Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
J OURNAL OF C RUSTACEAN B IOLOGY, 34(4), 460-466, 2014 THE INHERITANCE OF AUTOSOMAL AND SEX-LINKED CUTICULAR PIGMENTATION PATTERNS IN THE MARINE ISOPOD, PARACERCEIS SCULPTA HOLMES, 1904 (ISOPODA: SPHAEROMATIDAE) Stephen M. Shuster ∗ , Saundra J. Embry, Carla R. Hargis, and Adrianna Nimer Department of Biological Sciences, Northern Arizona University, Flagstaff, AZ 86011-5640, USA ABSTRACT Cuticular pigmentation is highly variable in Paracerceis sculpta Holmes, 1904, a Gulf of California isopod. Individuals expressing L2r (= Laterals-2 red) have red pigmentation on the lateral margins of their 6-7th body segments. Individuals expressing 3rs (= Three red stripes) have three red pigmentation zones running the length of their bodies. Individuals expressing Cd (= Cephalon dark) have black head capsules. In 12 years of field collections, L2r and 3rs represented less than 0.3% of all individuals (N = 5491) with the frequency of each sex proportional to population frequencies. Of all individuals scored as Cd (N = 178), 98% were females. We crossed marked and unmarked parents in all possible combinations. Progeny ratios for L2r and 3rs met Mendelian expectations within families as did adult expression of 3rs despite observed biases in family sex ratio. In three generations, Cd females crossed to unmarked males produced families with 1:1 sex ratios, 100% Cd daughters and no Cd sons. Sons from these families never produced Cd daughters. Our results suggest that L2r and 3rs are controlled by dominant, autosomal alleles. The sex-limited expression of Cd corroborates previous results suggesting female heterogamety in P. sculpta and in other flabelliferan isopods. The appearance of rare phenotypes controlled by dominant alleles is paradoxical given the hypothesis that allelic dominance evolves in response to positive selection. However, this combination might persist if apostatic selection imposed by visual predators occurs in this species’ structured populations, thereby favoring dominance modifiers that suppress fitness losses by heterozygotes. This species with its large number of cuticular markers could provide tests of this hypothesis. K EY W ORDS: allelic dominance, apostatic selection, color patterns, female heterogamety, genetics DOI: 10.1163/1937240X-00002251 I NTRODUCTION Cuticular pigmentation patterns in isopod crustaceans are diverse and variable (Legrand-Hamelin, 1976; Heath, 1979; Shuster, 1989). Most patterns appear to be controlled by dominant Mendelian alleles at autosomal loci, which persist at low frequency in natural populations. Four specific predictions follow from these results: 1) most individuals bearing cuticular markers in nature are expected to be heterozygous at the marker locus; 2) both males and females are expected to bear these markers in equal frequency; 3) marked individuals are expected to produce 1:1 ratios of marked:unmarked progeny when crossed to unmarked individuals; and 4) marked individuals are expected to produce 3:1 ratios of marked:unmarked progeny when crossed to other individuals bearing the same marker. Certain other cuticular patterns in isopod crustaceans appear to be sex linked (Legrand-Hamelin, 1976; Shuster and Levy, 1999). Crustacean sex determination mechanisms are known to involve combinations of allelic, chromosomal and extrachromosomal factors (Ginsberger-Vogel and Charniaux-Cotton, 1982; Bull, 1983; Legrand et al., 1987; Heath and Ratford, 1990; Juchault et al., 1992; Rousset et al., 1992; Hurst, 1993; Rigaud and Juchault, 1993; Juchault and Rigaud, 1995). Unless sex change is possible (Shuster and Sassaman, 1997) only members of the heterogametic sex are expected to express the marker when cuti∗ Corresponding cular marker loci are located on heterochromosomes influencing sex determination. Most studies demonstrating chromosomal sex determination in isopods suggest that females are the heterogametic sex (ZW = females; ZZ = males; Legrand-Hamelin, 1976; Ginsberger-Vogel and CharniauxCotton, 1982; Legrand et al., 1987; Juchault and Rigaud, 1995; Shuster and Levy, 1999; although see Tomaszkiewicz et al., 2010). The sex-limited expression of cuticular pigmentation patterns has been documented in three genera of flabelliferan isopods to date (Legrand et al., 1987; Shuster and Levy, 1999). Paracerceis sculpta Holmes, 1904 is a sphaeromatid isopod native to the central Pacific coast of North America, with well-studied populations inhabiting the Texas coast of the Gulf of México (Munguia and Shuster, 2013) and the northern Gulf of California in the Republic of México (Shuster et al., 2001). Previous studies of this latter population