EFFECT OF DIFFERENT FORMULATED DIETS AND REARING CONDITIONS ON GROWTH PARAMETERS IN THE SEA URCHIN PARACENTROTUS LIVIDUS

(1)

HAL Id: hal-01768545

https://hal.archives-ouvertes.fr/hal-01768545

Submitted on 17 Apr 2018

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

EFFECT OF DIFFERENT FORMULATED DIETS AND REARING CONDITIONS ON GROWTH

PARAMETERS IN THE SEA URCHIN PARACENTROTUS LIVIDUS

Catherine Fernandez, Gérard Pergent

To cite this version:

Catherine Fernandez, Gérard Pergent. EFFECT OF DIFFERENT FORMULATED DIETS AND

REARING CONDITIONS ON GROWTH PARAMETERS IN THE SEA URCHIN PARACENTRO-

TUS LIVIDUS. Journal of Shellfish Research, National Shellfisheries Association, 1998, 17, pp.1571 -

1581. �hal-01768545�

(2)

Journal of Shellfish Research, Vol. 17, No . 5, 1571-1581, 1998 .

EFFECT OF DIFFERENT FORMULATED DIETS AND REARING CONDITIONS ON GROWTH

PARAMETERS IN THE SEA URCHIN PA RACENTROTUS LIVID US

CATHERINE FERNANDEZ 1,2 AND GERARD PERGENT 1 1

EQEL Faculty of Sciences

University of Corsica, BP 52 . F-20250 Corte, France 2 CRITT Corse Technologie BP 111

F-20250 Corte, France

ABSTRACT World annual production of sea urchins has increased regularly since 1980, and many countries, such as France, are currently confronted with the problem of overexploitation of stocks . For this reason, aquaculture of the sea urchin is a possible solution for the future . The study undertaken here aimed to determine physiological responses (growth) ofParacentrotus lividusgiven three different feed types varying in quality (vegetable or animal) and biochemical composition and under variable environmental conditions (temperature, light) . Experiments in land-based tanks revealed that growth was greater when individuals were fed a mixed formulated feed containing both animal meal and plant meal or containing only animal meal, than when fed a feed containing only plant meal or a natural food, the seaweedCymodocea nodosa .Maintaining a temperature of20°Cand rearing the sea urchins in the darkness did not have any noticeable effect on growth rate . The growth model established with data obtained from the tank studies for individuals fed the mixed feed indicated that the growth of the sea urchin can be maximized by appropriate feeding . Finally, the different treatments did not bring about any differences in the relative importance of the various body parts (gonads, gut, test, lantern) although the gonadal index in reared sea urchins was very high, even for small individuals (20-25 mm) .

KEY WORDS : Sea urchin, aquaculture, growth, formulated feeds, Paracentrotus lividus

The human consumption of a variety of a sea urchin species dates back to ancient times (Giot 1970, Le Direac'h 1987) . In France, the species that is the most widely consumed is Paracen- trotus lividus (Lamarck, 1816) . The fisheries for this species is extensive and highly organized (Le Direac'h 1987) . Since the col- lapse of the French sea urchin fisheries (Le Direac'h 1987, Le Gall and Bucaille 1987), the shortage has been met by increased Irish and Spanish imports (Birais and Le Gall 1986, Kelly et al . 1998) . In light of this demand, and fueled by the high market value of the sea urchin roe, sea urchin farming has developed in France (Le Gall and Bucaille 1987, Fernandez et al . 1995, Fernandez 1996) . The development of improved sea urchin rearing techniques re- quires finding an adapted feed and rearing system as well as a thorough understanding of the organism's growth processes .

In the present study, the technology chosen to develop sea urchin aquaculture is land-based rearing with the use of formulated diets . Although more expensive, such land-based systems are interesting, because they offer the possibility of controlling environmental conditions and may be the only approach possible in areas of limited access to sea-based resources . The advantages of using formulated feeds include the elimination of problems linked to food availability and nutritional content observed for natural food sources (marine plants), the limitation of environmental impacts associated with the utilization of these natural food sources, and the facilitation of storage and transport .

To date, most studies dealing with formulated diets have examined the effect of these diets on either gonad production (De Jong-Westman et al. 1995, Lawrence et al . 1997, McBride et al . 1997, Goebel and Barker 1998) or nutrition (for review see Klinger et al. 1998) . Studies concerning the effect of formulated diets on growth are less numerous . Most of these examine sea urchin production as a function of different feed . The results generated generally reveal a greater growth with formulated diets than with their

INTRODUCTION natural counterparts (Lawrence et al . 1992, Fernandez and Calt- agitone 1994, Williams and Harris 1998) .

In addition to food quantity and quality, growth of echinoids depends upon several abiotic factors . Two of these are temperature and photoperiod (Kenner 1992, Guillou and Michel 1994, Pearse et al . 1995) . In temperate waters, growth is seasonal and is characterized by an increase in growth during spring and summer (high temperature) and a decrease in growth in autumn and winter (low temperature) (Guillou and Michel 1994) . For Paracentrotus lividus, the optimal temperature seems to be between 18 to 23°C (Le Gall et al . 1990, Grosjean et al . 1998) . Photoperiod seems to influence partitioning between somatic and gonad growth (Pearse et al . 1986, Beyer et al . 1998) .

The aim of the present study is to compare growth parameters in Paracentrotus lividus when this species is maintained under rearing conditions for long periods of time (6 months) and fed different types of formulated feed . The effect of various rearing conditions, such as light and temperature, is also examined .

1571

MATERIALS AND METHODS Collection of Echinoids

The experiments were conducted on the shore of the Urbinu lagoon (Corsica, Mediterranean) . The site is characterized by a large population of Paracentrotuslividus(Fernandez and Caltagi- rone 1990), which can be used as a source of sea urchins for the farming scheme. For the purposes of this investigation, all sea urchins were collected from this lagoon, on pebbly bottoms (Fernandez and Boudouresque 1997), or produced in an experimental laboratory (for the smallest size class) . Five size classes were used ; 3 to 9 mm, 10 to 15 mm, 20 to 25 mm, 30 to 35 mm, and 40 to 45 mm (test diameter at the ambitus without the spines) . After a 5-day acclimation period, the sea urchins were separated into groups (two replicates of 30 sea urchins per treatment except for the 3 to 9 mm size class, for which only one group was studied)

(3)

1572 FERNANDEZ AND PERGENT using plastic baskets with a mesh of 1 to 0 .5 mm . The baskets were

suspended 1 cm above the bottom of the tanks to prevent the sea urchins from reingesting their feces .

