1 INTRODUCTION

The donkey's ear abalone Haliotis asinina is one of the most, if not the most, commercially important gastropods in the Philippines. As the largest and the fastest growing among the tropical species of abalone (Capinpin, 1995; McNamara & Johnson, 1995), its potential for large-scale commercial culture is promising. Aside from this, the culture of abalone should be encouraged to avoid further decline in natural stocks and fishery collapse, as wild stocks remain the major source of abalone fished for consumption and trade. For more than two decades, the Aquaculture Department of the Southeast Asian Fisheries Development Center (SEAFDEC/AQD) in Tigbauan, Iloilo, Philippines, has been actively doing research in developing and improving aquaculture technologies for H. asinina. Feed development has been, and still is, an important research area at SEAFDEC/AQD. Nevertheless, abalone culture existing sporadically in the Philippines depends mainly on natural food (Lebata-Ramos & Solis, 2021). While in most developed countries, the large-scale grow-out culture of abalone is dependent on formulated diets (Buss et al., 2015), the use of traditional feed, the seaweeds, remains a common practice in the country. Formulated diets, which are more nutrition balanced, economical and convenient, are highly recommended over macroalgae for practical abalone production in farming (Choi et al., 2018; Kirkendale et al., 2010).

Abalone are known grazers (Hernández & Siqueiros-Beltrones, 2010). Typically, they scrape surfaces with their mouthpart called the radula (Kirkendale et al., 2010) and forage by actively grazing on attached benthic algae (Shepherd, 1973). In a study of wild abalone population, despite their patchy distribution, abalone density showed a significantly positive correlation with substrates of dead branching corals with encrusting algae (Lebata-Ramos et al., 2013). It is evident that in their natural environment, abalone preferred substrates, which can provide their basic needs, shelter and food. While the complex branching corals provide a good hiding place from predators, these very thin benthic algae forming flattened mats on dead corals make a very convenient grazing field for abalone to feed.

Kirkendale et al. (2010) listed the attributes of a good feed as reliability and quality of supply, convenience and ease of application, balanced nutritional composition, cost efficiency, digestibility, optimum particle size and feeding methods, attractiveness to abalone, and ease of storage and transportation. But looking at most of the available formulated diets, the grazing behaviour of abalone seemed to have not been considered during feed development. Aside from the noodle form commonly used in feeding experiments at SEAFDEC/AQD, other abalone diet forms include 0.15 cm thick sheets, cut into 1 cm² (Ansary et al., 2019; Choi et al., 2018; Jung et al., 2016; Kim et al., 2016); 2 × 2 × 3.5 cm cubes (Angell et al., 2012); 0.5 cm dia tablets (Susanto et al., 2016); 4 × 3 × 2 mm flat pellets (Bansemer et al., 2014); pellets (Yun et al., 2018); and powder, crumble and chips (Kirkendale et al., 2010), among others. For a crawling grazer, however, a very thin and flattened feed may be easier to access and forage. To check on what has been done on the use of flakes as feed for the abalone, Google Scholar was searched using the keywords “Haliotis asinina flake diet” (Google, 2021). This resulted in only 32 articles, none of which reported the use of flaked diet for H. asinina, mostly on other species of Haliotis.

For the past two decades, most of the studies done in the Philippines on developing feeds for abalone were done in containers not intended for use in the actual grow-out culture – trays or 60-L fiberglass tanks provided with halved polyvinyl chloride (PVC) tubes as shelters; and mostly the feeds were in the form of noodles or flat squares (Bautista-Teruel et al., 2011, 2016; Bautista-Teruel & Millamena, 1999). This study was conducted using two pre-existing diet formulations for abalone grow-out culture, refined and unrefined (Bautista-Teruel et al., 2016). Two diet forms were prepared, flakes and the commonly used form noodles, and fed to H. asinina reared in culture containers designed for large-scale grow-out culture (Lebata-Ramos et al., 2021), not the usual experimental containers.

2 METHODS

2.4 Statistical analysis

Statistical analysis and graphs were done using the Minitab 17.0 software package (Minitab, http://www.minitab.com). Growth parameters and FCR from different treatment diets (MBTr-F = refined flakes; MBTr-N = refined noodles; MBTur-F = unrefined flakes; MBTur-N = unrefined noodles) were tested for normality using the Anderson–Darling test and compared using analysis of variance (ANOVA). Tukey's test was used to compare differences between diets, where growth parameters and FCR significantly differ at p < .05 statistical significance level. Survival data were arcsine-transformed before proceeding with the same analyses used for the growth parameters.

Monthly growth rates both for length (GR_L) and weight (GR_W) were obtained using the following equations:

()

()

where GR is the growth rate; L is the length (mm); W is the weight (g); T is the sampling date; f is the present sampling; i is the preceding sampling; 30 = constant, days in a month.

Survival (%) and feed conversion ratio (FCR) were obtained using Equations 3 and 4 respectively:

()

where N_f is the number of abalone in each container during harvest; N_i is the number of abalone initially stocked per container (30).

