Growth and survival of juvenile greenlip abalone (Haliotis laevigata) feeding on germlings of the macroalgae ulva sp.

ABSTRACT Germlings of the green alga Ulva sp. were developed as a diet for juvenile Haliotis laevigata ([greater than or equal to]3.5 mm shell length) and compared with a current commercial diet consisting of Ulvella lens plus the diatom species Navicula cf. jeffreyi. The utilization of macroalgae germlings (juvenile gametophyte and sporophyte) allowed 3-dimensional growth and subsequently provided greater feed biomass in comparison with the current 2-dimensional commercial feed for the later nursery phase consisting of 5-10 mm (shell length) juvenile abalone. The juvenile abalone showed active feeding on both the Ulva germling diet and the current commercial diet. The Ulvella lens/Navicula cf. jeffreyi diet resulted in abalone of significantly larger shell length at the end of the 14-wk feeding trial. However, the Ulva germling diet recorded significantly larger abalone for the first 4-5 wk, whereas the commercial diet produced significantly larger abalone from week 6 to the end of the trial. The growth rate on both diets exceeded 100 [micro]m.[day.sup.-1] and the specific growth rates were maintained above 1%.[day.sup.-1] for the duration of the feeding trial with neither measure portraying significant differences between diets. There was no significant difference in juvenile abalone mortality feeding on the two diets. The Ulva germling consumption exhibited a spike (500 germling blades.[abalone.sup.-1].[day.sup.-1]) in consumption at week three then, once reduced, a gradual increase occurred until the end of the trial. Ulvella lens consumption demonstrated a similar pattern to Ulva germlings consumption and was significantly, positively correlated. Consumption rates for the two green algae both correlated with juvenile abalone growth. The diatom (Navicula cf. jeffreyi) consumption was affected by plate rotation (light intensity and grazing pressure) rather than juvenile abalone.

KEY WORDS: juvenile abalone, Haliotis laevigata, Ulva, germlings, Ulvella lens, diatoms, dietary value

INTRODUCTION

To culture quality abalone to commercial harvest size within an economical time frame, current culture protocols and in particular, juvenile nutrition need to be improved. To advance this area of production, new juvenile diets must be explored that supply sufficient biomass and provide greater nutritional benefits.

The main diet of postlarval and early juvenile abalone (up to -5 mm) in the natural environment consists of epiphytic and epilithic diatoms, crustose coralline algae, turf algae and bacteria, whereas larger juveniles consume macroalgae (Dunstan et al. 2002, Kawamura et al. 1995, Kawamura 1996, Kawamura & Takami 1995, McShane et al. 1994, Takami et al. 1998). Once abalone reach the transition phase from a diatom-based diet to a macroalgae diet, diatoms such as Cylindrotheca closterium (Ehrenberg) alone are no longer sufficient to maintain adequate growth rates in an aquaculture system (Takami et al. 2003). At this stage additional algal food is required to sustain maximum growth rates and reduce the variability of growth and survival rates. Maintenance of an adequate food supply to the 5-10 mm juveniles is seen as a major limiting factor in the intensification of abalone nurseries (Krsinich et al. 2000).

Currently in Australian commercial abalone nurseries, postlarvae are supplied with diatoms (e.g., Navicula cf. jeffreyi) and as they develop into juveniles they are provided with the crustose green alga Ulvella lens crouch (Daume & Ryan 2004, Daume et al. 2004). U. lens has been shown to induce higher settlement rates of abalone larvae compared with monospecific benthic diatom films (Daume et al. 2000, Krsinich et al. 2000). By itself U. lens only supports moderate growth rates but, when combined with an easily digestible diatom such as N. jeffreyi, the diet can sustain high growth rates (Daume & Ryan 2004, Kawamura et al. 1998). Takami et al. (1997) also found that rapid abalone growth is only achievable on crustose coralline algae (Lithophyllum yessoense) if diatoms are present. Once abalone exceed about 5 mm in length, the combined diet of U. lens and N. jeffreyi is unable to adequately support the high abalone biomass per plate (Daume & Ryan 2004).

A potential alternative commercial feed for juvenile abalone (5-10 mm) may be macroalgae sporelings. The majority of juvenile abalone dietary studies have been conducted with mature macroalgae; however, they may have different nutritional and structural properties to juvenile macroalgae of the same species (Van Alstyne et al. 1999). The juvenile macroalgae (germlings) can potentially provide a greater biomass per plate because of their 3-dimensional morphology compared with the 2-dimensional encrusting algae and have been shown to support moderate to rapid growth of 90-130 [micro]m.[day.sup.-1] (Maesako et al. (1984) as cited in Kawamura et al. (1998)).

The 3-dimensional growth reduces the surface area required and gives the feed the potential to combat the juvenile abalone's ability to consume 5% to 30% of their body weight in algae each day (Corazani & Illanes 1998, Hahn 1989). Ulva bas been used in numerous studies, both individually or as part of mixed/rotation diets but is considered a relatively poor nutrition source (Simpson & Cook 1998). Shpigel et al. (1999) bas shown that specific growth rates of 0.6 to 1%.[day.sup.-1] can be attained for juveniles 8-15 mm in shell length and that some abalone species grow better on Ulva cultured in high ammonia-N enriched seawater, underlying the importance of the feed's nutritional value.

In this study, the dietary value of an Ulva sp. germling diet was compared with a currently used commercial diet consisting of the green alga U. lens plus the diatom species N. jeffreyi on the growth and survival of juvenile greenlip abalone (Haliotis laevigata Donovan).

MATERIALS AND METHODS

Location

The feeding trial was conducted in a greenhouse at the Aquaculture Development Unit, Challenger TAFE, Fremantle, Western Australia between March and August 2004. Juvenile greenlip abalone (Haliotis laevigata) were supplied by Great Southern Marine Hatcheries in Albany, Western Australia.

Algal Culture--Diets

Ulva sp. Germling Diet

Ulva sp. thalli were collected from submerged limestone rocks on South Mole in Fremantle and exposed to a cold (4[degrees]C) treatment to induce gametogenesis. Ulva thalli were arranged in layers in-between moist newspaper then refrigerated. After 7 days of cold treatment, 10 kg blotted wet weight of Ulva thalli was placed into each of the five, 400 L tanks filled with a modified f/2 culture medium (Guillard & Ryther 1962), that lacked PII metals, sodium metasilicate and vitamin stock solutions. Each tank held 3 baskets of 12, 30 x 60 cm PVC plates lying horizontally. The tanks received only light aeration to reduce water motion and allow maximum spore attachment.

