Utah State University

From the SelectedWorks of Karen H. Beard

2010

Genetic Basis of a Color Pattern in the Frog Eleutherodactylus coqui Karen H. Beard, Utah State University

Available at: https://works.bepress.com/karenh_beard/28/ Journal of Heredity 2010:101(6):703–709 Ó The American Genetic Association. 2010. All rights reserved. doi:10.1093/jhered/esq082 For permissions, please email: [email protected]. Advance Access publication July 19, 2010 Genetic Basis of a Color Pattern Polymorphism in the Coqui Frog

Eleutherodactylus coqui Downloaded from https://academic.oup.com/jhered/article/101/6/703/1032960 by Utah State University Libraries user on 29 September 2020

ERIC M. O’NEILL AND KAREN H. BEARD

From the Department of Biology, Utah State University, Logan, UT 84322-5305 (O’Neill); and the Department of Wildland Resources and the Ecology Center, Utah State University, Logan, UT (Beard). O’Neill is now at the Department of Biology, University of Kentucky, 101 Thomas Hunt Morgan Building, Lexington, KY 40506-0225.

Address correspondence to Eric M. O’Neill at the address above, or e-mail: [email protected].

Abstract Many species of frog exhibit striking color and pattern polymorphisms, but the genetic bases of these traits are not known for most species. The coqui frog, Eleutherodactylus coqui, a species endemic to the island of Puerto Rico, exhibits a wide variety of color and pattern polymorphisms including 4 discrete stripe patterns on its dorsal surface and an unstriped morph. We conducted breeding experiments to determine the mode of inheritance for these 5 dorsal color patterns in E. coqui.We analyzed results from 14 different cross types, which included 1519 offspring from 71 clutches. We found that color patterns segregate at ratios consistent with a single autosomal locus, 5-allele model, in which all alleles coding for stripes are codominant and the allele coding for the unstriped morph is recessive. We propose that this locus be named ‘‘stripes’’ with alleles B (interocular bar), L (dorsolateral stripes), N (narrow middorsal stripe), W (wide middorsal stripe), and u (unstriped). The results of this experiment suggest the genetic basis of stripe patterns in this well-studied species and provide a model for studying the evolution and maintenance of this phenotypic polymorphism. Key words: Amphibian, codominance, Mendelian inheritance, Puerto Rico

Polymorphism is the occurrence of multiple discontinuous polymorphisms has been investigated in at least 28 species phenotypes within a single interbreeding population (Mayr (Hoffman and Blouin 2000) but conclusively demonstrated 1963). Visible polymorphisms have attracted much attention in only 2, Discoglossus pictus (Lantz 1947) and Rana pipiens from evolutionary biologists because they are easily obser- (Volpe 1956, 1961; Anderson and Volpe 1958). Single ved and can be used to study the fundamental processes that generation crosses suggest that in some species background maintain genetic variation (Ford 1975; Gray and McKinnon colors are genetically determined (e.g., Fogleman et al. 1980; 2007), especially when their genetic architecture is known. Blouin 1989; Summers et al. 2004), whereas in other species, For example, Gillespie and Tabashnik (1989) and Oxford there is some element of environmental control (e.g., Wente and Gillespie (1996) used breeding experiments to deter- and Phillips 2005). Similarly, in some species, color patterns, mine the mode of inheritance for a color pattern polymor- such as stripes in Eleutherodactylus (Goin 1950, 1960) and phism in the Hawaiian , grallator. Gillespie melanistic patterns in Dendrobates (Summers et al. 2004), and Oxford (1998) then used this model of inheritance to exhibit simple Mendelian inheritance, whereas patterns in demonstrate that the polymorphism is maintained by other species, such as spots in R. pipiens, are partly deter- balancing selection within populations rather than gene mined by environmental factors (Davidson 1964). flow among locally adapted populations. Genetic architec- Eleutherodactylus coqui is a diploid (Bogart 1981) frog ture for most phenotypic polymorphisms is not known, but endemic to Puerto Rico (Rivero 1978). This species exhibits it is likely that understanding the inheritance of these traits considerable variation in background color and both spot will shed light on the roles of different evolutionary and stripe patterns, and a large number of morphs have processes in maintaining them. been described (e.g., Schwartz and Henderson 1991; Joglar Frogs exhibit a wide array of color and pattern 1998). Woolbright (2005) simplified this variation into 6 polymorphisms including variation in background color morphs including 4 striped morphs, one spotted morph, and and presence or absence of stripes and spots (Hoffman and one morph lacking both stripes and spots. This poly- Blouin 2000). The mode of inheritance for these color-based morphism appears to be maintained, at least in part, by

