CSIRO PUBLISHING Australian Journal of Zoology http://dx.doi.org/10.1071/ZO14076 Reconstructed paternal genotypes reveal variable rates of multiple paternity at three rookeries of loggerhead sea turtles (Caretta caretta) in Western Australia J. N. Tedeschi A,B,F, N. J. Mitchell A,B, O. Berry C, S. Whiting D, M. Meekan B,E and W. J. Kennington A A School of Animal Biology (M092), The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. B Oceans Institute (M470), The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. C CSIRO Oceans and Atmosphere Flagship, PMB 5, Floreat, WA 6014, Australia. D Marine Science Program, Department of Parks and Wildlife, Locked Bag 104, Bentley Delivery Centre, WA 6983, Australia. E Australian Institute of Marine Science (M096), The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. F Corresponding author. Email: jamie.tedeschi@research.uwa.edu.au Abstract. Female sea turtles are promiscuous, with clutches of eggs often sired by multiple males and rates of multiple paternity varying greatly within and across species. We investigated levels of multiple paternity in loggerhead sea turtles (Caretta caretta) from three rookeries in Western Australia by analysing polymorphic species-specific genetic markers. We predicted that the level of multiple paternity would be related to female population size and hence the large rookery at Dirk Hartog Island would have higher rates of multiple paternity than two smaller mainland rookeries at Gnaraloo Bay and Bungelup Beach. Contrary to our prediction, we found highly variable rates of multiple paternity among the rookeries that we sampled, which was unrelated to female population size (25% at Bungelup Beach, 86% at Gnaraloo Bay, and 36% at Dirk Hartog Island). Approximately 45 different males sired 25 clutches and the average number of sires per clutch ranged from 1.2 to 2.1, depending on the rookery sampled. The variance in rates of multiple paternity among rookeries suggests that operational sex ratios are variable in Western Australia. Periodic monitoring would show whether the observed patterns of multiple paternity for these three rookeries are stable over time, and our data provide a baseline for detecting shifts in operational sex ratios. Received 9 September 2014, accepted 19 December 2014, published online 15 January 2015 Introduction Unlike many bird and mammal species, parental care beyond nesting is absent in most reptiles (Shine 2005; Uller and Olsson 2008). Males do not provide any resources to females other than sperm, yet multiple paternity in clutches has been recorded in most reptile species to date (Uller and Olsson 2008). Multiple paternity has been detected in all seven extant species of sea turtle, with one or two sires being the most common number for a single clutch (reviewed by Bowen and Karl 2007, and Lee 2008). In sea turtles, multiple paternity can arise in two ways: either a female can mate with more than one male during the same reproductive cycle or, alternatively, a female may utilise sperm stored from a previous breeding season (Pearse and Avise 2001; Lara-De La Cruz et al. 2010; Phillips et al. 2014a). Many explanations for multiple paternity have been proposed, including increased fertilisation success, improved Journal compilation
CSIRO 2015 offspring fitness, and harassment of receptive females by males (Jensen et al. 2013). Ireland et al. (2003) and Lee and Hays (2004) suggested that the phenomenon was a product of male density and avoidance of aggressive mating behaviour by females, causing females to mate with more than one male (convenience polyandry). A study on solitary and mass-nesting (arribada) Olive Ridley turtles (Lepidochelys olivacea) by Jensen et al. (2006) attributed the higher rate of multiple paternity in the arribada females to their high density of nesting. As males rarely come ashore and are difficult to catch at sea, genetic analyses of nesting females and their offspring can both identify the number of sires per clutch and provide data on the number of breeding males and females from which operational sex ratios (OSRs) can then be calculated (Wright et al. 2012a, 2012b; Hawkes et al. 2014). The OSR of a given population should be proportional to the number of males at the breeding www.publish.csiro.au/journals/ajz B Australian Journal of Zoology J. N. Tedeschi et al. third-largest population of C. caretta in the world (Baldwin et al. 2003; Reinhold and Whiting 2014), we know relatively little about the population demographics. A description of mating systems, quantification of the incidence of multiple paternity, and quantification of genetic variation is a first step towards understanding the implications of climate change and changing sex ratios of this globally important population. Genetic analyses offer a means to indirectly sample the male component of a population of breeding turtles (Lee 2008; Phillips et al. 2014b; Stewart and Dutton 2011) and there are several methods for estimating multiple paternity using genetic data (Table 1). For sea turtles, such studies show that rates of multiple paternity are highly variable (Table 1), though it is unclear whether the reported variability among species and area before the nesting season (Hays et al. 2010; Stewart and Dutton 2011), and therefore reflect the underlying genetic variation of the population. Relative to other parts of the world, little is known about the population dynamics of loggerhead turtles (Caretta caretta) nesting in the eastern Indian Ocean. In Australia, there are two genetically distinct populations of C. caretta, one in Western Australia and the other in Queensland (Baldwin et al. 2003). All rookeries in Western Australia comprise a single genetic stock (Pacioni et al. 2012), spanning ~520 km of coastline from Dirk Hartog Island (25.49827S, 