Ophiodon Elongatus, Hexagrammidae)

Molecular Ecology (2005) 14, 653–660 doi: 10.1111/j.1365-294X.2005.02438.x

BlackwellMale Publishing, Ltd. nest site fidelity and female serial polyandry in lingcod (Ophiodon elongatus, Hexagrammidae)

JACQUELYNNE R. KING and RUTH E. WITHLER Pacific Biological Station, Fisheries and Oceans Canada, Nanaimo, B.C., V9T 6N7, Canada

Abstract Nest site fidelity and serial polyandry were examined in lingcod, Ophiodon elongatus, a teleost fish in which the nest-guarding male parent invests more heavily in parental care than the elusive female parent. Lingcod parental and progeny genotypes were established for fish spawning on a 200 m2 section of Snake Island reef, British Columbia in two successive years to evaluate male and female mate choice (monogamy or polygamy) and nest site reuse by the same parents (nest site fidelity) and/or different parents (nest site affinity). Thirteen nests (egg masses) guarded by nine males and 14 nests guarded by seven males were observed in 2002 and 2003, respectively. No female laid more than one nest per season or spawned in the study area in both years. In contrast, at least six (86%) and possibly all seven (100%) of the 2003 guardian males had been guardian or auxiliary males in 2002. Both nest site affinity and extreme male nest site fidelity were observed, with at least four males reusing the exact same nest site. Serial polyandry resulting from the high male and low female nest site fidelity is consistent with predictions based on a low female parental investment and high rate of progeny loss to predation and cannibalism. Male polygyny, achieved primarily by cuckoldry within seasons, was enhanced by the lack of female fidelity between seasons. Polygamy in both sexes of nest-tending marine fish may minimize reproductive skew and maximize genetic diversity within populations. Keywords: lingcod, nest site fidelity, polyandry, polygyny Received 24 June 2004; revision received 4 November 2004; accepted 29 November 2004

offspring, presumably have the choice between committing Introduction to monogamy with an apparently superior male or hedg- Among oviparous teleost fish, parental care may be pro- ing their bets by breeding with multiple males within a vided to offspring by either, both or neither of the parents, breeding season (polyandry) or between breeding seasons providing incentives for both sexes to maximize their (sequential polyandry). Documented examples of poly- reproductive success through a wide range of mating beha- andry are few (DeWoody & Avise 2001) and little is known viours ranging from complete monogamy to extensive poly- about the occurrence of sequential polyandry in teleost gamy. Males are the primary caregivers in many species fish. (Blumer 1979), responsible for establishing a nest or nesting Lingcod (Ophiodon elongatus) is a demersal marine teleost territory, wooing females and defending the fertilized eggs species of the eastern Pacific Ocean that inhabits nearshore resulting from their courtships. Genetic studies on nest- rocky reefs predominantly at depths from 10 to 100 m. It is tending species indicate that guardian males are commonly a large carnivore with maximum sizes of approximately polygynous within and between breeding seasons, mating 90 cm for males and 152 cm for females. Males mature at with multiple females who deposit eggs in nests guarded age two and have been known to live up to 14 years, whereas by themselves or other males (DeWoody & Avise 2001). females mature at ages 3–5 and may live to an age of 20 Females, whose fecundity is limited by body size and who years. Spawning occurs annually in January through March depend upon their partner(s) to adequately protect their but males arrive in spawning areas and establish territories as early as November. During spawning, which likely Correspondence: Jacquelynne R. King, Fax: 1(250) 756-7053; occurs at night (Low & Beamish 1978), females lay a single E-mail: [email protected] large mass of up to 500 000 eggs (referred to as a nest) in

