RESEARCH ARTICLE

An observational study on changes in biometryand generation time of Odontophora villoti (Nematoda,Axonolaimidae) related to petroleum pollutionin Bizerte bay, Tunisia

Fehmi Boufahja & Amor Hedfi & Naceur Essid &

Patricia Aïssa & Ezzeddine Mahmoudi &Hamouda Beyrem

Received: 6 June 2011 /Accepted: 29 August 2011 /Published online: 10 September 2011# Springer-Verlag 2011

AbstractIntroduction We conducted a yearly polluted-referencesampling to assess the effects of petroleum pollution on lifecycle characteristics of the meiobenthic nematode Odonto-phora villoti. Samples were taken every 15 days between 26November 2004 and 25 November 2005 from two beachesof Bizerte bay (Tunisia), Rimel and Tunisian RefiningIndustries Company (TRIC). The latter site is located infront of the “Tunisian Refining Industries Company” runoff.Discussion When compared to the reference site, the meanbody dimensions of O. villoti from the impacted site weresignificantly lower. The small size of affected nematodeswas represented both by the length and width as a functionof the life stage. It was also established that changes inlengths of body parts during molts were different betweenthe two study sites. The low availability of oxygen fromApril to August seems to prevent the formation of embryosof O. villoti. Thus, two annual reproductive cycles withdifferent durations were observed in Rimel and TRIC.Under stress, juvenile phase and egg production weregenerally shorter. Globally, the impact of petroleumpollution on O. villoti was expressed by a short egg-to-eggdevelopment time.Conclusion Our study assessed the usefulness of life cyclecharacteristics (biometry and life stage durations) of O.

villoti in biomonitoring, and the results are generallyconsistent suggesting that this species may be consideredas an efficient bioindicator.

Keywords Bizerte bay . Biometry . Life cycle .

Odontophora villoti . Petroleum pollution

1 Introduction

The meiofauna is defined as those benthic metazoans thatpass through a 1-mm sieve mesh size but are retained by40 μm (Wieser 1960; Mahmoudi et al. 2005). Nematodestypically are by far the most abundant and diverseconstituents of the meiobenthic community and comprise60–90% of the total numbers of this size class of benthicorganisms. Up to 130 species can be identified at anindividual sampling site (Vanhove et al. 1999). As a resultof their high abundance and diversity and their widespreaddistribution, nematodes show a spectrum from high to lowtolerances to various kinds of pollution and disturbances.Having short reproductive cycles (days/weeks) and fastmetabolic rates, free-living marine nematodes play animportant role in ecosystem functioning and may beeffective indicators of marine environmental health onvarious scales (Vanhove et al. 2004).

Research into anthropogenic influences on meiobenthicnematodes has focused on the effects of organic enrichment,hydrocarbon spills, heavy metals, pesticides, lubricants, etc.using classic tools concerning quantitative (numbers, size),feeding, and taxonomic data on nematode communities andspecies (Steyaert et al. 2003; Gheskiere et al. 2005).However, the relationships between life cycles of nematodes

Responsible editor: Philippe Garrigues

F. Boufahja (*) :A. Hedfi :N. Essid : P. Aïssa : E. Mahmoudi :H. BeyremCoastal Ecology and Ecotoxicology Unit,Laboratory of Environment Biomonitoring,Faculty of Sciences of Bizerte, University of Carthage,7021, Zarzouna, Tunisiae-mail: fehmiboufahja@yahoo.fr

Environ Sci Pollut Res (2012) 19:646–655DOI 10.1007/s11356-011-0609-y

and pollution are still largely unknown. Moreover, life cyclesin the field of only a few free-living marine species aredocumented of a known number of about 4,000 species(Heip et al. 1982) which have been identified.

In this study, the effects of petroleum compounds fromthe “Tunisian Refining Industries Company (TRIC)” runoffon life cycle characteristics of the Axonolaimid nematodespecies Odontophora villoti Luc and De Coninck (1959)were examined. The male of this species, as described bythese authors, is a typical slender nematode of about 4 mmlength and a maximum width of 0.06 mm, with a smoothcuticle over almost the total body length, a small roundedloop amphid and a conical tail. The holotype specimen ischaracterized by the following De Man’s ratios: a (totallength/maximum width)=63, b (total length/pharyngeallength)=15.5, and c (total length/tail length)=19.7. Thebuccal cavity is conical with posterior walls stronglycuticularized and containing six well-developed odontia.Due to this buccal structure and feeding habits known fromAxonolaimids, O. villoti is thought to be a non-selectivedeposit feeder, as described by Wieser (1953). The speciesis known from the Mediterranean Sea and the NorthAtlantic Ocean. The present study is designed (1) todistinguish larval stages of O. villoti from polluted andnon-polluted sampling sites and (2) to investigate biometryand durations of life cycles from both sites.

2 Materials and methods

2.1 Study area

The study area, the Bizerte bay, is located along thenortheast coast of Tunisia (Fig. 1). Two sets of data were

obtained from samples collected every 2 weeks between26 November 2004 and 25 November 2005 at two sandytidal beaches, both exposed to waves and tides. The siteswere located at 3.7 km distance from each other, Rimel(37°15.18′ N, 9°55.34′ E), being practically not affectedby oil-polluted seawater, and TRIC (37°16.07′ N, 9°53.58′ E), located at about 40 m distance from the oilcompany’s runoff, exposed to oil-polluted tidal watersince 1963.

2.2 Abiotic environment

Dissolved oxygen concentrations in the overlying waterwere measured in situ at the water/sediment interface usingan Oxymeter (WTW OXI 330/SET, WTW, Weilheim,Germany). Salinity and temperature were measured with athermosalinity meter (WTW LF 196, Weilheim, Germany).