()

where F_t is the total feed given per container (g); W_t is the total weight gained per container, W_f − W_i; W_f is the total final weight of abalone per container (g); W_i is the total initial weight of abalone per container (g).

3 RESULTS

Mean shell length and body weight of abalone started showing significant differences between treatment diets used 1 month from the start of the experiment (ANOVA: p < .05, F = 3.02, df = 3 and p < .05, F = 3.46, df = 3, respectively). This trend was observed every 15 days for both growth parameters until the termination of the experiment on day 92 (Figure 1), where abalone fed MBTr-F had significantly higher mean shell length (35.92 ± 0.36 mm) compared with those fed MBTr-N (34.63 ± 0.37 mm) and MBTur-N (34.23 ± 0.36 mm), but did not significantly differ with those fed MBTur-F (35.40 ± 0.30 mm; Figure 1a; ANOVA: p < .01, F = 4.77, df = 3). The same was observed for mean body weight, where abalone fed MBTr-F were significantly heavier (9.24 ± 0.29 g) compared with those fed MBTr-N (8.22 ± 0.28 g) and MBTur-N (7.66 ± 0.26 g), but did not significantly differ with those fed MBTur-F (8.57 ± 0.24 g; Figure 1b; ANOVA: p < .001, F = 6.15, df = 3).

Mean growth rates of abalone, both in terms of length and weight, were significantly highest when fed with MBTr-F at 5.31 ± 0.08 mm/month and 2.59 ± 0.06 g/month, respectively (Table 3). Abalone fed MBTr-F had significantly higher mean growth rates, in terms of length, compared with those fed MBTr-N and MBTur-N, but did not significantly differ with those fed MBTur-F (Table 3; ANOVA: p < .05, F = 6.34, df = 3). Moreover, abalone fed MBTr-F had significantly higher growth rates, in terms of weight, compared with those fed MBTr-N and MBTur-N, but did not significantly differ with those fed MBTur-F (Table 3; ANOVA: p < .01, F = 9.16, df = 3).

Abalone survival did not significantly differ between treatment diets during the bi-monthly samplings from day 15 until day 92. Upon termination of the experiment, survival was highest in abalone fed MBTr-F at 98.89 ± 1.11%, while those fed with MBTr-N, MBTur-F and MBTur-N each had equal survival of 96.67 ± 0.00, 96.67 ± 1.9, and 96.67 ± 3.33%, respectively (Table 3). Feed conversion ratios (FCR) were lower in abalone fed with flakes as compared with those fed with noodles (Table 3). Nevertheless, FCR did not show significant differences among treatment diets.

Abalone fed unrefined formulated diets, regardless of form, developed a distinct turquoise shell colour (Figure 2). On the other hand, those fed with refined formulated diets, regardless of form, maintained the natural colour of their shells, with some indistinct cream to light green streaks.

4 DISCUSSION

Growing abalone from 10 to 15 mm to approximately 30-mm shell length falls under the nursery culture phase. In the Philippines, the usual practice, starting from nursery culture until harvest, is to feed H. asinina with G. heteroclada. In the present study, the highest growth rates of abalone were attained when they were fed with MBTr-F – 5.31 ± 0.08 mm/month and 2.59 ± 0.06 g/month (Table 3). Nursery culture of abalone using the same culture containers in tanks and the same culture duration, but fed with G. heteroclada resulted in growth rates of 6.28 mm/month and 2.39 g/month (Lebata-Ramos, 2018). Likewise, in a sea-based nursery culture experiment using G. heteroclada as feed, growth rates were 6.65 mm/month and 2.31 g/month (Encena et al., 2013). In a feeding trial comparing G. heteroclada and formulated diet, Capinpin and Corre (1996) reported higher growth rates in terms of length in abalone fed with G. heteroclada (5.79 mm/month) compared with those fed with formulated diet (5.75 mm/month), but opposite in terms of weight (G. heteroclada fed = 2.01 g/month; formulated diet fed = 2.21 g/month). The results of both the present and previous studies have been consistent in exhibiting faster growth rates in the length of abalone fed with G. heteroclada, while faster growth rates in the weight of those fed with formulated diet. It can be inferred that formulated diet, which is more nutritionally balanced and complete, promotes muscle development, while natural food seems to promote shell growth.

Survival of H. asinina was high and did not significantly differ among all four treatment diets, with means ranging 96.7%–98.9%. Although Bautista-Teruel and Millamena (1999) reported lower survival (85%–95%) in a tank experiment of the same duration for almost the same size of abalone, the results of the study were almost parallel to the survival rates obtained in tank culture experiments of Capinpin and Corre (1996) on diet (98.3%–100.0%); Setyono and Aswandy (2007) and Jarayabhand et al. (2010) on density (100% and 92.9%–95.7%, respectively); Setyono (2015) on substrate (98%–100%) and Lebata-Ramos et al. (2021) on culture container (94.88%–98.64%). These were higher and less variable than the survival obtained in sea-based culture systems reported by Setyono (2007) and Encena et al. (2013) at 83.3%–97.9% and 79.8%–87.0%, respectively. High survival in tanks may be attributed to a more controlled environment and maintained good water quality. Hence, the use of these formulated diets in a sea-based culture system should be also tested before promoting their use in open water culture systems as results may differ between culture systems.

