Towards a Characterization of the Locomotor Activity Rhythm of theSupralittoral Isopod Tylos europaeus

Dhouha Bohli-Abderrazak,1 Amel Ayari,1 Elfed Morgan,2 and Karima Nasri-Ammar1

1Unité de recherche “Biologie Animale et Systématique Evolutive”, Faculté des Sciences de Tunis, Campus Universitaire de TunisEl Manar I, Tunis, Tunisie, 2School of Biosciences, The University of Birmingham, Birmingham, UK

Freshly collected samples of Tylos europaeus from Korba beach (northeast of Tunisia) were housed in an environmentalcabinet at controlled temperature (18°C ± .5°C) and photoperiod. Locomotor activity was recorded under twophotoperiodic regimens by infrared actography every 20 min by multichannel data loggers. One regimen simulated thenatural light-dark cycle on the day of collection, whereas the second imposed a state of continuous darkness on allindividuals. Under entraining conditions, the animals displayed rhythmic activity, in phase with the period of darkness,whereas in continuous darkness these isopods exhibited a strong endogenous rhythm with circadian and semidiurnalcomponents at mean periods of τ (h:min) = 25:09 ± 01:02 h and τ = 12:32 ± 00:26 h, respectively. Under free-runningconditions, this endogenous rhythm showed significant intraspecific variability. (Author correspondence: bohli.dhouha@gmail.com)

Keywords: Circadian, Free-running, Isopod crustacean, Semidiurnal rhythm, Tylos europaeus

INTRODUCTION

Modification of behavior constitutes an animal’s initialresponse to environmental change, and the biologicaloscillators that regulate the timing of these behavior pat-terns play a pivotal role in the adaptive reaction of thespecies. Their most obvious manifestation is in the peri-odic change in locomotor activity, and the characteristicfeatures of these endogenous rhythms have been welldefined (Aschoff, 1981). These rhythms are common-place in the animal kingdom and provide an opportunityto study a fundamental aspect of the animal’s adaptivecapacity. One way in which this may be achieved is bycomparing representatives of the same or closelyrelated species from the same and different habitats.The marine littoral is of particular interest in thisrespect, comparing a dynamic gradient of geophysicalchanges recurring over a range of different substrataand microhabitats. Amphipod and isopod crustaceansconstitute major components of these habitats. For themost part they share a common basic morphology, andtheir ecological diversity may be attributed, essentially,to changes in behavior and physiology.

Activity rhythms of many littoral and near-littoralspecies have been studied, and both circadian and

circatidal rhythms, occurring independently and in com-bination, have been described. However, the predomi-nant period and phase of the active state varies with thegeophysical cycles prevailing in the habitat. Supralittoralamphipods, notably talitrids, generally show a nocturnalcircadian activity rhythm, such as shown by Orchestiagammarellus, Orchestia mediterranea and Orchestia cavi-mana (Wildish, 1970), Talorchestia deshayesii (Ayariet al., 2007; Willams, 1982, 1983), and Talitrus saltator(Bohli et al., 2006; Bregazzi & Naylor, 1972; Nasri &Morgan, 2005; Williams, 1980a, 1980b, 1980c), whereasthe tidal estuarine species Corophium volutator showsalmost exclusively a semidiurnal tidal pattern (Holm-strom &Morgan, 1983) that in the sand beach amphipodBathyporeia pelagica is modulated over a circadianperiod (Fincham, 1970a, 1970b).

Similar diurnal modulation of the circatidal pattern ofswimming has been recorded in the surf-migrant isopodEurydice pulchra from the European sand beaches of theNorth Atlantic, and a circa-semilunar monthly pattern oftotal daily swimming is also evident (Reid & Naylor,1985). However, in Excirolana chiltoni, an isopod thatoccupies a similar ecological niche on Americanshores, alternate peaks of the endogenous rhythm of

Dhouha Bohli-Abderrazak and Amel Ayari contributed equally to the conduct of the research and writing of the manuscript.Address correspondence to Dhouha Bohli-Abderrazak, Unité de recherche “Biologie Animale et Systématique Evolutive”, Faculté desSciences de Tunis, Campus Universitaire de Tunis El Manar I 2092, Tunis, Tunisie. Email: bohli.dhouha@gmail.com

Submitted April 15, 2011, Returned for revision May 7, 2011, Accepted December 12, 2011

Chronobiology International, 29(2): 166–174, (2012)Copyright © Informa Healthcare USA, Inc.ISSN 0742-0528 print/1525-6073 onlineDOI: 10.3109/07420528.2011.652327

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Con

nect

icut

on

01/0

1/14

For

pers

onal

use

onl

y.

swimming activity are of unequal amplitude, shadowingthe amplitude of the prevailing semidiurnal tides(Enright, 1965, 1972; Klapow, 1972).

More recently, Yannicelli et al. (2000) studied theactivity rhythm of two Uruguayan cirolanid isopods,which occupy different beach levels, using both fieldobservations and laboratory experiments. Excirolanaarmatawas seen to be activemost of the time, but labora-tory results showed that emergence under constant con-ditions was rare. However, individuals of Excirolanabraziliensis were always observed during the night, andthey displayed an endogenous circadian activity.