112.98719E) at the southern limit to the Muiron Islands north-east of Exmouth (21.39156S, 114.21205E) at the northern limit of the range (Baldwin et al. 2003). Although the rookeries within this area constitute the Table 1. Variation in rates of multiple paternity in sea turtles within species and across studies MP, multiple paternity No. of clutches analysed Mean no. of offspring genotyped per clutch No. of loci analysed Minimum no. of males Frequency of MP Green turtle (C. mydas) Ascension Island Ascension Island Southern Great Barrier Reef Tortuguero, Costa Rica Algadi, Cyprus Algadi, Cyprus Sri Lanka 18 3 22 8 20 94 24 38.9 15.3 41.3 – 21.9 21.7 10 2–5 2 5 2 14 13 6 2 2 1 2 1.4 1.1 1.7 61% 100% 9% 63% 36% 23% 63% Loggerhead turtle (C. caretta) Zakynthos, Greece Melbourne Beach, Florida Mon Repos, Queensland Melbourne Beach, Florida Nagoya, JapanC 15 70 24 3 7 40.7 10 21 20.7 29 4 4 0 2 2 3.2 1.4 – – – Olive Ridley turtle (L. olivacea) Ostional, Costa RicaA Playa Hermosa, Costa RicaB Galibi, SurinameB 13 13 10 22 22.6 70.3 2 2 2 Kemp’s Ridley turtle (L. kempi) Tamaulipas, Mexico 26 7.8 Hawksbill turtle (E. imbricata) Gulisaan, Sabah, Malaysia Cousine Island, Seychelles Seychelles (various islands) 10 43 249 Leatherback turtle (D. coriacea) Las Baulas, Costa Rica Sandy Point, Virgin Islands Sandy Point, Virgin Islands Playa Grande, Costa Rica Flatback turtle (N. depressus) Mon Repos and Peak Island, Queensland A Methods Citation DADSHARE, GERUD REAP GENEPOP 3.1 Irwin COLONY 2.0 COLONY 2.0 GERUD 2.0 Lee and Hays (2004) Ireland et al. (2003) Fitzsimmons (1998) Peare and Parker (1996) Wright et al. (2012b) Wright et al. (2012a) Ekanayake et al. (2013) 93% 31% 33% 33% 43% GERUD 1.0 PARENTAGE Allozymes – – Zbinden et al. (2007) Moore and Ball (2002) Harry and Briscoe (1988) Bollmer et al. (1999) Sakaoka et al. (2011) 2.8 1.4 1.2 92% 30% 20% GERUD 1.0 GERUD 1.0 Initial inference Jensen et al. (2006) Jensen et al. (2006) Hoekert et al. (2002) 3 – 58% – Kichler et al. (1999) 27 18.8 22.6 3 33 32 1.3 – – 20% 9.3% 9.2% GERUD 1.0 COLONY 2.0 COLONY 2.0 Joseph and Shaw (2011) Phillips et al. (2013) Phillips et al. (2014b) 4 38 17 20 – 26.8 10.5 19.5 2 7 6 3 1 – – – 0% 42% 0% 10% – GERUD 1.0 – – Rieder et al. (1998) Stewart and Dutton (2011) Dutton et al. (2000) Crim et al. (2002) 16 26.7 4 – 69% Initial inference, Chi-square, PARENTAGE 1.0, GERUD 2.0, MER 3.0 Theissinger et al. (2009) Arribada nesting beach. Solitary nesting beach. C Captive population, paternal genotype known. B Variable rates of paternity in loggerheads Australian Journal of Zoology populations is due to the use of different types of genetic markers, differences in multiple paternity estimation methods, or if indeed it reflects natural variability among populations. To date, only one study (Jensen et al. 2006) has concurrently examined the frequency of multiple paternity in two different rookeries of the same species. Here, we investigated patterns of multiple paternity in clutches sampled from three locations spread across the geographic range of rookeries of C. caretta in Western Australia. The southern-most rookery was on Dirk Hartog Island (DHI), one of the world’s largest rookeries (Baldwin et al. 2003; Reinhold and Whiting 2014), while we also sampled clutches from near the northern-most edge (Bungelup Beach, BB) and from a smaller mainland rookery approximately midway in the breeding range (Gnaraloo Bay, GB). We aimed to describe: (1) the presence of multiple paternity, and (2) spatial variation in multiple paternity rates among rookeries across the range of the nesting population. Because these rookeries differed in size, we predicted that the frequency of multiple paternity should be higher in clutches from the larger nesting rookery at DHI compared with the smaller, mainland rookeries (GB and BB) C based on the density-dependence convenience polyandry model. To exclude the possibility that any variation in multiple paternity we detected was an artefact of methodology, we analysed paternity using identical statistical methods and the same genetic markers for samples from all rookeries. Our results are discussed in the context of estimating population size and the implications of climate change on the demography of the Western Australian population of C. caretta. Methods Egg collection and tissue sampling Eggs of C. caretta were collected from three rookeries in Western Australia during peak nesting periods between 2011 and 2013. Collection sites and dates were Turtle Bay on Dirk Hartog Island (25.49827S, 112.98719E) in January 2013, Gnaraloo Bay on the Western Australian mainland (23.82618S, 113.52629E) in January 2011 (Woolgar et al. 2013), and Bungelup Beach in the Cape Range National Park on the Exmouth Peninsula (22.282331S, 113.831570E) in December 2013 (Fig. 1). The Dirk Hartog Island rookery hosts the largest nesting numbers, Bungelup Beach 22.282331\S, 113.831570\E Tropic of Capricorn Gnaraloo Bay 23.82618\S, 113.52629\E N Indian Ocean 0 0.15 0.3 0.6 0.9 1.2 Kilometres TURTLE BAY BEA CH Cape Inscription 1 BEACH BEAC H2 CH 4 BEACH 3 BEA 5 Cape Levillain Sammys Dirk Hartog Island DIRK HARTOG ISLAND 25.49827\S, 112.98719\E N Kilometres 0 15 30 60 90 120 Fig. 1. Locations of the three collection sites: Dirk Hartog Island, Gnaraloo Bay, and Bungelup Beach. More than 2000 females nest per season on Dirk Hartog Island, most notably on Beach 1 and Beach 5. Map adapted from Trocini (2013), and Reinhold and Whiting (2014). D Australian Journal of Zoology with ~2000 nesting females per season (Trocini 2013; Reinhold and Whiting 2014), while an estimated 700–1200 females nest per season at Bungelup Beach (Trocini 2013). In contrast, nesting at the Gnaraloo Bay rookery is comparatively infrequent, with ~100 females per season (Hattingh et al. 2011). All offspring samples in this study were