 12 0.0010 Unlinked Absent 0.17 (0.00–0.21) 23 B5.180.40 Present 0.28 (0.00–0.62) 13 1 9.3 11 0.0022 Unlinked Absent 0.16 (0.00–0.17) 18 B6.315.40 Present 0.44 (0.08–1.00) 8 8.8 10 0.0030 Unlinked Absent 0.51 (0.09–1.00) 44 B5.157.42 Present 0.25 (0.00–0.33) 16 1 7.7 9 0.0056 Unlinked Absent 0.17 (0.00–0.13) 15 G7.101.62 Present 0.25 (0.00–0.51) 16 1 7.5 9 0.0062 Unlinked Absent 0.17 (0.00–0.16) 15 B6.096.42 Present 1.00 (1.00–1.00) 1 7.4 9 0.0067 Unlinked Absent 0.49 (0.10–1.00) 37 H5.311.13 Present 0.37 (0.05–0.67) 6 1 10.5 13 0.0012 Group 2 Absent 0.17 (0.00–0.22) 25 G3.475.55 Present 0.50 (0.17–0.86) 15 1 11.6 13 0.0007 Group 3 Absent 0.19 (0.00–0.28) 24 B6.069.43 Present 0.33 (0.12–0.63) 10 1 15.1 17 0.0001 Group 3 Absent 0.15 (0.00–0.19) 21 B5.193.24 Present 0.31 (0.00–0.64) 8 1 11.8 13 0.0006 Group 3 Absent 0.17 (0.00–0.19) 23 Both positive (‘‘Effect 1’’; enhancing fertilization by A. fasciatus sperm) and negative (‘‘Effect ’’; reducing fertilization by A. fasciatus sperm) effects were found to be associated with linked and unlinked AFLP markers. The average percentage of A. fasciatus offspring produced when a given marker is either present or absent is followed by the interquartile range (IQR) (25th and 75th percentiles), which is used as a measure of variance. Averages and IQR are based on raw trait data, while LRS, P, and percentages are based on square-root transformations. LRS, likelihood-ratio statistic (see text); n, sample size; %, percentage of the trait var- iance explained by a QTL at this marker; P, P-value of the LRS. Refer to Figure 4 for the map of linkage groups. Original single- marker regression analysis was done at P ¼ 0.05, but markers were excluded that fell below the experimentwide LRS threshold of ‘‘suggestive.’’ a Critical LRS values are 6.8 (suggestive), 11.8 (significant), and 19.5 (highly significant). of reducing the LRS of B6.069.43 from 15.1, the most A. socius linkage group 4 that gave a highly significant significant LRS of the single-marker regression analyses, LRS of 24.6, the highest in any analysis of either ex- to the borderline suggestive value of 6.5. This was prob- periment, and explained 69% of the trait variance (posi- ably due to concerted background effects of the re- tive effect). Interestingly, B5.140.99 by itself had been maining markers on linkage group 3 (Figure 4). Overall, detected by single-marker regression, as well as by sim- the range of the percentage of trait variance explained ple interval analysis, but showed only a suggestive-to- by markers found with interval mapping and composite- significant LRS of 7.2, while still accounting for a very interval mapping in A. fasciatus was 9–19%, comparable large 29% of the trait variance (positive effect). But by to that with single-marker regression. controlling the effect of the nearby H5.292.14 as a Turning to A. socius, composite-interval mapping re- background QTL, a genetic factor between the two vealed a new marker, H5.107.37, with only weak LRS markers was permitted to show its full effect on the trait support but moderate negative effect on sperm utiliza- (Kearsey 1998). Even in the absence of this marker tion (Table 4). One of the most surprising results came interval the range of the percentage of trait variance from the interval between B5.140.99 and H5.292.14 on explained by markers found with interval mapping and 1216 S. C. Britch et al.

TABLE 2

Single-marker QTL regression analysis results for A. socius, calculated in Map Manager (MEER et al. 2002)