Sediment samples were taken for coarse fraction(>63 μm) evaluation by means of wet sieving on a 63-μmmesh. Hydrocarbon analyses were carried out by means ofinfrared spectrophotometry according to Danovaro et al.(1995). Before determination of organic contents, sedimentswere treated with an excess of 10% HCl to removecarbonates. Organic matter contents were established afterdesiccation at 450°C for at least 6 h until weight hadstabilized and were expressed as percentage dry weight(Fabiano and Danovaro 1994). All analyses were carriedout in triplicate.

2.3 Meiofauna processing in the laboratory

To investigate the meiofauna, three 20-cm sedimentsamples were taken using cylindrical Plexiglas coringtubes (diameter 3.6 cm). Samples were gently rinsed

Mediterranean Sea

TRIC

9°55’

37°15’

TUNISIA

Rimel

Tunisian Refining Industries Company

0 0,5 1

Km

0 0,5 1

Km

Fig. 1 Sampling area andsite locations

Environ Sci Pollut Res (2012) 19:646–655 647

with freshwater to extract meiofauna from the sedimentby means of the resuspension–decantation technique(Wieser 1960), using a 1-mm sieve for the upper and a40-μm sieve for the lower size limit. Samples were fixedin 4% formalin and stained in Rose Bengal (0.2 g/L).While sorting under the stereomicroscope, 100 nematodesper sample were picked out, transferred to glycerolthrough a series of ethanol–glycerine solutions, andmounted in glycerine slides (Somerfield et al. 1994).Finally, nematodes were identified to species level usingdescriptions downloaded from the web site http://nemys.ugent.be/ developed by nematologists of Ghent University(Belgium). All nematodes were subsequently mounted onslides for microscopic research of O. villoti at magnificationsbetween ×100 and ×1,000.

2.4 Biometry and life cycle of O. villoti

Nematode life cycles are simple: Undifferentiated cells inthe oval egg develop into first-stage slender juvenile worms(J1) which molt (shed cuticle) four times, into J2, J3, J4,and adult male and female successively. Females produceeggs generally after copulation/fertilization, and the cycle isrepeated. In our study, we consider generation time to be theegg-to-egg development period during one reproductivecycle, thus the time period from gravid mother to graviddaughter (Vranken et al. 1981).

As body size is an important factor in determining thecourse of life of aquatic invertebrates and thus in establish-ing the impact of pollutants, body dimensions of alljuvenile and adult O. villoti during one sampling year weredetermined. Unfortunately no qualitative criteria to differ-entiate the four larval stages exist. Seven body dimensions(in micrometers) were measured, and ten biometric ratioswere calculated:

TL Total lengthPL Pharyngeal length (distance between

the base of the stoma andthe cardia)

IPL Intermediate piece length (distancebetween the cardia and proximallimit of the tail)

CL Caudal lengthMD Maximum width (measured

on the middle region of the body(juveniles and males) or in front ofthe vulvae (females))

HW Head width (at the level of thecephalic setae)

AW Anal widthDe Man’s index a Total length divided by maximum

width

De Man’s index b Total length divided bypharyngeal length

De Man’s index c Total length divided by caudallength

Index d Total length divided byintermediate piece length

CL/AW Caudal length divided by analwidth

PL/CL Pharyngeal length divided bycaudal width

IPL/PL Intermediate piece length dividedby pharyngeal length

IPL/CL Intermediate piece length dividedby caudal length

AW/MW Anal width divided by maximumwidth

HW/MW Head width divided by maximumwidth

In the current study, we used five biometric descriptorspreviously proposed by Boufahja et al. (2011) for meio-benthic nematodes (IPL, d, IPL/PL, IPL/CL and AW/MW)and assessed one new ratio (HW/MW).

2.5 Egg volume

Egg volume can be quantified using the followingequation: V=(4/3) π abc, where V is the volume (in cubicmicrometers, 10−6 nl) and a, b, and c are the lengths of theellipsoid radii (in micrometers). Since only two-dimensionalobservations are possible with light microscopy, V wasapproximated as (4/3) π (x/2)2(y/2), where x and y are theminor and major axes of the egg.

2.6 Statistical analysis

At each site, all biometric data (body dimensions andbiometric ratios) of juveniles were included in the k-meansanalysis in order to generate four categories of statisticallyhomogenous data; each corresponds to one juvenile stage(Boufahja et al. 2011). Differences in body dimensionsbetween sites were tested using the Z test (n≥30). Prior toanalysis, variables were assessed for normality with theKolmogorov–Smirnov test and homogeneity of varianceusing the Bartlett test. This assumption was met followingtransformation using natural logarithms. Linear allometricregressions were performed on natural logarithm-transformed lengths of the whole body and that of theintermediate piece. This dimension was also used for thescaling relative growth of the ovaries, eggs and the uterus.Differences in regression slopes were tested with analysisof covariance (ANCOVA) according to White (2003) todetect differences in relative rates of growth of the

648 Environ Sci Pollut Res (2012) 19:646–655

intermediate piece. An α of 0.05 was used to assesssignificant differences. All of the above analyses wereperformed using xlstat 1.6 software.

3 Results

3.1 Abiotic environment

Variation in water depths was similar at both study sites.This parameter varied from 0.2 to 1.54 m at Rimel and from0.33 to 1.47 m at TRIC. Temperature showed a cleartemporal development at both sites, Rimel (14.1–31.1°C)and TRIC (14.2–32.7°C), with lower values in October–March (wet seasons) than in April–September (dry seasons).Salinity varied slightly with time at both study sites, Rimel(37.1–40.1 psu) and TRIC (36.9–40.3 psu), with the lowestvalues in winter. Dissolved oxygen concentration variedstrongly over time at both sites, Rimel (3.9–12.8 mg/L) andTRIC (3.91–13.22 mg/L). The highest oxygen levelsmeasured were observed in winter.

