Fish populations are often treated as homogeneous units in typical fishery management, thereby tacitly ignoring potential intraspecific variation which can lead to imprecise management rules. However, intraspecific variation in life-history traits is widespread and related to a variety of factors. We investigated the comparative age-based demography of the two main colour patterns of Labrus bergylta (plain and spotted, which coexist in sympatry), a commercially valuable resource in the NE Atlantic. Individuals were aged based on otolith readings after validating the annual periodicity of annuli deposition. The relationships between the otolith weight and fish age and between otolith length and fish length were strong but differed between colour patterns. The fit of the growth models to the age and length data resulted in divergent growth curves between colour morphotypes and between sexes. Males and spotted individuals attained larger mean asymptotic sizes (Linf) than females and plain individuals, respectively, but converged to them more slowly (smaller k). Estimates of mortality based on catch curves from two independent datasets provided a global total mortality (Z) of 0.35 yr–1, although Z was larger in plain and female individuals. Overall, the results of this research have direct implications for management of L. bergylta and, as a precautionary measure, we recommend considering both colour patterns as two different management units.
Citation: Villegas-Ríos D, Alonso-Fernández A, Fabeiro M, Bañón R, Saborido-Rey F (2013) Demographic Variation between Colour Patterns in a Temperate Protogynous Hermaphrodite, the Ballan Wrasse Labrus bergylta. PLoS ONE 8(8): e71591. doi:10.1371/journal.pone.0071591
Editor: Athanassios C. Tsikliras, Aristotle University of Thessaloniki, Greece
Received: May 31, 2013; Accepted: July 5, 2013; Published: August 23, 2013
Copyright: © 2013 Villegas-Ríos et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the project Artevigo (Ref: 09MMA022402PR), funded by the Regional Government of Galicia (Xunta de Galicia), Spain. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Age-based demographic studies of marine fishes are one of the foundations of population biology and provide essential information for ecosystem and fisheries management . Aged-based information enables assessment of population stability and exploitability and allows for the evaluation of management initiatives . Demographic parameters have been found to differ between populations of the same species geographically separated by tens to thousands of kilometres , , , highlighting the need to investigate local populations rather than apply models based on parameters from other regions for management purposes . The rationale behind this is that fish demography is influenced by a number of variables including temperature, density of competitors, presence of predators, resource availability, fishing pressure and habitat quality , . Demographic variation may also result from the evolution of alternative life-history strategies within and among populations, i.e. life-history variants, which may confer species increased resilience to environmental fluctuations , . Furthermore, some species have developed into reproductively isolated sympatric morphotypes which can eventually promote population evolutionary divergence and ultimately, speciation . Failing to consider this level of intraspecific variation in the demographic parameters can result in inadvertent undesirable outcomes and can have important implications for the sustainability and productivity of fisheries , .
The ballan wrasse Labrus bergylta is a commercially valuable wrasse with a large contribution to both recreational and artisanal commercial catches throughout its distribution range. The species exists in coastal areas of the NE Atlantic and the Mediterranean, and is primarily associated with rocky reefs and kelp beds. The colour pattern of L. bergylta is highly variable , , but unlike other protogynous wrasses , ,  it is not related to the sex of the individuals , , . Two main colour patterns are usually differentiated and coexist in sympatry. Plain coloured individuals (Figure 1a) are characterized by a uniform, although variable, body colour (mainly greenish, brownish or reddish), darker in the back and whitish in the abdomen. Spotted coloured individuals (Figure 1b) display a dark orange or reddish body patterned with white dots. In NW Spain, there is a strong local belief that these are two different species and accordingly, spotted and plain individuals are usually commercialized separately (spotted individuals are noticeably more expensive). The presence of both colour patterns has been confirmed in Galicia , Azores Islands (J. Azevedo, pers. comm.), France (K. Nedreaas, pers. comm.), English Channel (C. Meyer, pers. comm.) and Norway (K. Nedreaas, pers. comm.). However, there are no reports on the biological significance of colour variation in L. bergylta in the scientific literature and the extent of ecological and/or life-history differences between colour morphs remains unknown.
Example of (a) plain coloured individual (total length = 39.2 cm) and (b) spotted coloured individual (total length = 43.6 cm). Images are not in the same scale.
Labrus bergylta has being described as a protogynous hermaphrodite species , . However, management is based solely on a minimum landing size which varies along its distribution range (typically <23 cm), in spite of the evidence that in sequential hermaphrodite populations larger individuals should be also protected to ensure sex ratios are sufficient for effective mating and fertilization . Management fails to consider the intraspecific variation in life-history traits. One reason for this mismatch between life-history and management is the existence of serious gaps in our knowledge of the age based-demographic parameters of L. bergylta, specially the potential differences between plain and spotted individuals. Previous studies were limited to report age based on scales and opercular bones, revealing that L. bergylta was a slow growing fish and males were larger at age than females. Maximum reported ages were 29 yr for males and 25 yr for females . In this regard, the present study provides the first estimation of the otolith-based demographic parameters for the species. Our primary objective was to investigate the comparative demography of the plain and spotted individuals in the south of its distribution range. Since this is a protogynous fish , differences between sexes were also explored. As an essential first step, we validated the ageing criteria and the periodicity of annuli deposition in the otoliths. This information will be highly relevant to assess the status of the population and the fishing impacts, and to evaluate alternative management strategies of this important resource in the NE Atlantic.