The Ulva thalli were removed from the five tanks after 6 days and the germling seeded PVC plates redistributed into three 400 L tanks each containing 3 baskets of 20 plates now orientated vertically. The germlings were then cultured over 5 wk in the modified f/2 medium, which was exchanged twice weekly.

Ulvella lens Pius Navicula cf. jeffreyi Diet

The diatom Navicula cf. jeffreyi (CSIRO Hobart, CS-514) was cultured in standard f/2 medium (Guillard & Ryther 1962) with cultures starting indoors in 4 petri dishes that were then scaled up through four, 1.5 L horizontally laid culture bags and finally to one, 60 L, shallow tank outdoors. 20 L of the N. jeffreyi inoculum was added to each of the three U. lens tanks.

Sixty U. lens seed plates (30 x 60 cm PVC) (Daume & Ryan 2004, Daume et al. 2004) were placed at regular intervals between clean 30 x 60 cm PVC plates and exposed to sunlight for 5 days, then removed. The aeration was low to allow the U. lens spores to attach to the plates and the modified f/2 medium was exchanged twice weekly.

Feeding Trial

For each of the 2 treatments (Ulva germling diet and U. lens/N. jeffreyi diet), three, 400 L tanks were stocked with three baskets of 20 vertically arranged seeded plates (30 x 60 cm). The tanks were aerated by three airlines spaced evenly along the bottom and shaded with 70% shade cloth; 1 [micro]m filtered bore seawater was supplied at 10 L.[min.sup.-1] via a spray bar above the water surface. The water temperature over the 3 mo feeding trial started at 20.8 [+ or -] 0.13[degrees]C (May) then reduced to 19.7 [+ or -] 0.18[degrees]C (June) and finished at 19.0 [+ or -] 0.08[degrees]C (July).

Juvenile greenlip abalone (H. laevigata) were taken off an U. lens/naturally occurring diatoms diet and transported (4 h) on PVC plates seeded with U. lens between wet sponge sheets in insulated containers. The PVC plates with juveniles attached were placed across the top of the baskets in each tank and left for 2 wk to allow the juveniles to migrate onto the experimental diet plates. Seventeen hundred juveniles of 3.5-4 mm shell length were stocked in each of the six tanks giving approximately 28 juveniles per 30 x 60 cm plate.

The feeding trial was for a period of 14 wk and the three U. lens tanks were reinoculated with N. jeffreyi during weeks 2, 4 and 8. The plates in all tanks were rotated twice, 180[degrees] about the horizontal in weeks 3 and 10.

Measurements

Abalone shell length (mm) and weight (g) were measured at the beginning of the trial and then weekly by collecting a sub sample of 50 juveniles from 10 randomly selected plates in each tank. After the juvenile abalone had been measured the contents of each tank were siphoned through 50 [micro]m mesh and the dead abalone counted.

Ulva germling abundance was determined by counting the number of germling blades per [cm.sup.2] of plate at weekly intervals. Every fifth plate was counted with 5 randomly selected fields of view (0.785 [cm.sup.2]) counted diagonally across the plate. The density of the U. lens was determined by estimating percentage cover along a graticule using the same sampling procedure as the Ulva germlings. The density of N. jeffreyi was measured on 2 removable notches cut from the side of every sixth plate. The notches were approximately 16 [cm.sup.2] and positioned 3 cm from the top and bottom. The number of cells present on these notches was then counted for a defined area in 20 randomly chosen fields of view and the number of cells.[cm.sup.-2] calculated.

Biochemical Analysis

Samples were taken by scraping diagonally across the plates that were used for determining weekly algal abundance. Scrapings were stored at -20[degrees]C until needed.

Algal Dry Weight

Five milliliters of the U. lens/N. jeffreyi samples and 0.05 g of the Ulva germling sample were filtered through Whatman GF/C (2.5 cm) glass microfiber filters that had been washed, precombusted and preweighted. The filtrate was then washed with 10 mL of ammonium formate solution (0.65 M) to remove excess salts, dried in an oven for 12 h (80[degrees]C) and placed in a vacuum desiccator overnight. They were then weighted to 4 decimal places on an analytical balance.

Lipid Determination

The lipid content of the algal diets were determined based on the method of Bligh and Dyer (1959) as modified by Kates and Volcani (1966) and adapted by Mercz (1994).

Five milliliters of the U. lens/N. jeffreyi samples and 0.025 g of the Ulva germling samples were filtered onto Whatman GF/C (2.5 cm) glass microfiber filters, rinsed with 10 mL ammonium formate (0.65 M) and stored at -20[degrees]C for approximately 2 mo.

Once thawed, filters were homogenized in a glass mortar and pestle with 5 mL of a methanol:chloroform:deionised water solution (2: 1:0.8 v/v/v). The extract was centrifuged at 3000 rpm for 5 min and the supernatant transferred to a second, 10 mL graduated glass centrifuged tube. The volume was made up to 5.7 mL with fresh methanol:chloroform:deionised water, then 1.5 mL chloroform and 1.5 mL deionized water were added while mixing well.

The tubes were recentrifuged (3,000 rpm for 5 min), after which phase separation was complete and the lower green chloroform layer containing the lipids were carefully transferred into dry, preweighted 4 mL glass vials. A few drops of toluene were added and the extract dried under ultra pure nitrogen. The vials were placed in a vacuum desiccator (KOH pellets) overnight and weighted to 4 decimal places.

Protein Determination

The protein content of the algal diets were determined utilizing a modification of the Lowry et al. (1951) method by Dorsey et al. (1978) and Mercz (1994). Samples were prepared as in the Lipid Determination procedure (above) with 0.0125 g of Ulva germlings used for each sample.

Filters were homogenized with 5 mL Biuret reagent in a glass mortar and pestle, then transferred into 10 mL graduated glass centrifuged tubes and 0.14 mL of deionized water added. Protein standards (Bovine Serum Albumin) of 0, 10, 20, 30, 40, 50, 60 and 70 [micro]g were made up to 0.14 mL with deionized water, and 5 mL Biuret reagent was added.

All tubes were incubated at 100[degrees]C for 60 min and immediately after 0.5 mL Folin Phenol reagent was added while mixing on a Vortex stirrer. The tubes were cooled for 15 min at 10[degrees]C to 15[degrees]C and 15 min at room temperature, then centrifuged (3,000 rpm for 5 min). The absorbance of the supernatant was read at 660 nm and the protein content determined from the standard curve.