The sediments consisted largely of fine sand throughoutthe year of sampling. The coarse fraction (>63 μm) loads atRimel (81 samples, 99.65±0.24%) and TRIC (81samples, 99.63±0.19%) were statistically similar (Ztest: p=0.757). Conversely, the average hydrocarbon loadat TRIC (42 samples, 2.781±0.446 mg/g) was muchhigher (Z test: p<0.0001) than at Rimel (42 samples,0.076±0.008 mg/g). At TRIC, the temporal variation inhydrocarbon and organic matter contents was relativelysmall (CV≈16 and 18%, respectively), and correlationbetween both parameters was significant (log-transformeddata: df=12, r=0.53, p=0.047).

3.2 Species choice

Based on unpublished data, Oncholaimus campylocercoidesand O. villoti were the longest species (3–4 mm) presentsimultaneously at both study sites. The two species arepresent in sufficient numbers in both polluted as well asunpolluted sediment, thus supporting the statistical rigor ofthe analyses. Only data reflecting the impact of petroleumpollution on the life cycle of O. villoti are presented in thecurrent paper.

3.3 Biometry of O. villoti

3.3.1 Body dimensions

Dimensions of O. villoti from non-polluted Rimel weresignificantly larger than those of O. villoti from pollutedTRIC except for pharyngeal length and anal width ofjuveniles J1 and head width of juveniles J4 (Z test, Table 1).

3.3.2 Differential effects of pollution on body dimensions

Ratios relating mean body dimensions of the sevendemographic classes (juveniles J1–4, males, females, andgravid females) from the two study sites (Table 1) wereused to compare the magnitudes of negative effects ofpollution on these size descriptors:

– Total length vs. maximum width: The significance ofdifferences in size between specimens from thepolluted and from the non-polluted site was greaterfor maximum width than for total length with respect tojuveniles J1 and J4 and to adults. For juveniles J3, thereverse was observed. The reduction in body size ofjuveniles J2 from TRIC was due to both reduced totallength and maximum width.

– Pharyngeal length vs. caudal length and intermedi-ate piece length: For juveniles J1–3 and males,intermediate piece length was affected more nega-tively as compared to pharyngeal and caudallengths. For these demographic classes, the siteeffects on pharyngeal and caudal lengths weresimilar. The highest ratios between study sitescharacterized lengths of the tail and the pharynxfor juveniles J4 and those of the intermediate pieceand the pharynx for females.

– Maximum width vs. head width and anal width: Thepossible impact of pollution was particularly apparentfor maximum body width and for head width ofjuveniles J1–2 and for maximum body width and analwidth of juveniles J3–4 and gravid females. Withrespect to males and non-gravid females, differencesbetween anal widths and head widths of specimens O.villoti from the polluted site and from the non-pollutedsite were more pronounced as compared to maximumbody width.

3.3.3 Proportionality between body part dimensionsduring molts and egg production

An increase of the ratios IPL/PL and IPL/CL of juvenilesJ2–3 from Rimel compared with those of J1 was progres-sively noted (Table 2). After the third molt (J3 to J4), theseratios showed a small decrease. In the time period betweenthe three molts, PL/CL remained more or less constant. Themolt into males initiated an increase of the values forIPL/PL and IPL/CL whereas PL/CL decreased. The lastmolt into females and their consequent fertilizationresulted in an increase of the values for the IPL/PL,IPL/CL, and PL/CL ratios.

At TRIC, an increase of the ratios IPL/PL and IPL/CLwas observed throughout the complete development fromJ1 to males and gravid females. The ratio PL/CL increased

Environ Sci Pollut Res (2012) 19:646–655 649

Tab

le1

Bod

ydimension

sof

juveniles(J1,

J2,J3,andJ4)andadultsof

O.villo

tifrom

Rim

elandTRIC

(Bizerte

bay,

Tun

isia)

Dem

ographic

groups

J1J2

J3J4

Males

Non-gravidfemales

Gravidfemales

Sites

Rim

elTRIC

Rim

elTRIC

Rim

elTRIC

Rim

elTRIC

Rim

elTRIC

Rim

elTRIC

Rim

elTRIC

Num

berof

individuals

221

161

206

96164

81228

57245

95201

96203

110

TL(μm)

Mean

912.52

(a)

793.53

(b)

1,481.90

(a)

1,205.35

(b)

2,142.16

(a)

1,484.19

(b)

2,449.60

(a)

2,287.55

(b)

3,511.14

(a)

2,991.19

(b)

3,640.22

(a)

3,175.52

(b)

3,736.84

(a)

3,299.29

(b)

SD

196.66

73.88

158.04

57.86

127.77

169.90

52.46

389.49

508.79

277.10

344.59

252.08

386.85

457.97

Ratio

ofmeans

1.14

1.22

1.44

1.07

1.17

1.14

1.13

PL(μm)

Mean

96.69(a)

97.34(a)

135.07

(a)

128.12

(b)

194.77

(a)

150.23

(b)

236.51

(a)

178.40

(b)

238.99

(a)

226.33

(b)

277.13

(a)

235.23

(b)

275.71

(a)

238.42

(b)

SD

13.51

7.81

9.74

8.90

12.72

5.99

15.77

11.07

51.03

6.02

63.06

19.41

59.36

33.74

Ratio

ofmeans

0.99

1.05

1.29

1.32

1.05

1.17

1.15

CL(μm)

Mean

69.81(a)

67.16(a)

99.17(a)

86.86(b)

140.57

(a)

107.74

(b)

172.99

(a)

125.05

(b)

191.89

(a)

178.20

(b)

198.77

(a)

179.18

(b)

193.36

(a)

181.67

(b)

