Journal of Fish Biology (2010) 76, 2370–2381

doi:10.1111/j.1095-8649.2010.02627.x, available online at www.interscience.wiley.com

Differences in δ13C and δ15N stable isotopes in the pearlyrazorfish Xyrichtys novacula related to the sex, location

and spawning period

A. Box*†, S. Deudero*‡, A. Blanco*, A. M. Grau§ and F. Riera§

*Laboratorio de Biología Marina, Universidad de las Islas Baleares, Ctra Valldemossa Km7.5, CP 07122, Balearic Islands, Spain, ‡Instituto Espanol de Oceanografia, Centro

Oceanografico de Baleares, P.O. Box 291, CP 07015, Palma, Balearic Islands, Spain and§Direccio General de Pesca, Govern de les Illes Balears, C/ Foners 10, 07006, Palma,

Balearic Islands, Spain

(Received 17 March 2009, Accepted 10 February 2010)

In the present study, Xyrichtys novacula (Labridae) were sampled at five locations around the islandsof Ibiza and Formentera (western Mediterranean Sea). Isotopic signatures of δ13C, δ15N and theC:N ratio were analysed in relation to locality, sex and size differences. δ13C and δ15N partitioningwas also studied in the reproductive spawning period. There were significant differences in the δ13Csignature between localities for both sexes, but not for δ15N. Sex differences were also found witha mean ± s.e. value of −17·38 ± 0·06‰ δ13C and 8·36 ± 0·05‰ δ15N for females and −17·17 ±0·07‰ δ13C and 8·80 ± 0·06‰ δ15N for males. Increasing total length in both sexes was positivelycorrelated with δ15N enrichment and a significant positive linear regression was established for bothvariables. During the reproductive spawning period, there were changes in δ13C fractioning withenrichment in postspawning females and males (with respect to prespawning and spawning periods)and δ15N impoverishment in postspawning females (with respect to prespawning and spawningperiods). Xyrichtys novacula uses local food sources, as confirmed by δ13C and δ15N, and femalesand males use different food sources, thus avoiding intraspecific competition. This was confirmedby δ15N enrichment as size increased. Spawning leads to special requirements for gonad maturation,which is reflected in the isotopic signatures for both sexes. © 2010 The Authors

Journal compilation © 2010 The Fisheries Society of the British Isles

Key words: isotopic signature; ontogenic changes; reproduction; western Mediterranean.

INTRODUCTION

The pearly razorfish, Xyrichtys novacula (L.), lives in the Mediterranean Sea andAtlantic Ocean (FishBase 2007; www.fishbase.org). It is a sequential hermaphrodite,beginning life as a female and becoming a male as size increases (Candi et al., 2004).Xyrichtys novacula live on sandy bottoms and in Cymodocea nodosa seagrass mead-ows, feeding mainly on benthic invertebrates (Riera & Linde, 2001; Castriota et al.,2005; Beltrano et al., 2006). This fish exhibits territorial behaviour, especially during

†Author to whom correspondence should be addressed. Tel.: +34 971 17 31 38; fax: +34 971 17 31 84;email: boxtoni@yahoo.es

2370© 2010 The Authors

Journal compilation © 2010 The Fisheries Society of the British Isles

S TA B L E I S OT O P E S I N X Y R I C H T Y S N OVA C U L A 2371

the spawning season when males have a harem of four to five females in a territorythat is defended against other males (Katsanevakis, 2005). Sexual dimorphism isapparent between males and females, the males being larger with different coloura-tion patterns and a different head shape (Cardinale et al., 1998). The reproductiveperiod of X. novacula, based on gonad histology, starts in May and extends until lateSeptember with a reproductive peak in mid-July (Candi et al., 2004). Other stud-ies based on macroscopic observations have established a reproductive period fromJuly to September with maximum spawning activity in August (Cardinale et al.,1998). After the reproductive period, some individuals can be found in an initialphase of sexual inversion that seems to last for 2–4 months (Candi et al., 2004).The sex inversion size is directly related to the fishing efforts for this species inthe area (Riera & Linde, 2001), with sex inversion being found in total length(LT) classes from 10·5 to 15·0 cm (Candi et al., 2004; Castriota et al., 2005) or15·0–17·0 cm (Cardinale et al., 1998). The capture of the dominant male in the ter-ritory causes one of the biggest females in the territory to change sex (Riera & Linde,2001).