Average % A. socius AFLP marker State offspring (IQR) n Effect (1/) LRSa % P Linkage group B6.197.02 Present 0.67 (0.39–1.00) 6 1 10.4 13 0.00128 Unlinked Absent 0.00 (0.00–0.00) 4 B6.098.65 Present 0.76 (0.64–1.00) 3 1 8.7 12 0.00319 Unlinked Absent 0.22 (0.00–0.19) 8 B5.406.67 Present 0.69 (0.31–1.00) 5 7.5 8 0.00605 Unlinked Absent 0.82 (0.75–1.00) 17 G7.200.72 Present 0.68 (0.39–1.00) 6 1 6.2 6 0.01245 Unlinked Absent 0.08 (0.00–0.13) 4 B5.402.55 Present 0.49 (0.14–1.00) 5 5.9 6 0.01532 Unlinked Absent 0.78 (0.50–1.00) 17 G7.255.32 Present 0.86 (0.87–1.00) 7 5.8 7 0.01607 Unlinked Absent 0.79 (0.50–1.00) 17 G7.260.15 Present 0.90 (1.00–1.00) 5 5.7 7 0.01649 Unlinked Absent 0.77 (0.56–1.00) 18 B5.195.64 Present 0.77 (0.60–1.00) 35 1 8.2 14 0.00419 Group 4 Absent 0.44 (0.12–0.75) 19 B5.140.99 Present 0.85 (0.85–1.00) 13 1 7.2 29 0.00732 Group 4 Absent 0.46 (0.13–0.63) 8 H5.325.71 Present 0.86 (0.80–1.00) 17 1 6.1 17 0.01377 Group 4 Absent 0.60 (0.14–1.00) 16 Description and format are similar to Table 1, but here positive effects (‘‘Effect 1’’) indicate enhancing fertilization by A. socius sperm and negative effects (‘‘Effect ’’) indicate reducing fertilization by A. socius sperm. Refer to Figure 5 for map of linkage groups. a Critical LRS values are 5.6 (suggestive), 10.5 (significant), and 18.2 (highly significant). composite-interval mapping in A. socius was 8–40%, the marker was derived. This marker, although possess- higher than that found in single-marker regression. ing a significant LRS and accounting for 12% of the Although the backcross mapping approach focuses trait variance in both experiments, appeared to have a on species-specific markers, there were 14 AFLP frag- stronger effect in A. socius where the average trait value ments shared between the 116 A. fasciatus and 275 A. associated with its presence was 76% fertilization by A. socius fragment sets. These shared fragments along with socius sperm (IQR 0.64–1.00) vs. 26% fertilization by A. the hundreds of others from which they were filtered fasciatus sperm (IQR 0.03–0.43) in the A. fasciatus underscore the close relatedness and recent divergence experiment (Tables 1 and 2). of the two species. Of these 14 shared fragments, only one, A. fasciatus marker B6.252.15 and A. socius marker DISCUSSION B6.252.03, mapped to linkage groups in both species, but with no trait association: group 4 in A. fasciatus and Compared to an earlier study (Howard et al. 2002) we group 10 in A. socius (Figures 4 and 5). The slight dis- were successful in increasing the number of species- crepancy in fragment sizes reflects small vagaries in the unique AFLP markers available for creating linkage ABI 3100 sequencer’s binning algorithms. Even though maps and for characterizing QTL in A. fasciatus.We B6.252.15/B6.252.03 was a common marker between raised 25 AFLP markers on 8 linkage groups (318.4 cM) the species, we retained the linkage group numbering found in the original experiment to 64 markers on 13 (4 in A. fasciatus and 10 in A. socius) output from Map linkage groups (833.9 cM) in the new analysis (Figure Manager since linkage groups were derived separately in 4). With the new analyses we also substantially increased the A. fasciatus and A. socius QTL experiments. Another the number of QTL detected in the A. fasciatus ex- shared fragment, B6.088.93 in A. fasciatus and B6.088.98 periment (Tables 1 and 3). The preliminary A. fasciatus in A. socius, appeared in A. socius linkage group 6 and data had yielded 6 significant QTL (only 2 with a was associated with the CSP trait in single-marker re- positive effect on conspecific sperm utilization), where- gression in A. fasciatus. One other fragment, B6.098.51 as in the new analysis we identified 14 QTL of positive present in 31 A. fasciatus individuals and B6.098.65 effect and 4 of negative effect. One interesting outcome present in 11 A. socius individuals, was associated with of the expanded analysis of the A. fasciatus data was that the CSP trait in single-marker regression in both species. as more QTL markers were added, the genetic archi- In both sets of backcrosses, the presence of this marker tecture of the trait remained fairly constant. Specifically, increased the use of sperm from the species from which in the old and the new analyses of A. fasciatus we located Conspecific Sperm Precedence 1217

QTL of positive and negative effect (although the proportion of negative QTL was greatly reduced in the new A. fasciatus analysis), QTL associated with linked markers were found on a small proportion of the total number of linkage groups, and linked and unlinked QTL accounted for a range of effects on the trait variances (i.e., some QTL with a small effect and some QTL with a more moderate effect). Of particular in- terest was the fact that the range of effects of positive