654 J. R. KING and R . E . WITHLER rocky crevices or under rocks within a male territory. The monogamous within a breeding season, they may practice female leaves the nesting area after spawning but the male serial polyandry in part to counter the effects of predation remains in close proximity to the nest and actively guards and cannibalism. Observations made during nest density it against predation for 6 weeks until hatching. Two types surveys suggested that the same locations were being used of guarding behaviours have been observed: (i) males by lingcod as nesting sites between years (King & Winchell lying directly on or beside the nest and remaining nearly 2002) but provided no information on exact nest location or motionless unless touched, and (ii) males lying on a on male and female site fidelity. sentry post and defending the nest when other fish swam In this study, we have examined nest site fidelity in male close (O’Connell 1993). Males typically guard a single and female lingcod by comparing the genotypes of parents nest, but have been reported to guard up to three nests in that spawned on the Snake Island reef in 2003 with those close proximity (Low & Beamish 1978; King & Winchell observed in the 2002 study (Withler et al. 2004). Male and 2002). female genotypes for the parental lingcod were again Lingcod in the Strait of Georgia, British Columbia is determined by microsatellite analysis of tissue samples considered an overfished population (Cass 1985) with from guardian males and egg samples from their nests. early 1990s estimates of spawning biomass at only 5% of The analysis allowed us to distinguish between nest site observed historic levels (Martell & Wallace 1998). Recent fidelity (i.e. when an individual selects the same nest site success in year – class strength has been observed but the each year) and nest site affinity (i.e. when a nest site, as a stock is still considered to be of conservation concern (King result of habitat characteristics, is selected each year by dif- 2001). Nevertheless, a high degree of genetic diversity at ferent lingcod). We also tested our prediction that females microsatellite loci was observed in adult lingcod that is would practice a bet-hedging strategy of serial polyandry sampled from the region and attributed, at least in part, to by mating with different males in the second spawning a highly polygynous mating structure (Withler et al. 2004). season. Genetic analyses of tissue samples from lingcod guardian males and fertilized eggs from their 13 nests on Snake Study site Island reef in 2002 revealed that females were monogamous within breeding seasons, with each egg mass laid by a sin- Snake Island reef, situated in the Strait of Georgia (Fig. 1), gle female and no female spawning more than once. Males is the subject of ongoing lingcod nest density surveys guarded an average of fewer than 1.5 nests but achieved a (King & Beaith 2001; King & Winchell 2002). The study high degree of polygyny through cuckoldry. Thus each area, approximately 10 by 20 m with an average depth nest, consisting of the eggs of a single female, had been fer- of 10 m, was noted for recent relatively high nest densities tilized by between one and five males. Cuckolding males (King & Beaith 2001) and marked with a dive buoy included neighbouring male guardians and, likely, non- (49°12′43.7″ N; 123°53′04.7″ W) throughout the study. The guarding males. In all, the genotypes of 21 different males study site tended to be flat and open without large flora were identified among the progeny of the 13 nests attended such as Agarum spp. but containing several large rocks and by nine guardian males. boulders to provide nesting sites. Lingcod are territorial and have a narrow home range (Matthews 1992), making it possible that lingcod return to Methods the same reef each year to spawn. However, the observed polygyny may serve as a mechanism to maintain diversity Nest site identification in lingcod populations and a further increase in effective population size might occur if either males or females, As part of a genetic study investigating polygamous or both, failed to demonstrate interannual fidelity to nest mating in lingcod (Withler et al. 2004), nests were located sites, choosing to spawn on different reefs or different by scuba divers in the study site 1– 27 February 2002. Each areas of the same reef, in different years. For females, serial nest was marked with a numbered galvanized spike and polyandry would not only increase the genetic diversity the relative orientation (i.e. under a rock or in a crevice) among progeny, but also reduce the costs of a mating with was noted. The nest volume was estimated using length, a poor guardian in any one season. Lingcod eggs and width and height (nearest cm) measurements made under- juveniles often experience high levels of cannibalism and water. Given the difficulty in measuring irregularly shaped predation mainly by invertebrates, such as gastropods nests under rocks, measurements were made several times (Calliostoma ligatum) and sea stars (Pycnopodia helianthoides), and the average (nearest 0.5 L) was used. At the end of the and by sculpins (Artedius harringtoni, Jordania zonope), kelp nesting season, the galvanized spikes were replaced with greenling (Hexagrammos decagrammus) and striped seaperch rocks with plastic numbers attached. Distances between (Embiotoca lateralis) (Low & Beamish 1978; King & Beaith nests were measured to the nearest 0.5 m to develop a map 2001). This leads to the prediction that whereas female are of nest locations.