Experimental Set-up

The experimental set-up consisted of a rearing apparatus mea- suring 9 m x 4 m and housing three series of three superimposed tanks (3 m x 0 .15 m x 0 .5 m) (see Fig . 2) . For each tank series, the force of gravity was used to circulate the water among the three

tanks . The resulting cascade oxygenated the water . For two of these tank series, water flowed in an open circuit to the lagoon . The water turnover was 10 min . Experimental water temperature varied with the season (from 9 to 29°C, Fig . 1) and did not differ from lagoon water temperature . For the last tank series, the water flowed through a semiclosed circuit with a 50% new water turnover every 24 h . In this last tank series, the water temperature was maintained at 18 to 20°C year around .

Rearing Conditions

Figure 2 . A. Photograph of two series of superimposed tanks (open circuit), with water circulation . B . Photograph of plastic suspended baskets in which sea urchins are placed (30 individuals per basket) .

ture to the binder (gelatin previously dissolved in water, 12 .5%

solution) . The paste thus obtained was spread onto a plaque at a thickness of 1 .5 cm . Once hardened, this food plaque was cut into cubes of 1 cm x 1 cm x 1 .5 cm . Sea urchins were always provided surplus food .

Each tank was assigned particular environmental conditions and a specific diet . The exact characteristics of the different tanks are provided in Table 2 . For increased clarity, the tanks were given a three letter code . Each letter represents one of the studied treatments . The first letter corresponded to the nature of the food dis- tributed : V = vegetable feed ; M = mixed formulated feed ; A = animal formulated feed ; C = Cymodocea nodosa fronds . The second letter indicated the rearing temperature : R = temperature maintained at 18 to 20°C (in the semiclosed circuit) ; and N = natural temperature (in the open circuit) . The third letter corresponded to the light conditions : L = natural light; and D = total darkness .

Growth Parameters

Rearing began on February 1, 1994 . The experiment lasted 6 months . Each month the test diameter (without the spines) of all the individuals was measured using a sliding caliper . At the end of the rearing period (August 1994), 10 sea urchins of the 40 to 45, 30 to 35, and 20 to 25 mm size classes were sampled from each of the tanks . These sea urchins were measured and weighed . They were then dissected into various components : gonads, gut (without Figure 1 . Evolution of water temperature within the tanks (open and

semiclosed circuits) throughout the 6-month rearing period . Sea urchin growth was studied as a function of three rearing parameters : (1) temperature (ambient or maintained at 18-20°C) ; (2) light (natural or darkness) ; and (3) feed (three formulated feeds and one natural food source) . The natural food source was made up of living fronds of the seaweed Cymodocea nodosa (Ucria) Ash- erson sampled from the lagoon ; whereas, the three formulated feeds were prepared (Table 1) . The first feed consisted of vegetable meal and vegetable oils . This food is rich in soluble carbohydrates (58%) and was referred to as "vegetable feed ." The second feed consisted of fish meal and vegetable meal in equal quantities mixed with fish oil and vegetable oil and was referred to as

"mixed feed ." Its biochemical composition is a mixture of soluble proteins (29%) and soluble carbohydrates (35%) . The third feed consisted of fish meal and fish oil, was rich in soluble proteins (47%) and was referred to as "animal feed ." The level of soluble proteins, carbohydrates, and lipids in the feeds was determined using the techniques of Lowry et al . (1951), Dubois et al . (1956), and Folch et al . (1957), respectively . Ash was measured by placing the dried tissues in a muffle furnace at 500°C for 4 h . The energetic values were calculated by multiplying the level of each organic constituent by its energy equivalent (Brody 1945) . Feeds were prepared by progressively adding the meal, oil, and vitamin mix-

(4)

TABLE 1 .

Ingredients used in the preparation of the formulated feeds (expressed as a percentage) .

The formulated feed contain 56% of base meal mixed into a 12 .5% gelatin solution . Biochemical composition is also expressed as a percentage of dry weight . Energetic levels are expressed in kJ .g-1

dry weight

° The meal was produced by Norsildmel© and is made up of fresh cold . water Norwegian fish (herring, capelin, mackerel) .

e

The mixture is made up of (expressed in mg or UI/kg of feed)

: tocopherol acetate: 70 .8 UI, ascorbic acid : 283 mg; thiamin : 7 .1 mg; riboflavin : 7 .6 mg ; pyridoxine : 9 .4 mg ; cyanobalanine : 0 .014 mg ; biotine : 0 .47 mg ; folic acid : 1 .89 mg; calcium pantothenate : 23 .6 mg ; vit . A : 710 FT ; vit D3 : 700 UI; niacin : 14.6 mg ; CaCO3 : 2 .1 mg ; Cu SO4 : 9 .4 ; Fe SO4 : 4.7 mg ; NaF : 7 .1 mg ; Mg C03 : 174 mg; Mn SO4 : 18 .9 mg ; CaHPO4 : 75 .5 mg ; Zn SO4 : 7 .7 mg .

gut contents), test (without Aristotle's lantern), Aristotle's lantern, and gut contents . Each component was drained on filter paper and weighed to the nearest 0 .1 mg . The relation between components was then c alculated . CI

(component index) is the ratio between the

wet weight of the component (WWC) expressed in mg and the total wet weight of the sea urchin (TWW) expressed in mg : CI (%)

= WWC x 100/TWW (Lawrence et al . 1965) . Statistics

The growth for each treatment and size class was analyzed using a simple linear regression (Zar 1984) . The goodness of fit was tested by the Chi square test or linearity test . Differences in the rate of growth were demonstrated using an analysis of covariance followed by a multiple comparisons among slopes (Zar 1984) . The comparisons of mean sizes and physiological indices as a function of treatment were processed by one-way or two-way analysis of variance (ANOVA) followed by Tukey tests (Zar 1984) . Previ- ously, normality and homoscedasticity were verified by Kolmog- orov-Smirnov and Bartlett tests, respectively (Zar 1984) . The soft- ware Statgraphics Plus (v . 2.1) for Windows was used. For those treatments where the five size classes were present, the growth equations given by von Bertalanffy (1938) were used. The model parameters were estimated using the method of Gulland and Holt (1959) as modified by Lockwood (1974) . This linear method was chosen, because it allows growth measurements to be obtained without having to determine the age of the individuals .

RESULTS

Rearing mortality was fairly low (7 .5%) throughout the 6 months of the experiment . These mortalities can be attributed to a

EFFECT OF DIET AND REARING ON P . LIVIDUS GROWTH 1 57 3

temporary (1 day) decrease in the flow of incoming water in one tank (CNL) which brought about an oxygen shortage .