1219 Group 1 Group 3 QTL in A. fasciatus was stable as we increased the number of markers available for study: the previous A. fasciatus analysis had revealed 2 unlinked positive QTL

-values for LRS were not available. The explaining 7 and 24% of the trait variance; the new P single-marker analysis identified 6 positive QTL explain- . 2002) ing 9–17% of the trait variance. Moreover, the current et al b b interval analysis identified 4 positive QTL explaining 9– (peak) % at peak LRS Linkage group 7.36.7 9 Group 2 6.5 a EER 19% of the trait variance. Thus, as we added markers to the analysis we did not find that the trait variance became more divided between a greater number of QTL of smaller effect or was confined to fewer markers ) LRS of large effect; rather, the pattern of QTL of moderate to / large effect was simply expanded across more markers 1 1 1 1 (Tables 1 and 3). This stability held true for the negative QTL in A. fasciatus as well, where, except for a single

Effect ( outlier (a negative QTL accounting for 38% of the trait variance), effects of negative QTL ranged from 8 to 18% in the old analysis, comparable to the 9–16% found in n , calculated in Map Manager (M the new analysis. The number of A. fasciatus linkage groups found in Howard et al.’s (2002) study was consistent with the hap- TABLE 3

A. fasciatus loid number of chromosomes in Allonemobius (N ¼ 8; Lim 1971); however, in the present study, both for A. socius and A. fasciatus, the numbers of linkage groups are higher. The most straightforward explanation for the

offspring (IQR) discrepancy is that the greater number of AFLP markers identified in the present study allowed clusters of linked Interval mapping (no control for other QTL) markers to form that could not form in the previous Composite-interval mapping (control for other QTL) study. Markers formed small groups rather than ap- A. fasciatus pending to larger groups because they represent seg- ments of large chromosomes physically distant from more easily detected, closely adjacent groups of markers. The karyotype of Allonemobius is characterized by a disproportionately large X chromosome (Lim 1971) that may be the source of one or more of the larger linkage groups as well as one or more of the smaller Interval QTL mapping analysis results for linkage groups in both A. fasciatus and A. socius. The power to gain uniform coverage of AFLP markers throughout the genome decreases when the mapping population falls below 400 or 500 individuals (Lynch and Walsh 1998), and on a large chromosome such as the X chromosome there may be enough variation in the population that rare, physically isolated clusters of linked markers would be detectable only with either a larger mapping population or a larger repertoire of AFLP markers. Critical LRS values are 6.8 (suggestive), 11.8 (significant), and 19.5 (highly significant). Value is slightly below minimum LRS, but the presence of a QTL atIn this marker is supportedfour by a significant LRS bootstrap valueinstances (see text). in the A. fasciatus linkage groups and a b PresentAbsent 0.37 (0.05–0.67) 0.17 (0.00–0.22) 6 25 PresentAbsentPresentAbsentDescription and format are similar to Table 1, but LRS and percentage are given for the peak value of the QTL closest to the marker. 0.61 (0.23–1.00) 0.36 (0.02–0.64) 0.33 (0.12–0.63) 0.15 (0.00–0.19) 29 23 10 21 H5.311.13 B5.190.45 (H5.225.52) AFLP marker or marker interval Control marker Average % B6.069.43 (G3.445.79) control QTL marker is given in parentheses for instances of composite-interval mapping. Refer to Figure 4six for map of linkage groups. instances in the A. socius linkage groups two to three 28S .Britch C. S. 1218

TABLE 4

Interval QTL mapping analysis results for A. socius, calculated in Map Manager (MEER et al. 2002)

AFLP marker or marker interval Control marker Average % A. socius offspring (IQR) n Effect (1/) LRSa (peak) % at peak LRS Linkage group Interval mapping (no control for other QTL) B5.304.15 Present 0.76 (0.51–1.00) 18 Absent 0.59 (0.13–1.00) 8 G3.324.66 1 5.8 26 Group 1 Present 0.89 (0.98–1.00) 12 Absent 0.70 (0.50–1.00) 11 ) B5.140.99 1 7.2 29 Group 4 Present 0.85 (0.85–1.00) 13 Absent 0.46 (0.13–0.63) 8 B5.195.64 1 11.9 40 Group 4 Present 0.77 (0.60–1.00) 35 Absent 0.44 (0.12–0.75) 19 tal et