have documented the inheritance of three autosomal cuticular pigmentation markers (Shuster, 1989) and one cuticular marker whose inheritance is consistent with sex linkage and female heterogamety (Shuster and Sassaman, 1997; Shuster and Levy, 1999). In this paper, we document the inheritance of three additional cuticular markers in this species. Individuals bearing L2r (= Laterals-2 red), have red pigmentation on the lateral margins of each of their sixth and seventh body segments (Fig. 1a-c). Individuals bearing 3rs author; e-mail: [email protected] © The Crustacean Society, 2014. Published by Brill NV, Leiden DOI:10.1163/1937240X-00002251 SHUSTER ET AL.: INHERITANCE OF CUTICLE PATTERNS IN PARACERCEIS 461 Fig. 1. a, dorsal view of female Paracerceis sculpta bearing the autosomal pigmentation pattern L2r (= Laterals-2 red); b, ventral view of female P. sculpta bearing the autosomal pigmentation pattern L2r; c, dorsal view of female P. sculpta bearing the W-linked pigmentation pattern, Cd (= Cephalon dark). (= Three red stripes), have three bands of red pigmentation running the length of their bodies, one located medially and one on each lateral body margin (Fig. 2). Individuals bearing Cd (= Cephalon dark) have darkly pigmented cephalons and bright, white bodies (Fig. 1c). We present five specific results: 1) the population frequencies of L2r and 3rs were consistent with those reported for heritable pigmentation markers in this, as well as in other sphaeromatid isopod species (Shuster, 1989; Shuster and Levy, 1999); 2) Mendelian inheritance was confirmed among newly-released progeny in L2r and 3rs; 3) expression of 3rs among adults in laboratory-reared families confirmed that this locus is autosomal; 4) a sex-bias in the expression of Cd existed in population samples of P. sculpta collected over a 12-year period; and 5) inheritance of Cd was limited to females in three generations of laboratory reared isopods, indicating that sex determination in P. sculpta involves female heterogamety (ZW = females; ZZ = males) and that Cd is W-linked. We also discuss the hypothesis that rare phenotypes controlled by dominant alleles might persist in this population because visual predators impose apostatic selection on distinctive isopod color patterns (Jormalainen et al., 1995; Bourguet, 1999; Bond and Kamil, 1998, 2006). Apostatic selection occurs when predators form search images that cause distinctive prey to be taken in disproportionate numbers relative to their actual abundance. This condition favors prey phenotypes that do not resemble preferred prey, until preferred prey become extremely rare. Then, predators switch to more abundant prey, modifying their search images to match the phenotype of these now preferred individuals. We suggest that apostatic selection, combined with this species’ structured populations, may favor dominance modifiers that suppress fitness losses by heterozygotes (Wright, 1929; Bourguet, 1999) and so allow distinctive pigmentation polymorphisms to arise and persist at low population frequencies. M ATERIALS AND M ETHODS Field Collections Fig. 2. Diagrams. a, α-male Paracerceis sculpta bearing the autosomal pigmentation pattern L2r; b, female P. sculpta bearing the autosomal pigmentation pattern, 3rs (= Three red stripes); in life, shaded areas appear bright red (redrawn after Shuster, 1991b). Isopods were collected from the spongocoels of the intertidal sponge, Leucetta losangelensis, in the northern Gulf of California between 1985 and 1997. All individuals were sexed, measured to the nearest 0.125 mm and identified by cuticular pigmentation pattern (Shuster, 1989). We tabulated these observations by sex and month. We summed all observations, and using a Chi-squared test (Sokal and Rohlf, 1995), we compared the number of individuals of each sex bearing L2r, 3rs and Cd with the numbers of individuals that were unmarked or which bore some marker other than these markers (indicated ‘+’). 462 JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 34, NO. 4, 2014 Laboratory Experiments Field-collected females bearing the above markers were maintained in seawater and were paired with the individuals bearing the same marker or with unmarked individuals as described in Shuster and Sassaman (1997). Progeny were separated from females at parturition and placed into individual, sterilized glass Petri dishes. Individual isopods were reared to maturity at 24°C on coralline algae (Amphiroa sp.) and brine shrimp flakes with seawater changes every four days, as described in Shuster and Sassaman (1997). We investigated Mendelian inheritance at L2r and 3rs (Figs. 1a-b, 2) by comparing the frequency of marked and unmarked individuals