Growth Rate

The monthly growth rates in terms of test diameter for the eight diets and rearing treatments and five size classes are given in Table 3 . For the calculation of mean monthly growth, the data for each of the paired basket were pooled (60 sea urchins/treatment) . In- deed, for each treatment and size class, the paired sets exhibited an identical mean size and weight at each of the data acquisition periods (Student's t-tests, t = 0 .4 to 1 .6, df = 46 to 58, p > .05) . The results revealed that growth varied as a function of the treat- ment, time of year and sea urchin size [ANOVA and analysis of covariance (ANCOVA) p < .05] .

The evolution in mean size in the eight tanks exhibited a linear tendency (Table 4) (growth curves are usually logarithmic, but over relatively short periods of time (6 months) the curve takes on a linear appearance) . The statistical analyses revealed that growth was significantly linear, and this was regardless of the treatment and size class examined (testing for linearity, F = 0 .08 to 3 .80, p

> .05) . In addition, the slopes were all positive (testing for slope, F = 19 .1 to 737, p < .05) . The rate of growth for each size class, as represented by the slopes of the linear regressions, varied ac- cording to treatment (ANCOVA, F = 9 .8 to 35 .1, p < .05) . The results of the multiple comparison among slopes test performed a posteriori are presented in Table 4 . Generally, the slowest growth was obtained in tanks CNL, VRL, and VNL (vegetable feed) ; whereas, the fastest growth occurred in tanks MNL and ANL (animal or mixed feed) .

Two environmental parameters were examined ; namely, tem- perature and light . Regardless of size class and diet, growth rates were lower in the semiclosed circuit (18-20°C) than in the open circuit at ambient environmental temperatures (slope comparison test, p < .05) . At ambient environmental temperatures, the lowest growth was usually observed in winter (February) ; whereas, the highest growth occurred in June (ANOVA, F = 11 .2 to 45 .6, p <

.05) . The general tendency seems to be toward an increase in growth when temperature is in the vicinity of 20°C . To demon- strate this result, a correlation was made between growth rate and the difference between tank temperatures and the optimal growth temperature in Paracentrotus lividus (20°C) . The results obtained for each treatment reveal that only half of the relationships estab- lished are significant (r > 0 .90) . Statistical analysis of the data concerning the effects of light revealed that total darkness did not result in any significant differences in growth as compared to light conditions (slope comparison test, p > .05 for all size classes) .

The results also indicated that a significant negative correlation existed between total growth and the initial mean size of the in- dividuals (all treatments considered) (y = -0 .156x + 8 .132, r = 0 .90, n = 30 ; testing of correlation coefficient, p < .05) . The ANCOVA performed, for each treatment, among the different size classes indicated that the 40 to 45 mm size class exhibited the lowest growth regardless of treatment . The next two size classes, in increasing order of growth, were 30 to 35 mm and 20 to 25 mm . The rate of growth for the 3 to 10 mm and 10 to 15 mm size classes were identical and were the highest recorded of the five size classes considered .

Growth Modeling

The modeling of Paracentrotus lividus growth under rearing

conditions was performed using the data generated by providing

(5)

the mixed formulated feed (by using the five size classes, we can represent the different size classes present in the lagoon) . Two models were drawn up : the first makes use of the data obtained at the regulated temperature (semiclosed circuit, data from tanks MRD and MRL were pooled) . The second model was drawn up using the data obtained at environmental temperatures (open circuit, data from tanks MND and MNL) . The method described by Gulland and Holt (1959), applied to the five size classes for each of the temperature treatments, allowed the parameters of two von Bertalanffy equations to be calculated (Fig . 3) ; (1) for those individuals reared at the regulated temperature : k = 0 .0245 ; L,, = 52.355 ; to = -2 .409 ; (2) for sea urchins maintained at environmental temperatures : k = 0 .0252; L 0 = 59 .480 ; to = -2 .053 .

The validity of the models, drawn up using the data generated by the rearing experiment, was tested by performing a linear regression between the experimental and theoretic data (these last being calculated from the models' equations) . The correlation and linearity were significant (r = 0 .99 and linearity test, p > .05 ; for both models) . In addition, the slopes and linear regressions obtained were both not significantly different from 1 (comparison test between the slope obtained and a theoretic slope of 1, p > .05 in both cases) . It can, therefore, be concluded that the two theoretic models are in accordance with the data generated experimentally . Physiological Indices

The relative importance of the various components of the individuals taken from the eight tanks is represented in Figure 4 . Statistical analysis of the gonadal index reveals that important

variations in this index occurred as a function of both "diet- rearing" treatment and size class . In addition, a significant interaction existed between these two factors (two-way ANOVA, "diet- rearing" factor : F = 33 .00, p < .001 ; size factor : F = 18 .16, p

< .001 ; interaction, F = 4 .68, p < .001) . This significant interaction signifies that the evolution in gonadal index as a function of size differs depending upon the "diet-rearing" regime. The gonadal indices for the 20 to 25 mm size class were very high and were even above those of the larger individuals (in particular in tanks MNL and MND ; one-way ANOVA, F = 4 .1 and 7 .7, respectively ; and Tukey tests, p < .05) . Conversely, in the CNL tank (where individuals were fed Cymodocea nodosa), the gonadal index/sea urchin size ratio was identical to that observed in sea urchins living in the lagoon . The differences observed among the different rearing tanks allows three distinct groups to be distinguished . The first was composed of individuals fed Cymodocea nodosa, the second was individuals reared in tanks VRL, MRL, and MRD (regulated temperature), and the third group (having the highest gonadal index) contained those individuals reared in the tanks VNL, MND, ANL, and MNL (environmental temperatures) ; (one-way ANOVA, F = 8 .8 to 21 .6 ; and Tukey tests, p < .05) . The relative importance of the gut, test, and lantern in the reared sea urchins did not differ significantly as a function of the "diet-rearing" treatment administered (two-way ANOVA, "diet-rearing" factor : F = 1 .38, 1 .34, and 1 .47 respectively, p > .05 in all three cases) . Conversely, a significant difference in these three components was observed as a function of size class (size class factor : F = 29 .99, 43 .01, and 55 .67 respectively, p < .001) . The interaction between these two

1574 FERNANDEZ AND PERGENT

TABLE 2.

Rearing conditions in the different land-based tanks .

(6)

factors ("diet-rearing" and size class) was not significant, indicat- ing that the variations observed among the different size classes were identical for each of the tanks . Individuals of the 40 to 45 mm size class possessed a gut index that was lower than that observed for the two other size classes . The lowest test index was observed for the 40 to 45 mm size class, followed by the 30 to 35 size class and, finally, the highest value was seen for the 20 to 25 mm size class (Tukey test, p < .05) . The lowest lantern index was observed for the 40 to 45 size class followed by the 30 to 35 mm size class, and, finally, the 20 to 25 mm size class (Tukey tests, p < .05) .