In the Mediterranean, species of the genus Tylosoccupy a similar niche as Excirolana. Tylos species aremorphologiquely similar but inhabit semipermanentburrows on the upper reaches of intertidal sandybeaches (Hayes, 1969). Kenslay (1974) suggested thatTylos granulatus displays both diurnal and tidal cycles.This species emerges only nocturnally, at low tide, toforage for 2–3 h, so that the possibility of animalsbeing swept away by the rising tide is minimized, andmeasurements of the respiration rate of T. granulatusshow that the respiratory rhythm is closely related tothe activity cycle. Oxygen consumption is low duringthe day, but there is a circatidal rhythmic increaseduring the nocturnal low-tide period (Marsh & Branch,1979). Moreover, field observation of a Chilean popu-lation of Tylos spinolosus has shown higher locomotoractivity during neap tide than during spring tide. Further-more, laboratory studies revealed a circadian locomotorrhythm with a probable circatidal component (Jaramilloet al., 2003).

Tylos europaeus was found to be mainly active insummer and autumn and during the night, and waszoned along the eulittoral (Fallaci et al., 1996). Thesurface activity was influenced by almost all the environ-mental factors when they were limiting, but especially bythe relative humidity of the air. These authors reported anegative temperature effect on activity by indicating thatmore animals were surface active when air temperaturewas between 13°C and 20°C. A decreased activity at rela-tive humidity ≤70% or ≥90% was also reported. Thepurpose of this study is firstly to determine the differentcharacteristics of the endogenous locomotor activityrhythm of Tylos europaeus in laboratory controlled con-ditions and then to interpret these results from an eco-logical perspective, in order to understand the behaviorof animals on the ground.

MATERIALS AND METHODS

The study location (Figure 1) was the large beach (100 mlength) of Korba (36°36′N, 10°52′E; Gulf of Hammamet,Tunisia) that is characterized by a considerable dunesystem. Adults of Tylos europaeus were collected byhand during the early morning (08:30 h) at 22°C on 15April 2006 and brought to the university in isolated con-tainers. The investigative protocol conformed to the

ethical principles stipulated by the journal for theconduct of animal biological rhythm research (Portalup-pi et al., 2010).

Once in the laboratory, 30 randomly selected animalswere individually transferred with humid sand from thelocality of origin and food, in 30 annular recordingchambers. Actographs weremade from opaque Plexiglas,9.5 cm high and with internal and external diameters of 4and 11 cm, respectively (Bohli et al., 2006; Nardi et al.,2003; Rossano & Scapini, 2008). Locomotor activity wasphotoelectrically monitored as the frequency of interrup-tion of an infrared beam. This interruption caused anevent to be registered on a data-logging device thatdownloaded the number, every 20 min, of each beaminterruption to a computer. The recording apparatusand controlled environment cabinet were designed andconstructed in the School of Biosciences University ofBirmingham, Birmingham, UK (Figure 2).

To characterize the locomotor activity rhythm, theexperiment was conducted under two photoperiodicregimens at a temperature of 18°C ± 1°C. The experimen-tal conditions were similar to those described by Nardiet al. (2003) and Scapini et al. (2005). For the first 6 d,samples were kept in a light (L)-dark (D) regimenapproximating the day of collection. The nLD (naturallight dark) cycle (h:min) was of 13:16/10:44 h, and bothsunrise and sunset were, respectively, at 05:45 and06:56 h. The light intensity was 60 lux. For the rest ofthe experiment (7 d), these animals were kept in constantdarkness (DD).

Raw data were displayed as a double-plot actogram bythe software Chart35 (D.D. Green, Biosciences Work-shop, University of Birmingham), showing cumulativeactivity values per 20-min intervals. This provided theoption of normalizing the data by expressing each loco-motor value as a percentage of the highest value for the24 h over which it occurred or the highest value overthe entire time series; (Cummings &Morgan, 2001). Sub-sequent periodogram analysis was performed using theprogram based on the method of Dorscheidt and Beck(Harris & Morgan, 1983). Each time series was scannedfor periods (τ) between 10 and 30 h and was accepted

FIGURE 1. Map representing the study site location; Korba,North East of Tunisia (36°36′N, 10°52′E).

Locomotor Activity Rhythm of the Isopod T. europaeus

© Informa Healthcare USA, Inc.

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Con

nect

icut

on

01/0

1/14

For

pers

onal

use

onl

y.

as rhythmic if the periodogram peaks exceeded the 95%confidence level. This technique also gives a goodmeasure of the rhythm definition, in the form of thesignal-to-noise ratio (SNR), measured as the differencebetween the periodogram peak value and its correspond-ing 95% confidence limit. The waveform curves, obtainedby calculating the mean activity (in %) per hour per dayfor each individual, were plotted relative to astronomictime under entraining conditions and circadian timeunder free-running conditions.

Other factors of potential adaptive significanceinclude the phase relationship of the active state, andthe relative duration of the activity/rest-time componentsof the oscillation. Phase values were calculated using atechnique in which the median point of the main activitybloc is correlated to the time of onset of the dark phase,expressed in degrees. The correlation coefficient betweenthe active phase (α) and its corresponding rest time (ρ) isnegative, resulting in a stabilization of the free-runningperiod (τ). This relationship has been investigated hereusing the method of Aschoff et al. (1971). The percentageof survival, number of active animals, number of animalsshowing periodicity, mean periods, activity and resttimes, phase shift, and SNR (a measure of rhythm defi-nition) were calculated for each individual and for thewhole population. Statistically, differences between per-centages were analyzed by χ2 test, and differencesbetween all mean values were calculated using the non-parametric Wilcoxon rank test (Nardi et al., 2003; Nasri &Morgan, 2005, 2006).