at birth, and for 3rs, at maturity, within and among families. All parents of L2r crosses (N = 8) and F1 parents of 3rs crosses (N = 10) were field-collected, as were all + males in F2 and F3 crosses (N = 3). All sires of these F1-3 families were α-males (Shuster, 1987). The sire and dam of the single 3rs × 3rs cross were also field collected but this sire was a β-male (Shuster and Sassaman, 1997). Consistent with the predictions of Mendelian inheritance, in all crosses involving one marked and one unmarked individual, we expected 1:1 ratios of marked: unmarked individuals. In all crosses involving two marked individuals, we expected 3:1 ratios of marked: unmarked progeny. We also investigated the inheritance of family sex ratio by comparing the number of male and female individuals within and among families. Consistent with the predictions of chromosomal sex determination, we expected 1:1 sex ratios in all families. All comparisons were performed using heterogeneity G-tests (Sokal and Rohlf, 1995). The additive properties of this test provide a means for examining four possible sources of deviation of observed from expected frequencies in genetic hypotheses: 1) Gi measures the deviation within individual i-th crosses of observed phenotypic frequencies from Mendelian 2 expectation, where Gi = X(df=(a−1);α=0.05) = 3.84 and a = the number of classes within each test; 2) GTotal measures the deviation of the sum of all Gi values from their expected magnitude given the number of tests of the genetic hypothesis, where GTotal = GT = Gi(df=b(a−1)) and b = the number of tests of the genetic hypothesis; 3) GPooled measures the deviation of the observed from expected phenotypic frequencies, when observed frequencies are pooled across all tests of the genetic hypothesis within an experiment, e.g., across all 3rs crosses, and where 2 GPooled = GP = X(df=(a−1);α=0.05) ; and 4) GHeterogeneity measures the difference between GT and GP , thereby revealing whether observed frequencies within individual tests of the genetic hypothesis contribute disproportionately to the value of GT , and where GHeterogeneity = GH = 2 X(df=(b(a−1)−1);α=0.05) . A field-collected, gravid female (Shuster, 1991) bearing Cd (Fig. 1c) was maintained in seawater until she released her progeny. Progeny were separated from the female at parturition, placed into individual, sterilized glass Petri dishes, and reared to maturity as described above. Three F1 Cd females were crossed to unmarked sires (+; N = 3) from a laboratory lineage (α-1) that consistently produced families with 1:1 sex ratios. Individuals from this ‘α-1’ lineage were homozygous for the Amsα allele at Ams (= Alternative mating strategy), an autosomal locus which controls male maturation rate, male external morphology and male mating behavior (Shuster and Sassaman, 1997). α-1 individuals were also homozygous for the Tfr1 allele at Tfr (= Transformer), another autosomal locus that causes sex reversal, depending on an individual’s genotype at Ams and at primary sex determination loci (Shuster et al., 2001). Since Amsα and Tfr1 alleles do not interact, the use of Amsα Amsα , Tfr1 Tfr1 sires in this and in subsequent crosses (see below) minimized the possibility of sex ratio distortion within families. The F2 generation was reared to maturity as described above. Four F2 females were crossed to unmarked α1 sires, seven F2 males were crossed to unmarked females from laboratory stocks, and the F3 generation was reared to maturity as well. Within each generation, as in all other crosses, all Cd individuals were recorded at birth as well as at maturity, and all surviving individuals were measured and sexed, allowing us to determine whether sex- or phenotype-specific mortality had occurred before maturity. We investigated Mendelian inheritance of Cd by comparing the frequency of marked and unmarked individuals at birth and at maturity, within and among families. In all comparisons, we expected 1:1 ratios of marked: unmarked individuals. We also investigated the inheritance of family sex ratio by comparing the number of male and female individuals within and among families. Under chromosomal sex determination, we expected 1:1 sex ratios in all families. All comparisons were performed using heterogeneity G-tests (Sokal and Rohlf, 1995). R ESULTS Field Collections In monthly samples collected over a 12-year period, the frequency of L2r in the northern Gulf of California P. sculpta population never exceeded 0.01% within one month, and was less than 0.001 overall (N = 5491). Over the same period, the frequency of 3rs in the P. sculpta population never exceeded 1% within one month, and was less than 0.003 overall. Similar, low