703 Journal of Heredity 2010:101(6) Downloaded from https://academic.oup.com/jhered/article/101/6/703/1032960 by Utah State University Libraries user on 29 September 2020

Figure 1. Stripe patterns in Eleutherodactylus coqui showing (A) interoccular bar, (B) dorsolateral stripes, (C) wide middorsal stripe, (D) narrow middorsal stripe, (E) unstriped, and (F) a combination of dorsolateral stripes and a narrow middorsal stripe. selection from visual predators and habitat matching frogs were scored for stripe patterns and individually marked (Woolbright and Stewart 2008). Although these morphs using a standard toe-clipping method. Terraria included 1–2 appear to be highly heritable, the specific genetic architec- cm of moist sphagnum moss, two 10 cm lengths of PVC ture is not known. Understanding the genetic architecture pipe (diameters: 2.54 and 3.81 cm), half of a 0.47-l plastic underlying these traits would assist further investigation into cup (cut lengthwise), and a potted plant (Pothos sp.). Relative the evolutionary processes affecting this striking poly- humidity inside the terraria was maintained at levels greater morphism. Our objective was to determine the mode of than 95%, temperature at 25 °C, and the photoperiod was inheritance including the number of loci, number of alleles, maintained at a constant 12:12 h light:dark. Frogs were fed hierarchy, and autosomal versus for vitamin-dusted crickets ad libitum. stripe patterns found in E. coqui. Terraria were checked daily for eggs, and each clutch was removed and placed in a petri dish (95 mm diameter) on Materials and Methods moist paper towel. When clutches were removed, stripe patterns of adults were confirmed. Petri dishes were watered Frogs were collected from 4 populations in Puerto Rico (El and checked for hatching eggs every 2 days. Infertile eggs or Yunque Low: lat 18°20#01N, long 65°45#38W; El Yunque eggs showing evidence of fungal infection were removed High: lat 18°17#54N, long 65°47#15W; Rio Abajo Low: lat from clutches. Five to seven days after hatching, juvenile 18°21#28N, long 66°41#02W; Rio Abajo High lat frogs were removed to individual petri dishes (95 mm 18°12#59N, long 66°44#51W) in May 2006 and sent to diameter) lined with moist paper towel and sphagnum moss a laboratory at Utah State University. Controlled breeding and fed Collembola ad libitum. Color patterns of offspring experiments were conducted to determine the mode of were scored at 1 week from the hatching date and confirmed inheritance for stripe patterns and the unstriped morph. for each after one month. Adults from the same population were established in mixed The 4 striped and single unstriped color patterns include: pairs and housed in half of a 37.85-l terrarium using a light colored stripe between the eyes (B 5 interocular bar); corrugated plastic board as a divider. Prior to the breeding, a light hairline stripe extending from the tip of the snout to

704 O’Neill and Beard  Genetics of Color Patterns in Coqui Frogs

the vent that branches to continue along the leg to the foot P (N 5 narrow middorsal stripe); a wide stripe extending from the tip of the snout to the vent, not branching and not

2 continuing to the foot (W 5 wide middorsal stripe); 2 light v stripes extending from the snout or eye to the insertions of the hind limbs (L 5 dorsolateral stripes); and a solid or mottled design with no noticeably strong clear stripes (U 5

unstriped; Figure 1). Throughout the text, the same letters Downloaded from https://academic.oup.com/jhered/article/101/6/703/1032960 by Utah State University Libraries user on 29 September 2020 are used to abbreviate phenotypes and their alleles, but alleles are written in italics. The spotted pattern described by Woolbright (2005) and Woolbright and Stewart (2008) was not included in this study because it shows continuous variation and therefore may be a quantitative trait. 1:0 1:0 NA NA 1:1 1:1.02 0.008 0.93 0:1 0:1 NA NA 1:1 1:1.12 0.34 0.56 1:1 1:1.13 0.37 0.54 1:1 1:1.06 0.06 0.81 3:13:1 3.2:1 3.02:1 0.055 0.001Unstriped 0.82 0.97 frogs were crossed with other unstriped frogs Expected ratio Observed ratio 1:1:1:1 1:0.9:0.7:0.8 1.60 0.66 1:1:1:1 1:0.8:0.9:0.7 1.11 0.77 1:1:1:1 1:0.6:0.8:0.6 1.69 0.64 and with single striped frogs to determine the dominance L