Composite-interval mapping (control for other QTL) . H5.107.37 (G3.314.13) 5.6 8 Group 2 Present 0.58 (0.15–1.00) 30 Absent 0.68 (0.27–1.00) 33 B5.140.99 (H5.292.14) ) Present 0.85 (0.85–1.00) 13 Absent 0.46 (0.13–0.63) 8 H5.292.14 (H5.292.14) 1 24.6 69 Group 4 Present 0.85 (0.86–1.00) 18 Absent 0.63 (0.50–1.00) 5 Description and format are similar to Table 1, but LRS and percentage are given for the peak value of the QTL closest to the marker. P-values for LRS were not available. The control QTL marker is given in parentheses for instances of composite-interval mapping. In instances where the peak QTL is in a marker interval, the LRS and percentage values at this peak are shown and the two markers enclosing this interval are grouped together. Refer to Figure 5 for map of linkage groups. a Critical LRS values are 5.6 (suggestive), 10.5 (significant), and 18.2 (highly significant). Conspecific Sperm Precedence 1219

AFLP markers mapped to the same location. In only one fasciatus. Assuming additive polygenic control of CSP, case, on linkage group 2 in the A. fasciatus map, did a backcross females are expected to produce, on average, marker associated with a QTL, H5.311.13, map to the 25% A. fasciatus offspring if mated once to an A. socius same location as another AFLP marker, G7.164.65. male and once to an A. fasciatus male. The average per- Aside from basic karyotypic work documented by Lim centage of offspring sired by A. fasciatus and produced (1971) there are no studies of chromosome polymor- by females from backcrosses to A. socius (i.e., in the phisms in Allonemobius that we are aware of. The fact A. fasciatus experiment) was 39%, with an IQR of 0.00– that some markers map to the same location in both 0.68. While not a close match to the expected percent- species may mark the presence of inversions, but if there age of 25%, more eggs were fertilized by A. socius males are inversions there is little evidence in this study that than by A. fasciatus males. In contrast, the average per- QTL for conspecific sperm precedence are located centage of offspring sired by A. socius and produced by within them or that they underlie the process of spe- females from backcrosses to A. fasciatus (in the A. socius ciation in A. fasciatus and A. socius. experiment) was 58% with an IQR of 0.15–1.00. In other The genetic control of CSP in A. socius presents a words, in the A. socius experiment, backcross females, slightly different picture from the one gleaned from the which were expected on average to preferentially use QTL analysis of CSP in A. fasciatus. In A. socius, AFLP loci A. fasciatus sperm, were on average preferentially using accounting for CSP are fewer and of much larger effect A. socius sperm. In addition to this, the IQR was skewed than those in A. fasciatus. In addition, A. socius females to the high end, and many females produced 100% possess a higher frequency of loci that hinder the use of A. socius offspring (data not shown). conspecific sperm, but the loci tend to have weaker In most mating trials in the A. socius experiment, due negative effects (Tables 2 and 4). The finding that QTL to a shortage of males, both A. socius and A. fasciatus accounted for a higher percentage of the trait variance males were used in multiple matings (mean number of in the A. socius experiment may have been due to the matings ¼ 2.7, SD 1.6; range ¼ 1–7 matings). Matings Beavis effect. The Beavis effect refers to the expectation are physiologically costly to males in the A. fasciatus– that additive genetic effects will be overestimated in A. socius system since males yield large protein-rich QTL experiments with smaller sample sizes (i.e., n > spermatophores to the female and allow the female to 500; Beavis 1994, 1998). The Beavis effect also states consume hemolymph. About 24 hr was allotted to males that QTL of small effect are unlikely to be detected in between matings to allow sperm supplies and hemo- studies with small sample sizes. Thus, there may be many lymph to regenerate. Moreover, if we used a multiply more loci involved with CSP in both A. fasciatus and A. mated male in the first mating, we used a multiply mated socius than were detected in this study. In other words, male in the second mating that had been used in a whether the underlying genetics controlling CSP in similar number of matings as the first male. However, it females involve the accumulated effect of many small may be the case that males of the two species recover at loci or a few loci of strong effect cannot be fully an- different rates, giving A. socius males an advantage in swered by the work presented here. However, the num- sperm competition because of concentration effects, bers of QTL found in both experiments, 18 in A. fasciatus effects of sperm quality, or effects of senescence. Also, and 13 in A. socius, are comparable to those summarized although the target time period between first and in Table 1 of Orr (2001), which shows a range of 1–19 second matings was 24 hr, once-mated females were genes accounting for species differences across five often unwilling to mate with a second male in that time groups of insects and plants. Also, the high phenotypic span. In cases of a more extended time between first and variances accounted for by the QTL (Tables 1–4) sug- second matings, one would expect first-mated A. fascia- gest that the apparent major effect of these QTL is real. tus to have a fertilization advantage. We used PROC One finding worthy of discussion is the difference in GLM in SAS v.8 (SAS Institute, Cary, NC) to perform an sperm utilization patterns exhibited by backcross fe- analysis of variance between the frequency of A. socius males in the A. fasciatus experiment and the A. socius offspring and the class variables ‘‘number of prior experiment. Although second-male sperm precedence matings A. fasciatus male’’ and ‘‘number of prior matings is a phenomenon observed in many insect species, the A. socius males’’ and the continuous variable ‘‘numberof work of Howard and Gregory (1993) and Gregory days between first and second matings.’’ We tested for and Howard (1994) showed that sperm mixing occurs effects of each independent variable as well as inter- in both A. fasciatus and A. socius (the second male fer- actions between independent variables. PROC GLM tilizes the same proportion of eggs as the first male in indicated that no relationships exist between any com- conspecific matings). Moreover, the conspecific male bination of mating experience of males of either species, sires .90% of the offspring in matings between A. socius or the lapse between first and second matings, and the and A. fasciatus, regardless of whether the conspecific variance in the number of A. socius offspring (ANOVA, male mates first or second. The genomic content of F ¼ 0.13–2.53, P ¼ 0.12–0.95). a female offspring of a backcross to A. socius (in the At the very least, the outcome of the ANOVA suggests A. fasciatus experiment) is 75% A. socius and 25% A. that the use of males of both species multiple times did 1220 S. C. Britch et al. not influence our results in the A. socius experiment. The presence of negative QTL in both species could However, this phenomenon of more fertilization of also be explained by epistasis (Howard et al. 2002). backcrosses to A. fasciatus by A. socius males than by Epistatic interactions are clearly the basis of hybrid male A. fasciatus males carries implications for the A. fascia- sterility and inviability in Drosophila (Wuand Hollocher tus–A. socius hybrid zone. The zone has been moving 1998; Coyne and Orr 1999). In the case of Drosophila, north over 14 years of sampling, as inferred by the alleles that behave normally in the genetic background of increased presence of A. socius in mixed populations their own species cause hybrid sterility and inviability and in A. fasciatus populations, sometimes to the point when introduced into the genetic background of another of extinction of pure A. fasciatus types (Britch et al. species. In the present situation, the negative QTL could 2001; S. C. Britch and D. J. Howard, unpublished represent the effects of any kind of gene (i.e.,notnec- results). A. socius CSP alleles in hybrid females may have essarily a locus that has anything whatever to do with the a stronger effect than CSP alleles from A. fasciatus, biochemistry of conspecific sperm precedence when pre- causing hybrid females to behave like A. socius females sent in its ‘‘home’’genetic background) that has a negative with regard to patterns of sperm utilization. Although fertilization side effect when bred into a genome that is changing climate has been implicated in the northward 75% the opposite species (Figure 2). movement of the zone (Britch et al. 2001), if both types Orr (2001) summarized important questions regard- of backcross females in mixed populations preferen- ing the genetic architecture of species differences, two tially utilize A. socius sperm in fertilizing eggs, A. socius of which, the number of genes involved and the mag- will become the dominant species in mixed populations nitude of their phenotypic effects, we have been able to over time. Thus, skewed patterns of fertilization in address. Orr (2001) also noted that relative to natural backcross females may be another factor leading the selection, sexual selection may increase the complexity zone of contact between the species to shift north. of genetic changes accompanying the evolution of Patterns of flow across the hybrid zone of AFLP markers species differences. As evidence for this point of view, associated with CSP are being looked at in a separate Orr (2001) cited two Drosophila studies that found 19 ritch oward study (S. C. B and D. J. H , unpublished and 11 QTL controlling divergence in male genitalia (a data). rapidly evolving trait in insects that is thought to be Some discussion of the negative QTL found in both under strong sexual selection). This is a greater number species is warranted since they too were observed to ex- of QTL than identified in Drosophila and other taxa for ert a force on the phenotypic variances, albeit in favor species-specific traits that are not subject to sexual of the heterospecific sperm (Tables 1, 2, and 4). The selection. The numbers of CSP QTL that we found in phenomenon of conspecific sperm precedence in both both species are comparable to the numbers found in species is very strong (Howard and Gregory 1993; the studies cited in Orr (2001), namely 18 in A. fasciatus Gregory and Howard 1994; Howard et al. 1998a,b) and 14 in A. socius. As far as the distribution of pheno- and the efficacy of the barrier in nature is supported by typic effects among genes important in species differ- the strong bimodality of character index scores ob- ences (Orr 2001), we do not have the experimental served in mixed populations of the two species (Britch resolution to distinguish whether a strong QTL is ac- et al. 2001). Why is it that not all QTL contribute to the tually one gene or the combined effect of several ad- effect and that some appear to have an antagonistic jacent genes that are individually weak. The majority of effect on sperm utilization? Howard et al. (2002) dis- QTL in both A. fasciatus and A. socius appear to cover a cussed one possible explanation for the negative QTL, small range of effects, from low to moderate; although that is, sexual conflict (reviewed in Howard 1999 and a few QTL exist in A. socius with very large effects (Table Panhuis et al. 2001). Briefly, the argument is that the 4). The QTL in both species are spread across many multiple mating that is characteristic of Allonemobius unlinked markers, yet the linked markers associated females establishes sperm competition as an important selective pressure on males and leads to antagonistic with QTL are on a relatively small number of linkage coevolution between the two sexes in traits related to groups. Some evidence points to genes controlling prezygotic isolation being concentrated on the sex chro- fertilization. This sexual conflict leads to females pos- ervedio aetre sessing genes that have antagonistic effects on the sperm mosomes (S and S 2003) and it may be of conspecific males. Thus, uncovering QTL with antag- that these few linkage groups in A. fasciatus and A. socius onistic effects on conspecific sperm may be seen as are actually fragments of the large sex chromosome im providing additional empirical support for the concept characteristic of Allonemobius (L 1971). of sexual conflict and its importance in driving the We thank T. Parchman and S. Long for discussions regarding evolution of barriers to fertilization between closely re- interpretation and troubleshooting of AFLPs, K. Hopper for discus- lated species (Rice 1996, 1998a,b; Howard 1999; sions regarding QTL analysis, and K. Edwards and C. Johnson for avrilets nowles arkow iklund expert help in the lab. This work was supported in part by a New G 2000; K and M 2001; W Mexico State University Department of Biology Excellence in Re- et al.2001;Miller and Pitnick 2003; Morrow and search Fellowship to S.C.B. and in part by National Science Founda- Arnqvist 2003; Orteiza et al. 2005). tion grants DEB 011613 and DEB 0316194 to D.J.H. Conspecific Sperm Precedence 1221