Carbohydrate Determination

The carbohydrate content of the algal diets were determined using the method of Kochert (1978) incorporating modifications by Ben-Amotz et al. (1985) and Mercz (1994). Samples were prepared as in the lipid determination procedure (earlier) with 0.012 g of Ulva germlings being used.

Five milliliters of [H.sub.2]S[O.sub.4] (1 M) was used to homogenize filters in a glass mortar and pestle before being transferred into 10 mL graduated glass centrifuged tubes and incubated at 100[degrees]C for 60 min. After cooling to room temperature and centrifuging (3,000 rpm for 5 min) a known volume of supernatant (<50 [micro]g total carbohydrate, which is between 0.1-0.5 mL, depending on initial algal concentration) was taken and made up to 2 mL with deionized water in 10 mL graduated glass centrifuged tubes. Carbohydrate standards (Glucose) of 0, 10, 20, 30, 40 and 50 [micro]g were made up to 2 mL with deionized water.

One milliliter of 5% (w/v) phenol solution was added and mixed well on a Vortex stirrer. Five milliliter of concentrated [H.sub.2]S[O.sub.4] (98%, 18 M) was added rapidly and then the tubes left for 30 min to cool. Absorbance was measured at 485 nm and the carbohydrate content determined from the standards.

Data Analysis

Juvenile abalone growth and density for the 2 dietary treatments Ulva germling and U. lens/N, jeffreyi were compared by analysis of variance (1-way ANOVA). A univariate analysis of variance, posthoc (Tukey HSD) test was applied to test for differences between mean abalone interval sizes (shell length) on the two diet treatments at each time interval. Comparisons of the algae diets consumption and biochemical composition were achieved through Bivariate Correlation and analysis of variance (l-way ANOVA) respectively. The plate rotation was analyzed with an Independent t-test.

RESULTS

Abalone Growth

At the commencement of the feeding trial there was no significant difference between the average shell lengths of the abalone distributed to the 2 diets ([F.sub.(df 1298)] = 0.49 [P = 0.484]). The juvenile abalone grew on both diets with the Ulva germling diet producing significantly larger abalone (shell length) for the first 5 wk ([F.sub.(df 1,2398)] = 6.779 [P < 0.05]) (Fig. 1).

[FIGURE 1 OMITTED]

During the first 4 wk the mean weekly increase in shell length of the abalone on the Ulva germling diet was 0.51 [+ or -] 0.1 mm with each increase shown to be significant (Table 1). The abalone on the U. lens/N, jeffreyi diet only averaged a weekly increase in shell length of 0.41 [+ or -] 0.1 mm for the first 4 wk but were able to maintain significant increases in shell length until week 7 (Table 1). The subsequent extended period of significantly faster growth resulted in the abalone on the U. lens/N, jeffreyi diet surpassing the size (shell length) of the abalone on the Ulva germling diet and this transition is evident in Figure 1 where the two growth profiles intersect between weeks 5 and 6. The U. lens/N. jeffreyi diet then proceeded to yield significantly larger abalone (shell length) ([F.sub.(df 1,2698)] = 24.671 (P < 0.05)).

The weekly growth rates of juvenile abalone were very variable (reaching over 100 [micro]m.[day.sup.-1]) on both the Ulva germling diet and the U. lens/N. jeffreyi diet during the first 6 wk (Table 1, Table 2). The Ulva germling diet sustained a higher specific growth rate for abalone over the first 4 wk, reaching a maximum of 1.5%.[day.sup.-1] (Fig. 2).

[FIGURE 2 OMITTED]

After the 8th wk of the feeding trial, shell length was not significantly different between adjacent weeks on either diet (Table 1), indicating a reduction in absolute growth rate (Table 2) and specific growth rate (Fig. 2). From Table 2 it is evident that the Ulva germling diet produced slightly lower, not significantly lower, ([F.sub.(df 1,82)] : 0.583 [P = 0.448]) growth rates over the entire feeding trial. Consequently, the specific growth rate of the juvenile abalone was not significantly affected by diet ([F.sub.(df 1,82)]= 1.968 [P = 0.164]) with both recording 1%.[day.sub.-1] by the end of the 14 wk trial (Fig. 2).

The juveniles consuming the Ulva germling diet were smaller at the completion of the trial with an average of 9.61 [+ or -] 0.1 mm shell length, compared with the average shell length of 10.29 [+ or -] 0.1 mm the abalone on the U. lens/N. jeffreyi achieved (Fig. 1). The final abalone shell lengths were significantly different indicating that the U. lens/N. jeffreyi diet produced significantly larger abalone than the Ulva germling diet ([F.sub.(df 1298)] : 10.335 [P < 0.05]).

Abalone Survival

In conjunction with growth, mortality and the subsequent abalone density are important in comparing the two diets effectiveness as a feed for juvenile abalone. Weekly mortality on both diets exhibited very similar profiles with the U. lens/N. jeffreyi diet producing an average of 91 mortalities in week 3 but thereafter the Ulva germling diet recorded slightly higher mortalities until week 10 (Table 3). Calculating the progressive abalone density from the weekly mortality indicated there was no significant difference in abalone density between the two diets ([F.sub.df 1,88] = 0.569 (P = 0.453)). Crushed shells and escapees were unable to be considered in the weekly mortality giving a discrepancy with the final densities. Even though the number of abalone at the end of the 14 wk trial was lower on the Ulva germling diet, 1,948 abalone (38.2% survival), compared with the U. lens/N. jeffreyi diet, 2,390 abalone (46.9% survival), the difference was not significant ([F.sub.(df 1,4)] = 3.911 (P = 0.119)).

Algal Consumption

The juvenile abalone consumed entire Ulva germlings, both blade and attachment regions. During the first month of the experiment, consumption of germlings peaked at 500 germling blades.[abalone.sup.-1].[day.sup.-1] but by week 6 the consumption had decreased to 100 germling blades.[abalone.sup.-1].[day.sup.-1]. Consumption gradually increased after that (week 6), doubling by the end of the feeding trial (Fig. 3). During the last two months there was a positive correlation between the increase in Ulva germling consumption and the increase in abalone shell length (R = 0.583; P < 0.05).