SD

10.41

5.05

7.03

6.49

8.51

4.37

14.30

9.59

21.27

6.31

24.03

24.70

27.76

24.60

Ratio

ofmeans

1.03

1.14

1.30

1.38

1.07

1.10

1.06

IPL(μm)

Mean

735.25

(a)

622.03

(b)

1,229.99

(a)

961.74

(b)

1,781.76

(a)

1,209.11

(b)

2,010.68

(a)

1,751.27

(b)

3,041.98

(a)

2,554.64

(b)

3,124.96

(a)

2,726.78

(b)

3,226.36

(a)

2,843.22

(b)

SD

170.60

65.78

140.15

43.83

106.32

168.70

22.85

33.60

436.65

261.98

259.41

208.11

294.44

395.62

Ratio

ofmeans

1.18

1.27

1.47

1.14

1.19

1.14

1.13

MW

(μm)

Mean

27.44(a)

21.94(b)

34.23(a)

28.31(b)

36.60(a)

30.74(b)

39.80(a)

34.45(b)

55.45(a)

43.87(b)

64.22(a)

53.57(b)

71.43(a)

59.03(b)

SD

4.55

1.12

0.68

0.98

0.61

0.52

1.30

1.75

6.52

2.81

6.53

5.73

4.52

7.75

Ratio

ofmeans

1.25

1.20

1.19

1.15

1.26

1.19

1.21

HW

(μm)

Mean

12.23(a)

9.76

(b)

14.77(a)

13.44(a)

15.16(a)

15.08(a)

16.03(a)

17.69(a)

24.68(a)

17.26(b)

28.71(a)

23.61(b)

29.58(a)

26.24(b)

SD

2.01

0.64

0.15

0.53

0.09

0.53

0.40

0.77

2.88

1.05

3.06

2.50

2.05

3.52

Ratio

ofmeans

1.25

1.09

1.00

0.90

1.42

1.21

1.12

AW

(μm)

Mean

15.84(a)

17.65(a)

23.84(a)

23.42(a)

29.56(a)

24.73(b)

34.52(a)

27.74(b)

44.72(a)

26.88(b)

60.08(a)

43.14(b)

64.47(a)

43.38(b)

SD

3.58

0.90

1.53

0.82

1.71

0.41

1.60

1.76

5.22

3.25

3.98

3.40

4.51

5.82

Ratio

ofmeans

0.89

1.01

1.19

1.24

1.66

1.39

1.48

Letters(a)and(b)indicatesign

ificantdifference

inthemeanvalues

ofthebo

dydimension

s(ln-transformed

data,Z

test)betweenRim

elandTRIC

(p<0.00

01).Valuesin

italic

referto

thehigh

est

meanvalues

ofthebo

dydimension

s

TLtotalleng

th,PLph

aryn

geal

leng

th,CLcaud

alleng

th,IPLinterm

ediate

pieceleng

th,MW

maxim

umwidth,HW

head

width,AW

anal

width

650 Environ Sci Pollut Res (2012) 19:646–655

throughout the life cycle but with a decline after the second(J2–3) and fourth (J4–adults) molt.

3.4 Life cycle of O. villoti

Figure 2 shows the abundance of juvenile and adult O. villoticollected over the sampling year 2004–2005. Abundance ofJ1 and of gravid females is presented by curves with two andthree peaks respectively. At Rimel, O. villoti was totallyabsent from the end of June till the end of August and fromearly April till the end of July at TRIC. These observationswere used for the construction of a reproduction model. Toimprove model fits, two sets of data were taken into account:(1) the variation in proportion of every juvenile stage and (2)linking abundance of gravid females and J1.

3.4.1 Population dynamics at Rimel

The following are the annual emergence of two reproductivecycles:

First reproductive cycle: During the winter period,gravid females of late December 2004 generated

juveniles J1 of January 2005 and consequently thestages J2–4 appeared (J4: 18 March 2005–1 April2005), becoming adults in April 2005. Males appeared15 days before females, and these became gravid in theperiod from 13 May to 10 June 2005.Second reproductive cycle: The first generation juve-niles appeared on 2 September 2005 and developedinto J4 in October 2005. Males and femalesappeared on 14 and 28 October 2005, respectively,and continued to appear during the following weeks.Fertilization of females is suggested to have startedat about the last sampling date (i.e., 25 November2005).

3.4.2 Population dynamics at TRIC

The following are the annual emergence of two reproductivecycles:

First reproductive cycle: Peaks of maximum abundanceof gravid females were noticed during the first twosamplings (26 November 2004 and 10 December2004) and preceded the increase in numbers of small

Table 2 Mean values of biometric ratios of juveniles (J1, J2, J3, and J4) and adults of O. villoti from Rimel and TRIC (Bizerte bay, Tunisia)

Demographic groups J1 J2 J3 J4 Males Non-gravid females Gravid females

Sites Rimel TRIC Rimel TRIC Rimel TRIC Rimel TRIC Rimel TRIC Rimel TRIC Rimel TRIC

Number of individuals 221 161 206 96 164 81 228 57 245 95 201 96 203 110

a Mean 33.04 36.13 43.22 42.57 58.48 48.23 61.57 66.07 63.24 68.08 56.73 59.50 52.19 55.86

SD 2.26 2.08 3.75 0.85 2.56 4.78 0.75 8.06 3.39 2.54 0.71 2.84 2.10 1.17

b Mean 9.34 8.16 10.95 9.42 11.00 9.86 10.39 12.76 14.78 13.19 13.50 13.50 13.86 13.85

SD 0.74 0.51 0.39 0.31 0.19 0.82 0.47 1.46 0.98 0.87 1.64 0.34 1.46 0.27

c Mean 12.96 11.83 14.91 13.91 15.24 13.75 14.23 18.21 18.24 16.75 18.37 17.87 19.44 18.16