Materials and Methods
No specific permission for sampling was required for this study as all the individuals sampled were obtained from commercial fishing at the local fish markets. No protected species were sampled. Captive conditions and experiments with captive fish were carried out as part of an official research project, according to the European and Spanish legislation and specifically approved by Spanish authorities (09MMA022402PR).
Two different datasets were used to accomplish the objectives of this research. The main one (markets –MKT- dataset) was composed of information on 1583 individuals regularly obtained from the local fish markets in the south of Galicia, NW Spain (42°15’N, 8°50’W), from December 2009 to January 2012. These individuals were fished with gillnets at depths from 0 to 60 m. The sampling was conducted randomly to obtain a representative sample of the population, with the exception of the smaller size classes (typically under 15 cm) which are not captured by gillnets. For all the individuals in the MKT dataset, the colour pattern (plain vs. spotted), total length (TL;±1 mm) and total weight (TW; ±0.1 g) were recorded. Sex of the individuals was determined microscopically after histological processing of the gonads (see details in ) and sagittal otoliths were removed, washed in freshwater and stored dry until processing. The second dataset (UTPB dataset), composed by the historical (2000–2011) artisanal fishing monitoring data from the Coastal Fisheries Technical Unit of the Galician Regional Government (Xunta de Galicia), was used for mortality estimates. This dataset is considered a random sampling of the commercial catch. It included 10,163 records of Labrus bergylta (with no reference to the colour pattern or sex) measured (TL) to the nearest 1 cm. UTPB data were obtained by fisheries observers on board of randomly selected artisanal boats who registered all the individuals of L. bergylta captured in each fishing trip.
Data processing and all analyses were conducted in R . In all statistical analyses, significance level was set to α = 0.05, and selection of the optimal models was based on Akaike’s Information Criteria .
Validation of ageing criteria
To be useful for ageing purposes, otoliths must 1) grow continuously throughout the life of the fish, 2) display visible growth increments, and 3) those increments must be formed on a regular and determinable timescale . Continuous growth of otoliths was assessed by analysing the relationship between fish age and otolith weight and between fish TL and otolith length using linear models. Both analyses included fish colour pattern and sex (and their interaction) as factors in order to assess intraspecific variation in these relationships. A strong relationship was expected if otoliths accreted calcium carbonate throughout the life of the fish. For this purpose, a subsample of 498 otoliths chosen randomly for each pair was weighed (±0.1 mg). Another random subsample of 243 complete and intact otoliths was photographed with a digital camera attached to a stereomicroscope and measured in length (±0.01 mm) with image analysis software.
A visual inspection of the whole otoliths evaluated the existence of visible growth increments. The rate at which those increments were formed in the otoliths was assessed using oxytetracycline (OTC) marking . Eight individuals were injected with 50 mg of OTC per kg of fish on February 1st 2011, and kept in 700 L tanks for a growth period of 433 d after OTC injection. They were maintained under controlled natural photoperiod and water temperature and fed ad libitum with mussels and sandworms 3 times per week. After the growth period, the 8 individuals were euthanized and otoliths removed and examined with a fluorescence stereomicroscope. Photos were taken with a digital camera attached to the microscope, and measurements (±0.01 mm) were made with image analysis software. The model of Cappo et al.  modified by Pears et al.  was used to obtain estimates of the periodicity of annuli formation. This model proposes that if each annulus (constituted by one opaque and one translucent zone) is formed in 1 year, then the distance from the OTC mark to the outer edge of the otolith, divided by the width of the last complete cycle, should be equal to the time elapsed from mark to death. In this model the periodicity of increment formation, P (d–1), is:
where x is the distance from the OTC mark to the edge of the otolith (mm), y is the distance of the last complete increment (mm) and G is the growth period (d).
Complementarily, marginal increment analysis and edge-type analysis  were undertaken. We measured and analysed the marginal increments for a subsample of individuals of age classes 5 (n = 68) and 6 (n = 68). For edge type analysis otoliths with indeterminate edge type (due to their inherent variability) were excluded. Thus, only otoliths classified as “G” (Table 1; ) from ages 3 to 12 were retained for analysis. After validating the annual nature of increment deposition, otoliths of 919 individuals from the MKT dataset covering the full range of fish TL, colour patterns, sexes and months were examined for age determination.
Age readings were conducted on a stereomicroscope using reflected light over the whole otolith immersed in mineral oil. Age counting was made along a consistent axis near the sulcus. Age assignment was based on the annulus count, the edge type in relation to collection date and assigned birth date (by convention: January 1st). The primary reader examined each otolith (n = 919) on two occasions without prior knowledge of the fish characteristics, and assigned a reader confidence index (adapted from ) to the readings (Table 1). A second reader examined a random subsample of the otoliths (n = 222) on one occasion. The counts from both readers were compared, and the precision of age estimates calculated using the Index Average Percent Error (IAPE ). In addition, all the otoliths classified as “F”, “FP” or “P” were analysed two more times by the primary reader until consensus was reached. In the case that no agreement was achieved after these readings, these otoliths were considered unreadable and disregarded for the analysis.
The relationship between fish TL and fish TW was described by a power equation. The relationship was linearized by log transforming data and a linear model was adjusted. Differences in the relationship between sexes and colour patterns were assessed by including these variables (and their interaction) as factors in the model.