[FIGURE 3 OMITTED]

The consumption of U. lens followed a similar trend and was significantly, positively correlated (R = 0.422, P < 0.05) to the Ulva germling consumption but with a reduced rate of decline after the period of high consumption (Fig. 4). Consumption of U. lens also significantly correlated to the grow rate (R = 0.544, P < 0.05) and subsequently the specific growth rate (R = 0.618, P < 0.05) of the juvenile abalone.

[FIGURE 4 OMITTED]

Diatom consumption (Fig. 5) exhibited a similar profile to that of the Ulva germlings and U. lens consumption including the slow increase after week 7. However this increase fluctuated under or on zero diatoms.[abalone.sup.-1][day.sup.-1] for weeks 7-12 because the positive growth of algae was greater than the consumption by abalone. Subsequently the diatom consumption did not correlate with either the Ulva germlings or the U. lens consumption, nor did it appear to relate to abalone growth.

[FIGURE 5 OMITTED]

During the feeding trial N. jeffreyi was reinoculated and the PVC plates rotated as illustrated in Table 4 and Figure 5. The consumption of diatoms corresponds to the plate rotation rather than to reinoculation. When comparing the two rotational profiles (1 = start at top - bottom - top and 2 = start at bottom - top - bottom) of diatom consumption they were shown to be statistically different ([t.sub.(df 41)] = -2.986 [P = 0.005]). However, if the plates were not rotated then consumption was not significantly different. This indicates that rotating the plates had a considerable effect on the consumption of N. jeffreyi.

Two other species of diatoms, Melosira cf. moniliformis and Synedra sp. contaminated the Ulva germling treatment at various stages (Table 4). These diatoms were present throughout the 14 wk, however they only bloomed at the top of the plates on 3 separate occasions (start, week 11 and week 13). The contamination was quickly removed by physically detaching (hand abrasion) it from the substrate then siphoning the tanks' contents.

Biochemical Composition of Algal Diets

The proximate biochemical composition of the U. lens/N. jeffreyi diet and Ulva germling diet can be seen in Table 5. The two diets both exhibit dry weights of approximately 12.5% with the U. lens/N. jeffreyi only slightly higher. Even though the lipid and carbohydrate components were greater in the Ulva germling diets the difference was not significant ([(F.sub.(df 1,15)] = 2.141 [P = 0.164]) and ([F.sub.(df 1,14)] = 1.767 [P = 0.205]) respectively). The protein level in the Ulva germling diet however was shown to be significantly higher ([F.sub.(df 1,16)] = 10.893 [P = 0.005]). The total extractable component (sum of protein, lipid and carbohydrate) was higher for the Ulva germlings diet.

DISCUSSION

The experimental juvenile macroalgae diet of Ulva germlings was comparable to the current commercial diet consisting of Ulvella lens and Navicula cf. jeffreyi for the growth and survival of juvenile Haliotis laevigata. The two diets demonstrated similar absolute and specific abalone growth rates but the U. lens/N. jeffreyi diet produced significantly larger abalone at the completion of the 14 wk feeding trial. Juveniles feeding on the U. lens/N. jeffreyi diet reached 10 mm (SL) in less than 13 wk whereas the Ulva germling diet produced juveniles of 9.61 mm (SL) at week 14. Daume and Ryan (2004) found that for abalone of a similar initial size to the present trial (4 mm SL), it took less than 15 wk to reach 10 mm (SL) on just U. lens with a stocking density of approximately 50 animals per plate.

The transition at week 5, between the Ulva germling diet and U. lens/N. jeffreyi diet, producing significant larger abalone (Fig. 1) indicates that Ulva germlings were a more successful diet for H. laevigata in the range of 3.5-6 mm. The failure to sustain a growth advantage to week 14 indicates better performance of the U. lens/ N. jeffreyi diet for H. laevigata in the range of 6-10 mm. The Ulva germling diet can therefore be considered an acceptable commercial diet for juvenile abalone (<6 mm) and used either as an alternative or in conjunction with the current commercial diet of U. lens/N. jeffreyi.

In the later phase of the trial, Ulva germlings were either no longer able to supply the juvenile abalone with specific nutrients or there was not enough biomass. Lack of biomass was an unlikely cause as only 25% of the Ulva germlings had been consumed at this point. To overcome either of these problems, freshly seeded plates could be cycled through to maintain a constant supply of new Ulva germlings. Daume et al. (2004) utilized this procedure for U. lens, which enabled the high initial growth rates of newly settled Haliotis rubra to be maintained for 114 d.

The abalone on the Ulva germling diet recorded 9 out of the last 10 weekly increases as not significant, which was further compounded by the reduction in abalone weekly growth to only 0.08 mm (SL) for the last 2 wk (Table 1). However the U. lens/N. jelfreyi diet also sustained a low weekly abalone growth rate of 0.24 mm (SL) for the last 2 wk. H. rubra has been shown to achieve steady growth for 100 d and then fail to grow further on some monospecific algal diets (Day & Fleming 1992).

The reduction of juvenile growth towards the end of the feeding trial occurred at the peak of the winter season with water temperatures dropping from 20.8[degrees]C to 19[degrees]C. The colder water temperatures may have resulted in the metabolic activity of the abalone reducing, causing less consumption of both algal diets and subsequently slower growth rates.

By week 12 both diets had produced abalone of approximately 9.5 mm (SL) with decreasing growth rates and at this stage could be weaned onto formulated feed (Dunstan et al. 2002, Fleming et al. 1996). However, it could be beneficial to maintain the abalone on its original diet for as long as possible by incorporating fresh seeded plates to reduce competition for food as well as the stress caused by handling (Daume et al. 2004, Fleming 1995).

An alternative to weaning may be to incorporate the Ulva germling diet as part of a mixed/rotational diet when it no longer supports adequate growth by itself (Day & Fleming 1992). The consumption of the subsequent algae, rotated through, may account for the deficiencies in the initial diet, in this case Ulva germlings (Simpson & Cook 1998). The plate method of feeding juvenile abalone directs itself to diet rotation or a mixed diet plan whereby plates seeded with different diets can be interspersed throughout the tanks. Simpson and Cook (1998) and Stuart and Brown (1994) demonstrated that Ulva sp. as a singular diet produced the lowest abalone growth rates but when presented in a rotational/mixed diet it sustained excellent growth rates.