DISCUSSION Effect of Food Type on Growth

Statistical analysis of all of the size classes revealed that food type had a significant effect on the rate of growth . The highest growth rate was obtained using the mixed and animal feeds . The growth observed was higher than has been found by a number of authors (Allain 1978, Le Gall and Bucaille 1987, Azzolina 1988, Turon et al . 1995), particularly for the 30 to 35 mm size class . A number of studies have demonstrated that growth rates in the Echi- noidea differ with the food type available and/or ingested (quality

EFFECT OF DIET AND REARING ON P . LIVID US GROWTH

TABLE 3 .

Mean monthly growth rate (± standard error) of reared Paracentrotus lividus (in mm/month) for the eight treatments and five size classes . 1575

Tank codes, the letter corresponds to the food : V : vegetable feed; M : mixed feed; A: animal feed ; C : Cymodocea nodosa. The second letter indicates the rearing temperature ; R : temperature maintained at 18 to 20°C ; N : natural temperature. The third letter corresponds to the light conditions : L : natural light ; D : total darkness .

and quantity) (Lawrence 1975, Vadas 1977, Rowley 1990, Frantzis and Grémare 1992) . Although Cymodocea nodosais considered to be the preferred food source ofParacentrotus lividus(Traer 1980), the use of this food type in the present study did not yield maximal growth . It should be noted, however, that the decrease in oxygen levels (for 1 day) in the CNL tank may have led to a temporary decrease in growth rate and, thus, lower the total growth observed during the 6-month rearing period . The increase in sea urchin growth rates when fed formulated feeds (mixed or animal) as compared to the growth observed on a diet made up of Cymodocea nodosa has been observed in a previous study using similar rearing conditions (Fernandez and Caltagirone 1994, Fernandez 1996).

The present study, therefore, confirms that the quality of the mixed and animal formulated feeds allows growth to be maximized . This food quality is fairly difficult to assess, because it encompasses a number of characteristics, such as quantity, digestibility, absorption, and composition of the food (Lawrence and Lane 1982) . A possible first observation concerns somatic growth, which was always higher for these two feeds, the biochemical composition of which was characterized by a high level of soluble proteins . In fact, it has been demonstrated that the growth of certain Echi- noidea depends upon the levels of ingested protein (Lowe and

(7)

For all size classes n = 14, except for the 3 to 9 mm class size, where n = 7 . b: the slope [the results of the multiple comparison among slopes (for each size class) is given by an exponent letter : values that do not differ at the 0.05 level are noted with a same letter] ; a : intercept, r: correlation coefficient.

Tank codes, the letter corresponds to the food : V : vegetable feed; M : mixed feed; A : animal feed;C : Cymodocea nodosa .The second letter indicates the rearing temperature : R : temperature maintained at 18 to 20°C, N : natural temperature . The third letter corresponds to the light conditions : L : natural light, D : total darkness .

Lawrence 1976, Frantzis and Grémare 1992) . The results presented here would, therefore, seem to be in agreement with these studies . They also reveal that formulated feeds made up of fish meal represent good food sources for Paracentrotus lividus . It should also be noted that an increase in the percentage of fish meal in the feed (which causes an increase in the level of protein) does not, in the present study, bring with it a significant increase in somatic growth. Part of the energy used must, therefore, be lost or used to other ends (reproduction, building of reserves) .

Effect of Type Water System (Open Circuit/Semiclosed Circuit)

An attempt was made to compare growth rates obtained in the open circuit (natural environmental temperatures) with those obtained in the semiclosed circuit (regulated temperature) . It is ob- vious that such a comparison between two very different water systems is only of value as a simple observation . Nevertheless, the results obtained are interesting . Indeed, these results (lower growth rates at the regulated than at nonregulated temperatures) are in contradiction with pre-existing data on Paracentrotus lividus . In- deed, Le Gall et al. (1990) suggested that the growth of_Paracen- trotus lividuswas at a maximum at 18 to 20°C . In addition, it has been demonstrated that maximal growth of Paracentrotus lividus in Urbinu lagoon occurs when water temperature is also approxi- mately 18 to 20°C (Fernandez 1996) . It can, therefore, be hypoth- esized that factors other than temperature affected sea urchin growth . Under rearing conditions, the pollution generated by the

TABLE 4.

Comparison of equation parameters describing the increase in mean test size (y, in mm) as a function of rearing time (x, in days) for sea urchins based on their initial size and rearing treatment .

rearing itself (disintegration of food, biodegradation of this disin- tegrated food, and urchin metabolic waste) is a parameter that would seem to be of some importance . In the present study, the water renewal in the semiclosed circuit was voluntarily low in order to lower the energy costs associated with regulating the rearing temperature, and a decanting tank was the only means used to eliminate a portion of the suspended matter . As a result, nitrite peaks were sometimes observed (>0 .2 mg/L), the level of which were above the level generally considered detrimental to marine invertebrates (<O .1 mg/L, Alzieu 1989) . The accumulation of rearing waste products (ammonia and nitrite) are toxic to the animals and increases their oxygen demand (Gonzalez et al . 1993) and, thus, represent stress factors capable of inhibiting growth . Grosjean et al . (1996) have observed that, under rearing conditions, a number of individuals exhibit very low growth rates (usually generating a multimodal size frequency distribution) and concluded that the lower productivity of the small individuals is not genetically induced but rather the result of poor water quality . The low growth observed in the semiclosed circuit may, therefore, be partially attributable to the poor water quality present in these tanks .

Effect of Variations in Water Temperature on Growth in the Open Circuit of Temperature

The growth of sea urchins in the open circuit differs substan- tially between months (Table 3) . Water temperature in the open

(8)

circuit presents the same seasonal variations as those observed in situ in the lagoon (from 9-29°C, Fernandez 1996) . A number of authors have noted the importance of temperature on the growth of Echinoidea (Le Gall et al . 1990, Lares and McClintock 1991, Guillou and Michel 1994) . The rearing of Paracentrotus lividus in the natural environment (immerged cases in situ) has demonstrated that growth is dependent on temperature (Fernandez 1996) with a general tendency that seems to be toward an increase in growth when temperature is in the vicinity of 20°C . The high proportion of nonsisnificant relationships observed (between growth and temperature) may be attributable to the low number of datapoints available (n = 5) . In addition, it should be noted that these correlations have been drawn after eliminating the growth value for the month of May (for all tanks and size classes) . These growth values are very low when the temperature is 19 .1 °C . It should be noted that the month of May was characterized by several massive and successive spawning events (May 3, 6, and 17) . All of the size classes were involved in these spawns and this in all of the tanks (whether in open or semiclosed circuits) . It is possible that after these spawning events, a portion of the energy ingested went toward gonadal production to replenish the gonad and gametes, and this at the expense of somatic growth . These results reveal that temperature was not the sole parameter responsible for the observed seasonal variations in growth . The reproductive cycle seems to be one of the more important parameters . Indeed, the low somatic growth observed during the spawning period would seem to be in agreement with the observations made by a number of researchers . The latter have observed an inverse relationship between gonadal and somatic growths (Vadas 1977, Greenwood 1980, Lawrence and Lane 1982, Azzolina 1988, Guillou and