L hierarchy between the unstriped pattern and each stripe L 5 5 pattern. Frogs with single stripe patterns were crossed 5 u u u u u L N W B N with frogs with single stripe patterns to further test the . . . . . , , , , , dominance hierarchy and develop a working model of Dominance hierarchy inheritance of all patterns. Finally, unstriped frogs were crossed with 2-patterned frogs to test the simplest model of Lu u Nu N uu u NN N Wu W Bu B Lu L Lu u Lu u Lu u Nu u inheritance against several more complex alternative models            including multiple loci and sex linkage. Parental genotypes uu uu uu uu uu uu Bu Lu Nu Nu Wu The heterogametic sex has not been identified in E. coqui; therefore, we tested for sex linkage using 2 alternative models of male and female heterogamy (XY and ZW) that Total are common among anurans (Hillis and Green 1990). Only linkage to the homogametic (X or Z) chromosome was tested for each of these models because the presence of each stripe pattern in members of both sexes (O’Neill E, personal observation) allowed us to reject a model of linkage to the heterogametic chromosome (Y or W). To compare the results of our crosses with expected values under alternative models of inheritance, chi-square tests for goodness-of-fit were used in SAS 9.1 (SAS Institute, Cary, NC). When expected values were zero and observed values were nonzero, no statistical tests were necessary to reject the null hypothesis. For example, if the model predicted there should be no offspring with stripes, but we observed offspring with stripes, the model could not be true. Some crosses were performed using multiple pairs, in these cases the different families did not differ in their observed offspring ratios; phenotypes 1 F UNWBLNLWLBL therefore, these families were combined for all analyses. All P values from the chi-squared tests are presented in the tables.

No. of clutches Results A total of 71 clutches yielded 1519 offspring from 14 different cross types. Stripe patterns scored after 1 week of No. of families hatching did not change after one month. Patterns N and W were visible at hatching, but B and L were not fully visible

L 1 1until 16 about — 121 week — 14 after — hatching. 11 — Patterns 53 in offspring were L1 1 95——75——26 N 7 14 60 181 — — — — — — 241 N 5 5 63 64 — — — — — — 127 U 12 17 368 — — — — — — — 368 N 3 8 0 158 — — — — — — 158 W 3 7 56 — 50 — — — — — 106 B2 4 46——52————98 L 3 3 36 — — — 34 — — — 70 L23 20——1813——1667 L 2 2 13 — — — 42 — — — 55          similar to those in adults suggesting that there is little   Results from crosses testing the dominance hierarchy between the unstriped phenotype (U) and the 4 striped phenotypes (B, L, N, and W)

Parental phenotypes ontogenetic change in these phenotypes after the first week. Two percent of the offspring exhibited 2 of the 4 striped patterns simultaneously but never more than 2 stripe 2U 1U 3U 4U 5U 6U 8W 7B 9N Table 1 Genotypes were assigned to all parents using the proposed model and segregation ratios of offspring. Cross no. 10 L 11 N patterns were observed on a single frog.

705 Journal of Heredity 2010:101(6)

Crosses 1–9 tested the dominance hierarchy between U and the 4 stripe patterns: B, L, N, and W (Table 1). Cross 1, 0.0001 0.0001 0.0001 0.0001 0.0001 , , , , , P between 2 unstriped parents (U Â U), resulted in all unstriped offspring, which suggests that either the allele for

2 U is recessive and both parents were homozygous or that v the allele for U is dominant and at least one parent was homozygous. Crosses 2–6, between unstriped (U) and tripe) differently striped frogs (N, W, W, B, and L), resulted in Downloaded from https://academic.oup.com/jhered/article/101/6/703/1032960 by Utah State University Libraries user on 29 September 2020 some offspring, within each clutch, with the parental patterns at ratios that were not different from 1:1 or 0:1