Growth rates of H. laevigata fed the Ulva germling diet were not significantly different from those produced on the U. lens/N. jeffreyi diet (Table 2). The growth rate profile was similar to that obtained by Daume and Ryan (2004) utilizing U. lens, where once the first 2 wk had been removed, the next 6 wk recorded 84 [micro]m.[day.sup.-1] and the final 6 wk 63 [micro]m.[day.sup.-1]. The growth rates during the first 2 wk were removed from Table 2 as any nutrient deficiency in a diet may be masked by the abalone ability to utilize its own energy stores (Fleming et al. 1996). As the juvenile abalone were taken off an U. lens/naturally occurring diatom diet, the weaning process was considered minimal compared with the recommendation of approximately 50 d (Day & Fleming 1992). However it was important to run the feeding trial for as long as possible to detect any effects of nutrient limitation and to determine the capacity of an alga to maintain acceptable abalone growth (Day & Fleming 1992).

The Ulva germling diet achieved growth rates of over 100 [micro]m.[day.sup.-1] during the first 6 wk. This was comparable to Haliotis discus discus growth rates attained over a month on a variety of macroalga germlings including Colpomenia sinuosa, Ectocarpus siliculosus and Enteromorpha sp. (Maesako et al. (1984) as cited in Kawamura et al. [1998]). Takami et al. (2003) showed that Haliotis discus hannai of approximately 1.8-2.2 mm and 2.8-2.9 mm shell length could reach growth rates of 80 and 100 [micro]m.[day.sup.-1] respectively on juvenile sporophytes of Laminaria japonica.

The specific growth rate reached over 1.3%.[day.sup.-1] and finished at 1%.[day.sup.-1] on both diets with no significant difference between them. The Ulva germling diet achieved 1.5%.[day.sup.-1] at week 4 but then exhibited a slow decline. Corazani and Illanes (1998) reported that H. discus hannai obtained a higher specific growth rate (0.69%.[day.sup.-1]) utilizing adult Ulva rigida than other macroalgal diets, whereas Haliotis rufescens achieved the lowest specific growth rate. This was comparable to the 0.71%.[day.sup.-1] achieved by H. discus hannai on an Ulva sp. (Uki & Watanabe 1992). Ulva lactuca has been found to have reasonable dietary value for Haliotis tuberculata (1.16%.[day.sup.-1]) but significantly lower for H. discus hannai (0.75%.[day.sup.-1]) (Mai et al. 1996). Haliotis iris was only able to achieve 0.1%.[day.sup.-1] on U. lactuca (Stuart & Brown 1994). Simpson and Cook (1998) also found that the suitability of Ulva sp., as a feed was dependent on the abalone species.

The Ulva sp. being used in the present study was not manipulated through nutrient enrichment during the 14 wk feeding trial. Enriched Ulva rigida has been shown to produce growth rates of juvenile Haliotis roei comparable to those achieved on the best performing artificial diets (Boarder & Shpigel 2001). Taylor and Tsvetnenko (2004) showed that only 15 mgN.[L.sup.-1] enriched U. rigida produced significantly higher specific growth rates (0.28 [micro]m.[day.sup.-1]) than unenriched U. rigida. Shpigel et al. (1999) reported growth rates of 44.47 and 121.47 [micro]m.[day.sup.-1] for H. discus hannai and H. tuberculata respectively on a high ammonia-N enriched U. lactuca compared with 31.7 and 80.72 [micro]m.[day.sup.-1] on low ammonia-N enriched U. lactuca.

The growth rate for H. tuberculata produced on the high-enriched U. lactuca (Shpigel et al. 1999) was the only growth rate to exceed that obtained on the Ulva germling diet in this study. During the culturing process, before the feeding trial began, the Ulva germlings were grown in f/2 media (minus the PII metals, sodium metasilicate and vitamin stock solutions) (Guillard & Ryther 1962). The elevation in nutrients at the start may have led to the extremely high level of consumption (week 3, Fig. 3) resulting in the significantly larger abalone size during the first 5 wk and in turn the rapid 1.5%.[day.sup.-1] specific growth rate. Therefore it is important to investigate the benefit of culturing nutrient enriched (high ammonia-N seawater) Ulva germlings to achieve the best juvenile abalone growth rates.

The consumption rate of Ulva germlings was extremely high during the first month and subsequently the abalone grew rapidly. However, once the consumption rate reduced so did the growth rates giving a significant correlation (Fig. 3). Hone (1992) showed that Ulva australis was rapidly consumed by abalone. On a quantitative basis Ulva sp. had the lowest consumption in g.[abalone.sup.-1].[day.sup.-1] compared with five other adult macroalge and subsequently produced the lowest growth rates for Haliotis midae (Simpson & Cook 1998). Boarder and Shpigel (2001) reported that inorganically enriched U. rigida had the lowest consumption rate but was still able to produce growth rates of H. roei comparable to that achieved on some of the best artificial diets. As mentioned before, ammonia-N enriched U. lactuca produced the highest growth rates for both H. discus hannai and H. tuberculata but these rates were recorded while consuming significantly less biomass (Shpigel et al. 1999). This indicates that nutrient enriched Ulva sp. produces greater growth rates, while requiring less biomass to achieve them.

Daume and Ryan (2004) reported that at the start of a feeding trial U. lens had 55% coverage and decreased until 11% was left at 9 wk, when new plates were introduced. That consumption pattern is considerably faster than recorded in this study with 14 wk needed to achieve approximately 11% cover from a similar start value. The different stocking density accounts for the difference in U. lens consumption rates with the present study starting approximately 20 fewer animals per plate (Daume & Ryan 2004).

The U. lens consumption was significantly correlated with the Ulva germling consumption indicating that the juvenile abalone exhibited a similar preference for the two species of green algae (Fig. 4). This is understandable because the biochemical profiles of the two diets were fairly similar (Table 5). The dry weight of both diets, even though similar (12.5%), was lower than [approximately equal to] 15% expected for the majority of algae such as Ulva sp. (Mercer et al. 1993, Shpigel et al. 1999). This may be caused by the sampling method incorporating all the biofilm/moisture from the plates rather than just the two green algae.

The Ulva germling diet exhibited a higher overall percentage extracted, possibly because of the high ash content of diatoms including Navicula (Brown & Jeffrey 1995). The individual biochemical components were also higher. Specifically, the protein level (32.3%) which was significantly larger than that of the U. lens/N. jeffreyi diet. It was nearly identical to that of U. rigida when enriched from 11.4% to 32.2% protein by using high nutrient water (5 gN.[m.sup.-2].[day.sup.-1]; 0.6 gP.[m.sup.-3].[day.sup.-1]) (Boarder & Shpigel 2001). The Ulva germlings as a 3-dimensional juvenile macroalgae are in a phase of high growth and therefore may be able to utilize the limited nutrient supply in the water extremely well compared to what the 2-dimensional U. lens/N. jeffreyi diet can.