used opportunistically, as they were collected for other research projects (Woolgar et al. 2013; Tedeschi et al., unpubl. data). As a consequence, sample sizes varied among clutches and among rookeries. We had access to 80 eggs per clutch (n = 15 clutches) from the Dirk Hartog Island rookery, and 20 eggs per clutch (n = 4 clutches) from Bungelup Beach. Eggs sampled from these rookeries were incubated in the laboratory and embryos were euthanised before hatching. Maternal tissue was collected from these two rookeries during oviposition by sampling from the trailing edge of the back flipper with a sterile 3-mm biopsy punch (Bydand Medical, NSW, Australia). At the Gnaraloo Bay rookery, hatchlings were collected from nests for a study conducted by Woolgar et al. (2013), and we used available samples to assess paternity (GB; n = 10–22 eggs per clutch from eight clutches). No maternal tissue samples were available for the GB rookery because the collection permit for the study on this population did not cover sampling of adult females. All samples from the DHI (n = 859) and BB (n = 66) rookeries were stored at room temperature in 2 mL Longmire buffer until processing, whereas samples from the GB rookery (n = 119) were stored at 4C in 1.5–2.0 mL of 100% EtOH. Microsatellite analysis and genotyping Fourteen of the 15 clutches collected from DHI were genotyped, as one clutch was unfertilised. Total DNA was extracted from 1026 samples of offspring (minimum of 10 offspring per clutch) and 18 maternal samples using a standard salting-out method (Sunnucks and Hales 1996), with the exception of proteinase K digestion [200 mg mL–1] at 56C overnight. The DNA pellet was resuspended in 100 mL nuclease-free sterile water and quantified by a NanoDrop® Spectrophotometer (ND1000, Thermo Fisher Scientific, Australia). All samples were normalised to 10 ng DNA mL–1 with nuclease-free water before polymerase chain reaction (PCR). Four loci designed for C. caretta (Cc8E07, Cc7B07, Cc5F01, Cc7C04: see Shamblin et al. 2007) were run in a single PCR multiplex. PCR was performed in 10-mL reactions with 1 ng DNA template, 7.8 mL Platinum Supermix (Invitrogen, Life Technologies, Vic., Australia), 0.2 mL MgCl2 [50 mM], and 0.25 mL each primer [7 mM]. PCR products were denatured at 95C for 3 min, (40) 30 s at 95C, 45 s at 53C, 30 s at 72C, and 8 min extension at 72C. All PCR products were analysed on an ABI 3730 Sequencer against GeneScan 500 LIZ internal size standard and DNA fragments were scored manually with GeneMarker 1.91 software (SoftGenetics, LLC®, USA). Data analysis Levels of genetic variation among the 18 maternal genotypes (DHI and BB rookeries) were assessed by calculating the number of alleles per locus, and allele frequencies at each locus using the GENEALEX 6.5 software package (Peakall and Smouse 2012). We also used this program to assess Hardy– J. N. Tedeschi et al. Weinberg equilibrium and calculate the probability of two different females having identical multilocus genotypes. Observed and expected heterozygosity for the four loci for each rookery were estimated with CERVUS 3.0.3 (Marshall et al. 1998). The presence of null alleles was tested at each locus in only the 14 maternal genotypes from the DHI rookery using the software package MICROCHECKER (Van Oosterhout et al. 2004); the sample size from the BB rookery was insufficient for detecting null alleles with reliability. We assessed paternity within each clutch sample using initial inference, the GERUD 2.0 software package (Jones 2005), and the COLONY 2.0 software package (Wang 2004; Wang and Santure 2009). Neither GERUD 2.0 nor COLONY 2.0 require population allele frequencies in order to calculate the minimum number of sires (Jones 2005; Wang and Santure 2009), so these packages were ideal for our purposes given that other adult females from the rookeries were not sampled. To evaluate paternity with initial inference we used the maternal genotypes to identify maternal contributions to each offspring and inferred paternal alleles by excluding maternal alleles in the offspring genotypes (Jones et al. 2010). Multiple paternity was determined when three or more non-maternal alleles were found at a single locus. Since maternal genotypes were not available for the GB rookery, we inferred maternal allelic contribution based on the frequency of alleles in the offspring within each clutch. The GERUD 2.0 analyses were performed using all four loci with the parameter for the maximum number of sires set to four. Runs were conducted with and without maternal genotypes. When the GERUD program returned multiple solutions for progeny arrays, they were ranked by likelihood based on the segregation of paternal alleles and their deviation from Mendelian expectations (Jones 2005). The combination of sires with the highest probability score was used to calculate the minimum number of sires for the clutch. The COLONY analyses were also performed using all four loci. COLONY assigns sibships and parentage based on a maximum-likelihood model. Offspring are clustered by full-sib and half-sib (maternal and paternal), and parent–offspring relationships are determined, with parents assigned to full-sib groups. Unknown genotypes for either parent can be inferred (Wang 2004; Wang and Santure 2009). For each rookery, all genotyped offspring were analysed in a single dataset to identify any paternal half-sibs, which would indicate males that sired offspring with more than one female. COLONY was set to the default parameters, a single medium-length run, with fulllikelihood analysis, assuming polygamy for both males and females. Parallel to the GERUD analysis, COLONY runs were performed with and without maternal genotypes. COLONY can estimate paternity with datasets containing missing or rare alleles, but GERUD cannot. Offspring for which