LINGCOD NEST SITE FIDELITY 655

Fig. 1 Snake Island reef is located in the Strait of Georgia, British Columbia. The location of the reef in proximity to Snake Island is indicated in the inset.

On 4–21 February 2003, the numbered rocks were located products were run on gels on an ABI 377 automated DNA and the locations of the 2003 nests were again marked with sequencer. Allele sizes were determined with genescan 3.1 numbered galvanized spikes. Nest sites and volumes were and genotyper 2.5 software (PE Biosystems). These loci calculated as in 2002. possessed between eight and 19 alleles each and heterozy- gosity values ranging from 83% to 91% in a sample of 80 adult lingcod genotypes from the sample area (Withler et al. 2004). Guardian male identification We calculated the probability of identity (PID) (i.e. the pro- Guardian males were attributed to a nest if they were within bability of obtaining identical genotypes from two differ- 2 m of it and exhibited protective or territorial behaviour ent individuals) for each locus according to equation: when a diver approached. We used a coding system for =+ ∑+∑−2224 labelling guardian male data and samples: Ny; where N PpppID 025. (. 025 iii ) [.( 05 )] (. 025 ) (Waits et al. denotes the nest number the male was guarding, and if 2001). guarding mültiple nests, the male was given only one number from the group of nests; y denotes the year that observa- This conservative estimate of PID is actually the estimated tions were made. All nests tended by a guardian male were likelihood of two full-sib individuals carrying the same noted and the total length (to the nearest cm) of each genotype and was recommended for use in studies of guardian estimated by pulling a tape measure alongside natural populations in which the possibility of encountering the fish as it rested on the bottom. Tissue samples of relatives is high (Waits et al. 2001). The PID across all five guardian males were collected in situ by scuba divers using loci was calculated as the product of the values obtained at sharp scissors from the terminal tip of the caudal fins. Fin the individual microsatellite loci. tissue samples were placed in a labelled tube filled with The allelic combinations observed in the progeny at each ambient seawater that was later replaced with ethanol. locus were used to reconstruct parental genotypes. Egg samples from all nests included some eggs fertilized by one or more of the guardian males that were genotyped Egg sample collection and microsatellite analysis from fin tissue. This made reconstruction of the maternal Divers cut egg samples (approximately 1 mL of eggs) from genotype apparent by subtraction of paternal alleles and each nest in 2003 and placed in a tube filled with seawater, also enabled determination of genotypes for auxiliary later with 95% ethanol on shore. Fifteen eggs were removed males that contributed to fertilization. from each egg sample and subjected to microsatellite analysis. DNA was obtained from individual eggs or fin tissue Results (from guardian male) using QIAGEN DNeasy Kits. Five microsatellite loci were assayed: Oel 32, Oel 35, Oel 41, Oel Nest site affinity 42 and Oel 45 (Sewall Young, Washington Department of Fish and Wildlife, personal communication). Primers were In 2002, nests 1 through 16 were located, although three labelled fluorescently and PCR (polymerase chain reaction) of them (numbers 7, 9 and 14) disappeared before genetic

656 J. R. KING and R . E . WITHLER

Probability of identity

The PID values for the five lingcod microsatellite loci ranged from 0.311 to 0.351, and provided an overall PID value of 0.004 across all loci. Thus, these highly polymorphic loci provided the ability to distinguish even closely related individuals in the study area.

Male nest site fidelity In 2002, six of the nine guardian males were genotyped from tissue samples and putative genotypes for three other

unsampled males (402, 602 and 1202) observed guarding nests were generated by extensive analysis of embryo genotypes from their nests (Withler et al. 2004). Genotyping of all seven 2003 guardian males revealed that four of the six guardians genotyped in 2002 returned to the same nest sites (Fig. 3). Thus, the guardian of nest 2 (along with nests 1 and 3 in 2002) was the same male in 2002 and 2003, i.e.