The diet of this species also varies according to sex and size, with larger prey inlarger size classes (Castriota et al., 2005). Studies on the food habits of X. novaculahave shown that they feed mainly on molluscs, but also on shrimps, crabs, mysids,polychaetes, sipunculans, echinoderms and teleosts (Cardinale et al., 1997; Riera &Linde, 2001; Castriota et al., 2005; Beltrano et al., 2006).The low displacement ratesaccording to its described behaviour and the sex-related feeding habits are reflectedin the local use of food sources, which is expected to be mirrored as a local signalof stable isotopes.

Xyrichtys novacula is one of the principal recreational target species (Morales-Ninet al., 2001, 2005). A decrease in the size of individuals in recent years has beenobserved due to the high recreational angling pressure (Riera & Linde, 2001) andhas made it necessary to establish temporal and spatial regulations for this species(Balearic Islands Government Regulation 69/1999; www.caib.es). In the islands ofIbiza and Formentera, regulations have established a period from 1 April until 31August during which fishing for this species by professional and recreational fishersis totally prohibited.

Stable isotopes of carbon and nitrogen have been widely used to reconstruct trophicwebs in coastal waters (Vizzini & Mazzola, 2002; Deudero et al., 2004) and estuaries(Guelinckx et al., 2008; Leakey et al., 2008). Isotopic signatures have also shownthe potential for use in the study of ontogenic changes in marine fishes (Deuderoet al., 2004) and population connectivity (Menard et al., 2007; Acolas et al., 2008;Fry, 2008; Rooker et al., 2008; Vinagre et al., 2008). Carbon is considered a goodtracer of the food source used by the organism (Pinnegar & Polunin, 2000) andnitrogen values are useful to establish trophic levels in the ecosystem (Schmidtet al., 2003; Pakhomov et al., 2004; Bearhop et al., 2006). Stable isotopes have alsobeen used to study size-related dietary habits, which are reflected in changes in theisotopic signatures (Deudero et al., 2004; Dubois et al., 2007; Parry, 2008). The foodsources and the isotopic fractionation during the feeding process are the origin of thecarbon and nitrogen isotopic ratios that the organism’s tissues assimilate (Pinnegar& Polunin, 2000). Discrimination between organic matter sources is made possibleby the addition of C:N ratios to 13C determinations (Meyers, 1994). The use ofδ13C values as evidence of dietary and trophic differences among species can be

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

2372 A . B OX E T A L .

validated by the examination of the relationship between biomass δ13C and biomassC:N (Dunton, 2001).

The aim of this study was to analyse differences in the δ13C and δ15N signaturesat five locations in the Balearic Islands in order to determine: (1) spatial variationin the isotopic signal of X. novacula, (2) ontogenic changes, (3) differences relatedto sexual dimorphism and behaviour and (4) changes in the isotopic signature inspawning and postspawning periods.

MATERIALS AND METHODS

S A M P L I N G L O C AT I O N S

Five sandy areas where X. novacula are usually captured by recreational anglers wereselected around Ibiza and Formentera with a depth ranging from 15 to 34 m. These areaswere Cala Saona, Freu Petit marine protected area (MPA), Clot des Llamp and Portinatx,all of them being subject to temporal regulations, while Freu Gros MPA also has a banagainst recreational anglers (Fig. 1). This sampling design allowed the isotopic signatures ofX. novacula populations to be studied from north to south in the Pitiusas Archipelago.

Two fishing periods were investigated at the five sampling locations. The first samplingperiod was from May to August in the prespawning to spawning season, during which fishingfor this species is banned. The second period was during the postspawning season fromSeptember to November, during which fishing is allowed. The LT of captured fish wasmeasured to the nearest 1 mm.

1° 30'1° 20'1° 10'

39° 00'

38° 50'

38° 40'

Portinatx

Clot des Llamp

Freu PetitFreu Gros

La Mola

Cala Saona

0 5 10 15 km

N

Fig. 1. Sampling locations in Ibiza and Formentera Islands (western Mediterranean Sea).

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

S TA B L E I S OT O P E S I N X Y R I C H T Y S N OVA C U L A 2373

I S OTO P I C A NA LY S I S

Dorsal white muscle of X. novacula was dissected, cut into small pieces, dried at 60◦ Cfor 24 h and then ground to a fine powder using a mortar and pestle. Water samples (5 l) forparticulate organic matter (POM) determination were collected at each site using a Niskinbottle and filtered through pre-combusted fibreglass filters (Whatman GF/C) at 450◦ C for4 h. Muscle samples and POM were analysed to measure carbon and nitrogen isotope ratios.Homogeneous dried powder (2·0 mg, range ± 0·1 mg) from each sample was introduced intoa cadmium tin cup, and then combusted to measure 15N and 13C stable isotope composi-tion by continuous-flow isotope-ratio mass spectrometry (CF-IRMS) using a Thermo DeltaX-PLUS mass spectrometer (www.thermo.com). One sample of an internal reference materialwas analysed after every eight samples in order to calibrate the system and compensate fordrift over time. Bovine liver standard (BLS; 1577b, U.S. Department of Commerce, NationalInstitute of Standards and Technology; www.nist.gov) was the reference material used forcarbon and nitrogen stable-isotopes analysis. The analytical precision was based on the s.d.of internal standard replicates; these deviations were 0·09 and 0·08‰ for BLS for δ13C andδ15N, respectively.