maternal alleles or data were missing were therefore excluded from the GERUD analysis. The reduced dataset for the DHI rookery included genotypes for 14 females and 791 offspring; full dataset included 813 offspring. For BB, the reduced dataset was for 4 females and 60 offspring; full dataset included 62 offspring. Finally, for the GB rookery, the reduced dataset included 84 offspring while the full dataset analysed included 92 offspring. Following the consensus approach proposed by Theissinger et al. (2009) and Stewart and Dutton (2011), multiple Variable rates of paternity in loggerheads Australian Journal of Zoology paternity was identified in each clutch if two of the three methods used had detected more than one sire. Results All four loci were polymorphic, with the number of alleles per locus ranging from 6 to 22, with observed heterozygosity ranging from 0.70 to 1.00 (Table 2). The probability of females from the BB and DHI rookeries sharing a multilocus genotype ranged from 1.6  10 2 to 7.5  10 2. Genotypic frequencies for the DHI rookery at all loci were in agreement with Hardy–Weinberg equilibrium (P > 0.05), and no null alleles were detected. The estimated proportions of multiple paternity varied among rookeries (Table 3). On the basis of initial inference, the frequency of multiple paternity was 25.0% (1 of 4 clutches) at BB, 35.7% (5 of 14 clutches) at DHI and 85.7% (6 of 7 clutches) at GB. The mean minimum number of sires per clutch estimated using initial inference ranged from 1.2 to 1.9 (Table 3). Estimates of multiple paternity and the minimum number of sires per clutch were slightly higher when calculations were based on the GERUD and COLONY analyses. The frequency of multiple paternity ranged from 25% (1 of 4) to 100% (7 of 7) and the minimum number of sires per clutch ranged from 1.1 to 2.1 (Table 3). Nevertheless, a similar pattern to the initial inference estimates was apparent, with both estimates for the large rookery at DHI being closer to the lower range values. The GERUD estimates of minimum number of sires per clutch and frequency of multiple paternity were identical when calculated with or without maternal genotypes. Two additional instances of multiple paternity were detected in the COLONY analyses when runs were conducted without maternal genotype (Table 3). Reconstructed paternal genotypes from GERUD and COLONY agreed six out of the 12 instances (50%) where multiple paternity was determined across the three rookeries. The analyses indicated that 16–25 individual males sired offspring in the clutches sampled from the DHI rookery (n = 14), Table 2. Descriptive statistics of the four polymorphic microsatellite markers n, sample size; A, mean number of alleles per locus; HO, observed heterozygosity; HE, expected heterozygosity Rookery n Locus Allele size range (bp) A HO HE 791 Cc8E07 Cc5F01 Cc7C04 Cc7B07 248–299 115–178 184–233 212–304 13 18 14 22 0.901 0.886 0.804 0.833 0.873 0.916 0.862 0.915 Bungelup (BB) 60 Cc8E07 Cc5F01 Cc7C04 Cc7B07 248–291 116–169 192–233 216–304 9 11 6 13 0.883 0.967 0.700 1.000 0.827 0.891 0.769 0.909 Gnaraloo (GB) 84 Cc8E07 Cc5F01 Cc7C04 Cc7B07 248–315 115–190 188–233 216–308 13 15 11 17 0.833 0.917 0.905 0.940 0.885 0.929 0.844 0.911 Dirk Hartog (DHI) E 5 males sired offspring in the clutches from BB (n = 4), and 11–15 males sired offspring in the clutches from the GB rookery (n = 7). None of the males were identical across the three rookeries. Where maternal genotypes were known, the probability of two males from the BB and DHI rookeries sharing a multilocus genotype ranged from 5.5  10 2 to 9.2  10 3. Where the maternal genotype was not known, the probability of two males from all three rookeries sharing a multilocus genotype ranged from 1.6  10 2 to 9.2  10 3. Discussion At three C. caretta rookeries in Western Australia, females laid clutches that were sired by multiple males 25–86% of the time during peak nesting periods between 2011 and 2013. This result is consistent with estimates of multiple paternity in populations of C. caretta from the Northern Hemisphere and eastern Australia, where rates of 25–33% are typical (Table 1). The highest rate of multiple paternity was found at Gnaraloo Bay (GB) where multiple males sired 86% of clutches (assuming paternal genotypes were correctly deduced from correctly inferred maternal genotypes). It is unclear why such a high rate of multiple paternity should occur in a low-nesting rookery such as GB, although the size of the offshore breeding area occupied by males and females may impact the nesting density, and hence male–female encounters. For example, Zbinden et al. (2007) reported that 93% of C. caretta clutches from the Laganas Bay rookery on Zakynthos Island in Greece exhibited multiple paternity. They attributed this rate to the small size of the bay that bordered the nesting beach, which confined the population and increased densities of breeding males and females. Lasala et al. (2013) also report a high rate of multiple paternity (75%) for nests on Wassaw Island, Georgia. The authors suggest that this may be due to a large number of males migrating along the coastline and crossing nesting beach boundaries. The GB rookery is situated on a wide and open bay with near-continuous fringing reef (Short 2005; Hattingh et al. 2011), so it is plausible that the high rate of multiple paternity found at this rookery is a result of large numbers of males migrating along the fringing reefs. However, as we do not know how much of the offshore area comprises the breeding grounds, tracking of sea turtles in the water during the breeding season would indicate the density of turtles at sea, and permit estimation of the probability of male–female encounters (Schofield et al. 2013). In contrast to the rookery at GB, the frequency of multiple paternity was lower