Nursery culture of the abalone Haliotis laevigata: larval settlement and juvenile production using cultured algae or formulated feed.

ABSTRACT This study was conducted to investigate the settlement, growth, survival, and size variability of the abalone Haliotis laevigata on commercial scale. At settlement, two plate conditioning times for the green alga Ulvella lens (conditioned for 6 or 8 wk) were evaluated and compared with plates colonized by different diatom species (Navicula cf. jeffreyi and Cocconeis sp.). In a choice experiment an overall settlement rate of 87% was estimated 3 days after larval release. The majority of the larvae chose to settle on U. lens (61% on 8-wk-old and 14% on 6-wk-old U. lens, with average algal cover of 97% and 82% respectively). Larvae showed a clear preference for older than for younger U. lens, with similar percentage cover, indicating that the developmental stage of the alga and not percentage cover per se is important in settlement induction. Only 7% and 5% respectively of the settled larvae were found on the plates colonized by the diatom species N. cf. jeffreyi and Cocconeis sp. Results clearly demonstrate that U. lens provides a suitable substrate to improve the settlement of Haliotis laevigata larvae on commercial scale. Juvenile growth ([micro]m [day.sup.-1]) on a mixed diet of U. lens and N. cf. jeffreyi was not constant and increased with size to 70-80 [micro]m [day.sup.-1]. Both growth rate and size variability increased over time until juveniles reached approximately 5 mm in shell length. The survival declined to ca. 30% at 44 days after settlement but stabilized thereafter. Distinct increases in size variability were observed between 3 and 16 and between 30 and 44 days after settlement, followed by elevated growth rates and a change from monomodal to multimodal distribution at day 44. The latter coincided with a shift in diet from a diatom-dominated diet to a macroalgal diet of U. lens at day 44. The increase in growth rate between 3 and 16 days is believed to reflect a shift in nutrition with postlarvae switching from absorbing yolk reserves to efficient exogenous feeding. The growth, survival, and size variability of juveniles was assessed when feeding on algal diets as well as a formulated diet. Juveniles grew faster on U. lens than on a formulated feed or plates colonized with the diatom N. cf. jeffreyi. Results indicate that growth of juveniles may be more variable when feeding on natural feed (Navicula cf. jeffreyi., U. lens) than on a formulated teed. However, until a feed is formulated that can match the mean growth rates achieved on U. lens, we suggest keeping juveniles on plates colonized with U. lens as long as possible. Juvenile H. laevigata should not be weaned onto formulated feed until they reach at least 5 mm in shell length.

KEY WORDS: abalone, algae, diatoms, growth, size variability, settlement, Ulvella lens

INTRODUCTION

Australia has one of the world's largest abalone fisheries and its abalone aquaculture industry is expanding (Fleming 2000). With wild abalone fisheries declining in other countries, the interest in abalone aquaculture has increased substantially.

Since the inception of abalone aquaculture in Australia, research has focused on achieving the best growth rates possible in growout systems. This is because any increase in growth rates reduces production time and will usually have large cost benefits for an abalone farm. Recently it has been shown that differences in early growth of the abalone Haliotis rubra Leach persisted and were amplified by the end of a 4-too growth trial (Daume et al. 2004), suggesting that early growth is important in determining later performance. Despite this result and the consequent benefits to the abalone aquaculturist, very little emphasis has been placed on improving growth rates of juvenile abalone.

Laboratory experiments have shown that the macroalgae Ulvella lens Crouch is a suitable settlement inducer for larvae of the abalone Haliotis rubra Leach and H. laevigata Donovan (Daume et al. 2000). These findings were recently verified in the nursery for H. rubra (Daume et al. 2004). However, the growth rates of postlarvae (up to 3 mm in shell length) feeding on U. lens have been less promising and significantly better growth was achieved with benthic diatom species (Daume et al. 2000).

Formulated feeds have long been recognized to have beneficial effects for the growth and survival of juvenile abalone (Fleming et al. 1996). Experiments conducted in Australia showed that juvenile abalone can achieve maximum growth rates of up to 53 [micro]m [day.sup.-1], and 90 [micro]m [day.sup.-1] for animals ranging between 3 and 18 mm and 7 and 20 mm in shell lengths respectively when feeding on an unspecified formulated diet (Table 6 in Fleming et al. 1996). There are several studies on growth and survival of abalone juveniles when feeding on different algal species and growth rates of 60-100 [micro]m [day.sup.-1] can be expected for juveniles at 3-4 mm in shell length (see review by Kawamura et al. 1998a). Some studies compared formulated diets to an algal diet. For example higher growth-rates were achieved with juvenile abalone (H. fulgens), of approximately 13 mm in shell length, feeding on a formulated diet compared with juveniles feeding on macroalgae (Viana et al. 1993). Corazani and Illanes (1998) showed a similar trend with larger abalone of H. rufescens (~21 mm in shell length). Knauer et al. (1996) found no significant difference in growth rates of Haliotis midae juveniles (3-11 mm in shell length) when comparing a diatom diet (50 [micro]m [day.sup.-1]) and a formulated diet (59 [micro]m [day.sup.-1]). However the species composition and density of diatom species was not given in this study. Results of these studies indicate that formulated diets may support similar or better growth in larger juveniles, however abalone farmers still have to rely on algae as a settlement inducer and as a food source for recently settled postlarvae. Significantly better growth rates on an algal diet are expected for postlarvae and young juveniles.