but persistent frequencies of cuticular pigmentation markers have been reported in this and in other isopod species (Heath, 1979; Shuster, 1989). In monthly samples collected over the same period, the frequency of Cd in the northern Gulf of California P. sculpta population never exceeded 8% within one month, and was less than 0.02 overall (mean ± 95% CI = 0.015 ± 0.006, N = 5491). Ninety-eight percent of all Cd individuals collected were females (N = 178; G = 124.1, P < 0.001, N = 5491), indicating a significant sex-bias in the expression of Cd in nature. Similar sex-biases in cuticular marker expression are reported in Idotea, Dynamene and Paracerceis (Legrand et al., 1987; Shuster and Levy, 1999). Laboratory Experiments The frequency of L2r among the progeny of four sets of field-collected parents met Mendelian expectations (Table 1). Among the three crosses involving one marked and one unmarked parent, both marked males and marked females produced similar 1:1 ratios of marked and unmarked progeny ( Gi(df=3) = GTotal(df=3) = 0.71, P > 0.1; GPooled = GP = 0.54, P > 0.1, GHeterogeneity(df=2) = GH(df=2) = 0.17, P > 0.10, Table 1). The 3:1 ratio of marked:unmarked progeny in the single cross involving two L2r parents was consistent with Mendelian expectations (G = 0.09, P > 0.1, Table 1). The frequencies of marked and unmarked progeny among all seven crosses involving one parent bearing 3rs and one unmarked parent also conformed to Mendelian expectations. (GT(df=7) = 6.93, P > 0.10, N = 438; Table 2). As with L2r, marked parents of either sex produced similar progeny ratios (GH(df=6) = 5.61, P > 0.1, Table 2). The ratios of marked and unmarked progeny among the six families reared to adulthood were also consistent with Table 1. Heterogeneity G-tests for Mendelian inheritance of L2r in Paracerceis sculpta. 1 GT(df=3) = 0.71, P > 0.10; GP = 0.54, P > 0.10, GH(df=2) = 0.17, P > 0.10. 2 Comparison of observed and expected ratios of L2r: + at birth; (Ho = 3:1), G(df=1,α=0.05) = 3.84; P > 0.10. Parental genotype At birth Gi Sire Dam L2r + N L2r L2r + Total + + L2r 10 25 14 49 10 21 11 42 20 46 25 91 0.00 0.35 0.36 0.711 L2r L2r 41 15 56 0.092 463 SHUSTER ET AL.: INHERITANCE OF CUTICLE PATTERNS IN PARACERCEIS Table 2. Heterogeneity G-tests for Mendelian inheritance of 3rs in Paracerceis sculpta (F1 -F3 ). 1 Comparison of observed and expected ratios of 3rs:+ at birth within families; (Ho = 1:1), G(df=1,α=0.05) = 3.84. 2 Comparison of observed and expected ratios of 3rs:+ at maturity within families; (Ho = 1:1), G(df=1,α=0.05) = 3.84. 3 Comparison of observed and expected sex ratios within families; (Ho = 1:1), Gi(SR)(df=1,α=0.05) = 3.84. 4 Heterogeneity analysis of at birth ratios of 3rs:+ for crosses involving one marked and one unmarked parent: GT(df=7) = 6.93, P > 0.10, GP = 1.32, P > 0.10, GH(df=6) = 5.61, P > 0.10. 5 Heterogeneity analysis of at maturity ratios of 3rs:+ for crosses involving one marked and one unmarked parent; GT(df=6) = 2.92, P > 0.50, GP = 1.96, P > 0.10, GH(df=5) = 0.96, P > 0.90. 6 Heterogeneity analysis of sex ratio; GT(SR)(df=6) = 15.29, P < 0.025, GP(SR) = 0.60, P > 0.10, GH(SR)(df=5) = 14.69, P < 0.025; with first F2 cross excluded, GT(SR)(df=5) = 8.43, P > 0.10, GP(SR) = 0.01, P > 0.90, GH(SR)(df=4) = 8.42, P > 0.10. 7 Comparison of observed and expected ratios of 3rs:+ at birth within families; (Ho = 3:1), G(df=1,α=0.05) = 3.84. ∗ P < 0.05. Generation Genotype Sire F1 F2 F3 Total F1 At birth Dam 3rs + At maturity G1i N + 3rs F M F M N G2i G3i(SR) 3rs + 3rs 3rs 3rs + + + + 3rs + + + 3rs 3rs 3rs 11 54 – 80 41 6 8 7 207 12 44 – 94 51 9 16 5 231 23 98 – 174 92 15 24 12 438 0.04 1.02 – 1.13 1.09 0.60 2.72 0.33 6.934 – – 6 24 9 5 0 3 47 – – 2 12 6 1 5 1 27 – – 6 9 9 6 9 2 41 – – 4 38 6 1 0 2 51 – – 18 83 30 13 14 8 166 – – 0.22 1.46 0.00 0.08 1.16 0.00 2.925 – – 2.04 3.51 1.21 6.86∗ 1.16 0.51 15.29∗,6 3rs 3rs 94 22 116 2.39 15 28 2 8 52 3.15 8.98∗ Mendelian inheritance ( Gi(df=7) = 2.92, P > 0.50; GP = 1.96, P > 0.10; GH(df=6) = 0.96, P > 0.90; Table 2), despite the appearance of deviations in family sex ratio ( Gi(SR)(df=6) = 15.29, P < 0.025, GP(SR) = 0.60, P > 0.10, GH(SR)(df=5) = 14.69, P < 0.025). With the first F2 cross excluded (see Discussion), no sex ratio deviation was apparent (GiSR(df=5) = 8.43, P > 0.10, GP(SR) = 0.01, P > 0.90, GH(SR)(df=4) = 8.42, P > 0.10). The deviant cross showed a significant excess of females (85%). We observed no significant deviation from 3:1 Mendelian expectations in the frequencies of 3rs:+ progeny in the single 3rs × 3rs cross, either at birth (G = 2.39, P > 0.10, N = 116) or at maturity (G = 3.15, P > 0.05, N = 52). However, this family showed a significant