RESULTS

After 14 d of recording and at the end of the experiment,the 30 animals were found alive and well active with

several burrows in their actographs. The activity of oneanimal was not recorded, probably due to a recordingbox failure. Using the actogram analysis, three differentactivity patterns were identified, i.e., uni-, bi-, and pluri-modal (Figure 3), and the most representative one waspresented for each type as an example. Under nLD(natural LD) cycle, these patterns were equally rep-resented, with a small advantage for the bimodal one.In contrast, under free-running conditions, the bimodalprofile became the least represented, whereas the pluri-modal pattern became the most common. On all theseactograms, the locomotor activity still concentratedexclusively during the experimental night under entrain-ing conditions, and drifted to the right under continuousdarkness. These actograms indicate the presence of a cir-cadian period in locomotor activity as confirmed by per-iodogram analysis. In addition, many ultradian periodswere revealed, the most frequent one being close to 12 h.

Activity under nLD cycleFor the first profile (Figure 4A), the activity shows anunimodal aspect, and the activity peak is observed essen-tially just after lights-off, with an exception during the

FIGURE 2. Recording apparatus: a = environmental chamber; b = data logger; c = computer; d = actograph; e = photophase programmableswitch; f = thermometer; g = variable light intensity. 1 = humid substrate; 2 = infrared beam.

FIGURE 3. Percentage of different patterns (uni-, bi-, and pluri-modal) under both nLD (white) and DD (gray) conditions, N = 29.Data are percentage + standard error; * p < .05.

D. Bohli-Abderrazak et al.

Chronobiology International

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Con

nect

icut

on

01/0

1/14

For

pers

onal

use

onl

y.

FIGURE 4. Double plotted actograms (A, B, C) corresponding to the different locomotor activity patterns of Tylos europaeus under bothnLD and DD conditions. a1, a2; b1, b2; c1, c2 and a3, a4; b3, b4; c3, c4 represent, respectively, relative waveforms and periodograms. The grayarea corresponds to the dark phases (experimental night d1–6; continuous darkness d7–13). Dotted lines (- - - -) represent the activity drift.

Locomotor Activity Rhythm of the Isopod T. europaeus

© Informa Healthcare USA, Inc.

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Con

nect

icut

on

01/0

1/14

For

pers

onal

use

onl

y.

second day. The relative waveform (Figure 4a1) confirmsthis unimodal aspect. In the second actogram(Figure 4B), activity displays a bimodal pattern; the firstpeak is observed just after the experimental dusk and thesecond, which is the most prominent one, with the exper-imental dawn. This bimodal aspect is illustrated by therelative mean activity curve (Figure 4b1). The locomotoractivity exhibited many large peaks in the third actogram(Figure 4C), where a plurimodal pattern and the relativewaveform describe a serrated profile (Figure 4c1). For thefirst actogram, periodogram analysis (Figure 4a3) revealeda significant ultradian period equal to τ = 12 h. This ultra-dian rhythmicity was found in 24% of tested specimens(Table 1), with mean period (h:min) of τLD12h = 11:57 ±00:08 h (Table 2). On the other hand, the same analysis(Figure 4a3, 4b3, 4c3) determined an entrained circadianperiod, respectively, equal to τ = 24 h, τ = 23:40 h, and τ= 24 h. The circadian rhythmicity (% variance accountedfor) was 96% (Table 1), and the mean period for allsamples was equal to τLD24h = 23:49 ± 00:16 h (Table 2).

Activity during Continuous DarknessIn the second part of the experiment, and despite theabsence of the synchronizer, the locomotor patterns dis-played appreciable rhythmicity. On the first pattern(Figure 4A), the activity was sometimes unimodal andsometimes multimodal, but the waveform curve de-scribed a unimodal profile, and the activity peak was ob-served just after subjective dusk (Figure 4a3). The globaldrift seemed to move towards the right, but at d 7, 10,and 11 it shows an attempt of resynchronization with aleft drift. The second one (Figure 4B) and the relative wa-veform (Figure 4b2) show a serrated pattern. During thefirst day, the activity rhythm preserved its synchroniza-tion. Starting from the second day, the activity profiledrifted to the right. The last actogram (Figure 4C) de-scribes overall a bimodal profile as evident by its respect-ivemean activity curve (Figure 4c2). Periodogramanalysis(Figure 4a4, 4b4, 4c4) revealed two different periods in all

patterns. One was a circadian respectively equal to τ =24:20, τ = 25, and τ = 24:20 h. The mean period of τDD24hwas 25:09 ± 01:02 h (Table 2). The other was ultradianrespectively equal to τ = 12:20, τ = 12:40, and τ = 12:20 h,and the mean period of τDD12h was 12:32 ± 0:26 h(Table 2). Both semidiurnal and circadian periods length-ened fromentrainment to free-running conditions, and thenonparametric Wilcoxon test revealed highly significantdifferences (Table 2). Under DD conditions, the circadianrhythmicity percentage remained unchanged (96%). Inaddition, the semidiurnal one significantly increasedfrom 24% to 52% (χ2 = 4.6869, df = 1, p = .0304; Table 1).

The SNR ratio was estimated for both the circadianand semidiurnal components (Table 2). The first oneremained clearly defined in LD and DD with meanvalues of .4670 ± .1352 and .4828 ± .2368, respectively.However, the second one became well defined underfree-running conditions. In fact, the SNR value consist-ently increased from .0269 ± .0403 (nLD) to .2501± .1032 (DD), and theWilcoxon test reveals highly signifi-cant difference ( p = .0077; Table 2).