Growth in abalone is highly variable (Day & Fleming 1992). The patchiness of natural food on the settlement plates is likely to contribute to the large size variation of juveniles experienced in nurseries. Formulated feed may allow a more constant form of nutrition potentially resulting in less size variation at the end of the nursery phase. By using formulated feed, farmers would become independent from seasonal variation in food supply and reduce variation between sites, tanks, and even plates. Formulated feed would also provide a great advantage when algal food becomes limited at the later stages of the nursery phase when juvenile grazing pressure is very high.

In this study three experiments were conducted to investigate settlement as well as growth and survival during the early development of Haliotis laevigata when feeding on algal as well as a formulated diet. The size variability of animals in all treatments was assessed over time to determine if juvenile populations feeding on formulated feed show less size variation.

Experiment 1: A large-scale settlement experiment was conducted to test the suitability of different algal species as settlement cues for H. laevigata larvae in a commercial abalone nursery. At settlement, diatom films dominated by the cultured diatom Navicula cf. jeffreyi or Cocconeis sp. were compared with plates covered by the green alga U. lens at two different developmental stages (6 and 8 wk old). Experiment 2: Settlement, growth and survival were estimated on a mixed diet of cultured algae (U. lens and Navicula cf. jeffreyi). Previous work has shown that nursery plates seeded with U. lens result in high and consistent settlement, whereas the diatom N. cf. jeffreyi support rapid growth of young postlarvae (Daume et al. 2000, Daume et al. 2004). This experiment was conducted to investigate if regular inoculation at commercial scale with the diatom N. cf. jeffreyi can sustain rapid growth until juveniles reach ca. 4 mm in shell length. Experiment 3: The potential for a shorter algal-reliant nursery approach was investigated by weaning young juveniles (3.79 [+ or -] 0.06 mm in shell length) onto formulated feed 10 wk after settlement. This short-term approach was compared with the traditional long-term nursery system utilizing cultured algae. The natural food was provided separately to determine which of the algal diets U. lens or N. cf. jeffreyi achieves the better growth-rates in juveniles >3 mm in shell length.

METHODS

Location

Experiments were conducted at ambient temperature at a commercial abalone farm, Great Southern Marine Hatcheries (GSMH), Albany, Western Australia between November 2001 and June 2002. The farm provided the abalone larvae and the diatom inocula for the trials.

Experiment 1: Larval Settlement on Different Algal Species

Preparation of Tanks and Plates

Three semicommercial nursery tanks (490 L) were each set up with 3 baskets in series that each held 20 settlement plates (= 60 plates/tank). Airlines were installed on the tank bottom along each side and down the center of the line of baskets (3 airlines/tank). Four treatments were tested: 6- and 8-wk-old U. lens and the diatom species Cocconeis sp. and Navicula cf. jeffreyi, both cultured for 2-3 wk on the plates. Five plates of each treatment were alternated so that each tank received 15 plates of each treatment, thus providing 45 plates for each treatment in total.

Algal culture

Ulvella lens seeding: To produce mass spore release, plates with large mature patches of U. lens and well-developed sporangia were selected, wiped clean to remove any diatom film, and stored in 1-[micro]m filtered seawater under two layers of 70% shading cloth, 2 wk prior to starting the conditioning of the experimental plates. U. lens seed plates were then placed between the baskets of the experimental tanks (4 seed plates per 1000 L tank), while the tanks were maintained with no water flow, low aeration, and without shading. A complete f/2 mix, including silicates, was applied (40 g 1000 [L.sup.-1] Microalgae Food, Manutech, Port Lincoln, Australia). These methods for spore collection were adapted from Takahashi and Koganezawa (1988). The release of zoospores is triggered by the increase in water temperature, nutrients and light. The largest release occurred 4-5 days after the introduction of seed plates.

Diatom Culture and Inoculation

Cultures of Navicula cf. jeffreyi and Cocconeis sp. were established in horizontally laid algal bags progressively increased in size up to commercial size bags of ca. 1 x 2 m. The diatom culture was harvested during the exponential growth phase (4-6 days after inoculation) and mixed into suspension. The diatom density was determined before tanks were inoculated. A 15 L inoculum ([10.sup.5]-[10.sup.6] cells [mL.sup.-1]) was used in each tank. A complete f/2 mix was applied (40 g 1000 [L.sup.-1]). The tanks remained static with low aeration for 24 h and then received low water flow with light aeration for 2-3 days.

Diatom Cell Density and Percentage Cover of Ulvella lens

The number of diatom cells and the percentage cover of U. lens were estimated under an inverted compound microscope in 15 randomly chosen fields of view on three sub samples (2 x 2 cm) of six plates per treatment at the time of settlement.