The feed conversion ratios (FCR) were relatively low (2.2–2.5) in all four treatment diets. Although these did not significantly differ among treatments, it is notable that they were lower in abalone fed flaked diet, regardless of formulation (Table 3). Using natural food, usually Gracilariopsis spp., in nursery culture, FCR can greatly vary – 12.58–15.54 (Capinpin et al., 1999), 13.2 (Nguyen et al., 2010), 7.8–11.3 (Encena et al., 2013) and 9.37–12.69 (Amin et al., 2020). This may even double in grow-out culture, at 19.63–22.57 (Capinpin et al., 1999), 16.9–25.8 (Encena et al., 2013), or as high as 31–40 (Fermin & Buen, 2000). Contrastingly, using formulated diet, FCR ranged 1.01–5.62 (Bautista-Teruel et al., 2011, 2016; Bautista-Teruel & Millamena, 1999). With the availability of MBTur-F (US$2.08 per kg), between natural diet (US$0.40 per kg) and formulated diet, the use of the latter in large-scale culture is the more practical option. Apart from the lower FCR, formulated diets do not necessitate large storage space, whereas natural food needs to be kept alive, hence may require yet another large facility for rearing and storing and entails additional expense. This applies regardless of whether culture is land based or sea based.

The use of refined (MBTr-F and MBTr-N) and unrefined (MBTur-F and MBTur-N) diets in this experiment resulted in abalone with two distinct shell colours; natural colour and turquoise, respectively (Figure 2). The effect of diet, whether natural or formulated, on the shell colour of abalone had also been documented in other species of Haliotis (Liu et al., 2009; Marchais et al., 2017). A light blue–green shell colouration had been reported, by Capinpin and Corre (1996) and Bautista-Teruel and Millamena (1999), in H. asinina fed with formulated diet. The latter formulation was later used and recommended by Gallardo et al. (2003) for sea ranching and stock enhancement of abalone as they resulted in permanent shell tags. This method of permanently marking abalone shells through diet tagging had been adopted and successfully employed in separate stock enhancement efforts in two reef areas in Sagay Marine Reserve, Philippines (Lebata-Ramos et al., 2013; Salayo et al., 2020). The unrefined diet formulation (MBTur) used in this experiment (Bautista-Teruel et al., 2016) has almost the same major ingredients as the earlier formulation of Bautista-Teruel and Millamena (1999), which also resulted in the turquoise shell colour of the abalone fed with this diet. MBTr differed from the other diet because of the inclusion of Spirulina spp. and sodium alginate in its formulation (Table 1), which may have maintained the natural shell colour of abalone. The presence of these two or their interactions may have neutralized the effect of the other ingredients in changing the natural colour of the abalone's shell. However, just to maintain the natural colour of the shell, it is not economical to incorporate them in the feed as it increases its cost fivefold.

5 CONCLUSION

Abalone fed MBTr-F were significantly larger and heavier than those fed with MBTr-N and MBTur-N, but did not significantly differ than those fed with MBTur-F. The same was observed for growth rates in terms of both length and weight. However, when used independently, MBTr, regardless of form, significantly promotes muscle development compared with MBTur, while regardless of formulation, flakes significantly promote both shell and muscle growth compared with noodles. Although there were no significant differences with the other treatment diets, MBTr-F fed abalone had the highest survival and lowest FCR. Based on these results, MBTr-F is the best among the four diets tested. However, with no significant differences in all parameters tested between MBTr-F and MBTur-F, except for weight, it can be deduced that diets in flake form are more efficient as feed for the grazing abalone. Hence, practically, the cheaper MBTur-F, which fared almost the same as MBTr-F, may be a better feed option because producing MBTur-F costs only one-fifth the price to produce the same volume of MBTr-F.

The high growth rates and survival of H. asinina in the present study, using formulated flaked diets and culture containers designed for use in land- and sea-based commercial production systems, show the strong potential of the species for large-scale culture. Moreover, formulated diets may also be manipulated by adding necessary ingredients to improve abalone's meat quality through enhanced taste and texture (Brown et al., 2008). This brings a glimpse of hope to the Philippine abalone industry, as the promising results of this experiment may open the doors of acceptance and use of formulated feed in large-scale abalone culture in the country, which barely exists because of its dependence on seaweeds as feeds. This will help boost abalone production in the country, which according to FAO statistics is mainly dependent on capture fisheries since 1976 (FAO, 2020b), but none recorded from aquaculture (FAO, 2020a). Based on the results of this experiment, it is high time to introduce the use of flaked formulated diet to the abalone industry, if not as a sole diet, as an alternate or mixed feed to the more popular and commonly used, G. heteroclada.