The curve of Figure 5 represents the rest time in termsofthe activity time. As shown in Table 2, the animalsremained active longer in DD conditions, with thehighest mean activity time of 10:26 ± 2:35 h, than undernLD conditions (8:14 ± 1:07 h). The Wilcoxon testrevealed highly significant difference, with p = .0007(Table 2). Besides examining the different standarderrors under both nLD and DD, the large distribution ofvalues around themean activity time in the latterconditionreflects great individual variability. This variability is alsoobserved in the curve by comparing the R2 (0 ≤ R2 ≤ 1).R2 close to 1 indicates the dispersion around the mean islow and, therefore, the variability is of low importance.

For the whole sample, the increase in the α/ρ ratio isstatistically significant (p = .01; Table 2). In theseisopods, the α/ρ ratio is a function of τ, and the correlationis illustrated in Figure 6. This ratio decreased withincrease in τ under entraining conditions (negative

TABLE 1. Percentage of rhythmic individuals under both nLD and DD conditions.

LD DD χ2 test

R12h (%) 24%(n = 7/29) 52%(n = 15/29) p = .0304R24h (%) 96%(n = 28/29) 96%(n = 28/29) NS

R12h and R24h correspond, respectively, to semidiurnal and circadian rhythmicity. NS = nonsignificant.

TABLE 2. Locomotor activity rhythm parameters of Tylos europaeus

Regimens τ12h ± SE τ24h ± SE SNR12h ± SE SNR24h ± SE α/ρ ± SE Activity time ± SE

LD (h:min) 11:57 ± 00:08 23:49 ± 00:16 .0269 ± .0403 .4670 ± .1352 .53 ± .108 08:14 ± 01:07DD (h:min) 12:32 ± 00:26 25:09 ± 01:02 .2501 ± .1032 .4828 ± .2368 .84 ± .424 10:26 ± 02:35Wilcoxon test p = .0015 p < .001 p = .0077 NS p = .0013 p < .001

τ12h = semidiurnal mean period; τ24h = circadian mean period; SNR = signal to noise ratio; α = activity time; ρ = rest time; SE = standarderror; NS = nonsignificant.

D. Bohli-Abderrazak et al.

Chronobiology International

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Con

nect

icut

on

01/0

1/14

For

pers

onal

use

onl

y.

correlation), and increased with increase in τ under free-running conditions (positive correlation). On the otherhand, as seen in the Figure 4, R2 was lower in DD; there-fore, there is a greater variability between individualsunder this condition (R2

nLD = .0277 vs. R2DD = .0033).

Phase shift analysis was technically possible for only23 individuals. Most of them (82.6%) showed a negativephase shift, i.e., phase delay (Figure 7), averaging−46.47° ± 43.5°. This explains the activity drift observedon the actograms under DD conditions. In fact, thisphase shift makes periods longer from day to day. Themore negative is this value, the longer is the period.

DISCUSSION

The locomotor activity pattern recorded for Tylos euro-paeus in the present study is typical of many small

crustaceans that inhabit semipermanent burrows in theupper reaches of intertidal sandy beaches. Nephrops nor-vegicus is a burrowing decapod that inhabits the muddycontinental shelf in the northeast Atlantic. Its distributionextends into much deeper areas on the continental slopein the Mediterranean Sea (Aguzzi et al., 2004b). Thisdecapod exhibits a circadian locomotor activity rhythmwith an ultradian component during the expected nightphase and day-night transition, respectively. Talitrussaltator is widely distributed on the Mediterranean andeastern Atlantic shores, where it burrows in the supralit-toral zone during the day, emerging from the sand duringthe night to feed and breed (Nasri & Morgan, 2005, 2006;Scapini et al., 1997; Williams, 1995). This nocturnalactivity pattern is regulated by an endogenouscircadian oscillator.

In the nLD entraining regimen, the animals wereactive mainly during the dark span, a nocturnal behaviorconsistent with the field observation of Kensley (1974)on Tylos latreille. Just as for the nocturnal supralittoralAmphipod Talitrus saltator, the nocturnal activity mini-mizes the danger of desiccation for Tylos europaeus(Bregazzi, 1972; Williamson, 1951). During the day, theanimals remain in burrows where temperature andhigh relative humidity (90%) are constant (Kensley,1974). At night, the relative surface humidity increasesdue to the decline of temperature below that inside theburrows. At this time, oxygen consumption of Tylos gran-ulatus is higher than during quiescence (Marsh &Branch, 1979), and the low temperature restricts respirat-ory energy loss. These conditions encourage the animalsto leave their burrow and to start their activity outside.

The persistence and precision of rhythmic locomotoractivity under free-running conditions, with mean circa-dian period (τDD = 25:09 ± 1:02 h) different than the exter-nal daily cycle (T = 24 h), is an indication of a strongendogenous component (Aschoff, 1967). This has alsobeen suggested for Tylos granulatus (Imafuku, 1976) forthe sympatric amphipod Talitrus saltator (Bohli & et al.,2006; Bregazzi & Naylor, 1972; Williams, 1979), and forother supralittoral talitrids (Ayari & Narsi-Ammar, 2008;Benson & Lewis, 1976; Wildish, 1970).