Stable-isotope abundances were measured by comparing the ratio of the most abundantisotope (13C:12C and 15N:14N) in the samples with the international isotopic standards. Car-bon and nitrogen stable-isotope ratios were expressed in δ notation as parts per thousand(‰) deviations from the standards, according to the following equation: δX = 103 [(Rsample

R−1reference) −1], where X is 13C or 15N and R is the corresponding 13C:12C or 15N:14N ratio.The following formula was applied in order to determine the trophic level (TL) of the organ-

isms: TL = (δ15Nconsumer − δ15NFirst trophic level) 3·4−1 + 1, where 3·4% is the assumed δ15Ntrophic enrichment factor (Minagawa & Wada, 1984; Le Loc’h et al., 2008) andδ15NFirst trophic level is considered that of the lowest isotopic nitrogen value, representing thebaseline of the trophic food web (Le Loc’h et al., 2008). The baseline for TL determinationwas taken as the mean of POM, collected at each studied location (due to the high localvariability). Mean POM value was considered to be the first trophic level (Le Loc’h et al.,2008). The quality of the diet had been evaluated by means of the C:N ratio (Waddington &MacArthur, 2008).

DATA A NA LY S I S

Differences in the capture LT for males and females between sites were analysed by meansof non-parametric procedures (Kruskal–Wallis test). This test was also applied to the size andspawning periods.

Non-parametric procedures (Kruskal–Wallis test) were applied to the isotopic signaturesof X. novacula specimens. The isotopic values for X. novacula were compared for malesand females separately to check for differences between localities. Isotopic values werealso compared by sex excluding transitional individuals with pooled data (Kruskal–Wallis)and considering separately each location (Kolmogorov–Smirnov). Finally, the effects of thespawning period on the isotopic signal for δ13C and δ15N were studied in each sex.

The size-dependent effects on the isotopic signature in X. novacula were studied usinglinear regressions for each location and pooled data. ANCOVA was applied to test whetherthe slopes of the regression lines were significantly different and Spearman correlations wereperformed between X. novacula LT and the isotopic signatures δ13C and δ15N.

RESULTS

There were size differences between localities for both sexes [Kruskal–Wallistest: H (n = 85) = 12·68, P < 0·05 for females, and H (n = 56) = 9·64, P < 0·05for males], with larger X. novacula females and males captured in Clot des Llampcompared to other localities. Similar LT of X. novacula females and males were

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

2374 A . B OX E T A L .

20

19

18

17

16

15

14

13

12

11

10Females Females Females Females FemalesMales Males Males Males Males

Cala Saona Freu Gros Freu Petit PortinatxClot des Llamp

LT (

cm)

Fig. 2. Mean ± s.e. total length (LT) classes at the sampling locations (see Fig. 1) of female and male Xyrichtysnovacula.

found at Cala Saona, Portinatx and Freu Gros, while smaller individuals of bothsexes were found at Freu Petit (Fig. 2).

Isotopic signals differed between localities for δ13C in females [Kruskal–Wallistest, H (n = 85) = 35·72, P < 0·001] and males [Kruskal–Wallis test: H (n =56) = 38·09, P < 0·001], but not for δ15N in females [Kruskal–Wallis test: H

(n = 85) = 6·05, P > 0·05] and males [Kruskal–Wallis test: H (n = 56) = 8·16,P > 0·05] (Table I), which was more homogeneous between localities. The mostenriched δ13C values for males and females were found in the Freu Gros area.Higher δ15N in females was found in Freu Gros, while males in Freu Petit showedhigher δ15N values. Kruskal–Wallis test results showed that there were no differencesbetween localities in terms of the C:N ratio and TL for either sex.