Observed phenotypic ratio (Table 1), which supports the hypothesis that the allele for unstriped is recessive to each allele for stripes and that all parents in cross 1 were homozygous recessive. Finally, and critically, crosses 7–11, between 2 striped parents resulted in some unstriped offspring, which also suggests that the allele 1:1 1:1.03 0.419 0.5176 for unstriped is recessive to all alleles for striped patterns. Expected phenotypic ratio 1:1:1:10:0:0:10:1:0:11:0:0:1 1:1.03:0:0 1:1.03:0:0 1:1.03:0:0 1:1.03:0:0 29.534 NA NA NA NA NA NA 0:1:0:1 1:1.03:0:0 NA NA 1:1:1:10:0:0:11:0:0:10:1:0:11:1:1:1 1:1.03:0:00:0:0:1 1:1.03:0:00:1:0:1 1:1.03:0:01:0:0:1 1:1.03:0:0 29.534 1:1.03:0:0 NA 1:1.03:0:0 NA NA 1:1.03:0:0 NA NA 1:1.03:0:0 29.534 NA NA NA NA NA NA NA 1:1:1:11:1:1:1 1:1.03:0:0 1:1.03:0:0 29.534 29.534 1:0:0:1 1:1.03:0:0Crosses NA 7–9 NA (B Â L, W Â L, and N Â L, respectively) L u N u u L u N W X X resulted in offspring ratios that were not different from W X X W W X X X X L L L N N N L L L N N N Z X X

Z a 1:1:1:1 ratio, and included offspring with no stripes, stripe Z X X Z X X X X u N u L u N u N u L u L A A A A A A A A F patterns from one parent, and stripe patterns from both A A A A L L N N N N N N L L L L Â A A A A

A A A A A A A A parents (Table 1). The simplest explanation for these ratios             is a single-locus model with stripe patterns coded by u u u u LuNu LLNN LuNN LLNu Y Y Y Y Y Y Y Y Z Z Z Z u u u u u u u u

u u u u codominant alleles and the unstriped pattern coded by LN X X X X X X X X Z Z Z Z     u u u u u u u u u u u u a recessive allele. Alternatively, a 2-locus model with each  A A A A A A A A A A A A u u u u u u u u u u u u parent heterozygous at one locus and homozygous recessive Putative parental genotypes M uu uuuu uuuu uuuu uuuu A A A A A A A A A A A A at the other locus may also explain these results. Crosses 12–14 (U  NL, U  BN, and LW  U, respectively) tested the single versus 2-locus model, in- cluding autosomal versus sex linkage, for three 2-pattern combinations (Tables 2–4). All 3 crosses resulted in is autosomal, is X-linked is Z-linked is autosomal, is X-linked is autosomal, is autosomal, is Z-linked Autosomal versus sex-linked N L L N L N N L offspring ratios that were not different from 1:1 for each parental stripe pattern (Tables 2–4), which was predicted by the single-locus model. These ratios were inconsistent with a 2-locus model, either with both loci autosomal or one locus as sex linked (Tables 2–4). Hetero- gametic sex Male XY Female ZW Although the above crosses did not directly test for the allelic relationships between combinations W and B, L, and No. of Loci 2 NA Both autosomal B, or N and W because frogs were not available to test these combinations, the conclusion that these patterns are coded by different alleles at a single locus can be deduced from the results of other crosses. For example, cross 12 suggests that L and N are alleles the same locus, and cross 13 suggests that N and B are alleles at the same locus; therefore, L and B should also be alleles at the same locus. phenotypes 1 F L N U LN Total Discussion Our crosses suggest that dorsal color pattern variation in E. No. of clutches

F coqui is genetically determined and likely the result of a single

 autosomal locus with 4 codominant alleles coding for various stripe patterns and one universally recessive allele coding for the unstriped pattern. All patterns were found on frogs of both

LN 2 40 46 0 0 86 1 NAsexes, Autosomal further supporting the hypothesis that the locus for Results from crosses testing the number of loci and sex linkage for the phenotypes: U (unstriped), L (dorsolateral stripes), and N (narrow middorsal s  stripe patterns is not linked to a heterogametic chromosome. Parental phenotypes M We propose that this locus be named ‘‘stripes’’ with alleles B (interoccular bar), L (dorsolateral stripes), N (narrow mid- Cross no. Table 2 12 U dorsal stripe), W (wide middorsal stripe), and u (unstriped).