Larval Settlement

One hundred thousand larvae were released into each tank and left with very low water flow and low aeration for 3 days. Banjo sieves (118 [micro]m) were fitted onto each standpipe until settlement to prevent the escape of larvae. The settlement rates of larvae were estimated on six whole plates (30 x 40 cm) per tank after 3 days. Plates were kept submerged in seawater during counts and replaced into tanks immediately after measurements. A grid with equal size squares was placed inside the tray and a stereomicroscope was mounted above the tray to count and assess the settled larvae.

Experiment 2: Settlement and Growth on a Combined Algal Diet (Ulvella lens and Navicula cf. jeffreyi)

Preparation of Tanks and Plates

The same tank set up was used as in the previous experiment. Three tanks with 60 plates each were conditioned with U. lens 7 days before larval settlement and then inoculated with 15 L of cultured N. cf. jeffreyi per tank, 4 days before settlement. The seawater was filtered to 1 [micro]m and provided with a flow rate of 5-7 L [min.sup.-1].

Larval Settlement

One hundred thousand larvae were released into each tank and left with very low water flow and low aeration for 3 days. The settlement rates of larvae were estimated on whole plates after 3 days (see earlier).

Algal Inoculation

Tanks were inoculated with 15 L of Navicula cf. jeffreyi culture, 4 days before and 30 days after larval settlement and every 2 wk thereafter until the end of the trial. The details of algal culture and inoculation are described earlier.

Diatom Cell Density and Percentage Cover of U. lens

The number of diatom cells and the percentage cover of U. lens were estimated in 12 randomly chosen fields of view of six plates per tank under an inverted compound microscope at x200 magnification.

Growth and Survival of Juveniles

Juveniles were reared on a combined diet of U. lens and diatoms until they reached approximately 3 mm in shell length. Counts and measurements of the abalone were undertaken under a dissection microscope on six plates per tank at approximately 2-wk intervals.

Experiment 3: Assessment of Different Nursery Approaches

Early Weaning Onto Formulated Feed Versus Algal Feed

After 10 wk, when juveniles reached an average of 3 mm in shell length in the preceding experiment, all animals were taken off the plates and randomly assigned to 3 different treatments. (1) A subsample of 7,500 animals were transferred into three experimental raceways (1 m x 300 mmx 30 mm) and stocked at a density of 2,500 animals each (70% cover). Abalone were weighed and gradually introduced to a commercial weaner diet over the next 2-3 days. Animals were fed to excess at 1% of their body weight every day for the duration of the experiment. (2) Three semicommercial nursery tanks (see previous experiments) were set up with plates covered by Ulvella lens. (3) Three additional nursery tanks were stocked with plates colonized by Navicula cf. jeffreyi. The abalone were placed onto one horizontal settlement plate on top of each basket and left for 24 h allowing them to redistribute across all feed plates. Animals reached a stocking density of ca. 50 animals per plates (ca. 3,000 per tank). All tanks and raceways had separate inlets and outlets. The seawater was filtered to 5 [micro]m and provided at a flow rate of 5-7 L per [min.sup.-1].

Algal Inoculation

Tanks were inoculated with I 5 L of Navicula cf. jeffreyi culture, 2 wk before the experiment started and every 2 wk thereafter. The details of algal culture and inoculation are described above.

Diatom Cell Density and Percentage Cover of Ulvella Lens

The number of diatom cells and the percentage cover of U. lens were estimated in 12 randomly chosen fields of view of six plates per tank under an inverted compound microscope at x200 magnification.

Proximate Composition of Juvenile Diets

Diatom samples (Navicula cf. jeffreyi) were harvested and washed onto preweight 47 mm fiberglass filter paper using 0.5 M ammonium formate to remove residual salts. Samples of Ulvella lens were scraped of the settlement plates and washed with ammonium formate. All samples were then freeze-dried, reweighed and, together with samples of the formulated feed, forwarded to the State Chemistry Laboratory, Werribee, Victoria, for a proximate analysis of protein, carbohydrate, and fat content.

Growth and Survival of Juveniles

Growth-rates and survival of abalone were determined by measuring the size of 72 juveniles per tank and raceway as well as counting the number of juveniles per tank and raceway every 2 wk. At 10 wk, when the algal food supply became limiting, animals were taken off the old plates and moved onto a new set of plates with the same algal diet. The experiment was terminated after an additional 6 wk when animals reached l0 mm in shell length in one of the treatments.

Data Analysis

Statistical analyses were carried out using the STATISTICA computer packages. The assumptions of normality and homogeneity of variance of residuals and normal quartile plots were checked graphically for each data set using boxplots.

Daume et al. (1999) showed that the attractiveness of the substratum changes with the presence of recently settled conspecifics. Consequently, individual settlement choices are not independent, so that the numbers of larvae on the plates were analyzed using a paired t-test. Larvae started settling earlier on U. lens (old Ulvella-OU, young Ulvella-YU) than on the two diatom species (Navicula cf. jeffreyi. -N, Cocconeis sp. -C). Paired t-tests were performed on (1) the difference between OU and YU within each tank, (2) the difference between the average of the Ulvella treatments (OU + YU/2) and N, and (3) the difference between the average of the Ulvella treatments (OU + YU/2) and C. Total settlement rates per tank were analyzed using a 2-way ANOVA with tank and basket position as factors.