male 202 was male 203 (Fig. 3). Similarly, the 2002 guardian of nest 8 returned to nest site 8 and additionally guarded nests 20 and 21 in 2003. The 2002 nest 10 (and 11) guardian returned to nest site 10 in 2003. Finally, the same male guarded nest 15 in 2002 and 2003, and also guarded nests 16 and 19 in 2003 (Fig. 3). Of the three 2003 guardian males whose genotypes did not correspond with sampled 2002 guardian males, one

(603) matched one of the putative guardian genotypes established in 2002 (whose progeny were detected in 2002

nests 4, 8 and 12). Thus, male 603 likely guarded the same or a nearby site in 2002. Another of the 2003 males (403) who did not correspond with a genotyped 2002 male did match Fig. 2 Schematic map of study site with nests represented by a 2002 reconstructed male genotype (auxiliary male A) numbered circles. Dotted circles represent nest sites used in 2002 detected in eggs sampled from nest 1. Although male 202 only; solid line circles represent nest sites used in 2002 and was the guardian of nest 1, male A fertilized half the eggs 2003; grey shaded circles represent nest sites used in 2003 only. in the nest and may have been a neighbouring guardian Distances between nest sites are to the nearest 0.5 m. The letter B denotes location of the dive buoy (49°12′43.7″ N; 123°53′04.7″ W). outside the study area (Withler et al. 2004). Only one of the seven 2003 guardians (1403) had a genotype that was not detected among 2002 sampled males or recon- structions of paternal genotypes from 2002 egg sampling. sampling, presumably resulting from predation or becoming This male guarded nests 7 and 14 in 2003, located in the dislodged as a result of wave action (Fig. 2, Table 1). Nine area from which three nests and a male guardian dis- male lingcod guarded the 13 nests and fin tissue samples appeared in 2002 before sampling. Thus, among the seven were obtained from six of them (Table 1). In 2003, 14 nests 2003 guardian males, at least six returned from 2002 and were observed, nine of which were located at the exact four are known to have returned to the same nest sites. same rock or crevice as nests observed in 2002 (Fig. 2, Two guardian males genotyped in 2002 (1302 and 1602) Table 1). The numbers used to identify the locations of were not identified in from 2003 guardian male samples or these nine nests were retained from the 2002 season and reconstructed from 2003 nest egg samples. the five nests located at rocks or crevices not used by spawning lingcod in 2002 were assigned nest locations Female nest site fidelity 17 through 21. Each of the five nests in a new site was associated with a guardian male from a nest site that was In 2002, 13 different female genotypes (one for each nest) used in both seasons (Table 1). In total, there were seven were identified. In 2003, reconstruction of female genotypes males guarding 14 nests in the 2003 spawning season. was again possible for all 14 nests from embryonic genotyping

LINGCOD NEST SITE FIDELITY 657

Table 1 Nest data collected in 2002 (Withler et al. 2004) and 2003. The nest number was consistent between years. An empty cell means that the nest corresponding to the numbered location was not present in that sampling year. The guardian male for each nest is identified by a number associated with the first (if multiple) nests it was guarding. The males detected in egg samples from each 2003 nest are shown, with a plus sign indicating paternity by the guardian. Paternity by neighbouring guardians is indicated by their 2003 number and by males with genotypes not previously detected indicated a letter. Asterisks identify nests and guardian males that disappeared in a season. Numbers in bold denote guardian males that were genotyped from which fin tissue samples

Nest orientation Guardian male number Fertilizing males 2003 Nest number 2002 2003 2002 2003 Guardian Auxiliary

1 Under rock 202 – 2 Under rock Under rock 202 203 +403, A 3 Under rock 202 4 Under rock Under rock 402 403 +B 5 Under rock 402 6 Under rock Under rock 602 603 +C 7* In crevice Need location 702* 1403 +1503, D 8 Under rock In crevice 802 803 +— 9* Under rock 902* 10 In crevice In crevice 1002 1003 +— 11 In crevice 1002 12 Under rock 1202 13 In crevice 1402 14* Under rock Under rock 1402 1403 +— 15 In crevice In crevice 1502 1503 +— 16 Under rock Under rock 1602 1503 +603, 1403 17 Under rock 1503 +— 18 Under rock 403 –803 19 Under rock 403 –1503, E 20 Under rock 803 –1003 21 Under rock 803 –403