SD 1.03 0.75 0.60 0.50 0.16 1.18 0.86 2.09 0.76 0.96 0.65 1.18 0.90 0.37

d Mean 1.25 1.28 1.21 1.25 1.20 1.23 1.22 1.30 1.15 1.17 1.16 1.16 1.16 1.16

SD 0.02 0.02 0.01 0.03 0.00 0.04 0.01 0.20 0.01 0.01 0.02 0.01 0.02 0.00

CL/AW Mean 4.50 3.80 4.16 3.71 4.76 4.36 5.00 4.51 4.30 6.70 2.68 4.14 2.69 4.19

SD 0.45 0.12 0.11 0.17 0.10 0.11 0.19 0.15 0.23 0.62 0.19 0.28 0.22 0.10

PL/CL Mean 1.39 1.45 1.36 1.48 1.39 1.39 1.37 1.43 1.23 1.27 1..38 1.32 1.41 1.31

SD 0.04 0.02 0.02 0.02 0.02 0.01 0.04 0.06 0.13 0.01 0.15 0.10 0.11 0.02

IPL/PL Mean 7.51 6.40 9.08 7.53 9.15 8.03 8.53 9.84 12.89 11.27 11.62 11.60 11.99 11.93

SD 0.73 0.50 0.39 0.42 0.18 0.89 0.47 0.43 0.90 0.85 1.54 0.32 1.40 0.25

IPL/CL Mean 10.42 9.27 12.37 11.11 12.68 11.21 11.69 14.06 15.80 14.30 15.80 15.35 16.81 15.65

SD 1.00 0.74 0.58 0.63 0.16 1.28 0.82 0.79 0.67 0.96 0.71 1.08 0.97 0.35

HW/MW Mean 0.45 0.44 0.43 0.47 0.41 0.49 0.40 0.51 0.45 0.39 0.45 0.44 0.41 0.44

SD 0.00 0.01 0.00 0.01 0.00 0.01 0.00 0.01 0.00 0.00 0.01 0.01 0.00 0.01

AW/MW Mean 0.57 0.80 0.70 0.83 0.81 0.80 0.87 0.80 0.81 0.61 1.16 0.81 1.00 0.73

SD 0.04 0.01 0.03 0.01 0.03 0.01 0.01 0.01 0.01 0.04 0.06 0.04 0.00 0.01

a, b, and c De Man’s ratios, TL total length, IPL intermediate piece length, d TL/IPL, PL pharyngeal length, MW maximum width, HW headwidth, AW anal width, CL caudal length

Environ Sci Pollut Res (2012) 19:646–655 651

juveniles. In the period from 26 November 2004 to 18February 2005, juveniles J1 developed to fourth stagejuvenile J4. Peaks of males, females, and gravid femaleswere recorded from 18 February to 1 April 2005.Second reproductive cycle: The appearance of J1occurred during the first sampling in August 2005,and these juveniles matured (J1–4) in the course of thefollowing weeks while first males were observed inOctober and first females a couple of weeks later at theend of October and the first weeks of November. Alsopeaks in abundance of females and gravid females wereobserved at the same sampling dates, and it is assumedthat most females are fertilized during a period of lessthan 15 days. After deposition and hatching of eggs,small juveniles J1–2 were observed in November.

3.5 Approximate life table

The time span of juvenile stages of O. villoti generally wasshorter at TRIC than at Rimel, except for the time span of J4of the second reproductive cycle (Fig. 3). The molt into malespreceded the molt into females by 2 weeks, and males arebelieved to live longer than females. Mature organisms started

to copulate immediately after the last molt and continuedintermittently. Females started egg production within 45 daysafter fertilization at Rimel and within circa 2 weeks at TRIC(both reproductive cycles). After having obtained maximumsize, O. villoti started oviposition at both TRIC and Rimel inthe two annual reproductive cycles. Most J1 appeared within30 days (cycle 1) or 90 days (cycle 2) after appearance ofgravid females. At TRIC, the time required for oviposition,embryo development, and finally egg hatch accounted for15 days (cycle 1) and 150 days (cycle 2).

Significant differences were observed in the egg-to-eggdevelopment time between Rimel and TRIC. With respectto the first cycle, generation time at unpolluted Rimel took97–171 days; however, at polluted TRIC, it took only 52–112 days. The second cycle required more time than thefirst one at both sampling sites: 172–217 days at Rimel and225–240 days at TRIC; this slower development was due tothe increase in time the eggs needed for hatching.

3.6 Relative growth of the intermediate piece

The slopes, intercepts, and regression fit (R2) of the linearregressions between total length and intermediate piece

0

55

110

165

22026

NO

V.

10 D

EC

.

24 D

EC

.

07 J

AN

.

21 J

AN

.

04 F

EB

.

18 F

EB

.

04 M

AR

.

18 M

AR

.

01 A

PR

.

15 A

PR

.

29 A

PR

.

13 M

AY

.

27 M

AY

.

10 J

UN

.

24 J

UN

.

08 J

UL

.

22 J

UL

.

05 A

UG

.

19 A

UG

.

02 S

EP

.

16 S

EP

.

30 S

EP

.

14 O

CT

.

28 O

CT

.

11 N

OV

.

25 N

OV

.

Density (ind. 30 cm-2)

2004 2005

0

20

40

60

80

26 N

OV

.

10 D

EC

.

24 D

EC

.

07 J

AN

.

21 J

AN

.

04 F

EB

.

18 F

EB

.

04 M

AR

.

18 M

AR

.

01 A

PR

.

15 A

PR

.

29 A

PR

.

13 M

AY

.

27 M

AY

.

10 J

UN

.

24 J

UN

.