LITERATURE CITED Howard, D. J., P. G. Gregory,J.Chu and M. L. Cain, 1998a Con- specific sperm precedence is an effective barrier to hybridiza- eavis B , W. D., 1994 The power and deceit of QTL experiments: les- tion between closely related species. Evolution 52: 511–516. sons from comparative QTL studies, pp. 255–256 in Proceedings of Howard, D. J., M. Reece,P.G.Gregory,J.Chu and M. L. Cain, the Corn and Sorghum Industry Research Conference. American Seed 1998b The evolution of barriers to fertilization between closely Trade Association, Washington, DC. related organisms, pp. 279–288 in Endless Forms: Species and Speci- eavis B , W. D., 1998 QTL analyses: power, precision and accuracy, ation, edited by D. J. Howard and S. H. Berlocher. Oxford Uni- pp. 145–162 in Molecular Dissection of Complex Traits, edited by versity Press, New York. A. H. Paterson. CRC Press, Boca Raton, FL. oward arshall ampton ritch enedix r oward H , D. J., J. L. M ,D.D.H ,S.C.B ,M.L. B , J. H., J ., and D. J. H , 1991 Calling song displace- Draney et al., 2002 The genetics of reproductive isolation: a ret- ment in a zone of overlap and hybridization. Evolution 45: rospective and prospective look with comments on ground crick- 1751–1759. radshaw tettler ets. Am. Nat. 159: S8–S21. B , H. D., and R. F. S , 1995 Molecular genetics of Kearsey, M. J., 1998 The principles of QTL analysis (a minimal growth and development in Populus. 4. Mapping QTLs with large mathematics approach). J. Exp. Bot. 49: 1619–1623. effects on growth, form, and phenology traits in a forest tree. Knowles, L. L., and T. A. Markow, 2001 Sexually antagonistic co- Genetics 139: 963–973. radshaw ilbert tto chemske evolution of a postmating-prezygotic reproductive character B , H. D. J., S. M. W ,K.G.O and D. W. S , in desert Drosophila. Proc. Natl. Acad. Sci. USA 98: 8692–8696. 1995 Genetic mapping of floral traits associated with reproduc- Lillie, F. R., 1921 Studies on fertilization. VIII. On the measure of tive isolation in monkeyflowers (Mimulus). Nature 376: 762– specificity in fertilization between two associated species of the 765. sea-urchin genus Strongylocentrotus. Biol. Bull. 40: 1–22. Bradshaw, H. D., K. G. Otto,B.E.Frewen,J.K.McKay and D. W. Lim, H.-C., 1971 Note on the chromosomes of the (Or- Schemske, 1998 Quantitative trait loci affecting differences in thoptera: ). Can. J. Zool. 49: 391–395. floral morphology between two species of monkeyflower (Mimu- Lincoln, S. E., M. J. Daly and E. S. Lander, 1993 MAPMAKER/ lus). Genetics 149: 367–382. EXP v. 3.0, Ed. 3. Whitehead Institute for Biomedical Research Britch, S. C., M. L. Cain and D. J. Howard, 2001 Spatio-temporal Technical Report. Whitehead Institute, Cambridge, MA. dynamics of the Allonemobius fasciatus—A. socius mosaic hybrid Loeb, J., 1915 On the nature of the conditions which determine or zone: a 14-year perspective. Mol. Ecol. 10: 627–638. prevent the entrance of the spermatozoon into the egg. Am. Nat. Brown, D. V., and P. E. Eady, 2001 Functional incompatibility be- 49: 257–285. tween the fertilization systems of two allopatric populations of Lynch, M., and B. Walsh, 1998 Genetics and Analysis of Quantitative Callosobruchus maculatus (Coleoptera: Bruchidae). Evolution 55: Traits. Sinauer Associates, Sunderland, MA. 2257–2262. ac onald oldstein Civetta, A., H. M. Waldrip-Dail and A. G. Clark, 2002 An intro- M D ,S.J.,andD.B.G , 1999 A quantitative genetic gression analysis to mapping differences in mating success and analysis of male sexual traits distinguishing the sibling species Dro- sophila simulans and D. sechellia.Genetics153: 1683–1699. sperm competitive ability in Drosophila simulans and D. sechellia. eer anly udmore Genet. Res. 79: 65–74. M , J. M., K. F. M and R. H. J. C , 2002 MAP- Coyne, J. A., and H. A. Orr, 1999 The evolutionary genetics of spe- MANAGER QTXb16: software for genetic mapping of Mendelian ciation, pp. 1–36 in Evolution of Biological Diversity, edited by A. E. markers and quantitative trait loci. Roswell Park Cancer Institute, agurran ay Buffalo, NY. M and R. M. M . Oxford University Press, London. iller itnick Coyne, J. A., and H. A. Orr, 2004 Speciation. Sinauer Associates, M , G. T., and S. P , 2003 Functional significance of sem- Sunderland, MA. inal receptacle length in Drosophila melanogaster. J. Evol. Biol. 16: oherty oward 114–126. D , J. A., and D. J. H , 1996 Lack of preference for orrow rnqvist conspecific calling songs in female crickets. Anim. Behav. 