706 O’Neill and Beard  Genetics of Color Patterns in Coqui Frogs

Although some of our crosses had relatively small sample sizes, we had sufficient power to reject multiple 0.0001 0.0001 0.0001 0.0001 0.0001 , , , , P , alternative models that were testable with our data. Furthermore, sample sizes for tests in this study were 2

v comparable with those in similar studies (e.g., Oxford and Gillespie 1996; Summers et al. 2004). Further support for the single autosomal locus model is that no cross produced e) any offspring that deviated from the proposed model (e.g., Downloaded from https://academic.oup.com/jhered/article/101/6/703/1032960 by Utah State University Libraries user on 29 September 2020 an individual with 3 stripe patterns or a striped individual from a U Â U cross), and frogs with more than 2 different

Observed phenotypic ratio stripe patterns have never been found in the field (O’Neill E, personal observation; Woolbright and Stewart 2008; Peacock et al. 2009). Other alternative models (e.g., epistasis) were not testable as null hypotheses because these did not provide specific expected offspring ratios. Addi- 1:1 1:0.7 0.67 0.4100 tionally, the possibility of close linkage between multiple loci Expected phenotypic ratio 1:0:0:1 1:0.7:0:0 NA NA 1:1:1:11:1:1:1 1:0.7:0:0 1:0.7:0:0 13.33 13.33 0:1:0:1 1:0.7:0:0 NA NA 1:1:1:10:0:0:11:0:0:10:1:0:11:1:1:1 1:0.7:0:00:0:0:1 1:0.7:0:00:1:0:1 1:0.7:0:01:0:0:1 1:0.7:0:0 13.33 1:0.7:0:0 NA 1:0.7:0:0 NA NA 1:0.7:0:0 NA NA 1:0.7:0:0 13.33 NA NA NA NA NA NA NA 1:1:1:1 1:0.7:0:0 13.33 0:0:0:1 1:0.7:0:0 NA NA 0:1:0:1 1:0.7:0:0 NA NA 1:0:0:1 1:0.7:0:0 NA NA remains, but we could not reject the simpler single-locus B u N u u B u N W X X W X X W W X X X X model based on the available data. B B B B B B N N N N N N Z X X Z Z X X Z X X X X u N Dorsal stripe pattern variation is found in at least 80 u N u N u B u B u B A A A A A A F A A A A A A

B B N N other species of Eleutherodactylus (Hoffman and Blouin N N N N B B B B  A A A A A A A A A A A A 2000). Of these, the number of loci coding for different            Â

u u u u stripe patterns has been investigated in only one other BuNu BBNN BuNN BBNu Y Y Y Y Y Y Y Y Z Z Z Z u u u u u u u u u u u u

BN species, E. nubicola (Goin 1960). Goin (1960) suggested, X X X X X X X X Z Z Z Z Â Â Â Â u u u u u u u u u u u u Â

A A A A A A A A A A A A from a single clutch collected from a wild-mated female, u u u u u u u u u u u u Putative parental genotypes M A A A A A A A A A A A uu uuuu uuuu uuuu uuuu A that dorsolateral stripe (similar to our L) and middorsal stripes (similar to our N) were coded by separate loci in E. nubicola. This suggests that similar stripe patterns may not be homologous among closely related species of Eleutherodactylus. Further studies are necessary to is autosomal, is X-linked is Z-linked is autosomal, is Z-linked is autosomal, is autosomal, is X-linked determine to what extent the stripe patterns in this genus Autosomal versus sex-linked N B B N B N N B are inherited from a common ancestor representing homologous parallel evolution (Bull 1975)ortheresult of convergent evolution. The dominance hierarchy found in this study, alleles Hetero- gametic sex Female ZW Male XY coding for dorsal stripes dominant to unstriped, is similar to that of other amphibians with similar stripe patterns including frogs of the genera Acris (Pyburn 1961a), No. of Loci 2 NA Both autosomal Discoglossus (Lantz 1947), Eleutherodactylus (Goin 1947, 1950, 1960), Rana (Moriwaki 1953; Browder et al. 1966; Ishchenko and Schupak 1974) and as well as in the salamander, Plethodon cinereus (Highton 1959). To our knowledge, there are no studies in which the stripe patterns of amphibians have been shown to be recessive to unstriped patterns. Two of the stripe patterns (B and L) exhibited some ontogenetic change during the first week after hatch but phenotypes 1 F B N U BN Total showed no noticeable change over the next month. Subtle changes of background and stripe color sometimes occurred but these did not affect the scoring of phenotypes