Growth parameters were estimated using the iterative non-linear least squares method within the nlstools package using the Gauss-Newton algorithm  to fit the von Bertalanffy growth function (VBGF):where Lt = size at age t; Linf = asymptotic length; K = curvature parameter that represents the rate at which the asymptotic length is approached; t = age of fish and t0 = the age at which fish have theoretical length of 0 cm. Because the VBGF parameters estimate can be sensitive to the range of ages and sizes used and no fish of age 1 were collected, the intercept was constrained to 0 cm , . Parameters Linf and K and their confidence intervals were obtained by bootstrapping with 1000 iterations . Potential differences in growth curves between colour patterns, between sexes, and between sexes within each colour pattern were investigated using bivariate 95% confidence ellipses surrounding the K and Linf estimates . Non-overlapping confidence regions indicate differences in the growth parameters , , . For comparison, we fitted and calculated the parameters of the unconstrained VBGF, which is provided as Supporting Information. In addition, since the biological interpretation of the VBGF parameters has been questioned when comparing them between populations , , , the VBGF was reparametrized after Francis  (Text S1).
For the remaining demographic analysis, age was assigned to the 682 individuals in the MKT dataset that were not directly aged using sex- and colour-specific age-length keys created with the aged individuals. Differences in the mean size and age between sexes and colour patterns of the individuals of the MKT dataset were then compared using t-tests. Mean maximum age (longevity) and mean maximum size were calculated as the average age (yr) and average TL (cm) of the oldest 10% individuals  of the MKT dataset. Wilcoxon test was used to detect differences in longevity and mean maximum size between plain and spotted individuals.
Age-based catch curves were used to estimate the instantaneous rate of total mortality (Z) for L. bergylta ,  under the following assumptions: (1) no trend in recruitment over time, (2) no trend in the fishing mortality rate over time, (3) constant natural mortality at age for the analysed ages, and (4) constant selectivity at age for the analysed ages , . Catch curves were generated by plotting the logarithm of the number of individuals by age class. A linear model was then fitted to the descending limb of the catch curve. The slope of this regression is an estimate of Z. Linear models were fitted from a minimum age of one year older than the maximum catch frequency (assumed full selectivity) up to the oldest age class that was preceded by no more than 2 consecutive zero frequencies , . The annual mortality rate (A), i.e., the proportion of the population that suffers mortality in a given year, was calculated as:
Catch curves were analysed with the FSA package .
A global value of Z for the species was calculated using data from 2011, including the dataset (MKT and UTPB) as a factor to test for differences between them (age of the individuals of the UTPB dataset was assigned from an age-key length created with the aged individuals of the MKT dataset). In addition, Z was estimated by cohort using the UTPB dataset. Eight cohorts (1997–2004) were selected based on the existence of a sufficient number of data and age classes. Sex- and colour-based specific mortality was investigated for the individuals in the MKT dataset. Due to the low number of males, an analysis of the interaction between sex and colour patterns was not possible. Therefore, we compared Z estimates by sex and colour pattern independently. Finally, colour-based differences in Z amongst females were evaluated.
Validation of ageing criteria
A significant positive linear relationship was observed between fish age and otolith weight (r2 = 0.79, F = 453.2, p<0.001, n = 498) but it differed significantly between plain and spotted individuals and between sexes (Figure 2a and Table 2). A positive significant relationship was also obtained between fish TL and otolith length (r2 = 0.75, F = 243.8, p<0.001, n = 243) with significant differences between colour patterns (Figure 2b and Table 2).
Linear model fit on (a) otolith weight on fish age and (b) otolith length on fish total length (TL).
A visual inspection of whole otoliths under the stereomicroscope revealed a consistent and regular distribution of presumed annuli in the otoliths, with an alternation of opaque and translucent zones in all age classes. All the 8 injected individuals successfully revealed the OTC mark as a bright green line (Figure 3). Captivity conditions decreased the differentiation between opaque and translucent zones. Therefore, identification of annuli outside the OTC mark (and subsequent estimation of periodicity) was only possible for 6 individuals (Table 3). The calculated periodicity of annuli formation was 346±24 d (hereinafter all values expressed as mean ± standard deviation). The marginal increment displayed a cyclical trend (Figure 4a) with minimum values from June to September and a subsequent increase until the start of formation of a new opaque zone in March. The plot of monthly percentages of otoliths with translucent marginal increments (Figure 4b) revealed maximum values from December to March (slow growth season) and minimum values from July to September (rapid growth season).
Details of OTC-marked otolith showing (a) 5 annual bands in a captive individual and (b) OTC mark on the same area under fluorescent light. Note the presence of a secondary OTC mark irrelevant for the purpose of this study (OTC injected later as part of another experiment).
Plot of (a) mean and standard deviation of marginal increment for the 5 and 6 yr classes and (b) monthly percentage of otoliths with translucent margin increments.
The log transformed fish TL and fish TW fitted to a positive linear model (r2 = 0.97, F = 16740, p<0.001, n = 1583), with significantly different results between sexes and colour patterns (Table 4).