in clutches sampled from the BB (25%) and DHI (36%) rookeries. Although the estimate of low multiple paternity for the BB rookery may reflect the relatively small clutch and offspring sample sizes, this was not the case for DHI, where sample sizes and the number of clutches analysed were comparatively large (see Table 1). Uller and Olsson (2008) proposed that, all else being equal, the degree of multiple paternity should be positively correlated with the probability of mate encounters. If this model applies to C. caretta in Western Australia, it would imply that male density is higher in the centre of the species’ Western Australian distribution. However, additional survey and molecular work would be required to verify this possibility. F Australian Journal of Zoology J. N. Tedeschi et al. Table 3. Minimum number of sires per clutch in C. caretta as estimated by initial inference, GERUD 2.0, and COLONY 2.0 runs with and without maternal genotype Multiple paternity (MP) was concluded when at least two of the three methods detected a minimum of two sires per clutch (shown in bold). Percentages (% MP) and mean values ( s.e.m.) of estimated rates of multiple paternity are indicated for each method across all clutches analysed Rookery DHI Clutch No. of embryos analysed (clutch size) Initial inference A B C D E F G H J L M N P R 41 (45) 60 (69) 37 (42) 45 (46) 52 (58) 53 (62) 50 (54) 61 (62) 71 (72) 73 (77) 55 (56) 73 (75) 65 (69) 55 (58) 1 2 1 2 1 1 1 1 1 2 1 2 1 2 1 3 1 2 1 1 3 1 1 3 1 3 1 3 1 3 1 2 1 1 3 1 1 3 1 3 1 3 1 2 1 1 1 1 1 1 1 1 1 2 1 1 1 2 2 1 1 2 1 1 1 1 1 2 1 1 N Y N Y N N N N N Y N Y N Y 35.7% (5/14) 1.26 ± 0.13 42.9% (6/14) 1.79 ± 0.26 42.9% (6/14) 1.79 ± 0.26 14.3% (2/14) 1.14 ± 0.10 28.6% (4/14) 1.29 ± 0.13 35.7% (5/14) 1 2 1 1 1 2 1 1 1 2 1 1 1 2 1 1 1 2 1 1 N Y N N 25% (1/4) 1.25 ± 0.25 25% (1/4) 1.25 ± 0.25 25% (1/4) 1.25 ± 0.25 25% (1/4) 1.25 ± 0.25 25% (1/4) 1.25 ± 0.25 25% (1/4) % MP Mean BB A C D H 11 (11) 18 (19) 14 (14) 17 (18) % MP Mean GB A B C D F G H % MP 14 (15) 10 (11) 12 (12) 10 (10) 13 (20) 13 (15) 12 (14) GERUD 2.0 Without With maternal maternal genotype genotype COLONY 2.0 Without With maternal maternal genotype genotype Multiple paternity? 2 1 2 1 2 2 3 2 2 2 2 2 2 3 3 2 1 1 1 1 2 Y Y Y N Y Y Y 71.4% (5/7) 100% (7/7) 42.9% (3/7) 85.7% (6/7) We found no evidence of a relationship between rookery size and the incidence of multiple paternity as has been reported elsewhere in sea turtles (Lee 2008). The C. caretta rookery at DHI is one of the largest in the world (Baldwin et al. 2003; Reinhold and Whiting 2014), but has one of the lowest rates of multiple paternity reported. Phillips et al. (2013, 2014b) found a similar pattern in a population of hawksbill turtles (Eretmochelys imbricata) nesting in the Seychelles Islands. They found a high number of males contributing to the clutches sampled (47 males fertilised 43 clutches), but the frequency of multiple paternity was low (9.3%), which they attributed to a low rate of mate encounter over a widely dispersed breeding area (Phillips et al. 2013, 2014b). The lack of a relationship between rookery size and rates of multiple paternity might also reflect a declining number of males associated with feminisation of primary sex ratios due to climate change (Wright et al. 2012b; Hawkes et al. 2014). However, no data on the trends in male abundance in this genetic stock are available to support this assumption. If rookery topography is not a factor in determining the frequency of multiple paternity at GB, perhaps population demographics can explain this high estimate. Fitzsimmons et al. (1997a) found that both male and female green turtles (C. mydas) in eastern Australia exhibit similar levels of philopatry to their native beaches. If this behaviour is also common to loggerhead turtles in Western Australia, the high rate of multiple paternity occurring at the GB rookery may reflect a greater number of males returning to breed than at the DHI and BB rookeries. Alternatively, more male offspring may be produced at the GB rookery. Tentative support for this idea comes from a study that compared empirical and modelled nest temperatures at each of our three study rookeries, where the midrange GB rookery had cooler beach temperatures relative to the two range-edge rookeries at DHI and BB (Woolgar 2012). Variable rates of paternity in loggerheads Hence, as C. caretta has temperature-dependent sex determination, with males being produced at cooler incubation temperatures (Miller 1985; Standora and Spotila 1985; Mrosovsky 1994), the GB rookery, at the centre of the species’ range, may produce relatively more male offspring than at the other two rookeries studied (see Woolgar et al. 2013). This, in turn, could drive differences in the OSRs of each nesting population. Further, male-mediated gene flow is promoted by mating on migration routes and possibly feeding grounds (Fitzsimmons et al. 1997b), which may contribute to sex ratio differences between rookeries, especially if females and males travel different routes (Fitzsimmons et al. 1997b; Wright et al. 2012b). Measuring nest incubation temperatures across years to assess long-term changes in hatchling sex ratios (Laloë et al. 2014) in combination with long-term genetic monitoring of the nesting females will show whether the pattern we observed for these three rookeries is temporally stable, and our data can be used as a baseline for determining whether OSRs change over time. Two common methods for estimating OSRs in sea turtle populations are to count the number of females and males encountered along a transect (Hays et al. 2010) or to estimate paternal contributions of clutches sampled from nesting beaches (Stewart and Dutton 2011; Hawkes et al. 2014). OSRs change as the nesting season progresses, as males and females arrive at breeding