No. of clutches (O’Neill E, personal observation). Ontogenetic change in

F similar stripe patterns has not been reported in other frogs

 (reviewed in Hoffman and Blouin 2000), but changes in stripe color between juvenile and adult stages do occur in other frogs (e.g., Pyburn 1961b; Gray 1972). Stripe patterns

BN 1 14 10 0 0 24 1 NAare not Autosomal sexually dimorphic in E. coqui. Most stripe patterns Results from crosses testing the number of loci and sex linkage for the phenotypes: U (unstriped), B (interoccular bar), and N (narrow middorsal strip  in other species of frog are not sexually dimorphic Parental phenotypes M (Hoffman and Blouin 2000), but sexual dimorphism of dorsolateral stripes occurs in Hyla bokermanni and Cross no. Table 3 13 U H. luteocellata (Rivero 1969).

707 Journal of Heredity 2010:101(6)

In natural populations of E. coqui, striped frogs are less common compared with unstriped frogs (Woolbright and 0.0001 0.0001 0.0001 0.0001 0.0001 , , , , P , Stewart 2008). This inverse relationship between the domi- nance hierarchy and phenotypic frequency has been found 2

v in fish (Winge 1927), locusts (Haldane 1930), snails (Fisher 1930), (Oxford and Gillespie 1998), and 6 species of frog (Hoffman and Blouin 2000) and is predicted by some Downloaded from https://academic.oup.com/jhered/article/101/6/703/1032960 by Utah State University Libraries user on 29 September 2020 ipe) models of frequency-dependent selection (Clarke 1964; Clarke and O’Donald 1964). Whether or not frequency-dependent selection is responsible for maintaining the stripe polymor- Observed phenotypic ratio phism in E. coqui is not known, but some form of selection appears to be involved (Woolbright and Stewart 2008). The stripe pattern polymorphism described here for E. coqui is easily scored in the field and appears to be inherited

1:1 1:1.4 0.9 0.3428 in a simple Mendelian fashion. Therefore it should be valuable Expected phenotypic ratio 1:0:0:1 1:1.4:0:0 NA NA 1:1:1:11:1:1:1 1:1.4:0:0 41.8 1:1.4:0:0 41.8 1:0:0:11:0:0:1 1:1.4:0:0 1:1.4:0:0 NA NA NA NA 1:1:1:11:0:0:11:1:1:10:1:0:1 1:1.4:0:0 1:1.4:0:0 1:1.4:0:0 1:1.4:0:0 41.8 NA NA 41.8 NA NA 0:1:0:1 1:1.4:0:00:1:0:1 NA NA 1:1.4:0:0 NA NA 0:0:0:10:0:0:1 1:1.4:0:0 NA 1:1.4:0:0 NA NA NA 1:1:1:10:1:0:1 1:1.4:0:0 1:1.4:0:0 41.8 NA NA 0:0:0:1 1:1.4:0:0 NA NA as a readily observable genetic marker for future studies of u

u population genetics in Puerto Rico and for introduced u u W W W W X u u W W X u u u W W X X u u u u u u u Z Z populations in areas where E. coqui has invaded (e.g., Hawaii, Z Z X u u Z Z X u u Z Z u X X u u u u u u u A A A A A u u A A see Velo-Anto´n et al. 2007; Beard et al. 2009). Because stripe A u u A A u A A u u u u u F u u A A A A A A A A A A patterns are thought to be maintained by selection in Puerto A A Â Â Â Â Â Â Â Â Â Â Â Â Â

uuuu Rico (Woolbright and Stewart 2008), they will be of interest to L u W u uuuu u W u L uuuu Y Z Z uuuu Y Z Z Y Y Z Z Z Z L  L L those studying potential effects of recent human-mediated W L  W W L L W uu  W W X Z Z  X X Z Z Z Z X Z Z u W u W u W

u L u L u L introductions on the ecology and evolution of E. coqui. For  A A A A A A A A A A A A