Size and age distributions of the MKT samples showed different degrees of overlap between sexes and colour patterns (Figure 5). Males were larger (t = 14.91, p<0.001, df = 663.4) and older (t = 18.84, p<0.001, df = 498.5) than females. Yet females persisted in the larger and older groups. As for the colour pattern, spotted were larger than plain individuals (t = –5.74, p<0.001, df = 611.2). Conversely, plain individuals were older than spotted individuals (t = 4.45, p<0.001, df = 859.8). Differences in age and size structure by sex were additionally investigated separately for each colour pattern. Males were significantly larger (t = 18.32, p<0.001, df = 912.9) and older (t = 16.61, p<0.001, df = 424.2) than females within the plain group. These differences were even more noticeable in the spotted individuals, with males also significantly larger (t = 11.16, p<0.001, df = 33.4) and older (t = 9.17, p<0.001, df = 34.1) than females.
Independent plots are presented for the sex-specific size (a) and age (b) and colour-specific size (c) and age (d) distributions of all the individuals pooled. Additional plots are provided for sex-specific size (e) and age (f) distributions for plain individuals and sex-specific size (g) and age (h) distributions for spotted individuals.
Age and growth.
The precision of opaque-zone counts for Labrus bergylta was high with an IAPE of 1.8% (n = 222) between readers. There was good agreement among counts across all age classes, with 66.5% of the two counts being identical, 21.7% differing in one year and the other 11.8% differing in two to four years. The number of otoliths assigned within each class of the confidence index by the primary reader (Table 1) indicated that most otoliths were fair or good to read, with relatively minor and inconsistent differences between otoliths. Eighteen otoliths were considered as unreadable.
Size-at-age plots were created with the 901 individuals ultimately aged. Size variability increased with age in all the analysed groups (Figure 6). Fitting VBGF to each dataset gave the parameter values shown in Table 5, and the presence of non-overlapping 95% bivariate confidence intervals surrounding the parameter estimates (Figure 7) revealed significant differences between the compared datasets. Growth trajectories differed between sexes (Figure 6a) with males resulting in significantly larger Linf and smaller K estimations (Table 5, Figure 7). Partitioning the plot by colour pattern revealed increasingly divergent growth trajectories (Figure 6b) with plain individuals attaining significantly smaller Linf and larger K estimations than spotted individuals (Table 5, Figure 7). Within each colour pattern (Figures 6c and 6d), sex-specific differences were more evident in the spotted than in the plain individuals (Table 5, Figure 7). For comparative purposes, the parameters of the unconstrained VBGF are presented in Table S1, and the parameters of the rVBGF are presented in Table S2.
Von Bertalanffy growth functions (VBGF) plotted by (a) sex, b) colour pattern, c) sex within plain individuals and d) sex within spotted individuals.
Ellipses surround the growth coefficient (K) and mean asymptotic length (Linf) parameter estimates for the different sexes and colour patterns of Labrus bergylta. Non-overlapping confidence regions indicate differences in the growth parameters.
Longevity and mortality.
Estimated mean maximum age (longevity) for L. bergylta was 12.70 yr (±2.32) with no differences between colour patterns (W = 2940, p = 0.156, n = 158). Conversely, estimated mean maximum size was different for plain and spotted individuals (W = 554.5, p<0.001, n = 158), being 36.54±3.36 cm and 46.76±6.46 cm, respectively.
Estimate of Z with data from 2011 were 0.35±0.03 and with no significant differences between datasets (p = 0.666), and as a result A = 29.5%. Significant differences in mortality among cohorts were observed (r2 = 0.89, p = 0.029, F = 2.501, n = 61) (Figure 8). Analysis of the MKT dataset revealed significant differences in mortality between colour patterns (r2 = 0.90, F = 8.644, p = 0.008, n = 26) with Zplain = 0.42±0.03 and Zspotted = 0.28±0.05, and as a result Aplain = 34.3% and Aspotted = 24.4%. Differences in sex-specific mortality were also significant (r2 = 0.92, F = 16.07, p<0.001, n = 25), with Zmales = 0.32±0.05, Zfemales = 0.51±0.04, Amales = 27.4% and Afemales = 39.9%. Within the females, non-significant differences were found in Z estimates between colour patterns (p = 0.490).
This study represents the first estimate of the age-based demographic parameters of Labrus bergylta based on otoliths, and revealed striking demographic differences between the two main colour patterns of the species. Plain and spotted individuals displayed differences in the length-weight relationship, otolith and body growth, size and age structures and mortality. We estimated the age-based demographic parameters of L. bergylta using whole otoliths, which are preferred to scales and opercular bones as they are not reabsorbed or metabolically reworked and are not exposed to external damage . Whole otoliths, together with the break and burn method, is one of the most rapid and widespread methods for fish age determination , .
The relationship between otolith weight and fish age and between otolith length and fish size showed that individuals accreted calcium carbonate in the otoliths throughout the lifetime of the fish. The robust relationship between fish size and otolith length permits the back calculation of fish body size at younger ages in individual growth studies , but specific parameters for each colour pattern should be used. The whole otoliths under reflected light showed a clear internal structure of alternating narrow translucent and wide opaque increments, as observed in other labroids . The analysis of fish marked with OTC confirmed that L. bergylta produced 1 annulus per year on the otoliths. This result was confirmed by the presence of a single yearly minimum and maximum in the edge type and marginal increment analysis. The maximum percentage of fish with translucent margins found in winter (slow growth period) corresponds to the coldest water temperatures in the study area .