Abalone shell length measurements in the early weaning experiment were analyzed by repeated measure analyses of variance. Data at the end of the experiment was also analyzed using a 1-way ANOVA with Tukey HSD test. Relationships between the growth rates of the juveniles and the % cover of the feed species were explored with a simple regression analysis. Size-frequency distribution were drawn and compared over time and between treatments where appropriate using descriptive statistics (size range, mean, standard deviation [SD], and coefficient of variation [CV]). The shape of the distributions was compared with a Normal distribution using Shapiro-Wilks W tests (STATISTICA). Inter quartile differences IQD between the first and third quartile were used as a measure of spread of the size-frequency distributions.

RESULTS

Experiment 1: Larval Settlement on Different Algal Species

Larval Settlement

An overall settlement rate of 87% was estimated 3 days after larval release (Table 1). Sixty-one percent of the larvae chose to settle on the old U. lens, 14% on the young U. lens, 7% on the diatom species Navicula cf. jeffreyi and 5% on the diatom Cocconeis sp. We detected a high variation between tanks (ANOVA, P = 0.025) and within tanks (ANOVA, P < 0.001). Plates closest to the outlet had more settled larvae than plates closest to the inlet. Old and new U. lens were significantly different (paired t = 4.64, df = 2, P = 0.043). In addition, there was a significant difference between the combined U. lens treatments (OU + YU/2) and both diatom treatments, N. cf. jeffreyi (t = 4.412, d = 2, P = 0.048) and Cocconeis sp. (t = 4.804, df = 2, P = 0.041).

At the time of settlement, the estimated percentage cover of the old U. lens was 97% and 82% on young U. lens. A cell density of 4,978 cells [cm.sup.-2] was estimated on plates in the Cocconeis sp. treatment and 150,000 cells [cm.sup.-2] on plates with N. cf. jeffreyi.

Experiment 2: Settlement and Growth on a Combined Algal Diet (Ulvella lens and Navieula cf. jeffreyi)

Larval Settlement

A settlement rate of 30% [+ or -] 6.5 was estimated 3 days after larvae release. There was no significant difference between tanks (ANOVA, P = 0.47) or within tanks (P = 0.36).

Ulvella Lens Cover and Diatom Density

Ulvella lens covered between 50% and 70% of the plates for most of the growing period but around 58 days after larval release the cover decreased substantially (Fig. 1). Tanks were first inoculated with the cultured diatom Navicula cf. jeffreyi 4 days before larval settlement. Inoculations were repeated 30 days after settlement and then every 14 days thereafter.

[FIGURE 1 OMITTED]

At the time of settlement U. lens covered approximately 55% of the plate area (Fig. 1). Nitzschia sp. was the dominant species in the diatom assemblage on the plates with an estimate of 10,489 cells [cm.sup.-2]. This species developed naturally on the plates. The cultured diatom N. cf. jeffreyi slowly increased over time from 533 cells [cm.sup.-2] at the time of settlement to ca. 36,000 cells [cm.sup.-2] and was the dominant diatom species from day 44 until the end of the experiment.

Growth, Survival and Size Variability of Juveniles <4 Mm in Shell Length

The water temperature during this trial ranged between 17 [degrees]C and 24[degrees]C; with an average of 20[degrees]C. Juveniles growing on a mixed diet of cultured algae (Ulvella lens, Navicula cf. jeffreyi) reached 3.9 mm in shell length 72 days after larval release (Table 2). The growth-rates during the first 3 days after larval release were low; averages of approximately 40-[micro]m [day.sup.-1] were achieved during the following 41 days and ca. 69-[micro]m [day.sup.-1] during the last 28 days of the experiment. Post-settlement survival was estimated at 46%, 16 days after larval release, decreasing to 29% at 44 days and to 27% at the end of the experiment. The size range of juveniles increased substantially over time (Fig. 2, Table 3). After 10 wk the size of juveniles ranged between 2 and 5.1 mm in shell length and was nearly twice as large as ca. 2 wk earlier and about 37 times larger than at 3 days after larval release (Table 3). The SD, CV, and IQD increased over time. A large increase in at least one of the above measures was observed between 3 and 16 days, 30 and 44 days, and between 58 and 72 days. The skewness of the size frequency distribution changed from negative to positive between 16 and 30 days and between 44 and 58 days.

[FIGURE 2 OMITTED]

Experiment 3: Assessment of Different Nursery Approaches

Early Weaning Onto Formulated Feed Versus Algal Feed

The average water temperature declined from 19 [degrees]C during the first 8 wk to 17 [degrees]C during the second half of the experiment (Table 4). Growth rates and survival rates were highest on U lens. Juveniles feeding on U. lens reached 10 mm in shell length in <15 wk (Table 5). Juveniles feeding on the diatom Navicula cf. jeffreyi or the formulated feed only reached about 5 mm in shell length at the end of the experiment, resulting in a significant difference between the treatments (ANOVA, P < 0.001). Animals feeding on U. lens were significantly larger than the ones feeding on the diatom N. cf. jeffreyi (Tukey posthoc test, P < 0.001) and on the formulated feed (Tukey posthoc test, P < 0.001).