L L W W L L L L W W W W example, this polymorphism may be useful for studying the Putative parental genotypes M A A A A A A LLWu A A A A A A LW LuWu LLWW LuWW roles of adaptive and nonadaptive processes in the mainte- nance of genetic variation in native and introduced populations (e.g., Eckert et al. 1996; Hoffman et al. 2006). is X-linked is autosomal, is autosomal, is Z-linked is autosomal, is Z-linked is X-linked is autosomal, Funding Autosomal versus sex-linked L W W L W L L W Jack H. Berryman Institute; United States Department of Agriculture (USDA) and Plant Health Inspection Service (APHIS) Wildlife Service (WS) Hilo Field Station. Hetero- gametic sex Male XY Female ZW Acknowledgments No. of loci 2 NA Both autosomal Permits were provided by the Puerto Rico Departemento de Recourses y Naturales (Permit number: 06-IC-019) and USU IACUC (no. 1145 and 1251). We thank E. D. Brodie Jr for the use of laboratory space and supplies, and M. Pfrender and P. Wolf for discussions about inheritance. We thank L. Giovanetto, J. Poulos, and A. Huff who assisted with animal collection, G. Jones, M. Cooke, E. Lytle, Y. Kajita, K. Bakkegard, L. Latta, and K. Latta who assisted with animal husbandry, and 2 anonymous reviewers for helpful suggestions and comments. phenotypes 1 L W U LW Total F References Anderson SC, Volpe EP. 1958. Burnsi and kandiyohi genes in the leopard frog Rana pipiens. Science. 127:1048–1049.

No. of clutches Beard KH, Price EA, Pitt WC. 2009. Biology and impacts of Pacific Island

F invasive species: Eleutherodactylus coqui, the Coqui frog (Anura: Leptodacty-

 lidae). Pac Sci. 63:297–316. Blouin MS. 1989. Inheritance of a naturally occurring color polymorphism in the ornate chorus frog, Pseudacris ornata. Copeia. 1989:1056–1059. U 1 17 23 0 0 40 1 NA Autosomal

 Bogart JP. 1981. Chromosome studies in Sminthillus from Cuba and Results from crosses testing the number of loci and sex linkage for the phenotypes: U (unstriped), L (dorsolateral stripes), and W (wide middorsal str Eleutherodactylus from Cuba and Puerto Rico (Anura, Leptodactylidae). Life Parental phenotypes M Sci Contrib R Ontario Museum. 129:1–22. Browder LW, Underhill JC, Merrel JC. 1966. Mid-dorsal stripe in the wood Table 4 Cross no. 14 LW frog. J Hered. 57:65–67.

708 O’Neill and Beard  Genetics of Color Patterns in Coqui Frogs

Bull MC. 1975. Parallel polymorphism in Australian frogs of the genus Mayr E. 1963. Animal species and evolution. Cambridge (MA): Belknap Ranidella. Heredity. 35:273–278. Press of Harvard University Press. Clarke B. 1964. Frequency dependent selection for the dominance of rare Moriwaki T. 1953. The inheritance of the dorso-lateral median stripe in polymorphic genes. Evolution. 18:364–369. Rana limnocharis Weigmann. J Sci Hiroshima Univ (Zool). 14:159–164. Clarke B, O’Donald P. 1964. Frequency dependent selection. Heredity. Oxford GS, Gillespie RG. 1996. Genetics of a colour polymorphism in 19:201–206. Theridion grallator (Araneae, ), the Hawaiian happy-face spider, Davidson J. 1964. A study of spotting patterns in the leopard frog. III. from greater . Heredity. 76:238–248.