There were local significant differences between sexes in all localities (Kol-mogorov–Smirnov test, Clot des Llamp P < 0·01, Freu Petit P < 0·05 and FreuGros P < 0·01) except for Cala Saona. For pooled data, significant differencesbetween sexes were found also in terms of δ13C, with mean ± s.e. values of−17·38 ± 0·06‰ for females and −17·17 ± 0·07‰ for males [Kruskal–Wallis test:H (n = 141) = 4·79, P < 0·05]. δ15N was significantly different between sexes withmean ± s.e. values of 8·36 ± 0·05‰ for females and 8·80 ± 0·06‰ for males[Kruskal–Wallis test: H (n = 141) = 29·11, P < 0·001]. No differences were foundin the C:N ratio between sexes. For the TL, however, significantly higher valueswere obtained for males [Kruskal–Wallis test: H (n = 141) = 29·11, P < 0·001](Table I).

The isotopic composition of females and males changed according to repro-ductive status (Fig. 3). Postspawning females showed significant δ13C enrichment[Kruskal–Wallis test: H (n = 85) = 22·56, P < 0·001] and a significant decrease inδ15N [Kruskal–Wallis test: H (n = 85) = 11·66, P < 0·01] compared with spawning

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

S TA B L E I S OT O P E S I N X Y R I C H T Y S N OVA C U L A 2375

Table I. Isotopic values for δ13C, δ15N, C:N ratio and trophic level (TL) for Xyrichtysnovacula females and males captured in Ibiza and Formentera. Values are expressed as

mean ± s.e.

Site n δ13C δ15N C:N TL

FemalesCala Saona 21 −17·36 ± 0·12 8·29 ± 0·10 1·98 ± 0·01 1·95 ± 0·03Clot des Llamp 12 −18·06 ± 0·11 8·25 ± 0·11 1·99 ± 0·01 1·94 ± 0·03Freu Gros 13 −16·80 ± 0·11 8·58 ± 0·07 1·98 ± 0·01 2·03 ± 0·02Freu Petit 21 −17·56 ± 0·09 8·45 ± 0·08 1·99 ± 0·01 2·00 ± 0·02Portinatx 18 −17·17 ± 0·08 8·27 ± 0·13 1·99 ± 0·01 1·94 ± 0·02Overall mean value 85 −17·38 ± 0·06 8·36 ± 0·05 1·98 ± 0·01 1·97 ± 0·01

MalesCala Saona 6 −17·12 ± 0·11 8·78 ± 0·17 1·97 ± 0·02 2·10 ± 0·05Clot des Llamp 17 −17·75 ± 0·08 8·75 ± 0·11 1·98 ± 0·01 2·09 ± 0·03Freu Gros 12 −16·74 ± 0·07 8·87 ± 0·17 1·97 ± 0·01 2·12 ± 0·05Freu Petit 7 −16·77 ± 0·18 9·03 ± 0·11 2·00 ± 0·01 2·17 ± 0·03Portinatx 14 −17·07 ± 0·11 8·71 ± 0·12 1·99 ± 0·01 2·08 ± 0·03Overall mean value 56 −17·17 ± 0·07 8·80 ± 0·06 1·98 ± 0·01 2·10 ± 0·02

n, number of individuals used to calculate the mean values.

females. Similar trends were observed in males with a significant δ13C enrichmentin postspawning males [Kruskal–Wallis test: H (n = 56) = 17·70, P < 0·001] com-pared with spawning individuals and a decrease in δ15N in postspawning malescompared with spawning males, although the latter was not significant [Fig. 3(b)].The quality of the diet analysed by means of changes in C:N ratio did not dif-fer among localities both for females [Kruskal–Wallis test: H (n = 85) = 6·05,P > 0·05] and males [Kruskal–Wallis test: H (n = 56) = 8·16, P > 0·05].

There were size-related changes in the δ15N signature, which increased withincreases in the LT of X. novacula. A positive linear regression was establishedbetween LT and δ15N for each locality and for the pooled data. This linear regres-sion was site dependent (ANCOVA, P < 0·001) (Fig. 4). Spearman correlations werenot found for δ13C and size, but for δ15N, a significant positive correlation was foundwith size (n = 141, r = 0·65, P < 0·001).