08 J

UL

.

22 J

UL

.

05 A

UG

.

19 A

UG

.

02 S

EP

.

16 S

EP

.

30 S

EP

.

14 O

CT

.

28 O

CT

.

11 N

OV

.

25 N

OV

.

Density (ind. 30 cm-2)Juveniles

GF2004 2005

0

20

40

60

80

100

26 N

OV

.

10 D

EC

.

24 D

EC

.

07 J

AN

.

21 J

AN

.

04 F

EB

.

18 F

EB

.

04 M

AR

.

18 M

AR

.

01 A

PR

.

15 A

PR

.

29 A

PR

.

13 M

AY

.

27 M

AY

.

10 J

UN

.

24 J

UN

.

08 J

UL

.

22 J

UL

.

05 A

UG

.

19 A

UG

.

02 S

EP

.

16 S

EP

.

30 S

EP

.

14 O

CT

.

28 O

CT

.

11 N

OV

.

25 N

OV

.

Percentage (%)

0

20

40

60

80

100

26 N

OV

.

10 D

EC

.

24 D

EC

.

07 J

AN

.

21 J

AN

.

04 F

EB

.

18 F

EB

.

04 M

AR

.

18 M

AR

.

01 A

PR

.

15 A

PR

.

29 A

PR

.

13 M

AY

.

27 M

AY

.

10 J

UN

.

24 J

UN

.

08 J

UL

.

22 J

UL

.

05 A

UG

.

19 A

UG

.

02 S

EP

.

16 S

EP

.

30 S

EP

.

14 O

CT

.

28 O

CT

.

11 N

OV

.

25 N

OV

.

Percentage (%)J1

J2

J3

J4

0

20

40

60

26 N

OV

.

10 D

EC

.

24 D

EC

.

07 J

AN

.

21 J

AN

.

04 F

EB

.

18 F

EB

.

04 M

AR

.

18 M

AR

.

01 A

PR

.

15 A

PR

.

29 A

PR

.

13 M

AY

.

27 M

AY

.

10 J

UN

.

24 J

UN

.

08 J

UL

.

22 J

UL

.

05 A

UG

.

19 A

UG

.

02 S

EP

.

16 S

EP

.

30 S

EP

.

14 O

CT

.

28 O

CT

.

11 N

OV

.

25 N

OV

.

Sampling dates

Density (ind. 30 cm-2)

0

20

40

60

26 N

OV

.

10 D

EC

.

24 D

EC

.

07 J

AN

.

21 J

AN

.

04 F

EB

.

18 F

EB

.

04 M

AR

.

18 M

AR

.

01 A

PR

.

15 A

PR

.

29 A

PR

.

13 M

AY

.

27 M

AY

.

10 J

UN

.

24 J

UN

.

08 J

UL

.

22 J

UL

.

05 A

UG

.

19 A

UG

.

02 S

EP

.

16 S

EP

.

30 S

EP

.

14 O

CT

.

28 O

CT

.

11 N

OV

.

25 N

OV

.

Sampling dates

Density (ind. cm-2) M

NGF

♣

♣ ♣ ♣♣

♣

♣♣ ♣ ♣

Reproductive cycle 1

Reproductive cycle 2

Reproductive cycle 1

1

Reproductive cycle 2

♣

♣♣

Fig. 2 Abundances of juveniles and adults of O. villoti from Rimel(left) and TRIC (right) during one sampling year (November 2004–November 2005). J1 juvenile of the first stage, J2 juvenile of thesecond stage, J3 juvenile of the third stage, J4 juvenile of the fourth

stage, M males, NGF non-gravid females, GF gravid females. Blackclub suit major periods of egg-laying at Rimel and TRIC. Continuousarrows link pools of gravid females and juveniles J1

652 Environ Sci Pollut Res (2012) 19:646–655

length are given in Table 3. There were significant differ-ences (p<0.05) between the two study populations in thecase of J1 and J2 juveniles, males, non-gravid females, andgravid females, with steeper slopes at the TRIC site than atRimel.

3.7 Egg volume

Gravid females at the TRIC site produced significantlysmaller eggs relative to Rimel (0.30±0.04 vs 0.46±0.08 nl, respectively, p<0.001). The significance of thispattern was tested by the Z test (Z test: p<0.001). Therewas absolutely no difference in fecundity of specimensbetween the two sites, and all examined gravid femalespossessed two eggs ready to be expelled independent oftheir native site.

4 Discussion

4.1 Abiotic environment

TRIC and Rimel sites were similar in temporal fluctuationsof water parameters and grain composition of sediments.This was expected because the sites were close to eachother (only 3.7 km). In contrast, hydrocarbon and organicloads were far more important in sediments at TRIC.

4.2 Biometry

Z test (Table 1) revealed that nematodes from TRIC weregenerally significantly smaller in size. Differences in bodyvolume may be due to differences in length and/or width. Thelengths most informative as bioindicators include the interme-

Juveniles J4

Juveniles J1

Males

Non-gravid females

Gravid females

0.5 - 1 month (RC 1)Less than 0.5 month (RC 2)

0.5 - 1.5 month (RC 1)0.5 - 1 month (RC 2)

Less than 1.5 month0.5 - 1 month (RC 1)

3 months (RC 2)

Rimel

Juveniles J2

Juveniles J3

0.5 - 2 months (RC1) Less than 1 month (RC 2)

0.5 - 1 month

0.5 month (RC 1)0.5 - 1 month (RC 2)

Juveniles J4

Juveniles J1

Males

Non-gravid females

Gravid females

Less than 0.5 month (RC 1)0.5 - 1.5 month (RC 2)

≈ 0.5 month (RC 1)1.5 - 2 months (RC 2)

≈ 0.5 month (RC 1)Less than 0.5 month (RC 2)