The observed maximum age in Galicia in this study (22 yr) is smaller than previously reported in the Irish Sea (29 yr; ) and represents a modest longevity within the family Labridae , . Reasons behind this difference in longevity may be related to the lower water temperature in the Irish Sea (http://iridl.ldeo.columbia.edu last accessed: 23/05/2013). It’s recognized that adult longevity increases when there is a decline in average water temperature during the individuals’ development , . Variation could also be related to differing selective pressures between populations from each location , different hard structures used for ageing (otoliths vs. opercular bones) or sampling bias.
Size-at-age plots revealed that L. bergylta invested a relatively small proportion of its life span in an initial rapid somatic growth as other labrid species , , . Growth curves revealed sex-specific differences with males attaining larger sizes (Linf) than females but at a slower rate (K), as observed in other labroids , . This might be explained by a growth spurt after sex change, as previously suggested , . The fit of VBGF separately for each colour pattern also revealed an important divergence in the growth curves after a rapid initial growth (ages <5 yr). Differences in Linf estimates between colour patterns were as much as 11 cm which represents an increase of ∼31% in spotted individuals. K values were higher for plain and female individuals, indicating more rapid convergence on their asymptotic sizes (but do not necessarily grow more rapidly) than spotted individuals and males, respectively. The same conclusions can be drawn from the unconstrained fit of the VBGF, although a complete comparison between the two fits was not possible as the models failed to converge in some cases. The constrained and unconstrained fits of the VBGF to the size-at-age data likely provided an indication of the extremes of the growth pattern for L. bergylta, although the true growth pattern most likely lies somewhere between the two fits.
Size structures revealed that spotted individuals were larger than plain individuals, which is in agreement with the differences in the described growth pattern. However, the age distributions completely overlapped, with both plain and spotted individuals present in almost all age classes. The hypothesis of a colour shift with age  is thus improbable. The sex-specific size and age distributions support  the conclusion of protogyny in L. bergylta , , and the same conclusion of protogyny can be drawn for each colour if they are considered separately.
The observed larger mortality in plain individuals (∼1.5 larger than in spotted individuals) and in females (∼1.6 larger than in males) may be related to size-related predation risk (i.e. reduction of predation with size ). Spotted individuals attain larger sizes-at-age, and would therefore be subject to a lower predation risk during their life span, especially in the male phase, since no colour-based differences in mortality were found within the females. However, the mortality pattern of the species seems to be complex, as indicated by the differences in cohort-specific mortality rates in 8 consecutive year-classes (without any clear temporal pattern). These data suggest that shifts in exploitation occur and that these changes may differentially alter sex ratios and colour patterns. In this regard, a more comprehensive population dynamic model that considers sex change and colour patterns should be conducted.
This study revealed striking colour-based demographic variation between plain and spotted morphotypes of the temperate wrasse Labrus bergylta. The observed age-based demographic differences are tightly related to the recently reported reproductive traits of the species. For example, Villegas-Ríos et al.  observed a difference of 11.2 cm in the size at 50% sex change (36.0 cm for plain individuals, 47.2 cm for spotted individuals) that may be explained by the larger size-at-age of the spotted individuals. In the same study, it was found that the gonadosomatic index (GSI) of the spotted females was significantly lower than the GSI of the plain females. This is consistent with our results from the point of view of the general theory of energy allocation, which states that a trade-off should exist between the energy allocated to growth and reproduction . For a given size, we have demonstrated that plain individuals attained smaller sizes-at-age likely at the cost of their greater reproductive output . However, the basis of the demographic differences between plain and spotted individuals remains unknown. Do plains and spotted morphotypes belong to reproductively isolated populations of the same species? Are the observed differences related to habitat or diet preferences? On one hand, it has been recognized that the combination of assortative mating and disruptive natural selection on a single trait (for example body colour pattern , , ) can initiate speciation in marine environments, even in the absence of geographic barriers (sympatry) and high gene flow , , . The benthic spawning and nesting behaviour of the species , together with likely local larval retention and low adult dispersion  would enhance a potential speciation process . However, only a detailed genetic analysis with specific molecular markers would help to elucidate the existence and degree of divergence or disruptive selection between colour patterns and the presence of gene flow between them , . On the other hand, differences in diet composition or habitat preference between plain and spotted individuals are currently unexplored. In the Gulf of Maine some cod individuals are resident and feed primarily on benthic fauna rich in carotenoids, which is thought to provide them the red colouration. Other cod individuals are transient, feed on forage fish and display an olive colouration . Red and olive cods are considered two life-history variants as they also display differences in growth, habitat preferences, and body shape. Differences in movement behaviour between colour patterns of L. bergylta have never been reported and plain and spotted individuals seem to completely overlap in their depth ranges and habitat preferences (pers. obs.). Therefore it seems unlikely that differences in habitat utilization would explain the colour-based demographic variation. In addition, a comparative analysis of the diet of plain and spotted L. bergylta has never been analysed. In conclusion, although the demographic differences (this study) and the reproductive traits  are noticeable, the identity of the colour patterns of Labrus bergylta remains unresolved and deserves additional research effort. However, the demographic differences between the two morphotypes reported here should be directly applied to the management of this valuable resource in the NE Atlantic, which is solely regulated by minimum landing sizes. We recommend considering plain and spotted morphotypes as two independent management units throughout the distribution range of L. bergylta, at least while the status of the species is unresolved.
Parameters of the unconstrained Von Bertalanffy growth functions with upper and lower 95% confidence intervals and Akaike Information Criteria (AIC) for each model.