Under entraining conditions (nLD), the mean periodof τLD is 23:49 ± 00:16 h, close to the external daily cycleperiod. During continuous darkness, the circadianperiod lengthens, thereby explaining the important drifttowards the right. In fact, the animals start their activitylater from day to day. This drift, coupled with the day-to-day stability under entraining conditions, show thatthe nLD cycle is a powerful synchronizer of the Tyloseuropaeus locomotor activity rhythm. Imafuku (1976)reports also that the locomotor activity rhythm of Tylosgranulatus is predominantly synchronized by the exper-imental LD cycle. The relationship of the exogenousand endogenous elements is expressed by the phaseshift Δφ. The free-running period must be capable of cor-rection (Δφ) in order to match the period of the entrain-ing cycle (Ashoff, 1967). In this experiment, most Tylos

FIGURE 5. Correlation between the activity time (α) and the resttime (ρ) of each individual under both nLD and DD conditions.

FIGURE 6. Correlation between the α/ρ ratio and the entrained(nLD) and the circadian (DD) locomotor activity periods ofeach individual.

FIGURE 7. Phase delay (negative values) and phase advance(positive values) of the endogenous rhythm of Tylos europaeus.

Locomotor Activity Rhythm of the Isopod T. europaeus

© Informa Healthcare USA, Inc.

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Con

nect

icut

on

01/0

1/14

For

pers

onal

use

onl

y.

europaeus showed a phase delay; so, the synchronizer(nLD cycle) imposes a phase advance to maintain thelocomotor activity confined during the experimentalnight, as shown in Talitrus saltator (Williams, 1980c).

Under natural or entrained conditions, the degree offatigue produced during activity time demands resttime for recovery, as observed in Nephrops norvegicus(Aguzzi et al., 2003, 2004). Under constant conditionswhere no zeitgeber forces the subject to emerge at agiven time, a positive correlation between α and ρshould become relevant. However, the opposite effecthas been observed in man and birds (Aschoff et al.,1971), as well as for Tylos europaeus in this study. Infact, α and ρ always add up to the same mean period.

Rhythm characteristic analyses display more intraspe-cific variation under DD conditions than during nLDcycles. This variability was probably masked by the syn-chronizer effect (Chiesa et al., 2010). This result confirmsthat the LD cycle is a strong zeitgeber of the locomotorrhythm. Thus, under continuous darkness each animalexpresses its own rhythm. Bohli et al. (2006) and Nasriand Morgan (2006) demonstrated this variability withTalitrus saltator.

In addition to the circadian period, periodogramanalysis revealed significant peaks between 11:00 and13:20 h, suggesting that a circatidal component probablyexists. This pattern was more expressed in the free-running condition and was evident in 52% of samples(15/29). The main locomotor activity rhythm observedin amphipods is the circadian one, although semidiurnaltidal rhythms are also common (Rossano, 2004). Otherorganisms show tidal signals in their activity despite theabsence of appreciable tidal excursions in the CentralMediterranean, e.g., the fan worm Sabella spallanzanii(Aguzzi et al., 2007; Costa et al., 2008).

The Mediterranean is known for its weak tidal regime.However, contrary to this belief, tides do exist; they aremainly semidiurnal and have an average variation of 40cm, but they are often obscured by atmospheric con-ditions. Indeed, a headwind or, especially, atmosphericpressure higher than average reduces the effect of tidesup to making them invisible. The tidal range is moderate(4–12 cm) everywhere near the Tunisian coast, except inthe Gulf of Gabes (south of Tunisia) where it is amplifiedand reaches 1 to 1.5 m. The tidal wave crosses the straitnear the Tunisian coast while diminishing (12 cm at thenorthern coast, 5 cm at the Gulf of Hammamet) (GEF/IHEE, 2008).

Depending on the habitat, a tidal rhythm is rarelypresent and usually shown by subtidal and intertidalspecies. Enright (1963) described an evident circatidalendogenous locomotor activity rhythm in the intertidalamphipod Synchelidium sp. An interesting example isprovided by Corophium volutator (Morgan & Harris,1986), which inhabits estuaries with high tidal excursionsand shows exclusively semidiurnal circatidal rhythms,corresponding to daily tides. Hastings (1981) showedthat the exposure of Eurydce pulchra to artificial tidal

agitation in phase with the timing during the day ofspring high tides entrained a pronounced circatidalswimming rhythm. Reid and Naylor (1986) confirmedthis result. Tylos spinolosus is active after the expectedtimes of high tides and exhibits higher locomotor activityduring neap tides. The activity peaks of the periodogramclose to 11–14 h that were found probably represent a cir-catidal component (Jaramillo et al., 2003).

The presence of an endogenous oscillator necessitatesits entrainment to environmentally appropriate stimuli.Possible entraining cues that are associated with naturaltidal fluctuations include temperature, salinity, turbu-lence, current, turbidity, and depth; therefore, any ofthese could serve as cues to synchronize and entraintidal rhythms. Water level changes have been mostthoroughly investigated, and this cue can entrain circati-dal rhythms in amphipods Corophium volutator (Harris& Morgan, 1984), Portunid crabs Liocarcinus holsatus(Abello et al., 1991), and Fiddler crabs Uca crenulata(Honegger, 1973), as well as many other species. Coro-phium volutator collected intertidally and subjected tocyclical pressure changes of tidal amplitude and fre-quency, but opposite in phase to the naturally occurringtides, rapidly came into phase with this artificial cycle. It,therefore, seems that the rhythm of activity is probablyentrained by the decrease in pressure as the tide ebbs(Morgan, 1965).