The size range of juveniles from all three treatments increased substantially over time (Fig. 3, Table 5). After 16 wk the size range of juveniles from all three treatments was highest and varied between 7.6 (formulated feed) and 8.1 mm (N. cf. jeffreyi) to 13.9 mm (U. lens). The CV ranged between 0.2 and 0.3 in all treatments and was highest in the population feeding on the diatom N. cf. jeffreyi after 4 wk. There was a significant difference in SD between the treatments throughout the experiment (ANOVA, P = 0.049). The SD was significantly lower among juveniles feeding on formulated feed than those feeding on U. lens at 4 wk (Tukey posthoc test, P = 0.01), 6 wk (P = 0.02), and 8 wk (P = 0.002) and higher among juveniles feeding on U. lens than in populations of the other two treatments at 10 wk (Tukey posthoc, P < 0.05), 12 wk (Tukey posthoc, P < 0.001), and 16 wk (Tukey posthoc, P < 0.05). The IQD was higher in the juvenile population feeding on U. lens than in the juvenile populations of the other two treatments at 12 wk (Tukey posthoc, P < 0.001) and 16 wk (P < 0.05). However, when the mean size of the populations was taken into account, the IQD/mean was highest in the N. cf. jeffreyi treatment.

[FIGURE 3 OMITTED]

Ulvella Lens Cover and Diatom Density

At the start of the early weaning trial, U. lens covered about 55% of the plate area. The cover decreased continuously until only 11% of the plates were covered after 9 wk (Table 6). The cell density on the plates in the Navicula cf. jeffreyi treatment followed a similar trend. New plates were introduced at 10 wk with a higher % cover in the U. lens treatment and higher cell densities in the N. cf. jeffreyi treatment than at the start of the trial. U. lens cover and N. cf. jeffreyi cell density declined rapidly.

Proximate Composition of Juvenile Diets

Protein was highest in U lens and lowest in samples from the diatom Navicula cf. jeffreyi (Table 7). All diets contained similar amounts of fat and ranged from 4% to 7%. The formulated diet was high in carbohydrates compared with other diets.

DISCUSSION

Settlement, Growth, Survival, and Size Variability

The findings of the first experiment clearly demonstrate that the green alga Ulvella lens provides a suitable substrate to improve the settlement of Haliotis laevigata larvae on a commercial scale. Settlement was significantly higher in both U. lens treatments when compared with the diatom treatments, indicating that U. lens is more suitable to induce settlement of H. laevigata larvae than a monospecific diatom film. It also indicates that larvae can distinguish between different developmental stages of U. lens. Similar to these findings, the settlement of H. rubra larvae was higher on older compared with younger U. lens (Daume et al. 2001). In contrast to the previous study, U. lens covered the settlement plates almost completely in both treatments of experiment 1 (97% and 82% for old and young U. lens respectively), indicating that the developmental stage of the alga (such as the development of sporangia) may be more important for the settlement induction than the percentage cover per se.

About 30 days after the initial inoculation with the cultured diatom Navicula cf. jeffreyi this species out-competed the naturally developing diatom species Nitzschia sp., which dominated at the start of the experiment. This demonstrates that the species composition of a biofilm can be altered by inoculation of a cultured diatom. However, if competitive species are present at the time of inoculation, this process might take some time.

In this study, growth rates of 40 [micro]m [day.sup.-1] were recorded on a combined diet of U. lens and N. cf jeffreyi until juveniles reached 2 mm in shell length and 70 [micro]m [day.sup.-1] until juveniles reached 4 mm in shell length, which compares favorably to other studies reviewed in Kawamura et al. (1998). Although diatoms are a good food source for young post larvae (Daume et al. 2000), U. lens may prove to be just as nutritious as diatoms; however smaller juveniles may not be able to graze on it until they reach a certain size. In this study we provide some evidence that juveniles of 2-3 mm in shell length (between 44 and 58 days after larval release) are able to actively remove U. lens from the plates and access it as a food source. This was evident by a sharp decline in percentage cover of U. lens between these two measurements points. The cell density of the cultured diatom N. cf jeffreyi was low at the start of the experiment, which may have contributed to the lower growth-rate of the abalone at this stage. However, it is more likely that higher growth-rates are observed towards the end of the trial because juveniles are able to access U. lens as a food source. As abalone grow, not only does their ability to access food change usually in accordance to their mouth size (Fleming et al. 1996), but also the apparent efficiency of the animal's radula. The radula undergoes morphologic changes during the animal's development and these changes may be linked to a shift in feeding habits from microalgae to macroalgae (Kawamura et al. 2001). The teeth may function as scoops when postlarvae are less then 1 mm in shell length whereas postlarvae larger than 1 mm may be able to scrape and feed on larger particles (Roberts et al. 1999a).