Reparametrized von Bertalanffy growth function (rVBGF) parameters estimates with upper and lower 95% confidence intervals.
We would like to thank all the people from the Fisheries Ecology Group (IIM-CSIC) who helped with the logistic of this research. Special thanks to F. Pinto for her suggestions to improve the manuscript, Sarah Walters for English corrections as well as the reviewers by their helpful comments. Also thanks to personal of UTPB that provide data of the monitoring programme of the artisanal fishery of Galicia. D. Villegas-Ríos was partially financed by a FPU Scholarship from the Spanish Ministry of Economy and Competitiveness. This paper is dedicated to Nico.
Conceived and designed the experiments: DVR AAF FSR. Performed the experiments: DVR MF RB AAF. Analyzed the data: DVR AAF MF RB FSR. Contributed reagents/materials/analysis tools: DVR MF. Wrote the paper: DVR FSR.
- 1. Stearns SC (1992) The evolution of life histories. New York: Oxford University Press. 249 p.
- 2. Caldow C, Wellington GM (2003) Patterns of annual increment formation in otoliths of pomacentrids in the tropical western Atlantic: Implications for population age-structure examination. Mar Ecol Prog Ser 265: 185–195. doi: 10.3354/meps265185
- 3. Meekan MG, Ackerman JL, Wellington GM (2001) Demography and age structures of coral reef damselfishes in the tropical eastern Pacific Ocean. Mar Ecol Prog Ser 212: 223–232. doi: 10.3354/meps212223
- 4. Gust N, Choat JH, Ackerman JL (2002) Demographic plasticity in tropical reef fishes. Mar Biol 140: 1039–1051. doi: 10.1007/s00227-001-0773-6
- 5. Choat JH, Axe LM (1996) Growth and longevity in acanthurid fishes; An analysis of otolith increments. Mar Ecol Prog Ser 134: 15–26. doi: 10.3354/meps134015
- 6. Caselle JE, Hamilton SL, Schroeder DM, Love MS, Standish JD, et al. (2011) Geographic variation in density, demography, and life history traits of a harvested, sex-changing, temperate reef fish. Can J Fish Aquat Sci 68: 288–303. doi: 10.1139/f10-140
- 7. Berumen M, Trip E, Pratchett M, Choat JH (2012) Differences in demographic traits of four butterflyfish species between two reefs of the Great Barrier Reef separated by 1,200 km. Coral Reefs 31: 169–177. doi: 10.1007/s00338-011-0838-z
- 8. Jonsson B, Jonsson N (1993) Partial migration: niche shift versus sexual maturation in fishes. Rev Fish Biol Fish 3: 348–365. doi: 10.1007/bf00043384
- 9. Sherwood GD, Grabowski JH (2010) Exploring the life-history implications of colour variation in offshore Gulf of Maine cod (Gadus morhua). ICES J Mar Sci 67: 1640–1649. doi: 10.1093/icesjms/fsq094
- 10. Bernatchez L, Vuorinen JA, Bodaly R, Dodson JJ (1996) Genetic evidence for reproductive isolation and multiple origins of sympatric trophic ecotypes of whitefish (Coregonus). Evolution 50: 624–635. doi: 10.2307/2410836
- 11. Morita K, Morita SH, Yamamoto S (2009) Effects of habitat fragmentation by damming on salmonid fishes: lessons from white-spotted charr in Japan. Ecol Res 24: 711–722. doi: 10.1007/s11284-008-0579-9
- 12. Costello MJ (1991) Review of the biology of wrasse (Labridae:Pisces) in Northern Europe. Prog Underw Sci 16: 29–51.
- 13. Villegas-Ríos D, Alonso-Fernández A, Dominguez-Petit R, Saborido-Rey F Intraspecific variability in reproductive patterns in the temperate hermaphrodite Labrus bergylta. Mar Freshw Res. In press.
- 14. Alonso-Fernández A, Alós J, Grau A, Domínguez-Petit R, Saborido-Rey F (2011) The Use of Histological Techniques to Study the Reproductive Biology of the Hermaphroditic Mediterranean Fishes Coris julis, Serranus scriba, and Diplodus annularis. Mar Coast Fish 3: 145–159. doi: 10.1080/19425120.2011.556927
- 15. Porteiro F, Barreiros JP, Santos RS (1996) Wrasses (Teleostei: Labridae) of the Azores. Archipelago: Life and Marine Sciences 14A: 23–40.
- 16. Hamilton SL, Caselle JE, Standish JD, Schroeder DM, Milton SL, et al. (2007) Size-selective harvesting alters life histories of a temperate sex-changing fish. Ecol Appl 17: 2268–2280. doi: 10.1890/06-1930.1
- 17. Dipper FA, Pullin RSV (1979) Gonochorism and sex-inversion in British Labridae (Pisces). J Zool 187: 97–112. doi: 10.1111/j.1469-7998.1979.tb07716.x
- 18. Talbot C, Medeiros MV, Davie A (2012) In vivo gender determination in captive ballan wrasse (Labrus bergylta Ascanius 1767). Final Report to the Scottish Salmon Producers Organisation 54 Available: http://www.nafc.ac.uk/WebData/Files/Wrasse%20gender_SSPO%20final%20report.pdf.
- 19. Alonzo SH, Mangel M (2004) The effects of size-selective fisheries on the stock dynamics of and sperm limitation in sex-changing fish. Fish Bull 102: 1–13.