EFFECT OF DIET AND REARING ON P . LIVID US GROWTH 1577

Figure 3 . Calculation of the growth models : A and B . Gulland-Holt diagrams used to determine the parameters for the von Bertalanffy model generated based on the growth data obtained for sea urchins reared at both natural and regulated temperatures. C and D. The von Bertalanffy equations produced.

Michel 1994) . In the present study, it was not possible to correlate gonadal and somatic growths, because the cycle of the gonadal index was not recorded in the tanks . Indeed, the substantial differences existing between the in situ and reared gonadal indices was such as to make an extrapolation of the latter values impos- sible to perform . The last parameter to have an effect on somatic growth is the sea urchin's ingestion levels . Indeed, this parameter varies with sampling month, and these variations are correlated with those observed for growth (Fernandez 1996, Fernandez and Boudouresque 1997) . Guillou and Michel (1994) also observed that the resumption of growth in Sphaerechinus granularis (Lamark, 1816) is attributable, at least in part, to a resumption in feeding activity (because the repletion index is correlated to temperature) . In conclusion, the variations in somatic growth observed in the tanks would seem to be linked to a number of interacting factors including temperature, reproductive cycle, and ingestion . Effect of Light on Growth

Several species of sea urchins, including Paracentrotus lividus, are very active at night (Gamble 1966, Shepherd and Boudour- esque 1979, Dance 1987) especially in so far as trophic activity is concerned (Powis de Tembossche 1978, Rico et al . 1990) . Grosjean and Jangoux (1994) have observed a higher level of ingestion and absorption in individuals of this species reared in total darkness (this implies a higher level of food available for growth) . In light of this, Le Gall (1989) has proposed rearing sea urchins in total darkness in order to increase somatic growth . Pearse et al . (1995) and Beyer et al . (1998) have also observed

(9)

1 57 8 FERNANDEZ AND PERGENT

Figure 4. Mean gonadal, gut, test, and lantern indices (±95 % confidence interval) for three size classes ofParacentrotus lividusreared in tanks and fed either formulated feeds or Cymodocea nodosa . Tank codes, the first letter corresponds to the food : V : vegetable feed; M : mixed feed;

A : animal feed ;C : Cymodocea nodosa.The second letter indicates the rearing temperature : R : temperature maintained at 18 to 20°C ; N : natural temperature. The third letter corresponds to the light conditions : L : natural light ; D : total darkness .

greater somatic growth and a low gonadal growth in Strongylo- centrotus fransiscanus (Stimpson 1857) and Strongylocentrotus purpuratus(Stimpson 1857) when placed under dark conditions as compared to individuals reared in continuous light . The results obtained here do not confirm these results : total darkness did not bring about an increase in somatic growth as compared to con-

tinuous light conditions . Dark conditions in the present study were obtained by placing an opaque black cover over the appropriate tanks (tanks MRD and MLD) . It was necessary to remove the cover twice a week for a period of about 20 minutes in order to clean the tanks . This operation was always performed at night using a low watt formulated light . Nevertheless, it is possible that these cleaning operations disrupted the experiment . Indeed, Grosjean and Jangoux (1994) observed that the sea urchin food ration was lowest when the individuals were subjected to a 12 h light/12 h dark photoperiod, and they attributed this phenomenon to the sudden passage from light to dark conditions (Grosjean pers . comm .) . It is possible that in the present study the sudden passage from dark to light conditions was disrupting and, thus, did not allow an increase in growth to be observed .

Effect of Initial Size on Growth

Growth in the Echinoidea, such asParacentrotus lividus, varies as a function of age and, therefore, size of the individuals (cf . growth models : Allain 1978, Fenaux et al . 1987, Azzolina 1988, Turon et al . 1995) . In the present study, five different size classes

were used. The results generated revealed that, for the 40 to 45 mm size class, growth is very low and is most probably not profitable in a commercial aquaculture context . Indeed, despite the fact that large sea urchins ingest important quantities of food, this does not allow for a rapid or significant size increase . Although small sea urchins are not commercialized, it would seem that the rational- ization of sea urchin aquaculture necessitates that rearing be stopped once the urchins have reached a size of 40 mm (this being the minimum size required for the commercialization of this species) .

Growth Curves

The models presented here must be considered with a certain degree of caution . Because no individuals larger than 45 mm were examined, the model is not valid beyond this size . In addition, growth was studied over a period of only 6 months (February-July

1994) . Therefore, the seasonal variations in growth have not been included in the models . Nevertheless, the models presented are quite similar to those reported by other authors (Fig . 3) and, in particular, when the von Bertalanffy model was used (Allain 1978, Azzolina 1988) . The growth curves calculated for natural temperature present a growth more rapid than those generated by other models (with the exception of the logistic model proposed by Fenaux et al . 1987), at least for the size range considered in the present study (3-45 mm) .

(10)

Physiological Indices

From an aquacultural point of view, what is important is the gonadal index . Indeed, the gonadal index usually increases with increasing sea urchin size (Fuji 1967, Nichols et al . 1985, Pearse and Cameron 1991, Lumingas 1994) and this even in lagoon en- vironments (Fernandez and Boudouresque 1997) . The results obtained in the tanks of the present study were in direct contradiction with what is reported in the literature ; namely, the gonadal indices for the 20 to 25 mm size classes were very high and even superior to those observed in larger individuals (particularly in tanks MNL and MND) . Conversely, in the CNL tank (individuals fed _Cymo- docea nodosa),the gonadal index/sea urchin size ratio was identical to that observed in wild individuals . The rearing of sea urchins in tanks using a formulated diet, therefore, causes substantial changes in gonadal growth . The energy allocated to gonad development was, therefore, more important under rearing conditions, especially in small sized individuals (20-25 mm), than was observed in the natural environment . InEvechinus chloroticus (Va- lenciennes), Barker (pers . comm .) also observed high gonadal indices in small individuals . The increase in gonad weight may, in part, be attributable to an increase in nutritive phagocytes rather than gametes (Walker 1982, Pearse and Cameron 1991) . As a general rule, once a portion of the energy has been allocated to maintenance, the energy remaining is divided between reproduction and somatic growth (Calow 1981) . In the small individuals of the present study (in particular in the MNL and MND tanks), a large amount of energy, therefore, remains available for reproduction that leads to gonads the weight of which were very high for these size classes .