DISCUSSION

There is little available information on isotopic values in X. novacula. The resultsobtained here, however, are consistent with values reported previously for the Bale-aric Islands (Cardona et al., 2007). No information is available for other localities.In the present study, there were differences in the δ13C and δ15N isotopic signaturesbetween sexes and for δ13C between localities. The main food items consumedby X. novacula are small invertebrates such as shrimps, mysids, molluscs, crabs,sipunculans, echinoderms and teleosts (Cardinale et al., 1997; Riera & Linde, 2001).δ13C has been used to track animal movements between areas with different foodsources (Kurle & Worthy, 2001) and to evaluate the importance of different food

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

2376 A . B OX E T A L .

9·4

9·2

8·8

8·6

8·4

8·2

−19 −18·5 −17·5 −16·5−17 −16−188

9

Spawning

Spawning

Prespawning

Prespawning

Postspawning

Postspawning

(5)

(12)

(26)(39)

(54)

(5)

d 15

N:14

N

d 13C/12C

(a)

Spawning

−16·0

−16·5

−17·0

−17·5

−18·5

−18·0

1·96 1·97 1·98 1·99 2·00 2·01

Spawning

PrespawningPrespawning

Postspawning

Postspawning(39)

(54)

(5)(5)

(12)

(26)

d13C

:12C

C:N

(b)

Fig. 3. (a) Mean ± s.e. isotopic signatures δ13C and δ15N for females ( ) and males ( ) during the differentspawning periods studied. (b) Dietary and trophic differences by δ13C and ratio C:N in females ( ) andmales ( ) in the different spawning seasons. Numbers of individuals are given in parentheses.

sources (Pinnegar & Polunin, 2000). In the present study, low, but significant, δ13Cvariations between localities (1·26‰ females and 1·01‰ males) were found, the mostdifferent δ13C signatures being those obtained on the Ibiza south coast (Freu Gros)and Ibiza north coast (Clot des Llamp). These differences may be due to habitatcharacteristics such as the extent of the sandy area, proximity of other habitats(rocky bottom or seagrass meadows) and depth, and the consequent differences infood resources. Xyrichtys novacula males are considered to be haremic (Cardinaleet al., 1998) with a territory that hosts four to six females (Katsanevakis, 2005). Theprotection strategies against predators consist of burrowing into the sand, alwaysusing the same dive site (Katsanevakis, 2005). As a result, X. novacula shows lowdisplacement rates and small displacement distances, and feeds on local food sources,which is reflected in the isotopic signatures.

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

S TA B L E I S OT O P E S I N X Y R I C H T Y S N OVA C U L A 2377

8 10 12 14 16 18 20 22

(a)

8 10 12 14 16 18 20 22

(b)9·69·8

9·49·29·08·88·68·48·28·07·87·67·47·27·0

9·69·8

9·49·29·08·88·68·48·28·07·87·67·47·27·0

8 10 12 14 16 18 20 22

(c)

8 10 12 14 16 18 20 22

(d)9·69·8

9·49·29·08·88·68·48·28·07·87·67·47·27·0

9·69·8

9·49·29·08·88·68·48·28·07·87·67·47·27·0

9·69·8

9·49·29·08·88·68·48·28·07·87·67·47·27·0

8 10 12 14 16 18 20 22

(e) 9·69·8

9·49·29·08·88·68·48·28·07·87·67·47·27·0

8 10 12 14 16 18 20 22

(f)

d15N

:14N

LT (cm)

Fig. 4. The relationship between total length (LT) (including all females and males) and δ15N at (a) Cala Saona(y = 6·618 + 0·123x; r2 = 0·505, P < 0·001), (b) Clot des Llamp (y = 6·367 + 0·131x; r2 = 0·543,P < 0·001), (c) Freu Gros (y = 7·059 + 0·117x; r2 = 0·655, P < 0·001), (d) Freu Petit (y = 7·151 +0·113x; r2 = 0·507, P < 0·001), (e) Portinatx (y = 6·026 + 0·164x; r2 = 0·666, P < 0·001) and (f) allsites combined (y = 6·937 + 0·112x; r2 = 0·439, P < 0·001).

The δ15N value is more representative of the assimilation of food sources due toits relation to metabolic and physiological processes (Sweeting et al., 2007). δ15N isalso related to sewage and anthropogenic activity the marine environment (Leakeyet al., 2008). The fact that there were no differences between sites implies a similarfractionation for δ15N of the food sources and the absence of important human

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

2378 A . B OX E T A L .

nitrogen sources. Freus Gros and Freu Petit are located inside MPAs that lack sewageor human nitrogen input, and the other studied locations are relatively far away fromimportant human settlements.

The analysis of the stomach contents reported in previous studies revealed a certaindiet change related to increasing size and sex conversion, which is reflected in thechange in the predator trophic level from secondary to secondary–tertiary consumerat 18·0 cm LT (Cardinale et al., 1997). The results obtained in the present studyshow a size range from 9·0 to 16·6 cm for females and 13·7 to 20·7 cm for males,which represents the full range for adults, excluding small females that were notcaptured. Females, which are smaller than males, feed on smaller prey (Castriotaet al., 2005; Beltrano et al., 2006) and present a lower trophic level than males.The present results on stable isotopes are consistent with previous studies (Cardinaleet al., 1997; Castriota et al., 2005; Beltrano et al., 2006), demonstrating a progressiveδ15N enrichment and increase in trophic level as size increases, along with theassociated sex inversion, which leads to feeding among higher trophic levels.