excess of males (69%; G = 8.98, P < 0.01, N = 52). Although the frequency of Cd among the progeny of the field-collected parental female did not meet Mendelian expectations due to an excess of Cd progeny (G = 16.98, P < 0.001, N = 67; Table 3), the adult ratio of Cd and + progeny within this family was consistent with Mendelian inheritance (G = 0.33, P > 0.95, N = 12). Moreover, in the two generations in which Cd females descended from the field-collected Cd female were crossed to unmarked males, Cd showed Mendelian inheritance at birth within and among all families (with the F1 family excluded: Gi(df=7) = 2.62, P > 0.90; GP = 0.05, P > 0.50; GH(df=6) = 2.57, P > 0.90; Table 3). Table 3. Heterogeneity G-tests for Mendelian inheritance of Cd and sex ratio in Paracerceis sculpta (F1 -F3 ). 1 P < 0.001. 2 All crosses included: GP = 3.64, P > 0.05, GH(df=7) = 15.96, P < 0.05. 3 F1 at birth ratio excluded: GP = 0.05, P > 0.50; GH(df=6) = 2.57, P > 0.50. 4 GP = 0.10, P > 0.50, GH(df=7) = 2.92, P > 0.90. Generation At birth Cd F1 F2 F3 + At maturity N + Cd Gi Sex ratio/Cd F M F M N Gi 50 11 9 11 15 18 8 17 17 17 8 10 10 19 9 19 67 28 17 21 25 37 17 36 16.981 1.30 0.06 0.05 1.01 0.03 0.06 0.11 7 7 4 5 2 9 2 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 6 7 2 2 11 2 7 12 13 11 7 4 20 4 16 0.33 0.08 0.83 1.33 0.00 0.20 0.00 0.25 139 89 109 92 248 181 19.602 2.623 45 0 0 42 87 3.024 464 JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 34, NO. 4, 2014 Among adults, both Cd and family sex ratio showed Mendelian inheritance within and among all families ( Gi(df=8) = 3.02, P > 0.90; GP = 0.10, P > 0.50; GH(df=7) = 2.92, P > 0.90; Table 3). All females expressed Cd, whereas all males were unmarked, indicating close linkage between Cd and sex (Exact X 2 probability = 2.55 × 10−19 , N = 87). This result casts doubt on the hypothesis that Cd is a recessive Z-linked marker. Females bearing such a marker would have produced all unmarked progeny in both sexes. This was never the case in 8 crosses. Moreover, unmarked F2 males (N = 7) crossed to unmarked females produced no Cd progeny of either sex (N = 258), indicating that Cd is not a recessive allele transmitted through male lineages. These results are consistent with Mendelian inheritance of Cd and with chromosomal sex determination involving female heterogamety in this species. D ISCUSSION Our results suggest that 3rs and L2r in P. sculpta are autosomal traits and are consistent with reports of the inheritance of cuticular pigmentation patterns in this, as well as in two other species of flabelliferan isopods (reviews in Ginsberger-Vogel and Charniaux-Cotton, 1982; Legrand et al., 1987). Our results also indicate that Cd in P. sculpta is a sex-linked trait. In particular, the appearance of Cd in three consecutive generations of daughters, never in sons, and never among the progeny of F2 sons of Cd mothers, indicates that females in this species are heterogametic, and that Cd is located on the W-chromosome (ZW = female; ZZ = male). The appearance of sex ratio biases within certain families (Table 2) is consistent with other results indicating that female biased sex ratios occur primarily in families sired by α-males, whereas male biased sex ratios occur primarily in families sired by β-males (Shuster and Sassaman, 1997; Shuster et al., 2001). Sex ratio biases of this sort appear to be caused by interactions between the primary sex determination mechanism and two other autosomal loci (Ams and Tfr; Shuster and Sassaman, 1997) in P. sculpta. Note that despite the observed sex ratio biases, the ratios of 3rs:+ individuals met Mendelian expectations. The decreased fecundity of F2-3 families compared to F1 families is explained by the negative relationship between body size and fecundity in P. sculpta (Shuster, 1991). Isopods maintained in incubators at 24°C (F2-3 ) were smaller in size and less fecund than the field-collected parental Cdfemale, who had matured at a cooler sea surface temperature (17-20°C in March; Shuster and Guthrie, 1999). Despite these differences in fecundity among generations, no differential mortality was detectable between Cd and non-Cd individuals, or between males and females (Table 3). Although genetic factors are known to cause reversal of sexual phenotype in P. sculpta and in other isopods (Legrand et al., 1987; Juchault and Rigaud, 1995; Shuster and Sassaman, 1997), our use of α-sires from families with unbiased sex ratios (lineage α-1) and the lack of biased sex ratios within and among our F1-3 Cd families indicate that factors responsible for sex reversal were not present in these crosses. The excess of Cd progeny among the progeny of the fieldcollected