The total percentage extracted ([approximately equal to] 65%) from the U. lens/N. jeffreyi diet may have been reduced because of the combination of microalgae present within the diet. Brown and Jeffery (1995) extracted only 56% from N. jeffreyi with protein as the major constituent and carbohydrate the lowest, whereas 12% lipid, 28% protein and 7% carbohydrate have been extracted for a combination of diatoms (Brown et al. 1997). The biochemical composition can vary considerably between diatom species let alone a diet containing U. lens, diatoms and biofilm components (Brown 1991, Brown et al. 1997, Lewin & Guillard 1963).

Importantly the lower level of lipid, protein and carbohydrate present within the U. lens/N. jeffreyi diet produced significantly larger abalone at the end of the 14 wk feeding trial. Therefore the higher amounts of the biochemical components do not increase growth but rather an optimal level may be responsible. Lipid levels of 4% to 5% have been shown to be optimal for abalone, which corresponds with the U. lens/N. jeffreyi diet, whereas the Ulva germling diet was higher (Dunstan et al. 2000, Uki & Watanabe 1992). High levels ([greater than or equal to] 5%) of dietary lipid have been shown to be detrimental to abalone growth and are believed to depress the digestibility of other nutrients (Britz & Hecht 1997, Uki & Watanabe 1992, Van Barneveld et al. 1998). Therefore the high lipid level present in the Ulva germling diet may have restricted the optimal growth rates obtained in the first month of the trial.

Optimal protein levels of 28% have been reported but can range from 20% to 35% depending on abalone species (Britz & Hecht 1997, Coote et al. 2000, Mai et al. 1995, Uki & Watanabe 1992, Vandepeer & Van Barneveld 2002). However to maximize protein utilization not only should the diet contain sufficient readily digestible protein but a well balanced mixture of essential and nonessential amino acids (Britz & Hecht 1997, Mai et al. 1995). Therefore the U. lens/N. jeffreyi diet may provide biochemical components closer to the optimal levels for juvenile H. laevigata because it produced larger individuals.

The N. jeffreyi consumption (Fig. 5) did not correlate with either of the algae species nor any of the abalone results. Weeks 7-10 and 12 all produced negative consumption indicating that the N. jeffreyi was reproducing faster than the abalone could consume it. The consumption rate of N. jeffreyi did not correspond with the reinoculation but rather the plate rotation 180[degrees] about the horizontal.

This can be accounted for by two reasons; firstly because the light gradient through the tanks allowed N. jeffreyi situated at the top to receive greater light intensity inducing faster growth and secondly because changes in grazing pressure caused by light sensitivity/migration of abalone. These notions were substantiated through visual observation during the trial because diatom counts increased when at the top and juvenile abalone were found on the bottom of the tanks during the day. Positive relationships between feed intake and the duration of darkness have been shown by Dixon (1992) and Fleming et al. (1996), hence, when the juvenile abalone migrate from the bottom to feed high density of diatoms are closer reducing the effort expended to graze. Daume et al. (2004) indicated that the light intensity tended to be higher at the top of plates and migration to the bottom of the tank was evident.

When the plates were rotated at week 3 the high N. jeffreyi densities at the top were transferred to the bottom. This caused the high densities to be closer to the majority of abalone and subsequently caused a spike in consumption (Fig. 5). N. jeffreyi consumption dropped as the density at the bottom declined, whereas the density increased at the top because of greater light intensity and less grazing pressure. At week 7 the consumption became negative because the growth at the top exceeded the consumption at the bottom. Once the second rotation at week 10 was performed the consumption began to increase again as the high N. jeffreyi density was available to the juvenile abalone at the bottom. The two rotational profiles (i.e., top-bottom-top vs. bottom-top-bottom) were shown to be significantly different. Therefore it would be beneficial to rotate diatom-cultured plates at least weekly to maintain high diatom densities at the bottom where the majority of abalone reside.

The contaminating diatom species only occurred in the Ulva germling diet tanks indicating that it was present from the creation of the diet. It was not determined if the juvenile abalone utilized these contaminating diatoms as a food source although Synedra sp. was probably of a suitable size. It may have been difficult for the juveniles to deal with the Melosira cf. moniliformis because along with its large cell size it proceeded to rapidly form into dense mats with chains exceeding 5 cm in length, The large blooms of M. moniliformis occurring at week 11 and 13 may have had some impact on the declining growth rates because the juveniles are susceptible to smothering and entanglement (Daume et al. 2004).

Ulva germlings are a suitable feed for juvenile H. laevigata because they produced comparable absolute and specific growth rates to the Ulvella lens and Navicula cf. jeffreyi diet currently used in commercial aquaculture. Further investigation into the theoretical and procedural principles behind the development of the Ulva germling diet could allow for the diet to incorporate a variety of different algal species.

ACKNOWLEDGMENTS

The authors thank Fiona Graham, Sam Hair and William Strong for their assistance in material collection and routine measurements; Challenger TAFE for the space to run the feeding trial and Great Southern Marine Hatcheries for supplying the juvenile greenlip abalone. This study was funded as part of an FRDC project (2002/203) and conducted as part of a PhD by the first author.

LITERATURE CITED

Ben-Amotz, A., T. G. Tornabene & W. H. Thomas. 1985. Chemical profile of selected species of microalgae with emphasis on lipids. J. Phycol. 21:72-81.

Bligh, E. G. & W. J. Dyer. 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37:911-917.

Boarder, S. & M. Shpigel. 2001. Comparative performances of juvenile Haliotis roei fed on enriched Ulva rigida and various artificial diets. J. Shellfish Res. 20:653-657.

Britz, P. J. & T. Hecht. 1997. Effect of dietary protein and energy level on growth and body composition of South African abalone Haliotis midae. Aquaculture 156:195-210.

Brown, M. R. 1991. The amino-acid and sugar composition of 16 species of microalgae used in mariculture. J. Exp. Mar. Biol. Ecol. 145:79-99.

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

Brown, M. R., S. W. Jeffery, J. K. Volkman & G. A. Dunstan. 1997. Nutritional properties of microalgae for mariculture. Aquaculture 151: 315-331.

Coote, T. A., P. W. Hone, R. J. Van Bameveld & G. B. Maguire. 2000. Optimal protein level in a semipurified diet for juvenile greenlip abalone Haliotis laevigata. Aqua. Nutrition 6:213-220.