grounds at different times, have different periods of residence, and different remigration intervals (Limpus 1993; Godley et al. 2002; Hays et al. 2010, 2014). As the remigration interval for male C. caretta is shorter than for females, future scenarios of climate change may not decrease population viability even with increased feminisation of offspring (Hays et al. 2010; Phillips et al. 2014b; but see Wright et al. 2012a). As long as males return frequently to breeding grounds, fertilisation success should be stable (Hays et al. 2010; Wright et al. 2012b). Hence, healthy and genetically diverse populations should be able to absorb a reduction in males given the polyandrous nature of sea turtles, but periodic monitoring of OSRs (e.g. every 5–10 years) is critical for detecting ratios that could reduce population viability. In summary, it is clear that multiple paternity is the normal mating system in most species of sea turtle, and that rates vary by species, population, and by the method of detection (Bowen and Karl 2007). Despite the lack of a maternal genotype for one of our three rookeries, all methods used in this study led us to conclude that multiple paternity rates ranged from 25 to 86% in C. caretta clutches sampled from Western Australian rookeries. Additional samples from BB and GB (including maternal tissue), as well as from the Muiron Islands at the northern limit of the nesting range, would be valuable for assessing whether our estimates of the rates of multiple paternity are representative of the Western Australian population. Further, reconstruction of the genotypes of males that successfully mated with females, as we did in this study, allows the indirect estimation of the number of males contributing to this population and hence more realistic estimation of the adult population size. The rates of multiple paternity we have detected provide a snapshot of the mating system of the Western Australian population, and it will be important to repeat our sampling in order to detect changes in OSRs over time. Australian Journal of Zoology G Acknowledgements We thank the Editor and two anonymous reviewers for providing comments that significantly improved the quality of an earlier version of the manuscript. We thank the Western Australian Department of Parks and Wildlife (DPaW) turtle-monitoring program coordinators and staff at Shark Bay and Cape Range National Parks for facilitating our access to the rookeries and for outstanding logistical support, in particular Dave Holley, Dave Charles, Chris McMonagle, Peter Barnes and Keely Markovina. We also thank the many turtle-monitoring volunteers from DPaW for their assistance in the field. Paul Richardson, Karen Hattingh and the 2011 Gnaraloo Turtle Monitoring team are also thanked for their assistance with our fieldwork. Sherralee Lukehurst and Yvette Hitchen provided considerable guidance and support for our molecular work. This study was conducted under licenses SF007689, SF008414, SF009051 and SF009392 issued by the Western Australian Department of Parks and Wildlife, and was approved by the UWA Animal Ethics Committee (3/100/968, 3/100/1046, 3/100/1081, and 3/100/1195). References Baldwin, R., Hughes, G. R., and Prince, R. I. T. (2003). Loggerhead turtles in the Indian Ocean. In ‘Loggerhead Sea Turtles’. (Eds B. E. Bolten and A. B. Witherington.) pp. 218–234. (Smithsonian Books: Washington, DC.) Bollmer, J. L., Irwin, M. E., Rieder, J. P., and Parker, P. G. (1999). Multiple paternity in loggerhead turtle clutches. Copeia 1999, 475–478. doi:10.23 07/1447494 Bowen, B., and Karl, S. (2007). Population genetics and phylogeography of sea turtles. Molecular Ecology doi:10.1111/j.1365-294X.2007.03542.x Crim, J. L., Spotila, D., Spotila, J. R., O’Connor, M., Reina, R., Williams, C. J., and Paladino, F. V. (2002). The leatherback turtle, Dermochelys coriacea, exhibits both polyandry and polygyny. Molecular Ecology 16, 4886–4907. Dutton, P., Bixby, E., and Davis, S. K. (2000). Tendencey towards single paternity in leatherbacks detected with microsatellites. In ‘Proceedings of the Eighteenth International Symposium on Sea Turtle Biology and Conservation’. (Eds F. A. Abreu-Grobois, R. Briseno-Duenas, R. Marquez, and L. Sarti.) p. 39. NOAA Technical Memorandum NMFSSEFSC-436. Technical Information Service, Springfield, Virginia. Ekanayake, E. M. L., Kapurusinghe, T., Saman, M. M., Rathnakamura, D. S., Samaraweera, P., Ranawana, K. B., and Rajakaruna, R. S. (2013). Paternity of green turtle (Chelonia mydas) clutches laid in Kosgoda, Sri Lanka. Herpetological Conservation and Biology 8, 27–36. Fitzsimmons, N. N. (1998). Single paternity of clutches and sperm storage in the promiscuous green turtle (Chelonia mydas). Molecular Ecology 7, 575–584. doi:10.1046/j.1365-294x.1998.00355.x Fitzsimmons, N. N., Limpus, C. J., Norman, J. A., Goldizen, A. R., Miller, J. D., and Moritz, C. (1997a). Philopatry of male marine turtles inferred from mitochondrial DNA markers. Proceedings of the National Academy of Sciences of the United States of America 94, 8912–8917. doi:10.10 73/pnas.94.16.8912 Fitzsimmons, N. N., Moritz, C., Limpus, C. J., Pope, L., and Prince, R. (1997b). Geographic structure of mitochondrial and nuclear gene polymorphisms in Australian green turtle populations and male-biased gene flow. Genetics 147, 1843–1854. Godley, B. J., Broderick, A. C., Fraunstein, R, Glen, F, and Hays, G. C. (2002). Reproductive seasonality and sexual dimorphism in green turtles. Molecular Ecology Progress Series 226, 125–133. doi:10.3354/meps22 6125 Harry, J. L., and Briscoe, D. A. (1988). Multiple paternity in the loggerhead turtle (Caretta caretta). The Journal of Heredity 79, 96–99. Hattingh, K., Boureau, M., Duffy, M., and Wall, M. (2011). Gnaraloo Turtle Conservation Program. Gnaraloo Bay Rookery, Final Report, Program 2010/11. Day monitoring program with night checks and crab burrow surveys. 