During the juvenile phase, growth of wild H. laevigata has been found to be linear (Shepherd 1988) and size variability should remain constant. However, at 44 days after larval release, we observed a clear shift in diet, which caused an increase in growth of the larger juveniles. This resulted in a greater variability (SD) and larger spread (IQD), and change from a uni-modal to a multimodal distribution. Another distinctive increase in SD and IQD was observed between 3 and 16 days after larval release, which coincides with an increase in growth-rate. This is believed to be a shift in nutrition from absorption of yolk reserves to efficient exogenous feeding (Kawamura et al. 1998a). The change in skewness from a longer left tail to a longer right tail after 16 and 44 days indicates that larger juveniles were growing faster. After day 44 the survival stabilized, suggesting that potential stresses associated with morphologic, physiologic, and nutritional changes were overcome.

Early Weaning onto Formulated Feed Versus Algal Feed

Several factors need to be considered when comparing the growth and the survival of juvenile abalone feeding on different food items; the food abundance and availability (e.g., patchiness of the food provided, the effort needed to remove a food item, and biomass ingested per unit grazing effort), and the digestibility and the nutritional value of the food item.

In this study all food was offered in excess but the cell density of N. cf jeffreyi was low at times. Other studies have acknowledged the difficulty to maintain high cell densities in the presence of heavily grazing juveniles (Kawamura et al. 1998b, Roberts et al. 1999b) and the results of this study support their findings. Thus the full potential of diatoms as a food source for juvenile abalone might have been underestimated in this experiment. However, the growth-rates of juveniles were very low despite the high cell density of N. cf. jeffreyi at the start of the experiment and when new plates were introduced. In contrast, post larvae 0.4-3.3 mm in shell length show superior growth when feeding on diatom species such as Navicula spp. when compared with U. lens (Daume et al. 2000). This indicates that diatoms are suitable for small juveniles but N. cf jeffreyi is not sufficient to maintain adequate growth for juveniles larger than 3 mm in shell length. In comparison to the other diets U. lens had a relatively high protein content. This could have contributed to the better growth on this diet. Brown & Jeffrey (1995) reported a slightly higher protein concentration, higher fat content, and lower carbohydrate content in Navicula jeffreyi compared to the proximate composition of this study. However, it is well known that the biochemical composition of an algal species can change depending on culture conditions such as light intensity and nutrients (Thompson et al. 1993, Fabregas et al. 1996). In our study samples were harvested from the nursery and thus growing conditions were dependent on abalone culture conditions in the outdoor nursery environment that is certainly different from the culture condition in a controlled culture room.

The decline in growth rates, after juveniles were transferred onto new plates, indicates that juveniles might have been stressed and feeding might have slowed due to the handling. However, feeding must have been substantial because there was a subsequent rapid decline in U. lens cover and diatom cell density. It may be beneficial to introduce new plates, with the desired algae present, allowing juveniles to move onto the new plates to prevent handling stress. Animals on formulated feed were not transferred at this stage and growth-rates rose slowly.

Growth-rates of juveniles feeding on formulated feed were even lower than on the N. cf jeffreyi at the start and slightly better at the end of the experiment but still remarkably lower than growth-rates of juveniles feeding on U. lens. The growth-rates of juveniles feeding on formulated feed improved towards the end of the trial indicating that the feed may be suitable for juveniles larger than 4 mm in shell length. The formulated feed used in this experiment may not be well matched for the nutritional requirements of small juveniles. The formulated diet was lower in protein and higher in carbohydrates than U. lens. Unless a suitable formulated feed is found, which can match growth-rates achieved with the macroalga U. lens, we suggest keeping animals on plates colonized by U. lens as long as possible. However, the size variability was particularly high in the juvenile population fed on N. cf jeffreyi followed by U. lens towards the end of the 16-wk experiment. In this study we provided initial evidence that juveniles feeding on natural food showed higher size variability than juveniles feeding on formulated feed. Feeding formulated diets instead of natural diets such as N. cf jeffreyi may provide a more uniform supply of nutrition throughout the nursery phase (independent of season), growing conditions, and size of the animals; and may ultimately lead to higher growth rates and lower mortality and size variability if the food can be improved. However, culturing U. lens as a food source may prove to be more cost effective because there is only a small cost involved in culture set-up and fertilizer. In addition, once U. lens is in the nursery system it regenerates in situ and no costly algal culture is required. At the present stage we regard U. lens as the preferred food species for juveniles >3 mm in shell length, because of the ease with which U. lens can be cultured in the nursery and superior growth achieved on this alga.

TABLE 1.

Percentage settlement ([+ or -] SE) of Haliotis laevigata
after 3 days (n = 3).

                      Old             Young           Navicula
Treatments          U. lens          U. lens             sp.

% Settlement    61 [+ or -]  14   14 [+ or -] 1    7 [+ or -] 0.3

                   Cocconeis
Treatments            sp.           [summation]

% Settlement     5 [+ or -] 0.5           87

TABLE 2.

Shell length ([micro] m) and growth-rates ([micro] m [day.sup.-1])
and survival (%) of post-larvae as well as water average water
temperature 3, 16, 30, 44, 58, and 72 days after larval release.