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. 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.

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. Ryan. 2004. Nursery culture of the abalone Haliotis laevigata: Larval settlement and juvenile production using algae or formulated feed. J. Shellfish Res. 23:1019-1026.

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

Dixon, M. G. 1992. The effect of temperature and photoperiod on the digestive physiology of the South African abalone Haliotis midae. Master Thesis. Rhodes University, South Africa. pp. 85.

Dorsey, T. E., P. McDonald & D. A. Roels. 1978. Measurement of phytoplankton protein content with the heated biuret-folin assay. J. Phycol. 14:167-171.

Dunstan, G. A., M. R. Brown, G. B. Maguire & J. K. Volkman. 2002. Formulated feeds for newly settled juvenile abalone based on natural feeds (diatoms and crustose coralline algae), pp. 78. FRDC Project No. 1996/386. CSIRO Marine Research.

Dunstan, G. A., J. K. Volkman & G. B. Maguire. 2000. Optimisation of essential lipids in artificial feeds for Australian abalone, pp. 68. FRDC Project 94/85. CSIRO Marine Research.

Fleming, A. E. 1995. Growth, intake, feed conversion efficiency and chemosensory preference of the Australian abalone, Haliotis rubra. Aquaculture 132:297-311.

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

Guillard, R. R. L. & J. H. Ryther. 1962. Studies of marine planktonic diatoms. I. Cyclotella nana Hustedt, and Detonula confervacea (Cleve) Gran. Can. J. Microbiol. 8:229-239.

Hahn, K. O. 1989. Nutrition and growth of abalone. In: K. O. Hahn, editor. Handbook of culture of abalone and other marine gastropods. Boca Raton, Florida: CRC Press. pp. 135-156.

Hone, P. W. (1992). Some aspects of the nutritional biology of the blacklip abalone Haliotis rubra. In: Biotechnology Aquaculture Workshop, Abalone artificial diets.

Kates, M. & B. E. Volcani. 1966. Lipid composition of diatoms. Biochim. Biophys. Acta 116:264-278.

Kawamura, T. 1996. The role of benthic diatoms in the early life stages of the Japanese abalone (Haliotis discus hannai). In: Y. Watanabe, Y. Yamashita & Y. Oozeki, editors. Survival strategies in early life stages of marine resources. Rotterdam: A. A. Balkema. pp. 355-367.

Kawamura, T., R. D. Roberts & H. Takami. 1998. A review of the feeding and growth of postlarval abalone. J. Shellfish Res. 17:615-525.

Kawamura, T., T. Saido, H. Takami & Y. Yamashita. 1995. Dietary value of benthic diatoms for the growth of post-larval abalone Haliotis discus hannai. J. Exp. Mar. Biol. Ecol. 194:189-199.

Kawamura, T. & H. Takami. 1995. Analysis of feeding and growth rate of newly metamorphosed abalone Haliotis discus hannai fed on four species of benthic diatom. Fish. Sci. 61:357-358.

Kochert, G. 1978. Carbohydrate determined by the phenol-sulfuric acid method. In: J. A. Hellebust, & J. J. Craigie, editors. Handbook of physiological methods: physiological and biochemical methods. Cambridge: Cambridge University Press. pp. 95-97.

Krsinich, A., S. Daume, S. Farrell, M. Gervis & P. Thompson. 2000 Towards intensification of an abalone (Haliotis rubra) nursery operation via inoculation with a benthic diatom Navicula sp. and seeding with the macroalgae Ulvella lens. In: A. E. Fleming, editor. Proceedings of the 7th Annual Abalone Aquaculture Workshop. Fisheries Research Development Corporation's Abalone Aquaculture Subprogram, Dunedin, New Zealand. pp. 86-99.

Lewin, J. C. & R. R. L. Guillard. 1963. Diatoms. Annu. Rev. Microbiol. 17:373-414.

Lowry, O. H., N. J. Rosebrough, A. Farr & R. Randell. 1951. Protein measurement with Folin-phenol reagent. J. Biol. Chem. 193:265-275.

Maesako, N., S. Nakamura & T. Yotsui. 1984. Food effect of brown and green algae of early developmental stage and blue green algae for the growth of the juvenile abalone, Haliotis discus Reeve (in Japanese with English abstract). Bull. Nagasaki Pref Inst. Fish. 10:53-56.

Mai, K., J. P. Mercer & J. Donlon. 1995. Comparative studies on the nutrition of two species of abalone, Haliotis tuberculata L. and Haliotis discus hannai Ino IV. Optimal dietary protein level for growth. Aquaculture 136:165-180.

Mai, K., J. P. Mercer & J. Donlon. 1996. Comparative studies on the nutrition of two species of abalone, Haliotis tuberculata L. and Haliotis discus hannai Ino. V. The role of polyunsaturated fatty acids of macroalgae in abalone nutrition. Aquaculture 139:77-89.

McShane, P. E., H. K. Gorfine & I. A. Knuckey. 1994. Factors influencing food selection in the abalone Haliotis rubra (Mollusca: Gastropoda). J. Exp. Mar. Biol. Ecol. 176:27-37.

Mercer, J. P., K. Mai & J. Donlon. 1993. Comparative studies on the nutrition of two species of abalone, Haliotis tuberculata L. and Haliotis discus hannai Ino. I. Effects of algal diets on the growth and biochemical composition. Invert. Reprod. Devel. 23:75-88.

Mercz, T. I. 1994. A study of high lipid yielding microalgae with potential for large-scale production of lipids and polyunsaturated fatty acids. PhD Thesis. Murdoch University, Perth, Australia. pp. 278.

Shpigel, M., N. L. Ragg, I. Lupatsch & A. Neori. 1999. Protein content determines the nutritional value of the seaweed Ulva lactuca L. for the abalone Haliotis tuberculata L. and Haliotis discus hannai Ino. J. Shellfish Res. 18:227-233.

Simpson, B. J. & P. A. Cook. 1998. Rotation diets: A method of improving growth of cultured abalone using natural algal diets. J. Shellfish Res. 17:635-640.

Stuart, M. D. & M. R. Brown. 1994. Growth and diet of cultivated black-footed abalone, Haliotis irris (Martyn). Aquaculture 127:329-337.

Takami, H., T. Kawamura & Y. Yamashita. 1997. Contribution of diatoms as food sources for post-larval abalone Haliotis discus hannai on crustose coralline alga. Moll. Res. 18:143-151.