T. europaeus forms burrows at the upper limits of thetidal range (Hayes, 1969), which at the site of collectionis small in comparison with that of the oceanic sea-board. Adverse geophysical changes associated with in-undation and exposure are, therefore, less predictablethan those encountered by surf migrant crustaceans,and a strong tidal component is less likely to be selectedfor (Aguzzi & Company, 2010). Where both circadianand circatidal periods in activity have been reported,the latter drift progressively later relative to the dielcycle of light and darkness, as, for example, in theshore crab Carcinus maenas (Naylor, 1958) and periwin-kle Littorina nigrolineata (Petpiroon & Morgan, 1983). Inthe present study, bimodality is associated with a singlebout of nocturnal activity, and may be of a crepuscularnature. The importance of the bimodal pattern underLD conditions can also be explained by the abruptlight-dark transitions at both experimental dusk anddawn (Bohli et al., 2006). Jones et al. (1966) and Pageand Larimer (1972) have shown gross variation in theexpression of the locomotor rhythm of the fly Anophelesgambiae and the freshwater Procambarus clarkii,respectively, correlated with the brusque twilight tran-sition. Harris and Morgan (1984) suggest that the circa-tidal rhythm of swimming activity shown by Corophiumis under two clock controls. However, circadian and cre-puscular rhythms have been reported for adults andjuveniles of the intertidal amphipod Orchestia sp.(Kennedy et al., 2000). Rossano and Scapini (2006)report similar bimodality in free-running rhythms inthe sympatric supralittoral amphipod Talitrus saltator.

D. Bohli-Abderrazak et al.

Chronobiology International

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Con

nect

icut

on

01/0

1/14

For

pers

onal

use

onl

y.

Moreover, bimodality was more frequent in individualsfrom Atlantic coasts with a high tidal range than inthose from the Mediterranean, an observation that isconsistent with a tidal influence.

In conclusion, the Tunisian population of Tylos euro-paeus exhibits a circadian locomotor activity rhythmwith an ultradian component (close to 12 h). With refer-ence to the data on tides in the Mediterranean Sea, thelatter can be considered to be a circatidal rhythm.Change in bathymetry (Harris & Morgan, 1984) andhydrostatic pressure (Morgan, 1965) can be consideredas putative controllers of the reported circatidal rhythm.However, this is the first trial with this species and allhypotheses remain plausible; so, further experimentsare necessary to confirm or refute them.

ACKNOWLEDGMENTS

This study was supported by WADI project (INCO-CT2005-015226).

Declaration of Interest: The authors report no conflictsof interest. The authors alone are responsible for thecontent and writing of the paper.

REFERENCES

Abelló P, Reid DG, Naylor E. (1991). Comparative locomotor activitypatterns in the portunid crabs Liocarcinus holsatus and L. depura-tor. J. Mar. Biol. Assoc. U.K. 71:1–10.

Aguzzi J, Company JB. (2010). Chronobiology of deep-water decapodcrustaceans on continental margins. Adv. Mar. Biol. 58:155–225.

Aguzzi J, Company JB, Sarda F, Abello P. (2003). Circadian oxygenconsumption patterns in continental slope Nephrops norvegicus(Decapoda: Nephropidae) in the western Mediterranean. J. Crust.Biol. 23:749–757.

Aguzzi J, Abelló P, Depledge M. (2004a). Endogenous cardiac activityrhythm of the Northwester Mediterranean Nephrops norvegicus(L.) inhabiting the upper slope. Mar. Fresh. Behav. Physiol.37:55–64.

Aguzzi J, Company JB, Abello P. (2004b). Locomotor activity rhythm ofcontinental slope Nephrops norvegicus (Decapoda: Nephropidae).J. Crus. Biol. 24:282–290.

Aguzzi J, Chiesa JJ, Caprioli R, Cascione D, Magnifico G, Rimatori V,Costa C. (2006). Preliminary evidences of circadian fan activityrhythm in Sabella spallanzanii (Gmelin, 1791) (Polychaeta: Sabel-lidae). Sci. Mar. 70:727–724.

Aschoff J. (1967). Circadian rhythms in birds. Proceeding of the XIVInternational Ornithological Congress, Oxford 24–30 July 1966,Snow D.W. (ed), pp. 81–105.

Aschoff J. (1981). Free-running and entrained circadian rhythms. InHandbook of behavioral new biology. Biological rhythm. 4th ed.New York: Plenum Publishing, 81–93.

Aschoff J, Gerecke U, Kureck A, Pohl H, Reiger P, Saint Paul U, Wever R.(1971). Interdependent parameters of circadian activity rhythms inbirds and man. In Menaker M (ed.). Biochronometry.Washington,DC: National Academy of Sciences, 3–29.

Ayari A, Nasri-Ammar K. (2007). Variation saisonnière du rythme loco-moteur de Talorchestia deshayesii, issu de la plage de Bizerte(Tunisie). Rapp. Comm. Int. Mer. Médit. 38:422.

Ayari A, Nasri-Ammar K. (2008). Etude expérimentale Comparativedu rythme locomoteur de deux amphipodes supralittoraux

sympatriques de la plage de Bizerte (TUNISIE): Talitrus saltator& Talorchestia deshayesii. Bull. Soc. Zool. Fr. 133:159–167.

Benson JA, Lewis RD. (1976). An analysis of the locomotor activityrhythm of the sand beach amphipod, Talorchestia quoyana.J. Comp. Physiol. 105:339–352.