20 July 2011. Gnaraloo Station Trust, Western Australia. H Australian Journal of Zoology Hawkes, L. A., Broderick, A. C., Godfrey, M. H., Godley, B. J., and Witt, M. J. (2014). The impacts of climate change on marine turtle reproductive success. In ‘Coastal Conservation’. (Eds B. Masalo and J. L. Lockwood.) pp. 287–310. (Cambridge University Press: Cambridge.) Hays, G. C., Fossette, S., Katselidis, K. A., Schofield, G., and Gravenor, M. B. (2010). Breeding periodicity for male sea turtles, operational sex ratios, and implications in the face of climate change. Conservation Biology 24, 1636–1643. doi:10.1111/j.1523-1739.2010.01531.x Hays, G. C., Mazaris, A. D., and Schofield, G. (2014). Different male vs. female breeding periodicity helps mitigate offspring sex ratio skews in sea turtles. Frontiers in Marine Science 1, 1–9. doi:10.3389/fmars.2014. 00043 Hoekert, W. E. J., Neuféglise, H., Schouten, A. D., and Menken, S. B. J. (2002). Multiple paternity and female-biased mutation at a mirosatellite locus in the Olive Ridley sea turtle (Lepidochelys olivacea). Heredity 89, 107–113. doi:10.1038/sj.hdy.6800103 Ireland, J. S., Broderick, A. C., Glen, F., and Godley, B. J. (2003). Multiple paternity assessed using microsatellite markers, in green turtles Chelonia mydas (Linnaeus, 1758) of Ascension Island, south Atlantic. Journal of Experimental Marine Biology and Ecology 291, 149–160. doi:10.1016/ S0022-0981(03)00118-7 Jensen, M. P., Abreu-Grobois, F. A., Frydenberg, J., and Loeschcke, V. (2006). Microsatellites provide insight into contrasting mating patterns in arribada vs. non-arribada Olive Ridley sea turtle rookeries. Molecular Ecology 15, 2567–2575. doi:10.1111/j.1365-294X.2006. 02951.x Jensen, M. P., Fitzsimmons, N. N., and Dutton, P. H. (2013). Molecular genetics of sea turtles. In ‘Biology of Sea Turtles. Vol III’. (Eds J. Wyneken, K. J. Lohmann, and J. A. Musick) pp. 155–182. (CRC Press: Boca Raton, FL.) Jones, A. G. (2005). Gerud 2.0: a computer program for the reconstruction of parental genotypes from half-sib progeny arrays with known or unknown parents. Molecular Ecology Notes 5, 708–711. doi:10.1111/ j.1471-8286.2005.01029.x Jones, A. G., Small, C. M., Paczolt, K. A., and Ratterman, N. L. (2010). A practical guide to methods of parentage analysis. Molecular Ecology Resources 10, 6–30. doi:10.1111/j.1755-0998.2009.02778.x Joseph, J., and Shaw, P. W. (2011). Multiple paternity in egg clutches of hawksbill turtles (Eretmochelys imbricata). Conservation Genetics 12, 601–605. doi:10.1007/s10592-010-0168-7 Kichler, K., Holder, M. T., Davis, S. K., Márquez-M, R., and Owens, D. W. (1999). Detection of multiple paternity in the Kemp’s Ridley turtle with limited sampling. Molecular Ecology 8, 819–830. doi:10.1046/j.1365294X.1999.00635.x Laloë, J.-O., Cozens, J., Renom, B., Taxonera, A., and Hays, G. C. (2014). Effects of rising temperature on the viability of an important sea turtle rookery. Nature Climate Change 4, 513–518. doi:10.1038/ nclimate2236 Lara-De La Cruz, L. I., Nakagawa, K. O., Cano-Camacho, H., Zavala-Paramo, M. G., Vazquez-Marrufo, G., and Chassin-Noria, O. (2010). Detecting patterns of fertilization and frequency of multiple paternity in Chelonia mydas of Colola (Michoacán, Mexico). Hidrobiológica 20, 85–89. Lasala, J. A., Harrison, J. S., Williams, K. L., and Rostal, D. C. (2013). Strong male-biased operational sex ratio in a breeding population of loggerhead turtles (Caretta caretta) inferred by paternal genotype reconstruction analysis. Ecology and Evolution 3, 4736–4747. doi:10.10 02/ece3.761 Lee, P. L. M. (2008). Molecular ecology of sea turtles: new approaches and future directions. Journal of Experimental Marine Biology and Ecology 356, 25–42. doi:10.1016/j.jembe.2007.12.021 Lee, P. L. M., and Hays, G. C. (2004). Polyandry in a sea turtle: females make the best of a bad job. Proceedings of the National Academy of Sciences of the United States of America 101, 6530–6535. doi:10.1073/ pnas.0307982101 J. N. Tedeschi et al. Limpus, C. J. (1993). The green turtle, Chelonia mydas, in Queensland: breeding males in the southern Great Barrier Reef. Wildlife Research 20, 513–523. doi:10.1071/WR9930513 Marshall, T. C., Slate, J., Kruuk, L. E. B., and Pemberton, J. M. (1998). Statistical confidence for likelihood-based paternity. Molecular Ecology 7, 639–655. doi:10.1046/j.1365-294x.1998.00374.x Miller, J. D. (1985). Embryology of marine turtles. In ‘Biology of the Reptilia. Vol. 14’. (Eds C. Gans, F. Billet, and P. Maderson.) pp. 269–328. (John Wiley and Sons: New York.) Moore, M. K., and Ball, R. M. Jr. (2002). Multiple paternity in loggerhead turtle (Caretta caretta) nests on Melbourne Beach, Florida: a microsatellite analysis. Molecular Ecology 11, 281–288. doi:10.1046/ j.1365-294X.2002.01426.x Mrosovsky, N. (1994). Sex ratios of sea turtles. The Journal of Experimental Zoology 270, 16–27. doi:10.1002/jez.1402700104 Pacioni, C., Trocini, S., Heithaus, M., Burkholder, D., Thomson, J., Warren, K., and Krutzen, M. (2012). Preliminary assessment of the genetic profile of the Western Australian loggerhead turtle population using mitochondrial DNA. In ‘Proceedings of the First Western Australian Sea Turtle Symposium’. p. 19. Department of Parks and Wildlife, Government of Western Australia. Peakall, R., and Smouse, P. E. (2012). GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research – an update. Bioinformatics 28, 2537–2539. doi:10.1093/bioinformatics/bts460 Peare, T., and Parker, P. G. (1996). Local genetic structure within two rookeries of Chelonia mydas (the green turtle). Heredity 77, 619–628. doi:10.1038/hdy.1996.189 Pearse, D. E., and Avise, J. C. (2001). Turtle mating systems: behaviour, sperm storage, and genetic paternity. The Journal of Heredity 92, 206–211. doi:10.1093/jhered/92.2.206 Phillips, K. P., Jorgensen, T. H., Jolliffe, K. G., Jolliffe, S.