In so far as the treatment differences are concerned, the results revealed that over a long time period (6 months), all of the food types brought about high gonadal growth . Finally, it should be noted that the gonadal index generated under rearing conditions was higher than that observed in wild sea urchins at the same time of year (Fernandez 1996) .

The gut, test, and lantern indices were identical in the light diet-rearing treatments . Therefore, neither the food type nor the environmental parameters seemed to affect, over a 6-month period, the partitioning of weight into these various body components . It has similarly been observed in other species that food quality does not have any effect on the size of the gut (Klinger et al . 1994 1995) . The different body compartments in the reared sea urchins contain storage material in the form of lipids and carbohydrates (Fernandez 1997) .

Allain, J . Y . 1975 . Structure des populations de Paracentrotus lividus (Lamarck) (Echinodermata, Echinoïdea) soumises à la pêche sur les côtes nord de Bretagne .Rev. Trav. Inst. Pêche Marit. 39 :171-212 . Allain,J . Y . 1978 .Age et croissance deParacentrotus lividus(Lmk.) et de

Psammechinus miliaris (Gmelin) des côtes nord de Bretagne (Echi- noïdea) .Cah. Biol. Mar. 19 :11-21 .

Alzieu, C .1989 .L'eau milieu de culture .pp . 15-43 . In:G. Barnabé (ed .) . Aquaculture 1st edit . Lavoisier Publishing, Paris .

Azzolina,J.F . 1988 . Contribution à l'étude de la dynamique des populations de l'oursin comestibleParacentrotus lividus (Lamarck) . croissance, recrutement, mortalité, migrations . Thèse Doct . 3 ème Cycle Ecol ., Univ . Aix- MarseilleII, 225 pp .

Beyer, D ., J. S .Pearse & M . E . Steele . 1998. Both photoperiod and diet influence resource partitioning between somatic and gonad growth

EFFECT OF DIET AND REARING ON P . LIVIDUS GROWTH

LITERATURE CITED

CONCLUSION

The results obtained reveal that the mixed formulated (vegetable and animal meal) and animal feeds (animal meal) allow for a higher somatic growth than do the vegetable feed types (natural food source made up ofCymodocea nodosaor formulated feed) . In addition, observed growth was similar to or higher than that calculated from other studies on this species . The formulated feed was capable of inducing a higher growth rate than was seen with natural food sources .

The growth observed under rearing conditions exhibited distinct seasonal and monthly variations . These seem to be mostly attributable to water temperature (with a slower growth when temperatures were either very low or very high) but also to the organism's reproductive cycle (in particular, growth was very low during spawning events) . Those tanks with a regulated temperature did not generate satisfactory results . Indeed, the semiclosed circuit used, which is without water renewal, produces a water quality that is insufficient to maximize growth . In particular, the level of ni- trites were often high, which increases the sea urchins energy expenditure . Future experiments at a regulated temperature will need to include a water purifying system in order to generate irreproachable water quality .

The efficiency of formulated feeds is very important with re- spect to gonadal production . These feeds can generate a substantial increase in gonadal growth that is of far greater importance than that observed in the natural environment . This growth occurs even in small-sized individuals for which the gonadal index can reach up to 25% of the sea urchin's total mass . It is, therefore, possible that sea urchins smaller than those usually eaten may be able to generate a gonadal production equivalent to that produced by larger individuals . It would, therefore, be possible to reduce rearing time required before commercialization .

ACKNOWLEDGMENTS

The authors thank C .F . Boudouresque and J .M . Lawrence for their comments on an earlier draft of this manuscript . We also thank L . Bronzini of Caraffa, manager of the SCORSA, for col- laborating with us on this project in the Urbinu lagoon . Thanks go out to D . Viale, of the Université de Corse, for the interest she has shown and J . H . Balbi, President and M . Filippi, Director of the CRITT Corse Technologie for their assistance with this program . Many thanks to Dr . M . Johnson for translating the manuscript.

This research was supported by a grant from the regional govern- ment of Corsica and the ANVAR .

1579

in sea urchin, Strongylocentrotus franciscanus . p . 577 . In : R . Mooi and M. Telford (eds .), Echinoderms : San Francisco . Balkema, Rotter- dam .

Birais D . & P . Le Gall . 1986 . Deux cents mètres de toboggan pour les oursins .Aquarevue 5 :31-35 .

Brody, S . 1945 . Bioenergetics and growth . Hafner Publ . Co . Inc ., New York. 1023 pp .

Calow,P. 1981 . Resource utilization and reproduction .pp . 245-270 . In:

Townsend C . R. and P . Calow (eds .), Physiological Ecology : An Evo- lutionary Approach to Resource Use . Blackwell Science, Oxford, UK . Dance,C . 1987 .Patterns of activity of the sea urchinParacentrotus lividus in the bay of Port-Cros (Var, France, Mediterranean) .P.S.Z.N.I Mar.

Ecol. 8 :131-142.

(11)

De Jong-Westman, M ., P . Y. Qian, B . E . March & T. H . Carefoott . 1995 . The effect of different nutrient formulation in artificial diets on gonad growth in the sea urchin Strongylocentrotus droebachiensis . Can . J.

Zool. 73 :1495-1502 .

Dubois, M ., K . A. Gilles, J. K . Hamilton, P . A . Reber & F. Smith. 1956 . Colorimetric method for determination of sugar and related substance . Analyt . Chem. 28 :350-356 .

Fenaux, L ., M . Etienne & G . Quelart. 1987 . Suivi écologique d'un peuple- ment de Paracentrotus lividus (Lamarck) dans la baie de Villefranche- sur-mer (France) . pp . 187-197 . In: C. F . Boudouresque (ed .) . Colloque International sur Paracentrotus lividus et les Oursins Comestibles . GIS Posidonie, Marseille .

Fernandez, C. 1996 . Croissance et nutrition de Paracentrotus lividus dans le cadre d'un projet aquacole avec alimentation artificielle. These Doct . Univ . Corse ., 279 p .

Fernandez, C . 1997 . Effect of diet on the biochemical composition of Paracentrotus lividus (Echinodermata : Echinoidea) under natural and rearing conditions . Comp. Biochem. Physiol . (A)118 :1377-1384 . Fernandez, C . & C . F . Boudouresque. 1997 . Phenotypic plasticity of Para-

centrotus lividus (Echinodermata : Echinoidea) in a lagoonal environment. Mar. Ecol . Prog. Ser. 152 :145-154 .