because all nest egg samples included progeny sired by Nest volume and guardian male size one or more of the seven guardian males (Table 1). Again, the eggs of each nest were derived from a single female There was a significant relationship between length (cm) and no female laid eggs in more than one nest. There of the guardian male and estimated volume (L) of the guarded was no overlap between the 13 females responsible for nest (or mean volume when more than one nest was nests in 2002 and those that produced 14 nests in the same guarded): volume = 0.15 length – 6.08 (F = 5.30; P = 0.036; area, and in many of the same nest sites, in 2003. Female r2 = 0.26; n = 17). Nest volume appears to increase with lingcod exhibited no interannual site fidelity on spatial increasing size of the guardian male (Fig. 4). However, it scales encompassed by the study area. should be noted that the proportion of variance explained by the regression line is small (r2 = 0.26). Small nests were guarded by either small or relatively large lingcod while Polygyny, cuckoldry, and serial polyandry there was some indication that large nests were guarded The limited egg sampling in 2003 revealed progeny attri- only by relatively large lingcod (Fig. 4). The apparent sig- butable to the guardian male in 10 of the 14 nests sampled nificance of the regression may be an artefact of the right- (Table 1). All seven guardian males contributed to fertil- hand portion of the scatter plot (Fig. 4). The number of ization in at least one of their guarded nests and six nests guarded by a male lingcod did not vary with size guardians fertilized eggs in at least one nest guarded by a of the guardian male (F = 1.24; P = 0.32). neighbour. Progeny fertilized by one or more neighbouring guardian males were observed in samples from seven Discussion nests, including the four nests in which the guardian was not detected. In addition, fertilization attributable to at The interannual fidelity to nest locations exhibited by least five auxiliary males was detected (Table 1). These male lingcod is, to our knowledge, the first record of males could not be matched with males observed in 2002. such behaviour reported for a temperate marine fish

Fig. 3 Schematic representation of guardian males associated with nests, grouped within ovals. Nest sites are represented by numbered circles as in Fig. 2. Fish symbols represent guardian males genotyped from fin tissue samples: dotted line fish are guardian males present in 2002 only; solid line fish are guardian males present in both years; grey shaded fish are guardian males present in 2003 only. Black fish represent guardian males present in 2003 but observed as spawners in 2002. Fin tissue samples were not collected for guardians of nests 7, 9 and 12 in 2002 and are therefore unknown

Guardians02. The genotypes of guardian males of nests 4 and 6 in 2003 matched genotypes of two spawners present in 2002. Guardian males of nests 2, 8, 10 and 15 were the same lingcod in 2002 and 2003.

species. The high rate and extreme precision of nest site some were as far as 1200 m from their previous sites. fidelity exhibited by guardian males was remarkable. At Among species more closely related to lingcod, kelp green- least six and possibly all seven of the guardian males ling (Hexagrammos decagrammus), also exhibits male nest present in 2003 had established territories on the same guarding behaviour and the females have been reported to section of the Snake Island reef in 2002, indicating a level of be polyandrous batch spawners, producing at least three nest site fidelity of at least 86%. At least four, and possibly egg masses per spawning season (Crow et al. 1997). Male six, of the males used the same territory in successive years kelp greenling guard an average of four nests, with each and the spatial resolution of nest fidelity was high. In some nest produced by a different female (Crow et al. 1997). Poly- cases, the orientation of the nest under a boulder or within gamy has also been reported for painted greenling (Oxylebius a crevice was identical between years although the female pictus) (DeMartini 1987). However, male or female nest site parents always differed. Nest site fidelity in males but not in fidelity is yet to be studied for these species. females indicates that the males are selecting the nesting sites Nest site affinity was also apparently high, with nine of and undertaking prespawning mate attraction behaviour. 14 nests in 2003 located in sites used in 2002. The one poten- Ridgway et al. (1991) observed that 48% of Lake Opeongo tially new 2003 male guarded a nest site occupied in 2002 male smallmouth bass returning to spawn in a second year by a different guardian. Similarly, the guardian of nest 15 nested within 20 m of the previous year’s site. However, in both spawning seasons took over previous nest locations only 81% nested within 200 m of their previous site and in the second spawning season when a neighbouring guardian