0.5 month (RC 1) 5 months (RC 2)

TRIC

Juveniles J2

Juveniles J3

0.5 - 1.5 months (RC 1) Less than 0.5 month (RC 2)

Less than 0.5 month

0.5 month (RC 1)Less than 0.5 month (RC 2)

Fig. 3 Approximate life cyclesof O. villoti from Rimel andTRIC (Bizerte bay, Tunisia).RC reproductive cycle

Environ Sci Pollut Res (2012) 19:646–655 653

diate piece length for juveniles J1, J2, and J3 and males, thecaudal and pharyngeal lengths for juveniles J4, and lengths ofthe intermediate piece and pharynx for females.With respect towidths, the maximum body width and head widths forjuveniles J1 and J2 were the best tools in biomonitoring. Forjuveniles J3 and J4 and gravid females, the best tools were themaximum body width and anal width. The anal and headwidths were the best tools for males and non-gravid females.All these measures showed greater differences between sites.

Juveniles of nematodes molt four times, and generallyfemales become longer when gravid. At Rimel, the referencesite, all juvenile stages except the third and the egg productionwere accompanied with an increase in length of theintermediate piece. In addition, the third stage was character-ized by an increase in the relative length of the pharynx andthe tail; however, the fourth stage showed a notable increase inthe relative length of the tail. Some differences were recordedin the case of O. villoti from TRIC. Commonly, juvenilestages and egg production were accompanied by a longerintermediate piece. The second and the fourth stages wereassociated with a clear increase in tail length.

4.3 Life cycle

During the sampling year 2004–2005, O. villoti was notpresent for a period of 2 and 4 months at Rimel and TRIC,respectively (Fig. 2). This absence of specimens suggestspattern of seasonal reproduction. Peaks of high abundanceof juveniles and females reinforce this hypothesis. Lifecycles of parasite nematodes may include protectivestrategies relevant to understanding survival of eggs of O.villoti during the period of low oxygen availability (earlyspring to late summer). In fact, studies of Imbriani andPlatzer (1981) and Saunders et al. (2000) demonstrate thateggs of parasite nematodes cannot develop in media devoidof oxygen and may undergo no division even after 2–3.5 months of hypoxia, although undeveloped eggs mayremain viable in such conditions.

Little is known about the life cycles of nematodes. O.villoti is one of the longer species belonging to the genus

Odontophora together with Odontophora armata, Odonto-phora furcata, Odontophora mercurialis, and Odontophoraparavilloti. Body lengths of these species range between 3and 4 mm. Thus, it is inferred that O. villoti possesses arelatively slower development and consequently a longergeneration time. There was no difference in the number ofreproductive cycles per year (i.e., two yearly reproductivecycles) between Rimel and TRIC. The egg-to-eggdevelopment time of the first reproductive cycle waslower than that of the second reproductive cyclebecause of the slowness of the egg maturation. Twoannual reproductive cycles with different egg-to-eggdevelopment times were also observed in Axonolaimusarcuatus from Banyuls-Sur-Mer (France) (De Bovée 1981).

Under stress, juveniles of O. villoti might have a shorterduration, with the exception of the transition from juvenilesJ4 to females during the second reproductive cycle (Fig. 3).The duration was also shortened for the period up to egghatch and appearance of juveniles J1 in the first reproduc-tive cycle. In contrast, the rest of life cycle steps were ofshorter durations at TRIC (Fig. 3). Yet, it is clear that theegg-to-egg development time of O. villoti was shorter atTRIC. Such results may be linked to the effective differencein relative growth of the intermediate piece. The ANCOVAanalysis (Table 3) showed that the relative growth of thispart of the body and reproductive organs (ovaries anduterus) were significantly faster for juveniles J1 and J2,males, non-gravid females, and gravid females from TRICcompared with their counterparts from Rimel. The signif-icantly larger eggs (Z test) produced by females of Rimelseemed to be linked to slower reproductive activity, and thismay be because eggs took more time to form outer layersfrom uterine secretions.

5 Conclusions

This study showed, comparing two different beaches (Rimeland TRIC), that “Tunisian Refining Industries Company”-related activities reduce the body dimensions of O. villoti and

Table 3 Allometric regressionparameters of linearregressions performed on naturallogarithm-transformedlengths of the whole bodyand that of the intermediatepiece of O. villoti andslope comparisons withANCOVA

Significant differences(p<0.05) were indicatedwith bold values

Rimel TRIC ANCOVA

Slope Intercept R2 Slope Intercept R2 p value

J1 1.0806 0.912 0.9580 1.2271 −0.8835 0.9994 <0.001

J2 0.9114 1.8836 0.8961 1.1084 1.7820 0.9224 0.027

J3 1.0064 0.5001 0.8604 0.9768 1.1160 0.7875 0.420

J4 1.1003 0.9882 0.9926 1.0952 −1.5611 0.9971 0.865

Males 0.7403 0.9964 0.8032 0.8758 1.7956 0.9940 <0.001

Non-gravid females 0.8033 1.2229 0.9953 0.9303 0.4195 0.9174 <0.001

Gravid females 0.8225 0.9203 0.8599 0.9561 1.3112 0.8549 <0.001

654 Environ Sci Pollut Res (2012) 19:646–655

result in an overall decrease in generation time. This may bea functional adaptation of O. villoti in response to pollutionby reducing the exposure time to toxicants and speeding upmost of the steps in its life cycle. However, while this maywell be the case, the potential presence of other influencingfactors should not be excluded. For example, worms thatexperience high mortality rates as adults may put moreenergy into reproduction at earlier ages and less into growth(and therefore grow more slowly). For example, highmortality rates as adults may be the case at Rimel whereinter-specific competition stress might occur due to the highdensity of the community (unpublished data). Elsewhere,predation can reduce the productivity of O. villoti fromRimel where a predator species, Mesacanthion monhystera,was the most dominant species (unpublished data).