Bohli D, Morgan E, Charfi-Cheikhrouha F, Nasri-Ammar K. (2006).Etude de l’activité locomotrice de Talitrus saltator (Crustacea,Amphipoda) issu de la plage de Zouarâa (Tunisie). In Proceedingsof Final Conference of MEDCORE Project, Florence, 17–21 Novem-ber 2005. Firenze University Press, pp. 209–219.

Bregazzi PK, Naylor E. (1972). The locomotor activity rhythm ofTalitrus saltator (Montagu) (Crustacea, Amphipoda). J. Exp. Biol.57:375–391.

Chiesa JJ, Aguzzi J, García JA, Sardà F, de la Iglesia H. (2010). Lightintensity determines temporal niche switching of behavioralactivity in deep water Nephrops norvegicus (Crustacea: Decapoda).J. Biol. Rhythms 25:277–287.

Costa C, Aguzzi J, Chiesa JJ, Mmgnifico G, Cascione D, Capriolli R.(2008). Evidences on the transient disruption of Sabella spallanza-nii (Polychaeta: Sabellidae) endogenous fan activity rhythm asresult of stress of capture and acclimation to laboratory.Ital. J. Zool. 75:337–344.

Cummings SM,Morgan E. (2001). Time-keeping system of the eel poutZoarces viviparous. Chronobiol. Int. 18:1–9.

Enright JT. (1963). The tidal rhythm of activity of a sand-beach amphi-pod. Z. Vergl. Physiol. 46:276–313.

Enright JT. (1965). Entrainment of a tidal rhythm. Science 147:864–867.Enright JT. (1972). Avirtuoso isopod: circalunar rhythms and their tidal

fine structure. J. Comp. Physiol. 77:141–62.Fallaci M, Colombini I, Taiti, S, Chelazzi L. (1996). Environmental

factors influencing the surface activity and zonation of Tyloseuropaeus (Crustacea: Oniscidea) on a Tyrrhenian sandy beach.Mar. Biol. 125:751–763.

Fincham AA. (1970a). Amphipods in the surf plankton. J. Mar. Biol.Assoc. U.K. 50:177–198.

Fincham AA. (1970b). Rhythmic behaviour of the intertidal amphipodBathyporeia pelagica. J. Mar. Biol. Assoc. U.K. 50:1057–1068.

Harris JG, Morgan E. (1983). Estimates of significance in periodogramanalyses of damped oscillations in a biological time series. Behav.Anal. Lett. 3:221–230.

Harris JG, Morgan E. (1984). Entrainment of the circatidal rhythm ofthe estuarine amphipod Corophium volutator (Pallas) to non-tidal cycles of inundation and exposure in the laboratory. J. Exp.Biol. Ecol. 80:235–245

Hastings MH. (1981). The entraining effect of turbulence on the circa-tidal activity rhythm and its semilunar modulation. J. Mar. Biol.Assoc. U.K. 61:151–160.

Hayes WB. (1969). Ecological studies on the high beach Isopod Tylospunctatus Holmes and Gay. PhD dissertation, University of Califor-nia, San Diego, California.

HolmstromWF, Morgan E. (1983). Variation in the naturally occurringrhythm of the estuarine amphipod Corophium volutator (Pallas).J. Mar. Biol. Assoc. U.K. 63:833–850.

Honegger HW. (1973). Rhythmic activity responses of the fiddler crabUca crenulata to artificial tides and artificial light. Mar. Biol.21:169–202.

Imafuku M. (1976). On the nocturnal behaviour of Tylos granulatusMiers (Crustacea, Isopoda). Publs. Seto Mar. Biol. Lab. 23:299–340

Ingénierie de l’Hydraulique, de l’Equipement et de l’Environnement(IHEE). (2008). Etude de la vulnérabilité environnementale etsocio-économique du littoral tunisien face à une élévation accéléréedes niveaux de la mer due aux changements climatiques et identifi-cation d’une stratégie d’adaptation. Direction générale de l’envir-onnement et de la qualité de vie. Tunisia: GEF/UNDP, 58 pp.

Jaramillo E, Contreras H, Duarte C, Avellanal MH. (2003). Locomotoractivity and zonation of upper shore arthropods in a sandy beachof north central Chile. Estuar. Coast Shelf S. 58:177–197

Locomotor Activity Rhythm of the Isopod T. europaeus

© Informa Healthcare USA, Inc.

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Con

nect

icut

on

01/0

1/14

For

pers

onal

use

onl

y.

Jones MDR, Ford MG, Gillett JD. (1966). Light-on and light-off effectson the circadian flight activity of the mosquito Anopheles gambiae.Nature 211:871–872.

Kennedy F, Naylor E, Jaramillo E. (2000). Ontogenetic differences inthe circadian locomotor activity rhythm of talitrid amphipod crus-tacean Orchestoidea tuberculata. Mar. Biol. 137:511–517.

Kensley B. (1974). Aspects of the biology and ecology of the genus TylosLatrielle. Ann. S. Afr. Mus. 65:401–471.

Klapow LA. (1972). Natural and artificial re-phasing of a tidal rhythm.J. Comp. Physiol. 79:233–258.

Marsh BA, Branch G M. (1979). Circadian and circatidal rhythms ofoxygen consumption in the sandy-beach Isopod Tylos granulatusKrauss. J. Exp. Mar. Biol. Ecol. 37:77–89.