parental female, followed by Mendelian inheri- tance and sex-linkage of Cd among her progeny and grandprogeny, could be explained if this female had been mated in the field by an α-male bearing Cpd (= Cephalon, pleotelson dark; Shuster, unpublished data). Individuals bearing this cuticular pattern possess cephalons as well as pleotelsons that are both darkly pigmented. Newly released Cpdbearing mancas (juveniles) are sometimes difficult to distinguish from Cd mancas, although Cd and Cpd are usually distinct in laboratory-reared adults. However, field-collected α-males bearing Cpd can sometimes appear to bear only Cd, when the pleotelsons of older α-males become encrusted with bacterial plaques (Shuster, 1987). The appearance of three Cd males (0.0006, N = 5491) among field collected individuals may be due to such misdiagnoses. The cuticular pigmentation markers within the Isopoda, whose mode of inheritance is known, are controlled by dominant alleles at Mendelian loci and are observed within natural populations at low population frequencies (LegrandHamelin, 1976; Shuster, 1989; Shuster and Levy, 1999). Because these markers appear to be conspicuous to visual predators such as fish, their low population frequency has been attributed to directional selection (Jormalainen et al., 1989), and also to environmental factors such as salinity or temperature which appear to select against these patterns (Heath, 1979; Khazaeli and Heath, 1979). However, this explanation alone is at odds with the hypothesis that dominance evolves because such alleles are favored by selection (Nachman, 2005). The possibility exists that these markers are associated with metabolic pathways whose kinetic structure leads to marker expression as a dominant trait (Bourguet, 1999), or that past selection has favored an excess of enzymatic activity in response to unpredictable environmental fluctuation (Wright, 1929; Haldane, 1930; Orr, 1991; Forsdyke, 1994). However, another possible explanation for the unusual observed combination in these species could be that these predators impose apostatic selection on isopod pigmentation patterns. Apostatic selection, a form of negative frequency-dependent selection, occurs when visual predators form search images on distinctive prey phenotypes (Allen and Clarke, 1968). Predators continue to seek preferred prey phenotypes even after they become rare, allowing prey bearing other phenotypes to increase in frequency due to their higher fitness relative to preferred prey. However, when preferred prey become extremely rare, predators shift to more common prey, in turn forming search images on them, which relaxes selection against the previously preferred prey phenotype and intensifies it against the new prey phenotype. Such selection allows multiple prey phenotypes to persist in nature at comparatively low frequency, and also is thought to cause prey phenotypes to undergo periodic frequency oscillations (Cook, 2005), a combination of population-wide phenomena that is well documented in isopod populations bearing polymorphic cuticular markers (Shuster, 1989; Isocladus sp., J. Dale, personal communication). Bond and Kamil (2006) showed, using computer images of moths that were preyed upon by blue jays, that when moth phenotypes were variable and were allowed to evolve in response to predation, jays were less likely to detect atypical moths, particularly if these moths were cryptic when SHUSTER ET AL.: INHERITANCE OF CUTICLE PATTERNS IN PARACERCEIS viewed against a complex background. Furthermore, this tendency resulted in apostatic selection imposed by birds on moth populations even under conditions of high moth variability, and was not necessarily restricted to the maintenance of a limited number of discrete morphs, e.g., in classic cases of apostatic selection as is observed in land snails (Clark, 2005), stick insects (Sandoval, 1994) and water boatmen (Popham, 1941). Evolving moths showed significantly greater phenotypic variance than that of unselected controls over successive generations, indicating that apostatic selection can encourage the evolution of phenotypic diversity, not simply dimorphism. Wright (1929) argued that while the intensity of selection on dominant mutant alleles can only be on the order of the mutation rate itself, even weak selection can favor alleles that modify the expression of rare dominant alleles and cause both alleles to increase in frequency. Such modifiers appear to be especially effective when balanced polymorphisms exist and when populations are spatially structured (review in Bourguet, 1999). Isopod populations with polymorphic pigmentation patterns do appear to experience apostatic selection