Based on the current work, it is apparent that O. villotican persist in high levels of hydrocarbons by using severaladaptive strategies. These findings could be important basesfor possible future studies under controlled laboratoryconditions exploring the bioindicator potential of O.villoti. A measurement of tissue levels of hydrocarbons (orother biomarker of exposure to hydrocarbons) would bevery helpful, even if many nematodes were required forsuch analysis. Nevertheless, the required experiments maybe of value only at monospecific level. This imposes todirect the future studies to find a method to select O.villoti from rest of the community. This work could beinteresting in biomonitoring and also for bioremediationbecause of three reasons:

– The optimum requirements of experiments (duration,temperature, oxygen, salinity, etc.) could be deducedfrom such a field study.

– O. villoti has a high egg-to-egg development time(about 100 days) compared to majority of nematodespecies (mostly days to weeks according to Kennedyand Jacobi (1999)). The low mortality rate makescompounds the difficulties of the proliferation of moldsand pathogens that are difficult to manage in anexperimental system (e.g., microcosms).

– O. villoti is a non-selective deposit feeder (Wieser1953). Food may be experimentally obtained usinghydrocarbons as the sole energy source especially whenwe take into account that this worm can have highreproductive and growth rates when exposed to thesekinds of toxicants.

Acknowledgments Financial support was provided by theTunisian Ministry of Scientific Research and Technology. Weare thankful to Pr G. Boucher (France) and Pr P. Vitiello (France)for assistance on identification of O. villoti, especially juvenileforms. The detailed suggestions and comments of Dr D. Leduc(New Zealand), Pr. J. G. Baldwin (USA), Pr S. Ravichandran

(India), Pr L. A. Bouwman (Netherlands), Pr S. Fayyaz (Pakistan),and Pr G. Chinnadurai (India) helped substantially in improving thequality of the revised manuscript.

References

Boufahja F, Hedfi A, Amorri J, Aïssa P, Beyrem H, Mahmoudi E(2011) Examination of the bioindicator potential of Oncholaimuscampylocercoides (Nematoda, Oncholaimidae) from Bizerte bay(Tunisia). Ecol Indic 11:1139–1148

Danovaro R, Fabiano M, Vincx M (1995) Meiofauna response to theAgip Abruzzo oil spill in subtidal sediments of the Ligurian Sea.Mar Poll Bull 39(2):133–145

De Bovée F (1981) Ecologie et dynamique des nématodes d’une vasesublittorale (Banyuls-sur-mer). Thèse de Doctorat d’Etat,Université Pierre et Marie Curie, Paris VI, 194 p

Fabiano M, Danovaro R (1994) Composition of organic matter insediment facing a river estuary (Tyrrhenian Sea): relationshipswith bacteria and microphytobenthic biomass. Hydrobiologia277:71–84

Gheskiere T, Vincx M, Weslawski JM, Scapini F, Degraer S (2005)Meiofauna as descriptor of tourism-induced changes at sandybeaches. Mar Environ Res 60(2):245–265

Heip C, VincxM, Smol N, VrankenG (1982) The systematics and ecologyof free-living marine nematodes. Helminth Abstr, Ser B 51:1–31

Imbriani JL, Platzer EG (1981) Gaseous requirements for postparasiticdevelopment of Romanomermis culicivorax. J Nematol 13(4):470–476

Kennedy AD, Jacoby CA (1999) Biological indicators of marineenvironmental health: meiofauna a neglected benthic component.Environ Monit Assess 54:47–68

Mahmoudi E, Essid N, Beyrem H, Hedfi A, Boufahja F, Vitiello P,Aïssa P (2005) Effects of hydrocarbon contamination on a freeliving marine nematode community: results from microcosmexperiments. Mar Poll Bull 50:1197–1204

Saunders LM, Tompkins DM, Hudson PJ (2000) The role of oxygenavailability in the embryonation of Heterakis gallinarum eggs.Int J Parasitol 30:1481–1485

Somerfield PJ, Dee JM, Warwick RM (1994) Soft sediment meiofaunalcommunity structure in relation to a long-term heavy metal gradientin the Fal estuary system. Mar Ecol Prog Ser 105:79–88

SteyaertM, Vanaverbeke J, Vanreusel A, Barranguet C, Lucasc C, VincxM(2003) The importance of fine scale, vertical profiles in characterisingnematode community structure. Est Coast Shelf Sci 58:353–366

Vanhove S, Arntz W, Vincx M (1999) Comparative study of thenematode communities on the southeastern Weddell Sea shelf andslope (Antarctica). Mar Ecol Prog Ser 181:237–256

Vanhove S, Vermeeren H, Vanreusel A (2004) Meiofauna towards theSouth Sandwich Trench (750–6300 m), focus on nematodes.Deep-Sea Res II 51:1665–1687

Vranken G, Thielemans LK, Heip C, Vandycke M (1981) Aspectsof the life-cycle of Monystera parelegantula (Nematoda,Monhysteridae). Mar Ecol Prog Ser 6:67–72

White CR (2003) Allometric analysis beyond heterogeneousregression slopes: use of the Johnson–Neyman technique incomparative biology. Physiol Biochem Zool 76(1):135–140

Wieser W (1953) Die Beziehung zwischen Mundh6hlengestalt.Ernahrungsweise und Vorkommen bei freilebenden marinenNematoden. Eine skologisen-morphologische studie. Arkiv flirZoologie 4:439–484

Wieser W (1960) Benthic studies in Buzzards Bay. II. The meiofauna.Limnol Oceanogr 5:121–137

Environ Sci Pollut Res (2012) 19:646–655 655