There are several studies in fishes and invertebrates involving different environ-ments which show that larger individuals exist at a higher trophic level due toontogenic changes in the diet and thus higher δ15N (Schmidt et al., 2003; Pakhomovet al., 2004; Bearhop et al., 2006). For X. novacula, size dimorphism could be afeasible cause of the sexual differences. Males, which are larger overall and havelarger jaws than females, feed on the largest prey, reducing intraspecific competi-tion with females and juveniles (Castriota et al., 2005). This sexual dimorphism isreflected in the isotopic signatures, with a higher trophic level and δ15N for malesthan females. Small sex-related differences were also observed in the δ13C, reflectingsmall changes in the energy sources used by sexes and its prey species.

Changes in reproductive status involve changes in food sources both for malesand females, as indicated by δ13C enrichment in non-spawning individuals. Severalfactors could explain these changes in δ13C, such as the low displacement of femalesin the reproductive season when they live in a male harem (Cardinale et al., 1998;Riera & Linde, 2001; Candi et al., 2004). Another plausible explanation is the energyexpended on gonad maturation (Kjesbu et al., 1991), which leads to δ13C impov-erishment in prespawning individuals (Sherwood et al., 2007). A decrease in δ15Nwas observed from prespawning to spawning and from spawning to postspawning.These changes in δ15N could be related to the life cycle of the fish, which showshigh feeding activity during prespawning to spawning periods (summer), decreasingprogressively until postspawning (winter), when, according to fishing activity reports(Riera & Linde, 2001), feeding activity is very low. Another possible explanationcould be changes in the nitrate concentrations, which could affect the δ15N signalof aquatic organisms (Montoya et al., 1990). The C:N ratio, however, did not differbetween sexes and reproductive specimens, implying a similar nutritional status forboth sexes in the different reproductive periods.

Although no differences in TL were found between localities, those sites located ina MPA showed higher TL compared with open angling areas, as previously reported(Pinnegar & Polunin, 2000). The use of similar POM values as the baseline for alllocalities implies that these results are based only on the local nitrogen signature,and that the TL results are similar to the δ15N, which presents sex-related differencesin this species. On the other hand, the C:N ratio in X. novacula did not differbetween locations due to the high local variability of this value. In the present study,

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

S TA B L E I S OT O P E S I N X Y R I C H T Y S N OVA C U L A 2379

the C:N ratio of X. novacula did not differ between locations, sex and spawningperiods. Moreover, the C:N ratio indicates diet quality and explains the degree offractionation among the individual compounds of C:N within diets (Waddington &MacArthur, 2008). Thus, similar values of C:N ratios imply that X. novacula had asimilar dietary quality at all locations.

In conclusion, X. novacula uses local food sources as a result of its territorialbehaviour (Katsanevakis, 2005). Sexual partitioning of the food source is observedprobably due to the increase in size and behaviour; as males are more aggressiveand attain larger size than females, they feed on larger and more δ15N-enriched prey.Stable isotopes can be used to trace ontogenic and sexual changes in the diet ofX. novacula and to evaluate differences between locations. An interesting findingof the present study is the effect of the reproductive period on isotopic signatures.This period involves special requirements for gonad maturation, as reflected by δ15Nenrichment, which may be due to higher feeding activity, and also by δ13C impover-ishment. These results represent a starting point for further studies related to gonadmaturation and isotopic signatures.

This work was supported by the research ‘Avaluacio I seguiment dels recursos marinsde la CAIB, 2008’ of the Direccio General de Pesca (Balearic Islands). We appreciate thecollaboration of the Mar-I-Pi II and A. Box Lorenzo, and J. L. Lakarra for his help during fieldsampling and the collaboration for the isotopic analysis to B. Martorell (Serveis CientificoTecnics, Balearic Islands University) and M. Ribas (IUNICS).

References

Acolas, M. L., Roussel, J. M. & Bagliniere, J. L. (2008). Linking migratory patterns and dietto reproductive traits in female brown trout (Salmo trutta L.) by means of stable isotopeanalysis on ova. Ecology of Freshwater Fish 17, 382–393.