Beheler et al. 2003; Howlett & Stutchbury 2003). For species in which both genders may show site fidelity, pairs in which at least one parent is known to be experienced (i.e. nested in that site the previous year) may have greater success in fledging progeny (Woodward & Murphy 1999). For nest-guarding teleost fish in which the paternal care does not persist past hatching, such as lingcod, any advantages to females of mating with a known experienced male may be outweighed by maximizing the genetic diversity of their progeny, at least under the existing conditions of lingcod density, sex ratio and environmental conditions near the Snake Island reef. Additional support for the suggestion that the lingcod females may hedge their bets by mating Fig. 4 Length (cm) of guardian male vs. estimated volume (L) of with different males rather than trying to identify superior nest the fish was guarding (mean estimated volume if fish was males was provided by the weak support for a positive guarding > 1 nest). Squares are 2002 data and circles are 2003 data. relationship between male size and nest size and lack Linear regression (y = 0.15x – 6.08; r2 = 0.26) for pooled data was of relationship between male size and numbers of nests significant (P = 0.036). guarded. It is important to note that our study was limited to a small area at a single locality, therefore further studies will be required to determine how levels of polygyny, failed to return. There are likely habitat characteristics that cuckoldry and serial polyandry vary within and among ling- influence nest site choice with the presence of crevices an cod populations. However, anectodal evidence suggests obvious possibility and other factors such as water move- that nest site fidelity exhibited by males may be a common ment (Low & Beamish 1978) or the lack of flora coverage, occurrence for lingcod, at least for the whole reef of this particularly Agarum spp. (King & Winchell 2002), possibly study (King & Winchell 2002). also important. Polygamy in both sexes may maintain a relatively large Fewer males were identified as parents in the 14 nests effective population size of lingcod relative to census size, of 2003 (15 males) than in the 13 nests of 2002 (21 males). especially at low abundances. However, in some species Moreover, the guardian male was not detected in eggs in which males compete for females, the reproductive skew sampled from four nests in 2003 compared to only one nest of males (variance in lifetime reproductive output) is so in 2003. The much smaller numbers of eggs analysed in great that genetic diversity is reduced rather than enhanced 2003 (15 per nest) likely accounts for both observations. relative to monogamy (Storz et al. 2001). The strong site Male contributions to egg fertilization varied greatly in fidelity exhibited by male lingcod in this study provides a samples collected from locations throughout the egg mass means by which to estimate lifetime reproductive outputs for (Withler et al. 2004). However, sampling in both years indi- males and test hypotheses regarding population structure, cated that guardian males with multiple nests tended mate choice and genetic diversity in nest-tending marine fish. to dominate fertilization in only one nest, with auxiliary males achieving more success in the others (Withler et al. Acknowledgements 2004). Guardians may spend less time courting and guard- Several divers assisted J. King with fieldwork: B. Beaith, L. ing females after they have one nest to protect, providing MacDougall, K. Mathias, and P. Winchell. Janine Supernault con- fertilization prospects for neighbouring guardians and non- ducted the microsatellite analyses. J. Martin assisted in production territorial males in subsequent nests. of Fig. 3. D. Haggarty and G. A. McFarlane provided useful The extreme nest site fidelity coupled with polygynous comments on an initial version of this manuscript. mating within and between seasons in male lingcod and complete lack of site fidelity at this scale in female lingcod References leads to polygamy for both sexes in this species. This is con- Beheler AS, Rhodes OE, Weeks HP (2003) Breeding site and mate sistent with the suggestion that the advantages of promis- fidelity in eastern phoebes (Sayornis phoebe) in Indiana. Auk, 120, cuity increase as the level of parental investment decreases, 990–999. regardless of parental gender (Knight 2002). Although female Blumer LS (1979) Male parental care in the bony fishes. Quarterly 54 lingcod do not appear to be batch spawners as are closely Review of Biology, , 149–161. Cass AJ (1985) Lingcod. In: Groundfish Stock Assessments for the related kelp greenling females, this study indicates that they West Coast of Canada in 1984 and Recommended Yield Options for frequently practice serial polyandry by failing to return to 1985 (eds Tyler AV, McFarlane GA), pp. 25–26. Department of the same nesting areas in successive years. In birds, parents Fisheries and Oceans, Canadian Manuscript Report of Fisheries of one or both genders may show nest site fidelity (e.g. and Aquatic Sciences 1813, Ottawa, Canada.