Fernandez, C . & A . Caltagirone . 1990 . Données préliminaires sur la population de Paracentrotus lividus de l'étang d'Urbinu (Corse) . Rapp . Comm. int. Mer Médit. 32B-V10 :37 .

Fernandez, C . & A. Caltagirone . 1994 . Growth rate of adult Paracentrotus lividus in a lagoon environment : the effect of different diet types . pp . 655-660 . In : B . David, A. Guille, J . P . Féral and M . Roux (eds .) . Echinoderms through Time . Balkema, Rotterdam .

Fernandez, C ., E . Dombrowski & A . Caltagirone. 1995 . Gonadic growth of adult sea urchin Paracentrotus lividus (Echinodermata : Echinoidea) in rearing : the effect of different diet type . pp . 269-275 . In : R . Emson, A . Smith and A. Campbell (eds.) . Echinoderms Research 1995 . Balkema, Rotterdam .

Folch, J ., M . Lees & S . G. H . Sloane. 1957 . A simple method for isolation and purification of total lipids from animal tissues . J. Biol. Chem . 226 :493-509 .

Frantzis, A. & A . Grémare. 1992 . Ingestion, absorption, growth rate of Paracentrotus lividus (Echinodermata : Echinoidea) fed different mac- rophytes . Mar. Ecol. Prog. Ser. 95 :169-183 .

Fuji, A . 1967 . Ecological studies on the growth and food consumption of Japanese common littoral sea urchin Strongylocentrotus intermedius (Agassiz) . Mcm. Fac . Fish., Univ . Hokkaido 15 :83-160.

Gamble, J . G . 1966. Ecological studies on Paracentrotus lividus (Lmk) . pp . 85-88 . In: J . N . Lythgoe and J . D . Woods (eds .) . Symposium of Un- derwater Ass . Rep . 1966-1967, Malta .

Giot, P . R . 1970. De l'antiquité des talus et des dunes armoricaines . Penn . ar Bed. 60 :249-256 .

Goebel, N . & M . F . Barker . 1998 . Artificial diets supplemented with ca- rotenoid pigments as feeds for sea urchins . pp . 667-672 . In: R . Mooi and M . Telford (eds .) . Echinoderms : San Francisco . Balkema, Rotter- dam .

Gonzalez, M . L ., M . C. Perez, D . A. Lopez & C . A. Pino . 1993 . Effect of algal diet on the energy available for growth of juvenile sea urchin Loxechinus albus (Molina, 1782) . Aquaculture 115 :87-95 .

Greenwood P. J. 1980 . Growth, respiration, and tentative energy budget for two population of the sea urchins Parechinus angulosus (Leske) . Es- tuar. Coast. Mar. Sci . 10 :347-367 .

Grosjean, P. & M . Jangoux . 1994. Effect of light on feeding in cultivated echinoids (Paracentrotus lividus) . p . 691 . In : B . David, A . Guille, J . P . Feral and M . Roux (eds .) . Echinoderms through Time . Balkema, Rot- terdam .

Grosjean, P ., C . Spirlet & M . Jangoux . 1996 . Experimental study of growth in the echinoid Paracentrotus lividus (Lamarck, 1816) (Echinoder- mata) . J. Exp . Mar. Biol. Ecol. 201 :173-184 .

Grosjean, P., C . Spirlet & M . Jangoux . 1998 . Closed-circuit cultivation of the edible sea urchin Paracentrotus lividus : optimization of somatic growth through the control of abiotic environment . p . 673 . In : R . Mooi

and M . Telford (eds .) . Echinoderms : San Francisco . Balkema, Rotter- dam .

Guillou M . & C . Michel . 1994 . The influence of environmental factors on the growth of Sphaerechinus granularis (Lamarck) (Echinodermata : Echinoidea) . J. Exp. Mar. Biol . Ecol. 178 :97-111 .

Gulland, J. A . & S . L. Holt. 1959 . Estimation of growth parameters for data at unequal time interval. J. Cons. Perm. Int. Explor. Mer . 25 :47-49 . Kelly, M . S ., J . D . McKenzie & C . C . Brodie . 1998 . Sea urchins in poly-

culture : the way to enhanced gonad growth? pp . 707711 . In : R. Mooi and M . Telford (eds .) . Echinoderms : San Francisco. Balkema, Rotter- dam .

Kenner M . C . 1992 . Population dynamics of the sea urchin Strogylocen- trotus purpuratus in a central California kelp forest : recruitment, mortality, growth, and diet. Mar. Biol. 112 :107-118 .

Klinger, T . S ., J . M . Lawrence & A . L . Lawrence . 1994 . Digestive characteristics of the sea urchin Lytechinus variegatus (Lamarck) (Echino- dermata : Echinoidea) fed prepared feeds. J. World Aquacult. Soc . 25 :489-496 .

Klinger, T. S ., J. M. Lawrence & A. L . Lawrence. 1998 . Digestion, absorption, and assimilation of prepared feeds by echinoids . pp . 713-721 . In : R . Mooi and M . Telford (eds .) . Echinoderms : San Francisco . Balkema, Rotterdam .

Klinger, T . S ., J . M . Lawrence, A . L. Lawrence, R. Price & R. Boudreau . 1995 . Somatic and gonadal growth of Strongylocentrotus droebachiensis (Echinodermata :Echinoidea) fed manufactured feed . Am . Zool . 109 : 433 .

Lares, M. T. & J . B . McClintock. 1991 . The effect of temperature on the survival, organismal activity, nutrition, growth, and reproduction of the carnivorous Eucidaris tribuloides . Mar. Behav . Physiol. 19 :75-96 . Lawrence, J . M . 1975 . On the relationships between marine plants and

sea-urchins . Oceanogr . Mar. Biol. Ann . Rev. 13 :213-286 .

Lawrence, J . M . & J . M . Lane. 1982 . The utilization of nutrients by post- metamorphic echinoderms . pp . 331-371 . In: M . Jangoux and J . M . Lawrence (eds .) . Echinoderm Nutrition . Balkema, Rotterdam.

Lawrence, J . M ., A. L . Lawrence & N . D . Holland . 1965 . Annual cycle in the size of the gut of the purple sea urchin, Strongylocentrotus purpuratus (Stimpson) . Nature 205 :1238-1239 .

Lawrence, J . M ., R . Otaiza, A . L. Lawrence, & E . Bustos . 1997 . Enhance- ment of gonad production in the sea urchin Loxechinus albus in Chile fed extruded feeds . J. World Aquat. Soc . 28 :91-96 .