Environmental control of genic expression. J Hered. 55:47–56. Oxford GS, Gillespie RG. 1998. Evolution and ecology of spider Downloaded from https://academic.oup.com/jhered/article/101/6/703/1032960 by Utah State University Libraries user on 29 September 2020 Eckert CG, Manicacci D, Barrett SCH. 1996. Genetic drift and founder coloration. Ann Rev Entomol. 43:619–643. effects in natural versus introduced populations of an invasive plant Lythrum Peacock MM, Beard KH, O’Neill EM, Kirchoff V, Peters MB. 2009. Strong salicaria. Evolution. 50:512–519. founder effects and low genetic diversity in introduced populations of Fisher RA. 1930. The evolution of dominance in certain polymorphic Coqui frogs. Mol Ecol. 18:3603–3615. species. Am Nat. 64:385–406. Pyburn WF. 1961a. Inheritance of the green vertebral stripe in Acris Fogleman JC, Corn PS, Pettus D. 1980. The genetic basis of a dorsal color crepitans. Southwestern Nat. 6:164–167. polymorphism in Rana pipiens. J Hered. 71:439–440. Pyburn WF. 1961b. The inheritance and distribution of vertebral stripe Ford EB. 1975. Ecological genetics. London: Chapman and Hall. color in the cricket frog. In: Blair WF, editor. Vertebrate Speciation. Austin (TX): University of Texas Press. p. 235–261. Gillespie RG, Oxford GS. 1998. Selection on the color polymorphism in Hawaiian happy-face spiders: evidence from genetic structure and temporal Rivero JA. 1969. On the identity and relationships of Hyla luteocellata. fluctuations. Evolution. 52:775–783. Herpetologica. 25:126–134. Gillespie RG, Tabashnik BE. 1989. What makes a happy face? Rivero JA. 1978. Los anfibios y reptiles de Puerto Rico. San Juan (PR): Determinants of colour pattern in the Hawaiian happy face spider Theridion Universidad de Puerto Rico Editorial Universitaria. grallator (Araneae, Theridiidae). Heredity. 62:355–363. Schwartz A, Henderson RW. 1991. Amphibians and reptiles of the West Goin CJ. 1947. Studies in the life history of Eleutherodactylus ricordii planirostris Indies: descriptions, distributions, and natural history. Gainesville (FL): (Cope) in Florida. Gainesville (FL): University of Florida Press. University of Florida Press. Goin CJ. 1950. Color pattern inheritance in some frogs of the genus Summers K, Cronin TW, Kennedy T. 2004. Cross-breeding of distinct Eleutherodactylus. Bull Chic Acad Sci. 9:1–15. color morphs of the strawberry poison frog (Dendrobates pumilio) from the Bocas del Toro Archipelago, Panama. J Herpetol. 38:1–8. Goin CJ. 1960. Pattern variation in the frog Eleutherodactylus nubicola Dunn. Bull Fla State Mus Biol Sci. 5:243–258. Velo-Anto´n G, Burrowes PA, Joglar RL, Martinez-Solano I, Beard KH, Parra-Olea G. 2007. Phylogenetic study of Eleutherodactylus coqui (Anura: Acris Gray RH. 1972. Metachrosis of vertebral stripe in Cricket Frog, Leptodactylidae) reveals deep genetic fragmentation in Puerto Rico and crepitans . Am Midl Nat. 87:549–551. pinpoints origins of Hawaiian populations. Mol Phylogenet Evol. Gray SM, McKinnon JS. 2007. Linking color polymorphism maintenance 45:716–728. and speciation. Trends Ecol Evol. 22:71–79. Volpe EP. 1956. Mutant color patterns in leopard frogs, a possible allelic Haldane JBS. 1930. A note on Fisher’s theory of the origin of dominance series. J Hered. 47:79–85. and on correlation between dominance and linkage. Am Nat. 64:87–90. Volpe EP. 1961. Variable expressivity of a mutant gene in the leopard frog. Highton R. 1959. The inheritance of the color phases of Plethodon cinereus. Science. 134:102–104. Copiea. 1959:33–37. Wente WH, Phillips JB. 2005. Seasonal color change in a population of Hillis DM, Green DM. 1990. Evolutionary changes of heterogametic sex in pacific treefrogs (Pseudacris regilla). J Herpetol. 39:161–165. the phylogenetic history of amphibians. J Evol Biol. 3:49–64. Winge O. 1927. The location of eighteen genes in Lebistes reticulatus. J Genet. Hoffman EA, Blouin MS. 2000. A review of colour and pattern 18:1–43. polymorphisms in anurans. Biol J Linn Soc. 70:633–665. Woolbright LL. 2005. A plot based system of collecting population Hoffman EA, Schueler FW, Jones AG, Blouin MS. 2006. An analysis of information on terrestrial breeding frogs. Herpetol Rev. 36:139–142. selection on a colour polymorphism in the northern leopard frog. Mol Ecol. Woolbright LL, Stewart MM. 2008. Spatial and temporal variation in color 15:2627–2641. pattern morphology in the tropical frog, Eleutherodactylus coqui. Copeia. Ishchenko VG, Shchupak EL. 1974. Ecological differences of individual 2008:431–437. genotypes in a frog population. Sov J Ecol. 5:379–380. Joglar RL. 1998. Los coquı´es de Puerto Rico: su historia natural y Received March 20, 2010; Revised March 20, 2010; conservatio´n. San Juan (PR): Universidad de Puerto Rico. Accepted June 10, 2010 Lantz LA. 1947. Note (appendix to HM Bruce and AS Parkes, observations of Discoglossus pictus Otth.). Proc R Soc Lond B Biol Sci. 134:52–58. Corresponding Editor: William Modi

709