Morgan E. (1965). The activity rhythm of the Amphipod Corophiumvolutator (Pallas) and its possible relationship to change in hydro-static pressure associated with the tides. J. Anim. Ecol. 34:731–746.

Morgan E, Harris GJ. (1986). The role of tidal activity rhythms in themigrations of an estuarine amphipod. In Queinnec Y, Devolvé N(eds.). Behavioural rhythms. Toulouse: Privat, I.E.C, 45–53.

Nardi M, Morgan E, Scapini F. (2003). Seasonal variation in the free-running period in two Talitrus saltator populations from Italianbeaches differing in morphodynamics and human disturbance.Estuar. Coast Shelf S. 58:199–206.

Nasri-Ammar K, Morgan E. (2005). Variation saisonnière du rythme del’activité locomotrice de Talitrus saltator issu de la plage de Korba(Cap Bon, Tunisie). Bull. Soc. Zool. Fr. 130:19–29.

Nasri-Ammar K, Morgan E. (2006). Seasonality of the endogenousactivity rhythm in Talitrus saltator (Montagu) from a sandy beachin northeastern Tunisia. Biol. Rhythm Res. 37:479–488.

Naylor E. (1958). Tidal and diurnal rhythms of locomotory activity inCarcinus maenas (L.). J. Exp. Biol. 35:602–610.

Page TL, Larimer JL. (1972). Entrainment of the circadian locomotoractivity rhythm in crayfish. The role of the eyes and caudal photo-receptor. J. Comp. Physiol. 78:107–120.

Petpiroon S, Morgan E. (1983). Observations on the tidal activityrhythm of the periwinkle Littorina nigrolineata (Gray). Mar.Fresh. Behav. Physiol. 9:171–192.

Portaluppi F, Smolensky MH, Touitou Y. (2010). Ethics and methodsfor biological rhythm research on animals and human beings.Chronobiol. Int. 27:1911–1929.

Reid DG, Naylor E. (1985). Free-running endogenous semilunar rhyth-micity in a marine isopod crustacean. J. Mar. Biol. Assoc. U.K.65:85–91.

Reid DG, Naylor E. (1986). An entrainment model for semilunar rhyth-mic swimming behaviour in the marine isopod Eurydice pulchraLeach. J. Exp. Mar. Biol. Ecol. 100:25–35.

Rossano C. (2004). Studio degli orologi biologici in anfipodi. Attivitàlocomotoria ed orientamento. PhD dissertation, Università degliStudi di Firenze, Florence, Italy.

Rossano C, Scapini F. (2006). Evidence of endogenous circadianactivity rhythms of talitrids from different environments. In Pro-ceedings of Final Conference of MEDCORE Project, Florence,17–21 November 2005. Firenze University Press, pp. 199–208.

Scapini F, Audoglio M, Chelazzi L, Colombini I, Fallaci M. (1997).Astronomical, landscape and climatic factors influencing orientedmovements of Talitrus saltator in nature. Mar. Biol. 128:63–72.

Scapini F, Rossano C, Marchetti GM, Morgan E. (2005). The role of thebiological clock in the sun compass orientation of free-runningindividuals of Talitrus saltator. Anim. Behav. 69:835–843.

Wildish DJ. (1970). Locomotor activity rhythms in some littoralOrchestia (Crustacea: Amphipoda). J. Mar. Biol. Assoc. (U.K.)50:241–252.

Williams JA. (1979). A semi-lunar rhythm of locomotor activity andmoult synchrony in the sand-beach amphipod Talitrus saltator.Proceeding of the 13th European Marine Biology Symposium27 Sep–04 oct 1978, Isel of Man, Pergamon Press, 13:407–414.

Williams JA. (1980a). Environmental Influence on the locomotoractivity rhythm of Talitrus saltator (Crustacea, Amphipoda). Mar.Biol. 57:7–16.

Williams JA. (1980b). The effect of dusk and dawn on the locomotoractivity rhythm of Talitrus saltator (Montagu) (Crustacea, Amphi-poda). J. Mar. Biol. Assoc. U.K. 42:285–297.

Williams JA. (1980c). The light-response rhythm and seasonalentrainment of the endogenous circadian locomotor rhythm of Tali-trus saltator (Crustacea, Amphipoda). J. Mar. Biol. Assoc. U.K.60:773–785.

Williams JA. (1982). Environmental influence on the locomotor activityrhythm of the sand shore amphipod, Talorchestia deshayesi. Mar.Biol. 69:65–71.

Williams JA. (1983). The endogenous locomotor activity rhythm of foursupralittoral peracarid crustaceans. J. Mar. Biol. Assoc. U.K.63:481–492.

Williams JA. (1995). Burrow-zone distribution of the supralittoralamphipod Talitrus saltator on Derbyhaven Beach, Isle of Man—apossible mechanism for regulating desiccation stress? J. Crust.Biol. 5:466–475.

Williamson DI. (1951). Studies on the biology of Talitridae (Crustacea:Amphipoda): Visual orientation in Talitrus saltator. J. Mar. Biol.Assoc. U.K. 30:91–99.

Yannicelli B, Palacio R, Giménez L. (2000). Activity rhythms of two cir-olanid isopods from an exposed microtidal sandy beach inUruguay. Mar. Biol. 138:187–197.

D. Bohli-Abderrazak et al.

Chronobiology International

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Con

nect

icut

on

01/0

1/14

For

pers

onal

use

onl

y.