51: M , E. H., and G. A , 2003 Costly traumatic insemina- 981–989. tion and a female counter-adaptation in bed bugs. Proc. R. Soc. Fishman, L., A. J. Kelly and J. H. Willis, 2002 Minor quantitative Lond. Ser. B. Biol. Sci. 270: 2377–2381. urnberger ofman org rey raetzel trait loci underlie floral traits associated with mating system diver- N , B., S. H ,B.F -B ,G.P ,A. aclean gence in Mimulus. Evolution 56: 2138–2155. M et al., 2003 A linkage map for the hybridising toads Fricke, C., and G. Arnqvist, 2004 Conspecific sperm precedence Bombina bombina and B. variegata (Anura: Discoglossidae). Hered- in flour beetles. Anim. Behav. 67: 729–732. ity 91: 136–142. rr Gavrilets, S., 2000 Rapid evolution of reproductive barriers driven O , H. A., 2001 The genetics of species differences. Trends Ecol. by sexual conflict. Nature 403: 886–889. Evol. 16: 343–350. rteiza inder ice Gregory, P. G., and D. J. Howard, 1994 A postinsemination barrier O , N., J. E. L and W. R. R , 2005 Sexy sons from re- to fertilization isolates two closely related ground crickets. Evolu- mating do not recoup the direct costs of harmful male interac- tion 48: 705–710. tions in the Drosophila melanogaster laboratory model system. J. Gregory, P. G., M. D. Remmenga and D. J. Howard, 1998 Patterns Evol. Biol. 18: 1315–1323. of mating between two closely related ground crickets are not Panhuis, T. M., R. K. Butlin,M.Zuk and T. Tregenza, 2001 Sexual influenced by sympatry. Entomol. Exp. Appl. 87: 263–270. selection and speciation. Trends Ecol. Evol. 16: 364–371. Hartl, D. L., and A. G. Clark, 1997 Principals of Population Genetics, Price, C. S. C., 1997 Conspecific sperm precedence in Drosophila. Ed. 3. Sinauer Associates, Sunderland, MA. Nature 388: 663–666. Howard, D. J., 1983 Electrophoretic survey of eastern North Amer- Price, C. S. C., K. A. Dyer and J. A. Coyne, 1999 Sperm competition ican Allonemobius (: Gryllidae): evolutionary relation- between Drosophila males involves both displacement and inca- ships and the discovery of three new species. Ann. Entomol. Soc. pacitation. Nature 400: 449–452. Am. 76: 1014–1021. Rice, W. R., 1996 Sexually antagonistic male adaptation triggered by Howard, D. J., 1986 A zone of overlap and hybridization between experimental arrest of female evolution. Nature 381: 232–234. two ground cricket species. Evolution 40: 34–43. Rice, W. R., 1998a Male fitness increases when females are elimi- Howard, D. J., 1999 Conspecific sperm and pollen precedence and nated from gene pool: implications for the Y chromosome. Proc. speciation. Annu. Rev. Ecol. Syst. 30: 109–132. Natl. Acad. Sci. USA 95: 6217–6221. Howard, D. J., and D. G. Furth, 1986 Review of the Allonemobius Rice, W. R., 1998b Requisite mutational load, pathway epistasis and fasciatus (Orthoptera: Gryllidae) complex with the description deterministic mutation accumulation in sexual versus asexual of two new species separated by electrophoresis, songs, and mor- populations. Genetica 102/103: 71–81. phometrics. Ann. Entomol. Soc. Am. 79: 472–481. Servedio, M. R., and G. P. Saetre, 2003 Speciation as a positive Howard, D. J., and P. G. Gregory, 1993 Post-insemination signal- feedback loop between postzygotic and prezygotic barriers to ing systems and reinforcement. Philos. Trans. R. Soc. Lond. Ser. B gene flow. Proc. R. Soc. Lond. Ser. B 270: 1473–1479. 340: 231–236. Simmons, L. W., 2001 Sperm Competition and Its Evolutionary Conse- Howard, D. J., G. L. Waring,C.A.Tibbets and P. G. Gregory, quences in the Insects. Princeton University Press, Princeton, NJ. 1993 Survival of hybrids in a mosaic hybrid zone. Evolution Tanksley, S. D., 1993 Mapping polygenes. Annu. Rev. Genet. 27: 47: 789–800. 205–233. 1222 S. C. Britch et al.