Lawrence, J . M ., L . Fenaux, M . C . Corre & A. L. Lawrence. 1992 . The effect of quantity and quality of prepared diets on production in Para- centrotus lividus (Echinodermata : Echinoidea) . pp . 107-110 . In : L . Scalera-Liaci and C . Canicatti (eds .) . Echinoderm Research 1991 . Balkema, Rotterdam .

Le Direac'h, J . P. 1987 . La pêche des oursins en Méditerranée : historique, technique, législation, production . pp . 335-362. In : C . F . Boudou- resque (ed .) . Colloque International sur Paracentrotus lividus et les Oursins Comestibles . GIS Posidonie, Marseille .

Le Gall, P . 1989 . Echinoculture . pp . 468-491 . In: G . Barnabé (ed .) . Aqua- culture, vol . 1, 2nd ed. Lavoisier Tech and Doc, Paris .

Le Gall, P . & D . Bucaille . 1987 . Intérêt d'un élevage intensif de l'oursin violet Paracentrotus lividus. pp . 399-405 . In : C . F . Boudouresque (ed .) . Colloque International sur Paracentrotus lividus et les Oursins Comestibles . GIS Posidonie, Marseille .

Le Gall, P ., D . Bucaille & J. B . Grassin. 1990 . Influence de la température sur la croissance de deux oursins comestibles, Paracentrotus lividus et Psammechinus miliaris . pp. 183-188 . In : C . De Ridder, P. Dubois, H . Lahaye and M . Jangoux (eds .) . Echinoderm Research. Balkema . Rot- terdam .

Lockwood, S . J. 1974 . The use of the von Bertalanffy growth equation to describe the seasonal growth of fish . J. Cons. Int. Explor. Mer. 35 :

175-179 .

Lowe, E . F . & J. M . Lawrence . 1976. Absorption efficiencies of Lytechinus variegatus (Lamarck) (Echinodermata: Echinoidea) for selected marine plants . J. Exp. Mar. Biol. Ecol . 21 :223-234 .

(12)

Lowry, O . H ., N . J . Rosebrough, A . L . Farr & R . J . Randall . 1951 . Protein measurement with folin reagent . J. Biol. Chem. 193 :265-275 . Lumingas, L . J . L . 1994 . La plasticité chez l'oursin Sphaerechinus granu-

laris en rade de Brest (Bretagne, France) . Thèse Doct. Océanol . Biol ., Univ . Bretagne Occidentale, Brest. 193 pp .

McBride, S . C ., W. D . Pinnix, J. M . Lawrence, A . L . Lawrence & T . M.

Mulligan . 1997 . The effect of temperature on production of gonads by the sea urchin Strongylocentrotus franciscanus fed natural and prepared diets . J. World Aquat. Soc . 28 :357-365 .

Nichols, D ., G . M . Bishop & A . A . T. Sime. 1985 . Reproductive and nutritional periodicities in populations of European sea urchin Echinus esculentus (Echinodermata : Echinoidea) from the English channel .J.

Mar. Biel. Ass . U.K.65 :203-220 .

Pearse, J . S ., D . Beyer & M. E. Steele . 1995 . Both photoperiod and diet influence resource partitioning between somatic and gonad growth in

sea urchins (Strongylocentrotus franciscanus and Strongylocentrotus purpuratus) . Am . Zool. 35(A) :435 .

Pearse, J . S . & R . A. Cameron . 1991 . Echinodermata Echinoidea. pp . 513- 662 . In : A . C . Giese, J. S . Pearse & V . B . Pearse (eds .) . Reproduction of Marine Invertebrates, vol VI. Echinoderms and Lophophorates . The Boxwood Press, Pacific Grove, CA .

Pearse, J . S ., V . B . Pearse & K. Davis . 1986. Photoperiodic regulation of gametogenesis and growth in the sea urchin Strongylocentrotus purpu-

ratus . J. Exp . Zool. 237 :107-118 .

Petit, J . 1989 . L'approvisionement en eau, le traitement de l'eau, le recy- clage en aquaculture . pp . 45-180 . In : G . Barnabé (ed .) . Aquaculture,

1st edit . vol . 1 . Lavoisier, Paris.

Powis De Tenbossche, T . 1978 . Comportment alimentaire et structure di- gestivedeParacentrotus lividus (Lmk.) (Echinodermata, Echinoidea) . Mem. Lic. Zool ., Fac . Sci . Bruxelles . 82 pp.

EFFECT OF DIET AND REARING ON P . LIVIDUS GROWTH 1581

Rico V ., A . Caltagirone & C . Fernandez. 1990 . Comportement de Para- centrotus lividus : quelques exemples de budget temps . Rapp . Commiss.

Int. Explor. Sci . Me r Medit. 32B-V10 :38 .

Rowley, R . J . 1990. Newly settlement sea urchin in a kelp bed and urchin barren ground: a comparison of growth and mortality . Mar. Ecol. Pror.

Ser. 62 :229-240 .

Shepherd, S . A. & C . F . Boudouresque . 1979 . A preliminary note on the movement of the sea urchin Paracentrotus lividus. Trav. Sci . Pare Nation . Port-Gros 5 :155-158 .

Traer, K . 1980 . The consumption of Posidonia oceanica Delile by echinoids at the isle of Ischia . pp. 241-22<44 . In : M . Jangoux (ed .) . Echino- derm : Present and Past . Balkema, Rotterdam.

Turon, X ., G . Giribet, S . Lopez & C . Palacin. 1995 . Growth and population structure of Paracentrotus lividus (Echinodermata : Echinoidea) in two contrasting habitats . Mar. Ecol. Prog. Ser . 122 :193-204 .

Vadas, R. L. 1977 . Preferential feeding : an optimization strategy in sea urchins . Ecol . Monogr. 47 :337-371 .

von Bertalanffy, L . 1938 . A quantitative theory of organic growth (inquires of growth laws II) . Human Biology : a record of research . 10 :181-213 . Walker, C . W. 1982 . Nutrition of gametes . pp . 449-468. In: M . Jangoux and J. M . Lawrence (eds .) . Echinoderm Nutrition . Balkema, Rotter- dam .

Williams C . T . & L . G . Harris . 1998 . Growth of juvenile green sea urchins on natural and artificial diets . pp . 887-892 . In: R . Mooi and M . Telford (eds .) . Echinoderms : San Francisco. Balkema, Rotterdam .

Zar, J. H. 1984. Biostatistical analysis . 2nd ed. Prentice Hall International, Upper Saddle River, NJ . 718 pp .