Takami, H., T. Kawamura & Y. Yamashita. 1998. Development of polysaccharide degradation activity in postlarval abalone Haliotis discus hannai. J. Shellfish Res. 17:723-727.

Takami, H., D. Muraoka, T. Kawamura & Y. Yamashita. 2003. When is the abalone Haliotis discus hannai Ino 1953 first able to use brown macroalgae? J. Shellfish Res. 22:795-800.

Taylor, M. H. & E. Tsvetnenko. 2004. A growth assessment of juvenile abalone Haliotis laevigata fed enriched macroalgae Ulva rigida. Aqua. Int. 12:467-480.

Uki, N. & Y. Watanabe. 1992. Review of the nutritional requirements of abalone (Haliotis sp.) and development of more efficient artificial diets. In: S. A. Shepherd, M. J. Tegner & S. A. Guzman del Proo, editors. Abalone of the world: biology, fisheries and culture. Oxford: Blackwell Scientific Publishing. pp. 504-517.

Van Alstyne, K. L., J. M. Ehlig & S. L. Whitman. 1999. Feeding preferences for juvenile and adult algae depend on algal stage and herbivore species. Mar. Ecol. Prog. Ser. 180:179-185.

Van Barneveld, R. J., A. E. Fleming, M. E. Vandepeer, J. A. Kruk & P. W. Hone. 1998. Influences of dietary oil type and oil inclusion level in manufactured feeds on the digestibility of nutrients by juvenile greenlip abalone (Haliotis laevigata). J. Shellfish Res. 17:649-655.

Vandepeer, M. E. & R. J. Van Barneveld. 2002. The effect of different dietary digestible protein: digestible energy ratios on growth of juvenile blacklip abalone (Haliotis rubra). In: A. E. Fleming, editor. Proceedings of the 9th Annual Abalone Aquaculture Workshop. Fisheries Research and Development Corporation's Abalone aquaculture Subprogram, Queenscliff, Australia. pp. 50-60.

LACHLAN W. S. STRAIN, (1) * MICHAEL A. BOROWITZKA (1) AND SABINE DAUME (2)

(1) School of Biological Sciences & Biotechnology, Murdoch University, South Street, Murdoch, WA 6150, Australia; (2) Research Division, Department of Fisheries, Western Australia, PO Box 20, North Beach, WA 6920, Australia

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

TABLE 1.

Weekly changes in shell length of juvenile Haliotis laevigata grown
on an Ulva germling diet or an U. lens/N. jeffreyi diet, indicated by
the mean increase difference and significance (Univariate Analysis
of Variance, post-hoc Tukey HSD tests).

                    Ulva Germling Diet       U. lens/N. jeffreyi Diet

                   Mean                        Mean
                Difference   Significance   Difference   Significance
Week Interval      (mm)       (P-Value)        (mm)       (P-Value)

Start-1            0.24         0.904          0.00         1.000
    1-2            0.40         0.019          0.40         0.018
    2-3            0.67         0.000          0.67         0.000
    3-4            0.72         0.000          0.56         0.019
    4-5            0.52         0.052          0.80         0.000
    5-6            0.44         0.212          0.60         0.009
    6-7            0.59         0.010          0.70         0.000
    7-8            0.34         0.636          0.44         0.219
    8-9            0.41         0.331          0.46         0.151
    9-10           0.50         0.085          0.39         0.663
   10-11           0.41         0.322          0.66         0.002
   11-12           0.47         0.143          0.38         0.461
   12-13           0.08         1.000          0.23         0.972
   13-14           0.08         1.000          0.25         0.949

TABLE 2.

Weekly growth rates for juvenile Haliotis laevigata combined into
monthly periods (n = 3) for both an Ulva germling diet and an U.
lenslN. jeffreyi diet. The first 2 weeks were excluded to allow time
for the juvenile abalone to adapt to the experimental conditions
and diets.

                                   Week 3-6

                             Mean GR
        Diet          ([micro]m.[day.sup.-1])    SE

Ova germling                  84.1               9.5
U. lens/N. jeffreyi           94.1               5.0

                                   Week 7-10

                            Mean GR
        Diet          ([micro]m.[day.sup.-1])    SE

Ova germling                  66.1               8.7
U. lens/N. jeffreyi           68.2               7.5

                                   Week 11-14

                            Mean GR
        Diet          ([micro]m.[day.sup.-1])     SE

Ova germling                  35.7               11.8
U. lens/N. jeffreyi           52.9               11.7

TABLE 3.

Weekly mortality of juvenile Haliotis laevigata for both the Ulva
germling diet and the U. lenslN. jeffreyi diet. Mean [+ or -] std.
error (n = 3). (Initial number of abalone per replicate tank was
1,700).

        U. lens/N. jeffreyi       Ulva Germling

Week     Mean           SE      Mean           SE

 1        63           5.86      71           5.81
 2        81          23.16      85          22.15
 3        91          22.40      56           6.44
 4        64          14.19      72          12.01
 5        53           3.21      57           3.53
 6        27           2.91      59          22.21
 7        16           3         45          11.35
 8        18           2.65      28           2.33
10        24           1.73      25           1.67
12        17           4.63      12           2
14        15           1.76      12           0.88

TABLE 4.

The weeks, in which inoculation of N. jeffreyi occurred, the plates
were rotated and when contaminant diatom species were observed,
including their relative size.

                                1st          2nd        3rd

Inoculation                 Week 2         Week 4    Week 8
Plates rotated              Week 3         Week 10
Contamination present
Melosira cf. moniliformis   Before start   Week 11   Week 13
Synedra sp.                 Before start   Week 11   Week 13
Contamination size          Cell length    SE        Cell width    SE
                            ([micro]m)               ([micro]m)
Melosira cf. moniliformis   109.5          2.09       8.9         0.29
Synedra sp.                  21.5          0.81      19.7         1.42

TABLE 5.

The biochemical composition of both the Ulva germling diet and the
U. lens/N. jeffreyi diet. Values are on dry matter basis and expressed
as g/100g dry weight with standard errors in parentheses (n = 9).

Diet            Dry Weight       Lipid        Protein      Carbohydrate

Ova germling   12.44 (0.54)   7.12 (1.55)   32.30 (1.84)   43.86 (5.87)
U. lens/N.
  jeffreyi     12.82 (0.57)   4.37 (0.95)   24.17 (1.64)   35.85 (2.76)