Tao, Y., Z. B. Zeng,J.Li,D.L.Hartl and C. C. Laurie, Wiklund, C., B. Karlsson and O. Leimar, 2001 Sexual conflict and 2003 Genetic dissection of hybrid incompatibilities between cooperation in butterfly reproduction: a comparative study of Drosophila simulans and D. mauritiana. II. Mapping hybrid male polyandry and female fitness. Proc. R. Soc. Lond. Ser. B Biol. sterility loci on the third chromosome. Genetics 164: 1399–1418. Sci. 268: 1661–1667. Voorrips, R. E., 2002 MapChart: software for the graphical presen- Wu, C.-I., and H. Hollocher, 1998 Subtle is nature: the genetics of tation of linkage maps and QTLs. J. Hered. 93: 77–78. species differentiation and speciation, pp. 339–351 in Endless Vos, P., R. Hogers,M.Bleeker,M.Reijans,T.Van De Lee et al., Forms: Species and Speciation, edited by D. J. Howard and S. H. 1995 AFLP: a new technique for DNA fingerprinting. Nucleic Berlocher. Oxford University Press, New York. Acids Res. 23: 4407–4414. Wade, M. J., H. Patterson,N.W.Chang and N. A. Johnson, 1994 Postcopulatory, prezygotic isolation in flour beetles. He- redity 72: 163–167. Communicating editor: M. K. Uyenoyama