Bearhop, S., Phillips, R. A., McGill, R., Cherel, Y., Dawson, D. A. & Croxall, J. P. (2006).Stable isotopes indicate sex-specific and long-term individual foraging specialisationin diving seabirds. Marine Ecology Progress Series 311, 157–164.

Beltrano, A. M., Cannizzaro, L., Vitale, S. & Milazzo, A. (2006). Preliminary study on thefeeding habits of cleaver wrasse Xyrichtys novacula (Pisces: Labridae) in the Strait ofSicily (Mediterranean Sea). Electronic Journal of Ichthyology 2, 50–54.

Candi, G., Castriota, L., Andaloro, F., Finoia, M. G. & Marino, G. (2004). Reproductivecycle and sex inversion in razor fish, a protogynous labrid in the southern MediterraneanSea. Journal of Fish Biology 64, 1498–1513.

Cardinale, M., Colloca, F. & Ardizzone, G. D. (1997). Feeding ecology of Mediterraneanrazorfish Xyrichtys novacula in the Tyrrhenian Sea (Central Mediterranean Sea). Jour-nal of Applied Ichthyology-Zeitschrift fur angewandte Ichthyologie 13, 105–111.

Cardinale, M., Colloca, F. & Ardizzone, G. D. (1998). Growth and reproduction of Xyrichtysnovacula (Pisces : Labridae) in the Mediterranean Sea. Scientia Marina 62, 193–201.

Cardona, L., Revelles, M., Sales, M., Aguilar, A. & Borrell, A. (2007). Meadows of theseagrass Posidonia oceanica are a significant source of organic matter for adjoiningecosystems. Marine Ecology Progress Series 335, 123–131.

Castriota, L., Scarabello, M. P., Finoia, M. G., Sinopoli, M. & Andaloro, F. (2005). Foodand feeding habits of pearly razorfish, Xyrichtys novacula (Linnaeus, 1758), in thesouthern Tyrrhenian Sea: variation by sex and size. Environmental Biology of Fishes72, 123–133.

Deudero, S., Pinnegar, J. K., Polunin, N. V. C., Morey, G. & Morales-Nin, B. (2004). Spatialvariation and ontogenic shifts in the isotopic composition of Mediterranean littoralfishes. Marine Biology 145, 971–981.

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

2380 A . B OX E T A L .

Dubois, S., Jean-Louis, B., Bedrtrand, B. & Lefebvre, S. (2007). Isotope trophic-step fractionof suspension-feeding species: implications for food partitioning in coastal ecosystems.Journal of Experimental Marine Biology and Ecology 351, 121–128.

Dunton, K. H. (2001). δ15N and δ13C measurements of antarctic peninsula fauna: trophicrelationships and assimilation of benthic seaweeds. American Zoologist 41, 99–112.

Fry, B. (2008). Open bays as nurseries for Louisiana brown shrimp. Estuaries and Coasts 31,776–789.

Guelinckx, J., Maes, J., Geysen, B. & Ollevier, F. (2008). Estuarine recruitment of a marinegoby reconstructed with an isotopic clock. Oecologia 157, 41–52.

Katsanevakis, S. (2005). Habitat use by the pearly razorfish, Xyrichtys novacula (Pisces:Labridae). Scientia Marina 69, 223–229.

Kjesbu, O. S., Klungsoyr, J., Kryvi, H., Witthames, P. R. & Walker, M. G. (1991). Fecun-dity, atresia, and egg size of captive atlantic cod (Gadus morhua) in relation toproximate body-composition. Canadian Journal of Fisheries and Aquatic Sciences 48,2333–2343.

Kurle, C. M. & Worthy, G. A. J. (2001). Stable isotope assessment of temporal and geo-graphic differences in feeding ecology of northern fur seals (Callorhinus ursinus) andtheir prey. Oecologia 126, 254–265.

Le Loc’h, F., Hily, C. & Grall, J. (2008). Benthic community and food web structure onthe continental shelf of the Bay of Biscay (North Eastern Atlantic) revealed by stableisotopes analysis. Journal of Marine Systems 72, 17–34.

Leakey, C. D. B., Attrill, M. J., Jennings, S. & Fitzsimons, M. F. (2008). Stable isotopes injuvenile marine fishes and their invertebrate prey from the Thames Estuary, UK, andadjacent coastal regions. Estuarine Coastal and Shelf Science 77, 513–522.

Menard, F., Lorrain, A., Potier, M. & Marsac, F. (2007). Isotopic evidence of distinct feed-ing ecologies and movement patterns in two migratory predators (yellowfin tuna andswordfish) of the western Indian Ocean. Marine Biology 153, 141–152.