                           3         16         30

Shell length            338.36     749.48    1356.48
Growth-rates *           26.12      37.37      43.36
Survival ([dagger])                 46         37
Temperature *            18.37      18.75      19.43

                          44         58         72

Shell length           1921.58    2890.33    3850.22
Growth-rates *           40.36      69.20      68.56
Survival ([dagger])      29         28         27
Temperature *            20.42      20.45      19.62

* Between Consecutive sampling times.

([dagger]) Cumulative Survival based on 100% post-settlement.
Larval settlement at day 3 was 30%.

TABLE 3.

Descriptive statistics of juvenile size ([micro]m) 3, 16,
30, 44, 58, and 72 days after larval release.

                    3         16        30

Mean              338       749       1361
Min               295       610       1158
Max               379       863       1600
Size range         84       253        442
Standard
  Deviation        15.11     60.32      92.83
CV (%)              4         8          7
Inter-quartile
  differences      21.05    105.25     152.61
Skewness           -0.34     -0.37       0.18

                    44        58        72

Mean              1926      2893      3850
Min               1184      2000      2048
Max               2421      3738      5100
Size range        1237      1690      3100
Standard
  Deviation        260.75    360.99    663.85
CV (%)              14        12        17
Inter-quartile
  differences      374.92    460.8    1025
Skewness            -0.35      0.15      0.27

TABLE 4.

Growth-rates ([micro] m [day.sup.-1] [+ or -] SE) and percent survival
of juveniles feeding on Ulella lens or Navicula sp. in comparison to
formulated feed as well as average water temperature.

                         Week                  Week
                          2-8                  9-14

Ulvella lens      83.56 [+ or -] 8.7    62.67 [+ or -] 9.1
Navicula sp.      20.44 [+ or -] 8.2    16.05 [+ or -] 9.6
Formulated feed   13.54 [+ or -] 7.5    26.63 [+ or -] 8.8
Temperature              19.14                 16.67

                   Percent survival
                    up to 14 weeks

Ulvella lens      82.96 [+ or -] 1.4
Navicula sp.      76.16 [+ or -] 2.9
Formulated feed   67.06 [+ or -] 3.0
Temperature

TABLE 5. Descriptive statistics of size variability juvenile
populations feeding on Ulvella lens, Navicula sp., or formulated
feed after 1, 4, 8, 12, and 16 weeks (Experiment 3).

                           1        4        8       12      16

Ulvella lens
  Mean                   3.91     5.34     7.78     9.20    10.13
  Side range             3.70     5.00     7.30     7.70    13.90
  S.D.                   0.79     1.07     1.34     1.49     2.05
  C.V. (%r)             20       21       17       16       20
  Inter-quartile
    differences (IQD)    1.10     1.50     1.70     2.20     2.70
  IQD/mean               0.28     0.29     0.22     0.24     0.27
  Skewness              -0.44    -0.26    -0.57    -0.31    -0.36
Navicula sp.
  Mean                   3.86     3.66     4.88     5.01     5.63
  Size range             3.50     4.80     5.50     6.00     8.10
  S.D.                   0.77     1.07     1.15     0.97     1.41
  C.V. (%)              20       29       24       19       25
  Inter-quartile
    differences (IQD)    1.00     1.80     1.63     1.20     1.63
  IQD/mean               0.36     0.49     0.33     0.24     0.29
  Skewness              -0.22     0.12    -0.08     0.56     0.74
Formulated feed
  Mean                   3.56     3.53     4.15     5.00     6.77
  Size range             3.50     3.30     3.50     6.30     7.60
  S.D.                   0.86     0.77     0.78     1.10     1.31
  C.V. (%)              24       22       19       22       19
  Inter-quartile
    differences (IQD)    1.23     1.13     1.10     1.20     1.63
  IQD/mean               0.34     0.32     0.26     0.24     0.24
  Skewness               0.15     0.02     0.13     0.70     0.25

TABLE 6.

Average cover of Ulvella lens (%) and cell density of Navicula cf.
jeffreyi (cells [cm.sup.-2]) after 1, 2, 4, 6, 8, 10, 12, and 14 weeks.

   Week       Ulvella lens    Navicula cf. jeffreyi

    1              55               297,890
    2              51               110,443
    4              16                72,957
    6              15                20,596
    8              11                11,783

New plates
    10             62               485,126
    12             34                57,549
    14             28               117,549

TABLE 7.

Proximate composition of the juvenile diets (% [+ or -] S.E., n = 3).

                      U. lens           N. cf. jeffreyi

Protein          34.3 [+ or -] 1.6     25.1 [+ or -] 2.6
Carbohydrates    16.9 [+ or -] 1.4     16.3 [+ or -] 2.2
Fat               5.3 [+ or -] 1.6      6.7 [+ or -] 0.7

                  Formulated feed

Protein          28.5 [+ or -] 1.1
Carbohydrates    51.7 [+ or -] 1.7
Fat               4.1 [+ or -] 0.2

ACKNOWLEDGMENTS

The authors thank the staff for providing the infrastructure, the abalone larvae, and diatom cultures to inoculate the tanks. They also thank Dr. Rob Day for providing statistical advice and Dr. Greg Maguire and Andrew Hancock for useful comments to improve the manuscript. Great Southern Marine Hatcheries in Albany Western Australia, hosted this experiment. Fisheries Research and Development Corporation supported this study as part of a larger program (FRDC 98/306).