- 20. Dipper FA, Bridges CR, Menz A (1977) Age, growth and feeding in the ballan wrasse Labrus bergylta Ascanius 1767. J Fish Biol 11: 105–120. doi: 10.1111/j.1095-8649.1977.tb04103.x
- 21. R Development Core Team (2011) R: a language and environment for statistical computing. Vienna, Austria. ISBN 3-900051-07-0. Available: http://www.R-project.org.: R Foundation for Statistical Computing.
- 22. Zuur AF, Ieno EN, Smith GM (2007) Analysing ecological data: New York: Springer. 672 p.
- 23. Fowler AJ (1990) Validation of annual growth increments in the otoliths of a small, tropical coral reef fish. Mar Ecol Prog Ser 64: 25–38. doi: 10.3354/meps064025
- 24. Campana SE (2001) Accuracy, precision and quality control in age determination, including a review of the use and abuse of age validation methods. J Fish Biol 59: 197–242. doi: 10.1006/jfbi.2001.1668
- 25. Cappo M, Eden P, Newman SJ, Robertson S (2000) A new approach to validation of periodicity and timing of opaque zone formation in the otoliths of eleven species of Lutjanus from the central Great Barrier Reef. Fish Bull 98: 474–488.
- 26. Pears RJ, Choat JH, Mapstone BD, Begg GA (2006) Demography of a large grouper, Epinephelus fuscoguttatus, from Australia's Great Barrier Reef: Implications for fishery management. Mar Ecol Prog Ser 307: 259–272. doi: 10.3354/meps307259
- 27. ICES (2006) Workshop on Age Determination of Redfish (WKADR), 28 August – 1 September 2006, Vigo, Spain. ICES CM 2006/RMC 09: 43.
- 28. Beamish RJ, Fournier DA (1981) A Method for Comparing the Precision of a Set of Age Determinations. Can J Fish Aquat Sci 38: 982–983. doi: 10.1139/f81-132
- 29. Baty F, Delignette-Muller ML (2011) nlstools: tools for nonlinear regression diagnostics.
- 30. Choat JH, Davies CR, Ackerman JL, Mapstone BD (2006) Age structure and growth in a large teleost, Cheilinus undulatus, with a review of size distribution in labrid fishes. Mar Ecol Prog Ser 318: 237–246. doi: 10.3354/meps318237
- 31. Williams AJ, Davies CR, Mapstone BD, Currey LM, Welch DJ, et al. (2009) Age-based demography of humpback grouper Cromileptes altivelis: Implications for fisheries management and conservation. Endang Species Res 9: 67–79. doi: 10.3354/esr00237
- 32. Kimura DK (1980) Likelihood methods for the von Bertalanffy growth curve. Fish Bull 77: 765–774.
- 33. Rhodes K, Taylor B, McIlwain J (2011) Detailed demographic analysis of an Epinephelus polyphekadion spawning aggregation and fishery. Mar Ecol Prog Ser 421: 183–198. doi: 10.3354/meps08904
- 34. Choat JH, Robertson DR, Ackerman JL, Posada JM (2003) An age-based demographic analysis of the Caribbean stoplight parrotfish Sparisoma viride. Mar Ecol Prog Ser 246: 265–277. doi: 10.3354/meps246265
- 35. Mulligan TJ, Leaman BM (1992) Length-at-Age Analysis: Can You Get What You See? Can J Fish Aquat Sci 49: 632–643. doi: 10.1139/f92-072
- 36. Trip EL, Choat JH, Wilson DT, Robertson DR (2008) Inter-oceanic analysis of demographic variation in a widely distributed Indo-Pacific coral reef fish. Mar Ecol Prog Ser 373: 97–109. doi: 10.3354/meps07755
- 37. Craig P (2005) The von Bertalanffy growth curve: when a good fit is not good enough. Naga 22: 28–29.
- 38. Francis R (1988) Are growth parameters estimated from tagging and age-length data comparable? Can J Fish Aquat Sci 45: 936–942. doi: 10.1139/f88-115
- 39. Beverton RJH, Holt SJ (1957) On the dynamics of exploted fish populations. London: Invest., Ser. II, Vol XIX. 533 p.
- 40. Pauly D (1984) Fish population dynamics in tropical waters. ICLARM Stud Rev 8: 1–325. doi: 10.1086/414824
- 41. Chapman DG, Robson DS (1960) The analysis of a catch curve. Biometrics 16: 354–368. doi: 10.2307/2527687
- 42. Thorson JT, Prager MH (2011) Better catch curves: incorporating age-specific natural mortality and logistic selectivity. Trans Am Fish Soc 140: 356–366. doi: 10.1080/00028487.2011.557016
- 43. Wakefield CB, Newman SJ, Molony BW (2010) Age-based demography and reproduction of hapuku, Polyprion oxygeneios, from the south coast of Western Australia: Implications for management. ICES J Mar Sci 67: 1164–1174. doi: 10.1093/icesjms/fsq021
- 44. Ogle DH (2012) FSA: Fisheries Stock Analysis. R package version 0.2–8.
- 45. Campana SE, Neilson JD (1985) Microstructure of fish otoliths. Can J Fish Aquat Sci 42: 1014–1032. doi: 10.1139/f85-127
- 46. Panfili J, de Pontual H, Troadec H, Wright P (2002) Manual of Fish Sclerochronology. BrestFrance: Ifremer-IRD coedition. 464 p.