Crow KK, Powers DA, Bernardi G (1997) Evidence for multiple Martell SJD, Wallace SS (1998) Estimating historical lingcod bio- maternal contributors in nests of kelp greenling (Hexagrammos mass in the Strait of Georgia. In: Back to the Future: Reconstructing decagrammus, Hexagrammidae). Copeia, 1997 (1), 9–15. the Strait of Georgia Ecosystem (eds Pauly D, Pitcher T, Preikshot DeMartini EE (1987) Paternal defense, cannibalism and polygamy: D), pp. 45–57. University of British Columbia, Fisheries Centre factors influencing the reproductive success of painted green- Research Report 6(5), Vancouver, Canada. ling (Pisces, Hexagrammidae). Animal Behaviour, 35, 1145– Matthews KR (1992) A telemetric study of the home ranges and 1158. homing routes of lingcod (Ophiodon elongatus) on shallow rocky DeWoody JA, Avise JC (2001) Genetic perspectives on the natural reefs off Vancouver Island, British Columbia. Fishery Bulletin, history of fish mating systems. Journal of Heredity, 92, 167– 90, 784–790. 172. O’Connell VM (1993) Submersible observations on lingcod, Howlett JS, Stutchbury BJM (2003) Determinants of between- Ophiodon elongatus, nesting below 30 m off Sitka. Alaska Marine season site, territory, and mate fidelity in hooded warblers Fisheries Review, 55, 19–24. (Wilsonia citrina). Auk, 120, 457–465. Ridgway MS, MacLean JA, MacLeod JC (1991) Nest-site fidelity in King JR (2001) Assessment of lingcod in the Strait of Georgia. Fish- a centrarchid fish, the smallmouth bass (Micropterus dolomieui). eries and Oceans Canada, Canadian Scientific Advice Secretariat Canadian Journal of Zoology, 69, 3103–3105. Research Document 2001/132, Ottawa, Canada. Storz JF, Bhat HR, Kunz TH (2001) Genetic consequences of poly- King JR, Beaith BW (2001) Lingcod (Ophiodon elongatus) nest den- gyny and social structure in an Indian fruit bat, Cynopterus sphinx. sity scuba survey in the Strait of Georgia, January 16-April 26, II. Variance in male mating success and effective population 2001. Fisheries and Oceans Canada, Canadian Technical Report of size. Evolution, 55, 1224–1232. Fisheries and Aquatic Sciences 2374, Ottawa, Canada. Waits LP, Luikart G, Taberlet P (2001) Estimating the probability King JR, Winchell PM (2002) Lingcod (Ophiodon elongatus) nest of identity among genotypes in natural populations: cautions density scuba survey in the Strait of Georgia, January 15-March and guidelines. Molecular Ecology, 10, 249–256. 13, 2002. Fisheries and Oceans Canada, Canadian Technical Withler RE, King JR, Marliave JB et al (2004) Polygamous mating Report of Fisheries and Aquatic Sciences 2437, Ottawa, Canada. and high levels of genetic variation in lingcod (Ophiodon elongatus) Knight J (2002) Sexual stereotypes. Nature, 415, 254–256. of the Strait of Georgia, British Columbia. Environmental Biology Low CJ, Beamish RJ (1978) A Study of the Nesting Behaviour of Ling- of Fishes, 69, 345–357. cod (Ophiodon elongatus) in the Strait of Georgia, British Colum- Woodward JD, Murphy MT (1999) Sex roles, parental experience bia. Department of Fisheries and Oceans, Fisheries Marine Science and reproductive success of eastern kingbirds, Tyrannus tyrannus. Technical Report 843, Ottawa, Canada. Animal Behaviour, 57, 105–115.