LITERATURE CITED

Brown, M. R. & S. W. Jeffrey. 1995. The amino acid and gross composition of marine diatoms potentially useful for mariculture. J. of Appl. Phycol. 7:521-527.

Corazani, D. & J. E. Illanes. 1998. Growth of juvenile abalone Haliotis discus hannai Ino 1953 and Haliotis rufescens Swainson 1822, fed with different diets. J. Shellfish Res. 17:663-666.

Daume, S., S. Brand-Gardner & W. J. Woelkerling. 1999. Preferential settlement of abalone larvae: diatom films vs non-geniculate coralline red algae. Aquaculture 174:243-254.

Daume, S., A. Krsinich, S. Farrell & M. Gervis. 2000. Settlement, early growth and survival of Haliotis rubra in response to different algal species. J. Appl. Phycol. 12:479-488.

Daume, S., S. Huchette, S. Ryan & R. W. Day. 2001. Settlement and growth of abalone on cultured algae in commercial abalone nurseries: A preliminary report. In: A. E. Fleming, editor. Proceedings of the 8th Annual Abalone Aquaculture Subprogram Workshop. Canberra, Australia: Fisheries Research and Development Corporation's Abalone Aquaculture Subprogram. pp. 139-148.

Daume, S., S. Huchette, S. Ryan & R. W. Day. 2004. Nursery culture of Haliotis rubra: The effect of cultured algae and larval density on settlement and juvenile production. Aquaculture 236:221-239.

Day, R. W. & A. E. Fleming. 1992. The determinants and measurement of abalone growth. In: S. A. Shepherd, M. J. Tegner & S. A. Guzman de Proo, editors. Abalone of the world: biology, fisheries and culture. Oxford: Blackwell. pp. 141-168.

Fabregas, J., A. Otero, E. Morales, B. Cordero & M. Patino. 1996. Tetraselmis suecica cultured in different nutrient concentrations varies in nutritional value to Artemia. Aquaculture 143:197-204.

Fleming, A. E., R. J. V. Barneveld & P. W. Hone. 1996. The development of artificial diets for abalone: a review and future directions. Aquaculture 140:5-53.

Fleming, A. E. 2000. The current status of the abalone aquaculture industry in Australia. In: A. E. Fleming, editor. Proceedings of the 7th Annual Abalone Aquaculture Workshop. Canberra, Australia: Fisheries Research and Development Corporation's Abalone Aquaculture Subprogram. pp. 1-15.

Kawamura, T., R. D. Roberts, H. Takami & C. M. Nicholson. 1998a. A review of the feeding and growth of postlarval abalone. J. Shellfish Res. 17:615-625.

Kawamura, T., R. D. Roberts & C. M. Nicholson. 1998b. Factors affecting the food value of diatom strains for post-larval abalone Haliotis iris. Aquaculture 160:81-88.

Kawamura, T., H. Takami, R. D. Roberts & Y. Yamashita. 2001. Radula development in abalone Haliotis discus hannai from larva to adult in relation to feeding transitions. Fisheries Science 67:596-605.

Knauer, J., P. J. Britz & T. Hecht. 1996. Comparative growth performance and digestive enzyme activity of juvenile South African abalone, Haliotis midae, fed on diatoms and a practical diet. Aquaculture 140:75-85.

Roberts, R. D., T. Kawamura & H. Takami. 1999a. Morphological changes in the radula of abalone (Haliotis iris) during post-larval development. J. Shellfish Res. 18:637-644.

Roberts, R. D., T. Kawamura & C. M. Nicholson. 1999b. Growth and survival of post-larval abalone (Holiotis iris) in relation to development and diatom diet. J. Shellfish Res. 18:243-250.

Shepherd, S. A. 1988. Studies on southern Australian abalone (genus Haliotis). VIII. Growth of, juvenile H. laevigata. Aust. J. Mar. Freshwater Res. 39:177-183.

Takahashi, K. & A. Koganezawa. 1988. Mass culture of Ulvella lens as a teed for abalone Haliotis discus hannai. NOAA Technical Report NMFS 70:29-36.

Thompson, P. A., M. Guo & P. J. Harrison. 1993. The influence of irradiance on the biochemical composition of three phytoplankton species and their nutritional value for larvae of the Pacific Oyster (Crassostrea gigas). Mar. Biol. 117:259-268.

Viana, M. T., L. M. Lopez & A. Salas. 1993. Diet development for juvenile abalone Haliotis fulgens. Evaluation of two artificial diets and macroalgae. Aquaculture 117:149-156.

SABINE DAUME (1), * AND STEPHEN RYAN (1,2)

(1) Department of Fisheries, Research Division, PO Box 20, North Beach, WA 6920, Australia; (2) c/- Great Southern Marine Hatcheries, PO Box L34, Little Grove, Albany WA 6330, Australia

* Corresponding author. E-mail: [e-mail]