from predation by fish (Jormalainen et al., 1995; Maskell et al., 1977). Furthermore, there is evidence that limited dispersal by P. sculpta mancas may cause significant population structure in this species over relatively short (<1 km) spatial scales (Johnson and Shuster, 1999). If selection on pigmentation patterns in P. sculpta is indeed apostatic as described above, and if such selection can increase the frequency of rare phenotypic variants in visually complex habitats, particularly when prey populations are structured, then selection on cuticular pigmentation variants, perhaps through the influences of modifiers, could not only favor dominant expression in these markers, but maintain them at low population frequencies as well. Although Bond and Kamil (1998, 2006) did not explicitly explore the evolution of dominance in their model, their computer algorithm used haploid chromosomes and so simulated this phenotypic effect. Our results provide evidence of Mendelian inheritance of three additional cuticular pigmentation patterns in P. sculpta, they support existing evidence of chromosomal sex determination and female heterogamety in this species (Shuster and Sassaman, 1997; Shuster and Levy, 1999), and they are consistent with reports of these mechanisms in two other species of flabelliferan isopods (reviews in Ginsberger-Vogel and Charniaux-Cotton, 1982; Legrand et al., 1987). Our results also indicate that the three cuticular patterns described here, like other similar markers described for this species, are rare within the population and segregate as dominant alleles. These results suggest that this experimental system could provide a useful model for examining the evolution of dominance in a marine species. ACKNOWLEDGEMENTS This research was supported by NSF grants OCE-8401067, BSR-8700112, BSR-9106644, DEB-9726504, the NSF/DOE Research Internships Program in Neural and Behavioral Sciences, and the NSF Undergraduate Research Mentorships Program at Northern Arizona University, grants DBI9988009 and DBI-1041255. Additional support was provided by the Department of Biological Sciences, the Organized Research Program and the Henry O. Hooper Undergraduate Fellowship program at NAU. Assistance 465 in maintaining laboratory animals was provided by K. Johnson, J. Learned, E. Omana, K. Perry, K. Ressel, H. Wildey and S. Vuturo. Permission to study P. sculpta populations in the Gulf of California was granted by the Republic of México, permits 412.2.1.3.0.2315, A00-702-06296 and DAN 02384. R EFERENCES Allen, J. A., and B. C. Clarke. 1968. Evidence for apostatic selection by wild passerines. Nature 220: 501-502. Bond, A. B., and A. C. Kamil. 1998. Apostatic selection by blue jays produces balanced polymorphism in virtual prey. Nature 395: 594-596. , and . 2006. Spatial heterogeneity, predator cognition, and the evolution of color polymorphism in virtual prey. Proceedings of the National Academy of Sciences of the United States of America 103: 3214-3219. Bourguet, D. 1999. The evolution of dominance. Heredity 83: 1-4. Brusca, R. C., and G. D. F. Wilson. 1991. A phylogenetic analysis of the Isopoda with some classificatory recommendations. Memoirs of the Queensland Museum 31: 143-204. Bull, J. J. 1983. Evolution of Sex Determining Mechanisms. Benjamin Cummings, Menlo Park, CA. Cook, L. M. 2005. Disequilibrium in some Cepaea populations. Heredity 94: 497-500. Fordyke, D. R. 1994. The heat-shock response and the molecular basis of genetic dominance. Journal of Theoretical Biology 167: 1-5. Ginsberger-Vogel, T., and H. Charniaux-Cotton. 1982. Sex determination, pp. 257-283. In, L. G. Abele (ed.), The Biology of Crustacea. Vol. 2: Embryology, Morphology and Genetics. Academic Press, New York, NY. Haldane, J. B. S. 1930. A note on Fisher’s theory of the origin of dominance, and on a correlation between dominance and linkage. American Naturalist 64: 87-90. Heath, D. J. 1979. Colour polymorphism in the salt marsh isopod, Sphaeroma rugicauda; evidence for stable equilibrium frequencies. Oceologia 44: 95-97. , and J. R. Ratford. 1990. The inheritance of sex ratio in the isopod, Sphaeroma rugicauda. Heredity 64: 419-425. Holmes, S. 1904. Remarks on the sexes of sphaeromids, with a description of a new species of Dynamene. Proceedings of the California Academy of Sciences, Zoology 3: 295-306. Hurst, L. D. 1993. The incidences, mechanisms and evolution of cytoplasmic sex ratio distorters in animals. Biological Reviews 68: 121-193. Johnson, K., and S. M. Shuster. 1999. The genetic population structure of Paracerceis sculpta in the Gulf of California. Journal of the Arizona and Nevada Academy of Science 34: 14. Jormalainen, V., S. Merilaita, and J. Tuomi. 