Meyers, P. A. (1994). Preservation of elemental and isotopic source identification of sedi-mentary organic matter. Chemical Geology 114, 289–302.

Minagawa, M. & Wada, E. (1984). Stepwise enrichment of 15N along food chains: furtherevidence and the relation between δ15N and animal age. Geochimica et CosmochimicaActa 48, 1135–1140.

Montoya, J. P., Horrigan, S. G. & McCarthy, J. J. (1990). Natural abundance of 15N in par-ticulate nitrogen and zooplankton in the Chesapeake Bay. Marine Ecology ProgressSeries 65, 35–61.

Morales-Nin, B., Moranta, J., Garcia, C., Tugores, M. P., Grau, A. M., Riera, F. & Cerda, M.(2005). The recreational fishery off Majorca Island (western Mediterranean): someimplications for coastal resource management. ICES Journal of Marine Science 62,727–739.

Pakhomov, E. A., McClelland, J. W., Bernard, K., Kaehler, S. & Montoya, J. P. (2004). Spa-tial and temporal shifts in stable isotope values of the bottom-dwelling shrimp Nauti-caris marionis at the sub-Antarctic archipelago. Marine Biology 144, 317–325.

Parry, M. (2008). Trophic variation with length in two ommastrephid squids, Ommastrephesbartramii and Sthenoteuthis oualaniensis. Marine Biology 153, 249–256.

Pinnegar, J. K. & Polunin, N. V. C. (2000). Contributions of stable-isotope data to elucidatingfood webs of Mediterranean rocky littoral fishes. Oecologia 122, 399–409.

Riera, F. & Linde, M. (2001). El Raor i la Cirviola. Coneixer per preservar. Palma de Mal-lorca: Govern de les Illes Balears.

Rooker, J. R., Secor, D. H., DeMetrio, G., Kaufman, A. J., Rios, A. B. & Ticina, V. (2008).Evidence of trans-Atlantic movement and natal homing of bluefin tuna from stableisotopes in otoliths. Marine Ecology Progress Series 368, 231–239.

Schmidt, K., Atkinson, A., Stubing, D., McClelland, J. W., Montoya, J. P. & Voss, M.(2003). Trophic relationships among Southern Ocean copepods and krill: some uses andlimitations of a stable isotope approach. Limnology and Oceanography 48, 277–289.

Sherwood, G. D., Rideout, R. M., Fudge, S. B. & Rose, G. A. (2007). Influence of diet ongrowth, condition and reproductive capacity in Newfoundland and Labrador cod (Gadusmorhua): insights from stable carbon isotopes (delta C-13). Deep-Sea Research I 54,2794–2809.

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381

S TA B L E I S OT O P E S I N X Y R I C H T Y S N OVA C U L A 2381

Sweeting, C. J., Barry, J., Barnes, C., Polunin, N. V. C. & Jennings, S. (2007). Effects ofbody size and environment on diet-tissue delta N-15 fractionation in fishes. Journal ofExperimental Marine Biology and Ecology 340, 1–10.

Vinagre, C., Salgado, J., Costa, M. J. & Cabral, H. N. (2008). Nursery fidelity, food webinteractions and primary sources of nutrition of the juveniles of Solea solea andS. senegalensis in the Tagus Estuary (Portugal): a stable isotope approach. EstuarineCoastal and Shelf Science 76, 255–264.

Vizzini, S. & Mazzola, A. (2002). Stable carbon and nitrogen ratios in the sand smelt froma Mediterranean coastal area: feeding habits and effect of season and size. Journal ofFish Biology 60, 1498–1510.

Waddington, K. & MacArthur, L. (2008). Diet quality and muscle tissue location influ-ence consumer-diet discrimination in captive-reared rock lobsters (Panulirus cygnus).Marine Biology 154, 569–576.

Electronic Reference

Morales-Nin, B., Moranta, J., Garcia, C., Grau, A. M., Riera, F., Bosch, T., Martino, S.,Cerda, M., Cardona, L., Lopez, D., Sales, M., de Caralt, S. & Díez, I. (2001). Segui-miento de la pesca recreativa en las Islas Baleares. Determinacion del esfuerzo y delas Capturas. Les Illes Balears: Govern de les Illes Balears, Conselleria d’Agriculturai Pesca e Instrument Financer Orientacio en Pesca. Available at http://www.caib.es/govern/archivo.do?id=37981

© 2010 The AuthorsJournal compilation © 2010 The Fisheries Society of the British Isles, Journal of Fish Biology 2010, 76, 2370–2381