- 47. Committee of Age Reading Experts (2006) Manual on generalized age determination procedures for groundfish: Pacific States Marine Fisheries Commission. 52 care.psmfc.org/docs/CareManual2006.pdf p.
- 48. Pilling GM, Kirkwood GP, Walker SG (2002) An improved method for estimating individual growth variability in fish, and the correlation between von Bertalanffy growth parameters. Can J Fish Aquat Sci 59: 424–432. doi: 10.1139/f02-022
- 49. Gordoa A, Molí B, Raventós N (2000) Growth performance of four wrasse species on the north-western Mediterranean coast. Fish Res 45: 43–50. doi: 10.1016/s0165-7836(99)00094-6
- 50. Nogueira E, Pérez FF, Ríos AF (1997) Seasonal patterns and long-term trends in an estuarine upwelling ecosystem (Ria de Vigo, NW Spain). Estuar Coast Shelf Sci 44: 285–300. doi: 10.1006/ecss.1996.0119
- 51. Cossington S, Hesp SA, Hall NG, Potter IC (2010) Growth and reproductive biology of the foxfish Bodianus frenchii, a very long-lived and monandric protogynous hermaphroditic labrid. J Fish Biol 77: 600–626. doi: 10.1111/j.1095-8649.2010.02706.x
- 52. Atkinson D, Begon M, Fitter AH (1994) Temperature and organism size -a biological law for ectotherms? Adv Ecol Res Volume 25: 1–58.
- 53. Munch SB, Salinas S (2009) Latitudinal variation in lifespan within species is explained by the metabolic theory of ecology. P Natl Acad Sci USA 106: 13860–13864. doi: 10.1073/pnas.0900300106
- 54. Choat JH, Axe LM, Lou DC (1996) Growth and longevity in fishes of the family Scaridae. Mar Ecol Prog Ser 145: 33–41. doi: 10.3354/meps145033
- 55. Linde M, Palmer M, Alós J (2011) Why protogynous hermaphrodite males are relatively larger than females? Testing growth hypotheses in Mediterranean rainbow wrasse Coris julis (Linnaeus, 1758). Environ Biol Fish 92: 337–349. doi: 10.1007/s10641-011-9844-9
- 56. Artüz ML (2005) Age and growth of the ballan wrasse in the Sea of Marmara Labrus bergylta Ascanius 1767 Hidrobiologica. 03: 17–21. doi: 10.4172/2155-9546.1000146
- 57. Sadovy Y, Shapiro DY (1987) Criteria for the diagnosis of hermaphroditism in fishes. Copeia 1: 136–156. doi: 10.2307/1446046
- 58. Muñoz RC, Warner RR (2004) Testing a new version of the size-advantage hypothesis for sex change: Sperm competition and size-skew effects in the bucktooth parrotfish, Sparisoma radians. Behav Ecol 15: 129–136. doi: 10.1093/beheco/arg086
- 59. Roff DA (1983) An allocation model of growth and reproduction in fish. Can J Fish Aquat Sci 40: 1395–1404. doi: 10.1139/f83-161
- 60. Puebla O, Bermingham E, Guichard F, Whiteman E (2007) Colour pattern as a single trait driving speciation in Hypoplectrus coral reef fishes? Proc R Soc Lond B 274: 1265–1271. doi: 10.1098/rspb.2006.0435
- 61. Puebla O (2009) Ecological speciation in marine v. freshwater fishes. J Fish Biol 75: 960–996. doi: 10.1111/j.1095-8649.2009.02358.x
- 62. Elmer KR, Lehtonen TK, Meyer A (2009) Color assortative mating contributes to sympatric divergence of neotropical cichlid fish. Evolution 63: 2750–2757. doi: 10.1111/j.1558-5646.2009.00736.x
- 63. Barreto FS, McCartney MA (2008) Extraordinary AFLP fingerprint similarity despite strong assortative mating between reef fish color morphospecies. Evolution 62: 226–233. doi: 10.1111/j.1558-5646.2007.00285.x
- 64. Choat JH, Klanten OS, Van Herwerden L, Robertson DR, Clements KD (2012) Patterns and processes in the evolutionary history of parrotfishes (Family Labridae). Biol J Linn Soc 107: 529–557. doi: 10.1111/j.1095-8312.2012.01959.x
- 65. Sjolander S, Larson H, Engstrom J (1972) On the reproductive behaviour of two labrid fishes, the ballan wrasse (Labrus bergylta) and Jago's goldsinny (Ctenolabrus rupestris). Rev Comp Anim 6: 43–51.
- 66. Villegas-Ríos D, Alós J, March D, Palmer M, Mucientes G, et al. (2013) Home range and diel behaviour of the ballan wrasse, Labrus bergylta, determined by acoustic telemetry. J Sea Res 80: 61–71. doi: 10.1016/j.seares.2013.02.009
- 67. Palumbi SR (1992) Marine speciation on a small planet. Trends Ecol Evol 7: 114–118. doi: 10.1016/0169-5347(92)90144-z
- 68. Planes S, Doherty PJ (1997) Genetic relationships of the colour morphs of Acanthochromis polyacanthus (Pomacentridae) on the northern Great Barrier Reef. Mar Biol 130: 109–117. doi: 10.1007/s002270050230