1995. Differential predation on sexes affects colour polymorphism of the isopod Idotea baltica (Pallas). Biological Journal of the Linnean Society 55: 45-68. Juchault, P., and T. Rigaud. 1995. Evidence for female heterogamety in two terrestrial crustaceans and the problem of sex chromosome evolution in isopods. Heredity 75: 466-471. , , and J.-P. Mocquard. 1992. Evolution of sex-determining mechanisms in a wild population of Armadillidium vulgare Latr. (Crustacea: Isopoda): competition between two feminizing parasitic sex factors. Heredity 69: 382-390. Khazaeli, A. A., and D. J. Heath. 1979. Colour polymorphism, selection and the sex ratio in the isopod Sphaeroma rugicauda (Leach). Heredity 42: 187-199. Legrand-Hamelin, E. 1976. Sur le polychromatisme de l’isopode Flabellifere Dynamene bidentata (Adams) III. Relations entre les genes responsables de phenotypes bimaculata et lineata. Archive of Experimental Genetics 117: 325-343. Legrand, J. J., E. Legrand-Hamelin, and P. Juchault. 1987. Sex determination in Crustacea. Biological Reviews 62: 439-470. Maskell, M., D. T. Parkin, and E. Verspoor. 1977. Apostatic selection by sticklebacks upon a dimorphic prey. Heredity 39: 83-89. Munguia, P., and S. M. Shuster. 2013. Established populations of Paracerceis sculpta (Isopoda) in the Northern Gulf of Mexico. Journal of Crustacean Biology 33: 137-139. Nachman, M. W. 2005. The genetic basis of adaptation: lessons from concealing coloration in pocket mice. Genetica 123: 125-136. 466 JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 34, NO. 4, 2014 Orr, H. A. 1991. A test of Fisher’s theory of dominance. Proceedings of the National Academy of Sciences of the United States of America 88: 11413-11415. Popham, E. J. 1941. The variation in colour of certain species of Artocorisa (Hemiptera: Corixidae) and its significance. Proceedings of the Zoological Society London, Series A 111: 135-172. Rigaud, T., and P. Juchault. 1993. Conflict between feminizing sex ratio distorters and an autosomal masculinizing gene in the terrestrial isopod, Armadillidium vulgare Latr. Genetics 133: 247-252. Rousset, F., D. Bouchon, B. Pintureau, P. Juchault, and M. Solignac. 1992. Wolbachia endosymbionts responsible for various alterations of sexuality in arthropods. Proceedings of the Royal Society London, Series B: Biological Sciences 250: 91-98. Sandoval, C. P. 1994. Differential visual predation on morphs of Timema cristinae (Phasmatodeae: Timemidae) and its consequences for host range. Biological Journal of the Linnean Society 52: 341-356. Sassaman, C. 1978. Dynamics of a lactate dehydrogenase polymorphism in the wood louse Porcellio scaber Latr.: evidence for partial assortative mating and heterosis in natural populations. Genetics 88: 591-609. Shuster, S. M. 1987. The reproductive biology of Paracerceis sculpta (Crustacea: Isopoda). Unpublished Ph.D. Dissertation, University of California, Berkeley, CA, 254 pp. . 1989. Male alternative reproductive behaviors in a marine isopod crustacean (Paracerceis sculpta): The use of genetic markers to measure differences in fertilization success among α-, β- and γ -males. Evolution 34: 1683-1698. . 1991. Changes in female anatomy associated with the reproductive molt in Paracerceis sculpta (Holmes), a semelparous isopod crustacean. Journal of Zoology 225: 365-379. , J. O. W. Ballard, G. Zinser, C. Sassaman, and P. Keim. 2001. The influence of genetic and extrachromosomal factors on population sex ratio in Paracerceis sculpta. Crustacean Issues 13: 313-326. , and E. E. Guthrie. 1999. The effects of temperature and food availability on adult body size in natural and laboratory populations of Paracerceis sculpta, a Gulf of California isopod. Journal of Experimental Marine Biology and Ecology 233: 269-284. , and L. Levy. 1999. Sex-linked inheritance of a cuticular pigmentation marker in a marine isopod, Paracerceis sculpta Holmes (Crustacea: Isopoda: Sphaeromatidae). Journal of Heredity 90: 305-307. , and C. Sassaman. 1997. Genetic interaction between male mating strategy and sex ratio in a marine isopod. Nature 388: 373-376. Sokal, R. R., and J. F. Rohlf. 1995. Biometry. 3rd Edition. W. H. Freeman, San Francisco, CA. Tinturier-Hamelin, E. 1963. Polychromatisme et détermination génétique du sexe chez l’espèce polytypique Idotea balthica (Pallas) (Isopode Valvifère). Cahiers de Biologie Marine 4: 473-591. Tomaszkiewicz, M., K. Smolarz, and M. Wołowicz. 2010. Heterogamety in the Baltic glacial relict, Saduria entomon (Isopod: Valvifera). Journal of Crustacean Biology 30: 758-762. Wright, S. 1929. Fisher’s theory of dominance. American Naturalist 63: 274-279. R ECEIVED: 7 February 2014. ACCEPTED: 15 March 2014.