-M., Henwood, J., and Richardson, D. S. (2013). Reconstructing paternal genotypes to infer patterns of sperm storage and sexual selection in the hawksbill turtle. Molecular Ecology 22, 2301–2312. doi:10.1111/mec.12235 Phillips, K. P., Jorgensen, T. H., Jolliffe, K. G., and Richardson, D. S. (2014a). Potential inter-season sperm storage by a female hawksbill turtle. Marine Turtle Newsletter 140, 13–14. Phillips, K. P., Mortimer, J. A., Jolliffe, K. G., Joregensen, T. H., and Richardson, D. S. (2014b). Molecular techniques reveal cryptic life history and demographic processes of a critically endangered marine turtle. Journal of Experimental Marine Biology and Ecology 455, 29–37. doi:10.1016/j.jembe.2014.02.012 Reinhold, L., and Whiting, A. (2014). High-density loggerhead sea turtle nesting on Dirk Hartog Island, Western Australia. Marine Turtle Newsletter 141, 7–10. Rieder, J. P., Parker, P. G., Spotila, J. R., and Irwin, M. E. (1998). The mating system of the leatherback turtle: a molecular approach. In ‘Proceedings of the Sixteenth Annual Symposium on Sea Turtle Biolgy and Conservation’. (Eds R. Byles and Y. Fernandez.) pp. 120–121. NOAA Technical Memorandum NMFS-SEFSC-412. National Technical Information Service, Springfield, Virginia. Sakaoka, K., Yoshii, M., Okamoto, H., Sakai, F., and Nagasawa, K. (2011). Sperm utilization patterns and reproductive success in captive loggerhead turtles (Caretta caretta). Chelonian Conservation and Biology 10, 62–72. doi:10.2744/CCB-0878.1 Schofield, G., Scott, R., Dimadi, A., Fossette, S., Katselidis, K. A., Koutsoubas, D., Lilley, M. K. S., Pantis, J. D., Karagouni, A. D., and Hays, G. C. (2013). Evidence-based marine protected area planning for a highly mobile endangered marine vertebrate. Biological Conservation 161, 101–109. doi:10.1016/j.biocon.2013.03.004 Shamblin, B. M., Faircloth, B. C., Dodd, M., Wood-Jones, A., Castleberry, S. B., Carroll, J. P., and Nairn, C. J. (2007). Tetranucleotide microsatellites from the loggerhead sea turtle (Caretta caretta). Molecular Ecology Notes 7, 784–787. doi:10.1111/j.1471-8286.2007.01701.x Variable rates of paternity in loggerheads Australian Journal of Zoology Shine, R. (2005). Life-history and evolution in reptiles. Annual Review of Ecology Evolution and Systematics 36, 23–46. doi:10.1146/annurev. ecolsys.36.102003.152631 Short, A. D. (2005). Gnaraloo Bay. In ‘Beaches of the Western Australian Coast – Eucla to Roebuck Bay: A Guide to Their Nature, Characteristics, Surf and Safety’. pp. 322–323. (Sydney University Press: Sydney.) Standora, E. A., and Spotila, J. R. (1985). Temperature dependent sex determination in sea turtles. Copeia 1985, 711–722. doi:10.2307/1444 765 Stewart, K. R., and Dutton, P. H. (2011). Paternal genotype reconstruction reveals multiple paternity and sex ratios in a breeding population of leatherback turtles (Dermochelys coriacea). Conservation Genetics 12, 1101–1113. doi:10.1007/s10592-011-0212-2 Sunnucks, P., and Hales, D. F. (1996). Numerous transposed sequences of mitochondrial cytochrome oxidase I–II in aphids of the genus Sitobion (Hemiptera: Aphididae). Molecular Biology and Evolution 13, 510–524. doi:10.1093/oxfordjournals.molbev.a025612 Theissinger, K., Fitzsimmons, N. N., Limpus, C. J., Parmenter, C. J., and Phillott, A. D. (2009). Mating system, multiple paternity and effective population size in the endemic flatback turtle (Natator depressus) in Australia. Conservation Genetics 10, 329–346. doi:10.1007/s10592-0089583-4 Trocini, S. (2013). Health assessment and hatching success of two Western Australian loggerhead turtle (Caretta caretta) populations. Ph.D. Thesis, Murdoch University, Perth. Uller, T., and Olsson, M. (2008). Multiple paternity in reptiles: patterns and processes. Molecular Ecology 17, 2566–2580. doi:10.1111/j.1365-294X. 2008.03772.x Van Oosterhout, C., Hutchinson, W. F., Wills, D. P. M., and Shipley, P. (2004). Micro-Checker: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4, 535–538. doi:10.1111/j.1471-8286.2004.00684.x I Wang, J. (2004). Sibship reconstruction from genetic data with typing errors. Genetics 166, 1963–1979. doi:10.1534/genetics.166.4.1963 Wang, J., and Santure, A. W. (2009). Parentage and sibship inference from multilocus genotype data under polygamy. Genetics 181, 1579–1594. doi:10.1534/genetics.108.100214 Woolgar, L. (2012). A comparison of two techniques used to model sand temperatures and sex ratios at loggerhead turtle (Caretta caretta) rookeries in Western Australia. M.Sc. Thesis, The University of Western Australia, Perth. Woolgar, L., Trocini, S., and Mitchell, N. (2013). Key parameters describing temperature-dependent sex determination in the southernmost population of loggerhead sea turtles. Journal of Experimental Sea Biology and Ecology 449, 77–84. doi:10.1016/j.jembe.2013.09.001 Wright, L. I., Fuller, W. J., Godley, B. J., McGowan, A., Tregenza, T., and Broderick, A. C. (2012a). Reconstruction of paternal genotypes over multiple breeding seasons reveals male green turtles do not breed annually. Molecular Ecology 21, 3625–3635. doi:10.1111/j.1365-294X. 2012.05616.x Wright, L. I., Stokes, K. L., Fuller, W. J., Godley, B. J., McGowan, A., Snape, R., Tregenza, T., and Broderick, A. C. (2012b). Turtle mating patterns buffer against disruptive effects of climate change. Proceedings of the Royal Society B: Biological Sciences 279, 2122–2127. doi:10.1098/ rspb.2011.2285 Zbinden, J. A., Largiadèr, C. R., Leippert, F., Margaritoulis, D., and Arlettaz, R. (2007). High frequency of multiple paternity in the largest rookery of Mediterranean loggerhead sea turtles. Molecular Ecology 16, 3703–3711. doi:10.1111/j.1365-294X.2007.03426.x Handling Editor: Paul Cooper www.publish.csiro.au/journals/ajz